Manual Therapy Journal - Volume 13, Issue 3, Pages 181-276 (June 2008)

VOLUME 13 NUMBER 3 PAGES 181–276 June 2008 Editors International Advisory Board Ann Moore PhD, GradDipPhys, FCSP, Cer...

Author: Editors: Ann Moore and Gwen Jull

17 downloads 1005 Views 3MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

VOLUME 13 NUMBER 3 PAGES 181–276 June 2008

Editors

International Advisory Board

Ann Moore PhD, GradDipPhys, FCSP, CertEd, FMACP Clinical Research Centre for Health Professions University of Brighton Aldro Building, 49 Darley Road Eastbourne BN20 7UR, UK

K. Bennell (Victoria, Australia) K. Burton (Hudders¢eld, UK) B. Carstensen (Frederiksberg, Denmark) M. Coppieters (Queensland, Australia) E. Cruz (Setubal Portugal) L. Danneels (Mar|¤ akerke, Belgium) S. Durrell (London, UK) S. Edmondston (Perth, Australia) J. Endresen (Flaktvei, Norway) L. Exelby (Biggleswade, UK) D. Falla (Aalborg, Denmark) T.W. Flynn (Denver, CO, USA) J. Greening (London, UK) C. J. Groen (Utrecht,The Netherlands) A. Gross (Hamilton, Canada) T. Hall (West Leederville, Australia) W. Hing (Auckland, New Zealand) M. Jones (Adelaide, Australia) S. King (Glamorgan, UK) B.W. Koes (Amsterdam,The Netherlands) J. Langendoen (Kempten, Germany) D. Lawrence (Davenport, IA, USA) D. Lee (Delta, Canada) R. Lee (Brighton, UK) C. Liebenson (Los Angeles, CA, USA) L. Ma¡ey-Ward (Calgary, Canada) E. Maheu (Quebec, Canada) C. McCarthy (Coventry, UK) J. McConnell (Northbridge, Australia) S. Mercer (Queensland, Australia) D. Newham (London, UK) J. Ng (Hung Hom, Hong Kong) S. O’Leary (Queensland, Australia) L. Ombregt (Kanegem-Tielt, Belgium) N. Osbourne (Bournemouth, UK) M. Paatelma (Jyvaskyla, Finland) N. Petty (Eastbourne, UK) A. Pool-Goudzwaard (The Netherlands) M. Pope (Aberdeen, UK) G. Rankin (London, UK) D. Reid (Auckland, New Zealand) A. Rushton (Birmingham, UK) C. Shacklady (Manchester, UK) M. Shacklock (Adelaide, Australia) D. Shirley (Lidcombe, Australia) V. Smedmark (Stenhamra, Sweden) W. Smeets (Tongeren, Belgium) C. Snijders (Rotterdam,The Netherlands) R. Soames (Dundee, UK) P. Spencer (Barnstaple, UK) M. Sterling (St Lucia, Australia) P. Tehan (Victoria, Australia) M. Testa (Alassio, Italy) M. Uys (Tygerberg, South Africa) P. van der Wur¡ (Doorn,The Netherlands) P. van Roy (Brussels, Belgium) B.Vicenzino (St Lucia, Australia) H.J.M.Von Piekartz (Wierden,The Netherlands) M.Wallin (Spanga, Sweden) M.Wessely(Paris, France) A.Wright (Perth, Australia) M. Zusman (Mount Lawley, Australia)

Gwendolen Jull PhD, MPhty, Grad Dip ManTher, FACP Department of Physiotherapy University of Queensland Brisbane QLD 4072, Australia Associate Editor’s Darren A. Rivett PhD, MAppSc, (ManipPhty) GradDipManTher, BAppSc (Phty) Discipline of Physiotherapy Faculty of Health The University of Newcastle Callaghan, NSW 2308, Australia Tim McClune D.O. Spinal Research Unit. University of Hudders¢eld 30 Queen Street Hudders¢eld HD12SP, UK Editorial Committee Masterclass Editor Karen Beeton PhD, MPhty, BSc(Hons), MCSP MACP ex o⁄cio member Associate Head of School (Professional Development) School of Health and Emergency Professions University of Hertfordshire College Lane Hat¢eld AL10 9AB, UK Case reports & Professional Issues Editor Je¡rey D. Boyling MSc, BPhty, GradDipAdvManTher, MCSP, MErgS Je¡rey Boyling Associates Broadway Chambers Hammersmith Broadway LondonW6 7AF, UK Book Review Editor Raymond Swinkels MSc, PT, MT Ulenpas 80 5655 JD Eindoven The Netherlands

Visit the journal website at http://www.intl.elsevierhealth.com/journals/math doi:10.1016/S1356-689X(08)00066-0

ARTICLE IN PRESS

Manual Therapy 13 (2008) 181–182 www.elsevier.com/locate/math

Editorial

Specialization in musculoskeletal physiotherapy—the Australian model In many countries around the world, a number of factors over the past decade or more have stimulated changes in the tasks of the health workforce and methods of healthcare delivery. One outcome is that physiotherapists are assuming greater and more diverse roles and responsibilities in provision of healthcare and many are providing specialist expertise to patients in both the private and public sectors. There is a demand for specialists in the sub-disciplines of physiotherapy in this current environment. There is also a need for clear and attainable career pathways for clinicians to provide, and to be recognized, for these specialist services. Around the world, there are different processes and requirements for clinical specialization in musculoskeletal physiotherapy. In this editorial we present an overview of the specialization process in Australia, where the past 5 years in particular have witnessed an impetus in the process and collaborative development of a revised model for clinical specialization. The Australian College of Physiotherapists is the arm of the Australian Physiotherapy Association responsible for the specialization process and for awards of Fellowship in Specialization in the sub-disciplines of physiotherapy, inclusive of musculoskeletal physiotherapy. Rather than being a mid or end point of a career, the strong vision of the new process is that specialization is the beginning of a career as a specialist in a ﬁeld of physiotherapy practice, akin to the medical model. It is possible for a physiotherapist in the new model to fulﬁl all requirements of the process and 5 years after graduation, to begin a specialist career. The specialization process in physiotherapy in Australia is designed and operates for all sub-disciplines of physiotherapy, but musculoskeletal physiotherapy is highlighted in this instance. The revised model of the specialization process considers the new graduate. The initial phase after graduation requires at least 2 years of clinical experience and relevant professional development activities in musculoskeletal physiotherapy. The middle phase of the process requires successful completion of a university postgraduate coursework masters 1356-689X/$ - see front matter r 2008 Published by Elsevier Ltd. doi:10.1016/j.math.2008.03.005

program in musculoskeletal physiotherapy, which has been accredited by Musculoskeletal Physiotherapy Australia (MPA). Alternately, a physiotherapist may successfully complete a more independent program of continuing professional development and sit the MPA challenge examinations in both clinical and theoretical domains. The aim of either route is to attain what is termed ‘titled’ membership status of the MPA and Australian Physiotherapy Association. Following attainment of titled membership, the musculoskeletal physiotherapist can become an associate member of the Australian College of Physiotherapists and, as a candidate for specialization, embark on the ﬁnal stages of the specialization process. The ﬁnal stage of the specialization process involves a 2 year training period under the direction of the College as well as ﬁnal specialist examinations. In the 2 year training period, the candidate works in clinical practice and undertakes a mentored program of professional development, which has a strong emphasis on clinical development. During this training period, the candidate is required to provide evidence of development of quality and specialist-level practice as well as professional leadership through contributions to education of students or peers, a commitment to life-long learning and participation in and support of research activity. Once this period is completed and, on the recommendation of the College mentor, the candidate can present for the ﬁnal examinations for specialization. These examinations consist of clinical examinations of two patients evaluated and treated over 2 days as well as an oral examination to allow the candidate to demonstrate their clinical expertise and knowledge. An alternate pathway for the training process is successful completion of a postgraduate clinical doctorate in the ﬁeld of specialization. The College works with the Australian universities offering postgraduate clinical doctorates to ensure the requirements within the clinical doctorate parallel those of the specialization process before the candidate sits for the ﬁnal examinations in the Fellowship process.

ARTICLE IN PRESS 182

Editorial / Manual Therapy 13 (2008) 181–182

In the transition to the new model, consideration has been given to existing titled members of the MPA who can provide evidence that they are working at an advanced level in clinical practice and that they have fulﬁlled the requirements of the specialization training period. In recognition of their expertise, a provision was made for a 3 year period (ending December 2009) for current musculoskeletal physiotherapists to apply immediately for the ﬁnal examination process. In 2007, the College inducted 26 new Specialist Musculoskeletal Physiotherapists and it is estimated that between 30 and 40 candidates will undertake the ﬁnal examinations in 2008. In addition, calls will be made this year for applications to join the training program.

There is a need for physiotherapists to have high-level expertise to provide leadership in the delivery of specialist assessment and care of patients with musculoskeletal disorders. Specialization provides the avenue for training and importantly the recognition of the specialist skills of physiotherapists and has been enthusiastically embraced in the Australian context.

Editors Gwendolen Jull & Ann Moore (Gwendolen Jull is President, Australian College of Physiotherapists)

ARTICLE IN PRESS

Manual Therapy 13 (2008) 183–191 www.elsevier.com/locate/math

Original article

Standing balance: A comparison between idiopathic and whiplash-induced neck pain Sandra Field, Julia Treleaven, Gwendolen Jull Division of Physiotherapy, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, 4072 Queensland, Australia Received 24 April 2006; received in revised form 7 August 2006; accepted 1 December 2006

Abstract Disturbances of balance have been found both in patients with whiplash-associated disorders and idiopathic neck pain. This study directly compared balance between these groups to determine if neck pain precipitated by trauma resulted in greater or different balance impairments. The study was a comparative, observational design. Thirty subjects with whiplash, 30 with idiopathic neck pain and 30 healthy controls, took part in the study. Subjects performed balance tests in comfortable, narrow and tandem stances. Balance disturbances (sway energy and/or root mean squared (RMS) amplitude) were evident in several tests between subjects with neck pain and controls. Direct comparison between the neck pain groups revealed that the whiplash group had signiﬁcantly greater sway energy and RMS amplitude than the idiopathic group in comfortable stance tests on a soft surface (F44.4, po0.04). Further, the whiplash group had greater RMS, but signiﬁcantly less sway energy than the idiopathic group in most narrow stance tests in the anterior posterior direction F45.8, po0.02). Both neck pain groups were also signiﬁcantly less able to complete the eyes closed, tandem test compared to control subjects. In conclusion, the study has found that balance deﬁcits exist in both subjects with whiplash-associated disorders and idiopathic neck pain compared to controls; however, differences in balance strategies may exist between the neck pain groups. Overall, subjects who have experienced trauma appear to have greater balance disturbances. r 2007 Elsevier Ltd. All rights reserved. Keywords: Balance; Whiplash; Neck pain; Proprioception; Wavelet analysis; RMS

1. Introduction Recently, neck pain has been broadly classiﬁed as idiopathic or trauma-induced neck pain (e.g. neck pain from a whiplash injury)1 as is not possible to make a deﬁnitive pathoanatomical diagnosis in most cases. Such a classiﬁcation recognizes a difference in mechanism of onset, which implies that with the involvement of trauma, there might be differences in the nature or magnitude of pathophysiological features between the two neck pain types. Recent research is supporting this Corresponding author. Tel./fax: +61 7 3365 1622.

E-mail address: [email protected] (J. Treleaven). Australian Acute Musculoskeletal Pain Guidelines. Evidence Based Management of Acute Musculoskeletal Pained. Brisbane: Australian Academic Press, 2003 /http://www.nhmrc.gov.auS. 1

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.12.005

assumption. For example, Scott et al. (2005) found greater and more widespread mechanical and thermal hyperalgesia in a chronic whiplash group compared to an idiopathic neck pain group. There is also some preliminary evidence to suggest that measured impairments reﬂective of the sensorimotor control system (eye movement control and cervical joint position error) are of greater magnitude when trauma has precipitated the neck pain (Kristjansson et al., 2003; Michaelson et al., 2003; Tjell et al., 2003). Our interest in this study is in balance disturbances. Patients with neck pain of both idiopathic and whiplash origin have been found to have deﬁcits in standing balance (Karlberg et al., 1995, 1996a, b; Koskimies et al., 1997; McPartland et al., 1997; El-Kahky et al., 2000; Michaelson et al., 2003; Schieppati et al., 2003; Sjostrom et al., 2003; Treleaven

ARTICLE IN PRESS 184

S. Field et al. / Manual Therapy 13 (2008) 183–191

et al., 2005a). In the absence of vestibular pathology, this has been attributed to altered cervical somotosensory input (Karlberg et al., 1996a). A link between neck pain and impairment in balance is suggested in several studies, which have demonstrated improvements in standing balance following treatment localized to the cervical spine (Alund et al., 1993; Fattori et al., 1996; Karlberg et al., 1996b; Persson et al., 1996). The co-occurrence of neck pain and balance disturbances is not surprising as the abundant cervical receptors in the muscles and joints of the cervical spine as well as their central and reﬂex connections to the vestibular, visual and postural control systems suggest that they have an important role in providing information for general postural control (Karlberg et al., 1995; Lekhel et al., 1998; Bove et al., 2002; Peterson, 2004). The association between disturbances to cervical afferentation and disturbances to standing balance has been demonstrated in several ways. At the extreme, sectioning of the cervical dorsal root ganglions or anaesthetization of deep neck structures causes severe ataxia (Ishikawa et al., 1998) and disequilibrium (DeJong and DeJong, 1977). Less extremely, neck muscle vibration, which primarily stimulates the muscle spindle afferents, increases body sway (Kavounoudias et al., 1999) and inﬂuences velocity and direction of gait and running (Bove et al., 2002; Courtine et al., 2003). Vibration of neck muscles has a greater inﬂuence on postural sway when compared to most other muscles in the body (Pyykko et al., 1989). Studies have also demonstrated the direct deleterious effects of neck extensor muscle fatigue on standing balance (Schieppati et al., 2003; Gosselin et al., 2004), and McPartland et al. (1997) determined a signiﬁcant correlation between poor balance control and fatty inﬁltration of the rectus capitis posterior minor muscle. There are several mechanisms via which neck pain might cause altered cervical somotosensory input and integration to the postural control system. These include: direct trauma to the cervical receptors or the functional impairment of cervical muscle and joint receptors that may result from the trauma (Heikkila and Astrom, 1996); inﬂammatory mediators may activate chemosensitive nerve endings in joints and muscles leading to altered muscle spindle activity (Wenngren et al., 1998; Thunberg et al., 2001); the effects of pain on nociceptor and mechanoreceptor activity locally at the spinal cord and within the central nervous system may inﬂuence the central modulation of afferent input and in consequence, neuromuscular and postural control (Le Pera et al., 2001; Ageborg, 2002). We questioned whether balance disturbances were greater when neck pain was associated with trauma (whiplash in this instance) when compared to an idiopathic or non-trauma origin. In trauma, a larger number of structures and thus sources of afferent input

could be damaged and a greater number of mechanisms may be involved. As a consequence, there may be differences in either the nature or magnitude of pathophysiological features between the two neck pain types, which may have implications for clinical assessment and management. This study was conducted to provide a direct comparison of balance responses between these groups to inform practice. We hypothesized that there would be differences between the groups and balance deﬁcits would be more pronounced in the whiplash group.

2. Methods 2.1. Participants Thirty subjects with idiopathic neck pain, 30 subjects with persistent whiplash associated disorders and 30 healthy control subjects participated in the study. Idiopathic neck pain and healthy subjects were recruited from advertisements within the community. Whiplash subjects were recruited from eligible consecutive patients attending a Whiplash Research Unit and were included if categorized as WAD II according to the Quebec Task Force classiﬁcation (Spitzer et al., 1995). The age range of subjects was restricted to 18–45 years inclusively, to exclude the variable of aging on balance measures (Speers et al., 1998). To be included in either neck pain group, pain was to be of greater than 3 months duration with a score of at least 10 out of 100 on the Neck Disability Index (NDI) (Vernon, 1996). In our previous research on subjects with whiplashassociated disorders, we have determined that those reporting dizziness and unsteadiness have greater deﬁcits in balance than those not complaining of these symptoms (Treleaven et al., 2005a). To better standardize the three groups, no volunteer was considered if they reported dizziness or unsteadiness. To control for other variables, which could inﬂuence balance responses, no subject was considered if they had current or past lower limb problems, known vestibular pathology, signiﬁcant visual or hearing deﬁcits, neurological deﬁcits, Type II diabetes, abnormal blood pressure, or diagnosed psychiatric disorders. Whiplash subjects were also excluded if there was a loss of consciousness at the time of injury and volunteers for the idiopathic neck pain and control groups were not considered if they had a past history of whiplash. All volunteers accepted into the study were asked to refrain from taking medication such as antipsychotic and narcotic medication that may inﬂuence balance and from consuming alcohol for 24 h prior to testing (Alund et al., 1991). Ethical clearance for the study was obtained from the institutional Medical Ethics Committee and all procedures

ARTICLE IN PRESS S. Field et al. / Manual Therapy 13 (2008) 183–191

were conducted according to the Declaration of Helsinki. All the subjects gave their written informed consent to undertake the study. 2.2. Measurements Questionnaires were administered to collect demographic data, levels of current resting pain (VAS, Huskisson, 1983), and self-reported neck disability (NDI, Vernon, 1996). Neck pain subjects also completed the State Trait Anxiety Inventory—short form (Marteau and Bekker, 1992), which included measures of both ‘state’ (how they felt at the time of the investigation) and ‘trait’ (how they generally felt) anxiety levels. Scores were prorated to the full score. Both features were scored from 20 (little anxiety) to 80 (maximum anxiety). 2.3. Computerized posturography Postural sway was measured on a 40 60 cm stable computerized force platform. Centre of pressure force changes in the medio-lateral (ML) and antero-posterior (AP) directions were measured, over time, by ground reaction forces registering on four, corner strain gauges mounted within the ﬂoor. The force changes were converted to electrical signals by the force transducers and charge ampliﬁers. An analogue low pass ﬁlter was used to restrict the frequency content on the signals to within 0–5 Hz. The force signals were then AD converted at a sampling rate of 15 Hz and recorded using a Labview (2000 National Instruments) programme. The raw signal was produced both numerically and graphically. A modiﬁed Clinical Test of Sensory Integration and Balance (CTSIB) was used to assess standing balance (Shumway-Cook and Horak, 1986). Ten tests were performed in total (Shumway-Cook and Horak, 1986). The tests consisted of four tests in comfortable stance—standing on a ﬁrm surface and on a soft surface (a piece of high density 10-cm-thick foam rubber placed on the force platform) with eyes open and eyes closed. These four tests were then repeated in narrow stance. Each subject also performed two tests in tandem stance—ﬁrm surface with eyes open and closed. Only pass/failure rates were recorded for tandem tests, as it was known from previous research that limited numbers of subjects from the neck pain groups would be able to complete these tests (Treleaven et al., 2005a). 2.4. Procedure Inclusion/exclusion criteria were determined via telephone interview and questionnaires were completed at the time of testing. For the tests of balance, the subjects stood on the force platform. They were instructed to stare at a dot clearly marked on the wall at a distance of 1.5 m and asked to stand as steadily as possible with

185

their arms by their sides. Consistent clear instruction was given for each test. Each subject performed ten tests. Foot position for comfortable stance was repositioned exactly using a paper trace as described by McIlroy and Maki (1997). For narrow stance, the subjects were asked to place the middle of their right foot to the right of the marked centre point of the force plate, their left foot was placed parallel and as close as possible to their right foot. For tandem stance, the dominant foot, deﬁned as that which would be most likely used to kick a ball, was placed directly behind the non-dominant foot (McPartland et al., 1997; Riemann and Guskiewicz, 2000). One 30-s trial was performed for each balance condition. For all tests, an inability to stand without losing balance for a 30 s time period was recorded as failure to complete the particular test. 2.5. Statistical analysis Failure rates and percentages were calculated for tandem stance conditions. Fischer’s exact test was used to determine any signiﬁcant differences between groups. Wavelet analysis using Daubechies ﬁlter 6 was chosen for the analysis of postural sway, as our previous studies of balance in comfortable stance had determined that this analysis was better able to distinguish between whiplash and asymptomatic subjects than the measure of total sway distance (Treleaven et al., 2005b). Analysis was conducted for both the AP and ML traces in comfortable and narrow stance tests and both directions were considered separately. Normality of the data set was assessed using Q–Q plots to verify parametric test use and data were logged. Differences between the chronic neck pain groups for age, VAS, anxiety (state and trait) and NDI scores were assessed initially using a series of one-way ANOVA’s. Group differences due to signal energies were examined using a generalized linear model, MANOVA. Current and general anxiety levels, the NDI, current pain level (VAS) and age were included as separate factors in the MANOVA for both neck pain groups. Where these had a signiﬁcant inﬂuence on within subjects’ balance measures, they were included into the ﬁnal between groups analysis as co-variates.

3. Results Group characteristics for age, gender, VAS, NDI, anxiety (state and trait) scores are presented in Table 1. There were no signiﬁcant between group differences for age (p ¼ 0.20) and anxiety scores (state, p ¼ 0.47; trait p ¼ 0.12). The whiplash group had signiﬁcantly greater resting pain (VAS p ¼ 0.05) and NDI scores (p ¼ 0.00) and these were included as co-variates in analyses of the neck pain groups.

ARTICLE IN PRESS S. Field et al. / Manual Therapy 13 (2008) 183–191

186

3.1. Comfortable stance

comparison between the whiplash and idiopathic groups demonstrated that the whiplash subjects had signiﬁcantly greater sway energy in tests of both eyes open (F ¼ 4.4, p ¼ 0.04) and closed (F ¼ 5.7, p ¼ 0.02) on the soft surface in the ML direction.

The means, standard errors and the signiﬁcance between group differences for the logged energy values are presented for each test condition in the anterior posterior and medial lateral directions of the tests in comfortable stance (Fig. 1a and b) for the whiplash, idiopathic neck pain and control groups. As a general observation, there was a trend for more impaired balance (greater mean total energy of sway) for both the whiplash and idiopathic neck pain groups compared to the control group. Differences between the whiplash and control subjects were signiﬁcant only in the tests of eyes open, ﬁrm surface and eyes closed, soft surface in the AP direction and eyes open and closed on the soft surface in the ML direction (Fig. 1a and b). Direct

3.2. Narrow stance The means, standard errors and the signiﬁcant between group differences for the logged energy values are presented for each test condition in the AP and ML directions for the tests in narrow stance (Fig. 2a and b) for the whiplash, idiopathic neck pain and control groups. In narrow stance, subjects with idiopathic neck pain had greater sway energy in most narrow stance tests in the AP direction when compared to both subjects with whiplash and control subjects. Speciﬁcally differences between the subjects with whiplash and subjects with idiopathic neck pain revealed greater energy of sway in the idiopathic neck pain subjects for all tests in the AP direction in narrow stance apart from the eyes open on the ﬁrm surface condition (F45.8, po0.02). Conversely, in the narrow stance test of eyes open on a ﬁrm surface in the ML direction, subjects with whiplash had signiﬁcantly greater energy than the idiopathic neck pain subjects (F ¼ 5.0, p ¼ 0.03). There were no other signiﬁcant differences between the whiplash subjects and the control subjects in energy of sway in narrow stance. These results in narrow stance for the whiplash subjects were unexpected and against the general trend observed in comfortable stance. In order to determine if this was a true ﬁnding or a factor of the type of analysis

Table 1 Subject demographics for the control, idiopathic neck pain and whiplash groups

Age (years) Gender (% female) Resting pain (VAS) NDI (%) Anxiety State (/80) Trait (/80)

Control (n ¼ 30) Mean (SE)

Idiopathic (n ¼ 30) Mean (SE)

Whiplash (n ¼ 30) Mean (SE)

26.8 (1.3) 77 — — — — —

27.9 (1.3) 77 2.2 (0.2) 21.5 (1.4)

30.3 (1.3) 80 3.2 (0.4)* 36.9 (2.8)*

34 (2) 41(2)

32 (2) 45 (2)

*po0.05.

a

b 2

*

*

* Controls IDP WAD

1.8 1.6

*

*

2 1.8 1.6 Logged Energy

1.4 Logged Energy

*

1.2 1 0.8 0.6

1.4 1.2 1 0.8 0.6

0.4

0.4

0.2

0.2 0

0 EOF

ECF

EOS

ECS

EOF

ECF

EOS

ECS

Test Test *=p<0.05 significant difference between subjects EOF= Eyes open firm, ECF= Eyes closed firm, EOS= Eyes open soft, ECS= Eyes closed soft Fig. 1. Comparison of total energy between subject groups for comfortable stance tests in the anterior posterior and medial lateral direction.

ARTICLE IN PRESS S. Field et al. / Manual Therapy 13 (2008) 183–191

*

*

*

*

*

*

*

2.5

2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

*

* Controls IDP WAD

2 Logged energy

Logged Energy

*

187

1.5 1 0.5 0

EOF

ECF

EOS

ECS

EOF

ECF

EOS

ECS

Test Test *=p<0.05 significant difference between subjects EOF= Eyes open firm, ECF= Eyes closed firm, EOS= Eyes open soft, ECS= Eyes closed soft

Fig. 2. Comparison of total energy between subject groups for narrow stance tests in the anterior posterior and medial lateral direction.

a

b * 4.5

*

*

*

*

*

8

4

7

3.5

6

3

RMS (mm)

RMS (mm)

*

*

2.5 2 1.5

5 4 3

1

2

0.5

1

0

Controls IDP WAD

0 EOF

ECF

EOS

ECS

Test

EOF

ECF

EOS

ECS

Test

*=p<0.05 significant difference between subjects EOF= Eyes open firm, ECF= Eyes closed firm, EOS= Eyes open soft, ECS= Eyes closed soft Fig. 3. Comparison of root mean square distance between subject groups for comfortable stance tests in the anterior posterior and medial lateral direction.

of postural sway, the data were reanalyzed post-hoc using the root mean square (RMS) distance travelled in millimetres in both the AP and ML directions. Amplitude was measured, as together with measures of frequency and velocity, amplitude is considered an important component to reﬂect the strategy of postural control (Dault et al., 2001). RMS was chosen to demonstrate the average amplitude traveled by the centre of pressure as it is a robust measure of amplitude (Prieto et al., 1996; Rocchi et al., 2004) and has demonstrated differences in postural control between clinical populations (Rocchi et al., 2002). The means,

standard errors and the signiﬁcant between group differences for the RMS values are presented for each test condition in the AP and ML directions of the tests in comfortable stance (Fig. 3a and b) and narrow stance (Fig. 4a and b) for the whiplash, idiopathic neck pain and control groups. The results of this analysis revealed similar results to the total energy analysis in comfortable stance; however, in narrow stance the results were opposite to those found for total energy. Subjects with whiplash demonstrated signiﬁcantly greater RMS in the AP direction in most tests in narrow stance compared to control subjects (F44.4, po0.04) and in the test of eyes

ARTICLE IN PRESS S. Field et al. / Manual Therapy 13 (2008) 183–191

188

a

b 9

*

8

Controls IDP WAD

7 6

12

*

10 RMS (mm)

RMS (mm)

*

*

5 4 3 2

8

* Controls IDP WAD

6 4 2

1 0

0 NEOF

NECF

NEOS

NECS

NEOF

NECF

NEOS

NECS

Test

Test

*=p<0.05 significant difference between subjects N= Narrow, EOF= Eyes open firm, ECF= Eyes closed firm, EOS= Eyes open soft, ECS= Eyes closed soft Fig. 4. Comparison of root mean square distance between subject groups for narrow stance tests in the anterior posterior and medial lateral direction.

open on the ﬁrm surface, values for the whiplash subjects were also signiﬁcantly greater than those for the idiopathic neck pain group (F ¼ 5.5, p ¼ 0.005) (Fig. 4). Further, and opposite to the energy results, there were no signiﬁcant RMS differences between the idiopathic neck pain and control groups.

Table 2 Percentages of subjects who failed the tandem stance test conditions for control, idiopathic and whiplash groups Tandem stance test

Control (n ¼ 30)

Idiopathic (n ¼ 30)

Whiplash (n ¼ 30)

Eyes open (ﬁrm) Eyes closed (ﬁrm)

3.33 10.0

3.33 36.7*

13.33 36.7*

3.3. Tandem stance *po0.05.

The number of subjects who were unable to complete the 30-s tandem stance test for the control, idiopathic and whiplash groups and the probability of difference of failure rates between groups for each test is depicted in Table 2. In tests in tandem stance, both the neck pain groups lost balance signiﬁcantly more often in the test without vision. There was no signiﬁcant difference between the neck pain groups on their ability to complete tandem stance tests.

4. Discussion The results of this study conﬁrm that neck disorders are associated with deﬁcits in standing balance when compared to asymptomatic control subjects, whether the neck pain is of idiopathic or whiplash in origin; however, overall the impairments in balance are generally greater in subjects with neck pain following a whiplash injury. The differences in sway energy values determined for the AP and ML directions in comfortable stance for the asymptomatic control and whiplash groups reﬂect those which we have previously found for these subjects (Treleaven et al., 2005a). There was a trend for differences in energy values between the idiopathic neck

pain and control groups but these failed to reach signiﬁcance. In tandem stance, both idiopathic neck pain and whiplash subjects had difﬁculty completing tandem stance positions in agreement with other studies (Michaelson et al., 2003; Treleaven et al., 2005a). In narrow stance, the subjects with idiopathic neck pain had signiﬁcantly higher energy in the AP direction than control subjects and, unexpectedly, the subjects with whiplash presented with similar energy values to control subjects. This ﬁnding in narrow stance for the whiplash subjects was against the general trends in comfortable and tandem stance. Our previous research had determined that the measurement of total energy of sway (wavelet analysis) was more sensitive than total sway distance to determine impairments in balance in whiplash subjects, but this study had only considered comfortable stance (Treleaven et al., 2005b). In the review of our ﬁndings, we reﬂected on whether the measure of mean total energy alone, was useful for narrow stance. It was reasoned that consideration of the average amplitude of the sway trace, by way of the RMS distance, together with total energy may provide further information to assist in the interpretation of these

ARTICLE IN PRESS S. Field et al. / Manual Therapy 13 (2008) 183–191

ﬁndings (Prieto et al., 1996; Rocchi et al., 2004). When the outcomes of both analyses are considered together, there were increases in both amplitude and energy of sway in the whiplash group compared to both controls and the idiopathic group in most comfortable stance tests. In narrow stance, however, subjects with whiplash had greater amplitude of sway but less total energy in the AP direction when compared to the idiopathic neck pain group. Conversely, subjects with idiopathic neck pain had similar amplitude to control subjects but greater energy in the AP direction in narrow stance. It is salient to note that the use of energy or RMS in isolation in narrow stance would have led to false negative results in both neck pain groups and highlights the importance of an adequate combination of feature selection in sway analysis. The combined results for energy and RMS may reﬂect altered strategies of control between the neck pain groups. It is possible that subjects with idiopathic neck pain manage comfortable stance conditions well, but experience difﬁculty in the AP direction in the more challenging stance condition of narrow stance, choosing a stiffening strategy, ((high energy (frequency) at a reduced amplitude (Dault et al., 2001)) to maintain stability. Subjects with whiplash may already have difﬁculty in comfortable stance and have less awareness of altered stability in narrow stance, further increasing the amplitude of sway using less overall energy than idiopathic neck pain subjects. Further research will be required to test this hypothesis. The results as a whole indicate that the deﬁcits in balance in subjects with neck pain are more widespread when the neck pain is trauma related. Whiplash subjects had greater energy and amplitude in comfortable stance and greater amplitude and possibly a less effective control strategy in narrow stance when compared to the idiopathic neck pain subjects. It could be argued that anxiety levels and increased pain in the whiplash group might contribute to these more widespread balance disturbances. However, anxiety levels were similar between the neck pain groups (Table 1) and while the whiplash group in this current study reported signiﬁcantly higher pain and disability levels than the idiopathic neck pain group, these were factored into the analysis and thus pain and disability did not account for the observed group differences. There is also the possibility of a vestibular component contributing to the deﬁcits in subjects with whiplash. Some researchers have postulated that the vestibular system may be primarily causally related to balance disturbances in whiplash (Chester, 1991; Rubin et al., 1995; Mallinson et al., 1996; El-Kahky et al., 2000), while others suggest that the vestibular deﬁcits are secondary to disturbed cervical afferent input (Hinoki, 1975; Fischer et al., 1995). The debate still continues. In this study, subjects with reported or suspected

189

vestibular pathology were excluded, although exclusion of undiagnosed deﬁcits in the vestibular system was not within the scope of the study. However, the similarities with respect to the principle sway direction of abnormality (AP sway) between the neck pain groups in this study and in others with low back pain (Della Volpe et al., 2005) would suggest that the cause of the balance disturbances are more indicative of somatosensory impairment (Shumway-Cook and Horak, 1986). Abnormalities in the ML direction would be more consistent with patterns expected for vestibular dysfunction (Horak et al., 1990; Nashner and Peters, 1990; Allum et al., 2001; Treleaven et al., submitted for publication). Thus, the ﬁndings of this study suggest that the somatosensory impairment and its resultant effects are greater in those where trauma has precipitated the neck pain, with evidence of more widespread difﬁculties with standing balance over a variety of stances and conditions and possibly a decreased ability to compensate for challenges to the postural control system. The ﬁndings of this study have implications for the assessment and management of balance in those with neck pain. The need for a comprehensive assessment of both the amplitude and energy of sway in tests in comfortable, narrow and tandem stances to detect balance disturbances in neck pain subjects is highlighted. From a clinical perspective, the results indicate that a battery of balance tests including comfortable, narrow and tandem stance with eyes open and closed should be included in the routine examination of all neck pain patients even in those not speciﬁcally complaining of any unsteadiness or balance difﬁculties. Our previous research has demonstrated that subjects with whiplash reporting dizziness and unsteadiness have greater deﬁcits than those not experiencing these symptoms (Treleaven et al., 2005a). The present ﬁndings conﬁrm balance deﬁcits also exist in subjects with whiplash not reporting dizziness unsteadiness or balance difﬁculties (Treleaven et al., 2005a), and reveals that idiopathic neck pain subjects without these complaints also have balance disturbances.

5. Conclusion Neck disorders are associated with deﬁcits in standing balance particularly in the anterior posterior direction when compared to asymptomatic control subjects, whether the neck pain is of idiopathic or whiplash origin. This study has determined that overall, the impairments in balance are greater in patients with neck pain resulting from a whiplash injury than idiopathic neck pain and this is not related to anxiety or pain and disability levels. The differences in balance responses between subjects with whiplash and idiopathic neck pain in narrow stance may suggest altered strategies

ARTICLE IN PRESS 190

S. Field et al. / Manual Therapy 13 (2008) 183–191

of control. The results also suggest somatosensory impairment is the most likely cause of the balance disturbances in both neck pain groups. The results have implications for feature selection of sway analysis in future studies and for the clinical assessment and management of balance disturbances in neck pain.

References Ageborg E. Consequences of a ligament injury on neuromuscular function and relevance to rehabilitation—using the anterior cruciate ligament-injured knee as a model. Journal of Electromyography and Kinesiology 2002;12:205–12. Allum JHJ, Bloem BR, et al. Differential diagnosis of proprioceptive and vestibular deﬁcits using dynamic support-surface posturography. Gait and Posture 2001;14(3):217–26. Alund M, Larsson SE, et al. Dynamic posturography in cervical vertigo. Acta Otolaryngology Supplement 1991;481:601–2. Alund M, Ledin T, et al. Dynamic posturography among patients with common neck disorders. A study of 15 cases with suspected cervical vertigo. Journal of Vestibular Research Equilibrium and Orientation 1993;3(4):383–9. Bove M, Courtine G, et al. Neck muscle vibration and spatial orientation during stepping in place in humans. Journal of Neurophysiology 2002;88(5):2232–41. Chester JB. Whiplash, postural control, and the inner-ear. Spine 1991;16(7):716–20. Courtine G, Papaxanthis C, et al. Gait-dependent integration of neck muscle afferent input. Neuroreport 2003;14(18):2365–8. Dault MC, Geurts ACH, et al. Postural control and cognitive task performance in healthy participants while balancing on different support-surface conﬁgurations. Gait and Posture 2001;14(3): 248–55. DeJong PI, DeJong JM. Ataxia and nystagmus induced by injection of local anaesthetics in the neck. Annals of Neurology 1977; 1977(1):240–6. Della Volpe R, Popa T, et al. Changes in coordination of postural control during dynamic stance in chronic low back pain patients. Gait Posture 2005 Nov 24 [Epub ahead of print]. El-Kahky A, Kingma H, et al. Balance control near the limit of stability in various sensory conditions in healthy subjects and patients suffering from vertigo or balance disorders: impact of sensory input on balance control. Acta Otolaryngology 2000;120:508–16. Fattori B, Borsari C, et al. Acupuncture treatment for balance disorders following whiplash injury. Acupuncture and ElectroTherapeutics Research 1996;21(3-4):207–17. Fischer A, Huygen PLM, et al. Vestibular Hyperreactivity and hyperventilation after whiplash injury. Journal of the Neurological Sciences 1995;132(1):35–43. Gosselin G, Rassoulian H, et al. Effects of neck extensor muscles fatigue on balance. Clinical Biomechanics 2004;19(5):473–9. Heikkila H, Astrom PG. Cervicocephalic kinesthetic sensibility in patients with whiplash injury. Scandinavian Journal of Rehabilitation Medicine 1996;28(3):133–8. Hinoki M. Vertigo due to whiplash injury: a neuro-otological approach. Acta Otolaryngology (Stockholm) 1975;419:9–29. Horak F, Nashner L, et al. Postural strategies associated with somatosensory and vestibular loss. Experimental Brain Research 1990;82(1):167–77. Huskisson E. In: R. Melzack, Visual analogue scales. Pain measurement and assessment. New York: Raven Press; 1983. pp. 33–6. Ishikawa K, Matsuzaki Z, et al. Effect of unilateral section of cervical afferent nerve upon optokinetic response and vestibular nystagmus

induced by sinusoidal rotation in guinea pigs. Acta Otolaryngology Supplement 1998;537:6–10. Karlberg M, Johansson R, et al. Dizziness of suspected cervical origin distinguished by posturographic assessment of human postural dynamics. Journal of Vestibular Research—Equilibrium and Orientation 1996a;6(1):37–47. Karlberg M, Magnusson M, et al. Postural and symptomatic improvement after physiotherapy in patients with dizziness of suspected cervical origin. Archives of Physical Medicine and Rehabilitation 1996b;77(9):874–82. Karlberg M, Persson L, et al. Impaired postural control in patients with cervico-brachial pain. Acta Oto-Laryngology 1995:440–2. Kavounoudias A, Gilhodes JC, et al. From balance regulation to body orientation: two goals for muscle proprioceptive information processing? Experimental Brain Research 1999;124(1): 80–8. Koskimies K, Sutinen P, et al. Postural stability, neck proprioception and tension neck. Acta Oto-Laryngology 1997:95–7. Kristjansson E, Dall’Alba P, et al. A study of ﬁve cervicocephalic relocation tests in three different subject groups. Clinical Rehabilitation 2003;17(7):768–74. Lekhel H, Popov K, et al. Postural responses to vibration of neck muscles in patients with uni- and bilateral vestibular loss. Gait and Posture 1998;7(3):228–36. Le Pera D, Graven-Nielsen T, et al. Inhibition of motor system excitability at cortical and spinal level by tonic muscle pain. Clinical Neurophysiology 2001;112(9):1633–41. Mallinson AI, Longridge NS, et al. Dizziness, imbalance, and whiplash. Journal of Musculoskeletal Pain 1996;4(4):105–12. Marteau T, Bekker H. The development of a six item short-form of the Spielberger State Trait Anxiety Inventory (STAI). British Journal of Clinical Psychology 1992;3:301–6. McIlroy W, Maki B. Preferred foot placement during quite stance: development of a standardized foot placement for balance testing. Clinical Biomechanics 1997;12(1):66–70. McPartland JM, Brodeur RR, et al. Chronic neck pain, standing balance, and suboccipital muscle atrophy—a pilot study. Journal of Manipulative Physiological Therapeutics 1997;20(1):24–9. Michaelson P, Michaelson M, et al. Vertical posture and head stability in patients with chronic neck pain. Journal of Rehabilitation Medicine 2003;35(5):229–35. Nashner LM, Peters JF. Dynamic posturography in the diagnosis and management of dizziness and balance disorders. Neurologic Clinics 1990;8(2):331–49. Peterson BW. Current approaches and future directions to understanding control of head movement. Brain Mechanisms for the Integration of Posture and Movement 2004;143:369–81. Persson L, Karlberg M, et al. Effects of different treatments on postural performance in patients with cervical root compression— a randomized prospective study assessing the importance of the neck in postural control. Journal of Vestibular Research-Equilibrium and Orientation 1996;6(6):439–53. Prieto TE, Myklebust JB, et al. Measures of postural steadiness: differences between healthy young and elderly adults. IEEE Transactions on Biomedical Engineering 1996;43(9):956–66. Pyykko I, Aalto H, et al. Hierarchy of different muscles in postural control. Acta Otolaryngology Supplement 1989;468:175–80. Riemann B, Guskiewicz K. Effects of mild head injury on postural stability as measured through clinical balance testing. Journal of Athletic Training 2000;35(1):19–25. Rocchi L, Chiari L, et al. Effects of deep brain stimulation and levodopa on postural sway in Parkinson’s disease. Journal of Neurology Neurosurgery and Psychiatry 2002;73(3):267–74. Rocchi L, Chiari L, et al. Feature selection of stabilometric parameters based on principal component analysis. Medical and Biological Engineering and Computing 2004;42(1):71–9.

ARTICLE IN PRESS S. Field et al. / Manual Therapy 13 (2008) 183–191 Rubin AM, Woolley SM, et al. Postural stability following mild head or whiplash injuries. American Journal of Otology 1995;16(2): 216–21. Schieppati M, Nardone A, et al. Neck muscle fatigue affects postural control in man. Neuroscience 2003;121(2):277–85. Scott D, Sterling M, et al. A psychophysical investigation of pain processing mechanisms in chronic neck pain. Clinical Journal of Pain 2005;21(2). Shumway-Cook A, Horak F. Assessing the inﬂuence of sensory integration on balance. Physical Therapy 1986;66:1548–50. Sjostrom H, JH JA, et al. Trunk sway measures of postural stability during clinical balance tests in patients with chronic whiplash injury symptoms. Spine 2003;28(15):1725–34. Speers RA, Ashton-Miller JA, et al. Age differences in abilities to perform tandem stand and walk tasks of graded difﬁculty. Gait and Posture 1998;7(3):207–13. Spitzer W, Skovron M, et al. Scientiﬁc monograph of Quebec task force on whiplash associated disorders: redeﬁning ‘‘whiplash’’ and its management. Spine 1995;20(8S):1–73. Thunberg J, Hellstrom F, et al. Inﬂuences on the fusimotor-muscle spindle system from chemosensitive nerve endings in the cervical

191

facet joints in the cat; possible implications for whiplash induced disorders. Pain 2001;91:15–22. Tjell C, Tenenbaum A, et al. Smooth pursuit neck torsion test— a speciﬁc test for whiplash associated disorders? Journal of Whiplash and Associated Disorders 2003;1(2):9–24. Treleaven J, Jull G, et al. Standing balance in persistent WAD— comparison between subjects with and without dizziness. Journal of Rehabilitation Medicine 2005a;37(4):224–9. Treleaven J, Jull G, et al. Is the method of signal analysis and test selection important for measuring standing balance in chronic whiplash? Gait and Posture 2005b;21(4):395–402. Treleaven J, Low Choy N, et al. Comparison of postural control disturbance between subjects with persistent whiplash associated disorder and subjects with unilateral vestibular loss. Archives of Physical Medicine and Rehabilitation, in review. Vernon H. The Neck Disability Index: patient assessment and outcome monitoring in whiplash. Journal of Musculoskeletal Pain 1996;4:95–104. Wenngren B, Pedersen J, et al. Bradykinin and muscle stretch alter contralateral cat neck muscle spindle output. Neuroscience Research 1998;32:119–29.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 192–199 www.elsevier.com/locate/math

Effects of patellar taping on knee joint proprioception in patients with patellofemoral pain syndrome Michael J. Callaghana,, James Selfeb, Alec McHenryc, Jacqueline A. Oldhama a

Centre for Rehabilitation Science, Manchester Royal Infirmary, University of Manchester, Oxford Road, Manchester, M13 9WL, UK b Department of Allied Health Professions University of Central Lancashire Preston, Lancashire, PR1 2HE, UK c Physio & Therapies, 61 Halifax Road, Todmorden, Lancashire, OL14 5BB, UK Received 13 September 2005; received in revised form 14 August 2006; accepted 30 November 2006

Abstract The aim of this study was to assess the effect of patellar taping of the proprioceptive status of patients with patellofemoral pain syndrome (PFPS). A total of 32 subjects (18 males, 14 females of age 31.9711.2, body mass index 25.875.3) with PFPS were tested for Joint Position Sense (JPS) using a Biodex dynamometer. Outcomes of interest were the absolute error (AE), the variable error (VE) and the relative error (RE) of the JPS values for both active (AAR) and passive (PAR) angle reproduction at an angular velocity of 21/s with a start angle at 901 and target angles of 601 and 201. Taping was applied in random order across the patella of each subject with each of the subjects acting as their own internal control. Results indicated initially that application of patellar tape did not enhance and in some cases worsened the JPS of the subjects (P40.05). However, when the subjects’ proprioceptive status was graded according to their closeness to the target angles into ‘good’ (p51, N ¼ 10) and ‘poor’ (451, N ¼ 22) taping enhanced nearly all values of those with ‘poor’ proprioception, with AE at 201 to statistical signiﬁcance (P ¼ 0.021). In conclusion, this study has shown that patellar taping did not improve the AAR and PAR JPS tests of a whole sample of 32 PFPS patients. It also has shown that a subgroup of PFPS patients with poor proprioception may exist and be helped by patellar taping. r 2007 Elsevier Ltd. All rights reserved. Keywords: Anterior knee pain; Taping; Joint position sense

1. Introduction Although patellar taping is an inexpensive technique readily used by physiotherapists in the treatment of patients with patellofemoral pain syndrome (PFPS), the mechanism for its success is still unresolved. McConnell (1986) originally described patellar taping as part of a treatment programme for PFPS and theorized that this technique could alter patellar position, enhance contraction of the vastus medialis oblique (VMO) muscle, and hence decrease pain. Studies on patients with PFPS Corresponding author. Tel./fax: 0161 276 6672.

E-mail address: [email protected] (M.J. Callaghan). 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.11.004

have thus far been inconclusive regarding taping realignment of patellar position (Crossley et al., 2000) and patellar taping enhancement of VMO contractions (Cerny, 1995). However, a number of studies have shown that patellar taping helps decrease pain in patients with PFPS (e.g., Powers et al., 1997) although the mechanism for this symptomatic improvement remains largely unknown (for review see Callaghan, 1997). Proprioception, deﬁned as the acquisition of stimuli from conscious and unconscious processes in the sensorimotor system (Lephart and Fu, 2000), is now thought to play a more signiﬁcant role than pain in preventing injury in the aetiology of chronic injury and in degenerative joint disease (Lephart, 1995). As long

ARTICLE IN PRESS M.J. Callaghan et al. / Manual Therapy 13 (2008) 192–199

ago as 1986, Wilson and Lee proposed that various knee injuries may affect knee joint proprioception, due to damage of the position sense receptors (Wilson and Lee, 1986). Proprioception can be appreciated and measured consciously by a complex system involving quick adapting and slow adapting mechanoreceptors; these are thought to mediate the sensations of joint movement (often referred to as kinaesthesia) and joint position (referred to as joint position sense (JPS)) (Lephart et al., 1992). Proprioceptive deﬁcits have been found in anterior cruciate deﬁcient knees (Beynnon et al., 1999), in osteoarthritic knees (Sharma et al., 1997; Hewitt et al., 2002) and in knees with a chronic effusion (Guido et al., 1997). It was also thought that patients who suffered patellofemoral dislocation may suffer from a proprioceptive deﬁcit due to disrupted neuroproprioceptive structures in the medial retinaculum, capsule, bursae and vastus medialis (Jerosch and Prymka, 1996). It has been further speculated (Sanchis-Alfonso et al., 1999) that PFPS patients with more subtle forms of chronic patella malalignment may exhibit dysfunction of the peripatellar plexus, detectable with proprioceptive testing. These researchers found histological evidence of neuromata and nerve damage to the peripatellar soft tissues particularly the lateral retinaculum that suggested altered proprioception, subsequent patellar instability with resultant patellar pain. They recommended that proprioception training with tape should form part of a rehabilitation programme in patients with these symptoms. There have been few clinical studies to investigate proprioception status in PFPS. One found no differences between 24 PFPS patients and matched controls for JPS testing in weight and non-weight bearing (Kramer et al., 1997). Another, in abstract form (Prymka et al., 1998), conversely found signiﬁcant differences between 43 PFPS patients and 30 controls in isolated JPS testing of the knee. The poor proprioception performance associated with PFPS was improved after applying a simple elastic bandage. The most recent study to test the theory of poor proprioception associated with PFPS was performed on a group of 20 patients (Baker et al., 2002). Compared to 20 healthy control subjects, Baker et al. (2002) found signiﬁcant differences in the proprioceptive ability of the PFPS group as measured by absolute, relative and variable error (VE) during weight bearing and non-weight bearing active JPS tests. Regarding the effects of interventions for PFPS, Callaghan et al. (2002) in a multicentre study showed that the application of patellar tape signiﬁcantly improved the proprioceptive status using active and passive JPS tests of a sub-sample (n ¼ 26) of 52 healthy subjects whose proprioceptive status was graded as ‘poor’. The tape did not improve those healthy subjects (n ¼ 26) whose status was graded as ‘good’. This gave some insight into the enhanced proprioceptive effect of

193

taping. To date, the efﬁcacy of tape on the proprioception of a symptomatic group of PFPS patients has not been studied. We were interested if a similar subclassiﬁcation could be applied to the group of patients with PFPS. Just as the restoration to good proprioception status is widely accepted as a key component in the rehabilitation of other knee pathologies, so it may be that improving proprioception in patients with PFPS may help towards normal knee function and accelerate the rehabilitation process. Therefore, the purpose of this study was to determine the effect of application of patellar taping on the JPS proprioceptive ability of the knee in a group of patients with PFPS. In addition, we wanted to investigate the possibility of sub-classiﬁcation of PFPS proprioception values and whether or not this inﬂuenced the effect of patellar taping on JPS results. The null hypothesis was that there would be no difference in JPS proprioceptive ability between the taped and untaped conditions.

