Manual Therapy Journal - Volume 12, Issue 1, Pages 1-84 (February 2007)

VOLUME 12 NUMBER 1 PAGES 1–84 FEBRUARY 2007 Editors International Advisory Board Ann Moore PhD, GradDipPhys, FCSP, Ce...

Author: Editors: Ann Moore and Gwen Jull

13 downloads 975 Views 2MB Size Report

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

Report copyright / DMCA form

DOWNLOAD PDF

VOLUME 12 NUMBER 1 PAGES 1–84 FEBRUARY 2007

Editors

International Advisory Board

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

K. Bennell (Victoria, Australia) K. Burton (Hudders¢eld, UK) B. Carstensen (Frederiksberg, Denmark) E. Cruz (Setubal Portugal) L. Danneels (Mar|¤ akerke, Belgium) S. Durrell (London, UK) S. Edmondston (Perth, Australia) J. Endresen (Flaktvei, Norway) L. Exelby (Biggleswade, UK) J. Greening (London, UK) C. J. Groen (Utrecht,The Netherlands) A. Gross (Hamilton, Canada) T. Hall (West Leederville, Australia) W. Hing (Auckland, New Zealand) M. Jones (Adelaide, Australia) S. King (Glamorgan, UK) B.W. Koes (Amsterdam,The Netherlands) J. Langendoen (Kempten, Germany) D. Lawrence (Davenport, IA, USA) D. Lee (Delta, Canada) R. Lee (Brighton, UK) C. Liebenson (Los Angeles, CA, USA) L. Ma¡ey-Ward (Calgary, Canada) E. Maheu (Quebec, Canada) C. McCarthy (Coventry, UK) J. McConnell (Northbridge, Australia) S. Mercer (Queensland, Australia) D. Newham (London, UK) J. Ng (Hung Hom, Hong Kong) L. Ombregt (Kanegem-Tielt, Belgium) N. Osbourne (Bournemouth, UK) M. Paatelma (Jyvaskyla, Finland) N. Petty (Eastbourne, UK) A. Pool-Goudzwaard (The Netherlands) M. Pope (Aberdeen, UK) G. Rankin (London, UK) D. Reid (Auckland, New Zealand) M. Rocabado (Santiago, Chile) C. Shacklady (Manchester, UK) M. Shacklock (Adelaide, Australia) D. Shirley (Lidcombe, Australia) V. Smedmark (Stenhamra, Sweden) W. Smeets (Tongeren, Belgium) C. Snijders (Rotterdam,The Netherlands) R. Soames (Leeds, UK) P. Spencer (Barnstaple, UK) M. Sterling (St Lucia, Australia) P. Tehan (Victoria, Australia) M. Testa (Alassio, Italy) M. Uys (Tygerberg, South Africa) P. van Roy (Brussels, Belgium) B.Vicenzino (St Lucia, Australia) H.J.M.Von Piekartz (Wierden,The Netherlands) M.Wallin (Spanga, Sweden) M.Wessely(Paris, France) A.Wright (Perth, Australia) M. Zusman (Mount Lawley, Australia)

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

Visit the journal website at http://www.intl.elsevierhealth.com/journals/math doi:10.1016/S1356-689X(07)00008-2

ARTICLE IN PRESS

Manual Therapy 12 (2007) 1–2 www.elsevier.com/locate/math

Editorial

Dealing with heterogeneity in clinical trials Clinical trials would be a simpler enterprise if patients presented with homogenous clinical presentations to which we could assign simple diagnoses, if therapists uniformly applied standardised therapies using mechanistic decision rules based on objective and universally accepted criteria, and if patients responded in more or less uniform ways to intervention. Unfortunately, the clinical presentations for each diagnosis are varied, diagnosis can be difﬁcult, therapists choose to intervene very differently for the same condition or presentation, and patients’ outcomes often appear hard to predict. For almost any clinical problem, clinical presentations, diagnoses, interventions and outcomes are heterogenous. This makes clinical trials difﬁcult. Clinical research is a messy business. One way to deal with heterogeneity is to attempt to minimise it. For example a trial could recruit from homogeneous populations by deﬁning stringent inclusion and exclusion criteria. The process of clinical decision-making could be tightly constrained by requiring that the experimental intervention is always administered in a particular way, or by deﬁning precise algorithms for decisions about intervention. Trials with narrowly deﬁned populations and tightly constrained interventions are sometimes called ‘‘explanatory’’ clinical trials (Schwartz and Lellouch, 1967; McMahon, 2002; Herbert et al., 2005). Typically this approach maximises the effects of intervention and reduces variability of outcomes. So the explanatory approach is often preferred by researchers who are intent on proving the efﬁcacy of intervention. Alternatively, a trial might recruit from the diverse populations for whom therapy is usually provided in the course of normal clinical practice. Such populations will not usually be those in whom the intervention is most effective because therapists often offer intervention even when they are not optimistic of success. Therapists could be given freedom in exactly how they provide the experimental intervention, and they might be allowed to customise the intervention to particular needs of individual patients. These ‘‘pragmatic’’ trials reﬂect the way intervention is administered in the course of normal clinical practice. By giving in to heterogeneity, prag1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.11.001

matic trials may be less likely to ﬁnd large effects of interventions, but they have the advantage of telling us about the real-world effects of intervention. The primary purpose of clinical trials is to estimate ‘‘the effect’’ of intervention. But interventions do not have one effect on all patients; typically they are very effective for some patients and ineffective or even harmful for others. In other words, effects of intervention are heterogenous. Unfortunately, clinical trials cannot provide unbiased estimates of the effects of intervention on each participant in a trial. Alternative methods, such as single case experiments (also called n-of-1 designs) may permit conclusions to be drawn about effects of interventions on individuals (Barlow and Hersen, 1984), but these methods do not provide a basis for robust inference about effects of interventions on individuals other than those who participated in the study. Ultimately neither clinical trials nor single case experiments can provide what we most want: the capacity to make speciﬁc predictions about what the effect of intervention will be on our next patient. The best we can hope for from most clinical trials is an unbiased estimate of the average effect of the experimental intervention in a population. This has been a frequent source of criticism. Some have argued that by focusing on averages we ignore the heterogeneity of effects of interventions. After all, it is argued, we treat individual patients, not average patients. Of course that is true, but it fails to recognise why we might be interested in the average effect of an intervention: in the absence of better information about how individuals will respond, the average effect of intervention provides us with a ‘‘best guess’’ of what the effect of intervention will be on any individual (Herbert, 2000). Clinical trials cannot give us speciﬁc estimates of the effects of an intervention on our next patient but they can help us make unbiased guesses about effects of intervention that can form an appropriate basis for clinical decision making. That is not to say we should not try to identify those patients most likely and least likely to beneﬁt from an intervention. One of the next big challenges for clinical researchers investigating manual therapies is to identify

ARTICLE IN PRESS 2

Editorial / Manual Therapy 12 (2007) 1–2

characteristics of people who respond to therapy. (Sometimes such characteristics are called ‘‘effect modiﬁers’’, and in the context of clinical trials identiﬁcation of responders is sometimes referred to as ‘‘subgroup analysis’’.) Recently there has been a ﬂurry of interest in development of new taxonomies of low back pain (O’Sullivan, 2006) driven partly by a desire to be able to target interventions at people who will beneﬁt most from intervention. Manual therapy researchers have begun to think about the best methodologies for identifying effect modiﬁers (Beattie and Nelson, 2006). And the ﬁrst well designed studies have begun to identify which patients respond well to manual therapy interventions (Childs et al., 2004). Unfortunately, identiﬁcation of effect modiﬁers is a methodologically hazardous undertaking. A simple and common mistake is to confuse prognostic factors (predictors of outcomes) with effect modiﬁers (predictors of response to therapy). Prognostic factors can be identiﬁed using cohort studies, but effect modiﬁers can only be identiﬁed with controlled clinical trials. Rigorous identiﬁcation of effect modiﬁers involves contrasting effects of interventions across subgroups in randomised trials. The perils of naı¨ ve analyses have been widely discussed (Yusuf et al., 1991) and extensively analysed (Brookes et al., 2004). The message from this literature is that robust identiﬁcation of effect modiﬁers can only be carried out within the context of a randomised trial. Identiﬁcation of effect modiﬁers must involve prior speciﬁcation of a small number of speciﬁc hypotheses rather than undisciplined dredging of numerous hypotheses. Analysis must involve examination of the magnitude of the interaction between patient characteristics and intervention (Brookes et al., 2004). A consequence of the need to examine interactions is that sample size requirements are quadrupled (Brookes et al., 2004). Rothman and Greenland (1998) have pointed out that particular care must be taken in deﬁning what is meant by an interaction, because the magnitude of any interaction will depend on how the effect of intervention is measured. (For example, an interaction observed when the effect of an intervention is measured as an absolute risk reduction may evaporate when the effect is re-expressed as a relative risk.) A consequence is that

certain patient characteristics may appear to predict effects of intervention when effects are measured with one metric, but not when effects are measured with another. Heterogeneity is a universal feature of clinical practice that presents a challenge for clinical trialists. Careful consideration of sampling, intervention and analysis should make it possible to design trials which can support real-world clinical decision-making. References Barlow DH, Hersen M. Single case experimental designs: strategies for studying behavior change. Boston: Allyn and Bacon; 1984. Beattie P, Nelson R. Clinical prediction rules: what are they and what do they tell us? Australian Journal of Physiotherapy 2006;52(3):157–63. Brookes ST, Whitely E, Egger M, Smith GD, Mulheran PA, Peters TJ. Subgroup analyses in randomized trials: risks of subgroup-speciﬁc analyses; power and sample size for the interaction test. Journal of Clinical Epidemiology 2004;57(3):229–36. Childs JD, Fritz JM, Flynn TW, Irrgang JJ, Johnson KK, Majkowski GR, et al. 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(12):920–8. Herbert RD. How to estimate treatment effects from reports of clinical trials. I: continuous outcomes. Australian Journal of Physiotherapy 2000;46(3):229–35. Herbert RD, Jamtvedt G, Mead J, Hagen KB. Practical evidencebased physiotherapy. Oxford: Elsevier; 2005. McMahon AD. Study control, violators, inclusion criteria and deﬁning explanatory and pragmatic trials. Statistics in Medicine 2002;21(10):1365–76. O’Sullivan P. Classiﬁcation of lumbopelvic pain disorders—why is it essential for management? Manual Therapy 2006;11(3):169–70. Rothman KJ, Greenland S. Modern epidemiology. Philadelphia, PA: Lippincott-Raven; 1998. Schwartz D, Lellouch J. Explanatory and therapeutical attitudes in therapeutical trials. Journal of Chronic Diseases 1967;20: 637–48. Yusuf S, Wittes J, Probstﬁeld J, Tyroler HA. Analysis and interpretation of treatment effects in subgroups of patients in randomized clinical trials. Journal of the American Medical Association 1991;266(1):93–8.

Rob Herbert School of Physiotherapy, University of Sydney, Sydney, Australia E-mail address: [email protected]

ARTICLE IN PRESS

Manual Therapy 12 (2007) 3–11 www.elsevier.com/locate/math

Review

An evidence-based review on the validity of the Kaltenborn rule as applied to the glenohumeral joint Corlia Brandta,, Gisela Soleb, Maria W. Krausea, Mariette Nelc a

Department of Physiotherapy, Faculty of Health Sciences, University of the Free State, South Africa Musculoskeletal and Sports Physiotherapy, School of Physiotherapy, University of Otago, New Zealand c Department of Biostatistics, Faculty of Health Sciences, University of the Free State, South Africa

b

Received 25 January 2005; received in revised form 26 January 2006; accepted 15 February 2006

Abstract Kaltenborn’s convex–concave rule is a familiar concept in joint treatment techniques and arthrokinematics. Recent investigations on the glenohumeral joint appear to question this rule and thus accepted practice guidelines. An evidence-based systematic review was conducted to summarize and interpret the evidence on the direction of the accessory gliding movement of the head of the humerus (HOH) on the glenoid during physiological shoulder movement. Five hundred and eighty-one citations were screened. Data from 30 studies were summarized in ﬁve evidence tables with good inter-extracter agreement. The quality of the clinical trials rated a mean score of 51.27% according to the Physiotherapy Evidence Database scale (inter-rater agreement: k ¼ 0:6111). Heterogeneity among studies precluded a quantitative meta-analysis. Weighting of the evidence according to Elwood‘s classiﬁcation and the Agency for Health Care Policy and Research classiﬁcation guidelines indicated that evidence was weak and limited. Poor methodological quality, weak evidence, heterogeneity and inconsistent ﬁndings among the reviewed studies regarding the direction of translation of the HOH on the glenoid, precluded the drawing of any ﬁrm conclusions from this review. Evidence, however, indicated that not only the passive, but also the active and control subsystems of the shoulder may need to be considered when determining the direction of the translational gliding of the HOH. The indirect method, using Kaltenborn’s convex–concave rule as applied to the glenohumeral joint, may therefore need to be reconsidered. r 2006 Elsevier Ltd. All rights reserved. Keywords: Glenohumeral; Translational glide; Evidence-based; Kaltenborn

Contents 1. 2.

3.

Introduction/background . . . . . . . . . . . . . . . Methodology. . . . . . . . . . . . . . . . . . . . . . . . 2.1. The search strategy and data selection . 2.2. Quality assessment of the clinical trials. 2.3. Meta-analysis. . . . . . . . . . . . . . . . . . . 2.4. Weighting of the evidence . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Study characteristics . . . . . . . . . . . . . . 3.2. Methodological quality . . . . . . . . . . . . 3.3. Meta-analysis. . . . . . . . . . . . . . . . . . . 3.4. Level of the evidence . . . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

Corresponding author. P.O. Box 339 (G30), Bloemfontein 9300, South Africa. Tel: +51 4013297; fax: +51 4013290.

E-mail address: [email protected] (C. Brandt). 1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.02.011

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

.4 .4 .4 .5 .5 .5 .5 .5 .6 .6 .6

ARTICLE IN PRESS 4

C. Brandt et al. / Manual Therapy 12 (2007) 3–11

4.

Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Methodological quality of the clinical trials . . . . . . . . . . . . . . . 4.2. The evidence on the arthrokinematics of the glenohumeral joint. 4.3. Relating the ﬁndings to Kaltenborn‘s rule and theory . . . . . . . . 4.4. Implications and recommendations . . . . . . . . . . . . . . . . . . . . . 4.5. Limitations of this review . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction/background Dysfunction of the shoulder girdle is one of the most common musculoskeletal conditions to be treated in primary care. Thirty-four per cent of the general population may suffer from shoulder pain at least once in their lifetime (Green et al., 2002). In addition to the high incidence rate, shoulder dysfunction is often persistent and recurrent (Winters et al., 1999). Physiotherapy for shoulder dysfunction may include manual therapy joint techniques to treat pain or stiffness. Various approaches to treatment have been proposed, such as the Maitland approach (Maitland, 1998), movement with mobilization (Mulligan, 1999), and the application of passive mobilization techniques following the convex–concave rule (Kaltenborn and Evjenth, 1989). The latter approach is based on direct and indirect assessment of translational glides. Using the direct method, the passive translational gliding movements are performed by the therapist to the patient’s painful and/or stiff joint to determine which direction may be limited (Kaltenborn and Evjenth, 1989). Joint mobilizations would then be performed as a treatment method in the decreased direction to restore normal movement. The indirect method of determining the direction of translational glide was termed the ‘‘Kaltenborn convex–concave rule’’ (Kaltenborn and Evjenth, 1989). This rule was ﬁrst described by MacConaill (1953). Following this method, the therapist examines active and passive physiological movements such as ﬂexion, extension, abduction and lateral rotation (Kaltenborn and Evjenth, 1989). The direction of the glide would then be determined by considering the geometry of the moving articular surfaces. In the glenohumeral joint, the glenoid fossa (concave surface) was considered to be stable (ﬁxed) while the humeral head (convex surface) would be moved (mobilized) during a physiological shoulder movement. According to the convex–concave rule, the convex surface (humeral head) would glide in the opposite direction to the bone movement. Thus, during abduction of the arm, the humeral head would glide caudally. Kaltenborn and Evjenth (1989) proposed that for restricted shoulder extension and lateral rotation, the humeral head should be glided ventrally (anteriorly), and for restricted ﬂexion and medial rotation, the humeral head should be glided dorsally (posteriorly).

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

.7 .7 .8 .9 .9 10 10 10

Kaltenborn and Evjenth (1989) thus based the clinical reasoning of appropriate direction of translational glide mainly on the anatomy of the osseous articulating surfaces. More recently it has been suggested that other factors, such as the concept of functional stability (Panjabi, 1992), may also need to be considered in the assessment of the arthrokinematics of the glenohumeral joint (Hess, 2000). The question thus arose whether the convex–concave rule is valid in the clinical reasoning of the most appropriate direction of translational glide applied in the assessment and treatment of shoulder dysfunction. The aim of this study was to investigate the evidence on the arthrokinematics of the glenohumeral joint supporting or negating the validity of the MacConaill and Kaltenborn rule and theory.

2. Methodology 2.1. The search strategy and data selection An academic, computerized search was conducted. CINAHL, MEDLINE, The Cochrane Controlled trials register of randomized controlled trials, Kovsiedex, South African Studies and Sport Discussion were searched from 1966 to October 2003. The search was limited to English and human studies. Keywords such as shoulder, glenohumeral, kinematics, arthrokinematics, mechanics, translation(al), roll(-ing) and/or glide(-ing), accessory movement, and Kaltenborn were optimally combined. The search was continued over a period of ten months (Hoepﬂ, 2002). The titles and the abstracts of the retrieved citations were screened for relevance by the primary investigator. The reference lists of the relevant articles were checked by one reviewer to identify additional publications. Five clinical experts in the ﬁeld of shoulder orthopaedics were also contacted in order to retrieve data (Oxman et al., 1994; Mays and Pope, 1999; Green et al., 2002; Tugwell et al., 2003). The second screening consisted of the blinded assessment of the full papers’ Method and Results sections by two independent reviewers. The reports were numbered at random and the authors‘ names and afﬁliations, the name of the journal, the date of publication, and the acknowledgements were erased to ensure blinded assessment. All types of study designs were included in the systematic

ARTICLE IN PRESS C. Brandt et al. / Manual Therapy 12 (2007) 3–11

review to increase its clinical value (Mays and Pope, 1999; Elwood, 2002; Hoepﬂ, 2002; Fritz and Cleland, 2003). In vivo and in vitro studies were assessed. The investigated population had to be human (male and/or female), a mean age of 15 years or older, with or without shoulder pathology. The study had to investigate a variable factor regarding glenohumeral joint translation and had to measure the direction of translation of the humeral head on the glenoid fossa during normal or simulated, active or passive physiological shoulder movement. The reviewers decided upon inclusion by means of consensus (Oxman et al., 1994; Jadad et al., 1996). Data were extracted from the included reports and summarized on a standardized data collection form by two independent, masked reviewers. The form provided for the gathering of information on the study design, subgroups, exposure or intervention, study population, research methodology, data analysis, main results, hypotheses, and any other relevant data (Oxman et al., 1994; Elwood, 2002; Scholten-Peeters et al., 2003; Tugwell et al., 2003). The data were recorded (by means of consensus) as stated in the report. Where data were unclear and biased recording a possibility, it was clearly indicated (Scholten-Peeters et al., 2003). 2.2. Quality assessment of the clinical trials The quality of the clinical trials were assessed by means of the 11-item Physiotherapy Evidence Database (PEDro) scale which was developed by the Centre for evidencebased Physiotherapy, University of Sydney. The PEDro scale measures the internal validity and the sufﬁciency of the statistical information provided by a clinical trial. The scale assesses criteria such as random allocation, concealment of allocation, comparibility of groups at baseline, blinding of patients, therapists and assessors, analysis by intention to treat, adequacy of follow-up, between group statistical comparisons, report of point estimates, and measures of variability. Though the PEDro scale does not usually assess the external validity of a trial, this item from the Delphi list (upon which the PEDro scale is based), was included in the assessment. Verhagen et al. (1998) reported that external validity should form part of any concept of quality (Verhagen et al., 1998; Woolf, 2000; ‘‘PEDro: frequently asked questions’’, 2003). Two masked reviewers independently scored the quality of the studies (Jadad et al., 1996; Moher et al., 1996; Dickersin & Berline, 1997). Criteria were rated as yes when they were clearly satisﬁed on reading of the report, as no when an unbiased decision could be made that the criteria were not satisﬁed, and as don’t know when the information was insufﬁcient or unclear and a biased decision possible. Points were allocated for all the clearly satisﬁed items (Verhagen et al., 1998; ‘‘PEDro: the PEDro scale’’, 2003). The mean quality score, the total frequency results, as well as the frequency results on each item were

5

calculated. A study was considered as high quality if it satisﬁed at least 50% of the criteria (X5.5 points) (Maher et al., 2003; Scholten-Peeters et al., 2003). The k statistic and the 95% conﬁdence level provided for measurement of interobserver agreement (Maher et al., 2003; Scholten-Peeters et al., 2003). 2.3. Meta-analysis Clinical trials were considered for meta-analysis regardless of their quality score in order to reduce bias (Guyatt et al., 1995; Woolf, 2000). The following study characteristics were compared by two independent reviewers in order to identify the possibility of statistical pooling of results: (i) the study populations, (ii) the interventions, (iii) the sample sizes, (iv) the availability and format of the results, (v) the statistical methodology used for analysis, and (vi) the hypotheses tested (Dickersin and Berline, 1997). 2.4. Weighting of the evidence The strength of the scientiﬁc evidence was rated by two analysts according to two classiﬁcation systems (Moher et al., 1996; Elwood, 2002; Mays and Pope, 2002) namely, (i) a hierarchy of evidence (Table 1) relevant to human health studies (Elwood, 2002) and (ii) the modiﬁed classiﬁcation of the Agency for Health Care Policy and Research (AHCPR) guidelines (Table 2) on acute low back problems in adults (Ejnisman et al., 2002). 3. Results 3.1. Study characteristics Fig. 1 depicts the results yielded by the search and selection process. Eighteen clinical trials, seven comparative, and ﬁve descriptive studies were included in the review. Summary of the data indicated major methodological heterogeneity. Researchers used various protocols and measuring instruments such as magnetic tracking devices or position sensors (n ¼ 11), three-dimensional magnetic resonance imaging (n ¼ 4), computertomography (n ¼ 3), ultrasonic devices (n ¼ 2), potentiometers (n ¼ 3), radiographs (n ¼ 6), and arthroscopy (n ¼ 1) for investigation. Eleven studies were conducted in vivo and 19 in vitro. Movements were either done passively (n ¼ 15) or actively (n ¼ 14); simulated, static or continuous, while the plane of motion also varied. Data were gathered on eight different physiological movements performed through a variety of ranges of motion. The movements of active ﬂexion, active extension, and passive horizontal extension were not included in any investigation. The literature indicated six main factors to explain the translational behaviour of the humeral head namely, the inﬂuence of (i) the capsulo-ligamentous structures (n ¼ 17), (ii) neuromuscular control (n ¼ 17),

ARTICLE IN PRESS C. Brandt et al. / Manual Therapy 12 (2007) 3–11

6 Table 1 Elwood’s hierarchy of evidence Level

Deﬁnition of type of evidence

1 1m 1s

Randomized intervention trials, properly performed on an adequate number of subjects, in a human situation. Results from a meta-analysis of trials. One or more individual trials.

2 2m 2s

Observational studies, namely cohort and case–control designs, of appropriately selected groups of subjects. Results from a meta-analysis of such studies. One or more individual studies.

3

Comparative studies that compares groups of subjects representative of different populations or subject groups. For example: correlation studies of populations in which data on each individual are not assessed separately and informal comparisons between patients. Case series, descriptive studies, professional experience. The evidence is largely anecdotal, unsystematically recollected (for example ‘‘clinical judgement’’ and ‘‘experience’’), conclusions based on traditional practice, information derived from other species, in vitro testing, basic physiological principles and indirect assessments.

4

Table 2 The modiﬁed classiﬁcation of the AHCPR guidelines on acute low back problems in adults Level

Deﬁnition of type of evidence

A

Strong research-based evidence provided by generally consistent ﬁndings in multiple (more than one) high-quality randomized clinical trial (RCT). Moderate research-based evidence provided by generally consistent ﬁndings in one high-quality RCT and one or more low-quality RCT, or generally consistent ﬁndings in multiple low quality RCTs. Limited research-based evidence provided by one RCT (either high or low quality) or inconsistent or contradictory evidence ﬁndings in multiple RCTs. No research-based evidence: no RCTs.

B C D

(iii) articular geometry/congruency/conformity (n ¼ 8), (iv) negative intra-articular pressure (n ¼ 4), (v) rigidiﬁcation of musculature (n ¼ 1), and (vi) gravity (n ¼ 1). Agreement between the reviewers were 100% for the data extracted on the sample and methodological characteristics. Disagreement occurred only on the study design in two of the studies which was resolved by means of consensus. 3.2. Methodological quality The mean PEDro score of the clinical trials equalled 51.27%. Table 3 summarizes the individual results. The inter-rater agreement for quality assessment was poor (k ¼ 0:611). This was conﬁrmed by the 95% conﬁdence level of [0.8661;0.3562]. 3.3. Meta-analysis Heterogeneity among studies, insufﬁcient reported data, and poor study quality precluded statistical pooling of results. 3.4. Level of the evidence Twenty-ﬁve of the reviewed studies were analysed qualitatively. Five studies were excluded due insufﬁcient information provided for classiﬁcation purposes.

According to Elwood’s classiﬁcation (Table 1), one study fulﬁlled the criteria for level 2 s evidence, ﬁve for level 3 and 19 studies for level 4 evidence. The level 2 s evidence found (i) translation to be in the opposite direction during active physiological movement in pathological joints and (ii) the humeral head to remain centered during active physiological movement in normal joints (Paletta et al., 1997). For all other stratiﬁed movement planes, only levels 3 and 4 evidence were found. Table 4 summarizes the amount and level of evidence found on the direction of the translational movement of the humeral head. According to the AHCPR rating system (Table 2), level C evidence is contradictory on the direction of translation during active and passive lateral rotation in 901 of elevation in normal and reconstructed joints (Karduna et al., 1997; Williams et al., 2001). Only inconsistent, level D evidence could be found on the translation occurring during physiological movements in other planes. Inclusion of only higher quality clinical trials (quality score X54.5%) in the weighting of the evidence induced the following changes: according to Elwood‘s classiﬁcation, only level 4 evidence was now available, while the level of evidence according to the AHCPR rating system, remained unchanged.

ARTICLE IN PRESS C. Brandt et al. / Manual Therapy 12 (2007) 3–11

7

COMPUTER-BASED SEARCH of databases: 555 citations

First screening: retrieved and read 56 articles

6 articles not available internationally

Articles included for quality assessment = 11 21 articles were selected, summarized, and data extracted 10 articles were excluded from quality assessment because of study design

REFERENCE CHECKING: 26 articles identified, 21 retrieved and screened

5 articles not available internationally

Articles included for quality assessment = 7 9 articles were selected, summarized, and data extracted 2 articles were excluded from quality assessment because of study design

RESPONSE FROM EXPERTS: 00 articles

Total relevant articles reviewed = 30 Fig. 1.

4. Discussion 4.1. Methodological quality of the clinical trials Analysis of the methodology used by some of the included studies lead to serious concerns regarding the biomechanical and neurophysiological validity of their

results (to be discussed in the next section). According to the PEDro scale, methodological shortcomings of the clinical trials concerned mostly the insufﬁcient reporting of random allocation, insufﬁcient reporting of concealment of allocation, and insufﬁcient or unclear description of blinding of therapists and assessors. This may indicate that many of the clinical trials were, in fact, not

ARTICLE IN PRESS C. Brandt et al. / Manual Therapy 12 (2007) 3–11

8

randomized, which may raise some concern regarding the appropriateness of the PEDro scale for assessing these trials (Verhagen et al., 1998). It should be noted, though, that poor reporting does not necessarily imply that the criteria were not satisﬁed during the execution of the trial (Elwood, 2002). Table 3 Summary of the quality scores of clinical trials Study

Mean quality scores (out of 11)

Level 2s evidence Paletta et al. 1997 Level 4 evidence Karduna et al. (1997) Harryman et al. (1992) Harryman et al. (1990) McMahon et al. (1995) Gohlke et al. (1994) Vaesel et al. (1997) Novotny et al. (1998) Williams et al. (2001) Apreleva et al. (1998) Wuelker et al. (1994) Loehr et al. (1994) Karduna et al. (1996) Thompson et al. (1996) Helmig et al. (1993) Wuelker et al. (1998) Debski et al. (1995)

5 7 6 6 4.5 6 5 4.5 6.5 6 6 5 6 5 6 6 5

Total mean score

A quality score of 50–60% have been suggested as a cut-off to distinguish between good and poor quality studies (Maher et al., 2003; Scholten-Peeters et al., 2003). The mean quality score of 51.27% together with the poor inter-rater agreement (k ¼ 0:611) necessitated careful consideration regarding the methodological quality of the included clinical trials (Oxman et al., 1994; Elwood, 2002; Scholten-Peeters et al., 2003). The best approach when comparing the agreement between two raters is to calculate the k statistic. Similar to other methods, such as McNemar’s test which was also calculated (0.3103), small frequency tables (in this study n ¼ 30) present difﬁculties associated with the use and interpretation of kappa (Altman, 1996; Elwood, 2002). The problem most cited is that the value of k depends upon the proportion of subjects in each category. Landis and Koch (1977), as well as Elwood (2002), have characterized ranges of values for kappa with respect to the degree of agreement they suggest. Values greater than 0.75 may be taken to represent excellent agreement beyond chance, values below 0.40 may be taken to represent poor agreement beyond chance, and values between 0.40 and 0.75 may be taken to represent fair to good agreement beyond chance. 4.2. The evidence on the arthrokinematics of the glenohumeral joint The best evidence (level 2 s), as well as many of the selected studies (n ¼ 17), supported the hypotheses of

5.64

Table 4 Levels of evidence Physiological movement

Active: normal joints

Active: pathological joints

Passive: normal joints

Passive: pathological joints

Direction of translation of humeral head Same

Opposite

Centered

Non-uniform

n¼8 Level: 3 (n ¼ 1) 4 (n ¼ 7)

n¼2 Level: 3 (n ¼ 1) 4 (n ¼ 1)

—

Level: C

Level: C

n¼5 Level: 2s (n ¼ 1) 3 (n ¼ 1) 4 (n ¼ 3) Level: D

n¼7 Level: 3 (n ¼ 1) 4 (n ¼ 6) Level: C

n¼3 Level: 2s (n ¼ 1) 3 (n ¼ 1) 4 (n ¼ 1) Level: C

n¼6 Level: 3 (n ¼ 1) 4 (n ¼ 5) Level: C n¼7 Level: 3 (n ¼ 1) 4 (n ¼ 6) Level: C

Level: D

n¼2 Level: 4 (n ¼ 2)

—

Level: D

Level: D

n¼2 Level: 3 (n ¼ 1) 4 (n ¼ 1) Level: C

n¼1 Level: 4 (n ¼ 1)

n¼4 Level: 3 (n ¼ 1) 4 (n ¼ 3) Level: D

n¼3 Level: 4 (n ¼ 3)

n¼1 Level: 4 (n ¼ 1)

Level: C

Level: D

Level: D

n¼2 Level: 3 (n ¼ 1) 4 (n ¼ 1) Level: D

Levels of evidence are indicated according to Elwood’s classiﬁcation system (normal print) and according to the AHCPR’s guidelines (in italics). —, No evidence; n ¼ amount of studies.

