Dedication Professor Ian R. Griffiths, Glasgow University Medical School and Professor Joe N. Kornegay, College of Vete...
185 downloads
1553 Views
65MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Dedication Professor Ian R. Griffiths, Glasgow University Medical School and Professor Joe N. Kornegay, College of Veterinary Medicine, University of Missouri For their inspiration, encouragement and guidance
For Elsevier: Commissioning Editor: Joyce Rodenhuis Senior Development Editor: Zoë A Youd Project Manager: Ailsa Laing Designer: Andrew Chapman
© 2005, Elsevier Limited. All rights reserved.
Copyright 1994 Times Mirror International Publishers Limited Copyright 2000 Harcourt Publishers Limited © 2004 Elsevier Limited. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without either the prior permission of the publishers or a licence permitting restricted copying in the United Kingdom issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Permissions may be sought directly from Elsevier’s Health Sciences Rights Department in Philadelphia, USA: phone: (⫹1) 215 238 7869, fax: (⫹1) 215 238 2239, e-mail: healthpermissions @elsevier.com. You may also complete your request on-line via the Elsevier Science homepage (http://www.elsevier.com), by selecting ‘Customer Support’ and then ‘Obtaining Permissions’.
First edition 1994 Second edition 2005 ISBN 0723432090 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book is available from the Library of Congress Notice Knowledge and best practice in this field are constantly changing. As new research and experience broaden our knowledge, changes in practice, treatment and drug therapy may become necessary or appropriate. Readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of the practitioner, relying on their own experience and knowledge of the patient, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the publisher nor the editors assumes any liability for any injury and/or damage. The Publisher
Printed in China
Preface
The second edition of Small Animal Spinal Disorders now expands considerably on the foundations of neuroanatomy, clinical neurology and basic neurosurgery provided in the first edition. The primary aim of the second edition remains to assist students, general practitioners and specialists in the diagnosis and understanding of the latest neurosurgical procedures used to manage dogs and cats with spinal disease. The book has now been completely rewritten and reorganized for 2005. Most of the pre-1994 literature has been reviewed again so that we do not forget lessons learned previously and also so that this knowledge can be placed in proper context with more recent information. Many exciting contributions have been made in the last 10 years, but not surprisingly veterinary neurosurgery still tends to suffer from limited case numbers, with follow-up periods often less than a residencytraining period in duration. Extensive reference has therefore been made to the human neurosurgical literature, both in an attempt to gain a different perspective and to stimulate a continued reassessment of our current state of knowledge. Despite obvious differences and limitations, the greater maturity and collective knowledge of the human specialty provides important insights into veterinary neurosurgery. In addition to being completely rewritten, the second edition now contains 50% more figures. A much greater emphasis has now been placed on the advanced imaging techniques of CT and MRI; the chapters on functional anatomy and diagnostic aids in particular reflect these changes. Advanced imaging is also used liberally in the other chapters to illustrate surgical anatomy, pathology and complications. Another new innovation emphasizes the problem of postoperative complications. A specific focus is the wide variety of complications that can occur during the management of spinal disorders. Complications are
generally divided into intraoperative, early postoperative and late postoperative. These are listed in tabular format and are also discussed and illustrated where possible in the main text. Furthermore, prognostic information continues to be derived mainly from the literature rather than from anecdote, at least where information is available. Physiotherapy now receives four pages of special emphasis in the postoperative care chapter. An additional change is the separation of the procedures section from the main chapter, so that photographs of surgical techniques no longer interrupt the text. Finally, at the end of most chapters, key issues for future investigation have been identified, to highlight areas where we need to focus our creative energy and future research efforts. Additional information relating to this text and to neurosurgery in general has been made available on the Internet at www.vetneurosurgery.com. This includes direct links that provide abstract information to many of the citations used in the second edition, along with further information on CT and MRI anatomy. It also provides many links to other websites that contain information aimed at owners, referring veterinarians and specialists. There will come a time when veterinary neurosurgery becomes a specialty recognized in its own right, just as it is in human healthcare. Until that time it will fall upon individuals, mainly from the European and North American specialties of neurology and surgery, to advance our knowledge one step at a time. This book aims to bring all of the various quite disparate sources of information together, in order to provide a comprehensive summary of the current state of knowledge of veterinary neurosurgery.
Vancouver and Hertfordshire 2005
Nick Sharp Simon Wheeler
List of Abbreviations
ACE b.i.d. CDV CK CMC CNS COX CSM CSF CT DISH DVT ECG EMG FCE FeLV FIP FIV GI GME IM
angiotensin-converting enzyme bis in die (twice daily) canine distemper virus creatine kinase cerebello medullary cistern central nervous system cyclooxygenase cervical spondylomyelopathy cerebrospinal fluid computed tomography disseminated idiopathic skeletal hyperostosis deep vein thrombosis electrocardiogram electromyography fibrocartilaginous embolism feline leukemia virus feline infectious peritonitis feline immunodeficiency virus gastrointestinal granulomatous meningoencephalomyelitis intramuscular
IV LMN MPSS MR MRI NSAID OCD PCV PgE PO PTE RBC SQ STIR TMJ TSH UMN UTI VW WBC
intravenous lower motor neuron methylprednisolone sodium succinate magnetic resonance magnetic resonance imaging non-steroidal anti-inflammatory drug osteochondritis dissecans packed cell volume prostaglandin E per os pulmonary thromboembolism red blood cells subcutaneous short tau inversion recovery temporomandibular joint thyroid-stimulating hormone upper motor neuron urinary tract infection von Willebrand white blood cell
Functional anatomy
Skeleton 6 Vertebrae 6 Articulations 9 Blood supply 14 Vertebral column 14 17
Further reading
17
Knowledge of functional anatomy is important both for understanding the neurological examination and for performing spinal surgery. This chapter concentrates on clinically relevant points of anatomy and physiology, including surgical landmarks. Radiographs, CT scans and MRIs have been used for illustration, and the reader is also directed to the normal radiographic anatomy illustrated in Chapter 4. For more detail see ‘References’, page 17.
c a
d
b e
1
NERVOUS TISSUE Spinal cord
Nervous tissue 1 Spinal cord 1 Relationship of spinal cord segments to vertebrae 2 Cauda equina 3 Meninges 3 Cerebrospinal fluid 4 Spinal cord white matter tracts 5 Spinal cord nerve fibers and the effect of compression 6
References
Chapter
The spinal cord lies within the vertebral canal, fitting snugly within the thoracolumbar spine, but with more space in the cervical spine. The residual space is filled with epidural fat. The spinal cord extends from the caudal limit of the brainstem at the foramen magnum to the caudal lumbar vertebrae, terminating in the sixth lumbar vertebra (L6) in most dogs, and in L7 in cats, with some variations. The spinal cord is divided into segments: • Cervical C1–C8. • Thoracic T1–T13. • Lumbar L1–L7. • Sacral S1–S3. • Caudal (variable number). The spinal cord is wider at the cervical and lumbar intumescences (segments C6–T2 and L4–S3 respectively), from which the lower motor neurons (LMN) to the thoracic and pelvic limbs arise. These segments contain the cell bodies for the LMNs, also known as ventral horn cells, thus the spinal cord is thicker in these areas. The spinal cord is composed of central gray matter and peripheral white matter (1.1). A dorsal sulcus and ventral fissure, lined by pia mater, divide the spinal cord into two halves. Dorsal and ventral nerve roots exit the spinal cord at each segment and join to form the segmental spinal nerves. There are eight cervical segments, but seven cervical vertebrae. The C1 spinal nerves leave through the lateral foramina in C1 vertebra. 1.1 Spinal cord in transverse section. The gray matter forms an H shape, with two dorsal horns (a) and two ventral horns (b). The white matter tracts are divided into dorsal funiculi, between the dorsal roots (c); lateral funiculi, between dorsal and ventral roots (d); and ventral funiculi, between the ventral roots (e).
2
Small Animal Spinal Disorders
1.2 Position of the spinal cord segments in the cervical and cranial thoracic vertebrae. The cervical intumescence (C6–T2) lies within vertebrae C4–T2. Thus, lesions as far cranial as C4/C5 vertebrae may cause LMN signs in the thoracic limbs.
4 5 6
7
A
B
1.3 Normal cervical MRI. A: Sagittal, T2-weighted MRI of the cervical spine in a 7-year-old Doberman. It had undergone a ventral slot at C6/7 2 years ago and is now clinically normal. The other discs show the expected high signal intensity of the nucleus pulposus (arrow). B: Transverse, T2-weighted MRI of the cervical spine through mid-C5 vertebra; the gray matter is visible clearly within the spinal cord, which is surrounded by a hyperintense rim of cerebrospinal fluid (arrow). The black crescent on one side between CSF and epidural fat is a chemical shift artifact.
C5 T2
C6
T1
C8
C7
The rest of the cervical spinal nerves exit the vertebral canal cranial to the vertebrae of the same annotation, except C8 nerves, which exit between C7 and T1 vertebrae. The thoracic and lumbar spinal nerves exit behind the same-named vertebrae. The nerve roots are partly ensheathed by meninges, which are continuous with the epineurium.
Relationship of spinal cord segments to vertebrae
f
e
a d
b
c
1.4 Brachial plexus: lateral thoracic (a); ulnar (b); median (c); radial (d); axillary (e); and musculocutaneous nerves (f). The subscapular and suprascapular nerves arise just cranial to the musculocutaneous nerve.
Some spinal cord segments lie in the vertebra of the same annotation, but others do not (1.2, 1.5). Neurological lesion localization refers to spinal cord segments. It is therefore important to understand the relationship between vertebrae and spinal cord segments. Although regional nerves of the brachial and lumbosacral plexuses are relatively constant in their distribution they can vary considerably in their origins. The plexuses arise from C6 to T1 and from L5 to S3 spinal cord segment in the majority of animals (1.4, 1.8A). However, the brachial plexus arises from C5 to T1 in about 20% of dogs and from C6 to T2 in another 20% (Evans, 1993). Similarly the lumbosacral plexus arises from L3 to S1 in 20% and from L6 to S3 in 20% (Fletcher, 1970). This can have some impact on the neurological localization within these regions.
Functional anatomy
1.5 Position of spinal cord segments within the lumbar vertebrae. Segments L1 and L2 lie in their respective vertebrae. The lumbar intumescence lies within vertebrae L3–L5; lesions as far cranial as L3/4 intervertebral disc may cause LMN signs in the pelvic limbs. The sacral segments S1–S3 are within L5 vertebra in most dogs (the ‘5’ in L5 resembles the ‘S’ for sacral). The spinal cord ends in L6 in most dogs; L7 in cats. The cauda equina runs from L5 vertebra into the sacrum (1.7, 1.8B).
A
B
1.6 Normal thoracolumbar MRI. A: Sagittal, T2-weighted MRI of the thoracolumbar spine of a normal 6-year-old Golden retriever. Nucleus pulposus in this pulse sequence shows intermediate signal intensity; the thin layer of cerebrospinal fluid (CSF) is of intermediate signal intensity (arrow); epidural fat is high signal intensity (arrowhead). Note the nutrient vessels arising from the aorta (arrowheads) (Parker, 1973). B: Transverse, T1-weighted MRI of the thoracic spine at the level of T13. The spinal cord is surrounded by high signal epidural fat (arrow); CSF is of low signal intensity and does not show clearly. Note the aorta (arrowhead) (1.23). a
h
f
b
g
c
d
e
1.7 Nerve roots of the cauda equina. Conus medullaris (a). Nerve roots and rootlets (b). Intervertebral foramen (c). Spinal ganglion (d). Spinal nerve (e). L6 vertebra (f). L7 vertebra (g). Sacrum (h).
Cauda equina
Meninges
The nerves of the cauda equina have a typical peripheral nerve structure and are partly ensheathed by the meninges (1.7). They tolerate deformation better than spinal cord, and there is also a large epidural space in the region of the cauda equina (1.8B, 1.9). Thus they are usually more resistant to injury than spinal cord tissue, but if severe damage occurs, recovery is unlikely (13.22).
The meninges (1.10) surround the central nervous system (CNS). The arachnoid mater and pia mater together are termed the leptomeninges. Between the pia mater and the arachnoid mater is a space, the subarachnoid space, which is filled with cerebrospinal fluid (CSF). The subarachnoid space is traversed by the arachnoid trabeculae, which suspend the spinal
3
4
Small Animal Spinal Disorders
L5
L6
L5
L7
L6 a
b
1.8 A: Lumbosacral plexus: femoral (a); obturator (b); sciatic (c); and pudendal nerves (d). B: Dorsal, T1-weighted MRI of a normal 6-year-old German short-haired pointer to show the nerve roots of the cauda equina surrounded by high signal epidural fat (same dog as 1.9).
L7 d c
A
B 1.9 Normal lumbosacral MRI. A: Sagittal, T2-weighted MRI of the lumbosacral spine in extension (same dog as 1.8). There is loss of signal of the L7/S1 nucleus pulposus (arrowhead) but there is no dorsal displacement of the disc. B: Transverse, T1-weighted MRI through the L7/S1 disc space and foramen of the same dog; the nerve roots (arrowheads) are surrounded by high-signal epidural fat. See also 1.19B.
A
B
1.11 Section of spinal cord to show the subdural space (between dura and arachnoid mater, 1.10) filled with Indian ink. White denticulate ligaments (arrows) anchor the pia and spinal cord to the dura; they also restrict ventral extension of the ink within the subdural space (from Penderis et al., 1999). 1.10 There are three layers of meninges. The most superficial is the dura mater, which is composed of dense connective tissue (blue). The thin arachnoid mater (red) is inside the dura mater and lies adjacent to it. These two membranes follow the larger contours of the spinal cord. The pia mater (green) is a delicate layer that lies directly on the surface of the spinal cord.
cord within the CSF. The denticulate ligaments also traverse the subarachnoid space to anchor the spinal cord (1.11). The meninges caudal to the conus medullaris form the filum terminale, which extends into the
sacrocaudal vertebrae. The meningeal sac is outlined by myelography or MRI. The caudal limit to its extension varies between animals; it can terminate anywhere between L7 and the caudal vertebrae, but usually ends in the sacrum (see Chapter 10).
Cerebrospinal fluid Cerebrospinal fluid is formed in the brain, mainly by the choroid plexuses, with contributions from the
Functional anatomy
Sensory to brain
1.12 The reflex arc of the LMN is influenced by the UMN systems.
Upper motor neuron
Spinal ganglion
Sensory fiber
Lower motor neuron
leptomeninges (pia-arachnoid mater) and the ependymal lining. Cerebrospinal fluid flows mainly in a caudal direction. Most leaves the fourth ventricle of the brain through the lateral apertures into the subarachnoid space, with some entering the central canal of the spinal cord. It is absorbed through the arachnoid villi in cerebral venous sinuses as well as by venules in the subarachnoid space, lymphatics around the spinal nerves, and the ependymal lining. The CSF is normally a clear, colorless fluid with a very low protein and cellular content. It suspends and protects the brain and spinal cord against shock, allows some variation in the volume of the CNS without altering pressure, and has some nutritional and metabolic functions. The caudal direction of flow of CSF has some clinical relevance (14.3A). Cerebrospinal fluid collected caudal to a lesion is also more likely to provide diagnostic information (see Chapter 4).
Spinal cord white matter tracts ASCENDING SENSORY TRACTS Sensory information is gathered from the peripheral nervous system via sensory axons. The cell bodies of these axons lie in the spinal ganglia (dorsal root ganglia). Central projections of these axons ascend in the spinal cord to the brain (1.12). Proprioception is transmitted in the tracts of the dorsal and lateral funiculi. Axons project to either the somesthetic cerebral cortex or to the cerebellum. Temperature and superficial pain sensation are transmitted by the myelinated fibers of several tracts, including the lateral spinothalamic in the lateral funiculus. Severe pain sensation (deep pain or nociception) is
carried by non-myelinated fibers, particularly in the propriospinal and spinoreticular tracts. These tracts lie close to the junction of gray and white matter and therefore a lesion has to be extensive in order to damage all deep pain fibers at a given level. Pain fibers cross and re-cross the midline in a multisynaptic arrangement throughout the spinal cord, providing a diffuse bilateral pattern of ascending pain fibers from each limb. Information on the degree of urinary bladder filling is transmitted to the brain in the spinothalamic tract.
DESCENDING MOTOR TRACTS Two systems are responsible for the transmission of motor function—the upper motor neuron (UMN) and LMN systems (1.12, 1.13). The LMN is the effector neuron of the reflex arc. The cell bodies are the ventral horn cells, which lie in the ventral gray matter of the spinal cord. The axons leave the spinal cord in the ventral roots and pass through the brachial and lumbosacral plexuses to form the peripheral nerve trunks of the limbs. The sensory arm of the reflex arc is the sensory neuron. It arises in the periphery and enters the spinal cord via the dorsal root. It projects to the LMN (via an interneuron in some reflex pathways) and a branch also ascends in the spinal cord. Function of flexor muscles is facilitated by the corticospinal and rubrospinal tracts. The fibers of the corticospinal tract arise in the cerebral cortex, and most decussate at the spinomedullary junction and descend in the lateral corticospinal tracts of the lateral funiculi. Fibers that do not decussate descend in the ventral corticospinal tracts, which lie in the ventral funiculi. The rubrospinal fibers originate in the red nucleus of the brainstem, cross the midline and descend in the
5
6
Small Animal Spinal Disorders
1.13 The UMN system is the sum effect of the various individual descending pathways. The UMN system in general moderates LMN activity, initiates voluntary movement and maintains normal muscle tone.
UMN
LMN LMN
rubrospinal tract of the lateral funiculus. The vestibulospinal tracts and reticulospinal tracts also influence motor function. The function of extensor muscles is facilitated by these tracts, which lie in the ventral funiculi. The vestibulospinal fibers arise in the ipsilateral vestibular nuclei. They facilitate extensors and inhibit flexors on the ipsilateral side, and have the opposite effect on the muscles of the contralateral limbs. Voluntary bladder emptying is mediated through fibers in the tectospinal and reticulospinal tracts of the ventral funiculi.
ASCENDING MOTOR TRACT In dogs, an ascending motor tract originates in the border cells of the dorsolateral gray matter of the L1–L7 spinal cord segments. Their axons pass cranially in the fasciculus proprius of the lateral funiculus to inhibit the extensor muscles of the thoracic limbs. Interference with this pathway, as seen in some peracute, severe thoracic spinal cord lesions, is manifest as the Schiff– Sherrington sign (2.17).
Spinal cord nerve fibers and the effect of compression The white matter tracts of the spinal cord are composed of nerve fibers of different sizes, most of which have a myelin sheath. The largest fibers are myelinated, which are the most rapidly conducting; they transmit proprioception. Motor fibers are intermediate-sized myelinated fibers. Pain perception is transmitted by the smallest myelinated fibers and by non-myelinated fibers. Larger diameter fibers are more susceptible to injury than fibers of lesser diameter; small fibers are the most resistant. The progression of clinical signs seen with increasing spinal cord damage is explained largely by this feature. Mild lesions cause loss of proprioception. Increasingly severe lesions cause loss of the ability to
bear weight, loss of voluntary movement and, finally, loss of deep pain sensation. The position of the spinal cord tracts also contributes to the progression of signs. The ascending proprioceptive tracts lie superficially in the spinal cord and, therefore, are most susceptible to compression. In contrast, the spinothalamic tracts and ascending propriospinal pathways, which carry pain perception, are more deeply positioned, and the fibers cross the spinal cord at various levels. Thus a lesion must involve most of the diameter of the spinal cord for the patient to lose deep pain sensation (6.2) (Olby et al., 2003). This point and the fact that pain fibers are the most resistant to pressure explain why loss of deep pain sensation is such a severe clinical sign (see ‘Assessing the severity of the lesion’, page 31). Animals with lesions severe enough to damage transmission of deep pain along the cervical spinal cord either do not survive their injury or are at high risk of death from hypoventilation because of loss of respiratory muscle function (see Chapter 2).
SKELETON The vertebral column is composed of a series of vertebrae, most of which are joined by the intervertebral discs and by synovial joints between the articular processes.
Vertebrae The numbers of different types of vertebrae are as follows: cervical 7; thoracic 13; lumbar 7; sacral 3; and caudal 20 (approximately). Variations are possible, particularly in the transition zones between thoracic to lumbar, and lumbar to sacral vertebrae. The most common variations are in the number of ribs (8.19, 8.21), and abnormal articulations with the ilium (see Chapter 10). The importance of this lies in recognizing
Functional anatomy
g
b
f
e d
1.14 The atlas C1 (a) articulates with the skull (b) via the atlanto-occipital joints. C1 has prominent transverse processes (c), which can be palpated easily. There is no spinous process on the dorsal arch (d). There are lateral vertebral foramina in the vertebral arch (e), through which pass the C1 spinal nerves. The axis or C2 has a large spinous process (f), which extends cranially over the atlas and is connected to it by the dorsal atlantoaxial ligament (g). The atlas and axis are connected by multiple ligaments (1.35) and by synovial joints between the articular processes (h), which lie ventral to the vertebral canal.
a
c h
1.15 CT scans of a normal dog through A: the mid-body of the atlas C1: note the transverse foramina and the dens on the floor of the vertebral canal (9.1), and B: the cranial portion of the axis C2.
A
B
surgical landmarks (8.19–8.21). The total number of thoracic and lumbar vertebrae is generally 20. The vertebrae have various common features, but there are differences between the groups. Each vertebra has a vertebral body, which lies ventral to the spinal cord, and is joined to its neighbors by intervertebral discs. In immature animals, the vertebral bodies have cranial and caudal growth plates, which close by about 11 months of age in dogs (Hare, 1961). The center of the vertebral body is composed of cancellous bone, which is red and relatively soft. The margins of the vertebral body are made of hard, dense, white cortical bone, which also forms the vertebral end plates adjacent to the intervertebral discs. The types of bone provide an important guide to the depth of penetration in surgical procedures (8.32, 8.33, 10.30). Each vertebra has a vertebral arch, which forms the dorsal and lateral parts of the vertebral canal enclosing the spinal cord. The arch is made up of the pedicles
lateral to the vertebral canal and of the lamina dorsally. The vertebral arch also has cortical and cancellous bone, although the cancellous bone may be thin in small dogs and cats. Most vertebrae have transverse processes projecting laterally from the vertebral body, a spinous process projecting dorsally from the lamina, and cranial and caudal articular processes on the vertebral arch. Mammillary processes of the thoracic vertebrae are short, knob-like dorsal projections from the transverse processes. On lumbar vertebrae they are dorsal projections of the cranial articular processes (Evans, 1993). Other bony processes vary with the group of vertebrae. Between each pair of vertebrae there is an intervertebral foramen, through which pass the spinal nerves and blood vessels.
CERVICAL VERTEBRAE There are seven cervical vertebrae. The first two are distinct: the atlas (C1) and axis (C2) (1.14, 1.15). The vertebral body of C1 is very small, the bulk of the vertebra
7
8
Small Animal Spinal Disorders
being composed of lateral masses. Caudally on the body of C1, there are two articular processes, which articulate with C2. There is no intervertebral disc between C1 and C2. A prominent ventral tubercle lies on the caudoventral aspect of C1, just cranial to the intervertebral space, which can be a useful landmark at surgery. There are transverse foramina in the transverse processes, through which pass the vertebral arteries (1.36). The dens projects cranially from the vertebral body of C2 into the atlas, lying on the floor of the vertebral canal (9.1). The dens originates embryologically as part of C1. It has a growth plate at its attachment to the body of C2, which can separate after trauma. The atlas has three ossification centers at birth; the axis has four at birth and another two or three that develop after birth (Hare, 1961; Cook and Oliver, 1981). The other cervical vertebrae each have a similar morphology (1.16, 1.17). The large transverse processes of
C6 are particularly large and project ventrally; they are important surgical landmarks. The C5/6 intervertebral disc lies between the cranial edges of these transverse processes (4.6). A vascular channel runs through the center of these vertebrae and can give rise to hemorrhage and other complications (1.18, 1.19).
THORACIC AND LUMBAR VERTEBRAE The 13 thoracic vertebrae articulate with the ribs (1.20–1.22A). The spinous processes of the cranial thoracic vertebrae (to T10) slant caudally; those of the last two thoracic vertebrae slant cranially. The site of this change in direction (which may vary) is termed the anticlinal vertebra (1.20B). The articular processes from T1 to T10 sit at the base of the spinous process. They are in the same oblique horizontal plane as in the cervical vertebrae, with the caudal process of the cranial-most
a d
c b 1.16 Cervical vertebrae. The spinous processes are small (a), and the transverse processes (b) project laterally (with the exception of C6, where they are directed ventrally (4.6, 7.37B). There are transverse foramina (c), through which pass the vertebral arteries (1.17, 1.36), except in C7. The articular processes (d) lie in an oblique dorsal plane.
A
B
1.18 Photograph of the floor of the vertebral canal of C4, C5 and C6 with the lamina and pedicles removed. Note the large vascular foramen in the center of each vertebra (arrows). This can result in significant hemorrhage when drilling into bone and can impact screw placement (11.16).
C
1.17 CT scans through the mid-bodies of A: C5; B: C6; and C: C7 vertebrae. Note the transverse foramen in C5 and C6. The transverse processes of C7 have no foramina and have a characteristic horizontal orientation.
Functional anatomy
1.19 Nutrient foramen (arrows) in the center of lumbar vertebrae. A: Transverse CT scan of L4 vertebra. B: Transverse, T1-weighted MRI of L7 vertebra after fat suppression (same dog as 1.8). Bleeding from this structure is a common cause of severe hemorrhage when drilling into a vertebra; it can be stopped by direct pressure or with bone wax (1.6A, 1.18, 1.38, 8.2B).
A
B
a
b c
A b a
c
B 1.20 A: Thoracic vertebrae. The vertebral bodies are small with large spinous processes (a). The articular processes vary with the location along the thoracic spine (b). The transverse processes are short and have a fovea (c) that articulates with a rib. Caudal to the mid-thoracic area, caudally projecting accessory processes are present on the pedicle (13.55A). B: Radiograph of the anticlinal region. Spinous processes of cranial thoracic vertebrae (a); caudal thoracic and lumbar vertebrae (b); and T11, the anticlinal vertebra (c). This can be a useful surgical landmark, but the position must be ascertained in each animal from the radiographs.
vertebra overlying the cranial process of the caudalmost vertebra (1.20A). In the caudal thoracic vertebra, the articular processes adopt a more vertical orientation, as between the lumbar vertebrae (13.55). The change in orientation occurs near the anticlinal vertebra. The location of the vertebral canal relative to external vertebra landmarks at the thoracolumbar junction is variable, due largely to the orientation of the ribs (1.21, 1.22). It appears fairly constant within each breed of dog and also in the cat. The aorta and vena cava are in close proximity to many of the thoracic and lumbar vertebrae. Damage could be caused to these vascular structures when drilling or placing implants into vertebral bodies (Garcia et al., 1994). The relationship of the aorta to T13 vertebra is shown in 1.6B and 1.23A; to L1 in 1.23B and 4.44B; and to L2 in 4.43. There are seven lumbar vertebrae (1.24–1.28A). The L1 transverse processes are usually small, an important feature in identifying the thoracolumbar junction at surgery (8.19, 8.20). They are also difficult to palpate as they are obscured by the last rib. More caudally, the transverse processes are narrower and longer. The three sacral vertebrae are fused into one body, which articulates with the pelvis via the sacroiliac joint (1.28B). There is a marked notch in the cranial lamina of the sacrum, such that the lamina is not complete over the cauda equina at the lumbosacral junction (10.48). There are two pairs of dorsal and ventral sacral foramina, through which pass spinal nerves and blood vessels. The sacrum articulates caudally with the first caudal vertebra (1.26).
Articulations SYNOVIAL ARTICULATIONS The articular processes of the vertebral bodies have dorsal articulations (except for C1/2 where the articulations
9
10
Small Animal Spinal Disorders
A
B
1.21 CT scans through the mid-bodies of A: T11, T12, T13 and L1 vertebrae in a Dachshund (same dog as 8.60, 8.61B); and B: T12, T13, L1 and L2 vertebrae in a Springer spaniel after myelography. There is accumulation of contrast within the center of the spinal cord at T12, as well as leakage through a presumed dural tear (arrow) after trauma (13.10). The rib articulates progressively higher up the pedicle from T11 vertebra to T13. 1.22 CT scans through the mid-bodies of A: T12 in a cat; and B: L2 in a cat. T12 vertebra of a cat is shown in 1.23A. Compare these images with the dogs shown in 1.21.
A
B 1.23 The proximity of the aorta (arrowheads) is shown relative to A: T12–T13 intervertebral space in a cat (shown on a CT scan); and B: L1 vertebra in a dog (shown on a T1-weighted MRI). The vena cava is also evident (arrow). The cava lies to the right and somewhat ventral to the aorta.
A
B
Functional anatomy
d a
a
c d
e
c
b
b
1.24 Lumbar vertebrae. The vertebral bodies are long (particularly in cats). The spinous processes are short and blunt and slant cranially (a). The transverse processes project laterally and cranially (b). The articular processes are vertically oriented (c); the caudal process of the cranial-most vertebra lies medial to the cranial process of the caudal-most vertebra. There are accessory processes on the pedicles (13.55C), which project caudally (d).
1.26 Lumbosacral vertebrae. L7 differs from the other lumbar vertebrae in that the spinous process is shorter (a). The intervertebral foramen (b) lies cranial to the lumbosacral disc (1.27). The sacrum comprises the three fused sacral vertebrae (c), the spinous processes of which are fused and form a continuous ridge of bone (d). The lateral wing of the sacrum (e) articulates laterally with the wing of the ilium.
Sacrum
L7
A
1.27 Sagittal MRI of a dog to show that the intervertebral foramen of L7 (arrows) extends much more cranially than it does in other vertebrae (L6/7 and L7/S1 disc spaces shown by arrowheads). This foramen is thought to act as a stress riser for L7 fractures (1.26, 13.14, 13.21).
are ventral, and between the fused sacral vertebrae). These joints have a capsule, articular cartilage and synovial fluid.
INTERVERTEBRAL DISCS
B 1.25 CT scans through the mid-bodies of A: L6; and B: L7 vertebrae. Note that the transverse process is located a little more dorsally on L7 relative to the vertebral canal than on L6 (13.56B).
The vertebral bodies are joined by intervertebral discs (1.3A, 1.6A, 1.9A, 1.29), with the exception of C1/2 along with the fused sacral vertebrae. The intervertebral discs provide flexibility to the vertebral column, and act as shock absorbers for the spine. The capacity to absorb shock is diminished by age and degenerative changes. Intervertebral discs have a poor blood supply; nutrients gain access by diffusion. The anulus fibrosus is supplied with pain fibers, mainly in the outer laminae (Forsythe and Ghoshal, 1984).
11
12
Small Animal Spinal Disorders
A
B
1.28 The L7 nerve root runs in a lateral recess in the floor of the vertebral canal of L7 before emerging from the intervertebral foramen. A: The lateral recesses can be seen clearly (arrows) in this CT scan taken through the L7 vertebra. B: CT scan taken through the sacrum and sacroiliac joints.
a
b 1.29 The divisions of the intervertebral disc are the outer anulus fibrosus (a) and the inner nucleus pulposus (b). The anulus fibrosus is made up of concentric fibrous laminae. The laminae may not form complete rings, but they are interconnected with the adjacent laminae, thus forming a strong complete structure. The anulus fibrosus is thicker ventrally and laterally than it is dorsally and the nucleus pulposus is therefore eccentrically positioned (1.9A). The anulus fibrosus is attached firmly to the vertebral end plates by deeply penetrating (Sharpey’s) fibers. The nucleus pulposus is a gelatinous structure in young dogs, but its characteristics change with age (see below).
Degeneration of the intervertebral discs occurs with age and may precede disc herniation (Hansen, 1952). In humans there is a general loss of water content (Jenkins et al., 1985; Hickey et al., 1986) and of proteoglycans with age (Taylor et al., 2000). Similar changes have also been reported in dogs (Gysling, 1985; Cole et al., 1986; Bray and Burbidge, 1998). The two main types of disc degeneration in dogs are chondroid and fibroid metamorphosis: • Chondroid metamorphosis occurs in chondrodystrophoid breeds during the first 2 years of life. As the disc degenerates, it dehydrates and at the same time the nucleus pulposus is invaded by hyaline cartilage. These two processes interfere with the shock absorbing capacity of the disc by reducing the hydrostatic properties of the nucleus pulposus, and by weakening the fibers of the anulus fibrosus. In most Dachshunds, by 2 years of age the majority of discs have undergone chondroid metamorphosis, and many nuclei have also mineralized, changing from the former jelly-like consistency into a dry, gritty substance. Normal wear and tear often causes severe weakening of the intervertebral discs, especially at the thoracolumbar junction. This explains why the peak incidence of disc disease is between 3 and 6 years of age for most of the chondrodystrophoid breeds of dog. Herniation of this type of disc is often termed Hansen type I or a disc extrusion (1.30, 1.33). • Fibroid metamorphosis occurs in nonchondrodystrophoid breeds after middle age. The nucleus pulposus dehydrates but is invaded by fibrocartilage rather than hyaline cartilage. This process has a much later onset than in chondroid metamorphosis, and the discs are usually quite normal while the dog is young and active. The nucleus pulposus does not undergo mineralization as frequently as in discs that undergo chondroid metaplasia. Clinical problems occur primarily in older dogs and this type of herniation is often termed Hansen type II disease (1.31). Hansen type II lesions may or may not be responsible for clinical signs (see below).
DEFINITION OF DISC LESIONS Although the Hansen classification system is useful as a general guide, there is considerable overlap between type I and type II disc lesions. Each type of lesion can also occur in both chondrodystrophoid and nonchondrodystrophoid breeds (Cudia and Duval, 1997). Furthermore, there is a spectrum of chondrodystrophoid breeds ranging from extreme like the Shih Tzu,
Functional anatomy
1.30 Hansen type I disc extrusion is most common in chondrodystrophoid breeds. Following chondroid metamorphosis, the nucleus pulposus extrudes into the vertebral canal through the damaged anulus fibrosus. The nucleus may take a tortuous route through the anular fibers or may explode through a large defect (1.33).
et al., 1995). The classification scheme used most commonly for CT and MRI employs the terms normal, bulge, protrusion, or extrusion. Extrusion is a fairly distinct category used to describe disc material that has clearly escaped the normal boundaries of the disc, often to cause injury to neural structures. Disc extrusions are rare in asymptomatic humans, occurring in only 2 of 98 people in one study (BrantZawadzki et al., 1995). However the terms normal, bulge or protrusion include discs with identical degrees of disruption and associated symptoms in humans. Furthermore, there is only moderate interobserver agreement when applying these three terms to MRIs of disc lesions (Brant-Zawadzki et al., 1995; Milette et al., 1999). A purely morphological description of the displacement of disc material is therefore preferred to use of the terms normal, bulge and protrusion. An example would be ‘dorsolateral displacement of the L7/S1 disc to the right causing impingement on the L7 nerve root’ (Gorman and Hodak, 1997; Milette et al., 1999). Some intervertebral discs may have an internal derangement without obvious change in the contour of the disc. These changes can nevertheless be an important cause of clinical signs in humans and can only be diagnosed by using either MRI or discography (Milette et al., 1999) (see Chapter 4).
LIGAMENTS
1.31 Hansen type II disc herniation. This occurs mainly following fibroid metamorphosis, in the non-chondrodystrophoid breeds. The anulus fibrosus is damaged and there is displacement of the intervertebral disc into the vertebral canal.
classical like the Beagle, to less obvious like the Basset hound. An additional confounding factor is the high incidence of small- and medium-sized disc lesions in clinically normal animals. There is also a lack of a standardized nomenclature with which to describe disc abnormalities and this is a serious hindrance to proper communication and particularly to the accurate use of MRI (see Chapters 4 and 10) (Brant-Zawadzki et al., 1995; Milette et al., 1999). The term herniation is poorly defined and is therefore misused frequently. It is probably best reserved as a general term denoting a non-specific type of disc abnormality and should not be used to indicate clinical significance (Brant-Zawadzki
The nuchal ligament extends from the dorsal arch of the axis to the spinous processes of the cranial thoracic vertebrae (11.47, 11.48). This ligament lies deep in the dorsal cervical musculature. It is a large structure but can be sectioned at surgery if necessary without impairing head and neck support. The supraspinous ligament (1.32) continues along the tips of the spinal processes. The lumbodorsal fascia blends with this structure in the thoracolumbar region. The interspinous ligament (1.32) is a fascial sheet that is found between the spinous processes and is continuous with the lumbodorsal fascia in the lumbosacral region. The ligaments inside and outside the vertebral canal have a significant role in spinal stability and mobility (1.32–1.35). The ligamentum flavum (10.27) is found in the roof of the vertebral canal and in the space between adjacent vertebral laminae. It is continuous with the joint capsules of the articular processes, and may be significantly thickened in some disease conditions, particularly lumbosacral disease and cervical spondylomyelopathy (CSM). The fibers of the dorsal longitudinal ligament (1.33, 1.34) merge with those of the anulus fibrosus of the intervertebral disc; both carry pain fibers (Forsythe and Ghoshal, 1984). The presence of the intercapital ligament (1.34)
13
14
Small Animal Spinal Disorders
b
c
f
d
a b
e a
c
1.32 The cut-away portion reveals the weak ventral longitudinal ligament (a) running along the ventral surface of the vertebral bodies and the much more substantial dorsal longitudinal ligament (b) on the floor of the vertebral canal (1.33, 1.34). Also shown are the supraspinous ligament (c); interspinous ligament (d); intertransverse ligaments (e); and joint capsule covering the articular processes (f).
d
1.34 The dorsal longitudinal ligament (a) lies on the floor of the vertebral canal. The ligament is compact and narrow over the vertebral bodies (b). It diverges and is thus thinner over the intervertebral disc (c). Between the heads of the ribs (except T1, T12 and T13) there is a reinforcing intercapital ligament (d), which lies under the dorsal longitudinal ligament. 1.33 Dorsal view of the floor of the vertebral canal with the spinal cord removed to show a type I disc extrusion. The calcified nuclear material lies at the level of the intervertebral foramen, to one side of the dorsal longitudinal ligament. Note how this ligament fans out over, and then merges with, the dorsal surface of the anulus fibrosus.
b
c
contributes to the low incidence of intervertebral disc extrusions in the thoracic spine between T2 and T11 (Wilkens et al., 1996).
a
BLOOD SUPPLY Vertebral column The arterial supply to the vertebral column is segmental, with a spinal branch entering the vertebral canal via the intervertebral foramen, closely associated with the spinal nerve. The origin of the branches varies between the regions of the spine (1.36–1.39) (Forsythe and Ghoshal, 1984). Venous drainage is via the internal vertebral venous plexus, which comprises two valveless veins on the floor of the vertebral canal (often termed the venous sinuses). The veins converge at mid-vertebral body (and
1.35 The most significant ligament between C1 and C2 is the transverse ligament of the atlas (a), which runs between the sides of the atlas and over the dens, holding it down on to the floor of the atlas. Less significant are the apical ligament of the dens (b) (dens to foramen magnum) and the alar ligaments (c) (dens to occipital bones). The dorsal atlantoaxial ligament runs between the dorsal arch of C1 and the spinous process of C2 (1.14). There are two ventral synovial articulations between C1 and C2.
Functional anatomy
i
g
e k l m
h
j d
c f b a
1.36 Blood supply to the cervical spine. The arterial supply to the cervical vertebrae is from the paired vertebral arteries (a), which run cranially from their origin on the subclavian arteries in the thorax. The arteries run through the transverse foramina (b) in the transverse processes of the vertebrae (except C7). At each segment, there are dorsal (c) and ventral (d) muscular branches. A significant vessel runs near the caudal edge of the articular processes (e). A spinal branch (f) enters the vertebral canal at each intervertebral foramen; these supply the spinal cord. The vertebral artery branches at the atlas. The dorsal branch (g) runs over the transverse process of C1, anastomoses with a branch of the occipital artery (h), and enters the vertebral canal through the lateral vertebral foramen of C1 (i). The ventral branch runs under the transverse process and also anastomoses with a branch of the occipital artery ( j). The internal vertebral venous plexus (venous sinus) (k) lies on the floor of the vertebral canal. The veins converge at midvertebral body (and sometimes join) (l), and then diverge again over the intervertebral disc (m). In the atlas and axis, the veins of the internal vertebral venous plexus are much more laterally positioned. Thus, in attempting to collect CSF from the cerebellomedullary cistern, the veins may be perforated if the needle strays from the midline. 1.37 The thoracic spine is supplied by spinal branches (a) from the intercostal arteries (b), which enter the vertebral canal via the intervertebral foramina. The internal vertebral venous plexus drains into the major veins of the dorsal thorax (c), mainly the azygous vein. a
b
c
15
16
Small Animal Spinal Disorders
d a
e
1.38 The lumbar spine is supplied by spinal branches (a) of the lumbar arteries (b), which arise from the aorta (c). Each lumbar artery also gives rise to a nutrient vessel that enters the vertebral body (Evans, 1993). A dorsal branch runs caudally behind the articular processes in the musculature (d). The lumbar internal vertebral venous plexus drains into major veins of the abdomen (e), mainly the azygous vein and the caudal vena cava.
b
c
d
b
c
h
f
a
g e
i
1.39 The blood supply to the spinal cord arises from the spinal arteries (a), which enter the vertebral canal through the intervertebral foramina. These branch into dorsal (b) and ventral (c) radicular arteries. These arteries supply an anastomotic network on the surface of the spinal cord, deep to the dura mater. There are paired dorsolateral spinal arteries (d), which run on the dorsal spinal cord. These vessels may be tortuous and not recognizable as a distinct entity. There is a ventral spinal artery (e), which runs in the ventral fissure. There are multiple anastomotic arteries connecting the main vessels. The segmental arteries are inconsistently present, such that several segments may be supplied by one spinal artery. The distribution is also not symmetrical. The spinal cord substance is supplied by various arteries that penetrate the surface. The vertical arteries (f) arise from the ventral spinal artery and pass dorsally through the ventral fissure. They supply most of the gray matter and some white matter. Radial arteries (g) pass centrally from the arteries on the cord surface, and enter the spinal cord substance. They supply the white matter and the peripheral gray matter. The areas of spinal cord supplied by each branch have been described (de Lahunta and Alexander, 1976). The venous drainage of the cord is also in a radial pattern, to a network of surface veins (h). These drain into the internal vertebral venous plexus on the floor of the vertebral canal (i). These are large, valve-free vessels that have occasional anastomoses in the midline. The plexus drains at the intervertebral foramina through the intervertebral veins. There are also veins draining the vertebral bodies into the plexus, and drainage through the vertebral bodies (1.19).
Functional anatomy
sometimes join) and diverge over the intervertebral disc (7.43). They are thin walled and easily damaged. The venous plexus drains at the intervertebral foramina via the intervertebral veins into the vertebral veins. The intervertebral veins may be single at each foramen, or may be paired, in which case they surround the spinal nerve. The intervertebral veins are fragile and can bleed profusely if damaged.
REFERENCES Brant-Zawadzki, M.N., Jensen, M.C., Obuchowski, N., Ross, J.S., Modic, M.T. (1995) Interobserver and intraobserver variability in interpretation of lumbar disc abnormalities. A comparison of two nomenclatures. Spine 20, 1257–1263; discussion 1264. Bray, J.P., Burbidge, H.M. (1998) The canine intervertebral disk. Part Two: degenerative changes—non-chondrodystrophoid versus chondrodystrophoid disks. Journal of the American Animal Hospital Association 34, 135–144. Cole, T.C., Ghosh, P., Taylor, T.K.F. (1986) Variations of the proteoglycans of the canine intervertebral disc with ageing. Biochimica et Biophysica Acta 880, 209–219. Cook, J.R., Oliver, J.E. (1981) Atlantoaxial luxation in the dog. Compendium on Continuing Education for the Practicing Veterinarian 3, 242–250. Cudia, S.P., Duval, J.M. (1997) Thoracolumbar intervertebral disk disease in large, nonchondrodystrophic dogs: a retrospective study. Journal of the American Animal Hospital Association 33, 456–460. de Lahunta, A., Alexander, J.W. (1976) Ischemic myelopathy secondary to presumed fibrocartilaginous embolism in nine dogs. Journal of the Americal Animal Hospital Association 12, 37–48. Evans, H.E. (1993) Miller’s Anatomy of the Dog, 3rd edn. Philadelphia: WB Saunders. Fletcher, T.F. (1970) Lumbosacral plexus and pelvic limb myotomes of the dog. American Journal of Veterinary Research 31, 35–41. Forsythe, W.B., Ghoshal, N.G. (1984) Innervation of the canine thoracolumbar vertebral column. The Anatomical Record 208, 57–63. Garcia, J.N.P., Milthorpe, B.K., Russell, D., Johnson, K.A. (1994) Biomechanical study of canine spinal fracture fixation using pins or bone screws with polymethylmethacrylate. Veterinary Surgery 23, 322–329. Gorman, W.F., Hodak, J.A. (1997) Herniated intervertebral disc without pain. Journal of the Oklahoma State Medical Association 90, 185–190. Gysling, C. (1985) Ageing process of intervertebral disks in the German Shepherd dog [Abstract of Dissertation, Zurich Univ.]. Schweizer Archiv fur Tierheilkunde 127, 53–54. Hansen, H.J. (1952) A pathologic–anatomical study on disc degeneration in dogs. Acta Orthopaedica Scandinavia Suppl. 11, 1–117.
Hare, W.C.D. (1961) Radiographic anatomy of the cervical region of the canine vertebral column. Part I: Fully developed vertebrae II: Developing vertebrae. Journal of the American Veterinary Medical Association 139, 209–220. Hickey, D.S., Aspden, R.M., Hukins, D.W., Jenkins, J.P., Isherwood, I. (1986) Analysis of magnetic resonance images from normal and degenerate lumbar intervertebral discs. Spine 11, 702–708. Jenkins, J.P., Hickey, D.S., Zhu, X.P., Machin, M., Isherwood, I. (1985) MR imaging of the intervertebral disc: a quantitative study. British Journal of Radiology 58, 705–709. Milette, P.C., Fontaine, S., Lepanto, L., Cardinal, E., Breton, G. (1999) Differentiating lumbar disc protrusions, disc bulges, and discs with normal contour but abnormal signal intensity. Magnetic resonance imaging with discographic correlations. Spine 24, 44–53. Olby, N.J., Harris, T., Munana, K.R., Skeen, T.M., Sharp, N.J.H. (2003) Long-term functional outcome of dogs with severe injuries of the thoracolumbar spinal cord: 87 cases (1996–2001). Journal of the American Veterinary Medical Association 222, 762–769. Parker, A.J. (1973) Distribution of spinal branches of the thoracolumbar segmental arteries in dogs. American Journal of Veterinary Research 34, 1351–1353. Penderis, J., Sullivan, M., Schwarz, T., Griffiths, I.R. (1999) Subdural injection of contrast medium as a complication of myelography. Journal of Small Animal Practice 40, 173–176. Taylor, T.K., Melrose, J., Burkhardt, D., Ghosh, P., Claes, L.E., Kettler, A., Wilke, H.J. (2000) Spinal biomechanics and aging are major determinants of the proteoglycan metabolism of intervertebral disc cells. Spine 25, 3014–3020. Wilkens, B.E., Selcer, R., Adams, W.H., Thomas, W.B. (1996) T9–T10 intervertebral disc herniation in three dogs. Veterinary and Comparative Orthopaedics and Traumatology 9, 177–178.
FURTHER READING Boyd, J.S., Patterson, C. (1991) A Colour Atlas of Clinical Anatomy of the Dog and Cat. London: Wolfe Publishing. Caulkins, S.E., Purinton, P.T., Oliver, J.E. (1989) Arterial supply to the spinal cord of dogs and cats. American Journal of Veterinary Research 50, 425–430. de Lahunta, A. (1983) Veterinary Neuroanatomy and Clinical Neurology. Philadelphia: WB Saunders. Jenkins, T.W. (1978) Functional Mammalian Neuroanatomy. Philadelphia: Lea & Febiger. King, A.S. (1987) Physiological and Clinical Anatomy of the Domestic Mammals, Vol. 1, Central nervous system. Oxford: Oxford University Press. Worthman, R.P. (1956) The longitudinal vertebral venous sinuses of the dog. American Journal of Veterinary Research 17, 341–363.
17
Patient examination
Approach to the patient History
Chapter
2
age should be considered, but again this information must be used with care.
19
19
HISTORY Physical examination
19
Neurological examination 20 Stage 1: Patient in upright position 20 Stage 2: Patient in lateral recumbency 25 Localization of lesions 28 Assessment of the brachial and lumbosacral plexus 30 Assessing the severity of the lesion Determining the etiology References
31
32
Taking a history and performing a full clinical examination are prerequisites to the neurological examination. The history often leads to a provisional diagnosis. Of particular note is evidence of trauma; whether the condition is progressive, static, or episodic; previous episodes of disease; signs of pain; vaccination status; travel history; and urinary function. In particular, does the animal let the owner know that it needs to urinate, can it void a stream of urine or does it just dribble urine without being aware of it? Owners sometimes say that their dog is incontinent because every time they pick it up it urinates, when in fact this just occurs because of pressure being placed on its abdomen.
32
Further reading
33
The clinical syndromes seen in animals with spinal disease are recognized generally from either the history or the physical findings. Spinal disease should also be considered in animals with non-specific pain, exercise intolerance or lameness not caused by orthopedic disease (McDonnell et al., 2001). This chapter discusses the approach to a patient in which spinal disease is suspected.
PHYSICAL EXAMINATION A general physical examination must be made in all patients. If there has been trauma or if anesthesia is contemplated, involvement of other body systems must be determined (Neer, 1992) (2.1). Also, some patients in which spinal disease is suspected have disorders of other
APPROACH TO THE PATIENT The aims of patient examination are as follows: • Determine whether the problem is spinal in origin. • Locate the site of the disorder. • Assess the severity of the neurological deficit. • Identify the disease process. • Determine the most appropriate form of treatment. • Predict the prognosis. Knowledge of the breed incidence of spinal diseases is useful in the initial assessment, but it is a mistake to use such information as the only basis for diagnosis. Similarly,
2.1 Damage to thoracic or abdominal viscera is common after trauma. Here bile duct rupture has given rise to extensive peritoneal contamination (see page 282).
20
Small Animal Spinal Disorders
systems. It is not at all unusual for orthopedic disorders to mimic spinal conditions; examples are given in Table 2.1. Careful clinical examination should identify such problems, and particular note should be made of joint pain or enlargement, as these signs are present in many dogs misdiagnosed as having neurological disorders. The presence of any spinal pain or deformity should also be noted. The quality of the femoral pulse must be determined, particularly in acutely paralysed cats although thromboembolism also occurs in dogs (Boswood et al., 2000) (see Chapter 14). It is straightforward to perform a screening neurological examination as part of a general physical examination (Table 2.2). If abnormalities are detected, carry out a more complete neurological examination.
NEUROLOGICAL EXAMINATION The neurological examination is carried out with the aims of determining the precise location of the spinal lesion and its severity. The neurological examination described here is performed readily with the animal upright in the first instance, and later placed in lateral recumbency.
It is useful to have a form on which to write the findings of the neurological examination (2.2). This insures that no aspect of the examination is missed, provides a permanent record, prevents errors and permits more accurate comparisons of any serial examinations. A video can also be made of the patient for this purpose.
Stage 1: Patient in upright position ASSESS ATTITUDE, POSTURE AND GAIT Watch the patient as it relaxes in the examination room. Let it move to the best of its ability, unless it has an acute spinal injury, when movement should be restricted. Note the degree of motor function, the gait (particularly noting any asymmetry) and the general demeanor. In cats, this part of the examination is particularly important as later parts may be difficult to perform. It is useful to listen to dogs as they walk on a hard surface; if proprioceptive deficits are present, the examiner may hear the claws scuff.
DETERMINE THE LOCOMOTOR STATUS The animal is encouraged to move, except where an acute spinal injury has occurred or if there is severe
Table 2.1 Disorders that can mimic spinal disease Systemic disorders
Uni- or Bilateral orthopedic disorders
Generalized orthopedic disorders
Neuromuscular disorders
Endocarditis Cardiac insufficiency Hypertension Upper airway disease Hyperkalemia Hypokalemia Hypocalcemia Hypoglycemia Hyperthyroidism Addison’s disease Pheochromocytoma
Osteochondritis dissecans Cranial cruciate ligament injury Tibial crest avulsion Fractures Coxofemoral osteoarthritis Patellar luxation Septic arthritis Biceps tendonitis Infraspinatus contracture Gracilis contracture Achilles tendon rupture Psoas muscle injury
Hypertrophic osteodystrophy Polyarthritis Panosteitis
Generalized myopathies Ischemic neuromyopathy Neuropathies Radiculopathies Junctionopathies
Table 2.2 Screening neurological examination
Observation Mental status Posture Gait
Postural reactions
Cranial nerves
Spinal reflexes
Spinal hyperesthesia
Bladder function
Paw position Hopping
Menace Pupillary light reflex Oculovestibular response Jaw tone; temporal muscle mass Facial sensation Palpebral reflex
Patellar Withdrawal Perineal Cutaneous trunci reflexes
Cervical Thoracolumbar Lumbosacral
Historical Leakage? Voiding?
Patient examination
SPINAL REFLEXES
HISTORY PHYSICAL EXAMINATION OBSERVATION
Reflex (Nerve) (Spinal cord segments) LEFT
RIGHT Triceps (Radial) (C7-T1) Biceps (Musculocutaneous) (C6-C8) Withdrawal (thoracic limb) (Multiple) (C6-T2) Patellar (Femoral) (L4-L6) Gastrocnemius (Tibial, sciatic) (L6-S1) Withdrawal (pelvic limb) (Sciatic) (L6-S1)
Mental status (e.g. alert, depressed, stupor, coma) Posture (e.g. normal, paraparesis, hemiparesis, head tilt, tremor) Gait (e.g. ataxia, circling)
POSTURAL REACTIONS LEFT
RIGHT Hopping Front
Rear Paw position Front Rear Reflex step Front Rear Tactile placing Front Rear Visual placing Front Rear
Perineal (S1-S2)
URINARY FUNCTION
Hemistanding
Voluntary urination?
Hemiwalking Wheelbarrowing
Bladder distention?
Extensor postural thrust Overflow / ease of manual expression
CRANIAL NERVES
SPINAL HYPERAESTHESIA
Test (Innervation) LEFT
RIGHT Menace response (II & VII) Vision (II)
SML
Pupil size Pupillary light reflex (II & III) Stimulate left eye Stimulate right eye Strabismus (III, IV & VI) Spontaneous nystagmus (III, IV, VI & VIII) Positional nystagmus (III, IV, VI & VIII) Oculovestibular response (III, IV, VI & VIII) Facial sensation (V) Jaw tone (V) Temporal muscle mass (V) Corneal reflex (V, VI & VII) Facial symmetry (VIII) Palpebral reflex (V & VII) Hearing (hand clap) (VIII) Swallowing or ‘gag reflex’ (IX & X) Tongue (XII)
EYES
PANNICULUS REFLEX Level of cut-off (dermatome) LEFT
SML
RIGHT
DEEP PAIN PERCEPTION Thoracic limb Pelvic limb Tail
LESION LOCALIZATION BRAIN
Side Forebrain Brainstem Cerebellum Vestibular - peripheral Vestibular - central Multifocal
SPINAL CORD C1 - C5 C6 - T2 T3 - L3 L4 - S3 MULTIFOCAL CNS PERIPHERAL NERVE Local Generalised
Horner’s Syndrome (Sympathetic)
NEUROMUSCULAR
Fundic examination
MUSCULAR
MUSCLE PALPATION
NORMAL
Tone Atrophy
KEY: ABSENT 0; REDUCED ⫹1; NORMAL ⫹2; INCREASED ⫹3; CLONUS ⫹4
2.2 Form for neurological examination. The DAMNIT scheme can be listed on the reverse side with room to add differential diagnoses and the diagnostic plan(s).
21
22
Small Animal Spinal Disorders
A 2.4 Proprioception: paw position response where an animal’s body-weight is supported fully and then each paw is turned over individually to bring the dorsal surface into contact with the ground. Normal animals return the paw to an upright position almost immediately; those with neurological disease cranial to the limb may leave the paw flexed. If proprioception is normal, spinal cord disease is unlikely to be present.
B 2.3 A: Hopping in the thoracic limb. The patient’s body is supported with only one limb bearing weight on the ground, and then it is moved laterally. The animal will not be able to hop normally if it has a proprioceptive deficit or impaired motor function. In a large dog this can be done by just lifting one thoracic limb and then moving the dog sideways. B: Hopping in the pelvic limb. Large dogs can be hopped as described for the thoracic limb.
pain. Patients that appear paraplegic at rest may show some voluntary movement if supported by a sling (15.9A) or if held by the base of their tails. Unilateral weaknesses and sensory deficits may be revealed by hopping (2.3), hemistanding and hemiwalking tests. As hopping also requires good motor function and intact proprioception, it is included in the section on proprioceptive testing below. Assess muscle strength, if the patient is able to stand, by pressing down on the shoulders and hips. Following these tests, the locomotor status can then be classified, for example paraparetic, hemiparetic and tetraparetic.
ASSESS PROPRIOCEPTION This is evaluated in the standing animal by the paw position test and the reflex step (2.4, 2.6). Animals with deficits of proprioception that can still walk often wear the dorsum of their digit(s) or claw(s).
2.5 Boxer dog with proprioceptive loss in the left pelvic limb. Severe deficits such as this are evident even if the animal’s weight is not supported properly.
Tactile placing can be tested as in 2.10 but with the animal’s eyes covered or, in small patients, with the animal’s body curved around the examiner’s abdomen so that its head is pointing away from the table. The thoracic and pelvic limbs on the side nearest to the table are then touched gently against the table edge. The animal should immediately lift up the foot; the contralateral limbs are then repositioned so that they are adjacent to the table and the process is repeated.
Patient examination
2.6 A, B: Proprioception can also be tested by the ‘reflex step’, where a piece of paper is placed under the foot and pulled laterally. The animal should return the foot briskly to a normal position; an abnormal response is to let the foot slide away from the body. This test probably assesses proprioception more in the proximal part of the limb than the test shown in 2.4. A
B
2.7 Wheelbarrowing test. The pelvic limbs are lifted off the ground as shown and the animal made to walk forward. This test can reveal thoracic limb paresis and exacerbates asymmetrical lesions. Elevating the head on this test will sometimes unmask hypermetria.
2.9 Extensor postural thrust. The patient is held up, as shown, and lowered to the surface. The normal animal will push away with the pelvic limbs and step backwards. This test is useful in revealing pelvic limb deficits.
2.8 Wheelbarrow testing in large dogs is particularly useful to identify subtle weakness in the thoracic limbs that might indicate a cervical lesion in a dog that at first sight appears to be only weak in its rear limbs.
PALPATE THE ABDOMEN Determine the degree of bladder filling, and the ease with which urine is expressed by palpating the abdomen. Urinary incontinence is often a feature of spinal disorders, and some assessment of urinary function should have been gained from the history.
2.10 Placing test. Here visual placing is being evaluated as the animal can see the table edge. Visual placing evaluates vision and motor function. Tactile placing evaluates sensation, proprioception and motor function.
23
24
Small Animal Spinal Disorders
If neurogenic urinary incontinence is present, it is important to determine if it is LMN or UMN in nature (Table 15.6): • The LMN bladder is typically large, flaccid, and easily expressed. This is usually associated with lesions of the sacral spinal cord segments or nerve roots. Some animals, particularly cats with sacrocaudal injuries, can nevertheless be difficult to express due to high sphincter tone, probably from unopposed sympathetic tone via the uninjured hypogastric nerve (see page 351). • The UMN bladder is characteristically tense and difficult to express unless grossly distended, because urethral sphincter tone is often increased. This is seen with lesions cranial to the sacral segments, usually those affecting the T3–L3 spinal cord region (see ‘Disorders of micturition’, page 351). The importance of determining the type of incontinence is two-fold. Appropriate drug therapy can be determined from this information (see ‘Pharmacological manipulation of micturition’, page 351). Also, the prognosis for recovery of urinary function is often worse for LMN incontinence compared to most UMN lesions.
intact on the contralateral side. Thoracolumbar lesions may interfere with the afferent part of the cutaneous trunci reflex, which will be intact only in dermatomes supplied by segments cranial to a spinal cord lesion and will be absent in the segments caudal to it. Lesions caudal to L1 spinal cord segment usually have a normal cutaneous trunci response as the L1 dermatome extends to the level of the tuber coxae (2.12).
PALPATE THE SPINE Determine the presence of spinal hyperesthesia by palpating the vertebral column and evaluating the patient’s response (2.13, 2.14). This is an important step in the examination. The degree of pressure to be
T13
L1 T13
a
T12 T11
THE CUTANEOUS TRUNCI REFLEXES This is tested by pinching the skin along the dorsal surface of the trunk with fine forceps and observing the twitch of the cutaneous trunci muscle on both the ipsilateral and, to a lesser extent, the contralateral side (2.11, 2.12). There is crossing of the pathway within the spinal cord, leading to a bilateral response after unilateral stimulation. In thoracic limb paresis, the cutaneous trunci reflex will also be absent on the affected side if the lesion involves the C8 and T1 spinal cord segments, roots or nerves (1.4 and page 30), but will be
2.11 The cutaneous trunci reflex is activated by pinching the skin over the lumbar spine with forceps or by a gentle needle prick, which leads to a twitch of the cutaneous trunci muscle.
2.12 The cutaneous trunci reflex. Stimulation of the skin of the back activates the reflex, the efferent arm of which is the lateral thoracic nerve arising from segments C8 and T1 and leaving the caudal part of the brachial plexus (a). The dermatomes are positioned some distance caudal to their respective vertebrae as illustrated here.
2.13 Examining the neck for evidence of pain. Palpating the neck muscles and moving the neck is usually adequate to reveal pain, as evidenced by an increase in tension (guarding) and by muscle fasciculations.
Patient examination
applied when determining the presence of pain varies between patients. In a dog with neck pain caused by a cervical disc extrusion, it is often adequate to palpate the cervical muscles gently. It is not usually necessary to put the head in extreme positions to reach this conclusion. Some animals do not show pain but guard against neck movement with an increase in resistance compared to normal. In some more stoic dogs, greater force may be required either by downward pressure on the spinous processes or along the transverse processes. The transverse processes of C6 (4.6) can also be manipulated percutaneously in most dogs to assess for low cervical pain. Spinal pain can arise as a result of discogenic pain (Morgan et al., 1993); dorsal longitudinal ligament damage; nerve root irritation; stretching or inflammation of the meninges; and bone pain. In humans, back pain can also arise from the sacroiliac and facet joints; these are likely to be potential causes of pain in animals as well but would be much harder to confirm than in humans (Pang et al., 1998) (see Chapter 14). In some animals with spinal disease, pain is the only clinical abnormality, neurological function being normal.
CRANIAL NERVE EXAMINATION Even though abnormal cranial nerve findings may not be expected in spinal disorders, some animals with multifocal neurological disease present predominantly with signs of spinal dysfunction (14.15). It is therefore important to assess the entire nervous system (de Lahunta, 2001; Braund and Sharp, 2003). In addition, particular note should be made of the presence of any component of Horner’s syndrome (ptosis, miosis, enophthalmos and third eyelid protrusion), which occurs with interference to the sympathetic supply to the eye. Note that some animals will only show miosis without the other components. The resultant anisocoria is missed easily unless tested specifically (2.15). The pupils should be assessed carefully for anisocoria by looking through an ophthalmoscope held about 60 cm from the animal’s face. This will produce a tapetal reflection that allows easy comparison of pupil sizes even in a well-lit room; it is particularly important in animals with a dark iris. Isolated miosis may be a feature of spinal disease if the cervical or cranial thoracic spinal cord segments, or the nerve roots of the brachial plexus, are involved. An ophthalmoscopic examination should also be carried out to look for fundic lesions indicative of inflammatory CNS disease.
Stage 2: Patient in lateral recumbency
2.14 Palpating the thoracolumbar spine for pain. For assessment of lumbosacral hyperesthesia see page 184.
The patient is then placed in lateral recumbency and each limb evaluated, with the aim of placing it into one of the following categories: • Normal. • LMN-type abnormality. • UMN-type abnormality. An understanding of functional anatomy is required to appreciate the difference between these types of deficit (1.12, 1.13). The effect of lesions on the LMN and UMN systems can be considered in terms of motor function, muscle atrophy, muscle tone and local reflexes. The clinical
2.15 A: Isolated miosis due to loss of sympathetic outflow after brachial plexus injury. B: Iatrogenic Horner’s syndrome in a cat that has just had inflammatory polyps removed from the middle ear. The ptosis and prolapsed third eyelid are visible although the pupil is somewhat obscured. Tumors of the brachial plexus may demonstrate several components of Horner’s syndrome but often only show miosis in the early stages.
A
B
25
26
Small Animal Spinal Disorders
Table 2.3 Differentiation of LMN vs UMN abnormalities Lower motor neuron (LMN)
Upper motor neuron (UMN)
Motor function
Paresis or paralysis
Paresis or paralysis
Reflexes
Absent or reduced
Normal or increased
Muscle tone
Reduced
Normal or increased
Muscle atrophy
Severe, early—neurogenic
Late, mild—disuse
2.16 Neurogenic atrophy. It is useful to evaluate muscles that have a definitive bony border, for example the spinatus muscles as shown, the cranial tibial muscle and the muscles over the pelvis.
2.17 Paraplegic dog with stiff, hyperextended thoracic limbs and digits; sometimes the neck is also extended—the Schiff–Sherrington sign. Thoracic spinal cord lesions may interfere with inhibitory neurons that have their cell bodies (the border cells) in the L1–L7 spinal cord segments (Braund and Sharp, 2003). Axons from these cells pass cranially to inhibit neurons supplying the thoracic limb extensor muscles.
MUSCLE MASS signs that allow differentiation between UMN and LMN abnormalities are summarized in Table 2.3. LMN deficits are characterized by paresis or paralysis; severe (neurogenic) muscle atrophy (2.16); reduced tone in the affected muscles; and depressed or absent reflexes. UMN deficits show paresis or paralysis; mild (disuse) muscle atrophy; normal or increased muscle tone; and normal or hyperactive reflexes. There is some variation in mild cases, but from the neurological examination, it should be possible to categorize each limb as being ‘normal’, ‘UMN-type abnormality’, or ‘LMN-type abnormality’.
Muscle atrophy is assessed by observing and palpating major muscle groups (2.16).
MUSCLE TONE Test muscle tone by gently flexing and extending the joints. Normally there is some resistance to such manipulation. An incorrect impression of increased tone may be gained in excitable or fractious animals, or if the animal has a painful orthopedic condition. Increased tone in the thoracic limbs is seen in the Schiff–Sherrington sign (2.17). This sign (see ‘Ascending motor tract’, page 6) does not necessarily denote a hopeless prognosis (13.7).
MOTOR FUNCTION
REFLEX TESTING
This has been evaluated previously. With the animal in lateral recumbency, the examination proceeds as follows.
There are a number of local reflexes available for examination, but it is usual to concentrate on the patellar,
Patient examination
2.20 In the pelvic limb critical assessment should be made of hock flexion to assess for evidence of LMN weakness (2.27). It may also be necessary to stimulate the withdrawal reflex with forceps placed over the nail bed or digit in order to evoke a behaviural response for deep pain evaluation (2.21). 2.18 The flexor or withdrawal reflex is stimulated by pinching the toe, which results in flexion of the limb. It is important to persist with the stimulus until it is clear that all the limb joints are flexing. Loss of elbow flexion is often the most sensitive indicator of a weak thoracic limb withdrawal reflex. While the flexor reflex is being evoked, the contralateral limb should also be observed for reflex extension—the crossed extensor reflex.
2.19 The patellar reflex is evoked by tapping the straight patellar ligament. It can be done on either the upper or the lower limb, in contrast to the withdrawal reflex that is unreliable in the lower limb.
flexor (withdrawal) and perineal reflexes (2.18–2.20). The perineal reflex can be elicited by squeezing the base of penis or the perineal region. Anal tone can be assessed by a rectal examination or using a thermometer. Other reflexes such as the triceps, biceps, and gastrocnemius may also be tested. However, they are found inconsistently in normal animals and their main significance is in finding hyperactive responses in UMN disorders. Increased patellar reflexes are often seen in UMN lesions cranial to the L4–L6 spinal cord segments. When this occurs there are sustained contractions known
as clonus, where the limb vibrates briefly after the first initial kick. It must be remembered, however, that many animals with UMN lesions have a normal patellar reflex. A different type of exaggerated patellar reflex can be seen in animals with LMN lesions involving the sciatic nerve or its origins. The hamstring muscle group innervated by the sciatic nerve normally counteracts or damps the reflex kick of the patellar reflex. Interference with sciatic nerve function can reduce this damping effect and lead to an increased reflex, so that the lower limb will oscillate after an initial brisk kick. This is termed ‘pseudohyperreflexia’ or an oscillating patellar reflex. In tense patients, the patellar reflex may not respond at all in the upper limb whereas it does in the lower limb; it is then important to test the reflexes with the dog on its opposite side. In most animals the reflex can also be elicited in the standing position if the limb is relaxed.
DEEP PAIN SENSATION OR NOCICEPTION On testing the withdrawal reflex, note the patient’s behaviural response to the stimulus (2.21). Formerly called deep pain sensation, this is more correctly referred to as nociception although the terms are used interchangeably in this text (de Lahunta, 2001). A turn of the head or vocalization should be seen in response to the pinch, indicating that the painful stimulus has been transmitted up the spinal cord to the brain. If there is no such response then the stimulus must be increased, usually by using larger instruments such as needle holders, pliers, or even with an electrical stimulator (13.36B) across the digit or nail bed. Confirmation of a loss of deep pain is often provided by the animal showing a line of analgesia over its flank. The animal responds to
27
28
Small Animal Spinal Disorders
a b
SITE OF INJURY
2.21 Testing nociception or deep pain sensation using pliers (Scott, 1997; Scott and McKee, 1999). The reflex withdrawal of the limb must not be mistaken for a behaviural response. Similarly, some animals will react to the change in body position induced by the withdrawal reflex. Ideally, the withdrawal reflex should be initiated first, and then further pressure applied to evaluate nociception (6.2, 13.36B).
pinching skin of dermatomes arising cranial to a lesion but shows no response in those arising caudally. Animals with cervical injuries very rarely lose deep pain. This is because animals that suffer spinal cord injuries severe enough to interfere with nociception usually are left with no motor function, which results in rapid asphyxiation. This emphasizes the importance of assessing for hypoventilation in any severely tetraparetic or tetraplegic animal (6.1, 7.11). One example where there may be a unilateral decrease in deep pain after cervical injury is following catastrophic failure of a disc after trauma. Here the nucleus pulposus explodes dorsolaterally to cause devastating, asymmetrical neurological deficits (Griffiths, 1970). Absence of deep pain sensation therefore indicates severe spinal cord damage (see ‘Assessing the severity of the lesion’, page 31; ‘Spinal cord nerve fibers and the effect of compression’, page 6; ‘Mechanisms of recovery after spinal cord injury’, page 87).
LOCALIZATION OF LESIONS On the basis of the neurological examination, it is usually possible to identify the location of the spinal cord lesion (2.22). Functionally, the spinal cord may be divided into four regions: • A: C1–C5. • B: C6–T2 (cervical intumescence). • C: T3–L3. • D: L4–S3 (lumbar intumescence). Areas A and C, the cervical and thoracolumbar spinal cord, convey primarily the UMNs. Areas B and D, the cervicothoracic and lumbosacral spinal cord, provide
c
THORACIC LIMB DEFICIT
PELVIC LIMB DEFICIT
UMN
UMN
LMN
UMN
NORMAL
UMN
NORMAL
LMN
C1–C5
C6–T2
T3–L3
L4–S3
d
2.22 Lesions in specific regions will produce different combinations of neurological signs. Lesions in the cervical spinal cord (a) produce UMN signs in all four limbs. Lesions in the segments of the cervical intumescence (b) produce LMN deficits in the thoracic limbs and UMN signs in the pelvic limbs. Lesions in the thoracolumbar cord (c) produce UMN signs in the pelvic limbs only, with normal thoracic limbs (the Schiff–Sherrington sign may be present in peracute lesions). Lesions in the lumbar intumescence (d) produce LMN signs in the pelvic limbs, tail and perineum; the thoracic limbs are normal.
innervation to the thoracic and pelvic limbs respectively. Area B also conveys UMNs to the pelvic limbs. It can be seen in 1.12 that part of the LMN lies within the spinal cord; lesions in areas B and D of the spinal cord will therefore produce LMN signs in the limbs. Variations are possible, for example in some animals with myelopathy affecting the caudal cervical segments, the thoracic limbs often show a mixture of UMN and LMN signs. They may show increased thoracic limb muscle tone, which is an UMN effect on the elbow and carpal extensor muscles (Seim and Withrow, 1982). In addition, there is often an associated LMN weakness of the elbow flexors resulting in a weak withdrawal reflex. Dogs with severe C6–T2 signs also have a short-strided, choppy or ‘disconnected’ thoracic limb gait and a long-strided, ataxic pelvic limb gait (see Chapter 11). In contrast, dogs with C1–C5 signs often show a long-strided or ‘floating’ thoracic limb gait together with a long-strided, ataxic pelvic limb gait (Baum et al., 1992).
Patient examination
When considering the neurological localization, remember that the spinal cord segments are not all contained in the vertebra of the same number, especially in the cervical and lumbar region (1.2, 1.5). Examples of potential pitfalls in neurological examination include: • LMN deficits in all limbs generally indicate diffuse peripheral nerve or neuromuscular disease. Occasionally, deficits may only be present in either the thoracic limbs or the pelvic limbs early in the course of generalized LMN diseases. • If there are any cranial nerve deficits, or signs such as a change in mentation or behavior, then it is very unlikely that the animal only has a spinal cord lesion.
•
•
•
•
• •
• • 2.23 Eight-year-old Golden retriever with acute onset of left hemiparesis and inability to stand. Neurological deficits localized to C1–C5 spinal cord segments. No lesions were detected in the cervical spine; CT scan revealed a mass in the left medulla. In retrospect, the only clue to a brainstem lesion was a subtle decrease in mental status.
A
•
Rarely an animal will have no cranial nerve deficits and will appear to localize to the C1–C5 spinal cord segments when it actually has a lesion in its brainstem (2.23). Animals with cerebellar disease as well as those with spinal cord lesions affecting the spinocerebellar tracts often show hypermetric and dysmetric limb movements. Animals with primary cerebellar disease (see ‘Neuroaxonal dystrophy’, page 320), can usually be distinguished as they have additional signs such as head tremors, dysmetric movements or menace deficits (Holliday, 1980). Animals with cervical lesions can sometimes show vestibular signs. Local nerve blocks or neurectomy of the first cervical dorsal nerve roots produces vestibular signs in monkeys and in horses (Biemond and De Jong, 1969; Mayhew, 1999). ‘Central cord syndrome’ can cause unusual signs. With this condition a high-cervical lesion causes tetraparesis with much more severe deficits in the thoracic limbs than in the pelvic limbs; spinal reflexes are preserved in all four limbs. An LMN lesion may obscure a concomitant UMN lesion (2.24). Animals with some spinal cord lesions, such as arachnoid cysts or intramedullary tumors, may show urinary or fecal incontinence while still able to walk (12.4). Apparent paraparesis due to a mild cervical lesion (see Chapters 7 and 11). A lesion in the sacral spinal cord segments can mimic nerve root compression at the lumbosacral junction (2.25). Animals with a choppy thoracic limb gait and paraparesis may have a lesion between T1–T6 (3.36A).
B
2.24 Aged Dalmatian with mild, progressive paraparesis, dilated anus and urinary incontinence. Neurological deficits localized to the L7–S2 spinal cord segments. A CT scan of the lumbosacral space was unremarkable (10.2B, 10.10A). A: Myelography revealed masses (arrowheads) in the subarachnoid space at C1/2 on the right; caudal C2 on the left. B: Cranial C6 on the right; mid-C7 on the right; and a number of small filling defects over the lumbar cord (not shown). The mass at C2 was a nerve sheath tumor with subarachnoid metastasis (4.27, 4.28).
29
30
Small Animal Spinal Disorders
A
B
2.25 Seven-year-old Shih Tzu that presented initially with loss of tail tone and then mild paraparesis, anal arreflexia and incontinence over the next 8 months. Localization was to the cauda equina or L4–S3 spinal cord segments. T1-weighted MRIs post contrast medium. A: sagittal and B: transverse images show that the dog has an enhancing mass (arrows), most likely a nerve sheath tumor, involving the right L5 nerve root and invading the sacral spinal cord segments. The lumbosacral disc space appears normal.
2.26 A: Dog with a brachial plexus avulsion injury affecting the caudal portion of its left brachial plexus. It is unable to fix the carpal or elbow joint but can still flex its elbow joint due to preserved cranial plexal function. It has also begun to mutilate its foot. B: This dog sustained an injury to both the cranial and caudal portions of its right brachial plexus. It is unable to flex the elbow or to bear weight on the limb (note dropped elbow and collapsed carpal joints) and is at high risk of abrading its foot.
A
B
Assessment of the brachial and lumbosacral plexus Lesions affecting these areas may be missed if specific testing is not performed. Animals must be assessed for brachial plexus lesions after trauma (13.1) or when they have a chronic, progressive thoracic limb lameness (12.2). This is done by testing for anisocoria (2.15); unilateral loss of the cutaneous trunci reflex (2.11, 2.12); reduced thoracic limb flexor response (2.18, 2.26); and reduced nociception (2.21). Animals with lesions of the brachial plexus may have deficits referable to the cranial brachial plexus, the caudal brachial plexus, or both (1.4, 2.26). The suprascapular, subscapular, musculocutaneous and axillary nerves derive mainly from the cranial portions of
the plexus (C5, C6 and C7). Deficits of the cranial plexus cause loss of elbow flexion and may result in subluxation of the shoulder joint. The radial, median and ulnar nerves derive mainly from more caudal portions of the plexus (C7, C8, T1 and T2). Deficits of the caudal plexus cause an inability to fix the elbow or carpus and therefore the animal cannot bear weight on the limb. In addition they often have ipsilateral (full or partial) Horner’s syndrome and cutaneous trunci reflex deficit (1.4, 2.15). Elbow flexion is preserved when the cranial plexus is intact (2.26). Animals must be assessed for lumbosacral plexus lesions after trauma by testing for sciatic, pudendal or pelvic nerve deficits. One of the most specific tests for a sciatic nerve deficit is to assess hock flexion (2.27).
Patient examination
2.28 Sensory deficits secondary to a brachial plexus lesion may give rise occasionally to trophic ulceration as shown here (arrowheads). Trophic ulcers occur due to an inability to protect the foot and also from abnormal forces placed on the foot due to digital paralysis (Hunt and Chapman, 1991).
2.27 When assessing the withdrawal reflex in the pelvic limb, all three joints should be assessed. An animal can flex both its hip and stifle joints by using the femoral and sciatic nerves whereas it can only flex its hock joint by using the peroneal division of the sciatic nerve. Weak or absent hock flexion can be appreciated best by pulling the hock joint gently into extension and then applying a stimulus to the digits.
Loss of proprioception is a more sensitive indicator although this does not differentiate an UMN lesion from a LMN lesion. Sensory loss in small animals usually reflects spinal cord or nerve root lesions. Occasionally, it reflects peripheral nerve injury and specific test sites for the various cutaneous sensory nerves have been defined (Bailey and Kitchell, 1987). For the thoracic limb the main sites are: • Musculocutaneous nerve—just distal to the medial epicondyle of the humerus. • Radial nerve—dorsal aspect of the third or fourth digit. • Ulnar nerve—just distal to the olecranon on the caudolateral aspect of the antebrachium and on the lateral surface of the fifth digit. For the pelvic limb the main sites are: • Saphenous (femoral) nerve—4 cm distal to the medial condyle of the tibia.
•
Superficial peroneal—dorsal aspect of the 3rd digit, proximal end. • Deep peroneal—skin between the 2nd and 3rd digits, dorsal aspect. • Tibial nerve—skin at the proximal border of the tarsal pad. Sensory loss following brachial or lumbosacral plexus injury occasionally gives rise to trophic ulceration (2.28).
ASSESSING THE SEVERITY OF THE LESION Assessing the severity of a lesion plays a major part in the diagnostic procedure. In certain patients, it has as much bearing on prognosis as the etiology; if a poor prognosis is suggested from the neurological examination further investigations may be deemed unnecessary. In general, patients with spinal diseases that show LMN deficits have a worse prognosis for a return to function than those showing UMN deficits. One exception to this is for dogs with thoracolumbar disc disease (Dhupa et al., 1999) (see Chapter 8). In UMN injuries, rate of onset, duration, and degree of spinal cord damage all affect the clinical signs. Rapid progression of acute signs tends to reflect acute decompensation, which if treated early has a favorable outcome unless the damage is irreversible. Slow progression
31
32
Small Animal Spinal Disorders
generally reflects a chronic process that is gradually outstripping secondary compensation mechanisms (Gilson, 2003). For thoracic and lumbar lesions, the degree of dysfunction can be classified as grades 1–5 (Scott, 1997; Scott and McKee, 1999): 1. Pain only. 2. Paraparesis—walking. 3. Paraparesis—not walking. 4. Paraplegia. 5. Paraplegia with loss of deep pain sensation. The animal usually progresses through these stages as it deteriorates unless the disease is peracute in onset, such as trauma or a vascular lesion. The scheme applies to both UMN and LMN lesions. Control of micturition is often lost between grade 4 and grade 5 but full control of continence may be incomplete as early as grade 2. Recovery of spinal cord function usually progresses steadily in the reverse order, unless the animal develops either a reflex bladder or spinal walking (6.2, 13.30, 13.34). Recovery after spinal injury is usually defined as recovery of continence and of the ability to walk unaided. The severity of the neurological deficits depends on two anatomical features: the position of the tracts in the spinal cord that carry the respective function and the diameter of the fibers transmitting that function (see ‘Spinal cord nerve fibers and the effect of compression’, page 6). The prognosis worsens with increasing neurological deficit, as in general this reflects greater spinal cord damage. The presence of a crossed extensor reflex (seen in UMN lesions) and the Schiff–Sherrington sign indicate severe lesions, but are not in themselves prognostic indicators. Schiff–Sherrington syndrome in particular is often thought to indicate a poor prognosis but this is largely because it is only seen after severe spinal cord injury. However, some animals with Schiff– Sherrington syndrome make a good recovery (13.7) and so prognosis should be inferred from the presence or absence of nociception. The prognosis for patients without deep pain sensation (grade 5) is guarded for most disorders, especially if it has been absent for longer than 48 h (see also page 132). Unfortunately, although it is usually obvious how long an individual animal has been paraplegic it is seldom clear exactly when that animal lost deep pain. Loss of deep pain sensation after disc disease is associated with a 50–60% chance of recovery following surgery (see Chapter 8, page 138 and Table 8.1a). Animals that have absent deep pain sensation after trauma carry a much worse prognosis whatever the duration of the clinical signs (Olby et al., 2003) (see Chapter 13, page 302). In other grades of dysfunction, the prognosis is also somewhat dependent on the etiology. For example, a Dachshund with grade 3 signs could possibly have a
thoracolumbar disc extrusion or a spinal tumor. Clearly, the prognosis for such disorders could differ markedly. Spinal shock is a phenomenon that affects primarily humans and higher primates. It is where an UMN lesion is associated with a temporary loss of spinal reflexes below the lesion; in humans this can last for a week or more. If it occurs in dogs and cats it is transient and is only seen within the first few hours after injury. It could confound neurological localization and therefore in animals examined immediately after trauma a repeat examination should be performed a few hours later (Walmsley and Tracey, 1983; Gopal and Jeffery, 2001). Cervical spinal cord lesions display a similar progression of signs to that described above but urinary incontinence or loss of nociception is very rare. Severe cervical spinal cord lesions may result in respiratory failure caused by interference with respiratory control, although this is uncommon (6.1, 7.11).
DETERMINING THE ETIOLOGY Once the lesion has been located, a list of differential diagnoses can be made. Many components go into this assessment, including breed and age of the patient, history, presenting signs, progression, and physical and neurological findings. It is important not to depend entirely on one feature to make a firm diagnosis. It is reasonable to start from the assumption that the most common disease is the cause, but this should ideally either be confirmed or attempts made to exclude less likely conditions, particularly if the patient is not progressing as expected. Differential diagnosis is discussed in Chapter 3.
REFERENCES Bailey, C.S., Kitchell, R.L. (1987) Cutaneous sensory testing in the dog. Journal of Veterinary Internal Medicine 1, 128–135. Baum, F., Trotter, E.J., de Lahunta, A.D. (1992) Cervical fibrotic stenosis in a young Rottweiler. Journal of the American Veterinary Medical Association 201, 1222–1224. Biemond, A., De Jong, J.M. (1969) On cervical nystagmus and related disorders. Brain 92, 437–458. Boswood, A., Lamb, C.R., White, R.N. (2000) Aortic and iliac thrombosis in six dogs. Journal of Small Animal Practice 41, 109–114. Braund, K.G., Sharp, N.J.H. (2003) Neurological examination and localization. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1092–1107. Philadelphia: Elsevier Science. de Lahunta, A. (2001) Neurological examination. In: K.G. Braund (ed.), Clinical Neurology in Small Animal–Localization, Diagnosis and Treatment. Ithaca: International Veterinary Information Service. http://www.ivis.org/ special_books/Braund/deLahunta/chapter_frm.asp?LA⫽1 Dhupa, S., Glickman, N.W., Waters, D.J., Dhupa, S. (1999) Functional outcome in dogs after surgical treatment of caudal lumbar intervertebral disk herniation. Journal of the American Animal Hospital Association 35, 323–331. Gilson, S.D. (2003) Neuro-oncologic surgery. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1277–1286. Philadelphia: Elsevier Science.
Patient examination
Gopal, M.S., Jeffery, N.D. (2001) Magnetic resonance imaging in the diagnosis and treatment of a canine spinal cord injury. Journal of Small Animal Practice 42, 29–31. Griffiths, I. (1970) A syndrome produced by dorso-lateral ‘explosions’ of the cervical intervertebral discs. Veterinary Record 87, 737–741. Holliday, T.A. (1980) Clinical signs of acute and chronic experimental lesions of the cerebellum. Veterinary Science Communications 3, 259–278. Hunt, G.B., Chapman, B.L. (1991) ‘Trophic’ ulceration of two digital pads. Australian Veterinary Practitioner 21, 196, 205. Mayhew, I.G. (1999) The healthy spinal cord. American Association of Equine Practitioners 45, 56–66. McDonnell, J.J., Platt, S.R., Clayton, L.A. (2001) Neurologic conditions causing lameness in companion animals. Veterinary Clinics of North America, Small Animal Practice 31, 17–38. Morgan, P.W., Parent, J., Holmberg, D.L. (1993) Cervical pain secondary to intervertebral disc disease in dogs; radiographic findings and surgical implications. Progress in Veterinary Neurology 4, 76–80. Neer, T.M. (1992) A review of disorders of the gallbladder and extrahepatic biliary tract in the dog and cat. Journal of Veterinary Internal Medicine 6, 186–192. Olby, N.J., Harris, T., Munana, K.R., Skeen, T.M., Sharp, N.J.H. (2003) Long-term functional outcome of dogs with severe spinal cord injuries. Journal of Neurotrauma (in press). Pang, W.W., Mok, M.S., Lin, M.L., Chang, D.P., Hwang, M.H. (1998) Application of spinal pain mapping in the diagnosis of low back pain— analysis of 104 cases. Acta Anaesthesiology Singapore 36, 71–74.
Scott, H.W. (1997) Hemilaminectomy for the treatment of thoracolumbar disc disease in the dog: a follow-up study of 40 cases. Journal of Small Animal Practice 38, 488–494. Scott, H.W., McKee, W.M. (1999) Laminectomy for 34 dogs with thoracolumbar intervertebral disc disease and loss of deep pain perception. Journal of Small Animal Practice 40, 417–422. Seim, H.B., Withrow, S.J. (1982) Pathophysiology and diagnosis of caudal cervical spondylo-myelopathy with emphasis on the Doberman Pinscher. Journal of the American Animal Hospital Association 18, 241–251. Walmsley, B., Tracey, D.J. (1983) The effect of transection and cold block of the spinal cord on synaptic transmission between Ia afferents and motoneurones. Neuroscience 9, 445–451.
FURTHER READING de Lahunta, A. (2001) Neurological examination. In: K.G. Braund (ed.), Clinical Neurology in Small Animals—Localization, Diagnosis and Treatment. Ithaca: International Veterinary Information Service (www.ivis.org), B0201.1001. http://www.ivis.org/special_books/Braund/ deLahunta/chapter_frm.asp?LA⫽1
33
Diagnosis and differential diagnosis
Differential diagnosis A: C1–C5 35 B: C6–T2 35 C: T3–L3 36 D: L4–S3 37 Further reading
35
39
On the basis of the neurological examination the lesion is localized to one of the following four areas of the spinal cord (2.22): • A: C1–C5. • B: C6–T2. • C: T3–L3. • D: L4–S3. These areas relate to spinal cord segments. Refer to 1.2 and 1.5 to identify the location of these areas within the vertebral column. A differential diagnosis list is then drawn up, which embraces all the disease conditions that could be present. It is usual to create this list using the DAMNIT format (Table 3.1), the elements of which can also be expressed using the acronym VITAMIN D. Conditions that occur frequently in each of the four spinal cord regions are listed in Tables 3.2–3.5. When the animal has no neurological deficits, consider the conditions listed in Table 2.1 (disorders mimicking spinal disease) and Table 3.6 (diffuse or non-localizable pain). Other differential diagnoses are listed in Box 7.2 (neck pain), Box 8.1 (thoracolumbar disc disease), Box 9.1 (atlantoaxial disease), Table 10.2 (lumbosacral disease), Table 11.1 (cervical spondylomyelopathy), and Box 13.1 (trauma). Non-spinal causes of neurological disease and differentials for exercise intolerance are included in Table 3.6. The list of differential diagnoses for any given patient can be shortened considerably if some basic information is taken into account, such as the following: • The age of the patient. • Whether the condition is acute or chronic, progressive or static. • Presence or absence of spinal pain.
•
Chapter
3
The neurological localization will also exclude some conditions (see Chapter 2).
The breed of the patient may also indicate that certain diseases are more likely to be present, but this should not be the only information on which the diagnosis is based. An additional way to narrow the list is by a process of elimination. For the dog in 3.1, the signalment and history were not suggestive of degenerative, anomalous or metabolic conditions; there was no pain to suggest trauma; and the pattern of progression made vascular disease unlikely. The two most probable differential diagnoses were therefore either neoplasia or infectious and inflammatory disease.
DIFFERENTIAL DIAGNOSIS A: C1–C5 In adult dogs, intervertebral disc disease (Table 3.2) is the most common condition affecting this region. In dogs less than 2 years of age the differential diagnoses are different. Here, atlantoaxial subluxation, inflammatory CNS disease, discospondylitis and trauma are the most likely causes. Spinal tumors can occur in dogs of any age. The most common tumors in the cervical spine are meningiomas and nerve sheath tumors; both are more common in older dogs. Some dogs with cervical spondylomyelopathy (CSM) have neurological signs with a C1–C5 pattern of dysfunction (see page 28), although the lesion is often more caudal in the cervical spine. Acute, non-painful, non-progressive deficits usually result from ischemic myelopathy due to fibrocartilaginous embolism (FCE). Signs are often asymmetrical and severe. In cats, clinical disc disease is rare in the neck. The likely diagnoses are trauma, neoplasia (usually lymphoma) and inflammatory diseases, particularly feline infectious peritonitis (FIP). Atlantoaxial subluxation is rare in cats.
B: C6–T2 Similar considerations apply here to those causing C1–C5 signs, although atlantoaxial subluxation does not occur.
36
Small Animal Spinal Disorders
Table 3.1 Differential diagnoses for spinal disease using the DAMNIT scheme * Category
Diseases (common conditions in bold)
Degenerative
Degenerative myelopathy Synovial cysts Leukodystrophies Lumbosacral disease Facet joint pain
Cervical spondylomyelopathy Disc disease Lysosomal storage disease Cervical fibrotic stenosis Sacroiliac joint pain
Anomalous
Syringo(hydro)myelia Pilonidal (dermoid) sinus Spinal arachnoid cyst Epidermoid cyst Atlantoaxial subluxation Spina bifida Tethered cord syndrome Sacral osteochondritis dissecans
Vertebral malformations Cartilaginous exostoses Meningo(myelo)celes Spinal dysraphism Hereditary myelopathy Sacrocaudal dysgenesis Schmorl’s node Atlantooccipital dysplasia
Metabolic
See Table 3.6, Exercise intolerance
Lysosomal storage disease
Neoplastic
Primary or secondary tumor
Epidural lipomatosis
Nutritional
Hypervitaminosis A
Thiamine deficiency
Idiopathic
Tumoral calcinosis
Disseminated idiopathic skeletal hyperostosis
Iatrogenic injury
Peridural scar
Diagnostic, radiation or surgical injury
Inflammatory and Infectious
Discospondylitis Spinal epidural empyema Gelfoam reaction
Meningo(encephalo)myelitis Cauda equina neuritis Foreign body migration
Traumatic
Sacrocaudal injury Brachial plexus avulsion myelopathy Gunshot injury
Dural tear Traumatic disc injury Fracture/luxation
Toxic
See Table 3.6, Exercise intolerance
Vascular
Fibrocartilaginous embolism Spinal cord hematoma Vascular malformation Fat graft necrosis Ischemic neuromyopathy
Traumatic feline ischemic myelopathy Neurogenic claudication Ascending myelomalacia Spinal cord hemorrhage
* Common conditions are in bold.
Intervertebral disc disease is less frequent in the caudal cervical spine, but does occur, occasionally even at C7/T1. Cervical spondylomyelopathy is most prevalent in the caudal cervical spine of Dobermans and Great Danes. Ischemic myelopathy also occurs with some frequency in this region (see Table 3.3).
C: T3–L3 This region of the spine accounts for most cases of spinal disease. Disc herniation is the most likely diagnosis in dogs older than 1 year. In younger dogs, inflammatory CNS disease, discospondylitis and trauma are common. In cats, disc disease is uncommon but does occur,
Diagnosis and differential diagnosis
Table 3.2 Diseases causing signs that localize to C1–C5 (see also Table 3.1)
Painful
Non-painful
Acute
Chronic
Atlantoaxial subluxation CSM Disc disease Neoplasia Discospondylitis Meningomyelitis Fracture/luxation Spinal cord hematoma
Atlantoaxial subluxation Atlantooccipital dysplasia CSM Cervical fibrotic stenosis Synovial cysts Syringohydromyelia Hypervitaminosis A
FCE
Spinal arachnoid cyst Tumoral calcinosis Neoplasia
CSM, cervical spondylomyelopathy.
Painful
Non-painful
Chronic
Painful
Disc disease Neoplasia Discospondylitis Meningomyelitis Fracture/luxation Ascending myelomalacia
Disc disease Synovial cysts Tumoral calcinosis
Non-painful
Traumatic feline ischemic myelopathy FCE
Degenerative myelopathy Spinal arachnoid cyst Neoplasia
Acute
Chronic
Painful
Neoplasia Discospondylitis Meningomyelitis Fracture/luxation Disc disease Ascending myelomalacia Psoas muscle injury
Lumbosacral disease Tethered cord syndrome Sacral OCD
Non-painful
Cauda equina neuritis Ischemic neuromyopathy FCE
Degenerative myelopathy Spina bifida Sacrocaudal dysgenesis Dermoid sinus Neoplasia
Chronic
CSM CSM Disc disease Synovial cysts Neoplasia Discospondylitis Meningomyelitis Fracture/luxation Spinal cord hematoma Brachial plexus avulsion FCE
Acute
Table 3.5 Diseases causing signs that localize to L4–S3 (see also Table 3.1)
Table 3.3 Diseases causing signs that localize to C6–T2 (see also Table 3.1) Acute
Table 3.4 Diseases causing signs that localize to T3–L3 (see also Table 3.1)
Spinal arachnoid cyst Dermoid sinus Neoplasia
CSM, cervical spondylomyelopathy.
particularly in aged animals. Trauma, neoplasia and inflammatory diseases are the most likely causes of feline thoracolumbar spinal disease. In large-breed, older dogs with chronic signs, the primary differential diagnosis is degenerative myelopathy. Chronic disc herniation (Hansen type II), synovial cyst(s), neoplasia and mild L4–S3 lesions must be eliminated before this diagnosis is reached (Table 3.4).
D: L4–S3 Subacute or chronic presentations with signs localizing to this region are usually referable to the lumbosacral junction (Table 10.2). Neoplasia or discospondylitis
OCD, osteochondritis dissecans.
may be present. Lumbosacral disease is common in large breeds of dog, but is less common in smaller dogs and is rare in cats. In sacrocaudal injuries, signs referable to the bladder, perineum, and even paraparesis may be seen. Degenerative myelopathy may appear to localize to this region if the patellar reflexes are absent. This is due to nerve root involvement; the lesion is still largely T3–L3 in nature. Ischemic myelopathy is also seen here. In young animals, congenital defects of the vertebrae or spinal cord are likely (Table 3.5). Exterior signs of spina bifida may be present, but this disease is uncommon. Taking the signalment and neurological localization into account can therefore reduce the differential diagnosis
37
38
Small Animal Spinal Disorders
3.1 Eight-year-old Akita that presented with progressive tail paralysis, loss of anal reflex, incontinence and pelvic limb weakness over 3 weeks. The myelogram indicates a discrete filling defect within the dural sac over L6 vertebra. Final diagnosis was an intradural fibrosarcoma.
list. It may be reasonable, in some circumstances, to proceed with treatment without further investigations, for example in a middle-aged Dachshund with mild T3–L3 signs and where conservative treatment is planned. However, if the patient does not progress as expected, or if there is any doubt about the diagnosis, the provisional diagnosis should be reassessed and more
3.2 German shepherd dog with severe lumbosacral pain. Rectal examination revealed an asymmetrical prostate gland fixed to the pelvis. Ultrasound revealed an enlarged prostate (arrowheads) with mineralization (arrow) and small cystic cavities suggestive of prostatitis or carcinoma. The dog also had lumbosacral disease (10.4) and dermatofibrosis with renal cystadenocarcinoma (4.35). Nevertheless, it tested positive for Ehrlichia canis and returned to work as a search and rescue dog after treatment with doxycycline.
definitive diagnostic tests instituted (see Chapter 4). Occasionally, an animal may present with more than one disease that could explain its clinical signs (3.2). Neurological localization should define the relevant
Table 3.6 Differential diagnoses for diffuse pain, non-spinal neurological disease and exercise intolerance Diffuse pain
Non-spinal neurological disease
Exercise intolerance
Polyarthritis Polymyositis Pancreatitis Renal or ureteral calculi Gallstones Gastrointestinal parasites Kidney worm Osteoporotic vertebral fracture Mid-thoracic vertebral lesion Other abdominal pain Meningoencephalitis Thalamic pain syndrome Prostatic disease Urethral tumor
Myositis or myopathy Peripheral neuropathies Radiculopathies Myasthenia gravis Tick paralysis Subacute organophosphate toxicity
Myasthenia gravis Mild cervical myelopathy Addison’s disease Hypoglycemia Polymyositis Toxic myopathy Hypokalemic polymyopathy Hyperthyroidism Congenital myopathy Tick paralysis Coonhound paralysis Lumbosacral pain Idiopathic polyradiculoneuritis Protozoal myositis and polyradiculoneuritis Subacute organophosphate toxicity Botulism Peripheral neuropathy Ischemic neuromyopathy Intermittent claudication Cardiac disease Upper airway disease
Botulism Localized tetanus Brain tumor Intermittent claudication Ischemic neuromyopathy Psoas muscle hemorrhage
Diagnosis and differential diagnosis
lesion but occasionally each disease may localize to the same region. This shows that the clinician must strive to keep an open mind and try to rule out each potential diagnosis in step-wise fashion. Response to treatment is one useful means of doing this, in which case the simplest therapy should be tried first. The case discussed in 3.2 illustrates the importance of considering a full range of differential diagnoses for each case and not
simply assuming that the most common disease process is the one responsible for the clinical signs.
FURTHER READING Braund, K.G. (2003) Clinical Neurology in Small Animals: Localization, Diagnosis and Treatment. http://www.ivis.org/special_books/Braund/ toc.asp
39
Diagnostic aids
Routine laboratory analysis Hematology 41 Biochemistry 41 Urinalysis 42 Serology 42 Microbiology 42 Other 42
41
4 Biopsy
61
Key issues for future investigation References
Cerebrospinal fluid 42 Indications and contraindications for CSF collection 43 Collection of CSF 43 (64) Sample handling and laboratory analysis 43 Normal CSF findings 45 Abnormal CSF findings 45 Normal findings in the face of disease 45 Radiography 45 Survey radiographs 45 Radiographic positioning and normal spinal radiographs 46 Special radiographic procedures Myelography 48 (70) Epidurography 50 Discography 50
48
Principles of spinal radiology 50 Myelographic interpretation 50 Complications of myelography
Chapter
62
Procedures 64 Collection of CSF 64 Myelography 70
The exact selection of diagnostic tests to be employed varies with the circumstances. In the following chapters, recommendations as to the tests most likely to provide a diagnosis in a particular disease are given. Clearly, individual preference, experience and the availability of equipment will influence these decisions.
ROUTINE LABORATORY ANALYSIS A complete blood count, chemistry profile and urinalysis should be checked in all patients undergoing nonemergency procedures. They rarely provide definitive diagnostic information in neurological diseases, but can be important for detecting concurrent disorders (Braund and Sharp, 2003) (Table 3.6). In emergency situations, analysis may be restricted to packed cell volume, total protein, serum glucose and urea concentrations.
Hematology In the majority of animals with spinal disease, the hemogram is unremarkable, although a stress leukogram (lymphopenia, eosinopenia and leukocytosis) is a common finding. Patients with inflammatory diseases of the spinal cord and meninges usually have normal hemograms. Dogs with discospondylitis or rickettsial disease and cats with feline infectious peritonitis (FIP) may have inflammatory hemograms, but this is not consistent.
51
Other imaging techniques 54 Ultrasonography 54 Computed tomography 55 Magnetic resonance imaging 57 Scintigraphy 59 Clinical electrophysiology 60 Electromyography 60 Spinal cord evoked response 60 F waves and cord dorsum potentials
62
Biochemistry 61
Metabolic diseases may cause generalized weakness, which could be mistaken for spinal disease. Examples
42
Small Animal Spinal Disorders
include Addison’s disease, Cushing’s disease, hypothyroidism, hypoglycemia, hypocalcemia, hyponatremia and hypokalemic polymyopathy. Raised serum creatine kinase concentrations are usually indicative of muscle injury. This increases the index of suspicion for protozoal disease, especially when multifocal neurological deficits, pulmonary or hepatic disease coexist (4.46). Note that serum creatine kinase concentrations may be raised in patients that suffer trauma or are recumbent for lengthy periods. Abnormal biochemical findings may be seen in certain spinal diseases. Hypocalcemia is seen in some patients with generalized bone disorders, for example in secondary hyperparathyroidism. Hypercalcemia is seen in some patients with lymphoma or myeloma; this is a paraneoplastic effect. Hypergammaglobulinemia may be seen in myeloma and in chronic, infectious diseases such as FIP; serum electrophoresis is useful to differentiate the two.
Urinalysis Urinalysis may provide specific diagnostic information regarding renal and hepatic function or urinary tract infection. If bladder function is in doubt in patients with spinal disease, urinalysis should be performed on a sample collected by cystocentesis when the patient is first evaluated. Ultrasound may aid collection of urine if the bladder is not palpable. Subsequent urinalysis should be performed periodically until normal urinary function returns (see Chapter 15). Urinary tract infection (UTI) associated with urine retention is present frequently in animals with spinal disease. Urinalysis reveals high white blood cell counts and elevated total protein, and bacteria may be present in the sediment. If so, urine culture should be performed and antibiotic sensitivity determined. Fungal hyphae may be detected in dogs with systemic aspergillosis, especially if the urine has been at room temperature for more than 12 h. For routine urinalysis, aseptic catheterization or cystocentesis is suitable. For culture, cystocentesis is preferred. Bence–Jones proteinuria may be a feature of myeloma.
Serology Serology is useful in many diseases, and should be considered in any spinal patient where the origin of the disorder is obscure. Specific examples include Neospora caninum, Cryptococcus neoformans, Coccidioides immitis, Bartonella sp., Ehrlichia sp., and also Brucella canis in dogs with discospondylitis. Acute and convalescent
titers are usually necessary for Rickettsia sp. and Toxoplasma gondii. Canine distemper virus (CDV) and FIP both cause neurological lesions, which may present with spinal involvement. The presence of serum IgG is not confirmatory of central nervous system (CNS) involvement by these viruses unless it can be shown to be due to intrathecal production rather than caused by leakage from serum. Ante-mortem confirmation of these diseases is challenging. Non-spinal causes of weakness may be detected by serum tests such as cholinesterase concentrations in organophosphate toxicity or acetylcholine receptor antibody titers in animals with myasthenia gravis (Table 3.6).
Microbiology Urine culture is indicated in patients with cystitis related to neurological disease. Sensitivity testing should also be performed to determine the most appropriate antibiotic with which to treat infections. However, bladder infections usually only resolve when urinary function returns to normal. In patients with discospondylitis, urine and blood culture may be attempted to isolate the causative organism along with direct aspiration from the affected disc space(s). If wound infections occur following surgery, culture from the depths of the wound should be performed after aseptic preparation of the skin surface (see ‘Wound infection’, page 357).
Other Endocrine testing, analysis of Von Willebrand (VW) factor, and bleeding times may all be deemed necessary during preoperative evaluation (see Chapters 6 and 11). Coagulation profiles and testing for warfarin are indicated in animals with coagulopathies. Blood gas analysis should be performed in tetraplegic animals where hypoventilation is suspected (7.11).
CEREBROSPINAL FLUID Cerebrospinal fluid (CSF) analysis is an important tool in the investigation of neurological patients, but there are many limitations to CSF analysis. Abnormal CSF is strongly indicative of neurological disease, but it is a relatively non-specific finding. Cerebrospinal fluid should be collected routinely from either the cerebello-medullary cistern (CMC— ‘cisterna magna’) or the lumbar region. It is important to do this prior to performing myelography as injection of contrast medium induces a sterile meningitis that
Diagnostic aids
will confound subsequent interpretation of CSF analysis for 1 week or more (Widmer et al., 1992). When radiography and myelography are normal, and in patients with multifocal signs, CSF should always be analysed. Even when the diagnosis is obvious, the information gained may be very useful on a retrospective basis. An example would be a dog with a thoracolumbar disc extrusion that has a very high protein level in CSF and that subsequently develops myelomalacia after surgery. Collection of CSF is a very low-risk procedure when performed with care and when taking appropriate precautions. It is nevertheless an invasive test and informed consent should always be obtained. Owners should also be appraised of any specific risk factors before the procedure (see below). Collection of CSF should always be performed with the patient intubated and under general anesthesia. Ideally, corticosteroids should not be given prior to CSF collection as they may alter results and render interpretation inaccurate.
• •
•
Following spinal trauma unless a myelogram is to be performed. Patients that could have atlantoaxial subluxation should not have CSF collected from the CMC because of the danger associated with patient positioning. If raised intracranial pressure or brain herniation could be present then the risk of collection is increased significantly. This may be so in any patient with a severe intracranial lesion, but particularly following head trauma, with spaceoccupying lesions, or with severe inflammatory disease. Signs indicative of raised intracranial pressure include depression, stupor or coma; hypoventilation; bradycardia; dilated pupils or anisocoria; and cranial nerve deficits such as decreased or absent physiologic nystagmus and poor or absent pupillary light reflexes. Collection of CSF from either of the spinal sites in the face of increased intracranial pressure increases the likelihood of brain herniation.
Indications and contraindications for CSF collection INDICATIONS
Collection of CSF
•
See ‘Procedures’, page 64.
• • • •
Suspected intracranial disease (see ‘Contraindications’, below). Suspected multifocal disease. Spinal disease where survey radiographs are normal. Suspected polyneuropathy. Ideally in every patient unless contraindicated.
CONTRAINDICATIONS
•
A
Where general anesthesia is contraindicated.
Sample handling and laboratory analysis Analysis of CSF should include gross examination, total and differential white blood cell (WBC) count and total protein (4.1). A sedimentation technique (4.2) should be used unless the sample can be sent to a specialized laboratory that will perform cytocentrifugation (Cellio,
B
4.1 A: Left to right—normal CSF; xanthochromia (which is caused by hemoglobin breakdown products and indicates previous intrathecal hemorrhage); and hemorrhage at collection. B: Degenerate neutrophils containing bacteria from the CSF of a dog with severe, diffuse pain and fever secondary to myelography performed 24 h earlier using iohexol from a contaminated, multi-use bottle. A filter should be used with multi-use containers.
43
44
Small Animal Spinal Disorders
A
B 4.2 Hans Wolfgang Kolmel Sedimentation Chamber (Free University of Berlin, with permission). A: Glass microscope slide is placed on the apparatus, B: filterpaper is placed on the glass slide, and C: the chamber is fitted. CSF is placed in the chamber and the filter paper absorbs fluid, while cells remain on the slide. The slide is then stained for cytology and differential count.
C
2001). White cell analysis should be performed ideally on fresh CSF within 30 min of collection as the cells deteriorate rapidly (Bienzle et al., 2000). If this is not possible, the specimen should be split in two aliquots and then one aliquot is put into a small EDTA tube or a drop of serum or an equal volume of 4% formalin solution added in order to preserve the cells (Evans, 1989; Bienzle et al., 2000). Other analyses performed on CSF include immunology and electrophoresis.
automatic means. Undiluted CSF is used, unless cell counts are particularly high in which case the CSF will appear turbid. Normal CSF is free of red blood cells. Normal white cell counts vary by laboratory, but are generally less than 5 cells per l (⬍5 ⫻ 106 per liter). An increase in WBCs is termed pleocytosis (4.1B).
GROSS EXAMINATION
Mild blood contamination has little effect on WBC counts in CSF. There are formulae for correcting CSF cell counts that take the peripheral blood WBC count into consideration (Rossmeissl et al., 2002). However, contaminating red blood cells (RBCs) do not greatly alter WBC counts, especially those already markedly elevated (Wilson and Stevens, 1977). It is probably adequate to remember that WBCs and RBCs are in a ratio of approximately 1 : 500 in blood, and to take this into account when viewing cell counts in blood-contaminated CSF.
Normal CSF is clear and colorless. Color change and turbidity may be noted in some diseases but these do not occur unless the abnormalities in cell count or protein are marked. The most frequent color changes are due to hemorrhage or xanthochromia (4.1A).
CELL COUNTS These are performed using a hemocytometer; the cell numbers are too low to be counted accurately by
CELL COUNTS AND BLOOD CONTAMINATION
Diagnostic aids
DIFFERENTIAL WHITE CELL COUNTS AND CYTOLOGY This is an important part of the examination of CSF, even in the face of a normal total WBC count. The WBCs must be concentrated by centrifugation or sedimentation (4.2), mounted on a slide and stained (Cellio, 2001). Most WBCs in normal CSF are mononuclear— lymphocytes, monocytes and occasional neutrophils or macrophages (Duque et al., 2002). The differential WBC count is most useful in distinguishing acute and chronic inflammation, for example between non-infectious suppurative meningoencephalitis and granulomatous meningoencephalitis (Thomson et al., 1989, 1990; Chrisman, 1992). Occasionally, high white cell counts will be seen in disc disease as well (Thomson et al., 1989).
PROTEIN CONTENT Protein content of CSF is estimated by a number of methods; use of a professional laboratory is best. Normal CSF collected from the CMC has a protein concentration of less than 30 mg/dl; it may be up to 45 mg/dl at the lumbar site. Total protein may be increased in many diseases but is non-specific. Electrophoresis provides information regarding the composition of the CSF protein.
MICROBIOLOGY Examination for the presence of bacteria may be performed by microscopy of stained samples (4.1B) and by culture. If CSF culture is attempted, enhancement using blood culture bottles is recommended. Bacterial diseases of the CNS are uncommon in dogs and cats and even when present, culture of CSF tends to be unrewarding (Remedios et al., 1996; Lavely et al., 2002; Radaelli and Platt, 2002) (see below). Care should be exercised in interpreting positive results in the absence of pleocytosis, as bacteria may be contaminants.
Normal CSF findings Normal CSF findings are given in Table 4.1.
Table 4.1 Normal values for commonly evaluated CSF parameters Color Specific gravity White cell count Differential white cells
Clear, colorless 1.004–1.006 ⬍5/l Lymphocytes and monocytes
Total protein Cerebello-medullary cistern Lumbar
10–30 mg/dl (0.1–0.3 g/l) 10–45 mg/dl (0.1–0.45 g/l)
Mononuclear pleocytosis is generally indicative of chronic inflammation, for example canine distemper or granulomatous meningoencephalomyelitis. Neutrophilic pleocytosis is seen mostly with infectious or non-infectious meningoencephalomyelitis (‘aseptic meningitis’) but may also be seen with meningiomas and vascular disease. Mixed pleocytosis is non-specific, but can be seen in some tumors, particularly meningiomas. Identification of neoplastic cells in CSF is unusual. It is most likely in lymphoma and choroid plexus tumors (see Chapter 12). Increased protein in the absence of pleocytosis (‘albuminocytological dissociation’) is indicative of noninflammatory disorders such as neoplasia or vascular disease. Bacteria may be seen in CSF occasionally (4.1B). These may be pathogenic, but if present in the absence of a neutrophilic pleocytosis, they are usually the result of reagent contamination. Intracellular bacteria should always be viewed as being significant (Munana, 1996).
Normal findings in the face of disease Many animals with CNS disease, including spinal diseases, have normal CSF (Thomson et al., 1989, 1990; Chrisman, 1992; Cellio, 2001). This is particularly so if the CSF sample is collected cranial to the lesion; that is, from the CMC in thoracolumbar and lumbosacral lesions. Thus, finding normal CSF does not rule out the possibility of a lesion being present.
Abnormal CSF findings Abnormal CSF is indicative of CNS (or nerve root) disease (Murray and Cuddon, 2002). High numbers of RBCs in the CSF are generally indicative of contamination at collection. Hemorrhage in the CNS may be inferred from the presence of xanthochromia (4.1A) or erythrophagocytosis (macrophages containing RBCs).
RADIOGRAPHY Survey radiographs Radiography is a valuable diagnostic aid for spinal patients. Good quality survey radiographs will provide the diagnosis in some cases. To obtain diagnostic radiographs of the spine, general anesthesia is usually
45
46
Small Animal Spinal Disorders
4.3 Radiographic positioning aids. A: Radiolucent foam wedges covered in plastic, ‘floppy’ sandbags and ties. B: Foam wedges, ties, ‘floppy’ sandbag along with 2.5 cm tape and a trough (see examples of use in 4.4–4.14).
A
B
4.4 Lateral cervical spine. The patient is in lateral recumbency with the head extended and the thoracic limbs pulled caudally. A wedge is placed under the sternum to prevent rotation (visible in 4.10). It is usually best to take two views, one centered at C2 and one at C6.
necessary to position the patient correctly and to minimize radiation exposure of personnel. In many animals with spinal disease, the radiological features are subtle so that accurate positioning is essential. An exception is in acute spinal injuries, where general anesthesia or manipulations could exacerbate the neural damage. The area of interest must be centered properly and the beam collimated closely. Survey radiographs of large areas of the spine are rarely useful. Stress radiography may be helpful but is not without risk (Farrow, 1982) (9.5, 10.12, 11.10, 13.17).
Radiographic positioning and normal spinal radiographs Various positioning aids facilitate spinal radiography (4.3). Techniques for radiographing the spine are given in the following section.
CERVICAL SPINE (4.4–4.9) Stressed survey views of the cervical spine are not generally of value except when done with care to assess the atlantoaxial joint (9.3).
4.5 Lateral cervical spine. A foam wedge is placed under the middle part of the neck to avoid sagging, which could result in distortion of the intervertebral spaces.
a b
4.6 Lateral cervical radiograph of a normal dog. Note that the transverse processes are superimposed (a); the intervertebral spaces near the middle of the film are clear and the end plates are parallel to the beam. The transverse processes of C6 are large and project ventrally (b), (7.37B). See 4.16 for cranial cervical radiograph. See also 11.13B.
ATLANTOAXIAL JOINT (4.8, 4.9) Projections are as for the cervical spine with the beam centered appropriately. For the lateral projection, a mild degree of flexion may be employed to demonstrate
Diagnostic aids
a
4.9 Lateral atlantoaxial radiograph of a normal dog. Note the normal relationship between the dorsal arch of C1 and the spinous process of C2. (a) Wing of atlas. Rotating the head slightly makes the dens more visible.
4.7 Ventrodorsal cervical spine. The patient is placed in dorsal recumbency, with the whole body aligned vertically. The beam is centered on the area of interest. It is useful to remove the endotracheal tube for this view, particularly in myelography.
4.8 Ventrodorsal atlantoaxial radiograph of a normal dog. Note the dens (arrow) and the atlantoaxial articulations (arrowheads).
subluxation, but this must not be excessive (9.3, 9.5). The use of the open-mouth view to radiograph the dens is not recommended, as flexing the neck can be dangerous. The ventrodorsal projection is adequate and the cervical extension also alleviates cord compression. The spinal deviation present in many patients with atlantoaxial subluxation can make positioning difficult. A common error is to try to evaluate the atlantoaxial joint in poorly positioned films taken of a conscious
4.10 Lateral thoracolumbar spine. The patient is in lateral recumbency with the limbs positioned as shown. Foam wedges are placed under the sternum (arrow) and between the limbs to prevent rotation and under the lumbar vertebrae to avoid sagging. The beam is centered on the area of interest and collimated closely over the spine to reduce soft tissue scatter and improve radiographic quality.
patient with the area of interest near the edge of the film. Accurate interpretation is often impossible (see Chapter 9).
THORACOLUMBAR SPINE (4.10–4.13) A common error is to center the beam in the midlumbar region when evaluating the spine for disc disease; most herniations occur in the T11–L1 region, and the beam should be centered accordingly.
47
48
Small Animal Spinal Disorders
4.11 Ventrodorsal thoracolumbar spine. The patient is in dorsal recumbency with the limbs extended. It is important that the body is upright and the beam collimated.
4.14 Lateral lumbosacral spine. Positioning is crucial here as rotation may induce artefacts. The patient is positioned as for the thoracolumbar spine, taking particular care to keep both pelvic limbs parallel to the table top. a b
4.12 Lateral thoracolumbar spinal radiograph of a normal dog. The dog is well positioned, with the ribs (a) and transverse processes (b) superimposed. The vertebral end plates in the middle of the image are parallel to the beam, whereas those near the right edge are not.
4.15 Lateral lumbosacral radiograph of a normal dog. In a well-positioned radiograph, the wings of the ilium will be directly superimposed.
LUMBOSACRAL SPINE (4.14–4.15)
SPECIAL RADIOGRAPHIC PROCEDURES Myelography
4.13 Ventrodorsal thoracolumbar spinal radiograph of a normal dog.
In myelography the spinal cord is outlined by positive contrast medium injected into the subarachnoid space. The indications for myelography are: • Where the neurological examination indicates a spinal lesion, but none is visible on survey radiographs. • To determine the significance of multiple lesions identified on survey radiographs. • To determine the presence of spinal cord compression and especially to evaluate dynamic lesions (10.7, 11.4–11.6). • To assist in deciding whether or not to perform surgery as well as the type of procedure to be performed.
Diagnostic aids
This procedure is contraindicated if general anesthesia or spinal puncture is unsafe (see ‘Cerebrospinal fluid’, page 42), or where inflammatory disease of the CNS is present. The presence of mild CSF pleocytosis is not in itself a contraindication, as it is present in many compressive spinal diseases, including disc herniation (Thomson et al., 1989, 1990). The choice of contrast medium is important, as many are extremely irritating to neural tissue. Hyperosmolar or hypertonic solutions must be avoided. A non-ionic, water-soluble contrast medium must be used—Iohexol (Omnipaque, Nycomed) is the contrast medium of choice (Wheeler and Davies, 1985; Allen and Wood, 1988). Concentrations in the range of 180–350 mg iodine/ml are used—most often 240 or 300 mg iodine/ ml. The contrast medium should be warmed to body temperature before injection (Lamb, 1994). Performance of a myelogram does not preclude the need to take high-quality survey radiographs. The practice of using the survey films only to establish radiographic exposure factors is deplored, as significant information that would be visible on good survey films may be masked by the myelogram. The region of interest is dictated by the neurological examination, remembering that occasionally a lower motor neuron (LMN) lesion can mask an upper motor neuron (UMN) lesion (2.24). Myelography is carried out with the patient under general anesthesia. Injection of contrast medium is either at the CMC or in the lumbar subarachnoid space. The techniques for spinal puncture are described above in CSF collection. Contrast is injected following collection of the CSF sample (4.64, 4.66).
4.16 Normal lateral myelogram of the cranial cervical region. Note that the contrast columns are wide in C1 and C2. The ventral column thins and is elevated over the C2/3 intervertebral space and to a lesser degree over the other spaces.
4.17 Normal lateral myelogram of the caudal cervical region.
CERVICAL MYELOGRAPHY (4.16, 4.17) Technique is described under ‘Procedures’, page 70. LUMBAR MYELOGRAPHY (4.18–4.20) Technique is described under ‘Procedures’, page 71. Lumbar myelography is required in many circumstances, the most common being an acute disc extrusion. Here, injection of contrast medium requires relatively high pressures to delineate the lesion, as the spinal cord is often swollen or under pressure. Some radiologists prefer lumbar myelography for all patients. Epidural leakage is often a problem in lumbar myelography, which makes interpretation of the study very difficult (4.67). Slow injection of the contrast medium may reduce the chance of epidural leakage. Another common complication is poor filling of the subarachnoid space, especially in animals with spinal cord swelling (8.1).
4.18 Normal lumbar lateral myelogram. Note the streaked appearance of the contrast medium where it outlines the nerves of the cauda equina.
4.19 Lateral lumbosacral myelogram from a normal dog. Note that the contrast column passes well into the sacrum.
49
50
Small Animal Spinal Disorders
Normal
Extradural
Intradural– extramedullary
4.20 Ventrodorsal lumbosacral myelogram of a normal dog. The separate contrast columns can be seen in L5 (arrows). The columns converge and cross the lumbosacral junction. Superimposition of the spinous processes and the presence of colonic contents can make interpretation difficult. Intramedullary
Epidurography In epidurography, contrast medium is introduced into the epidural space. For some radiologists this was the preferred method for evaluating the lumbosacral region (Barthez et al., 1994; Ramirez and Thrall, 1998) (10.8B). CT or MRI are superior in most cases.
Discography In discography, contrast medium is injected into the nucleus pulposus of the intervertebral disc (Sisson et al., 1992; Barthez et al., 1994). Normally, only a very small amount can be introduced (0.1–0.2 ml in the lumbosacral disc). If there is damage to the anulus fibrosus, more contrast can be injected and the leakage will be evident on subsequent radiographs (10.8A). The technique may be particularly useful if combined with CT (Ohnmeiss et al., 1997; Milette et al., 1999).
PRINCIPLES OF SPINAL RADIOLOGY It is important to take a systematic approach to evaluating spinal radiographs. The clinician has the advantage of having similar adjacent structures with which to compare the suspected lesion. Accuracy of interpretation is limited by the radiographs, and it is well worth paying attention to obtaining good images as discussed above. Also, knowledge of the normal radiographic anatomy is required, as discussed in this chapter and in Chapter 1. One system used to evaluate lateral radiographs is to assess each structure in turn from cranial to caudal: start ventrally with the hypaxial muscles and end with
4.21 Spinal lesions are classified according to their location relative to the spinal cord and the dura mater. Examples of potential causes are given in Table 4.2.
the spinous processes dorsally. For ventrodorsal radiographs, start on the left side and work cranial to caudal. This technique should be employed even if there is an obvious lesion, to avoid overlooking other important features or additional lesions. General radiological principles dictate that images are evaluated for the following features: position, size, number, contour, architecture and opacity. Many of these alterations are seen in spinal disease and most are shown in this book.
Myelographic interpretation From the myelogram, it is usually possible to gain an impression of the location of the lesion relative to the spinal cord (4.24–4.31, 14.4). Note that lesions in all locations may give the appearance of an expanded or swollen cord in an image taken ‘face on’ to the lesion. For this reason, it is essential that perpendicular projections (lateral and ventrodorsal) are made (4.30, 4.31, 7.2). Interpretation is relatively straightforward if there are changes in the skeleton at the site of the myelographic abnormality, for example in disc herniation or vertebral neoplasia (4.47). Generally, it is possible to make some estimate of the location of the lesion as shown in 4.21. Causes of these compression patterns are given in Table 4.2. Extradural lesions are shown in
Diagnostic aids
Table 4.2 Common causes of myelographic abnormalities
Extradural
Intradural– extramedullary
Intramedullary
Degenerative
Disc herniation Synovial cyst
—
—
Congenital
Atlantoaxial subluxation
Arachnoid cyst
Syringohydromyelia
Neoplasia
Primary or metastatic Vertebral or soft tissue
Meningioma Nerve sheath tumor Nephroblastoma
Glioma Ependymoma Metastatic
Inflammatory
Discospondylitis Epidural abscess
—
Myelitis
Trauma
Disc Bone fragments
—
Spinal cord hemorrhage
Vascular
Hematoma
—
Ischemic myelopathy (acute stage) Hematoma
4.23 Extradural lesion. Myelogram of a Doberman with ‘Wobbler syndrome’ and a large, extradural lesion due to what appears to be a simple disc extrusion at C6/7. The ventral contrast column is split (arrow and arrowhead) over the C6 vertebra (4.24, 4.25).
4.22 Extradural lesion. Diagram of the myelographic pattern of an extradural lesion.
4.22–4.25, 7.2, 8.1; intradural–extramedullary lesions are shown in 4.26–4.28; and intramedullary lesions are shown in 4.29–4.31. A CT scan often enhances the utility of the myelogram markedly and also aids surgical planning (4.25, 11.5).
EXTRADURAL LESIONS INTRADURAL–EXTRAMEDULLARY LESIONS INTRAMEDULLARY LESIONS Spinal cord swelling may be evident although myelography rarely delineates parenchymal lesions. One
exception is dogs with generalized malacia of the spinal cord (13.10, 14.18). CT reconstruction can further enhance lesion detection after myelography (4.41) but MRI is superior for most parenchymal lesions.
COMPLICATIONS OF MYELOGRAPHY The most common complications after cervical injection are either that the contrast has entered the subdural space (1.11); that the needle has been displaced from the vertebral canal; that excessive contrast has entered the cranial vault (4.65); or either that the contrast has not reached the lesion or that it has flowed past the lesion (Lamb, 1994; Penderis et al., 1999)(4.32). The most common problem after lumbar myelography is
51
52
Small Animal Spinal Disorders
4.24 Extradural lesion causing column splitting. Diagram of column splitting, which generally indicates an asymmetrical extradural lesion (4.25, 4.33).
4.26 Intradural–extramedullary lesion. Diagram of the typical ‘golf-tee’ pattern (arrows).
A
4.27 Intradural–extramedullary lesion. Ventrodorsal myelogram at L1 of a 10-year-old mixed-breed dog with recurrence of paraparesis after removal of a nerve sheath tumor at L3/4. There was evidence of spread along the subarachnoid space (4.28). Necropsy revealed an anaplastic sarcoma of meningeal origin. An MRI from this dog is shown in 4.44B.
B 4.25 Same dog as in 4.23. A: Column splitting was due to a probable left-sided synovial cyst (its location shifted slightly into C7 with the dog on its back for the CT). A ventral slot was done at C6/7 but little disc material was removed and the dog was much worse the next day. B: A postoperative CT myelogram showed more marked compression, due in part to collapse of the interspace. Compression in this dog was probably exacerbated by the synovial cyst (11.8). In retrospect this dog should have undergone a distraction-stabilization procedure.
4.28 Intradural–extramedullary lesions. Ventrodorsal myelogram of this dog is shown in 4.27. An MRI of this area is shown in 4.44B.
Diagnostic aids
4.31 Intramedullary lesion. Lateral myelogram of the dog shown in 4.30. The myelographic columns diverge in this view as well as in the ventrodorsal view. 4.29 Intramedullary lesion. Diagram showing spinal cord swelling.
4.32 Poor contrast filling is common over the caudal cervical spine. The cranial cervical and thoracic spines can be elevated in order to cause the heavier contrast agent to pool in the caudal cervical area (from McKee et al., 2000).