2. Methods 2.1. Subjects Thirty-two patients with patellofemoral pain were referred from the Physiotherapy departments of Manchester Royal Inﬁrmary and St. Luke’s Hospital, Bradford. None of the subjects had commenced physiotherapy treatment, but were examined by two authors (MJC, AMcH) to conﬁrm the diagnosis clinically and exclude other causes of their symptoms. Each patient served as their own control with the no-taping condition being the internal control. All patients gave verbal and written informed consent. Ethical approval was obtained by the appropriate LRECs. 2.2. Inclusion criteria Patients had retropatellar pain greater than 6 months brought on by two (or more) of the following without traumatic onset: prolonged sitting (theatre goer’s sign); stair climbing, descending; running; kneeling; hopping/ jumping; pain on palpation of patellar facets; a step down (25 cm step) or double legged squat (Crossley et al., 2002). Patients were also included if they had a normal radiograph, normal MR scan or normal arthroscopy, if performed. 2.3. Exclusion criteria Patients were excluded from the study if they had previous knee surgery (not including arthroscopy),

ARTICLE IN PRESS 194

M.J. Callaghan et al. / Manual Therapy 13 (2008) 192–199

previous knee trauma that was still symptomatic or had an allergy to adhesive tape. Further clinical examination determined the presence of other lower extremity dysfunction that may account for the knee symptoms. These include referred pain from the lumbar spine and hip joint, severe leg length discrepancy, knee ligament, quadriceps tendon and meniscal pathologies; patella tendinitis (‘Jumper’s knee’); tibial tubercle apophysitis (Osgood Schlatters disease); bursitis; infrapatella fat pad lesion (Hoffa’s syndrome); medial plica syndrome; femoral anteversion and tibial torsion, bipartite patella (James, 1979) and also osteochondritis dessicans patella (Bentley, 1989). 2.4. Materials JPS testing was performed on the Biodex system 2 Dynamometer (Biodex Corp. Shirley NY, USA) using this system’s electrogoniometer, sensitive to 11 increments. This was calibrated before the sessions in accordance with the manufacturer’s instructions. Data were processed using the Biodex Advantage software (v4.5). In all tests, visual cues were eliminated by a blindfold. The tape was a 10 cm wide strip of Hypaﬁx adhesive tape (Smith & Nephew, Hull, UK). A sphygmomanometer cuff provided equal sensory input to the lower limb of each patient from the dynamometer’s tibial pad (SP Services, Telford, UK). 2.5. Procedure Wearing shorts, barefoot and blindfolded for each test, patients were seated with hip ﬂexion at 901 and a starting position for knee ﬂexion of 901. In the event of bilateral PFPS, the most symptomatic leg was tested. The tibial pad was secured to the shank of the leg 3 cm superior to the lateral malleolus. The sphygmomanometer cuff was wrapped around the tibia under the tibial pad and inﬂated to 40 mmHg, with constant checking to ensure equal pressure throughout the study. To avoid any learning effect the order of conditions (tape or no tape) was randomly allocated for each subject using a random number generator (www. mathgoodies.com/calculators/random_number.html). After each test condition the patient left the seat and walked around the room for approximately 5 min in order to reduce any possibility of proprioceptive carry over to the next test. 2.6. Measurement of proprioception In order to detect JPS aspects of proprioception, we used a family of methods adopted from our previous study and widely used in other studies on knee proprioception, namely, passive angle reproduction

(PAR) (Perlau et al., 1995), active angle reproduction (AAR) (Friden et al., 1996). The knee was moved from a 901 start position to each of the target angles of 201 and 601 in random order. These target angles were chosen for several reasons. Firstly, at 201 of knee ﬂexion the distal patella contacts the proximal femoral trochlea (Fulkerson, 2004) so that any proprioceptive deﬁcit at this angle may be related to patella mal-tracking, which is widely accepted to be a major causative factor in PFPS symptomology (Powers, 1998). Secondly, 601 of knee ﬂexion has been highlighted as a pertinent angle in PFPS pathology. When the critical test for the patellofemoral joint was ﬁrst described (McConnell, 1986) it included the angle of 601; the signiﬁcance of this angle has since underlined by functional motion analysis (Selfe et al., 2001). Thirdly, choosing non-weight bearing angles at 201 and 601 facilitates comparison with another proprioception study on PFPS patients (Baker et al., 2002). 2.7. Passive angle reproduction For PAR, starting at 901 of knee ﬂexion the lever arm extended the test limb, without resistance to the movement, to the target angles. Passive movement occurred at an angular velocity of 21/s to limit reﬂexive muscle contractions. Subjects were instructed not to voluntarily contract their muscles. The limb was maintained at the target angle for 10 s to enable the subject to remember the position. After passively returning to 901, and after a pause of 5 s, the same cycle was repeated. This time the subject activated a hand-held stop button when they felt the target angle had been reached. Once the angle had been reached, patients were not permitted to correct the angle. The angle was noted from the on-screen goniometer. A total of six readings were taken, and the difference between the perceived angle and each of the target angles calculated for each reading and saved for subsequent analysis. 2.8. Active angle reproduction In the same seated conditions, the subject’s limb was passively moved to the target angles. The leg was held there for 10 s for the subject to memorize the position and then returned to 901 knee ﬂexion. After a pause of 5 s, the subject moved the lower limb by active contraction at an angular velocity approximating 21/s and stopped when he/she perceived the target angle had been reached. Once the angle was achieved, patients were not permitted to correct the angle. A total of six readings were taken and the difference between the perceived angle and each of the target angles noted for each trial.

ARTICLE IN PRESS M.J. Callaghan et al. / Manual Therapy 13 (2008) 192–199

2.9. Patellar taping Patellar taping was applied by the two lead researchers (MJC and JS) following the methods of our previous study (Callaghan et al., 2002). With the patient in supine with a relaxed extended knee, one strip of tape was applied without tension across the centre of the patella. The centre of the tape was as near as possible to the centre of the patella, with its medial and lateral edges aligned with the medial and lateral knee joint lines. The length of tape was calculated at 50% of the total circumference of the subject’s knee. This was in order to account for anthropometric differences between patients, which may have meant some smaller patients getting proportionally greater amounts of tape than others. In order to standardize the taping technique and because of difﬁculties in assessing patella malalignment (Watson et al., 1999), no attempt was made to correct patellar position. The theory was that afferent cutaneous sensation changes from the tape may alter proprioception without the need to correct patellar malalignment. Pain was not formally assessed during testing as it has already been established that there are no signiﬁcant correlations between JPS errors and levels of patellar pain (Baker et al., 2002). Therefore, it could be assumed that patients were not using pain as a strategy to position their knee (Kramer et al., 1997).

195

interpret in a clinically relevant manner. Therefore, data are presented as Medians (IQR) in order to aid comparison with other studies and illustrate clinical relevance. The mean of the six trials was taken for each subject at each angle and for each condition, and used to calculate the difference between the actual angles achieved and the target angles (Beynnon et al., 1999). Three dependent variables were analysed for AAR and PAR tests (Baker et al., 2002). (1) Absolute error (AE)— the difference between the actual angle relative to the target angle; this has no directional bias. (2) Relative (or real) error (RE)—the difference between the actual angle relative to the target angle; this had a directional bias. (3) VE—the standard deviation of the mean of the subject’s score of six trials; this reﬂects the consistency of the actual angle achieved. Patients who had JPS scores further away from the target angles would have high AE and RE values and could be said to be less accurate. Those with high VE scores could be said to be less consistent. Analysis consisted of a 3-way ANOVA (two movements (PAR and AAR), two conditions (Tape and NoTape) and two angles (201 and 601)) to assess the presence of any interactions. The level of probability was set at Po0.05. A Wilcoxon signed ranks test was used to analyse differences within groups for the nonparametric data.

2.10. Sample size calculation

3. Results

A sample size calculation based on means and standard deviations from our previous study (Callaghan et al., 2002) determined that the number of subjects in each group (taped and untaped) needed to detect similar differences would be 30 for PAR and 23 for AAR with a power of 80% (Po0.05).

A total of 32 patients with PFPS were recruited; their characteristics can be seen in Table 1. During testing, no patients complained of increased patellar pain. There were no adverse events attributed to the procedure or the application of tape. Data for all 32 subjects are presented in Table 2. These results showed that, for the whole group of 32 patients in all tests, the application of patellar tape made no difference to the error scores. Most scores remained the same or were less accurate and less consistent with the tape. A 3-way ANOVA showed no signiﬁcant interactions for AE, RE and VE regardless of movement type, target

2.11. Data analysis Statistical analysis was performed using SPSS (Statistical Package for Social Sciences) for Windows (v11.5). Prior to analysis, the data were tested for normal distribution using the Kolmogorov–Smirnov test with a signiﬁcance level set at Po0.05. Additionally, histograms were produced to visualize any non-normally distributed data. Results from these methods revealed that the assumption for normality was violated (K–S test P40.05). Therefore, the data were transformed using natural logarithmic transformation (Atkinson and Nevill, 1998). The appropriateness of the logarithmic transformation was checked visually by histograms of the standardized residuals. Unfortunately, although this transformation permits a more robust statistical analysis, it converts the raw data into percentage ﬁgures and makes the data difﬁcult to

Table 1 Subjects’ characteristics Characteristics

Mean (SD)

Age BMI Duration of PFPS symptoms (months) Gender Pain at rest (VAS)

31.9 (11.2) 25.8 (5.3) 14.6 (18.2) 18 m; 14f 1.0 (1.6)

BMI ¼ body mass index; PFPS ¼ patellofemoral pain syndrome; VAS ¼ visual analogue scale.

ARTICLE IN PRESS M.J. Callaghan et al. / Manual Therapy 13 (2008) 192–199

196

Table 2 Whole PFPS sample (n ¼ 32), medians (IQR) Target angle

Test

Error type

601

AAR

Absolute Relative Variable Absolute Relative Variable

7.66 7.33 0.49 3.67 0.0 2.63

Absolute Relative Variable Absolute Relative Variable

7.00 3.33 2.47 4.75 1.25 2.79

PAR

201

AAR

PAR

No tape

Tape

(4.5–12.3) (3.0–12.3) (0.4–0.8) (2.1–6.8) (2.4–4.1) (2.0–3.9) (3.8–10.6) (9.9–3.6) (1.1–3.6) (3.2–6.9) (5.2 to 2.9) (2.1–4.2)

Outcome

P valuez

7.4 4.92 0.49 4.92 0.67 3.41

(4.9–10.9) (3.3–3.3) (0.3–0.8) (3.3 -8.6) (3.3–4.3) (2.1–4.9)

Better Better Same Worse Worse Worse

.795 .178 .984 .164 .793 .161

6.33 4.25 2.42 4.75 2.92 3.25

(3.9–9.7) (9.7–3.3) (1.5–4.2) (3.3–11.7) (8.2–1.7) (1.9–4.2)

Better Worse Same Same Worse Worse

.228 .152 .719 .184 .814 .248

Outcome

P valuez

Better Better Better Better Better Better

.142 .086 .664 .158 .969 .814

Better Better Better Better Worse Better

.021* .968 .526 .646 .445 .445

zAnalysis by Wilcoxon signed ranks test. AAR ¼ active angle reproduction; PAR ¼ passive angle reproduction. Table 3 Poor proprioception status (N ¼ 22) (451), medians (IQR) Target angle

Test

Error type

No tape (deg.)

601

AAR

Absolute Relative Variable Absolute Relative Variable

10.83 10.83 0.60 8.08 5.66 3.81

(7.7–14.3) (7.2–14.3) (.40–.86) (5.7–14.2) (6.6–10.6) (2.4–5.3)

Absolute Relative Variable Absolute Relative Variable

9.83 8.00 3.04 10.25 3.83 4.1

(7.7–15.4) (11.7–5.6) (1.3–3.8) (6.9–16.3) (16.3–9.5) (2.5–5.9)

PAR

201

AAR

PAR

Tape (deg.) 9.00 4.83 0.53 5.83 0.66 3.73

(4.9–13.2) (3.3–13.2) (.42–.98) (4.8–13.4) (3.7–10.7) (2.2–5.4)

9.00 (5.5–12.5) 7.41 (12.5–2.5) 2.81 (1.9–4.6) 9.25 (3.2–18.2) 5.25 (15.1 to 2.1) 3.01(2.0–4.1)

zAnalysis by Wilcoxon signed ranks test; *statistically signiﬁcant. AAR ¼ active angle reproduction; PAR ¼ passive angle reproduction.

angle or use of patellar tape (P40.05). Thus, there was no evidence that the application of tape had a signiﬁcant effect on any condition for the whole sample group. As stated earlier, further analysis was performed by sub-classifying the proprioceptive test results into ‘good’ and ‘poor’ based on the accuracy of the scores from the target angle (Callaghan et al., 2002; Perlau et al., 1995). This revealed two subgroups of patients: those with ‘good’ proprioception (p51 from the target angles, N ¼ 10) and those whose proprioception could be classed as ‘poor’ (i.e. 451 from the target angles, N ¼ 22). The descriptive statistics now revealed that whereas proprioceptive values were hardly affected by the application of tape in those patients with ‘good’ proprioception, it improved all but one of the test values for those with ‘poor’ status, one to statistical signiﬁcance when analysed by a non-parametric test. Table 3 shows the results of those patients whose median values were 451 from the target angles and were classiﬁed as having ‘poor’ proprioception.

4. Discussion Initial results using logarithmically transformed data revealed that there were no signiﬁcant interactions for using patellar tape on target angle and active or passive testing. However, careful analysis of the non-transformed data suggested that not all patients with PFPS had the same proprioception as detected by the JPS tests of AAR and PAR. Indeed, there appeared to be subgroups of patients whose status was regarded as ‘poor’ (N ¼ 22) or ‘good’ (N ¼ 10) (i.e. p51 or 451 from the target angles of 201 and 601). After applying a single strip of tape across the patella with no directional pull or attempted realignment of the patella, PFPS patients with ‘poor’ proprioception were more accurate and more consistent with their AAR and PAR JPS tests at both angles of 201 and 601. Only one of these improvements reached statistical signiﬁcance. The subgroup with ‘good’ proprioception, on the other hand, did not ﬁnd any beneﬁt from the addition of taping, and

ARTICLE IN PRESS M.J. Callaghan et al. / Manual Therapy 13 (2008) 192–199

in several tests were actually found to be worse. Two other studies (Perlau et al., 1995; Kaminski and Perrin, 1996) also noticed this phenomenon when testing JPS pre- and post-application of knee bracing in healthy subjects. Perlau et al. (1995) explained that far from ﬁnding extra cutaneous stimuli helpful, those with inherently ‘good’ proprioception derive no beneﬁt and may even ﬁnd a brace or taping confusing when asked to perform a positioning task. A further explanation can be the concept of treatment effect. Put simply, patients with ‘good’ proprioception do not need any extraneous help, whereas those who are ‘poor’ have more chance of beneﬁting from taping treatment. This concurs with our previous ﬁnding when using patellar tape on healthy subjects using a target angle of 451. Those whose proprioception was categorized as ‘good’ by the same deﬁnition found no statistically signiﬁcant beneﬁt from patellar taping on AAR AE, and were actually made signiﬁcantly worse on PAR AE (Callaghan et al., 2002). Compared to other knee conditions, there have been fewer studies to assess proprioception in patients complaining of PFPS. Baker et al. employed a kinematic system to compare healthy and PFPS active knee JPS testing in weight bearing and non-weight bearing (Baker et al., 2002). Although the difference in testing equipment and protocol makes comparison of all test values difﬁcult, they noticed signiﬁcant between-group differences in AE, RE and VE at 201 and 601 in NWB AAR trials. They found, as we did, that the greatest error value was in AE. Their maximum mean value for AE was 2.91 at 601 compared to our equivalent AAR median value of 7.71. In general, our median AE, RE and VE values for all 32 subjects were higher than their mean values for NWB AAR. Using an electrogoniometer, Kramer et al. (1997) found no differences between 24 PFPS and healthy controls using WB and NWB AAR. At a 601 target angle, the mean AE of the PFPS group was only 1.81 from the target angle; our median AAR equivalent score for all 32 subjects was 7.71. Apart from employment of different equipment, a further explanatory factor of these differences is that their data were taken from the average values from 2 separate test days. Prymka et al. (1998) described in abstract form that patients with ‘chondropathia patellae’ averaged 13.21 from the target angle for AAR tests. Their error scores are higher than any other work in this area and may simply reﬂect the different knee pathology of their patients.

197

ﬁbres as a response to pain (Capra and Ro, 2000) or microscopic small nerve damage in the lateral retinaculum (Sanchis-Alfonso et al., 2001; Fulkerson, 2004). The application of some form of knee support is thought to augment afferent input via the enhancement of cutaneous stimulation (Lephart et al., 1992). It has also been proven that cutaneous afferents in the hand provide proprioceptive feedback information when stretched with tape resulting in the perception and detection of ﬁnger movement (Collins et al., 2000). Therefore, it is possible that the loss of afferent information in the subgroup of PFPS patients was improved by tape-enhancing feedback information from the muscle spindles, soft tissue and skin. The fact that both AAR and PAR were affected suggests that both active and passive systems of patellar control were affected and concurs with the detailed discussions of Baker et al. (2002, p. 213). It was deemed impractical for the present study to correlate patellar malalignment with the proprioception values of our PFPS patients, as a clinically reliable method of assessing patellar malalignment has yet to be developed. These results raise the issue of sub-classiﬁcation of PFPS patients in the domain of proprioception. There appear to be some PFPS patients with poorer JPS proprioceptive status than others, and treatment of these patients may be more appropriately applied if they could be easily identiﬁed and appropriately categorized. There is the intriguing possibility that the subgroup helped by tape is comprised of patients with neural damage within the lateral retinaculum or nerve senitization due to pain. A three-way comparison of malalignment, proprioception and histological ﬁndings would be an intricate but useful area of further research. 4.2. Clinical significance The small values involved in JPS testing (both in this study and others previously) raise the issue of clinical signiﬁcance. There are two schools of thought. Firstly, that patellar taping can bring about improvements in JPS, which although statistically signiﬁcant are so small as to be of doubtful clinical signiﬁcance (in the present study—0.831 for AE AAR 201). Secondly, that proprioception as detected by JPS is such an inherently precise task requiring high levels of precision that even very small changes achieved with a treatment technique such as taping are clinically important.

4.1. Rationale for efficacy of patellar tape 4.3. Study limitations A possible explanation as to why patellar tape has the capacity to improve JPS testing in PFPS may be either in chemical sensitizing of small and large diameter nerve

A post hoc power calculation showed that for AE at 601 the study was underpowered (29% power; P40.05;

ARTICLE IN PRESS 198

M.J. Callaghan et al. / Manual Therapy 13 (2008) 192–199

mean ¼ 2.01, pooled SD ¼ 6.11, N ¼ 22) and would need a sample size of N ¼ 74 (Po0.05; 80% power) to see signiﬁcant differences between the tape and no-tape conditions. Despite our best efforts to recruit an appropriate sample size following an a priori power calculation, this study probably suffered from a type 2 statistical error.

5. Conclusion This study has shown that patellar taping did not improve the AAR and PAR JPS tests of a whole sample of 32 PFPS patients. However, it did improve the AAR and PAR JPS tests of 22 PFPS patients who were 451 from the target angles of 201 and 601. Only one of these reached statistical signiﬁcance. Those patients with p51 accuracy were not improved by the tape. This raises the issue of proprioception ability being a sub-classiﬁcation of PFPS and may help appropriate treatment.

Acknowledgements We wish to thank Dr. Steve Roberts for statistical help, the Physiotherapy Department at St. Luke’s Hospital, Bradford and the Wellcome Trust Clinical Research Facility, Manchester where the data were collected. References Atkinson G, Nevill AM. Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Medicine 1998;26(4):217–38. Baker V, Bennell K, Stillman B, Cowan SM, Crossley K. Abnormal knee joint position sense in individuals with patellofemoral pain syndrome. Journal of Orthopaedic Research 2002;20:208–14. Bentley G. Anterior Knee Pain: diagnosis and management. Journal of the Royal College of Surgeons of Edinburgh 1989;34(Suppl): 2–3. Beynnon BD, Ryder SH, Konradsen L, Johnson RJ, Johnson K, Renstro¨m PA. The effect of anterior cruciate ligament trauma and bracing on knee proprioception. American Journal of Sports Medicine 1999;27(2):150–5. Callaghan MJ. Patellar taping, the theory versus the evidence: a review. Physical Therapy Reviews 1997;2:181–3. Callaghan MJ, Selfe J, Bagley P, Oldham JA. The effect patellar taping on knee joint proprioception. Journal of Athletic Training 2002;37(1):19–24. Capra NF, Ro JY. Experimental muscle pain produces central modulation of proprioceptive signals arising from jaw muscles spindles. Pain 2000;86:151–62. Cerny K. Vastus medialis oblique/vastus lateralis muscle activity for selected exercises in persons with and without patellofemoral pain syndrome. Physical Therapy 1995;75(8):672–83. Collins DF, Refshauge KM, Gandevia SC. Sensory integration in the perception of movements at the human metacarpophalangeal joint. Journal of Physiology 2000;529(2):505–15.

Crossley K, Cowan SM, Bennell KL, McConnell J. Patellar taping: is clinical success supported by scientiﬁc evidence? Manual Therapy 2000;5(3):142–50. Crossley K, Bennell K, Green S, Cowan S, McConnell J. Physical therapy for patellofemoral pain: a randomized, double-blinded, placebo-controlled trial. American Journal of Sports Medicine 2002;30(6):857–65. Friden T, Roberts D, Zatterstrom R, Lindstrand A, Moritz U. Proprioception of the nearly extended knee. Knee Surgery, Sports Traumatology, Arthroscopy 1996;4(4):217–24. Fulkerson JP. Disorders of the patellofemoral joint, 4th ed. Baltimore: Williams & Wilkins; 2004. Guido J, Voight ML, Blackburn TB, Kidder JD, Nord S. The effects of chronic effusion on knee joint proprioception: a case study. Journal of Orthopaedic and Sports Physical Therapy 1997;25(3): 208–12. Hewitt BA, Refshauge KM, Kilbreath SL. Kinesthesia at the knee: the effect of osteoarthritis and bandage application. Arthritis Care and Research 2002;47(5):479–83. James SL. Chondromalacia of the patella in the adolescent. In: Kennedy JC, editor. The injured adolescent knee. Baltimore: Williams & Wilkins; 1979. p. 205–51. Jerosch J, Prymka M. Knee joint proprioception in patients with posttraumatic recurrent patellar dislocation. Knee Surgery, Sports Traumatology, Arthroscopy 1996;4:14–8. Kaminski TW, Perrin DH. Effect of prophylactic knee bracing on balance and joint position sense. Journal of Athletic Training 1996;31(2):131–6. Kramer J, Handﬁeld T, Keifer G, Forwell L, Birmingham TB. Comparisons of weight bearing and non weight bearing tests of knee joint proprioception performed by patients with PFPS and asymptomatic. Clinical Journal of Sport Medicine 1997;7(2):113–8. Lephart SM. The role of proprioception in the treatment of sports injuries. Sports Exercise and Injury 1995;1:96–102. Lephart SM, Fu FH. Proprioception and neuromuscular control in joint stability, 1st ed. Human Kinetics Champaign; 2000. Lephart SM, Kocher MS, Fu FH, Borsa PA, Harner CD. Proprioception following anterior cruciate ligament reconstruction. Journal of Sport Rehabilitation 1992;1:188–96. McConnell J. The management of chondromalacia patellae: a long term solution. Australian Journal of Physiotherapy 1986;32(4): 215–23. Perlau R, Frank C, Fick G. The effect of elastic bandages on human knee proprioception on the uninjured population. American Journal of Sports Medicine 1995;23(2):251–5. Powers CM. Rehabilitation of patellofemoral joint disorders: a critical review. Journal of Orthopaedic and Sports Physical Therapy 1998;28(5):345–54. Powers CM, Landel R, Sosnick T, Kirby J, Mengel K, Cheney A, et al. The effects of patellar taping on stride characteristics and joint motion in subjects with patellofemoral pain. Journal of Orthopaedic and Sports Physical Therapy 1997;26(6):286–91. Prymka M, Schmidt K, Jerosch J. Proprioception in patients suffering from chondropathia patellae. International Journal of Sports Medicine 1998;19(S 60). Sanchis-Alfonso V, Rosello-Sastre E, Martinez-Sanjuan V. Pathogenesis of anterior knee pain syndrome and functional patellofemoral instability in the active young. American Journal of Knee Surgery 1999;12(1):29–40. Sanchis-Alfonso V, Rosello-Sastre E, Revert F. Neural growth factor expression in the lateral retinaculum in painful patellofemoral malalignment. Acta Orthopaedica Scandinavica 2001;72(2): 146–9. Selfe J, Harper L, Pedersen I, Breen-Turner J, Waring J. Four outcome measures for patellofemoral joint problems. Part1; development and validity. Physiotherapy 2001;87(10):507–15.

ARTICLE IN PRESS M.J. Callaghan et al. / Manual Therapy 13 (2008) 192–199 Sharma L, Pai YC, Holtkamp K, Rymer WZ. Is knee joint proprioception worse in the arthritic knee versus the unaffected knee in unilateral knee osteoarthritis? Arthritis Rheum 1997;40(8):1518–25. Watson CJ, Propps M, Galt W, Redding A, Dobbs D. Reliability of McConnell’s classiﬁcation of patellar orientation in symptomatic

199

and asymptomatic subjects. Journal of Orthopaedic and Sports Physical Therapy 1999;29(7):378–85. Wilson AS, Lee HB. Hypothesis relevant to defective position sense in a damaged knee. Journal of Neurology Neurosurgery and Psychiatry 1986;49:1462–3.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 200–205 www.elsevier.com/locate/math

Original article

Agreement of measures obtained radiographically and by the OSI CA-6000 Spine Motion Analyzer for cervical spinal motion Cheryl M. Petersena,,1, Dale Schuitb,1, Robert D. Johnsonc,1, H. Knechtd,2, Phyllis Levinee,1 a Department of Physical Therapy, Concordia University Wisconsin, 12800 North Lake Shore Drive, Mequon, 53097 WI, USA Department of Physical Therapy, Rosalind Franklin University of Medicine and Science, 3333 Green Bay Road, North Chicago, IL, USA c Achieve Orthopedic Rehabilitation Institute, Chicago, IL, USA d Department of Physical Therapy, University of Illinois at Chicago, Chicago, IL, USA e Functional Therapy Rehabilitation Services Inc., Homer Glen, IL, USA

b

Received 7 December 2003; received in revised form 18 October 2006; accepted 4 December 2006

Abstract The purpose of this study was to determine the agreement between angular measures of cervical spinal motion obtained from radiographs and from measures recorded by the OSI CA 6000 Spine Motion Analyzer (OSI SMA) in asymptomatic subjects. Fourteen subjects performed each of the following motions two times while wearing the OSI SMA: cervical ﬂexion, extension, side bending to the right and left. Each motion was performed once for the cervical radiograph. The difference between the values obtained by the two methods was plotted against the average of those values for each subject to illustrate the level of agreement of the two methods. The plotted points were widely scattered, with a large range between the limits of agreement. Range of motion values taken from the OSI SMA were not similar to those obtained from radiographs for the motions of the cervical spine. r 2007 Elsevier Ltd. All rights reserved. Keywords: Cervical spine; Cervical radiograph; Spinal motion

1. Introduction The OSI Spinal Motion Analyzer (OSI SMA) (Orthopedic Systems Inc., Union City, CA, USA) is an apparatus designed to provide measurements of spinal motion in multiple planes of movement at the Corresponding author. Tel.: +1 262 243 4347; fax: +1 262 243 4506. E-mail address: [email protected] (C.M. Petersen). 1 Ms. Petersen and Dr. Schuit were at the Northwestern University, Department of Physical Therapy and Human Movement Science, Mr. Johnson was at the Chicago Institute of Neurosurgery/Neuroresearch, Chicago, IL, USA, and Ms. Levine was coordinator of Research at Chicagoland Orthopedic Rehabilitation Services Inc., Palos Heights, IL, USA when this study was conducted. 2 Retired Head of Department of Physical Therapy and Chief of Clinical Services, UIC Hospital, Chicago, IL, USA.

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.12.004

same time. It is important to determine that the measures of cervical motion provided by the OSI SMA are reliable and valid, if the instrument is to be used as a clinical and research measurement tool. Previous research has shown that the OSI SMA data is reliable (Dvorak et al., 1992; Johnson et al., 1992; Johnson and Schuit, 1993) and provides some support regarding the validity of the OSI SMA (Dvorak et al., 1992; Schuit et al., 1997; Lantz et al., 1999). There is limited information available on the validity of measures of cervical spinal motion obtained from the OSI SMA. It is important to ascertain the level of agreement between cervical ROM measures from this device and from radiographs as measures of spinal motion from radiographs are often regarded as a ‘‘gold standard’’. Also, previous studies have utilized correlation coefﬁcients to compare the measures between

ARTICLE IN PRESS C.M. Petersen et al. / Manual Therapy 13 (2008) 200–205

201

radiographs and various measurement devices. We felt it was important to analyze our data using the method comparison process (Altman and Bland, 1983). This process illustrates the actual differences between the two measurement methods, and has been used to compare measures of lumbar spinal motion between radiographs and the OSI SMA (Schuit et al., 1997). Therefore, the purpose of this study was to determine the level of agreement between angular measures of cervical spinal motion from asymptomatic subjects obtained by two methods: cervical radiographs and the OSI SMA.

2. Methodology Subjects (11 female, 3 male) were volunteers from the Department of Physical Therapy and Human Movement Science, Northwestern University Medical School, Chicago, IL, USA, and from staff at Chicagoland Orthopedic Rehabilitation Services, Inc., Palos Heights, IL, USA (mean age ¼ 27.3 years, standard deviation (SD) ¼ 6.01; mean height ¼ 1.71 m, SD ¼ 0.11; mean weight ¼ 68.04 kg, SD ¼ 12.83). Criteria for inclusion consisted of no history of cervical spine dysfunction, surgery, disease, or complaint of pain in the cervical spine region within the last 6 months. Subjects who might be pregnant or had existing cervical pain were excluded from the study. Subjects read and signed the consent form approved by the Institutional Review Board at Northwestern University and the Chicagoland Orthopedic Rehabilitation Services. 2.1. Apparatus The OSI SMA has been described in detail in a previous investigation (Schuit et al., 1997). The apparatus was linked by cable to a Gateway 486/25 personal computer (Gateway 2000, South Dakota, USA). The radiographic exposure equipment was a Siemens Polyphos 30M model (Siemens Medical Systems, Inc., IL, USA), and the radiographic ﬁlm was Konica MG daylight ﬁlm (Konica Medical Corporation, IL, USA).

Fig. 1. Starting position for a subject with the Spine Motion Analyzer in situ.

seated position with a stabilization belt placed around the abdomen (Fig. 1). Data collection was done in pairs to avoid the possibility that the order of randomized testing would affect the results. 2.3. OSI SMA testing The SMA linkage was attached to the hook piece on the cap and onto the body strap. The potentiometers on the linkage were electronically zeroed once at the beginning of the measurement session with the subject sitting in a ‘‘comfortable’’ upright posture. For each cervical motion repetition, gravity goniometers attached to the cranial cap were visualized to limit the amount of motion testing to 501 of ﬂexion, 301 of extension and 301 of right and left side bending, as indicated by a verbal command by the 2 examiners (DS and RDJ). Two repetitions were performed for each motion tested. For each repetition, the subject moved to the appropriate position and remained for 5 seconds to allow recording of the angular data, which was stored in a computer linked to the OSI SMA system. 2.4. Radiographs

2.2. Procedure Subjects were dressed in shorts (men) and swimsuit tops (women). The lower linkage hook piece of the OSI SMA was placed at the level of each subject’s T1 spinous process. The location of the T1 and C2 spinous processes were determined by palpation in the standing position as described by Hoppenfeld (1976) and Magee (2002). The upper linkage hook piece was attached to the cranial cap and was positioned such that the plastic portion of the ring sat snugly just above the ears. Two of the examiners (DS and RDJ) ﬁtted the OSI SMA hardware on the subjects. Testing was completed in a

Subjects were positioned for the radiographs in the same manner as for the OSI SMA testing. The cranial cap and the thoracic harness with the linkage hooks, visible on radiographs, were worn throughout both methods of testing to allow radiographic determination of the actual position of the OSI SMA linkage hooks. Each subject bit onto a mouth stick with metal markers on each end. Seated subjects were asked to maintain an upright posture and A/P and lateral view ﬁlms were taken. The subject then actively moved his/her head/ neck to the verbally indicated 501 ﬂexion, 301 extension and 301 right and left side bending positions.

ARTICLE IN PRESS 202

C.M. Petersen et al. / Manual Therapy 13 (2008) 200–205

2.5. Measurements taken from radiographs For measurement of the actual ﬁlms of cervical spine motion, tracing paper was placed over the neutral ﬁlms for the A/P or lateral views. Tracing paper was utilized because a ﬁner measurement line could be drawn on the paper than with a grease pencil on the radiograph, possibly affecting angular measurement values. All measurements were made by one examiner (CMP) but the initial drawing of the cephalic and caudal lines was done through consultation of two examiners (CMP and DS). Each ﬁnal ﬁlm measurement was the average of two separate random measurements. For both the lateral and AP ﬁlms, a caudal line was drawn either across the inferior surface of the T1 vertebra or when not visible, the inferior surface of C6, the inferior surface of C7 or the superior surface of T1. A cephalic line was drawn connecting the superior surface of the mandibular condyle and the occipital protuberance (OP) for the lateral ﬁlms, and parallel to the mouth stick line for the AP ﬁlms. Perpendicular lines were drawn from each of these horizontal lines. The angle of intersection of the perpendiculars was taken as the neutral position. The caudal line from the neutral view was superimposed over the caudal line on both the ﬂexion/extension or the side bending ﬁlms (Figs. 2 and 3).

Fig. 3. Method for determining measures of side bending right and left from the cervical radiograph; occipital line ¼ parallel to markers on mouth stick and vertebral line at superior surface of T1 (for variations see text).

2.6. Data collection and analysis Data from two trials from each movement were recorded with the OSI SMA. Data from one set of radiographs was recorded as the mean of the two measurements, measured by one examiner (CMP). We calculated the ICC (2, 2) for the two repetitions of each motion performed by each subject for within subject reliability for OSI SMA testing (Shrout and Fleiss, 1979). Standard errors of measurement were calculated per Portney and Watkins (2000). A method of analysis, proposed by Altman and Bland (1983), used in our previous study of lumbar spinal motion (Schuit et al., 1997), was also utilized.

3. Results

Fig. 2. Method for determining the measures of ﬂexion and extension from the cervical radiograph; occipital line ¼ superior surface of condylar head of mandible to occipital protuberance and vertebral line at superior surface of T1 (for variations see text).

ICC values between the two trials of each motion with the OSI SMA were as follows: ﬂexion ¼ 0.995, extension ¼ 0.995, right side bending ¼ 0.984 and left side bending ¼ 0.987 and indicate a high degree of repeatability between the two trials (Munro and Vistintainer,

ARTICLE IN PRESS C.M. Petersen et al. / Manual Therapy 13 (2008) 200–205

203

the plus and minus 1 and 1.96 SD lines on the graph (Fig. 4). For extension, radiographic measurements were on average 21 higher than that recorded by the OSI SMA. There is a difference of about 371 between the upper and lower limits of agreement (Fig. 5). For both side bending movements, systematic bias in both directions was found as well. For all motion graphs, the plotted points were scattered up to and sometimes beyond the upper and lower limits of agreement. Therefore, the magnitudes of the values at the limits of agreement are too large to state that these two methods provide similar clinical measures of cervical spinal motion. The wide limits of agreement indicate a high degree of random error (Table 1). Fig. 4. Bland–Altman plot of differences in the range of motion recorded radiographically and by the OSI SMA versus the average of the two methods for each subject for the motion of ﬂexion (in degrees from the results of 14 subjects).

Fig. 5. Bland–Altman plot of differences in the range of motion recorded radiographically and by the OSI SMA versus the average of the two methods for each subject for the motion of cervical extension (in degrees from the results of 14 subjects).

1986). The standard error of measurement (SEM) ranged from 0.741 to 1.351 reﬂecting acceptable reliability of the response (Portney and Watkins, 2000). Figs. 4 and 5 illustrate the values from the radiographs and the OSI SMA for the respective motions of ﬂexion and extension, comparing the mean difference versus the means of the data for the individual subjects. Table 1 contains the actual ROM recorded for the SMA and radiographs and the differences between the two measures. None of the mean differences were equal to zero, so bias was present in the measurement of each of the motions. The following ﬁndings represent the systematic bias present. For ﬂexion, radiographic measurements were on average 131 lower than that recorded by the SMA. In examining the magnitudes of the limits of agreement for ﬂexion, there is a difference of 361 between the upper and lower limits, as noted by

4. Discussion To analyze the magnitude of the differences between the two methods, limits of agreement that corresponded to 1 and 1.96 SD were used. These limits indicate only statistical signiﬁcance, as the acceptable magnitude of those differences is based on clinical judgment. By examining the differences between methods at the limits of 1 SD and at 1.96 SD, a decision can be made regarding the clinical acceptability of the differences. The present study did not ﬁnd agreement between values for the radiographs and the values obtained from the OSI SMA. Because one measure is taken from the internal bony skeleton (radiographs) and another measure is taken at the surface of the bony skeleton, we did not expect absolute agreement. But the range of the limits of agreement between values, for the radiographs and for the OSI SMA for all motions, was beyond what should be considered clinically acceptable. Random error can come from different sources including incorrect hook base placement, skin movement due to errors in palpation, and hook base or cranial cap movement with the OSI SMA. Table 2 shows the level of the lower hook base as noted on radiographs for each subject. There are no consistent differences for any of the levels indicated. Cranial cap movement, noted in an earlier reliability study (Petersen et al., 2000), had been corrected. Error is possible with placing lines on the radiographs. In an attempt to minimize potential error in locating landmarks on the cervical radiographs, two examiners (CMP and DS) studied the radiographs concurrently and determined together landmarks that were used for measurements. Polly et al. (1996) reported angular measurement differences of 101 for motion values from lumbar radiographs, even though interexaminer ICC’s ranged from 0.73 to 0.91. Acceptable reliability values for measures of cervical spinal motion from radiographs, measuring C2–C7, have been reported by Jackson et al. (1993). Measures involving the

ARTICLE IN PRESS C.M. Petersen et al. / Manual Therapy 13 (2008) 200–205

204

Table 1 Range of motion recorded for the OSI and radiographs (RAD) and the differences (DIFF) between these values for each subject; mean values for the OSI and radiograph and the limits of agreement for cervical motions S#

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Flexion

F F F F F F F M F F F F M M

X¯ UL LL

Extension

OSI

Rad

Diff

54.1 49.8 54 48.5 55.5 45.6 57.5 55.7 55.3 54.2 48.2 57.6 59.5 52.2

43 44 22 51 42 36 40 46 29 37 36 50 41 49

11.1 5.8 32 2.5 13.5 9.6 17.5 9.7 26.3 17.2 12.2 7.6 18.5 3.2

53.4 32 4

40.4

OSI 30.2 36.2 35 35.2 26.1 31.4 27 33.3 32.9 25.4 34.4 22.5 22.5 25.5

Left side bend

Right side bend

Rad

Diff

OSI

Rad

Diff

OSI

Rad

Diff

38 46 37 31 40 33 28 31 40 43 31 25 2 25

7.8 9.8 3.9 4.2 13.9 1.6 1 2.3 7.1 17.6 3.4 2.5 24.5 .5

34.7 30.8 33.7 39.5 31 30.1 32.1 29 34.1 32.7 28.4 34.2 31.8 33.1

21 38 33 33 20 32 35 32 28 26 26 31 32 25

13.7 7.2 .7 6.5 9 1.9 2.9 3 6.1 6.7 2.4 3.2 .2 8.1

38.6 28.6 33.5 38.3 30.7 31.8 36 32.8 42 33.5 32.3 32.1 32.8 30.1

34 27 22 24 31 35 28 20 25 29 33 29 28 27

4.6 1.6 11.5 14.3 .3 3.2 8 12.8 17 4.5 .7 3.1 4.8 3.1

32.5 18 6

26.3

33.8 16 3

27.5

29.8 17 20

31.9

S# ¼ subject number; F ¼ female; M ¼ male; X¯ ¼ mean; UL ¼ 95% upper limits of agreement; LL ¼ 95% lower limits of agreement. Table 2 Differences between the OSI SMA and radiograph values (degrees) per each subject by level of lower hook base as viewed on the radiograph Level

Subject

FLEX

EXT

SB(R)

SB(L)

Bottom of C6 Top of C7

8 1 9 6 2 3 4 5 7 10 11 12 13 14

9.7 11.1 26.3 9.6 5.8 32.0 2.5 13.5 17.5 17.2 12.2 7.6 18.5 3.2

2.3 7.8 7.1 1.6 9.8 3.9 4.2 13.9 1.0 17.6 3.4 2.5 24.5 0.5

3.0 13.7 6.1 1.9 7.2 0.7 6.5 9.0 2.9 6.7 2.4 3.2 0.2 8.1

12.8 4.6 17.0 3.2 1.6 11.5 14.3 0.3 8.0 4.5 0.7 3.1 4.8 3.1

Bottom of C7 Top of T1

angle between the occiput and C3 have been reported as being reliable by Phillips et al. (1999). Neither of these two studies incorporated measurements including both the occiput and lower cervical area. Because we had difﬁculty visualizing the occiput markings used by Phillips et al. (1999), we chose to use the markings as described in this paper. Our goal was to clearly visualize the superior surface of T1. Even though all radiographs were exposed to the T2 level, the superior surface of T1 was not clearly seen on four of the radiographs. As a result, we marked the vertebral surface most clearly seen closest to T1 (see Table 2).

Since gravity goniometers were used to deﬁne the limits of cervical motion, there is a certain amount of random error associated with the use of such devices. We feel that this potential error was minimized because the goniometers were ﬁxed to the cranial cap. We chose to require the subjects to stop at a ﬁxed point because the length of time that the subjects remained at maximum motion would be minimal with the OSI SMA and longer with radiographs. Prolonged positioning was occasionally required in preparation for the radiographic exposure. Prolonged stretching has been shown to result in an increase in range of motion of the spine (McGill and Brown, 1992) as well as the extremities (Madding et al., 1987). Another issue associated with the radiographic measurements was whether using the two superior landmark references (TMJ/OP) and the metal references on the mouth stick resulted in an accurate determination of the actual motion occurring. These landmarks were used because they were observable on the radiograph. Values obtained with these landmarks had shown high reliability (Herrmann, 1990) and measurement consistency (Ordway et al., 1999).

5. Conclusion For sagittal and frontal plane motions of the cervical spine, the values of the limits of agreement are too large to state that cervical radiographs and the OSI SMA provide similar clinical cervical spinal motion measures. The differences in values between the radiographs may

ARTICLE IN PRESS C.M. Petersen et al. / Manual Therapy 13 (2008) 200–205

be due to error in obtaining measures from the radiographs, and also due to variability of subject performance and variability of placement of the OSI SMA on the subject. The values obtained from this device will not be similar to values obtained from cervical radiographs, as measured in this study.

Acknowledgments The authors would like to express their appreciation to Meryl Mioni, RT (R) BS. Meryl Mioni is the Radiology Manager, Parkview Radiology, Palos Heights, Illinois, USA and took all the radiographs for the study. Funding for this study was provided primarily by a research grant from The Chicagoland Orthopedic Physical Therapy Study Group, Oak Park, Illinois, USA. References Altman DG, Bland JM. A note on the use of the intraclass correlation coefﬁcient in the evaluation of agreement between two methods of measurement. Computers in Biology and Medicine 1983;20(5): 337–40. Dvorak J, Antinnes JA, Manomar P, Loustalot D, Bonomo N. Age and gender related normal motion of the cervical spine. Spine 1992;17(10S):S393–8. Herrmann DB. Validity study of head and neck ﬂexion-extension motion comparing measurements of a pendulum goniometer and roentgenograms. Journal of Orthopaedic and Sports Physical Therapy 1990;11(9):414–8. Hoppenfeld S. Physical examination of the spine and extremities. New York: Appleton-Century-Crofts; 1976 (p. 109). Jackson BL, Harrison DD, Robertson GA, Barker WF. Chiropractic biophysics lateral cervical ﬁlm analysis reliability. Journal of Manipulative and Physiological Therapeutics 1993;16(6):384–91.

