Thyroid Ultrasound and Ultrasound-Guided FNA Second Edition
Thyroid Ultrasound and Ultrasound-Guided FNA Second Edition
H. Jack Baskin, M.D., MACE Orlando, FL, USA
Daniel S. Duick, M.D., FACE Phoenix, AZ, USA
Robert A. Levine, M.D., FACE Nashua, NH, USA
Foreword by Leonard Wartofsky, M.D., MACP Washington, DC, USA
Editors H. Jack Baskin 1741 Barcelona Way Winter Park FL 32789 USA
[email protected]
Daniel S. Duick 3522 North 3rd Avenue Phoenix AZ 85613 USA
[email protected]
Robert A. Levine Thyroid Center of New Hampshire 5 Coliseum Avenue Nashua NH 03063 USA
[email protected]
ISBN 978-0-387-77633-0
e-ISBN 978-0-387-77634-7
© 2008 Springer Science+Business Media, LLC All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper 9 8 7 6 5 4 3 2 springer.com
Foreword Ultrasound has become established as the diagnostic procedure of choice in guidelines for the management of thyroid nodules by essentially every professional organization of endocrinologists. In this, the second edition of their outstanding text on thyroid ultrasound, Baskin, Duick, and Levine have provided an invaluable guide to the application of gray-scale and color Doppler ultrasonography to state-of-the-art diagnostic evaluation of thyroid nodules, and to the management of thyroid cysts, benign thyroid and parathyroid nodules, and thyroid cancer. Differences with, and additions to, the first edition highlight the extraordinary and dramatic advances in applications of ultrasonography that have occurred in the past decade. The high yield of malignancy in ultrasound-guided fine-needle (FNA) aspirates of nondominant nodules in multinodular glands has altered our mistaken complacency in assuming that palpation-guided FNA only of palpable dominant nodules was adequate for diagnosis. Rather, ultrasound has taught us that the commonly held belief that malignancy is less likely in a multinodular gland is incorrect. Utility of ultrasound has gone far beyond just the initial diagnostic approach, as improved highly sensitive probes allow accurate characterization of the nature of thyroid nodules or lymph nodes, setting priorities for FNA and for serial monitoring for changes in size that could imply malignancy. Ultrasound is also informing us as to the frequency and significance of thyroid microcarcinomata. The greater sensitivity of modern ultrasonographic (US) technique has opened a Pandora’s box in facilitating the detection of small nodules, which then mandate FNA (or serial follow-up at a minimum). Awareness that certain ultrasound characteristics of nodules (e.g., hypoechogenicity, microcalcifications, and blurred nodule margins) are associated with malignancy has allowed us to focus our interest in FNA primarily and selectively on nodules with these characteristics. Many such small nodules with these characteristics are found to constitute microcarcinomas, and their natural history teaches us that they can be as aggressive as tumors that are > 1 cm in size. As a consequence, their earlier detection employing ultrasound has facilitated better
v
vi
FOREWORD
outcomes and potential cures. Thus, modern management of thyroid nodules demands the skilled use of ultrasound to identify all nodules in a given thyroid gland and to more definitively guide the needle for aspiration. The evidence is clear that an ultrasound-based strategy has been shown to be cost-effective in reducing nondiagnostic FNA rates, particularly by targeting those nodules with ultrasonographic characteristics that are more suggestive of malignancy. As a result, unnecessary thyroid surgeries can be avoided and a greater yield of thyroid cancer can be found at surgery. Moreover, in patients with FNA positive for cancer, preoperative baseline neck ultrasound has been shown to be of significant value for the detection of nonpalpable lymph nodes or for guiding the dissection of palpable nodes. Ultrasoundguided FNA of lymph nodes has taught us that anatomic characteristics and not size are better determinants of regional thyroid cancer metastases to lymph nodes. This book is replete with critical assessments of the recent literature on which the above statements are based, and includes the most up-to-date descriptions of newer applications of ultrasound to distinguish benign from malignant nodules such as elastography, as well as practical analytic appraisal of the utility of incorporation of ultrasound to the ablation of both benign and malignant lesions by ethanol instillation, high frequency ultrasound, laser, or radiofrequency techniques. In my view, given the extremely important current and future role of ultrasonography in the diagnosis and management of our patients, endocrinologists, cytopathologists, surgeons, and radiologists are obligated to become familiar with and adopt the approaches and advances described in this volume. Leonard Wartofsky, MD, MACP Washington Hospital Center Washington, DC
Preface to First Edition Over the past two decades, ultrasound has undergone numerous advances in technology, such as gray-scale imaging, realtime sonography, high resolution 7.5–10 Mtz transducers, and color-flow Doppler that make ultrasound unsurpassed in its ability to provide very accurate images of the thyroid gland quickly, inexpensively, and safely. However, in spite of these advances, ultrasound remains drastically underutilized by endocrinologists. This is due in part to a lack of understanding of the ways in which ultrasound can aid in the diagnosis of various thyroid conditions, and to a lack of experience in ultrasound technique by the clinician. The purpose of this book is to demonstrate how ultrasound is integrated with the history, physical examination, and other thyroid tests (especially FNA biopsy) to provide valuable information that can be used to improve patient care. Numerous ultrasound examples are used to show the interactions between ultrasound and tissue characteristics and explain their clinical significance. Also presented is the work of several groups of investigators worldwide who have explored new applications of ultrasound that have led to novel techniques that are proving to be clinically useful. To reach its full potential, it is critical that thyroid ultrasound be performed by the examining physician. This book instructs the physician on how to perform the ultrasound at the bedside so that it becomes part of the physical examination. Among the new developments discussed are the new digital phased-array transducers that allow ultrasound and FNA biopsy to be combined in the technique of ultrasound-guided FNA biopsy. Over the next decade, this technique will become a part of our routine clinical practice and a powerful new tool in the diagnosis of thyroid nodules and in the follow-up of thyroid cancer patients. H. Jack Baskin, MD Editor
vii
Preface to Second Edition In the eight years since the publication of the first edition of this book, ultrasound has become an integral part of the practice of endocrinology. Ultrasound guidance for obtaining accurate diagnostic material by FNA is now accepted normal procedure. As the chief editor of Thyroid wrote in a recent editorial: “I do not know how anyone can see thyroid patients without their own ultrasound by their side.” The widespread adoption of this new technology by clinicians in a relatively short span of time is unprecedented. While most endocrinologists now feel comfortable using ultrasound for the diagnosis of thyroid nodules, many are reluctant to expand its use beyond the thyroid. Its value as a diagnostic tool to look for evidence of thyroid cancer in neck lymph nodes, or to evaluate parathyroid disease is at least as great as it is in evaluating thyroid nodules. In this second edition, we continue to explore these diagnostic techniques that are readily available to all clinicians. Since the first edition, clinical investigators have continued to discover new techniques and applications for thyroid and neck ultrasound. Power Doppler has replaced color flow Doppler for examining blood flow in the tissues of the neck. Other new advances in diagnosis include ultrasound contrast media, ultrasound elastography, and harmonic imaging. The only ultrasound-guided therapeutic procedure addressed in the 2000 edition was percutaneous ethanol injection (PEI), which had not been reported from the United States but was commonly practiced elsewhere in the world. Today, other ultrasoundguided therapeutic procedures such as laser, radiofrequency, and high intensity focused ultrasound (HIFU) are being used for ablation of tissue without surgery. These innovative procedures are discussed by the physicians who are developing them. We hope that this second edition will inspire clinicians to proceed beyond using ultrasound just for the diagnosis of nodular goiter. The benefits to patients will continue as clinicians advance neck ultrasound to its full potential. H. Jack Baskin, MD Editor, 2008 ix
Contents Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Leonard Wartofsky
v
Preface to First Edition . . . . . . . . . . . . . . . . . . . . . . . . . . . . H. Jack Baskin
vii
Preface to Second Edition . . . . . . . . . . . . . . . . . . . . . . . . . . H. Jack Baskin
ix
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xiii
1
History of Thyroid Ultrasound . . . . . . . . . . . . . . . . . Robert A. Levine
1
2
Thyroid Ultrasound Physics . . . . . . . . . . . . . . . . . . . Robert A. Levine
9
3
Doppler Ultrasound . . . . . . . . . . . . . . . . . . . . . . . . . . Robert A. Levine
27
4
Anatomy and Anomalies . . . . . . . . . . . . . . . . . . . . . . H. Jack Baskin
45
5
Thyroiditis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reagan Schiefer and Diana S. Dean
63
6
Ultrasound of Thyroid Nodules . . . . . . . . . . . . . . . . Susan J. Mandel, Jill E. Langer and Daniel S. Duick
77
7
Ultrasound-Guided Fine-needle Aspiration of Thyroid Nodules . . . . . . . . . . . . . . . . . . . . . . . . . . Daniel S. Duick and Susan J. Mandel
97
8
Ultrasound in the Management of Thyroid Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 H. Jack Baskin
9
Parathyroid Ultrasonography . . . . . . . . . . . . . . . . . . 135 Devaprabu Abraham
xi
xii
CONTENTS
10 Contrast-Enhanced Ultrasound in the Management of Thyroid Nodules . . . . . . . . . . . . . . 151 Enrico Papini, Giancarlo Bizzarri, Antonio Bianchini, Rinaldo Guglielmi, Filomena Graziano, Francesco Lonero, Sara Pacella, and Claudio Pacella 11 Percutaneous Ethanol Injection (PEI): Thyroid Cysts and Other Neck Lesions . . . . . . . . . . . . . . . . . 173 Andrea Frasoldati and Roberto Valcavi 12 Laser and Radiofrequency Ablation Procedures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Roberto Valcavi, Angelo Bertani, Marialaura Pesenti, Laura Raifa Al Jandali Rifa’Y, Andrea Frasoldati, Debora Formisano, and Claudio M. Pacella 13 High Intensity Focused Ultrasound (HIFU) Ablation Therapy for Thyroid Nodules. . . . . . . . . . 219 Olivier Esnault and Laurence Leenhardt 14 Ultrasound Elastography of the Thyroid . . . . . . . . 237 Robert A. Levine Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Contributors Devaprabu Abraham, MD, MRCP Salt Lake City, UT H. Jack Baskin, MD, MACE Orlando, FL Angelo Bertani, MD Reggio Emilio, Italy Antonio Bianchini, MD Albano (Rome), Italy Giancarlo Bizzarri, MD Albano (Rome), Italy Diana S. Dean, MD, FACE Rochester, MN Daniel S. Duick, MD, FACE Phoenix, AZ Olivier Esnault, MD Paris, France Debora Formisano, MS Reggio Emilio, Italy Andrea Frasoldati, MD Reggio Emilio, Italy Filomena Graziano, MD Albano (Rome), Italy Rinaldo Guglielmi, MD Albano (Rome), Italy Jill E. Langer, MD Philadelphia, PA
xiii
xiv
CONTRIBUTORS
Laurence Leenhardt, MD, PhD Paris, France Robert A. Levine, MD, FACE Nashua, NH Francesco Lonero, MD Albano (Rome), Italy Susan J. Mandel, MD, MPH Philadelphia, PA Claudio M. Pacella, MD Albano (Rome), Italy Sara Pacella, MD Albano (Rome), Italy Enrico Papini, MD Albano (Rome), Italy Marialaura Pesenti, MD Reggio Emilio, Italy Laura Raifa Al Jandali Rifa’y, MD Reggio Emilio, Italy Reagan Schiefer, MD Rochester, MN Roberto Valcavi, MD, FACE Reggio Emilio, Italy
CHAPTER 1
History of Thyroid Ultrasound Robert A. Levine
The thyroid is well suited to ultrasound study because of its superficial location, vascularity, size and echogenicity (1). In addition, the thyroid has a very high incidence of nodular disease, the vast majority benign. Most structural abnormalities of the thyroid need evaluation and monitoring, but not intervention (2). Thus, the thyroid was among the first organs to be well studied by ultrasound. The first reports of thyroid ultrasound appeared in the late 1960s. Between 1965 and 1970 there were seven articles published specific to thyroid ultrasound. In the last five years there have been over 1,300 published. Thyroid ultrasound has undergone a dramatic transformation from the cryptic deflections on an oscilloscope produced in A-mode scanning, to barely recognizable B-mode images, followed by initial low resolution gray scale, and now modern high resolution images. Recent advances in technology, including harmonic imaging, contrast studies, and three-dimensional reconstruction, have furthered the field. In 1880, Pierre and Jacques Curie discovered the piezoelectric effect, determining that an electric current applied across a crystal would result in a vibration that would generate sound waves, and that sound waves striking a crystal would, in turn, produce an electric voltage. Piezoelectric transducers were capable of producing sonic waves in the audible range and ultrasonic waves above the range of human hearing. The first operational sonar system was produced two years after the sinking of the Titanic in 1912. This system was capable of detecting an iceberg located two miles distant from a ship. A low-frequency audible pulse was generated, and a human operator listened for a change in the return echo. This system was able to detect, but not localize, objects within range of the sonar (3). Over the next 30 years navigational sonar improved, and imaging progressed from passive sonar, with an operator
1
2
R.A. LEVINE
listening for reflected sounds, to display of returned sounds as a one-dimensional oscilloscope pattern, to two-dimensional images capable of showing the shape of the object being detected. The first medical application of ultrasound occurred in the 1940s. Following the observation that very high intensity sound waves had the ability to damage tissues, lower intensities were tried for therapeutic uses. Focused sound waves were used to mildly heat tissue for therapy of rheumatoid arthritis, and early attempts were made to destroy the basal ganglia to treat Parkinson’s disease (4). The first diagnostic application of ultrasound occurred in 1942. In a paper entitled “Hyperphonagraphy of the Brain,” Karl Theodore Dussic reported localization of the cerebral ventricles using ultrasound. Unlike the current reflective technique, his system relied on the transmission of sound waves, placing a sound source on one side of the head, with a receiver on the other side. A pulse was transmitted, with the detected signal purportedly able to show the location of midline structures. While the results of these studies were later discredited as predominantly artifact, this work played a significant role in stimulating research into the diagnostic capabilities of ultrasound (4). Early in the 1950’s the first imaging by pulse–echo reflection was tried. A-mode imaging showed deflections on an oscilloscope to indicate the distance to reflective surfaces. Providing information in a single dimension, A-mode scanning indicated only distance to reflective surfaces (See Fig. 2.7) (5). A-mode ultrasonography was used for detection of brain tumors, shifts in the midline structures of the brain, localization of foreign bodies in the eye, and detection of detached retinas. In the first presage that ultrasound may assist in the detection of cancer, John Julian Wild published the observation that gastric malignancies were more echogenic than normal gastric tissue. He later studied 117 breast nodules using a 15MHz sound source, and reported that he was able to determine their size with an accuracy of 90%. During the late 1950s the first two-dimensional B-mode scanners were developed. B-mode scanners display a compilation of sequential A-mode images to create a two-dimensional image (See Fig. 2.2). Douglass Howry developed an immersion tank B-mode ultrasound system, and several models of immersion tank scanners followed. All utilized a mechanically driven transducer that would sweep through an arc, with an image reconstructed to demonstrate the full sweep. Later
HISTORY OF THYROID ULTRASOUND
3
advances included a hand-held transducer that still required a mechanical connection to the unit to provide data regarding location, and water-bag coupling devices to eliminate the need for immersion (6). Application of ultrasound for thyroid imaging began in the late 1960s. In July 1967 Fujimoto et al. reported data on 184 patients studied with a B-mode ultrasound “tomogram” utilizing a water bath (8). The authors reported that no internal echoes were generated by the thyroid in patients with no known thyroid dysfunction and nonpalpable thyroid glands. They described four basic patterns generated by palpably abnormal thyroid tissue. The type 1 pattern was called “cystic” due to the virtual absence of echoes within the structure, and negligible attenuation of the sound waves passing through the lesion. Type 2 was labeled “sparsely spotted,” showing only a few small echoes without significant attenuation. The type 3 pattern was considered “malignant” and was described as generating strong internal echoes. The echoes were moderately bright and were accompanied by marked attenuation of the signal. Type 4 had a lack of internal echoes but strong attenuation. In the patients studied, 65% of the (predominantly follicular) carcinomas had a type 3 pattern. Unfortunately, 25% of benign adenomas were also type 3. Further, 25% of papillary carcinomas were found to have the type 2 pattern. While the first major publication of thyroid ultrasound attempted to establish the ability to determine malignant potential, the results were nonspecific in a large percentage of the cases. In December 1971 Manfred Blum published a series of A-mode ultrasounds of thyroid nodules (Fig. 2.1) (5). He demonstrated the ability of ultrasound to distinguish solid from cystic nodules, as well as accuracy in measurement of the dimensions of thyroid nodules. Additional publications in the early 1970s further confirmed the capacity for both A-mode and B-mode ultrasound to differentiate solid from cystic lesions, but consistently demonstrated that ultrasound was unable to distinguish malignant from benign solid lesions with acceptable accuracy (9). The advent of gray scale display resulted in images that were far easier to view and interpret (7). In 1974 Ernest Crocker published “The Gray Scale Echographic Appearance of Thyroid Malignancy” (10). Using an 8MHz transducer with a 0.5 mm resolution, he described “low amplitude, sparse and disordered echoes” characteristic of thyroid cancer
4
R.A. LEVINE
when viewed with a gray scale display. The pattern felt to be characteristic of malignancy was what would now be considered “hypoechoic and heterogeneous.” Forty of the eighty patients studied underwent surgery. All six of the thyroid malignancies diagnosed had the described (hypoechoic) pattern. The percentage of benign lesions showing this pattern was not reported in the publication. With each advancement in technology, interest was again rekindled in ultrasound’s ability to distinguish a benign from a malignant lesion. Initial reports of ultrasonic features typically describe findings as being diagnostically specific. Later, reports followed showing overlap between various disease processes. For example, following an initial report that the “halo sign,” a rim of hypoechoic signal surrounding a solid thyroid nodule, was seen only in benign lesions (11), Propper reported that two of ten patients with this finding had carcinoma (12). As discussed in Chap. 6 the halo sign is still considered to be one of the numerous features that can be used in determining the likelihood of malignancy in a nodule. In 1977 Wallfish recommended combining fine-needle aspiration biopsy with ultrasound in order to improve the accuracy of biopsy specimens (13). Recent studies have continued to demonstrate that biopsy accuracy is greatly improved when ultrasound is used to guide placement of the biopsy needle. Most patients with prior “nondiagnostic” biopsies will have an adequate specimen when ultrasound-guided biopsy is performed (14). Ultrasound-guided fine-needle aspiration results in improved sensitivity and specificity of biopsies as well as a greater than 50% reduction in nondiagnostic and false negative biopsies (15). Current resolution allows demonstration of thyroid nodules smaller than 1 mm; thus ultrasound has clear advantages over palpation in detecting and characterizing thyroid nodular disease. Nearly 50% of patients found to have a solitary thyroid nodule by palpation will be shown to have additional nodules by ultrasound, and more than 25% of the additional nodules are larger than 1 cm (16). With a prevalence estimated between 19% and 35%, the management of incidentally detected, nonpalpable thyroid nodules remains controversial. Several guidelines have been developed to assist in deciding which nodules warrant biopsy and which may be monitored without tissue sampling. These guidelines are discussed in Chap. 7. Over the past several years the value of ultrasound in screening for suspicious lymph nodes prior to surgery in
HISTORY OF THYROID ULTRASOUND
5
patients with biopsy proven cancer has been established. Current guidelines for the management of thyroid cancer indicate a pivotal role for ultrasound in monitoring for locoregional recurrence (17). During the 1980’s Doppler ultrasound was developed, allowing detection of flow in blood vessels. As discussed in Chapter 3 the Doppler pattern of blood flow within thyroid nodules has an important role in assessing the likelihood of malignancy. Doppler imaging may also demonstrate the increased blood flow characteristic of Graves’ disease (18), and may be useful in distinguishing between Graves’ disease and thyroiditis, especially in pregnant patients or patients with amiodaroneinduced hyperthyroidism (19). Recent technological advancements include intravenous sonographic contrast agents, three-dimensional ultrasound imaging and elastography. Intravenous sonographic contrast agents are available in Europe, but remain experimental in the United States. All ultrasound contrast agents consist of microbubbles, which function both by reflecting ultrasonic waves and, at higher signal power, by reverberating and generating harmonics of the incident wave. Ultrasound contrast agents have been predominantly used to visualize large blood vessels, with less utility in enhancing parenchymal tissues. They have shown promise in imaging peripheral vasculature as well as liver tumors and metastases (20), but no studies have been published demonstrating an advantage of contrast agents in thyroid imaging. Three-dimensional display of reconstructed images has been available for CT scan and MRI for many years and has demonstrated practical application. While three-dimensional ultrasound has recently gained popularity for fetal imaging, its role in diagnostic ultrasound remains unclear. While obstetrical ultrasound has the great advantage of the target being surrounded by a natural fluid interface, 3D thyroid ultrasound is limited by the lack of a similar interface distinguishing the thyroid from adjacent neck tissues. It is predicted that breast biopsies will soon be guided in a more precise fashion by real time 3D imaging (21), and it is possible that, in time, thyroid biopsy will similarly benefit. At the present time, however, 3D ultrasound technology does not have a demonstrable role in thyroid imaging. Elastography is a new technique in which the compressibility of a nodule is assessed by ultrasound as external pressure is applied. With studies showing a good predictive
6
R.A. LEVINE
value for prediction of malignancy in breast nodules, recent investigations of its role in thyroid imaging have been promising. Additional prospective trials are ongoing to assess the role of elastography in predicting the likelihood of thyroid malignancy. With the growing recognition that real time ultrasound performed by an endocrinologist provides far more useful information than that obtained from a radiology report, office ultrasound by endocrinologists has gained acceptance. The first educational course specific to thyroid ultrasound was offered by the American Association of Clinical Endocrinologists (AACE) in 1998. Under the direction of Dr. Jack Baskin, 53 endocrinologists were taught to perform diagnostic ultrasound and ultrasound-guided fine-needle aspiration biopsy. By the turn of the century 300 endocrinologists had been trained. Endocrine University, established in 2002 by AACE, began providing instruction in thyroid ultrasound and biopsy to all graduating endocrine fellows. By the end of 2006 over 2,000 endocrinologists had completed the AACE ultrasound course. In 2007 AACE and the American Institute of Ultrasound Medicine (AIUM) began a collaborative effort for certification and accredidation in thyroid ultrasound. In the 35 years since ultrasound was first used for thyroid imaging, there has been a profound improvement in the technology and quality of images. The transition from A-mode to B-mode to gray scale images was accompanied by dramatic improvements in clarity and interpretability of images. Current high-resolution images are able to identify virtually all lesions of clinical significance. Ultrasound characteristics cannot predict benign lesions, but features including irregular margins, microcalcifications, and central vascularity may deem a nodule suspicious (3). Ultrasound has proven utility in the detection of recurrent thyroid cancer in patients with negative whole body iodine scan or undetectable thyroglobulin (17, 22). Recent advances including the use of contrast agents, tissue harmonic imaging, elastography, and multiplanar reconstruction of images will further enhance the diagnostic value of ultrasound images. The use of Doppler flow analysis may improve the predictive value for determining the risk of malignancy, but no current ultrasound technique is capable of determining benignity with an acceptable degree of accuracy. Ultrasound guidance of fine-needle aspiration biopsy has been demonstrated to improve both diagnostic yield and accuracy, and will likely become the standard of
HISTORY OF THYROID ULTRASOUND
7
care. Routine clinical use of ultrasound is often considered an extension of the physical examination by endocrinologists. High quality ultrasound systems are now available at prices that make this technology accessible to virtually all providers of endocrine care (3). References 1. Solbiati L, Osti V, Cova L, Tonolini M (2001) Ultrasound of the thyroid, parathyroid glands and neck lymph nodes. Eur Radiol 11(12):2411–2424 2. Tessler FN, Tublin ME (1999) Thyroid sonography: current applications and future directions. AJR 173:437–443 3. Levine RA (2004) Something old and something new: a brief history of thyroid ultrasound technology. Endocr Pract 10(3): 227–233. 4. Woo JSK Personal Communication. 5. Blum M, Weiss B, Hernberg J (1971) Evaluation of thyroid nodules by A-mode echography. Radiology 101:651–656 6. Skolnick ML, Royal DR (1975) A simple and inexpensive water bath adapting a contact scanner for thyroid and testicular imaging. J Clin Ultrasound 3(3):225–227 7. Scheible W, Leopold GR, Woo VL, Gosink BB (1979) Highresolution real-time ultrasonography of thyroid nodules. Radiology 133:413–417 8. Fujimoto F, Oka A, Omoto R, Hirsoe M (1967) Ultrasound scanning of the thyroid gland as a new diagnostic approach. Ultrasonics 5:177–180 9. Thijs LG (1971) Diagnostic ultrasound in clinical thyroid investigation. J Clin Endocrinol Metab 32(6):709–716 10. Crocker EF, McLaughlin AF, Kossoff G, Jellins J (1974) The gray scale echographic appearance of thyroid malignancy. J Clin Ultrasound 2(4):305–306 11. Hassani SN, Bard RL (1977) Evaluation of solid thyroid neoplasms by gray scale and real time ultrasonography: the “halo” sign. Ultrasound Med 4:323 12. Propper RA, Skolnick ML, Weinstein BJ, Dekker A (1980) The nonspecificity of the thyroid halo sign. J Clin Ultrasound 8:129–132 13. Walfish PG, Hazani E, Strawbridge HTG, Miskin M, Rosen IB (1977) Combined ultrasound and needle aspiration cytology in the assessment and management of hypofunctioning thyroid nodule. Ann Intern Med 87(3):270–274 14. Gharib H (1994) Fine-needle aspiration biopsy of thyroid nodules: advantages, limitations, and effect. Mayo Clin Proc 69:44–49 15. Danese D, Sciacchitano S, Farsetti A, Andreoli M, Pontecorvi A (1998) Diagnostic accuracy of conventional versus sonographyguided fine-needle aspiration biopsy in the management of nonpalpable and palpable thyroid nodules. Thyroid 8:511–515
8
R.A. LEVINE
16. Tan GH, Gharib H, Reading CC (1995) Solitary thyroid nodule: comparison between palpation and ultrasonography. Arch Intern Med 155:2418–2423 17. Cooper DS, Doherty GM, Haugen BR et al (2006) Management guidelines for patients with thyroid nodules and thyroid cancer. Thyroid 16(2)1–33 18. Ralls PW, Mayekowa DS, Lee KP et al (1988) Color-flow Doppler sonography in Graves’ disease: “thyroid inferno.” AJR 150:781– 784 19. Bogazzi F, Bartelena L, Brogioni S et al (1997) Color flow Doppler sonography rapidly differentiates type I and type II amiodaroneinduced thyrotoxicosis. Thyroid 7(4)541–545 20. Grant EG (2001) Sonographic contrast agents in vascular imaging. Semin Ultrasound CT MR 22(1):25–41 21. Lees W (2001) Ultrasound imaging in three and four dimensions. Semin Ultrasound CT MR 22(1):85–105 22. Antonelli A, Miccoli P, Ferdeghini M (1995) Role of neck ultrasonography in the follow-up of patients operated on for thyroid cancer. Thyroid 5(1):25–28
CHAPTER 2
Thyroid Ultrasound Physics Robert A. Levine
SOUND AND SOUND WAVES Some animal species such as dolphins, whales, and bats are capable of creating a “visual” image based on receiving reflected sound waves. Man’s unassisted vision is limited to electromagnetic waves in the spectrum of visible light. Humans require technology and an understanding of physics to use sound to create a picture. This chapter will explore how man has developed a technique for creating a visual image from sound waves (1). Sound is transmitted as mechanical energy, in contrast to light, which is transmitted as electromagnetic energy. Unlike electromagnetic waves, sound waves require a propagating medium. Light is capable of traveling through a vacuum, but sound will not transmit through a vacuum. The qualities of the transmitting medium directly affect how sound is propagated. Materials have different speeds of sound transmission. Speed of sound is constant for a specific material and does not vary with sound frequency (Fig. 2.1). Acoustic impedance is the inverse of the capacity of a material to transmit sound. Acoustic impedance of a material depends on its density, stiffness and speed of sound. When sound travels through a material and encounters a change in acoustic impedance a portion of the sound energy will be reflected, and the remainder will be transmitted. The amount reflected is proportionate to the degree of mismatch of acoustic impedance. Sound waves propagate by compression and rarefaction of molecules in space (Fig. 2.2). Molecules of the transmitting medium vibrate around their resting position and transfer their energy to neighboring molecules. Sound waves carry energy rather than matter through space. As shown in Fig. 2.2, sound waves propagate in a longitudinal direction, but are typically represented by a sine wave
9
R.A. LEVINE
10 4500
4080
4000 3500
speed of sound
3000 m/sec
2500 2000
1580
1550
1540
1560
1570
1500
1450
1480
1000 330
500 0
Bone Muscle Liver
Soft Kidney Blood Tissue average
Fat
Water
Air
FIG. 2.1. Speed of sound. The speed of sound is constant for a specific material and does not vary with frequency. Speed of sound for various biological tissues is illustrated
FIG. 2.2. Sound waves propagate in a longitudinal direction but are typically represented by a sine wave where the peak corresponds to the maximum compression of molecules in space, and the trough corresponds to the maximum rarefaction
where the peak corresponds to the maximum compression of molecules in space, and the trough corresponds to the maximum rarefaction. Frequency is defined as the number of cycles per time of the vibration of the sound waves. A Hertz (Hz) is defined as one cycle per second. The audible spectrum is between 30 and 20,000 Hz. Ultrasound is defined as sound waves at a higher frequency than the audible spectrum. Typical frequencies used in diagnostic ultrasound vary between five million and 15 million cycles per second (5 MHz and 15 MHz).
THYROID ULTRASOUND PHYSICS
11
Diagnostic ultrasound uses pulsed waves, allowing for an interval of sound transmission, followed by an interval during which reflected sounds are received and analyzed. Typically three cycles of sound are transmitted as a pulse. The spatial pulse length is the length in space that three cycles fill (Fig. 2.3). Spatial pulse length is one of the determinants of resolution. Since higher frequencies have a smaller pulse length, higher frequencies are associated with improved resolution. As illustrated in Fig. 2.3, at a frequency of 15 MHz the wavelength in biological tissues is approximately 0.1 mm, allowing an axial resolution of 0.15 mm. As mentioned above, the speed of sound is constant for a given material or biological tissue. It is not affected by frequency or wavelength. It increases with stiffness and decreases with density of the material. As seen in Fig. 2.1, common biologic tissues have different propagation velocities. Bone, as a very dense and stiff tissue, has a high propagation velocity of 4,080 meters per second. Fat tissue, with low stiffness and low density, has a relatively low speed of sound of 1,450 m per second. Most soft tissues have a speed of sound near 1,540 m per second. Muscle, liver and thyroid have a slightly faster speed of sound. By convention, all ultrasound equipment uses an average speed of 1,540 meters per second. The distance to an object displayed on an ultrasound image is calculated by multiplying the speed of sound by the time interval for a sound signal to
FIG. 2.3. Diagnostic ultrasound uses pulsed waves, allowing for an interval of sound transmission, followed by an interval during which reflected sounds are received and analyzed. Typically three cycles of sound are transmitted as a pulse
12
R.A. LEVINE
FIGS. 2.4–2.6. Most biological tissues have varying degrees of inhomogeneity both on a cellular and macroscopic level. Connective tissue, blood vessels, and cellular structure all provided mismatches of acoustic impedance that lead to the generation of characteristic ultrasonographic patterns. FIG. 2.4. demonstrates the echotexture from normal thyroid tissue. It has a ground glass appearance and is brighter than muscle tissue.
THYROID ULTRASOUND PHYSICS
13
return to the transducer. By using the accepted 1,540 m per second as the assumed speed of sound, all ultrasound equipment will provide identical distance or size measurements. Reflection is the redirection of a portion of a sound wave from the interface of tissues with unequal acoustic impedance. The greater the difference in impedance, the greater the amount of reflection. A material that is homogeneous in acoustic impedance does not generate any internal echoes. A pure cyst is a typical example of an anechoic structure. Most biological tissues have varying degrees of inhomogeneity both on a cellular and macroscopic level. Connective tissue, blood vessels and cellular structure all provide mismatches of acoustic impedance that lead to the generation of characteristic ultrasonographic patterns (Figs. 2.4–2.6). Reflection is categorized as specular when reflecting off of smooth surfaces such as a mirror. In contrast, diffuse reflection occurs when a surface is irregular, with variations at or smaller than the wavelength of the incident sound. Diffuse reflection results in scattering of sound waves and production of noise. CREATION OF AN ULTRASOUND IMAGE The earliest ultrasound imaging consisted of a sound transmitted into the body, with the reflected sound waves displayed on an oscilloscope. Referred to as A-mode ultrasound, these images in the 1960s and 1970s were capable of providing measurements of internal structures such as thyroid lobes, nodules and cysts. Fig. 2.7a shows an A-mode ultrasound image of a solid thyroid nodule. Scattered echoes are present from throughout the nodule. Fig. 2.7b shows the image from a cystic nodule. The initial reflection is from the proximal wall of the cyst, with no significant signal reflected by the cyst fluid. The second reflection originates from the posterior wall. Fig. 2.7c shows the A-mode image from a complex nodule with solid and cystic components. A-mode ultrasound was capable of providing size measurements in one dimension, but did not provide a visual image of the structure.
FIGS. 2.4–2.6. (Continued) FIG. 2.5. shows the thyroid from a patient with the acutely swollen inflammatory phase of Hashimoto’s thyroiditis. Massive infiltration by lymphocytes has decreased the echogenicity of the tissue resulting in a more hypoechoic pattern. FIG. 2.6. shows a typical heterogeneous pattern from Hashimoto’s thyroiditis with hypoechoic inflammatory regions separated by hyperechoic fibrous tissue
14
R.A. LEVINE
FIG. 2.7. A-mode ultrasound images. a. shows an A-mode ultrasound image of a solid thyroid nodule. Scattered echoes are present from throughout the nodule. b. shows the image from a cystic nodule. The initial reflection is from the proximal wall of the cyst, with no significant signal reflected by the cyst fluid. The second reflection originates from the posterior wall. c. shows the A-mode image from a complex nodule with solid and cystic components
In order to provide a visual two-dimensional image, a series of one-dimensional A-mode images are aligned as a transducer is swept across the structure being imaged. Early thyroid ultrasound images were created by slowly moving a transducer across the neck. By scanning over a structure and aligning the A-mode images, a two-dimensional image is formed. The twodimensional image formed in this manner is referred to as a B-Mode scan (Fig. 2.8). Current ultrasound transducers use a series of piezoelectric crystals in a linear array to electronically simulate a sweep of the transducer. Firing sequentially, each crystal sends a pulse of sound wave into the tissue and receives subsequent reflections. The final ultrasound image reflects a cross sectional image through the tissue defined by the thin flat beam of sound emitted from the transducer. Resolution is the ability to distinguish between two separate, adjacent objects. For example, with a resolution of 0.2 mm, two adjacent objects measuring <2 mm would be shown as a single object. Objects smaller than the resolution will not be realistically imaged. THE USEFULNESS OF ARTIFACTS IN ULTRASOUND IMAGING A number of artifacts commonly occur in ultrasound images. Unlike most other imaging techniques, artifacts are very helpful in interpreting ultrasound images (2). Artifacts, such as shadows behind objects or unexpected areas of brightness,
THYROID ULTRASOUND PHYSICS
15
FIG. 2.8. A B-mode ultrasound image is composed of a series of A-mode images aligned to provide a two dimensional image
can provide additional understanding of the properties of the materials being imaged. When sound waves impact on an area of extreme mismatch of acoustic impedance, such as a tissue-air interface or a calcification, the vast majority of the sound waves are reflected, providing a very bright signal from the object’s surface and an absence of imaging beyond the structure. Fig. 2.9 demonstrates acoustic shadowing behind a calcified nodule. Fig. 2.10 illustrates a coarse calcification within the thyroid parenchyma with acoustic shadowing behind the calcification. Fig. 2.11 shows the typical appearance of the trachea on an ultrasound image. Because there is no transmission of sound through the air-tissue interface of the anterior wall of the trachea, no imaging of structures posterior to the trachea occurs. Conversely, a cystic structure transmits sound with very little attenuation, resulting in a greater intensity of sound waves behind it, compared to adjacent structures. This results in acoustic enhancement with a brighter signal behind a cystic or anechoic structure. This enhancement can be used to distinguish between a cystic and solid nodule within the thyroid. Fig. 2.12 illustrates enhancement behind a cystic nodule. Enhancement is not limited to cystic nodules, however. Any structure that causes minimal attenuation of the ultrasound signal will have enhancement posterior to it. Fig. 2.13 illustrates enhancement behind a solid parathyroid adenoma. Fig. 2.14 illustrates enhancement behind a benign colloid nodule. Due to the high content of fluid
16
R.A. LEVINE
FIG. 2.9. Acoustic shadowing. When sound waves impact on an area of extreme mismatch of acoustic impedance, such as a calcification, the vast majority of the sound waves are reflected, resulting in a shadow beyond the structure. This calcified nodule is from a patient with familial papillary carcinoma
FIG. 2.10. Acoustic shadowing. A shadow is observed behind a coarse calcification within the thyroid parenchyma. Unlike calcification within a nodule, amorphic calcification within the parenchyma is not typically associated with malignancy
and colloid within the nodule, and resultant decrease in cellularity, there is less attenuation of signal within the nodule than within the surrounding thyroid tissue. Shadowing and enhancement, as described above, are examples of attenuation artifacts. Shadowing occurs behind
THYROID ULTRASOUND PHYSICS
17
FIG. 2.11. Acoustic shadowing. Due to the extreme reflection from the tissue-air interface of the trachea, no image is seen behind the trachea on an anterior ultrasound
FIG. 2.12. Enhancement. A cystic structure transmits sound with very little attenuation, resulting in a greater intensity of sound waves behind it. Enhancement is typical behind a cystic nodule
structures with extreme acoustic mismatch due to the attenuation of transmission of sound waves caused by nearly complete reflection. Enhancement occurs behind structures with little to no attenuation, with higher intensity sound waves present behind the structure in comparison to the adjacent tissues.
18
R.A. LEVINE
FIG. 2.13. Enhancement. Parathyroid adenomas have relatively homogeneous tissue and, like a parathyroid cyst, may demonstrate enhancement behind them
FIG. 2.14. Enhancement. This benign colloid nodule has a high content of fluid and colloid with a resulting decrease in cellularity. The decreased attenuation of signal within the nodule results in enhancement despite it being a solid nodule
Fig. 2.15 shows a nodule exhibiting “eggshell” calcification. A layer of calcium surrounding the nodule results in an absence of reflected signal behind the nodule. As can be seen in the figure, reflection is greatest from the surfaces perpendicular to the sound waves: the front and back walls.