ARTICLE IN PRESS C. Brandt et al. / Manual Therapy 12 (2007) 3–11

capsulo-ligamentous structures and neuromuscular control inﬂuencing the translation of the head of the humerus (HOH). The capsulo-ligamentous structures may be responsible for an obligatory translation of the humeral head at the end range of motion when the capsule and/or ligaments are tensioned. This was especially observed during passive motion in the absence of rotator cuff activity (Howell et al., 1988; Harryman et al., 1990, 1992; Gohlke et al., 1994; Debski et al., 1995; Karduna et al., 1996, 1997; Paletta et al., 1997; Novotny et al., 1998; Rhoad et al., 1998; Baeyens et al., 2000; Williams et al., 2001). During active movement the stabilizing effect of the rotator cuff on the humeral head causes a centring motion (Poppen and Walker, 1976; Howell et al., 1988; Gohlke et al., 1994; Wuelker et al., 1994, 1998; Debski et al., 1995; Karduna et al., 1996; Thompson et al., 1996; Karduna et al., 1997; Paletta et al., 1997; Apreleva et al., 1998; Rhoad et al., 1998; Graichen et al., 2000; Williams et al., 2001; Von Eisenhart-Rothe et al., 2002). Any loss of or defect in the stabilizing mechanism of the shoulder joint may increase or disrupt normal translational patterns, depending on the involved structure and its role in the gliding of the humeral head (Poppen and Walker, 1976, 1978; McGlynn and Caspari, 1984; Howell et al., 1988; Ozaki, 1989; Harryman et al., 1990; Helmig et al., 1993; Loehr et al., 1994; Debski et al., 1995; McMahon et al., 1995; Deutsch et al., 1996; Thompson et al., 1996; Karduna et al., 1997; Paletta et al., 1997; Apreleva et al., 1998; Novotny et al., 1998; Wuelker et al., 1998; Baeyens et al., 2000, 2001; Graichen et al., 2000; Von Eisenhart-Rothe et al., 2002). Pain, muscle spasm, and loss of proprioception associated with shoulder dysfunction may lead to neurophysiological responses. Imbalance/incoordination of the shoulder musculature may inﬂuence the translation of the humeral head (Poppen and Walker, 1976; Wuelker et al., 1994, 1998; Bertoft, 1999; Graichen et al., 2000; Von Eisenhart-Rothe et al., 2002). In correllation with the original theory of MacConaill and Kaltenborn, some studies did report that geometrical factors, such as the size of the humeral head, may determine translation. Increased head size seems to distension the capsule and thus reduce translation (Vaesel et al., 1997; Rhoad et al., 1998). To relate the ﬁndings of this review on the translational direction of the humeral head to the Kaltenborn rule, the best evidence will be considered (Elwood, 2002). The level 2 s evidence (quality score o50%) found translation to be in the opposite direction during active horizontal extension with lateral rotation and in the same direction during active abduction in anterior unstable joints and joints with rotator cuff tears. The humeral head remained centred during active abduction in normal shoulder joints (Paletta et al., 1997). According to the AHCPR classiﬁcation, level C evidence (n ¼ 2, quality scores

9

450%) were contradicting regarding the translational direction during active and passive lateral rotation in 901 of elevation in normal and reconstructed joints (Karduna et al., 1997; Williams et al., 2001). Considering Table 4, interpretations with regards to the convex–concave rule need to be made with caution due to the following limitations: (i) the table is not representative of all physiological movements since certain motion planes were not investigated by any of the studies; (ii) ﬁndings regarding the direction of translation were inconsistent for different physiological motion planes, and (iii) heterogeneous shoulder pathologies were grouped together, although these may affect translation in different manners (Burkhart, 1994; Meister, 2000). 4.3. Relating the findings to Kaltenborn‘s rule and theory Kaltenborn and MacConaill based their hypotheses of normal and abnormal intra-articular dynamics on the geometry of the articulating surfaces and location of the movement axis alone (MacConaill, 1953; Kaltenborn and Evjenth, 1989). The evidence indicates (i) different arthrokinematic behaviour for normal and dysfunctional joints and (ii) that not only the passive subsystem, but also the active and control subsystems may determine intra-articular gliding motion. It appears that Kaltenborn’s rule for the treatment of restricted joint motion may be valid if the intention of the treatment is to stretch a tight capsulo-ligamentous structure causing limitation of the physiological joint motion. By gliding the humeral head in the opposite direction of the restricted physiological bone movement, the restricting capsulo-ligamentous structure may be stretched. According to the evidence, however, this motion performed by the therapist may not necessarily mimic the true gliding taking place due to the tight structure. 4.4. Implications and recommendations Clinically authors postulate that the validity of the Kaltenborn rule might not be accepted dogmatically. The arthrokinematics of each patient might need to be considered in the context of existing neuro-musculoskeletal and biopsychosocial dysfunction which requires the process of clinical reasoning. Scientiﬁcally such a recommendation still lacks evidence. Methodologically sound, randomized, clinically controlled, in vivo, and homogeneous primary studies are needed on this subject. As such studies emerge, this review should be updated and reproduced. To ensure a meta-analysis in future reviews, the following criteria need to be considered: (i) movement should be classiﬁed as active or passive, (ii) the plane and the range of motion investigated should be similar, (iii) homogeneous pathologies should be grouped, and (iv) measuring

ARTICLE IN PRESS 10

C. Brandt et al. / Manual Therapy 12 (2007) 3–11

instruments, exposures or interventions, as well as the hypotheses tested, should be similar. 4.5. Limitations of this review Bias needs to be considered. Only one reviewer was involved in the initial screening of the 555 citations. A few articles could not be retrieved internationally and attempts to retrieve unpublished literature yielded no results. Working with such considerable amounts of evidence could not exclude the possibility of including multiple publications from the same large trial. Careful inspection, though, did not reveal any such errors. Information from papers concerning the same variables or cohorts, may inﬂuence the quality rating of similar papers later on. Earlier papers can provide the reviewers with additional information on validity. This review lacks statistical strength due to the preclusion of a meta-analysis and the poor kappa value calculated for inter-rater agreement. The ﬁndings should be interpreted with caution due to the limitations of a qualitative/categorical analysis. 5. Conclusion Inconsistent evidence, poor methodological quality and heterogeneity among the reviewed studies precluded the drawing of any ﬁrm conclusions regarding the direction of translation of the humeral head on the glenoid. The indirect method using Kaltenborn‘s convex–concave rule, as applied to the glenohumeral joint, need to be investigated appropriately by primary studies to determine its validity. It can only be postulated that not only the passive subsystem, as proposed by Kaltenborn, but also the active and control subsystems may need to be considered when determining the direction of the translational gliding movement of the humeral head. It is suggested that clinical decisions of appropriate gliding directions in the assessment and treatment of a patient with shoulder dysfunction should be considered carefully at this stage. References Altman DG. Practical statistics for medical research. London: Chapman and Hall; 1996. p. 403–409. Apreleva M, Hasselman CT, Debski RE, Fu FH, Woo SLY, Warner JJP. A dynamic analysis of glenohumeral motion after simulated capsulolabral injury. A cadaver model. Journal of Bone and Joint Surgery 1998;80A:474–80. Baeyens JP, Van Roy P, Clarys JP. Intra-articular kinematics of the normal glenohumeral joint in the late preparatory phase of throwing: Kaltenborn’s rule revisited. Ergonomics 2000;43(10): 1726–37. Baeyens JP, Van Roy P, De Schepper A, Declercq G, Clarijs JP. Glenohumeral joint kinematics related to minor anterior instability

of the shoulder at the end of the late preparatory phase of throwing. Clinical Biomechanics 2001;16:752–7. Bertoft ES. Painful shoulder disorders from a physiotherapeutic view: a review of literature. Physical and Rehabilitation Medicine 1999;11:229–77. Burkhart SS. Reconciling the paradox of rotator cuff repair versus debridement: a uniﬁed biomechanical rationale for the treatment of rotator cuff tears. Arthroscopy 1994;10:4–19. Debski RE, McMahon PJ, Thompson WO, Woo SLY, Warner JJP, Fu FH. A new dynamic testing apparatus to study glenohumeral joint motion. Journal of Biomechanics 1995;28(7):869–74. Deutsch A, Altchek DW, Schwartz E, Otis JC, Warren RF. Radiologic measurement of superior displacement of the humeral head in the impingement syndrome. Journal of Shoulder and Elbow Surgery 1996;5(3):186–493. Dickersin K, Berline T. Combining the results of several studies. In: Lang TA, Secic M, editors. How to report statistics in medicine. American College of Physicians; 1997. p. 171–84 [Chapter 11]. Ejnisman B, Carrera EF, Fallopa F, Peccin MS, Cohen M. Interventions for tears of the rotator cuff in adults (Protocol for a Cochrane Review). In: The Cochrane Library, issue 4. Oxford: Update Software; 2002. Elwood JM. Critical appraisal of epidemiological studies and clinical trials, 2nd edn. Oxford: Oxford University Press; 2002. p. 105–115, 198–244. [Chapters 5,8–9]. Fritz JM, Cleland J. Effectiveness versus efﬁcacy: more than a debate over language. Journal of Orthopaedic and Sports Physical Therapy 2003;33(4):163–5. Gohlke FE, Barthel T, Daum P. Inﬂuence of T-shift capsulography on rotation and translation of the glenohumeral joint: an experimental study. Journal of Shoulder and Elbow Surgery 1994;3: 361–70. Graichen H, Stammberger T, Bone`l H, Englmeier K- H, Reiser M, Eckstein F. Glenohumeral translation during active and passive elevation of the shoulder–a 3D open-MRI study. Journal of Biomechanics 2000;33:609–13. Green S, Buchbinder R, Glazier R, Forbes A. Interventions for shoulder pain (Cochrane review). In: The Cochrane library, issue 2. Oxford: Update Software; 2002. Guyatt GH, Sackett DL, Sinclair JC, Hayward R, Cook DJ, Cook RJ. User’s guide to the medical literature: IX. A method for grading health care recommendations. Journal of the American Medical Association 1995;274(22):1800–4. Harryman DT, Sidles JA, Clark JM, Mcquade KJ, Gibb TD, Matsen FA. Translation of the humeral head on the glenoid with passive glenohumeral motion. The Journal of Bone and Joint Surgery 1990;72A(9):1334–43. Harryman DT, Sidles JA, Harris SL, Matsen FA. The role of the rotator interval capsule in passive motion and stability of the shoulder. The Journal of Bone and Joint Surgery 1992;74A(1): 53–66. + JF, Østgaard SE, Suder P. Helmig P, Søjbjerg JO, Sneppen O, Lohr Glenohumeral movement patterns after puncture of the joint capsule: an experimental study. Journal of Shoulder and Elbow Surgery 1993;2:209–15. Hess SA. Functional stability of the glenohumeral joint. Manual Therapy 2000;5(2):63–71. Hoepﬂ MC. Choosing qualitative research: A primer for technology education researchers. Acrobat reader: 1–15. Retrieved May 28, 2002 from the World Wide Web: http://www.curriculum.edu.au/ tech/articles/choose.htm, 2002. Howell SM, Galinat BJ, Renzi AJ, Marone PJ. Normal and abnormal mechanics of the glenohumeral joint in the horizontal plane. The Journal of Bone and Joint Surgery 1988;70A(2):227–32. Jadad AR, Moore A, Carroll D, Jenkinson C, Reynolds DJM, Gavaghan DJ, McQuay HJ. Assessing the quality of reports of

ARTICLE IN PRESS C. Brandt et al. / Manual Therapy 12 (2007) 3–11 randomized clinical trials: Is blinding necessary? Controlled Clinical Trials 1996;17:1–12. Kaltenborn FM, Evjenth O. Manual mobilization of the extremity joints. Basic examination and treatment techniques (I), 4th edn. Oslo: Olaf Norlin Bokhandel; 1989. p. 26–27. Karduna AR, Williams GR, Williams JL, Iannotti JP. Kinematics of the glenohumeral joint: inﬂuences of muscle forces, ligamentous constraints, and articular geometry. Journal of Orthopaedic Research 1996;14:986–93. Karduna AR, Williams GR, Williams JL, Iannotti JP. Glenohumeral joint translations before and after total shoulder arthroplasty. The Journal of Bone and Joint Surgery 1997;79A(8):1166–74. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. Loehr JF, Helmig P, Søjberg JO, Jung A. Shoulder instability caused by rotator cuff lesions: an in vitro study. Clinical Orthopaedics and Related Research 1994;304:84–90. MacConaill MA. The movements of bones and joints. The signiﬁcance of shape. The Journal of Bone and Joint surgery 1953;35B(2):290–7 [Chapter 5]. Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M. Reliability of the PEDro scale for rating quality of randomized controlled trials. Physical Therapy 2003;83:713–21 Retrieved January 14, 2004 from the World Wide Web: http://www.ptjournal.org/includes/printit.cfm, p. 1–9. Maitland GD. Vertebral manipulation. Oxford: Butterworth-Heinemann; 1998. p. 3–13. [Chapter 1]. Mays N, Pope C. Quality in qualitative health research. In: Pope C, Mays N, editors. Qualitative research in health care. London: British Medical Journal publishing group; 1999 [Chapter 9]. McGlynn FJ, Caspari RB. Arthroscopic ﬁndings in the subluxating shoulder. Clinical Orthopaedics and Related Research 1984;183: 173–8. McMahon PJ, Debski RE, Thompson WO, Warner JJP, Fu FH, Woo SLY. Shoulder muscle forces and tendon exursions during glenohumeral abduction in the scapular plane. Journal of Shoulder and Elbow Surgery 1995;4(3):199–208. Meister K. Injuries to the shoulder in the throwing athlete. Part one: Biomechanics/pathophysiology/classiﬁcation of injury. The American Journal of Sports Medicine 2000;28(2):265–75. Moher D, Jadad AR, Tugwell P. Assessing the quality of randomized controlled trials. International Journal of Technology Assessment in Health Care 1996;12(2):195–208. Mulligan BR. Manual Therapy: ‘‘NAGS’’, ‘‘SNAGS’’, ‘‘MWMS’’, etc, 4th edn. Wellington: Plane View Services Limited; 1999. Novotny JE, Nichols CE, Beynnon BD. Normal kinematics of the unconstrained glenohumeral joint under coupled moment loads. Journal of Shoulder and Elbow Surgery 1998;7:629–39. Oxman AD, Cook DJ, Guyatt GH. User‘s guides to the medical literature: IV. How to use an overview. Journal of the American Medical Association 1994;272(17):1367–71. Ozaki J. Glenohumeral movements of the involuntary inferior and multidirectional instability. Clinical Orthopaedics and Related Research 1989;238:107–11. Paletta GA, Warner JJP, Warren RF, Deutsch A, Altchek DW. Shoulder kinematics with two plane X-ray evaluation in patients

11

with anterior instability or rotator cuff tearing. Journal of Shoulder and Elbow Surgery 1997;6(6):516–27. Panjabi MM. The stabilising system of the spine: part I–Function, dysfunction, adaptation and enhancement. Journal of Spinal Disorders 1992;5:383–9. ‘‘PEDro: Frequently asked questions’’. PEDro 2003: 1–4. 5 March 2003: http://www.pedro.fhs.usyd.edu.au/FAQs/faqs.htm Poppen NK, Walker PS. Normal and abnormal motion of the shoulder. The Journal of Bone and Joint Surgery 1976;58A(2):195–201. Poppen NK, Walker PS. Forces at the glenohumeral joint in abduction. Clinical Orthopaedics and Related Research 1978;135: 165–70. Rhoad RC, Klimkiewicz JJ, Williams GR, Kesmodel SB, Udupa JK, Kneeland JB, et al. A new in vivo technique for three-dimensional shoulder kinematics analysis. Skeletal Radiology 1998;27:92–7. Scholten-Peeters GGM, Verhagen AP, Bekkering GE, Van der Windt DAWM, Barnsley L, Oostendorp RAB, et al. Prognostic factors of whiplash-associated disorders: a systematic review of prospective cohort studies. Pain 2003;104:303–22. Thompson WO, Debski RE, Boardman III ND, Taskiran E, Warner JJ, Fu FH, et al. A biomechanical analysis of rotator cuff deﬁciency in a cadaveric model. The American Journal of Sports Medicine 1996;24(3):286–92. Tugwell P, Brooks P, Wells G, Davies J, Shea B, De Bie R, et al. Cochrane Musculosceletal Group. In The Cochrane Library, issue 2. Oxford: Update software; 2003. Vaesel MT, Olsen BS, Søjbjerg JO, Helmig P, Sneppen O. Humeral head size in shoulder arthroplasty: a kinematic study. Journal of Shoulder and Elbow Surgery 1997;6(6):549–55. Verhagen AP, De Vet HCW, De Bie RA, Kessels AGH, Boers M, Bouter LM, et al. The Delphi list: a criteria list for quality assessment of randomized clinical trials for conducting systematic reviews developed by Delphi consensus. Journal of Clinical Epidemiology 1998;51(12):1235–41. Von Eisenhart-Rothe RMO, Jager A, Englmeier K-H, Vogl TJ, Graichen H. Relevance of arm position and muscle activity on three-dimensional glenohumeral translation in patients with traumatic and atraumatic shoulder instability. The American Journal of Sports Medicine 2002;30(4):514–22. Williams GR, Wong KL, Pepe MD, Tan V, Silverberg D, Ramsey ML, et al. The effect of articular malposition after total shoulder arthroplasty on glenohumeral translations, range of motion, and subacromial impingement. Journal of Shoulder and Elbow Surgery 2001;10(5):399–409. Winters JC, Jorritsma W, Groenier KH, Sobel JS, Meyboom-de Jong B, Arendzen HJ. Treatment of shoulder complaints in general practice: long term results of a randomised, single blind study comparing physiotherapy, manipulation and corticosteroid injection. British Medical Journal 1999;318(7195):1395–6. Woolf H. Evidence-based medicine and practice guidelines: an overview. JMCC 2000;7(4):362–7. Wuelker N, Korell M, Thren K. Dynamic glenohumeral joint stability. Journal of Shoulder and Elbow Surgery 1998;7:43–52. Wuelker N, Schmotzer H, Thren K, Korell M. Translation of the glenohumeral joint with simulated active elevation. Clinical Orthopaedics 1994;309:193–200.

ARTICLE IN PRESS

Manual Therapy 12 (2007) 12–16 www.elsevier.com/locate/math

Original article

Work-related thumb pain in physiotherapists is associated with thumb alignment during performance of PA pressures Anne Wajona,b,, Louise Adaa, Kathryn Refshaugea a

School of Physiotherapy, The University of Sydney, NSW, Australia Hand Therapy at Hornsby, 2/49, Palmerston Road, Hornsby, NSW 2077, Australia

b

Received 26 October 2004; received in revised form 9 April 2005; accepted 20 September 2005

Abstract Pain is common in the thumbs of physiotherapists. The purpose of this observational study was to investigate whether there is an association between the alignment of the thumb during performance of postero-anterior (PA) pressures and the presence of thumb pain. One hundred and twenty-nine physiotherapists who attended the Musculoskeletal Physiotherapy Association Conference (2003) participated. After providing a history of any work-related thumb pain, participants applied a PA pressure mimicking the technique they would use on a cervical spine, while the position of their metacarpophalangeal (MP) and interphalangeal (IP) joints was photographed. There was an association (po0.05) between work-related thumb pain and alignment of the thumb during performance of PA pressures: participants who were able to maintain their MP and IP joints in extension were less likely to report pain. These ﬁndings serve as a guide to the safe performance of mobilization techniques, both for beginning practitioners and for experienced therapists complaining of thumb pain. r 2006 Elsevier Ltd. All rights reserved. Keywords: Thumb; Pain; ‘‘physical therapy (specialty)’’; Manipulation; Spinal; Physical therapy techniques; Musculoskeletal manipulations

1. Introduction Occupational musculoskeletal pain in physiotherapists is extremely common in Australia. Cromie et al. (2000) surveyed 824 physiotherapists and found that 91% of respondents had experienced work-related musculoskeletal pain or discomfort at some time in their working lives. West and Gardner (2001) surveyed 445 physiotherapists and reported that 40% had experienced musculoskeletal pain (lasting more than 3 days) during the previous year, with the hand being the second most prevalent site. Furthermore, they found 82% of their respondents with hand pain reported that Corresponding author. Hand Therapy at Hornsby 2/49, Palmerston Road, Hornsby, NSW 2077, Australia. Tel.: 61 2 9482 5522; fax: 61 2 6482 5533. E-mail address: [email protected] (A. Wajon).

1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2005.09.003

performing manual therapy techniques was a risk factor. Wajon and Ada (2003) attempted to determine which particular manual therapy technique aggravated pain in the thumbs of musculoskeletal physiotherapists, and found that central postero-anterior (PA) vertebral mobilizations using thumb tips aggravated thumb pain in 85% of their respondents with pain. The technique of central PA vertebral mobilizations using thumb tips is a rhythmical oscillatory passive accessory movement performed on the patient’s spine (Watson and Burnett 1990). The vertebral movement is produced by applying an anteriorly directed force to the spinous process of the prone patient (Maher & Adams 1995) using both thumb tips. It is recommended that the thumbs are in contact with each other and that the pressure is applied by the arms combined with the trunk (Maitland, 1986). Positioning the MP and IP joints in extension during the application of force should

ARTICLE IN PRESS A. Wajon et al. / Manual Therapy 12 (2007) 12–16

13

encourage the direct transfer of pressure to more proximal joints, reducing the tendency for a zigzag collapse of the polyarticular chain (Nordin and Frankel 1989) with associated joint subluxation and pain. Observation of physiotherapists performing this technique reveals that some therapists are able to comfortably position their MP and IP joints in extension. However, many others ﬁnd this impossible due to hypermobility of their MP and/or IP joints, and their position varies from one of MP hyperextension and IP ﬂexion to one of MP ﬂexion and IP hyperextension. These therapists may be subject to increased pain, as Moulton et al. (2001) have identiﬁed MP joint hypermobility to be associated with eventual osteoarthritis at the CMC joint. It was the aim of this paper to determine if there is an association between the position of the MP and IP joints of the thumb during the performance of PA pressures, and work-related thumb pain.

2. Methods This observational study involved measuring a cross-section of the population of musculoskeletal physiotherapists. Participants were recruited from physiotherapists attending the Musculoskeletal Physiotherapy Australia (MPA) 13th Biennial Conference (2003), with no limit on age, sex, or years of clinical experience. The study was approved by the relevant university ethics committee and participants gave informed consent prior to data collection. 2.1. Measurement of work-related thumb pain Those physiotherapists’ who reported aggravation of thumb pain by performing manual therapy to the cervical spine in the last 12 months, were asked for a detailed history of their pain presentation, including the location, severity (0–10 pain scale), frequency and duration of their symptoms, and the most aggravating technique. 2.2. Measurement of thumb alignment during performance of PA pressures Participants were asked to perform a PA pressure while being photographed. In order to standardize the procedure, they were asked to press on a dense foam pad (3.2 mm) adhered to a wooden block (height 5 cm, width 10 cm), which was in turn resting on a force platform1 (Fig. 1). The participant was asked to apply a PA pressure to a force of 40 N. A pilot study with an 1 E-Link Computerized Rehabilitation & Clinical Testing System (http://www.biometricsltd.com/exercise.htm)

Fig. 1. Participant performing a PA pressure with their thumbs postured in MP and IP joint extension (Alignment 5 (EE)), the alignment least associated with thumb pain.

instrumented couch (Chiradejnant et al. 2001) had previously determined that 40 N was an average amount of force used by experienced musculoskeletal physiotherapists when performing a Grade III central PA pressure on the cervical spine. A digital photograph was taken of the thumbs as the participant mimicked their clinical technique of performing a PA pressure on the cervical spine. The photograph was taken at a ﬁxed distance to the lateral aspect of the thumbs. This has been shown to be a valid technique to determine the alignment of the hand in a non-invasive way (Vergara et al. 2003). The photographs were subsequently downloaded to a computer. To eliminate bias, the position of the joints of the thumb was determined by a measurer blinded to thumb pain status of participants. The positions were categorized as hyperextended, extended or ﬂexed, resulting in nine possible combinations of joint positions at the MP and IP joints (Table 1). Test–retest reliability was examined several months later, with 95% exact agreement. 2.3. Statistical analysis There was no pattern of similarity between the alignment of each pair of right and left thumbs, conﬁrmed by the absence of any correlation between them (r ¼ 0:25), and so they were considered independent of each other. Uncommon thumb alignments (those used by o5% of participants) were excluded from analysis. Chi square was then calculated to determine whether there was an association between the presence of work-related thumb pain and thumb alignment during performance of PA pressures.

ARTICLE IN PRESS 14

A. Wajon et al. / Manual Therapy 12 (2007) 12–16

Table 1 Number and percentage of thumbs which were categorized into each thumb alignment according to the positions of the metacarpophalangeal (MP) and interphalangeal (IP) joints during performance of PA pressures Thumb alignment

MP joint position

IP joint position

Code

Number

%

1 2 3 4 5 6 7 8 9

Hyperextension Hyperextension Hyperextension Extension Extension Extension Flexion Flexion Flexion

Hyperextension Extension Flexion Hyperextension Extension Flexion Hyperextension Extension Flexion

HH HE HF EH EE EF FH FE FF

2 4 59 11 20 20 57 69 16

0.7 1.5 22.9 4.3 7.8 7.8 22.1 26.7 6.2

FF

HE

3. Results The majority of the 129 participants were female (64%), right-hand dominant (94%), with a mean age of 42 (SD 7.9) years. The mean number of hours worked per week was 29 (SD 16.6) and participants had a mean of 17.3 (SD 7.3) years of experience in musculoskeletal physiotherapy. Eighty-two (64%) participants complained of having pain in the preceding 12 months in one or both thumbs, aggravated by performing spinal manipulative therapy techniques.

FE

EE FH EF

3.1. Description of work-related thumb pain In total, 123 of the 258 (48%) thumbs were reported to have been aggravated by performance of PA pressures to the cervical spine in the preceding 12-month period. Of the 82 participants with thumb pain, 75 (91%) reported pain in their right thumb and 48 (59%) pain in their left, including 41 (50%) with pain in both thumbs. The mean score for severity of thumb pain that was aggravated by performing manual therapy techniques was 4.2/10 (SD 1.9). The majority (67%) reported pain less frequently than once per month, but 21% of thumbs were painful more than once per day. The mean duration of each episode of pain was 3.2 days (SD 9.6) per thumb. Fiftyeight (47%) thumbs were painful at the MP joint, 45 (37%) thumbs were painful at the trapeziometacarpal joint, 7 (6%) thumbs were painful at both the trapeziometacarpal and MP joints, and 13 (11%) thumbs were painful at the IP joint. Of the respondents who reported thumb pain, 43% reported the unilateral PA pressure to be the most aggravating technique. 3.2. Description of thumb alignment during performance of PA pressures There were nine possible thumb alignments according to the positions of the MP and IP joints (Table 1).

Fig. 2. The relative contribution of the common thumb alignments to the total Chi square.

The most common thumb alignments were Alignment 3 (HF) used by 23%, Alignment 7 (FH) used by 22%, and Alignment 8 (FE) used by 27% of participants. The least common thumb alignments were Alignment 1 (HH), Alignment 2 (HE) and Alignment 4 (EH), each used by less than 5% of participants, and these were removed from subsequent analysis. 3.3. Association between work-related thumb pain and thumb alignment during performance of PA pressures There was a signiﬁcant association between the presence of work-related thumb pain and thumb alignment during performance of PA pressures (X 2 ¼ 12:49, df ¼ 5, p ¼ 0:03). This association was signiﬁcant because there was a large discrepancy between the number of participants reporting ‘pain’ and the number of participants ‘expected to report pain’, for each alignment. The relative contribution of each alignment to the total Chi square is demonstrated in Fig. 2. Alignment 5, with MP and IP joint extension (Fig. 1), contributes almost 50% to the total Chi square. Furthermore, Alignment 5 made this large contribution to the Chi

ARTICLE IN PRESS A. Wajon et al. / Manual Therapy 12 (2007) 12–16

square because there were far fewer participants reporting pain than expected.

4. Discussion This is the ﬁrst published study to identify the alignment of the thumb as an intrinsic factor in the development of work-related thumb pain in physiotherapists. Whilst many have reported the incidence of thumb pain in physiotherapists (Cromie et al. 2000, West and Gardner, 2001, Snodgrass and Rivett, 2002, Branton et al., 2003, Wajon and Ada, 2003), few have identiﬁed factors associated with its onset. In their casecontrol study of 44 physiotherapists, Snodgrass et al. (2003) categorized photographs of physiotherapists performing a standard PA pressure. They rated the hand positions according to whether the thumb was supported by the index ﬁnger, by the other thumb at the MP joint, or whether the thumb pads overlapped. They did not report the alignment of the MP and IP joints during the performance of PA pressures. This is the ﬁrst paper that has identiﬁed an association between the alignment of the thumb during performance of PA pressures and work-related thumb pain in physiotherapists. The trapeziometacarpal joint is a common site of pain in the hand (Armstrong et al. 1994), with classic patterns of joint surface wear and arthritis resulting from compression loading and shear forces applied to the joint (Pellegrini, 1991, Pellegrini et al., 1993, Ateshian et al., 1995, Uchiyama et al., 1999). Moulton et al. (2001) associated these patterns of joint surface wear with the position of the MP joint, and concluded that loading through a hyperextended MP joint would encourage reciprocal ﬂexion of the metacarpal shaft and dorsal subluxation of its base, contributing to pain in the thumb. We were therefore surprised to ﬁnd that those physiotherapists who hyperextended their MP joint during the application of PA pressure did not demonstrate an increase in the presence of work-related thumb pain. Perhaps this is due to the force being applied longitudinally through the tips of the thumb, whereas other studies which analysed patterns of trapeziometacarpal joint contact, used lateral pinch as their model (Brand, 1985, Pellegrini et al., 1993, 1996, Moulton et al., 2001). Extension of the MP and IP joints during performance of PA pressures was least likely to be associated with thumb pain. This alignment minimises tendon and joint moment arms (Brand, 1985), promoting optimal joint loading and maximizing the surface area of joint contact (McConnell, 1993). The compressive force is then transmitted longitudinally up the thumb to larger more proximal joints, minimizing deforming forces which could cause collapse of the polyarticular chain.

15

Various authors have reported the tendency for musculoskeletal physiotherapists to reduce their working hours, leave the workforce, or change their area of specialization due to pain and instability in their thumbs (Reglar and James, 1999, Snodgrass and Rivett, 2002; Wajon and Ada, 2003). It is proposed that positioning the MP and IP joints in extension during the performance of PA pressures should reduce the likelihood of the development of thumb pain in physiotherapists. It may be useful for those physiotherapists unable to change their alignment to be ﬁtted with a thermoplastic splint to control either or both the MP and IP joints (Colditz, 2000; Poole and Pellegrini, 2000; Wajon, 2000; Veldhoven and Lede, 2002; Fess et al., 2005).

5. Summary This study found a signiﬁcant association between the alignment of the thumb during performance of PA pressures and thumb pain. These ﬁndings have important clinical implications for the practice and teaching of manual therapy techniques. Enhancing the longitudinal transmission of force to more proximal joints by maintaining the MP and IP joints in extension during the performance of PA pressures appears to be associated with less thumb pain in physiotherapists.

Acknowledgements We are grateful to Surgical Synergies for lending us the Biometrics Force Plate, and allowing us to set up this study at their stand for the duration of the conference. In particular, we would like to thank Jenny Crane for helping us incorporate the E-link software with the study. References Armstrong A, Hunter J, Davis T. The prevalence of degenerative arthritis of the base of the thumb in post-menopausal women. Journal of Hand Surgery 1994;19B(3):340–1. Ateshian G, Ark J, Rosenwasser M, Pawluk R, Soslowsky L, Mow V. Contact areas in the thumb carpometacarpal joint. Journal of Orthopaedic Research 1995;13:450–8. Brand PW. Clinical mechanics of the hand. St. Louis, MO: C.V. Mosby Company; 1985. p.342. Branton A, Egan K, Conroy R, Horgan F. A study examining the prevalence of thumb signs and symptoms among physiotherapists working in private practice in Ireland. 14th world congress physical therapists, Barcelona, 2003. Chiradejnant A, Maher C, Latimer J. Development of an instrumented couch to measure forces during manual physiotherapy treatment. Manual Therapy 2001;64(4):229–34. Colditz J. The biomechanics of a thumb carpometacarpal immobilisation splint. Journal of Hand Therapy 2000;13(3):228–35.

ARTICLE IN PRESS 16

A. Wajon et al. / Manual Therapy 12 (2007) 12–16

Cromie J, Robertson V, Best M. Work-related musculoskeletal disorders in physical therapists: prevalence, severity, risks, and responses. Physical Therapy 2000;80(4):336–51. Fess E, Gettle K, Philips C, Janson J. Hand and upper extremity splinting principles and methods. 3rd ed. St Louis, Missouri: Elselvier, Mosby; 2005. Maher C, Adams R. A psychophysical evaluation of manual stiffness discrimination. Australian Journal of Physiotherapy 1995;41(3):161–7. Maitland G. Vertebral manipulation. London: Butterworth & Co.; 1986. McConnell J. Promoting effective segmental alignment. In: Crosbie J, McConnell J, editors. Physiotherapy: foundations for practice. Jordan Hill, Oxford: Butterworth-Heinemann, Ltd; 1993. p. 172–94. Moulton M, Parentis M, Kelly M, Jacobs C, Naidu S, Pellegrini V. Inﬂuence of metacarpophalangeal joint position on basal jointloading in the thumb. Journal of Bone and Joint Surgery 2001;83A(5):709–16. Nordin M, Frankel V. Basic biomechanics of the musculoskeletal system. Philadelphia, USA: Lea & Febiger; 1989. Pellegrini V D. Osteoarthritis of the trapeziometacarpal joint: the pathophysiology of articular cartilage degeneration. II. Articular wear patterns in the osteoarthritic joint. Journal of Hand Surgery 1991;16A(6):975–82. Pellegrini V D, Olcott C W, Hollenberg G. Contact patterns in the trapeziometacarpal joint: the role of the palmar beak ligament. Journal of Hand Surgery 1993;18A(2):238–44. Pellegrini V D, Parentis M, Judkins A, Olmstead J, Olcott C. Extension metacarpal osteotomy in the treatment of trapeziometacarpal osteoarthritis: a biomechanical study. Journal of Hand Surgery 1996;21(1):16–23. Poole J, Pellegrini V. Arthritis of the thumb basal joint complex. Journal of Hand Therapy 2000;13(2):91–107.

Reglar P, James G. Thumb pain in physiotherapists: a preliminary study. British Journal of Therapy and Rehabilitation 1999;6(10):505–9. Snodgrass S, Rivett D. Thumb pain in physiotherapists: potential risk factors and proposed prevention strategies. Journal of Manual and Manipulative Therapy 2002;10(4):206–17. Snodgrass S, Rivett D, Chiarelli P, Bates A, Rowe L. Factors related to thumb pain in physiotherapists. Australian Journal of Physiotherapy 2003;49:243–50. Uchiyama S, Cooney W, Niebur G, An K- N, Linscheid R. Biomechanical analysis of the trapeziometacarpal joint after surface replacement arthroplasty. Journal Hand Surgery 1999;24(3):483–90. Veldhoven G, Lede PV. Therapeutical Hand Splints, a rational approach, vol. 2. Leuven, Belgium: Provan bvba; 2002. Vergara M, Sancho-Bru J, Perez-Gonzalez A. Description and validation of a non-invasive technique to measure the posture of all hand segments. Journal of Biomechanical Engineering 2003;125(6):917–22. Wajon A. The thumb strap splint for dynamic instability of the trapeziometacarpal joint. Journal of Hand Therapy 2000;13(3): 236–7. Wajon A, Ada L. Prevalence of thumb pain in physical therapists practicing spinal manipulative therapy. Journal of Hand Therapy 2003;16(3):237–44. Watson M, Burnett M. Equipment to evaluate the ability of physiotherapists to perform graded postero-anterior central vertebral pressure type passive movements of the spine by thumb pressure. Physiotherapy 1990;76(10):611–4. West D, Gardner D. Occupational injuries of physiotherapists in North and Central Queensland. Australian Journal of Physiotherapy 2001;47(3):179–86.

ARTICLE IN PRESS

Manual Therapy 12 (2007) 17–21 www.elsevier.com/locate/math

Original article

Repeatability of skin displacement and pressure during ‘‘inhibitory’’ vastus lateralis muscle taping J.U. McCarthy Perssona,, A.C.B. Hooperb, H.E. Fleminga a

School of Physiotherapy and Performance Science, University College Dublin, UCD Health Sciences Centre, Belfield, Dublin 4, Ireland b Centre for Sports Studies, School of Public Health and Population Studies, University College Dublin, Belfield, Dublin 4, Ireland Received 5 November 2004; received in revised form 18 August 2005; accepted 16 December 2005

Abstract The use of taping to modify pain and muscle activity has become common practice among musculoskeletal physiotherapists. The aim of this study was to evaluate the repeatability of two variables, skin displacement and pressure, produced by a standardized taping procedure designed to inhibit the vastus lateralis (VL) muscle in patellofemoral pain. Measurements were recorded in 10 healthy volunteers. The effects of the taping procedure were assessed on the two lower limbs of each individual, and on measurements taken on the same limb in ﬁve subjects on two different days. On two-way analysis of variance no signiﬁcant variable or interaction effect (Po0.05) was found. The coefﬁcient (limit) of repeatability demonstrated that 95% of the differences measured for skin displacement and pressure were less than 6% and 94% of their respective means. The absolute pressures found were without exception very small and not repeatable. The results demonstrated that the VL ‘‘inhibitory’’ taping procedure used produced a reproducible effect for skin displacement. The validity of this taping technique is discussed. r 2006 Elsevier Ltd. All rights reserved. Keywords: Taping; Skin displacement; Pressure; Repeatability

1. Introduction Despite debate about the efﬁcacy of the procedures, a number of muscle taping techniques have been developed in attempts to alter pain (McConnell, 2002; O’Leary et al., 2002) or muscle activity (Morin et al., 1997; Tobin and Robinson, 2000; Cools et al., 2002; Alexander et al., 2003). The use of rigid tape applied directly on the skin at right angles to the muscle ﬁbres of an underlying muscle was developed initially in the context of patellofemoral pain by McConnell (1992) and termed ‘‘inhibitory’’ taping. Tobin and Robinson (2000) examined the effects of taping the vastus lateralis (VL) muscle and demonstrated a decrease in the EMG activity of the VL in Corresponding author. Tel.: +353 17166517; fax: +353 17166501.