4.30 Intramedullary lesion. Ventrodorsal myelogram at L1/2 of a 7-year-old Doberman with progressive paraparesis of 2 weeks’ duration. The myelographic columns diverge in this ventrodorsal view, as they do in the lateral view (4.31). Diagnosis was a highly-invasive neurofibrosarcoma with extensive spinal cord invasion but only a very small mass around one dorsal nerve rootlet.
poor myelographic quality (8.1A); this can be due to several reasons. Epidural leakage may obscure the subarachnoid contrast medium (4.67). Fortunately, epidural contrast is usually absorbed more rapidly than the subarachnoid contrast medium and so resolution may improve after 10–20 min. Another potential problem is inadvertent injection into the subdural space rather
than the subarachnoid space. This is not uncommon— one study found a higher incidence with injections at the CMC (13%) than after lumbar injection (2.6%). Subdural contrast is usually restricted to the dorsal area and this is probably due to a restriction of ventral flow caused by the denticulate ligaments (Lamb, 1994, 1997; Scrivani et al., 1997; Penderis et al., 1999; Scrivani, 2000) (1.11). Contrast in the subarachnoid space may need to be pooled by manipulating the animal’s position (4.32). Positioning the animal in sternal recumbency with its head elevated slightly can cause contrast to collect in the caudal cervical area for a dorsoventral view; analogous pooling techniques can be applied to the lumbar spine. One final potential problem is that the lesion may only be visible when the animal is lying on one side but not when it is on the other side (4.33). Neurological deterioration can occur in some animals after myelography (Allen and Wood, 1988). Largebreed dogs with significant cervical cord lesions; dogs with chronic spinal cord compression, meningitis or extradural tumors; and those with degenerative myelopathy seem to be affected most often. Fortunately, this deterioration is usually transient, most patients returning to their pre-myelogram status within a few
53
54
Small Animal Spinal Disorders
4.33 Some lesions may be missed if only one lateral view is taken. A: There is a generalized loss of contrast over the L1/2 space in this radiograph taken with the dog in right lateral recumbency. B: Same dog in left lateral recumbency. There is marked splitting of the ventral contrast column due to an asymmetric lesion. This indicates that the lesion lies on the left side of the vertebral canal (Matteucci et al., 1999). A
B
4.34 Injection of contrast into the spinal cord parenchyma. This can cause severe neurological deficits and focal malacia, which is related to the volume injected. If injection is performed without fluoroscopic guidance, a test injection should be used (Parker et al., 1975; Servo and Laasonen, 1985; Kirberger and Wrigley, 1993).
days. Clearly, if the spinal puncture or injection technique itself is at fault, significant neurological damage may result (4.34). Injection of contrast into the central canal can occur in cisternal myelography; the effect on the patient varies but is generally serious. Cardiovascular effects are usually seen and neurological deterioration is likely. Central canal injection in the lumbar spine can cause temporary deterioration of deficits and the severity is related to the volume injected. The likelihood of this complication increases with injections at sites cranial to L5/6 (Kirberger and Wrigley, 1993). This is particularly true in small dogs where the spinal cord terminates more caudally than in large dogs (Morgan et al., 1987). Injection of contrast medium can be made between T13 and L2 vertebrae but the incidence of epidural leakage (4.67), and of central canal injection is then higher compared to more caudal injections (McCartney, 1997). Nevertheless, provided that a test injection is employed, this technique can be useful when injection is not successful at more caudal sites. It is also useful in dogs with lesions that localize to the L4–S3 spinal cord segments as there is then much less risk of injecting into the lesion itself; use of this site is much less desirable in dogs when the lesion is located in the thoracolumbar region.
Manipulations involved in obtaining flexion, extension or traction radiographs during myelography can lead to neurological deterioration and it is wise not to maintain these positions for excessive periods of time, particularly when extending the cervical spine (11.9, 11.10). Seizures occur infrequently on recovery from anesthesia following myelography (Wheeler and Davies, 1985; Allen and Wood, 1988). The site of injection appears not to influence the frequency of seizures; dogs with CSM may be prone to this complication (Lewis and Hosgood, 1992). Seizures are best managed by intravenous diazepam or barbiturates and so it is prudent to leave an intravenous catheter in place in any patient recovering after a myelogram to facilitate medication.
OTHER IMAGING TECHNIQUES Ultrasonography This technique has two broad indications. The first and most important is to evaluate the heart and abdomen. The abdomen should be examined in animals that might have concurrent lesions caused by trauma or by neoplastic, inflammatory or endocrine disease (4.35). Echocardiography is indicated in dogs with cardiac disease secondary to degenerative, neoplastic, inflammatory or traumatic disease. The second indication is to visualize the nervous system. This can be done either through natural portals such as a fontanelle or the atlantoaxial (4.36) and lumbosacral spaces. It can also be done through thin areas of the skull in toy breed dogs with hydrocephalus (a condition that may coexist with syringohydromyelia). In addition, the examination can be performed at surgery through a laminectomy defect or ventral slot. The main value of intraoperative ultrasonography is that the surgeon may be able to visualize a tumor or disc material that is obscured by the spinal cord. It can therefore serve as a real-time aid to localizing and defining the extent of a lesion. This could make the difference between a complete or incomplete tumor
Diagnostic aids
4.35 German shepherd dog that presented with mild neurological deficits and lumbosacral pain. The dog had small dermal nodules all over its skin suggestive of dermatofibrosis, an inherited condition that is associated with renal cystadenocarcinoma (Moe and Lium, 1997). Ultrasound revealed multiple masses within each kidney (arrow). This dog also had several other disorders (3.2, 10.4); it died 6 months later.
A
B
resection or prevent the surgeon leaving behind significant amounts of disc material (Nakayama, 1993; Finn Bodner et al., 1995; Hudson et al., 1995; Ham et al., 1995) (7.8, 8.6). Sonography has also proven useful in determining the extent of arachnoid cysts (Galloway et al., 1999). In addition, it can show irregular new bone on vertebrae affected by tumors or discospondylitis.
Computed tomography Computed tomography (CT) of the vertebral column is very useful in certain circumstances, particularly in patients with mineralized disc extrusions, vertebral tumors or cervical spondylomyelopathy. Contrast enhancement by myelography (7.3A), rather than intravenous contrast administration as performed in intracranial imaging, outlines the subarachnoid space if needed. A much lower dose of contrast medium is required for CT myelography than for conventional myelograms. If only a CT study is planned, the dose of contrast medium is reduced to about one quarter of the
4.36 Ultrasonography can be used to examine the spinal cord; normal C1 region seen through the atlantooccipital space. The dorsal and ventral dura mater is visible (arrows); the two lines ventrally probably represent the dura and pia mater. The hyperechoic area centrally represents the central canal; CSF in the subarachnoid space is anechoic (Finn Bodner et al., 1995). 4.37 A: Dog in dorsal recumbency, intubated and connected to a ventilator prior to undergoing a CT scan; this minimizes movement artefacts caused by breathing (for both CT and MRI). Breathing artefact is less of a problem for lesions in the cranial cervical or lumbosacral regions. B: Normal L1/2 intervertebral space of a dog to show epidural fat within each foramen (arrowheads). See also 1.23A. This pattern is lost after disc extrusion (4.40A).
dose for conventional myelography. If the CT study is performed directly after a conventional myelogram, the natural dilution and absorption of the contrast medium is sometimes adequate to reduce the contrast concentration in some animals or the patient can be positioned to allow contrast to flow away from the lesion (taking care that it does not run into the head). Animals are usually positioned in dorsal recumbency (4.37). Animals in ventral recumbency tend to lean to one side or the other, especially those with a narrow chest but some fractures may be more stable in this position (4.38). In general CT images should be made at right angles to the vertebral canal and not angled as shown in 4.39. For non-mineralized disc extrusions, either CT myelography or intravenous contrast-enhanced CT may be required to see the lesion (Sharp et al., 1995) (11.52). Use of contrast medium is usually unnecessary for chondrodystrophoid breeds of dog (8.2A, 8.3) and will also tend to obscure some of the more subtle features of disc disease. The study is therefore non-invasive and
55
56
Small Animal Spinal Disorders
A
4.38 Positioning for a CT scan demands the same precautions as for radiography; unstable lesions and dynamic studies warrant particular care. A: This dog had an L2 fracture (arrowhead), which was displaced markedly by putting the dog into dorsal recumbency (scout CT image). B: It was repositioned in ventral recumbency to reduce the displacement (arrowhead) and made a good recovery after surgery (13.48).
B
4.39 A: Cranial and caudal extent of extruded disc material shown on a scout image. This also identifies the exact disc or vertebra(e) involved, which is particularly useful in surgical planning. B: Transverse CT scan through the C2/3 disc to show the mineralized extrusion (arrow).
A
B 4.40 A: CT scan through the T12/13 intervertebral disc space from the same dog as shown in 4.37B. There is loss of the normal epidural fat pattern at the level of the right foramen (arrow). Disc extrusion was confirmed at surgery. B: Hemorrhage in the epidural or subdural space (arrow) following a disc extrusion in the thoracolumbar area. Often the rim of hyperattenuation is less distinct than this (Olby et al., 2000). Contrast has not been administered.
A
B
also very fast (5–10 min depending on the scanner). Although CT scans are usually more expensive than myelograms, the reduced anesthesia time means that the final cost is often similar. The region from caudal T9 vertebra to cranial L4 vertebra is scanned for dogs with lesions that localize to the T3–L3 spinal cord segments. The region from caudal L3 vertebra to mid-sacrum is scanned for dogs with lesions that localize to the L4–S3 spinal cord segments. Three-millimetre slices are usually adequate although additional 1 mm slices can be made through the region of interest (Olby et al., 2000). CT is more useful than myelography for surgical planning and can also be used to depict the cranial and caudal extent of disc material clearly (Olby et al., 1999) (4.39). Survey radiographs are recommended in addition to the CT scan in order to allow the surgeon to identify any anatomical anomalies (8.19–8.21).
Interpreting CT scans in dogs with thoracolumbar disc disease has been reviewed (Olby et al., 2000). Normal CT images are shown in Chapter 1 (1.17, 1.21, 1.22, 1.23A, 1.25, 1.28). Images should be viewed initially using soft-tissue parameters (4.40) and then subsequently using bone parameters (1.17). The normal spinal cord has intermediate attenuation on transverse images, equivalent to that of soft tissue such as adjacent kidney. Spinal cord causes more attenuation than epidural fat, which is usually most evident at the level of the disc space, especially adjacent to the intervertebral foramen (1.23A, 4.37B). Displacement of epidural fat at this level (4.40A), or an increase in the attenuation characteristics of the epidural fat, is a useful indicator of a disc extrusion. Another very common feature is the presence of a focal, heterogenous, hyperattenuating mass of mineralized disc material in an
Diagnostic aids
A
B
C
4.41 The A: lateral and B: ventrodorsal myelogram in this dog revealed slight expansion of the spinal cord over the cranial C5/6 disc space. The myelogram was followed by a CT scan, which was inconclusive (not shown). C: 3D reconstruction of the CT myelogram identified an expansile lesion within the spinal cord. The dog was euthanized; necropsy was not performed. Had MRI been available it would have been a more efficient way to demonstrate this lesion.
epidural location (8.3). A less common presentation occurs when disc material and blood are spread more diffusely along and around the spinal cord. In such cases the epidural mass is relatively small in individual transverse images and is only slightly more attenuating than the spinal cord. It does, however, still cause more attenuation than epidural fat, which is either displaced or infiltrated with blood and nuclear material (Olby et al., 2000). In some animals a rim of increased attenuation is visible around the spinal cord, which almost certainly represents epidural or subdural hemorrhage (Penderis et al., 1999; Olby et al., 2000; Tidwell et al., 2002) (4.40B). The exact appearance of hemorrhage may depend on how recent it is and if the hemoglobin has degraded (Tidwell et al., 1994). Adjacent chronic disc lesions, which are frequent and often incidental findings in dogs with acute extrusions, are recognized by their extreme hyperattenuation, which may approach that of cortical bone. The degree of attenuation can be quantified by measuring the Hounsfield units (CT numbers) within a defined region of interest. For dogs with suspected cervical disc extrusions, the scan extends from the mid-atlas (useful to help rule out atlantoaxial instability by demonstrating a normal dens) to cranial T1 vertebra. Attenuation characteristics of cervical disc extrusions in chondrodystrophoid breeds are similar to those seen in the thoracolumbar area (4.39B, 7.4A) except that extruded disc material rarely extends over more than one vertebra and diffuse, subdural hemorrhage is unusual. Reconstruction techniques can be used to enhance further the diagnostic value of CT in disc disease (7.4, 7.51) as well as for other conditions (4.41). Comparisons between CT and MRI in humans show that each has its own strengths and weaknesses (see also Chapter 10, pages 186–187). CT is superior for identifying (cervical) fractures whereas MRI is better for soft tissue
lesions (Cirillo et al., 1988; Kent et al., 1992; Klein et al., 1999). MRI is more accurate for showing the extent of medullary bone involvement in osteosarcoma (O’Flanagan et al., 1991); is superior to myelography and CT myelography in cervical myelopathy (Masaryk et al., 1986); is at least as accurate as CT myelography for identifying disc disease and for differentiating scar tissue from recurrent disc (Modic et al., 1984; Sotiropoulos et al., 1989; Yousem et al., 1992); and is superior to CT at identifying epidural abscess (Angtuaco et al., 1987). CT myelography is superior to MRI in detecting multiple myeloma (Mahnken et al., 2002); and in delineating meningeal carcinomatosis (Krol et al., 1988) CT myelography is equivalent to MRI for diagnosis of nerve root avulsion in brachial plexus injury (Doi et al., 2002), and is a very accurate means of identifying cervical disc herniations (Houser et al., 1995). CT usually gives superior spatial resolution to MRI, which can be important in identifying spondylosis and the lateral extent of disc extrusion (Karnaze et al., 1988; Jones et al., 2000a).
Magnetic resonance imaging Magnetic resonance imaging (MRI) has an important role in imaging spinal diseases, particularly in lumbosacral disease and parenchymal spinal cord lesions (deHaan, 1993; Kippenes et al., 1999). Normal anatomy and basic changes in canine disc disease have been reviewed (Sether et al., 1990a,b; Kippenes et al., 1999) (see pages 11–13). Imaging of spinal lesions is covered below and in the relevant chapters (7.3B, 8.2B, 9.7, 10.12, 12.3, 12.6B, 12.41A, 13.7, 14.3A, 14.15, 14.17, 14.19). The region of interest to be covered by the scan is dictated by the neurological localization, bearing in mind that occasionally a LMN lesion can obscure an UMN lesion (2.24). The animal is usually positioned in dorsal recumbency to minimize movement artefact
57
58
Small Animal Spinal Disorders
4.42 A: T2-weighted, sagittal MRI showing spinal cord edema (arrow) caused by an acute disc herniation at C5/6 in a dog with ‘Wobbler syndrome’. There is also a large, more chronic lesion at C6/7. Both disc spaces have reduced central signal intensity in this T2-weighted image. B: T1-weighted, transverse image at C5/6 to show the spinal cord (arrowhead) being compressed by a dorsally displaced, non-mineralized disc (arrow). Compare to 4.44A, 7.3B and 11.52. Some scanners indicate the level of the transverse image on each image (inset). A
B 4.43 A: Transverse, T1-weighted MRI of the L2/3 disc in an 11-year-old paraparetic Labrador (same dog as in 8.2B). The spinal cord is deformed by a large, somewhat right-sided, low signal mass. B: Corresponding image after gadoteridol (Prohance, Bracco Diagnostics, Mississauga, Ontario) administration. There is a rim of enhancement to the mass (arrow) (Vroomen et al., 1998; Saifuddin et al., 1999). Mini-hemilaminectomy with corpectomy was used to remove the chronic disc material.
A
B
from breathing (4.37A) and sagittal scans using T1 and T2 weighting are used to identify the lesion. T2weighted images highlight water and are especially useful for this as many lesions have an associated spinal cord edema that highlights the lesion clearly (4.42A, 13.7, 14.15, 14.19). Transverse images are then made through the region of interest (4.42B, 4.43, 4.44A). Intravenous contrast (gadolidium [Gd-DPTA, Magnavist, Berlex laboratories, Liberty Corner, NJ] or gadoteridol [Prohance, Bracco Diagnostics, Mississauga, Ontario] both used at 0.1–0.2 mmol/kg) is administered if needed, followed by another series of sagittal and transverse T1 scans (4.43B, 4.44B). Additional pulse sequences may also be added as required (Kippenes et al., 1999) (9.7B). The MRI features of cervical disc lesions in dogs are similar to those described in humans. Loss of signal intensity of the nucleus pulposus on T2-weighted images is a common finding but is also non-specific. A more reliable feature is displacement of epidural fat by extruded disc material as shown in both sagittal and transverse plane images. Extruded disc material has low signal intensity and it may alter the shape of the spinal cord in a transverse plane (Levitski et al., 1999b; Yamada et al., 2001). These features are also consistent
with the changes seen in the lumbosacral region (Adams et al., 1995) (see Chapter 10). In acute thoracolumbar disc extrusions there is often an associated hemorrhage at the site of extrusion secondary to tearing of the internal venous plexus (Sether et al., 1990a; Olby et al., 2000). Mineralized disc material may show as a signal void (i.e. hypointense) and the extradural spinal cord compression is usually caused by a combination of disc material and hematoma. The hematoma is initially signal hypointense on T1- and T2-weighted images but between 2 and 7 days after onset the signal becomes more hyperintense as deoxyhemoglobin is converted to methemoglobin (Tidwell et al., 2002). There is often only minimal contrast enhancement of disc material or hematoma but this can identify which lesion is more recent (Sether et al., 1990a; Tidwell and Jones, 1999) (4.43, 8.2B). The terminology used to describe disc lesions is discussed on page 12: • Herniation is a general term denoting a nonspecific type of disc abnormality and should not be used to indicate clinical significance. • Extrusion describes disc material that has clearly escaped the normal boundaries of the disc and is usually significant clinically.
Diagnostic aids
A
•
4.44 A: Transverse image of the dog shown in 7.3B. There is a large, hypointense extrusion of mineralized disc material (arrow) causing marked unilateral compression of the spinal cord (arrowhead). The inset shows the level of this image (C4–5). B: T1-weighted MRI of the dog shown in 4.27 and 4.28 after contrast. The enhancing mass (arrowhead) is causing marked compression of the spinal cord. It is not clear from the MRI if the tumor is extraor intra-medullary (Kippenes et al., 1999).
B
The terms ‘bulge’ and ‘protrusion’ are also nonspecific and a morphological description of the disc displacement is preferred. Criteria for differentiating clinically insignificant lesions from incidental, age-related changes (7.15A) are discussed below and in Chapter 10. It must be remembered that some disc herniations are incidental, age-related findings (4.42A) that are not responsible for the clinical signs (Jensen et al., 1994; Milette et al., 1999; Jones and Inzana, 2000b) (see Chapters 1, 10, page 188 and 11, page 213). MRI of people without back pain has shown that 50% have one disc herniation and 25% have two (Jensen et al., 1994). Interpretation must therefore be made using clinical signs, electromyographic (EMG) changes in local muscles, and imaging characteristics (Ramirez and Thrall, 1998). Indicators of clinical relevance in dogs include changes in posture or evidence of pain such as spontaneous vocalization or discomfort on palpation of local paralumbar muscles (Chrisman, 1975; Nardin et al., 1999). Overall, MRI shows excellent anatomical detail with reasonably high sensitivity but only low specificity, resulting in a high rate of false positives. In contrast to MRI, electromyography has high specificity and is therefore useful to confirm which lesions are physiologically significant, thereby avoiding unnecessary interventions (Nardin et al., 1999; Robinson, 1999). Imaging features of disc lesions that suggest physiological relevance include edema in bone marrow adjacent to the disc (Sether et al., 1990a; Morrison et al., 2000); edema within the spinal cord over the affected disc space (4.42, 13.7); contrast enhancement, which shows a high correlation with rupture of the anulus (Vroomen et al., 1998; Saifuddin et al., 1999) (4.43); and changes within the disc itself (Jenkins et al., 1985; Milette et al., 1999) (8.2B). MRI signal characteristics within the disc itself are useful in characterizing internal derangement, which in humans with low back pain can indicate that the lesion is symptomatic even in the absence of a change in disc contour. Decreased central disc signal intensity
on T2-weighted images is highly predictive of anular tears extending into or beyond the outer anulus and a majority of these are responsible for symptoms. A localized area of hypersignal on T2-weighted images is also a reliable marker of symptomatic, outer anular disruption. However, a disc can also have normal MRI characteristics and yet still be symptomatic (Milette et al., 1999). Changes in disc signal intensity on T2weighted images may prove to be useful for dogs and cats when a lesion does not involve clear extrusion of disc material, but this remains to be proven. MRI features of spinal cord tumors (4.44B, 12.6B, 12.41A) have been reviewed (Kippenes et al., 1999). Localization of the mass is initially made using sagittal T2weighted images. Identification of tumor extent and its relationship to surrounding tissues is best done with T1weighted images after contrast administration. Extradural tumors can usually be identified accurately although their signal characteristics and enhancement patterns are not consistent. Contrast enhancement of vertebral tumors may be of less value than for tumors in other locations. Assessment of bone infiltration is facilitated by use of additional pulse sequences such as fat suppression or T2-weighted gradient echo, which delineate tumor by its hyperintense signal characteristics. Determining the intradural–extramedullary compartment is not always straightforward and it is not uncommon for this classification to be inaccurate even though most of these tumors show marked contrast enhancement (4.44B). CSF imaging studies (MR myelography) may improve anatomic classification (Kippenes et al., 1999). Other indications for MRI include degenerative (Levitski et al., 1999a,b; Lipsitz et al., 2001; Webb et al., 2001), anomalous (14.3A), inflammatory (Kraft et al., 1998) (14.15), traumatic (13.7), and vascular (14.17, 14.19) lesions (Gopal and Jeffery, 2001).
Scintigraphy Bone scintigraphy evaluates certain functional aspects of bone, particularly related to blood supply and metabolic
59
60
Small Animal Spinal Disorders
4.45 Nuclear bone scan from a 10year-old German shepherd dog with paraparesis and back pain of 1 week duration. A: An osteoproductive lesion is visible at T5 (arrow); note also the soft tissue mass ventral to the affected vertebral body. B: Increased uptake of technetium is evident on the bone scan. Final diagnosis was osteosarcoma.
A
A
B
B
activity. Certain types of lesions will produce a ‘hot spot’ on a bone scan (4.45); this may be evident before radiographic changes are visible. Particular uses of bone scintigraphy include discospondylitis, vertebral tumors, and small pathological fractures secondary to osteoporosis (Lamb et al., 1990; Stefanacci and Wheeler, 1991; Cook et al., 2002).
CLINICAL ELECTROPHYSIOLOGY Electromyography Electromyography (EMG) is the method by which the electrical activity of muscle is studied and analyzed. This technique can assist the neurosurgeon by helping to localize some of the more subtle lesions as being either UMN or LMN in nature. If LMN deficits are present they will generally be associated with spontaneous electrical activity in those muscles supplied by the injured spinal cord segment or nerve root (4.46). It is important to remember that this spontaneous activity takes between 4 and 7 days to appear, which is the time taken for the axons to degenerate from the site of injury to the neuromuscular junctions (Cuddon et al., 2003). Two broad categories of LMN lesions can be seen. The first relates to lesions that are restricted to either the brachial or lumbosacral regions, such as brachial plexus avulsion injury or damage to the cauda equina. In this case, the spontaneous activity will be restricted to a discrete muscle group(s). Use of EMG in radiculopathy
4.46 Spontaneous electrical activity recorded from the gastrocnemius muscle of a 7-month-old Sharpei with protozoal polyradiculoneuritis and myositis. The dog was paraplegic with LMN deficits. A: The two main types of spontaneous activity are visible: 1. Fibrillation potential. 2. Positive sharp wave. The dog tested positive for Neospora caninum; creatine kinase (CK) was elevated. B: Muscle biopsy identified numerous protozoal organisms.
has high specificity and as such is a very useful complement to MRI. In particular, the combination of these two tests is one way to overcome the high rate of falsepositive diagnosis that can occur using MRI alone (Nardin et al., 1999; Robinson, 1999). The second category of LMN disease is when the animal is suffering from a generalized peripheral neuropathy, or possibly a myopathy, in which case spontaneous activity will be widespread throughout the body and will not be restricted to any particular muscle group. If an UMN lesion is present, electromyography of the paraspinal muscles can sometimes be useful to help localize the lesion. Although UMN deficits are evident mainly through their effects on white matter tracts, they will also impair neurons in the local gray matter at the same level. Damage to these neurons often produces spontaneous activity in epaxial and hypaxial spinal muscles adjacent to the injured spinal cord segment. This activity seems to occur as a result of both degenerative and irritative effects, so the latter may be evident on an acute basis (Chrisman, 1975).
Spinal cord evoked response In this technique, a recording electrode is placed on the dorsal lamina of a cervical, thoracic, or lumbar vertebra in order to detect an electrical response in the spinal cord. The impulse in the spinal cord is created by stimulating a peripheral sensory nerve distal to the lesion. The impulse or evoked response needs to undergo signal
Diagnostic aids
averaging to remove background information, much like a brainstem auditory evoked response. The spinal cord evoked response has the potential to give prognostic information about the functional state of the spinal cord after trauma. The technique is performed routinely in humans undergoing surgery for scoliosis and can reduce the incidence of postoperative paraparesis and paraplegia by more than 60% (Dawson et al., 1991). Various parameters can be recorded from the evoked waveforms traveling in the spinal cord, but no single one has proved to be predictive of the eventual outcome in dogs with acute spinal cord injuries. Mathematical manipulations of the data may provide useful information but it can be difficult to identify and interpret the waveforms accurately in dogs with severe spinal cord injuries, as the waveforms are often small and dispersed (Sylvestre et al., 1993). It would be a tremendous advantage to have a more objective criterion than nociception to predict the eventual outcome in an animal with severe spinal cord injury. The main challenges are the need for one person to be dedicated solely to the monitoring technique, the variation in size of canine and feline patients, and the fact that pre-existing neurological deficits make it hard to get useful potentials (Cuddon et al., 2003). Further study is required before the spinal cord evoked response can make a significant contribution to the management of clinical cases (Holliday, 1992; Poncelet et al., 1993; Cuddon et al., 2003).
A
B 4.47 A: Radiograph of the lumbar spine of a 4-year-old Golden retriever with a 2-week history of lumbar pain and asymmetrical paraparesis. Loss of bone within the L4 vertebral body is seen. B: Myelogram of the same dog showing ventral extradural compression. A spinal radiographic survey, bone scan and thoracic radiographs revealed no other lesions.
F waves and cord dorsum potentials F waves are probably of more value than peripheral motor nerve conduction studies for most spinal diseases. They represent delayed action potentials that arrive a few milliseconds after the compound muscle action potential. F waves are initiated by antidromic conduction of a stimulus along a motor axon that elicits an action potential from the ventral horn cell in the spinal cord. This secondary action potential is then conducted orthodromically down the motor axon in the ventral nerve root and peripheral nerve to produce a second muscle-evoked action potential, which is known as an F wave. These waveforms therefore provide a specific way to assess nerve roots; the longer path taken by the F wave also means that any abnormalities tend to be magnified compared to standard nerve conduction studies. They are particularly useful in assessing nerve root disease in cauda equina syndrome (Cuddon et al., 2003) (see Chapter 10). Cord dorsum potentials are a spinal-cord evoked response that arise from purely sensory input of peripheral nerves. They are useful both for evaluating radiculopathies and also for myelopathies that involve the intumescences (Cuddon et al., 2003).
4.48 Fine needle aspiration was performed under fluoroscopy. The material collected was diagnostic of myeloma. The dog was treated by radiation and chemotherapy (4.49).
BIOPSY
(4.47–4.49)
Tumors are the most likely reason for biopsy and are discussed in Chapter 12. Most biopsies of spinal tissue will be performed following surgical exposure of the lesion.
61
62
Small Animal Spinal Disorders
4.49 Same dog as shown in 4.47–4.48, above. A: Sixty five months after therapy it was euthanized due to a pathologic fracture of C7 caused by recurrence of myeloma. B: The original lesion at L4 (arrowhead) had not changed although neoplastic plasma cells were still present at this location as well as in the spleen and kidney (Rusbridge et al., 1999).
A
B
Surgical exposure can sometimes be avoided by fine needle aspiration of tissue from within the vertebral canal or vertebral body, preferably under fluoroscopic (4.48) or CT (12.4) guidance (Irving and McMillan, 1990). A core of vertebral bone can also be taken using a Jamshidi needle (5.35).
Key issues for future investigation Can MRI signal characteristics and contrast enhancement be used to differentiate disc lesions that are significant clinically from those that are not significant?
REFERENCES Adams, W.H., Daniel, G.B., Pardo, A.D., Selcer, R.R. (1995) Magnetic resonance imaging of the caudal lumbar and lumbosacral spine in 13 dogs (1990–1993). Veterinary Radiology and Ultrasound 36, 3–13. Allen, G.S., Wood, A.K.W. (1988) Iohexol myelography in the dog. Veterinary Radiology 29, 78–82. Angtuaco, E.J., McConnell, J.R., Chadduck, W.M., Flanigan, S. (1987) MR imaging of spinal epidural sepsis. American Journal of Roentgenology 149, 1249–1253. Barthez, P.Y., Morgan, J.P., Lipsitz, D. (1994) Discography and epidurography for evaluation of the lumbosacral junction in dogs with cauda equina syndrome. Veterinary Radiology and Ultrasound 35, 152–157. Bienzle, D., McDonnell, J.J., Stanton, J.B. (2000) Analysis of cerebrospinal fluid from dogs and cats after 24 and 48 hours of storage. Journal of the American Veterinary Medical Association 216, 1761–1764. Braund, K.G., Sharp, N.J.H. (2003) Neurological examination and localization. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1092–1107. Philadelphia: Elsevier Science. Cellio, B.C. (2001) Collecting, processing, and preparing cerebrospinal fluid in dogs and cats. Compendium on Continuing Education for the Practicing Veterinarian 23, 786–794. Chrisman, C.L. (1975) Electromyography in the localization of spinal cord and nerve root neoplasia in dogs and cats. Journal of the American Veterinary Medical Association 166, 1074–1079. Chrisman, C.L. (1992) Cerebrospinal fluid analysis. Veterinary Clinics of North America, Small Animal Practice 22, 781–810.
Cirillo, S., Simonetti, L., La Tessa, G., Elefante, R., Smaltino, F. (1988) [MR vs CT: which is the diagnostic advantage in neuroradiology]. Radiologia Medica (Torino) 76, 390–398. Cook, G.J., Hannaford, E., See, M., Clarke, S.E., Fogelman, I. (2002) The value of bone scintigraphy in the evaluation of osteoporotic patients with back pain. Scandinavian Journal of Rheumatology 31, 245–248. Cuddon, P.A., Murray, M., Kraus, K. (2003) Electrodiagnosis. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1108–1117. Philadelphia: Elsevier Science. Dawson, E.G., Sherman, J.E., Kanim, L.E., Nuwer, M.R. (1991) Spinal cord monitoring. Results of the Scoliosis Research Society and the European Spinal Deformity Society survey. Spine 16, S361–364. De Haan, J.J., Shelton, S.B., Ackerman, N. (1993) Magnetic resonance imaging in the diagnosis of degenerative lumbosacral stenosis in four dogs. Veterinary Surgery 22, 1–4. Doi, K., Otsuka, K., Okamoto, Y., Fujii, H., Hattori, Y., Baliarsing, A.S. (2002) Cervical nerve root avulsion in brachial plexus injuries: magnetic resonance imaging classification and comparison with myelography and computerized tomography myelography. Journal of Neurosurgery 96, 277–284. Duque, C., Parent, J., Bienzle, D. (2002) The immunophenotype of blood and cerebrospinal fluid mononuclear cells in dogs. Journal of Veterinary Internal Medicine 16, 714–719. Evans, R.E. (1989) Haematology, biochemistry, cerebrospinal fluid analysis and other clinicopathological investigations. In: S.J. Wheeler (ed.), Manual of Small Animal Neurology, 49–62. Cheltenham: BSAVA Publications. Farrow, C.S. (1982) Stress radiography: applications in small animal practice. Journal of the American Veterinary Medical Association 181, 777–784. Finn Bodner, S.T., Hudson, J.A., Coates, J.R., Sorjonen, D.C., Simpson, S.T., Cox, N.R., Wright, J.C., Garrett, P.D., Steiss, J.E., Vaughn, D.M., Miller, S.C., Brown, S.A. (1995) Ultrasonographic anatomy of the normal canine spinal cord and correlation with histopathology after induced spinal cord trauma. Veterinary Radiology and Ultrasound 36, 39–48. Galloway, A.M., Curtis, N.C., Sommerlad, S.F., Watt, P.R. (1999) Correlative imaging findings in seven dogs and one cat with spinal arachnoid cysts. Veterinary Radiology and Ultrasound 40, 445–452. Gopal, M.S., Jeffery, N.D. (2001) Magnetic resonance imaging in the diagnosis and treatment of a canine spinal cord injury. Journal of Small Animal Practice 42, 29–31. Ham, L.M.v., Bree, H.J.v., Tshamala, M., Thoonen, H. (1995) Use of computed tomography and computed tomographic myelography for assessment of spinal tumoral calcinosis in a dog. Veterinary Radiology and Ultrasound 36, 115–118. Holliday, T.A. (1992) Electrodiagnostic examination. Somatosensory evoked potentials and electromyography. Veterinary Clinics of North America, Small Animal Practice 22, 833–857.
Diagnostic aids
Houser, O.W., Onofrio, B.M., Miller, G.M., Folger, W.N., Smith, P.L. (1995) Cervical disk prolapse. Mayo Clinic Proceedings 70, 939–945. Hudson, J.A., Finn Bodner, S.T., Coates, J.R., Sorjonen, D.C., Simpson, S.T., Wright, J.C., Steiss, J.E., Cox, N.R., Vaughn, D.M., Miller, S.C., Brown, S.A., Garrett, P.D. (1995) Color doppler imaging and doppler spectral analysis in the spinal cord of normal dogs. Veterinary Radiology and Ultrasound 36, 542–547. Irving, G., McMillan, M.C. (1990) Fluoroscopically guided percutaneous fine-needle aspiration biopsy of thoracolumbar spinal lesions in cats. Progress in Veterinary Neurology 1, 473–475. Jenkins, J.P., Hickey, D.S., Zhu, X.P., Machin, M., Isherwood, I. (1985) MR imaging of the intervertebral disc: a quantitative study. British Journal of Radiology 58, 705–709. Jensen, M., Brant, Z.M., Obuchowski, N., Modic, M., Malkasian, D., Ross, J. (1994) Magnetic resonance imaging of the lumbar spine in people without back pain. New England Journal of Medicine 331, 69–73. Jones, J.C., Banfield, C.M., Ward, D.L. (2000a) Association between postoperative outcome and results of magnetic resonance imaging and computed tomography in working dogs with degenerative lumbosacral stenosis. Journal of the American Veterinary Medical Association 216, 1769–1774. Jones, J.C., Inzana, K.D. (2000b) Subclinical CT abnormalities in the lumbosacral spine of older large-breed dogs. Veterinary Radiology and Ultrasound 41, 19–26. Karnaze, M.G., Gado, M.H., Sartor, K.J., Hodges, F.J., 3rd (1988) Comparison of MR and CT myelography in imaging the cervical and thoracic spine. American Journal of Roentgenology 150, 397–403. Kent, D.L., Haynor, D.R., Larson, E.B., Deyo, R.A. (1992) Diagnosis of lumbar spinal stenosis in adults: a metaanalysis of the accuracy of CT, MR, and myelography. American Journal of Roentgenology 158, 1135–1144. Kippenes, H., Gavin, P.R., Bagley, R.S., Silver, G.M., Tucker, R.L., Sande, R.D. (1999) Magnetic resonance imaging features of tumors of the spine and spinal cord in dogs. Veterinary Radiology and Ultrasound 40, 627–633. Kirberger, R.M., Wrigley, R.H. (1993) Myelography in the dog: review of patients with contrast medium in the central canal. Veterinary Radiology and Ultrasound 34, 253–258. Klein, G.R., Vaccaro, A.R., Albert, T.J., Schweitzer, M., Deely, D., Karasick, D., Cotler, J.M. (1999) Efficacy of magnetic resonance imaging in the evaluation of posterior cervical spine fractures. Spine 24, 771–774. Kraft, S.L., Mussman, J.M., Smith, T., Biller, D.S., Hoskinson, J.J. (1998) Magnetic resonance imaging of presumptive lumbosacral discospondylitis in a dog. Veterinary Radiology and Ultrasound 39, 9–13. Krol, G., Sze, G., Malkin, M., Walker, R. (1988) MR of cranial and spinal meningeal carcinomatosis: comparison with CT and myelography. American Journal of Roentgenology 151, 583–588. Lamb, C., Berg, J., Bengtson, A. (1990) Preoperative measurement of canine primary bone tumors, using radiography and bone scintigraphy. Journal of the American Veterinary Medical Association 196, 1474–1478. Lamb, C.R. (1994) Common difficulties with myelographic diagnosis of acute intervertebral disc prolapse in the dog. Journal of Small Animal Practice 35, 549–558. Lamb, C.R. (1997) Subdural collection of contrast medium during cervical myelography. Veterinary Radiology and Ultrasound 38, 479–481. Lavely, J., Vernau, K.M., LeCouteur, R.A. (2002) Spinal epidural empyema in seven dogs. Journal of Veterinary Internal Medicine 16, 369. Levitski, R., Chauvet, A., Lipsitz, D. (1999a) Cervical myelopathy associated with extradural synovial cysts in 4 dogs. Journal of Veterinary Internal Medicine 13, 181–186. Levitski, R.E., Lipsitz, D., Chauvet, A.E. (1999b) Magnetic resonance imaging of the cervical spine in 27 dogs. Veterinary Radiology and Ultrasound 40, 332–341. Lewis, D.D., Hosgood, G. (1992) Complications associated with the use of iohexol for myelography of the cervical vertebral column in dogs: 66 cases (1988–1990). Journal of the American Veterinary Medical Association 200, 1381–1384. Lipsitz, D., Levitski, R.E., Chauvet, A.E., Berry, W.L. (2001) Magnetic resonance imaging features of cervical stenotic myelopathy in 21 dogs. Veterinary Radiology and Ultrasound 42, 20–27. Mahnken, A.H., Wildberger, J.E., Gehbauer, G., Schmitz-Rode, T., Blaum, M., Fabry, U., Gunther, R.W. (2002) Multidetector CT of the
spine in multiple myeloma: comparison with MR imaging and radiography. American Journal of Roentgenology 178, 1429–1436. Masaryk, T.J., Modic, M.T., Geisinger, M.A., Standefer, J., Hardy, R.W., Boumphrey, F., Duchesneau, P.M. (1986) Cervical myelopathy: a comparison of magnetic resonance and myelography. Journal of Computer Assisted Tomography 10, 184–194. Matteucci, M.L., Ramirez, O., III, Thrall, D.E. (1999) Radiographic diagnosis: effect of right vs. left lateral recumbency on myelographic appearance of a lateralized extradural mass. Veterinary Radiology and Ultrasound 40, 351–352. McCartney, W.T. (1997) Lumbar myelography in 79 dogs, using different puncture sites. Veterinary Record 141, 417–419. McKee, W., Penderis, J., Dennis, R. (2000) Radiology corner: obstruction of contrast medium flow during cervical myelography. Veterinary Radiology and Ultrasound 41, 342–343. Milette, P.C., Fontaine, S., Lepanto, L., Cardinal, E., Breton, G. (1999) Differentiating lumbar disc protrusions, disc bulges, and discs with normal contour but abnormal signal intensity. Magnetic resonance imaging with discographic correlations. Spine 24, 44–53. Modic, M.T., Pavlicek, W., Weinstein, M.A., Boumphrey, F., Ngo, F., Hardy, R., Duchesneau, P.M. (1984) Magnetic resonance imaging of intervertebral disk disease. Clinical and pulse sequence considerations. Radiology 152, 103–111. Moe, L., Lium, B. (1997) Hereditary multifocal renal cystadenocarcinomas and nodular dermatofibrosis in 51 German shepherd dogs. Journal of Small Animal Practice 38, 498–505. Morgan, J.P., Atilola, M., Bailey, C.S. (1987) Vertebral canal and spinal cord mensuration: a comparative study of its effect on lumbosacral myelography in the Dachshund and German Shepherd dog. Journal of the American Veterinary Medical Association 191, 951–957. Morrison, J.L., Kaplan, P.A., Dussault, R.G., Anderson, M.W. (2000) Pedicle marrow signal intensity changes in the lumbar spine: a manifestation of facet degenerative joint disease. Skeletal Radiology 29, 703–707. Munana, K.R. (1996) Encephalitis and meningitis. Veterinary Clinics of North America, Small Animal Practice 26, 857–874. Murray, M., Cuddon, P.A. (2002) Cerebrospinal fluid analysis in acute canine polyradiculoneuritis: albumin quotient and immunoglobulin index determination using polyacrylamide gel electrophoresis. Journal of Veterinary Internal Medicine 16, 370. Nakayama, M. (1993) Intraoperative spinal ultrasonography in dogs: normal findings and case-history reports. Veterinary Radiology and Ultrasound 34, 264–268. Nardin, R.A., Patel, M.R., Gudas, T.F., Rutkove, S.B., Raynor, E.M. (1999) Electromyography and magnetic resonance imaging in the evaluation of radiculopathy. Muscle Nerve 22, 151–155. O’Flanagan, S.J., Stack, J.P., McGee, H.M., Dervan, P., Hurson, B. (1991) Imaging of intramedullary tumor spread in osteosarcoma. A comparison of techniques. The Journal of Bone and Joint Surgery—British volume 73, 998–1001. Ohnmeiss, D.D., Vanharanta, H., Ekholm, J. (1997) Degree of disc disruption and lower extremity pain. Spine 22, 1600–1605. Olby, N.J., Munana, K.R., Sharp, N.J.H., Flegel, T., Van Camp, S., Berry, C.R., Thrall, D.G. (1999) A comparison of computed tomography and myelography in the diagnosis of acute intervertebral disc disease in dogs. Journal of Veterinary Internal Medicine 13, 239. Olby, N.J., Munana, K.R., Sharp, N.J.H., Thrall, D.E. (2000) The computed tomographic appearance of acute thoracolumbar intervertebral disc herniations in dogs. Veterinary Radiology and Ultrasound 41, 396–402. Parker, A.J., Cusick, P.K., Stowater, J.L., Park, R.D. (1975) Effects of intramedullary injection of methiodal and lidocaine on spinal cords of dogs. American Journal of Veterinary Research 36, 1529–1532. Penderis, J., Sullivan, M., Schwarz, T., Griffiths, I.R. (1999) Subdural injection of contrast medium as a complication of myelography. Journal of Small Animal Practice 40, 173–176. Poncelet, L., Michaux, C., Balligand, M. (1993) Somatosensory potentials in dogs with naturally acquired thoracolumbar spinal cord disease. American Journal of Veterinary Research 54, 1935–1941. Radaelli, S.T., Platt, S.R. (2002) Bacterial meningoencephalomyelitis in dogs: a retrospective study of 23 cases (1990–1999). Journal of Veterinary Internal Medicine 16, 159–163.
63
64
Small Animal Spinal Disorders
Ramirez, O., III, Thrall, D.E.(1998) A review of imaging techniques for canine cauda equina syndrome. Veterinary Radiology and Ultrasound 39, 283–296. Remedios, A.M., Wagner, R., Caulkett, N.A., Duke, T. (1996) Epidural abscess and discospondylitis in a dog after administration of a lumbosacral epidural analgesic. Canadian Veterinary Journal 37, 106–107. Robinson, L.R. (1999) Electromyography, magnetic resonance imaging, and radiculopathy: it’s time to focus on specificity. Muscle Nerve 22, 149–150. Rossmeissl, J.H., Troy, G.C., Inzana, K.D., Jortner, B., Boon, G.D. (2002) Effects of blood contamination on canine cerebrospinal fluid white blood cell counts and total protein concentrations. Journal of Veterinary Internal Medicine 16, 370. Rusbridge, C., Wheeler, S.J., Lamb, C.R., Page, R.L., Carmichael, S., Brearley, M.J., Bjornson, A.P. (1999) Vertebral plasma cell tumors in 8 dogs. Journal of Veterinary Internal Medicine 13, 126–133. Saifuddin, A., Mitchell, R., Taylor, B.A. (1999) Extradural inflammation associated with annular tears: demonstration with gadolinium-enhanced lumbar spine MRI. European Spine Journal 8, 34–39. Scrivani, P.V. (2000) Myelographic artifacts. Veterinary Clinics of North America, Small Animal Practice 30, 303–314, vi. Scrivani, P.V., Barthez, P.Y., Leveille, R., Schrader, S.C., Reed, S.M. (1997) Subdural injection of contrast medium during cervical myelography. Veterinary Radiology and Ultrasound 38, 267–271. Servo, A., Laasonen, E.M. (1985) Accidental introduction of contrast medium into the cervical spinal cord. A case report. Neuroradiology 27, 80–82. Sether, L.A., Nguyen, C., Yu, S.N., Haughton, V.M., Ho, K.C., Biller, D.S., Strandt, J.A., Eurell, J.C. (1990a) Canine intervertebral disks: correlation of anatomy and MR imaging. Radiology 175, 207–211. Sether, L.A., Yu, S., Haughton, V.M., Fischer, M.E. (1990b) Intervertebral disk: normal age-related changes in MR signal intensity. Radiology 177, 385–388. Sharp, N.J.H., Cofone, M., Robertson, I.D., DeCarlo, A., Smith, G.K., Thrall, D.E. (1995) Computed tomography in the evaluation of caudal cervical spondylomyelopathy of the Doberman Pinscher. Veterinary Radiology and Ultrasound 36, 100–108. Sisson, A.F., LeCouteur, R.A., Ingram, J.T., Park, R.D., Child, G. (1992) Diagnosis of cauda equina abnormalities by using electromyography, discography, and epidurography in dogs. Journal of Veterinary Internal Medicine 6, 253–263. Sotiropoulos, S., Chafetz, N.I., Lang, P., Winkler, M., Morris, J.M., Weinstein, P.R., Genant, H.K. (1989) Differentiation between postoperative scar and recurrent disk herniation: prospective comparison of MR, CT, and contrastenhanced CT. American Journal of Neuroradiology 10, 639–643. Stefanacci, J.D., Wheeler, S.J. (1991) Skeletal scintigraphy in canine discospondylitis. American College of Veterinary Radiology 66.
Sylvestre, A.M., Cockshutt, J.R., Parent, J.M., Brooke, J.D., Holmberg, D.L., Partlow, G.D. (1993) Magnetic motor evoked potentials for assessing spinal cord integrity in dogs with intervertebral disc diseases. Veterinary Surgery 22, 5–10. Thomson, C.E., Kornegay, J.N., Stevens, J.B. (1989) Canine intervertebral disc disease: changes in the cerebrospinal fluid. Journal of Small Animal Practice 30, 685–688. Thomson, C.E., Kornegay, J.N., Stevens, J.B. (1990) Analysis of cerebrospinal fluid from the cerebellomedullary and lumbar cisterns of dogs with focal neurologic disease: 145 cases (1985–1987). Journal of the American Veterinary Medical Association 196, 1841–1844. Tidwell, A.S., Jones, J.C. (1999) Advanced imaging concepts: a pictorial glossary of CT and MRI technology. Clinical Techniques in Small Animal Practice 14, 65–111. Tidwell, A.S., Mahony, O.M., Moore, R.P., Fitzmaurice, S.N. (1994) Computed tomography of an acute hemorrhagic cerebral infarct in a dog. Veterinary Radiology and Ultrasound 35, 290–296. Tidwell, A.S., Specht, A., Blaeser, L., Kent, M. (2002) Magnetic resonance imaging features of extradural hematomas associated with intervertebral disc herniation in a dog. Veterinary Radiology and Ultrasound 43, 319–324. Vroomen, P.C., Van Hapert, S.J., Van Acker, R.E., Beuls, E.A., Kessels, A.G., Wilmink, J.T. (1998) The clinical significance of gadolinium enhancement of lumbar disc herniations and nerve roots on preoperative MRI. Neuroradiology 40, 800–806. Webb, A.A., Pharr, J.W., Lew, L.J., Tryon, K.A. (2001) MR imaging findings in a dog with lumbar ganglion cysts. Veterinary Radiology and Ultrasound 42, 9–13. Wheeler, S.J., Davies, J.V. (1985) Iohexol myelography in the dog and cat: a series of one hundred cases, and a comparison with metrizamide and iopamidol. Journal of Small Animal Practice 26, 247–256. Widmer, W.R., DeNicola, D.B., Blevins, W.E., Cook, J.R., Jr, Cantwell, H.D., Teclaw, R.F. (1992) Cerebrospinal fluid changes after iopamidol and metrizamide myelography in clinically normal dogs. American Journal of Veterinary Research 53, 396–401. Wilson, J.W., Stevens, J.B. (1977) Effects of blood contamination on cerebrospinal fluid analysis. Journal of the American Veterinary Medical Association 171, 256–258. Yamada, K., Nakagawa, M., Kato, T., Shigeno, S., Hirose, T., Miyahara, K., Sato, M. (2001) Application of short-time magnetic resonance examination for intervertebral disc diseases in dogs. Journal of Veterinary Medical Science 63, 51–54. Yousem, D.M., Atlas, S.W., Hackney, D.B. (1992) Cervical spine disk herniation: comparison of CT and 3DFT gradient echo MR scans. Journal of Computer Assisted Tomography 16, 345–351.
PROCEDURES Collection of CSF General anesthesia is required for CSF collection in dogs and cats; the depth must be suitable so that the animal will not move during the procedure. Patients should be intubated and ventilator support must be available. The collection site must be clipped and prepared aseptically. Sterile surgical gloves should be worn. An assistant is required to hold the patient in the correct position when collecting from the cerebello-medullary cistern (CMC). Which collection site to use warrants some consideration? The two sites available are the CMC and the lumbar region. Collection from the CMC is easier and is less likely to produce a sample contaminated with blood. However, as CSF flows in a cranial to caudal direction, abnormal CSF is more likely to be present caudal to a lesion. Thus, lumbar CSF is more likely to be useful diagnostically. However, lumbar collection is more difficult and blood contamination occurs more often. If raised intracranial pressure is present, lumbar collection is somewhat safer. This is not because brain herniation is any less likely, but because the risk of direct damage to already-herniated tissue is not a factor with lumbar collection. If CSF collection is necessary when intracranial pressure could be elevated (see page 43), then
Diagnostic aids
the animal should be ventilated prophylactically and a capnometer used to maintain end-tidal pCO2 at 30–35 mmHg. Mannitol, lasix and dexamethasone should also be on hand for emergency use (see below). Spinal needles are preferred for CSF collection (4.50). The CSF is collected into sterile vials; plain vials without anticoagulant are generally used.
COLLECTION FROM THE CEREBELLOMEDULLARY CISTERN (4.50–4.58) Because the neck is flexed severely (4.51), a kink-proof endotracheal tube should be used. Alternatively, the cuff is deflated to allow space around the tube for ventilation should the tube obstruct. An imaginary line is drawn between the wings of the atlas (4.52, 4.53). The ideal site is in the midline of the patient, half way between the occipital protuberance and the line joining the wings of the atlas. Just behind the occipital protuberance, there is a slight depression in the muscles. This is not the site of needle penetration but is cranial to it; inserting the needle there usually leads to it striking the bone of the skull. The needle is inserted perpendicular to the skin in the midline (4.54, 4.58). In small dogs and especially in cats, it is much safer to pick up the skin and first penetrate the skin with the needle pointing away from the spinal cord. There are two main methods for advancing the needle: 1.
Remove the stilette once the tip of the needle enters muscle and then advance the needle until CSF appears in the hub of the needle.
2.
Advance the needle with the stilette still in place in small increments, removing the stilette between each movement to check whether CSF is present in the needle. A slight ‘pop’ can sometimes be felt when the needle enters the subarachnoid space but this sensation can also be produced by movements of the needle if the tip has been blunted on bone.
Method (1) is simpler for most clinicians and is less hazardous. If CSF does not flow or is blood tinged, a number of possible complications could have arisen: •
If the tip of the needle hits bone, it is most common for it to be too far rostral. The point of the needle is redirected caudally by withdrawing it to a subcutaneous position and moving the hub rostrally.
•
If CSF flows, but is tinged with a trickle of bright red blood in the base of the hub, a dural vessel has been penetrated. This may clear after a few seconds, and then the CSF may be collected. Rotating the needle helps to clear this type of hemorrhage. The initial blood contamination can be evacuated from the hub of the needle using a separate, small needle and syringe although it is often better to just remove the needle and start again.
•
Dark venous blood flows from the needle. This indicates that a venous structure has been penetrated, usually because the needle has strayed from the midline. This causes no harm but it is best to start again with a fresh needle.
•
No CSF flows, even though all the features of a successful tap appear to be present and the needle appears to have penetrated to an adequate depth. The stilette should be replaced in case the needle has plugged with tissue or blood clot. If CSF still does not flow, this may be because either the needle is far off the midline; the brain has herniated; a mass occupies part of the cisterna magna; or the spinal cord has been penetrated. The latter is a significant risk if the needle is advanced with the stilette in place. The needle should be removed, the patient placed in a normal position and the respiratory pattern observed. If respiration is normal after several minutes, the procedure may be repeated or a lumbar puncture performed. In some patients with soft tissue lesions in the cranial cervical vertebral canal, CSF cannot be collected from the CMC.
•
The patient moves suddenly; this is usually because the spinal cord has been damaged. The needle must be removed and the patient ventilated. This is a potentially serious situation leading to marked neurological deterioration or death. It may be useful to give the patient methylprednisolone sodium succinate (MPSS) (30 mg/kg IV) in an attempt to limit the spinal cord damage.
•
The CSF flows forcefully from the needle. It is not always clear why this happens, but it clearly indicates high CSF pressure at the CMC. Some dogs do not suffer any apparent deleterious effects from this, but others do deteriorate. The needle should be removed and the precautions mentioned above taken. It may
65
66
Small Animal Spinal Disorders
also be useful to hyperventilate the patient as discussed below if increased intracranial pressure is suspected as the cause. •
CSF enters the hub but then flow stops abruptly and the animal’s heart rate drops below 40/min. This indicates that brain herniation has occurred; the needle must be removed and the animal ventilated to maintain end-tidal pCO2 at 30–35 mmHg. Mannitol (1.0 g/kg) and furosemide should be administered intravenously. This strategy is often successful in reversing the herniation if instituted immediately; a capnometer and emergency drugs should therefore always be available and ready for immediate use for any patient considered to be at increased risk of herniating.
4.50 Spinal needles have a stilette and a shallow bevel. A notch in the hub indicates the side of the bevel, which in myelography should be pointing in the direction in which contrast is intended to flow.
4.50
4.51 Patient positioned in right lateral recumbency for collection of CSF by a right-handed operator. The head is held by an assistant in 90° flexion with the nose parallel to the tabletop. The neck is positioned close to the edge of the table. A foam wedge may be used to support the nose. 4.51
a
4.52 Diagram to show the site of CMC puncture. Landmarks are the lateral margin of the wings of the atlas (a), the occipital protuberance (b), and the midline (c). The location of the cisterna magna is outlined by contrast in 4.65.
b
c
4.52
a
4.53 Landmarks for CMC puncture. These are the lateral margin of the wings of the atlas (a), the occipital protuberance (b), and the midline.
a
b a
4.53
Diagnostic aids
4.54 Needle insertion. The operator is wearing sterile gloves, and the site has been prepared aseptically. The left hand is identifying the landmarks. The thumb and upper fingers are palpating the wings of the atlas. The middle finger is on the midline, just behind the intended site. The 4th or 5th fingers of the right hand should always be rested on the animal’s head in case the table or dog move inadvertently. 4.54
4.55 The needle is in place with the stilette removed; a drop of CSF is seen. Spinal needles become plugged only rarely. If plugging is suspected, the stilette should be inserted then removed.
4.55
4.56 Fluid is collected into a sterile vial.
4.56
4.57 Fluid can also be collected into a sterile syringe, which is not used to aspirate CSF but just to catch drops as they fall.
4.57
67
68
Small Animal Spinal Disorders
4.58 Diagram to show position of needle. The landmarks are as in 4.52 (a) and (b). Technique for injection of contrast at the end of CSF collection is shown in 4.64.
b
a
4.58
COLLECTION FROM THE LUMBAR SPINE (4.59–4.63) Collection of CSF from the lumbar site is more likely to fail than from the CMC. Contamination is also more frequent. Collection is usually from L4/5 or L5/6 in dogs (4.59); in the cat L6/7 can be used. If CSF does not flow or is blood tinged, a number of possible complications may have arisen: •
The needle strikes bone. The space available for penetration of the vertebral canal is relatively small, particularly in large dogs with significant vertebral new bone formation. Also, the ligamentous tissue may be mineralized and have the feel of bone. Repeated attempts may be required to position the needle and, in some patients, the technique may fail. Flexing the spine by drawing the pelvic limbs forward may open up the interarcuate spaces.
•
The considerations for hemorrhage are similar to those mentioned above. If the needle appears to be positioned correctly and yet CSF does not appear, the needle can be rotated slowly or withdrawn slightly. Provided that intracranial pressure is normal, one or both jugular veins may be occluded to cause CSF to flow. If these approaches are not successful then a second attempt at collection can be made. Aspiration of CSF with a syringe is possible, but blood contamination is more likely.
4.59 Lumbar puncture between L5 and L6. The tip of the L6 spinous process is palpated. The needle is inserted alongside the caudal edge of the L6 spinous process and directed cranially and ventrally through the ligamentum flavum into the vertebral canal (4.67).
4.59
Diagnostic aids
70
Small Animal Spinal Disorders
Myelography As a guide, a dose of iohexol contrast medium from 0.25 to 0.5 ml/kg body weight is used, although this can vary with the site of the injection and the expected location of the lesion. Better contrast may be obtained when using a concentration of 300 mg iodine/ml compared to 240 mg iodine/ml.
CERVICAL MYELOGRAPHY (4.64, 4.65) If possible the table is tilted by 5–10° at injection. The bevel of the needle (indicated by a notch on the hub) should be directed caudally. Injection should be performed preferably by connecting the syringe to the spinal needle using a flexible tube, which is pre-filled with contrast before injection. Contrast should ideally be filtered if multi-use vials are used routinely (4.1B). It is vital that the needle is not advanced into the spinal cord substance; use of an extension tube can help to avoid needle movement during injection and the needle should be held as described in 4.64. After injection the needle is removed and the patient’s head is elevated. The legs and maxilla are tied to the table edge using rope ties. Initially, the patient is tilted at 30° with the head up and an immediate radiograph is taken, further radiographs being taken depending on the progress of the contrast column. If the X-ray tube can be tilted parallel to the table, radiographs may be taken in this position. The angle of tilt may be increased to 45° or more and further radiographs taken until a lesion is demonstrated or the contrast ceases to flow. When the lesion is delineated, ventrodorsal and oblique views should be taken. If a tilting table is not available, the patient must be held up by the thoracic limbs, with the head above the body. This can be difficult in large dogs and use of a tilting table is recommended. In normal dogs, the column will usually reach the lumbosacral region in a maximum of 10 min on a suitably tilted table. Occasionally, the column stops in a totally unexpected position, often the low cervical region, and this should not be considered diagnostic until further time has elapsed and the table tilted to a steeper angle.
4.64 Position of patient and needle. In this patient, an extension tube has been applied to the spinal needle for the injection of contrast medium for myelography. The position of the right hand as shown is dangerous and NOT recommended. It is much safer to maintain the needle in position by resting the 4th and 5th finger of the right hand on the dog’s head as shown for CSF collection (4.54).
4.64
Diagnostic aids
4.65 Close-up of a myelogram with the neck flexed to show the atlanto-occipital space and the cerebellomedullary cistern distended with contrast (arrow). The cranial border of the atlas is shown (arrowhead). The cerebellar folia are outlined by contrast ( ); cranial flow of contrast * should be avoided by pointing the bevel caudally while injecting and then elevating the head promptly.
4.65
LUMBAR MYELOGRAPHY (4.66–4.68) To expedite imaging and minimize the amount of repositioning, the order for radiographs should be as follows: ventrodorsal survey radiographs; lateral survey radiographs, then injection of contrast with lateral images; followed by a final repositioning for ventrodorsal and oblique images. Ideally, injection should be made at the L5/L6 space as the incidence of complications increases with injections at sites cranial to this (Kirberger and Wrigley, 1993). Injections at L6/L7 are prone to epidural injection, although this tends to be true more in large dogs and L6/L7 can be more reliable than L5/L6 in small dogs (4.68). Epidural leakage may also be more likely when the pelvic limbs are flexed. The bevel of the needle (indicated by a notch on the hub) should be directed cranially. Injection may be performed directly by applying a syringe to the spinal needle, or via a flexible tube, which is pre-filled with contrast before injection. A test injection of 0.2–0.4 ml of contrast should always be made to insure that it is not being delivered into the spinal cord parenchyma (4.34). Provided that this is not the case, a lateral radiograph is then taken once most of the contrast dose has been injected but with the needle still in place. If the spinal cord is swollen then radiographs should ideally be taken as contrast is being injected in order to best outline the subarachnoid space. In such cases it is useful to take the dorsoventral radiograph by repositioning the animal with the needle still in situ to permit a further injection of contrast to be made. If the needle has been inserted with the pelvic limbs flexed, they should be maintained in this position as the animal is repositioned. Once the spinal needle has been removed, ventrodorsal and oblique radiographs should be taken rapidly as epidural leakage occurs through the puncture hole, especially when the spinal cord is markedly swollen. When epidural leakage does occur it tends to clear faster than the contrast in the subarachnoid space and so serial images should be taken in the hope that resolution improves; if it does not then the study will need to be repeated.
4.66 Contrast injection using an extension tube. Injection should be slow to avoid epidural leakage. The apparatus may need to be held in place to prevent the extension tube and needle from disconnecting. 4.66
71
72
Small Animal Spinal Disorders
4.67 Radiograph to show position of needle for lumbar puncture. A myelogram has been performed; contrast can be seen in the subarachnoid space. Note that there is epidural contrast (arrow) in the cranial lumbar vertebral canal. This is a common complication in lumbar myelograms and when present makes interpretation much harder.
4.67
4.68 T2-weighted MRI of the lumbar region of a largebreed dog to show the high-signal epidural (arrowheads) and subarachnoid spaces (arrow). Note that the epidural space is usually much wider relative to the subarachnoid space at L6/7 than at L5/6. The L4/5, L5/6 and L6/7 discs show normal signal intensities but there is loss of signal of the L7/S1 disc along with extension of the disc dorsally compressing the nerve roots of the cauda equina.
5
6
7
4.68
SPECIAL MYELOGRAPHIC CONSIDERATIONS The simple technique of taking a radiograph with the animal lying on the contralateral side should also be considered (4.33). Oblique projections may also be very useful in myelography; they are of most value from the ventrodorsal position, although interpreting these projections can sometimes be challenging. A follow-up CT scan is preferred after myelography when available, when clarification is required. In some circumstances, special ‘stressed’ positions may give more information about a lesion seen in a neutral position, or they may reveal a lesion not apparent previously.
Cervical spine
In cervical spondylomyelopathy special positions during myelography may be useful:
•
The traction view, where tension is applied to the cervical spine. It can indicate whether the lesion is ‘dynamic’ or ‘static’, which can then have a bearing on surgical planning (11.4, 11.5).
•
The extension view, where the neck is extended dorsally, can reveal other lesions that may become significant in the future; that is, the ‘domino’ effect (11.23). This view is not without hazard and must be done with care (11.9, 11.10).
•
Flexed and extended views may reveal lesions that cause minor trauma as the animal moves its neck around (11.6).
Lumbosacral spine
In general a cervical injection is preferred to a lumbar injection to evaluate this region, even though this necessitates that contrast be made to flow caudally by using gravity. The reason is because after lumbar puncture there is a risk of either injecting into the lesion (2.25) or of epidural leakage (4.67). An alternative is to use an L1/2 injection site (McCartney,1997). Flexion and extension myelography may reveal lumbosacral lesions not apparent on neutral positions (10.7), but it is possible to get false-positive results with this technique. Dynamic views may also be used during CT (10.11) and MRI (10.12) although the patient will then need to be repositioned and a second scan conducted.
Instrumentation
Draping
Chapter
5
78
References
79
Spinal surgery requires some special instruments and many of the items are illustrated in this chapter (5.1– 5.35). However, the starting point is a well-equipped general surgical pack, as would be used for most soft
5.1 Operating loupes. Some form of magnification is almost essential for neurosurgery.
5.2 An operating microscope is extremely useful when dissecting close to the spinal cord.
5.3 Headlight—good lighting is also essential for neurosurgery and this is facilitated by a fiberoptic light source (Goring et al., 1991), ideally one that attaches to the surgeon’s head and can be flipped up or down as needed.
5.4 Anatomical specimens such as this canine spine are invaluable for reference and for surgical planning.
74
Small Animal Spinal Disorders
5.5 Sew in, waterproof drape. See also 12.32.
5.8 Blunt self-retaining retractors are invaluable in the ventral approach to the cervical spine. Either Gosset (illustrated) or pediatric Balfour (9.18).
5.6 Gelpi self-retaining retractors. These are particularly useful in all dorsal and dorsolateral approaches to the spine (10.26, 13.60).
5.7 Multi-toothed self-retaining retractors. Illustrated here are Weitlander and Adson–Baby (top) retractors. West retractors are similar (9.20). These facilitate ventral cervical approaches.
5.9 Hand-held retractors. The small Hohmann retractor (above) is useful in ventral repair of atlantoaxial subluxation. The Senn (below) or Langenbeck retractor is useful in dorsal, dorsolateral and lateral approaches to the spine (8.15–8.18).
5.10 A laminectomy spreader is very useful to reduce some fractures and to distract over-ridden vertebrae (Boudrieau, 1997) (13.64).
Instrumentation
5.14 #7 scalpel handle with #11 blade. This is particularly useful for disc fenestration and ligament removal (7.34, 7.45).
5.11 Electrosurgical instruments. The monopolar system is used for coagulation and for incising tissues (8.14). Monopolar cautery should not be used in close proximity to the spine, as the current travels through the patient, and this can lead to spinal cord damage. Close to the spinal cord, bipolar or microbipolar cautery (5.12) must be used for coagulation (8.24).
5.15 Instruments for fenestration and disc removal from the vertebral canal. From top: Rosen mobilizer, Shea curette and House curette (5.16). The latter is also very useful for removal of bone during the final stages of laminectomy (8.37) and ventral spinal decompression (7.51).
5.12 Microbipolar cautery should be used when working close to neural tissues.
A
B 5.13 Periosteal elevator and freer, for removal of muscle from the vertebrae (7.38) and for vertebral distraction (11.33).
5.16 Close up of House curette.
75
76
Small Animal Spinal Disorders
5.17 Nerve hook (below) for retraction of spinal nerves. Dental tartar scrapers (blunt and pointed) for fenestration and removal of disc material from the vertebral canal (5.18). 5.20 Small instruments for use with higher-powered magnification.
A
B
5.21 These delicate rongeurs are used for fine bone removal. Larger, double action rongeurs are available for heavier bone removal (10.28).
5.18 Close up of dental instruments (8.35, 8.36, 8.39 and 8.41).
5.19 25 gauge hypodermic needles. These can be held at the hub by needle holders. They can be used for cutting or the tip may be twisted off using needle holders and then bent to size as a blunt retractor or scraper. Blunt instruments made in this way are particularly useful when working down a ventral slot, especially in small dogs (11.55B).
5.22 Mini chuck for insertion of small pins, such as those used for atlantoaxial subluxation (Chapter 9) and to stabilize articular facets after spinal trauma (13.48A).
Instrumentation
5.23 Pneumatic systems are preferable for bone removal in neurosurgical procedures. Illustrated here is the Hall Surgairtome 2 with long bur guard, and round and oval burs (8.30, 11.53).
5.24 Angled bur guard is useful for the cement plug technique used in Wobbler surgery (11.38).
5.25 An electrical drill is an alternative to the pneumatic drill. The Dremel model is an example; it is considerably less expensive than the pneumatic system and does have acceptable performance. Sterilization is a problem; ethylene oxide systems can be used but their availability is restricted. Methods of wrapping the instrument in sterile drapes are available, but are less satisfactory, and we do not recommend them in view of the requirement for asepsis in spinal surgery. Disadvantages of an electric drill are that it is slower; more pressure may be needed, which can cause bone necrosis and possibly inadvertent injury to the spinal cord; and the burs may clog more rapidly (Shires et al., 1986; Bitetto and Kapatkin, 1989).
tissue and orthopedic procedures (Cockshutt, 2003). With experience, surgeons find which instruments they prefer for spinal surgery, but there is no doubt that certain instruments make the tasks easier and, thus, more efficient.
5.26 Adson suction tip and bulb syringe for bur irrigation (8.29, 8.30, 8.39–8.41).
5.27 Bone wax is used for hemostasis; it is used to plug small vessels in bone by pressing over the bleeding site (8.31, 13.58).
5.28 Cellulose surgical spears (Ultracell sponges, Ultracell medical Technologies Inc., North Stonington, CT.) are a very useful type of sponge that does not shed material like standard sponges. They are recommended to absorb fluid in contact with the spinal cord.
5.29 Surgicel. If this product is used, it must always be removed from the laminectomy site once hemostasis is achieved as it swells and may cause compression. Surgifoam hemostat, however, may be left in situ since it does not swell.
77
78
Small Animal Spinal Disorders
5.30 Gelfoam (Pharmacia, Kalamazoo, MI) absorbable gelatin sponge is an alternative product for hemostasis that can be left in place.
A
5.31 Hemoclips (Pilling Weck Inc., Research Triangle Park, NC) are often the only way to control hemorrhage from largerdiameter veins in proximity to the intervertebral foramen (11.51).
B 5.33 Methylmethacrylate bone cement is used in several of the procedures described in this book. It is worth emphasizing the requirement for sterility when using these products (13.60).
Basic surgical techniques for neurosurgery share most of the principles of soft tissue and orthopedic surgery (Cockshutt, 2003; Shmon, 2003). 5.32 Positive profile pins (see Chapter 9 and also 13.48B); small diameter pins are best inserted using a mini chuck (5.22). These pins also have a roughened end opposite the threaded end so that they do not need to be notched like smooth pins. Small diameter negative profile pins can also be used as even they have 4–6 times the pullout strength of equivalent diameter smooth pins (Degernes et al., 1998; Sandman et al., 2001).
DRAPING The patient should be completely covered by sterile, waterproof surgical drapes. Their additional expense must be weighed against the morbidity and the cost of
Instrumentation
5.34 Bone graft is used in several situations to promote vertebral fusion. Illustrated is a curette and bowl for graft collection (11.24).
placed on top of the waterproof drapes prevent tissues from drying out under surgical lights. They also help self-retaining retractors maintain exposure of the surgical field (10.26, 12.36) and are less likely than standard sponges to be left in the wound inadvertently. Tissues should be irrigated regularly with sterile saline to prevent desiccation and to reduce airborne contamination. After completion of surgery, the site must be inspected to insure that nothing has been left in the wound. Retractors are invaluable for providing adequate exposure to perform the procedure. This can be supplemented by use of fine stay sutures, particularly those applied to the dura (see 14.5–14.8).
REFERENCES
5.35 A Jamshidi needle is invaluable for taking a core biopsy of a vertebral mass.
a surgical wound infection caused by organisms penetrating a porous drape. Sew-in or clip-in drapes may be used to isolate the wound further; ideally these should also be waterproof (5.5). Moistened laparotomy sponges or towel drapes
Bitetto, W.V., Kapatkin, A.S. (1989) Intraoperative problems associated with intervertebral disc disease. Problems in Veterinary Medicine 1, 434–444. Boudrieau, R.J. (1997) Distraction-stabilization using the ScovilleHaverfield self-retaining laminectomy retractors for repair of 2nd cervical vertebral fractures in 3 dogs. Veterinary and Comparative Orthopaedics and Traumatology 10, 71. Cockshutt, J. (2003) Principles of surgical asepsis. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 149–154. Philadelphia: WB Saunders. Degernes, L.A., Roe, S.C., Abrams, C.F., Jr (1998) Holding power of different pin designs and pin insertion methods in avian cortical bone. Veterinary Surgery 27, 301–306. Goring, R.L., Beale, B.S., Faulkner, R.F. (1991) The inverted cone decompression technique: a surgical treatment for cervical vertebral instability ‘Wobbler syndrome’ in Doberman Pinschers. Part 1. Journal of the American Animal Hospital Association 27, 403–409. Sandman, K.M., Smith, C.W., Harari, J., Manfra Maretta, S., Pijanowski, G.J. (2001) Comparison of pull-out resistance of Kirschner wires and Imex miniature interface fixation pins in polyurethane foam. Veterinary and Comparative Orthopaedics and Traumatology 15, 18–22. Shires, P.K., Roberts, E.D., Hulse, D.A., Zeman, D.H., Kearney, M.T. (1986) Hemilaminectomy using an autoclavable electrical drill compared to a pneumatic drill. Journal of the American Animal Hospital Association 22, 25–29. Shmon, C. (2003) Assessment and preparation of the surgical patient and operating team. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 162–178. Philadelphia: WB Saunders.
79
Preoperative assessment
Pharmacological considerations 83 Antibiotics 83 Non-steroidal anti-inflammatories 83 Corticosteroids 83 Preoperative analgesia 84 Adverse drug reactions and interactions
85
85
References
Skin or periodontal disease
■
Cushing’s disease
■
Orthopedic disease
■
Hypothyroidism
■
Cardiac disease
■
Diabetes mellitus
■
Hepatic disease
■
Disorders of hemostasis
■
Renal disease
■
Urinary tract infection
■
Neoplasia – Metastatic disease
87
■
Prostatic disease
■
Pyometra
(Box 6.1). This is particularly true for older patients, following trauma (Box 13.2) and dogs with cervical spondylomyelopathy (see Chapter 11, page 213).
Concurrent disease
88
Key issues for future investigation
■
– Undetected primary
Surgical considerations 86 Laminectomy healing 86 Durotomy 87 Myelotomy 87 Mechanisms of recovery after spinal cord injury Client communication
6
Box 6.1 Conditions that may complicate the management of a neurological patient
Clinical assessment 81 Concurrent disease 81 Hemostasis 82 Hypoventilation 82
Anesthetic considerations Premedication 85 Induction 85 Maintenance 85 Recovery 85 Complications 86
Chapter
88
88
Accurate preoperative assessment of the neurosurgical patient is extremely important, particularly in view of the time, expense and complexity of the procedures performed. Anticipation of the most likely complications may lessen their impact or even allow them to be avoided altogether.
CLINICAL ASSESSMENT It is easy to overlook concurrent clinical or subclinical disorders that may have an important bearing on the case
The physical examination must be thorough. Skin disorders may necessitate delaying elective surgery because pyoderma, clipper rash, or povidone–iodine reactions may predispose to wound infection. Fleas should be eliminated to prevent them from entering the surgical field. Osteoarthritic joints may cause major interference with rehabilitation from neurological deficits, especially for large or obese dogs. Laboratory evaluation should include a hematological and biochemical profile, along with a urinalysis (see Chapter 4). In cats, the feline leukemia virus (FeLV) and feline immunodeficiency virus (FIV) status should be determined. When positive, the likelihood of lymphoma as a cause for neurological deficits increases markedly. If there is suspicion of cardiac dysfunction in large- or giant-breed dogs, suitable investigations should
82
Small Animal Spinal Disorders
be performed, such as an electrocardiogram (ECG) and echocardiogram (Calvert and Wall, 2001). Chest radiographs are essential in these patients: • For older animals (⬎7 years of age). • When neoplasia is a differential diagnosis. • For any recumbent, tetraparetic patient. • Following trauma. Similarly, an ultrasound or CT scan of the abdomen is useful to identify intercurrent disease in the first two categories, above. Endocrine disorders may predispose a patient to complications, for example, urinary tract infections (UTI) and delayed wound healing in Cushing’s disease and diabetes mellitus or pathological fractures in Cushing’s disease (Gehlbach et al., 2000; Hosgood, 2003). Neuropathies or myopathies associated with hypothyroidism may complicate the patient’s neurological status. The assessment and, if possible, stabilization of endocrine disorders is always recommended prior to any neurosurgical procedure.
Hemostasis Hemostasis is divided into primary and secondary events (Kerwin and Maudlin, 2003). Primary hemostasis depends on platelet aggregation and adhesion, and usually causes most concern to the neurosurgeon. It may be disturbed in the following conditions: • von Willebrand disease (see Chapter 11, page 217). • Severe thrombocytopenia (platelet count ⬍ 20 000/l). • Azotemia (serum creatinine ⬎ 5.6 mg/dl). • Non-steroidal anti-inflammatory drug (NSAID)induced platelet dysfunction, including aspirin at all therapeutic doses and especially above 25 mg/kg. • Anticoagulant intoxication. Disorders of primary hemostasis should be identified by a platelet count and bleeding time evaluation (11.11). Perturbed hemostasis can cause decreased visualization during surgery, postoperative bruising (15.2), hematoma formation and other more serious complications (15.40). The dog shown in 6.1, represented 4 months after surgery with a packed cell volume (PCV) of 12% from a bleeding gastric ulcer but survived after a transfusion. The owner had treated the dog with aspirin at 40 mg/kg for 2 weeks.
Hypoventilation This is a serious potential complication that can occur in any animal with a serious cervical spinal cord injury. Patients that are severely tetraparetic on initial presentation should be evaluated for hypoventilation using blood gas analysis or a capnometer. Tetraplegic animals are at even higher risk. Occasionally an animal may need
urgent ventilatory assistance prior to surgery in order to survive long enough to benefit from definitive treatment (Boudrieau, 1997). Any severely tetraparetic animal can also develop this complication in the postoperative period and the owner must be appraised of this risk prior to surgery (6.1, 7.11). Hypoventilation after spinal cord injury can occur due to three main mechanisms: • Hemorrhage or edema affecting the respiratory centers in the medulla and C1 spinal cord, which lead to decreased respiratory drive. • Cervical myelopathy severe enough to interrupt conduction along all motor fibers in the spinal cord, which causes near-paralysis of the respiratory muscles (see pages 28, 216). • Diaphragmatic paralysis due to a lesion of the phrenic LMNs (C5 spinal cord segment or nerve roots); this is the least important mechanism unless combined with an injury to the white matter at that level (Smith and Walter, 1985; Blass et al., 1988; Beal et al., 2001). The end result of these injuries is either that there is decreased respiratory drive or there is insufficient communication between the respiratory center and the respiratory muscles (Beal et al., 2001). Apnea can also occur following the injection of subarachnoid contrast at the cerebello-medullary cistern (CMC) (Widmer et al., 1992a), or after manipulation during imaging (Seim and Prata, 1982) (11.10).
6.1 Myelogram from a 6-month-old Great Dane with sudden onset tetraplegia and dyspnea. It had neck pain with deficits localized to the C1–C5 spinal cord. Blood gas analysis showed a paCO2 of 45 mmHg (N35–45). Ventral slot at C3/4 retrieved a large amount of nucleus pulposus from the vertebral canal. On extubation the dog did not breathe spontaneously; its paCO2 was 91 mmHg. It was ventilated overnight but attempts to wean it from the ventilator were unsuccessful until 4 days later. It was able to walk 2 weeks after surgery and made a good recovery (see Hemostasis).
Preoperative assessment
PHARMACOLOGICAL CONSIDERATIONS Antibiotics
(Reed, 2002), or given together with corticosteroids, as potentially severe gastrointestinal erosions or ulcerations are likely.
Most neurosurgical procedures can be classified as clean and uncontaminated. However, prophylactic antibiotics are indicated if sterility is broken or if one or more of the factors in Box 6.2 apply. In such cases, perioperative cefazolin (20 mg/kg IV) is recommended for its good tissue penetration and broad spectrum of activity against staphylococci and other Gram-positive organisms (Dunning, 2003). A short, decisive window for prophylactic antibiotic use extends from the start of surgery to a maximum of 3 h afterwards. It is at this time that bacterial contamination of tissue can be suppressed by antimicrobial therapy. There is no advantage to using IV antibiotics before the start of surgery, or for continuing them beyond its completion, except in concomitant diseases such as pyoderma or UTI (Rosin, 1988). A penicillin-derivative or a cephalosporin is usually a suitable initial choice for retention cystitis. Final antibiotic selection for UTI should be based on urine culture and sensitivity whenever possible, especially when the animal has been hospitalized and a nosocomial infection is much more likely (Dunning, 2003).
Corticosteroids
Box 6.2 Some factors that can predispose to wound infection ■
Dermatitis
■
Periodontal disease
■
Urinary tract infection
■
Obesity
■
Shock or sepsis
■
Use of a surgical implant
■
Cushing’s disease
■
Corticosteroid or non-steroidal anti-inflammatory drug (NSAID) use
■
Diabetes mellitus
■
Surgical time over 90 min
■
Excessive use of electrocautery
Non-steroidal anti-inflammatories These drugs are valuable analgesics although they are often less effective than corticosteroids as specific anti-inflammatory agents for neurological disease. They can have a number of important adverse effects. Such effects are most marked in drugs that inhibit mainly cyclooxygenase-1 (COX-1) as opposed to COX-2 (Kay-Mugford et al., 2000) (see Chapter 15, page 341). NSAIDs must not be used at more than the recommended dose, used in combination with other NSAIDs
Corticosteroids are used widely in all types of spinal cord conditions, mainly because they are so effective at reducing the associated inflammation and at relieving pain. However, there is a widespread misunderstanding that these beneficial effects also apply to the neural injury itself. While corticosteroids may reduce inflammation associated with, for example, extruded disc material, they can also prejudice the survival of any injured neurons by interfering with their glucose metabolism (Sapolsky, 1994; Smith-Swintosky et al., 1996). This may be of little consequence with less severe spinal cord lesions but it could be critical in severe injuries. In such cases the animal might feel better but may have a decreased chance for neurological recovery. In addition, there is almost certainly no neuroprotective benefit provided by any corticosteroid with the exception of methylprednisolone sodium succinate (MPSS) (Heary et al., 1997; Olby, 1999; Hurlbert, 2000; Bracken and Holford, 2002). Corticosteroids can also precipitate gastrointestinal bleeding in as many as 15% of neurosurgical patients, with mortality rates of up to 2%. Dexamethasone is most likely to cause problems and has no role in the management of spinal trauma (Moore and Withrow, 1982). Standard gastrointestinal protectant agents may not be effective in preventing corticosteroid-induced side-effects (Hanson et al., 1997). Duodenal or colonic perforation are the most serious potential complications (Toombs et al., 1986; Hinton et al., 2002). Routine glucocorticoid therapy in spinal patients is strongly discouraged unless these drugs are used for short periods at anti-inflammatory doses in animals with mild neurological deficits (LeCouteur and Sturgess, 2003).
METHYLPREDNISOLONE SODIUM SUCCINATE (MPSS) MPSS has received wide interest in human, and, to a lesser extent, veterinary medicine in the past 10 years because of proposed benefits in spinal cord injury. After acute injury, the blood supply to the spinal cord is progressively reduced. When the injured tissue is reperfused, massive amounts of highly reactive chemicals called free radicals are liberated. These free radicals are especially damaging to the plasma membrane of cells via a process called lipid peroxidation. Free radical-induced lipid peroxidation is now recognized as a key pathophysiological mechanism for irreversible tissue loss following spinal cord trauma and ischemia (Brown and Hall, 1992).
83
84
Small Animal Spinal Disorders
The neuroprotective effect of MPSS is exerted by its actions as a free-radical scavenger. These benefits are not due to its glucocorticoid activity and only occur in doses far exceeding those that saturate all glucocorticoid receptors. The optimal neuroprotective dose of MPSS has been determined to be 30 mg/kg, whereas doses of 60 mg/kg were detrimental and doses of 15 mg/kg had no effect. Benefit has only been observed in humans with spinal cord injury who received treatment within 8 h of injury (Bracken et al., 1990, 1992). In animals, the suggested dosage regime is an initial IV bolus of 30 mg/kg MPSS, followed by 15 mg/kg IV 2 and 6 h later, then 2.5 mg/kg IV per hour for a further 24 h (Brown and Hall, 1992). The bolus doses should be given slowly to avoid vomiting and hypotension. The continuous IV dose is not widely used in small animal patients; humans are given an initial bolus of 30 mg/kg, followed by an infusion of 5.4 mg/kg/h for 24 h (Bracken et al., 1990). A similar regime might also provide some benefit when given prior to spinal cord decompression for lesions that have been present for longer than 8 h. However, this hypothesis has not yet been tested and might even have an adverse effect (Olby, 1999; Bracken, 2000a,b, 2001). Vitamin E also has protective effects on the spinal cord when given at 1000–2000 IU per animal per day for 5 days and is an alternative prior to elective surgery (Fehlings et al., 1989; Olby, 1999). No serious side-effects were reported in one study using high dose MPSS therapy in 86 dogs with thoracolumbar disc disease (Siemering and Vroman, 1992). In another study, 35 of 108 dogs developed complications such as diarrhea or melena but none were considered serious and they usually resolved without therapy (Culbert et al., 1998). However, endoscopy revealed severe, subclinical gastric hemorrhage in 90% of dogs after MPSS treatment (Rohrer et al., 1999a,b); and 90% of dogs treated with MPSS prior to spinal surgery get occult gastrointestinal bleeding, although it is rarely severe enough to warrant intervention (Hanson et al., 1997; Rohrer et al., 1999a). Caution is advisable if considering MPSS therapy after treatment with other types of corticosteroid or with NSAIDs. Gastrointestinal barrier disruption and bloody diarrhea can lead to bacteremia, which is undesirable in a surgical patient (Epstein et al., 1992). Complications of MPSS reported in humans include pneumonia, sepsis, immunosuppression and pancreatitis (Levy et al., 1996; Bracken et al., 1997; Gerndt et al., 1997; Matsumoto et al., 2001). A number of studies have now raised serious questions about the value of MPSS in humans (George et al., 1995; Gerhart et al., 1995; Levy et al., 1996; Heary et al., 1997; Nesathurai, 1998; Hurlbert, 2000; LeCouteur and
Sturgess, 2003). The benefits reported originally were mainly to upper body function and not a regaining of the ability to walk (Bracken et al., 1990, 1992, 1997). These relatively small changes are crucial in people but are of much less significance in animals. They may also represent an improvement in gray matter function whereas white matter survival is much more important in animals (Jeffery and Blakemore, 1999a). Furthermore, the statistical basis of the original human clinical trials has now been questioned (Hurlbert, 2000). Although highdose MPSS therapy may be an advance in the management of acute spinal cord injury, it is not a panacea. Administration after the 8-h therapeutic window worsens the outcome in humans while the length of this window in dogs and cats has not been determined (Bracken, 2000a,b, 2001; Hurlbert, 2000). The deleterious effect of delayed administration of MPSS is probably due to interference with neuronal glucose metabolism (Sapolsky, 1994; Smith-Swintosky et al., 1996; LeCouteur and Sturgess, 2003). Therapy with MPSS must therefore be looked upon at best as a way to complement, but not replace, current veterinary neurosurgical techniques (LeCouteur and Sturgess, 2003).
Preoperative analgesia Fentanyl patches provide a useful way to deliver preoperative analgesia and they can also be combined with one of the analgesics discussed under postoperative care (see Chapter 15), however the rate of absorption of fentanyl can vary considerably in an individual animal over time and can also vary at different times in the same animal (Kyles et al., 1996; Egger et al., 1998). The dose range is 2–5 g/kg/h. Increasing the patch size did not increase the plasma fentanyl concentration in one study but did in a second (Egger et al., 1998; Welch et al., 2002a). The analgesic effect provided by fentanyl in dogs is equivalent to that provided by intramuscular oxymorphone and is often superior to epidural morphine; in cats it is superior to butorphanol (Kyles et al., 1998; Robinson et al., 1999; Franks et al., 2000). The onset of action is faster in cats (2–6 h) than in dogs (24–36 h) (Scherk Nixon, 1996; Robinson et al., 1999). The patch should therefore be placed 24 h prior to surgery in dogs and supplemental analgesia is often required during this period (Egger et al., 1998). Concentrations fall rapidly after patch removal and are often below therapeutic concentrations within 1 h (Egger et al., 1998). Respiratory depression can be a serious side-effect in humans but was not seen in dogs at high doses (5 g/kg/h) following thoracotomy (Welch et al., 2002a). Caution should be used in dogs that are hypoventilating due to head or spinal cord injury (see Chapter 15). Other potential side-effects in dogs include bradycardia,
Preoperative assessment
dysphoria and vomiting; acepromazine and glycopyrrolate may alleviate these effects. Skin reactions may also occur (15.1). The patch must not be placed on a heating pad as this will increase the rate of absorption from the patch (Egger et al., 1998). Fentanyl is a useful way to provide non-invasive, inexpensive, long-lasting analgesia that is tolerated well (Scherk Nixon, 1996; Kyles, 1998). However, careful monitoring is needed to insure adequate analgesia and minimize adverse effects (Scherk Nixon, 1996; Kyles, 1998) (see also ‘Postoperative analgesia’, page 339).
Adverse drug reactions and interactions Adverse effects of drugs are reported only rarely in animals but are likely to become a more serious problem, especially with the increasing awareness of such events in humans (Stillman, 1989; Weinblatt, 1989). One potential adverse effect is a drug interaction, for example renal failure as has been reported in people using NSAIDs and angiotensin-converting enzyme (ACE) inhibitors together (Seelig et al., 1990). Cyclosporin can cause adverse interactions in dogs when given together with either ketoconazole or with ivermectin (Myre et al., 1991; Roulet et al., 2003). A second type of adverse effect is an unexpected reaction to an individual drug, for example enrofloxacin as a (dose-related) cause of blindness in cats or metronidazole as a cause of central vestibular disease (Dow et al., 1989; Gelatt et al., 2001). Hepatotoxicosis, keratoconjunctivitis and aplastic anemia have been reported in dogs given trimethoprimsulfonamide combinations (Diehl and Roberts, 1991; Rowland et al., 1992; Fox et al., 1993). Anaphylactictype reactions can occur in dogs given cefalosporins during surgery; to avoid this complication occurring intraoperatively it is preferable to give the first dose along with the premedication where possible (Anderson and Adkinson, 1987; Grouhi et al., 1999). Adverse drug events have been reviewed extensively in humans and in animals (Boothe, 1990; Scott and Miller, 1998, 1999; Ament et al., 2000; Maddison et al., 2000; Papadogiannakis, 2000).
ANESTHETIC CONSIDERATIONS Many neurosurgical patients will be dehydrated at presentation, usually when pain or weakness reduce fluid intake. They must be rehydrated adequately prior to anesthetic induction, because of the detrimental effect of hypotension on spinal cord perfusion (Tator and Fehlings, 1991; Nuwer, 1999). Another way to reduce dehydration is to increase preoperative fluid intake; fasting recommendations for humans now permit water up
until 2 h prior to anesthesia for this reason (Crenshaw and Winslow, 2002).
Premedication Premedication should have a calming effect and relieve pain. Hypotension and loss of protective muscle tone should be avoided in an animal with severe neurological deficits or an unstable vertebral column. Glycopyrrolate is the preferred anticholinergic as it is less likely to induce tachycardia; atropine may still be required to treat severe bradyarrhythmias (Stauffer et al., 1988).
Induction A laryngoscope should be used for intubation with as little movement of the spine as possible, especially in dogs with unstable lesions or lesions in the cervical area. An armoured endotracheal tube is recommended if cerebrospinal fluid (CSF) is to be taken from the cerebellomedullary cistern (CMC), if stress radiographs are to be taken, or if a ventral approach to the neck is to be used.
Maintenance Isofluorane is the usual inhalation agent of choice, because it is both less depressant to the cardiovascular system and less arrhythmogenic than halothane. Methoxyfluorane provides good muscle relaxation and better postoperative analgesia than isofluorane, but is contraindicated in animals receiving opioids or NSAIDs, or in those with renal disease. Sevofluorane is a good inhalational agent in dogs and may be superior to isofluorane (Branson et al., 2001). Autoregulation in the brain is preserved better with sevofluorane than with isofluorane (Summors et al., 1999; Endoh et al., 2001); both agents maintain good spinal cord blood flow (Hoffman et al., 1991; Crawford et al., 1992). Maintenance of anesthesia should be at a depth sufficient to prevent movement of the patient, which could be dangerous during surgery. Mechanical ventilation is recommended in large or debilitated animals, those that are positioned in dorsal recumbency and those with neurological deficits cranial to the thoracolumbar junction. Barotrauma must be avoided, especially in small animals (Manning and Brunson, 1994; Parent et al., 1996). Nitrous oxide is not recommended as it may result in a rapid onset of hypoxia if the patient hypoventilates during CSF collection or if it will be disconnected temporarily from the gas supply during radiography.
Recovery The recovery from anesthesia should be smooth, and this may be helped by the use of narcotics, diazepam, or both. If the animal has undergone myelography, its head
85
86
Small Animal Spinal Disorders
must be kept elevated during the entire recovery period. Seizures should be treated promptly with diazepam (0.4 mg/kg IV), repeated as necessary. The use of iohexol for myelography means that the risk of seizures is low, although Doberman pinschers and other large-breed dogs with cervical spondylomyelopathy may be at increased risk and warrant careful monitoring (Lewis and Hosgood, 1992). It is prudent to leave an IV catheter in place until the patient is fully recovered.
Complications Anesthesia must optimize both cardiovascular and respiratory function in order to minimize ischemia of the spinal cord. Spinal cord blood flow is autoregulated in a manner analogous to cerebral blood flow; volatile anesthetic agents may depress this autoregulation and trauma to the spinal cord can abolish it altogether (Tator, 1991; Nuwer, 1999; Olby and Jeffery, 2003). Any cardiac arrhythmia or systemic hypotension will compound this by reducing spinal cord blood flow further and so should be avoided if at all possible. Iohexol is a safe agent for myelography (Wheeler and Davies, 1985), although the incidence of seizures can be around 7% in large-breed dogs with ‘Wobbler syndome’ (Lewis and Hosgood, 1992; Widmer et al., 1992b). Seizures can cause the animal to extend its neck, which might have an adverse effect on a dynamic cervical lesion (11.10). It is recommended that the patient be at an adequate depth of anesthesia before the injection is made, that the contrast be used at body temperature, and that it is injected at a steady rate (Lamb, 1994). In one study, six out of 66 dogs developed bradycardia during iohexol injection and one died; two also developed bradycardia during recovery (Lewis and Hosgood, 1992). The ECG must be monitored closely during the study and any problems treated promptly. Hypothermia is another potential problem, particularly in small animals or if the anesthetic period is prolonged. Surgical techniques can also affect the patient during anesthesia. Cervical spinal surgery is associated with a much higher risk of arrhythmias and ventricular premature contractions than thoracolumbar surgery (Stauffer et al., 1988). This may result from manipulation of either the spinal cord itself, or of nerves in the ventral neck. Both positioning the spine in extension and manipulation of the vagosympathetic trunk should be minimized. A paramedian approach to the neck provides better protection for vital structures and may therefore reduce the incidence of intraoperative arrhythmias (11.25–11.27). Magnification is recommended when performing surgery on the cervical spinal cord in order to reduce unnecessary manipulation (see Chapters 7 and 11). Severe cervical myelopathies can result in sympathetic blockade (11.10), which can be fatal
(Rosenbluth and Meirowsky, 1953; Seim and Prata, 1982; Clark, 1986). Surgery in the cranial and mid-thoracic spine can also impair autonomic control of cardiovascular function and so the patient must again be monitored closely.
SURGICAL CONSIDERATIONS Laminectomy healing A laminectomy heals as the hematoma forms a fibrous callus, which then undergoes metaplasia to cartilage and bone (Trotter et al., 1988). Adhesions can involve the dura and nerve root(s) if no attempt is made to protect them after laminectomy (Cook et al., 1994). Adhesions around local nerve roots can cause considerable postoperative morbidity in humans, probably by causing a tethering effect (Songer et al., 1990; Geisler, 1999). This has not been identified specifically in animals although those with postoperative back pain and evidence of probable peridural fibrosis on CT scan have been reported (Olby et al., 2000). The preferred way to assess epidural scarring is by using MRI (Ross et al., 1999). The more scar tissue that is evident on MRI, the more pain is reported in humans after lumbar discectomy (Ross et al., 1996; Maroon et al., 1999). It is advisable to keep the exposed dura mater separate from the damaged epaxial muscles during healing in order to minimize the development of adhesions, which can even cause subsequent spinal cord compression (Trotter et al., 1988; Cook et al., 1994). Several implants have been used to minimize adhesions but none is ideal (Cook et al., 1994). One study using Gelfoam (Pharmacia, Kalamazoo, MI) showed no reduction of peridural scarring in dogs; two more studies showed that scarring was actually increased (Gill et al., 1979; Songer et al., 1990; Robertson et al., 1993) and a fourth study showed reduction in scar using Gelfoam (LaRocca and Macnab, 1974). Better overall results seem to be obtained in dogs using autogenous fat (Gill et al., 1979; Cook et al., 1994). Histopathological evaluation of free fat grafts after laminectomy for disc disease in 21 dogs showed that 50–90% of the graft was made up of fat with no scar. In a further eight dogs, myelography was used to verify the lack of scar formation in the subarachnoid space over the surgical site. Overall follow-up times ranged from 1 month to 5 years (Biggart, 1988). CT scans can also be used to verify the presence and viability of fat graft at a previous laminectomy site (Biggart, 1988; Olby et al., 2000). However, two other studies found either no benefit from free fat grafts (Songer et al., 1995), or that grafts do not prevent dural adhesions (Trevor et al., 1991). Free fat grafts must revascularize and so should be
Preoperative assessment
no more than 3–5 mm thick in order to minimize the risk of aseptic necrosis (8.8, 8.54, 12.10). An initial inflammatory phase resolves as fibrosis increases over 8–16 weeks. By this time the graft has contracted to roughly half its original size (Trevor et al., 1991; Cook et al., 1994). One study has shown that pedicle grafts in dogs have no advantage over free fat grafts; a second study showed pedicle grafts to be superior (Gill et al., 1979; Trevor et al., 1991). This discrepancy could simply reflect the small number of dogs in the studies. A commercially available alternative is ADCON-L (Gliatech Inc., Cleveland, OH), a novel, porcine-derived, polyglycan implant that blocks in-growth of fibroblasts. It was shown to be effective in dogs (Einhaus et al., 1997; BenDebba et al., 1999; Geisler, 1999); it can even be used to deliver morphine locally to the surgical site (Mastronardi et al., 2002). There is still some controversy about its value and price may be an issue for veterinary use (Richter et al., 2001). A resorbable, polymer barrier film (Hydrosorb TS, Macropore Biosurgery Inc., San Diego, CA) has proven to be superior to ADCON-L in rats and was also moderately effective in dogs compared to controls (Welch et al., 2002b).
Durotomy Durotomy, or incision of the dura mater, results in mild, temporary deficits in normal dogs (Parker and Smith, 1972). As the dural incision heals, the edges rejoin but they often adhere to the spinal cord as well (Trevor et al., 1991). Durotomy is necessary to evaluate intradural and intramedullary lesions (Jeffery and Phillips, 1995) (12.40). It is sometimes performed at laminectomy in an attempt to decompress the spinal cord further (8.49, 8.50). Although this may improve decompression, the relative risks and benefits are unclear (Perkins and Deane, 1988). In experimental situations, durotomy may have value when performed immediately after trauma, but was found to have no benefit 2 h after the injury (Parker and Smith, 1974, 1975). However, studies of intracranial pressure show that craniotomy and durotomy lower pressure by 15 and 65%, respectively (Bagley et al., 1996). This suggests that durotomy might also produce a significant decompressive benefit for spinal cord. Durotomy is not recommended in order to provide prognostic information after spinal cord injury and euthanasia should not be based solely on the gross appearance of the spinal cord (Salisbury and Cook, 1988) (8.50).
Myelotomy Myelotomy involves incising the spinal cord on the dorsal midline to the level of the central canal. It has been performed in normal dogs at the T3–L3 region with minimal
residual neurological sequelae (Teague and Brasmer, 1978). It has also been used with good results to aid removal of a spinal cord hematoma and a tumor (Martin et al., 1986; Jeffery and Phillips, 1995). No therapeutic value has been demonstrated in acute spinal trauma and the technique is difficult to perform in cats without causing additional injury (Hoerlein et al., 1985).
Mechanisms of recovery after spinal cord injury Recovery occurs mainly by the re-establishment of a normal spinal cord microenvironment and forming new patterns of central nervous system (CNS) circuitry (Jeffery and Blakemore, 1999b). Nociception is normally regained first then motor function and continence followed by proprioception (see Chapter 2). Interestingly, some animals that fail to regain deep pain and continence still recover the ability to walk (13.30, 13.34). Although often termed spinal reflex walking, this type of recovery probably requires some input from higher centers mediated through a few intact axons surviving across the lesion. Survival of small groups of axons can occur because many severe spinal cord injuries are centered on the gray matter and there is relative sparing of peripheral white matter (Griffiths, 1978; Olby et al., 2003) (6.2). Evidence of higher input in dogs that walk without recovering deep pain is inferred from the fact that many regain a voluntary tail wag (Olby et al., 2003). However, these dogs nearly always remain incontinent and the recovery of walking ability takes at least 4 months and usually much longer (Olby et al., 2003).
6.2 Thoracic spinal cord from a dog that did not regain deep pain sensation after a spinal fracture 6 weeks previously. There is a large area of central necrosis with small groups of surviving axons in the periphery (arrows). These axons may mediate late recovery of motor function in dogs that never regain deep pain sensation and that remain incontinent (Olby et al., 2002).
87
88
Small Animal Spinal Disorders
CLIENT COMMUNICATION The client should always be offered the ‘textbook’ approach for managing a patient in terms of the diagnostic evaluation and treatment. This should be offered regardless of the clinician’s perception of whether the client will pursue treatment, and even when it involves referral. In this manner, the client is made aware that the very best is available for their animal, and any compromises arrived at will be seen in the proper perspective. Whenever possible, the clinical problems and the diagnostic and therapeutic options should be explained to the client in non-scientific terms. Anatomical specimens, websites, drawings and the animal’s own laboratory data and images should be used to illustrate the situation. On the other hand, care must be taken not to overload the client with excessive information, and they should be given time and the opportunity to think privately about their decision. It is also important that the client be given as accurate a prognosis as possible, both for the degree of recovery and the time required. Giving the client prior knowledge of the probable prognosis and of the most likely complications is invaluable should the patient suffer either a protracted recovery or setbacks in its postoperative course. The client should never be pressured into consenting to a course with which they are not comfortable, even if the clinician feels that it is in the animal’s best interests. If complications do arise the client may then hold the clinician responsible, not only for the decision but also for the complications. Finally, the client should be given, and asked to sign, a consent form and a written estimate of the likely cost for the entire procedure. It is vital that the client be made aware that there is some risk even in a routine general anesthetic and scan. This estimate should include a price range to take into account any foreseeable variations in the postoperative course. Should the bill begin to approach the upper end of the estimate, or if unforeseen complications develop, the client must be notified immediately (Thacher, 1989). Key issues for future investigation 1. Does MPSS provide significant neuroprotection in dogs after spinal cord trauma? 2. Does MPSS provide benefit when given prophylactically prior to surgery? 3. What are the relative merits of MPSS and Vitamin E?
REFERENCES Ament, P.W., Bertolino, J.G., Liszewski, J.L. (2000) Clinically significant drug interactions. American Family Physician 61, 1745–1754.
Anderson, J.A., Adkinson, N.F., Jr (1987) Allergic reactions to drugs and biologic agents. Journal of the American Medical Association 258, 2891–2899. Bagley, R.S., Harrington, M.L., Pluhar, G.E., Keegan, R.D., Greene, S.A., Moore, M.P., Gavin, P.R. (1996) Effect of craniectomy/durotomy alone and in combination with hyperventilation, diuretics, and corticosteroids on intracranial pressure in clinically normal dogs. American Journal of Veterinary Research 57, 116–119. Beal, M.W., Paglia, D.T., Griffin, G.M., Hughes, D., King, L.G. (2001) Ventilatory failure, ventilator management, and outcome in dogs with cervical spinal disorders: 14 cases (1991–1999). Journal of the American Veterinary Medical Association 218, 1598–1602. BenDebba, M., Augustus van Alphen, H., Long, D.M. (1999) Association between peridural scar and activity-related pain after lumbar discectomy. Neurological Research 21 Suppl 1, S37–42. Biggart, J.F., III (1988) Prevention of laminectomy membrane by free fat grafts after laminectomy in dogs with disk herniations. Veterinary Surgery 17, 28–29. Blass, C.E., Waldron, D.R., Van Ee, R.T. (1988) Cervical stabilization in three dogs using Steinmann pins and methylmethacrylate. Journal of the American Animal Hospital Association 24, 61–68. Boothe, D.M. (1990) Drug therapy in cats: mechanisms and avoidance of adverse drug reactions. Journal of the American Veterinary Medical Association 196, 1297–1305. Boudrieau, R.J. (1997) Distraction–stabilization using the ScovilleHaverfield self-retaining laminectomy retractors for repair of 2nd cervical vertebral fractures in 3 dogs. Veterinary and Comparative Orthopaedics and Traumatology 10, 71. Bracken, M.B. (2000a) Methylprednisolone and spinal cord injury. Journal of Neurosurgery 93 Suppl 1, 175–179. Bracken, M.B. (2000b) The use of methylprednisolone. Journal of Neurosurgery 93 Suppl 2, 340–341. Bracken, M.B. (2001) High dose methylprednisolone must be given for 24 or 48 hours after acute spinal cord injury. British Medical Journal 322, 862–863. Bracken, M.B., Holford, T.R. (2002) Neurological and functional status 1 year after acute spinal cord injury: estimates of functional recovery in National Acute Spinal Cord Injury Study II from results modeled in National Acute Spinal Cord Injury Study III. Journal of Neurosurgery 96, 259–266. Bracken, M.B., Shepard, M.J., Collins, W.F., Holford, T.R., Young, W., Baskin, D.S., Eisenberg, H.M., Flamm, E., Leo-Summers, L., Maroon, J., et al. (1990) A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury. Results of the Second National Acute Spinal Cord Injury Study. New England Journal of Medicine 322, 1405–1411. Bracken, M.B., Shepard, M.J., Collins, W.F., Jr, Holford, T.R., Baskin, D.S., Eisenberg, H.M., Flamm, E., Leo-Summers, L., Maroon, J.C., Marshall, L.F., et al. (1992) Methylprednisolone or naloxone treatment after acute spinal cord injury: 1-year follow-up data. Results of the Second National Acute Spinal Cord Injury Study. Journal of Neurosurgery 76, 23–31. Bracken, M.B., Shepard, M.J., Holford, T.R., Leo-Summers, L., Aldrich, E.F., Fazl, M., Fehlings, M., Herr, D.L., Hitchon, P.W., Marshall, L.F., et al. (1997) Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute spinal cord injury. Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial. National Acute Spinal Cord Injury Study. Journal of the American Medical Association 277, 1597–1604. Branson, K.R., Quandt, J.E., Martinez, E.A., Carroll, G.L., Trim, C.M., Dodam, J.R., Hartsfield, S.M., Matthews, N.S., Mackenthun, A., Beleau, M.H. (2001) A multisite case report on the clinical use of sevoflurane in dogs. Journal of the American Animal Hospital Association 37, 420–432. Brown, S.A., Hall, E.D. (1992) Role of oxygen-derived free radicals in the pathogenesis of shock and trauma, with focus on central nervous system injuries. Journal of the American Veterinary Medical Association 200, 1849–1859. Calvert, C.A., Wall, M. (2001) Results of ambulatory electrocardiography in overtly healthy Doberman Pinschers with equivocal echocardiographic evidence of dilated cardiomyopathy. Journal of the American Veterinary Medical Association 219, 782–784.
Preoperative assessment
Clark, D.M. (1986) An analysis of intraoperative and early postoperative mortality associated with cervical spinal decompressive surgery in the dog. Journal of the American Animal Hospital Association 22, 739–744. Cook, S., Prewett, A., Dalton, J., Whitecloud, T.R. (1994) Reduction in perineural scar formation after laminectomy with Polyactive membrane sheets. Spine 19, 1815–1825. Crawford, M.W., Lerman, J., Pilato, M., Orrego, H., Saldivia, V., Carmichael, F.J. (1992) Haemodynamic and organ blood flow responses to sevoflurane during spontaneous ventilation in the rat: a dose–response study. Canadian Journal of Anaesthesia 39, 270–276. Crenshaw, J.T., Winslow, E.H. (2002) Original Research: Preoperative Fasting: Old Habits Die Hard: Research and published guidelines no longer support the routine use of ‘NPO after midnight,’ but the practice persists. American Journal of Nursing 102, 36–44. Culbert, L.A., Marino, D.J., Baule, R.M., Van-Knox, I.-W. (1998) Complications associated with high-dose prednisolone sodium succinate therapy in dogs with neurological injury. Journal of the American Animal Hospital Association 34, 129–134. Diehl, K.J., Roberts, S.M. (1991) Keratoconjunctivitis sicca in dogs associated with sulfonamide therapy: 16 cases (1980–1990). Progress in Veterinary and Comparative Ophthalmology 1, 276–282. Dow, S.W., LeCouteur, R.A., Poss, M.L., Beadleston, D. (1989) Central nervous system toxicosis associated with metronidazole treatment of dogs: Five cases (1984–1987). Journal of the American Veterinary Medical Association 195, 365–368. Dunning, D. (2003) Surgical wound infection and the use of antimicrobials. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 113–121. Philadelphia: Elsevier Science. Egger, C.M., Duke, T., Archer, J., Cribb, P.H. (1998) Comparison of plasma fentanyl concentrations by using three transdermal fentanyl patch sizes in dogs. Veterinary Surgery 27, 159–166. Einhaus, S.L., Robertson, J.T., Dohan, F.C., Jr, Wujek, J.R., Ahmad, S. (1997) Reduction of peridural fibrosis after lumbar laminotomy and discectomy in dogs by a resorbable gel (ADCON-L). Spine 22, 1440–1446; discussion 1446–1447. Endoh, H., Honda, T., Ohashi, S., Hida, S. (2001) [Cerebral autoregulation during sevoflurane or isoflurane anesthesia: evaluation with transient hyperemic response]. Masui 50, 1316–1321. Epstein, M.D., Banducci, D.R., Manders, E.K. (1992) The role of the gastrointestinal tract in the development of burn sepsis. Plastic Reconstructive Surgery 90, 524–531. Fehlings, M.G., Tator, C.H., Linden, R.D. (1989) The effect of nimodipine and dextran on axonal function and blood flow following experimental spinal cord injury. Journal of Neurosurgery 71, 403–416. Fox, L.E., Ford, S., Alleman, A.R., Homer, B.L., Harvey, J.W. (1993) Aplastic anemia associated with prolonged high-dose trimethoprim– sulfadiazine administration in two dogs. Veterinary Clinical Pathology 22, 89–92. Franks, J.N., Boothe, H.W., Taylor, L., Geller, S., Carroll, G.L., Cracas, V., Boothe, D.M. (2000) Evaluation of transdermal fentanyl patches for analgesia in cats undergoing onychectomy. Journal of the American Veterinary Medical Association 217, 1013–1019. Gehlbach, S.H., Bigelow, C., Heimisdottir, M., May, S., Walker, M., Kirkwood, J.R. (2000) Recognition of vertebral fracture in a clinical setting. Osteoporosis International 11, 577–582. Geisler, F.H. (1999) Prevention of peridural fibrosis: current methodologies. Neurological Research 21 Suppl 1, S9–22. Gelatt, K.N., van der Woerdt, A., Ketring, K.L., Andrew, S.E., Brooks, D.E., Biros, D.J., Denis, H.M., Cutler, T.J. (2001) Enrofloxacinassociated retinal degeneration in cats. Veterinary Ophthalmology 4, 99–106. George, E.R., Scholten, D.J., Buechler, C.M., Jordan-Tibbs, J., Mattice, C., Albrecht, R.M. (1995) Failure of methylprednisolone to improve the outcome of spinal cord injuries. American Surgeon 61, 659–663. Gerhart, K.A., Johnson, R.L., Menconi, J., Hoffman, R.E., Lammertse, D.P. (1995) Utilization and effectiveness of methylprednisolone in a populationbased sample of spinal cord injured persons. Paraplegia 33, 316–321. Gerndt, S.J., Rodriguez, J.L., Pawlik, J.W., Taheri, P.A., Wahl, W.L., Micheals, A.J., Papadopoulos, S.M. (1997) Consequences of high-dose steroid therapy for acute spinal cord injury. Journal of Trauma 42, 279–284.
Gill, G., Sakovich, L., Thompson, E. (1979) Pedicle fat grafts for the prevention of scar formation after laminectomy. An experimental study in dogs. Spine 4, 176–186. Griffiths, I.R. (1978) Spinal cord injuries: a pathological study of naturally occurring lesions in the dog and cat. Journal of Comparative Pathology 88, 303–315. Grouhi, M., Hummel, D., Roifman, C.M. (1999) Anaphylactic reaction to oral cefaclor in a child. Pediatrics 103, e50. Hanson, S.M., Bostwick, D.R., Twedt, D.C., Smith, M.O. (1997) Clinical evaluation of cimetidine, sucralfate, and misoprostol for prevention of gastrointestinal tract bleeding in dogs undergoing spinal surgery. American Journal of Veterinary Research 58, 1320–1323. Heary, R.F., Vaccaro, A.R., Mesa, J.J., Northrup, B.E., Albert, T.J., Balderston, R.A., Cotler, J.M. (1997) Steroids and gunshot wounds to the spine. Neurosurgery 41, 576–583. Hinton, L.E., McLoughlin, M.A., Johnson, S.E., Weisbrode, S.E. (2002) Spontaneous gastroduodenal perforation in 16 dogs and seven cats (1982–1999). Journal of the American Animal Hospital Association 38, 176–187. Hoerlein, B.F., Redding, R.W., Hoff, E.J., McGuire, J.A. (1985) Evaluation of naloxone, crocetin, thyrotropin releasing hormone, methylprednisolone, partial myelotomy, and hemilaminectomy in the treatment of acute spinal cord trauma. Journal of the American Animal Hospital Association 21, 67–77. Hoffman, W.E., Edelman, G., Kochs, E., Werner, C., Segil, L., Albrecht, R.F. (1991) Cerebral autoregulation in awake versus isofluraneanesthetized rats. Anesthesia and Analgesia 73, 753–757. Hosgood, G. (2003) Wound repair and specific tissue response to injury. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 66–86. Philadelphia: Elsevier Science. Hurlbert, R.J. (2000) Methylprednisolone for acute spinal cord injury: an inappropriate standard of care. Journal of Neurosurgery 93 (1 Suppl), 1–7. Jeffery, N.D., Blakemore, W.F. (1999a) Spinal cord injury in small animals 2. Current and future options for therapy. Veterinary Record 145, 183–190. Jeffery, N.D., Blakemore, W.F. (1999b) Spinal cord injury in small animals. 1. Mechanisms of spontaneous recovery. Veterinary Record 144, 407–413. Jeffery, N.D., Phillips, S.M. (1995) Surgical treatment of intramedullary spinal cord neoplasia in two dogs. Journal of Small Animal Practice 36, 553–557. Kay-Mugford, P., Benn, S.J., LaMarre, J., Conlon, P. (2000) In vitro effects of nonsteroidal anti-inflammatory drugs on cyclooxygenase activity in dogs. American Journal of Veterinary Research 61, 802–810. Kerwin, S.C., Maudlin, G.E. (2003) Hemostasis, surgical bleeding, and transfusion. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 44–65. Philadelphia: Elsevier Science. Kyles, A.E. (1998) Transdermal fentanyl. Compendium on Continuing Education for the Practicing Veterinarian 20, 721–722, 724–726. Kyles, A.E., Hardie, E.M., Hansen, B.D., Papich, M.G. (1998) Comparison of transdermal fentanyl and intramuscular oxymorphone on post-operative behaviour after ovariohysterectomy in dogs. Research in Veterinary Science 65, 245–251. Kyles, A.E., Papich, M., Hardie, E.M. (1996) Disposition of transdermally administered fentanyl in dogs. American Journal of Veterinary Research 57, 715–719. Lamb, C.R. (1994) Common difficulties with myelographic diagnosis of acute intervertebral disc prolapse in the dog. Journal of Small Animal Practice 35, 549–558. LaRocca, H., Macnab, I. (1974) The laminectomy membrane. Studies in its evolution, characteristics, effects and prophylaxis in dogs. The Journal of Bone and Joint Surgery—British volume 56B, 545–550. LeCouteur, R.A., Sturgess, B. (2003) Spinal fractures and luxations. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1244–1260. Philadelphia: Elsevier Science. Levy, M.L., Gans, W., Wijesinghe, H.S., SooHoo, W.E., Adkins, R.H., Stillerman, C.B. (1996) Use of methylprednisolone as an adjunct in the management of patients with penetrating spinal cord injury: outcome analysis. Neurosurgery 39, 1141–1148; discussion 1148–1149. Lewis, D.D., Hosgood, G. (1992) Complications associated with the use of iohexol for myelography of the cervical vertebral column in
89
90
Small Animal Spinal Disorders
dogs: 66 cases (1988–1990). Journal of the American Veterinary Medical Association 200, 1381–1384. Maddison, J., Ettinger, S.J., Feldman, E.C. (2000) Adverse drug reactions. In: S.J. Ettinger and E.C. Feldman (eds), Textbook of Veterinary Internal Medicine: Diseases of the Dog and Cat, Vols 1 and 2, 5th edn, 320–325. Philadelphia: WB Saunders. Manning, M.M., Brunson, D.B. (1994) Barotrauma in a cat. Journal of the American Veterinary Medical Association 205, 62–64. Maroon, J.C., Abla, A., Bost, J. (1999) Association between peridural scar and persistent low back pain after lumbar discectomy. Neurological Research 21 Suppl 1, S43–46. Martin, R.A., Shell, L., Dodds, W.J. (1986) Focal intramedullary spinal cord hematoma in a dog. Journal of the American Animal Hospital Association 22, 545–550. Mastronardi, L., Pappagallo, M., Puzzilli, F., Tatta, C. (2002) Efficacy of the morphine-Adcon-L compound in the management of postoperative pain after lumbar microdiscectomy. Neurosurgery 50, 518–524; discussion 524–525. Matsumoto, T., Tamaki, T., Kawakami, M., Yoshida, M., Ando, M., Yamada, H. (2001) Early complications of high-dose methylprednisolone sodium succinate treatment in the follow-up of acute cervical spinal cord injury. Spine 26, 426–430. Moore, R.W., Withrow, S.J. (1982) Gastrointestinal hemorrhage and pancreatitis associated with intervertebral disk disease in the dog. Journal of the American Veterinary Medical Association 180, 1443–1447. Myre, S.A., Schoeder, T.J., Grund, V.R., Wandstrat, T.L., Nicely, P.G., Pesce, A.J., First, M.R. (1991) Critical ketoconazole dosage range for ciclosporin clearance inhibition in the dog. Pharmacology 43, 233–241. Nesathurai, S. (1998) Steroids and spinal cord injury: revisiting the NASCIS 2 and NASCIS 3 trials. Journal of Trauma 45, 1088–1093. Nuwer, M.R. (1999) Spinal cord monitoring. Muscle and Nerve 22, 1620–1630. Olby, N. (1999) Current concepts in the management of acute spinal cord injury. Journal of Veterinary Internal Medicine 13, 399–407. Olby, N.J., Jeffery, N.D. (2003) Pathogenesis of diseases of the central nervous system. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1132–1147. Philadelphia: Elsevier Science. Olby, N.J., Munana, K.R., Sharp, N.J.H., Thrall, D.E. (2000) The computed tomographic appearance of acute thoracolumbar intervertebral disc herniations in dogs. Veterinary Radiology and Ultrasound 41, 396–402. Olby, N.J., Harris, T., Munana, K.R., Skeen, T.M., Sharp, N.J.H. (2003) Long-term functional outcome of dogs with severe injuries of the thoracolumbar spinal cord. Journal of the American Veterinary Medical Association 222, 762–769. Papadogiannakis, E.I. (2000) Cutaneous adverse drug reactions in the dog and cat. European Journal of Companion Animal Practice 10, 71–77. Parent, C., King, L.G., Van Winkle, T.J., Walker, L.M. (1996) Respiratory function and treatment in dogs with acute respiratory distress syndrome: 19 cases (1985–1993). Journal of the American Veterinary Medical Association 208, 1428–1433. Parker, A.J., Smith, C.W. (1972) Functional recovery following incision of spinal meninges in dogs. Research in Veterinary Science 13, 418–421. Parker, A.J., Smith, C.W. (1974) Functional recovery from spinal cord trauma following incision of spinal meninges in dogs. Research in Veterinary Science 16, 276–279. Parker, A.J., Smith, C.W. (1975) Functional recovery from spinal cord trauma following delayed incision of spinal meninges in dogs. Research in Veterinary Science 18, 110–112. Perkins, P.G., Deane, R.H. (1988) Long-term follow-up of six patients with acute spinal injury following dural decompression. Injury 19, 397–401. Reed, S. (2002) Nonsteroidal anti-inflammatory drug-induced duodenal ulceration and perforation in a mature rottweiler. Canadian Veterinary Journal 43, 971–972. Richter, H.P., Kast, E., Tomczak, R., Besenfelder, W., Gaus, W. (2001) Results of applying ADCON-L gel after lumbar discectomy: the German ADCON-L study. Journal of Neurosurgery 95, 179–189. Robertson, J., Meric, A., Dohan, F.J., Schweitzer, J., Wujek, J., Ahmad, S. (1993) The reduction of postlaminectomy peridural fibrosis in rabbits by a carbohydrate polymer. Journal of Neurosurgery 79, 89–95.
Robinson, T.M., Kruse Elliott, K.T., Markel, M.D., Pluhar, G.E., Massa, K., Bjorling, D.E. (1999) A comparison of transdermal fentanyl versus epidural morphine for analgesia in dogs undergoing major orthopedic surgery. Journal of the American Animal Hospital Association 35, 95–100. Rohrer, C.R., Hill, R.C., Fischer, A., Fox, L.E., Schaer, M., Ginn, P.E., Casanova, J.M., Burrows, C.F. (1999a) Gastric hemorrhage in dogs given high doses of methylprednisolone sodium succinate. American Journal of Veterinary Research 60, 977–981. Rohrer, C.R., Hill, R.C., Fischer, A., Fox, L.E., Schaer, M., Ginn, P.E., Preast, V.A., Burrows, C.F. (1999b) Efficacy of misoprostol in prevention of gastric hemorrhage in dogs treated with high doses of methylprednisolone sodium succinate. American Journal of Veterinary Research 60, 982–985. Rosenbluth, P.R., Meirowsky, A.M. (1953) Sympathetic blockade, an acute cervical cord syndrome. Journal of Neurosurgery 10, 107–112. Rosin, E. (1988) The timing of antibiotic administration for antimicrobial prophylaxis in surgery. Veterinary Surgery 17, 181. Ross, J., Obuchowski, N., Modic, M. (1999) MR evaluation of epidural fibrosis: proposed grading system with intra- and inter-observer variability. Neurological Research 21 Suppl 1, S23–26. Ross, J.S., Robertson, J.T., Frederickson, R.C., Petrie, J.L., Obuchowski, N., Modic, M.T., deTribolet, N. (1996) Association between peridural scar and recurrent radicular pain after lumbar discectomy: magnetic resonance evaluation. ADCON-L European Study Group. Neurosurgery 38, 855–861; discussion 861–863. Roulet, A., Puel, O., Gesta, S., Lepage, J.F., Drag, M., Soll, M., Alvinerie, M., Pineau, T. (2003) MDR1-deficient genotype in Collie dogs hypersensitive to the P-glycoprotein substrate ivermectin. European Journal of Pharmacology 460, 85–91. Rowland, P.H., Center, S.A., Dougherty, S.A. (1992) Presumptive trimethoprim–sulfadiazine-related hepatotoxicosis in a dog. Journal of the American Veterinary Medical Association 200, 348–350. Salisbury, S.K., Cook, J.R., Jr (1988) Recovery of neurological function following focal myelomalacia in a cat. Journal of the American Animal Hospital Association 24, 227–230. Sapolsky, R.M. (1994) Glucocorticoids, stress and exacerbation of excitotoxic neuron death. Seminars in Neuroscience 6, 323–331. Scherk Nixon, M. (1996) A study of the use of a transdermal fentanyl patch in cats. Journal of the American Animal Hospital Association 32, 19–24. Scott, D.W., Miller, W.H., Jr (1998) Idiosyncratic cutaneous adverse drug reactions in the cat: literature review and report of 14 cases (1990–1996). Feline Practice 26, 10–15. Scott, D.W., Miller, W.H., Jr (1999) Idiosyncratic cutaneous adverse drug reactions in the dog: literature review and report of 101 cases (1990–1996). Canine Practice 24, 16–22. Seelig, C.B., Maloley, P.A., Campbell, J.R. (1990) Nephrotoxicity associated with concomitant ACE inhibitor and NSAID therapy. Southern Medical Journal 83, 1144–1148. Seim, H.B., Prata, R.G. (1982) Ventral decompression for the treatment of cervical disk disease in the dog: a review of 54 cases. Journal of the American Animal Hospital Association 18, 233–240. Siemering, G.B., Vroman, M.L. (1992) High dose methylprednisolone sodium succinate: an adjunct to surgery for canine intervertebral disc herniation. Veterinary Surgery 21, 406. Smith, G.K., Walter, M.C. (1985) Fractures and luxations of the spine. In: C.D. Newton and D. Nunamaker (eds), Textbook of Small Animal Orthopedics, 307–322. Philadelphia: JB Lippincott. Smith-Swintosky, V.L., Pettigrew, L.C., Sapolsky, R.M., Phares, C., Craddock, S.D., Brooke, S.M., Mattson, M.P. (1996) Metapyrone, an inhibitor of glucocorticoid production, reduces brain injury induced by focal and global ischemia and seizures. Journal of Cerebral Blood Flow and Metabolism 16, 585–598. Songer, M., Ghosh, L., Spencer, D. (1990) Effects of sodium hyaluronate on peridural fibrosis after lumbar laminotomy and discectomy. Spine 15, 550–554. Songer, M.N., Rauschning, W., Carson, E.W., Pandit, S.M. (1995) Analysis of peridural scar formation and its prevention after lumbar laminotomy and discectomy in dogs. Spine 20, 571–580; discussion 579–580.
Preoperative assessment
Stauffer, J.L., Gleed, R.D., Short, C.E., Erb, H.N., Schukken, Y.H. (1988) Cardiac dysrhythmias during anesthesia for cervical decompression in the dog. American Journal of Veterinary Research 49, 1143–1146. Stillman, M.T. (1989) Interaction and selection of therapeutic agents in the elderly: NSAIDs and the ageing kidney. Scandinavian Journal of Rheumatology Supplement 82, 33–38. Summors, A.C., Gupta, A.K., Matta, B.F. (1999) Dynamic cerebral autoregulation during sevoflurane anesthesia: a comparison with isoflurane. Anesthesia and Analgesia 88, 341–345. Tator, C.H. (1991) Review of experimental spinal cord injury with emphasis on the local and systemic circulatory effects. Neurochirurgie 37, 291–302. Tator, C.H., Fehlings, M.G. (1991) Review of the secondary injury theory of acute spinal cord trauma with emphasis on vascular mechanisms. Journal of Neurosurgery 75, 15–26. Teague, H.D., Brasmer, T.H. (1978) Midline myelotomy of the clinically normal canine spinal cord. American Journal of Veterinary Research 39, 1584–1590. Thacher, C. (1989) Problems with owner communications. Problems in Veterinary Medicine 1, 478–490. Toombs, J.P., Collins, L.G., Graves, G.M., Crowe, D.T., Caywood, D.D. (1986) Colonic perforation in corticosteroid-treated dogs. Journal of the American Veterinary Medical Association 188, 145–150. Trevor, P.B., Martin, R.A., Saunders, G.K., Trotter, E.J. (1991) Healing characteristics of free and pedicle fat grafts after dorsal laminectomy and durotomy in dogs. Veterinary Surgery 20, 282–290.