THYROID ULTRASOUND PHYSICS
19
FIG. 2.15. Eggshell calcification. A layer of calcium surrounding the nodule results in reflection from the surface, along with marked posterior acoustic shadowing
Because the angle of incidence approaches 180° along the side walls, most of the reflected waves are reflected away from the transducer, resulting in a decreased signal corresponding to the sides of the structure. Edge artifacts are extremely useful in identifying nodules in the thyroid. Fig. 2.16 shows dark lines extending posteriorly from the sides of a nodule, aligned with the ultrasound beam. This is another example of a reflection artifact. As described above, the sound waves striking the object along the side are reflected away, rather than back toward the transducer. When two parallel dark lines are seen aligned vertically in an image they can be followed “up” to help identify a nodule or other structure. Several artifacts arise due to reverberation. When sound waves reflect off of a very reflective surface some may be re-reflected from the skin surface, producing multiple phantom images beyond the actual image. Fig. 2.17 illustrates the very common reverberation artifact that occurs due to this reverberation of sound waves between the skin surface and deeper tissue interfaces. Since some of the reflected sound waves will bounce back from the skin surface into the tissue multiple times, phantom images are produced. As shown, it is very common to see this artifact in the anterior aspect of cysts, raising doubt as to whether the lesion is a true cyst or partly solid. Changing the angle at which the sound strikes the lesion
20
R.A. LEVINE
FIG. 2.16. Edge artifact. Dark lines are seen extending posteriorly from the sides of a nodule. This artifact can be used to help identify a nodule or other structure
FIG. 2.17. Reverberation artifact. It is very common to see this artifact in the anterior aspect of cysts. This arises due to reverberation of signal between the skin surface and the anterior wall of the cyst, resulting in the late signals being received, and giving the appearance of solid tissue in the anterior aspect of the cyst
will usually clarify the situation. Fig. 2.18 shows this common artifact behind the anterior wall of the trachea. The “comet tail” artifact is another extremely common finding caused by reverberation (Figs. 2.19–2.20). Colloid nodules
THYROID ULTRASOUND PHYSICS
21
FIG. 2.18. Reverberation artifact. Numerous parallel lines are seen posterior to the anterior wall of the trachea. These are commonly misconstrued as the tracheal rings, but are actually reverberation artifacts
FIG. 2.19. “Comet tails.” Colloid nodules may contain tiny crystals resulting from the desiccation of the gelatinous colloid material. Reflection of the sound waves off of the crystal results in a bright spot. However, in contrast to a soft tissue calcification, the crystals begin to vibrate under the influence of the ultrasound energy. The vibration generates sound waves that return to the transducer after the initial reflected signal
may contain tiny crystals resulting from the desiccation of the gelatinous colloid material. Reflection of the sound waves off of a crystal results in a bright spot. However, in contrast to a soft tissue calcification, the crystals begin to vibrate under the
22
R.A. LEVINE
FIG. 2.20. “Comet tails.” Another example of comet tail artifacts within a benign colloid nodule
influence of the ultrasound energy. The vibration generates sound waves, which return to the transducer after the initial reflected signal. Also referred to as a ringdown artifact, a cat’s eye (Fig. 2.21), or stepladder artifact, these comet tails help differentiate between the typically benign densities found in a colloid nodule and highly suspicious microcalcifications. While comet tail artifacts most commonly arise within a benign colloid nodule, they may also be seen in resolving hematomas, and have rarely been described within papillary carcinoma. Refraction is the alteration of direction of the transmitted sound at an acoustic interface when the angle of incidence is not 90°. A sound wave striking an interface at 90° is reflected straight back. When waves strike at an angle other than 90° the transmitted wave is bent as it propagates through the interface. A greater degree of mismatch of acoustic impedance between tissues results in a greater degree of refraction. While not seen in near field ultrasound as used in thyroid and other small parts imaging, refraction artifacts can result in a second “ghost” image when a refracting object exists in the path of an ultrasound beam. As sound waves propagate through any tissue, the intensity of the wave is attenuated. Attenuation of acoustic energy results from a combination of reflection, scattering and absorption, with conversion of sound energy to heat. Attenuation is frequency dependent, with higher frequencies having greater
THYROID ULTRASOUND PHYSICS
23
FIG. 2.21. “Cat’s-eye” artifact. The comet tail artifact is also referred to as a ringdown artifact, a stepladder artifact, or, when a single lesion is seen within a small cyst, a cat’s-eye
attenuation. As a result, while higher frequencies provide improved resolution, the depth of imaging decreases with increasing frequency. Current ultrasound technology utilizes sound waves as high as 16 MHz for thyroid imaging. However, imaging is limited to less than 5 cm of depth at this frequency. Visualization of deeper structures, as with abdominal or pelvic ultrasound, requires lower frequencies. In obese patients, or when imaging very deep structures, frequencies of 7.5 MHz– 10 MHz may be needed for adequate penetration and visualization of the deep neck structures. Figs. 2.22 and 2.23 compare images made at 7.5 MHz and 15 MHz. Loss of detail of proximal structures is evident with the lower frequency. Resolution is the ability to discriminate two adjacent small structures from one larger mass. Lateral resolution refers to the ability to discriminate in a transverse, or side to side, direction. Azimuthal resolution refers to the image perpendicular to the axis of the ultrasound beam. Axial resolution is the ability to discriminate objects along the path of the ultrasound beam. Axial resolution is determined by the spatial pulse length and therefore frequency. Lateral and azimuthal resolution are dependent on the focusing of the ultrasound beam. Ultrasound transducers consist of an array of crystals capable of transmitting and receiving ultrasound energy. Piezoelectric crystals vibrate when exposed to an electrical current. Conversely, when energy strikes the crystal, it results in an electrical signal,
24
R.A. LEVINE
FIGS. 2.22–2.23. Comparison of images produced with frequencies of 7.5 MHz and 13 MHz. The nodule is much more defined in the high frequency image, but the posterior structures are far more evident in the low frequency image
with a frequency corresponding to the frequency of the incident sound wave. Thyroid ultrasound typically uses a linear array of crystals within a transducer, most often utilizing 128 aligned crystals in a linear array. The transverse width of the
THYROID ULTRASOUND PHYSICS
25
image produced is equal to the length of the array of crystals in the transducer. Curved array transducers are less often used in thyroid ultrasound (but are commonly used in abdominal, pelvic and cardiac imaging). By producing a divergent beam of ultrasound they allow visualization of structures larger than the transducer. They are occasionally used as an aid in fine needle aspiration biopsy, but the image produced has spatial distortion due to the lack of a linear relationship between the transverse and longitudinal planes. Once received, the ultrasound signal undergoes image reconstruction, followed by image enhancement. Noise reduction and edge sharpening algorithms are used to clarify the image. Most ultrasound equipment allows the user to select the degree of noise reduction, dynamic range and edge sharpening to optimize image quality. Ultrasound equipment allows for user adjustment of the gain of the received signal. Overall gain can be adjusted, and separate channels corresponding to individual depths may be adjusted (time gain compensation) to provide the best image quality at the region of interest. Most ultrasound equipment also allows for user adjustment of the focal zone, the depth at which the ultrasound beam is ideally focused. Multiple focal zones may also be selected on most ultrasound equipment. While providing a slight increase in image sharpness, the use of multiple focal zones typically slows the refresh rate of the image, resulting in a more jumpy image when visualized during real time scanning. While standard ultrasound receives only the frequency identical to that transmitted for imaging, tissue harmonic imaging capitalizes on the tendency of tissues to reverberate when exposed to higher power ultrasound energy. Different tissues have a different degree of reverberation and produce unique signatures of tissue harmonics (multiples of the original frequency). Selective detection of the harmonic signal produces an alternative image. Because higher frequencies are being detected, the resolution may be improved, but the original transmitted frequency is typically lower when using tissue harmonic imaging. Since the distance traveled by a harmonic signal is one half that of the transmitted and received signal there is less noise. The increased resolution and decreased noise may result in increased conspicuity of some objects, but tissue harmonic imaging has not had widespread application in thyroid imaging. In summary, sound transmission is dependent on the conducting medium. Sound is reflected at interfaces of mismatch
26
R.A. LEVINE
of acoustic impedance. The resolution of an ultrasound image is dependent on the frequency, the focused beam width, and the quality of the electronic processing. Resolution improves with higher frequencies, but the depth of imaging suffers. Image artifacts such as shadowing and enhancement provide useful information, rather than just interfering with creation of a clear image. The current image quality and ease of performance make real time ultrasound an integral part of the clinical evaluation of the thyroid patient. References Levine RA (2004) Something old and something new: a brief history of thyroid ultrasound technology. Endocr Pract 10(3):227–233 Meritt CRB (1998) Physics of Ultrasound. In: Rumack CM, Wilson SR, Charboneau JW (eds) Diagnostic Ultrasound, 2nd edn. Mosby, St. Louis, pp 3–34
CHAPTER 3
Doppler Ultrasound Robert A. Levine
DOPPLER ULTRASOUND, PHYSICAL PRINCIPLES The Doppler shift is a change in frequency that occurs when sound (or light) is emitted from, or bounced off of, a moving object. When a moving target reflects a sound the frequency of the reflected sound wave is altered. The frequency is shifted up by an approaching target and shifted down by a receding target. This is illustrated in Fig. 3.1. The amount the frequency is shifted is proportional to the velocity of the moving object. Because the Doppler shift was originally described for energy in the visible light spectrum, an upward Doppler shift is referred to as a blue shift, (a shift to a higher visible light frequency) and a downward Doppler shift is referred to as a red shift. Ultrasound utilization of the Doppler shift falls into three main categories. Analysis of the Doppler frequency spectrum allows for calculation of velocity, and is used in vascular studies. Color-flow Doppler and power Doppler superimpose a color image representing motion onto a B-mode image to illustrate location of motion (blood flow). In thyroid ultrasound, Doppler imaging is used predominantly to assess the vascularity of tissues. The leading use is to help determine the likelihood of a thyroid nodule being malignant. However, other applications of Doppler imaging include assessing the etiology or subtype of amiodarone thyrotoxicosis, clarifying images and helping to assess the etiology of hyperthyroidism. Analysis of the Doppler spectrum allows for determination of flow velocity and calculation of resistance to flow. By analyzing the waveform, the peak systolic velocity and diastolic velocity can be calculated. Resistive index and pulsatility index can be derived from these measurements. While these values are typically used in studies of peripheral vascular disease, the peak flow velocity and resistive index are occasionally used in reporting the degree of vascularity of thyroid tissue.
27
28
R.A. LEVINE
FIG. 3.1. Illustration of the Doppler shift. When a moving target reflects a sound, the frequency of the reflected sound wave is altered. The frequency is shifted up by an approaching target and shifted down by a receding target. The amount the frequency is shifted is proportional to the velocity of the moving object
For most thyroid imaging, color-flow Doppler and power Doppler are used. In color-flow Doppler a unique color (or brightness) is assigned to an individual frequency. Typically a greater frequency shift (corresponding to a higher velocity) is assigned a brighter color. Analysis of the color-flow image gives a graphic illustration of the direction and speed of blood flow within soft tissue. In contrast, power Doppler considers all frequency shifts to be equivalent, integrating the total amount of motion detected. The assigned color represents the total amount of flow present, independent of the velocity. The color image, therefore, is indicative of the total amount of flow present, without information regarding velocity. Color-flow Doppler provides information regarding both direction and velocity, and is more useful in vascular studies. In contrast, power Doppler does not provide information regarding velocity. However, it has increased sensitivity for the detection of low degrees of flow, has less noise interference, and is less dependent on the angle of incidence between the
DOPPLER ULTRASOUND
29
FIG. 3.2. Color Doppler. In color-flow Doppler a unique color (or brightness) is assigned to an individual frequency. Typically a greater frequency shift (corresponding to a higher velocity) is assigned a brighter color. Analysis of the color-flow image gives a graphic illustration of the direction and speed of blood flow within soft tissue
FIG. 3.3. Power Doppler. In contrast, power Doppler considers all frequency shifts to be equivalent, integrating the total amount of motion detected. The assigned color represents the total amount of flow present, independent of the velocity. Power Doppler has increased sensitivity for the detection of low degrees of flow, has less noise interference and is less dependent on the angle of incidence between the ultrasound waves and the moving object
30
R.A. LEVINE
ultrasound waves and the moving object. Power Doppler is generally the preferred imaging technique for assessing the vascularity of thyroid tissue (1). DOPPLER ULTRASOUND OF THYROID NODULES Fig. 3.4 shows a follicular carcinoma in the inferior pole of the thyroid, with a very high degree of blood flow. The inferior thyroid artery can be seen feeding the nodule. In contrast, Fig. 3.5 shows a nodule with no significant intranodular vascularity, with only scattered blood vessels around the periphery. This nodule was a benign follicular adenoma. Color and power Doppler imaging have been shown to have predictive value for determining the probability of malignancy in thyroid nodules. Most benign nodules have absent intranodular blood flow on power Doppler analysis, and most malignancies have demonstrable central flow (2,3). However, the negative predictive value is only 88%, and a negative study does not eliminate the need for biopsy (3,4). Several studies have assessed whether Doppler imaging plays a role in the prediction of probability of malignancy. Papini (2) studied 494 consecutive patients with nonpalpable nodules measuring 8 to 15 mm. All patients had a Doppler ultrasound study performed prior to fine needle aspiration biopsy. An intranodular vascular pattern was observed in 74% of all nodules with thyroid cancer. Eighty-seven percent of the cancers were solid and hypoechoic, and 77% of the cancers
FIG. 3.4. Vascular nodule. A follicular carcinoma is present in the inferior pole of the thyroid, with a very high degree of blood flow. The inferior thyroid artery can be seen feeding the nodule
DOPPLER ULTRASOUND
31
FIG. 3.5. Avascular nodule. This benign follicular adenoma has no intranodular blood flow
had irregular or blurred margins. Only 29% of the cancers had microcalcifications. Independent risk factors for malignancy included irregular margins (RR = 16.8%), intranodular Doppler blood flow (RR = 14.3) and microcalcifications (RR = 5). Berni (3) analyzed 108 patients with thyroid nodules demonstrated to be hypofunctioning on nuclear medicine study. All of the patients had subsequent surgical excision of the nodule. Half of the patients were found to have malignancy, so this clearly was not a random population. Of the 108 patients, 92 would have been diagnosed based on their color Doppler pattern. There were six false negative cancers with no blood flow, and 10 false positive benign lesions with significant intranodular flow. The calculated sensitivity was 88.8%, and the specificity was 81.5%. The positive predictive value of blood flow was 83%, and the negative predictive value was 88%. Fukunari (5) studied 310 patients with a solitary thyroid nodule in which a prior fine needle aspiration biopsy had demonstrated a follicular lesion. All patients underwent a color Doppler flow mapping study prior to surgery. The amount of flow in the nodule was classified on a four-point scale. Grade 1 nodules had no flow detectable. Grade 2 nodules had only peripheral flow, without intranodular flow. Grade 3 nodules had low velocity central flow, and grade 4 nodules had highintensity central flow. (Figs. 3.6–3.9) For purposes of statistical analysis the absence of intranodular flow (grades 1–2) was
32
R.A. LEVINE
FIG. 3.6. Grade 1 Doppler flow. Grade 1 lesions have no intranodular flow and no flow to the periphery
FIG. 3.7. Grade 2 Doppler flow. Grade 2 lesions have peripheral flow only, without intranodular flow
considered a negative result, and the presence of central flow (grades 3–4) was considered a positive result. Of 177 benign adenomatous nodules, 95% were grade 1 or 2, and only 5% were grade 3. No benign adenomatous nodules had grade 4 blood flow. Of 89 benign follicular adenomas, 66% showed grade 1 or 2 Doppler flow, and 34% showed grade 3 or 4 flow.
DOPPLER ULTRASOUND
33
FIG. 3.8. Grade 3 Doppler flow. Great 3 lesions have low to moderate velocity central flow
Of the 44 follicular carcinomas none showed grade 1 Doppler flow, 13.6% showed grade 2 flow and 86.4% showed either grade 3 or 4 flow. Using the data of Fukunari, the sensitivity of intranodular blood flow in predicting malignancy was 86%. The specificity was 85%, and the diagnostic accuracy was 81%. The prevalence of cancer in this group of follicular nodules was 14%. In a similar analysis, De Nicola (6) studied 86 patients in whom nodules had prior follicular biopsies. The flow pattern was characterized on a scale from 0 to 4, with 0 defined as no visible flow, 1 as peripheral flow only, 2 as peripheral flow with a small amount of central flow, 3 as peripheral flow plus extensive intranodular flow and 4 as central flow only. Patterns 0–2 were grouped as negative results, and nodules with pattern 3–4 were considered positive. Of 59 non-neoplastic nodules, 93% were grade 0–2, and only 7% were grade 3. No non-neoplastic nodules had grade 4 blood flow. Of 14 benign follicular adenomas 71% showed grade 0–2 Doppler flow, and 29% showed grade 3 or 4 flow. Of the 10 carcinomas, 20% showed grade 0–2 Doppler flow, and 80% showed grade 3 or 4 flow. Based on this analysis, sensitivity was 80%, and the specificity was 89%. Applying Bayes’ theorem to the data of Fukinari and De Nicola suggests that follicular nodules with no intranodular flow have only a 3% probability of malignancy rather than
34
R.A. LEVINE
FIG. 3.9a–b. Grade 4 Doppler flow. Grade 4 lesions have high-intensity central blood flow
the generally accepted 15%–20% likelihood in unselected follicular nodules. Conversely, vascular follicular nodules have a probability of malignancy approaching 50% (7). Color and power Doppler imaging provide useful information regarding the likelihood of malignancy in thyroid nodules. The predictive power may be stronger for nodules with a prior follicular biopsy than for all unselected nodules, as benign hyperplastic colloid nodules frequently show very intense intranodular flow (Fig. 3.10). The negative predictive value of a Doppler study is not sufficient to obviate the need for fine needle aspiration biopsy. Absence of flow in a nodule with a follicular biopsy makes malignancy less likely, however.
DOPPLER ULTRASOUND
35
FIG. 3.10. Grade 4 Doppler flow in a benign nodule. While high-intensity blood flow within the nodule raises the likelihood of malignancy, benign hyperplastic colloid nodules may also show very intense intranodular blood flow. Note the similarity of this image of a benign nodule to the malignant nodule shown in Fig. 9a
The power Doppler flow pattern should be interpreted along with other ultrasonographic characteristics including echogenicity, edge definition and calcifications, as well as clinical features such as nodule size and sex of the patient, to help in the decision regarding the need for biopsy and the need for, and extent of, surgery. DOPPLER ULTRASOUND OF THYROIDITIS Doppler ultrasound may have a role in the differentiation of the etiology of amiodarone-induced thyrotoxicosis. Type 1 amiodarone thyrotoxicosis resembles Graves’ disease. It typically occurs in patients with pre-existing thyroid autoimmunity. The gland is hyperthyroid, overproducing thyroid hormone. It may respond to treatment with thionomides and perchlorate. Typically type 1 amiodarone-induced thyrotoxicosis is associated with normal or increased vascularity on Doppler ultrasound. Type 2 amiodarone-induced thyrotoxicosis more closely resembles painless thyroiditis. In this entity, inflammation and destruction of thyroid tissue results in the release of preformed thyroid hormone. It may respond to glucocorticoid therapy, and typically does not respond to thionomides or perchlorate. An elevated interleukin-6 level has been described as indicative of type 2 amiodarone thyrotoxicosis, but its
36
R.A. LEVINE
predictive value is poor. Typically, type 2 amiodarone thyrotoxicosis is associated with absent or a very low degree of vascularity on power Doppler analysis (8). Bogazzi demonstrated a 58% response to steroid therapy when flow was absent on color Doppler imaging, and only a 14% steroid response rate when flow was present (9). A suggested treatment algorithm, using power Doppler analysis, suggests the use of steroid therapy when no flow is present. If flow is present, and especially if extremely vascular, thionomides with or without perchlorate are recommended. Combined therapy or surgery should be considered for any patient who does not respond to the initial treatment protocol. With the recognition that the Doppler image was useful in amiodarone-induced thyrotoxicosis, it seemed likely that Doppler would be useful for distinguishing hyperthyroidism of Graves’ disease from thyrotoxicosis due to thyroiditis. Graves’ disease has been described as the “thyroid inferno” (10), with intense blood flow and peak systolic velocity up to 20 cm/second (Fig. 11). However, as discussed in Chap. 5, the vascular pattern observed in thyroiditis is extremely variable, ranging from totally absent to extremely hypervascular. Fig. 3.11 illustrates hypervascularity associated with Graves’ disease. Fig. 3.12 shows the thyroid from a patient with postpartum thyroiditis, in this case demonstrating extreme hypervascularity. Fig. 3.13 shows intense blood flow in a patient with Hashitoxicosis,
FIG. 3.11. Graves’ disease. Graves’ disease has been described as the “thyroid inferno,” typically showing very intense blood flow
DOPPLER ULTRASOUND
37
FIG. 3.12. Postpartum thyroiditis. Thyroiditis may be associated with any level of vascular flow, ranging from totally absent to intense, as is seen in this patient with postpartum thyroiditis
FIG. 3.13. Hashitoxicosis. Hashimoto’s thyroiditis may also be associated with any degree of vascular flow. This patient with the hyperthyroid phase of early Hashimoto’s thyroiditis (hashitoxicosis) has intense blood flow and could be easily confused with Graves’ disease
and Fig. 3.14 shows low vascularity in Subacute thyroiditis. As Graves’ disease typically has very intense flow, the absence of flow in a thyrotoxic patient would be suggestive of thyroiditis. However, the presence of normal or increased blood flow may
38
R.A. LEVINE
FIG. 3.14. Subacute thyroiditis. Subacute thyroiditis may also be associated with any degree of vascular flow. This image shows decreased flow associated with subacute thyroiditis
reflect either Graves’ disease or thyroiditis in a thyrotoxic patient. Doppler imaging may be useful in cases of thyrotoxicosis factitia in which minimal intrathyroidal vascular flow is observed. DOPPLER ULTRASOUND FOR IMAGE CLARIFICATION Doppler imaging may be useful to clarify images. For example, the margins of an isoechoic nodule may be difficult to discern, but Doppler imaging often shows peripheral vascularity, helping to identify the boundaries of the nodule (Figs. 3.15–3.16). What appear to be small hypoechoic nodules may in fact be small intrathyroidal blood vessels, apparent only when Doppler imaging is used. Doppler imaging is also useful prior to biopsy, to avoid laceration of large feeding vessels (Fig. 3.17). Doppler imaging may help distinguish benign from malignant lymph nodes. As discussed in Chap. 7, in a normal lymph node the vascular supply enters centrally at the hilum and spreads along the long axis within the hilar line. In malignant lymph nodes, aberrant vessels enter peripherally through the lymph node capsule. Increased and disordered vascularity may be seen both peripherally and centrally. Figs. 3.18 and 3.19 demonstrate the vascular pattern in benign and malignant lymph nodes. Doppler flow study may indicate compression of the jugular vein by a malignant lymph node, as seen in Fig. 3.20. As discussed in Chap. 9, parathyroid adenomas frequently have a pulsatile polar artery, and Doppler imaging may be
DOPPLER ULTRASOUND
39
FIGS. 3.15–3.16. The margins of the nodule are not entirely clear in FIG. 3.15. With application of power Doppler ultrasound the boundaries of the nodule become evident, as seen in FIG. 3.16
useful in establishing that a nodule posterior to the thyroid represents a parathyroid rather than a central compartment lymph node. SUMMARY In summary, Doppler ultrasound plays an important role in thyroid imaging. Power Doppler imaging plays an important role in the prediction of the likelihood of malignancy in a thyroid nodule and should be documented for all significant thyroid
40
R.A. LEVINE
FIG. 3.17. Doppler imaging is useful prior to biopsy. The high-intensity signal associated with a thyroidal artery (arrow) is noted in the projected path of the biopsy needle in this image. Repositioning prior to biopsy can help avoid the rare complication of lacerating the artery
FIG. 3.18. Normal lymph node. In a normal lymph node the vascular supply enters centrally at the hilum and spreads along the long axis, within the hilar line
nodules. However, the flow pattern does not have sufficient predictive value to eliminate the need for biopsy of a nodule. Doppler imaging is useful in the evaluation of goiter, thyroid nodules, lymph nodes and parathyroid glands.
DOPPLER ULTRASOUND
41
FIG. 3.19. Lymph node with metastatic papillary carcinoma. In contrast to the preceding figure, this malignant lymph node has disordered and chaotic blood flow
FIG. 3.20. A malignant lymph node in a patient with papillary carcinoma is seen compressing the jugular vein in this Doppler image. Benign lymph nodes may cause deviation of the major vessels but typically do not indent the vessel or cause abnormality in the blood flow
42
R.A. LEVINE
FIG. 3.21. Parathyroid adenomas frequently have a pulsatile polar artery, and Doppler imaging may be useful in establishing that a nodule posterior to the thyroid represents a parathyroid rather than a central compartment lymph node
References 1. Cerbone G, Spiezia S, Colao A, Di Sarno et al (1999) Power Doppler improves the diagnostic accuracy of color Doppler ultrasonography in cold thyroid nodules: follow-up results. Hormone Research 52(1):19–24 2. Papini E, Guglielmi R, Bianchini A, Crescenzi A et al (2002) Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color Doppler features. J Clin Endocrinol Metab 87(5):1941–1946 3. Berni A, Tromba L, Falvo L, Marchesi M et al (2002) Malignant thyroid nodules: comparison between color Doppler diagnosis and histological examination of surgical samples. Chir Ital 54(5):643–647 4. Frates MC, Benson CB, Doubilet PM, Cibs ES, Marqusee E (2003) Can color Doppler sonography aid in the prediction of malignancy of thyroid nodules? J Ultrasound Med 22:127–131 5. Fukunari N, Nagahama M, Sugino K et al (2004) Clinical evaluation of color Doppler imaging for the differential diagnosis of thyroid follicular lesions. World J Surg 28(12):1261–1265 6. De Nicola H, Szejnfeld J, Logullo AF et al (2005) Flow pattern and vascular resistance index as predictors of malignancy risk in thyroid follicular neoplasms. J Ultrasound Med 24:897–904. 7. Levine RA (2006) Value of Doppler ultrasonography in management of patients with follicular thyroid biopsies. Endocr Pract 12(3): 270–274
DOPPLER ULTRASOUND
43
8. Macedo TA, Chammas MC, Jorge PT et al (2007) Differentiation between the two types of amiodarone-associated thyrotoxicosis using duplex and amplitude Doppler sonography. Acta Radiol 48(4):412–421 9. Bogazzi F, Bartelena L, Brogioni S et al (1997) Color flow Doppler sonography rapidly differentiates type I and type II amiodaroneinduced thyrotoxicosis. Thyroid 7(4):541–545 10. Ralls PW, Mayekowa DS, Lee KP et al (1988) Color-flow Doppler sonography in Graves disease: “thyroid inferno.” AJR 150:781– 784
CHAPTER 4
Anatomy and Anomalies H. Jack Baskin
INTRODUCTION Real-time ultrasound was introduced in the 1980’s and rapidly proved to be the most sensitive and efficient method to evaluate thyroid anatomy. The superficial location of the soft tissue structures in the anterior neck also allows the physician to study the anatomy of the entire area and detect anatomic variations as well as identify extra thyroidal masses (1,2). Often patients present with a neck mass thought to be of thyroid origin, but ultrasound quickly and easily identifies it of extrathyroidal origin. This use of ultrasound to define the anatomy of the neck provides an excellent teaching tool in medical education to augment the physical examination. In this chapter we review the normal anatomy along with some common anomalies one might encounter in performing ultrasound. ULTRASOUND TECHNIQUE Patients are usually scanned in the supine position with the neck mildly hyperextended by an “oatmeal” pillow. Both lobes are scanned individually in the transverse and in the longitudinal planes. Any specific abnormalities should be studied in both planes by rotating the transducer 90 degrees over the area of concern. Remember that an ultrasound exam of the thyroid should always include the entire neck, looking for abnormal lymph nodes, enlarged parathyroid glands and abnormal masses. Ultrasound of the post-operative neck and evaluation for parathyroid disease will be discussed in later chapters. NORMAL ANATOMY In order to recognize neck pathology, it is important to be familiar with the anatomy and ultrasound appearance of the normal neck. A normal healthy thyroid lobe is pear-shaped in the transverse view and resembles “ground glass” in 45
46
H.J. BASKIN
appearance on the ultrasound monitor. It is bordered anteriorly by the strap muscles (sternohyoid, sternothyroid and omohyoid). Lateral to the thyroid lie the large sternocleiodomastoid muscle, the carotid artery and the internal jugular vein. The longus colli muscle is posterior and the trachea is medial to the thyroid lobe. The parathyroid glands are posterior to the thyroid and usually not seen unless they are enlarged. The esophagus can also be seen protruding from behind the tracheal shadow posterior to the left lobe. Real-time ultrasound shows the vessels pulsating, and peristalsis can be seen in the esophagus when the patient is asked to swallow. Very rarely, the esophagus will be seen on the right. Measurement of the volume of the thyroid gland is sometimes difficult using ultrasound because most modern small parts transducers have a footpad of only 4 cm or less, and the normal thyroid lobe is over 4 cm long. If the lobe is longer than the transducer, a “split screen technique” can be used to
FIG. 4.1. Normal appearing thyroid in transverse view. Thyroid is homogeneous and slightly hyperechoic. The lobes are bordered anteriorly by the strap muscles (SM), posteriorly by the longus colli muscle (LC), medially by the trachea, and laterally by the sternocleidomastoid muscle (SCM), carotid artery and jugular vein. A portion of the esophagus (ESO) protrudes behind the tracheal shadow against the medial border of the left lobe
ANATOMY AND ANOMALIES
47
measure the length of the lobe. Transducers with a footpad of 6cm or more are used when thyroid volume is important, such as epidemiological studies screening for endemic goiter. The normal thyroid gland is 2 cm or less in both the transverse dimension and depth, and is 4.5–5.5 cm in length. In clinical practice in North America, routine measurement of the volume of the entire gland is not always necessary; however, measuring the volume of thyroid nodules is important and is done in the same manner. Measurement of the thyroid (or a nodule) involves three measurements: the width, depth and length. The volume can then be calculated using the formula for a prolate ellipse: Volume = π/6 (W × D × L). The width (W) of a thyroid lobe is measured from an imaginary vertical line drawn along the lateral edge of the trachea to the most lateral border of the thyroid gland. The depth (D) is measured on the same screen and is the maximum anterior-posterior distance in the middle third of the lobe. The length (L) is measured in the longitudinal view and is the maximum distance from the most cranial to the most caudal part of the lobe. Most ultrasound equipment has an onboard computer to calculate the volume of each lobe (or nodule) from the three measurements. After calculating the volumes of each lobe, they are added together for the total volume of the gland. The isthmus is ignored unless a nodule is present.
FIG. 4.2. Measurement of thyroid lobe. The width (W) is measured from an imaginary line drawn along the lateral edge of the trachea to the lateral border of the thyroid. The depth (D) is measured on the same view and is the maximum anterior-posterior distance in the middle third of the lobe
48
H.J. BASKIN
FIG. 4.3. In the longitudinal view of the thyroid, the length (L) is measured from the cranial to the caudal ends of the lobe
FIG. 4.4. Measure the volume of a thyroid nodule using the same formula used to calculate the volume of a thyroid lobe: Volume = π/6 (W × D × L)
DIFFUSE GOITER Diffuse goiter refers to enlargement or swelling of the thyroid without nodules. The size of a normal thyroid gland correlates with the amount of iodine in the diet. Most areas of the earth have relatively less dietary iodine than North America and,
ANATOMY AND ANOMALIES
49
FIG. 4.5. A diffusely enlarged thyroid gland with an isthmus width of 1.5 cm. Although this could be a hyperplastic normal thyroid, the parenchyma is hypoechoic, suggesting that there is inflammation of the thyroid tissue indicating thyroiditis (Hashimoto’s or Graves’ disease)
therefore, their inhabitants have larger thyroid glands. Normal thyroid glands in the United States are usually 15–20 grams in size, while glands up to 40 grams are common in Europe. A quick check for enlargement can be done by measuring the width of the isthmus; a width >5 millimeters generally indicates thyroid enlargement. A diffusely enlarged gland may be a “simple goiter” due to hyperplasia of normal appearing tissue or due to thyroiditis. Thyroiditis causes inflammation and follicular changes in the parenchyma resulting in the gland losing its characteristic uniform “ground glass” appearance on ultrasound. Various types of thyroiditis and their ultrasound changes are discussed in the next chapter. THYROID ANOMALIES Hemiagenesis of the thyroid is a common anomaly seen with ultrasound. It is usually found when ultrasound is being done for some other cause, such as evaluation of a nodule in the
50
H.J. BASKIN
FIG. 4.6. Hemiagenesis of the left lobe. Ultrasound done to evaluate the palpable nodule (N) in the right lobe reveals the thyroid ends at the isthmus (arrow). The strap muscles (SM) have filled in the space where the left lobe would be. Physical examination of the left neck was normal
FIG. 4.7. Hemiagenesis in patient with Graves’ disease. The isthmus is intact and ends where the left lobe should be (arrow). The isthmus is present in 50% of patients with hemiagenesis. Note the diffuse enlargement and hypoechogenicity of the gland
ANATOMY AND ANOMALIES
51
contralateral lobe. The anomaly cannot be detected by physical examination. In the past it was suspected to be related to Graves’ disease because it was often found while doing an isotope scan for hyperthyroidism. However since the introduction of ultrasound, most cases are discovered by this form of imaging, and there is no correlation with hyperthyroidism. The incidence is 1:2,500 with 95% involving hemiagenesis of the left lobe. While the condition is benign and not known to
FIGS. 4.8 and 4.9. Hemiagenesis of the right lobe. Initially, this appeared to be a multinodular goiter, but the physical examination was normal. Doppler revealed a venous plexus occupying the space where the right lobe was absent
52
H.J. BASKIN
predispose to any type of pathology, knowledge of the condition would be very important to someone undergoing surgery for a thyroid nodule. This is a good reason to always do an ultrasound before sending a patient for neck surgery. Aberrant Thyroid can occur anywhere in the neck. The thyroid gland develops at the base of the tongue and descends to below the larynx where it bifurcates into two lobes connected by an isthmus. Sometimes this descent fails to occur (lingual thyroid) or is interrupted, resulting in a failed bifurcation. This may appear to be a goiter on physical examination, but ultrasound will reveal that the thyroid is above the larynx and undivided but otherwise normal in size. Such “pseudogoiters” function well and do not typically grow, cause problems or require surgery. Conversely, thyroid glands may also descend into the superior mediastinum, as in the case of substernal goiter, and may be difficult to evaluate with ultrasound. However, by using a small transducer and having the patient hyperextend his or her neck and swallow, one can generally visualize a portion of the thyroid. Thyroid tissue can occasionally
FIG. 4.10. Failed bifurcation of the thyroid. This teenage female patient presented with an apparent “goiter” located 1 cm above the larynx. Ultrasound reveals a normal amount of thyroid tissue, and thyroid function was normal
ANATOMY AND ANOMALIES
53
FIG. 4.11. Failed bifurcation of thyroid with a small cyst (calipers)
be found in the lateral neck. Careful evaluation of this “lateral aberrant thyroid” with ultrasound will often show a pedicle or some type of connection to the thyroid. A fine-needle aspiration (FNA) may be necessary to rule out a metastatic lymph node from a nonvisualized microcarcinoma of the thyroid. NONTHYROIDAL ANOMALIES Thyroglossal Duct is formed of embryonic tissue running from the base of the tongue to the larynx, and portions of the duct near the isthmus may persist as a pyramidal lobe of the thyroid, which can be seen with ultrasound. Sometimes the entire thyroglossal duct persists, and protein material secreted by the lining epithelium may form a thyroglossal duct cyst that manifests itself clinically as a midline mass in the anterior aspect of the neck above the isthmus. Brachial cleft cysts will have a similar appearance except they will be more lateral in the neck. Benign Masses in the neck can be evaluated by ultrasound. A large hypoechoic mass posterior to the left lobe of the thyroid should raise the suspicion of an esophageal diverticulum, and it is important to avoid mistaking it for a thyroid nodule
54
H.J. BASKIN
FIGS. 4.12 and 4.13. Lateral aberrant thyroid. This patient had a long history of a small mass (calipers) in the right lateral neck that moved up and down with swallowing. FNA of the mass revealed normal thyroid cells. A small pedicle attached the mass to the right lobe on both transverse and longitudinal views (arrows)
needing an FNA. The longitudinal view will reveal that the “mass” is posterior and separate from the thyroid lobe. An undescended thymus gland having echogenicity of thyroid tissue may be seen inferior to the thyroid but attached by a short thyrothymic ligament. By having the patient swallow, the thymus can be seen to move up and down with the
ANATOMY AND ANOMALIES
55
FIG. 4.14. Thyroglossal duct cyst in the midline. These fluid filled cysts are very superficial. A reverberation artifact in the anterior third of the cyst resembles debris in the fluid (white arrow). Posterior to the cyst is enhancement artifact (black arrow) indicating the sound waves have passed through fluid
FIG. 4.15. This small esophageal diverticulum (arrow) could be mistaken for a posterior nodule in the left lobe
thyroid. Hemangiomas of neck muscles are uncommon, but may present as a hypoechoic heterogeneous mass imbedded in the muscle. Enlarged inflammatory lymph nodes are frequently seen with ultrasound. They can occur in patients
56
H.J. BASKIN
FIG. 4.16. This series of enlarged inflamed lymph nodes (calipers) beneath the sternocleidomastoid muscle could represent infection or signal the presence of autoimmune thyroiditis
FIG. 4.17. Undescended thymus gland attached to the lower pole of the thyroid by a short thyrothymic ligament (arrow)
having sarcoidosis and are a common finding in patients with Hashimoto thyroiditis. Lymph node enlargement is common with infection. These benign lymph nodes are often prominent in the upper neck (Level II) in patients having upper respiratory
ANATOMY AND ANOMALIES
57
FIG. 4.18. Muscle anomaly. Patient was thought to have a nodule in the right lobe by physical examination. Ultrasound revealed enlargement of the strap muscle in the right neck (arrow) causing asymmetry; no nodule is present
infections. Although inflammatory lymph nodes can get very large, they maintain their shape and have a short/long axis ratio less than 0.7. They usually have an echogenic hilum running through the center that shows arteriolar blood flow with power Doppler. Malignant Masses may also be found in the neck by ultrasound. Lymphoma frequently presents as enlarged lymph nodes in the neck. While any malignancy may metastasize to the neck, squamous cell carcinoma from an occult primary is the most common nonthyroid cancer. Ultrasound and FNA have been found to complement each other, and the technique of ultrasound-guided FNA has emerged as a powerful diagnostic tool for evaluating thyroid nodules. The technique is equally beneficial in evaluating other masses in the neck, both benign and malignant. When an abnormal mass is found by ultrasound, ultrasound-guided needle placement is essential in obtaining accurate diagnostic material.
58
H.J. BASKIN
FIGS. 4.19 and 4.20. Hemangioma. This heterogeneous mass imbedded in the left sternocleidomastoid muscle between the thyroid, carotid artery and jugular vein is very vascular on power Doppler, indicating it is a hemangioma. It might be confused with a malignant lymph node except that it is too flat (short/long ratio <0.5)
ANATOMY AND ANOMALIES
59
FIG. 4.21. This complex mass (calipers) appears to be extraneous to the left lobe of the thyroid; however, several ultrasound-guided FNA biopsies showed only benign thyroid cells
FIG. 4.22. This hypoechoic mass seems to be extraneous to the left lobe of the thyroid and compressing the normal thyroid (arrow). Ultrasound-guided FNA revealed the mass to be a squamous cell carcinoma
60
H.J. BASKIN
FIG. 4.23. Two views of an intrathyroidal hypoechoic homogeneous mass (calipers) in the upper pole of the left lobe that proved to be a squamous cell carcinoma on ultrasound-guided FNA
FIG. 4.24. Twenty-two-year-old male with large rounded extremely hypoechoic mass (calipers) in close proximity to the carotid artery and replacing the left lobe of the thyroid. Ultrasound-guided FNA revealed metastatic testicular cancer. Patient was cured by chemotherapy
ANATOMY AND ANOMALIES
61
References 1. Baskin HJ (1997) Thyroid ultrasonography—A review. Endocrine Pract 3:153–157 2. Baskin HJ (2004) New applications of thyroid and parathyroid ultrasound. Minerva endocrinol 29:195–206
CHAPTER 5
Thyroiditis Reagan Schiefer and Diana S. Dean
INTRODUCTION Chronic autoimmune thyroiditis and Graves’ disease are two forms of autoimmune thyroid disease (AITD). It has been said that Hashimoto thyroiditis and Graves’ disease are the same autoimmune thyroid disease but at different ends of the spectrum. Transition between the two autoimmune thyroid diseases may occur, which adds to the difficulty in differentiating between the two (1). The ultrasonographic appearance of both Graves’ disease and Hashimoto thyroiditis are similar as well, with both having a hypoechoic and heterogeneous echotexture. While Graves’ disease typically shows marked hypervascularity with power Doppler analysis, the vascularity of Hashimoto thyroiditis is variable, ranging from avascular to hypervascular. Chronic autoimmune thyroiditis has two clinical forms: a goitrous form referred to as Hashimoto disease and an atrophic form called atrophic thyroiditis, which may represent the end stage of the former. The presence of serum thyroid autoantibodies, varying degrees of thyroid dysfunction and lymphocytic infiltration characterize both forms of chronic autoimmune thyroiditis (2). Both silent (or painless) thyroiditis and post-partum thyroiditis are transient disorders thought to be manifestations of chronic autoimmune thyroiditis (2). The diagnosis of AITD is easy to make when patients are clinically symptomatic. However, AITD may be missed when thyroid autoantibodies are negative (4) or have not been ordered in the diagnostic work-up due to normal thyroid function tests and normal palpation, making the diagnosis of AITD appear less likely (3). Ultrasonography, which has proven valuable in clarifying number and size of thyroid nodules (5–8), can also be of value in the diagnosis of AITD due to radiographic findings often associated with AITD (9–17).
63
64
R. SCHIEFER AND D.S. DEAN
PATHOLOGY The diagnosis of chronic autoimmune thyroiditis by cytology is not based just on the presence of chronic lymphocytic infiltration. When lymphocytic infiltration alone is the only histologic finding, chronic autoimmune thyroiditis can be diagnosed with confidence only if the patient has high serum titers of antithyroid autoantibodies. Pathology-proven chronic autoimmune thyroiditis requires diffuse lymphocytic infiltration with occasional germinal centers; small, colloid-deplete follicles; fibrosis; and Hürthle cells (2). Hurthle cells are enlarged-appearing thyroid cells containing oxyphilic cytoplasm (granular and pink) (26). ULTRASONOGRAPHY Since its introduction in clinical practice, ultrasonography has proven to be a useful tool in the management of patients with thyroid disease (3, 12, 16, 17). Besides identification of thyroid nodules (19), ultrasonography is able to characterize the echostructure of thyroid tissue in patients with AITD (4, 20). In AITD, lymphocytic infiltration and disruption of tissue architecture cause a reduction in thyroid echogenicity (4). Hashimoto thyroiditis and Graves’ disease appear similar on gray-scale ultrasound, but power Doppler will demonstrate increased blood flow in Graves’ disease (“thyroid inferno”), with Hashimoto thyroiditis showing the full spectrum from absent to normal to increased blood flow. In granulomatous or deQuervain’s thyroiditis the hypoechogenicity may be localized to one lobe or even a portion of a lobe that is involved. The ultrasound appearance returns to normal when the subacute thyroiditis resolves. Several studies have indicated that reduction in thyroid echogenicity occurs at a relatively early stage in the AITD process, often before overt thyroid failure (3). Reduced thyroid echogenicity detected by thyroid ultrasonography is a strong predictor of AITD even when these disorders have not been suspected clinically (3). Subclinical thyroid dysfunction has gained importance due to greater knowledge of its negative effect on cholesterol, bone mineral density, heart rhythm and depression (18). Hypoechogenicity is a major characteristic finding on ultrasound of patients having Hashimoto thyroiditis. In most of the studies, modifications in thyroid echogenicity were described subjectively as compared with the hyporeflective surrounding neck muscles (3, 12, 16, 21, 22, 23). Recently, the gray-scale
THYROIDITIS
65
histogram analysis has been proposed for a quantitative measurement of decrease in echogenicity of thyroid glands affected by autoimmune disease with respect to the normal glands (24, 25). However, other ultrasonographic findings in addition to hypoechogenicity occur in Hashimoto’s thyroiditis.