E-mail address: [email protected] (J.U. McCarthy Persson). 1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2005.12.002

normal subjects during a step down task. A number of studies have also examined the effects of taping the upper trapezius muscles; Morin et al. (1997) produced a decrease in the activity of the upper trapezius during an isometric muscle contraction, while Brown (2001) demonstrated a decrease in upper trapezius:lower trapezius EMG ratio during shoulder abduction. However, using the same taping technique, Cools et al. (2002) were unable to replicate an inhibitory effect on the upper trapezius muscle. Descriptions of the methods of tape application are subjective and variable. The application of tape needed to be ‘‘ﬁrm’’ (Cools et al., 2002; O’Leary et al., 2002), ‘‘snug’’ (Morin et al., 1997) or ‘‘under tension’’ (Alexander et al., 2003). Tobin and Robinson (2000) produced a ‘‘furrow’’ in the skin, McConnell (1996) created ‘‘a tuck or a fold in the skin’’ and O’Leary et al. (2002) described causing ‘‘a pitted, orange peel like appearance’’ of the skin. However, these studies described the tension of the

ARTICLE IN PRESS 18

J.U. McCarthy Persson et al. / Manual Therapy 12 (2007) 17–21

tape as producing a desired end effect on the skin without attempts to measure or objectively quantify. Other studies failed to give any description of the actual taping procedure but only stated that the person taping was trained in ‘‘McConnell taping techniques’’ (Wilson et al., 2003; Pfeiffer et al., 2004). Comparisons between such studies are therefore questionable. Therefore, studies of the effects of taping are of little consequence if the taping procedure itself cannot be replicated, and a repeatable mechanical effect produced. To date, the consistency and repeatability of taping techniques have not been established. The extent of skin displacement and the compressive pressure produced by taping have both been suggested as possible mechanisms of muscle inhibition (Morin et al., 1997; Tobin and Robinson, 2000). However, these parameters have not been quantiﬁed in previous studies. The aim of this study was to assess the repeatability of VL ‘‘inhibitory’’ taping in a clinical situation using a standardized technique.

zinc oxide tape (Leukotape-Beiersdorf, Milton Keynes, UK) were laid on top of the hypoallergenic tape from the anterior line extending over the lateral line without any tension being applied. Lines were drawn on the zinc oxide tape as they crossed over the reference lines on the skin of the thigh. The tape was carefully removed and reapplied starting anteriorly over the hypoallergenic tape. Tension was applied to the zinc oxide tape laterally and posteriorly with one hand. The lateral thigh tissues were collected with the other hand while applying a downward pressure with the thumb over the VL between the reference lines causing a furrow in the skin. The tension applied on the tape was standardized to cause a ‘‘skin roll’’ in front of the thumb with the same height as the width (20 mm) of the researchers thumb. A total of three zinc oxide tape strips were applied, starting with the most superior tape, followed by the middle, and ﬁnishing with the distal tape, each overlapping the other by one-third of the tape width. 2.3. Measurement of pressure

2. Methodology 2.1. Subjects Ten healthy subjects (3 male and 7 female) aged 19–34 years (mean 26.4) participated in the study. Informed written consent was received from each participant prior to the commencement of the study, which was approved by University College Dublin Human Research Ethics Committee. Subjects with previous injury or pathology of the knee, quadriceps muscles, skin or connective tissues of the region were excluded. The mean height of the subjects was 167 cm, mean body mass 65 kg and mean calculated body mass index 23.0 kg/m2. The mean skin fold thickness (48715.2 mm) was obtained from caliper measurements at four sites: biceps, triceps, subscapular and suprailiac and was within normal limits (Durnin and Womersley, 1974). Two reference lines were drawn longitudinally on each thigh: (1) an anterior line joined the anterior superior iliac spine (ASIS) to the midpoint of the superior border of the patella, and (2) a lateral line joined the greater trochanter to the lateral femoral epicondyle. The midpoint was marked on each line. The area was shaved and wiped with alcohol. The subject was positioned on the side with a pillow between the knees, which were ﬂexed to an angle of 301.

The downward pressure produced between the tape and the skin was measured using a force-sensing resistor (FSR) (Interlink electronics, Camarillo, CA, USA) with a variability of 5.8% (McCarthy Persson, unpublished data) (Fig. 1). The calibrated FSR was inserted underneath the zinc oxide tape, between two slips of paper at the middle of the application between the anterior attachment and the skin furrow formed. The FSR displayed an initial plasticity in response to a static load. Pressure recordings were therefore taken after 10 min, the time taken for the sensor to give a stable reading. 2.4. Measurement of skin displacement The distance between the new position of the reference line on the zinc oxide tape and the original line on the skin was measured after 10 min using a

2.2. Taping procedure Two lengths of ﬂexible hypoallergenic tape, 5 cm in width (Fixomull-Beiersdorf, Milton Keynes, UK) were applied superior and inferior to the previously marked midpoints of the anterior and lateral reference lines extending past both lines by 2 cm. Three strips of 3.8 cm

Fig. 1. The force-sensing resistor (FSR) (Interlink electronics, Camarillo, CA, USA) used for the study of pressure (thickness 0.2 mm, area 4.15 cm2).

ARTICLE IN PRESS J.U. McCarthy Persson et al. / Manual Therapy 12 (2007) 17–21

caliper (Harpenden, Burgess Hill, UK) (Fig. 2). Each measurement was repeated three times and the mean calculated. This parameter represented the distance the skin had been displaced due to the tension of the tape application. 2.5. Data collection All procedures and measurements were carried out by the same researcher. The taping procedure and measurements were made on both limbs in 10 subjects. The FSR failed in one instance so that recordings of pressure from only one limb were collected in that subject. The taping and measurements was repeated 2 days later in ﬁve subjects to enable analysis of the repeatability of the taping procedure on the same limb within a subject. All subjects met the study inclusion criteria at the time of testing. 2.6. Statistical analysis Statistical analysis was carried out using Statistical Package for Social Sciences (SPSS) version 11.0 and Excel 2000. Data was normally distributed as conﬁrmed by the Kolmogorv–Smirnov test, thus allowing parametric tests to be employed. Group means were

19

calculated for each parameter along with their standard deviations. A two-way analysis of variance was performed to analyse the effects of taping the right and left leg, taping the same leg on two different days and the interaction between the day and side taped. The coefﬁcient (limit) of repeatability (CR) (Bland and Altman, 1986) was used to analyse the repeatability of the data. The 95% CR was given by 1.96 standard deviation of the differences of the two repeated measurements (Daly and Bourke, 2000).

3. Results No interactions were found between the day tested and the side taped. There were no signiﬁcant differences between the values of skin displacement or pressure measurements when measured on the right and left limbs or between measurements in the same limb on different days (P40.05) (Table 1). Calculation of the CR, demonstrated that 95% of the differences in skin displacement measured were less than 2.3 mm (6% of the mean) both between the right and the left limbs and in the same limb on different days. However, the CR for pressure measurements was not repeatable as indicating that 95% of the differences were less than 8.21 mmHg (94.1% of the total mean).

4. Discussion

Fig. 2. Measurement of skin displacement as shown by the distance between the original reference line on the skin (A), and its new position following taping (B).

This is the ﬁrst study to establish the repeatability of a clinically useful and objective measure of tape tension during an ‘‘inhibitory’’ vastus lateralis taping procedure. Standardization of taping procedures and their clear, detailed description are essential, both to permit valid comparisons between different studies and to facilitate the reﬁnement and modiﬁcation of techniques. In the present study, an objectively reproducible technique was used resulting in the production of a standardized height of skin fold when applying tension on the tape.

Table 1 Skin displacement and pressure produced by taping (Mean7SD) Day 1 (n ¼ 10 or 9a)

Day 2 (n ¼ 5)

Day 1+2

Signiﬁcance level

Distance (mm) Right Left

39.870.8 39.171.0

40.070.4 39.670.6

39.670.9

P40.05

Pressure (mmHg) Right Left

10.874.5* 7.376.5*

9.074.7 6.774.3

8.775.3

P40.05

a

n ¼ 9 due to failure of FSR during measurements of one limb.

ARTICLE IN PRESS 20

J.U. McCarthy Persson et al. / Manual Therapy 12 (2007) 17–21

The data collected in this study did not vary in a systematic way, i.e. differences between the measurements did not become larger as the measurements increased. The differences were random in nature, which led to the use of a statistical analysis for absolute repeatability rather than relative repeatability, which assumes consistency of the differences (Bruton et al., 2000). There were no signiﬁcant differences found between any of the measurements of skin displacements produced, using the taping technique described. The amount of skin displacement was found to be repeatable regardless of subject height, weight, sex or skin fold thickness. This study examined the intra-tester reliability of measurements comparing the left and the right leg in one session within the same subject. Comparison of the same leg measurements was not possible in the same session, as skin properties change due to viscoelasticity (Khatyr et al., 2004). A comparison of measurements on the same leg on two different days was therefore performed to ensure that no differences existed when comparing the right and left legs. There was no interaction found between the day and side measured. No signiﬁcant differences were found between measurements of pressure from the right and the left limb on the same day or on the same limb on different days. However, the pressures produced underneath the tape were not repeatable as the CR indicated that 95% of the differences were within 94.1% of the mean. Although the measurement of pressure at skin interfaces is notoriously difﬁcult (Patterson and Fisher, 1979; Ferguson-Pell et al., 1985), the absolute values found were, without exception, very small. Pressures of this order (o10 mmHg) do not appear to exert any compressive effect (Sockalingham et al., 1990), furthermore, pressure pain thresholds in the thigh have been found to be in excess of 300 kPa, i.e. 2250 mmHg (Ordeberg, 2004). It is therefore unlikely that the inhibitory taping procedure would have exerted sufﬁcient downward pressure underneath the tape to cause muscle inhibition as suggested by Tobin and Robinson (2000). MacGregor et al. (2005) have recently shown that a small controlled amount of tension on the skin, applied through a tape, can alter muscle activity. The results of the present study have shown that a repeatable mechanical effect on the skin can be produced by the use of a standardized technique of inhibitory taping. The measurement of skin displacement, as described in this study, is simple to quantify in the clinical situation. This will permit the conduct of further studies into the previously demonstrated inhibitory effects of the procedure on muscle activity and will permit valid assessments of the procedures and their clinical efﬁcacy. It is proposed initially to investigate the consistency of changes in EMG muscle activity with repeated applications of tape.

Acknowledgements This work was supported by a research grant from Chartered Physiotherapists in Manipulative Therapy, a special interest group of the Irish Society of Chartered Physiotherapists. Thanks to Dr. Catherine Blake and Dr. Deirdre Hurley-Osing for the statistical and editorial support.

References Alexander CM, Stynes S, Thomas A, Lewis J, Harrison PJ. Does tape facilitate or inhibit the lower ﬁbres of trapezius? Manual Therapy 2003;8(1):37–41. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. The Lancet 1986(8):307–10. Brown LM. The effect of taping the glenohumeral joint on scaulohumeral resing position and trapezius activity during abduction—A pilot study. 1st International Conference on Movement Dysfunction Conference proceeding, Edinburgh; 2001. p4. Bruton A, Conway JH, Holgate ST. Reliability: what is it, and how is it measured? Physiotherapy 2000;86(2):94–9. Cools AM, Witvrouw EE, Dannieels LA, Cambier DC. Does taping inﬂuence electromyographic muscle activity in he scapular rotators in healthy shoulders? Manual Therapy 2002;7(3):154–62. Daly LE, Bourke GJ. Interpretation and uses of medical statistics, 5th ed. Oxford: Blackwell Science; 2000. p. 397–9 [Chapter 12]. Durnin JVGA, Womersley J. Body fat assessment form total body density and its estimation from skinfold thickness on 481 men and women aged from 16 to72 years. British Journal of Nutrition 1974;32:77–97. Ferguson-Pell MW, Reddy NP, Stewart SFC, Palmierie V Cochran GVD. Measurement of physical parameters of the patient–support interface. In: Whittle M, Harris D, editors. Oxford medical engineering series: 5 biomechanical measurement in orthopaedic practice. Oxford: Clarendon Press; 1985 [Chapter 17]. Khatyr F, Imberdis C, Vescovo P, Varchon D, Lagarde JM. Model of the viscoelastic behavior of skin in vivo and study of anisotropy. Skin Research and Technology 2004;10:96–103. MacGregor K, Gerlach S, Mellor S, Hodges PW. Cutaneus stimulation from patella tape causes a differential increase in vasti muscle activity in people with patellofemoral pain. Journal of Orthopedic Research 2005;23:351–8. McConnell J. McConnell patellofemoral course—course notes. London: McConnell Institute; 1992. McConnell J. Recalcitrant chronic low back and leg pain—a new theory and different approach to management. Manual Therapy 2002;7(4):183–92. Morin GE, Tibero D, Austin G. The effect of upper trapezius taping in electromyographic activity in the upper and middle trapezius region. Journal of Sport Rehabilitation 1997(6):309–18. Ordeberg G. Characterization of joint pain in human OA. In: Novarthritis Foundation Symposium, editor. Osteoarthritic joint pain. Chichester: Wiley; 2004. p. 105–21. O’Leary S, Carroll M, Mellor R, Scott A, Vicenzino B. The effect of soft tissue deloading tape on thoracic spine pressure pain thresholds in asymptomatic. Manual Therapy 2002;7(3):150–3. Patterson RP, Fisher SV. The accuracy of electrical transducers for the measurement of pressure applied to the skin. IEEE Transactions on Biomedical Engineering BME 1979;26(8):450–6. Pfeiffer RP, DeBeliso M, Shea KG, Kelley L, Irmischer B, Harris C. Kinematic MRI assessment of McConnell taping before and after exercise. American Journal of Sports Medicine 2004;32(3):621–8.

ARTICLE IN PRESS J.U. McCarthy Persson et al. / Manual Therapy 12 (2007) 17–21 Sockalingham S, Barbenel JC, Queen D. Ambulatory monitoring of the pressures beneath compression bandages. CARE Science and Practice 1990;8(2):75–9. Tobin S, Robinson G. The effect of vastus lateralis inhibition taping technique on vastus lateralis and vastus medialis obliquus activity. Physiotherapy 2000;86(4):173–83.

21

Wilson T, Carter N, Thomas G. A multicenter, single-masked of medial, neutral, and lateral patellar taping in individuals with patellofemoral pain syndrome. Journal of Orthopaedic and Sports Physical Therapy 2003;33(8):437–48.

ARTICLE IN PRESS

Manual Therapy 12 (2007) 22–28 www.elsevier.com/locate/math

Original article

Cervicocephalic kinesthetic sensibility in young and middle-aged adults with or without a history of mild neck pain$ C.-C. Tenga, H. Chaia, D.-M. Laib, S.-F. Wanga, a

Graduate Institute and School of Physical Therapy, College of Medicine, National Taiwan University, No. 17, Xu-Zhou Road, Zhong Zheng District, Taipei, 100, Taiwan, ROC b Department of Surgery, National Taiwan University Hospital, National Taiwan University, Taipei, Taiwan, ROC Received 28 December 2004; received in revised form 16 January 2006; accepted 3 February 2006

Abstract Previous research has shown that there is no signiﬁcant relationship between the degree of structural degeneration of the cervical spine and neck pain. We therefore sought to investigate the potential role of sensory dysfunction in chronic neck pain. Cervicocephalic kinesthetic sensibility, expressed by how accurately an individual can reposition the head, was studied in three groups of individuals, a control group of 20 asymptomatic young adults and two groups of middle-aged adults (20 subjects in each group) with or without a history of mild neck pain. An ultrasound-based three-dimensional coordinate measuring system was used to measure the position of the head and to test the accuracy of repositioning. Constant error (indicating that the subject overshot or undershot the intended position) and root mean square errors (representing total errors of accuracy and variability) were measured during repositioning of the head to the neutral head position (Head-to-NHP) and repositioning of the head to the target (Head-to-Target) in three cardinal planes (sagittal, transverse, and frontal). Analysis of covariance (ANCOVA) was used to test the group effect, with age used as a covariate. The constant errors during repositioning from a ﬂexed position and from an extended position to the NHP were signiﬁcantly greater in the middle-aged subjects than in the control group (b ¼ 0:30 and b ¼ 0:60, respectively; Po0:05 for both). In addition, the root mean square errors during repositioning from a ﬂexed or extended position to the NHP were greater in the middle-aged subjects than in the control group (b ¼ 0:27 and b ¼ 0:49, respectively; Po0:05 for both). The root mean square errors also increased during Head-to-Target in left rotation (b ¼ 0:24; Po0:05), but there was no difference in the constant errors or root mean square errors during Head-to-NHP repositioning from other target positions (P40:05). The results indicate that, after controlling for age as a covariate, there was no group effect. Thus, age appears to have a profound effect on an individual’s ability to accurately reposition the head toward the neutral position in the sagittal plane and repositioning the head toward left rotation. A history of mild chronic neck pain alone had no signiﬁcant effect on cervicocephalic kinesthetic sensibility. r 2006 Elsevier Ltd. All rights reserved. Keywords: Age; Neck proprioception; Neck repositioning test

Abbreviations: H, head attachment with a triple marker; S, shoulder cap with a reference triple marker; U, miniature ultrasound transmitters; T, transducer sensor stand; M, microphones of the transducer sensor for sound wave; B, basic unit (A/D converter) of the ultrasound motion analysis; Com, computer system for real time monitoring of the neck motions $ Study approval was obtained from the Institutional Human Study and Ethics Committee, National Taiwan University Hospital. Corresponding author. Tel.: +11 886 233228139; fax: +11 886 233228161. E-mail address: [email protected] (S.-F. Wang). 1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.02.003

1. Introduction The number of individuals with cervical dysfunction has markedly increased in recent years. The prevalence of neck pain among adults has been reported to be 12–34% of the total population, with approximately 10–19% of the total population having such pain for more than 6 months (Bovim et al., 1994; Dvorak, 1998; Cote et al., 2000). Furthermore, for patients with

ARTICLE IN PRESS C.-C. Teng et al. / Manual Therapy 12 (2007) 22–28

neck-shoulder pain, the recurrence rate is high (Waling et al., 2002; Cote et al., 2004), and the condition frequently becomes chronic. The pathogenesis of chronic nontraumatic neck pain is still unclear. Several investigators have reported no signiﬁcant relationship between the degree of structural pathology and cervical pain (Boden et al., 1990; Marchiori and Henderson, 1996; Harrison et al., 2002). Thus, we sought to evaluate the role of sensoromotor dysfunction in the pathogenesis of neck pain (Sterling et al., 2003a, b). It has been demonstrated that proper function of the head–neck system relies on proprioceptive information provided from receptors in the zygapophysial joints and small intrinsic muscles (Cordo et al., 1995). Proprioception, including joint positioning sense, protects the joint by regulating joint stiffness through the activation of mechanoreceptors and the muscle spindle system (Johansson et al., 1991; Solomonow et al., 1998). A frequently adapted method of determining cervical positioning sense (Revel et al., 1991, 1994; Heikkila and Anstrom, 1996; Loudon et al., 1997; Heikkila and Wenngren, 1998) was designed to examine the subject’s ability to reposition the head after it had been moved away from the reference position, otherwise known as the cervicocephalic kinesthetic sensibility test. Despite the high prevalence of nontraumatic neck pain in middle-aged adults, limited information is available about impaired cervicocephalic kinesthetic sensibility in this population. Only one study, by Rix and Bagust (2001), has been carried out with individuals with non-traumatic chronic neck pain; these authors found decreased kinesthetic sensibility only during repositioning of the head to a zero position (straight ahead) from a ﬂexed position. However, it is unknown whether the decreased cervicocephalic kinesthetic sensibility was related to the presence of current or previous pain. The ageing process is another important factor related to proprioception. Age-related reduction in proprioception in weight-bearing joints, such as the knee or ankle joint, of asymptomatic adults has been documented extensively (Skinner et al., 1984; Gilsing et al., 1995; Robbins et al., 1995; Pai et al., 1997). However, no agerelated increases in cervical repositioning error were reported in asymptomatic adults. Interestingly, Heikkila and Wenngren (1998) reported a positive correlation between repositioning errors and age in patients who were 18–66 years old and had whiplash syndrome. This ﬁnding implies that age is a factor in the pathogenesis of chronic neck pain. The purpose of this study was to determine whether age and previous neck pain experience affected the cervicocephalic kinesthetic sensibility by evaluating repositioning errors in middle-aged adults with and without a history of neck pain. The hypotheses of this research included (1) a history of neck pain contributes

23

to the alteration of cervicocephalic kinesthetic sensibility and (2) cervicocephalic kinesthetic sensibility decreases with increasing age.

2. Methods 2.1. Participants The study included 40 middle-aged adults (range, 45–65 years) who had no current cervical pain; 20 of whom had a history of chronic mild neck pain and 20 of whom had no history of neck pain; 20 asymptomatic young adults (range, 18–30 years) served as a control group. All subjects reported that they had no neck pain at the present time. A history of neck pain was deﬁned as pain for at least 6 months during the past years. To be considered asymptomatic, a subject could not have had any previous treatment for neck pain, and no current neck pain. Exclusion criteria included the following: traumatic spinal injury, whiplash associated disorder; central nervous system impairment, demonstrated by paraethesia; vestibular impairment, demonstrated by vertigo, dizziness, or motor imbalance; neck pain induced by cervical motion in the range tested for the study; neck disability index of more than 15, an index below 15 indicates mild disability (Vernon and Mior, 1991); antigravity neck muscle weakness demonstrated to be less than fair (Hislop and Montgomery, 2002) for the purpose to exclude subjects could not actively moving his/her head against gravity freely; and limitation in the cervical range of motion below two standard deviations compared with previously established means for asymptomatic middle-aged adults (Wang et al., 2005). Participants completed a questionnaire to provide basic demographic data (age, gender, weight, and height), medical history of hypertension and diabetes, most frequently adopted posture during the daytime (sitting or standing; Table 1). The reason for including this information is that posture might relate to the muscle function; and hypertension and diabetes might potentially be a risk factor for changes of sensory threshold. Due to the small case number of subject with hypertension and diabetes, these factors are listed as demographic data and not for statistical analysis. All participants signed a written consent form prior to participating in the experiment. The study was approved by the local Institutional Human Study and Ethics Review Board. 2.2. Instrumentation The range of neck motion was measured with an ultrasound-based coordinate measuring system (CMS 70P, Zebris Medizintecknik GmbH, Tuebingen,

ARTICLE IN PRESS 24

C.-C. Teng et al. / Manual Therapy 12 (2007) 22–28

Table 1 Basic data of participants in asymptomatic young adults as control (C), asymptomatic middle-aged adults with no history of neck pain (NP), and middle-aged adults with history of neck pain (P). (Mean7SD)

Age (years) Gender (male/female) Height (cm) Weight (kg) Posture during daytime (sitting/standing/non-speciﬁc) Hypertension Diabetes

Fig. 1. The experimental setup of the ultrasonic based motion analysis system.

Germany). This motion analysis system includes data acquisition software (WinData v2.11), two sets of triple markers on the head attachment and the shoulder cap, a transducer sensor stand for the sound wave, and a computer processing basic unit system (Wang et al., 2005). The triple marker set includes three ultrasound transmitters with frequencies of 35 KHz that were used to measure the kinematics of cervical motions. The sampling frequency of each sound wave during cervical motion was set at 25 Hz (Fig. 1). The intra-class correlation coefﬁcient (ICC) for intra-session test–retest reliability ranged between 0.85 and 0.95. The intersession reliability ranged from 0.58 and 0.88 (Wang et al., 2005). The validity of CMS has been demonstrated by comparing with the gravity-based goniometer and showed signiﬁcant correlation (R2 ranging from 0.66 to 0.88 in a regression model, Wang et al., 2005). 2.3. Testing procedure The subjects were asked to sit upright in a comfortable position and look straight ahead to be determined as the neutral head position (NHP). Thus, it is a subjectadopted head zero position and may be inﬂuenced by the head to shoulder relationship. However, during the reposition test the shoulder position was not allowed to move. Therefore, the inﬂuence of shoulder position during repositioning test is controlled. The investigator

(C) N ¼ 20

(NP) N ¼ 20

(P) N ¼ 20

21.973.9 11/ 9 160.5721.9 58.579.1 20/0/0 0 0

54.575.0 3/17 157.374.5 56.474.7 11/4/5 1 0

58.875.7 6/ 14 157.776.4 61.279.2 9/1/10 3 1

then calibrated the motion analysis system to a neutral head position. The subject performed the maximum range of motion in each direction, which was recorded. For the cervicocephalic kinesthetic sensibility test (Lee et al., 2006), subjects ﬁrst held the head in the NHP and were told to close their eyes throughout the subsequent tests. The investigator moved the subject’s head at a speed of 351/s or slower to the predetermined target position, 65% of maximum range of motion. The speed of passive neck motion was controlled to be less than that of rotational vestibular stimulation (351/s), as speeds higher than this have been associated with signiﬁcant differences in vestibular function according to age (Goebel et al., 1994). Passive movement of the neck was guided by the real-time display of the neck angle on the computer screen of the motion analysis system (Fig. 1). The head was maintained in the target position for 3 s, and the subject was asked to remember that position for the following two repositioning tests. The ﬁrst test (Head- to-NHP) involved having the subject move the head to the NHP. Our procedure for this test differed from that used by Revel et al. (1991) in that in our study, the investigator did not reposition the subject’s head to the starting position between each trial of each movement test. Instead, the subjects actively moved the head from the target position back to the NHP. The subject kept the head in the NHP for 3 s and then moved the head to the target position (Head-toTarget), a second repositioning test adapted from Loudon et al. (1997). The two repositioning tests were performed in the sagittal, transverse, and frontal planes. The intra-rater reliability coefﬁcients demonstrated in general fair to excellent reliability of the root mean square error and the constant error (Lee et al., 2006). The standard error of measurement (SEM) of repositioning error test in our laboratory expressed by RMSE ranged from 0.7 to 2.61 and by CE ranged from 0.3 to 4.01 (Lee et al., 2006). 2.4. Data reduction and analysis The agreement between the predetermined angle (neutral or target position) and the observed angle

ARTICLE IN PRESS C.-C. Teng et al. / Manual Therapy 12 (2007) 22–28

during the repositioning test in each movement was expressed by constant error and root mean square error (Schmidt and Lee, 1999). A positive constant error indicated that the subject overshot the target position (that is, moved beyond it). The root mean square error indicated the total error, including accuracy for each trial and variability among the trials for each subject (Lee et al., 2006). The equations for determining constant error and root mean square error are presented in the Appendix A. 2.5. Statistical analysis The comparison of the age, body height, and weight among three groups was performed by analysis of variance (ANOVA). The group effect for the constant error and root mean square error for each repositioning test among the three groups of subjects was tested by analysis of covariance (ANCOVA), with age used as a covariate. The b values were used to indicate the group and age effect. All statistical analyses were performed using SPSS for Windows (Version 11.0, SPSS Inc., Chicago, Illinois, USA). The level of signiﬁcance was set at a ¼ 0:05.

3. Results The mean age of the subjects in the two middle-aged groups was signiﬁcantly higher than that of the subjects in the control group, but there was no difference between these two middle-aged groups (Table 1). Body height and weight did not signiﬁcantly differ among all three groups. 3.1. Group effect After controlling for age as a covariate, the ANCOVA indicated that there was no signiﬁcant group effect for constant error or for root mean square error during Head-to-NHP repositioning in all three cardinal planes (P40:05 for all; Table 2) or for Head-toTarget repositioning in all three planes (P40:05 for all; Table 3).

25

Table 2 Repositioning errors (degree) of Head-to-Neutral Head Position (NHP) repositioning tests of asymptomatic young adult as control (C), asymptomatic middle-aged adult with no history of neck pain (NP) and middle-aged adults with previous neck pain (P; n ¼ 20 for each group) Head- To- NHP (deg)

Control (C)

No pain (NP)

Pain (P)

Age b value

CE Flexed Extended Left-rotated Right-rotated Left-side-bended Right-side-bended

2.973.4 5.874.6 3.474.7 3.673.8 1.373.1 2.073.4

6.876.7 12.2710.5 4.777.3 5.576.8 3.773.3 4.874.2

10.275.7 11.676.8 7.676.8 5.1710.5 5.673.9 7.575.5

0.30* 0.60* 0.02 1.59 0.05 0.07

RMS Flexed Extended Left-rotated Right-rotated Left-side-bended Right-side-bended

4.372.2 7.173.6 5.273.1 4.972.7 3.171.6 3.272.5

9.174.2 13.279.4 7.374.8 7.674.9 4.472.7 5.773.2

10.775.4 12.276.5 9.175.2 10.377.4 6.073.6 7.875.3

0.27* 0.49* 0.08 0.28 0.03 0.11

ANCOVA was used to test the effect of groups with age as a covariate (positive number indicating overshooting; negative, undershooting; CE, constant error and RMS, root mean square; unit: degree; mean7SD, SD,standard deviation).* Po0:05.

Table 3 Repositioning errors of Head-to-Target repositioning tests of asymptomatic healthy young as control (C), healthy middle-aged adult with no history of neck pain (NP) and middle-aged adults with previous neck pain (P; n ¼ 20 for each group) Head-To-Target (deg)

Control (C)

No pain (NP)

Pain (P)

Age b value

CE Flexion Extension Left-rotation Right-rotation Left-side-bending Right-sidebending

0.5074.7 1.475.0 0.874.3 0.674.4 0.372.7 0.574.5

2.675.4 3.178.0 4.977.2 5.577.8 2.8704.0 1.673.3

4.076.1 4.876.3 3.877.2 5.077.2 1.674.5 1.775.5

0.18 0.32 0.05 0.22 0.04 0.122

RMS Flexion Extension Left-rotation Right-rotation Left-side-bending Right-sidebending

2.772.1 3.172.2 3.072.3 2.471.9 1.471.0 2.071.8

5.173.7 6.775.7 7.571.1 7.476.2 4.072.8 3.172.6

6.674.0 7.174.1 7.374.1 7.975.0 3.973.0 4.173.8

0.15 0.02 0.24y 0.23 0.05 0.12

3.2. Age effect The decrease in the accuracy of repositioning according to age was signiﬁcant only for (Head-to NHP) repositioning to the neutral position in the sagittal plane (Table 2). These ﬁndings indicate that age had an effect on the cervicocephalic kinesthetic sensibility in the sagittal plane but not in either the transverse or frontal planes. The root mean square error was greater in

ANCOVA was used to test the effect of groups with age as a covariate (positive number indicating overshooting; negative, undershooting; CE, constant error and RMS, root mean square; unit: degree; mean7SD, standard deviation) yPo0:05.

middle-aged subjects only for movement from the neutral position to left rotation (b ¼ 0:24, Po0:05; Table 3).

ARTICLE IN PRESS 26

C.-C. Teng et al. / Manual Therapy 12 (2007) 22–28

4. Discussion The present study is the ﬁrst, to our knowledge, to demonstrate that age is related to reduced cervicocephalic kinesthetic sensibility. Head-to-NHP reposition in the sagittal plane and the Head-to-Target reposition in the transverse plane were both less accurate for middle-aged adults than for the control subjects, regardless of the history of mild neck pain. The activemotion components involved in the repositioning testing require the activation of bilateral dorsal and ventral neck muscle at various layers (Vitti et al., 1973; Costa et al., 1990; Richmond et al., 1992; Mayoux-Benhamou et al., 1997). Given that muscle spindles signiﬁcantly contribute to the sense of body position (Gandevia, 1992; Cordo et al., 1995), the present results probably indicate the decreases of muscle spindle function as age increases in middle-aged adults. Because both the activated agonists and the lengthened antagonists contribute to the proprioceptive information (Ribot-Ciscar and Roll, 1998), this reduced cervicocephalic kinesthetic sensibility in various directions indicates decreases of sensory function of multiple neck muscles. Considering that one of the main functions of the cervical spine is to bear the weight of the head, the ﬁndings of this study are consistent with those of other studies in which proprioception in weight-bearing joints decreased with increasing age (Skinner et al., 1984; Gilsing et al., 1995; Robbins et al., 1995; Pai et al., 1997). Similar ﬁndings have been reported for the area of the lumbar spine, where the movement threshold was found to be changed with age (Parkhurst and Burnett, 1994). However, the present results contradict the ﬁndings of a study by Heikkila and Wenngren (1998), which showed no age-related decrease of cervical proprioception in asymptomatic adults but a signiﬁcant age-related decrease in patients with whiplash. One limitation of the present study is that the age range of the young and middle age of the present study is from 20–30 and 45–65, respectively. Thus, the age range of our analysis is only clustering in these two ranges of age. This might explain the different ﬁndings from the present study and the study of Heikkila and Wenngren (1998). The difference in results might also be attributed to two methodological reasons. Although the procedure for the repositioning testing was nearly identical in the two studies, both the tools for measuring the position angle and the participants’ conditions were different. The use of a three-dimensional motion analysis system in the present study was able to detect small changes in reposition errors better than the light-beam methods used in the study by Heikkila and Wenngren (1998). Additionally, Heikkila and Wenngren (1998) did not determine the effect of existing pain on cervicocephalic kinesthesia, an important point, given that the partici-

pants in their study had various levels of pain during repositioning testing. It is unclear whether the middle-aged subjects in the present study were more subject to age-related changes in the cervical spine because of biologic factors in the intervertebral disc and neck muscles or physical ﬁtness factors, such as less physical activity and prolonged sedentary lifestyle. These biologic factors may be associated with various age-related decreases in the musculoskeletal system, including disc degeneration (Buckwalter, 1995) and/or muscle atrophy (Clement, 1974; Lexell, 1995). It also remains to be determined whether age-related disc degeneration (Buckwalter, 1995), which may not be symptomatic, resulted in muscle active insufﬁciency because of changes in the length–tension relationship of the muscles that cross the inter-vertebral disc or indirectly affected the sensitivity of the muscle spindles in the deep spindle-rich muscles. In fact, recent research has demonstrated that the size of the dorsal neck muscles did not decrease with age (Rankin et al., 2005); and Boyd-Clark et al. (2002) have reported that muscle spindle distribution, morphology and density in some of the neck muscles do not change with age (Boyd-Clark et al., 2002). However, inﬁltration of fat into the dorsal neck muscle that occurs during the aging process cannot be excluded as a factor in reduced cervicocephalic kinesthetic sensibility. The contribution of joint receptor in capsule to the cervicocephalic kinesthetic sensibility could not be neglected, based on the facts that its close relationship among the inter-vertebral disc, facet joints, and the muscular control of the neck (Anderson et al., 2005). Any degenerative change in the inter-vertebral disc or segmental muscles might affect the function of receptors in the capsule of facet joints and vice versa. Furthermore, the different ﬁndings of Head-to NHP and Headto-Target repositioning tests might relate to the different contribution of muscle or joint afferents. During the Head-to-NHP reposition in the sagittal plane, the capsules on both sides are presumably loose and posterior neck muscles in both sides are highly activated (Mayoux-Benhamou et al., 1997). In contrast, during the Head-to-Target reposition in the horizontal plane, the capsule in contralateral side is presumably taut and only one side of neck muscles are more activated (Mayoux-Benhamou et al., 1997). Thus, the possibility of change of cervicocephalic sensibility contributed from the combined effects of joint receptors and muscle spindles could not be ruled out. The present results indicated that a history of mild chronic neck pain did not contribute to decreases in cervicocephalic kinesthetic sensibility in middle-aged adults. The subjects in our study did not have neck pain during the period of testing, which means that pain sensation was not an interfering factor. Thus, previous mild neck pain itself may not result in alteration of

ARTICLE IN PRESS C.-C. Teng et al. / Manual Therapy 12 (2007) 22–28

27

cervicocephalic kinesthetic sensibility. However, it is still possible that the measuring system is not accurate enough to detect group differences. Furthermore, this conclusion may not be generalizable to the population with acute or severe chronic pain. Whether more severe acute or chronic pain affects cervicocephalic kinesthetic sensibility remains to be determined. The ﬁndings of our study, taken together with the results of previous studies showing decreased cervicocephalic kinesthetic sensibility (Revel et al., 1991; Heikkila and Anstrom, 1996; Heikkila and Wenngren, 1998; Rix and Bagust, 2001), suggest that less precise repositioning of the head may be a primary or secondary result of the interaction of traumatic muscular/ligamentous/capsular injury, presence of pain, and increased age. Although evidence has demonstrated that acute muscle pain caused immediate excitatory response of muscle spindle (Johansson et al., 1993; Matre et al., 1998), so far there is no direct evidence of causal relationship between neck proprioception and chronic non-traumatic neck pain. However, the deﬁcit of neck proprioception had been documented in the chronic neck pain (Revel et al., 1991), including whiplash patients (Heikkila and Anstrom, 1996; Loudon et al., 1997). While examining the severity of experimental muscle pain around ankle, the pain of the subjects reached high intensity and the pain distribution covered the agonist and antagonist muscles, then the decrease of proprioception was found (Matre et al., 2002). Thus, greater age range and mild pain severity of subject recruited in the present study might explain the ﬁnding of prominent age effect and no signiﬁcant pain effect on proprioception. There was a tendency for the subjects to overshoot the target position, as evidenced by greater constant errors during repositioning from both the ﬂexed and extended positions to the neutral position. Several other investigators have reported a similar overshooting phenomenon occurring in patients with low back pain (Lam et al., 1999), patients with fewer large afferent ﬁbers, such as patients with large-ﬁbre sensory neuropathy (Gordon et al., 1995) or patients with deafferentation (Blouin et al., 1995). Thus, the overshooting phenomenon in the present study indicates decreases in proprioceptive afferent inputs, presumably from the activating neck muscles. Furthermore, additional studies are needed to determine whether a compensatory strategy is used by middle age or older patients (more than 65 years old) who have neck pain of different severity and a progressive decrease in cervicocephalic kinesthetic sensibility.

sensibility, thus, resides in examining the functional performance of spindle-rich neck muscles and joints under the inﬂuence of multiple clinical factors including age, pain, and trauma. Furthermore, an exercise program to enhance the cervicocephalic kinesthetic sensibility should be considered for middle-aged adults regardless of previous neck pain experiences.