Trotter, E.J., Crissman, J., Robson, D., Babish, J. (1988) Influence of nonbiologic implants on laminectomy membrane formation in dogs. American Journal of Veterinary Research 49, 634–643. Weinblatt, M.E. (1989) Drug interactions with non steroidal anti-inflammatory drugs (NSAIDs). Scandinavian Journal of Rheumatology Supplement 83, 7–10. Welch, J.A., Wohl, J.S., Wright, J.C. (2002a) Evaluation of postoperative respiratory function by serial blood gas analysis in dogs treated with transdermal fentanyl. Journal of Veterinary Emergency and Critical Care Medicine 12, 81–87. Welch, W.C., Cornwall, G.B., Toth, J.M., Turner, A.S., Thomas, K.A., Gerszten, P.C., Nemoto, E.M. (2002b) Use of polylactide resorbable film as an adhesion barrier. Orthopedics 25 (10 Suppl), S1121–1130. Wheeler, S.J., Davies, J.V. (1985) Iohexol myelography in the dog and cat: a series of one hundred cases, and a comparison with metrizamide and iopamidol. Journal of Small Animal Practice 26, 247–256. Widmer, W.R., DeNicola, D.B., Blevins, W.E., Cook, J.R., Jr, Cantwell, H.D., Teclaw, R.F. (1992a) Cerebrospinal fluid changes after iopamidol and metrizamide myelography in clinically normal dogs. American Journal of Veterinary Research 53, 396–401. Widmer, W.R., Blevins, W.E., Jakovljevic, S., Teclaw, R.F., Han, C.M., Hurd, C.D. (1992b) Iohexol and iopamidol myelography in the dog: a clinical trial comparing adverse effects and myelographic quality. Veterinary Radiology and Ultrasound 33, 327–333.
91
Cervical disc disease
Clinical signs
94
Treatment options 96 Non-surgical treatment 96 Surgical treatment 96 Fenestration 96 Ventral decompression 96 Ventral decompression and fixation Dorsal decompression 98
CLINICAL SIGNS
97
Complications 98 Intraoperative 98 Early postoperative 99 Postoperative care Prognosis
102
102
Cervical disc disease in cats
103
Key issues for future investigation References
7
et al., 1992). Dachshunds, Beagles, Poodles, Spaniels, Shih Tzus, Pekingese and Chihuahuas are affected most often. Large-breed dogs also suffer from cervical disc disease, usually as part of the syndrome of cervical spondylomyelopathy (see Chapter 11). Most small dogs suffer signs after 2 years of age, with a mean onset at 6 years (Goggin et al., 1970; Gage, 1975).
93
Diagnosis 94 Survey radiography CSF analysis 94 Myelography 95 CT and MRI 95
Chapter
The predominant clinical sign is severe neck pain (7.1 and Box 7.1), which may be acute or chronic (Fry et al., 1991; Smith et al., 1997). In making a diagnosis, it is usually adequate to palpate the spine and muscles of the neck rather than flex and extend the neck to confirm pain (2.13). This is one of the few conditions that cause dogs to scream spontaneously. Often the pain is unremitting and not responsive to medication although dogs with more marked neurological deficits often show less pain. Affected dogs may be reluctant to eat unless the food is raised off the floor. Nerve ‘root signature’ (pain on traction of the limb) is another frequent finding (Seim and Prata, 1982; Fry et al., 1991; Morgan et al., 1993). It may present as
103
104
Procedures 106 Approach to the ventral neck 106 Fenestration 109 Ventral decompression 111 Ventral decompression with distraction-stabilization 117 Dorsal hemilaminectomy 119 Dorsal laminectomy 120 Cervical disc disease is a frequent disorder of dogs. Small dogs are affected commonly, particularly those with chondrodystrophoid characteristics although the condition can occur in any breed (Gage, 1975; Dallman
7.1 Dachshund with neck pain caused by a cervical disc extrusion. Note low head position and thoracic limb held up due to root signature. The hunched posture can be mistaken for thoracolumbar pain; history and physical examination differentiates the two.
94
Small Animal Spinal Disorders
an apparent orthopedic lameness but nerve root pain can usually be elicited by neck palpation or limb traction. This sign usually indicates that more caudal discs are affected but can also be seen with C2/3 discs (Morgan et al., 1993). Neurological deficits may be restricted to one thoracic limb or the dog may show Box 7.1 Clinical signs of cervical disc disease ■
Neck pain
■
Spontaneous screaming
■
Low head carriage
■
Thoracic limb lameness or paresis
■
Thoracic limb ‘nerve root signature’
■
Hemiparesis
■
Tetraparesis
hemiparesis, tetraparesis or even tetraplegia with hypoventilation (6.1, 7.11). Neurological deficits are more common with lesions at C4/5 to C6/7 inclusive, while neck pain without deficits is more common with lesions at C2/3 and C3/4 (Lemarie et al., 2000). This may reflect the greater degree of space in the cranial vertebral canal compared to more caudally. Most dogs have Hansen type I extrusions. Hansen type II herniations do occur, generally in larger breed dogs (7.56). The C2/3 disc is involved most frequently, with the incidence decreasing caudally (Seim and Prata, 1982; Fry et al., 1991). Lesions at C6/7 are the least common, with the exception of large-breed dogs as part of cervical spondylomyelopathy and also in Pekingese (Fry et al., 1991). The C7/T1 disc herniates occasionally (7.56). Intracranial lesions can cause neck pain on rare occasions and should be considered under differential diagnosis (Coates and Dewey, 1998) (Box 7.2).
Box 7.2 Differential diagnoses for neck pain
DIAGNOSIS Survey radiography
■
Schmorl’s node
■
Facet joint pain
■
Synovial cyst
■
Bicipital bursitis
■
Temporomandibular joint (TMJ) lesion or oropharyngeal pain
■
Otitis media
■
Cervical spondylomyelopathy
■
Atlantoaxial subluxation
■
Syringohydromyelia
■
Osteoporotic pathological fracture
■
Soft tissue tumor in the neck or bone tumor
■
Intracranial lesion
■
Mid-thoracic lesion, e.g. T5/6 discospondylitis
CSF analysis
■
Thoracic lesion, e.g. pleuritic pain
■
Meningomyelitis
■
Polyarthritis
Analysis of CSF is useful to eliminate inflammatory CNS disease. Results of CSF analysis may be abnormal in disc disease, but elevations of protein and cells are usually mild (Thomson et al., 1989). Conditions other than disc disease are more common in dogs less than 2 years old or in aged animals (Box 7.2).
■
Polymyositis
■
Spinal cord hematoma or hemorrhage
Diagnosis is based on the clinical signs described above. Survey radiographic features of chronic disc disease are common incidental findings in older dogs. Narrowing of the intervertebral space or dorsal displacement of mineralized disc material is suggestive of disc extrusion (Morgan et al., 1993) (7.2). Discography may be useful for non-mineralized, lateral extrusions (Felts and Prata, 1983; Wrigley and Reuter, 1984). Myelography or advanced imaging is necessary for definitive diagnosis (Somerville et al., 2001).
7.2 A: Lateral myelogram from a 10-year-old Shih Tzu that presented with neck pain and no neurological deficits. There is a narrow intervertebral space at C5/6 with extruded, mineralized disc material and splitting of the ventral contrast column at this level (4.23–4.25). B: Dorsoventral myelogram. Further images from the same dog are shown in 7.3, 7.9 and 7.37.
A
B
Cervical disc disease
Myelography
CT and MRI
Myelography or advanced imaging is necessary to confirm the diagnosis and because multiple discs are sometimes involved. In some lateral or intraforaminal extrusions the myelogram is normal but oblique views may reveal the offending disc (Felts and Prata, 1983). Oblique views are also useful in determining the side on which an asymmetrical extrusion lies, although CT or MRI are better.
Depending on the scanner, CT is usually much faster to perform than myelography and is also more accurate for surgical planning (Olby et al., 1999). When available, CT is the imaging modality of choice for chondrodystrophoid dogs with suspected cervical disc disease (4.39B, 7.4, 7.5). The cross-sectional image helps the surgeon to plan the ventral slot and is also useful to identify lateralized or foraminal lesions (Hara et al., 1994;
C3
A
C4
C5
C6
7.3 A: CT scan through C5/6 made after myelography (same dog as shown in 7.2). The connection between the disc extrusion and the rest of the nucleus pulposus is seen clearly. B: T2-weighted, sagittal MRI to show a disc extrusion at C4/5 (transverse image, 4.44A). The C3/4 and C5/6 discs are of relatively normal signal intensity but C2/3 and especially C4/5 and C6/7 discs are degenerate with low signal intensity (Sether et al., 1990; Levitski et al., 1999).
B 7.4 A: Transverse CT and B: 3D reconstruction of CT scan from a 5-year-old Lhasa Apso that had neck pain and was unable to walk. The scan revealed a large mineralized extrusion at the C4/5 disc space. The postoperative scan is shown in 7.5.
A
B
7.5 A: Transverse and B: 3D reconstruction of the post-surgical CT scan from the same dog as shown in 7.4. A ventral slot was performed, which permitted the surgeon to remove the majority of the extruded material. Note the small amount of residual material present to one side of the slot (arrowhead and arrow). The width of this slot is 39% of the vertebral body width.
A
B
95
96
Small Animal Spinal Disorders
7.6 Dog with severe neck pain and root signature but no neurological deficits. A: Transverse CT scan and B: 3D reconstruction reveals disc material within the intervertebral foramen at C5/6 (arrowheads). The disc material was removed using a blunt probe after a standard ventral approach with elevation of the longus colli muscles from the ventral and lateral aspects of the anulus fibrosus. Fenestration was also performed. The dog was pain free within 2 days. A
B
Bagley et al., 1996) (7.6). The scanner can also be used to show the exact location and extent of disc material on a scout image (4.39A). A heterogenous, hyperattenuating, extradural mass with loss of epidural fat is a characteristic feature of a mineralized disc extrusion (Olby et al., 2000). If CT is not diagnostic it can be followed by a myelogram or a CT myelogram (Hara et al., 1994). For more information on performing and interpreting CT images in disc disease, see Chapter 4, page 55–57. MRI may be better than CT, particularly when the disc material is not mineralized, but scan times are usually longer and the cost is higher (Levitski et al., 1999) (4.42, 4.44A, 7.3B).
TREATMENT OPTIONS Treatment may be non-surgical or surgical.
Non-surgical treatment This entails cage rest and use of anti-inflammatory medications. It can be tried in any patient unless marked neurological deficits are present. Either non-steroidal anti-inflammatory drugs (NSAIDs), low-dose prednisolone, or narcotics may be used, sometimes combined with diazepam or methocarbamol (Tables 15.1, 15.2). Acupuncture may also be of benefit (Janssens, 1985). The catastrophic worsening of neurological status that can occur with medical treatment of thoracolumbar discs is rare with cervical disc disease. However, the neck pain in cervical disc disease seems to be less responsive to non-surgical treatment than does the pain from thoracolumbar disc disease. Progression of signs or lack of response in 1–2 weeks indicates treatment failure. A dog that is responding well to non-surgical treatment should be kept rested for at least 6 weeks after clinical signs have resolved. Recurrence of clinical signs after non-surgical treatment has been reported in over 30% of patients (Russell and Griffiths, 1968; Janssens, 1985). The role of chemonucleolysis is unclear and cervical discs must first be exposed surgically if this procedure is planned (Atilola et al., 1993).
Surgical treatment Indications for surgical treatment include: • Failure of non-surgical treatment. • Unremitting pain. • Severe or progressive neurological deficits. Ventral decompression is the preferred procedure. It may need to be combined with stabilization for caudal disc extrusions. Dorsal or dorsolateral decompression may be required in extrusions that cannot be reached via ventral slot or if there is doubt about the diagnosis.
Fenestration (7.30–7.36) Although the value and desirability of fenestration has been questioned (Fingeroth, 1989), it should prevent further extrusion of disc material into the vertebral canal and so reduce the recurrence rate (Russell and Griffiths, 1968). It also appears effective for dogs with discogenic pain (Morgan et al., 1993 (Algorithm 7.1)). It is usual to fenestrate the discs from C2/3 to C5/6 inclusive; C6/7 is fenestrated if there is evidence of disease. Advantages and disadvantages of fenestration compared to ventral slot decompression are shown in Table 7.1. Fenestration is only recommended as a primary procedure for dogs with discogenic pain and as a prophylactic procedure in combination with ventral decompression. In small breed dogs that develop signs suggestive of ‘Wobbler syndrome’, fenestration is probably contraindicated as it may exacerbate bulging of the dorsal anulus (7.15A). Ventral decompression (7.37–7.52) General indications for decompression include: Presence of neurological deficits. Spinal cord compression on neuroimaging. Failure of fenestration. Removal of disc material by ventral slot decompression provides the most rapid resolution of clinical signs and it is therefore the treatment of choice. Accurate
• • •
Cervical disc disease
Neuroimaging No spinal cord compression Traction non-responsive Reassess diagnosis
Traction responsive (small dogs*)
Single lesions
Fenestrate for discogenic pain
Cranial discs C2/3 to C3/4
Multiple lesions
Caudal discs C4/5 to C6/7 Dorsal laminectomy
C7/T1 or very lateral extrusion
Ventral slot
Dorsolateral hemilaminectomy or dorsal laminectomy
Ventral slot and stabilize if width close to 50%
Distraction (7.55)
*less than 10 kg body weight
Algorithm 7.1 Surgical decision-making in cervical disc disease.
Table 7.1 Comparison of ventral fenestration and slot Fenestration
Slot
Technically
Easy
Difficult
Accurate identification of disc involved
Not required
Required
Special equipment required
No
Yes
Potential for iatrogenic damage
Unlikely
Possible
Removal of disc material from vertebral canal
No
Yes
Resolution of neck pain
Slow in many dogs
Usually within a few days
identification of the disc involved is an obvious prerequisite. The width of the slot should be about one third the width of the vertebral body and certainly no more than 50% (7.10, 7.16). These parameters are especially important for the C4/5 to C6/7 discs inclusive (Fitch et al., 2000; Lemarie et al., 2000). The inverted cone technique can be used to improve access while minimizing any potential instability (Goring et al., 1991) (7.50).
Postoperative fusion occurs in a proportion of dogs following ventral decompression but how well may depend on the width of the slot (11.12); narrow slots are less likely to fuse than wide slots (Gilpin, 1976; Seim and Prata, 1982). If osseous fusion does occur across the slot it can then predispose even small breed dogs to domino lesions, although probably to a lesser degree than it does in large-breed dogs (Prata and Stoll, 1973; Bagley et al., 1993) (7.14).
Ventral decompression and fixation (7.53–7.55) Postoperative instability or subluxation are important potential complications that can be prevented by fixation of the interspace at the time of ventral decompression (Fitch et al., 2000; Lemarie et al., 2000). Fixation is not usually necessary for cranial disc lesions (C2/3 and C3/4) but should be considered for caudal lesions (C4/5–C6/7) when the slot width is nearing 50% of the vertebral body width (Lemarie et al., 2000) (7.10, 7.16). Fixation is probably not necessary for caudal lesions if the slot dimensions are less than 50% of the vertebral body width. When subluxation does occur following ventral decompression the rescue technique of choice is distraction-stabilization. Several methods can be used for fixation after ventral slot (see also pages 118, 220). Cement plugs are very difficult to use in small dogs (7.13, 7.54). Bone autografts
97
98
Small Animal Spinal Disorders
from the ilium are effective but necessitate two surgical approaches (Prata and Stoll, 1973). A more recent study preferred a bone allograft block placed in the interspace without use of additional implants (Lemarie et al., 2000). Autografts give better radiographic fusion than allografts but other factors such as graft site morbidity also affect final outcome in humans (Floyd and Ohnmeiss, 2000). Whatever method is used, bony fusion should be encouraged so that long-term fixation does not depend solely on the implant(s). The prognosis is good for dogs that subluxate after a ventral slot provided that the site can be stabilized (Lemarie et al., 2000). In general, adherence to the recommendation to limit the width of the slot to near 33% is preferable to managing potential or subsequent complications.
Dorsal decompression (7.56–7.59) Dorsal laminectomy is much easier technically in small dogs than large dogs and short-term morbidity is less of a problem (Gill et al., 1996; De Risio et al., 2002). Dorsal laminectomy has been proposed as an alternative to ventral decompression for small dogs but ventral slot has the advantage of permitting removal of disc material (Gill et al., 1996; Fitch et al., 2000). Dorsal laminectomy is therefore best reserved for lesions at C7/T1, where there is doubt regarding the diagnosis, or for dogs with multiple lesions (7.15, 7.59). Hemilaminectomy can be performed in the cervical region using a lateral approach but a dorsolateral approach is easier (Lipsitz and Bailey, 1995) (7.56–7.58, 12.15–12.29). Hemilaminectomy is indicated where disc material is situated too laterally to access via ventral slot, for some lesions at C7/T1 (7.56), or if there is doubt about the diagnosis (Seim and Prata, 1982). Intraforaminal extrusions approached in this way or laterally (Lipsity & Bailey, 1995) may occasionally be accessible without entering the vertebral canal (Prata and Stoll, 1973; Felts and Prata, 1983) (7.6).
occur if the intervertebral space is explored recklessly. Other technical errors can occur during fenestration, ventral decompression (7.7, 7.8), or implant placement (Box 7.3). Hemorrhage can be problematical at various stages. Dorsal laminectomy can damage vessels in the epidural space, especially at the level of a foramen (Hurov, 1979) (7.59, 11.51, 11.52). During a ventral approach the longus colli muscles tend to bleed when removed
C4
C5
7.7 This dog presented with neck pain and no neurological deficits. Myelography revealed a mineralized extrusion at the C4/5 space. A ventral slot was performed with fenestration between C2/3 and C6/7 inclusive. There was considerable hemorrhage from the vertebral venous plexus and no disc material was retrieved from the vertebral canal. It was assumed that the response would be at least equivalent to a fenestration; re-examination was scheduled in 2 weeks.
C4 C5
COMPLICATIONS Intraoperative The ventral surgical approach itself has few complications unless the surgeon does not identify the midline correctly and damages the vertebral artery. It is also possible to damage vital structures such as the recurrent laryngeal nerve (7.25, 7.54); the esophagus can be perforated if mistaken for the longus colli muscle and pleura or other vital structures may be damaged at the level of the thoracic inlet (Funkquist and Svalastoga, 1979). Spinal cord damage during fenestration can
7.8 Two weeks after surgery the dog was tetraparetic and unable to stand. The ventral slot had been performed at the correct interspace; the cranial margin is just visible in the caudal portion of C4 (arrowhead). However, the degree of spinal cord compression is now worse than before. The slot was re-explored; it was noted to have been directed to one side. The slot was redirected to the midline and a large amount of disc material was removed; hemorrhage was minimal. The dog made a good recovery and was able to walk 2 weeks later.
Cervical disc disease
from their insertion on the ventral process. Muscle dissection should be restricted to the midline to avoid bleeding from the vertebral artery or its branches. The vertebral arteries may also be damaged during removal of lateral or foraminal disc material from either a ventral or a lateral approach (Felts and Prata, 1983) (7.6). Bleeding from the bone during drilling can be controlled with bone wax. The vertebral venous plexus (7.43, 7.44, 11.29B) is damaged easily and hemorrhage can be so severe that it is impossible to continue to explore the vertebral canal (7.7); it may occasionally even be fatal (Clark, 1986). Severe hemorrhage may occur in as many as one quarter of dogs undergoing ventral slot, even without pre-existing coagulopathy (Smith et al., 1997). Concurrent use of aspirin or the presence of any coagulopathy increases the risk (15.2). Steps recommended in case of severe venous plexus hemorrhage are outlined in Table 7.2. Venous plexus hemorrhage can lead on rare occasions to complications in the early postoperative period. Two dogs that had venous plexus hemorrhage during ventral decompression developed acute, progressive tetraparesis within 12 h. Both had large blood clots compressing the spinal cord but recovered well after evacuation combined with hemostasis using thrombin (Seim and Prata, 1982) (Table 7.2).
Box 7.3 Intraoperative complications
Early postoperative These relate mainly to continued pain; neurological deterioration (7.8); or respiratory complications (Box 7.4). After fenestration neck pain can persist in a significant number of dogs and may take 1–2 months to resolve completely (Denny, 1978) (see ‘Prognosis’, page 102). Neurological deterioration after fenestration can also occur, probably when incorrect fenestration technique leads to disc material being forced into the vertebral canal (Tomlinson, 1985). After ventral slot decompression neck pain should improve within a few days. Moderate or severe pain has been reported 3 days after surgery in up to 65% of dogs although this could have been due to excessive slot width in some of these dogs (Fitch et al., 2000). Persistence of severe pain beyond 72 h, or progressive neurological deficits, warrant repeat imaging (Seim and Box 7.4 Early postoperative complications ■
Residual or increased compression at the site (7.8)
■
Residual material within a foramen
■
Subluxation or instability at the surgical site (7.10, 7.16, 7.52)
■
Extradural hematoma
■
Implant complications (7.13, 7.54)
■
New extrusion at another site
■
Hyperflexion injury
■
Dyspnea (7.12)
■
Infection or sepsis Vascular decompensation of spinal cord
■
Iatrogenic injury
■
■
Cardiopulmonary arrest
■
Hypoventilation (6.1, 7.11)
■
Implant complications
■
Edema (7.12)
■
Hemorrhage (7.7)
■
Seroma or hematoma of wound
■
Disc extrusion due to fenestration
■
Horner’s syndrome
■
Ventral slot too wide (7.16)
■
Megaesophagus
■
Surgery performed at the wrong site
■
Laryngeal paralysis
■
Redistribution of disc material
■
Aspiration
Inadequate decompression (7.8)
■
Pneumonia
■
Table 7.2 Hemorrhage from vertebral venous plexus Preoperative
Intraoperative
Postoperative
Evaluate for coagulopathy Cross-match for high-risk patients
Place finger over the slot if severe Relieve retraction pressure on jugular veins Plug slot with macerated muscle tissue Pack slot with Gelfoam Occlude plexus with direct pressure (11.29B) Avoid long-term continuous suction of hemorrhage Increase fluid rate, consider plasma or blood product
Check for neurological deterioration
99
100
Small Animal Spinal Disorders
R
A R
7.10 Ventrodorsal radiograph of a dog where subluxation had occurred following a very wide ventral slot (59%). The dog had severe neck pain for a week after surgery, which prompted repeat radiography. Treatment by distraction-stabilization was attempted but the dog died (7.16, 7.52). B 7.9 Residual disc material does not always cause persistent pain. A: Preoperative 3D reconstruction of the CT myelogram shown in 7.3B. There is a large mass of mineralized material on the left side of the floor of the vertebral canal. B: 3D reconstruction made immediately following surgery. A ventral slot has been performed but removal of disc material was suboptimal. Despite this, the dog was almost pain free the day after surgery and was normal within a few days. The slot width in this dog is approximately 38%.
Prata, 1982; Smith et al., 1997). This may mean that some dogs are imaged unnecessarily (7.9) but will also identify quickly any dogs that need further surgery. Potential causes of severe postoperative pain are shown in Box 7.4. One common cause is when some of the disc material has simply been moved to a new location during attempts at retrieval, which sets up a new area of irritation. A poorly executed slot is actually no better than a fenestration and can be worse (Chambers et al., 1986) (4.25, 7.8, 11.21). Vertebral instability is another important cause of persistent pain after surgery and can occur if the slot is made too wide (Seim and Prata, 1982). Range of motion increases in cadaver spines by 30–40% after fenestration and by 66% after a ventral slot, even if the slot is only one third of the vertebral width (Macy
et al., 1999; Wolf and Roe, personal communication). Subluxation leads to severe pain or marked deterioration in neurological status due to nerve root, meningeal or spinal cord compression (7.10, 7.16, 7.52). Signs associated with this usually occur within a week of surgery but can be delayed for up to 3 months (Lemarie et al., 2000). If subluxation does occur, the lesion should be managed by distraction-stabilization, as for a traumatic fracture or subluxation. It is likely that instability is under-recognized as a cause of continued pain after surgery (Macy et al., 1999; Fitch et al., 2000, Lemarie et al., 2000). Infection at the surgical site may cause either incisional drainage or systemic signs (Chambers et al., 1982; Fry et al., 1991; Lipsitz and Bailey, 1995). It can occur following dorsal or ventral decompression, especially if the dog becomes bacteremic for any reason (see page 84). Discospondylitis (14.11) or even epidural abscessation (14.14) may also develop. Respiratory arrest during or after surgery has been reported in several studies (Clark, 1986; Waters, 1989; Smith et al., 1997; Beal et al., 2001). This can be due to either cardiopulmonary or neurological abnormalities (see Chapter 6, page 82). Some dogs develop more severe neurological deficits following surgery so that
Cervical disc disease
C7 A
7.11 Some dogs only need to be ventilated for a day or two but others require ventilation for 1–2 weeks before they can be weaned off mechanical support (Beal et al., 2001) (6.1).
C7
B
7.12 This dog shows profound swelling 2 days after a ventral slot; the base of the tongue was also affected. It caused this dog no problems but can cause dysphagia or dyspnea in some animals. Hot packing and anti-inflammatory medications are indicated (Jerram et al., 1997) (see page 357). The possibility of infection should be investigated.
they are unable to breathe spontaneously. In these animals the only option is to put them on a mechanical ventilator. Ventilation should be considered once the paCO2 exceeds 50 mmHg and the dog should either be ventilated or euthanized on humane grounds if the paCO2 is greater than 70 mmHg (7.11). Swelling or edema in the ventral neck can also cause problems after surgery (Fry et al., 1991; Smith et al., 1997) (7.12). Care with hemostasis and wound closure helps to avoid this. Seroma is a particular risk after dorsal decompression (15.34). Technical problems may occur when trying to place implants in small dogs (7.13, 7.54). Other potential
7.13 Nine-year-old pug with chronic tetraparesis localizing to C6–T2 spinal cord. A: There is severe spinal cord compression at C5/6 and C6/7, which was reduced partially by traction. Cement plugs were placed at C5/6 and C6/7 using anchor holes in the end plates and the dog was put in a splint (Dixon et al., 1996) (11.39). B: It developed severe neck pain 5 days later; neurological status was unchanged. A myelogram showed the lesion at C6/7 had improved but C5/6 was worse. Disc material could have extruded spontaneously, been forced into the canal during plug placement, or the distraction had failed. There was no evidence of cement in the canal. The dog was euthanized; no necropsy was performed.
complications include trauma during recovery (11.22); megaesophagus (15.40) and Horner’s syndrome (Boydell, 1995); seizures after myelography; urinary tract infection; diarrhea and sepsis (Fry et al., 1991). Any dog that has undergone ventral decompression in the past, or that has congenitally fused vertebrae, could go on to develop a domino lesion. The incidence is much lower than in ‘Wobbler syndrome’ but it does occur (Prata and Stoll, 1973; Bagley et al., 1996) (7.14). The overall risks of serious complications or mortality with cervical surgery are higher than in thoracolumbar surgery. The highest mortality rate is for dogs that are unable to walk prior to surgery, especially those with pre-existing disease. Nine of 37 such dogs (25%) in one study died or were euthanized. Causes of death
101
102
Small Animal Spinal Disorders
T1 T1
A
B
included cardiac arrest, pulmonary thromboembolism and respiratory dysfunction (Waters, 1989). Recumbent dogs are also at high risk of developing pneumonia, as are those that develop laryngeal paralysis (Prata and Stoll, 1973; Lemarie et al., 2000) (see Chapter 15). Three of 52 dogs in another study died after ventral decompression for cervical disc herniation: one from uncontrollable venous plexus hemorrhage and two from acute bradycardia and hypotension (Clark, 1986). Two of these dogs presented with only neck pain. Dysrhythmias are common during cervical disc surgery. They were reported in 15 of 48 dogs (31%), two of which died, emphasizing the importance of good anesthetic monitoring (Stauffer et al., 1988). Hypotension is also common during ventral decompression and may influence outcome (Griffiths, 1973; Cybulski, 1988; Smith et al., 1997) (see page 86). In summary, a high index of suspicion for complications must be maintained if the surgery does not go as planned, if the response to surgery is poor and certainly if the animal’s neurological status worsens after surgery (7.7). In such cases, repeat imaging is strongly recommended (7.7–7.9).
POSTOPERATIVE CARE
7.14 A: Eight-year-old Silky terrier (5 kg) with a disc extrusion at C5/6. A ventral slot was performed and the dog made a good recovery. B: Eighteen months later the dog showed recurrence of neck pain, root signature, mild tetraparesis and ataxia. A repeat myelogram revealed no residual compression at C5/6 but extrusion of disc material at C4/5 and C6/7. Note the fusion at C5/6 (7.53, 7.54).
(see Chapter 15)
Most patients should be much less painful within a day or two of ventral slot surgery. Two weeks of restricted exercise should be enforced, even if improvement has been marked. Leash exercise for urination and defecation is allowed, but a harness should be used rather than a collar. If discomfort persists, anti-inflammatory drugs may be needed (Table 15.2) along with diazepam, methocarbamol, ultrasound (15.13B), laser (15.14A), or acupuncture. Severe pain beyond 48 h warrants repeat imaging (see ‘Complications’, page 99). Dogs with preoperative neurological deficits benefit from physiotherapy once the pain has subsided. Largebreed dogs are prone to problems related to recumbency,
and strict attention to bedding and cleanliness is required (see Chapter 15).
PROGNOSIS Resolution of clinical signs may be slow after fenestration. Of 12 dogs with neck pain, six recovered rapidly but five continued to have intermittent pain for 1–4 weeks. Of a further 16 dogs with thoracic limb paresis, 12 (75%) recovered in an average of 3–8 weeks. Of nine with severe deficits (tetraparesis or tetraplegia), five (56%) recovered in 1–6 weeks (Denny, 1978). Fenestration has been compared to ventral slot decompression in 111 dogs that were able to walk prior to surgery. Intraoperative and postoperative complications, other than simple incisional swelling, were more common after ventral decompression (30% vs 12%), resulting in longer hospital stays (7.12). However, neurological recovery was slower following fenestration. Fenestrated dogs were three times more likely to have static signs at discharge and twice as likely to be unchanged at their last hospital visit (Fry et al., 1991). Ventral slot decompression gives much better results overall (Table 7.3). This is probably because even dogs with no neurological deficits usually have substantial amounts of disc material in the vertebral canal. The prognosis for dogs with pain or moderate neurological deficits is usually good after a ventral slot (Table 7.3). The prognosis for dogs that cannot walk is more guarded. Complete recoveries are reported in 55–60% of such dogs, a further 15–20% recover with residual deficits and 20–25% die or are euthanized. Dogs that do not walk within 2 weeks are likely to continue to have residual deficits (Waters, 1989; Smith et al., 1997). Others have reported better recovery rates for severely affected dogs (Seim and Prata, 1982). The outcome after ventral slot decompression in small-breed dogs appears to be better for dogs with cranial cervical lesions (C2/3 or C3/4) than for those
Cervical disc disease
Table 7.3 Results of ventral slot decompression for cervical disc Neurological status
No. of dogs
No. recovered* by 2 days (%)
No. recovered* by 7 days (%)
No. recovered* by 28 days (%)
No. recovered* by 365 days (%)
No recovery
Pain ⫹/⫺ RS1
33
16 (48)
24 (73)
30 (91)
33 (100)
0
Can walk1
14
6 (43)
10 (71)
12 (86)
14 (100)
0
Can not walk1,2
18
5 (28)
10 (56)
14 (78)
15 (83)
3 (17)
1 Seim and Prata (1982); 2 Waters (1989). * Recovered defined as no neck pain; ability to walk for dogs unable to walk before surgery. RS, Root signature.
with caudal cervical lesions (C4/5–C6/7 inclusive) (Waters, 1989; Fitch et al., 2000). Long-term resolution of signs after ventral slot decompression was seen in 31 of 47 dogs with cranial lesions (66%) compared to only 10 of 48 with caudal lesions (21%). The poor outcome for dogs with caudal lesions undergoing a ventral slot contrasts with the long-term resolution seen in 12 of 15 dogs (80%) that were also stabilized and distracted intraoperatively at the ventral slot site (Fitch et al., 2000). One of the main reasons for failure to improve, or for recurrence of signs after ventral slot decompression, is postoperative instability or subluxation. These problems have only been reported in dogs with caudal lesions (Seim and Prata, 1982; Wheeler and Sharp, 1994; Smith et al., 1997; Fitch et al., 2000; Lemarie et al., 2000). One factor that increases the risk considerably is the width of the slot. In dogs that suffer subluxation the slot is usually too wide; reported as 39–80% of the vertebral width and a median of 50%
(Fitch et al., 2000; Lemarie et al., 2000) (7.10, 7.16). Subluxation does appear to be a significant risk when a wide slot is performed for caudal lesions. It is not clear if this is also true for small dogs when the slot width is kept to about one third of vertebral body width (Lemarie et al., 2000). If ventral slot decompression is to be performed in dogs with caudal lesions, the slot proportions should be close to 33% or an inverted cone should be considered (Goring et al., 1991). If the width of a caudal slot is nearing 50% then the site should also be stabilized (Lemarie et al., 2000).
CERVICAL DISC DISEASE IN CATS Disc herniation is not uncommon in cats, particularly in the cervical region, but clinical signs related to these lesions are rare (Heavner, 1971; Littlewood et al., 1984; Kathmann et al., 2000; Knipe et al., 2001; Muñana et al., 2001). The principles of diagnosis and treatment are discussed above.
Key issues for future investigation 1. How to manage dogs with multiple extrusions of mineralized disc material (7.15A)? Options include dorsal laminectomy (Gill et al., 1996), or several ventral slot(s) with bone graft distraction. For the latter procedure it would help to be able to decide which lesion(s) is most significant (4.43). 2. How to deal with multiple, dynamic lesions in small dogs (7.15B)? Options include dorsal laminectomy, fusion using cement plugs (but see 7.13), or possibly several ventral slots, each with bone graft or cement wedges (page 118). 3. What is the maximum recommended width for a slot and what are the risks of instability (7.16)? 4. Is there a need for routine fixation to prevent luxation in dogs with caudal lesions or is this only necessary when the slot width is ⬎50% (Lemarie et al., 2000)?
103
104
Small Animal Spinal Disorders
A
B
7.15 A: Multiple herniations in a Pomeranian with neck pain, mild tetraparesis and a disconnected gait (see page 28). Extradural compression is evident from C2/3 to C7/T1 inclusive; several lesions were of mineralized material. Fenestration of these disc spaces led to recurrence of signs within a year. B: Miniature pinscher with neck pain and thoracic limb root signature. Traction-responsive, ventral extradural compression is present from C2/3 to C6/7 inclusive.
A
B
7.16 A: Ventrodorsal survey radiograph from a 6-year-old Yorkshire terrier that suffered a vertebral subluxation at the site of a previous ventral slot at C4/5 (arrow). The width of the slot is 55% (same dog as in 7.52). B: Seven-year-old Dachshund with neck pain and root signature that was unable to walk after ventral slot at C3/4; slot width is 66%. An external splint was applied; the dog could walk within 2 weeks. 5. Are cranial lesions more resistant to subluxation (Lemarie et al., 2000) (7.16B)? 6. What is the best way to fuse interspaces in small-breed dogs? Options include metal and bone cement (page 118) or bone allografts combined with ventral slot (Fitch et al., 2000). Non-biological materials that permit bone in-growth may become available in the future (Cook et al., 1994; Emery et al., 1996).
REFERENCES Atilola, M.A.O., Cockshutt, J.R., McLaughlin, R., Cochrane, S.M., Pennock, P.W. (1993) Collagenase chemonucleolysis—a long term radiographic study in normal dogs. Veterinary Radiology and Ultrasound 34, 321–324. Bagley, R.S., Forrest, L.J., Cauzinille, L., Hopkins, A.L., Kornegay, J.N. (1993) Cervical vertebral fusion and concurrent intervertebral disc extrusion in four dogs. Veterinary Radiology and Ultrasound 34, 336–339. Bagley, R.S., Tucker, R., Harrington, M.L. (1996) Lateral and foraminal disk extrusion in dogs. Compendium on Continuing Education for the Practicing Veterinarian 18, 795–804. Beal, M.W., Paglia, D.T., Griffin, G.M., Hughes, D., King, L.G. (2001) Ventilatory failure, ventilator management, and outcome in dogs with cervical spinal disorders: 14 cases (1991–1999). Journal of the American Veterinary Medical Association 218, 1598–1602. Boydell, P. (1995) Horner’s syndrome following cervical spinal surgery in the dog. Journal of Small Animal Practice 36, 510–512.
Chambers, J.N., Oliver, J.E., Jr, Kornegay, J.N., Malnati, G.A. (1982) Ventral decompression for caudal cervical disk herniation in large- and giant-breed dogs. Journal of the American Veterinary Medical Association 180, 410–414. Chambers, J.N., Oliver, J.E., Jr, Bjorling, D.E. (1986) Update on ventral decompression for caudal cervical disk herniation in Doberman Pinschers. Journal of the American Animal Hospital Association 22, 775–778. Clark, D.M. (1986) An analysis of intraoperative and early postoperative mortality associated with cervical spinal decompressive surgery in the dog. Journal of the American Animal Hospital Association 22, 739–744. Coates, J.R., Dewey, C.W. (1998) Cervical spinal hyperesthesia as a clinical sign of intracranial disease. Compendium on Continuing Education for the Practicing Veterinarian 20, 1025–1037. Cook, S.D., Dalton, J.E., Tan, E.H., Tejeiro, W.V., Young, M.J., Whitecloud, T.S. 3rd (1994) In vivo evaluation of anterior cervical fusions with hydroxylapatite graft material. Spine 19, 1856–1866. Cybulski, G., D’Angelo, C.M. (1988) Neurological deterioration after laminectomy for spondylotic cervical myeloradiculopathy: the putative
Cervical disc disease
role of spinal cord ischaemia. Journal of Neurology, Neurosurgery and Psychiatry 51, 717–718. Dallman, M.J., Palettas, P., Bojrab, M.J. (1992) Characteristics of dogs admitted for treatment of cervical intervertebral disk disease: 105 cases (1972–1982). Journal of the American Veterinary Medical Association 200, 2009–2011. De Risio, L., Muñana, K.R., Murray, M., Olby, N.J., Sharp, N.J.H., Cuddon, P. (2002) Dorsal laminectomy for caudal cervical spondylomyelopathy: postoperative recovery and long-term follow-up in 20 dogs. Veterinary Surgery 31, 418–427. Denny, H.R. (1978) The surgical treatment of cervical disc protrusions in the dog: a review of 40 cases. Journal of Small Animal Practice 19, 251–257. Dixon, B.C., Tomlinson, J.L., Kraus, K.H. (1996) Modified distractionstabilization technique using an interbody polymethyl methacrylate plug in dogs with caudal cervical spondylomyelopathy. Journal of the American Veterinary Medical Association 208, 61–68. Emery, S.E., Fuller, D.A., Stevenson, S. (1996) Ceramic anterior spinal fusion. Biologic and biomechanical comparison in a canine model. Spine 21, 2713–2719. Felts, J.F., Prata, R.G. (1983) Cervical disk disease in the dog: intraforaminal and lateral extrusions. Journal of the American Animal Hospital Association 19, 755–760. Fingeroth, J.M. (1989) Fenestration. Pros and cons. Problems in Veterinary Medicine 1, 445–466. Fitch, R.B., Kerwin, S.C., Hosgood, G. (2000) Caudal cervical intervertebral disk disease in the small dog: role of distraction and stabilization in ventral slot decompression. Journal of the American Animal Hospital Association 36, 68–74. Floyd, T., Ohnmeiss, D. (2000) A meta-analysis of autograft versus allograft in anterior cervical fusion. European Spine Journal 9, 398–403. Fry, T.R., Johnson, A.L., Hungerford, L., Toombs, J. (1991) Surgical treatment of cervical disc herniations in ambulatory dogs. Ventral decompression vs. fenestration, 111 cases (1980–1988). Progress in Veterinary Neurology 2, 165–173. Funkquist, B., Svalastoga, E. (1979) A simplified surgical approach to the last two cervical discs of the dog. Journal of Small Animal Practice 20, 593–601. Gage, E.D. (1975) Incidence of clinical disc disease in the dog. Journal of the American Animal Hospital Association 11, 167–174. Gill, P.J., Lippincott, C.L., Anderson, S.M. (1996) Dorsal laminectomy in the treatment of cervical intervertebral disk disease in small dogs: a retrospective study of 30 cases. Journal of the American Animal Hospital Association 32, 77–80. Gilpin, G.N. (1976) Evaluation of three techniques of ventral decompression of the cervical spinal cord in the dog. Journal of the American Veterinary Medical Association 168, 325–328. Goggin, J.E., Li, A., Franti, C.E. (1970) Canine intervertebral disk disease: Characterization by age, breed, sex and anatomic site of involvment. American Journal of Veterinary Research 31, 1687. Goring, R.L., Beale, B.S., Faulkner, R.F. (1991) The inverted cone decompression technique: a surgical treatment for cervical vertebral instability ‘Wobbler syndrome’ in Doberman Pinschers. Part 1. Journal of the American Animal Hospital Association 27, 403–409. Griffiths, I.R. (1973) Spinal cord blood flow in dogs: the effect of blood pressure. Journal of Neurology, Neurosurgery and Psychiatry 36, 914. Hara, Y., Tagawa, M., Ejima, H., Orima, H., Fujita, M. (1994) Usefulness of computed tomography after myelography for surgery on dogs with cervical intervertebral disc protrusion. Journal of Veterinary Medical Science 56, 791–794. Heavner, J.E. (1971) Intervertebral disc syndrome in the cat. Journal of the American Veterinary Medical Association 159, 425–427. Hurov, L. (1979) Dorsal decompressive cervical laminectomy in the dog: surgical considerations and clinical cases. Journal of the American Animal Hospital Association 15, 301–309. Janssens, L.A.A. (1985) The treatment of canine cervical disc disease by acupuncture: a review of 32 cases. Journal of Small Animal Practice 26, 203–212. Jerram, R.M., Hart, R.C., Schulz, K.S. (1997) Postoperative management of the canine spinal surgery patient—part I. Compendium on Continuing Education for the Practicing Veterinarian 19, 147–161.
Kathmann, I., Cizinauskas, S., Rytz, U., Lang, J., Jaggy, A. (2000) Case report spontaneous lumbar intervertebral disc protrusion in cats: literature review and case presentations. Journal of Feline Medicine and Surgery 2, 207–212. Knipe, M.F., Vernau, K.M., Hornof, W.J., LeCouteur, R.A. (2001) Intervertebral disc extrusion in six cats. Journal of Feline Medicine and Surgery 3, 161–168. Lemarie, R.J., Kerwin, S.C., Partington, B.P., Hosgood, G. (2000) Vertebral subluxation following ventral cervical decompression in the dog. Journal of the American Animal Hospital Association 36, 348–358. Levitski, R.E., Lipsitz, D., Chauvet, A.E. (1999) Magnetic resonance imaging of the cervical spine in 27 dogs. Veterinary Radiology and Ultrasound 40, 332–341. Lipsitz, D., Bailey, C.S. (1995) Clinical use of the lateral cervical approach for cervical spinal cord and nerve root disease: eight cases. Progress in Veterinary Neurology 6, 60–65. Littlewood, J.D., Herrtage, M.E., Palmer, A.C. (1984) Intervertebral disc protrusion in a cat. Journal of Small Animal Practice 25, 119–127. Macy, N.B., Les, C.M., Stover, S.M., Kass, P.H. (1999) Effect of disk fenestration on sagittal kinematics of the canine C5–C6 intervertebral space. Veterinary Surgery 28, 171–179. Morgan, P.W., Parent, J., Holmberg, D.L. (1993) Cervical pain secondary to intervertebral disc disease in dogs; radiographic findings and surgical implications. Progress in Veterinary Neurology 4, 76–80. Muñana, K.R., Olby, N.J., Sharp, N.J.H., Skeen, T.M. (2001) Intervertebral disk disease in 10 cats. Journal of the American Animal Hospital Association 37, 384–389. Olby, N.J., Munana, K.R., Sharp, N.J.H., Flegel, T., Van Camp, S., Berry, C.R., Thrall, D.G. (1999) A comparison of computed tomography and myelography in the diagnosis of acute intervertebral disc disease in dogs. Journal of Veterinary Internal Medicine 13, 239. Olby, N.J., Munana, K.R., Sharp, N.J.H., Thrall, D.E. (2000) The computed tomographic appearance of acute thoracolumbar intervertebral disc herniations in dogs. Veterinary Radiology and Ultrasound 41, 396–402. Prata, R.G., Stoll, S.G. (1973) Ventral decompression and fusion for the treatment of cervical disc disease in the dog. Journal of the American Animal Hospital Association 9, 462–472. Russell, S.W., Griffiths, R.C. (1968) Recurrence of cervical disc syndrome in surgically and conservatively treated dogs. Journal of the American Veterinary Medical Association 153, 1412–1416. Seim, H.B., Prata, R.G. (1982) Ventral decompression for the treatment of cervical disk disease in the dog: a review of 54 cases. Journal of the American Animal Hospital Association 18, 233–240. Sether, L.A., Yu, S., Haughton, V.M., Fischer, M.E. (1990) Intervertebral disk: normal age-related changes in MR signal intensity. Radiology 177, 385–388. Smith, B.A., Hosgood, G., Kerwin, S.C. (1997) Ventral slot decompression for cervical intervertebral disc disease in 112 dogs. Australian Veterinary Practitioner 27, 58–64. Somerville, M.E., Anderson, S.M., Gill, P.J. (2001) Accuracy of localization of cervical intervertebral disk extrusion or protrusion using survey radiography in dogs. Journal of the American Animal Hospital Association 37, 563–572. Stauffer, J.L., Gleed, R.D., Short, C.E., Erb, H.N., Schukken, Y.H. (1988) Cardiac dysrhythmias during anesthesia for cervical decompression in the dog. American Journal of Veterinary Research 49, 1143–1146. Thomson, C.E., Kornegay, J.N., Stevens, J.B. (1989) Canine intervertebral disc disease: changes in the cerebrospinal fluid. Journal of Small Animal Practice 30, 685–688. Tomlinson, J. (1985) Tetraparesis following cervical disk fenestration in two dogs. Journal of the American Veterinary Medical Association 187, 76–77. Waters, D.J. (1989) Nonambulatory tetraparesis secondary to cervical disk disease in the dog. Journal of the American Animal Hospital Association 25, 647–653. Wheeler, S.J., Sharp, N.J.H. (1994) Small Animal Spinal Disorders: Diagnosis and Surgery, 1st edn. St Louis: Mosby, page 82. Wrigley, R.H., Reuter, R.E. (1984) Canine cervical discography. Veterinary Radiology 25, 274–279.
105
106
Small Animal Spinal Disorders
PROCEDURES Approach to the ventral neck (7.17–7.29) The cervical spine is extended over a sandbag; the thoracic limbs are pulled in a caudal direction, and tape is used to immobilize the head and thorax (7.17). It is important to have the dog straight to insure that the neck is aligned correctly. Note that extension of the caudal portion of the neck as shown here will tend to close the dorsal intervertebral space of C5/6 and C6/7 spaces. Ventral slot decompression at these sites is easier to perform if the neck is put in a more neutral position. If access needs to be improved to C6/7, it is preferable to put the neck in gentle traction using a weight on the maxilla (11.24), or consider a dorsal (11.44–11.55, 7.59) or dorsolateral laminectomy (7.56–7.58, 12.15–12.29). Deep anatomy includes the carotid sheaths, recurrent laryngeal nerves and esophagus (7.25–7.27). Careful dissection of the loose fascial layers in a longitudinal direction will expose these structures. Finger dissection here is safe and effective. Further dissection of the deep fascia reveals the longus colli muscles, which lie ventral to the cervical vertebrae (7.28). If preferred, the paramedian approach to the ventral neck can be used instead of the approach described here in order to provide greater protection for blood vessels, nerves, trachea and esophagus (11.25–11.27). Blunt-jawed self-retaining retractors (5.8) are used initially to maintain exposure (7.29). The retractors must not interfere with gas flow in the airway. Palpate the end of the endotracheal tube and insure that it is positioned distal to the retractors. Moist towels or sponges are used to protect the tissues. The longus colli muscles join in the midline and insert on the ventral process of the vertebra, thus overlying the intervertebral discs (7.29). Accurate identification of the intervertebral discs at this stage is crucial. The ventral process of C5 vertebral body lies in the midline between the cranial borders of the large C6 transverse processes (4.6, 7.37B); the C5/6 intervertebral disc lies immediately behind this ventral process. The ventral process of C1 is particularly prominent and sharp (7.19); this also can or may be palpated. There is no intervertebral disc at C1/2.
7.17
7.17 Positioning of dog for ventral approach to the neck. Excessive extension will close the dorsal disc space; mild traction may be preferable (11.24A).
7.18 Landmarks are (a) larynx; (b) wing of atlas; and (c) manubrium of sternum. Incision site is depicted on the midline.
c a
b
7.18
7.19 Landmarks can be palpated; in this illustration the surgeon is palpating the wings of the atlas (left hand) and the prominent transverse process of C6 (4.6, 7.37B).
7.19
Cervical disc disease
7.20 Superficial anatomy—see also 7.21. The skin and superficial fascia have been divided to reveal the sternocephalicus muscles (a), and sternohyoid muscles (b). The cervical vertebrae C1 through C7 are seen in the background.
b a
7.20
7.21 The skin and superficial fascia have been divided. This reveals the sternocephalicus muscles (a), and the sternohyoid muscles (b). The sternocephalicus muscles should be divided to the manubrium, particularly if access to the caudal cervical vertebrae is required.
b
a
7.21
7.22 The sternohyoid muscles are divided in the midline. If the median raphe is not apparent, apply finger pressure to the muscles over the trachea and the raphe will become visible.
7.22
7.23 Making a small incision in the fascia and then dissecting the fascia bluntly usually prevents bleeding. The caudal thyroid vein lies in the connective tissue with small branches on each side. This vein should be preserved if possible; the branches may be cauterized. The trachea is immediately under the fascia. 7.23
107
108
Small Animal Spinal Disorders
7.24 Once the sternohyoid muscles have been divided the trachea is visible. Here the separation is made with fingers.
7.24
b
7.25 Close up view of the trachea (a); recurrent laryngeal nerve (b); and carotid sheath (c). The approach is between the recurrent laryngeal nerve (which remains associated with the trachea) and the carotid sheath at the level indicated by the arrow. At this stage the esophagus must be identified or the surgeon could damage it inadvertently (7.26).
a c
7.25
7.26 There are several layers of loose fascia that must be penetrated before reaching the longus colli muscles. The esophagus is identified by its pink color compared to the darker red of the longus colli muscles; differentiation is aided by having an esophageal stethoscope in place (see 7.27 for labels). The recurrent laryngeal nerve has been separated from the trachea. This is for illustration only and is not done normally.
7.27 Note the recurrent laryngeal nerve (a), and the carotid sheath containing the carotid artery along with the more prominent, white vagosympathetic trunk (b). The trachea and esophagus have been retracted away from the surgeon; they should be kept together on the same side of the incision.
a
b 7.26
a
b 7.27
Cervical disc disease
7.28 Retraction of the vital structures exposes the longus colli muscles. The transverse processes of C6 are large and are directed ventrally; they are palpated readily as shown here (4.6, 7.37B).
7.28
7.29 Deep anatomy. Note the pattern of longus colli muscle bellies running cranially. Once landmarks have been identified and any fenestrations completed, pressure on the jugular veins should be relieved by substituting smaller Gelpi retractors over the space of interest if a ventral slot is to be performed.
7.29
Fenestration (7.30–7.36) During fenestration it is important not to confuse the ventral process of C2–C5 vertebrae with the transverse processes (see under ‘Approach’, page 106). The longus colli muscles run cranially to insert on the ventral process in the midline. The transverse processes can be palpated on each side of the ventral process. The three rows of structures (ventral processes in the midline, transverse processes laterally on each side) should be identified before proceeding.
7.30 To gain access to the disc, the longus colli muscles are divided in the midline, immediately caudal to the ventral process. They may be cut or division may be achieved with a pair of small, curved hemostats.
7.30
109
110
Small Animal Spinal Disorders
7.31 Method of fenestration. The anulus fibrosus is cut in the form of a window to allow access to the nucleus pulposus (7.34).
7.31
7.32
7.32 The nuclear material is removed (7.35). It is important not to push the instrument deeply through the dorsal anulus fibrosus into the vertebral canal. Radiographs provide some guide to the depth of the intervertebral space as long as magnification is taken into account. The window in the anulus must be larger than the instrument to avoid creating a piston effect, which could force disc material dorsally into the vertebral canal.
b
7.33 If greater exposure of the disc is required, the muscle separation is maintained by self-retaining retractors. The longus colli tendons may also be cut. The ventral surface of the vertebra (a), and the ventral anulus fibrosus (b), can be seen. The surgeon’s finger is on the ventral process of the vertebra cranial to the disc.
7.33
7.34 Fenestration is done by cutting a hole in the anulus fibrosus; this should be made as large as possible to allow complete evacuation of the nucleus pulposus. A fresh #11 scalpel blade is used to make the opening, either by making four cuts in a rectangular pattern in the anulus fibrosus, or by using a sawing motion in an oval pattern until the piece of anulus is free. The surgeon’s finger is on the ventral process in this picture.
7.34
a
Cervical disc disease
7.35 Disc fenestrated. Nucleus pulposus is oozing from the intervertebral space. The removed piece of anulus fibrosus lies cranial to the disc space (arrow). A small curette or blunt instrument is used to remove the nucleus pulposus by entering the space, dragging towards the surgeon and 7.35 then lifting the material out. This maneuver should be repeated until there is a reasonable certainty that all the nuclear material has been removed. Care must be taken to avoid penetrating the vertebral canal dorsally (7.32).
7.36
7.36 Sagittal diagram shows that fenestration only allows access to disc material in the intervertebral space. Disc material lying in the vertebral canal cannot be removed by fenestration.
Ventral decompression (7.37–7.52) Sometimes the dorsal longitudinal ligament remains after removal of the anulus fibrosus. The ventral surface (the side nearest the surgeon) of the ligament is slightly roughened and has longitudinal fibers; the dorsal surface is smooth and shiny. The identity of any tissue exposed in the vertebral canal must be established before further manipulations or incisions are made. It is often necessary to complete bone removal before tissue can be identified clearly. If hemorrhage from the venous plexus does occur, it may be possible to control it with Gelfoam, direct pressure (11.29), or a small piece of muscle (Table 7.2). Pressure on the jugular veins should be relieved. Work should then continue at the other end of the slot. Suction may be maintained while disc material is being removed, but careful note must be taken of the amount of blood aspirated. If hemorrhage is severe, the slot is packed and time allowed for it to stop before re-exploration.
111
112
Small Animal Spinal Disorders
7.37 A: View of cervical vertebrae showing ventral slot decompression. The procedure allows access to the vertebral canal through the vertebral body. Correct identification of the intended space is crucial (see ‘Approach’, page 106). Where concomitant fenestration is to be performed this should be done first in order to verify landmarks. B: 3D reconstruction of a ventral slot (same dog as shown in 7.9) to illustrate the appearance from a ventral aspect at C5/6. The overall width of this slot is 38%. Note the prominent wings of C6 (arrowheads).
7.37 A
B
7.38 The site for the ventral slot is prepared by removing the musculature from the vertebrae on both sides of the midline. The longus colli muscles are separated from the ventral process. Hemostasis at this stage is important, as excessive bleeding will obscure the site of the slot. Muscle separation is maintained with selfretaining retractors and tissues are protected with moist sponges.
7.38
Cervical disc disease
7.39
7.39 Sagittal section shows that the slot should be started more in the vertebra cranial to the disc than the one caudal, because of the angulation of the disc space relative to the vertebral canal. This will also allow room for screws or pins to be placed should the slot collapse or subluxate (7.10, 7.16, 7.52, 11.21).
7.40
7.40 The slot is commenced in the position described in 7.39. Initially drilling is only into the cranial vertebra. The cortical bone has been penetrated to reveal purple cancellous bone; the intervertebral disc is caudal to this. The aim is to create a slot approximately one third of the width of the vertebra and one third of its length, with the disc space at its center once the vertebral canal is reached (7.39, 7.41). This size of slot should not result in instability and if kept on midline should avoid the internal vertebral venous plexus (7.8, 7.43, 7.44).
7.41
7.41 Partially completed slot. Somewhat more advanced than in the previous illustration. Hemorrhage from the bone is controlled with bone wax. It is important to keep the bur irrigated and the site free of debris. It is useful to stop drilling periodically to clean the site and to assess progress.
7.42 The floor of the slot is shown. It is important to judge the depth of the slot accurately. Cortical bone is white and hard (a), in contrast to the dark cancellous bone (b). The cancellous bone is removed over the whole slot area, using a small bur. The remaining cortical bone is then thinned to allow easy removal; it is best to thin the whole area before entering the vertebral canal (8.36, 10.34). The dorsal anulus fibrosus is visible (c). The shape of the opening into the vertebral canal must take into account the location of the venous plexus (7.43, 7.44).
b a
7.42
c
113
114
Small Animal Spinal Disorders
7.43 Latex-infused preparation to demonstrate the location of vertebral venous plexus on the floor of the vertebral canal. The dorsal lamina has been removed for this view from above the spine. The discs are represented in orange. Note how the venous plexus converges in the mid-portion of the vertebra and diverges over the disc (1.36, 7.44). 7.43
7.44 This series of CT scans was made after intravenous contrast to highlight the vertebral venous plexus (arrows). A: Middle of C6 vertebra. B: At the C6/7 disc space. C: Middle of C7. Note how the vessels diverge at the level of the disc space but converge again over the vertebral bodies. 7.44 A
B
C
Cervical disc disease
7.45 The dorsal anulus fibrosus is incised on each side of the slot using a fresh #11 blade. One incision can be seen uppermost in the picture (arrow). The near-side incision is being made. The incision is then extended to where the cortical bone has been removed.
7.45
7.46 The anulus fibrosus is lifted with a suitable instrument (in this case, a pointed tartar scraper) and dissected free. Removal can be facilitated by preserving a small knub of anulus to grasp with fine rongeurs (11.29A).
7.46
7.47 The dorsal anulus has been removed, allowing a view into the vertebral canal. Here the glossy dura mater is visible. When extruded disc material is present, it is wise to remove it from the midline first to avoid damaging the venous plexus early in the procedure.
7.47
7.48 The slot is almost complete. The floor of the slot is excised with fine rongeurs or a small curette after bone has been removed and ligamentous tissue cut with a scalpel. Great patience is required when removing herniated disc material; magnification and additional lighting are preferred (5.1–5.3). Once all material has been removed, the spinal cord will lie adjacent to the edges of the slot as shown. If the disc is asymmetrically positioned, it is possible to shape the base of the slot to allow more exposure on one side although this may increase the risk of bleeding (7.50B).
7.48
115
116
Small Animal Spinal Disorders
7.49 Removal of mineralized disc material is generally straightforward although patience is required. As in the thoracolumbar region, additional disc material usually remains in the vertebral canal unless the dura returns to lie level with the bone.
7.49
7.50
7.50 In small dogs or dogs with very large extrusions the inverted cone technique can improve access to the vertebral canal A: This 5-year-old Dachshund A B had severe neck pain and mild tetraparesis. The 3D reconstruction of the preoperative CT shows a huge extrusion at C2/3, viewed looking caudally. B: 3D reconstruction made from the immediate postoperative CT scan looking cranially to C1. The slot has been extended to one side (arrowhead) as an inverted cone in order to give greater access without causing instability (Goring et al., 1991). Nearly all of the material has been removed. The width is approximately 33%. 7.51
7.51 Access to disc material in the vertebral canal via a ventral slot. A: Sagittal diagram. Note the starting position of the slot relative to the disc. B: Inverted preoperative (left) and postoperative (right) 3D reconstruction of the dog shown in 7.50. The extrusion at C2/3 is indicated in the left-hand panel (*). The margins of the slot are outlined in the right-hand panel by arrowheads.
A
*
B
Cervical disc disease
7.52 A: Postoperative radiograph of a dog that had a subluxation after ventral slot at C4/5. The width of the slot was 55% of the vertebral body (7.16). B: Stabilization was performed using pins and cement. The dog suffered respiratory arrest and did not recover from surgery.
C6
C6 7.52 A
B
Ventral decompression with distraction-stabilization (7.53–7.55) (see also Chapter 11) Preferred methods of distraction-stabilization use a corticocancellous autograft, allograft (Veterinary Transplant Services, Seattle, WA), or metal and bone cement (Fitch et al., 2000); cement plugs are difficult to use in small dogs (7.13, 7.54). Initial distraction can be accomplished as shown in 11.31–11.34, by K-wires (page 235), or by a rope around the dog’s upper canine teeth and tied to the surgical table (7.55, 11.24). After distraction a bone allograft (Veterinary Transplant Services, Seattle, WA), cement (page 118), or an autograft from the wing of C6, sternum or ilium, can be wedged into the slot to maintain distraction (Fitch et al., 2000). Metal and cement implants may also be used (7.55).
7.53 A: This 8-kg Silky terrier had undergone a ventral slot at C5/6 and then suffered domino lesions at C4/5 and C6/7 some18 months later (same dog as 7.14). B: Both lesions were traction-responsive.
T1 T1
7.53 A
B
7.54 A: Cement plugs at C4/5 and C6/7 using anchor holes in the end plates (11.38) with cancellous grafting from C4 to T1 T1 C7 (Dixon et al., 1996). An external splint was not used. B: Six-week follow-up shows end7.54 plate fracture at C4/5 and A B ventral displacement of the implant at C6/7 (7.13). The dog was doing well clinically but had developed iatrogenic laryngeal paralysis.
117
118
Small Animal Spinal Disorders
7.55
7.55 Miniature pinscher with severe neck pain and left thoracic limb root signature. A: Preoperative myelogram shows collapse and mineralized material at C6/7. B: CT myelogram shows mineralized extrusion or osteophytes within the left foramen also (arrowhead) (Prata and Stoll, 1973). C: Postoperative radiograph of stabilization using 2.7-mm screws and bone cement following traction on the maxilla as the cement hardened (compare disc space to A). The caudal screw is positioned poorly but the dog has shown no clinical signs for 6 years.
A L
B
C
Distraction can be maintained with a small amount of bone cement wedged into a partial ventral slot. The edges of the slot are undercut to prevent slippage of the cement. This method can also be used to rescue a wide or subluxated ventral slot, as the main goal here is to prevent excessive collapse or subluxation as seen in 7.10. Gelfoam (Pharmacia, Kalamazoo, MI) should be used to insulate the spinal cord (page 235 and 239). The cement wedge should be kept to the ventral half of the interspace.
Cervical disc disease
Dorsal hemilaminectomy (7.56–7.58) (see also page 266)
7.56
L
L
7.56 A: Transverse and B: 3D reconstruction of a CT scan from a 12-year-old Labrador retriever with paraparesis, worse on the right side. Neurological examination revealed a disconnected thoracic limb gait (see page 28) and guarding of the neck. A B These findings suggested a caudal cervical lesion; a C7/T1 ventral, extradural compression was confirmed by myelography. CT scan revealed the mass to be mainly right-sided. Differential diagnoses were a disc, nerve root tumor or meningioma.
7.57 A: Preoperative myelogram and B: 3D reconstruction of the postoperative CT scan to show the hemilaminectomy (arrowheads) used to access the mass. The gap between C6 and C7 (arrow) is an artefact of reconstruction. Head is to the right in these images.
7.57
A
B
7.58
7.58 Postoperative A: Transverse and B: 3D reconstruction of a CT scan from the same dog as shown in 7.56–7.57. The mass was confirmed to be a large, fibrous disc herniation; about 75% of it was removed using a B A combination of scalpel and rongeurs. A pocket of gas is visible at the surgical site in A. The extent of the hemilaminectomy is shown clearly in B. The dog made a rapid recovery.
119
120
Small Animal Spinal Disorders
Dorsal laminectomy (7.59) The surgical approach is described in Chapter 11 (11.44–11.55). Large veins may be encountered in the dorsal epidural space, especially close to the intervertebral foramen (Hurov, 1979) (11.51, 11.52).
R 7.59
7.59 A: Ventrodorsal myelogram from a dog with hemiparesis and neck pain shows left-sided, extradural compression over C7 vertebra. B: CT myelogram shows mineralized material to the left of the dural tube over mid-C7 vertebral body. C: Intraoperative photograph shows mineralized disc material to one side of the dural tube (arrowheads). The dura was retracted using stay sutures to improve access. Fenestration was performed using a ventral approach after the laminectomy. The dog’s neurological status was improved the next day and it was almost normal 1 month later.
A
L
B
C
Thoracolumbar disc disease
Clinical signs
122
CLINICAL SIGNS
Treatment 123 Non-surgical 123 Decompression (hemilaminectomy, mini-hemilaminectomy or pediculectomy) Fenestration without decompression 127
126
Complications 127 Intraoperative complications 127 Early postoperative complications 127 Late postoperative complications 130 Postoperative care Prognosis
132
132
Thoracolumbar disc disease in cats Key issues for future investigation References
8
frequently, usually after middle age. However, disc disease should not be overlooked in large dogs as they can have acute, Hansen Type I extrusions as well as Type II lesions (see pages 12 and 56–58). L1/2 is affected most often in large dogs (Cudia and Duval, 1997).
121
Diagnosis 122 Radiography 122 Cerebrospinal fluid analysis Myelography 122 CT and MR imaging 123
Chapter
133 133
134
Procedures 136 Dorsolateral hemilaminectomy 136 Pediculectomy and mini-hemilaminectomy Lateral fenestration 154
151
Thoracolumbar disc disease is a common disorder in dogs that affects mainly chondrodystrophoid breeds. Peak incidence in these breeds is between 3 and 6 years of age. Over 50% of lesions occur at the T12/13 and T13/L1 discs and more than 85% occur between T11/12 and L2/3 inclusive; disc extrusion occurs occasionally as far cranial as T9/10 (Wilkens et al., 1996). Non-chondrodystrophoid breeds are affected less
Back pain and neurological deficits in the pelvic limbs are features of thoracolumbar disc disease; urinary dysfunction may occur with more severe lesions (see page 31). The back pain is usually less dramatic than that associated with cervical disc disease. The dog may show kyphosis and a reluctance to run or jump; discomfort can usually be elicited by palpation over the thoracolumbar region. Pain without neurological deficits can be misinterpreted as being cervical, orthopedic or abdominal in origin (Table 2.1, Table 3.6, Box 7.2). The pain in thoracolumbar disc disease arises because of combinations of anulus fibrosus and dorsal longitudinal ligament damage with associated meningeal and nerve root irritation. Neurological deficits range from mild ataxia and paraparesis to paraplegia, which may be accompanied by depressed or absent nociception caudal to the lesion (Olby et al., 2001) (see ‘Assessing the severity of the lesion’, page 31). Neurological deficits usually become more severe with increasing spinal cord compression (see ‘Spinal cord nerve fibers and the effect of compression’, page 6). In addition to the mass effect of the disc material, the rate at which spinal cord compression occurs is also very important. In extreme cases there may be a combination of mass effect and impact injury to the spinal cord, resulting from explosive disc rupture. In the majority of dogs the deficits are upper motor neuron (UMN) in nature and there may be an associated cutaneous trunci reflex cut-off (see page 24). Approximately 10–15% of dogs show lower motor neuron (LMN) deficits because of lesions between L3/4 and L6/7 discs inclusive (see 1.5). The most common LMN sign is loss of the patellar reflex indicating L5 (range L4 to L6) spinal cord segment involvement, usually due to L3/4 disc herniation. Differential diagnoses are listed in Box 8.1.
122
Small Animal Spinal Disorders
Box 8.1 Differential diagnoses for thoracolumbar disc disease and for back pain ■
Lumbar synovial cyst
■
Lumbar facet joint pain
■
Congenital, e.g. arachnoid cyst
■
Osteoporotic pathological fracture
■
Neoplasia
■
Tumoral calcinosis
■
Discospondylitis
■
Inflammatory CNS disease
■
Epidural abscess
■
Polyarthritis
■
Polymyositis
■
Trauma
■
Dural tear
■
Ischemic myelopathy
■
Ischemic neuromyopathy
■
Psoas muscle injury
A
Abdominal and pelvic pain due to: ■
Pancreatitis
■
Renal pain
■
Ureteral calculi
■
Gallstones
■
Gastrointestinal parasites
■
Kidney worm
■
Prostatic disease
■
Urethral tumor
DIAGNOSIS Radiography Survey radiographs may indicate if disc disease is present but are only 60–70% accurate in identifying the exact location (Kirberger et al., 1992; Olby et al., 1994). Survey radiographs must not be used as the sole means of confirming the diagnosis if decompressive surgery is planned. The main roles of radiography are to help rule out differential diagnoses (Box 8.1) and to confirm anatomical landmarks (8.19–8.21).
B 8.1 A: Lateral myelogram from a dog with a disc extrusion at T12/13. There is poor filling of the subarachnoid space over T12 vertebral body because of a ventral, extradural mass. Filling can often be improved by injecting contrast as the radiograph is being taken. B: The ventrodorsal view of the same dog reveals poor contrast filling of the subarachnoid space over T12 and T13 vertebral bodies. The mass is mainly left-sided over the T12/13 space (arrow) (from Olby et al., 2000).
do show mild abnormalities (Thomson et al., 1989). Routine collection and analysis from the CMC insures that the clinician is able to detect meningitis. Analysis of CSF may show that imaging is unnecessary or that myelography is contraindicated and it also provides a sample for further diagnostic testing should the myelogram prove to be non-diagnostic.
Cerebrospinal fluid analysis Ideally, CSF should be collected routinely from the cerebello-medullary cistern (CMC) unless fluid can be obtained from the lumbar region prior to contrast injection. CSF taken after myelography is almost impossible to interpret as most contrast agents induce a short-term, sterile meningitis (Widmer et al., 1992) (see page 43). Analysis of CSF can assist in refining the differential diagnosis (Box 8.1) although some dogs with disc disease
Myelography Either myelography or advanced imaging should be performed for definitive diagnosis. A lumbar injection is preferred for myelography because there is often considerable spinal cord swelling, which tends to cause cervical myelograms to stop cranial to the lesion (see Chapter 4, page 71). Lateral and ventrodorsal images should be taken (8.1A, B). If it is not clear on which
Thoracolumbar disc disease
8.2 A: CT scan from the dog shown in 8.1. This image, performed prior to myelography, shows a centrally located, mineralized mass occupying much of the vertebral canal (arrows) (from Olby et al., 2000). B: Sagittal T2-weighted MRI through the lumbar region of an 11-year-old paraparetic Labrador with disc herniations at T12/13–L3/4; L2/3 and L3/4 have decreased signal intensity (see 4.43). Note the nutrient arteries entering L4 (arrow) (Parker, 1973).
R
A
B 8.3 A: Transverse CT image and B: 3D reconstruction from different dogs, each demonstrating disc material occupying both sides of the vertebral canal (arrows). These images show why it can sometimes be hard to judge on which side to perform surgery (Schulz et al., 1998; Grevel and Schwartau, 1997).
A
B
side the disc material is located, oblique views or a CT myelogram should be used.
some disc herniations that are only incidental findings and that are not responsible for causing clinical signs (Milette et al., 1999; Olby et al., 2000) (see Chapter 4).
CT and MR imaging CT is more accurate and usually much faster than myelography, especially in chondrodystrophoid breeds (Olby et al., 1999). CT is non-invasive as mineralized disc material shows clearly without the need for contrast, even when it is not visible on survey radiographs. It is usually much easier to decide what side(s) the disc material is on from a CT than from a myelogram (see Chapter 4, page 55; 8.2A, 8.3). A heterogenous, hyperattenuating, extradural mass with loss of epidural fat are characteristic features of mineralized disc extrusions (Olby et al., 2000) (4.37B, 4.40A, 8.3A). Contrast medium may be necessary if the extruded material is not mineralized (4.40A, 11.52). CT has become increasingly popular with neurologists for diagnosing disc disease in chondrodystrophoid breeds and is the modality of choice for many. It does not show the extent of spinal cord swelling as well as myelography but, conversely, swelling does not degrade the CT image. If a CT is not diagnostic it can always be followed by a myelogram. Scanners can also be used to show the exact location and extent of disc material on a scout image (4.39, 4.44A). MRI also provides transverse imaging and is superior to CT when the disc material is not mineralized (4.42, 4.43, 8.2B). The increased sensitivity of CT and MRI reveal
TREATMENT Many dogs will recover from moderate neurological deficits following either non-surgical or surgical treatment. Certain generalizations can be made regarding the advantages of each type of therapy. An algorithm for surgical decision-making is shown in Algorithm 8.1. Patients with marked deficits (see ‘Assessing the severity of spinal cord injury’, page 31) seen within 8 h of spinal cord injury may benefit from concomitant methylprednisolone sodium succinate (MPSS) therapy (see page 83).
Non-surgical Strict cage rest is the overriding principle here although judicious use of analgesics or anti-inflammatory drugs may also be needed. These drugs should be withheld when feasible in order to encourage the animal to rest. The animal must rest quietly in a confined space (traveling cage size) for at least 4 weeks, during which time it should only be removed to urinate and defecate. A satisfactory response should be followed by a further 2 weeks’ rest and a gradual increase in activity between
123
124
Small Animal Spinal Disorders
Neurological examination
Grades 1 or 2
First episode
Grades 3 or 4
Grade 5 or rapid progression
Persistent or recurrent
Neuroimaging Rest and monitor
Elective neuroimaging URGENT ELECTIVE
Fenestration alone
Decompression
Large dog
Acute
Pediculectomy mini-hemilaminectomy, or hemilaminectomy and stabilize DO NOT FENESTRATE AFFECTED DISC
Small dog
Chronic
Pediculectomy mini-hemilaminectomy, or hemilaminectomy and stabilize DO NOT FENESTRATE AFFECTED DISC?
Acute
Decompress in the least invasive way FENESTRATE
Algorithm 8.1 Surgical decision-making in thoracolumbar disc disease.
Table 8.1 Results of treatment for thoracolumbar disc disease (See also Table 8.1a) Treatment as % success (no. of dogs) Neurological grade
Conservative4
Decompression1,2,5,6,7,8
Fenestration3
1—no deficits
100 (8/8)
97 (29/30)
92 (1/12)
2—paresis – walking
84 (32/38)
95 (36/38)
93 (40/43)
3—paresis – not walking
84 (32/38)
93 (43/46)
85 (22/26)
4—paraplegia
81 (13/16)
95 (37/39)
88 (1/8)
64 (86/135)*
33 (2/6)
5—no deep pain 1
7 (1/14)
Anderson et al., 1991; 2 Black, 1988; 3 Butterworth and Denny, 1991; 4 Davies and Sharp, 1983; 5 Olby et al., 2003; Scott and McKee, 1999; 7 Sukhiani et al., 1996; 8 Yovich et al.,1994. *See Table 8.1a.
6
Thoracolumbar disc disease
the 6th and 8th weeks. This is the minimum time needed for an avascular structure like the anulus fibrosus to repair. Activities like jumping should be avoided for 4–6 months. Animals that will not rest or are confined inadequately may fail to respond or get worse. The patient must be evaluated regularly for any deterioration, which indicates treatment failure, as does a lack of improvement within 2 weeks. Advantages of non-surgical treatment are minimal expense and equipment-needs. Treatment can be continued at home, ideally after an initial few days of direct observation. Overall recovery rates are good for dogs with grade 1 to 3 deficits (Table 8.1). About 50% of paraplegic animals that are also incontinent will recover,
Table 8.1a Results of hemilaminectomy for dogs with no deep pain (grade 5)—percentage (nos) Deep pain checked in all dogs at follow-up*
57 (38/67)2,3
Deep pain not checked in all dogs at follow-up
71 (48/68)1,4,5
Total grade 5 dogs recovering
64 (86/135)1,2,3,4,5
1
Anderson et al., 1991; 2 Olby et al., 2003; 3 Scott, 1997; 4 Scott and McKee, 1999; 5 Yovich et al., 1994. *An additional 8/67 dogs (12%) regained voluntary motor function without regaining deep pain (Olby et al., 2003).
but non-surgical treatment is ineffective for dogs with grade 5 lesions (Davies and Sharp, 1983). Although a useful initial option for some dogs with grade 1 or 2 lesions, non-surgical therapy is rarely the treatment of choice for dogs with grade 3 lesions or worse, especially if there are no financial constraints. The major long-term problem is that over one third of dogs will suffer recurrence (Table 8.2). Another disadvantage is that the dog can deteriorate during treatment, possibly as far as grade 5, often due to poor owner compliance. In addition, there is a natural tendency to minimize the diagnostic evaluations for a dog to be treated by cage rest and consequently other causes for the neurological deficits can be overlooked. Physical therapy must also be delayed until the latter part of the treatment period and recovery of the neurological deficits may be slow or incomplete. Dogs that suffer recurrences after nonsurgical treatment may also have more severe deficits compared to dogs suffering recurrences after surgical therapy (Davies and Sharp, 1983; Dhupa et al., 1999a). A short course of corticosteroids without cage rest does not constitute effective non-surgical treatment. A high proportion of dogs referred for emergency decompression have been treated in the preceding days or weeks using corticosteroids without cage confinement (8.4). Humans experience a euphoric effect when on steroids and a similar phenomenon may also make dogs more active (Swinburn et al., 1988; Assimes and Lessard, 1999). Unrestrained activity renders the dog susceptible to further herniation of disc material and severe neurological deficits.
Table 8.2 Recovery times and recurrence rates after treatment for thoracolumbar disc disease Mean recovery time in weeks Neurological grade
Conservative4
Decompression7,8,9,10
Fenestration2
1—no deficits
3
⬍2
3
2—paresis – walking
6
⬍2
4
3—paresis – not walking
6
⬍2 (25, 6–73)
6–8
4—paraplegia
9–12
1–4
6–8
5—no deep pain
N/A
5–10 (59, 77, 363)
10
Recurrence of signs (%) (see ‘Prognosis’)
344, 406
193,10, 166, 138, 67, 41*
04, 22, 156
1
Brisson et al., 2002; 2 Butterworth and Denny, 1991; 3 Cudia and Duval, 1997; 4 Davies and Sharp, 1983; 5 Davis and Brown, 2001; 6 Levine and Caywood, 1984; 7 Olby et al., 2003; 8 Scott, 1997; 9 Scott and McKee, 1999; 10 Yovich et al., 1994. *After hemilaminectomy with fenestration of a variable number of disc spaces. N/A, not available.
125
126
Small Animal Spinal Disorders
8.4 Paraparetic Dachshund given large doses of corticosteroids without confinement. The dog had diarrhea, its packed cell volume (PCV) was 21% and its serum albumin 2.4 g/dl (normal ⬎ 3.0). PCV was 15% the next day and the albumin 1.7 g/dl, presumably secondary to corticosteroid-induced gastrointestinal bleeding. Melena was only evident after 4 days. The dog required 7 days intensive care prior to laminectomy at T11/12. One year previously it had been decompressed at T13/L1 and fenestrated from T12/13 to L3/4 inclusive.
Decompression (hemilaminectomy, mini-hemilaminectomy or pediculectomy) Although decompression is the treatment of choice for disc disease it does require good-quality imaging. Imaging usually identifies the affected interspace but determining which side to decompress can be more problematic. Myelography is more reliable than neurological signs, but CT or MR imaging are superior (Smith et al., 1997; Schulz et al., 1998; Olby et al., 1999). This is because spinal cord swelling can confound myelographic interpretation regarding the side of the lesion and because the disc material may actually be on both sides (Schulz et al., 1998) (8.3). Good surgical technique is needed to retrieve as much disc material as possible once the spinal cord has been exposed. Any material on the contralateral side must be removed by probing over or under the spinal cord; alternatively bilateral decompression may be needed (8.3). Bilateral hemilaminectomy does not appear to prejudice the outcome in small dogs despite causing an increase in range of motion or even overt instability (Anderson et al., 1991; Shires et al., 1991; Grevel and Schwartau, 1997; Corse et al., 2002; Viguier et al., 2002). Instability will be worsened by concomitant fenestration (Shires et al., 1991). Preservation of facet joints is therefore recommended on at least one side (Yovich et al., 1994). Decompression is the treatment of choice for dogs with spinal cord compression causing persistent or recurrent grade 1 signs and for most dogs with neurological deficits. The rate of recovery is faster after decompression
than after either non-surgical treatment or fenestration (Table 8.2) and there is less likelihood of residual neurological deficits (see ‘Prognosis’, page 132). Corticosteroids provide no overall benefit when used with decompressive surgery, except possibly MPSS for dogs that present within the first 8 h of injury (Olby, 1999) (see page 83). Disadvantages of decompression relate mainly to the need for advanced imaging and for special equipment and expertise (8.11–8.62). Specific recommendations for decompression include: • Progressive, persistent, or recurrent clinical signs. • Evidence of severe spinal cord compression. • Grade 5 lesions—these should be decompressed as soon as possible (see page 132). Concomitant fenestration should be performed at the time of the decompression (8.53). Probable exceptions include chronic disc herniations or extrusions in large-breed dogs because of the destabilizing effect (Shires et al., 1991) (8.51, 8.52). As many discs as possible should be fenestrated in chondrodystrophoid breeds, especially the discs between T11/12 and L2/3 inclusive. Decompression without fenestration can result in recurrence rates as high as 32% (Levine and Caywood, 1984; McKee, 1992). Up to 10% of dogs where adjacent discs were not fenestrated routinely at the time of decompression subsequently require another decompression due to a second herniation (Smith et al., 1997; Dhupa et al., 1999a). The re-operation rate falls to 4% when prophylactic fenestration of adjacent discs is performed (Brisson et al., 2002). • Dorsal laminectomy is not recommended as it has no advantages over hemilaminectomy and causes considerably more biomechanical instability (Smith and Walter, 1988). It may also increase intradiscal pressure, which could affect recurrence rates adversely (Lin et al., 1978; Shires et al., 1991). • Hemilaminectomy gives better access to extruded disc material than dorsal laminectomy and also makes fenestration easier (McKee, 1992; Muir et al., 1995). Compared to standard hemilaminectomy, pediculectomy and mini-hemilaminectomy have the advantage of preserving the articular processes (8.57), which is likely to retain stability especially when the disc is also fenestrated (Shires et al., 1991; Hill et al., 2000; Viguier et al., 2002). The decreased bone removal should also reduce surgery time and overall morbidity (Jeffery, 1988; Lubbe et al., 1994; McCartney, 1997). • Mini-hemilaminectomy removes less bone than a standard hemilaminectomy but only gives good access to the ventral portion of the vertebral canal (8.57–8.59). The greater tendency for hemorrhage when working around the foramen can also
Thoracolumbar disc disease
•
exacerbate the decreased exposure with mini-hemilaminectomy. Pediculectomy also preserves the facet joint and removes less bone than a standard hemilaminectomy (8.60–8.61). In contrast to a mini-hemilaminectomy, pediculectomy avoids the region of the foramen and its vessels. However, it also gives restricted access, especially as most disc material tends to be concentrated over the foramen (8.61B). To improve access, one or two pediculectomies can be merged with a mini-hemilaminectomy (Biggart, 1988; Lubbe et al., 1994; McCartney, 1997) (8.62). This combination can then readily be expanded to a standard hemilaminectomy if additional exposure is necessary.
Fenestration without decompression Some surgeons believe that fenestration alone has no merit as the sole surgical procedure. Decompression is the preferred treatment for extradural compression caused by disc disease because even dogs that present with back pain alone usually have significant spinal cord compression (Sukhiani et al., 1996). There are, however, a few situations in which to consider fenestration alone: • Presumed discogenic pain. • Recurrent back pain with minimal spinal cord compression. • Recurrent, mild ataxia and paresis associated with multiple small herniations. When fenestration is not combined with decompression it is done most easily in small dogs using a lateral approach. Although relatively straightforward in principle, fenestration requires a thorough understanding of anatomy and, like decompression, should be undertaken only after careful preparation and practice on cadavers. It is usual to fenestrate the discs from T11/12 to L3/4 via the lateral approach (8.63–8.78). The more caudal lumbar discs should be fenestrated in dogs with LMN deficits, taking particular care not to damage the large ventral branches of the spinal nerves at this level. The main advantage of fenestrating the majority of high-risk discs is that it reduces recurrence rates (Table 8.2). There is also no need for special instrumentation. In comparison to non-surgical therapy, fenestration has the added advantage that physiotherapy can be instituted immediately. The main disadvantages of fenestration are that recovery rates and times are prolonged for dogs with grade 3 and 4 deficits compared to decompression (Butterworth and Denny, 1991; Davies and Brown, 2001) (Tables 8.1, 8.2). Results are much worse than decompression for dogs with grade 5 deficits (Table 8.1). Residual neurological deficits are also more common than after decompression. Therefore fenestration alone
cannot be regarded as the treatment of choice for dogs with marked spinal cord compression. Other potential problems are that improper technique can occasionally force more disc material into the vertebral canal (8.77) and fenestration of chronic discs or in large dogs may decrease stability (Shires et al., 1991) (8.52).
COMPLICATIONS Complications common to both surgical and non-surgical management include gastrointestinal (GI) ulceration, iatrogenic hyperadrenocorticism, pancreatitis, pulmonary thromboembolism (PTE), deep vein thrombosis (DVT), decubitus, urine scald, urine retention and urinary tract infection (UTI) (see Chapters 6 and 15).
Intraoperative complications These are listed in Box 8.2. The main problems are finding the correct disc space and removing all of the disc material. The problems are discussed below. Box 8.2 Intraoperative complications ■
Improper identification of the surgical site (8.6, 8.21)
■
Inability to find disc material (8.6, 8.46, 8.61)
■
Inadequate removal of disc material (8.42, 8.61B)
■
Diffuse disc material (8.5, 8.55)
■
Adhesion of disc material to the dura (8.51)
■
Excessive hemorrhage (8.43)
■
Hemilaminectomy done on wrong side (8.6, 8.46, 8.61B)
■
Severe spinal cord swelling (8.49, 8.50, 8.55)
Early postoperative complications Several factors can account for neurological deterioration in the early postoperative period (Box 8.3). Box 8.3 Early postoperative complications ■
Myelomalacia (8.5)
■
Self-mutilation
■
Inadequate decompression (8.6)
■
Scoliosis
■
Second disc extrusion (8.7)
■
Body wall flaccidity
■
Fat graft (or Gelfoam) reaction (8.8)
■
Pneumothorax
■
Instability
■
Femoral nerve paralysis
■
Wound infection
■
Cutaneous fistula
127
128
Small Animal Spinal Disorders
•
Progressive myelomalacia (8.5) affects about 10% of dogs that present without nociception (Scott and McKee, 1999; Olby et al., 2003). Most affected dogs develop paralysis over less than 12 h; occasionally dogs present with grade 4 deficits but then go on to develop myelomalacia (Griffiths, 1972). The condition usually develops within 5 days of initial paralysis, with a range of 1–10 days; signs may therefore only become evident in the postoperative period. The condition then progresses over 3–7 days (Funkquist, 1962; Olby et al., 2003). Warning signs include depression, anorexia, vomiting, hypotension, toxemia, profound hyperesthesia, a cutaneous trunci cut-off that moves cranially, progression from UMN to LMN deficits and tetraparesis with abdominal breathing. Myelography reveals diffuse contrast medium infiltration into the cord parenchyma (Lu et al., 2002) (14.18). CSF usually shows very high protein levels even when taken from the CMC. Widespread malacia with epidural and subarachnoid hemorrhages develop, although these are not always evident at the time of surgery (8.5, 8.50). In most dogs with progressive
•
•
•
• 8.5 Dorsal laminectomy performed in a dog with clinical signs of progressive myelomalacia shows a swollen and grossly hemorrhagic spinal cord (arrow). It is very important to distinguish this situation from the more common one of focal malacia shown in 8.50 (see ‘Prognosis’, page 132).
A
B
myelomalacia the disc material spreads extensively along the epidural space, often encircling the dura mater but causing no direct spinal cord compression. It is likely that release of catecholamines and other substances causes severe, progressive vasospasm (Griffiths, 1972). As soon as clinical signs of this condition are recognized, euthanasia should be performed on humane grounds as patients usually progress from hypoventilation to asphyxiation. Occasionally the condition stops progressing before it kills the dog. Differential diagnoses for progressive myelomalacia include any coagulopathy that could cause intradural hemorrhage. If little or no disc material is retrieved and the animal shows no improvement, then repeat imaging should be considered promptly. It is possible that the wrong site was decompressed (8.6); more disc material was hidden on the opposite side of the spinal cord (Schulz et al., 1998) (8.6); disc material was missed at surgery (Dhupa et al., 1999a) (8.40, 8.61B); or that the dog is developing progressive myelomalacia (8.5, 8.55). Occasionally a dog will herniate a second disc in the early postoperative period. This can occur as a complication of improper fenestration (8.7) or from manipulation and loss of muscle tone under anesthesia. Aseptic necrosis of an excessively thick fat graft can cause deterioration within a few days of surgery (8.8). Adverse reactions have also been reported after Gelfoam (Pharmacia, Kalamazoo, MI) (Muir et al., 1995; Songer et al., 1990). Some dogs will deteriorate after surgery yet none of the above factors are present. Iatrogenic damage to the spinal cord is rare but can occur either during myelography (4.34) or at surgery. Postoperative deterioration seems to be a particular problem in dogs with chronic compression (8.51, 8.52). This apparent decompensation may be related to poor 8.6 This dog deteriorated after hemilaminectomy the previous day. Myelography had identified an L2/3 disc; the surgeon believed that decompression was from L2 to L4 on the right but no disc material was detected. 3D reconstruction of a CT scan made 24 h after hemilaminectomy revealed decompression was actually from L1 to L3. A: Coronal reconstruction shows material situated over L2/3 (arrow) at the caudal end of the decompression (arrowheads). The line shows the level of the transverse image in B. B: Transverse view looking forward to show the residual compression. A left-sided mini-hemilaminectomy was performed over L2/3; the dog made an excellent recovery.
Thoracolumbar disc disease
A
B
C
D
A
8.7 Series of four CT scans from a dog with acute paraparesis. Images in A and C were made at presentation; A shows a disc extrusion at T12/13. Images in B and D were made when the dog lost deep pain 36 h after surgery at T12/13; D shows a second disc extrusion at T11/12. A: Right-sided, extradural disc material (arrows) at T12/13. Material was retrieved by right-sided hemilaminectomy; fenestration was from T12/13 to L3/4 inclusive. Deep pain was present 12 h after surgery but was absent after 36 h. B: T12/13 at 36 h after surgery showing the hemilaminectomy defect with no residual disc material at this site (the increased opacity was a hematoma).C: T11/12 shown prior to surgery, made at the same time as the image in A. D: T11/12 shown 36 h after surgery, at the same time as the image in B. Mineralized disc material is now visible at T11/12 (arrows); this was not present prior to surgery. A second hemilaminectomy revealed new material from a herniation of the T11/12 disc (not fenestrated previously).
B
8.8 This dog had good deep pain 24 h after hemilaminectomy at L1/2 but had lost it after 48 h. Myelogram at 48 h reveals extradural compression over the hemilaminectomy defect. A: Marked ventral deviation of the dorsal contrast column (arrowheads). B: Thinning of right contrast column between arrowheads. The fat graft was swollen and edematous (12.10). It was replaced by Gelfoam (Pharmacia, Kalamazoo, MI) and the dog made a good recovery.
129
130
Small Animal Spinal Disorders
Table 8.3 Outcome for dogs according to size following decompression for thoracolumbar disc disease Neurological grade
As % success (total number)—ALL dogs*
As % success (total number)—LARGE dogs
2—paresis, walking
95 (38) 2,5,7
92 (13) 3
3, 4—not walking, paraplegia
94 (85) 2,5,7
90 (31) 3
5—no deep pain
64 (135) 1,4,5,6,7
25 (4) 3
1
Anderson et al., 1991; 2 Black, 1988; 3 Cudia and Duval, 1997; 4 Olby et al., 2003; 5 Scott, 1997; 6 Scott and McKee, 1999; 7 Yovich et al., 1994. *Mostly small-breed dogs.
•
perfusion in injured spinal cord segments, a reperfusion injury, or because dogs with chronic compression may have only just enough surviving axons to walk and so cannot afford to lose any more (see page 293, 132) (Blight and Decrescito, 1986; Cybulski and D’Angelo, 1988; Basso et al., 1996; Jeffery and Blakemore, 1999; Olby et al., 2003). Another potential cause for either early deterioration or long-term morbidity is low-grade instability after facet joint removal. Facet removal has minimal effects on lateral bending but significant effects during rotation, especially when it is combined with fenestration (Shires et al., 1991; Schulz et al., 1996; Viguier et al., 2002). Unilateral facetectomy has no effect on strength in studies of flexion and extension but does cause a significant increase (11%) in range of motion, which could have adverse effects on a spinal cord that is already severely injured (Smith and Walter, 1988). The biomechanical studies performed to date may not address all types of forces acting on the spine and have only been done in normal cadavers (Smith and Walter, 1988); the effects of severe disc degeneration on stability are unknown. In addition, animals with more severe injuries can probably generate only minimal protection from their paraspinal muscles. These various factors may be even more important in large-breed dogs, and may help to explain their apparent lower recovery rates, more frequent residual deficits and longer recovery times following grade 5 lesions compared to smaller dogs (Cudia and Duval, 1997; Olby et al., 2003) (Tables 8.2, 8.3). Facet preservation or concomitant vertebral stabilization are recommended in active, large-breed dogs (McKee, 2000) (see ‘Prognosis’, page 132). In humans, the more extensive the removal of bone the greater the subsequent morbidity (Eule et al., 1999; Papagelopoulos et al., 1997); therefore surgeries that preserve the facet
•
•
joints are a logical development for dogs and are in keeping with the trend towards microdiscectomy in humans (Hermantin et al., 1999) (8.6, 8.56–8.62). Fenestration can decrease stability, which may be of clinical relevance in large-breed dogs or at sites of chronic compression (Shires et al., 1991). It could also cause collapse of an interspace thereby increasing compression (McKee, 2000). Other potential problems during the early postoperative period include postoperative pain; wound discharge (14 of 264 dogs or 5%) (Hosgood, 1992); UTI (9 of 36 dogs or 26% recovering after decompression for grade 5 deficits) (Olby et al., 2003); or self-mutilation of the feet or penis (Olby et al., 2003) (see page 359). Scoliosis or flaccidity of the body wall is reported in up to 10% of dogs undergoing fenestration (Bartels et al., 1983; Black, 1988) (8.9). Pneumothorax (6 of 127 dogs or 5%) and femoral nerve paresis (4 of 127 dogs or 3%) may also occur (Bartels et al., 1983; Sukhiani et al., 1996). These problems are usually transient (8.9), but scoliosis can be permanent (Yovich et al., 1994).
Late postoperative complications The main problems in the late postoperative period are listed in Box 8.4.
•
•
Restrictive, peridural fibrosis or laminectomy scar is recognized only rarely in dogs (Applewhite et al., 1999)(see Chapter 6). It is most likely in a dog that has had a previous, wide, dorsal laminectomy; if hemilaminectomy was combined with excessive removal of bone dorsally; following hemilaminectomy over several interspaces (8.56); or if there has been a chronic reaction to the material placed at the surgery site (Muir et al., 1995). Infection of an intervertebral space may occur, which can be iatrogenic (Funkquist, 1978) (see ‘Discospondylitis’, page 326, 13.34, 14.14).
Thoracolumbar disc disease
•
8.9 Mild kyphosis and body wall flaccidity (arrowheads) in a Cocker spaniel the day after a right-sided hemilaminectomy and fenestration for intervertebral disc disease. Signs resolved gradually over a few days and were assumed to have been due to neurapraxic injury to nerve root(s) or spinal nerve(s).
Box 8.4 Late postoperative complications ■
Peridural fibrosis
■
Residual neurological deficits
■
Infection
■
Fecal incontinence
■
Disc extrusion at a new site
■
Recurrent UTI
•
•
There may be recurrence of signs, which usually occurs between 1 month and 2 years after surgery due to late disc herniation at a new space (Dhupa et al., 1999a) (8.4). Although still controversial, it appears that fenestration reduces the recurrence rate and that the more discs fenestrated, the greater this reduction (Funkquist, 1978; Fingeroth, 1989; Yovich et al., 1994; Olby et al., 2003 (Table 8.2). Twelve of 265 dogs suffered recurrences (4%) in one study; in 10 out of 12 of these dogs (83%) recurrence was at an interspace that had not been fenestrated at the original surgery (Brisson et al., 2002). Residual neurological deficits, usually mild paraparesis or pelvic limb ataxia, affect about 20–25% of dogs presenting with severe deficits (McCartney, 1997; Scott, 1997). This rate appears to be higher (39%) for large-breed dogs (Cudia and
•
Duval, 1997) (see ‘Prognosis’, page 132). One factor that may contribute to the residual neurological deficits in some dogs is inadequate spinal cord decompression. Postoperative vertebral canal stenosis can cause development of syringohydromyelia in humans and may be an overlooked cause of residual neurological deficits in animals (PerrouinVerbe et al., 1998; Fischbein et al., 1999; Bains et al., 2001). Fecal incontinence has been reported in 5 to 39% of dogs recovering from surgery (Anderson et al., 1991; Cudia and Duval, 1997; Dhupa et al., 1999a; Olby et al., 2003). This problem is usually only an intermittent one. Of 36 dogs with no deep pain that recovered nociception after surgery, 14 (39%) had fecal incontinence but the owners only perceived this to be a problem in three (8%) (Olby et al., 2003). Mild urinary incontinence also occurred in most of the dogs that suffered from postoperative fecal incontinence (31%) (Olby et al., 2003). Recurrent UTI occurs in some dogs recovering from grade 5 deficits (Olby et al., 2003). This may be due to an underlying problem, such as pyelonephritis or cystic calculi. However, a more important cause of recurrent UTI is when a dog recovers the ability to walk without recovering continence or nociception (also termed spinal reflex walking, see Chapter 6, page 87). This problem is almost certainly underrecognized. It was generally believed that useful motor function only recovers once deep pain had already returned and so most dogs are not checked for deep pain once they begin to walk (Wheeler and Sharp, 1994; Oliver et al., 1997). However, 8 of 19 dogs (42%) that never regained deep pain sensation after surgery still recovered the ability to walk. These dogs took between 4 and 18 months (mean 9 months) to walk and yet none had deep pain sensation on re-examination and all suffered from incontinence and recurrent UTI. Interestingly, despite these problems the owner of each dog was happy with the outcome. If all dogs that regained motor function in this study are considered to have had a successful outcome, then 44 of the 64 dogs (69%) presenting with grade 5 lesions recovered overall (Olby et al., 2003). The small population of dogs that walk without regaining deep pain could explain the apparent discrepancy in outcome between studies that checked all dogs specifically for recovery of nociception at long-term follow-up and studies that used telephone follow-up for some dogs (Table 8.1a). Dogs with recurrent UTI should therefore be assessed critically for deep pain sensation.
131
132
Small Animal Spinal Disorders
POSTOPERATIVE CARE
(see Chapter 15)
Extended cage rest is not usually necessary although it may be recommended for large-breed dogs, after bilateral hemilaminectomy or after decompression of chronic disc lesions. Postoperative physical therapy together with restricted exercise on a leash are indicated (see Chapter 15). Corticosteroids are not recommended other than possibly low-dose prednisone for analgesia. By far the most important aspect of postoperative care is to insure that the bladder is emptied regularly of all urine. Urinary retention is the most common postoperative problem in dogs paralyzed due to thoracolumbar disc disease. Such dogs must have regular urinalysis and may require pharmacological intervention (see ‘Control of urinary function’, page 350; Table 15.7).
PROGNOSIS The prognosis is excellent for dogs with grade 1, 2 and 3 deficits, especially following decompression. Dogs with grade 3 and 4 deficits have a better outcome after surgery and the response is better after decompression than after fenestration alone. Decompression is clearly the treatment of choice for dogs with grade 5 lesions, with between 60 and 70% making a functional recovery in most studies (Tables 8.1, 8.1a). When deep pain was assessed using bone clamps or pliers the rate was 62% (23/37) (Scott and McKee, 1999). Specific evaluation of all dogs for recovery of deep pain, rather than the ability to walk, also brings the recovery rate close to 60% (Olby et al., 2003) (Table 8.1a). Most dogs that make a functional recovery have regained deep pain within 2 weeks. However, 19% (8/42) recovered deep pain in the 3rd or 4th weeks and so a final assessment should not be made prior to 1 month post-surgery (Scott and McKee, 1999; Olby et al., 2003). Recovery of deep pain may take even longer for large-breed dogs (Olby et al., 2003). Studies show conflicting results for both the prognostic value of the speed of onset of clinical signs and of spinal cord swelling at myelography (Duval et al., 1996; Scott and McKee, 1999; Olby et al., 2003). The prognosis for dogs with LMN signs was no worse than those with UMN signs (Dhupa et al., 1999b; Olby et al., 2003). Furthermore, prognosis did not correlate with the duration of signs prior to surgery. Good results were still obtained if surgery was performed within 48 h, or even more than 72 h after onset of signs (Anderson et al., 1991; Scott and McKee, 1999; Olby et al., 2003), which is in contrast to the recommendations in many standard textbooks. Taken together, these data suggest that a delay in decompression of a few hours seems to make little difference to outcome, with the possible exception of dogs
with rapidly progressive signs (Anderson et al., 1991; Scott and McKee, 1999; Olby et al., 2003). However, logic would still dictate that early decompression be a high priority when a dog presents with severe deficits. Prognosis has been inferred from the presence of spinal cord liquifaction following a durotomy. However, euthanasia based on a malacic, toothpaste-like spinal cord at durotomy almost certainly results in the death of some dogs that would otherwise have recovered (Olby et al., 2003) (8.50). In a study where durotomy was routine and dogs with malacic cords were euthanized, only 20 of 46 dogs (43%) with grade 5 lesions recovered. If the 16 dogs that were euthanized at surgery are excluded, the recovery rate for the other 30 dogs becomes 66% (Duval et al., 1996). This is more in line with the overall recovery rate for dogs with grade 5 lesions shown in Table 8.1. As few as 5–10% of axons surviving within a lesion appear to allow functional recovery and these would be impossible to identify at durotomy (Blight and Decrescito, 1986; Basso et al., 1996; Jeffery and Blakemore, 1999; Olby et al., 2003). Therefore the significance of focal malacia, as well as the difference between it and the syndrome of progressive myelomalacia, should not be determined at surgery. Dogs or cats with extensive malacia should only be euthanized based on either a lack of recovery within 4 weeks or if they develop clinical signs of progressive myelomalacia (Salisbury and Cook, 1988; Muir et al., 1995; Olby et al., 2003). Recurrence rates of clinical signs once a dog has recovered from surgery ranged from 6–9% (Table 8.2). Confirmed recurrences at a new disc space (8.4, 8.7) occur in 5–8% of dogs but these are probably underestimates of the overall figure (Muir et al., 1995; Smith et al., 1997; Dhupa et al., 1999a). When a recurrence does occur it is often at a site that has not been fenestrated (Levine and Caywood, 1984; Muir et al., 1995; Smith et al., 1997) (8.4, 8.7). Fenestration is a low-risk procedure that, when combined with decompression, appears to lower the recurrence rate. However, this assertion remains to be proven by randomized, prospective studies; the role of fenestration in chronic disc disease also needs to be defined (Levine and Caywood, 1984; McKee, 1992; Lubbe et al., 1994; Yovich et al., 1994; Brisson et al., 2002; Olby et al., 2003) (8.51, 8.52). Fortunately, if dogs do require a second decompression for recurrent extrusion the prognosis appears to be no worse than after the first surgery (Dhupa et al., 1999a). Although the recovery rate for large-breed dogs is reported as 91%, this only applies to dogs that present with good deep pain (grades 2–4) (Cudia and Duval, 1997). Results for dogs with grade 5 lesions seem to be much worse. Only 1 of 4 large dogs (25%) with grade 5 signs recovered compared to 86 of 135 small dogs (64%,
Thoracolumbar disc disease
8.10 Lateral myelogram of a cat with a disc extrusion causing extradural compression at T13/L1. The CT image taken at this interspace following myelography shows a mineralized, left-sided extradural mass (arrows).
A
B
Table 8.3). Recovery times after grade 5 lesions also appear to be prolonged for large-breed dogs, with a mean of 36 weeks compared to 5–7 weeks for small breeds (Cudia and Duval, 1997; Scott and McKee, 1999; Olby et al., 2003) (Table 8.2). Furthermore, of the 41 largebreed dogs that recovered following decompression, 16 (39%) had residual neurological deficits compared to 20–25% of small-breed dogs (Cudia and Duval, 1997; McCartney, 1997; Scott, 1997) (Table 8.2). In addition, 8 of the 41 large dogs (19%) that recovered then suffered a recurrence of signs, of which 5 became paralyzed (Cudia and Duval, 1997). This recurrence rate is higher than the rates from most other studies (Table 8.2). If this trend towards disappointing results is borne out in subsequent studies of large-breed dogs then results might be improved through facet joint preservation using pediculectomy or mini-hemilaminectomy, provided that all disc material can be removed. An alternative is for some form of stabilization to be applied following a standard hemilaminectomy. In addition, fenestration should probably not be performed at the affected site as it can compromise stability (Shires et al., 1991; McKee, 2000). The most important points relating to prognosis are that: • The clinician is able to assess the neurological status of the dog accurately. • Any dog presenting with, or subsequently developing, a grade 5 neurological status should undergo decompression as soon after presentation as possible. • If deep pain does not return within 4 weeks the prognosis for full recovery is poor (Olby et al., 2003). • MPSS produces only minor improvements in human spinal cord injury and is unlikely to cause significant improvements in outcome for dogs or cats (Hurlbert, 2000) (see page 83). It may help some animals with grade 5 lesions but it must be combined with surgery. • Other corticosteroids are contraindicated as they may worsen neurological outcome and increase the
complication rate (see page 83). They relieve pain and inflammation but do not lessen spinal cord injury. Recovery from spinal cord injury takes time; this should be combined with decompression where indicated.
THORACOLUMBAR DISC DISEASE IN CATS Neurological deficits caused by disc disease in cats are more common than was thought previously. Affected cats tend to be older but often have acute, type I extrusions (8.10). Cats with neurological deficits may not show back pain. Lesions are visible on CT scan although subarachnoid contrast may be helpful. Differential diagnoses to be considered include trauma, neoplasia (lymphoma), inflammatory CNS disease (feline infectious peritonitis—FIP) and ischemic neuromyopathy (see Chapter 14). A diagnosis of feline disc disease should only be made after thorough patient evaluation and neuroimaging (Knipe et al., 2001; Munana et al., 2001). The response to surgery appears to be equivalent to that reported in dogs.
Key issues for future investigation 1. Is there any clinical advantage for either small- or large-breed dogs in preserving the articular facets? 2. Does durotomy provide any additional decompression for spinal cord swelling? 3. Does the long-term prognosis depend on the extent of bone removal after extensive hemilaminectomy? 4. How effectively do we remove 100% of extruded disc material and does this affect outcome? 5. Should a chronic disc be fenestrated; how do we define ‘chronic disc’?
133
134
Small Animal Spinal Disorders
REFERENCES Anderson, S.M., Lippincott, C.L., Gill, P.J. (1991) Hemilaminectomy in dogs without deep pain perception. A retrospective study of 32 cases. California Veterinarian 45, 24–28. Applewhite, A.A., Wilkens, B.E., McDonald, D.E., Radasch, R.M., Barstad, R.D. (1999) Potential central nervous system complications of von Willebrand’s disease. Journal of the American Animal Hospital Association 35, 423–429. Assimes, T.L., Lessard, M.L. (1999) The use of perioperative corticosteroids in craniomaxillofacial surgery. Plastic Reconstructive Surgery 103, 313–321; quiz 322. Bagley, R.S., Harrington, M.L., Pluhar, G.E., Keegan, R.D., Greene, S.A., Moore, M.P., Gavin, P.R. (1996) Effect of craniectomy/durotomy alone and in combination with hyperventilation, diuretics, and corticosteroids on intracranial pressure in clinically normal dogs. American Journal of Veterinary Research 57, 116–119. Bains, R.S., Althausen, P.L., Gitlin, G.N., Gupta, M.C., Benson, D.R. (2001) The role of acute decompression and restoration of spinal alignment in the prevention of post-traumatic syringomyelia: case report and review of recent literature. Spine 26, E399–402. Bartels, K.E., Creed, J.E., Yturraspe, D.J. (1983) Complications associated with the dorsolateral muscle-separating approach for thoracolumbar disk fenestration in the dog. Journal of the American Veterinary Medical Association 183, 1081–1083. Basso, D.M., Beattie, M.S., Bresnahan, J.C. (1996) Graded histological and locomotor outcomes after spinal cord contusion using the NYU weightdrop device versus transection. Experimental Neurology 139, 244–256. Biggart, J.F., III (1988) Prevention of laminectomy membrane by free fat grafts after laminectomy in dogs with disk herniations. Veterinary Surgery 17, 28–29. Bitetto, W.V., Thacher, C. (1987) A modified lateral decompressive technique for treatment of canine intervertebral disk disease. Journal of the American Animal Hospital Association 23, 409–413. Black, A.P. (1988) Lateral spinal decompression in the dog; a review of 39 cases. Journal of Small Animal Practice 29, 581–588. Blight, A.R., Decrescito, V. (1986) Morphometric analysis of experimental spinal cord injury in the cat: the relation of injury intensity to survival of myelinated axons. Neuroscience 19, 321–341. Braund, K.G., Taylor, T.K.F., Ghosh, P., Sherwood, A.A. (1976) Lateral spinal decompression in the dog. Journal of Small Animal Practice 17, 583–592. Brisson, B.A., Moffat, S.L., Parent, J.M. (2002) Canine intervertebral disc disease; retrospective evaluation of 265 cases: 1995–1999. Veterinary Surgery 31, 500. Butterworth, S.J., Denny, H.R. (1991) Follow-up study of 100 cases with thoracolumbar disc protrusions treated by lateral fenestration. Journal of Small Animal Practice 32, 443–447. Corse, M.R., Renberg, W.C., Friis, E.A. (2002) Biomechanical effects of multiple hemilaminectomy in the canine lumbar spine. Veterinary Sugery 31, 500. Cudia, S.P., Duval, J.M. (1997) Thoracolumbar intervertebral disk disease in large, nonchondrodystrophic dogs: a retrospective study. Journal of the American Animal Hospital Association 33, 456–460. Cybulski, G.R., D’Angelo, C.M. (1988) Neurological deterioration after laminectomy for spondylotic cervical myeloradiculopathy: the putative role of spinal cord ischemia. Journal of Neurology 51, 717–718. Davies, J.V., Sharp, N.J.H. (1983) A comparison of conservative treatment and fenestration for thoracolumbar intervertebral disc disease in the dog. Journal of Small Animal Practice 24, 721–729. Davis, G.J., Brown, D.C. (2001) Prognostic indicators for time to ambulation in non-ambulatory dogs with acute thoracolumbar disk extrusions: 115 cases. Veterinary Surgery 30, 491. Dhupa, S., Glickman, N., Waters, D.J., Sarit, D. (1999a) Reoperative neurosurgery in dogs with thoracolumbar disc disease. Veterinary Surgery 28, 421–428. Dhupa, S., Glickman, N.W., Waters, D.J., Dhupa, S. (1999b) Functional outcome in dogs after surgical treatment of caudal lumbar intervertebral disk herniation. Journal of the American Animal Hospital Association 35, 323–331. Duval, J., Dewey, C., Roberts, R., Aron, D. (1996) Spinal cord swelling as a myelographic indicator of prognosis: a retrospective study in dogs with
intervertebral disc disease and loss of deep pain perception. Veterinary Surgery 25, 6–12. Eule, J., Breeze, R., Kindt, G. (1999) Bilateral partial laminectomy: a treatment for lumbar spinal stenosis and midline disc herniation. Surgical Neurology 52, 329–337; discussion 337–338. Fingeroth, J.M. (1989) Fenestration. Pros and cons. Problems in Veterinary Medicine 1, 445–466. Fischbein, N.J., Dillon, W.P., Cobbs, C., Weinstein, P.R. (1999) The ‘presyrinx’ state: a reversible myelopathic condition that may precede syringomyelia. American Journal of Neuroradiology 20, 7–20. Funkquist, B. (1962) Thoraco-lumbar disk protrusion with severe cord compression in the dog. I: Clinical and pathoanatomical observations with special reference to the role of development of the symptoms of motor loss. Acta Veterinaria Scandinavia 3, 256–274. Funkquist, B. (1978) Investigations of the therapeutic and prophylactic effects of disc evacuation in cases of thoraco-lumbar herniated discs in dogs. Acta Veterinaria Scandinavica 19, 441–457. Grevel, V., Schwartau, K. (1997) Hemilaminectomy in thoracolumbar disc disease in the dog. Part 2: Intraoperative findings and surgical results. Kleintierpraxis 42, 173–196. Griffiths, I.R. (1972) The extensive myelopathy of intervertebral disc protrusions in dogs (‘the ascending syndrome’). Journal of Small Animal Practice 13, 425–438. Henry, A., Tunkel, R., Arbit, E., Ku, A., Lachmann, E. (1997) Tethered thoracic cord resulting from spinal cord herniation. The Archives of Physical Medicine and Rehabilitation 78, 530–533. Hermantin, F., Peters, T., Quartararo, L., Kambin, P. (1999) A prospective, randomized study comparing the results of open discectomy with those of video-assisted arthroscopic microdiscectomy. The Journal of Bone and Joint Surgery—American volume 81, 958–965. Hill, T.P., Lubbe, A.M., Guthrie, A.J. (2000) Lumbar spine stability following hemilaminectomy, pediculectomy, and fenestration. Veterinary and Comparative Orthopaedics and Traumatology 13, 165–171. Holmberg, D.L., Palmer, N.C., Vanpelt, D., Willan, A.R. (1990) A comparison of manual and power-assisted thoracolumbar disc fenestration in dogs. Veterinary Surgery 19, 323–327. Hosgood, G. (1992) Wound complications following thoracolumbar laminectomy in the dog: a retrospective study of 264 procedures. Journal of the American Animal Hospital Association 28, 47–52. Hurlbert, R.J. (2000) Methylprednisolone for acute spinal cord injury: an inappropriate standard of care. Journal of Neurosurgery 93 (1 Suppl), 1–7. Jeffery, N.D. (1988) Treatment of acute and chronic thoracolumbar disc disease by ‘mini hemilaminectomy’. Journal of Small Animal Practice 29, 611–616. Jeffery, N.D., Blakemore, W.F. (1999) Spinal cord injury in small animals 2. Current and future options for therapy. Veterinary Record 145, 183–190. Kirberger, R.M., Roos, C.J., Lubbe, A.M. (1992) The radiological diagnosis of thoracolumbar disc disease in the Dachshund. Veterinary Radiology and Ultrasound 33, 255–261. Knipe, M.F., Vernau, K.M., Hornof, W.J., LeCouteur, R.A. (2001) Intervertebral disc extrusion in six cats. Journal of Feline Medicine and Surgery 3, 161–168. Levine, S.H., Caywood, D.D. (1984) Recurrence of neurological deficits in dogs treated for thoracolumbar disk disease. Journal of the American Animal Hospital Association 20, 889–894. Lin, H.S., Liu, Y.K., Adams, K.H. (1978) Mechanical response of the lumbar intervertebral joint under physiological (complex) loading. The Journal of Bone and Joint Surgery—American volume 60, 41–55. Lu, D., Lamb, C.R., Targett, M.P. (2002) Results of myelography in seven dogs with myelomalacia. Veterinary Radiology and Ultrasound 43, 326–330. Lubbe, A.M., Kirberger, R.M., Verstraete, F.J.M. (1994) Pediculectomy for thoracolumbar spinal decompression in the Dachshund. Journal of the American Animal Hospital Association 30, 233–238. Maroon, J.C., Abla, A., Bost, J. (1999) Association between peridural scar and persistent low back pain after lumbar discectomy. Neurological Research 21, Suppl 1, S43–46. McCartney, W. (1997) Partial pediculectomy for the treatment of thoracolumbar disc disease. Veterinary and Comparative Orthopaedics and Traumatology 10, 117–121. McKee, M. (2000) Intervertebral disc disease in the dog: 2. Management options. In Practice 22, 458–471.
Thoracolumbar disc disease
McKee, W.M. (1992) A comparison of hemilaminectomy (with concomitant disc fenestration) and dorsal laminectomy for the treatment of thoracolumbar disc protrusion in dogs. Veterinary Record 130, 296–300. Milette, P.C., Fontaine, S., Lepanto, L., Cardinal, E., Breton, G. (1999) Differentiating lumbar disc protrusions, disc bulges, and discs with normal contour but abnormal signal intensity. Magnetic resonance imaging with discographic correlations. Spine 24, 44–53. Moissonnier, P., Carozzo, C., Meheust, P. (2002) Lateral corpectomy as a treatment of chronic disc herniation in 15 dogs. Veterinary Surgery 31, 296. Muir, P., Johnson, K.A., Manley, P.A., Dueland, R.T. (1995) Comparison of hemilaminectomy and dorsal laminectomy for thoracolumbar intervertebral disc extrusion in Dachshunds. Journal of Small Animal Practice 36, 360–367. Munana, K.R., Olby, N.J., Sharp, N.J., Skeen, T.M. (2001) Intervertebral disk disease in 10 cats. Journal of the American Animal Hospital Association 37, 384–389. Olby, N. (1999) Current concepts in the management of acute spinal cord injury. Journal of Veterinary Internal Medicine 13, 399–407. Olby, N.J., Dyce, J., Houlton, J.E.F. (1994) Correlation of plain radiographic and lumbar myelographic findings with surgical findings in thoracolumbar disc disease. Journal of Small Animal Practice 35, 345–350. Olby, N.J., Munana, K.R., Sharp, N.J.H., Flegel, T., Van Camp, S., Berry, C.R., Thrall, D.G. (1999) A comparison of computed tomography and myelography in the diagnosis of acute intervertebral disc disease in dogs. Journal of Veterinary Internal Medicine 13, 239. Olby, N.J., Munana, K.R., Sharp, N.J.H., Thrall, D.E. (2000) The computed tomographic appearance of acute thoracolumbar intervertebral disc herniations in dogs. Veterinary Radiology and Ultrasound 41, 396–402. Olby, N.J., de Risio, L., Munana, K.R., Wosar, M.A., Skeen, T.M., Sharp, N.J.H., Keene, B.W. (2001) Development of a functional scoring system in dogs with acute spinal cord injuries. American Journal of Veterinary Research 62, 1624–1628. Olby, N.J., Harris, T., Munana, K.R., Skeen, T.M., Sharp, N.J.H. (2003) Long-term functional outcome of dogs with severe spinal cord injuries. Journal of the American Veterinary Medical Association 222, 762–769. Oliver, J.E., Lorenz, M.D., Kornegay, J.N. (1997) Handbook of Veterinary Neurology, 3rd edn. Philadelphia: WB Saunders. Osterholm, J. (1974) The pathophysiological response to spinal cord injury. The current status of related research. Journal of Neurosurgery 40, 5–33. Papagelopoulos, P., Peterson, H., Ebersold, M., Emmanuel, P., Choudhury, S., Quast, L. (1997) Spinal column deformity and instability after lumbar or thoracolumbar laminectomy for intraspinal tumors in children and young adults. Spine 22, 442–451. Parker, A.J. (1973) Distribution of spinal branches of the thoracolumbar segmental arteries in dogs. American Journal of Veterinary Research 34, 1351–1353. Perrouin-Verbe, B., Lenne-Aurier, K., Robert, R., Auffray-Calvier, E., Richard, I., Mauduyt de la Greve, I., Mathe, J.F. (1998) Post-traumatic syringomyelia and post-traumatic spinal canal stenosis: a direct relationship: review of 75 patients with a spinal cord injury. Spinal Cord 36, 137–143. Ross, J.S., Robertson, J.T., Frederickson, R.C., Petrie, J.L., Obuchowski, N., Modic, M.T., deTribolet, N. (1996) Association between peridural scar and recurrent radicular pain after lumbar discectomy: magnetic resonance evaluation. ADCON-L European Study Group. Neurosurgery 38, 855–861; discussion 861–863. Salisbury, S.K., Cook, J.R., Jr (1988) Recovery of neurological function following focal myelomalacia in a cat. Journal of the American Animal Hospital Association 24, 227–230.
Schulz, K.S., Waldron, D.R., Grant, J.W., Shell, L., Smith, G., Shires, P.K. (1996) Biomechanics of the thoracolumbar vertebral column of dogs during lateral bending. American Journal of Veterinary Research 57, 1228–1232. Schulz, K.S., Walker, M., Moon, M., Waldron, D., Slater, M., McDonald, D.E. (1998) Correlation of clinical, radiographic, and surgical localization of intervertebral disc extrusion in small-breed dogs: a prospective study of 50 cases. Veterinary Surgery 27, 105–111. Scott, H.W. (1997) Hemilaminectomy for the treatment of thoracolumbar disc disease in the dog: a follow-up study of 40 cases. Journal of Small Animal Practice 38, 488–494. Scott, H.W., McKee, W.M. (1999) Laminectomy for 34 dogs with thoracolumbar intervertebral disc disease and loss of deep pain perception. Journal of Small Animal Practice 40, 417–422. Shires, P.K., Waldron, D.R., Hedlund, C.S., Blass, C.E., Massoudi, L. (1991) A biomechanical study of rotational instability in unaltered and surgically altered canine thoracolumbar vertebral motion units. Progress in Veterinary Neurology 2, 6–14. Smith, G.K., Walter, M.C. (1988) Spinal decompressive procedures and dorsal compartment injuries: comparative biomechanical study in canine cadavers. American Journal of Veterinary Research 49, 266–273. Smith, J.D., Newell, S.M., Budsberg, S.C., Bennett, R.A. (1997) Incidence of contralateral versus ipsilateral neurological signs associated with lateralised Hansen type I disc extrusion. Journal of Small Animal Practice 38, 495–497. Songer, M., Ghosh, L., Spencer, D. (1990) Effects of sodium hyaluronate on peridural fibrosis after lumbar laminotomy and discectomy. Spine 15, 550–554. Sukhiani, H.R., Parent, J.M., Atilola, M.A.O., Holmberg, D.L. (1996) Intervertebral disk disease in dogs with signs of back pain alone: 25 cases (1986–1993). Journal of the American Veterinary Medical Association 209, 1275–1279. Swinburn, C.R., Wakefield, J.M., Newman, S.P., Jones, P.W. (1988) Evidence of prednisolone induced mood change (‘steroid euphoria’) in patients with chronic obstructive airways disease. British Journal of Clinical Pharmacology 26, 709–713. Thomson, C.E., Kornegay, J.N., Stevens, J.B. (1989) Canine intervertebral disc disease: changes in the cerebrospinal fluid. Journal of Small Animal Practice 30, 685–688. Trevor, P.B., Martin, R.A., Saunders, G.K., Trotter, E.J. (1991) Healing characteristics of free and pedicle fat grafts after dorsal laminectomy and durotomy in dogs. Veterinary Surgery 20, 282–290. Viguier, E., Petit-Etienne, G., Magnier, J., Diop, A., Lavaste, F. (2002) Mobility of T13-L1 after spinal cord decompression procedures in dogs (an in vitro study). Veterinary Surgery 31, 297. Wheeler, S.J., Sharp, N.J.H. (1994) Small Animal Spinal Disorders: Diagnosis and Surgery, 1st edn. St Louis: Mosby. Widmer, W.R., DeNicola, D.B., Blevins, W.E., Cook, J.R., Jr, Cantwell, H.D., Teclaw, R.F. (1992) Cerebrospinal fluid changes after iopamidol and metrizamide myelography in clinically normal dogs. American Journal of Veterinary Research 53, 396–401. Wilkens, B.E., Selcer, R., Adams, W.H., Thomas, W.B. (1996) T9–T10 intervertebral disc herniation in three dogs. Veterinary and Comparative Orthopaedics and Traumatology 9, 177–178. Yovich, J.C., Read, R., Eger, C. (1994) Modified lateral spinal decompression in 61 dogs with thoracolumbar disc protrusion. Journal of Small Animal Practice 35, 351–356.
135
136
Small Animal Spinal Disorders
PROCEDURES Dorsolateral hemilaminectomy (8.11–8.56)
8.11 Completed hemilaminectomy with extruded disc material in situ. Ideal prerequisites to perform this procedure are: •
Identification of the affected interspace(s).
•
Knowledge of how disc material is distributed threedimensionally across the vertebral canal at the affected interspace(s).
•
Adequate exposure of the interspace(s) to be decompressed.
•
Removal of a minimum of bone in order to access disc material and decompress the spinal cord. In retrospect, the surgeon could have removed this disc material by mini-hemilaminectomy.
•
Removal of extruded disc material without traumatizing the spinal cord.
8.11
8.12 Incision site for caudal thoracic and cranial lumbar intervertebral spaces. The incision can be modified according to the exact site of compression. Mini-hemilaminectomy is made easier if the dog is rotated somewhat more to one side than for a standard hemilaminectomy. 8.12
8.13
Skin and superficial tissues are incised 1 cm from midline so that the incision does not rest over the spinous processes on closure. Subcutaneous fat is reflected for 1 cm on either side of midline to facilitate closure. This reveals the lumbodorsal fascia, which is incised on the near side of each spinous process over five vertebrae. A periosteal elevator is then used to lever muscle away from the near side of each spinous process. Muscular insertions on the cranial and caudal ends of each process are cut. One process is exposed already (arrow).
8.13
Thoracolumbar disc disease
8.14 Here a second spinous process has been exposed (arrow). Electrocautery to cut muscular insertions on the spinous processes reduces minor hemorrhage during these superficial stages of vertebral exposure.
8.14
8.15 A Langenbeck or Senn retractor (5.9) is placed adjacent to the spinous process and pulled laterally, and slightly cranially, to facilitate exposure of the dorsal surface of the articular facet joint.
8.15
8.16 Continued retraction has exposed the articular processes (arrow) with muscles still attached. While maintaining traction, the muscular attachments onto the articular facets are cut as close as possible to the joint capsule (white arrows). Bipolar electrocautery is helpful to retard hemorrhage. 8.16
8.17 The muscular attachments have been severed, and the isolated articular facets are visible clearly (arrow). This process is now repeated for one to two facet joints on each side of the site to be decompressed. Gelpi retractors are placed to facilitate exposure.
8.17
137
138
Small Animal Spinal Disorders
8.18 Landmarks must now be evaluated carefully to insure that the final dissection is done at the correct location. Landmarks include the spinous process of the anticlinal vertebra (1.20B); the difference between the transverse process of the first lumbar vertebra 8.18 and the head of the last rib (8.19, 8.20); and the relationship of the spinous process of the sixth lumbar vertebra to the wings of the ilium (4.15).
8.19 3D reconstruction of a CT scan to show that the caudally directed, proximal portion of the thirteenth rib is distinct from the short, somewhat cranially directed, first lumbar transverse process. It is crucial that landmarks are identified accurately prior to further dissection (8.6). Vertebral anomalies occur in 10–15% of Dachshunds (Jeffery, 1988; Kirberger et al., 1992) (8.21).
8.19
8.20 The proximal portion of the thirteenth rib is much longer (arrow) than the short transverse process of the first lumbar vertebra. The tip of the transverse process can also be palpated with an elevator, in distinction to the ribs. The angle of the last rib and the size of the first transverse process do vary considerably between animals. The transverse process is also deeper than the rib. This dog has a neurovascular bundle running over the rib.
8.20
Thoracolumbar disc disease
8.21 This CT reconstruction is from a dog with an anomalous rib; the surgeon was misled into performing the hemilaminectomy at the wrong interspace. On finding no disc material, the surgeon rechecked the scout image of the CT scan and noticed that there was a vestigial thirteenth rib. This had been misidentified as the first lumbar transverse process.
8.21
8.22 Once landmarks are ascertained, a Gelpi retractor is placed dorsal to the articular facet on either side of the interspace(s) to be decompressed. Retractors are positioned between an interspinous space and the epaxial muscles; a moistened laparotomy sponge is placed over the muscle for protection prior to distraction. 8.22
8.23 If the anatomy around the chosen intervertebral space is not clear, a periosteal elevator and dry gauze sponge can be used to clean away tags of muscle. This should not be done excessively as a large area of periosteal irritation could increase postoperative pain. Here the sponge has clarified an accessory process.
8.23
139
140
Small Animal Spinal Disorders
8.24 There is a long, tendinous attachment of the longissimus muscle (arrowhead) to the accessory process (arrow) of the vertebra just in front of each disc. There is frequently a small artery located dorsal to this process that requires bipolar cautery. (*) Articular process. 8.24
8.25 Cutting the longissimus tendon (arrowhead) close to its insertion on the accessory process (asterisk) by cutting away from the underlying spinal artery and nerve. The tendon is exposed by putting pressure on it with an elevator (arrow). Tendons are cut over one to three interspace(s). 8.25
8.26 The tendinous attachment to the accessory process has now been cut and the main branches of the spinal artery, vein and nerve appear as a common neurovascular bundle (arrow). (*) Articular process. Saline-soaked laparotomy sponges are repositioned in such a way as not to interfere with drilling.
8.27 Diagram to show the relationship of the accessory process to the intervertebral foramen, through which the neurovascular bundle passes.
8.27
8.26
Thoracolumbar disc disease
8.28 Rongeurs are used to remove the articular processes at the site of entry into the vertebral canal. Rongeurs can be used to perform the entire hemilaminectomy if desired; adjacent spinous processes are grasped with bone clamps, which are then used to lever the two vertebrae apart. The rongeurs can then be introduced carefully into the intervertebral foramen to start bone removal.