FIG. 5.1. Extremely hypoechoic thyroid gland caused by lymphocytic infiltration of thyroid tissue. The echogenicity is similar to the surrounding musculature
FIG. 5.2. fibrosis
This hypoechoic right lobe shows small dense areas of early
66
R. SCHIEFER AND D.S. DEAN
These findings illustrate heterogeneity in the ultrasound appearance of the thyroid caused by destruction of the normal homogeneous “ground glass” architectural pattern of thyroid tissue, resulting in the formation of pseudonodules that may
FIG. 5.3. “Bag of marbles.” The areas of fibrosis appear hyperechoic in comparison to the hypoechogenicity of the rest of the gland, and could be mistaken for hyperechoic nodules (pseudonodules)
FIG. 5.4. This enlarged thyroid is typical of Hashimoto thyroiditis with a hypoechoic but heterogeneous pattern
THYROIDITIS
67
be numerous and resemble a “bag of marbles.” These pseudonodules are not well outlined or defined. They may also come and go, meaning that they are there today but may disappear if ultrasonography is done even a week later. Pseudonodules are
FIG. 5.5. This enlarged right lobe has lost the typical ground glass appearance of normal thyroid tissue. Infiltration by lymphocytes and other inflammatory cells causes the tissue to be less dense and appear more hypoechoic
FIG. 5.6. Left lobe of patient with Hashimoto thyroiditis. Fibrosis has advanced in a sheet-like pattern with layers of connective tissue running through the hypoechoic thyroid parenchyma
68
R. SCHIEFER AND D.S. DEAN
thought to be “diffuse lakes of lymphocytes.” Another frequent finding is the presence of tiny cystic lesions, which may be described as a “swiss cheese” appearance of the gland. These diffuse cystic lesions are usually 2–3 mm in size. Yet another
FIG. 5.7. “Swiss cheese.” Diffuse small cystic lesions scattered throughout normal appearing thyroid represent an early stage of Hashimoto thyroiditis
FIG. 5.8. Pseudonodules and fibrosis lead to disruption of the architectural pattern of this enlarged thyroid, which causes the gland to appear very heterogeneous
THYROIDITIS
69
characteristic feature of Hashimoto thyroiditis is echogenic strands, or septa. These strands have been described as thin echogenic septa traversing the thyroid tissue, sometimes giving the thyroid a lobulated appearance (3). They are thought to be due to fibrosis within the gland. Fibrosis can also develop in pseudonodules, changing them from hypoechoic to hyperechoic.
FIG. 5.9. Left lobe of a patient with chronic thyroiditis. Over a period of time areas of amorphous calcification and shadowing have developed
FIG. 5.10. This patient with longstanding thyroiditis shows a small atrophic hypoechoic gland that blends in with the surrounding muscles
70
R. SCHIEFER AND D.S. DEAN
FIG. 5.11. This gland shows several pseudonodules (arrows) that might initially be mistaken as nodules. However these pseudonodules have no halo or sharp border, are not palpable and will change in appearance over time
FIG. 5.12. This patient had a painful upper right lobe with elevated sedimentation rate typical of deQuervain’s thyroiditis. Note the line of demarcation (arrow) between the inflamed upper lobe and normal appearing lower lobe. After six weeks the entire gland appeared normal
Often the first indication of thyroiditis one finds on ultrasound is the presence of enlarged lymph nodes. These may be unilateral or bilateral; they may be in the central compartment or the lateral compartments of the neck, and they may be single or multiple. Except for their size they typically resemble
THYROIDITIS
71
normal lymph nodes with a short/long ratio < 0.7. While they generally have a hilum with hilar blood flow on power Doppler, it is not unusual to see confluent, enlarged, atypical appearing lymph nodes in active Hashimoto thyroiditis. Nodes exhibiting highly suspicious features such as chaotic vascularity may warrant UGFNA.
FIG. 5.13. These enlarged flattened lymph nodes under the sternocleidomastoid muscle are commonly seen in early Hashimoto thyroiditis and are often a clue to early diagnosis. See also Fig. 4.16
FIG. 5.14. This enlarged flattened paratrachael lymph node (calipers) in the central compartment is a common finding in Hashimoto thyroiditis
72
R. SCHIEFER AND D.S. DEAN
FIG. 5.15. This patient with Hashimoto thyroiditis also had a true nodule (calipers) in the right lobe that required FNA
FIG. 5.16. This palpable nodule in the isthmus of a patient with Hashimoto thyroiditis proved to be a papillary carcinoma by FNA and surgical excision
Care must be taken in performing ultrasound on a patient with Hashimoto thyroiditis, since there may be coincidental occurrence of malignancy. Although the incidence of papillary carcinoma does not appear to be increased in AITD, it does occur and may initially appear to be a pseudonodule.
THYROIDITIS
73
FIG. 5.17. Lymphoma of the thyroid occurs almost exclusively in patients having Hashimoto thyroiditis. The tumor is extremely hypoechoic with only a thin area of thyroid tissue seen (arrow). Diagnosis was made by FNA with flow cytometry, and an operation was avoided. When lymphoma is localized to the thyroid, it usually responds well to chemotherapy and/or radiation
If a nodule is palpable or there is doubt about it being a true nodule, ultrasound-guided FNA is recommended. Lymphoma is known to be predisposed to Hashimoto thyroiditis. Virtually all patients with a lymphoma of the thyroid had preexisting AITD. The diagnosis can be confirmed by ultrasound-guided FNA with flow cytometry, and surgery can be avoided. SUMMARY Diffuse hypoechogenicity and heterogeneity are the two hallmarks of thyroid autoimmunity; however, the ultrasound appearance varies dramatically among patients depending upon the severity and duration of the disease. Negative immunologic tests do not rule out the diagnosis of AITD, as illustrated by patients with histologically proven Hashimoto thyroiditis, who have a hypoechoic, heterogeneous echotexture on thyroid ultrasound despite negative immunologic tests. Reduced echogenicity, lymph node enlargement, and the other ultrasonographic findings mentioned in this chapter can occur prior to overt thyroid dysfunction (3). These characteristic findings can be used to identify the risk of overt thyroid dysfunction many years before its development.
74
R. SCHIEFER AND D.S. DEAN
References 1. Utiger RD (1991) The pathogenesis of autoimmune thyroid disease. N Engl J Med 325:278–279 2. Dayan DM, Daneils GH (1996) Chronic autoimmune thyroiditis. N Engl J Med 335:99–107 3. Pedersen et al (2000) The value of ultrasonography in predicting autoimmune thyroid disease. Thyroid 10:251–259 4. Gutekunst R, Hafermann W, Mansky T, Scriba PC (1989) Ultrasonography related to clinical and laboratory findings in lymphocytic thyroiditis. Acta Endocrinol (Copenh) 121:129–135 5. Tan GH, Gharib H (1997) Thyroid incidentalomas: management approaches to nonpalpable nodules discovered incidentally on thyroid imaging. Ann Intern Med 126:226–231 6. Schneider AB, Bekerman C, Leland J, Rosengarten J, Hyun H, Collins B, Shore-Freedman E, Gierlowski TC (1997) Thyroid nodules in the follow-up of irradiated individuals: comparison of thyroid ultrasound with scanning and palpation. J. Clin Endocrinol Metab 82:4020–4027 7. Danses D, Sciacchitano S, Farsetti A, Andreoloi M, Pantecorvi A (1998) Diagnostic accuracy of conventional versus sonographyguided fine-needle aspiration biopsy of thyroid nodules. Thyroid 8:15–21 8. Leenhardt L, Hejblum G, Franc B, Fediaevski LGP, Delbout T, Guillouzic DL, Menegaux F, Guillausseau C, Hoang C, Turpin G, Aurengo A (1999) Indications and limits of ultrasound-guided cytology in the management of nonpalpable thyroid nodules. J Clin Endocrinol Metab 84:24–28 9. Yoshida A, Adachi T, Noguchi T, Urabe K, Onoyama S, Okamura Y, Shigemasa C, Abe K, Mashiba H (1985) Echographic findings and histological feature of the thyroid: a reverse relationship between the level of echo-amplitude and lymphocytic infiltration. Endocrinol Jpn 32:681–690 10. Hayashi N, Tamaki N, Konishi J, Yonekura Y, Senda M, Kasagi K, Yamamoto K, Lida Y, Misaki T, Endo K, Torizuka K, Mori T (1986) Sonography of Hashimoto’s thyroiditis. J Clin Ultrasound 14:123–126 11. Nordmeyer J.P, Shafeh TA, Heckmann C (1990) Thyroid sonography in autoimmune thyroiditis: a prospective study on 123 patients. Acta Endocrinol (Copenh) 122:391–395 12. Marcocci C, Vitti P, Cetani F, Catalano F, Concetti R, Pinchera A (1991) Thyroid ultrasonography helps to identify patients with diffuse lymphocytic thyroiditis who are prone to develop hypothyroidism. J Clin Endocrinol Metab 72:209–213 13. Sostre S, Reyes MM (1991) Sonographic diagnosis and grading of Hashimoto’s thyroiditis. J Endocrinol Invest 14:115–121 14. Adams H, Jones MC, Othman S, Lazarus JH, Parkes AB, Hall R, Phillips DIW, Richards CJ (1992) The sonographic appearances in postpartum thyroiditis. Clin Radiol 45:311–315
THYROIDITIS
75
15. Miyakawa M, Tsushima T, Onoda N, Etoh M, Isozaki O, Arai M, Shizume K, Demura H (1992) Thyroid ultrasonography related to clinical and laboratory findings in patients with silent thyroiditis. J Endocrinol Invest 15:289–295 16. Vitt P, Rago T, Mancusi F, Pallini S, Tonacchera M, Santini F, Chiovato L, Marcocci C, Pinchera A (1992) Thyroid hypoechogenic pattern at ultrasonography as a tool for predicting recurrence of hyperthyroidism after medical treatment in patients with Graves’ disease. Acta Endocrinol 126:128–131 17. Vitt P, Lampis M, Piga M, Loviselli A, Brogioni S, Rago T, Pincherra A, Martino E (1994) Diagnostic usefulness of thyroid ultrasonography in atrophic thyroiditis. J Clin Ultrasound 22:375–379 18. Woeber KA (1997) Subclinical thyroid dysfunction. Arch intern med 157:1065–1068 19. Hegedus L, Karstrup S (1998) Ultrasonography in the evaluation of cold thyroid nodules. Euro J Endocrin 128:30–31 20. Muller HW, Schroder S, Schneider C, Seifert G (1985) Sonographic tissues characterization in thyroid gland diagnosis. Klinische Wochenschrif 63:706–710 21. Premawardhana LD, Parkes AB, Ammari F, John R, Darke C, Adams H, Lazarus JH (2000) Post partum thyroiditis and longterm thyroid status: prognostic influence of thyroid peroxidase antibodies and ultrasound echogenicity. J Clin Endocrinol Metab 85:71–75 22. Rago T, Chiovato L, Grasso L, Pinchera A, Vitti P (2001) Thyroid ultrasonography as a tool for detecting thyroid autoimmune diseases and predicting thyroid dysfunction in apparently healthy subjects. J Endocrinol Invest 24:763–769 23. Raber W, Gessl A, Nowotny P, Vierhapper H (2002) Thyroid ultrasound versus antithyroid peroxidase antibody determination: a cohort study of 451 subjects. Thyroid 12:725–731 24. Schiemann U, Geellner R, Riemann B, Schierbaum G, Menzel J, Domschke W, Hengst K (1999) Standardized grey scale ultrasonography in Graves’ disease: correlation of autoimmune activity. Euro J Endocrin 141:332–336 25. Vitti P (2000) Grey scale thyroid ultrasonography in the evaluation of patients with Graves’ disease. Euro J Endocrin 142:22–24 26. Livolsi VA (1994) The pathology of autoimmune thyroid disease: a review. Thyroid 4:333–339
CHAPTER 6
Ultrasound of Thyroid Nodules Susan J. Mandel, Jill E. Langer, and Daniel S. Duick
INTRODUCTION Thyroid ultrasound is an exquisitely sensitive technique for the detection of thyroid nodules and is able to image nodules as small as 2–3 mm. The prevalence of sonographically detected nodules that cannot be palpated is up to 50%–60% in individuals older than 60 (1). Therefore, the challenges confronting the clinician are the identification of those nodules that have a higher probability of being clinically relevant malignancies so that these can be targeted for fineneedle aspiration biopsy, and the recognition of those that may undergo sonographic surveillance. Furthermore, diffuse thyroid disorders, such as Graves’ disease and Hashimoto’s thyroiditis, image differently than the normal thyroid parenchyma, but it may be challenging to differentiate asymmetric involvement of the thyroid by one of these diffuse processes from a discrete thyroid nodule. DIAGNOSTIC THYROID ULTRASOUND Palpable thyroid nodules: After excluding those with low serum TSH levels, recently published evidence-based guidelines from the American Thyroid Association and the American Association of Clinical Endocrinologists recommend a diagnostic thyroid ultrasound for patients with palpable thyroid nodules (2, 3). The rationale for this includes: a) Confirmation of a sonographically identifiable nodule corresponding to the palpable abnormality. A thyroid nodule is discrete lesion in the thyroid that is distinct from the surrounding thyroid parenchyma. For one out of every six patients with a palpable thyroid nodule, ultrasound fails to demonstrate a corresponding nodule and, therefore, fineneedle aspiration (FNA) is unnecessary (4, 5). 77
78
S.J. MANDEL ET AL.
b) Detection of additional nonpalpable nodules for which FNA may be indicated. Ultrasound identifies additional nonpalpable thyroid nodules in almost 50% of patients with an index-palpable and sonographically confirmed thyroid nodule (4–6). However, only about 20% of these nodules are supracentimeter in size. c) Determination of accuracy of FNA by palpation. For nodules that are found to be more than 50% cystic (7) or are situated in the posterior aspect of the thyroid (8), palpation FNA is less accurate because of the potential for either nondiagnostic cytology or sampling error. Therefore, FNA with ultrasound guidance is the preferred technique. d) Identification of the sonographic characteristics of the thyroid nodule(s). Different sonographic features are associated with a higher likelihood of thyroid malignancy (Table 1). Therefore, if a nodule is considered borderline in size for FNA or if several thyroid nodules are present, the sonographic appearance of a nodule may aid in making decisions about performance of FNA (9, 10). Normal thyroid on physical examination. Thyroid ultrasound should neither be routinely performed in patients whose thyroid is normal by palpation nor be exploited as a substitute for physical examination. However, some exceptions exist. Diagnostic ultrasound should be considered in two groups of patients with a higher prevalence of thyroid cancer:
TABLE 6.1. Reported sensitivities and of sonographic features for detection of thyroid cancer
Microcalcifications Absence of Halo Irregular Margins Hypoechoic Increased Intranodular Flow (9, 10, 26–33)
Median Sensitivity [Range]
Median Specificity [Range]
50% [26–73%] 66% [46–91%] 55% [17–77%] 80% [49–90%] 67% [57–74%]
85% [69–96%] 54% [30–72%] 76% [63–85%] 53% [36–66%] 81% [49–89%]
ULTRASOUND OF THYROID NODULES
79
a) Prior history of head and neck irradiation. Patients who received external beam radiation as children for benign conditions more frequently develop both benign and malignant thyroid nodules, and ultrasound detects these tumors and facilitates intervention (11). In addition, because of the increased incidence of thyroid cancer in children younger than age 16 who were exposed to radiation after the Chernobyl accident, ultrasound has been instrumental for screening and identification of those affected (12). b) Family history of thyroid cancer, including papillary thyroid cancer. Recently emergent evidence reports that papillary cancer may be familial in up to 10% of cases (13, 14). When family members from 53 families with nonmedullary thyroid cancer were examined by thyroid ultrasound, thyroid cancer (median size 10 mm, range 3–21mm) was discovered in 10%, of which half were multifocal (15). For familial medullary thyroid cancer (either isolated or as part of the multiple endocrine neoplasia II syndromes), thyroid ultrasound should be performed only in patients with mutations in the RET proto-oncogene to evaluate the thyroid and the cervical lymph nodes. THYROID ULTRASOUND OF DIFFUSE THYROID DISORDERS Diffuse thyroid disorders, most commonly chronic autoimmune thyroiditis (Hashimoto’s thyroiditis) and Graves’ disease, may alter the sonographic appearance of normal thyroid tissue. Histologically, a thyroid affected by Hashimoto’s thyroiditis has lymphocytic infiltration, variable degrees of fibrosis and, at least in the early stages, preservation of some normal follicular areas. This correlates with a spectrum of sonographic images ranging from diffuse hypoechogenicity to a hypoechoic “micronodular” appearance (Fig. 6.1b) (16) to isolated interspersed patches of hypoechogenicity. The majority of thyroid glands from hypothyroid patients with Hashimoto’s thyroiditis appear diffusely heterogeneous, with admixed hypo- and hyperechoic areas (17, 18) (Fig. 6.1a). In addition, some patients may present with a focal, often palpable abnormality simulating a thyroid nodule. The focal lesion may be hyper- rather than hypoechoic, but FNA cytology reveals only lymphocytic thyroiditis (19) (Fig. 6.1b). Lastly, vascularity is variable, but it may appear increased as determined by color-flow Doppler (CFD) ultrasonography. Even the thyroid glands from euthyroid patients who have measurable serum
80
S.J. MANDEL ET AL.
FIG. 6.1. a Hashimoto’s lymphocytic thyroiditis. The echo pattern is heterogeneous, with interspersed discrete white lines representing fibrosis b Longitudinal view demonstrating the micronodular appearance of Hashimoto’s thyroiditis superiorly and a hyperechoic nodule that was aspirated and was consistent with lymphocytic thyroiditis
FIG. 6.2. a Gray-scale image of Graves’ disease with heterogeneous echogenicity. b Color-flow Doppler image of Graves’ disease demonstrating increased vascularity
antithyroid peroxidase antibodies, the serologic evidence of Hashimoto’s thyroiditis, will demonstrate some degree of heterogeneity on ultrasonography (20). The thyroid gland is enlarged in most patients with Graves’ disease, but the echogenicity is usually decreased. The hypoechogenicity reflects three factors: a variable amount of lymphocytic infiltration (similar to that seen in Hashimoto’s thyroiditis), increased intrathyroidal blood flow, and decreased intraglandular colloid from rapid thyroid hormone turnover, which reduces the cell–colloid interface (21). Furthermore, the degree of hyperthyroidism correlates with increased vascularity determined by CFD (22) (Fig. 6.2).
ULTRASOUND OF THYROID NODULES
81
Although the clinical diagnosis of subacute thyroiditis is generally apparent, there are well-described sonographic changes in the thyroid gland. Hypoechogenicity may be localized or generalized and corresponds to the areas of affected tissue, which are usually painful. However, vascularity is not increased in the affected swollen thyroid in comparison with a Graves’ goiter. In the recovery stage, sonography demonstrates return to isoechogenicity with slightly increased vascularization (23). Given the sonographic abnormalities associated with each of the above conditions, it may be difficult to discern if a discrete nodule is superimposed upon the abnormal thyroid parenchyma. In one investigation of patients with Hashimoto’s disease and coincident thyroid nodules, papillary carcinoma imaged as hypoechoic even compared to the background thyroid (24). Importantly, the incidence of thyroid cancer arising in a thyroid with chronic lymphocytic thyroiditis is not increased. In patients with Graves’ disease, several studies have reported a higher incidence of thyroid cancer, but almost all are retrospective and subject to selection bias. In one of the few systematic prospective thyroid ultrasound surveys in Graves’ disease, neither the frequency of thyroid nodules nor that of thyroid cancer differed from the general population (25). However, the sonographic appearance of these malignancies is not reported. ULTRASOUND CHARACTERISTICS OF THYROID NODULES Ultrasound not only detects the presence, location, and size of nodules within the thyroid gland, but it identifies imaging characteristics of these nodules. Over the last decade, multiple reports have evaluated sonographic features of thyroid nodules as predictors of malignancy. However, these studies neither use consistent methodologies nor uniformly address all characteristics. Some inconsistencies may be related to technical improvements in thyroid sonography, with earlier reports using a 7MHz probe while more recent studies use a 12–14MHz probe, but differences in classification criteria account for the largest proportion of variability among the studies. For example, benign nodules may be identified by cytology or by histology; consequently, the proportion of thyroid cancers in these series varies from 4% to 32% (9, 10, 26–33). Furthermore, classification of sonographic features differs among the series. Some only consider echogenicity for solid nodules; others include cystic nodules, and echogenicity is determined by the
82
S.J. MANDEL ET AL.
solid portion. Most studies dichotomize a halo as absent or present, but some separate those with a partial halo from those with a complete halo (33). Some series group all calcifications together (9, 26); others divide them into subtypes (32, 33). Finally, identification of these qualitative ultrasound features is highly operator dependent, especially for characterization of nodule margins (34). A description of the sonographic characteristics of thyroid nodules follows, with a focus on the features that are associated with thyroid cancer. Table 1 lists these sonographic characteristics and summarizes their published median sensitivities and specificities for detection of malignancy summarized from 10 published reports (9, 10, 26–33). For inclusion in this analysis, each study must (1) contain at least 100 nodules, (2) analyze at least three sonographic characteristics and (3) report both sensitivity and specificity for thyroid cancer. Echogenicity. The echogenicity of a thyroid nodule refers to its brightness compared to the normal thyroid parenchyma. The normal thyroid images as homogeneously hyperechoic and bright compared to the surrounding strap muscles of the neck. A nodule is generally characterized as hypo-, iso-, or hyperechoic. Hypoechogenicity is associated with thyroid malignancy and is thought to represent a microfollicular structure on histology, whereas macrofollicular lesions may image as iso- or hyperechoic (29) (Fig. 6.3). Cystic fluid is classified as anechoic, with through transmission of sound waves and posterior acoustic enhancement. True cysts over 1.5–2 cm are rare, <2% of all lesions, but if
FIG. 6.3. a Hypoechoic solid nodule 3. b Iso- to hyperechoic solid nodule
ULTRASOUND OF THYROID NODULES
83
FIG. 6.4. a Pure cystic lesion without any internal vascular flow. b Cyst with comet tail artifact
present, these are reported to always be benign (33). However, usually a solid component is present. However, often multiple small <1 cm cystic nodules may be imaged that are either simple cysts or may contain internal bright echogenic foci. The bright foci are often associated with “comet tail” or reverberation artifacts within the cystic nodule (Fig. 6.4). With or without comet tail artifact, these small cysts are thought to represent nonneoplastic benign nodular hyperplasia with its associated colloid-filled cysts. In fact, the comet tail artifact is hypothesized to result from sound wave interaction with the condensed colloid protein (35). For complex nodules that are predominantly cystic but have an associated solid component, the fluid represents degeneration and possible hemorrhage. There may be associated areas of internal debris that appear the same on gray-scale imaging as viable tissue, but with color-flow Doppler, the debris is avascular. Aspiration of bloody fluid does not reliably differentiate benign from malignant lesions (36). Although clear yellow fluid is more likely to be associated with a benign lesion, this can rarely be associated with thyroid cancer. From a recent Mayo clinic review of ultrasound findings in 360 consecutive operated thyroid cancers, only 3% were more than 50% cystic, and of these, all except one had another suspicious ultrasound feature that included either microcalcifications, intranodular vascularity, a mural nodule or a thick, irregular wall surrounding the cystic area (37) (Fig. 6.5). Therefore, for complex nodules with discrete solid areas, sonographic features should be classified based upon these solid areas.
84
S.J. MANDEL ET AL.
FIG. 6.5. a Gray-scale image of cystic papillary cancer. Note microcalcification in solid area. b Color-flow Doppler image of same nodule demonstrating increased vascularity in the papilliform solid area
Nodule echogenicity may be challenging to determine in two situations. First, the extranodular thyroid may be affected by Hashimoto’s thyroiditis and is, therefore, itself more heterogeneous in appearance, making classification of the nodule’s echogenicity more difficult. Second, one-third of nodules are more than 25% cystic, and an additional quarter of nodules are up to 25% cystic (38). Therefore, 55% of nodules have some cystic composition, and the classification of echogenicity is made by examination of the solid component. This can be more straightforward if the cystic area is distinct from the solid area. However, some nodules have mixed echogenicity throughout with minute <5 mm cystic areas separated by thin septations that are interspersed within solid tissue, a so-called “spongiform” or “honeycomb” pattern (39, 40). This appearance is often found in benign hyperplastic nodules (Fig. 6.6a). These spongiform nodules may also have echogenic foci that are associated with the septations or the back wall of the small internal cystic spaces. These bright foci should not be confused with microcalcifications. In the suspicious hypoechoic solid nodules, the bright spots, representing calcifications, are located within the hypoechoic solid stroma itself (Fig. 6.7). Furthermore, a spongiform nodule may relay the overall impression of iso- to hyperechogenicity, but this must be distinguished from a complex nodule with discrete cystic areas that is otherwise iso- or hyperechoic. This second appearance may represent a true neoplastic growth, either follicular or Hürthle cell, and would require histology to discriminate
ULTRASOUND OF THYROID NODULES
85
FIG. 6.6. a “Spongiform” nodule. Solid nodule with small interspersed cystic areas. FNA cytology is benign. b Predominantly solid nodule with discrete cystic area. FNA cytology is indeterminate, histology is benign follicular adenoma
FIG. 6.7. a Echogenic foci (some with comet tail artifact) that are associated with the septations or the back wall of the small internal cystic spaces in a spongiform nodule. FNA cytology is benign. b Microcalcifications in the solid stroma of a hypoechoic solid nodule with irregular microlobulated margins. FNA cytology is papillary thyroid cancer
between benign and malignant (Fig. 6.6b). A recent ultrasonographic–FNA series of 800 nodules reported that only pure cystic nodules had no risk of cancer, but that complex noncalcified nodules harbored a 3% risk of malignancy (33). The distinction between complex nodules that were spongiform in appearance and those with isolated discrete cystic areas was not made however.
86
S.J. MANDEL ET AL.
Calcifications. Calcifications may be present in up to 30% of nodules and can be divided into different categories. Microcalcifications appear as small (<1 mm) echogenic foci without acoustic shadowing and are more specific (in some studies, up to 96%) than sensitive for thyroid cancer (Fig. 6.7b). Furthermore, the interobserver variability for the identification of microcalcifications is quite good (34). It is hypothesized that these microcalcifications are the imaging equivalent of aggregates of psammoma bodies, the laminated spherical concretions characteristic of most papillary cancers, but occasionally found in benign nodules and Hashimoto’s thyroiditis (41). Coarse or dense calcifications are larger than 2 mm and cause posterior acoustic shadowing. Occurring within either benign or malignant nodules, these dystrophic calcifications are present in areas of fibrosis and tissue degeneration and necrosis. However, coarse calcifications, either associated with microcalcifications or appearing in the center of a hypoechoic nodule, may be worrisome for malignancy (33, 39) (Fig. 6.8a). The third type of calcification is the peripheral or “egg-shell” calcification, once thought to indicate a benign nodule (Fig. 6.8b). However, this can be found in malignant nodules (41), and a particular worrisome finding is interruption of this rim calcification, indicating probable invasion by the cancer (Fig. 6.8c). Margins. The margins of a thyroid nodule may appear either regular and well defined or blurred and irregular, sometimes with a microlobulated appearance (Figs. 6.7b, 6.8a). Malignant nodules that are invasive into surrounding thyroid parenchyma have irregular margins (median sensitivity 55%), but since most benign nodules have regular margins, the median specificity is higher: 76%. Interobserver variability for identification of this feature is the poorest (34), which may account for its reported lack of association with malignancy in several studies. Halo. A halo is a sonolucent ring that surrounds a nodule and is thought to represent compressed perinodular blood vessels. Since benign hyperplastic nodules grow slowly and are generally not neoplastic, they displace the surrounding blood vessels as they expand. Lacking a true capsule, hyperplastic nodular tissue may even merge into the surrounding thyroid parenchyma in some areas. The thin halo, which demonstrates the nodule’s peripheral vascularity on color-flow Doppler, is found in about half of benign nodules but is less common in thyroid cancer that is invasive (absent halo median 66%
ULTRASOUND OF THYROID NODULES
87
FIG. 6.8. a Hypoechoic solid nodule with irregular margins and microand macrocalcifications. The macrocalcification demonstrates posterior acoustic shadowing. This was a papillary thyroid cancer. b Eggshell calcifications. c Interrupted rim calcification in a follicular variant of papillary thyroid cancer. Note the interruption of the anterior calcified border that corresponded with localized invasion into the surrounding thyroid
sensitivity for detection of thyroid cancer) (Figs. 6.9a and 6.9b). However, with high-resolution ultrasound, a second type of halo is now described, a thick irregular avascular halo (30), which may signify the fibrous capsule surrounding a neoplastic growth, either follicular or Hürthle cell, and is therefore, more concerning (Fig. 6.9c). Vascularity. The vascularity of a thyroid nodule is evident with color-flow Doppler (CFD) imaging. Based upon the mean Doppler shift, CFD is a measure of the directional component of the velocity of blood moving through the sample volume. The technical shortcomings of CFD include the interference by noise and angle dependence. More recently, power Doppler (PD) analysis has been applied to nodule assessment. By amplification of the Doppler shift signal, it can record information about the Doppler signal amplitude. PD is more sensitive for
88
S.J. MANDEL ET AL.
FIG. 6.9. a Gray-scale image of isoechoic nodule with thin regular halo. Cytology is benign. b Color-flow Doppler image of same nodule indicating the halo corresponds with peripheral vascularity. c Thick, irregular and incomplete halo surrounding solid iso- to hyperechoic nodule. Histology is Hürthle cell cancer
detection of flow in small vessels that would not be detected by CFD. Furthermore, PD imaging is relatively independent of the angle of the probe and the sound beam and noise can be assigned to a homogeneous background rather than appearing as random color on CFD (30). Using CFD, nodule vascularity is categorized as absent (type I), perinodular (type II) (Figs. 6.9b, 6.10a) or peri- and intranodular (type III) (Fig. 6.10b). Intranodular vascularity is associated with malignancy (median 67% sensitivity). However, CFD cannot discriminate among types of intranodular blood flow. With PD, two types of intralesional vascularization can be distinguished: moderate blood flow with homogeneous structure and regular caliber of blood vessels (type B1), from high blood flow with anarchical structure with winding caliber and flow of vessels (type B2). Furthermore, a third pattern of blood flow is that of a large peripheral afferent blood vessel entering the nodule with intralesional vascularization (type C). A recent study reported that all 23 cancers had either type B2 or C vascularity on PD (sensitivity 100%), but the specificity was only 40% (30).
ULTRASOUND OF THYROID NODULES
89
FIG. 6.10. a Peripheral vascularity in spongiform nodule. Cytology is benign. b Increased intranodular vascularity in hypoechoic nodule
Taller than wide. The association of a nodule’s anterioposteriorto-transverse diameter (A/T) ratio ≥1 and malignancy has been evaluated by two series. Why should the shape of a nodule matter? The ratio of the surface area to the volume is maximized by spherical shape, which would optimize exposure of the tumor cells to nutrient delivery (42). By considering all three dimensions of a nodule (anterioposterior, transverse and longitudinal) to calculate shape as the ratio of the longest to shortest diameter, one study found that the least spherical nodules, defined as a long-to-short axis ratio of >2.5, were all benign (42). The current report by Cappelli et al (26) only calculates the ratio of two of the three dimensions, so this may not truly reflect the nodule shape. Nevertheless, they found that an A/T ratio ≥1 has both high sensitivity (84%) and specificity (82%) for detection of thyroid cancer, significantly higher than the sensitivity of 33% reported by the only other prior investigation of this feature (31). Therefore, the A/T ratio requires additional validation before widespread use to select nodules for FNA. Elastography. Ultrasonographic elastography is a recently developed dynamic technique that provides an estimate of tissue stiffness by measuring its degree of distortion under the application of an external force. This technology is based upon the principle that the softer areas of tissue deform more easily than the harder parts under compression, thus allowing an objective determination of tissue consistency. Because of the observation that malignant lesions are often associated with changes in tissue mechanical properties that render it stiffer,
90
S.J. MANDEL ET AL.
FIG. 6.11. Hypoechoic solid nodule with irregular borders. Note anterior border of lesion penetrates through anterior thyroid capsule into surrounding tissue
elastography may have potential as an adjunctive tool for the diagnosis of thyroid cancer, especially for indeterminate cytology nodules. A recent study from Italy examined the performance of elastography in such nodules and found high elasticity scores in 6 of 7 patients with malignant histology, and low scores in all 25 patients with benign lesions (43). Other features. Invasion of the surrounding structures can occasionally be seen with invasive thyroid cancers. The thyroid capsule is interrupted at the level of the tumor, and the tumor tissue is seen to penentrate into surrounding strap muscles (Fig. 6.11). Only rarely is intratracheal growth seen. Thyroid cancer may involve cervical lymph nodes in up to 35% of cases. Therefore, abnormal lymph nodes in the ipsilateral cervical chain may be apparent at the time of nodule diagnosis. Evaluation of cervical lymph nodes should be done pre-operatively in any patient with a nodule that has malignant or suspicious-for-malignant cytology (2). Because of the presence of the thyroid, it is more difficult to evaluate the central neck or paratracheal lymph nodes. Recognizing the limitations of individual sonographic features for prediction of thyroid cancer, some series have explored the association of combinations of these features with cancer risk. In general, as the specificity increases, the sensitivity decreases. For example, although very few (<4%)
ULTRASOUND OF THYROID NODULES
91
benign nodules will be hypoechoic with microcalcifications, only a minority of thyroid cancers (26%–31%) will have this appearance (10, 27, 32). Therefore, if only such nodules were to be aspirated, over 70% of cancers would be missed. The combination of sonographic features that maximizes sensitivity and specificity is a solid hypoechoic nodule, which identifies ~65% of all cancers but still describes the appearance of 30% of benign nodules (9, 28, 32) and excludes any attempt at classification of complex nodules. In addition, some of the variability associated with the reported sensitivities of individual sonographic features for prediction of thyroid malignancy may depend upon the histology of the thyroid cancer. Papillary thyroid cancers are more likely to be solid, hypoechoic and lacking a halo compared with follicular thyroid cancers. Follicular cancers most commonly have a halo (90%), which is irregular (60%), and are iso- to hyperechoic (44). Therefore, it is critical to recognize that ultrasound does not replace FNA cytology; rather, the two modalities are complementary. Two clinical scenarios illustrate the synergy of these modalities in clinical decision-making. First, if multiple thyroid nodules are present as potential candidates for FNA, sonographic appearance can assist in nodule selection. For example, FNA would be first performed for a 1.6 cm hypoechoic solid vascular nodule, even in the presence of a larger 3.7 cm mixed echogenicity nonvascular nodule. Second, if a nodule is borderline in size for FNA, it may be reasonable to pursue FNA if the nodule has a suspicious sonographic appearance, but to continue medical surveillance for one that lacks suspicious features. In fact, for small nodules <1.5 cm where the cost-benefit analysis of FNA is unclear, decisionmaking based upon suspicious sonographic features of hypoechogenicity, microcalcifications, irregular margins or increased vascularity is superior to using an arbitrary size cutoff of >1 cm to identify thyroid cancers (9, 10). Change in size. After a benign FNA cytology result, surveillance recommendations include periodic assessment of nodule size with re-aspiration if growth is observed. Since ultrasound is superior to physical examination for nodule size determination, size changes should be determined by serial sonography. However, agreement does not exist for definition of nodule growth by ultrasonography, and the recent consensus statement on thyroid nodules published by the Society of Radiologists in Ultrasound acknowledged that “the panelists
92
S.J. MANDEL ET AL.
did not come to a consensus on how to define substantial … nor on how to monitor growth” (40). Some groups have suggested a 15% increase in nodule volume (38). However, interobserver variation for ultrasound determination of nodule volume is reported to be about 45% (45). Reproduction of the same planar image of the nodule for follow-up may be difficult. The American Thyroid Association guidelines proposed as a reasonable definition of growth a 20% increase in 2 of the 3 nodule diameters, with a minimum increase of at least 2 mm (2). This methodology is actually quite useful because it equates with at least a 44% increase in nodule volume, which would overcome the reported interobserver variability allowing for determination of true change in size. With the exception of the finding of suspicious cervical lymphadenopathy, no single sonographic feature or combination of features is adequately sensitive to identify all malignant nodules. However, certain features and combinations of features have high predictive value to indicate if a nodule is likely to be malignant or benign. Clinical judgment, assessment of personal risk factors and ultrasound appearance combined with FNA cytology provide the optimal process for nodule diagnosis. References 1. Mazzaferri EL (1993) Management of a solitary thyroid nodule. N Engl J Med 328(8):553–559 2. Cooper DS, Doherty GM, Haugen BR et al (2006) Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 16(2):109–142 3. American Association of Clinical Endocrinologists and Associazione Medici Endocrinologi (2006) Medical guidelines for clinical practice for the diagnosis and management of thyroid nodules. Endocr Pract 12(1):63–102 4. Marqusee E, Benson CB, Frates MC et al (2000) Usefulness of ultrasonography in the management of nodular thyroid disease. Ann Intern Med 133(9):696–700 5. Brander A, Viikinkoski P, Tuuhea J, Voutilainen L, Kivisaari L (1992) Clinical versus ultrasound examination of the thyroid gland in common clinical practice. J Clin Ultrasound 20(1):37–42 6. Tan GH, Gharib H, Reading CC (1995) Solitary thyroid nodule: comparison between palpation and ultrasonography. Arch Intern Med 155(22):2418–2423 7. Alexander EK, Heering JP, Benson CB et al (2002) Assessment of nondiagnostic ultrasound-guided fine needle aspirations of thyroid nodules. J Clin Endocrinol Metab 87(11):4924–4927
ULTRASOUND OF THYROID NODULES
93
8. Hall TL, Layfield LJ, Philippe A, Rosenthal DL (1989) Sources of diagnostic error in fine needle aspiration of the thyroid. Cancer 63(4):718–725 9. Leenhardt L, Hejblum G, Franc B et al (1999) Indications and limits of ultrasound-guided cytology in the management of nonpalpable thyroid nodules. J Clin Endocrinol Metab 84(1):24–28 10. Papini E, Guglielmi R, Bianchini A et al (2002) Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clin Endocrinol Metab 87(5):1941–1946 11. Schneider AB, Ron E, Lubin J, Stovall M, Gierlowski TC (1993) Dose-response relationships for radiation-induced thyroid cancer and thyroid nodules: evidence for the prolonged effects of radiation on the thyroid. J Clin Endocrinol Metab 77(2):362–369 12. Shibata Y, Yamashita S, Masyakin VB, Panasyuk GD, Nagataki S (2002) 15 years after Chernobyl: new evidence of thyroid cancer. Lancet 358(9297):1965–1966 13. Hemminki K, Eng C, Chen B (2005) Familial risks for nonmedullary thyroid cancer. J Clin Endocrinol Metab 90(10):5747–5753 14. Malchoff CD, Malchoff DM (2002) The genetics of hereditary nonmedullary thyroid carcinoma. J Clin Endocrinol Metab 87(6):2455–2459 15. Uchino S, Noguchi S, Yamashita H et al (2004) Detection of asymptomatic differentiated thyroid carcinoma by neck ultrasonographic screening for familial nonmedullary thyroid carcinoma. World J Surg 28(11):1099–1102 16. Yeh HC, Futterweit W, Gilbert P (1996) Micronodulation: ultrasonographic sign of Hashimoto thyroiditis. J Ultrasound Med 15(12):813–819 17. Vejbjerg P, Knudsen N, Perrild H et al (2006) The association between hypoechogenicity or irregular echo pattern at thyroid ultrasonography and thyroid function in the general population. Eur J Endocrinol 155(4):547–552 18. Gutekunst R, Hafermann W, Mansky T, Scriba PC (1989) Ultrasonography related to clinical and laboratory findings in lymphocytic thyroiditis. Acta Endocrinol (Copenh) 121(1):129–135 19. Langer JE, Khan A, Nisenbaum HL et al (2001) Sonographic appearance of focal thyroiditis. AJR Am J Roentgenol 176(3):751–754 20. Pedersen OM, Aardal NP, Larssen TB, Varhaug JE, Myking O, VikMo H (2000) The value of ultrasonography in predicting autoimmune thyroid disease. Thyroid 10(3):251–259 21. Zingrillo M, D’Aloiso L, Ghiggi MR et al (1996) Thyroid hypoechogenicity after methimazole withdrawal in Graves’ disease: a useful index for predicting recurrence? Clin Endocrinol (Oxf) 45(2):201–206 22. Baldini M, Orsatti A, Bonfanti MT, Castagnone D, Cantalamessa L (2005) Relationship between the sonographic appearance of the thyroid and the clinical course and autoimmune activity of Graves’ disease. J Clin Ultrasound 33(8):381–385
94
S.J. MANDEL ET AL.