4.1. Clinical application

References

Comprehensive Head-to-NHP and Head-to-Target repositioning tests in three cardinal planes could be used to reveal the afferent function of the neck structure. The clinical application of the cervicocephalic kinesthetic

Anderson JS, Hsu AW, Vasavada AN. Morphology, architecture and Biomechanics of human cervical multiﬁdus. Spine 2005;30:E86–91. Blouin J, Vercher JL, Gauthier GM, Paillard J, Bard C, Lamarre Y. Perception of passive whole-body rotations in the absence of neck and body proprioception. Journal of Neurophysiology 1995;74:2216–9.

5. Conclusion On the basis of the data in this study, we conclude that cervicocephalic kinesthetic sensibility in the sagittal plane is signiﬁcantly reduced in middle-aged individuals. This reduction was demonstrated by increased errors during repositioning tests that required greater contribution of muscular activity, such as repositioning of the head to the neutral position (Head-to-NHP) in the sagittal plane and (Head-to-Target) to a target position of left rotation. Thus, these results may imply that the number of proprioceptive afferents from multiple neck muscles and/or joints decrease with increasing age. Additionally, a history of mild chronic neck pain had no signiﬁcant effect on cervicocephalic kinesthetic sensibility.

Acknowledgements This study was supported by grants from the National Health Research Institute, Taiwan, ROC (NHRI-GT89-E823C).

Appendix A Deﬁne : E ij ¼ X ij T ij

i ¼ 1; . . . ; n; j ¼ 1; . . . ; m,

where n denotes the subject number, m ¼ 3 denotes the trial number and Xij and Tij denote the reposition angle of ith subject in jth trial, and the target position angle of ith subject in jth trial, respectively. Deﬁne ð1Þ CE ¼

ð2Þ

n 1X E¯ i n i¼1

RMSE2 ¼

where

n 1X RMSE2i n i¼1

CE i ¼ E¯ i ¼

where

m 1X E ij m j¼1

RMSE2i ¼

m 1X E2 . m j¼1 ij

ARTICLE IN PRESS 28

C.-C. Teng et al. / Manual Therapy 12 (2007) 22–28

Boden SD, McCowin PR, Davis DO, Dina TS, Mark AS, Wiesel S. Abnormal magnetic-resonance scans of the cervical spine in asymptomatic subjects. A prospective investigation. Journal of Bone and Joint Surgery (American) 1990;72:1178–84. Bovim G, Schrader H, Sand T. Neck pain in the general population. Spine 1994;19:1307–9. Boyd-Clark LC, Briggs CS, Galea MP. Muscle spindle distribution, morphology, and density in longus colli and multiﬁdus muscles of the cervical spine. Spine 2002;27:694–701. Buckwalter JA. Aging and degeneration of the human intervertebral disc. Spine 1995;20:1307–14. Clement FJ. Longitudinal and cross-sectional assessments of age changes in physical strength as related to sex, social class, and mental ability. Journal of Gerontology 1974;29:423–9. Cordo P, Gurﬁnkel VS, Bevan L, Kerr GK. Proprioceptive consequences of tendon vibration during movement. Journal of Neurophysiology 1995;74:1675–88. Costa D, Vitti M, de Oliveira T. Electromyographic study of the sternocleidomastoid muscle in head movements. Electromyography and Clinical Neurophysiology 1990;30:429–34. Cote P, Cassidy JD, Carroll L. The factors associated with neck pain and its related disability in the Saskatchewan population. Spine 2000;25:1109–17. Cote P, Cassidy JD, Carroll LJ, Kristman V. The annual incidence and course of neck pain in the general population: a population-based cohort study. Pain 2004;112:267–73. Dvorak J. Epidemiology, physical examination, and neurodiagnostics. Spine 1998;23:2663–73. Gandevia SC, McCloskey DI, Burke D. Kinaesthetic signals and muscle contraction. Trends in Neurosciences 1992;15:62–5. Gilsing MG, Van den Bosch CG, Lee SG, Ashton-Miller JA, Alexander NB, Schultz AB, et al. Association of age with the threshold for detecting ankle inversion and eversion in upright stance. Age Ageing 1995;24:58–66. Goebel JA, Hanson JM, Fishel DG. Age-related modulation of the vestibulo-ocular reﬂex using real and imaginary targets. Journal of Vestibular Research 1994;4:269–75. Gordon J, Ghilardi MF, Ghez C. Impairments of reaching movements in patients without proprioception. I. Spatial errors. Journal of Neurophysiology 1995;73:347–60. Harrison DE, Bula JM, Gore DR. Roentgenographic ﬁndings in the cervical spine in asymptomatic persons: a 10-year follow-up. Spine 2002;27:1249–50. Heikkila H, Anstrom PG. Cervicocephalic kinesthetic sensibility in patients with whiplash injury. Scandinavian Journal of Rehabilitation Medicine 1996;28:133–8. Heikkila HV, Wenngren BI. Cervicocephalic kinesthetic sensibility, active range of cervical motion, and oculomotor function in patients with whiplash injury. Archives of Physical Medicine & Rehabilitation 1998;79:1089–94. Hislop HJ, Montgomery J. Testing the muscles of the neck. In: Hislop HJ, Montgomery J, editors. Daniels and Worthingham’s muscle testing techniques of manual examination. Philadelphia: Saunders WB; 2002. p. 13–34. Johansson H, Djupsjobacka M, Sjolander P. Inﬂuences on the gammamuscle spindle system from muscle afferents stimulated by KCl and lactic acid. Neuroscience Research 1993;16:49–57. Lam SS, Jull G, Treleaven J. Lumbar spine kinesthesia in patients with low back pain. Journal of Orthopaedics and Sports Physical Therapy 1999;29:294–9. Lee H-Y, Teng C-C, Chai H-M, Wang S-F. Test-retest reliability of cervicocephalic kinesthetic sensibility in three cardinal planes. Manual Therapy 2006;11:61–8. Lexell J. Human aging, muscle mass, and ﬁber type composition. Journal of Gerontology 1995;50:11–6.

Loudon JK, Ruhl M, Field E. Ability to reproduce head position after whiplash injury. Spine 1997;22:865–8. Matre DA, Sinkjaer T, Svensson P, Rendt-Nielsen L. Experimental muscle pain increases the human stretch reﬂex. Pain 1998;75:331–9. Marchiori DM, Henderson CN. A cross-sectional study correlating cervical radiographic degenerative ﬁndings to pain and disability. Spine 1996;21:2747–51. Mayoux-Benhamou MA, Revel M, Vallee C. Selective electromyography of dorsal neck muscles in humans. Experimental Brain Research 1997;113:353–60. Pai YC, Rymer WZ, Chang RW, Sharma L. Effect of age and osteoarthritis on knee proprioception. Arthritis and Rheumatism 1997;40:2260–5. Parkhurst TM, Burnett CN. Injury and proprioception in the lower back. Journal of Orthopaedic and Sports Physical Therapy 1994;19: 282–95. Rankin G, Stokes M, Newham DJ. Size and shape of the posterior neck muscles measured by ultrasound imaging: normal values in males and females of different ages. Manual Therapy 2005;10:108–15. Revel M, Andre-Deshays C, Minguet M. Cervicocephalic kinesthetic sensibility in patients with cervical pain. Archives of Physical Medicine & Rehabilitation 1991;72:288–91. Revel M, Minguet M, Gregoy P, Vaillant J, Manuel JL. Changes in cervicocephalic kinesthesia after a proprioceptive rehabilitation program in patients with neck pain: a randomized controlled study. Archives of Physical Medicine and Rehabilitation 1994;75:895–9. Ribot-Ciscar E, Roll JP. Ago-antagonist muscle spindle inputs contribute together to joint movement coding in man. Brain Research 1998;791:167–76. Richmond FJ, Thomson DB, Loeb GE. Electromyographic studies of neck muscles in the intact cat. I. Patterns of recruitment underlying posture and movement during natural behaviors. Experimental Brain Research 1992;88:41–58. Rix GD, Bagust J. Cervicocephalic kinesthetic sensibility in patients with chronic, nontraumatic cervical spine pain. Archive of Physical Medicine and Rehabilitation 2001;82:911–9. Robbins S, Waked E, McClaran J. Proprioception and stability: foot position awareness as a function of age and footwear. Age Ageing 1995;24:67–72. Schmidt RA, Lee TD. Methodology for studying motor performance. In: Schmidt RA, Lee TD, editors. Motor control and learning: a behavioral emphasis. New York: Human Kinetics; 1999. p. 15–40. Skinner HB, Barrack RL, Cook SD. Age-related decline in proprioception. Clinical Orthopaedics and Related Research 1984:208–11. Solomonow M, Zhou BH, Harris M, Lu Y, Baratta RV. The ligamento-muscular stabilizing system of the spine. Spine 1998;23:2552–62. Sterling M, Jull G, Vicenzino B, Kenardy J. Sensory hypersensitivity occurs soon after whiplash injury and is associated with poor recovery. Pain 2003a;104:509–17. Sterling M, Jull G, Vicenzino B, Kenardy J, Darnell R. Development of motor system dysfunction following whiplash injury. Pain 2003b;103:65–73. Vernon H, Mior S. The neck disability index: a study of reliability and validity. Journal of Manipulative and Physiological Therapeutics 1991;14:411. Vitti M, Fujiwara M, Basmanjian JM, Iida M. The integrated roles of longus colli and sternocleidomastoid muscles: an electromyographic study. Anatomic Record 1973;177:471–84. Waling K, Jarvholm B, Sundelin G. Effects of training on female trapezius Myalgia: An intervention study with a 3-year follow-up period. Spine 2002;27:789–96. Wang S-F, Teng C-C, Lin KW. Measurement of cervical pattern during cyclic neck movement by ultrasound-based motion system. Manual Therapy 2005;10:68–72.

ARTICLE IN PRESS

Manual Therapy 12 (2007) 29–33 www.elsevier.com/locate/math

Original article

Myofascial trigger points in subjects presenting with mechanical neck pain: A blinded, controlled study C. Ferna´ndez-de-las-Pen˜as, C. Alonso-Blanco, J.C. Miangolarra Department of Physical Therapy, Occupational Therapy, Physical Medicine and Rehabilitation, Universidad Rey Juan Carlos (URJC), Alcorco´n, Spain Received 12 November 2004; received in revised form 4 January 2006; accepted 3 February 2006

Abstract The aim of this study was to describe the differences in the presence of myofascial trigger points (TrPs) in the upper trapezius, sternocleidomastoid, levator scapulae and suboccipital muscles between patients presenting with mechanical neck pain and control healthy subjects. Twenty subjects with mechanical neck pain and 20 matched healthy controls participated in this study. TrPs were identiﬁed, by an assessor blinded to the subjects’ condition, when there was a hypersensible tender spot in a palpable taut band, local twitch response elicited by the snapping palpation of the taut band, and reproduction of the referred pain typical of each TrP. The mean number of TrPs present on each neck pain patient was 4.3 (SD: 0.9), of which 2.5 (SD: 1.3) were latent and 1.8 (SD: 0.8) were active TrPs. Control subjects also exhibited TrPs (mean: 2; SD: 0.8). All were latent TrPs. Differences in the number of TrPs between both study groups were signiﬁcant for active TrPs (Po0.001), but not for latent TrPs (P40.5). Moreover, differences in the distribution of TrPs within the analysed cervical muscles were also signiﬁcant (Po0.01) for all muscles except for both levator scapulae. All the examined muscles evoked referred pain patterns contributing to patients’ symptoms. Active TrPs were more frequent in patients presenting with mechanical neck pain than in healthy subjects. r 2006 Elsevier Ltd. All rights reserved. Keywords: Mechanical neck pain; Myofascial trigger points; Myofascial pain; Blinded controlled study

1. Background Mechanical neck pain affects 45–54% of the general population at some time during their lives and can result in severe disability (Coˆte´ et al., 1998). The exact pathology of mechanical neck pain is not clearly understood and has been purported to be related to various anatomical structures including, uncovertebral or intervertebral joints, neural tissues, discs, muscular disorders and ligaments (Simons et al., 1999; Maitland et al., 2000). Different authors often assume that mechanical neck pain is associated with muscular, joint Corresponding author. Ce´sar Ferna´ndez de las Pen˜as, Facultad de Ciencias de la Salud, Universidad Rey Juan Carlos, Avenida de Atenas s/n, 28922 Alcorco´n Madrid, Spain. Tel.: + 34 91 488 88 84; fax: +34 91 488 89 57. E-mail addresses: [email protected], [email protected] (C. Ferna´ndez-de-las-Pen˜as).

1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.02.002

and neural impairments (Simons et al., 1999; Maitland et al., 2000). Soft tissue therapies and spinal manipulation/mobilization are manual therapies commonly used in the management of mechanical neck pain and associated impairments (Gross et al., 2002). Simons et al. (1999) have claimed that myofascial trigger points (TrPs) from neck and shoulder muscles might play an important role in the genesis of mechanical neck pain. There are epidemiological studies suggesting that TrPs represent an important source of musculoskeletal disorder (Chaiamnuay et al., 1998). A TrP is a hyperirritable spot within a palpable taut band of a skeletal muscle that is painful on compression, stretch or overload of the affected tissues and that can give rise to a typical referred pain pattern (Simons et al., 1999). TrPs are typically located by the following physical examination ﬁndings: presence of a palpable taut band in a skeletal muscle, tender spot within the

ARTICLE IN PRESS 30

C. Ferna´ndez-de-las-Pen˜as et al. / Manual Therapy 12 (2007) 29–33

taut band, local twitch response provoked by snapping palpation of the TrP, and referred pain pattern (Simons et al., 1999). The formation of TrPs may result from a variety of factors, such as severe trauma, overuse, mechanical overload or psychological stress (Simons et al., 1999). Recent studies have hypothesized that the pathogenesis of TrPs results from injured or overloaded muscle ﬁbres. This leads to involuntary shortening, loss of oxygen supply, loss of nutrient supply and increased metabolic demand on local tissues (Mense et al., 2000). Although previous studies have investigated the prevalence of TrPs in benign chronic headaches (Jaeger, 1989; Marcus et al., 1999), we were unable to locate any study in the peer-reviewed literature analysing the prevalence of TrPs in the cervical musculature (levator scapulae, sternocleidomastoid, upper trapezius muscle). These TrPs have the potential to refer pain contributing to neck and shoulder symptoms in patients presenting with mechanical neck pain. Our aim was to describe the differences in the presence of TrPs in the upper trapezius, sternocleidomastoid, levator scapulae and suboccipital muscles between subjects with mechanical neck pain and healthy controls. In addition, we examined the possible correlation in the presence of TrPs among the aforementioned muscles.

2. Material and methods 2.1. Subjects Twenty subjects presenting with mechanical neck pain for at least 4 months and 20 healthy age- and sexmatched controls without neck pain during the prior 6 months participated in this study from January to September of 2004. For the purpose of this study, mechanical neck pain was deﬁned as generalized neck and/or shoulder pain with mechanical characteristics including: symptoms provoked by maintained neck postures or by movement, or by palpation of the cervical muscles. Patients were excluded if they exhibited any of the following: (1) diagnosis of ﬁbromyalgia syndrome according to the American College of Rheumatology (Wolfe et al., 1990); (2) previous history of a whiplash injury; (3) history of cervical spine surgery; (4) diagnosis of cervical radiculopathy or myelopathy determined by their primary care physician; or (5) therapeutic intervention for myofascial pain within the past month before the study. The health status of all patients was clinically stable, without current symptoms of any other concomitant chronic disease. The clinical history for each patient was solicited from their primary care physician to assess the exclusion criteria and to check the presence of ‘‘red ﬂags’’, i.e. infection, malignancy. Subjects with neck pain were examined on days in which the

neck pain intensity was less than four points on a 10-cm horizontal visual analogue scale. This study was supervised by the Department of Physical Therapy, Occupational Therapy, Physical Medicine and Rehabilitation of the Universidad Rey Juan Carlos. The research project was approved by the local human research committee of the Universidad Rey Juan Carlos. All subjects signed an informed consent prior to their inclusion.

2.2. Procedure Subjects were examined for TrPs by an assessor who had more than 4 years experience in TrPs diagnosis, and who was blinded to the subjects’ condition. The diagnosis of the TrP was performed following the latter ﬁve diagnostic criteria described by Simons et al. (1999) and by Gerwin et al. (1997): (1) presence of a palpable taut band in a skeletal muscle; (2) presence of a hypersensible tender spot in the taut band; (3) local twitch response elicited by the snapping palpation of the taut band; (4) reproduction of the typical referred pain pattern of the TrP in response to compression; and (5) spontaneous presence of the typical referred pain pattern and/or patient recognition of the referred pain as familiar. If the ﬁrst four criteria were satisﬁed the TrP was considered to be latent. If all of the aforementioned criteria were present the TrP was considered to be active (Gerwin et al., 1997; Simons et al., 1999). Tender points were also diagnosed when subjects reported local tenderness but they did not report referred pain to compression and/or overload of the affected tissues, so minimum criteria for TrP diagnosis were not fulﬁlled (Gerwin et al., 1997; Simons et al., 1999). In criteria four and ﬁve, pressure on the TrP was assessed using a Pressure Threshold Meter (PTM). The assessor applied continuous pressure approximately at a rate of 1 kg/cm2/seg until 2.5 kg/cm2. A PTM distributed by ‘‘Pain Diagnosis and Rehabilitation’’ commercial home (233 East Shore Road, Suite 108, Great Neck, New York 11023) was used in this study. The PTM consists on a rubber disk with 1 cm2 surface The rubber disk is connected to a pressure pole inserting into a gauge which records pressure in kilograms (kg). Pressure measurements are expressed in kg/cm2. The range of pressure is between 0 and 10 kg/cm2 recording values each 0.1 kg. Previous papers reported an intra-examiner (I.C.C.) reliability of the PTM ranging from 0.6 to 0.97, and an inter-examiner reliability (I.C.C.) ranging from 0.4 to 0.98 (Takala, 1990; Levoska, 1993). Pressure thresholds lower than 3 kg, are considered abnormal (Fischer, 1996). Fig. 1 details the location and the referred pain patterns evoked by TrPs in the examined cervical muscles based on the comprehensive research performed by Simons et al. (1999).

ARTICLE IN PRESS C. Ferna´ndez-de-las-Pen˜as et al. / Manual Therapy 12 (2007) 29–33

2.3. Analysis of data Descriptive data was collected on all patients. The number of active and latent TrPs was recorded for each patient and then the group mean was calculated. The inter-group comparison between the number of TrPs (active or latent) was analysed with the unpaired t-test.

31

The w2 of association was used to assess the differences in the distribution of TrPs within each muscle between both study groups and the presence of TrPs among the analysed cervical muscles. The statistical analysis was conducted at a 95% conﬁdence level. A P-value less than 0.05 were considered as statistically signiﬁcant.

3. Results

Fig. 1. Referred pain pattern from myofascial trigger points in some cervical muscles based on the comprehensive research performed by Simons et al. (1999).

A total of 20 neck pain subjects, 7 men and 13 women, 20–44 years old (mean age: 2877 years), were studied. The duration of neck complaints ranged from 7 to 15 months (mean ¼ 9.2573 months). The mean level of neck pain according to the 10 cm visual analogue scale on the day of the examination was 2.5 cm (SD 0.7). Control subjects were 20 healthy volunteers, 10 men and 10 women, aged 20–50 (mean age: 2979 years). Each of the 20 neck pain patients exhibited at least three TrPs in the analyzed muscles. The mean number of TrPs on each patient was 4.3 (SD: 0.9), of which 2.5 (SD: 1.3) were latent and 1.8 (SD: 0.8) were active TrPs. On the other hand, each control subject also exhibited TrPs (mean: 2; SD: 0.8). All were latent TrPs. Differences in the total number of TrPs (active and latent) and the number of active TrPs between both study groups reached the statistical signiﬁcance (Po0.001). Differences in the number of latent TrPs were not signiﬁcant (P40.05). Moreover, differences in the distribution of active TrPs within each cervical muscle were also signiﬁcant for all muscles except for both levator scapulae (see Table 1). Within the neck pain group, TrPs in the suboccipital muscles were the most prevalent (n ¼ 18; 90%), following by TrPs in the right sternocleidomastoid muscle (n ¼ 17; 85%), and TrPs in the left upper trapezius muscle (n ¼ 14; 70%). Surprisingly, TrPs in the suboccipital muscles were the most prevalent in our neck pain patients (90%). As the referred pain evoked

Table 1 Distribution of subjects with myofascial trigger points (active or latent) in both study groups Suboccipital muscles

Upper trapezius muscle

Levator scapulae muscle

Sternocleidomastoid muscle

Left side

Right side

Left side

Right side

Left side

Right side

Subjects with mechanical neck pain Active TrPs (n) 10 Latent TrPs (n) 8

7 7

8 5

0 6

3 3

3 10

5 12

Control healthy subjects Active TrPs (n) 0 Latent TrPs (n) 5 P-value 0.001

0 8 0.01

0 10 0.006

0 6 NS

0 7 NS

0 1 0.001

0 4 0.001

TrP, myofascial trigger point; NS, non-signiﬁcant; n, number of subjects, P-values express differences between active TrPs. Differences between latent TrPs were not signiﬁcant.

ARTICLE IN PRESS 32

C. Ferna´ndez-de-las-Pen˜as et al. / Manual Therapy 12 (2007) 29–33

by these TrPs spreads to the head and it is usually perceived as headache (Simons et al., 1999), patients were asked for the presence of headache. Half of these patients reported tension-type headache concomitant with their neck symptoms, especially when their neck pain were aggravated by stress. Within the control group, the most prevalent TrPs were located in both upper trapezius muscles (n ¼ 10; 50% in the right side; n ¼ 8; 40% in the left side) and in the right levator scapulae muscle (n ¼ 7; 35%). Finally, w2 analysis by contingency tables showed a signiﬁcant relationship between the presence of TrPs in the left upper trapezius and left sternocleidomastoid muscles (P ¼ 0.03), between TrPs in the left upper trapezius and left levator scapulae muscles (P ¼ 0.005) and between TrPs in the right upper trapezius and suboccipital muscles (P ¼ 0.05). Other relationships, i.e. TrPs in the right or left sternocleidomastoid and suboccipital muscles, did not reach a signiﬁcant level (P ¼ 0.06).

4. Discussion Our study is the ﬁrst to provide preliminary evidence suggesting that active myofascial trigger points (TrPs) are more common in subjects presenting with mechanical neck pain than in healthy controls. Active TrPs of the examined muscles evoked referred pain patterns contributing to neck symptoms seen in our patients. Simons et al. (1999) claimed that neck pain might be usually provoked by TrPs in the upper trapezius and levator scapulae muscles. Almost all neck pain patients showed TrPs in the upper ﬁbres of the trapezius muscle, in the right and/or left sides. Most of these TrPs were active TrPs, as patients were familiar with the location (posterior-lateral region of the neck) and the quality of the referred pain (tightening and burning) that was elicited by pressure applied to the TrP. When the assessor applied pressure to that TrP, many patients reported: ‘‘Yes, this is exactly the pain that I usually feel’’. On the other hand, active TrPs in the control healthy group were scarce (Po0.001). TrPs in this group never evoked a familiar ache, and therefore were classiﬁed as latent TrPs (Gerwin et al., 1997; Simons et al., 1999). Signiﬁcant differences between neck pain subjects and healthy controls were found for active TrPs, but not for latent TrPs. This is expected, as latent TrPs, have been commonly observed in healthy, normal subjects (Chaiamnuay et al., 1998). Hong (1994) claimed that the treatment of ‘‘key’’ TrPs in some muscles could also relieve the pain arising from satellite TrPs in other muscles. This was one of the reasons for assessing the possible relationship on the presence of TrPs in the cervical musculature. Our results

have showed different relationships among the presence of TrPs in some of the analysed cervical muscles. One hypothesis to justify these relationships might be that muscles located in the region of the referred pain pattern of a TrP might also develop secondary TrPs (Hong, 1994; Simons et al., 1999). This hypothesis might explain our results: the presence of TrPs in the sternocleidomastoid was associated to the presence of TrPs in the homo-lateral upper trapezius muscle, or TrPs in the upper trapezius muscle associated to TrPs in the homolateral levator scapulae. However, our study design did not enable a cause and effect relationship to be established so further studies are required on that topic. TrPs diagnosis needs adequate innate ability, training, and clinical practice to develop a high degree of reliability in the examination (Gerwin et al., 1997; Sciotti et al., 2001). Moreover, some muscles are consistently more reliably examined than others. Simons et al. (1999) and Gerwin et al. (1997) recommend that the minimum acceptable criteria for active TrP diagnosis is the combination of the presence of a spot tenderness in a palpable taut band in a skeletal muscle and patient recognition of referred pain that is elicited by pressure applied to the tender spot. These criteria had obtained a good inter-examiner reliability (k) ranging from 0.84 to 0.88 (Gerwin et al., 1997). In the present study, these two minimum criteria identiﬁed active TrPs. Furthermore, the local twitch response, a conﬁrmatory sign of TrP diagnosis (Simons et al., 1999), was also an inclusion requirement in the diagnosis of TrPs in all muscles except in the suboccipital muscles, in which it is difﬁcult or impossible to elicit a local twitch response by snapping palpation. Suboccipital muscle TrPs were explored bilaterally in order to evoke bilateral referred pain (Simons et al., 1999), and also to avoid the palpation of TrPs in other cervical posterior muscles. The high incidence of suboccipital muscle TrPs in neck pain patients might be provoked because these muscles could not be explored unilaterally. Tender points were also diagnosed when subjects did not report referred pain elicited by compression and/or overload of the affected tissues, so minimum criteria for TrP diagnosis were not fulﬁlled (Gerwin et al., 1997; Simons et al., 1999). In the present study the presence of TrPs in subjects with mechanical neck pain has been demonstrated. However, it is possible that other tissues might also contribute to symptoms associated with mechanical neck pain. Edgar et al. (1994) reported that decreased extensibility of the upper quadrant neural structures as assessed by the median nerve tension test was associated with decreased length of the upper trapezius muscle. Ferna´ndez-de-las-Pen˜as et al. (2005) have recently found a signiﬁcant relationship between the presence of TrPs in the upper trapezius muscle and the presence of intervertebral joint dysfunctions at C3 and C4 vertebrae.

ARTICLE IN PRESS C. Ferna´ndez-de-las-Pen˜as et al. / Manual Therapy 12 (2007) 29–33

Since the relationship of TrPs to impairments in the articular or neural systems has not been well established, further research is required.

5. Conclusion Active TrPs are more frequent in neck pain patients than in healthy subjects. The prevalence of latent TrPs is similar in neck pain patients and healthy subjects. From a clinical standpoint, the results from the study supports the clinical practice of assessing TrPs in the cervical musculature as one important element of the clinical reasoning process performed by physical therapists in patients with mechanical neck pain.

Acknowledgements We would like to acknowledge Dr. David Simons for his kind encouragement and support. We would also like to thank to each patients who participated in the study. References Chaiamnuay P, Darmawan J, Muirden KD, Assawatanabodee P. Epidemiology of rheumatic disease in rural Thailand: a WHOILAR COPCORD study. Community Oriented Programme for the Control of the Rheumatic Disease. Journal of Rheumatology 1998; 25:1382–7. Coˆte´ P, Cassidy JD, Carroll L. The Saskatchewan health and back pain survey. The prevalence of neck pain and related disability in Saskatchewan adults. Spine 1998;23:1689–98. Edgar D, Jull G, Sutton S. Relationship between upper trapezius muscle length and upper quadrant neural tissue extensibility. Australian Journal of Physiotherapy 1994;40:99–103.

33

Ferna´ndez-de-las-Pen˜as C, Ferna´ndez J, Miangolarra JC. Musculoskeletal disorders in mechanical neck pain: myofascial trigger points versus cervical joint dysfunctions. A clinical study. Journal of Musculoskeletal Pain 2005;13(1):27–35. Fischer AA. Algometry in diagnosis of musculoskeletal pain and evaluation of treatment outcome: an update. Journal of Musculoskeletal Pain 1996;6:5–33. Gerwin RD, Shanon S, Hong CZ, Hubbard D, Gevirtz R. Interrater reliability in myofascial trigger point examination. Pain 1997;69:65–73. Gross AR, Kay T, Hondras M, Goldsmith C, Haines T, Peloso P, et al. Manual therapies for mechanical neck disorders: a systematic review. Manual Therapy 2002;7:131–49. Hong CZ. Considerations and recommendations regarding Myofascial trigger point injection. Journal of Musculoskeletal Pain 1994;2: 29–59. Jaeger B. Are cervicogenic headaches due to myofascial pain and cervical spine dysfunction? Cephalalgia 1989;9:157–64. Levoska S. Manual palpation and pain threshold in female ofﬁce employees with and without neck – shoulder symptoms. Clinical Journal of Pain 1993;9:236–41. Marcus DA, Scharff L, Mercer S, Turk DC. Musculoskeletal abnormalities in chronic headache: a controlled comparison of headache diagnostic groups. Headache 1999;39:21–7. Maitland G, Hengeveld E, Banks K, English K. Maitland0 s vertebral manipulation, 6th ed. London: Butterworths Heineman; 2000. Mense S, Simons DG, Russell IJ. Muscle pain: understanding its nature, diagnosis and treatment. Philadelphia: Lippincontt Williams & Wilkins; 2000. Sciotti VM, Mittak VL, DiMarco L, Ford LM, Plezbert J, Santipadri E, et al. Clinical precision of myofascial trigger point location in the trapezius muscle. Pain 2001;93:259–66. Simons DG, Travell J, Simons LS. Myofascial pain and dysfunction. The trigger point manual. Volume 1. 2nd ed., Baltimore: Williams & Wilkins, 1999. Takala EP. Pressure pain threshold on upper trapezius and levator scapulae muscles. Scandinavian Journal of Rehabilitation Medicine 1990;22:63–8. Wolfe F, Smithe HA, Tunus MB, Bennet RM, Bombardier C, Goldenberg DL, et al. The American College of Rheumatology 1990 criteria for clasiﬁcation of ﬁbromyalgia: report of the multicenter criteria committee. Arthritis and Rheumatism 1990;33: 160–70.

ARTICLE IN PRESS

Manual Therapy 12 (2007) 34–39 www.elsevier.com/locate/math

Original article

Cranio-cervical ﬂexor muscle impairment at maximal, moderate, and low loads is a feature of neck pain Shaun O’Leary, Gwendolen Jull, Mehwa Kim, Bill Vicenzino Division of Physiotherapy, The University of Queensland, St Lucia Queensland 4072, Australia Received 20 May 2005; received in revised form 19 December 2005; accepted 13 February 2006

Abstract Impairment of the cranio-cervical ﬂexor (CCF) muscles is a feature of painful cervical spine disorders. The aim of this study was to investigate if CCF muscle impairment is present over a range of contraction intensities (maximal, moderate, low) in neck pain sufferers compared to individuals with no history of neck pain. Isometric CCF muscle strength (isometric maximal voluntary contraction (MVC)), and endurance at moderate (50% of MVC), and low (20% of MVC) loads was compared in 46 participants with neck pain (Neck Disability Index (NDI): mean7SD; 22.875.2) and 47 control participants (NDI: 2.672.6). Compared to the control group, the neck pain group had a signiﬁcant deﬁcit (15.9%, P ¼ 0:037) in their MVC peak torque recordings, as well as a signiﬁcantly reduced capacity to sustain isometric CCF muscle contractions to task failure at 20% of MVC (35% deﬁcit, P ¼ 0:03) and 50% of MVC (27% deﬁcit, P ¼ 0:002). Neck pain participants also demonstrated poorer accuracy in maintaining their MVC20 contraction at the nominated isometric CCF torque amplitude (P ¼ 0:02), compared to control participants. It would appear that impairment in isometric CCF muscle performance exists over a range of contraction intensities in neck pain sufferers, which may beneﬁt from speciﬁc therapeutic intervention. r 2006 Elsevier Ltd. All rights reserved. Keywords: Cranio-cervical ﬂexor muscles; Neck pain; Dynamometry

1. Introduction Cranio-cervical ﬂexor (CCF) muscle impairment is a feature of painful cervical spine disorders (Watson and Trott, 1993; Falla et al., 2004a, b; Jull et al., 2004b), and their rehabilitation is effective in reducing the symptoms of cervicogenic headache (Jull et al., 2002). In accordance, speciﬁc assessment and rehabilitation of their performance is an accepted practice in the clinical management of neck pain and cervicogenic headache (Jull et al., 2002, 2004a). Speciﬁc assessment of these muscles is warranted because, compared to other cervical ﬂexor muscles, the attachment of the CCF muscles (primarily the longus capitis and rectus capitis Corresponding author. Tel.: +61 7 3365 4587; fax: +61 7 3365 2775. E-mail address: [email protected] (S. O’Leary).