205

Johnson RD, Schuit D. Reliability of cervical range of motion testing with the OSI CA-6000 Spine Motion Analyzer on subjects with cervical pathology. Journal of Orthopaedic and Sports Physical Therapy 1993;17:56. Johnson RD, Schuit D, Petersen CM. Reliability of the OSI CA-6000 Spine Motion Analyzer system for active cervical range of motion measurement of normal subjects. Physical Therapy 1992; 72(Suppl):S62. Lantz CA, Chen J, Buch D. Clinical validity and stability of active and passive cervical range of motion with regard to total and unilateral uniplanar motion. Spine 1999;24(11):1082–9. Madding SW, Wong JG, Hallum A, Medieros JM. Effect of duration of passive stretch on hip abduction range of motion. Journal of Orthopaedic and Sports Physical Therapy 1987;8(8):410–6. Magee DJ. Orthopedic physical assessment, 4th ed. Philadelphia: WB Saunders; 2002 (p. 168). McGill SM, Brown S. Creep response of the lumbar spine to prolonged full ﬂexion. Clinical Biomechanics 1992;7:43–6. Munro BH, Vistintainer JL, Page EB. Statistical methods for health care research. JB Philadelphia: Lippincott Company; 1986 (p. 70). Ordway NR, Seymour RJ, Donelson RG, Hojnowski LS, Edwards WT. Cervical ﬂexion, extension, protrusion, and retraction: a radiographic segmental analysis. Spine 1999;24(3):240–7. Petersen CM, Johnson RD, Schuit D. Reliability of cervical range of motion using the OSI CA 6000 Spine Motion Analyzer on asymptomatic and symptomatic subjects. Manual Therapy 2000;5(2):82–8. Phillips FM, Phillips CS, Wetzel FT, Gelinas C. Occipitocervical neutral position: possible surgical implications. Spine 1999;24(8): 775–8. Polly DW, Kilkelly FX, McHale KA, Asplund LM, Mulligan M, Chang AS. Measurement of lumbar lordosis: evaluation of intraobserver, interobserver, and technique variability. Spine 1996; 21(13):1530–6. Portney LG, Watkins MP. Foundations of clinical research: applications to practice. 2nd ed. Stamford, CT: Appleton & Lange; 2000. Schuit D, Petersen C, Johnson R, Levine P, Knecht H, Goldberg D. Validity and reliability of measures obtained from the OSI CA6000 Spine Motion Analyzer for lumbar spinal motion. Manual Therapy 1997;2(4):206–15. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychological Bulletin 1979;86:420–8.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 206–212 www.elsevier.com/locate/math

Original article

Urinary incontinence in women with low back pain Kerstin Eliasson, Britt Elfving, Birgitta Nordgren, Eva Mattsson Department of Neurobiology, Care Sciences and Society, Division of Physiotherapy, Karolinska Institutet, Huddinge, SE-141 83 Stockholm, Sweden Received 15 September 2005; received in revised form 22 June 2006; accepted 7 December 2006

Abstract Urinary incontinence (UI) is a common female dysfunction, affecting women in all ages. Experienced physiotherapists and experts on low back pain (LBP) have since long observed and discussed the empirical association between LBP and UI. The aim of this study was to describe the occurrence of UI in women with LBP and to compare this group with a reference group of comparable age, language, culture and parity. The authors of this study had previously collected all original data from the reference group. A validated questionnaire concerning UI was answered by 200 consecutive women with LBP visiting specialised physiotherapy clinics in the area of Stockholm. Inclusion criteria were LBP, female, not pregnant, Swedish speaking, age between 17 and 45 years. Seventy-eight percent of the women with LBP reported UI. In comparison with the reference group, the prevalence of UI and signs of dysfunctional pelvic ﬂoor muscles (PFM) were greatly increased (po0.001) in the LBP group. Logistic regression analysis showed that the condition LBP and PFM dysfunction i.e. inability to interrupt the urine ﬂow (po0.001) were risk factors for UI irrespective of parity. Physiotherapists treating patients with LBP should be aware of possible incontinence problems in this patient group. r 2007 Elsevier Ltd. All rights reserved. Keywords: Female urinary incontinence; Low back pain; Pelvic ﬂoor dysfunction

1. Introduction Urinary incontinence (UI) is a common condition in women (Hunskaar et al., 2000, 2003). The prevalence of UI increases with age, for young adults the prevalence is reported to be 20–30% and around middle age 30–40% (Hannestad et al., 2000; Hunskaar et al., 2003). The prevalence of UI has, however, varied with the populations studied and the deﬁnitions and methods used. UI is since 2002 categorised as ‘‘the complaint of any involuntary leakage of urine’’ (Abrams et al., 2002). This deﬁnition has replaced the former deﬁnition of the International Continence Society (ICS); ‘‘involuntary loss of urine, which is objectively demonstrable and a social or hygienic problem’’ (Abrams et al., 1988). The most frequent form of UI in women is stress urinary incontinence (SUI), categorised as ‘‘the complaint of Corresponding author. Tel.: +46 8 524 888 30; fax: +46 8 52488813. E-mail address: [email protected] (K. Eliasson).

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.12.006

involuntary leakage on effort or exertion or on sneezing or coughing’’. Urge urinary incontinence (UUI) is categorised as ‘‘the complaint of involuntary leakage accompanied by or immediately proceeded by urgency’’ (Abrams et al., 2002) and has been more frequently reported in elderly women (Hunskaar et al., 2000). Since 2002 (Abrams et al., 2002) UI and urinary leakage (UL) are mostly used synonymously. Age, childbirth, lower urinary tract infections, pelvic surgery and factors increasing the intraabdominal pressure (IAP) such as overweight, straining at stool and physical exertion are traditionally considered to be risk factors for UI, alone or in combination (Hunskaar et al., 2000). The condition of the pelvic ﬂoor (PF) and particularly the pelvic ﬂoor muscles (PFM) are considered to be of importance (DeLancey, 1994), although no single factor completely explains UI aetiology. The association between the PFM and abdominal muscle activity has been suggested in former physiotherapeutic research (Wennergren et al., 1991) and activity in the PFM is associated with abdominal muscle activity

ARTICLE IN PRESS K. Eliasson et al. / Manual Therapy 13 (2008) 206–212

in general (Hemborg et al., 1985; Bo¨ and Stien, 1994). From that point of view, the PFM are part of the muscles surrounding the abdomen and necessary for development of IAP (Hemborg et al., 1985). Recent research has furthermore conﬁrmed a synergistic response between the deep abdominal muscles and the PFM (Sapsford et al., 2001; Neumann and Gill, 2002). Hence, the PFM seem to be an integral part of trunk and lumbo-pelvic stability, in addition to contributing to continence (Richardson et al., 1999). In healthy people, control of increased IAP is performed automatically as a feed-forward loop via the recruitment of the M Transversus Abdominis (TrA) together with the diaphragm and the PFM (Hodges and Gandevia, 2000; Sapsford et al., 2001) and lack of this early muscular recruitment is believed to cause instability in the trunk. Recently, Pool-Goudzwaard et al. (2005) reported symptoms of PF dysfunction and UI in women with post-pregnancy instability and pain in the pelvis. Low back pain (LBP) has been deﬁned as a condition of pain localised to the lumbar spine with or without radiation to the hip or leg (Waddell, 1996), which can be the result of several concurrent conditions whose aetiology is unknown. Experienced physiotherapists and experts on LBP have since long observed and discussed the empirical association between LBP and UI, while also observing the beneﬁts of the muscular stabilisation treatment on both LBP and UI (Richardson et al., 1999). Hypothetically, there might exist a relationship between LBP and UI but research regarding the relationship between UI and LBP is scarce. To our knowledge there are no studies describing the occurrence of UI in women with LBP. The aim of this study was therefore to describe the occurrence of UI in young women visiting physiotherapy clinics for treatment of LBP and compare the results with a reference group.

2. Material and methods 2.1. The study group Physiotherapy clinics in the Stockholm area specialising in musculoskeletal disorders were contacted by telephone and nine clinics agreed to participate. Participation included distribution of written information on the study and a questionnaire to be handed out to the women who met the inclusion criteria. The inclusion criteria were female, seeking physiotherapy for LBP, not pregnant, Swedish speaking, aged between 17 and 45 years. The ﬁrst 200 women who agreed to participate and who answered the questionnaire (see below) were included in the study. Seven women declined participation.

207

The physiotherapists were also asked to collect the completed questionnaires and record the number of patients who declined participation. The information letter stated that participation was voluntary and anonymous, and that the women could decline participation without any effect on their physiotherapy treatment and care. The completed questionnaire was put into an envelope, which was sealed by the patient before being given to the physiotherapist. The Ethics Committee at Karolinska Hospital, Stockholm, Sweden, approved the study. 2.2. The reference group In a prospective questionnaire study of 725 primiparous women with a mean age of 28 (range 17–43) years in Stockholm, the prevalence of UI before and during pregnancy and one year postpartum was surveyed (Eliasson et al., 2004, 2005). Before pregnancy the prevalence of UI was found to be 39%. Two percent of the women could be categorised having a more severe ‘‘signiﬁcant UI’’ according to the former deﬁnition of UI (Abrams et al., 1988), and most of them reported SUI. One year postpartum, the prevalence rates had increased to 49% and 7%, respectively (Eliasson et al., 2005). Risk factors signiﬁcantly associated with UI were found to be; connective tissue disorders, high impact physical activity, pelvic load, symptoms of dysfunctional and micturition habits (Eliasson et al., 2004, 2005). 2.3. Questionnaire The same questionnaire, used for the reference group (Eliasson et al., 2004, 2005), was answered by the 200 women with LBP. The questionnaire consisted of demographic questions and questions concerning etiological medical factors, UI, physical activity and micturition habits. The validity and reliability concerning the questions used and analysed in these studies were found to be good (X90% agreement) (Nordlander, 1994). Some questions concerning pregnancy and delivery were not considered relevant to this study and were excluded, while questions concerning LBP were added. The revised questionnaire was tested on ﬁve women aged 20–45 years. 2.4. Data analyses In the reference study the former deﬁnitions of the ICS were used (Abrams et al., 1988) but in the current study the new deﬁnition of UI (Abrams et al., 2002) was used and to describe a more severe leakage the expression ‘‘significant UI’’. UI was categorised as a positive response to the initial questions ‘‘Have you experienced UL?’’ with the

ARTICLE IN PRESS 208

K. Eliasson et al. / Manual Therapy 13 (2008) 206–212

alternatives occasionally, several times and often, and ‘‘Do you experience UL today’’ with the same alternatives. SUI was categorised as a positive response to the question ‘‘Do you leak urine during coughing, sneezing, lifting or physical activities?’’ UUI was categorised as a positive response to the question ‘‘Do you experience UL on your way to the bathroom?’’ Mixed urinary incontinence (MUI) was categorised as a positive response to both questions concerning SUI and UUI. Those who answered ‘‘no’’ or failed to answer but answered afﬁrmatively regarding leakage occasions and type of leakage were categorised as having UI. ‘‘Significant UI’’ was deﬁned according to the ICS deﬁnition (Abrams et al., 1988), including a positive response to the question ‘‘Do you leak more than a few drops’’, and a positive response to one of the questions ‘‘Do you use sanitary pads due to your leakage?’’, ‘‘Does leakage have a negative impact on your social life?’’ and ‘‘Does leakage have an impact on your psychological well-being?’’ Recurrent LBP was categorised as a positive response to the question ‘‘Have you experienced repeated periods of LBP?’’ Regular physical activity/exercise was categorised as a regular activity performed every week. The authors classiﬁed the physical activity according to its impact (loading effect) on the PF. Low impact activities were walking, bicycling, swimming and horseback riding. High impact activities, i.e. activities that raise the intraabdominal pressure and thus exert great force on the PF, were: gymnastics, running, jumping, dancing and ball sports. Due to the study population of patients with LBP, strength training was classiﬁed as low impact activity forming a natural part of the physiotherapy treatment. Overweight was categorised as a Body Mass Index (BMI) X25 kg/m2 and obese X30 kg/m2 (WHO, 1997). 2.5. Statistical analysis Results are presented as absolute and relative numbers. Univariate analyses using the Chi-square test were performed to test etiological factors and possible risk factors between continent and incontinent women with LBP. Univariate analyses using the Chi-square test were performed comparing the LBP group with original data from the reference group (Eliasson et al., 2004, 2005), separately according to parity. Data from the reference group and the study group were then included in the same data ﬁle for analysis. Signiﬁcant variables from these Chi-square tests were entered in multiple backward logistic regression analyses in order to ﬁnd the most important predicting factors for UI. Variables in

the regression model were removed in order of highest pvalue. Remaining variables were signiﬁcant at po0.05. The multiple logistic regression analyses were performed for nulliparous and parous women separately. For the statistical calculations, Statistica 7.1 and SOLO Statistical System 4.0 have been used.

3. Results 3.1. Demographics for women with LBP A total of 200 women answered the questionnaire. Their mean age was 36 (range 17–45) years. The majority of the women with LBP reported recurrent LBP (87%). Thirty-two percent (n ¼ 63) were nulliparous with a mean age of 30 (range 17–45) years and 68% (n ¼ 137) were parous with a mean age of 36 (range 21– 45) years, whereas 28 had delivered two, and one women three children. Of the parous women, 88% (n ¼ 120) had vaginal deliveries. BMI averaged 24 (range 14–41) kg/m2. Thirty-one percent were overweight and 9% were obese. Forty-ﬁve percent reported university studies and 39% hard work including heavy lifting. Twenty-four percent reported a chronic disease (most frequent were asthma and hypertension). Seventy-nine percent (n ¼ 158) of the women exercised regularly, 27% with high impact (most frequent; aerobics and jogging) and 73% with low impact activities (most frequent; walking and strength training). 3.2. Prevalence of UI in women with LBP Seventy-eight percent (n ¼ 155) of the women with LBP reported UI, of whom 73% occasionally, 23% several times and 4% often. Twenty-three percent (n ¼ 46) of the women could be classiﬁed as having ‘‘signiﬁcant UI’’. Nineteen percent used sanitary pads because of the leakage. Thirty-two percent were affected in their daily life, and 45% were psychologically affected. Of the 155 women with UI, 72% reported SUI, 1% UUI and 27% MUI. Seventy-seven percent reported UI when coughing, 45% when laughing, 39% when exercising, 30% during heavy lifting and 27% on their way to the bathroom. Nullipara did not report signiﬁcantly less UI compared with parous women, and parous women, delivered vaginally, did not report signiﬁcantly more UI than women delivered by Caesarean section. The prevalence of UI was 79% in women with recurrent LBP (n ¼ 173) and 65% in women with occasional LBP (n ¼ 26). In the ‘‘signiﬁcant UI’’ group (n ¼ 46), 93% (n ¼ 43) reported recurrent LBP. Univariate analyses with respect to UI for the women with LBP are presented in Table 1 and prevalence rates for nullipara and para, respectively, in Table 2.

ARTICLE IN PRESS K. Eliasson et al. / Manual Therapy 13 (2008) 206–212

209

Table 1 Demographic data, micturition habits and physical activity in continent and incontinent women with low back pain (n ¼ 200) Continent, n ¼ 45 (%)

Incontinent, n ¼ 155 (%)

P-value

Demographic data BMIX25 kg/m2 Hard work Chronic disease Varicose veins Pelvic load

8 15 8 11 9

(18%) (33%) (18%) (24%) (20%)

53 62 39 34 70

(34%) (40%) (25%) (22%) (46%)

0.034 0.419 0.304 40.5 0.002

Micturition habits Urethrotract infection Straining during micturition Inability to interrupt urine ﬂow ‘‘Preventive’’ micturition

26 10 13 32

(59%) (23%) (29%) (71%)

105 71 89 129

(68%) (46%) (58%) (84%)

0.261 0.005 o0.001 0.057

Physical activity Low impact High impact

23 (51%) 14 (31%)

93 (60%) 28 (18%)

40.5 0.423

Number (percent). P-values from the Chi-square test.

Table 2 Incontinence (UI and ‘‘signiﬁcant UI’’), demographic data, micturition habits and physical activity for the reference group when nulliparous (Eliasson et al., 2004) and parous (Eliasson et al., 2005) and women with low back pain (LBP) Nulliparous

UI ‘‘Sign UI’’ BMIX25 kg/m2 Chronic disease Varicose veins Pelvic load Urethrotract infection Straining during micturition Inability to interrupt urine ﬂow Low impact High impact

Parous

Ref group n ¼ 725

LBP group n ¼ 63

P-value

Ref group n ¼ 665

LBP group n ¼ 137

P-value

280 (39%) 14 (2%) 111 (17%) 72 (10%) 56 (8%)

44 11 10 14 8 21 31 22 30 32 19

o0.001 o0.001 40.5 0.004 0.208

323 (49%) 38 (6%) 189 (29%) 67 (10%) 55 (9%) 114 (18%) 332 (50%) 248 (37%) 192 (29%) 271 (41%) 201 (30%)

111 35 51 33 37 58 100 59 72 84 23

o0.001 o0.001 0.043 o0.001 o0.001 0.001 o0.001 0.174 o0.001 o0.001 0.001

364 232 217 88 355

(51%) (32%) (30%) (12%) (49%)

(70%) (18%) (16%) (22%) (13%) (33%) (50%) (35%) (48%) (51%) (30%)

40.5 40.5 0.004 o0.001 0.004

(81%) (26%) (37%) (24%) (27%) (42%) (72%) (44%) (53%) (61%) (18%)

Number (percent). P-value from the Chi-square test.

3.3. Comparison with the reference group Women with LBP were 36 years of age versus 29, and they reported a higher educational level than the reference group (po0.001). Differences in other variables between the women with LBP and the reference women when they were nulliparous and parous are shown in Table 2. The women with LBP reported signiﬁcantly more UI (po0.001) and more ‘‘signiﬁcant UI’’ (po0.001) than the reference group, and this applied to nulliparous as well as parous women (Table 2). Nulliparous women with LBP reported more MUI (po0.001) than the reference group. The multiple logistic regression analyses revealed that suffering from LBP and inability to interrupt the urine

flow signiﬁcantly increased the risk for UI in both the nulliparous (Table 3) and parous groups (Table 4). 4. Discussion In this explorative study including 200 women visiting physiotherapy clinics for treatment of LBP, the prevalence of UI as well as the rates of ‘‘signiﬁcant UI’’ showed to be higher compared with the reference group, and the statistical analysis showed that the condition LBP increased the risk for UI almost three times for parous women, and even more for nulliparous women. The prevalence of UI was extremely high, and has previously only been seen in studies of elderly women (Hellstro¨m et al., 1990; Sandvik et al., 1995) and trampolinists at the elite level (Eliasson et al., 2002), who load their PF extremely.

ARTICLE IN PRESS K. Eliasson et al. / Manual Therapy 13 (2008) 206–212

210

Table 3 Result from stepwise backward multiple logistic regression analysis for the nulliparous women with low back pain (LBP) (n ¼ 63) and the nulliparous reference group (n ¼ 725) with respect to urinary leakage

LBP condition Inability to interrupt urine ﬂow Straining during micturition Chronic disease

Log odds ratio (b)

SE

OR

CI

P-value

1.15 0.61 0.57 0.58

0.29 0.16 0.16 0.24

3.1 1.8 1.8 1.8

1.8–5.6 1.3–2.5 1.3–2.4 1.1–2.9

o0.001 o0.001 o0.001 0.015

Log odds ratio (b) values, standard error (SE), odds ratio (OR), 95% conﬁdence intervals (CI) and p-values are presented.

Table 4 Result from stepwise backward multiple logistic regression analysis for the parous women with low back pain (LBP) (n ¼ 137) and the parous reference group (n ¼ 665) with respect to urinary leakage

Inability to interrupt urine ﬂow Pelvic load LBP condition Urethrotract infection

Log odds ratio (b)

SE

OR

CI

P-value

1.37 1.30 1.00 0.61

0.19 0.22 0.26 0.17

3.9 3.6 2.7 1.8

2.7–5.7 2.3–5.6 1.6–4.5 1.3–2.5

o0.001 o0.001 o0.001 o0.001

Log odds ratio (b) values, standard error (SE), odds ratio (OR) and 95% conﬁdence intervals (CI) and p-values are presented.

A cross-sectional study design was chosen, as our intention was to explore whether UI was present in a group of younger women with LBP. There are studies within this topic reporting UI as sidetracks (Eisenstein et al., 1994; Dangaria, 1998; Pool-Goudzwaard et al., 2005) but no cross-sectional studies. This is a new research area and our study is a ﬁrst attempt to describe whether women with LBP suffer from UI more or less than ordinary women, or the same. A comparison with the reference group revealed that the ﬁndings should be followed up in a larger epidemiological population study. Recently, Smith et al. (2006) found, when they analysed data from an Australian study on women’s health, that disorders of breathing and continence had an association with self-reported back pain in 38,050 women. However, in our study the diagnose LBP was established, and the women were referred to specialised physiotherapist. A limitation in this study was lack of information as to whether the reference group suffered from LBP. LBP sufferers were most likely present in the reference group as well, but it is not known if the pain was of the recurrent or occasional kind. Possible association with UI is also unknown. Nevertheless, LBP was found to be a signiﬁcant risk factor for UI, and the signiﬁcance might have been even greater if compared to women without LBP complaints. The strength of this study was the use of a questionnaire utilised previously in a crosssectional study of women belonging to a normal Swedish population of the same age, culture and from the same geographical area as the LBP group.

There were some differences between the study samples. Most of the women in the parous LBP group had delivered one or two children, in comparison with the corresponding reference group, where the women had delivered one child (Eliasson et al., 2005). However, childbirth per se was not associated with UI, nor was the mode of delivery. There are diverging opinions on childbirth as a risk factor for UI. Most studies report that UI is most likely to occur in parous rather than nulliparous women (Jolleys, 1988) with an increased risk for every vaginal delivery (Jolleys, 1988). However, Thomas et al. (1980) found UI to be most common in parous women, but not until after four or more children, while in a recent Swedish study, Uustal Fornell et al. (2004) found an increased risk after more than two children. Overweight is reported to be a risk factor for UI (Dwyer et al., 1988; Hunskaar et al., 2000). The LBP group was more overweight (po0.001), but overweight did not inﬂuence the prevalence of UI neither in this study nor in the reference study (Eliasson et al., 2004, 2005). Hence, we do not regard differences between the LBP group and the reference group as responsible for the higher leakage rates in women with LBP. Surprisingly, women with LBP were signiﬁcantly less able to interrupt the urine ﬂow than women in the corresponding reference group (Table 2). In earlier studies (Eliasson et al., 2004, 2005), this inability has been reported to be closely associated with UI. This association was furthermore evident in the present study, where the statistical analysis found inability to interrupt the urine ﬂow to be a strong predictor of UI

ARTICLE IN PRESS K. Eliasson et al. / Manual Therapy 13 (2008) 206–212

increasing the risk four times for parous and twice for nulliparous women. Whether this aggravated dysfunction in the study group is due to LBP or not might be discussed, and if so, the causes could be delay in the muscular reaction or pain inhibition—present, former or both. The timing of the PFM response in relation to increases in IAP is probably more important than the strength of the PFM in promoting continence (Deindl et al., 1993). Another sign of dysfunctional PFM was perceived pelvic load, signiﬁcantly more reported by parous women with LBP than by the reference group. In the nulliparous group with LBP, every third woman reported this symptom. Due to lack of information from the nulliparous reference group this variable was, however, not included in the statistical analysis. Pelvic load is a symptom, described by women with genital prolapse, and has been found to be associated with decreased PFM strength (Samuelsson et al., 1999) and UI (Uustal Fornell et al., 2003; Eliasson et al., 2005). The question of whether subjectively reported pelvic load is another symptom of dysfunctional PFM due to LBP needs to be further illuminated. Although stress symptoms still were dominant, nulliparous women with LBP reported urge symptoms signiﬁcantly more often than the corresponding reference group and nulliparous women are usually not expected to exhibit this kind of symptoms. Interestingly, the association between LBP and urge symptoms has been reported previously (Eisenstein et al., 1994; Dangaria, 1998; Pool-Goudzwaard et al., 2005). The differences between the groups have been assessed, with the result that it cannot be excluded that the variation of incontinence rates is due to the occurrence of LBP. However, further research is needed to evaluate the association between LBP and UI. 5. Conclusion UI was reported by 78% of women with LBP. In comparison with the reference group, the prevalence of UI and ‘‘signiﬁcant UI’’ as well as signs of dysfunctional PFM was greatly increased. Logistic regression analysis showed that suffering from LBP and inability to interrupt the urine ﬂow increased the risk for UI irrespective of parity. Physiotherapists treating patients with LBP should be aware of possible leakage problems within this patient group. References Abrams P, Blaivas JG, Stanton SL, Andersson JT. The standardization of terminology of lower urinary tract function. Scandinavian Journal of Urology and Nephrology Supplementum 1988;114:5–19. Abrams P, Cardozo L, Fall M, Grifﬁths D, Rosier P, UImsten U, et al. The standardisation of terminology of lower urinary tract function:

211

report from the Standardisation Sub-committee of the International Continence Society. Neurourology and Urodynamics 2002; 21:167–78. Bo¨ K, Stien R. Needle EMG registration of striated urethral wall and pelvic ﬂoor muscle activity patterns during cough, Valsalva, abdominal, hip adductor and gluteal muscle contractions in nulliparous healthy females. Neurourology and Urodynamics 1994; 13:35–41. Dangaria TR. A case report of sacro-iliac joint dysfunction with urinary symptoms. Manual Therapy 1998;3:220–1. Deindl FM, Vodusek DB, Hesse U, Schussler B. Activity patterns of pubococcygeal muscles in nulliparous continent women. British Journal of Urology 1993;72:46–51. DeLancey JOL. Structural support of the urethra as it relates to stress urinary incontinence: the hammock hypothesis. American Journal of Obstetrics and Gynecology 1994;170:1713–23. Dwyer PL, Lee ECT, Hay DM. Obesity and urinary incontinence in women. British Journal of Obstetrics and Gynaecology 1988;95: 91–6. Eisenstein SM, Engelbrecht DJ, El Masry WS. Low back pain and urinary incontinence. A hypothetical relationship. Spine 1994;19: 1148–52. Eliasson K, Larson T, Mattsson E. Prevalence of stress incontinence in nulliparous elite trampolinists. Scandinavian Journal of Medicine and Science in Sports 2002;12:106–10. Eliasson K, Nordlander I, Mattsson E, Larson B, Hammarstro¨m M. Prevalence of urinary leakage in nulliparous women with respect to physical activity and micturition habits. International Urogynecology Journal 2004;15:149–53. Eliasson K, Nordlander I, Larson B, Hammarstro¨m M, Mattsson E. Inﬂuence of physical activity on urinary leakage in primiparous women. Scandinavian Journal of Medicine and Science in Sports 2005;15:87–94. Hannestad YS, Rortveit G, Sandvik H, Hunskaar S. A communitybased epidemiological survey of female urinary incontinence: the Norwegian EPINCONT Study. Journal of Clinical Epidemiology 2000;53:1150–7. Hellstro¨m L, Ekelund P, Milsom I, Mellstro¨m D. The prevalence of urinary incontinence and use of incontinence aids in 85-year-old men and women. Age Ageing 1990;19:386–9. Hemborg B, Moritz U, Lo¨wing H. Intra-abdominal pressure and trunk muscle activity during lifting. Scandinavian Journal of Rehabilitation Medicine 1985;17:25–38. Hodges PW, Gandevia SC. Changes in intra-abdominal pressure during postural and respiratory activation of the human diaphragm. Journal of Applied Physiology 2000;89:967–76. Hunskaar G, Lose G, Sykes D, Voss S. The prevalence of urinary incontinence in women in four European countries. BJU International 2003;93:324–30. Hunskaar S, Arnold EP, Burgio K, Diokno AC, Herzog AR, Mallett VT. Epidemiology and natural history of urinary incontinence. International Urogynecology Journal 2000;11:301–19. Jolleys JV. Reported prevalence of urinary incontinence in women in a general practice. British Medical Journal 1988;296:1300–2. Neumann P, Gill V. Pelvic ﬂoor and abdominal muscle interaction: EMG activity and intraabdominal pressure. International Urogynecology Journal 2002;13:125–32. Nordlander I. Construction and reliability-test of two questionnaires to register origin factors of urinary incontinence in primiparas. Project within Master of Science in Physiotherapy. Stockholm: Karolinska Institutet; 1994. Pool-Goudzwaard AL, Slieker ten Hove MCP, Vierhout ME, Mulder PH, Pool JJM, Snijders CJ, et al. Relations between pregnancy-related low back pain, pelvic ﬂoor activity and pelvic ﬂoor dysfunction. International Urogynecology Journal 2005;16: 468–74.

ARTICLE IN PRESS 212

K. Eliasson et al. / Manual Therapy 13 (2008) 206–212

Richardson C, Jull G, Hodges P, Hides J. Therapeutic exercise for spinal segmental stabilization in low back pain. Edinburgh: Churchill Livingstone; 1999. p. 134. ISBN 0 443 058024. Samuelsson EC, Victor FTA, Tibblin G, Sva¨rdsudd KF. Signs of genital prolaps in a Swedish population of women 20–59 years of age and possible related factors. American Journal of Obstetrics and Gynecology 1999;180:299–305. Sandvik H, Hunskaar S, Vanvik A, Bratt H, Seim A, Hermstad R. Diagnostic classiﬁcation of female urinary incontinence: an epidemiological survey corrected for validity. Journal of Clinical Epidemiology 1995;48:339–43. Sapsford RR, Hodges PW, Richardson CA, Cooper DH, Markwell SJ, Jull GA. Co-activation of the abdominal and pelvic ﬂoor muscles during voluntary exercises. Neurourology and Urodynamics 2001; 20:31–42. Smith MD, Russell A, Hodges PW. Disorders of breathing and continence have a stronger association with back pain than obesity and physical activity. Australian Journal of Physiotherapy 2006; 52:11–6.

Thomas TM, Plymat KR, Blannin J, Meade TW. Prevalence of urinary incontinence. British Medical Journal 1980;281: 1243–5. Uustal Fornell E, Wingren G, Kjolhede P. Prevalence of urinary and fecal incontinence and symptoms of genital prolapse in women. Acta Obstetricia et Gynecologica Scandinavica 2003;82: 280–6. Uustal Fornell E, Wingren G, Kjolhede P. Factors associated with pelvic ﬂoor dysfunction with emphasis on urinary and fecal incontinence and genital prolapse: an epidemiological study. Acta Obstetricia et Gynecologica Scandinavica 2004;83:383–9. Waddell G. Low back pain. A twentieth century health care enigma. Spine 1996;21:2820–5. Wennergren H, O¨berg BE, Sandstedt P. The importance of leg support for relaxation of the pelvic ﬂoor muscles. Scandinavian Journal of Urology and Nephrology 1991;25:205–13. WHO, World Health Organisation. Obesity. Preventing and managing the global epidemic. Report of a WHO consulting on obesity, Geneva; 1997. p. 3–5.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 213–221 www.elsevier.com/locate/math

Original article

Do ‘sliders’ slide and ‘tensioners’ tension? An analysis of neurodynamic techniques and considerations regarding their application Michel W. Coppietersa,b,, David S. Butlerb a

Division of Physiotherapy, School of Health and Rehabilitation Sciences, The University of Queensland, QLD 4072, St. Lucia Brisbane, Australia b Neuro-Orthopaedic Institute, 19 North Street, Adelaide 5000, Australia Received 7 April 2006; received in revised form 8 December 2006; accepted 15 December 2006

Abstract Despite the high prevalence of carpal tunnel syndrome and cubital tunnel syndrome, the quality of clinical practice guidelines is poor and non-invasive treatment modalities are often poorly documented. The aim of this cadaveric biomechanical study was to measure longitudinal excursion and strain in the median and ulnar nerve at the wrist and proximal to the elbow during different types of nerve gliding exercises. The results conﬁrmed the clinical assumption that ‘sliding techniques’ result in a substantially larger excursion of the nerve than ‘tensioning techniques’ (e.g., median nerve at the wrist: 12.6 versus 6.1 mm, ulnar nerve at the elbow: 8.3 versus 3.8 mm), and that this larger excursion is associated with a much smaller change in strain (e.g., median nerve at the wrist: 0.8% (sliding) versus 6.8% (tensioning)). The ﬁndings demonstrate that different types of nerve gliding exercises have largely different mechanical effects on the peripheral nervous system. Hence different types of techniques should not be regarded as part of a homogenous group of exercises as they may inﬂuence neuropathological processes differently. The ﬁndings of this study and a discussion of possible beneﬁcial effects of nerve gliding exercises on neuropathological processes may assist the clinician in selecting more appropriate nerve gliding exercises in the conservative and post-operative management of common neuropathies. r 2007 Elsevier Ltd. All rights reserved. Keywords: Neurodynamic test; Nerve biomechanics; Nerve gliding exercises; Nerve inﬂammation

1. Introduction The prevalence of carpal tunnel syndrome (CTS) is around 3% in the general population (Atroshi et al., 1999) and up to 15–20% in occupations that involve repetitive forceful hand tasks, such as meat processing (21%) (Gorsche et al., 1999) and ski manufacturing (15%) (Barnhart et al., 1991). Despite this high prevalence and the major socio-economic impact of Corresponding author. Division of Physiotherapy, School of Health and Rehabilitation Sciences, The University of Queensland, Building 84A, St. Lucia, QLD 4072 Brisbane, Australia. Tel.: +61 0 7 3365 4590; fax: +61 0 7 3365 1622. E-mail address: [email protected] (M.W. Coppieters).

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.12.008

CTS, the quality of clinical practice guidelines in hand therapy is poor (MacDermid, 2004). The Cochrane Database of Systematic Reviews evaluated the clinical efﬁcacy of the conservative management of CTS. O’Connor et al. (2003) concluded that there is signiﬁcant short-term beneﬁt from oral steroids, ultrasound, splinting, yoga and carpal bone mobilisation. However, Gerritsen et al. (2002) concluded that there is conﬂicting evidence for oral steroids and ultrasound, and that diuretics, vitamin B6, yoga, non-steroidal anti-inﬂammatory drugs, and laser acupuncture are ineffective in providing short-term symptom relief. Local corticosteroid injections provide shortterm relief, but there is no beneﬁt compared to placebo beyond one month (Marshall et al., 2002). In general,

ARTICLE IN PRESS 214

M.W. Coppieters, D.S. Butler / Manual Therapy 13 (2008) 213–221

long-term outcomes are poor (Wilson and Sevier, 2003), which prompts the question whether traditionally advocated modalities for CTS are adequate (Seradge et al., 2002). Most studies which examined the clinical efﬁcacy of nerve gliding exercises in CTS (Rozmaryn et al., 1998; Tal-Akabi and Rushton, 2000; Akalin et al., 2002; Seradge et al., 2002; Pinar et al., 2005, Baysal et al., 2006) were not included in the Cochrane review. However, several other reviews do suggest the use of nerve and tendon gliding exercises in the conservative treatment of CTS (Osterman et al., 2002; Burke et al., 2003; Michlovitz, 2004; Muller et al., 2004). Also in the post-operative management following carpal tunnel release, nerve and tendon gliding exercises are recommended (Nathan et al., 1993; Cook et al., 1995). The nerve gliding exercises included in the abovementioned studies aimed to induce sliding of the median nerve relative to its surrounding structures by performing joint movements that elongate the nerve bed (the tract formed by the structures that surround the nerve). The nerve bed was elongated at either the hand and wrist (Rozmaryn et al., 1998; Akalin et al., 2002; Pinar et al., 2005; Baysal et al., 2006) or over a longer section of the median nerve (Seradge et al., 2002), or by using the neurodynamic test for the median nerve as a mobilisation manoeuvre (Tal-Akabi and Rushton, 2000). There is indeed ample evidence that elongation of the nerve bed induces nerve gliding (Szabo et al., 1994; Byl et al., 2002; Dilley et al., 2003; Coppieters et al., 2006). Lengthening of the nerve bed also elongates the nerve which increases nerve tension and intraneural pressure. Whereas sustained elevated intraneural ﬂuid pressure reduces intraneural blood ﬂow in oedematous neuropathies (Myers et al., 1986), a dynamic variation in intraneural pressure when correctly applied may facilitate evacuation of intraneural oedema and reduce symptoms (Burke et al., 2003). In contrast, the increase in nerve strain associated with elongation of the nerve bed may also trigger ectopic discharges from mechanosensitive abnormal impulse generating sites (Dilley et al., 2005) and exacerbate symptoms. Techniques which facilitate nerve gliding by elongation of the nerve bed no longer cover the wide spectrum of nerve gliding exercises currently advocated. Combinations of movements in which elongation of the nerve bed at one joint is simultaneously counterbalanced by a reduction in the length of the nerve bed at an adjacent joint (‘sliding techniques’) have been promoted (Butler, 2000; Coppieters et al., 2004; Shacklock, 2005). The clinical assumption is that these sliding techniques result in a larger longitudinal excursion of the nerve with a minimal increase in strain. Although anatomical/biomechanical studies contributed to the validation of neurodynamic tests (Kleinrensink et al., 2000; Byl et al., 2002; Coppieters et al., 2006), to the best of our

knowledge, excursion and strain in peripheral nerves have never been evaluated from a therapeutic perspective. The aim of this study was to evaluate excursion and strain in the median and ulnar nerve for different types of nerve gliding exercises for CTS and cubital tunnel syndrome. The selected techniques reﬂect the different types of exercises suggested by Butler (2000), Coppieters et al. (2004) and Shacklock (2005).

2. Methods Longitudinal excursion and strain in the median and ulnar nerve during tensioning and sliding techniques and during isolated movements of the wrist and elbow were measured in two embalmed undisturbed male cadavers (age at time of death: 78 and 85 years). The study was approved by the Institutional Ethics Committee. 2.1. Excursion A digital Vernier calliper was used to measure longitudinal excursion of the median and ulnar nerve in relation to surrounding structures. A ﬁxed marker was screwed into the humerus and into the distal end of the radius. This marker consisted of a metal L-shaped pin which was placed perpendicularly over the nerve bed. As a mobile marker, a suture was placed around the peripheral nerve in the vicinity of the ﬁxed marker (Byl et al., 2002; Coppieters et al., 2006). 2.2. Strain Linear displacement transducers (Microstrain, Burlington, USA) (Fig. 1) with a stroke length of 6 mm and

Fig. 1. The linear displacement transducer (differential variable reluctance transducer) used to calculate nerve strain shown in a fresh cadaver. Note that the nerve which runs diagonally across the image is not the median or ulnar nerve. To insert the transducers, small windows of approximately 7 4 cm were made into the skin and underlying subcutaneous tissues. Disruption of surrounding soft tissues was minimised to limit alteration of local biomechanics.

ARTICLE IN PRESS M.W. Coppieters, D.S. Butler / Manual Therapy 13 (2008) 213–221

a resolution of 1.5 mm were used to measure strain in the median and ulnar nerve. One transducer was inserted into the median nerve just proximal to the carpal tunnel, while a second transducer was inserted in the nerve approximately 10 cm proximal to the medial epicondyle of the elbow. For the ulnar nerve, one transducer was inserted just proximal to the elbow. Previous studies have demonstrated the usefulness of these transducers to measure strain in peripheral nerves (Wright et al., 1996; Byl et al., 2002; Coppieters et al., 2006). For the median nerve, the anatomical position was used as the arbitrary reference position to which changes in strain were expressed. In this reference position, the arm was positioned in 101 shoulder abduction, with the elbow in submaximal extension (1701), the forearm in supination and the wrist in a neutral position (01 extension). Because the ulnar nerve buckled at the elbow in the anatomical position, the reference position for the ulnar nerve was 901 shoulder abduction, 901 elbow ﬂexion and the wrist in a neutral position. 2.3. Goniometry To guarantee accurate repositioning, twin axis electrogoniometers (SG65 and SG110, Biometrics, Blackwood, UK) were attached to the wrist, elbow and shoulder. 2.4. Mobilisation techniques (median nerve) 2.4.1. Tensioning technique With a tensioning technique, nerve gliding is obtained by moving one or several joints in such a manner that the nerve bed is elongated. In this study, the tensioning technique (Fig. 2A) consisted of simultaneous extension of the wrist (from 01 to 601) and elbow (from 901 to 1651), followed by a return to the starting position (wrist from 601 extension to neutral (01) and elbow from 1651 extension to 901). Full elbow extension was deﬁned as 1801. 2.4.2. Sliding technique A sliding technique consists of an alternation of combined movements of at least two joints in which one movement lengthens the nerve bed thus increasing tension in the nerve while the other movement simultaneously decreases the length of the nerve bed which unloads the nerve. These techniques aim to mobilise a nerve with a minimal increase in tension and are thought to result in a larger longitudinal excursion than techniques which simply elongate the nerve bed, such as tensioning techniques. In this study, the sliding technique (Fig. 2B) consisted of the alternation of elbow extension (loads the median nerve) and wrist ﬂexion (unloads the median nerve), with elbow ﬂexion (unloading) and wrist extension (loading). The

215

range of motion (ROM) was identical to the amplitudes in the tensioning technique (wrist: between 01 and 601 extension; elbow: between 901 and 1651 extension). 2.4.3. Single joint movements In addition to the combined movements (tensioning and sliding technique), the impact of isolated wrist and elbow movements was investigated. These isolated movements were performed with the neighbouring joint in a position which either unloaded (Fig. 2C and E) or pre-tensioned (Fig. 2D and F) the median nerve. In principle, these movements can be regarded as tensioning techniques as no simultaneous movement limits the lengthening of the nerve bed. Amplitudes for the wrist and elbow were identical to the ROM for the tensioning and sliding techniques mentioned above. 2.5. Mobilisation techniques (ulnar nerve) 2.5.1. Tensioning technique With the wrist in 601 extension, the elbow was ﬂexed (from 1501 to 651) and the shoulder abducted (from 601 to 1001). The technique was performed with supination as the forearm could not be pronated due to stiffness of the elbow joint. 2.5.2. Sliding technique Elbow extension (unloads the ulnar nerve) and shoulder abduction (loads the ulnar nerve) were alternated with elbow ﬂexion (loading) and shoulder adduction (unloading). Throughout the sliding technique, the wrist was maintained in 601 extension and the forearm in supination. The ROM for the elbow and shoulder were identical to the amplitudes used in the tensioning technique (elbow: between 651 and 1501; shoulder: between 601 and 1001 abduction). 2.6. Data collection and analysis The output from the strain gauges and electrogoniometers was connected to a data acquisition system (Micro 1401, Cambridge Electronic Design, Cambridge, UK) which sampled at 100 Hz using Spike 2 software. The uniqueness of this set-up was that continuous strain recordings could be made throughout the entire ROM. This allowed the construction of line ﬁgures expressing nerve strain in function of ROM, rather than only reporting discrete values associated with the start or end position of a technique. To our knowledge, this method has not yet been employed by other research groups to document variations in nerve strain. Another advantage of the set-up was that the investigator received real time feedback of the position of the wrist, elbow and shoulder via a computer screen, which also displayed the target angles. This method promoted accurate

ARTICLE IN PRESS 216

M.W. Coppieters, D.S. Butler / Manual Therapy 13 (2008) 213–221

Fig. 2. Mobilisation techniques and changes in strain in the median nerve. For each mobilisation technique, the corresponding diagrams in the middle and right column consist of three waveforms: the top waveform ( ) represents the change in strain in the median nerve at the wrist (middle column) or at the humerus (right column). The middle waveform (- -) shows the angle at the elbow and the bottom waveform (—) demonstrates the angle at the wrist. For the elbow, 1801 corresponds with full extension; for the wrist, 601 represents extension. For a detailed discussion, see text.

ARTICLE IN PRESS M.W. Coppieters, D.S. Butler / Manual Therapy 13 (2008) 213–221

repositioning of joint angles which justiﬁed comparison of excursion and strain between techniques. Throughout the experiment, the investigator was blinded to the output of the strain gauges. Note that we use the term ‘nerve gliding exercises’ to refer to a variety of different techniques, whereas the term ‘sliding technique’ is exclusively used to refer to techniques where movements in adjacent joints are combined to limit elongation of the nerve bed. 3. Results 3.1. Median nerve The amount of longitudinal excursion and differences in strain between the starting and end position for each mobilisation technique are summarised in Table 1. Fig. 2 demonstrates the continuous strain recordings in the median nerve in relation to the angles at the elbow and wrist for two consecutive repetitions for each mobilisation technique. Clear differences were observed between the sliding and tensioning technique and isolated single joint movements. Longitudinal excursion of the median nerve at the wrist was approximately twice as large for the sliding technique (12.6 mm) than for the tensioning technique (6.1 mm). In addition, strain in the median nerve at the wrist remained relatively constant during the sliding technique (variation of 0.8%) whereas it varied strongly during the tensioning technique (6.8%).

217

Peak strain in the median nerve was also substantially larger for the tensioning technique (+4.7%) than for the sliding technique (+2.7%). For single joint movements, wrist extension resulted in a slightly larger excursion of the median nerve at the wrist when the elbow was ﬂexed (9.8 mm) compared to when the elbow was extended (8.1 mm). Wrist extension with the elbow in ﬂexion was associated with a smaller peak strain (+2.8%) than when the wrist was mobilised while the nerve bed was already elongated by elbow extension (+4.6%). A similar pattern could be observed for elbow movements with the wrist in neutral and extension, although the excursion was larger when the elbow was moved with the wrist in extension. Paradoxically, longitudinal excursion of the median nerve at the humerus was larger for the tensioning (16.1 mm) than for the sliding (11.1 mm) technique. However, as anticipated, the tensioning technique resulted in a larger peak strain (+5.0%) than the sliding technique (+3.5%). For single joint movements, wrist extension resulted in small longitudinal movements of the median nerve relative to the humerus (0.8 and 1.8 mm) and small changes in strain (0.3% and 0.9%). In contrast, elbow movements were associated with large excursions (12.0 and 14.9 mm) and large changes in strain (5.2% and 5.3%). 3.2. Ulnar nerve The longitudinal movement of the ulnar nerve associated with the sliding technique (8.3 mm) was approximately

Table 1 Excursion and strain in the median nerve at the level of the wrist (A) and at the level of the humerus (B), and in the ulnar nerve, just proximal to the elbow (C) Excursion (mm)

Strain increase during the technique; from minimal to maximal strain value (relative to reference position)

A. Median nerve at the wrist Tensioning technique Sliding technique Wrist movement, with elbow in ﬂexion Wrist movement, with elbow in extension Elbow movement, with wrist in neutral Elbow movement, with wrist in extension

6.1 12.6 9.8 8.1 1.7 4.4

6.8%; 0.8%; 9.5%; 3.0%; 3.0%; 1.9%;

from 2.0% to+4.7% from+1.9% to+2.7% from 6.7% to+2.8% from+1.6% to+4.6% from 0.5% to+2.4% from+2.8% to+4.7%

B. Median nerve at the humerus Tensioning technique Sliding technique Wrist movement, with elbow in ﬂexion Wrist movement, with elbow in extension Elbow movement, with wrist in neutral Elbow movement, with wrist in extension

16.5 11.1 0.8 1.8 12.0 14.9

6.0%; 4.1%; 0.3%; 0.9%; 5.2%; 5.3%;

from 1.0% to+5.0% from 0.6% to+3.5% from+0.9% to+1.2% from+4.9% to+5.8% from 1.6% to+3.6% from 0.3% to+5.0%

C. Ulnar nerve proximal to the elbow Tensioning technique Sliding technique

3.8 8.3

(9.8%)a; from ( 6.6%)a to+3.2% 0.4%; from+0.3% to+0.7%

The values represent the mean of three consecutive repetitions. a Due to buckling of the ulnar nerve in the relaxed position of the tensioning technique (shoulder adduction–elbow extension), the minimal strain value and the derived increase in strain are not precise, hence they have been placed between brackets. The maximal strain value was not affected.

ARTICLE IN PRESS 218

M.W. Coppieters, D.S. Butler / Manual Therapy 13 (2008) 213–221

double the amount of excursion observed during the tensioning technique (3.8 mm). Strain in the ulnar nerve during the tensioning and sliding technique is illustrated in Fig. 3. As for the median nerve, changes in strain with the sliding technique were minimal (0.4%) and peak strain (+0.7%) was substantially smaller than during the tensioning technique (+3.2%). Due to buckling of the ulnar nerve in the relaxed position of the tensioning technique, it was impossible to accurately determine the change in strain during the tension technique. It was however obvious that the increase in strain was substantially larger than during the sliding technique. 4. Discussion The ﬁndings clearly demonstrate that different types of nerve gliding exercises have largely different mechanical effects on the peripheral nervous system. Longitudinal excursion and strain associated with a particular joint movement is strongly inﬂuenced by the position or simultaneous movement of an adjacent joint. For example, when considering the median nerve at the wrist, wrist extension resulted in a distal glide of approximately 9 mm. This excursion increased by 30% (to 12.6 mm) if wrist extension was accompanied by elbow ﬂexion, a movement that reduces the length of the nerve bed and decreases strain in the median nerve around the elbow and thus facilitates the distal glide of the nerve at the wrist (sliding technique). Similarly, distal excursion decreased by 30% (to 6.1 mm) if wrist extension was accompanied by elbow extension, which increases the length of the nerve bed and increases tension in the nerve at the elbow and thus hinders distal excursion (tensioning technique). A similar trend was observed for the ulnar nerve at the elbow: nerve gliding was substantially larger for the sliding technique than for the tensioning technique (8.3 versus 3.8 mm).

Fig. 3. Changes in strain in the ulnar nerve just proximal to the elbow during a tensioning (left panel) and sliding technique (right panel). The top waveform ( ) represents the change in strain in the ulnar nerve; the middle waveform (- -) shows the angle at the elbow and the bottom waveform (—) demonstrates the angle at the shoulder. Continuous data for half a cycle are presented (from the starting position to the end position).