23. Hiromatsu Y, Ishibashi M, Miyake I et al (1999) Color Doppler ultrasonography in patients with subacute thyroiditis. Thyroid 9(12):1189–1193 24. Takashima S, Matsuzuka F, Nagareda T, Tomiyama N, Kozuka T (1992) Thyroid nodules associated with Hashimoto thyroiditis: assessment with US. Radiology 185(1):125–130 25. Cantalamessa L, Baldini M, Orsatti A, Meroni L, Amodei V, Castagnone D (1999) Thyroid nodules in Graves’ disease and the risk of thyroid carcinoma. Arch Intern Med 159(15):1705–1708 26. Cappelli C, Pirola I, Cumetti D et al (2005) Is the anteroposterior and transverse diameter ratio of nonpalpable thyroid nodules a sonographic criteria for recommending fine-needle aspiration cytology? Clin Endocrinol (Oxf) 63(6):689–693 27. Rago T, Vitti P, Chiovato L et al (1998) Role of conventional ultrasonography and color-flow Doppler sonography in predicting malignancy in “cold” thyroid nodules. Eur J Endocrinol 138(1):41–46 28. Takashima S, Fukuda H, Nomura N, Kishimoto H, Kim T, Kobayashi T (1995) Thyroid nodules: re-evaluation with ultrasound. J Clin Ultrasound 23(3):179–184 29. Brkljacic B, Cuk V, Tomic-Brzac H, Bence-Zigman Z, DelicBrkljacic D, Drinkovic I (1994) Ultrasonic evaluation of benign and malignant nodules in echographically multinodular thyroids. J Clin Ultrasound 22(2):71–76 30. Cerbone G, Spiezia S, Colao A et al (1999) Power Doppler improves the diagnostic accuracy of color Doppler ultrasonography in cold thyroid nodules: follow-up results. Horm Res 52(1):19–24 31. Kim EK, Park CS, Chung WY et al (2002) New sonographic criteria for recommending fine-needle aspiration biopsy of nonpalpable solid nodules of the thyroid. AJR Am J Roentgenol 178(3):687–691 32. Nam-Goong IS, Kim HY, Gong G et al (2004) Ultrasonographyguided fine-needle aspiration of thyroid incidentaloma: correlation with pathological findings. Clin Endocrinol (Oxf) 60(1):21–28 33. Frates MC, Benson CB, Doubilet PM et al (2006) Prevalence and distribution of carcinoma in patients with solitary and multiple thyroid nodules on sonography. J Clin Endocrinol Metab 91(9):3411–3417 34. Wienke JR, Chong WK, Fielding JR, Zou KH, Mittelstaedt CA (2003) Sonographic features of benign thyroid nodules: interobserver reliability and overlap with malignancy. J Ultrasound Med 22(10):1027–1031 35. Ahuja A, Chick W, King W, Metreweli C (1996) Clinical significance of the comet-tail artifact in thyroid ultrasound. J Clin Ultrasound 24(3):129–133 36. de los Santos ET, Keyhani-Rofagha S, Cunningham JJ, Mazzaferri EL (1990) Cystic thyroid nodules: the dilemma of malignant lesions. Arch Intern Med 150(7):1422–1427
ULTRASOUND OF THYROID NODULES
95
37. Henrichsen T, Reading CC, Charboneau JW, Donovan DJ, Hay ID (2005) Cystic change in thyroid carcinoma: frequency and extent in 360 cancers by sonography. In: Annual Meeting of Radiology Society of North America, p. 245 38. Alexander EK, Hurwitz S, Heering JP et al (2003) Natural history of benign solid and cystic thyroid nodules. Ann Intern Med 138(4):315–318 39. Reading CC, Charboneau JW, Hay ID, Sebo TJ (2005) Sonography of thyroid nodules: a “classic pattern” diagnostic approach. Ultrasound Q 21(3):157–165 40. Frates MC, Benson CB, Charboneau JW et al (2005) Management of thyroid nodules detected at US: Society of Radiologists in Ultrasound consensus conference statement. Radiology 237(3):794–800 41. Taki S, Terahata S, Yamashita R et al (2004) Thyroid calcifications: sonographic patterns and incidence of cancer. Clin Imaging 28(5):368–371 42. Alexander EK, Marqusee E, Orcutt J et al (2004) Thyroid nodule shape and prediction of malignancy. Thyroid 14(11):953–958 43. Rago T, Santini F, Scutari M, Pinchera A, Vitti P (2007) Elastography: new developments in ultrasound for predicting malignancy in thyroid nodules. J Clin Endocrinol Metab 92(8):2917–2922 44. Jeh SK, Jung SL, Kim BS, Lee YS (2007) Evaluating the degree of conformity of papillary carcinoma and follicular carcinoma to the reported ultrasonographic findings of malignant thyroid tumor. Korean J Radiol 8(3):192–197 45. Brauer VF, Eder P, Miehle K, Wiesner TD, Hasenclever H, Paschke R (2005) Interobserver variation for ultrasound determination of thyroid nodule volumes. Thyroid 15(10):1169–1175
CHAPTER 7
Ultrasound-Guided Fine-needle Aspiration of Thyroid Nodules Daniel S. Duick and Susan J. Mandel
INTRODUCTION There are many benefits to performing an ultrasound examination prior to performing a fine-needle aspiration (FNA) as discussed in Chap. 6. They include determining the size and position of a nodule, which allows better selection of needle length and needle size. In patients having multinodular goiter, ultrasound assures biopsy of the dominant nodule or the nodules most likely to be malignant—those having microcalcifications, increased vascularity, marked hypoechogenicity, blurred irregular borders or other characteristics associated with malignancy. Finally, ultrasound may redirect the FNA to other areas of suspicion, such as an enlarged lymph node or a parathyroid adenoma. Once the endocrinologist becomes accustomed to performing thyroid ultrasound at the time of doing an FNA, it is a simple progression to combine the two procedures into an ultrasoundguided FNA (UGFNA). Indeed, this technique is essential to biopsy non-palpable nodules and most nodules less than 1.5 cm in size. UGFNA is also necessary in many obese, muscular or large frame patients or in patients having osteoporosis where a nodule is felt in the upright position, but it is difficult to palpate when the patient is supine. UGFNA is indicated for the biopsy of complex or cystic nodules in order to obtain material from the mural or solid component of the nodule and assure adequate cytology. In solid nodules, the best cytology material is usually obtained from the entire nodule. However, many nodules undergo changes centrally as they grow and this chapter will describe a number of ultrasound-guided FNA techniques utilized to diagnose problematic nodules. In heterogeneous nodules, the biopsy should be taken from the hypoechoic area of 97
98
D.S. DUICK AND S.J. MANDEL
the nodule and any area with any additional suspicious findings (e.g., regions of intranodular Doppler blood flow, microcalcifications, etc). UGFNA allows for this more precise placement of the needle tip within the nodule. Multiple investigators have revealed that combining ultrasound and FNA into a single procedure, UGFNA, leads to a three-fold to five-fold increase in satisfactory cellular yields for cytology interpretation compared to conventional FNA (1, 2). Others have demonstrated an increase in both FNA specificity and sensitivity when UGFNA was performed (3, 4). UGFNA assures the needle tip is in the nodule (avoiding false negatives) and allows the operator to avoid the trachea and great vessels in the neck. The technique will usually allow the operator to avoid passing the needle through the sternocleidomastoid muscle and thus significantly decreases the discomfort of the procedure. Because UGFNA maximizes the quality and quantity of the cytology, it has become the single best tool with which to evaluate and manage thyroid nodules. MICRONODULES The question of whether to biopsy nodules less than 1.5 cm (micronodules or “incidentalomas”) is controversial. Many argue that nodules this size seldom present a threat and are so common in the population that routine biopsy of all such nodules is not cost effective. However, several investigators have shown that the incidence of malignancy in small nonpalpable nodules is the same as in palpable nodules (5, 6). In addition, others have shown that cancers that present less than 1.5 cm in size are often as aggressive as larger cancers (7). Clearly some judgment is required in deciding which nodules require FNA. The American Association of Endocrinologists (AACE) recommends performing FNA on nodules over 1 cm in size (8). Smaller nodules in patients who received external radiation to the head or neck during childhood or in patients with a family history of medullary or papillary thyroid cancer also need UGFNA. Patients who have had a hemithyroidectomy for thyroid cancer are candidates for UGFNA if a micronodule should be found in the remaining lobe. Small nodules that appear taller than wide in the transverse view on ultrasound or have an increased intranodular vascular pattern with Doppler interrogation have a higher risk of malignancy and should have an UGFNA. Other characteristics such as microcalcifications, an irregular/blurred border, or marked hypoechogenicity which is comparable to strap muscles also
ULTRASOUND-GUIDED FINE-NEEDLE ASPIRATION
99
indicate micronodules that may require UGFNA (9, 10). Most other nodules 1 cm or less in size can safely be observed over a period of time using ultrasound, and FNA can be avoided if there is no significant increase in size. PREPARATION Prior to consideration of a thyroid nodule aspiration, a history of relative contraindications should be obtained. These are the same as with a conventional FNA and include patients who may not be able to lie recumbent due to physical problems, or who have difficulty in controlling the rate and depth of respiration, as well as patients who are uncooperative because of anxiety. The former may be able to be aspirated at 45–60 degrees elevation of the upper body or in a semi-sitting position. Patients with breath rate control issues or anxiety may require mild sedation or an anxiolytic medication in order to obtain a satisfactory biopsy. An informed written and signed consent should be obtained after a verbal discussion during which all questions have been answered. The consent form should contain in lay language all details and additional information regarding the reason for the procedure, who will be performing the UGFNA, a description of the procedure and risks, as well as patient and witness signatures. An absolute contraindication is the presence of a severe, uncorrected bleeding, platelet or coagulopathy disorder rendering the patient incapable of homeostasis. Relative contraindications include the use of injectable heparin products, warfarin with an above-therapeutic-range INR, clopidrogrel and large-dose aspirin therapy. These latter situations can be associated with an increased frequency of local puncture site bleeding, ecchymoses and hematoma formation. These can usually be treated by manual tamponage followed by a pressure-taped dressing and an ice pack. If a hematoma occurs, it should be observed by ultrasound to assure stabilization prior to the patient’s departure. The individual physician operator’s judgment and experience in performing UGFNA in these situations of potentially increased risk is most important. Full disclosure of risks to the patient is mandatory prior to proceeding with UGFNA. If withholding anticoagulation or anti-platelet therapy for greater than 24 hours is deemed appropriate, the procedure should be deferred and the referring or treating physician should be contacted regarding concerns, risks and plans for problem resolution and UGFNA disposition.
100
D.S. DUICK AND S.J. MANDEL
MATERIALS The ultrasound laboratory should consist of an ultrasound machine with a probe and linear transducer that has a 3.5–5.0 cm footprint and multiple frequency settings ranging between 7.5–14MHz. The machine should also have Doppler imaging capabilities (e.g., color-flow Doppler and preferably power Doppler also). Larger footprint transducers are cumbersome and may impede aspiration capabilities. An additional or utility probe with a similar or lower frequency range is a 1–2 cm curvilinear or curved linear array transducer (a linear transducer with a convex-curved footprint that produces an image with an increased field of view in a sector format). The smaller curvilinear transducer is useful for imaging difficult locations, especially in the low neck at the level of the manubrium, clavicles and insertions of the sternocleidomastoid muscle. Additional laboratory items should include a mobile ultrasound machine, set up tray and imaging table each of which can be moved for optimum positioning visualization and utilization during an UGFNA or other procedure. The setup tray should include all materials required for topical cleansing, as well as transducer covers, sterile coupling gel and an assortment of needles readily assembled and accessible to perform UGFNA. A detachable needle guide adapted for the transducer
FIG. 7.1. From L to R demonstrates different footprint size, linear array transducers with the transducer on the right demonstrating a smaller, curvilinear array transducer
ULTRASOUND-GUIDED FINE-NEEDLE ASPIRATION
101
is usually not necessary for routine UGFNA but is often used and helpful in specialized and prolonged procedures such as drainage of a large cyst followed by percutaneous ethanol injection. Almost any needle will be visible on modern highresolution ultrasound equipment; this makes use of echogenic needles unnecessary. An assortment of small needles (25–27 g) and medium needles (21–23 g) and specialty needles (25 g, 23 g or 21 g stylet-type needles or spinal needles) of variable lengths and types should be part of the routine setup (Fig. 7.2). The stylet-type needles are used for prolonged fluid aspirations or aspirating structures posterior to the thyroid, which may or may not lie within the thyroid (e.g., thyroid nodule versus parathyroid tumor or lymph node). The stylet can be left in while advancing the needle into the lesion of interest and prevents the uploading of thyroid cells into the needle. The stylet also stiffens the needle, making it easier to maneuver prior to withdrawing the stylet when in the nodule and performing an aspirate. Use of commercially available heparinized needles is not necessary for properly obtained specimens, and heparin may interfere with cytology interpretation. For aspiration technique, a 10 cc slip-on tip or Luer lock syringe is recommended—preferably with a peripheral or eccentric tip (enhances visibility of the syringe hub and bevel
FIG. 7.2. Tray of needles used for UGFNA
102
D.S. DUICK AND S.J. MANDEL
tip viewing of the needle). Pistol grip holders are not recommended since they are cumbersome and often apply excessive negative aspiration pressure, inducing bleeding and poor aspirates. A useful variation of the pistol grip holder is a smaller, spring-loaded aspiration device (e.g., Tao aspirator), which combines both a syringe holder for stabilization and allows for the presetting of the aspiration pressure. This may be especially useful when UGFNA is performed without assistance and when both hands are required for imaging and aspirating. The preset aspiration pressure setting is triggered after needle insertion into the nodule and the other hand continues to hold the transducer and monitor the procedure (Fig. 7.3). The use of injectable or topical anesthesia is optional for a 27 g or 25 g needle procedure. For aspirations with larger needles (e.g., less than 25 g), the operator may choose to use one or more of the following: ethyl chloride spray, topical lidocaine gel or patch (applied prior to procedure) or injectable 1% or 2% lidocaine. These should all be readily available depending on the patient’s request or the perceived need of a procedure that is technically difficult or may involve multiple nodules or repetitive aspirations. Spray fixative or transport fixative solution for thin smears of slides should also be readily available on the setup tray. A written laboratory protocol for the performance of UGFNA and all laboratory associated procedures should be available for reference.
FIG. 7.3. Demonstrates use of a small, spring-loaded aspiration device (Tao) during UGFNA without the use of an assistant for stabilizing the transducer and monitor image during the procedure
ULTRASOUND-GUIDED FINE-NEEDLE ASPIRATION
103
TECHNIQUE The patient should be positioned supine with the neck extended and soft pillow or pad inserted beneath the shoulders to optimize extension of the neck. An additional, small, soft pillow may be placed behind the head for patients who have known neck problems or discomfort with extension of the head and/or rotation of the extended neck. Based on prior knowledge of the planned procedure, the operator should position oneself optimally for the performance of the aspiration procedure. The monitor should be clearly visible to the operator/physician during the entire procedure. Prior to prepping the neck, an extended field of view should be performed before every needle biopsy. Both lobes of the thyroid, the isthmus, low central region and the lateral neck should all be observed for any abnormalities or lymphadenopathy not previously detected. Coupling gel is applied to the transducer face, and the transducer is then enclosed in a sheath or cot to avoid
FIG. 7.4. Use of monitor for imaging UGFNA
104
D.S. DUICK AND S.J. MANDEL
direct contact with any blood products. A low cost alternative transducer cover is Parafilm. The covered transducer is dipped in alcohol, and the neck area is prepped with alcohol swabs. Sterile coupling gel can be applied to the covered transducer face or directly to the prepped neck area. The key to utilizing ultrasound guidance for performance of the FNA is understanding the orientation of the azimuthal plane, which is the mid-sagittal plane of the transducer face. The transducer sends and receives high frequency ultrasound waves from and along the azimuthal or mid-sagittal plane of the transducer face. Utilization of the azimuthal plane during UGFNA imaging allows the operator to visualize the needle pathway or approach, angle of needle insertion and either track the entire needle by a parallel approach or locate only the bevel of the needle within a nodule by a perpendicular approach. Thus, there are basically two approaches for performing the UGFNA based on orientation to the azimuthal plane. PARALLEL APPROACH The ultrasound-guided parallel approach tracks the needle from the point of insertion down and along the azimuthal plane to the nodule. The needle is oriented and introduced at either end of the mid portion of the transducer, which is in parallel to the mid-plane of the long axis or sagittal plane (Fig 7.5.). On the monitor screen, the nodule is positioned off-center and closer to the screen’s lateral border on the side of planned needle insertion. The needle is best inserted with the bevel up towards the transducer since this has angular edges with a flat surface producing greater reflectance and a “brighter” ultrasound image of the tip of the needle. The transducer and needle need to be maintained in the same plane. Upon needle penetration of the skin the needle tip appears at the upper right or upper left corner (depending on orientation of the transducer) of the monitor screen. As the needle is advanced forward and into the nodule, it is carefully guided along and within or adjacent to the azimuthal plane in parallel fashion. This approach allows the operator to observe needle penetration, location and pathway of the entire needle within the neck, thyroid and nodule, which remain visible on the monitor (Fig 7.6.). If the needle course veers laterally or away from or out of the azimuthal plane even a few degrees, it will be lost from the monitor screen. The parallel technique requires practice and experience to utilize successfully.
ULTRASOUND-GUIDED FINE-NEEDLE ASPIRATION
105
FIG. 7.5. Longitudinal orientation and alignment of transducer, needle and syringe during performance of parallel approach for UGFNA
FIG. 7.6. The upper panel graphically depicts parallel approach with needle and tip visualization during UGFNA. The lower panel demonstrates ultrasound image of needle and tip (arrows) during parallel approach for UGFNA
106
D.S. DUICK AND S.J. MANDEL
PERPENDICULAR APPROACH In the perpendicular approach, the nodule is imaged and positioned in the mid portion of the screen rather than off center to either lateral side of the monitor. In this way, the point of needle introduction and the nodule beneath for aspiration are both centered in the mid point of the transducer’s side or long axis in order to transversely cross the azimuthal plane at 90 degrees (Fig 7.7). This again requires experience and skill since the needle itself will not be visualized during the performance of the biopsy. The needle bevel is again introduced with the bevel facing upward toward the transducer to reflect the ultrasound waves and detect its bright image as it crosses the azimuthal plane during needle penetrance of the nodule (Fig 7.8.). Understanding and visualizing the various angles of needle descent needed to match the depth of the nodule in the neck is most important when performing UGFNA in the perpendicular approach. The angle of descent dictates whether the needle bevel will be visualized within the nodule (necessary to perform FNA) or above the nodule (descent angle too shallow) or below the nodule (descent angle too steep) as the bevel crosses the narrow beam of the azimuthal plane. Repetitive practice and utilization of both the parallel approach and the perpendicular approach will result in optimizing the orientation and skills of the operator to enhance the performance of UGFNA. ASPIRATION AND NON-ASPIRATION TECHNIQUES In general, the use of ultrasound at the time of planned FNA allows the operator to assess for solid, partially cystic and multi compartmental cystic nodules. Based on this assessment and the initial pass of an UGFNA, different approaches may be required to obtain adequate sampling and aspirated material for cytologic interpretation. UGFNA can localize tissue areas with vascularity by Doppler interrogation in partially or mostly cystic nodules and enhance acquisition of material for cytology interpretation. There are two basic techniques for obtaining cellular material from a nodule during UGFNA: (1) the closed suction, “free hand” technique is performed with a 27 or 25 g needle attached to a 10 cc syringe. This needle is introduced into the nodule, under ultrasound guidance and the plunger is withdrawn for 1–3 cc of negative pressure to induce aspiration. The needle is moved back and forth within diameter of a solid nodule at 3 cycles per second over 5 to 6 seconds, the aspiration pres-
ULTRASOUND-GUIDED FINE-NEEDLE ASPIRATION
107
FIG. 7.7. Transverse orientation of transducer at 90 degrees to alignment of needle and syringe during perpendicular approach for UGFNA
FIG. 7.8. The upper panel graphically depicts perpendicular approach with needle tip only visualization as it crosses azimuthal plane in a nodule during UGFNA. The lower panel demonstrates ultrasound image of the needle tip only as it crosses azimuthal plane (arrow) within a nodule during perpendicular approach for UGFNA
108
D.S. DUICK AND S.J. MANDEL
sure is released and the needle withdrawn. When liquid or low viscosity diluted material is encountered, this technique may be modified to perform the aspiration in the 2–5 mm subcapsular region (peripherally located tissue is less likely to undergo degenerative changes or dilute the aspirate specimen from complex nodules). The syringe is then detached from the needle, the plunger withdrawn (allowing 2 or 3 cc of air into the syringe), the needle reattached and the plunger moved slowly forward to extrude aspirated material onto a glass slide for smear and fixation preparation. Many times, however, the nodule is composed of loosely formed microcystic and degenerating tissue and fluid, or the nodule is highly vascularized internally. In these situations a more dilute material rapidly appears in the syringe above the level of the needle hub. If this continues after repeat aspiration attempts with less negative pressure, switching to the “needle only” (Zajdela) technique usually improves acquisition of optimum cellular material for slide preparation (13). The “needle only” or Zajdela technique utilizes a 27 or 25 g needle and the principles of needle bevel nodule penetration and capillary action uploading of cellular material into the needle without aspiration. The needle is grasped at the hub between the thumb and index finger and introduced into the nodule with one–three quick up and down motions over 1–3 seconds. The index fingertip is then placed over the hub to close the system and the needle is withdrawn, reattached to a syringe with the plunger retracted approximately 2–3 cc, and the material is extruded onto a slide for smear preparation and fixation. Usually two to four separate needle passes are made. A modification of this approach is to remove the plunger from a 10 cc syringe, attach the needle to the syringe (for enhanced control of the needle) and perform an open-system aspirate. The thumb pad is placed over the end of the syringe at the time of needle withdrawal. The needle is detached and reattached to a syringe with a partially withdrawn plunger, and material is extruded onto a slide for smear and fixation. Another modification is to leave the plunger in the syringe and draw up 2–3 cc of air in the syringe prior to aspiration. Cellular material enters the needle via capillary action during the procedure, and the aspirate can then be extruded directly onto a slide. If the material obtained is frank blood or a watery mixture of interstitial, cystic and degenerative or bloody fluid, a modification of the “needle only” technique is often helpful. Again, two–four individual needle passes are performed, but an
ULTRASOUND-GUIDED FINE-NEEDLE ASPIRATION
109
exceedingly rapid forward penetration approximating a fraction of a second is utilized, and the hub is immediately capped by the fingertip. This allows for maximum needle bevel cutting and cellular acquisition with only minimal capillary action time to avoid fluid uploading into the needle. If the extruded material continues to be a watery texture or if it is optimal texture material but the physician is inexperienced in smear and slide preparation, the material can be extruded into a transport media (check with your reference laboratory for the required or preferred transport media) and forwarded to the laboratory to process, cytospin and prepare slides for interpretation. When slides are produced on sight, they are either fixed with a spray or placed in a 95% ethanol solution bottle and capped. Additional air dried smears may be requested by the cytopathologist. Cellularity may also be checked by microscopy at the time of FNA by air drying 1 or 2 smears and performing a rapid staining technique with Diff Quick or a similar product. The essence and desired outcome of UGFNA is the acquisition of cellular material and the production of smears on glass slides for fixation, future staining and interpretation. The capability to produce interpretable slides of aspirated material cannot be overemphasized. If the physician has poor technical skills in slide production, the entire procedure is a worthless exercise and a wasted opportunity. The repetitive need to depend on or utilize diluted material in transport media for subsequent cytospin or cell block preparations usually reduces diagnostic accuracy which is more optimally accomplished by having the capability of producing direct smears and fixation of aspirated materials on glass slides. Although slide production is not intrinsic to this chapter, the reader who is poorly trained or repeatedly has non-interpretable slides should either enroll in a slide-making cytology course or attend a training course to learn this and all skills associated with UGFNA. SUMMARY There are different methods of performing UGFNA, but there is no single best method. The techniques described are those widely utilized; they are not meant to be prescriptive but to provide a starting point for those who desire to start learning this procedure. You will discover many adaptations that can be customized to individual situations. It is important that the endocrinologist develop expertise in UGFNA in order to optimize patient care, safety and outcomes.
110
D.S. DUICK AND S.J. MANDEL
References 1. Takashima S, Fukuda H, Kobayashi T (1994) Thyroid nodules: clinical effect of ultrasound-guided fine needle aspiration biopsy. J Clin Ultrasound 22:536–542 2. Danese D, Sciacchitano S, Farsetti A, Andreoli M, Pontecorvi A (1998) Diagnostic accuracy of conventional versus sonographyguided fine needle aspiration biopsy of thyroid nodules. Thyroid 8:15–21 3. Carmeci C, Jeffery RB, McDougall IR, Noweis KW, Weigel RJ (1998) Ultrasound-guided fine-needle aspiration biopsy of thyroid masses. Thyroid 8:283–289 4. Yang GCH, Liebeskind D, Messina AV (2001) Ultrasound-guided fine-needle aspiration of the thyroid assessed by ultrafast papanicolaou stain: data from 1,135 biopsies with a two to six year follow-up. Thyroid 11:581–589 5. Hagag P, Strauss S, Weiss M (1998) Role of ultrasound-guided fine-needle aspiration biopsy in evaluation of nonpalpable nodules. Thyroid 8:989–995 6. Leenhardt L, Hejblum G, Franc, Fediaevsky LD, Delbot T, Le Guillozic D, Menegaux F, Guillausseau C, Hoang C, Turpin A, Aurengo A (1999) Indications and limits of ultrasound-guided cytology in the management of nonpalpable thyroid nodules. J Clin Endo Metab 84:24–28 7. Rosen I, Azadian A, Walfish P, Salem S, Lansdown E, Bedard Y (1995) Ultrasound-guided fine-needle aspiration biopsy in the management of thyroid disease. Am J Surg 166:346–349 8. Gharib H, Papini E, Valcavi R et al (2006) American Association of Clinical Endocrinologist/Associazone Medici Endocrinologi Medici guidelines for clinical practice for diagnosis and management of thyroid nodules. AACE/AME Task Force on Thyroid Nodules. Endocrine Practice 12:63–192 9. Papini E, Guglielmi R, Bianchini A, Crescenzi A, Taccogna O, Nardi F, Panunzi C, Rinaldi R, Toscano V, Parcella CM (2002) Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clin Endo Metab 87:1941–1946 10. Kim E, Park CS, Chung WY, Oh KK, Kim DI, Lee JT, Yoo HS (2002) New sonographic criteria for recommending fine-needle aspiration biopsy of nonpalpable solid nodules of the thyroid. AJR 178:687–691 11. Marqusee E, Benson CB, Frates MC et al (2000) Usefulness of ultrasonography in the management of nodular thyroid disease. Ann Int Med 133: 696–700 12. Baudin E, Travagli JP, Ropers J et al (1998) Microcarcinoma of the thyroid gland: the Gustave Roussy Institute experience. Cancer 83: 553–559 13. Zajdela A, de Maublanc MA, Schlienger P, Haye C (1986) Cytologic diagnosis of orbital and periorbital palpable tumors using fineneedle sampling without aspiration. Diagn Cytopathol 2:17–20
Chapter 8
Ultrasound in the Management of Thyroid Cancer H. Jack Baskin
INTRODUCTION The strategic value of ultrasound in the postoperative surveillance of patients with thyroid cancer and in the preoperative surgical planning of patients undergoing thyroid cancer surgery has become increasingly appreciated over the past decade. In this chapter we will focus on how to recognize and differentiate malignant lymph nodes from benign lymphadenopathy. Once you become familiar with the appearance of metastatic lymph nodes in thyroid cancer, you will find that ultrasound is a more specific tool for separating benign from malignant lymph nodes than it is for separating benign and malignant thyroid nodules. However, we still must rely on ultrasound-guided FNA for a definitive diagnosis. POSTOPERATIVE SURVEILLANCE FOR THYROID CANCER Ultrasound has assumed a primary role in the management of patients who have been treated for thyroid cancer. In spite of better surgical techniques, the acceptance of total and near-total thyroidectomy, and the increasing use of radioiodine, the mortality rate from well-differentiated thyroid cancer has changed very little over the past thirty years. Because of its propensity to occur at any age, even in the very young, and to recur many years later, thyroid cancer must be monitored for the lifetime of the patient. Surveillance of these patients in a cost-effective manner has been a challenge. Until the 1990s the only diagnostic tool available was a 131I whole body scan (WBS) done after withdrawing the patient from thyroid hormone replacement. The sensitivity of a WBS in the early detection of residual, recurrent, or metastatic thyroid cancer is poor. This is apparent 111
112
H.J. BASKIN
from the many patients who have increased thyroglobulin (Tg) but negative diagnostic scans who are treated with 131I and have positive post-treatment scans (1–4). Park et al. have also shown that the doses of 131I used for WBS can stun the uptake of iodine in metastatic lesions and interfere with the subsequent treatment dose of 131I (5). The expense, poor sensitivity, and risk of stunning with a WBS make it an unsatisfactory test with which to follow patients with thyroid cancer. In the last decade several new probes have been developed that aid in the early detection of recurrent thyroid cancer. These include: (1) sensitive, reliable, reproducible Tg assays that biochemically detect the earliest sign of cancer recurrence; (2) development of recombinant TSH (rhTSH) that allows scanning and Tg stimulation without thyroid hormone withdrawal; and (3) high-resolution ultrasound of the postoperative neck to identify early lymph node recurrence. Using these new tools, especially neck ultrasound combined with UG FNA of suspicious lymph nodes, has greatly improved the sensitivity of cancer surveillance in these patients. Hopefully, their use will result in lower mortality from thyroid cancer. Physical examination of the neck of a patient who has undergone a thyroidectomy for thyroid cancer is seldom helpful in the early detection of a recurrence. The scar tissue following surgery, combined with the propensity of metastatic lymph nodes to lie deep in the neck beneath the sternocleidomastoid muscle, make palpation of enlarged lymph nodes in the neck difficult. Even lymph nodes several centimeters in diameter are often not palpable. High-resolution ultrasound has solved this problem by proving to be a very sensitive method to find and locate early recurrent cancer and lymph node metastasis. Frasoldati et al. (6) studied 494 patients with a history of low risk well-differentiated thyroid cancer by a withdrawal WBS, stimulated Tg, and ultrasound, and found by at least one test that 51 had had a recurrence. The WBS was positive in 23 patients (45%), the Tg was positive in 34 patients (67%), and the ultrasound with FNA was positive in 48 patients (94%). Since most thyroid cancer metastasizes to the neck, it is rare for it to spread elsewhere without neck lymph node involvement. Therefore, neck ultrasound has proven to be the most sensitive test available in locating early recurrent disease, even before serum Tg is elevated. ULTRASOUND OF THE POSTOPERATIVE NECK Identifying and evaluating lymph nodes should be done with high-resolution ultrasound using a 10–15MHz transducer with power Doppler capability to assess vascularity. When performing
ULTRASOUND MANAGEMENT OF THYROID CANCER
113
ultrasound of the neck in a patient who has undergone a thyroidectomy, one sees that the carotid artery and jugular vein have migrated medially close to the trachea, and that the thyroid bed has been filled with a varying amount of hyperechoic connective tissue that appears white (dense) on ultrasound. This serves well in demarcating a recurrence of cancer or a metastatic lymph node, which will appear dark or hypoechoic. Someone unfamiliar with the appearance of the postoperative neck on ultrasound should begin by examining the neck of someone who underwent a thyroidectomy or hemithyroidectomy for benign disease. This allows one to become accustomed to the neck structures and the altered anatomy of the postoperative neck without worrying about recurrent thyroid cancer. The commonest areas for detecting cancer are the thyroid bed and the jugular chain of lymph nodes, but metastatic lymph nodes may occur anywhere in the neck. In performing ultrasound looking for metastatic lymph nodes, the entire length of the internal jugular vein from the head of the clavicle up to the mandible is searched, paying close attention to the area between the carotid artery and the jugular vein. Special attention should be given to the thyroid bed and the central compartment medial to the common carotid artery. Malignant paratrachael lymph nodes in this area are likely to metastasize more quickly to the mediastinum and lungs.
FIG. 8.1. Normal postoperative left neck. Note that the common carotid artery and the internal jugular vein have migrated medially next to the trachea. The vein is anterior to the artery but closely adhered to it. Hyperechoic connective tissue has filled in the thyroid bed.
114
H.J. BASKIN
FIG. 8.2. Normal postoperative right neck. In this patient the vein remains lateral to the artery, but still lies adjacent to it. The strap muscles (sm) have helped fill in the space left by removal of the thyroid.
The normal neck contains approximately 300 lymph nodes. Except for the pharyngeal area, they are usually less than 0.5 cm in their short axis and flattened or oval in the transverse view of the neck, with a long axis two or more times the short axis. If they become inflamed or hyperplastic, they enlarge but generally maintain this flattened or oval shape. High-resolution ultrasound often shows a white line of fat and intranodal blood vessels running through the center of the lymph node referred to as a hilar line. The hilar line is present in most benign lymph nodes greater than 0.5cm and is also more prominent in older patients. A hilar line is seldom seen in malignant lymph nodes. Because lymph node hyperplasia is so common in the neck, only those lymph nodes >0.5cm in the short axis are usually biopsied. Those with a short axis 0.5cm (0.8cm in the pharyngeal area) or less should have their location marked and be reexamined in six months. Metastatic lymph nodes generally have a fuller or more rounded appearance in the transverse view with a short/long axis ratio >0.5. Postoperative ultrasound surveillance for cancer is done in the transverse view, since all lymph nodes may appear elongated in the longitudinal view. In addition to a rounded shape and the absence of a hilar line, there are other ultrasound findings that suggest a lymph node is malignant (7) (Table 8.1). The internal jugular vein
ULTRASOUND MANAGEMENT OF THYROID CANCER
115
TABLE 8.1. Neck lymph node characteristics Benign Short/Long Axis Hilar line Jugular Deviation or Compression Microcalcifications Cystic Necrosis Vascularity
<0.5 Present Absent Absent Absent Central
Malignant >0.5 Absent Present Present Present Chaotic/peripheral
FIG. 8.3. Benign lymph node. The normal neck contains scores of lymph nodes, some of which are easily seen with ultrasound. This lymph node (calipers) appears benign because it is flat with a short/ long axis ratio <0.5.
remains lateral or migrates anterior next to the carotid artery in the postoperative neck. Since metastatic nodes commonly occur in proximity to the jugular vein or in the carotid sheath, any deviation of the jugular vein away from the carotid artery strongly suggests malignancy. The entire length of the vessels should be surveyed closely with particular attention given to any area where the artery and vein diverge. In addition to causing deviation of the internal jugular vein, malignant lymph nodes tend to compress the vein and cause a partial obstruction to blood flow. Benign lymph nodes rarely do this until they become quite large. Calcifications in the lymph node, either microcalcifications or amorphous calcium with shadowing, are indicative
116
H.J. BASKIN
FIG. 8.4. This lymph node beneath the sternocleidomastoid muscle (scm) is slightly more oval but still maintains a short/long axis ratio <0.5. It also has a distinct hilar line (arrow), a strong indication that it is benign.
FIG. 8.5. Power Doppler of the previous lymph node shows vascularization of the hilum, which contains small arterioles. Note there is no vascularization seen in the periphery of the node.
of malignancy. Cystic necrosis within the lymph node, often recognized because of distal enhancement, is another sign of metastatic involvement with thyroid cancer, although it may also occur with tuberculosis.
ULTRASOUND MANAGEMENT OF THYROID CANCER
117
FIG. 8.6. Malignant lymph node. This lymph node (calipers) is slightly more rounded, with a short/long axis ratio > 0.5 in the transverse view. Note the absence of a hilar line, which makes this node suspicious. An UG FNA was needed to confirm malignancy.
FIG. 8.7. Same lymph node in longitudinal view. It appears more benign in this view because it is flatter; even malignant lymph nodes can be long in longitudinal view. Therefore, always take the short/long axis measurement in the transverse view.
Power Doppler is very useful in evaluating lymph nodes because of its sensitivity to arteriolar blood flow. Normal nodes generally show only hilar vascularization, but malignant nodes have chaotic vascularization throughout the cortex. This is due to the recruitment of vessels into the periphery of the node (8, 9).
118
H.J. BASKIN
FIG. 8.8. On transverse view, this small rounded lymph node (calipers) without a hilar line is in close proximity to the great vessels.
FIG. 8.9. Same lymph node (calipers) in longitudinal view shows compression of the jugular vein against the carotid. UG FNA confirmed malignancy.
Characteristics that were helpful in deciding if a thyroid nodule is benign or malignant may not apply to lymph nodes. For example, metastatic lymph nodes may have sharp borders until they become quite large. Both normal and malignant lymph nodes are generally hypoechoic compared to thyroid,
ULTRASOUND MANAGEMENT OF THYROID CANCER
119
FIG. 8.10. This 0.5cm lymph node (calipers) lies between the carotid and the jugular. Its location and shape (short/long axis ratio >1) strongly suggest that it is malignant, which was confirmed by UG FNA.
FIG. 8.11. Although this lymph node (arrow) measures only 2.5mm, its location and shape lead to UG FNA, and Tg was found in the needle washout, confirming metastatic thyroid cancer.
but they have varying degrees of echogenicity. Early papillary metastases are sometimes dense and may be relatively hyperechoic. As they enlarge up to 1cm they develop cystic necrosis and become hypoechoic. Therefore, echogenicity may not be helpful in determining malignancy. Matting of lymph nodes
120
H.J. BASKIN
FIG. 8.12. This irregular rounded lymph node (arrow) was discovered because of the separation of the jugular from the carotid. The calcification at 3:00 o’clock indicates it is malignant, but UG FNA is necessary before surgery.
FIG. 8.13. Transverse view of a metastatic lymph node (calipers) in right neck beneath the sternocleidomastoid muscle (scm) and lateral to the carotid artery. The node is impinging upon the jugular vein (J). The short/long axis ratio is >0.5 and no hilar line is seen. UG FNA had positive cytology, and Tg was found in the needle washout.
occurs with malignancy but is not a helpful sign, since it is also seen with inflammation or in patients who have had radiation.
ULTRASOUND MANAGEMENT OF THYROID CANCER
121
FIG. 8.14. Longitudinal view of the same lymph node (calipers), showing partial obstruction of the jugular vein. In this view, the short/long axis ratio is <0.5, emphasizing the need to measure the short/long ratio in the transverse view.
FIG. 8.15. This markedly heterogeneous lymph node (calipers) contains scattered calcifications indicating metastatic papillary carcinoma.
Because the sonographic features of malignant lymph nodes are not always present, and there is overlap in the ultrasound appearance of benign and malignant lymph nodes, UG FNA of suspicious lesions is essential for a definitive diagnosis and
122
H.J. BASKIN
FIG. 8.16. This 2cm rounded lymph node in the right neck is 80% cystic; note the distal enhancement. Although occasionally seen in tuberculosis, cyst formation within a lymph node usually indicates metastatic papillary carcinoma.
FIG. 8.17. This rounded lymph node (calipers) without a hilar line is hypoechoic and has begun to develop cystic necrosis. Liquid formation within a solid lymph node is often first suspected because of distal enhancement (arrow).
before recommending surgery. Lymph nodes with a short axis >0.5cm (>0.8cm in pharyngeal area), with a short/long axis ratio >0.5, and that do not have a hilar line must have an UG FNA.
ULTRASOUND MANAGEMENT OF THYROID CANCER
123
FIG. 8.18. This metastatic lymph node, less than 1cm in size, contains cystic necrosis on the medial side which is hypoechoic (calipers) and shows enhancement. The other side (arrow) is solid and hyperechoic. UG FNA of the hypoechoic area yielded negative cytology, but high levels of Tg in needle washout—a finding not unusual if there is cystic necrosis.
FIG. 8.19. Typical small metastatic lymph node (calipers) near the jugular vein (J). Note the rounded shape with a short/long axis of 1, absence of a hilar line, calcification (white arrow), and enhancement (black arrow), indicating early cystic necrosis.
124
H.J. BASKIN
FIG. 8.20. Power Doppler of previous lymph node shows chaotic vascularization of the periphery of the node, rather than the normal hilar vascular pattern. Although cytology from the FNA was negative, a high level of Tg was found in the needle washout.