8.29 Relationship of the suction tip (arrow) in the ‘gutter’ formed between the vertebral body and the epaxial muscle mass, the tip of the irrigator over the hemilaminectomy site, and the drill itself. Note how the surgeon is holding the drill, steadying the bur guard with the other hand to resist excessive downward force. One hand should also rest gently on the dog, taking care not to compromise ventilation. The articular facets have been removed.
8.28
8.29
8.30 Bone debris can be irrigated away while drilling by using a jet of water as shown (arrow). Alternatively, a slow drip of saline can be used to cool the bur with periodic flushing of debris. Here the surgeon has repositioned a finger lower down the bur guard to improve fine motor control.
8.30
8.31 Bleeding from cancellous bone is controlled using bone wax. Note that cancellous bone is visible over both pedicles but not centrally over the inner articular facet joint (arrowheads). The surgeon will not encounter any cancellous bone at the level of the joint prior to entering the vertebral canal.
8.31
141
142
Small Animal Spinal Disorders
8.32 Bone has been removed more extensively, revealing the gray/pink cancellous bone (arrows) of the caudal vertebral body. Drilling is advanced further over the cranial vertebral body, where cancellous bone has been removed to reveal the inner cortical bone plate (asterisk). Note the seam of cortical bone at the facet joint (arrowheads).
8.32
8.33 The inner cortical bone plate of each vertebral body has been exposed in this dog (arrows). The next stage is to drill to eggshell thickness over the entire hemilaminectomy defect.
8.33
In order to remove the inner cortical bone, the surgeon makes fine movements of the bur as shown in 8.34. Holding the drill close to the remaining plate of bone the surgeon makes fine movements, approximately 1 mm in excursion, until the bur is felt to bite and cut away fragments of bone. The softness of the inner plate should be tested often using the bur as a probe but with the power off. This method of drilling is continued until the inner plate yields to gentle pressure, at which point it can be broken down with fine instruments. The surgeon should again note that there is a seam of cortical bone at the level of the articular facet joint that unites the outer and inner plates with no intervening cancellous bone.
8.34 In this diagram, the surgeon has started drilling for a mini-hemilaminectomy. The aim should be to perform the least invasive surgery possible for the distribution of disc material (see page 126). When feasible, start with a pediculectomy or mini-hemilaminectomy and either combine the two or extend subsequently to a standard hemilaminectomy if necessary (8.57, 8.62).
8.34
Thoracolumbar disc disease
8.35 Once through the inner cortical plate, the surgeon may want to palpate the floor of the canal with instruments to clarify exactly how the defect needs to be enlarged.
8.35
8.36 Here the thinned, inner cortical plate has been broken. All bone should be removed from within the defect before starting to remove disc material. It is easier to drill all remaining bone away when disc material still covers the spinal cord rather than once the cord has moved back to the edges of the defect.
8.36
8.37 Extruded disc is often mixed with old hemorrhage; it is still covered here by endosteum (arrows). The surgeon is using a 8.37 House curette (5.16) to clear bone from around B A the entire circumference of the bone window. All force applied to this instrument is directed away from the spinal cord.
143
144
Small Animal Spinal Disorders
8.38 A: After all bone has been removed, a thin endosteum (arrow) often remains over the extruded disc material and dura 8.38 (8.37). B: The endosteum must be broken down A B before disc material can be accessed. Dura (arrow) is usually much whiter than endosteum. Endosteum is thinner than dura and tears easily. 8.44 shows the same dog after decompression.
8.39 Instruments are then used to tease the extruded disc material (arrows) away from the dura mater. The tough, medial portion of joint capsule (arrowhead) 8.39 may tear or, in bigger dogs, A it may need to be cut.
B
C
8.40 A space is visible (vertical line) between dura and bone when significant disc material remains. The gray mass under the spinal cord is disc material (arrowhead); it is separating the dura from the floor of the vertebral canal. 8.62A is from the same dog and shows how the dura returns to the floor of the canal once all disc material has been evacuated.
D
8.40
Thoracolumbar disc disease
8.41 Gentle exploration beneath the spinal cord with a blunt probe shown in three successive images. This permits ventrally located material to be retrieved even from the opposite side of the vertebral canal, as shown. The suction tip should not touch the dura; a soft, red rubber catheter tip should be attached to the metal tip when in close proximity to spinal cord. 8.41 A
B
C
8.42 CT reconstruction of a hemilaminectomy defect to show the idealized motion for retrieving material from underneath the spinal cord. A long, circular scooping motion retrieves material better than more linear movements that often just serve to push material farther away. The venous plexus (‘venous sinus’) may be lacerated accidentally at this stage.
8.42
145
146
Small Animal Spinal Disorders
8.43 Control of venous hemorrhage, often arising from the junction of the spinal vein and the venous plexus, using a small piece of muscle tissue (arrow) cut from the adjacent epaxial musculature. The muscle is macerated then pressed firmly over the vein; it should be removed once decompression is finished. Alternatives include Gelfoam (Pharmacia, Kalamazoo, MI), direct pressure on either side of the tear, or a small Hemoclip (Pilling Weck Inc., Research Triangle Park, NC) for a tear in a spinal vein.
8.44 When all disc material has been removed the dura returns to lie adjacent to the floor and pedicle of the vertebral canal; the spinal cord may remain indented as here (same dog as in 8.38). Other clues to the adequacy of decompression are reappearance of epidural fat at each end of the bony defect along with pulsation of the dura (caused by CSF pulsing in synchrony with venous flow).
8.43
8.44
When no disc material can be found, it may be that: •
The surgeon has approached at the wrong site (8.6).
•
The surgeon has approached on the wrong side.
•
There has been a small volume, high-energy disc herniation causing an impact injury.
•
The dog could have progressive myelomalacia.
•
The original diagnosis could be wrong.
If landmarks confirm that the site is correct (8.20), the defect must be extended in whichever direction the spinal cord appears most compressed or swollen (8.56, 8.61A). If the spinal cord still appears compressed after removal of all visible disc material then it is possible that more material exists on the far side of the spinal cord (8.3, 8.6). The defect may then have to be extended either to one side (8.45), dorsally (8.46) or bilaterally (Schulz et al., 1998).
8.45 To extend the hemilaminectomy, the surgeon should start drilling a short distance away from the existing opening. The two should then be merged to form one larger opening by breaking the intervening bone bridge (arrow). This is safer than trying to drill outwards from the edge of the original hemilaminectomy defect as the drill bit tends to slip off the edge and could injure the spinal cord.
8.45
Thoracolumbar disc disease
8.46 Drilling can continue dorsally if disc material lies on the opposite side of the vertebral canal. Excessive bone removal could cause delayed injury from restrictive fibrosis; to avoid this the cancellous and inner cortical bone can sometimes be removed while sparing the outer plate. If the disc material is still not visible, it is safer to start a mini-hemilaminectomy on the contralateral side. 8.46
8.47 A: This hemilaminectomy has been extended ventrally to expose the spinal nerve and ganglion (arrow). Bleeding from the spinal artery may require bipolar cautery. The venous plexus is just visible on the floor of the vertebral canal (arrowhead); it is damaged easily, especially where the spinal vein joins the plexus (8.43). B: Lateral corpectomy (arrowheads) of one or two vertebral bodies improves ventral access further, especially to chronic disc herniations (Moissonnier et al., 2002). Drilling should proceed under the cortical bone plate that forms the floor of the canal before collapsing the cortical plate ventrally along with the disc. The area of the corpectomy is occupied by air; a large air pocket also lies adjacent to the fat graft placed over the mini-hemilaminectomy site.
8.47 A
B 8.48
8.48 Access to the vertebral canal varies with the location. Hemilaminectomy at A: lumbar vertebrae (L2) often gives better access than at B: thoracic vertebrae (T10/11) due mainly to the rib heads. A
B
147
148
Small Animal Spinal Disorders
8.49 A durotomy has been performed using a 25-gauge hypodermic needle to reveal the spinal cord, small pial blood vessels and dorsal nerve rootlets (arrow). A guarded scalpel or fine scissors may also be used to incise the dura. Collapse of the subarachnoid space has caused the dura mater to become translucent (8.5, 8.50). The venous plexus is visible clearly on the floor of the vertebral canal (arrowhead).
8.49
Although durotomy may provide additional spinal cord decompression, the relative risks and benefits are not clear (see page 87): •
Local necrosis of gray matter over one or two spinal cord segments is common after severe injury and it is not unusual for this material to ooze out after durotomy (8.50).
•
This does not necessarily indicate a poor prognosis, such as with complete spinal cord transection, or that there is progressive myelomalacia (Salisbury and Cook, 1988).
•
It is only necessary for 5–10% of the white matter to survive in order for an animal to recover; this would be very hard to quantify after durotomy (Blight and Decrescito, 1986; Basso et al., 1996; Jeffery and Blakemore, 1999; Olby et al., 2003) (see ‘Prognosis’, page 132).
•
Durotomy should therefore not be used to infer prognosis (Muir et al., 1995); its use purely as an additional decompressive procedure may warrant further study (Anderson et al., 1991; Bagley et al., 1996).
•
Routine use of durotomy is not yet recommended in dogs with grade 5 deficits, particularly as more than half of these dogs recover without one (Muir et al., 1995) (see ‘Prognosis’, page 132).
•
There is a small risk of spinal cord herniation through the durotomy defect, which can cause severe distortion and compression (Osterholm, 1974; Henry et al., 1997).
8.50 Minihemilaminectomy and durotomy (arrowheads) performed in a 25 kg dog presenting with no deep pain sensation. There is extensive bruising with focal malacia (white arrow) of the spinal cord. This dog regained nociception three days after surgery and made a good recovery within six weeks of surgery. The facet joint is shown by an asterisk. 8.50
8.51 Nerve roots can cause significant tethering of spinal cord over the disc. This can be relieved by rhizotomy (arrow) except at the lumbosacral plexus. Great care must be taken when removing a chronic disc like this (see ‘Early postoperative complications’, page 127). If a plane of dissection can be developed between dura and adhesions, the extrusion can then be cut away by scalpel. Corpectomy facilitates this (8.47B). Otherwise the spinal cord is just decompressed as shown without prolonged attempts at removal.
8.51
Thoracolumbar disc disease
Rhizotomy improved the degree of decompression although it is not known if cutting the associated segmental blood supply has any adverse effect on perfusion at that level.
8.52 Diagram to show the relationship of the spinal nerve, the completed hemilaminectomy defect and the intervertebral disc space. These relationships should be understood clearly prior to fenestration (8.53). The surgeon must also take care not to injure the exposed spinal cord inadvertently. 8.52
8.53 Fenestration is usually performed after hemilaminectomy. If the surgeon is not familiar with fenestration, then it is best to fenestrate the hemilaminectomy site first as shown here (arrow), in order to understand the relative positions of the disc and spinal cord. In practice the spinal nerve is seen only rarely. Here a scalpel blade has been used to make a deep cut into the anulus (8.70, 8.78). It is not clear if a chronic disc should be fenestrated (page 126). 8.53
8.54 Prior to closure, the hemilaminectomy site is covered by a 3–5 mm thick layer of subcutaneous fat demonstrated here lying over a cloth drape. The graft shrinks to 50% of its original size and so should be considerably larger than the laminectomy defect (Trevor et al., 1991). It is held in place by the epaxial musculature after retractors are removed; alternatively saline-soaked Gelfoam (Pharmacia, Kalamazoo, MI) can be used. The fat graft must not be too thick (8.8, 12.10). 8.54
149
150
Small Animal Spinal Disorders
8.55 3D CT reconstruction showing the postoperative appearance of a standard hemilaminectomy. This image, along with several other reconstructions (8.48, 8.56, 8.60–8.62), shows postoperative laminectomies from several dogs. This scan was performed in a dog that lost deep pain 2 days after a hemilaminectomy at L1/2; no residual disc material was identified and the dog had progressive myelomalacia at necropsy.
8.55
Decompression may be necessary over three or four vertebrae if: •
The extrusion cannot be found (8.61).
•
Disc material is spread over a wide area.
•
The spinal cord is very swollen.
Extensive hemilaminectomy does not appear to affect recovery rates but movement at the surgical site is increased (Corse et al., 2002); this may cause delayed morbidity in some dogs (Anderson et al., 1991; Grevel and Schwartau, 1997; Applewhite et al., 1999; Scott and McKee, 1999).
8.56
8.56 This dog had CT for recurrent back pain 4 years after hemilaminectomy from T13 to L3. No cause was identified; the pain may have been related to laminectomy scar (Applewhite et al., 1999). There is a direct correlation in humans between persistent low back pain and extensive peridural scarring (Ross et al., 1996; Maroon et al., 1999). Where possible, less extensive decompression should be considered (8.57–8.62).
Thoracolumbar disc disease
Pediculectomy and mini-hemilaminectomy (8.57–8.62) Approaches include lateral (8.58–8.59) or standard dorsolateral (8.11–8.27) muscle separation. The lateral approach only works well for lean, small dogs. With a standard dorsolateral approach the dog should be rotated away from the surgeon in order to perform these surgeries as this makes it much easier to gain access under the facet joint (Bitetto and Thacher, 1987).
8.57 Difference between A: hemilaminectomy, B: mini-hemilaminectomy, and C: pediculectomy. Hemilaminectomy sacrifices a facet joint; the other procedures do not. Mini-hemilaminectomy enlarges the region around the foramen. Pediculectomy spares the foramen but removes bone between adjacent facet joints. As access is restricted with mini-hemilaminectomy (8.59B) and pediculectomy (8.61), it is often better to combine them (8.62). 8.57 A
B
C
151
152
Small Animal Spinal Disorders
8.58 Transverse section through the lumbar region to show the lateral approach for mini-hemilaminectomy or pediculectomy. The transverse processes are palpated and then the epaxial muscles are elevated dorsally to reveal the lateral aspect of the vertebrae (Braund et al., 1976) (8.59, 8.63–8.65). A curette is shown retrieving extruded nucleus pulposus following a mini-hemilaminectomy.
8.58
8.59 A: Army–Navy retractors elevate the epaxial muscles and expose the vertebrae laterally (articular processes are obscured by the muscle and retractors). B: Mini-hemilaminectomy provides good access to the mid-ventral vertebral canal. The floor of the vertebral canal is obscured here by a large, chronic disc (arrow) extruding dorsolaterally.
8.59 A
B
Thoracolumbar disc disease
8.60 Pediculectomy via a standard dorsolateral approach (8.11–8.27). A: Intraoperative photograph, and B: 3D reconstruction of a postoperative CT scan from the same dog. Pediculectomy gives good access over the mid-vertebral body and preserves the articular processes. Access to the intervertebral space is poor compared to mini- and standard hemilaminectomy (8.61B).
8.60 A
B
8.61 A: When imaging is inconclusive, multiple pediculectomies (arrows) can be used to search for disc, hemorrhagic fat, or spinal cord swelling; the defect can be enlarged once the correct location is found. Although not ideal, this is often quicker than extensive laminectomy (8.56). B: A disadvantage of pediculectomy is that material can be missed easily. Mid-sagittal 3D reconstruction through the vertebral canal of a dog where disc material (arrow) was missed after pediculectomy (same dog as 1.21A, 8.60).
8.61 A
B
153
154
Small Animal Spinal Disorders
8.62 A mini-hemilaminectomy has been combined with two pediculectomies to give much better access than either procedure alone (Biggart, 1988; Lubbe et al., 1994; McCartney, 1997). A: Intraoperative photograph, and B: a 3D reconstruction of a postoperative CT scan from the same dog as shown in 8.40. The dorsolateral approach used here makes it easier to extend the bony defect and is therefore preferred over the lateral approach shown in 8.58 and 8.59.
8.62 A
B
Lateral fenestration (8.63–8.78) The lateral approach for fenestration can be made from either side but is normally undertaken from the left side. The depth of the epaxial muscle makes the approach progressively more challenging in dogs greater than 10 kg bodyweight.
8.63 Positioning and skin incision for lateral fenestration. Thoracic limbs are tied forwards with the pelvic limbs extended backwards. A thin sandbag is placed under the dog at the level of the thoracolumbar junction to open up the disc spaces of 8.63 interest; this may be moved as required to approach the lumbar discs. The skin incision is made at the level of the lumbar transverse processes and extends from approximately T8 to L5.
8.64 The thick, shiny, lumbodorsal fascia lies under the subcutaneous fat and fascia. The subcutaneous fat is reflected for 1 cm on either side of the proposed incision. The lumbodorsal fascia is then incised as shown. Reflection of fat creates additional dead space but a wide exposure makes it easier to repair the fascia later. 8.64
Thoracolumbar disc disease
8.65 The deep layer of fat, which can be substantial even in lean dogs, has been incised to reveal the iliocostalis lumborum and the thirteenth rib (arrow). The longissimus dorsi muscle is covered by a fascial sheath, which is just visible under the layer of fat (arrowhead). Exposure dorsally is more than is usually needed but has been made to show the longissimus muscle. 8.65
8.66 Diagram to show the deep anatomy and muscle separation through the iliocostalis lumborum muscle, in this case over the L1/2 intervertebral space. The longissimus muscle is in the dorsocaudal part of the surgical field (arrow). The iliocostalis muscle fibers are 8.66 seen running obliquely (arrowhead) to insert on the ribs.
8.67 Close up to show separation of the iliocostalis lumborum muscle over the T13/L1 disc by opening a pair of Metzenbaum scissors in the same direction as the muscle fibers. The thirteenth rib (with periosteum reflected for clarity) is shown clearly in the lower left-hand side of the picture. 8.67
8.68 Retraction of the iliocostalis muscle. This reveals the body (*) and transverse process (arrowhead) of the L1 vertebra along with the T13/L1 disc (arrow). Note the fibers of the anulus fibrosus (over which lies a fine layer of connective tissue). This is best removed by using a periosteal elevator covered with a surgical swab (8.23), pushing the tissue in a craniodorsal direction. 8.68
155
156
Small Animal Spinal Disorders
8.69 Close up diagram of the site approached in 8.68. Muscle separation has revealed the lateral aspect of the anulus fibrosus, which lies just cranial to the transverse process of the lumbar vertebra. Note the small vein that lies over the craniodorsal anulus; this is retracted as described in 8.68. The incision in the anulus is shown by the dotted line. Note also the accessory process, which marks the dorsal margin of the intervertebral foramen (arrow). The exact dorsal limit for the fenestration varies slightly depending on the disc space and also from dog to dog. A skeleton is a useful reference (5.4). A blunt periosteal probe can be used to palpate the curving lateral surface of each disc before fenestrating it. A fresh approach is made for each disc space.
8.69
8.70 Incising the disc with a #11 scalpel blade reveals the window of anulus and the jelly-like nucleus pulposus (arrow); angle same as in 8.68.
8.70
Thoracolumbar disc disease
To fenestrate a disc: •
The window is made by four separate stab incisions, which are joined at the corners.
•
A sawing motion is used to cut instead of direct pressure.
•
The hole in the anulus must be made larger than any instrument used for removal of nucleus pulposus.
•
Disc removal is by small curette, a Rosen mobilizer or, failing that, by twisting the #11 blade.
•
It is important to remove as much nuclear material as possible, as any left will remain in the intervertebral space and could cause a clinical problem later.
•
Power fenestration with a small drill bit may allow more complete evacuation (Holmberg et al., 1990). Care must be taken to prevent the drill bit from slipping toward the foramen.
8.71 The approach to thoracic discs is different. They may be approached by separating the iliocostalis lumborum muscle as for lumbar discs. Alternatively, the muscle can be cut close to its insertions on the 12th and 13th ribs.
8.71
8.72 Retraction of the iliocostalis lumborum muscle dorsally reveals the levator costarum muscle (arrow). This will be separated from the rib and retracted in a cranial direction.
8.72
8.73 The periosteum of the rib has been incised at the caudal border of the levator costarum muscle and elevated using a thin periosteal elevator. Periosteum is elevated adjacent to the neck of the rib, a small retractor is being used to retract the levator costarum muscle cranially. The plane of dissection is medial and dorsal to the pleural reflection. Dissection of the deep fascia that attaches to the cranial margin of the rib (arrow) is then continued proximally, ventral to the head of the rib.
8.73
157
158
Small Animal Spinal Disorders
8.74 Diagram to show the features described in 8.73. The levator costarum muscle is being elevated from the rib with a periosteal elevator. A hand-held retractor is positioned to keep the iliocostalis muscles retracted dorsally. Gelpi retractors can be used here but they must not tear the pleura. A change in respiratory character or volume indicates possible pneumothorax. 8.74
8.75 Diagram to show the lateral aspect of the disc exposed, with the levator costarum muscle retracted cranioventrally, and the epaxial muscles retracted dorsally. The site of the fenestration is shown.
8.75
Thoracolumbar disc disease
8.76 Blunt dissection has been completed and the disc is now visible clearly (*). Handheld retractors are being used to elevate the epaxial muscle away from the last two ribs.
8.76
8.77 A #11 scalpel blade cutting deeply into the anulus. The fenestration is then completed as described in 8.70. The surgeon should insure that the window in the anulus is slightly larger than the instrument to be used for curettage or material might be forced inadvertently into the vertebral canal. 8.77
8.78 Diagram of a transverse section through the lumbar region of a dog to show the lateral approach for fenestration. This approach allows access to the lateral anulus fibrosus and to disc material in the intervertebral space. However, there is no access to disc material in the vertebral canal unless a pediculectomy or mini-hemilaminectomy is also performed (8.57–8.62).
8.78
159
Atlantoaxial subluxation
Clinical signs
161
Diagnosis 163 Examination 163 Differential diagnosis Radiography 163
163
Treatment 164 Non-surgical treatment Surgery 165 Complications 167 Non-surgical treatment Ventral fusion 167 Dorsal fixation 168 Postoperative care Prognosis
164
167
169
169
Atlantoaxial subluxation in cats
169
Key issues for future investigation References
170
170
Procedures 171 Ventral transarticular fixation 171 Multiple ventral implants and bone cement Dorsal wire fixation 178 Dorsal cross-pin fixation 180
177
The atlantoaxial joint allows rotation of the head; C1 pivots around the dens of C2 but the joint permits little flexion. There is no intervertebral disc between C1 and C2 (4.8, 4.9) and the relationship between these vertebrae is maintained largely by ligaments (1.14, 1.35). Clinical signs in congenital atlantoaxial subluxation are usually seen in immature patients although signs can develop at any age (Thomas et al., 1991; McCarthy et al., 1995; Beaver et al., 2000) (9.1, 9.32). The disorder is encountered most often in small breeds of dog,
Chapter
9
particularly Yorkshire terriers, Chihuahuas and Miniature poodles (Denny et al., 1988; Thomas et al., 1991; McCarthy et al., 1995; Beaver et al., 2000). Rare cases occur in cats and in large-breed dogs (9.4). Atlantoaxial subluxation can cause clinical signs in breeds such as the Rottweiler (Rochat and Shores, 1999; Wheeler, 1992), Doberman (Huibregtse et al., 1992; LeCouteur and Child, 1995), Basset hound (Hurov, 1979), Standard poodle (Knipe et al., 2002), Weimaraner, and German shepherd dog (Read et al., 1987). A number of pathological processes can lead to atlantoaxial subluxation: • Absence of the dens (9.2B). • Fracture or separation of the dens (9.2C). • Failure of the ligaments due to either malformation or rupture (9.2D). Most dogs with congenital lesions have either absence or hypoplasia of the dens (46%; 9.4), 30% have a malformed dens (9.1B), and 24% have a normal dens (Beaver et al., 2000). In dogs with a normal or malformed dens, abnormalities of the transverse ligament of the atlas can lead to subluxation (Watson et al., 1989). This is serious as the dens tends to protrude dorsally into the spinal cord (9.1B, 9.2D, 9.19B).Occipito-atlantoaxial malformation can also occur (Read et al., 1987). Most patients have an underlying congenital abnormality but trauma can cause failure of normal elements in this region (9.2C) (see Chapter 13). Minor trauma may also precipitate a crisis in a dog that has a congenital abnormality but has shown no clinical signs previously (Thomas et al., 1991; Beaver et al., 2000).
CLINICAL SIGNS Neck pain is seen in most dogs following traumatic lesions and in 30–60% of dogs with congenital lesions (Thomas et al., 1991; Beaver et al., 2000). Neurological signs reflect cervical spinal cord compression. In mild cases only proprioceptive deficits are seen. Tetraparesis indicates more significant spinal cord compression. Asymmetry of signs, or preferential involvement of
162
Small Animal Spinal Disorders
either thoracic or pelvic limbs, may occur. Tetraplegia is rarely encountered but if present the dog must be checked for respiratory failure (Beaver et al., 2000) (see Chapter 2). Signs referable to brain involvement have
A
also been reported and should be evaluated carefully. Hydrocephalus has been reported in dogs with atlantoaxial subluxation and could cause the animal to show forebrain signs (Chambers et al., 1977; Denny
B
9.1 3D reconstruction of CT scans; the dens is indicated by arrows. A: A normal toy-breed dog. B: A dog with atlantoaxial subluxation and a malformed dens, which occupies much of the vertebral canal. The spinous process of C2 is also angled away from the arch of the axis (arrowhead). Two-dimensional images of these dogs are shown in 9.6; sagittal 3D reconstructions in 9.19; postoperative radiographs for dog B in 9.32.
A
B
C
D
9.2 A: Diagram to show the normal relationship between C1 and C2 (1.14, 1.35, 9.19). B: Congenital absence or hypoplasia of the dens is the most common abnormality. C: There is an ossification center between the dens and the body of C2, which predisposes to fracture or separation. D: Rupture or malformation of the ligaments.
Atlantoaxial subluxation
et al., 1988). Hydrocephalus may also be accompanied by syringohydromyelia (9.12). Another potential explanation for signs of forebrain disease in dogs with atlantoaxial subluxation is hepatic encephalopathy, which is over-represented in toy-breed dogs and coexisted in two of six dogs undergoing surgery for atlantoaxial subluxation (Schulz et al., 1997). Forebrain signs of disorientation and behavior change, along with vestibular deficits, have also been associated with basilar artery compression caused by the dens. The clinical signs resolved completely after surgery (Jaggy et al., 1991). Torticollis has been described with atlantoaxial lesions and could be due to syringohydromyelia or a vestibular sign secondary to a high cervical lesion (Johnson and Hulse, 1989; Gibson et al., 1995; Mayhew, 1999) (see page 29).
DIAGNOSIS Examination Atlantoaxial subluxation should be considered in any young, small-breed dog with the clinical signs described. Neurological examination indicates a lesion between C1 and C5 and palpation of the neck often localizes the origin of the pain to the C1–C2 region. It is unwise to flex the neck forcibly in a patient where atlantoaxial
Box 9.1 Primary differential diagnoses for atlantoaxial subluxation (see also Box 7.2) ■
Cervical disc extrusion (older than 1 year)
■
Syringohydromyelia
■
Neoplasia
■
Inflammatory CNS disease
■
Intracranial disease
■
Discospondylitis
■
Polyarthritis
■
Polymyositis
■
Trauma
subluxation is suspected, as this may worsen the situation considerably.
Differential diagnosis The differential diagnosis for a small-breed dog with cranial cervical signs is given in Box 9.1. Inflammatory CNS disease is the most likely consideration in immature dogs. Cervical disc disease is more likely in older dogs but is rare in dogs less than 2 years of age. Discospondylitis or fractures could be present at any age. Atlantoaxial subluxation is encountered in cats rarely (see page 169).
Radiography SURVEY RADIOGRAPHY Survey radiographs provide the diagnosis in most cases (9.3). General anesthesia is usually required, although great care must be taken when intubating the patient. If non-surgical management is to be used following trauma then it may be worth trying to obtain diagnostic images without anesthesia (see page 283). However, accurate positioning is essential to evaluate the cranial cervical region and this may be impossible in the conscious patient, particularly if severe neck pain is present (see page 46). It is a common error to diagnose congenital atlantoaxial subluxation on radiographs of conscious dogs where the positioning is inadequate and the region of interest is far from the center of the film. Fluoroscopic observation while flexing the neck gently in a conscious animal can provide rapid and accurate diagnosis by revealing the dynamic nature of the lesion while allowing the animal to maintain some protective muscle tone. The lateral projection is the most useful. Mild flexion of the cranial cervical region may be required to demonstrate misalignment but this must not be excessive. A ventrodorsal view will highlight the dens; it is safe to position the dog in dorsal recumbency with the neck extended for the ventrodorsal projection, which is preferable to the open-mouth view. Oblique radiographs can also provide an excellent image of the dens (9.3B, 9.4B) (Cook and Oliver, 1981).
9.3 Survey radiographs from a 1-year-old Toy poodle showing a marked increase in the gap between the dorsal arch of C1 and the spinous process of C2 when the neck is flexed (B). Postoperative radiographs are shown in 9.8.
A
B
163
164
Small Animal Spinal Disorders
A
B
9.4 Eight year-old Bull mastiff with acute pelvic limb ataxia. Survey radiographs revealed severe hypoplasia of the dens; a lumbar myelogram revealed no other abnormality. A: Flexed view and B: extended view of C1 and C2. Note the narrowing of the dorsal subarachnoid space and the hypoplastic dens (arrowhead). Neurological deficits worsened after the myelogram but the dog made a good recovery and had no neurological deficits 4 months later. Subclinical atlantoaxial malformation has been reported in humans (McKeever, 1968).
9.5 Myelogram in a 4-year-old Toy poodle demonstrates that there may be little spinal cord compression if the dens is absent or hypoplastic. Postoperative radiographs of this dog are shown in 9.29.
A
B 9.6 A: CT scan through the atlas of a normal toy-breed dog. The dens is of normal size and occupies the floor of the vertebral canal. B: Dog with a malformed dens. The same dogs are also shown in 9.1 and 9.19.
A
B
MYELOGRAPHY Myelography should not be necessary for diagnosis and any post-myelographic seizures could be disastrous. Cerebello-medullary cistern puncture either for myelography or CSF sampling should not be performed in dogs that could have atlantoaxial subluxation (9.5); lumbar puncture is preferable.
CT AND MRI Although much simpler techniques are available to assess atlantoaxial stability, advanced imaging can add important preoperative information (Johnson and Hulse, 1989) (9.6, 9.12). CT provides excellent bone
imaging, which assists surgical planning (Johnson and Hulse, 1989) (9.33). It will also reveal abnormal dens conformation, which is seen in over 70% of dogs (Denny et al., 1988; Beaver et al., 2000) (9.6). An MRI (9.7) might prove to be prognostic if there is extensive spinal cord malacia and it will also reveal syringohydromyelia (Sanders et al., 2000) (9.12).
TREATMENT Non-surgical treatment Non-surgical treatment entails cage rest, application of a neck brace (13.18) and use of analgesics (Tables 15.1,
Atlantoaxial subluxation
B
A
9.7 Sagittal MRIs of a dog with atlantoaxial subluxation. A: T2-weighted and B: short tau inversion recovery (STIR) images. Note the high signal intensity (edema) and severe spinal cord compression at C1–C2 (arrowheads).
Algorithm 9.1 Surgical decision-making in atlantoaxial subluxation.
Survey radiographs
Malformed dens
Normal or hypoplastic dens
Consider odontoidectomy
Fusion with multiple implants and cement
Failure
Success
CT scan
Dorsal cross pinning
External splint
Replace implant
15.2). This approach can produce surprisingly good results. Six toy-breed dogs were managed non-surgically, four of which had been unable to walk. After 14 weeks all six dogs could walk without neurological deficits (Hawthorne et al., 1998). Four out of another six dogs also did well with non-surgical management (Lorinson et al., 1998). Despite these excellent results after nonsurgical management, the concern is that any improvement might be lost after brace removal and return to normal activity (9.32). Nevertheless, it can provide a very useful alternative for very young dogs, for dogs that cannot walk, or for other high-risk patients. Surgical management should be recommended for congenital lesions once the neurological deficits improve or the dog
is large enough, at least until the long-term outcome of non-surgical management is known. Non-surgical treatment does give excellent long-term outcomes for animals that fracture a normal atlantoaxial articulation; results are often superior to surgery (see page 295).
Surgery Surgical treatment is indicated in most patients with congenital lesions. Even dogs with profound neurological deficits are likely to benefit from stabilization (Thomas et al., 1991; Beaver et al., 2000). The two main options are either ventral fusion or dorsal stabilization. Ventral fusion is the treatment of choice. An algorithm for surgical decision-making is shown in Algorithm 9.1.
165
166
Small Animal Spinal Disorders
9.8 Reduction after transarticular screw fixation is good. Note the hypoplastic dens. Same dog as in 9.3. Angles are mediolateral: left 41° and right 38°; ventrodorsal 33° (9.25).
R
A
B
Table 9.1 Results of ventral fixation for atlantoaxial subluxation Transarticular fixation
Multiple implant fixation
(1)
(2)
(6)
Total transarticular
(5)
(4)
(3)
Total multiple
Success rate after 1st surgery (%)
31/40 (78)
9/10 (90)
8/18 (44)
48/68 (71)
5/6 (83)
5/6 (83)
12/13 (92)
22/25 (88)
Median follow-up months
7
12 (mean)
N/A
21
N/A
36
Residual ataxia (%)
6/31 (19)
N/A
N/A
6/31 (19)
0/5
N/A
4/13 (31)
4/18 (22)
Residual neck pain (%)
3/31 (10)
N/A
N/A
3/31 (10)
0/5
N/A
0/13
0/18
Second surgery (%)
4/35 (11)
0/9
3/18 (17)*
7/62 (11)
0/5
1/5** (20) 0/13
1/23 (4)
Mortality (%)
5/40 (13)
1/10 (10)
7/18 (39)
13/68 (19)
1/6 (17)
0/6
2/25 (8)
1/13 (8)
(1) Beaver et al., 2000; (2) Denny et al., 1988; (3) Knipe et al., 2002; (4) Sanders et al., 2000; (5) Schulz et al., 1997; (6) Thomas et al., 1991. * Two of these recovered. ** Signs recurred but did not actually undergo second surgery. N/A, not available.
VENTRAL FUSION Fusion of the atlantoaxial joints can be performed using either transarticular fixation (9.8) or using multiple implants and bone cement (9.32–9.34). Good results can be obtained using transarticular fixation; threaded pins or screws give superior results to smooth pins. However, there is little margin for error with only two implants and overall failure rates approach 30% with transarticular fixation (Table 9.1). Furthermore, radiographic evidence suggests that fusion is often delayed or incomplete (Thomson and Read, 1996; Beaver et al., 2000). Multiple implant techniques should reduce the failure rate; they add little technical difficulty and almost certainly provide more rigid fixation (Schulz et al., 1997; Sanders et al., 2000; Knipe et al., 2002) (9.32–9.34). Use of multiple
implants is therefore the technique of choice for ventral fixation. Following ventral fixation, odontoidectomy may rarely be indicated via a ventral slot in C1 (9.6, 9.27, 9.32).
DORSAL STABILIZATION Dorsal wiring has a high failure rate (Table 9.2), such that many animals need repeat surgery (9.11). Reinforcement with bone cement may be useful but any thermal necrosis could weaken the bone further (Renegar and Stoll, 1979; Cook and Oliver, 1981; Martinez et al., 1997). It does provide an option should ventral fusion fail (Beaver et al., 2000), but a second attempt at fusion using multiple implants or by dorsal cross-pinning are more logical approaches (Thomas et al., 1991; Jeffery, 1996; Schulz et al., 1997; Sanders et al., 2000).
Atlantoaxial subluxation
Table 9.2 Results of dorsal suture fixation for atlantoaxial subluxation Dorsal wire or suture fixation (1)
(2)
(3)
(4)
Total dorsal
Success rate after 1st surgery (%)
9/12 (75)
6/6 (100)
7/13 (54)
2/7 (29)
24/38 (63)
Median follow-up months
23
9
N/A
N/A
Residual ataxia (%)
4/9 (44)
5/6 (83)
N/A
N/A
9/15 (60)
Residual neck pain (%)
1/9 (11)
0/6
N/A
N/A
1/15 (6)
Second surgery (%)
2/11 (18)
2/6 (33)
N/A
3/7 (43)*
7/24 (29)
Mortality (%)
1/12 (8)
0/6
5/13 (38)
0/7
6/38 (16)
(1) Beaver et al., 2000; (2) Chambers et al., 1977; (3) Denny et al., 1988; (4) Thomas et al., 1991. * One dog recovered.
COMPLICATIONS The most common non-surgical and surgical complications are listed in Table 9.3.
Non-surgical treatment The splinted animal must initially be assessed daily for problems (Table 9.3), including the tendency for moisture to seep into the splint around the mouth. Dyspnea can occur if the splint is too restrictive in the pharyngeal or thoracic regions but it must also be tight enough to stay in place (Schulz et al., 1997). Aspiration is a particular problem if the neck is extended excessively as it is almost impossible for the dog to swallow in this position. Nasogastric or gastrostomy feeding may reduce the risk of aspiration and also helps to keep the splint clean (Shelton et al., 1991).
Ventral fusion The ventral approach to the neck is straightforward but care must be taken to avoid damage to vital structures, particularly the recurrent laryngeal nerve and the vascular supply to the thyroid gland. Particular care must be taken to prevent excessive traction or compression on the trachea and esophagus (Thomas et al., 1991). The main complications are death, or implant failure that necessitates a second surgery (Table 9.3). The primary causes of death are cardiac or respiratory arrest and pulmonary edema (Thomas et al., 1991; Schulz et al., 1997; Beaver et al., 2000). One potential cause of pulmonary edema is barotrauma; care must be taken when ventilating toy-breed dogs (see page 85). Implant failure is more likely after ventral transarticular fixation than after use of multiple implants. The main
Table 9.3 Complications of non-surgical and surgical treatment Non-surgical complications
Surgical complications
Dyspnea Aspiration Abrasions Decubitus Otitis externa Dermatitis Recurrence
Cardiac arrest Respiratory arrest Pulmonary edema Implant failure (9.11, 9.31) Soft tissue injury Tracheal collapse
causes of implant failure are K-wire migration or loss of reduction; these tend to occur in the first 3 weeks (Johnson and Hulse, 1989; Thomas et al., 1991; Schulz et al., 1997; Beaver et al., 2000). Implant failure in some dogs may not cause clinical signs if preceded by sufficient fusion (Wheeler, 1992). Threaded pins are preferred over Kirschner wires as they are much less likely to migrate and have greater pullout strength (Johnson and Hulse, 1989; Sandman et al., 2001). In one dog a K wire migrated into the oral cavity although it caused no clinical signs other than a cough (Schulz et al., 1997) (9.9). The disadvantage of using only two implants in transarticular fashion is the relatively high failure rate (Table 9.1). Failure rates should be lower when screws or threaded pins are used instead of K wires but failure can still occur even when implant position is excellent (9.10). In ventral fusion the implants are on the compression side of the vertebral column and so are subject to greater stresses than implants on the dorsal (tension) side (Jeffery, 1996; Rochat and Shores, 1999). Even
167
168
Small Animal Spinal Disorders
L
A
B
9.9 A: Postoperative radiographs from a 1-year-old Toy poodle show K-wires with bone cement over the exposed ends. Angles are mediolateral: left 53° and right 55°; ventrodorsal angle is 25–30°. B: Failure occurred as one implant is anchored only poorly in C1. An external splint was applied but both pins had failed by 5 weeks. The splint was maintained for 12 weeks in all; the dog was pain-free with no deficits 4 years later apart from a lump on its neck from a migrated pin. A metallic skin staple is visible.
L
A
L
9.10 A: Transarticular fixation using lag screws and cancellous bone graft; postoperative alignment is excellent. Angles are mediolateral: left 22° and right 45°. B: The dog was also placed in a splint but despite this one screw was loose at the 6-week recheck and there was no bridging callus. After a further 6 weeks in the splint good fusion was evident and the dog made an excellent recovery.
B
when failure does not occur, fusion is often delayed (Sorjonen and Shires, 1981; Thomson and Read, 1996; Beaver et al., 2000). Nine of 12 immature dogs showed no bony fusion 6 weeks after transarticular fixation despite use of a bone graft. Nevertheless, the fixations were stable in 10 of the 12 dogs, usually due to fibrocartilaginous tissue (Sorjonen and Shires, 1981). If implant failure does occur, either a splint can be used or a second surgery may be performed. Good results can often be obtained using non-surgical management when the failure is only partial (9.9, 9.10). A second surgery is recommended for complete failure of fixation (Thomas et al., 1991; Beaver et al., 2000). After failure of either a dorsal or ventral surgery, rescue is probably best attempted using either multiple implants and bone cement or dorsal cross-pinning (Thomas et al., 1991; Jeffery, 1996; Schulz et al., 1997; Sanders et al., 2000; (9.32–9.34; 9.43). Mini plates are another alternative but there is little margin for error when placing the screws in C2 (Stead et al., 1993). Other postoperative complications include gagging or coughing; laryngeal paralysis (Beaver et al., 2000;
Sanders et al., 2000); tracheal collapse (Thomas et al., 1991); tracheal necrosis (Thomas et al., 1991); pneumonia (Rochat and Shores, 1999; Beaver et al., 2000); dyspnea and sudden death (Denny et al., 1988; Thomas et al., 1991). Tracheostomy may be necessary to relieve iatrogenic laryngeal paresis or paralysis (Sanders et al., 2000). Respiratory dysfunction can also result from spinal cord injury impairing respiratory drive or motor function (Thomas et al., 1991; Schulz et al., 1997) (7.11). Non-respiratory complications include torticollis (Beaver et al., 2000); delayed fracture of C1 (Johnson and Hulse, 1989; Beaver et al., 2000); Horner’s syndrome (Jaggy et al., 1991); and esophageal stricture secondary to gastric reflux (Schulz et al., 1997).
Dorsal fixation The main complications are death or failure of fixation. Death is due to cardiac or respiratory arrest, which may occur during placement or tightening of the wires (Denny et al., 1988). The wire suture can also break or ‘cheese-wire’ though C2 spinous process (Beaver et al., 2000; Thomas et al., 1991)(9.11). Almost 30% of dogs
Atlantoaxial subluxation
9.11 Failure of dorsal fixation. 2/0 Teflon-coated, polypropylene suture was used instead of wire in this immature dog to avoid ‘cheese wiring’ through the spinous process of C2 but it broke after 2 days. The radiograph shows loss of reduction; the hole drilled in C2 spinous process is visible (arrowheads). The first suture was replaced with 0.6-cm Dacron tape; it broke at the cranial edge of C1 after a further 7 days. Number 2 Mersiline was used for the third surgery but almost tore through the spinous process of C2. The spinous process was then reinforced with bone cement and the dog placed in a splint for 4 weeks, which caused severe otitis externa. The dog was walking normally 4 months later.
require a second surgery after standard dorsal fixation (Table 9.2). Other failures reported after dorsal wiring include wire suture breakage and fracture of the spinous process of C2 (Chambers et al., 1977; Thomas et al., 1991; Beaver et al., 2000).
POSTOPERATIVE CARE
(see Chapter 15)
Pain is common postoperatively and adequate analgesia should be provided, avoiding respiratory depressant drugs in tetraplegic or severely tetraparetic animals (Tables 15.1, 15.2). Strict rest is enforced for 6–12 weeks following surgery. Osseous fusion, or at least stability of implants for a minimum of 6 weeks, is the goal. External support is useful after surgery, especially with transarticular or dorsal wire fixation. The splint should be removed periodically for a thorough inspection.
PROGNOSIS The prognosis for dogs with congenital lesions is good if the animal survives the perioperative period (Beaver et al., 2000). The best predictor of a successful outcome is when the onset of signs is prior to 2 years of age (Beaver et al., 2000). The final outcome also tends to be better if signs have been present for less than
10 months, if the dog can still walk, and if the reduction after surgery is good (Beaver et al., 2000). Despite the guarded prognosis for dogs with severe neurological deficits, 9 of 13 dogs (69%) that were unable to walk before surgery had a good outcome (Thomas et al., 1991; Knipe et al., 2002). Neck pain persists in about 10% of dogs and residual ataxia in about 20% after transarticular pin fixation (Beaver et al., 2000). The rate of residual ataxia is two to three times as high after dorsal wire fixation, presumably as the fibrosis at the surgical site provides less stability than osseous or fibrocartilaginous fusion ventrally (Beaver et al., 2000) (Tables 9.1, 9.2). Overall failure rates are as follows: • After dorsal wire fixation techniques the failure rate is 37%; the mortality rate is 16% and 29% of dogs need a second surgery (Table 9.2). A literature summary reported a similar failure rate in 20/52 dogs (38%) (McCarthy et al., 1995). Outcome after dorsal wiring is no different to that after ventral transarticular fixation using smooth pins (McCarthy et al., 1995; Beaver et al., 2000). • After transarticular fixation the overall failure rate is 29%; the mortality rate is 19% and 11% of dogs need a second surgery. However, when results for lag screw transarticular fixation are considered specifically the results are much better than for smooth pins (Denny et al., 1988; McCarthy et al., 1995). Threaded pins should also give much better results than smooth pins due to their higher pullout strengths (Sandman et al., 2001). • After multiple ventral implants and bone cement the failure rate is 12%; the mortality rate is 8% and 4% of dogs need a second surgery (Table 9.1). Reported outcomes are therefore best for techniques that employ multiple ventral implants and bone cement. Dogs requiring a second surgery have no worse outcomes than dogs that undergo only one (Beaver et al., 2000).
ATLANTOAXIAL SUBLUXATION IN CATS Clinical signs of congenital luxation are similar to those seen in dogs. All three cats described had good outcomes after transarticular pin fixation (Jaggy et al., 1991; Shelton et al., 1991; Thomson and Read, 1996). Differential diagnoses for cats include disc disease (see Chapter 8), mucopolysaccharidosis, hypervitaminosis A (see Chapter 14), lymphoma or other spinal cord tumors (see Chapter 12), trauma (see Chapter 13), meningoencephalitis, discospondylitis and possibly fibrocartilaginous embolism (FCE) (see Chapter 14).
169
170
Small Animal Spinal Disorders
Key issues for future investigation 1. What is the long-term outcome after non-surgical management for atlantoaxial subluxation? 2. Is a splint necessary to provide additional stability after surgery? 3. Is there any advantage to performing routine odontoidectomy for a malformed dens? 4. What is the prevalence and significance of syringohydromyelia in atlantoaxial subluxation (
9.12 Syringohydromyelia has been reported as an incidental finding on MRI of dogs with atlantoaxial subluxation (Sanders et al., 2000). It could explain residual signs (9.32) or failure to improve after stabilization. This dog shows high signal within the spinal cord over the first and second cervical vertebrae (arrow). There is also severe compression due to instability at the atlantoaxial joint (arrowhead).
REFERENCES Beaver, D.P., Ellison, G.W., Lewis, D.D., Goring, R.L., Kubilis, P.S., Barchard, C. (2000) Risk factors affecting the outcome of surgery for atlantoaxial subluxation in dogs: 46 cases (1978–1998). Journal of the American Veterinary Medical Association 216, 1104–1109. Chambers, J.N., Betts, C.W., Oliver, J.E. (1977) The use of nonmetallic suture material for stabilization of atlantoaxial subluxation. Journal of the American Animal Hospital Association 13, 602–604. Cook, J.R., Oliver, J.E. (1981) Atlantoaxial luxation in the dog. Compendium on Continuing Education for the Practicing Veterinarian 3, 242–250. Denny, H.R., Gibbs, C., Waterman, A. (1988) Atlantoaxial subluxation in the dog: a review of thirty cases and an evaluation of treatment by lag screw fixation. Journal of Small Animal Practice 26, 37–47. Gibson, K.L., Ihle, S.L., Hogan, P.M. (1995) Severe spinal cord compression caused by a dorsally angulated dens. Progress in Veterinary Neurology 6, 55–57. Hawthorne, J.C., Cornell, K.K., Blevins, W.E., Waters, D.J. (1998) Nonsurgical treatment of atlantoaxial instability: A retrospective study. Veterinary Surgery 27, 526. Huibregtse, B.A., Smith, C.W., Fagin, B.D. (1992) Atlantoaxial luxation in a Doberman Pinscher. Canine Practice 17, 7–10. Hurov, L. (1979) Congenital atlantoaxial malformation and acute subluxation in a mature Basset Hound—surgical treatment by wire stabilization. Journal of the American Animal Hospital Association 15, 177–180.
Jaggy, A., Hutto, V.L., Roberts, R.E., Oliver, J.E. (1991) Occipitoatlantoaxial malformation with atlantoaxial subluxation in a cat. Journal of Small Animal Practice 32, 366–372. Jeffery, N.D. (1996) Dorsal cross pinning of the atlantoaxial joint: new surgical technique for atlantoaxial subluxation. Journal of Small Animal Practice 37, 26–29. Johnson, S.G., Hulse, D.A. (1989) Odontoid dysplasia with atlantoaxial instability in a dog. Journal of the American Animal Hospital Association 25, 400–408. Knipe, M.F., Sturges, B.K., Vernau, K.M., Berry, W.L., Dickinson, P.J., Anor, S., LeCouteur, R.A. (2002) Atlantoaxial instability in 17 dogs. Journal of Veterinary Internal Medicine 16, 368. LeCouteur, R.A., Child, G. (1995) Diseases of the spinal cord. In: S.J. Ettinger (ed.), Textbook of Veterinary Internal Medicine, 629–695. Philadelphia: WB Saunders. Lorinson, D., Bright, R.M., Thomas, W.B., Selcer, R.R., Wilkens, B.A. (1998) Atlantoaxial subluxation in dogs: the results of conservative and surgical therapy. Canine Practice 23, 16–18. Martinez, S.A., Arnoczky, S.P., Flo, G.L., Brinker, W.O. (1997) Dissipation of heat during polymerization of acrylics used for external skeletal fixator connecting bars. Veterinary Surgery 26, 290–294. Mayhew, I.G. (1999) The healthy spinal cord. American Association of Equine Practitioners; 56–66. McCarthy, R.J., Lewis, D.D., Hosgood, G. (1995) Atlantoaxial subluxation in dogs. Compendium on Continuing Education for the Practicing Veterinarian 17, 215–226. McKeever, F.M. (1968) Atlantoaxial instability. Surgical Clinics of North America 48, 1375–1390. Read, R., Brett, S., Cahill, J. (1987) Surgical treatment of occipitoatlantoaxial malformation in the dog. Australian Veterinary Practitioner 17, 184–189. Renegar, W.R., Stoll, S.G. (1979) The use of methylmethacrylate bone cement in the repair of atlantoaxial subluxation stabilization failures. Case report and discussion. Journal of the American Animal Hospital Association 15, 313–318. Rochat, M.C., Shores, A. (1999) Fixation of an atlantoaxial subluxation by use of cannulated screws. Veterinary and Comparative Orthopaedics and Traumatology 12, 43–46. Sanders, S.G., Bagley, R.S., Silver, G.M. (2000) Complications associated with ventral screws, pins and polymethylmethacrylate for the treatment of atlantoaxial instability in 8 dogs. Journal of Veterinary Internal Medicine 14, 339. Sandman, K.M., Smith, C.W., Harari, J., Manfra Maretta, S., Pijanowski, G.J. (2001) Comparison of pull-out resistance of Kirschner wires and Imex miniature interface fixation pins in polyurethane foam. Veterinary and Comparative Orthopaedics and Traumatology 15, 18–22. Schulz, K.S., Waldron, D.R., Fahie, M. (1997) Application of ventral pins and polymethylmethacrylate for the management of atlantoaxial instability: results in nine dogs. Veterinary Surgery 26, 317–325. Shelton, S.B., Bellah, J., Chrisman, C., McMullen, D. (1991) Hypoplasia of the odontoid process and secondary atlantoaxial luxation in a Siamese cat. Progress in Veterinary Neurology 2, 209–211. Sorjonen, D.C., Shires, P.K. (1981) Atlantoaxial instability: a ventral surgical technique for decompression, fixation, and fusion. Veterinary Surgery 10, 22–29. Stead, A.C., Anderson, A.A., Coughlan, A. (1993) Bone plating to stabilise atlantoaxial subluxation in four dogs. Journal of Small Animal Practice 34, 462–465. Swaim, S.F., Greene, C.E. (1975) Odontoidectomy in a dog. Journal of the American Animal Hospital Association 11, 663–667. Thomas, W.B., Sorjonen, D.C., Simpson, S.T. (1991) Surgical management of atlantoaxial subluxation in 23 dogs. Veterinary Surgery 20, 409–412. Thomson, M.J., Read, R.A. (1996) Surgical stabilisation of the atlantoaxial joint in a cat. Veterinary and Comparative Orthopaedics and Traumatology 9, 36–39. Watson, A.G., de Lahunta, A. (1989) Atlantoaxial subluxation and absence of transverse ligament of the atlas in a dog. Journal of the American Veterinary Medical Association 195, 235–237. Wheeler, S.J. (1992) Atlantoaxial subluxation with absence of the dens in a Rottweiler. Journal of Small Animal Practice 33, 90–93.
Atlantoaxial subluxation
PROCEDURES The patient is placed with the neck in extension; this helps to reduce the subluxation. Although positioning the neck like this helps, reduction of C1 and C2 is usually incomplete. The mid-body of C2 must be grasped with bone holding forceps and pulled in a caudoventral direction (9.19). Breaking down the joint capsule may also be necessary to complete reduction. Occasionally malformation or secondary changes make complete reduction impossible (9.32). The approach can either be the standard ventral approach (see page 106) or the paramedian approach to the neck (see page 232). Preserving the thyroid artery (9.18) is much easier when using the paramedian approach as the carotid artery and its thyroid branch are reflected along with the trachea and larynx. The dog’s head is to the left in all illustrations.
Ventral transarticular fixation (9.13–9.31)
9.13 Positioning for surgery. The area is prepared and draped, including the proximal humerus to allow bone graft collection (11.24).
9.13
9.14 Site of incision. Note that the incision extends cranial to the larynx. At the cranial edge of the skin incision is the hyoid venous arch, which should be divided.
9.14
9.15 Incision through the superficial fascia reveals the sternohyoid muscles (7.21, 7.22). The larynx (a) is at the cranial end of the incision, and the trachea is visible (b).
a b
9.15
171
172
Small Animal Spinal Disorders
9.16 The sternothyroid muscle inserts on the thyroid cartilage. The muscle is sectioned along the dotted line. The vascular bundle shown in 9.18 can be seen under the muscle insertion.
9.16
9.17 The sternothyroid muscle is mobilized and divided close to the larynx. Leave an adequate portion near the thyroid cartilage for repair. The thyroid gland is visible (*).
9.17
9.18 A vascular bundle supplies the thyroid gland; only one artery supplies each gland and so it must be preserved. Adequate padding with moist towels or sponges must be provided for the trachea, esophagus and thyroid gland. Insufficient protection has been provided here to the trachea, which could lead to necrosis (Thomas et al., 1991). 9.18
9.19
9.19 These sagittal 3D reconstructions of CT scans show how subluxation can make the initial approach challenging. A: Normal dog to show congruity of B A the ventral surfaces of C1 and C2. B: In most dogs with atlantoaxial subluxation, the dens, atlantoaxial joints and the cranial portion of C2 are obscured under the body of C1. Reduction is in the direction of the arrow. Images from the same dogs are shown in 9.1 and 9.6.
Atlantoaxial subluxation
9.20 Deep fascia dissected with C1 and C2 reduced to show the tendons of the longus colli muscles inserting on the ventral process of C1 (arrow).
9.20
9.21 Diagram of deep anatomy as shown in 9.20. Note the relationship of the soft tissues to the underlying skeletal structures.
9.21
9.22 The tendons of the longus colli muscles are elevated from the ventral process of C1. The muscles are elevated caudolaterally from the body of C2 (exposed here). Dissection of the fascia reveals the joint capsule of the C1/C2 joints. Here the capsule has been incised and removed partially on the dog’s right side (arrow).
9.22
9.23 The joint spaces can be seen clearly. The articular cartilage is removed with a curette, #11 blade or a small bur. If a bur is used, care must be taken not to weaken the cortical bone on either side of the joint.
9.23
173
174
Small Animal Spinal Disorders
9.24
9.24 The joint space can be opened using a dental tartar scraper or small Hohmann retractor. This is a useful maneuver in order to support C2 while drilling. Here the cranial articular cartilage has been removed revealing the subchondral bone (arrow).
9.25 Position of screws. Ideally, the screws are angled away from the midline at approximately 30°, towards the medial angle of the alar notch, and downward (i.e. dorsally) at approximately 20° from the horizontal, which in practice means using as flat a trajectory as possible (Sorjonen and Shires, 1981).
9.25
9.26 Placement of screws. In most miniature dogs, 1.5-mm cortical bone screws are used. A 1.1-mm hole is drilled in C2 and across the articulation. The body of C2 has a tendency to move down, away from the surgeon; use of the stabilization technique shown in 9.19 and 9.24 can help to prevent this. Once the hole has been drilled in one side of C2, a tartar scraper may be inserted in the hole to stabilize the vertebra while the other hole in C2 is drilled.
9.26
Atlantoaxial subluxation
The following order for drilling, tapping and screw placement is preferred: 1. Drill both 1.1-mm holes in C2. 2. Drill 1.1-mm hole through one of these holes into C1 on one side only. Insure that far cortex is penetrated. 3. Drill 1.5-mm glide hole on the same side in C2. 4. Measure depth of the hole through C2 into C1. 5. Tap this hole with 1.5-mm tap. 6. Place screw through C2 into C1. Do not tighten fully. 7. Repeat steps 2 to 6 on the other side. Performing the operations in the order shown above avoids the problem of drilling holes and not being able to locate one after the other screw has been tightened. A similar sequence can be used with threaded pins. These still require a pilot hole and should be inserted using a low power setting or ideally with a small hand-held chuck (5.22). Pins must also be encased completely within bone cement (Johnson and Hulse, 1989). Cannulated screws provide another alternative as they permit repositioning of the guide wires if necessary before placing screws. These screws are self-tapping and come in 4.0- and 3.0-mm diameters (Rochat and Shores, 1999). The average mediolateral pin angle in dogs undergoing a successful stabilization was between 22° and 27°, with a theoretical ideal of 29° and a practical range of 10° and 45°. If implants are directed too laterally across the atlantoaxial joints, they may damage the vertebral artery; if directed too medially they may damage spinal cord (Rochat and Shores, 1999). The average ventrodorsal angle in the same dogs was between 28.5° and 34.5° with a theoretical ideal of 22° and a practical range of 15° and 45° (9.8, 9.29, 9.31). The theoretical ideal ventrodorsal angle is hard to approach and in practice the angle should be as near horizontal as possible. Starting the implants as far caudally as possible in C2 may facilitate this (Sorjonen and Shires, 1981; Jaggy et al., 1991). Malformation of C1 or C2 can make assessment of the correct angles very difficult (Thomas et al., 1991). Use of a goniometer during surgery may help improve accuracy (Thomson and Read, 1996). Vertebral positioning must also be perfectly symmetrical. Odontoidectomy is usually described prior to reduction but may be easier and safer once the joint is reduced and stabilized (Johnson and Hulse, 1989; Jaggy et al., 1991; Gibson et al., 1995) (9.27). Access to the dens is through a short slot in C1. Odontoidectomy is recommended if the dens has a marked dorsal angulation (Swaim and Greene, 1975; Johnson and Hulse, 1989; Jaggy et al., 1991; Gibson et al., 1995) but is not necessary in most dogs (Schulz et al., 1997).
9.27 Here one screw is in position, and the other is being inserted prior to final tightening.
9.27
175
176
Small Animal Spinal Disorders
9.28 Cancellous bone is harvested from the proximal humerus and placed in and around the joint space before tightening the screws (11.24). The longus colli is then apposed with absorbable sutures and the sternothyroid muscle is repaired.
9.28
R
9.29 A: The ventrodorsal screw angle is nearly perfect (25°). B: The mediolateral angles are a little high but still acceptable (left 43° and right 40°). This dog, also shown in 9.5, was normal at 3-month follow-up.
9.29 A
9.30 If the ventrodorsal angle is suboptimal then failure occurs due to insufficient bone purchase (Rochat and Shores, 1999). Despite implant failure, this 1-year9.30 old Toy poodle was managed in a splint and was A normal 3 years later.
B
B
L
9.31 The screw on the left is not angled adequately away from the midline and was probably not crossing the joint. It snapped soon after surgery and the dog had a partial recurrence of signs. The dog was managed by addition of an external support and made a good recovery. Angles are mediolateral: left 14° and right 41°.
9.31
Atlantoaxial subluxation
Multiple ventral implants and bone cement (9.32–9.34) Screws can be directed laterally into the thick bone just rostral to each caudal articular surface of C1 (1.15, 9.33A) and also into the bone caudal to the cranial articular surface of C2 (1.15B). The screw heads are then united with bone cement (9.32). Variations on this technique include placement of two implants in C1, two in C2, and two more in C3 (Sanders et al., 2000); and the pattern shown in 9.34 that incorporates transarticular implants (Schulz et al., 1997). Implants of the largest possible size should be chosen and they should penetrate two cortices for greatest pullout strength (Sandman et al., 2001). All implants should be covered completely by cement. Threaded pins are superior to smooth pins and should ideally be put in by hand using a mini-chuck (5.22). Whatever pattern of multiple implant is used, the length of implant and the volume of cement must be considered in relation to the amount of room existing at this location (9.34). Care must be taken to protect soft tissues from the heat of polymerization and to avoid rough edges that might abrade tissues.
9.32 This Miniature poodle was tetraparetic at 6 weeks of age but improved dramatically in a splint. The dog presented with neck pain, tetraparesis and fecal incontinence at 3.5 years. Fixation was with 9.32 multiple implants and bone B cement. A and B: Despite A excellent reduction dorsally there is a marked step on the floor of the vertebral canal; the malformed dens probably prevented complete reduction. Odontoidectomy may have helped but the risk of iatrogenic spinal cord trauma was considered too high (9.6). The dog was pain-free but still incontinent at 1-year follow-up.
9.33 CT scans to show A: site of implant placement in the thick bone just rostral to the caudal articular surfaces of C1 (arrows) and B: the method for cross-pinning into the caudal vertebral body of C2 (arrows). 9.33 Implants should penetrate two cortices for maximum A B holding power (Sandman et al., 2001). When available, intraoperative radiography is recommended to confirm implant placement (Rochat and Shores, 1999). See also 1.14, 1.15.
177
178
Small Animal Spinal Disorders
9.34 Postoperative A: lateral and B: ventrodorsal radiographs to show fixation in a 1-year-old Yorkshire terrier using six Kirschner wires, cancellous bone graft and bone cement. Two wires were placed into each of the pedicles of C1 (9.33A). Two wires were placed across the 9.34 atlantoaxial joints as described in B 9.24–9.27. Two wires were placed into A the caudal body of C2 at approximately 30° to the transverse plane (9.33B); ideally they should not cross C2/3 disc space (from Schulz et al., 1997).
Dorsal wire fixation (9.35–9.42)
9.35 Incision site relative to the skeleton. The finger is on the occipital protuberance; the incision is made just off the midline. Note that the neck must be flexed, which is not ideal (page 171).
9.35
9.36 Exposure of the spinous process of C2. Manipulation and movement of the vertebrae must be kept to a minimum; it is therefore preferable to use sharp dissection.
9.36
Atlantoaxial subluxation
9.37 Diagram to show the relationship between skeletal, vascular and nervous structures. (a) Dorsal notch of the foramen magnum; (b) dorsal arch of C1; (c) spinous process of C2; (d) C2 nerve roots and vessels. Note also the vertebral artery and its branches (1.36).
c
b
d
a
9.37
9.38 Two holes are drilled in the spinous process of C2.
9.38
9.39 The dorsal atlantoaxial ligament between C1 arch and C2 spinous process is disrupted in atlantoaxial subluxation. The periosteum and soft tissues are removed to allow access to the vertebral canal between C1 and C2. A double loop of wire is passed under the arch of C1 in a cranial direction. Pressure on the spinal cord must be avoided. The internal periosteum of the vertebral canal may be continuous with the dura mater at this stage, and gentle dissection is required to allow passage of the wire.
9.39
9.40 The loop of wire is retrieved from the atlantooccipital space. This space may need to be enlarged in order to grasp the wire. This is best achieved by removing bone from the occiput, thus preserving the dorsal arch of C1 required for the fixation. 9.40
179
180
Small Animal Spinal Disorders
9.41 Arrangement of wires after tightening and cutting the ends. If the spinous process is too small to drill holes into, the suture can be passed under the C2 spinous process (Chambers et al., 1977).
9.41
9.42 Postoperative radiograph showing reduction of the subluxation. Note that the wire under the lamina of C1 is causing some spinal cord compression.
9.42
Dorsal cross-pin fixation (9.43) This should be stronger than dorsal wiring; it does not carry the risk of spinal cord trauma and provides a useful rescue technique for any fixation failure. A disadvantage is that it still relies on the spinous process of C2. Care must be taken during exposure of the wings of the atlas to avoid the C1 nerve root and also the vertebral artery as it exits the transverse foramen (1.36). With the vertebrae reduced, a K-wire is driven in a ventrolateral direction across each side of the spinous process of C2 to engage and penetrate the caudal half of the wing of C1. Predrilling a small initial hole and then using a bone gouge is necessary to stop the pin from slipping caudally off the sloped surface of the wing. Each K-wire is then removed and replaced by a positive profile pin. The pin usually has to curve slightly between C1 and C2 in order to make good bone contact; the curve is best induced using the bone gouge. Bone cement is then applied to the exposed aspects of the pins but direct contact should be avoided between the cement and the spinous process of C2 in order to reduce the risk of thermal necrosis weakening the bone of toy-breed dogs. Bone graft is also applied between the arch of C1 and the spinous process of C2 to promote a fibro-osseous union (Jeffery, 1996).
9.43 A,B: Dorsal cross-pin fixation.
9.43 A
B
Lumbosacral disease
Clinical signs
Chapter
10
183
Diagnosis 183 Examination 183 Differential diagnosis 184 Electrophysiology 184 Radiography 185 Treatment 188 Non-surgical treatment 188 Surgical treatment 188 Surgery 189 Dorsal laminectomy 189 Foraminal decompression or facetectomy Dorsal fusion–fixation 190
190
Complications 191 Intraoperative 191 Early postoperative 191 Late postoperative 191
10.1 Dog in a typical posture of low back pain. It also had lameness in one pelvic limb and marked pain on palpation over the lumbosacral junction.
Postoperative care
cauda equina. The spinal cord ends within L6 vertebra in most dogs or in L7 in cats and some small dogs (Fletcher and Kitchell, 1966) (1.5, 1.8B, 1.9). The L7 nerve roots run in unique troughs called the lateral recesses within L7 vertebra (1.28B). Motion in the normal lumbar spine is greatest at the lumbosacral joint (Burger and Lang, 1993). In some dogs, abnormal motion probably leads to degenerative changes such as spondylosis deformans, osteophyte proliferation and soft tissue overgrowth of the joint capsules. These degenerative changes appear to reduce the overall range of motion and clinical problems may arise due to subsequent compression of neural structures in the vertebral canal and intervertebral foramina (Mattoon and Koblik, 1993; Schmid and Lang, 1993). This sequence of events resembles those thought to occur in hip dysplasia (Chambers et al., 1988). A number of abnormalities may combine to cause compression of the cauda equina or L7 nerve roots (10.2–10.4). These include:
Prognosis
192
192
Key issues for future investigation References
193
193
Procedures 195 Dorsal laminectomy 195 Foraminal decompression and facetectomy Dorsal fixation–fusion 204
202
The clinical signs seen with lumbosacral lesions differ from those seen at other locations of the spine (10.1), mainly because of the unique anatomical structure of the region. The vertebral canal in this region contains only
182
Small Animal Spinal Disorders
A
B
10.2 A: Sagittal section to show the normal relationship between L7 and the sacrum. B: 3D reconstruction from a dog with a normal lumbosacral spine (same dog as in 2.24). The dog was free of signs for lumbosacral disease and a transverse CT study through this region was normal (10.10A).
10.3 A: Hansen type II disc herniation compressing the cauda equina (10.10B). B: 3D reconstruction of a CT scan from a 7-year-old Mastiff with chronic, lower back pain, reduced hock flexion and low tail carriage. A large, disc herniation was excised via dorsal laminectomy. The dog then developed urinary retention due to increased sphincter tone; this resolved after 2 weeks of catheterization (O’Brien, 1988; Coates, 1999). A
A
B
B
10.4 A: Ventral subluxation of the sacrum relative to L7. This may occur in normal dogs, although a step of more than 4 mm suggests an abnormal lumbosacral junction (Wright, 1980; Denny et al., 1982; Schmid and Lang, 1993). B: Six-year-old working German shepherd dog presenting with mild paraparesis, weak hock flexion and lumbosacral pain. The dog was seropositive for Ehrlichia canis and returned to work after antibiotics; the subluxation did not seem to cause any clinical signs (3.2, 4.35).
Lumbosacral disease
•
Stenosis (multilevel) of the vertebral canal (Meij, 1993; Jones et al., 1995a,b, 1996b). • Hansen type II disc herniation at the L7/S1 intervertebral space (Jones et al., 1996b). • Subluxation, osteophytosis or thickening of the articular processes (Jones et al., 1994). • Epidural fibrosis (Oliver et al., 1978; Sisson et al., 1992; Jones et al., 1996b, 1999). • Soft tissue proliferation, usually of the joint capsule or ligamentous structures (Jones et al., 1999). • Vascular compromise of the spinal nerves (Tarvin and Prata, 1980; Jones et al., 1996a, 1999; Porter, 1996). • Osteochondrosis of the sacrum (Lang et al., 1992; Hanna, 2001). • Instability and misalignment between L7 and S1 (Schmid and Lang, 1993). The exact role of lumbosacral instability is unclear and it is difficult to quantify regardless of the imaging technique (Schmid and Lang, 1993; Fox et al., 1996). This can pose a problem for surgical decision-making (Algorithm 10.1). Larger-breed dogs, particularly German shepherd dogs, are affected most frequently although signs are also reported in small dogs (Tarvin and Prata, 1980; Lang et al., 1992; Jones et al., 1995b). Young, working dogs that have been trained heavily are particularly prone to this disorder (Wheeler, 1992; Danielsson and Sjostrom, 1999; Jones et al., 2000). The condition is rare in cats (Hurov, 1985; Stoll, 1996). Although degenerative and congenital conditions are the focus of this chapter, other disorders can affect the lumbosacral spine. Tumors (see Chapter 12) and fractures (see Chapter 13) should be considered in the differential diagnosis. Discospondylitis also occurs at the lumbosacral junction (14.12); its management is discussed in Chapter 14 (page 327). Articular (synovial) cysts have also been described as a cause of lumbosacral pain in a German shepherd dog (Webb et al., 2001).
CLINICAL SIGNS Lumbosacral lesions can cause pelvic limb gait abnormalities, lameness, or lower motor neuron (LMN) neurological deficits. Pain is common but additional signs vary depending on the nature and severity of the neurological impairment (McKee, 1993) (Table 10.1). There may be pain with no deficits; mild paresis with proprioceptive deficits; or paraparesis, tail paralysis and incontinence (Oliver et al., 1978; de Risio et al., 2001). Limb signs include lameness or sciatic deficits affecting the caudal thigh muscles and those distal to the stifle (Tarvin and Prata, 1980; Meij et al., 1993). Lameness in performance animals is often exacerbated by work (Jones et al., 2000).
Table 10.1 Clinical signs of lumbosacral disease Mild
Severe
Low back pain – palpation, tail elevation, per rectum Difficulty sitting Difficulty jumping Difficulty climbing Pelvic limb lameness, worsening with exercise Hyperesthesia, pruritus
Mild pelvic limb paresis Pelvic limb muscle atrophy Tail paresis Fecal incontinence Anal hyporeflexia Urinary incontinence Self-mutilation
Incontinence results from pelvic and pudendal nerve dysfunction. Urinary incontinence is usually LMN in nature with dribbling of urine and a bladder that is easily expressed by manual pressure (10.3) (see ‘Control of urinary function’, page 350). Rarely, urinary dysfunction may be the only clinical sign (Tarvin and Prata, 1980). Fecal incontinence seems to be related mainly to poor anal tone, which may be present even when the anal reflex is intact. Dogs with chronic degenerative lumbosacral lesions may present with non-specific clinical signs, but low back pain is quite different from that seen in thoracolumbar lesions. Diagnosis of lumbosacral disease depends on recognizing the historical features and clinical signs and on a careful physical examination, which should pinpoint the source of pain.
DIAGNOSIS Examination Clinical signs may be vague in some dogs, which makes accurate diagnosis difficult. A thorough physical, orthopedic and neurological examination is essential, along with a rectal examination. In view of the difficulty in interpreting some of the ancillary diagnostic tests the clinical signs may provide the main basis for reaching a diagnosis (Table 10.1). The anal reflex, sphincter tone and tail tone should be evaluated. Reduced or absent hock flexion during the withdrawal reflex (2.27) is a sensitive indicator of motor dysfunction in the sciatic nerve (but mild lesions will only cause loss of proprioception). The patellar reflex may also appear exaggerated if sciatic nerve function is depressed. This phenomenon of ‘pseudohyperreflexia’ must be differentiated from the increased reflex that occurs with upper motor neuron (UMN) deficits seen with lesions cranial to the L4 segment. Pseudohyperreflexia results from decreased tone in the muscles innervated by the sciatic nerve, which normally
183
184
Small Animal Spinal Disorders
counteract the extension of the stifle induced by the patellar reflex (see page 27). Hyperesthesia is a frequent finding and the hindquarters should be palpated carefully to locate the focus of pain. Manipulation of the limbs and spine may provoke a pain response, but it may be difficult to distinguish pain associated with spinal disease from that caused by orthopedic problems. Direct pressure over the lumbosacral joint, either with the examiner’s knee between the dog’s rear limbs to support its pelvis and lift its feet off the ground slightly or with the dog lying on its side (or per rectum), may help to differentiate lumbosacral pain from orthopedic disease. Pain may also be elicited by elevation of or traction on the tail and when rotating the lumbosacral joint by swinging both rear limbs from one side to the other (Sjostrum, 2003). A final test is to have the dog stand up on its rear legs, with its front feet resting on a chair and its spine extended, and then apply pressure over the lumbosacral space. If lameness or weakness worsen with exercise and resolve to some degree with rest then they may be due to neurogenic claudication, a condition well recognized in humans. This is due to failure of arterial vasodilation in affected nerve roots during exercise. As well as lameness, paresthesia and sphincter disturbance may also occur (Markwalder, 1993; Porter, 1996). Neurogenic claudication has also been recognized in dogs, one of which had concomitant stranguria (Tarvin and Prata, 1980).
Differential diagnosis Differentials vary depending on whether the dog has non-specific signs or obvious neurological deficits (Table 3.5 and 10.2). Dogs with the orthopedic diseases listed in Table 10.2 have a normal neurological examination. Dogs with pain and no neurological deficits could have synovial cysts (Webb et al., 2001); tethering of the dural sac (Huttmann et al., 2001; Shamir et al., 2001); inflammation of disc, meninges or nerve root; or possibly either facet or sacroiliac joint pain (Pang et al., 1998) (see Chapter 14). Dogs with degenerative myelopathy are pain free and the withdrawal reflex is normal, but the patellar reflex may be depressed because of dorsal nerve root involvement (see Chapter 14). It may not be possible to differentiate degenerative lumbosacral disease, discospondylitis and neoplasia on physical examination. In some dogs, disorders such as degenerative myelopathy (10.53) or a thoracolumbar disc(s) may coexist with lumbosacral disease and this can confound the neurological localization (2.24). The deficits of lumbosacral disease are referable to the L4–S3 nerve roots and so may be difficult to differentiate from L4–S3 lesions that are situated within the dural sac (2.25). In such dogs,
imaging should extend as far forward as the L4/5 disc space to include the sacral spinal cord.
Electrophysiology Clinical electrophysiological studies are useful to confirm LMN disease and nerve root dysfunction (Sisson et al., 1992; Meij, 1993; de Risio et al., 2001; Cuddon, 2002; Cuddon et al., 2003) (see page 60). Electromyography of the limbs, tail and perineum may reveal spontaneous activity, which is consistent with LMN involvement. However, a normal electromyogram does not eliminate the possibility of lumbosacral disease (de Risio et al., 2001). Electromyography is particularly useful to reduce the number of false-positive diagnoses associated with MRI evidence of nerve root disorders (Nardin et al., 1999) (see page 60). Nerve conduction studies and Fwave latencies may also be useful (Cuddon, 2002; Cuddon et al., 2003) (see page 61). Abnormal findings confirm LMN disease but do not specify the etiology.
Table 10.2 Differential diagnosis of lumbosacral disease
Mild clinical signs Neurological disorders Degenerative myelopathy Synovial cyst Schmorl’s node Discogenic pain Facet joint pain Sacroiliac joint pain Congenital anomaly—spina bifida, dermoid sinus Tethering of the terminal spinal cord Peripheral neuropathy Neoplasia—spinal cord or nerve root Discospondylitis Occult discospondylitis Meningomyelitis Polyradiculoneuritis Orthopedic disorders Coxofemoral arthritis Cruciate rupture Gracilis contracture Psoas muscle injury Pelvic limb lameness, other Other Prostatic disease Urethral neoplasia Anal sac adenocarcinoma
Severe neurological deficits
Degenerative myelopathy Ischemic myelopathy Ischemic neuromyopathy Neoplasia Discospondylitis or epidural abscess Polyradiculoneuritis Myelitis Trauma
Lumbosacral disease
Radiography SURVEY RADIOGRAPHY It is advisable that the patient be anesthetized or sedated heavily when radiographing the lumbosacral joint. Rotation of the spine and pelvis must be avoided (4.14, 4.15). The value of flexed and extended positional survey radiographs appears limited (Mattoon and Koblik, 1993; Schmid and Lang, 1993). The main role of survey radiographs is to rule out neoplasia (12.11A), trauma (13.21) and discospondylitis (14.12). Many clinically normal dogs have radiographic abnormalities of the lumbosacral junction (10.5). Conversely, occasional dogs with lumbosacral disease will have
normal survey radiographs (Ness, 1994). Identification of sacral osteochondrosis or transitional vertebrae increases substantially the likelihood that the dog’s signs are due to cauda equina compression (Lang et al., 1992; Morgan et al., 1993; Morgan, 1999) (10.6). Neurological localization of a patient may indicate a lesion of L4–S3 spinal cord or nerve roots. It may be possible to define the location more accurately, for example by the anal reflex. Consideration of the diagnostic imaging in many of these patients frequently focuses on the lumbosacral joint alone. This policy runs the risk of missing lesions elsewhere in the vertebral column that are involving the L4–S3 cord segments (2.25) or ascribing significance to incidental lesions (10.4).
MYELOGRAPHY
10.5 Lumbosacral spondylosis deformans and narrowing of the intervertebral space. This type of change is seen frequently in older, large-breed dogs but in itself is not diagnostic of disease (Morgan et al., 1989; Ramirez and Thrall, 1998).
Myelography (or MRI) is particularly useful in dogs that might have a spinal cord lesion (2.25). The subarachnoid space extends beyond the lumbosacral junction in about 80% of dogs (Lang, 1988; Ness, 1994; de Risio et al., 2001). Myelography can therefore be useful to assess the low lumbar region as well as the rest of the spinal cord (Ramirez and Thrall, 1998; Sjostrum, 2003). Cervical injection is preferred, as it avoids the potential for epidural contrast leakage in the area of interest. Filling of the dural sac is improved by flexing the spine (Lang, 1988). Lesions may cause dorsal elevation or attenuation of the column; flexion–extension studies may also be useful (10.7), but false-positive results occur (Lang, 1988; Watt, 1991; Danielsson and Sjostrom, 1999). Furthermore, a normal study does not preclude cauda equina involvement (Ramirez and Thrall, 1998).
R
A
B
C
10.6 A, B: Five-year-old German shepherd dog with lumbosacral pain. Sacral OCD is present; the fragment moves with flexion and extension (Lang et al., 1992). Subluxation of the articular facets (arrowhead) with foraminal narrowing is also evident (10.11, 10.46). C: Four-year-old dog with transitional vertebra and left nerve root signature. 3D reconstruction of the CT scan shows foraminal narrowing on the left due to a large articular facet (10.9). There is only one transverse process (arrow); the other is fused to the ilium.
185
186
Small Animal Spinal Disorders
Sjostrom, 1999) (10.8). Discography is abnormal if more than 0.3 ml of contrast medium can be injected (Barthez et al., 1994). When both are to be done the discogram should precede the epidurogram. Positional views may be employed with epidurography (Selcer et al., 1988; Ramirez and Thrall, 1998). If both myelography and epidurography are planned, the myelogram should be performed first, as the presence of epidural contrast complicates myelographic interpretation. The utility of discography may be enhanced if it is followed by CT and discography in humans may be more sensitive in detecting anular tears than MRI (Ohnmeiss et al., 1997; Milette et al., 1999).
DISCOGRAPHY AND EPIDUROGRAPHY These two methods can be very useful, especially when used in combination (Selcer et al., 1988; Barthez et al., 1994; Ramirez and Thrall, 1998; Danielsson and
COMPUTED TOMOGRAPHY CT is particularly useful as it shows clearly the vertebral canal (10.10), lateral recesses (1.28), intervertebral foramina (1.26, 1.27) and articular processes (10.9) in cross-sectional images (Jones et al., 1995a, 1996b). Reformatting can be used to create dorsal and sagittal images (10.11) and facet joint subluxation may be visible using bone window images or 3D reconstructions (Jones et al., 1994; Meij et al., 1996) (10.6, 10.48). CT can also be used for dynamic studies (10.11) and is helpful when performed after surgery (Meij et al., 1996; Jones et al., 2000) (10.54). The primary abnormalities visible on CT are a replacement of epidural fat with soft tissue density (10.10), which often represents epidural fibrosis, and multilevel compression (Jones et al., 1996b) (10.17). Use of intravenous contrast may improve surgical decision-making by identifying compressed tissues, which tend to enhance (Jones et al., 1999) (10.40). Some CT findings, such as loss of epidural fat, may not be of clinical significance in older animals (Jones and Inzana, 2000). In contrast to MRI, CT offers lower cost
A
B 10.7 Seven-year-old German shepherd dog with mild paraparesis and lumbosacral pain. There is a change in the degree of compression at the lumbosacral space between flexion and extension. A: Good filling of the dural sac when the spine is flexed; mild disc herniation is present at L6/7. B: There is dorsal displacement of the dural sac in extension and marked attenuation of contrast filling from L6/7 caudally. Fixation fusion was done using two screws and a bone graft but fixation was suboptimal (10.16).
A
B
10.8 A: Discogram demonstrating a large disc herniation. B: The needle was then withdrawn slightly to perform the epidurogram, which also outlines the bulging disc. Attenuation of more than 50% of the canal is considered abnormal (Ramirez and Thrall, 1998).
Lumbosacral disease
and thinner slice thickness along with better discrimination of bone spurs, articular process disease, soft tissue calcification, and soft tissue gas opacities (Jones et al., 2000). Scans should include at least the L6 vertebral body and, ideally, additional images should also be made through the L4/5 and L5/6 disc spaces (10.17).
component to the compression (10.12). One potential disadvantage of MRI is over-diagnosis. Even expert, human neuroradiologists tend to over-interpret the significance of lesions seen on MRI (Deyo, 1994; Jensen
MRI This provides better soft tissue resolution than CT as well as an ability to acquire images in multiple planes without image degradation; earlier detection of disc degeneration (1.9, 4.68) and evaluation of the entire lumbar spine in a single sagittal examination (Ramirez and Thrall, 1998; Jones et al., 2000) (1.9, 2.25, 10.12). Slice thickness is often greater than for CT, however, which increases volume averaging artefacts (Jones et al., 2000). Transverse images provide the best visualization of disc or foraminal anatomy (Adams et al., 1995) (1.9). They also reveal lesions in a foramen that cannot be detected by myelography or epidurography (Chambers et al., 1997). Sagittal views made with the spine in flexion and then extension help to identify a dynamic
A
B 10.10 A: Transverse CT scan at the level of the L7/S1 disc in a dog with a normal lumbosacral spine. The nerve roots and dural sac are surrounded by epidural fat. Same dog as in 10.2. B: In this dog (also shown in 10.3) there is almost complete loss of epidural fat and a general increase in soft tissue opacity.
A
A
B 10.9 Marked foraminal compression is evident on A: transverse CT scan, and B: 3D reconstruction. Same dog as in 10.6C.
B
10.11 Sagittal reformatted CT scan made with the dog in an A: extended position and B: flexed position. There is stenosis of the vertebral canal and foramina with a bulging L7/S1 disc. An OCD lesion is visible on the sacrum and the degree of compression increases with extension. Same dog as in 10.6 A, B and 10.46.
187
188
Small Animal Spinal Disorders
et al., 1994). Therefore, the clinician should not rely exclusively on imaging to confirm the diagnosis (BrantZawadzki et al., 1995; Gorman and Hodak, 1997; Milette et al., 1999). Part of the problem is a confusion in terminology. Classification of lesions into a disc bulge, protrusion or extrusion is not very helpful in humans (Milette et al., 1999) (see pages 13 and 58). In particular, bulges or protrusions are common incidental findings in people without back pain (Deyo, 1994; Jensen et al., 1994). Criteria in humans for disruption of disc anatomy include a loss of disc height; decrease in central disc signal on T2-weighting; or localized, peripheral hyperintensity on T2-weighting. Most outer anulus disruptions in humans are symptomatic (Milette et al., 1999). The choice of diagnostic tests is largely dependent on availability and clinician preference. Physical and neurological examination, survey radiography, CSF sampling at both the cerebello-medullary cistern (CMC) and lumbar sites, and clinical electrophysiology are recommended. Imaging by CT or MR is done when available as they provide cross-sectional images (Ramirez and Thrall, 1998; de Risio et al., 2001). CT and MRI can also be used to compare the degree of compression with the spine in flexed and extended positions (Bagley, 2003) (10.6A,B, 10.11, 10.12, 10.46). Either myelography or MRI should be performed if the dog has signs of spinal cord disease (2.25). MRI is the best, single imaging modality overall for humans (Kent et al., 1992). The position of the dog in the scanner can be very important; a neutral or flexed position reduces compression compared to an extended position (Adams et al., 1995; Jones et al., 2000). If CT or MRI is not available, myelography is performed followed by discography and epidurography if necessary (Sisson et al., 1992; Sjostrum, 2003). Some lesions detected by CT or MRI do not cause
A
clinical signs and this must be considered prior to instituting treatment (Deyo, 1994; Jensen et al., 1994; Jones and Inzana, 2000). A herniated disc is detected in a high proportion of people who have never had spinal pain and so imaging alone, both in dogs and in humans, should not serve as confirmation of a clinical diagnosis (Gorman and Hodak, 1997). Electrophysiological testing provides an additional means of independent confirmation of lesions (Nardin et al., 1999).
TREATMENT Non-surgical treatment Most dogs are treated initially with rest and antiinflammatory medication (Table 15.2). This may be successful if pain is the main clinical sign; 3–4 months of exercise restriction produced improvement in 8/16 dogs (50%) (Ness, 1994). Such a course is seldom effective in working dogs and signs often recur when normal activities are resumed (Danielsson and Sjostrom, 1999; Janssens et al., 2000; Sjostrum, 2003).
Surgical treatment Surgical treatment is indicated when non-surgical treatment has failed; in working dogs; and in those with pain or neurological deficits (Sjostrum, 2003). Further indications for surgery include CT or MR findings of increased soft tissue suggestive of epidural fibrosis, especially if it enhances with contrast. The choice of surgical procedure is then between dorsal laminectomy, distraction and fusion, or a combination of the two. Definitive criteria for these procedures are lacking. Laminectomy alone is not effective for dogs with chronic incontinence (de Risio et al., 2001). Fusion alone may be insufficient for dogs with neurological deficits or severe pain (Slocum and Devine,
B
10.12 T2-weighted MRIs of a 7-year-old Labrador with lumbosacral pain. A: Image made with the spine in flexion. There is loss of signal in the L7/S1 disc but little compression of the cauda equina (arrowhead). B: Image from the same dog with the spine in extension. There is severe compression at L7/S1 caused by both disc and ligamentum flavum (arrowhead). A large disc herniation was confirmed at surgery.
Lumbosacral disease
excised once the cauda equina is retracted laterally when there is marked bulging of the disc (10.3, 10.39). Routine fenestration may also be warranted (Danielsson and Sjostrom, 1999). Any redundant joint capsule should also be resected. Laminectomy often provides rapid relief of pain with improvement of mild gait abnormalities and minor neurological deficits (Danielsson and Sjostrom, 1999; Risio et al., 2001). It does not address instability if this is a contributing factor. Decompression by laminectomy can be combined with fixation and fusion (Slocum and Devine, 1998; Bagley, 2003) (10.46–10.54). As dogs with incontinence of more than 6 weeks duration respond poorly to laminectomy alone they should also undergo fixation and fusion (de Risio et al., 2001). Results of dorsal laminectomy are shown in Table 10.3.
1998). These presentations warrant combined dorsal laminectomy and fixation–fusion. Additional indications for distraction and fusion include a marked change in the degree of compression between flexion and extension or a telescoping of the facets, which suggest instability (10.6A,B, 10.46–10.48). An algorithm for surgical decision-making is shown in Algorithm 10.1.
SURGERY Dorsal laminectomy (10.18–10.39) Decompression of the cauda equina and spinal nerves can be achieved by dorsal laminectomy (10.13), which can be combined with foraminal decompression or even facetectomy (10.40–10.45). The anulus fibrosus should be
Pain, mild
Neurological deficits or severe pain without incontinence
Dorsal laminectomy +/– foraminal decompression
Fixation– fusion
A
Incontinence
Severe radiculopathy
Algorithm 10.1 Surgical decision-making in lumbosacral disease.
Severe instability on imaging
Dorsal laminectomy, +/– foraminal decompression AND fixation– fusion
Dorsal laminectomy, foraminal decompression or facetectomy +/– fusion
B
10.13 Postoperative appearance 1 year after dorsal laminectomy; this dog remained painful until it underwent a fixation fusion (10.53). An asterisk indicates the L7/S1 disc space. A: Mid-sagittal 3D reconstruction after CT scan. B: Dorsal view of the 3D reconstruction shown in A. Same dog as shown in 10.6A,B, 10.11 and 10.46.
189
190
Small Animal Spinal Disorders
Epidural fibrosis is a common finding; this is probably a response to chronic compression (10.17B). Magnification is recommended if attempts are made to break down adhesions around nerve roots (Oliver et al., 1978; Fingeroth et al., 1989; Sisson et al., 1992; Markwalder, 1993; Jones et al., 1996b, 1999).
Foraminal decompression or facetectomy (10.40–10.45) An increase in soft tissue within the foramen, loss of epidural fat or facet osteophytosis are indicators of probable radiculopathy (Adams et al., 1995; Jones et al., 1996b). Contrast enhancement after CT or MRI further increases preoperative suspicion of foraminal disease (Jones et al., 1999, 2000) (10.40). Together with radicular pain and delayed late waves (see Chapter 4), these features are indicators for some form of foraminal decompression (Oliver et al., 1978; Chambers et al., 1988). The laminectomy can be simply undercut or widened over the foramen and any osteophytes can also be removed by undercutting (10.41–10.43). However, the foramen can be difficult to examine fully until the facet is removed (Watt, 1991). The risks and benefits of facetectomy are unclear as is the need for subsequent stabilization (Slocum and Devine, 1998) (10.45). Lumbosacral stiffness is decreased markedly after bilateral facetectomy (Smith and Bebchuk, 2002).
Dorsal fusion–fixation (10.46–10.54) There are a number of similarities between lumbosacral disease and caudal cervical spondylomyelopathy (see
Chapter 11). A rationale can therefore be made for lumbosacral distraction and stabilization (Oliver et al., 1978). The principle is the same, namely to open the vertebral canal and intervertebral foramen, relieve pressure on neural tissues, and prevent abnormal motion. Distraction of the dorsal aspect of L7 and S1 is achieved using screws placed through the articular processes of L7/S1 and into the body of the sacrum (10.14). Fusion is promoted by removing the articular cartilage from the facet joints and by a cancellous bone graft (10.52). This can be combined with dorsal laminectomy, especially for dogs with marked neurological deficits, severe pain,
10.14 Middle-aged German shepherd dog that presented with mild pelvic limb weakness, lumbosacral pain, anal hyporeflexia and dribbling urine. CT scan revealed marked telescoping of the sacrum relative to L7. Screw fixation was performed using 4.0 mm cancellous screws. The screws had loosened slightly at 6-week follow-up but the dog had an excellent outcome with resolution of all signs for 3 years.
Table10.3 Results after dorsal laminectomy (⫹ or ⫺ foraminal decompression) Oliver et al., 1978
Chambers et al., 1988
Danielsson and Sjostrom, 1999
Risio et al., 2001
Dogs (n)
10
26
131
69
Good outcome* n (%)
9 (90)
19 (73)
122 (93)
54 (78)
Mean follow-up months (range)
13 (1–48)
21 (2–55)
26 ⫹/⫺17 (5–73)
38 ⫹/⫺22 (6–96)
Median follow-up months
6
18
N/A
36
Incontinence resolved
3/4
1/8
N/A
5/11
*Substantial improvement and resolution of any incontinence. N/A, not available.
Lumbosacral disease
or a large disc herniation (Slocum and Devine, 1998; Bagley, 2003).
distal cortex in order to avoid damage to the lumbosacral trunk of the sciatic nerve (1.8, 10.15).
COMPLICATIONS
Early postoperative
These have been reviewed extensively after lumbar disc surgery in humans but have not been reviewed for dogs (Stolke et al., 1989; Fritsch et al., 1996; Gibson et al., 2002). Intraoperative, early postoperative and late complications are listed in Table 10.4.
Seroma formation is particularly likely in the lumbosacral region if there is any dead space (Oliver et al., 1978; Slocum and Devine, 1986; Chambers et al., 1988; Watt, 1991; Ness, 1994). Infection can occur, especially as some dogs have established urinary tract infection (UTI) (Oliver et al., 1978). Occasionally, a dog may be unable to void after surgery because of increased sympathetic tone (10.3), analogous to that seen after sacrocaudal injury (see page 351). If pain persists after surgery this suggests implant failure, residual compression or possibly instability (10.53).
Intraoperative The surgeon must be vigilant while using a scalpel in the vertebral canal to fenestrate the L7/S1 disc, and also while using a bur to undercut a facet, to avoid inadvertent damage to the cauda equina. In some cases an AO drill guard can be used to protect neural tissues from the bur. The base of the L7 articular process must not be thinned excessively (10.41, 10.42), especially when foraminal decompression is performed. In addition, over-tightening must be avoided when placing a screw across the L7/S1 facet or the facet could fracture. When placing screws, the drill bit, tap and screw must not extend beyond the
Table 10.4 Intraoperative, early postoperative and late postoperative complications
Intraoperative Iatrogenic nerve root injury Fracture of L7 articular facet Poor implant position (10.16) Spinal nerve injury Inadequate decompression
Early postoperative
Late postoperative
Implant failure Instability (10.16) Dorsal seroma Infection Increased sphincter tone
Implant failure Recurrence of signs (10.53) Dural herniation Scar formation Infection (10.45)
Late postoperative Implant loosening or failure is always a risk after instrumented fusion (Fox et al., 1996; Gibson et al., 2002). It is usually due to poor implant selection or suboptimal technique (10.16). Sometimes implant failure does not cause clinical problems, especially if physiological fusion progresses fast enough (Slocum and Devine, 1986; Bagley, 2003) (10.14, 10.53). Postoperative imaging for recurrent lumbosacral pain may reveal residual disc material or fibrous scar, especially when intravenous contrast is used (Jones et al., 1999, 2000). Postoperative scarring has been reported in several studies (Adams et al., 1995; Danielsson and Sjostrom, 1999; Risio et al., 2001). Herniation of the dural sac through the laminectomy defect has also been reported; this may be secondary to tearing of the dura (Lang, 1988; Markwalder, 1993; Fox et al., 1996). Patient age and operative time are predictors of complications in humans undergoing lumbar disc surgery (Stolke et al., 1989). Additional reasons for failure to improve include surgery done at the wrong site, insufficient
10.15 Nine-year-old Pug that underwent dorsal laminectomy and fixation fusion using 2.7 mm screws and a bone graft. These screws are actually still contained within bone because of the contour of the ventral sacrum (1.28B, 10.46, 10.50, 10.54). However, if they were any longer there would be risk of damaging the proximal sciatic nerve, which lies in very close proximity to the ventral surface of the sacroiliac joint. Same dog as in 10.47. A
B
191
192
Small Animal Spinal Disorders
10.16 A: Eight week follow-up radiograph reveals implant failure due to poor technique. The screws were put in at too steep an angle in the sagittal plane. Immediate postoperative radiographs and fluoroscopy showed the screws to be in bone and not the disc space; the dog was confined in the hope that fusion would occur before implant failure. One screw has now backed out (10.14, 10.15, 10.53). B: 3D reconstruction shows the screws also do not diverge sufficiently from each other in a transverse plane (10.50). A
B
decompression or removal of disc material, a second disc lesion, extravertebral compression or nerve root trauma (Fritsch et al., 1996; Janssens et al., 2000). Facet fracture can cause persistent signs after dorsal laminectomy. It can occur after excessive thinning of the base of the L7 facet at laminectomy or from overtightening of a screw (Adams et al., 1995). Repeat imaging is recommended when postoperative problems develop as very good outcomes have been reported after a second surgery (Danielsson and Sjostrom, 1999; Jones et al., 2000; Risio et al., 2001).
POSTOPERATIVE CARE
(see Chapter 15)
Strict rest is enforced for 3 months following surgery, followed by a gradual return to fitness over a further 2 months. Dogs that are allowed to move freely too soon after surgery are at risk of making a poor recovery (de Risio et al., 2001). For working and athletic dogs, an additional month of gradual transition to full work is recommended (Sjostrum, 2003). After fixation–fusion the dog is confined for at least 6–8 weeks or until there is radiographic evidence of fusion. Physical therapy, leash walking and swimming are used with a gradual return to normal activity over a further 2–3 months (Bagley, 2003). Long-term non-steroidal anti-inflammatory drugs (NSAIDs) have been suggested to reduce postoperative scarring and improve outcome but this is unsubstantiated (Janssens et al., 2000).
PROGNOSIS Laminectomy provides rapid relief of pain in most dogs. Similarly, lameness and mild neurological deficits usually improve rapidly. More severe deficits probably carry a less favorable prognosis (Denny et al., 1982; Chambers et al., 1988; Chambers, 1989). Few studies document long-term follow-up after lumbosacral surgery. Studies with mean follow-up periods beyond 1 year show overall success rates from 73 to 93% (Table 10.3). The presence
of an osteochondritis dissecans (OCD) lesion does not appear to affect prognosis after surgery (Hanna, 2001). The prognosis for working dogs is similar with 67 of 88 (76%) returning to normal duties after dorsal decompression (Danielsson and Sjostrom, 1999; Jones et al., 2000). Some of these working dogs had also undergone foraminal decompression or facetectomy (Jones et al., 2000). The presence of fecal incontinence, and both the presence and the duration of urinary incontinence prior to surgery, are negative prognostic factors in dogs with lumbosacral disease. Incontinence for longer than 6 weeks carries a guarded prognosis (Risio et al., 2001). Success rates after surgery are known to decline over time in humans and this also seems to be true in dogs (Fox et al., 1996; Javid and Hadar, 1998; Janssens et al., 2000). Few specific results are available for either foraminal decompression or facetectomy. Foraminal decompression in 12 dogs did not improve outcome compared to 15 dogs in which laminectomy alone was performed (de Risio et al., 2001). Facetectomy gave good long-term results in 15/15 dogs but many were of small breeds (Tarvin and Prata, 1980). Nine of 11 larger-breed dogs improved after facet removal but were often left with residual signs (Denny et al., 1982; Watt, 1991; Ness, 1994). Repeat surgery is indicated for dogs with a poor initial outcome or that suffer a recurrence of signs following dorsal laminectomy. Good results were reported for four of six dogs undergoing a second surgery; with scar tissue being the most common finding on re-operation (Danielsson and Sjostrom, 1999; de Risio et al., 2001). There are only two limited studies with long-term follow-up after fixation–fusion. All eight dogs in one study and five of five in another had good outcomes (Slocum and Devine, 1986; Meheust, 2000). Although one study in humans reported better results after fusion than after decompression alone, a large meta-analysis of all available data in people found that fusion did not improve outcome and was associated with a higher complication rate (Fox et al., 1996; Gibson et al., 2002).
Lumbosacral disease
Key issues for future investigation 1. What is the significance of instability and how should it be assessed (Fox et al., 1996; Gibson et al., 2002) (10.53)? 2. What is the role of facetectomy and should it be followed by stabilization (10.45)? 3. Is routine dorsal anulectomy of benefit or does it increase collapse of the lumbosacral space (Danielsson and Sjostrom, 1999; Janssens et al., 2000; de Risio et al., 2001)? 4. Does grafting at the disc space promote fusion (page 328) (Auger et al., 2000)? 5. Why is epidural fibrosis so common and is it a consequence of multilevel compression (Jayson, 1992; Porter, 1996)? 6. How should lesions at more than one location be dealt with; that is, should decompression be extended over L6/7 for the dog in 10.17A? 7. What are the long-term outcomes after fixation–fusion? 8. When is fusion indicated and when is decompression indicated? 10.17 A: This dog has a lesion at L7/S1 and another at L6/7. B: This dog has an isolated L7/S1 lesion. Several studies have emphasized the role of multilevel compression (Jayson, 1992; Markwalder, 1993; Jones et al., 1996b; Porter, 1996).
A
B
REFERENCES Adams, W.H., Daniel, G.B., Pardo, A.D., Selcer, R.R. (1995) Magnetic resonance imaging of the caudal lumbar and lumbosacral spine in 13 dogs (1990–1993). Veterinary Radiology and Ultrasound 36, 3–13. Auger, J., Dupuis, J., Quesnel, A., Beauregard, G. (2000) Surgical treatment of lumbosacral instability caused by discospondylitis in four dogs. Veterinary Surgery 29, 70–80. Bagley, R.S. (2003) Surgical stabilization of the lumbosacral joint. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1238–1243. Philadelphia: Elsevier Science. Barthez, P.Y., Morgan, J.P., Lipsitz, D. (1994) Discography and epidurography for evaluation of the lumbosacral junction in dogs with cauda equina syndrome. Veterinary Radiology and Ultrasound 35, 152–157. Brant-Zawadzki, M.N., Jensen, M.C., Obuchowski, N., Ross, J.S., Modic, M.T. (1995) Interobserver and intraobserver variability in interpretation of lumbar disc abnormalities. A comparison of two nomenclatures. Spine 20, 1257–1263; discussion 1264. Burger, R., Lang, J. (1993) Kinematic study of the lumbar and lumbosacral spine in the German Shepherd Dog. Part 2. Results of experiments. Schweizer Archiv fur Tierheilkunde 135, 35–43. Chambers, J.N. (1989) Lumbosacral stenosis. Confusion, surgery, prognosis. Veterinary Medicine Report 1, 228–229. Chambers, J.N., Barbera, A., Selcer, B.A. (1988) Results of treatment of degenerative lumbosacral stenosis in dogs by exploration and excision. Veterinary Comparative Orthopaedics and Traumatology 3, 130–133.
Chambers, J.N., Selcer, B.A., Sullivan, S.A., Coates, J.R. (1997) Diagnosis of lateralized lumbosacral disk herniation with magnetic resonance imaging. Journal of the American Animal Hospital Association 33, 296–299. Coates, J.R. (1999) Urethral dyssynergia in lumbosacral syndrome. American College of Veterinary Internal Medicine 1999; 299–301. Cuddon, P.A. (2002) Electrophysiology in neuromuscular disease. Veterinary Clinics of North America, Small Animal Practice 32, 31–62. Cuddon, P.A., Murray, M., Kraus, K. (2003) Electrodiagnosis. In: D. Slatter (ed.) Textbook of Small Animal Surgery, 3rd edn, 1108–1117. Philadelphia: Elsevier Science. Danielsson, F., Sjostrom, L. (1999) Surgical treatment of degenerative lumbosacral stenosis in dogs. Veterinary Surgery 28, 91–98. Denny, H.R., Gibbs, C., Holt, P.E. (1982) The diagnosis and treatment of cauda equina lesions in the dog. Journal of Small Animal Practice 23, 425–443. de Risio, L., Sharp, N.J.H., Olby, N.J., Munana, K.R., Thomas, W.B. (2001) Predictors of outcome after dorsal decompressive laminectomy for degenerative lumbosacral stenosis in dogs: 69 cases (1987–1997). Journal of the American Veterinary Medical Association 219, 624–628. Deyo, R. (1994) Magnetic resonance imaging of the lumbar spine. Terrific test or tar baby? New England Journal of Medicine 331, 115–116. Ebraheim, N.A., Haman, S.P., Xu, R., Stanescu, S., Yeasting, R.A. (2000) The lumbosacral nerves in relation to dorsal S1 screw placement and their locations on plain radiographs. Orthopedics 23, 245–247. Fingeroth, J.M., Johnson, G.C., Burt, J.K., Fenner, W.R., Cain, L.S. (1989) Neuroradiographic diagnosis and surgical repair of tethered cord syndrome
193
194
Small Animal Spinal Disorders
in an English Bulldog with spina bifida and myeloschisis. Journal of the American Veterinary Medical Association 194, 1300–1302. Fletcher, T.F., Kitchell, R.L. (1966) Anatomical studies on the spinal cord segments of the dog. American Journal of Veterinary Research 27, 1759–1767. Fox, M., Onofrio, B., Hanssen, A. (1996) Clinical outcomes and radiological instability following decompressive lumbar laminectomy for degenerative spinal stenosis: a comparison of patients undergoing concomitant arthrodesis versus decompression alone. Journal of Neurosurgery 85, 793–802. Fritsch, E.W., Heisel, J., Rupp, S. (1996) The failed back surgery syndrome: reasons, intraoperative findings, and long-term results: a report of 182 operative treatments. Spine 21, 626–633. Gibson, J.N.A., Waddell, G., Grant, I.C. (2002) Surgery for degenerative lumbar spondylosis (Cochrane Review). Issue 1 ed: The Cochrane Library. http:/www.update-software.com/cochrane/ Gorman, W.F., Hodak, J.A. (1997) Herniated intervertebral disc without pain. Journal of the Oklahoma State Medical Association 90, 185–190. Hanna, F.Y. (2001) Lumbosacral osteochondrosis: radiological features and surgical management in 34 dogs. Journal of Small Animal Practice 42, 272–278. Hurov, L. (1985) Laminectomy for treatment of cauda equina syndrome in a cat. Journal of the American Veterinary Medical Association 186, 504–505. Huttmann, S., Krauss, J., Collmann, H., Sorensen, N., Roosen, K. (2001) Surgical management of tethered spinal cord in adults: report of 54 cases. Journal of Neurosurgery 95 (2 Suppl), 173–178. Janssens, L.A.A., Moens, Y., Coppens, P., Peremans, K., Vinck, H. (2000) Lumbosacral degenerative stenosis in the dog: the results of dorsal decompression with dorsal anulectomy and nuclectomy. Veterinary and Comparative Orthopaedics and Traumatology 13, 97–103. Javid, M., Hadar, E. (1998) Long-term follow-up review of patients who underwent laminectomy for lumbar stenosis: a prospective study. Journal of Neurosurgery 89, 1–7. Jayson, M.I. (1992) The role of vascular damage and fibrosis in the pathogenesis of nerve root damage. Clinical Orthopaedics and Related Research 279, 40–48. Jensen, M., Brant,–Z.M., Obuchowski, N., Modic, M., Malkasian, D., Ross, J. (1994) Magnetic resonance imaging of the lumbar spine in people without back pain. New England Journal of Medicine 331, 69–73. Jones, J.C., Inzana, K.D. (2000) Subclinical CT abnormalities in the lumbosacral spine of older large-breed dogs. Veterinary Radiology and Ultrasound 41, 19–26. Jones, J.C., Wilson, M.E., Bartels, J.E. (1994) A review of high resolution computed tomography and a proposed technique for regional examination of the canine lumbosacral spine. Veterinary Radiology and Ultrasound 35, 339–346. Jones, J.C., Cartee, R.E., Bartels, J.E. (1995a) Computed tomographic anatomy of the canine lumbosacral spine. Veterinary Radiology and Ultrasound 36, 91–99. Jones, J.C., Wright, J.C., Bartels, J.E. (1995b) Computed tomographic morphometry of the lumbosacral spine of dogs. American Journal of Veterinary Research 56, 1125–1132. Jones, J.C., Hudson, J.A., Sorjonen, D.C., Hoffman, C.E., Braund, K.G., Wright, J.C., Garrett, P.D., Bartels, J.E. (1996a) Effects of experimental nerve root compression on arterial blood flow velocity in the seventh lumbar spinal ganglion of the dog: measurement using intraoperative Doppler ultrasonography. Veterinary Radiology and Ultrasound 37, 133–140. Jones, J.C., Sorjonen, D.C., Simpson, S.T., Coates, J.R., Lenz, S.D., Hathcock, J.T., Agee, M.W., Bartels, J.E. (1996b) Comparison between computed tomographic and surgical findings in nine large-breed dogs with lumbosacral stenosis. Veterinary Radiology and Ultrasound 37, 247–256. Jones, J.C., Shires, P.K., Inzana, K.D., Sponenberg, D.P., Massicotte, C., Renberg, W., Giroux, A. (1999) Evaluation of canine lumbosacral stenosis using intravenous contrast-enhanced computed tomography. Veterinary Radiology and Ultrasound 40, 108–114. Jones, J.C., Banfield, C.M., Ward, D.L. (2000) Association between postoperative outcome and results of magnetic resonance imaging and computed tomography in working dogs with degenerative lumbosacral stenosis. Journal of the American Veterinary Medical Association 216, 1769–1774. Jones, J.C., Shires, P.K., Inzana, K.D., Mosby, A.D., Sponenberg, D.P., Lanz, O.I. (2002) Use of computed tomographic densitometry to quantify contrast enhancement of compressive soft tissues in the canine lumbosacral vertebral canal. American Journal of Veterinary Research 63, 733–737.
Kent, D.L., Haynor, D.R., Larson, E.B., Deyo, R.A. (1992) Diagnosis of lumbar spinal stenosis in adults: a metaanalysis of the accuracy of CT, MR, and myelography. American Journal of Roentgenology 158, 1135–1144. Lang, J. (1988) Flexion–extension myelography of the cauda equina. Veterinary Radiology and Ultrasound 33, 69–76. Lang, J., Hani, H., Schawalder, P. (1992) A sacral lesion resembling osteochondrosis in the German Shepherd Dog. Veterinary Radiology and Ultrasound 33, 69–76. Markwalder, T.M. (1993) Surgical management of neurogenic claudication in 100 patients with lumbar spinal stenosis due to degenerative spondylolisthesis. Acta Neurochirurgica (Wein) 120, 136–142. Mattoon, J.S., Koblik, P.D. (1993) Quantitative survey radiographic evaluation of the lumbosacral spine of normal dogs and dogs with degenerative lumbosacral stenosis. Veterinary Radiology and Ultrasound 34, 194–206. McKee, W.M. (1993) Differential diagnosis of cauda equina syndrome. In Practice 15, 243–244. McKee, W.M., Mitten, R.W., Labuc, R.H. (1990) Surgical treatment of lumbosacral discospondylitis by a distraction–fusion technique. Journal of Small Animal Practice 31, 15–20. Meheust, P. (2000) A new surgical technique for lumbosacral stabilization: arthrodesis using the pedicle screw fixation. A clinical study of 5 cases. Pratique Medicale and Chirurgicale de l’Animal de Compagnie 35, 201–207. Meheust, P., Mallet, C., Marouze, C. (2000) A new surgical technique for lumbosacral stabilization: arthrodesis using the pedicle screw fixation. Anatomical aspects. Pratique Medicale and Chirurgicale de l’Animal de Compagnie 35, 193–199. Meij, B.P. (1993) Pelvic limb lameness associated with nerve root compression: diagnosis, neurosurgery and follow-up in two dogs. Veterinary Surgery 22, 249. Meij, B.P., Voorhout, G., Wolvekamp, W.T.C. (1996) Epidural lipomatosis in a six-year-old dachshund. Veterinary Record 138, 492–495. Milette, P.C., Fontaine, S., Lepanto, L., Cardinal, E., Breton, G. (1999) Differentiating lumbar disc protrusions, disc bulges, and discs with normal contour but abnormal signal intensity. Magnetic resonance imaging with discographic correlations. Spine 24, 44–53. Morgan, J.P. (1999) Transitional lumbosacral vertebral anomaly in the dog: a radiographic study. Journal of Small Animal Practice 40, 167–172. Morgan, J.P., Hansson, K., Miyabayashi, T. (1989) Spondylosis deformans in the female Beagle dog: a radiographic study. Journal of Small Animal Practice 30, 457–460. Morgan, J.P., Bahr, A., Franti, C.E., Bailey, C.S. (1993) Lumbosacral transitional vertebrae as a predisposing cause of cauda equina syndrome in German Shepherd Dogs; 161 cases (1987–1990). Journal of the American Veterinary Medical Association 202, 1877–1882. Nardin, R.A., Patel, M.R., Gudas, T.F., Rutkove, S.B., Raynor, E.M. (1999) Electromyography and magnetic resonance imaging in the evaluation of radiculopathy. Muscle and Nerve 22, 151–155. Ness, M.G. (1994) Degenerative lumbosacral stenosis in the dog: a review of 30 cases. Journal of Small Animal Practice 35, 185–190. O’ Brien, D. (1988) Neurogenic disorders of micturition. Veterinary Clinics of North America, Small Animal Practice 18, 529–544. Ohnmeiss, D.D., Vanharanta, H., Ekholm, J. (1997) Degree of disc disruption and lower extremity pain. Spine 22, 1600–1605. Oliver, J.E., Jr., Selcer, R.R., Simpson, S. (1978) Cauda equina compression from lumbosacral malarticulation and malformation in the dog. Journal of the American Veterinary Medical Association 173, 207–214. Palmer, R.H., Chambers, J.N. (1991) Canine lumbosacral diseases. Part II. Definitive diagnosis, treatment, and prognosis. Compendium on Continuing Education for the Practicing Veterinarian 13, 213–222. Pang, W.W., Mok, M.S., Lin, M.L., Chang, D.P., Hwang, M.H. (1998) Application of spinal pain mapping in the diagnosis of low back pain— analysis of 104 cases. Acta Anaesthesiological Sinica 36, 71–74. Porter, R.W. (1996) Spinal stenosis and neurogenic claudication. Spine 21, 2046–2052. Ramirez, O., III, Thrall, D.E. (1998) A review of imaging techniques for canine cauda equina syndrome. Veterinary Radiology and Ultrasound 39, 283–296. Schmid, V., Lang, J. (1993) Measurements on the lumbosacral junction in normal dogs and those with cauda equina compression. Journal of Small Animal Practice 34, 437–442.
Lumbosacral disease
Selcer, B.A., Chambers, J.N., Schwensen, K. (1988) Epidurography as a diagnostic aid in canine lumbosacral compressive disease: 47 cases (1981–1986). Veterinary Comparative Orthopaedics and Traumatology 2, 97–103. Shamir, M., Rochkind, S., Johnston, D. (2001) Surgical treatment of tethered spinal cord syndrome in a dog with myelomeningocele. Veterinary Record 148, 755–756. Sisson, A.F., LeCouteur, R.A., Ingram, J.T., Park, R.D., Child, G. (1992) Diagnosis of cauda equina abnormalities by using electromyography, discography, and epidurography in dogs. Journal of Veterinary Internal Medicine 6, 253–263. Sjostrum, L. (2003) Decompression of lumbosacral disease. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1227–1237. Philadelphia: Elsevier Science. Slocum, B., Devine, T. (1986) L7–S1 fixation–fusion for treatment of cauda equina compression in the dog. Journal of the American Veterinary Medical Association 188, 31–35. Slocum, B., Devine, T. (1998) L7–S1 fixation–fusion technique for cauda equina syndrome. In: M.J. Bojrab (ed.), Current Techniques in Small Animal Surgery, 5th edn. 861–864. Philadelphia: Lea & Febiger.
Smith, M.E.H., Bebchuk, T. (2002) An in vitro biomechanical study of the effects of surgical modification upon the canine lumbosacral spine. Veterinary Surgery 31, 505. Stolke, D., Sollmann, W.P., Seifert, V. (1989) Intra- and postoperative complications in lumbar disc surgery. Spine 14, 56–59. Stoll, S.G. (1996) Alternative method of traction stabilization of L7–S1. San Francisco: American College of Veterinary Surgeons Symposium, 116–117. Tarvin, G., Prata, R.G. (1980) Lumbosacral stenosis in dogs. Journal of the American Veterinary Medical Association 177, 154–159. Watt, P.R. (1991) Degenerative lumbosacral stenosis in 18 dogs. Journal of Small Animal Practice 32, 125–134. Webb, A.A., Pharr, J.W., Lew, L.J., Tryon, K.A. (2001) MR imaging findings in a dog with lumbar ganglion cysts. Veterinary Radiology and Ultrasound 42, 9–13. Wheeler, S.J. (1992) Lumbosacral disease. Veterinary Clinics of North America, Small Animal Practice 22, 937–950. Wright, J.A. (1980) Spondylosis deformans of the lumbosacral joint in dogs. Journal of Small Animal Practice 21, 45–55.
PROCEDURES Dorsal laminectomy (10.18–10.39) Discectomy may or may not provide additional benefit to laminectomy although it is logical for dogs in which the disc is causing significant compression (Danielsson and Sjostrom, 1999; Janssens et al., 2000; de Risio et al., 2001). A cancellous bone graft can be placed into the disc space following curettage and this may enhance fusion (Auger et al., 2000) (10.52, 14.12).
10.18 Positioning of dog for lumbosacral surgery. Note that the pelvic limbs are drawn forward. The interarcuate space and dorsal anulus are widened by either supporting the pelvis over a 10.18 sandbag or by pulling the hocks further forward with the hips flexed (Oliver et al., 1978; Watt, 1991; Slocum and Devine, 1998; Bagley, 2003; Sjostrum, 2003). The dog’s head is to the left in all illustrations.
10.19 The important landmarks are being palpated. The surgeon’s right hand is on the cranial dorsal iliac spine. The left index finger is on the spinous process of L6. The spinous process of L7 is shorter and often cannot be palpated (10.2B, 10.24, 10.26).
10.19
195
196
Small Animal Spinal Disorders
10.20 Here the upper instrument is over the lumbosacral space, and the lower instrument on the wing of the left ilium.
10.20
10.21 The skin incision is made just lateral to the midline. It extends from the spinous process of L5 to the caudal end of the fused spinous processes of the sacrum. Here the skin and superficial fascia (a) have been incised revealing the deep lumbodorsal fascia (b). The fascia, which may be quite thick, is undermined slightly to facilitate closure (8.13) prior to incision.
10.22 The lumbodorsal fascia is incised around and between the spinous processes (a). This reveals the epaxial musculature. Here the fascia has been incised and retracted on the lower side of the illustration but is still intact on the upper side (b). The lumbodorsal fascia merges with the interspinous ligament that lies between the spinous processes. (c) Transverse process L6. (d) Transverse process L7. (e) Wing of ilium.
a
b
10.21
b a
c
a
d
10.22
10.23 Here the fascia has been retracted on both sides, revealing the epaxial muscles. The spinous processes are seen in the midline (a). The interspinous fascia between the spinous processes is thick.
a
10.23
a
e
Lumbosacral disease
10.24 The muscles are elevated from the spinous processes and retracted. The L6 (a), L7 (b), and sacral (c) spinous processes are visible. Note that the spinous process of L6 is taller than L7.
c b
a
10.24
10.25 A periosteal elevator is used to remove the muscles from the spinous processes on both sides of the vertebral column. It is useful to insert self-retaining retractors to maintain the exposure (see 10.24 for key).
c
b
a
10.25
10.26 The muscles have been retracted further. In the illustration, L6 (a), L7 (b) and the sacrum (c) have been exposed.
c b
a
10.26
10.27 Closer view of 10.26. The curette is on the ligamentum flavum. (a) L7 vertebrae. (b) Sacrum. In some dogs with lumbosacral disease the sacrum telescopes cranioventrally and the dorsal lamina of the sacrum lies adjacent to, or even under, the lamina of L7 (10.46). There is then no interarcuate space.
b
a 10.27
197
198
Small Animal Spinal Disorders
10.28 The spinous processes of L7 (shown here) and S1 are removed with rongeurs or bone cutters.
10.28
10.29 The laminectomy is made with a power bur. In this illustration the laminectomy has been commenced in L7; note the dark red cancellous bone (a). At least half of L7 is removed and most of the sacrum (Danielsson and Sjostrom, 1999). The remnant of sacral spinous process is visible (b). The laminectomy should be restricted to the lamina at this stage, preserving the articular processes.
b a
10.29
10.30 The laminectomy has been continued into both L7 and the sacrum down to the inner cortical bone. The margin between dark cancellous bone and white cortical bone is seen clearly (arrows).
10.30
Lumbosacral disease
10.31 The ligamentum flavum is removed by sharp dissection. Great care must be taken not to penetrate too deeply, risking damage to the cauda equina (Markwalder, 1993; Fox et al., 1996), as there is no bony protection overlying the neural structures at this level. In dogs with marked compression and dorsal elevation of the cauda equina it is safer to enter the vertebral canal over mid-L7 or mid-sacrum and then peel away the ligamentum flavum from either side.
10.31
10.32 Removal of the ligamentum flavum exposes the epidural fat (a) and the cauda equina (b). Epidural fat should be removed only when necessary.
a
b
10.32
10.33
10.33 The final shelf of bone is thinned with the bur.
199
200
Small Animal Spinal Disorders
10.34 It is best to bur the bone to eggshell thickness to allow easy removal. The tip of the probe is visible through the thinned bone (arrow).
10.34
10.35 The inner shelf of bone is removed carefully with rongeurs. The width of this laminectomy is greater than is ideal if screw fixation were to be performed subsequently (compare to 10.41, 10.51 and 10.52)
10.35
10.36 The laminectomy is completed with rongeurs or a curette. The cauda equina is then inspected. The S1 nerve roots (a) are usually seen adjacent to the dural tube of the cauda equina (b). The L7 nerve root lies further laterally, in the lateral recess and under the articular process (1.27, 1.28). A herniated disc may either be visible directly (10.44), or palpable by running a probe along the floor of the vertebral canal (which is flat in normal dogs) (10.38A).
a
b
10.36
10.37 Laminectomy has been completed. Part of L7 nerve root (a) is seen in the lateral recess, before exiting the intervertebral foramen. (b) S1 nerve root. (c) S2 nerve root. (d) Remainder of cauda equina.
a
10.37
b
c d
Lumbosacral disease
10.38 Ten-year-old Rottweiler that had been dribbling urine for 2 months and had lumbosacral pain, pelvic limb ataxia, and marked reduction of right-sided hock flexion with a lick granuloma over the right tarsus. A CT myelogram revealed that the terminal dural sac was displaced to the left by a large soft tissue mass. A: The mass was palpable just in front of the lumbosacral joint. B: Retraction of the right S1 nerve root revealed a huge disc extrusion (*). C: Continued retraction reveals the extent of the mass compressing the right L7 nerve root. D: After fenestration with a scalpel the mass was removed using rongeurs to leave a large crater (arrowhead), which was then covered with a thin fat graft. The clinical signs resolved slowly after surgery. Two years later the dog was pain free, continent and its hock lesion had healed.
10.38 A
B
C
D
201
202
Small Animal Spinal Disorders
10.39 Here the cauda equina is being retracted medially to allow a bulging disc to be fenestrated on one side (arrowheads). Half of the anulus fibrosus is incised carefully with a #11 blade and removed with fine rongeurs. The outermost incision is just medial to the venous plexus (Danielsson and Sjostrom, 1999). The cauda equina is then retracted to the other side and the second half of the disc is fenestrated. The nucleus pulposus is then removed with a curette. 10.39
Foraminal decompression and facetectomy (10.40–10.45) Enhancement of soft tissues within a foramen after intravenous contrast is an indication for foraminal decompression (see page 186, 190; 10.40). Densitometry can be used to assess the degree of contrast enhancement more accurately although it does not predict the nature of the enhancing tissue (Jones et al., 2002). Facetectomy can be followed by stabilization using screws and bone graft (Slocum and Devine, 1998), screws and bone cement (10.45), or screws with connecting bars (Meheust, 2000; Meheust et al., 2000). Wound closure is routine. A free fat graft, pedicle fat graft or Gelfoam may be placed over the laminectomy site. Closure must avoid leaving any dead space.
L
10.40 A: Pre-contrast, and B: post-contrast transverse CT images obtained at the level of the L7/S1 intervertebral foramina in a dog with 10.40 degenerative lumbosacral stenosis. A B Contrast enhancing tissue fills the left foramen. A small area of enhancement can also be seen on the right side of the ventral canal (Jones et al., 1999).
10.41 Reconstruction following a dorsal laminectomy. If imaging indicates nerve root compression, the laminectomy can be widened still further, initially only to the medial limit of the facet joint capsule (Danielson and Sjostrom, 1999). The intervertebral foramen can be explored with a probe; the normal L7 nerve root should move a few millimetres without undue tension.
10.41
L
Lumbosacral disease
10.42 If a foraminotomy is to be performed, the laminectomy is extended laterally on that side dorsal to the foramen. The relationship between the lumbosacral disc and the foramen is shown in 1.26 and 1.27; the foramen is cranial to the disc (10.44). The articular processes should be preserved and care taken not to weaken the base of the L7 facet when decompressing the foramen. Extending the laminectomy all the way into the foramen will isolate L7 articular process.
10.43 Undercutting of a foramen to remove osteophytes is illustrated on this transverse CT image. The level of the laminectomy is shown by white arrowheads. Additional undercutting of the L7 or S1 facet can be performed with a curette or bur, as shown by the area outlined with black arrowheads (Palmer and Chambers, 1991). Great care is taken to protect the nerve roots while drilling. The facet should not be weakened excessively, particularly if screw fixation is also to be performed.
10.42
10.43
10.44 Facetectomy may be considered if foraminal decompression does not expose enough of the L7 nerve root. Here the dural tube is displaced medially and S1 spinal nerve laterally by an extruded L7/S1 disc (arrow). The articular process of L7 has been removed to expose cartilage of the sacral facet (asterisk). The dorsal root ganglion of L7 nerve root can be seen 10.44 within the foramen (arrowhead) although it is covered partly by remnants of joint capsule. More bone may need to be removed cranially to decompress the nerve root within the lateral recess as the foramen is located some distance cranial to the disc space (1.27).
203
204
Small Animal Spinal Disorders
10.45 This dog had shown intermittent lameness for one year with no neurological deficits. A: After facetectomy, screws were placed into the L7 pedicle; this is easier if a dorsal laminectomy 10.45 is done first so that both lateral and medial walls of A B the pedicle can be palpated. B: The cement bridge is visible clearly between L7 and S1; a single screw has also been placed across the other L7/S1 facet joint (10.50–10.53). Six weeks after surgery it was much improved; at 11 months it developed a draining tract from the wound but was otherwise normal.
Dorsal fixation–fusion (10.46–10.54) Indications for fixation–fusion include telescoping of the sacrum under L7 and marked instability (10.46). Facet subluxation must be reduced prior to fixation by either positioning the dog as in 10.18 or by using a laminectomy spreader (10.49, 13.64). Initial reduction can also be obtained using bone holding forceps or towel clamps (10.47).
10.46 This dog initially underwent dorsal laminectomy for lumbosacral pain (10.6A,B, 10.13). Preoperative 3D reconstruction of CT scans performed with the dog in A: flexion, and B: extension revealed marked compression of the cauda equina (10.6A,B, 10.11). The L7 articular facets and ventral osteophytes on L7 are seen in a constant position but the sacral lamina has telescoped under the roof of L7 (arrowhead). The disc is also bulging upwards and there is severe foraminal narrowing (10.6). Postoperative images are shown in 10.53.
10.46 A
B
Lumbosacral disease
10.47 A: Collapse of the dorsal lumbosacral articulation with telescoping of the sacrum under the L7 lamina. B: The telescoping is reduced and the distance between the towel clamps is increased markedly as the L/S articulation is distracted. Same dog as in 10.15.
10.47 A
B
205
206
Small Animal Spinal Disorders
10.48 The degree of telescoping of the sacrum relative to L7 increases from A to D as shown in these 3D reconstructions from four different dogs (dog A is normal). The black line is drawn at the cranial edge of each sacral lamina. The space between the lamina of L7 and S1 decreases and the facet overlap increases. The best guide to the degree of subluxation is provided by opening the joint capsule (10.49).
10.48 A
B
C
D
Lumbosacral disease
10.49 A: Subluxation of facet joints is evident by the amount of articular cartilage exposed on the sacral facet (arrow). B: Reduction with a laminectomy spreader anchored at the laminectomy edges has brought the L7 facet back to cover the cartilage of the sacral facet (arrow). The cartilage on both sides of this joint is removed with a small curette prior to screw fixation. Bone debris or graft is then packed into and around the joint (Bagley, 2003).