1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.02.010

anterior muscles) to the head, affords them functional autonomy in orientation and stability of the specialized upper cervical motion segments (Vasavada et al., 1998; Moore and Dalley, 1999; Kettler et al., 2002). Theoretically, deﬁcits in the contractile capacity of the CCF muscles would destabilize the cranio-cervical region with a tendency for it to extend and as such, poor performance of these muscles has been implicated in abnormal head on neck posture (Janda, 1988; Jull, 1988; Watson and Trott, 1993; Grimmer and Trott, 1998). While tests of isometric cervical ﬂexor muscle performance using various dynamometry methods have been widely described (Jordan et al., 1999; Peolsson et al., 2001; Chiu and Lo, 2002; Garces et al., 2002; Seng et al., 2002; Gabriel et al., 2004; Ylinen et al., 2004), speciﬁc measurements of isometric CCF muscle performance are less common (Watson and Trott, 1993; Jull et al., 2004a; O’Leary et al., 2005b). Watson and Trott

ARTICLE IN PRESS S. O’Leary et al. / Manual Therapy 12 (2007) 34–39

(1993) using an isometric dynamometry method showed deﬁcits in CCF muscle maximal strength and endurance in cervicogenic headache sufferers compared to control participants. Deﬁcits in low load CCF muscle performance have also been shown in neck pain sufferers compared to control participants when performing the cranio-cervical ﬂexion test (Jull et al., 2004b). The cranio-cervical ﬂexion test method utilizes a pneumatic pressure sensor placed behind the upper cervical spine to monitor the capacity of the CCF muscles to ﬂatten the cervical lordosis. Larger pressure shortfalls (Jull et al., 1999, 2004b; Jull, 2000), and altered coordination of the CCF muscles within the cervical ﬂexor synergy, characterized by reduced activity of primary CCF muscles (longus capitis) that coincided with elevated activity of superﬁcial muscles that are not primary CCF muscles (sternocleidomatoid, anterior scalenes) (Falla et al., 2004b), have been demonstrated in participants with neck pain when compared to control participants. Such ﬁndings of CCF muscle impairment at low load have underpinned strategies for their rehabilitation (Jull et al., 2002, 2004a). The purpose of this study was to investigate isometric CCF muscle performance at maximal (maximal voluntary contraction—MVC), moderate (50% of MVC), and low (20% of MVC) contraction intensities in neck pain sufferers compared to control participants using a cranio-cervical ﬂexion dynamometry method (O’Leary et al., 2005b). The aim of these isometric tests was to challenge the CCF muscles over a spectrum of contraction intensities as would be required for cranio-cervical postural function. The hypothesis was that participants with neck pain would demonstrate poorer performance over the spectrum of contraction intensities (maximal, moderate, low) compared to control participants with no history of neck pain.

2. Methods 2.1. Participants Ninety-three female volunteers participated in this study including 46 participants with a history of neck pain (age 37.0710.1 years, weight 64.0710.6 kg, height 166.276.7 cm), and 47 control participants with no history of neck pain (age 27.877.7 years, weight 62.679.3 kg, height 167.576.1 cm). Groups were similar in their height and weight characteristics but there was a signiﬁcant difference in age (Po0:0001) with the neck pain group being older. Females only were included in this comparative cohort study so as to eliminate a potential confounding factor of gender to strength measures (Jordan et al., 1999; Kumar et al., 2001; Portero and Genries, 2003).

35

All participants were recruited via electronic and written advertising within the university and general community. Participants with neck pain were included if they reported neck pain of greater than 3 months duration of either a traumatic or non-traumatic origin, scored greater than 10/100 on the Neck Disability Index (NDI) (NDI 22.875.2) (Vernon, 1996), and demonstrated positive ﬁndings on a physical examination of the cervical spine such as altered joint motion, and painful reactivity to palpation (Jull, 1994). Control participants were included if they reported no history of neck pain for which they had sought treatment, scored less than 10 on the NDI (NDI 2.672.6), and had no positive ﬁndings on a physical examination of the cervical spine. Participants in either group were excluded if they had speciﬁcally trained their neck or shoulder girdle muscles in the preceding 6 months, had neck pain from nonmusculoskeletal causes, neurological signs, or any medical disorder contraindicating physical exercise. After receiving verbal and written information each participant signed a consent form. This study was ethically approved by the University’s Medical Research Ethics Committee and was in accordance with the declaration of Helsinki.

2.2. Instrumentation and measurement procedure CCF muscle performance was measured in supine using a cranio-cervical ﬂexion dynamometer (Fig. 1) that has been shown to have good test–retest measurement reliability (ICC 0.7–0.92) (O’Leary, 2005). The dynamometer measures isometric CCF muscle torque about the axis of rotation (AOR) of the C0/1 motion segment that lies in close proximity to the anterior mastoid process (Harms-Ringdahl et al., 1986; White and Panjabi, 1990; vanMameran et al., 1992). Due the occlusion of the anterior mastoid process from direct vision by the ear, the axis of the dynamometer was aligned to the concha of the ear as this best approximated the anterior mastoid process (O’Leary et al., 2005b). The CCF effort was resisted at the under-surface of the mandible by the dynamometer resistance arm producing a torque at the dynamometer axis that was measured in Newton meters (N m). Any tendency for the participant to push their head into or lift it off the supporting surface, thought to be a possible strategy to enhance CCF muscle torque (O’Leary et al., 2005a), was monitored with a force platform (Watson and Trott, 1993; O’Leary, 2005). All measurements of CCF muscle torque were recorded in an anthropometric neutral cranio-cervical ﬂexion/extension position (Montagu, 1960; Norton and Olds, 1996), with the participant’s knees and hips positioned in 451 and the arms folded

ARTICLE IN PRESS 36

S. O’Leary et al. / Manual Therapy 12 (2007) 34–39

elevate the visual display graph. Following the MVC trials, 5 min rest was permitted before performing the endurance tests. MVC20 and MVC50 tests: The MVC20 test was performed ﬁrst. Participants performed cranio-cervical ﬂexion onto the application pad of the dynamometer to elevate the visual display graph to the marker indicating 20% of their MVC effort. Participants were asked to accurately maintain this level of contraction effort until they perceived that their muscles were too fatigued to sustain the contraction any longer. No expectations were given to participants of any maximal time limit for the test and they were not informed that the Labview program was set to default after a 10-min time period. Participants were warned that while some discomfort associated with muscle fatigue was to be expected, they were to discontinue the test if they experienced their reported neck pain symptoms, or pain in the jaw, head, or upper limbs. Following completion of the MVC20 test and 10 min rest, the MVC50 test was performed using the same procedure as for the MVC20 test with the exception that the target torque was set at 50% of MVC. All participants received standardized verbal encouragement during the dynamometry tests. 2.3. Data management and statistical analysis Fig. 1. Cranio-cervical ﬂexion dynamometry device for the measurement of isometric cranio-cervical ﬂexor muscle torque about the AOR of the C0/1 motion segment. The dynamometer axis is aligned to the participants’ C0/1 AOR (concha of the ear) with the assistance of the web camera. The dynamometer resistance arm is extended from the dynamometer axis so that the application pad sat comfortably at the under-surface of the mandible. When the participant performed a cranio-cervical ﬂexion effort it resulted in torque at the dynamometer axis that was recorded in Newton-meters (N m). The head was supported on a padded platform supported on ball bearings to minimize friction at the point of head contact and suspended on a mechanical transducer to monitor changes in dorsal head force on the supporting surface.

across the chest to minimize the effects of limb leverage (O’Leary et al., 2005b). Two custom-written LabView programs (LabView 6i Virtual Instruments) recorded the CCF muscle torque data at 20 Hz. The ﬁrst was for the measurement of isometric MVC peak torque, and the second for endurance measurements of sustained torque at 20% of MVC (MVC20) and 50% of MVC (MVC50). Visual feedback of CCF muscle torque was provided to participants on a visual display unit. Maximal voluntary contraction (MVC) test: Five MVC trials were performed following a standardized warm up procedure. A rest period of 1 m was given between repetitions. Participants were instructed to nod their head (‘yes’ type action) such that their jaw pushed down onto the padded bar in an effort to maximally

Data were excluded for measurements that reproduced participants’ painful neck symptoms, or produced pain in the head, jaw, or upper limb during the dynamometry procedure, or if the participant found it too difﬁcult to control the dual task of CCF and control of their dorsal head force on the supporting surface. For the MVC measurements the highest amplitude torque measurement (N m) of the ﬁve MVC trials was recorded as the MVC peak torque score. For the MVC20,50 tests, two measures were extracted from each trial. The time (s) until the participant terminated the test was recorded as the time to task failure measure. The second measure, contraction accuracy, was calculated as the percentage of the recorded samples that remained within a previously detailed amplitude margin (73%) either side of the expected torque task for the duration of the test (O’Leary, 2005). This calculation was performed ofﬂine using a custom written program (LabView 6i Virtual Instruments). The ﬁrst 10 s of data for each test was discarded as this time was permitted for the participant to reach and stabilize the contraction. Group means (7standard deviation) were computed for both the neck pain and control groups for all measures (MVC peak torque, MVC20,50 time to task failure, contraction accuracy). Between group data were compared using a general linear model univariate analysis for the MVC peak torque measure (N m), and a multivariate analysis for the MVC20 and MVC50

ARTICLE IN PRESS S. O’Leary et al. / Manual Therapy 12 (2007) 34–39

measures of time to task failure (seconds), and contraction accuracy (% accuracy). All group comparative analyses included the age as a covariate to account for any discrepancies due to the signiﬁcant age difference between the groups.

3. Results 3.1. MVC test All 93 participants completed the MVC trials with no report of neck pain during these measures. The neck pain group (n ¼ 46) showed signiﬁcantly less CCF muscle strength (P ¼ 0:037) than the control group (n ¼ 47), demonstrating a deﬁcit of 15.9% of peak torque (Table 1). 3.2. MVC20 test Between-group comparisons were made between 31 neck pain and 36 control participants for the MVC20 test following the exclusion of data due to the onset of pain (14 neck pain, 8 control), and inability to control the dual task of CCF muscle torque and dorsal head force (1 neck pain, 3 controls). Six of the control and two of the neck pain participants maintained the contraction for the 10-min time limit of the test. These data were retained in the analysis. The time to task failure measure revealed that the neck pain group could not sustain the MVC20 contraction for as long as the control participants (mean difference ¼ 86.7 s (35% deﬁcit), P ¼ 0:03, Table 1). The neck pain group were signiﬁcantly less accurate at sustaining the MVC20 contraction within the set margin (P ¼ 0:02), demonstrating a deﬁcit of 10.4% compared to the control group. Table 1 Comparison of group (neck pain, control) means7standard deviations for the isometric maximal voluntary contraction (MVC) peak torque measurement, and the 20% MVC (MVC20) and 50% MVC (MVC50) measurements of time to task failure (seconds), and contraction accuracy (%) Measurements MVC Peak torque (N m)

Neck pain

Control

5.371.5

6.371.7

MVC20 Time to task failure (s) Contraction accuracy (%)

158.47154.2 33.1715.7

245.17198 43.5714.5

MVC50 Time to task failure (s) Contraction accuracy (%)

48.1723.6 35.7717.3

66731.5 37.7714

Denotes signiﬁcant differences between groups (Po0:05).

37

3.3. MVC50 test Between-group comparisons were made between 37 neck pain and 43 control participants for the MVC50 test following the exclusion of data due to the onset of pain (7 neck pain, 1 control) and inability to control the dual task of CCF torque and dorsal head force (2 neck pain, 3 controls). The group with neck pain exhibited a 17.8 s (27%) deﬁcit in the MVC50 contraction compared to controls (P ¼ 0:002, Table 1). There were no signiﬁcant group differences for the contraction accuracy measure at MVC50.

4. Discussion Knowledge of the characteristics of muscle impairment is a prerequisite to the appropriate prescription of therapeutic exercise when managing painful neck disorders. The results of this study indicate that neck pain sufferers have deﬁcits in CCF muscle strength, endurance at low and moderate intensity contractions, and contraction precision at low intensity contractions. Each of these deﬁcits may require speciﬁc exercise strategies to restore as adaptations in the neuromuscular system to exercise appear to be related to the speciﬁcity of training (Conley et al., 1997). Changes in muscle performance appear to be speciﬁc to exercise characteristics such as movement pattern, velocity of contraction, type of contraction and joint angle (Rasch and Morehouse, 1957; Sale and MacDougall, 1981; Kanehisa and Miyashita, 1983; Portero et al., 2001). Studies are now needed to evaluate training effects and adaptations of various cervical muscle training protocols in order to better match them to speciﬁc impairments. Deﬁcits in maximal CCF muscle strength in neck pain sufferers in this study (15.9% deﬁcit) were similar to those found by Watson and Trott (1993) (14.6% deﬁcit) in cervicogenic headache sufferers. It could be postulated that when head on neck orientation is challenged under larger loads, these muscles may be less capable of controlling cranio-cervical joint orientation and absorbing the forces of load. Poorer ability to exert maximal CCF muscle torque in the presence of musculoskeletal impairment may be intrinsic to the muscle tissue itself such as reductions in cross-sectional area (Mayoux Benhamou et al., 1989), or due to other factors such as neural drive (Gandevia, 2001), which may also be inﬂuenced by factors such as pain tolerance, fear avoidance, determination, and competitiveness (Vlaeyen and Linton, 2000; Mannion et al., 2001). A large (35%) and signiﬁcant deﬁcit (P ¼ 0:03) in the capacity of the neck pain participants to sustain an MVC20 contraction until task failure and poorer

ARTICLE IN PRESS 38

S. O’Leary et al. / Manual Therapy 12 (2007) 34–39

accuracy in sustaining the contraction compared to controls was observed. Poorer contraction accuracy when sustaining a contraction may reﬂect other muscle fatigue manifestations such as muscle tremor (Gandevia, 2001). These ﬁndings are consistent with the deﬁcits found at low load in studies utilizing the cranio-cervical ﬂexion test (Jull et al., 1999, 2004b; Jull, 2000; Falla et al., 2004b). It is feasible that poorer performance by neck pain sufferers during the MVC20 test may be explained by altered coordination of the CCF muscles previously demonstrated during the cranio-cervical ﬂexion test (Falla et al., 2004b), however this needs to be investigated. A signiﬁcant deﬁcit (27%, P ¼ 0:002) was also found for the neck pain group in their capacity to sustain the MVC50 test to task failure. This is the ﬁrst study to demonstrate CCF muscle impairment at moderate loads. The results of the MVC20 and MVC50 measures suggest that when the head on neck postural orientation is challenged under prolonged or repetitive circumstances in a neck pain sufferer, the CCF muscles may fatigue prematurely and not be capable of controlling cranio-cervical orientation potentially exposing cervical spine tissues to abnormal mechanical load. It is acknowledged that the MVC20,50 measures of time to task failure are affected by factors other than the intrinsic fatigue of the CCF muscle group and would include multiple neural contributions (Hunter et al., 2004), supraspinal inﬂuences (Gandevia 2001), and factors such as pain tolerance, boredom, determination, competitiveness and fear-avoidance (Vlaeyen and Linton, 2000; Mannion et al., 2001). Surprisingly a far greater proportion of data were excluded due to the onset of pain during the low load MVC20 test (30% of symptomatic participants, and 17% of control participants) than during the moderate load MVC50 test (15% of symptomatic participants, and 2% of control participants), or during the maximal load MVC test (no exclusions). These ﬁndings may reﬂect the far greater duration for which the gentler MVC20 contractions were sustained. Perhaps pain sensitive neck structures adversely respond to gentler sustained mechanical load more so than loads that are more substantial in magnitude, but are of shorter duration. Interestingly, a relatively high proportion of control participants (17%) also ceased the MVC20 test due to the onset of pain, these ﬁndings perhaps reﬂecting the difﬁculty participants have in distinguishing pain resulting from exercise-induced fatigue, to that perceived to be a signal of potential injury. Some participants also discontinued their test due to the onset of jaw and head pain from the sustained pressure to the mandible from the dynamometry procedure, this may account for some of the control participant exclusions. Irrespective, the high proportion of data excluded from

the MVC20 test due to pain, jeopardizes the tests clinical usefulness as measure of muscle fatigue. For all measurements (MVC, MVC20, MVC50), age was entered as a covariate in statistical analysis to account for the neck pain group being older (9.2 years) than the control group. With regard to the MVC measures, isometric cervical spine strength has been shown to decrease with age (Jordan et al., 1999; Garces et al., 2002), however perhaps not signiﬁcantly so until the seventh decade (Jordan et al., 1999; Ylinen et al., 2004). Barber (1994), using the same isometric dynamometry method for the CCF muscles as Watson and Trott (1993), demonstrated no signiﬁcant effects of age on CCF muscle strength in women over a 20–65 years age range. In our study, age did not have a signiﬁcant effect on the MVC peak torque measures (P ¼ 0:14) or for any of the MVC20 measures (P40:36). However, age was a signiﬁcant factor for the MVC50 measure of time to task failure (P ¼ 0:04), but did not alter betweengroup differences in this measure. It should be noted that the neck pain participants used in this study were of a mild neck pain severity according to the NDI score (Vernon, 1996). Participants had an average NDI of 22.8 points out of a possible 100. We chose to use participants with only mild symptoms to optimally scrutinize the presence of CCF muscle impairment in neck pain. We could reasonably expect that participant populations with greater NDI scores would show larger differences in measurements.

5. Conclusion This study found deﬁcits in isometric CCF muscle performance across a spectrum of contraction intensities in persons with neck pain. These ﬁndings suggest that rehabilitative exercise of the CCF muscles in neck pain sufferers may need to incorporate various intensities, durations, and precision of CCF muscle contraction efforts.

References Barber A. Upper cervical spine ﬂexor muscles: age related performance in asymptomatic women. Australian Journal of Physiotherapy 1994;40:167–72. Chiu TT, Lo SK. Evaluation of cervical range of motion and isometric neck muscle strength: reliability and validity. Clinical Rehabilitation 2002;16:851–8. Conley MS, Stone MH, Nimmons M, Dudley GA. Speciﬁcity of resistance training responses in neck muscle size and strength. European Journal of Applied Physiology 1997;75:443–8. Falla D, Jull G, Hodges P. Feedforward activity of the cervical ﬂexor muscles during voluntary arm movements is delayed in chronic neck pain. Experimental Brain Research 2004a;157:43–8. Falla D, Jull G, Hodges P. Patients with neck pain demonstrate reduced electromyographic activity of the deep cervical ﬂexor

ARTICLE IN PRESS S. O’Leary et al. / Manual Therapy 12 (2007) 34–39 muscles during performance of the craniocervical ﬂexion test. Spine 2004b;29(19):2108–14. Gabriel DA, Matsumato JY, Davis DH, Currier BL, An K-N. Multidirectional neck strength and electromyographic activity for normal controls. Clinical Biomechanics 2004;19:653–8. Gandevia SC. Spinal and supraspinal factors in human muscle fatigue. Physiological Reviews 2001;81(4):1725–89. Garces G, Medina D, Milutinovic L, Garavote P, Guerado E. Normative database of isometric cervical strength in a healthy population. Medicine and Science in Sports and Exercise 2002;33:464–70. Grimmer K, Trott P. The association between cervical excursion angles and cervical short ﬂexor muscle endurance. Australian Journal of Physiotherapy 1998;44:201–7. Harms-Ringdahl K, Ekholm J, Schuldt K, Nemeth G, Arborelius U. Load moments and myoelectric activity when the cervical spine is held in full ﬂexion and extension. Ergonomics 1986;29(12): 1539–52. Hunter SK, Duchateau J, Enoka RM. Muscle fatigue and mechanisms of task failure. Exercise and Sports Sciences Reviews 2004;32(2):44–9. Janda V. Muscle and cervicogenic pain syndromes. In: Grant RE, editor. Physical therapy for the cervical and the thoracic spine. New York: Churchill Livingstone; 1988. p. 153–66. Jordan A, Mehlsen J, Bulow PM, Ostergaard K, Danneskiold-Samsoe B. Maximal isometric strength of the cervical musculature in 100 healthy volunteers. Spine 1999;24(13):1343–8. Jull G. Examination of the articular system. In: Boyling JD, Palastanga N, editors. Grieve’s modern manual therapy. 2nd ed. Edinburgh: Churchill Livingstone; 1994. p. 511–27. Jull G, Barrett C, Magee R. Further characterization of muscle dysfunction in cervical headache. Cephalalgia 1999;19: 179–85. Jull G, Trott P, Potter H, Zito G, Niere K, Shirley D, et al. A randomised controlled trial of exercise and manipulative therapy for cervicogenic headache. Spine 2002;27(17):1835–43. Jull G, Falla D, Treleavan J, Sterling M, O’Leary S. A therapeutic exercise approach for cervical disorders. In: Boyling JD, Jull G, editors. Grieve’s modern manual therapy. 3rd ed. Edinburgh: Churchill Livingstone; 2004a. p. 451–69. Jull G, Kristjansson E, Dall’Alba P. Impairment in the cervical ﬂexors: a comparison of whiplash and insidious onset neck pain patients. Manual Therapy 2004b;9:89–94. Jull GA. Headaches of cervical origin. In: Grant RE, editor. Physical therapy of the cervical and thoracic spine, vol. 17. New York: Churchill Livingstone; 1988. p. 195–217. Jull GA. Deep cervical ﬂexor muscle dysfunction in whiplash. Journal of Musculoskeletal Pain 2000;8:143–54. Kanehisa H, Miyashita M. Speciﬁcity of velocity in strength training. European Journal of Applied Physiology 1983;52: 104–6. Kettler A, Hartwig E, Schultheiss M, Claes L, Wilke HJ. Mechanically simulated muscle forces strongly stabilize intact and injured upper cervical spine specimens. Journal of Biomechanics 2002;35(3):339–46. Kumar S, Narayan Y, Amell T. Cervical strength of young adults in sagittal, coronal, and intermediate planes. Clinical Biomechanics 2001;16:380–8. Mannion AF, Taimela S, Muntener M, Dvorak J. Active therapy for chronic low back pain. Part 1. Effects on back muscle activation, fatigability, and strength. Spine 2001;26(8):897–908.

39

Mayoux Benhamou MA, Wybier M, Revel M. Strength and crosssectional area of the dorsal neck muscles. Ergonomics 1989;32:513–8. Montagu MF. An introduction to physical anthropology. 3rd ed. Springﬁeld: Charles C Thomas; 1960. Moore KL, Dalley AF. Clinically orientated anatomy, 4th ed. Philadelphia: Lippincott Williams & Wilkins; 1999. p. 1025–7 [chapter 8]. Norton K, Olds T. Anthropometrica. Sydney: University of New South Wales Press; 1996. p. 37 [chapter 2]. O’Leary S. Development of a new method of measurement of craniocervical ﬂexor muscle performance. PhD thesis, Physiotherapy Division, University of Queensland, Brisbane, 2005. O’Leary S, Jull G, Vicenzino B. Do dorsal head contact forces have the potential to identify impairment during graded craniocervical ﬂexor muscle contractions? Archives of Physical Medicine and Rehabilitation 2005a;86:1763–6. O’Leary S, Vicenzino B, Jull G. A new method of isometric dynamometry for the craniocervical ﬂexor muscles. Physical Therapy 2005b;85:556–64. Peolsson A, Oberg B, Hedlund R. Intra- and inter-tester reliability and reference values for isometric neck strength. Physiotherapy Research International 2001;6(1):15–26. Portero P, Genries V. An update of neck muscle strength: from isometric to isokinetic assessment. Isokinetics and Exercise Science 2003;11:1–8. Portero P, Bigard AX, Gamet D, Flageat JR, Guezennec CY. Effects of resistance training in humans on neck muscle performance, and electromyogram power spectrum changes. European Journal of Applied Physiology 2001;84(6):540–6. Rasch P, Morehouse L. Effect of static and dynamic exercises on muscular strength and hypertrophy. Journal of Applied Physiology 1957;11:29–34. Sale D, MacDougall D. Speciﬁcity in strength training: a review for the coach and athlete. Canadian Journal of Applied Sport Sciences 1981;6:87–92. Seng K-Y, Peter V-S, Lam P-M. Neck muscle strength across the sagittal and coronal planes: an isometric study. Clinical Biomechanics 2002;17:545–7. vanMameran H, Sanches H, Beursgens J, Drukker J. Cervical spine motion in the sagittal plane II: position of segmental averaged instantaneous centers of rotation—a cineradiographic study. Spine 1992;17(5):467–74. 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(4):412–22. Vernon H. The neck disability index: patient assessment and outcome monitoring in whiplash. Journal of Musculoskeletal Pain 1996;4(4):95–104. Vlaeyen JWS, Linton SV. Fear-avoidance and its consequences in chronic musculoskeletal pain: a state of the art. Pain 2000;85:317–32. Watson DH, Trott PH. Cervical headache: an investigation of natural head posture and upper cervical ﬂexor muscle performance. Cephalalgia 1993;13:272–84. White AA, Panjabi MM. Clinical biomechanics of the spine. 2nd ed. Philadelphia: J.B. Lippincott Company; 1990. p. 95–7 [chapter 2]. 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.

ARTICLE IN PRESS

Manual Therapy 12 (2007) 40–49 www.elsevier.com/locate/math

Original article

The clinical reasoning of pain by experienced musculoskeletal physiotherapists Keith Smarta,, Catherine Doodyb a

Physiotherapy Department, St Vincent’s University Hospital, Elm Park, Dublin 4, Ireland b School of Physiotherapy, University College Dublin, Ireland

Received 25 February 2005; received in revised form 26 January 2006; accepted 15 February 2006

Abstract There is currently no research within Physiotherapy to explain the extent to which current theories and models of pain inﬂuence clinicians’ reasoning related to clinical presentations of pain. The purpose of this qualitative study was to investigate the clinical reasoning of experienced musculoskeletal physiotherapists in relation to three different presentations of pain. A qualitative multiple-case studies method was used in this study. A purposive sample of seven experienced musculoskeletal physiotherapists viewed three videotaped patient-therapist clinical interviews describing three different pain presentations. An audio taped, semi-structured interview was carried out with each participant during which the participants were encouraged to verbalize their thoughts regarding aspects of each patient’s pain presentation. All interviews were subsequently transcribed, coded and analysed. Results showed a dynamic, multidimensional nature to the therapists’ clinical reasoning, which was found to be grounded in a number of established models of pain. Five main categories of pain-based clinical reasoning were identiﬁed. These were (i) biomedical, (ii) psychosocial, (iii) pain mechanisms, (iv) chronicity and (v) irritability/severity. Reasoning within these categories inﬂuenced therapists’ prognostic decision-making as well as the planning of physical assessments and treatment. The clinical reasoning of pain by the participants in this study appeared to reﬂect the integration of diverse models and theories of pain into current clinical practice. Mechanisms-based clinical reasoning has not been previously observed amongst physiotherapists. r 2006 Elsevier Ltd. All rights reserved. Keywords: Clinical reasoning; Pain; Physiotherapists

1. Introduction It has been suggested that pain is the main symptom with which patients present to musculoskeletal physiotherapists (Watson, 1996; Cheing and Cheung, 2002). It has also been suggested that within physiotherapy considerable importance is placed upon the patient’s report of pain, as elicited through the patient-therapist clinical interview, by physiotherapists when reasoning and decision-making with regard to its nature and treatment (Main and Watson, 1999). Corresponding author. Tel.: +353 1 2094354; fax: +353 1 2696018.

E-mail address: [email protected] (K. Smart). 1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.02.006

Epidemiological studies and reviews suggest that between 10% and 20% of the populations of industrialized western societies have a persistent pain problem (Magni et al., 1993; Verhaak et al., 1998; Pain in Europe, 2003). The personal cost to sufferers of ‘chronic pain’ can be devastating. One in ﬁve chronic pain sufferers have lost a job as a result of their pain, and one in ﬁve sufferers have been diagnosed with depression as a result of their pain (Pain in Europe, 2003). Financially, the costs to societies both directly (through expenditure on healthcare) and indirectly (through lost productivity and tax revenue and disability compensation) run into the billions (US dollars) (Turk, 2002).

ARTICLE IN PRESS K. Smart, C. Doody / Manual Therapy 12 (2007) 40–49

1.1. Theories of pain A review of the literature shows a variety of theories and clinically applicable models to account for the experience and presentations of pain. The Cartesian/ Medical model and speciﬁcity theories of pain explain pain as a direct correlate of physical disease or injury (Melzack and Wall, 1991). The gate control theory of pain (Melzack and Wall, 1965) described the neurophysiological mechanisms of pain transmission and modulation centring on the dorsal horn of the spinal cord. The gate control theory of pain together with more contemporary approaches such as the biopsychosocial model (Waddell, 1998), and mechanisms-based approaches (Jones, 1995; Gifford and Butler, 1997) have contended the usefulness of earlier theories, such as the medical model, by attempting to explain and account for the often variable and inconsistent relationship between pathology and pain. 1.2. Clinical reasoning and pain For physiotherapists, a hypothesis-oriented, mechanisms-based approach to the clinical reasoning and categorization of pain has been proposed (Jones, 1995; Gifford and Butler, 1997). This approach suggests that clinical presentations of pain may be categorized according to ﬁve classes of pain mechanisms: (1) nociceptive, (2) peripheral neurogenic, (3) central pain, (4) autonomic and motor mechanisms and (5) affective mechanisms. This approach has been based upon the perceived limitations of the medical model of pain and illness and recent advances in understanding of the neurophysiological basis of pain. According to Gifford and Butler (1997) a pain mechanisms approach could aid judgements regarding the assessment, treatment and prognosis of patients’ pain. The ‘mature organism model’ proposed by Gifford (1998) further expanded the mechanisms-based approach by integrating knowledge of the neurophysiological mechanisms of pain with the science of stress biology and the biopsychosocial model of pain and disability. The ‘mature organism model’ describes the numerous and interrelated biological systems and processes involved in the initiation, maintenance and perception of pain together with the physiological and behavioural reactions to it. This type of broad understanding of pain, the model suggests, is required in order that clinical presentations of pain might be better managed. Whilst the clinical reasoning of pain has not been directly studied in physiotherapy, a limited body of literature exists with respect to the study of pain knowledge amongst physiotherapists and health professionals. Wolff et al. (1991) in their survey of 500 orthopaedic physiotherapists, conducted by postal

41

questionnaire, found speciﬁc deﬁciencies in clinicians’ knowledge regarding pain mechanisms. In addition, 72% of respondents perceived their graduate/entry level of pain education, with respect to pain theory and management, as very inadequate or less than adequate. Moseley (2003) also found health professionals to have poor knowledge of the neurophysiology of pain but that with appropriate training were capable of improving their understanding. Rivett and Higgs (1997), in a study of the hypothesis categories used by 19 manual therapists, found no evidence for clinical reasoning concerning neurophysiological pain mechanisms. Other authors have also highlighted the discrepancy that exists between published information on the neurobiology and psychosociology of pain and the knowledge and actions of clinicians (Woolf and Decosterd, 1999; Champion, 2000). Theories and models of pain provide a conceptual framework with which to investigate and interpret current methods of clinical reasoning of pain. It has been argued that within manual therapy (Main and Watson, 1999) and medicine generally (Waddell and Main, 1998) clinical reasoning with respect to pain remains dominated by the medical model with its structure/pathology-oriented explanations of pain and disability. However clinical reasoning in relation to pain in physiotherapy practice has not been subject to focused investigative study. The extent to which clinicians incorporate and utilize other theories and models of pain into clinical practice, such as the biopsychosocial model (Waddell and Main, 1998) or mechanism-based methods (Jones, 1995; Gifford and Butler, 1997) has not been studied in physiotherapy. The aims of this study were: 1. To investigate the clinical reasoning processes of experienced musculoskeletal physiotherapists in relation to three different pain presentations. 2. To determine how such reasoning may inform or inﬂuence other areas of clinical decision-making in physiotherapy.

