VOLUME 14 NUMBER 4 PAGES 353–460 August 2009
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 Gwendolen Jull PhD, MPhty, Grad Dip ManTher, FACP Department of Physiotherapy University of Queensland Brisbane QLD 4072, Australia
K. Bennell (Melbourne, Australia) K. Burton (Huddersfield, UK) B. Carstensen (Frederiksberg, Denmark) J. Cleland (Concord, NH, USA) M. Coppieters (Brisbane, Australia) E. Cruz (Setubal, Portugal) L. Danneels (Maríakerke, Belgium) I. Diener (Stellenbosch, South Africa) S. Durrell (London, UK) S. Edmondston (Perth, Australia) L. Exelby (Biggleswade, UK) J. Greening (London, UK) A. Gross (Hamilton, Canada) T. Hall (Perth, Australia) W. Hing (Auckland, New Zealand) M. Jones (Adelaide, Australia) B.W. Koes (Amsterdam, The Netherlands) J. Langendoen (Kempten, Germany) D. Lawrence (Davenport, IA, USA) D. Lee (Delta, Canada) R. Lee (London, UK) C. Liebenson (Los Angeles, CA, USA) L. Maffey-Ward (Calgary, Canada) E. Maheu (Quebec, Canada) C. McCarthy (Coventry, UK) J. McConnell (Northbridge, Australia) S. Mercer (Brisbane, Australia) P. Michaelson (Luleå, Sweden) D. Newham (London, UK) J. Ng (Hung Hom, Hong Kong) S. O’Leary (Brisbane, Australia) 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) E. Rasmussen Barr (Stockholm, Sweden) D. Reid (Auckland, New Zealand) A. Rushton (Birmingham, UK) M. A. Schmitt (Amersfoort, The Netherlands) M. Shacklock (Adelaide, Australia) D. Shirley (Sydney, Australia) C. Snijders (Rotterdam, The Netherlands) P. Spencer (Barnstaple, UK) M. Sterling (Brisbane, Australia) M. Stokes (Southampton, UK) P. Tehan (Melbourne, Australia) M. Testa (Alassio, Italy) P. van der Wurff (Doorn, The Netherlands) P. van Roy (Brussels, Belgium) O. Vasseljen (Trondheim, Norway) B.Vicenzino (Brisbane, Australia) M. Wessely (Paris, France) A. Wright (Perth, Australia) M. Zusman (Perth, 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 E-mail:
[email protected] Deborah Falla PhD, BPhty(Hons) Department of Health Science and Technology Aalborg University, Fredrik BajersVej 7, D-3, DK-9220 Aalborg Denmark Email:
[email protected] Tim McClune D.O. Spinal Research Unit. University of Huddersfield 30 Queen Street Huddersfield HD12SP, UK E-mail:
[email protected] Editorial Committee Timothy W Flynn PhD, PT, OCS, FAAOMPT RHSHP-Department of Physical Therapy Regis University Denver, CO 80221-1099 USA Email:
[email protected] Masterclass Editor Karen Beeton PhD, MPhty, BSc(Hons), MCSP MACP ex officio member Associate Head of School (Professional Development) School of Health and Emergency Professions University of Hertfordshire College Lane Hatfield AL10 9AB, UK E-mail:
[email protected] Case Reports & Professional Issues Editor Jeffrey D. Boyling MSc, BPhty, GradDipAdvManTher, MCSP, MErgS Jeffrey Boyling Associates Broadway Chambers Hammersmith Broadway London W6 7AF, UK E-mail:
[email protected] Book Review Editor Raymond Swinkels PhD, PT, MT Ulenpas 80 5655 JD Eindoven The Netherlands E-mail:
[email protected] Visit the journal website at http://www.elsevier.com/math doi:10.1016/S1356-689X(09)00071-X
Manual Therapy 14 (2009) 353–354
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Editorial
The primacy of clinical reasoning and clinical practical skills
Intellectual enquiry and the desire for delivery of practice which is evidence based have seen a surge in research and an explosion of knowledge in the musculoskeletal field. There is a constantly increasing volume of research related to musculoskeletal practice. It ranges from studies designed to unravel the mechanisms underpinning pain dysfunction and functional loss, to studies evaluating clinical assessment methods, the physiological effects of interventions, the evidence for the efficacy and cost effectiveness of interventions as well as studies of prognostic indicators. In the clinical arena, clinicians in line with their responsibilities for continuing professional development are attending conferences and courses where they are exposed to information and instruction on a variety of management methods, all providing some evidence of, or claiming effectiveness. All activity aims to enhance clinical practice and the outcomes for patients. However the outcomes of this research activity to date and its synthesis for the clinical setting emphasise the critical element of practitioners’ clinical reasoning and practical skills to ensure good clinical outcomes for patients. To encourage evidence based practice, clinical practice guidelines have been developed from the research evidence to guide patient management decisions. Guidelines provide direction at a ‘first base’ or ‘in principal’ level. For example, we know that encouraging a patient to stay active after an episode of back or neck pain is better than prescribing prolonged bed rest or immobilisation in a collar. However, there is widespread recognition of the limitations of clinical practice guidelines for neck and back pain because of their generic nature in the face of the heterogeneity in patient presentation encountered in clinical practice. Evidence from the results of clinical trials which inform the clinical practice guidelines clearly demonstrate the dilemma. It is readily obvious from the numbers needed to treat or treatment effect size data that a certain treatment modality or method can be highly effective for some patients but ineffective for others, affirming that a ‘one size fits all’ approach to management of patients with neck and back pain is neither possible nor best practice. In the clinical situation, the evidence based approach of the guidelines must be applied in the context of the patient and a thorough understanding of their presenting syndrome which relies on clinician’s high level clinical reasoning, evaluation and therapeutic skills. Recognition of the heterogeneity in patient presentation, causative mechanisms and moderating factors has spurred a growing body of clinical and research work which is aimed at identifying groups of patients who have commonalities in presentation. Such subgrouping is reasoned to better direct management approaches. We are now seeing the development
1356-689X/$ – see front matter Ó 2008 Published by Elsevier Ltd. doi:10.1016/j.math.2009.05.002
of classification systems which have emanated from various perspectives, for example, from movement and motor control characteristics (Dankaerts et al., 2006; van Dillen et al., 2009), pain and movement response characteristics (Clare et al., 2005), or from pathophysiological features using a biopsychosocial framework (Sterling, 2004). An alternate approach to classification is the development of clinical prediction rules which define the primary clinical features of patients who are likely respond to a certain intervention (Tsenga et al., 2006; Cleland et al., 2007; Vicenzino et al., 2008). These classification systems and clinical prediction rules require further validation but have the potential to improve guidelines for the management for the subgroups identified. However on the downside, these classification systems will be relevant for a certain percentage of patients and they do not inform on management of patients who fall outside the classification or clinical rule. Successful outcomes for these patients rely even more heavily on the practitioner’s clinical reasoning and practical skills. A plethora of treatment approaches for musculoskeletal disorders such as neck and back pain are still undergoing scientific scrutiny and this reflects the absence of conclusive evidence for one approach. It confirms the heterogeneity in patient presentation and clinical expectations that some approaches or techniques, even in the spirit of evidence informed practice, stand to be efficacious for some but not all patients. It is the practitioner’s clinical reasoning, assessment and clinical practical skills that are a crucial nexus between the patient, the research evidence and successful clinical outcomes. It might not be too incorrect to assert that the evidence will lose its impact for enhanced patient care in the absence of high level clinical reasoning and practical skills in the practitioner. We are strong advocates for research informed practice, but assert that the fundamental clinical skills of the practitioner should never be undervalued and must attract equal attention in education at professional and post professional levels. References Dankaerts W, O’Sullivan P, Straker L, Burnett A, Skouen J. The inter-examiner reliability of a classification method for non-specific chronic low back pain patients with motor control impairment. Manual Therapy 2006;11:28–39. Clare H, Adams R, Maher C. Reliability of McKenzie classification of patients with cervical or lumbar pain. Journal of Manipulative and Physiological Therapeutics 2005;28:122–7. Cleland J, Childs J, Fritz J, Whitman J, Eberhart S. Development of a clinical prediction rule for guiding treatment of a subgroup of patients with neck pain: use of thoracic spine manipulation, exercise, and patient education. Physical Therapy 2007;87:9–23.
354
Editorial / Manual Therapy 14 (2009) 353–354
Sterling M. A proposed new classification system for whiplash associated disorders. Manual Therapy 2004;9:60–70. Tsenga Y-L, Wanga W, Chena W-Y, Houb T-J, Chenc T-C, Lieuc F-K. Predictors for the immediate responders to cervical manipulation in patients with neck pain. Manual Therapy 2006;11:306–15. van Dillen L, Maluf K, Sahrmann S. Further examination of modifying patientpreferred movement and alignment strategies in patients with low back pain during symptomatic tests. Manual Therapy 2009;14:52–60. Vicenzino B, Smith D, Cleland J, Bisset L. Development of a clinical prediction rule to identify initial responders to mobilisation with movement and exercise for lateral epicondylalgia. Manual Therapy 2008. doi:10.1016/j.math.2008. 08.004.
Gwendolen Jull* Ann Moore NHMRC Centre for Clinical Research Excellence -Spine, School of Health and Rehabilitation Sciences, The University of Queensland, Queensland 4072, Australia Corresponding author. Tel.: þ61 7 3365 1114; fax: þ61 7 3365 1622. E-mail address:
[email protected] Manual Therapy 14 (2009) 355–362
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Systematic Review
The reliability and validity of assessing medio-lateral patellar position: a systematic review Toby O. Smith a, *, Leigh Davies a, Simon T. Donell b a
Orthopaedic Physiotherapy Research Unit, Physiotherapy Department – Out-Patients East, Norfolk and Norwich University Hospital, Colney Lane, Norwich, NR4 7UY, UK b Faculty of Health, University of East Anglia, Norwich, NR4 7TJ, UK
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 May 2008 Received in revised form 18 July 2008 Accepted 2 August 2008
Medio-lateral patellar position is regarded as a sign of patellofemoral pain syndrome and patellar instability. Its assessment is important in accurately performing patellofemoral therapeutic taping techniques. The purpose of this paper is to systematically review the literature to determine the reliability and validity of evaluating medio-lateral patellar position. An electronic database search was performed accessing AMED, British Nursing Index, CINAHL, the Cochrane database, EMBASE, Ovid Medline, Physiotherapy Evidence Database (PEDro), PubMed and Zetoc to July 2008. Conference proceedings and grey literature were also scrutinised for future publications. All human subject, clinical trials, assessing the inter- or intra-tester reliability, or the criterion validity, were included. A CASP tool was employed to evaluate methodological quality. Nine papers including 237 patients (306 knees) were reviewed. The findings of this review suggest that the intra-tester reliability of assessing medio-lateral patellar position is good, but that inter-tester reliability is variable. The criterion validity of this test is at worse moderate. These are based on a limited evidence-base. Further study is recommended to compare the McConnell (1986) [McConnell J. The management of chondromalacia patellae: a long term solution. Australian Journal of Physiotherapy 1986;32(4):215–23] and Herrington (2002) [Herrington LC. The inter-tester reliability of a clinical measurement used to determine the medial/lateral orientation of the patella. Manual Therapy 2002;7(3):163–7] methods of assessing medio-lateral patellar position in patients with well-defined patellofemoral disorders. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Assessment Patellar position Reliability Validity
1. Introduction The aetiology of patellofemoral pain syndrome (PFPS) and patellar instability are multi-factorial (Sutlive et al., 2004). One factor indicated in both patellofemoral disorders is abnormal patellar tracking. The patella in said to be frequently lateralised in both disorders (Mizuno et al., 2001). It is hypothesised that this can cause an increase in retropatellar pressure over the articular surfaces, contributing to articular cartilage degeneration and subsequent pain (Powers et al., 1999; Ota et al., 2006; Fulkerson and Shea, 1990; Hughston, 1968; Insall, 1979). Similarly, patellar maltracking within the femoral sulcus can cause instability symptoms, and predispose the patella to dislocation (Arendt et al., 2002). Taping is one physiotherapeutic strategy aimed at correcting patellar mal-tracking (Warden et al., 2008). This has gained widespread acceptance as a viable treatment option for patients with
* Corresponding author. Tel.: þ44 1603 286990; fax: þ44 1603 287369. E-mail address:
[email protected] (T.O. Smith). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.08.001
PFPS and patellar instability (Powers et al., 1999; McConnell, 1986, 2007). The patella is taped specifically to address the individual’s abnormal glide, rotation and tilt, and to maintain the patellar correctly within the femoral trochlea throughout range (Warden et al., 2008; Crossley et al., 2001). Since the medial and lateral translation have been associated with PFPS and patellar instability, it is important that the extent and direction of such translation can be accurately assessed. The medial and lateral displacement of the patellar can be measured by two means (Figs. 1 and 2). McConnell (1986) first described assessing this through palpation and visual estimation. She suggested that both the medial and lateral femoral epicondyles should be palpated and identified with both index fingers. The midpatellar point should then be recognised using both thumbs. In normal cases, the distance between index fingers and thumbs will be approximately equal. If the patella is laterally displaced, the distance between the index finger palpating the lateral epicondyle to thumb, will be less than the other fingers measuring the medial patellar position. This is reversed for medial displacement (McConnell, 1986). More recently, Herrington (2002) has assessed
356
T.O. Smith et al. / Manual Therapy 14 (2009) 355–362
a review of the literature suggested that such an evaluation had yet to be performed.
2. Methodology 2.1. Study eligibility criteria
Fig. 1. The McConnell (1986) approach for the assessment of medio-lateral patellar position.
this distance by marking the epicondyles and mid-patellar position on zinc tape and measuring the medial and lateral distance with a tape measure. The validity of a test is the extent to which a test measures what it is intended to measure (Edwards and Talbot, 1994; Polgar and Thomas, 2000). One aspect of validity is criterion validity. This assesses how the test under investigation compares against an established or gold-standard measure (Evans et al., 2004). In the case of patellar position, such gold-standard tests would include magnetic resonance imagery (MRI), computed tomography (CT) or plain radiographs (Grelsamer et al., 1998; Herrington, 2006; Tolouei et al., 2005). Reliability is the extent to which a test is reproducible (Polgar and Thomas, 2000; Edwards and Talbot, 1994). This is subcategorised into two types. Inter-tester reliability assesses the degree to which different examiners give consistent estimates of the same test (Portney and Watkins, 2000). Intra-tester reliability assesses the consistency of a measure on two different occasions (Polgar and Thomas, 2000; Portney and Watkins, 2000). The objective of this study is to systematically review the evidence-base to determine the inter- and intra-tester reliability and criterion validity of medio-lateral patellar position. This has considerable clinical importance given that this assessment forms the basis of therapeutic taping of the patellofemoral joint, a widely used and acceptable intervention in clinical practice for patients with patellofemoral disorders. This is further justified since
Fig. 2. The Herrington (2002) approach for the assessment of medio-lateral patellar position.
The inclusion criteria included all full text papers assessing medio-lateral patellar position by two or more examiners, at one or more time points (inter- or intra-tester reliability). Papers comparing the clinical assessment of medio-lateral patellar position to a radiological assessment using MRI, CT or plain radiograph (criterion validity) were also included. Papers of any language were included, as well as unpublished material including university theses and dissertations and conference proceedings, in an attempt to limit publication bias from impacting on this systematic review’s findings. Papers were excluded if they presented insufficient data on their method of assessing medio-lateral patellar position. Single-subject case reports, comments, letters, editorials, protocols, guidelines, or review papers were excluded. The reference lists of review papers were scrutinised for any clinical papers deemed relevant to the research question. No exclusion was made to subject age or gender. Animal and cadaver studies were excluded.
2.2. Search strategy The primary search was a search of the electronic databases AMED, British Nursing Index, CINAHL, the Cochrane database, EMBASE, Ovid Medline, Physiotherapy Evidence Database (PEDro), PubMed and Zetoc from their inception to July 2008. Key terms and Boolean operators adopted included: patella AND position; orientation. A secondary search of the following specialist journals was undertaken: Knee Surgery Sports Traumatology Arthroscopy (1993– July 2008), The Knee (1994–July 2008), the British and American editions of the Journal of Bone and Joint Surgery (1988–July 2008), American Journal of Sports Science (1988–July 2008), and Journal of Orthopaedic Sports and Physical Therapy (1991–July 2008). Unpublished or grey literature was assessed using the databases SIGLE (System for Information on Grey Literature in Europe), the National Research Register (UK), the National Technical Information Service, the British Library’s Integrated Catalogue, and Current Controlled Trials database for recently completed trials. Conference proceedings from the British Orthopaedic Association Annual Congress and British Association for Surgery of the Knee were searched from 2002 to 2008, for additional studies pertaining to this research question. Using the predefined eligibility criteria, two investigators (TS, LD) independently assessed all identified titles and abstracts. Full manuscripts of citations adhering to the criteria were ordered. Full manuscripts were ordered of those citations the reviewers were uncertain about after reading the abstracts. Reference lists from each full manuscript were scrutinised to identify any publications not initially identified. Each full text was then screened against the eligibility criteria by the same two reviewers. In cases of disagree, a census was reached through discussion. No paper was excluded on poor methodological quality. The two investigators were not blinded to the source or authors of the papers reviewed. The corresponding author of each paper included in the review was then contacted. They were asked whether they knew of any additional papers which had not been identified by the search strategy, to ensure that every paper potentially answering this research question, had been considered in this systematic review.
T.O. Smith et al. / Manual Therapy 14 (2009) 355–362
357
Articles recovered from the search strategy. (n= 1425)
Title or abstract not pertaining to the research question. (n= 1328) Abstracts which appeared relevant to the research question (n=97) Articles deemed not related to the research question after consulting the full abstract. (n= 62) Appropriate studies related to the research question, permitting full manuscripts to be ordered for further scrutiny. (n= 35)
Articles excluded as not adhering to the eligibility criteria. (n= 26) Appropriate studies related to the research question, and adhering to the eligibility criteria. (n =9) Articles excluded due to replication of] data presented. (n = 0) Final included articles. (n =9) Fig. 3. A QUORUM flow-chart.
2.3. Data extraction Data from all studies fully satisfying the eligibility criteria was entered into a spreadsheet by a single investigator (TS), and verified by a second investigator (LD). This spreadsheet tabulated:
Author names and publication date Study design Sample size Population characteristics including diagnosis, subject age and gender Method of assessing Tester details including number of tester, frequency of testing, experience of tester, teaching of tester to the measurement procedure Method of reference test for criterion validity Statistical analysis Results Any relevant methodological limitations
2.4. Critical appraisal All included papers were evaluated against an appraisal based on the Critical Appraisal Skills Programme (CASP, 2007) appraisal tool for diagnostic test studies. This appraisal tool comprises of three sections: an assessment of study validity; an evaluation of methodological quality and presentation of results; an assessment of external validity. Each paper was assessed independently by two reviewers (TS, LD). Any differences in appraisal results were settled
through discussion. There was a difference between the reviews over six items, in two papers (Tomsich et al., 1996; Fitzgerald and McClure, 1995), which was resolved by discussion. 3. Results 3.1. Search results Fig. 3 outlines the results of the search strategy. The search yielded 1425 articles whose titles and abstracts were read. Of these, 9 studies met the eligibility criteria, these are summarised in Tables 1 and 2. The papers have been subdivided by their study aims and discussed below. A total of 213 patients and 282 knees were reviewed. This included 29 patients (37 knees) diagnosed as PFPS, and 104 patients (164 knees) asymptomatic control subjects. One paper did not specify the pathology of its sample, whilst 76 patients (77 knees) were collectively assessed as general knee pathologies. Eighty-four males, and 137 females were reviewed, whilst three studies did not specify the gender of their cohorts. In those with patellofemoral disorders, the age of subjects ranged from 18 to 41 years, with a mean of 30.6 years. This differed from the asymptomatic samples where age ranged from 18 to 28 years, with a mean of 24.2 years. As expected, the method for assessing medio-lateral patellar position was made using two techniques. The majority of studies assessed patellar position following McConnell’s (1986) method. Only Tomsich et al. (1996) assessed patellar orientation differently using visual estimation or pluri-cal callipers. There was some variability in assessment position using this method. McEwan et al. (2007), Herrington (2008), Herrington (2002), Herrington et al.
358
T.O. Smith et al. / Manual Therapy 14 (2009) 355–362
Table 1 Summary of the papers included in this systematic review. Study Design Sample size Population
Medial/lateral position test
Reliability assessment Validity assessment Tester details
Statistical analysis Study Design Sample size Population
Medial/lateral position test
Reliability assessment Validity assessment Tester details Statistical analysis Study Design Sample size Population Medial/lateral position test
Reliability assessment Validity assessment
Tester details
Statistical analysis Study Design Sample size Population
Fitzgerald and McClure (1995) Observational 66 (66 knees) 66 symptomatic subjects; 31 males, 35 females; mean age 29.7 13.1 years (range 14–74); mean weight 73.4 19.6 kg; mean height 171.2 10.2 cm 40 diagnosed with patellar pain syndrome, anterior knee pain, chondromalacia patellae, subluxing patellar, patellar tendonitis or patellar fracture 26 diagnosed with meniscal pathology, ligamentus pathology, femoral or tibia fracture Subjects excluded if they had received a surgical procedure specifically to realign the patella (e.g. lateral retinacular release) McConnell assessment: subject supine, tibiofemoral joint in full extension, quadriceps contraction or lower limb rotation not documented. Palpation of the medial and lateral femoral epicondyles with the index finger and simultaneously palpating the mid-patellar with the thumbs. Normally. Distance between the fingers and thumb should be equal. If a lateral displacement is present, the distance between the index finger palpating the lateral epicondyle to the thumbs will be less than the distance from the fingers palpating the medial epicondyle to the thumbs. If a medial displacement is evident, the distance between the medial epicondyle to the thumbs will be less than the distance from the lateral eipcondyle to the thumbs Subjects were independently assessed once, by 2 different examiners, most often during the same clinic session Not assessed 12 physical therapists from 4 clinics who frequently treat patients with knee or patellofemoral disorders. All testers were familiar with the patellar orientation test prior to participation in the study. One tester had learnt the patellar orientation test from McConnell’s course. The other tester had learnt the test from colleges or from reading texts on the method of assessment. For this study, each tester received a written and photographic description of how to perform the tests based on McConnell (1986). This was provided approximately 2 weeks prior to testing Inter-tester reliability, kappa Herrington (2008) Observational, matched pairs 24 (24 knee) 12 asymptomatic subjects; 12 females; mean age 21.6 2.8 years (range 18–25); mean body mass 62.3 8.4 kg 12 subjects diagnosed with patellofemoral pain syndrome for at least 1 month; 12 females; mean age 21.6 2.6 years (range 18–25); mean body mass 64.5 9.3 kg Subjects were excluded if they had reported previous knee surgery or arthritis, history of patellar dislocation, subluxation or ligament laxity, patellar tendonopathy, chondral damage, spinal referred pain, lower limb abnormalities such as leg length discrepancy (>2 cm), were taking medication as part of their knee treatment, or had received previous knee physiotherapy and acupuncture treatment within the previous 30 days McConnell (1986) and Herrington and Nester (2004) assessment: subject general position, quadriceps contraction or lower limb rotation not documented, knee in 20 degrees flexion. Centre of patella, medial and lateral femoral epicondyle marked on a piece of folded zinc oxide tape placed on subject’s knee. Distance between the medial epidcondyle to mid-point of patella, and lateral epicondyle to mid-point of patella measured .. Not stated what used to measure distances. An assessor marked these distances on the zinc oxide tape, and a second assessor blinded to diagnosis, measured the distance between the markings. This whole procedure was repeated 3 times, with the average of the 3 measurement recorded This procedure was then repeated to assess the 12 matched control subjects on a separate occasion 1–2 days after the original measurement Not assessed The physiotherapist identifying the relevant bony landmarks and marking the zinc oxide tape was an experienced physical therapist. Details for the independent assessor blinded to diagnosis, who measured the markings, was not documented Intra-tester reliability, ICC AND SEM Herrington (2002) Observational 1 (1 knee) 1 subjects; pathology, gender, age not specified McConnell (1986) assessment: subject general position, quadriceps contraction or lower limb rotation not documented, knee in 20 degrees flexion. Centre of patella, medial and lateral femoral epicondyle marked on a piece of folded zinc oxide tape placed on subject’s knee. Distance between the medial epidcondyle to mid-point of patella, and lateral epicondyle to mid-point of patella measured .. Not stated what used to measure distances. Each assessor repeated this procedure 3 times each, re-palpating and applying tape on each occasion. Average of the 3 measurement recorded This procedure was then repeated to assess the subject on 2 separate occasions. Not specified how long duration was between assessments MRI assessment of medial/lateral patellar position with patient supine, in 20 degrees knee flexion, no details on limb rotation. Medio-lateral position determined by assessing lateral patellar displacement in relation to femoral condyles. LPD measured 3 times, with average taken. Not specified which MRI slice used to assess LPD Clinical test: medio-lateral patella orientation measured by 20 chartered physiotherapists, experience musculoskeletal physiotherapists at MACP examination approved level, and a minimum of 5 years specialising in musculoskeletal physiotherapy. Details for this assessor on experience of this test or of teaching of this technique were not detailed. MRI test: all made from one investigator blinded to the clinical examination findings Inter-tester reliability and criterion validity, means of the ICC
Reliability assessment Validity assessment Tester details Statistical analysis
Herrington and Nester (2004) Observational 10 (20 knees) 10 asymptomatic, gender, age not specified Asymptomatic described as physically active asymptomatic individuals with no history of lower limb, spinal or neurological injury Herrington (2002) assessment: subject general position, quadriceps contraction or lower limb rotation not documented, knee in 20 degrees flexion. Centre of patella, medial and lateral femoral epicondyle marked on a piece of folded zinc oxide tape placed on subject’s knee. Distance between the medial epidcondyle to mid-point of patella, and lateral epicondyle to mid-point of patella measured .. Not stated what used to measure distances. Each assessor repeated this procedure 3 times each, re-palpating and applying tape on each occasion. Average of the 3 measurement recorded This procedure was then repeated to assess the subjects on 2 separate occasions. Not specified how long duration was between assessments Not assessed Not documented Intra-tester reliability, ICC and SEM
Study Design
Herrington et al. (2006) Observational
Medial/lateral position test
T.O. Smith et al. / Manual Therapy 14 (2009) 355–362
Sample size Population Medial/lateral position test
Reliability assessment Validity assessment Tester details Statistical analysis Study Design Sample size Population Medial/lateral position test
Reliability assessment Validity assessment
Tester details
Statistical analysis Study Design Sample size Population
Medial/lateral position test
Reliability assessment Validity assessment Tester details Statistical analysis Study Design Sample size Population Medial/lateral position test
Reliability assessment Validity assessment Tester details Statistical analysis Study Design Sample size Population
359
5 (5 knees) 5 asymptomatic subjects; males/females not specified; mean age not specified Asymptomatic described as physically active asymptomatic individuals McConnell (1986) and Herrington (2002) assessment: subject general position, quadriceps contraction or lower limb rotation not documented, knee in 20 degrees flexion. Centre of patella, medial and lateral femoral epicondyle marked on a piece of folded zinc oxide tape placed on subject’s knee. Distance between the medial epidcondyle to mid-point of patella, and lateral epicondyle to mid-point of patella measured .. Not stated what used to measure distances. 1 assessor repeated this procedure 3 times each, re-palpating and applying tape on each occasion. Average of the 3 measurement recorded This procedure was then repeated to assess the subjects on 2 separate occasions. Not specified how long duration was between assessments Not assessed Clinical test: all measures taken by 1 assessor, Details for this assessor on experience of this test or of teaching of this technique was not detailed Intra-tester reliability, means of the ICC and SEM McEwan et al. (2007) Observational 24 (24 knees) 24 asymptomatic subjects; 16 males, 8 females; mean age 24.5 7.9 years, range 18–42 years Asymptomatic consist of no current or previous history of knee or lower extremity injury McConnell (1986) and Herrington and Nester (2004) assessment: subject general position, quadriceps contraction or lower limb rotation not documented, knee in 20 degrees flexion. Centre of patella, medial and lateral femoral epicondyle marked on a piece of folded zinc oxide tape placed on subject’s knee. Distance between the medial epidcondyle to mid-point of patella, and lateral epicondyle to mid-point of patella measured Not stated with what measured. 2 assessors repeated this procedure 3 times each, re-palpating and applying tape on each occasion. Average of the 3 measurement recorded This procedure was then repeated 1 day later to assess the subjects on 2 separate occasions MRI assessment of medial/lateral patellar position with patient supine, in 20 degrees knee flexion, no details on limb rotation. Medio-lateral position determined by assessing lateral patellar displacement in relation to femoral condyles. LPD measured 3 times, with average taken. Not specified which MRI slice used to assess LPD Clinical test: all measures taken by 2 independent assessor. Details for 1 assessor provided as a musculoskeletal physiotherapist with 15 years experience, not specified how long undertaken this testing procedure, or details of teaching of this technique to the assessor. MRI test: all made by one investigator blinded to the clinical examination findings Intra-tester reliability, means of the ICC criterion validity, Pearson’s product moment Powers et al. (1999) Observational 24 (38 knees) Intra-tester assessment: 10 subjects (20 knees) asymptomatic; 4 males, 6 females; mean age 26 2 years. Asymptomatic consisted of pain-free status, not specified of what Criterion validity: 4 subjects (7 knees) asymptomatic – unspecified criteria. 10 subjects (11 knees) symptomatic – described as either tibiofemoral joint osteoarthritis, anterior knee pain, meniscal injury – not specified how these diagnoses were determined. In total: 10 females, 4 males; mean age 41 16 years McConnell (1986) assessment: subject supine, knee extended, quadriceps relaxed, limb rotation not documented. Centre of patellar determined and marked, with a soft tape measure, and bisecting the distance between the most medial and lateral borders of the patella. Distance from centre of patellar to medial femoral epicondyle, and from centre of patella to lateral femoral epicondyle was then assessed using the tape measure 4 measurements of medial/lateral position median and averaged, 2 for medial and 2 for lateral femoral epicondyle distance. Measurements made on 2 separate occasions at least 2 weeks apart MRI assessment of medial/lateral patellar position with patient supine, full extension with quadriceps relaxed, in natural limb rotation for each patient (10–15 degrees external rotation). The image containing the largest patellar cross-section (mid-patellar slice) was used for analysis Clinical test: all measures taken by 1 assessor, who had less than 1 year’s experience of this technique. MRI test: all made by one investigator and measured according to procedure Intra-tester reliability, ICC criterion validity, ANOVA Tomsich et al. (1996) Observational 27 (27 knees) 27 asymptomatic subjects; 7 males, 20 females; mean age 21 5.5 years Asymptomatic defined as no history of knee pathology Assessed by visual estimation and pluri-cal calliper, each subject in supine position, in 0 degrees tibiofemoral flexion, quadriceps relaxed, foot position to maintain leg in neutral rotation using a KT-100 foot stabilizer. (1) Visual estimation of medio-lateral position assessed by positioning the index fingers and thumbs on the sides of the femoral epicondyles and over the patellar midpoint. The distance between the index finger and thumb medially and laterally is an estimate. (2) Medio-lateral position assessed by callipers by positioning the callipers along both femoral epicondyles, the midpoint marker on the ruler of the pluri-cal callipers was placed over the patellar midpoint. The distance between the epicondyle to the patellar midpoint, medially and laterally, was then recorded Subjects assessed 3 times each, by the three different examiners. Assessors were blinded to subject’s identification Not assessed 3 physical therapists who spent a total of 2 h practicing the measurement procedure before the study. Therapist’s ages ranged from 25 to 27 with 2.5 to 5.5 years experience of orthopaedic practice, graduating from 3 different physiotherapy schools Intra- and inter-tester reliability, kappa, ICC and SEM Watson et al. (1999) Observational 56 (101 knees) 56 subjects; 22 males, 34 females; mean age 29 8.0, range 22–34 years 39 (76 knees) asymptomatic subjects mean age 28 6.2 17(25 knees) symptomatic subjects mean age 30 11.4 Asymptomatic defined as no knee pain or pathology Symptomatic defined as patellofemoral pain classified as patellofemoral joint pain reproducible following at least 2 of the following activities in the last month: ascending or descending stairs, prolonged sitting, squatting Subjects excluded if they had a history of knee surgery or patellar dislocation, or if there was evidence on clinical examination of knee ligamentus injury, meniscal injuries, patellar tendonopathy, major joint effusion or plica syndrome (continued on next page)
360
T.O. Smith et al. / Manual Therapy 14 (2009) 355–362
Table 1 (continued) Medial/lateral position test
Reliability assessment Validity assessment Tester details Statistical analysis
McConnell assessment: subject supine, tibiofemoral joint in full extension and femur parallel to the plinth, quadriceps relaxed through lower limb rotation not documented. Distance measured using a tape measure from the mid-patella to the medial femoral epicondyle and to the lateral femoral epicondyle. The midpoint was determined on visual estimation, and marked with a grease pencil. A score of 0 was allocated if the distance from the medial epicondyle to the mid-patella point was equal to the distance from the lateral epicondyle to mid-patella; 1 was allocated if the distance from the medial epicondyle to the mid-patella point was >0.5 cm from the lateral epicondyle to mid-patella Subjects assessed once, by 2 different examiners, on 2 separate occasions, 3–7 days after the initial measure. Assessors were blinded to whether subjects had knee pain Not assessed 2 senior physical therapy students who had received approximately 2 h of didactic instructions and 2 h of practice on the McConnell patellofemoral classification system of medio-lateral position Intra- and inter-tester reliability, kappa
ANOVA, analysis of variance; ICC, intra-class correlation coefficient; LPD, lateral patellar displacement; MACP, Manipulation Association of Chartered Physiotherapists; MRI, magnetic resonance imaging; SEM, standard error of the measurements.
(2006) and Herrington and Nester (2004) assessed the knee in 20 degrees flexion, whereas Powers et al. (1999), Tomsich et al. (1996), Fitzgerald and McClure (1995) and Watson et al. (1999) assessed the knee in full extension. Lower limb rotation was only documented as controlled in Tomsich et al.’s (1996) study in neutral. Only Watson et al. (1999), Tomsich et al. (1996), and Powers et al. (1999) acknowledged that the quadriceps muscles were relaxed during testing.
3.2. Reliability Intra-tester reliability of medio-lateral patellar position as assessed in seven studies. Six studies reported either substantial or near perfect agreement between assessment periods, with intraclass coefficient (ICC)/kappa results ranging from 0.70 to 0.99. Four studies all reported almost perfect agreement between test procedures in McEwen et al. (2007), Powers et al. (1999), Herrington et al. (2006), and Herrington and Nester’s (2004) results. Only Watson et al.’s (1999) study reported poor to fair agreement with 0.11–0.35 kappa results. Four studies assessed inter-tester reliability. These studies reported differing results. Two studies reported near perfect agreement in Herrington (2002) results with 0.91 and 0.94. In contrast, Tomsich et al. (1996), Fitzgerald and McClure (1995) and Watson et al. (1999) reporting poor agreement with results of 0.14, 0.10 and 0.02 respectively.
3.3. Criterion validity The criterion validity of assessment of medio-lateral patellar orientation was evaluated in studies by Herrington (2002), McEwan et al. (2007) and Powers et al. (1999). These studies reported variable agreement between clinical medio-lateral patellar position assessment and MRI evaluation. Herrington (2002) reported near perfect agreement between the measures with an ICC of 0.9. McEwan et al. (2007) reported substantial agreement with an ICC of 0.61, whilst Powers et al. (1999) reported moderate agreement with an ICC of 0.44. 3.4. Critical appraisal results The findings of the CASP appraisal are presented in Table 3. These suggest that the methodological quality of the papers was limited in a number of areas. The CASP review highlighted that all studies stated appropriate research questions and applied suitable study designs to answer their research questions. As Table 1 outlines, three studies used a references test (MRI) to assess the criterion validity of the medio-lateral patella position. Only McEwan et al. (2007) stated that assessors were blinded to the results of this test, whilst Herrington (2002) indicated that different assessors were used for their MRI and clinical findings. Population characteristics such as patient’s knee history and pathology, gender, age or weight and height were poorly described in five papers. Whilst all papers identified the basic method of assessing medio-lateral patellar
Table 2 Summary of results from studies included in this systematic review. Author (date)
Mean clinically assessed medio-lateral position in mm (SD)
Mean reference test medio-lateral position (SD)
Inter-tester reliability (ICC with SEM)
Intra-tester reliability (ICC with SEM)
Criterion validity ICC (p value)
Fitzgerald and McClure (1995) Herrington (2008)
N/D
N/A
0.10 (44%)a
N/A
N/A
PFPS group: lateral 7.5 (2.6); asymptomatic group: lateral 3.8 (2.4) Medial 8.98 (0.51); lateral 8.35 (0.66)
N/A
N/A
0.86, SEM 0.2 mm
N/A
LPD 5.0(2.8)
N/A
0.9
3 mm (6 mm) lateralised from central Medial females 2.5 (1.8)/males 2.3 (1.8); lateral females 6.6 (4.5)/males 5.9 (4.5) Medial 8.9 (0.1); lateral 8.3 (0.1) 6.8 9.6% lateral displacement as percentage of patella width
N/A
Medial measure 0.91; lateral measure 0.94 N/A
0.99 (p < 0.01), SEM 6 mm
N/A
N/A
N/A
0.99 (p < 0.015), SEM 0.1 mm
N/A
LPD 8.1 (2.8) 16.1 12.3% lateral displacement as percentage of patella width N/D N/A
N/A N/A
0.86 (0.1, CI 8.1–8.5) 0.91
0.61 (p ¼ 0.002) 0.44
0.14 (0.55) 0.02 (70%)a
0.70 (0.28) 0.11 (8.4) to 0.35 (0.74)a
N/A N/A
Herrington (2002)
Herrington et al. (2006) Herrington and Nester (2004) McEwan et al. (2007) Powers et al. (1999)
Tomsich et al. (1996) Watson et al. (1999)
N/D Frequency between testers of: score 0, 135–149; score 1, 47–53
ICC, intra-class correlation co-efficient; N/D, not documented; mm, millimetres; SD, standard deviation; N/A, not assessed; SEM, standard error of the measurements. a Kappa coefficient (percentage of agreement).