As anticipated, the peak in nerve strain was large for techniques which involved simultaneous elongation of the nerve bed at adjacent joints. Previous research has demonstrated that nerve strain can be transmitted along a long section of a peripheral nerve (e.g., an increase in tibial nerve strain at the tarsal tunnel following hip ﬂexion in a modiﬁed straight leg raising test; Coppieters et al., 2006). Similarly, a decrease in length of the nerve bed reduces nerve strain at adjacent joints (e.g., shoulder adduction reduces strain in the median nerve at the elbow and wrist; Wright et al., 1996). One of the advances of this study is that we demonstrated that when movements which increase and decrease the length of the nerve bed are performed simultaneously at adjacent joints, nerve gliding occurs with almost no increase in nerve strain. Facilitation of nerve gliding in this manner (sliding technique) is markedly different to inducing nerve gliding by elongating the nerve bed and increasing nerve strain (tensioning technique or isolated joint movements). Overall, sliding techniques resulted in the largest excursion. However, median nerve gliding at the humerus revealed a cumulative effect of joint movements that elongate the nerve bed if both movements are located distally from the location of the excursion measurements. This can be explained by the fact that a nerve slides toward the joint where the nerve bed is elongated (Wright et al., 2001; Boyd et al., 2005; Coppieters et al., 2006) and a cumulative effect occurs if both movements facilitate nerve gliding in the same direction. However, this excursion was associated with a relatively large increase in nerve strain, which may be contraindicated in more acute conditions. Reduced transverse gliding of the median nerve in the carpal tunnel has been demonstrated in patients with CTS (Nakamichi and Tachibana, 1995; Allmann et al., 1997; Erel et al., 2003), but the ﬁndings regarding reduced longitudinal gliding are less conclusive. Smaller differences in latencies of action potentials between measurements with the wrist in ﬂexion and extension in patients with CTS compared to healthy controls were interpreted to reﬂect a smaller longitudinal glide during wrist movements (Valls-Sole et al., 1995). Erel et al. (2003) showed that median nerve movement following metacarpophalangeal ﬂexion was 20% smaller in patients with CTS. However, this difference failed to reach the level of signiﬁcance and the authors concluded that longitudinal excursion in patients with CTS is normal. Tuzuner et al. (2004) measured longitudinal nerve gliding before and immediately after endoscopic carpal tunnel release and noted no immediate difference. Although these studies do not irrefutably indicate or rule out that longitudinal excursion is restricted in CTS, the beneﬁcial effects of nerve gliding exercises for CTS and other neuropathies are unlikely to relate (solely) to restoration of restricted longitudinal nerve motion. An exploration of therapeutic

ARTICLE IN PRESS M.W. Coppieters, D.S. Butler / Manual Therapy 13 (2008) 213–221

potential needs to include effects on pathophysiological processes. These effects should be considered at the site of injury, but also remotely at the dorsal root ganglion and central nervous system. The awareness that different types of nerve gliding exercises have markedly different mechanical effects on the peripheral nervous system may result in the selection of safer and pathology-targeted techniques. The data supports the contention and would allow the suggestion that a sliding technique is less aggressive and may be more appropriate for acute injuries, post-operative management and situations which may lead to nerve irritation and entrapment such as bleeding and inﬂammation around the nerve. A tensioning technique may reduce intraneural swelling and circulatory compromise via ﬂuctuating effects on intraneural pressure. Dynamically altering intraneural pressure may result in a ‘pumping action’ or ‘milking effect’ with beneﬁcial effects on nerve hydration (Rozmaryn et al., 1998). This ‘milking effect’ may also be present with a sliding technique, when mobilising the median nerve through areas of increased pressure, such as the carpal tunnel in patients with CTS. We propose that the likely milking or pumping effects of sliding techniques performed with respect to reasoned pathobiology and individual patient presentation may enhance dispersal of local inﬂammatory products in and around nerves. Nerve inﬂammation is frequently associated with damaged and diseased nerves. The ‘inﬂammatory soup’ comprises ﬂuids and cells including enzymes, acids, prostaglandins, histamine and macrophages. It creates an acidic environment which is known to enhance peripheral nerve sensitivity (Maves et al., 1995; Steen et al., 1996). Inﬂamed nerves are also immunoreactive, with proinﬂammatory cytokines such as tumour necrosis factor alpha (TNFa) capable of producing spontaneous discharge in sensory ﬁbres by forming its own ion channels (Baldwin et al., 1996; Sorkin et al., 1997), a process enhanced by the acidic environment of inﬂammation. TNFa and other proinﬂammatory cytokines may also damage myelin and alter the blood nerve barrier (Watkins and Maier, 2002). Nerve gliding exercises may also limit ﬁbroblastic activity and minimise scar formation via normal and early use of mesoneurial gliding tissues (Millesi et al., 1995). They may prevent post-operative adhesions and may decrease venous engorgement and elevated endoneurial ﬂuid pressure. Injured and irritated nerves frequently become pressurised as endoneurial ﬂuid pressure increases. This is associated with endoneurial oedema, ischaemia, slowing and pooling of axoplasmic ﬂow, and disruption of pressure gradients which normally allow adequate perfusion of blood into neurones (Sunderland, 1976; Myers et al., 1986; Lundborg, 1988). Patients with CTS typically show higher carpal tunnel pressures (Gelberman et al., 1981), which may have a detrimental effect on nerve function

219

and integrity (Rempel et al., 1999; Mackinnon, 2002; Diao et al., 2005). However, exercises that induce dynamic changes in pressure may have a positive effect. Upon full excursion of the ﬂexor tendons of the ﬁngers, the lumbrical muscles travel in and back out of the carpal tunnel, inﬂuencing carpal tunnel pressure (Cobb et al., 1995). When performed dynamically, the pumping effect may facilitate venous return, oedema dispersal and decrease of pressure inside the perineurium (Totten and Hunter, 1991; Burke et al., 2003). Blood ﬂow to the wrist and hand also increased after hand exercises (Hansford et al., 1986) thereby increasing circulation, axonal transport, nutrition, and oxygenation to the median nerve in the carpal tunnel. There may also be beneﬁcial remote effects of sliding techniques. Ideal management of a local nerve injury reduces sensitivity and restores function, thus easing the threat value of the injury. This would be likely to minimise the potential for ion channel upregulation in dorsal root ganglia and the central nervous system, and limit the potential for dorsal horn and brain changes. Sliding techniques involve large amplitudes, can be performed passively or actively, and can be integrated into metaphorical movements or dance and as such can distract the patient from the condition (Butler, 2005). Patients with CTS are known to have altered somatosensory hand representations in the brain (Druschky et al., 2000; Tecchio et al., 2002). Sliding techniques allow large range neurally non-aggressive movements to be constructed, often allowing movement to be presented in novel ways to the brain, uncoupling learnt expectations of pain. The larger range movements are likely to decrease fear of movement and they may well assist in remapping altered representations. A limitation of the present study is that only two cadavers were available at the time of testing. However, because a similar trend was observed for the median and ulnar nerve, and because the ﬁndings are in line with the theoretical construct, we have no reasons to believe that the ﬁndings are not representative. Additional testing is currently carried out to further evaluate the strength of the concepts presented in this manuscript. Another potential limitation is the use of embalmed cadavers. However, the magnitude of strain reported in this study is very similar to the values reported by Byl et al. (2002) who used fresh cadavers. This suggests that embalmment may not dramatically alter the mechanical properties of nerves. The use of a repeated-measures design and the large relative differences between techniques also adds to the legitimacy of the main ﬁndings.

Acknowledgements The authors wish to thank Ali Alshami for his assistance during the measurements and the Department

ARTICLE IN PRESS 220

M.W. Coppieters, D.S. Butler / Manual Therapy 13 (2008) 213–221

of Anatomy and Developmental Biology of The University of Queensland for their assistance and the use of the facilities.

References Akalin E, El O, Peker O, Senocak O, Tamci S, Gulbahar S, et al. Treatment of carpal tunnel syndrome with nerve and tendon gliding exercises. American Journal of Physical Medicine & Rehabilitation 2002;81(2):108–13. Allmann KH, Horch R, Uhl M, Guﬂer H, Altehoefer C, Stark GB, et al. MR imaging of the carpal tunnel. European Journal of Radiology 1997;25(2):141–5. Atroshi I, Gummesson C, Johnsson R, Ornstein E, Ranstam J, Rosen I. Prevalence of carpal tunnel syndrome in a general population. Journal of the American Medical Association 1999;282(2):153–8. Baldwin RL, Stolowitz ML, Hood L, Wisnieski BJ. Structural changes of tumor necrosis factor alpha associated with membrane insertion and channel formation. Proceedings of the National Academy of Sciences of the United States of America 1996;93(3):1021–6. Barnhart S, Demers PA, Miller M, Longstreth WT, Rosenstock L. Carpal tunnel syndrome among ski manufacturing workers. Scandinavian Journal of Work, Environment & Health 1991;17(1):46–52. Baysal O, Altay Z, Ozcan C, Ertem K, Yologlu S, Kayhan A. Comparison of three conservative treatment protocols in carpal tunnel syndrome. International Journal of Clinical Practice 2006; 60(7):820–8. Boyd BS, Puttlitz C, Gan J, Topp KS. Strain and excursion in the rat sciatic nerve during a modiﬁed straight leg raise are altered after traumatic nerve injury. Journal of Orthopaedic Research 2005;23(4):764–70. Burke FD, Ellis J, McKenna H, Bradley MJ. Primary care management of carpal tunnel syndrome. Postgraduate Medical Journal 2003;79(934):433–7. Butler DS. The sensitive nervous system. Adelaide: Noigroup Publications; 2000. Butler DS. The neurodynamics techniques. DVD and handbook. Adelaide: Noigroup Publications; 2005. Byl C, Puttlitz C, Byl N, Lotz J, Topp K. Strain in the median and ulnar nerves during upper-extremity positioning. Journal of Hand Surgery [Am] 2002;27(6):1032–40. Cobb TK, An KN, Cooney WP. Effect of lumbrical muscle incursion within the carpal tunnel on carpal tunnel pressure: a cadaveric study. Journal of Hand Surgery [Am] 1995;20(2):186–92. Cook AC, Szabo RM, Birkholz SW, King EF. Early mobilization following carpal tunnel release. A prospective randomized study. Journal of Hand Surgery [Br] 1995;20(2):228–30. Coppieters MW, Alshami AM, Babri AS, Souvlis T, Kippers V, Hodges PW. Strain and excursion of the sciatic, tibial and plantar nerves during a modiﬁed straight leg raising test. Journal of Orthopaedic Research 2006;24(9):1883–9. Coppieters MW, Bartholomeeusen KE, Stappaerts KH. Incorporating nerve-gliding techniques in the conservative treatment of cubital tunnel syndrome. Journal of Manipulative and Physiological Therapeutics 2004;27(9):560–8. Diao E, Shao F, Liebenberg E, Rempel D, Lotz JC. Carpal tunnel pressure alters median nerve function in a dose-dependent manner: a rabbit model for carpal tunnel syndrome. Journal of Orthopaedic Research 2005;23(1):218–23. Dilley A, Lynn B, Greening J, DeLeon N. Quantitative in vivo studies of median nerve sliding in response to wrist, elbow, shoulder and neck movements. Clinical Biomechanics 2003;18(10):899–907. Dilley A, Lynn B, Pang SJ. Pressure and stretch mechanosensitivity of peripheral nerve ﬁbres following local inﬂammation of the nerve trunk. Pain 2005;117(3):462–72.

Druschky K, Kaltenhauser M, Hummel C, Druschky A, Huk WJ, Stefan H, et al. Alteration of the somatosensory cortical map in peripheral mononeuropathy due to carpal tunnel syndrome. Neuroreport 2000;11(17):3925–30. Erel E, Dilley A, Greening J, Morris V, Cohen B, Lynn B. Longitudinal sliding of the median nerve in patients with carpal tunnel syndrome. Journal of Hand Surgery [Br] 2003;28(5): 439–43. Gelberman RH, Hergenroeder PT, Hargens AR, Lundborg GN, Akeson WH. The carpal tunnel syndrome. A study of carpal canal pressures. Journal of Bone and Joint Surgery [Am] 1981;63(3): 380–3. Gerritsen AA, de Krom MC, Struijs MA, Scholten RJ, de Vet HC, Bouter LM. Conservative treatment options for carpal tunnel syndrome: a systematic review of randomised controlled trials. Journal of Neurology 2002;249(3):272–80. Gorsche RG, Wiley JP, Renger RF, Brant RF, Gemer TY, Sasyniuk TM. Prevalence and incidence of carpal tunnel syndrome in a meat packing plant. Occupational and Environmental Medicine 1999; 56(6):417–22. Hansford T, Blood H, Kent B, Lutz G. Blood ﬂow changes at the wrist in manual workers after preventive interventions. Journal of Hand Surgery [Am] 1986;11(4):503–8. Kleinrensink GJ, Stoeckart R, Mulder PG, Hoek G, Broek T, Vleeming A, et al. Upper limb tension tests as tools in the diagnosis of nerve and plexus lesions. Anatomical and biomechanical aspects. Clinical Biomechanics 2000;15(1):9–14. Lundborg G. Nerve injuries and repair. Edinburgh: Churchill Livingstone; 1988. MacDermid JC. The quality of clinical practice guidelines in hand therapy. Journal of Hand Therapy 2004;17(2):200–9. Mackinnon SE. Pathophysiology of nerve compression. Hand Clinics 2002;18(2):231–41. Marshall S, Tardif G, Ashworth N. Local corticosteroid injection for carpal tunnel syndrome. The Cochrane Database of Systematic Reviews 2002;4:CD001554. Maves TJ, Gebhart GF, Meller ST. Continuous infusion of acidiﬁed saline around the rat sciatic nerve produces thermal hyperalgesia. Neuroscience Letters 1995;194(1–2):45–8. Michlovitz SL. Conservative interventions for carpal tunnel syndrome. The Journal of Orthopaedic and Sports Physical Therapy 2004; 34(10):589–600. Millesi H, Zoch G, Reihsner R. Mechanical properties of peripheral nerves. Clinical Orthopaedics 1995;314):76–83. Muller M, Tsui D, Schnurr R, Biddulph-Deisroth L, Hard J, MacDermid JC. Effectiveness of hand therapy interventions in primary management of carpal tunnel syndrome: a systematic review. Journal of Hand Therapy 2004;17(2):210–28. Myers RR, Murakami H, Powell HC. Reduced nerve blood ﬂow in edematous neuropathies: a biomechanical mechanism. Microvascular Research 1986;32(2):145–51. Nakamichi K, Tachibana S. Restricted motion of the median nerve in carpal tunnel syndrome. Journal of Hand Surgery [Br] 1995; 20(4):460–4. Nathan PA, Meadows KD, Keniston RC. Rehabilitation of carpal tunnel surgery patients using a short surgical incision and an early program of physical therapy. Journal of Hand Surgery [Am] 1993;18(6):1044–50. O’Connor D, Marshall S, Massy-Westropp N. Non-surgical treatment (other than steroid injection) for carpal tunnel syndrome. The Cochrane Database of Systematic Reviews 2003;1:CD003219. Osterman AL, Whitman M, Porta LD. Nonoperative carpal tunnel syndrome treatment. Hand Clinics 2002;18(2):279–89. Pinar L, Enhos A, Ada S, Gungor N. Can we use nerve gliding exercises in women with carpal tunnel syndrome? Advances in Therapy 2005;22(5):467–75.

ARTICLE IN PRESS M.W. Coppieters, D.S. Butler / Manual Therapy 13 (2008) 213–221 Rempel D, Dahlin L, Lundborg G. Pathophysiology of nerve compression syndromes: response of peripheral nerves to loading. Journal of Bone and Joint Surgery [Am] 1999;81(11):1600–10. Rozmaryn LM, Dovelle S, Rothman ER, Gorman K, Olvey KM, Bartko JJ. Nerve and tendon gliding exercises and the conservative management of carpal tunnel syndrome. Journal of Hand Therapy 1998;11(3):171–9. Seradge H, Parker W, Baer C, Mayﬁeld K, Schall L. Conservative treatment of carpal tunnel syndrome: an outcome study of adjunct exercises. The Journal of the Oklahoma State Medical Association 2002;95(1):7–14. Shacklock MO. Clinical neurodynamics: a new system of musculoskeletal treatment. Edinburgh: Elsevier Butterworth-Heinemann; 2005. Sorkin LS, Xiao WH, Wagner R, Myers RR. Tumour necrosis factoralpha induces ectopic activity in nociceptive primary afferent ﬁbres. Neuroscience 1997;81(1):255–62. Steen KH, Steen AE, Kreysel HW, Reeh PW. Inﬂammatory mediators potentiate pain induced by experimental tissue acidosis. Pain 1996;66(2-3):163–70. Sunderland S. The nerve lesion in the carpal tunnel syndrome. Journal of Neurology, Neurosurgery, and Psychiatry 1976;39(7):615–26. Szabo RM, Bay BK, Sharkey NA, Gaut C. Median nerve displacement through the carpal canal. The Journal of Hand Surgery [Am] 1994;19(6):901–6. Tal-Akabi A, Rushton A. An investigation to compare the effectiveness of carpal bone mobilisation and neurodynamic mobilisation as

221

methods of treatment for carpal tunnel syndrome. Manual Therapy 2000;5(4):214–22. Tecchio F, Padua L, Aprile I, Rossini PM. Carpal tunnel syndrome modiﬁes sensory hand cortical somatotopy: a MEG study. Human Brain Mapping 2002;17(1):28–36. Totten PA, Hunter JM. Therapeutic techniques to enhance nerve gliding in thoracic outlet syndrome and carpal tunnel syndrome. Hand Clinics 1991;7(3):505–20. Tuzuner S, Ozkaynak S, Acikbas C, Yildirim A. Median nerve excursion during endoscopic carpal tunnel release. Neurosurgery 2004;54(5):1155–60. Valls-Sole J, Alvarez R, Nunez M. Limited longitudinal sliding of the median nerve in patients with carpal tunnel syndrome. Muscle & Nerve 1995;18(7):761–7. Watkins LR, Maier SF. Beyond neurons: evidence that immune and glial cells contribute to pathological pain states. Physiological Reviews 2002;82(4):981–1011. Wilson JK, Sevier TL. A review of treatment for carpal tunnel syndrome. Disability and Rehabilitation 2003;25(3):113–9. Wright TW, Glowczewskie Jr. F, Cowin D, Wheeler DL. Ulnar nerve excursion and strain at the elbow and wrist associated with upper extremity motion. The Journal of Hand Surgery [Am] 2001; 26(4):655–62. Wright TW, Glowczewskie F, Wheeler D, Miller G, Cowin D. Excursion and strain of the median nerve. Journal of Bone and Joint Surgery [Am] 1996;78(12):1897–903.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 222–231 www.elsevier.com/locate/math

Original article

Rasch analysis of three versions of the Oswestry Disability Questionnaire Megan Davidson,1 School of Physiotherapy, La Trobe University, Vic. 3058, Australia Received 11 October 2006; received in revised form 9 January 2007; accepted 17 January 2007

Abstract The purpose of the study was to explore the construct validity of three versions of the Oswestry Disability Questionnaire for low back pain using Rasch analysis. The three versions of the ODQ share 9 items and differ on one other. About 100 patients with non-speciﬁc low back pain seeking physiotherapy treatment at hospital outpatient departments and physiotherapy private practices completed the 12 Oswestry items as part of a battery of questionnaires. Rasch analysis revealed that four items (Personal Care, Standing, Sex Life and Social Life) had disordered response thresholds and one item (Walking) showed differential item functioning by age. The 10 standard Oswestry items and a modiﬁed version in which Sex Life is replaced by Work/Housework showed adequate overall ﬁt to the Rasch model (w2 P4.01). The third version, in which Sex Life is replaced by Changing Degree of Pain, did not ﬁt the model (w2 P ¼ .006) and the Changing Degree of Pain item was misﬁtting (residual 2.34, P ¼ .007). These ﬁndings suggest that either of the ﬁrst two of the three versions of this widely used low back pain outcome measure should be selected over the third. Users should also be aware that for some items the rating scale steps do not perform as intended. r 2007 Elsevier Ltd. All rights reserved. Keywords: Questionnaires; Validity; Low back pain; Rasch analysis

1. Introduction The Oswestry Disability Questionnaire (ODQ) is one of the oldest self-report questionnaires for measuring functional outcomes in patients with low back pain and remains widely used (Fairbank et al., 1980; Grotle et al., 2004). The ODQ was developed as a clinical assessment tool that would provide an estimate of disability expressed as a percentage score. Ten sections or items assess pain, personal care, lifting, walking, sitting, standing, sleeping, sex life, social life and travelling. The developers provided little detail on how the items Corresponding author. Tel.: +61 3 9479 5798; fax: +61 3 9479 5766. E-mail address: [email protected]. 1 Department/Institution to which the work should be attributed: Musculoskeletal Research Centre, School of Physiotherapy, La Trobe University, Vic. 3086, Australia.

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.01.008

were selected, saying only that the activities chosen were those most relevant to people with low back pain. Each item of the ODQ has 6 response choices arranged in order of difﬁculty and the respondent is asked to select the response ‘‘that most closely describes you today’’. For example, the Sitting section responses are I can sit in any chair as long as I like, I can only sit in my favourite chair as long as I like, Pain prevents me sitting more than 1 hr, Pain prevents me from sitting more than 30 min, Pain prevents me from sitting more than 10 min and Pain prevents me from sitting at all. A score of 0 is awarded if the ﬁrst response option is selected, through to 5 for the last option. A total score is calculated by summing the individual items scores, dividing by the total possible score (adjusted if any items are missed) and multiplied by 100. The possible score range is 0–100 and a higher score indicates greater disability. The ODQ is therefore an atypical

ARTICLE IN PRESS M. Davidson / Manual Therapy 13 (2008) 222–231

questionnaire because there is no consistent rating scale used across all items: instead, each step of each item has its own deﬁnition. The ODQ was modiﬁed by Baker et al. (1989) who removed references to medication from the Pain and Sleeping items, thereby improving the relevance of these items to people not taking medication. Davidson and Keating (2002) further modiﬁed this version by replacing miles with kilometres in the Walking section. A modiﬁed version sometimes called the Chiropractic version (Hudson-Cook et al., 1989) replaced Sex Life with a new item called Changing Degree of Pain. This version has been criticised for including a transitional rating, which is conceptually different from the other items that ask about pain intensity and activity limitations (Fairbank and Pynsent, 2000). More recently, Fritz and Irrgang (2001) reported a version that replaced Sex Life with a new item called Employment/ Homemaking. This modiﬁcation added an aspect of activity/participation that is otherwise absent from the ODQ. The developers recommend Version 2.0 of the Oswestry (Fairbank and Pynsent, 2000), which instructs patients to answer the questions in relation to how their back problem is affecting them ‘‘today’’, rather than the original instructions, which do not specify a time-frame. Selection of any particular version of the ODQ is at present based solely on preference for content and no studies have directly compared different versions. The aim of this study was to explore the construct validity of three versions of the ODQ using Rasch analysis. Rasch analysis is a useful tool for exploring the validity of questionnaires that have been developed using traditional methods. Developed by the Danish mathematician Georg Rasch (Rasch, 1960). Rasch analysis is a probabilistic model that tests the extent to which the observed pattern of responses ﬁts the pattern expected by the model. Rasch analysis calibrates person ability and item difﬁculty onto an interval scale in units called logits (log-odds units). Because logits are interval units, Andrich (2004) argues that Rasch analysis ‘‘y provides an operational criterion for fundamental measurement of the kind found in the physical sciences’’ (pI-12). The Rasch model provides evidence of scale validity by determining whether data derived from questionnaires can be validly summed, and if polytomous scoring categories work as intended. In addition Rasch analysis tests for invariance of items across external group characteristics by differential item functioning (DIF) analysis. Thus, a vigorous protocol is used to test what, in effect, is the internal construct validity of the scale, including unidimensionality, through a test of local independence. This analysis provides a strong diagnostic for the scale and provides a mechanism for comparison of different versions of a questionnaire.

223

2. Method 2.1. Sample and procedures Consecutive eligible patients at 7 public hospital outpatient or community-based physiotherapy departments and 9 physiotherapy private practices in Australia were invited to participate. Ambulatory patients receiving their ﬁrst or second consultation for an episode of low back pain were invited to participate. Eligible patients were those aged 18 years or older who were able to read and write English. People with back pain related to pregnancy, traumatic injury or rheumatic diseases were excluded. Participants completed a 12-item Oswestry (see Appendix A) as part of a larger battery of questionnaires and were mailed the questionnaires to complete a second time 4 weeks later. The ODQ version by Baker et al. (1989) was used for items 1–10 with metric distances replacing imperial units in the Walking item. The ﬁnal two items were the Changing Degree of Pain item from the Hudson-Cook version (1989) and the Work/Housework item, which is an adaptation of the Employment/Homemaking item from Fritz and Irrgang (2001). These changes were made to make the items more culturally relevant for Australia. Ethics approvals were gained from the Faculty of Health Sciences Human Ethics Committee of La Trobe University and from the ethics committees of the participating clinics where such committees existed. 2.2. Analysis For analysis, data for each of the three 10-item versions of the ODQ were extracted from the 12-items completed by respondents. Version 1 was the standard 10-items of the ODQ which includes the section Sex Life. Version 2 replaced Sex Life with a Work/Housework section and Version 3 replaced Sex Life with Changing Degree of Pain. Rasch analysis locates item difﬁculty and person ability on a logit scale. A logit (log-odds unit) is the natural logarithm of the odds of a person endorsing a particular rating scale step in an item. Due to the scoring direction of the ODQ, persons of higher ability and items of greater difﬁculty are located on the negative side of the logit scale, while persons of lower ability and items of less difﬁculty are located on the positive side. There are a series of components to Rasch anaysis. 2.2.1. Threshold order A threshold occurs where there is a transition between possible response options. The threshold is reached when the likelihood of endorsing one level of the scale is the same as the likelihood of endorsing the next level. Each item in the ODQ has 6 statements of increasing difﬁculty

ARTICLE IN PRESS 224

M. Davidson / Manual Therapy 13 (2008) 222–231

and therefore 5 thresholds. Each threshold has a location on the logit scale and each item has an average location. For each item one would expect that with decreasing ability the probability of selecting each statement in turn would increase in an ordered fashion from least to most difﬁcult. In the Sitting item, for example, one would expect the probability of selecting a particular response would increase from easiest (I can sit in any chair as long as I like) to hardest (pain prevents me from sitting at all) in a step-wise manner with decreasing person ability. Rasch analysis allows us to identify whether the steps in the response scale attract this expected pattern of responses or whether the thresholds are disordered, that is when the probability of selecting each level does not rise in the manner predicted. Disordered thresholds can be a source of item misﬁt. In the analysis, if items had disordered thresholds, response categories were collapsed until the ordered thresholds were achieved and the items showed adequate ﬁt. 2.2.2. Differential item functioning (DIF) A useful questionnaire can be used with a broad spectrum of patients so it is important that the items function similarly for persons at the same level of ability. Some items may attract systematically different responses on the basis of some characteristic other than item difﬁculty. DIF by gender and by age in the groupings 18–44, 45–64 and 65-plus years was conducted to explore whether these characteristics had a confounding effect on item responses. DIF by time (the ﬁrst and second administrations of the test) was tested so that the responses at the two administrations could be pooled, provided no DIF by time was evident.(Chang and Chan, 1995) For the DIF analysis the sample is divided into three equal-sized groups or ‘‘class intervals’’ classifying persons of low, medium and high ability. Uniform DIF is exhibited when there is a consistent deviation of observed from expected responses across all class intervals. Non-uniform DIF occurs when class intervals differ on their deviation from expected scores. 2.2.3. Overall fit and person separation The extent to which the overall questionnaire data for the class intervals ﬁt the Rasch model is tested with a w2 statistic. The w2 probability values greater than the chosen alpha value indicates no signiﬁcant deviation of the data from the model. The Person Separation Index provides an indication of how many groups or strata of ability the test can discriminate amongst (Wright and Masters, 1982). The higher the reliability of person separation, the more groups the test is able to detect. A reliability coefﬁcient of .8 indicates that two groups can be identiﬁed, and .9 four or more groups.2 2

Separation ¼ (reliability/(1reliability))0.5.

2.3. Unidimensionality Rasch analysis examines the unidimensionality of the scale, that is, the extent to which all the items in a scale are measuring the same underlying construct or latenttrait variable. Scale unidimensionality, or local independence, is a requirement of Rasch models. Item ﬁt statistics are an indicator of whether or not each item contributes to the measurement of a single underlying construct. Fit residuals are the distance of the observed item data from the expected data where a perfect ﬁt would result in a mean of zero and a standard deviation of one. Items with a residual greater than 72 are considered signiﬁcantly misﬁtting (Masters and Keeves, 1999). Items with large negative residual values indicate a high level of predictability in responses and therefore information redundancy. Items with large positive residual values indicate an unacceptable level of ‘‘noise’’ in the responses. Item ﬁt statistics are often the only indicator used to establish scale unidimensionality. A second strategy is to formally test the assumption of local independence by examining the principal components analysis (PCA) results of the Rasch analysis (Smith, 2002). If a set of items truly are unidimensional, then if a person takes any subset of items in the questionnaire their responses should provide the same person ability estimates as if they had taken the entire test. Two item subsets were determined by examining the ﬁrst component of the PCA (after extraction of the Rasch component). Items with negative and positive loadings comprised the two subsets of items, which were anchored to the original item locations. Anchoring equates the two tests by calibrating them on the same logit-scale ‘‘ruler’’. The person ability locations from each subset of items were then compared, using paired t-tests, with the locations derived from the full set of items. 2.4. Targetting Targetting refers to the extent to which the item threshold difﬁculties have adequately targeted the abilities of the persons in the sample. Poor targetting occurs when item thresholds are clustered at certain points along the logit scale leaving large gaps, and where many respondents have a higher or lower ability than the most or least difﬁculty item threshold. Targetting is judged by visual inspection of the distribution of persons and item thresholds on the logit scale. Descriptive statistics were calculated using SPSS for windows V11.0 and Rasch analysis using RUMM2020 software.3 RUMM2020 applies the unrestricted Partial Credit Model. Formulae for Rasch models can be found 3 RUMM2020, RUMM Laboratory Pty Ltd., 14 Dodonaea Court, Duncraig W.A. 6023, Australia.

ARTICLE IN PRESS M. Davidson / Manual Therapy 13 (2008) 222–231

in numerous publications (Andrich, 1978; Masters, 1982; Wright and Masters 1982; Masters and Keeves, 1999). Acceptable overall ﬁt of the data to the model was set as P4.01 (item-trait interaction, w2 probability). Items were considered misﬁtting if ﬁt residuals exceeded 72.0 or w2 probability was o.01. As three person factors (age, gender and time) were tested for twelve items, DIF was considered signiﬁcant if the w2 probability was o.001. Difference in person locations between item subsets and the full test was considered signiﬁcant if Po.01.

3. Results One hundred patients completed the questionnaires initially and 74 of these also returned the questionnaires a second time 4 weeks later. Sample characteristics (Table 1) reveal a predominantly female (64%) sample with ages ranging from 19 to 80 years. Few individuals were on sick leave (2%) or receiving compensation (7%). The majority were either employed (31%), retired (24%) or homemakers (19%). There was a high prevalence of recurrent back pain with 88% reporting they had experienced prior episodes of back pain. Table 2 shows the response frequencies for each item. There was no DIF by time so the two questionnaire data sets totalling 174 were combined for further analysis. Initial Rasch analysis of the whole data set of 174 cases identiﬁed 3 persons who either had not completed entire pages of the questionnaire or had scored zero on all items of the test. The analysable sample was therefore 171. Table 3 provides a comparison of the statistics for the three versions of the Oswestry.

225

Table 1 Sample characteristics Characteristic

Count

Age in years

Mean (sd) 52.69 (14.66) range 19–80

Gender Male Female

36 64

Work status Employed at usual job Light duty or restricted work Paid leave/sick leave Unpaid leave Unemployed due to health problems Unemployed due to other reason Student Keeping house/homemaker Retired On disability beneﬁt

31 2 5 0 4 3 2 19 24 10

On compensation Yes No

7 92

Duration of current episode Less than 6 weeks 6 weeks to 3 months 3–6 months More than 6 months

31 17 12 40

Pain location Back only Refers to buttock, groin or thigh Refers to leg below knee

24 40 35

Number of previous episodes None 1–5 More than 5

12 25 63

Note that because n ¼ 100 the percentage is equal to the count. Totals may not equal 100 due to missing data.

3.1. Threshold order Four items had disordered thresholds. The ﬁrst and second thresholds for item 9 Social Life and the second and third thresholds for item 6 Standing were reversed. For item 2 Personal Care the ﬁrst and second, and fourth and ﬁfth thresholds were reversed. The thresholds for item 8 Sex Life were ordered 1,4,3,2,5. Ordered thresholds were achieved for Standing and Sex Life, by combining response scores 2 and 3. For Personal Care scores 1 and 2, and 3 and 4 were combined. For Social Life scores 0 and 1, and 3 and 4 were combined. Fig. 1 compares the regular, ordered thresholds of the Pain item, with the disordered thresholds of the Personal Care item. In the Pain item the most likely response for a person of high ability located at 4 logits, is 0, the best possible score. As person ability decreases there is a step-wise change in the most probable response from 0 to 5. The most likely response for a person of low ability at 4 logits, is 5, the worst possible score. In the Personal

Table 2 Response frequency Item

1 2 3 4 5 6 7 8 9 10 11 12

Pain Personal care Lifting Walking Sitting Standing Sleeping Sex life Social life Traveling Work/housework Changing degree of pain

Response category 0

1

2

3

4

5

5 57 9 34 14 18 7 20 23 10 5 3

26 18 18 22 14 28 59 25 14 38 47 23

39 22 14 26 40 15 23 1 23 26 23 26

21 3 36 15 17 20 3 11 29 11 12 29

7 0 18 2 12 17 0 9 2 6 3 11

1 0 2 0 2 1 0 2 1 0 2 0

Note that because n ¼ 100 the percentage is equal to the count. Totals may not equal 100 due to missing data.

ARTICLE IN PRESS M. Davidson / Manual Therapy 13 (2008) 222–231

226

Table 3 Comparison of ﬁt statistics and DIF for 3 versions of the Oswestry Disability Questionnaire Version 1 Standard

Version 2 Work/housework

Version 3 Changing degree of pain

Total item w2 Person separation Item ﬁt residuals

36.39 (P ¼ .014) 0.87 None 472

22.19 (P ¼ .329) 0.88 None 472

Item ﬁt w2

All items P4.01

All items P4.01

DIF w2

Item 4 Walking: uniform DIF by age (P ¼ .00004)

Item 4 Walking: uniform DIF by age (P ¼ .00006)

39.43 (P ¼ .006) 0.87 Item 12 changing degree of pain ¼ 2.34 Item 12 changing degree of pain P ¼ .007 Item 4 Walking: uniform DIF by age (P ¼ .00008)

Total item w2 Po.01 indicates poor overall ﬁt. Item ﬁt w2 Po.01 indicates poor item ﬁt. DIF w2 Po.001 indicates deviation of observed from expected values.

Fig. 1. Threshold probability curves for item 1 pain and 2 personal care.

Care item, however, respondents are not using the available responses in a consistent manner: responses 1 and 4 are never the most probable choices no matter what the person’s ability level.

3.2. Differential item functioning (DIF) In all three versions of the Oswestry item 4 Walking consistently exhibited uniform DIF by age (Table 3).

ARTICLE IN PRESS M. Davidson / Manual Therapy 13 (2008) 222–231

The 65 and over age group had an observed score higher than expected. No other items had uniform or nonuniform DIF by age or gender (P4.001). 3.3. Fit Table 3 shows that while Version 1 (Standard) and Version 2 (Work/Housework) showed adequate overall ﬁt to the Rasch model (P4.01), Version 3 (Changing Degree of Pain) did not (w2 P ¼ .006). Person separation for all three versions indicates that at least 3 strata of person ability can be differentiated. 3.4. Unidimensionality The item Changing Degree of Pain in Version 3 was misﬁtting (residual 2.34, w2 9.82, df2, P ¼ .007). The item ﬁt statistics for the two other versions indicate that each set of items form a unidimensional scale. Paired t-tests of person locations from the full item set compared with the item subsets revealed no signiﬁcant difference for any of the Oswestry versions (Table 4).

227

3.5. Targetting Average item locations for the three versions are shown in Table 5. The threshold map for Version 1 (Standard Oswestry) (Fig. 2) shows the match of item threshold difﬁculty, on the lower part of the graph, to person ability on the upper part, on a common logitscale. Person ability and item difﬁculty move from highest (the negative side of the logit scale) to lowest (the positive side of the logit scale). The threshold located furthest to the left of the scale is the statement I can lift heavy weights without extra pain. The threshold located furthest to the right of the scale is the statement Pain prevents me from sleeping at all. Note that the threshold maps for the other two versions are not shown here as there is little difference in targeting between the versions.

4. Discussion This study directly compared three versions of the Oswestry Disability Questionnaire, which was achieved by administration of a test version containing the 10-item Oswestry plus two additional items contained

Table 4 Test of person locations for anchored item subsets compared to the full set Subset 1 cf total item set

Version 1 Standard Version 2 Work/housework Version 3 changing degree of pain

Subset 2 cf total item set

Mean difference

95% CI

P

Mean difference

95% CI

P

.013 .012 .065

.109–.083 .096–.071 .159–.03

.787 .770 .180

.029 .0002 .014

.112–.054 .094–.094 .059–.088

.489 .997 .705

Note: subset 1 comprised items 3,4,6,9 and subset 2 1,2,5,7,10 for all three version. Item 11 was part of subset 1 for version 2, and item 12 was part of subset 2 for version 3.

Table 5 Item logit locations (from easiest to hardest) for the 3 Oswestry versions Item

Version 1 Standard Location (SE)

Version 2 Work/housework Location (SE)

Version 3 Changing degree of pain Location (SE)

7 Sleeping 2 Personal care 4 Walking 9 Social life 10 Traveling 11 Work/housework 1 Pain 8 Sex life 6 Standing 12 Changing degree of pain 5 Sitting 3 Lifting

2.505 2.465 0.309 0.023 0.460

2.390 2.475 0.338 0.032 0.439 0.638 0.657

(0.135) (0.172) (0.094) (0.118) (0.103) (0.110) (0.102)

2.423 2.440 0.318 0.027 0.439

0.876 (0.103)

0.820 0.826 1.148 1.272

(0.136) (0.172) (0.095) (0.119) (0.103)

0.657 (0.103) 0.703 (0.125) 0.881 (0.103) 1.207 (0.087) 1.347 (0.087)

1.198 (0.088) 1.361 (0.087)

(0.133) (0.168) (0.091) (0.115) (0.100)

0.650 (0.100) (0.100) (0.094) (0.085) (0.084)

ARTICLE IN PRESS 228

M. Davidson / Manual Therapy 13 (2008) 222–231

Fig. 2. Threshold Map Oswestry Disability Questionnaire. Note: items of greater difﬁculty and persons of greater ability are located to the left side of the logit scale; items of lesser difﬁculty and persons of lesser ability are located to the right side of the logit scale.

in two modiﬁed Oswestry versions. Including all 12 items in the test version allowed all three versions to be extracted. The version that replaces Sex Life with Work/ Housework had the best overall ﬁt to the model (P ¼ .329) and the standard Oswestry showed adequate ﬁt to the Rasch model (P ¼ .014), while the version that replaces Sex Life with Changing Degree of Pain did not (P ¼ .006). The results conﬁrm that the item Changing Degree of Pain does not belong with the other items and is measuring a different underlying construct to the other items. Testing that sub-sets of items provide an equivalent estimate of person ability to the entire set of items provides a robust indication of whether departures from unidimensionality signiﬁcantly distort estimates of person location. Despite the presence of one misﬁtting item and overall poor ﬁt to the model, Version 3 showed that the deviation from unidimensionality did not result in signiﬁcantly different person location estimates calculated for two anchored subsets of items. It remains to be demonstrated how robust the estimates of person location are and the extent of departure from unidimensionality that can be tolerated before signiﬁcant deviations occur. Two previous studies that have used Rasch analysis to examine the ODQ reported that the Pain item did not ﬁt the model (Page et al., 2002; White and Velozo, 2002). That the current study did not ﬁnd this item misﬁtting may reﬂect the different wording of the item in the versions administered. The version administered in the current study asks only about pain intensity, while the two previous studies have administered a version that relates pain to analgaesic medication. If the content

of the ODQ is mapped to the WHO International Classiﬁcation of Functioning the pain item is a measure of impairment while the other items reﬂect activity limitations (WHO, 2001). However, these items all relate activity limitation to pain, and the pain item has been shown to have a linear relationship with the other items (Fairbank et al., 1980). The existence of disordered thresholds for Personal Care, Standing, Sex Life and Social Life is evidence that, at least for these items, the response options do not perform as intended. White and Velozo (2002) and Page et al. (2002) both proposed modiﬁed versions of the Oswestry in which the Pain item is deleted and response levels 2 and 3, and 4 and 5 for all items are combined, reducing the number of response options from 6 to 4. Neither study reports which individual items had disordered thresholds. Citation tracking has failed to ﬁnd any subsequent studies that have administered or further tested either of these versions. Due to the low frequency of responses to response options 4 and 5 in an ambulatory population (Table 2) there is some merit in suggesting a reduction in response options at the upper end of the scale. Item 4 Walking displayed DIF by age in all three versions. On this item, persons in the 65-plus age group, at the same level of ability as the younger groups, had higher (worse) scores than expected. This indicates that something other than the difﬁculty of walking as an activity is inﬂuencing older persons’ responses to this item. Fear of falling and various sociodemographic variables have been reported to be associated with reduced mobility in elderly persons (Arfken et al., 1994; Tinetti et al., 1994; Simonsick et al., 1999; Murphy et al., 2002). Neither of the previous Rasch studies reported if

ARTICLE IN PRESS M. Davidson / Manual Therapy 13 (2008) 222–231

they examined DIF (Page et al., 2002; White and Velozo, 2002). The Oswestry item thresholds for the persons in this sample are a reasonable match in that there are thresholds for all persons except for a small number of persons of very high ability (Fig. 2). This reﬂects the fact that total Oswestry scores in an ambulatory population are often skewed toward the lower (better functioning) end of the scale, with few persons scoring in the top 1/5th of the available total score range. This is because the highest responses options of some items are rarely or never selected (Table 2). Some gaps are evident in the item difﬁculty threshold placement on the logit scale on the far right (lower functioning) end of the scale. Ideally, item thresholds should be evenly spread along the logit scale. Although two versions showed adequate overall ﬁt to the Rasch model, the problems of disordered thresholds for some items, DIF for the Walking item, and gaps in targeting are typical of the limitations of ordinal scales designed using classical test theory and which are only revealed using Rasch anlysis. A limitation of the study is that the number of eligible patients who were not invited or who refused to participate in the study is unknown. It is also not known the extent to which the sample is representative of ambulatory patients seeking physiotherapy treatment for low back pain, as there is no data available to describe this population. However, participants were recruited from a number of private and public agencies in both metropolitan and rural settings and this would maximise the likelihood that the sample is representative. As the data were collected from ambulatory patients with low back pain no generalisations can be made to non-ambulatory or admitted patients.

229

5. Conclusion The standard version of the Oswestry and the version that replaces Sex Life with Work/Housework both form unidimensional scales in which all items are measuring a single underlying variable. The item Changing Degree of Pain that replaces Sex Life in the Hudson–Cook version does not measure the same underlying construct as the other items. These ﬁndings suggest that either of the ﬁrst two of the three versions of this widely used low back pain outcome measure should be selected over the third. Users should also be aware that for some items the rating scale steps do not perform as intended.

Acknowledgements Professor Alan Tennant, Academic Unit of Musculoskeletal & Rehabilitation Medicine, University of Leeds provided advice on Rasch analysis.

Appendix A This questionnaire has been designed to give us information as to how your back or leg pain has affected your ability to manage in everyday life. Please answer by checking one box in each item for the statement which best applies to you. We realise you may consider that two of the statements in any one item relate to you, but please just mark the box for the statement that most clearly describes your problem (Table A1 here).

Table A1 A 12-item test version of the Oswestry Disability Questionnaire Item 1: pain intensity & I have no pain at the moment & The pain is very mild at the moment & The pain is moderate at the moment & The pain is fairly severe at the moment & The pain is very severe at the moment & The pain is the worst imaginable at the moment Item 2: personal care (washing, dressing, etc.) & I can look after myself normally without causing extra pain & I can look after myself normally but it is very painful & It is painful to look after myself and I am slow and careful & I need some help but manage most of my personal care & I need help every day in most aspects of self-care & I do not get dressed, wash with difﬁculty and stay in bed Item 3: lifting & I can lift heavy weights without extra pain & I can lift heavy weights but it gives extra pain & Pain prevents me lifting heavy weights off the ﬂoor but I can manage if they are conveniently positioned, e.g. on a table & Pain prevents me lifting heavy weights but I can manage light to medium weights if they are conveniently positioned & I can only lift very light weights & I cannot lift or carry anything

ARTICLE IN PRESS 230

M. Davidson / Manual Therapy 13 (2008) 222–231

Table A1 (continued ) Item 4: walking & Pain does not prevent me walking any distance & Pain prevents me from walking more than 2 km & Pain prevents me from walking more than 1 km & Pain prevents me from walking more than 500 m & I can only walk using a stick, crutches or other support & I am unable to walk at all Item 5: sitting & I can sit in any chair as long as I like & I can only sit in my favourite chair as long as I like & Pain prevents me sitting for more than 1 h & Pain prevents me from sitting for more than 30 min & Pain prevents me from sitting more than 10 min & Pain prevents me from sitting at all Item 6: standing & I can stand as long as I want without extra pain & I can stand as long as I want but it gives me extra pain & Pain prevents me from standing for more than 1 h & Pain prevents me from standing for more than 30 min & Pain prevents me from standing for more than 10 min & Pain prevents me from standing at all Item 7: sleeping & My sleep is never disturbed by pain & My sleep is occasionally disturbed by pain & Because of pain I have less than 6 h sleep & Because of pain I have less than 4 h sleep & Because of pain I have less than 2 h sleep & Pain prevents me from sleeping at all Item 8: sex life (if applicable) & My sex life is normal and causes no extra pain & My sex life is normal but causes some extra pain & My sex life is nearly normal but is very painful & My sex life is severely restricted by pain & My sex life is nearly absent because of pain & Pain prevents any sex life at all Item 9: social life & My social life is normal and gives me no extra pain & My social life is normal but increases the degree of pain & Pain has no signiﬁcant effect on my social life apart from limiting my more energetic interests e.g. sport, etc. & Pain has restricted my social life and I do not go out as often & Pain has restricted my social life to my home & I have no social life because of pain Item 10: travelling & I can travel anywhere without pain & I can travel anywhere but it gives extra pain & Pain is bad but I manage journeys over 2 h & Pain restricts me to journeys of less than 1 h & Pain restricts me to short necessary journeys under 30 min & Pain prevents me from travelling except to receive treatment Item 11: work/housework & My normal work/housework does not cause pain & My normal work/housework increase my pain, but I can still perform all that is required of me & I can perform most of my work/housework, but pain prevents me from performing more physically demanding activities (e.g., lifting, vacuuming) & Pain prevents me from doing anything but light work/housework & Pain prevents me from doing even light work/housework & Pain prevents me from performing any work/housework Item 12: changing degree of pain & My pain is rapidly getting better & My pain ﬂuctuates but overall is deﬁnitely getting better & My pain seems to be getting better but improvement is slow at present & My pain is neither getting better or worse & My pain is gradually worsening & My pain is rapidly worsening

ARTICLE IN PRESS M. Davidson / Manual Therapy 13 (2008) 222–231

References Andrich D. A rating formulation for ordered response categories. Psychometrika 1978;43(4):561–73. Andrich D. Controversy and the Rasch model: a characteristic of incompatible paradigms? Medical Care 2004;42(Suppl 1):I7–I16. Arfken CL, Lach HW, Birge SJ, Miller JP. The prevalence and correlates of fear of falling in elderly persons living in the community. American Journal of Public Health 1994;84(4):565–70. Baker DJ, Pynsent PB, Fairbank JCT. The Oswestry Disability Index revisited: its reliability, repeatability and validity, and a comparison with the St Thomas Disability Index. In: Roland M, Jenner JR, editors. Back pain: new approaches to rehabilitation and education. Manchester: Manchester University Press; 1989. p. 174–86. Chang WC, Chan C. Rasch analysis for outcomes measures: some methodological considerations. Archives of Physical Medicine and Rehabilitation 1995;76(10):934–9. Davidson M, Keating JL. A comparison of ﬁve low back disability questionnaires: reliability and responsiveness. Physical Therapy 2002;82(1):8–24. Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine 2000;25(22):2940–52. Fairbank JC, Couper J, Davies JB, O’Brien JP. The Oswestry low back pain disability questionnaire. Physiotherapy 1980;66(8):271–3. Fritz JM, Irrgang JJ. A comparison of a modiﬁed Oswestry Low Back Pain Disability Questionnaire and the Quebec Back Pain Disability Scale. Physical Therapy 2001;81(2):776–88. Grotle M, Brox JI, Vollestad NK. Functional status and disability questionnaires: what do they assess? Spine 2004;30(1):130–40. Hudson-Cook N, Tomes-Nicholson K, Breen A. A revised Oswestry Disability Questionnaire. In: Roland M, Jenner JR, editors. Back pain: new approaches to rehabilitation and education. Manchester: Manchester University Press; 1989. p. 187–204.