FIG. 8.21. Ultrasound of a 16-year old-female one year post-thyroidectomy. Power Doppler of a small lymph node (arrow) found in the central compartment between the trachea and the left carotid shows chaotic vascularization.
UG FNA of a suspicious lymph node in the neck is carried out in the same manner as an UG FNA of a thyroid nodule with aspirate slides prepared and sent for cytology interpretation.
ULTRASOUND MANAGEMENT OF THYROID CANCER
125
FIG. 8.22. Larger lymph node found in the lateral compartment of the same patient. Power Doppler again shows an abnormal pattern with recruitment of vessels into the cortex of the lymph node.
FIG. 8.23. 2cm nonpalpable lymph node (calipers) in 47-year-old male seven years post-thyroidectomy. Enhancement distal to the node (arrow) indicates cystic necrosis has started. FNA found negative cytology, but very high levels of Tg in the needle washout.
Lymph node cytology is sometimes difficult to interpret (10). However, thyroid cancer metastases contain Tg, which can be measured and used as a tissue marker. Therefore, the biopsy
126
H.J. BASKIN
FIG. 8.24. Ultrasound of 54-year-old female 36 years after her thyroidectomy reveals a paratrachael lymph node (arrow) in the right central compartment. Note that the short/long axis is >1 and several calcifications are seen, indicating malignancy. FNA showed positive cytology, but Tg in needle washout was negative, demonstrating the need to do both tests when lymph nodes are biopsied.
FIG. 8.25. A 1cm lymph node (1) and a 0.5cm lymph node (2) in the lateral neck are both rounded without a hilar line. Both nodes had papillary cancer at surgery.
needle(s) is then washed with 1cc normal saline and the washout sent for Tg assay (11, 12). A normal saline control is also sent for Tg assay. Since most patients are on thyroid hormone suppression, serum Tg is usually low or non-detectable, and
ULTRASOUND MANAGEMENT OF THYROID CANCER
127
FIG. 8.26. Ultrasound of a 57-year-old female 13 years after a thyroidectomy shows a small oval node (calipers) in the left central compartment that is suspicious.
FIG. 8.27. Power Doppler of previous lymph node shows a malignant vascular pattern. Cytology was negative on FNA, but Tg in needle washout was high, confirming malignancy.
the material in the needle is diluted approximately a hundredfold to a thousandfold; therefore finding a Tg >10 in the needle washout is considered positive for malignancy. Because the intracellular Tg is not exposed to circulating anti-TgAB, a positive test for anti-TgAB in the serum does not interfere with
128
H.J. BASKIN
FIG. 8.28. This lymph node (calipers) was suspicious because of its borderline short/long axis of 0.5 and absent hilum. Surgery confirmed metastatic papillary cancer.
FIG. 8.29. Metastatic lymph node (calipers) in left central compartment with short/long axis >1 and no hilar line.
the measurement of Tg obtained from lymph nodes, as it does with serum Tg (13). Either a positive cytology report, or finding Tg present in the needle washout, confirms the lymph node is malignant. Using either positive cytology or the presence of Tg as proof of recurrent cancer, Lee et al. reported 100%
ULTRASOUND MANAGEMENT OF THYROID CANCER
129
FIG. 8.30. Another lymph node (calipers) less than 0.5cm that was biopsied because of its shape and location. FNA cytology showed papillary carcinoma.
FIG. 8.31. Ultrasound of 50-year-old female 18 years after total thyroidectomy revealed a paratrachael lymph node (calipers) in the central compartment with a short/long axis of 1.
sensitivity and specificity of UG FNA in detecting recurrent thyroid cancer (14). Studies have found the Tg in the needle washout to be more sensitive than cytology in detecting malignancy (15). This is probably due to poor cellular material in lymph nodes having cystic necrosis (see Fig. 18–20).
130
H.J. BASKIN
FIG. 8.32. Power Doppler of previous lymph node showed peripheral vascular pattern of malignancy. FNA cytology was positive and Tg in the needle washout was >10,000.
Any suspicious lymph nodes should be mapped using the surgical areas defined by the American Joint Commission on Cancer, even if no FNA is done. This will allow one to go back and restudy the node at a later date to see if any changes have occurred. Mapping of malignant lymph nodes is particularly important when FNA is performed, in order that the surgeon can be directed to the site of the recurrence and plan the surgery appropriately. A modified lateral neck dissection, rather than removing a few nodes, is usually necessary to return an elevated serum Tg to a nondetectable level. Routine isotope scanning is no longer an acceptable method of following patients for recurrent thyroid cancer. Extrapolating from the Chernobyl experience, early detection using ultrasound offers our best hope of eradicating residual or recurrent disease. Ultrasound in conjunction with sensitive Tg monitoring provides endocrinologists with the tools to do this. Since 90% of recurrent cancer is initially in the neck, ultrasound of this area is essential. Searching for small hypoechoic lymph nodes or masses in the neck using ultrasound is not difficult, but it does require some effort and patience that is quickly rewarded. Annual ultrasound for at least 5 years after surgery, and ultrasound on all patients with detectable serum Tg or with positive anti-TgAB seems a minimum requirement if we are to make a significant reduction in the mortality rate of thyroid cancer.
ULTRASOUND MANAGEMENT OF THYROID CANCER
131
ULTRASOUND OF THE PREOPERATIVE NECK Obviously, if ultrasound is helpful in the management of thyroid cancer patients after surgery, it should be equally beneficial before surgery. Patients who have been selected for thyroidectomy because of FNA of a nodule or other reasons should be brought back prior to surgery and undergo ultrasound of the lateral neck and central compartment in a search for suspicious lymph nodes. Confirming a malignant lymph node(s) in the lateral neck will extend the surgery beyond a simple thyroidectomy to a modified lateral neck dissection. Likewise, finding a malignant lymph node in the central compartment assures the surgeon that a total thyroidectomy and central node dissection is required. At surgery, well differentiated thyroid cancer involves lymph nodes in 20–50% of patients. Using preoperative ultrasound, suspicious lymph nodes are found in 20–30% of patients. This discrepancy is in part due to ultrasound not being as sensitive as surgery in detecting micrometastasis. Ultrasound may also be unable to see enlarged lymph nodes in the central compartment because they are hidden by the thyroid gland or pushed behind the trachea. Nevertheless, ultrasound will detect most lymph node involvement prior to the initial thyroid surgery. One may question the significance of finding preoperative lymph node involvement in these patients. Ito studied 560 patients who underwent ultrasound prior to having a thyroidectomy for thyroid cancer, and he followed the patients for an average of ten years (16). The recurrence rate of thyroid cancer was 3.1% in the ultrasound negative patients, but the recurrence rate in the ultrasound positive patients was 24.8%. This indicates that recurrence-free survival is definitely worse if lymph node involvement is present on ultrasound at the time of the initial surgery. One may still question the need to perform a lateral neck dissection on such patients, unless it can be shown to improve long-term survival. Noguchi reported on a series of patients found to have macroscopic lymph node involvement at the time of surgery, who were followed for an average of 40 years (17). The long-term survival rate of the group that underwent a modified lateral neck dissection (MLND) was 84%, while that of the group not having MLND was only 66%. Therefore it appears that MLND does alter the outcome in patients known to have lateral lymph node metastasis at the time of surgery. Evidence of the value of preoperative ultrasound of the central compartment is less conclusive. It may be less important
132
H.J. BASKIN
FIG. 8.33. Preoperative ultrasound of a patient seen for a thyroid nodule in the right lobe (N) revealed a large cystic lymph node in the right lateral neck (box). Note the short/long axis >0.5, absent hilum, and abnormal vascularity. Although FNA of the lymph node had negative cytology, Tg in needle washout was positive.
FIG. 8.34. Preoperative ultrasound of patient with a thyroid nodule (N) whose FNA was positive for papillary cancer shows a suspicious lymph node (calipers) adjacent to the lower pole of the thyroid. A central node dissection at the time of surgery revealed that this was metastatic papillary cancer.
ULTRASOUND MANAGEMENT OF THYROID CANCER
133
because many endocrine surgeons now routinely perform a central node dissection at the time of the thyroidectomy. Preoperative ultrasound of the central compartment is much less sensitive than it is in the lateral neck, for the reasons mentioned above. Preoperative FNA of central compartment lymph nodes is fraught with the risk of contamination if the needle inadvertently passes through the thyroid, which might cause a false positive cytology and/or Tg in the washout. It is generally recommended that the central compartment be examined at the time of lateral neck ultrasound. Because 70–90% of positive lymph nodes in the central neck are not seen by ultrasound before the thyroid is removed, a “negative” ultrasound examination is of no value. Suspicious lymph nodes should be reported to the surgeon, but FNA is not recommended. SUMMARY The critical importance of ultrasound in thyroid cancer management is recognized in the 2006 Management Guidelines for Patients with Thyroid Cancer published by the American Thyroid Association. Recommendation 21 is that all patients undergoing surgery for thyroid cancer have a preoperative ultrasound of the neck for lymph nodes. Recommendations 46 and 48 are that routine whole body scans be abandoned, and that periodic ultrasound of the neck be used for the followup of thyroid cancer patients. Implementation of these recommendations by endocrinologists may ultimately lead to a reduction in the rate of recurrence of thyroid cancer and to an increase in the long-term survival of patients. References 1. Pineda J, Lee T, Ain K et al (1995) Iodine-131 therapy for thyroid cancer patients with elevated thyroglobulin and negative diagnostic scan. J Clin Endocrinol Metab 80:1488–1492 2. Schumberger M, Arcangioli O, Piekarski J et al (1988) Detection and treatment of lung metastases of differentiated thyroid carcinoma in patients with normal chest x-ray. J Nucl Med 29:1790–1794 3. Torre E, Carballo M, Erdozain R, Lienas L, Iriarte M, Layana J (2004) Prognostic value of thyroglobulin and I-131 whole-body scan after initial treatment of low –risk differentiated thyroid cancer. Thyroid 14:301–306 4. Pacini F, Lari R, Mazzeo S (1985) Diagnostic value of a single serum thyroglobulin determination on and off thyroid suppressive therapy in the follow-up of patients with differentiated thyroid cancer. Clin Endocrinol 23:405–411
134
H.J. BASKIN
5. Park H, Perkins O, Edmondson J (1994) Influence of diagnostic radioiodines on the uptake of ablative dose of iodine-131. Thyroid 4:49–54 6. Frasoldati A, Presenti M, Gallo M, Coroggio A, Salvo D, Valcavi R (2003) Diagnosis of neck recurrences in patients with differentiated thyroid carcinoma. Cancer 97:90–96 7. Ahuja A, Ying M, Phil M, King A, Yuen HY (2001) Lymph node hilus—gray scale and power Doppler sonography of cervical nodes. J Ultrasound Med 20:987–992 8. Ahuja A, Ying M, Yuen H, Metreweli C (2001) Power Doppler sonography of metastatic nodes from papillary carcinoma of the thyroid. Clinical Radiology 56:284–288 9. Ahuja A, Ying M (2002) An overview of neck node sonography. Investigative Radiology 37:333–342 10. Ballantone R, Lombardi C, Raffaelli M, Traini E, Crea C, Rossi E et al (2004) Management of cystic thyroid nodules: the role of ultrasound-guided fine-needle aspiration biopsy. Thyroid 14:43–47 11. Frasoldati A, Toschi E, Zini M, Flora M, Caroggio A, Dotti C et al (1999) Role of thyroglobulin measurement in fine-needle aspiration biopsies of cervical lymph nodes in patients with differentiated thyroid cancer. Thyroid 9:105–111 12. Pacini F, Fugazzola I, Lippi F, Ceccarelli C, Centoni R, Miccoli P, Elisei R, Pinchera A (1992) Detection of thyroglobulin in the needle aspirates of nonthyroidal neck masses: a clue to the diagnosis of metastatic differentiated thyroid cancer. J Clin Endocrinol Metab 74:1401–1404 13. Baskin HJ (2004) Detection of recurrent papillary thyroid carcinoma by thyroglobulin assessment in the needle washout after fine-needle aspiration of suspicious lymph nodes. Thyroid 14:959–963 14. Lee M, Ross D, Mueller P, Daniels G, Dawson S, Simeone J (1993) Fine-needle biopsy of cervical lymph nodes in patients with thyroid cancer: a prospective comparison of cytopathologic and tissue marker analysis. Radiology 187:851–854 15. Cignarelli M, Ambrosi A, Marino A, Lamacchia O, Campo M, Picca G (2003) Diagnostic utility of thyroglobulin detection in fine-needle aspiration of cervical cystic metastatic lymph nodes from papillary thyroid cancer with negative cytology. Thyroid 13:1163–1167 16. Ito Y,Tomoda C, Uruno T et al (2005) Ultrasonographically and anatomopathologically detectable node metastases in the lateral compartment as indicators of worse relapse-free survival in patients with papillary carcinoma. World J Surg 29:917–920 17. Noguchi S, Murakami N, Yamashita H et al (1998) Prognostic factors in patients with differentiated thyroid carcinoma. Eur J Surg 166:29–33
CHAPTER 9
Parathyroid Ultrasonography Devaprabu Abraham
INTRODUCTION Primary hyperparathyroidism (PHPT) is a common endocrine condition affecting approximately 100,000 new patients each year in the United States (1). An apparent increase in the incidence of PHPT can be traced to the wide availability and use of multi-channel analyzers leading to earlier detection of the disease since 1970 (2). In more than 85 percent of these cases, a solitary adenoma is the cause of the problem. Accurate preoperative localization of the adenoma allows minimally invasive surgery to be performed, leading to shorter hospitalization and recuperative times (3). ANATOMY OF THE PARATHYROID GLANDS Precise understanding of the normal location and the anatomical variations of the parathyroid glands is the cornerstone to successful identification of parathyroid adenomas. Normal parathyroid glands are ovoid, or bean-shaped, and measure approximately 3 by 5 mm in size with the superior glands being smaller than the inferior glands. They have an anatomically distinct vascular supply from that of the thyroid gland. Normal parathyroid glands are enveloped in a pad of a fibro-fatty capsule, and are seldom seen on ultrasound (4). However, this capsule may become compressed when the parathyroid enlarges, and is often seen on ultrasound as a hyperechoic line between the thyroid gland and the parathyroid adenoma. Post-mortem examination reveals that four glands are found in 91% of subjects, three glands are found in 5%, and five glands in 4% (5). More supernumerary glands appear to be a rare occurrence (6). The anatomic location of parathyroid glands varies widely due to the embryonic origin of the glands from the 4th and 3rd pharyngeal pouches with eventual migration to the lower neck. The superior parathyroid glands
135
136
D. ABRAHAM
FIG. 9.1. Inferior parathyroid adenoma in transverse view.
develop in the 4th pharyngeal pouch, and migrate caudally to situate along the upper two thirds of the posterior margin of the thyroid lobes. The superior parathyroid glands are more constant in their location in relation to the thyroid gland due to their common developmental origins. The third pharyngeal pouch gives rise to the inferior parathyroid glands and the thymus gland and, together, they migrate to the lower neck. Fortyfour percent of these glands are located within 1 cm of the inferior pole of the thyroid gland, 17% are in close proximity to the inferior margin of the thyroid gland, 26% are found in relation to the superior portion of the thymus along the thyrothymic ligament, and 2% remain within the mediastinum portion of the thymus (7). Unusual variations in the location of the parathyroid glands include the carotid bifurcation, within the carotid sheath, intrathyroidal and retropharyngeal locations. Due to these anatomical variations, accurate preoperative localization becomes crucial for the success of minimally invasive parathyroid surgery. LOCALIZATION STUDIES It is important to remember that the diagnosis of PHPT is made by chemical tests (i.e., elevated serum calcium and parathormone levels, after ruling out hereditary hypocalcuric hypercalcemia and other causes of hypercalcemia). Localizing studies are not useful for diagonising PHPT and should be obtained after the diagnosis of PHPT has been confirmed. The two most
PARATHYROID ULTRASONOGRAPHY
137
FIG. 9.2. Inferior parathyroid adenoma seen in longitudinal view.
FIG. 9.3. Superior parathyroid adenoma seen in transverse view. Note thyroid nodule in anterior lobe.
widely used studies for locating abnormal parathyroid gland(s) have been 99 Tc MIBI—which is a functional study—and ultrasonography—an anatomical study. The respective studies have their strengths and weaknesses. To a practicing endocrinologist, the use of ultrasonography to study a patient with a suspected parathyroid adenoma poses several advantages. These include the ease of availability of ultrasound equipment, the lack of ionizing irradiation, the short duration of the procedure, and the
138
D. ABRAHAM
FIG. 9.4. Superior parathyroid adenoma seen in longitudinal view.
potential cost savings realized by the patient. Unfortunately, not all parathyroid adenomas can be visualized by ultrasound. Very small adenomas and those located in aberrant locations, such as the mediastinum or posterior to the trachea, are not amenable to ultrasound examination. The limitations of parathyroid localization using ultrasonography also include operator variability—however, an experienced sonographer can be expected to identify 85% of parathyroid adenomas. Similar results have been reported using 99 Tc MIBI scans (60–94%), but they are equally operator dependent. While false positive results are sometimes seen with 99 Tc MIBI scans due to thyroid nodules that trap the isotope, UGFNA of parathyroid adenomas seen with ultrasound have 100% specificity. One should also keep in mind that isotope scans lateralize the adenoma while ultrasound studies localize the lesion. TECHNIQUE OF PARATHYROID ULTRASOUND There are many good reasons to perform an ultrasound on a patient having surgery for hyperparathyroidism. Several investigators have shown a 2% incidence of thyroid cancer in patients with a parathyroid adenoma. Not infrequently, a preoperative ultrasound will reveal an unsuspected coexisting thyroid nodule, which, after FNA, may alter the entire surgery. Finally, ultrasound allows localization of the parathyroid adenoma in most patients, thereby facilitating minimally invasive surgery.
PARATHYROID ULTRASONOGRAPHY
139
Proper positioning of the patient is critical to the successful visualization of enlarged parathyroid glands. The subject should be made to lie flat on a firm table with one or two pillows placed under the shoulders to enable full extension of the neck. The linear probe (3–5 cm) is placed along with the coupling gel close to the skin and the thyroid gland is located first. The structures of the neck should be carefully studied in two or more axis at multiple levels of the neck. Most clinicians use multifrequency probes (5–15 MHz) for the study of parathyroid glands. Although high frequency increases resolution, the lower frequency probes are more effective at visualizing the deeper reaches of the neck. The most common locations (posterior border of the thyroid lobes, tracheoesophogeal groove, thyroid-thymic ligament) are searched first, looking for a hypoechoic mass with the features of a parathyroid adenoma. The following are the most distinctive sonographic features of parathyroid adenomas. 1. Extrathyroidal Location: The majority of parathyroid adenomas are located adjacent to, but separate from, the posterior aspect of the thyroid. It is common to see an indentation made by the parathyroid adenoma on the posterior capsule of the thyroid gland (Figures 9.4 and 9.10). Look for the echogenic line separating the adenoma from thyroid tissue. This represents the compressed fibro-fatty capsule surrounding the enlarged parathyroid. Parathyroid adenomas are completely embedded within the thyroid gland in about 2–5% of cases (9, 10). The incidence of intraglandular location is higher in patients with multigland disease—in one series, 3% in patients with single-gland disease versus 15% in those with hyperplasia (10). 2. Hypoechoic and Homogeneous Appearance: The most typical imaging characteristic of parathyroid adenomas is the homogeneously hypoechoic echogenicity in relation to the thyroid gland (11). 3. Oval, Elongated, Triangular, or Oblong Shapes: Parathyroid adenomas conform to the pressures of surrounding anatomical structures and therefore variation in the shapes is common. An excellent method of learning to recognize enlarged parathyroid glands is by performing ultrasound on patients with renal failure. These patients frequently have very enlarged parathyroid glands that are easily seen with ultrasound. This exercise will allow the beginner to become acquainted with the endless variety of shapes and locations of parathyroid glands.
140
D. ABRAHAM
4. Vascular Pedicle and Blood Flow Seen with Doppler: The presence of an extrathyroidal artery (polar artery) feeding an adenoma may be found in 83% of parathyroid adenomas (12). Besides the visualization of the polar artery, other patterns such as the vascular arc pattern and diffuse flow within the adenoma have also been described (13). PARATHYROID INCIDENTALOMA Subclinical parathyroid tumors can be incidentally discovered during neck ultrasonography. The frequency of observing these incidental tumors is <1% (17, 18). Fine needle aspiration with PTH estimation in syringe washings can identify these lesions
FIG. 9.5. A and B Polar vascular pedicle.
FIG. 9.6. A Polar blood flow depicted by color flow Doppler.
FIG. 9.6. B Dual Polar blood flow depicted by Power Doppler.
FIG. 9.6. C Arc pattern of blood flow.
142
D. ABRAHAM
FIG. 9.7. Diffuse blood flow seen within adenoma.
FIG. 9.8. Polar blood flow. This adenoma was found within the carotid sheath during surgery.
as parathyroid tumors. Their prognosis for developing hyperparathyroidism in the future is presently not known. PARATHYROID CYST Cystic parathyroid adenomas are rare. Simple cysts of the parathyroid glands without hypercalcemia are occasionally seen during ultrasound assessment of the thyroid. Partial cystic change of an adenoma is depicted in Fig. 9.15. Syringe washout PTH estimation is useful to prove the origin of these cysts.
PARATHYROID ULTRASONOGRAPHY
143
FIG. 9.9. a, b, and c Parathyroid adenoma visualized along side of incidental thyroid disease. (a. multinodular goiter b. colloid nodule c. Hashimoto thyroiditis)
144
D. ABRAHAM
FIG. 9.10. Parathyroid adenoma indenting the posterior capsule of the thyroid gland.
FIG. 9.11. Double adenoma visualized in longitudinal view. Note also PA located within the thyroid gland capsule.
ULTRASOUND-GUIDED FNA OF PARATHYROID LESIONS UGFNA of suspected parathyroid adenomas can be performed in the office setting, and syringe washings can been submitted for parathormone (PTH) estimation (14, 15). Lesions >1.5 cm
PARATHYROID ULTRASONOGRAPHY
145
FIG. 9.12. Double inferior parathyroid adenoma in panoramic view. The findings were confirmed during surgery.
FIG. 9.13. Cystic PA.
with obvious ultrasound features of a parathyroid adenoma may not need biopsy confirmation. FNA confirmation is most useful when multiple lesions are identified, in which case the lesion with the least identification characteristics of a parathyroid adenoma should be sampled. FNA also allows one to differentiate a parathyroid adenoma from coexistent posterior thyroid nodules (15). (See Fig. 9.X) Other situations that are candidates for FNA include suspected parathyroid masses in patients with: 1) prior failed surgery, 2) negative 99 Tc MIBI scans, 3) atypical location, or 4) coexistent multinodular goiter.
146
D. ABRAHAM
Elevation of intact PTH in syringe washings provides confirmation with virtually 100% specificity. Parathyroid cytology has a very limited role in the diagnosis of parathyroid adenomas (15). The technique of parathyroid FNA is similar to that of thyroid FNA. One or two passes using 27 G needles is recommended. Due to the location depth of these lesions, longer needles may be necessary to enable a biopsy to be performed. Parathyroid tumors can be mobile and may need a sharp, brief, and abrupt jab to penetrate the capsule. Parathyroid lesions provide bloody aspirates and the fluid should enter the hub of the needle. Absence of bloody aspirate during FNA is typically encountered with non parathyroid lesions. The syringe aspirate can be processed as follows: 1. Make one slide and rinse the remainder of the specimen in 2 ml of normal saline. 2. Centrifuge the specimen, remove and freeze the supernatant before transporting to the laboratory. 3. Do not submit the slide for cytological evaluation until the syringe-washing PTH results become available.
GRAPH. 1. Parathyroid hormone (PTH) levels in syringe washing specimens obtained after fineneedle aspiration of parathyroid adenomas of study subjects and proven thyroid nodules as controls, represented in log scale. Findings in the 2 groups were significantly different (p<0.001).
PARATHYROID ULTRASONOGRAPHY
147
FIG. 9.14. Biopsy of suspected PA.
FIG. 9.15. PA located in the proximal portion of thyro-thymic ligament.
If the PTH level is high in the syringe aspirate, the cytology is not necessary. If the PTH level is low in the syringe aspirate, the smear is submitted for cytological analysis. The latter technique provides an additional measure of safety in case of sampling a metastatic lymph node either from thyroid or other coincidental occult cancers. Most laboratories in the United States are willing to perform intact PTH estimations in tissue specimens if prior arrangements
148
D. ABRAHAM
FIG. 9.16. Fluid from a parathyroid cyst
are made with the supervisor of the lab. In our laboratory, the phenomenon of hook effect has not been detected in over 100 specimens subjected to PTH estimation by the above-described technique, despite the PTH levels being enormously elevated (15). In summary, most parathyroid adenomas can readily be seen with ultrasound and do not require FNA. Sonography provides an inexpensive method of localizing the lesion prior to surgery without ionizing radiation, and facilitates minimally invasive parathyroid surgery. References 1. Kebebew E, Clark OH (1998) Parathyroid adenoma, hyperplasia and carcinoma: Localization, technical details of primary neck exploration and treatment of hypercalcemic crisis. Surg Oncol Clin N Am. 7:721–748 2. Heath H III, Hodgson SF, Kennedy M (1980) Primary hyperparathyroidism: incidence, morbidity and potential economic impact in a community. N Engl J Med 302:189–193 3. Udelsman R, Donovan PI (2004) Open minimally invasive parathyroid surgery. World J Surg. Dec. 28(12):1224–6 4. Gilmore JR (1938) The gross anatomy of parathyroid glands. J Pathol 46:133, 5. Alveryd A (1968) Parathyroid glands in thyroid surgery. Acta Chir Scand 389:1 6. Wang CA, Mahaffey JE, Axelrod L, et al (1979) Hyperfunctioning supernumerary parathyroid glands. Surg Gynecol Obstet 148:711 7. Akerstrom G, Malmaeus J, Bergstrom R (1984) Surgical anatomy of human parathyroid glands. Surgery 95:14 8. Yeh MW, Barraclough BM, Sidhu SB, Sywak MS, Barraclough BH, Delbridge LW (2006) Two hundred consecutive parathyroid ultrasound studies by a single clinician: the impact of experience. Endocr Pract. May-Jun.12(3):257–63
PARATHYROID ULTRASONOGRAPHY
149
9. Andre V, Andre M, Le Dreff P, Granier H, Forlodou P, Garcia JF (1999) Intrathyroid parathyroid adenoma J Radiol. Jun. 80(6):591–2 10. McIntyre R Jr, Eisenach J, Pearlman N, Ridgeway C, Liechty RD (Date?) Intrathyroidal parathyroid glands can be a cause of failed cervical exploration for hyperparathyroidism. The American Journal of Surgery 174(6):750–754 11. Kamaya A, Quon A, Jeffrey RB (2006) Sonography of the abnormal parathyroid gland. Ultrasound Q. Dec. 22(4):253–62 12. Lane MJ, Desser TS, Weigel RJ, Jeffrey RB Jr (1998) Use of color and power Doppler sonography to identify feeding arteries associated with parathyroid adenomas. AJR Am J Roentgenol. Sep. 171(3):819–23 13. Wolf RJ, Cronan JJ, Monchik JM (1994) Color Doppler sonography: an adjunctive technique in assessment of parathyroid adenomas. J Ultrasound Med. Apr 13(4):303–8 14. Doppman JL, Krudy AG, Marx SJ, Saxe A, Schneider P, Norton JA, Spiegel AM, Downs RW, Schaaf M, Brennan ME, Schneider AB, Aurbach GD (1983) Aspiration of enlarged parathyroid glands for parathyroid hormone assay. Radiology Jul 148(1):31–5 15. Abraham D, Sharma PK, Bentz J, Gault PM, Neumayer L, McClain DA (2007) The utility of ultrasound guided FNA of parathyroid adenomas for pre-operative localization prior to minimally invasive parathyroidectomy. Endocrine Practice. Jul/Aug 13(4) pages? 16. Krause UC, Friedrich JH, Olbricht T, Metz K (1996) Association of primary hyperparathyroidism and non-medullary thyroid cancer. Eur J Surg. Sep 162(9):685–9 17. Pesenti M, Frasoldati A, Azzarito C, Valcavi R (1999) Parathyroid incidentaloma discovered during thyroid ultrasound imaging. J Endocrinol Invest. Nov 22(10):796–9 18. Frasoldati A, Pesenti M, Toschi E, Azzarito C, Zini M, Valcavi R (Year?) Detection and diagnosis of parathyroid incidentalomas during thyroid sonography. Journal of Clinical Ultrasound. 27(9): 492 – 498
Chapter 10
Contrast-Enhanced Ultrasound in the Management of Thyroid Nodules Enrico Papini,* Giancarlo Bizzarri, Antonio Bianchini, Rinaldo Guglielmi,* Filomena Graziano,* Francesco Lonero, Sara Pacella, and Claudio Pacella
INTRODUCTION Thyroid nodules are discovered by palpation in 3–7% of subjects in the general population, but an epidemic of clinically unapparent thyroid lesions is detected by high-resolution ultrasonography (US) of the cervical region (1–3). The clinical importance of thyroid nodules, besides the infrequent local compressive symptoms or thyroid dysfunction, is the possibility of thyroid cancer, which occurs in about 5% of all thyroid nodules (4–5). Thus it is essential to improve our diagnostic tools to avoid the use of unnecessary diagnostic surgery. Brightness-mode US is currently the most accurate imaging test to evaluate solitary thyroid nodules or multinodular goiters (6–8). Thyroid US results in improved management for patients, with clinical findings suggestive of thyroid nodules (9). Many patients either have a palpable but not suspicious nodule, or have incidentally revealed but sonographically relevant nodules that warrant fine needle aspiration biopsy (10). Unfortunately, in most cases US characteristics cannot unequivocally distinguish benign and malignant lesions (10–12). Color Doppler US was proposed to evaluate nodule
*
Department of Endocrine, Metabolic and Digestive Diseases & Department of Diagnostic Imaging Albano (Rome), Italy
151
152
E. PAPINI ET AL.
vascularity, since hypervascularity with an intranodular chaotic arrangement of blood vessels is supposed to be associated with malignancy. However, several reports have failed to consistently identify cancer on color Doppler alone (10–12). Thyroid fine needle aspiration (FNA) biopsy is established as the most accurate test for detecting malignant lesions (6–8, 13). In several series, 70% of FNA specimens are benign, 5% malignant, 10% indeterminate or suspicious, and 15% unsatisfactory (13–14). The cytological report is critical in dictating whether the patient’s management should be medical or surgical. Benign and malignant findings are reliably identified by an experienced cytopathologist, but in a few patients indeterminate FNA results occur because the morphologic criteria used to identify malignant lesions appear ill-defined. Overall, about 20% of these indeterminate specimens are malignant (15–16). Patients with indeterminate nodules (“follicular lesions”) should undergo radioisotope scanning to rule out a hyperfunctioning, and usually benign, lesion. However, most follicular lesions are nonfunctional and radioisotope scans add no further information. As clinical criteria and US appearance fail to consistently identify a malignant nodule within this group, it is generally agreed that cytologically indeterminate thyroid nodules are best managed with surgical treatment (6–8, 17). Moreover, despite repeat US-guided FNA biopsy, a residual 5–10% of thyroid lesions remain nondiagnostic. Currently, solid nodules with nondiagnostic cytology and progressive growth are surgically excised (18–19). Recently, new US techniques, such as harmonic and pulse inversion imaging, have been developed that are extremely sensitive to the nonlinear effects of US interactions with contrast agents (20–21). Contrast-enhanced US (CEUS) was reported to improve the identification of malignant focal liver lesions (22–23). Quantitative parametric curve analysis of hypervascularized lesions aided the differential diagnosis of focal nodular hyperplasia, the second most common benign tumor of the liver, from hepatocellular carcinomas (HCC) and hypervascular hepatic metastases (24). Because of the persistence of relevant “gray zones” in the diagnostic workup of thyroid nodules (6–9), the potential usefulness of CEUS for the assessment of the risk of malignancy of thyroid lesions (25–28), and for an improved definition of the extent of tissue ablation induced by percutaneous mini-invasive techniques (29–38), is currently under evaluation.
CONTRAST-ENHANCED ULTRASOUND IN THE MANAGEMENT
153
TECHNICAL BACKGROUND Conventional clinical US imaging consists of a single or multidimensional representation of the acoustic impedance of tissues. The difference in acoustic impedance between normal and pathological tissue may be so small that even high sensitivity US equipment cannot discriminate it. In these cases, the vascularity pattern may help in the detection and characterization of pathological tissue. Doppler examination can successfully separate echoes reflected from blood or tissue (39). This technique relies on the relative high velocity of blood compared to the movements of the surrounding tissues, and is effective in detecting the rapid blood flow in large vessels, but not the slow flow within the small parenchymal vessels. Subtraction techniques traditionally used for digital subtraction angiography seem to be poorly suited for the dynamic and interactive nature of US imaging. In the near future, new possibilities may be offered in this field by advances in techniques such as coded imaging and B-flow imaging (40). The characterization of lesion vascularity is a diagnostic problem, not only for US, but for computerized tomography (CT) and magnetic resonance imaging (MRI) as well (41). So the limited sensitivity of CT and MRI in visualizing blood vessels and in detecting density or signal differences between normal and pathological tissue led to the introduction into clinical practice of specific and effective contrast agents. In US imaging a contrast agent introduced intravenously increases the signals reflected from both the large and the small parenchymal vessels and provides additional information about the presence and the characteristics of a pathological tissue. The first example of echo enhancement was the use of US examination after the injection of CO2 into the hepatic artery to improve the visualization of HCCs (42). Cardiologists have subsequently used an isotonic solution, injected intravenously, to increase the blood echogenicity within the right heart chambers, because the air microbubbles contained in the solution were not able to pass through the lung vessels. These first experiences, although of limited clinical use because of the short life of the microbubbles and the use of conventional US techniques, shed light on possible future solutions. These included: a) the use of gas microbubbles enclosed in a lipid or proteinous shell, with a low solubility in biological fluid, and with an elevated resistance to external pressure changes
154
E. PAPINI ET AL.
(venous vs. arterial pressure) (Figures 10.1 A, B, C); and b) the use of advanced US equipment that is able to detect the harmonic component of ultrasounds scattered by a contrast agent perfusing the tissue under examination (Fig. 10.2). Microbubble stability is a function of both gas solubility and shell resistance and, theoretically, well engineered microbubbles could persist in the blood stream for some minutes. Microbubble size is a critical parameter, since they have to be small enough to allow free circulation through the capillaries and, at the same time, large enough to influence the resonance frequency (Fig. 10.3). Resonating microbubbles absorb a large amount of the colliding energy and become scattering bodies. The scattered waves are spherically dispersed around the microbubbles, with a scattering pattern that depends on the peak pressure of the incident ultrasound field. At a low incident pressure, contrast agents produce a linear backscatter enhancement (“undistorted backscatter”) that results in an augmentation of the echoes from the blood. As the transmitted intensity increases beyond a pressure of 50–100 KPa (below the level usually employed for standard diagnostic applications), the backscatter begins to be nonlinear and shows harmonic characteristics. Finally, if the incident pressure reaches 1 MPa, the maximum intensity transmitted by most equipment, the backscatter of the microbubbles shows a transient and nonlinear pattern due to their destruction (Fig. 10.4). This pattern is the basis for the “triggered imaging technique.” Because of the variability in diameter of the microbubble population contained in contrast media, the borders between these patterns are not sharply defined in clinical practice (43). The approximate exposure to ultrasound pressure at the US beam focus can be mathematically calculated in an average tissue. The mechanical index (MI) is defined as the peak rarefactional (negative) pressure divided by the square root of the ultrasound frequency. During clinical diagnostic use the MI usually lies between 0.1 and 2.0. Although a single value is displayed, MI level varies through the whole field of view of each image. This index is broadly approximated depending on the tissue, the beam, and the attenuation characteristics. So, in everyday practice, the same MI may correspond to different negative peak pressures on different scanners, and each manufacturer gives specific recommendations for machine settings. Microbubbles are designed and constructed to resonate at a frequency close to the one used in US scan evaluation (typically 2.0–5.0 MHz for abdominal examination). As aforementioned,
AIR BUBBLES
A
PALMITIC ACID SHELL
CRYSTALLINE GALACTOSE PARTICLES
Crosslink obtained by sonification
albumin chains
Albumin shell octofluoropropane
B
PHOSPHOLIPID
Sulfur Hexafluoride
C FIG. 10.1. A-C Structure of first- (A) and second-generation (B and C) contrast media: A) the air microbubbles are coated with palmitic acid absorbed to crystalline galactose particles; (B and C) microbubbles are composed of an external lipidic or proteinous shell and are filled with sulphur hexafluoride or octofluoropropane
E. PAPINI ET AL.
156
TRANSDUCER BANDWIDTH Fundamental component
First harmonic (contrast media)
3 MHz
6 MHz FREQUENCY
FIG. 10.2. The figure represents the spectrum of ultrasound scattered under an acoustic field by the microbubbles of contrast media. The variable pattern of microbubble behaviour during the contraction and rarefaction phases produces multiple harmonic components that can be detected with a specific US equipment. The first harmonic component is currently used for diagnostic applications
BUBBLE DIAMETER (µm)
10 9 8 7 6 5 4 3 2 1 1
2
3
4
5
6
7
8
9
10
FREQUENCY (MHz) FIG. 10.3. Microbubbles in an acoustic field resonate at a frequency that is a function of their diameter. A bubble with a 3 µm size (the median diameter of a transpulmonary microbubble agent) resonates at about 3 MHz, which is the approximate center frequency for an abdominal US examination
CONTRAST-ENHANCED ULTRASOUND IN THE MANAGEMENT
157
Power (mW/cm2)
500
100
Non-linear transient Scattering (scintillation) Non-linear stationary Scattering (harmonic) Linear scattering (resonant)
50 Reflection
1
Inactive
FIG. 10.4. Microbubble response to different incident pressures. In a very low acoustic pressure microbubbles can be inactive, while in a strong acoustic field a transient nonlinear behaviour due to the microbubble rupture can be observed
the ability of microbubbles to scatter a distorted copy of the colliding energy can be described as a nonlinear or “harmonic” property. When microbubbles are driven by an ultrasound field at an adequately elevated acoustic pressure, the oscillatory excursions reach a level characterized by asymmetrical expansions and contractions (Fig. 10.5). Latest generation wideband transducers have low-noise electronic circuits and are designed for transmitting a fundamental US frequency equal to the resonance frequency of the microbubbles (1.5–3.0 MHz), and for receiving the first harmonic scattered from them (3–6 MHz.). The harmonic response produced by contrast media can be revealed in two ways. The first technique is a filtering of the received signal that conceals the fundamental component. The limit of this technique is its reduced sensitivity, as some amount of the harmonic component may be filtered out with an increase in the noise and a distortion of the original signal. The second technique consists of a firing of sequential US impulses (usually a series of two or three impulses) that are the inverse of one another, followed by the subtraction of the received signal. In this way only the
E. PAPINI ET AL.
158
PRESSURE
+
TIME
−
BUBBLE DIAMETER
+
D0 TIME
− FIG. 10.5. Nonlinear pattern of response of contrast agents in an acoustic field. During the positive phase the stiffness of the microbubbles increases, producing only a small volume contraction. During the rarefaction phase the stiffness decreases, making possible greater changes in microbubble diameter
nonlinear component is detected. The limits of this technique are the reduced time resolution (a reduced frame-rate of US scan), the increase of the transmitted energy which causes an increased destruction of the microbubbles, and the frequent movement artifacts. Currently, a few commercially available software programs allow for the correction of the uncontrolled movements that may occur during US examination (21, 43). Manufacturers frequently use mixed techniques that employ several phase- and amplitude- modulated impulses such as power pulse inversion (PPI) and contrast pulse sequence (CPS). These advances in CEUS have been made possible by the introduction of coded-pulse techniques that increase the quality of standard US images and reduce the drawbacks of the multipulses approach.