10.49 A
B
10.50 Transverse CT image of the LS articulation from a dog after screw fixation to show the approximate angles for screw insertion (13.57). The ideal screw angle is between 30 and 45° from the sagittal plane (Bagley, 2003). Compare these angles to the suboptimal angles shown in 10.16. Ideally, screws should not compromise the sacroiliac joint or cause injury to the lumbosacral trunk of the sciatic nerve (Ebraheim et al., 2000).
10.50
10.51 Screw entry points are shown (*) in a 3D CT reconstruction from a dog that had undergone previous dorsal laminectomy (10.13). The entry point should be equal in distance from each side and from the tip of the facet. The tip of the 10.51 L7 spinous process can also be used as a guide to the angle of insertion (McKee et al., 1990). Adequate reduction of the facets is assessed by observing joint congruity (10.49).
207
208
Small Animal Spinal Disorders
10.52 Screw placement. A: One screw is in position; the pilot hole is being drilled for the second. A dorsal laminectomy has already been performed. Ideally the facet nearest the surgeon (L7) is over-drilled. If the facet is too narrow, this step can be skipped rather than risk facet fracture. B: The hole is then countersunk. Washers can also be used (10.14, 10.15). C: The hole is measured so that the screw will only just exit the sacrum (10.15); the hole is then tapped. The tip of the L7 spinous process is shown (arrowhead). D: Cancellous bone from the wing of the ilium is packed over the screw; the L7 spinous process can also be used as a source of corticocancellous graft (Stoll, 1996). Normally, the laminectomy would be covered first with a fat graft but it has been left exposed here for orientation.
10.52 A
B
C
D
Lumbosacral disease
10.53 This dog had a poor response to dorsal laminectomy done 1 year previously (same dog as in 10.6A, B, 10.11, 10.13, 10.46). It had episodes of severe pain; fluoroscopy 10.53 suggested marked lumB bosacral instability. A: After A fusion the dog was free of pain for 4 months but then developed degenerative myelopathy. B: Necropsy confirmed degenerative myelopathy; it also revealed that the site had fused and was immobile despite lysis around one screw (arrow) and bending of the other (arrowhead).
10.54 3D reconstruction of a CT scan made after fusion using bone screws and graft (same dog as 10.53). Areas of new bone production are shown (white arrows). The black arrowhead indicates the position of the sacral spinous process; above it is the bent spinous process of L7 (white arrowhead).
10.54
209
Cervical spondylomyelopathy
Clinical signs
Chapter
11
• • • •
Ligamentous hypertrophy. Joint capsule proliferation or cyst formation. Osteophyte production. Disc herniation. Vertebral malformation and canal stenosis may be present at birth in the Doberman, suggesting either a congenital or inherited disorder (Burbidge, 2001; Drost et al., 2002) (11.1). An early onset of clinical signs is most common in giant-breed dogs. Most other breeds show clinical signs from middle age onwards. These usually develop due to
212
Diagnosis 212 Radiography 213 Presurgical evaluation 216 Hypothyroidism 217 Bleeding disorders 217 Treatment 217 Non-surgical treatment 218 Surgical treatment 218 Complications 224 Intraoperative complications 224 Early postoperative complications 226 Late postoperative complications 228 Postoperative care 228 Prognosis
229
Key issues for future investigation References
229
229
Procedures 232 Paramedian approach to the ventral neck 232 Ventral decompression 233 Vertebral distraction 235 Cement plug 237 Metal implant and bone cement method 239 Dorsal decompression 241 Laminoplasty 245
Cervical spondylomyelopathy (CSM or ‘Wobbler’ syndrome) is predominantly a syndrome of large- and giant-breed dogs, particularly Doberman pinschers and Great Danes. The cause of this disorder is multifactorial. Important contributing factors are: • Vertebral canal malformation, stenosis, or both. • Vertebral instability.
11.1 CT images of C6 and C7 vertebrae from neonatal puppies A: Doberman B: other large breed. There are significant differences in vertebral dimensions between the two breed groups: the cranial vertebral canal is relatively narrow in the Doberman group and they have vertebral body asymmetry. Changes were most marked for C7 but were also present at both C5 and C6 (Burbidge, 2001).
212
Small Animal Spinal Disorders
acquired soft tissue or osseous lesions, which are probably a consequence of low-grade instability. Compression is seen mainly at C5/6 and C6/7 in the Doberman but lesions in other breeds often affect more cranial disc spaces. Both C5/6 and C6/7 sites are at high risk of causing spinal cord compression; twenty per cent of dogs present with both C5/6 and C6/7 lesions at the time of initial diagnosis. In addition, if either one of these intervertebral spaces fuses then the other one seems even more likely to cause compression (Bruecker et al., 1989a). Any surgical procedure should therefore address all high-risk spaces in an individual animal (Dixon et al., 1996; Hilibrand et al., 1999).
CLINICAL SIGNS The most common presentation is a gait disturbance, which is most severe in the pelvic limbs and ranges from
mild ataxia to marked paresis and dysmetria. Cervical hyperesthesia, guarding of the neck, pain on manipulation of the prominent transverse process of C6 (4.6), or a low carriage of the head may also be seen (11.2). Lameness and shoulder muscle atrophy in one or both thoracic limbs, or pain when traction is applied to the limb (root signature), suggest nerve root compression (see Chapter 7, page 93). Neurological deficits localize to either the C1–C5 region or to the cranial portion of C6–T2. Dogs with C1–C5 signs often show a ‘floating’ thoracic limb gait in addition to tetraparesis and dysmetria (Baum et al., 1992). Dogs with C6–T2 signs have more of a short-stepping thoracic limb gait. This causes the thoracic and pelvic limbs to be advanced at different rates and produces a characteristic ‘disconnected’ gait. The short thoracic limb steps may be due to either increased thoracic limb muscle tone, which is an upper motor neuron (UMN) effect on the elbow and carpal extensor muscles, lower motor neuron (LMN) weakness, or both (Seim and Withrow, 1982). There is often an associated LMN weakness of the elbow flexors resulting in a weak withdrawal reflex. Pelvic limb muscle tone and reflexes are normal or exaggerated. Tetraplegia is uncommon but when present (15.7) the dog must be assessed for hypoventilation (6.1, 7.11, 11.10).
DIAGNOSIS
11.2 Doberman with CSM. Note the broad-based pelvic limb stance and lowered head position.
Although the gait abnormalities are often suggestive of CSM, careful evaluation of the history and thorough physical and neurological examinations are important to help rule out the differential diagnoses listed in Table 11.1 (see also Box 7.2 and Chapter 14). The clinician should also consider the possibility of a
Table 11.1 Differential diagnoses for cervical spondylomyelopathy (CSM) Degenerative disorders
Anomalous/developmental disorders
Neoplastic disorders
Inflammatory/infectious disorders
Ischemic disorders
Degenerative myelopathy Leukodystrophies Synovial cyst(s) Orthopedic disease Disc extrusion Cervical fibrotic stenosis
Congenital vertebral malformations Atlantoaxial subluxation Multiple cartilaginous exostoses Tumoral calcinosis (calcinosis circumscripta) Meningocele/ meningomyelocele Spinal dysraphism Hydromyelia, syringomyelia, or both Pilonidal sinus (dermoid sinus) Epidermoid cyst Spinal arachnoid cyst
Meningioma, Nerve sheath tumor Other tumors
Discospondylitis Epidural abscess Meningomyelitis
Fibrocartilaginous embolic myelopathy (FCE) Spinal cord hematoma Ischemic neuromyopathy
Consider also thoracolumbar lesion; brainstem lesion.
Cervical spondylomyelopathy
thoracolumbar lesion in a dog with signs restricted to the pelvic limbs (2.8, 4.30, 4.31). Brainstem lesions can also mimic cervical spinal cord lesions on rare occasions (2.23). A further diagnostic challenge is the potential for over-diagnosis using an MRI without any reference to the clinical signs. In humans a disc herniation and even spinal cord compression appear in many asymptomatic people older than 30 years (Teresi et al., 1987; Hayashi et al., 1988; Gorman and Hodak, 1997). By the age of 60 years, some 12% of asymptomatic individuals show compression of the spinal cord on MRI (Gorman and Hodak, 1997). Incidental, mild disc lesions also occur in the cervical region of large-breed dogs. Prior to anesthesia, several potential complicating conditions should be ruled out: • Chronic active hepatitis of Doberman pinschers is usually evident on serum biochemistry. • Cardiomyopathy is seen in many large- and giant-breed dogs, and is often fatal within 6 months of diagnosis, particularly in male dogs. Even a subtle arrhythmia should not be discounted (Calvert et al., 1996). An electrocardiogram (ECG) and echocardiogram should be performed, realizing that this still cannot rule out occult cardiomyopathy. A 24-h ECG may also be of value as a screening test (Calvert and Wall, 2001). • Hypothyroidism. • Bleeding disorders (page 217).
Radiography SURVEY RADIOGRAPHY Survey radiographs are useful to rule out potential differential diagnoses but are not definitive for CSM. Severe articular facet changes or vertebral body malformation do raise the index of suspicion for CSM, especially in giant-breed dogs (11.3).
flexion or extension. Lesions that did not improve were termed static and those that did improve were termed dynamic (Seim and Withrow, 1982). Dynamic lesions can be subdivided further; first by whether or not they respond to traction (traction-responsive or traction non-responsive) and then by whether or not they change during flexion and extension (positional). This subdivision of lesion types helps the surgeon to decide on the best procedure to perform. Although there may be some overlap, lesions fall primarily into one of three basic types (11.4–11.6). Compressive lesions that improve with traction are termed tractionresponsive (McKee and Sharp, 2003) (11.4). Traction usually decreases spinal cord compression caused by anulus fibrosus or ligamentous tissue and therefore increases dural tube diameter (Rusbridge et al., 1998). Such traction-responsive lesions would be expected to benefit from a distraction-stabilization surgery. Such lesions have also been called ‘dynamic’ (Seim and Withrow, 1982; Wheeler and Sharp, 1994).
Traction-responsive lesions
Some compressive lesions do not improve with traction (11.5). Such lesions are usually caused by new bone formation or extrusion of nucleus pulposus. They are most likely to respond either to a ventral slot or dorsal decompression. Such lesions have also been called ‘static’ (Seim and Withrow, 1982; Wheeler and Sharp, 1994). They are less common than traction-responsive lesions (Seim and Withrow, 1982). Differentiation between lesions that do or do not respond to traction can be subjective at times, but this is still the most logical means of deciding on the best
Traction non-responsive lesions
MYELOGRAPHY This is the standard means of confirming a diagnosis of CSM and has the advantage that the lesion can be observed readily in different positions of the spine. Contrast should be first concentrated at the lesion by positioning the site of interest at the low point of the spine for several minutes (McKee et al., 2000) (4.32). Lateral, ventrodorsal (for cranial cervical vertebrae), dorsoventral (for caudal cervical vertebrae), flexion and traction views should then be taken (Rendano and Smith, 1981; Lamb, 1995). The lesion may alter appearance as the relative positions of adjacent vertebrae are changed. Formerly, lesions were categorized based on whether or not compression changed in the ‘stressed’ positions of traction,
11.3 Survey radiographs must be taken under general anesthesia to be of diagnostic quality. This 6-year-old tetraparetic Great Dane has multiple sites of periarticular new bone around the facet joints, especially at C5/6 and C6/7. The CT scan of this dog is shown in 11.7.
213
214
Small Animal Spinal Disorders
11.4 Traction-responsive type of dynamic lesion. A: Ventral spinal cord compression at C6/7 due to redundant anulus fibrosus in a middle-aged Doberman. B: Same dog but with traction applied to the vertebral column. There is marked relief of spinal cord compression and widening of the dural tube dorsally.
A
A
B
B
C
11.5 Traction non-responsive type of static lesion in a middle-aged Doberman (same dog as 11.30B). A: Ventral spinal cord compression at C6/7 due to disc herniation. B: Same dog but with traction applied to the vertebral column. C: CT myelogram of the same dog to show extruded, mineralized disc material at C6/7 (arrow).
type of surgical procedure (Rusbridge et al., 1998). Some lesions will also have a positional component and then a judgement must be made about which is most significant.
Positional lesions Some dogs have minimal spinal cord compression in a neutral neck position and therefore show little change with traction. Nevertheless they may show significant compression in different neck positions (11.6). In such dogs the degree of compression changes specifically as the neck is moved between flexed, neutral and gently extended positions (see below) (Dueland et al., 1973; McKee and Sharp, 2003). These types of dynamic lesion are best termed ‘positional’ as they are worsened by positions that reflect normal neck motion. In dogs with positional lesions the spinal cord is probably suffering repeated, minor trauma during everyday life. Such dogs may therefore benefit from a stabilization procedure to prevent this repeated, low-grade injury. Positional studies are not without risk. The extension view can cause severe exacerbation of spinal cord compression and should be undertaken with extreme care
and preferably only under fluoroscopy (11.9, 11.10). If this is not available, the dog should be positioned in mild (not extreme) extension for as short a time as possible. If the dog is not being ventilated, its ability to breathe spontaneously must be checked carefully as respiratory arrest can occur in this position (Seim and Withrow, 1982). Use of the extension view should be restricted to dogs with suspected positional lesions or possibly to identify subclinical lesions. Advanced imaging techniques are preferable to diagnose subclinical lesions when conventional myelography does not provide sufficient information (4.42, 11.7–11.10).
CT-MYELOGRAPHY Ideally all conventional myelograms for CSM should be followed by a CT scan (Sharp et al., 1992). The CT provides excellent bone imaging (11.7) and when used with contrast it also gives a good transverse image of the spinal cord (11.8, 11.10). This information can improve surgical planning and may provide prognostic information by detecting spinal cord atrophy (Chambers and Betts, 1977; Sharp et al., 1995). Sagittal or three-dimensional reconstructions may also prove useful (11.8B). Finally,
Cervical spondylomyelopathy
11.6 Dynamic, positional lesions in a 9-month-old Doberman. A: Neutral view showing minimal spinal cord compression. B: There is no change in appearance with traction. C: In flexion the craniodorsal aspects of both C4 and C5 vertebral bodies cause mild attenuation of the ventral subarachnoid space. D: There is moderate attenuation of the dorsal subarachnoid space over C3/4 and C4/5 intervertebral spaces in extension (arrows) (Dueland et al., 1973).
11.7 CT scan at C5/6 from the dog shown in 11.3. The conventional myelogram was of poor quality. A: Transverse CT image after myelography reveals dramatic overgrowth of the right articular facet joint (arrowhead). B: 3D reconstruction demonstrates the overgrown right facets and also shows marked impingement of both right and left facet joints into the vertebral canal (arrows) (Massicotte et al., 1999).
L
A
L
B
11.8 A: Transverse CT image at C5/6 in a 2-year-old Mastiff reveals marked, asymmetrical, extradural compression caused by a soft tissue mass adjacent to the left articular facet joint (same dog as in 11.56, 11.57). B: 3D reconstruction suggests that this is cystic in nature (arrow). This was suspected to be a synovial cyst (Levitski et al., 1999; Dickinson et al., 2001; Lipsitz et al., 2001). See page 320.
215
216
Small Animal Spinal Disorders
L
A
B
L
A
11.9 Myelogram of a 5-year-old Rottweiler with tetraparesis. The dog had deteriorated after a ventral slot at C6/7 1 week previously. A: The lateral view shows no compression. B: The ventrodorsal view shows moderate extradural compression at C6/7 on the right side. The radiopaque object over the spinal cord is a pet identification chip. CT was performed with the dog first in a neutral and then an extended position (11.10).
L
11.10 A: Same dog as in 11.9. CT image with the neck in a neutral position showing moderate attenuation of the subarachnoid space over C6/7. The central defect in the vertebral body of C6 is due to the ventral slot. B: CT image from the same dog with the neck in an extended position. Spinal cord compression is much worse; this positional effect also led to an unanticipated complication (see below).
B
postoperative imaging can be used to gauge the effectiveness of a surgical procedure (4.25, 7.5, 7.8, 7.9, 11.10, 11.12, 11.56, 11.57). CT can also be used to study positional lesions. However, keeping the neck in an extended position for the duration of the scan can cause severe injury (11.10). The dog shown in 11.9 and 11.10 suffered seizures during recovery, which were very difficult to control. He was put on a ventilator but became profoundly hypotensive and died 12 h after the study. The cause of death was not determined but was probably complicated by sympathetic blockade secondary to severe cervical spinal cord compression (see page 82) (Rosenbluth and Meirowsky, 1953; Seim and Withrow, 1982; Clark, 1986). It is likely that maintaining the neck in an extended position for the time that it took to complete the second CT scan caused severe spinal cord injury; this was probably exacerbated by the seizures.
disadvantages are the lack of general availability and, depending on the machine, the time required to complete a study. Dynamic studies are possible but do require that the patient be repositioned and then re-imaged (10.12, 11.10). However, dynamic studies are less important if the cement plug distraction-stabilization technique is to be used because this surgery can be used for nearly all types of lesion (Algorithms 11.1A,B). It is then only important for MRI to differentiate a disc extrusion from other lesion types (4.42, 4.43). A temporary worsening of neurological deficits may occur after imaging dogs with CSM, especially after using subarachnoid contrast. Imaging ideally should be undertaken 48 h before elective surgery in order to allow the patient to recover. This period also gives ample opportunity to decide on treatment options, to discuss these with the owner and to perform a thorough presurgical evaluation.
MR IMAGING
PRESURGICAL EVALUATION
This is the technique of choice for imaging humans with degenerative diseases of the cervical spine. It is non-invasive and also provides superior soft tissue resolution to CT-myelography (Lipsitz et al., 2001). The
During the recovery period following imaging, medical problems that may complicate the situation should be assessed. Post-myelographic seizures may be more common in Doberman pinschers with CSM than in
Cervical spondylomyelopathy
other breeds (Lewis and Hosgood, 1992). This highlights the need for constant monitoring during recovery, with diazepam on hand to control any seizures as they could also exacerbate the spinal cord injury (Lipsitz et al., 2001) (11.10). Care should be taken to maintain adequate systemic blood pressure and spinal cord perfusion throughout the procedure (see page 86). Dogs with chronic spinal cord lesions are particularly susceptible to postoperative deterioration, especially following dorsal laminectomy or ventral slot (Kohno et al., 1997; Rusbridge et al., 1998; de Risio et al., 2002). Preoperative methylprednisolone sodium succinate (MPSS) may be useful but the risks and benefits of this strategy remain unproven (Olby, 1999; Pietila et al., 2000). Preoperative vitamin E may be a useful alternative for elective surgeries (see page 85).
Hypothyroidism Doberman pinschers and Great Danes are predisposed to hypothyroidism although this disorder is probably over-diagnosed. Lethargy, muscle weakness and peripheral neuropathy may occur, all of which are undesirable in a surgical candidate. The best diagnostic test is thyroid-stimulating hormone (TSH) stimulation but this is not widely available. In its absence a high endogenous TSH level with a low total T4 is highly suggestive of hypothyroidism if the history and clinical signs are compatible. Glucocorticoids, phenobarbitone and some non-steroidal anti-inflammatory drugs (NSAIDs) may interfere with thyroid function testing (Gieger et al., 2000). Hypothyroid dogs should probably receive supplementation for at least 48 h prior to surgery although hypothyroidism does not induce VW disease as was once thought, nor does it cause defects in primary hemostasis (Panciera and Johnson, 1996).
Bleeding disorders It has been estimated that 16% of Doberman pinschers in the USA have a bleeding tendency related to VW disease (Dodds, 1989). Bleeding from the internal vertebral venous plexus is a potential problem during ventral decompression and can be almost impossible to arrest unless the dog has normal hemostatic abilities. After a ventral approach to the neck, the inability to close dead space under the strap muscles can lead to hematoma formation several days after surgery (15.40). The easiest way to test an animal’s VW status is to perform a standardized bleeding time test (11.11). A stopwatch is started just as the incisions are made. Blood is blotted at 5-s intervals using filter paper, taking care not to touch the incision itself. The mean buccal mucosal bleeding time for normal dogs is
11.11 Illustration of a buccal mucosal bleeding time being performed in a dog. The gauze strip has been used to keep the lip turned over and cause slight venous engorgement. A twoblade, spring-loaded device (Simplate II, Organon Teknika, General Diagnostics, Cambridge, UK) has been used to make two 6-mm long by 1-mm deep incisions in the upper lip mucosa. Any obvious blood vessels should be avoided.
2.62 min, with a range of 1.7–4.2 min (Jergens et al., 1987). The cuticle bleeding time is harder to standardize than the buccal mucosa bleeding time, but should identify severe bleeding disorders. A cut is made using a guillotine nail clipper at the apex of the cuticle. In normal dogs, bleeding stops within 8 min, but occasional normal animals will bleed for up to 12 min (Giles et al., 1982; Stokol and Parry, 1998). Dobermans with prolonged bleeding times, or those with known VW disease, can be given desmopressin (DDAVP®—Rhone-Poulenc Rorer) (1.0 g/kg SQ) immediately prior to surgery (Callan and Giger, 2002; Kraus et al., 1989). Cryoprecipitate is the optimal therapy, although fresh or frozen plasma (10 ml/kg of plasma, taken 30–60 min after the donor has been given desmopressin) may also be useful (Kraus et al., 1989; Ching et al., 1994; Stokol and Parry, 1998). As a precaution, dogs at known risk can be cross-matched if they are to undergo ventral decompression. There does not appear to be any benefit in giving thyroid hormone supplementation to euthyroid dogs with prolonged bleeding times as this does not increase plasma VW factor antigen (Panciera and Johnson, 1994).
TREATMENT The decision on the best way to treat each patient is based on the presenting history, neurological status, results of imaging, and on the owner’s expectations and their ability to undertake any necessary aftercare. Most dogs that show neurological deficits are surgical candidates, but consideration will also be given here to nonsurgical treatment.
217
218
Small Animal Spinal Disorders
Table 11.2 Reported results for surgical treatment of cervical spondylomyelopathy (CSM)
Ventral slot
Total
Metal and cement
Cement plug
Dorsal laminectomy
Total
Dogs (n)
182
273
147
59
411
225
188
146
214
53
Successes (as %)
14 (78)
18 (67)
13 (91)
45 (76)
37 (90)
18 (82)
14 (78)
11 (79)
20 (93)
45 (85)
Follow-up (months) (Mean)
10–60 (29)
6–48 N/A
6–40 N/A
3–50 (20)
3–60 (21)
1.5–53 (17)
N/A N/A
7–108 (38)
No. of repeat episodes (as %)
4/14 (28)
N/A N/A
5/13 (38)
8/37 (22)
2/18 (11)
0/14 (0)
N/A N/A
4/21 (19)
Months to repeat episode (Mean)
12–60 (32)
N/A N/A
16–33 (23)
5–42 N/A
8–33 N/A
N/A N/A
N/A N/A
4–48 (23)
9/27 (33)
4/35 (11)
1
Bruecker et al., 1989a; 2 Bruecker et al., 1989b; 3 Chambers et al., 1986; 4 de Risio et al., 2002; 5 Dixon et al., 1996; 6 Lipsitz et al., 2001; Rusbridge et al., 1998; 8 Trotter et al., 1976. N/A, not available. 7
Non-surgical treatment Non-surgical treatment is warranted in two situations. The first situation is when a normal dog develops neurological deficits following minor trauma; these may resolve completely within a few weeks unless severe injury is sustained. The second is when a dog develops CSM before it is skeletally mature and so may benefit from correction of nutritional imbalances together with severe caloric reduction. This is analogous to the strategy used for management of young horses with ‘Wobbler’ syndrome but it is unproven in dogs (Donawick et al., 1993). For most other dogs, however, surgery is the treatment of choice as the majority probably show progressive deterioration without treatment (Jeffery and McKee, 2001). For most dogs with CSM the surgery is elective. A 2–4 week trial period of severe exercise reduction and use of a chest harness is often justified (15.11); if this fails it often emphasizes the need for surgery to the owner. Anti-inflammatory doses of prednisolone may also be used, but for short periods only and preferably on an alternate day basis (VanGundy, 1988). One additional role for corticosteroids is to use the resultant response, or lack of response, as a crude indicator of the reversibility of any neurological deficit prior to performing surgery.
Surgical treatment A large number of different surgical techniques have been proposed for CSM, with many of the authors claiming between 70 and 90% success rates (Table 11.2).
The most logical way to obtain the best overall results is to consider three basic types of surgery and to perform these for certain, relatively well-defined indications. The three types of surgery are: • Ventral decompression. • Vertebral distraction-stabilization. • Dorsal decompression. The basic indications for each procedure are summarized in Table 11.3. The main factor governing the choice of surgical procedure is the appearance of the spinal cord on imaging, particularly the traction and flexion views after myelography. Some lesions show a combination of different types of compression and then a judgement must be made as to which is the major component (see ‘Myelography’ and ‘MRI’, pages 213 and 216). Algorithms 11.1A and B are algorithms for surgical decision-making based on whether the dog has single or multiple lesions. In general: • Dogs with single, ventral lesions that do not respond to traction should undergo a ventral slot. • Single and multiple lesions that respond well to traction should undergo cement plug distractionstabilization. A ventral slot can be as good for single lesions if performed well. • Single and multiple lesions that are positional should undergo cement plug distractionstabilization. Some surgeons may prefer dorsal decompression, particularly for dorsal positional lesions or synovial cysts (page 320) (Jeffery, 1995; Dickinson et al., 2001; de Risio et al., 2002).
Cervical spondylomyelopathy
Table 11.3 General indications for surgical procedures in cervical spondylomyelopathy (CSM) Procedure
Indication(s) (see also Algorithms 11.1A,B)
Lesion(s) addressed
Ventral decompression Ventral slot
Single, ventral, traction non-responsive (static) lesion*
Disc extrusion Can be used for anulus
Ventral slot with metal and bone cement
To prevent (or to treat) disc space collapse after ventral slot decompression*
Disc extrusion and anulus
Distraction-stabilization Cement plug
Single lesions of all types except ventral, traction non-responsive lesion* Multiple lesions of all types
Metal and bone cement
Single traction-responsive (dynamic) lesion* Rescue after failed ventral slot decompression
Synthes locking plate
Single lesions—as for cement plug Rescue after failed ventral slot decompression Multiple lesions—needs further evaluation
Dorsal decompression Dorsal laminectomy
Laminoplasty (needs further evaluation)
Single or multiple, dorsal, traction non-responsive (static) lesion(s) Single or multilpe, dorsal positional lesion(s) Single traction non-responsive (static) lesion Single dorsal, positional lesion
} }
Osseous or soft tissue compression Anulus fibrosus Ligamentum flavum
Osseous vertebral canal stenosis, articular facet osteophytosis, synovial cyst(s)
*In addition to the operated site, any additional subclinical lesion, in particular at C5/6 and C6/7 interspaces, should be addressed routinely in an attempt to reduce the incidence of domino lesions (see 11.30 and page 221).
L
A
•
L
11.12 A: Preoperative CT myelogram at C5/6 in a tetraparetic Doberman. There is a soft tissue mass on the right, ventral to the spinal cord. Radiopaque material within the right foramen represents mineralized disc or new bone. A ventral slot retrieved nucleus pulposus and anulus fibrosus; the slot was filled with fat and then cancellous bone. B: Eight weeks later the spinal cord and ventral subarachnoid space have expanded. New bone occupies the ventral portion of the slot. Contrast material within the spinal cord may represent a syrinx (Faiss et al., 1990), (from Sharp et al., 1995).
B
In all dogs that show compression at only one interspace, consideration should also be given to a strategy that will prevent compression from developing at adjacent high-risk spaces in the future (see page 221).
VENTRAL DECOMPRESSION (11.28–11.30) (see Chapter 7, page 96) Ventral decompressive surgery can be very challenging in CSM because of various combinations of ventral
osteophytosis, misshapen vertebrae, and intraoperative hemorrhage (Table 7.2). Short-term deterioration is common even in dogs that have good long-term results (Rusbridge et al., 1998). This technique is best reserved for the relief of single-level spinal cord compression caused by midline, dorsal extrusion of disc material. In such lesions, a ventral slot should permit retrieval of a large amount of nucleus pulposus or torn anulus fibrosus from the vertebral canal (11.5, 11.12). Retrieval of extruded nucleus pulposus is not possible in some dogs because the compressive lesion consists
219
220
Small Animal Spinal Disorders
instead solely of anulus fibrosus. In this situation it is essential that a large portion of this anulus be excised so that on completion, the dura is visible clearly across the entire width of the slot (7.48, 11.29B). This may prove to be very difficult, as the fibers of the anulus tend to fragment before they can be removed and are often attached to the internal venous plexus. Removal is facilitated by leaving a plug of anulus to serve as a handle during removal (11.29A) and by using the inverted cone technique (Goring et al., 1991) (7.50). If the surgeon is in doubt about the adequacy of decompression after a ventral slot then it should be converted to a distraction-stabilization technique. This is recommended in order to prevent intervertebral collapse with its attendant risk of anular buckling and increased spinal cord compression (Chambers et al., 1986) (4.25, 11.21). The worsening noted after surgery even in dogs that do well long term may suggest that some degree of collapse is common after ventral slot (Rusbridge et al., 1998). It is likely that the signs then improve as the site stabilizes subsequently due to the development of a fibrous or osseous union (Sharp et al., 1995). Postoperative imaging can be used to assess the decompressive effect of a ventral slot and is certainly indicated if the neurological deficits are markedly worse after surgery (11.10, 11.21). If a distraction-stabilization procedure is needed following a ventral slot, then the metal and bone cement technique (11.13A, 11.40–11.43) or a Synthes locking plate (11.13B,C) are recommended (Wilson et al., 1994). A full cement plug should not be used to distract the site after a ventral slot as there is a high risk of cement entering the vertebral canal through the defect created in the dorsal anulus. A further potential problem is that
A
B
the cement also tends to collapse into the cancellous bone with subsequent loss of distraction (Dixon et al., 1996). A block of corticocancellous autograft (see page 292), small cement wedge (page 118), cancellous bone allograft (Veterinary Transplant Services, Seattle, WA), or a rope (11.31), can be used to distract a ventral slot temporarily prior to fixation (page 117). The major long-term disadvantage in using a ventral slot for CSM is that about 30% of dogs undergoing single level decompression suffer a second episode of neurological signs within 2–3 years (Bruecker et al., 1989b; Rusbridge et al., 1998) (Table 11.2). In most cases this is presumed to be due to recurrence at the original site or a domino lesion at an adjacent space. There are probably several reasons for this: • Even disc spaces that appear normal often show subclinical or histological abnormalities in dogs with CSM, especially the high incidence spaces C5/6 and C6/7 (1.3) (Seim and Withrow, 1982; Rusbridge et al., 1998). • Such intervertebral discs may not respond to fusion of an adjacent interspace in the same way that a disc would in a normal dog (Cole et al., 1987; Bruecker et al., 1989a; VanGundy, 1989; Hilibrand et al., 1999). Signal changes on MRI in adjacent discs have been detected in humans within 12 months of fusion (Iseda et al., 2001). • Fusion does occur at many disc spaces after a single site ventral slot (11.12). This occurs mainly after wide slots but the mobility at the slotted space probably changes even for more narrow slots (Gilpin, 1976; Chambers et al., 1982; VanGundy, 1989).
C
11.13 A: A ventral slot has been performed at C5/6 intervertebral space. Bone screws have been placed in C5 and C6 ready for bone cement to be applied once the space has been distracted. Distraction using a Gelpi retractor in adjacent disc spaces is no longer recommended (Wilson et al., 1994) (see ‘Vertebral distraction’, page 235). B, C: Seven-month postoperative radiographs of a Doberman treated using Syncage-C intervertebral implant and Cervical Spine Locking Plate (AO) at C6/7. Outcome at 1 year was excellent (Matis, 2001).
Cervical spondylomyelopathy
Taken together, this would suggest that high incidence disc spaces adjacent to the ventral slot site should undergo some sort of prophylactic procedure in order to reduce the incidence of domino lesions (see also ‘Vertebral decompression’, page 234). The choices of prophylactic procedure for high incidence disc spaces adjacent to a ventral slot are: 1. Place a cancellous bone graft at the adjacent space(s), combined with forage of ventral cortical bone (11.30), in order to promote fusion (Dixon et al., 1996). The aim is to improve stability and so reduce the likelihood of this space producing a domino lesion in the future. 2. Place a cement plug at the adjacent space(s). However, this approach may increase collapse of the slotted space. Even if this collapse does not compress the spinal cord it may cause foraminal narrowing and radiculopathy. 3. Do nothing at the adjacent space(s) but then the risk of domino lesions will remain high (7.14, 11.18, 11.23A). Grafting and forage are recommended as they carry the lowest risk of the three options, although at present there are no data to support their efficacy (11.30). The threat of a domino lesion is also present after a single site ventral slot combined with metal and bone cement fixation. The same three prophylactic options also apply for this situation, but inducing osseous fusion at the adjacent space will be more difficult due to the mass of cement ventral to the slot (see ‘Metal and bone cement’, page 239; 11.18, 11.43). Fenestration is not a suitable treatment for any dog with CSM (Lincoln and Pettit, 1985; Jeffery and McKee, 2001). It hastens intervertebral collapse and anular buckling, which cause spinal cord compression. Likewise, fenestration of discs to position a distraction instrument is not recommended (Wilson et al., 1994) (11.13A). Others suggest that concomitant forage and grafting may be sufficient to stabilize an interspace that has been fenestrated for distraction purposes (M. McKee, personal communication).
VERTEBRAL DISTRACTION-STABILIZATION (11.31–11.43) The primary indications for distraction and stabilization are the presence of a traction-responsive lesion and to relieve nerve root compression. Lesions may be either single or multiple and cause either dorsal or ventral spinal cord compression (Algorithms 11.1A,B). Distraction-stabilization has been attempted in the past using a number of different techniques, but the ones now recommended are the cement plug, Synthes
locking plate (11.13B), and metal implant and bone cement techniques. The latter technique is limited to single space distraction. The locking plate may be suitable for multiple lesions although this needs to be studied further in dogs. Cement plugs can be used for nearly all types of lesion. An advantage of all distraction techniques is that they often provide rapid relief of cervical hyperesthesia related to nerve root decompression. Unlike the ventral slot, these distraction techniques do not involve entry into the vertebral canal but this advantage is partly offset by the risk of implant failure or other implant-associated complications. Future fixation devices will almost certainly be made of absorbable materials (Vaccaro et al., 2002).
Metal implant and bone cement method (11.40–11.43) Metal implant and bone cement distraction is a well-tested technique with good long-term follow up results (Table 11.2). It is also the standard rescue technique for a failed ventral slot. The metal implants can either be Steinmann pins (11.43B), threaded pins (McKee and Sharp, 2003), or bone screws (11.13A, 11.19, 11.40–11.43). Attempts to bridge more than one interspace usually result in implant failure (Ellison et al., 1988; VanGundy, 1988) (11.17B). If two spaces must be bridged with cement, implants should be placed in all three vertebrae and the cement reinforced with a thick Steinmann pin (13.25). The main disadvantage of single-level metal and bone cement is the high rate of domino lesions. Subclinical lesions at adjacent interspaces should therefore be addressed for the reasons described under ventral decompression. This is difficult because of the mass of cement (11.19), but may be best-accomplished using forage (11.30), a cement plug, or possibly a locking plate (11.13B).
Cement plug (11.35–11.39) Cement plugs can be applied to many types of lesion but the most logical indication is for traction-responsive lesions (page 213). These implants retain the advantages of the earlier metal washer technique but have overcome many of the disadvantages (Wheeler and Sharp, 1994; Dixon et al., 1996; Rusbridge et al., 1998; McKee et al., 1999). Catastrophic collapse from end-plate fracture is unusual with the cement plug (11.20A), presumably as forces are distributed more evenly over the end plates than they were using washers (McKee et al., 1999). The main advantage of cement plugs is that they can be applied easily to more than one intervertebral space. The surgeon can therefore be more aggressive in dealing with a dog that has more than one lesion, any one of which could go on to cause a domino problem in the future if not treated (Dixon et al., 1996; Rusbridge et al., 1998).
221
222
Small Animal Spinal Disorders
Cement plugs should be placed routinely at the highest risk disc spaces (usually C5/6 and C6/7), and ideally to any other space that shows subclinical lesions. An alternate approach is to implant cement plugs at sites of spinal cord compression and fuse other sites that are considered to be at risk by forage and grafting (11.30). Obviously not every cervical intervertebral space can be fused and the maximum is probably three interspaces (Bolesta et al., 2000). It may also be difficult to decide exactly which sites to fuse in any given dog. For the Great Dane and other giant breeds, C4/5 or even C3/4 may be involved along with C5/6 and C6/7 (Olsson et al., 1982; Lewis, 1989; Lipsitz et al., 2001). For the Doberman usually only C6/7 and C5/6 discs are at high risk although in some dogs C4/5 should probably also be addressed (11.14). The aim of fusing multiple spaces routinely is to reduce the incidence of domino lesions (Dixon et al., 1996; Jeffery and McKee, 2001) (Table 11.2). Support for this more aggressive approach also comes from work in humans with cervical spondylosis. In humans, the risk of new disease at an adjacent level was found to be significantly lower following a multilevel arthrodesis than it was following a single-level arthrodesis. Therefore, it has been proposed in humans that all
degenerated segments causing radiculopathy or myelopathy should be included in an anterior cervical arthrodesis (Hilibrand et al., 1999). This is also consistent with reports suggesting that normal canine discs can adjust to the fusion of adjacent segments, whereas abnormal discs can not (Cole et al., 1987; Bruecker et al., 1989a). Cement plugs are also indicated for dogs with positional lesion(s) (page 214, Algorithms 11.1A,B). The aim of stabilization in these dogs is to reduce or abolish movement at the affected interspace. Stabilization may even benefit dorsally located soft tissue or osseous lesions by limiting movement locally. Fusion has been shown to cause the regression of ligamentum flavum lesions in dogs and of bony articular lesions in horses (Grant et al., 1985; Seim, 1986; McKee et al., 1990). Therefore cement plugs should be considered for dogs with certain types of traction non-responsive lesions such as those caused by osseous compression of the dorsal vertebral canal (11.15). The advantage of using cement plugs over dorsal laminectomy in these situations is that distraction-stabilization should avoid the short-term morbidity associated with laminectomy and will also allow the surgeon to address adjacent, subclinical lesions (de Risio et al., 2002).
11.14 A: Cement plugs have been inserted in this Doberman at the C4/5, C5/6 and C6/7 intervertebral spaces. The degree of cement filling is good for C4/5 and C5/6 but suboptimal for C6/7. The dog wore an external splint for 8 weeks (11.39). B: Two-month follow-up radiograph reveals good osseous fusion at all three spaces. The presenting complaint of neck pain and exercise intolerance had resolved. A
B
11.15 Young Great Dane with severe, osseous stenosis from C3/4 to C6/7. A: Preoperative CT myelogram from site of maximum compression at C5/6. Cement plugs were applied at C4/5 and C5/6; C6/7 was only foraged and grafted. A restraint device was applied (13.18B). B: Two-year follow-up, again from site of maximum compression at C5/6 (arrow ⫽ cement plug); there is bony remodeling and less spinal cord compression (arrowheads). C6/7 had not fused; compression here was worse, as was the dog’s neurological status. A
B
Cervical spondylomyelopathy
The main disadvantage of the cement-plug technique is that it is unclear if an external splint must be used in the postoperative period. Splints are not tolerated by all dogs, they may require frequent modification or replacement and can cause pressure sores (VanGundy, 1988) (11.39). An alternative is to use a Halti (13.18B) or fixation screw or pin to prevent the cement from falling out (McKee and Sharp, 2003), (11.16, 11.36–11.38), or to use a locking plate instead (11.13B). Preliminary results with this technique are encouraging (11.13B,C). A swivel ring in the plate hole means that screws may be inserted at any angle within a range of ⫾20º and the screws lock in the plate via a unique locking mechanism. The Syncage is designed to maintain distraction; it lies in the intervertebral space and is packed with cancellous bone. These implants are extremely strong and can even bridge more than one interspace (McLaughlin et al., 1997; Matis, 2001). Failure rates are lower in humans than for implants that lack the screw-locking feature (Lowery and McDonough, 1998). However, their utility for multiple lesions in dogs is not yet clear.
Locking plate
DORSAL DECOMPRESSION This technique is an alternative for dogs with single dorsal lesions that do not respond to traction as well as those with multiple dorsal lesions (Dickinson et al., 2001; Lipsitz et al., 2001) (11.8, 11.56, 11.57). This technique also provides an option for dogs with ventral lesions at multiple intervertebral spaces (Lyman, 1991) (7.15).
Dorsal laminectomy (11.44–11.57)
Traction non-responsive
Ventral lesion
Ventral slot*
Dorsal lesion
Positional
Ventral lesion
Cement plug(s)*, locking plate, laminoplasty or dorsal laminectomy
Long-term results of dorsal laminectomy appear to be very good (Lipsitz et al., 2001; de Risio et al., 2002). The major disadvantage is that there is significant, short-term morbidity with deterioration in neurological status, which can cause considerable nursing problems in giant-breed dogs (Trotter et al., 1976; VanGundy, 1988; de Risio et al., 2002). An extensive soft-tissue approach is also needed and ventrally located disc material cannot be removed (VanGundy, 1988). Although dorsal laminectomy should not cause domino lesions, recurrence of clinical signs occurs in about 10% of dogs and is reported to be due to restrictive fibrosis (Trotter et al., 1976; de Risio et al., 2002) (Table 11.2). Because of the disadvantages of laminectomy, and because cement plugs also allow the surgeon to address multiple lesions, including subclinical ones, the cement-plug technique is preferred for most types of lesion (Algorithms 11.1A,B). Dorsal laminectomy has given good results for dogs with synovial cysts, with four of six dogs being followed for more than 1 year (Dickinson et al., 2001). It is not clear if these results will be maintained long term in a larger series of dogs (Dickinson et al., 2001; Lipsitz et al., 2001). When it is desirable to provide additional stability following dorsal decompression, fusion can be encouraged by screwing and then bone grafting the facet joints (11.56, 11.57). Potential risks include fracture of a facet and trauma to nerves at the level of the foramen. Stability of this fixation technique was reported to be good when spines were assessed grossly at post mortem 2 weeks after surgery (Swaim, 1975). Long-term results of this approach have not been reported.
Traction responsive
Ventral lesion
Cement plug(s)* or locking plate
Dorsal lesion
Cement plug(s)* or locking plate
Cement plug(s)* or locking plate
*Consider a strategy to address all high-risk disc spaces
Algorithm 11.1A a single lesion.
Surgical decision-making in dogs with
223
224
Small Animal Spinal Disorders
Traction non-responsive
Ventral lesions
Dorsal lesions
Positional
Ventral lesions
Dorsal lesions
Ventral slot*
Traction responsive
Ventral lesions
Dorsal lesions
Cement plugs
Cement plugs
Algorithm 11.1B Surgical decision-making in dogs with multiple lesions.
Cement plugs or dorsal laminectomy
Laminoplasty (11.58)
This technique has been reported just once and was then only used as a singlelevel procedure. A 5-year-old Great Dane with marked ataxia and tetraparesis improved markedly 2 weeks after surgery. It was normal neurologically 6 months later but then died of a gastric torsion (McKee, 1988). Laminoplasty is used widely in humans with severe vertebral canal stenosis for both single- and multi-level compression (Kohno et al., 1997; Shaffrey et al., 1999). Its main advantage over laminectomy is that it provides protection to the spinal cord from the recurrent compression that can arise from re-growth of a fibrous or osseous lamina. This is often termed restrictive fibrosis (Trotter et al., 1976; de Risio et al., 2002). Laminoplasty may also induce less change in adjacent discs compared to fusion (Iseda et al., 2001) and warrants further evaluation.
COMPLICATIONS Some risks are either inherent to surgery on the cervical spine or to a particular breed and as such should be explained clearly to the owner. Many of the respiratory and cardiovascular problems discussed in Chapter 7 also occur in CSM (Boxes 7.3, 7.4; Table 11.4). These problems emphasize further the need for a thorough presurgical evaluation (Calvert et al., 1996). Complications can be divided into three main categories (Table 11.4).
Intraoperative complications These are mainly due to either iatrogenic injury or technical errors (Table 11.5). Overzealous retraction of
Table 11.4 Main sources of complications
Intraoperative Iatrogenic injury Technical errors
Early postoperative
Late postoperative
Implant failure Postoperative morbidity Infection Adjacent segment disease
Adjacent segment disease Recurrence of signs
soft tissues during a ventral approach to the neck can damage important nerves in the cervical region. This can induce intraoperative arrhythmias or postoperative laryngeal paralysis (11.20B), Horner’s syndrome (Boydell, 1995), or megaesophagus (15.40). Bleeding from the venous plexus can be a major problem during ventral decompression and may require a blood transfusion. Excessive retraction or improper patient positioning can exacerbate the bleeding by compressing the jugular veins (11.24). Hemorrhage can usually be arrested by relieving pressure on the jugular veins in combination with use of a piece of muscle to promote coagulation (Table 7.2; 8.43). Direct pressure can also be used (11.29B). Hematoma formation is most likely to occur in the Doberman and can cause delayed damage to the vagus nerve or its branches (15.40) or even spinal cord compression (Seim and Prata, 1982) (see page 100). Technical errors during surgery are the other main source of complications in CSM. Examples include improper selection of the implant (McKee et al., 1990) (11.16) and poor implant positioning (11.17A, 11.37).
Cervical spondylomyelopathy
Table 11.5 Specific complications
A
Intraoperative
Early postoperative
Late postoperative
Iatrogenic neural injury (11.20B) Wrong surgical site Hemorrhage Poor implant selection (11.16) Poor implant position (11.17A, 11.37) Pneumomediastinum Prolonged extension (11.18) Vertebral fracture
Implant failure (11.19, 11.20A) Morbidity after ventral slot (11.21) Morbidity after dorsal laminectomy End-plate failure (11.20A) Self-induced trauma (11.22) Ventral or epidural hematoma (15.40) Dorsal seroma Ischemic injury Adjacent segment disease Discospondylitis Epidural abscess (14.14) Recumbency complications
Adjacent segment disease (11.23A) Restrictive fibrosis Recurrence of signs (11.8, 11.23B) Discospondylitis (14.11) Delayed compression from implant
B
C
11.16 A, B: A single midline screw has been used to keep each cement plug in place (11.35–11.38). Unfortunately each screw passed through a vascular foramen (1.18) to enter the vertebral canal and cause dramatic compression of the spinal cord, especially over C5. C: The screws were removed and replaced by shorter ones. The dog walked without assistance the next day and had an excellent outcome; the screw probably caused only mild, transient spinal cord compression.
A
B
C
11.17 A, B: These screws are all inserted too caudally in their respective vertebral bodies. Two have penetrated an end plate and all four have the potential to damage the nerve roots or spinal nerves (Swaim, 1975). C: An excess of bone cement can compress structures within the thoracic inlet to cause dysphagia or even esophageal perforation (Halligan and Hubschmann, 1993; Dixon et al., 1996). Note that in this dog, the implants span two intervertebral spaces. If this is done, more implants and cement reinforcement are needed (page 221) (13.25).
225
226
Small Animal Spinal Disorders
The surgeon may also cause iatrogenic spinal cord damage (Read et al., 1983; Lipsitz et al., 2001), vertebral fracture (McKee et al., 1989), induce pneumomediastinum (Marchevsky and Richardson, 1999), perform surgery at the wrong site, or simply fail to understand certain limitations of the technique itself (11.17C). Prolonged or excessive extension of the neck during surgery is undesirable as it causes spinal cord compression (11.10). In addition, it could lead to fusing the vertebrae in extension if the position is not corrected prior to cement hardening (Bruecker et al., 1989b) (11.18).
Early postoperative complications The main complications at this stage are postoperative deterioration and implant failure (Table 11.5). Refractory neck pain or a marked deterioration in neurological status are indications for repeat imaging and possibly for a second surgery (Rusbridge et al., 1998; McKee et al., 1999). Implant failure can occur such as loss of distraction after a variety of techniques (Bruecker et al., 1989a; Wilson et al., 1994; Marchevsky and Richardson, 1999; McKee et al., 1999; (11.19–11.20A). Implant failure may also be subclinical at times (11.19B, 11.20B, 11.37B). 11.18 A: There is a lesion at C5/6 and a probable subclinical lesion at C4/5. B: The C5/6 lesion was distracted manually and stabilized using screws and bone cement. Only minimal distraction is visible on the postoperative myelogram; the caudal spine was also fixed inadvertently in extension. The dog did well after surgery and had no deficits or neck pain 7 months later. The thread for the cranial screw in C6 vertebra was stripped and it was replaced with a 6.5-mm screw. This dog is at a particularly high risk for a domino lesion at C4/5 or C6/7.
A
B
A
A
B
11.19 A: Distraction at C6/7 intervertebral space using two small 2.7mm distraction screws (11.40). Each of the four 4.5-mm screws only penetrates one cortex (11.43). B: Loss of distraction at 7 months; one of the distraction screws has broken. In addition the 4.5mm screws have changed orientation probably due to bone remodeling. The cement has not failed and the dog was doing well clinically although it had an episode of neck pain 10 months post-surgery.
B
11.20 A: Loss of distraction after a cement plug was used at C5/6 (11.38B). The dog showed sudden deterioration 4 weeks after surgery. A small fracture is visible at the ventral portion of the plug (arrow) and there is failure of the cranial end plate of C6 (arrowhead) with recurrent spinal cord compression. B: This dog did well until it fell down the stairs 6 weeks after distraction at C6/7. Failure of a screw in C7 did not cause clinical signs (McKee et al., 1990). The dog had also shown a change in bark since surgery.
Cervical spondylomyelopathy
End-plate failure can occur after the cement-plug technique (11.20A); potential factors include excessive motion, thermal necrosis and possibly weakening from too large an anchor hole (Boker et al., 1989; Martinez et al., 1997; Williams et al., 1997). Inadequate removal of disc material during a ventral slot decompression can increase spinal cord compression as the intervertebral space collapses (Chambers et al., 1986; Ellison et al., 1988) (11.21). This complication was seen in up to 20% of dogs that failed to respond after ventral slot (Chambers et al., 1986). Even if it does not compress the spinal cord, collapse can compress the nerve roots in the intervertebral foramen, thereby increasing cervical hyperesthesia. Collapse at an interspace can also be exacerbated by the presence of an unsuspected synovial cyst (4.25). Even when surgery goes well there can be a deterioration in neurological status after either dorsal laminectomy or ventral slot (Rusbridge et al., 1998; de Risio et al., 2002). This deterioration increases a dog’s susceptibility to self-induced trauma, particularly as it recovers from anesthesia or tries to stand up (Bruecker et al., 1989b) (11.22). Even without injury, there may be deterioration secondary to decreased stability at the operated site(s)
(Queen et al., 1998; Rusbridge et al., 1998; Macy et al., 1999). Range of motion increases in cadaver spines by 30–40% after fenestration and by 66% after a ventral slot, even if the slot is only one third of the vertebral width (Macy et al., 1999; Wolf and Roe, personal communication). Ventral slots with dimensions of greater than 50% of the vertebral width produce instability in small dogs and are likely to do the same in larger dogs with CSM (Seim and Withrow, 1983; Fitch et al., 2000; Lemarie et al., 2000). Dorsal laminectomy in the lumbar region is known to reduce stability significantly (see Chapter 8) but data are not available for the neck. Postoperative deterioration could also be due to an ischemic or reperfusion injury of the spinal cord (Cybulski and D’Angelo, 1988). Vascular injuries can occur either during surgery or in the immediate postoperative period (see pages 86 and 130). Early development of a second lesion at a new disc space has been described at 2, 4 and 12 weeks after surgery (Chambers et al., 1982; Wilson et al., 1994; Rusbridge et al., 1998). One of these lesions developed at a space that had been fenestrated in order to place a Gelpi for distraction and in another the disc was affected subclinically at the time of the first surgery (Wilson et al., 1994; Rusbridge et al., 1998).
11.21 A: The preoperative myelogram reveals ventral, extradural compression at C6/7 that did not improve with traction. B: The dog was worse the day after ventral slot at C6/7; the margins of which are visible clearly in C7 (arrowhead) and to a lesser extent in C6. Compression was now more severe than before surgery, probably because of inadequate removal of disc material and intervertebral collapse (Chambers et al., 1986). A
11.22 A: This dog made a good recovery after ventral slot at C5/6 but became severely tetraparetic 36 h later. A small fracture is visible at the ventral aspect of C5 (arrow), presumably due to hyperflexion injury (Bruecker et al., 1989b; McKee et al., 1989). B: Distraction-stabilization were performed using screws and bone cement. The dog made a slow recovery and was able to walk 4 months after surgery (15.11).
A
B
227
228
Small Animal Spinal Disorders
Discospondylitis can also occur on either an early or delayed basis after surgery (Chambers et al., 1982). It can sometimes be associated with an epidural abscess (14.14).
Late postoperative complications The most important of these is recurrence of clinical signs (Table 11.5; 11.8, 11.23B). Domino lesions or adjacent segment disease (11.23A) are a particular problem and result at least in part from abnormal stresses imposed on one intervertebral space by fixation of an interspace adjacent to it (Fox et al., 1996; Hilibrand et al., 1999). These stresses can exacerbate any pre-existing subclinical instability and so produce either disc extrusion or hypertrophy of anular or ligamentous structures (Bruecker et al., 1989a).
A
Recurrence of paraparesis or tetraparesis occurs in up to one third of dogs after either ventral decompression or metal implant and bone cement fixation (Table 11.2). Recurrence can be caused by compression at the original site or by a domino lesion at an adjacent site (Jeffery and McKee, 2001). It usually occurs between 6 months and 4 years after the original surgery, with a mean of around 2 years (Seim, 1986; Bruecker et al., 1989a,b; Rusbridge et al., 1998; McKee et al., 1999). Recurrence seems to be less of a problem for the cement-plug technique, which is probably because all high-risk interspaces can be stabilized at the same time (Dixon et al., 1996; Hilibrand et al., 1999). Recurrence after dorsal decompression is often due to constrictive fibrosis at the surgical site, also termed the ‘laminectomy membrane’ (LaRocca and Macnab, 1974) (page 86). When the roof of the vertebral canal is removed at laminectomy, the spinal cord is often displaced upwards when ventral extradural compression is present (Lyman, 1991). However, the scar that forms as the laminectomy heals often does so in the position of the original lamina, which can then cause a recurrence of compression (Trotter et al., 1975, 1976; Lyman, 1991; de Risio et al., 2002). A second surgery may be successful when signs recur following a dorsal laminectomy (de Risio et al., 2002) (11.23B). Overall recurrence rates appear to be similar between dorsal and ventral decompressive techniques (Jeffery and McKee, 2001). Spinal cord compression can also be caused by a relative change in the position of an implant secondary to bony remodeling and collapse of a distracted interspace (McKee et al., 1990). In addition, late-onset discospondylitis has been recorded 6 and 26 months after a surgery and it can also occur as a complication of an unrelated surgical procedure (Ellison et al., 1988; Dixon et al., 1996) (14.11).
Postoperative care
B 11.23 A: This dog underwent a ventral slot decompression at C6/7 1 year previously. It had recurrent tetraparesis due to a new lesion at C5/6. This is the domino effect. B: CT myelogram made at C5/6 in a Great Dane that underwent a dorsal laminectomy 1 year previously. The dog improved but then deteriorated 10 months later. Proliferation of new bone has almost reformed the lamina of the vertebral canal resulting in severe spinal cord compression with atrophy (A and B from Sharp et al., 1992).
Activity must be restricted severely in the immediate postoperative period. Ataxic dogs should be confined to a small kennel for 1–2 weeks to prevent them from stumbling and hyperflexing their neck (11.22). Activity can be increased gradually after this period. As a minimum precaution, neck collar restraint and vigorous exercise should be avoided for 4 months. The first 6–8 weeks after surgery are the most crucial until significant osseous or fibrous healing has occurred. Protective neck braces can be used for this period but may be tolerated poorly (11.39, 13.18). Dogs that remain recumbent after surgery require a high level of nursing care (see Chapter 15). Pneumonia
Cervical spondylomyelopathy
is a particular risk for these dogs (Seim, 1986; Bruecker et al., 1989b). Recumbent dogs are also prone to muscle atrophy, decubitus, gastric volvulus (Trotter et al., 1976; Mason, 1979; Seim, 1986; McKee, 1988) seroma formation at the surgical site (15.34), urinary tract infection (UTI) (Rusbridge et al., 1998), and urine scald (15.35) (VanGundy, 1989). Some of these factors can be complicated further by hypothyroidism (VanGundy, 1989; de Risio et al., 2002). Additional postoperative problems in CSM include hematoma formation due to VW disease (Rusbridge et al., 1998) (15.40), diarrhea and gastrointestinal hemorrhage, especially after excessive corticosteroid use (VanGundy, 1988), pancreatitis (Read et al., 1983; VanGundy, 1988); excoriation of digits (Trotter et al., 1976) (15.7, 15.10), discospondylitis or epidural abscess (Chambers et al., 1982; Dixon et al., 1996) (14.11, 14.14), sepsis (Black, 1986) (14.14), and multiple abscessation (Seim, 1986).
PROGNOSIS The seriousness of this condition is best illustrated by the fact that a quarter of dogs with CSM in one series were euthanized within 6 weeks of surgery as a result of neurological problems (Seim, 1986). Overall longterm mortality rates for CSM vary from 19 to 43% (Seim et al., 1986; Dixon et al., 1996; Rusbridge et al., 1998; McKee et al., 1999; de Risio et al., 2002). Dogs with more than one lesion generally have a worse prognosis than dogs with single lesions, and dogs with chronic tetraparesis have a guarded prognosis (Seim, 1986). Most severely tetraparetic dogs that are going to recover will do so within 6 weeks. In some dogs surgery will only halt the progression of disease (VanGundy, 1988). This is presumably because there is often significant loss of neural tissue at the site of the lesion (Read et al., 1983; Sharp et al., 1995). It is therefore likely that the outcome will be better if surgery is done earlier in the disease process (Chambers et al., 1986; VanGundy, 1988; McKee et al., 1990). The general estimate provided by Seim is still useful to predict the likely outcome after surgery for CSM. For dogs with single lesions, about 80% of those that are walking prior to surgery will have a favorable outcome (Seim, 1986). Success rates quoted in the literature vary from 70 to 90% but these figures do not include dogs that undergo euthanasia later for reasons related to CSM, usually due to recurrence of signs (Chambers et al., 1986; de Risio et al., 2002) (Table 11.2). Any apparent differences in success rates between the various techniques are not statistically significant (Jeffery and McKee, 2001). There are conflicting results from the limited studies conducted to date
on whether dogs that are unable to walk before surgery have a poor prognosis; the opposite may even be true (Seim, 1986; Ellison et al., 1988; Bruecker et al., 1989a; Dixon et al., 1996; Jeffery and McKee, 2001; de Risio et al., 2002). Clearly there is a need to reduce the overall mortality rate for dogs with this condition and in particular to reduce the high risk of recurrent spinal cord compression. Key issues for future investigation 1. Does grafting an interspace that is judged to have subclinical disease increase its stability and thereby result in a decreased incidence of domino lesions (11.30B)? 2. Does forage of the ventral cortex of adjacent vertebrae enhance the degree of fusion at a site grafted with cancellous bone (11.30A)? 3. Does the routine fusion of C5/6 and C6/7 interspaces decrease the incidence of domino lesions or will this approach just serve to transfer the problem along to C4/5 or C7/T1? If this approach works, how does the surgeon decide how many spaces to fuse (Bolesta et al., 2000) (11.14)? 4. Is there any objective means to decide whether an interspace is at risk of a domino lesion (Mitchell et al., 2001)? Will MRI be able to identify such an interspace and does this technique have any advantage over use of an extension view under myelography? 5. If the aim is to induce fusion between adjacent vertebrae, at what stage should this be considered successful (Fox et al., 1996; Gibson et al., 2002)? Does the presence of a radiolucent line in the bridging spondyle mean that there is insufficient stability?
REFERENCES Anderson, G.I. (1988) Polymethylmethacrylate: a review of the implications and complications of its use in orthopaedic surgery. Veterinary and Comparative Orthopaedics and Traumatology 2, 74–79. Bagley, R.S., Silver, G.M., Connors, R.L., Harrington, M.L., Cambridge, A.J., Moore, M.P. (2000) Exogenous spinal trauma: surgical therapy and aftercare. Compendium on Continuing Education for the Practicing Veterinarian 22, 218–230. Baum, F., de Lahunta, A., Trotter, E.J. (1992) Cervical fibrotic stenosis in a young Rottweiler. Journal of the American Veterinary Medical Association 201, 1222–1224. Black, A.P. (1986) Results of arthrodesis as a treatment for caudal cervical spondylo-myelopathy in nine dogs. Australian Veterinary Practitioner 16, 147–148. Boker, D., Schultheiss, R., Probst, E. (1989) Radiologic long-term results after cervical vertebral interbody fusion with polymethyl methacrylat (PMMA). Neurosurgical Review 12, 217–221. Bolesta, M.J., Rechtine, G.R., 2nd, Chrin, A.M. (2000) Three- and fourlevel anterior cervical discectomy and fusion with plate fixation: a prospective study. Spine 25, 2040–2044; discussion 2045–2046. Boydell, P. (1995) Horner’s syndrome following cervical spinal surgery in the dog. Journal of Small Animal Practice 36, 510–512.
229
230
Small Animal Spinal Disorders
Bruecker, K.A., Seim, H.B., Blass, C.E. (1989a) Caudal cervical spondylomyelopathy: decompression by linear traction and stabilization with Steinmann pins and polymethyl methacrylate. Journal of the American Animal Hospital Association 25, 677–683. Bruecker, K.A., Seim, H.B., Withrow, S.J. (1989b) Clinical evaluation of three surgical methods for treatment of caudal cervical spondylomyelopathy of dogs. Veterinary Surgery 18, 197–203. Burbidge, H.M. (2001) Caudal Cervical Vertebral Malformation in the Dobermann Pinscher, 121–135. PhD thesis. Massey University, New Zealand. Callan, M.B., Giger, U. (2002) Effect of desmopressin acetate administration on primary hemostasis in Doberman Pinschers with type-1 von Willebrand disease as assessed by a point-of-care instrument. American Journal of Veterinary Research 63, 1700–1706. Calvert, C.A., Wall, M. (2001) Results of ambulatory electrocardiography in overtly healthy Doberman Pinschers with equivocal echocardiographic evidence of dilated cardiomyopathy. Journal of the American Veterinary Medical Association 219, 782–784. Calvert, C.A., Jacobs, G.J., Pickus, C.W. (1996) Unfavorable influence of anesthesia and surgery on Doberman Pinschers with occult cardiomyopathy. Journal of the American Animal Hospital Association 32, 57–62. Cechner, P.E. (1980) Ventral cervical disc fenestration in the dog: a modified technique. Journal of the American Animal Hospital Association 16, 647–650. Chambers, J.N., Betts, C.W. (1977) Caudal cervical spondylopathy in the dog: a review of 20 clinical cases and the literature. Journal of the American Animal Hospital Association 13, 571–576. Chambers, J.N., Oliver, J.E., Jr, Kornegay, J.N., Malnati, G.A. (1982) Ventral decompression for caudal cervical disk herniation in large- and giant-breed dogs. Journal of the American Veterinary Medical Association 180, 410–414. Chambers, J.N., Oliver, J.E., Jr, Bjorling, D.E. (1986) Update on ventral decompression for caudal cervical disk herniation in Doberman Pinschers. Journal of the American Animal Hospital Association 22, 775–778. Ching, Y., Meyers, K.M., Brassard, J.A., Wardrop, K.J. (1994) Effect of cryoprecipitate and plasma on von Willebrand factor multimeters and bleeding time in Doberman Pinchers with type-1 von Willebrand’s disease. American Journal of Veterinary Research 55, 102–110. Clark, D.M. (1986) An analysis of intraoperative and early postoperative mortality associated with cervical spinal decompressive surgery in the dog. Journal of the American Animal Hospital Association 22, 739–744. Cole, T., Ghosh, P., Hannan, N., Taylor, T., Bellenger, C. (1987) The response of the canine intervertebral disc to immobilization produced by spinal arthrodesis is dependent on constitutional factors. Journal of Orthopedic Research 5, 337–347. Cybulski, G.R., D’Angelo, C.M. (1988) Neurological deterioration after laminectomy for spondylotic cervical myeloradiculopathy: the putative role of spinal cord ischaemia. Journal of Neurology, Neurosurgery and Psychiatry 51, 717–718. De Risio, L., Munana, K., Murray, M., Olby, N., Sharp, N.J., Cuddon, P. (2002) Dorsal laminectomy for caudal cervical spondylomyelopathy: postoperative recovery and long-term follow-up in 20 dogs. Veterinary Surgery 31, 418–427. Dickinson, P.J., Sturges, B.K., Berry, W.L., Vernau, K.M., Koblik, P.D., LeCouteur, R.A. (2001) Extradural spinal synovial cysts in nine dogs. Journal of Small Animal Practice 42, 502–509. Dixon, B.C., Tomlinson, J.L., Kraus, K.H. (1996) Modified distractionstabilization technique using an interbody polymethyl methacrylate plug in dogs with caudal cervical spondylomyelopathy. Journal of the American Veterinary Medical Association 208, 61–68. Dodds, W.J. (1989) Acquired von Willibrand’s disease in dogs. American Animal Hospital Association 1989; 614–619. Donawick, W.J., Mayhew, I.G., Galligan, D.T., Green, S.L., Stanley, E.K., Osborne, J. (1993) Results of a low-protein, low-energy diet and confinement on young horses with wobbles. Proceedings of the Thirty Ninth Annual Convention of the American Association of Equine Practitioners, Texas, USA, December 5–8, 1993, 125–127. Drost, W.T., Lehenbauer, T.W., Reeves, J. (2002) Mensuration of cervical vertebral ratios in Doberman pinschers and Great Danes. Veterinary Radiology and Ultrasound 43, 124–131.
Dueland, R., Furneaux, R.W., Kaye, M.M. (1973) Spinal fusion and dorsal laminectomy for midcervical spondylolishthesis in a dog. Journal of the American Veterinary Medical Association 162, 366–369. Ellison, G.W., Seim, H.B., Clemmons, R.M. (1988) Distracted cervical spinal fusion for management of caudal cervical spondylomyelopathy in large-breed dogs. Journal of the American Veterinary Medical Association 193, 447–453. Faiss, J.H., Schroth, G., Grodd, W., Koenig, E., Will, B., Thron, A. (1990) Central spinal cord lesions in stenosis of the cervical canal. Neuroradiology 32, 117–123. Fitch, R.B., Kerwin, S.C., Hosgood, G. (2000) Caudal cervical intervertebral disk disease in the small dog: role of distraction and stabilization in ventral slot decompression. Journal of the American Animal Hospital Association 36, 68–74. Fox, M., Onofrio, B., Hanssen, A. (1996) Clinical outcomes and radiological instability following decompressive lumbar laminectomy for degenerative spinal stenosis: a comparison of patients undergoing concomitant arthrodesis versus decompression alone. Journal of Neurosurgery 85, 793–802. Garcia, J.N.P., Milthorpe, B.K., Russell, D., Johnson, K.A. (1994) Biomechanical study of canine spinal fracture fixation using pins or bone screws with polymethylmethacrylate. Veterinary Surgery 23, 322–329. Gibson, J.N.A., Waddell, G., Grant, I.C. (2002) Surgery for degenerative lumbar spondylosis (Cochrane Review). Issue 1 ed: The Cochrane Library. http://www.update-software.com/cochrane/ Gieger, T.L., Hosgood, G., Taboada, J., Wolfsheimer, K.J., Mueller, P.B. (2000) Thyroid function and serum hepatic enzyme activity in dogs after phenobarbital administration. Journal of Veterinary Internal Medicine 14, 277–281. Giles, A.R., Tinlin, S., Greenwood, R. (1982) A canine model of hemophilic (factor VIII:C deficiency) bleeding. Blood 60, 727–730. Gilpin, G.N. (1976) Evaluation of three techniques of ventral decompression of the cervical spinal cord in the dog. Journal of the American Veterinary Medical Association 168, 325–328. Goring, R.L., Beale, B.S., Faulkner, R.F. (1991) The inverted cone decompression technique: a surgical treatment for cervical vertebral instability ‘Wobbler syndrome’ in Doberman Pinschers. Part 1. Journal of the American Animal Hospital Association 27, 403–409. Gorman, W.F., Hodak, J.A. (1997) Herniated intervertebral disc without pain. Journal of the Oklahoma State Medical Association 90, 185–190. Grant, B., Hoskinson, J.J., Barbee, D.D., Gavin, P.R., Sande, R.D., Bayly, W.M. (1985) Ventral stabilization for decompression of caudal cervical spinal cord compression in the horse. Proceedings of 31st Annual Convention of the American Association of Equine Practitioners 1985; 75–103. Halligan, M., Hubschmann, O. (1993) Short-term and long-term failures of anterior polymethylmethacrylate construct with esophageal perforation. Spine 18, 759–761. Hayashi, H., Okada, K., Hashimoto, J., Tada, K., Ueno, R. (1988) Cervical spondylotic myelopathy in the aged patient. A radiographic evaluation of the aging changes in the cervical spine and etiologic factors of myelopathy. Spine 13, 618–625. Hilibrand, A., Carlson, G., Palumbo, M., Jones, P., Bohlman, H. (1999) Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. The Journal of Bone and Joint Surgery—American volume 81, 519–528. Iseda, T., Goya, T., Nakano, S., Kodama, T., Moriyama, T., Wakisaka, S. (2001) Serial changes in signal intensities of the adjacent discs on T2weighted sagittal images after surgical treatment of cervical spondylosis: anterior interbody fusion versus expansive laminoplasty. Acta Neurochirurgica (Wein) 143, 707–710. Jeffery, N.D. (1995) The ‘wobbler’ syndrome. In: N.D. Jeffery (ed.), Handbook of Small Animal Spinal Surgery, 169–186. London: WB Saunders. Jeffery, N.D., McKee, W.M. (2001) Surgery for disc-associated wobbler syndrome in the dog—an examination of the controversy. Journal of Small Animal Practice 42, 574–581. Jergens, A.E., Turrentine, M.A., Kraus, K.H., Johnson, G.S. (1987) Buccal mucosa bleeding times of healthy dogs and of dogs in various pathologic states, including thrombocytopenia, uremia, and von Willebrand’s disease. American Journal of Veterinary Research 48, 1337–1342.
Cervical spondylomyelopathy
Kohno, K., Kumon, Y., Oka, Y., Matsui, S., Ohue, S., Sakaki, S. (1997) Evaluation of prognostic factors following expansive laminoplasty for cervical spinal stenotic myelopathy. Surgical Neurology 48, 237–245. Krag, M.H. (1991) Biomechanics of thoracolumbar spinal fixation. A review. Spine 16 (3 Suppl), S84–99. Kraus, K.H., Turrentine, M.A., Jergens, A.E., Johnson, G.S. (1989) Effect of desmopressin acetate on bleeding times and plasma von Willebrand factor in Doberman Pinscher dogs with von Willebrand’s disease. Veterinary Surgery 18, 103–109. Lamb, C.R. (1995) The dorsoventral cervical myelogram. Veterinary Radiology and Ultrasound 36, 201–202. LaRocca, H., Macnab, I. (1974) The laminectomy membrane. Studies in its evolution, characteristics, effects and prophylaxis in dogs. The Journal of Bone and Joint Surgery—British volume 56B, 545–550. Lemarie, R.J., Kerwin, S.C., Partington, B.P., Hosgood, G. (2000) Vertebral subluxation following ventral cervical decompression in the dog. Journal of the American Animal Hospital Association 36, 348–358. Levitski, R., Chauvet, A., Lipsitz, D. (1999) Cervical myelopathy associated with extradural synovial cysts in 4 dogs. Journal of Veterinary Internal Medicine 13, 181–186. Lewis, D.D., Hosgood, G. (1992) Complications associated with the use of iohexol for myelography of the cervical vertebral column in dogs: 66 cases (1988–1990). Journal of the American Veterinary Medical Association 200, 1381–1384. Lewis, D.G. (1989) Cervical spondylomyelopathy (‘wobbler’ syndrome) in the dog: a study based on 224 cases. Journal of Small Animal Practice 30, 657–665. Lincoln, J.D., Pettit, G.D. (1985) Evaluation of fenestration for treatment of degenerative disc disease in the caudal cervical region of large dogs. Veterinary Surgery 14, 240–246. Lipsitz, D., Levitski, R.E., Chauvet, A.E., Berry, W.L. (2001) Magnetic resonance imaging features of cervical stenotic myelopathy in 21 dogs. Veterinary Radiology and Ultrasound 42, 20–27. Lowery, G.L., McDonough, R.F. (1998) The significance of hardware failure in anterior cervical plate fixation. Patients with 2- to 7-year followup. Spine 23, 181–186; discussion 186–187. Lyman, R. (1991) Wobbler syndrome. Continuous dorsal laminectomy is the procedure of choice. Progress in Veterinary Neurology 2, 143–146. Macy, N.B., Les, C.M., Stover, S.M., Kass, P.H. (1999) Effect of disk fenestration on sagittal kinematics of the canine C5–C6 intervertebral space. Veterinary Surgery 28, 171–179. Marchevsky, A.M., Richardson, J.L. (1999) Disc extrusion in a Rottweiler dog with caudal cervical spondylomyelopathy after failure of intervertebral distraction/stabilisation. Australian Veterinary Journal 77, 295–297. Martinez, S.A., Arnoczky, S.P., Flo, G.L., Brinker, W.O. (1997) Dissipation of heat during polymerization of acrylics used for external skeletal fixator connecting bars. Veterinary Surgery 26, 290–294. Mason, T.A. (1979) Cervical vertebral instability (wobbler syndrome) in the dog. Veterinary Record 104, 142–145. Massicotte, C., Jones, J., Newman, S., Moon, M. (1999) Wobbler syndrome due to cervical stenosis in a Great Dane puppy. Canine Practice 24, 18–21. Matis, U. (2001) AO spinal implants for canine wobbler syndrome. McKee, M., Sharp, N. (2003) Cervical spondylopathy. In: D. Slatter (ed.), Small Animal Surgery, 3rd edn, 1180–1192. Philadelphia: WB Saunders. McKee, W., Penderis, J., Dennis, R. (2000) Radiology corner: obstruction of contrast medium flow during cervical myelography. Veterinary Radiology and Ultrasound 41, 342–343. McKee, W.M. (1988) Dorsal laminar elevation as a treatment for cervical vertebral canal stenosis in the dog. Journal of Small Animal Practice 29, 95–103. McKee, W.M., Lavelle, R.B., Mason, T.A. (1989) Vertebral stabilisation for cervical spondylopathy using a screw and washer technique. Journal of Small Animal Practice 30, 337–342. McKee, W.M., Lavelle, R.B., Richardson, J.L., Mason, T.A. (1990) Vertebral distraction-fusion for cervical spondylopathy using a screw and double washer technique. Journal of Small Animal Practice 31, 21–26. McKee, W.M., Butterworth, S.J., Scott, H.W. (1999) Management of cervical spondylopathy-associated intervertebral disc protrusions using metal washers in 78 dogs. Journal of Small Animal Practice 40, 465–472.
McLaughlin, M.R., Purighalla, V., Pizzi, F.J. (1997) Cost advantages of twolevel anterior cervical fusion with rigid internal fixation for radiculopathy and degenerative disease. Surgical Neurology 48, 560–565. Mitchell, R.A.S., Innes, J.F., McNally, D. (2001) Pressure profilometry of the lumbosacral disk in dogs. American Journal of Veterinary Research 62, 1734–1739. Olby, N. (1999) Current concepts in the management of acute spinal cord injury. Journal of Veterinary Internal Medicine 13, 399–407. Olsson, S.E., Stavenborn, M., Hoppe, F. (1982) Dynamic compression of the cervical spinal cord. A myelographic and pathologic investigation in Great Dane dogs. Acta Veterinaria Scandinavica 23, 65–78. Panciera, D.L., Johnson, G.S. (1994) Plasma von Willebrand factor antigen concentration in dogs with hypothyroidism. Journal of the American Veterinary Medical Association 205, 1550–1553. Panciera, D.L., Johnson, G.S. (1996) Plasma von Willebrand factor antigen concentration and buccal mucosal bleeding time in dogs with experimental hypothyroidism. Journal of Veterinary Internal Medicine 10, 60–64. Pietila, T.A., Stendel, R., Schilling, A., Krznaric, I., Brock, M. (2000) Surgical treatment of spinal hemangioblastomas. Acta Neurochirurgica (Wein) 142, 879–886. Queen, J.P., Coughlan, A.R., May, C., Bennett, D., Penderis, J. (1998) Management of disc-associated wobbler syndrome with a partial slot fenestration and position screw technique. Journal of Small Animal Practice 39, 131–136. Read, R.A., Robins, G.M., Carlisle, C.H. (1983) Caudal cervical spondylomyelopathy (wobbler syndrome) in the dog: a review of thirty cases. Journal of Small Animal Practice 24, 605–621. Rendano, V.T., Jr, Smith, L.L. (1981) Cervical vertebral malformationmalarticulation (wobbler syndrome)—the value of the ventrodorsal view in defining lateral spinal cord compression in the dog. Journal of the American Animal Hospital Association 17, 627–634. Roosen, K., Grote, W., Liesegang, J., Linke, V. (1978) Epidural temperature changes during anterior cervical interbody fusion with polymethylmethacrylate. Advances in Neurosurgery 5, 373–375. Rosenbluth, P.R., Meirowsky, A.M. (1953) Sympathetic blockade, an acute cervical cord syndrome. Journal of Neurosurgery 10, 107–112. Rusbridge, C., Wheeler, S.J., Torrington, A.M., Pead, M.J., Carmichael, S. (1998) Comparison of two surgical techniques for the management of cervical spondylomyelopathy in Dobermanns. Journal of Small Animal Practice 39, 425–431. Seim, H.B. (1986) Caudocervical spondylomyelopathy. 14th Annual Veterinary Surgical Forum 1986; 72–78. Seim, H.B., Prata, R.G. (1982) Ventral decompression for the treatment of cervical disk disease in the dog: a review of 54 cases. Journal of the American Animal Hospital Association 18, 233–240. Seim, H.B., Withrow, S.J. (1982) Pathophysiology and diagnosis of caudal cervical spondylo-myelopathy with emphasis on the Doberman Pinscher. Journal of the American Animal Hospital Association 18, 241–251. Seim, H.B., Withrow, S.J. (1983) Ventral decompression for the treatment of herniated cervical intervertebral disk in the dog. In: M.J. Bojrab (ed.), Current Techniques in Small Animal Surgery, 2nd edn, 544–548. Philadelphia: WB Saunders. Shaffrey, C., Wiggins, G., Piccirilli, C., Young, J., Lovell, L. (1999) Modified open-door laminoplasty for treatment of neurological deficits in younger patients with congenital spinal stenosis: analysis of clinical and radiographic data. Journal of Neurosurgery 90 (4 Suppl), 170–177. Sharp, N.J.H., Wheeler, S.J., Cofone, M. (1992) Radiological evaluation of ‘wobbler’ syndrome—caudal cervical spondylomyelopathy. Journal of Small Animal Practice 33, 491–499. Sharp, N.J.H., Cofone, M., Robertson, I.D., DeCarlo, A., Smith, G.K., Thrall, D.E. (1995) Computed tomography in the evaluation of caudal cervical spondylomyelopathy of the Doberman Pinscher. Veterinary Radiology and Ultrasound 36, 100–108. Stokol, T., Parry, B.W. (1998) Efficacy of fresh-frozen plasma and cryoprecipitate in dogs with von Willebrand’s disease or hemophilia A. Journal of Veterinary Internal Medicine 12, 84–92. Swaim, S.F. (1975) Evaluation of four techniques of cervical spinal fixation in dogs. Journal of the American Veterinary Medical Association 166, 1080–1086. Teresi, L.M., Lufkin, R.B., Reicher, M.A., Moffit, B.J., Vinuela, F.V., Wilson, G.M., Bentson, J.R., Hanafee, W.N. (1987) Asymptomatic
231
232
Small Animal Spinal Disorders
degenerative disk disease and spondylosis of the cervical spine: MR imaging. Radiology 164, 83–88. Trotter, E.J., Brasmer, T.H., de Lahunta, A. (1975) Modified deep dorsal laminectomy in the dog. Cornell Veterinarian 65, 402–427. Trotter, E.J., de Lahunta, A., Geary, J.C., Brasmer, T.H. (1976) Caudal cervical vertebral malformation-malarticulation in Great Danes and Doberman Pinschers. Journal of the American Veterinary Medical Association 168, 917–930. Vaccaro, A.R., Venger, B.H., Kelleher, P.M., Singh, K., Carrino, J.A., Albert, T., Hilibrand, A. (2002) Use of a bioabsorbable anterior cervical plate in the treatment of cervical degenerative and traumatic disk disruption. Orthopedics 25 (10 Suppl), s1191–1199; discussion s1199. VanGundy, T. (1989) Canine wobbler syndrome. Part II. Treatment. Compendium on Continuing Education for the Practicing Veterinarian 11, 269–284. VanGundy, T.E. (1988) Disc-associated wobbler syndrome in the Doberman Pinscher. Veterinary Clinics of North America, Small Animal Practice 18, 667–696.
Wheeler, S.J., Sharp, N.J.H. (1994) Caudal cervical spondylomyelopathy. In: S.J. Wheeler and N.J.H. Sharp (eds), Small Animal Spinal Disorders—Diagnosis and Surgery, 135–155. London: Mosby-Wolfe. Williams, N., Tomlinson, J.L., Hahn, A.W., Constantinescu, G.M., WagnerMann, C. (1997) Heat conduction of fixator pins with polymethylmethacrylate external fixation. Veterinary and Comparative Orthopaedics and Traumatology 10, 153–159. Wilson, E.R., Aron, D.N., Roberts, R.E. (1994) Observation of a secondary compressive lesion after treatment of caudal cervical spondylomyelopathy in a dog. Journal of the American Veterinary Medical Association 205, 1297–1299. Zindrick, M.R., Wiltse, L.L., Widell, E.H., Thomas, J.C., Holland, W.R., Field, B.T., Spencer, C.W. (1986) A biomechanical study of intrapeduncular screw fixation in the lumbosacral spine. Clinical Orthopaedics and Related Research 203, 99–112.
PROCEDURES Paramedian approach to the ventral neck Excessive extension should be avoided during positioning as it tends to close the dorsal part of the disc space and also increases spinal cord compression. Rather than position the neck in extension to improve access to the caudal vertebrae, the dog is positioned with the neck in a relatively neutral position and with a weight tied to the rostral maxilla. The mild traction also tends to open the disc space, keep the anulus under tension and reduce spinal cord compression (Goring et al., 1991) (11.24). Exposure to the caudal vertebrae may also be improved using the lateral muscle separating approach and the strap muscles then serve to protect the trachea, recurrent laryngeal nerve, carotid sheath and jugular vein (11.25–11.27). Regardless of the approach used it is important to take great care during dissection near the thoracic inlet and to protect all exposed tissues with damp sponges (Cechner, 1980).
11.24
11.24 A: Patient positioning for a ventral decompression or distraction-stabilization using traction. The sandbag shown under the dog’s head can be removed to produce mild cervical extension if needed. B: Harvesting of cancellous bone from a proximal humerus (inset). Both shoulders should be included in the operative field. Bone graft can be used to promote fusion following a ventral slot, distraction-stabilization or after vertebral forage at an adjacent site (11.30). A
B
Cervical spondylomyelopathy
11.25 Initially the standard, midline, ventral approach to the neck is followed as described in Chapter 7 (7.17–7.21). Once the strap muscles have been exposed the surgeon can use a more lateral approach. The right sternothyroideus muscle is first separated from the right sternocephalicus muscle (Cechner, 1980). 11.25
11.26 An Army Navy retractor is used to retract all structures ventral to the longus colli muscles away from the surgeon except for the right sternocephalicus, which is pulled towards the surgeon (Cechner, 1980).
11.26
11.27 The dissection is continued through the loose cervical fascia until the longus colli muscles are exposed. Balfour self-retaining retractors are used to keep structures away from the midline while the dissection is continued to expose the vertebrae of interest. The esophagus is visible at the caudal end of the dissection (arrow). 11.27
Ventral decompression The ventral slot technique can be challenging in CSM because of ventral osteophytosis, malformed vertebral bodies, intraoperative hemorrhage, friable anulus fibrosus, or a combination of these (11.28, 11.29). Surgery at C6/7 may be especially difficult but access can be improved by taking particular care with patient positioning, by use of traction (11.24) and possibly by using a paramedian approach (11.25–11.27). Decompression should be performed as described under 7.37–7.51. If the slot is too narrow the decompression will be inadequate and the outcome will be more like a fenestration (Chambers et al., 1986; Ellison et al., 1988). However, a slot that approaches 50% of the vertebral body carries a high risk of subluxation (Fitch et al., 2000; Lemarie et al., 2000). In general the slot should be about one third of the vertebral width unless it is shaped like an inverted cone (Goring et al., 1991; Lemarie et al., 2000) (7.50).
233
234
Small Animal Spinal Disorders
Osseous fusion is common following a ventral slot (Swaim, 1975; Gilpin, 1976; Sharp et al., 1995) (11.12). This may be beneficial in preventing motion at the operated site but probably causes increased stress at adjacent spaces (Fox et al., 1996; Hilibrand et al., 1999). Normal disc spaces can tolerate this stress but subclinically affected spaces suffer a high incidence of domino lesions (Bruecker et al., 1989a; Hilibrand et al., 1999) (Table 11.2). A technique is therefore needed to prevent domino lesions after single site ventral slot surgery. One technique that might accomplish this is to try to promote fusion at subclinically affected spaces by forage and grafting (11.30).
11.28 3D reconstruction of a CT scan made the day after a ventral slot was performed at C5/6 intervertebral space (same dog as in 4.25). Suction and excellent lighting, such as a fiberoptic source (5.3), are essential; access may be improved by modifying the slot into an inverted cone (7.50) (Goring et al., 1991).
11.28
11.29 A: Removal of anulus can be especially difficult in CSM. It can be aided by preserving a small knub of anulus during drilling (arrow). This can then be grasped like a handle to facilitate its removal after a window has been cut into the vertebral canal (Jeffery, 1995) (7.45–7.49). B: Hemorrhage from the vertebral venous plexus can be hard to control (7.44; Table 7.2); occlusion of the vertebral venous plexus is by pressure or a small piece of muscle tissue (8.43). The most likely puncture sites for the venous plexus are at each corner. The width of the slot in this illustration is much greater than generally recommended unless some type of stabilization is to be applied afterwards (11.41–11.43).
11.29 A
B
Cervical spondylomyelopathy
11.30 A: After a ventral slot at C6/7 (not shown) cortical bone was removed over the ventral aspect of C5 and C6 vertebrae (arrowheads) without damaging the C5/6 disc (arrow). Cancellous bone was then 11.30 applied over both disc spaces A B (Dixon et al., 1996; Hilibrand et al., 1999). B: A ventral slot at C6/7 with forage of the ventral cortical bone on either side of C5/6 (different dog to A, see 11.5). Both sites show good ventral bridging with new bone at 6 weeks; this progressed to complete fusion by 4 months. The dog’s neurological deficits had resolved completely by 6 months.
Vertebral distraction (11.31–11.34) The temporary distraction device can be custom-made as shown in 11.33A but a Freer or Synthes periosteal elevator is just as effective. Use of this initial distraction technique allows one side of the space to be curetted; the distractor is then repositioned to curette the other side. It is especially important to remove as much material as possible from the craniodorsal portion of the disc space, although this process is harder to perform properly at C6/7 than at other spaces (11.14). The dorsal anulus must be left intact to prevent cement from entering the vertebral canal. If the anulus is torn, it is possible to protect the dura prior to introducing cement using Gelfoam (Pharmacia, Kalamazoo, MI) as this has good thermo-insulating properties (Roosen et al., 1978; Boker et al., 1989). Distraction can be done using Gelpi retractors in the vertebral bodies as shown in 11.38A (Dixon et al., 1996) but ideally not by placing a retractor in the adjacent disc spaces (Bruecker et al., 1989a; Dixon et al., 1996; Wilson et al., 1994) (11.13A). Distraction can also be accomplished by an assistant pulling on a rope around the dog’s upper canine teeth (7.55) but is unreliable and often produces inadequate distraction (11.18). Finally, small K-wires may be used to maintain distraction like the screws in 11.40, but are then removed once the cement plug hardens. a
11.31 A: Sagittal section through the cervical spine to show compression ventrally from anulus (a) and dorsally from ligamentum flavum (b). B: Distraction relieves compression caused by redundant tissues. Gelpi retractors maintain distraction; one tip is against the caudal end plate of the cranial vertebra and the other in a small hole drilled in the body of the caudal vertebra (11.34). A rope pulling on the dog’s maxilla (11.24A), and tied to a hook on the surgical table, with or without the Gelpi distraction shown here, are the most reliable techniques to permit cement hardening.
11.31
b
Ventral
Dorsal
A Ventral
Dorsal B
235
236
Small Animal Spinal Disorders
11.32
11.32 A: Distraction has been started by making a wide fenestration at the C6/7 disc. The window in the ventral anulus fibrosus (a) and the nucleus pulposus (b) are visible. B: The C6/7 disc is curetted thoroughly following fenestration. This is easier with the space distracted as shown in 11.33B.
b
A
B
11.33 A: A distraction device is introduced into the intervertebral space to aid curettage. B: The device is then turned through 90° to distract the two vertebrae and widen the intervertebral space. When all nucleus and end-plate cartilage have been removed (Dixon et al., 1996), a rope or Gelpi retractor is used to maintain distraction as shown in 11.31B and 11.34.
11.33 A
B
a
Cervical spondylomyelopathy
11.34 A: A channel is drilled with a fine bur just under the cortical bone on the ventral surface of the vertebra. This hole starts about two thirds the distance along the vertebra caudal to the disc space to be distracted. One end of the Gelpi is hooked into this hole; the other is set against the caudal end plate of the vertebra in front of the disc space to be distracted (11.31B). The space is distracted as shown in 11.33B and the Gelpi is locked. B: The Gelpi retractor is locked. The initial distraction device is now removed and the space checked again to insure that no nuclear material remains. The space is now ready to be filled with cement. Care must be taken not to dislodge the Gelpi once it is in place; sudden collapse of the distracted space could injure the spinal cord (Dixon et al., 1996).
11.34 A
B
Cement plug (11.35–11.39)
11.35 Cement is mixed until the powder is dissolved and then it is transferred quickly to a 60-ml catheter-tipped syringe. Attached to the end of the syringe is an 8 French red rubber tube with the end cut short so that its tip just fits into the disc space. Cement is injected into the space taking care to fill the dorsal portion first without trapping air under the cement. The Gelpi is then left in place for 10–12 min.
11.35
The handle of a small instrument is used to compress the cement on either side of the Gelpi tip; this tip must be kept clear of cement to facilitate its removal after cement hardens. Ideally the two or three high-risk spaces are distracted using this procedure as shown in 11.16 and 11.37. Alternatively, any subclinically affected spaces can be foraged and then grafted as shown in 11.30. If no hole has been drilled into the end plate to anchor the cement plug, then a single 4.5-mm screw can be used to prevent the plug from falling out. A single screw is inserted by drilling and tapping the first cortex only; the screw then taps its own thread (11.36) as tapping may weaken fixation in cancellous bone (Krag, 1991; Bagley et al., 2000). The second (inner) cortex should prevent the screw from entering the vertebral canal (see 11.16 for potential complication). The advantages of using this technique to anchor the plug is that the vertebral end plates should be less likely to collapse as they are not weakened (11.20A, 11.38) and there is also no need for the dog to wear a splint after surgery. Some collapse of the space does still occur, presumably from thermal damage to the end plates (Anderson, 1988; Boker et al., 1989), and to pressure-induced resorption. This technique has worked
237
238
Small Animal Spinal Disorders
well in the small number of dogs studied provided that the screw is positioned so the head sits over the cement plug (11.36, 11.37). A similar approach uses threaded pins to anchor the plug instead (McKee and Sharp, 2003). The original method for anchoring a cement plug is to drill holes in each end plate and to fill these with cement (Dixon et al., 1996) (11.38). Having an angled bur guard for the drill (5.24) makes it easier to drill the anchor hole into the cranial end plate. It is not clear if an external splint provides additional benefit following the standard cement-plug technique. Although dogs that were not splinted had outcomes as good as those that were, most surgeons who use this technique recommend use of a splint (Dixon et al., 1996). The dog should be assessed closely for the appearance of rub sores, discomfort, odor or any type of malaise that might signify infection under the splint. If there is any doubt it should be replaced. A Halti and chest harness are an easier way to restrict movement (13.18B).
11.36 A: Cement has hardened (arrow). A screw, then cancellous bone is applied over the distracted space. B: Good new bone production is evident at C6/7 5 weeks after surgery 11.36 with nearly complete bridging callus. Note the B A radiolucent area ventral to the cement (arrow), which also occurs with the standard technique (Dixon et al., 1996). The implant survived this male St Bernard jumping a fence to reach a bitch in heat and falling on the far side 2 weeks after surgery. The dog was normal at 4 months.
11.37 A: These screw heads do not cover the cement plug adequately (11.16, 11.36). B: Follow-up radiograph 12 weeks after surgery reveals that the plugs have 11.37 fallen out (Marchevsky and Richardson, 1999). Despite A this the dog did well clinically and its neurological deficits had resolved by 4 months.
B
11.38 A: Distraction can be accomplished using a Gelpi retractor positioned within each vertebral body as shown here (Dixon et al., 1996). B: Excellent filling of the C5/6 disc space; cement can be seen clearly 11.38 where it has entered the caudal anchor hole A B (arrow). Despite this the plug failed due to end-plate collapse 4 weeks after surgery (11.20A) and the dog was euthanized. It had not worn an external splint although this is not a prerequisite for a successful outcome (Dixon et al., 1996).
Cervical spondylomyelopathy
11.39 A splint can be made in one piece from fiberglass; it is easiest to apply under general anesthesia but can also usually be done under heavy sedation. Whatever material is used has to be well padded and the entire splint should be replaced every 2 weeks, or sooner should problems arise. A Halti headcollar tied to a chest harness is a simple and effective way to limit neck movement (13.18B).
11.39
Metal implant and bone cement method (11.40–11.43) Initial vertebral distraction for the metal and bone cement technique cannot be performed using the method shown in 11.31 and 11.34 because of the final position of the cement. The disc space is first distracted as shown in 11.33B or 11.40. Then a bone allograft (Veterinary Transplant Services, Seattle, WA) or an autograft from either the wing of C6, the sternum or the ilium can be wedged into the slot to maintain distraction prior to cement application (Fitch et al., 2000). Use of such a graft has the additional advantage that it will promote an osseous union. Alternatively, distraction can be maintained by a rope around the maxillary teeth (11.24) that is then tied to the surgical table.
11.40 A: Another distraction option is to use two, 2.7-mm distraction screws. With the space distracted (11.33) these are directed from the ventral vertebral body to cross the disc space and rest against the end plate of the caudal vertebra (Queen et al., 11.40 1998). B: The distraction is asymmetrical as one of the B A distraction screws was aimed too laterally; this is exacerbated by the curved shape of the end plate (11.19). The anchoring pins or screws should enter the bone close to the midline (11.41) and are then driven towards the contralateral articular facets, away from the vertebral canal. Two implants are placed in the vertebrae on each side of the affected space. The vertebrae are then distracted and this position is maintained by molding bone cement carefully around each one of the screws. Even when a ventral slot is not needed to produce decompression, a shallow slot is useful in order to pack with bone graft and so promote fusion across the interspace. Gelfoam (Pharmacia, Kalamazoo, MI) can be used to insulate the graft as the cement hardens (Roosen et al., 1978; Boker et al., 1989). If pins or screws are to be combined with a ventral slot it is important to position the screws precisely as there is only a limited amount of bone available. There is more room for implants in the cranial vertebra if they are inserted a little more parallel to the end plates than shown in 11.43 (Jeffery and McKee, 2001). The screws have better purchase if they penetrate two cortices but there is less risk of penetrating the vertebral canal or damaging nerve roots if they do not (Swaim, 1975; Zindrick et al., 1986) (11.40, 11.43). If only one cortex is penetrated then the implant will be strengthened considerably by use of three screws in each vertebra (Garcia et al., 1994).
239
240
Small Animal Spinal Disorders
11.41 A: Placement of pins or screws in cervical vertebra for distraction-stabilization (13.56A). B: The diverging path of the screw in each vertebra is shown clearly in this dorsoventral radiograph. This dog had also undergone a ventral slot decompression; the cranial portion of the slot is visible (arrowhead).
11.41 B
A
11.42 Sagittal section shows how the ventral slot should be started almost entirely in the vertebra cranial to the disc. Owing to the angle of the end plates, and the position of the slot, there is often less room for screws in the cranial vertebra than in the caudal vertebra (11.17–11.19, 11.22, 11.40–11.43).
11.43 A: Postoperative radiograph of screws and bone cement in position after a ventral slot. A hole was only drilled and tapped for the first (ventral) cortex. Screw position is good but still there was only room for one 4.5-mm and one 3.5-mm screw in C6 due to the angle used and the proximity of the cranial edge of the slot (arrow). B: Pins can also be used as the metal implant. Positive profile pins are preferred to the smooth pins shown here. Fully threaded screws are also preferred to the partially threaded ones shown in 11.22B as the thread junction can act as a stress riser.
11.42
11.43 A
B
Cervical spondylomyelopathy
Dorsal decompression (11.44–11.55) Dorsal decompression may be limited to two vertebrae or it can extend over multiple vertebrae (Lyman, 1991). It is advisable to preserve the facet joints in dogs with CSM and so the limit of bone removal is usually the medial aspect of the facet joint(s) (Trotter et al., 1976; VanGundy, 1988). After the initial approach (11.44–11.47), the nuchal ligament is divided in the midline (11.48). The approach can be alongside the ligament rather than dividing it and it can even be sectioned if additional exposure is required. It can be repaired later but this is not necessary. There is a layer of thick fascia and ligamentum flavum between adjacent vertebrae. It can be removed using sharp dissection but this must be done with great care as the spinal cord is not protected by bone between the vertebral arches. It is probably safer to remove this layer of fascia after entering the vertebral canal on either side. Screws are preferred over wire for providing additional stability after dorsal laminectomy (Swaim, 1975; Trotter et al., 1976) (11.56, 11.57). Cartilage is debrided from the joint surfaces and bone debris, or ideally cancellous bone graft, is packed around the joints. The main risks with the screw fixation technique are facet fracture or nerve root damage. The latter could have severe consequences if major nerves to the brachial plexus are injured.
11.44 The dog is positioned with its limbs pulled cranially and its caudal neck supported by a sandbag. The site of incision when approaching the caudal cervical vertebrae is from C3 to T2.
11.44
11.45 Skin and superficial fascia incised and retracted to reveal the aponeurosis of the rhomboideus and trapezius muscles.
11.45
11.46 Diagram to show the position of muscles relative to underlying skeletal structures. The muscles are divided in the midline. Bleeding from the large neurovascular bundles that penetrate the fascia must be controlled to prevent postoperative hematoma (15.34). The nuchal ligament can be palpated at this stage. 11.46
241
242
Small Animal Spinal Disorders
11.47 The fascia overlying the nuchal ligament is divided and cleared. The ligament is freed back to the prominent spinous process of T1. Here the nuchal ligament has been mobilized, revealing the paired spinalis muscles (arrows).
11.47
11.48 A: The nuchal ligament attaches to the spinous process of T1 and then continues caudally with the interspinous ligament. The spinous process of C7 can usually be palpated at this stage. The relative size of C7 and T1 11.48 should be ascertained from lateral radiographs A as they may vary between animals. B: Nuchal ligament being divided.
B
11.49 Closer view showing spinalis muscles attached to spinous process of C7 (arrow).
11.49
Cervical spondylomyelopathy
11.50 The muscles are elevated from the spinous process (a) and lamina of the vertebra. The spinous process is then removed with bone cutters. a
11.50
11.51 The spinal cord is now visible (a). Large veins may be present in the epidural space (11.52). Bleeding may respond to bipolar cautery but a Hemoclip (Pilling Weck Inc., Research Triangle Park, NC) or fine suture is needed if a vein is torn.
a 11.51
11.52 CT scans of a tetraplegic Doberman after IV contrast. A: Image at the unaffected C6/7 disc space shows how the vertebral venous plexus communicates with intervertebral veins at the level of the disc space 11.52 (arrows). These veins may extend into the dorsal epidural A B space; the exact pattern is variable. B: Image through C5/6 disc space to show the spinal cord compressed by a lateralized, non-mineralized disc extrusion (arrowhead) that was confirmed at surgery (from Sharp et al., 1995).
11.53 The laminectomy is performed with a bur. Close attention is paid to the layers of bone as they are removed (8.32, 10.29, 10.30).
11.53
243
244
Small Animal Spinal Disorders
11.54 A: Spinal cord exposed (see also 7.59). The laminectomy can be continued laterally to remove proliferated articular processes and synovial tissue (11.57). B: The laminectomy can be continued to one side to expose the nerve root more fully. However, this almost certainly decreases stability. 11.54 A
B
11.55 A: Myelogram of a 6-year-old Rottweiler with marked tetraparesis of 12 months’ duration. There is a dilation of the ventral and, to an even greater extent, dorsal subarachnoid space. B: Intraoperative photograph after dorsal laminectomy to show the exposed spinal cord and dilation of the subarachnoid space (arrow). Marsupialization of the arachnoid cyst is in progress (14.5–14.8).
11.55 A
B
Cervical spondylomyelopathy
11.56 Two-year-old mastiff with severe osseous compression at C2/3 (11.57A) that had a dorsal laminectomy and 4.5-mm lag screws inserted across the facet joints. The 11.56 dog recovered well but developed mild left-sided A hemiparesis 3 months later. This follow-up myelogram shows good decompression at C2/3 (11.57B) compared to the preoperative myelogram (not shown) but a left-sided synovial cyst at C5/6 (11.8). The owner declined further surgery; the dog improved and was doing well 3 years later.
11.57 3D reconstructions of the (A) preoperative, and (B) 3-month postoperative CT myelograms at C2/3; same dog as shown in 11.8 and 11.56. Undercutting was used to remove compressive periarticular bone on the medial aspects of the facet 11.57 joints; this has widened the A vertebral canal significantly (10.43) (Trotter et al., 1976).
B
B
Laminoplasty Laminoplasty can be performed using a variety of techniques (McKee, 1988; Kohno et al., 1997; Shaffrey et al., 1999). In the method illustrated in 11.58, a fine-tipped bur is used to make a cut three quarters of the way along the midline of the C6 dorsal lamina. Great care is needed to avoid the spinal cord, especially cranially. The cut is extended into a T-shape caudally. Grooves are cut down to the inner cortical bone at the junction of the lamina with each pedicle. The laminar flap is elevated sufficiently to make a small bur hole in its medial edge for 0 polypropylene suture. The inner cortical bone is weakened further by drilling until each flap can be elevated almost vertically. The interarcuate ligament and any dural adhesions must be separated with care. 3.5-mm screws are placed in lag fashion across the C5/6 articular facets in order to attach and tighten the sutures. A cancellous bone graft is packed around the joints and a fat graft then placed over the spinal cord (McKee, 1988).
245
246
Small Animal Spinal Disorders
11.58 Laminoplasty of the cervical vertebrae as performed in a 5-year-old Great Dane with stenosis at C5/6 (McKee, 1988). Many different types of laminoplasty are used in humans (Shaffery et al., 1999) (from McKee, 1988).
11.58
Neoplasia
Clinical signs
Chapter
12
247
References
Tumor biology 248 Extradural tumors 248 Intradural/extramedullary Intramedullary 250
Procedures 266 Hemilaminectomy 266 Dorsal laminectomy 270 Vertebrectomy 273 Nerve sheath tumor resection
249
Diagnosis 250 Radiography 250 Ultrasound 250 Electrophysiology 250 Myelography 251 Computed tomography 251 Magnetic resonance imaging 251 Biopsy 251 Staging
263
275
Although tumors are uncommon causes of spinal disease, they are significant once the more common problems such as disc disease (in dogs) and trauma are eliminated. Older animals are usually affected, although certain tumor types do occur in young individuals.
252
CLINICAL SIGNS
Treatment 254 Medical treatment 254 Radiation treatment 254 Surgical treatment 255 Complications
258
Postoperative care
259
Prognosis 259 Extradural tumors 260 Intradural/extramedullary tumors Intramedullary tumors 261 Feline spinal tumors Diagnosis 262 Treatment 263 Prognosis 263
260
261
Key issues for future investigation
263
Animals with vertebral column tumors present with a fairly typical pattern of initial non-specific discomfort, followed by development of progressive neurological deficits and more definitive evidence of spinal pain. Severe pain and significant neurological dysfunction develop eventually if the condition is allowed to progress. Marked weight loss or localized muscle atrophy is often seen. One exception to this pattern is a precipitous decline following initial mild or even imperceptible signs. This can occur with vertebral body tumors where pathological fracture develops (12.12), or in soft tissue tumors where the animal tolerates a slowly progressing degree of spinal cord compression until it decompensates suddenly. Vascular interference by the tumor may also be responsible for acute deterioration. Most tumors of the spinal cord cause little or no discomfort. Although incontinence usually develops only after an animal has lost all motor function (see Chapter 2), some animals with intramedullary tumors develop urinary or fecal incontinence while they are still able to walk. This can occur even for lesions affecting
248
Small Animal Spinal Disorders
T3–L3 segments (Prata, 1977; Jeffery and Phillips, 1995) (12.4). Tumors involving the brachial or lumbosacral plexus may present initially as a slowly progressive unilateral lameness (McDonnell et al., 2001). This can present a diagnostic challenge and may lead to extended efforts to find a non-neurological abnormality; even including surgery for a suspected orthopedic lesion (McCarthy et al., 1993). Rectal examination may identify a mass within the pelvic canal (such as an enlarged lumbosacral trunk, the intrapelvic portion of the sciatic nerve, 1.8A) (Niles et al., 2001). Animals with brachial plexus tumors may have an axillary mass or show pain on palpation of the area (Carmichael and Griffiths, 1981; Brehm et al., 1995). The miosis of partial Horner’s syndrome (Bradley et al., 1982) (2.15A), unilateral loss of the cutaneous trunci reflex with a consensual response (Carmichael and Griffiths, 1981) (2.11, 2.12), or diaphragmatic hemiparalysis (13.12) may be other important clues to brachial plexus neoplasia (Bradley et al., 1982; Wheeler et al., 1986; Bailey, 1990). Signs tend to be insidious in onset, with spinal cord dysfunction occurring only if the vertebral canal is invaded. Even then half of all dogs with nerve sheath tumors and abnormal myelograms show no neurological deficits in the pelvic limbs (Bradley et al., 1982). The lower motor neuron (LMN) deficits make electromyography a useful test, first for confirming that the animal has neurological disease and then in identifying which nerve roots are involved (see ‘Electromyography’, page 250). Involvement of other body systems with diffuse neoplasia or metastasis may cause clinical signs unrelated to the nervous system. This is most frequent in animals with lymphoma. Paraneoplastic effects may also be seen, particularly lymphoma-related hypercalcemia.
TUMOR BIOLOGY There are a large number of tumor types that affect the spine. Many tumors are only reported in small numbers and these have been reviewed elsewhere (LeCouteur and Child, 1995; LeCouteur, 2001; Oakley and Paterson, 2003). Common tumor types are listed in Table 12.1. Spinal tumors may be classified by histopathology (Table 12.1) or according to their anatomic location in the spine (4.21): • Extradural. • Intradural/extramedullary. • Intramedullary. Extradural tumors are the most prevalent type in dogs, accounting for approximately half of all cases. Intradural/extramedullary tumors make up another third and intramedullary tumors the remainder (LeCouteur, 2001). Tumors of bone and nerve roots are most common in dogs; extradural lymphoma is most common in cats (Oakley and Patterson, 2003). The biology and expected behavior of a tumor must be understood if it is to be diagnosed, staged, and treated properly.
Extradural tumors As the name implies these tumors lie outside the dura mater. They typically cause pain with rapid neurological deterioration and include vertebral body tumors and those that occupy the epidural space. Most vertebral body tumors are osteosarcomas; fibrosarcomas are the next most common type (Dernell et al., 2000). Chondrosarcoma, hemangiosarcoma and myeloma are less frequent (Dernell et al., 2000; LeCouteur, 2001; Withrow and MacEwan, 2001). Meningiomas and nerve sheath tumors can be extradural in location
Table 12.1 Classification of major spinal tumor types according to location within the spine Extradural
Intradural/extramedullary
Intramedullary
Primary Osteosarcoma Fibrosarcoma Chondrosarcoma Lymphoma Hemangiosarcoma
Meningioma Nerve sheath tumours (schwannoma, neurofibroma, neurofibrosarcoma, lymphoma) Nephroblastoma Sarcoma Lymphoma
Ependymoma Glioma Lymphoma Hemangiosarcoma Reticulosis (focal granulomatous meningoencephalomyelitis (GME))
Metastatic
Metastatic
Metastatic Carcinoma Sarcoma Melanoma Lymphoma Myeloma
Neoplasia
but are discussed under intradural tumors. Extradural tumors may also be metastatic in origin (osteosarcomas or carcinomas). Discovery of an extradural tumor indicates the possibility of a primary mass elsewhere, particularly the mammary or thyroid glands, kidney or urinary bladder. Osteosarcoma and fibrosarcoma of the spine show similar biological behavior. The usual cause of treatment failure is local recurrence (Dernell et al., 2000). Spinal osteosarcomas are moderately aggressive with a metastatic rate reported as 17% (Heyman et al., 1992). A new grading system has been developed for osteosarcoma; most tumors are high grade (Kirpensteijn et al., 2002). Grading for fibrosarcoma is less clear cut (Ciekot et al., 1994; Hung et al., 2000). Vertebral plasma cell tumors may exist as a solitary plasmacytoma or as multiple myeloma. Solitary plasmacytoma is not associated with paraneoplastic syndromes and usually does not cause monoclonal gammopathy. It has a better prognosis than multiple myeloma but in humans about 50% go on to develop disseminated disease (Rusbridge et al., 1999) (12.12). This may be because a third of humans thought to have solitary disease actually have occult multiple myeloma on MRI (Moulopoulos et al., 1993). Multiple myeloma does cause monoclonal gammopathy and additional clinical signs can occur from hyperviscosity or paraneoplastic syndromes (Rusbridge et al., 1999). Lymphoma has the propensity to involve the dura mater and the spinal cord substance itself but is included here as the usual site is extradural (Britt et al., 1984). Canine lymphoma is usually a manifestation of multicentric disease. Lymphoma can affect nerve roots. Spinal cord compression is often extradural and the mass can sometimes be mistaken for epidural fat (Summers et al., 1994). Some dogs show meningeal infiltration and most have neoplastic lymphocytes in the CSF (Rosin, 1982; Britt et al., 1984; Couto et al., 1984). Grading techniques for multicentric lymphoma can predict outcome and these may be applicable to spinal lymphoma (Kiupel et al., 1999; Dobson et al., 2001). Malignant histiocytosis and mastocytosis also affect the spinal cord occasionally (Moore and Rosin, 1986; Tyrrell and Davis, 2001; Uchida et al., 2001; Moore et al., 2002). Feline lymphoma is discussed on page 261.
Intradural/extramedullary These tumors lie within the dura mater but outside the spinal cord parenchyma. The most common tumors in this category are meningiomas and nerve sheath tumors (neurofibroma, neurofibrosarcoma, schwannoma, lymphoma). Both types can also be in an extradural location.
Meningiomas usually cause pain or chronic discomfort with slowly progressive neurological deficits. Some meningiomas show an intramedullary pattern on myelography due to associated edema, even though they are located outside the spinal cord (Fingeroth et al., 1987). They tend to be discrete, firm to rubbery extramedullary masses. Meningiomas usually compress rather than infiltrate spinal cord but some show invasion along the perivascular space (Summers et al., 1994). Four of 13 spinal cord meningiomas in one report were invasive (Fingeroth et al., 1987). Distant metastasis has been reported for intracranial but not for spinal meningioma. Meningiomas show a remarkable histopathological diversity, probably because the meninges arise from both neural crest and mesodermal cells. They originate from arachnoid granulations and cap cells, and often have receptors for progesterone (Summers et al., 1994; Theon et al., 2000). Histopathological grading is not standardized although characteristics of benign versus malignant meningiomas have been reviewed (Ribas et al., 1991; Summers et al., 1994; Theon et al., 2000). Nerve sheath tumors usually cause a chronic, progressive lameness. The two most common sites of origin are from the plexus or from a nerve root within the vertebral canal (Brehm et al., 1995). Brachial plexus tumors are fusiform thickenings of one or more nerve trunks, which sometimes coalesce into a common mass. Spinal cord compression is usually a late event for brachial plexus tumors but is often a presenting sign for tumors that arise from within a nerve root (Brehm et al., 1995). Malignant peripheral nerve sheath tumor is the term preferred to malignant schwannoma or neurofibrosarcoma as it is usually not possible to discriminate the cell of origin. These tumors are usually malignant both cytologically and biologically (Summers et al., 1994). Nerve sheath tumors often invade extensively along the nerve (Oliver et al., 1965; Bradley et al., 1982; McCarthy et al., 1993; Brehm et al., 1995; Jones et al., 1995). Invasion into the spinal cord can occur but local or distant metastasis is unusual (Brehm et al., 1995; Sharp, 1988; Summers et al., 1994). They also metastasize occasionally along the subarachnoid space (2.24, 4.27, 4.28, 4.44B). Histopathologically, nerve sheath tumors are often anaplastic with abundant mitosis and necrosis. Specific grading systems are not yet standardized (Summers et al., 1994; Kuntz et al., 1997). Nephroblastoma is another type of intradural/ extramedullary tumor that occurs between T10 and L2 vertebrae in young dogs (Pearson et al., 1997; Summers et al., 1988; Terrell et al., 2000) (12.40). It was referred to formerly as neuroepithelioma but this tumor is really an extrarenal nephroblastoma equivalent to Wilms’ tumor in humans (Pearson et al., 1997).
249
250
Small Animal Spinal Disorders
Spinal cord compression is often severe but the mass tends to be well-circumscribed and the prognosis has been very good after surgical excision in some dogs (Ferretti et al., 1993; Siegel et al., 1996). Metastasis is rare but some of these tumors are invasive locally (Baumgartner et al., 1988; Terrell et al., 2000).
Intramedullary Intramedullary tumors occur within the spinal cord substance and are the least common type. They typically cause neurological deficits that progress rapidly. Tumors are either gliomas (12.4) or, very occasionally, metastatic tumors (Waters and Hayden, 1990; LeCouteur, 2001). Staging relies mainly on neuroimaging and MRI gives the best definition for intramedullary lesions. However, MRI may not always differentiate intramedullary tumors from those in the intradural/ extramedullary compartment. This is due partly to the limitations of slice thickness and partly because some tumors are only classified accurately using histopathology (Kippenes et al., 1999). A rare, non-neoplastic differential diagnosis for an intramedullary spaceoccupying lesion is epidermoid cyst (Tomlinson et al., 1988).
Radiography Spinal tumors are usually relatively easy to diagnose using standard methods. Following a physical and neurological examination, routine blood work and urinalysis are performed. Radiography is often the technique with the highest diagnostic yield. Survey radiographs of the spine may not be diagnostic of soft tissue tumors, part of the value of negative findings lies in the elimination of obvious, destructive lesions and of differential diagnoses such as discospondylitis (14.11). The axial and appendicular skeleton should be surveyed in suspected plasma cell tumors in order to stage the tumor as solitary or disseminated (Rusbridge et al., 1999). Bony changes will be seen mainly in vertebral tumors (12.1) but other tumors, such as meningiomas and nerve sheath tumors can cause mineralization or bone remodeling. Spinal tumors do not metastasize frequently to the thorax but chest radiographs (three projections—both right and left laterals with ventrodorsal or dorsoventral views) should always be taken. Chest radiography is followed by abdominal ultrasound; in some animals the spinal tumor itself may be a metastatic lesion (Bentley et al., 1990; Jeffery, 1991).
Ultrasound DIAGNOSIS Diagnosis of a tumor is usually straightforward but degenerative conditions like disc disease (see Chapters 7 and 8) and degenerative myelopathy, congenital anomaly with late onset such as an arachnoid cyst, infectious inflammatory disease including discospondylitis and focal granulomatous meningoencephalomyelitis (GME), or vascular diseases such as ischemic neuromyopathy in both dogs and cats, warrant consideration (see Chapter 14). Bone tumors may cross a disc space but this is rare; this helps to differentiate them from discospondylitis (Moore et al., 2000).
A general examination of the abdomen is indicated to identify other sites of neoplastic involvement and evaluate intercurrent disease (Platt et al., 1998). Ultrasonography can also be very useful to identify some nerve sheath tumors (Platt et al., 1999; Niles et al., 2001) and may even be superior to MRI (Donner et al., 1994; Simonovsky, 1997; Platt et al., 1999).
Electrophysiology Evidence of denervation on electromyography (EMG) is highly suggestive of nerve sheath tumor in a dog with suspicious clinical signs (Wheeler et al., 1986; Steinberg, 1988; Brehm et al., 1995). Nerve conduction
12.1 A: Four-year-old German shepherd dog presenting with back pain and paraparesis. Survey radiographs show a destructive process affecting the T3 spinous process (arrows). Myelography revealed extradural compression of the spinal cord at this site (12.6A). B: Tenyear-old paraparetic Cocker spaniel with an osteoproductive process within and ventral to the T5 vertebral body (arrow-heads) (12.6B).
A
B
Neoplasia
studies, late wave assessment and cord dorsum potentials may also provide supportive evidence (Cuddon et al., 2003).
Myelography This provides useful information about the location of the tumor and its position within the vertebral canal relative to the dura mater and spinal cord (4.27–4.31). It is important to take two views (lateral and ventrodorsal) for correct localization (4.30, 4.31). Although it will remain an important imaging modality, myelography is invasive and image quality can vary due to technical problems such as contrast leakage or poor filling of the subarachnoid space (see Chapter 4). False-negative results can also occur, especially with nerve sheath tumors (Carmichael and Griffiths, 1981; Bradley et al., 1982; McCarthy et al., 1993) and myelography may miss as many as 42% of intramedullary tumors (4.41) (Grem et al., 1985; Waters and Hayden, 1990). Transverse imaging by CT or MRI in general allows earlier detection and better surgical planning than myelography (Waters and Hayden, 1990; Brehm et al., 1995).
Computed tomography Computed tomography excels in imaging tumors that destroy cortical bone (Dernell et al., 2000; Moore et al., 2000) (12.6A, 12.7). Administration of intravenous contrast may improve detection further, especially for some nerve sheath tumors (McCarthy et al., 1993; Niles et al., 2001). Beam-hardening artefact from the humerus can interfere with resolution of brachial plexus lesions (Platt et al., 1999). False-positive results can occur in some animals that show thickening of nerve roots on CT myelography suggestive of tumor (12.2).
L
A
Thickening can also be due to edema and not to neoplastic invasion (Simpson et al., 1999). MRI is probably better than CT myelography at differentiating tumor from edema (Simpson et al., 1999).
Magnetic resonance imaging This is now considered the gold standard imaging modality for human oncosurgery (Gilson, 2003) (see page 57). It is superior to CT for imaging metastases and occult myeloma within medullary bone (O’Flanagan et al., 1991; Rusbridge et al., 1999; Moore et al., 2000). Sagittal images with T2-weighting provide excellent lesion localization by identifying peritumoral edema. Nerve sheath tumors are an exception as they do not always show abnormalities on T2-weighting although they usually enhance well with contrast (Kippenes et al., 1999) (2.25, 12.3). As nerve sheath tumors are often located away from the midline within the axilla the study must be centerd to include soft tissues on the affected side (Kippenes et al., 1999; Levitski et al., 1999; Platt et al., 1999). For most tumor types, T1weighted images post-contrast are preferred for defining tumor volume and the relationship between the tumor and surrounding tissues. Assessment of bone marrow infiltration requires additional pulse sequences such as fat suppression. The main limitation of MRI is that some lower field-strength magnets are unable to image thin enough tissue slices to permit fine anatomic resolution (Kippenes et al., 1999).
Biopsy This may be done as a needle aspirate or as an incisional or excisional procedure. The choice is important and biopsy tracts must be within the final field of surgical or radiation treatment. Percutaneous fine needle aspiration
L
B
12.2 CT myelogram from a dog with chronic lameness, LMN deficits, Horner’s syndrome, and a painful mass in the right axilla. Myelography was unremarkable. A: A CT scan showed subtle, right-sided, extradural compression at the cranial edge of T1 vertebra (arrow). B: Filling defects caused by C8 nerve roots as they cross the subarachnoid space (arrows). The right ventral nerve root is much thicker than the left side. In this dog the ventral C8 and T1 nerve roots were invaded by tumor (12.49).
251
252
Small Animal Spinal Disorders
A
B
12.3 MRI of a dog with a 5-month history of unilateral pelvic limb lameness. It was unable to extend its stifle due to severe atrophy and paralysis of the quadriceps muscle, the patellar reflex was absent but the withdrawal reflex was intact. A: Post-contrast, T1-weighted image shows a high signal area in the dorsal portion of one psoas muscle (arrowhead). B: T2-weighted image with fat suppression reveals a high signal mass at the same location as shown in A (arrowhead). Diagnosis was a nerve sheath tumor.
A
B
C
12.4 Five-year-old dog with a 1-week history of mild paraparesis and dribbling urine. Myelography revealed lack of subarachnoid contrast from T10 to T13. A: CT-guided needle biopsy (arrowhead) at T11/12 was non-diagnostic. B: Hemilaminectomy revealed discolored spinal cord and a needle aspirate was done over T11. C: Cytological diagnosis was ependymoma or oligodendroglioma. Euthanasia was performed and histopathological diagnosis was poorly differentiated glioma (Fernandez et al., 1997). In humans the histological grade of ependymoma does not necessarily correlate with prognosis (Mork and Loken, 1977; Ritter et al., 1998).
of the spinal cord was shown to produce no neurological deficits when conducted at L1/2 or L2/3 in seven normal dogs (Schulz et al., 1994). It has also been used safely in five cats: one with an intramedullary lesion at L2/3 and four with epidural lesions (Irving and McMillan, 1990). Overall, percutaneous needle biopsy in humans is rapid and minimally invasive; accuracy rates range from 80 to 95% with image-guidance (Ayala and Zornosa, 1983; Schiff et al., 1997; Ashizawa et al., 1999; Gilson, 2003)(12.4). Incisional biopsy provides a higher diagnostic yield than percutaneous techniques and larger tissue samples make histological diagnosis more accurate. However, incisional biopsy necessitates two-stage procedures unless combined with frozen sections and it can also disseminate tumor cells if done poorly (Gilson and Stone, 1990). Excisional biopsy is better for spinal cord tumors as most are well localized and excision is also the treatment
of choice except for lymphoma or plasma cell tumors. Needle biopsy is preferable when lymphoma or plasma cell tumor is suspected (Rusbridge et al., 1999; Gilson, 2003).
STAGING General staging includes neurolocalization, screening for any concurrent or paraneoplastic disorders, radiographic survey of the entire vertebral column, abdominal ultrasound, and screening for pulmonary metastases using thoracic radiographs or spiral CT. Aspiration of lymph nodes and bone marrow may also be indicated (Withrow and MacEwan, 2001). Screening for bone metastases in humans is done best with a bone scan as this is more accurate than survey radiography although false-positive results have been reported in dogs (Bentley et al., 1990; Berg and Lamb, 1990; Rusbridge et al., 1999; Dernell
Neoplasia
12.5 Diagram of an anatomic classification system for the extent of tumor involvement within and adjacent to a vertebral body (modified from Gilson, 2003). See below for an explanation of zones I–IV.
A
A
B
B
12.6 A: CT scan of the dog shown in 12.1A reveals an osteolytic, expansile mass within the spinous process of T3 (zone I, arrowhead). B: Transverse, T1-weighted MRI of the dog shown in 12.1B reveals a large mass ventral to the vertebra (zone III, white arrowheads); it has also extended dorsally around the rib (*) into zone II and into the vertebral canal (white arrows). The spinal cord (black arrow) is displaced by tumor within the vertebral canal (zone IV) (12.41).
et al., 2000) (see page 59). Bone scan may also fail to identify plasma cell tumors (Rusbridge et al., 1999). A CSF sample should be collected as neoplastic cells will be identified occasionally. Neoplastic cells were found in CSF from 10 of 12 dogs with lymphoma, which is more often than for feline lymphoma (one third of cats, see ‘Feline spinal tumors’, page 262) (Rosin, 1982; Couto et al., 1984). Local and regional tumor imaging is done by myelography, CT or MRI. CT is very useful for looking at cortical bone but MRI is now the gold standard modality for both neural and vertebral tumors (Gilson, 2003) (12.3, 12.6B).
Anatomical classification of the extent of vertebral tumors is used to determine surgical approach and resection technique and may also assist in determining the subsequent fixation to be employed. Classification is based on the region of spine involved and on the portion(s) of vertebral body affected. A guide for anatomic classification of vertebral body involvement is shown in 12.5 (Gilson, 2003). For most tumors of zone I and for smaller tumors of zone II, complete en bloc removal is recommended via a dorsal or dorsolateral approach (Klopp and Quinn, 2001; Gilson, 2003) (12.6A).
253
254
Small Animal Spinal Disorders
A
B
12.7 A: Survey radiograph of the lumbar spine in a 9-year-old Rottweiler with lumbar pain and severe paraparesis. There is loss of bony detail in the vertebral arch (arrow). B: CT scan shows destruction of the pedicle and part of the vertebral body (zones II and IV). The extent of bony involvement is clearer on this than on either the survey radiograph or the myelogram (not shown). An intraoperative view is shown in 12.9 (see page 256); the necropsy appearance in 12.8.
consideration of results in humans, there is a great deal of room for improvement in our management of animals with spinal tumors. It is probable that improvements will be incremental based on gradual advances in diagnostic and surgical techniques combined with earlier diagnosis and better adjuvant therapies (Gilson, 2003).
Medical treatment
12.8 Section through the lumbar vertebra at necropsy performed 4 weeks after imaging and laminectomy revealed the extent of the tumor re-growth, particularly involving the vertebral body and the lamina (zones II, III and IV). The CT image in 12.7 is very similar to this picture.
Resections in zone II at certain sites may be challenging because of the vertebral artery, ribs or wings of the ilium. Depending on their size, many tumors in zones II, III and IV will either require piecemeal intracapsular resection, vertebrectomy or both (Chauvet et al., 1999; Dernell et al., 2000) (12.6–12.8). Tumors in zone III must be assessed critically for adherence to cervical soft tissues or to the thoracoabdominal great vessels or organs (12.41) (Gilson, 2003).
TREATMENT Treatment is not possible in every animal with a spinal tumor, but an aggressive approach will prove rewarding in some patients. With the exception of myeloma, vertebral body tumors are difficult to treat, but the fundamental techniques are well established. Based on
Medical treatment of spinal tumors is unlikely to provide curative therapy in most patients. The exception is for animals with multiple myeloma where good responses are obtained with chemotherapy alone or in combination with radiotherapy; melphalan and prednisone are the standard chemotherapeutic agents (Rusbridge et al., 1999). In lymphoma, chemotherapeutic regimens are employed ideally without surgical intervention. Agents such as cytosine arabinoside may be added due to the difficulty of getting drugs across the blood brain barrier (Couto et al., 1984; Withrow and McEwan, 2001). Lomustine is of value for CNS tumors (Jeffery and Brearley, 1993; Moore et al., 1999; Fan and Kitchell, 2000) and oral hydroxyurea may prove useful for meningioma (Mason et al., 2002). Doxorubicin or cisplatin (in either systemic or slowrelease formulations) have been used to treat various other tumor types (Jeffery, 1991; Dernell et al., 2000). Palliative therapy (analgesics, anti-inflammatory agents) may also be used to alleviate pain and peritumoral edema, although when used alone they only relieve clinical signs for relatively short periods.
Radiation treatment Ideally, radiation treatment is used as an adjunct to surgery, unless the tumor is small, well localized and particularly sensitive to radiation (e.g. lymphoma or plasma cell tumor) in which case radiation may be used as the only therapy (LeCouteur, 2001; Rusbridge et al., 1999; Withrow and MacEwan, 2001). An important consideration is that previous surgery decreases the radiation tolerance of brain (and probably spinal cord) by causing injury to local blood vessels. If surgery reduces tumor
Neoplasia
burden to microscopic levels, then this is not crucial. If incomplete resection leaves a larger tumor burden, this factor may prevent the aggressive, postoperative radiation therapy necessary to control macroscopic disease (Smith and LaRue, 2000). Radiation can also cause injury to the spinal cord (Tiller-Borcich et al., 1987; Powers et al., 1992; Rusbridge et al., 1999): • Early delayed effects occur from 2 weeks to 3 months after therapy. They are probably due to transient demyelination; signs are often temporary and may respond to systemic corticosteroids. • Late effects occur 6 months or more after radiotherapy and are caused by vascular endothelial and glial cell injury. Signs are usually progressive; this is the main factor limiting radiation dose. More accurate radiation planning allows higher treatment doses to be delivered to the tumor while sparing normal tissues (Smith and LaRue, 2000). Several tumors affecting the spinal cord appear to respond to radiation (Siegel et al., 1996; Oakley and Patterson, 2003) (Tables 12.4–12.6). These include meningioma, ependymoma, nephroblastoma and some but not all nerve sheath tumors (12.49, 12.50).