T.O. Smith et al. / Manual Therapy 14 (2009) 355–362
361
Table 3 A summary of the CASP results. CASP factors
Fitzgerald and McClure (1995)
Herrington Herrington Herrington (2008) (2002) et al. (2006)
Herrington and Nester (2004)
McEwan Powers Tomsich Watson et al. (2007) et al. (1999) et al. (1996) et al. (1999)
Clearly focused question stated Appropriate design Appropriate reference test available Did all receive reference and diagnostic test Could reference test findings influence diagnostic test result Population characteristics clearly defined Diagnostic test clearly defined Appropriate results analysis Precise statistical results presented Appropriate interpretation Ability to generalise results Were the results applicable to clinical practice
Y Y N
Y Y N
Y Y Y
Y Y N
Y Y N
Y Y Y
Y Y Y
Y Y N
Y Y N
N
N
Y
N
N
Y
Y
N
N
N/A
N/A
N
N/A
N/A
N
Y
N/A
N/A
N
Y
N
N
N
Y
Y
Y
Y
N Y N
N Y N
N Y N
N Y N
N Y N
N Y Y
N Y N
Y Y N
N Y N
Y Y Y
Y Y Y
Y N N
N N N
N N N
Y N N
Y Y Y
N N N
Y Y Y
Y, yes; N, no; N/A, not applicable.
position, all studies except Tomsich et al. (1996) did not clearly state either the position of the knee or lower limb, or whether the quadriceps were relaxed or contracted, confounding variables in assessing medio-lateral patellar position. The evidence-base used appropriate statistical tests with ICC and kappa analysis, but only McEwan et al. (2007) documented confidence intervals to assess the precision of their statistical findings. Since only Fitzgerald and McClure (1995), Powers et al. (1999), Watson et al. (1999) and Herrington (2008) recruited patients with patellofemoral pathology, only these studies were regarded as having any clinical significance to be generalisable to the clinic setting. 4. Discussion The findings of this review suggest that the intra-tester reliability of medio-lateral position tests is good, but that inter-tester reliability is variable. The criterion validity of these tests is at worst moderate. However, such conclusions should be interpreted with caution since the evidence-base presently exhibits a number of limitations. The two most important weaknesses identified were the poor documentation of the actual medio-lateral patella positioning test methods and the limited description of subject characteristics. Two distinct methods of assessing medio-lateral patellar position were recognised (Figs. 1 and 2). This review would suggest that Herrington’s (2002) methods appears to have better inter-tester reliability and criterion validity than McConnell’s (1986). This difference may account for the substantial difference in intra-tester results between the findings of Watson et al. (1999) and Herrington et al. (2006) and Herrington and Nester (2004), or inter-tester findings between Tomsich et al. (1996) and Herrington (2002). However, this cannot be categorically stated given the limited size of evidence presently available. Furthermore, with the exception of 12 subjects in Herrington’s (2008) study, all other studies which assessed Herrington’s (2002) method were undertaken on asymptomatic populations. As a result, it is not possible to generalise with confidence, these results to patients with patellofemoral disorders. Accordingly, a direct comparison of these two methods of assessing medio-lateral position is warranted to determine the optimal method of assessing patients with different patellofemoral disorders. A considerable weakness in the evidence-base is the poor description of the medio-lateral patellar position test. Studies did not demonstrate whether they controlled confounding factors such
as lower limb rotation, quadriceps contraction or knee flexion (McEwan et al., 2007; Herrington et al., 2006; Herrington, 2002, 2008; Herrington and Nester, 2004; Fitzgerald and McClure, 1995). These variables can influence the patellar position within the trochlear groove (Herrington and Pearson, 2008; Muhle et al., 1999). Variations in these factors between study methodologies may account for the differences in results between studies for intertester reliability. This should be considered when presenting the findings of similar studies in the future. The degree of knee flexion may be an important factor. Five papers (McEwan et al., 2007; Herrington et al., 2006; Herrington, 2002, 2008; Herrington and Nester 2004) assessed the knee in 20 degrees flexion, compared to full extension. Considering the patella engages with the femoral trochlear at approximately 10–30 degrees of knee flexion (Beasley and Vidal, 2004; Senavongse et al., 2003), it may be hypothesised that the good reliability results of these studies may be related to greater patellar osseous constraint compared to full extension. Further study may therefore be indicated to investigate this assumption assessing medio-lateral patellar position in full extension compared to difference ranges of tibiofemoral flexion. The papers reviewed poorly documented important subject characteristics, particularly in view of weight and height. Papers poorly distinguished between those subjects with patellar instability, patellofemoral pain syndrome or other patellofemoral disorders. Consequently, it was not possible to ascertain the heterogeneity of these papers. In response to this, it was deemed unstable to formally compare these paper’s results using a metaanalysis design. In addition, Egger et al. (2001) suggested that meta-analysis should not be undertaken for observational studies, which was the methodology design used in all the studies reviewed. It appeared that only Fitzgerald and McClure (1995) assessed medio-lateral patellar position in patients with subluxed patellar. The present evidence-base does not state whether the reliability or validity of this test is related to the severity of patellar displacement. Further study is recommended to investigate whether the accuracy of these measurements is related to the degree of patellar displacement in well-defined populations. With the exception of Herrington (2008), Herrington and Nester (2004) and Herrington et al. (2006), the assessors’ experience of the medial-lateral patellar position tests was well described. The assessors in each study appeared to have broadly similar levels of experience and training in each assessment method. However, it remains unclear whether the reliability or validity of the tests to
362
T.O. Smith et al. / Manual Therapy 14 (2009) 355–362
assess medio-lateral position is dependent on the method of teaching these measures. This may inform clinicians as to whether attendance at courses is warranted, or whether text and photographic teaching is sufficient to accurately assess this measurement. The results of this review would suggest that the intra-tester reliability of medio-lateral patellar orientation tests are satisfactory. Accordingly, clinicians can have some confidence that they have consistency in their measures between treatment sessions. It remains unclear whether the results of this measure are reproducible if patients change from one physiotherapist to another with such variable inter-tester reliability. Furthermore, further study is indicated to determine the clinical relevance of these findings. The reliability of physiotherapist’s abilities to translate their mediolateral displacement findings to specific taping placement to address abnormal patellar translation is presently unclear and should be addressed with future study. Although studies frequently described the age and gender of their subjects, only Fitzgerald and McClure (1995) and Herrington (2008) detailed the weight of their subjects to indicate body mass. Body mass may have influenced medio-lateral patellar position measurements. Fitzgerald and McClure (1995) acknowledged that it appeared more difficult to palpate the bony landmarks in those subjects with greater body mass. Accordingly, this may have contributed to variability in the measurement of these patients, and an additional variable which should be considered. 5. Conclusions The findings of this study indicate that the intra-tester reliability of medio-lateral patellar position tests are good, but that intertester reliability is variable. The criterion validity of these tests is at worse moderate. These are based on a limited evidence-base. Further study is recommended to compare the McConnell (1986) and Herrington (2002) methods of assessing medio-lateral patellar orientation. After this rigorous assessment, clinicians will then be better informed on the appropriateness of these tests when assessing patellar orientation in patients with patellofemoral disorders. Acknowledgements We thank the library staff at the Norfolk and Norwich University Hospital’s Sir Benjamin Gooch Library for their assistance in paper retrieval. We also thank Mr Mark Rowlands for his assistance with the photographs used in this paper. References Arendt EA, Fithian DC, Cohen E. Current concepts of lateral patella dislocation. Clinical Sports Medicine 2002;21:499–519. Beasley LS, Vidal AF. Traumatic patellar dislocation in children and adolescents: treatment update and literature review. Current Opinion in Paediatrics 2004;16:29–36. Critical Skills Appraisal Programme (CASP) Homepage on the Internet. Oxford, UK: Learning & Development Public Health Resource Unit; c. 2007. Available from: http://www.phru.nhs.uk/casp/critical_appraisal_tools.htm (accessed 1 May 2007). Crossley K, Bennell K, Green S, McConnell J. A systematic review of physical interventions for patellofemoral pain syndrome. Clinical Journal of Sports Medicine 2001;11(2):103–10.
Edwards A, Talbot R. The hard pressed researcher. A research handbook for the caring professions. London: Longman; 1994. Egger M, Davey Smith G, Schneider M. Systematic reviews of observational studies. In: Egger M, Davey Smith G, Altman DG, editors. Systematic reviews in health care. London: BMJ Books; 2001. p. 211–28. Evans R, Elwyn G, Edwards A. Review of instruments for peer assessment of physicians. BMJ 2004;328:1240. Fitzgerald GK, McClure PW. Reliability of measurements obtained with four tests for patellofemoral alignment. Physical Therapy 1995;75(2):84–92. Fulkerson JP, Shea KP. Disorders of patellofemoral alignment. Journal of Bone and Joint Surgery 1990;72-A:1424–9. Grelsamer RP, Newton PM, Staron RB. The medial-lateral position of the patella on routine magnetic resonance imaging: when is normal not normal? Arthroscopy 1998;14:23–8. Herrington LC. The inter-tester reliability of a clinical measurement used to determine the medial/lateral orientation of the patella. Manual Therapy 2002;7(3):163–7. Herrington L. The effect of corrective taping of the patella on patella position as defined by MRI. Research in Sports Medicine 2006;14:215–23. Herrington LC. The difference in a clinical measure of patella lateral position between individuals with patellofemoral pain and matched controls. Journal of Orthopaedic and Sports Physical Therapy 2008;38:59–62. Herrington L, Nester C. Q-angle undervalued? The relationship between Q-angel and medio-lateral position of the patella. Clinical Biomechanics 2004;19: 1070–3. Herrington L, Pearson S. The applicability of ultrasound imaging in the assessment of dynamic patella tracking: a preliminary investigation. The Knee 2008;15(2):125–7. Herrington L, Rivett N, Munro S. The relationship between patella position and length of the iliotibial band as assessed using Ober’s test. Manual Therapy 2006;11:182–6. Hughston JC. Subluxation of the patella. Journal of Bone and Joint Surgery 1968;50-A:1003–26. Insall J. Chondromalacia patellae: patellar malalignment syndrome. Orthopedic Clinics of North America 1979;10:117–27. McConnell J. The management of chondromalacia patellae: a long term solution. Australian Journal of Physiotherapy 1986;32(4):215–23. McConnell J. Rehabilitation and nonoperative treatment of patellar instability. Sports Medicine and Arthroscopy Reviews 2007;15:95–104. McEwan I, Herrington L, Thom J. The validity of clinical measures of patella position. Manual Therapy 2007;12:226–30. Mizuno Y, Kumagai M, Mattessich SM, Eilas JJ, Ramrattan N, Cosgarea AJ, et al. Q-angle influences tibiofemoral and patellofemoral kinematics. Journal of Orthopaedic Research 2001;19:834–40. Muhle C, Brossmann J, Heller M. Kinematic CT and MR imaging of the patellofemoral joint. European Radiology 1999;9(3):508–18. Ota S, Ward SR, Chen T-J, Tsai Y-J, Powers CM. Concurrent criterion-related validity and reliability of a clinical device used to assess lateral patellar displacement. Journal of Orthopaedic and Sports Physical Therapy 2006;36(9):645–52. Polgar S, Thomas SA. Introduction to research in the health sciences. 4th ed. London: Churchill Livingstone; 2000. Portney LG, Watkins MP. Foundations of clinical research. Applications to practice. 2nd ed. New Jersey: Prentice Hall; 2000. Powers CM, Mortenson S, Nishimoto D, Simon D. Criterion-related validity of a clinical measurement to determine the medial/lateral component of patellar orientation. Journal of Orthopaedic and Sports Physical Therapy 1999;29(7):372–7. Senavongse W, Farahmand F, Jones L, Andersen H, Bull AM, Amis AA. Quantitative measurement of patellofemoral joint stability: force-displacement behaviour of the human patella in vitro. Journal of Orthopaedic Research 2003;21:780–6. Sutlive TG, Mitchell SD, Maxfield SN, McLean CL, Neumann JC, Swiecki CR, et al. Identification of individuals with patellofemoral pain whose symptoms improved after a combined program of foot orthosis use and modified activity: a preliminary investigation. Physical Therapy 2004;84(1):49–61. Tolouei FM, Afshar A, Salarilak S, Sina A. CT patellar cortex tilt angle: a radiological method to measure patellar tilt. Iran Journal of Radiology 2005;3(1):17–21. Tomsich DA, Nitz AJ, Threlkeld AJ, Shapiro R. Patellofemoral alignment: reliability. Journal of Orthopaedic and Sports Physical Therapy 1996;23(3):200–8. Warden SJ, Hinman RS, Watson Jr MA, Avin KG, Bialocerkowski AE, Crossley KM. Patellar taping and bracing for the treatment of chronic knee pain: a systematic review and meta-analysis. Arthritis and Rheumatology 2008;15(1):73–83. Watson CJ, Propps M, Galt W, Redding A, Dobbs D. Reliability of McConnell’s classification of patellar orientation in symptomatic and asymptomatic subjects. Journal of Orthopaedic and Sports Physical Therapy 1999;29(7):378–85.
Manual Therapy 14 (2009) 363–368
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Original Article
Clinical measurement of craniovertebral angle by electronic head posture instrument: A test of reliability and validityq Herman Mun Cheung Lau a, Thomas Tai Wing Chiu a, *, Tai-Hing Lam b a b
Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, Hung Hom, Hong Kong Department of Community Medicine, The University of Hong Kong, Hong Kong
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 August 2007 Received in revised form 6 May 2008 Accepted 19 May 2008
The study was a cross-sectional reliability study with the objective of assessing the reliability and validity of the Electronic Head Posture Instrument (EHPI) in measuring the craniovertebral (CV) angle for subjects with or without neck pain. Twenty-six subjects (mean age ¼ 36.88, SD 9.95) with chronic neck pain and 27 subjects (mean age ¼ 31.85, SD 7.63) without neck pain were recruited. The CV angle was measured by the EHPI which consists of an electronic angle finder, a transparent plastic base and a camera stand. Two therapists were recruited to assess the intra- and inter-rater reliability of the EHPI in two separate sessions of measurement. The difference in CV angle between the two groups was determined. The CV angle of the patient group (mean 43.94, SD 3.61) was significantly smaller (p < 0.001) than that of the normal group (mean 50.58, SD 2.09). Intra-rater (intra-class correlation coefficient (ICC) ranged from 0.86 to 0.94) and inter-rater (ICC ranged from 0.85 to 0.91) reliability of the EHPI in measuring CV angle for both groups of subjects were high. In conclusion the EHPI was found to be reliable and valid in measuring the CV angle for subjects with or without neck pain. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Craniovertebral angle Reliability Validity
1. Introduction Neck pain is a common musculoskeletal disorder in the general population. A systematic review of neck pain over the world showed that the one-year prevalence ranged from 16.7% to 75.1% for the entire adult population with a mean of 37.2% (Fejer et al., 2006). In a recent telephone survey done in Hong Kong, Chiu and Leung (2006) reported that the 12 month prevalence was 53.6%. Posture of the head and neck has long been recognized as a factor contributing to the onset and perpetuation of cervical pain and dysfunction (Harrison et al., 2005; Persson et al., 2007). Forward head posture is one of the common poor head postures seen in patients with neck disorders (Hickey et al., 2000). A number of studies suggested that forward head posture predisposes individuals towards pathological conditions such as thoracic outlet syndrome and cervical spondylogenic changes (Rocabado, 1983; Ayub et al., 1984). There are many instruments used for assessing the head posture, such as the Rocabado Posture Gauge
q The various parts of the EHPI are readily available in the commercial market and we had no financial support from the manufacturers. No conflict of financial interest was involved in this measurement tool. * Corresponding author. Tel.: þ852 27666709; fax: þ852 23308656. E-mail address:
[email protected] (T.T. Wing Chiu). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.05.004
designed by Mariano Rocabado which is a T-shaped instrument. It measures the horizontal distance from the tangent of the most posterior thoracic spinous process to the most anterior cervical spinous process in standing position. The distance measured is the amount of the forward head posture of the subject (Willford et al., 1996). Moreover head posture can be measured by the plumb line and photographic imaging (Wilmarth and Hilliard, 2003). The use of plumb line is, however, limited to the subjective nature of determining the degree of forward head posture and discrepancies can be caused by the viewing angulation of the examiner in relation to the patient. Problems of photographic imaging are related to the time expenditure required for accurate assessment and it does not allow for immediate results (Wilmarth and Hilliard, 2003). One of the objective methods of assessing head posture is through measuring the craniovertebral (CV) angle. This is the angle between a horizontal line through the spinous process of C7 and a line from spinous process of C7 through the tragus of the ear (Raines and Twomey, 1994; Joe et al., 2003). Joe et al. (2003) studied the reliability of measuring the CV angle with Coutts overlay sheet (a transparent grid sheet for mapping) and a protractor in a lateral photograph in 29 female students, and the intra-rater reliability was very high. They also suggested that smaller CV angles indicate greater protraction of the head and larger angles are more representative of ‘ideal’ sagittal plane head/neck alignment. However, the result was limited to young female students only. In many
364
H.M. Cheung Lau et al. / Manual Therapy 14 (2009) 363–368
previous studies, investigators suggested that there were associations between the forward head posture and neck pain and disability (Griegel-Morris et al., 1992; Szeto et al., 2002). They found that those subjects with head, neck and shoulder discomfort or pain are more likely to have a smaller CV angle. In a recent study Yip et al. (2008) concluded that patients with small CV angle had a greater forward head posture and the greater the forward head posture, the greater the disability. Yip et al. (2008) also recommended that the CV angle could provide clinicians with further objective information on the disability and severity of patients with neck pain. An instrument, the Head Posture Spinal Curvature Instrument (HPSCI) was developed to measure both the CV angle and cervical curvature (Willford et al., 1996; Wilmarth and Hilliard, 2003). It was designed to provide more efficient assessment tool with immediate feedback. The HPSCI is a non-invasive, inexpensive measurement method which has been demonstrated to produce consistent and stable intra-rater results (intra-class correlation coefficient, ICC ¼ 0.9) across days and trials in 27 healthy subjects (Willford et al., 1996). However, its measuring scale was accurate to the whole number only; there is potential error in attempting to choose which way to read the scale when the indicator falls between two whole numbers. Despite the good intra-rater reliability, Wilmarth and Hilliard (2003) did not investigate the validity of the instrument which is essential for a clinical measuring tool, because strong reliability does not suggest strong validity (Portney and Watkins, 2000). Moreover, the inter-rater reliability in assessment of patients with neck pain has not been elucidated. Manual therapists utilize different methods to achieve correction of forward head posture in patients with neck pain (Harrison et al., 1996). Assessment of posture is an important component of evaluation and affects the design of a treatment regimen. However, clinical evaluation of head posture is generally based on the clinician’s subjective visual impression. Besides, it is difficult to compare patients with each other and to quantify the improvements. Clinicians are continually looking for easier, safer, more objective and reliable devices to measure the head posture which is not yet available in the literature. Hence, the Electronic Head Posture Instrument (EHPI) was developed. The measuring scale of this instrument is accurate to one decimal place and the reading can be taken automatically by the electronic sensor. The objective of this study was to examine the reliability and validity of the EHPI in subjects with and without neck pain.
months and without any treatment were recruited in the patient group. Any person who had history of fracture injury at the cervical and shoulder region or vertebral column, scoliosis, severe thoracic kyphosis, spasmodic torticollis, rheumatic disease, temporomandibular joint dysfunction, neurological motion disorder or back pain, or loss of standing balance were excluded. Explanation and informed consent were obtained from each subject. The project was approved by the University’s Review Board for Health Science Research involving Human Subjects. 2.2. Apparatus The CV angle was measured by the EHPI which consists of an electronic angle finder, a transparent plastic base and a camera stand. The electronic angle finder ‘SmartTool Angle Finder’ made by M-D Building Products (United States) was fixed on a transparent plastic base. The combined Angle Finder and the plastic base (now named as Angle Finder) were mounted on a tripod camera or video camera stand – HAMA ‘Gamma 74’ which was made in Germany (Fig. 1). The Angle Finder could be used to identify and digitally display degrees/percent slope quickly and pitch to 1/10 degree accuracy. Two parallel lines were marked on the opposite sides of the transparent base. The targeted markers were aligned with the two parallel lines of the plastic base simultaneously, in order to measure the CV angle (Fig. 1). The camera stand was used to adjust the height and the tilting angle of the Angle Finder to measure the CV angle. Measurement validity defines the extent to which an instrument measures what it is intended to measure (Portney and Watkins, 2000). It is therefore necessary to establish the validity of the EPHI to detect different angles of rotation from the horizontal. An ‘index table’, validated in angle measurement was used as a criterion to test the measurement validity of the Angle Finder and the steps were as follows. The Angle Finder was reset to ‘0 ’ and mounted to the index table as shown in Fig. 2. The index table was rotated counter-clockwise and then clockwise again such that the readout from Angle Finder was 0.0 . The pointer of the index table was set to 0 . After that, the index table was rotated clockwise manually by hand to 90 with stops over each 5 rotation. The readings from the index table and the Angle Finder were recorded at each step. The index table was then rotated counter-clockwise back to 0 with stops over on each 15 rotation with the reading recorded. 2.3. Procedure of measuring the CV angle
2. Materials and methods 2.1. Subjects This study was a cross-sectional reliability study with a convenient sampling from the out-patient clinic of the Physiotherapy Department in the Prince of Wales Hospital, Shatin, Hong Kong. In order to carry out a contrast-group comparison to investigate the validity of the EHPI, the present authors adopted the following approach for calculation of the sample size. Using a software package from Number Cruncher Statistical System (NCSS) – power analysis and sample size for Windows (1996), it was estimated that a total of 52 subjects, i.e. 26 subjects in each of the patient group and normal group, would be required (Hintze, 1996). The following parameters were used as inputs to the program: (1) 0.05 alpha; (2) 90% power; (3) two-sided alternative; (4) an effect size of 0.5 (this was chosen because an effect size of 0.5–0.7 was considered moderate by Cohen (1977)). Twenty-seven volunteers aged 19–53 years (mean 31.85 years, SD 7.63) and without neck pain during the past 6 months were recruited in the normal group (Ylinen et al., 2004). Twenty-six subjects aged 20–55 years (mean 36.88 years, SD 9.95) with diagnosis of more than one episode of neck pain during the past 3
The EPHI was put on the standardized marking on the floor and the HAMA stand was adjusted into the position until the bubble of the horizontal indicator and the central marking overlapped. The distance from the subject to the centre of the HAMA stand was standardized to 0.3 m while the distance between the operator and the HAMA stand was 0.5 m because this was the longest distance that the testers can reach. The subjects were asked to put on sportswear in order to expose their neck and the upper thoracic spine. They were also required to remove their socks and shoes. The seventh cervical (C7) spinous process was palpated and identified and an adhesive pin marker (Fig. 1) was attached over its midpoint of the most prominent part. The subject was then asked to stand with his/her left shoulder in front of the EPHI. Another pin marker was fixed at tragus of his/her left ear. The subject was instructed to stand comfortably with their weight distribution evenly on both feet and to keep their eyes looking straight ahead. He/she was then instructed to flex and extend the head for three times and then rest it in a comfortable position. A virtual line was drawn between the two pin makers from midpoints of the tragus to C7. The therapist adjusted the EHPI until the two indicator lines were aligned with the markers. The reading from the Angle Finder represented the CV angle, as seen in Fig. 1.
H.M. Cheung Lau et al. / Manual Therapy 14 (2009) 363–368
365
Fig. 1. Measuring CV angle by the EHPI. When two indicator lines were aligned with the markers (C7 spinous process and tragus), the reading from the Angle Finder would represent the CV angle.
Two physiotherapists with five years of clinical experience in treating patients with neck pain were involved in this experiment to test the intra- and inter-rater reliability of the EHPI. They were trained by one of the authors (HL) to operate the EHPI for measuring CV angles. Two sessions were arranged to measure the CV angle. The reading of each session was taken by two therapists, respectively. In each session, the subject was asked to stand up for the first therapist to take the measurement. He/she was then required to walk around the room and then resume standing for the second therapist, who was blinded to the results of the first measurement, to repeat the measuring procedure. The second session was arranged 7 days later, in order to minimize the memory effect of the therapists. This amounted to four trials in two sessions [i.e. 2(1 þ1)]. Each subject was informed to avoid any unusual activities within the 7 days between the two sessions in order to minimize the changes in his/her neck pain conditions. They were also required to report any change of the neck condition before the second session.
2.4. Statistical analysis 2.4.1. Validity Paired t-test and Pearson’s correlation test were used to examine if there was any significant difference and the correlation between the readings of the index table and the Angle Finder as a test of validity. 2.4.2. Inter/intra-tester reliability ICC Model 1, Form 1 was used to determine the intra-rater and Model 2, Form 1 for the inter-rater reliability of using the EHPI. It has been suggested that ICCs below 0.50 represent poor reliability, ICCs from 0.50 to 0.75 represent moderate reliability and ICCs above 0.75 indicate good reliability (Portney and Watkins, 2000).
Fig. 2. Validity test of smart tool with the index table. The Angle Finder was put on top of the arm of the index table. In the lower figure, the Angle Finder was moved along with the arm of the index table to 30.0 in order to check the measurement validity of the Angle Finder.
2.4.3. Descriptive analysis on normal group and patient group Independent sample t-tests were used to determine if there was any difference in the demographic characteristics and the CV angle between the normal group and the patient group. The minimal level of detectable change (MDC) was calculated according to the formula: standard error of measurement (SEM) z-score at the
366
H.M. Cheung Lau et al. / Manual Therapy 14 (2009) 363–368
showed that the intra-rater (ICC ranged from 0.86 to 0.94) and inter-rater (ICC ranged from 0.85 to 0.91) reliability of the EHPI in measuring the CV angle for both groups of subjects were high. The SEM was 1.193 and the MDC was 3.31.
Table 1 Demographic characteristics of normal group and patient group. Study part
Normal group (n ¼ 27)
Patient group (n ¼ 26)
Age Mean (standard deviation) (years) Range (years)
31.9 (7.6) 19–53
36.5 (9.7) 20–55
3.4. CV angle of normal group vs patient group
Gender Male Female
n 12 15
n 11 15
The CV angle of the patient group (group mean 43.9, SD 3.6) was significantly smaller (p < 0.001) than that of the patient group (group mean 50.6, SD 2.1) (Table 3)
% 22.6 28.3
% 20.8 28.3
two-sided 95% confidence intervals (z ¼ 1.96) O2, where SEM ¼ SD O1 ICC (Beaton, 2000). The level of significance was set to be 0.05. SPSS version 14.0 program was used for statistical analysis.
4. Discussion 4.1. Validity The validity of the EPHI to detect different angles of rotation from the horizontal was demonstrated by the non-significant difference and the high correlation between the readings from the index table which acted as a standard for measurement. Also, the validity of a clinical measuring tool depends on its ability to differentiate the symptomatic subjects from non-symptomatic ones. The EHPI is also valid in this aspect which can be demonstrated by the contrast-group comparison as discussed in the later section.
3. Results 3.1. Descriptive analysis A total of 26 subjects (11 males, 15 females) were recruited in the patient group and 27 subjects (12 males, 15 females) in the normal group. The demographic characteristics of the subjects are shown in Table 1. It was found that there was no significant difference in the age (p ¼ 0.061) or gender (p ¼ 0.88) between the two groups.
4.2. Reliability 3.2. Validity Result of the paired t-test demonstrated that there was no significant difference (p ¼ 1.000) between the readings of the index table and the Angle Finder at different angles of rotation. They are highly correlated with each other as indicated by the result of Pearson’s correlation test (p ¼ 0.000, r ¼ 1.000) and the X–Y plot as shown in Fig. 3.
It was found that the intra- and inter-rater reliability of using the EHPI to measure the CV angle were high, which demonstrated that this CV angle measurement method was highly repeatable for both the normal group and the patient group. The inter-tester reliability results for trial 2 were better than that of trial 1; this may be due to the practice effect which can help to decrease the random errors of measurement (Portney and Watkins, 2000).
3.3. Reliability
4.3. Normal group vs patient group
Reliability of the EHPI was analyzed separately in the normal group and the patient group as shown in Table 2. The results
The contrast-group comparison of the present study confirmed that the CV angle of the non-symptomatic subjects as measured by
100.00
Smart Tool
80.00
60.00
40.00
20.00 R Sq Linear =1
0.00 0.00
20.00
40.00
60.00
80.00
100.00
Index Table Fig. 3. X–Y plot of index table and Angle Finder. Validity of the Angle Finder in correlation to the index table is shown where the linear line represented that these two tools were highly correlated with each other.
H.M. Cheung Lau et al. / Manual Therapy 14 (2009) 363–368
367
Table 2 Intra-rater and inter-rater reliability of therapists A and B in measuring CV angle for subjects of patient group and normal group. Patient group ICC (95% CI) n ¼ 26
Normal group ICC (95% CI) n ¼ 27
Intra-rater reliability
Therapist A Therapist B
0.87 (0.73–0.94) 0.94 (0.86–0.97)
0.87 (0.74–0.94) 0.86 (0.72–0.93)
Inter-rater reliability
First trial Second trial
0.86 (0.71–0.94) 0.91 (0.83–0.96)
0.85 (0.71–0.93) 0.88 (0.76–0.95)
the EHPI was significantly smaller than that of the patients with neck pain. This concurred with the previous findings that patients with neck pain have a forward head posture [bad posture] (Edmondston et al., 2007; Cho, 2008; Yip et al., 2008) as compared to non-symptomatic subjects (Haughie et al., 1995; Hickey et al., 2000). It is well accepted that forward head posture predisposed individual towards pathological condition over the cervical and cranial region. Johnson (1998) suggested that persisted forward head posture might increase loading of the cervical joints and abnormal stress on the posterior cervical structures and cause myofascial pain. However, the exact mechanism is still unknown, further study in this aspect is warranted. As there was no statistically significant difference in terms of the age and gender between the two groups, these two factors should not contribute to the difference in the CV angle between the two groups. Raines and Twomey (1994) reported that, regardless of the difference in population samples, the CV angle is a reliable indicator of variation in the head and neck posture. To be clinically useful and valid an instrument should be able to distinguish normal subjects from the patients. The present study has demonstrated significant difference in the CV angle between the patient group and the normal group. This helps strengthen the validity of the EHPI in clinical measurement. The MDC for the measurement of CV angle by the EHPI was 3.31. This implies that when the CV angle measured by the EHPI demonstrated a difference of 3.31 or more, the therapist can be confident that there is a real change of the CV angle and not just a measurement error. In comparison with previous studies (Joe et al., 2003; Wilmarth and Hilliard, 2003; Yip et al., 2008) the present study achieved the following improvements: the automatic display of all the measurement data in the Angle Finder helped reduce the possibility of transcription and calculation errors. In addition, all the measurement data were digitalized to one decimal place. This helped minimize the inaccuracies caused when reading an analogue scale where the indicator falls between two whole numbers. Moreover, the Angle Finder can quickly identify any angle with good accuracy within 0.1. 4.4. Clinical implications Clinicians always try to correct their patients’ forward head postures by various treatment approaches. In order to assess the
effectiveness of these approaches, it is vital to develop an objective method to measure the posture of the patients. The device involved should be practical, user-friendly, reliable and objective in its quantification. The EHPI serves the above purpose as it is an inexpensive, portable and convenient apparatus that allows clinicians to collect an accurate, reliable and objective reading. The instrument provides clinicians with additional useful and reliable information to monitor patients’ condition and progression. 4.5. Limitations of the study The sagittal plane position of the cervical, thoracic or lumbar spine was not measured. This was one of the limitations of this study because the CV angle depends on the relative position of the entire spine. It must be borne in mind that the CV angle is an index reflective of only one part of the total picture of the cervical and head posture. Accurate assessment of complete head and neck posture requires a cephalometric radiographic analysis which was not available in this study. Moreover, further study may be required to elucidate if there are any differences when the subject is in a standing vs a sitting position for analysis of head posture. Due to the limited number of subjects who participated in the study, and the possibility of selection bias in the recruitment of normal subjects, the CV angle measurement results cannot be regarded as a baseline reference for the particular group. In addition, the age group of this study was limited to 19–55 years; therefore further investigation with larger sample size involving different age groups is indicated. Also, responsiveness of the CV angle as measured by the EHPI should be tested in a longitudinal study for subjects with neck pain. 5. Conclusion We have demonstrated that the EHPI was valid in measuring the CV angle. There was a high degree of test–retest reliability in measuring the CV angle for both the normal subjects and those with neck pain. The CV angle of subjects with neck pain was significantly smaller than that of the normal subjects. With increasing demand being placed on evidence-based practice, a valid and reliable outcome measuring tool should contribute to better clinical service and evaluation.
Table 3 CV angle of subjects in normal group and patient group measured by therapists A and B. Minimum CV angle
Maximum CV angle
Individual mean SD
Group mean SDa
Normal (n ¼ 27)
First session therapist A Second session therapist A First session therapist B Second session therapist B
47.4 46.3 47.1 46.5
56.7 57.0 55.4 55.5
51.02.4 51.22.3 50.12.1 50.02.1
50.62.1
Patient (n ¼ 26)
First session therapist A Second session therapist A First session therapist B Second session therapist B
33.7 35.3 35.6 33.7
49.5 48.2 49.5 50.1
43.83.4 44.13.6 44.03.7 43.93.8
43.93.6
Normal/patient
a
Group mean is the mean CV angle measured by both therapists in two sessions of each group.
368
H.M. Cheung Lau et al. / Manual Therapy 14 (2009) 363–368
References Ayub E, Glasheen-Wray M, Kraus S. Head posture: a case study of the effects on the rest position of the mandible. Journal of Orthopaedic and Sports Physical Therapy 1984;5:179–83. Beaton DE. Understanding the relevance of measured change through studies of responsiveness. Spine 2000;25:3192–9. Chiu TW, Leung SL. Neck pain in Hong Kong: a telephone survey on prevalence, consequences and risk groups. Spine 2006;31(16):E540–4. Cho CY. Survey of faulty postures and associated factors among Chinese adolescents. Journal of Manipulative and Physiological Therapeutics 2008;31(3):224–9. Cohen J. Statistical power analysis for the behavioural sciences. New York: Academic Press; 1977. Edmondston SJ, Chan HY, Chi Wing Ngai G, Warren MLR, Williams JM, Glennon S, et al. Postural neck pain: an investigation of habitual sitting posture, perception of good posture and cervicothoracic kinaesthesia. Manual Therapy 2007;12(4):363–71. Fejer R, Kyvik K, Hartvigsen J. The prevalence of neck pain in the world population: a systemic critical review of the literature. European Spine Journal 2006;15: 834–48. Griegel-Morris P, Larson K, Mueller-Klaus K, Oatis CA. Incidence of common postural abnormalities in the cervical, shoulder, and thoracic region and their association with pain in two age groups of healthy subjects. Physical Therapy 1992;72:425–31. Harrison AL, Barry-Greb T, Wojtowicz GW. Clinical measurement of head and shoulder posture variables. The Journal of Orthopaedic and Sports Physical Therapy 1996;23(6):353–61. Harrison DE, Haas JW, Cailliet R, Harrison DD, Holland B, Janik TJ. Concurrent validity of flexicurve instrument measurements: sagittal skin contour of the cervical spine compared with lateral cervical radiographic measurements. Journal of Manipulative and Physiological Therapeutics 2005;28: 597–603. Haughie L, Fiebert IM, Roach KE. Relationship of forward head posture and cervical backward bending to neck pain. Journal of Manual and Manipulative Therapy 1995;3:91–7.
Hickey ER, Rondeau MJ, Corrente JR, Abysalh CJ. Reliability of the cervical range of motion (CROM) device and plumb line techniques in measuring resting head posture (RHP). Journal of Manual and Manipulative Therapy 2000;8:10–7. Hintze JL. Power analysis and sample size for Windows. PASS. Kaysville, Utah, USA: NCSS; 1996. Joe B, Elizabeth B, Aoife NM. Reliability of measuring natural head posture using the craniovertebral angle. Irish Ergonomics Review 2003:37–41. Johnson GM. The correlation between surface measurement of head and neck posture and the anatomic position of the upper cervical vertebrae. Spine 1998; 23:921–7. Persson P, Hirschfeld H, Nilsson-Wikmar L. Associated sagittal spinal movements in performance of head pro- and retraction in healthy women: a kinematic analysis. Manual Therapy 2007;12:119–25. Portney LG, Watkins MP. In: Foundations of clinical research applications to practice. 2nd ed. Upper Saddle River, New Jersey: Prentice Hall Health; 2000. p. 61–77. Raines S, Twomey L. Posture of the head, shoulders and thoracic spine in comfortable erect standing. Australian Journal of Physiotherapy 1994;40:25–32. Rocabado M. Biomechanical relationship of the cranial, cervical and hyoid regions. Journal of Cranio-mandibular Practice 1983;1:1–6. Szeto G, Straker L, Raines S. A field comparison of neck and shoulder posture in symptomatic and asymptomatic office workers. Applied Ergonomics 2002; 33(1):75–84. Willford G, Kisner C, Glenn TM, Sachs L. The interaction of wearing multifocal lenses with head posture and pain. The Journal of Orthopaedic and Sports Physical Therapy 1996;23(3):194–9. Wilmarth MA, Hilliard TS. Measuring head posture via the craniovertebral angle. Orthopaedic Physical Therapy Practice 2003;14:13–5. Yip CHT, Chiu TTW, Poon ATK. The relationship between head posture and severity and disability of patients with neck pain. Manual Therapy 2008;13(2):148–54. Ylinen J, Salo P, Nykanen M, Kautiainen H, Hakkinen A. Decreased isometric neck strength in women with chronic neck pain and the repeatability of neck strength measurements. Archives of Physical Medicine and Rehabilitation 2004;85:1303–8.
Manual Therapy 14 (2009) 369–374
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Original Article
Mobilizations of the asymptomatic cervical spine can reduce signs of shoulder dysfunction in adults Lynda McClatchie a, Judi Laprade b, Shelley Martin c, Susan B. Jaglal a, b, Denyse Richardson a, Anne Agur d, * a
Graduate Department of Rehabilitation Sciences, University of Toronto, 500 University Avenue, Toronto, Ontario, Canada M5G 1V7 Department of Physical Therapy, University of Toronto, 160-500 University Avenue, Toronto, Ontario, Canada M5G 1V7 Spine and Sport Physiotherapy Centre, 123 Edward Street, Suite 500, Toronto, Ontario, Canada M5G 1E2 d Division of Anatomy, Department of Surgery, University of Toronto, Medical Sciences Building, 1 King’s College Circle, Toronto, Ontario, Canada M5S 1A8 b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 July 2007 Received in revised form 18 February 2008 Accepted 19 May 2008
Generalized shoulder pain is a common problem that is difficult to treat and frequently recurrent. The asymptomatic cervical spine must be ruled out as a cause of any shoulder pain, as it can have a similar presentation to an isolated shoulder disorder. Previous studies have shown that lateral cervical glide mobilizations to the asymptomatic cervical spine at C5/6 can affect peripheral pain, but none have examined shoulder pain. A randomized, blinded, placebo-controlled, cross-over trial was used to examine the immediate effects of cervical lateral glide mobilizations on pain intensity and shoulder abduction painful arc in subjects with shoulder pain. Twenty-one subjects received interventions of both cervical mobilization and placebo over two sessions. Pain intensity using a visual analog scale (VAS) and painful arc were assessed prior to and following application of cervical mobilization or placebo intervention. Evaluation of cervical mobilization revealed the shoulder abduction painful arc (12.5 15.6 , p ¼ 0.002) and shoulder pain intensity (1.3 1.1 cm, p < 0.001) were significantly decreased. The results of this study suggest that any immediate change in shoulder pain or active shoulder range of motion following cervical mobilizations indicate that treatment directed toward the asymptomatic cervical spine may expedite recovery. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Manual therapy Mobilization Radiculopathy Painful arc
1. Introduction The prevalence of shoulder pain as reported in the literature ranges from 7 to 34% in the adult population (Chard et al., 1991; Badley and Tennant, 1992; Van der Windt et al., 1996). Despite its prevalence, the pathophysiology of shoulder pain has not been well defined (de Winter et al., 1999; Groenier et al., 2003), and the diagnosis is complicated by the large number of structures in the shoulder region (Bamji et al., 1996; Pope et al., 1997). Differentiation between various shoulder disorders is important to implementing effective treatment (Green et al., 1998), but the differential diagnosis is often complex as a broad spectrum of intrinsic and extrinsic conditions may produce shoulder pain (Neviaser, 1983). The tendons of the rotator cuff or long head of biceps (Lyons and Orwin, 1998; Van der Heijden, 1999), calcium deposits within tendons (Turner-Stokes, 1996), degenerative changes in the glenohumeral or acromioclavicular joint (Prescher, 2000), or inflammation in the bursae surrounding the shoulder (Koester et al.,
* Corresponding author. Tel.: þ1 416 978 8855; fax: þ1 416 978 3844. E-mail address:
[email protected] (A. Agur). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.05.006
2005) are some of the structures which cause pain around the shoulder. Generalized shoulder pain is frequently recurrent (Winters et al., 1999) and can remain present for over three years after onset (Chakravarty and Webley, 1993; Croft et al., 1996; Badcock et al., 2002), with one study indicating that 50% of subjects with shoulder pain had persistent problems three years later (MacFarlane et al., 1998). Research has shown that long-term shoulder pain can lead to a considerable restriction of work and leisure activities (Wells, 1982). Persistence of the problem can lead to lengthy treatment in physiotherapy and frequent general practitioner appointments (Wells, 1982). Shoulder pain can arise from a cervical spine disorder, and therefore the neck must be ruled out as a potential cause of pain (Manifold and McCann, 1999). Radiculopathy arising from the cervical spine is difficult to differentiate from localized shoulder pathology because the sensory distribution extends from the base of the neck to the outer edge of the shoulder (Wilson, 2005). Sobel et al. (1997) found that restricted mobility in the cervicothoracic spine in patients with shoulder pain did not seem to recover significantly after 26 weeks. These investigators suggested that if intervention at the cervicothoracic spine was included in
370
L. McClatchie et al. / Manual Therapy 14 (2009) 369–374
management of patients with shoulder pain, the tendency for shoulder disorders to recur could be decreased. Cervical spine mobilization techniques can be used during therapy to affect more peripheral symptoms (Vicenzino et al., 1996; Sterling et al., 2001). Previous studies have shown that chronic elbow pain and temporomandibular joint (TMJ) pain can be reduced by a mobilization intervention to the asymptomatic neck (Vicenzino et al., 1996; Stiesch-Scholz et al., 2003). However, utilizing this technique for the treatment of shoulder symptoms has rarely been studied (Bergman et al., 2004). Bergman et al. (2004) examined the effectiveness of manual therapy directed toward the cervicothoracic spine and adjacent ribs for patients with shoulder pain. Participants were randomized into groups receiving manual therapy in addition to usual medical care (advice, analgesics, nonsteroidal anti-inflammatory drugs), or usual medical care alone. Pain radiating to the neck region was not criterion for exclusion, and the manipulative therapy included specific mobilizations or manipulations at the discretion of the therapist. The primary outcome measure was patient-perceived recovery, which was recorded on a seven point scale. Other outcomes included the severity of shoulder pain and functional disability. The results showed that at 12, 26 and 52 weeks follow-up, the manual therapy group demonstrated significant improvement for all outcome measures. A link between the asymptomatic cervical spine and the painful shoulder has not been clearly established in the literature. The objective of this study was to determine whether the utilization of cervical lateral glide mobilizations of the asymptomatic cervical spine at C5, C6, and C7 could immediately reduce the intensity and/ or range of the painful arc in subjects with generalized shoulder pain who were previously unresponsive to traditional shoulder treatment.