2. Method 2.1. Study design A qualitative multiple case studies design was used in this investigation (Yin, 1994). Yin (1994) deﬁnes a case study as ‘An empirical inquiry that investigates a contemporary phenomenon within its real-life context’. According to Yin (1994) case study inquiry can be guided by the theoretical propositions that lead to the study. Current models and theories of pain as outlined in the introduction provided the conceptual framework

ARTICLE IN PRESS 42

K. Smart, C. Doody / Manual Therapy 12 (2007) 40–49

for this investigation and were the guiding propositions for the study. 2.2. Subjects and setting A purposive sample of seven experienced physiotherapists took part in the study. In this multiple-case study the ‘cases’ were seven experienced musculoskeletal physiotherapists and the units of analysis were the therapists’ clinical reasoning processes. All subjects had a minimum of 10 years experience in musculoskeletal physiotherapy and had engaged in formal postgraduate study. Proﬁles of study participants are displayed in Table 1. The study took place in the physiotherapy department of a large teaching hospital in Dublin. Approval was obtained from the hospital’s Ethics Committee prior to the commencement of the study. All patients and physiotherapists gave signed consent prior to their participation in the study. All physiotherapists were told the purpose of the study was to gain access to their thoughts regarding aspects of different patients’ pain presentations. 2.3. Procedure Multiple data collection methods were used in this study and included audio taped semi-structured interviews, participant personal proﬁle data and recorded ﬁeld notes. The use of multiple sources of evidence allows for triangulation of data sources (Yin, 1994). Data collection and analysis was carried out by one researcher who was a senior musculoskeletal physiotherapist. Each physiotherapist viewed, in the same sequence, three separate videotaped patient-therapist clinical interviews describing three different pain presentations. Patient 1 presented with chronic low back and leg pain, patient 2 with Complex Regional Pain Syndrome (CRPS Type I) 3 months post distal radius fracture and patient 3 with an acute ankle sprain. Patients with differing presentations of pain were chosen in order to allow for study participants to reveal potentially diverse methods of reasoning. Table 1 Proﬁles of study participants N ¼ 7 Mean years since qualiﬁcation Practice setting

Postgraduate Teaching experience Formal research experience

16.1 (range 12–25) 3 hospital outpatient dept., 3 private practice, 1 private practice/university lecturer 5 taught MSc, 2 taught MSc in progress 1 undergraduate, 1 postgraduate, 3 under/postgraduate, 2 none 4

Before viewing the three videotaped patient-therapists interviews, each study participant was given written instructions similar to those used by Barrows and Tamblyn (1980) whereby they were asked to verbalize their thoughts regarding the nature of the patients’ pain presentation. As participants viewed each of the three pain presentations, an audio taped semi-structured interview was carried out with the principal researcher. At pre-determined intervals the video recordings were paused at which point the researcher asked open-ended questions in order to stimulate participants to verbalize their thoughts regarding aspects of each patient’s pain presentation. Further questioning was designed to encourage participants to relate aspects of pain reasoning to clinical decision-making associated with prognosis, physical examination and treatment. An audio taped semi-structured interview format was used in order to generate verbal protocols (Patel and Arocha, 2000) and each interview was subsequently transcribed (Ericcson and Simon, 1984). The transcripts combined each patient-therapist interview with the subjects’ comments. Each of the seven participants generated three transcripts. These 21 transcriptions along with the seven physiotherapist proﬁle documents and ﬁeld notes formed the ﬁnal case study database (Yin, 1994). In order to maintain a chain of evidence (Yin, 1994) each transcript was given both a subject and patient presentation number. In addition each transcript was line numbered to allow for identiﬁcation of all citations from the case study database. 2.4. Analysis Data analysis was guided by the approach described by Miles and Huberman (1994). The process of data analysis allows for the development of analytic and conceptual frameworks for the purpose of generating description and theory of the phenomenon under investigation (Creswell, 1998). According to (Gwyer et al., 2004), ‘a ﬁnal component of the multiple case study design is the development of grounded theory or theory building’. All case study reports were read and re-read a number of times in order to obtain an overall sense of the data (Creswell, 1998). A tentative coding framework was devised based on theories and models of pain from the literature previously outlined, and subsequently expanded as further themes and categories emerged from the expanding database of transcripts. The coding scheme was thus revised to reﬂect the multiple perspectives of study participants. In the ﬁrst stage of analysis ﬁve primary codes were used (Table 2). In the second stage of analysis a further three secondary level codes (Table 3) were used. Inter-coder and intra-coder reliability of the coding scheme was checked on a random selection of transcripts and was calculated using

ARTICLE IN PRESS K. Smart, C. Doody / Manual Therapy 12 (2007) 40–49

43

Table 2 Deﬁnition of primary level codes with examples

Table 3 Deﬁnition of secondary level codes with examples

Code

Deﬁnition and examples

Code

Deﬁnition and examples

Biomedical

Reasoning of pain related to: (a) its structural source or anatomical distribution, (b) biomechanical principles, (c) its aetiology, (d) pathological processes (e.g. inﬂammation) and (e) diagnostic labelling. Example: ‘‘there’s a very mechanical nature to his pain’’ (7)

Objective

Reasoning of pain as related to aspects of the physical examination of patients. Example: ‘‘I am going to choose in my physical examination to reproduce it’’ (4)

Prognosis

Reasoning of pain related to concurrent thinking concerning a patient’s prognosis or outcome. Example: ‘‘prognostically it may take longer to settle’’ (7)

Treatment

Reasoning of pain related to reasoning surrounding issues of physiotherapeutic and/or other intervention Example: ‘‘Four years on, I think we would be looking at rehabilitation as oppose to getting rid of her pain’’ (2)

Psychosocial

Pain mechanisms

Reasoning of pain related to patients’ (a) thoughts (cognitions), (b) feelings (emotions), (c) behaviours, (d) attitudes and coping styles, and (e) to sociological factors such as work, family, social life, ﬁnancial issues, and their impact on any pain presentation. Example: ‘‘She is afraid of moving it so there is an element of fear avoidance’’ (1) Reasoning of pain related to the underlying pathophysiological mechanisms responsible for its generation and/or maintenance. More speciﬁcally this relates to reasoning surrounding the ﬁve classes of pain (nociceptive, peripheral neurogenic, central neurogenic, autonomic/ motor and affective) as outlined by Jones (1995) and Gifford and Butler (1997). Example: ‘‘it’s almost 100% peripheral nociceptive in origin’’ (3)

Chronicity

Reasoning related to the temporal aspects of pain as indicated by the use of descriptors such as ‘chronic’ or ‘acute’. Example: ‘‘Well she’s a chronic pain state’’ (5)

Severity/ irritability

Reasoning of pain as related to it’s (a) severity, i.e. to the degree of pain (Maitland, 1991) or its intensity (Maitland, 1986) and (b) its ‘irritability’, as conceptualised and deﬁned by Maitland (1986), that is a disorder’s susceptibility to become painful, how painful it becomes and the length of time this pain takes to subside. Example: ‘‘I would go quite a bit on the severity and irritability of somebody’s condition’’ (2)

the formula suggested by Miles and Huberman (1994). Intra-coder reliability showed 88% agreement. Intercoder reliability was checked by two separate coders, the primary researcher and one other musculoskeletal physiotherapist and showed 78% agreement. The Kappa coefﬁcients for intra and inter-coder agreement were 0.83 and 0.76 suggesting excellent and good agreement respectively (Daly and Bourke, 2000). The third stage of analysis involved within-case analysis where each case was analysed across all three settings or patient presentations. The fourth stage of analysis involved a cross case analysis across the seven cases and three settings. A number of data analysis methods, as described by Miles and Huberman (1994),

were utilized, including, noting patterns and themes, making contrasts and comparisons, and noting relations between variables. Six veriﬁcation procedures, as outlined by Creswell (1998), were used in order to ensure the integrity of the research including, prolonged engagement and persistent observation, peer review or debrieﬁng as an external check of the research process, triangulation of data through use of multiple data sources, negative case analysis (to purposely seek any inconsistent or disconﬁrming evidence for conclusions), rich, thick description and clarifying researcher bias, such that the researcher’s background, position and motivation is known and how these may impact upon the inquiry.

3. Results Data are presented in the form of verbatim quotes. All citations are followed with a number from 1 to 7, corresponding to each participant physiotherapist, in order to give a sense of the spread of the data. Five main categories of pain-oriented reasoning were identiﬁed in this investigation. These were; biomedical, psychosocial, pain mechanisms, chronicity and severity/irritability. In addition, reasoning within all categories occurred interchangeably and simultaneously. 3.1. Biomedical All seven therapists demonstrated extensive biomedical-oriented reasoning of pain in relation to each of the three patient presentations. Biomedical-oriented reasoning included reasoning of pain related to structural/anatomical source, biomechanical principles, aetiology, pathological processes and diagnostic labelling.

ARTICLE IN PRESS K. Smart, C. Doody / Manual Therapy 12 (2007) 40–49

44 Table 4 Biomedical reasoning

Table 5 Psychosocial reasoning

Subclassiﬁcation

Examples

Sub-classiﬁcation

Examples

Cognitions/thoughts

Structure/ Source

‘‘She’s mapping it out quite clearly, it’s maybe indicating a level of the lumbar spine that might be of interest to me.’’(6)

‘‘And the factors that inﬂuence the pain would be like you know the thoughts the belief system’’(2)

Emotions

Biomechanical

‘‘I would think of mechanical pain as pain that is aggravated by certain activities and eased by rest’’(2)

‘‘It’s the emotive component of her pain that is striking me mostybut I’m thinking that her whole response to her fracture is very emotive rather than the commoner pain descriptors that we hear.’’(3)

Aetiology

‘‘Well the mechanism of injury is the sudden macro traumayI didn’t catch which way he went but it’s probably an inversion type of strain.’’ (3)

Behaviours

‘‘He doesn’t have fear avoidance. He is a little bit nervous himself which I would perceive as within expected limits.’’(1)

‘‘There may be an element of degeneration or early degeneration here’’ (4)

Attitudes and coping

‘‘The pain seems to be a big obstacle to her moving on with doing the exercises. I’m fearful of her getting complex regional pain syndrome.’’(5)

‘‘Well he’s coping with the pain pretty well, he’s not letting it interfere too much.’’(5)

Sociological

‘‘As well as ﬁnancial implications she is looking at her career implications, which are again factors that will inﬂuence her pain’’ (2)

Pathology Diagnostic

Examples of biomedical based pain reasoning are displayed in Table 4. In addition, biomedical based reasoning appeared to inﬂuence four participants’ prognostic reasoning. For example, one participant stated in relation to biomechanical principles: ‘‘He has a pretty good prognosis because he has a simple mechanical type presentation.’’ (5) Biomedical based reasoning was also found to inﬂuence thinking concerning treatment and also in the planning or implementation of the physical examination. For example, one therapist suggested: ‘‘Already in my mind I’m thinking of the structure at faultymy physical examination would be directed at that structure.’’ (7) In relation to biomedical reasoning, three study participants commented on the apparent ‘normality’ of patient presentation 3. For example: ‘‘There is a very mechanical nature to this pain ywhat he’s describing is ﬁtting with the structures that you would think would be at fault, so again it’s ﬁtting a normal case presentation.’’ (7) 3.2. Psychosocial All seven therapists demonstrated extensive and diverse psychosocial oriented reasoning in relation to all three presentations of pain. This included reasoning related to patients’ cognitions, emotions, behaviours, attitudes and coping styles and sociological factors as shown in Table 5.

Reasoning associated with the psychosocial aspects of pain appeared to be signiﬁcant to six study participants when considering approaches to treatment, as illustrated by one therapist: ‘‘So now my management strategy is going to have to consider her home life, her work life, her capacity to live her life and to manage her pain.’’ (6) In addition, reasoning associated with the psychosocial aspects of pain appeared to inform the prognostic decision-making of ﬁve participants, whereby the presence or absence of psychosocial factors appeared to act as negative and positive prognostic indicators respectively, as illustrated by the following comments: ‘‘From the psychosocial factors there, it has interfered with her everyday activities. She can’t work at a job she enjoys, she’s been missing out on socialising as well, so all that’s going to hamper recovery. (5) If he was starting to say some of the things that some of the other patients said like ‘‘I can’t do this, I’m off work, I can’t get back’’, if things were starting to progress like that then I’d be really worried about his prognosis. Alternatively the fact that he has very little in the way of psychosocial components to this problem, that’s all going well for his prognosis.’’ (4) 3.3. Pain mechanisms All seven study participants demonstrated clinical reasoning associated with the neurophysiological basis of pain, speciﬁcally nociceptive, peripheral neurogenic,

ARTICLE IN PRESS K. Smart, C. Doody / Manual Therapy 12 (2007) 40–49

central and autonomic/sympathetic mechanisms of pain. Five of the seven study participants showed evidence of reasoning directly related to nociceptive mechanisms of pain. Such reasoning was expressed more often in relation to patient presentation 3. For example one therapist stated: ‘‘It’s a soft tissue injuryyHe’s still got quite a bit of inﬂammation thereyso it’s a nociceptive type of pain mechanism, pain presentation.’’ (5) There was also some evidence that reasoning surrounding nociceptive pain mechanisms was closely integrated with various subcategories of biomedical based reasoning, as demonstrated by one therapist: ‘‘There’s a mechanical nature to his pain he’s describing speciﬁc activities that would be telling me that there’s a nociceptive source of pain.’’ (7) Five of the seven therapists demonstrated clinical reasoning related to peripheral neurogenic mechanisms of pain, almost exclusively in relation to patient presentation 1. All seven therapists showed evidence of reasoning with regard to central mechanisms of pain and were expressed in relation to patient presentations 1 and 2. For example, one therapist explained: ‘‘From a pain processing point of view I would say that she’s probably got a primary and secondary hyperalgesia happening andyshe has an ongoing pain presentation. So rather than a peripheral nociceptive source, I would deﬁnitely be thinking of some kind of central component happening, central nervous system pain.’’ (7) Reasoning with respect to autonomic/sympathetic mechanisms of pain was demonstrated by all seven study participants, exclusively in relation to patient presentation 2. Such reasoning was also expressed in conjunction with diagnostic based reasoning, as suggested by the following therapist: ‘‘Well again it’s ﬁtting into the sympathetic element of a causalgia or an RSD.’’ (4) There was some evidence that mechanism-based reasoning appeared to inﬂuence participants’ thinking surrounding aspects of the physical examination. Three study participants described how the presence of neurogenic mechanisms would lead them into conducting a more cautious physical examination. Mechanismbased reasoning of pain also appeared to inﬂuence four participants’ prognostic reasoning, where nociceptive and central mechanisms of pain were associated with more and less favourable prognoses correspondingly. In relation to treatment, the identiﬁcation of predominantly neurogenic and central mechanisms of pain by four participants appeared to discourage the use of predominantly manual therapy based approaches.

45

3.4. Chronicity Six of the seven study participants showed evidence of reasoning linked to the relative chronicity of pain i.e. inferences regarding the duration of patients’ pain. The study participants appeared to hold clear and relatively consistent frames of reference as to the cut-off time after which pain may be labelled or deﬁned as ‘chronic’. i.e. after 3–6 months duration. There was evidence of less consistency between subjects regarding a suitable cut off time by which to distinguish or deﬁne ‘acute’ pain, ranging from less than 6 weeks to less than 6 months duration. In addition, there was some evidence of the integration of chronicity based reasoning with biomedical, psychosocial and pain mechanisms oriented reasoning. For example one subject highlighted the inﬂuence of a patient’s thought processes: ‘‘She’s constantly thinking about her back, so that just feeds into chronicity.’’ (1) 3.5. Severity/irritability All seven study participants showed evidence of reasoning with respect to the severity and irritability of patients’ pain, as conceptualized by Maitland (1986, 1991). Judgements about the severity and irritability of patients’ pain appeared to be based on cues derived from the pain reports, particularly night pain and sleep disturbance, aggravating and easing factors of pain and the use of medication. In addition, such judgements were in some instances graded by participants using descriptors such as mild, moderate and severe. For example, one participant stated: ‘‘So the aggravating and relieving factors are giving me an indication of his irritability which I think is mild to moderate.’’ (6) Reasoning associated with the irritability (three participants) and severity (two participants) of patients’ pain also appeared to inﬂuence thinking concerning the extent of any physical examination, as evidenced by the following comment: ‘‘I am not overly concerned about aggravating her during the physical examination part. I mean that would be one of the main reasons why you would want to assess irritability up front.’’ (4) In addition there was limited evidence from two participants that reasoning related to the irritability and severity of patients’ pain had an inﬂuence on decisionmaking concerning the planned extensiveness of subsequent treatment.

ARTICLE IN PRESS K. Smart, C. Doody / Manual Therapy 12 (2007) 40–49

46 Table 6 Summary of results Category Biomedical

Psychosocial

Mechanisms

Chronicity

Sev/irrit

Structure biomechanical aetiology pathology diagnosis

Cognitive emotive behavioural attitudes sociological factors

Nociceptive neurogenic central autonomic/ sympathetic

Chronic acute

Severity irritablity

7 1,2,3 O

7 1,2,3 X

7 1,2,3 O

6 1,3 X

7 1,2,3 O

O O

O O

O O

X X

O X

Sub-category

No. of therapists Settinga Implications for Phys. Ex. Treatment Prognosis a

Setting ¼ patient presentation.

3.6. Summary The results of this study suggest a broad and multidimensional nature to the clinical reasoning of pain by the experienced musculoskeletal physiotherapists observed in this study; which appeared to be grounded within a number of different theories and models of pain. Reasoning within categories appeared to occur simultaneously and interchangeably suggesting a dynamic reasoning process associated speciﬁcally with the clinical reasoning of pain. The main categories of pain based reasoning identiﬁed are summarized in Table 6. The therapists clinical reasoning of pain appeared to inﬂuence reasoning associated with other aspects of clinical practice such as the planning of physical examinations and treatment, and prognostication.

4. Discussion Evidence was found from all participants that the clinical reasoning of pain was grounded, in part, within what has been termed the Medical/Disease model of pain and illness (Waddell and Main, 1998; Main and Watson, 1999). The medical model refers to a tissue and pathology oriented approach towards the explanation of pain and dysfunction (Watson, 2000) and functions on the premise that all pain has a dominant tissue or structural source. These ﬁndings of this study lend some support to the assertion by Gifford (1998) that physiotherapists often attempt to validate somatic tissues and nerves as deﬁnitive sources of patients’ pain. Biomedical oriented reasoning has also been identiﬁed in other studies investigating the clinical reasoning of physiotherapists, however the focus of these studies was physiotherapists’ pain beliefs (Daykin and Richardson,

2004), diagnostic reasoning (King and Bithell, 1998), the identiﬁcation of hypothesis categories (Rivett and Higgs, 1997) and the investigation of hypotheticodeductive reasoning (Payton, 1985) rather than the clinical reasoning of pain speciﬁcally. In addition, the ﬁndings from this study also provide some evidence that clinicians made inferences regarding the ‘normality’ of some presentations of pain based on reasoning grounded within the medical model. The description of some presentations of pain as ‘normal’ may suggest an implicit view on the part of some clinicians that presentations of pain that do not readily ﬁt the medical model may in turn be regarded as ‘abnormal’ and perhaps, therefore, somehow more difﬁcult to understand or explain. That some presentations of pain may be regarded in this way may perhaps reﬂect the extent to which the ‘traditional’ medical/ disease model (Waddell and Main, 1998) continues to inﬂuence clinical thinking, possibly as a result of physiotherapists’ prior education. Substantial evidence was found from all study participants of reasoning related to the psychosocial aspects of pain consistent with the biopsychosocial model of pain and disability (Waddell and Main, 1998). This model recognizes that the clinical expression of pain and any resultant disability is multiply determined by the interaction of physical disease with psychological and social factors. Speciﬁcally, the participants in this study were found to recognize and acknowledge the importance of the cognitive, emotive, behavioural, attitudinal and sociological aspects of patients’ pain, suggesting that the multiple determinants of and inﬂuences on patients’ experience of pain were appraised for each patient. This ﬁnding appears to provide evidence that the therapists in this study adopted a more holistic approach towards the interpretation of patients’ pain. Evidence of psychosocial oriented

ARTICLE IN PRESS K. Smart, C. Doody / Manual Therapy 12 (2007) 40–49

reasoning has also been found in other studies investigating clinical reasoning within physiotherapy (Payton, 1985; Embrey et al., 1996; Jensen et al., 2000; Jette et al., 2003; Edwards et al., 2004). Whilst none of these studies investigated the clinical reasoning of pain speciﬁcally they do provide some evidence that psychosocial oriented reasoning is an established part of the clinical reasoning of experienced or expert physiotherapists. Reasoning in relation to the psychosocial determinants of and inﬂuences on pain appeared to be particularly important and directly linked to ﬁve participants’ prognostic- based decision-making. The ﬁndings from this study therefore appear to reﬂect the growing consensus of opinion that psychosocial factors rather than physical factors may be more important as prognostic indicators for predicting outcome and determining which patients are most at risk of developing chronic pain (Dworkin, 1997; Waddell, 1998; Turner et al., 2000; Watson, 2000; Picavet et al., 2002). The ﬁndings from this study also appear to concur with those of Overmeer et al. (2004) whose survey found Swedish physiotherapists to be well aware of the psychosocial risk factors for developing chronic pain and disability. The results from this study provide clear evidence of mechanism-based reasoning of pain in four out of the ﬁve classes of pain mechanisms described by Jones (1995) and Gifford and Butler (1997) and therefore provide some evidence to support the use of this method of mechanism-based reasoning amongst experienced physiotherapists. Minimal evidence was found of reasoning related to the ﬁfth or affective class of pain mechanisms. Whilst attention to the emotive aspects of pain was clearly demonstrated by study participants as part of a psychosocial oriented approach to the reasoning of pain, they appeared not to consider the emotive (affective) dimension of pain from an inherently neurophysiological perspective as outlined in the literature (Price, 2002; Zusman, 2002). Reasoning within these ﬁve classes of pain mechanisms, it has been suggested, is necessary in order for clinicians to better understand clinical presentations of pain (Jones, 1995) and to inform decision making associated with the assessment, treatment and prognosis of patients’ pain problems (Gifford and Butler, 1997). The results of the present study suggest that the mechanism-based reasoning of pain employed by the experienced physiotherapists in this investigation served similar ends. Mechanism-based reasoning of pain has not previously been identiﬁed or described within the existing body of literature on clinical reasoning within (musculoskeletal) physiotherapy. A study by Rivett and Higgs (1997), which investigated the clinical reasoning of manual therapists found no evidence of reasoning related to the neurophysiological mechanisms of pain.

47

One possible explanation for the difference in ﬁndings may be the greater dissemination and awareness of knowledge within physiotherapy concerning the mechanisms of pain since the publication of the study by Rivett and Higgs (1997). Substantial evidence was found of clinical reasoning related to the relative chronicity of pain and for the use of traditional terms, such as ‘acute’ and ‘chronic’ to describe pain from the perspective of its duration (Waddell, 1998). The use and meaning of such terms generally coincides with deﬁnitions and perspectives from the literature (Waddell, 1998; Linton, 1999; Pain in Europe, 2003). Although based mainly on the duration of pain, reasoning and judgements associated with the relative chronicity of pain also appeared, at times, to be integrated with biomedical, psychosocial and mechanisms oriented reasoning. For the therapists in this study therefore, the determinants of acute or chronic pain appeared to be more multidimensional and not singularly dependant on judgements related to time scale as otherwise suggested in the literature (Waddell, 1998; Linton, 1999; Pain in Europe, 2003). Perhaps surprisingly, there was no evidence from the participants in this study that such reasoning in itself informed any subsequent decision-making related to treatment or prognosis. Chronicity based reasoning appeared to serve solely as a means to describe patients’ pain. An alternative style of questioning may have elicited different results. Despite consistent demonstrations of clinical reasoning related to the ‘irritability’ and ‘severity’ of patients’ pain by the experienced physiotherapists observed in this study, such concepts as deﬁned in this study (Table 2) appear to have no clear grounding in the wider literature on pain. Attention to the irritability and severity of patients’ pain has been advocated as an important part of assessment and treatment planning in musculoskeletal physiotherapy (Maitland, 1991). Reasoning associated with the concepts of irritability and severity of pain appear to have their origins in physiotherapy based texts (Corrigan and Maitland, 1983; Maitland, 1991, 1986) and may represent a unique conceptual approach within physiotherapy with regards to the clinical reasoning of pain. Doody (2003) found evidence of clinical reasoning related to the irritability and severity of pain in a study of expert and novice physiotherapists in an outpatient orthopaedic setting. The ﬁndings from this study show that clinical reasoning associated with the irritability and severity of patients’ pain had a clear purpose, that being to provide a conceptual framework with which to aid clinicians’ decision-making regarding the planned extensiveness of any physical examination and to a lesser degree, treatment. Such planning appeared to act a precautionary inﬂuence against the unnecessary exacerbation of patients’ pain as a direct consequence of physical assessment and/or

ARTICLE IN PRESS 48

K. Smart, C. Doody / Manual Therapy 12 (2007) 40–49

treatment procedures. The ﬁndings from this study therefore support the usefulness of irritability and severity based clinical reasoning as described by Maitland (1991). 4.1. Implications for physiotherapy The ﬁndings from this study may have implications for both physiotherapy practitioners and educators. It is recognized that within the discipline of musculoskeletal physiotherapy, clinicians place considerable importance upon the patient’s report of pain when making decisions regarding its nature, cause and treatment (Main and Watson, 1999). The ﬁndings from this study may encourage clinicians and educators to consciously reﬂect upon those theories and models of pain that inform, guide and underpin their practice procedures and teaching. Such reﬂection may lead to an appraisal of those theories and models of pain, together with the assumptions and suppositions that underlie them, against alternative or less familiar approaches to pain based reasoning and the actuality of clinical practice. A process of critical reﬂection and appraisal may then assist clinicians in their attempts to better understand the nature of patients’ pain in the hope that it might be better managed. The goal of establishing a clinical reasoning model for pain that can be shown to improve patient outcomes and promote the effective use of healthcare resources may be considered a priority given the prevalence of pain (Magni et al., 1993; Verhaak et al., 1998; Pain in Europe, 2003), its cost to society (Turk, 2002) and the personal suffering that can be its legacy (Pain in Europe, 2003). 4.2. Limitations of the study Due to the speciﬁcity of the group under investigation i.e. experienced musculoskeletal physiotherapists, the results cannot be considered transferable across other clinical specialities within physiotherapy or to other groups of physiotherapists within the musculoskeletal ﬁeld, such as clinicians without postgraduate education or with less years experience. In addition, the study design involved one researcher in data collection and analysis. Different results may have emerged with multiple researchers.

5. Conclusion The results of this study show a multidimensional nature to the clinical reasoning of pain by the experienced musculoskeletal physiotherapists observed in this study, reﬂective of the multidimensional nature of pain itself. Five main categories of pain-based clinical

reasoning were identiﬁed which were grounded in a number of models of pain. These were (1) biomedical, (2) psychosocial, (3) mechanisms, (4) chronicity and (5) irritability/severity. Reasoning within these categories appeared to be useful in helping participants understand and account for clinical presentations of pain. Such reasoning was also found to inﬂuence prognostic decision-making as well as the planning of physical assessments and treatment. Future research might seek to describe, compare and contrast the clinical reasoning of pain by physiotherapists with varying levels of experience and educational backgrounds, such as undergraduates and novice clinicians and of physiotherapists with and without formal postgraduate education. Longitudinal studies could also be carried out to investigate if and how the clinical reasoning of pain changes from an initial assessment through to discharge, to observe if such reasoning changes through clinical encounters which include treatment and reassessment. Finally and perhaps most desirably, it may also be possible to test the effectiveness of different methods of reasoning against the outcomes of clinical intervention in order to inform and guide evidenced based practice. Acknowledgements The author (KS) would like to thank the patients and physiotherapists who participated in this study and to acknowledge with gratitude receipt of the Coyle Hamilton Research Bursary, 2002 and the Chartered Physiotherapists in Manual Therapy, Research Bursary, 2003.

References Barrows HS, Tamblyn RM. Problem based learning: an approach to medical education. New York: Springer Publishing Company; 1980. Champion DG. Emerging inﬂuences of pain medicine on clinical reasoning. Clinical Therapeutics 2000;41:8–11. Cheing GLY, Cheung KSH. Placebo analgesia: clinical considerations. Physiotherapy 2002;88:735–43. Corrigan B, Maitland GD. Practical orthopaedic medicine. Oxford: Butterworth Heinneman; 1983. Creswell JW. Five qualitative traditions of inquiry. Qualitative inquiry and research design. Choosing among ﬁve traditions. Thousand Oaks: Sage Publications; 1998. Daly LE, Bourke GJ. Interpretation and uses of medical statistics. Oxford: Blackwell Science; 2000. Daykin AR, Richardson B. Physiotherapists’ pain beliefs and their inﬂuence on the management of patients with chronic low back pain. Spine 2004;29:783–95. Doody C. The clinical reasoning of expert and novice physiotherapists in an outpatient orthopaedic setting. University College Dublin. PhD Thesis, 1050. M0132238UD, 2003. Dworkin RH. Which individuals with acute pain are most likely to develop a chronic pain syndrome? Pain Forum 1997;6:127–36.

ARTICLE IN PRESS K. Smart, C. Doody / Manual Therapy 12 (2007) 40–49 Edwards I, Jones M, Carr J, Braunack-Mayer A, Jensen GM. Clinical reasoning strategies in physical therapy. Physical Therapy 2004;84:312–35. Embrey DG, Gutherie MR, White OR, Dietz J. Clinical decision making by experienced and inexperienced paediatric physical therapists for children with diplegic cerebral palsy. Physical Therapy 1996;76:20–33. Ericcson KS, Simon HA. Protocol analysis: verbal reports as data. Cambridge, MA: The MIT Press; 1984. Gifford L. Pain, the tissues and the nervous system: a conceptual model. Physiotherapy 1998;84:27–36. Gifford LS, Butler DS. The integration of pain science into clinical practice. The Journal of Hand Therapy 1997;10:86–95. Gwyer J, Jensen G, Hack L, Shepard K. Using a multiple case studies design to develop an understanding of clinical expertise in physical therapy. In: Whalley Hammell K, Carpenter C, editors. Qualitative research in evidence-based rehabilitation. Edinburgh: Churchill Livingstone; 2004. p. 103–15 [chapter 9]. Jensen GM, Gwyer J, Shepard KF, Hack LM. Expert practice in physical therapy. Physical Therapy 2000;80:28–52. Jette DU, Grover L, Keck CP. A qualitative study of clinical decision making in recommending discharge from the acute care setting. Physical Therapy 2003;83:224–36. Jones M. Clinical reasoning and pain. Manual Therapy 1995;1:17–24. King CA, Bithell C. Expertise in diagnostic reasoning: a comparative study. British Journal of Therapy and Rehabilitation 1998;5:78–87. Linton SJ. Prevention of disability due to chronic musculoskeletal pain. In: Wall PD, Melzack R, editors. Textbook of pain. 4th ed. Edinburgh: Churchill Livingstone; 1999. p. 1535–48 [chapter 68]. Magni G, Marchetti M, Moreschi C, Merskey H, Luchini SR. Chronic musculoskeletal pain and depressive symptoms in the national health and nutrition examination. 1. epidemiologic follow-up study. Pain 1993;53:163–8. Main CJ, Watson PJ. Psychological aspects of pain. Manual Therapy 1999;4:203–15. Maitland GD. Vertebral manipulation, 5th ed. Oxford: Butterworth Heinemann; 1986. Maitland GD. Peripheral manipulation. 3rd ed. Oxford: Butterworth Heinemann; 1991. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971–9. Melzack R, Wall PD. The challenge of pain. 3rd ed. London: Penguin Books; 1991. Miles MB, Huberman AM. Qualitative data analysis. An expanded source book. 2nd ed. Thousand Oaks: Sage Publications; 1994. Moseley GL. Unravelling the barriers to reconceptualisation of the problem in chronic pain: the actual and perceived ability of patients and health professionals to understand the neurophysiology. The Journal of Pain 2003;4:184–9. Overmeer T, Linton SJ, Boersma K. Do physical therapists recognise established risk factors? Swedish physical therapists’ evaluation in comparison to guidelines. Physiotherapy 2004;90:35–41.

49

Pain in Europe. A 2003 Report, www.painineurope.com Patel VL, Arocha JF. Methods in the study of clinical reasoning. In: Higgs J, Jones M, editors. Clinical reasoning in the health professions. 2nd ed. Oxford: Butterworth Heineman; 2000. p. 78–91 [chapter 9]. Payton OD. Clinical reasoning process in physical therapy. Physical Therapy 1985;65:924–8. Picavet SH, Vlaeyen JWS, Schouten JSAG. Pain catastrophising and kinesiophobia: predictors of chronic low back pain. American Journal of Epidemiology 2002;156:1028–34. Price DD. Central neural mechanisms that interrelate sensory and affective dimensions of pain. Molecular Interventions 2002;2:392–402. Rivett DA, Higgs J. Hypothesis generation in the clinical reasoning behaviour of manual therapists. Journal of Physiotherapy Education 1997;11:40–5. Turk DC. Clinical effectiveness and cost-effectiveness of treatment for patients with chronic pain. The Clinical Journal of Pain 2002;18:355–65. Turner JA, Jensen MP, Romano JM. Do beliefs, coping and catastrophizing independently predict functioning in patients with chronic pain? Pain 2000;85:115–25. Verhaak PF, Kerssens JJ, Dekker J, Sorbi MJ, Bensing JM. Prevalence of chronic benign pain disorder among adults: a review of the literature. Pain 1998;77:231–9. Waddell G. Pain and disability. In: Waddell G, editor. The back pain revolution. Edinburgh: Churchill Livingstone; 1998. p. 27–44 [chapter 3]. Waddell G, Main CJ. A new clinical model of low back pain and disability. In: Waddell G, editor. The back pain revolution. Edinburgh: Churchill Livingstone; 1998. p. 223–40 [chapter 14]. Watson G. Neuromusculoskeletal physiotherapy: encouraging selfmanagement. Physiotherapy 1996;82:352–7. Watson P. Psychosocial predictors of outcome from low back pain. In: Gifford L, editor. Topical issues in pain 2. Biopsychosocial assessment and management. Relationships and pain. Falmouth: CNS Press; 2000. p. 85–109 [chapter 2]. Woolf CJ, Decosterd I. Implications of recent advances in the understanding of pain neurophysiology for the assessment of pain in patients. Pain 1999(Suppl 6):S141–7. Wolff MS, Hoskins Michel T, Krebs DE, Watts NT. Chronic painassessment of orthopaedic physical therapists’ knowledge and attitudes. Physical Therapy 1991;71:207–14. Yin RK. Case study research. Design and methods. 2nd ed. Thousand Oaks: Sage Publications; 1994. Zusman M. Forebrain-mediated sensitisation of central pain pathways: non-speciﬁc pain and a new image for manual therapy. Manual Therapy 2002;7:80–8.

ARTICLE IN PRESS

Manual Therapy 12 (2007) 50–55 www.elsevier.com/locate/math

Original article

The clinical examination of neck pain patients: The validity of a group of tests Willem J. De Hertogha,b,c,, Peter H. Vaesa,b, Veerle Vijvermanb, Ann De Cordtb, William Duqueta,b a

Faculty of Physical Education and Physical Therapy, Department of KINE, Vrije Universiteit Brussel, Laarbeeklaan 103, 1090 Brussel, Belgium b Postgraduate Education in Manual Therapy, Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Brussels, Belgium c Department of Health Sciences, Higher Institute of Physiotherapy, Hogeschool Antwerpen, Antwerp, Belgium Received 18 November 2004; received in revised form 30 January 2006; accepted 15 February 2006

Abstract We evaluated whether a blinded observer could identify the neck pain patients in a sample of 42 subjects consisting of neck pain patients and asymptomatic controls. The allocation of subjects to either the control or patient group was based on the scoring of a VAS scale for pain intensity, a Bournemouth Questionnaire (BQ), a manual examination of the rotation of C0-2-7 (rated for Range Of Motion, end feel, onset of pain), an adapted Spurling test and Cervical Range Of Motion (CROM) measurements. The VAS and BQ resulted in a high % of correct allocations (X77.5%) and a high speciﬁcity (90.9%). The Manual Examination Procedures (MEPs) have similar results especially when clustered. The combination of the VAS score, BQ and MEPs resulted in a sensitivity and speciﬁcity of 100% and 86.4%, respectively. Except for the ﬂexion movement all CROM allocation percentages are around 50%, indicating a lesser diagnostic value. Our ﬁndings reinforce the validity of MEPs. Clustering pain measurements, BQ and MEPs provides the highest diagnostic value to identify neck pain patients or necks in need of treatment. r 2006 Elsevier Ltd. All rights reserved. Keywords: Neck pain; Diagnosis; Sensitivity; Speciﬁcity

1. Introduction Neck pain is a common complaint. Patients frequently contact their general practitioner (GP) and are frequently referred to physiotherapists (Borghouts et al., 1999; Korthals-de Bos et al., 2003). The clinical examination of neck pain patients consists of a thorough history taking followed by a functional examination.

Corresponding author. Faculty of Physical Education and Physical Therapy, Department of KINE, Vrije Universiteit Brussel, Laarbeeklaan 103, 1090 Brussel, Belgium. Tel.: +32 2 477 45 29; fax: +32 2 477 44 21. E-mail address: [email protected] (W.J. De Hertogh).