CONTRAST-ENHANCED ULTRASOUND IN THE MANAGEMENT
159
FIG. 10.6. Time-intensity curves calculated with a continuous dynamic recording after contrast injection on a defined ROI (blue line). A mathematical function can be fitted on the time-intensity curve (green line) and perfusional parameters can be derived from the shape of the resultant curve
Time-intensity curves obtained by a dynamic recording of the region of interest are fitted to a mathematical curve on which different parameters can be calculated. These parameters may describe the different characteristics of tissue vascularity in a numeric fashion. The shape of the curves depends mainly on the modality of the contrast injection (intravenous infusion or bolus). During CEUS examination, the baseline conditions should be recorded and the correct interval of time for quantitative evaluation should be chosen (Fig. 10.6). Mathematical curves describing the pattern of US enhancement (time-intensity curves), or parametric maps that give an overview of a chosen parameter in the scanned area, can be obtained (Fig. 10.7). In the past, wash-in and wash-out curves have been extensively used to compare the different vascularity of tissues. The currently evaluated parameters are the peak value (PEAK), the time to peak (TTP), the regional blood volume (RBV), the mean transit time (MTT), the regional blood flow (RBF), the refilling time (RT), and the refilling velocity (beta). CLINICAL USE The usefulness of CEUS in abdominal imaging is well established, but its use for small parts US examination is still under investigation (25–28). As for other anatomical districts,
160
E. PAPINI ET AL.
FIG. 10.7. Example of a parametric map representing the regional blood flow curve (RBF) in a follicular adenoma of the thyroid gland
CEUS reveals the microscopic vascularity of thyroid parenchyma, obtaining a spatial and temporal resolution superior to the traditional color Doppler and power Doppler techniques. Moreover, time-intensity curves can be calculated, allowing both qualitative and quantitative evaluation. The parameters that describe the characteristics of blood supply to the tissue under evaluation can be calculated. These parameters have been demonstrated to be of practical use in medical fields such as brain and cardiac imaging, but it is not clear if they can add clinically relevant information to the characterization of thyroid nodular disease. We evaluated the current evidence on the use of CEUS in the US diagnosis of malignant thyroid lesions, and in the assessment of the extent of coagulative necrosis induced by thermal ablation treatments of thyroid nodules. EVALUATION OF THYROID NODULES WITH CONTRAST-ENHANCED SONOGRAPHY TIME-INTENSITY CURVES In 2001, Spiezia et al. (25) evaluated the diagnostic accuracy of a first-generation galactose-based US contrast agent (Levovist, Schering AG, Berlin, Germany) in differentiating benign from malignant thyroid nodules by analysis of the time-intensity curves. correlating the variation of the value of the intensity signal during the contrast transit time. Fifty-four patients scheduled for surgery underwent a basal color and power Doppler evaluation, and then a color Doppler examination after an intravenous bolus injection of contrast. Thyroid carcinomas were reported to have a significantly earlier arrival time
CONTRAST-ENHANCED ULTRASOUND IN THE MANAGEMENT
161
of Levovist than hyperplastic nodules or adenomas (8.1 +/− 1.4, versus 19.6 +/− 2.2 and 16.1 +/− 2.8 seconds, respectively; p < 0.0001). Malignant lesions and adenomas showed an earlier time to peak than benign nodules (14.6 +/− 1.2 and 23.1 +/− 3.8, versus 33.0 seconds, respectively; p < 0.0001), while no significant difference was detected in baseline, peak, and final intensity signal. Similar results were reported in a recent study on 20 patients with benign and malignant thyroid nodules (26). An early wash-in with a homogeneous peripheral enhancement pattern and a rapid centripetal progression was detected in all neoplastic lesions. These data were not confirmed by a 2002 study by Argalia et al. (27) that evaluated the time-intensity curves in 61 patients with solitary cold thyroid nodules before surgical removal. The power Doppler vascular pattern before and after a 60-second intravenous injection of Levovist, and the timeintensity curves, were assessed during a 5-minute period. At contrast-enhanced power Doppler examination, 83% of the malignant nodules and 91% of the benign lesions were found to be hypervascularized. All the nodules showed similar rapid wash-in curves with no difference in the time of appearance of contrast enhancement. By contrast, the wash-out curves were regular and monophasic in most (93%) of the benign nodules, while they appeared irregular and polyphasic in most (89%) of the malignant lesions. Bartolotta et al. (28) recently assessed the potential of CEUS for characterizing solitary thyroid nodules on gray-scale images. A series of 18 patients with solitary thyroid nodules underwent pulse inversion US at low mechanical index, after an intravenous injection of a 2.4 ml bolus of a second-generation contrast agent (SonoVue, Bracco, Italy). Signal intensity values on gray-scale images were registered at baseline, 30, 60, and 120 seconds. Following administration of SonoVue, benign nodules showed a diffuse pattern of contrast enhancement, either homogeneous or inhomogeneous, while malignant nodules showed in most cases (8/13, 61.5%) an absent or faint dotted pattern of contrast enhancement. Unfortunately, irrespective of their histological diagnosis, all nodules with detectable intranodular vascular signals at the baseline power Doppler evaluation revealed a diffuse contrast enhancement after contrast injection, while all nodules with a perinodular vascular pattern revealed an absent or faint contrast enhancement at CEUS. Thus, CEUS did not change the precontrast diagnostic information based on the results of gray scale and power Doppler evaluation.
162
E. PAPINI ET AL.
We tested the efficacy of a second-generation contrast agent in the differential diagnosis of thyroid lesions in a series of 28 patients with thyroid nodules scheduled for surgery (unpublished data). After a preliminary assessment with high resolution US and power Doppler, the US evaluation was repeated after an intravenous bolus injection of 2.4 ml of sulphur hexafluoride microbubbles. The wash-in and time to peak curves, calculated on the entire nodular area, showed overlapping values between the 8 malignant and the 20 benign nodules. In malignant lesions, the washout curves were characterized by an earlier disappearance of the contrast signals. However, 25% of benign thyroid nodules presented an overlap with malignant lesions, and the difference in the time of disappearance did not reach statistical significance (Fig. 10.8). On the basis of these findings, the study of time-intensity curves seems not to enable a clear differentiation between benign and malignant lesions. The major limit of
FIG. 10.8. B-mode US images recorded in different phases after contrast injection. The lesion on the right with a hypoechoic core and a hypervascular ring (small arrow) is a papillary carcinoma; while the lesion on the left (large arrow), with a central fluid collection is a benign colloid nodule. On closer analysis, the malignant lesion showed a hypervascular ring with ill-defined margins and an early enhancement. Note the central hypovascular portion, corresponding to a necrotic area of the tumor at histological evaluation
CONTRAST-ENHANCED ULTRASOUND IN THE MANAGEMENT
163
the previously described reports was probably due to the examination as a whole of the area of the nodule under investigation. In fact, in our experience, the vascular pattern of thyroid nodules is usually inhomogeneous and changes widely within the same nodule from area to area. Thus, quantitative information of the mean perfusion pattern adds little to the traditional US and Doppler features for the identification of thyroid nodules to be submitted to fine needle aspiration biopsy. On the basis of these findings, we performed a feasibility study on thyroid nodules with new specifically designed software (Qontrast, Esaote, Genoa, Italy), after the injection of a second-generation contrast agent. The aim of the study was to improve the differential diagnosis of complex thyroid lesions by means of a 3D imaging of the perfusion parameters (parametric maps). In all cases, a diagrammatic correlation between the B-mode images after contrast injection and the parametric maps representing PEAK, RBF, RBV, and TTP was obtained (Fig. 10.9). Colors used for the construction of 3D images went from black (zero value) to red (highest value), passing through
FIG. 10.9. Spatial correlation between a B-mode image registered after contrast injection and a 3D diagram representing the RBF. Note that the central portion of the two nodules is characterized by different RBF indexes. In the benign nodule on the left, the center of the lesion is visualized as a blue area corresponding to an RBF near to zero, while in the malignant nodule on the right the partially necrotic area is represented with an intermediate (yellow) color
164
E. PAPINI ET AL.
the blue and yellow colors. Thus, on a parametric map representing RBF, a colloid collection with an RBF near to zero was shown as a black area, while a partially necrotic or hypovascularized tissue was represented with an intermediate yellow color. Vascularized thyroid lesions with an elevated RBF index were visualized as red areas. A blinded US study on 10 hypofunctioning thyroid nodules with benign, indeterminate, or malignant cytological findings was performed before their surgical removal (unpublished data). CEUS demonstrated a high RBF in the well vascularized areas of all thyroid lesions, irrespective of their malignant (four cases of papillary carcinoma and one case of follicular carcinoma) or benign (two cases of follicular adenoma and three cases of benign hyperplastic nodule) final histological diagnosis. Thus, contrast-enhanced evaluation of the whole area of the nodule was confirmed not to add any diagnostic information to the precontrast vascular diagnosis. However, the assessment of the parametric maps showed a few potentially relevant findings in complex thyroid lesions. In benign thyroid nodules, the areas surrounding a fluid collection showed a hypervascular pattern characterized by thin and regular peripheral signals. By contrast, in malignant lesions, a more complex morphology was detected with thick, irregular, and asymmetric peripheral vascular signals. Finally, a low grade but well detectable and homogeneous RBF was detected in papillary microcarcinomas, characterized by a nearly avascular central area at color or power Doppler evaluation (Fig. 10.9). These sonographic findings corresponded to areas of partial tissue necrosis or fibrosis at histology. By contrast, three cases of benign nodules with a central area devoid of vascular signals due to the presence of a colloid collection demonstrated a complete absence of RBF in their central zone. In this respect, 3D CEUS was more effective than power Doppler in differentiating fluid areas in benign lesions, which are completely devoid of blood supply, from necrotic or degenerative zones within malignant lesions, which appear characterized by a low and irregular but well detectable regional blood flow. In conclusion, the 3-D CEUS elaboration did not provide a clear-cut differentiation between benign and malignant lesions. CEUS added some information to the precontrast power Doppler evaluation, demonstrating the presence of a marked heterogeneity of blood flow within neoplastic lesions when compared with benign thyroid nodules, and differentiating the partially necrotic but still viable tumor areas from fluid collections.
CONTRAST-ENHANCED ULTRASOUND IN THE MANAGEMENT
165
CONTRAST-ENHANCED ULTRASONOGRAPHY IN THE ASSESSMENT OF THE EFFICACY OF THERMAL ABLATION TREATMENTS US evaluation and FNA, selecting thyroid nodules with a low risk of malignancy, have reduced the rate of thyroid surgery for nonfunctioning lesions (6–8). A high number of nonfunctioning nodules are managed with a regular long term control, but in cases of progressive growth of the nodule, surgery is recommended. Surgery is an expensive therapeutic procedure that is not devoid of side effects, including subclinical hypothyroidism and cosmetic damage. Thus, percutaneous thermal ablation with laser (PLA) or radiofrequency (RFA) has been proposed as an office-based procedure for a reduction in size of enlarging benign lesions (30, 38). Two RCT showed that PLA treatment results in a satisfactory clinical response in the majority of patients with a benign solid cold thyroid nodule, with a median reduction in nodule volume of over 40% and a significant decrease of pressure symptoms (33–34). The major limit in the clinical use of PLA is the lack of real-time monitoring of the extent of damage induced by thermal ablation. At the end of laser illumination, the treated zone appears as an irregular and ill-defined hyperechoic area. After 24 hours, PLA-induced thyroid damage is characterized by specific US features (29), with a small central hypoechoic area (zone of vaporization), surrounded by a hyperechoic rim (zone of carbonization) and by an outer hypoechoic zone with ill-defined margins (zone of coagulative necrosis) (17). Color or power Doppler evaluation performed 6 hours after PLA treatment shows a loss of vascularization of the treated zone, but fails to provide an accurate definition of the area of necrosis (30) (Figure 10.10). The exact definition of the area of thermalinduced necrosis is critical for the mini-invasive debulking of cervical recurrences or distant metastases of thyroid cancer, or endocrine tumors that are not amenable to surgical or radioiodine treatments (35–37, 44). These malignant lesions are usually close to, encase, or invade vital structures in neck, liver, or bones. Thus, the greatest extent of debulking should be achieved with the least risk of a major damage to the contiguous nervous or vascular structures. Currently, PLA treatment may be performed in real time under MR guidance using fast gradient echo sequences. Laser application should be blocked when MR images demonstrate the complete coagulation of the lesion. Contrast-enhanced T1-weighted sequences are subsequently performed to visualize the exact margins of the volume of thermal ablation. The clinical advantages of MR monitoring
166
E. PAPINI ET AL.
FIG. 10.10. (A) B-mode US and (B) power Doppler images of a large benign nodule registered 24 hours after a PLA treatment. The ablated area is in some way detectable on a B-Mode image, but no definite information can be derived about the viability of the treated tissue on the basis of the gray-scale pattern. Relevant information about vascularity is added by color Doppler examination, but the persistence of a slow flow parenchymal vascularity is not assessable. A further limit of color Doppler examination is its sensitivity to movement artifacts
for achieving a near complete destruction of a malignant lesion are relevant, but this imaging technique is cumbersome and of difficult access, while US procedures are by far less expensive and are more widely available. On the basis of these considerations, we tested the usefulness of CEUS for obtaining an early and accurate definition of the volume of ablated tissue in 10 consecutive patients with benign or malignant conditions. The volume of coagulative necrosis was initially assessed six hours after the procedure with US, power Doppler, and CEUS examination. The results were compared with the volume of ablated tissue demonstrated by means of a 24-hour contrast-enhanced CT (unpublished data). In all cases, the assessment of the volume of thermal necrosis and the definition of the limits of the area of tissue damage were more accurate and clearly depicted by the CT scan than by the sonographic techniques (Fig. 10.11). However, CEUS was more effective than B-mode US and power Doppler examination in the early detection, and in the definition of the margins of the ablated area that appeared completely devoid of blood supply and well demarcated from the surrounding viable tissue (Fig. 10.12). CONCLUSIONS Color and power Doppler provide relevant information on thyroid blood flow direction and velocity but, on state of the art Doppler, no data are available on the slowly flowing blood
FIG. 10.11. CT scan of a 80-year old woman with an unresectable cervical recurrence of a poorly differentiated thyroid cancer. No radioiodine uptake was detected at a post-therapeutic dose total body scan with 131I. The patient was treated with PLA to induce a preliminary debulking of the lesion before treatment with external radiotherapy. (A) After the first PLA treatment, a large area of necrosis was demonstrated in the central part of the tumor recurrence (arrowhead). Viable areas of thyroid cancer were still present in the retrotracheal space and in the superficial layers of the neck (arrows). Note the encasement of the right carotid and vertebral arteries. (B) After the second PLA treatment, a large part of the tumor mass underwent coagulative necrosis (white arrowheads). Note the gas bubbles within the necrotic tissue (arrow). Still viable tissue is present in the retrotracheal space (black arrowheads). No PLA treatment was performed on this part of the tumor because of its close proximity to the airway and the oesophagus, putting thermal ablation at risk of causing major complications
FIG. 10.12. The same patient as in Fig. 10.11: dynamic contrastenhancement US performed after the second laser ablation. (A) In the early phase (8 seconds after contrast injection), the US image shows the arrival of the contrast agent in the carotid artery (arrow), and the presence of gas bubbles (short arrows) within the treated mass. (B) The image recorded 27 seconds after the injection shows an enhancing halo in the peripheral ring of still viable tumor tissue (arrows) that encloses a zone of nonenhancing necrosis. The extent and margins of the area of necrosis are quite similar in the CEUS and the CT images
168
E. PAPINI ET AL.
at the capillary level. Contrast-enhanced ultrasound (CEUS) may supply, for the first time, information on tissue perfusion, and may play a role in the characterization of lesions at risk of malignancy in solid viscera, with results similar to the images provided by contrast-enhanced CT and MR scans. First- and second-generation contrast agents seem to provide only ancillary data for the diagnosis of malignant nodules. The variation of time-intensity curves during the transit time of the injected echo-enhancers offers a modest and uncertain improvement over the prediction of malignancy obtainable with traditional color Doppler or power Doppler findings. The evaluation of signal intensity changes on gray-scale images of the thyroid gland after contrast injection is a feasible technique, but the overlapping findings in the characterization of thyroid nodules, together with the cost and time expenditure of the procedure, seem to limit its usefulness in clinical practice. Preliminary data on second-generation contrast media demonstrate their effectiveness in obtaining early and welldefined evidence of the volume of thyroid tissue destruction induced by mini-invasive thermal ablation procedures on benign and malignant thyroid lesions. CEUS is more effective than B-mode US and power Doppler examination in the early detection of the margins of the ablated area of malignant tissue, that appear completely devoid of blood supply and well demarcated from the surrounding viable tissue. A feasibility study with a 3D CEUS representation obtained with specifically designed software did not provide a clearcut differentiation between benign and malignant lesions. Preliminary results suggest that 3D CEUS may add some information to the precontrast color and power Doppler evaluation of vascular signals in thyroid nodules, demonstrating the presence of a marked heterogeneity of regional blood flow within the neoplastic lesions when compared with benign thyroid nodules. However, further study is required to evaluate the impact of this new US procedure on the management of thyroid nodules. US contrast agents are easy to use, have a very low incidence of adverse events, are unaffected by renal function, and add no radiation for their use. Using CEUS results in an improved depiction of the vascular pattern of thyroid nodules, but its usefulness in the management of thyroid nodules is currently restricted to the assessment of the efficacy of ablation procedures. New specifically designed microbubbles and new models of US equipment with specific software are needed to
CONTRAST-ENHANCED ULTRASOUND IN THE MANAGEMENT
169
improve the malignancy-predictive value of contrast-enhanced sonography for the management of thyroid lesions. References 1. Tan GH, Gharib H (1997) Thyroid incidentalomas: management approaches to nonpalpable nodules discovered incidentally on thyroid imaging. Ann Intern Med 126:226–31 2. Ezzat S, Sarti DA, Cain DR, et al (1994) Thyroid incidentalomas: prevalence by palpation and ultrasonography. Arch Intern Med 154:1838–40 3. Hegedus L, Bonnema SJ, Bennedbaek FN (2003) Management of simple nodular goiter: current status and future perspectives. Endocrine Rev 24:102–132 4. Filetti S, Durante C, Torlontano M (2006) Nonsurgical approaches to the management of thyroid nodules. Nat Clin Pract Endocrinol Metab 2:384–94 5. Belfiore A, Giuffrida D, La Rosa GL, et al (1989) High frequency of cancer in cold thyroid nodules occurring at young age. Acta Endocrinol (Copenh) 121:197–202 6. AACE/AME Task Force on Thyroid Nodules, American Association of Clinical Endocrinologists, and Associazione Medici Endocrinologi (2006) Medical guidelines for clinical practice for the diagnosis and management of thyroid nodules. Endocr Pract 12:63–102 7. Cooper DS, Doherty GM, Haugen BR, et al (2006) The American Thyroid Association Guidelines Taskforce. Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 16:109–142 8. Pacini F, Schlumberger M, Dralle H et al (2006) European consensus for the management of patients with differentiated thyroid carcinoma of the follicular epithelium. Eur J Endocrinol 154: 787–803 9. Marqusee E, Benson CB, Frates MC, et al (2000) Usefulness of ultrasonography in the management of nodular thyroid disease. Ann Intern Med 133:696–700 10. Papini E, Guglielmi R, Bianchini A, et al (2002) Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clin Endocrinol Metab 87:1941–6 11. Mandel SJ (2004) Diagnostic use of ultrasonography in patients with nodular thyroid disease. Endocr Pract 10:246–52 12. Frates MC, Benson CB, Charboneau JW, et al (2005) Society of Radiologists in Ultrasound. Management of thyroid nodules detected at US: Society of Radiologists in Ultrasound consensus conference statement. Radiology 237:794–800 13. Hamberger B, Gharib H, Melton LJ III, et al (1982) Fine-needle aspiration biopsy of thyroid nodules: impact on thyroid practice and cost of care. Am J Med 73:381–4
170
E. PAPINI ET AL.
14. Goellner JR, Gharib H, Grant CS, et al (1987) Fine needle aspiration cytology of the thyroid, 1980 to 1986. Acta Cytol 31:587–90 15. Gharib H, Goellner JR (1993) Fine-needle aspiration biopsy of the thyroid: an appraisal. Ann Intern Med 118:282–9 16. Schlinkert RT, van Heerden JA, Goellner JR, et al (1997) Factors that predict malignant thyroid lesions when fine-needle aspiration is “suspicious for follicular neoplasm.” Mayo Clin Proc 72:913–6 17. Rago T, Di Coscio G, Basolo F et al (2007) Combined clinical, thyroid ultrasound and cytological features help to predict thyroid malignancy in follicular and Hurthle cell thyroid lesions: results from a series of 505 consecutive patients. Clin Endocrinol (Oxf) 66:13–20 18. MacDonald L, Yazdi HM (1996) Nondiagnostic fine needle aspiration biopsy of the thyroid gland: a diagnostic dilemma. Acta Cytol 40:423–8 19. McHenry CR, Walfish PG, Rosen IB (1993) Non-diagnostic fine needle aspiration biopsy: a dilemma in management of nodular thyroid disease. Am Surg 59:415–9 20. Foster FS, Burns PN, Simpson DH, Wilson SR, Christopher DA, Goertz DE (2000) Ultrasound for the visualization and quantification of tumor microcirculation. Cancer Metastasis Rev. 19:131–8 21. Wilson SR, Burns PN (2006) Microbubble contrast for radiological imaging: 2. Applications. Ultrasound Q 22:15–8 22. Burns PN, Wilson SR (2000) Simpson DH et al. Pulse inversion imaging of liver blood flow: improved method for characterizing focal masses with microbubble contrast. Invest Radiol 35:58–71 23. Burns PN, Wilson SR (2007) Focal liver masses: enhancement patterns on contrast-enhanced images: concordance of US scans with CT scans and MR images. Radiology 242:162–74 24. Huang Wei C, Bleuzen A, Bourlier P, et al (2006) Differential diagnosis of focal nodular hyperplasia with quantitative parametric analysis in contrast-enhanced sonography. Invest Radiol 41:353–368 25. Spiezia S, Farina R, Cerbone G et al (2001) Analysis of color Doppler signal intensity variation after levovist injection: a new approach to the diagnosis of thyroid nodules. J Ultrasound Med 20:223–231 26. Appetecchia M. Bacaro D, Brigida R, Milardi D, Bianchi A, Solivetti F (2006) Second generation ultrasonographic contrast agents in the diagnosis of neoplastic thyroid nodules. J Exp Clin Cancer Res 25:325–30 27. Argalia G, De Bernardis S, Mariani D et al (2002) Ultrasonographic contrast agent: evaluation of time intensity curves in the characterisation of solitary thyroid nodules. Radiol Med 103: 407–413 28. Bartolotta TV, Midiri M, Galia M, et al (2006) Qualitative and quantitative evaluation of solitary thyroid nodules with contrastenhanced ultrasound: initial results. Eur Radiol 16:2234–2241
CONTRAST-ENHANCED ULTRASOUND IN THE MANAGEMENT
171
29. Pacella CM, Rossi Z, Bizzarri G, et al (1993) Ultrasound-guided percutaneous laser ablation of liver tissue in a rabbit model. Eur Radiol 3:26–32 30. Pacella CM, Bizzarri G, Guglielmi R, et al (2000) Thyroid tissue: US-guided percutaneous interstitial laser ablation – A feasibility study. Radiology 217:673–677 31. Dossing H, Bennedbaek FN, Karstrup S, Hegedus L (2002) Benign solitary solid cold thyroid nodules: US-guided interstitial laser photocoagulation – Initial experience. Radiology 225:53–57 32. Pacella CM, Bizzarri G, Spiezia S, et al (2004) Thyroid tissue: US-guided percutaneous laser thermal ablation. Radiology 232:272–80 33. Dossing H, Bennedbaek FN, Hegedus L (2005) Effect of ultrasound guided interstitial laser photocoagulation on benign solitary cold thyroid nodules – a randomised study. Eur J Endocrinol 152:341–5 34. Papini E, Guglielmi R, Bizzarri G et al (2007) Treatment of benign cold thyroid nodules: a randomized clinical trial of percutaneous laser ablation versus levothyroxine therapy or follow-up. Thyroid 17:229–35 35. Guglielmi R, Pacella CM, Dottorini ME, et al (1999) Severe thyrotoxicosis due to hyperfunctioning liver metastasis from follicular carcinoma: treatment with 131I and Interstitial laser ablation. Thyroid 9:173–177 36. Pacella CM, Stasi R, Bizzarri G, et al (2007) Percutaneous laser ablation of unresectable primary and metastatic adrenocortical carcinoma. Eur J Radiol. May 9; [Epub ahead of print] 37. Wood BJ, Abraham J, Hvizda JL, Alexander HR, Fojo T (2003) Radiofrequency ablation of adrenal tumors and adrenocortical carcinoma metastases. Cancer 97:554–60 38. Kim YS, Rhim H, Tae K, Park DW, Kim ST (2006) Radiofrequency ablation of benign cold thyroid nodules: initial clinical experience. Thyroid 16:361–7 39. Taylor KJW, Burns PN, Wells PNT (1987) Clinical Applications of Doppler Ultrasound. New York: Raven Press 40. Wachsberg HR (2007) B-Flow imaging of the hepatic vasculature: correlation with color Doppler sonography. AJR 188:522–533 41. Terrier F, Grossholz M, Becker CD (1999) Spiral CT of the abdomen. New York: Springer 42. Matsuda Y, Yabuuchi I (1986) Hepatic tumors: US contrast enhancement with CO2 microbubbles. Radiology 161, 701–705 43. Chin CT, Burns PN (2000) Predicting the acoustic response of a microbubble population for contrast imaging in medical ultrasound. Ultrasound Med Biol 26:1293–300 44. Mazzaglia PJ, Berber E, Milas M, Siperstein AE (2007) Laparoscopic radiofrequency ablation of neuroendocrine liver metastases: a 10-year experience evaluating predictors of survival. Surgery 142:10–9
CHAPTER 11
Percutaneous Ethanol Injection (PEI): Thyroid Cysts and Other Neck Lesions Andrea Frasoldati and Roberto Valcavi
BACKGROUND Ethanol injection causes irreversible tissue damage through cellular dehydration, protein denaturation, coagulative necrosis, and small vessel thrombosis leading to hemorrhagic infarct and reactive fibrosis. The sclerosing properties of ethanol have prompted its use in the treatment of various malignant or benign lesions such us hepatocellular carcinoma, adrenal adenoma, parathyroid adenoma, or hyperplasia (1,2,3,4,5). Percutaneous ethanol injection (PEI) for the treatment of thyroid lesions was introduced into clinical practice in 1990 (6). This technique was initially proposed as an alternative to surgery and radioiodine administration in the management of autonomous functioning nodules. While the use of PEI for this purpose has sensibly decreased, ethanol sclerosing properties have been successfully applied in the treatment of thyroid cystic lesions. The present chapter will deal with PEI applications to thyroid lesions, focusing on the treatment of thyroid cysts. PEI therapy of other neck lesions (e.g., parathyroid, lymph nodes) will also be briefly addressed. PEI OF THYROID LESIONS Autonomous Functioning Thyroid Nodules The cumulative evidence collected in the last fifteen years indicates that PEI treatment of hyperfunctioning nodules is followed by restoration of normal thyroid function in the majority (90%) of patients with subclinical hyperthyroidism, Endocrinology Unit, Arcispedale Santa Maria Nuova, Reggio Emilia, Italy
173
174
A. FRASOLDATI AND R. VALCAVI
and in a large proportion (50–75%) of patients with thyrotoxicosis (6,7,8,9,10). However, it should be underlined that, in spite of encouraging short-term results, disappearance of the hot nodule is obtained in a minority (4.2–11.1%) of patients (11,12). This discrepancy suggests that PEI therapy may not eradicate hot nodules, even though there is transient normalization of thyroid hormone levels. In line with this hypothesis, a 20.8% rate of re-expansion and recurrence of thyroid hot nodules after PEI has been reported (12). The need for multiple (2–10) treatments, the risk of side effects, and the consistent rate of unsatisfactory long-term results, have lead to questioning the cost-benefit ratio of PEI in the management of hot thyroid nodules. According to most studies, the rate of successful treatments is correlated to the initial nodular volume (13): in a large Italian multicenter study, a 90–100% success rate was reported in small (≤15 ml) nodules, whereas the treatment showed far lower efficacy when applied to larger (≥ 30 ml) nodules (10). A promising new approach in large toxic nodules could be provided by the combination of PEI plus radioiodine administration (14); according to this strategy, the goal of PEI is to obtain a shrunken nodule, thus allowing the use of lower doses of 131I. Cold Solid Thyroid Nodules PEI has been used in the treatment of solid cold nodules with controversial results. The reported mean volume reduction is 38–47%, with relief of compressive symptoms experienced by up to 56% of patients (15,16,17). According to the results of a randomized trial, PEI offered far better results than l-thyroxine administration (15). This promising line of evidence, however, must be weighted against some consistent concerns: a) potentially harmful side effects (see below); b) induction of perinodular fibrosis causing a major obstacle to subsequent surgery; and c) unrecognized occurrence of cancer due to limitations and pitfalls of thyroid cytology (18). Thyroid Cysts Cystic thyroid lesions with a predominant fluid component (Figs. 11.1–11.3) are a frequent finding, representing up to 32% of all thyroid nodules studied by the ultrasound (US)(19). In most cases, the cystic portion is caused by hemorrhage and necrotic changes. Pure thyroid cysts are less frequent, corresponding to about 1% of thyroid nodules (20). The majority of these lesions are asymptomatic, yet due to their size and/or
THYROID CYSTS AND OTHER NECK LESIONS
175
FIG. 11.1. Large thyroid cyst located in the left lobe (upper panel). Results of PEI at 1-month (middle panel) and 1-year (lower panel) control ultrasound. The volume of the lesion decreased from 17.5 ml to less than 0.1 ml. A single PEI session was performed (6 ml of ethanol injected); no side effects were observed
176
A. FRASOLDATI AND R. VALCAVI
FIG. 11.2. Thyroid cyst in the left lobe (12 × 12.4 × 20 mm = 1.5 ml volume) (upper panel). Results of PEI at 1-month (middle panel) and 1-year (lower panel) control ultrasound. The final result of the procedure was a minute scar in the thyroid tissue. A single PEI session was performed (1.4 ml of ethanol injected); no side effects were observed
THYROID CYSTS AND OTHER NECK LESIONS
177
FIG. 11.3. Large thyroid cyst located in the right lobe (upper panel). Results of PEI at 1-month (middle panel) and 1-year (lower panel) control ultrasound. The volume of the lesion decreased from 7.3 to 0.1 ml. A single PEI session was performed (4 ml of ethanol injected); no side effects were observed
178
A. FRASOLDATI AND R. VALCAVI
location, they may produce compressive and/or cosmetic discomfort. Fine-needle aspiration (FNA) of the fluid content may reduce the lesion size, but this effect is usually transient, and in most cases (up to 80%) thyroid cysts recur after FNA (21,22). In the last few years, PEI has been widely used in the management of thyroid cysts (23,24,25,26). In our series of patients with empty body cystic nodules, PEI achieved an 88.8% median volume reduction, while in mixed nodules (Fig.11.4), a 65.8% median volume reduction was recorded (26). Compressive and cosmetic symptoms respectively disappeared in 74.8% and 80.0% of patients treated with PEI, whereas similar results were obtained in 24.4% and 37.4% of patients treated with simple evacuation (26).
FIG. 11.4. Large mixed thyroid nodule before (upper panel) and two months after (lower panel) a PEI procedure. A 5 ml volume of ethanol was injected. The volume of the lesion showed a 70% decrease (from 8.9 to 2.7 ml). No side effects were noted apart from a transient pain irradiated to the ears arising immediately after the needle extraction.
THYROID CYSTS AND OTHER NECK LESIONS
179
PEI therapy is more effective in reducing size of thyroid cystic nodules in comparison to solid nodules. In a recent controlled study, the volume reduction rate of thyroid cysts (64%) was greater than that achieved in benign solid nodules (35%) (17). PEI success rate, defined as a > 50% reduction of thyroid cyst size, varies from 72 to 95% (27). More importantly, the recurrence rate of thyroid cysts is significantly reduced (< 20%) after PEI therapy (27). Therefore, PEI is currently regarded as the first-line and definitive treatment for cystic lesions causing local discomfort and/or cosmetic concern to patients. PEI OF THYROID CYSTS: TECHNICAL ASPECTS Disposable echoic Chiba needles (18–22 gauge) are inserted through a sterile needle pointing device, and connected to a 20 ml syringe held by a Cameco® pistol (The Business Village, London, U.K.) (Fig. 11.5). Sterility of the maneuver is ensured by a sterile cover for the probe, sterile gel, and sterile gloves for operators. After complete fluid extraction,
FIG. 11.5. The set of materials needed for performing PEI procedure. All the materials should be prepared in advance and be readily available during the maneuver. This way, PEI can be easily performed by two skilled operators (e.g., one physician plus one nurse)
180
A. FRASOLDATI AND R. VALCAVI
95% sterile ethanol is immediately injected into the cyst. Ethanol is contained in 10 ml syringes connected to the needle by 20–25 cm. pipes—allowing precise and gentle repositioning of the needle during the maneuver. (Fig. 11.6) The infused ethanol can be seen on US as a hyperechoic material progressively refilling the cyst (Fig.11.7). The operator should carefully monitor the correct positioning of the needle tip during ethanol infusion, avoiding major trauma to the cyst wall. The amount of ethanol injected is usually equal to 50–70 % of the fluid extracted. Before needle extraction, the pipeline and the needle are rinsed with normal saline to minimize ethanol reflux into subcutaneous tissues, thus preventing a potential cause of pain to the patient. In the procedure adopted at our hospital, the injected ethanol
FIG. 11.6. PEI procedure in a 35-year old male patient: view of the operating field. The 20 ml ethanol-containing syringe is connected to the pipe, and ethanol is injected into the lesion (upper panel). At the end of the maneuver, the pipe is clamped, the ethanol-containing 20 ml syringe is disconnected, and a 5 ml saline-containing syringe is connected to the line (lower panel). The saline rinse prevents ethanol leakage through the superficial tissues—a potential cause of intense pain to the patient
THYROID CYSTS AND OTHER NECK LESIONS
181
Fig. 11.7. (Continued)
is not evacuated and is left within the lesion. Some authors claim that complete evacuation of infused ethanol avoids the risk of ethanol leakage and any related complications. Available data comparing the two different options, however, apparently do not show any differences in term of successful results (28).
182
A. FRASOLDATI AND R. VALCAVI
Fig. 11.7.
(Continued)
SIDE EFFECTS OF PEI Relevant side effects of PEI are rare. The most frequently reported complaint (10–20%) refers to pain at the injection site. Pain, which may irradiate to the ears and jaw, is often quite intense but self-limiting—usually lasting for a few seconds. In a minority of patients, pain may be followed by neck tenderness
THYROID CYSTS AND OTHER NECK LESIONS
183
FIG. 11.7. (a) PEI procedure in a large (44 × 32 × 52 mm) empty body thyroid cyst. Phase 1: Fluid extraction followed by disappearance of the cystic lesion (b) Phase 2: Ethanol injection and refilling of the cystic lesion. Ethanol spreading within the cyst walls is immediately visualized as the popping-up of hyperechoic inhomogeneous material. Shortly afterwards, the cyst reassumes its typical anechoic appearance (c) PEI procedure results at a 6-months control ultrasound examination. The lesion volume showed a 90% decrease (from 38.8 to 3.8 ml). A single PEI session was performed. The amount of the extracted fluid was 32 ml, while 16 ml of ethanol were injected. The entire procedure lasted 10 min. No side effects were noted
that generally recedes within some hours. In most cases, pain is due to leakage of ethanol into subcutaneous tissue during needle extraction—therefore, it can be prevented by rinsing the needle tip with a small amount of saline before extracting the needle. Alternatively, local anesthesia can be used to minimize the local discomfort that may occur during and/or immediately after PEI. The most serious complication of PEI is unilateral cord paralysis—usually transient—due to the toxic action of
184
A. FRASOLDATI AND R. VALCAVI
absolute ethanol on the recurrent laryngeal nerve. In a series of autonomous hyperfunctioning nodules treated with PEI, unilateral cord palsy has been reported in up to 3.9% of patients (10). Occasionally, other severe complications—such as permanent ipsilateral facial dysesthesia or jugular vein thrombosis—have also been described (10,16). Relevant side effects are far less frequent, if not anecdotic, in PEI therapy of thyroid cystic lesions (13). The presence of a fibrous capsule surrounding the cyst allows direct control on ethanol infusion and acts as a barrier against its uncontrolled spreading. This makes the PEI procedure of cystic nodules safer compared to solid nodules. In addition, PEI of cystic lesions usually requires one to two sessions, while PEI of autonomous functioning nodules is a more complex procedure—demanding many sessions—therefore, the risk of side effects is proportionally increased. PEI OF PARATHYROID CYSTS AND OTHER NECK LESIONS Parathyroid Cysts Parathyroid (PT) cysts (29) are commonly regarded as rare lesions of the neck and superior mediastinum (less than 300 cases reported in the literature). However, their occurrence is probably underestimated, as PT cysts are often mistaken for thyroid cysts (30). PT cysts may have different origins. Only in a minority of cases do they represent functioning lesions—seemingly pseudocystic changes of parathyroid adenomas—whereas, most PT cysts are non-functional derivates from embryological remnants (31). Furthermore, microcystic changes are frequently (40–50%) detected at autopsy in otherwise normal parathyroid glands, and it has been suggested that some PT cysts may arise from a gradual retention of secretions (31). Ultrasound-guided, fine-needle aspiration and parathyroid hormone (PTH) assay in the fluid within the cyst (32,33) are the main tools for diagnosis of PT cysts. The finding of a water-clear fluid in a neck cyst strongly suggests a PT origin and should prompt measurement of PTH in the fluid sample, as intra-cystic PTH levels are usually elevated many times above serum PTH levels (32,33). As with thyroid cysts, PT cysts tend to recur after fine-needle aspiration—therefore, injection of ethanol or other sclerosing agents (e.g., tetracycline) has been proposed as an alternative to surgery (34,35,36,37). In line with reports from the literature, while neck pain and recurrent nerve palsy have been reported after tetracycline injection (37), ethanol injection therapy of PT cysts has proved in our experience to be safe and effective (75.0% mean volume
THYROID CYSTS AND OTHER NECK LESIONS
185
reduction; 84.6% cure rate in a series of 13 PT cysts, unpublished data) (Fig.11.8). Larger series are necessary to more accurately define the pros and cons of PEI of PT cysts. Other Neck Lesions In case of thyroglossal duct cysts (Fig. 11.9), PEI effectiveness seems lower in comparison to the results obtained with thyroid and parathyroid cysts (38,39). Besides, a higher risk of infectious
FIG. 11.8. Parathyroid (PT) cyst located at the inferior pole of thyroid right lobe before (upper panel) and after (lower panel) PEI procedure. PT nature of the lesion was suggested by the water-clear appearance of the fluid during diagnostic fine-needle aspiration. As expected, PTH levels in the needle wash-out were exceedingly high (> 1600 pg/ml). The PT cyst was nonfunctional (serum PTH levels = 23 pg/ml; serum calcium levels 9.5 mg/dl). After PEI, the volume of the lesion showed a 90% decrease (from 2.6 ml to 0.25 ml). No side effects were observed
186
A. FRASOLDATI AND R. VALCAVI
FIG. 11.9. Thyroglossal duct cyst before (upper panel) and after (lower panel) a single session of PEI procedure (1.5 ml of ethanol injected). The cyst volume decreased from 1.8 to 0.05 ml. No side effects were observed
complications is to be taken into account in treating these lesions by PEI. For these reasons, PEI cannot be considered the first-line treatment for thyroglossal duct cysts, and should rather be viewed as a therapeutic option in patients not amenable to surgery. PEI has been used for treating parathyroid adenomas in high surgical risk patients with primary hyperparathyroidism with partial success (40,41). Furthermore, in patients with chronic renal failure suffering from secondary or tertiary hyperparathyroidism, PEI of hyperplastic PT glands has gained the role of an established and powerful adjunct to medical therapy (42,43). In patients with ultrasound evidence of neck recurrences from papillary thyroid carcinoma, ethanol injection has been
THYROID CYSTS AND OTHER NECK LESIONS
187
proposed as a conservative alternative treatment to radio-iodine therapy (often unsuccessful) and to surgical neck exploration (sometimes regarded as overly aggressive). According to two large published series (44,45). PEI treatment is effective in decreasing the volume of most metastatic lymph nodes, with 4.5% of treated lesions disappearing during the follow-up period and no relevant side effect observed. Therefore, in experienced hands, percutaneous ethanol injection seems a valuable option for patients who are not candidates for second surgery and/or radio-iodine therapy.