2. Methods Approval for this study was granted by the University of Toronto Research Ethics Board (Protocol reference #14011).
2.1. Subjects Subjects were recruited from a private orthopaedic physical therapy practice in Toronto, Canada, from August 2005 through May 2006. Each subject was currently being treated for generalized unilateral shoulder pain. Both male and female subjects were included in this study if they were age 18 or older, had an insidious onset of unilateral shoulder pain of at least six weeks duration, demonstrated a painful arc with shoulder abduction, and had no current or previous complaints of neck pain within the past year. Patients with shoulder pain were excluded if they had symptoms of paresthesia or neurological deficits, previous surgery or dislocation of the affected shoulder, clinically definitive arthritis of the shoulder on X-ray or had a cortisone injection for the current episode of shoulder pain. Prior to entering the study, subjects must have been unresponsive to 2–4 recent physiotherapy sessions addressing shoulder pain through ‘‘traditional’’ methods of movement patterns, strengthening and modalities such as ultrasound and cryotherapy. The short duration of shoulder treatment to patients prior to admittance into this study was purposeful, as a change in treatment direction would likely be warranted if the patient’s shoulder signs and symptoms had not changed following 2–3 weeks of treatment and home exercises. A sample size calculation of 21 subjects was determined using data from the first 12 subjects to detect a statistically significant difference in shoulder pain with 80% power.
2.2. Study design This was a randomized placebo-controlled cross-over trial. It was an exploratory pilot study to determine if a subsequent RCT would be appropriate. Subjects were randomized by a coin toss to receive either the cervical lateral glide mobilization or the placebo intervention during the first session. The subject returned for the second session within four days and underwent the same measurement protocol, but the second examiner performed the intervention that was not received during the first session. All outcome measures were assessed both before and after the intervention and conducted by the first investigator, who was blinded to which treatment intervention was received. The second examiner performed the predetermined cervical lateral glide mobilization or placebo treatment condition. Informed consent was obtained as per protocol, and all participants were required to attend two sessions of approximately 40 min duration (Fig. 1). 2.3. Intervention 2.3.1. Technique of lateral cervical glide mobilization A lateral cervical glide mobilization was the technique chosen for this study, with the subject seated and the thoracic spine resting against the back of the chair, head in a neutral position, feet resting flat on the floor, and arms relaxed with hands in their lap. The lateral aspect of the spinous processes of C5, C6, and C7 was landmarked on the ipsilateral side of the subject’s painful shoulder. The examiner’s thumb remained on the lateral aspect of the spinous process of C5, with the opposite hand placed on the subject’s non-affected shoulder or head for counterbalance as a lateral movement toward the non-painful side was applied with the mobilizing hand (Mulligan, 1995) (Fig. 3). Mobilizations were conducted for 2 min each at C5, C6 and C7, with small amplitude end range movements (Grade IVþ). The placebo treatment condition involved the examiner resting her hands in the same positions as the mobilization technique, but without the application of external force. 2.3.2. Outcomes measured A 10 cm visual analog scale (VAS) for pain measurement was completed by the subject following the shoulder abduction trials. The VAS has been shown to be a valid, reliable and responsive measure of a subject’s perceived level of pain (Price et al., 1994; Guerra de Hoyos et al., 2004). Active cervical spine range of motion in all planes (flexion, extension, bilateral side-bending) was measured using a cervical range of motion goniometer (CROM), which has been shown to be a valid and reliable tool for the measurement of cervical range of motion in the sagittal and frontal planes (Ordway et al., 1997; Tousignant et al., 2000). A Myrin goniometer was used to measure bilateral cervical rotation, and has been shown to be a reliable tool for measurement in the transverse plane (Balogun et al., 1989; Malstrom et al., 2003). Manual muscle testing of shoulder abduction at 90 was tested on the unaffected side followed by the affected side using an electrodynamometer. The resistance was applied by the examiner proximal to the subject’s elbow joint. Active shoulder range of motion into abduction was performed to determine the presence and extent of a painful arc. Reflective adhesive stickers were placed in five locations on the subject: sternal notch, anterior tip of the right and left shoulder at the acromion process, affected side elbow crease and proximal wrist crease. These points were used as landmarks from which the angle of the painful arc within the shoulder abduction range of motion could be accurately measured. Shoulder abduction was videotaped with the subject facing the video camera so that the reflective markers were clearly seen. The
L. McClatchie et al. / Manual Therapy 14 (2009) 369–374
371
Subject 1
Day 1
Objective Measurements
Randomization
Day 2 Cervical AROM Abduction MMT FHP Arc of Pain
Objective Measurements
Placebo Condition
Mobilization Condition
Re-measure
Re-measure
Fig. 1. Summary of methodology.
starting position for this test was with the palm of the affected arm facing outward (external rotation of the shoulder) and the arm by their side. While performing active shoulder abduction, the subject kept the thumb on the uninvolved side down until the start of their shoulder pain, then kept the thumb raised until the end of the painful arc (Fig. 2A and B). The digital video camera (Samsung SCD353, China) was mounted on a standard adjustable tripod at a standard position 220 cm from the subject. The beginning and end of the arc of pain with shoulder abduction was later quantified using software (Virtual Dub freeware) to identify the precise frames in which the subject began to raise and lower their thumb on the uninvolved side to indicate the start and end of their shoulder pain,
respectively. The abduction angle was determined on those chosen frames by using the measuring tool in Adobe PhotoshopÔ CS2. Cervical AROM was performed first, followed by shoulder abduction manual muscle testing. The subject was then positioned in front of the camera with adhesive markers as outlined previously to perform shoulder abduction. Three trials of shoulder abduction were recorded, with the subject indicating the start and end of the arc of pain each time. The measurements from the painful arc from each of the trials were averaged together to represent the true arc of pain. Subjects indicated the intensity of their pain during shoulder abduction on the VAS. Following these measures, the predetermined intervention (i.e. mobilization or placebo) was then performed by the second examiner as outlined previously. Subsequent to this intervention, the first examiner returned for re-measurement of all pre-intervention outcome measures. A second VAS was then completed by each subject to indicate the intensity of post-intervention shoulder pain. After the completion of the second session, each subject was
Fig. 2. Measurement of painful arc during shoulder abduction.
Fig. 3. Positioning for cervical lateral glide mobilization.
372
L. McClatchie et al. / Manual Therapy 14 (2009) 369–374
asked by the examiner to describe the perceived difference between the two treatment techniques. Answers were recorded on subject data sheets. 2.4. Data analysis The Statistical Package for the Social Sciences (SPSS) version 14.0 was used for data analysis. Paired t-tests were used to evaluate prepost mobilization as well as pre-post placebo for the following outcomes: cervical spine range of motion, shoulder abduction manual muscle testing, shoulder pain intensity and shoulder range of motion. A paired t-test was used to determine if there was a significant difference (p < .05) between the mean averages of the placebo and cervical mobilization conditions. A Pearson correlation test was used to determine if a relationship existed between shoulder abduction arc of pain and shoulder pain intensity. 3. Results 3.1. Subjects Twenty-one subjects (14 females, seven males) with an average age of 49.8 (9.8) years volunteered and consented to participate in the study. Forty-three percent of subjects reported a unilateral shoulder problem, and 57% reported a previous resolved neck problem more than one year prior to commencing participation in the study. Seven subjects were randomized to the mobilization condition during the first session, while 14 subjects received the placebo condition (Fig. 1). Each subject recognized that the mobilization and placebo interventions were different from each other, however, no subject realized that the placebo intervention was not therapeutic. 3.2. Pain measures There was a significant difference (p < 0.001) in the intensity of shoulder pain as measured by VAS before and after the mobilization treatment condition (Table 1). A 1.3 1.1 cm decrease in the mean VAS score post-mobilization was recorded when compared with the pre-mobilization pain score. Eighteen subjects (86%) showed an average decrease of 1.5 cm of shoulder pain postmobilization condition. Following the placebo condition, the 0.2 0.6 cm decrease in the mean VAS score was not statistically significant (p ¼ 0.078) (Table 1). Using a paired t-test, the mobilization and placebo conditions were found to be significantly different (p ¼ 0.0002), with a treatment effect of 1.038. The radius of the arc of pain with shoulder abduction diminished significantly after both the mobilization (12.5 15.6 , p ¼ 0.002) and the placebo (8.8 12.7, p ¼ 0.005) conditions (Table 1). In comparing pre- and post-mobilization shoulder arc of pain to pre- and post-mobilization VAS scores, 14 subjects (66.7%)
Table 1 Pre-post condition (placebo and mobilization) scores. Pre-condition mean SD VAS – placebo (cm) 3.5 2.3 VAS – mobilization 3.7 2.0 (cm) Arc of pain – placebo 31.4 22.3 ( ) Arc of pain – 33.0 21.6 mobilization ( )
Post-condition mean SD
Mean difference SD
P* [twotailed]
3.2 2.5 2.4 2.1
0.2 0.6 1.3 1.1
0.078 20% of their professional time caring for people with NSLBP and no time researching NSLBP. ‘Predominantly researchers’ was arbitrarily defined as participants who spend 20% of time caring for people with NSLBP and 20% of their time researching NSLBP.
Fig. 2. Relationship of aims of this research in developing a conceptual framework for NSLBP.
390
P.M. Kent et al. / Manual Therapy 14 (2009) 387–396
3. Results 3.1. Survey response Overall, there was an 84% response rate (n ¼ 544), consisting of an 89.4% response rate from attendees at the Melbourne meeting and a 66.7% response from attendees at the Amsterdam Forum. Participant demographics are shown in Table 1. Overall, most participants worked predominately as clinicians, most had a special interest in low back pain and the largest professional discipline represented was physiotherapy (70.3%). As the descriptive characteristics of people who declined participation in the survey were not measured, non-responder bias cannot be determined. The number of participants from different professional disciplines ranged from 383 physiotherapists to three psychologists. However, sample size calculations in previous work (Kent and Keating, 2004) indicate that there were too few participants in the current study for the results of comparisons between professional groups to be generalizable to the broader population. Therefore, although a summary of the modal responses of participating professional groups are included as Appendix 2, no analysis of these data using inferential statistics was performed. Of the responders, 273 (50.2%) were classified as working predominately as clinicians and 69 (12.7%) were classified as working predominately as researchers. A further 202 (37.1%) did not fall within either classification, as their clinical and research workload was too mixed to be classified with our arbitrary classification criteria. Regardless of classification, the views of all responders were included in the data used to model a conceptual framework.
3.2. Question responses Participant responses to questions 1, 2, 4–7 are shown in Fig. 3. Their responses to these questions showed central tendencies, with most respondents being prepared to commit to responses that favoured one end of the response scale. Most respondents favoured the beliefs that NSLBP is many conditions (88.4%), that clinically useful subgroups are most likely to be classified on the basis of symptoms and signs rather than
Table 1 Demographic details of responders. Overall (n ¼ 544) Age: mean (SD), range Years of clinical practice: mean (SD), range Physiotherapists General Medical Practitioners Specialist Physicians (e.g. Rheumatologists, etc.) Musculoskeletal Medicine Practitionersa Surgeons Chiropractors Osteopaths Psychologists Other occupation Missing occupation Predominantly clinicians (20% of time is spent caring for people with NSLBPb & no time spent researching NSLBP) Predominantly researchers (20% of time is spent caring for people with NSLBP & >20% time spent researching NSLBP) Special interest in low back pain a
38.5 (11.2), 21–80 13.2 (13.2), 0–44 384 (70.3%) 41 (7.5%) 27 (4.9%) 16 (2.9%) 14 (2.6%) 9 (1.6%) 4 (0.7%) 3 (0.5%) 42 (7.7%) 4 (0.7%) 273 (50.2%) 69 (12.7%) 374 (68.8%)
Musculoskeletal Medicine Practitioners either have a university-based postgraduate qualification in orthopaedic, spinal and musculoskeletal evaluation and treatment, or have a special interest in orthopaedic, spinal and musculoskeletal evaluation and treatment. b NSLBP ¼ nonspecific low back pain.
pathoanatomy (62.5%), that the pain (65.1%) and disability (57.7%) associated with acute NSLBP are more physiological than psychological, and that the pain (68.1%) and disability (72.1%) associated with chronic NSLBP are more psychological than physiological. However, within these proportions of respondents, participants varied in their strength of belief. Similarly, although participants’ responses showed these central tendencies, their beliefs also traverse the full spectrum of response options, indicating a diversity of beliefs. Where present, missing data were less than 1.0% and were therefore ignored. Table 2 shows responses to the question 3 ‘For providing care in nonspecific low back pain, how useful are measures of physical impairment, pain, activity limitation, participation restriction, psychosocial function?’ Most responses indicate that participants believe all these domains of health status to be quite useful, however, responses also ranged across the full spectrum from ‘not useful at all’ to ‘extremely useful’. Where present, missing data were less than 0.6% and were therefore ignored. Overall, for the 11 questions and part-questions about the nature and behaviour of NSLBP, participant responses ranged across nearly all the possible response options for each question. As all these questions about NSLBP had seven possible response options, there were a total of 77 possible responses. Participants collectively used 74 (96.1%) of these response options, illustrating the very diverse views that were expressed. 3.3. Clusters of belief Probabilistic data-mining did not detect the presence of multiple ‘clusters of belief’ within respondents’ answers. The SNOB software reported that the variability in participants’ scores across questions 1–2 and 4–7 was best explained by all participants belonging to one population of respondents and not by any statistical models containing subgroups of participants. As there were 544 subjects in the study, it is likely that this data-mining had adequate statistical power. 3.4. Comparisons between groups Participants who had a special interest in low back pain reported similar beliefs to those without this special interest. In the same way, participants who were predominately clinicians reported similar beliefs to those who were predominantly researchers, however, there was one question where they differed. For the question ‘For providing care in NSLBP how clinically useful are measures of physical impairment?’ the modal response from clinicians was 6 and for researchers was 2 (P ¼ .000) indicating that people working predominantly as clinicians think that physical impairment measures are much more useful than people working predominantly as researchers. 4. Discussion This research aimed to develop a conceptual framework of NSLBP from the views of people who treat and research NSLBP. The first step in this process was to test the hypothesis that beliefs within that framework are diverse but discretely ‘clustered’. Very diverse beliefs were reported but multiple ‘clusters of belief’ were not detected by data-mining. Therefore, it was not possible to undertake the further aims of testing if such clusters were associated with demographic factors or with views regarding the clinical usefulness of different measures of health status. Reinforcing earlier findings (Kent and Keating, 2004), there was variability in the beliefs commonly reported by participants from
38.4%
40% 30.7%
30% 19.3%
20% 10% 2.4% 2.8% 1.1%
0%
Proportion of respondents
50%
1
2
5.3%
3
One condition
4
5
6
7
Proportion of respondents
50% 36.5%
30%
26.7%
26.3%
20% 10%
6.5% 1.9%
0%
1
2.0%
2
3
4
5
21.5%
20% 10% 0%
16.9% 12.2%
8.9%
7.6%
2
3
4.1%
1
4
5
6
Pathoanatomy Unsure which method
0.0%
6
7
50% 40% 31.9%
30%
25.8%
24.5%
20% 12.1%
10%
4.5%
1.3%
0%
All psychological
1
0.0%
2
3
All physiological
4
5
Equally both
6
Question 5: In ACUTE NSLBP the DISABILITY is based in physiological or psychological phenomena?
50% 42.4%
40% 25.1%
23.1%
20% 10% 0%
6.3% 0.4%
1
2.2%
0.6%
2
3
All physiological
4 Equally both
5
6
7
All psychological
Question 6: In CHRONIC NSLBP the PAIN is based in physiological or psychological phenomena?
7
All psychological
Question 4: In ACUTE NSLBP the PAIN is based in physiological or psychological phenomena?’
30%
7
Symptoms & signs
Proportion of respondents
Equally both
28.8%
30%
Proportion of respondents
All physiological
40%
Question 2: Subgrouping NSLBP is likely to be on the basis of pathoanatomic diagnoses or on clusters of symptoms and signs (with indistinct pathoanatomy)?
Question 1: NSLBP is one condition or a number of conditions?
40%
391
50%
Many conditions
Uncertain
Proportion of respondents
Proportion of respondents
P.M. Kent et al. / Manual Therapy 14 (2009) 387–396
50% 40%
33.7%
36.7%
30% 21.1%
20% 10% 0%
5.9% 0.0%
0.9%
1
2
All physiological
1.7%
3
4 Equally both
5
6
7
All psychological
Question 7: In CHRONIC NSLBP the DISABILITY is based in physiological or psychological phenomena?
Fig. 3. Responses to questions regarding nonspecific low back pain (NSLBP).
different professional disciplines (Appendix 2) and these may be useful in hypothesis setting for further studies designed to sample sufficient participants for the testing of significant differences of opinion across disciplines. The finding that most respondents favoured the belief that NSLBP is many conditions verifies findings from our earlier study of clinician beliefs (Kent and Keating, 2004). The current study additionally found that low back pain researchers attending a primary care research conference had similar views. Other new insights were: that many respondents believed that clinically useful
subgroups are most likely to be classified on the basis of symptoms and signs rather than pathoanatomy; that the pain and disability associated with acute NSLBP are mostly physiological; and that the pain and disability associated with chronic NSLBP are mostly psychological. However, these findings are not sufficient in themselves to explain the diversity of beliefs. The beliefs of people with and without a special interest in LBP were similar. The beliefs of people working predominantly as clinicians and researchers only markedly diverged over one question. Those who were predominantly clinicians tended to rate the
392
P.M. Kent et al. / Manual Therapy 14 (2009) 387–396
Table 2 Responses to question: ‘For providing care in nonspecific low back pain, how useful are measures of.’.
Physical impairment Pain Activity limitation Participation restriction Psychosocial function
Median
25th–75th percentile
Range
5 5 6 6 6
4–6 4–6 5–7 5–6 5–7
1–7 1–7 1–7 1–7 1–7
1 ¼ not at all useful and 7 ¼ extremely useful.
clinical usefulness of measures of physical impairment as ‘extremely useful’ while those who were predominantly researchers tended to rate physical impairment measures as ‘not useful at all’. The criteria used to classify people based on their NSLBP workload as predominantly clinicians or researchers were arbitrary and not everyone will agree that these criteria were ideal. However, these criteria were set prior to data collection and the results reinforce earlier findings that clinicians highly value measures of physical impairment in NSLBP (Kent et al., 2009). To our knowledge, these data are the first to quantify that people active in NSLBP research place a much lower value on these measures. The strength of evidence for the clinical utility of measures of physical impairment varies depending on the purpose for which these measures are applied and a brief summary of that evidence follows. Symptoms and signs are usually assessed for four clinical purposes: making a diagnosis, informing treatment decisions, assisting estimation of likely prognosis and monitoring outcome. Despite popular sentiment, the evidence is weak for physical impairment assisting with definitive pathonatomic diagnoses in NSLBP. In mixed LBP that includes neurocompressive LBP and NSLBP, an association has been demonstrated between centralization/peripheralisation and pain reproduction on provocative discography. However, the overall accuracy of this test has been variable and moderate at best: 51.8% (Laslett et al., 2005), 66.7% (Young et al., 2003) and 70.1% (Donelson et al., 1997). Similarly, investigators have found only very weak associations between particular physical impairments and a positive response to anaesthetic blocks of the lumbar zygapophyseal joints (Schwarzer et al., 1995; Laslett et al., 2004; Laslett, 2005). For example, ipsilateral rotation and extension, which historically have been promoted as an indicator of zygapophyseal joint pain, have an overall accuracy of 41.1% (Laslett, 2005). Furthermore, the use of diagnostic injections as reference standards of definitive diagnoses remains controversial, with some arguing that pain reproduction or ablation does not unequivocally isolate the site of primary pain generation (North et al., 1996; Carragee et al., 1999). There is preliminary evidence that some physical impairments can inform the NSLBP treatment decisions of manual therapists. In mixed LBP that includes neurocompressive LBP, centralization/peripheralisation (directional preference) has been shown to be strongly predictive of the type of exercise most likely to benefit people (Long et al., 2004). Physical impairments also are statistically significant components of clinical prediction rules shown to predict response to either manipulation or stabilization exercises (Flynn et al., 2002; Childs et al., 2004; Hicks et al., 2005). There is also preliminary evidence that some physical impairments, such as centralization, limited flexion range of movement (ROM) and greater body mass are as strongly associated with prognosis (when defined as poor outcome from an episode of NSLBP) as factors from other domains of health status (Kent and Keating, 2009). However, due to disparate methods, contradictory
findings, the likely interaction between factors and highly variable study quality, considerable uncertainty still remains regarding prognostic factors. Similarly, there is also evidence (Hahne et al., 2004) supporting the common practice of manual therapists using within-treatment-session change in impairment of range of movement as a predictor of sustained improvement (overall accuracy 74–88%). Change in particular physical impairments associated with NSLBP can be reliably measured (Van Dillen et al., 1998; Hahne et al., 2004). However, change in physical impairment is very weakly associated with change in other measures of health status (such as activity limitation or participation restriction) suggesting that it cannot be used as a proxy for directly measuring change in these other domains of health status (Deyo, 1986; Mellin, 1986; Riddle, 1997; Sullivan et al., 2000). This has led to some disagreement about the utility of measurements of impairment, as third party payers are predominantly interested in reductions of activity limitation and participation restriction. However, as changes in measures of activity and participation do not typically occur within a treatment session they often do not provide information that assists with early decisions regarding the best choice of manual therapy treatment. In contrast, immediate changes in impairments can be observed in response to clinical interventions. These varying needs of outcome measures by different users may account for some observed differences in beliefs. The overall finding of the current research is that despite there being very diverse views of the nature and behaviour of NSLBP expressed by those participating in this survey, there was no apparent unifying framework that explained the diversity of their beliefs. There may be a number of reasons for this. We asked simple closed questions that sought to elicit beliefs about contentious concepts in NSLBP. It may have been that these questions were insensitive to the subtleties of participants’ beliefs and that research using qualitative methods might better elucidate a framework that explains current beliefs. It is also possible that providing a variety of specific clinical scenarios may have provided a richer data set to model a NSLBP conceptual framework. Alternatively, it may be that there is no underlying framework that explains the diversity of current beliefs. The observed diversity of beliefs may also reflect historical views coupled with limited knowledge about the etiology of NSLBP.
5. Conclusion Conflicting views exist amongst those that participated in this survey regarding the underlying nature of NSLBP. However, within the constructs that were sampled in these surveys, there does not appear to be any unifying framework that explains the diversity of current beliefs. This is likely to reflect pervasive uncertainty about the etiology and best practice assessment of NSLBP.
Acknowledgement Peter Kent is supported by a NHMRC Health Professional Fellowship (384366) and Rachelle Buchbinder by a NHMRC Practitioner Fellowship (334010). No benefits in any form have been, or will be, received from a commercial party related directly or indirectly to the subject of this manuscript. Ethics approval provided by the Cabrini Human Research Ethics Committee (05111206).
P.M. Kent et al. / Manual Therapy 14 (2009) 387–396
Appendix 1. Questionnaire
393
394
P.M. Kent et al. / Manual Therapy 14 (2009) 387–396
P.M. Kent et al. / Manual Therapy 14 (2009) 387–396
395
Appendix 2. Modal responses (most common answer) by participants from professional groups
Question
Response options
Physio (n ¼ 383)
GP (n ¼ 41)
Physic (n ¼ 27)
Musc Med (n ¼ 16)
Surg (n ¼ 14)
Chiro (n ¼ 9)
Osteo (n ¼ 4)
Psych (n ¼ 3)
Is NSLBP one condition or a number of conditions?
1¼one condition 4¼uncertain 7¼many conditions
7
7
7
7
7
7
4
4
If NSLBP were able to be classified into clinically useful subgroups, is this likely to be on the basis of pathoanatomy or symptoms & signs?
1¼pathoanatomy 4¼unsure which method 7¼symptoms & signs
6
6
2
6
6
7
1
4
For providing care for NSLBP, measures of Physical Impairment are. For providing care for NSLBP, measures of Pain are . For providing care for NSLBP, measures of Activity Limitation are . For providing care for NSLBP, measures of Participation Restriction are . For providing care for NSLBP, measures of Psychosocial Function are.
1¼not at all useful 7¼extremely useful
6
6
6
5
4
5
2
3
5
5
4
5
4
6
7
2
6
6
6
7
4
5
4
7
6
6
7
5
5
5
5
7
6
6
6
7
5
5
6
7
The pain in acute NSLBP is . The pain in chronic NSLBP is . The disability in acute NSLBP is . The disability in chronic NSLBP is .
1¼all physiological 4¼equally both 7¼all psychological
3 5 3
3 5 3
4 5 3
2 5 4
3 4 2
4 4 3
3 4 4
2 3 2
6
5
6
6
6
4
6
6
References Brennan GP, Fritz JM, Hunter SJ, Thackeray A, Delitto A, Erhard RE. Identifying subgroups of patients with acute/subacute ‘‘nonspecific’’ low back pain – results of a randomized clinical trial. Spine 2006;31(6):623–31. Carragee EJ, Tanner CM, Yang B, Brito JL, Truong T. False-positive findings on lumbar discography. Reliability of subjective concordance assessment during provocative disc injection. Spine 1999;24(23):2542–7. Childs JD, Fritz JM, Flynn TW, Irrgang JJ, Johnson KK, Majkowski GR, Delitto A. A clinical prediction rule to identify patients with low back pain most likely to benefit from spinal manipulation: a validation study. Annals of Internal Medicine 2004;141: 920–8. Delitto A, Erhard RE, Bowling RW. A treatment-based classification approach to low back syndrome: identifying and staging patients for conservative treatment. Physical Therapy 1995;75(6):470–89. Deyo R, Rainville J, Kent D. What can the history and physical examination tell us about low back pain. Journal of the American Medical Association 1992;268(6):760–5. Deyo RA. Comparative validity of the sickness impact profile and shorter scales for functional assessment in low-back pain. Spine 1986;11(9):951–4. Dillman D. Mail and telephone surveys, the total design method. New York: John Wiley and Sons; 1978. Donelson R, Aprill C, Medcalf R, Grant W. A prospective study of centralization of lumbar and referred pain: a predictor of symptomatic discs and annular competence. Spine 1997;22(10):1115–22. Donicar S. A review of unquestioned standards in using cluster analysis for datadriven market segmentation. Australia: Faculty of Commerce, University of Wollongong, http://ro.uow.edu.au/commpapers/273; 2002 [accessed 24.06.08]. Flynn T, Fritz JW, Whitman M, Wainner RS, Magel J, Rendeiro D, Butler B, Garber M, Allison S. A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improvement with spinal manipulation. Spine 2002;27(24):2835–43. Fritz JM, Delitto A, Erhard RE. Comparison of classification-based physical therapy with therapy based on clinical practice guidelines for patients with acute low back pain a randomized clinical trial. Spine 2003;28(13):1363–72. Hahne AJ, Keating JL, Wilson SC. Do within-session changes in pain intensity and range of motion predict between-session changes in patients with low back pain? Australian Journal of Physiotherapy 2004;50(1):17–23. Hicks GE, Fritz JM, Delitto A, McGill SM. Preliminary development of a clinical prediction rule for determining which patients with low back pain will respond to a stabilization exercise program. Archives of Physical Medicine & Rehabilitation 2005;86(9):1753–62.
Kent PM, Keating J. Do primary-care clinicians think that non-specific low back pain is one condition? Spine 2004;29:1022–31. Kent PM, Keating JL. Can we predict poor recovery from recent-onset nonspecific low back pain? A systematic review. Manual Therapy 2008;13:12–28. Kent PM, Keating JL, Taylor NF. Methods used by primary care clinicians to assess recent-onset non-specific low back pain. Manual Therapy 2009;14(1):88–100. Laslett M. Diagnostic accuracy of the clinical examination compared to available reference standards in chronic low back pain patients. Division of Physiotherapy, Linkoping, Sweden: Linkopings Universitet; 2005. p. 113. Laslett M, Oberg B, Aprill CN, McDonald B. Zygapophysial joint blocks in chronic low back pain: a test of Revel’s model as a screening test. BMC Journal of Musculoskeletal Disorders 2004;5(1):43. Laslett M, Oberg B, Aprill CN, McDonald B. Centralization as a predictor of provocation discography results in chronic back pain, and the influence of disability and distress on diagnostic power. The Spine Journal 2005;5:370–80. Leboeuf-Yde C. Low back pain: time to get off the treadmill. Journal of Manipulative and Physiological Therapeutics 2001;24(1):63–6. Long A, Donelson R, Fung T. Does it matter which exercise? A randomized control trial of exercise for low back pain. Spine 2004;29:2593–602. Mellin G. Chronic low back pain in men 54–63 years of age. Correlations of physical measurements with the degree of trouble and progress after treatment. Spine 1986;11(5):421–6. Monash Data Mining Centre, Monash University, Clayton, VIC, Australia. SNOB datamining software. Murphy DR, Hurwitz EL. A theoretical model for the development of a diagnosisbased clinical decision rule for the management of patients with spinal pain. BMC Musculoskeletal Disorders 2007;8:75. doi:10.1186/1471-2474-8-75. North RB, Kidd DH, Zahurak M, Piantadosi S. Specificity of diagnostic nerve blocks: a prospective, randomized study of sciatica due to lumbosacral spine disease. Pain 1996;65(1):77–85. O’Sullivan PB, Beales DJ. Diagnosis and classification of pelvic girdle pain disorders – Part 1: A mechanism based approach within a biopsychosocial framework. Manual Therapy 2007a;12:86–97. O’Sullivan PB, Beales DJ. Diagnosis and classification of pelvic girdle pain disorders, Part 2: Illustration of the utility of a classification system via case studies. Manual Therapy 2007b;12:e1–12. Petersen T, Laslett M, Thorsen H, Manniche C, Ekdahl C, Jacobsen S. Diagnostic classification of non-specific low back pain. A new system integrating patho-anatomic and clinical categories. Physiotherapy Theory & Practice 2003;19(4):213–37. Petersen T, Thorsen H, Manniche C. Classification of non-specific low back pain: a review of the literature on classifications systems relevant to physiotherapy. Physical Therapy Reviews 1999;4(4):265–81.
396
P.M. Kent et al. / Manual Therapy 14 (2009) 387–396
Riddle D. Relationships between disability and physical impairment in patients with low back pain. Virginia Commonwealth University; 1997. Rosner A. Fables or foibles: inherent problems with RCTs. Journal of Manipulative & Physiological Therapeutics 2003a;26(7):460–7. Rosner A. Response to meta-analysis published by Assendelft et al., in the Annals of Internal Medicine. Dynamic Chiropractic 2003b;21(19):34–6. Schwarzer AC, Wang SC, Bogduk N, McNaught PJ, Laurent R. Prevalence and clinical features of lumbar zygapophysial joint pain: a study in an Australian population with chronic low back pain. Annals of the Rheumatic Diseases 1995;54(2):100–6. Sullivan MS, Shoaf LD, Riddle DL. The relationship of lumbar flexion to disability in patients with low back pain. Physical Therapy 2000;80(3):240–50. Upal MA, Neufeld EM. Comparison of unsupervised classifiers. In: Proceedings of the first international conference on information, statistics and induction in science (ISIS). River Edge, NJ, USA:: World Scientific; 1996. p. 342–53. Van Dillen LR, Sahrmann SA, Norton BJ, Caldwell CA, Fleming DA, nnell MK, Woolsey NB. Reliability of physical examination items used for classification of patients with low back pain. Physical Therapy 1998;78(9):979–88.
Victorian WorkCover Authority. Clinical framework for the delivery of physiotherapy services to injured workers. Melbourne, Australia: Victorian WorkCover Authority; 2004. p. 1–15. Waddell G. A new clinical model for the treatment of low-back pain. Spine 1987;12(7):632–44. Waddell G. The back pain revolution. Edinburgh: Churchill Livingstone; 1998. p. 438. Wallace CS. Statistical and inductive inference by minimum message length. New York, USA: Springer; 2005. p. 429. Wallace CS, Boulton DM. An information measure for classification. Computer Journal 1968;11(2):185–94. Wallace CS, Dowe DL. MML clustering of multi-state, Poisson, von Mises circular and Gaussian distributions. Statistics and Computing 2000;10(1):73–83. World Health Organisation. International classification of functioning, disability and health. Geneva, Switzerland: World Health Organisation; 2001. Young S, Aprill C, Laslett M. Correlation of clinical examination characteristics with three sources of chronic low back pain. The Spine Journal 2003;3:460–5. Zusman M. Structure-oriented beliefs and disability due to back pain. Australian Journal of Physiotherapy 1998;44(1):13–20.
Manual Therapy 14 (2009) 397–403
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Original Article
The validity and intra-tester reliability of a clinical measure of humeral head position Leanda McKenna*, Leon Straker, Anne Smith Curtin University of Technology, School of Physiotherapy, G.P.O. Box U1987, Perth, WA 6845, Australia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 December 2007 Received in revised form 6 May 2008 Accepted 29 June 2008
The purpose of this study was to determine the degree of criterion validity and intra-tester reliability of humeral head palpation in subjects with shoulder pathology. The study also sought to determine whether there was any effect of arm position on humeral head position in subjects with shoulder pathology. In a same day repeated measures design, 27 subjects had the distance between the most anterior portion of the humeral head and the anterior edge of the acromion measured by a radiologist using MRI (supine), and by a physiotherapist using palpation and photography (supine, sit with arm in neutral and in abduction). The Standard Error of Measurement (SEM) for the difference between MRI and palpation ranged from 3.4 to 4.4 mm and correlated significantly with palpation measures in sit (r ¼ 0.57–0.64, p 0.002). The Intraclass Correlation Coefficients (ICCs) and SEMs for intra-tester reliability were 0.85 and 2.6 mm for supine, 0.86 and 2.2 mm for sit (glenohumeral neutral), and 0.91 and 3.0 mm for sit (glenohumeral abduction). Significant differences between the positions of sit neutral and sit with abduction were found (p < 0.001). Humeral head palpation in sit abduction demonstrates sufficient validity and reliability for clinical use. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Shoulder Palpation Validity Reliability
1. Introduction Subacromial impingement is a well recognized, painful and potentially limiting condition. Theoretically, anterior position of the humeral head in relation to the acromion may compromise the subacromial space (Michener et al., 2003), because the humerus is in greater proximity to anterior structures in elevation (Ludewig and Cook, 2002; Werner et al., 2004; Bach and Goldberg, 2006). Measuring the habitual anterior position of the humeral head in relation to the acromion may therefore be important when assessing patients with impingement syndromes. Measurement of in vivo humeral head position usually quantifies the minimal acromiohumeral or coracohumeral interval by imaging or complicated methods. These methods are not practical for repeated clinical use and a simple, palpatory technique may be more appropriate. Published methods have concentrated on evaluating the superior position of the humeral head in relation to the acromion. However, the superior aspect of the humeral head under the acromion is difficult to palpate and assessment of the superior position of the humeral head in relation to the acromion may not be possible in the clinic. Given
* Corresponding author. Tel.: þ61 8 9266 3660; fax: þ61 8 9266 3699. E-mail address:
[email protected] (L. McKenna). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.06.004
that anterior and superior migrations of the humeral head (and resultant proximity to subacromial structures) during flexion appear to be closely linked (Harryman et al., 1990a,b; Werner et al., 2004), anterior assessment may also provide insight into acromiohumeral proximity. The anterior portion of the humeral head and acromion are easily located by palpation may provide a useful technique to enhance the therapeutic decision-making process and has been used previously as an assessment tool (Mckenna et al., 2001). 1.1. Validity and reliability of a palpatory method Humeral head palpation has been examined for reliability (Bryde et al., 2005) in a healthy population, but not in a population with pathology. Additionally, there does not appear to be any validity study examining anterior humeral head position assessment that is applicable to impingement. 1.2. Criterion standard X-rays have previously been used as the criterion of gold standard against which palpatory techniques were judged in live subjects, but has possible errors of magnification, projection, patient positioning and identification of landmarks (Vaatainen et al., 1991; Sutherland and Bresina, 1992; Kladny et al., 1996; Duralde and Gauntt, 1999; Graichen et al., 1999b; Burckhardt et al.,
398
L. McKenna et al. / Manual Therapy 14 (2009) 397–403
2000; Schulze and d’Hoedt, 2001). Ultrasound accuracy is highly operator dependent (Tirman et al., 1997; Torriani and Kattapuram, 2003; Roemer et al., 2005) and CT involves high levels of radiation exposure. As a criterion standard, MRI appears to have the advantages of high tissue contrast (Kladny et al., 1996; Eckstein et al., 2001) with high resolution (Torriani and Kattapuram, 2003) and therefore may be theoretically superior in measuring distances.
1.3. Influence of testing position One disadvantage of using MRI as a validation tool is that closed MRI machines do not allow shoulder joint elevation and images are generally taken in a neutral position that can be sustained by the patient. Closed MRI is often the standard choice for patients with impingement type symptoms who require imaging. Thus any validity study in patients with impingement that uses closed MRI as the reference comparison should use the same glenohumeral joint position as that utilized within the MRI machine. However, neutral positions may not be provocative enough to elicit a pathological habitual humeral head position. Elevation is recognized as a provocative position for glenohumeral impingement because the subacromial space narrows in elevation (Warner et al., 1994; Allmann et al., 1997; Moffet et al., 1998; Graichen et al., 1999a,b; Hinterwimmer et al., 2003) and may be reduced to 1.0 mm in patients with impingement (Allmann et al., 1997; Graichen et al., 1999b). Measurement of humeral head position in elevation is therefore likely to be used where clinicians wish to elicit symptoms that are difficult to provoke. Validity therefore needs to include both the neutral and the elevation positions to address the dual issues of appropriate validation and clinical need.
1.4. Purpose of this study Firstly, to determine the degree of criterion validity of a palpatory method of humeral head position measurement in subjects who have shoulder pathology. Secondly, to determine the degree of intra-tester reliability of a palpatory method of humeral head position measurement in subjects who have shoulder pathology. Thirdly, to determine the effect of arm position on humeral head position (measured by a palpatory method) in subjects who have shoulder pathology. 2. Methods 2.1. Study design This was a double blind validity and intra-tester reliability study utilizing a qualified physiotherapist (LMcK) and consultant radiologist. Curtin University of Technology Human Research Ethics Committee approved this study and all rights of the individual were protected. 2.2. Subjects Subject recruitment was conducted over one year as shown in Fig. 1. Participating surgeons invited patients undergoing shoulder MRI into the study if they were aged above 18, and had a provisional diagnosis of impingement or rotator cuff disease, as determined by an orthopaedic surgeon. Subjects were excluded from the
Patients referred to Orthopaedic Specialist by General Practitioner
Some patients referred for MRI of the shoulder by Orthopaedic Specialist Patients that met the inclusion no criteria invited to join study
yes
1 patient was unable to allow extra time for testing
32 patients indicated they would like to join the study.
no
3 patients could not be contacted 1 patient could not be tested as the testing room was double booked
yes 25 patients tested immediately yes prior to MRI 27 subjects in study yes
2 patients tested immediately post MRI
Patients underwent MRI Fig. 1. Flowchart of study participants.