1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.02.007

The history taking should be structured covering the items of the International Classiﬁcation of Functioning that are relevant for physiotherapy. Bolton and Humphreys (2002) recently adapted their original BQ for low back pain patients (Bolton and Breen, 1999) into a disease-speciﬁc questionnaire for neck pain patients. It consists of seven questions addressing pain (impairment level), disability and participation, the mental level (fear, depression, locus of control) and environmental inﬂuence. Covering these dimensions it can be seen as a short-form medical history. Pain is an important symptom in neck pain patients and is measured using the Visual Analogue Scale (VAS). This is a commonly used measuring tool, and is accepted as a reliable and valid instrument (McCormack et al., 1988).

ARTICLE IN PRESS W.J. De Hertogh et al. / Manual Therapy 12 (2007) 50–55

The functional examination can contain manual examination procedures (MEPs) and instrumental measurements. Different types of MEPs are described by various authors (Maitland, 1986; Van der El, 2001). Mainly the reliability of these procedures has been studied, with disappointing results. When scored purely on mobility, the reliability is moderate with kappa values between 0.40 and 0.60 (Fjellner et al., 1999; Strender et al., 1997). Also, Pool et al. (2004) found low kappa values (o0.75) despite the use of very standardized procedures. Reliability tends to increase (kappa 40.78) if the provocation of pain during the mobility test is also scored (Jull et al., 1997). This lack of reliability has led to frustration among practitioners and has made the profession vulnerable to criticism. The validity of MEPs has hardly been studied, mainly as a result of the lack of a gold standard. Humphreys et al. (2004) studied the validity of motion palpation using the presence of a congenital block vertebra as a gold standard. Twenty fourth-year chiropractic students had to identify the hypomobile segments in three subjects with a congenital block vertebra. They found a sensitivity of 74% and a speciﬁcity of 98% for the general detection of all blocks and a kappa value of 0.67 which is considered as good according to Landis and Koch (1977). The Spurling test (passive lateral ﬂexion, homolateral rotation and axial compression at the end) is described as a pain provocation test, and has been shown to be both useful and clinically feasible (Spurling and Scoville, 1944). Kaltenborn (1989) introduced a minor adaptation and added passive extension. This test has been shown to have good sensitivity (77%) and speciﬁcity (92%) and a good positive and negative predictive value (respectively, 80% and 91%) (Sandmark and Nisell, 1995). The Cervical Range of Motion Device (CROM) is used to measure neck mobility and to express the results numerically. It is a highly reliable (inter- and intra-rater reliability, inter-instrument reliability) and valid tool (Youdas et al., 1991, 1992; Hole et al., 1995; Nilsson et al., 1996; Tousignant et al., 2000). It is used in clinical and experimental settings to quantitate test results and to document the progress of patients. The aim of this study was to examine the diagnostic value of pain assessment using a VAS, a short form history using the BQ and of a selection of tests, both manual and instrumental, in neck pain patients. These components will be analysed seperately and in combinations. The identiﬁcation of patients that require a cervical treatment is essential for clinical practise. As a result, we shall look for test sensitivity and speciﬁcity. Sensitivity is the ability of a test to identify a patient; test speciﬁcity is the ability to identify the non-patients.

51

Our research question was: to what extent can a blinded rater identify the neck patients in a sample of volunteers?

2. Materials and methods 2.1. Subjects Subjects were recruited from multiple physio/manual therapy practices. Ten therapists were asked to participate in the study. Their participation consisted in the referral of both patients and asymptomatic controls. All ten therapists were contacted personally and the study setup was explained. Eight therapists were willing to participate. They received additional written information and a ﬁle containing a yellow bookmark with a ﬂow chart of the protocol as well as two checklists including the in- and exclusion criteria. The colour of the bookmark was chosen to focus the attention of the referring therapists (Eastwood, 1940). The ﬁrst checklist was developed to screen the patients, the other for the screening of the asymptomatic controls. All subjects had to be willing to participate in the study and were required to complete an informed consent. Neck patients were included if they experienced pain in the cervical region, with or without irradiation to the head and/or arms. They were excluded if they had a history of neck surgery, took medication because of a neck complaint or experienced dizziness caused by neck or head movements. Subjects for the asymptomatic control group were excluded if they had experienced neck complaints during the past 5 years, irradiating complaints in the shoulder-arm region, headaches, had a history of neck trauma or were in treatment for a spinal disorder. Subjects were generally excluded in case of dizziness provoked by neck or head movements, in case they were mentally disabled, if they had a neuromuscular disorder, rheumatoı¨ d arthritis or Downs syndrome. Of the eight therapists who had agreed to participate, only four actually referred subjects, resulting in 42 subjects entering the study. 2.2. Test protocol Prior to the actual tests, a training session to standardize and familiarize the raters with the procedures was organized. Not using a training session and a standardized protocol inﬂuences the reliability negatively as was experienced, e.g. by van Suijlekom et al. (1999). In this session the entire test procedure was performed on ﬁve subjects in the presence of an external observer (WDH). Feedback was provided and corrections were made when necessary. These ﬁve subjects were not part of the sample and their results were not used for further analysis.

ARTICLE IN PRESS 52

W.J. De Hertogh et al. / Manual Therapy 12 (2007) 50–55

All 42 subjects referred by the therapists were tested by the same two raters (VV and ADC). Both raters are physiotherapists with an additional master in manual therapy. They, respectively, had 5 and 17 years experience as physiotherapists. The test started with the completion of a VAS score on a paper sheet. This VAS-score concerned pain intensity at the moment of testing. Collins et al. (1997) compared the VAS scores of 1080 patients with their scores on a four-point category scale (no pain–mild– moderate–severe). They found that moderate pain intensity starts from 30 mm on the VAS scale and severe pain from 54 mm. These reference values are important both for clinical practise and research to interpret reported VAS scores. Next they were asked to complete the BQ (Bolton and Humphreys, 2002). It consists of seven items which are scored on an 11-point scale. Its internal consistency is good (coefﬁcients: 0.87–0.92). A good internal consistency (Cronbach’s alfa40.70) indicates that all items cover the same dimensions and that the results of the different items can be added to achieve a total score. The test–retest reliability was assessed in patients who reported no change in their neck complaint after treatment. It was found to be moderate for the total score (ICC: 0.65). The construct validity was acceptable with the Neck Disability Index (Pearson’s correlations: 0.50) and with the Copenhagen Neck Functional Disability Scale (Pearson’s correlations: 0.44). The completed forms of both the VAS and BQ were folded inwards and stapled by the subjects themselves to prevent the raters from seeing the completed forms. The next step was the manual and instrumental examination. This included a manual examination of both rotations on the C0-2 and C2-7 levels, the adapted Spurling test on the C1-2–C6-7 levels performed bilaterally and an instrumental mobility examination using the CROM. Both raters were blinded: they were unaware of the subjects status (patient or control). To ensure the blinding, subjects were instructed not to give information to the raters about a possible neck problem.All tests were performed in one single test session, and took about 30 min. One rater performed the manual examination of the rotation and the CROM measurements (VV), the other the adapted Spurling test (ADC). Rater 1 took notes while rater 2 performed his test and vice versa. 2.3. Manual examination procedures and CROM measurements 2.3.1. Manual examination procedures Subjects were seated with backrest, hands resting on their laps. Prior to testing six active warming up movements were made: cervical ﬂexion, extension, both lateral ﬂexions and rotations. Subjects were instructed to

move in one plane, and if necessary they were corrected. Consequently, C7 was marked by rater 1. The position of C7 was checked by rater 2. Next, the manual examination was performed for the rotation of C0-2 and C2-7. First, the head is rotated until the movement arrives at the C2 level. The rotation is then continued from the C2 to the C7 level. This test is scored for mobility (hypermobile, normal, hypomobile) end feel (hard, normal, soft or empty) and possible pain provocation (yes/no). The results were scored on a registration form. Only rotation was tested, and only on these segments. Testing all the movements would have made the entire test procedure too long and too exacting. The bony reference points of C0, C2 and C7 are the most easily recognisable of the entire cervical spine, and allow for easy assessment of rotation. Second, the mobility was measured using the CROM, following the protocol delivered with the CROM (Professional Medical Technologies, Inc., 702, North McColl Road, McCallen, TX 78504, USA). Tests were repeated three times for all tested movements, and the average result was calculated. Third, the adapted Spurling test was performed, starting at the C1-2 level proceeding downwards to C6-7. It was performed bilaterally. The subjects were asked whether the technique caused pain. If so, they were asked to score the pain on a VAS-scale. The rater who performed the test was unable to see the VASscoring. 2.4. Scoring of the tests Based on the test results, the raters allocated subjects to either the patient or the control group. This decision was made following previously determined criteria for each test. A subject was allocated to the patient group if:

the original VAS-score was420 mm.We considered that people, though they may not be suffering from a true neck complaint could also experience a feeling of discomfort in their neck, which might be scored40 on a VAS scale. Therefore a score unequal to 0 does not make them neck patients. We wished to eliminate any feeling of discomfort that scored40 on the VAS, though at the same time include mild pain. A VAS score of 30 mm is described as the borderline between mild and moderate pain intensity (Collins et al., 1997). This was too high for our purpose. Therefore the cut-off point was set at 20 mm. The subject had a score of 414/70 on the BQ. The same rationale was applied to the interpretation of the BQ as to the VAS-score. the manual examination of rotation was positive for at least one movement, which means if two out of three criteria were positive. These positive criteria were hyper/hypomobility and/or a hard or empty end feel

ARTICLE IN PRESS W.J. De Hertogh et al. / Manual Therapy 12 (2007) 50–55

and/or the provocation of pain during the movement. At the C0-2 segments, normal mobility was set at an estimated 451 rotation for subjects younger than 40 years, and an estimated 351 rotation for those older than 40 years. At the C2-7 segments, normal rotation was set at an estimated 25–301 rotation. This was derived from biomechanical studies (Penning and Wilmink, 1987; Mimura et al., 1989) and the fact that rotation decreases with age (Hole et al., 1995). the adapted Spurling test was positive on at least one level (if a pain was provoked with a VAS420 mm). if the average CROM score exceeded the mean71 SD from the results of Hole et al. (1995). This study was chosen because it provides normative ROM data for each decade and gender for each tested movement separately.

The effect of clustering test results on the percentage of correct allocations was investigated using different combinations of tests. Subjects were allocated to the patient group if at least one test of the combination was positive. The referring therapists were asked to register whether the referred subjects were neck patients or asymptomatic controls. This classiﬁcation was set as the gold standard. The classiﬁcation made by the raters was compared with this gold standard. 2.5. Statistical analysis The results were analysed using SPSS 10.0 for Macintosh. Descriptive statistics were calculated for the variables age and CROM. For the manual mobility test and the pain provocation test, data were analysed

53

using 2 2 contingency tables, w2-tests and the calculation of Phi-coefﬁcients (j). This j-value can be considered as a correlation between test result and actual status of the subject. From these contingency tables the sensitivity and speciﬁcity of the tests were calculated. No reference values with regard to sensitivity and speciﬁcity can be provided. They should both be ‘as high as possible’. A lower limit of 75% is used, e.g. by Antonaci et al. (2001).

3. Results Of all 42 participants, two scored the initial VAS scale inadequately, and were therefore excluded from further analysis. The referring therapists indicated 18 neck patients (M/F: 3/15; age: 50.3711.74 yr) and 22 controls (M/F: 10/12; age: 31.68712.18 yr). The median VAS score for the controls was 0 mm (IQR: 0–10 mm). The median VAS for the neck patients was 24 mm (IQR: 14–36 mm). 3.1. Allocation of subjects The result of the allocation of the subjects is presented in the following tables. The way of allocating is mentioned, the percentage of correct allocations, the j-value and its signiﬁcance level. The sensitivity and speciﬁcity of the test are also given (Table 1). The percentage of correct allocations is X75% for all components of the physio/manual therapy assessment. Accumulation of ﬁndings increases the correct allocations as well as test sensitivity and speciﬁcity.

Table 1 Results of the allocation of the subjects based on the VAS, BQ and MEPs % Correct allocations

j

P

VAS BQ VAS and BQ

77.5 87.5 92.5

0.553 0.747 0.850

o 0.0001 o0.0001 o0.0001

Manual Rotation Adapted Spurling MEPs

77.5

0.548

77.5

VAS and MEPs BQ and MEPs VAS, BQ and MEPs

Sensitivity (%)

Speciﬁcity (%)

61.1 83.3 89.5

90.9 90.9 95.2

0.001

77.8

77.3

0.548

0.001

77.8

77.3

82.5

0.649

o0.0001

72.2

90.9

87.5 90.0 92.5

0.750 0.804 0.860

o0.0001 o0.0001 o0.0001

88.9 94.4 100

86.4 86.4 86.4

Results of the allocation of subjects based on the VAS, BQ, Manual Rotation Test, Adapted Spurling, Manual Examination Procedures (MEPS, combination of Manual Rotation and Adapted Spurling) and combinations of the VAS, BQ and MEPs. The percentage of correct allocations, j-value, signiﬁcance level (P), sensitivity and speciﬁcity are displayed.

ARTICLE IN PRESS W.J. De Hertogh et al. / Manual Therapy 12 (2007) 50–55

54

Table 2 Results of the allocation of subjects based on the CROM measurements % Correct allocations Flexion Extension Right lateral ﬂexion Left lateral ﬂexion Right rotation Left rotation Total CROM Total CROM+MEPs

j

70.0 50.0 65.0 52.5 52.5 45.0 45.0 45.0

0.39 0.01 0.29 0.06 0.05 0.12 0.04 0.02

P

Sensitivity (%)

Speciﬁcity (%)

0.014 0.949 0.064 0.775 0.775 0.436 0.832 0.884

55.6 55.5 61.1 50.0 50.0 33.4 88.9 94.4

81.8 45.5 68.2 54.5 54.5 54.5 9.1 4.5

Results of the allocations based on the CROM measurements. All six measured movements are displayed. The percentage of correct allocations, j-value, signiﬁcance level (P), sensitivity and speciﬁctiy are displayed. Total CROM: total score of the CROM measurements. Total CROM+MEPs: cluster of the total CROM and the MEPs.

The allocation based on the CROM measurements results in less differentiating between asymptomatic controls and neck patients (Table 2). Combination of CROM results (total CROM) increased sensitivity only. The number of correct allocations was low. The addition of the MEPs did not increase the percentage of correct allocations.

4. Discussion We evaluated the diagnostic value of pain assessment using a VAS, of a short form history using the BQ and of a selection of tests, both manual and instrumental, in neck pain patients. These components were analysed seperately and in combinations. The presence of a neck complaint was set as the gold standard. The presence and intensity of neck pain at the moment of testing appears to be a highly speciﬁc parameter. Pain is indeed a major symptom of neck pain patients. We used a cut off point of 20 mm to decide whether a subject belonged to the patient group (VAS420 mm) or to the control group (VASo20 mm). Not all controls scored zero on the VAS. A feeling of discomfort can be scored above zero, without representing pain. Given the high speciﬁcity of the VAS, we recommend the use of a cut off point set at 20 mm. Scoring the BQ resulted in a high percentage of correct allocations and a high sensitivity and speciﬁcity (Table 1). It appears to be extremely useful in differentiating between neck patients and asymptomatic controls. The BQ can be seen as a structured short-form medical history. The good results of the BQ here emphasise the importance of the medical history. The percentages of correct allocations based on the Manual Rotation and adapted Spurling are high, as are their sensitivity and speciﬁcity. This indicates that physiotherapists can succesfully identify neck pain patients based on MEPs. Age and gender differed

between the neck pain patients and the control group, which might be a confounding factor as ROM decreases with age. This inequity was accounted for by providing reference values, based on biomechanical analyses of cervical mobility (Penning and Wilmink, 1987; Mimura et al., 1989). Moreover the MEPs were considered positive if 2/3 criteria were positive, and ROM was only one of them. Although the allocation percentages are high, the j values of the Manual Rotation and adapted Spurling remain rather low (Table 1). A larger number of subjects with the same percentages would result in higher j values. Clustering test results increases the diagnostic value. In clinical practise, therapists reach a conclusion after an entire test procedure consisting of more than one diagnostic test. Our study supports this clinical practise. The high speciﬁcity of the MEPs is important: it enables us to exclude subjects (or patients) that do not require treatment. This is a better option than to include subjects (or patients) with no justiﬁcation. The validity of MEPs was recently documented (Humphreys et al., 2004) and is now conﬁrmed by our study. The percentages of correct allocations based on the CROM measurements are low. Except for the ﬂexion movement the allocation percentages are around 50%. Neither clustering the test results nor adding the MEPs had a positive inﬂuence on the situation. The j values were low and the P-values insigniﬁcant. The age and gender inequity between the neck pain patients and the control group is less confounding for the CROM measurements, as Hole et al. (1995) provide reference values for each decade and gender. Our results indicate that the CROM—though a reliable and valid instrument—is less suitable as a diagnostic tool, but mainly can be used as an outcome measure. Our ﬁndings reinforce the validity of MEPs. Clusters of pain measurements, BQ and MEPs provide the highest diagnostic value to identify neck pain patients or necks in need of treatment.

ARTICLE IN PRESS W.J. De Hertogh et al. / Manual Therapy 12 (2007) 50–55

Acknowledgment This research was supported ﬁnancially by the Vrije Universiteit Brussels, R&D Research Grant no. OZR 997. References Antonaci F, Ghirmai S, Bono G, Sandrini G, Nappi G. Cervicogenic headache: evaluation of the original diagnostic criteria. Cephalalgia 2001;21(5):573–83. Bolton J, Breen AC. The Bournemouth Questionnaire: a short-form comprehensive outcome measure.I. Psychometric properties in back pain patients. Journal of Manipulative and Physiological Therapeutics 1999;22(8):503–10. Bolton J, Humphreys K. The Bournemouth Questionnaire: a shortform comprehensive outcome measure. Psychometric properties in neck pain patients. Journal of Manipulative and Physiological Therapeutics 2002;25(3):141–8. Borghouts J, Janssen H, Koes B, Muris J, Metsemakers J, Bouter L. The management of chronic neck pain in general practice, a retrospective study. Scandinavian Journal of Primary Health Care 1999;17:215–20. Collins S, Moore R, McQuay H. The visual analogue pain intensity scale: what is moderate pain in millimeters. Pain 1997;72:95–7. Eastwood RP. Sales Control by Quantitative Methods. New York: Columbia University Press; 1940. Fjellner A, Bexander C, Faleij R, Strender LE. Interexaminer reliability in physical examination of the cervical spine. Journal of Manipulative and Physiological Therapeutics 1999;22(8):511–6. Hole DE, Cook JM, Bolton JE. Reliability and concurrent validity of two instruments for measuring cervical range of motion: effects of age and gender. Manual Therapy 1995;1:36–42. Humphreys BK, Delahaye M, Peterson CK. An investigation into the validity of cervical spine motion palpation using subjects with congenital block vertebrae as a ‘gold standard’. BMC Musculoskeletal Disorders 2004;5(19) http://www.biomedcentral.com/14712474/5/19. Jull G, Zito G, Trott P, Potter H, Shirley D. Inter-examiner reliability to detect painful upper cervical joint dysfunction. Australian Journal of Physiotherapy 1997;43(2):125–9. Kaltenborn FM. Manual Mobilisation of the Spine. Norway: Olaf Norlis Bokhandel; 1989. Korthals-de Bos IB, Hoving JL, van Tulder MW, Rutten-van Molken MP, Ader HJ, de Vet HCW, et al. Cost effectiveness of physiotherapy, manual therapy, and general practitioner care for

55

neck pain: economic evaluation alongside a randomised controlled trial. BMJ 2003;326(7395):911. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. Maitland G. Vertebral Manipulation, ﬁfth ed. Butterworth: Boston, MA; 1986. McCormack HM, Horne DJL, Sheather S. Clinical applications of Visual Analogue Scales: a critical review. Psychological Medicine 1988;18:1007–19. Mimura M, Moriya H, Watanabe T, Takahashi K, Yamagata M, Tamaki T. Three-dimensional motion analysis of the cervical spine with special reference to the axial rotation. Spine 1989;14(11):1135–9. Nilsson N, Christensen HW, Hartvigsen J. The interexaminer reliability of measuring passive cervical range of motion, revisited. The Journal of Manual and Manipulative Therapy 1996;19: 302–5. Penning L, Wilmink J. Rotation of the cervical spine. A CT study in normal subjects. Spine 1987;12(8):732–8. Pool JJ, Hoving JL, de Vet HC, van Mameren H, Bouter LM. The interexaminer reproducibility of physical examination of the cervical spine. Journal of Manipulative and Physiological Therapeutics 2004;27(2):84–90. Sandmark H, Nisell R. Validity of ﬁve common manual neck pain provoking tests. Scandinavian Journal of Rehabilitation Medicine 1995;27:131–6. Spurling RG, Scoville WB. Lateral rupture of the cervical intervertebral disc. Surgery, Gynaecology and Obstetrics 1944;78: 350–8. Strender LE, Lundin M, Nell K. Interexaminer reliability in physical examination of the neck. Journal of Manipulative and Physiological Therapeutics 1997;10(8):516–20. Tousignant M, De Bellefeuille L, O’Donoughue S, Grahovac S. Criterion validity of the Cervical Range Of Motion (CROM) goniometer for cervical ﬂexion and extension. Spine 2000;25(3):324–30. Van der El A. Manuele Diagnostiek Wervelkolom, 5e ed. Rotterdam: Manthel; 2001. van Suijlekom JA, de Vet HC, van den Berg SG, Weber WE. Interobserver reliability of diagnostic criteria for cervicogenic headache. Cephalalgia 1999;19(9):817–23. Youdas JW, Carey JR, Garrett TR. Reliability of measurements of cervical spine range of motion—comparison of three methods. Physical Therapy 1991;71(2):98–106. Youdas JW, Garrett TR, Suman VJ, Bogard CL, Hallman HO, Carey JR. Normal range of motion of the cervical spine : an initial goniometric study. Physical Therapy 1992;72(11): 770–80.

ARTICLE IN PRESS

Manual Therapy 12 (2007) 56–62 www.elsevier.com/locate/math

Original article

Palpation identiﬁcation of spinous processes in the lumbar spine Joanne C. Harlick, Stephan Milosavljevic, Peter D. Milburn School of Physiotherapy, University of Otago, P.O. Box 56, Dunedin, New Zealand Received 16 February 2005; received in revised form 24 January 2006; accepted 15 February 2006

Abstract The purpose of this study was to determine the accuracy of manipulative physiotherapists in palpating radiologically identiﬁed lumbar spinous processes (SPs). Five experienced manipulative physiotherapists were each allocated a cohort of 15 consecutive low back pain (LBP) patients presenting for X-rays and were asked to use surface palpation to identify the L1, L3 and L5 SPs. Spherical radio-opaque markers were taped to the skin over these palpated points and standard lateral radiographs taken. Measurements were made to determine the proximity of these nominated markers to identiﬁed SPs. Seventy-two percent of markers were either placed accurately over the nominated SP or consistently within one SP of the nominated level. Forty-seven percent were accurately placed over the nominated SPs. A greater SP height at L3 and L5, and decreased soft tissue thickness over L5, were associated with an increase in palpation accuracy levels, yet the patient variables of age, sex and body mass index (BMI) had no effect. The strongest effect on accuracy was between-therapist variability. The manipulative physiotherapists used in this study appear to be moderately successful in either palpating a nominated SP or being no more than one spinal level in error. Further research will focus on the choice of palpation procedure and a larger sample. r 2006 Elsevier Ltd. All rights reserved. Keywords: Lumbar spine; Clinical palpation; Validity; Accuracy

1. Introduction Identiﬁcation of lumbar vertebrae by palpation is an important component in the assessment and management of a dysfunctional lumbar spine. However, the literature indicates that clinicians and researchers have difﬁculty in identifying a vertebral segment by surface palpation (Mayer et al., 1997; Downey et al., 1999; Ebraheim et al., 1999). Inaccurate identiﬁcation of key anatomical landmarks is a crucial source of error when measuring spinal movement (Burton, 1986; Stokes et al., 1987; Paquet et al., 1991; Dopf et al., 1994; Gracovetsky et al., 1995; Mayer et al., 1995, 1997; Schuit et al., 1997). Importantly non-invasive spinal range of motion Corresponding author. Tel.: +64 3 479 7193; fax: +64 3 479 8414.

E-mail address: [email protected] (S. Milosavljevic). 1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.02.008

(ROM) and dynamic movement analyses are used as objective measures in the assessment of physical impairment in LBP (American Medical Association, 1990; Burton et al., 1990; Davis and Marras, 2000). It’s clear spinal motion analysis requires valid and reliable identiﬁcation of spinal landmarks to accurately determine available movement. A small number of investigators have examined the inter-rater palpation reliability of physiotherapists in identifying a marked vertebral level by non-invasive means. Binkley et al. (1995) assessed the reliability of six therapists in identifying a marked lumbar vertebral level on 18 symptomatic subjects palpated in prone lying. Although they argue an inter-rater agreement to only within 1.4 segmental levels in the lumbar spine it is difﬁcult to apply such a result to the clinical environment as the clinical situation will allow a number of assessment tools, positions, and procedures for spinal identiﬁcation. McKenzie and Taylor (1997) also measured

ARTICLE IN PRESS J.C. Harlick et al. / Manual Therapy 12 (2007) 56–62

the reliability of 14 physiotherapists who were asked to palpate lumbar spinal levels using their preferred method of palpation on a small sample of ﬁve asymptomatic subjects. They describe an inter-therapist reliability as only fair with a Kappa value of 0.28. Lack of standardization of palpation techniques was identiﬁed as a potential source of inter-therapist error and the use of averaged outlines of spinous processes (SPs) from anatomical specimens as a guideline for SP identiﬁcation is questionable as a reﬂection of the age, ethnicity and sex variances in a such a small population sample. Simmonds and Kumar (1993) investigated the palpation reliability of 20 paired physiotherapists in locating and marking lumbar locations on only two subjects lying in a prone position. They found a mean distance error of 16 mm between therapist’s marks for the L4 SP and suggest that their results are a measure of the palpation accuracy variability in a sample of physiotherapists. Using a similar method to that of McKenzie and Taylor (1997) three pairs of manipulative physiotherapists were used by Downey et al. (1999) to palpate the lumbar spines of 20 symptomatic subjects per pair. Dependent on the pair being evaluated the results show fair-to-high unweighted Kappa values of 0.44–0.88. The authors conclude that an averaged unweighted Kappa value of 0.69 indicates that manipulative physiotherapists demonstrate palpation competence. In summary, these studies have not explored the validity of spinal palpation procedures and the small sample sizes of therapists and subjects makes it difﬁcult to extrapolate the reliability results to either the population of physiotherapists or the subjects/patients who are palpated for spinal identiﬁcation in the clinical environment. In a recent systematic review of spinal palpation tests Sefﬁnger et al. (2004) argue that symptom provocation is a more sensitive indicator for within and between therapist palpation reliability tests and that spinal landmark identiﬁcation by soft tissue and SP palpation is neither reliable nor valid. Although the non-invasive reliability studies suggest that palpation reliability is not high and that identiﬁcation inaccuracy exists, it is surprising there is no research which has tested the accuracy and validity of lumbar spinal palpation against a criterion-based gold standard of radiological identiﬁcation of lumbar vertebrae. The aim of this research is therefore to use standard two-dimensional (2-D) X-ray investigation of radiologically opaque, palpated surface points to determine the lumbar SP palpation accuracy of a sample of manipulative physiotherapists. The use of single plane lateral radiographs has recently been identiﬁed with potential sagittal postural and movement measurement error particularly from out of plane position (Harvey and Hukins, 1998; Marks et al., 2003) and the results of this

57

study may need to be interpreted in the light of such 2-D limitations. Despite these limitations our research will help clarify whether manipulative physiotherapists can accurately identify lumbar intervertebral levels under clinical investigation.

2. Methods 2.1. Research design This is a quasi-experimental, cross-sectional study of experienced manipulative physiotherapists who attempted to identify nominated lumbar SPs using surface palpation on a sample of LBP patients attending a radiology clinic for standard lumbar X-rays. This design allowed the use of an invasive procedure to determine whether manipulative physiotherapists could accurately place radiologically opaque surface markers over radiologically identiﬁed SPs. The determination of accuracy was categorical (yes or no) and inaccuracy was quantiﬁed by distance. Patient variables were identiﬁed that allowed the use of the odds ratio and conﬁdence interval in a logistic regression analysis to determine whether any of these variables were associated with accuracy. 2.2. Subjects Following ethical approval 75 consenting patients presenting with a standard non-urgent request for lumbar spine X-ray were recruited over a 3-month period and distributed sequentially among ﬁve manipulative physiotherapists. This process gave each physiotherapist a separate cohort of 15 subjects for palpation identiﬁcation. Although there were more females (n ¼ 46) than males (n ¼ 29) in the subject group, the between therapist distribution of patients demonstrated no statistically signiﬁcant differences (p4:05) for age, sex, or body mass index (BMI). The physiotherapists had postgraduate manipulative physiotherapy qualiﬁcations attained in either Australia or New Zealand and were experienced in postgraduate manipulative physiotherapy education at a tertiary level. Although all were experienced clinicians only two still maintained their own clinical caseload. They also had a minimum of 10 years’ experience since gaining their Australian or New Zealand undergraduate qualiﬁcation in physiotherapy. Only non-urgent standard X-ray referrals were accepted for inclusion. Subjects with previous spinal surgery had the potential to confound palpation skills due to possible alteration in tissue structure and were excluded during the clinical screening process. Other exclusion criteria were a severe level of pain and previous contact with the physiotherapist. The minimal

ARTICLE IN PRESS J.C. Harlick et al. / Manual Therapy 12 (2007) 56–62

58

acceptable age was 15 years as skeletal and developmental immaturity below this age may be a confounding issue for structural identiﬁcation. 2.3. Instrumentation The radiographic equipment was a SiemensTM Tridoros Tomography unit using Lanex regular cassette with Kurix ﬁlm utilizing KodakTM processing techniques. The distance from the X-ray tube to the ﬁlm tray was 110 cm and standardized for each radiograph. 2.4. Procedure Each subject completed a clinical history questionnaire and measurements were taken of height and weight. The physiotherapist was then given the completed clinical history to identify the subject’s symptom duration, severity of pain and sites of symptoms. This was done to enhance the simulation of the clinical environment and was aimed at assisting each physiotherapist in their choice of palpation procedures. As the height of the X-ray examination table was ﬁxed at 76.5 cm a step-stool was made available for the physiotherapists if required. This allowed ﬁnal identiﬁcation and placement of radio-opaque markers by the physiotherapist without the subject having to move any further than was necessary and helped minimize skin movement of markers prior to X-ray. Each physiotherapist was free to use their preferred clinical assessment techniques for palpating and identifying the L1, L3 and L5 SPs. The choice of techniques was recorded for each subject. The assessment techniques included palpation of the iliac crests, the 12th rib, the posterior superior iliac spines (PSIS), passive accessory intervertebral movement (PAIVM), and passive physiological intervertebral movement (PPIVM). Following identiﬁcation and surface marking of the landmark, each subject was positioned in side lying in the protocol position for a lateral spinal X-ray (Fig. 1). At this point the physiotherapist re-evaluated the

Fig. 1. Final patient position immediately prior to lateral X-ray.

identiﬁed spinal levels in case of potential skin movement associated with ﬁnal subject positioning. Spherical (7.5 mm) lead radio-opaque markers (buckshot) were placed on the centre of the nominated SP skin marks and were adhered with non-allergenic radiotransparent tape (Fig. 1). This marker size was chosen for both ease of view and delineation of the marker as well as the marker to skin surface interface. A further three spherical radio-opaque markers were positioned caudal to the nominated L1, L3 and L5 markers, respectively (Fig. 1) to allow for geometric analysis. These markers were positioned at the estimated midpoint between the L1 and L3, and L3 and L5 markers, as well as over a point caudal to L5 estimated to be similar to the midpoint distance between L3 and L5. The placement of these additional markers allowed for geometric skin curvature construction between each of the nominated (L1, L3 and L5) skin markers. The linear distance (mm) between the markers placed at the estimated midpoints of the L1–L3 and the L3–L5 markers (Fig. 2[II]) was measured with the use of a hand-held metric caliper. This reference distance calculated was used as a scaling factor for each individual to enable true distance measurements. Two lead sheets were then placed behind the subject to establish a deﬁnite line on the ﬁlm to minimize scattering of X-rays. Four rice ﬂour bags were taped over the lumbar lordosis to create tissue equivalent density surrounding the marker. The lateral X-ray was taken with an aluminium ﬁlter to assist in identifying the lead markers on the ﬁlm. Prior to marker removal each X-ray was checked for the appropriate clarity required for a normal radiological report. The lateral radiographs were viewed and outlines of vertebrae and radio-opaque markers transferred onto transparency for analysis. The lines joining the six markers were connected through the identiﬁed midpoint of each marker to obtain an indication of lumbar sagittal skin curvature (lordosis) in side lying. The angle (a) between the drawn lines that intersected at the positioned SP marker was measured manually using a protractor at the L1, L3 and L5 levels. The distance from the midpoint of the L1, L3 and L5 markers to the midpoint of the reference markers immediately above (a) and below (b), were also measured and are illustrated in Fig. 2[III]. Using the intersection of distances a and b and angle a, basic trigonometry was applied to calculate and construct an angle (y) that allowed determination of the radius passing through the marker as well as the tangent of skin curvature. This process was used to create a direction of clinical palpation that was assumed as being perpendicular to the skin surface. All measurements were taken from this perpendicular palpation point and the process is graphically represented in Fig. 2[IV]. The palpable boundaries of the correct SP were then

ARTICLE IN PRESS J.C. Harlick et al. / Manual Therapy 12 (2007) 56–62

59

Fig. 2. X-ray and measurement example (SPs and skin surface digitally enhanced): Key: a ¼ measured line; b ¼ measured line; a ¼ measured angle alpha; A ¼ measured distance from superior margin of SP to inferior margin of nominated marker; B ¼ measured height of SP; C ¼ measured subcutaneous soft tissue depth; D ¼ calculated radius of curvature; E ¼ tangent to curvature; y ¼ calculated angle.

determined and the height of the SP was measured between upper and lower perpendicular lines to the tangent of curvature (Fig. 2[IV]). As skin marker movement may have occurred during set up for lateral X-rays it was decided to accept any nominated marker as being accurately placed where geometric overlap was seen to occur between the superior or inferior boundaries of the marker and the inferior and superior boundaries of the SPs, respectively. This criterion was thus slightly less stringent than a measurement from the centre of a marker to the centre of a SP. As the literature revealed that this measurement procedure for palpation accuracy had not been used previously a two-way mixed model of intraclass correlation coefﬁcient (ICC) was applied to determine reliability of measurement (Shrout and Fleiss, 1979). Two independent examiners made separate measurements from X-rays of the same subjects in a randomly selected subgroup of 20 patients. The primary measurements of angle a and distance a were chosen for this comparative analysis. The ICC scores for a and distance a were 0.96 (CI 0.89–0.98) and 0.98 (CI 0.94–0.99), respectively, showing a high level of inter-tester reliability. Distances were calculated for the following variables:

The vertical distance (mm) from the top of the marker to the inferior border of the nominated SP if the marker was placed inferior to the SP, or

The vertical distance (mm) from the bottom of the marker to the superior border of the nominated SP if the marker was placed superior to the SP, or If any part of the marker overlapped with the SP boundaries a zero displacement was recorded. The vertical distance (mm) between the superior and inferior borders of the nominated SP (SP height). The horizontal distance (mm) between the midpoint of the anterior border of the marker and the midpoint of the posterior surface of the SP (subcutaneous soft tissue thickness).