Surgical treatment Owing to the inherent anatomical limitations of dealing with the CNS, aggressive surgical margins are rarely possible for vertebral or spinal cord tumors. Nevertheless, removal of vertebral tumors using wide margins for most of a mass, with a more limited margin for the portion nearest the spinal cord, may not affect adversely the success of local tumor control (Gilson, 2003). Effective biological barriers to neoplastic tissue should be used to advantage; these include ‘collagen rich/vascular poor’ structures such as intervertebral disc, cortical bone, dorsal longitudinal ligament, dura mater, interspinous and intertransverse ligaments, and fascial sheaths of paraspinal muscles (1.29, 1.32–1.35, 8.65). General principles of surgical oncology must be adhered to in order to insure a successful outcome (Gilson, 2003). Normal tissues must be protected by impenetrable barriers and copious lavage with suction used to clear the surgery field of tumor cells. Contaminated instruments should be discarded after tumor excision and new materials and instruments are used for closure. The most proximal portion of the specimen is identified and the tumor margins are assessed histologically (Gilson, 2003). The goal of surgery is to achieve complete resection or, at minimum, reduce tumor burden to residual microscopic disease (Algorithm 12.1). The best results are obtained in humans when the surgical approach and type of resection is targeted to
the exact location of the mass. Only 4% of patients with anterior (i.e. ventral) tumors worsened after an anterior approach whereas 26% worsened after posterior (i.e. dorsal) laminectomy. Similarly, over 30% of patients with posterior tumors could still walk after a posterior approach compared to less than 16% of patients with anterior tumors approached posteriorly (Gilson, 2003).
Vertebral tumors Dogs may show good initial improvement after debulking surgery (Moore et al., 2000), but then often worsen markedly (Bentley et al., 1990; Heyman et al., 1992; Chauvet et al., 1999) (12.9). Of 20 dogs undergoing debulking surgery, seven were unchanged (35%), eight were better (40%), and five (25%) worse after surgery. Improvement is usually temporary unless combined with effective adjunctive therapy (Dernell et al., 2000). Stabilization is required after most partial and all complete vertebral body resections (Chauvet et al., 1999; Dernell et al., 2000; Gilson, 2003); temporary stabilization is required prior to removing an entire vertebra (Chauvet et al., 1999; Gilson, 2003) (12.41). Principles of spinal stabilization after tumor resection are similar to those used for spinal trauma (Sundaresan et al., 1990) (see Chapter 13). The most versatile technique is metal implants and bone cement. An external fixator has some advantages but is better suited to techniques that do not use a cement spacer (Walker et al., 2002). Some studies report a high incidence of cement failure in humans over time (McAfee et al., 1986; Sundaresan et al., 1990; Gilson, 2003). This does not appear to be the case in animals following spinal trauma, probably because of the relative difference in life span, although this issue has not been studied in spinal neoplasia (see Chapters 11 and 13). Whether or not postoperative radiation will affect implants is also unclear (D.E. Thrall, personal communication). Titanium implants may be affected less than other metals and have the advantage that they also permit postoperative MRI. Following vertebrectomy the surgeon has three main options: 1. Replace excised bone using grafts such as freezedried femoral allograft (Veterinary Transplant Services, Seattle, WA) packed with autogenous, corticocancellous bone from a vertebral spinous process. The graft replaces the vertebral body and is anchored by plates (12.42) or with metal implants and bone cement. Autologous bone can also be harvested from ilium, rib, fibula, ulna or possibly a coccygeal vertebra (Yeh and Hou, 1994). Additional pieces of cortical graft or an adjacent rib are split longitudinally to make
255
256
Small Animal Spinal Disorders
Algorithm 12.1 Surgical decision-making in spinal tumours.
Neurological localization
Vertebral tumour
Survey radiographs of spine and thorax, plus abdominal ultrasound
Spinal cord tumour
Biopsy
Neuroimaging - CT or MRI or (myelography)
Laminectomy
Lymphoma, myeloma, nerve sheath tumor, see text
Resection under magnification
Anatomical classification
Surgical approach and margins
Zone I, Zone II (small)
Dorsal or dorsolateral approach – en bloc resection
Zone III
Ventral or lateral approach – intracapsular or vertebrectomy
Zone IV or crossing Zones II or III
Combined approach – vertebrectomy
b
d
A
c
a
B
12.9 Intraoperative photographs from the dog in 12.7 and 12.8. A: Exposure of the vertebral column revealed a mass of abnormal tissue on one vertebra (arrow). B: Hemilaminectomy was performed; much of the vertebral arch was absent. Debulking of the soft tissues revealed a large defect in the vertebral body (a). The spinal cord (b), spinal nerve (c) and spinal ganglion (d) are seen. The dog recovered well and was able to walk within 2 days. Histopathological diagnosis was anaplastic carcinoma. Attempts to find the primary tumor failed, and the dog was paraparetic again within 4 weeks, when it was euthanized (12.8).
protective strut grafts placed dorsal to the spinal cord (Chauvet et al., 1999; Gilson, 2003) (12.42). For further information see bone grafts (see page 292).
2. Fashion a spacer from bone cement to replace the vertebral body. This is anchored using metal implants and additional bone cement. The advantage of the spacer is that it provides
Neoplasia
immediate stability and, because there is no graft to revascularize, this technique is not affected by adjunctive therapies such as radiation (Brasmer and Lumb, 1972; Sundaresan et al., 1985; Matsui et al., 1994; Gilson, 2003). Ideally, some stabilization must also be provided to dorsal spinal elements. This could be fashioned from cement although thermal injury to the spinal cord must be avoided, such as by using Gelfoam (Roosen et al., 1978; Boker et al., 1989). An alternative is to add modified segmental fixation (12.41). 3. Simply close the gap and fuse the adjacent vertebrae to leave a shortened vertebral column. In one study the L2 vertebra was removed from eight normal dogs; five of these were clinically normal 11 months later (Yturraspe and Lumb, 1973). The advantage is that there is no spacer to loosen or revascularize and there is good protection dorsally for the spinal cord. Some histological changes were reported at the surgical site in normal dogs and it is not known how well a compromised spinal cord would accept this technique (Yturraspe et al., 1975). This is probably the best approach of the three although the ribs may complicate its use in thoracic areas. Nerve sheath tumors present a difficult surgical challenge but are also potentially curable if three criteria can be satisfied: 1. Several centimeters of normal nerve can be excised with complete histological margins proximal and distal to each branch of the mass. Simple visual inspection of margins usually underestimates the extent of invasion markedly (Bradley et al., 1982). 2. There is no invasion into the sympathetic nerves and ganglia or the spinal cord (Bradley et al., 1982; Bradney and Forsyth, 1986) (12.49, 12.50). 3. There is no local (McCarthy et al., 1993), or distant metastasis (Oliver et al., 1965; Bradley et al., 1982; Brehm et al., 1995). These tumors can sometimes invade through the epineurium and even into adjacent bone (Jones et al., 1995). It is usually necessary to amputate the forequarter or hemipelvis in order to satisfy these criteria (Bailey, 1990; Targett et al., 1993; Simpson et al., 1999) (Table 12.6). This is also necessary due to the diffuse nature of the tumor within nerves, the high recurrence rate, and the critical nature of much of the plexus for adequate limb function. For the thoracic limb, neoplastic infiltration of C8 and T1 spinal nerves, of more than one spinal nerve or a major portion of the plexus,
Nerve sheath tumors
necessitates that the limb be amputated as the resultant deficit to the radial, median and ulnar nerves will be too disabling. Good preoperative imaging is especially helpful in surgical planning, such as in determining the exact nerve roots involved and where to perform the laminectomy (Sakai et al., 1996; Platt et al., 1999).
Spinal cord tumors For spinal cord tumors, resection at the pseudocapsule between the mass and normal neural tissue can be curative for many tumor types. There is much less risk of postoperative instability for tumors of the spinal cord than for vertebral tumors. Extensive dorsal laminectomy warrants some form of stabilization, especially after pedicle or articular process removal (Siegel et al., 1996) (12.39). Ideally, surgery for spinal cord tumors should combine good preoperative imaging, preferably using MRI, intraoperative imaging with ultrasound and magnification, together with ‘notouch’ techniques employing lasers and ultrasonic aspiration (Gilson, 2003). As a minimum, good preoperative imaging with intraoperative magnification is preferred for intradural/extramedullary tumors (Ferretti et al., 1993). Magnification is essential for resection of intramedullary tumors in order to differentiate tumor, pseudocapsule and normal tissue; to minimize spinal cord manipulation; and to allow recognition and ligation of blood vessels that supply the tumor and not the normal spinal cord (Jeffery and Phillips, 1995; Gilson, 2003). Mass removal may lead to reperfusion injury of the spinal cord, and spinal cord manipulation can also cause damage. Use of preoperative vitamin E or perioperative methylprednisolone sodium succinate (MPSS) may be beneficial (Pietila et al., 2000) (see page 83). As microscopic tumor burden often remains after resection of many tumors, adjunct therapy using chemotherapy or radiation should be considered for follow-up treatment (Jeffery, 1991; Siegel et al., 1996). One possible exception is a well-circumscribed nerve sheath tumor resected with wide margins of normal nerve on either side or in dogs with more extensive nerve sheath tumors that undergo forequarter amputation and for which histological margins show no evidence of residual tumor (12.50).
TUMOR DISSECTION Gentle tissue handling is vital for tumors in or near the spinal cord. Manipulation must be kept to a minimum to avoid iatrogenic damage; it is preferable to avoid touching the spinal cord at all. If manipulation is necessary then dural sutures are used (Yturraspe and Lumb, 1973) (12.40). The spinal nerve and vessels can be sacrificed in
257
258
Small Animal Spinal Disorders
order to improve access if a tumor is difficult to remove (8.51), but this should be avoided if possible in the brachial and lumbosacral plexuses unless the limb can be sacrificed (Fingeroth et al., 1987). The tumor site is isolated with cottonoid pledgets (5.28) to both prevent dissemination of tumor cells and retard migration of hemorrhage into the subarachnoid space. The goal of surgery is to remove the tumor without causing damage to the spinal cord rather than to preserve a beautiful specimen (Fingeroth et al., 1987). If the tumor involves dura mater, sharp dissection is used to decrease spinal cord manipulation. Large dural defects seem to be of no consequence in animals and can be left open (14.8) (Fingeroth et al., 1987; Ferretti et al., 1993). Soft tumors can be removed by gentle suction but it is important to retain a piece of the mass for histopathology. For tumors that involve nerve it is crucial to mark the proximal margins of the mass accurately (12.50) (Targett et al., 1993; Niles et al., 2001). Intramedullary tumors are best approached via durotomy, piaotomy and dorsal myelotomy using magnification. An ultrasonic aspirator, microsurgical bipolar cautery and a laser are particularly helpful at this stage but good results may be obtained using bipolar cautery, loupes and copious irrigation with suction to preserve visibility (Jeffery and Phillips, 1995). Tumor removal is either by blunt dissection to create a plane between normal tissue and the mass, or the mass is debulked with the aim that the pseudocapsule then falls in on itself and a plane can then be developed between normal tissue and mass (Jeffery and Phillips, 1995; Gilson, 2003). Although considered highly desirable in people, intraoperative monitoring of spinal cord evoked potentials requires dedicated personnel and expertise that are rarely available for animals (Cuddon et al., 2003).
COMPLICATIONS The main sources of complications are listed in Table 12.2. Intraoperative complications include iatrogenic spinal cord injury, which can occur from manipulation, instability or vascular compromise (Brasmer and Lumb, 1972; Fingeroth et al., 1987) (Table 12.3). The surgeon may also be unable to find the tumor, especially if it is intramedullary. Intraoperative ultrasound is particularly useful in this situation (Nakayama, 1993). The tumor may be invasive and therefore difficult to resect, or access to the tumor can be difficult if it extends ventral to the spinal cord. Access may be improved by performing a partial corpectomy of the vertebral body (Moissonnier et al., 2002b) (8.47B). If nerve roots are involved at an intumescence it may be impossible to
Table 12.2 Main sources of complications
Intraoperative Technical problems Technical errors
Early postoperative
Late postoperative
Infection Lack of stability Non-neurological problems
Infection Implant failure
Table 12.3 Complications
Intraoperative Iatrogenic spinal cord injury Difficulty locating tumor Invasive tumor Poor access to tumor Incomplete resection
Early postoperative
Late postoperative
Instability Pathological fracture (12.12, 12.12) Implant failure Wound infection Sepsis Fat graft necrosis (12.10) Diagnostic error
Implant failure Tumor recurrence (12.11, 12.50) Peridural scar
achieve complete resection without amputating the limb (Fingeroth et al., 1987). Technical errors during surgery may result in instability in the early postoperative period (Table 12.3). Dorsal laminectomy, especially if combined with removal of the articular processes, causes a significant reduction in spinal stability (see page 286) and in humans up to 20% of patients experience neurological deterioration (Findlay, 1984, 1987; Gilson, 2003). Preoperative vertebral body collapse in humans also has a negative effect on outcome after laminectomy (Findlay, 1987; Santen et al., 1991; Chauvet et al., 1999) (12.11). Laminotomy has been used in children instead of laminectomy to avoid late deformity due to instability and may be warranted in certain situations for animals (Fingeroth and Smeak, 1989; Cristante and Herrmann, 1994; Gilson, 2003). Implant failure may occur due to inadequate fixation (Brasmer and Lumb, 1972; Yturraspe and Lumb, 1973; Onimus et al., 1996; Miller et al., 2000). Failure of bone cement tends to occur in humans over time although this is less likely in animals, probably due in part to differences in life span (McAfee et al., 1986; Jonsson et al., 1994; Yerby et al., 1998; Lu et al., 2001; Jang et al., 2002) (see page 291). However, the incidence of metal implant and bone cement failure could prove to be higher following tumor irradiation than it is in spinal trauma (D.E. Thrall, personal communication).
Neoplasia
12.10 Aseptic necrosis of fat graft after dorsal laminectomy. The dog’s neurological status deteriorated several days after surgery. Myelography revealed extradural compression. The graft was removed and the dog made a good recovery (8.8).
12.12 This dog had been diagnosed 5 years previously with a solitary plasma cell tumor in L4 when it was treated with melphalan, prednisone and irradiation (4.47–4.49); back pain recurred after 63 months. A lytic lesion on the spinous process of T7 was removed but the sample was non-diagnostic. Two months later the dog had sudden onset of tetraparesis due to a pathological fracture of C7. Plasma cells were found at necropsy in C7, T10, L4 and the spleen and kidney (Rusbridge et al., 1999).
A
a 10% rate of histopathological misdiagnosis has been reported and the rate may be up to 15% in humans (Levy et al., 1997a). Some problems only become apparent later in the postoperative period (Table 12.3). The most important problem is tumor recurrence. This can also cause instability but unless vertebral column failure is catastrophic (12.11, 12.12), it may respond to repeat surgery or adjunctive therapy (Bell et al., 1992; Schueler et al., 1993).
B 12.11 A: Lateral radiograph of the lumbar spine of a paraparetic 3-year-old mixed-breed dog with a destructive lesion in the L5 vertebral body (arrow). It underwent a dorsal laminectomy and biopsy. B: The dog became paraplegic with reduced deep pain sensation 2 days later. There is marked collapse of the L5 vertebral body compared to the preoperative radiographs. Histological diagnosis was fibrosarcoma (12.42) (from Chauvet et al., 1999).
Fat grafts used to cover dorsal laminectomies must be thin enough to revascularize (8.8, 12.12; see also ‘Laminectomy healing’, page 86). Further potential complications include sepsis, which is a particular risk for debilitated animals with cancer (Chauvet et al., 1999), and diagnostic errors. For example,
POSTOPERATIVE CARE Routine nursing considerations are discussed in Chapter 15. Any postoperative deterioration in neurological status will necessitate a high standard of nursing care (Chauvet et al., 1999). If radiation treatment is planned, it is usual to delay this for 10 days or until the wound has healed (Siegel et al., 1996).
PROGNOSIS In some circumstances where a tumor is not amenable to treatment because of its location or because a pathological fracture has occurred, euthanasia is indicated. Alternatively, palliative radiation therapy may be considered along with analgesia as discussed in Chapter 15.
259
260
Small Animal Spinal Disorders
Table 12.4 Outcome of dogs with extradural tumors Myxoma/ Myxosarcoma3 (n ⫽ 3)
Osteosarcoma3
Osteosarcoma2
Fibrosarcoma2
Fibrosarcoma1
Adjuvant treatment
None
None
Variable
Variable
Vaccine
Chemotherapy ⫾ radiation
Survival (months)
11⫹; 19; 35
⬍1; 10
4.4
3.7
24⫹
17, 26, 65*
Final outcome
N/A
N/A
Variable
Variable
N/A
Recurrence
Tumor type
Plasma cell tumors4
1 Chauvet et al., 1999; 2 Dernell et al., 2000; 3 Levy et al., 1997a; 4 Rusbridge et al., 1999. * One dog euthanized at 4 months due to radiation myelopathy. N/A, not available.
Treatment may only lead to short-term remission in some patients, but the quality of life during that period can be acceptable if pain is controlled adequately. Remission may be achieved for CNS lymphoma in some dogs. Survival times, however, range from days to a few months (Couto et al., 1984; Turner et al., 1992). Results may be improved by use of radiation therapy (LeCouteur, 2001).
•
Surgery without curative intent for soft tissue sarcomas may produce useful remission if combined with other treatment modalities (Dernell et al., 2000). However, the prognosis is poor if the animal has marked neurological deficits (Dernell et al., 2000; Gilson, 2003) (12.7–12.9), and is also influenced negatively by pathological fracture (Findlay, 1987; Chauvet et al., 1999) (12.11).
Extradural tumors Extradural tumors, with the exception of plasma cell tumors and lymphomas, carry a guarded prognosis (Dernell et al., 2000; Gilson, 2003) (Table 12.4). • Preoperative neurological status is one important determinant of final outcome in both humans and dogs (Dernell et al., 2000; Gilson, 2003). In humans, between 60 and 95% of patients who could walk well before surgery still did so after surgery but only 35–65% of paretic patients and less than 25% of paraplegic patients walked again (Gilson, 2003). • Another factor is anatomic location; four dogs with tumors in the dorsal spinal compartment were alive from 7 to 30 months after excisional surgery; a fifth died from complications after 2 months (Klopp and Quinn, 2001). • Local disease control is also an important prognostic factor for overall survival (Kuntz et al., 1997; Dernell et al., 2000). Dogs with soft tissue sarcomas that are resected with incomplete margins are 10 times more likely to have local recurrence than dogs with complete margins (Kuntz et al., 1997). One dog with a fibrosarcoma treated by vertebrectomy survived for over 2 years (Chauvet et al., 1999) (12.42). Another dog that underwent a tail amputation for a caudal osteosarcoma survived 2.5 years (Heyman et al., 1992).
Intradural/extramedullary tumors Intradural/extramedullary tumors have a variable prognosis depending largely on tumor type: • Meningiomas treated by surgical excision alone in nine dogs gave good long-term outcomes in six (Fingeroth et al., 1987). Survival is probably improved further by use of adjunctive therapies (Bell et al., 1992; Siegel et al., 1996). Even incomplete resection of meningioma can give a good outcome when followed by radiation therapy; three such dogs had 8-, 15- and 25-month survival times (Siegel et al., 1996) (12.28). A mean survival of 19.5 months was reported for another 10 dogs with meningioma (Moissonnier et al., 2002a). Survival data for dogs with meningioma are shown in Table 12.5. • Nerve sheath tumors on the other hand have given very disappointing results overall. This is particularly true for dogs with tumors that arise within the vertebral canal and for those undergoing incomplete resections (Bradley et al., 1982; Brehm et al., 1995; Jones et al., 1995; Kuntz et al., 1997; McCarthy et al., 1993; Schueler et al., 1993; Targett et al., 1993). Only 6 of 51 dogs (12%) where margins were not assessed remained disease free after 1 year (Brehm et al., 1995). Ideally all
Neoplasia
Table 12.5 Outcome of dogs with intradural/extramedullary lesions (excluding nerve sheath tumors (12.6)) Tumor type
Hemangio-4 sarcoma
Nephroblastoma5
Nephroblastoma2
Nephroblastoma8
Meningioma8
Meningioma7
Meningioma1
Meningioma3
Meningioma6
No. of dogs
1
1
1
1
6
2
1
9
10
Adjuvant treatment
Doxorubicin
Radiation
None
Radiation
Radiation
N/A
Radiation
None
None
Survival (months)
11
6⫹
36⫹
6.5
8–25 median ⫽ 13.5
46 & 47
19
5 live ⬍6 months; 1 alive at 36 months
Mean ⫽ 19.5
Final outcome
Metastasis
N/A
N/A
Recurrence
See citation
N/A
Recurrence
See citation
N/A
1
Bell et al., 1992; 2 Ferretti et al., 1993; 3 Fingeroth et al., 1987; 4 Jeffery, 1991; 5 Jeffery and Phillips, 1995; 6 Moissonier et al., 2002a; 7 Levy et al., 1997a; Siegel et al., 1996. N/A, not available. 8
Table 12.6 Outcome for nerve sheath tumors following complete excision Targett et al., 1993
Bailey, 1990
Platt et al., 1999
Niles et al., 2001
Siegel et al., 1996
Unpublished (NCSU)
Number of dogs
2
1
1
1
1
3
Survival (months)
9*, 18*
42*
9*
16*
25*
24*, 36*, 8**
Adjuvant
None
None
None
None
Radiation
None
* Without recurrence. ** Probable recurrence, no necropsy; histological margins had not been evaluated. NCSU, North Carolina State University.
involved branches are resected along with a generous margin of normal nerve and histological margins are evaluated. For tumors involving the plexus this nearly always necessitates concomitant amputation (Targett et al., 1993; Simpson et al., 1999) (12.49, 12.50). Usually nerve sheath tumors are contained within the nerve by an effective tissue barrier of epineurium. Therefore all neoplastic tissue can often be excised with complete margins, in which case these tumors should be curable (Targett et al., 1993; Ganju et al., 2001). Results for dogs where apparent complete excision was performed are shown in Table 12.6. Nerve sheath tumors show an unreliable response to radiation (Siegel et al., 1996) (12.49, 12.50). Malignant nerve sheath tumors in humans also have poor outcomes and the best survival times follow amputation (Ganju et al., 2001).
Intramedullary tumors Intramedullary tumor resection has only been reported in three dogs. Two had ependymomas and both underwent surgery and radiation therapy (Jeffery and Phillips, 1995; Siegel et al., 1996). One dog made a full recovery and survived for 70 months (Siegel et al., 1996). The second had a poorly differentiated tumor; it could walk well 3 months after surgery but was euthanized because it remained incontinent (Jeffery and Phillips, 1995) (see also 12.4). A third dog had a paraganglioma that was removed and the dog was still alive 10 months later (Poncelet et al., 1994).
FELINE SPINAL TUMORS Spinal tumors account for over half of all cases of feline spinal disease not associated with trauma (Wheeler, 1989). Several types of spinal tumors have been reported, with lymphoma being the most common.
261
262
Small Animal Spinal Disorders
Neurological signs are seen in 5–12% of cats with systemic lymphoma (Spodnick et al., 1992; Lane et al., 1994). Most lymphomas are solitary and extradural but in some cats the tumor may extend over multiple vertebrae or there may be distant meningeal infiltration (12.13). Nerve root infiltration is also reported (Spodnick et al., 1992; Lane et al., 1994). About two thirds of cats with lymphoma are feline leukemia virus (FeLV) positive (35/54) and have involvement of other organs (32/50), mostly in the kidney (Spodnick et al., 1992; Lane et al., 1994; Noonan et al., 1997). A less common cause of spinal lymphoma is feline immunodeficiency virus (FIV). In contrast to FeLVassociated lymphoma, which is a disease of T cells, most FIV-associated lymphomas are B cell in origin (Callanan, 1996) (12.14). Other tumor types
12.13 Myelography showing extradural spinal cord compression due to FeLV-associated lymphosarcoma in a 1-year-old cat that presented with a sudden onset of lower motor neuron signs. Treatment was not attempted.
A
include: meningioma (most common), osteosarcomas, chondrosarcoma, myeloma, lipoma, glioma and nerve sheath tumor (Mills et al., 1982; Shell et al., 1987; Wheeler, 1989; Levy et al., 1997b).
Diagnosis Some extradural tumors may affect the vertebral bodies, causing bony changes on survey radiographs, but this is not typical. Myelography or transverse imaging is required to confirm the diagnosis in most cats (Asperio et al., 1999). Compression is usually extradural in lymphoma (12.13, 12.14) although occasionally other patterns are seen, particularly if nerve roots are affected. Other findings that may indicate possible neoplasia include soft tissue opacities in the cranial thorax with mediastinal lymphosarcoma, organomegaly or pulmonary metastases. Other diagnostic tests are indicated to confirm the presence of a tumor and to define its nature. The bone marrow is involved in approximately 75% of cats (14/19) with extradural lymphoma and bone marrow aspiration is indicated for cytological analysis (Spodnick et al., 1992; Noonan et al., 1997). CSF contains malignant lymphocytes in only 11 of 31 of cats with CNS lymphoma (Spodnick et al., 1992; Lane et al., 1994; Noonan et al., 1997). In some instances, fine-needle aspiration (page 252) or biopsy at exploratory surgery may be the only means of making a definitive diagnosis
B
12.14 Myelogram from a cat with acute ataxia, paraparesis, generalized lymphadenopathy and herpes keratitis; it was FeLV negative but FIV positive. Neurological deficits localized to T3–L3 spinal cord; the thoracic limbs were normal. Neoplastic lymphocytes were detected in the CSF; flow cytology identified these as B-cells. A: Myelogram reveals extradural compression at T1–2, which was unexpected given the normal thoracic limbs. B: CT myelography confirms the mass as extradural (arrow). Response to chemotherapy was excellent and remission lasted 1 year. Radiation produced a second remission; lomustine produced a third, brief remission but the cat was euthanized 18 months after presentation.
Neoplasia
(Irving and McMillan, 1990; Spodnick et al., 1992). In an FeLV-positive cat with spinal disease related to an extradural soft tissue mass demonstrated by neuroimaging, it may be reasonable to make a diagnosis of lymphoma and to treat accordingly.
Treatment Chemotherapy often leads to a rapid improvement in neurological function but remission times may be brief. Chemotherapeutic protocols are described elsewhere (Withrow and MacEwan, 2001). Clearly, surgical treatment alone in lymphoma has only a limited role because of the multifocal nature of the disease (Spodnick et al., 1992). Radiotherapy may be of benefit in lymphoma, but should probably be combined with chemotherapy for the same reason (Lane et al., 1994). If lymphoma is not suspected strongly, biopsy is required to diagnose tumor type and can be combined with surgical excision of the mass (Levy et al., 1997b).
Prognosis Although initial improvement may be seen after therapy for lymphoma, long-term prognosis is unfavorable because of the propensity for multifocal involvement, local recurrence or the development of systemic FeLVrelated disease. Three of six cats treated with chemotherapy went into complete remission but this usually lasted less than 6 months (mean 3.5 months). Remission for more than 1 year was seen in only 2 of 14 cats, one of which underwent surgery and chemotherapy (Spodnick et al., 1992; Lane et al., 1994). Cats usually succumb to systemic effects of the disease (Spodnick et al., 1992; Lane et al., 1994).
Key issues for future investigation 1. How are the sympathetic nerves that arise from the brachial plexus best assessed for neoplastic involvement and what is the best way to dissect them out? 2. Should dogs with nerve sheath tumors that are resected with complete histological margins undergo postoperative radiation therapy? 3. Is a bone allograft preferable to a cement spacer following vertebrectomy? 4. Will radiation therapy adversely affect metal implant and bone cement stabilization after vertebrectomy? 5. Is spondylectomy without vertebral replacement preferable to replacement of the resected vertebral body with an allograft or cement spacer?
One cat with meningioma survived almost 4 years; median survival for four other cats was 6 months (Levy et al., 1997b). One cat with myeloma survived 3 months on prednisone therapy (Mills et al., 1982). Complete excision of a nerve sheath tumor resulted in a 6-year survival in one cat. Vertebral osteosarcoma may be less aggressive in cats than in dogs and a survival time of 4.5 years was reported in one cat even after incomplete excision. Non-lymphoid tumors in cats may therefore have a better prognosis than vertebral canal lymphoma (Levy et al., 1997b).
REFERENCES Ashizawa, R., Ohtsuka, K., Kamimura, M., Ebara, S., Takaoka, K. (1999) Percutaneous transpedicular biopsy of thoracic and lumbar vertebrae— method and diagnostic validity. Surgical Neurology 52, 545–551. Asperio, R.M., Marzola, P., Zibellini, E., Villa, W., Sbarbati, A., Osculati, F., Addis, F. (1999) Use of magnetic resonance imaging for diagnosis of a spinal tumor in a cat. Veterinary Radiology and Ultrasound 40, 267–270. Ayala, A.G., Zornosa, J. (1983) Primary bone tumors: percutaneous needle biopsy. Radiologic–pathologic study of 222 biopsies. Radiology 149, 675–679. Bailey, C.S. (1990) Long-term survival after surgical excision of a schwannoma of the sixth cervical spinal nerve in a dog. Journal of the American Veterinary Medical Association 196, 754–756. Baumgartner, W., De Vargas Peixoto, P.F. (1988) Spinal nephroblastoma (neuroepithelioma) in the dog. Kleintierpraxis 33, 75–76, 78. Bell, F.W., Feeney, D.A., Caywood, D.D., Hultgren, B.D. (1992) External beam radiation therapy for recurrent intraspinal meningioma in a dog. Journal of the American Animal Hospital Association 28, 318–322. Bentley, J.F., Simpson, S.T., Hathcock, J.T., Brawner, W.R., Powers, R.D. (1990) Metastatic thyroid solid-follicular carcinoma in the cervical portion of the spine of a dog. Journal of the American Veterinary Medical Association 197, 1498–1500. Berg, J., Lamb, C. (1990) Bone scintigraphy in the initial evaluation of dogs with primary bone tumors. Journal of the American Veterinary Medical Association 196, 917–920. Boker, D., Schultheiss, R., Probst, E. (1989) Radiologic long-term results after cervical vertebral interbody fusion with polymethyl methacrylat (PMMA). Neurosurgical Review 12, 217–221. Bradley, R.L., Withrow, S.J., Snyder, S.P. (1982) Nerve sheath tumors in the dog. Journal of the American Animal Hospital Association 18, 915–921. Bradney, I.W., Forsyth, W.M. (1986) A Schwannoma causing cervical spinal cord compression in a dog. Australian Veterinary Journal 63, 374–375. Brasmer, T.H., Lumb, W.V. (1972) Lumbar vertebral prosthesis in the dog. American Journal of Veterinary Research 33, 493–500. Brehm, D.M., Vite, C.H., Steinberg, H.S., Haviland, J., Van Winkle, T. (1995) A retrospective evaluation of 51 cases of peripheral nerve sheath tumors in the dog. Journal of the American Animal Hospital Association 31, 349–359. Britt, J.O., Jr, Simpson, J.G., Howard, E.B. (1984) Malignant lymphoma of the meninges in two dogs. Journal of Comparative Pathology 94, 45–53. Callanan, J.J., Jones, B.A., Irvine, J., Willett, B.J., McCandlish, I.A., Jarrett, O. (1996) Histologic classification and immunophenotype of lymphosarcomas in cats with naturally and experimentally acquired feline immunodeficiency virus infections. Veterinary Pathology 33, 264–272. Carmichael, S., Griffiths, I.R. (1981) Tumours involving the brachial plexus in seven dogs. Veterinary Record 108, 435–437. Chauvet, A., Hogge, G., Sandin, J., Lipsitz, D. (1999) Vertebrectomy, bone allograft fusion, and antitumor vaccination for the treatment of vertebral fibrosarcoma in a dog. Veterinary Surgery 28, 480–488. Ciekot, P.A., Powers, B.E., Withrow, S.J., Straw, R.C., Ogilvie, G.K., LaRue, S.M. (1994) Histologically low-grade, yet biologically high-grade, fibrosarcomas of the mandible and maxilla in dogs: 25 cases (1982–1991). Journal of the American Veterinary Medical Association 204, 610–615.
263
264
Small Animal Spinal Disorders
Couto, C.G., Cullen, J., Pedroia, V., Turrel, J.M. (1984) Central nervous system lymphosarcoma in the dog. Journal of the American Veterinary Medical Association 184, 809–813. Cristante, L., Herrmann, H.D. (1994) Surgical management of intramedullary spinal cord tumors: functional outcome and sources of morbidity. Neurosurgery 35, 69–74; discussion 74–76. Cuddon, P.A., Murray, M., Kraus, K. (2003) Electrodiagnosis In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1108–1117. Philadelphia: Elsevier Science. Dernell, W., Van, V.B., Straw, R., LaRue, S., Powers, B., Withrow, S. (2000) Outcome following treatment of vertebral tumors in 20 dogs (1986–1995). Journal of the American Animal Hospital Association 36, 245–251. Dobson, J.M., Blackwood, L.B., McInnes, E.F., Bostock, D.E., Nicholls, P., Hoather, T.M., Tom, B.D.M. (2001) Prognostic variables in canine multicentric lymphosarcoma. Journal of Small Animal Practice 42, 377–384. Donner, T.R., Voorhies, R.M., Kline, D.G. (1994) Neural sheath tumors of major nerves. Journal of Neurosurgery 81, 362–373. Fan, T.M., Kitchell, B.E. (2000) Lomustine. Compendium on Continuing Education for the Practicing Veterinarian 22, 934–936. Fernandez, F.R., Grindem, C.B., Brown, T.T., Jr, Sharp, N.J.H., Saulnier, M. (1997) Cytologic and histologic features of a poorly differentiated glioma in a dog. Veterinary Clinical Pathology 26, 182–186. Ferretti, A., Scanziani, E., Colombo, S. (1993) Surgical treatment of a spinal cord tumor resembling nephroblastoma in a young dog. Progress in Veterinary Neurology 4, 84–87. Findlay, G.F. (1984) Adverse effects of the management of malignant spinal cord compression. Journal of Neurology, Neurosurgery and Psychiatry 47, 761–768. Findlay, G.F. (1987) The role of vertebral body collapse in the management of malignant spinal cord compression. Journal of Neurology, Neurosurgery and Psychiatry 50, 151–154. Fingeroth, J.M., Smeak, D.D. (1989) Laminotomy of the axis for surgical access to the cervical spinal cord. Veterinary Surgery 18, 123–129. Fingeroth, J.M., Prata, R.G., Patnaik, A.K. (1987) Spinal meningiomas in dogs: 13 cases (1972–1987). Journal of the American Veterinary Medical Association 191, 720–726. Ganju, A., Roosen, N., Kline, D.G., Tiel, R.L. (2001) Outcomes in a consecutive series of 111 surgically treated plexal tumors: a review of the experience at the Louisiana State University Health Sciences Center. Journal of Neurosurgery 95, 51–60. Gilson, S.D. (2003) Neuro-oncologic surgery In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1277–1286. Philadelphia: Elsevier Science. Gilson, S.D., Stone, E.A. (1990) Surgically induced tumour seeding in eight dogs and two cats. Journal of the American Veterinary Medical Association 196, 1811–1815. Grem, J.L., Burgess, J., Trump, D.L. (1985) Clinical features and natural history of intramedullary spinal cord metastasis. Cancer 56, 2305–2314. Heyman, S.J., Diefenderfer, D.L., Goldsschmidt, M.H., Newton, C.D. (1992) Canine axial skeletal osteosarcoma: a retrospective study of 116 cases (1986 to 1989). Veterinary Surgery 21, 304–310. Hill, T.P., Lubbe, A.M., Guthrie, A.J. (2000) Lumbar spine stability following hemilaminectomy, pediculectomy, and fenestration. Veterinary and Comparative Orthopaedics and Traumatology 13, 165–171. Hung, L.C., Pong, V.F., Cheng, C.R., Wong, F.I., Chu, R.M. (2000) An improved system for quantifying AgNOR and PCNA in canine tumors. Anticancer Research 20, 3273–3280. Irving, G., McMillan, M.C. (1990) Fluoroscopically guided percutaneous fine-needle aspiration biopsy of thoracolumbar spinal lesions in cats. Progress in Veterinary Neurology 1, 473–475. Jang, J.S., Lee, S.H., Rhee, C.H. (2002) Polymethylmethacrylateaugmented screw fixation for stabilization in metastatic spinal tumors. Technical note. Journal of Neurosurgery 96, 131–134. Jeffery, N., Brearley, M.J. (1993) Brain tumours in the dog: treatment of 10 cases and review of recent literature. Journal of Small Animal Practice 34, 367–372. Jeffery, N.D. (1991) Treatment of epidural haemangiosarcoma in a dog. Journal of Small Animal Practice 32, 359–362. Jeffery, N.D., Phillips, S.M. (1995) Surgical treatment of intramedullary spinal cord neoplasia in two dogs. Journal of Small Animal Practice 36, 553–557. Jones, B.R., Alley, M.R., Johnstone, A.C., Jones, J.M., Cahill, J.I., McPherson, C. (1995) Nerve sheath tumours in the dog and cat. New Zealand Veterinary Journal 43, 190–196.
Jonsson, B., Jonsson, H., Jr., Karlstrom, G., Sjostrom, L. (1994) Surgery of cervical spine metastases: a retrospective study. European Spine Journal 3, 76–83. Kippenes, H., Gavin, P.R., Bagley, R.S., Silver, G.M., Tucker, R.L., Sande, R.D. (1999) Magnetic resonance imaging features of tumors of the spine and spinal cord in dogs. Veterinary Radiology and Ultrasound 40, 627–633. Kirpensteijn, J., Kik, M., Rutteman, G.R., Teske, E. (2002) Prognostic significance of a new histologic grading system for canine osteosarcoma. Veterinary Pathology 39, 240–246. Kiupel, M., Teske, E., Bostock, D. (1999) Prognostic factors for treated canine malignant lymphoma. Veterinary Pathology 36, 292–300. Klopp, L.S., Quinn, M.K. (2001) En-Bloc excision and comparative surgical staging of dorsal compartment vertebral tumors. Veterinary Surgery 30, 497. Knecht, C.D., Greene, J.A. (1977) Surgical approach to the brachial plexus in small animals. Journal of the American Animal Hospital Association 13, 592–594. Kuntz, C.A., Dernell, W.S., Powers, B.E., Devitt, C., Straw, R.C., Withrow, S.J. (1997) Prognostic factors for surgical treatment of softtissue sarcomas in dogs: 75 cases (1986–1996). Journal of the American Veterinary Medical Association 211, 1147–1151. Lane, S.F., Kornegay, J.N., Duncan, J.R., Oliver, J.E. (1994) Feline spinal lymphosarcoma: A retrospective evaluation of 23 cats. Journal of Veterinary Internal Medicine 8, 99–104. LeCouteur, R.A. (2001) Tumors of the nervous system. In: S.J. Withrow and E.G. MacEwan (eds) Small Animal Clinical Oncology, 3rd edn, 514–531. Philadelphia: WB Saunders. LeCouteur, R.A., Child, G. (1995) Diseases of the spinal cord. In: S.J. Ettinger (ed.) Textbook of Veterinary Internal Medicine, 629–695. Philadelphia: WB Saunders. Levitski, R., Lipsitz, D., Chauvet, A. (1999) Magnetic resonance imaging of the cervical spine in 27 dogs. Veterinary Radiology and Ultrasound 40, 332–341. Levy, M.S., Kapatkin, A.S., Patnaik, A.K., Mauldin, G.N., Mauldin, G.E. (1997a) Spinal tumors in 37 dogs: clinical outcome and long-term survival (1987–1994). Journal of the American Animal Hospital Association 33, 307–312. Levy, M.S., Mauldin, G., Kapatkin, A.S., Patnaik, A.K. (1997b) Nonlymphoid vertebral canal tumors in cats: 11 cases (1987–1995). Journal of the American Veterinary Medical Association 210, 663–664. Lipsitz, D., Bailey, C.S. (1995) Clinical use of the lateral cervical approach for cervical spinal cord and nerve root disease: eight cases. Progress in Veterinary Neurology 6, 60–65. Lu, W.W., Cheung, K.M., Li, Y.W., Luk, K.D., Holmes, A.D., Zhu, Q.A., Leong, J.C. (2001) Bioactive bone cement as a principal fixture for spinal burst fracture: an in vitro biomechanical and morphologic study. Spine 26, 2684–2690; discussion 2690–2691. Mason, W.P., Gentili, F., Macdonald, D.R., Hariharan, S., Cruz, C.R., Abrey, L.E. (2002) Stabilization of disease progression by hydroxyurea in patients with recurrent or unresectable meningioma. Journal of Neurosurgery 97, 341–346. Matsui, H., Tatezaki, S., Tsuji, H. (1994) Ceramic vertebral body replacement for metastatic spine tumors. Journal of Spinal Disorders 7, 248–254. McAfee, P.C., Bohlman, H.H., Ducker, T., Eismont, F.J. (1986) Failure of stabilization of the spine with methylmethacrylate. A retrospective analysis of twenty-four cases. The Journal of Bone and Joint Surgery— American volume 68, 1145–1157. McCarthy, R.J., Feeney, D.A., Lipowitz, A.J. (1993) Preoperative diagnosis of tumors of the brachial plexus by use of computed tomography in three dogs. Journal of the American Veterinary Medical Association 202, 291–294. McDonnell, J.J., Platt, S.R., Clayton, L.A. (2001) Neurologic conditions causing lameness in companion animals. Veterinary Clinics of North America, Small Animal Practice 31, 17–38. Miller, D.J., Lang, F.F., Walsh, G.L., Abi-Said, D., Wildrick, D.M., Gokaslan, Z.L. (2000) Coaxial double-lumen methylmethacrylate reconstruction in the anterior cervical and upper thoracic spine after tumor resection. Journal of Neurosurgery 92 (2 Suppl), 181–190. Mills, J.N., Eger, C.E., Robinson, W.F., Penhale, W.J., McKenna, R.P. (1982) A case of multiple myeloma in a cat. Journal of the American Animal Hospital Association 18, 79–82. Moissonier, P., Crigel, M.–H., Timmerman, P. (2002a) Spinal tumors in dogs: Retrospective study of 18 cases. Veterinary Surgery 31, 301.
Neoplasia
Moissonnier, P., Carozzo, C., Meheust, P. (2002b) Lateral corpectomy as a treatment of chronic disc herniation in 15 dogs. Veterinary Surgery 31, 296. Moore, A.S., London, C.A., Wood, C.A., Williams, L.E., Cotter, S.M., DA, L.H., Frimberger, A.E. (1999) Lomustine (CCNU) for the treatment of resistant lymphoma in dogs. Journal of Veterinary Internal Medicine 13, 395–398. Moore, G., Mathey, W., Eggers, J., Estep, J. (2000) Osteosarcoma in adjacent lumbar vertebrae in a dog. Journal of the American Veterinary Medical Association 217, 1038–1040. Moore, L.E., Garrett, L.D., Debey, B., Biller, D.S. (2002) Spinal mast cell tumor in a dog. Journal of the American Animal Hospital Association 38, 67–70. Moore, P.F., Rosin, A. (1986) Malignant histiocytosis of Bernese mountain dogs. Veterinary Pathology 23, 1–10. Mork, S.J., Loken, A.C. (1977) Ependymoma: a follow-up study of 101 cases. Cancer 40, 907–915. Moulopoulos, L.A., Dimopoulos, M.A., Weber, D., Fuller, L., Libshitz, H.I., Alexanian, R. (1993) Magnetic resonance imaging in the staging of solitary plasmacytoma of bone. Journal of Clinical Oncology 11, 1311–1315. Nakayama, M. (1993) Intraoperative spinal ultrasonography in dogs: normal findings and case-history reports. Veterinary Radiology and Ultrasound 34, 264–268. Niles, J.D., Dyce, J., Mattoon, J.S. (2001) Computed tomography for the diagnosis of a lumbosacral nerve sheath tumour and management by hemipelvectomy. Journal of Small Animal Practice 42, 248–252. Noonan, M., Kline, K., Meleo, K. (1997) Lymphoma of the central nervous system: a retrospective study of 18 cats. Compendium on Continuing Education for the Practicing Veterinarian 19, 497–502. Oakley, R.E., Patterson, J.S. (2003) Tumors of the central and peripheral nervous system In: D. Slatter (ed.) Textbook of Small Animal Surgery, 3rd edn, 2405–2424. Philadelphia: Elsevier Science. O’Flanagan, S.J., Stack, J.P., McGee, H.M., Dervan, P., Hurson, B. (1991) Imaging of intramedullary tumour spread in osteosarcoma. A comparison of techniques. The Journal of Bone and Joint Surgery—British volume 73, 998–1001. Oliver, J.E., Eubank, N.H., Geary, J.C. (1965) Neurofibrosarcoma in a dog. Journal of the American Veterinary Medical Association 146, 965–970. Onimus, M., Papin, P., Gangloff, S. (1996) Results of surgical treatment of spinal thoracic and lumbar metastases. European Spine Journal 5, 407–411. Parker, A.J. (1973) Distribution of spinal branches of the thoracolumbar segmental arteries in dogs. American Journal of Veterinary Research 34, 1351–1353. Pearson, G.R., Gregory, S.P., Charles, A.K. (1997) Immunohistochemical demonstration of Wilms tumour gene product WT1 in a canine ‘neuroepithelioma’ providing evidence for its classification as an extrarenal nephroblastoma. Journal of Comparative Pathology 116, 321–327. Pietila, T.A., Stendel, R., Schilling, A., Krznaric, I., Brock, M. (2000) Surgical treatment of spinal hemangioblastomas. Acta Neurochirurgica ( Wein) 142, 879–886. Platt, S.R., Sheppard, B.J., Graham, J., Uhl, E.W., Meeks, J., Clemmons, R.M. (1998) Pheochromocytoma in the vertebral canal of two dogs. Journal of the American Animal Hospital Association 34, 365–371. Platt, S.R., Graham, J., Chrisman, C.L., Collins, K., Sundeep, C., Sirninger, J., Newell, S.M., Chandra, S. (1999) Magnetic resonance imaging and ultrasonography in the diagnosis of a malignant peripheral nerve sheath tumor in a dog. Veterinary Radiology and Ultrasound 40, 367–371. Poncelet, L., Snaps, F., Jacobovitz, D., Pippendop, B., Balligand, M. (1994) Successful removal of an intramedullary spinal paraganglioma from a dog. Journal of the American Animal Hospital Association 30, 213–216. Powers, B.E., Beck, E.R., Gillette, E.L., Gould, D.H., LeCouter, R.A. (1992) Pathology of radiation injury to the canine spinal cord. International Journal of Radiation Oncology Biology Physics 23, 539–549. Prata, R.G. (1977) Diagnosis of spinal cord tumors in the dog. Veterinary Clinics of North America, Small Animal Practice 7, 165–185. Ribas, J.L., Carpenter, J.L., Mena, H., Schulman, F.Y., Lee, K. (1991) Central nervous system meningioma in the dog: a review of 50 cases. Journal of Neuropathology and Experimental Neurology 50, 373. Ritter, A.M., Hess, K.R., McLendon, R.E., Langford, L.A. (1998) Ependymomas: MIB-1 proliferation index and survival. Journal of Neurooncology 40, 51–57.
Roosen, K., Grote, W., Liesegang, J., Linke, V. (1978) Epidural temperature changes during anterior cervical interbody fusion with polymethylmethacrylate. Advances in Neurosurgery 5, 373–375. Rosin, A. (1982) Neurologic disease associated with lymphosarcoma in ten dogs. Journal of the American Veterinary Medical Association 181, 50–53. Rusbridge, C., Wheeler, S.J., Lamb, C.R., Page, R.L., Carmichael, S., Brearley, M.J., Bjornson, A.P. (1999) Vertebral plasma cell tumors in 8 dogs. Journal of Veterinary Internal Medicine 13, 126–133. Sakai, F., Sone, S., Kiyono, K., Maruyama, Y., Oguchi, K., Imai, N., Li, F., Matsubara, M., Ueda, H., Haniuda, M., Kubo, K., Honda, T., Ishii, K. (1996) Magnetic resonance imaging of neurogenic tumors of the thoracic inlet: determination of the parent nerve. Journal of Thoracic Imaging 11, 272–278. Santen, D.R., Payne, J.T., Pace, L.W., Kroll, R.A., Johnson, G.C. (1991) Thoracolumbar vertebral osteochondroma in a young dog. Journal of the American Veterinary Medical Association 199, 1054–1056. Schiff, D., O’Neill, B.P., Suman, V.J. (1997) Spinal epidural metastasis as the initial manifestation of malignancy: clinical features and diagnostic approach. Neurology 49, 452–456. Schueler, R.O., Roush, J.K., Oyster, R.A. (1993) Spinal ganglioneuroma in a dog. Journal of the American Veterinary Medical Association 203, 539–541. Schulz, K.S., Chickering, W.R., Shell, L., Doherty, M.A., Moon, N., Shires, P. (1994) Fluoroscopic–guided and direct spinal cord aspiration in the normal dog. Progress in Veterinary Neurology 5, 46–48. Sharp, N.J.H. (1988) Craniolateral approach to the canine brachial plexus. Veterinary Surgery 17, 18–20. Shell, L., Dallman, M.J., Sponenberg, P. (1987) Chondrosarcoma in a cat presenting with forelimb monoparalysis. Compendium on Continuing Education for the Practicing Veterinarian 9, 391–397. Shires, P.K., Waldron, D.R., Hedlund, C.S., Blass, C.E., Massoudi, L. (1991) A biomechanical study of rotational instability in unaltered and surgically altered canine thoracolumbar vertebral motion units. Progress in Veterinary Neurology 2, 6–14. Siegel, S., Kornegay, J.N., Thrall, D.E. (1996) Postoperative irradiation of spinal cord tumors in 9 dogs. Veterinary Radiology and Ultrasound 37, 150–153. Simonovsky, V. (1997) Peripheral nerve schwannoma preoperatively diagnosed by sonography: report of three cases and discussion. European Journal of Radiology 25, 47–51. Simpson, D.J., Beck, J.A., Allan, G.S., Culvenor, J.A. (1999) Diagnosis and excision of a brachial plexus nerve sheath tumour in a dog. Australian Veterinary Journal 77, 222–223. Smith, G.K., Walter, M.C. (1988) Spinal decompressive procedures and dorsal compartment injuries: comparative biomechanical study in canine cadavers. American Journal of Veterinary Research 49, 266–273. Smith, M.O., LaRue, S.M. (2000) Lesssons learned from 10 years of irradiating brain tumors. American College of Veterinary Internal Medicine Annual Conference, Seattle, WA, 2000; 305–307. Spodnick, G.J., Berg, J., Moore, F.M., Cotter, S.M. (1992) Spinal lymphoma in cats: 21 cases (1976–1989). Journal of the American Veterinary Medical Association 200, 373–376. Steinberg, H.S. (1988) Brachial plexus injuries and dysfunctions. Veterinary Clinics of North America, Small Animal Practice 18, 565–580. Summers, B.A., de Lahunta, A., McEntee, M., Kuhajda, F.P. (1988) A novel intradural extramedullary spinal cord tumor in young dogs. Acta Neuropathologica 75, 402–410. Summers, B.A., Cummings, J.F., de Lahunta, A. (1994) Veterinary Neuropathology. St Louis: Mosby. Sundaresan, N., Galicich, J.H., Lane, J.M., Bains, M.S., McCormack, P. (1985) Treatment of neoplastic epidural cord compression by vertebral body resection and stabilization. Journal of Neurosurgery 63, 676–684. Sundaresan, N., Schmidek, H.H., Schiller, A.L. (1990) Tumors of the spine: Diagnosis and Clinical Management. Philadelphia: WB Saunders. Targett, M.P., Dyce, J., Houlton, J.E.F. (1993) Tumours involving the nerve sheaths of the forelimb in dogs. Journal of Small Animal Practice 34, 221–225. Terrell, S.P., Platt, S.R., Chrisman, C.L., Homer, B.L., de Lahunta, A., Summers, B.A. (2000) Possible intraspinal metastasis of a canine spinal cord nephroblastoma. Veterinary Pathology 37, 94–97. Theon, A.P., LeCouteur, R.A., Carr, E.A., Griffey, S.M. (2000) Influence of tumor cell proliferation and sex-hormone receptors on effectivenessof radiation therapy for dogs with incompletely resected menin-giomas. Journal of the American Veterinary Medical Association 216, 701–707.
265
266
Small Animal Spinal Disorders
Tiller-Borcich, J.K., Fike, J.R., Phillips, T.L., Davis, R.L. (1987) Pathology of delayed radiation brain damage: an experimental canine model. Radiation Research 110, 161–172. Tomlinson, J., Higgins, R.J., LeCouteur, R.A., Knapp, D. (1988) Intraspinal epidermoid cyst in a dog. Journal of the American Veterinary Medical Association 193, 1435–1436. Turner, J.L., Luttgen, P.J., VanGundy, T.E., Roenigk, W.J., Hightower, D., Frelier, P.F. (1992) Multicentric osseous lymphoma with spinal extradural involvement in a dog. Journal of the American Veterinary Medical Association 200, 196–198. Tyrrell, D., Davis, R.M. (2001) Progressive neurological signs associated with systemic mastocytosis in a dog. Australian Veterinary Journal 79, 106–108. Uchida, K., Morozumi, M., Yamaguchi, R., Tateyama, S. (2001) Diffuse leptomeningeal malignant histiocytosis in the brain and spinal cord of a Tibetan Terrier. Veterinary Pathology 38, 219–222. Viguier, E., Petit-Etienne, G., Magnier, J., Diop, A., Lavaste, F. (2002) Mobility of T13-L1 after spinal cord decompression procedures in dogs (an in vitro study). Veterinary Surgery 31, 297. Walker, T.M., Pierce, W.A., Welch, R.D. (2002) External fixation of the lumbar spine in a canine model. Veterinary Surgery 31, 181–188.
Waters, D., Hayden, D. (1990) Intramedullary spinal cord metastasis in the dog. Journal of Veterinary Internal Medicine 4, 207–215. Wheeler, S.J. (1989) Spinal tumors in cats. Veterinary Annual 29, 270–277. Wheeler, S.J., Clayton-Jones, D.G., Wright, J.A. (1986) The diagnosis of brachial plexus disorders in dogs: a review of twenty-two cases. Journal of Small Animal Practice 27, 147–157. Withrow, S.J., MacEwan, E.G. (2001) Small Animal Clinical Oncology, 3rd edn. Philadelphia: WB Saunders. Yeh, L.S., Hou, S.M. (1994) Repair of a mandibular defect with a free vascularized coccygeal vertebra transfer in a dog. Veterinary Surgery 23, 281–285. Yerby, S.A., Toh, E., McLain, R.F. (1998) Revision of failed pedicle screws using hydroxyapatite cement. A biomechanical analysis. Spine 23, 1657–1661. Yturraspe, D., Lumb, W., Young, S., Gorman, H. (1975) Neurological and pathological effects of second lumbar spondylectomy and spinal column shortening in the dog. Journal of Neurosurgery 42, 47–58. Yturraspe, D.J., Lumb, W.V. (1973) Second lumbar spondylectomy and shortening of the spinal column of the dog. American Journal of Veterinary Research 34, 521–525.
PROCEDURES Hemilaminectomy (12.15–12.29) Cervical hemilaminectomy is described below. The ventral slot approach is rarely useful for spinal tumor exploration as access to the spinal cord and nerve roots is severely limited. Thoracolumbar hemilaminectomy is described in 8.11–8.37.
12.15
12.15 Patient positioning for cervical hemilaminectomy via a dorsolateral approach.
12.16 To approach the midcervical vertebrae, the incision can extend from the occipital protuberance to the spinous process of T2. The skin incision is to one side of the midline, which reduces any tendency for wound breakdown by avoiding tension over the 12.16 spinous processes. Good wound healing is important in case postoperative radiation therapy is planned. Good hemostasis is also vital; any seroma or hematoma formation should be attended to carefully (see 15.34).
Neoplasia
12.17 The skin and superficial fascia are incised to reveal the superficial muscles. Note the dorsal branches of the cervical nerves emerging in the midline and diverging laterally. Each is usually combined with an artery and vein, which must be ligated or cauterized as necessary. Further details of anatomy are covered in Chapters 7, 9 and 11.
12.17
12.18 The incision is continued in the midline through the muscular aponeurosis. The nuchal ligament is exposed. This may be retracted away from the surgeon, transected or divided in the midline.
12.18
12.19 The nuchal ligament has been divided in the midline as have the spinalis muscles. This reveals the muscular attachments to the spinous process of the cervical vertebra (arrow).
12.19
12.20 The spinalis muscles have been elevated from the spinous processes (a) and vertebral laminae (b) on the side of the spine to be approached.
a
a
b
12.20
b
267
268
Small Animal Spinal Disorders
12.21 The paraspinal muscles are elevated first from the articular process (a) and then from further ventrally. Branches of the vertebral artery emerge here and must be ligated if encountered. Bleeding from spinal veins is best occluded with a Hemoclip (Pilling Weck Inc., Research Triangle Park, NC) (5.31, 11.51). The spinous process (b) and lamina (c) are seen here. (1.16).
b c a
12.21
12.22 The articular process is removed with rongeurs. The laminectomy is commenced with a bur. Here the cartilage of the articular surfaces can be seen (arrow).
12.22
12.23 Site of hemilaminectomy. Here the bone has been removed in the cranialmost vertebra. The site of bone removal is shown in the caudal-most vertebra.
12.23
Neoplasia
12.24 Bone is removed until the vertebral canal is entered. Here the inner cortical bone is visible, and ligamentum flavum is exposed in the craniodorsal corner (arrow). The bone will be burred to eggshell thickness over the entire defect before the vertebral canal is entered (8.34–8.36, 10.34). The inner portion of the joint capsule is visible in the center of the defect (a).
a
12.24
12.25 The laminectomy is continued to reveal the spinal cord, nerve root (a) and spinal ganglion (b). Large branches of the vertebral artery and vein are present at this level; any large tear in a vein is best closed using a Hemoclip (Pilling Weck Inc., Research Triangle Park, NC) (5.31). Hemorrhage may also occur from the vertebral plexus. This is best controlled with Gelfoam (Pharmacia, Kalamazoo, MI), direct pressure, or, as in this case, a piece of macerated muscle (c).
a c
b 12.25
12.26 The laminectomy is continued into the intervertebral foramen to reveal the nerve root and spinal ganglion. The laminectomy may be extended as appropriate for removal of neoplasms (7.56–7.59, 11.54).
12.26
12.27 Myelogram of a dog that presented with a 3-month history of neck pain and right-sided weakness. An extradural mass is compressing the spinal cord at the level of C1 (arrowheads). 12.27 A
B
269
270
Small Animal Spinal Disorders
12.28 CT scan without subarachnoid contrast, same dog as in 12.27. A: Mineralized mass to one side of the vertebral canal prior to surgery (arrowhead). B: After surgery the mass is 12.28 still visible along with the laminectomy defect in C1. A B Complete resection was not possible as the mass had a consistency similar to bone. Histological diagnosis was meningioma; tissue consisted of cartilage, mature bone with bone marrow and a few scattered fusiform cells.
12.29 3D reconstruction of the dorsolateral hemilaminectomy (arrows) performed over the right side of C1 and the cranial aspect of C2. The residual mass of the meningioma is evident within the vertebral canal (arrowhead).
12.29
Dorsal laminectomy (12.30–12.40) Cervical dorsal laminectomy is described in 11.44–11.55. Thoracolumbar dorsal laminectomy is described below. Dorsal laminectomy is the approach of choice for most spinal tumors, although hemilaminectomy may be suitable in some (12.29, 12.40). Good access is often essential for complete tumor resection and to permit the extensive removal of surrounding dura necessary with meningioma. Access to ventrally located masses may be somewhat restricted and use of dural stay sutures (12.40, 14.5–14.8) or rhizotomy (8.51) may be necessary (Fingeroth et al., 1987). Partial corpectomy of the vertebral body adjacent to the mass (with occlusion of the vertebral venous plexus using bone wax) may improve access further (8.47B) (Moissonnier et al., 2002b). Dorsal laminectomy with removal of an articular process does cause significant instability (Smith and Walter, 1988; Shires et al., 1991) (12.39).
12.30
12.30 Positioning of dog for thoracolumbar dorsal laminectomy.
Neoplasia
12.31 The skin incision is made 1 cm to one side of the midline. The precise site of the incision is governed by the location of the lesion.
12.31
12.32 The skin is incised just off the midline, at least two vertebrae on each side of the vertebra to be approached. The superficial fascia is mobilized and retracted to reveal the lumbodorsal fascia. Here the fascia has been incised bilaterally to reveal the spinous processes.
12.33 Exposure of the vertebral bodies is done bilaterally using the methods described under thoracolumbar hemilaminectomy (8.12–8.27). The multifidus muscles have been elevated from the spinous processes (a) and the vertebral arch. This reveals the articular processes (b), and the longissimus tendon attaching to the accessory process (c). The muscle retraction is maintained with Gelpi retractors. The dissection is completed on one side before starting the other side. The spinous processes are then removed with rongeurs or bone cutters.
12.32
a b c
12.33
12.34 The spinous processes have been removed. The articular processes are still present (a) and the ligamentum flavum is visible between the vertebrae (b).
b
a
12.34
271
272
Small Animal Spinal Disorders
12.35 The laminectomy is commenced in the midline with a bur, preserving the articular processes at this stage. Here the cortical bone has been removed over two vertebrae, revealing dark cancellous bone. See Chapter 8 for details on technique for using the bur (8.34).
12.35
12.36 The cancellous bone has been removed with the bur to reveal the white, inner cortical bone. This is then thinned to eggshell thickness over the entire defect (8.33–8.36).
12.36
12.37 The final layer of bone is removed to reveal the spinal cord (arrow). This is achieved with fine rongeurs and a curette. This degree of exposure may be adequate for some circumstances.
12.37
12.38 The approach can be modified depending on requirements. Wide exposure of the tumor and spinal cord is desirable. If the spinal cord is swollen, care should be taken to remove enough bone to prevent injury by impingement of the bone edges (Fingeroth et al., 1987). Here the pedicle is removed between the articular processes to expose the lateral aspect of the spinal cord (arrow). The articular processes are preserved if at all possible.
12.38
Neoplasia
12.39 Here the left articular processes have been removed as have the pedicles of two vertebrae. This gives good access to the nerve root (a). Note the vertebral plexus lying on the floor of the vertebral canal (arrows); this should be avoided. Constrictive fibrosis is a potential risk with this degree of bone removal. Subsequent stabilization is advisable (Hill et al., 2000; Viguier et al., 2002; Gilson, 2003).
12.40 Nephroblastoma in a young dog with acute paraparesis. A: The dura has been opened and the edges retracted using silk stay sutures. The spinal cord, covered by nerve rootlets, has herniated through the durotomy. It is important to make a long enough laminectomy and durotomy to prevent distortion of spinal cord under pressure. B: Gentle dissection has teased out a large mass (arrows). One rootlet was enveloped by the mass but there was no attachment to the spinal cord. The dog underwent radiation therapy. It improved substantially but was then lost to follow-up.
273
Trauma
Chapter
13
Initial assessment 281 Other injuries 281 Cardiothoracic 282 Urinary tract 282 Other 282
Box 13.1 Differential diagnosis of spinal trauma
Neurological examination
■
Dural tear
■
Ischemic neuromyopathy (cats and dogs, see Chapter 14)
■
Traumatic feline ischemic myelopathy (see Chapter 14)
■
Spinal cord hemorrhage
■
Fibrocartilaginous embolus
■
Psoas muscle injury
■
Bilateral pelvic/long bone fracture or cruciate rupture
282
Radiography 283 Survey radiography 283 Additional imaging 283 Pathophysiology 286 Spinal fracture biomechanics 286 Biomechanics of fixation devices 289 Treatment 293 Choice of therapy 293 Anatomical location of the injury Postoperative care
295
297
Complications 297 Intraoperative complications 297 Early postoperative complications 299 Late postoperative complications 299 Feline spinal injuries Prognosis
301
301
Key issues for future investigation References
303
Procedures 305 Non-surgical 305 Metal and bone cement 309 Modified segmental fixation 315 External fixation 316
302
■
Congenital atlantoaxial instability
■
Degenerative disc disease
■
Cervical spondylomyelopathy
■
Pathological fracture, neoplasia
Trauma can damage the vertebra(e), disc(s), meninges, spinal cord, or any combination of these. Diagnosis is by history and physical findings in most dogs, but may be less obvious in cats (Box 13.1). After initial assessment of a spinal trauma patient, consider methylprednisolone sodium succinate (MPSS) if within 8 h of the injury (see page 83). General principles of treatment can be summarized as follows: • Establish a prognosis and stabilize other injuries. • Provide analgesia and consider whether or not to use MPSS (see page 83). • Reduce vertebral misalignment to relieve spinal cord compression. • Provide stability by cage confinement, an external splint, or an internal or external fixator. • Decompression by laminectomy is not usually necessary when vertebral alignment is good. It is recommended for marked extradural compression from hematoma, disc material or bone fragments.
INITIAL ASSESSMENT Other injuries The first priority is to treat shock and other lifethreatening disorders (13.1). A thorough and meticulous
282
Small Animal Spinal Disorders
reduce spinal cord perfusion (page 86), complicate anesthesia and may also be fatal (Murtaugh and Ross, 1988; Snyder et al., 2001). Dyspnea is also common after trauma and some animals may require ventilation (Campbell and King, 2000; Beal et al., 2001).
Urinary tract Bladder rupture usually causes discomfort on abdominal palpation, vomiting, depression and hematuria within 24–48 h. Damage to the more proximal urinary tract may take longer to manifest itself (Weisse et al., 2002). Pubic or perineal swelling indicates potential urethral trauma. 13.1 This dog sustained spinal trauma 2 h previously. It has been strapped to a rigid board for evaluation and transport. The head can also be taped down for cervical injuries. Early immobilization is very important as up to 50% of animals continue to deteriorate after injury, particularly when referral is delayed (McKee, 1990; Hawthorne et al., 1999). A muzzle would be a useful precaution. Aggressive intravenous fluid therapy is in progress.
Box 13.2 Other injuries associated with spinal trauma ■
Shock
■
Pulmonary or pleural lesions
■
Diaphragmatic hernia
■
Traumatic cardiomyopathy
■
Damaged urinary tract
■
Ruptured bile duct
■
Injury to other abdominal organs
■
Long bone or pelvic fractures
■
Multiple spinal injuries
■
Brachial or lumbosacral plexus injury
■
Soft tissue injuries
■
Head injuries
Other Bile duct rupture, although rare, can result in patient deterioration up to 15 days after injury (Neer, 1992) (2.1). As some injuries are not apparent immediately, the owner should be forewarned of the potential need for further diagnostic evaluation and treatment (Lanz et al., 2000). After initial assessment (including nociceptive evaluation—see below) the patient is given analgesics. Narcotic agonists are preferred unless there is concomitant hypoventilation; they should also be used with care after head injury (Table 15.1). Analgesia must be combined with some form of stabilization, so that the animal does not injure itself further. If presented within 8 h of injury high-dose MPSS should be considered although the available evidence for its use is not compelling (Hurlbert, 2000) (see page 83). Sedatives should not be used routinely because they may decrease protective muscle tone; adverse effects on the cardiovascular system may also exacerbate shock and compromise spinal cord blood flow (see page 86).
NEUROLOGICAL EXAMINATION physical examination is essential at this stage as many animals have non-neural injuries, which could be overlooked (Box 13.2). In one study, 20% of dogs with injuries to lumbar vertebrae had concomitant pelvic fractures; 33% had cardiopulmonary lesions (Turner, 1987). Brachial plexus injury must also be ruled out (page 30, 2.15, 2.26, 13.1).
Cardiothoracic Post-traumatic cardiac arrhythmias are common and are easily overlooked as they can be delayed in onset by up to 24 h. Continuous electrocardiogram (ECG) monitoring is recommended for 24 h or until any arrhythmia resolves. Arrhythmias can decrease cardiac output,
The neurological examination is essential to localize the deficit, to identify multiple lesions, and to determine the prognosis. The single, most important prognostic factor following spinal trauma is the presence or absence of nociception or deep pain sensation (2.21, 13.36B). If nociception is absent caudal to a traumatic lesion the prognosis for return of neurological function is poor (page 302) (Olby et al., 2003). Analgesics should be given only after this parameter has been assessed as they could alter the findings. Animals with cervical injuries very rarely lose nociception (see page 28) but if severely tetraparetic they must be assessed for hypoventilation (Griffiths, 1970; Boudrieau, 1997) (6.1). The Schiff– Sherrington response has no prognostic value (see pages 26, 32, 13.7) although severe vertebral displacement
Trauma
does (Bagley, 2000; Papadopoulos et al., 2002) (13.4). The animal must always be assessed for injury to the head and brachial plexus (Griffiths et al., 1974; AANS, 2000; Platt et al., 2001) (see pages 29, 30). Screening for brachial plexus avulsion injury is done by assessing pupil diameter, cutaneous trunci reflex, thoracic limb withdrawal and sensation to the digits (Griffiths et al., 1974; Faissler et al., 2002) (2.20, 2.26).
RADIOGRAPHY Survey radiography Once the patient has been stabilized and a neuroanatomical diagnosis has been reached, survey radiographs are taken. The clinician must remember that radiographs are no substitute for the neurological examination; it is not possible to estimate the neurological status from the radiographs alone unless there is 100% vertebral displacement (McKee, 1990; Bagley, 2000) (13.6, 13.35). Lateral views are taken first. Oblique views to assess the articular facets may be taken with the animal in lateral recumbency and the beam angled (13.2, 13.3), although CT is more accurate (13.12, 13.16). Ventrodorsal radiographs are best taken using a horizontal beam. Failing that, extreme care must be taken when positioning the animal for ventrodorsal or dorsoventral radiographs (4.38). Temporary stability should be provided by a wood or Perspex splint. Anesthesia or heavy sedation is undesirable prior to this as it will reduce the stabilizing effect of paravertebral muscle spasm (Blass et al., 1988). Multiple fractures are not common except when adjacent vertebrae are involved or with injuries caudal to the lumbosacral junction (Feeney and Oliver, 1980; Turner, 1987; McKee, 1990; Selcer et al., 1991; Hawthorne et al., 1999). The neurological localization and assessment for focal hyperesthesia should help to rule out multiple injuries. As lower motor neuron (LMN) lesions can mask upper motor neuron (UMN) deficits, the T3–L3 region should be surveyed when L4–S3 deficits are present, likewise the C1–5 region with C6–T2 deficits (see page 29).
Additional imaging Specific indications for spinal cord imaging by myelography, CT or MRI include: • Identification of additional lesions, such as when survey radiographs do not correlate with the neurological localization. • Identification of lesions that require removal such as extradural bone fragments, blood clots or disc material.
13.2 Diagram to illustrate assessment of the articular facets with the animal in lateral recumbency and with the X-ray beam angled. The facet joint closest to the film is highlighted with the beam angled as in a; the other by the beam angle b.
13.3 This oblique radiograph reveals a fracture of the articular facets (13.12).
These techniques require general anesthesia. Great care must be taken when intubating animals with cervical fractures and when positioning any animal with a fracture that might be unstable (4.38). Myelography and MRI often add significant time to the overall procedure. MRI and CT have the advantage of being non-invasive. Of the two, CT is faster but MRI provides better images of extradural and intradural injuries (Fehlings et al., 1999) (13.7). Additional imaging is not generally indicated in an animal with good deep pain sensation that is to be managed non-surgically.
MYELOGRAPHY Advantages of myelography are that: • The entire spine can be evaluated easily. • It may rule out differential diagnoses (Box 13.1). • Dural tears (Hay and Muir, 2000) (13.10) and spinal cord transection (13.4) can be
283
284
Small Animal Spinal Disorders
identified. The spinal cord may be transected even if radiographs show little evidence of subluxation as there could have been a transient yet complete luxation at the time of the original trauma (13.4). Routine myelography is not recommended for surgical candidates that have only mild or moderate neurological deficits. These animals should make a very good recovery provided that they do not deteriorate prior to definitive stabilization. The benefit of identifying a significant extradural lesion in such animals is outweighed by the risk of neurological deterioration secondary to
13.4 Myelographic appearance of a dog that has suffered an anatomical spinal cord transection following vertebral luxation. Note that there is now only minimal vertebral displacement.
A
B
the added manipulation and anesthetic time. In more severely affected animals the benefits of invasive imaging may outweigh the risks. However, care must be taken not to manipulate the spine excessively during injection of contrast. One important goal of managing animals with spinal trauma is to prevent deterioration, which is why non-invasive imaging by CT or MRI is preferred (Bagley, 2000).
COMPUTED TOMOGRAPHY Computed tomography should detect bone fragments within the vertebral canal (13.5, 13.12, 13.16). It is better than survey radiography at detecting articular facet fractures and it also allows fracture stability to be assessed using a three-compartment or column model (Denis, 1984; Shires et al., 1991). In addition, images can be used to make 2D and 3D reconstructions of the lesion (13.5, 13.6). After localizing the lesion, scans should be made through the region of interest (see page 55). A CT can aid the choice of entry points and trajectories if an implant is to be placed in the vertebral bodies (13.49– 13.57). The scan should identify bone fragments, mineralized disc and acute hemorrhage (Tidwell et al., 1994; Lanz et al., 2000). Use of Hounsfield units (i.e. CT numbers, a measure of tissue attenuation properties) might also distinguish spinal cord edema, which is potentially
C
13.5 A CT scan was used in this dog to look for bone fragments in the vertebral canal. A: Survey radiograph. Note the instability evident between this and image B. B: Sagittal view of 3D reconstruction. A small fragment is visible on the floor of the canal but it is not causing compression (arrowhead). C: Transverse view of 3D reconstruction. Fracture of the transverse process is visible. The postoperative radiograph for this dog is shown in 13.24. A myelogram was not performed.
Trauma
reversible, from intramedullary hemorrhage, which carries a poor prognosis (Ramon et al., 1997). CT may miss traumatic disc herniations although these are usually impact injuries with little residual mass effect. A disadvantage of CT is that it may underestimate significantly the extent of spinal cord compression and so should ideally be combined with an MRI (Tator et al., 1999). Both CT and MRI still carry the risk of fracture displacement during positioning. Animals with failure of the disc or end plate (13.5) may be prone to hyperextension injury (4.38). A temporary (non-magnetic for MRI, below) splint could be used to reduce the risk.
MR IMAGING A mid-sagittal, T2-weighted MRI allows a large area of the spinal cord to be surveyed for injury and can also provide a quantitative assessment of compression. MR
may provide prognostic information by distinguishing edema from hematoma. Hematoma often carries a poor prognosis whereas edema has the potential to be reversible (Ramon et al., 1997; Gopal and Jeffery, 2001) (13.7). The following are prognostic indicators in humans (Ramon et al., 1997; Selden et al., 1999): • Presence and extent of any intra-axial hematoma. • The extent of spinal cord edema. • The extent of spinal cord compression by extra-axial hematoma. Spinal cord swelling does not interfere with MRI as it can for myelography (Gopal and Jeffery, 2001) (see pages 57, 123, 8.1). The main problems with MRI are availability and cost; greater scan times and slice thickness may also be factors. The ideal way to image human spinal cord injury is using both MRI and a CT scan (Fehlings et al., 1999; Tator et al., 1999). 13.6 A: Survey radiograph of a 5-year-old Labrador that was hit by a car to show 100% displacement of L2 vertebra relative to L3 in the vertical plane. B: 3D reconstruction of a CT scan in the same dog, which illustrates 100% displacement in the lateral plane as well. The dog was euthanized as there was no hope of a functional recovery.
B
HL
A
R
A
B
13.7 This dog was paraplegic and incontinent with Schiff–Sherrington syndrome and UMN deficits 2 days after vehicular trauma. A: 3-mm, sagittal, T2-weighted image. T13–L1, L1–L2 and L2–L3 discs show subtle loss of signal and extend dorsally; there is high signal within the spinal cord adjacent to these discs consistent with edema. B: 3-mm, axial, T2-weighted image at L1–L2, same dog. Spinal cord is displaced to the left but not compressed. A low signal, extra-dural mass thought to be a hematoma is ventrolateral to the spinal cord (arrow). Diagnosis was traumatic disc herniation with focal spinal cord edema, malacia, or both (see also Gopal and Jeffery, 2001). Treatment was non-surgical; 8 days after trauma the dog could stand and it was normal after 1 month.
285
286
Small Animal Spinal Disorders
PATHOPHYSIOLOGY Huge forces are necessary to disrupt the vertebral column and so the spinal cord often sustains a much more severe injury than it does after a simple disc extrusion (Tator et al., 1999). Although secondary injury mechanisms are the same, the prognosis is much worse after trauma than it is after a disc extrusion for animals that have severe injuries (see page 302). In some animals the bony elements are spared but catastrophic failure of the disc can still cause serious spinal cord injury (traumatic disc herniation). The sudden trauma may tear the anulus causing a previously healthy nucleus pulposus to rupture explosively into the vertebral canal. This type of extrusion does not produce the usual mass effect associated with the mineralized material of degenerative disc disease. In the cervical region the explosion of nucleus pulposus is directed dorsolaterally by the dorsal longitudinal ligament and can cause devastating, asymmetrical neurological deficits. Hemiplegia results together with loss of sympathetic function and poor nociception on the affected side (Griffiths, 1970). Catastrophic disc failure in the thoracolumbar region can also cause severe spinal cord injury and the sudden change in subdural pressure may tear the dura mater (1.21, 13.10). Severe spinal cord injuries with dural tears can even occur after vigorous running or struggling without any vertebral injury (Hay and Muir, 2000; Yarrow and Jeffery, 2000).
Spinal fracture biomechanics Forces that act on the vertebral column include bending (dorsoventral and lateral), rotation, shear, and axial loading. These can interact to cause a variety of injuries, but regardless of the exact mechanism one of the most important clinical questions is ‘How stable is the fracture/luxation site?’ Estimates of stability can be made from knowledge of the elements in the spine that are compromised. The simplest method is to classify injuries into three types based on the anatomical compartment that has been compromised. Injuries may damage the dorsal compartment, the ventral compartment or both (Smith and Walter, 1985) (13.8). Another method to assess stability is to divide the spine into three compartments; however, the exact elements of each compartment vary depending on whether or not the soft tissues are included (Denis, 1984; Shires et al., 1991). A modification of these various schemes assesses the ability of the vertebral column to resist forces applied to it (13.9–13.16). This scheme assesses specifically the integrity of the vertebral body, which acts as a buttress to resist bending and axial loading, and of the articular facets, which resist all forces (Smith and Walter, 1985; Patterson and Smith, 1992). Articular facet integrity is
Dorsal
Ventral
13.8 Diagram to illustrate the dorsal and ventral compartments of the vertebral column, as referred to in 13.9–13.16. The major drawback of this method of classification is that the majority of injuries damage both dorsal and ventral compartments (Smith and Walter, 1985).
judged using lateral and oblique radiographs of the spine (13.2, 13.3) or preferably by CT or MRI (13.12, 13.16). The vertebral body is assessed readily by survey radiographs (13.14). The disc must also be assessed as it is the single most important stabilizing factor against rotation and lateral bending (Shires et al., 1991; Schulz et al., 1996). The main advantage of this scheme is that it gives some indication of stability and can serve as a guide to the fixation procedure best suited to each injury (see ‘Biomechanics of fixation devices’, page 289). The main categories of injury are listed in Table 13.1:
Table 13.1 Main categories of injury Failed component(s)
Intact component(s)
I Intervertebral disc
Vertebral buttress and articular facets Vertebral buttress and intervertebral disc Articular facets and intervertebral disc Vertebral buttress
II Articular facet III Vertebral buttress IV Articular facets and intervertebral disc IV Vertebral buttress and intervertebral disc IV Vertebral buttress and articular facets IV Vertebral buttress, articular facets and intervertebral disc
Articular facets Intervertebral disc None
Trauma
•
I: The disc has failed; the buttress and articular facets are intact. This injury is surprisingly unstable due to the importance of the disc for stability against rotation, lateral bending and
•
13.9 I: Vertebral buttress and articular facets are intact but the anulus has failed (arrow) (13.10).
•
extension (Smith and Walter, 1988; Shires et al., 1991; Schulz et al., 1996) (13.6, 13.10). It is sometimes combined with minor fracture(s) of the vertebral body, but in such cases the spine can usually still resist axial compression and some degree of dorsoventral bending (13.5, 13.17, 13.24). An external splint can provide adequate stability for many fractures in this category (Patterson and Smith, 1992). Metal and bone cement or an external fixator provide the best means of stabilization (Walker et al., 2002). II: A facet has failed; the buttress and disc are intact. Articular facet fractures that preserve the ventral buttress are relatively stable provided that the disc is intact (Shires et al., 1991; Schulz et al., 1996) (13.12). Facet joints are important for rotational stability; the metal and bone cement technique best resists this force (Shires et al., 1991). Failure of one facet alone causes minimal instability although cicatrix formation has been recorded as a late complication of articular facet fracture (Waters et al., 1994). III: The buttress has failed; the facets and disc are intact. Fractures of the ventral buttress are often unstable and are particularly susceptible to bending and collapse in axial compression (Walter et al., 1986) (13.14, 13.32A). Both external and internal fixators can give excellent results for this type of injury (Walter et al., 1986; Bagley, 2000; Walker et al., 2002; LeCouteur and Sturgess, 2003) (13.21, 13.26, 13.65). An external splint is not ideal for this type of injury.
A
B 13.10 This dog presented with paraplegia and a lack of deep pain sensation in one limb. A: Myelography revealed extensive spinal cord malacia with leakage of contrast over T12 (Lu et al., 2002) (13.52A, 14.18). There is an isolated ventral compartment injury; the ventral buttress is intact but the disc has failed. B: 3D reconstruction of a CT scan from the same dog. There is leakage of contrast (arrowhead) secondary to a presumed dural tear (1.21, 13.52). See 13.20 and 13.30 for follow-up.
13.11 II: The vertebral buttress and disc are intact; an articular facet is fractured (13.12).
287
288
Small Animal Spinal Disorders
L
A
13.12 II: This cat was tetraparetic and paralyzed on its left thoracic and pelvic limbs along with the ipsilateral diaphragm after being bitten in the neck by a dog. A: A triangular bone fragment is overlaying the vertebral canal at C5/6 (arrow). B: Transverse CT image at C5/6 shows an impacted left articular process. This fracture was considered stable. The facet was removed; no fixation was applied. The cat recovered well but had severe, residual LMN deficits in its left thoracic limb.
B
IV: Two or more components fail. An injury is usually very unstable when two components fail (13.15, 13.19, 13.32). Failure of two components is often associated with failure of the third (13.16). Any fixation technique must now withstand almost all the forces acting on the vertebral column. For most dogs the choice is between metal implants and bone cement or an external fixator. An external splint does not resist axial compression and is not ideal when there is major failure of the buttress (Patterson and Smith, 1992). Cage rest is also suboptimal for this injury. The scheme in 13.9–13.16 can only serve as a guide to stability . An additional way to assess vertebral stability is by the use of stress radiography. In its simplest form, this is done by just comparing vertebral positions in subsequent radiographs (13.17). The risks of more deliberate attempts to move the fracture are obvious and this is best done with caution under fluoroscopy. Regardless of the method used, our understanding of fracture stability remains incomplete (Walter et al., 1986; Schulz et al., 1996). Shear force and axial
•
13.13 III: The vertebral buttress is fractured; the articular facets and disc are intact (13.14).
A
B
C
13.14 This dog jumped out of a truck at 20 mph. It could walk and was continent but had decreased anal tone. There is a fracture of the L7 buttress; the facets are intact. A: Survey radiograph. B: Mid-sagittal 3D reconstruction of the same dog. C: Dorsal view of the 3D reconstruction. The facets are intact but overridden (10.48). Postoperative radiographs are shown in 13.21.
Trauma
13.15 IV: Vertebral buttress and articular facet fracture; the disc may also fail.
loading have not been tested; most models have not looked at the effect of vertebral body failure; the definition of instability has been questioned (Fox et al., 1996); and the strength that a fixator requires to counter each force is not known (Waldron et al., 1991; Schulz et al., 1996). Most internal fixation methods tested to date are far from ideal (Walter et al., 1986). The relative contribution of the four main forces acting on the vertebral column is also unknown, and almost certainly varies between patients. For example, dorsoventral bending occurs when a paraplegic animal is lifted or as it attempts to raise its hindquarters. Rotation may occur as an animal rises from lateral
A
B
recumbency or as it moves with its pelvic limbs dragging to one side. Axial loading occurs from paravertebral muscle spasm. Only the effects of isolated dorsoventral, lateral bending and rotational forces have been studied experimentally and these studies may not reflect the clinical situation accurately. It is very hard to predict stability for most injuries except for rare isolated facet lesions. Furthermore, good outcomes may be obtained sometimes using non-surgical management or with less rigid internal fixation techniques even in unstable fractures (Selcer et al., 1991) (13.63). Despite these shortcomings, the least stable fractures must be where there has been catastrophic failure of two or three components (Schulz et al., 1996; Shires et al., 1991) (13.16, 13.32). Although in vitro results for metal and bone cement implants have been conflicting, it is the internal fixation technique with the best overall success in clinical use (Garcia et al., 1994; Sharp et al., 1998; Bagley, 2000; Walker et al., 2002; LeCouteur and Sturgess, 2003). An external fixator can also be very strong and a splint can be successful in certain situations (Patterson and Smith, 1992; Walker et al., 2002).
Biomechanics of fixation devices NON-SURGICAL MANAGEMENT Animals with injuries in the cervical or lumbosacral regions often respond well to non-surgical management (Smith and Walter, 1985; Hawthorne et al., 1999). Analgesia is not usually required beyond 96 h. Intractable pain or neurological deterioration warrants a reassessment of therapy.
C
13.16 Images of a 1.6 kg dog with articular facet fractures; the vertebral buttress has also failed through the disc. The dog had decreased anal tone but could walk and was continent despite the marked displacement. A: Survey radiograph. B: Mid-sagittal 3D reconstruction of a CT scan through the fracture site. C: Transverse CT scan to show articular facet fractures. Spinal stapling was successful although not at all ideal for this fracture (13.63).
289
290
Small Animal Spinal Disorders
A
B
13.17 A, B: Severe instability of a spinal fracture as demonstrated by a change in position between two radiographs (McKee, 1990). Although there is only a small fracture, loss of the disc removes 30–40% of the spine’s overall stability (Shires et al., 1991).
It is inexpensive, there are no complications related to implants and the spinal cord is not subject to myelography, anesthetic hypotension or manipulation. Hospitalization times may be less than for internal fixation (Selcer et al., 1991).
Advantages
Significant instability cannot be addressed, reduction is difficult, some animals suffer prolonged discomfort, and the neurological deficits worsen in some patients or even become irreversible. The main problem with medical management of thoracic or lumbar fractures is the longer recovery period compared to surgery and the much longer time required to reach optimal neurological status (Selcer et al., 1991). Cage rest alone is only recommended for the most stable fractures in the thoracolumbar region and certainly not when there is severe compression or when two or three components have failed (13.16, 13.32). Non-operative management in humans is not really comparable as it involves reduction by traction and a degree of immobilization that is not possible in most animals (Fehlings and Tator, 1999; Tator et al., 1999). An external splint is the nearest equivalent in animals and this may also help to reduce some displaced fractures (Patterson and Smith, 1992).
Disadvantages
EXTERNAL SPLINT (13.37–13.47) An external splint has been shown to prevent dorsoventral angulation of thoracolumbar spines subjected to large bending forces. The forces used were in excess of those causing failure of five internal fixation techniques in a spinal fracture model and were in the range of those experienced by a large paralyzed dog undergoing routine nursing care. External splinting works best with
injuries that preserve the ventral buttress; it can sometimes work with severe compromise of the ventral buttress although it is not the ideal technique for such injuries (Patterson and Smith, 1992). The advantages are similar to conservative management but this technique provides much greater stability (13.46). It can be used as the sole means of fixation or as a supplement to internal fixation. It is most applicable to mid-thoracic or lumbar fractures but can also be used for cervical or lumbosacral injuries.
Advantages
Disadvantages The disadvantages are: the nursing care needed; variable individual tolerance for splints; the possibility of neurological deterioration (13.24); and the development of complications such as urine scald or decubitus (13.36A, 13.47). When employed as an adjunct to internal fixation it is difficult to observe the surgical wound unless the splint is attached by Velcro straps (13.41). Good candidates for external fixation include those with normal nociception, an intact ventral buttress (13.9, 13.11), and no pelvic, thoracic or major soft tissue injuries (Patterson and Smith, 1992). Marginal candidates are those with a compromised ventral buttress (13.13, 13.15) and those with pelvic, thoracic or soft tissue injuries (Patterson and Smith, 1992).
EXTERNAL FIXATOR (13.64–13.66) External constructs have been compared to both intact spines and to Steinmann pin and bone cement implants. Type II external constructs and internal fixation using eight Steinmann pins with bone cement were stronger
Trauma
than type I external constructs and internal fixation using four pins with cement. All implants tested were as strong as an intact spine in extension and rotation and were stronger in flexion; the type II external construct using a parabolic arch was the strongest single device. Although there was no overall difference between the arch construct and internal fixation using eight pins with cement, the arch was the strongest in flexion. Flexion is considered the main mode of failure for most implants (Lanz et al., 2000; Walker et al., 2002). The advantages are that these fixators can be made to be very strong (Walker et al., 2002); they work well if the ventral buttress is compromised (13.65); dissection can be minimized and pins can even be placed fluoroscopically (Wheeler et al., 2002); and the implant can be removed easily after healing (Ullman and Boudrieau, 1993; Lanz et al., 2000) (13.65). It can also be a very useful rescue technique if other methods fail (13.65).
The advantages are that it is economical, most clinics have the necessary orthopedic implants and it can work well when the vertebral buttress is intact.
Advantages
The disadvantages are that it immobilizes a long segment of the vertebral column with the weak link being the orthopedic wire, not the Steinmann pin. In addition, the wire can cause fracture of the articular facets. It is not recommended for stabilizing a compromised vertebral buttress (13.13, 13.15).
Disadvantages
Advantages
Disadvantages include the potential to introduce infection and place excessive tension on the skin or fascia (Shores et al., 1989; Lanz et al., 2000); the pin tracts need regular cleaning; additional postoperative management can be labor intensive (Lanz et al., 2000); limited pin lengths for large dogs using some systems (14.13); and the risk of traumatic removal (13.65). Removal of the device prior to optimal fracture healing could result in catastrophic failure.
Disadvantages
INTERNAL FIXATION The procedures described are usually performed via a standard dorsal approach to the thoracolumbar spine (12.30–12.33). The dorsal spinal plate, vertebral body plate, transilial pin techniques (Lewis et al., 1989), and spinal stapling are no longer recommended. Bilateral plating has similar biomechanical properties to pins and bone cement but is harder to perform over the thoracic and caudal lumbar regions (Viguier et al., 2002). AO locking plates may be useful in some situations (11.13B); these can now be made from bioabsorbable materials that avoid stress protection by transferring load gradually to the spine (Yerby et al., 1998; Vaccaro and Madigan, 2002; Vaccaro et al., 2002) (13.33).
MODIFIED SEGMENTAL FIXATION (13.61–13.63) Although the original spinal stapling technique is inadequate and no longer recommended for all but the smallest dogs, it is improved by incorporation of the articular facets into the fixation and by use of several pins in parallel (McAnulty et al., 1986).
METAL AND BONE CEMENT This technique was originally described using Steinmann pins placed in the vertebral bodies (Rouse and Miller, 1975; Blass and Seim, 1984). Threaded metal implants and bone cement are preferred for most types of spinal fracture/luxation and have been used in animals of all sizes with very good results, including those with significant compromise of the ventral buttress (13.32). A study using negative threaded, 4-mm Steinmann pins showed that four-pin and eight-pin patterns were at least as stiff as intact spines. Eight-pin constructs were significantly stronger than four-pin constructs (Walker et al., 2002). The optimum configuration of pins or screws has yet to be determined by biomechanical testing and a variety of configurations work well in a clinical setting (13.25, 13.31, 13.48). Failures in clinical cases are unusual despite mixed results in biomechanical testing (Garcia et al., 1994; Sharp et al., 1998; Bagley et al., 2000; Magnier and Lavaste, 2002; Walker et al., 2002; LeCouteur and Sturgess, 2003). Implant failure occurred in only one of 19 dogs in one series (Sharp et al., 1998) (cement fracture shown in 13.28A).
Advantages The advantages are the immobilization of only a short segment of the vertebral column and its utility in thoracic and caudal lumbar areas because the ribs and spinal nerves can be avoided easily. The implants provide excellent strength and stability against rotation, flexion and extension (Waldron et al., 1991; Walker et al., 2002). The disadvantages include the low resistance of some configurations to dorsoventral bending (Walter et al., 1986; Walker et al., 2002); migration of smooth pins (13.29), difficulty in closing soft tissues over cement if placed in a doughnut pattern, difficulty of removal; potential for implant failure (13.27, 13.28) or stress protection (13.33), and the risk of infection (10.45, 13.28, 13.34). Failure of Steinmann pins and bone cement occurs mainly because of pin pullout from the bone (Walter et al., 1986; Waldron et al., 1991). The cement bar must
Disadvantages
291
292
Small Animal Spinal Disorders
be thick enough (13.59) and the implants must be strong enough for each patient (Willer et al., 1991) (13.27). Although one study showed poor results with screw and cement implants, it compared 3.2-mm Steinmann pins to 3.5-mm screws but these screws have only a 2.4-mm core diameter (Garcia et al., 1994). The strength of a metal implant is proportional to its radius to the power of four, so small increases in diameter dramatically increase implant strength (Beaver et al., 1996). When pins of the same diameter are used to produce external fixation and internal fixation of the lumbar spine in a canine model the techniques have similar mechanical properties (Walker et al., 2002). The actual strength required in any given animal is not known but will obviously depend on the fracture, and the weight and temperament of the animal. In general, the largest possible implant should be used. Use of threaded in place of smooth implants should increase the strength of the fixator because: • Threaded implants have greater resistance to pullout than equivalent-sized smooth pins. • Threaded implants are less likely to migrate than smooth pins. • Threaded implants do not need to be notched to provide anchorage for the cement. Additional advantages of screws over threaded pins are that: • Screws do not require cutting, which could cause rocking and loosening of a pin. • Screws are interchangeable, so the optimum implant length can be selected. • Screw heads provide additional purchase for the cement. In summary: • Positive profile threaded pins or screws should be used instead of Steinmann pins. • The implant pattern shown in 13.54A (see also 13.20, 13.25, 13.65) is preferred over the pattern shown in 13.49A. This is because three implants per fragment are 66% stronger than two per fragment (Garcia et al., 1994). Implants can be placed in two vertebrae on each side of the lesion if necessary (Bagley, 2000; Bagley et al., 2000), but this immobilizes four or five vertebrae compared to just the two or three shown in 13.20 and 13.48. • The implant should penetrate two cortices for greater holding power (Zindrick et al., 1986). The risk of damage to the aorta or vena cava is low but this must still be borne in mind when drilling and when tapping the thread (see page 9, 1.23). The risks are lowered by using a relatively flat trajectory (13.52, 13.54B).
•
The implant diameter should be as large as possible. 4.5-mm cortical screws should be used for dogs weighing over 30 kg (Beaver et al., 1996) (13.27).
BONE GRAFTS These have a wide range of potential uses, which include to enhance the healing of fractures and luxations. Grafts are either cortical or cancellous in origin; they can be derived from the host (autograft) or from a donor (allograft) (Millis and Martinez, 2003). Autogenous cancellous graft is the most effective material to improve bone healing. It does not provide useful mechanical strength but up to 10% of transplanted cells survive to produce bone (osteogenesis). It also induces adjacent cells to make new bone (osteoinduction) and it forms a weak scaffold for bone ingrowth (osteoconduction). Harvesting autogenous cancellous bone requires a separate surgical approach; the proximal humerus is the most accessible site during ventral surgical approaches (11.24) and the iliac crest during dorsal approaches. • Cortical autografts can be harvested from the ilium, rib, fibula or ulna. • Caudal vertebra can be used as a free vascularized graft (Yeh and Hou, 1994). • Corticocancellous autograft can be obtained from the ilium, sternum, wing of C6 vertebra, or from adjacent spinous processes (Chauvet et al., 1999). • Cancellous allograft (Veterinary Transplant Services, Seattle, WA) can be used like fresh cancellous graft; it can also be combined with a small amount of autogenous graft. Cancellous allografts contain no live cells; they are not osteogenic but are both osteoconductive and osteoinductive. Small blocks of allograft can also be used to temporarily distract a ventral slot during fixation. • Cortical allografts (Veterinary Transplant Services, Seattle, WA) are used to provide some mechanical support and a strong scaffold for new host bone ingrowth. They may be packed with cancellous bone to enhance healing further. • Demineralized bone matrix can also be used to enhance bone healing. It is available as a powder (Veterinary Transplant Services, Seattle, WA) made from cortical bone after acid-extraction of the mineral to expose osteoinductive proteins. These proteins then cause adjacent mesenchymal cells to differentiate into osteogenic cells. Potential indications for bone grafts in neurosurgery are to improve healing potential such as in a diabetic or geriatric animal; for arthrodesis (9.28, 10.52D, 11.30A), to replace bone loss such as after vertebrectomy (12.42) and to manage infection (13.28, 14.12).
Trauma
TREATMENT Choice of therapy
•
Non-surgical management is indicated for most cervical injuries and for some lumbosacral injuries (see ‘Anatomical location of the injury’, page 295). Surgical management is best for most thoracic and lumbar injuries. Severe shock and hypotension can exacerbate the neurological deficit and so it may be worth reassessing a patient that shows a lack of deep pain sensation after a 24-h period of circulatory support. Definitive treatment in such animals should only be undertaken once the owner has been made fully aware of the prognosis and the likely time course for recovery (see ‘Prognosis’, page 301).
PATIENTS WITH NO DEEP PAIN SENSATION The clinician has several options for animals with no deep pain due to thoracic or lumbar injuries (Algorithm 13.1): • Perform euthanasia because of the poor prognosis (Olby et al., 2003).
Presentation
•
•
Reassess deep pain after 24 h of circulatory support combined with an external splint. Image by myelogram or MRI in case these provide clear evidence of spinal cord transection, which carries a hopeless prognosis for recovery (13.4). Severe compression or parenchymal hemorrhage on MRI are also likely to be poor prognostic signs in animals much as they are in humans (Ramon et al., 1997; Fehlings et al., 1999). Perform a durotomy as this might identify that the spinal cord is not in continuity (13.4). However, a severely compromised spinal cord that nevertheless remains physically intact may be even less likely to survive when subjected to additional manipulation. If there is any tissue in continuity, the animal should be given the benefit of the doubt. Only about 5% of axons need to survive across the injury site in order for a dog or cat to regain the ability to walk (Blight and Decrescito, 1986; Basso et al., 1996; Jeffery and Blakemore, 1999; Olby et al., 2003). The presence of severe, localized malacia is
Euthanasia
Reassess after 24 hours support with a splint (+/– MPSS)
Deep pain present
No deep pain
See Algorithm 13.2
Examine for spinal cord transection by durotomy
Euthanasia
Imaging Cord intact
Cord transected
Euthanasia
No compression
Severe compression
Stabilize Decompress
External fixation
Internal fixation
Cord transected
Euthanasia
Algorithm 13.1 Surgical decision-making when deep pain is absent.
293
294
Small Animal Spinal Disorders
to be expected after any serious injury and on its own does not warrant euthanasia (Salisbury and Cook, 1988) (8.50 and see page 132). If imaging or direct inspection rules out spinal cord transection, the surgeon can: • Stabilize the spine using internal or external fixation and hope that the dog recovers or can live in a cart. • Apply an external splint and reassess over 7 days. If deep pain sensation has not returned by then, and certainly if it does not return within 1 month, then it probably never will. The only hope is that the animal will have a delayed recovery of motor function but without recovery of continence or deep pain (Olby et al., 2003) (see pages 87, 131, 302). Either approach is acceptable in this situation provided that the owner is aware the chances for a full recovery are probably less than 5%.
PATIENTS WITH INTACT DEEP PAIN SENSATION The decision between surgical and non-surgical management for thoracic or lumbar injuries should be made
on an individual basis. A suggested approach is shown in Algorithm13.2. If the animal satisfies criteria for a splint (see page 290) then non-surgical therapy can be considered. With surgery, the risks of implant failure or infection, overall hospitalization time and cost are factors to be considered. Surgical patients may stay in hospital longer than medically treated patients and can therefore incur more than twice the expense. Final neurological status may not differ between animals treated medically and those treated by surgery (Selcer et al., 1991). However, in general: • Surgery is the preferred treatment for thoracic or lumbar injuries due to the low overall risk of complications and the difficulties in instituting non-surgical therapy in many animals. Reduction of misalignment, additional decompression if necessary and rigid fixation are the goals of surgery. • For most animals a simple re-alignment provides sufficient decompression (Bagley, 2000; Bagley et al., 2000). Immediate decompression has the most relevance in patients with severe deficits (Tator et al., 1999; Fehlings et al., 2001; Papadopoulos et al., 2002).
Algorithm 13.2 Surgical decision-making when deep pain is present.
Consider anatomical location
Surgical
Non-surgical
Advanced imaging
External fixation (Splint)
Compression
No compression
Decompress by realignment
Improves
Deteriorates
Continue
Surgical
Bone, disc hematoma
Decompress by (mini) hemilaminectomy
Fixation
Internal fixation
External fixator
Trauma
•
Adding a laminectomy will increase surgical time and decrease stability further. It is only indicated when there is a large extradural mass of bone fragment, disc or blood clot (Lanz et al., 2000).
•
• •
Hemilaminectomy should be used rather than dorsal laminectomy (Smith and Walter, 1988). Where possible, less invasive surgeries that preserve the facet are preferred over standard hemilaminectomy because facet removal is very destabilizing, particularly when the disc is also damaged (Shires et al., 1991; Schulz et al., 1996) (see page 126). The vertebral column must always be stabilized after dorsal laminectomy. A final factor that may modify the decision regarding surgery is the anatomical location of the injury.
Anatomical location of the injury CERVICAL SPINE
A
B 13.18 A: Dog with a fracture/luxation of C2 treated with an external splint. The splint should ideally extend from near the lateral canthus of the eyes to the mid-thoracic region. B: An alternative means of stabilizing the upper cervical area using a chest harness and Halti collar; the Halti with harness can also be used in cervical spondylomyelopathy (CSM).
This region has the largest ratio of vertebral canal to spinal cord diameter. Cage rest, with external support in unstable or displaced fractures (13.18), is the treatment of choice for most cervical fractures unless the animal is deteriorating neurologically (Hawthorne et al., 1999). Mortality rates may be as high as 35–40% with surgery (Stone et al., 1979; Hawthorne et al., 1999). Severe intraoperative hemorrhage can also occur with C2 fractures and reduction can be challenging (Boudrieau, 1997; Schulz et al., 1997) (13.64). This is not the easiest location to apply a splint but one can be made from various materials (13.18). Surgery may reduce recovery times but is best reserved for animals that: • Are tetraplegic or have poor ventilatory function. • Show neurological deterioration despite adequate confinement or external fixation. • Remain very painful beyond an initial 48–72 h. The most useful surgical technique for cervical injuries is ventral placement of pins or screws and bone cement (Rouse, 1979; Schulz et al., 1997) (11.22, 13.19 and see page 221). The main disadvantage of this technique is that it may fail if used to span more than one intervertebral space, which can be a problem if a vertebral 13.19 This dog was unable to walk after running into a tree. The C5 fracture was stabilized with bone screws and cement (11.22, 13.56A); the dog could walk well within 2 weeks. Note that the screws in C7 are placed too far caudally.
A
B
295
296
Small Animal Spinal Disorders
13.20 Dog shown in 13.10 after stabilization with six 3.5-mm screws and bone cement. This dog suffered an unusual complication (shown in 13.30).
A
B
13.21 Postoperative radiographs from the dog shown in 13.14. A: L7 fracture stabilized with screws and bone cement. B: Two cancellous screws have been inserted into the wing of each ilium. Better purchase would have been obtained by angling these screws vertically as shown in 13.65 and 14.12. The dog was clinically normal 6 weeks after surgery.
A
B
body is shattered (see 11.17C). In this case at least three implants should be on either side of the fracture and Steinmann pins used to reinforce the cement (13.25). If dorsal stability is required, such as after facet luxation, screws may be placed across the articular facets (Swaim, 1975; Basinger et al., 1986).
THORACIC AND LUMBAR SPINE These are the most common regions of the vertebral column to be injured in animals, and are also at most risk for devastating neurological consequences because of the relatively small diameter of the vertebral canal compared to other areas. Metal implants and bone cement (13.20) or an external fixator (13.65) are preferred unless the dog is a good candidate for an external splint (Patterson and Smith, 1992) (see page 305).
L6 AND L7 VERTEBRAE AND LUMBOSACRAL JUNCTION The vertebral canal is relatively spacious in the caudal lumbar area because the spinal cord ends around the 6th lumbar vertebra, and only nerve roots occupy the caudal part of the vertebral canal (1.5, 1.9 and see page 1). For these reasons, injuries causing severe
displacement may result in only mild neurological deficits (13.16, 13.26). Fractures in this region can heal satisfactorily without surgical intervention (Smith and Walter, 1985; Patterson and Smith, 1992). Candidates for internal fixation include animals with persistent pain, marked deficits or deteriorating neurological status. Screws or threaded pins and bone cement are the preferred technique, with implants in the wing of the ilium if necessary (13.21, 13.26). An external fixator also works well in this area (13.65, 13.66, 14.13). A splint or modified segmental fixation can be effective but ideally the ventral buttress should then be intact (McAnulty et al., 1986; Patterson and Smith, 1992) (13.63).
SACRAL, SACROCAUDAL AND TAIL INJURIES Fractures of the sacrum have a high incidence of neurological deficits if they traverse the vertebral canal or involve the sacral foramina (Kuntz et al., 1995; Anderson and Coughlan, 1997; Kuntz and Bonagura, 2000) (13.22). Surgery may improve outcome although the risk of iatrogenic injury is high (Kuntz et al., 1995). Therefore surgery should be restricted to animals with severe or progressive deficits. Realignment of the
Trauma
A
B
13.22 A: Cat with a sacrocaudal injury and transverse sacral fracture (arrow). Six weeks post trauma the cat still had a paralysed tail, no anal reflex and was incontinent. Dorsal laminectomy revealed fibrosis; neurolysis produced no improvement and the cat was euthanized at 18 weeks. Histopathology revealed lesions in the S2, S1 and L7 nerve roots caused by traction injury to the cauda equina (Smeak and Olmstead, 1985; Kuntz and Bonagura, 2000). B: Necropsy specimen from a dog with a transverse sacral fracture that has severed all nerve roots at this level (arrowheads).
L7–S1 articular facets facilitates reduction of a sacral fracture (Pare et al., 2001). Sacrocaudal injuries occur most commonly in cats (Feeney and Oliver, 1980). The role of tethering in this type of injury is unclear (Taylor, 1981; Smeak and Olmstead, 1985; LeCouteur and Sturgess, 2003). Cats that do not recover continence by 4 weeks post-injury have a poor prognosis (Smeak and Olmstead, 1985). Fractures of the caudal vertebra are usually treated non-surgically or by tail amputation.
POSTOPERATIVE CARE
(see also
Chapter 15) The surgical wound must be examined at least daily for any sign of infection. A urinary tract infection (UTI) should be anticipated in any animal whose deficits are severe enough that it cannot walk. Urine obtained by cystocentesis should be analysed every 2 days or if the urine appears cloudy; culture should be done if there is any sign of infection. Bacteremia from the urinary and gastrointestinal tracts or from skin lesions increases the risk of infection at the surgical site (see page 355). Any sudden increase in pain (13.27) or deterioration in neurological status (13.30) suggests infection or implant failure and requires repeat radiographs. A deterioration in patient status can also occur from non-neurological complications of the original trauma (Box 13.2).
COMPLICATIONS These are discussed for three major time points— intraoperative, early and late postoperative (Table 13.2).
Table 13.2 Intraoperative complications
Intraoperative Poor reduction (13.24–13.26) Poor implant selection (13.27) Poor implant placement (13.23) Inadequate cement (13.27, 13.28A) Pneumothorax Fat embolism Cardiac arrythmias
Early postoperative
Late postoperative
Complications of trauma Implant failure (13.28, 13.29) Wound infection (13.28) Gastrointestinal ulceration Urinary tract infection Decubitus Urine scald Pneumonia Vascular complications Sepsis
Implant failure Infection (13.34) Callus encroachment Syringomyelia Arachnoid cyst (13.32) Late deformity Adjacent segment disease (13.28)
Intraoperative complications These fall into two main groups: cardiopulmonary complications and technical errors (Table 13.2). Cardiovascular complications are most likely in animals with cervical and thoracic injuries (Hawthorne et al., 1999) (see page 86). Pneumothorax is common and may be iatrogenic (Swaim, 1971; Blass and Seim, 1984; Selcer et al., 1991). Fat embolism may occur during fracture repair (Schwarz et al., 2001). Technical errors in fixation include improper placement of implants (Blass and Seim, 1984) (13.19, 13.23), poor postoperative alignment (13.24–13.26) that can even lead to a delayed onset
297
298
Small Animal Spinal Disorders
A
13.23 One Steinmann pin has been inserted too far and could damage vital structures in the retroperitoneal space or abdomen (see 1.23 for relationship of vertebrae to the aorta and vena cava). This dog was normal at 5-year follow-up.
B
A
13.25 A: Mild deformity in extension, and B: significant lateral misalignment. Use of K wires through the articular facets helps to prevent this (Berry et al., 1999). The cement bars have been reinforced with Steinmann pins (13.59). Six 4.5-mm screws have been used in this 30-kg dog. Despite the poor alignment the dog recovered from paraparesis to walking within 48 h and was normal at 6 weeks. See also 13.51.
B 13.24 This dog was progressively paraparetic 7 days after a lowimpact injury. An external splint was applied but the neurological status deteriorated. A: Preoperative radiograph shows an injury at L4/5 (same dog as 13.5). B: Although facet joint alignment was good, the surgeon exerted excessive ventral pressure causing a misalignment in extension (Matthieson, 1983). Use of a towel or sandbag under the abdomen helps to prevent this as does gentle traction on the spinous processes as the cement hardens (Blass and Seim, 1984; Lewis et al., 1989; Berry et al., 1999). Fortunately the dog recovered quickly and was normal within 2 months.
syringomyelia in humans (Perrouin-Verbe et al., 1998; Fischbein et al., 1999; Bains et al., 2001), poor choice of implant (13.27), and too little cement or an uneven cement bar (13.27, 13.28A).
13.26 This 11 kg dog was paraparetic with no anal tone 24 h after trauma. Reduction using a laminectomy spreader, curved hemostat or Senn retractor could have overcome the misalignment shown here (Harrington and Bagley, 1998) (13.64). Correct alignment is not quite so crucial in the caudal lumbar or cervical regions because there is more space at these sites. Nerve roots of the cauda equina also tolerate deformity better than does spinal cord. The dog was normal within 8 weeks and remained so 2 years later.
Trauma
13.27 Implant failure 4 days after surgery; the cement did not cover the cranial screws adequately (13.57). The 3.5-mm screws were also not strong enough for this 44 kg dog (Beaver et al., 1996). They were chosen in order to fit down the pedicle into the base of the articular facet of L7. This is a useful way to add an additional implant but should not be used as the sole fixation because the pedicle limits the implant diameter. The rescue procedure is shown in 13.64–13.66.
A
B
A
B
13.28 Implant failure 5 days after surgery. A: The cement has broken; either the bar was not thick enough (13.59) or a bubble (13.33), fold or thin area was present at this site (McAfee et al., 1986) (13.27). Wound culture yielded Streptococcus fecalis. B: Wound discharge resolved 2 weeks after the implant was replaced with screws and cement plus modified segmental fixation (13.61–13.63). The dog did well for 2 years when pain and discharge returned. The implant was removed and discharge resolved again. The dog did well for 3 more years but then developed a marked kyphotic deformity.
Late postoperative complications Early postoperative complications A high index of suspicion must be maintained during this period for complications relating to other organ systems such as leakage from the urinary tract or bile duct (Matthieson, 1983). Some complications, especially vascular disorders like pulmonary thromboembolism and deep vein thrombosis, will benefit from early mobilization and physical therapy (Selcer et al., 1991). If the animal becomes very painful or deteriorates neurologically then implant failure or infection must be considered (13.28–13.30; see also Table 13.2). External fixation (13.64–13.66) is an excellent rescue technique in such situations. It would have been a much better rescue technique for the dog shown in 13.28B; bone grafting may also help to improve healing at infected sites (Auger et al., 2000).
The most likely problems at this stage again relate to either the implant or to delayed infection (Table 13.2). Of 19 dogs managed by metal and bone cement, five had infections (13.34). All were treated successfully although one suffered a late recrudescence (13.28). Three of the five did not undergo implant removal (Sharp et al., 1998). Infection can also be a problem underneath external splints and may even be fatal if detected too late (McAfee et al., 1986) (13.36A). Callus has been reported to encroach into the vertebral canal but this is a very unusual problem (Carberry et al., 1989). Internal fixators do not generally require removal; late removal could lead to kyphotic deformity or even pathological fracture (13.28, 13.33). Other potential problems include adjacent segment disease (13.31), late-onset syringomyelia (PerrouinVerbe et al., 1998), or arachnoid cyst formation (13.32).
299
300
Small Animal Spinal Disorders
13.29 This dog died 1 week after surgery of unknown causes. The pin was loose in the abdomen but did not appear to be the cause of death. Implant migration is most likely with smooth pins (Blass and Seim, 1984; Blass et al., 1988; Wong and Emms, 1992). Every effort should be made to incorporate all implants into the cement, including any K wires (compare 13.48 and 13.63).
13.31 Three-year follow-up of the dog shown in 13.32. Although domino lesions have not been reported after fracture repair there is evidence that additional strain is imposed on segments adjacent to the fixation (arrows) (Fox et al., 1996; Wheeler et al., 2002). This could help to explain the failure shown in 13.30 (Goffin et al., 1995).
A
A
B
13.30 This dog lost deep pain sensation 8 days after surgery at T12/13 (13.20). This unusual failure may be a type of adjacent segment disease (MacMillan and Stauffer, 1991; Fox et al., 1996; Lanz et al., 2000; Wheeler et al., 2002) (11.23, 13.31). A: Radiography revealed luxation at the adjacent T13/L1 interspace. The implant was intact. B: Marked rotational deformity was also present. The preoperative myelogram and CT scan were re-examined but no injury was found at T13/L1. 3D CT scan reconstructions were also examined to view the articular facets but no abnormality was detected (13.10B). The dog recovered motor function at 3 months; it could walk by 6 months but did not recover continence or deep pain (Olby et al., 2003) (see pages 32, 131, 302).
B 13.32 A: Preoperative myelogram of the dog shown in 13.31 and 13.33. The T12 vertebral buttress has failed and the facet joints have luxated. Despite the severity of the injury the dog made an excellent recovery and did very well for 3 years when it developed progressive paraparesis. B: A myelogram then revealed an arachnoid cyst (arrowhead) at T12/13 (see pages 321, 323). The dog remained stable for a further 21 months without intervention (Skeen et al., 2003).
Trauma
13.35 Example of the devastating type of spinal injury commonly encountered in cats (13.6, 13.12). 13.33 A CT scan through the arachnoid cyst shown in 13.32 confirmed a focal dilation of the subarachnoid space (arrowhead). Note also the large bubble in the cement (arrow). This can weaken the implant and could predispose it to failure (McAfee et al., 1986; Anderson, 1988; Beaver et al., 1996) (13.28). There is also significant loss of cortical bone in the vertebral body when compared to the ribs. This could be due to stress protection, which could complicate late implant removal (Craven et al., 1994; Vaccaro and Madigan, 2002) (13.28).
FELINE SPINAL INJURIES Although non-surgical therapy is useful it usually is limited to cage confinement as cats tolerate splints poorly (Carberry et al., 1989). In theory, any internal fixation technique can be applied to cats although feline spinal injuries are often severe (13.35).
PROGNOSIS
13.34 Infection is a potential problem when using bone cement. A: Reduction of the luxation was poor. B: The dog developed pain after 6 weeks due to discospondylitis at T13/L1. The entire implant was removed and the pain resolved after 2 months of antibiotics (Blass and Seim, 1984; Wong and Emms, 1992) (13.28). At 2-year follow-up the dog was walking but had no deep pain sensation (Olby et al., 2003).
This varies somewhat with the location of injury. Prognosis is very good after cervical injury if the animal does not die acutely from respiratory arrest (see pages 28, 82). In the largest series to date, 32 of 39 dogs (82%) with neck injuries made a functional recovery; results were much better after non-surgical treatment because the mortality rate was only one third of that in the surgically treated dogs. Twenty-five of 28 dogs recovered after non-surgical management compared with seven of 11 after surgery (89% vs 64%). Results were good after non-surgical management even for dogs that were unable to walk. Delay in instituting definitive treatment was a predictor of poor outcome (Hawthorne et al., 1999). Immediate decompression by vertebral realignment is likely to benefit animals with tetraplegia or hypoventilation secondary to cervical injury (Papadopoulos et al., 2002). The prognosis following thoracic and lumbar injury is similar to that for disc disease provided that the animal has good deep pain at presentation. However, studies with good follow-up are limited. Surgical management of 24 such animals gave good results in 21 (88%), although 10 had residual deficits and one was euthanized due to implant migration after 22 months (Blass and Seim, 1984; McKee, 1990). In another study, 19 of 23 dogs (83%) presenting with intact deep pain made good recoveries (Sharp et al., 1998). Non-surgical management in a further 25 animals, comprising either cage confinement or splinting, gave good results in 24 (96%). Three had residual deficits (Carberry et al.,
301
302
Small Animal Spinal Disorders
1989; McKee, 1990). Non-surgical management using a splint in 16 dogs gave good results in all animals although six had residual deficits. Six dogs also developed serious complications and three deteriorated but these problems were temporary and resolved in all cases (Patterson and Smith, 1992). Animals presenting with thoracic or lumbar injuries and no deep pain rarely recover after trauma. Of nine such dogs, none regained deep pain over follow-up periods of 6 weeks to 2 years (Olby et al., 2003). Two did recover motor function without recovering deep pain (13.30, 13.34). This unusual phenomenon, which can also occur after thoracolumbar disc disease (see pages 32, 131), presumably is mediated by a peripheral rim of surviving axons at the edge of an extensive central lesion (Griffiths, 1978; Olby et al., 2003) (6.2). Recovery of motor function without deep pain usually takes 3–4 months; it rarely produces a normal gait and animals are usually incontinent with recurrent urinary tract infections. Based on extrapolation of data from disc disease, a full recovery is unlikely unless deep pain returns within 4 weeks of injury (Olby et al., 2003).
Although MPSS may have some benefit in severe injuries, its contribution is likely to be marginal (see page 83). Adequate reduction and fixation by internal or external means is far more important. The prognosis after sacral fracture is good with 26 of 32 dogs (81%) recovering (Kuntz et al., 1995). Cats with sacrocaudal injuries usually recover urinary continence if they retain good anal tone and perineal sensation on initial examination. Cats that do not become continent within 1 month generally fail to regain urinary function (Moise and Flanders, 1983; Smeak and Olmstead, 1985). Animals that remain incontinent may be candidates for permanent cytostomy tubes (see 15.32). The prognosis following brachial plexus injury is guarded to poor. The best prognostic indicator is the presence or absence of deep pain sensation in the distal limb. Five of seven dogs with normal sensation made complete recoveries whereas only six of 21 with reduced sensation recovered or remained in stable condition. Abnormal motor nerve conduction is also associated with a negative outcome (Faissler et al., 2002).
Key issues for future investigation 1. What is the best way to manage cranial thoracic lesions (13.36A)? 2. Is there any advantage to using more severe stimuli to assess deep pain (13.36B)? 3. Will MRI prove to be prognostic in animals as it may be in humans by detecting spinal cord hemorrhage (Ramon et al., 1997; Fehlings et al., 1999; Selden et al., 1999)? 4. What can electrophysiology contribute to determining prognosis?
A
B
13.36 A: This dog showed progressive paraparesis after trauma. It had subluxation at T2/3, which was managed with an external splint. This is a difficult area to stabilize so the sides of the splint were extended down each flank to just above the elbow but this made it impossible to check beneath the splint without removing it. The dog would not eat 5 days after splinting and was found in extremis on day 6. A huge, phlegmon-like, subcutaneous abscess was present associated with a small decubital lesion. Bone cement and threaded implants would have been preferable in this dog. B: The prognosis for dogs with traumatic injuries and without nociception on presentation is very poor (Olby et al., 2003). It would be helpful to be able to differentiate animals without deep pain that are capable of functional recovery from those with irreversible injuries. Here an electrical stimulator from an electromyography (EMG) machine is being used but there was no response in this dog. Alternative methods include strong pliers or a cattle prod (see page 132).
Trauma
REFERENCES AANS (2000) Guideline to Improve the Outcome of Brain Trauma Patients: Management and Prognosis of Severe Traumatic Brain Injury, 2nd edn. Brain Trauma Foundation: American Association of Neurological Surgeons. http://www.braintrauma.org/index.nsf/Pages/Guidelines-main. Anderson, A., Coughlan, A.R. (1997) Sacral fractures in dogs and cats: a classification scheme and review of 51 cases. Journal of Small Animal Practice 38, 404–409. Anderson, G.I. (1988) Polymethylmethacrylate: a review of the implications and complications of its use in orthopaedic surgery. Veterinary and Comparative Orthopaedics and Traumatology 2, 74–79. Auger, J., Dupuis, J., Quesnel, A., Beauregard, G. (2000) Surgical treatment of lumbosacral instability caused by discospondylitis in four dogs. Veterinary Surgery 29, 70–80. Bagley, R.S. (2000) Spinal fracture or luxation. In: W.B. Thomas (ed.), Veterinary Clinics of North America Small Animal Practice, 133–154. Philadelphia: WB Saunders. Bagley, R.S., Silver, G.M., Connors, R.L., Harrington, M.L., Cambridge, A.J., Moore, M.P. (2000) Exogenous spinal trauma: surgical therapy and aftercare. Compendium on Continuing Education for the Practicing Veterinarian 22, 218–230. Bains, R.S., Althausen, P.L., Gitlin, G.N., Gupta, M.C., Benson, D.R. (2001) The role of acute decompression and restoration of spinal alignment in the prevention of post-traumatic syringomyelia: case report and review of recent literature. Spine 26, E399–402. Basinger, R.R., Bjorling, D.E., Chambers, J.N. (1986) Cervical spinal luxation in two dogs with entrapment of the cranial articular process of C6 over the caudal articular process of C5. Journal of the American Veterinary Medical Association 188, 865–867. Basso, D.M., Beattie, M.S., Bresnahan, J.C. (1996) Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight-drop device versus transection. Experimental Neurology 139, 244–256. Beal, M.W., Paglia, D.T., Griffin, G.M., Hughes, D., King, L.G. (2001) Ventilatory failure, ventilator management, and outcome in dogs with cervical spinal disorders: 14 cases (1991–1999). Journal of the American Veterinary Medical Association 218, 1598–1602. Beaver, D.P., MacPherson, G.C., Muir, P., Johnson, K.A. (1996) Methylmethacrylate and bone screw repair of seventh lumbar vertebral fracture-luxations in dogs. Journal of Small Animal Practice 37, 381–386. Berry, W.L., LeCouteur, R.A., Vernau, K.M., Sturges, B.K. (1999) Trans-articular process pinning for the initial stabilization of vertebral fracture/luxations in 18 dogs. Journal of Veterinary Internal Medicine 13, 240. Blass, C.E., Seim, H.B. (1984) Spinal fixation in dogs using Steinmann pins and methylmethacrylate. Veterinary Surgery 13, 203–210. Blass, C.E., Waldron, D.R., Van Ee, R.T. (1988) Cervical stabilization in three dogs using Steinmann pins and methylmethacrylate. Journal of the American Animal Hospital Association 24, 61–68. Blight, A.R., Decrescito, V. (1986) Morphometric analysis of experimental spinal cord injury in the cat: the relation of injury intensity to survival of myelinated axons. Neuroscience 19, 321–341. Boker, D., Schultheiss, R., Probst, E. (1989) Radiologic long-term results after cervical vertebral interbody fusion with polymethyl methacrylate (PMMA). Neurosurgical Review 12, 217–221. Boudrieau, R.J. (1997) Distraction-stabilization using the Scoville–Haverfield self-retaining laminectomy retractors for repair of 2nd cervical vertebral fractures in 3 dogs. Veterinary and Comparative Orthopaedics and Traumatology 10, 71. Campbell, V.L., King, L.G. (2000) Pulmonary function, ventilator management, and outcome of dogs with thoracic trauma and pulmonary contusions: 10 cases (1994–1998). Journal of the American Veterinary Medical Association 217, 1505–1509. Carberry, C.A., Flanders, J.A., Dietze, A.E., Gilmore, D.R., Trotter, E.J. (1989) Nonsurgical management of thoracic and lumbar spinal fractures and fracture luxations in the dog and cat: a review of 17 cases. Journal of the American Animal Hospital Association 25, 43–54. Chauvet, A., Hogge, G., Sandin, J., Lipsitz, D. (1999) Vertebrectomy, bone allograft fusion, and antitumor vaccination for the treatment of vertebral fibrosarcoma in a dog. Veterinary Surgery 28, 480–488.
Craven, T.G., Carson, W.L., Asher, M.A., Robinson, R.G. (1994) The effects of implant stiffness on the bypassed bone mineral density and facet fusion stiffness of the canine spine. Spine 19, 1664–1673. Denis, F. (1984) Spinal instability as defined by the three-column spine concept in acute spinal trauma. Clinical Orthopaedics and Related Research 189, 65–76. Egger, E.L., Histand, M.B., Blass, C.E., Powers, B.E. (1986) Effect of fixation pin insertion on the bone-pin interface. Veterinary Surgery 15, 246–252. Ethell, M.T., Bennett, R.A., Brown, M.P., Merritt, K., Davidson, J.S., Tran, T. (2000) In vitro elution of gentamicin, amikacin, and ceftiofur from polymethylmethacrylate and hydroxyapatite cement. Veterinary Surgery 29, 375–382. Faissler, D., Cizinauskas, S., Jaggy, A. (2002) Prognostic factors for functional recovery in dogs with suspected brachial plexus avulsion. Journal of Veterinary Internal Medicine 16, 370. Feeney, D.A., Oliver, J.E. (1980) Blunt spinal trauma in the dog and cat: neurologic, radiologic, and therapeutic correlations. Journal of the American Animal Hospital Association 16, 664–668. Fehlings, M., Tator, C. (1999) An evidence-based review of decompressive surgery in acute spinal cord injury: rationale, indications, and timing based on experimental and clinical studies. Journal of Neurosurgery 91, 1–11. Fehlings, M.G., Rao, S.C., Tator, C.H. et al. (1999) The optimal radiologic method for assessing spinal canal compromise and cord compression in patients with cervical spinal cord injury. Part II: Results of a multicenter study. Spine 24, 605–613. Fehlings, M.G., Sekhon, L.H., Tator, C. (2001) The role and timing of decompression in acute spinal cord injury: what do we know? What should we do? Spine 26, S101–110. Fischbein, N.J., Dillon, W.P., Cobbs, C., Weinstein, P.R. (1999) The ‘presyrinx’ state: a reversible myelopathic condition that may precede syringomyelia. American Journal of Neuroradiology 20, 7–20. Fox, M., Onofrio, B., Hanssen, A. (1996) Clinical outcomes and radiological instability following decompressive lumbar laminectomy for degenerative spinal stenosis: a comparison of patients undergoing concomitant arthrodesis versus decompression alone. Journal of Neurosurgery 85, 793–802. Garcia, J.N.P., Milthorpe, B.K., Russell, D., Johnson, K.A. (1994) Biomechanical study of canine spinal fracture fixation using pins or bone screws with polymethylmethacrylate. Veterinary Surgery 23, 322–329. Goffin, J., van Loon, J., Van Calenbergh, F., Plets, C. (1995) Long-term results after anterior cervical fusion and osteosynthetic stabilization for fractures and/or dislocations of the cervical spine. Journal of Spinal Disorders 8, 500–508; discussion 499. Gopal, M.S., Jeffery, N.D. (2001) Magnetic resonance imaging in the diagnosis and treatment of a canine spinal cord injury. Journal of Small Animal Practice 42, 29–31. Griffiths, I. (1970) A syndrome produced by dorso-lateral ‘explosions’ of the cervical intervertebral discs. Veterinary Record 87, 737–741. Griffiths, I.R. (1978) Spinal cord injuries: a pathological study of naturally occurring lesions in the dog and cat. Journal of Comparative Pathology 88, 303–315. Griffiths, I.R., Duncan, I.D., Lawson, D.D. (1974) Avulsion of the brachial plexus. 2. Clinical aspects. Journal of Small Animal Practice 15, 177–183. Harrington, M.L., Bagley, R.S. (1998) Realignment of a seventh lumbar vertebral fracture/luxation using a Senn retractor in two puppies. Journal of the American Animal Hospital Association 34, 377–380. Hawthorne, J.C., Blevins, W.E., Wallace, L.J., Glickman, N., Waters, D.J. (1999) Cervical vertebral fractures in 56 dogs: a retrospective study. Journal of the American Animal Hospital Association 35, 135–146. Hay, C.W., Muir, P. (2000) Tearing of the dura mater in three dogs. Veterinary Record 146, 279–282. Hurlbert, R.J. (2000) Methylprednisolone for acute spinal cord injury: an inappropriate standard of care. Journal of Neurosurgery 93, 1–7. Jeffery, N.D., Blakemore, W.F. (1999) Spinal cord injury in small animals 2. Current and future options for therapy. Veterinary Record 145, 183–190. Kriaras, I., Michalopoulos, A., Turina, M., Geroulanos, S. (2000) Evolution of antimicrobial prophylaxis in cardiovascular surgery. European Journal of Cardiothoraic Surgery 18, 440–446. Kuntz, C.A., Bonagura, J.D. (2000) Sacral fractures and sacrococcygeal injuries in dogs and cats. In: R.W. Kirk and J.D. Bonagura (eds), Kirk’s Current Veterinary Therapy XIII: Small Animal Practice, 1023–1026. Philadelphia: WB Saunders.