L. McKenna et al. / Manual Therapy 14 (2009) 397–403
study if they had a suspected shoulder fracture or grossly unstable glenohumeral joint. Potentially interested patients received written information from the surgeon’s secretary to read in the waiting room. Those interested in participating in the study indicated to the secretary that their phone number could be given to the chief investigator. Inclusion to the study was dependent upon the volunteers’ reading and signing an informed consent immediately prior to measurement at the MRI centre. A priori analysis suggested that a minimum of 24 subjects was needed to achieve 80% power to detect a systematic difference between MRI and palpatory measures of 3.0 mm (5.0 mm) using a 2-tailed t-test (a ¼ 0.05). A sample of 26 subjects has 80% power (a ¼ 0.05) to detect a correlation of 0.52.
399
weighted sagittal and proton density weighted fat saturated axial sequences in patients requiring gadolinium injection (see Fig. 3). Sagittal image resolution matrix was 256 192 pixels (repetition time 500–3250 ms, echo time 14.0–35.4 ms) and 384 224 for the axial views (repetition time 640–3000 ms, echo time 13.5– 28.5 ms). Slice thickness was 3.0 mm (1.0 mm inter-slice gap) and field of view 14.0 cm 14.0 cm. Using digital MRI images, the radiologist identified the anterior margin of the acromion and the anterior margin of the Deltoid where it overlaid the most anterior portion of the humeral head (see Fig. 3). Values drawn from images were made available to the chief investigator at the conclusion of the study. 2.4. Data processing
2.3. Data collection 2.3.1. Palpatory method Subjects completed a questionnaire, immediately prior to palpation testing at the MRI centre, which enquired about date of birth, hand dominance and shoulder pathology. Height and weight were measured. The shoulder was prepared with the application of a patch of clear adhesive plastic dressing (Tegaderm, 3M Health Care, St. Paul, Minnesota, USA). Subjects were tested using a random order for position: (1) supine with arm by side and slight external rotation, (2) sitting, neutral glenohumeral joint, and (3) sitting, arm supported in abduction. The arm was supported in abduction using a removable wooden platform set at axillary height and attached to the camera tripod. The anterior acromion was palpated and marked with a line. The examiner placed a mark on their index finger that corresponded with the point of maximal palpatory pressure. The most anterior aspect of the subject’s humeral head was palpated and the examiner placed their marked finger on this point. A photograph (see Fig. 2) was taken using a digital camera (Kodak DC4800, Eastman Kodak Company, NY) mounted 20.0 cm superiorly above the shoulder. All subject marks were removed and the procedure was repeated twice for each position.
Digital photographic images were corrected for lens pincushion distortion using a 32% correction from Colour Science Image factory Home Edition program (http://www.colour-science.com/if/ registration.htm). A custom program (High-Tech Laboratories, Perth, Western Australia) was used to calculate the humeral head distance following correction for scaling and parallax error using the formulas shown in Appendix 1. The scaling factor was taken from a rule included in the field of view. Calculation of parallax error required the vertical distance (dv in Appendix 1) between the acromion and most anterior portion of the anterior humeral head as measured from MRI by the chief investigator. Using the computer software, the investigator placed digital points on either end of the marked anterior margin of the acromion and at the marked investigator’s finger (placed against the anterior point of the humeral head). The software calculated the line of best fit along the acromion and the perpendicular distance from the line to the most anterior point of the humeral head (see Appendix 1 and Fig. 2). 2.5. Statistical analysis Statistical processing was performed using SPSS (Statistical Packages for the Social Sciences) version 13 for Macintosh. A trial
2.3.2. MRI The subject underwent MRI in supine with arm by side and in slight external rotation. A closed MRI system (General Electric Sigma 1.5T, Milwaukee, USA) with dedicated phased array shoulder coil was used. Two image sequences were used – either proton density weighted sagittal and fat saturated axial sequences or T1
Fig. 2. Photograph showing palpation technique for measurement of anterior humeral head to acromion distance (viewed from the superior aspect of the shoulder). Dashed line indicates the extrapolated acromion line. The continuous line (points 1–2) indicates the perpendicular distance between the examiner’s finger overlaying the anterior humeral head and the dashed acromial line.
Fig. 3. MRI showing measurement of anterior humeral head to acromion distance. Dashed line extrapolates the most anterior portion of the humeral head to level with the acromion. Continuous line (points 1–2) indicates the distance between anterior acromion and the dashed line.
400
L. McKenna et al. / Manual Therapy 14 (2009) 397–403
order effect was examined with a Repeated measures Analysis of Variance (RANOVA), with posthoc examination of individual trial comparisons. Validity was examined for agreement between the two methods of measuring humeral head position by calculating a mean difference between methods (Bland and Altman, 1986), Pearson’s correlation coefficient (r) and Standard Error of Measurement (SEM). SEM ¼ OMSE, where MSE is the square root of the RANOVA Mean Square Error (Stratford and Goldsmith, 1997). Intra-tester reliability of the palpatory method was examined using Intraclass Correlation Coefficient (ICC) for consistency and for absolute agreement (McGraw and Wong, 1996), and SEM. Comparison between positions was analysed using a one factor RANOVA. Posthoc analysis of contrasts between individual positions was performed to identify significant differences in position. Bivariate correlations between arm positions were also performed.
Table 2 Anterior humeral head to acromion distances (mm) as measured by palpatory and MRI methods (mean (sd)). Palpation
Supine Sit neutral Sit abduction
MRI
Trial 1
Trial 2
Trial 3
Mean of 3 trials
18.7 (6.5) 17.0 (6.4) 22.9 (7.8)
18.3 (6.4) 17.0 (6.6) 21.5 (7.8)
17.8 (5.5) 16.0 (6.1) 20.7 (8.0)
18.2 (5.8) 16.7 (6.1) 21.7 (7.5)
14.0 (5.3) NA NA
Larger values indicate a more anterior humeral head, in comparison to the acromion.
(F ¼ 49.43, p < 0.001). These two positions were also highly correlated to each other (r ¼ 0.89, p < 0.001), whereas there appeared to be no correlation with supine measures (sit neutral–supine r ¼ 0.26, p ¼ 0.200; sit abduction–supine r ¼ 0.12, p ¼ 0.555).
4. Discussion
3. Results One subject’s data were excluded from analysis as the MRI showed evidence of previous anterior dislocation (marrow oedema was evident within the posterosuperior aspect of the humeral head, with slight flattening) and therefore possible instability, despite fulfilling initial inclusion criteria. Patient demographic data are presented in Table 1. There were 21 males and 5 females. 54% of subjects had dominant arm symptoms. Pathology found at MRI included calcification, atrophy, thickening, fraying and tears of the rotator cuff tendons, bursal thickening and effusion, mild subluxation, tearing and rupture of the biceps tendon, acromioclavicular joint degeneration, glenohumeral joint synovitis, SLAP lesions, and paralabral cysts. 7 Subjects had been injected with gadolinium to enhance diagnostic findings. Mean anterior humeral head distance from the anterior acromion measured between 14.0 and 21.7 mm, depending on position and method of assessment (see Table 2). There was a trial effect within the palpatory method for sit abduction (F ¼ 3.49, p ¼ 0.039) due to a difference between trial 1 and trial 3 (F ¼ 5.31, p ¼ 0.030). Other positions did not demonstrate a significant trial effect. The mean of three trials was used for further analysis. Soft tissue overlay measured between the anterior humeral head and skin on MRI was on average 17.7 mm thick (Standard deviation (sd) ¼ 0.5 mm, range ¼ 8.0–32.0 mm). MRI measured the average anterior difference between the humeral head and acromion as 4.2, 2.7 (see Fig. 5) and 7.7 mm smaller than the palpatory measurement of the humeral head in supine, sit neutral and sit abduction, respectively (see Table 3). A pattern of higher validity for palpation in sit positions compared to supine positions was demonstrated by lower SEMs and greater correlations between palpation in sit and MRI (see Table 3 and Fig. 4). Intra-tester reliability revealed ICCs (repeated measures in each position) that were all above 0.85. The SEMs were all below 3.0 mm (see Table 4). There was a significant difference between positions (F ¼ 7.07, p ¼ 0.009) with sit abduction demonstrating the most anterior humeral head position (see Table 2). This overall difference was due to the significant difference between sit abduction and sit neutral
Table 1 Subject attributes. Attribute (unit)
Mean (standard deviation)
Range
Age (years) Weight (kg) Height (cm) Chronicity of pain (months)
52.0 87.2 175.3 30.5
22.4–76.3 60.0–146.0 157.0–188.5 2.0–120.0
(13.6) (15.8) (8.6) (32.3)
4.1. Validity Previous criterion validity studies of palpation assessment techniques of the humeral head were confined to assessment of inferior subluxation. These studies reported slightly higher correlation coefficients (rho ¼ 0.70–0.76) than that found in this study (r ¼ 0.64), which may be a result of the differences in recording (fingerbreadths versus mm) and the type of correlation coefficients used (Spearman versus Pearson). Despite finding similar correlations to each other, these authors have concluded the technique to be both valid (Prevost et al., 1987; Boyd et al., 1993) and invalid (Hall et al., 1995). These conclusions were based only on correlations and demonstrate the subjective nature of using an association to decide whether to use a technique or not. This demonstrates the need for reporting additional statistical analyses such as SEM and mean difference to allow adequate assessment of validity. Comparison of MRI to anterior palpation in the present study demonstrated that MRI values were consistently smaller than palpation values, which is to be expected as palpation included overlying soft tissue. This systematic mean difference was greatest in sit abduction (see Table 3), possibly due to the superior migration of deltoid during abduction. The systematic differences between MRI and palpation in upright positions were also expected, given the changing orientation of structures in the different body and glenohumeral positions. Systematic differences between MRI and palpation in supine measures may also have been caused by the longer measurement period for MRI, which may have allowed greater soft tissue creep. Random error factors that may have affected validity in all positions include soft tissue creep, the use of gadolinium contrast injection in 7 patients, phasic muscle firing, examiner line of vision difficulties and MRI accuracy. Patients may have different rates of creep (e.g. patients with occult glenohumeral laxity versus those with rotator cuff thickening) and this may have contributed to random error. Contrast injection was rejected as a contribution to
Table 3 Differences and relationships of palpatory and MRI measures for anterior humeral head to acromion distance. Position
Mean difference (mm) (LOA)
SEM (mm)
r (p)
MRI – supine MRI – sit neutral MRI – sit abduction
4.2 (7.3 to 1.2) 2.7 (4.6 to 0.7) 7.7 (10.2 to 5.2)
5.3 3.4 4.4
0.10 (0.630) 0.64 ( 0.10). Mean angles (95 % CI) for imitated and facilitated conditions were: flat 19.0 (15.0–23.0) and 18.0 (13.5–22.5) deg, long lordosis 16.2 (12.2–20.2) and17.6 (13.4–21.8) deg, short lordosis 18.9 (14.3–24.5) and 22.0 (17.1–26.9) deg. Thoraco-lumbar angle: At the thoraco-lumbar angle (Fig. 3), slump was kyphotic (19.2 (16.2–22.2) deg), flat showed a small degree of kyphosis in most subjects (imitated: 4.6 (0.0–9.2) deg, facilitated: 3.4 (0.5–6.3) deg conditions), long lordosis showed a small degree of lordosis (imitated: 2.6 (7.0 to 0.7) deg, facilitated: 3.0 (6.8 to 0.8) deg), and short lordosis showed a small degree of kyphosis at thoraco-lumbar angle (imitated: 3.1 (0.6 to 6.8) deg, facilitated: 3.8 (0.3–7.3) deg). Analysis of variance showed that the thoraco-lumbar angle was significantly more kyphosed in slump than the other postures (p < 0.001); similar between flat and short lordosis (p > 0.40); and significantly more lordosed in long lordosis than in the other postures (p < 0.001). Subjects were able to
Long lordosis
Short lordosis
Fig. 2. The four postures examined in this study. Postures were defined by the curve directions at thoraco-lumbar and lumbar regions. Angles were measured at thoraco-lumbar and lumbar regions as indicated by the arcs.
A.P. Claus et al. / Manual Therapy 14 (2009) 404–408
Lordosis
Kyphosis
Slump Imitated Facilitated
Flat
Long lordosis Short lordosis
407
the imitated 4.1 (7.6 to 0.6) deg, but a larger degree of lordosis in the facilitated condition 15.0 (18.3 to 11.7) deg (p < 0.001). Analysis of variance showed that the lumbar angle was significantly more kyphosed in slump than the other postures (p < 0.001); imitated flat and imitated short lordosis angles were similar (p ¼ 0.14, NS in Fig. 3) and close to 0 ; but the facilitated short lordosis and facilitated long lordosis angles were similar (p ¼ 0.32, NS in Fig. 3) and more lordosed than the other postures (p < 0.001). At the lumbar angle, results show that subjects achieved a lordotic curve in the imitated and facilitated long lordosis, but for the short lordosis most subjects required facilitation to achieve an angle that was beyond flat. 4. Discussion
-15
-5
0
15
5
25
35
15
25
Thoracic Angle (deg)
Slump
Flat
Long lordosis
NS
Short lordosis -25
-15
-5
0
5
Thoraco-lumbar Angle (deg)
Slump
Flat
NS
Long lordosis
* NS
Short lordosis
* -25
-15
-5
0
5
15
25
Lumbar Angle (deg) Fig. 3. Thoracic, thoraco-lumbar and lumbar angles are shown with mean and 95% confidence intervals for each of the postures, in the imitated and facilitated intervention conditions. *p < 0.01 – comparison between imitate and facilitate conditions that was statistically different; imitated and facilitated were similar within posture for all other comparisons. Grey bars with NS – comparison between postures where there was no difference; thoraco-lumbar and lumbar angles were different between postures for all other comparisons.
imitate thoraco-lumbar angles similar to those in the facilitated condition for each posture. Lumbar angle: At the lumbar angle (Fig. 3), slump was kyphotic 10.8 (5.8–15.8) deg, flat was close to zero (imitated: 1.5 (5.2 to 2.2) deg, facilitated: 0.1 (1.8 to 2.0) deg), long lordosis showed a lordotic curve (imitated: 9.2 (12.4 to 6.0) deg, facilitated: 13.4 (17.3 to 9.5) deg), and short lordosis showed a small degree of lordosis for
The results show that most subjects could not attain the short lordosis spinal curves (kyphotic/flat thoraco-lumbar, and lordotic lumbar region) with visual and verbal description alone. Facilitation and feedback were needed in the short lordosis to achieve a lumbar angle that was more lordotic than the flat posture. Yet most subjects were able to imitate the slump, flat and long lordosis postures (similar curve directions at thoraco-lumbar and lumbar regions) without manual facilitation. The short lordosis posture was unique in demanding different directions of spinal curve at thoraco-lumbar and lumbar regions. Results from the thoraco-lumbar angle were closely matched between the imitated and facilitated condition for all postures, and achieved the appropriate directions of spinal curve. It is interesting to note that the failure to intuitively imitate a spinal curve mostly occurred at the lumbar angle (short lordosis required facilitation to lordose at the lumbar angle). This raises the question, if a short lordosis posture is commonly adopted in standing (Berthonnaud et al., 2005) why wouldn’t it be easily achievable in sitting? The difference in hip positions between standing and sitting could be a reason for lordosis to be commonly achieved in standing but not in sitting. In a radiographic study from 1953, it was observed that hip flexion to 90 in side-lying (the hip angle commonly advocated in sitting) caused the subjects to adopt a kyphotic lumbar curve, and hip extension caused a lordotic lumbar curve (Keegan, 1953). Although subjects in the current study sat with their knees below the height of their hips, the observation that hip flexion encourages a kyphotic lumbar curve might explain why subjects in the current study had difficulty in achieving a lumbar lordosis in sitting, and subjects in another study (Scannell and McGill, 2003) had difficulty in maintaining a lumbar lordosis in sitting. These results give reason to reconsider what is the natural posture for the lumbar spinal joints. Although the anatomical position is often assumed to be a natural or ‘ideal’ lordosed posture for the lumbar spine, it is derived from postures with the hip in an extended position (standing, supine or prone). The anatomical position is not necessarily a mid-range or natural resting position for joints (e.g. glenohumeral, tibiofemoral, hip or talocrural). Despite the wedge-shape of vertebral bodies and intervertebral discs, the assumed natural or ‘ideal’ lordosis could be at least partly due to anterior tilt of the sacral base and pelvis, as a result of hip extension. The relative merits of various spinal curves need to be understood. What then, is a ‘good’ spinal posture to adopt in sitting? Kyphosed lumbar postures require less muscle activity than upright postures (Floyd and Silver, 1951; O’Sullivan et al., 2006a), but may cause greater stress to articular and ligamentous structures (Gracovetsky et al., 1990). Upright lumbar postures such as flat, long lordosis and short lordosis examined in this study are likely to approach mid-range for the lumbar joints. Although mid-range postures avoid end-range stress to ligaments, they are prone to bend, twist and shear (buckling) (Crisco and Panjabi, 1992; Adams,
408
A.P. Claus et al. / Manual Therapy 14 (2009) 404–408
1995). Computational modeling of spinal stability with the objective to prevent buckling of spinal segments has shown that midrange low-load postures place more demands upon neuromuscular control than end-range postures or high-load tasks do (Cholewicki and McGill, 1996). Neuromuscular control of spinal segments in mid-range was hypothesised to be an important determinant for safety of loading/movement of the spine (Panjabi, 1992a,b). In support of this hypothesis, a recent study with subjects in an upright semi-seated posture showed that a deficit in one aspect of neuromuscular control was a predictor for consequent development of low back pain (Cholewicki et al., 2005). The present study showed that sitting in the flat and long lordosis postures was intuitive, but further facilitation was needed for the neuromuscular control to adopt the short lordosis posture, as advocated in clinical rehabilitation texts (Lee, 2003; O’Sullivan, 2004). This suggests that the short lordosis posture is attainable, but we cannot recommend whether the posture is a realistic goal of intervention or useful training for neuromuscular control of the spine. Further research is needed to determine whether the sitting postures quantified in this study present advantages for safety or efficiency, to prevent and/or coordinate spinal movement. There are several important considerations for interpretation of the results of this study. First, this study examined only males. Replication with female subjects would be required to determine whether results are applicable to females. Second, the laboratory environment and adhesive tape used to attach electrodes to subjects’ spine, chest and pelvis may have influenced performance, although one would not expect this to compromise or enhance any specific postures. Third, although skin measures accurately represent changes in lumbar flexion/extension, skin surface measures may show smaller angles of lordosis than measures made with spinal imaging (Gracovetsky et al., 1990). Lastly, postures in this study focused only on sagittal spinal curves without back support. Other functional postures (including asymmetrical, backrest or armrest supported postures) remain to be investigated. 5. Conclusions Sitting in the short lordosis posture (thoraco-lumbar region kyphotic/flat, and lumbar region lordotic) required facilitation and feedback, but subjects imitated the slump, flat and long lordosis postures (consistent spinal curve directions at thoraco-lumbar and lumbar regions) without facilitation. Postures quantified in this study provide a foundation to examine whether particular spinal curves are advantageous in sitting. Acknowledgements (1) Paul Hodges is supported by the National Health and Medical Research Council of Australia. (2) G. Lorimer Moseley is supported by the Nuffield Medical Research Fellowship from the University of Oxford. (3) The Dorothy Hopkins award is acknowledged for financial assistance with reimbursing research subjects.
References Adams MA. Mechanical testing of the spine. An appraisal of methodology. Spine 1995;20(19):2151–6. Berthonnaud E, Dimnet J, Roussouly P, Labelle H. Analysis of the sagittal balance of the spine and pelvis using shape and orientation parameters. Journal of Spinal Disorders and Techniques 2005;18(1):40–7. Bogduk N. Clinical anatomy of the lumbar spine and sacrum. New York: Elsevier/ Churchill Livingstone; 2005. p. 53. Bullock MI, Bullock-Saxton JE. Control of low back in the workplace using an ergonomic approach. In: Twomey LT, Taylor JR, editors. Physical therapy of the low back. 3rd ed. New York: Churchill Livingstone; 2000. p. 297–326. Cholewicki J, McGill SM. Mechanical stability of the in vivo lumbar spine: implications for injury and chronic low back pain. Clinical Biomechanics (Bristol, Avon) 1996;11(1):1–15. Cholewicki J, Silfies SP, Shah RA, Greene HS, Reeves NP, Alvi K, Goldberg B. Delayed trunk muscle reflex responses increase the risk of low back injuries. Spine 2005;30(23):2614–20. Crisco JJ, Panjabi MM. Euler stability of the human ligamentous lumbar spine. Part II: experiment. Clinical Biomechanics (Bristol, Avon) 1992;7(1):27–32. Floyd WF, Silver PH. Function of erectores spinae in flexion of the trunk. Lancet 1951;1(3):133–4. Gracovetsky S, Kary M, Levy S, Ben Said R, Pitchen I, Helie J. Analysis of spinal and muscular activity during flexion/extension and free lifts. Spine 1990;15(12):1333–9. Hodges P, Cresswell A, Thorstensson A. Preparatory trunk muscle motion accompanies rapid upper limb movement. Experimental Brain Research 1999;124(1):69–79. Keegan JJ. Alterations of the lumbar curve related to posture and seating. The Journal of Bone and Joint Surgery (Am) 1953;35-A(3):589–603. Kendall FP. Posture. In: Muscles: testing and function with posture and pain. 5th ed. Baltimore: Lippincott Williams & Wilkins; 2005. p. 49–117. Kendall FP, McCreary EK, Kendall HO. Muscles, testing and function. Baltimore: Williams & Wilkins; 1983. p. 280. Kendall HO, Kendall FP, Boynton DA. Posture and pain. Baltimore: Williams & Wilkins; 1952. p. 5. Lee L. Ch 7: restoring force closure/motor control of the thorax. In: Lee D, editor. The thorax: an integrated approach. 2nd ed. Minneapolis: OPTP; 2003. p. 103–35. Magee DJ. Thoracic (dorsal) spine. In: Orthopedic physical assessment. 4th ed. Philadelphia: Saunders Elsevier; 2006. p. 425–65. Mandal AC. The correct height of school furniture. Physiotherapy 1984;70(2):48–53. Morl F, Blickhan R. Three-dimensional relation of skin markers to lumbar vertebrae of healthy subjects in different postures measured by open MRI. European Spine Journal 2006;15(6):742–51. O’Sullivan PB. ‘Clinical instability’ of the lumbar spine: its pathological basis, diagnosis and conservative management. In: Boyling JD, Jull GA, editors. Grieve’s modern manual therapy: the vertebral column. 3rd ed. Edinburgh: Churchill Livingstone; 2004. p. 311–31. O’Sullivan PB, Dankaerts W, Burnett A, Chen D, Booth R, Carlsen C, Schultz A. Evaluation of the flexion relaxation phenomenon of the trunk muscles in sitting. Spine 2006a;31(17):2009–16. O’Sullivan PB, Dankaerts W, Burnett AF, Farrell GT, Jefford E, Naylor CS, O’Sullivan KJ. Effect of different upright sitting postures on spinal–pelvic curvature and trunk muscle activation in a pain-free population. Spine 2006b;31(19):E707–12. Panjabi MM. The stabilizing system of the spine. Part I. Function, dysfunction, adaptation, and enhancement. Journal of Spinal Disorders 1992a;5(4):383–9. Panjabi MM. The stabilizing system of the spine. Part II. Neutral zone and instability hypothesis. Journal of Spinal Disorders 1992b;5(4):390–6. Roussouly P, Gollogly S, Berthonnaud E, Dimnet J. Classification of the normal variation in the sagittal alignment of the human lumbar spine and pelvis in the standing position. Spine 2005;30(3):346–53. Sahrmann S. Movement impairment syndromes of the lumbar spine. In: Diagnosis and treatment of movement impairment syndromes. St Louis: Mosby; 2002. p. 51–119. Scannell JP, McGill SM. Lumbar posture: should it and can it be modified? A study of passive tissue stiffness and lumbar position during activities of daily living. Physical Therapy 2003;83(10):907–17. Singer KP, Edmondston SJ, Day RE, Breidahl WH. Computer-assisted curvature assessment and Cobb angle determination of the thoracic kyphosis. An in vivo and in vitro comparison. Spine 1994;19(12):1381–4. Sprague RB. Differential assessment and mobilisation of the cervical and thoracic spine. In: Donatelli R, Wooden MJ, editors. Orthopaedic physical therapy. 3rd ed. New York: Churchill Livingstone; 2001. p. 108–43.
Manual Therapy 14 (2009) 409–414
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Original Article
Reliability and validity of a palpation technique for identifying the spinous processes of C7 and L5 Roar Robinson a, b, *, Hilde Stendal Robinson c, Gustav Bjørke b, Alice Kvale a a
University of Bergen, Department of Public Health and Primary Health Care, Section for Physiotherapy Science, Bergen, Norway Hans and Olaf Physiotherapy Clinic, Oslo, Norway c University of Oslo, Faculty of medicine, Institute for nursing and health sciences, Oslo, Norway b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 May 2007 Received in revised form 5 June 2008 Accepted 29 June 2008
The objective was to examine inter-tester reliability and validity of two therapists identifying the spinous processes (SP) of C7 and L5, using one predefined surface palpation procedure for each level. One identification method made it possible to examine the reliability and the validity of the procedure itself. Two manual therapists examined 49 patients (29 women). Aged between 26 and 79 years, 18 were cervical and 31 lumbar patients. An invisible marking pen and ultraviolet light were used, and the findings were compared. X-rays were taken as an objective measure of the correct spinal level. Percentage agreement and kappa statistics were used to evaluate reliability and validity. The best inter-therapist agreement was found for the skin marks. Percentage agreement within 10 mm and 20 mm was 67% and 85%, respectively. The inter-tester reliability for identifying a radiological nominated SP by palpation was found to be poor for C7 and moderate for L5, with kappa of 0.18 and 0.48, respectively. The results indicated acceptable inter-therapist surface palpation agreement, but the chosen procedures did not identify the correct SP. This indicates that the procedures are not precise enough. Future reliability studies should test other non-invasive palpation procedures, both individually and in combination, and compare these with radiological investigation. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Palpation procedure Reliability Validity Spinous process
1. Introduction The ability to palpate a spinous process (SP) is considered to be a basic skill and a prerequisite for other manual therapy techniques (Jull, 1986; Downey et al., 1999, 2003). If physiotherapists are unable to locate the same SP by palpation, it would be unreasonable to assume that other spinal manual therapy techniques have better reproducibility (Billis et al., 2003). For a variety of reasons, colleagues often examine the same patient, yet often their findings differ or are conflicting. Several studies have assessed the reliability of palpation tests for locating bony landmarks in the lumbar and sacral spine (Burton et al., 1990; Keating et al., 1990; Byfield et al., 1992; Simmonds and Kumar, 1993; McKenzie and Taylor, 1997; O’Haire and Gibbons, 2000; Harlick et al., 2007). Studies of static palpation of the L5 SP have shown acceptable intra-tester reliability, but generally poor inter-tester reliability (Burton et al.,1990; Breen,1992; Russell,1993;
* Corresponding author. Hans og Olaf fysioterapi A/S, Torggt 16, N – 0181 Oslo, Norway. Tel.: þ47 22993177; fax: þ47 22203019. E-mail address:
[email protected] (R. Robinson). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.06.002
Simmonds and Kumar, 1993). Some studies have demonstrated a trend for better agreement when using highly experienced physiotherapists specialised in manual therapy (Byfield et al., 1992; Jull et al., 1994; Downey et al., 1999; Billis et al., 2003). However, a systematic review of the reliability of spinal palpation for diagnosing back and neck pain concluded that neither the examiners’ education nor their experience improved reliability (Seffinger et al., 2004). In contrast to this, a recently published study by Harlick et al. (2007) concluded that inter-therapist variability had a greater effect on accuracy than any patient-defined factor. Due to common anatomical variations of the spine, advised palpation procedures may fail (Lewit, 1985; Grieve, 1994). The assumption that SPs are points rather than having a surface area (Burton et al., 1990; Simmonds and Kumar, 1993) might influence palpation results. McKenzie and Taylor (1997) attempted to correct this source of error and included an average surface area for each of the SPs of L1–L5. They examined 13 cadavers and found that the height of the SPs ranged from 16.4 to 20.4 mm and the width from 7.4 to 9.4 mm. Harlick et al. (2007) measured the height of the SPs on L1–L5 by means of X-ray and reported the mean height of L5 to be 14.1 mm. They suggested a mean SP height of 18.3 mm as level of acceptance for agreement.
410
R. Robinson et al. / Manual Therapy 14 (2009) 409–414
Binkley et al. (1995) had six orthopaedic physiotherapists to examine the mobility of the lumbar spine of 18 patients. The examination included identification of an arbitrary marked SP, and agreement on numbering this SP was found to be low. With no predefined method, they demonstrated inter-rater agreement only within 1.4 segments. Downey et al. (1999) reported a kappa of 0.69 and a weighted kappa of 0.92 for three pairs of therapists palpating the appropriate levels in the lumbar spine. If physiotherapists are unable to identify the same level, treatment and evaluation may be applied to the wrong level. Hence, there is no need to use weighted kappa. McKenzie and Taylor (1997) reported a lower overall kappa, even though they used the same palpation method as Downey et al. (1999). Christensen et al. (2002) studied the thoracic spine using the term expanded agreement; which also included the neighbouring segment. They argued that pain does not come from one segment, but from an area. However, we find this problematic, particularly when therapists use palpation to identify and treat symptomatic spinal levels, and also when evaluating patient progress. The literature on the topic is sparse, and with conflicting levels of intra- and inter-tester reliability it is difficult to argue for palpation as a reliable assessment tool. Najm et al. (2003) claimed that most research results are not comparable, due to variability in the tests, terminology, design, methodology, and statistical analyses utilised. Such inconsistencies make it difficult to rate the value of reliability studies and the results of effect studies. Furthermore, to our knowledge, no former studies have reported inter-tester reliability results for C7 identification. Although a wide variety of studies have related to radiological examination of the spine, very few have investigated either the reliability or the validity of locating SPs by palpation in relation to X-ray (Harlick et al., 2007). Hence, earlier studies might have shown satisfactory reliability, but the wrong spinal level may nevertheless have been palpated. Consequently; validity cannot be examined without confirmation. Postural assessment instruments and radiographic measurement are suggested as valid and reliable objective tools for identifying spinal levels (Haldeman et al., 1993; Troyanovich and Harrison, 1999). The SPs of C7 and L5 are easily identified on X-rays and are considered accurate if standard procedures are followed (A. Høiseth radiologist, personal communication 2004). Accordingly, we decided to use X-ray as the gold standard for identifying SPs, as used in a recently published study (Harlick et al., 2007). The main purpose of this study was to examine the inter-tester reliability of experienced manual therapists (MTs) using a predefined surface palpation technique to identify the SPs of C7 and L5, and to verify whether the markings were at the correct SP by means of X-ray (concurrent validity). We wanted to test the actual procedures described in textbooks and taught in physiotherapy schools. Hence, the focus was not to optimise the MTs’ possibilities of identifying the correct SP, but to evaluate the procedure itself. Allowing more than one procedure for each segment would have made this impossible. The amount of subcutaneous fat varies among patients and might influence palpation results (Harlick et al., 2007). Accordingly, we also decided to investigate whether patient’s body mass index (BMI) and gender influenced the findings. 2. Methods In clinical practice, the SPs of C7 and L5 are commonly used key points for identifying cervical and lumbar levels, respectively, before starting motion assessment (Magee, 2002). Guidelines for clinical identification of segmental levels are based on published descriptions (Hoppenfeld, 1976; Lewit, 1985; Grieve, 1994; Magee,
2002). In this study, C7 SP was identified through an assisted movement of the cervical spine into extension, where the C6 SP appears to move anterior (or ‘‘disappear’’) and C7 is thus the first cervical SP remaining stationary during the movement (Lewit, 1985; Magee, 2002). L5 was identified as the first SP under an imaginary line connecting the two iliac crests (Hoppenfeld, 1976; Lewit, 1985; Magee, 2002). The MTs used these procedures, among others, in their clinical practice. They attended three training sessions with the researcher to test the procedure details, including standard positioning of the participant, palpation, use of marking pen, blinding, and time consumption. Twelve healthy volunteers were tested, but no X-rays taken. The Regional Committee for Medical Research Ethics, Western Norway, provided formal ethical approval for this study. 2.1. Testers The participating MTs had between16 and 18 years in practice since completing the Norwegian postgraduate qualification in manual therapy (IFOMT standards). One radiologist with 30 years experience examined all the X-rays. 2.2. Subjects Patients referred to a particular radiology institute in Oslo, Norway, for X-ray examination of either their cervical or lumbar spine, were invited to participate in the study. Patients who had undergone surgery in the area of interest were not invited. If the MTs could neither identify the iliac crests properly nor palpate the SP in question, the patients were excluded. Cervical patients who could not maximally extend the lower cervical spine were also excluded. Information regarding diagnosis, age, gender, height and weight was collected. 2.3. Testing procedure Prior to radiographs, each participant was examined separately and in random order by both of the two MTs (GB and HSR). For practical reasons, examination was conducted in a different room to the X-ray table. Lumbar patients were examined in side lying (the same position in which the X-rays were subsequently taken). Cervical patients had their X-rays taken in the standing position, but the palpation procedures were performed with the participant seated. Reliability: Each MT palpated the SP in question and marked the position on the skin with a pen containing ink visible only under ultraviolet light, thus making it possible to blind the result for the next tester. The method has been utilised in previous studies (Burton et al., 1990; Simmonds and Kumar, 1993; McKenzie and Taylor, 1997; Downey et al., 1999; Billis et al., 2003). The researcher (RR) measured the distance between the marks using a hand held calliper and an UV lamp. Validity: To visualize the marks on X-rays, one small magnet was taped on each skin mark (Accu-Band, Magnetic Plaster, Gauss, Ito co., Ltd). The X-rays were taken and examined according to standard procedures with one anterior–posterior picture and one lateral projection; no extra X-rays were taken for the study. The outlines of the vertebrae were inspected on the X-rays, and lines were drawn perpendicular to the skin when defining the sector for C7 (L5). The area between two lines defined the sector (Figs. 1 and 2); line one midway from the lower demarcation of the C6 (L4) and the upper demarcation of C7 (L5) to the skin and line two from the upper demarcation of Th1 (S1) and the lower demarcation of C7 (L5) to the skin. The markers (magnets) were inspected and noted as being within or outside the sector. The
R. Robinson et al. / Manual Therapy 14 (2009) 409–414
411
results were reported as positive or negative identification of the actual SP marked by each of the therapists. 2.4. Statistical analysis SPSS version 12.0 was used. The percentage agreement was calculated as a measure of reliability between the two therapists, firstly based on skin markings alone and secondly on X-ray identification of the magnets and the SPs. To evaluate palpation precision, we defined and used three levels of agreement: marking at the same point (0 mm), within 10 mm and within 20 mm, respectively. These distances are within formerly reported sizes of the SPs (McKenzie and Taylor, 1997; Harlick et al., 2007). Kappa statistics were not used, since no contingency (2 2) table could be made with this data format. However, when the palpation marks were compared with the gold standard (X-ray), Cohen’s kappa (k) was used to calculate agreement between the two testers, the marks being either within or outside the sector. k effectively discounts the proportion of agreement expected by chance, and ranges in values from 1 to þ1 (Landis and Koch, 1977). Both inter-tester reliability and concurrent validity could be examined using these calculations. We also split the participants into three groups according to their BMI and used the Kruskal–Wallis test to compare the palpation results for the groups to see if BMI influenced the results. 3. Results Fig. 1. The identification sector of C7 on the X-ray was defined as: the area between one line drawn from a midpoint between the upper demarcation of C7 SP and the lower demarcation of C6 SP, and another line drawn from the midpoint between the lower demarcation of C7 SP and upper demarcation of Th1 SP, the lines being perpendicular to the skin.
The 52 patients (aged 26–79 years) who were asked to participate included 20 cervical and 32 lumbar patients. All participants were examined twice prior to radiographs, one patient was excluded because one magnet had fallen off and two patients did not meet the inclusion criteria. The study sample thus contained 49 patients; 18 cervical patients (8 females), and 31 lumbar patients (21 females) (Table 1). 3.1. Reliability: therapist agreement on skin marks The therapists had marked the same point in 18 out of 49 patients (37%). The agreement increased to 33 (67%) and 40 (82%) of 49 patients for markings within 10 and 20 mm, respectively. When the results for C7 and L5 were analysed separately, we found corresponding results (Tables 2 and 4). 3.2. Validity: therapist agreement examined on X-rays The magnets’ positions were examined using two X-ray projections; the anterior–posterior projection was used to check that both magnets were in place, and the lateral projection to evaluate the position of the magnets in relation to the defined sector. The two MTs identified the correct SP in a total of 25 (51%) and 24 (49%) of 49 participants, respectively. C7 was correctly identified in 10 (55%) and 13 (72%) participants, respectively. There was agreement between the two MTs in 8 (44%) out of 18 cervical participants. For 3 (17%) of the cervical participants, both MTs
Table 1 Demographic data. Demographic data
Fig. 2. The identification sector of L5 on the X-ray was defined as: the area between one line drawn from a midpoint between the upper demarcation of L5 SP and the lower demarcation of L4 SP, and another line drawn from the midpoint between the lower demarcation of L5 SP and the upper demarcation of S1 SP, both lines being perpendicular to the skin.
Age, range (mean) Women Normal weight (BMI 18.5–24.9) Overweight (BMI 25–29.9) Obese (BMI >30)
Cervical
Lumbar
n ¼ 18
n ¼ 31
Total n ¼ 49
26–79 (53.6) 8 7 7 4
26–79 (49) 21 14 12 5
26–79 (50.7) 29 21 19 9
412
R. Robinson et al. / Manual Therapy 14 (2009) 409–414
Table 2 Agreement between the two therapists, according to skin marks only.