231

Masters GN. A Rasch model for partial credit scoring. Psychometrika 1982;47(2):149–74. Masters GN, Keeves JP. Advances in measurement in educational research and assessment. Amsterdam: Pergamon; 1999. Murphy SL, Williams CS, Gill TM. Characteristics associated with fear of falling and activity restriction in community-living older persons. Journal of the American Geriatric Society 2002; 50(3):516–20. Page SJ, Shawaryn MA, Cernich AN, Linacre JM. Scaling of the Revised Oswestry Low Back Pain Questionnaire. Archives of Physical Medicine and Rehabilitation 2002;83(11): 1579–84. Rasch G. Probabilistic models for some intelligence and attainment tests. Copenhagen: Danmarks Paedogogiske Institute; 1960. Simonsick EM, Guralnik JM, Fried LP. Who walks? Factors associated with walking behavior in disabled older women with and without self-reported walking difﬁculty. Journal of the American Geriatrics Society 1999;47(6):672–80. Smith EV. Detecting and evaluating the impact of multidimensionality using item ﬁt statistics and principal component analysis of residuals. Journal of Applied Measurement 2002;3(2): 205–31. Tinetti ME, Mendes de Leon CF, Doucette JT, Maker DI. Fear of falling and fall-related efﬁcacy in relationship to functioning among community-living elders. Journal of Gerontology 1994; 49(3):M140–7. White LJ, Velozo CA. The use of Rasch measurement to improve the Oswestry classiﬁcation scheme. Archives of Physical Medicine and Rehabilitation 2002;83(6):822–31. WHO. International classiﬁcation of functioning, disability and health. Geneva: World Health Organization; 2001. Wright BD, Masters GN. Rating scale analysis. Chicago: Mesa Press; 1982.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 232–238 www.elsevier.com/locate/math

Original article

The inﬂuence of breathing type, expiration and cervical posture on the performance of the cranio-cervical ﬂexion test in healthy subjects Barbara Cagnie, Lieven Danneels, Ann Cools, Nele Dickx, Dirk Cambier Department of Rehabilitation Sciences and Physiotherapy, Ghent University, De Pintelaan 185, 6K3, B-9000 Ghent Received 16 August 2006; received in revised form 12 January 2007; accepted 17 January 2007

Abstract The cranio-cervical ﬂexion test (CCF-T) is used as a clinical evaluation tool for the deep cervical ﬂexors (DCF). The inﬂuence of breathing type, expiration and cervical posture on the performance of the test is evaluated in asymptomatic subjects. Thirty volunteers participated in the study and were classiﬁed according to their breathing type: costo-diaphragmatic breathing type and upper costal breathing type. Sternocleidomastoid (SCM) electromyographic (EMG) activity was recorded during ﬁve incremental levels of CCF during normal breathing as well as during expiration. The cranio-vertebral angle of each subject was measured to quantify cervical posture. During normal inspiration, higher EMG activity of the SCM muscles was observed in subjects with an upper costal breathing pattern compared to costo-diaphragmatic breathing subjects. This difference was statistically signiﬁcant (Po 0.05) at the three lowest stages of the test. In the upper costal breathing group a signiﬁcantly lower EMG activity of the SCM muscles was observed while performing the CCF-T during slow expiration compared to normal breathing. No signiﬁcant correlation was found between the cranio-vertebral angle and the EMG activity of the SCM muscles. Performing the CCF-T during slow expiration diminishes the activity of the SCM muscles in subjects with a predominantly upper costal breathing pattern. Using a costo-diaphragmatic breathing pattern while performing the test will optimize the performance. Studies on neck pain patients are required to further clarify this issue. r 2007 Elsevier Ltd. All rights reserved. Keywords: Cranio-cervical ﬂexion test; Sternocleidomastoid; Breathing type; Cervical posture

1. Introduction In recent years research has focused on identifying and quantifying deﬁcits in the deep cervical ﬂexor (DCF) muscles in patients with neck pain disorders (Jull, 2000; Falla, 2004; Jull et al., 2004; O’Leary et al., 2006). The cranio-cervical ﬂexion test (CCF-T) has been advocated as the method of choice to assess and retrain the contractile performance of these muscles. The cranio-cervical movement aims to ﬂex the upper cervical spine in association with a mild ﬂattening effect of the cervical lordosis, which is an anatomical action of the deep longus capitis and longus colli muscles (MayouxCorresponding author. Tel.: +32 9 240 52 65; fax: +32 9 240 38 11.

E-mail address: [email protected] (B. Cagnie). 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.01.009

Benhamou et al., 1994; Vasavada et al., 1998; O’Leary et al., 2006). In contrast, superﬁcial cervical ﬂexor muscles such as the sternocleidomastoid (SCM) muscles are not prime movers of CCF-T and are structurally more suited to assist in ﬂexing the lower cervical spine on the thorax (O’Leary et al., 2006). In the CCF-T, the subject performs ﬁve increments of increasingly inner range cranio-cervical ﬂexion in a supine lying position (Jull, 2000; Falla et al., 2003a). Patients are guided to the test level by feedback from a pressure unit which is placed behind the neck to monitor the progressive ﬂattening of the cervical lordosis which results from the contraction of longus colli (Jull et al., 2004). Amplitude of muscle signals can be measured in the SCM during the test, as an increased activity of the superﬁcial muscles could be a measurable compensation

ARTICLE IN PRESS B. Cagnie et al. / Manual Therapy 13 (2008) 232–238

for poor segmental stability or poorer activation of the longus colli (Cholewicki et al., 1997). Besides its postural function, the SCM is also considered to be an accessory inspiratory muscle (De Mayo et al., 2005). One could assume that performance of the CCF-T during slow expiration will inhibit the action of the SCM and thus facilitate the DCF more. Several breathing types have been deﬁned depending on the expansion of the abdomino-thoracic region during inspiration at rest (Costa et al., 1994; De Mayo et al., 2005): (1) Costo-diaphragmatic breathing type is observed when the abdominal and lateral costal expansion is predominant over the superior thoracic expansion, during inspiration at rest. This is considered the optimum breathing type because it allows maximal lung expansion, and therefore, maximum lung capacity and gas exchange. (2) Upper costal breathing type takes place when superior thoracic expansion exceeds the abdominal and lateral costal expansion, during inspiration at rest. This breathing type produces a smaller expansion of the rib cage and therefore, smaller lung capacity and gas exchange. Hence, the use of accessory muscles may be required in order to breathe properly. (3) Mixed breathing type is observed when there is no clear predominance of superior thoracic expansion or abdominal and lateral costal expansion. In order to increase the diameter of the thorax, one could assume that the SCM is more activated in subjects with an upper costal breathing pattern. To the best of our knowledge, the effect of breathing type on the performance of the CCF-T has not been studied until now. One of the main functions of the SCM is to pull the head forward and down with associated extension of the upper cervical spine, which well reﬂect the deﬁnition of a forward head posture (FHP) (Chiu et al., 2002; Fernandez-de-las-penas et al., 2006b). Habitual FHP has been considered important in the aetiology of postural neck pain, as an increased forward neck ﬂexion may result in increased tension in the regions postural stabilizing muscles as well as increased compressive forces in the articulations of the cervical spine (Szeto et al., 2005). Recent research has demonstrated that headache and neck pain patients may adopt habitual postures with greater FHP and upper cervical extension, although the differences are small compared to asymptomatic individuals (Watson and Trott, 1993; Szeto et al., 2005; Fernandez-de-las-penas et al., 2006a, b; Edmondston et al., 2006). As different studies have found higher measures of electromyographic (EMG) signal amplitude in the SCM in neck pain patients compared to healthy controls (Falla et al., 2004), a positive correlation between SCM activity during the performance of the CCF-T and FHP could be assumed. The purpose of this study was to evaluate the abovementioned hypotheses: the inﬂuence of (1) breath-

233

ing type, (2) expiration and (3) cervical posture on the performance of the CCF-T.

2. Materials and methods 2.1. Subjects 30 volunteers (17 females, 13 males) with no history of neck pain and a mean age of 32.178.3 years (range 22– 52 years) participated in the study. After receiving verbal and written information each subject signed a consent form containing information about the nature of the study. This study was approved by the local Ethics Committee of the Ghent University Hospital. 2.2. Protocol 2.2.1. Determination of the cervical posture A picture of the lateral view of each subject was taken to objectively assess FHP. The base of the camera was set at the height of the subject’s shoulder. The subjects were instructed to stand comfortably, in their ‘normal, loose, or habitual’ posture, with their weight evenly on both feet and looking straight ahead. They were asked not to stand erect, or in a ‘best posture’, because the purpose of the photograph was to capture their habitual or usual standing posture. A plumb line was marked on the wall. The index ﬁnger of the examiner was placed to the skin indicating the spinous process of the seventh cervical vertebra (C7). The examiner located the C7 spinous process by the following procedure. First, he palpated the most prominent spinous process (C6 or C7) at the base of the cervical spine. After it was identiﬁed, he passively ﬂexed and extended the lower cervical spine to verify which one moved ﬁrst: C6 vertebra should be more mobile, whereas C7 should demonstrate less motion. Once the picture was obtained, it was used to measure the cranio-vertebral angle: the angle between the horizontal line passing through C7 and a line extending from the midpoint of the tragus of the ear to C7. A smaller cranio-vertebral angle indicated a greater FHP. The reliability of this procedure is reported as high (ICC: 0.880.98; r2 ¼ 0.973) (Raine and Twomey, 1997; Fernandez-de-las-penas et al., 2006b). 2.2.2. Determination of the breathing type Subjects were classiﬁed according to their breathing type in two groups: costo-diaphragmatic breathing type and upper costal breathing type. They were asked to remain standing, look straight ahead, with their feet 10 cm apart, and to breathe normally for the duration of 2 min as a baseline. Two examiners determined the breathing type as follows: ﬁrst, they placed their left hand on the upper chest and their right hand on the

ARTICLE IN PRESS 234

B. Cagnie et al. / Manual Therapy 13 (2008) 232–238

upper back; next, they placed their left hand on the lower right costal region and their right hand on the upper abdomen. After checking 10 inspirations on each step of the clinical examination, the subject was classiﬁed to be of the upper costal breathing type if, during inspiration at rest, the superior thoracic expansion was predominant, and the costo-diaphragmatic breathing type when the abdominal and lateral costal expansion was predominant (De Mayo et al., 2005). To classify subjects into a certain breathing type, an agreement among both examiners was required. Five subjects did not show a clear predominance of superior thoracic expansion or abdominal and lateral costal expansion and were excluded from the study. The upper costal breathing type group included 12 subjects, 8 females and 4 males, ranging in age from 22 to 46 years with a mean age of 31.4 years. The costodiaphragmatic breathing type group included 13 subjects, 5 females and 8 males, ranging in age from 22 to 44 years, with a mean age of 30.7 years. 2.2.3. Cranio-cervical flexion test Subjects were positioned in supine lying position with the head and neck in a mid position such that the face line was horizontal and an imaginary horizontal line bisected the neck longitudinally. If necessary, layers of towel were placed under the head to gain the position. Myoelectric signals were detected from the left and right SCM by a surface EMG system with 8 channels (MyoSystem 1400, Noraxon Inc, Scotsdale, AZ, USA) using disposable Ag/AgCl surface electrodes (Blue Sensor, Medicotest A/S, Ølstykke, Denmark). Following careful skin preparation, a marker point was placed at the distal one third of the line between the mastoid process and the sternal notch. Surface electrodes were positioned just above and below this point in the direction of the line joining the mastoid process and sternal notch. This electrode positioning has proved to be appropriate in the detection of EMG signals for the SCM muscle (Falla et al., 2002). The ground electrode was placed on the sternum. The raw surface EMG signals were preampliﬁed (overall gain ¼ 1000, common rate rejection ratio 115 dB, ﬁltered to produce a bandwidth of 10–1000 Hz) and analogue/digital converted (12-bit resolution) at 1000 Hz. To allow subsequent normalization of the EMG data, subjects ﬁrst performed a reference maximal voluntary isometric contraction, which consisted of a combined cranio-cervical ﬂexion and cervical ﬂexion. The subject was asked to ﬂex his head on his neck followed by a lift of his head while resistance was given on the forehead by the investigator. This position was sustained for 5 s. The pressure cuff of the Stabilizer (Chattanooga Group Inc.) was placed suboccipitally behind the subject’s cervical spine and inﬂated until a stable pressure of 20 mmHg was achieved. Subjects were

instructed in the action of cranio-cervicial ﬂexion and practiced progressive targeting of ﬁve incremental levels (increment, 2 mmHg) between 22 and 30 mmHg. Subjects then performed the ﬁve incremental stages (22–30 mmHg) of the cranio-cervical ﬂexion maintaining the pressure steady on each target for 10 s in two conditions: during normal breathing and during expiration. In the latter condition, the subjects were asked to maximally inspire immediately followed by the performance of the CCF-T at the different levels during slow expiration. Both conditions were randomly executed. In each trial, data collection commenced at the point at which the subject reached the pressure target. A 30 s rest was allowed between trials. 2.2.4. Data management and statistical analysis To obtain a measure of EMG signal amplitude, maximum root mean square (RMS) was calculated for 8 s. For normalization, EMG amplitude at baseline and for each stage of the C-CFT was expressed as a percentage of the 1-s RMS values obtained during the reference voluntary contraction. The normalized baseline measure was subtracted from the normalization score of each stage tested, in order to represent real differences. All statistical analyses were performed using SPSS 12.0 for windows. Preliminary analysis identiﬁed no difference between sides for the SCM muscle, both during normal breathing and expiration, allowing pooling of the data. The normality of variables was evaluated by the Kolmogorov–Smirnov test, which demonstrated a normal distribution (P40.05). A general linear model was used to investigate within and between group differences in the normalized RMS values for the SCM muscles for the factors breathing type and stages of the CCF-T during normal breathing and expiration. Signiﬁcant main or interaction effects were further evaluated by examining means and 95% conﬁdence intervals for each level in each condition between groups. Independent sample t tests were conducted to compare for group differences (breathing type) and paired sample t tests were conducted to determine if EMG values were signiﬁcantly different between normal breathing and expiration during the performance of the CCF-T. Results are reported as mean and 95% conﬁdence interval. A value of Po0.05 was used as an indicator of statistical signiﬁcance. A Pearson correlation coefﬁcient was used to determine whether a correlation existed between the craniovertebral angle and the EMG activity of the SCM during the different stages of the CCF-T. 3. Results The multivariate analysis of variance yielded a signiﬁcant interaction effect between the factors respiration and breathing type [F ¼ 6.90, P ¼ 0.014] and a

ARTICLE IN PRESS B. Cagnie et al. / Manual Therapy 13 (2008) 232–238

signiﬁcant interaction effect between the factors respiration and stage [F ¼ 6.98, Po0.001]. Therefore, normalized RMS values, after subtracting the normalized baseline values for the SCM are presented for each stage of the C-CFT during normal breathing and expiration, for both the upper costal and costodiaphragmatic breathing group (Table 1). 3.1. Breathing type Subjects with an upper costal breathing pattern (5.41%72.76) were shown to have a greater EMG activity at rest when compared to subjects with a costodiaphragmatic breathing pattern (2.69%71.43) (P ¼ 0.026). Considering the baseline values, higher EMG activity of the SCM was observed in subjects with an upper costal breathing pattern than in the costodiaphragmatic breathing subjects during normal breathing (Fig. 1), which was statistically signiﬁcant (Po 0.05) at the three lowest stages of the test. The two highest stages of the test showed a tendency towards signiﬁcance (P ¼ 0.074 and 0.050, respectively). No signiﬁcant difference was found between both groups during expiration.

235

lowest stages of the test (22, 24 and 26 mm Hg) (Fig. 2). No differences between normal respiration and expiration were found in the costo-diaphragmatic breathing group. 3.3. Cervical posture The mean cranio-vertebral angle was 49.41 (77.01). The angle was similar in both breathing types (49.2176.3 in the upper costal breathing type group versus 49.7177.8 in the costo-diaphragmatic breathing type group). There was no signiﬁcant correlation between the cranio-vertebral angle and the EMG activity of the SCM at the ﬁve different stages of the CCF-T during normal breathing as well as during expiration.

4. Discussion This study investigated the inﬂuence of breathing type, expiration and cervical posture on the performance of the CCF-T in asymptomatic subjects. 4.1. Breathing type

3.2. Expiration In the upper costal breathing group, lower EMG activity of the SCM was observed while performing the CCF-T during slow expiration compared to normal breathing, which was statistically signiﬁcant at the three

In this study, we found an increased baseline activity in subjects with an upper costal breathing pattern compared to costo-diaphragmatic breathers, which remained during the performance of the CCF-T, after subtracting the baseline values. However, Costa et al.

Table 1 Normalized RMS values, after subtracting the normalized baseline values for the SCM for each stage of the C-CFT during normal breathing and expiration, for the upper costal and costo-diaphragmatic breathing group (mean and SD) Normal breathing

Expiration

p-value

22 mmHg

Upper costal Costo-diaphragmatic P-value

2.05 (0.99) 0.99 (1.27) 0.003

0.78 (2.61) 0.69 (1.53) 0.893

0.048 0.463

24 mmHg

Upper costal Costo-diaphragmatic P-value

3.95 (2.05) 2.37 (1.89) 0.034

0.66 (2.56) 1.58 (1.88) 0.169

0.000 0.485

26 mmHg

Upper costal Costo-diaphragmatic P-value

5.72 (3.27) 3.33 (2.97) 0.013

2.18 (2.90) 2.64 (3.51) 0.641

0.001 0.277

28 mmHg

Upper costal Costo-diaphragmatic P-value

5.37 (3.48) 3.50 (3.58) 0.074

4.99 (3.51) 3.93 (2.86) 0.255

0.573 0.516

30 mmHg

Upper costal Costo-diaphragmatic P-value

7.34 (5.01) 4.46 (3.29) 0.050

5.85 (5.47) 5.81 (4.97) 0.977

0.503 0.133

Bold font reﬂect those p-values which are signiﬁcant (po0.05). Signiﬁcance of difference between upper costal breathing type and costo-diaphragmatic breathing type. Signiﬁcance of difference between normal breathing and expiration.

ARTICLE IN PRESS B. Cagnie et al. / Manual Therapy 13 (2008) 232–238

236

Upper costal breathing type costo-diaphragmatic breathing type 10 normalised RMS value (%)

9 8

*

7 6 *

5 4 3

*

2 1 0 22

24

26

28

30

stage of C-CFT (mmHg)

Fig. 1. Group data for normalized RMS values, after subtracting the normalized baseline values (mean and 95% conﬁdence intervals) for the SCM for each stage of the C-CFT during normal breathing. * indicates signiﬁcant difference between upper costal breathing type and costo-diaphragmatic breathing type (Po 0.05).

normal inspiration expiration

muscles may be required in order to breathe properly as upper costal breathing is characterized by an outward, upward movement of the chest wall that requires more work to be done in lifting the rib cage. An important function of the SCM, as an accessory muscle for respiration, is to increase the anterior–posterior diameter of thorax by elevation of the sternum (Raper et al., 1966). In order to increase the diameter of the thorax, one could assume that the SCM is more activated in subjects with an upper costal breathing pattern. At the two highest stages of the CCF-T, the difference between costo-diaphragmatic and upper costal breathing diminished. This could be attributed to the increasing demands of the SCM. A positive linear relationship exists between SCM normalized RMS values and stage of the CCF-T (Falla et al., 2003b). The presence of progressively increasing SCM normalized RMS values in each test stage in both subject groups suggests that these muscles are recruited to further stabilize the neck as the contractile demand of the DCF increases in the inner range of cranio-cervical ﬂexion. These increasing demands of the SCM may minimize its respiratory function and support its important role as a head postural muscle.

normalised RMS value (%)

10 *

4.2. Expiration

8 * 6

*

4 2 0 22

24

26

28

30

stage of C-CFT (mmHg)

Fig. 2. Normalized RMS values, after subtracting the normalized baseline values (mean and 95% conﬁdence interval) for the SCM for each stage of the C-CFT in the upper costal breathing type group. * indicates signiﬁcant difference in the performance of the CCF-T between normal breathing and expiration (Po 0.05).

(1994) as well as De mayo et al. (2005) found no signiﬁcant difference in resting SCM activity between costo-diaphragmatic and upper costal breathing types. According to Costa et al. (1994) a higher SCM activity was only observed in the upper costal types when they breathed rapidly, roughly and during breathing effort. It is not clear why differences in resting values were found between this and previous studies, although differences in data management and analysis could reﬂect these discrepancies. A possible explanation for the difference between both breathing types may be that the use of accessory

Performing the CCF-T during slow expiration tends to lower the EMG activity of the SCM in asymptomatic subjects. This difference is statistically signiﬁcant only in the upper costal breathing group at the three lowest stages of the test (22, 24 and 26 mm Hg). This can be explained by the fact that upper costal breathing subjects demonstrate a higher activity of the SCM during the performance of the CCF-T. Slow expiration may inhibit the action of the SCM resulting in a lower recruitment of this muscle. Cholewecki et al. hypothesized that an increased activity of the superﬁcial muscles could be a measurable compensation for poorer activation of the longus colli (Cholewicki et al., 1997). Conversely, one could state that a decreased activity of the superﬁcial muscles is a measurable indication for stronger activation of the DCF. This may have some important clinical implications. Testing and retraining the cervical ﬂexor synergy as a component of a speciﬁc active stabilization program for the cervico-brachial region are now used widely in clinical practice in the treatment of patients with various neck pain syndromes. Speciﬁc emphasis should ﬁrst be placed on reeducating the deep and postural muscles whereas general strengthening exercises are only introduced once the imbalance between the deep and superﬁcial neck synergists has been addressed (Falla, 2004).

ARTICLE IN PRESS B. Cagnie et al. / Manual Therapy 13 (2008) 232–238

Subjects with a predominantly upper costal breathing pattern should be taught to perform the CCF-T during slow expiration after a deep inspiration. It is hypothesized that this will inhibit the action of the SCM and facilitate the DCF more. However, one should bear in mind that a forced expiration requires a good coordination, which makes the test complex and maybe less feasible for some subjects. According to this research, this is only relevant during the ﬁrst three stages of the CCF-T. Additionally, subjects should be trained a costo-diaphragmatic breathing pattern to decrease activity in the SCM. Further studies on neck pain patients are required to clarify this issue. 4.3. Cervical posture The mean cranio-vertebral angle (49.6177.01) was similar to that reported for healthy subjects by Watson and Trott (1993) (49.172.91) and Treleaven et al. (1994) (50.7177.91). As the SCM muscles pull the head forward and down, one could assume that a greater activity of these muscles will be present in subjects with an FHP. However, the current investigation does not conﬁrm this as there was no correlation found between the cranio-vertebral angle and the EMG activity of the SCM at the different stages during the performance of the CCF-T. This could be attributed to the fact that only healthy volunteers participated in this study. FHP has been related to neck pain, temporomandibular disorders, cervico-genic headache and postconcussional headache (Watson and Trott, 1993; Szeto et al., 2005; Fernandezde-las-penas et al., 2006a, b; Edmondston et al., 2006). In addition, different studies have found higher measures of EMG signal amplitude in the SCM in neck pain patients compared to healthy controls which were hypothesized to be related to impaired performance of the DCF muscles (Falla et al., 2004). It remains to be tested whether there is any association between performance in the cranio-cervical test and FHP in neck pain subjects.

5. Conclusion The inﬂuence of breathing type, expiration and cervical posture on the performance of the CCF-T was evaluated in asymptomatic subjects. This appears to have potential clinical applications; (1) subjects with a predominantly upper costal breathing pattern should be taught to perform the CCF-T during slow expiration after a deep inspiration as this lowers the EMG activity of the SCM. (2) Teaching a subject to use a costodiaphragmatic breathing pattern while performing the test will optimize the performance. However, studies on

237

neck pain patients are required to further clarify this issue.

References Chiu T, Ku W, Lee M, et al. A study on the prevalence and risk factors for neck pain among university academic staff in Hong Kong. Journal Of Occupational Rehabilitation 2002;12:77–91. Cholewicki J, Panjabi M, Khachatryan A. Stabilizing function of the trunk ﬂexor–extensor muscles around a neutral spine. Spine 1997; 22:2207–12. Costa D, Vitti M, de Olivera Tossello D, Costa R. Participation of the sternocleidomastoid muscle on deep inspiration in man. An electromyographic study. Electromyography And Clinical Neurophysiology 1994;34:315–20. De Mayo T, Miralles R, Barrero D, et al. Breathing type and body position effects on sternocleidomastoid and suprahyoid EMG activity. Journal Of Oral Rehabilitation 2005;32:487–94. Edmondston S, Chan H, Ngai G, et al. Postural neck pain: an investigation of habitual sitting posture, perception of ‘good’ posture and cervicothoracic kinaesthesia. Manual Therapy 2006. Falla D, Dall’Alba P, Rainoldi A, et al. Location of innervation zones of sternocleidomastoid and scalene muscles—a basis for clinical and research electromyography applications. Clinical Neurophysiology 2002;113:57–63. Falla D, Campbell C, Fagan A, et al. An investigation of the relationship between upper cervical ﬂexion range of motion and pressure change during the cranio-cervical ﬂexion test. Manual Therapy 2003a;8:92–6. Falla D, Jull G, Dall’Alba P, Rainoldi A, Merletti R. An electromyographic analysis of the deep cervical ﬂexor muscles in performance of craniocervical ﬂexion. Physical Therapy 2003b;83:899–906. Falla D. Unraveling the complexity of muscle impairment in chronic neck pain. Manual Therapy 2004;9:125–33. Falla D, Jull G, Hodges P. Patients with neck pain demonstrate reduced electromyographic activity of the deep cervical ﬂexor muscles during performance of the craniocervical ﬂexion test. Spine 2004;29(19):2108–14. Fernandez-de-las-penas C, Alonso-Blanco C, Cuadrado M, et al. Forward head posture and neck mobility in chronic tension-type headache: a blinded, controlled study. Cephalalgia 2006a;26:314–9. Fernandez-de-las-penas C, Alonso-Blanco C, Cuadrado M, et al. Trigger points in the suboccipital muscles and forward head posture in tension-type headache. Headache 2006b;46(3):454–60. Jull G. Deep cervical neck ﬂexor dysfunction in whiplash. Journal of Musculoskeletal Pain 2000;8:143–54. Jull G, Kristjansson E, Dall’Alba P. Impairment in the cervical ﬂexors: a comparison of whiplash and insidious onset neck pain patients. Manual Therapy 2004;9:89–94. Mayoux-Benhamou M, Revel M, Vallee C, et al. Longus colli has a postural function on cervical curvature. Surgical And Radiologic Anatomy 1994;16:367–71. O’Leary S, Falla D, Jull G, Vicenzino B. Muscle speciﬁcity in tests of cervical ﬂexor muscle performance. Journal of Electromyography and Kinesiology 2006 (epub). Raine S, Twomey L. Head and shoulder posture variations in 160 asymptomatic women and men. Archives of Physical Medicine and Rehabilitation 1997;78:1215–23. Raper AJ, Thompson Jr. WT, Shapiro W, Patterson JL. Scalene and sternomastoid muscle function. Journal of Applied Physiology 1966;21:497–502. Szeto G, Straker L, O’Sullivan P. A comparison of symptomatic and asymptomatic ofﬁce workers performing monotonous keyboard work-2: neck and shoulder kinematics. Manual Therapy 2005;10: 270–80.

ARTICLE IN PRESS 238

B. Cagnie et al. / Manual Therapy 13 (2008) 232–238

Treleaven J, Jull G, Atkinson L. Cervical musculoskeletal dysfunction in post-concussional headache. Cephalalgia 1994;14:273–9. Vasavada A, Li S, Delp S. Inﬂuence of muscle morphometry and moment arms on the moment-generating capacity of human neck muscles. Spine 1998;23:412–22.

Watson DH, Trott PH. Cervical headache: an investigation of natural head posture and upper cervical ﬂexor muscle performance. Cephalalgia 1993;13:272–84.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 239–248 www.elsevier.com/locate/math

Original article

The use of fear-avoidance beliefs and nonorganic signs in predicting prolonged disability in patients with neck pain Merrill R. Landers, Rachel V. Creger, Carrie V. Baker, Karl S. Stutelberg Department of Physical Therapy, School of Allied Health Sciences, Division of Health Sciences, University of Nevada, Las Vegas, 4505 Maryland Parkway, Box 453029, Las Vegas, NV 89154-3029, USA Received 28 April 2006; received in revised form 9 January 2007; accepted 23 January 2007

Abstract Psychological factors, such as fear-avoidance beliefs and nonorganic signs, have been postulated to play a role in the development of prolonged disability. The purpose of this study was to determine if fear-avoidance beliefs and nonorganic behavior are predictive of disability in patients with neck pain. Seventy-nine patients, with neck pain, were recruited from ﬁve outpatient physiotherapy clinics. Each of the patients completed a modiﬁed Fear-Avoidance Beliefs Questionnaire (FABQ) and was evaluated for the presence of cervical nonorganic signs (CNOS). The FABQ consists of two subscales pertaining to work (FABQ-W) and physical activity (FABQ-PA). The patients also completed the Neck Disability Index (NDI) during the initial examination and 12 weeks later. A 12-week NDI score X15 was operationally deﬁned as prolonged disability. In order to determine the overall predictive ability of the FABQ and CNOS, receiver operator characteristic (ROC) curves were used. The areas under the ROC curve were 0.782 (CNOS), 0.833 (FABQ-Total), 0.782 (FABQ-W) and 0.814 (FABQ-PA). Results from this study suggest that the FABQ and testing for CNOS are both good tools for predicting patients who may develop prolonged disability. r 2007 Elsevier Ltd. All rights reserved. Keywords: Neck pain; Fear-avoidance beliefs; Nonorganic signs; Neck disability index; Psychological factors; Physiotherapy

1. Introduction Neck pain, much like low back pain, is prone to chronicity and frequently results in prolonged disability (Makela et al., 1991; Pietri-Taleb et al., 1994). Moreover, the healthcare costs associated with this prolonged disability present a signiﬁcant economic drain on society (Luo et al., 2004b) and pose a signiﬁcant challenge to the healthcare provider. It has been suggested that, in addition to addressing the pathology and physical impairment of a condition, the healthcare provider should also consider the role that psychological factors play in the development of prolonged disability (Vlaeyen and Crombez, 1999; Luo et al., 2004a). It is now well Corresponding author. Tel.: +1 702 895 1377; fax: +1 702 895 4883. E-mail address: [email protected] (M.R. Landers).

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.01.010

recognized that many psychological factors are important determinants for risk of prolonged disability. Two such psychological factors, fear-avoidance beliefs and nonorganic signs, have received attention as predictors of prolonged disability in low back pain (Rose et al., 1992; Waddell et al., 1993; Klenerman et al., 1995; Karas et al., 1997; Crombez et al., 1999a,b; Vlaeyen and Crombez, 1999; Al-Obaidi et al., 2000; Fritz et al., 2001; Fishbain et al., 2003; Verbunt et al., 2003). However, these psychological factors have received relatively little attention for prolonged disability associated with neck pain. The fear-avoidance model of exaggerated pain perception suggests that patients with an exaggerated fear of pain will avoid physical activities that are anticipated to cause or increase pain (Lethem et al., 1983; Slade et al., 1983). This model proposes that there is a normal balance between sensation and emotion that occur with

ARTICLE IN PRESS 240

M.R. Landers et al. / Manual Therapy 13 (2008) 239–248

injury or disease. However, when this balance becomes disrupted, it can lead to prolonged disability. This maladaptive coping strategy (i.e. exaggerated fear leading to avoidance of physical activities) may stimulate self-imposed immobilization which may potentiate the ﬁbrosis and atrophy of the underlying injury. In addition, these patients may be less likely to engage in movement-related activities that promote healing and recovery. It has been said that this fear of pain is actually more disabling than the condition itself (Waddell et al., 1993; Crombez et al., 1999b). While research has suggested that a relationship exists between fear-avoidance beliefs and the chronicity of low back pain, there have been only a couple of studies that have speciﬁcally investigated fear-avoidance beliefs with cervical neck pain. Nederhand et al. (2004) found that fear-avoidance beliefs were predictive of chronic neck pain disability. However, George et al. (2001) reported a weak association between fear-avoidance beliefs and disability for those with chronic cervical pain. Thus, the relationship between fear-avoidance beliefs and neckrelated disability at the present is inconclusive. Another psychological factor, unrelated conceptually to fear-avoidance beliefs, that has received considerable attention in the literature is the concept of nonorganic signs. Nonorganic signs are ﬁndings, during a patient examination, that are suggestive of a physical illness or disease for which there is no demonstrable organic cause or physiological dysfunction. Simply put, nonorganic signs are ﬁndings that do not seem to be consistent with the nature of a particular pathology. In contrast, organic signs are ﬁndings that are consistent with the presence of pathology or disease. Intuitively, patients presenting with musculoskeletal dysfunction (e.g. neck pain) most commonly present with organic signs. However, occasionally a patient will, in addition to organic signs, exhibit nonorganic signs (i.e. signs not consistent with organic pathology). A high nonorganic presentation may be suggestive of psychological factors related to abnormal illness behavior, which may predispose the patient to a protracted recovery. The presence of a high nonorganic sign component in a patient with low back pain is thought to be indicative of psychological distress (Waddell et al., 1984, 1989; Main et al., 1992; Novy et al., 1998). Nonorganic signs have been reported to have a consistent correlation with the ‘neurotic triad’ (hypochondriasis, depression and hysteria scales) of the Minnesota Multiphasic Personality Inventory (Waddell et al., 1980; Novy et al., 1998). In addition, nonorganic signs have been shown to identify patients with chronic low back pain who may be depressed or anxious (Weaver et al., 2003/2004). More importantly, from a healthcare perspective, the presence of a high nonorganic component has been linked to prolonged disability (Waddell et al., 1980; Klenerman et al., 1995) and negative treatment out-

comes (McCulloch, 1977; Waddell et al., 1980, 1986; Lehmann et al., 1983; Dzioba and Doxey, 1984; Doxey et al., 1988; Werneke et al., 1993). Waddell et al. (1980) found a correlation between nonorganic signs and a number of treatment outcomes, including work loss, treatment failure, poor postsurgical results, and disability in patients with low back pain. Waddell and Main (1987) reported that nonorganic signs were found more frequently in patients with chronic low back pain as compared to those with shorter symptom duration. They also found that NOS were more prevalent in those with failed treatment. In addition, a number of other researchers have reported a correlation between NOS and poor return to work outcomes (Vallfors, 1985; Lancourt and Kettelhut, 1992; Werneke et al., 1993; Gaines and Hegmann, 1999). Until recently, there has been no standardized method for assessing nonorganic signs in patients with neck pain. In 2000, Sobel et al. (2000) developed and standardized nonorganic signs for cervical spine pain. These cervical nonorganic signs (CNOS) were patterned after the nonorganic signs for low back pain that were originally described by Waddell et al. (1980). While Sobel et al. (2000) reported good inter-rater reliability of CNOS, there have been no studies that have examined any aspects of validity of these signs in patients with neck pain. The purpose of this study was to determine if two different psychological factors (i.e. fear-avoidance beliefs and nonorganic behavior), both of which have been shown to be related to disability in low back pain, played a role in prolonged disability in patients with neck pain. If an association between these psychological factors and disability can be established, it will offer some initial evidence to the validity of these measures for patients with neck pain. As Waddell et al. (1980) originally noted, the ﬁnal proof of the validity of nonorganic signs (and, logically fear-avoidance beliefs) rests on their ability to predict outcome. In order to determine the role that fear-avoidance beliefs and nonorganic behavior play in prolonged neck disability, two research questions were considered: 1. Are fearavoidance beliefs and nonorganic behavior predictive of prolonged disability in patients with neck pain? and 2. Which of the cut-off values for fear-avoidance beliefs and nonorganic behavior maximizes prediction of prolonged disability of neck pain?

2. Methods 2.1. Patients Seventy-nine consecutive patients with neck pain (males ¼ 23, females ¼ 56) aged 19–71 (mean ¼ 49.6, SD ¼ 12.7), presenting to ﬁve outpatient physiotherapy clinics were recruited to participate in this study. Of this

ARTICLE IN PRESS M.R. Landers et al. / Manual Therapy 13 (2008) 239–248

total, 5 were acute (7 days or less since the onset of pain), 18 were sub-acute (between 8 days and 7 weeks since the onset of pain), and 56 were chronic (more than 7 weeks since the onset of pain). Because the entry point of patients to outpatient physiotherapy occurs at various stages from the onset of pain, we included all levels of acuity in our study. Each patient signed a consent form prior to participation in the study. Factors that excluded patients from this study were contraindications to physiotherapy intervention (i.e. neoplasm, congenital upper cervical instability, etc.), requirements for further diagnostic tests, and non-English speaking patients. 2.2. Procedures In order to determine the presence of fear-avoidance beliefs, patients completed a standardized questionnaire, the Fear-Avoidance Beliefs Questionnaire (FABQ), during the initial examination (Waddell et al., 1993). The FABQ consists of a total of 16 statements describing the effects of work and physical activity on back pain. Each statement has a six-point Likert scale, with anchors being ‘‘completely agree’’ and ‘‘completely disagree.’’ The total FABQ (FABQ-T) score ranges from 0 to 96, with higher scores being indicative of a high, self-assessed level of fear-avoidance beliefs. The

241

FABQ-T is separated into two subscales, a physical activity subscale (FABQ-PA) and a work subscale (FABQ-W). The FABQ-PA subscale consists of four statements regarding the fear of physical activity. Scores on this scale range from 0 to 24, with a higher score being indicative of high fear-avoidance beliefs related to general physical activity. The FABQ-W subscale consists of seven statements regarding the fear of work activity. Scores on this subscale range from 0 to 42, with higher scores being indicative of high fear-avoidance beliefs associated with work. The FABQ was slightly modiﬁed for this study: the word ‘back’ was changed to ‘neck’. During the initial evaluation, the patients were also evaluated for the presence of CNOS by the physiotherapist. CNOS are classiﬁed into ﬁve categories: palpation, simulation, cervical range of motion, regional disturbances, and overreaction (Table 1). One positive sign in a category equated to a positive sign for that category. Because there are ﬁve categories, scores ranged from 0 to 5, with a higher score being indicative of a higher level of nonorganic behavior. In order to assess neck disability, the Neck Disability Index (NDI) was completed by the patient during the initial evaluation and 12 weeks later (Vernon and Mior, 1991). Scores on this self-report scale range from 0 to 50, with higher scores being indicative of greater

Table 1 Cervical nonorganic signs Sign

Test site

Criteria for a positive test

Palpation of the cervical spine region and upper thoracic region Deep palpation of the cervical, thoracic, lumbar and brachial regions

Patient complains of pain with light touch or light pinching of the skin Patient complains of widespread tenderness, i.e. outside of the cervical and upper thoracic region

2. Simulation

Examiner rotates the patient’s head, shoulders, trunk and pelvis in the same plane

Patient complains of neck pain with trunk rotation

3. Cervical range of motion

Patient rotates head as far as possible to the right and then left

Rotation is less than 50% of normal in each direction

Light touch or pinprick

Patient reports diminished sensation in a pattern that does not correspond to a speciﬁc dermatome of a nerve root(s) or peripheral nerve(s) Weakness detected in a nonanatomic pattern; the hallmark being ‘‘giveaway weakness’’. Also positive if the patient is observed to have normal muscle strength but on formal test exhibits weakness

1. Palpation (A) Superﬁcial tenderness (B) Nonanatomic tenderness

4. Regional disturbances (A) Sensory loss

(B) Motor loss

5. Overreaction

Formal manual muscle testing, observation

Examiner’s observation

Examiner feels the patient is ‘‘overreacting’’ during the examination. Reliable behaviors include: 1. Moderate to extremely stiff, rigid, or slow movements 2. Rubbing the affected area for more than 3 s 3. Clutching, grasping, or squeezing the area for more than 3 s 4. Grimacing due to pain 5. Sighing

ARTICLE IN PRESS 242

M.R. Landers et al. / Manual Therapy 13 (2008) 239–248

neck-related disability. After the initial examination, patients participated in a physiotherapy treatment program. This treatment plan was not protocol driven; rather, it was determined by the treating physiotherapist and consisted of standard impairment-based physiotherapy intervention. Therapists were instructed to treat the patient as their individual impairment dictated. Therefore, the dosing and the modalities of the intervention were left to the discretion of the treating physiotherapist. The 12-week NDI measurement was used to determine prolonged disability, which was operationally deﬁned as an NDI score of X15. Based on our clinical experience, a score greater than 15 is consistent with a moderate to high level of disability. In addition, this cutoff score has been used to represent prolonged disability in a previous study (Nederhand et al., 2004). 2.3. Data analysis In order to determine the predictive value of fearavoidance beliefs and nonorganic signs in prolonged disability, the data for the total sample were analyzed using sensitivity and speciﬁcity values for each level of the FABQ-T, FABQ-PA, FABQ-W, and CNOS. The overall predictive accuracy of these measures was assessed using the area under the receiver operator characteristic (ROC) curve. These curves were calculated by using Dorfman’s and Alf’s (1969) method, wherein each curve is graphed with the true-positive rate (sensitivity) against the false-positive rate (1—speciﬁcity) for each cut-off score. The area under the ROC curve, which can range from 0.50, having no prognostic ability, to 1.00, having perfect prognostic ability, was computed for the dichotomous result (Xoro15 NDI) at 12 weeks for each of the variables. The area under the ROC curve represents the percentage of the time one would correctly classify a patient as disabled given a randomly chosen pair of patients. Thus, an area under the ROC of 0.800 would mean that one would correctly classify 80% of the time when presented with a randomly chosen pair of patients, one with prolonged disability and one without. The ROC curve also allows inference about which value on the scale or questionnaire serves as the optimal cut-off point for prediction (i.e. sensitivity and speciﬁcity). Usually the best cut-off point for the curve is where the curve makes a marked turn after the steep initial section. In addition, positive and negative likelihood ratios were calculated for each potential cut-off point. The positive likelihood ratio (+LR) indicates the change in odds favoring the condition given a positive test result. That is, the odds that someone with high fear-avoidance beliefs or CNOS would go on to develop disability. The negative likelihood ratio (LR) indicates the change in odds favoring the condition given a negative test result. That is, the odds that someone with low fear-avoidance

beliefs or CNOS would go on to develop disability. Therefore, the higher the +LR, the higher the chance of developing prolonged disability with a positive test result. Likewise, the lower the LR, the lower the odds of developing prolonged disability with a positive test result. Using the odds-likelihood formulation of Bayes’ theorem and the pretest probability of prolonged disability, the +LR will be used to calculate the posttest probability of prolonged disability (Gallagher, 1998). In order to better assess the role that acuity played in prolonged disability, the overall sample was divided into two parts: acute/sub-acute and chronic. Because of the small number of subjects in the acute and sub-acute subsamples, they were collapsed into one group (acute/subacute). All of the aforementioned statistical procedures were analyzed for the 23 patients that were in the acute/ sub-acute sub-sample. The 56 patients in the chronic sub-sample were also analyzed separately. In order to determine what variables were the best predictors of the 12-week NDI score, data were analyzed using multiple regression analysis (stepwise solution). Age, FABQ-PA, FABQ-W, FABQ-T, total CNOS, length of onset (acute, sub-acute and chronic), and initial NDI were included in the analysis. 3. Results Sensitivity, speciﬁcity, likelihood ratio, and posttest probability values were calculated for each possible cutoff score for each of the variables (FABQ-T, FABQ-W, FABQ-PA, and CNOS) for the overall sample and broken down into the acute/sub-acute and chronic subsamples (Table 2). Of the 79 patients for the overall sample, 29 would go on to develop prolonged disability (60.4% pretest probability). The areas under the ROC curves for the FABQ variables for the overall sample were 0.833 (95% CI: 0.741, 0.924) (FABQ-T), 0.782 (95% CI: 0.673, 0.891) (FABQ-W), and 0.814 (95% CI: 0.712, 0.915) (FABQ-PA) (Figs. 1–3). The area under the ROC curve for the CNOS variable for the total sample was 0.782 (95% CI: 0.668, 0.895) (Fig. 4). Based on the ROC curves, the following cut-off scores are best: 48 for the FABQ-T, 18 for the FABQ-W, 15 for the FABQ-PA, and 2 for CNOS (Table 2). Of the 23 patients in the acute/sub-acute sub-sample, eight were classiﬁed in the prolonged disability category (53.3% pretest probability). The areas under the ROC curves for the FABQ variables for the acute/sub-acute sample were 0.767 (95% CI: 0.566, 0.967) (FABQ-T), 0.654 (95% CI: 0.409, 0.900) (FABQ-W), and 0.783 (95% CI: 0.567, 0.999) (FABQ-PA) (Figs. 1–3). The area under the ROC curve for the CNOS variable was 0.758 (95% CI: 0.536, 0.980) (Fig. 4). The best cut-off scores for this sub-sample were 51 for the FABQ-T, 18 for the FABQ-W, and 15 for the FABQ-PA (Table 2). The best cut-off score for the CNOS was 2 (Table 2).

ARTICLE IN PRESS M.R. Landers et al. / Manual Therapy 13 (2008) 239–248

243

Table 2 Cut-off, sensitivity, speciﬁcity, likelihood values, and posttest probability for each of the clinical tools on their prediction of prolonged disability Cut-off score

Sensitivity

Speciﬁcity

+LR

LR

Posttest probability (%)

Total Acute/sub-acute Chronic

2 2 2

0.476 0.429 0.448

0.970 0.933 0.959

15.714 6.429 10.759

0.540 0.612 0.576

96.0 88.0 95.0

FABQ-T

Total Acute/sub-acute Chronic

48 51 41

0.655 0.625 0.714

0.854 0.800 0.788

4.493 3.125 3.367

0.404 0.469 0.363

87.3 78.1 85.5

FABQ-W

Total Acute/sub-acute Chronic

18 18 19

0.690 0.625 0.714

0.792 0.733 0.818

3.310 2.344 3.929

0.392 0.511 0.349

83.5 72.8 87.3

FABQ-PA

Total Acute/sub-acute Chronic

15 15 19

0.793 0.750 0.619

0.688 0.733 0.909

2.538 2.813 6.810

0.301 0.341 0.419

79.5 76.3 92.3

1

1

0.9

0.9

0.8

0.8

0.7

0.7

0.6

0.6

Sensitivity

Sensitivity

CNOS

0.5 0.4 Acute/Subacute

0.3

Acute/Subacute Chronic

0.2

Overall 0.1

0.4 0.3

Chronic

0.2

0.5

Overall 0.1

Reference

0

Reference

0 0

0.2

0.4

0.6

0.8

1

1-Specificity

0

0.2

0.4

0.6

0.8

1

1-Specificity

Fig. 1. ROC curve for the FABQ-T.

Fig. 2. ROC curve for the FABQ-W.

Of the 54 patients in the chronic sub-sample, 21 went on to develop prolonged disability (63.6% pretest probability). The areas under the ROC curves for the FABQ variables were 0.858 (95% CI: 0.757, 0.959) (FABQ-T), 0.821 (0.700, 0.942) (FABQ-W), and 0.826 (95% CI: 0.710, 0.943) (FABQ-PA) (Figs. 1–3). The area under the ROC curve for the CNOS variable was 0.790 (95% CI: 0.657, 0.923) (Fig. 4). The best cut-off scores for patients in the chronic sub-sample were 41 for the FABQ-T, 19 for the FABQ-W, and 19 for the FABQ-PA (Table 2). Again, the best cut-off score for the CNOS was 2 (Table 2). Results from the multiple regression analysis indicated that FABQ-PA, CNOS, and initial NDI contributed to the prediction of prolonged disability,

F(3,69) ¼ 47.736, po0.0005 (Table 3). These three independent variables accounted for 67.5% of the variance in 12-week NDI scores, R2 ¼ 0.675. Age, length of onset, FABQ-W, and FABQ-T were not included in the ﬁnal model. The stepwise solution indicated that the initial NDI (56.0% of the variance in 12-week NDI) was the best predictor of 12-week NDI scores, followed by FABQ-PA (additional 6.2% of the variance) and CNOS (additional 5.3% of the variance). Because length of onset (acute, sub-acute and chronic) was not included in the ﬁnal regression model, we can infer that length of onset did not play a signiﬁcant role in predicting the 12-week NDI score. To further explore this, we compared the initial presentation of the variables using acuity (acute/sub-acute versus chronic)

ARTICLE IN PRESS M.R. Landers et al. / Manual Therapy 13 (2008) 239–248

244

and found no statistically signiﬁcant differences between the two in initial FABQ total (p ¼ 0.651), initial FABQW (p ¼ 0.369), initial FABQ-PA (p ¼ 0.955), and total

initial CNOS (p ¼ 0.994). Likewise, there was no difference between the acute/sub-acute and chronic sub-samples for initial NDI scores (p ¼ 0.443).