303
304
Small Animal Spinal Disorders
Kuntz, C.A., Waldron, D., Martin, R.A., Shires, P.K., Moon, M., Shell, L. (1995) Sacral fractures in dogs: a review of 32 cases. Journal of the American Animal Hospital Association 31, 142–150. Lanz, O.I., Jones, J.C., Bergman, R. (2000) Use of an external fixator to correct spinal fracture/luxation and instability in three dogs. Veterinary Neurology and Neurosurgery, http://www.neurovet.org/LanzSpinalFixator/ LANZ1final.htm#spinal%20instability,%20traumatic. LeCouteur, R.A., Sturgess, B. (2003) Spinal fractures and luxations. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1244–1260. Philadelphia: Elsevier Science. Lewis, D.D., Stampley, A., Bellah, J.R., Donner, G.S., Ellison, G.W. (1989) Repair of sixth lumbar vertebral fracture-luxations, using transilial pins and plastic spinous-process plates in six dogs. Journal of the American Veterinary Medical Association 194, 538–542. Lu, D., Lamb, C.R., Targett, M.P. (2002) Results of myelography in seven dogs with myelomalacia. Veterinary Radiology and Ultrasound 43, 326–330. MacMillan, M., Stauffer, E.S. (1991) Traumatic instability in the previously fused cervical spine. Journal of Spinal Disorders 4, 449–454. Magnier, J., Lavaste, F. (2002) In vitro biomechanical evaluation of unstable T13–L1 stabilization procedures in dogs. Veterinary Surgery 31, 288. Marcellin Little, D.J., Papich, M.G., Richardson, D.C., DeYoung, D.J. (1996) Pharmacokinetic model for cefazolin distribution during total hip arthroplasty in dogs. American Journal of Veterinary Research 57, 720–723. Matthieson, D.T. (1983) Thoracolumbar spinal fractures/luxations: surgical management. Compendium on Continuing Education for the Practicing Veterinarian 5, 867–878. McAfee, P.C., Bohlman, H.H., Ducker, T., Eismont, F.J. (1986) Failure of stabilization of the spine with methylmethacrylate. A retrospective analysis of twenty-four cases. The Journal of Bone and Joint Surgery— American volume 68, 1145–1157. McAnulty, J.F., Lenehan, T.M., Maletz, L.M. (1986) Modified segmental spinal instrumentation in repair of spinal fractures and luxations in dogs. Veterinary Surgery 15, 143–149. McKee, W.M. (1990) Spinal trauma in dogs and cats: a review of 51 cases. Veterinary Record 126, 285–289. Millis, D.L., Martinez, S.A. (2003) Bone grafts. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1875–1890. Philadelphia: Elsevier Science. Moise, N.S., Flanders, J.A. (1983) Micturation disorders in cats with sacrocaudal vertebral lesions. In: R.W. Kirk (ed.), Current Veterinary Therapy VIII, 772–776. Philadelphia: WB Saunders. Murtaugh, R.J. and Ross, J.N. (1988) Cardiac arrhythmias: pathogenesis and treatment in the trauma patient. Compendium on Continuing Education for the Practicing Veterinarian 10, 332–339. Neer, T.M. (1992) A review of disorders of the gallbladder and extrahepatic biliary tract in the dog and cat. Journal of Veterinary Internal Medicine 6, 186–192. Novelli, A. (1999) Antimicrobial prophylaxis in surgery: the role of pharmacokinetics. Journal of Chemotherapy 11, 565–572. Olby, N.J., Harris, T., Munana, K.R., Skeen, T.M., Sharp, N.J.H. (2003) Long-term functional outcome of dogs with severe injuries of the thoracolumbar spinal cord; 87 cases. Journal of the American Veterinary Medical Association 222, 762–769. Papadopoulos, S.M., Selden, N.R., Quint, D.J., Patel, N., Gillespie, B., Grube, S. (2002) Immediate spinal cord decompression for cervical spinal cord injury: feasibility and outcome. Journal of Trauma 52, 323–332. Pare, B., Gendreau, C.L., Robbins, M.A. (2001) Open reduction of sacral fractures using transarticular implants at the articular facets of L7–S1: 8 consecutive canine patients (1995–1999). Veterinary Surgery 30, 476–481. Patterson, R.H., Smith, G.K. (1992) Backsplinting for treatment of thoracic and lumbar fracture/luxation in the dog: principles of application and case series. Veterinary and Comparative Orthopaedics and Traumatology 5, 179–187. Perrouin-Verbe, B., Lenne-Aurier, K., Robert, R., Auffray-Calvier, E., Richard, I., Mauduyt de la Greve, I., Mathe, J.F. (1998) Post-traumatic syringomyelia and post-traumatic spinal canal stenosis: a direct relationship: review of 75 patients with a spinal cord injury. Spinal Cord 36, 137–143. Platt, S.R., Radaelli, S.T., McDonnell, J.J. (2001) The prognostic value of the modified Glasgow Coma Scale in head trauma in dogs. Journal of Veterinary Internal Medicine 15, 581–584.
Ramon, S., Dominguez, R., Ramirez, L., Paraira, M., Olona, M., Castello, T., Garcia Fernandez, L. (1997) Clinical and magnetic resonance imaging correlation in acute spinal cord injury. Spinal Cord 35, 664–673. Roosen, K., Grote, W., Liesegang, J., Linke, V. (1978) Epidural temperature changes during anterior cervical interbody fusion with polymethylmethacrylate. Advances in Neurosurgery 5, 373–375. Rouse, G.P. (1979) Cervical spinal stabilization with methylmethacrylate. Veterinary Surgery 8, 1. Rouse, G.P., Miller, J.I. (1975) The use of methylmethacrylate for spinal stabilization. Journal of the American Animal Hospital Association 11, 418–425. Salisbury, S.K., Cook, J.R., Jr (1988) Recovery of neurological function following focal myelomalacia in a cat. Journal of the American Animal Hospital Association 24, 227–230. Sandman, K.M., Smith, C.W., Harari, J., Manfra Maretta, S., Pijanowski, G.J. (2001) Comparison of pull-out resistance of Kirschner wires and Imex miniature interface fixation pins in polyurethane foam. Veterinary and Comparative Orthopaedics and Traumatology 15, 18–22. Schulz, K.S., Waldron, D.R., Grant, J.W., Shell, L., Smith, G., Shires, P.K. (1996) Biomechanics of the thoracolumbar vertebral column of dogs during lateral bending. American Journal of Veterinary Research 57, 1228–1232. Schulz, K.S., Waldron, D.R., Fahie, M. (1997) Application of ventral pins and polymethylmethacrylate for the management of atlantoaxial instability: results in nine dogs. Veterinary Surgery 26, 317–325. Schwarz, T., Crawford, P.E., Owen, M.R., Stork, C.K., Thompson, H. (2001) Fatal pulmonary fat embolism during humeral fracture repair in a cat. Journal of Small Animal Practice 42, 195–198. Selcer, R.R., Bubb, W.J., Walker, T.L. (1991) Management of vertebral column fractures in dogs and cats: 211 cases (1977–1985). Journal of the American Veterinary Medical Association 198, 1965–1968. Selden, N.R., Quint, D.J., Patel, N., d’Arcy, H.S., Papadopoulos, S.M. (1999) Emergency magnetic resonance imaging of cervical spinal cord injuries: clinical correlation and prognosis. Neurosurgery 44, 785–792; discussion 792–793. Sharp, N.J.H., Gilson, S.D., Kornegay, J.N., Wheeler, S.J., Hopkins, A.L., Lane, S.D. (1998) Long term follow-up of spinal fracture fixation using screws or pins and bone cement in 30 dogs. Veterinary Orthopedic Society 1998; Snowmass, CO. Shires, P.K., Waldron, D.R., Hedlund, C.S., Blass, C.E., Massoudi, L. (1991) A biomechanical study of rotational instability in unaltered and surgically altered canine thoracolumbar vertebral motion units. Progress in Veterinary Neurology 2, 6–14. Shores, A., Nichols, G., Rochat, M., Fox, S.M., Burt, G.J., Fox, W.R. (1989) Combined Kirschner–Ehmer device and dorsal spinal plate fixation technique for caudal lumbar vertebral fractures in dogs. Journal of the American Veterinary Medical Association 195, 335–339. Skeen, T.M., Olby, N.J., Munana, K.M., Sharp, N.J.H. (2003) Spinal arachnoid cysts in 17 dogs. Journal of the American Animal Hospital Association 39, 271–282. Smeak, D.D., Olmstead, M.L. (1985) Fracture/luxations of the sacrococcygeal area in the cat. A retrospective study of 51 cases. Veterinary Surgery 14, 319–324. Smith, G.K., Walter, M.C. (1985) Fractures and luxations of the spine. In: C.D. Newton and D. Nunamaker (eds), Textbook of Small Animal Orthopedics, 307–322. Philadelphia: JB Lippincott. Smith, G.K., Walter, M.C. (1988) Spinal decompressive procedures and dorsal compartment injuries: comparative biomechanical study in canine cadavers. American Journal of Veterinary Research 49, 266–273. Snyder, P.S., Cooke, K.L., Murphy, S.T., Shaw, N.G., Lewis, D.D., Lanz, O.I. (2001) Electrocardiographic findings in dogs with motor vehiclerelated trauma. Journal of the American Animal Hospital Association 37, 55–62. Stone, E.A., Betts, C.W., Chambers, J.N. (1979) Cervical fractures in the dog: a literature and case review. Journal of the American Animal Hospital Association 14, 463–471. Swaim, S.F. (1971) Vertebral body plating for spinal immobilization. Journal of the American Veterinary Medical Association 158, 1683–1695. Swaim, S.F. (1975) Evaluation of four techniques of cervical spinal fixation in dogs. Journal of the American Veterinary Medical Association 166, 1080–1086.
Trauma
Tator, C.H., Fehlings, M.G., Thorpe, K., Taylor, W. (1999) Current use and timing of spinal surgery for management of acute spinal cord injury in North America: results of a retrospective multicenter study. Journal of Neurosurgery 91 (1 Suppl), 12–18. Taylor, R.A. (1981) Treatment of fractures of the sacrum and sacrococcygeal region. Veterinary Surgery 10, 119–124. Tidwell, A.S., Mahony, O.M., Moore, R.P., Fitzmaurice, S.N. (1994) Computed tomography of an acute hemorrhagic cerebral infarct in a dog. Veterinary Radiology and Ultrasound 35, 290–296. Turner, W.D. (1987) Fractures and fracture-luxations of the lumbar spine: retrospective study in the dog. Journal of the American Animal Hospital Association 23, 459–464. Ullman, S.L., Boudrieau, R.J. (1993) Internal skeletal fixation using a Kirschner apparatus for stabilization of fracture/luxations of the lumbosacral joint in six dogs. A modification of the transilial pin technique. Veterinary Surgery 22, 11–17. Vaccaro, A.R., Madigan, L. (2002) Spinal applications of bioabsorbable implants. Orthopedics 25 (10 Suppl), s1115–1120. Vaccaro, A.R., Venger, B.H., Kelleher, P.M., Singh, K., Carrino, J.A., Albert, T., Hilibrand, A. (2002) Use of a bioabsorbable anterior cervical plate in the treatment of cervical degenerative and traumatic disk disruption. Orthopedics 25 (10 Suppl), s1191–1199; discussion s1199. Viguier, E., Petit-Etienne, G., Magnier, J., Lavaste, F. (2002) In vitro biomechanical evaluation of unstable T13–L1 stabilization procedures in dogs. Veterinary Surgery 31, 288. Waldron, D.R., Shires, P.K., McCain, W., Hedlund, C., Blass, C.E. (1991) The rotational stabilizing effect of spinal fixation techniques in an unstable vertebral model. Progress in Veterinary Neurology 2, 105–110. Walker, T.M., Pierce, W.A., Welch, R.D. (2002) External fixation of the lumbar spine in a canine model. Veterinary Surgery 31, 181–188. Walter, M.C., Smith, G.K., Newton, C.D. (1986) Canine lumbar spinal internal fixation techniques. A comparative biomechanical study. Veterinary Surgery 15, 191–198.
Waters, D.J., Wallace, L.J., Roy, R.G. (1994) Myelopathy in a dog secondary to scar tissue (cicatrix) formation: a complication of vertebral articular facet fracture. Progress in Veterinary Neurology 5, 105–108. Weisman, D.L., Olmstead, M.L., Kowalski, J.J. (2000) In vitro evaluation of antibiotic elution from polymethylmethacrylate (PMMA) and mechanical assessment of antibiotic-PMMA composites. Veterinary Surgery 29, 245–251. Weisse, C., Aronson, L.R., Drobatz, K. (2002) Traumatic rupture of the ureter: 10 cases. Journal of the American Animal Hospital Association 38, 188–192. Wheeler, J.L., Cross, A.R., Rapoff, A.J. (2002) A comparison of the accuracy and safety of vertebral body pin placement using a fluoroscopically guided versus an open surgical approach: an in vitro study. Veterinary Surgery 31, 468–474. Willer, R.L., Egger, E.L., Histand, M.B. (1991) Comparison of stainless steel versus acrylic for the connecting bar of external skeletal fixators. Journal of the American Animal Hospital Association 27, 541–548. Wong, W.T., Emms, S.G. (1992) Use of pins and methylmethacrylate in stabilisation of spinal fractures and luxations. Journal of Small Animal Practice 33, 415–422. Yarrow, T.G., Jeffery, N.D. (2000) Dura mater laceration associated with acute paraplegia in three dogs. Veterinary Record 146, 138–139. Yeh, L.S., Hou, S.M. (1994) Repair of a mandibular defect with a free vascularized coccygeal vertebra transfer in a dog. Veterinary Surgery 23, 281–285. Yerby, S.A., Toh, E., McLain, R.F. (1998) Revision of failed pedicle screws using hydroxyapatite cement. A biomechanical analysis. Spine 23, 1657–1661. Zindrick, M.R., Wiltse, L.L., Widell, E.H., Thomas, J.C., Holland, W.R., Field, B.T., Spencer, C.W. (1986) A biomechanical study of intrapeduncular screw fixation in the lumbosacral spine. Clinical Orthopaedics and Related Research 203, 99–112.
PROCEDURES Non-surgical (13.37–13.47) Cage confinement can give excellent results and is the only non-surgical option for most cats (Carberry et al., 1989; Selcer et al., 1991). A splint is preferred for dogs if certain criteria are satisfied including intact deep pain sensation, an intact ventral buttress and no pelvic, thoracic or soft tissue injuries (Patterson and Smith, 1992) (see page 290). It is essential that as much of the area under the splint as possible along with the skin of the groin and axilla be examined daily. Urine soiled straps must be replaced. Areas at risk from urine scalding or fecal soiling benefit from application of Desitin ointment (Pfizer Inc., New York). If there is any doubt, the straps or splint must be loosened sufficiently to allow proper inspection and then reattached. If doubt remains the splint must be removed. Useful indicators of potential problems are the animal’s mental attitude, its appetite, and the presence of pain or fever (13.36A). Most animals are surprisingly comfortable in a splint unless problems such as decubital ulcers develop. The splint should be maintained for a minimum of 4 weeks (3 weeks for immature animals), and is preferably then followed by a further 2–4 weeks of strict cage confinement.
305
306
Small Animal Spinal Disorders
13.37 Diagram to illustrate the construction of an external splint that is to be applied to a dog with a spinal injury. The splint is made from sheet aluminum, 0.54 mm in thickness, double or triple layered if necessary for added rigidity. In animals less than 10 kg, materials such as Orthoplast (Johnson and Johnson, New Brunswick, NJ) may be substituted. To make the splint, measure and cut the splint to the dimensions A–B, 2A–C and 2B–D (Patterson and Smith, 1992).
13.37
13.38 The splint extends from between the scapulae to the tail base, measured directly along the spine. Most animals have marked kyphosis following spinal injury, which must be considered during measurement.
13.38
13.39 The material is bent along its longitudinal axis to form a ridge; the splint is flattened slightly over the pelvic area. At the sides, the splint should extend laterally to the mid-scapula and the hip joint (13.46). The splint should be well padded with cotton, especially at the edges, and covered with Elastoplast (Johnson and Johnson, New Brunswick, NJ). 13.39
13.40 Some animals tolerate application of the splint while fully conscious, but most require sedation and a few need general anesthesia (Patterson and Smith, 1992). With the animal on its side, the splint is slid under the dog with the ridge aligned along the animal’s dorsal midline.
13.40
Trauma
13.41 The animal and splint are then rotated
simultaneously until the animal is lying on its back in the splint (Patterson and Smith, 1992). The dog usually relaxes and the fracture may reduce to some degree. The axillae and groin are padded heavily with cotton wool prior to attaching the splint using Elastoplast (Johnson and Johnson, New 13.41 Brunswick, NJ). This has the advantage of being light, porous and adhesive. Velcro straps are much easier to re-adjust but more likely to cause abrasions. Velcro is preferred when the splint needs to be removed frequently.
13.42 To secure the dog to the splint, the caudal half of the animal is first extended over the end of a table.
13.42
13.43 Elastoplast (Johnson and Johnson, New Brunswick, NJ) is applied in a cruciate pattern over the pelvis, avoiding the anus and vulva. In male animals it is important that any strap crossing the base of the penis is not too tight, or pressure necrosis of the urethra is possible (Patterson and Smith, 1992). 13.43
13.44 In male dogs, a cod-piece roughly the shape of a half flowerpot can be made from aluminum, Orthoplast (Johnson and Johnson, New Brunswick, NJ) or an empty saline bag to protect the base of the penis and to try to funnel urine away from the abdomen and inguinal straps (Bagley et al., 2000). 13.44
307
308
Small Animal Spinal Disorders
13.45 In the pectoral region, the straps are applied as a figure of eight, extending in front of the shoulder and under the neck like a harness. The straps around the chest must be snug but must not impede ventilation. Radiographs can be taken through the splint to check fracture alignment.
13.45
13.46 The main advantages of an external splint are that it is inexpensive and strong (Patterson and Smith, 1992). It also facilitates lifting the animal, turning it from side to side, moving it to clean the kennel, or taking it out to void. Handles made from aluminum rods can also be incorporated into the splint or may be taped on afterwards (Bagley, 2000). 13.46
13.47 The main disadvantage is the potential for pressure sore formation along the edges and underneath the splint (13.36A). In addition, the straps can cause skin excoriation in the axilla or inguinal region. They may also hinder manual expression of the bladder, and prevention of urine scald can be challenging.
13.47
Trauma
Metal and bone cement (13.48–13.60) This is the internal fixation technique of choice. Threaded pins or screws should be used (Sandman et al., 2001); the implant diameter should be as large as possible (Beaver et al., 1996), each implant should penetrate two cortices (Zindrick et al., 1986), and ideally three implants should be used on each side of the lesion (Garcia et al., 1994). The cement bar must also be strong enough (Rouse and Miller, 1975; Willer et al., 1991) (13.59). Variables in using implants include: •
The pattern in which implants are placed (13.49, 13.54).
•
The angle or trajectory at which they penetrate the vertebrae (13.52).
•
Their entry points (13.55).
Intravenous antibiotic effective against staphylococci (such as a cephalosporin) is given and repeated every 1–2 h during surgery (Marcellin Little et al., 1996; Novelli, 1999; Kriaras et al., 2000). Fracture reduction can be very difficult in some injuries without some type of mechanical assistance (Boudrieau, 1997; Schulz et al., 1997; Bagley et al., 2000) (13.64). A laminectomy spreader, Senn retractor, or a curved hemostat can be used for reduction of L7 fractures or manual traction can be put on the tail (Blass and Seim, 1984; Beaver et al., 1996; Harrington and Bagley, 1998; Bagley et al., 2000). Cervical injuries can be reduced using traction on the maxilla or a Gelpi retractor can be placed in holes drilled in the middle of a vertebra (Blass and Seim, 1984; Blass et al., 1988) (11.38). A Scoville–Haverfield or similar retractor can also be placed in adjacent disc spaces or between a disc space and the base of the skull for C2 fractures (Boudrieau, 1997). Initial stabilization is provided by bone-holding forceps (13.48, 13.60) while K-wires are placed across the articular facets to provide additional stability and maintain alignment (Blass and Seim, 1984; Beaver et al., 1996; Berry et al., 1999) (13.28A, 13.48). When a hemilaminectomy must be performed, when a facet is fractured, and when a ventral approach is used for cervical fractures, the K-wire can instead be placed across a disc space to maintain reduction (Blass et al., 1988). Additional methods for reducing a fracture are discussed in 13.64. A small K wire is used initially to make a point of purchase for the drill bit. It can also be used to make a test hole to evaluate bone quality over the chosen path of the implant. If bone quality is poor this small test hole can easily be redirected without prejudicing the final pilot hole. Pins should be placed using a low power setting to reduce bone necrosis; a pilot hole is recommended for threaded pins (Egger et al., 1986; Walker et al., 2002). Muscle should be protected during drilling with a drill guard. If brisk hemorrhage arises from the pilot hole then a finger or bone wax should be placed over the hole, which should be tapped and screwed as soon as possible (1.19). A flatter implant trajectory increases implant holding strength due to the greater bone contact (Garcia et al., 1994). Such an angle is also necessary if a hemilaminectomy has been performed so that the cement can be positioned well away from the spinal cord. Landmarks for lumbar vertebrae are the ventral portion of the base of the accessory process and the junction of the transverse process with the vertebral body (Rouse and Miller, 1975; Blass and Seim, 1984; Wong and Emms, 1992; Bagley et al., 2000; LeCouteur and Sturgess, 2003). Entry points should be between these landmarks and are discussed in 13.55. A skeleton should always be available for reference. The implant entry point should be no higher than the base of the accessory process. If a flat trajectory is to be used then the entry point should ideally be no higher than the floor of the vertebral canal otherwise the vertebral canal may be entered inadvertently (Walker et al., 2002). There is some variation based on the exact vertebra (1.21–1.23, 1.25, 13.56). Also, due to the concave ventral surface in the middle of each vertebral body, implant placement must be more precise centrally. A CT scan is invaluable for planning the entry point as well as the trajectory. The bone cement is applied in a cylindrical pattern on each side of the spine. Wound closure is then much easier than when cement is applied in a doughnut pattern. Implants must be encased fully by the cement (13.57). The wound is irrigated with saline to dissipate heat generated by the curing of the cement. The spinal cord and nerves must not contact the cement, especially following hemilaminectomy. Gelfoam (Pharmacia, Kalamazoo, MI) has good insulating properties and should be used to protect exposed spinal cord (Roosen et al., 1978; Boker et al., 1989). Excision of epaxial muscle facilitates wound closure (Blass and Seim, 1984). Potential causes of cement failure include a fold or thin region of the cement bar. Cement can be reinforced with a pin provided that this is encased fully by the cement (Bagley et al., 2000) (13.25, 13.28, 13.48). Antibiotic should not be added to the methylmethacrylate unless done in a very carefully controlled fashion as it tends to reduce strength (Ethell et al., 2000; Weisman et al., 2000). It is preferable to give a second dose of intravenous antibiotic just prior to bone cement application.
309
310
Small Animal Spinal Disorders
13.48 A: Counter pressure is applied by bone clamps on the spinous processes during the initial approach, K-wire fixation of facets (shown here) and while the cement hardens (Blass and 13.48 Seim, 1984) (13.24). Note the torn supraspinous A B ligament (arrow). B: All K-wires must be incorporated fully into the cement to prevent migration (13.63). If this is not possible then small screws should be used. In this Pomeranian, three screws and three positive profile pins were used for fixation; several hairline fractures dictated the implant pattern. The dog (4.38) was almost normal 6 months later.
13.49 Diagrams to illustrate one possible implant pattern. The preferred pattern is 13.49 to drive implants perpendicular to A the vertebral body (13.54A), which allows placement of three implants in each vertebra instead of the two shown here (Garcia et al., 1994). A: Dorsoventral view. B: Transverse view.
B
13.50 Steinmann pins must be cut and notched and the ends then bent to provide purchase for the bone cement. Positive profile pins are preferred due to the tendency for smooth pins to migrate (Wong and Emms, 1992) (13.29). They also have a roughened surface for better cement attachment and do not need to be notched (5.32). 13.50
Trauma
13.51 A: Lateral, and B: ventrodorsal radiograph of pins and bone cement fixation. This dog developed a late postoperative infection (13.34). Alignment is suboptimal. In theory this dog is at risk for late onset syringomyelia (Perrouin-Verbe, et al., 1998).
13.51 A
B
13.52 Range of angles that can be used for vertebral implants (same dog as in 13.10). A 30° angle from horizontal has been recommended as standard (Walker 13.52 et al., 2002; Wheeler et al., 2002). A A: Transverse CT scan B of T12 vertebra to show the range of possible implant trajectories. The angles have to be more vertical in thoracic vertebrae because of the ribs. Landmarks for thoracic vertebrae are the ventral portion of the base of the accessory process and the tubercle of the rib. B: Transverse CT scan of L2 vertebra to show the range of possible implant trajectories (1.21–1.23).
311
312
Small Animal Spinal Disorders
13.53 A: This 3D reconstruction is from the dog shown in 13.31–13.33. The screws are visible in blue and were placed using the more vertical angle shown in 13.49B than the flat angle shown in 13.54B. B: Corresponding ventrodorsal radiograph to show screw position.
13.53 A
B
13.54
13.54 Diagrams to show the preferred implant pattern. A: Dorsoventral view. The advantage of this pattern is that it makes it easier to place three implants in each vertebral body (Garcia et al., 1994). B: Transverse view to show a flat trajectory for screws. Implants can also be put in using the more vertical angle shown in 13.49 and 13.53 but a flatter trajectory provides slightly more bone purchase as well as more room for cement (13.52).
A
B
Trauma
13.55 Illustration of potential entry points in the vertebral bodies. A, B: T12 vertebra. C, D: L2 vertebra. Implants should enter bone between the two lines; the higher the entry point, the more vertical the trajectory and vice versa. If further clarification of anatomy is needed, the floor of the vertebral canal can be palpated with care at the intervertebral foramen using a fine, curved dental instrument.
13.55 A
B
C
D
13.56 The entry points and implant trajectory will vary with the anatomical location of the fixation. A: Entry points and potential implant trajectories are shown on a CT scan of C6 vertebra (11.41–11.43). The disc spaces and intervertebral 13.56 foramina should be avoided (Rouse, 1979; A B Wong and Emms, 1992). B: Entry points and potential implant trajectories are shown on a CT scan of L6 vertebra. The range of possible angles is less here than can be used at L2 and is even less for L7 vertebra (1.25). Variations may also be marked between different types of dog.
313
314
Small Animal Spinal Disorders
13.57 Fractures or luxations involving L7 vertebra may warrant different implant patterns. Radiographs of this 13.57 dog are shown in 13.27 and A B 13.65. A: Implants can be placed in L7 down the pedicle and into the base of an articular facet as shown here. Note that there is insufficient cement around the screws; the technique should also not have been used with a compromised ventral buttress. B: Screw placement in the sacrum (two most lateral screws on each side) is also discussed on page 204. Implants can also be placed into the wings of the ilium, but should be placed vertically (13.65) rather than horizontally (13.21, 14.12) for better bone purchase.
13.58 Bone cement is best applied when it no longer sticks to the surgeon’s gloves. It can be used when more liquid, but then bone wax wrappers should be used as a mold. Wrappers are first coated with bone wax to retard cement adhesion and they must be removed before it hardens.
13.58
13.59 Vertebral bodies A: before, and B: after bone cement application (13.33, 13.53, 13.57, 13.60). The cement bar must be of sufficient thickness; as a rough guide the bar should be between 1.25 and 2.5 cm in diameter depending on the weight of the animal. This usually requires 40 g of cement for large dogs and 20 g for smaller animals (Rouse and Miller, 1975; Willer et al., 1991). Ideally vacuum mixing should be used to prevent bubble formation and to lessen exposure of personnel (Anderson, 1988) (13.33). Steinmann pins of various sizes can be incorporated into the cement to reinforce it (13.25). This is useful for large dogs or at sites subject to excessive forces (11.17C).
13.59 A
B
Trauma
13.60 Intraoperative view A: after implant placement, and B: after bone cement application. Bone wax has been used to fill the screw heads to facilitate implant removal if necessary. Kirschner wires should ideally be placed through the articular facets once the fracture has been apposed to maintain alignment (13.28, 13.48). Bone holding forceps maintain reduction as cement hardens (13.24, 13.25, 13.48).
13.60 A
B
Modified segmental fixation (13.61–13.63) The original spinal stapling technique should only be used for cats or for dogs weighing less than 7 kg (Matthieson, 1983) (13.63). The improved modified segmental fixation has been used in dogs over 40 kg (McAnulty et al., 1986) (13.28B). Both techniques should ideally be used only when the ventral buttress is intact.
13.61 Diagram to show application of modified segmental fixation. The orthopedic wire penetrates the base of the articular facet and is twisted around the Steinmann pins. Wires can also be placed through the base of each spinous process.
13.61
315
316
Small Animal Spinal Disorders
13.62 Close-up of an articular facet joint to show the position of the wire in modified segmental fixation. Additional pins are used depending on the estimated stability of the fracture and on the weight of the dog (McAnulty et al., 1986).
13.62
13.63 Spinal stapling as shown in this 1.6 kg dog (13.16) and modified segmental fixation provide useful alternatives for fixation in the lumbosacral region (McAnulty et al., 1986). A: The bone fragment visible at L6/7 could not be reached after a dorsal laminectomy. B: The dog was normal and there was good fracture healing 6 weeks later although one K-wire had migrated. The dog was normal 6 months after surgery.
13.63 A
B
External fixation (13.64–13.66) Although dependent on the exact design of the fixator and on the type of implants used, this can be an extremely reliable method for use in even very large dogs (Walker et al., 2002) (13.65, 14.13). It is also a useful rescue procedure (13.27) and can be used at infected sites (14.12). Placement of pins under fluoroscopic guidance may give more consistent insertion angles and better bone purchase; a closed technique makes fracture stabilization during implant insertion of paramount importance (Wheeler et al., 2002).
Trauma
13.64 Reduction of an overridden L7 fracture using a laminectomy spreader (same dog as in 13.27) (Bagley et al., 2000). The positions of the laminectomy spreader jaws are marked by white arrowheads. * ⫽ articular facet of L7. A: Before reduction. A short dorsal laminectomy has been made in the center of the caudal lamina of L7; the laminectomy spreader has been anchored cranially at this point and caudally under the roof of S1. B: Fracture reduced. Torn articular facet joint capsule is now visible (black arrow).
13.64 A
B
317
318
Small Animal Spinal Disorders
13.65 After failure of the implant shown in 13.27, an external apparatus was applied as shown. A: Lateral view. B: Dorsoventral view. Implants in each ilial wing 13.65 are angled from dorsal to A B ventral to obtain better purchase into soft bone. Skin and lumbodorsal fascia must be pulled towards the midline prior to pin penetration or there will be excessive tension at wound closure (Shores et al., 1989; Lanz et al., 2000).
13.66 External fixator (SK External Fixation System, Imex Veterinary Inc., Longview, TX) in place (14.13). Postoperative management can be labor intensive. There is also a risk of traumatic removal; this dog panicked and removed the entire device as it tried to enter its kennel during a thunderstorm 5 weeks after fixation. Incredibly, the dog’s neurological status, which was almost normal, did not deteriorate (Lanz et al., 2000; Walker et al., 2002). 13.66
Miscellaneous conditions
Degenerative 319 Degenerative myelopathy (chronic degenerative radiculomyelopathy, CDRM) 319 Synovial cyst 320 Facet joint pain 320 Sacroiliac joint pain 320 Leukodystrophies 320 Arachnoid cyst 321 Spondylosis deformans 321 Anomalous 321 Congenital vertebral anomalies Sacrocaudal dysgenesis 322 Spina bifida 322 Tethered spinal cord 322 Spinal dysraphism 322 Syringomyelia and hydromyelia Cartilaginous exostoses 323 Dermoid sinus 323 Epidermoid cyst 323 Arachnoid cyst 323 Metabolic 326 Lysosomal storage diseases Osteoporosis 326 Nutritional 326 Hypervitaminosis A
321
331
Vascular 332 Fibrocartilaginous embolism (ischemic myelopathy) 332 Ascending myelomalacia 332 Ischemic neuromyopathy (aortic embolism, iliac thrombosis) 332 Spinal cord hemorrhage, hematoma 333 Intermittent claudication 334 Key issues for future investigation
334
334
Further reading
337
322
326
326
Infectious/inflammatory 326 Discospondylitis 326 Epidural empyema 328 Inflammatory CNS diseases 329 Epidural steatitis 331 331
14
Psoas muscle injury
References
Idiopathic 326 Tumoral calcinosis 326 Disseminated idiopathic skeletal hyperostosis (DISH) 326
Trauma
Chapter
This chapter covers various other important conditions of the spine that are likely to be encountered but are not covered specifically elsewhere in this book. Surgical treatment is not indicated in many of these conditions, thus they must be diagnosed correctly to avoid unnecessary operations (Braund and Sharp, 2003). Key references are given, but the list is not intended to be encyclopedic. See Dewey et al. (2003), Braund (Further reading) and LeCouteur et al. (2000), which are comprehensive and have excellent bibliographies. The diseases are listed according to the DAMNIT scheme (Table 3.1).
DEGENERATIVE Degenerative myelopathy (chronic degenerative radiculomyelopathy, CDRM) This is a degenerative condition of the spinal cord of older dogs; it is mostly seen in large breeds and particularly German shepherd dogs. A similar disorder has been described in a cat (Mesfin et al., 1980). There are probably several different types of canine degenerative myelopathy that are grouped together and cannot be differentiated at present. Phenotypes in the various breeds of dog may vary somewhat from that of the classic condition found in the German shepherd dog (Matthews et al.,
320
Small Animal Spinal Disorders
1985; Salinas and Martinez, 1993). Onset of signs in German shepherd dogs is from 5 years of age. Progressive pelvic limb ataxia, loss of proprioception, and paraparesis occur. The neurological examination usually indicates a T3–L3 lesion, but some dogs lose their patellar reflex because of dorsal nerve root involvement. Urinary and fecal function is normal. Spinal pain is not seen. Signs in affected dogs progress over several months, and eventually the thoracic limbs will be involved (Averill, 1973; Griffiths and Duncan, 1975). The diagnosis is confirmed by the absence of structural spinal cord disease on myelography or MRI. Even if other lesions are found, for example a disc extrusion, the possibility of degenerative myelopathy also being present should be considered (10.53). Analysis of lumbar CSF often reveals a moderate elevation in protein. The cause is unknown and there is no known treatment. Degenerative changes are found at necropsy throughout the spinal cord and in nerve roots, mainly affecting the thoracolumbar region. There is a loss of both myelin and axons, with cellular infiltration in the most severely affected regions (Johnston et al., 2000).
animals (Cook et al., 2002). Diagnosis is not simple and depends largely on nerve blocks in humans (Manchikanti, 1999). Signal intensity changes in lumbar pedicles on MRI may be a diagnostic tool that is applicable to animals (Morrison et al., 2000).
Synovial cyst
Leukoencephalomalacia is a degenerative CNS disease, reported in Rottweilers in the USA, the Netherlands and Australia. There is malacia throughout the spinal cord, particularly in the cervical region, due to demyelination and cavitation. The etiology is unclear. Affected dogs show clinical signs from 1 to 4 years of age. There is pelvic limb ataxia and hypermetria, progressing through paraparesis to tetraparesis over a period of months. Proprioceptive deficits are marked. Limb reflexes are intact or hyperactive. Differentiation from neuroaxonal dystrophy and other CNS conditions is important. All routine diagnostic tests are normal in leukoencephalomalacia. The major differential features are the presence of proprioceptive deficits in leukoencephalomalacia, and tremor or dysmetric head movements together with decreased menace responses in neuroaxonal dystrophy (see below). There is no treatment and the prognosis is poor; most dogs with leukoencephalomalacia are euthanized within 1 year of presentation (Chrisman, 1992).
These cysts arise from the synovial joints between the articular facets. Often the affected joint(s) also show(s) degenerative changes on survey radiographs. The cysts tend to occur mainly in the cervical area of giant-breed dogs but can be found in the thoracolumbar area (Dickinson et al., 2001); they have also been reported in a German shepherd dog presenting with lumbosacral pain (Webb et al., 2001). Synovial cysts in humans are thought to be an under-diagnosed problem in the elderly and this problem will probably be recognized more frequently in dogs with the increased use of cross-sectional imaging and especially MRI (Charest and Kenny, 2000). CT myelography and MRI reveal discrete, round, cystic structures (Levitski et al., 1999a,b; Lipsitz et al., 2001) (11.8). Imaging by myelography usually shows dorsolateral compression of the spinal cord due to soft tissue rather than bone. Cysts in the caudal neck may be located more laterally or even ventrolaterally within the vertebral canal (4.25, 11.8). Dorsal laminectomy and removal of the cyst or distraction–stabilization (page 219) can be performed. Prognosis has been good over a mean follow-up of 17 months although it is not known how young dogs do over a longer period of time (Dickinson et al., 2001). In humans, synovial cysts are usually associated with facet joint degenerative disease and so there may be overlap with facet joint pain (see below) (Howington et al., 1999).
Facet joint pain Facet joint pain is a common cause of both neck and low back pain in humans and is a potential source of pain in
Sacroiliac joint pain This is another common cause of low back pain in humans that can mimic disc pain and that might have a counterpart in animals (Pang et al., 1998; Swezey, 1998; Hodge and Bessette, 1999). CT evidence of osteoarthritis in the joint is suggestive but MRI will probably prove to be more useful as an indicator of physiological significance (Morrison et al., 2000).
Leukodystrophies Several breed-specific disorders are described in Afghan hounds, Dalmatians, Miniature poodles and Dutch kooiker dogs. The most common examples are seen in Rottweilers.
ROTTWEILER LEUKOENCEPHALOMALACIA
ROTTWEILER NEUROAXONAL DYSTROPHY Neuroaxonal dystrophy is a degenerative CNS disease seen mainly in Rottweilers, but has also been reported occasionally in cats. Axonal dystrophy with spheroids is seen in parts of the CNS, including the dorsal horn gray matter of the spinal cord, and in the gracile, cuneate and dorsal spinocerebellar tract nuclei. Cerebellar atrophy may also be seen. The etiology is unknown and there may be overlap in some dogs between this condition and leukoencephalomalacia.
Miscellaneous conditions
Affected dogs show clinical signs from puppyhood, but these may not be noticed until the dog is adult. There is pelvic limb ataxia and hypermetria of the thoracic limbs. Proprioception is normal. Limb reflexes are intact or even hyperactive. Head incoordination, tremor, positional nystagmus, and loss of menace response (due to cerebellar involvement) develop later, often after several years. Weakness is not seen. Differentiation from other CNS conditions is important (see above). Routine diagnostic tests are normal. There is no treatment and the long-term prognosis is poor, although affected dogs may survive as active pets for several years (Chrisman, 1992).
Arachnoid cyst In an arachnoid cyst there is a focal accumulation of CSF in the subarachnoid space, which compresses the spinal cord (Dyce et al., 1991; Skeen et al., 2003). These can be acquired secondary to some type of injury (13.32) although they are usually congenital lesions (see below).
14.1 Spondylosis deformans was an incidental finding in this 7-year-old Boxer dog.
A
Spondylosis deformans Spondylosis deformans is a common radiographic finding in older dogs, but it is rarely associated with clinical signs (Morgan et al., 1989). Generally the osteophytes develop ventrally and laterally on the vertebral body, and they may grow to the point that they bridge the intervertebral space (Larsen and Selby, 1981) (14.1). Osteophytes around the articular facet joints may be of more significance (see ‘Facet joint pain’, above). The thoracolumbar junction and lumbosacral joint are particularly affected. Where these changes are seen at the lumbosacral joint they may be related to clinical signs of lumbosacral disease. However, this diagnosis should not be reached on the basis of survey radiographs alone, as such changes are seen in many normal dogs (see page 188). It is important to differentiate spondylosis deformans from the changes seen in discospondylitis (14.11–14.13).
ANOMALOUS Congenital vertebral anomalies Vertebral malformations are common findings in dogs and are also seen occasionally in cats. Some do cause compressive myelopathy or are associated with anomalies of the spinal cord (Bailey, 1975; Bailey and Morgan, 1992). Many cases of atlantoaxial subluxation also have an underlying congenital vertebral malformation (see Chapter 9). Anomalies such as hemivertebrae, butterfly vertebrae and block vertebrae are relatively common but they rarely cause clinical signs. Hemivertebrae are seen usually in the small, brachycephalic breeds. They
B
14.2 Dog with severe ataxia and paraparesis due to a hemivertebra at T7. Threaded pins and bone cement were used to stabilize the site prior to performing a dorsal laminectomy. Neurological function was unchanged postoperatively and the dog was walking well within a few days.
321
322
Small Animal Spinal Disorders
are wedge-shaped deformities that may cause spinal deviation in the lateral or dorsoventral plane depending on the orientation of the wedge (14.2). When the deviation is marked they tend to cause a chronic, progressive myelopathy. Skeletal abnormality may be apparent on physical examination but the exact site should be confirmed by neuroimaging (Dewey et al., 2003). Animals that have severe spinal cord compression may show a marked deterioration of neurological deficits after invasive tests such as myelography, so MRI is preferred. Deterioration is also common when animals with these lesions are subject to decompressive laminectomy (page 126). Stabilization of the vertebral column should be the primary aim of surgery and decompression should not be performed without concomitant fixation (Aikawa, 2001). Nevertheless, the prognosis is guarded in dogs with severe neurological deficits. Vertebral anomalies would be an ideal condition in which to use electrophysiological monitoring during spinal cord surgery (see Chapter 4, page 60).
Sacrocaudal dysgenesis Cats and dogs with congenital tail defects often have vertebral abnormalities of the sacrum and caudal vertebrae. Manx cats, Pugs and Bulldogs are affected most often and the condition is inherited in the Manx. The vertebral abnormalities may themselves cause neurological deficits that affect the pelvic limbs, lower urinary tract and anus or there may be malformations of the spinal cord such as spina bifida. Diagnosis is suspected from the clinical signs. Radiography will demonstrate vertebral abnormalities and myelography or MRI may reveal spinal cord malformations. Treatment is not possible and the prognosis is poor.
Spina bifida This is a developmental defect resulting from failure of the embryonic vertebral arch to fuse normally. There may be protrusion of the meninges or spinal cord into a meningocele, a myelocele or a meningomyelocele; this is termed spina bifida aperta. Alternatively, there may be no protrusion of nervous tissue, which is termed spina bifida occulta (Wilson et al., 1979; Wilson, 1982). There is a high incidence of these conditions in English bulldogs and Manx cats. Spina bifida usually involves the caudal lumbar spine. Clinical signs indicative of L4–S3 spinal cord dysfunction occur. Radiography may reveal defects in the dorsal vertebral arch, such as paired spinous processes, and myelography or MRI may demonstrate a meningocele. Treatment is not possible but untethering may help (see below). When there is an opening in the skin, closure should be considered to prevent meningomyelitis.
Tethered spinal cord This condition is usually associated either with spina bifida and meningo(myelo)cele or with an intradural lipoma (Fingeroth et al., 1989; Plummer et al., 1993; Huttmann et al., 2001; Shamir et al., 2001). It can also be caused by previous surgery that leaves a residual defect in the dura mater with subsequent spinal cord herniation (Henry et al., 1997).
Spinal dysraphism Malformations of the spinal cord have been described in several breeds of dogs, particularly Weimaraners (Broek et al., 1991). Various lesions of the central canal, gray matter, dorsal sulcus and ventral fissure have been described and a syringohydromyelia may also be present. A bunny-hopping pelvic limb gait, abnormalities of the hair coat, a depression of the sternum, and head tilt may be seen. Neurological deficits localize to the T3–L3 spinal cord and signs are often evident in affected puppies. The signs are usually non-progressive. Abnormalities may be detected using MRI (Dewey et al., 2003).
Syringomyelia and hydromyelia Syringomyelia and hydromyelia are fluid-filled cavitations of the spinal cord and central canal respectively. Syringomyelia can be a congenital or an acquired (11.12) spinal cord lesion (Cauzinille and Kornegay, 1992; Perrouin-Verbe et al., 1998), whereas hydromyelia is more often associated with congenital malformations. Hydromyelia can also be a complication of myelography (Kirberger and Wrigley, 1993), and has been described in a cat secondary to feline infectious peritonitis (FIP) (Tamke et al., 1988). Clinical differentiation is often impossible and these two conditions may occur together. In humans these conditions probably result from overcrowding within the caudal fossa of the skull, such as occurs in Chiari type-1 malformations. The overcrowding is thought to obstruct CSF flow and lead to the syringohydromyelia (Nishikawa et al., 1997). Early surgical intervention to decompress the caudal fossa and restore the flow of CSF in susceptible individuals might prevent the development of syringomyelia (Fischbein et al., 1999). The onset of clinical signs is usually in adult dogs. Diagnosis is aided considerably by MRI (Kirberger et al., 1997; Levitski et al., 1999b; Taga et al., 2000). Associated changes such as spinal hyperesthesia, torticollis or scoliosis are common (Child et al., 1986; Bagley et al., 1997; Dewey et al., 2003). Syringohydromyelia has been reported in Cavalier King Charles spaniels, which usually present with a characteristic scratching at the shoulder region (14.3A). Other signs include apparent neck, thoracic limb or ear pain and thoracic limb lower motor neuron (LMN) deficits. In these dogs the disorder is
Miscellaneous conditions
A
ribs and limb bones. It is caused by abnormal differentiation of cartilage cells in bones that develop by endochondral ossification, leading to the production of large masses composed of a thin cortex lined by cartilage and with a core of cancellous bone. The etiology is unclear (Gambardella et al., 1975; Meomartino et al., 1997). The masses continue to grow until skeletal maturity is reached and may also continue thereafter in some dogs. Diagnosis is by radiography (14.3B). Surgical decompression of compressive lesions may be necessary but the prognosis is guarded in dogs and is considered to be poor in cats. Neoplastic transformation can result in clinical signs in adult animals and this may actually be more common than was thought previously (Jacobson and Kirberger, 1996; Dewey et al., 2003).
Dermoid sinus
B 14.3 A: Syringohydromyelia (arrow), mild hydrocephalus and caudal occipital malformation in a Cavalier King Charles spaniel with persistent scratching at the shoulder. B: Multiple cartilaginous exostoses affecting the last rib and L4 vertebra.
thought to develop secondary to compression at the level of the foramen magnum (Rusbridge et al., 2000). Anti-inflammatory medications may improve the clinical signs temporarily but early surgical intervention is recommended in humans with Chiari type-1 malformations (Rusbridge, 1997; Fischbein et al., 1999). There is a trend away from shunting of the syrinx in humans towards decompressing the caudal fossa by an occipital craniectomy (Sgouros and Williams, 1995; Nishikawa et al., 1997; Sakamoto et al., 1999). This approach has also been used with success in a small number of dogs with syringomyelia (W.B. Thomas, personal communication). When syringomyelia develops following trauma, a decompressive laminectomy with subarachnoid space reconstruction is recommended (Sgouros and Williams, 1996).
Cartilaginous exostoses Cartilaginous exospores (osteochondromatosis) may cause spinal cord compression at any site of the vertebral column of dogs and cats. Lesions may also occur on the
In dermoid sinus (pilonidal sinus) the skin over the dorsal midline is inverted and, in some dogs, the invagination communicates with the dura mater. Rhodesian ridgebacks and Shih tzus have a high incidence (Tshamala and Moens, 2000). Infection from the cyst may extend to the spinal cord, causing meningitis and myelitis with associated clinical signs. Diagnosis is based on physical examination and clinical signs. If a cyst is suspected to be in communication with the vertebral canal, myelography or MRI is preferable to fistulography. Infected lesions are treated by antibiotics and surgical excision; it may be necessary to perform a laminectomy to retrieve all the tissue. Careless exploration of this type of lesion, without a full appreciation of its extent, can lead to the development of marked neurological deficits.
Epidermoid cyst This is a rare cystic lesion that arises from entrapment of epithelial cells within the neural tube. This lesion can be congenital or it can be acquired secondary to mechanical implantation of epithelial cells, such as by puncture of the spinal cord by a needle (Tomlinson et al., 1988). Myelography or MRI will show an intramedullary lesion, which is likely to be mistaken for a spinal cord tumor such as nephroblastoma of young dogs (see page 249).
Arachnoid cyst These are not true cysts as they do not have an epithelial lining (Dyce et al., 1991). Rather there is a focal accumulation of CSF, probably due to adhesions within the subarachnoid space (Dyce et al., 1991; Moissonier et al., 2002) (14.6). The natural pulsation of CSF is sufficient to produce a bulbous or tear-drop enlargement of the subarachnoid space and this can then cause marked spinal
323
324
Small Animal Spinal Disorders
cord compression (Dewey and Coates, 2003) (14.4). Dysraphism or syringomyelia may accompany this condition (Dyce et al., 1991; Galloway et al., 1999). The condition occurs at two main locations: the cervical area or the thoracolumbar junction. It should be suspected in a young dog with progressive signs of myelopathy, but which is pain free. Rottweilers may be over-represented (Moissonier et al., 2002; Rylander et al., 2002; Skeen et al., 2003). The condition is also described in cats (Shamir et al., 1997). Arachnoid cysts can also be acquired secondary to trauma or disc lesions (Rylander
et al., 2002; Skeen et al., 2003) (13.32). Diagnosis is by myelography or MRI (Galloway et al., 1999) (14.4). Sonography is helpful during surgery to provide intraoperative orientation (Galloway et al., 1999). Surgical decompression may be effective. A dorsal laminectomy is usually needed in order to gain adequate exposure of the cyst (11.55, 14.5). Seven of 11 dogs improved in one series after fenestration and resection of the cyst. However, three dogs with cervical lesions required postoperative ventilatory assistance, two because of hematoma formation (Rylander et al., 2002). Successful long-term
B
A
14.4 Ten-month-old Retriever with progressive paraparesis for 2 months. A: Well-defined, bulbous dilation of the dorsal subarachnoid space at C2/3. This communicates with the rest of the subarachnoid space as contrast flowed caudally within a few minutes (14.8B). B: Dorsoventral view confirms the discrete border of this lesion. Surgical appearance of the lesion is shown in 14.8A.
A
B
14.5 Seven-month-old Labrador with tetraparesis for 1 month. Myelography revealed a subarachnoid cyst at C2/3. A: The dura has been opened over the cyst; the wall is still intact and compressing the spinal cord (arrow). B: The cyst wall (arrow) has been incised to reveal the spinal cord beneath a pool of CSF. The dog was normal 2 years after surgery (Skeen et al., 2003).
Miscellaneous conditions
14.6 Three-year-old Toy poodle with a cervical arachnoid cyst. The dura has been opened to reveal an extensive network of fine adhesions (arrow) in the subarachnoid space. The dura mater and arachnoid adhesions are being manipulated (arrowhead) with two, modified 25-gauge needles (5.19). This dog’s gait improved but it remained fecally incontinent 4 months later. It then developed urinary incontinence and its gait deteriorated 26 months postoperatively (Skeen et al., 2003).
A
outcomes were obtained in 8 out of another 13 dogs that were followed for more than 1 year. There were no significant predictors of a good outcome although there was a trend towards a good outcome in dogs of less than 3 years of age, those that had a duration of signs of less than 4 months, and when marsupialization was used as the surgical technique (Skeen et al., 2003). Marsupialization is one potential technique for treating arachnoid cysts (McKee and Renwick, 1994). Durotomy is performed and the dural edges are sutured to periarticular soft tissues using magnification and nonabsorbable suture material (4/0 to 8/0 depending on the breed). A wide laminectomy gives excellent access but runs the risk of putting undue tension on the dura, especially at the cranial and caudal ends of the durotomy. If this happens a release incision should be made (14.7A). Another technique used to treat arachnoid cysts is to make a wide fenestration in the dura. Here the dura is excised over the cyst (14.8) and ideally visible subarachnoid adhesions (14.6) are also removed using meticulous hemostasis (Frykman, 1999; Rylander et al., 2002). A rare
B
14.7 A: Marsupialization at T12/13 in a 9-year-old Westie with progressive paraparesis and fecal incontinence for 2 months. A release incision may be needed if marsupialization puts tension on the dura (arrow). The dog was doing well 22 months after surgery (Skeen et al., 2003). B: The laminectomy in this dog is not as wide but still gave enough access to suture the dura accurately.
A
B
14.8 A: A wide dural fenestration has been made after hemilaminectomy at C2/3. The dog was much improved 24 months after surgery (Skeen et al., 2003). B: Radiograph taken 10 min after the images shown in 14.4. Contrast has flowed from the previous point of obstruction (arrow) into the remainder of the subarachnoid space. CSF flow dynamics have been studied in one dog with an arachnoid cyst and showed normal communication with the subarachnoid space (Moissonier et al., 2002).
325
326
Small Animal Spinal Disorders
potential complication of not closing a dural incision is that the spinal cord could herniate through the defect; this has only been reported in humans (Osterholm, 1974; Henry et al., 1997; Watters et al., 1998).
METABOLIC Lysosomal storage diseases Lysosomal storage diseases occur where there is a defect of metabolism caused by dysfunction in a specific enzyme pathway. They are relatively common disorders in dogs and cats and frequently cause neurological signs, which are usually seen from early in life and are progressive. Most cause signs of intracranial disease but cats with mucopolysaccharidosis type VI can present with paraparesis. Neurological deficits are caused by vertebral exostoses that can resemble cartilaginous exostosis and hypervitaminosis A (Haskins et al., 1980, 1983). Mucopolysaccharidosis type IIIA in dogs can present as an adult onset pelvic limb ataxia or hypermetric gait (Fischer et al., 1998; Jolly et al., 2000). Cytoplasmic inclusions may be detected in hepatocytes but definitive diagnosis requires specialized techniques. There is no treatment and the prognosis is poor.
Osteoporosis This is a common problem in humans, the incidence of which increases with age as well as in women after menopause or ovariectomy (Riggs, 2002). Other risk factors include chronic renal disease, type I diabetes, corticosteroid use and Cushing’s disease (Khanine et al., 2000; Vestergaard et al., 2002). Vertebral fractures are a common sequel to osteoporosis in humans (Khanine et al., 2000); diagnosis is made by radiography, scintigraphy, CT or MRI (Cook et al., 2002; Tan et al., 2002). Even though vertebral fracture due to osteoporosis is a common problem in humans it is nevertheless underdiagnosed (Gehlbach et al., 2000). Osteoporosis has been produced experimentally in canine lumbar vertebrae as a result of ovariectomy or steroid administration (Norrdin et al., 1990; Yamaura et al., 1993); it has also been recognized as a clinical problem in dogs with Cushing’s disease as well as in routine post-mortem studies (Pellegrini et al., 1979; Huntley et al., 1982; Schleithoff, 1984). Vertebral fracture secondary to osteoporosis is therefore likely to be an under-recognized cause of spinal pain in dogs, especially in older, neutered animals with Cushing’s disease.
14.9 Hypervitaminosis A in a cat showing massive vertebral exostoses.
parts of the spine and the limbs may also be involved (Goldman, 1992). Clinical signs related to nerve root and spinal cord compression are seen. Neck pain and rigidity, ataxia, paresis of the thoracic limbs, and lameness may occur. The diagnosis is suspected in a cat fed exclusively on liver, and is confirmed by radiography (14.9). Treatment is difficult. Stopping vitamin A intake can arrest the development of further exostoses, and antiinflammatory drugs may relieve clinical signs.
IDIOPATHIC Tumoral calcinosis Also known as calcinosis circumscripta, this condition has been described as a cause of compressive spinal cord dysfunction in young dogs. The most common site is between C1 and C2 but it has also been reported in the thoracolumbar region (McEwan et al., 1992; de Risio and Olby, 2000). Diagnosis is by radiography; a mineralized mass is visible at the site of cord compression (14.10). The cause is not known. Surgical decompression may be successful.
Disseminated idiopathic skeletal hyperostosis (DISH) This is thought to be an exaggerated proliferation of bone as a result of minor stresses, which results in extensive ossification throughout the body. A characteristic feature in the spine is a ‘flowing pattern’ of ventrolateral new bone formation that extends over at least four adjacent vertebrae. Clinical signs usually just reflect the mechanical limitations to movement (Morgan and Stavenborn, 1991; Dewey et al., 2003).
NUTRITIONAL Hypervitaminosis A
INFECTIOUS/INFLAMMATORY Discospondylitis
Cats fed a diet with excessive vitamin A may suffer from a skeletal condition characterized by severe exostosis of the vertebrae, particularly in the cervical spine. Other
Discospondylitis is an inflammatory condition centered on the intervertebral disc and involving the vertebral end plates and adjacent bone of the vertebral body. Large
Miscellaneous conditions
breeds of dogs are affected most often. The condition is rare in cats (Malik et al., 1990; Watson and Roberts, 1993). The disease is usually caused by Staphylococcus intermedius but other organisms including Streptococcus sp., Escherichia coli and Brucella canis are also found. Fungal infections are uncommon but should be considered particularly in German shepherd dogs; hyphae may be detected in urine sediment (Butterworth et al., 1995; Watt et al., 1995; Greene, 1998). The bacteria usually gain access to the disc by hematogenous spread from other foci in the body, of which the bladder is most common. Occasional cases associated with foreign body migration are seen, usually grass seeds (awns). Iatrogenic cases following surgery are also recognized, especially when an implant has been used (13.34, 14.11, 14.14). The first clinical sign is usually pain, with concurrent or later development of neurological deficits. Certain sites are predisposed: lumbosacral, caudal cervical, thoracolumbar and mid-thoracic vertebrae. Multiple discs may be involved, either adjacent or distant to the disc
14.10 Tumoral calcinosis in a young Labrador retriever with progressive paraparesis. The myelogram reveals a large, mineralized epidural mass located dorsally between C1 and C2. Following surgical removal of the mass the dog returned to normal within a few weeks (Lewis and Kelly, 1990).
first identified. Systemic illness is common, typically with pyrexia, lethargy, inappetence, and dysuria related to cystitis. Diagnosis is by radiography (14.11, 14.12). Generally the radiographic changes are seen clearly, but occasionally the radiographs will be normal even though infection is present (occult discospondylitis). When the radiographs are inconclusive, a bone scan may reveal the lesion (Stefanacci and Wheeler, 1991) (see page 59). Blood or urine cultures are positive in about two thirds of patients (Kornegay, 1986). Brucella canis titers should be analyzed in endemic areas. Treatment of bacterial discospondylitis is initially by use of antibiotics. In view of the predominance of S. intermedius, clindamycin, cepazolin or cloxacillin are the best initial choices. Ampicillin and amoxycillin are not useful because the organism is often resistant to these drugs through the mechanism of -lactamase production. Further treatment is governed by clinical progression and results of bacteriological testing. Treatment must be continued for at least 6 weeks, even when the response is good. Most dogs respond well to antibiotics alone (Gilmore, 1987), or combined with surgical curettage (Kornegay and Barber, 1980). It is not known how many dogs suffer a recurrence of signs. Treatment for B. canis is with minocycline (25 mg/kg PO q24h for 2 weeks) and streptomycin (5 mg/kg IM or SQ q12h for 1 week) or gentamycin (2 mg/kg IM or SQ q12 h for 1 week) (Greene, 1998). There is potential for zoonotic spread, and recurrence is common. It may be wise to castrate male dogs with B. canis infection as the testes can act as a reservoir of infection (Kerwin et al., 1992). Surgical curettage of solitary, readily accessible lesions often leads to rapid resolution of signs (Kornegay and Barber, 1980). It also provides material for culture and promotes blood supply to the affected disc. Surgical intervention should always be considered in patients with an unsatisfactory response to antibiotics. The main risk is that surgery could destabilize the 14.11 Ten-year-old Doberman with S. intermedius discospondylitis at C6/7. It had undergone total hip replacement 9 months previously, which became infected and was removed after 4 months. A: Destructive lesion at C6/7. There is sclerosis and destruction of C6 and C7 end plates (arrowhead). B: CT scan through caudal C6 confirms bony destruction within the vertebral body (arrowhead).
A
B
327
328
Small Animal Spinal Disorders
14.12 Rottweiler with uncontrollable pain due to discospondylitis at L7/S1 that did not respond to 1 week of antibiotics and opioid analgesics. An external fixator was applied and the dog was pain-free the next day. It was normal 3 months later although L7/S1 had not fused; no bone graft had been used at surgery.
A
B
A
B
14.13 A: The dog shown in 14.12 at 6-week follow-up. B: Dorsal view of the external fixator (SK External Fixation System, Imex Veterinary Inc., Longview, TX). A potential limiting factor in large dogs is the length of pin; these were only just long enough to attach the clamps.
affected intervertebral space further; loss of stability can also occur spontaneously in animals with advanced bony destruction. Subluxation or collapse of the intervertebral space then results together with severe pain and neurological deficits. When collapse occurs in the lumbosacral area the animal often develops intractable pain (14.12). Fixation of the spine is then best provided by an external device that can be removed once the infection has resolved and the site has stabilized (Auger et al., 2000; Walker et al., 2002). In the lumbosacral area threaded pins or screws may be driven through the articular facets into the sacrum (McKee et al., 1990). Pins or bone cement have also been used successfully in the face of infection (Lavely et al., 2002). An autogenous bone graft placed into the affected interspace probably enhances complete fusion (Auger et al., 2000) although this does not seem to be crucial for a successful outcome (McKee et al., 1990) (14.13). Decompression of discospondylitis lesions by
laminectomy without stabilization is not recommended as it may lead to marked instability. In addition to instability and collapse of the disc space, another serious complication of discospondylitis is the development of epidural abscessation (see below) (Lobetti, 1994; Remedios et al., 1996). Fungal discospondylitis is a systemic infection that occurs in German shepherd dogs and occasionally in other breeds of dog; this disorder is not restricted to tropical or semi-tropical regions as was initially thought. It is caused usually by one of several Aspergillus species and has a poor prognosis. Diagnosis can often be obtained by identifying fungal hyphae in urine sediment. Itraconazole may control the disease but the long-term prognosis is poor (Butterworth et al., 1995; Watt et al., 1995).
Epidural empyema This is an emergency situation that requires rapid diagnosis and treatment (14.14). Clinical signs include fever,
Miscellaneous conditions
A
B
14.14 A: Doberman that underwent a ventral slot and distraction stabilization using screws and bone cement for a dynamic lesion at C5/6. The dog did well for 4 days but then became febrile and suffered progressive neurological deterioration. B: A repeat myelogram 2 weeks after surgery revealed small erosions of the C6/7 end plates and a diffuse epidural mass extending from the mid-body of C5 vertebra to mid-C7 (arrowheads). The dog died; epidural abscess was found at necropsy.
lethargy, pain and neurological deficits along with peripheral neutrophilia and a neutrophilic CSF pleocytosis. Less than half of dogs described have had concomitant discospondylitis. Myelography can provide an accurate diagnosis but is no longer recommended in humans if MRI is available because the latter permits earlier identification of lesions (Angtuaco et al., 1987; Reihsaus et al., 2000). Lumbar puncture is also no longer recommended in humans as it carries the risk of introducing infection into the subarachnoid space. Surgical decompression and drainage together with antibiotics are the treatment of choice in humans. The biggest problem is one of early diagnosis before massive neurological damage has occurred (Reihsaus et al., 2000). Results in dogs have been mixed. One study reported a good outcome in five of seven dogs whereas others report a poor outcome even after aggressive therapy (Dewey et al., 1998; Lavely et al., 2002).
Inflammatory CNS diseases STEROID-RESPONSIVE MENINGITIS— ARTERITIS Several aseptic meningitis syndromes have been described in dogs. Meningitis and polyarteritis has been described in a colony of young research Beagles; Bernese mountain dogs also suffer from a similar syndrome (Meric, 1988). These disorders are similar to necrotizing vasculitis of the meningeal arteries (Meric et al., 1986). The clinical signs are typical of meningitis, with depression, cervical pain and stiff gait; pyrexia is also seen in this condition. The disease may be acute or it can have a relapsing pattern. Analysis of CSF reveals marked pleocytosis (with CSF white blood cell counts in the hundreds or
even thousands per microliter), mainly comprising neutrophils along with an increase in protein concentration. Infectious agents are not seen in the CSF and culture is negative; an immune-mediated mechanism is suspected. The CSF may be relatively normal between bouts of the disease. Treatment with long-term corticosteroids is recommended (prednisolone 2–4 mg/kg/day initially, reducing to anti-inflammatory doses) until the clinical signs and CSF pleocytosis resolve. The prognosis is fair, although some dogs experience relapses. Older dogs with high IgA levels in their CSF require a longer duration of corticosteroid therapy and have a less favorable prognosis (Tipold and Jaggy, 1994; Cizinauskas et al., 2000).
CANINE DISTEMPER VIRUS INFECTION Canine distemper virus (CDV) infection is the most common infectious cause of neurological disease in dogs. Demyelination and inflammation occur in various sites throughout the CNS. The virulence of the virus strain and the immunocompetence of the dog are important factors in determining the severity of disease. Other CNS infections may be seen in association with the immunosuppression related to CDV. Systemic signs of disease may occur, although this is not a consistent feature (Tipold et al., 1992; Thomas et al., 1993). The neurological signs may be multifocal although in one third of cases they suggest a focal lesion. Signs may be acute in onset. CSF analysis usually reveals mild to moderate, lymphocytic pleocytosis (see page 45). Diagnostic methods include testing for local production of antibody in the CSF; fluorescent antibody testing of conjunctival cells, CSF leukocytes, or skin biopsy; or for viral RNA by
329
330
Small Animal Spinal Disorders
polymerase chain reaction (Tipold et al., 1992; Thomas et al., 1993; Haines et al., 1999; Sharp et al., 2000). Confirmation of a diagnosis of CDV infection is difficult ante mortem. Skin biopsy is easy and may be the most accurate method (Haines et al., 1999). Treatment is restricted to managing the clinical signs and providing supportive care. Prognosis is guarded.
FELINE INFECTIOUS PERITONITIS VIRUS INFECTION (FIP) Neurological signs, including spinal cord syndromes, may be seen in cats with FIP virus infection. Multisystemic signs, particularly ocular involvement, are seen in most affected cats. Neurological signs are most often associated with the dry form of the disease. CSF is usually abnormal with increased protein and a mixed, predominantly neutrophilic pleocytosis (Foley et al., 1998) (see page 45). There is no definitive treatment and the prognosis is poor (Kline et al., 1994).
OTHER INFECTIOUS AGENTS
GRANULOMATOUS MENINGOENCEPHALOMYELITIS (GME) This is an inflammatory disease of unknown etiology; any part of the CNS may be involved and the lesions can either be focal or diffuse in nature. The retina and optic nerves may also be affected. Perivascular accumulations of mononuclear cells are present throughout the CNS,
14.15 T2-weighted MRI of a 9-year-old Boxer dog with an extensive high signal area (arrow) in the cervical spinal cord. Histopathological diagnosis was GME.
which can lead to the formation of large granulomas (Cordy, 1979). Clinical signs are typical of meningoencephalomyelitis; onset is usually between 3 and 7 years of age. Spinal cord syndromes may be seen alone or as part of a multifocal presentation. The signs of spinal disease may be restricted to spinal hyperesthesia or may include neurological deficits. The course is usually chronic, but some cases show a rapid decline. Analysis of CSF reveals moderate, mainly mononuclear pleocytosis with increased protein, but the findings are non-specific. Imaging in dogs with spinal cord involvement may reveal an intramedullary lesion (14.15). Treatment with immunosuppressive doses of corticosteroids may lead to improvement (prednisolone 1–2 mg/kg/day) (Munana and Luttgen, 1998). The longterm prognosis used to be poor, although excellent longterm results have now been obtained after treatment with cytosine arabinose (Cuddon et al., 2002).
Meningomyelitis due to other agents is relatively rare in dogs and cats; infections are usually associated with brain involvement and therefore present with multifocal neurological signs. Many organisms have been implicated, including bacteria, rickettsial, fungal and helminth organisms; the distribution of some of these is regional (Munana, 1996; Stiles, 2000; Gasser et al., 2001). Some of the organisms that may affect the spinal cord in dogs and cats are listed in Table 14.1. The signs are typical of inflammatory CNS disease as described above. Some infections show mainly signs of intracranial disease, typically rickettsial infections when seizures, dullness, depression, and vestibular signs occur. Pelvic limb hyperextension is often seen in young dogs with toxoplasmosis or neosporosis (Nesbit et al., 1981). Systemic signs of infection may be apparent, for example gastrointestinal disturbance or respiratory signs, which may indicate the portal of entry into the CNS. Patients may be pyrexic, but this is variable. Confirmation of the diagnosis is usually by CSF analysis. Pleocytosis is variable, but very high cell counts (in the thousands per microliter) may be seen, mostly
Table 14.1 Known organisms associated with spinal cord disease Viruses
Bacteria
Rickettsia
Fungi
Protozoa
Helminth
Canine distemper Feline infectious peritonitis Rabies
Staphylococcus sp. Brucella canis
Rickettsia sp. Ehrlichia sp. Bartonella sp.
Cryptococcus neoformans Coccidioides imitis Blastomycoides dermatidis Histoplasma capsulatum
Toxoplasma gondii Baylisascaris sp. Neospora caninum
Miscellaneous conditions
composed of neutrophils; eosinophils may also be present. Organisms may be seen in the CSF. Culture may be attempted but is often unrewarding. Serological testing may be useful in some infections, for example in neosporosis, rickettsial disease or cryptococcosis. Those rare patients where bacterial infections are identified should be treated with antibiotics. The bacterial sensitivity and CNS penetration of the drug must be considered in choosing an antibiotic (Greene, 1998). Fungal infections are difficult to treat, but amphotericin B, itraconazole and fluconazole may be effective. Rickettsial infections are treated with tetracyclines, preferably doxycycline or chloramphenicol. Protozoal infections may be treated with clindamycin, azithromycin or trimethoprimsulfonamide combined with pyrimethamine (BoschDriessen et al., 2002). The use of corticosteroids in meningomyelitis is controversial. They are contraindicated in fungal infection, but may have a role in acute bacterial infections of the CNS (Vandecasteele et al., 2001; Gijwani et al., 2002). The potential consequences of corticosteroid use must be considered before their administration. Their use at the same time as antibiotics can also make it difficult to determine which is responsible for an improvement in the animal’s condition.
Epidural steatitis This is a common complication to epidural empyema but could also occur secondary to infection with Brucella, Bartonella or Mycobacterium sp. (Postacchini and Montanaro, 1980; Nas et al., 2001). Epidural fibrosis and fat necrosis are also seen secondary to vascular compromise in lumbosacral disease (see page 190).
TRAUMA The most common disorders are discussed on pages 30 and 283. Brachial plexus injuries are a common
consequence of trauma and it is imperative that they be recognized on admission of the patient (2.15, 2.26, 2.28, 13.1, 14.16B). Re-implantation of avulsed ventral nerve rootlets into the cervical spinal cord has recently been proposed as a new treatment for avulsion injuries in dogs and cats (Moissonnier et al., 1996, 2001). Although an effective treatment is needed desperately, surgical repair is unlikely to be useful for most animals with plexal injuries due to the difficulty in producing an early return of function to the distal limb and the likelihood of residual sensory deficits (Rodkey and Sharp, 2003). Trauma can occasionally produce neurological deficits by affecting the blood supply to the spinal cord, peripheral nerves or muscles. Acute pelvic limb paralysis is described in cats after severe abdominal trauma. It is thought to be a vasospasm or thrombosis of radicular blood vessels (Summers et al., 1994). Paralysis of a thoracic limb has been described after injury to the subclavian artery in a dog (MacCoy and Trotter, 1977). Gunshot injuries may involve the vertebral column or spinal cord and are usually devastating (Fullington and Otto, 1997).
Psoas muscle injury This unusual condition can cause acute onset of pelvic limb lameness with severe pain on extension of the hip and palpation of the psoas muscles along with reluctance to stand (Krawiecki and Puignero, 1995; Breur and Blevins, 1996). If inflammation and hemorrhage is severe there may also be unilateral femoral nerve deficits (N. Olby, personal communication). The clinical signs are caused by an injury or a hematoma within the psoas muscle; femoral neuropathy is due to compression within the muscle compartment (Robinson et al., 2001). Diagnosis is made by ultrasound or MRI. Most dogs improve with rest but a tenomyectomy may be necessary to resolve the pain.
14.16 A: Post-mortem appearance of a brachial plexus avulsion. Note the absent nerve roots and rootlets on the side with subdural hemorrhage. B: This dog was paraplegic due to an L1 vertebral fracture; its brachial plexus avulsion was almost overlooked before repairing the vertebral fracture.
A
B
331
332
Small Animal Spinal Disorders
VASCULAR Fibrinocartilaginous embolism (ischemic myelopathy) Fibrocartilaginous embolism (FCE) is a syndrome of acute, severe neurological dysfunction of dogs. It is an important differential diagnosis in cases of disc disease, trauma, and other acute spinal conditions (Cauzinille and Kornegay, 1996). The emboli are composed of fibrocartilage, identical to the material within the nucleus pulposus. In pathological studies of FCE, emboli are found in the vasculature of the spinal cord substance or nerve roots. The exact mechanism by which they gain access to these areas is not clear. Adult dogs of large and giant breeds are affected most often along with Miniature schnauzers, but this condition can affect any dog. FCE can occur in dogs as early as 8 weeks of age, although it is seen most often in adults between 3 and 7 years (Cauzinille and Kornegay, 1996; Junker et al., 2000; Hawthorne et al., 2001). The condition appears to be very rare in cats (Scott and O’Leary, 1996; Abramson et al., 2002). Peracute, severe neurological presentations occur, often following vigorous exercise or mild trauma. Owners may note a progression of the signs over a period of several hours, often from an initial lameness to eventual paralysis. However, progression beyond the first 24 h is very unusual and this helps to differentiate the condition from other myelopathies. Spinal hyperesthesia is not usually present on clinical examination, but sometimes severe discomfort is apparent during the development of the condition and this can persist for several hours after onset (Cauzinille and Kornegay, 1996). Any part of the spinal cord may be affected, and often the signs are markedly asymmetrical.
A
Confirmation of the diagnosis is by elimination of other causes. Myelography or MRI should be performed to rule out compressive spinal cord lesions (14.17). Generally, the myelogram is normal, but in some cases of FCE, it will show an intramedullary pattern of spinal cord swelling. MRI reveals spinal cord edema and swelling in the acute stage. Changes in CSF are often non-specific although xanthochromia is very suggestive of FCE (4.1). Treatment is by supportive care of the patient. Use of methylprednisolone sodium succinate (MPSS), as in spinal trauma, may be useful in the first few hours following onset (see page 83). Other corticosteroids are not indicated. The prognosis is variable and dependent largely on the size of the area that undergoes necrosis. Discrete upper motor neuron (UMN) lesions often improve, but loss of nociception or extensive LMN deficits carry a poor prognosis (Cauzinille and Kornegay, 1996). Dogs that recover may be left with residual neurological deficits unless the lesion is small (Hawthorne et al., 2001).
Ascending myelomalacia This is probably an ischemic lesion induced by the diffuse spread of extruded disc material; it is illustrated in 8.5 and 14.18 and discussed on page 128. Extensive malacia may also be evident following trauma (13.10).
Ischemic neuromyopathy (aortic embolism, iliac thrombosis) Ischemic neuromyopathy is a common cause of acute paraplegia in cats. Involvement of either one pelvic limb,
B
14.17 10-year-old cat that was normal prior to an acute onset of symmetrical tetraparesis localizing to the C6–T2 spinal cord. T2weighted MRIs show high signal within the spinal cord (arrows) that extends A from the level of C2 to C7 vertebral bodies and that B occupies much of the cross-sectional area of the spinal cord. Histological diagnosis was FCE.
Miscellaneous conditions
or of one thoracic limb, has been reported in 13 and 14% of cats, respectively (Smith et al., 2003). Hypothermia, tachypnea and signs of cardiac disease are common in affected cats and renal or gastrointestinal dysfunction may also be apparent. Affected limbs are usually arreflexic, hypothermic and hypalgesic; femoral pulses are often absent and the gastrocnemius muscles swollen and painful in pelvic limb involvement. The nail beds are characteristically cyanotic and the toes do not bleed after needle prick. Diagnosis is based on the clinical signs (Griffiths and Duncan, 1979; Laste and Harpster, 1995). Pre-existing cardiomyopathy underlies the thromboembolic episode, but the presence of a thrombus does not entirely explain the clinical signs. There also appears to be a failure of collateral circulation caused by release of vasoactive substances from the area of the thrombus.
The prognosis for long-term survival is guarded. Rectal temperature of 99°F (37.22°C) or higher was the best predictor of survival to discharge; 76% of 39 cats with a temperature of 99°F (37.22°C) or more survived to be discharged (Smith et al., 2003). Median long-term survival of 44 cats was 117 days. Recurrence occurred in 11 cats (25%). Cats with congestive heart failure at presentation had significantly shorter survival times than cats without heart failure (77 days vs 223 days). Low-dose aspirin therapy is recommended as an inexpensive, safe treatment that is at least as effective as high-dose aspirin or warfarin (Smith et al., 2003). This condition also occurs in dogs although it is less common than in cats. The presenting signs are also distinct from those in cats; the onset tends to be gradual and most dogs do not appear to be in discomfort. Hyperadrenocorticism, neoplasia and cardiac disease are predisposing factors (van Winkle et al., 1993). The prognosis depends on the underlying cause; the prognosis for dogs surviving the acute episode is favorable (Boswood et al., 2000).
Spinal cord hemorrhage, hematoma
14.18 Typical myelographic appearance of a dog with myelomalacia secondary to intervertebral disc extrusion. There is infiltration of contrast material into the parenchyma of the spinal cord (arrow) (Lu et al., 2002).
A
B
Hemorrhage and hematoma formation may occur within the spinal cord, in the subarachnoid space or in the epidural space. Epidural hemorrhage is a common complication of acute thoracolumbar disc extrusion (Olby et al., 2000). Hemorrhage can also cause marked deterioration after ventral slot surgery for cervical disc extrusion (Seim and Prata, 1982). Bleeding can occur with various coagulopathies such as hemophilia, rickettsial disease, trauma, anticoagulant toxicosis, or as a complication to any severe bleeding disorder (Stokol et al., 1994). If spinal hemorrhage is suspected in the absence of trauma, tests for clotting function should be performed. Analysis of CSF may reveal xanthochromia (4.1) or erythrophagocytosis.
C
14.19 T2-weighted MRI of a 9-month-old dog with neck pain and tetraplegia for one day caused by a hematoma (arrowheads). Platelet numbers, coagulation profile and a buccal mucosal bleeding time were normal. The dog failed to improve and was euthanized after 3 weeks; necropsy showed that the hematoma was subarachnoid; no cause was identified. A: Sagittal image. The dotted line shows the level of the transverse image. B: Transverse image. For MRI of blood, see page 58. C: Necropsy.
333
334
Small Animal Spinal Disorders
If fresh blood is seen in a CSF sample, it is most likely to be caused by puncture of a dural vessel rather than reflecting an underlying blood disease. Intramedullary hematoma formation is an unusual cause of tetraparesis in young, large-breed dogs (Martin et al., 1986) (14.19). The cause is not known; there may or may not be an associated coagulopathy. Neck pain is a variable finding. Surgical decompression is indicated in compressive lesions not associated with clotting disorders. Any underlying coagulopathy must also be addressed.
Intermittent claudication This is recognized in both humans and dogs with lesions in the lumbar and lumbosacral regions (Tarvin and Prata, 1980; Markwalder, 1993; Porter, 1996). It is characterized by circulatory impairment of neural structures that occurs secondary to exercise. Intermittent claudication has also been recognized in humans with cervical or thoracic spinal cord compression where it can complicate cervical spondylotic myelopathy (Kikuchi et al., 1996).
Key issues for future investigation 1. Is decompression indicated for severe vertebral deformities and if so, should it be performed at the same time as stabilization or at a second surgery? 2. What is the long-term recurrence rate following medical treatment for discospondylitis? Is this influenced by the duration of antibiotic therapy? 3. Is MPSS of any value in FCE? 4. Is any treatment regime useful in degenerative myelopathy?
REFERENCES Abramson, C.J., Platt, S.R., Stedman, N.L. (2002) Tetraparesis in a cat with fibrocartilaginous emboli. Journal of the American Animal Hospital Association 38, 153–156. Aikawa, T. (2001) Surgical management of vertebral deformity in four dogs. Journal of the American Animal Hospital Association (Japan) 16, 39–44. Angtuaco, E.J., McConnell, J.R., Chadduck, W.M., Flanigan, S. (1987) MR imaging of spinal epidural sepsis. American Journal of Roentgenology 149, 1249–1253. Auger, J., Dupuis, J., Quesnel, A., Beauregard, G. (2000) Surgical treatment of lumbosacral instability caused by discospondylitis in four dogs. Veterinary Surgery 29, 70–80. Averill, D.R., Jr (1973) Degenerative myelopathy in the aging German Shepherd dog: clinical and pathologic findings. Journal of the American Veterinary Medical Association 162, 1045–1051. Bagley, R.S., Silver, G.M., Seguin, B., Lincoln, J.D., Britt, L.G. (1997) Scoliosis and associated cystic spinal cord lesion in a dog. Journal of the American Veterinary Medical Association 211, 573–575.
Bailey, C.S. (1975) An embryological approach to the clinical significance of congenital vertebral and spinal cord abnormalities. Journal of the American Animal Hospital Association 11, 426–434. Bailey, C.S., Morgan, J.P. (1992) Congenital spinal malformations. Veterinary Clinics of North America, Small Animal Practice 22, 985–1015. Bosch-Driessen, L.H., Verbraak, F.D., Suttorp-Schulten, M.S., van Ruyven, R.L., Klok, A.M., Hoyng, C.B., Rothova, A. (2002) A prospective, randomized trial of pyrimethamine and azithromycin vs pyrimethamine and sulfadiazine for the treatment of ocular toxoplasmosis. American Journal of Ophthalmology 134, 34–40. Boswood, A., Lamb, C.R., White, R.N. (2000) Aortic and iliac thrombosis in six dogs. Journal of Small Animal Practice 41, 109–114. Braund, K.G., Sharp, N.J.H. (2003) Neurological examination and localization. In: D. Slatter (ed.) Textbook of Small Animal Surgery, 3rd edn, 1092–1107. Elsevier Science, Philadelphia. Breur, G.J., Blevins, W.E. (1996) Traumatic injury of the iliopsoas muscle in 25 dogs. Veterinary Surgery 25, 262. Broek, A.H.M.v.d., Else, R.W., Abercromby, R., France, M. (1991) Spinal dysraphism in the Weimaraner. Journal of Small Animal Practice 32, 258–260. Butterworth, S.J., Barr, F.J., Pearson, G.R., Day, M.J. (1995) Multiple discospondylitis associated with Aspergillus species infection in a dog. Veterinary Record 136, 38–41. Cauzinille, L., Kornegay, J.N. (1992) Acquired syringomyelia in a dog. Journal of the American Veterinary Medical Association 201, 1225–1228. Cauzinille, L., Kornegay, N. (1996) Fibrocartilaginous embolism of the spinal cord in dogs: review of 36 histologically confirmed cases and retrospective study of 26 suspected cases. Journal of Veterinary Internal Medicine 10, 241–245. Charest, D.R., Kenny, B.G. (2000) Radicular pain caused by synovial cyst: an underdiagnosed entity in the elderly? Journal of Neurosurgery 92 (1 Suppl), 57–60. Child, G., Higgins, R.J., Cuddon, P.A. (1986) Acquired scoliosis associated with hydromyelia and syringomyelia in two dogs. Journal of the American Veterinary Medical Association 189, 909–912. Chrisman, C.L. (1992) Neurological diseases of Rottweilers: neuroaxonal dystrophy and leukoencephalomalacia. Journal of Small Animal Practice 33, 500–504. Cizinauskas, S., Jaggy, A., Tipold, A. (2000) Long-term treatment of dogs with steroid-responsive meningitis-arteritis: clinical, laboratory and therapeutic results. Journal of Small Animal Practice 41, 295–301. Cook, G.J., Hannaford, E., See, M., Clarke, S.E., Fogelman, I. (2002) The value of bone scintigraphy in the evaluation of osteoporotic patients with back pain. Scandinavian Journal of Rheumatology 31, 245–248. Cordy, D.R. (1979) Canine granulomatous meningoencephalomyelitis. Veterinary Pathology 16, 325–333. Cuddon, P.A., Coates, J.R., Murray, M. (2002) New Treatments for Granulomatous Meningoencephalomyelitis. Proceedings of the ACVIM Veterinary Forum 2002; 319–321. American College of Veterinary Internal Medicine, Dallas, TX. de Risio, L., Olby, N.J. (2000) Tumoral calcinosis of the thoracic spine: Veterinary Neurology Neurosurgery. http://www.neurovet.org/ DeRisioandOlby1/Tumoralcalcinosis.htm Dewey, C.W., Hoffman, A., Coates, J.R. (2003) Miscellaneous spinal conditions. In: D. Slatter (ed.) Textbook of Small Animal Surgery, 3rd edn, 1148–1162. Philadelphia: Elsevier Science. Dewey, C.W., Kortz, G.D., Bailey, C.S. (1998) Spinal epidural empyema in two dogs. Journal of the American Animal Hospital Association 34, 305–308. Dickinson, P.J., Sturges, B.K., Berry, W.L., Vernau, K.M., Koblik, P.D., LeCouteur, R.A. (2001) Extradural spinal synovial cysts in nine dogs. Journal of Small Animal Practice 42, 502–509. Dyce, J., Herrtage, M.E., Houlton, J.E.F., Palmer, A.C. (1991) Canine spinal ‘arachnoid cysts’. Journal of Small Animal Practice 32, 433–437. Fingeroth, J.M., Johnson, G.C., Burt, J.K., Fenner, W.R., Cain, L.S. (1989) Neuroradiographic diagnosis and surgical repair of tethered cord syndrome in an English Bulldog with spina bifida and myeloschisis. Journal of the American Veterinary Medical Association 194, 1300–1302. Fischbein, N.J., Dillon, W.P., Cobbs, C., Weinstein, P.R. (1999) The ‘presyrinx’ state: a reversible myelopathic condition that may precede syringomyelia. American Journal of Neuroradiology 20, 7–20.
Miscellaneous conditions
Fischer, A., Carmichael, K.P., Munnell, J.F., Jhabvala, P., Thompson, J.N., Matalon, R., Jezyk, P.F., Wang, P., Giger, U. (1998) Sulfamidase deficiency in a family of Dachshunds: a canine model of mucopolysaccharidosis IIIA (Sanfilippo A). Pediatric Research 44, 74–82. Foley, J.E., Lapointe, J.M., Koblik, P., Poland, A., Pedersen, N.C. (1998) Diagnostic features of clinical neurologic feline infectious peritonitis. Journal of Veterinary Internal Medicine 12, 415–423. Frykman, O.F. (1999) Spinal arachnoid cyst in four dogs: diagnosis, surgical treatment and follow-up results. Journal of Small Animal Practice 40, 544–549. Fullington, R.J., Otto, C.M. (1997) Characteristics and management of gunshot wounds in dogs and cats: 84 cases (1986–1995). Journal of the American Veterinary Medical Association 210, 658–662. Galloway, A.M., Curtis, N.C., Sommerlad, S.F., Watt, P.R. (1999) Correlative imaging findings in seven dogs and one cat with spinal arachnoid cysts. Veterinary Radiology and Ultrasound 40, 445–452. Gambardella, P.C., Osborne, C.A., Stevens, J.B. (1975) Multiple cartilaginous exostoses in the dog. Journal of the American Veterinary Medical Association 166, 761–768. Gasser, A.M., Birkenheuer, A.J., Breitschwerdt, E.B. (2001) Canine rocky mountain spotted fever: a retrospective study of 30 cases. Journal of the American Animal Hospital Association 37, 41–48. Gehlbach, S.H., Bigelow, C., Heimisdottir, M., May, S., Walker, M., Kirkwood, J.R. (2000) Recognition of vertebral fracture in a clinical setting. Osteoporosis International 11, 577–582. Gijwani, D., Kumhar, M.R., Singh, V.B., Chadda, V.S., Soni, P.K., Nayak, K.C., Gupta, B.K. (2002) Dexamethasone therapy for bacterial meningitis in adults: a double blind placebo control study. Neurology India 50, 63–67. Gilmore, D.R. (1987) Lumbosacral diskospondylitis in 21 dogs. Journal of the American Animal Hospital Association 23, 57–61. Goldman, A.L. (1992) Hypervitaminosis A in a cat. Journal of the American Veterinary Medical Association 200, 1970–1972. Greene, C.E. (1998) Infectious Diseases of the Dog and Cat, 2nd edn. Philadelphia: WB Saunders. Griffiths, I.R., Duncan, I.D. (1975) Chronic degenerative radiculomyelopathy in the dog. Journal of Small Animal Practice 16, 461–471. Griffiths, I.R., Duncan, I.D. (1979) Ischaemic neuromyopathy in cats. Veterinary Record 104, 518–522. Haines, D.M., Martin, K.M., Chelack, B.J., Sargent, R.A., Outerbridge, C.A., Clark, E.G. (1999) Immunohistochemical detection of canine distemper virus in haired skin, nasal mucosa, and footpad epithelium: a method for antemortem diagnosis of infection. Journal of Veterinary Diagnostic Investigation 11, 396–399. Haskins, M.E., Aguirre, G.D., Jezyk, P.F., Patterson, D.F. (1980) The pathology of the feline model of mucopolysaccharidosis VI. American Journal of Pathology 101, 657–674. Haskins, M.E., Bingel, S.A., Northington, J.W., Newton, C.D., Sande, R.D., Jezyk, P.F., Patterson, D.F. (1983) Spinal cord compression and hindlimb paresis in cats with mucopolysaccharidosis VI. Journal of the American Veterinary Medical Association 182, 983–985. Hawthorne, J.C., Wallace, L.J., Fenner, W.R., Waters, D.J. (2001) Fibrocartilaginous embolic myelopathy in miniature schnauzers. Journal of the American Animal Hospital Association 37, 374–383. Henry, A., Tunkel, R., Arbit, E., Ku, A., Lachmann, E. (1997) Tethered thoracic cord resulting from spinal cord herniation. Archives of Physical Medicine and Rehabilitation 78, 530–583. Hodge, J.C., Bessette, B. (1999) The incidence of sacroiliac joint disease in patients with low–back pain. Canadian Association of Radiologists Journal 50, 321–323. Howington, J.U., Connolly, E.S., Voorhies, R.M. (1999) Intraspinal synovial cysts: 10-year experience at the Ochsner Clinic. Journal of Neurosurgery 91 (2 Suppl), 193–199. Huntley, K., Frazer, J., Gibbs, C., Gaskell, C.J. (1982) The radiological features of canine Cushing’s syndrome: a review of forty-eight cases. Journal of Small Animal Practice 23, 369–380. Huttmann, S., Krauss, J., Collmann, H., Sorensen, N., Roosen, K. (2001) Surgical management of tethered spinal cord in adults: report of 54 cases. Journal of Neurosurgery 95 (2 Suppl), 173–178. Jacobson, L.S., Kirberger, R.M. (1996) Canine multiple cartilaginous exostoses: unusual manifestations and a review of the literature. Journal of the American Animal Hospital Association 32, 45–51.
Johnston, P.E.J., Barrie, J.A., McCulloch, M.C., Anderson, T.J., Griffiths, I.R. (2000) Central nervous system pathology in 25 dogs with chronic degenerative radiculomyelopathy. Veterinary Record 146, 629–633. Jolly, R.D., Allan, F.J., Collett, M.G., Rozaklis, T., Muller, V.J., Hopwood, J.J. (2000) Mucopolysaccharidosis IIIA (Sanfilippo syndrome) in a New Zealand Huntaway dog with ataxia. New Zealand Veterinary Journal 48, 144–148. Junker, K., van den Ingh, T., Bossard, M.M., van Nes, J.J. (2000) Fibrocartilaginous embolism of the spinal cord [FCE] in juvenile Irish Wolfhounds. Veterinary Quarterly 22, 154–156. Kerwin, S.C., Lewis, D.D., Hribernik, T.N., Partington, B., Hosgood, G., Eilts, B.E. (1992) Diskospondylitis associated with Brucella canis infection in dogs: 14 cases (1980–1991). Journal of the American Veterinary Medical Association 201, 1253–1257. Khanine, V., Fournier, J.J., Requeda, E., Luton, J.P., Simon, F., Crouzet, J. (2000) Osteoporotic fractures at presentation of Cushing’s disease: two case reports and a literature review. Joint Bone Spine 67, 341–345. Kikuchi, S., Watanabe, E., Hasue, M. (1996) Spinal intermittent claudication due to cervical and thoracic degenerative spine disease. Spine 21, 313–318. Kirberger, R.M.M., Wrigley, R.H. (1993) Myelography in the dog. Veterinary Radiology and Ultrasound 34, 253–258. Kirberger, R.M., Jacobson, L.S., Davies, J.V., Engela, J. (1997) Hydromyelia in the dog. Veterinary Radiology and Ultrasound 38, 30–38. Kline, K.L., Joseph, R.J., Averill, D.R. (1994) Feline infectious peritonitis with neurologic involvement: clinical and pathological findings in 24 cats. Journal of the American Animal Hospital Association 30, 111–118. Kornegay, J.N. (1986) Discospondylitis. In: R.W. Kirk (ed.), Current Veterinary Therapy, 810–814. Philadelphia: WB Saunders. Kornegay, J.N., Barber, D.L. (1980) Diskospondylitis in dogs. Journal of the American Veterinary Medical Association 177, 337–341. Krawiecki, J.M., Puignero, V. (1995) Avulsion of the tubercle of the psoas minor muscle in a dog. Point Veterinaire 26, 1167–1169. Larsen, J.S., Selby, L.A. (1981) Spondylosis deformans in large dogs— relative risk by breed, age and sex. Journal of the American Animal Hospital Association 17, 623–625. Laste, N.J., Harpster, N.K. (1995) A retrospective study of 100 cases of feline distal aortic thromboembolism: 1977–1993. Journal of the American Animal Hospital Association 31, 492–500. Lavely, J., Vernau, K.M., LeCouteur, R.A. (2002) Spinal epidural empyema in seven dogs. Journal of Veterinary Internal Medicine 16, 369. LeCouteur, R.A., Grandy, J.L., Ettinger, S.J., Feldman, E.C. (ed.) (2000) Diseases of the spinal cord. Textbook of Veterinary Internal Medicine: Diseases of the Dog and Cat, Vols 1 and 2. Philadelphia: WB Saunders. Levitski, R.E., Chauvet, A.E., Lipsitz, D. (1999a) Cervical myelopathy associated with extradural synovial cysts in 4 dogs. Journal of Veterinary Internal Medicine 13, 181–186. Levitski, R.E., Lipsitz, D., Chauvet, A.E. (1999b) Magnetic resonance imaging of the cervical spine in 27 dogs. Veterinary Radiology and Ultrasound 40, 332–341. Lewis, D.G., Kelly, D.F. (1990) Calcinosis circumscripta in dogs as a cause of spinal ataxia. Journal of Small Animal Practice 31, 35–37. Lipsitz, D., Levitski, R.E., Chauvet, A.E., Berry, W.L. (2001) Magnetic resonance imaging features of cervical stenotic myelopathy in 21 dogs. Veterinary Radiology and Ultrasound 42, 20–27. Lobetti, R.G. (1994) Subarachnoid abscess as a complication of discospondylitis in a dog. Journal of Small Animal Practice 35, 480–483. Lu, D., Lamb, C.R., Targett, M.P. (2002) Results of myelography in seven dogs with myelomalacia. Veterinary Radiology and Ultrasound 43, 326–330. MacCoy, D.M., Trotter, E.J. (1977) Brachial paralysis subsequent to traumatic partial occlusion of the right subclavian artery. Journal of the Americal Animal Hospital Association 13, 625–630. Malik, R., Latter, M., Love, D.N. (1990) Bacterial discospondylitis in a cat. Journal of Small Animal Practice 31, 404–406. Manchikanti, L. (1999) Facet joint pain and the role of neural blockade in its management. Current Review of Pain 3, 348–358. Markwalder, T.M. (1993) Surgical management of neurogenic claudication in 100 patients with lumbar spinal stenosis due to degenerative spondylolisthesis. Acta Neurochirurgica (Wein) 120, 136–142. Martin, R.A., Shell, L., Dodds, W.J. (1986) Focal intramedullary spinal cord hematoma in a dog. Journal of the American Animal Hospital Association 22, 545–550.
335
336
Small Animal Spinal Disorders
Matthews, N.S., de Lahunta, A. (1985) Degenerative myelopathy in an adult Miniature Poodle. Journal of the American Veterinary Medical Association 186, 1213–1215. McEwan, J.D., Thomson, C., Sullivan, M., Callanan, S., Park, M. (1992) Thoracic spinal calcinosis circumscripta causing cord compression in two German Shepherd dog littermates. Veterinary Record 130, 575–578. McKee, W.M., Renwick, P.W. (1994) Marsupialisation of an arachnoid cyst in a dog. Journal of Small Animal Practice 35, 108–111. McKee, W.M., Mitten, R.W., Labuc, R.H. (1990) Surgical treatment of lumbosacral discospondylitis by a distraction-fusion technique. Journal of Small Animal Practice 31, 15–20. Meomartino, L., Fatone, G., De Vico, G. (1997) Osteochondromatosis (multiple cartilaginous exostoses) in a dog: report of a clinical case. Veterinaria Cremona 11, 95–99. Meric, S.M. (1988) Canine meningitis. Journal of Veterinary Internal Medicine 2, 26–35. Meric, S.M., Child, G., Higgins, R.J. (1986) Necrotizing vasculitis of the spinal pachyleptomeningeal arteries in three Bernese Mountain dog littermates. Journal of the American Animal Hospital Association 22, 459–465. Mesfin, G.M., Kusewitt, D., Parker, A. (1980) Degenerative myelopathy in a cat. Journal of the American Veterinary Medical Association 176, 62–64. Moissonnier, P., Duchossoy, Y., Lavieille, S., Reviron, T., Ye, J.H., Horvat, J.C. (1996) Reimplantation of ventral rootlets into the cervical spinal cord after their avulsion in the dog: a new surgical treatment of brachial plexus injuries? Veterinary Surgery 25 (3), 263. Moissonnier, P., Duchossoy, Y., Lavieille, S., Horvat, J.C. (2001) Evaluation of ventral root reimplantation as a treatment of experimental avulsion of the cranial brachial plexus in the dog. Revue de Medecine Veterinaire 152, 587–596. Moissonier, P., Carozzo, C., Gnirs, K., Ruel, Y., Blot, S., Begon, D., Rault, D., Delise, F., Boulouha, L., Colle, M.-A. (2002) What is the exact nature of ‘spinal subarachnoid cysts?’ Imaging and surgical investigation in 13 dogs. Veterinary Surgery 31, 301–302. Morgan, J.P., Stavenborn, M. (1991) Disseminated idiopathic skeletal hyperostosis (DISH) in a dog. Veterinary Radiology 32, 65–70. Morgan, J.P., Hansson, K., Miyabayashi, T. (1989) Spondylosis deformans in the female Beagle dog: a radiographic study. Journal of Small Animal Practice 30, 457–460. Morrison, J.L., Kaplan, P.A., Dussault, R.G., Anderson, M.W. (2000) Pedicle marrow signal intensity changes in the lumbar spine: a manifestation of facet degenerative joint disease. Skeletal Radiology 29, 703–707. Munana, K.R. (1996) Encephalitis and meningitis. Veterinary Clinics of North America Small Animal Practice 26, 857–874. Munana, K.R., Luttgen, P.J. (1998) Prognostic factors for dogs with granulomatous meningoencephalomyelitis: 42 cases (1982–1996). Journal of the American Veterinary Medical Association 212, 1902–1906. Nas, K., Gur, A., Kemaloglu, M.S., Geyik, M.F., Cevik, R., Buke, Y., Ceviz, A., Sarac, A.J., Aksu, Y. (2001) Management of spinal brucellosis and outcome of rehabilitation. Spinal Cord 39, 223–227. Nesbit, J.W., Lourens, D.C., Williams, M.C. (1981) Spastic paresis in two littermate pups caused by Toxoplasma gondii. Journal of the South African Veterinary Association 52, 243–246. Nishikawa, M., Sakamoto, H., Hakuba, A., Nakanishi, N., Inoue, Y. (1997) Pathogenesis of Chiari malformation: a morphometric study of the posterior cranial fossa. Journal of Neurosurgery 86, 40–47. Norrdin, R.W., Histand, M.B., Sheahan, H.J., Carpenter, T.R. (1990) Effects of corticosteroids on mechanical strength of intervertebral joints and vertebrae in dogs. Clinical Orthopaedics and Related Research 259, 268–276. Olby, N.J., Munana, K.R., Sharp, N.J.H., Thrall, D.E. (2000) The computed tomographic appearance of acute thoracolumbar intervertebral disc herniations in dogs. Veterinary Radiology and Ultrasound 41, 396–402. Osterholm, J. (1974) The pathophysiological response to spinal cord injury. The current status of related research. Journal of Neurosurgery 40, 5–33. Pang, W.W., Mok, M.S., Lin, M.L., Chang, D.P., Hwang, M.H. (1998) Application of spinal pain mapping in the diagnosis of low back pain— analysis of 104 cases. Acta Anaesthesiologica Sinica 36, 71–74. Pellegrini, N., Braca, G., Pozzo, F., Pambianco, L., Poli, A. (1979) Diabetic osteopathy in the dog. Post-mortem findings and pathogenesis. Annali della Facolta di Medicina Veterinaria di Pisa 31, 271–280.
Perrouin-Verbe, B., Lenne-Aurier, K., Robert, R., Auffray-Calvier, E., Richard, I., Mauduyt de la Greve, I., Mathe, J.F. (1998) Post-traumatic syringomyelia and post-traumatic spinal canal stenosis: a direct relationship: review of 75 patients with a spinal cord injury. Spinal Cord 36, 137–143. Plummer, S.B., Bunch, S.E., Khoo, L.H., Spaulding, K.A., Kornegay, J.N. (1993) Tethered spinal cord and an intradural lipoma associated with a meningocele in a Manx-type cat. Journal of the American Veterinary Medical Association 203, 1159–1161. Porter, R.W. (1996) Spinal stenosis and neurogenic claudication. Spine 21, 2046–2052. Postacchini, F., Montanaro, A. (1980) Tuberculous epidural granuloma simulating a herniated lumbar disk: a report of a case. Clinical Orthopaedics and Related Research 148, 182–185. Reihsaus, E., Waldbaur, H., Seeling, W. (2000) Spinal epidural abscess: a meta-analysis of 915 patients. Neurosurgical Review 23, 175–204; discussion 205. Remedios, A.M., Wagner, R., Caulkett, N.A., Duke, T. (1996) Epidural abscess and discospondylitis in a dog after administration of a lumbosacral epidural analgesic. Canadian Veterinary Journal 37, 106–107. Riggs, B.L. (2002) Endocrine causes of age-related bone loss and osteoporosis. Novartis Foundation Symposium Series 242, 247–259; discussion 260–264. Robinson, D.E., Ball, K.E., Webb, P.J. (2001) Iliopsoas hematoma with femoral neuropathy presenting a diagnostic dilemma after spinal decompression. Spine 26, E135–138. Rodkey, W.G., Sharp, N.J.H. (2003) Surgery of the peripheral nervous system. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 1218–1227. Philadelphia: WB Saunders. Rusbridge, C. (1997) Persistent scratching in Cavalier King Charles Spaniels. Veterinary Record 141, 179. Rusbridge, C., MacSweeny, J.E., Davies, J.V., Chandler, K., Fitzmaurice, S.N., Dennis, R., Cappello, R., Wheeler, S.J. (2000) Syringohydromyelia in Cavalier King Charles Spaniels. Journal of the American Animal Hospital Association 36, 34–41. Rylander, H., Lipsitz, D., Berry, W.L., Sturges, B.K., Vernau, K.M., Dickinson, P.J., Anor, S.A., Higgins, R.J., LeCouteur, R.A. (2002) Retrospective analysis of spinal arachnoid cysts in 14 dogs. Journal of Veterinary Internal Medicine 16, 690–696. Sakamoto, H., Nishikawa, M., Hakuba, A., Yasui, T., Kitano, S., Nakanishi, N., Inoue, Y. (1999) Expansive suboccipital cranioplasty for the treatment of syringomyelia associated with Chiari malformation. Acta Neurochirurgica (Wein) 141, 949–960; discussion 960–961. Salinas, E.M., Martinez, N.L. (1993) Description of a case compatible with degenerative myelopathy in an English Mastiff dog. Veterinaria Mexico 24, 159–162. Schleithoff, N.S. (1984) Frequency of occurrence of skeletal lesions among adult dogs Berlin: Fachbereich Veterinarmedizin der Freien Universitat Berlin. Scott, H.W., O’Leary, M.T. (1996) Fibro-cartilaginous embolism in a cat. Journal of Small Animal Practice 37, 228–231. Seim, H.B., Prata, R.G. (1982) Ventral decompression for the treatment of cervical disk disease in the dog: a review of 54 cases. Journal of the American Animal Hospital Association 18, 233–240. Sgouros, S., Williams, B. (1995) A critical appraisal of drainage in syringomyelia. Journal of Neurosurgery 82, 1–10. Sgouros, S., Williams, B. (1996) Management and outcome of posttraumatic syringomyelia. Journal of Neurosurgery 85, 197–205. Shamir, M., Rochkind, S., Johnston, D. (2001) Surgical treatment of tethered spinal cord syndrome in a dog with myelomeningocele. Veterinary Record 148, 755–756. Shamir, M.H., Shahar, R., Aizenberg, I. (1997) Subarachnoid cyst in a cat. Journal of the American Animal Hospital Association 33, 123–125. Sharp, N.J.H., Vitek, S., Steingold, S., Olby, N.J., Munana, K. (2000) A rapid, antemortem diagnostic test for canine distemper virus. Journal of Veterinary Internal Medicine 14, 341. Skeen, T.M., Olby, N.J., Munana, K.M., Sharp, N.J.H. (2003) Spinal arachnoid cysts in 17 dogs. Journal of the American Animal Hospital Association 39, 271–282. Smith, S.A., Tobias, A.H., Jacob, K.A., Fine, D.M., Grumbles, P.L. (2003) Arterial thromboembolism in cats: acute crisis in 127 cases (1992–2001)
Miscellaneous conditions
and long-term management with low-dose aspirin in 24 cases. Journal of Veterinary Internal Medicine 17, 73–83. Stefanacci, J.D., Wheeler, S.J. (1991) Skeletal scintigraphy in canine discospondylitis. American College of Veterinary Radiology, 66. Stiles, J. (2000) Canine rickettsial infections. Veterinary Clinics of North America, Small Animal Practice 30, 1135–1149. Stokol, T., Parry, B.W., Mansell, P.D., Richardson, J.L. (1994) Hematorrhachis associated with hemophilia A in three German Shepherd dogs. Journal of the American Animal Hospital Association 30, 239–243. Summers, B.A., Cummings, J.F., de Lahunta, A. (1994) Veterinary Neuropathology. St Louis: Mosby. Swezey, R.L. (1998) The sacroiliac joint. Nothing is sacred. Physical Medicine and Rehabilitation Clinics of North America 9, 515–519. Taga, A., Taura, Y., Nakaichi, M., Wada, N., Hasegawa, T. (2000) Magnetic resonance imaging of syringomyelia in five dogs. Journal of Small Animal Practice 41, 362–365. Tamke, P.G., Peterson, M.G., Dietze, A.E., de Lahunta, A. (1988) Acquired hydrocephalus and hydromelia in a cat with feline infectious peritonitis: a case report and brief review. Canadian Veterinary Journal 29, 997–1000. Tan, D.Y., Tsou, I.Y., Chee, T.S. (2002) Differentiation of malignant vertebral collapse from osteoporotic and other benign causes using magnetic resonance imaging. Annals of the Academy of Medicine Singapore 31, 8–14. Tarvin, G., Prata, R.G. (1980) Lumbosacral stenosis in dogs. Journal of the American Veterinary Medical Association 177, 154–159. Thomas, W.B., Sorjonen, D.C., Steiss, J.E. (1993) A retrospective evaluation of 38 cases of canine distemper encephalomyelitis. Journal of the American Animal Hospital Association 29, 129–133. Tipold, A., Jaggy, A. (1994) Steroid responsive meningitis-arteritis in dogs: long-term study of 32 cases. Journal of Small Animal Practice 35, 311–316. Tipold, A., Vandevelde, M., Jaggy, A. (1992) Neurological manifestations of canine distemper virus infection. Journal of Small Animal Practice 33, 466–470. Tomlinson, J., Higgins, R.J., LeCouteur, R.A., Knapp, D. (1988) Intraspinal epidermoid cyst in a dog. Journal of the American Veterinary Medical Association 193, 1435–1436. Tshamala, M., Moens, Y. (2000) True dermoid cyst in a Rhodesian ridgeback. Journal of Small Animal Practice 41, 352–353. Vandecasteele, S.J., Knockaert, D., Verhaegen, J., Van Eldere, J., Peetermans, W.E. (2001) The antibiotic and anti-inflammatory treat-
ment of bacterial meningitis in adults: do we have to change our strategies in an era of increasing antibiotic resistance? Acta Clinica Belgica 56, 225–233. van Winkle, T.J., Liu, S.M., Hackner, S.G. (1993) Clinical and pathological features of aortic thromboembolism in 36 dogs. Journal of Veterinary Emergency and Critical Care 3, 13–21. Vestergaard, P., Lindholm, J., Jorgensen, J.O., Hagen, C., Hoeck, H.C., Laurberg, P., Rejnmark, L., Brixen, K., Kristensen, L.O., FeldtRasmussen, U., Mosekilde, L. (2002) Increased risk of osteoporotic fractures in patients with Cushing’s syndrome. European Journal of Endocrinology 146, 51–56. Walker, T.M., Pierce, W.A., Welch, R.D. (2002) External fixation of the lumbar spine in a canine model. Veterinary Surgery 31, 181–188. Watson, E., Roberts, R.E. (1993) Discospondylitis in a cat. Veterinary Radiology and Ultrasound 34, 397–398. Watt, P.R., Robins, G.M., Galloway, A.M., O’Boyle, D.A. (1995) Disseminated opportunistic fungal disease in dogs: 10 cases (1982–1990). Journal of the American Veterinary Medical Association 207, 67–70. Watters, M.R., Stears, J.C., Osborn, A.G., Turner, G.E., Burton, B.S., Lillehei, K., Yuh, W.T. (1998) Transdural spinal cord herniation: imaging and clinical spectra. American Journal of Neuroradiology 19, 1337–1344. Webb, A.A., Pharr, J.W., Lew, L.J., Tryon, K.A. (2001) MR imaging findings in a dog with lumbar ganglion cysts. Veterinary Radiology and Ultrasound 42, 9–13. Wilson, J.W. (1982) Spina bifida in the dog and cat. Compendium on Continuing Education for the Practicing Veterinarian 4, 626–632, 636–637. Wilson, J.W., Kurtz, H.J., Leipold, H.W., Lees, G.E. (1979) Spina bifida in the dog. Veterinary Pathology 16, 165–179. Yamaura, M., Nakamura, T., Nagai, Y., Yoshihara, A., Suzuki, K. (1993) Reduced mechanical competence of bone by ovariectomy and its preservation with 24R,25-dihydroxyvitamin D3 administration in beagles. Calcified Tissue International 52, 49–56.
FURTHER READING Braund, K.G. (2003) Clinical Neurology in Small Animals: Localization, Diagnosis and Treatment. http://www.ivis.org/special_books/Braund/ toc.asp
337
Postoperative care
Chapter
15
Analgesia 339 Opioid analgesics 340 Non-steroidal anti-inflammatory drug (NSAID) analgesia 341 Analgesia in cats 343 Other agents 343
Box 15.1 Important clinical signs to recognize in the postoperative neurosurgical patient
Nursing care 343 Moving patients 343 Myelography 343 Treatment plans 343 Cleanliness 343 Urinalysis 344 Recumbency 344 Hydration and nutrition
344
Flooring for recumbent patients Physiotherapy
345
346
Control of urinary function 350 Anatomy and physiology 350 Disorders of micturition 351 Pharmacological manipulation of micturition Assisted emptying of the bladder 352
351
Postoperative complications 355 Urinary tract infections 355 Gastrointestinal disturbances 355 Pancreatitis 357 Wound complications 357 Urine scald 357 Decubital ulcers 358 Miscellaneous 359 References
360
The surgeon should accept that some form of postoperative complication will develop in many neurosurgical patients. This may range from a simple urinary tract infection (UTI) to less common but life-threatening
■
Post-myelographic seizures
■
Inappetance
■
Depression
■
Persistent fever
■
Wound discharge
■
Diarrhea or vomiting
■
Abdominal pain
■
Melena or ‘coffee ground’ vomit
■
Change in ability to void, or in the ease of manual expression
■
Blood or floccular material in urine
■
Odiferous urine
■
Urine scald
■
Decubital ulcers
■
Increase in spinal pain
■
Deterioration in neurological status
conditions such as pneumonia or pancreatitis. If complications are viewed as something to be expected the surgeon is more likely to have the patient monitored appropriately (Box 15.1). Care of neurosurgical patients can be labor intensive (Table 15.4). It is important that hospital nursing staff have a high standard of training and are properly informed about what parameters to monitor, how often to monitor them, and what the most likely complications are going to be in each individual patient. Clear written instructions should also be given to the client at the time of discharge.
ANALGESIA Neurosurgical procedures often cause a great deal of pain for human patients. It is likely that in the past veterinarians have tended to overlook the degree of pain suffered by our patients (Hansen and Hardie, 1993; Dohoo and Dohoo, 1996; Lascelles et al., 1999). Doubt as to whether an animal really is in pain often caused a
340
Small Animal Spinal Disorders
clinician to withhold analgesics, yet the same clinician would administer antibiotics without documented evidence of infection (Crane, 1987). When in doubt, a dose of opioid should be tested for effect.
Opioid analgesics Opioids are the most effective analgesic for severe pain. The approximate duration of action of these drugs, together with their advantages and disadvantages, are outlined in Table 15.1. Preoperative or intraoperative opioid administration is recommended to reduce the requirement for postoperative analgesia. Fentanyl patches are a convenient way to provide long-lasting analgesia (15.1). As these patches have a delayed onset of action (see Chapter 6, page 84), they should be applied the night before elective surgeries and are supplemented with additional intravenous opioid agonists as needed.
The pure opioid agonists are the most potent analgesics but they also have more potential side-effects. They should be given on a fixed schedule every 2–4 h; the interval can be decreased if greater analgesia is needed. The response varies markedly between patients. Therefore small, intravenous doses should be given every 5 min (for dogs and cats 0.1 mg/kg morphine; 0.05 mg/kg oxymorphone; 0.05 mg/kg hydromorphone) until a satisfactory level of analgesia is achieved (stop administration if mydriasis occurs in cats). The total dose required for that patient is then repeated on the fixed schedule thereafter. A continuous intravenous infusion can also be used by delivering the cumulative dose for each time interval as a constant rate infusion over that same period, but the animal must be re-evaluated frequently in case adjustments are needed to the rate. Common side-effects include ileus and nausea.
Table 15.1 Narcotic analgesic agents1 Drug*
Dose
Interval
Advantages
Disadvantages
Morphine (pure agonist)
Dog: 0.2–1.0 mg/kg IM, SQ Dog: 1–5.0 mg/kg PO Cat: 0.1–0.4 mg/kg IM, SQ
4h 4–6 h 4h
Inexpensive, sedative
Respiratory depression Bradycardia** Caution IV** Emesis
Oxymorphone (pure agonist)
Dog: 0.05–0.2 mg/kg IV, IM, SQ Cat: 0.02–0.1 mg/kg IV, IM, SQ
2–4 h
Less vomiting than with morphine
As for morphine Expense, availability
Hydromorphone (pure agonist)
Dog: 0.02–0.4 mg/kg IV, IM, SQ Cat: 0.02–0.2 mg/kg IV, IM, SQ
2–4 h
Similar to oxymorphone
As for morphine Expense
Butorphanol (agonist/antagonist)
Dog: 0.2–1.0 mg/kg IV, IM, SQ Cat: 0.1–0.6 mg/kg IV, IM, SQ; 0.5–1.0 mg/kg PO
1–2 h
Reduced respiratory or cardiovascular effects
Less potent analgesic in severe pain, sedation lasts longer than analgesia
Reduced respiratory or cardiovascular effects
If respiratory depression does occur it can be difficult to reverse
2–6 h 6–8 h
Buprenorphine (partial agonist)
Dog: 0.005–0.02 mg/kg 6–12 h IV, IM, SQ Cat: 0.005–0.01 mg/kg IV, IM, SQ
Fentanyl transdermal patch (pure agonist)
Dog: ⬍10 kg: 25 g 10–29 kg: 50 g ⬎30 kg: 75–150 g Cat: 25 g patch
Lasts up to Convenient, well 3 days; longer tolerated, apply before in cats surgery (see page 84)
Delay in onset 12–24 h dogs, 6–12 h cats. Ineffective in some animals
Codeine 60 mg with 300 mg acetaminophen
Dog: Dose at 10 mg/kg of the acetaminophen PO
8h
Sedation, GI sideeffects. Not for cats
Inexpensive Useful orally
1 B. Hansen, personal communication. * Consult appropriate source for contraindications and adverse reactions and to verify doses. Many opioids are controlled substances. ** Concomitant administration of atropine may be required to overcome bradycardia; this is only a problem if hypovolemic or under anesthesia. GI, gastrointestinal.
Postoperative care
Histamine release occurs in dogs after IV administration of morphine; this can cause problems for animals under general anesthesia or those with hypovolemia but not usually for hemodynamically stable, conscious dogs. The dose-dependent respiratory depression of opioid agonists means that they must be used with care in animals that have decreased ventilatory function, such as those with cranial cervical spinal cord lesions. Opioid agonists can increase intracranial pressure and should be used with care after severe head trauma unless paCO2 can be monitored and preferably intracranial and mean arterial blood pressure as well (Marik et al., 1999; de Nadal et al., 2000; Lam and Warner, 2001). Judicious use of these agents is justified in animals with less severe head injuries provided that their neurological status can be monitored closely and ideally that the effect on paCO2 can also be assessed. Animals that become hypercapnic will need to be ventilated in order to restore paCO2 to between 25 and 30 mmHg. Epidural morphine is a useful technique for some neurosurgical patients such as those with severe nerve root or cancer pain. Respiratory depression has not been a problem in normal dogs. It does not have central opioid effects and is not associated with sensory, sympathetic or motor blockade, so the patient can still walk and analgesia can be effective as far cranial as the thoracic limbs. Epidural morphine is equivalent both to intravenous morphine and to local bupivacaine nerve block for controlling pain after lateral thoracotomy (Pascoe and Dyson, 1993; Popilskis et al., 1993). The onset of action is from 20 to 60 min and mean duration of analgesia is 6 h; this
15.1 Skin reaction to a fentanyl patch. The patch has been removed as dermatitis can increase the rate of absorption. Fever can have the same effect (Bagley et al., 2000).
increases in dogs to a mean of 13 h with the addition of epidural medetomidine at 5 g/kg (Branson et al., 1993). Buprenorphine and morphine are equally effective when given epidurally to dogs; the dose for morphine is 0.1 mg/kg and for buprenorphine is 4 g/kg, diluted in warm saline to a volume of 1 ml per 5 kg and injected into the lumbosacral epidural space (Smith et al., 2001). Urinary retention can occur after epidural or systemic opioid (Herperger, 1998). Epidural abscessation has been reported as a rare complication of epidural analgesia (Remedios et al., 1996). In general, transdermal fentanyl lasts much longer than epidural morphine and is easier to use (Robinson et al., 1999).
Non-steroidal anti-inflammatory drug (NSAID) analgesia The NSAIDs are indicated either as a supplement to opioid analgesia or when opioids are unsuitable (Table 15.2). Carprofen is particularly useful as it can also be given intravenously; it is most effective when given preoperatively (Lascelles et al., 1998). All NSAIDs provide analgesia by blocking the cyclooxygenase enzymes that produce prostaglandins. Cyclooxygenase (COX) activity is either primarily constitutive (the COX-1 enzyme constantly produces prostaglandins necessary for normal metabolic functions in many tissues) or primarily inducible (the COX-2 enzyme is induced by inflammation to produce prostaglandins locally). Prostaglandins are integral to the inflammatory process and they enhance nociception both peripherally and centrally. Some prostaglandins, particularly prostaglandin E (PgE) are important for homeostasis in tissues like kidney and intestine. Newer NSAIDs have now been developed that only inhibit COX-1 weakly. Carprofen, ketoprofen, etodolac and meloxicam are COX-1 sparing; carprofen and ketoprofen being somewhat less so than meloxicam (KayMugford et al., 2000; Brideau et al., 2001). The analgesic effects of these newer NSAIDs are good and they are roughly equivalent to each other. Carprofen provides superior analgesia to the opioid pethidine in dogs and cats (Lascelles et al., 1994; Balmer et al., 1998); higher doses of most NSAIDs provide analgesia comparable to low doses of opioids in most studies. The COX-1 sparing NSAIDs in general cause less gastrointestinal ulceration than more potent COX-1 inhibitors. The gastrointestinal side-effects of etodoloc and carprofen are significantly less than aspirin (Reimer et al., 1999). Carprofen is known to preserve protective levels of PgE in the gut and no significant gastrointestinal problems were observed in more than 200 dogs with degenerative joint disease (Holtsinger et al., 1992). However, meloxicam in higher doses can cause severe
341
342
Small Animal Spinal Disorders
Table 15.2 Non-steroidal anti-inflammatory drug analgesia1 Drug*
Dose—dog
Dose—cat
Comment**
Aspirin
10–25 mg/kg PO q12 h
10 mg/kg PO q36–48 h
Coagulopathy
Ketoprofen
1 mg/kg PO, IV, IM, SQ q24 h for 5 days
1 mg/kg PO, IV, IM, SQ q24 h for 5 days
Meloxicam
Acute pain: 0.2 mg/kg PO, IV, SC once, then 0.1 mg/kg q24 h
Acute pain: 0.2 mg/kg PO, IV, SC once, then 0.1 mg/kg PO q24 h for 3–4 days Long-term therapy: up to 0.1 mg per cat q48–72 h
Long-term therapy: 0.1 mg/kg q24 h Etodolac
10–15 mg/kg PO q24 h Lower dose for long-term use
N/A
Meclofenamic acid
1.1 mg/kg PO q24 h Lower dose for long-term use
N/A
Deracoxib
3–4 mg/kg q24 h
N/A
Carprofen
2 mg/kg PO, IV q12 h
2–4 mg/kg q5 days PO, IV, SC
Rare, severe hepatic effects. Do not repeat within 5 days or use long-term for cats
1 B. Hansen, personal communication. * Consult appropriate source for full list of contraindications and adverse reactions and to verify doses. ** All NSAIDs have the potential to cause GI side-effects. N/A, not available.
ulceration despite its greater selectivity for COX-2. Acute renal failure has been reported in humans taking COX-1 sparing drugs (Goldstein et al., 2001; Ahmad et al., 2002). Owing to the tendency for neurosurgical patients to develop gastrointestinal disturbances, NSAIDs should not be used for more than a few days and this probably applies to COX-1 sparing drugs as well (Ahmad et al., 2002). A specific contraindication to NSAID use is when the animal has recently had corticosteroids, including methylprednisolone sodium succinate (MPSS), because of the high risk of gastrointestinal bleeding and even perforation (Toombs et al., 1986; Strombeck and Guilford, 1990; Hinton et al., 2002). Care in the use of NSAIDs is also needed in any animal with renal or hepatic dysfunction; especially during periods of hemodynamic instability. Furthermore, COX-1 inhibition by some of these drugs, such as aspirin and ketoprofen, prevents thromboxane A2 synthesis and causes decreased platelet aggregation with significantly increased bleeding times (Grisneaux et al., 1999; Mathews et al., 2001; Kerwin and Maudlin, 2003) (15.2). A synthetic form of prostaglandin E
15.2 Postoperative appearance of a dog given 12 mg/kg of aspirin 24 h before ventral slot. Severe bruising was probably due to poor platelet aggregation. Aspirin blocks COX-1 irreversibly, thereby inactivating platelets. New platelets must then be generated and so the effect on coagulation can last 7 days or more. Aspirin should not be used preoperatively.
Postoperative care
(misoprostol) can be used prophylactically to counter gastrointestinal side-effects of NSAIDs (Walt, 1992; Murtaugh et al., 1993). Misoprostol is also protective to the kidney and could be used for this purpose should NSAIDs be deemed necessary (Paller and Manivel, 1992; Shield, 1992; Davies et al., 2001).
Analgesia in cats Selected opioids can be used in cats (Table 15.1). Meloxicam and carprofen are also tolerated well by cats using oral (both drugs) and IV (carprofen) routes (Parton et al., 2000; Slingsby et al., 2000; Lascelles et al., 2001) (Table 15.2). Acetaminophen should not be used in cats.
Other agents These include: • Gabapentin is an anticonvulsant that has also been used to alleviate trigeminal pain in humans. It can be an effective analgesic for animals with pain caused by neoplasia or inflammation of a nerve root. The dose range starts at 5 mg/kg q8h and extends in dogs up to 50 mg/kg q8h given to effect; potential side-effects include sedation and ataxia. • Medetomidine can provide useful supplemental analgesic at low doses (1–3 g/kg/h). The drug’s cardiovascular effects are much shorter in duration when used at such low doses but are quite dramatic with an IV bolus. The dose should therefore be given over 10 min to avoid hypotension; it can also be delivered by syringe pump or in IV fluids. • The skeletal muscle relaxants methocarbamol (55–132 mg/kg PO in divided doses) or diazepam ( Table 15.7) are useful to relieve muscle spasm after spinal surgery. • Acupuncture may also be a helpful adjunctive means of providing analgesia (Haskins, 1987; Still, 1989; Janssens, 1992). • Physiotherapy such as heat, ultrasound (15.13B) and laser (15.14A) can provide additional pain relief. Pain relief after laser therapy is probably mediated mainly through serotonin release (Walker, 1983; Clokie et al., 1991).
Myelography It is important that the animal’s head is kept elevated at all times following the injection of contrast into the subarachnoid space, both during the myelogram and during recovery from anesthesia. This simple precaution is easily overlooked as a means of preventing postmyelographic seizures.
Treatment plans The best way to nurse a neurological patient, especially one with multiple problems, is to make a plan for each day as illustrated in Tables 15.3 and 15.4. In this manner, voiding requirements, physiotherapy needs, medications, laboratory work and routine tasks are planned out and not forgotten. A specific plan for physiotherapy is also useful (Table 15.5).