C7 (n ¼ 18) L5 (n ¼ 31) Total (n ¼ 49)
Total agreement (%)
Agreement within:
0 mm
10 mm
20 mm
7 (39) 11 (35) 18 (37)
11 (61) 22 (71) 33 (67)
14 (78) 27 (87) 40 (82)
The results are presented within three levels of agreement, 0 mm, 10 mm, and 20 mm, to permit determination of the degree of palpation precision.
marked outside the sector (Table 3). Kappa was 0.18 (95% CI; 0.25 to 0.62) for identifying C7 SP. L5 SP was correctly identified by the two MTs for 15 (48%) and 11 (36%) participants, respectively. There was agreement between the MTs in 9 (29%) out of 31 participants. For 14 (45%) participants, both MTs had marked outside the sector (Table 3). Kappa was 0.48 (95% CI: 0.18–0.78) for identifying L5 SP. For 15 participants, one MT marked within and the other outside the sector (Table 3). For 7 of these 15, the distance between the marks was 0.05) at all time intervals following intervention. The group that received a lumbopelvic joint manipulation demonstrated a significant increase in quadriceps force (3%) and activation (5%) (p < 0.05) immediately following intervention, but this effect was not present after the 20 min interval. Since participants in this study were free of knee joint pathology, it is possible that they did not have the capacity to allow for large changes in quadriceps muscle activation to occur. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Force output Manual therapy Muscle activation Sacroiliac
1. Introduction Manual therapeutic interventions such as joint mobilisation or manipulation have been shown to alter muscle force output and activation. Specific to the lower extremity, changes in muscle force output and activation have been demonstrated in the hip extensors (Yerys et al., 2002), hamstrings (Cibulka et al., 1986), quadriceps (Suter et al., 1999, 2000; Hillermann et al., 2006), soleus (Murphy et al., 1995), and gastrocnemius (Dishman and Bulbulian, 2000, 2001; Dishman et al., 2002b, 2005; Dishman and Burke, 2003). Joint mobilisation or manipulation is thought to stimulate sensory receptors in and around the joint and affects the central nervous system at the spinal segmental level (Suter et al., 1994; Murphy et al., 1995; Herzog et al., 1999; Pickar, 2002; Colloca et al., 2003, 2004; Sung et al., 2004) as well as the cortical level (Dishman et al., 2002a). The associated neurophysiological effect may be dependent on the forces (high vs. low grade joint mobilisations) applied during the manual intervention (Dishman et al., 2002a, 2005). Changes in muscle activation have been demonstrated in symptomatic (Suter et al., 1999, 2000) and healthy individuals * Corresponding author. University of Virginia, 290 Massie Road, McCue Center, PO Box 400834, Charlottesville, VA 22903, USA. Tel.: þ1 434 823 5031/þ1 434 243 2419; fax: þ1 434 243 2430. E-mail address:
[email protected] (T.L. Grindstaff). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.06.005
(Murphy et al., 1995; Dishman and Bulbulian, 2000, 2001; Dishman et al., 2002b; Dishman and Burke, 2003). A single lumbopelvic joint manipulation has been shown to acutely increase quadriceps force output (Suter et al., 1999, 2000; Hillermann et al., 2006) and quadriceps activation (Suter et al., 1999, 2000) in individuals with anterior knee pain. Unfortunately these studies are limited by not examining the underlying physiological mechanisms for changes in strength and function (Hillermann et al., 2006; Iverson et al., 2008) or the duration of effects (Suter et al., 1999, 2000). It is also unknown whether lower grade joint mobilisations would have a similar effect. Studying the effects of joint mobilisation and manipulation on asymptomatic individuals may provide additional insight into the neurophysiological muscle response of the intervention without the confounding, uncontrolled effects of altered muscle activation related to injury. Further understanding the neurophysiological response and duration of altered muscle activation of the quadriceps following manual intervention will help guide future studies and begin to provide scientific rational for treatment selections. Since previous studies (Suter et al., 1999, 2000) have only examined immediate changes in quadriceps force output and activation further investigation is necessary to determine if changes would be maintained over a 60 min period of time. Therefore, the purpose of this study was to determine the amount and duration of altered quadriceps force output and activation following a single high or
416
T.L. Grindstaff et al. / Manual Therapy 14 (2009) 415–420
low grade joint mobilisation/manipulation applied at the lumbopelvic region in healthy individuals over the course of 1 h. 2. Methods 2.1. Design A randomised controlled trial with one between factor, treatment group (lumbopelvic joint manipulation, passive lumbar range of motion, and prone extension) and one within factor, time (pre/ post 0, 20, 40, 60 min) was used to examine the effects of high and low grade joint mobilisation/manipulation on quadriceps force output and activation in this clinic-based study. Main outcome variables included quadriceps force output and percentage quadriceps activation. 2.2. Participants Forty-two healthy subjects volunteered for this study (Table 1). Subjects self-reported they had pain free lumbar spine and lower extremities for the past six months. Exclusion criteria included signs/symptoms indicating nerve root compression, previous spine or lower extremity surgery, osteoporosis, pregnancy, and spinal or neurological disorders. A brief health history form was completed by each subject and a standard musculoskeletal evaluation was performed and included assessment of the lumbar spine, sacroiliac, and knee joints to screen for exclusionary criteria. All subjects signed a consent form prior to participation and the study was approved by our Institutional Review Board. 2.3. Instrumentation 2.3.1. Quadriceps force output Isometric quadriceps force was measured using a load cell (Model 41, Range 1–1000 lbs; Sensotec, Columbus, OH) interfaced with a data acquisition system (MP150; Biopac Systems, Inc., Goleta, CA) and amplifier (DA100B; Biopac Systems Inc.), and sampled at 125 Hz. Subjects were seated in a custom-made chair with their hips flexed at 85 , knees flexed at 90 , and arms folded across their chest The pelvis was secured to the chair using Velcro straps, while a padded ankle strap was placed 3 cm proximal to the lateral malleolus and connected to the load cell via an ‘‘S’’ hook. 2.3.2. Burst-superimposition technique Quadriceps activation was estimated by utilising the burstsuperimposition technique on a maximum voluntary isometric contraction (MVIC). The burst-superimposition technique provides the muscle with a percutaneous supramaximal stimulus to recruit any remaining muscle fibres which have not been stimulated (Rutherford et al., 1986; Snyder-Mackler et al., 1994; Stevens et al., 2001). A superimposed burst (100 pulses/s, 600 ms pulse duration, 10 pulse tetanic train, 125 V, 100 ms duration) was manually applied to the quadriceps approximately 2 s after the beginning of the MVIC when the experimenter determined a plateau in force had occurred. Amount of muscle activation was quantified using the
Table 1 Subject demographics.
Age Height (cm) Mass (kg) Force (N) CAR (%)
Manipulation (n ¼ 15)
PROM (n ¼ 13)
Prone extension (n ¼ 13)
24.6 168.4 69.1 495.1 83.4
28.6 (8.2) 170.2 (7.0) 68.7 (8.1) 431.5 (105.7) 76.2 (12.3)
27.0 168.0 70.7 450.9 75.6
(6.2) (8.4) (16.1) (122.1) (9.9)
Values are mean (SD).
(5.9) (10.4) (14.9) (113.9) (11.9)
central activation ratio (CAR) and calculated by dividing the volitional MVIC force by total force (combined effect of the electrical superimposed burst stimulation upon the MVIC, Eq. (1)) (KentBraun and Le Blanc, 1996). A CAR of 1.00 represents complete quadriceps activation (Stackhouse et al., 2000, 2001; Mizner et al., 2003; Stevens et al., 2003; Fitzgerald et al., 2004; Lewek et al., 2004).
CAR ¼
Fvolitional Fvolitionalþelectrical
(1)
An S88 Grass Stimulator (Astro-Med, West Warwick, RI) was used with the SIU8T isolation unit (125 V stimulus) and two rubber–carbon electrodes (8 14 cm) to deliver the electrical stimuli over quadriceps. Electrode surfaces were covered with conductive gel and secured with an elastic bandage over the proximal lateral aspect and the distal medial aspect of the quadriceps muscle. The burst-superimposition technique has been shown to be highly reliable with repeated testing of healthy subjects (ICC ¼ 0.98) (Snyder-Mackler et al., 1993). 2.4. Study protocol After initial evaluation, all participants had baseline testing of quadriceps strength and quadriceps activation. Test leg was randomly determined by coin toss and all interventions and tests were performed on the same side. Participants performed a standardised warm-up consisting of four submaximal isometric contractions (50–75% MVIC) with submaximal electrical stimulation of the quadriceps and one MVIC with submaximal electrical stimulation to orient them to the test procedures. Participants were instructed to slowly build up force and hold an MVIC for 3–5 s. Verbal encouragement and visual feedback of real time force output were given. A superimposed burst (100 pulses/s, 600 ms pulse duration, 10 pulse tetanic train, 125 V, 100 ms duration) was manually applied to the quadriceps approximately 2 s after the beginning of the MVIC when the experimenter determined a plateau in force had occurred. If force did not plateau, a stimulus was not applied. A 90 s rest period was given between trials. Participants performed three trials with superimposed burst, with the average MVIC and CAR values used for data analysis. Following baseline testing, participants were randomised to one of three treatment interventions: lumbopelvic joint manipulation, side-lying lumbar mid-range flexion/extension PROM for 1 min, or lying prone (Prone Ext) on elbows for 3 min. The total duration to perform each of the three interventions was estimated at 3 min and accounted for subject positioning and intervention. Lumbopelvic joint manipulation was selected as a high grade mobilisation, while lumbar PROM was selected as a lower grade joint mobilisation. The prone on elbows intervention was selected as a sham treatment to reduce potential participant bias. Quadriceps strength and quadriceps activation were tested immediately following intervention (post 0), and at 20, 40, and 60-min post-intervention time intervals, using the same methods described above. Sixty minutes was chosen to coincide with a 60 min rehabilitation session based on common clinical practice. During rest periods between testing intervals, participants were asked to remain seated. Testing concluded after 60-min post-intervention data were collected. 2.4.1. Lumbopelvic joint manipulation The lumbopelvic joint manipulation (Flynn et al., 2006) was performed on the ipsilateral side of the test limb (Fig. 1). The term lumbopelvic was used to describe the targeted region since this manipulation technique is not exclusively specific to the lumbar, sacroiliac, or pelvic regions (Flynn et al., 2006). The manipulation procedure utilised in this study was consistent with previously
T.L. Grindstaff et al. / Manual Therapy 14 (2009) 415–420
417
Fig. 2. Passive range of motion.
2.5. Statistical analyses
Fig. 1. Lumbopelvic joint manipulation in supine with side bending (a) and rotation (b).
used methods (Flynn et al., 2002; Fritz et al., 2004; Iverson et al., 2008) and was performed by one of two physical therapists (initials, initials) with advanced manual therapy training. Subjects were positioned supine on a treatment table, while the experimenter stood on the opposite side to be manipulated. The participant was passively side-bent towards and rotated away from the selected lumbopelvic joint which was followed by the delivery of a posterior/inferior force through the opposite anterior superior iliac spine (ASIS). If a cavitation was not heard or felt by the subject or examiner, the technique was repeated. If the second attempt was not successful the procedure was repeated on the contralateral side using similar methods (Flynn et al., 2002; Fritz et al., 2004; Iverson et al., 2008). If cavitation was not heard or felt by the participant or examiner following the fourth attempt, the participant proceeded with the assessment of quadriceps activation as usual.
Subject demographics and baseline values for MVIC and CAR were compared using a one-way ANOVA. Two separate single factor repeated measures ANOVAs were performed to compare MVIC and CAR percent change scores from baseline between groups (manipulation, PROM, and Prone Ext) across each time period. A secondary analysis consisting of two post-hoc one-way ANOVAs was performed to analyse the percent change from baseline for quadriceps MVIC force and CAR values immediately following intervention. This analysis was performed to assess immediate effects of the intervention, allowing direct comparisons to similar studies. The level of statistical significance was set a priori at p < 0.05. Statistical analyses were performed with SPSS Version 14.0 (SPSS Inc., Chicago, IL).
3. Results There were no significant differences (p > 0.05) between any of the subject group demographics or baseline MVIC or CAR values (Table 1). Lumbopelvic joint cavitation was achieved in 86.7% of the individuals (54% with one attempt, 46% requiring 2–3 attempts). Only two of the subjects in the manipulation group were unable to achieve joint cavitation after four attempts (two per side), but were retained in the statistical analysis, since cavitation may not be
2.4.2. PROM Subjects were positioned side-lying on the opposite side of the test limb (Fig. 2). The experimenter held both knees with one arm while placing their opposite hand on the participant’s lumbar spine. The experimenter performed 1 min of flexion and extension PROM without reaching physiological end range in either direction of movement. 2.4.3. Prone extension on elbows Subjects were positioned prone with lumbar spine extension (Fig. 3) while using their elbows for support to maintain the position for 3 min.
Fig. 3. Prone extension on elbows.
T.L. Grindstaff et al. / Manual Therapy 14 (2009) 415–420
necessary to achieve clinically relevant changes (Flynn et al., 2003, 2006). For quadriceps MVIC force (Fig. 4) there was not a significant time by group interaction (F8,152 ¼ 1.41, p ¼ 0.20) or a significant difference between groups (F2,38 ¼ 2.55, p ¼ 0.09). When all subjects were examined independent of group assignment there was a significant difference across time intervals (F4,152 ¼ 18.45, p < 0.001) with a decrease (p < 0.01) in MVIC force from baseline at all time (0, 20, 40, 60 min) intervals following intervention. For quadriceps CAR (Fig. 5) there was not a significant time by group interaction (F8,152 ¼ 1.02, p ¼ 0.43) or significant differences in time (F4,152 ¼ 1.08, p ¼ 0.37) or between groups (F2,38 ¼ 1.12, p ¼ 0.34). To allow for comparison with previous studies (Suter et al., 1999, 2000; Hillermann et al., 2006) two post-hoc one-way ANOVAs were performed to analyse the percent change from baseline for quadriceps MVIC force and CAR values immediately following intervention. There was a significant difference between groups for MVIC (F2,38 ¼ 6.93, p ¼ 0.003) and CAR (F2,38 ¼ 3.98, p ¼ 0.03). The manipulation group demonstrated a significant increase in quadriceps force output (3.1%) compared to the PROM group (p ¼ 0.001, 95% CI ¼ 5.52, 19.44) and the Prone Ext group (p ¼ 0.02, 95% CI ¼ 1.36, 15.28). The manipulation group also significantly increased quadriceps activation (4.7%) compared to the PROM group (p ¼ 0.04, 95% CI ¼ 0.37, 9.92) and Prone Ext group (p ¼ 0.01, 95% CI ¼ 1.34, 10.90). Effect size for immediate changes in MVIC force (d ¼ 0.12) and CAR (d ¼ 0.38) following joint manipulation was also calculated.
4. Discussion The results of this study indicate that changes in quadriceps force output and activation are not present over the course of 1 h following high or low grade joint mobilisation/manipulation directed at the lumbopelvic region. The original data analysis suggested an immediate change in quadriceps force output and activation was not present following lumbopelvic joint manipulation and contrasted findings of similar studies (Suter et al., 1999, 2000; Hillermann et al., 2006). The secondary analysis was conducted to only examine immediate changes following intervention and allowed for direct comparisons with previous findings. This analysis indicated following lumbopelvic joint manipulation an acute increase in quadriceps force output (3.1%) and quadriceps activation (4.7%) was present. Although changes in quadriceps force output were less than previously reported values (11–17%) (Suter et al., 1999, 2000; Hillermann et al., 2006), changes in quadriceps activation (5–7.5%) were consistent with previously reported values (Suter et al., 1999, 2000). It appears with the
10
Percent Change
Manipulation PROM Prone Ext
*
5 0 -5 -10 -15 -20 -25 Baseline
Post 0
Post 20
Post 40
Post 60
Fig. 4. Quadriceps MVIC force. Values are expressed as percent change from baseline and standard error of the mean. *Significant from baseline p 0.05.
10 Manipulation PROM Prone Ext
* Percent Change
418
0
-5
-10 Baseline
Post 0
Post 20
Post 40
Post 60
Fig. 5. Quadriceps activation CAR. Values are expressed as percent change from baseline and standard error of the mean. *Significant from baseline p 0.05.
multifactorial design originally used in this study that an immediate increase in quadriceps force output and activation may have gone unrecognised. Thus we cannot discount the immediate findings which are in agreement with previous studies. Examination of effect sizes for the manipulation group demonstrates a small effect size (d ¼ 0.12) for immediate changes in quadriceps force output and a small, but approaching moderate (d ¼ 0.38) effect size for quadriceps activation. Due to the extremely short term effect following lumbopelvic joint manipulation the clinical relevance of these findings is questionable and interpretation should be left to the reader. The immediate increase in quadriceps force output and activation following lumbopelvic joint manipulation could be attributed to a facilitation of the motoneuron pool mediated at the spinal or cortical level. We hypothesise the underlying physiological mechanism for the distant response associated with lumbopelvic joint manipulation may be due to common sensory and motor nerve root levels with the same interneurons. Joint manipulation is thought to affect the central nervous system at the segmental level by activating structures in and around the manipulated joint (mechanoreceptors, proprioceptors, and free nerve endings) (Suter et al., 1994; Murphy et al., 1995; Herzog et al., 1999; Pickar, 2002; Colloca et al., 2003, 2004; Sung et al., 2004). Cortical changes have also been demonstrated following spinal manipulation (Dishman et al., 2002a) and may also subsequently affect motoneuron pool excitability. Since the sacroiliac joint (L2–S3), quadriceps (L2–4) and knee joints (L2–S2) share common nerve root levels (Moore and Dalley, 1999) it is possible that afferent information from one structure may alter efferent signals to all structures innervated by a similar nerve root level. Lumbopelvic joint manipulation has been shown to briefly decrease H-reflexes of the soleus (Murphy et al., 1995) and gastrocnemius (Dishman and Bulbulian, 2000, 2001; Dishman et al., 2002b, 2005; Dishman and Burke, 2003) muscles and increase quadriceps force output (Suter et al., 1999, 2000; Hillermann et al., 2006) and activation (Suter et al., 1999, 2000). This relationship has been demonstrated in reverse using a knee joint effusion model, where quadriceps inhibition and soleus facilitation were demonstrated using H-reflex measures (Hopkins et al., 2001). It is proposed that the increase in afferent information due to joint manipulation is mediated at the interneuron and can affect efferent motor output to the surrounding musculature. The clinical implication of this study is in agreement that lumbopelvic joint manipulation has the ability to immediately increase quadriceps force output and activation, but the effects in a healthy population
T.L. Grindstaff et al. / Manual Therapy 14 (2009) 415–420
are of limited duration and small, but approaching a moderate effect size (d ¼ 0.38) for changes in quadriceps activation. A limitation of this study is only healthy individuals without joint pathology were studied. It is possible since participants were free of joint pathology that their motoneuron pool availability could not be altered beyond the acute response observed in this study. An explanation for the differences in magnitude of quadriceps activation between this study (4.7%) and previous studies (Suter et al., 1999, 2000) demonstrating that a 5–7.5% increase in quadriceps activation occurs may be due to the fact that facilitation of the quadriceps may not occur as easily as disinhibition. Individuals with knee joint pathology typically have some level of quadriceps inhibition (Stratford, 1982; SnyderMackler et al., 1994; Maitland et al., 1999; Suter et al., 1999, 2000; Urbach et al., 1999, 2001; Hopkins et al., 2001; Palmieri et al., 2003, 2004; Williams et al., 2003; Chmielewski et al., 2004) which may allow for greater changes in quadriceps activation to occur following intervention. Participants in this study did have quadriceps activation levels which were below values previously published literature on healthy individuals (Stackhouse et al., 2001; Lewek et al., 2004; Hart et al., 2006a,b). CAR values greater than or equal to 0.95 have been used to describe a situation where all motoneurons have been activated during an MVIC (Stackhouse et al., 2001; Lewek et al., 2004). Approximately 80% of the individuals in this study could not achieve quadriceps activation levels above 90%. Previous studies (Stackhouse et al., 2001; Lewek et al., 2004) have estimated that 25% of younger to middle aged adults are not able to achieve full activation. Procedures used in this study including warm-up, verbal encouragement, visual feedback, and rest periods between trials were similar to other studies (Manal and Snyder-Mackler, 2000; Stackhouse et al., 2001; Lewek et al., 2002, 2004; Mizner et al., 2003, 2005; Stevens et al., 2003; Chmielewski et al., 2004; Williams et al., 2005; Hart et al., 2006a,b,c). Although the CAR values in this study were lower, measurements in the sham group were extremely stable (ICC3,k ¼ 0.97, 95% CI: 0.93–0.99; SEM ¼ 2.06%) over the course of an hour indicating stability in our measurement technique. Even if the sham intervention had a therapeutic effect, at best, this value would be underestimated and likely represents normal variability within each subject over the 60 min testing period. Results also indicated decreased quadriceps force output occurred over the course of 1 h for all groups. Obtaining valid and reliable CAR values is dependent on participants giving a maximal effort (Behm et al., 1996). Subjects were verbally encouraged during all trials to give 100% effort, but deceased force output occurred over the course of 1 h. This may have been due local muscle fatigue due to performing 15 MVICs augmented with a supramaximal electrical stimulus. The decrease in quadriceps force output could not be attributed to changes in CAR, but were more likely due to local muscle fatigue. Future studies should utilise electromyographic analysis of muscle median frequency to quantify local muscle fatigue associated with repeated MVICs (Hart et al., 2006a,b).
5. Conclusion The findings of this study indicate that there was an acute increase in quadriceps force output and quadriceps activation following lumbopelvic joint manipulation in healthy individuals, but these effects diminished within 20 min of intervention. All groups demonstrated a progressive loss of quadriceps force output over the course of testing, which was thought to be due to muscle fatigue. Additional research is necessary to accurately determine duration of changes in quadriceps force output and activation
419
following lumbopelvic joint manipulation in healthy and symptomatic individuals. References Behm DG, St-Pierre DM, Perez D. Muscle inactivation: assessment of interpolated twitch technique. Journal of Applied Physiology 1996;81:2267–73. Chmielewski TL, Stackhouse S, Axe MJ, Snyder-Mackler L. A prospective analysis of incidence and severity of quadriceps inhibition in a consecutive sample of 100 patients with complete acute anterior cruciate ligament rupture. Journal of Orthopaedic Research 2004;22:925–30. Cibulka MT, Rose SJ, Delitto A, Sinacore DR. Hamstring muscle strain treated by mobilizing the sacroiliac joint. Physical Therapy 1986;66:1220–3. Colloca CJ, Keller TS, Gunzburg R. Neuromechanical characterization of in vivo lumbar spinal manipulation. Part II. Neurophysiological response. Journal of Manipulative and Physiological Therapeutics 2003;26:579–91. Colloca CJ, Keller TS, Gunzburg R. Biomechanical and neurophysiological responses to spinal manipulation in patients with lumbar radiculopathy. Journal of Manipulative and Physiological Therapeutics 2004;27:1–15. Dishman JD, Ball KA, Burke J. Central motor excitability changes after spinal manipulation: a transcranial magnetic stimulation study. Journal of Manipulative and Physiological Therapeutics 2002a;25:1–9. Dishman JD, Bulbulian R. Spinal reflex attenuation associated with spinal manipulation. Spine 2000;25:2519–24. Dishman JD, Bulbulian R. Comparison of effects of spinal manipulation and massage on motoneuron excitability. Electromyography and Clinical Neurophysiology 2001;41:97–106. Dishman JD, Burke J. Spinal reflex excitability changes after cervical and lumbar spinal manipulation: a comparative study. The Spine Journal 2003;3:204–12. Dishman JD, Cunningham BM, Burke J. Comparison of tibial nerve H-reflex excitability after cervical and lumbar spine manipulation. Journal of Manipulative and Physiological Therapeutics 2002b;25:318–25. Dishman JD, Dougherty PE, Burke JR. Evaluation of the effect of postural perturbation on motoneuronal activity following various methods of lumbar spinal manipulation. The Spine Journal 2005;5:650–9. Fitzgerald GK, Piva SR, Irrgang JJ, Bouzubar F, Starz TW. Quadriceps activation failure as a moderator of the relationship between quadriceps strength and physical function in individuals with knee osteoarthritis. Arthritis & Rheumatism 2004;51:40–8. Flynn T, Fritz J, Whitman J, Wainner R, Magel J, Rendeiro D, et al. A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improvement with spinal manipulation. Spine 2002;27:2835–43. Flynn TW, Childs JD, Fritz JM. The audible pop from high-velocity thrust manipulation and outcome in individuals with low back pain. Journal of Manipulative and Physiological Therapeutics 2006;29:40–5. Flynn TW, Fritz JM, Wainner RS, Whitman JM. The audible pop is not necessary for successful spinal high-velocity thrust manipulation in individuals with low back pain. Archives of Physical Medicine and Rehabilitation 2003;84:1057–60. Fritz JM, Whitman JM, Flynn TW, Wainner RS, Childs JD. Factors related to the inability of individuals with low back pain to improve with a spinal manipulation. Physical Therapy 2004;84:173–90. Hart JM, Fritz JM, Kerrigan DC, Saliba EN, Gansneder BM, Ingersoll CD. Quadriceps inhibition after repetitive lumbar extension exercise in persons with a history of low back pain. Journal of Athletic Training 2006a;41:264–9. Hart JM, Fritz JM, Kerrigan DC, Saliba EN, Gansneder BM, Ingersoll CD. Reduced quadriceps activation after lumbar paraspinal fatiguing exercise. Journal of Athletic Training 2006b;41:74–8. Hart JM, Kerrigan DC, Fritz JM, Saliba EN, Gansneder B, Ingersoll CD. Contribution of hamstring fatigue to quadriceps inhibition following lumbar extension exercise. Journal of Sports Science & Medicine 2006c;5:70–9. Herzog W, Scheele D, Conway J. Electromyographic responses of back and limb muscles associated with spinal manipulative therapy. Spine 1999;24:146–53. Hillermann B, Gomes AN, Korporaal C, Jackson D. A pilot study comparing the effects of spinal manipulative therapy with those of extra-spinal manipulative therapy on quadriceps muscle strength. Journal of Manipulative and Physiological Therapeutics 2006;29:145–9. Hopkins JT, Ingersoll CD, Krause BA, Edwards JE, Cordova ML. Effect of knee joint effusion on quadriceps and soleus motoneuron pool excitability. Medicine & Science in Sports & Exercise 2001;33:123–6. Iverson CA, Sutlive TG, Crowell MS, Morrell RL, Perkins MW, Garber MB, et al. Lumbopelvic manipulation for the treatment of patients with patellofemoral pain syndrome: Development of a clinical prediction rule. Journal of Orthopaedic and Sports Physical Therapy 2008;33(6):297–309. Kent-Braun JA, Le Blanc R. Quantitation of central activation failure during maximal voluntary contractions in humans. Muscle & Nerve 1996;19:861–9. Lewek M, Rudolph KS, Axe MJ, Snyder-Mackler L. The effect of insufficient quad strength on gait after anterior cruciate ligament reconstruction. Clinical Biomechanics 2002;17:56–63. Lewek MD, Rudolph KS, Snyder-Mackler L. Quadriceps femoris muscle weakness and activation failure in patients with symptomatic knee osteoarthritis. Journal of Orthopaedic Research 2004;22:110–5.
420
T.L. Grindstaff et al. / Manual Therapy 14 (2009) 415–420
Maitland ME, Ajemian SV, Suter E. Quadriceps femoris and hamstring muscle function in a person with an unstable knee. Physical Therapy 1999;79:66–75. Manal TJ, Snyder-Mackler L. Failure of voluntary activation of the quadriceps femoris muscle after patellar contusion. Journal of Orthopaedic and Sports Physical Therapy 2000;30:655–60. Mizner RL, Petterson SC, Stevens JE, Vandenborne K, Snyder-Mackler L. Early quadriceps strength loss after total knee arthroplasty. The Contributions of muscle atrophy and failure of voluntary muscle activation. Journal of Bone and Joint Surgery American 2005;87:1047–53. Mizner RL, Stevens JE, Snyder-Mackler L. Voluntary activation and decreased force production of the quadriceps femoris muscle after total knee arthroplasty. Physical Therapy 2003;83:359–65. Moore KL, Dalley AF. Clinically oriented anatomy. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 1999. Murphy BA, Dawson NJ, Slack JR. Sacroiliac joint manipulation decreases the H-reflex. Electromyography and Clinical Neurophysiology 1995;35:87–94. Palmieri RM, Ingersoll CD, Edwards JE, Hoffman MA, Stone MB, Babington JP, et al. Arthrogenic muscle inhibition is not present in the limb contralateral to a simulated knee joint effusion. American Journal of Physical Medicine & Rehabilitation 2003;82:910–6. Palmieri RM, Tom JA, Edwards JE, Weltman A, Saliba EN, Mistry DJ, et al. Arthrogenic muscle response induced by an experimental knee joint effusion is mediated by pre- and post-synaptic spinal mechanisms. Journal of Electromyography and Kinesiology 2004;14:631–40. Pickar JG. Neurophysiological effects of spinal manipulation. The Spine Journal 2002;2:357–71. Rutherford OM, Jones DA, Newham DJ. Clinical and experimental application of the percutaneous twitch superimposition technique for the study of human muscle activation. Journal of Neurology. Neurosurgery & Psychiatry 1986;49:1288–91. Snyder-Mackler L, Binder-Macleod SA, Williams PR. Fatigability of human quadriceps femoris muscle following anterior cruciate ligament reconstruction. Medicine & Science in Sports & Exercise 1993;25:783–9. Snyder-Mackler L, De Luca PF, Williams PR, Eastlack ME, Bartolozzi ARI. Reflex inhibition of the quadriceps femoris muscle after injury or reconstruction of the anterior cruciate ligament. Journal of Bone and Joint Surgery American 1994;76:555–60. Stackhouse SK, Dean JC, Lee SCK, Binder-MacLeod SA. Measurement of central activation failure of the quadriceps femoris in healthy adults. Muscle & Nerve 2000;23:1706–12.
Stackhouse SK, Stevens JE, Lee SC, Pearce KM, Snyder-Mackler L, Binder-Macleod SA. Maximum voluntary activation in nonfatigued and fatigued muscle of young and elderly individuals. Physical Therapy 2001;81:1102–9. Stevens JE, Binder-Macleod S, Snyder-Mackler L. Characterization of the human quadriceps muscle in active elders. Archives of Physical Medicine and Rehabilitation 2001;82:973–8. Stevens JE, Mizner RL, Snyder-Mackler L. Quadriceps strength and volitional activation before and after total knee arthroplasty for osteoarthritis. Journal of Orthopaedic Research 2003;21:775–9. Stratford P. Electromyography of the quadriceps femoris muscles in subjects with normal knees and acutely effused knees. Physical Therapy 1982;62:279–83. Sung PS, Kang YM, Pickar JG. Effect of spinal manipulation duration on low threshold mechanoreceptors in lumbar paraspinal muscles: a preliminary report. Spine 2004;30:115–22. Suter E, Herzog W, Conway PJ, Zhang YT. Reflex response associated with manipulative treatment of the thoracic spine. Journal of the Neuromusculoskeletal System 1994;2:124–30. Suter E, McMorland G, Herzog W, Bray R. Decrease in quadriceps inhibition after sacroiliac joint manipulation in patients with anterior knee pain. Journal of Manipulative and Physiological Therapeutics 1999;22:149–53. Suter E, McMorland G, Herzog W, Bray R. Conservative lower back treatment reduces inhibition in knee-extensor muscles: a randomized controlled trial. Journal of Manipulative and Physiological Therapeutics 2000;23:76–80. Urbach D, Nebelung W, Becker R, Awiszus F. Effects of reconstruction of the anterior cruciate ligament on voluntary activation of quadriceps femoris a prospective twitch interpolation study. Journal of Bone and Joint Surgery British 2001;83:1104–10. Urbach D, Nebelung W, Weiler HT, Awiszus F. Bilateral deficit of voluntary quadriceps muscle activation after unilateral ACL tear. Medicine & Science in Sports & Exercise 1999;31:1691–6. Williams GN, Barrance PJ, Snyder-Mackler L, Axe MJ, Buchanan TS. Specificity of muscle action after anterior cruciate ligament injury. Journal of Orthopaedic Research 2003;21:1131–7. Williams GN, Buchanan TS, Barrance PJ, Axe MJ, Snyder-Mackler L. Quadriceps weakness, atrophy, and activation failure in predicted noncopers after anterior cruciate ligament injury. American Journal of Sports Medicine 2005;33:402–7. Yerys S, Makofsky H, Byrd C, Pennachio J, Conkay J. Effect of mobilization of the anterior hip capsule on gluteus maximus strength. Journal of Manual & Manipulative Therapy 2002;10:218–24.
Manual Therapy 14 (2009) 421–426
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Original Article
Validity and reliability of ultrasonography for the longus colli in asymptomatic subjectsq Barbara Cagnie a, *, Erwin Derese a, Leen Vandamme a, Koenraad Verstraete b, Dirk Cambier a, Lieven Danneels a a b
Ghent University, Department of Rehabilitation Sciences and Physiotherapy, De Pintelaan 185, 3B3, 9000 Ghent, Belgium Ghent University, Department of Radiology, Ghent University Hospital, De Pintelaan 185, 1K12, 9000 Ghent, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 September 2007 Received in revised form 9 July 2008 Accepted 28 July 2008
The purposes of this study were to evaluate the reliability and validity of ultrasound (US) for measuring the cross-sectional area (CSA) of the longus colli (LC) as compared with magnetic resonance imaging (MRI), and to determine the change in CSA of the LC during contraction. 27 healthy volunteers participated in the study. In order to assess the validity of US, the US measurements of the CSA of the LC were compared to those determined with MRI. Two testers established the measurements to ascertain intra- and interrater reliability. The widely spaced limits of agreement (2SD ¼ 0.45) reflect the large variability between the measurements by US and MRI. The ICC for the intra- and interrater reliability for the CSA of the LC was respectively 0.71 (95% CI, 0.57–0.81; SEM, 0.17; SDD, 0.48) and 0.68 (95% CI, 0.48–0.81; SEM, 0.18; SDD, 0.50). The CSA of the LC increased significantly during contraction of the LC (p ¼ 0.006). Results from this study show that the validity and reliability of US to evaluate the CSA of the LC is questionable, which may be due to both anatomical characteristics and methodological limitations. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Ultrasonography Longus colli Reliability Validity
1. Introduction In recent years evidence has accumulated of impairment in deep cervical flexor muscle (DCF) function in neck pain sufferers (Falla et al., 2004a; Falla et al., 2004b; Falla, 2004; Jull et al., 2004; O’leary et al., 2007b). Evidence suggests that these muscles are important for the control and support of the cervical lordosis and maintenance of cervical spine postural form (Mayoux-Benhamou et al., 1994; Conley et al., 1995; Vasavada et al., 1998; Boyd-Clark et al., 2001; Boyd-Clark et al., 2002). Furthermore, specific therapeutic retraining of the DCF muscles has demonstrated efficacy in the management of patients with chronic neck pain and cervicogenic headache (Jull et al., 2002; O’leary et al., 2003; Falla et al., 2006a). Unfortunately, because these muscles are deeply situated, traditional methods such as palpation and manual muscle testing are unreliable for assessment of their function. Secondly, it is difficult to reach the DCF with surface electromyography (EMG). Nevertheless, Falla et al. (2003) described a novel surface EMG technique for the detection of DCF muscle activity. However, this technique is not applicable for
q This study was supported by the Research Foundation – Flanders (FWO). * Corresponding author. Tel.: þ32 9 332 52 65; fax: þ32 9 332 38 11. E-mail address:
[email protected] (B. Cagnie). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.07.007
routine assessment of the DCF in clinical practice (Falla et al., 2006b; O’leary et al., 2007a). Quantitative measurements of paraspinal muscle size can be obtained with both real-time ultrasonography (US) and magnetic resonance imaging (MRI) and there is growing support for their use in investigations of patients with spinal pain (Stokes et al., 2007; Whittaker et al., 2007; Elliott et al., 2008). MRI can be regarded as the gold standard for muscle imaging. Real-time US, however, has the advantages of widespread accessibility and lower cost. The muscles are visualized in real-time and measurements can be obtained in a relaxed state and in different states of contraction as well as during movements (Rezasoltani et al., 2002; Kiesel et al., 2007). The disadvantages of real-time US are its relatively limited field of view and its inability to provide pilot sections for confirmation of vertebral levels when the spine is imaged. This means that strict protocols must be followed to allow accurate measurement of the soft tissues. Today, US is frequently used as a diagnostic tool and as a biofeedback method for the muscles in the lumbo-pelvic region (Van et al., 2006). However, the use of US for the evaluation of the cervical muscles is sparse. Previous researchers have used US to evaluate the involvement of the dorsal neck muscles in chronic pain, while, to the best of our knowledge, this has not been used for the evaluation for the cervical flexor muscles (Rezasoltani et al., 1998; Rezasoltani et al., 1999; Kristjansson, 2004; Rankin et al., 2005).
422
B. Cagnie et al. / Manual Therapy 14 (2009) 421–426
The purposes of this study were 1) to set up a clear standardized protocol for the evaluation of the longus colli (LC) with US, 2) to evaluate the validity of US as compared with MRI (CSA) of the LC during rest 3) to determine the reliability of US for measuring the cross-sectional area and 4) to determine the change in CSA of the LC during contraction. 2. Materials and methods 2.1. Subjects US images were obtained in 27 healthy subjects (14 men and 13 women). Demographic details are shown in Table 1. Subjects were either sedentary or moderately active (0–10 h of sports a week). Exclusion criteria were pronounced neck or back trauma, neurological or inflammatory disorders, dizziness, vestibular symptoms or diabetes. The subjects should not have had neck pain, back pain or headache from cervical origin in 6 weeks before the scanning and at the moment of the scanning. MRI images were obtained in 18 of the 27 volunteers. Exclusion criteria for MRI were a cardiac pacemaker, claustrophobia, implanted metals, unremovable piercings, aneurysm clips, carotid artery vascular clamp, neurostimulator, cochlear or ear implants, and (possible) pregnancy within the first 3 months. The project was approved by the local ethics committees. Written informed consent was obtained from all subjects. 2.2. Study design The CSA of the LC was measured using real-time US and MRI. 2.2.1. Validity study In order to assess the validity of US, one examiner compared the US measurements of the CSA of the LC to those determined with MRI, both at the C5-C6 level. 2.2.2. Reliability study One tester established the measurements on three different occasions to ascertain intrarater reliability of US at rest. The time interval between the tests was one week. To determine the interrater reliability, each subject was imaged and measured by a second tester on the first test day. Each rater independently established anatomical landmarks and subject positioning. One single measurement on each side was taken. In order to determine the interrater reliability of the CSA measurement of the LC on MRI, both testers evaluated independently the same MRI images. In order to interpret the values obtained, the reliability of the CSA of the sternocleidomastoid (SCM) was also determined.
range of motion while trying to keep the SCM relaxed and without lifting their head of the surface (Falla et al., 2004b). A pressure cuff was used (Chattanooga Group Inc.), which was placed suboccipitally behind the subject’s cervical spine and inflated until a stable pressure of 20 mmHg was achieved. Prior to the test day, subjects were instructed in the movement and practiced targeting their maximal pressure without excessive action of the SCM. The pressure level that the subject could achieve and hold in a steady manner for 10 10 s with the SCM relaxed and without lifting the head off the surface was determined. 2.3. Ultrasound imaging technique The US measurements were performed by two experienced manual therapists using a real-time US apparatus (MyLab 30 CV - Esaote) with a 12 MHz linear array transducer (type LA 532). Both raters were trained by an experienced ultrasonographer which included instruction in the measurement technique, visualization of the LC and processing of the data. The measurement protocol using US was designed through several pilot trials and based on a fundamental knowledge of cervical anatomy and US. The subjects were supine in a neutral position, with their knees and hips bent, and the arms lying along the sides of the body. The subjects were first measured at rest, followed by a measurement during contraction. At rest, the cervical lordosis was supported by a folded towel. LC was scanned at C5-C6 level. This level was selected because of its clear visualization. Secondly, at this level, there is no overlap between LC and longus capitis muscle, as this muscle runs more laterally and attaches to the anterior tubercle of C6. Axial images were obtained by placing the middle of the probe perpendicular to the long axis of the anterior neck (Fig. 1) The bottom of the laryngeal prominence of the thyroid cartilage, which corresponds with the C5 level, served as a reference point for the probe, to assure that all measurements took place at the same level (Putz & Pabst, 1994; Moeller & Reif, 2007). The transducer was then moved approximately 1 cm laterally from the bottom of the laryngeal prominence to each side to image the left and right LC. Images were captured, stored and measured afterwards with the device-linked program MyLabDesk. During contraction, participants were asked to accurately maintain the predetermined level of contraction for at least 10 s while US measurements of the LC were made. The probe was positioned at the same place as during rest. The CSA was measured by using on-screen callipers. The LC was localized on an average depth of 2–3 cm. The outlines of the LC were identified by the following landmarks: anterolaterally by the
2.2.3. Rest versus contraction In addition to the measurement at rest, the CSA of the LC was also measured during contraction on the third test day by one examiner. To isolate the contraction, subjects were asked to perform a gentle nodding action to reach full cranio-cervical flexion
Table 1 Anthropometric variables. N ¼ 27 Age (year) Length (cm) Weight (kg) BMI (kg/m2) Sports (h/week) Neck perimeter (cm)
22.4 1.2 174.3 7.7 66.2 8.1 21.7 2.1 4.5 3.1 35.0 2.8
Fig. 1. Position of the transducer for imaging of the longus colli.