References 1. Bean WJ (1981) Renal cysts: treatment with alcohol. Radiology 138:329–331 2. Bean WJ, Rodan BA (1985) Hepatic cysts: treatment with alcohol. Am J Radiol 144:237–241 3. Solbiati L, Giangrande A, DePra L, et al (1985) Percutaneous ethanol injection of parathyroid tumors under US guidance: treatment for secondary hyperparathyroidism. Radiology 155:607–610 4. Livraghi T, Giorgio A, Mario G, et al (1985) Hepatocellular carcinoma and cirrhosis in 746 patients: long term results of percutaneous ethanol injection. Radiology 197:101–108 5. Rossi R, Savastano S, Tommasselli AP, et al (1995) Percutaneous computer tomography-guided ethanol injection in aldosteroneproducing adrenal adrenocortical adenoma. Eur J Endocrinol 132:302–305 6. Livraghi T, Paracchi A, Ferrari C, et al (1990) Treatment of autonomous thyroid nodule with percutaneous ethanol injection: preliminary results. Radiology 175:827–829 7. Martino E, Murtas MI, Liviselli A, et al (1992) Percutaneous intranodular ethanol injection for treatment of autonomously functioning thyroid nodules. Surgery 112:1161–1165 8. Papini E, Panunzi C, Pacella CM, et al (1993) Percutaneous ultrasound-guided ethanol injection: a new treatment of toxic autonomously functioning thyroid nodules? J Clin Endocrinol Metab 76:411–416 9. Livraghi T, Paracchi MA, Ferrari C, et al (1994) Treatment of autonomous thyroid nodules with percutaneous ethanol injection – a 4 year experience. Radiology 190:529–533 10. Lippi F, Ferrari C, Manetti L, et al (1996) Treatment of solitary autonomous thyroid nodules by percutaneous ethanol injection. Results of an Italian multicenter study. J Clin Endocrinol Metab 81:3261–3264 11. Monzani F, Caraccio N, Goletti O (1997) Five year follow-up of percutaneous ethanol injection for the treatment of hyperfunctioning thyroid nodules: a study of 117 patients. Clin Endocrinol 46:9–15
188
A. FRASOLDATI AND R. VALCAVI
12. Lee SJ, Ahn I-M (2005) Effectiveness of percutaneous ethanol injection therapy in benign nodular and cystic thyroid disease: long-term follow-up experience. Endocr J 52:455–462 13. Pacini F (2003) Role of percutaneous ethanol injection in management of nodular lesions of the thyroid gland. J Nucl Med 44:211–212 14. Zingrillo M, Modoni S, Conte M, et al (2003) Percutaneous ethanol injection plus radioiodine versus radioiodine alone in the treatment of large toxic thyroid nodules. J Nucl Med 44:207–210 15. Bennedbaek FN, Nielsen LK, Hegedus L (1998) Effect of percutaneous ethanol injection therapy versus suppressive doses of l-thyroxine on benign solitary solid cold nodules: a randomized trial. J Clin Endocrinol Metab 83: 830–835 16. Bennendbaek FN, Hegedus L (1999) Percutaneous ethanol injection therapy in benign solitary cold thyroid nodules: a randomized trial comparing one injection with three injections. Thyroid 9:225–233 17. Kim JH, Lee HK, Lee JH et al (2003) Efficacy of sonographically guided percutaneous ethanol injection for treatment of thyroid cysts versus solid thyroid nodules. AJR 180: 1623–1726 18. Alexander EK, Heering JP, Benson CB, et al (2002) Assessment of nondiagnostic ultrasound-guided fine needle aspirations of thyroid nodules. J Clin Endocrinol Metab 87:924–927 19. De Los Santos ET, Keyhani-Rofagha S, Cunningham JJ, et al (1990) Cystic thyroid nodules the dilemma of malignant lesions. Arch. Intern. Med 150:422–427 20. Sheppard MC, Franklyn JA (1994) Management of the single thyroid nodule. Clin Endocrinol 41:719–724 21. Clark OH, Okerlund MD, Cavalieri RR, et al (1979) Diagnosis and treatment of thyroid parathyroid and thyroglossal duct cysts. J Clin Endocrinol Metab 48:983–988 22. Jensen F, Rasmussen SN (1976) The treatment of thyroid cysts by ultrasonographically-guided fine needle aspiration. Acta Chir Scand 142:209–211 23. Monzani F, Lippi F, Goletti O, et al (1994) Percutaneous aspiration and ethanol sclerotherapy for thyroid cysts. J Clin Endocrinol Metab 78:800–802 24. Verde G, Papini E, Pacella CM, et al (1994) Ultrasound guided percutaneous ethanol injection in the treatment of cystic thyroid nodules. Clin Endocrinol 41:719–724 25. Zingrillo M, Torlontano M, Chiarella R, et al (1999) Percutaneous ethanol injection may be a definitive treatment for symptomatic thyroid cystic nodules not treatable by surgery: five-year follow-up study. Thyroid 9:763–767 26. Valcavi R, Frasoldati A (2004) Ultrasound-guided percutaneous ethanol injection therapy in thyroid cystic nodules. Endocr Pract 10:269–275 27. Bennedbaek FN, Hegedus L (2003) Treatment of recurrent thyroid cysts with ethanol: a randomized double blind controlled trial. J Clin Endocrinol Metab 88:5773–5777
THYROID CYSTS AND OTHER NECK LESIONS
189
28. Kim DW, Rho MH, Kim HJ et al (2005) Percutaneous ethanol injection for benign cystic thyroid nodules: is aspiration of ethanolmixed fluid advantageous? AJNR 26:2122–2127 29. Clark OH (1978) Parathyroid cysts. Am J Surg 35:395–402 30. Ujiki MB, Nayar R, Sturgeon C, et al (2007) Parathyroid cyst: often mistaken for a thyroid cyst. World J Surg 31:60–64 31. Ippolito G, Palazzo F, Sebag F, et al (2006) A single institution 25-year review of true parathyroid cysts. Lagenbecks Arch Surg 391:13–18 32. Silverman JF, Khazanie PG, Norris HT, et al (1986) Parathyroid hormone (PTH) assay of parathyroid cysts examined by fine-needle aspiration biopsy. Am J Clin Pathol 86:776–780 33. Pacini F, Antonelli A, Lari R, et al (1985) Unsuspected parathyroid cysts diagnosed by measurement of thyroglobulin and parathyroid hormone concentration in fluid aspirates. Ann Intern Med 102:793–794 34. Akel M, Salti I, Azar ST (1999) Successful treatment of parathyroid cyst using Ethanol Sclerotherapy. Am J Med Sci 317:50–52 35. Baskin HJ (2004) New applications of thyroid and parathyroid ultrasound. Min Endocrinol 29:195–206 36. Okamura K, Ikenoue H, Sato K, et al (1992) Sclerotherapy for benign parathyroid cysts. Am J Surg 163:344–345 37. Sanchez A, Carretto H (1993) Treatment of a nonfunctioning parathyroid cysts with tetracycline injection. Head Neck 15:263–265 38. Fukumoto K, Kojima T, Tomonari H, et al (1994) Ethanol injection sclerotherapy for Baker’s cysts, thyroglossal duct cysts, and branchial cleft cysts. Ann Plast Surg 33:615–619 39. Baskin HJ (2006) Percutaneous ethanol injection of thyroglossal duct cysts. Endocr Pract 12:355–357 40. Harman CR, Grant CS, Hay ID, et al (1998) Indications, technique and efficacy of alcohol injection of enlarged parathyroid glands in patients with primary hyperparathyroidism. Surgery 124:1011–1020 41. Cercueil JP, Jacob D, Verges B, et al (1998) Percutaneous ethanol injection into parathyroid adenomas: mid- and long-term results. Eur Radiol 8:1565–1569 42. Solbiati L, Giangrande A, Pra LD, et al (1985) Ultrasound-guided percutaneous fine-needle ethanol injection into parathyroid glands in secondary hyperparathyroidism. Radiology 155:607–610 43. Fugakawa M Kitaoga M Tominaka Y, et al (2003) Guidelines for percutaneous ethanol injection therapy of the parathyroid glands in chronic dialysis patients. Nephrol Dial Transplant 18:(Suppl 3):31–33 44. Lewis BD, Hay ID, Charboneau JW, et al (2002) Percutaneous ethanol injection for treatment of cervical lymph node metastases in patients with papillary thyroid carcinoma. AJR 178:699–704. 45. Lim CY, Yum JS, Lee J, et al (2007) Percutaneous ethanol injection therapy for locally recurrent papillary thyroid carcinoma. Thyroid 17:347–350
Chapter 12
Laser and Radiofrequency Ablation Procedures Roberto Valcavi, Angelo Bertani, Marialaura Pesenti, Laura Raifa Al Jandali Rifa’Y, Andrea Frasoldati, Debora Formisano, and Claudio M. Pacella
INTRODUCTION Several ultrasound-guided interventional procedures have been proposed to treat benign thyroid nodules without open surgery. The basic principle is to destroy thyroid nodular tissue by physical means. Percutaneous ethanol injection (PEI) was introduced in 1990 (1) and it has since proven to be very helpful in cystic benign thyroid nodules, whereas it is poorly effective in solid lesions (2,3) and is discussed in Chapter 11. HIFU (High Intensity Focused Ultrasound) is a promising new interventional US-guided technique for solid nodules (4), illustrated in Chapter 13. RADIOFREQUENCY Ultrasound-guided radiofrequency (RF) ablation is a method utilizing high frequency (3.8 to 4 MHz) radio wave energy to coagulate tissues. As RF energy is applied, frictional heating of tissues results, with cell death occurring at temperatures
Endocrinology Division, §Statistics & Clinical Epidemiology Unit, Arcispedale Santa Maria Nuova, Reggio Emilia, Italy *
Department of Radiology and Diagnostics Imaging, Ospedale Regina Apostolorum, Albano Laziale, Rome, Italy Correspondence: Dr. Roberto Valcavi Director, Endocrinology Unit Arcispedale Santa Maria Nuova 42100 Reggio Emilia, Italy
[email protected]
191
192
R. VALCAVI ET AL.
between 60 and 100° C. RF is receiving increased attention as an effective minimally invasive approach for the treatment of patients with a variety of primary and secondary malignant neoplasms; liver tumor ablation has been the subject of most published reports (5). Some authors reported that RF may be effective and safe in treating thyroid nodules (6). However, as RF is based on the use of large needle electrodes (G 14–18), multiple needles, or hook-needles (Fig. 12.1A, B, C), RF may be too invasive for the thyroid gland, a small and delicate organ anatomically adjacent to neck vital structures. Large breaches of thyroid capsule and parenchyma caused by outsized needles may increase the risk of bleeding with sudden neck swelling and potential airway obstruction. This complication has been reported even using Fine Needle (G 22–27) Aspiration (FNA) biopsy (7). We decided not to use this method. PERCUTANEOUS LASER ABLATION (PLA) LASER is an acronym of Light Amplified Stimulated Emission of Radiation. Optical fibers deliver high energy laser radiation to the target lesion. Neodymiun:yttrium aluminum garnet lasers (Nd:YAG), with a wavelength of 1064 nm, are used for PLA because penetration of light is optimal in the nearinfrared spectrum. In recent years, diode lasers with suitable wavelength have also been used. The penetration of laser light is only a few millimeters as a result of scattering and absorption. Scattering results in a relatively uniform distribution of absorbed energy, and heat is produced by conversion of absorbed light (8, 9). Temperatures greater than 60° C result in rapid coagulation necrosis. Irreversible cell death, without preceding coagulation, also occurs at lower temperatures (40–45° C), but requires duration of treatment that inversely correlates with temperature (8, 9). The first report of percutaneous laser ablation (PLA) on human thyroid tissue is by Pacella, et al. (10). In Reggio Emilia, we started to use PLA in patients with benign thyroid cold nodules in 2002. Since then, several studies have been published confirming effectiveness and safety of this new technique (11–19). Fig. 12.2 shows macroscopic changes occurring in a thyroid nodule resected one month after laser ablation. Nodule section shows tissue degeneration and necrosis with tissue carbonization (arrows). Vaporization and charring are consequences of tissue overheating. Plane-cut tip fibers achieve temperatures up to 180–200° C where fibers are in direct contact with tissue. Microscopically (Fig. 12.3), laser ablated areas are characterized
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
FIG. 12.1. Needle electrodes used for percutaneous ultrasound-guided radiofrequency (RF) ablation. Multiple needle (A), needle with four retractable lateral hooks (B), needle with multiple lateral hooks resembling an open umbrella (C)
193
194
R. VALCAVI ET AL.
FIG. 12.2. Macroscopic changes in a thyroid nodule resected one month after laser ablation. Nodule section shows tissue degeneration and necrosis with tissue carbonization (arrows)
FIG. 12.3. Microscopic section of laser ablated tissue. Typical changes include: central cavity due to tissue vaporization (A) surrounded by a thin layer of carbonized tissue (B), coagulative necrotic area (C), and transition zone (D) that separates necrotic from viable tissue
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
195
by a central cavity due to tissue vaporization (A) surrounded by a thin layer of carbonized tissue (B). Coagulative necrosis develops in the outer stratum of tissue (C) surrounded by a rim of metabolic damage (transition zone, D) that separates necrotic from viable tissue. Flat tip technique, proposed and developed by Pacella, et al. (10), is based on the insertion of a 300 µm plane-cut optic fiber through the sheath of a 21 G Chiba needle, exposing the nude fiber in direct contact with thyroid tissue for a length of 7–10 mm, according to the size of the lesion (Fig. 12.4). Fiber lockers allow us to expose the tip of the fiber within the lesion for the appropriate length (Fig 12.5). A single optic fiber, maintained in a still position, destroys only a small amount of tissue (16–18 mm in length, 8–10 mm in width, 8–10 mm in thickness, i.e., about 1 ml volume) when an energy of 1,600–1,800 Joules with an output power of 2–4 Watts is delivered (10). Therefore, simultaneous insertion of multiple fibers is generally
FIG. 12.4. 21 G Chiba needle with 300 µm plane-cut optic fiber inserted through the sheath, exposing the nude fiber for 10 mm
FIG. 12.5. Nude optic 300 µm plane-cut optic fibers. Fiber lockers allow precise fiber exposure into tissue out of needle tip
196
R. VALCAVI ET AL.
needed. Fiber pullback from the bottom to the upper part of the nodule, along the cranio-cadual axis, achieve further tissue destruction. A square configuration with four fibers has been used by Pacella for round shaped liver lesions (Fig. 12.6) (20). Parallel insertion of two to four fibers allows an ellipsoid ablation, suitable for ellipsoid shape of most benign thyroid nodules (Fig. 12.7). Using this latter technique, we obtained
FIG. 12.6. Multiple fiber technique for laser ablation volume increase. A square configuration with four fibers is used for round shaped lesions in large organs like the liver
FIG. 12.7. Multiple fiber ellipsoid configuration, suitable for ellipsoid shape of benign thyroid nodules. Needles are inserted along the craniocaudal nodule axis, at a distance of about 10 mm. Two to four fibers may be simultaneously inserted, achieving up to 40–45 mm wide, and 18–22 mm thick ablation diameters
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
197
up to 40–45 mm wide, 18–22 mm thick ablation diameters. Combining multiple fiber placement, fiber pull-back, and energies up to 6,000 Joules per fiber, nodules up to 50–70 mm in length may be treated with PLA. Number of fibers, number of pull-backs and total energy delivered are tailored to nodule volume. A maximum amount of about 30 ml nodular tissue may be destroyed in a single session (21). PLA INTERVENTION PLA is an office-based intervention. We established careful precautions for patient safety. A sterile operative setting is arranged (Fig. 12.8). The operator stands on the left side of the patient, while US equipment is used by the ultrasonography assistant who sits on the right side. A sensitive color-Doppler/ power color US system is required. Power color images are important during PLA procedure and follow-up (see below). An auxiliary monitor permits direct US vision by the operator while the assistant looks in the US machine monitor. Usually we do imaging by multifrequence linear probes (7.5–13 MHz) with a footprint of 3.5 or 4.5 cm. Convex probes with 6–7.5 MHz frequency are used only for large nodules. A cardiac monitor with defibrillator is connected to the patient showing continuous ECG. A venous catheter is inserted in a peripheral forearm vein before starting the procedure to ensure continuous venous access. Emergency care facilities and materials are on hand in
FIG. 12.8. Sterile PLA operative setting
198
R. VALCAVI ET AL.
the operating room. An anesthesiologist is not present during PLA. However, should an emergency occur, an anesthesiologist is immediately available in our hospital. The patient is placed in the supine position with hyperextended neck with a pillow under her/his shoulders. Eyes are protected by special glasses. The laser machine is placed behind the patient’s head. There is room for the operator to move around the patient’s head between the bed and laser appliance (Fig. 12.9). Delimitation of the nodule by palpation with a marker pen helps to find the point of needle insertion and plan optimal needle trajectory (Fig. 12.10). Light conscious sedation is obtained by IV diazepam (2–3 mg, repeatable during procedure if necessary). Local anesthesia with lidocaine subcapsular and subcutaneous infiltration is performed under US assistance (Fig. 12.11). US visualization of needle used for local anesthesia allows correct tissue lidocaine infiltration. In addition, multiplanar scans of this non-traumatic, thin needle
FIG. 12. 9. Patient placement with hyperxtended neck. The laser machine is in the back of patient’s head. There is room for the operator to move between the bed and laser machine
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
199
(G 29–30) help in planning subsequent 21 G Chiba PLA needle point of insertion and trajectory (Fig. 12.12). Guidance attachment may be used for 21 G Chiba needles insertion (Fig. 12.13). We prefer manual needle placement as it permits to fit needles according to variable anatomy of the nodule (Fig. 12.14). US images show 21 G Chiba needle introduced into the nodule. Longitudinal scans allow clear vision of the tip of the needle (Fig. 12.15 A, arrow). The inserted optic fiber is
FIG. 12.10. Nodule delimitation with a marker pen
FIG. 12.11. US-assisted local anesthesia with lidocaine infiltration
200
R. VALCAVI ET AL.
FIG. 12.12. Linear 12.5 MHz probe, 4.5 cm footprint. Longitudinal (A) and axial (B) US scans of a left lobe colloid thyroid nodule. Needle (29–30 G) US visualization during tissue lidocaine infiltration. Needle is inserted through prethyroid muscles (arrow)
FIG. 12.13. 21 G Chiba needle insertion with guidance attachment
FIG. 12.14. Needle manual placement. Needles are fit along the longitudinal, cranio-caudal, nodule axis
FIG. 12.15. Linear 12.5 MHz probe, 4.5 cm footprint. Longitudinal US scans of a left lobe colloid thyroid nodule. 21 G Chiba needle is placed into the nodule. Tip of the needle (arrow) (A). Optic fiber exposed (B). Arrows mark needle and fiber tips
202
R. VALCAVI ET AL.
then thoroughly seen (Fig. 12.15 B). The distance between needles is checked by US images and external measurements (Fig. 12.16). Accurate needle placement is critical for procedure success. Sedation and local analgesia reduce patient anxiety, swallowing, cough or other untoward movements that could impede precise needle insertion. After needle placement, fibers are inserted through the needle sheath into the nodule (Fig. 12.17) and laser firing is started (Fig. 12.18). US images through continuous axial, longitudinal and multiplanar scans are performed by the assistant throughout laser illumination duration (10–30 minutes), allowing real time visual control of each fiber. A highly echogenic area due to tissue heating and vaporization slowly enlarges over time (Fig. 12.19). The hyperechoic image gradually increases until coalescence between fibers is observed (Fig. 12.20). Figure 12.21 demonstrates an example of a typical 2-fiber session in a 9.5 ml (40 mm long × 17 mm thick × 27 mm wide) benign nodule of the left thyroid lobe. Panel (A) shows fiber placement with a distance of 1 cm between needles and laser firing. Oval shaped, hypoechoic area, close to nodule margins, shows presumed tissue ablation produced by 2-fiber coalescence (panel B). Binocular anechoic spots with hyperechoic halo, corresponding to cavitation and charring produced by two laser fibers, surrounded by hypoechoic tissue, are clearly seen one week (B) and one month (C) after PLA intervention. Peripheral capsular
FIG. 12.16. External measurement of distance between 21G Chiba needles (10 mm)
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
203
FIG. 12.17. Fibers inserted through 21 G Chiba needle sheath. Lockers ensure exact fiber exposition into nodular tissue
vascularization is preserved, while no blood flow is observed inside the nodule. The nonvascular, hypoechoic necrotic tissue will be reabsorbed over several months following PLA with consequent nodule shrinkage. PLA: CLINICAL RESULTS AND INDICATIONS From 2003 to 2005, we studied 119 patients, average age 56 years, who had benign hypofunctioning thyroid nodules with an average pretreatment volume of 24.8 ml (range 1.3–104 ml). In Reggio Emilia, we developed PLA technique originally proposed by Pacella, et al. (12). In order to achieve maximum ablation in a single session, optimize time and costs, avoid multiple sessions, and increase patient satisfaction and compliance (21), we built up energies, used up to four fibers simultaneously in the ellipsoid configuration, and the pull-back procedure. Table 12.1 shows the characteristics of population studied and technical parameters. At variance with other authors using mean energies of 3,000 Joule per session (12) or less (14, 15, 22), we delivered remarkably higher energies with a mean 7,060 Joule per fiber. Output power was 2–4 Watts. US features, size (ml), presumed necrosis (ml), lab tests (TSH, fT3, fT4,
204
R. VALCAVI ET AL.
FIG. 12.18. Axial US scan, linear 12.5 MHz probe. Left lobe solid isoechoic benign nodule (width 25 mm, thickness 19.5 mm) before fiber insertion (A). Two hyperechoic spots correspond to needles (arrows) before (B) and a few seconds (C) after laser illumination start. Initial tissue vaporization enhances echogenic areas around needles (arrows)
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
FIG. 12.19. PLA fiber illumination. Longitudinal US scans, 12.5 MHz probe. Highly echogenic area due to tissue heating and vaporization (A). Color Doppler images show laser firing (B)
FIG. 12.20. PLA treatment. Axial US scans, 12.5 MHz probe. Hyperechoic areas due to tissue vaporization enlarge and coalesce (A) during laser illumination. At the end of PLA treatment (C) nodule is filled by hyperechoic images due to tissue infiltration by gas
205
206
R. VALCAVI ET AL.
FIG. 12.21. PLA 2-fiber session. Axial US scans, linear 12.5 MHz probe, 9.5 ml left thyroid lobe benign nodule. Laser firing (A), one week (B), and (C) one month changes after PLA. Binocular images are due to laser fiber cavitation surrounded by charring. Nonvascular hypoechoic area corresponds to necrotic tissue. Initial nodule shrinkage is visible (C)
Thyroglobulin, anti-Thyroglobulin and anti-Thyroperoxidase antibodies), side effects, compressive symptoms and cosmetic scores were recorded before and at days 1, 7, 30, 90, 180, and 365 after PLA. Mean nodule volume slightly increased to 25.2 ml after a week, then gradually decreased to 12.1 ml after one year (Fig. 12.22). Figure 12.23 shows the mean percent change in nodule volume during the observation period, with a final 55.5% decrease after one year. Mean ablation volume was estimated at 10.6 ml, while the actual volume decrease was
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
207
TABLE 12.1. PLA procedure in 119 patients (Reggio Emilia, years 2003–2005) Patients Age Cytology No. of optic fibres Energy delivered Output power Treatment time Pre-treatment volume
119 (M 23 F 96) 55.8 ± 13.3* years Benign hyperplasia 1–4 7,060 ± 4,299* Joule (1,062–22,000 J) 2 – 4 Watts (2.9 ± 0.5 W) 19 ± 8* min 24.8 ± 21.1 ml* (1.3–104 ml) *mean ± SD
FIG. 12.22. Mean nodule volume before and at several time intervals up to one year, following PLA intervention in 119 patients
FIG. 12.23. Mean percent changes in nodule volume (Delta %) following PLA procedure in 119 patients
208
R. VALCAVI ET AL.
greater after one year (12.5 ml), possibly due to necrosis volume underestimation (Fig. 12.24 A). A significant correlation was observed between energy delivered and necrosis volume (Fig. 12.24 B). To study relationships between PLA effects and nodule volume, patients’ population was stratified into quartiles. Overall, 30 patients had small class 1 nodules (volume 1.3–9.6, mean 6.3 ml), 30 medium to small class 2 nodules (9.8–18.0 ml, mean 13.4 ml), 30 medium to large class 3 nodules (18.1–33.7 ml, mean 24.6 ml), and 29 large class 4 nodules (33.8–104.2 ml, mean 55.3 ml). Data demonstrate that nodules up to 18 ml volume (class 1 and class 2) are best candidates for PLA treatment, with a percent decrease in volume one year after PLA of −66% and −61%, respectively. However, size-tailored energy delivery technique developed in Reggio Emilia also obtained good results in larger nodules. One year after PLA, the percent volume change was − 47% for class 3 and − 50% for class 4 nodules (Table 12.2). Figures 12.25 and 12.26 show the cases of medium-small (10.7 ml), and medium-large (24 ml) nodules treated with proportional energies (3,600 and 8,000 Joules, 3 and 4 watts, respectively). In both cases PLA was successful. After one year, volume decreased by −81% and −85%, respectively. Other representative cases are shown in Fig. 12.27 and Fig. 12.28. In the latter patient, PLA has been preceded by PEI. Some authors claim that combined aspiration and PLA may be beneficial in cystic benign nodules (23). However, in nodules that are
FIG. 12.24. Mean nodule volume, necrosis volume, and final (Delta at 12 months) volume in 119 patients treated with PLA (A). Correlation between energy delivered and necrosis volume (B)
209
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
TABLE 12.2. PLA effects on mean nodule volume and necrosis, according to mean nodule size Baseline
Class 1 – small N = 30 Class 2 – small to medium N = 30 Class 3 – medium to large N = 30 Class 4 – large N = 29
12 Months ∆%
Volume ml
Necrosis ml Volume ml
6.3 13.4
3.7 7.2
2.2 5.6
− 66.0 − 61.3
24.6
10.2
13.2
− 46.8
55.3
23.4
27.4
− 49.9
FIG. 12.25. PLA treatment, 3,600 Joules, 3 Watts, two fibers. Axial US scans, linear 12.5 MHz probe. Mixed, isoechoic 9.5 ml right thyroid lobe benign nodule. Nodule structure before PLA (A) and immediately after PLA intervention (B). Volume was 2.0 ml (−81%) after 12 months (C)
210
R. VALCAVI ET AL.
FIG. 12.26. PLA treatment 8,000 Joules, 4 Watts, two fibers. Axial US scans, linear 12.5 MHz probe. Solid isoechoic 24 ml right thyroid lobe benign nodule. Hyperechoic spots due to tissue vaporization develop around optic fibers (arrows) (A). One month after PLA intervention, large hypoechoic, avascular, necrotic area is visible. Nodule volume is 15.7 ml (−34.6% of the initial volume) (B). One year after PLA intervention, nodule volume is 3.5 ml (−85% of the initial volume). Hyperechoic spots with acoustic shadow due to tissue carbonization correspond to initial fiber placement (arrows) (C)
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
211
FIG. 12.27. PLA treatment 6,000 Joules, 3 Watts, two fibers. Axial US scans, linear 12.5 MHz probe. Solid isoechoic 6.7 ml right lobe benign nodule with mixed areas before PLA procedure (A). Longitudinal scan during PLA intervention with inserted needle (B). Two months after PLA: nodule turned hypoechoic, with dark halo and a few hyperechoic spots due to tissue carbonization. Volume is 3.5 ml (−48%) (C)
212
R. VALCAVI ET AL.
FIG. 12.28. PLA treatment 14,400 Joules, 3 Watts, three fibers. Axial US scans with linear 12.5 MHz probe. Solid isoechoic, large (34.6 ml) isthmus and left thyroid lobe benign nodule, previously treated with PEI. Dense hyperechoic areas are due to prior PEI (A). One month after PLA. Volume is 21.9 ml (−7%) (B). One year after PLA: nodule shows hyperechoic spots with acoustic shadow due to remnants of tissue carbonization and gross calcification. Volume is 7.6 ml (−78%) (C)
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
213
more than 20% liquid, we perform PEI treatment prior to PLA intervention. This induces fluid reabsorption with nodule shrinkage, improving the efficacy of subsequent PLA procedure. SIDE EFFECTS There were no side effects in 76 patients (63.9%). Thirty-six complained of cervical pain (30.3%), slight in 33 cases (27.7%), and intense in three cases (2.5%), requiring anti-inflammatory drug medication. Four patients (3.3%) had fever (38–38.5° C) for two to three days, one patient (0.8%) had subcapsular hematoma with spontaneous reabsorption in two weeks, and one patient (0.8 %) had skin burn. No patient had vocal palsy or other severe side effects. More recently, we introduced routine medication as follows. Immediately after PLA procedure patients receive 20 mg IV methylprednisolone and IV 100 mg ketoprofene infusion. The day after PLA they start on prednisone 25 mg for three days, 12.5 mg for three days, and 5 mg for four days. Pump inhibitors are simultaneously administered (lansoprazole 30 mg p.o.) for 10 days. This medication protocol greatly enhances tolerance, reducing pain, tenderness, and occasional fever. Mean TSH and FT4 values slightly varied immediately after PLA (peak at day 1: TSH 1.23 ± 0.88 µU/ml vs. 0.79 ± 0.79 µU/ml, p < 0.001; fT4 11.52 ± 1.91 pg/ml vs. 12.81 ± 3.52 pg/ml, p < 0.05) and returned to baseline within three months. FT3 did not change after PLA. A large thyroglobulin (Tg) peak occurred at day 1 (4657 ± 7595 ng/ml vs. baseline 75.4 ± 83.4 ng/ml, p < 0.001), and gradually returned to baseline within three months. AntiTg and anti-TPO antibodies were slightly, but not significantly, increased following PLA. Six patients (5%) with baseline negative Anti Tg antibodies became positive at 12 months (normal levels < 100 U/ml). Laboratory changes were not symptomatic. Fig. 12.29 shows clinical results of benign thyroid nodule PLA intervention after one year (21). Compressive symptoms disappeared in 57.5% of patients, improved in 17%, and were unchanged in 25.5%. Cosmetic concerns disappeared in 43%, improved in 43% and did not change in 23%. PLA IN AUTONOMOUS THYROID NODULES Autonomously functioning thyroid nodules (AFNT) have also been treated with PLA (22, 24–26). Scinti scan of a representative case is shown in Fig. 12.30. The nodule was small (2.5 ml). Almost total ablation was obtained in this case.
214
R. VALCAVI ET AL.
FIG. 12.29. Effects of PLA treatment on compressive and cosmetic symptoms after one year. Results are expressed as percent changes (disappearance, improvement or no change) in 119 patients
FIG. 12.30. Scinti scan of a small autonomous adenoma of the left thyroid lobe before and six months after PLA treatment. The hot spot disappeared and normal thyroid uptake was restored
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
215
TABLE 12.3. Advantages 1. New therapeutic option in benign thyroid nodule management 2. Effective: thyroid nodule shrinkage > 50% in a single PLA session 3. Office-based procedure 4. High patient satisfaction 5. Safe: no major side effects Limitations 1. Restricted to specialized centers 2. Operator-dependent 3. Long-term effects unknown 4. More studies needed
However, in larger nodules, ablation may be insufficient without complete disappearance of AFNT. Moreover, recurrence is frequent. Combined reduced radioiodine doses and PLA has been proposed (27). However, this modality does not attain the basic objective of PLA, which is patient cure without other therapies such as radioactive iodine. Further studies are needed in AFNT. CONCLUSIONS PLA is a new, office-based, safe, highly effective technique in benign thyroid nodule treatment. Nodule shrinkage > 50% may be obtained with a single PLA session in variably sized nodules. Small to medium size nodules (< 18 ml, i.e., maximum 40–50 mm in length, maximum 25–35 mm in width, and maximum 15–20 mm in thickness) are best candidates for PLA procedure. Large nodules (18–50 ml) may also be treated. Multiple PLA sessions may be required in outsized nodules. Conscious sedation with IV benzodiazepines and local analgesia with lidocaine permit compliant, well tolerated, PLA intervention. Post-ablation corticosteroid administration minimizes pain and other side effects. Patient satisfaction is greatly enhanced by this method. PLA may be an alternative choice to surgery in the clinical management of benign thyroid nodules. At present, PLA technique is restricted to specialized centers, should to be performed in safe conditions (i.e., hospital setting), and is operator-dependent. Long-term effects are unknown (Table 12.3). Technological improvements, technique standardization, medium-to long-term multicentric, controlled, clinical trials are needed.
216
R. VALCAVI ET AL.
Acknowledgment. Authors thank Sara Lenzi for her invaluable assistance in editing the manuscript and figures. References 1. Livraghi T, Paracchi A, Ferrari C, Bergonzi M, Garavaglia G, Raineri P, Vettori C (1990) Treatment of autonomous thyroid nodules with percutaneous ethanol injection: preliminary results.Work in progress. Radiology 175:827–9 2. Valcavi R, Frasoldati A (2004)Ultrasound-guided percutaneous ethanol injection therapy in thyroid cystic nodules. Endocr Pract 10:269–75 3. American Association of Clinical Endocrinoligists and Associazione Medici Endocrinologi (2006) Medical guidelines for clinical practice for the diagnosis and management of thyroid nodules. Endocr Pract 12:63–102 4. Esnault O, Franc B, Monteil JP, Chapelon JY (2004) High-intensity focused ultrasound for localized thyroid-tissue ablation: preliminary experimental animal study. Thyroid 14:1072–6 5. Gazelle GS, Goldberg SN, Solbiati L, Livraghi T (2000) Tumor ablation with radio-frequency energy. Radiology 217:633–46 6. Kim YS, Rhim H, Tae K, Park DW, Kim ST (2006) Radiofrequency ablation of benign cold thyroid nodules: initial clinical experience. Thyroid 16:361–7 7. Roh JL (2006) Intrathyroid hemorrhage and acute upper airway obstruction after fine needle aspiration of the thyroid gland. Laryngoscope 116:154–6 8. Jacques SL (1992) Laser-tissue interactions: photochemical, photothermal, and photomechanical. Surg Clin North Am 72:531–558 9. Thomsen S (1991) Pathologic analysis of photothermal and photomechanical effects of laser-tissue interactions. Photochem Photobiol 53:825–835 10. Pacella CM, Bizzarri G, Guglielmi R, Anelli V, Bianchini A, Crescenzi A, Pacella S, Papini E (2000) Thyroid tissue: USguided percutaneous interstitial laser ablation—a feasibility study. Radiology 217:673–7 11. Dossing H, Bennedbaek FN, Karstrup S, Hegedus L (2002) Benign solitary solid cold thyroid nodules: US-guided interstitial laser photocoagulation-initial experience. Radiology 225:53–7 12. Pacella CM, Bizzarri G, Spiezia S, Bianchini A, Guglielmi R, Crescenzi A, Pacella S, Toscano V, Papini E (2004) Thyroid tissue: USguided percutaneous laser thermal ablation. Radiology 232:272–80 13. Papini E, Guglielmi R, Bizzarri G, Pacella CM (2004) Ultrasoundguided laser thermal ablation for treatment of benign thyroid nodules. Endocr Pract 10:276–83 14. Dossing H, Bennedbaek FN, Hegedus L (2005) Effect of ultrasound-guided interstitial laser photocoagulation on benign solitary solid cold thyroid nodules— a randomised study. Eur J Endocrinol 152:341–5
LASER AND RADIOFREQUENCY ABLATION PROCEDURES
217
15. Amabile G, Rotondi M, Chiara GD, Silvestri A, Filippo BD, Bellastella A, Chiovato L (2006) Low-energy interstitial laser photocoagulation for treatment of nonfunctioning thyroid nodules: therapeutic outcome in relation to pretreatment and treatment parameters. Thyroid 16:749–55 16. Dossing H, Bennedbaek FN, Hegedus L (2006) Effect of ultrasound-gGuided iInterstitial laser photocoagulation on benign solitary solid cold thyroid nodules: one versus three treatments. Thyroid 16:763–768 17. Gambelunghe G, Fatone C, Ranchelli A, Fanelli C, Lucidi P, Cavaliere A, Avenia N, d’Ajello M, Santeusanio F, De Feo P (2006) A randomized controlled trial to evaluate the efficacy of ultrasoundguided laser photocoagulation for treatment of benign thyroid nodules. J Endocrinol Invest 29:RC23–6 18. Cakir B, Topaloglu O, Gul K, Agac T, Aydin C, Dirikoc A, Gumus M, Yazicioglu K, Ersoy RU, Ugras S (2006) Effects of percutaneous laser ablation treatment in benign solitary thyroid nodules on nodule volume, thyroglobulin and anti-thyroglobulin levels, and cytopathology of nodule in 1 yr follow-up. J Endocrinol Invest 29:876–84 19. Papini E, Guglielmi R, Bizzarri G, Graziano F, Bianchini A, Brufani C, Pacella S, Valle D, Pacella CM (2007) Treatment of benign cold thyroid nodules: a randomized clinical trial of percutaneous laser ablation versus levothyroxine therapy or follow-up. Thyroid 17:229–35 20. Pacella CM, Bizzarri G, Francica G, Bianchini A, De Nuntis S, Pacella S, Crescenzi A, Taccogna S, Forlini G, Rossi Z, Osborn J, Stasi R (2005) Percutaneous laser ablation in the treatment of hepatocellular carcinoma with small tumors: analysis of factors affecting the achievement of tumor necrosis. J Vasc Interv Radiol 16:1447–1457 21. Valcavi R, Pesenti M, Bertani A, Frasoldati A, Formisano D (2006) Percutaneous laser ablation (PLA) in 119 benign thyroid nodules. RSNA Annual Meeting SSC 15–04 22. Barbaro D, Orsini P, Lapi P, Pasquini C, Tuco A, Righini A, Lemmi P (2007) Percutaneous laser ablation in the treatment of toxic and pretoxic nodular goiter. Endocr Pract 13:30–6 23. Dossing H, Bennedbaek FN, Hegedus L (2006) Beneficial effect of combined aspiration and interstitial laser therapy in patients with benign cystic thyroid nodules: a pilot study. Br J Radiol 79:943–7 24. Dossing H, Bennedbaek FN, Hegedus L (2003) Ultrasound-guided interstitial laser photocoagulation of an autonomous thyroid nodule: the introduction of a novel alternative. Thyroid 13:885–8 25. Spiezia S, Vitale G, Di Somma C, Pio Assanti A, Ciccarelli A, Lombardi G, Colao A (2003) Ultrasound-guided laser thermal ablation in the treatment of autonomous hyperfunctioning thyroid nodules and compressive nontoxic nodular goiter. Thyroid 13:941–7
218
R. VALCAVI ET AL.
26. Dossing H, Bennedbaek FN, Bonnema SJ, Grupe P, Hegedus L (2007) Randomized prospective study comparing a single radioiodine dose and a single laser therapy session in autonomously functioning thyroid nodules. Eur J Endocrinol 157:95–100 27. Guglielmi R, Papini E, Pacella CM, Todino V, Bizzarri G (2006) Combined treatment of large toxic nodular goiter by percutaneous laser ablation and radioiodine 131I: a pilot study. American Association of Clinical Endocrinologists 15th Annual Meeting & Clinical Congress Abstract Book: 72
Chapter 13
High Intensity Focused Ultrasound (HIFU) Ablation Therapy for Thyroid Nodules Olivier Esnault and Laurence Leenhardt
INTRODUCTION AND RATIONALE High Intensity Focused Ultrasound (HIFU) is a unique process that delivers a large amount of heat energy to a confined space. HIFU can coagulate tissue at a distance. The effect of HIFU on tissues and tumors was first established in the early to mid1950s (1,2), and more recently partial or complete destruction of tumor (3–8) was shown in the animal; HIFU is already being used to treat localized prostate cancer. This technique has been shown to lower costs, shorten hospitalization stays and represents a valuable alternative for patients for whom surgery is contraindicated. Thyroid nodules are frequently discovered during routine physical examinations or during investigations for other purposes (cervical ultrasonography, carotid duplex exams, cervical scans). Recently, systematic ultrasonographic exploration of a French large adult cohort indicated that 14.5% of the subjects had nodular thyroid structures (9). In the United States, 40% of the female population age 50 or older presented with thyroid nodules at ultrasonography, and the prevalence of thyroid nodules increases throughout life (10). Fortunately, more than 95% of thyroid nodules are the result of benign disease processes and the incidence of thyroid cancer is low. Fine needle aspiration biopsy (FNAB) is considered the most reliable test for the diagnosis of malignant thyroid nodule. In large published series, adequate cytological material is classified as benign, malignant, or suspicious in 69%, 4% and 10% of cases, respectively (11). Benign cytological results correspond to colloid or macrofollicular adenomas, nodular and/or cystic goiters, 219
220
O. ESNAULT AND L. LEENHARDT
or thyroiditis. Patients presenting with such benign nodules are subjected to long follow-up. However, the best therapeutic strategy after the discovery of such nodules is still a matter of debate (12). Some nodules are treated with thyrotropin (TSH)suppressive levothyroxine (LT4) therapy. Unwanted effects of thyroxine treatment on the skeletal and cardiovascular systems (13) lead physicians to take a wait-and-see policy that is often preferred (12, 14). Moreover, the effectiveness of suppressive thyroxine therapy in reducing the volume of benign thyroid nodules is controversial (13, 15). Despite reassuring cytological results, some physicians advise removing them, especially in cases of increases in a nodule’s volume, pressure symptoms or cosmetic complaints. Then the goal of the physician is to delay the time of surgery, or even to switch to another treatment which could represent an alternative to surgery. Ideally a new method must be minimally invasive and be done in an ambulatory setting. HIFU meets these requirements. Compared to other alternative treatments such as percutaneous laser ablation, radiofrequency or ethanol injection, HIFU represents a promising non-invasive procedure that patients would favor. The project’s aim is four-fold: - firstly, to assess the feasibility of using HIFU to obtain localized ablation of thyroid tissue without affecting neighboring structures on sheep’s thyroid, - secondly, to evaluate the safety, feasibility and efficacy of HIFU for the destruction of thyroid nodules in patients who are indicated for thyroid surgery, - thirdly, to confirm safety of HIFU treatment and study ultrasonographic nodule changes after HIFU in non-operated patients (the hypothesis is that HIFU treatment would delay surgery and should, therefore, represent an alternative treatment for patients with benign nodular disease) - lastly, to study other applications of HIFU as treatment of toxic adenomas, primary or secondary hyperparathyroidism, treatment of recurrent thyroid cancer and define indications of HIFU treatment in the management of thyroid diseases. HIFU TECHNOLOGY Principle of HIFU High Intensity Focused Ultrasound (HIFU) is a unique process of delivering a large amount of heat energy to a confined space. The acoustic energy is produced by an ultrasonic transducer and concentrated on the tissue to be treated. The energy heats
HIGH INTENSITY FOCUSED ULTRASOUND
(HIFU)
221
the tissue to therapeutic levels, producing necrosis by a thermal effect, with minimum effect on surrounding structures. Simply stated, power ultrasound can coagulate tissue at a distance. A piezoelectric transducer emits a beam of convergent ultrasonic wave towards a tissue target. When ultrasound waves enter the tissue interface (the skin and the muscles) the beam is wide so the power density is low, ensuring that no damage is done to these superficial structures. However, as the beam converges into the target, the power density becomes so high that the tissue heats to over 85° C and coagulates in a few seconds. During an HIFU treatment, the computer controlled device will cause small adjacent lesions (2 mm diameter by 10 depth) to chosen portions of the thyroid gland, while preserving the surrounding structures. A typical treatment lasts less than half an hour, as the treatment head traverses the treatment zone. The imaging and targeting is carried out by a standard linear array included in the device. APPLICATIONS HIFU is intended to control localized tumors in patients with Stage T1–T2 prostate cancer. The focused energy is delivered from an endorectal probe containing an ultrasound treatment transducer and an imaging transducer. This medical device received CE Mark in 2000 for its prostate application. In Europe, Ablatherm® Treatment System (Technomed Medical Systems, Lyon, France), is used in daily practice for the treatment of prostate cancer (16–20). Prior experience (21) with the early development of the Ablatherm device for the prostate
FIG. 13.1. Principle of HIFU
222
O. ESNAULT AND L. LEENHARDT
FIG. 13.2. Principle of HIFU, dose delivery
served as a guide for the following studies: in patients presenting with a T1–2 N0 M0 prostate carcinoma, HIFU treatment was performed under general anesthesia seven to 12 days prior to radical prostatectomy. Only the lobe in which carcinoma was confirmed was treated. The radical prostatectomy specimen was examined histopathologically, and the changes were compared with the targeted zones. In all cases complete necrosis was seen in the treated region. It was shown that extensive coagulative necrosis can be obtained in the treated areas; however, exact targeting is crucial and a prerequisite for extended clinical application of HIFU. Long-term experience with Ablatherm confirmed the early results (16–20) Other potential applications of HIFU for benign prostate hypertrophy, gynecological tumors, and renal cell carcinoma are currently under investigation. In a logical way, we assessed the feasibility of HIFU for the localized ablation of thyroid tissue and conducted experimental and humans trials for the treatment of thyroid diseases, especially thyroid nodules. ANIMAL TRIALS The sheep model proved to be the best choice because of its cervical anatomy. The sheep’s thyroid gland is easily accessible with ultrasound. It is located in the middle of the animal’s neck, is rather superficial and its size is comparable to humans’. As
HIGH INTENSITY FOCUSED ULTRASOUND
(HIFU)
223
sheep do not develop thyroid nodules the treatments involved plain, healthy thyroid tissue. These trials have been supported by a grant from Fondation de l’Avenir pour la Recherche Médicale Appliquée, INSERM, and industrial support by EDAP-Technomed SA. FIRST TRIALS: PROVE FEASIBILITY The first series of trials was carried out with the same Ablatherm™ device (EDAP, Vaulx en Velin, France) as used for the prostate clinical treatments. The endorectal probe featured a 7.5 MHz sector scan imaging probe and a 3 MHz power transducer. Nonetheless the probe was attached to a gantry for positioning above the neck of the animal (Fig. 13.3). To target the lesions, the author overlaid a computer generated pattern which simulated the lesions over the ultrasonic image of the target. The ultrasound power parameters were the same as for human prostate. In the first animal trial both thyroid lobes of eight animals were treated (22). The trachea was hit four times on three animals, the oesophagus once, and the skin once. In all the remaining cases (nine last lobes) the lesion was confined to the gland and no adjacent tissue was damaged. The goal of the initial trials was to prove feasibility and to justify the development of a device for human use. This first study roughly defined the energy parameters, and the subsequent trials were carried out using several devices progressively adapted to the human thyroid.