Cleanliness Animals that soil themselves repeatedly may be easier to manage if their entire hindquarters are shaved. Soiled areas should be bathed as necessary; bathing of the perineal area may be needed several times a day in some animals. It is often suggested that patients with surgical wounds should not be bathed until a minimum of 5 days after surgery. However, in one study bathing Table 15.3 Nursing plan for a neurological patient* Task
0800
TPR
(X)
Express bladder
(X)
Feed 1/4 can
(X)
Check drinking water
(X)
Phenoxybenzamine 5 mg PO
(X)
( )
Diazepam 2.5 mg PO
(X)
( )
Trimethoprim-sulfa 120 mg PO
(X)
Urinalysis
NURSING CARE Moving patients Proper support must be provided for the spine when moving an animal with a spinal lesion, especially an anesthetized patient. Use of a board or stretcher is indicated (13.1).
1200
1600
2000(etc)
( )
( )
( )
( ) ( )
( )
( )
( ) ( )
( )
*Taken from the hospital record of a 4-year-old male Dachshund that had undergone a hemilaminectomy for thoracolumbar disc extrusion 4 days previously. The dog was paraplegic and incontinent, with reduced deep pain sensation. An E. coli, sensitive to trimethoprim/ sulfa, had been cultured from his urine on admission. TPR, temperature, pulse and respiratory rate; (X), task completed; ( ), task still to be completed.
343
344
Small Animal Spinal Disorders
and swimming did not influence the incidence of wound infection, even when performed in the first 3 days after surgery (Hosgood, 1992).
Urinalysis If urinary incontinence is present, a urinalysis should be done every 2–3 days, regardless of whether the patient is on antibiotics. Most animals with neurological deficits that are severe enough to interfere with motor function will also have inefficient bladder emptying that predisposes them to retention cystitis.
Recumbency Recumbent, tetraparetic animals need to be turned every 2 h because sustained hypostatic congestion predisposes the patient to pneumonia; alternating between lateral and sternal recumbency will also help to improve
ventilation. The chest should be auscultated twice daily to facilitate early detection of problems. Patients should be fed while supported in a sternal position to reduce the risk of aspiration (Nicoll and Remedios, 1995). They should be kept on a water bed if possible to minimize the development of decubitus.
Hydration and nutrition Tetraparetic animals dehydrate easily and should be offered water from a sternal position every 4 h and most also require intravenous fluids. Total protein, hematocrit, urine specific gravity and serum electrolyte concentrations should be monitored daily in these patients. Nutritional support for the neurosurgical patient should also be evaluated carefully. Stress, particularly from trauma or surgery, results in an increased metabolic rate and so enhanced nutritional intake is important
Table 15.4 Nursing plan for a neurological patient* Task
8 am 10 am 12 noon 2 pm 4 pm 6 pm
8 pm 10 pm 12 midnight 2 am 4 am 6 am
TPR b.i.d.
(X)
( )
Express bladder q 6 h; take outside
(X)
Nil per os except water
_
Offer water q4 h and place sternal
(X)
Sucralfate 500 mg per os q12 h
(X)
( )
375 mg Amoxycillin– clavulanic acid q12 h
(X)
( )
Rebandage feet
(X)
Turn q4 h
(X)
( )
_
Chest radiograph
( )
_
_
( )
_
( )
( ) ( )
( )
_
( )
( ) ( )
(X)
_
( )
Coupage and turn q4 h
Place in sling for 1/2 h
_
( )
( ) ( )
( )
_
_
_
( )
( )
( )
( ) ( )
_
( )
( )
* Taken from the hospital record of a 6-year-old male Doberman (same dog as shown in 15.7) with cervical spondylomyelopathy that had undergone a ventral decompression 1 week previously. The dog was recumbent, tetraparetic, and although continent, he needed manual expression of his bladder to initiate voiding. Food was being withheld as he had suffered severe bloody diarrhea the previous day, for which he was being treated with sucralfate. The dog also tended to chew his feet if they were left unbandaged. The chest radiograph was scheduled because the dog appeared depressed and auscultation of the lungs suggested early pneumonia. Coupage and turning the animal effectively every hour is preferable for pneumonia; propping the animal into sternal recumbency is also helpful. The nursing care for this patient is particularly complex, and illustrates the support needed for certain types of neurosurgical patients. TPR b.i.d., temperature, pulse and respiration rate, twice a day; (X), task completed; ( ), task still to be completed.
Postoperative care
to minimize immunosuppression and the tendency to deplete body protein. The patient should be given palatable, high-quality meals several times daily to cope with their increased requirement; ideally, their exact caloric requirements should be calculated. Conversely, an inactive, overweight, paraplegic dog with no serious complications may need a reduction in its nutritional intake (Strombeck and Guilford, 1990).
FLOORING FOR RECUMBENT PATIENTS (15.3–15.7) Different flooring materials and their indications are illustrated below. In general, a raised grate is best for most small paraparetic or paraplegic patients. The advantage of a grate is the reduced tendency for an animal to lie in its own urine and feces, particularly if non-retentive bedding is provided. A grate does not provide padding
15.5 A waterproof foam mattress; in general the foam should be as thick as possible. Non-retentive bedding is recommended as urine tends to pool in the depression made by the patient. Foam does distribute body weight more evenly but decubitus will still develop without additional precautions.
15.3 Paraplegic dog on a raised grate. This flooring allows urine and other fluids to pass through it keeping the dog cleaner and drier. It is suitable for most small patients and for larger patients with minimal neurological deficits.
15.6 An inflatable airbed also provides good protection against decubitus. A useful, cheap and disposable alternative is to use plastic bubble packing material; the largest-sized bubbles seem to work best.
15.4 Non-retentive bedding. Artificial fleece (Unreal Lamb Skin, Alpha Protech, North Salt Lake, UT) is an excellent material to use with a raised grate as it stays dry yet allows liquids to pass through (Swaim et al., 1996). It may make some patients uncomfortably hot (Nicoll and Remedios, 1995).
15.7 A waterbed is the ideal surface for a large recumbent patient. Problems include cost and the risk of cooling the animal unless the water is heated. Punctures can be a problem just as they can for airbeds. The patient also tends to roll off the edge unless the bed is surrounded by a barrier. See Table 15.4.
345
346
Small Animal Spinal Disorders
and is not suitable for large, severely tetraparetic patients or those with established decubitus. In such cases, a variety of padded flooring materials can be employed; each has its own advantages and disadvantages. Nursing staff must be aware that patients with disturbed pain perception are at extreme risk for the development of thermal burns. A thermostatically controlled
circulating water blanket is the only safe way to warm a neurological patient (15.8). Electric heating blankets can be unreliable and dangerous (Swaim et al., 1989). Direct heat must not be applied to patients with a fentanyl patch (page 85).
PHYSIOTHERAPY However good the flooring material, the aim is to minimize the recumbency period, because this is when the animal is most susceptible to complications. This period can be shortened by physiotherapy and early mobilization. Towel walking of paraplegic dogs or supporting a tetraparetic patient in a sling are often possible within 24–48 h of surgery (15.9, 15.10).
15.8 Neurological patients are susceptible to cooling excessively as they may have difficulty avoiding draughts or generating their own heat. Heat can only be supplied safely by warming the surrounding air (15.20), or via a thermostatically controlled water blanket.
A
A
B 15.9 A: Walking a paraplegic dog by supporting its hindquarters with a towel placed under the abdomen just in front of the pelvic limbs. The tail can also be used to provide support provided that it is held at the base to avoid injury. B: A Walkabout sling (Walkabout Harnesses, Santa Cruz, CA) can also be useful for early mobilization.
B 15.10 These two slings with frames provide an excellent way to rehabilitate tetraparetic dogs. A: A canvas sling and metal frame; this sling tends to rub more than the one shown in B. B: This sling is made from Neoprene (Dupont-Dow, Wilmington, DE); it is shown in 15.16 supporting a dog in a whirlpool bath.
Postoperative care
Tetraparetic animals with good motor function can be walked with assistance using a tracking harness (15.11). Assistants must take care not to injure themselves when lifting heavy patients (15.12). Standard physiotherapy techniques are often effective when modified for animals (Taylor, 1992; Bockstahler et al., 2002) (15.13–15.15). Massage and
15.11 A tracking harness gives effective support to a tetraparetic dog with good motor function. This is the same dog illustrated in 15.38 and 15.39, 4 months after surgery for cervical spondylomyelopathy (11.22). The decubital ulcer visible in 15.38 has contracted but is still open and eventually required surgical closure.
passive range of motion can usually begin almost immediately after recovery from surgery. Massage of the limbs for 15 min once or twice a day is usually well tolerated especially for recumbent animals. Massage should be performed in a distal to proximal direction in order to promote venous return (Berry and Reyers, 1990). Hot packing is useful to reduce swelling, pain and muscle spasm (7.12). Hot packs should be insulated from the animal’s skin; treatments should last 10–20 min repeated every 8–12 h (Jerram et al., 1997). Ultrasound is another way of applying warmth to deeper tissues and is useful to prevent and treat muscle spasm (Taylor, 1992) (15.13B). Ultrasound is converted to heat, mainly at the bone–tissue interface; it causes little temperature rise in superficial tissue. Intensities from 0.5 to 4 W/s are used; pain indicates excessive heat generation. Creative thinking can adapt other human techniques to veterinary patients (Taylor, 1992) (15.14, 15.15). Neuromuscular stimulator packs (Respond Dual Channel Neuromuscular Stimulator, Medtronic) can build specific muscle groups and can be useful (Taylor, 1992), for example after brachial plexus injury has caused shoulder muscle atrophy (see Chapter 2). As for nursing care, it is recommended to develop a specific plan for each animal (Table 15.5). An external splint (see
15.12 A: A mechanical hoist is invaluable for lifting large, recumbent dogs in a sling. The hoist is raised and lowered using a simple lever mechanism (arrowhead). B: This dog has been suspended from the same hoist but this time using two Walkabout slings (Walkabout Harnesses, Santa Cruz, CA).
A
B 15.13 A: Passive range of motion; this is done for 5–10 times in two or three sessions a day (Jerram et al., 1997). B: Ultrasound should not be used over nerves or directly over a laminectomy defect as it generates heat (Taylor, 1992). Here the energy is being directed away from the bony defect.
A
B
347
348
Small Animal Spinal Disorders
15.14 A: Low power, cold laser is used immediately after surgery to stimulate healing by evoking an intracellular, photochemical response; it also causes release of serotonin to provide pain relief. B: A treadmill is invaluable for tetraparetic dogs; it can be combined with a sling as here.
A
B
15.15 Balance training stimulates recovery of proprioceptive responses. A: Tetraparetic dog on a ball. B: Same dog on a balance board.
A
B
Table 15.5 Physical therapy plan for a neurological patient* Task
8 am
Range of motion
10 am
12 noon
X
2 pm
4 pm
X
6 pm
8 pm
X
10 pm
12 midnight
X
Proprioceptive sensory stimulation
X
X
X
X
X
Massage (effleurage)
X
X
X
X
X
Helium–neon laser; scan surgical site Standing balance in sling (15.10, 15.12). 10 min increasing to 30 min
X X
X
X
X
* Taken from the hospital record of an 11-year-old dog (similar to the dog shown in 15.10A) with cervical spondylomyelopathy that had undergone a ventral decompression 1 week previously. Active postoperative physical therapy begins postoperative Day 1. Active mobilization and sensory stimulation continues every 4 h until the dog is discharged. X, task completed.
13.46) can be applied as a temporary precaution to make physiotherapy safer for an animal with an unstable thoracolumbar lesion. Swimming in a whirlpool bath or a bathtub is invaluable (15.16–15.19) but the animal must be dried thoroughly afterwards (15.20).
In addition to the physical well-being of the patient, the mental attitude of the animal often has a significant effect on outcome, particularly when recovery is delayed (15.39). For dogs facing protracted recovery periods the temporary and intermittent use of a paraplegic cart can
Postoperative care
15.16 The sling shown in 15.10B has been used to support this dog inside a whirlpool bath. While in the pool, the patient’s limbs are flexed and it is encouraged to swim. Chlorhexidine has been added to the water, which should be at 38–40°C. The tank is either filtered or it must be drained daily.
15.17 A small dog can be suspended by hand in a shallow bath. Water levels can be varied to provide support along with some degree of weight bearing. At the end of the session, the patient must be dried with towels and a warm air drier to prevent hypothermia (15.20).
15.18 Life jackets are necessary when an animal is in a deep pool as it can still get into difficulty, especially once tired. A: The jacket has a handle on top and must fit properly so that the dog cannot slip out of it (K-9 Float coat, Ruffwear, Bend, OR). B: Hydrotherapy should begin slowly, an attendant must always be present and the animal must wear a life jacket (Taylor, 1992; Jerram et al., 1997). A
B
15.19 Here the patient is being made to swim against a jet of water; it can also be made to swim around the tank to retrieve a ball. Animals with cervical lesions may be reluctant to maintain their head in this somewhat extended position.
15.20 A warm air blower is invaluable as neurological patients are inefficient at drying themselves. The drier must not be too hot, as the patient may be unable to move away or may not even feel the heat on areas with poor nociception.
349
350
Small Animal Spinal Disorders
play a role in rehabilitation. In general, the risk of the dog coming to rely on the cart is usually outweighed by the stimulus of enhanced mobility although they should probably not be used for the first month in order to encourage the animal to walk. In special circumstances, some dogs will do well in a cart long-term (15.21). The permanent use of carts is suitable for selected patients in which there is no hope for return of neurological function. In such cases the owner must: • Be informed of the final prognosis. • Be able to cope with the physical demands of getting the dog into and out of the cart. • Be able to be with the dog while it is in the cart. • Be able to manage the incontinence and tendency for urine scald along with repeated infections.
•
Insure that the bladder is emptied at least three times daily. • Monitor the urine regularly using a dipstick at home. • The dog must be of a suitable temperament. Provided that these criteria are satisfied, many dogs can have an excellent quality of life and may live for years in this way.
CONTROL OF URINARY FUNCTION The clinician is usually aware of the animal’s preoperative urinary status, but the status can change several times in the days after surgery. For example, an incontinent dog that could be expressed easily before surgery might become impossible to express a few days after surgery. Several days of intermittent catheterization might then be necessary before manual expression again becomes possible using drugs to modify sphincter tone (page 24).
Anatomy and physiology Parasympathetic fibers run in the pelvic nerves, which originate from S1 to S3 spinal cord segments and cause the detrusor muscle of the bladder wall to contract (15.22). The detrusor muscle normally contracts in a co-ordinated fashion to empty the bladder. Each individual smooth muscle cell in the bladder wall is joined to its neighbors by so-called ‘tight junctions’. These junctions have low electrical resistance so that the wave of membrane depolarization spreads rapidly from cell to cell over the whole detrusor muscle. Over-distension of the bladder can disrupt the ‘tight junctions’ so preventing the normal co-ordinated contraction of the detrusor muscle fibers. This should be prevented by keeping bladder volumes low at all times and paying strict attention to assisted emptying of the bladder (see below).
15.21 A cart (K-9 Cart Company, Oak Harbor, WA; Dewey’s Wheelchairs for Dogs, Prineville, OR) designed for use in paraplegic dogs. These are made in a variety of sizes to suit different body weights.
Cortical neurons
UMN-sphincter L 1 2 34 Motordetrusor Sensorydetrusor Sympathetic
S 1 23
Sensorysphincter Motorsphincter
15.22 The physiological control of micturition is complex and involves the integration of spinal cord reflexes with the modulating effect of higher centres. Any lesion in the spinal cord is likely to alter either the reflexes, their higher control, or both. This can then change the bladder’s ability to store urine efficiently and to empty itself completely.
Postoperative care
Sympathetic fibers run in the hypogastric nerves, which originate from L1 to L4 or L5 spinal cord segments, and cause the urethral smooth muscle to contract. Somatic nerve fibers run in the pudendal nerves, which originate from S1 to S3 spinal cord segments and cause the urethral striated muscles to contract. Both the smooth and the skeletal muscle tone of the urethra contribute to maintaining a functional urethral sphincter mechanism. (These will be referred to collectively as the urethral sphincter mechanism unless stated otherwise.)
out when the bladder fills to the point where intravesicular pressure exceeds sphincter pressure causing urinary retention and overflow. After approximately 1 month, a reflexive emptying of the bladder develops. In LMN lesions, the sphincter tone is decreased and so urine will tend to leak continuously (Table 15.6). Occasionally some animals, such as cats with sacrocaudal injuries, have an LMN bladder that is difficult to express because of increased sympathetic tone (O’Brien, 1988).
Disorders of micturition
Pharmacological manipulation of micturition
Micturition disorders are divisible broadly into either upper motor neuron (UMN) deficits affecting the spinal cord proximal to the sacral segments, or lower motor neuron (LMN) deficits affecting the sacral spinal cord or nerve roots. In mild or moderate UMN spinal cord lesions (grade 2 or 3), increased urethral tone may prevent the patient from emptying the bladder fully. Initially, in severe UMN lesions (some grade 4 and all grade 5), the detrusor muscle is paralysed and urine is retained. It leaks
The most common problem is excessive urethral sphincter tone in UMN bladder dysfunction. The patient’s voluntary efforts to void, or attempts at manual expression of the bladder, may be unable to overcome this excessive urethral tone. The result is an increase in the residual volume of urine in the bladder. As it is not usually possible to distinguish smooth from striated muscle effects in any individual patient, the simplest approach is to block the activity of both sphincters (15.23, Table 15.7). Dogs
Table 15.6 Differences in urinary function associated with UMN or LMN lesions UMN (cranial to sacral segments) paraplegic—acute
UMN (cranial to sacral segments) paraplegic—chronic
LMN lesions (sacral segments)
Detrusor function
⫺
⫹
⫺
Striated muscle sphincter tone
⫹ or ⫹⫹
⫹ or ⫹⫹
⫺
Smooth muscle sphincter tone
⫹ or ⫹⫹
⫹ or ⫹⫹
⫹
⫹ Normal; ⫹⫹ increased; ⫺ decreased. LMN, lower motor neuron; UMN, upper motor neuron.
15.23 Innervation of the bladder and urethra. Detrusor muscle Urethral sphincter (smooth and striated muscle)
Somatic sympathetic α Contracts and closes Parasympathetic—contracts Sympathetic β—relaxes
351
352
Small Animal Spinal Disorders
with UMN lesions that have minimal motor function are therefore usually started on diazepam, along with either phenoxybenzamine or prazosin. This prophylactic approach is preferable to waiting for an UTI to develop; the drugs can easily be discontinued once the animal is able to void unaided. Drugs acting on the urinary tract are shown in Table 15.7: • Diazepam is used to reduce the striated muscle tone. It is not recommended in cats; dantrolene can be used as an alternative. • Phenoxybenzamine is used to reduce the smooth muscle tone (sympathetic antagonist). There is usually a delay of 2–3 days before phenoxybenzamine takes effect. The drug is potentially carcinogenic and may be discontinued. • Prazosin is now the preferred alpha antagonist. It is inexpensive but can cause marked hypotension and so animals should first be tried on half the calculated dose and ideally blood pressures are measured for the first few days of treatment (Lane and Thomas, 2000). An alternative is terazosin, which has also been used safely in dogs and cats (R. Schueler, personal communication).
Assisted emptying of the bladder The three crucial roles of assisted voiding are to: • Prevent leakage and urine scald by keeping intravesicular pressure and volume low. • Empty all urine from the bladder periodically to reduce the risk of retention cystitis. • Prevent overstretching of ‘tight junctions’ by avoiding bladder distension. Retention cystitis will develop in any patient that does not empty its bladder completely. No antibiotic, regardless of its potency or spectrum, can prevent this. Use of ‘antibiotic cover’ in these circumstances only succeeds in selecting the most resistant organism (Lees, 1986). Incontinent animals with either UMN or LMN bladder lesions require assisted emptying at least three times daily, with pharmacological assistance recommended for UMN lesions. Some clinicians prefer indwelling catheters to avoid urine leakage and the patient stress that manual expression can cause. Repeated cystocentesis is not recommended as a means of keeping the bladder empty. Bethanechol and possibly cisapride (Lane and Thomas, 2000), may improve voiding in animals with LMN lesions but are not recommended in UMN lesions.
Table 15.7 Drugs acting on the lower urinary tract1 Use
Drug*
Species
Dose
Notes and side-effects
Alpha antagonist
Phenoxybenzamine
Dog Cat
0.25–0.5 mg/kg PO q12 h Per cat—1.25–7.5 mg PO q12–24 h
Hypotension Start at lowest dose Carcinogen? Availability?
Alpha antagonist
Prazosin
Dog Cat
1 mg /15 kg PO q12–24 h Per cat—0.25–0.5 mg PO q12–24 h
Hypotension Start at lowest dose Care in CRF Seizures
Alpha antagonist
Terazosin
Dog Cat
Per dog—1–2 mg up to 11 kg, 2–5 mg up to 50 kg; PO q12 h Per cat—0.5–1 mg PO q12 h
Hypotension Start at lowest dose Priapism, rare, needs immediate treatment
Skeletal muscle relaxant
Diazepam
Dog Cat
Per dog—2–10 mg PO q8 h Not recommended
Give 10–20 min before expression
Skeletal muscle relaxant
Dantrolene
Dog Cat
1–5 mg/kg PO q8–12 h 0.5–2.0 mg/kg PO q8 h
Alternative to diazepam Potential for hepatic toxicity
Urinary antiseptic
Methenamine mandelate
Dog Cat
10–20 mg/kg PO q8–12 h Same or 250 mg PO q8–12 h
Urine pH must be below 6.5 Liver function must be normal
1
Lane and Thomas, 2000. *Consult appropriate source for full list of contraindications and adverse reactions. CRF, chronic renal failure.
Postoperative care
MANUAL EXPRESSION (15.24, 15.25) Probably the best method to try initially for assisted voiding is simple manual expression of the bladder. Drugs that decrease sphincter tone usually make expression easier. It can sometimes be difficult to estimate the completeness of bladder emptying without subsequent catheterization or ultrasound scanning. Occasional use of one of these objective methods for determining bladder volume is useful, but with experience the effectiveness of manual expression can usually be assessed quite well by palpation. The advantage of manual expression is that, in contrast to catheterization, it avoids the introduction of bacteria into the normally sterile bladder. Even if only performed twice daily, with the third emptying done by catheter, this approach is preferable to catheterization each time. The disadvantages of manual expression are that it can be difficult to perform effectively in some animals, especially those with large
15.24 Manual expression of the bladder relies on gentle, continuous, caudal abdominal pressure to overcome urethral resistance.
15.25 Excessive force should not be used to overcome high sphincter resistance; this might damage the bladder wall. It is better to maintain moderate pressure to cause the sphincter to fatigue.
or tense abdomens, those with external splints, or before increased urethral tone has been modified by drugs. Some animals become very difficult to express due to the development of UTI, presumably because it often induces urethral spasm. Animals managed with manual expression are prone to urine leakage and need frequent bathing of their hindquarters to prevent urine scald and decubitus (15.35).
INTERMITTENT ASEPTIC CATHETERIZATION (15.26–15.28) This may prove necessary in some difficult or fractious patients, or in dogs whose sphincter tone has not been modified successfully. Strict attention must be paid to aseptic technique. Even with these precautions, bacteria may still be introduced into the bladder because the
15.26 Aseptic technique is crucial when catheterizing the bladder. The tip of the penis or the vulva is irrigated with sterile saline and then dilute iodophor or chlorhexidine solution. In females a sterile speculum or a gloved finger is used to locate the urethral opening.
15.27 A sterile, soft rubber urinary catheter with sterile lubricating jelly is advanced using a ‘no touch’ technique, holding it instead by a short length of sterile catheter wrapping. The catheter is protected inside its wrapping to prevent contamination prior to insertion.
353
354
Small Animal Spinal Disorders
15.28 Once the catheter is in the bladder, a sterile syringe is used to aspirate urine. If large volumes are anticipated, then a sterile three-way tap or stopcock is useful to allow urine to be emptied from the syringe without repeated disconnection and potential contamination.
15.29 A Foley catheter can be sutured to the vulva and the bulb inflated as shown. The catheter can then be connected to a closed collection system (15.30, 15.31).
distal urethra has a normal flora that includes Staphylococcus intermedius and Escherichia coli in particular (Stone and Barsanti, 1992). Nosocomial organisms may also be introduced by poor technique. Urinary catheters are the second most common source of nosocomial bloodstream infection in people (Maki and Tambyah, 2001); the risk of infection also increases with the number of catheterizations (Lees, 1986; Lulich and Osborne, 1995). Irrigation of the prepuce or vestibule with 0.02% chlorhexidine prior to each catheterization may help to reduce the infection rate (see ‘Closed collection system’, below).
CLOSED COLLECTION SYSTEM (15.29–15.31) This system is useful in incontinent patients with severe UTI, in males that are hard to express manually, and
15.30 A commercially available urine collection container is preferred for the closed technique. This apparatus prevents reflux of urine into the bladder and has a much lower likelihood of introducing bacteria into the urinary tract than IV drip tubing and an empty fluid bag.
15.31 Drip tubing connects this catheter to an empty IV fluid bag. The bag should always be kept lower than the animal; if the animal is moved the tubing should be clamped to prevent reverse flow. These bags are difficult to empty aseptically; laying the bag on the floor increases the likelihood of contamination. The preferred collection apparatus is shown in 15.30.
to measure urine output in patients with reduced renal function. Closed collection has the advantage of permitting diuresis while keeping the bladder empty. Diuresis is useful if the patient has a severe infection as it helps to flush debris and bacteria from the bladder. This technique is also useful for the short-term management of incontinent animals as it keeps the animal clean and dry, which reduces both the development of urine scald and the chances for direct contamination of a surgical wound. Closed collection should ideally be replaced by another means of assisted voiding within 3 days. If a catheter is to be used for longer, it must be replaced every 3 days and use of urinary antiseptics should be considered. The risk of infection can be kept
Postoperative care
15.32 Tube cystostomy is best performed using a Bard Urological Catheter (C.R. Bard, Inc., Covington, GA) with a mushroom tip.
low for the first 4 days if good catheter management is practiced (Smarick et al., 2002); but increases significantly thereafter. The vestibule or prepuce should be irrigated every 8 h with sterile saline and then a very dilute solution of chlorhexidine (0.02%) (Smarick et al., 2002). Nosocomial bacteriuria or candiduria develops in up to 25% of humans requiring a urinary catheter for 7 days or more (Maki and Tambyah, 2001); corresponding rates for animals can be as high as 50% after 4 days (Barsanti et al., 1985).
TUBE CYSTOSTOMY In certain circumstances a prepubic tube cystostomy provides an alternative to repeated catheterization (15.32). An interlocking box suture pattern is recommended to anchor the tube (Daye et al., 1999). Tubes are left in place for a minimum of 7 days; animals are able to void normally after tube removal once the initiating problem has resolved (Williams and White, 1991). Tube cystostomy can also be used occasionally as a longterm option for animals that are permanently incontinent and difficult to express (Smith et al., 1994).
consequence of UTI is transient bacteremia, which may lead to wound infection or sepsis (Barsanti and Finco, 1984). Repeated UTI is common; it may be the presenting problem for animals that fail to recover nociception after a severe injury even though they recovered good motor function (Olby et al., 2003). The potential for increased residual urine and an associated UTI should be recognized in all patients with a lesion severe enough to disturb motor function. In such animals a urinalysis should be performed every 2–3 days on urine collected by cystocentesis. Evidence of inflammation on urinalysis, such as the presence of inflammatory cells or bacteria, warrants: • Urine culture. • Re-assessment of the method for assisted voiding. • Specific antibiotic therapy. In an animal with marked neurological deficits and a severe, established UTI, a 12–24 h period of diuresis combined with continuous evacuation of the bladder into a closed drainage system is recommended. Before urine culture results are available, the initial antibiotics of choice are trimethoprim-sulfa, amoxycillin–clavulanic acid or a cephalosporin. Treatment is maintained for 14 days and repeat culture is performed once the animal has been off antibiotics for 7 days. Multidrug-resistant organisms may be treatable with subcutaneous amikacin or imipenem (Barker et al., 2002). Candida sp. are an important cause of UTI in humans and may account for infections in animals where bacterial culture is negative, especially those on long-term antibiotic therapy (Maki and Tambyah, 2001). Methenamine mandelate (Table 15.7) is a urinary antiseptic agent that is useful in chronic UTI or in patients undergoing intermittent catheterization (Kevorkian et al., 1984; Krebs et al., 1984). The drug is hydrolyzed in the bladder to ammonia and formaldehyde at a pH ⬍6, which usually requires the mandelic acid supplementation (Pearman et al., 1978). It should only be used in animals with normal liver function.
Gastrointestinal disturbances POSTOPERATIVE COMPLICATIONS The most important potential complications are UTI, gastrointestinal disturbances, pancreatitis, surgical wound complications, urine scald and decubital ulcers.
Urinary tract infections Stagnant urine remaining in the bladder after voiding will predispose the patient to UTI. Animals with Cushing’s disease or diabetes mellitus are also predisposed to UTI and animals given corticosteroids may be at increased risk of pyelonephritis (Barsanti et al., 1992). An important
Up to 15% of dogs with disc disease will develop gastrointestinal problems and the mortality rate has been reported to be as high as 2%. The major risk factor in one study was use of dexamethasone; dose and duration of therapy were not important (Moore and Withrow, 1982). Vomiting can have a variety of causes including corticosteroids, antibiotics, NSAIDs or pancreatitis. If vomiting occurs, food, water and non-critical medications should be withheld for 24 h and intravenous fluids given to replace losses. A high index of suspicion should be maintained for pancreatitis (see below) and aspiration pneumonia (especially in recumbent animals). Drugs used to
355
356
Small Animal Spinal Disorders
treat gastrointestinal disturbances are listed in Table 15.8. Persistent vomiting warrants anti-emetic therapy and suggests an underlying cause such as pancreatitis. Diarrhea will tend to increase the risk of UTI or wound infection and every attempt should be made to prevent it, or at least to shorten its severity and duration. Food should be withheld for 12–24 h and then a high fiber diet offered. If this fails, opioids are the preferred anti-diarrheal drugs (Strombeck and Guilford, 1990). Bleeding into the gastrointestinal (GI) tract, presenting as ‘coffee ground’ vomit or melena, should be treated aggressively (8.4). This problem has a high potential mortality rate; compromise of the GI barrier can also lead to bacteremia and sepsis (Epstein et al., 1992). All oral
intake and non-essential drugs are stopped and symptomatic therapy is started. Misoprostol is indicated for NSAID-induced GI ulceration. It would also seem to be the logical therapy for corticosteroid-induced bleeding. However, it does not reduce the incidence of GI hemorrhage caused by corticosteroids; omeprazole, cimetidine and sucralfate are also ineffective (Hanson et al., 1997; Neiger et al., 2000). The lack of a suitable pharmacological agent emphasizes the importance of prevention, supportive care and dietary management. Barium sulfate may work to stop GI bleeding or diarrhea when all else fails. Colonic and gastroduodenal perforation carry the highest mortality rate of all GI complications. Furthermore,
Table 15.8 Drugs acting on the gastrointestinal system1 Use
Drug**
Species
Dose
Notes and side-effects
Antiemetic
Metoclopramide
Dog and cat
0.2–0.5 mg/kg SQ q8 h
Not if GI bleeding, seizures
Antiemetic
Chlorpromazine
Dog and cat
0.5 mg/kg IM q8 h
Hypotension, seizures
Ulceration
Ranitidine
Dog Cat
2 mg/kg PO, IV, q8–12 h 2.5 mg/kg PO, IV, q12 h
More effective than cimetidine
Ulceration
Famotidine
Dog
0.5–1.0 mg/kg PO, SQ, IV q24 h 0.5 mg/kg PO, SQ q24 h
Once daily dosing
Can bind other drugs
Cat
IV in cats not recommended
Ulceration
Sucralfate
Dog and cat
0.25–1.0 g per animal—PO q8 h
Ulceration
Omeprazole
Dog and cat
0.5–0.7 mg/kg PO q24 h
NSAID-induced ulceration
Misoprostol
Dog Cat
1–5 g/kg PO q8 h Unknown
Diarrhea, abortion
Diarrhea
Loperamide
Dog and cat
0.08 mg/kg PO q8 h
Narcotic overdose
Diarrhea
Bismuth salicylate
Dog
0.25 ml/kg PO q6 h
For enterotoxic diarrhea, not with NSAIDs or corticosteroids
Cat
Not recommended
Diarrhea and ulceration
Barium sulfate
Dog and cat
0.5 ml/kg q 24 h, max. 3 doses
Constipation Aspiration
Constipation
Bisacodyl
Dog and cat
5–10 mg PO q24 h per animal
Laxative
Constipation
Psyllium
Dog and cat
2–10 g PO in food q12–14 h
Laxative and stool softener
1
Strombeck and Guilford, 1990; Hart et al., 1997; Plumb, 1999. *Consult appropriate source for full list of contraindications and adverse reactions. GI, gastrointestinal. NSAID, non-steroidal anti-inflammatory drug.
Postoperative care
affected dogs often show none of the classic signs of an acute abdomen making early detection very difficult. Major risk factors are multiple doses of corticosteroids, particularly dexamethasone, or combination of corticosteroids with NSAIDs; enemas may also play a role in colonic perforation (Toombs et al., 1986; Hinton et al., 2002; Reed, 2002). Control of defecation rarely causes a problem, as reflexive emptying will occur periodically even in animals with functional transection of the spinal cord. However, constipation can become a problem. Good hydration status must be maintained but enemas should be used with caution due to the potential risk of colonic perforation associated with this procedure (Toombs et al., 1986). High fiber diets and stool softeners such as psyllium derivatives are safer alternatives (Hart et al., 1997). One potential late complication in dogs that fail to recover after intervertebral disc disease is fecal incontinence; this applies to dogs with UMN as well as LMN injuries (Olby et al., 2003).
Pancreatitis Acute pancreatitis has a high mortality rate. It should be considered along with GI perforation in any neurological patient that develops a sudden onset of vomiting, collapse and pyrexia during its postoperative course. Some dogs show a more insidious onset of signs. High doses of corticosteroids do seem to predispose dogs to pancreatitis, particularly those with neurological disease (Strombeck and Guilford, 1990; Williams, 1995). Diagnosis can be difficult; serum amylase and lipase are not specific but may help if markedly elevated and abdominal ultrasound can also be very useful (Hess et al., 1998). Treatment entails withholding all oral intake for 5 days until the clinical signs are in remission. Intravenous fluid therapy must keep up with fluid losses, which can be dramatic, and urine output should also be monitored. Metabolic acidosis and electrolyte derangements are common and warrant regular blood gas and electrolyte measurements. An anti-emetic such as chlorpromazine should be used at the lowest possible dose to control vomiting. Metaclopramide is not recommended. Procaine penicillin and an aminoglycoside are the antibiotics of choice (Strombeck and Guilford, 1990). The prognosis is guarded.
complication was swelling of the wound (7.12), which affected 7.5% of the dogs. Wound discharge occurred in another 5% but it was not determined what proportion of these was infected (Hosgood, 1992). Many factors play a role in wound infection (Hosgood, 2003) (see Box 6.2). Spinal surgery is classified as a clean surgical procedure; the infection rate for clean procedures is reported as 2.5% (Vasseur et al., 1988). However, a number of factors predispose spinal surgeries to a higher infection rate compared to other clean procedures, including frequent preoperative use of corticosteroids or NSAIDs, obesity, GI ulceration and UTI (Epstein et al., 1992; Hosgood, 1992) (8.4, 15.2). Routine use of intraoperative antibiotic is therefore recommended, especially when surgery lasts longer than 90 min (Vasseur et al., 1988; Novelli, 1999). Postoperative antibiotic is only recommended when intra- or postoperative infection is documented (Kriaras et al., 2000). A sterile adhesive dressing is useful to protect the wound from gross contamination (15.33). This should be kept in place for several days, but the wound must also be checked daily for seroma formation, swelling, redness or discharge. Large seromas should be drained and the wound repaired again to close dead space (15.34). If infection is documented, a culture should be taken from the depths of the wound after aseptic preparation of the skin. Antibiotic therapy should be instituted, initially directed at staphylococcal involvement (cefazolin 5–15 mg/kg IV, IM q8 h, or cefadroxil 20 mg/kg PO q12 h). Severe wound infections should be subject to surgical debridement and irrigation.
Urine scald Urine scald is an important cause of dermatitis and also predisposes to decubital ulcer formation. Any dog that soils itself with urine or feces must have the affected
Wound complications Wound complications have been recorded in as many as 14% of dogs undergoing spinal surgery (36/264). Surgical time over 90 min and use of multifilament, absorbable suture material increase the complication rate; monofilament, absorbable material is therefore recommended for wound closure. The most common
15.33 A sterile, self-adhesive dressing. Use of a bandage does not reduce the incidence of wound complications; it is not known if it lowers the actual wound infection rate (Hosgood, 2003).
357
358
Small Animal Spinal Disorders
15.34 A: Seroma after cervical dorsal laminectomy. This area and the lumbosacral region are at high risk of seroma or hematoma formation and so closure of dead space must be perfect. B: Closed suction apparatus draining a seroma. The protective bandage has been removed for illustration.
A
B
15.35 Early urine scald around the vulva in a dog incontinent due to an L6 fracture managed using an external splint. Regular replacement of the straps and washing and drying the skin is indicated. Desitin (Pfizer Inc., New York) is helpful although ingestion of large amounts can cause zinc toxicity.
area bathed and dried. Emollient cream and a water repellent ointment such as Desitin (Pfizer Inc., New York) are particularly useful (15.35). Animals that are unable to move can be placed on human infant absorbent diapers, which collect urine and prevent it from soaking the skin; these can be weighed if necessary to quantify urine output. An indwelling catheter may be needed to prevent leakage until the urine scald resolves.
Decubital ulcers Decubital ulcers result mainly from unrelieved compression of tissue between a hard surface and a bony prominence (15.36–15.39). Even small paraplegic animals can develop decubital ulcers (15.37). The skin should be kept clean and dry at all times and in recumbent animals the bony prominences should be examined at least daily for the onset of decubitus. The
15.36 The area over the ischiatic tuberosity in the early stages of decubitus formation, to show edema and the onset of hair loss. At a more advanced stage, areas of decubitus may appear simply as a wet area on the hair coat due to the exudation of serum.
skin overlying the shoulder, elbow, rib cage, pelvis, hip, and lateral stifle are the areas most at risk. The hair over high-risk areas must be parted and inspected regularly to look for moisture and hyperemia (Swaim et al., 1996). The early appearance of decubitus is development of erythema, edema and tenderness, followed by serum exudation and alopecia. Skin and subcutaneous tissue loss then develops rapidly. Decubital ulcers may give rise to episodes of bacteremia, with the potential risk of infecting the surgical wound or implant.
Postoperative care
15.37 Some paraplegic dogs, like this Miniature schnauzer, sit compulsively with their pelvic limbs extended rigidly. This posture puts constant pressure on the ischiatic tuberosity leading to decubitus. Foam rubber pads or aluminum splints can be fashioned to prevent continued contact (Coates et al., 1995; Swaim et al., 1996).
An appropriate flooring material (15.3–15.7) is essential to prevent, or at least retard, the onset of decubitus. Resolution of established decubitus is obviously difficult until the inciting cause is eliminated. Relief of pressure is an important principle of treatment, which has been reviewed in detail (Swaim et al., 1996). Neurological patients are at particular risk due to the high potential for urine and fecal soiling, which must be minimized by regular bathing and drying (15.17, 15.20). Prompt removal of devitalized tissue and regular irrigation with an antiseptic solution (such as 0.4% chlorhexidine or Domeboro solution (Domeboro Astringent Solution, Bayer, Morristown, NJ)) are recommended. Surgical debridement and primary closure, possibly using skin flaps or grafts, may be required to resolve an indolent ulcer, even after neurological function returns (Swaim et al., 1996) (15.11).
Miscellaneous
15.38 An advanced decubital ulcer over the greater trochanter that will require aggressive management.
15.39 The same dog illustrated in 15.11 and 15.38 is shown lying on inflatable rings, one placed under the shoulder and one under the hip. An alternative is to fashion a bandage into a similar doughnut shape (Nicoll and Remedios, 1995). These devices are difficult to keep in place for long; they were used to allow this dog to be moved from his waterbed into an area with lots of activity.
A number of other problems can arise during the postoperative period. These include: • Self-mutilation can occur in some animals with paresthesia or absent deep pain (2.26, 15.7). • Some male dogs with severe UMN lesions develop permanent erections; this can also lead to selfmutilation or the exposed penis may simply get traumatized (Olby et al., 2003). Gabapentin (page 343) has been used to treat spasticity in humans with chronic spinal cord injuries (Gruenthal et al., 1997); it may help to reduce the muscle spasm contributing to erection in dogs. Elizabethan collars or other restraint devices may also be helpful. • In humans, prolonged recumbency combined with inactivity increases significantly the risk of deep vein thrombosis and pulmonary thromboembolism. These are almost certainly under-recognized as causes of morbidity and mortality in veterinary neurosurgical patients (Feldman, 1986). Treatment is difficult so every effort should be directed toward avoiding circulatory stasis by providing adequate physiotherapy and intake of fluids (LaRue and Murtaugh, 1990). Aspirin may be protective. • Some problems are more likely to develop after surgery for specific disorders. Myelomalacia develops in some dogs with thoracolumbar disc disease (see page 128); pathological fractures can occur in dogs with spinal tumors; pneumonia and gastric dilation or torsion are more likely in recumbent, tetraparetic dogs after cervical surgery. Some complications can be prevented by a thorough presurgical evaluation (15.40). Finally, an animal’s mental status should not be overlooked in the recovery process. Affection from nursing
359
360
Small Animal Spinal Disorders
15.40 Cervical swelling illustrated by barium esophagram. This was a hematoma that arose 10 days after surgery in a Doberman with von Willebrand disease. The dog presented with intermandibular edema, Horner’s syndrome, laryngeal paralysis, megaesophagus and aspiration pneumonia.
staff, multiple physiotherapy sessions throughout the day, putting the animal in a location where it can see a lot of activity, regular trips outside and frequent contact with owners extending to overnight stays at home where feasible, can mean the difference between success and failure in some animals (15.39).
REFERENCES Ahmad, S.R., Kortepeter, C., Brinker, A., Chen, M., Beitz, J. (2002) Renal failure associated with the use of celecoxib and rofecoxib. Drug Safety: an International Journal of Medical Toxicology and Drug Experience 25, 537–544. Bagley, R.S., Silver, G.M., Connors, R.L., Harrington, M.L., Cambridge, A.J., Moore, M.P. (2000) Exogenous spinal trauma: surgical therapy and aftercare. Compendium on Continuing Education for the Practicing Veterinarian 22, 218–230. Balmer, T.V., Irvine, D., Jones, R.S., Roberts, M.J., Slingsby, L., Taylor, P.M., Waterman, A.E., Waters, C. (1998) Comparison of carprofen and pethidine as postoperative analgesics in the cat. Journal of Small Animal Practice 39, 158–164. Barker, C.W., Stevenson, M.A., Zhang, W., Sanchez, S., Boudinot, F.D., Budsberg, S.C. (2002) Pharmacokinetics and pharmacodynamics of imipenem in dogs. Veterinary Surgery 31, 475–476. Barsanti, J.A., Finco, D.R. (1984) Bacteremia of urinary tract origin (urosepsis). In: R.W. Kirk (ed.), Current Veterinary Therapy, 1150–1153. Philadelphia: WB Saunders. Barsanti, J.A., Blue, J., Edmunds, J. (1985) Urinary tract infection due to indwelling bladder catheters in dogs and cats. Journal of the American Veterinary Medical Association 187, 384–388. Barsanti, J.A., Shotts, E.B., Crowell, W.A., Finco, D.R., Brown, J. (1992) Effect of therapy on susceptibility to urinary tract infection in male cats with indwelling urethral catheters. Journal of Veterinary Internal Medicine 6, 64–70. Berry, W.L., Reyers, L. (1990) Nursing care of the small animal neurological patient. Journal of the South African Veterinary Association 61, 188–193. Bockstahler, B., Lorinson, D., Grossliner, K., Lendl, S. (2002) Effects of a rehabilitation program for dogs undergoing decompressive spinal cord surgery. Veterinary Surgery 31, 293.
Branson, K.R., Ko, J.C.H., Tranquilli, W.J., Benson, J., Thurmon, J.C. (1993) Duration of analgesia induced by epidurally administered morphine and medetomidine in dogs. Journal of Veterinary Pharmacology and Therapeutics 16, 369–372. Brideau, C., van Staden, C., Chan, C.C. (2001) In vitro effects of cyclooxygenase inhibitors in whole blood of horses, dogs, and cats. American Journal of Veterinary Research 62, 1755–1760. Clokie, C., Bentley, K.C., Head, T.W. (1991) The effects of the helium–neon laser on postsurgical discomfort: a pilot study. Journal of the Canadian Dental Association 57, 584–586. Coates, J.R., Sorjonen, D.C., Simpson, S.T., Cox, N.R., Wright, J.C., Hudson, J.A., Finn Bodner, S.T., Brown, S.A. (1995) Clinicopathologic effects of a 21-aminosteroid compound (U74389G) and high-dose methylprednisolone on spinal cord function after simulated spinal cord trauma. Veterinary Surgery 24, 128–139. Crane, S.W. (1987) Perioperative analgesia: a surgeon’s perspective. Journal of the American Veterinary Medical Association 191, 1254–1257. Davies, N.M., Longstreth, J., Jamali, F. (2001) Misoprostol therapeutics revisited. Pharmacotherapy 21, 60–73. Daye, R.M., Huber, M.L., Henderson, R.A. (1999) Interlocking box jejunostomy: a new technique for enteral feeding. Journal of the American Animal Hospital Association 35, 129–134. de Nadal, M., Munar, F., Poca, M.A., Sahuquillo, J., Garnacho, A., Rossello, J. (2000) Cerebral hemodynamic effects of morphine and fentanyl in patients with severe head injury: absence of correlation to cerebral autoregulation. Anesthesiology 92, 11–19. Dohoo, S.E., Dohoo, I.R. (1996) Postoperative use of analgesics in dogs and cats by Canadian veterinarians. Canadian Veterinary Journal 37, 546–551. Epstein, M.D., Banducci, D.R., Manders, E.K. (1992) The role of the gastrointestinal tract in the development of burn sepsis. Plastic Reconstructive Surgery 90, 524–531. Feldman, B.F. (1986) Thrombosis-diagnosis and treatment. In: R.W. Kirk (ed.), Current Veterinary Therapy IX, 505–509. Philadelphia: WB Saunders. Goldstein, J.L., Correa, P., Zhao, W.W., Burr, A.M., Hubbard, R.C., Verburg, K.M., Geis, G.S. (2001) Reduced incidence of gastroduodenal ulcers with celecoxib, a novel cyclooxygenase-2 inhibitor, compared to naproxen in patients with arthritis. American Journal of Gastroenterology 96, 1019–1027. Grisneaux, E., Pibarot, P., Dupuis, J., Blais, D. (1999) Comparison of ketoprofen and carprofen administered prior to orthopedic surgery for control of postoperative pain in dogs. Journal of the American Veterinary Medical Association 215, 1105–1110. Gruenthal, M., Mueller, M., Olson, W.L., Priebe, M.M., Sherwood, A.M., Olson, W.H. (1997) Gabapentin for the treatment of spasticity in patients with spinal cord injury. Spinal Cord 35, 686–689. Hansen, B.D., Hardie, E.M. (1993) Prescription and use of analgesics in dogs and cats in a veterinary teaching hospital: 258 cases (1983–1989). Journal of the American Veterinary Medical Association 202, 1485–1494. Hanson, S.M., Bostwick, D.R., Twedt, D.C., Smith, M.O. (1997) Clinical evaluation of cimetidine, sucralfate, and misoprostol for prevention of gastrointestinal tract bleeding in dogs undergoing spinal surgery. American Journal of Veterinary Research 58, 1320–1323. Hart, R.C., Jerram, R.M., Schulz, K.S. (1997) Postoperative management of the canine spinal surgery patient—Part II. Compendium on Continuing Education for the Practicing Veterinarian 19, 1133–1142. Haskins, S. (1987) Use of analgesics postoperatively and in a small animal intensive care setting. Journal of the American Veterinary Medical Association 191, 1266–1268. Herperger, L.J. (1998) Postoperative urinary retention in a dog following morphine with bupivacaine epidural analgesia. Canadian Veterinary Journal 39, 650–652. Hess, R.S., Saunders, H.M., van Winkle, T.J., Shofer, F.S., Washabau, R.J. (1998) Clinical, clinicopathologic, radiographic, and ultrasonographic abnormalities in dogs with fatal acute pancreatitis: 70 cases (1986–1995). Journal of the American Veterinary Medical Association 213, 665–670. Hinton, L.E., McLoughlin, M.A., Johnson, S.E., Weisbrode, S.E. (2002) Spontaneous gastroduodenal perforation in 16 dogs and seven cats (1982–1999). Journal of the American Animal Hospital Association 38, 176–187.
Postoperative care
Holtsinger, R.H., Parker, R.B., Beale, B.S., Friedman, R.L. (1992) The therapeutic efficacy of carprofen (Rimadyl–VTM) in 209 clinical cases of canine degenerative joint disease. Veterinary Comparative Orthopaedics and Traumatology 5, 140–144. Hosgood, G. (1992) Wound complications following thoracolumbar laminectomy in the dog: a retrospective study of 264 procedures. Journal of the American Animal Hospital Association 28, 47–52. Hosgood, G. (2003) Wound repair and specific tissue response to injury. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 66–86. Philadelphia: Elsevier Science. Janssens, L.A.A. (1992) Acupuncture for the treatment of thoracolumbar and cervical disc disease in the dog. Problems in Veterinary Medicine 4, 107–116. Jerram, R.M., Hart, R.C., Schulz, K.S. (1997) Postoperative management of the canine spinal surgery patient—part I. Compendium on Continuing Education for the Practicing Veterinarian 19, 147–161. Kay-Mugford, P., Benn, S.J., LaMarre, J., Conlon, P. (2000) In vitro effects of nonsteroidal anti-inflammatory drugs on cyclooxygenase activity in dogs. American Journal of Veterinary Research 61, 802–810. Kerwin, S.C., Maudlin, G.E. (2003) Hemostasis, surgical bleeding, and transfusion. In: D. Slatter (ed.), Textbook of Small Animal Surgery, 3rd edn, 44–65. Philadelphia: Elsevier Science. Kevorkian, C.G., Merritt, J.L., Ilstrup, D.M. (1984) Methenamine mandelate with acidification: an effective urinary antiseptic in patients with neurogenic bladder. Mayo Clinic Proceedings 59, 523–529. Krebs, M., Halvorsen, R.B., Fishman, I.J., Santos-Mendoza, N. (1984) Prevention of urinary tract infection during intermittent catheterization. Journal of Urology 131, 82–85. Kriaras, I., Michalopoulos, A., Turina, M., Geroulanos, S. (2000) Evolution of antimicrobial prophylaxis in cardiovascular surgery. European Journal of Cardiothoracic Surgery 18, 440–446. Lam, A.M., Warner, D.S. (2001) Opiates, intracranial pressure, and autoregulation. Anesthesiology 94, 177–178. Lane, I.F., Thomas, W.B. (2000) Diagnosis and management of urinary retention. Veterinary Clinics of North America, Small Animal Practice 30, 25–57. LaRue, M.J., Murtaugh, R.J. (1990) Pulmonary thromboembolism in dogs: 47 cases (1986–1987). Journal of the American Veterinary Medical Association 197, 1368–1372. Lascelles, B.D.X., Butterworth, S.J., Waterman, A.E. (1994) Postoperative analgesic and sedative effects of carprofen and pethidine in dogs. Veterinary Record 134, 187–191. Lascelles, B.D.X., Cripps, P.J., Jones, A., Waterman Pearson, A.E. (1998) Efficacy and kinetics of carprofen, administered preoperatively or postoperatively, for the prevention of pain in dogs undergoing ovariohysterectomy. Veterinary Surgery 27, 568–582. Lascelles, B.D.X., Capner, C.A., Waterman-Pearson, A.E. (1999) Current British veterinary attitudes to perioperative analgesia for cats and small mammals. Veterinary Record 145, 601–604. Lascelles, B.D.X., Henderson, A.J., Hackett, I.J. (2001) Evaluation of the clinical efficacy of meloxicam in cats with painful locomotor disorders. Journal of Small Animal Practice 42, 587–593. Lees, G.E. (1986) Risks of urinary catheterisation. In: R.W. Kirk (ed.), Current Veterinary Therapy, 1127–1129. Philadelphia: WB Saunders. Lulich, J.P., Osborne, C.A. (1995) Bacterial infections of the urinary tract. In: S.J. Ettinger and E.C. Feldman (eds), Textbook of Veterinary Internal Medicine: Diseases of the Dog and Cat, Vols 1 and 2, 4th edn, 1775–1788. Philadelphia: WB Saunders. Maki, D.G., Tambyah, P.A. (2001) Engineering out the risk for infection with urinary catheters. Emerging Infectious Diseases 7, 342–347. Marik, P., Chen, K., Varon, J., Fromm, R., Jr, Sternbach, G.L. (1999) Management of increased intracranial pressure: a review for clinicians. The Journal of Emergency Medicine 17, 711–719. Mathews, K.A., Pettifer, G., Foster, R., McDonell, W. (2001) Safety and efficacy of preoperative administration of meloxicam, compared with that of ketoprofen and butorphanol in dogs undergoing abdominal surgery. American Journal of Veterinary Research 62, 882–888. Moore, R.W., Withrow, S.J. (1982) Gastrointestinal hemorrhage and pancreatitis associated with intervertebral disk disease in the dog. Journal of the American Veterinary Medical Association 180, 1443–1447.
Murtaugh, R.J., Matz, M.E., Labato, M.A., Boudrieau, R.J. (1993) Use of synthetic prostaglandin E1 (misoprostol) for prevention of aspirininduced gastroduodenal ulceration in arthritic dogs. Journal of the American Veterinary Medical Association 202, 251–256. Neiger, R., Gaschen, F., Jaggy, A. (2000) Gastric mucosal lesions in dogs with acute intervertebral disc disease: characterization and effects of omeprazole or misoprostol. Journal of Veterinary Internal Medicine 14, 33–36. Nicoll, S.A., Remedios, A.M. (1995) Recumbency in small animals: pathophysiology and management. Compendium on Continuing Education for the Practicing Veterinarian 17, 1367–1374. Novelli, A. (1999) Antimicrobial prophylaxis in surgery: the role of pharmacokinetics. Journal of Chemotherapy 11, 565–572. O’Brien, D. (1988) Neurogenic disorders of micturition. Veterinary Clinics of North America, Small Animal Practice 18, 529–544. Olby, N.J., Harris, T., Munana, K.R., Skeen, T.M., Sharp, N.J.H. (2003) Long-term functional outcome of dogs with severe injuries of the thoracolumbar spinal cord. Journal of the American Veterinary Medical Association 222, 762–769. Paller, M.S., Manivel, J.C. (1992) Prostaglandins protect kidneys against ischemic and toxic injury by a cellular effect. Kidney International 42, 1345–1354. Parton, K., Balmer, T.V., Boyle, J., Whittem, T., Machon, R. (2000) The pharmacokinetics and effects of intravenously administered carprofen and salicylate on gastrointestinal mucosa and selected biochemical measurements in healthy cats. Journal of Veterinary Pharmacology and Therapeutics 23, 73–79. Pascoe, P.J., Dyson, D.H. (1993) Analgesia after lateral thoracotomy in dogs: epidural morphine vs. intercostal bupivacaine. Veterinary Surgery 22, 141–147. Pearman, J.W., Peterson, G.J., Nash, J.B. (1978) The antimicrobial activity of urine of paraplegic patients receiving methenamine mandelate. Investigative Urology 16, 91–98. Plumb, D.C. (1999) Veterinary Drug Handbook, 3rd edn. Ames, Iowa: Iowa State Press. Popilskis, S., Kohn, D.F., Laurent, L., Danilo, P. (1993) Efficacy of epidural morphine versus intravenous morphine for post-thoracotomy pain in dogs. Journal of Veterinary Anaesthesia 20, 21–25. Reed, S. (2002) Nonsteroidal anti-inflammatory drug-induced duodenal ulceration and perforation in a mature rottweiler. Canadian Veterinary Journal 43, 971–972. Reimer, M.E., Johnston, S.A., Leib, M.S., Duncan, R.B., Jr, Reimer, D.C., Marini, M., Gimbert, K. (1999) The gastroduodenal effects of buffered aspirin, carprofen, and etodolac in healthy dogs. Journal of Veterinary Internal Medicine 13, 472–477. Remedios, A.M., Wagner, R., Caulkett, N.A., Duke, T. (1996) Epidural abscess and discospondylitis in a dog after administration of a lumbosacral epidural analgesic. Canadian Veterinary Journal 37, 106–107. Robinson, T.M., Kruse Elliott, K.T., Markel, M.D., Pluhar, G.E., Massa, K., Bjorling, D.E. (1999) A comparison of transdermal fentanyl versus epidural morphine for analgesia in dogs undergoing major orthopedic surgery. Journal of the American Animal Hospital Association 35, 95–100. Shield, M.J. (1992) Misoprostol: new frontiers; benefits beyond the gastrointestinal tract. Scandinavian Journal of Rheumatology Supplement 92, 31–52. Slingsby, L.S., Waterman Pearson, A.E. (2000) Postoperative analgesia in the cat after ovariohysterectomy by use of carprofen, ketoprofen, meloxicam or tolfenamic acid. Journal of Small Animal Practice 41, 447–450. Smarick, S., Aldrich, J., Foley, J., Kass, P., Ling, G., Fudge, M., Haskins, S. (2002) Catheter-associated urinary tract infections in a small animal intensive care unit. Veterinary Emergency and Critical Care 12, 195. Smith, J.D., Stone, E.A., Gilson, S.D. (1994) Permanent cystostomy catheter as a palliative treatment for urine outflow obstruction from presumed transitional cell carcinoma. American College of Veterinary Surgeons Annual Conference 1994; 426. Smith, L.J., Yu, K.A.J. (2001) A comparison of epidural buprenorphine with epidural morphine for postoperative analgesia following stifle surgery in dogs. Veterinary Anaesthesia and Analgesia 28, 87–96. Still, J. (1989) Analgesic effects of acupuncture in thoracolumbar disc disease in dogs. Journal of Small Animal Practice 30, 298–301.
361
362
Small Animal Spinal Disorders
Stone, E.A., Barsanti, J.A. (1992) Urologic Surgery of the Dog and Cat. Malvern, PA: Lea & Febiger. Strombeck, D.R., Guilford, W.G. (1990) Small Animal Gastroenterology, 2nd edn. Davis, CA: Stonegate Publishing Company. Swaim, S.F., Lee, A.H., Hughes, K.S. (1989) Heating pads and thermal burns in small animals. Journal of the American Animal Hospital Association 25, 156–162. Swaim, S.F., Hanson, R.R., Jr, Coates, J.R. (1996) Pressure wounds in animals. Compendium on Continuing Education for the Practicing Veterinarian 18, 203–218. Taylor, R.A. (1992) Postsurgical physical therapy: the missing link. Compendium on Continuing Education for the Practicing Veterinarian 14, 1583–1593. Toombs, J.P., Collins, L.G., Graves, G.M., Crowe, D.T., Caywood, D.D. (1986) Colonic perforation in corticosteroid-treated dogs. Journal of the American Veterinary Medical Association 188, 145–150.
Vasseur, P.B., Levy, J., Dowd, E., Eliot, J. (1988) Surgical wound infection rates in dogs and cats. Data from a teaching hospital. Veterinary Surgery 17, 60–64. Walker, J. (1983) Relief from chronic pain by low power laser irradiation. Neuroscience Letters 43, 339–344. Walt, R.P. (1992) Misoprostol for the treatment of peptic ulcer and antiinflammatory-drug-induced gastroduodenal ulceration. New England Journal of Medicine 327, 1575–1580. Williams, D.A. (1995) Exocrine pancreatic disease. In: S.J. Ettinger and E.C. Feldman (eds), Textbook of Veterinary Internal Medicine: Diseases of the Dog and Cat, Vols 1 and 2, 4th edn, 1372–1392. Philadelphia: WB Saunders. Williams, J.M., White, R.A.S. (1991) Tube cystostomy in the dog and cat. Journal of Small Animal Practice 32, 598–602.
Index
Note: Numbers in bold refer to illustrations or tables A Abdominal palpation 23–24 Abrasions 167 Abscess formation 51, 184, 122, 212, 225, 229 Accessory process 141, 156 ACE inhibitors 85 Acetaminophen 343 Achilles tendon rupture 20 Acupuncture 96, 343 ADCON-L 87 Addison’s disease 20, 38 Adjacent segment disease 225 Adverse drug reactions analgesics 84–85, 340–342 anesthetics 85 anti-coagulants 82, 99, 333, 342 anti-convulsants 343 anti-emetics 356 anti-infectives 85 anti-inflammatories 83, 84, 85, 341–342 cardiovascular system agents 85, 352 gastrointestinal system agents 356 sedatives 282 Albuminocytological dissociation 45 Allograft 117, 220, 221, 274, 292 see also Fat graft Alpha-antagonists 352 Amikacin 355 Aminoglycoside 357 Amoxycillin 327, 344, 355 Amphotericin B 331 Ampicillin 327 Amputation, tail 297 Analgesia opiates 84–85, 340–341, 343, 356 palliative 254 postoperative care 131, 339–343 preoperative assessment 84–85 thoracolumbar disc disease 123, 131 Anal reflexes 183 Anal sac adenocarcinoma 184
Anaphylaxis 85 Anatomical location of the injury 295–297 Anatomical specimens 73, 74 Anatomy blood supply 14–17 nervous tissue 1–6 skeleton 6–14 urinary system 350–351 Anesthesia complications 86 CSF collection 64 induction 85 maintenance 85 premedication 85 recovery 85–86 side-effects 85 Angiotensin converting enzyme (ACE) inhibitors 85 Angled bur guard 77 Anomalous spinal disease 36 Antibiotics see also specific drugs drug side-effects 85, 355 prophylactic 83, 309, 327, 331, 355, 357 wound infection 357 Anticlinal vertebra 9 Anti-coagulant drugs 82, 99, 333, 342 Anti-convulsants 343 Anti-emetics 356, 357 Anti-infective agents 85 Anti-inflammatory drugs see also Corticosteroids; Non-steroidal anti-inflammatory drugs cervical disc disease 96 CSM 218 hypervitaminosis A 326 lumbosacral disease 188 palliative 254 side-effects 83, 85, 341–342 thoracolumbar disc disease 123 ventral slot surgery 102 Antiseptic, urinary 352 Anulus fibrosus approach to ventral neck 115
bulging 110, 124–125, 219 CSM 219, 234, 236 fenestration 154, 202 mini-hemilaminectomy 159 pediculectomy 159 removal 234 spinal fracture 288–289 thoracolumbar disc disease 124–125 Aorta 10 Aortic embolism 332–333 Apnea 82 Arachnoid cyst 321, 323–326 diagnosis 29, 36, 122 localization of signs 37 marsupialization 244 myelographic abnormalities 51 postoperative complications 297, 300, 301 spinal 212 subarachnoid 324 Arrhythmia, heart 297 Arteritis 329 Articular facet fracture 287–289 Articulations, synovial 9–11 Ascending motor tracts 6 Ascending myelomalacia 36, 37, 332 Ascending sensory tracts 5 Aseptic catheterization (intermittent) 353–354 Aseptic necrosis 259 Aspergillus infection 42, 328 Aspiration 167 Aspirin analgesia 342 bleeding disorders 99, 342 ischemic neuromyopathy 333 platelet dysfunction 82, 342 Assessment, clinical 81–82 brachial plexus 30–31 cranial nerves 20, 21, 25 DAMNIT formula 21, 35, 36 differential diagnosis 20 etiology 20–28 eyes 21 functional 20–22, 25–29, 184
364
Index
Assessment, clinical (contd.) grading 124, 125 history taking 19 lesion localization 21, 38–39 lesions 28–32 lumbosacral plexus 30–31 neurological 20–28 observation 20, 21 palpation 21, 22–25, 26 physical 19–20 Ataxia 167, 169 Atlantoaxial joint 46–47 Atlantoaxial subluxation (fracture) 161–180 bone cement 177–178 cats 169–170 clinical signs 37, 161–163 complications of treatment 166–167 cross-pins 180 CSF collection 43 decision-making algorithm 165 diagnosis 36, 37, 94, 163–164, 212 dorsal stabilization 166–167 fixation 166–180, 171–176 multiple ventral implants 177–178 myelographic abnormalities 51 non-surgical treatment 164–165, 167 postoperative care 169 prognosis 169 ventral fusion 166–168 wire fixation 178–180 Atlantooccipital dysplasia 37 Atlantooccipital space 71 Atlas (C1) 7, 14, 106, 161 Atropine 85 Axis (C2) 7, 14, 174 Azithromycin 331 Azotemia 82 Azygous vein 15, 16 B Bacterial infections see also specific organisms CSF analysis 45 postoperative complications 355 spinal cord disease 330 Balance training 348 Barium sulfate 356 Bartonella 42, 330, 331 Basset hound (atlantoaxial subluxation) 161 Bathing (postoperative) 343, 349 Baylisascaris 330
Beagle chondrodystrophy 13 meningitis 329 Bedding (postoperative) 346 Bence–Jones proteinuria 42 Bernese mountain dog (meningitis) 329 Bethanechol 352 Bile duct rupture 19, 282 Biomechanics of spinal fixation 286–290 Biopsies core 79 needle 61–62, 275 tumors 61–62, 251–252, 275, 276 Bisacodyl 356 Bismuth salicylate 356 Bladder assisted emptying 131, 343, 344, 352, 353 dysfunction 351–352 innervation 351 leakage 20 neurological screening 20 rupture 282 Blastomycoides dermatidis 330 Bleeding disorders see also Von Willebrand disease aspirin-induced 99 bleeding time test 217 blood clots 99 cervical disc surgery 99 cervical myelopathy 82 CSF analysis 65 CSM 217, 224, 225, 229, 234 epidural 57 gastrointestinal 83, 84, 229, 342, 356 Hemoclips 78 hypoventilation 82 laminectomy 98, 138, 146 neurovascular bundles 241 preoperative assessment 213, 217 prevention 138 respiratory 82 spinal cord 51, 94, 281, 333 subdural 57 surgical complications 98, 128, 224, 225, 229 thyroid hormone supplementation 217 venous plexus 99, 146, 234 Von Willebrand factor 42, 82, 217, 229, 360 Blood cell counts CSF 44, 45 routine laboratory analysis 41 white blood cells 42, 329
Blood contamination 44 Blood gas analysis 42 Blood supply 14–17 see also Vascular system Body position CSF collection 66 head position 29, 94 imaging 46, 47, 54, 55, 56 patient examination 20–25, 25–28 paw position 20, 22 perception of 5, 6, 22, 23, 348 position for radiography 46 postoperative 225, 228–229, 343, 344, 345–346 Body weight (cement bar thickness) 314 Bone diseases CSM 211, 212 DISH 326 lumbosacral 183, 184 mimicking spinal disease 20 neurological patient 81 osteoarthritis 20 osteochondritis dissecans 20, 36, 37, 183 osteochondritis 20, 36, 37, 183 osteoporosis 38, 122, 326 scintigraphy 59–60 tumors 94, 248, 249, 260 Bone physiology composition 7 demineralized matrix 292 production 209 Bone surgery allograft 117, 220, 221, 274, 292 atlantoaxial subluxation 168, 177–178 autograft 117 bleeding from cancellous bone 142 bone vessel plug 77 bone wax 77, 142, 313, 314 cement 238, 314 CSM 223, 232, 235, 238–240, 245 decompression 117, 203, 243 external fixation 290–292 facetectomy 203 fragments 51 fusion 234 hemilaminectomy 1, 143, 268, 269 infection risk 301 instrumentation 77, 79 laminectomy 195, 198, 200, 208, 272 lumbosacral disease 191, 198, 200
Index
Bone surgery (contd.) metal and cement fixation 313, 314 methylmethacrylate 78 rongeurs 76 trauma 291–292 tumors 255–256 vertebrectomy 274 Botulism 38 Boxer dog proprioception disorder 22 spondylosis deformans 320 Brachial plexus anatomy 2 assessment 30–31 avulsion 30, 37 331 miosis 25 myelopathy 36 nerve sheath tumor resection 275, 276 spinal trauma 282, 283, 302 Brain herniation 65, 66 Brainstem lesions 29, 212 Brain tumor 38 Brucella canis 42, 327, 330, 331 Buccal mucosal bleeding time 217 Bulldog 322 Bupivacaine 341 Buprenorphine 340, 341 Bursitis 94 Butorphanol 340 Buttress, vertebral 287–289 C Cage rest 125, 126, 131, 228, 288, 305 Calcinosis circumscripta 212 Calcinosis (tumoral) 37, 122, 326 Calcium metabolism 42, 212, 326 Calculi 38, 122 Callus encroachment 297, 299 Candida 355 Canine distemper virus (CDV) 42, 329–330 Carcinoma 248 Cardiovascular disorders arrhythmia 282, 297 bradycardia 86 cardiac arrest 167 cardiomyopathy 213 cardiopulmonary arrest 99 differential diagnosis 38 disorders mimicking spinal disease 20 endocarditis 20 hypertension 20 insufficiency 20 neurological patient 81
tachycardia 85 vascular diseases 20, 36, 329, 332–333 Carprofen 341, 343 Cartilaginous exostoses 36, 323 Cart (paraplegic) 350 Catheters 353–354 Cats analgesic drugs 340, 343 anti-inflammatory drugs 342 atlantoaxial subluxation 169–170 cervical disc disease 35, 103 feline immunodeficiency virus 81, 262 feline infectious peritonitis 35, 330 feline leukemia virus 81, 262, 263 gastrointestinal tract 356 Horner’s syndrome 25 hypervitaminosis A 326 ischemic myelopathy 37, 38 ischemic neuromyopathy 281, 332–333 lymphoma 263 Manx 322 pharmacology 340, 342, 343, 352, 356 sacrocaudal dysgenesis 322 spinal trauma 36, 37, 169, 287, 297, 301–302, 305 thoracolumbar disc disease 10, 127, 128, 129, 133 tumors 261–263 urinary tract 302, 352 Cauda equina compression 181–182 disc fenestration 202 functional anatomy 3 lumbosacral laminectomy 199, 200 nerve roots 3 neuritis 36, 37, 184 Caudal vertebra lumbosacral disease 183 sacrocaudal dysgenesis 322 tail injuries 296–297 Cauterization 75, 83, 275 Cavalier King Charles spaniel (syringomyelia) 322–323 Cefadroxil 357 Cefalosporin 83, 85, 355 Cefazolin 83, 327, 357 Cellulose surgical spears 77 Cement plugs CSM 218, 219, 220–223, 226, 237–239 ventral decompression 117
vertebral distraction–stabilization 221–223 Central cord syndrome 29 Cerebellar disease 29 Cerebellomedullary cistern (CMC) CSF collection 64, 65–68, 85 injection of contrast medium 53, 71 Cerebrospinal fluid (CSF) cell counts 44, 45 cervical disc disease 94, 99 collection 43, 64–69, 85 examination 44, 45 functional anatomy 4–5 meningitis 329 neutrophils 43 sedimentation chambers 44 thoracolumbar disc disease 122 ventral recumbancy 69 xanthochromia 45 Cervical disc disease 93–120 see also Cervical spondylomyelopathy cats 103–104 diagnosis 35–37, 94–96 clinical signs 37, 93–94 decision-making algorithm 97 palpation 24 dorsal decompression 98 laryngeal nerve paralysis 99 non-surgical treatment 96 pain 37, 94 postoperative care 102 prognosis 102–103 surgical treatment approach to ventral neck 106–109 complications 98–102 distraction–stabilization 117–118 dorsal decompression 98 dorsal laminectomy 120 fenestration 96, 109–110 hemilaminectomy 119, 266–270 indications 96 locking plate 220 paramedian approach to ventral neck 232–233 postoperative care 102 ventral decompression 96–98, 111–118 Cervical fibrotic stenosis 212 Cervical spine blood supply 15 diagnosis 35–37 anatomical location of injury 295–297 myelography 38, 70, 72 radiography 46, 49 ligaments 14
365
366
Index
Cervical spine (contd.) spinal cord segments 3 stenosis, fibrotic 36, 37 Cervical spondylomyelopathy (CSM) 211–246 see also Wobbler syndrome cement plug 218, 237–239 CSM 218 diagnosis 35, 36, 94, 71–72, 212–216, 281 decision-making algorithms 223–224 neurological examination 20 hypothyroidism 217 non-surgical treatment 218 surgical treatment 218–224 complications 224–228 dorsal decompression 223, 241–245 dorsal laminectomy 218 indications 219 laminoplasty 223–224, 245–246 metal and cement fixation 218, 239–240 paramedian approach to ventral neck 232–233 postoperative care 228–229 presurgical evaluation 216–217 prognosis 229 ventral decompression 218, 219–221, 233–235 vertebral distraction 221–223, 235–237 Cervical tumors 35, 37 Cervical vertebrae 7–8 C1/C2 7, 14, 106, 161,174 C1/C5 15, 28–29, 36, 37–38, 212 C1/C8 1 C5/C7 8 C5/C6 212 C6/C7 52, 211 C6/T2 15, 28, 28, 36, 212 Chemical shift artifact 2 Chemotherapeutic agents 254, 262–263 Chest harness 237, 239 Chihuahua (atlantoaxial fracture) 161 Chloramphenicol 331 Chlorhexidine 355 Chlorpromazine 356, 357 Chondrodystrophoid breeds 2–13, 93 Chondroid metamorphosis 12 Chondrosarcoma 248 Chronic degenerative radiculomyelopathy (CDRM) 319 Cicatrix formation 287 Cimetidine 356
Cisapride 352 Cisplatin 254 Claudication, intermittent 38, 334 neurogenic 36 Clavulanic acid 344, 355 Cleanliness, postoperative 343–344 Client communication 88 Client consent form 88 Clindamycin 327, 331 Closed collection system 354–355 Cloxacillin 327 Coccidioides imitis 330 Cocker spaniel kyphosis 131 thoracic tumor 250 Codeine 340 Cod-piece 307 Collar (head) 237, 239, 295 Column splitting 52 Computerized tomography (CT) imaging 55–57 see also Magnetic resonance imaging atlantoaxial subluxation 164 body position 55, 56 cervical disc disease 95–96 comparison to MRI 57 CSM 214–216 disc extrusion 56, 57, 94 dorsal recumbency 55 neoplasia 251 preoperative assessment 82 spinal cord expansion 57 thoracolumbar disc disease 56, 123 trauma 284–285 Confinement (cage) 125, 126, 131, 228, 288, 305 Congenital diseases arachnoid cyst 122, atlantoaxial instability 281 myopathy 38 vertebral anomalies 37, 212, 321–322 Consent form 88 Conservative treatment anti-inflammatory drugs 96 atlantoaxial subluxation 164–165, 166–167 cervical disc disease 96 complications 166–167 CSM 218 decompression 124, 125 fenestration 124, 125 immobilization 282 lumbosacral disease 188, 192 recovery times 124, 125 rest 123, 124, 125, 192, 282
Constipation 356 Contrast meda absorbtion 53 discography 50 epidural leakage 71 epidurography 50 hand position 70 iohexol 43, 49, 86 myelography 49, 72 poor filling 53 spinal cord parenchyma 54 sterile meningitis 42–43 Coonhound paralysis 38 Cord dorsum potentials 61 Corpectomy 58, 149, 147 Corticosteroids see also Antiinflammatory drugs canine distemper virus infection 330 CSM 218 drug side-effects 355, 356, 357 fungal infections 331 GME 330 meningitis 329 meningomyelitis 331 MPSS 83–84, 133, 217, 257, 282, 293, 332, 342 osteoporosis 326 preoperative assessment 83–84 thoracolumbar disc disease 125, 126, 131, 133 wound infection 83, 357 Costs of surgery 88 Coupage 344 Coxofemoral arthritis 184 Cranial nerves 20, 21, 25 Creatine kinase (CK) 42 Creatinine 82 Crossed extensor reflex 32 Cross-pin fixation see also Pin fixation; Screw fixation atlantoaxial subluxation 180 dorsal 180 spinal trauma 282 Cruciate ligament rupture 20, 184 Cryoprecipitate, bleeding disorders 217 Cryptococcus neoformans 42, 330 Curettage 236 Curettes 75, 79 bone graft collection 79 House 75, 144 Shea 75 Cushing’s disease osteoporosis 326 preoperative assessment 82 wound infection risk 83
Index
Cushing’s syndrome 81 Cutaneous trunci reflex 20, 24, 26–27 Cuticle bleeding time 217 Cyclosporin 85 Cystocentesis, postoperative 297 Cysts arachnoid 29, 37, 51, 122, 212, 297, 300, 301, 321, 323–326 cystocentesis 297 dermoid sinus 184, 212, 323 epidermoid 36, 212, 323 kidney 55 marsupialization 244 subarachnoid 324 synovial 36, 37, 52, 94, 122, 183, 215, 245, 320 tube cystostomy 355 Cytosine arabinoside 254 D Dachshund cervical disc extrusion 93 chondroid metamorphosis 12 hemilaminectomy 344 L4/S3 lesions 38 thoracolumbar disc disease treatment 126 thoracolumbar mid-bodies 10 ventral slot decompression 116 Dalmatian (L2/L7 defects) 28–29 DAMNIT scheme 21, 35, 36 Dantrolene 352 Decision-making algorithms atlantoaxial subluxation 165 cervical disc disease 97 cervical spondylomyelopathy 223 lumbosacral disease 189 neoplasia 256 thoracolumbar disc disease 124 trauma 293, 294 Decompression (dorsal) for CSM 218, 219 see also Ventral decompression Decubital ulcers 339, 347, 358–359 Decubitus 127, 167, 297 Deep pain sensation see also Nociception; Pain after durotomy 293 after hemilaminectomy 150 lack of 293 malacia 293–294 neurological examination 21 spinal fracture 87 thoracolumbar disc disease 125 trauma 293–295, 301–302 Deep vein thrombosis (DVT) 127, 359
Degenerative disc disease 36, 183, 184, 281, 320 Degenerative leukoencephalomalacia 320 Degenerative myelopathy 36, 37, 209, 212, 319, 319–320 Degenerative neuroaxonal dystrophy 320–321 Dehydration (fluid intake) 85, 343, 344–345, 355 Demineralized bone matrix 292 Dens hypoplasia 164 ligaments 14 malformation 164 normal CT 7 Dental instruments 76 Depression (post-surgical) 339 Dermatitis 83, 167 Dermatofibrosis 38 Dermoid sinus 36, 37, 184, 212, 323 Descending motor tracts 5–6 Desitin 358 Desmopressin (DDAVP) 217 Detrusor function 351 Dexamethasone 65, 83 Diabetes mellitus 81, 83 Diagnosis, differential cervical spine 35–39 CSM 281 lumbar spine 15, 37 lumbosacral spine 35–39, 184, 184 myelopathy 184 neurological examination 20 spinal trauma 184 synovial cyst 36, 94 thoracolumbar spine 35–39, 122 Diagnostic techniques 41–72 diagnostic imaging CT scan 55–57 MRI 57–59 myelography 51–54, 70–72 radiography 45–50 radiology 50–51 scintigraphy 59–60 ultrasonography 54–55 electrophysiology cord dorsum potentials 61 electromyography 60–61 F waves 61 spinal cord evoked response 60–61 laboratory analysis biochemistry 41–42 biopsy 61–62 CSF analysis 42–45, 64–69 hematology 41
microbiology 42, 45 serology 42 urinalysis 42 Diaphragm herniation 282 paralysis 82 Diarrhea 339, 356 Diazepam analgesia 85, 96, 102, 343 lower urinary tract effects 352 micturition disorders 352 post-myelographic seizure prevention 216 Disc anatomy 12–13 Disc degeneration 12, 36, 183, 184, 281, 320 Disc extrusion (bulge) see also Intervertebral discs; specific spinal regions age related 59 atlantoaxial 163 cervical 93, 94, 99 definition 13 diagnosis 56, 57, 94, 212 dorsolateral hemilaminectomy 144–145 Hansen type 13 intervertebral foramen 96 protrusion 59 surgical complications 127, 129, 131 surgical indications 219 thoracolumbar 122, 127, 129, 131, 136 Disc herniation see also Intervertebral discs CSM 211, 219 dorsal hemilaminectomy 119 Hansen type 12, 37 lumbosacral 186, 200 myelographic abnormalities 51 surgical indications 219 terminology 58 thoracic 123 thoracolumbar 36 traumatic 285 Disc mineralization 59 Discogenic pain 184 Discography 50 Discospondylitis 326–328 atlantoaxial subluxation 163 CSM 212, 225, 228 differential diagnosis 36, 122 fungal infection 328 localization of signs 37 lumbosacral disease 183, 184 myelographic abnormalities 51
367
368
Index
Disc trauma 36, 51 Disseminated idiopathic skeletal hyperostosis (DISH) 36, 326 Distraction–stabilization 117–118, 235–237 Doberman bleeding disorders 217 chronic active hepatitis 213 CSM 36, 211–212, 214–215, 219–220, 222, 243 epidural mass 329 hypothyroidism 217 intramedullary lesion 53 MRI of cervical spine 2 seizures during myelography 86 Staphylococcus intermedius 327 ventral decompression 344 von Willebrand disease 359–360 Dogs (general) see also specific breeds anticonvulsants 343 anti-inflammatory dose 342 carts 350 gastrointestinal tract pharmacology 356 narcotic analgesic dose 340 physical therapy 348 postoperative nursing plan 343, 344 urinary tract pharmacology 352 water therapy 349 Domino lesions after fracture repair 300 after ventral slot surgery 117, 220, 221, 228 cement plug 221–222 Wobbler syndrome 101 Dorsal funiculi 1 Dorsal longitudinal ligament 14 Doxorubicin 254 Doxycycline 331 Draping 74, 78–79 Dressings 344, 357 Drug interactions 85, 342 Drug side-effects ACE inhibitors 85 alpha-blockers 352 anesthetics 85 anti-coagulants 82, 99, 333, 342 anti-convulsants 343 anti-emetics 356 anti-infective agents 85 anti-inflammatory agents 83, 85, 341–342 barium sulfate 356 Baytril 85 bisacodyl 356 buprenorphine 340
butorphanol 340 codeine 340 dantrolene 352 fentanyl 84–85, 341 glycopyrrolate 85 halothane 85 iohexol 86 ivermectin 85 ketoconazole 85 ketoprofen 342 loperamide 356 metronidazole 85 misoprostol 356 morphine 340, 341 MPSS 84 nitrous oxide 85 non-steroidal 83, 85, 341–342 opiates 84–85, 340, 341 oxymorphone 340 phenoxybenzamine 352 prazosin 352 sedatives 282 steroids 83, 85, 326, 357 sucralfate 356 terazosin 352 trimethoprim–sulfonamide 85 Dura dorsolateral hemilaminectomy incision 148 herniation 191 lesions 51 stay suture 279 tears 36, 122, 281 Durotomy dorsolateral hemilaminectomy 148 patients with no deep pain sensation 293 preoperative assessment 87 thoracolumbar disc disease 132, 149 Dysgenesis, sacrocaudal 322 Dyspnea 82, 99, 167 Dysplasia antlantooccipital 36, 37 myelodysplasia 36, 212, 322 Dysraphism (spinal) 36, 212, 322
F waves 61 gastrocnemius muscle 60 neoplasia 250–251 spinal cord evoked response 60–61 Electrosurgery cautery 83 instruments 75 Embolism aortic 332–333 fat 297 fibrocartilaginous 35, 37, 212, 281, 332 Empyema, epidural 36, 328–329 Endocarditis 20 Endotracheal tube 85 End-plate failure 225, 227, 238 Enrofloxacin 85 Ependymoma 51, 248, 252 Epidermoid cyst 36, 212, 323 36 Epidural abscess 51, 122, 212, 225 Epidural empyema 328–329 Epidural fibrosis 183 Epidural hemorrhage 57 Epidural leakage 49, 53, 71 Epidural lipomatosis 36 Epidural steatitis 330–331 Epidurography 50 Epineurium 276 Erections 359 Erythrophagocytosis 45 Escherichia coli 327 Esophagus disease 98, 99 Etodolac 341 Euthanasia 229, 293 Euthyroidism 217 Evoked response, spinal cord 60–61 Exercise intolerance 36–37, 222 Exostoses, cartilaginous 36, 323 Extensor postural thrust 23 External fixation 237, 316–318, 328 External splints 239, 290, 305, 306 Extradural tumors 248–250, 259–263 Eye function 21
E
F
Edema 99 Ehrlichia 38, 42, 330 Electric drill 77 Electrocardiogram (ECG) 213, 282 Electrocautery 83 Electromyography (EMG) 60–61 Electrophysiology cord dorsum potentials 61
Facetectomy 192, 202 Facet joints fracture 191, 207 pain 36, 122, 184, 320 removal 130 Facial sensation 20 Famotidine 356 Fat embolism 297
Index
Fat graft autogenous fat 86–87 dorsolateral hemilaminectomy 149 fenestration of disc extrusion 201 foraminal decompression and facetectomy 202 free fat 86–87 laminectomy 208 minimization of scarring 86–87 necrosis 38 postoperative complications 127, 128, 129 thoracolumbar disc disease 127, 128, 129 Fecal incontinence 131, 183 Feline diseases see also Cats thoracolumbar disc disease 127, 128, 129 FeLV-associated lymphoma 262 feline immunodeficiency virus (FIV) 81, 262 feline infectious peritonitis (FIP) 35, 330 feline leukemia virus (FeLV) 81, 263 Femoral nerves 127 Fenestration and disc removal bulging disc 202 cervical disc disease 96, 109–111 comparison to ventral slot 97 CSM 221, 227 fenestration 127, 154–159 hemilaminectomy 149, 159 instrumentation 75 laminectomy 201 pediculectomy 159 postoperative complications 130, 221, 227 thoracolumbar disc disease 124, 127, 130, 154–159 Fentanyl 84, 340, 341 Fever 339 Fibrocartilaginous embolism (FCE) 35, 36, 37, 212, 281, 331–332 Fibroid metamorphosis 12 Fibrosarcoma 38, 248, 249, 260 Fibrosis cervical 36, 37, 212 epidural 183, 190 peridural 131 restrictive 225, 273 Fine needle aspiration 61 FIP (feline infectious peritonitis) 35, 330 Fistula, cutaneous 127 Fixation devices biomechanics 289–290
external 290–291 internal 291 multiple implant 166 Fixation–stabilization atlantoaxial subluxation 166–167 caudal disc extrusion 96 cervical disc disease 117–118 Flexor (withdrawal) reflex 27 Floating limb gait 28, 212 Flooring (cage) 345–346, 359 Fluconazole 331 Fluid intake 85, 343, 344–345, 355 Focal granulomatous meningoencephalomyelitis (GME) 248 Food intake 343, 344, 356 Forage (CSM surgery) 221, 232, 235 Foramen magnum 179 Foraminal decompression 192, 202 Foreign body migration 36 Free radicals 84 Fungal infection 42, 328, 330 Fusion–fixation atlantoaxial subluxation 166, 167–168 lumbosacral disease 189, 192 new bone production 209 sacral subluxation 204 F waves 61 G Gabapentin 343 Gadoteridol 58 Gagging 20 Gait abnormalities disconnected 212 ‘floating’ limb 28, 212 neurological examination 20 Gallstones 38, 122 Gastrocnemius muscle 60 Gastrointestinal tract disturbances bleeding 83, 84, 342, 356 diarrhea 339, 356 drug-induced 83, 84, 341, 342, 356 melena 339, 356 parasites 38, 122 perforation 83 pharmacology 356 postoperative 339, 355–357 ulceration 127, 297, 341, 356 vomiting 339, 356 Gelatin sponge 78 Gelfoam 36, 78, 86, 99, 269 Gentamicin 327
German shepherd chronic degenerative radiculomyelopathy 319–320 fungal discospondylitis 328 lumbosacral disease 38, 185, 186, 190 nuclear bone scan 60 sacral subluxation 182 thoracic tumor 250 German short-haired pointer 4 Giant breed dogs (CSM) 211, 222, 223 see also Large breed dogs; specific breeds Glioma myelographic abnormalities 51 poorly differentiated 252 tumor classification 248 Glucocorticoids 217 Glucose metabolism 20, 38 Glycopyrrolate 85 GME (focal granulomatous meningoencephalomyelitis) 248 Golden retriever bulbous dilation of dorsal arachnoid space 324 left hemiparesis 29 myelomalacia 61 thoracolumbar spinal cord segments 3 vertebral bone loss 61 Golf-tee pattern 52 Gracilis contracture 20, 184 Grafts bone 117, 220, 221, 274, 292 fat 38, 86–87, 127, 128, 129, 149, 201, 202, 208, Granulomatous meningoencephalomyelitis (GME) 330 Great Dane CSM 36, 211, 213, 222–224, 228, 246 dyspnea after imaging 82 hypothyroidism 217 postoperative complications 228 Gunshot injury 36, 331 H Halothane 85 Halti head collar 237, 239, 295 Hansen-type disc herniation 12, 13, 182, 183 Harnesses 218, 237, 239, 347 Headlight 73, 74 Head movements 29, 94
369
370
Index
Heart disease arrest 167 cardiomyopathy 213 cardiopulmonary arrest 99 differential diagnosis 20, 38 neurological patient 81 rhythm disorders 85, 86, 282, 297 Heat therapy 347 Helminths 330 Hemangiosarcoma 248, 261 Hematoma cervical disc surgery 99 CSM 212, 225, 229 extradural 99 postoperative complications 99, 225, 229 prognosis 284 spinal cord 36, 37 212, 333 spinal trauma 284 Hemilaminectomy see also Minihemilaminectomy cervical disc disease 98, 119, 266–270 dorsolateral 136–150 marsupialization 326 neoplasia 266–270 nerve sheath tumor 275–279 spinal trauma 294 thoracolumbar disc disease 125, 126, 151–154 Hemiparesis 94 Hemivertebra 320–321 Hemoclips 78 Hemocytometer 44 Hemogram 41 Hemorrhage see also Bleeding disorders aspirin-induced 99 cervical 82, 99 CSF analysis 65 CSM 224, 225, 229, 234 diaphragm paralysis 82 epidural 57 gastrointestinal 229 Hemoclips 78 hypoventilation 82 prevention 138 psoas muscle 38, 281 respiratory 82 spinal cord 51, 94, 281, 333 subdural 57 surgical 98, 128, 138, 146, 224, 225, 229 thoracolumbar 128 venous plexus 99, 146, 234 Hemostasis 81, 82 Hemostat 77
Hepatic disease 81, hepatitis (chronic) 213 hepatotoxicosis 85 Histoplasma capsulatum 330 History taking 15, 19 Hock flexion 27, 30, 182 Hopping 20, 22 Horner’s syndrome cats 25 cervical disc surgery 99 cervical tumor 251 CSM 224 Horses (Wobbler syndrome) 218 Hounsfield units (CT numbers) 283 Hydration 85, 343, 344–345, 355 Hydrocephalus 162–163 Hydromorphone 340 Hydromyelia 212, 322–323 Hydrosorb TS 87 Hydroxyurea 254 Hyoid venous arch 171 Hypercalcemia 42 Hyperesthesia 20, 21, 183, 184 see also Nociception; Pain Hyperflexion injury 99, 227 Hypergammaglobulinemia 42 Hyperkalemia 20, 38 Hyperreflexia (pseudo) 183 Hypertension 20 Hyperthyroidism 20, 38 Hypertrophic osteodystrophy 20 Hypervitaminosis 36, 37, 326 Hypocalcemia 42 Hypodermic needles 76 Hypoglycemia 20, 38 Hypokalemia 20 Hypothyroidism 81, 82, 213, 217 Hypoventilation 82, 99 I Iatrogenic injury cervical disc surgery 99 CSM 224, 225, 226 intraoperative complications 224 lumbosacral disease 191 neural 225, 226 surgical complications 99, 191, 225, 226 Idiopathic disorders 38, 326 Iliac thrombosis 38, 332–333 Imipenem 355 Immobilization 282 Implants see also Locking plates; Metal and cement fixation atlantoaxial subluxation 177–178
CSM 220, 224, 225, 226, 238, 239–240 failure 167, 191, 192, 224, 225, 226, 258–259, 299, 318 intervertebral 220 lumbar spinal injury 296 lumbosacral disease 191, 192 metal and cement fixation 239–240, 296 migration 300 multiple ventral implants 177–178 screw fixation 192 surgical complications 191, 224, 225, 226 thoracic spinal injury 296 threaded 292 titanium 255 wound infection risk 83 Infections see also Wounds and wound infections; specific organisms antibiotic prophylaxis 83 discospondylitis 36, 326–328 epidural empyema 328–329 infectious agents 330–331 inflammatory disorders 35, 330–331 surgical 83, 99, 131, 191, 357 Inflammatory disorders 35, 36 atlantoaxial subluxation 36, 163 canine distemper virus infection 329–330 cauda equina neuritis 36 central nervous system 122, 329–331 discospondylitis 36, 326–328 epidural empyema 36, 328–329 epidural steatitis 330–331 FIP 330 foreign body migration 36 Gelfoam reaction 36 GME 330 meningo(encephalo)myelitis 36 steroid-responsive meningitis–arteritis 329 Infraspinatus contracture 20 Instrumentation, surgical 73–79 Intercapital ligament 13 Intercostal artery 15, 278 Intermittent aseptic catheterization 353–354 Intermittent claudication 38, 334 Internal fixation devices 291 see also Metal and cement fixation Interspinous ligaments 14
Index
Intertransverse ligaments 14 Intervertebral discs see also Disc herniation anatomy 12 disc removal 75 functional anatomy 11–13 fusion 220, 221 Intervertebral foramen 3, 11, 96 Intracranial disease 163 lesions 94 raised pressure 64 Intradural–extramedullary lesions 29, 51, 52, 53, 248 Intradural–extramedullary tumors 248–252, 259–263 Intraoperative complications cervical disc disease 98–99 CSM surgery 224–226 thoracolumbar disc disease 127 trauma surgery 297, 297–299 tumor surgery 258 Intubation 85 Intumescence 2, 3, 258 Iohexol 43, 49, 86 Ischemic injury 225 Ischemic myelopathy 35, 36, 51, 122, 184, 281 Ischemic neuromyopathy 20, 38, 122, 212, 281, 332–333 Ischemic neuropathy 281, 332–333 Ischemic spinal cord 86 Isoflurane 85 Itraconazole 331 Ivermectin 85 J Jamshidi needle 79 Jaw tone 20 Joint capsule proliferation 211 Joint pain 36, 122, 184, 320 Jugular vein 69, 278 Junctionopathy 20 K Ketoconazole 85 Ketoprofen 341, 342 Kidney disease calculi 38 cysts 55 nephroblastoma 249–250 neurological patient 81 pain 122 worms 38, 122 Kirschner wires 178
K-wires 117, 167–168, 235 Kyphosis 131, 306 L Labrador disc herniation 123 displacement of L2, 284 dorsal hemilaminectomy 119 lumbosacral disease 188 spinal cord hematoma 36, 333 spinal cord mass 58 subarachnoid cyst 324 tumoral calcinosis 327 Lameness (pelvic limb) 183, 184, 204 Laminectomy cervical disc disease 98, 120 CSM 218, 219, 227, 243–244 dorsal 98, 120, 126, 189–190, 195–202, 218, 223, 272 healing 86–87 lumbosacral 188, 189–190, 192 neoplasia 270–273 postoperative complications 209 spinal trauma 294 technique compared to pediculectomy 153 dorsal 126, 189–190, 218, 219, 227 House curette 75 spreader 74 thoracolumbar 126, 270–273 Laminectomy membrane 228 Laminoplasty 219, 223–224, 245–246 Large breed dogs see also specific breeds cardiomyopathy 213 CSM 211 lumbosacral disease 183, 184 spondylosis deformans 185 thoracolumbar disease 132–133 Laryngeal nerve 98, 99, 108 Laryngoscope 85 Larynx 171 Laser therapy 348 Lasix 65 Laxatives 356 Lesions (examination) classification 50 decision-making algorithm 223, 224 localization 21, 28–31 severity 31–32 Leukodystrophy 36, 320, 320–321 Leukoencephalomalacia 320
Lhasa Apso 94 Ligamentous hypertrophy 211 Ligaments cruciate 20 dorsal longitudinal 14 functional anatomy 13–14 ligamentum flavum 197, 199 nuchal 242 ventral longitudinal 14 vertebral 13–14 Limb disorders functional assessment 25–26 gait, attitude and posture 212 lameness 183, 184, 204 neurological examination 28 Lipomatosis, epidural 36 Liver disease 81 Locking plates 219, 220, 220, 221, 223 Locomotor status 20, 20, 22 Lomustine 254 Longissimus tendon 271 Longus colli muscles 109, 112, 173 Loperamide 356 Lower motor neurons (LMN) bladder function 24, 351 diagnostic pitfalls 29 differentiation of abnormalities 26 EMG 60 lesion severity 31–32 limb function 25–26 motor system function 26 muscle atrophy 26 reflexes 5, 26 thoracolumbar disc disease 121 Lumbar arteries 16 Lumbar cord segments 3 Lumbar disc disease see also Lumbosacral disease anatomical location of the injury 296 CSF collection 68–69 differential diagnosis 15, 35, 37 lumbar puncture 68, 69, 72 myelography 71 radiography 47–48, 49–50 Lumbar nerve roots 3 Lumbar nutrient foramen 9 Lumbodorsal fascia 154 Lumbosacral anatomy 11, 182 Lumbosacral disease 181–209 see also Lumbar disc disease diagnosis clinical signs 183 CT 186–188 decision-making algorithm 189 differential 36, 184
371
372
Index
Lumbosacral disease (contd.) discography 186 electrophysiology 184 epidurography 186 examination 183–184 hyperesthesia 184 localization of signs 37 L4/S3 lesions 37 myelography 185 survey radiography 185 non-surgical treatment 188 surgical treatment complications 191–192, 192 dorsal fusion–fixation 190–191, 204–209 dorsal laminectomy 189–190, 195–202 facetectomy 190, 202–204 foraminal decompression 190, 202–204 prognosis 192 Lumbosacral junction 296 Lumbosacral myelography 48–49, 72 Lumbosacral plexus 4, 30, 148 Lumbosacral reflexes 183 Lumbosacral vascular system 183 Lumbosacral vertebrae anatomy 8–9, 10, 11 L1/L7, 1 L2, 284 L3, 36–37 L4/S3, 28, 37–38 L6/L7, 296 L7, 12, 296 Lymphoma see also Neoplasia;Tumors extradural 249 FeLV-associated 81, 263 spinal 262 tumor classification 248 Lysosomal storage disease 36, 326 M Magnetic resonance (MR) imaging atlantoaxial subluxation 164 cervical disc disease 95–96 comparison to CT 57 CSM 216 diagnostic imaging 57–59 thoracolumbar disc disease 123 trauma 285–286 Malacia, focal 148, 286, 293–294 Mannitol 65 Manx cat sacrocaudal dysgenesis 322 spina bifida 322 Marsupialization 244, 325
Massage (effleurage) 348 Mastiff CSM 215, 245 dens hypoplasia 164 Hansen-type disc herniation 182 Medetomidine 343 Medullary lesions 51 Melena (coffee ground) vomit 339, 356 Melanoma 248 Meloxicam 341–342, 343 Menace deficits 20, 29 Meninges 3–4 Meningioma see also Neoplasia; Tumors CSM 212 diagnosis differential diagnosis 35 localization of signs 37 myelographic abnormalities 51 intradural 249 prognosis 260 surgical dissection 273 treatment outcome 261 tumor classification 248 Meningitis–arteritis 329 Meningitis (FIP) 330 Meningocele 212 Meningoencephalitis 38 Meningomyelitis CSM 212 differential diagnosis 36 localization of signs 37 lumbosacral disease 184 neck pain 94 Meningomyelocele 36, 212 Mental status depression 339 neurological screening 20 postoperative care 359–360 Metabolic disease 36 Metaclopramide 356, 357 Metal and cement fixation, see also Implants CSM 218, 219, 221, 228, 239–240 failure 258–259 implant pattern 310, 312 implants 239–240, 296 postoperative complications 228 trauma 309–316 ventral decompression 117 Methenamine mandelate 352 Methocarbamol 96, 102, 343 Methoxyfluorane 85 Methylmethacrylate bone cement 78 see also Bone cement
Methylprednisolone sodium succinate (MPSS) see also Corticosteroids drug dose 84 drug interactions 342 drug side-effects 84 neuroprotective effect 83 spinal trauma 282, 293 treatment CSM 217 decision-making 293 fibrocartilaginous embolism 332 outcome 133 preoperative 83–84, 217, 257 spinal cord tumors 257 spinal trauma 282 thoracolumbar disc disease 133 Metronidazole 85 Micturition see also Urinary system; Urine neurological damage 24 physiological control 350 postoperative care 351 Miniature dogs see also Small breed dogs Chihuahua 161 pinscher 118 poodle 177, 325 schnauzer 332, 359 toy breeds 165 Mini-chuck 76 Mini-hemilaminectomy combined with pediculectomy 154 dorsolateral 143 lateral approach 151 thoracolumbar disc disease 126–127, 136, 151–154 Minocycline 327 Miosis 25 Misoprostol 343, 356 Mobilizers, surgical 75 Morphine 340, 341 Motor system 5–6 Motor neurons (LMN and UMN) 5, 20, 24–26, 29, 31–32, 60, 121, 351 Movement disorders 183 see also Walking ability Moving of patients, postoperatively 343 Mucopolysaccharidosis type VI 326 Mucosal bleeding time (buccal) 217 Multiple cartilaginous exostoses (osteochondromatosis) 212
Index
Multiple ventral implants 169, 177 Muscle 146 Muscle activity see also Electromyography EMG 60–61 gastrocnemius muscle 60 urinary sphincter 351 Muscle atrophy 26, 183 Muscle masses 20, 26 Muscle miosis 25 Muscle palpation 21, 24 Muscle relaxants 352 Muscles and muscle groups epaxial 147, 152, 159, 196, 197 iliocostalis 155, 156, 157, 158 intercostal 278 levator costarum 157, 158 longissimus 141, 155 longus colli 109, 112, 173, 233 multifidus 271 omotransversarius 276 pectoral 276 psoas 37, 38, 331 rhomboideus 241 scalenus 277, 278 spinalis 242, 243, 267 sternocephalicus 107, 233 sternohyoid 107,171 sternothyroid 172 strap 233 trapezius 241 Musculocutaneous nerve 31 Mycobacterium 331 Myelitis 184 Myelodysplasia (spinal dysraphism) 36 Myelography abnormalities common causes 51 CSM 213–216 interpretation 50–51 neoplasia 251 nerve sheath tumor 51 neurodegeneration 53–54 spinal cord expansion 57 trauma 283–284 complications 51–54, 217 contrast media 49, 70, 86 indications 48–49 postoperative care 343 spinal region cervical 49, 70–72, 95, 164 lumbar 49–50, 71 lumbosacral 72 normal lateral 49 thoracolumbar 122–123 Myeloma 42, 61, 248
Myelomalacia malacia 148 postoperative 127, 359 secondary to disc extrusion 333 thoracolumbar disc disease 127, 128 Myelopathy brachial plexus 36 cervical 38, 82, 212 degenerative 37, 184, 209 differential diagnosis 184, 212 fibrocartilaginous embolic 212 hemorrhage 82 hypoventilation 82 hereditary 36 ischemic 35, 37, 51, 184 lumbosacral disease 184 myelographic abnormalities 51 Myelotomy 87 Myesthenia gravis 38 Myopathy 20, 38 Myositis 20, 38 Myxoma 260 N Narcotic agents 85, 96, 340 Necrosis aseptic 259 fat graft 36 Needle biopsy 275 Neoplasia 247–279 see also specific tumor types diagnosis biopsy 251–252, 262 classification 248 clinical signs 247–248 differential 35, 163, 281 electrophysiology 250–251 imaging 250–251 localization of signs 37 staging 252–254 extradural 248–249, 259 extramedullary 249–250, 259–261 fibrosarcoma 38 intradural 38, 249–250, 259–261 intramedullary 250, 261 lumbosacral 184 nerve root tumors 251, 275, 279 nerve sheath tumor 257, 275–279 neurological patients 81 prognosis 259–261, 263 spinal 248, 261–263 spinal cord 59, 257–258 treatment decision-making algorithm 256 non-surgical treatment 254–255, 262–263
surgical treatment 255–259, 266–279 tumor biology 248–250 Neospora caninum 42, 330 Nephroblastoma 51, 248, 249–250, 261, 273 Nerve block 341 Nerve compression 6, 29 Nerve hook 76 Nerve roots 3, 12, 21, 131 nerve root compression 29 nerve root signature 93, 94, 103, 212, 279 nerve root tumors 251, 275, 279 Nerve sheath tumor clinical signs 248 diagnostic imaging 252 differential diagnosis 35, 212 myelographic abnormalities 51 prognosis 260–261 resection 275–279 surgery 257 tumor biology 248 Nerves cranial 20, 21, 25, 277, 278 femoral 127 laryngeal 98, 99, 108 musculocutaneous 31 pelvic 30 pudendal 30 radial 31, 275 saphenous 31 sciatic 30 superficial peroneal 31 thoracic 277, 278 tibial 31 ulnar 31 Nervous tissue, functional anatomy 1–6 Neuritis cauda equina 36, 37, 184 polyradiculoneuritis 38 Neurogenic muscle disease 24, 26 Neurological disease CSM 212 ischemic 38, 332–333 localization of signs 37 lumbosacral disease 184 neuroaxonal dystrophy 320–321 neurogenic atrophy 26 neuromyopathy 37, 38, 184, 212 neuropathy 38, 184, 332–333 patient management 81 peripheral 38, 184 trauma 282–283
373
374
Index
Neurological examination see also Reflexes brachial plexus 30–31 cranial nerves 20, 21, 25 DAMNIT formula 21, 35, 36 differential diagnosis 20 eyes 21 functional assessment attitude, posture and gait 20, 22, 29 hyperesthesia 20, 21, 184 limbs 28, 29 motor function 20, 22, 26 pain 21, 25, 27–28 proprioception 22 urinary function 20, 21 grading 124, 125 lesion localization 21, 38–39 lumbosacral plexus 30–31 observation 20, 21 palpation abdomen 22–24 muscle 21, 26 pain assessment 24 spine 24–25 Neuromuscular disorders 20 Neuromyopathy (ischemic) 20 Neuropathy 38, 184, 332–333 Neurovascular bundle 139, 141 Neutrophils 43, 45 Nitrous oxide 85 Nociception (pain perception) see also Pain decision-making algorithms 293, 294 deep pain sensation 20, 21, 87, 125, 127, 293–295, 301–302 localization 37, 38 neurological examination 27–28 pathophysiology 286 prognosis 32, 301–302 recurrent 150 trauma 87, 293–295 withdrawal reflex 28 Non-steroidal anti-inflammatory drugs (NSAIDs) see also Antiinflammatory drugs cervical disc disease 96 effect on thyroid function tests 217 pharmacology 341 postoperative care 192, 341–343 preoperative assessment 82, 83 side-effects 83, 85, 341–342, 355, 356 wound infection 357
Non-surgical treatment anti-inflammatory drugs 96 atlantoaxial subluxation 164–165, 166–167 cervical disc disease 96 complications 166–167 CSM 218 decompression 124, 125 fenestration 124, 125 immobilization 282 lumbosacral disease 188, 192 recovery times 124, 125 rest 125, 126, 131, 192, 228, 282, 288, 305 thoracolumbar disc disease 123, 124, 125 trauma 282, 289–290, 305–309 Nuchal ligament 241, 242 Nucleus pulposus 13, 111, 152, 156 Nursing care, postoperative 343–345 see also Postoperative complications Nutrient foramen 9 Nutrition see also Food intake hypervitaminosis A 36, 37, 326 postoperative nursing care 344–345 vitamin disorders 36, 37, 217, 326 O Obesity 83 Occipital artery 15 Occipital protuberance 178 Oculovestibular response 20 Odontoidectomy 171, 177 Oligodendroglioma 252 Omeprazole 356, 356 Operating loupes 73, 74 Operating microscope 73, 74 Operation duration 83 Opioid analgesics 340–341, 343, 356 side-effects 84–85, 340, 341 Organophosphate toxicity 38, 42 Oropharyngeal pain 94 Orthopedic disease CSM 211, 212 DISH 326 fracture 122 lumbosacral 183, 184 neurological patients 81 mimicking spinal disease 20 tumors 248, 249, 260 Osteoarthritis 20 Osteochondritis dissecans (OCD) 20, 36, 37, 183 Osteophytes 211 Osteoporosis 38, 326
Osteoporotic pathological fracture 122 Osteosarcoma 248, 249, 260 Otitis externa 167 Otitis media 94 Oxymorphone 340 P Packed cell volume (PCV) 82 Pain see also Deep pain sensation; Nociception abdominal 38, 122 discogenic 184 facet joint 36, 122, 184 lumbosacral 181, 184 neck 94 oropharyngeal 94 with paraplegia 32 pleuritic 94 postoperative after hemilectomy 149 after spinal trauma 301–302 after ventral slot surgery 103 sacroiliac joint 184 thoracolumbar disc disease 122, 125 Pain syndrome (thalamic) 38 Palliative therapy 254 Palpation (spine) 24–25 Palpebral reflex 20 Pancreatitis 38, 122, 357 Panniculus reflex 21 Panosteitis 20 Paraparesis 29, 32 Paraplegia 26, 32, 130 Paralysis Coonhound 38 progressive myelomalacia 127 tick 38 Patellar luxation 20 Patellar reflex 20, 26–27 Pathophysiology, trauma 290–292 Patient examination etiology 20–28 history taking 19 lesions 28–32 neurological 20–28 physical 19–20 Paw position 20 Pedicle grafts 87 Pediculectomy 127, 151–154 Pelvic nerve 30 Penicillin derivatives 83 Penile erection 359 Penrose drains 276 Perineal reflex 20
Index
Periodontal disease 83 Periosteal elevator 75, 136, 197, 235 Peritonitis 330 Peroneal nerve 31 Pharynx pain 94 Pheochromocytoma 20 Phenobarbitone 217 Phenoxybenzamine 343, 352 Physical examination 19–20 see also Neurological examination Physical therapy 131, 343, 346–350 Pilonidal sinus (dermoid sinus) 36, 37, 184, 212, 323 Pin fixation 78 see also Cross-pin fixation; Screw fixation angles for fixation 311, 312 entry points and trajectory 312 external fixation 292 failure rates 169 insertion 76 metal and cement fixation 309 stapling technique 315–316 Steinmann pins 221, 290–292, 296, 298, 310, 314, 315 vertebral distraction–stabilization 221 Pinscher, miniature 118 Placing test 23 Plasma cell tumor 259, 260 Platelets 82, 342 Pleocytosis 44, 45, 49, 329, 330 Pleuritic pain 94 Pneumatic system 77 Pneumomediastinum 225 Pneumonia 99, 297 Pneumothorax 127, 130, 158, 297 Polyarthritis 20, 38, 94, 122 Polymyositis 38, 94, 122, 163 Polyradiculoneuritis 38, 184 Poodle atlantoaxial subluxation 161, 163, 168 dens hypoplasia 164 Position (perception of) anatomy 5 nerve compression 6 neurological examination 22 postoperative sensory stimulation 348 reflex step 23 Positive profile pins 78 Postoperative care 339–360 analgesia 339–343 atlantoaxial subluxation 169 cervical disc disease 102 cleanliness 343–344 CSM 228–229
flooring and bedding 345–346 hydration and nutrition 344–345 moving patients 343 myelography 343 neoplasia 259 neurosurgery 339 physical therapy 346–350 recumbency 344 slings and harnesses 346, 347, 349 trauma 297 treatment plans 343, 344 urinary function 344, 350–352, 353–355 Postoperative complications 224 see also Intraoperative complications; Preoperative complications cervical disc surgery 99–102 CSM 225, 226–228 decubital ulcers 358–359 deep pain sensation (loss of) 150 deformity 258 fenestration 130 gastrointestinal 127, 297, 355–357 implant failure 258–259 infections 83, 258, 259, 354–355, 357 myelomalacia 359 osseous fusion 97 penile erection 258, 359 respiratory 168 self-mutilation 359 seroma formation 358 spinal cord tethering 149 thoracolumbar disc disease 127–132 thromboembolism 359 trauma 299, 299–301 tumor recurrence 259 urine scald 357–358 Posture, attitude and gait 20, 20, 23, 181 Prazosin 352 Prednisolone 96, 218, 329, 330 Prednisone 254 Preoperative assessment analgesia 84–85 anesthesia 85–86 client communication 87 clinical assessment 81–83 durotomy 87 imaging 82 laminectomy healing 86–87 myelotomy 87 pharmacological considerations 83–85 recovery after spinal cord injury 87
surgical considerations 86–87 Pressure sores 308 Procaine penicillin 357 Proprioception (perception of position) anatomical considerations 5 effects of compression 6 neurological examination 22 postoperative sensory stimulation 348 reflex step 23 Prostate disease 38, 81, 122, 184 Protein analysis 42, 45 Protozoal infections 38, 330 Pruritus 183 Pseudohyperreflexia 27, 183 Psoas muscle 37, 38, 122, 184, 281, 331 Psyllium 356 Ptosis 25 Pudendal nerve 30 Pug lumbosacral disease 191 sacrocaudal dysgenesis 322 tetraparesis 101 Pulmonary thromboembolism (PTE) 359 Pupillary light reflex 20 Pyoderma 81 Pyometra 81 Pyrimethamine 331 R Rabies 330 Radial nerve 31, 275 Radiation injury 36 Radiation therapy 254–255 Radicular arteries 16 Radiculopathy 20, 38 Radiography imaging techniques discography 50 epidurography 50 interpretation 50–51 myelography 48–50, 51–54 normal spinal 46 positioning 46, 47 principles 50–51 survey radiography 45–46, 283 x-rays 82, 99 lesions atlantoaxial joint 46–47 cervical 46, 49, 94 extradural 51 intradural 51 intramedullary 51
375
376
Index
Radiography (contd.) lesions (contd.) lumbar 49–50 neoplasia 250 thoracolumbar 122 trauma 283 postoperative care 344 Ranitidine 356 Recovery times, thoracolumbar surgery 125 Recumbency CSF collection 59 imaging 47, 55 postoperative 225, 228–229, 344 Red blood cell counts 44, 45 Reflexes 26–27 anal 183, 185 crossed extensor 32 cutaneous trunci 20, 24, 24, 26–27 hock flexion 27, 30, 182 LMN 5, 26 palpebral 20 panniculus 21 patellar 20, 26–27 perineal 20 postural 20 pseudohyperreflexia 27, 183 pupillary 20 reflex step 23 Schiff–Sherrington sign 32 spinal 20, 21, 87 UMN 6, 26 withdrawal 20, 27, 28, 31 Renal disease calculi 38 cysts 55 nephroblastoma 249–250 neurological patients 81 pain 122 worms 38, 122 Respiratory disorders dyspnea after imaging 82 fentanyl-induced 84 hypoventilation 82 pneumomediastinum 225 pneumonia 99, 297 pneumothorax 127, 130, 158, 297 postoperative 168 pulmonary edema 167 respiratory arrest 167 upper airway disease 20, 38 Rest see also Non-surgical treatment cage confinement 125, 126, 131, 228, 288, 305 lumbosacral surgery 192 spinal trauma 282, 288
thoracolumbar disc disease 123, 124, 125 Restrictive fibrosis 225, 273 Reticulosis (focal granulomatous meningoencephalomyelitis) 248 Retractors Adson–Baby 74 Army Navy 152, 233 blunt 74 Gelpi 74, 108, 138, 235, 236, 237, 238 Gosset 74 Langenbeck 138 multi-toothed 74 pediatric Balfour 74 self-retaining 74, 197 Senn 138 Weitlander 74 Retriever bulbous dilation of dorsal arachnoid space 324 left hemiparesis 29 myelomalacia 61 thoracolumbar spinal cord segments 3 vertebral bone loss 61 Rhizotomy 148 Rhodesian ridgeback 222 Rib (thirteenth) 140, 155 Rickettsia 42, 330 Rongeurs 76 Root signature 93, 94, 103, 212, 279 Rottweiler atlantoaxial subluxation 161 CSM surgery 244 discospondylitis 328 laminectomy 201, 244 leukodystrophy 320–321 leukoencephalomalacia 320 lumbar tumor 254 neuroaxonal dystrophy 320–321 spinal compression 216 S Sacral vertebrae S1/S3, 1 S3, 37, 37–38 Sacrocaudal spine dysgenesis 36, 37, 322 localization of signs 37 tail injury 296–297 Sacroiliac joint pain, cervical 36, 184, 320 see also Lumbosacral disease Sacrum 3 Saphenous nerve 31
Sarcoma 248, 261 Scalenus muscle 277, 278 Scalpel 75 Scar formation 86–87, 191, 192 Schiff–Sherrington sign 26, 32 Schiff–Sherrington syndrome 32, 284 Schmorl’s node 36, 94, 184 Schnauzer (miniature) decubital ulcers 359 fibrocartilaginous embolism 332 Sciatic nerve 31 Scintigraphy 59–60 Scoliosis 61, 127 Screaming (in dogs) 94 Screw fixation see also Fusion–fixation; Pin fixation atlantoaxial subluxation 166, 169, 174–176 CSM 220, 223, 225, 226, 237–240 failure rates 169 lumbar spine 274, 296 lumbosacral spine 190, 191, 192, 296 pelvic limb lameness 204 postoperative complications 208, 225, 226 techniques cement plug 237, 238 laminectomy 191, 192, 208 locking plates 223 metal and cement fixation 239–240 placement 207, 208 threaded screws 240, 292, 309 ventral slot 220 vertebrectomy 274 trauma 296 Sedatives 282 Sedimentation chambers 44 Seizures after myelography 53–54, 86, 339 after neurosurgery 339 prazosin 352 Self-mutilation 127, 183, 359 Sensory function see also Neurological examination; Nociception; Proprioception ascending sensory tracts 5 facial sensation 20 Sepsis (postoperative) 99, 229, 259, 297 Septic arthritis 20 Septic shock 83 Seroma formation 99, 191, 229, 358 Sevofluorane 85
Index
Sharpei 60 Shih Tzu chondrodystrophy 12 disc extrusion 94 L4/S3 neurological defects 30 neck pain 94 Shock septic 83 spinal 32, 282 Silky terrier (disc extrusion) 102 Sinus (pilonidal) 36, 37, 184, 212, 323 Skeletal hyperostosis 36, 326 Skeleton 6–14 Slings and harnesses 346, 347, 349 Small breed dogs see also Miniature dogs; Toy breed dogs atlantoaxial subluxation 161 thoracolumbar disc disease 151 Soft tissue tumors 94 Spears 77 Specific gravity (CSF) 45 Sphincter tone 191 Spina bifida 36, 37, 184, 322 Spinal arteries 16 Spinal cord anatomy 1, 1–6, 24 blood supply 16 evoked responses 60–61 nerve fibers 6 segments 1–3 ultrasound examination 55 white matter tracts 5–6 Spinal cord disease compression 227, 228 expansion 57 hematoma 36, 37, 94, 212, 333–334 hemorrhage 36, 51, 94, 281, 333–334 herniation 147, 273 infection 330 injury 87 ischemia 86 liquification 132 malacia 287 swelling 127 tethering 36, 37, 149, 184, 322 tumors 59, 257–258 Spinal dysraphism (myelodysplasia) 36, 212, 322 Spinal ganglion 3 Spinalis muscle 267 Spinal nerves 3, 183, 191, 277, 278 Spinal palpation 24–25 Splints, external 239, 290, 305, 306 Spondylitis 326–328
Spondylomyelopathy (cervical) 20, 35, 36, 94, 71, 211–246, 281 see also Wobbler syndrome Spondylosis deformans 185, 320, 321 Sponges 77, 78, 79, 141 Stabilization atlantoaxial subluxation 166–167 caudal disc extrusion 96 cervical disc disease 117–118 Staphylococcus 327, 330 Stapling (spinal) 316 St Bernard 238 Steatitis (epidural) 330–331 Steinmann pins 221, 290–292, 296, 298, 310, 314, 315 Stenosis cervical fibrotic 36, 37, 212 vertebral canal 183 Sternocephalicus muscles 107 Sternohyoid muscles 107 Sternothyroid muscles 172, 172 Steroid-responsive meningitis–arteritis 329 Steroids 83, 326, 329, 357 see also Corticosteroids; Nonsteroidal anti-inflammatory drugs Stilette 65 Stranguria 184 Streptococcus 327 Streptomycin 327 Subdural hemorrhage 57 Subdural space 4 Sucralfate 344, 356 Suction tip and bulb syringe 77 Supraspinous ligaments 14 Surgical instruments 73–79 Surgicel 77 Surgifoam 77 Survey radiography 163–164, 213, 283 Suturing 276 Swallowing 20 Swimming therapy 348–349 Synovial articulations 9–11 Synovial cyst 320 cervical 215 column splitting 52 CSM 212, 219, 245 differential diagnosis 36, 94, 122 localization of signs 37 lumbosacral 183 Synthes locking plate 219, 220, 221 Syringes collection of CSF 67 hypodermic needles 76 suction tip and bulb syringe 77
Syringohydromyelia 163 Syringomyelia 36, 37, 51, 94, 163, 212, 297, 311, 322 T Tail defects lumbosacral disease 183 sacrocaudal dysgenesis 322 Tail injuries 296–297 Tartar scrapers 76 Technical errors 258 Temperature, pulse, respiratory rate (TPR) 343, 344 Temporal muscle mass 20 Temporomandibular joint lesion 94 Tendonitis 20 Tendons 20, 271 Terazosin 352 Tetanus 38 Tethered spinal cord 36, 37, 149, 184, 322 Tetracyline antibiotics 331 Thalamic pain syndrome 38 Thiamine deficiency 36 Thoracic spine anatomy 8–9 aorta 10 blood supply 15 localization of injury 296 mid-bodies 10 radiography 47–48 spinal cord segments 3 Thoracolumbar disc disease diagnosis clinical signs 121–122 CSF analysis 122 decision-making algorithm 124 differential 122 imaging techniques 56, 122–123 kyphosis 131 non-surgical treatment 123–126 prognosis 132–133 surgery complications 127, 129, 130–132 disc extrusion 127, 129, 131 decompression 124, 126–127, 130 fenestration 127, 154–159 hemilaminectomy 126, 126–127, 136–150,151–154 kyphosis 131 laminectomy 126, 270–273 pediculectomy 126, 127, 151–154 recovery times 124–125
377
378
Index
Thoracolumbar vertebrae T1/T13, 1 T2, 36 T3/L3, 15, 28, 36, 37 Thrombin 99 Thrombocytopenia 82 Thromboembolism 359 Thrombosis (iliac) 332–333 Thyroid gland 20, 172, 217 Thyroid hormone supplementation 217 Thyroid-stimulating hormone (TSH) 217 Tibial crest avulsion 20 Tibial nerve 31 Tick paralysis 38 Toxicity organophosphate 38, 42 toxic disease 36 toxic myopathy 38 zinc 358 Toxoplasma gondii (toxoplasmosis) 42, 330 Toy breed dogs 165, 325 see also Miniature dogs TPR (temperature, pulse, respiratory rate) 343, 344 Trachea 108, 171 Traction (response to) 213, 214, 223, 224 Transverse foramen 7, 8 Transverse ligament of the atlas 14 Trauma 281–318 see also Atlantoaxial subluxation bile duct rupture 19 biomechanics 286–289 diagnosis anatomical location of the injury 295–297 decision-making algorithm 293, 294 differential 36, 184 imaging techniques 283, 283–286 initial assessment 281–282 neurological signs 28, 282–283, 293–295 fracture 21, 36, 87, 286–289 gunshot injury 36, 331 neurology deep pain sensation 293–295 examination 21 pathophysiology 289–292 psoas muscle 331 prognosis 301–302 recovery 87
spinal region cervical 295–297 lumbar 296 lumbosacral 184 sacral 36, 296–297 thoracic 282, 296 treatment bone grafts 292 choice 293–295 complications 297–301 fixation 2890–291, 316–318 metal and bone cement 291–292, 309–316 modified segmental 291 non-surgical treatment 289–290, 305–309 postoperative care 297 urinary tract 282 Traumatic feline ischemic myelopathy 36, 37, 281 Trimethoprim–sulfonamide 85, 331, 343, 355 Tube cystostomy 355 Tumoral calcinosis (calcinosis circumscripta) 36, 37, 122, 212, 326 Tumors 252–254 see also Neoplasia; specific types of tumor biopsy 61–62 classification 248, 253 seeding 276 spinal 59, 257–258, 261–263 U Ulcer formation decubital 339, 358–359 gastrointestinal 127, 297, 341, 356 Ulnar nerve 31 Ultrasonography 54–55, 250 Ultrasound therapy 347 Upper airway disease 20, 38 Upper motor neurons (UMN); see also Lower motor neurons bladder function 20, 24, 351 crossed extensor reflex 32 EMG 60 motor evaluation 25–26 reflexes 6, 26 severity 31–32 thoracolumbar disc disease 121 Ureter calculi 38, 122 Urethra innervation 351 neoplasia 184 tumor 38
Urinary system see also Bladder antiseptic 352 function 21, 22, 23–24, 351–352 incontinence 24, 183, 302 infection 42, 81, 82, 83, 131, 297, 355 micturition disorders 351 neurological damage 24 pharmacology 352 physiological control 350 postoperative care 351 Urine urine retention 182 urinalysis 42, 339, 343, 344 urine scald 127, 297, 339, 357–358 voiding change 339 V Vascular disease 20, 36, 329, 332–334; see also Cardiovascular disorders Vascular system see also Blood supply arteries aorta 10, 332–333 intercostal 15, 278 occipital 15 omocervical 276 superficial cervical 278 blood supply spinal regions cervical spine 15 lumbar spine 16 lumbosacral spine 183 spinal cord 16 thoracic spine 15 vertebral column 14–17 veins azygous 15 jugular 278 venous drainage 17 vena cava 10 venous plexus 114, 145–146, 148 Ventilation (mechanical) 101 Ventral decompression (slot) cervical disc disease 99, 103 comparison to fenestration 97 complications 98, 99, 227, 228 CSM 218, 218, 219–221, 227, 228 laryngeal nerve 108 lumbosacral disease 192 postoperative care 100, 101, 102 recovery 103
Index
Ventral fissure 16 Vertebrae (general) anticlinal 9 blood supply 14–17 column splitting 52 congenital anomalies 321–322 functional anatomy 1–15 intervertebral discs 11–12 ligaments 13–14 spinal cord segments 2–3 synovial articulations 9–11 Vertebral buttress 287–289 Vertebral canal 7, 8 cauda equina compression 181 CSM 211, 219 differential diagnosis 36 tumors 253 type I disc extrusion 14 Vertebral column 14–17 Vertebral distraction/stabilization 218, 219, 221–223 see also Fixation–stabilization; Internal fixation devices; Metal and cement fixation Vertebral foramen 141 Vertebral fracture 225, 331 Vertebral instability 211 Vertebral plexus 15, 269, 273 Vertebral tumors 253, 255–257 Vertebrectomy 255, 273–274 Vestibular disease 85 Viral infections 330 Visual function blindness 85 neurological examination 20, 21 placing test 23 ptosis 25 Vitamin A 36, 37, 326 Vitamin E 217, 257
Vomiting 339, 355–356 Von Willebrand (VW) factor/disease cervical hematoma 360 CSM surgery 229 diagnosis 217 hypothyroidism 217 preoperative assessment 82 routine laboratory analysis 42 W Walking ability CSM surgery 229 lumbar vertebrectomy 274 neurological examination 32 pain 32 paraparesis 32 paraplegia 32 spinal reflex 87 supported 346 thoracolumbar disc surgery 124, 130 ventral slot surgery 102, 103 Warfarin 333 Washing (postoperative) 343 Water bath therapy 348–349 Water intake 85, 343, 344–345, 355 Waterproof drapes 74 Wedges (foam) 46 Weimaraner atlantoaxial subluxation 161 spinal dysraphism 322 Westie (marsupialization) 325 Wheelbarrowing test 23 White blood cells (WBC) CSF analysis 44, 45 meningitis 329 urinalysis 42
White matter tracts 5–6 Wire fixation atlantoaxial subluxation 167, 169, 178–180 dorsal 178–180 failure rates 169 K-wires 117, 167–168, 235 Withdrawal reflex 20, 28, 31 Wobbler syndrome see also Cervical spondylomyelopathy domino lesions 101 drug-induced seizures 86 horses 218 MRI 58 surgery 77, 218 Working dogs (lumbosacral disease) 183, 188 Wounds and wound infections discharge 339 dressings 344, 357 obesity 83 postoperative 99, 127, 128, 297, 339, 357 predisposing factors 83, 357 prevention 357 routine laboratory analysis 42 septic shock 83 X Xanthochromia 45 Y Yorkshire terrier 43 Z Zinc toxicity 358
379