B. Cagnie et al. / Manual Therapy 14 (2009) 421–426
423
common carotid artery and internal jugular vein; anteromedially by the thyroid gland and posteriorly by the echogenic vertebra (Fig. 2). 2.4. MRI MRI was performed on a 3 Tesla magnet (Siemens Magnetom ‘Trio a Tim System’ with Syngo MR B13). A flexible surface coil, 20 50 cm, fixed over the anterior aspect of the participant’s neck was combined with the phased-array spine coil as a receiver coil combination. The subjects were placed in a comfortable and relaxed supine position, with their hips flexed to 45 and legs supported by foam wedges. The head was positioned in a neutral position, without rotation, lateral flexion or exaggerated lordosis. Compared to US, no towel was added as this was not possible due to the used coils. T2-weighted images of the cervical spine were obtained from the C0 to the C6 segmental level using the following imaging sequences: 20 slices with a slice thickness of 3 mm; Field of view read: 200 mm; relaxation time: 5130 ms; Echo time: 103 ms; Flip angle: 160 ; Acquisition time: 3:22 s. (Fig. 3). The measurements at the C5-C6 level were used to compare with the measurement results of the US. The location of the C5-C6 level was determined from parallel images in the mid-sagittal plane by identifying the spinous process of C5 and C6 and the C5–C6 intervertebral disc. Axial images were positioned parallel to the C2–C3 intervertebral disk, which may produce a slight measurement error for the CSA measures at the C5–C6 level due to the cervical lordosis. As a result, relative CSAs (rCSA) are reported. Muscle CSA was measured using on-screen callipers. The rCSA of LC was calculated by the number of pixels under each ROI in the x and y axes with DicomWorks. 2.5. Statistical analysis Data were analyzed with SPSS 15.0. All data are presented as mean standard deviation (SD). Because there were no significant differences between left and right side, the values of both sides were averaged for further analyses. For analysis of validity, differences between the measurements by the two methods for each subject were plotted against their mean (Bland & Altman, 1999). The 95% limits of agreement were
Fig. 3. MRI image on the C6-level. 1:longus colli, 2: sternocleidomastoid, 3: trachea, 4: vertebral body C6, 5: multifidus, 6: trapezius muscle, 7: common carotid artery, 8: internal jugular vein.
calculated, to indicate the extent the two methods are likely to differ due to observer error when used in practice. For analysis of intra- and intertester reliability, intra-class correlation coefficients [ICC1,1 and ICC2,1], standard error of measurement (SEM) and smallest detectable difference (SDD) were used. Defined with respect to a 95% level of confidence, the SDD is equal to 1.96 O2*SEM. The changes in CSA of the LC between rest and contraction were evaluated by a paired t-test with a 95% CI for the mean difference. 3. Results 3.1. Validity The within-subject differences between the MRI and US values plotted against the mean of measurements with 95% limits of agreement are seen in Fig. 4. The widely spaced limits of agreement (2SD ¼ 0.45) reflect the variability between the two methods. The average (SD) CSA of the LC on MRI (1.25 cm2 0.28) was larger than the CSA on US (1.22 cm2 0.37), although not significant (p ¼ 0.309). 3.2. Reliability The ICC for the intra- and interrater reliability for the CSA of the LC was respectively 0.71 (95% CI, 0.57–0.81) and 0.68 (95% CI, 0.48–0.81) (Table 2). The interrater measurements of LC on MRI showed good reliability (ICC ¼ 0.81; 95% CI, 0.64–0.90), whereas the interrater measurements of SCM showed excellent reliability (ICC ¼ 0.96; 95% CI, 0.87–0.98) (Table 3). 3.3. Rest versus contraction
Fig. 2. Ultrasound image of the longus colli at the C5/C6 level. 1: acoustic shadow of the trachea, 2: internal jugular vein, 3: sternocleidomastoid muscle, 4: common carotid artery, 5: longus colli, 6: thyroid gland.
The CSA of the LC during contraction (1.35 cm2 0.32) was significantly higher compared to rest (1.22 cm2 0.37) (p ¼ 0.006), with a 95% CI for the difference ranging from 0.04 to 0.20 (Table 4). This corresponds with an enhancement of 12%.
424
B. Cagnie et al. / Manual Therapy 14 (2009) 421–426 Table 3 Inter rater reliability of the CSA measurement of longus colli (LC) and sternocleidomastoid (SCM) on MRI.
0,60 +2SD
LC SCM
Difference US and MRI (cm2)
0,40
0,20
0,00
-0,20
-0,40 -2SD -0,60 0,00
0,50
1,00
1,50
2,00
2,50
Average US and MRI (cm2) Fig. 4. Agreement between US and MRI recordings plotted as mean values against differences between the two methods. The solid horizontal line illustrates the mean difference value with 2SD (dotted lines).
4. Discussion The present study is the first, to our knowledge, to document the evaluation of CSA of the longus colli with use of US. Results from this study show that the validity and reliability of US to evaluate the CSA of the LC is doubtful, which may be due to both the anatomic structure of the muscle and some methodological limitations. 4.1. Validity Compared with MRI, US seems to have a questionable validity. In previous comparable validity studies, Hides et al. (1995) did not found a difference between US and MRI measurements of the CSA of the lumbar multifidus, whereas Lee et al. found no acceptable validity for measuring the area of the cervical multifidus muscle (Hides et al., 1995; Lee et al., 2007). In this study, the widely spaced limits of agreement (2SD ¼ 44.58) reflects a large variability between the two methods. These differences may be explained by the size and anatomic structure of this muscle. The small CSA of the LC (about 1.20 cm2 compared with 2–7 cm2 for lumbar multifidus depending on the level) may amplify errors, thus influence the variability of measurements (Lee et al., 2007; Stokes et al., 2007). Secondly, the boundaries are not always distinct, which may influence the accuracy of the US measurement. This is especially true for the posteromedial boundary of the LC which is difficult to outline on US, due to the acoustic shadow of the trachea, which may explain the large variability. It should have been useful to watch the muscle contracting before taking Table 2 Inter- and intrarater reliability of the CSA measurement of longus colli (LC) on US.
Intra Inter
ICC (95% CI)
SEM
SDD
0.71 (0.57–0.81) 0.68 (0.48–0.81)
0.17 0.18
0.48 0.50
ICC (95% CI)
SEM
SDD
0.81 (0.64–0.90) 0.96 (0.87–0.98)
0.14 0.27
0.39 0.76
measurements at rest, as this technique has been found helpful for identifying the borders of multifidus more clearly (Stokes et al., 2005). It should also be noted that MRI as being the gold standard can be questioned for smaller muscles. MRI seems a very reliable method for large muscles, but it is less obvious to obtain reliable results for smaller and deeper muscles. This was also demonstrated in our study, in which the reliability of the LC was much lower on MRI than the reliability of the SCM. This fact should be kept in mind when interpreting the results of the validity study. The validity could have been affected by some methodology flaws. The positioning of the patient may have influenced the validity. In both the US and MRI study, patients were placed in a supine position, with their hips and knees bent. The head was positioned in a neutral position, without rotation, lateral flexion or exaggerated lordosis. However, in the US study, the cervical lordosis was supported by a folded towel, whereas this was not the case in the MRI scanner as this was not possible due to the fixed coils. Minimal differences in cervical lordosis probably may influence the results. Although not significant, the average CSA of the LC on MRI was larger than the CSA on US. This is in accordance with the study of Lee et al. who found that the values obtained by MRI are on average 2–3 mm2 larger than those obtained by US (Lee et al., 2007). This can be explained by the different scanning planes of US and MRI. In the cervical region, the lordosis may play a role in the measurement of muscle thickness. The T2-weighted MRI images were taken perpendicular to the C2–C3 vertebral disk, while the US images are supposed to be taken perpendicular to the spine and thus the LC. This gives an overestimation of the CSA on MRI which may explain why MRI shows a higher value for the CSA in comparison with US (Lee et al., 2007). In order to reduce measurement errors and thus increase validity, it would have been more accurate to take an image parallel to the C5–C6 intervertebral disk instead of the C2–C3 level. In addition, as the bottom of the laryngeal prominence of the thyroid cartilage was taken as a reference point for C5–C6, it is possible that the MRI and US levels were not exactly the same, which may also could have influenced the validity. Measurements were taken by two trained manual therapists. Although trained in this technique, it is possible that the fact that they were no experienced radiologists may have affected validity. 4.2. Reliability The reliability of the protocol was moderate (ICC: 0.68–0.71) (Shrout, 1998). Kiesel et al. (2007), Lee et al. (2007) and Pressler et al. (2006) demonstrated quite similar results as our study (Pressler et al., 2006; Kiesel et al., 2007; Lee et al., 2007). In contrast, Rankin et al. (2005), Stokes et al. (2005) and Van et al. (2006), did however found quite high ICC values (varying between 0.97 and 1.00) for the evaluation of the dorsal neck muscles and the lumbar multifidus (Rankin et al., 2005; Stokes
Table 4 CSA of the Longus colli (LC) (cm2) at rest and during contraction. LC rest
LC contraction
Mean difference (95% CI)
p-Value
1.22 0.27
1.35 0.32
0.04–0.20
0.006
B. Cagnie et al. / Manual Therapy 14 (2009) 421–426
et al., 2005; Van et al., 2006). The fact that in these studies mean measurements rather than single measurements were used, should be borne in mind. In addition to the findings of moderate intrarater and interrater reliability, the reported SEM and SDD values are quite high when compared to the resting CSA of the muscle. SEM and SDD values ranged from 0.17 to 018 and 0.48 to 0.50 respectively. The SDD values indicate that the CSA would change by at least 0.48–0.50 cm, to be 95% confident that true change occurred, which would require a 40% change. This is rather high compared to previous studies (Pressler et al., 2006; Van et al., 2006; Wallwork et al., 2007). These high SEM and SDD values will limit the ability to detect a change in muscle CSA with US in longitudinal studies. The fact that only one measurement was taken, could partly explain the moderate reliability. It is likely that a procedure which utilises more than one measurement would increase the SEM and SDD values. In order to reduce measurement error, averaged values of repeated trials should be considered to determine the CSA. Springer et al. demonstrated that by taking an average of three measures of the transversus abodminis muscle with US, the SEM reduced by approximately 50% (Springer et al., 2006). 4.3. Rest versus contraction During contraction, a significant increase in muscle CSA was found. However, the questionable validity and reliability should be kept in mind when interpreting the results. Although significant, the amount of increase is not comparable to previous studies, in which the lumbar and cervical multifidus were investigated during contraction (McMeeken et al., 2004; Kiesel et al., 2007; Lee et al., 2007). In these studies, an increase of up to 60% between rest and contraction was noted, compared to 12% in our study. One of the hypotheses of having a lower increase may be the function of the LC. In order to isolate the contraction, we asked the subjects to perform a cranio-cervical flexion. Previous studies have indicated that this method is specific to the anatomical action of the DCF, which encompasses the LC and longus capitis. In the studies of Conley et al. (1995), Falla et al. (2003) and O’leary et al. (2006), no difference is made between both muscles (Conley et al., 1995; Falla et al., 2003; O’leary et al., 2007b). However, a recent study has indicated the need to differentiate between the LC and longus capitis, since there is a clear difference in activation of both muscles (Cagnie et al., 2008). The longus capitis seems to play a more important role in the performance of the cranio-cervical flexion compared to the LC. This could be due to the fact that the primary anatomical action of Lca is flexion of the cranio-cervical junction, whereas the primary action of LC is flattening of the cervical lordosis (Falla et al., 2006b). It may be possible that greater changes in CSA would occur if 1) a different task was used to contract the parts of longus colli that were imaged or 2) the LC was measured at a higher segmental level. 4.4. Clinical implications As mentioned in the introduction section, a key advantage of US imaging is the ability to visualise a muscle contracting. If further testing suggests that measurement of changes in CSA is not reliable, the use of imaging as a biofeedback tool should still be considered. As it is not possible to palpate the longus colli, neither to capture this muscle with surface EMG, US may provide visual feedback in order to enhance motor learning for contracting this muscle. Evidence that feedback with US enhances motor learning holds promise for the future but studies are needed on subjects with neck pain.
425
5. Conclusion Results from this study show that the validity and reliability of US to evaluate the CSA of the LC is questionable, which may be due to both anatomical characteristics and methodological limitations. The small size of the muscle and difficulties in outlining the boundaries of the muscle may explain the large variability. Taking an average of different measures should be considered in further research. If further testing suggests that measurement of changes in CSA is not reliable, the use of imaging as a biofeedback tool should still be considered. Acknowledgements We would like to thank Dr. H. Vinck for his useful help during the set up of the ultrasound protocol. Reference Bland J, Altman D. Measuring agreement in method comparison studies. Stat Methods Res 1999;8:135–60. Boyd-Clark LC, Briggs CA, Galea MP. Comparative histochemical composition of muscle fibres in a pre- and a postvertebral muscle of the cervical spine. J Anat 2001;199:709–16. 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;27:694–701. Cagnie B, Dickx N, Peeters I, Tuytens J, Achten E, Cambier D, et al. The use of functional MRI to evaluate cervical flexor activity during different cervical flexion exercises. J Appl Physiol 2008;104:230–5. Conley MS, Meyer RA, Bloomberg JJ, Feeback DL, Dudley GA. Noninvasive analysis of human neck muscle function. Spine 1995;20:2505–12. Elliott J, Jull G, Noteboom JT, Galloway G. MRI study of the cross-sectional area for the cervical extensor musculature in patients with persistent whiplash associated disorders (WAD). Man Ther 2008;13:258–65. Falla D. Unravelling the complexity of muscle impairment in chronic neck pain. Man Ther 2004;9:125–33. 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. Phys Ther 2003;83:899–906. Falla D, Jull G, Hodges P, Vicenzino B. An endurance-strength training regime is effective in reducing myoelectric manifestations of cervical flexor muscle fatigue in females with chronic neck pain. Clin Neurophysiol 2006a;117: 828–37. Falla D, Jull G, O’leary S, Dall’Alba P. Further evaluation of an EMG technique for assessment of the deep cervical flexor muscles. J Electromyogr Kinesiol 2006b;16:621–8. Falla D, Jull G, Rainoldi A, Merletti R. Neck flexor muscle fatigue is side specific in patients with unilateral neck pain. Eur J Pain 2004a;8:71–7. Falla DL, Jull GA, Hodges PW. Patients with neck pain demonstrate reduced electromyographic activity of the deep cervical flexor muscles during performance of the craniocervical flexion test. Spine 2004b;29:2108–14. Hides JA, Richardson CA, Jull GA. Magnetic resonance imaging and ultrasonography of the lumbar multifidus muscle. Comparison of two different modalities. Spine 1995;20:54–8. Jull G, Kristjansson E, Dall’Alba P. Impairment in the cervical flexors: a comparison of whiplash and insidious onset neck pain patients. Man Ther 2004;9:89–94. Jull G, Trott P, Potter H, Zito G, Niere K, Shirley D, et al. A randomized controlled trial of exercise and manipulative therapy for cervicogenic headache. Spine 2002;27:1835–43. Kiesel KB, Uhl TL, Underwood FB, Rodd DW, Nitz AJ. Measurement of lumbar multifidus muscle contraction with rehabilitative ultrasound imaging. Man Ther 2007;12:161–6. Kristjansson E. Reliability of ultrasonography for the cervical multifidus muscle in asymptomatic and symptomatic subjects. Man Ther 2004;9:83–8. Lee JP, Tseng WY, Shau YW, Wang CL, Wang HK, Wang SF. Measurement of segmental cervical multifidus contraction by ultrasonography in asymptomatic adults. Man Ther 2007;12:286–94. Mayoux-Benhamou MA, Revel M, Vallqe C, Roudier R, Barbet JP, Bargy F. Longus colli has a postural function on cervical curvature. Surg Radiol Anat 1994;16: 367–71. McMeeken JM, Beith ID, Newham DJ, Milligan P, Critchley DJ. The relationship between EMG and change in thickness of transversus abdominis. Clin Biomech (Bristol, Avon) 2004;19:337–42. Moeller T, Reif E. Pocket atlas of sectional anatomy. Georg Thieme Verlag 2007. O’leary S, Falla D, Jull G. Recent advances in therapeutic exercise for the neck: implications for patients with head and neck pain. Aust Endod J 2003;29: 138–42. O’leary S, Falla D, Jull G, Vicenzino B. Muscle specificity in tests of cervical flexor muscle performance. J Electromyogr Kinesiol 2007a;17:35–40.
426
B. Cagnie et al. / Manual Therapy 14 (2009) 421–426
O’leary S, Jull G, Kim M, Vicenzino B. Cranio-cervical flexor muscle impairment at maximal, moderate, and low loads is a feature of neck pain. Man Ther 2007b;12:34–9. Pressler JF, Heiss DG, Buford JA, Chidley JV. Between-day repeatability and symmetry of multifidus cross-sectional area measured using ultrasound imaging. J Orthop Sports Phys Ther 2006;36:10–8. Putz R, Pabst R. Sobotta Atlas van de menselijke anatomie. Houten: Bohn Stafleu Van Loghum; 1994. 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. Man Ther 2005;10:108–15. Rezasoltani A, Kallinen M, Malkia E, Vihko V. Neck semispinalis capitis muscle size in sitting and prone positions measured by real-time ultrasonography. Clin Rehabil 1998;12:36–44. Rezasoltani A, Malkia E, Vihko V. Neck muscle ultrasonography of male weightlifters, wrestlers and controls. Scand J Med Sci Sports 1999;9:214–8. Rezasoltani A, Ylinen J, Vihko V. Isometric cervical extension force and dimensions of semispinalis capitis muscle. J Rehabil Res Dev 2002;39:423–8. Shrout P. Measurement of reliability and agreement in psychiatry. Statisical methods in medical research. Stat Methods Med Res 1998;7:301–17.
Springer BA, Mielcarek BJ, Nesfield TK, Teyhen DS. Relationships among lateral abdominal muscles, gender, body mass index, and hand dominance. J Orthop Sports Phys Ther 2006;36:289–97. Stokes M, Hides J, Elliott J, Kiesel K, Hodges P. Rehabilitative ultrasound imaging of the posterior paraspinal muscles. J Orthop Sports Phys Ther 2007;37:581–95. Stokes M, Rankin G, Newham DJ. Ultrasound imaging of lumbar multifidus muscle: normal reference ranges for measurements and practical guidance on the technique. Man Ther 2005;10:116–26. Van K, Hides JA, Richardson CA. The use of real-time ultrasound imaging for biofeedback of lumbar multifidus muscle contraction in healthy subjects. J Orthop Sports Phys Ther 2006;36:920–5. Vasavada AN, Li S, Delp SL. Influence of muscle morphometry and moment arms on the moment-generating capacity of human neck muscles. Spine 1998;23: 412–22. Wallwork TL, Hides JA, Stanton WR. Intrarater and interrater reliability of assessment of lumbar multifidus muscle thickness using rehabilitative ultrasound imaging. J Orthop Sports Phys Ther 2007;37:608–12. Whittaker JL, Teyhen DS, Elliott JM, Cook K, Langevin HM, Dahl HH, et al. Rehabilitative ultrasound imaging: understanding the technology and its applications. J Orthop Sports Phys Ther 2007;37:434–49.
Manual Therapy 14 (2009) 427–432
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Original Article
Evaluation of stretching position by measurement of strain on the ilio-femoral ligaments: An in vitro simulation using trans-lumbar cadaver specimens Egi Hidaka a, Mitsuhiro Aoki b, *, Takayuki Muraki a, Tomoki Izumi a, Misaki Fujii a, Shigenori Miyamoto b a b
Graduate School of Health Sciences, Sapporo Medical University, Sapporo, Japan Department of Physical Therapy, Sapporo Medical University School of Health Sciences, Sapporo, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 March 2008 Received in revised form 2 July 2008 Accepted 28 July 2008
The ilio-femoral ligament is known to cause flexion contracture of the hip joint. Stretching positioning is intended to elongate the ilio-femoral ligaments, however, no quantitative analysis to measure the effect of stretching positions on the ligament has yet been performed. Strains on the superior and inferior iliofemoral ligaments in 8 fresh/frozen trans-lumbar cadaveric hip joints were measured using a displacement sensor, and the range of movement of the hip joints was recorded using a 3Space Magnetic Sensor. Reference length (L0) for each ligament was determined to measure strain on the ligaments. Hip positions at 10 adduction with maximal external rotation, 20 adduction with maximal external rotation, and maximal external rotation showed larger strain for the superior ilio-femoral ligament than the value obtained from L0, and hip positions at 20 external rotation with maximal extension and maximal extension had larger strain for the inferior ilio-femoral ligament than the value obtained from L0 (p < 0.05). Superior and inferior ilio-femoral ligaments exhibited positive strain values with specific stretching positions. Selective stretching for the ilio-femoral ligaments may contribute to achieve lengthening of the ligaments to treat flexion contracture of the hip joint. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Fresh cadaver Hip Ligaments Biomechanics
1. Introduction The ilio-femoral ligament is known to be the strongest ligament in human body and provides stability to the hip in the standing position (Hewitt et al., 2001; Neumann, 2002; Werner, 2004). The ilio-femoral ligament, which is located on the anterior aspect of the hip joint limits hyperextension of the hip and is known to be a cause of flexion contracture of the hip joint (O’Malley, 1959; Fuss and Bacher, 1991; French, 2007). For the treatment of flexion contracture associated with adult hip osteoarthritis, soft tissue release surgery, as advocated by O’Malley, has been performed (O’Malley, 1959). To prevent recurrence of flexion contracture of the hip, passive stretching of the released ilio-femoral ligament is performed postoperatively. On the other hand, stretching of the hip has been used for conditioning the limbs and trunk to improve flexibility of the lower extremities prior to strenuous physical activities, such as soccer, football, and handball (Mo¨ller et al., 1985; Zakas et al., 2003; Zakas et al., 2006). Thus, it is important for therapists and trainers to understand effective stretching positions for the ilio-femoral ligament.
* Corresponding author. Department of Physical Therapy, Sapporo Medical University School of Health Sciences, South-3, West-17, Chuo-ku, Sapporo 0608556, Japan. Tel.: þ81 11 611 2111; fax: þ81 11 611 2150. E-mail address:
[email protected] (M. Aoki). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.07.006
With regard to stretching positions for the ilio-femoral ligaments, Kapandji (1970) reported that the superior ilio-femoral ligament was stretched during hip external rotation and adduction, whereas the inferior ilio-femoral ligament was stretched during hip extension. Werner (2004) reported that the superior ilio-femoral ligament was stretched during hip external rotation, adduction and extension, whereas the inferior ilio-femoral ligament was stretched during hip internal rotation and extension. Lee (2004) reported that the superior ilio-femoral ligament was stretched during the combination of hip external rotation and adduction, with slight extension, whereas the inferior ilio-femoral ligament was stretched during hip extension. These stretching positions were determined from anatomical observations of the origin and insertion of the ligaments or kinesiological assumptions based on motion analysis. On the basis of a cadaveric study, Fuss and Bacher (1991) reported that the superior ilio-femoral ligament was stretched mainly during hip external rotation, whereas the inferior iliofemoral ligament was stretched mainly during hip extension. Although several authors have advocated various stretching positions, no consensus has been reached. Therefore, quantitative analysis to measure the effect of stretching on the ilio-femoral ligament is required to resolve this question. The purpose of this study was to measure strain on the superior and inferior ilio-femoral ligaments of the hip joints during passive stretching by using trans-lumbar cadaver specimens, and to
428
E. Hidaka et al. / Manual Therapy 14 (2009) 427–432
determine joint positions with positive ligament strain for the selective stretching of the ilio-femoral ligaments. 2. Methods 2.1. Preparation of specimens Eight fresh/frozen trans-lumbar specimens (four left hips and four right hips) with no evidence of osteoarthritis or fracture were used in this experiment. The age of the specimens at death was 67– 91 years (average, 80.3 years). Within 24 h after death, the specimens were transferred from their respective hospitals to the Department of Anatomy at the University and were kept in a freezer at 20 C. Frozen specimens were separated into 2 parts through the waist at the L1 level; i.e., into trans-thoracic specimen and trans-lumbar specimens. Thawing of the trans-lumbar specimens at room temperature began 24 h before preparation. After removal of the internal organs in a saline solution containing 0.1% sodium azide, the specimens were washed with clean saline. The skin, fascia, muscles, nerves and vessels were all removed leaving the lumbar spine, pelvic ring, hip joints, and both femurs. Ligaments of the hip joint were clearly exposed. To achieve the full range of passive hip joint motion, each translumbar specimen was secured by Kirschner wires and screws onto a thick wooden pole while maintaining the pelvis in an upright position with 30 of anterior tilt (Saito and Nagasaki, 2002). The anterior torsion angle of the femoral neck was set at 0 , and this position was regarded as the neutral hip position in this experiment. A plastic rod was inserted into the proximal femur in the vertical direction. The distal femur was then amputated (Fig. 1). According to the classification advocated by Kapandji (1970), the ilio-femoral ligament is divided into two parts. The ligament that originates from the antero-superior acetabular wall and inserts into the supero-lateral aspect of the inter-trochanteric line of the femur is designated as the superior ilio-femoral ligament, and the ligament that originates from anterior wall of the acetabulum and inserts into the infero-medial aspect of the inter-trochanteric line of the femur is designated as the inferior ilio-femoral ligament (Fig. 2). 2.2. Measurement devices Strain on the ilio-femoral ligament was measured using a displacement sensor. This device consisted of a coil sensor and
3SPACE Pulse Coder
Fig. 2. The superior and inferior ilio-femoral ligaments. (a) The superior ilio-femoral ligament originates from the antero-superior acetabular wall, extends parallel to the femoral neck axis, and inserts into the supero-lateral aspect of the inter-trochanteric line of the femur. (b) The inferior ilio-femoral ligament originates from the anterior wall of the acetabulum, extends parallel to the femoral shaft axis, and inserts into the infero-medial aspect of the inter-trochanteric line of the femur. The displacement sensor, which had 10-mm long points, was mounted on the center of the superior and inferior ilio-femoral ligament parallel to the ligament fibers.
a brass pipe. Changes in the distance between the points of the coil and the brass pipe enabled the measurement of the strain on the ligament. The range of measurement was 3–15 mm. The sensor, equipped with 10-mm long points, was mounted on the center of the superior and inferior ilio-femoral ligament parallel to the ligament fibers. Fishhook-like barb points prevented the sensors from slipping out of the ligament (Fig. 2). By using 3-power digital images projected on SCION Images, the accuracy of the preliminary calibration of the Pulse Coder, which was attached to the isolated iliofemoral ligament, was found to be 0.01 mm root mean square (RMS). A six-degree-of-freedom electromagnetic tracking device 3Space Tracker System was used for the measurement and monitoring of the hip angles. The rotation angle was defined as the rotation of the femur along the longitudinal axis of the femoral shaft. Within 750-mm range from the source, the positional accuracy was 0.8 mm RMS, and the angular accuracy was 0.5 RMS (Muraki et al., 2006). 2.3. Experimental procedure
Rod
Wooden jig
X
Z Y
Fig. 1. Experimental set-up. A trans-lumbar specimen was fixed on the wooden pole keeping the pelvis in an upright position with 30 of anterior tilt. A six-degree-offreedom electromagnetic tracking device 3Space Tracker System was used to measure hip angles.
On the basis of data obtained from preliminary experiments, flexion, abduction, and internal rotation for the superior iliofemoral ligament, and flexion, adduction, and internal rotation for the inferior ilio-femoral ligament, were excluded from this study, as the ilio-femoral ligaments were loose in those hip positions. For measurement of the superior ilio-femoral ligament, external rotation was included as a primary stretching position as it was shown to stretch the ligament in preliminary observation. The 6 stretching positions for the superior ilio-femoral ligament were determined as follows: (a) maximal external rotation, (b) maximal adduction, (c) maximal extension, and (d) 10 adduction with maximal external rotation, (e) 20 adduction with maximal
E. Hidaka et al. / Manual Therapy 14 (2009) 427–432
external rotation, and (f) 10 extension with maximal external rotation. For measurement of the inferior ilio-femoral ligament, extension was included as a primary stretching position as it was shown to stretch the ligament in preliminary observations. Thus, the 7 stretching positions for the inferior ilio-femoral ligament were determined as follows: (a) maximal extension, (b) maximal abduction, (c) maximal external rotation, (d) 20 external rotation with maximal extension, (e) 40 external rotation with maximal extension, (f) 20 abduction with maximal extension, and (g) 40 abduction with maximal extension. Strain on the ligament at each hip position was measured at the end of range, which was determined by Grade III mobilization after Kaltenborn’s procedure (Kaltenborn, 1999). The extent of unidirectional passive motion, applied by therapists, on the joints was graded from I to III. In that grading, Grade III mobilization comprised the manual application of force at a point which therapist perceived end-feel of the joint and observed no further stretching of the capsule or ligament. As the stretching procedure described in the Manual (Evjenth and Hamberg, 1984) is applied to the joint for 10–12 s after passive motion of the hip joint has reached the end of the range of movement, each position was maintained for more than 10 s until no increase or decrease in strain values was observed. Strain measurement was randomized in order for the superior and inferior ligament. During the mobilizations of the hip joint, the interclass correlation coefficient (1,1) of strain values obtained from 3 independent measurements using a displacement sensor was 0.936 for the ilio-femoral ligament. During the stretching procedures, the 3Space Tracker System during stretching procedures demonstrated the following average maximal ranges of hip motion; 23.9 6.4 for extension, 41.8 11.3 for external rotation, 22.0 11.1 for adduction, and 42.1 19.5 for abduction. The inter-class correlation coefficient (1,1) of measured ranges of motion during mobilization was 0.954. Throughout the experiment, which was performed at room temperature (22 C), the specimens were kept moist by spraying with saline solution every 10 min. To preserve smooth joint motion and reduce hysteresis of the ligaments, passive hip movement to the end-range was performed more than 10 times as preconditioning (Woo et al., 1986). 2.4. Identification of reference length (L0) and data analysis Based on a previous report using cadaver shoulders (Urayama et al., 2001), reference length (L0) was determined for each ligament. L0 was the length at which the angle-deformation curve of the ligament started to indicate a sudden increase in angle (Fig. 3). The L0 was the point between the laxity and linear regions in the angle-deformation curve. The displacement of the ligament (DL) was defined as the change in length from the L0. Since the distance between the points of the displacement sensor attached on the ligaments was not always the same, the strain on each ligament was calculated using the following formula (Muraki et al., 2006).
Strainð%Þ ¼ DLðmmÞ=L0 ðmmÞ 100 Strain values greater than 0% indicated positive stretching of the ligament from the L0. Values less than 0% indicated no stretching, and were presented as 0% strain. 2.5. Statistical analysis Statistical analysis was performed using SPSS for Windows ver.11.5J. One-way analysis of variance and post hoc Dunnett’s
a
429
b a laxity region
c c linear region
angle
b L0 deformation Fig. 3. Identification of reference length (L0). Reference length is the length at which the angle-deformation curve of the ligament starts to indicate a sudden increase in angle.
multiple comparisons test were used to determine the hip joint positions that showed significantly greater strain than that observed at the L0. The significance level was set at 0.05. 3. Results 3.1. The superior ilio-femoral ligament Strain on the superior ilio-femoral ligament with the hip at 10 adduction with maximal external rotation (3.2 3.3%), 20 adduction with maximal external rotation (4.0 4.2%), and maximal external rotation (3.7 3.0%) was significantly larger than the value obtained for L0 (p < 0.05). Few other hip positions demonstrated any positive strain on the superior ilio-femoral ligament (Fig. 4). 3.2. The inferior ilio-femoral ligament Strain on the inferior ilio-femoral ligament with the hip at maximal extension (2.1 2.1%) and 20 external rotation with maximal extension (1.8 2.1%) was significantly larger than the value obtained for L0 (p < 0.05). Few other hip positions demonstrated any positive strain on the inferior ilio-femoral ligament (Fig. 5). 4. Discussion Based on the traditional concepts of stretching to obtain effective lengthening of the ligaments, joint positioning by separating the origin and insertion of the ligament is important. The present study demonstrated that a significant amount of strain was observed on both the superior ilio-femoral ligament and the inferior ilio-femoral ligament in several hip positions.
430
E. Hidaka et al. / Manual Therapy 14 (2009) 427–432
10
*
8
*
*
6
4
2
0
ER
Add
Ext
Ext10°+ER
Add10°+ER
Add20°+ER
Fig. 4. Strain on the superior ilio-femoral ligament. The Y-axis indicates strain (%). Error bars indicate standard deviations. *A statistically significant difference between the obtained strain value and L0 value (p < 0.05).
The superior ilio-femoral ligament originates from the anterosuperior acetabular wall, extends parallel to the femoral neck axis, and inserts into the supero-lateral aspect of the inter-trochanteric line of the femur (Werner, 2004). As for external rotation, Kapandji (1970), Oatis (2004), and Werner (2004) indicated that adduction also stretched the ligament. The limiting factor of hip adduction was not only the superior ilio-femoral ligament but also the ischiofemoral ligament which extends from the ischium to the posterior aspect of the femoral neck (Luttgens and Wells, 1982; Fuss and Bacher, 1991; Oatis, 2004). The inferior ilio-femoral ligament originates from the anterior wall of the acetabulum, extends parallel to the femoral shaft axis, and inserts into the infero-medial aspect of the inter-trochanteric line of the femur (Werner, 2004). As for extension, Werner (2004) indicated that internal rotation also stretched the ligament. The limiting factor of hip internal rotation was reported to be the ischiofemoral ligament (Kapandji, 1970; Luttgens and Wells, 1982; Fuss and Bacher, 1991; Oatis, 2004; Werner, 2004). Although several authors have advocated various stretching positions, no consensus has been reached. While the anatomy of the hip joint capsule and ligaments has been well described, little or no information had been available on the material properties of the ilio-femoral ligaments (Fuss and Bacher, 1991). Recently the tensile properties of the ilio-femoral ligaments were reported by Hewitt et al. (2002). By using isolated ilio-femoral ligaments from cadaver hips, breaking strength and stiffness were measured. According to the report, the breaking strength and stiffness of the superior and inferior ilio-femoral ligaments were 320 N and 351 N, and 98.7 N/mm and 100.7 N/mm, respectively. In addition, it was noted that the ilio-femoral ligaments were the strongest capsular ligaments in the human body.
Further, from the stress–strain curves which were obtained from the cross-sectional area of the ligaments, maximal stress and strain on the ilio-femoral ligaments were found to be 2.7 MPa and 6.2%, respectively (Hewitt et al., 2001). This strain value was small compared with the values reported for the inferior gleno-humeral ligaments of the shoulder, i.e., 10.9% (Bigliani et al., 1992) and 9.3% (Ticker et al., 1996). In order to apply these biomechanical values to mobilization of the hip in patients, it is supposed that passive mobilization of the joint is performed by the application of a degree of force that can be estimated form the cross-sectional area of the ligaments. However, accurate evaluation of the cross-sectional area of the ilio-femoral ligament in vivo for each subject is difficult and the joint torque that should be applied to the individual hip joint cannot be clearly determined. Excessive torque in an inappropriate direction on a painful and contracted hip joint can occasionally cause ligament and cartilage damage, or even fracture (French, 2007; Hewitt et al., 2001; Wainner, 2004; Izumi et al., in press). Therefore, the direct application of these biomechanical values to patients involves considerable risk. For these reasons, this experimental study was designed to reduce the disparity between the biomechanical values and clinical practice. Taking into consideration a suitable safety range for the mobilization force used for the treatment of patients, in this study Grade III uni-directional mobilization after Kaltenborn’s procedure (Kaltenborn, 1999) was applied to the hip to simulate adequate torque to stretch and measure strain on the ilio-femoral ligaments. As a result, the maximal strains on the superior and inferior iliofemoral ligaments obtained by the mobilization were 3.2–4.1% and 1.8–2.1%, respectively, and those values were consistent with the strain values of toe lesions of the ilio-femoral ligaments (2–4%)
10 8 6 4
*
*
2 0
Ext
Abd
ER
ER20° +Ext ER40° +Ext Abd20° +Ext Abd40° +Ext
Fig. 5. Strain on the inferior ilio-femoral ligament. The Y-axis indicates strain (%). Error bars indicate standard deviation. *A statistically significant difference between the obtained strain value and L0 value (p < 0.05).
E. Hidaka et al. / Manual Therapy 14 (2009) 427–432
(Hewitt et al., 2001). This suggests that Grade III mobilization can apply adequate torque to the hip joint so as to provide maximal strain without resulting in injury to the ilio-femoral ligaments (Woo et al., 1994; Tillan and Hertling, 1996). The cause of flexion contracture of the hip joint after hip osteoarthritis or cerebro-vascular disease is thought to be the coexisting contracture of the ilio-femoral ligaments and the iliopsoas, adductor and internal rotator muscles of the hip (Sage, 1987; Lee et al., 1997; French, 2007). If muscle contracture has progressed and the hip is in a flexed position, surgical release of both the psoas muscle and ilio-femoral ligaments is required (O’Malley, 1959; Morais Filho et al., 2006). To prevent recurrence of the contracture, postoperative stretching of the ilio-femoral ligament and ilio-psoas muscle will be indicated using this procedure. Patients may require long-term physical therapy to treat the contracture of the hip. To improve these pathological conditions, a selective stretching procedure for the ilio-femoral ligaments is considered to be very important. Selective stretching applied separately for the superior and inferior ilio-femoral ligaments is thought to be effective in reducing total stress on the contracted ligaments of the hip joint. This may diminish the risk of chronic inflammation generated by micro rupture due to excessive strain (Tillan and Hertling, 1996). Based on the data obtained from this study, selective stretching for the superior ilio-femoral ligament with the hip at 10 or 20 adduction with maximal external rotation, and maximal external rotation, and that for the inferior ilio-femoral ligament with the hip at 20 external rotation with maximal extension and maximal extension should be performed as specific procedures. A series of selective stretches for the ilio-femoral ligaments may contribute to an effective treatment for flexion contracture of the hip joint. There are limitations to this study. First, since the trans-lumbar specimens were harvested from aged cadavers, the range of motion of the specimens might be different from those of specimens from younger people. The range of passive hip motion measured in the current study was 23.9 6.4 for extension, 41.8 11.3 for external rotation, 22.0 11.1 for adduction, and 42.1 19.5 for abduction. These values were similar to the standard values obtained in clinical measurements (Boone and Azen, 1979; Walker et al., 1984; Roach and Miles, 1991). Second, the mechanical properties of the ligaments in aged specimens might be different from those of specimens from younger people. Because tendon strain also decreases with age, the strain on the ilio-femoral ligaments observed in aged cadavers was thought to be less than that on tendons in younger cadavers (Yamada, 1970). Third, the displacement sensor had no capacity to measure strain in all dimensions. Therefore, the reference length (L0) was determined for each fiber direction at the center of the superior and inferior ilio-femoral ligaments. We considered that strain values obtained by the displacement sensor in those location represented strains of the superior and inferior ilio-femoral ligaments. Although we adopted uni-directional Grade III hip mobilization after Kaltenborn’s procedure (Kaltenborn, 1999) to obtain consistent strain on the ilio-femoral ligaments in this experimental study, Grade III or IV mobilization advocated by Maitland with reciprocating motions near the end-range is also considered to be applicable to clinical cases (Maitland, 1991). To resolve contracture of the hip joint with joint mobilization, long-term mechanical stress through an adequate amount of stretching should be applied to the contracted ligaments. This process may induce remodeling of the collagen tissue of the ligament and joint capsule, which will, in turn, help transform the tight, stiff joint into a flexible, mobile joint (Tillan and Hertling, 1996). For this reason, together with fact that the basic data support the beneficial mechanical effect of this procedure, joint mobilization should become a widely accepted clinical practice for the treatment of hip joint contracture.