1

4. Discussion

0.9 0.8 Sensitivity

0.7 0.6 0.5 0.4

Acute/Subacute

0.3

Chronic

0.2

Overall

0.1

Reference

0 0

0.2

0.4

0.6

0.8

1

1-Specificity Fig. 3. ROC curve for the FABQ-PA.

1 0.9 0.8

Sensitivity

0.7 0.6 0.5 0.4

Acute/Subacute

0.3

Chronic

0.2

Overall

0.1

Reference

0 0

0.2

0.4

0.6

0.8

1

1-Specificity

Fig. 4. ROC curve for CNOS.

These results provide evidence that the presence of prolonged disability in patients with neck pain is at least partially inﬂuenced by fear-avoidance beliefs and nonorganic behavior. More speciﬁcally, these results indicate that the higher the level of fear-avoidance beliefs, the higher the risk of prolonged disability. Additionally, a high nonorganic presentation is associated with a higher risk for prolonged disability. In the least, these results offer evidence of the predictive validity of these two constructs in patients with neck pain and suggest that future studies are warranted. Results from this study are consistent with previous research on the predictive value of fear-avoidance behavior to future poor outcome and/or disability (Lancourt and Kettelhut, 1992; Waddell et al., 1993; Gaines and Hegmann, 1999; Fritz et al., 2001; George et al., 2001; Boersma and Linton, 2005a,b, 2006; Verbunt et al., 2005; Nederhand et al., 2003, 2004). Fear-avoidance issues in patients with neck pain have also been reported (George et al., 2001; Nederhand et al., 2003, 2004; Boersma and Linton, 2005b, 2006). Nederhand et al. (2004) found that fear of movement in conjunction with baseline disability was predictive of prolonged disability with a probability of 83.3%. They found that the area under the ROC curve was 0.770 for fear of movement using the Tampa Scale of Kinesiophobia (TSK). While the FABQ, used in the present study, and the TSK are different measurement instruments, they both purport to measure the same construct. Our results are the ﬁrst to offer evidence that nonorganic signs are predictive of prolonged disability

Table 3 Multiple regression coefﬁcients (stepwise solution) for predicting 12-week NDI score Unstandardized coefﬁcients

Standardized coefﬁcients

B

Beta

SE

t statistic

95% Conﬁdence Interval for B

Lower

Upper

Model 1 R2 ¼ 0.560

Constant Initial NDI

37.209 0.773

2.669 0.081

0.748

13.939 (po0.0005) 9.508 (po0.0005)

31.887 0.935

42.532 0.611

Model 2 R2 ¼ 0.622

Constant Initial NDI FABQ-PA

24.209 0.588 0.495

4.590 0.094 0.147

0.569 0.306

5.274 (po0.0005) 6.279 (po0.0005) 3.373 (p ¼ 0.001)

15.054 0.775 0.202

33.364 0.401 0.787

Model 3 R2 ¼ 0.675

Constant Initial NDI FABQ-PA CNOS

17.297 0.423 0.482 2.016

4.754 0.100 0.137 0.600

0.410 0.298 0.283

3.639 4.224 3.521 3.361

7.814 0.623 0.209 0.820

26.780 0.223 0.755 3.212

(p ¼ 0.001) (po0.0005) (p ¼ 0.001) (p ¼ 0.001)

ARTICLE IN PRESS M.R. Landers et al. / Manual Therapy 13 (2008) 239–248

in patients with neck pain. Results from this study parallel ﬁndings from the low back pain literature which have demonstrated an association between nonorganic behavior and greater pain levels (Fishbain et al., 2003). To our knowledge, no other study has been conducted on CNOS other than the original introductory research (Sobel et al., 2000). However, there has been a considerable amount of research using low back nonorganic signs. While the results from these low back pain studies have not produced any deﬁnitive conclusions, there appears to be a consistent association with nonorganic signs and poorer nonsurgical treatment outcome (Lehmann et al., 1983; Dzioba and Doxey, 1984; Waddell et al., 1984; Doxey et al., 1988; Klenerman et al., 1995; Fishbain et al., 2003). The areas under the ROC curve, ranging from 0.782 to 0.833, offer support of the notion that fear-avoidance beliefs and nonorganic behavior are associated with prolonged neck-related disability. Based on these ROC curves, we proposed the following cut-off scores: 48 for the FABQ-T, 18 for the FABQ-W, 15 for the FABQ-PA, and 2 for CNOS. Thus, scores on these questionnaires at or higher than these cut-off points offer the best value for prediction of prolonged disability and make meaningful shifts in posttest probability (Table 2). There were no differences in the presentation of fearavoidance beliefs and CNOS based on their acuity at initial presentation. Patients already in the chronic stages of neck pain at initial entry point did not have more fear-avoidance beliefs, nonorganic signs, or neck disability. Instead, analyses of these acuity sub-samples (acute/sub-acute and chronic) revealed parallel ﬁndings to the overall sample. The areas under the ROC curve for the two sub-samples (chronic ¼ 0.790–0.858, and acute/sub-acute ¼ 0.654–0.783) suggest that these two variables offer predictive value for patients in different stages of acuity. The areas under the ROC curves were the strongest for the chronic sub-sample. This suggests that those already presenting to outpatient physiotherapy in the chronic stage were more likely to be classiﬁed as having prolonged disability if they presented with high fear-avoidance beliefs or high nonorganic behavior. From a clinical perspective, it is important to consider that even patients presenting with chronic pain are still at risk for a protracted recovery. The areas under the ROC curve were lowest for the acute/sub-acute sub-sample suggesting that the utility of these constructs in the early stages may not be as strong as it is in the chronic stages. Work by Sieben et al. (2005a,b) cast doubt on the fear-avoidance model in the acute stage of low back pain. However, they suggest that the fear-avoidance model may serve as a perpetuating factor once chronicity has developed. In our study, all of the areas under the ROC curve were still quite

245

strong (40.758) with the exception of the FABQ-W subscale which had a marginal area of 0.654. Moreover, fear-avoidance beliefs were of more value in the chronic stage. The initial level of neck-related disability played the biggest role in predicting prolonged disability. Initial disability accounted for 56% of the 12-week NDI variance. This is logical since neck-related disability may likely be a function of the severity of the injury. The strength of the inﬂuence of the two constructs in our study added an additional 6.2% (FABQ-PA) and 5.3% (CNOS) variance to the ﬁnal model over and above the variance contributed by the initial level of disability. Theoretically, the remaining variance would be logically partitioned out into other psychological factors and other physical impairment variables that are not directly measured by the NDI, such as pain intensity and previous history of pain, which have been shown to offer the best predictive value in patients with low back pain (Sieben et al., 2005b). Based solely on our results, we would suggest that while initial disability is the best predictor of neck-related disability, other variables, such as FABQ-PA and CNOS, are also predictive. From a clinical perspective, it would mean that the focus of our treatment should be on the physical factors related to prolonged disability; however, fear-avoidance beliefs and nonorganic behavior should also play a role in clinical decision-making. Early identiﬁcation of patients at risk for prolonged disability allows the physiotherapist to employ treatment strategies that have been shown to be effective in patients with high fear-avoidance beliefs (Symonds et al., 1995; Burton et al., 1999; de Jong et al., 2005). Linton et al. (2005) demonstrated that a cognitivebehavior educational program enhanced the prevention of prolonged disability. Additionally, the implementation of physiotherapy treatment speciﬁcally designed to treat those with high fear-avoidance beliefs resulted in less disability than those receiving standard physiotherapy (George et al., 2003). These studies offer evidence that addressing fear-avoidance beliefs through treatment and education may lessen the development of prolonged neck disability. Importantly, the FABQ may help in early identiﬁcation of patients at risk before their fear of movement precipitates a protracted condition. Previous researchers have suggested that an isolated nonorganic ﬁnding should be considered with caution (Scalzitti, 1997). That is, nonorganic ﬁndings may be present with some organic conditions. For example, consider the third nonorganic ﬁnding of range of motion less than 50% (Table 1). This sign was based on the notion that the majority of cervical rotation comes from the upper cervical spine, whereas the majority of cervical spine lesions are in the lower cervical spine (Sobel et al., 2000). While this may indeed be true, we were unable to

ARTICLE IN PRESS 246

M.R. Landers et al. / Manual Therapy 13 (2008) 239–248

ﬁnd epidemiologic evidence from the literature to support this. Work by Kasch et al. (2001a,b) also highlights the problem with using range of motion as a nonorganic variable for patients with neck pain. They reported that the best predictor of prolonged disability following whiplash injury was neck range of motion. Therefore, we recommend cautious interpretation of this sign since supporting evidence is lacking. Careful consideration of the other nonorganic signs is also recommended since evidence from the literature suggests that some of these signs may, in fact, be organic. For instance, pain to superﬁcial palpation may represent a central nervous system-mediated hypersensitivity reaction (Koelbaek Johansen et al., 1999; Curatolo et al., 2001). Additionally, it is possible that widespread dermatomal and/or myotomal changes may be present with signiﬁcant lateral stenosis. Others have reported that pain-associated hyperalgesia may spread outside the territory of the affected peripheral nerve distribution (Moriwaki and Yuge, 1999). For a more in-depth critical analysis of the etiologies of these signs, including regional weakness and overreaction, we recommend the review article by Fishbain et al. (2003). In light of these concerns, we caution the clinician not to overinterpret the presence of nonorganic signs. However, in the case of a patient presenting with a high nonorganic component, we suggest targeting interventions at modifying limiting factors (Feuerstein and Beattie, 1995) and creating positive expectations about their anticipated progress (Symonds et al., 1995). From a clinical perspective, we advise the clinician to remember that subjects who present with high fearavoidance beliefs and nonorganic behavior simply may be more prone to developing prolonged disability. Although there may be a causal relationship between these constructs and prolonged disability, these simple associations do not prove it. It is possible that something spurious may account for the associations found in our study. One limitation of this study was that there were too few patients in the acute stage of neck pain. Another weakness is that, in addition to the constructs investigated in this study, there may have been additional physical impairment variables (e.g. range of motion, pain level, instability, previous history, nature of injury (traumatic versus non-traumatic and sudden onset versus gradual), etc.) that may have contributed to disability. It is recommended, therefore, that future studies include physical impairment variables, in addition to psychological variables, to better analyze their shared or unshared roles. A longer follow-up may have provided a more complete picture of the course of disability in neck pain. As CNOS and FABQ purport to be measures of psychological involvement, future research should also consider correlating these measures with standardized psychological tests.

5. Conclusion Results of this study support the predictive validity of CNOS and the FABQ in patients with neck pain. Importantly, these tests provide valuable clinical information in identifying patients with neck pain who are at risk for prolonged disability. Identiﬁcation of these patients early in the process can lead to improved clinical decision making by facilitating treatment options that will best address the physical and psychological impairment of the patient.

References Al-Obaidi SM, Nelson RM, Al-Awadhi S, Al-Shuwaie N. The role of anticipation and fear of pain in the persistence of avoidance behavior in patients with chronic low back pain. Spine 2000; 25(9):1126–31. Boersma K, Linton SJ. How does persistent pain develop? An analysis of the relationship between psychological variables, pain and function across stages of chronicity. Behaviour Research and Therapy 2005a;43(11):1495–507. Boersma K, Linton SJ. Screening to identify patients at risk: proﬁles of psychological risk factors for early intervention. Clinical Journal of Pain 2005b;21(1):38–43. Boersma K, Linton SJ. Expectancy, fear and pain in the prediction of chronic pain and disability: a prospective analysis. European Journal of Pain 2006;10:551–57. Boersma K, Linton SJ. Expectancy, fear and pain in the prediction of chronic pain and disability: a prospective analysis. European Journal of Pain 2006;10:551–57. Burton AK, Waddell G, Tillotson KM, Summerton N. Information and advice to patients with back pain can have a positive effect. A randomized controlled trial of a novel educational booklet in primary care. Spine 1999;24(23):2484–91. Crombez G, Eccleston C, Baeyens F, van Houdenhove B, van den Broeck A. Attention to chronic pain is dependent upon pain-related fear. Journal of Psychosomatic Research 1999a;47(5):403–10. Crombez G, Vlaeyen JW, Heuts PH, Lysens R. Pain-related fear is more disabling than pain itself: evidence on the role of pain-related fear in chronic back pain disability. Pain 1999b;80(1–2):329–39. Curatolo M, Petersen-Felix S, Arendt-Nielsen L, Giani C, Zbinden AM, Radanov BP. Central hypersensitivity in chronic pain after whiplash injury. Clinical Journal of Pain 2001;17(4):306–15. de Jong JR, Vlaeyen JW, Onghena P, Goossens ME, Geilen M, Mulder H. Fear of movement/(re)injury in chronic low back pain: education or exposure in vivo as mediator to fear reduction? Clinical Journal of Pain 2005;21(1):9–17. Dorfman DD, Alf E. Maximum-likelihood estimation of parameters of signal-detection theory and determination of conﬁdence intervals-rating-method data. Journal of Mathematical Psychology 1969;6:487–96. Doxey NC, Dzioba RB, Mitson GL, Lacroix JM. Predictors of outcome in back surgery candidates. Journal of Clinical Psychology 1988;44(4):611–22. Dzioba RB, Doxey NC. A prospective investigation into the orthopaedic and psychologic predictors of outcome of ﬁrst lumbar surgery following industrial injury. Spine 1984;9(6):614–23. Feuerstein M, Beattie P. Biobehavioral factors affecting pain and disability in low back pain: mechanisms and assessment. Physical Therapy 1995;75(4):267–80. Fishbain DA, Cole B, Cutler RB, Lewis J, Rosomoff HL, Rosomoff RS. A structured evidence-based review on the meaning of

ARTICLE IN PRESS M.R. Landers et al. / Manual Therapy 13 (2008) 239–248 nonorganic physical signs: Waddell signs. Pain Medicine 2003;4(2): 141–81. Fritz JM, George SZ, Delitto A. The role of fear-avoidance beliefs in acute low back pain: relationships with current and future disability and work status. Pain 2001;94(1):7–15. Gaines Jr. WG, Hegmann KT. Effectiveness of Waddell’s nonorganic signs in predicting a delayed return to regular work in patients experiencing acute occupational low back pain. Spine 1999;24(4): 396–400. Gallagher EJ. Clinical utility of likelihood ratios. Annals of Emergency Medicine 1998;31(3):391–7. George SZ, Fritz JM, Erhard RE. A comparison of fear-avoidance beliefs in patients with lumbar spine pain and cervical spine pain. Spine 2001;26(19):2139–45. George SZ, Fritz JM, Bialosky JE, Donald DA. The effect of a fearavoidance-based physical therapy intervention for patients with acute low back pain: results of a randomized clinical trial. Spine 2003;28(23):2551–60. Karas R, McIntosh G, Hall H, Wilson L, Melles T. The relationship between nonorganic signs and centralization of symptoms in the prediction of return to work for patients with low back pain. Physical Therapy 1997;77(4):354–60. Kasch H, Bach FW, Jensen TS. Handicap after acute whiplash injury: a 1-year prospective study of risk factors. Neurology 2001a;56(12): 1637–43. Kasch H, Stengaard-Pedersen K, Arendt-Nielsen L, Staehelin Jensen T. Headache, neck pain, and neck mobility after acute whiplash injury: a prospective study. Spine 2001b;26(11):1246–51. Klenerman L, Slade PD, Stanley IM, Pennie B, Reilly JP, Atchison LE, et al. The prediction of chronicity in patients with an acute attack of low back pain in a general practice setting. Spine 1995; 20(4):478–84. Koelbaek Johansen M, Graven-Nielsen T, Schou Olesen A, ArendtNielsen L. Generalised muscular hyperalgesia in chronic whiplash syndrome. Pain 1999;83(2):229–34. Lancourt J, Kettelhut M. Predicting return to work for lower back pain patients receiving worker’s compensation. Spine 1992;17(6): 629–40. Lehmann TR, Russell DW, Spratt KF. The impact of patients with nonorganic physical ﬁndings on a controlled trial of transcutaneous electrical nerve stimulation and electroacupuncture. Spine 1983;8(6):625–34. Lethem J, Slade PD, Troup JD, Bentley G. Outline of a fear-avoidance model of exaggerated pain perception—I. Behaviour Research and Therapy 1983;21(4):401–8. Linton SJ, Boersma K, Jansson M, Svard L, Botvalde M. The effects of cognitive-behavioral and physical therapy preventive interventions on pain-related sick leave: a randomized controlled trial. Clinical Journal of Pain 2005;21(2):109–19. Luo X, Edwards CL, Richardson W, Hey L. Relationships of clinical, psychologic, and individual factors with the functional status of neck pain patients. Value Health 2004a;7(1):61–9. Luo X, Pietrobon R, Sun SX, Liu GG, Hey L. Estimates and patterns of direct health care expenditures among individuals with back pain in the United States. Spine 2004b;29(1):79–86. Main CJ, Wood PL, Hollis S, Spanswick CC, Waddell G. The distress and risk assessment method. A simple patient classiﬁcation to identify distress and evaluate the risk of poor outcome. Spine 1992; 17(1):42–52. Makela M, Heliovaara M, Sievers K, Impivaara O, Knekt P, Aromaa A. Prevalence, determinants, and consequences of chronic neck pain in Finland. American Journal of Epidemiology 1991;134(11): 1356–67. McCulloch JA. Chemonucleolysis. Journal of Bone and Joint Surgery British Volume 1977;59(1):45–52.

247

Moriwaki K, Yuge O. Topographical features of cutaneous tactile hypoesthetic and hyperesthetic abnormalities in chronic pain. Pain 1999;81(1–2):1–6. Nederhand MJ, Hermens HJ, IJzerman MJ, Turk DC, Zilvold G. Chronic neck pain disability due to an acute whiplash injury. Pain 2003;102(1–2):63–71. Nederhand MJ, IJzerman MJ, Hermens HJ, Turk DC, Zilvold G. Predictive value of fear avoidance in developing chronic neck pain disability: consequences for clinical decision making. Archives of Physical Medicine and Rehabilitation 2004;85(3): 496–501. Novy DM, Collins HS, Nelson DV, Thomas AG, Wiggins M, Martinez A, et al. Waddell signs: distribution properties and correlates. Archives of Physical Medicine and Rehabilitation 1998; 79:820–2. Pietri-Taleb F, Riihimaki H, Viikari-Juntura E, Lindstrom K. Longitudinal study on the role of personality characteristics and psychological distress in neck trouble among working men. Pain 1994;58(2):261–7. Rose MJ, Klenerman L, Atchison L, Slade PD. An application of the fear avoidance model to three chronic pain problems. Behaviour Research and Therapy 1992;30(4):359–65. Scalzitti DA. Screening for psychological factors in patients with low back problems: Waddell’s nonorganic signs. Physical Therapy 1997;77(3):306–12. Sieben JM, Portegijs PJ, Vlaeyen JW, Knottnerus JA. Pain-related fear at the start of a new low back pain episode. European Journal of Pain 2005a;9(6):635–41. Sieben JM, Vlaeyen JW, Portegijs PJ, Verbunt JA, van Riet-Rutgers S, Kester AD, et al. A longitudinal study on the predictive validity of the fear-avoidance model in low back pain. Pain 2005b;117(1–2): 162–70. Slade PD, Troup JD, Lethem J, Bentley G. The fear-avoidance model of exaggerated pain perception—II. Behaviour Research and Therapy 1983;21(4):409–16. Sobel JB, Sollenberger P, Robinson R, Polatin PB, Gatchel RJ. Cervical nonorganic signs: a new clinical tool to assess abnormal illness behavior in neck pain patients: a pilot study. Archives of Physical Medicine and Rehabilitation 2000;81(2):170–5. Symonds TL, Burton AK, Tillotson KM, Main CJ. Absence resulting from low back trouble can be reduced by psychosocial intervention at the work place. Spine 1995;20(24):2738–45. Vallfors B. Acute, subacute and chronic low back pain: clinical symptoms, absenteeism and working environment. Scandinavian Journal of Rehabilitation Medicine Supplement 1985;11:1–98. Verbunt JA, Seelen HA, Vlaeyen JW, van der Heijden GJ, Knottnerus JA. Fear of injury and physical deconditioning in patients with chronic low back pain. Archives of Physical Medicine and Rehabilitation 2003;84(8):1227–32. Verbunt JA, Sieben JM, Seelen HA, Vlaeyen JW, Bousema EJ, van der Heijden GJ, et al. Decline in physical activity, disability and painrelated fear in sub-acute low back pain. European Journal of Pain 2005;9(4):417–25. Vernon H, Mior S. The Neck Disability Index: a study of reliability and validity. Journal of Manipulative and Physiological Therapeutics 1991;14(7):409–15. Vlaeyen JW, Crombez G. Fear of movement/(re)injury, avoidance and pain disability in chronic low back pain patients. Manual Therapy 1999;4(4):187–95. Waddell G, Main CJ. Chronic backache, distress, illness behaviour and surgery. Acta Orthopaedica Belgica 1987;53(2):265–8. Waddell G, McCulloch JA, Kummel E, Venner RM. Nonorganic physical signs in low-back pain. Spine 1980;5(2):117–25. Waddell G, Bircher M, Finlayson D, Main CJ. Symptoms and signs: physical disease or illness behaviour? British Medical Journal (Clinical Research ed.) 1984;289(6447):739–41.

ARTICLE IN PRESS 248

M.R. Landers et al. / Manual Therapy 13 (2008) 239–248

Waddell G, Morris EW, Di Paola MP, Bircher M, Finlayson D. A concept of illness tested as an improved basis for surgical decisions in low-back disorders. Spine 1986;11(7):712–9. Waddell G, Pilowsky I, Bond MR. Clinical assessment and interpretation of abnormal illness behaviour in low back pain. Pain 1989;39(1):41–53. Waddell G, Newton M, Henderson I, Somerville D, Main CJ. A FearAvoidance Beliefs Questionnaire (FABQ) and the role of fear-

avoidance beliefs in chronic low back pain and disability. Pain 1993;52(2):157–68. Weaver CS, Kvaal SA, McCracken L. Waddell signs as behavioral indicators of depression and anxiety in chronic pain. Journal of Back and Musculoskeletal Rehabilitation 2003/2004;17(1):21–6. Werneke MW, Harris DE, Lichter RL. Clinical effectiveness of behavioral signs for screening chronic low-back pain patients in a work-oriented physical rehabilitation program. Spine 1993;18(16): 2412–8.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 249–257 www.elsevier.com/locate/math

Original article

Postural taping decreases thoracic kyphosis but does not inﬂuence trunk muscle electromyographic activity or balance in women with osteoporosis Alison M. Greiga,b,, Kim L. Bennella, Andrew M. Briggsa,b, Paul W. Hodgesc a

Centre for Health, Exercise and Sports Medicine, School of Physiotherapy, University of Melbourne, Victoria 3010, Australia b Department of Medicine, Royal Melbourne Hospital, University of Melbourne, Victoria, Australia c Division of Physiotherapy, The University of Queensland, Australia Received 7 June 2006; received in revised form 28 December 2006; accepted 16 January 2007

Abstract Background: Greater thoracic kyphosis is associated with increased biomechanical loading of the spine which is potentially problematic in individuals with osteoporotic vertebral fractures. Conservative interventions that reduce thoracic kyphosis warrant further investigation. This study aimed to investigate the effects of therapeutic postural taping on thoracic posture. Secondary aims explored the effects of taping on trunk muscle activity and balance. Methods: Fifteen women with osteoporotic vertebral fractures participated in this within-participant design study. Three taping conditions were randomly applied: therapeutic taping, control taping and no taping. Angle of thoracic kyphosis was measured after each condition. Force plate-derived balance parameters and trunk muscle electromyographic activity (EMG) were recorded during three static standing tasks of 40 s duration. Results: There was a signiﬁcant main effect of postural taping on thoracic kyphosis (p ¼ 0.026), with a greater reduction in thoracic kyphosis after taping compared with both control tape and no tape. There were no effects of taping on EMG or balance parameters. Conclusions: The results of this study demonstrate that the application of postural therapeutic tape in a population with osteoporotic vertebral fractures induced an immediate reduction in thoracic kyphosis. Further research is needed to investigate the underlying mechanisms associated with this decrease in kyphosis. r 2007 Elsevier Ltd. All rights reserved. Keywords: Postural taping; Osteoporosis; Vertebral fracture; Thoracic kyphosis

1. Introduction Vertebral fractures are the hallmark of osteoporosis, and are associated with signiﬁcant physical impairments (Greendale et al., 1995; Huang et al., 1996; Hall et al., 1999). One such impairment is increased thoracic kyphoCorresponding author. Centre for Health, Exercise and Sports Medicine, School of Physiotherapy, University of Melbourne, Victoria 3010, Australia. Tel.: +61 3 8344 4171; fax: +61 3 8344 4188. E-mail addresses: [email protected] (A.M. Greig), [email protected] (K.L. Bennell), [email protected] (A.M. Briggs), [email protected] (P.W. Hodges).

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.01.011

sis, which itself is associated with increased spinal loading, back extensor muscle weakness, limitations in functional activities, and increased risk of further fracture. The relationship between vertebral fracture and thoracic kyphosis is recognised in the literature, with agreement that greater numbers of vertebral fractures are associated with increases in thoracic kyphosis (Ensrud et al., 1997; Cortet et al., 2002). Considering this association, interventions that decrease thoracic kyphosis may reduce spinal load, and therefore may potentially decrease the risk of further fracture. However, whether thoracic kyphosis and the associated risk for vertebral fracture can be changed with conservative interventions is unclear.

ARTICLE IN PRESS 250

A.M. Greig et al. / Manual Therapy 13 (2008) 249–257

Thoracic kyphosis is associated with increased compression loading through the spine, which results in greater vertebral and intervertebral disc loads (Keller et al., 2003). In a vertebral fracture population, thoracic kyphosis is correlated with increased vertebral loading (Briggs et al., 2007). Thus, theoretically, a reduction in thoracic kyphosis may have the potential to reduce fracture risk. In addition to kyphosis, spinal loading is also increased by co-contraction of trunk ﬂexor and extensor muscles (Marras et al., 2001; Diee¨n et al., 2003). Recently, we have shown that individuals with osteoporosis, both with and without vertebral fracture, have increased co-contraction of trunk muscles in association with tasks that challenge balance (Greig, 2006). Together, greater thoracic kyphosis and cocontraction of trunk muscles would substantially increase spinal loading, and when this occurs in individuals with fragile bones and vertebral fracture, this may predispose the individual to further fracture. Thus interventions would ideally modify both thoracic kyphosis and trunk muscle activity. A further issue that complicates the association between thoracic kyphosis and vertebral fracture is that increased thoracic kyphosis is also associated with impaired balance control in individuals with osteoporosis (Lynn et al., 1997; Cook, 2002; Balzini et al., 2003), and this may further contribute to the development of fracture as a result of falling. It has been argued that increased forward curvature of the spine displaces the centre of mass anteriorly towards the limits of stability, and thus increases the likelihood of loss of balance (Horak et al., 1989; Yuan et al., 2004). However, the relationship between thoracic kyphosis and balance is unclear as we have recently shown that impaired balance is more closely associated with vertebral fracture than thoracic kyphosis (Greig et al., 2006). When testing interventions that change thoracic kyphosis to reduce risk for further fracture, it is essential to also investigate the potential for these interventions to concurrently affect balance. In light of the previous research, treatments that aim to reduce thoracic kyphosis and alter trunk muscle activity may have positive effects on vertebral loading, trunk control, and balance, which in turn may reduce the risk for fracture. Conservative management techniques which aim to reduce thoracic kyphosis, such as therapeutic postural taping, spinal orthoses and postural retraining, warrant further investigation. Positive outcomes have been found after combined interventions of spinal orthoses and exercise (Sinaki and Lynn, 2002; Pfeifer et al., 2004; Sinaki et al., 2005). These studies report decreased thoracic kyphosis in association with reduced back extensor muscle activity, and positive effects on pain, mobility, and quality of life. Postural taping is used by physiotherapists in conjunction with exercise to manage individuals with

increased thoracic kyphosis (Bennell et al., 2000). Like spinal orthoses, postural taping aims to decrease forward curvature of the spine, reduce pain associated with thoracic kyphosis, and facilitate activity of the postural muscles in a more optimal spinal position. However, few studies have investigated the effect of taping on spinal posture, and no studies have investigated the effects of thoracic kyphosis on trunk muscle activity or other functional outcomes, such as balance. Therefore, the primary aim of this study was to investigate whether therapeutic postural taping changes thoracic posture in women with osteoporosis and vertebral fracture. The secondary aims were to explore whether changes in posture, if present, are associated with changes in trunk muscles activity and balance ability.

2. Methods 2.1. Participants Fifteen participants with osteoporotic vertebral fractures were recruited through advertisements in local newspapers, osteoporosis support groups, osteoporosis clinics, endocrinologists, and general practitioners. Sample size was calculated a-priori. Change in posture was selected as the outcome measure in which we based our calculation, as it was the primary aim of the postural taping. Based on previous literature reporting intervention-based changes in thoracic kyphosis (Wang et al., 1999; Itoi and Sinaki, 1994), we assumed a change of 4.072.01 in the magnitude of thoracic kyphosis to be clinically signiﬁcant. The type I error (a) was set to 0.05. Repeated measures analysis was performed and type II error (b) was set at 0.20. It was calculated that a total sample size of 15 participants would give 80% power to detect a difference in changes in thoracic kyphosis between the various conditions with variability between measures of r ¼ 0.6. The descriptive statistics for the population are presented in Table 1. Participants were included if they were community dwelling, 50 years or older, at least 5 years post-menopause, and had sustained a vertebral fracture within the past 2 years (greater than 3 months prior to testing) that was clinically recognised by the presence of pain at the time of fracture. Exclusion criteria included any medical conditions other than osteoporosis that could affect bone metabolism or balance, or participation in activities or rehabilitation that could affect trunk neuromuscular control or balance. Subjects participated in a concurrent study, and all participants provided written, informed consent. Ethical approval to conduct the study was granted by the institutional ethics committee.

ARTICLE IN PRESS A.M. Greig et al. / Manual Therapy 13 (2008) 249–257

251

Table 1 Descriptive statistics for participants expressed as mean (SD)

Height (cm) Weight (kg) Age (years) BMI (kg/m2) PASEa Tx kyphosis (deg.)

Pooled n ¼ 15

Decreased kyphosis n ¼ 11

Increased kyphosis n¼4

160.6 (1.5) 68.6 (2.8) 67.2 (2.5) 26.6 (0.9) 169 (50) 58.2 (3.2)

162.0 (5.7) 69.9 (10.7) 65.5 (8.9) 26.7 (3.5) 164 (50) 61.5 (3.5)

157.1 (5.1) 65.5 (10.7) 72.0 (12.1) 26.6 (4.1) 169 (50) 49.1 (7.5)

p ¼ 0.163 p ¼ 0.500 p ¼ 0.269 p ¼ 0.973 p ¼ 0.418 p ¼ 0.086b

The ‘‘decreased kyphosis’’ group includes participants that decreased kyphosis with therapeutic taping, and the ‘‘increased kyphosis’’ group includes those who increased with therapeutic taping. Paired t-tests examined differences between those individuals who decreased thoracic kyphosis with therapeutic tape compared to those individuals who increased kyphosis with therapeutic tape. Note the trend (b) towards greater thoracic kyphosis in the group that experience a decrease in thoracic kyphosis with postural taping compared to those who experienced an increase in kyphosis. a Physical Activity Scale for the Elderly (PASE) questionnaire. b Trend.

A diagnosis of osteoporosis was determined based on bone densitometry results in accordance with the guidelines established by the World Health Organisation (Alexeeva et al., 1994). Identiﬁcation of vertebral fracture was made from a standardised, plain ﬁlm, lateral radiograph of the lumbar and thoracic spine. These images were taken as participants adopted a selfdeﬁned, relaxed standing posture. A semi-quantitative assessment was used, following guidelines recommended by Genant et al. (1996). Anterior and posterior vertebral heights of the vertebrae from T1 to L5 were calculated. According to the protocol outlined by McCloskey et al. (1993) a vertebral body was classiﬁed as ‘‘fractured’’ when two criteria were fulﬁlled at each site. Vertebrae were classiﬁed as wedge-fractured if the anterior height was reduced by X30% compared with the posterior height and the posterior height of the adjacent superior or inferior vertebra. Two participants had sustained two vertebral fractures, and the remaining participants had one vertebral fracture. Radiographs were further reviewed qualitatively by a radiologist to ensure that compression fractures were not overlooked. Participants reported minimal or no pain in the trunk and lower limbs (less than 2/10 on the numerical rating scale) at the time of testing. 2.2. Thoracic kyphosis Thoracic kyphosis was measured using a dual inclinometer (The Dualer Electric Inclinometer, J. Tech, 1992, North American Fork, Utah). Measurement protocol was followed by an experienced physiotherapist (AG), and according to the instruction manual (Livingston and Livingston, 1992) (see Fig. 1). The use of inclinometers to measure spinal curvature has been found to be reliable (Mellin, 1986; Ng et al., 2001) and valid (Saur et al., 1996), and the use of inclinometers to measure thoracic kyphosis in the elderly population is reported in the literature (O’Brien et al., 1997).

Fig. 1. Use of the Dualer digital inclinometer to measure angle of thoracic kyphosis. Spinous processes of T1 and T12 were used as landmarks for positioning the inclinometer sensors. The angle of the intersection of the dashed lines demonstrates the angle of thoracic kyphosis.

Measurements were made prior to testing, and after each of the taping conditions (therapeutic tape, control tape, no tape). Same day, intra-rater reliability of the thoracic kyphosis measurement used in this study was established in 10 of the participants during their participation in a concurrent study (Greig, 2006). The concurrent study did not include any postural interventions or activities that would affect posture. The mean of two thoracic kyphosis measures were taken in the participants’ resting posture at each of two testing intervals. Anatomical landmarks (T1 and T12) were

ARTICLE IN PRESS 252

A.M. Greig et al. / Manual Therapy 13 (2008) 249–257

identiﬁed with stickers and removed between the intervals. Intra-class correlation coefﬁcients (ICC), standard error of measurement (SEM) and corresponding 95% conﬁdence intervals (CI) were calculated. Measurements taken one hour apart indicated a high reproducibility (ICC [1,1] ¼ 0.93), (95% CI ¼ 0.66– 0.99), and the SEM was 21 or 74.3% of total thoracic kyphosis.

involved application of the hypo-allergenic tape which was considered to provide minimal mechanical support. The tape was laid over the skin while participants were asked to stand in their normal resting posture. There was no tension applied through the control tape during application.

2.3. Postural taping application technique

Recordings of trunk muscle electromyography (EMG) and balance were collected while participants performed three quiet standing tasks. Participants stood with bare feet on a force platform within a frame consisting of adjustable rails on three sides, and were advised that they could use the rails at any time if they felt unsafe. Participants were instructed to stand in the centre of the force plate with feet shoulder-width apart, toes aligned forwards, and equal weight through both feet. Participants’ feet measured and marked to align with the centre markings on the force plate. The three quiet standing variations involved (i) standing on a ﬂat surface with eyes open, (ii) standing ﬂat surface with eyes closed, and (iii) standing on a short base with eyes open. The short base consisted of a piece of wood (4 cm in height by 12 cm in width). Participants were asked to stand on the short base so that their toes and heels did not make contact with the force plate below. The aim of the short base was to increase the difﬁculty of the balance task and to limit the use of ankle torque to maintain balance, thus forcing the use of hip and trunk movement (Horak and Nashner 1986). Participants performed each of the quiet standing tasks for each of the three taping conditions (postural tape, control tape, no-tape). The order of both the surface conditions and the taping conditions were randomised in a balanced latin-square design, and data were collected for 40 s during each task.

The postural taping technique was previously demonstrated by Bennell et al. (2000). Prior to the application therapeutic tape, participants were instructed to ‘‘elongate the crown of the head towards to ceiling and gently draw shoulder blades down and together’’. Participants were instructed to maintain the position during the application of the tape. Prior to the rigid tape application, hypo-allergenic tape (Fixomull Stretch, Beiersdorf Ltd., North Ryde, NSW, Australia) was applied to the skin to minimise negative skin reactions. The rigid therapeutic tape (Leuko Sportstape Premium Plus, Beiersdorf Ltd., North Ryde, NSW, Australia) was applied ﬁrmly from the anterior aspect of the acromioclavicular joint, coursed over the muscle bulk of upper trapezius, and then diagonally towards the spinous process of T6. As the tape was applied bilaterally, there was an intersection at T6 (Fig. 2). In contrast to the therapeutic tape, the control tape was designed to provide sensory input only, and

2.4. Task protocol

2.5. Electromyography

Fig. 2. Therapeutic postural taping: Tape is applied from the anterior aspect of the acromioclavicular joint, over the muscle bulk of upper trapezius, and then diagonally towards the spinous process of T6. As the tape was applied bilaterally, there was an intersection at T6. Electrode positions of UT, LT, and ES (T8 and L3) are depicted.

EMG was recorded with pairs of Ag/AgCl adhesive electrodes (diameter: 1 cm) (Meditrace, Kendall LTP), arranged in a bipolar conﬁguration with an interelectrode distance of 2 cm. EMG activity of seven trunk muscles was recorded: obliquus internus (OI) and externus abdominis (OE), rectus abdominis (RA), erector spinae (ES) at T8 and L3, and upper (UT) and lower trapezius (LT). Electrodes were placed over the left OI, OE and RA according to guidelines of Ng et al. (1998), over the left ES at T8 and L3 according Schultz et al. (1982), and over the right UT and LT according to Basmajian and DeLuca (1985). Surface EMG has good repeatability, especially if electrodes are not removed (Viitasalo and Komi, 1975; Hakkinen and Komi, 1983; Yang and Winter, 1983; Kabada et al., 1985), and as this study was a within-participant design, good reliability of

ARTICLE IN PRESS A.M. Greig et al. / Manual Therapy 13 (2008) 249–257

EMG measures was expected. A ground electrode was placed over the right iliac crest. EMG data were ampliﬁed 1000 times, band-pass ﬁltered between 20 and 1000 Hz using a second-order Butterworth 12 dB/ octave ﬁlter and sampled at 2000 Hz. A notch ﬁlter was used at 50 Hz to reduce electrical artefact. Data were recorded and stored using Spike 2, version 4.10 software (Cambridge Electronic Design Limited, Cambridge, England), and exported for analysis with Matlab 6.5.0 (The Mathworks, Natick, MA, USA). Mean EMG amplitude over a 5-s period was calculated by visually selecting a representative period from the raw EMG data from the 40-s trial. To remove bias, EMG traces were displayed individually and without indication of muscle. As performance of a maximum voluntary contraction for normalisation purposes was not appropriate in this population due to vertebral fracture and fragility, raw EMG of each muscle was normalised to the peak EMG for that muscle across all conditions to permit comparison between taping conditions and balance tasks. 2.6. Force plate data Ground reaction forces were recorded with a piezoelectric Kistler force plate (Type 9286AA, Kistler AG, Winterthur, Switzerland). Measures of centre of pressure (COP) were derived from ground reaction forces. COP data was calculated and stored using Bioware software (Version 3.21, Kistler Instrument Corp., Kistler AG, Winterthur, Switzerland). Force plate data were lowpass ﬁltered at 10 Hz to reduce measurement noise and to remove drifts in the strain gauge apparatus and sampled at 1000 Hz. COP data were exported from Bioware for analysis with Matlab 6.5.0 (The Mathworks, Natick, MA, USA). Measures used in this study were: range, root mean square (RMS) and total path of COP displacement (COPx, COPy, COPxy), range, RMS and resultant velocity (COPvx, COPvy, COPvxy), and range and RMS of shear force (COPfx, COPfy) (Table 2). Range is indicative of the greatest amplitude of the measure, whereas RMS measures reﬂect the average amount that the measure varies around its mean. The mean COP position in the AP direction (MeanCOPx) was also calculated. 2.7. Statistical analysis Change in magnitude of thoracic kyphosis was explored using repeated measures analysis of variance (ANOVA) with one repeated measure (tape condition). The per cent-change between normal standing posture (taken prior to testing) and the taping and control taping conditions were also compared using a paired t-test. Pearson’s correlation was calculated to examine the association between initial kyphosis and the percentage

253

Table 2 Table of abbreviations of balance parameters Abbreviation

Balance parameter

COPx COPy COPxy COPvx COPvy COPvxy COPfx COPfy MeanCOPx EOFB ECFB EOSB

COP displacement in the AP direction COP displacement in the ML direction COP resultant displacement (total path) COP velocity in the AP direction COP velocity in the ML direction COP resultant velocity COP shear force in the AP direction COP shear force in the ML direction Mean position of COP in the AP direction Eyes open, ﬂat base Eyes closed, ﬂat base Eyes open, short base

change with taping. Differences in EMG amplitude and COP parameters between taping conditions (therapeutic tape, control tape, no tape) and surface conditions (ﬂat base, eyes open; ﬂat base, eyes closed; short base) were explored with two-way repeated measures ANOVA. Signiﬁcant differences were explored with Duncan’s post hoc analyses and Sharpened Bonferroni adjustments were applied (Hochberg and Benjamini, 1990). The level of statistical signiﬁcance was set at po0.05. All statistical analyses were conducted using SPSS for windows, version 11.0.1. (SPSS Inc., Chicago, Illinois, USA).

3. Results A signiﬁcant main effect was found for the inﬂuence of tape on thoracic kyphosis (p ¼ 0.026). Post hoc analysis revealed that thoracic kyphosis was decreased immediately after application of therapeutic tape (mean7SD: 55.3713.51) compared with control tape (57.2713.81), (p ¼ 0.043) and no tape (58.2712.31), (p ¼ 0.024). There was no difference between control tape and no tape (p ¼ 0.377). Application of therapeutic tape reduced kyphosis by 5.270.9%, whereas control taping was associated with a mean reduction of 2.170.8%. The difference between therapeutic taping and control taping was statistically signiﬁcant (p ¼ 0.048). As the SEM of measurement for per cent change in thoracic kyphosis was calculated to be 4.3%, the change with taping is greater than the error in measurement. Eleven of the 15 participants (73%) demonstrated a mean reduction (8.578.0%) in thoracic kyphosis with application of therapeutic tape. There was a trend for the 11 participants who reduced thoracic kyphosis to have greater initial thoracic kyphosis (p ¼ 0.086) compared with the four participants who increased thoracic kyphosis with taping (Table 1). Of the four participants who demonstrated an increase in

ARTICLE IN PRESS A.M. Greig et al. / Manual Therapy 13 (2008) 249–257

254

thoracic kyphosis with therapeutic taping, all four had an increase of less than 21, and an increase in the per cent change of thoracic kyphosis of less than 3%. There was an inverse relationship between the magnitude of initial thoracic kyphosis and the change in kyphosis following the application of therapeutic tape (r ¼ 0.70), (p ¼ 0.018); that is, individuals with greater thoracic kyphosis had the greatest reduction in thoracic kyphosis with the therapeutic tape. There was no change in normalised EMG following the application of therapeutic tape for any muscle (all pX0.139) (Table 3). Despite the absence of a change in EMG with tape, there was a main effect of surface on muscle activity of RA (po0.001) and L3 (po0.001). EMG activity of RA and L3 was increased on the short base compared with the ﬂat surface, eyes open condition (po0.001 and p ¼ 0.003, respectively), and ﬂat surface, eyes closed position (po0.001 and p ¼ 0.032, respectively). In addition, EMG activity of RA was not different between the ﬂat surface conditions (p ¼ 0.680), while EMG activity of L3 was greater on the ﬂat surface, eyes open compared with eyes closed conditions (p ¼ 0.002). Application of therapeutic tape had no effect on any of the balance parameters (all pX0.269) (Tables 4 and 5). However, most balance parameters differed between surface conditions. Although the position of the MeanCOPx did not change with application of therapeutic tape (p ¼ 0.635), there was a main effect for MeanCOPx for surface (po0.001). The MeanCOPx was more posterior when participants stood on the short base compared with standing on a ﬂat surface with the eyes open (po0.001) and eyes closed (po0.001). However, there was no difference between the ﬂat surface conditions (eyes open and eyes closed) (p ¼ 0.425).