FIG. 13.3. Positioning above the neck of the animal
224
O. ESNAULT AND L. LEENHARDT
FIG. 13.4. Macroscopic examination
FIG. 13.5. On the left-hand side, the dark treated tissue demonstrates HIFU precision near the trachea.On the right, a bird’s eye view shows the magnified destroyed thyroid tissue
SECOND TRIAL: ADJUSTMENT OF TREATMENT PARAMETERS HIFU treatment was carried out on 27 additional animals to confirm our preliminary results. In a second series of experi-
HIGH INTENSITY FOCUSED ULTRASOUND
(HIFU)
225
ments, typical lesions were coagulative necrosis around the targeted area. The sacrifice of three animals at 30 days showed that the coagulation necrosis is later replaced by fibrosis. Macroscopic examination of adjacent organs revealed mainly skin lesions and muscle injuries. In a third series, the safety of the method was evaluated by repeated firings at the thyroid lobes’ periphery to explore consequences of the surrounding structures’ injuries. Macroscopic examination revealed a tracheal lesion, superficial oesophagus lesions and recurrent nerve lesions. When the recurrent nerves were damaged bilaterally we observed dysphagia and the ewe died three days after HIFU damage. Finally, the repeatability of the method was evaluated using an HIFU prototype designed specifically for human use. The desired thyroid lesions were obtained in the treated lobes. No damage to the nerves, trachea, oesophagus or muscle was observed. At this stage, the use of the HIFU device on sheep thyroid demonstrated several points: ●
●
●
●
●
● ● ●
It is possible to coagulate thyroid tissue at a distance without side effects In the weeks following the treatment fibrosis replaces the coagulated tissue Subject to minor improvements, the prototype was ready for clinical use The limitations of the animal models should be kept in mind when extrapolating those data to the human situation: In the sheep the dimensions of the gland and the distances to surrounding structures are about two-thirds of those in men The anterior part of the neck is flatter in men than in sheep No nodes can be found in sheep thyroids Many immuno histo chemical markers cannot be used in the animal, and must be tested in the next human trials
HUMAN TRIALS A modified device was built to treat the thyroid gland. Animal studies showed that thyroid tissue could be accurately targeted, with no damage to surrounding structures, and that the observed pathologic lesions were typical of HIFU effect (necrosis, haemorrhage and aspects of mummified tissue). The prototype device (Theraclion, Paris, France) was designed to deliver precise high intensity focused ultrasound (HIFU) to the thyroid, resulting in thermal destruction of thyroid tissue. The focused energy is delivered from an external probe containing both ultrasound
226
O. ESNAULT AND L. LEENHARDT
treatment and imaging transducers. INSERM and industrial partner Theraclion SAS have supported these trials. HUMAN FEASIBILITY STUDY The first human study was performed to evaluate the safety, feasibility and efficacy of HIFU for the destruction of thyroid nodules in patients who are indicated for thyroid surgery. Twenty-five patients scheduled for thyroidectomy were treated by the device. Ultrasound (US) examination was performed before and after HIFU treatment. The HIFU treatment lasted approximately 15 minutes per patient. Thyroidectomy was performed at two weeks followed by histopathological examination. During this study, the HIFU device was optimized and the acoustic energy was gradually increased in order to reach the goal of the complete destruction of the nodules, while minimizing complications. For the sake of caution this was done very progressively. Mild adverse events occurred, such as small skin burns and mild local pain during the HIFU shots. The skin burns resolved quickly and without any medication in all patients but one, who kept a punctiform scar. After adaptation of the treatment head no further skin lesions were observed. Local pain caused by HIFU shots increases with the intensity of the energy, ceasing immediately as the treatment is interrupted. As the energy density was increased, adapted analgesia using oral and local medications was given for pain relief during treatment. Post-HIFU US examination showed changes in echogenicity, a decrease of vascularization at power Doppler examination. Macroscopic and histological lesions were observed, and were precisely located in the targeted nodule without affect to the neighboring structures. Lesions in the treated nodule were thrombosis, diffuse lesion with cavitations, coagulative necrosis, hemorrhage and disappearance of the nuclei (data not shown). The extent of the damage varied among the patients according to the delivered energy. The histological assessment after two weeks is only preliminary evidence of the effect. These studies confirmed the feasibility and safety of the HIFU procedure. No serious adverse event was observed, particularly no affect to the recurrent nerves or to the trachea. Likewise the parathyroid remained untouched.
HIGH INTENSITY FOCUSED ULTRASOUND
FIG. 13.6. Operator’s interface
FIG. 13.7. HIFU lesion, macroscopic
(HIFU)
227
228
O. ESNAULT AND L. LEENHARDT
FIG. 13.8. HIFU lesion microscopic
ONGOING HUMAN STUDIES Benign Thyroid Nodules The objectives of this ongoing study are to evaluate the efficacy and safety of the ENT HIFU device for the destruction of thyroid nodules in patients presenting with at least one thyroid nodule with no signs of malignancy and who wish to avoid or to delay surgery. The primary objective is the evaluation of the nodule’s volume change following ENT HIFU treatment, as assessed by ultrasonography (US) performed at six-month follow-up (M6), compared with baseline results, and to describe the nodule volume change between M6 and baseline in the observation group. Toxic Nodules The effectiveness and safety of the ENT HIFU device in treating patients who present with at least one autonomous hyperfunctioning thyroid nodule is studied. The primary objective is to assess the proportion of patients with TSH normalization three months after the first course of treatment with HIFU. HIFU’S ROLE IN THE MANAGEMENT OF THYROID DISEASES In 2006, the American Thyroid Association (ATA) published treatment guidelines for patients with thyroid nodules and thyroid cancer (14). Recommendations pointed out that if the nodule is benign on cytology, further immediate diagnostic
HIGH INTENSITY FOCUSED ULTRASOUND
(HIFU)
229
FIG. 13.9. Ultrasound features before treatment
study or treatment is not routinely required. Nevertheless, it seems that the majority of benign non-functioning nodules grow, particularly those that are solid. Then, thyroid surgery is often indicated in such cases, even for benign nodules. The annual rate of evolution of a solitary functioning nodule into a hyperfunctioning nodule is as high as 6%. The risk is positively related to the size of the nodule and negatively related to the serum thyrotropin level. There is controversy as to whether a nodule should be treated and, if so, how. Figure 13.3 from Hegedus, et al. (23) shows a management algorithm that highlights the place of alternative treatments that contrasts with the ATA algorithm. These differences are probably explained,
230
O. ESNAULT AND L. LEENHARDT
FIG. 13.10. Ultrasound features after treatment
in part, by differences in iodine intake in populations that lead to significant differences in natural history and management of thyroid nodules. The therapeutic strategy for treating benign nodules includes several options. Standard therapeutic approaches comprise surgery, radioiodine treatment and levothyroxine therapy. Surgery is the most effective therapy for symptomatic thyroid nodules and should be regarded as standard treatment (23). Nevertheless, thyroidectomy is an invasive and expensive treatment option for a benign condition, which requires an expert surgeon to avoid major complications. Postoperative hypocalcemia or injury to the laryngeal nerve can occur in 1% and 1%–3%,
HIGH INTENSITY FOCUSED ULTRASOUND
(HIFU)
231
FIG. 13.11. Current prototype
respectively. Disadvantages include hospitalization, general anesthesia, postoperative scar formation and iatrogenic hypothyroidism. Still, the need for an effective alternative exists, especially if surgery is declined or the patient is a poor surgical risk. In these situations a nonsurgical approach may be indicated. Radioiodine is a simple, cost effective, safe procedure in the treatment of autonomously hyperfunctioning thyroid nodules. In a prospective study (24) there was a 45% decrease in the total thyroid volume within three months after radioiodine treatment, and 75% of patients with no previous antithyroid drug treatment normalized for thyroid function within three months. In another recently published prospective study (25) there was a decrease in hot nodules volume assessed by US of 28.8% three months, 46.2% six months and 54% 12 months after radioiodine treatment, and 66.7% of patients were euthyroid at three months, 71.8% at six months, and 76.9% at 12 months. Radioiodine may also be recommended to treat large nontoxic multinodular goiters in patients who decline surgery or who are at high surgical risk. When radioiodine treatment is proposed, physicians must follow regulations and
232
O. ESNAULT AND L. LEENHARDT
FIG. 13.12. Patient positioning
policies on 131I in Europe and the United States. The European Union recently adopted the main international commission on radiological protection (ICRP) recommendations on radiation protection. Such regulations may raise a number of practical problems for 131I treated patients. With the exception of a few countries like Germany and Switzerland, there often is no legislation defining the maximal dose of radioiodine that can be administered on an ambulatory basis. Policies and recommendations have been proposed to reduce the radiation hazards for the public or the family living around the radioiodine-treated patients. Therefore, because of these radioprotection recommendations, and because of contraindication to surgery and to radioiodine in some cases, an alternative treatment for toxic nodular nodules may be suggested. Levothyroxine therapy is still a common approach to the management of thyroid nodules in euthyroid patients with inducing thyroid suppression. However, only a subset of patients has a clinically significant reduction of nodule size (13, 15), and recurrence is seen after cessation of therapy. Moreover, LT4 treatment may cause adverse effects such as atrial fibrillation or reduced bone density. These limita-
HIGH INTENSITY FOCUSED ULTRASOUND
(HIFU)
233
Thyroid nodule
History, physical examination, and serum thyrotropin test
Low thyrotropin level
Scintigraphy
Normal or high thyrotropin level
Functioning nodule Strong suspicion of cancer
Clinical evaluation
Surgery
Radioiodine; alternatives include no treatment, surgery, ethanol injection, and laser treatment (experimental)
Ultrasonographically guided FNAB
Diagnostic results
Nondiagnostic results
Malignant
Suspicious
Benign
Surgery
Surgery
No treatment with clinical follow-up; alternatives include surgery, levothyroxine therapy, ethanol injection, and laser treatment (experimental)
Repeat ultrasonographically guided FNAB
Nondiagnostic results
Surgery
FIG. 13.13. Algorithm for the Cost-Effective Evaluation and Treatment of a clinically Detectable Solitary Thyroid Nodule
tions of standard therapeutic approaches raised interest for alternative treatments. Recently, several publications reported results of interstitial laser photocoagulation (26–29), percutaneous ethanol injection (30–31) and radiofrequency ablation (RF) (32). In a series published by Papini, et al. comparing clinical and ultrasound changes induced in cold nodules by US-guided percutaneous laser ablation (PLA) versus follow-up or LT4 suppressive therapy, PLA was more effective than LT4. A nodule reduction > 50% was found in 33.3% of cases in the PLA-treated group and was not significantly observed in LT4 group. Age, sex ultrasound pattern, pre-treatment volume, number of PLA treatments and total energy delivered did not show any significant correlation with therapeutic outcome (27). Therapeutic efficacy of RF ablation was reported (32). Nevertheless, thermal injury to
234
O. ESNAULT AND L. LEENHARDT
the recurrent laryngeal nerve was the most serious complication in this study (3.3%). Ethanol injection was reported to be more effective in cystic nodules. These side effects and invasiveness of such procedures raise questions and lead to proposed HIFU treatment in solid or mixed nodules. CONCLUSION The technical treatment parameters leading to a precise necrosis of the targeted nodule by HIFU were determined in feasibility studies. The ultimate goal of HIFU treatment is to induce a significant reduction in the volume of thyroid nodules, perhaps leaving scarring in longer term. Since other thermal therapies such as laser show that the decreases in nodule volume are fully visible after three months, it can be expected that a similar result will be obtained with HIFU technology. Ongoing studies are underway to confirm the efficacy and safety of HIFU treatment of thyroid nodule, parathyroid adenoma, cervical lymph nodes and thyroid cancer recurrences. References 1. Fry WJ, Barnard JW, Fry FJ, Brennan JF (1955) Ultrasonically produced localized selective lesions in the central nervous system. American Journal of Physical Medicine 34(3):413–23 2. Fry WJ, Mosberg WH, Jr., Barnard JW, Fry FJ (1954) Production of focal destructive lesions in the central nervous system with ultrasound. Journal of Neurosurgery 11(5):471–8 3. Fry FJ, Johnson LK (1978) Tumor irradiation with intense ultrasound. Ultrasound in Medicine & Biology 4(4):337–41 4. Moore WE, Lopez RM, Matthews DE, et al (1989) Evaluation of high-intensity therapeutic ultrasound irradiation in the treatment of experimental hepatoma. Journal of Pediatric Surgery 24(1):30– 3; discussion 3 5. Yang R, Reilly CR, Rescorla FJ, et al (1991) High-intensity focused ultrasound in the treatment of experimental liver cancer. Arch Surg 126(8):1002–9; discussion 9–10 6. Chapelon JY, Margonari J, Theillere Y, et al (1992) Effects of highenergy focused ultrasound on kidney tissue in the rat and the dog. European Urology 22(2):147–52 7. Chapelon JY, Margonari J, Vernier F, Gorry F, Ecochard R, Gelet A (1992) In vivo effects of high-intensity ultrasound on prostatic adenocarcinoma Dunning R3327. Cancer Research 52(22):6353–7 8. Margonari J, Chapelon JY, Gelet A, Cathignol D, Bouvier R, Gorry F (1992) Tumor ablation with focalized ultrasound. In vivo
HIGH INTENSITY FOCUSED ULTRASOUND
9.
10. 11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
(HIFU)
235
experiment with prostatic adenocarcinoma R3327 Mat-Ly-Lu. Prog Urol 2(2):207–19 Valeix P, Zarebska M, Bensimon M, et al (2001) Ultrasonic assessment of thyroid nodules, and iodine status of French adults participating in the SU.VI.MAX study. Ann Endocrinol (Paris) 62(6):499–506 Mazzaferri EL (1993) Management of a solitary thyroid nodule. The New England Journal of Medicine 328(8):553–9 Gharib H, Goellner JR, Johnson DA (1993) Fine-needle aspiration cytology of the thyroid. A 12-year experience with 11,000 biopsies. Clinics in Laboratory Medicine 13(3):699–709 Bennedbaek FN, Perrild H, Hegedus L (1999) Diagnosis and treatment of the solitary thyroid nodule. Results of a European survey. Clinical Endocrinology 50(3):357–63 Castro MR, Caraballo PJ, Morris JC (2002) Effectiveness of thyroid hormone suppressive therapy in benign solitary thyroid nodules: a meta-analysis. J Clin Endocrinol Metab 87(9):4154–9 Cooper DS, Doherty GM, Haugen BR, et al (2006) Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 16(2):109–42 Wemeau JL, Caron P, Schvartz C, et al (2002) Effects of thyroidstimulating hormone suppression with levothyroxine in reducing the volume of solitary thyroid nodules and improving extranodular nonpalpable changes: a randomized, double-blind, placebocontrolled trial by the French Thyroid Research Group. J Clin Endocrinol Metab 87(11):4928–34 Poissonnier L, Chapelon JY, Rouviere O, et al (2007) Control of prostate cancer by transrectal HIFU in 227 patients. European Urology 51(2):381–7 Beerlage HP, Thuroff S, Debruyne FM, Chaussy C, de la Rosette JJ (1999) Transrectal high-intensity focused ultrasound using the Ablatherm device in the treatment of localized prostate carcinoma. Urology 54(2):273–7 Rebillard X, Gelet A, Davin JL, et al (2005) Transrectal highintensity focused ultrasound in the treatment of localized prostate cancer. Journal of Endourology / Endourological Society 19(6):693–701 Blana A, Walter B, Rogenhofer S, Wieland WF (2004) Highintensity focused ultrasound for the treatment of localized prostate cancer: 5-year experience. Urology 63(2):297–300 Thuroff S, Chaussy C, Vallancien G, et al (2003) High-intensity focused ultrasound and localized prostate cancer: efficacy results from the European multicentric study. Journal of Endourology / Endourological Society 17(8):673–7 Beerlage HP, van Leenders GJ, Oosterhof GO, et al (1999) Highintensity focused ultrasound (HIFU) followed after one to two weeks by radical retropubic prostatectomy: results of a prospective study. Prostate 39(1):41–6
236
O. ESNAULT AND L. LEENHARDT
22. Esnault O, Franc B, Monteil JP, Chapelon JY (2004) High-intensity focused ultrasound for localized thyroid-tissue ablation: preliminary experimental animal study. Thyroid 14(12):1072–6 23. Hegedus L (2004) Clinical practice. The thyroid nodule. The New England Journal of Medicine 351(17):1764–71 24. Nygaard B, Hegedus L, Nielsen KG, Ulriksen P, Hansen JM (1999) Long-term effect of radioactive iodine on thyroid function and size in patients with solitary autonomously functioning toxic thyroid nodules. Clinical Endocrinology 50(2):197–202 25. Erdogan MF, Kucuk NO, Anil C, et al (2004) Effect of radioiodine therapy on thyroid nodule size and function in patients with toxic adenomas. Nuclear Medicine Communications 25(11):1083–7 26. Dossing H, Bennedbaek FN, Hegedus L (2005) Effect of ultrasound-guided interstitial laser photocoagulation on benign solitary solid cold thyroid nodules - a randomised study. Eur J Endocrinol 152(3):341–5 27. Amabile G, Rotondi M, Chiara GD, et al (2006) Low-energy interstitial laser photocoagulation for treatment of nonfunctioning thyroid nodules: therapeutic outcome in relation to pretreatment and treatment parameters. Thyroid 16(8):749–55 28. Dossing H, Bennedbaek FN, Hegedus L (2006) Effect of ultrasound-guided interstitial laser photocoagulation on benign solitary solid cold thyroid nodules: one versus three treatments. Thyroid 16(8):763–8 29. Papini E, Guglielmi R, Bizzarri G, et al (2007) Treatment of benign cold thyroid nodules: a randomized clinical trial of percutaneous laser ablation versus levothyroxine therapy or follow-up. Thyroid 17(3):229–35 30. Zieleznik W, Kawczyk-Krupka A, Barlik MP, Cebula W, Sieron A (2005) Modified percutaneous ethanol injection in the treatment of viscous cystic thyroid nodules. Thyroid 15(7):683–6 31. Lee SJ, Ahn IM (2005) Effectiveness of percutaneous ethanol injection therapy in benign nodular and cystic thyroid diseases: long-term follow-up experience. Endocrine Journal 52(4):455–62 32. Kim YS, Rhim H, Tae K, Park DW, Kim ST (2006) Radiofrequency ablation of benign cold thyroid nodules: initial clinical experience. Thyroid 16(4):361–7
Chapter 14
Ultrasound Elastography of the Thyroid Robert A. Levine
It has long been recognized that palpably hard thyroid nodules are suspicious for cancer (1). Conventional ultrasound provides information regarding characteristics shown to be correlated with risk of cancer, such as shape, echogenicity, edge definition, calcification, and vascular flow. However, it does not provide direct information corresponding to the hardness of a nodule. Elastography is a newly developed technique that utilizes ultrasound to analyze the stiffness of a nodule by measuring the amount of distortion that occurs when the nodule is subjected to external pressure. The technique was first described ten years ago, but has only recently been tested on thyroid nodules. Preliminary results have shown an excellent correlation between the hardness of a nodule determined by elastography and subsequent pathology determined by biopsy or excision. Two techniques have been employed to provide external pressure and strain to a nodule. The most common technique is to apply external pressure using the transducer. After placing a linear transducer over the region of interest, the ultrasonographer manually applies light pressure with the transducer. Multiple sites within and around the nodule are analyzed, and the ultrasound software compares the deformation of the nodule to the surrounding tissue (strain index). The relative stiffness is shown on a color display, superimposed on a B-mode image. An alternative technique uses pulsation from the carotid artery as the compression source. This may be useful, particularly when the nodule is in the lateral aspect of the gland, near the carotid. One preliminary study has indicated that carotid pulsation can be used as the pressure source for elastography (2). On the other hand, artifacts introduced by carotid pulsation have been reported to adversely affect image quality on real-time elastograms (3). 237
238
R.A. LEVINE
Ultrasound elastography has been used to analyze nodules and predict malignant potential in breast, prostate (4), pancreas, and lymph nodes (5). It has been used to measure liver fibrosis (5), as well as stiffness of cardiac tissue following myocardial infarction. It remains an ancillary technique in these organs with clinical application still predominantly in the research setting. Early experience with breast nodules has shown great promise for elastography in prediction of malignant potential. Early reports showed close to 100% sensitivity and specificity for this technique. However, as discussed below, this may be misleading and due to a bias in sample selection. A large study looking at breast elastography has shown a sensitivity of 86% and a specificity of 90% (7). These values suggest that the technique may be useful in selecting which nodules require biopsy. However, as with all ultrasonographic characteristics, the sensitivity may not be sufficient to eliminate the need for biopsy of a nodule. In 2007, there were only two published studies regarding thyroid elastography. Both reported extremely promising results. Lyshchik et al. (8) performed a prospective study involving 52 thyroid nodules in 31 consecutive patients. Of the 52 nodules, 22 were malignant and 30 were benign. They utilized both real-time elastography, and off-line processed ultrasound elastograms. The strain of the nodule was compared to the strain of the surrounding normal thyroid tissue. The results for the off-line analysis were far superior to the real-time studies. They reported that the off-line processed elastogram was the strongest independent predictor of thyroid gland malignancy, with 96% specificity and 82% sensitivity. However, they also report that off-line strain image processing is time-consuming and labor intensive. Rago et al. (9) recently published a study of real-time ultrasound elastography in 96 consecutive patients with a solitary thyroid nodule undergoing surgery for compressive symptoms or suspicion of malignancy on prior fine-needle aspiration biopsy. Tissue stiffness was scored from 1 to 5 based on subjective analysis of the elastogram image. They reported that scores of 1 or 2 were found in 49 cases—all benign lesions. A score of 3 was found in thirteen cases with one case of carcinoma, and twelve from benign lesions. Thirty cases had scores of 4 or 5, and all were carcinomas. They reported a sensitivity of 97% and a specificity of 100% for a score of 4 or 5 being predictive of malignancy. The general applicability of both of the above studies is limited due to selection bias. In the studies, the incidence of
ULTRASOUND ELASTOGRAPHY OF THE THYROID
239
malignancy was 31 – 43%. Most studies show an incidence of malignancy of 2–5% in nodules selected for biopsy, and the incidence of malignancy is much lower in all unselected nodules (1). The predictive value of the test will vary with the incidence of malignancy in the population studied, and will need to be studied in an unselected population with thyroid nodules. A single study has evaluated the utility of elastography in the assessment of cervical lymph nodes suspected of containing metastatic cancer. Lyshchik et al. (5) examined 141 peripheral lymph nodes in 43 consecutive patients referred for surgical treatment of suspected thyroid or hypopharyngeal cancer. By comparing the strain of lymph nodes and surrounding neck muscles, a strain index was calculated. An index cutoff of 1.5 resulted in a 98% specificity and 85% sensitivity. The results were superior to conventional grey-scale ultrasound criteria utilizing the short to long axis ratio. Figs. 14.1 – 14.5 provide illustration of the images provided by elastography. Fig. 14.1 shows a papillary carcinoma of the thyroid with peripheral psammomatous calcification. As would be expected, this nodule was very firm on physical examination. The figure shows that areas of the nodule are very hard, and very suggestive of a malignant nodule. Note the scale at the right edge of the image indicating a color scale ranging from soft (SF) to hard (HD). Fig. 14.2 shows a nodule with soft consistency on the elastogram. Figs. 14.3 and 14.4 are from the same patient. Bilateral nodules were present. The
FIG. 14.1. This nodule has several suspicious features including peripheral microcalcifications, scalloped margins and hypervascularity on power Doppler (not shown). The elastogram shows significant areas indicated as “Hard” (see scale located on the right of the image. HD = Hard, SF = Soft.) The pathology confirmed a papillary carcinoma
240
R.A. LEVINE
FIG. 14.2. This hypoechoic nodule appears very soft on elastography, suggesting a lower risk of malignancy
FIG. 14.3. This 34 year old woman had bilateral nodules. The figure shows the left nodule, which was the larger of the two, and was previously biopsied with benign cytology. The elastogram shows the nodule to be predominately soft
larger left (dominant) nodule had previously demonstrated benign aspiration cytology. It had a soft texture on elastography. The right-sided nodule had more suspicious sonographic features (echotexture, irregular margins, and microcalcifications), as well as a hard testure at elastography, and proved to be a papillary carcinoma. Fig. 14.5 is from a 38-year old male with diffusely multifocal infiltrative tall cell variant of papillary carcinoma—stage T3N1BM0. Multiple areas of hard tissue are shown on the elastogram.
ULTRASOUND ELASTOGRAPHY OF THE THYROID
241
FIG. 14.4. In the same patient as Image 14.3, the right nodule was smaller, but had several suspicious features including a heterogeneous echotexture, irregular margins, and microcalcifications. The elastogram shows the nodule to have a hard composition. Fine needle aspiration cytology demonstrated papillary carcinoma
FIG. 14.5. This image is from a 38 year old male with diffusely multifocal infiltrative tall cell variant of papillary carcinoma, stage T3N1BM0. Multiple areas of hard tissue are shown on the elastogram
Not all nodules are amenable to elastography. Due to an inability of the ultrasound beam to penetrate the nodule, elastography cannot be performed on nodules with peripheral rim calcification. Complex nodules with a large cystic component may provide misleading results because the elasticity is more dependent on the liquid content than the solid portion of the nodule. Rago included four cases in which intranodular cysts
242
R.A. LEVINE
made up less than 20% of the total nodule volume and did not appear to adversely affect the results. Small nodules probably can be measured accurately with elastography, but the limits of acceptable size have not been tested. The impact of background Hashimoto’s thyroiditis or other abnormalities of the thyroid parenchyma has not been adequately assessed. There are two potential roles for elastography in the analysis of thyroid nodules. The first is indicating the need for biopsy in a nodule that otherwise would be considered low suspicion and not be biopsied. Current guidelines state that nodules smaller than 1.5 cm with no suspicious features (indistinct margins, microcalcifications, taller than wide shape, extreme hypoechogenicity, or strong vascular flow) can be monitored without biopsy. However, if the positive predictive value of elastography is high, an otherwise nonsuspicious nodule demonstrated to be hard by elastography should be biopsied. On the other hand, if the negative predictive value of elastography is adequate, it could be used to help determine which nodules can be safely observed without biopsy. Approximately 4% of the population has a palpable thyroid nodule, and over 50% has a small nodule detectable by ultrasound (1). Clearly, all nodules found by physical examination—or as an incidental finding during other neck studies—cannot undergo fine needle biopsy. Any technique used to determine which nodules can be safely monitored without biopsy needs to have a sensitivity close to 100%. In each of the studies reported to date, the prevalence of malignancy far exceeded that found in an unselected population with thyroid nodules. While the initial reports of elastography of thyroid lesions are very exciting, additional large studies on unselected populations with thyroid nodules will be needed to determine whether the technique has sufficient sensitivity and predictive value to obviate the need for biopsy. References 1. Cooper D, Doherty G, Haugen B, et al (2006) American Thyroid Association Guidelines Task force 2006 Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid Feb 2:2–33 2. Bae U, Dighe M, Dubinsky T, et al (2007) Ultrasound Thyroid Elastography Using Carotid Artery Pulsation: Preliminary Study. J Ultrasound Med June 26(6): 797–805 3. Lyshchik A, Tatsuya H, Ryo A, et al (2004) Ultrasound Elastography in Differential Diagnosis of Thyroid Gland Tumors: Initial Clinical Results. RSNA Abstract
ULTRASOUND ELASTOGRAPHY OF THE THYROID
243
4. Pallwein L, Mitterberger M, Struve P et al (2007) Real-time Elastography for Detecting Prostate Cancer: Preliminary Experience. BJU Int. July 100(1):42–47 5. Lyshchik A, Higashi T, Asato R, et al (2007) Cervical Lymph Node Metastases: Diagnosis at Sonoelastography-Initial Experience. Radiology April 243(1):258–267 6. Friedrich-Rust M, Ong M, Herrman E, et al (2007) Real-Time Elastography for Noninvasive Assessment of Leiver Fibrosis in Chronic Viral Hepatitis. Am J Roentgenol. March 188(3):758–764 7. Itoh A, Venu E, Tohno E et al (2006) Breast Disease: Clinical applications of US Elastography for Diagnosis. Radiology May 239(2):341–350 8. Lyshchik A, Higashi T, Asato R, et al (2005) Thyroid Gland Tumor Diagnosis at US Elastography. Radiology 237(1):202–211 9. Rago T, Santini F, Scutari M, et al (2007) Elastography: New Developments in Ultrasound for Predicting Malignancy in Thyroid Nodules. J Clin Endocrinol Metab Aug 92:2917–2922
Index A A-mode images, 13–15 A-mode imaging, 2, 3, 6 Acoustic enhancement, 15–18 Acoustic impedance, 9 Acoustic shadowing, 15–17 Air microbubbles, 153–158 American Association of Clinical Endocrinologists (AACE), 6 American Institute of Ultrasound Medicine (AIUM), 6 Anesthesia, 102 Anterioposterior-totransverse diameter (A/T) ratio, 89 Attenuation, 22–23 Autoimmune thyroid disease (AITD), 63; see also Thyroiditis diagnosis, 63 Autonomously functioning thyroid nodules (AFNT); see also Nodules percutaneous ethanol injection, 173–174 percutaneous laser ablation, 213–215 Avascular nodules, 30, 31 Azimuthal plane, 104 B B-mode images, 15 B-mode imaging, 2, 3, 6 “Bag of marbles,” 66
Baskin, H. Jack, 6 Benign masses, 53–57; see also Nodules Biopsy; see also Fine-needle aspiration biopsy Doppler imaging prior to, 38, 40 C Calcification(s), 86 eggshell, 18–19 Cancer cystic papillary, 84 testicular (metastatic), 60 thyroid, 111, 133 postoperative surveillance for, 111–112 risk factors, 78–79 ultrasound of postoperative neck, 112–130 ultrasound of preoperative neck, 131–133 Cancer detection, 3–5, 60 commonest areas of detecting cancer, 113 Doppler ultrasound for, 30–33, 40, 41 follicular carcinoma, 30 indications for use of ultrasound for, 78–79 malignant masses, 57 sensitivities of sonographic features, 78 Cat’s eye artifact, 22, 23; see also “Comet tail” artifacts
245
246
INDEX
Closed suction, “free hand” technique, 106, 108 Color-flow Doppler (CFD), 28–30, 87–88 “Comet tail” artifacts, 20–23 Contrast-enhanced ultrasonography (CEUS), 152, 166, 168–169 in assessment of efficacy of thermal ablation treatments, 165–167 clinical use, 159–160 evaluation of nodules with CEUS timeintensity curves, 160–164 technical background, 153–159 Cysts, 53 D Doppler shift, 27, 28 Doppler ultrasound, 39–40 development, 5 for image clarification, 38–39 physical principles, 27–30 prior to biopsy, 38, 40 of thyroid nodules, 30–35 of thyroiditis, 35–38 Dussic, Karl Theodore, 2 E Echogenic strands, 69 Echogenicity, 82–85; see also Hypoechogenicity when it may be challenging to determine, 84 Edge artifacts, 19, 20 Eggshell calcification, 18–19 Elastography, 5–6, 237–242
nodules not amenable to, 241–242 roles in the analysis of nodules, 242 Enhancement: see Acoustic enhancement Esophageal diverticulum, 55 Ethanol injection: see Percutaneous ethanol injection F Fibrosis, 68 Fine-needle aspiration (FNA) of nodules, 78 palpation used to determine accuracy of, 78 Fine-needle aspiration biopsy (FNAB), 152, 219 ultrasound-guided, 4 Follicular carcinoma, 30 G Goiter, diffuse, 48–49 Graves’ disease, 5, 36, 50, 63, 64, 80, 81 Gray scale display, 3–4 “Ground glass” architectural pattern, 66, 67 H Halo, 86–87 Hashimoto’s lymphocytic thyroiditis, 79, 80 Hashimoto’s thyroiditis, 63–67, 71–73, 79 Hashitoxicosis, 36–38 Hemangioma, 58 Hemiagenesis of thyroid, 49–52 High intensity focused ultrasound (HIFU) ablation therapy, 219, 234 animal trials, 222–223
INDEX
first trials (proving feasibility), 223 second trial (adjustment of treatment parameters), 224–225 applications, 221–222 for benign nodules, 228 HIFU lesions, 227, 228 human trials, 225–226 human feasibility study, 226 ongoing studies, 228 operator’s interface, 227 rationale, 219–220 role in management of thyroid diseases, 228–234 technology/principle of, 220–222 for toxic nodules, 228 Hilar line, 114 Hilum, 116 “Honeycomb” pattern, 84 Hypoechogenicity, 64–65; see also Echogenicity Hypoechoic nodules, 90 I Isoechoic nodules, 88 L Levothyroxine therapy, 232–234 Lymph nodes, 38, 40, 115 characteristics, 115 enlarged/inflamed, 56–57, 125 malignant, 41, 117–123, 126–130, 132 Lymphoma, 73 M Mechanical index (MI), 154 Microbubbles, 153–158
247
Modified lateral neck dissection (MLND), 131 Muscle anomaly, 57 N Neck, normal postoperative, 113–114 “Needle only” technique, 108 Needles used for UGFNA, 101 Nodules, thyroid, 77, 132, 151; see also Autonomously functioning thyroid nodules; specific topics algorithm for cost-effective evaluation and treatment of, 233 avascular, 30, 31 evaluation with CEUS time-intensity curves, 160–164 hypoechoic, 90 isoechoic, 88 measuring volume of, 48 palpable, 77–78 treatment of benign, 230–234 ultrasound characteristics, 78, 81–82 calcifications, 86 change in size, 91–92 echogenicity, 82–85 elastography, 89–90; see also Elastography halo, 86–87 margins, 86 taller than wide, 89 vascularity, 87–88 vascular, 30 P Papillary carcinoma, 41 Parathyroid adenomas, 42, 136–138, 142–145, 214
248
INDEX
Parathyroid adenomas (continued) sonographic features, 139–140 Parathyroid glands, anatomy of, 135–136 Parathyroid hormone (PTH), 144, 146–148, 184 Parathyroid incidentaloma, 140, 142 Parathyroid lesions, ultrasound-guided FNA of, 144–148 Parathyroid (PT) cyst, 142, 145 Parathyroid ultrasound, technique of, 138–140 Percutaneous ethanol injection (PEI), 165–167 of cysts, 179–181 of lesions autonomous functioning nodules, 173–174 cold solid nodules, 173–174 cysts, 174–179 of neck lesions, 185–187 of parathyroid cysts, 184–185 side effects, 182–184 technical aspects, 179–181 Percutaneous laser ablation (PLA), 165–167, 192, 194–197, 215 advantages and limitations, 215 in autonomous nodules, 213–215 clinical results and indications, 203, 206–214 changes in nodule volume, 206–209 ellipsoid ablation, 196–197
history, 192 multiple fiber technique for laser ablation volume increase, 196 needle electrodes used for, 192, 193 PLA intervention, 197–206 side effects, 213 Power Doppler (PD), 28–30, 87–88 Primary hyperparathyroidism (PHPT), 135 localization studies, 136–138 Pseudonodules, 68, 70 Pulsed waves, 11 R Radiation exposure, 79 Radiofrequency (RF) ablation (RFA), 165, 191–192 Radioiodine, 231 Reflection, 13 Refraction, 22 Reverberation artifacts, 19–21 Ringdown artifact, 22; see also “Comet tail” artifacts S Shadowing: see Acoustic shadowing Sound and sound waves, 9–13, 25–26 sound wave propagation, 9–10 speed of sound, 9–11, 13 “Spongiform” pattern, 84, 85, 89 Squamous cell carcinoma, 60 Stepladder artifact, 22; see also “Comet tail” artifacts
INDEX
Strepta, 69 “Swiss cheese” appearance of thyroid, 68 T Testicular cancer, metastatic, 60 Three-dimensional (3D) ultrasound, 5 Thymus gland, undescended, 56 Thyroglossal duct, 53, 55 Thyroid; see also specific topics aberrant, 52–53 lateral, 54 anatomy (normal), 45–48 failed bifurcation of, 52, 53 longitudinal view of, 48 in transverse view, 46 Thyroid anomalies, 49–53 Thyroid disorders; see also specific disorders diffuse, 79–81 Thyroid lobe, measurement of, 47 Thyroid nodule (N): see Nodules Thyroidectomy, 124–127, 230–231 Thyroiditis, 63, 73 atrophic, 73 pathology, 64 postpartum, 36, 37 subacute, 37, 38, 81 ultrasonography, 64–73 Doppler ultrasound, 35–38
249
Thyrotoxicosis, 35–36 Tomogram, 3 U Ultrasound, thyroid; see also specific topics history, 1–7 resolution, 4 Ultrasound elastography: see Elastography Ultrasound-guided FNA (UGFNA), 97–98, 109 aspiration and non-aspiration techniques, 106–109 materials, 100–102 of micronodules, 98–99 parallel approach, 104–105 of parathyroid lesions, 144–148 perpendicular approach, 106, 107 preparation, 99 technique, 103–104 Ultrasound image, creation of an, 13–14 Ultrasound imaging, usefulness of artifacts in, 14–26 Ultrasound technique, 45 W Whole body scan (WBS), 111, 112 Z Zajdela technique, 108