431
5. Conclusion This study is designed to elucidate selective stretching positions for the superior and inferior ilio-femoral ligaments of the hip by measurement of strain. Based on the data obtained from this study, selective stretching for the superior ilio-femoral ligament with the hip at 10 or 20 adduction with maximal external rotation and maximal external rotation, and that for the inferior ilio-femoral ligament with the hip at 20 external rotation with maximal extension and maximal extension should be performed as specific procedures. A series of selective stretches for the ilio-femoral ligaments may contribute to an effective treatment for flexion contracture of the hip joint. Acknowledgments Authors sincerely thank E. Uchiyama, D. Suzuki, M. Fujii, S. Ohshiro, H. Takasaki, and H. Miyamoto for their useful suggestions and technical support. References Bigliani LU, Pollock RG, Soslowsky LJ, Flatow EL, Pawluk RJ, Mow VC. Tensile properties of the inferior glenohumeral ligament. Journal of Orthopaedic Research 1992;10:187–97. Boone DC, Azen SP. Normal range of motion of joints in male subjects. Journal of Bone and Joint Surgery 1979;61A:756–9. Evjenth O, Hamberg J. The extremities. In: Muscle stretching in manual therapy: a clinical manual, vol. 1. Alfta: Alfta Rehab Forlag; 1984. p. 7–12. French HP. Physiotherapy management of osteoarthritis of the hip: a survey of current practice in acute hospitals and private practice in the Republic of Ireland. Physiotherapy 2007;93:253–60. Fuss FK, Bacher A. New aspects of the morphology and function of the human hip joint ligaments. American Journal of Anatomy 1991;192:1–13. Hewitt J, Guilak F, Glisson R, Vail TP. Regional material properties of the human hip joint capsule ligaments. Journal of Orthopaedic Research 2001;19:359–64. Hewitt JD, Glisson RR, Guilak F, Vail TP. The mechanical properties of the human hip capsule ligaments. Journal of Arthroplasty 2002;17:82–9. Izumi T, Aoki M, Muraki T, Hidaka E, Miyamoto S. Stretching positions for the posterior capsule of the glenohumeral joint: strain measurement using cadaver specimens. American Journal of Sports Medicine, in press. Kaltenborn FM. The extremities. In: Manual mobilization of the joint: the Kaltenborn method of joint examination and treatment. 5th ed., vol. 1. Oslo: Olaf Norlis Bokhandel; 1999. p. 21–8. Kapandji IA. Lower limb. In: The physiology of the joints. 2nd ed., vol. 2. New York: Churchill Livingstone; 1970. p. 34–41. Lee D. The pelvic girdle: approach to be the examination and treatment of the lumbopelvic-hip. 3rd ed. New York: Churchill Livingstone; 2004. p. 105. Lee LW, Kerrigan DC, Della Croce U. Dynamic implications of hip flexion contractures. American Journal of Physical Medicine and Rehabilitation 1997;76:502–8. Luttgens K, Wells KF. Kinesiology: scientific basis of human motion. 7th ed. Philadelphia: Saunders College; 1982. p. 148–9. Maitland GD. Peripheral manipulation. 3rd ed. Oxford: Butterworth-Heinemann; 1991. p. 183–9. Mo¨ller MH, Oberg BE, Gillquist J. Stretching exercise and soccer: effect of stretching on range of motion in the lower extremity in connection with soccer training. International Journal of Sports Medicine 1985;6:50–2. Morais Filho MC, de Godoy W, Santos CA. Effects of intramuscular psoas lengthening on pelvic motion in patients with spastic diparetic cerebral palsy. Journal of Pediatric Orthopaedics 2006;26:260–4. Muraki T, Aoki M, Uchiyama E, Murakami G, Miyamoto S. The effect of arm position on stretching of the supraspinatus, infraspinatus, and posterior portion of deltiod muscles: a cadaveric study. Clinical Biomechanics 2006;21:474–80. Neumann D. Kinesiology of the musculoskeletal system: foundations for physical rehabilitation. London: Mosby Elsevier; 2002. p. 397–400. Oatis CA. Kinesiology: the mechanics and pathomechanics of human movement. Philadelphia: Lippincott; 2004. p. 667–9. O’Malley AG. Correspondence and preliminary communications. Journal of Bone and Joint Surgery 1959;41B:888–9. Roach KE, Miles TP. Normal hip and knee active range of motion: the relationship to age. Physical Therapy 1991;71(9):656–65. Saito H, Nagasaki H. In: Nakamura R, editor. Clinical kinesiology. 3rd ed. Tokyo: Ishiyaku Ltd; 2002. p. 455–6 [in Japanese]. Sage FP. Cerebral palsy. In: Crenshaw AA, editor. Campbell’s operative orthopaedics. 7th ed. St. Louis: Mosby; 1987. p. 2891–5. Ticker JB, Bigliani LU, Soslowsky LJ, Pawluk RJ, Flatow EL, Mow VC. Inferior glenohumeral ligament: goniometric and strain rate dependent properties. Journal of Shoulder and Elbow Surgery 1996;5:269–79. Tillan LJ, Hertling D. Properties of dense connective tissue and wound healing. In: Hertling D, Kessler RM, editors. Management of common musculoskeletal disorders. 3rd ed. Philadelphia: Lippincott; 1996. p. 8–21.
432
E. Hidaka et al. / Manual Therapy 14 (2009) 427–432
Urayama M, Itoi E, Hatakeyama Y, Pradhan RL, Sato K. Function of the 3 portions of the inferior glenohumeral ligament: a cadaveric study. Journal of Shoulder and Elbow Surgery 2001;10:589–94. Wainner RS. Passive versus active stretching of hip flexor muscles in subjects with limited hip extension: a randomized clinical trial. Physical Therapy 2004;84:800–7. Walker JM, Sue D, Mile-Elkousy N, Ford G, Trevelyan H. Active mobility of the extremities in older subjects. Physical Therapy 1984;64(6):919–23. Werner P. Locomotor system. In: Color atlas and textbook of human anatomy. 5th ed., vol. 1. Stuttgart, New York: George Thieme Verlag; 2004. p. 200–1. Woo S-LY, Orlando CA, Camp JF, Akeson WH. Effects of postmortem storage by freezing on ligament tensile behavior. Journal of Biomechanics 1986;10:399–404.
Woo SLY, An KN, Arnoczky SP, Wayne JS, Fithian DC. Anatomy, biology, and biomechanics of tendon, ligaments, and meniscus. In: Simon SR, editor. Orthopaedic basic science. Chicago: American Academy of Orthopaedic Surgeons; 1994. p. 45–74. Yamada H. In: Evans FG, editor. Strength of biological materials. Maryland: Williams and Wilkins; 1970. p. 248–80. Zakas A, Vergou A, Grammatikkopoulou NG, Zakas N, Sentelidis T, Vamvakoudis S. The effect of stretching during worming-up on the flexibility of junior handball players. Journal of Sports Medicine and Physical Fitness 2003;43:145–9. Zakas A, Grammatikkopoulou NG, Zakas N, Zahariadis P, Vamvakoudis E. The effect of active worm-up and stretching on the flexibility of adolescent soccer players. Journal of Sports Medicine and Physical Fitness 2006;46:57–61.
Manual Therapy 14 (2009) 433–438
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Original Article
Validity of the Neck Disability Index and Neck Pain and Disability Scale for measuring disability associated with chronic, non-traumatic neck pain Mark Chan Ci En a, Dean A. Clair b, Stephen J. Edmondston c, * a
Physiotherapy Department, Tan Tock Seng Hospital, Singapore Physiotherapy Department, Osborne Park Hospital, WA, Australia c School of Physiotherapy, Curtin University of Technology, GPO Box U1987, Perth, WA 6845, Australia b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 January 2008 Received in revised form 13 May 2008 Accepted 27 July 2008
The purpose of this study was to evaluate the construct and content validity of the Neck Disability Index (NDI) and the Neck Pain and Disability Scale (NPAD) in patients with chronic, non-traumatic neck pain. Twenty patients (mean age ¼ 64.5 years) completed a patient-specific questionnaire, the Problem Elicitation Technique (PET), followed by the NDI and NPAD. Content validity was assessed by comparing the items of the NDI and NPAD with problems identified from the PET. Construct validity of the fixed-item questionnaires was examined by establishing the correlation with each other, and with the PET score. Eleven common problems were identified by patients through the PET, of which six were included in the NDI and seven included in the NPAD. The NDI and NPAD scores were strongly correlated (r ¼ 0.86, p < 0.01), while the correlation between the PET and the fixed-item questionnaires was moderate (NDI: r ¼ 0.62, p < 0.01; NPAD: r ¼ 0.71, p < 0.01). Both the NDI and the NPAD include most of the functional problems common to this patient group, and display good content validity. The PET is better able to evaluate the problems specific to the individual patient and is therefore measuring a somewhat different construct to the fixed-item questionnaires. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Neck pain Disability Outcome assessment Questionnaires
1. Introduction Neck pain has a lifetime prevalence of about 70% in the general population (Makela et al., 1991; Bovim et al., 1994). Although acute neck pain often resolves, about 19% of the population may suffer from chronic neck pain at any given time (Bovim et al., 1994; Guez et al., 2002). Measurement of the impact of neck pain on the sufferer presents a challenge due to the variability between patients in pain intensity, and the effect of the disorder on physical and psychological functions (Clair et al., 2004). Measures of pain intensity and tissue sensitivity have been used to quantify the sensory dimension of neck pain disorders (Hubka and Phelan, 1994; Olson et al., 2000), while range of motion and muscle function has been used to measure impairments of physical function (Falla et al., 2004; Jull et al., 2004; Hoving et al., 2005; O’Leary et al., 2007). However, recent recommendations place greater emphasis on functional status and quality of life more broadly, in the evaluation of neck pain disorders (Philadelphia Panel, 2001; Pietrobon et al., 2002). Measurement of function has been a developing theme in neck pain research as this shifts the focus away from signs and symptoms towards the specific effects of the symptoms on patient * Corresponding author. Tel.: þ61 8 9266 3665; fax: þ61 8 9266 3699. E-mail address:
[email protected] (S.J. Edmondston). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.07.005
function (Pietrobon et al., 2002). In relation to neck pain, this includes neck function, physical function more generally, and psychological function. A range of neck pain-specific questionnaires have been developed for this purpose, and have been incorporated into recent clinical studies (Vernon and Mior, 1991; Leak et al., 1994; Jordan et al., 1998; Westaway et al., 1998; Wheeler et al., 1999). The value of questionnaires is dependent on a range of factors but of primary importance is the validity, particularly in relation to construct and content. A recent review of neck pain-specific questionnaires concluded that most have not been extensively validated, and recommended a comparative study to better define the psychometric properties of the commonly used instruments (Pietrobon et al., 2002). The Neck Disability Index (NDI) is the most commonly used questionnaire for the measurement of neck pain disability. It was originally developed to evaluate the activities of daily living in patients with disabling neck pain, particularly that resulting from whiplash trauma (Vernon and Mior, 1991). The NDI includes 10 questions of which 7 examine functional activities, 2 ask about symptoms and the final question considers concentration. The Neck Pain and Disability Scale (NPAD) was developed to provide clinicians with a tool to assess the multi-dimensional effects of the neck pain disorder (Wheeler et al., 1999). The scale consists of 20 questions relating to 4 domains (neck function, pain intensity, emotion/ cognition and activities of daily living) which look at the effects of
434
M. Chan Ci En et al. / Manual Therapy 14 (2009) 433–438
the neck pain disorder on patients’ physical and emotional functions. The potential limitation of these questionnaires, and others with fixed questions, is that they constrain the scope of the evaluation to the specific issues included. Therefore, the questionnaire may include questions not relevant to some patients, and may not include issues of importance. An alternative to the fixed-item questionnaire are the patientspecific techniques which require patients to generate their own, possibly unique, set of problems or items. The patient-specific methods offer the advantage of identifying the problems or issues relevant to each individual, and are therefore consistent with the approach to patient evaluation commonly employed in clinical practice (Jolles et al., 2005). Two patient-specific techniques which have been used to evaluate neck pain are the Problem Elicitation Technique (PET) and the Patient-Specific Functional Scale (PSFS) (Buchbinder et al., 1995; Westaway et al., 1998). The disadvantage of this approach, particularly in research, is that without standardisation of content, the scale is different for each patient (Jolles et al., 2005). The level of statistical correlation between patientspecific scales and fixed-item questionnaires has been found to be only moderate (Westaway et al., 1998; Hoving et al., 2003). Clinical studies in which multiple neck pain questionnaires are applied simultaneously in the same patient population have been identified as an important focus for research in this area (Pietrobon et al., 2002). Specifically, patient-specific questionnaires will help identifying the problems which are most common and relevant to specific sub-groups of patients with neck pain. This may assist the development or modification of fixed-item questionnaires, and enhance the psychometric evaluation, particularly the content and construct validity. Hoving et al. (2003) used the PET to evaluate the validity of the NDI and Northwick Park Neck Pain questionnaire in patients with neck pain associated with whiplash injury. They found that the two fixed-item questionnaires did not fully cover the problems considered important in patients with whiplash injury, especially those concerning emotional and social functions. More recently, the NDI has been shown to have poor construct validity and to be less responsive to change than the PSFS, in patients with cervical radiculopathy (Cleland et al., 2006). Further, the NDI has been shown to be less responsive to change than previously reported in patients with non-traumatic neck pain (Cleland et al., 2008). These findings suggest that analysis of the psychometric properties of the NDI and other fixed-item questionnaires in different groups of patients with neck pain should be an on-going process. To date there has been little evaluation of the common problems associated with chronic, non-traumatic neck pain, particularly in older patients. Recent studies of this patient group have examined patient variability and treatment dose (Clair et al., 2004; Clair et al., 2006), but an evaluation of the common functional problems and validity of fixed-item questionnaires has not been conducted. The purpose of this study was to examine the content and construct validity of two fixed-item questionnaires as measures of disability in patients with chronic, non-traumatic neck pain. This was achieved through a comparison of the NDI and NAPD questionnaires in a cohort of patients with this disorder, and comparison of the responses to the fixed-item questionnaires with a patient-specific questionnaire, the PET. 2. Methods 2.1. Study population A cross-sectional survey of 20 subjects with chronic, non-traumatic neck pain was performed. Subjects were recruited from two public hospitals in Perth, Western Australia. Nineteen subjects were referred to the physiotherapy department for treatment of chronic,
non-traumatic neck pain. The remaining subject was a physiotherapist who had a long-standing history of non-traumatic neck pain. All subjects had a current episode of neck pain of greater than three months duration, with pain predominantly located in the somatic referral zones of the cervical spine (Grubb and Kelly, 2000). Participants were excluded if the symptom duration did not exceed three months, or were unable to read or comprehend the questionnaires. Subjects with a specific diagnosis such as cervical radiculopathy, ankylosing spondylitis and rheumatoid arthritis, or who had a history of neck trauma were not included in the study. The institutional Human Research Ethics Committee granted approval for this study. 2.2. Procedure Upon consenting to be involved in the study, participants were asked general questions in relation to age, symptom duration, current medication intake, and average pain intensity over the previous week. The average pain intensity was measured using a 10 cm visual analogue scale (VAS). The PET was firstly performed for all subjects, and was administered by a skilled interviewer. After PET, all participants completed the NDI and NPAD in a random order. 2.3. Patient questionnaires The PET is a disability questionnaire designed to help clinicians to identify the most significant problems experienced by each individual patient (Bakker et al., 1995; Buchbinder et al., 1995). A skilled interviewer carries out the interview where subjects are allowed to spontaneously identify problems associated with their neck pain disorder. The interviewer then assists the subject with a series of open-ended questions, which cover areas such as selfcare and work, mobility, leisure activities, social activities, emotion, communication and sleep. Subjects are allowed to identify a maximum of 15 problems. The subject is then asked to rate the severity of each problem on an 11-point numerical scale (no severity ¼ 0 to extremely severe ¼ 10). Finally, the subject ranks the problems according to their importance, from the most to least significant. In this study, the overall PET score was defined as the sum of the severity scores for all problems divided by the number of problems identified. The PET score had a possible range between 0 and 10. The NDI is a 10-item questionnaire which asks patients about their symptoms and the effect of their neck pain on a range of functional activities (Vernon and Mior, 1991). The items in the questionnaire are pain intensity, personal care, lifting, reading, headache, concentration, work, driving, sleeping and recreation. The subject is instructed to circle one of the six options which describes the severity of each item (0–5). The NDI score is calculated as the sum of the scores for each question multiplied by two (range ¼ 0–100). A higher score is indicative of greater disability associated with the neck disorder (Vernon and Mior, 1991). The NPAD is a 20-item questionnaire which examines the effects of the neck pain disorder and covers four factors of neck function, pain intensity, cognitive and emotional affects and activities of daily living (Wheeler et al., 1999). The subject responds to each question on a 10 cm visual analogue scale where the subject indicates the severity or frequency of each item. Each question was measured using a ruler to provide each individual score. The total NPAD score is the sum of the scores of all 20 questions divided by 2, where the maximum score is 100 and the minimum is 0. Like the NDI, a higher score will indicate greater disability. The NPAD has the potential advantage as the four factors examine the various aspects of the effects of the neck pain disorder on patients’ physical, cognitive or emotional function (Goolkasian et al., 2002; Clair et al., 2004).
M. Chan Ci En et al. / Manual Therapy 14 (2009) 433–438
2.4. Statistical analyses
Table 2 Disability profile of the study population according to the PET dimensions.
The content validity of the NDI and NPAD was studied by comparing the items in the questionnaires with the problems identified by the PET. The frequency of each problem and ranking by importance were determined. The construct validity of the NDI and NPAD was examined by determining the association between the questionnaire scores and the PET score, using Pearson’s correlation coefficients. The correlation between the NDI and NPAD scores was also examined. 3. Results Twenty subjects (7 males, 13 females) with a mean age of 64.5 years (SD ¼ 12.8) were recruited over a three-month period. The mean symptom duration was 115.6 months (SD ¼ 119.5). The characteristics of the study population are summarised in Table 1. The average pain intensity over the week prior to the interview was 5.2 (SD ¼ 1.9). The mean NDI and NPAD scores were 33.1 (SD ¼ 17.2) and 47.7 (SD ¼ 23.7), respectively. The mean PET score was 6.6 (SD ¼ 1.7). The PET identified an average of 9.2 problems per patient (SD ¼ 3.7). 3.1. Content validity Table 2 presents the disability profile of the study population according to the PET dimensions. Over 80% of the patients identified one or more problems within the dimensions of sleep (90.0%), mobility (90.0%) and role activity (85.0%). There were 23, 30 and 53 individual problems identified within these three dimensions, respectively. More than half of the subjects identified problems within the dimensions of emotion (75.0%) and symptoms (75.0%). The individual patient problems identified by the PET and their mean severity are presented in Table 3. Only problems identified by two or more patients are presented. Of the individual problems identified by the PET, sleep disturbance had the highest prevalence (75.0%). More than half of the subjects identified frustration (65.0%), driving (60.0%) and lifting (60.0%). Other common problems (more than 33%) were looking into cupboards (45.0%), gardening (40.0%), headaches (40.0%), housework (35.0%), working overhead (35.0%), and general exercise (35.0%). The highest mean
Role activity Emotional Sleep Mobility Sport and leisure Social activity Personal care Communication Symptoms
Gender (%) Male Female Symptom duration (%) 3–6 months 6–12 months 12–24 months >24 months Medication (%) None Analgesics only Analgesics and NSAIDS Anti-depressants Outcome measures [mean (SD)] Average pain intensity over past week NDI percentage score (0–100) NPAD percentage score (0–100) Overall PET score (0–10) Mean number of problems elicited
64.5 12.8 22.0–83.0
(15.0) (5.0) (15.0) (65.0)
4 6 9 1
(20.0) (30.0) (45.0) (5.0)
5.2 33.1 47.7 6.6 9.2
(1.9) (17.2) (23.7) (1.7) (3.7)
Overall number of problems identified in each dimension
Number of different problems within each dimension
Mean severity per dimension (possible range 0–10)
17 15 18 18 6
(85.0) (75.0) (90.0) (90.0) (30.0)
53 19 23 30 6
12 4 2 5 1
6.8 7.1 7.1 6.8 6.5
10 8 10 15
(50.0) (40.0) (50.0) (75.0)
10 10 11 19
3 2 3 3
6.9 6.5 6.9 7.4
severity scores were found in depression (9.3), cooking (8.3), and sitting upright (8.0). The individual problems that were ranked most important by the subjects were driving (30.0%), sleep disturbance (30.0%) and frustration (20.0%). Of the 11 problems identified by most subjects, 6 of these are included in the NDI and 7 are included in the NPAD (Table 4). 3.2. Construct validity The PET was moderately correlated with the NDI (r ¼ 0.62, p < 0.01) and NPAD (r ¼ 0.71, p < 0.01). There was a stronger correlation between the NDI and NPAD (r ¼ 0.86, p < 0.01). The Table 3 Problems identified by patients using the PET, and the mean severity and ranking of importance. Dimension
Problem
Role Activity
Looking into cupboards Gardening Housework Working overhead Hanging up washing Work for wages Vacuuming Cooking
Number Mean of subjects (SD) (%) severity (0–10) 9 8 7 7 6 4 4 4
Number of subjects ranking problem as most important (%)
(45) (40) (35) (35) (30) (20) (20) (20)
5.3 6.6 7.7 6.6 7.8 6.5 7.0 8.3
0 0 3 0 2 3 1 1
(0) (15) (10) (15) (5) (5)
Emotional
Frustration Depression
13 (65) 4 (20)
6.5 9.3
4 (25) 0
Sleep
Affected sleep Fatigue throughout day
15 (75) 8 (40)
7.1 7.1
6 (30) 0
Mobility
Driving Lifting Crossing the road
12 (60) 12 (60) 3 (15)
6.2 7.6 7.0
6 (30) 2 (10) 1 (5)
Social Activities
Socialising with friends
6 (30)
6.5
0
Sports and leisure
General exercise Non athletic leisure activities
7 (35) 2 (10)
7.3 7.0
0 0
Personal Care
Dressing Hair care
6 (30) 4 (20)
6.0 7.3
2 (10) 0
Communication Computer use Reading
6 (30) 4 (20)
6.3 7.3
1 (5) 1 (5)
Symptoms
8 (40) 6 (30) 5 (25)
7.6 7.8 6.6
2 (10) 1 (5) 0
7 (35.0) 13 (65.0) 3 1 3 13
Number of patients who identified problems within each dimension (%)
Table shows the number of problems, the number of patients who identified problems within each dimension, and the mean severity of problems per dimension.
Table 1 Demographic and clinical characteristics of the study population (n ¼ 20). Age (years) Mean SD Range
435
Headaches Concentration Neck movements
436
M. Chan Ci En et al. / Manual Therapy 14 (2009) 433–438
Table 4 Comparison of commonly identified problems elicited by the PET with items in NDI and NPAD. Identified problem
Item included in NDI
Item included in NPAD
Affected sleep Frustration Driving Lifting Looking into cupboards Gardening Headaches Fatigue throughout day Housework Working overhead
Yes
Yes Yes Yes
Yes Yes
Yes Yes Yes
Yes Yes
association between the three questionnaires is presented graphically in Fig. 1. 4. Discussion In the development of neck pain questionnaires, assumptions were made as to the nature of the functional limitations associated with neck pain. However, patient input is considered paramount in the development and evaluation of an outcome measure (Guyatt et al., 1993). While there was some input from patients in the
construction of the NDI and NPAD, some questionnaires have been developed with little or no input from patients with neck pain (Pietrobon et al., 2002). Establishing the validity of a questionnaire is important to ensure that it reflects the nature and spectrum of the problems experienced by the majority of patients. The validity of the NDI and NPAD has been investigated in patients with whiplash-related neck pain (Goolkasian et al., 2002; Hoving et al., 2003), but this is the first study to specifically examine the validity of two commonly used disability questionnaires in an older cohort of patients with chronic, non-traumatic neck pain. The PET simulates clinical practice by asking patients to identify physical, emotional and cognitive problems specific to their neck pain disorder. It serves to elicit the problems specific to each patient, thereby reducing the ‘noise’ created when items not relevant to the patient are included (Buchbinder et al., 1995; Jolles et al., 2005). The most commonly reported functional problems in the present study were disturbed sleep, driving, and lifting, while frustration was the most common emotional problem. In patients with whiplash, who were significantly younger than the patients in the current study, the most common functional problems identified were work for wages, fatigue during the day, participation in sports, and driving, while the most common emotional problem was depression (Hoving et al., 2003). Driving or riding in a car was the only common functional problem experienced by both patient
80
100 r = 0.71
r = 0.86 80
NPAD Score
40
60
40
20 20
0
0 0
20
40
60
80
100
3
4
5
NPAD Score
6
7
PET Score
80 r = 0.62
60
NDI Score
NDI Score
60
40
20
0 3
4
5
6
7
8
9
PET Score Fig. 1. Scatter plots showing the relationships between total scores for the NDI, NPAD and PET questionnaires.
8
9
M. Chan Ci En et al. / Manual Therapy 14 (2009) 433–438
groups. This finding suggests that the impact of neck pain on physical and emotional functions may be somewhat different in older patients with non-traumatic neck pain compared to younger patients who have whiplash-related neck pain. This is consistent with the studies which have reported differences in the nature and severity of physical impairments in patients with neck pain of traumatic origin, compared to those with a non-traumatic onset (Dumas et al., 2001; Drottning et al., 2002). Of the 10 most common problems identified by the PET, 6 were included in the NDI and 7 were included in the NPAD, which supports the content validity of both questionnaires for this patient population. Sleep disturbance, driving and frustration were ranked as the three most commonly reported problems. All are included in the NPAD, while frustration is not addressed in the NDI. A significant impact on psychological function has been described in patients with chronic, non-traumatic neck pain, which was found to improve with improvements in pain intensity and functional limitation (Clair et al., 2006). The NPAD differs from the NDI in that it includes questions which relate specifically to emotion and social function. The NPAD uses sub-domains to identify more specifically the areas of physical and psychological functions most commonly indicated by the patient as being affected. While most of the common functional problems relevant to this patient group are included in the NDI and NPAD, the greater scope of the latter questionnaire may provide better information about the impact of the disorder on the patient more broadly. The results of the present study suggest that neck pain questionnaires should have a greater emphasis on neck function, activities of daily living and psychological function, and limited emphasis on symptoms such as pain intensity, tissue tenderness and movement restrictions which can be measured in other ways. The high correlation between the NDI and NPAD scores suggests that they measure the same construct in this patient group. While the questionnaire format and scoring systems are different, the questionnaires address a range of common items, which relate to function rather than symptoms. Previous studies have shown high correlation between the NDI and other fixed-item neck pain questionnaires, where the items in the questionnaires were very similar (Hoving et al., 2003; Wlodyka-Demaille et al., 2004; Gay et al., 2007). The only previous study to directly compare the NDI and NPAD found a moderate correlation (r ¼ 0.72) between questionnaires, in younger patients with neck pain of both traumatic and atraumatic origins (Goolkasian et al., 2002). Consistent with the study of Hoving et al. (2003), there was a moderate correlation between the PET and both fixed-item questionnaires, which suggests that the PET measures a somewhat different construct. This may be due to the PET only scoring items of relevance to each patient. For this reason, the PET reflects clinical practice as it identifies problems relevant to the individual, which may include issues not addressed in fixed-item questionnaires. A recommendation based on the results of the present study is that the PET should be used in conjunction with a fixed-item questionnaire, as each provides different information about the study population. While fixed-item questionnaires are relatively simple to administer, the PET requires some training and experience of the interviewer to ensure consistency in its application and adherence to the target concept (Jolles et al., 2005). During the study the relevance of the driving item in the NDI was raised by some patients, as many of the subjects did not drive, either due to their age or the neck pain disorder. The applicability of the driving item in the NDI to non-drivers has not previously been considered and may be an area for review. Previous studies have chosen to modify the NDI in an attempt to improve the relevance of the questionnaire to the specific study population (Hains, 1998; Riddle and Stratford, 1998). In the present study, the item was answered by either a driver or a passenger, consistent with the driving question in the NPAD. This modification to the NDI should
437
be considered if the questionnaire is used in future studies with this patient group. A potential limitation of this study is the relatively small study population. However, the characteristics of the patients were consistent with those of larger studies of patients with chronic, non-traumatic neck pain (Clair et al., 2004, 2006). The higher proportion of women in this study (65%) is consistent with gender ratios in other neck pain studies where the proportion of female subjects has been between 60 and 70% (Guez et al., 2003; Clair et al., 2004; Gay et al., 2007). The age and symptom duration suggest that degenerative pathology and the effects of aging may be important in the development of the symptoms, however, a review of radiological examinations and correlation with symptoms were not part of the present study. The patients in the study were receiving treatment in a public health system, and the problems identified may not reflect those of all patients with non-traumatic neck pain in the broader community. In conclusion, the NDI and NPAD both identified the common problems considered important by the patients, and the NPAD included all problems ranked as most important. Both questionnaires have good content validity and are therefore equally relevant for use in this patient group. The broader scope of the NPAD, particularly in relation to emotional and social functions, may be an advantage in future studies. In future research involving chronic, non-traumatic neck pain, it is recommended that a patient-specific questionnaire such as the PET should be used in conjunction with a fixed-item neck pain questionnaire, as each seems to measure a somewhat different construct. Acknowledgements The authors acknowledge the support of the staff of the Physiotherapy Outpatient Department, Osborne Park Hospital, Perth, Western Australia, for their assistance in recruitment of subjects for this study. References Bakker C, van der Linden S, van Santen-Hoeufft M, Bolwijn P, Hidding A. Problem elicitation to assess patient priorities in ankylosing spondylitis and fibromyalgia. Journal of Rheumatology 1995;22:1304–10. Bovim G, Schrader H, Sand T. Neck pain in the general population. Spine 1994;19:1307–9. Buchbinder R, Bombardier C, Yeung M, Tugwell P. Which outcome measures should be used in rheumatoid arthritis clinical trials? Arthritis & Rheumatism 1995;38:1568–80. Clair D, Edmondston S, Allison G. Variability in pain intensity, physical and psychological function in non-acute, non-traumatic neck pain. Physiotherapy Research International 2004;9:43–54. Clair DA, Edmondston SJ, Allison GT. Physical therapy treatment dose for nontraumatic neck pain: a comparison between 2 patient groups. Journal of Orthopaedic and Sports Physical Therapy 2006;36:867–75. Cleland JA, Childs JD, Whitman JM. Psychometric properties of the neck disability index and numeric pain rating scale in patients with mechanical neck pain. Archives of Physical Medicine and Rehabilitation 2008;89:69–74. Cleland JA, Fritz JM, Whitman JM, Palmer JA. The reliability and construct validity of the neck disability index and patient specific functional scale in patients with cervical radiculopathy. Spine 2006;31:598–602. Drottning M, Staff PH, Sjaastad O. Cervicogenic headache (CEH) after whiplash injury. Cephalalgia 2002;22:165–71. Dumas J-P, Arsenault AB, Boudreau G, Magnoux E, Lepage Y, Bellavance A, Loisel P. Physical impairments in cervicogenic headache: traumatic vs. non-traumatic onset. Cephalalgia 2001;21:884–93. Falla D, Bilenkij G, Jull G. Patients with chronic neck pain demonstrate altered patterns of muscle activation during performance of a functional upper limb task. Spine 2004;29:1436–40. Gay RE, Madson TJ, Cieslak KR. Comparison of the neck disability index and the Bournemouth neck questionnaire in a sample of patients with chronic uncomplicated neck pain. Journal of Manipulative and Physiological Therapeutics 2007;30:259–62. Goolkasian P, Wheeler AH, Gretz SS. The neck pain and disability scale: test–retest reliability and construct validity. Clinical Journal of Pain 2002;18:245–50. Grubb SA, Kelly CK. Cervical discography: clinical implications from 12 years of experience. Spine 2000;25:1382–9.
438
M. Chan Ci En et al. / Manual Therapy 14 (2009) 433–438
Guez M, Hildingsson C, Nilsson M, Toolanen G. The prevalence of neck pain. Acta Orthopaedica Scandinavica 2002;73:455–9. Guez M, Hildingsson C, Stegmayr B, Toolanen G. Chronic neck pain of traumatic and non-traumatic origin. Acta Orthopaedica Scandinavica 2003;74:576–9. Guyatt G, Feeney DH, Patrick DL. Measuring health-related quality of life. Annals of Internal Medicine 1993;118:622–9. Hains F. Psychometric properties of the neck disability index. Journal of Manipulative and Physiological Therapeutics 1998;21:75–80. Hoving JL, O’Leary EF, Niere KR, Green S, Buchbinder R. Validity of the neck disability index, Northwick Park neck pain questionnaire, and problem elicitation technique for measuring disability associated with whiplash-associated disorders. Pain 2003;102:273–81. Hoving JL, Pool JJ, van Mameren H, Deville WJ, Assendelft WJ, de Vet HC, de Winter AF, Koes BW, Bouter LM. Reproducibility of cervical range of motion in patients with neck pain. BMC Musculoskeletal Disorders 2005;6:59–67. Hubka MJ, Phelan SP. Inter-examiner reliability of palpation for cervical spine tenderness. Journal of Manipulative and Physiological Therapeutics 1994;17:591–5. Jolles BM, Buchbinder R, Beaton DE. A study compared nine patient-specific indices for musculoskeletal disorders. Journal of Clinical Epidemiology 2005;58:791–801. Jordan A, Manniche C, Mosdal C, Hindsberger C. The Copenhagen neck functional disability scale: a study of reliability and validity. Journal of Manipulative and Physiological Therapeutics 1998;21:520–7. 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. Leak AM, Cooper J, Dyer S, Williams KA, Turner-Stokes L, Frank AO. The Northwick Park neck pain questionnaire, devised to measure neck pain and disability. British Journal of Rheumatology 1994;33:469–74.
Makela M, Heliovarra M, Sievers K, Impivaara O, Knekt P, Aromaa A. Prevalence, determinants, and consequences of chronic pain in Finland. American Journal of Epidemiology 1991;124:1356–67. O’Leary S, Jull G, Kim M. Vicenzino B Cranio-cervical flexor muscle impairment at maximal, moderate, and low loads is a feature of neck pain. Manual Therapy 2007;12:34–9. Olson SL, O’Connor DP, Birmingham G, Broman P, Herrera L. Tender point sensitivity, range of motion and perceived disability in subjects with neck pain. Journal of Orthopaedic and Sports Physical Therapy 2000;30:13–20. Philadelphia Panel. Philadelphia Panel evidence-based clinical practice guidelines on selected rehabilitation interventions for neck pain. Physical Therapy 2001;81:1701–17. Pietrobon R, Coeytaux RR, Carey TS, Richardson WJ, DeVellis RF. Standard scales for measurement of functional outcome for cervical pain or dysfunction. Spine 2002;27:515–22. Riddle DL, Stratford PW. Use of generic versus region-specific functional status measures on patients with cervical spine disorders. Physical Therapy 1998;79:951–63. Vernon H, Mior S. The neck disability index: a study of reliability and validity. Journal of Manipulative and Physiological Therapeutics 1991;14:409–15. Westaway MD, Stratford PW, Binkley JM. The patient-specific functional scale: validation of its use in persons with neck dysfunction. Journal of Orthopaedic and Sports Physical Therapy 1998;27:331–8. Wheeler AH, Goolkasian P, Baird AC, Darden BV. Development of the neck pain and disability scale: item analysis, face, and criterion-related validity. Spine 1999;24:1290. Wlodyka-Demaille S, Poiraudeau S, Catanzariti J-F, Rannou F, Fermanian J, Revel M. The ability to change of three questionnaires for neck pain. Joint Bone Spine 2004;71:317–26.
Manual Therapy 14 (2009) 439–443
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Original Article
The use of osteopathic manipulative treatment as adjuvant therapy in patients with peripheral arterial disease Rita Lombardini a, *, Simona Marchesi a, Luca Collebrusco b, Gaetano Vaudo a, Leonella Pasqualini a, Giovanni Ciuffetti a, Matteo Brozzetti a, Graziana Lupattelli a, Elmo Mannarino a a b
Internal Medicine, Angiology and Atherosclerosis, University of Perugia, Italy Division of Physiotherapy, ‘‘S. Maria della Misericordia’’ Hospital, Perugia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history: Received 20 November 2007 Received in revised form 16 July 2008 Accepted 2 August 2008
Peripheral arterial disease (PAD) is a manifestation of systemic atherosclerosis associated with impaired endothelial function and intermittent claudication is the hallmark symptom. Hypothesizing that osteopathic manipulative treatment (OMT) may represent a non-pharmacological therapeutic option in PAD, we examined endothelial function and lifestyle modifications in 15 intermittent claudication patients receiving osteopathic treatment (OMT group) and 15 intermittent claudication patients matched for age, sex and medical treatment (control group). Compared to the control group, the OMT group had a significant increase in brachial flow-mediated vasodilation, ankle/brachial pressure index, treadmill testing and physical health component of life quality (all p < 0.05) from the beginning to the end of the study. At univariate analysis in the OMT group there was a negative correlation between changes in brachial flow-mediated vasodilation and IL-6 levels (r ¼ 0.30; p ¼ 0.04) and a positive one between claudication pain time and physical function score (r ¼ 0.50; p ¼ 0.05). In conclusion, despite the relatively few patients in our study, these results suggest that OMT significantly improves endothelial function and functional performance in intermittent claudication patients along with benefits in quality of life. This novel treatment combined with drug and lifestyle modification might be an effective alternative to traditional training based on exercise. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Intermittent claudication Osteopathic manipulative treatment Endothelial function
1. Introduction Peripheral arterial disease (PAD) is a manifestation of systemic atherosclerosis. Intermittent claudication, the predominant clinical symptom, is characterized by cramping, aching or fatigue, which typically involves the calf muscles, thighs and buttocks. PAD is associated with impaired endothelium-dependent vasodilation of peripheral vessels (Weiss et al., 2002) and by an increased expression of adhesion molecules and release of proinflammatory circulating molecules (Fiotti et al., 1999; Brevetti et al., 2001a). In patients with mild to moderate symptoms (Fontaine stage II; European Working Group on Critical Leg Ischemia, 1991) conservative treatment improves endothelial function and peripheral circulation. Results of functional tests improve and quality of life is better (Shepard and Balady, 1999; Brendle et al., 2001). Conservative treatment includes dietary and pharmacological risk factor
* Corresponding author. Medicina Interna, Angiologia e Malattie da Arteriosclerosi, Universita` di Perugia, Ospedale ‘‘S. Maria della Misericordia’’, Loc. S. Andrea delle Fratte, 06156 Perugia, Italy. E-mail address:
[email protected] (R. Lombardini). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.08.002
modification, mild invasive strategies such as spinal cord stimulation (SCS) and exercise therapy. Risk factor modification (lipid lowering drugs, glycaemic control, antihypertensive and antiplatelet treatment), is recognized to be beneficial. SCS is a neuromodulation technique using electricity which has shown beneficial impacts on quality of life in patients with ischemic heart disease and PAD (Ubbink et al., 2004; De Vries et al., 2007). Exercise training is a critical and debated part of PAD patient treatment. It is often plain advice to ‘‘walk more’’ rather than follow a closely supervised exercise programme. In fact, there is no consensus on the most effective form of exercise therapy. The kind and frequency of exercise, level of supervision and improvement scales have not yet been fully defined, and few trials have addressed these specific points (Bartelink et al., 2004). Several recent studies showed exercise training confers benefits (Gardner and Poehlman, 1995; Collins et al., 2007), but could enhance endothelial injury (Hickman et al., 1994; Brevetti et al., 2001b). During rest intervals in an exercise programme claudicant patients suffered repeated episodes of leg muscle ischaemia- and reperfusion injury which led to development of a systemic inflammatory response that was hypothesized to injure the endothelium, accelerate atherosclerosis and increase thrombotic risk
440
R. Lombardini et al. / Manual Therapy 14 (2009) 439–443
(Silvestro et al., 2002; Burns et al., 2003). Given these findings, along with few manageable supervised exercise programmes and major financial and logistical obstacles to improving exercise programmes, we turned our attention to a more cost-effective approach. Osteopathic manipulative therapy (OMT) is one of the hallmarks of osteopathic medicine, which was founded in 1874 by Andrew Taylor Still (Still, 1992). Osteopathic medicine focuses on the union of all body components and identifies the muscoloskeletal system as a key to health. In osteopathic manipulation, bones, muscles, and tendons are manipulated to promote blood flow through tissues and thus enhance the body’s own healing powers (DiGiovanna et al., 1997). The osteopathic manipulative techniques used may allow for a normalization of imbalances between the sympathetic and parasympathetic nervous systems and improved vascular motion which would result in a more balanced homeostatic mechanism. A recent report by Salamon et al. (2004) used their own findings in the area of nitric oxide (NO) research to explain the therapeutic vascular effects of OMT. Until recently, there was little scientific evidence to support these claims. Despite reports that OMT is successful in low back pain and in patients with cardiovascular diseases, including hypertension (Rogers and Rogers, 1976; Spiegel et al., 2003), there are no reports of its use in PAD. The present pilot study investigated whether OMT, when combined with lifestyle modifications and pharmacological therapy, could be of benefit to patients with intermittent claudication. 2. Patients and methods The study was carried out from January 2005 to February 2006. Consecutive eligible male patients with PAD were recruited from those attending the Unit of Internal Medicine, Angiology and Atherosclerosis at the University of Perugia – ‘‘Santa Maria della Misericordia’’ Hospital (Perugia, Italy). PAD was diagnosed by highresolution ultrasound screening of femoral tracts (HDI 3500, Advanced Technology Laboratories, USA, which is equipped with a linear multifrequency 5–12 MHz transducer) and Doppler velocimetry (including the treadmill test) and measurement of ankle/ brachial pressure index (ABPI) 0.05). Fig. 3 illustrates mean values (SEM) for each successive repetition across trial 1. Although between subject variability is considerable (mean SEM ¼ 3.23), within subject variability was small across the ten repetitions (mean SEM ¼ 1.70). Statistical analysis was performed on the data to determine the triad with the least variability across trial one. Repeated measures ANOVA for repetitions 1–3, 2–4, 3–5, 4–6, etc. showed that repetitions 2–4 had the smallest effect size and least variability (partial Eta squared 0.005 at p ¼ 0.95).