4. Discussion The results of this study demonstrate that the application of postural therapeutic tape to the thoracic spine induces an immediate reduction in thoracic kyphosis in a population with osteoporotic vertebral fractures. However, this reduction in kyphosis is not associated with signiﬁcant changes in the activity of the trunk muscles or balance parameters. The reduction in thoracic kyphosis associated with the therapeutic tape may have been mediated by either passive support from the tape, active support from muscle contraction, or a combination of both. Given that there was no change in muscle activity associated with the reduction of thoracic kyphosis, this suggests that thoracic extension was achieved passively (i.e. mechanical support from the tape), as opposed to through the facilitation of back extensor muscles. However, it is also possible that changes in muscle activity occurred, but were not detected because of limitations in the EMG measurement. Even though differences in RA, L3 and UT muscle activity were detected across surface conditions, it is possible that EMG recordings may have lacked the sensitivity to detect small changes in muscle activity. Furthermore, individual participants may have adopted different muscle activation strategies in order to achieve thoracic extension, and this would have been obscured in the statistical comparisons. The application of therapeutic tape was associated with the greatest reduction in thoracic kyphosis in individuals with greater initial kyphosis. This ﬁnding suggests that despite advanced thoracic curvature, spinal stiffness in this population does not preclude improvement with postural taping. Taken with the absence of

Table 3 Statistical comparisons of mean (SD) of normalised EMG of trunk muscles across taping and surface conditions OI

OE

RA

ES L3

ES T8

UT

LT

Tape EOFB ECFB EOSB

0.79(0.15) 0.84(0.15) 0.85(0.14)

0.95(0.04) 0.95(0.04) 0.97(0.03)

0.76(0.27) 0.76(0.26) 0.89(0.25)

0.81(0.16) 0.87(0.12) 0.93(0.12)

0.78(0.15) 0.77(0.20) 0.73(0.18)

0.87(0.13) 0.77(0.25) 0.74(0.26)

0.86(0.12) 0.85(0.13) 0.84(0.17)

Control EOFB ECFB EOSB

0.89(0.13) 0.89(0.14) 0.82(0.26)

0.93(0.06) 0.93(0.06) 0.95(0.06)

0.79(0.13) 0.78(0.16) 0.94(0.09)

0.79(0.16) 0.80(0.14) 0.89(0.13)

0.80(0.16) 0.80(0.12) 0.79(0.13)

0.72(0.17) 0.69(0.15) 0.69(0.15)

0.76(0.23) 0.79(0.22) 0.83(0.25)

No tape EOFB ECFB EOSB

0.85(0.17) 0.79(0.17) 0.81(0.15)

0.96(0.04) 0.95(0.04) 0.95(0.05)

0.78(0.17) 0.78(0.14) 0.84(0.22)

0.79(0.15) 0.88(0.13) 0.92(0.07)

0.82(0.15) 0.78(0.22) 0.77(0.20)

0.79(0.16) 0.70(0.22) 0.64(0.21)

0.76(0.23) 0.79(0.22) 0.75(0.25)

Tape

p ¼ 0.402

p ¼ 0.264

p ¼ 0.658

p ¼ 0.373

p ¼ 0.668

p ¼ 0.139

p ¼ 0.237

Surface

p ¼ 0.877

p ¼ 0.123

po0.001*

po0.001*

p ¼ 0.569

p ¼ 0.062

p ¼ 0.983

*

Signiﬁcant difference

ARTICLE IN PRESS A.M. Greig et al. / Manual Therapy 13 (2008) 249–257

255

Table 4 Mean (SD) of force plate parameters (range and RMS) of across taping and surface conditions EOFB

ECFB

EOSB

Range

RMS

Range

RMS

Range

RMS

Taping COPx (cm) COPy (cm) COPxy (cm) COPvx (cm/s) COPvy (cm/s) COPfx (N) COPfy (N)

1.69(0.55) 0.99(0.36) 28.27(6.96) 8.47(2.52) 8.52(4.23) 3.91(1.30) 2.59(0.97)

0.37(0.14) 0.21(0.09) N/A 0.93(0.28) 0.74(0.25) 0.66(0.21) 0.36(0.12)

2.50(0.85) 1.34(0.66) 40.00(12.28) 12.14(4.71) 8.31(2.91) 6.56(2.28) 3.91(1.36)

0.49(0.19) 0.26(0.15) N/A 1.41(0.52) 0.89(0.37) 1.00(0.35) 0.53(0.18)

2.24(0.66) 1.28(0.40) 52.50(24.61) 17.17(9.65) 8.53(3.52) 7.65(3.53) 3.56(1.15)

0.41(0.14) 0.24(0.08) N/A 2.09(1.19) 0.86(0.29) 1.08(0.41) 0.50(0.16)

Control tape COPx (cm) COPy (cm) COPxy (cm) COPvx (cm/s) COPvy (cm/s) COPfx (N) COPfy (N)

1.78(0.77) 1.28(0.61) 29.75(11.10) 7.95(2.72) 9.37(4.54) 4.15(1.39) 3.19(1.04)

0.38(0.18) 0.25(0.14) N/A 0.90(0.27) 0.74(0.28) 0.65(0.23) 0.42(0.11)

2.33(0.98) 1.43(0.81) 39.86(9.23) 11.46(4.08) 9.25(4.74) 6.30(2.77) 3.52(1.52)

0.48(0.23) 0.27(0.17) N/A 1.36(0.38) 0.81(0.25) 1.00(0.42) 0.49(0.22)

2.18(0.73) 1.35(0.40) 48.94(17.21) 16.72(6.95) 7.49(2.43) 7.33(2.92) 3.71(1.10)

0.41(0.13) 0.26(0.09) N/A 1.95(0.83) 0.81(0.24) 1.03(0.38) 0.53(0.16)

No tape COPx (cm) COPy (cm) COPxy (cm) COPvx (cm/s) COPvy (cm/s) COPfx (N) COPfy (N)

1.75(0.48) 1.23(0.52) 31.26(9.30) 8.50(2.05) 7.79(2.33) 4.37(1.51) 3.33(1.37)

0.34(0.11) 0.24(0.13) N/A 0.93(0.18) 0.73(0.14) 0.65(0.24) 0.43(0.11)

2.46(1.04) 1.31(0.50) 41.58(10.46) 10.72(3.13) 8.74(3.80) 5.89(1.88) 3.36(0.92)

0.52(0.24) 0.25(0.11) N/A 1.29(0.42) 0.86(0.22) 0.97(0.36) 0.49(0.11)

2.28(0.57) 1.20(0.46) 55.19(18.44) 18.29(9.27) 8.32(2.98) 7.57(2.94) 2.65(1.55)

0.42(0.10) 0.24(0.11) N/A 2.10(0.91) 0.87(0.28) 1.11(0.39) 0.50(0.18)

Table 5 Statistical comparisons of force plate parameters (range and RMS) across taping and surface conditions Tape effect

COPx (cm) COPy (cm) COPxy (cm) COPvx (cm/s) COPvy (cm/s) COPfx (N) COPfy (N) MeanCOPx (cm) *

Surface effect

Range

RMS

Range

RMS

p ¼ 0.876 p ¼ 0.269 p ¼ 0.418 p ¼ 0.690 p ¼ 0.871 p ¼ 0.944 p ¼ 0.842 p ¼ 0.635

p ¼ 0.991 p ¼ 0.425

po0.001* p ¼ 0.160 po0.001* po0.001* p ¼ 0.561 po0.001* p ¼ 0.054 po0.001*

p ¼ 0.002* p ¼ 0.427

p ¼ 0.547 p ¼ 0.584 p ¼ 0.939 p ¼ 0.814

po0.001* p ¼ 0.030* po0.001* p ¼ 0.002*

Signiﬁcant difference

any associated changes in trunk muscle activity, these results suggest that thoracic kyphosis may be associated with impairment of active thoracic extension. Indeed, back extensor weakness has been found to be associated with greater thoracic kyphosis (Itoi and Sinaki, 1993), and changes in spinal posture have been shown to compromise back extensor strength (Mika et al., 2005). In turn, back extensor muscle weakness and increased fatigability, or ineffectual muscle force generation in the sub-optimal kyphosed position, may further contribute

to the kyphotic posture. For instance, the magnitude of trunk muscle force has been shown to be inﬂuenced by architectural and positional changes of trunk muscles, and the ability of the erector spinae to support anterior shear loads in the spine is thought to be compromised by alterations in length–tension relationships, moment arm lengths, and force vector orientations (Tveit et al., 1994; McGill et al., 2000). Thus, in a kyphosed posture, changes in the orientation of the erector spinae muscles, and alterations in generation of force may discourage active thoracic extension. Furthermore, research has shown that exercises that strengthen back extensor muscles result in a reduction in thoracic kyphosis (Itoi and Sinaki, 1994). Taken together, the results from the present study and previous ﬁndings, suggest that interventions that aim to improve back extensor strength, endurance, and muscle function in a more optimal posture may effectively reduce thoracic kyphosis. This reduction in thoracic kyphosis, in turn, may have beneﬁcial effects on spinal loading and risk of vertebral fracture. In the present study, the reduction in thoracic kyphosis was not associated with any changes in the balance parameters. We hypothesised that a change in thoracic kyphosis may shift the COM posteriorly, and would affect the COP displacement and shear forces. This hypothesis was based on previous research that

ARTICLE IN PRESS 256

A.M. Greig et al. / Manual Therapy 13 (2008) 249–257

found an association between increased thoracic kyphosis and balance impairments (Lynn et al., 1997; Cook, 2002; Balzini et al., 2003; Sinaki et al., 2005). The present data indicate that although COP was shifted posteriorly when standing on the short base compared with ﬂat surface conditions, there was no effect of the taping condition on the mean AP COP position despite the changes in kyphosis angle. Thus, the reduction in thoracic kyphosis was either not sufﬁcient to induce an observable change in balance parameters, or individuals may have compensated for the immediate change in COM displacement. Spinal orthoses have been shown to induce changes, not only in posture, but also in muscle strength and balance. Combined interventions of spinal orthoses and exercises have been shown to improve balance in women with osteoporosis and increased thoracic kyphosis (Sinaki and Lynn, 2002), and to improve gait and falls risk (Sinaki et al., 2005) after 1 month. However, as previous studies have often combined both orthoses and exercise, it is not clear whether positive outcomes were associated with the spinal orthoses or the exercise components. In addition, the interventions in the spinal orthoses studies were conducted for periods of a month or more, which limits the comparison with the present study measuring immediate outcomes. However, the results from the present study suggest that interventions that apply mechanical support (i.e. tape) may be effective in reducing thoracic kyphosis, but further clinical trials are needed to explore the short and longterm effects of postural interventions, as well as the independent effects of exercise. In addition, larger studies with greater power may indeed demonstrate relationships between the change in kyphosis achieved with postural tape and changes in trunk muscle activity and balance measures.

5. Conclusions The results of this study demonstrate that the application of therapeutic postural tape induces an immediate reduction in thoracic kyphosis in individuals with osteoporotic vertebral fracture. However, the reduction in thoracic kyphosis is not associated with any changes in trunk muscle activity or balance during a quiet standing task. Future studies may consider investigating the underlying mechanisms associated with the reduction in thoracic kyphosis from postural taping, with an emphasis on potential active and passive mechanisms.

References Alexeeva L, Burckhardt P, Christiansen C, Cooper C, Delmas P, Johnell O, et al. Assessment of fracture risk and its application to

screening for postmenopausal osteoporosis. Report of a World Health Organization Study Group 1994;843. Balzini L, Vannucchi L, Benvenuti F, Benucci M, Monni M, Capozzo A, et al. Clinical characteristics of ﬂexed posture in elderly women. Journal of the American Geriatrics Society 2003;51:1419–26. Basmajian JV, DeLuca CJ. Muscles alive: their functions revealed by electromyography, 5th ed. Baltimore: Williams and Wilkins; 1985. Bennell K, Khan KM, McKay HA. The role of physiotherapy in the prevention and treatment of osteoporosis. Manual Therapy 2000;5(4):198–213. Briggs AM, Dieen JHv, Wrigley TW, Greig AM, Phillips B, Lo SK, et al. Thoracic kyphosis affects spinal loads and trunk muscle force. Physical Therapy 2007;87(5):548–53. Cook C. The relationship between posture and balance disturbances in women with osteoporosis. Physical and Occupational Therapy in Geriatrics 2002;20:37–49. Cortet B, Roches E, Logier R, Houvenagel E, Gaydier-Souquieres G, Puisieux F, et al. Evaluation of spinal curvatures after a recent osteoporotic vertebral fracture. Joint, Bone, Spine: Revue du Rhumatisme 2002;69(2):201–8. Diee¨n JHv, Cholewicki J, Radebold A. Trunk muscle recruitment patterns in patients with low back pain enhance the stability of the lumbar spine. Spine 2003;28:834–41. Ensrud KE, Black DM, Harris F, Ettinger B, Cummings SR. Correlates of kyphosis in older women. Journal of the American Geriatrics Society 1997;45(6):682–7. Genant HK, Jergas M, Palermo L, Nevitt M, SanValentin R, Black D, et al. Comparison of semiquantitative visual and quantitative morphometric assessment of prevalent and incident vertebral fractures in osteoporosis. Journal of Bone and Mineral Research 1996;11(7):984–96. Greendale GA, Barrett-Connor E, Ingles S, Haile R. Late physical and functional effects of osteoporotic fracture in women: The Rancho Bernardo Study. Journal of the American Geriatrics Society 1995;43:955–61. Greig AM. Relationships between vertebral fracture , thoracic kyphosis and postural control in individuals with osteoporosis. PhD thesis. Melbourne: University of Melbourne; 2006. Greig AM, Bennell KL, Briggs AM, Wark JD, Hodges PW. Balance impairment is related to vertebral fracture rather than thoracic kyphosis in individuals with osteoporosis. Osteoporosis International 2006, doi:10.1007/s00198-006-0277-g. Hakkinen K, Komi PV. Alterations of mechanical characteristics of human skeletal muscle during strength training. European Journal of Applied Physiology 1983;50:161–72. Hall SE, Criddle RA, Comito TL, Prince RL. A case-control study of quality of life and functional impairment in women with longstanding vertebral osteoporotic fracture. Osteoporosis International 1999;9:508–15. Hochberg Y, Benjamini Y. More powerful procedures for multiple signiﬁcance testing. Statistics in Medicine 1990;9(7):811–8. Horak FB, Nashner LM. Central programming of postural movements: adaptation to altered support-surface conﬁgurations. Journal of Neurophysiology 1986;55:1369–81. Horak FB, Diener HC, Nashner LM. Inﬂuence of central set on human performance. Journal of Neurophysiology 1989;62(4):841–53. Huang C, Ross P, Wasnich RD. Vertebral fracture and other predictors of physical impairment and health care utilization. Archives of Internal Medicine 1996;156:2469–75. Itoi E, Sinaki M. Thoracic kyphosis and back extensor strength in postmenopausal women. In: Paper presented at the 4th international symposium on osteoporosis and consensus development conference, Hong Kong, 1993. Itoi E, Sinaki M. Effect of back-strengthening exercise on posture in healthy women 49–65 years of age. Mayo Clinic Proceedings 1994;69:1054–9.

ARTICLE IN PRESS A.M. Greig et al. / Manual Therapy 13 (2008) 249–257 Kabada MP, Wootten ME, Gainey J, Cochran GVB. Repeatability of phasic muscle activity: performance of surface and intramuscular wire electrodes in gait analysis. Journal of Orthopaedic Research 1985;3:350–9. Keller TS, Harrison DE, Colloca CJ, Harrison DD, Janik TJ. Prediction of osteoporotic spinal deformity. Spine 2003;28(5): 455–62. Livingston R, Livingston T. Dualer electronic inclinometer: instruction manual. American Force: Jtech; 1992. Lynn SG, Sinaki M, Westerlind KC. Balance characteristics of persons with osteoporosis. Archives of Physical Medicine and Rehabilitation 1997;78(3):273–7. Marras WS, Davis KG, Ferguson SA, Lucas BR, Gupta P. Spine loading characteristics of patients with low back pain compared with asymptomatic individuals. Spine 2001;23:2566–74. McCloskey EV, Spector TD, Eyres KS, Fern ED, Orourke N, Vasikaran S, et al. The assessment of vertebral deformity: a method for use in population studies and clinical trials. Osteoporosis International 1993;3(3):138–47. McGill SM, Hughson RL, Parks K. Changes in lumbar lordosis modify the role of the extensor muscles. Clinical Biomechanics 2000;15:777–80. Mellin G. Measurement of thoracolumbar posture and mobility with a Myrin inclinometer. Spine 1986;11(7):759–62. Mika A, Unnithan VB, Mika P. Differences in thoracic kyphosis and in back muscle strength in women with bone loss due to osteoporosis. Spine 2005;30(2):241–6. O’Brien K, Culham E, Pickles B. Balance and skeletal alignment in a group of elderly female fallers and nonfallers. The Journals of Gerontology: Biological Sciences 1997;52(4):B221–6. Ng JK-F, Kippers V, Richardson CA. Muscle ﬁbre orientation of abdominal muscles and suggested surface EMG electrode positions. Electromyography and Clinical Neurophysiology 1998;38:51–8. Ng JK, Kippers V, Richardson CA, Parnianpour M. Range of motion and lordosis of the lumbar spine—reliability of measurement and normative values. Spine 2001;26(1):53–60.

257

Pfeifer M, Begerow B, Minne HW. Effects of a new spinal orthosis on posture, trunk strength, and quality of life in women with postmenopausal osteoporosis. A randomized trial. American Journal of Physical Medicine and Rehabilitation 2004;83(3): 177–86. Saur PM, Ensink FB, Frese K, Seeger D, Hildebrandt J. Lumbar range of motion: reliability and validity of the inclinometer technique in the clinical measurement of trunk ﬂexibility. Spine 1996;21(11):1332–8. Schultz A, Andersson GB, Ortengren R, Bjork R, Nordin M. Analysis and quantitative myoelectric measurements of loads on the lumbar spine when holding weights in standing postures. Spine 1982;7(4):390–7. Sinaki M, Lynn SG. Reducing the risk of falls through proprioceptive dynamic posture training in osteoporotic women with kyphotic posturing: a randomized pilot study. American Journal of Physical Medicine and Rehabilitation 2002;81(4):241–6. Sinaki M, Brey R, Hughes CA, Larson DR, Kaufman KR. Signiﬁcant reduction in risk of falls and back pain in osteoporotic-kyphotic women through a spinal proprioceptive extension exercise dynamic (SPEED) program. Mayo Clinic Proceedings 2005;80(7):849–55. Tveit P, Daggfeldt K, Hetland S, Thorstensson A. Erector spinae lever arm length variations with changes in spinal curvature. Spine 1994;19(2):199–204. Viitasalo JHT, Komi PV. Signal characteristics of EMG with special reference to reproducibility of measurements. Acta Physiologica Scandinavia 1975;93:531–9. Wang C, McClure P, Pratt NE, Nobilini R. Stretching and strengthening exercises: their effect on three-dimensional scapular kinematics. Archives of Physical Medicine and Rehabilitation 1999;80:923–9. Yang JF, Winter DA. Electromyography: reliability in maximal and submaximal isometric contractions. Archives of Physical Medicine and Rehabilitation 1983;64:417–20. Yuan HA, Brown CW, Phillips FM. Osteoporotic spinal deformity— a biomechanical rationale for the clinical consequences and treatment of vertebral body compression fractures. Journal of Spinal Disorders and Techniques 2004;17(3):236–42.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 258–265 www.elsevier.com/locate/math

Original article

MRI study of the cross-sectional area for the cervical extensor musculature in patients with persistent whiplash associated disorders (WAD) James Elliotta,b,c,, Gwendolen Julla, Jon Timothy Noteboomb, Graham Gallowayc a

Division of Physiotherapy, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia b Department of Physical Therapy, Rueckert-Hartman School for Health Professions, Regis University, Denver, CO, USA c Centre for Magnetic Resonance, The University of Queensland, Brisbane, Australia Received 20 August 2006; received in revised form 18 December 2006; accepted 2 January 2007

Abstract Cervical muscle function is disturbed in patients with persistent pain related to a whiplash associated disorder (WAD) but little is known about neck extensor muscle morphometry in this group. This study used magnetic resonance imaging to measure relative cross-sectional area (rCSA) of the rectus capitis posterior minor and major, multiﬁdus, semispinalis cervicis and capitis, splenius capitis and upper trapezius muscles bilaterally at each cervical segment. In total, 113 female subjects (79 WAD, 34 healthy control; 18–45 years, 3 months–3 years post-injury) were recruited for the study. Signiﬁcant main effects for differences in muscle and segmental level were found between the two groups (Po0.0001) as well as a signiﬁcant group * muscle * level interaction (Po0.0001). The cervical multiﬁdus muscle in the WAD group had signiﬁcantly larger rCSA at all spinal levels and in contrast, there were variable differences in rCSA measures across levels in the intermediate and superﬁcial extensor muscles when compared to the healthy controls (Po0.0001). There were occasional weak, although statistically signiﬁcant relationships between age, body mass index (BMI), duration of symptoms and the size of some muscles in both healthy control and WAD subjects (Po0.01). It is possible that the consistent pattern of larger rCSA in multiﬁdus at all levels and the variable pattern of rCSA values in the intermediate and superﬁcial muscles in patients with WAD may reﬂect morphometric change due to fatty inﬁltrate in the WAD muscles. Future clinical studies are required to investigate the relationships between muscular morphometry, symptoms and function in patients with persistent WAD. r 2007 Elsevier Ltd. All rights reserved. Keywords: MRI; Cervical; Muscle; CSA; Whiplash

1. Introduction Morphometric alterations are common in paraspinal muscles in patients with low back pain (Hides et al., 1994; Kader et al., 2000) and there are qualitative data supporting similar changes in the cervical paraspinal Corresponding author. Regis University, Rueckert-Hartman School for Health Professions, Department of Physical Therapy, 3333 Regis Blvd, G-4, Denver, CO 80221-1099, USA. Tel.: +1 303 458 4022; fax +1 303 964 5474. E-mail address: [email protected] (J. Elliott).

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.01.012

muscles of patients with chronic neck pain (McPartland et al., 1997; Kristjansson, 2004). Quantitative measurements of paraspinal muscle size can be obtained with both real-time ultrasonography and magnetic resonance imaging (MRI) and there is growing support for their use in investigations of patients with spinal pain (Kristjansson, 2004; Elliott et al., 2005; Rankin et al., 2005). While these studies have offered valuable reference data for measurable differences in paraspinal muscle size in both asymptomatic and symptomatic subjects, the value of diagnostic muscular imaging in patients with traumatic neck pain (whiplash associated

ARTICLE IN PRESS J. Elliott et al. / Manual Therapy 13 (2008) 258–265

disorders (WAD)) has yet to be determined. Since MRI can be regarded as the gold-standard for muscle imaging, it can be used to accurately determine if muscle size in patients with WAD varies signiﬁcantly from normative data. Such knowledge may assist the understanding of muscle changes in neck disorders as well as serve as an outcome measure in rehabilitation. The purpose of this cross-sectional study was to (1) measure the cross-sectional area (CSA) of cervical extensor muscles, (2) determine whether there was a difference in CSA between WAD and healthy control subjects within a deﬁned age-range (18–45 years), and (3) determine if age, neck disability index (NDI) scores, duration of symptoms and body mass index (BMI) had an inﬂuence on extensor muscle size. It was hypothesized that CSA alterations in the cervical extensor musculature would be greater in patients with persistent WAD than healthy control subjects. 2. Methods 2.1. Subjects Volunteer subjects with persistent WAD were gained upon referral from physical therapy and medical centers in both Denver (USA) and Brisbane (Australia). Healthy control subjects were recruited through university advertisement campaigns. Whiplash subjects were included provided they suffered from persistent neck pain and disability (duration from 3 months to 3 years) resulting from a motor vehicle crash (MVC) and were classiﬁed as Grade II per the Quebec Task Force Classiﬁcation (Spitzer et al., 1995). The cohort was limited to females (18–45 years) as this group best represents the predominant population of patients suffering from persistent whiplash related pain (Larsen and Holm, 2000). Volunteers were excluded when (1) classiﬁed as WAD I, III or IV, (2) suffered one or more MVC or (3) reported receiving treatment for neck pain within the past 10 years. Subjects were also excluded if they had been previously diagnosed with any central or peripheral nervous system disorder or were considered unsuitable for an MRI scan e.g. pregnancy, possible pregnancy, claustrophobia or presented with metallic objects in the body that may be moved when introduced to the magnetic ﬁeld. A total of 136 (101 WAD and 35 healthy controls) volunteers were originally considered. However, 22 potential WAD participants were excluded for the following reasons: older than 45 years (1); pregnancy (1); history of previous MVCs (16); refusal to be placed into the MR magnet due to claustrophobia (2) and MVC-related loss of consciousness (2). One healthy control subject was excluded following reports that she had suffered from intermittent neck pain with symptoms of vague dizziness for the past 5 years. Thus, the study

259

population consisted of 79 WAD subjects (78 Americans and 1 Australian) and 34 healthy controls (16 Americans and 18 Australians). Ethical approval was granted by local and Institutional Ethics Committees. Written informed consent was obtained from all subjects prior to their inclusion into the study. 2.2. Procedure of MRI acquisition In the USA, all subjects were scanned using a conventional spin-echo pulse sequence (656 ms TR and 14 ms TE) with a Horizon LX General Electric 1.5 T scanner (Milwaukee, WI, USA). In Australia a SONATA 1.5 T magnet (448 ms TR and 14 ms TE) (Siemens, Erlangen, Germany) was used using the same measurement methodology. There was no systematic difference found between images acquired in the USA and those acquired in Australia. Axial images of the cervical spine were obtained from the mid point of the cerebellum through to the T1 segmental level, thus ensuring proper capture of the paraspinal musculature. Fig. 1 illustrates the cervical extensor musculature on T1-weighted axial MRI scan at the C6 segmental level (healthy control). The suboccipital muscles (rectus capitis posterior minor (RCPmin) and major (RCPmaj)) were measured at the C1 and C2 segments and the cervical extensor muscles (multiﬁdus, semispinalis cervicis and capitis, splenius capitis and upper trapezius) were measured at each segment (C3-7). T1-weighted MR parameters were chosen in order to provide images of reasonable tissue contrast between fat and skeletal muscle (Murphy et al., 1986). 2.3. Procedure of image analysis The images were analyzed post hoc with MRIcro software (www.mricro.com) (Rorden and Brett, 2000) and securely stored on a laptop computer. Analysis was accomplished by manually tracing deﬁned regions of interest (ROI) within the fascial borders of the suboccipital and cervical extensor muscles bilaterally on the axial T1-weighted images. The most cephalad portion of each vertebral body (C3–C7) was the landmark used for measurement of the cervical extensors, multiﬁdus, semispinalis cervicis, capitis, splenius capitis and upper trapezius at each level. The axial MR slices were positioned parallel to the C2/3 intervertebral disk, which produced a slight measurement error for the CSA measures of muscles above and below. This was consistent for the muscles bilaterally at each vertebral segment and between each subject. As a result, relative CSAs (rCSA) are reported. The rCSA of cervical extensor musculature were calculated by the number of pixels under each ROI in the x and y axes (mm mm ¼ mm2) with MRIcro software. The rCSA measures for the RCPmin muscle were measured after

ARTICLE IN PRESS 260

J. Elliott et al. / Manual Therapy 13 (2008) 258–265

Fig. 1.

identifying the odontoid process and occipital condyles at the C1 segmental level on the MR axial scan. The rCSA measures for the RCPmaj muscle were measured from the most cephalad portion of the C2 segmental level on the MR axial scan. The rCSA measures for the bilateral segmental extensor muscles (C3–C7) were measured after identifying the individual muscle at the most cephalad portion of each particular vertebral level on the MR axial scan. This process ensured consistent representation of rCSA measures for each muscle at each level between the subjects and has been previously reported in detail (Elliott et al., 2005). The examiner (JE) was not blinded to the status of the subjects. To determine the ﬁdelity of the measures (as well as inter-examiner reliability), a second researcher undertook the rCSA measures for the right side multiﬁdus, semispinalis cervicis, capitis, and splenius capitis muscles at the 5 segments (C3–7 levels) on 5 subjects (100 measures). 2.4. Data analysis Data were analyzed with SPSS 13.0. Intraclass correlational coefﬁcients (ICC) were performed to examine inter-examiner reliability and the ICCs and standard error of the measure (SEM) were calculated for each muscle at each segmental level. Analysis of the main data was performed through several steps. In justifying the assumption of common distribution for the two groups, a Q–Q plot was graphed and revealed that the distribution of errors for the rCSA measures was skewed positively. As a result, raw data values (mm2) for rCSA were transformed and analyzed as

logged values. A multi-factorial repeated measures analysis of variance was used to investigate whether there were any differences between the whiplash and control groups in measures of rCSA of the cervical extensor muscles across muscle, side and segmental levels. Group differences between WAD and control subjects for rCSA measures in the RCPmin and RCPmaj muscles were compared using independent sample t-tests. The Spearman correlations were also performed to determine if the score on the NDI (Vernon and Mior, 1991) age, BMI and length of history from date of injury had an inﬂuence on muscle size. Signiﬁcance was set conservatively at Po0.01. 3. Results Table 1 presents the demographic characteristics for the two groups. The WAD group were marginally (albeit signiﬁcantly) older and had a higher BMI than the control group, although there were no signiﬁcant differences in height and weight between groups. In the reliability study, the measures of muscle rCSA ranged from 74.7–215.1 mm2. The ICCs for the interrater agreement of the rCSA measures of the right sided cervical extensors at the C3–C7 segmental levels ranged from 0.96 to 0.99 (SEM 0.96–4.87), indicating the measure to have acceptable accuracy. 3.1. rCSA of cervical extensor muscles The ANOVA revealed signiﬁcant main effects for muscle and segmental level (Po0.0001) with a signiﬁcant interaction for group muscle level, indicating

ARTICLE IN PRESS J. Elliott et al. / Manual Therapy 13 (2008) 258–265

(semispinalis cervicis, semispinalis capitis, splenius capitis and upper trapezius) yielded variable differences with regards to muscle size. Speciﬁcally, the rCSA measures were smaller in the WAD subjects for the semispinalis cervicis and the difference was signiﬁcant for C3 (P ¼ 0.005), C5 and C6 levels (Po0.001). A mixed pattern of muscle size was noted for the semispinalis capitis and splenius capitis muscles. In particular, rCSA values were signiﬁcantly larger in the WAD group for semispinalis capitis at C3 (Po0.001); and splenius capitis at the C3 level (Po0.001). However, at the C6 level the semispinalis capitis was larger in the healthy control group (P ¼ 0.001). No signiﬁcant differences were noted in rCSA of the upper trapezii musculature between the two groups (P ¼ 0.5678). Appendix A presents raw data for rCSA values (mm2) of the cervical extensor musculature side level group. Appendix B presents raw data for rCSA values (mm2) of the sub-occipital musculature side group.

that differences in rCSA between the two groups varied by muscle and level (Po0.0001). There was no signiﬁcant interaction between group side muscle and as a result, we have averaged the rCSA values for right and left sides for post hoc analysis. Fig. 2 presents the logged rCSA measures for each muscle of the cervical extensor group at each cervical level (C3–C7) for the WAD and control groups. Post hoc analysis revealed that there was a signiﬁcantly larger rCSA in the multiﬁdii at all levels (C3–C7) in the WAD compared to control subjects (Po0.0001). In contrast, the other intermediate to superﬁcial musculature

Table 1 Demographic details of whiplash and control subjects WAD (n ¼ 79)

Age (years) Height (m) Weight (kg) BMI (kg/m2) Duration of symptoms (months) NDI

Mean (SD)

Healthy controls (n ¼ 34) Mean (SD)

29.7 1.66 68.9 25.1 20.3

27.0 1.67 64.1 23.0 —

(7.7) (0.07) (15.6) (5.73) (9.55)

45.5 (15.9)

261

(5.6)* (0.06) (14.3) (4.44)*

3.2. rCSA of the subocippital musculature The independent samples t-test revealed no signiﬁcant between-group differences in rCSA (averaged between sides) for the RCPmin (P ¼ 0.07) and for the RCPmaj (P ¼ 0.02). The general trend was for larger rCSA in the WAD RCPmin and RCPmaj musculature.

—

NDI ¼ neck disability index (Vernon and Mior, 1991). *Po0.05.

Fig. 2.

ARTICLE IN PRESS 262

J. Elliott et al. / Manual Therapy 13 (2008) 258–265

3.3. rCSA and NDI, duration of symptoms, BMI and age The Spearman’s r correlation analyses indicated that there were no signiﬁcant associations between the rCSA measures in the WAD group extensor muscles (averaged between sides) and NDI scores or duration of symptoms. However, there were weak (albeit signiﬁcant) positive correlations between the rCSA measures in some of the extensor muscles for both groups and BMI scores (r2 range from 0.06 (semispinalis capitis—C7) to 0.14 (splenius capitis—C3)) and for age (r2 ¼ 0.05 (semispinalis cervicis—C5)) (Po0.01). Even though the correlations differed signiﬁcantly from zero, the amount of variation captured in the associations of age and BMI was very small and only explained between 5% and 14% of the variance (age: semispinalis cervicis—C5; BMI: splenius capitis—C3, respectively). 4. Discussion This study has identiﬁed changes in the MRI measure for rCSA in the cervical extensor musculature in patients with persistent WAD compared to healthy control subjects. Speciﬁcally, signiﬁcantly larger rCSAs were identiﬁed in the deep multiﬁdus muscle at the C3–C7 levels in the WAD group and a varying pattern in rCSA was found in the more intermediate musculature. Smaller rCSAs were found in the WAD group for the semispinalis cervicis (C3, C5 C6) while larger rCSAs were found in the more superﬁcial muscles; semispinalis capitis (C3) and splenius capitis (C3). The semispinalis capitis had a greater rCSA in the healthy control group at the C6 level only. No signiﬁcant differences were found in the upper trapezius or in the suboccipital muscles (RCPmin and RCPmaj). There was a weak correlation between rCSA, age and BMI (both groups) for some muscles. However, the muscle changes were independent of self-reported levels of pain and disability (NDI) and duration of symptoms in the WAD subjects. NDI scores (/100) in the WAD group ranged from 16 to 82 (mean 45.5715.9) indicating self-reported levels of pain and disability in this whiplash group ranged from mild to severe. 4.1. Increased rCSA in multifidus A consistent pattern of larger rCSA was observed in the WAD group in the multiﬁdii muscle at each segment (C3–7) and this contrasted to a less-consistent pattern in rCSA measures in the other more intermediate and superﬁcial muscles; semispinalis cervicis and capitis and splenius capitus. It is possible that fatty inﬁltrate is responsible for these ﬁndings in the multiﬁdus. We have previously established that the WAD subjects from this study also demonstrate widespread fatty inﬁltrate in all cervical extensor muscles (Elliott et al., 2006). However,

the highest fat indices were consistently found in the deep multiﬁdii at all segmental levels (C3–C7) and lesser fatty inﬁltrate was present in the other paraspinal muscles. This may help to explain the reversed pattern of larger rCSA in the multifdii musculature compared to the other muscles in this study. Higher fat content is likely to alter and expand the musculofascial borders thereby creating larger rCSA values with MRI measures. This suggests that measures of rCSA of the cervical extensor musculature have to be interpreted with caution at least in persons with persistent WAD. Increased fat content has also been found in the lumbar paraspinal muscles in patients with persistent low back pain (Alaranta et al., 1993; Kader et al., 2000) and observed in the suboccipital muscles in patients suffering from persistent neck pain (Hallgren et al., 1994; McPartland et al., 1997). It is possible that the measures of MR fatty inﬁltration may be more sensitive markers than CSA for changes in these muscles in patients with persistent WAD, or both measures should be taken concurrently. The involvement of multiﬁdus at all segments in the cervical region in this chronic whiplash group is notable as multiﬁdus has substantial muscle insertions into the cervical facet capsule (Winkelstein et al., 2001). Numerous reports have implicated the facet joint in neck pain and injury (Lord et al., 1996; Panjabi et al., 1998; Winkelstein et al., 2000; Siegmund et al., 2001). Anderson et al. (2002) have further substantiated that the cervical multiﬁdus is architecturally complex with deep and superﬁcial fascicular subgroups attaching directly to the facet capsules. Based on the location of the cervical multiﬁdus, any morphometrical changes may compromise its ability to play a role in controlling cervical segmental motion. These muscular changes may therefore be a contributing factor to ongoing neck pain and disability. 4.2. Altered rCSA in the other neck extensor muscles The consistent ﬁnding of reduced rCSA in the semispinalis cervicis may reﬂect general wasting, without the large volumetric inﬂuence of higher levels of fatty inﬁltrate as was noted to occur in the deep WAD multiﬁdus. The changes in the semispinalis capitis and splenius capitis were not as marked through the spine with the exception of the C3 level where paradoxically the rCSAs were larger in the WAD group. This cannot be explained by fatty inﬁltrate as these muscles were found to have less fat within muscle than the semispinalis cervicis (Elliott et al., 2006). The increased rCSA in semispinalis capitis and splenius capitis at the C3 level could represent a function adaptation in response to the changes in multiﬁdus and semspinalis cervicis at this level. Nevertheless, at the C6 level, the rCSA of semispinalis capitis was signiﬁcantly smaller in the WAD group, which might question this proposal. Further research is required to replicate this ﬁnding

ARTICLE IN PRESS J. Elliott et al. / Manual Therapy 13 (2008) 258–265

before more deﬁnitive explanations can be provided. No signiﬁcant changes in rCSA were observed in the upper trapezii musculature; a muscle that largely inﬂuences the shoulder girdle and upper limb mechanics rather than the neck (Johnson et al., 1994). It is known that a whiplash injury can damage any number of structures at any segment in the cervical spine (Grauer et al., 1997; Panjabi et al., 1998; Pearson et al., 2004). Nevertheless, Lord et al. (1996) examined the prevalence of cervical zygapophysial joint pain in patients with persistent WAD using diagnostic anesthetic blocks and found the highest incidence at the C2–3 and C5–6 cervical zygapophysial joints. While we found signiﬁcant rCSA differences existed throughout the cervical spine segments, the C3 level featured signiﬁcant rCSA changes in all four of the cervical muscles measured and the C6 level had changes in three of the four muscles (multiﬁdus, semispinalis cervicis and capitis). The ﬁndings of most altered muscular morphometry at the C3 (C2–3) and C6 (C5–6) levels in our WAD group are consistent with those of Lord et al. (1996). However, the level of symptoms was not determined clinically in this study. Future research is warranted to determine the relationship of segmental muscle size and symptoms.

263

included for consideration in future research of muscle measurements in the cervical region. The correlation analysis investigating the association between duration of symptoms and rCSA in the WAD subjects did not yield signiﬁcant ﬁndings. Our ﬁndings of muscle size changes indicate that they were established by 3 months following injury and it is currently unknown how quickly these changes occur after injury. Further longitudinal study involving acute WAD subjects is warranted to better answer that question. Regardless of the prevailing mechanisms for the observed muscle changes in this study, these morphometrical alterations in the deep muscles may provide valuable insight into the common functional impairments observed in patients with persistent WAD. The deep cervical muscles are predominantly comprised of Type I ﬁbers (Boyd-Clark et al., 2002) and have high muscle spindle density (Liu et al., 2003) suggestive of muscle-speciﬁc differences in the precise control of movement and proprioceptive function (Bakker et al., 1984; Edney and Porter, 1986; Porter, 1986). Disturbances in cervical kinesthetic sense (Treleaven et al., 2005) are also common in patients suffering from persistent WAD and these kinesthetic deﬁcits may relate to abnormal cervical afferent input due to alterations within the deep musculature.

4.3. rCSA in the sub-occipital muscles 5. Conclusion We found no signiﬁcant differences in rCSA in the suboccipital musculature. This was surprising as we have previously reported higher amounts of fatty inﬁltration in these muscles in those subjects suffering from persistent WAD (Elliott et al., 2006). Our proposal for the multiﬁdus muscle, that MRI markers of fat inﬁltration provide a more sensitive measure than CSA may thus apply to the analysis of sub-occipital muscle morphology of patients with possible upper cervical injury. 4.4. rCSA and NDI, BMI, age and duration of symptoms. The ﬁnding that rCSA changes in the WAD group were not associated with NDI scores was unexpected as injury severity is usually related to higher levels of pain and disability (Gargan and Bannister, 1990). BMI and age appeared to impact on some of the rCSA measures for the cervical extensor muscles in both the control and WAD subjects (18–45 years) and this was surprising as age-related reductions in lean body mass are reported not to commence until the ﬁfth or sixth decade of life (Forbes and Reina, 1970). However, future research is required to better understand the potential cultural inﬂuences, if any, on muscle size and BMI. Nevertheless, the relationships were weak and without a consistent pattern of muscle involvement. While the clinical relevance of these ﬁndings is not clear, it does indicate BMI and age should be

This is the ﬁrst MRI study to demonstrate that female patients (18–45 years) with persistent WAD (grade II) show quantiﬁable alterations in rCSA of the cervical paraspinal musculature that differ signiﬁcantly from subjects with no history of neck pain. In particular, this study demonstrated larger rCSA in all of the segmental levels (C3–C7) of the deep multiﬁdus muscle in the WAD group which is different from the pattern of variable segmental alterations in rCSA found in the other extensor muscles. The origin behind these ﬁndings is not completely understood and further work is warranted to investigate potential mechanisms contributing to these differences and their relevance to function and symptomatology. Finally, the muscle changes found in this study relate to persistent WAD. It is unknown if such changes are associated with chronic neck pain of idiopathic or a nontraumatic origin and this is to be investigated.

Acknowledgments We wish to sincerely thank Dr. Ross Darnell, Statistician; John Finnesy and Dorinda Gill, Radiology Technicians; Bao Nguyen, MD, Radiologist, Kathy Francis and Candice Valdez, research assistants, for their assistance with conducting this study.

ARTICLE IN PRESS J. Elliott et al. / Manual Therapy 13 (2008) 258–265

264

Appendix A The raw data for rCSA values (mm2) of the cervical extensor musculature side level group are shown in Table A1. Table A.1 rCSA values (mm2) of the cervical extensor musculature * side * level * group Group

WAD C3 C4 C5 C6 C7 Normal C3 C4 C5 C6 C7

Multiﬁdus

Semispinalis cervicis

Semispinalis capitis

Splenius capitis

Upper trapezius

R

L

R

L

R

L

R

L

R

L

73.9 (20.6) 103.4 (21.4) 130.3 (30.1) 148.1 (32.9) 125.0 (31.6)

70.8 (20.6) 96.3 (28.2) 123.4 (37.1) 131.8 (40.1) 121.6 (43.8)

113.9 (24.2) 150.0 (26.2) 153.9 (28.0) 156.8 (28.6) 165.0 (29.1)

110.8 (25.8) 143.6 (26.3) 150.7 (30.9) 152.2 (31.2) 156.2 (29.3)

215.9 (46.6) 196.9 (37.8) 166.0 (34.9) 131.1 (24.9) 119.7 (22.4)

220.5 (55.3) 201.0 (43.6) 168.2 (41.6) 130.6 (30.9) 105.6 (18.6)

180.0 (35.1) 175.3 (39.7) 174.7 (32.1) 178.8 (27.7) 186.8 (32.2)

186.2 (38.5) 181.2 (39.3) 180.2 (40.0) 184.7 (32.8) 185.7 (29.8)

80.1 (18.6) 107.5 (55.0) 215.9 (210.1) 538.6 (397.0) 907.4 (361.3)

87.1 (24.7) 113.8 (69.0) 251.2 (323.4) 610.4 (460.2) 1017.9 (390.5)

63.4 (10.6) 70.7 (22.1) 87.2 (30.8) 118.7 (48.4) 101.7 (28.3)

61.2 (10.2) 66.4 (21.3) 83.0 (31.9) 115.4 (47.9) 98.7 (27.2)

128.6 (37.4) 156.4 (31.1) 175.9 (32.9) 177.3 (27.2) 161.3 (20.7)

127.2 (36.4) 153.2 (29.5) 173.0 (30.1) 174.4 (30.1) 155.7 (22.3)

176.1 (36.9) 182.7 (29.2) 175.9 (27.8) 152.1 (24.8) 113.0 (28.2)

176.7 (40.4) 178.9 (32.3) 175.8 (29.8) 143.5 (23.9) 110.1 (27.4)

156.8 (30.9) 162.2 (31.7) 178.2 (30.3) 192.7 (35.1) 182.3 (24.1)

155.7 (35.7) 164.4 (30.9) 178.2 (31.9) 187.7 (33.3) 181.0 (26.0)

79.2 (29.1) 115.2 (54.9) 296.3 (377.6) 679.5 (491.2) 1106.2 (458.9)

75.3 (29.7) 109.8 (56.6) 296.9 (359.1) 687.2 (452.9) 1165.0 (358.3)

Data reported as mean (SD) (muscle size is reported as relative cross-sectional area (mm2)).

Appendix B The raw data for rCSA values (mm2) of the sub-occipital musculature side group are shown in Table B1. Table B.1 rCSA values (mm2) of the sub-occipital musculature * side * group Group

Rectus capitis posterior minor

Rectus capitis posterior major

R

L

R

L

117.0 (29.7)

102.6 (32.7)

140.2 (32.6)

122.1 (37.7)

106.1 (29.6)

105.3 (34.0)

123.8 (43.7)

122.3 (45.9)

WAD

Normal

Data reported as mean (SD) (muscle size is reported as relative crosssectional area (mm2)).

References Alaranta H, Tallroth K, Soukka A, Heliovaara M. Fat content of lumbar extensor muscles and low back disability: a radiographic and clinical comparison. Journal of Spinal Disorder 1993;6(2):137–40.

Anderson JS, Hsu AW, Vasavada AN. Morphology, architecture, and biomechanics of human cervical multiﬁdus. Spine 2005;30(4): E86–91. Bakker DA, Richmond FJ, Abrahams VC. Central projections from cat suboccipital muscles: a study using transganglionic transport of horseradish peroxidase. Journal of Comparative Neurology 1984;228:409–21. Boyd-Clark LC, Briggs CA, Galea MP. Muscle spindle distribution, morphology, and density in longus colli and multiﬁdus muscles of the cervical spine. Spine 2002;27(7):694–701. Edney DP, Porter JD. Neck muscle afferent projections to the brainstem of the monkey: implications for the neural control of gaze. Journal of Comparative Neurology 1986;250:389–98. Elliott JM, Jull GA, Noteboom JT, Durbridge GL, Gibbon WW. Magnetic resonance imaging study of cross-sectional area of the cervical extensor musculature in an asymptomatic cohort. Clinical Anatomy 2005;20:35–40. Elliott JM, Jull GA, Noteboom JT, Darnell R, Galloway GG, Gibbon WW. Fatty inﬁltration in the cervical extensor musculature in persistent whiplash: a MRI analysis. Spine 2006;31(22): E847–55. Forbes GB, Reina JC. Adult lean body mass declines with age: some longitudinal observations. Metabolism 1970;19:653–63. Gargan MF, Bannister GC. Long-term prognosis of soft-tissue injuries of the neck. Journal of Bone and Joint Surgery—British Volume 1990;72(5):901–3. Grauer JN, Panjabi MM, Cholewicki J, et al. Whiplash produces an S-shaped curvature of the neck with hyperextension at lower levels. Spine 1997;22(21):2489–94.

ARTICLE IN PRESS J. Elliott et al. / Manual Therapy 13 (2008) 258–265 Hallgren RC, Greenman PE, Rechtien JJ. Atrophy of suboccipital muscles in patients with chronic pain: a pilot study. Journal of the American Osteopathic Association 1994;94(12):1032–8. Hides JA, Stokes MJ, Saide M, Jull GA, Cooper DH. Evidence of lumbar multiﬁdus muscle wasting ipsilateral to symptoms in patients with acute/subacute low back pain. Spine 1994;19:165–77. Johnson G, Bogduk N, Nowitzke A, House D. Anatomy and actions of the trapezius muscle. Clinical Biomechanics 1994;9:44–50. Kader DF, Wardlaw D, Smith FW. Correlation between the MRI changes in the lumbar multiﬁdus muscles and leg pain. Clinical Radiology 2000;55:145–9. Kristjansson E. Reliability of ultrasonography for the cervical multiﬁdus in asymptomatic and symptomatic subjects. Manual Therapy 2004;9(2):83–8. Larsen LB, Holm R. Prolonged neck pain following automobile accidents. Gender and age related risk calculated on basis of data from an emergency department. Ugeskr Laeger 2000;162(2):178–81. Liu JX, Thornell LE, Domellof-Pedrosa F. Muscle spindles in the deep muscles of the human neck: a morphological and immunocytochemical study. Journal of Histochemistry & Cytochemistry 2003;51(2):175–86. Lord SM, Barnsley L, Wallis BJ, Bogduk N. Chronic cervical zygapophysial joint pain with whiplash: a placebo-controlled prevalence study. Spine 1996;21:1737–45. McPartland JM, Brodeur RR, Hallgren RC. Chronic neck pain, standing balance, and suboccipital muscle atrophy—a pilot study. Journal of Manipulative and Physiological Therapeutics 1997; 20(1):24–9. Murphy WA, Totty WG, Carrol JE. MRI of normal and pathologic skeletal muscle. American Journal of Roentgenology 1986;146(3): 565–74. Panjabi MM, Cholewicki J, Nibu K, et al. Mechanism of whiplash injury. Clinical Biomechanics 1998;13(4–5):239–49.

265

Pearson AM, Ivancic PC, Ito S, et al. Facet joint kinematics and injury mechanisms during simulated whiplash. Spine 2004;29(4): 390–7. Porter JD. Brainstem terminations of extraocular muscle primary afferent neurons in the monkey. Journal of Comparative Neurology 1986;247:133–43. Rankin G, Stokes M, Newham DJ. Size and shape of the posterior neck muscles measured by ultrasound imaging: normal values in males and females of different ages. Manual Therapy 2005;10(2): 108–15. Rorden C, Brett M. Stereotaxic display of brain lesions. Behavioral Neurology 2000;12:191–200. Siegmund GP, Myers BS, Davis MB, Bohnet HF, Winkelstein BA. Mechanical evidence of cervical facet capsule injury during whiplash: a cadaveric study using combined shear, compression, and extension loading. Spine 2001;26(19):2095–101. Spitzer WO, Skovron ML, Salmi LR, Cassidy JD, Duranceau J, Suissa S, et al. Scientiﬁc monograph of the Quebec Task Force on Whiplash-Associated Disorders: redeﬁning ‘‘whiplash’’ and its management. Spine 1995;20(Suppl 8):1S–73S. Treleaven J, Jull G, Lowchoy N. Standing balance in persistent whiplash: a comparison between subjects with and without dizziness. Journal of Rehabilitation Medicine 2005;37(4):224–9. Vernon H, Mior S. The neck disability index: a study of reliability and validity. Journal of Manipulative and Physiological Therapeutics 1991;14:409–15. Winkelstein BA, McLendon RE, Barbir A, Myers BS. An anatomical investigation of the human cervical facet capsule, quantifying muscle insertion area. Journal of Anatomy 2001;198(Part 4): 455–61. Winkelstein BA, Nightingale RW, Richardson WJ, Myers BS. The cervical facet capsule and its role in whiplash injury: a biomechanical investigation. Spine 2000;25(10):1238–46.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 266–275 www.elsevier.com/locate/math

Case Report

Sensorimotor disturbances in neck disorders affecting postural stability, head and eye movement control—Part 2: Case studies Julia Treleaven Neck Pain and Whiplash Research Unit, Division of Physiotherapy, University of Queensland, Brisbane, Q1d 4072, Australia Received 8 October 2007; accepted 8 November 2007

Abstract Recent research highlights sensorimotor control disturbances in those with neck disorders. Assessment and management of the symptoms of dizziness, altered cervical proprioception, eye and head co-ordination and disturbances to postural stability in those with neck disorders are important and are presented in a companion article. In this paper, four case studies are presented to illustrate the formulation and use of a tailored program designed to retrain balance, cervical proprioception and eye and head movement control in those with neck disorders. This program should be used in conjunction with a multi-modal approach to the management of neck disorders. Such a combined approach should address causes of abnormal cervical afferent input as well as the important links between the cervical, vestibular and ocular systems and adaptive changes in the sensorimotor control system. r 2007 Elsevier Ltd. All rights reserved. Keywords: Sensorimotor; Eye; Head; Postural stability; Cervical; Management; Case studies

1. Introduction In the preliminary article to this paper, the theoretical framework for the assessment and management of sensorimotor control disturbances affecting postural stability, head and eye movement control in neck disorders was presented. (Treleaven, 2007) Such disturbances are thought to be resultant of abnormal cervical afferent input and subsequent changes to the integration, timing and tuning of sensorimotor control. Recommendations for clinical assessment and management of such sensorimotor control disturbances in neck pain were presented based on the evidence available to date (Treleaven, 2007). This highlighted the need for a combined approach to address not only the possible causes of abnormal cervical afferent input, such as pain, inﬂammation, altered muscle spindle sensitivity and functional impairment and morphological changes of neck musculature, but to also consider the important links between the cervical, vestibular and ocular systems E-mail address: [email protected] 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.11.002

and any adaptive changes in the sensorimotor control system. Thus, local cervical treatment in conjunction with tailored programs including cervical joint position retraining, gaze stability and eye–head co-ordination exercises as well as walking and balance training to improve sensorimotor control in patients with neck pain was recommended. The current paper presents a series of case studies to speciﬁcally illustrate the formulation and use of the tailored program designed to retrain the sensory and motor aspects of sensorimotor control in those with neck pain. The case studies also highlight the assessment of sensorimotor control in neck pain including investigation of the symptoms of dizziness and unsteadiness, cervical joint position error (JPE), balance and oculomotor control and presents considerations for other possible causes of such disturbances. Particular emphasis is placed on the tailored program in the case studies to illustrate its use, but this should always be incorporated into a multi-modal approach. The tests used in the assessment of sensorimotor control and the speciﬁcs of the tailored exercises are

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 266–275 Table 1 Examples of exercises to improve sensorimotor control in neck disorders Activity

Task

Improve cervical

Relocate back to neutral, eyes closed, laser on headband, check with eyes open Relocate to pre-determined positions in range, laser on headband, dots along wall eyes closed, check eyes open Practice tracing intricate patterns on the wall with laser mounted on headband, eyes open

Joint position error (JPE)

Improve balance

Eyes open then closed, ﬁrm then soft surface Different stances—comfortable, narrow, tandem, single leg Walking with head movements—rotation, ﬂexion and extension maintaining direction and velocity of gait Performing oculomotor or JPE exercises while balance training

Improve oculomotor Eye follow Eyes follow laser light moving backwards and forwards across a wall while sitting in a neutral neck position, then neck torsion (move laser light with hand in lap) Saccades

Place numbered points at different points around concentric circles, quickly move and focus to randomized number

Gaze stability

Maintain gaze as passively move trunk, neck Maintain gaze as actively move trunk and neck all directions Change the focus point—few words, business card Fix gaze, close eyes, move head and open eyes to check have maintained gaze (imaginary gaze) Change the background of the target—plain, stripes, checks

Eye–head coordination

Move eyes focus and then move head same direction and return to neutral Move eyes one direction and head opposite direction Actively move head and eyes together Move head and eyes together when peripheral vision restricted Move hand, arm, head and trunk following with the eyes with or without vision restricted

outlined in Treleaven (2007); a summary of some of the exercises is presented in Table 1. Other tests are outlined in Jull et al. (2004). Brief descriptions of the questionnaires used are as follows.

The Neck Disability Index is a 10-question assessment tool, which records patient responses to questions about neck pain and other symptoms, as well as their effects on physical and social function. The composite score indicates the degree of pain and disability, with higher scores indicative of greater disability (Vernon, 1996).

267

The VAS provides a measure of pain intensity on a scale where 0 is equivalent to no pain and 10 represents the worst pain imaginable. Higher scores on VAS represent greater perceived pain. The General Health Questionnaire 28 (GHQ 28) is a 28-item measure of emotional distress in medical settings, which is divided into four subscales: somatic symptoms, anxiety/insomnia, social dysfunction and severe depression. The total score can be used as a measure of psychological distress. A score above 23 is thought to be indicative of elevated psychological distress (Goldberg, 1978). The Impact of Events Scale (IESR) is a questionnaire that measures current stress related to a speciﬁc life event, speciﬁcally the motor vehicle collision. Three response sets are reported to be associated with psychological reactions to stress: avoidance, intrusion and hyperarousal (Weiss and Maramar, 1997). The Patient Speciﬁc Functional Scale (PSFS) is a questionnaire used to quantify activity limitation and measure functional outcome for patients. The score ranges from 0 (unable to perform the activity) to 10 (able to perform activity) at pre-injury level (Westaway et al., 1998). The Dizziness Handicap Inventory short form (DHIsf) is a 13-item questionnaire developed to measure the self-perceived level of handicap associated with the symptom of dizziness. The DHIsf is scored between two statements; the ﬁrst is scored 1 and the second is scored 0, with a possible maximum score of 13, where 13 indicates no dizziness handicap and 0 maximum handicap (Tesio et al., 1999).

2. Case studies 2.1. Case 1: acute whiplash injury complaining of dizziness 2.1.1. Miss X, age 26 years, flight attendant 2.1.1.1. History. Miss X reported that she had been involved in a motor vehicle collision 5 days prior. She had been stopped on the highway in trafﬁc and a car behind her traveling at approximately 80 km/h was unable to brake in time and rear-ended her car and pushed it into the car in front. She felt immediate bilateral neck pain and stiffness and visited her GP. She had plain cervical X-rays that were unremarkable. Since then her neck pain has persisted and intermittent occipital headache is noted particularly if her neck pain is exacerbated. She reported moderate restriction in her range of cervical motion and several episodes of dizziness and some blurred vision when her neck pain is exacerbated. Dizziness was reported to last several seconds. True vertigo or spinning was not reported. Her symptoms were reportedly increased by neck

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 266–275

268

movements and sustained neck positions and eased with rest, heat and simple analgesic medications. She has not sought any other treatment to date. She is currently off work. She also reported nervousness related to driving as a passenger and has not tried to drive herself. She has also been having dreams and nightmares associated with the accident. She reported no past history of neck pain, dizziness or blurred vision or headaches prior to the motor vehicle collision.

2.1.1.2. Questionnaire scores

Neck Disability Index

30/100

Current pain VAS

4/10

Moderate neck pain

GHQ 28

18/84

IESR

5/12

Below threshold distress Moderate distress

Dizziness Handicap Inventory (sf)

8/13

Mild handicap

Patient speciﬁc functional scale /10

Moderate disability

Reading 5, Driving 6, Sleeping 5

2.1.1.5. Initial management. A multi-modal intervention may include pain-free manual therapy and exercise therapy to improve neuromuscular control and cervical range of motion and advice regarding pain relief, ergonomics and posture. Monitoring of the signs of speciﬁc distress relating to the accident (elevated IESR score and reports of nightmares and dreams about the accident) should be conducted over the next couple of visits and referral to a psychologist if symptoms persist. A tailored program to assist with improvement of sensorimotor control should also be integrated and would include

2.1.1.3. Relevant physical examination findings

dizziness, others difﬁcult to assess at this stage, due to poor cervical range of motion. Eye follow neutral—some dizziness reproduced, unable to assess torsion position due to lack of range of motion. Gaze stability—poor with left cervical rotation and ﬂexion. Saccadic eye movements—poor for eye movements to the left and downwards. Eye–head co-ordination—unable to assess accurately due to lack of cervical range of motion.

Balance comfortable stance eyes open and closed— up to 30-s attempts. Eye follow in neutral—in supine lying. Gaze stability—using target of small dot or mark on the ceiling—ﬂexion and rotation to the left performed passively during treatment and actively in supine lying.

Add next visit/visits Global moderate restriction in cervical range of motion. Symptomatic joint dysfunction (L) C 1/2 and 2/3 motion segments (pain VAS 7/10), decreased pressure pain thresholds over the left cervical spine. Neuromuscular control—poor neuromuscular control of the cervical and left shoulder girdle region. No evidence of nerve tissue mechanosensitivity, neurological deﬁcits or generalized hypersensitivity.

Cervical JPE—left rotation relocation to neutral practice in supine lying if able to perform pain free. Saccades—practicing quickly moving eyes to the left and downwards to ﬁxate on a point.

2.1.1.6. Home program. Two times per day, ﬁve repetitions per exercise. Stop any exercises if it exacerbates pain or headache. A slight exacerbation of dizziness/blurred vision is allowable.

2.1.1.4. Assessment of sensorimotor control

Balance—comfortable stance eyes closed reproduced dizziness and demonstrated marked increased sway. Higher-level balance tests not required at this stage. Cervical JPE in sitting—global limited range of motion to 201 most movements—inaccurate return to neutral from rotation to left productive of

2.1.1.7. Progress. Progression of the oculomotor exercises can be achieved by altering the focus point, background of the focus point, duration, repetitions, range and speed of the activity as well as performing the above relocation, gaze stability and eye follow exercises in sitting. Once cervical range of motion improves reassessment of each task and other tests unable to assess initially

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 266–275

should be performed and exercises tailored according to the ﬁndings. 2.1.1.8. Commentary. This case illustrates that the onset of sensorimotor disturbances can occur soon after injury (Sterling et al., 2003a). Often a full assessment cannot be conducted initially due to poor cervical range of motion, but should be introduced as soon as able and reassessed as range of motion improves. Some of the exercises can begin immediately even though the patient has a signiﬁcant reduction in cervical range of motion. Eye and head exercises are initially encouraged in supine to prevent neck pain and headache exacerbation. Exercises can particularly exacerbate headache and performing the exercises in a comfortable supported position should eliminate this. Limiting how many exercises are given for the home program as well as integrating the exercises into others such as while performing range of motion exercises will also avoid pain exacerbation and avoid an ‘‘overload’’ of exercises. This should assist with patient compliance with the home program. In this case, the description of the dizziness and the deﬁcits in sensorimotor control support the cervical spine as the primary cause of the disturbances rather than a vestibular or an anxiety-related cause. The patient does present with an elevated level of speciﬁc psychological distress relating to the accident but this is unlikely to be the primary cause of the sensorimotor disturbances although it might contribute to abnormal cervical afferent input, via sympathetic nervous system activation (Passatore and Roatta, 2006), and should be closely monitored and addressed as it has also been shown to be a prognostic indicator for a poorer outcome for whiplash-associated disorders and unlikely to change over time without intervention (Sterling et al., 2003b). 2.2. Case 2: chronic neck pain and headache— no reported dizziness 2.2.1. Mrs. X, age 46 years, counselor 2.2.1.1. History. Mrs. X reported bilateral neck pain right side greater than left, stiffness and tightness in the upper trapezius region as well as an occipital and frontal headache several times per week. Plain cervical X-rays demonstrate early degenerative changes in the cervical spine. She reported that her symptoms were increased by neck movements, computer work, sitting talking to clients when her neck was turned or in a rotated position, driving and sustained neck positions and eased with rest, heat and medications. She has not sought any other treatment to date. She works fulltime as a counselor. She has had no past history of trauma or injury to the neck or head.

269

2.2.1.2. Questionnaire scores.

Neck Disability Index Score

16/100

Mild disability

Current pain VAS

4/10

Moderate neck pain

GHQ 28

4/84

Dizziness Handicap Inventory (sf)

13/13

Below threshold distress No handicap

Patient speciﬁc functional scale

Reading 7, Driving 7, Sustained sitting 7

2.2.1.3. Relevant physical examination findings

Moderate restriction in cervical range of motion into right rotation and lateral ﬂexion. Symptomatic joint dysfunction at the (R) C 2/3 and 5/6 motion segment (pain VAS 3/10). Neuromuscular control—poor neuromuscular control of the cervical and right shoulder girdle regions. No evidence of nerve tissue mechanosensitivity, neurological deﬁcits or secondary hyperalgesia.

2.2.1.4. Assessment of sensorimotor control

Balance—comfortable stance within normal limits. Narrow stance eyes closed on foam—moderate increased sway. Tandem stance eyes closed unable to maintain 30 s, failed at 5 s. Cervical JPE (o4.5 cm normal)—extension 1 cm, rotation (L) 2.5 cm, rotation (R) 2 cm. All within normal limits. Eye follow with left neck torsion position difﬁcult to follow with increased saccades when compared to neutral neck position and right neck torsion position. Gaze stability—ﬂexion and rotation to the right poor. Eye–head co-ordination—normal. Saccades—normal.

2.2.1.5. Initial management. Management to address abnormal cervical afferent input—might include manual therapy, exercises to improve neuromuscular control and cervical range of motion and advice on pain relief, ergonomics and posture. A tailored program to assist with the improvement of sensorimotor

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 266–275

270

Fig. 1. Following a laser light backwards and forwards across a wall with the neck in relative left torsion in sitting. The patient moves the laser light to provide the target to follow.

control should also be added and would include the following: Fig. 2. Soft surface narrow stance practice.

Balance—narrow stance eyes closed on foam, tandem stance eyes open then closed—30-s attempts. Eye follow with neck in left torsion in sitting, ﬁve times. At home, the patient could move a laser light or torch backwards and forwards across a wall to provide the target to follow (Fig. 1). Gaze stability—sitting focus on dot or word—move head into ﬂexion and rotation to the right.

2.2.1.6. Home program. repetitions per exercise.

Two times per day, ﬁve

Adding a busy background to the focus point. Adding a soft surface to narrow stance tests (Fig. 2).

2.2.1.8. Commentary. This case illustrates the point that in the absence of any complaints of dizziness or unsteadiness, deﬁcits in sensorimotor control can still occur and will be an important addition to the management of this patient. The patient reported difﬁculty when sitting talking to clients when her head was in an awkward position, which could be related to the neck itself but could also be related to the eye movement disturbances. Addition of the tailored program is likely to enhance the management of this patient’s neck pain and headache.

2.2.1.7. Progress

Eye and head exercises in standing then in narrow, tandem and single leg stance. Altering the focus point from a dot to some words or a business card. Increasing speed or range of movements. Increasing duration or number of repetitions.

2.3. Case 3: chronic whiplash, complaints of dizziness, blurred vision and moderate to severe headaches 2.3.1. Ms. X, age 49, sales executive Ms. X presented with ongoing constant neck pain radiating to the right trapezius region and right arm pain associated with frequent headaches, pain and

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 266–275

vertigo following a motor vehicle collision 6 months ago. She has episodes of nausea and non-spinning vertigo with a constant feeling of unsteadiness worsened by increased neck pain and headaches. She also reported episodes of blurred vision, occasional tinnitus, fullness in the right ear and right ear pain. Her ear pain is associated with the neck pain and working, particularly reading and looking down. She has only had one episode of vomiting associated with the vertigo 2 days post the accident. She has not had any episodes of true ‘‘spinning’’ vertigo. She also reported difﬁculty with bumping into doorways, veering to the right when walking, blurred vision with reading and words moving around the page at times. She has no prior history of neck pain, headaches, dizziness or ear complaints. She is currently taking daily Celebrex and Panadol and Panadeine forte to control her headaches several times per week. Ms. X has sought opinions from her general practitioner and a neurologist. She attended physiotherapy and acupuncture initially following the accident but found that this tended to exacerbate her symptoms at times. CT and MRI scans of her head and cervical spine have been performed. A slight disk bulge at C5/6 was the only reported abnormal ﬁnding.

Neck Disability Index Score

56/100

Moderate/severe pain and disability

Current pain VAS

7/10

Moderate pain

GHQ 28

17/84

Impact of events revised score

2.83/12

Below threshold score Below threshold score

Dizziness Handicap Inventory (sf)

7

Mild/moderate handicap

Patient speciﬁc functional scale

/10

Fast walk 0, Cleaning 2, Reading 4

2.3.1.2. Relevant physical examination findings Moderate global range of cervical movement deﬁcits. Symptomatic joint dysfunction at the (R) C1/2, 2/3 and C4/5 motion segments (pain VAS 5/10). Neuromuscular control—poor neuromuscular

control of the cervical and right shoulder girdle regions complicated by muscle protection of nerve tissue mechanosensitivity. Decreased mechanical pain thresholds for the cervical spine. Decreased mechanical pain thresholds for upper and lower limb sites and increased cold pain thresholds over the cervical spine indicative of generalized hypersensitivity and altered central nervous system pain processing. Nerve tissue mechanosensitivity in the right upper limb and upper cervical region demonstrated on the brachial plexus provocation test and passive neck ﬂexion test.

2.3.1.3. Assessment of sensorimotor control

2.3.1.1. Questionnaire scores

271

Hallpike dix test (Herdman, 1997)—negative— no nystagmus reproduced. Cervical JPE in sitting (o4.5 cm normal)—marked increase in JPE from rotation to the left. Extension 3 cm, rotation (L) 12.5 cm, rotation (R) 2 cm. Extension also causes exacerbation of nausea and dizziness. Eye follow in sitting—markedly abnormal with neck torsion to the left and right, indicating disturbances to cervical afferent input is inﬂuencing eye movement control. This also reproduced vertigo, blurred vision and headache particularly with neck torsion to the left. Gaze stability in sitting—decreased with rotation to the right and ﬂexion. This reproduced dizziness, blurred vision and nausea. Eye–head co-ordination in sitting—decreased on the right side. This reproduced dizziness, blurred vision and nausea. Saccades in sitting—abnormal for eye movement to the right and downwards. Standing balance—evidence of increased sway on most tests of comfortable and narrow stance particularly on tests with the eyes closed. Ms. X was unable to perform tandem stance with eyes closed. Eyes closed tasks also increased headache and vertigo.

2.3.1.4. Initial management. Management to address abnormal cervical afferent input—in light of the signs of generalized hypersensitivity and nerve tissue sensitivity, management should be non-pain provocative. A review of pain management options by her general practitioner should be performed to ensure adequate pain control. Physiotherapy might include non-pain provocative manual therapy, techniques to improve nerve tissue sensitivity, exercises to improve neuromuscular control and cervical range of motion and ergonomic and postural advice. A tailored program to assist with the

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 266–275

272

improvement of sensorimotor control should also be added and will be important to address the symptoms of unsteadiness, vertigo, blurred vision and ear symptoms and would include the following:

Balance comfortable and narrow stance eyes open and closed—30-s attempts. Cervical JPE relocation back to neutral following rotation to the left in supine lying—comfortable movement only. Gaze stability—rotation to the right and ﬂexion active and passive movement in supine lying. Leave eye–head co-ordination, saccades and eye follow neck right torsion exercises until reassess any effect of the above exercises on pain exacerbation.

2.3.1.5. Home program

Two times per day, three repetitions each exercise. No increased pain during exercises. Stop if exacerbates pain or headache. Some dizziness and nausea is acceptable but should not last longer than a couple of minutes.

2.3.1.6. Progress

Carefully and slowly so not to exacerbate pain or headache. Passive trunk movement into left rotation in sitting or standing during treatment might also be helpful for gaze stability (Fig. 3). Add saccades, eye–head co-ordination and eye follow exercises one at a time. Add cervical JPE extension supine. Increasing number of repetitions of each exercise. Eye and head exercises in sitting then standing.

Altering the focus point from a dot to some words or a business card. Increasing speed or range of movements. Adding a busy background to the focus point. Restricting the peripheral vision. Balance—add soft surfaces, practice eyes open and closed in tandem and single leg stance, eye and head exercises in comfortable then in narrow stance, walking with head turns maintaining direction and velocity of gait.

2.3.1.7. Commentary. This case illustrates the complex patient issues of generalized hypersensitivity, nerve tissue mechanosensitivity, poor neuromuscular control, cervical joint dysfunction and altered sensorimotor control in those with persistent whiplash associated disorders. Such a combination of problems is likely to compound abnormal cervical afferent from a variety of sources. In this case, it is essential that any management be non-pain provocative and care with the progression of the exercises will be very important to avoid pain exacerbation. Adequate medication to provide effective pain management will be important. The moderate postural control deﬁcits are likely resultant of abnormal cervical afferent input to the postural control system. Her description of the dizziness would ﬁt a cervical cause of the dizziness. Subjective ear complaints such as these can also be associated with cervical disorders. However, secondary vestibular impairment or concomitant vestibular pathology may be present and to date she has not had any investigations associated with this (apart from the negative Hallpike Dix which can exclude benign paroxysmal positional vertigo (BPPV) as a possible cause). Thus she should be monitored carefully and if her symptoms of vertigo, ear complaints and blurred vision do not improve in line with improvements to the cervical physical function, review by an ENT specialist or vestibular rehabilitation physiotherapist would be appropriate. 2.4. Case 4: chronic dizziness, unsteadiness and nausea associated with symptoms of neck stiffness and headache

Fig. 3. Passive trunk movement into left rotation in standing to improve gaze stability.

2.4.1. Mr. X, age 60 years, self-employed company director 2.4.1.1. History. Mr. X reported a history of persistent bilateral orbital pain, tiredness and heaviness of the eyes associated with neck stiffness and discomfort following an accident 2 years ago where he walked into a plate glass door. He reported immediate headache, which settled. Ten days afterwards, he drove for 6 h and after falling asleep in an awkward neck position in the motel, he awoke with dizziness (not true vertigo) and had difﬁculty standing and walking. This settled with stematil after approximately 4 days. Incidentally, he had also been receiving treatment (antibiotics) for an ear

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 266–275

infection a few days prior to this. Since then he has had several episodes (lasting approximately 1 week) of mild to moderate dizziness and unsteadiness and or nausea preceded with activities involving sustained neck extension or ﬂexion. He also reported intermittent difﬁculty with visual conﬂict in shopping center aisles and bumping into objects. Symptoms were reported to be temporarily relieved by local cervical physiotherapy and chiropractic treatment, applying ice packs to the eyes and rest. Mr. X reported that he has not had any other symptoms suggestive of VBI or migraine. He also reported no incidents of true vertigo. Mr. X has consulted several specialists in an attempt to try to determine the cause of his symptoms including an ENT specialist, neurologist, maxillofacial surgeon, optometrist and a physician. He has had two unremarkable brain CT scans, one with IV contrast injection. Normal audiometry testing and a negative Hallpike Dix test for benign BPPV were also reported. In the last 6 months, Mr. X has been attending both physiotherapy and chiropractic treatment once every week to 2 weeks. He reported that both these treatments focusing on his upper cervical spine temporarily relieve his eye pain, neck discomfort and unsteadiness; however, he feels his overall condition remains the same. Physiotherapy has been focusing on manual therapy, and more recently deep cervical ﬂexor rehabilitation. Mr. X reported that he has not been vigilant with his home exercise program. Twenty-ﬁve years ago, he was hit on the top of the skull by a roller door, resulting in a loss of consciousness. He reported no neck pain or headaches as a result of this accident. 2.4.1.2. Questionnaire scores

Neck Disability Index Score

20/100

Mild disability

Current pain VAS

0.5/10

Mild neck pain

GHQ 28

17/84

Impact of events revised score

3.08/12

Dizziness Handicap Inventory (sf)

4/13

Patient speciﬁc functional scale

2.4.1.3. Relevant physical examination findings

Moderate restriction in upper cervical range of motion. Symptomatic joint dysfunction: (R) C 1/2, 2/3 and 3/4 motion segments (pain VAS 3/10). Poor neuromuscular control of the cervical spine, scapular control non-remarkable. No evidence of nerve tissue mechanosensitivity, neurological deﬁcits or generalized hypersensitivity.

2.4.1.4. Assessment of sensorimotor control

Cervical JPE (o4.5 cm normal)—increased for return from rotation to the left, right and extension. Extension 10.5 cm, rotation (L) 8.6 cm, rotation (R) 5.2 cm. Eye follow—abnormal (increased catch up saccades) when tested with neck torsion to the right and left when compared to the neutral neck position. Gaze stability—poor and blurred vision when the head moves into ﬂexion and extension. Eye–head co-ordination—within normal limits. Saccadic eye movements into ﬂexion and extension—blurred vision and dizziness reproduced. Standing balance—increase sway was noted on all tests in narrow stance.

2.4.1.5. Initial management. Management to address abnormal afferent input should continue especially exercises to improve cervical neuromuscular control and compliance with the home program will be vital. Cervical range of motion and ergonomic and postural advice to avoid unnecessary stresses placed on the upper cervical region will also be important. However, most importantly, a tailored program to assist with the improvement of sensorimotor control should also be added and would include the following:

Below threshold distress Mild distress

273

Balance—narrow stance tasks 30 s eyes open and closed. Cervical JPE retraining in sitting into extension and rotation left and right. Gaze stability in sitting into ﬂexion and extension.

Add next visit/visits Moderate handicap

Eye follow—with neck torsion to the left and right. Saccadic eye movements into ﬂexion and extension.

2.4.1.6. Home program Reading 4, Driving 4, Watching TV 4/10

Two to three times per day, ﬁve repetitions each exercise; stop if exacerbates pain or headache and

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 266–275

274

some dizziness and nausea is acceptable but should not last longer than a couple of minutes.

2.4.1.7. Progression

Cervical joint position retraining to different points in range of neck motion, tracing patterns with laser light on head (Fig. 4). Balance—add soft surfaces, eye and head exercises in comfortable then in narrow stance, walking with head turns maintaining direction and velocity of gait. Altering the focus point from a dot to some words or a business card and or adding a busy background to the focus point (Fig. 5). Increasing speed or range of movements. Increasing number of repetitions. Restricting peripheral vision performing active head and eye movements (Fig. 6).

Fig. 5. Gaze stability exercises with a business card as the focus point and a busy striped background.

2.4.1.8. Commentary. In this case, there was evidence of altered neuromotor control of the cervical spine, painful upper cervical segmental joint dysfunction in conjunction with moderate sensorimotor control disturbances relating to cervical JPE, neck-inﬂuenced eye follow, gaze stability, saccadic eye movement and balance. It is likely that the causes of his symptoms are due to abnormal afferent input from the cervical spine and has been improved temporarily with management addressing his cervical spine. It is unlikely that the vestibular system is the primary cause of the complaints but it is possible that some

Fig. 6. Restricting peripheral vision while performing active head and eye movements.

Fig. 4. Cervical joint position retraining tracing a pattern with a laser light mounted onto the head.

hypofunction of the vestibular system is present due to the patient’s age, past history of the vestibular neuritis and secondary adaptations due to inﬂuence from the cervical spine. Mild head injury is also possible. Mr. X has attended both physiotherapy and chiropractic in the past with only temporary results. More compliance with the home exercise program to improve cervical neuromuscular control will be important as this has shown to improve cervical JPE (Jull et al., 2007). However, in this case, it would seem that the introduction of the tailored program designed to improve sensorimotor control will be of most signiﬁcance to assist in the long term.

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 266–275

275

3. Integration of exercises

References

A number of the above exercises can also be easily integrated into the overall multi-modal physiotherapy management and home exercise program which can assist in patient compliance, for example, gaze stability while performing range of motion exercises, gaze stability while performing manual therapy techniques, e.g. PIVVMS, use of eye movement (eye–head coordination) to facilitate craniocervical ﬂexion (CCFT) or cervical range of motion exercises, targeting speciﬁc pressures with the eyes closed during CCFT retraining, gaze stability and eye–head co-ordination exercises while performing neck extensor exercises in four-point kneeling (Jull et al., 2004), and adding oculomotor and joint relocation exercises while standing in various positions and when walking.

Goldberg D. Manual of the general health questionnaire. Windsor: NFER-Nelson; 1978. Herdman S. Advances in the treatment of vestibular disorders. Physical Therapy 1997;77(6):602–17. Jull G, Falla D, Treleaven J, Sterling M, O’Leary SC. A therapeutic exercise approach for cervical disorders. Edinburgh: Churchill Livingstone, Elsevier; 2004. Jull G, Falla D, Treleaven J, Hodges P, Vicenzino B. Retraining cervical joint position sense: the effect of two exercise regimes. Journal of Orthopaedic Research 2007:5. Passatore M, Roatta S. Inﬂuence of sympathetic nervous system on sensorimotor function: whiplash associated disorders (WAD) as a model. European Journal of Applied Physiology 2006;98:423–49. Sterling M, Jull G, Vicenzino B, Kenardy J, Darnell R. Development of motor system dysfunction following whiplash injury. Pain 2003a;103(1/2):65–73. Sterling M, Kenardy J, Jull G, Vicenzino B. The development of psychological changes following whiplash injury. Pain 2003b; 106(3):481–9. Tesio L, Alpini D, Cesarani A, Perucca M. Short form of the dizziness handicap inventory. American Journal of Physical Medicine & Rehabilitation 1999;78(3):233–41. Treleaven J. Sensorimotor disturbances in neck disorders affecting postural stability, head and eye movement control. Manual Therapy 2007. Vernon H. The neck disability index: patient assessment and outcome monitoring in whiplash. Journal of Musculoskeletal Pain 1996;4:95–104. Weiss D, Maramar C. The impact of event scale—revised. In: Wilson J, Keane T, editors. Assessing psychological trauma and PTSD. New York: Guildford; 1997. Westaway MD, Stratford PW, Binkley JM. The patient-speciﬁc functional scale: validation of its use in persons with neck dysfunction. Journal of Orthopaedic & Sports Physical Therapy 1998;27(5):331–8.

4. Summary The four case studies presented highlight the importance of assessment and management of sensorimotor control disturbances in those with neck disorders. Multi-modal management for the cervical spine is advocated and as part of this, the formulation and use of a tailored program designed to retrain balance, cervical proprioception and eye and head movement control in those with neck disorders has been speciﬁcally considered.

Manual Therapy (2008) 13(3), 276

Available online at www.sciencedirect.com

Diary of events

2nd World Congress on Manual Therapy and Sport Rehabilitation, The Spine II, in Roma Italy 6th–8th of February 2009 www.newmaster.it

10th International IFOMT Congress, Rotterdam 8th–13th June 2008 The Scientiﬁc Committee wishes to invite abstract submissions for Platform and Poster Presentations. Instructions are now online and available at www.ifomt2008.nl

Janet G. Travell, MD Seminar Series, Bethesda, USA For information, contact: Myopain Seminars, 7830 Old Georgetown Road, Suite C-15, Bethesda, MD 20814-2432, USA. Tel.: +1 301 656 0220; Fax: +1 301 654 0333; website: www.painpoints.com/seminars.htm E-mail: [email protected]

5th Chiropractic Osteopathy Physiotherapy Annual Undergraduate/Pre-Registration Research Conference ‘‘Moving Forward Through Research and Practice’’ 25th October 2008, Bournemouth, UK. Keymote Speaker: Prof. Gordon Waddell, Rehabilitation What Works, For Who and When? Details on how to submit an abstract and register available soon at www. aecc.ac.uk

If you wish to advertise a course/conference, please contact: Karen Beeton, Associate Head of School (Professional Development), School of Health and Emergency Professions, University of Hertfordshire, College Lane, Hatﬁeld, Herts AL10 9AB, UK. There is no charge for this service.

276

ARTICLE IN PRESS

Manual Therapy 13 (2008) e1–e2 www.elsevier.com/locate/math

Letter to the Editor Landers M, Creger R, Baker C, Stutelberg K. The use of fear-avoidance beliefs and non-organic signs in predicting prolonged disability in patients with neck pain. Manual Therapy 2007; doi:10.1016/j.math.2007.01.010

To the Editor, Landers et al. (2007) again attempt the perpetuation of old myths regarding physical and psychological factors associated with pain-related disability. Contrary to their assertions, the authors do not seem aware of either the contemporary literature in this area or current understanding of neurophysiological processes associated with pain perception. The authors’ hackneyed version that ‘non-organic’ signs suggest ‘‘no demonstrable organic cause or physiological dysfunction’’; ‘‘may be suggestive of psychological factors related to abnormal illness behaviour’’ and ‘‘correlate with the neurotic triad (hypochondriasis, depression and hysteria scales)’’ is not supported by the recent literature. The authors have overlooked Waddell’s more recent publications in this area which readers are advised to peruse. To quote: ‘‘How often do I have to say this to stop people misquoting my work? yy Non-organic signs are simply one part of the current clinical presentation; non-organic signs do not tell us anything about the original cause of the pain; non-organic signs do not mean that the pain is not real, psychological or faked’’ (Waddell, 2004, p. 192). Although Landers et al. (2007) cite a recent review by Fishbain and colleagues (2003); they fail to acknowledge the evidence that leads Fishbain et al. to conclude: ‘nonorganic signs’ do not correlate with psychological distress; ‘non-organic signs’ may represent an organic phenomenon and whilst non-organic signs are associated with greater pain levels, they are not associated with secondary gain. In fact, the evidence to support the presence of augmented central pain processing mechanisms as an important contributor to some neck pain conditions is now substantial (Koelbaek-Johansen et al., 1999; Curatolo et al., 2001; Moog et al., 2002; Sterling et al., 2003). Perhaps more concerning is Landers et al.’s statement ‘‘non-organic signs are ﬁndings, during a 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.10.004

patient examination, that are suggestive of a physical illness or disease for which there is no demonstrable organic cause or physiological dysfunction. Simply put, non-organic signs are ﬁndings that do not seem to be consistent with the nature of a particular pathology’’. In this scenario, when a patient’s symptoms do not ﬁt the clinician’s prototypical construct of the condition in question, the patient’s presentation is deemed to be inappropriate. It has been argued that this prescriptive reasoning process allows clinicians to make moral or ethical claims of the patient’s behaviour (Cronje and Williamson, 2006), a situation that has no place in contemporary musculoskeletal practice. As most musculoskeletal practitioners would be aware, attempts to attribute the symptoms of patients as having a predominantly psychological basis are not only unhelpful for clinicians but may in fact adversely affect patient outcomes by discrediting the validity of their complaints. It could be argued as doing as much disservice to a patient as an over reliance on high-tech interventions to identify a physical cause of pain, which is now (rightly) frowned upon in the management of spinal pain. Only when we move past attempts to dichotomize the mind and body will a more sophisticated and humane approach to the management of spinal pain is realized.

References Cronje R, Williamson O. Is pain ever normal?. Clinical Journal of Pain 2006;22:692–9. Curatolo M, Petersen-Felix S, Arendt-Nielsen L, Giani C, Zbinden A, Radanov B. Central hypersensitivity in chronic pain after whiplash injury. Clinical Journal of Pain 2001;17:306–15. Fishbain D, Cole B, Cutler R, Lewis J, Rosomoff H, Rosomoff R. A structured evidence-based review on the meaning of non-organic physical signs: waddell signs. Pain Medicine 2003; 4:141–81. Koelbaek-Johansen M, Graven-Nielsen T, Schou-Olesen A, ArendtNielsen L. Muscular hyperalgesia and referred pain in chronic whiplash syndrome. Pain 1999;83:229–34. Landers M, Creger R, Baker C, Stutelberg K. The use of fearavoidance beliefs and non-organic signs in predicting prolonged disability in patients with neck pain. Manual Therapy 2007; doi:10.1016/j.math.2007.01.010. Moog M, Quintner J, Hall T, Zusman M. The late whiplash syndrome: a psychophysical study. European Journal of Pain 2002; 6:283–94.

ARTICLE IN PRESS e2

Letter to the Editor / Manual Therapy 13 (2008) e1–e2

Sterling M, Jull G, Vicenzino B, Kenardy J. Sensory hypersensitivity occurs soon after whiplash injury and is associated with poor recovery. Pain 2003;104:509–17. Waddell G. The back pain revolution. Edinburgh: Churchill Livingstone; 2004.

Michele Sterling, PhD, MPhty, Grad Dip Manip Physio, Bphty Department of Physiotherapy, Centre for National Research on Disability and Rehabilitation Medicine

(CONROD), The University of Queensland 4006, Australia E-mail address: [email protected] Owen D. Williamson, FFPMANCZA Department of Epidemiology and Preventative Medicine, Monash University, Victoria 3004, Australia

ARTICLE IN PRESS

Manual Therapy 13 (2008) e3–e4 www.elsevier.com/locate/math

Reply to letter to the editors Reply to ‘‘Letter to the editors’’: re: Landers M, Creger R, Baker C, Stutelberg K. The use of fear-avoidance beliefs and nonorganic signs in predicting prolonged disability in patients with neck pain. Manual Therapy 2007 Dear Editors, We are grateful to the authors of this letter, who have brought to light several issues related to our article. We agree with them on several points. In our article, we found, among other things, that cervical nonorganic signs were predictive of prolonged neck pain-related disability. However, it should be clear, that these authors have not made any assertions whatsoever that our ﬁndings were invalid. Instead, the substance of their assertions is simply that the deﬁnitions that were used in our article are ‘‘hackneyed’’ or perhaps more accurately, obsolete. While we agree on several points related to these deﬁnitions, we also disagree with some of their characterizations and assertions, most of which are based on assumptions. The authors state that ‘‘The authors’ hackneyed versionyyis not supported by the recent literature.’’ There is no disagreement on our part that this deﬁnition may be obsolete. However, the deﬁnition we used is consistent with the purported nature of the name of this construct (i.e., nonorganic), its original description, and historical deﬁnitions. We agree that cervical nonorganic signs (CNOS) are simply a small component of the clinical presentation and should not imply that the pain is psychological. We agree with Waddell that these types of signs or symptoms may simply represent an atypical presentation that does not ﬁt the common patterns of presentations to which clinicians are accustomed (Waddell, 2004). We also agree with the authors that nonorganic signs may represent augmented central pain processing. However, we are confused at this line of contention, since we have made this exact same point in the discussion section of our paper. In fact, we used two of the same references (Curatolo et al., 2001; KoelbaekJohnson et al., 1999) that the authors used in their letter. Consider our narrative on this subject: Careful consideration of the other nonorganic signs is also recommended since evidence from the literature suggests that some of these signs may, in fact, be 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.10.005

organic. For instance, pain to superﬁcial palpation may represent a central nervous system-mediated hypersensitivity reaction. (Curatolo et al., 2001; Koelbaek-Johnson et al., 1999). The authors also state that when a patient’s symptoms do not ﬁt the clinician’s prototypical construct of the condition in question, ‘‘the patient’s presentation is deemed to be inappropriate.’’ Nowhere in our paper do we make the assertion that a patient’s presentation is inappropriate. In fact, we actually caution the reader ‘‘y.not to overinterpret the presence of nonorganic signs’’ as ‘‘y.some of these signs may, in fact, be organic.’’ We also added several lines of discussion that are critical of the CNOS as a measure of nonorganic behavior. Lastly, in the next paragraph, we have advised the clinician to remember that ‘‘ysubjects who present with high fear avoidance beliefs and nonorganic behavior simply may be more prone to developing prolonged disability.’’ Additionally, we suggest that these associations may be spurious and should not imply cause and effect. This is an important distinction and one that should be considered in the context of the whole discussion, which was clearly critical of the use of CNOS as a tool for dichotomizing the organic nature of neck pain. The authors have also made a huge downstream assertion that, based on our paper, moral and ethical claims may be made about patients’ behavior. Not only was this never mentioned in our narrative, it goes inappropriately beyond our data. We do, however, appreciate this opportunity to reiterate that CNOS should be interpreted with caution. They are a part of discovery during the examination that may be suggestive of someone prone to prolonged disability. They should not be used in medicolegal contexts or to make moral or ethical claims. Another assertion made by the authors is that we have attempted ‘‘yto attribute the symptoms of patients as having a predominately psychological basisy .’’ Again, we have not done this. A reading of the discussion section illustrates our view on this issue. We have stated that 56% of the variance in the 12-week disability scores was the initial physical disability. CNOS only explained 5.3% of 12-week disability variance over and above the initial physical disability (56%) and fear-avoidance

ARTICLE IN PRESS e4

Reply to letter to the editors / Manual Therapy 13 (2008) e3–e4

beliefs (6.2%). We concluded that this would logically mean that the focus of our treatment should be on the physical factors related to prolonged disability. However, we had suggested that other factors, such as CNOS, should also play a role in clinical decisionmaking. We have appreciated this opportunity to participate in healthy scientiﬁc discourse on CNOS. As a matter of conclusion, we would like to make our position clear that CNOS: (A) are a part of discovery during the initial examination that assist a clinician in making decisions about patients prone to prolonged neck pain-related disability; (B) may actually represent an organic condition; (C) should not be overinterpreted to mean the patient has a psychological condition and/or to make moral/ethical judgements; and, (D) should be considered in the context of the whole examination. Lastly, we should point out that the evidence on nonorganic signs is, by no means, fully conclusive. These signs may or may not be related to a true nonorganic condition. However, one thing is clear, the

validity of these signs should not be made by opinion; rather, it should be made by reasoning gathered through the scientiﬁc process. Respectfully, Merrill Landers, DPT, OCS Rachel Creger, MSPT Carrie Baker, MSPT Karl Stutelberg, MSPT References Waddell G. The back pain revolution. Edinburgh: Churchill Livingstone; 2004. Curatolo M, Petersen-Felix S, Arendt-Nielsen L, et al. Central hypersensitivity in chronic pain after whiplash injury. Clin J Pain 2001;17:306–15. Koelbaek Johansen M, Graven-Nielsen T, Schou Olesen A, ArendtNielsen L. Generalised muscular hyperalgesia in chronic whiplash syndrome. Pain 1999;83:229–34.

Merrill Landers, Rachel Creger, Carrie Baker, Karl Stutelberg Division of Health Sciences, University of Nevada, Las Vegas, 4505 Maryland Parkway, P.O. Box 453029, Las Vegas, NV 89154, USA E-mail address: [email protected] (M. Landers)

ARTICLE IN PRESS

Manual Therapy 13 (2008) e5 www.elsevier.com/locate/math

Letter to the Editor Challenging editorial wisdom and raising the ‘‘VBI’’ debate Balancing objectively, editorial copy must be a difﬁcult task. From their position, editors are able to inﬂuence and even manipulate opinion. Whilst thanking Professor Rivett for his editorial, which kindly endorsed many of the points raised in the article which followed in the same issue (Kerry and Taylor, 2006), we feel that there is need for balance in the debate with regard to pre-treatment cervical spine risk assessment. Carney and Anderson (1981) stated that ‘‘the clinical approach to ischaemia of the brain is burdened by practices established by custom and conﬂicting terminologies and perspectives tailored to the individual needs of diverse medical specialties’’, they went on to suggest a ‘‘system based’’ approach. Their contention that ‘‘blood ﬂow within the carotid artery and vertebral arteries is interdependent via the circle of Willis, and it follows therefore that neither one hemisphere nor one system can be validly studied in isolation’’ is one that deserves attention. Manual therapy and physiotherapy in particular, seems burdened by its narrow focus on the vertebral artery in isolation, particularly the use of ‘‘vertebral artery tests’’. The continued defence of provocative positional testing in its current format, whilst admirable, ﬂies in the face of the current evidence. Whilst the assertion that provocative testing may detect the ‘‘occasional’’ patient, is technically correct, should this really make this test the cornerstone of risk assessment prior to treatment? Rather, the tests poor reliability could perhaps rank it alongside Homan’s test, which (due to its known poor reliability) is gradually disappearing from modern texts detailing risk assessment methods for venous thrombo-embolism (VTE). The carotid artery has been for some time considered in medicine to be a strong predictor of cardio-vascular health (Bots et al., 1997). Therefore if, as suggested by both ourselves and Professor Rivett, the therapist is to cultivate a high index of suspicion of arterial dissection, failure to consider carotid pathology would potentially lead to grave error, as numerous case series suggest (Biousse et al., 1994). Particularly as head and neck pain are frequently cited as the only early presenting symptoms (Arnold and Bousser, 2005). A review of medico-legal cases involving stroke following treatment, suggests that in many cases the 1356-689X/$ - see front matter r 2007 Published by Elsevier Ltd. doi:10.1016/j.math.2007.09.001

presenting features of arterial dissection (both carotid and vertebral) were already present at the time of the administered treatment. This further adds to the concept that attempts to predict ‘‘VBI’’ or potential arterial damage based on the current evidence, may be ﬂawed and that more effort should be applied to educating therapists as to the clear presenting symptoms and objective tests which together in combination, may reveal underlying arterial dissection. There is a need for a much wider understanding of the haemodynamic and biomechanical mechanisms leading to brain ischaemia together with a clear picture of the key presenting symptoms of arterial insult. This should incorporate a general vascular risk assessment, which puts vertebral artery testing into context and which incorporates carotid artery assessment into a ‘‘system based’’ approach, much like the concept applied to VTE assessment, a similarly potentially fatal condition. References Arnold M, Bousser MG. Carotid and vertebral artery dissection. Practical Neurology 2005;5:100–9. Bots ML, Hoes AW, Koudstaal PJ, Hofman A, Grobbee DE. Common carotid intima-media thickness and risk of stroke and myocardial infarction: the Rotterdam Study. Circulation 1997;2, 96(5):1432–7. Biousse V, D’Anglejan-Chatillon J, Massiou H, Bousser MG. Head pain in non-traumatic carotid artery dissection: a series of 65 patients. Cephalalgia 1994;14(1):33–6. Carney AL, Anderson EM. The system approach to brain blood ﬂow. In: Carney AL, Anderson EM editors. Advances in neurology, vol. 30. New York: Raven Press; 1981. p. 1–30. Kerry R, Taylor AJ. Cervical arterial dysfunction assessment and manual therapy. Manual Therapy 2006;11(4):243–53.

Alan J. Taylor Nottingham Nuffield Hospital, University of Nottingham, 748 Mansfield Road, Woodthorpe, Nottingham NG5 3 FZ, UK E-mail address: Alan.Taylor@NufﬁeldHospitals.org.uk Roger Kerry Division of Physiotherapy Education, University of Nottingham, Nottingham, UK

ARTICLE IN PRESS

Manual Therapy 13 (2008) e6 www.elsevier.com/locate/math

Letter to the Editor Letter to Editor response I would like to thank Dr. Kerry and Mr. Taylor for raising some important issues related to cervical spine pre-manipulative screening in response to my recent editorial (Rivett, 2006). Of course, an editorial is by nature brief and cannot fully cover all matters pertaining to this issue. It is also by deﬁnition an opinion or view, albeit an informed one. Before responding to some points made in their letter, I would respectfully disagree that the editorial suggests that provocative positional testing remains ‘‘the cornerstone of risk assessment prior to treatment’’. Indeed, the editorial only brieﬂy discusses these tests, highlighting their limited predictive validity, and stresses the importance of recognising the early symptoms of a dissecting vertebral artery (as do Kerry and Taylor, 2006) and the role of clinical reasoning in averting manipulative catastrophes. That is, the emphasis has shifted to greater weight being placed on interpretation of the ﬁndings elicited in the patient’s history, and substantially less on the results of the provocative tests themselves (Rivett et al., 2006). The provocative tests have not yet been discarded as the research evidence is not completely clear as to their value in detecting patients with vascular changes (Mitchell, 2007). There is still much debate in this regard (Thiel and Rix, 2005), and in the absence of a more valid alternative screening procedure it would seem somewhat premature to jettison them entirely. As discussed in the editorial, Doppler ultrasound may provide a more objective assessment of the vascular state of the vertebral artery but the promising work of Haynes (2000) has not been replicated to date. Dr. Kerry and Mr. Taylor suggest that a ‘‘general vascular risk assessment’’ would provide a better means of screening a patient prior to neck manipulation. In particular, they stress that the carotid arterial system should be evaluated as part of this procedure. While I do not have any problem with advocating a comprehensive vascular examination such as that described in their recent paper (Kerry and Taylor, 2006)), I wonder how busy clinicians will receive such time-consuming recommendations? In Australia, our experience has been that the time taken for pre-manipulative testing was a substantial disincentive for practitioners to comply with guidelines (Magarey et al., 2004). There is also little or no evidence to support the validity of the recommended vascular tests 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.09.002

(e.g. blood pressure examination, cranial nerve examination, eye examination (Kerry and Taylor, 2006)) in predicting vertebrobasilar (or even carotid) stroke following manipulation. This begs the question as to whether it makes sense to recommend alternative procedures open to the same criticism levelled at provocative positional testing? Indeed, of all the potential cardiovascular risk factors there is no evidence that any play a role in vertebral artery dissection, with the possible exception of atherosclerotic plaques (Dittrich et al., 2007). Finally, the ‘‘focus on the vertebral artery’’ to which Dr. Kerry and Mr. Taylor refer is simply a reﬂection of the fact that the overwhelming majority of strokes following neck manipulation are related to vertebral artery dissection. Notably, only about 5% of serious neurovascular complications are attributed to internal carotid artery pathology (Lee et al., 1995). References Dittrich R, Rohsbach D, Heidbreder A, Heuschmann P, Nassentein I, Bachmann R, et al. Mild mechanical traumas are possible risk factors for cervical artery dissection. Cerebrovascular Diseases 2007;23:275–81. Haynes MJ. Vertebral arteries and neck rotation: Doppler velocimeter and duplex results compared. Ultrasound in Medicine and Biology 2000;26:57–62. Kerry R, Taylor AJ. Cervical arterial dysfunction and manual therapy. Manual Therapy 2006;11:243–53. Lee KP, Carlini WG, McCormick GF, Albers GW. Neurologic complications following chiropractic manipulation: a survey of California neurologists. Neurology 1995;45:1213–5. Magarey ME, Rebbeck T, Coughlan B, Grimmer K, Rivett DA, Refshauge K. Pre-manipulative testing of the cervical spine. Review, revision and new clinical guidelines. Manual Therapy 2004;9:95–108. Mitchell J. Blood ﬂow changes associated with cervical spine rotation— implications for manual therapists. Physiotherapy 2007;93(S1):S702. Rivett DA. Adverse effects and the vertebral artery: can they be averted?. Manual Therapy 2006;11:241–2. Rivett DA, Shirley D, Magarey M, Refshauge K. Clinical guidelines for assessing vertebrobasilar insufﬁciency in the management of cervical spine disorders. Melbourne: Australian Physiotherapy Association; 2006. Thiel H, Rix G. Is it time to stop functional pre-manipulative testing of the cervical spine?. Manual Therapy 2005;10:154–8.

Darren A. Rivett School of Health Sciences, The University of Newcastle, Callaghan, NSW 2308, Australia E-mail address: [email protected]