Fig. 3. The mean value for the group for each successive repetition across trial 1.
Fig. 4. Bland–Altman plots for within and between day comparisons.
N.R. Heneghan et al. / Manual Therapy 14 (2009) 452–455
sensors, the preferred approach for motion analysis of this region in vivo (Willems and Jull, 1996; Theodoridis and Ruston, 2002; Edmonston et al., 2006), possibly lack accuracy and reliability due to relative movement between the sensor, skin and bone, despite reporting good stability (Willems and Jull, 1996; Edmonston et al., 2006). The stability of the measurements appeared relatively constant, despite some considerable variation between subjects. This may in part be due to the use of a standardised sitting posture resulting in some individuals performing axial rotation away from their ‘normal’ sitting posture. The experimental set up was used to ensure that subjects’ thoracic posture was consistent across repetitions and trials. Motion in other planes was recorded, however in this instance data was only analysed to determine axial rotation. Although minimal time was spent at the extremes of the range of motion stress relaxation may account for the slight trend for increase in range across each trial. Hysteresis may explain the lack of cumulative increase in range from trial 1 to trial 2 with tissues having a chance to ‘recover’ between trials. Whilst intra-tester reliability was found to be ‘excellent’ and ‘good to excellent’ for within and between day measures respectively, using the data from either the third repetition or the mean of the 2–4 repetitions, some caution should be taken when interpreting the results alongside the Bland–Altman plots. The Bland– Altman plots suggest that the difference between paired measures is up to 10% and 15% for within and between day agreements of measures respectively. Given that the mean range of motion was 85 degrees, this suggests that there could be an error as large as 8–10 degrees size for some subjects. Visual inspection of the graphs however suggests that the approach may be better suited to within day measures, where the percentage difference for the majority of subjects is less than 5%. Further research with a larger sample is indicated to explore the nature and extent of sources of error with this approach and perhaps using images from more than one spinal level. This is the first study that has utilised ultrasound imaging of the spine in dynamic and functional motion analysis as a means of ensuring that the start and end body positions/postures are truly representative of the underlying bony anatomy. As image acquisition of sufficient quality is operator dependent and the equipment reasonably expensive, there is currently little prospect for this as a clinical practice measurement tool. However, as our current understanding of biomechanics and effects of interventions in this region is considerably underdeveloped compared to other areas of the spine the need to consider alternative non-invasive measurement approaches remains. Future research could, with sufficient imaging expertise, explore regional
455
or segmental motion analysis using data from all three coordinates (x, y, z) or vertebral coupling which has in this region been debated. 4. Conclusion Ultrasound imaging combined with a motion analysis system has been shown to be a reliable method as a measure of active thoracic axial rotation in a seated position. Although the intratester reliability was shown to be ‘‘good to excellent’’ further work is indicated to explore the use of this approach in the evaluation biomechanics and clinical interventions in this region. Acknowledgments Alison Rushton, School of Health Sciences, University of Birmingham, Doug Carroll, School of Sport and Exercise Sciences, Chris Wright, School of Health Sciences, University of Birmingham. The Manipulative Association of Chartered Physiotherapists Doctoral Award 2007. References Bland CJ, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;327:307–10. Burwell RG, Kirby AS, Aujla RK, Krik EL, Pratt RK, Bailey MA, et al. Evaluation of vertebral rotation by ultrasound for the early detection of adolescent idiopathic scoliosis. In: Research into Spinal Deformities 2. Health Technology & Information 59. Amsterdam: IOS Press; 1999. Edmonston S, Aggerholm M, Elfving S, Flores N, Ng C, Smith R, et al. Influence of posture on the range of axial rotation and coupled lateral flexion of the thoracic spine. Journal of Manipulative and Physiological Therapeutics 2006;30(3): 193–9. Koerhuis CL, Winters JC, van der Helm FCT, Hof AL. Neck mobility measurement by means of the Flock of Birds electromagnetic tracking system. Clinical Biomechanics 2003;18:14–8. Kirby A, Burwell R, Cole A, Pratt R, Webb J, Moulton A. Evaluation of a new real-time ultrasound method for measuring segmental rotation of vertebrae and ribs in scoliosis. In: Research into Spinal Deformities 2. Health Technology & Information 59. Amsterdam: IOS Press; 1999. Polhemus Specifications. Available at http://www.polhemus.com/?page¼Motion_ Liberty (accessed November 22nd, 2007). Sim J, Wright C. (2000) Research in Health Care. Concepts, Designs, and Methods. Gloucester, UK: Stanley Thornes. Shrout PE. Measurement reliability and agreement in psychiatry. Statistical Methods in Medical Research 1998;7:301–17. Suzuki S, Yamamuro T, Shikata J, Shimizu K, Iida H. (1989) Ultrasound measurement of vertebral rotation in idiopathic scoliosis. Journal of Bone and Joint Surgery 71-B(2):252–5. Theodoridis D, Ruston S. The effect of shoulder movements on thoracic spine 3D motion. Clinical Biomechanics 2002;17:418–21. Walter SD, Eliasziw M, Donner A. Sample size and optimal designs for reliability studies. Statistics in Medicine 1998;17:101–10. Willems JM, Jull GA. An in vivo study of the primary and coupled rotations of the thoracic spine. Clinical Biomechanics 1996;11(6):311–6.
Manual Therapy 14 (2009) 456–459
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Professional Issue
Bibliometrics, impact factors and manual therapy: Balancing the science and the art Derek R. Smith*, Darren A. Rivett School of Health Sciences, Faculty of Health, The University of Newcastle, Australia
a r t i c l e i n f o Article history: Received 8 November 2008 Accepted 28 November 2008 Keywords: Impact factors Publishing Manual therapy Physiotherapy Bibliometrics Rehabilitation
Bibliometrics can be defined as a field of research that examines bodies of knowledge both within and across disciplines (Holden et al., 2005). Although many methods are commonly used, perhaps the most widely known is citation analysis; that is, tracking published articles to see whether they are subsequently cited by others (Smith, 2008a). Much of contemporary bibliometrics can be traced back to a seminal publication known as Shepard’s Citations, a tool first used by American lawyers in 1873 to establish whether a previous legal judgment had been referred to, overruled, or made invalid in some other way (Adair, 1955). By the early 20th century, citation analysis had attracted the attention of various scientific scholars, although most of their early work simply involved the counting and sorting of reference lists. Nevertheless, some trends were noticed early on in the journals of chemistry (Gross and Gross, 1927), engineering (Bradford, 1934) and physiology (Brodman, 1944). Perhaps the most striking observation was that not all journals were being equally cited; rather, only a few core periodicals appeared to be attracting the majority of all citations. On the other hand, while larger journals tended to gather more citations than smaller ones, some of the smaller periodicals still appeared to be performing well, relative to their actual size and circulation.
* Correspondence to: Derek R. Smith, WorkCover New South Wales Research Centre of Excellence, School of Health Sciences, Faculty of Health, University of Newcastle, Ourimbah, New South Wales 2258, Australia. Tel.: þ61 2 4348 4021; fax: þ61 2 4348 4013. E-mail address:
[email protected] (D.R. Smith). 1356-689X/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.11.004
With this phenomenon in mind, an information scientist named Eugene Garfield proposed calculating a journal’s relative ‘impact’ in 1955 (Garfield, 1955), whereby the number of citations received by a particular journal in a particular time period could be divided by the number of articles it had actually published during that time. The concept was refined with experience to include only ‘citable items’ in the calculation, that is, substantial types of articles that were most likely to be cited by others (Garfield, 1986). Garfield then founded a company known as the Institute for Scientific Information (ISI), and began publishing impact factors in the yearly Journal Citation ReportsÒ (JCRÒ) during the early 1970s (Smith, 2008c). Data from ISI-listed journals was initially collated in the Science Citation IndexÒ (SCIÒ), although by 1972 the scope of material had expanded to also encompass the social sciences, which led to formation of the Social Sciences Citation IndexÒ (SSCIÒ) (Garfield, 1972). Garfield’s idea caught on over time, the ISI was acquired by Thomson Scientific (later Thomson Reuters), and impact factors now represent a powerful influence in the world of modern publishing (Smith, 2006). Given that the impact factor calculation is fundamentally based on citation counts, citations themselves have now risen to become the ‘currency’ of modern scientific research (Joseph, 2003). Impact factors tend to change over time and are generally believed to be increasing in recent years. Such trends have been quantitatively demonstrated in both the larger medical journals (Chew et al., 2007) and also in some of the smaller health (Smith, 2008b) and medical (Boldt et al., 2000) sub-disciplines. There are a few potential reasons for this phenomenon. Firstly and fundamentally, although the basic calculation itself has not changed for
D.R. Smith, D.A. Rivett / Manual Therapy 14 (2009) 456–459
over 50 years (i.e. citations received divided by articles published), the actual number of citations being made each year is steadily increasing. This is probably due to various factors, such as an increased use of automatic referencing software, thereby making it easier to include a larger number of references in a journal article than before (Smith and Hazelton, 2008). This may reflect an increasing tendency for authors to cite their peers whenever possible, and thus increase the chances of their article being accepted. There is also increasing pressure on authors to include more references per article to help demonstrate a more comprehensive understanding of the topic. Secondly, journal impact factors are now being increasingly influenced by deliberate editorial practices in recent years, as more and more journals learn to play the ‘impact factor game’ (Tse, 2008). Citation analysis represents a subset of bibliometric research with which most researchers and academics are becoming increasingly familiar. Although relevancy of a published article is highly desirable, exactly how well it will achieve this goal is not predictable at the time of publication (Balon, 2005). Given time, however, citation trends will emerge and many scholars now consult the various electronic databases to see where and when their articles have been cited, and by whom. There are three main types of scientific article; those that present data, those that teach, and those which analyze, speculate or comment (D’Auria, 1999). Literature reviews tend to attract more citations than original articles, which themselves tend to attract more citations than editorials or letters. Even so, all forms of academic scholarship are important for clinicians in manual therapy, as evidenced in this journal. Although clinical experience may suggest that various kinds of physical therapies are worthwhile, evaluations still need to be based on research findings, especially empirical clinical research (Michels, 1982). Academic publishing is not just for scientists. Indeed, ten years ago D’Auria suggested that scholarship itself is the ‘intellectual counterpart of manual dexterity in clinical work’ (D’Auria, 1999, p. 277). Some of the earliest citation analysis in physical therapy journals appears to have been conducted around 25 years ago. Although it was not citation analysis as such, in 1982 Michels explored the issue of research evaluation in physical therapy, asserting the importance of a sound research base for clinical practice (Michels, 1982). In one of the first bibliometric investigations in our field, Dean and Davies (Dean and Davies, 1986) investigated the frequency of citations combined with a ‘Reputational Assessment’ of contributors in physical therapy. In their article, the authors performed a citation analysis of the journals Physical Therapy and Physiotherapy Canada between 1981 and 1982, concluding that the perception of therapists regarding the impact of eminent individuals was comparable to ratings of those individuals made by objective measures, such as citation analysis. Five individuals who were nominated as being ‘eminent’ in the profession also appeared regularly in the citation lists of these two journals. In the same year, 1986, Bohannon and Gibson published their analysis of journals cited in Physical Therapy, finding that Physical Therapy itself was the most highly-cited periodical, followed by the Archives of Physical Medicine and Rehabilitation (Bohannon and Gibson, 1986). Interestingly, the second ranked journal on the list had attracted less than half the number of citations as the first. In 1987, Bohannon proposed one of the first ‘core’ lists of physiotherapy journals by examining citation frequency in Physical Therapy, Physiotherapy, Physiotherapy Canada and Physiotherapy Practice (Bohannon, 1987). In 1989, Bohannon and Tiberio examined the medical index coverage of journals cited in physiotherapy periodicals, finding that it was neither complete nor consistent. The following year, 1990, an article looking at information accessing
457
behavior of physical therapists was published by Bohannon, who established that while books and journals were certainly being read, physical therapists also utilized patient protocols, medical communications, course notes and materials from company representatives (Bohannon, 1990). In 1991, another citation analysis was conducted by Bohannon and Roberts to help establish a ‘core list’ of rehabilitation journals, during which the authors found that information relating to rehabilitation was actually being published across a large number of different journals (Bohannon and Roberts, 1991). In 1992, Roberts conducted a study of the journal literature and its quality in relation to physiotherapy. His results suggested that while Medline was an excellent source of supplementary material relevant to physiotherapy, coverage was not complete and other information sources also needed to be consulted (Roberts, 1992b). In the following year, Roberts then looked at the coverage of core journals in rehabilitation and related topics by various online databases. In his study, it was revealed that although the number of core journals was very large, their actual coverage by information services was still very selective (Roberts, 1992a). In the same year, Kuhlemeier published a bibliometric analysis of the Archives of Physical Medicine and Rehabilitation, reporting that although it sat near the top of impact factor rankings for rehabilitation periodicals, its score was lower than for most medical journals (Kuhlemeier, 1992). In 1995, Tesio and colleagues investigated, from a bibliometric and citationist perspective, the drive of the neurology profession toward rehabilitation. In their study, the authors found that rehabilitation literature suffered from having a relatively small number of articles published, having a greater proportion of the literature being published in journals without impact factors, and having lower impact factors even when published in ISI-listed periodicals (Tesio et al., 1995). In a 1997 article, the literature of physical therapy was ‘mapped’ by examining citations in two established journals, Physical Therapy and the Archives of Physical Medicine and Rehabilitation (Wakiji, 1997). A skewed distribution of citations was clearly demonstrated, with only 14 journals being responsible for one-third of all references, whereas the next one-third came from 95 other journals. In 1999, Bohannon proposed another list of core physiotherapy journals established by means of citation analysis. In his study, the author looked at 5534 citations from 973 journals, finding that 48 journals had been cited 20 times or more in the time period 1997–1998. Over half of all citations received by this core group were from only 10 journals with the highest number of citations (Bohannon, 1999). In 2001, the issue of impact factors and their relationship with rehabilitation journals was explored by Lankhorst and Franchignoni, with the authors finding that Clinical Rehabilitation, was placed second in impact factor rankings among journals specifically dedicated to rehabilitation medicine (Lankhorst and Franchignoni, 2001). Although citation analysis has not yet been performed for Manual Therapy, a recent editorial pointed out that Masterclasses themselves now represent some of the most popular downloads (Beeton, 2008). Future bibliometric analysis of Manual Therapy would certainly be interesting to conduct, if only to establish whether article download trends in manual therapy mirror citation behavior. In the title of this paper we have called for a balance between the science and the art in manual therapy, and there are a few reasons why such an approach is necessary. Firstly, given the current obsession with journal impact factors, it is often forgotten that this measure contains various intrinsic limitations, and citation analysis itself is by no means perfect (Smith, 2008c). While the JCRÒ is known to be useful for those in the field of physical therapy (Bohannon, 1986), the impact factor calculation has not changed since it was first invented, and the two year ‘citation window’ may not be appropriate for every research
458
D.R. Smith, D.A. Rivett / Manual Therapy 14 (2009) 456–459
field, particularly some of the smaller sub-disciplines with longer publication lag-times (Smith, 2007). The impact factor itself may be due for an overhaul, perhaps as part of a general reflection and debate on where citation indexing is heading in future. From a social science perspective, Holden and colleagues have also suggested that ‘any research area worthy of investigation needs to have its methods continuously and critically reviewed’ (Holden et al., 2005, p. 4). Either way, most critics agree that impact factors tell only part of the story and the future use of alternative measures such as article download counts and internet-based journal sessions may offer more tangible alternatives (Favaloro, 2008). Secondly, there is the issue of journal coverage. While bibliographic databases clearly provide a means of assimilating common threads of ideas and data (Ebrahim, 2006), this can only be achieved if journals relevant to our field are actually being included. Recent years have also witnessed a large expansion of the various complementary disciplines in health care, although less than half of the studies in this area tend to be published in journals with impact factors (Raschetti et al., 2005). Rehabilitation research is published in a large number of different journals (Bohannon and Roberts, 1991), and not all will be included in citation-tracking databases. Not all physiotherapy journals have impact factors either. The longstanding journal Physiotherapy for example, was not ISI-listed until 2005 and thus did not receive an impact factor score until 2007 (Harms, 2006). As such it can be suggested that the publication of modern physiotherapy research may be a relatively new phenomenon in the scientific literature. The broader field of rehabilitation itself has a shorter tradition than some of the other medical disciplines, and as such, fewer human and financial resources are probably being dedicated to it (Tesio et al., 1995). Thirdly, there is the issue of ‘crossover’ given the fact that authors in multidisciplinary fields often have many different choices as to where their findings may be published. Manual therapy is well established as a multidisciplinary field and this is reflected in journal readership and article authorship. With regard to Manual Therapy journal itself, while most readers and authors are physiotherapists working in the field of manual therapy, the journal has also published work written by physicians, scientists, chiropractors and osteopaths (Beeton, 2008). Being interdisciplinary is not easy however, as individuals need to be familiar with other disciplines, and this often takes a great deal of time and effort (Lynch, 2006). Fourthly, there is the philosophical issue of why we conduct research in the first place. Achieving a professional balance between the science and the art of manual therapy is clearly important for the practitioner, and this point also needs to be remembered when publishing. Blind obsession with journal performance indicators is not always good for a journal and its readers, and it has been suggested that having a purely ‘impact factor centered approach’ can easily lead to a situation where everything practical, readable and entertaining is cut, in favor of material that will be cited (Smith, 2006). A periodical can easily fall into the trap of focusing more on those who might cite it, rather than those who will actually read it. On the other hand, journals still need to attract quality submissions on a regular basis, which is where having a high impact factor can be very useful. It is important for editors and readers to keep this balance in mind. Finally, and perhaps most critically, it is important to remember that while publication usually stems from research, becoming purely focused on research can be detrimental for the individual practitioner, as clinical skills might easily be forgotten. Although he was referring to dermatology, in a 1999 editorial that is equally relevant to modern manual therapy, Marks suggested that there
will always be a need to keep a balance between the art and science ‘lest we rely too much on the modern reductionist approach to defining clinical skills and rely too little on the lessons learned from history on the value of the bedside’ (Marks, 1999, p. 344). References Adair WC. Citation indexes for scientific literature? American Documentation 1955;6:31–2. Balon R. Reflections on relevance: psychotherapy and Psychosomatics in 2004. Psychotherapy and Psychosomatics 2005;74(1):3–9. Beeton K. Masterclass editorial. Manual Therapy 2008;13(5):373–4. Bohannon R. Core journals of physiotherapy. Physiotherapy Practice 1987;3(3): 126–8. Bohannon RW. Journal citation reports. Physical Therapy 1986;66(8):1275. Bohannon RW. Information accessing behaviour of physical therapists. Physiotherapy Theory and Practice 1990;6(4):215–25. Bohannon RW. Core journals of physiotherapy. Physiotherapy 1999;85(6):317–21. Bohannon RW, Gibson DF. Citation analysis of physical therapy. A special communication. Physical Therapy 1986;66(4):540–1. Bohannon RW, Roberts D. Core journals of rehabilitation: identification through index analysis. International Journal of Rehabilitation Research 1991;14(4): 333–6. Boldt J, Haisch G, Maleck WH. Changes in the impact factor of anesthesia/critical care journals within the past 10 years. Acta Anaesthesiologica Scandinavica 2000;44(7):842–9. Bradford SC. Sources of information on specific subjects. Engineering 1934;137:85–6. Brodman E. Methods of choosing physiology journals. Bulletin of the Medical Library Association 1944;32:479–83. Chew M, Villanueva EV, van der Weyden MB. Life and times of the impact factor: retrospective analysis of trends for seven medical journals (1994–2005) and their Editors’ views. Journal of the Royal Society of Medicine 2007;100(3):142– 50. D’Auria D. Occupational Medicine, publishing and the new millenium. Occupational Medicine (London) 1999;49(5):277. Dean E, Davies J. Frequency of citation and reputational assessment of contributors in physical therapy. Physical Therapy 1986;66(6):961–6. Ebrahim S. Entelechy, citation indexes, and the association of ideas. International Journal of Epidemiology 2006;35(5):1117–8. Favaloro EJ. Measuring the quality of journals and journal articles: the impact factor tells but a portion of the story. Seminars in Thrombosis and Hemostasis 2008;34(1):7–25. Garfield E. Citation indexes for science; a new dimension in documentation through association of ideas. Science 1955;122(3159):108–11. Garfield E. The new social sciences citation index (SSCI) will add a new dimension to research on man and society. Essays of an Information Scientist 1972;1: 317–9. Garfield E. Which medical journals have the greatest impact? Annals of Internal Medicine 1986;105(2):313–20. Gross PL, Gross EM. College libraries and chemical education. Science 1927;66(1713):385–9. Harms M. Impact factors. Physiotherapy 2006;92:73–4. Holden G, Rosenberg G, Barker K. Tracing thought through time and space: a selective review of bibliometrics in social work. Social Work in Health Care 2005;41(3–4):1–34. Joseph KS. Quality of impact factors of general medical journals. British Medical Journal 2003;326(7383):283. Kuhlemeier KV. A bibliometric analysis of the archives of physical medicine and rehabilitation. Archives of Physical Medicine and Rehabilitation 1992;73(2):126–32. Lankhorst GJ, Franchignoni F. The impact factor – an explanation and its application to rehabilitation journals. Clinical Rehabilitation 2001;15(2):115–8. Lynch J. It’s not easy being interdisciplinary. International Journal of Epidemiology 2006;35(5):1119–22. Marks R. The art, the science, and the practice of dermatology in the next millennium. International Journal of Dermatology 1999;38(5):343–4. Michels E. Evaluation and research in physical therapy. Physical Therapy 1982;62(6):828–34. Raschetti R, Menniti-Ippolito F, Forcella E, Bianchi C. Complementary and alternative medicine in the scientific literature. Journal of Alternative and Complementary Medicine 2005;11(1):209–12. Roberts D. Coverage by four information services of the core journals of rehabilitation and related topics. Scandinavian Journal of Rehabilitation Medicine 1992a;24(4):167–73. Roberts D. The journal literature of physiotherapy: quality through peer review and access through MEDLINE. Physiotherapy 1992b;78(1):29–33. Smith DR. Historical development of the journal impact factor and its relevance for occupational health. Industrial Health 2007;45(6):730–42. Smith DR. Bibliometrics, dermatology and contact dermatitis. Contact Dermatitis 2008a;59(3):133–6. Smith DR. Citation analysis and impact factor trends of 5 core journals in occupational medicine, 1985–2006. Archives of Environmental & Occupational Health 2008b;63(3):114–22.
D.R. Smith, D.A. Rivett / Manual Therapy 14 (2009) 456–459 Smith DR. Citation indexing and the development of academic journals in tropical medicine. Memorias do Instituto Oswaldo Cruz 2008c;103(3):310–2. Smith DR, Hazelton M. Bibliometrics, citation indexing, and the journals of nursing. Nursing & Health Sciences 2008;10(4):260–5. Smith R. Commentary: the power of the unrelenting impact factor – is it a force for good or harm? International Journal of Epidemiology 2006;35(5):1129–30.
459
Tesio L, Gamba C, Capelli A, Franchignoni FP. Rehabilitation: the Cinderella of neurological research? A bibliometric study. Italian Journal of Neurological Sciences 1995;16(7):473–7. Tse H. A possible way out of the impact-factor game. Nature 2008;454(7207):938–9. Wakiji EM. Mapping the literature of physical therapy. Bulletin of the Medical Library Association 1997;85(3):284–8.
Available online at www.sciencedirect.com
Manual Therapy 14 (2009) 460 www.elsevier.com/math
Diary of events NZMPA biennial scientific conference, Heritage Hotel, Rotorua, New Zealand 28, 29 & 30 August 2009. The theme is ‘Striving for Excellence in OMT’ & also celebrating 40 years of Manual Therapy in New Zealand. The conference co-coordinator is Vicki Reid, Phone 0800 646 000 or 09 476 5353 Fax 09 476 5354 e-mail:
[email protected] Website: www.nzmpa.org.nz
NOI International conference UK and Ireland Nottingham UK e April 15e17, 2010 Dublin IRELAND April 21e23, 2010 For further details www.noi2010.com Fax þ 3906 51882443
APA Conference Week, Sydney Convention Centre, Sydney, Australia 1-5 October 2009 For more information please visit www.apaconferenceweek09. asn.au
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.: þ1 301 656 0220; Fax: þ1 301 654 0333; website: www.painpoints.com/seminars.htm E-mail:
[email protected] Aaron Mattes 4 Day Active Isolated Stretching & Strengthening Seminar The Renaissance Hotel, Heathrow 15e18 October 2009 First ever UK Seminar. Further details regarding the contents of the seminar can be found at www.stretchingusa.com & registration details can be found at www.stretchinggb.com. Phone: 020 8897 0377 / 07984 005366. Email:
[email protected] The Diary of events will soon be moving to http://www.elsevier. com/math (click on the journal news tab).
1356-689X/$ - see front matter doi:10.1016/S1356-689X(09)00074-5
MUSCULOSKELETAL DISORDERS IN PRIMARY CARE RESEARCH CONGRESS 2010 Rotterdam 11-13 October 2010 For further information visit www.bsl.nl/primarycare
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. E-mail:
[email protected] There is no charge for this service.
Manual Therapy 14 (2009) e1
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Letter to the Editor
Scha¨fer A, Hall T, Briffab K. Classification of low back-related leg pain – A proposed patho-mechanism-based approach. Manual Therapy (2007) doi:10.1016/j.math.2007.10.003
Dear Editor We welcome the contribution to the debate about low backrelated leg pain, it’s origins and a possible means of sub-classification by Schafer et al. (2009) Whilst they raise many important and salient points, we do have some concerns about the model they propose and the evidence used to support it. Their classification of central sensitisation is characterised (see Figure 1) by a score of 12 or more on the Leeds Assessment of Neuropathic Symptoms and Signs (LANSS) a scale designed to identify pain of a probable neuropathic origin in a variety of neurological conditions (Bennett, 2001). A more appropriate tool to assess low back-related leg pain may be a questionnaire recently validated for low back and leg pain, painDETECT (Freynhagen et al., 2006). With regard to the interpretation of these and other neuropathic pain scales (Bennett et al., 2007), they have been designed to distinguish between neuropathic and non-neuropathic pain, and do not have the sensitivity or selectivity to distinguish between neuropathic pain of peripheral or central origin. Secondly, the proposal that Paraesthesia is a unique distinguisher between central and peripheral neuropathic pain lacks supporting evidence. Whilst Paraesthesia may well result from changes in central sensitivity it is commonly associated with irritation of peripheral nerves, nerve roots, and dorsal root ganglion (Devor, 2006). Furthermore, vascular compromise and the resultant ischaemia is a potent causative factor in the development of Paraesthesia, which occurs more commonly in peripheral nerves (Lundborg, 2005). Thirdly these authors propose that distal pain is a central phenomena and pain anywhere in the leg is peripherally mediated (Table 2). Such a distinction has not been demonstrated or reported. Finally it is proposed that Denervation is identified by ‘‘negative’’ symptoms. In fact positive symptoms including Paraesthesia are often present (a) in the first 6–12 weeks when Wallerian degeneration is an active process (Lundborg, 2005) and (b) during the recovery stages (tingling is the positive finding of the Tinel’s test for a recovering median nerve (Stewart and Eisen, 1978)). The algorithm summarising Schafer et al’s proposed model (Figure 1) is attractive and could easily be adopted in clinical practice. For the above reasons we have reservations about the model’s applicability, as the symptoms and signs used are not unique to any of the three neuropathic groupings suggested. It is our opinion that
1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.07.001
these inaccuracies are likely to lead to errors in clinical reasoning and at least some inappropriate management. Whilst the authors acknowledge that there is ‘‘significant overlap and causal interrelation between the different mechanisms’’ they suggest it is ‘‘feasible that there is a predominant mechanism primarily responsible for a patient’s complaints’’ which the proposed algorithm may help to detect. What little we currently do know about neuropathic pain and its management does not justify (a) the proposed groups included in this algorithm, or (b) the idea of a predominant mechanism. The suggested groupings of nervous system anatomy and physiological function/dysfunction may become valid as our knowledge increases, albeit with different underlying reasoning. The current model, however, is premature and may lead to errors in reasoning, and inappropriate management of this debilitating situation. References Bennett M. The LANSS pain scale: the Leeds assessment of neuropathic symptoms and signs. Pain 2001;92(1–2):147–57. Bennett MI, Attal N, Backonja MM, Baron R, Bouhassira D, Freynhagen R, et al. Using screening tools to identify neuropathic pain. Pain 2007;127(3):199–203. Devor M. Centralization, central sensitization and neuropathic pain. Focus on sciatic chronic constriction injury produces cell-type-specific changes in the electrophysiological properties of rat substantia gelatinosa neurons. J Neurophysiol 2006;96:522–3. Freynhagen R, Baron R, Gockel U, Tolle TR. painDETECT: a new screening questionnaire to identify neuropathic components in patients with back pain. Curr Med Res Opin 2006;22(10):1911–20. Lundborg G. Nerve and nerve injuries. 2nd ed. Oxford: Elsevier, 2005. Schafer A, Hall T, Briffa K. Classification of low back-related leg pain – a proposed pathomechanism-based approach. Manual Therapy 2009;14(2):222–30. Stewart JD, Eisen A. Tinel’s sign and the carpal tunnel syndrome. BMJ 1978;2(6145): 1125–6.
Iain Beith, PhD, MSc, MCSP, DipTP (Cert Ed)* Mick Thacker, MSc, MMACP, MCSP King’s College London, Division of Applied Biomedical Research, School of Biomedical & Health Sciences, Guy’s Campus, London SE1 1UL, UK Corresponding author. E-mail address:
[email protected] (I. Beith) 5 February 2008
Manual Therapy 14 (2009) e2
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Letter to the Editor
Letter to Editor Response: Scha¨fer A, Hall T, Briffa NK. Classification of low back-related leg pain – A proposed patho-mechanism-based approach. Manual Therapy (2007)
Dear Editor, We would like to thank Beith and Thacker for their interest in our paper (Scha¨fer et al., 2009). We cannot agree that the presentation of the classification system is premature but recognize it may need to be refined as evaluation studies progress. It is well recognized that a different treatment approach is required for patients with a predominant neuropathic pain component. For example the presence of sensory hypersensitivity, which is commonly associated with neuropathic pain, has been linked to a poor response to physiotherapy intervention (Sterling et al., 2003; Jull et al., 2007). For the purposes of our classification system, we selected a score of 12 on the Leeds Assessment of Neuropathic Symptoms and Signs (LANSS) scale as the criterion for the first level. This scale has demonstrated validity and reliability to identify neuropathic pain (Bennett, 2001) and has been used to detect neuropathic components in patients with low back-related leg pain (Kaki et al., 2005). Items within the LANSS scale are primarily concerned with identifying positive features of neuropathic pain arising from a mix of peripheral and central mechanisms. As there is ample evidence to support the role of central mechanisms in neuropathic pain (Woolf and Mannion, 1999; Jensen and Baron, 2003) we elected to use the term Central Sensitization when referring to the group with positive LANSS scores. At the time our classification system was developed, the painDETECT screening questionnaire was not yet published (Freynhagen et al., 2006). It is of interest, but not surprising, that the LANSS scale and the painDETECT questionnaires incorporate similar items. A strength of our classification system is that none of the subgroups are defined by a single clinical feature. In Table 2 we listed features that can be used clinically to align patients with categories but we did not mean to suggest that any individual features were unique to one diagnostic category. Although the classification system is based on the premise there will be a predominant underlying mechanism, the potential for overlap and causal interrelation between the different mechanisms acting in each patient was emphasized. Moreover, the hierarchical nature of the algorithm (Figure 1) does not preclude patients with features of a lower level classification being classed at a higher level. Beith and Thacker suggest the current understanding of pain mechanisms does not justify the concept of a predominant mechanism, yet it is widely acknowledged that neuropathic pain can rarely be attributed to a single mechanism (Finnerup et al., 2005). 1356-689X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2008.07.002
Where mixed mechanisms may be contributing, clinical judgment is required to establish the predominant mechanism (Bennett et al., 2006) to provide a foundation for clinical reasoning and treatment. We acknowledge the model may appear rudimentary in the complex framework of events occurring with nerve injury. If it is accepted that the predominant pathomechanisms active in each group differ, it follows that the classification system for the optimal treatment approach for each group is also likely to differ. Further evaluation will determine whether this form of classification has clinical utility and in time, we anticipate refinement of this classification system will occur.
References Bennett M. The LANSS pain scale: the Leeds assessment of neuropathic symptoms and signs. Pain 2001;92:147–57. Bennett MI, Smith BH, Torrance N, Lee AJ. Can pain can be more or less neuropathic? Comparison of symptom assessment tools with ratings of certainty by clinicians. Pain 2006;122:289–94. Finnerup NB, Otto M, McQuay HJ, Jensen TS, Sindrup SH. Algorithm for neuropathic pain treatment: an evidence based proposal. Pain 2005;118:289–305. Freynhagen R, Baron R, Gockel U, To¨lle TR. painDETECT: a new screening questionnaire to identify neuropathic components in patients with back pain. Curr Med Res Opin 2006;22:1911–20. Jensen TS, Baron R. Translation of symptoms and signs into mechanisms in neuropathic pain. Pain 2003;102:1–8. Jull G, Sterling M, Kenardy J, Beller E. Does the presence of sensory hypersensitivity influence outcomes of physical rehabilitation for chronic whiplash? A preliminary RCT. Pain 2007;129:28–34. Kaki AM, El-Yaski AZ, Youseif E. Identifying neuropathic pain among patients with chronic low-back pain: use of the Leeds assessment of neuropathic symptoms and signs pain scale. Reg Anesth Pain Med 2005;30:422–8. Scha¨fer A, Hall T, Briffa K. Classification of low back related leg pain – a proposed pathomechanism based approach. Manual Therapy 2009;14(10):222–33. 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. Woolf CJ, Mannion RJ. Neuropathic pain: aetiology, symptoms, mechanisms, and management. Lancet 1999;353:1959–64.
Axel Scha¨fer* Toby Hall Kathy Briffa Ruckenzentrum am Michel, Ludwig Erhart Strasse 18, 20459 Hamburg, Germany Corresponding author. Tel.: þ49 40 43280274. E-mail address:
[email protected] (A. Scha¨fer) 22 July 2008
Manual Therapy 14 (2009) e3
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Book Review Whiplash, Headache, and Neck Pain: Research Based Directions for Physical Therapies, G. Jull, M. Sterling, D. Falla, J. Treleaven, S. O’Leary. Churchill Livingstone (2008). 242 pp., ISBN: 9780443100475 The textbook by Gwendolen Jull et al. provides a contemporary overview of whiplash, headache and neck pain. As the authors note in the preface, research on cervical spine disorders ‘‘played second fiddle to low-back pain for a large part of the 20th century’’. In the last few decades, however, more extensive knowledge has become available on the cervical spine, and this book is designed to bring both students and clinicians up-to-date on recent developments. This book consists of 15 chapters covering three main areas, basic clinical science (including an analysis of psychosocial factors, sensory manifestations of neck pain, and disturbances in postural stability and movement control), differential diagnosis for the three conditions presented, and a discussion on clinical assessment (history, physical examination, and principles of management). The language used is reader friendly without sacrificing the science. Although ‘‘evidence-based medicine’’ has become a rallying cry in the last decade by clinicians of all disciplines, relatively few are able to bring this into practice because they lack the ability to dissect or understand the literature and apply it. This textbook attempts to tie the two together. As is appropriate or expected in
doi:10.1016/j.math.2009.04.002
such a textbook, there are some nice tables delineating, for example, the classification of whiplash as well as a proposed adaptation of this classification based on physical and psychological factors. However, I would have liked to have seen more tables scattered throughout the chapters. The authors demonstrate not only an appreciation of the scientific literature, but an understanding of the clinical aspect as well. For example, provocative testing is presented in many, even recent orthopaedic textbooks as rather matter of fact (i.e. a positive test indicates the presence of the disease while a negative test indicates its absence) without discussing the actual strategy or diagnostic accuracy of the test. Here Jull et al. report the latest information based upon a systematic review and discuss the nuances (or implications) of orthopaedic testing. In short, students as well as experienced clinicians who want a good overview of the vast body of literature in this area will appreciate this textbook. Sidney M. Rubinstein Department of Epidemiology and Biostatistics, EMGO Institute for Health and Care Research, VU University Center, Amsterdam, The Netherlands E-mail address:
[email protected] 2 April 2009
Manual Therapy 14 (2009) e4
Contents lists available at ScienceDirect
Manual Therapy journal homepage: www.elsevier.com/math
Book Review Clinical Sports Medicine – Medical Management and Rehabilitation, W.R. Frontera, S.A. Herring, L.J. Micheli, J.K. Silver. Saunders Elsevier (2007). Hard Cover, 512 pp., 336 ills, CD-ROM, $155.00 CAN, ISBN: 978-1-4160-2443-9 Advancement in sports medicine is an international effort shared by experts around the world. Clinical Sports Medicine – Medical Management and Rehabilitation is the culmination of the editors’ effort to bring a large group of international experts together within a 34 chapter, multi-authored (61 contributors), comprehensive sports medicine text. Within the preface, the editors state their intent to provide a comprehensive text for the medical and non-medical sports medicine professional with an emphasis on the rehabilitation management of the injured athlete. What we find, of their final product, is a commendable text that may have slightly missed its mark on its intended purpose. The text is organized into 3 sections. The first section, ‘‘General Scientific and Medical Concepts’’, overviews general topics in sports medicine such as physiology, conditioning, and nutrition. Following these topics are several chapters detailing management considerations for special populations in sports medicine. The second section, ‘‘Principles of Injury Care and Rehabilitation’’, reviews everything from pre-participation evaluations to prescribing medications. The last section is comprised of a discussion of common athletic injuries topographically organized into chapters. Accompanying the text is a CD-ROM that contains patient handouts and PDF files of diagrams taken directly from the textbook. Praiseworthy to the editors, considering 61 authors contributed to this text, each chapter is written with a consistent structure, and the multiple writing styles of each author blend seamlessly into the next chapter. The narration is well-written and easy to follow; however, there is a noticeable lack of visual aids to explain technically demanding concepts. As an intended comprehensive sports medicine text, many chapters lack sufficient information and detail to be classified as being comprehensive. For instance, there is an absence of review of SLAP lesions, internal impingement, and rotator interval tears within the shoulder chapter. The book is subtitled Medical Management and Rehabilitation. Of the 61 contributing authors, only 2 are not physicians. It is not surprising that the text is written from a physician’s perspective. The advantage of this authorship is the excellent content relating
doi:10.1016/j.math.2009.03.001
to the medical management of the athlete. Highlighting this strength are exceptional chapters on doping and sports, special populations in sports medicine, laboratory tests and diagnostic imaging, and prescribing medications for pain and inflammation. Where this text falters is in its lack of emphasis on rehabilitation. This is inconsistent with the text’s original intention. Considering the book is subtitled Medical Management and Rehabilitation, the textbook offers little to the practical application of rehabilitation for diagnosis-specific conditions. Reference is made consistently within chapters to the conservative management of the athlete. However, detailed diagnosis-specific protocols for rehabilitation are not mentioned in sufficient detail for in-office application. The reason for this omission may potentially be the result of not including experts from other rehabilitation and manual therapy health professions as contributing authors. Perhaps, the inclusion of such experts in a future edition may resolve this weakness. Ironically, due to the wealth of expert medical knowledge that contributed to the development of this text, this book is an excellent resource for the manual therapy practitioner seeking a textbook to further their understanding of the medical management of the injured athlete. When balancing the strengths and weaknesses of this resource, we have to keep in mind multi-authored textbooks will always have both excellent and mediocre chapters. While this text may not be detailed and broad enough to be considered a comprehensive reference text, it does have some chapters that are clinical ‘‘gems’’. While perhaps missing the mark on its intended focus on rehabilitation, the text is an admirable sports medicine resource written for the primary care sports medicine setting. From the perspective of a manual therapy health care professional, this text is a great resource to be included in the manual therapist’s sports medicine library. Alex Lee Graduate Education and Research Programs, Canadian Memorial Chiropractic College, 6100 Leslie Street, Toronto, Ontario, Canada M2H 3J1 Tel.: þ1 416 482 2340; fax: þ1 416 482 2560. E-mail address:
[email protected] 24 March 2009