All distances and angular measurements were made manually with a metric calliper and a precision protractor. The vertical distance measurement was made parallel to the tangent of curvature constructed about each nominated marker. The dependent variables of vertical distance of opaque radiological marker from nominated SP was then categorized for accuracy (acceptable and non-acceptable) and quantiﬁed for distance (mm) for the markers placed in a nonacceptable position. 2.5. Statistical analysis Statistical analyses were undertaken with SPSSTM (Version 11.0.0). The categorical accuracy data and the patient and physiotherapist variables are presented descriptively (Tables 1 and 3) and used in separate

ARTICLE IN PRESS J.C. Harlick et al. / Manual Therapy 12 (2007) 56–62

60

logistic regression calculations (Table 2) to determine their effect on accuracy at the L1, L3 and L5 spinal levels. The logistic regression results for the effect of these variables on accuracy are expressed as the odds ratio with its 95% conﬁdence interval.

3. Results As a group, the physiotherapists had a mean palpation accuracy of 47% (range 24–69%) across the three lumbar levels (Table 1) with Therapist 1 showing the highest level of accuracy (69%) and Therapist 5 the least (24%). Therapists 2–4 had similar levels of accuracy (44–48%). Therapists 1 and 4 most accurately identiﬁed (450%) the L1 SP while the L5 SP was most accurately identiﬁed (450%) by Therapists 1–3 (Table 1). The L3 SP was most accurately identiﬁed (450%) by Therapist 1.

This substantial variation in lumbar palpation accuracy among the ﬁve physiotherapists was statistically signiﬁcant at both the L3 (OR ¼ 1.56, CI ¼ 1.10–2.23, P ¼ 0:01) and L5 (OR ¼ 1.74, CI ¼ 1.21–2.51, Po0:01) levels (Table 2). Subject age, sex and BMI were also included in the logistic regression analyses for SP identiﬁcation yet did not demonstrate any statistically signiﬁcant associations (P40:05) with palpation accuracy (Table 2). The subcutaneous tissue thickness measurements (horizontal dimension) were greatest at L5 with a mean distance of 31.6 mm, L3 at 22.1 mm and L1 at 18.5 mm, yet a lesser soft tissue thickness only demonstrated a weak effect at L5 (P ¼ 0:05). Spinous process height (vertical dimension) measurements were greatest at L1 with a mean height of 20.5 mm, L3 at 20.2 mm and then L5 at 14.1 mm. A predictive effect for height of SP on palpation accuracy was demonstrated weakly at L5 (OR ¼ 0.87, CI ¼ 0.76–0.99, P ¼ 0:04) and more strongly at L3 (OR ¼ 0.71, CI ¼ 0.59–0.86, Po0:01).

Table 1 Percentage lumbar palpation accuracy per physiotherapist PT 1

PT 2

PT 3

PT 4

PT 5

Total

L1 % Acceptable % Non-acceptable

67 33

47 53

40 60

60 40

33 67

49 51

L3 % Acceptable % Non-acceptable

67 33

47 53

33 67

40 60

20 80

41 59

L5 % Acceptable % Non-acceptable

73 27

53 47

60 40

40 60

20 80

49 51

Combined % Acceptable % Non-acceptable

69 31

48 52

44 56

47 53

24 76

47 53

PT ¼ physiotherapist.

Table 2 Inﬂuence of subject and physiotherapist variables on categorical palpation accuracy (acceptable and non-acceptable) Variables

a

Age Sexa BMIa STTa SPHa Therapist

L1

L3

L5

Odds ratio

95% CI

P

Odds ratio

95% CI

P

Odds ratio

95% CI

P

1.01 0.68 1.05 1.03 0.93 1.25

0.99–1.04 0.27–1.74 0.95–1.16 0.98–1.07 0.83–1.05 0.90–1.73

0.39 0.42 0.37 0.22 0.24 0.19

1.01 0.50 1.09 1.04 0.71 1.56

0.99–1.04 0.19–1.29 0.98–1.21 0.99–1.09 0.59–0.86 1.10–2.23

0.31 0.15 0.13 0.14 o 0.01 0.01

1.00 0.59 1.03 1.04 0.87 1.74

0.97–1.02 0.23–1.52 0.93–1.14 1.00–1.07 0.76–0.99 1.21–2.51

0.82 0.28 0.61 0.05 0.04 o 0.01

STT ¼ subcutaneous tissue thickness (mm). SPH ¼ spinous process height (mm). a Subject variable.

ARTICLE IN PRESS J.C. Harlick et al. / Manual Therapy 12 (2007) 56–62

61

Table 3 Inaccurate marker placement by number and distance (mm) at L1, L3 and L5 Inaccurate

N Mean distance SD

Therapists 1–5

Therapists 1–4

L1

L3

L5

All

L1

L3

L5

All

38 19.7 21.4

44 18.5 18.2

38 20.4 15.7

120 19.3 18.6

28 15.1 10.5

32 17.0 14.2

26 17.8 14.4

86 16.6 12.9

This effect was such that a greater SP height was associated with an increase in palpation accuracy. Although a number of clinical procedures were used by the physiotherapists for locating SPs, including palpation of the iliac crest (87%), PPIVM (75%), PAIVM (53%), PSIS (37%) and 12th rib (24%) of subjects, further regression analysis revealed no evidence for either a given clinical procedure or a mix of procedures having any predictive inﬂuence on palpation accuracy (P40:05). The distance of the incorrect markers placed either above or below the identifying level was also descriptively analysed (Table 3). The mean inaccuracy was 19.7 mm (SD 21.4 mm) for L1; 18.5 mm (SD 18.2 mm) for L3; 20.4 mm (SD 15.7 mm) for L5; and for all levels was 19.3 mm (SD 18.6 mm). As Therapist 5 also showed the greatest amount of incorrectly placed markers by a considerable margin (Table 1) the analysis was also repeated with this therapist’s data removed (Table 3). Therapists 1–4 then demonstrated a mean inaccuracy of 15.1 mm (SD 10.5 mm) for L1; 17.0 mm (SD 14.2 mm) for L3; 17.8 mm (SD 14.4 mm) for L5; and a mean inaccuracy for all levels of 16.6 mm (SD 12.9 mm).

4. Discussion Although the mean accuracy of lumbar SP identiﬁcation was 47%, further interpretation of the data shows a mean distance error of 19.3 mm for incorrectly placed markers (Table 3), and a mean SP height of 18.3 mm across all levels. A post hoc analysis used this mean SP height as an upper limit for acceptance and demonstrated that 72% of all nominated markers (162 out of a possible 225) were either placed over the nominated SP or within one SP in height. Four of the manipulative physiotherapists (Therapists 1–4) also demonstrated a higher mean level of accuracy (52%) as well as closer mean distance measurements for inaccurately placed markers (16.6 mm). Using the same mean SP height as the upper limit for acceptance Therapists 1–4 placed 88% of markers (158 out of a possible 180) either accurately over the nominated SP or within one SP of the nominated level.

For the physiotherapists in the current study, palpation and identiﬁcation of the PSIS (37%) and the 12th rib (24%) appear to be procedures used least often yet these structures are landmarks directly associated with identiﬁcation of end segments of the lumbar spine. As SP height had a signiﬁcant effect on accuracy at both L5 and L3, and soft tissue thickness at L5 had a weak effect, it would seem logical to utilize multiple landmark identiﬁcation procedures to help minimize identiﬁcation error. However, no statistically signiﬁcant associations between use of palpation procedures and accuracy were observed and it appears that between therapist variability had the greatest effect on accuracy. These levels of palpation accuracy may be argued as clinically acceptable for non-invasive therapeutic intervention as it is obvious any given vertebrae articulates with vertebrae above and below and clinical intervention to one lumbar level in error may still have some inﬂuence. Furthermore the manipulative physiotherapists in this study were not asked to clinically intervene and were not asked to nominate or identify symptomatic spinal levels. Recent evidence suggests reproduction of symptoms is a strong feature for reliability of spinal palpation with a weaker effect for landmark identiﬁcation (Sefﬁnger et al., 2004). Although the present results suggest some success in palpation identiﬁcation of lumbar SPs it remains to be seen whether lumbar palpation accuracy has any effect on the outcome of clinical intervention particularly where the lumbar spine is often treated as an entity and not as separate vertebrae. It is difﬁcult to compare results as previous research has not used X-rays and has concentrated on the reliability of skin surface point identiﬁcation. To our knowledge this study is the ﬁrst attempt to determine the validity of palpation skills utilizing X-ray analysis. However the results of this study are consistent with the results and palpation accuracy arguments described by McKenzie and Taylor (1997) as well as Downey et al. (1999). Downey et al. (1999) also demonstrated that identiﬁcation of the L3 SP had the highest error with the mean distance of 20 mm. Conversely, McKenzie and Taylor (1997) had the poorest identiﬁcation for the L1 SP with mean error distance of 36 mm.

ARTICLE IN PRESS 62

J.C. Harlick et al. / Manual Therapy 12 (2007) 56–62

However some caution should be used in the interpretation of the present results. Although sample size was small the use of manipulative physiotherapists experienced in clinical and skills-based teaching at a tertiary institution was an attempt to highlight the external validity of this sample. These participants should be capable of a high level of palpation accuracy. Potential skin movement was minimized by allowing each physiotherapist to re-evaluate marker placement on spinal levels prior to X-ray positioning. However, the procedural issues of ﬁxed height of X-ray examination table, ﬁnal side-lying position, use of rice ﬂour bags, measurement from single plane radiographs (Harvey and Hukins, 1998), and the weak assumptions of standardized lordotic curvature calculation in side lying (Marks et al., 2003) had the potential to inﬂuence ﬁnal marker position and ultimately the determination and measurement of marker proximity to the SP. The error effect of all these factors on palpation accuracy is unknown however their potential effect was minimized by accepting as accurate any partial overlap between identiﬁed lead marker and nominated SP.

It appears that between therapist variability has a greater effect on accuracy than any patient deﬁned factor. Unfortunately this sample of physiotherapists is too small to consider any multivariate analysis using therapist-deﬁned factors such as professional qualiﬁcations, clinical experience, or choice of palpation procedure as co-variates that might help to explain such variability between physiotherapists. The need for accurate identiﬁcation of lumbar SPs and by association lumbar vertebral levels is not limited to the physiotherapy profession. Future research may include other health professionals who require surface palpation of spinal surface landmarks for diagnostic and/or intervention purposes. However, palpation is an integral component of the examination and therapeutic skills of manipulative physiotherapists. Although these results indicate a moderate ability to identify spinal structure, further research into clinical palpation skills may help identify the evaluation procedures necessary to improve palpation identiﬁcation.

Binkley J, Stratford PW, Gill C. Inter-rater reliability of lumbar accessory motion mobility testing. Physical Therapy 1995;75(9):786–92. Burton AK. Regional lumbar sagittal mobility; measurement by ﬂexicurves. Clinical Biomechanics 1986;1(1):20–6. Burton AK, Edwards V, Skyes DA. Invisible skin marking for testing palpatory reliability. Journal of Manual Medicine 1990;5:27–9. Davis KG, Marras WS. The effects of motion on trunk biomechanics. Clinical Biomechanics 2000;15(10):703–17. Dopf CA, Mandel SS, Geiger DF, Mayer PJ. Analysis of spine motion variability using a computerized goniometer compared to physical examination. A prospective clinical study. Spine 1994;19(5):586–95. Downey BJ, Taylor N, Niere KR. Manipulative physiotherapists can reliably palpate nominated lumbar spinal levels. Manual Therapy 1999;4(3):151–6. Ebraheim NA, Inzerillo C, Xu R. Are anatomic landmarks reliable in determination of fusion level in posterolateral lumbar fusion? Spine 1999;4(10):973–4. Gracovetsky S, Newman N, Pawlowsky M, Lanzo V, Davey B, Robinson L. A database for estimating normal spinal motion derived from noninvasive measurements. Spine 1995;20(9):1036–46. Harvey SB, Hukins DWL. Measurement of lumbar spinal ﬂexion–extension kinematics from lateral radiographs: simulation of the effects of out-of-plane movement and errors in reference point placement. Medical Engineering and Physics 1998;20(6):403–9. Marks MC, Stanford CF, Mahar AT, Newton PO. Standing lateral radiographic positioning does not represent customary standing balance. Spine 2003;28(11):1176–82. Mayer RS, Chen I-H, Lavender SA, Traﬁmow JH, Andersson GBJ, Shaffer WO. Variance in the measurement of sagittal lumbar spine range of motion among examiners, subjects, and instruments. Spine 1995;20(13):1489–93. Mayer TG, Kondraske G, Beals SB, Gatchel RJ. Spinal range of motion. Accuracy and sources of error with inclinometric measurement. Spine 1997;22(17):1976–84. McKenzie AM, Taylor NF. Can physiotherapists locate lumbar spinal levels by palpation? Physiotherapy 1997;83(5):235–9. Paquet N, Malouin F, Richards CL, Dionne JP, Comeau F. Validity and reliability of a new electrogoniometer for the measurement of sagittal dorsolumbar movements. Spine 1991;16(5):516–9. Schuit D, Petersen C, Johnson R, Levine P, Knecht H, Goldberg D. Validity and reliability of measures obtained from the OSI CA6000 spine motion analyzer for lumbar spinal motion. Manual Therapy 1997;2(4):206–15. Sefﬁnger MA, Najm WI, Mishra SI, Adams A, Dickerson VM, Murphy LS, et al. Reliability of spinal palpation for diagnosis of back and neck pain. Spine 2004;29(19):E413–23. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychological Bulletin 1979;86(2):420–8. Simmonds MJ, Kumar S. Part II: location of body structures by palpation—a reliability study. International Journal of Industrial Ergonomics 1993;11:145–52. Stokes IA, Bevins TM, Lunn RA. Back surface curvature and measurement of lumbar spinal motion. Spine 1987;12(4):355–61.

References

Further reading

American Medical Association. Guides to the evaluation of permanent impairment. 3rd ed. Chicago: American Medical Association; 1990.

Sicard C, Gagnon M. A geometric model of the lumbar spine in the sagittal plane. Spine 1993;18(5):646–58.

5. Conclusion

ARTICLE IN PRESS

Manual Therapy 12 (2007) 63–71 www.elsevier.com/locate/math

Original article

Effectiveness of speciﬁc soft tissue mobilizations for the management of Achilles tendinosis: Single case study—Experimental design Rebecca E. Christenson,1 Pure Sports Medicine, David Lloyd Club, Point West, 116 Cromwell Road, Kensington, London SW7 4XR, UK Received 6 January 2005; received in revised form 22 January 2006; accepted 15 February 2006

Abstract A single case study ABA design was used to evaluate the effectiveness of a treatment protocol of accessory and combined speciﬁc soft tissue mobilizations (SSTMs) in a 39-year-old female with a 5-year history of Achilles tendinosis. The study involved three phases each lasting 6 weeks: a pre-treatment, treatment and post-treatment phase. There was also a 3-month follow-up after the post-treatment phase. The VISA A questionnaire was used as a severity index for Achilles tendinosis to quantify symptoms and dysfunction. A visual analogue scale (VAS) was used to record pain. Dorsiﬂexion range was measured using a goniometer with the knee extended to reﬂect gastrocnemius length and a weight-bearing dorsiﬂexion lunge test was used as a measure for soleus length. Following treatment, considerable improvements were recorded in all outcomes. The subject returned to a full gym programme and recorded no pain on the VAS as well as scoring 100% on the index of severity for Achilles tendinosis during the post-treatment phase and follow-up. The single case study design limits generalization but the results support the use of SSTMs in the treatment of Achilles tendinosis and suggests that further research into this intervention is warranted. r 2006 Elsevier Ltd. All rights reserved. Keywords: Achilles tendinosis; Speciﬁc soft tissue mobilizations; Single case study

1. Introduction Achilles tendinosis involves pathological changes to the tendon, characterized by an absence of inﬂammatory cells (A˚stro¨m and Rausing, 1995; Khan et al., 1999). Achilles tendon disorders are common, particularly in sporting individuals but can occur in those with a sedentary lifestyle (Alfredson and Lorentzon, 2000). The history often involves overuse either by a sudden increase in training or sustained high-level training (Kannus, 1997; Almekinders and Temple, 1998). Symptoms are aggravated by activities that load the Achilles tendon, commonly running (McCrory Tel.: +870 2000 878.

E-mail address: [email protected]. Research carried out at: Hammersmith Hospital, DuCane Road, London W12 0HS, UK. 1

1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.02.012

et al., 1999; Humble and Nugent, 2001; Kader et al., 2002; Paavola et al., 2002). These patients often have pain on and off for many years. Rest can ease the symptoms but is normally insufﬁcient to allow return to the aggravating activity (Kvist, 1994; Maffulli and Kader, 2002). The pathogenesis is not entirely understood but mechanical load placed on the tendon, whether excessive or moderate, seems to result in microtrauma. The tendon then responds poorly with slow or incomplete healing and lacks extracellular organization (Cook et al., 2002). Normal tendon ﬁbres are composed of predominantly type I collagen (Hayem, 2001). Histology of pathologic Achilles tendons have shown a chronic increase in type III collagen as well as disordered ﬁbre arrangement (Ja¨rvinen et al., 1997; Khan et al., 1999; Maffulli et al., 2000; Eriksen et al., 2002). This thinner and less durable collagen compromises the function of the tendon and its ability to withstand load. Adding to

ARTICLE IN PRESS R.E. Christenson / Manual Therapy 12 (2007) 63–71

64

this picture of mechanical disadvantage are the observations of increased mucoid ground substance, abnormal appearance and activity of tenocytes and neovascularization (A˚stro¨m and Rausing, 1995; Movin et al., 1997; Khan and Maffulli, 1998; Cook et al., 2001). Conservative management remains a challenge because of the deﬁcient tendon repair and incomplete understanding of the pathology. Despite this, there does seem to be reasonable evidence for heavy-load eccentric calf muscle training in the management of Achilles tendinosis (Alfredson et al., 1998). A rational for this is that exercise may be able to affect and improve extracellular organization (Kannus et al., 1997; Wren et al., 2000). Convincing evidence behind manual therapy treatment of acute soft tissue injuries or tendinopathies is scant. However, empirically, the use of mobilizations to produce a mechanical effect has been advocated. Hunter (1994) developed speciﬁc soft tissue mobilizations (SSTMs) which tension the soft tissue via accessory soft tissue movement, physiological joint motion or a combination of both techniques. Hunter (1998) proposed a mechanical model to explain the potential effects of SSTM, suggesting that they may increase collagen synthesis and improve collagen alignment. There is evidence that application of a speciﬁc force to healing tissue can increase ﬁbroblast activation (Davidson et al., 1997; Gehlsen et al., 1999) and improve biomechanical, biochemical and morphological properties of healing tissue (Gomez et al., 1991). However, as there is no histological evidence of inﬂammation in tendinosis, this explanation is deﬁcient. It is possible that a purely mechanical explanation of the effects is too simplistic and a neurophysiological explanation is a plausible alternative. 1.1. Aim of the study The aim of this study was to examine the effectiveness of accessory and combined SSTM in the management of a subject diagnosed with Achilles tendinosis. The use of an accessory SSTM in conjunction with a stretch, isometric contraction, active or resisted movement is termed a combined SSTM (Hunter, 1998, 2000). Speciﬁcally, the objectives were to determine whether:

Pain, function and muscle length of gastrocnemius and soleus changed in response to a treatment protocol of accessory and then combined SSTM, as the subject progressed. Any changes in signs and symptoms were maintained once the programme stopped up to the 3-month follow-up.

2. Method 2.1. Research design A single case study ABA experimental design was chosen as no clinical trials have investigated the effects of SSTM. It is a useful starting point to establish the need for further research. The complete management of Achilles tendinosis would involve the identiﬁcation of external and internal causative factors as well as treatment directed towards improving strength and ﬂexibility of the Achilles tendon. However, the single case study allows evaluation of one component of the management in isolation. The Hammersmith and Queen Charlottes’ & Chelsea Hospitals Research Ethics Committee granted ethical approval. 2.2. Subject The main inclusion criterion for the study was Achilles tendon pain for more than 3 months (Alfredson et al., 1998; Silbernagel et al., 2001). The diagnosis was made from a complete history and objective examination. Presence of morning stiffness, pain on loading the tendon during a heel raise, pain on palpation (Galloway et al., 1992; Kvist, 1994), decreased length of soleus or gastrocnemius (Ja¨rvinen et al., 2001) and restricted accessory glide of the tendon (Paavola et al., 2002) were focal points of the diagnosis. A slump and SLR were used to screen for any obvious adverse neurodynamics. Exclusion criteria were: associated LBP, previous foot injury or surgery, evidence or diagnosis of an inﬂammatory arthropathy, diabetes or any other medical diagnosis that was thought to interfere with the study. A 39-year-old female met the criteria and consented to enter the study. She had a 5-year history of bilateral Achilles pain with no obvious cause. The side with the most severe pain varied but the left had been worse for a year. She had played regular club hockey for 13 years as a goalkeeper, training and playing a game each week. She described soreness over the Achilles tendon for a day after playing hockey and following her walk to work and had used analgesia before games for the last 6 months. She also went to the gym three times a week and had altered her programme to avoid aggravating the pain. Pain occasionally disturbed her sleep. Objectively she had tightness of the left gastrocnemius and soleus. The most comparable stretch biased soleus with weight-bearing dorsiﬂexion, knee ﬂexion and lateral rotation of the hip. A medial glide of the Achilles tendon was restricted, reproducing the most severe symptoms. Accessory joint movement at the talocrural, subtalar and midtarsal joint was similar on both sides

ARTICLE IN PRESS R.E. Christenson / Manual Therapy 12 (2007) 63–71

suggesting that the limitation to dorsiﬂexion was from triceps surae.

65

et al., 2002) as triceps surae is a limiting factor to dorsiﬂexion (Palastanga et al., 1993).

2.3. Procedure 2.5. Phase B: treatment The study was divided into three phases and a 3month follow–up. 2.4. Phase A: pre-treatment assessment Baseline measurements were taken of all the dependent variables once a week at the same time each week for 6 weeks. No treatment was given. 2.4.1. Function The VISA A is a valid and reliable index of severity of Achilles tendinopathy (Robinson et al., 2001). It was a useful guide to represent change and a desirable outcome for the subject as it included many of the subjective markers that arose in the patient’s history. A rest period was given to allow for resolution of pain between questions where symptoms were reproduced. Use of equipment and instructions given were standardized when answering questions. 2.4.2. Pain The subject was asked to mark an on a 10 cm line where 0 was no pain and 10 was the worst imaginable pain (Carlsson, 1983; Price et al., 1983). To ensure consistency, the subject was always asked to record her average pain intensity over the 24 h preceding the assessment. 2.4.3. Muscle length For soleus: Dorsiﬂexion of the foot in weight-bearing was measured by the distance of the great toe from a wall as described by Bennell et al. (1998). The distance was recorded using a tape measure, which was placed face down to minimize observer bias. The mean of three measurements was taken. For gastrocnemius: A goniometer was used to measure the range of dorsiﬂexion with knee extension. Goniometry has been found to have fair intratester reliability when used to measure ankle range (Elveru et al., 1988). The goniometer was blacked out with tape so that the assessor was unaware of the recording as it was taken. The mean of three measurements was recorded as the use of repeated measures further improved reliability (Youdas et al., 1993). These latter two measurements were taken to reﬂect a change in triceps surae length as assessed through ankle dorsiﬂexion. Although this is not a true indication of muscle length, this has been used previously (Singer

Treatment was carried out by the author and involved accessory and combined SSTM based on a treatment protocol outlined by Hunter (2000). Initially a sustained medial accessory SSTM to the left Achilles tendon at its insertion was used with the subject prone and the Achilles tendon in neutral. This was the most comparable direction based on quality, range and reproduction of symptoms. The force was applied perpendicular to the line of the Achilles tendon using the thumbs. The force of the application was governed by the point of marked tissue resistance and onset of mildto-moderate discomfort. The ﬁrst combined SSTM administered was a medial glide with the Achilles tendon on stretch to R1. This was progressed to an accessory glide with an isometric hold of triceps surae. Finally, the medial glide was applied during dynamic through range loading against theraband with the subject prone. Ethical approval was granted on the basis that the treatment would be altered if the patient was not improving. Objective measures, which were separate from the outcome measures, were therefore taken to guide treatment. The most comparable stretch on the Achilles tendon was used, along with number of calf raises and subsequently number of hops, as the objective markers to determine the response to treatment. The objective markers were only altered as the subject improved. As the response to the SSTM in neutral plateaued the treatment was progressed as described in the treatment table (Table 1). The timing of the progression was related to the subject’s response to the objective markers. The subject was given no home exercises or advice on how to manage her symptoms and was asked to continue activities as usual. She had excluded lower limb exercises that aggravated her symptoms from her gym programme prior to the study. The dependent variables were recorded once a week each week for 6 weeks at the same time as in phase A. When this coincided with treatment, the assessment always preceded the treatment.

2.6. Phase A: post-treatment assessment Measurements were taken of ndent variables once a week at the week for 6 weeks. No treatment the subject was advised to continue usual.

all the depesame time each was given and her activities as

ARTICLE IN PRESS R.E. Christenson / Manual Therapy 12 (2007) 63–71

66 Table 1 Treatment table Week

1–2

3

4

5–6

Treatment given

Accessory SSTM with pt prone and Achilles in neutral

Combined SSTM with Achilles on stretch

Combined SSTM with isometric plantarﬂexion in plantargrade

Repetitions

30 s, 15 reps (break after each set of 5) Two

30 s, 15 reps (break after each set of 5) Two

30 s, 15 reps (break after each set of 5) One

Combined SSTM with through range plantarﬂexion against theraband 3 20 (break after each set of 5). One

No treatments each week

VISA-A Questionnaire Percentage Scores for: pre-treatment (A), treatment (B), post-treatment (A) and three month follow-up 120

Percentage scores

100 80 percentage scores + two standard deviation

60

mean - two standard deviation

40 20 0 A1 A2 A3 A4 A5 A6 B1 B2 B3 B4 B5 B6 A7 A8 A9 A10 A11 A12 3 month Assessment period f/u

Fig. 1. VISA-A questionnaire percentage for: pre-treatment (A), treatment (B), post-treatment (A) and the 3-month follow-up.

2.7. Post-treatment follow-up All the dependent variables were reassessed 3 months after the ﬁnal A phase.

3. Results Line graphs are the traditional way to evaluate single case studies and have been used for analysis of continuous data (Ottenbacher, 1986). The interpretation of the results is through visual inspection and the twostandard deviation band (Nourbakhsh and Ottenbacher, 1994). Although visual interpretation is useful because it is insensitive to weak treatment effects, it is frequently criticized because there are no formal rules to interpret data. The two-standard deviation band was therefore used as an adjunct to the analysis. If two consecutive points fall outside of the two-standard deviation in the

treatment phase, this is recorded as a signiﬁcant change to the assessment phase. The three phases are represented by A1–6, B1–6, A7–13 and the 3-month followup. The VISA A questionnaire has a percentage as the overall score where a high percentage relates to better function and vice versa. In the pre-treatment assessment, the percentages were relatively static but slightly higher towards the end (Fig. 1). There was a sharp increase between B2 and B3 because the subject was able to exercise without pain (Appendix A, question 8). There was a signiﬁcant change between the ﬁrst two phases according to the two-standard deviation band. The score reached a maximum of a 100% by the end of the treatment phase which was maintained until the 3month follow-up. In the visual analogue scale (VAS) there was some variation in the pre-treatment assessment with a considerable drop at A4 (Fig. 2). This did not correlate with changes in the other dependent variables in the

ARTICLE IN PRESS R.E. Christenson / Manual Therapy 12 (2007) 63–71

67

Visual Analogue Scale For Pain (VAS) for: pre-treatment(A), treatment (B), post-treatment (A) and the 3 month follow-up 10 9 8 7 VAS

6

VAS + two standard deviation mean - two standard deviation

5 4 3 2 1

0 A1 A2 A3 A4 A5 A6 B1 B2 B3 B4 B5 B6 A7 A8 A9 A10 A11 A12 3 month Assessment period f/u Fig. 2. Visual analogue scale for pain (VAS) for: pre-treatment (A), treatment (B), post-treatment (A) and the 3-month follow-up.

Soleus Length For: pre-treatment (A), treatment (B), post-treatment (A) and 3 month follow-up

distance from great toe to wall in cm

7 6 5 4 Soleus length + two standard deviation

3

mean - two standard deviation

2 1 0 A1

A2 A3 A4 A5 A6 B1 B2 B3 B4 B5 B6 A7 A8 A9 A10 A11 A12 3 month Assessment period f/u

Fig. 3. Soleus length for: pre-treatment (A), treatment (B), post-treatment (A) and the 3-month follow-up.

same week. There was a slight decrease in pain rating at the end of the pre-treatment phase. However, there was a signiﬁcant change in the two-standard deviation across the pre-treatment A and B phase which was maintained until the end of the study. In Fig. 3 the distance for soleus was consistently small in the pre-treatment A phase, in contrast to a large increase between B2 and B3. The increase in distance

was recorded as signiﬁcant by the two-standard deviation band across the ﬁrst two phases. Visually, the results for gastrocnemius (Fig. 4) correlate with the measurements taken for soleus (Fig. 3). The angles measured for gastrocnemius were quite consistent in the initial A phase with a marked increase between B1 and B2. A maximum of 101 was reached towards the end of the B phase which was maintained in the ﬁnal A phase

ARTICLE IN PRESS R.E. Christenson / Manual Therapy 12 (2007) 63–71

68

Degrees of dorsiflexion with knee extension

Length of gastrocnemius for: pre-treatment (A), treatment (B), post-treatment (A) and 3 month follow-up 105

100

95

90

gastrocnemius length mean - two standard deviation + two standard deviation

85

80 A1 A2 A3 A4 A5 A6 B1 B2 B3 B4 B5 B6 A7 A8 A9 A10 A11 A12 3 month Assessment period f/u

Fig. 4. Length of gastrocnemius for: pre-treatment (A), treatment (B), post-treatment (A) and the 3-month follow-up.

at A7 and A9. This was followed by a slight decrease just before the 3-month follow-up. However, the angle was still far greater than the plus two-standard deviation band.

4. Discussion The results of this study have demonstrated that following a treatment protocol of accessory and combined SSTM this subject, with chronic Achilles tendinosis, had improvements in function, pain and length of gastrocnemius and soleus. These beneﬁcial effects were maintained through the ﬁnal A phase when no treatment was administered and at the 3-month follow-up. The baseline data collected in the pretreatment A phase did not demonstrate an overall trend of improvement, although there was a slight reduction in pain intensity. This suggests that the considerable improvements during the B phase were due to the intervention rather than a natural variation in symptoms. Spontaneous recovery also seemed unlikely because the condition was chronic and worsening. The subject had not had a period of remission for longer than a week prior to the intervention, and in this study maintained her improvements to the 3-month follow-up. Selection of the ABA rather than ABC design allowed the effects to be narrowed to the intervention without any contribution from home exercises. This was particularly important because of the evidence supporting the use of exercises that load the

Achilles tendon (Alfredson et al., 1998; Silbernagel et al., 2001). There was a possible ﬂaw in question 7 in the VISA A, which asked whether sport or physical activity was being undertaken. No single response was entirely appropriate for this subject. A maximum score was awarded for competing at the same or higher level as when symptoms began. As she was always able to play hockey this response could have been selected. However, her gym routine had been modiﬁed as a result of her symptoms which made the second response equally applicable. The author answered the question on the basis of her training to reﬂect change and only awarded her the maximum score when she started competing at a higher level than before the symptoms began. Even without this question there was a dramatic improvement in the VISA A questionnaire during the B and ﬁnal A phase. The VAS varied more than the other dependent variables. The substantial decrease in pain recorded in A4 was difﬁcult to explain. Looking at the VISA A questionnaire, it was possible to see that on the same day, the subject recorded a similar level of pain during her walk to work. However, the pain recorded on questions 4 and 5 (Appendix A) did not correspond in the same way. The highest pain in conjunction with the worst score in the VISA A was recorded the week before in A3. The only change of note between A3 and A4 was the subject’s purchase of a new pair of shoes. A major weakness of this study was the lack of an independent observer. However, efforts were made to

ARTICLE IN PRESS R.E. Christenson / Manual Therapy 12 (2007) 63–71

minimize observer bias as detailed in the methods. Another important limitation of any single case study is its lack of control. The design seeks to minimize the likelihood that results are due to spontaneous recovery by taking recordings over a sufﬁcient time period. A placebo effect is possible because of the subject’s faith in the intervention, possibly augmented by her participation in a research project with connotations that she was receiving the best, most up to date treatment. However, her faith and any placebo effect associated with hands on treatment seem insufﬁcient to explain the improvement maintained throughout the post-treatment phase and at 3-month follow-up.

69

5. Conclusion The results of this study demonstrated that a treatment protocol of accessory and combined SSTM was followed by an improvement in pain, function and length of gastrocnemius and soleus in a subject with Achilles tendinosis. The chronic nature of the subject’s condition and stable baseline measurements suggest that the improvements were not a result of spontaneous recovery. The single case study design limits generalization, but the results suggest that further research in the form of controlled trials is warranted.

Acknowledgement 4.1. Suggestions for further research The results from this study suggest that further research into SSTM for Achilles tendinosis is warranted. The evidence behind heavy-load eccentric exercises suggests that this should be part of the management of patients with Achilles tendinosis. Therefore, a randomized controlled trial looking at SSTM and heavy-load eccentric exercises would be a useful area for future research.

The author would like to thank Glenn Hunter for his comments on a previous version of the text.

Appendix A The VISA A questionnaire reproduced by kind permission of BMJ publishing group, British Journal of Sports Medicine (2001) 35: 335–341.

ARTICLE IN PRESS 70

R.E. Christenson / Manual Therapy 12 (2007) 63–71

References Alfredson H, Lorentzon R. Chronic Achilles tendinosis: recommendations for treatment and prevention. Sports Medicine 2000;29(2):135–46. Alfredson H, Pietila¨ T, Jonsson P, Lorentzon R. Heavy-load eccentric calf muscle training for the treatment of chronic Achilles tendinosis. The American Journal of Sports Medicine 1998;26(3):360–6. Almekinders LC, Temple JD. Etiology, diagnosis, and treatment of tendonitis: an analysis of the literature. Medicine & Science in Sports & Exercise 1998;30(8):1183–90. A˚stro¨m M, Rausing A. Chronic Achilles tendinopathy. A survey of surgical and histopathologic ﬁndings. Clinical Orthopaedics and Related Research 1995;316:151–4. Bennell KL, Talbot RC, Wajswelner H, Techovanich W, Kelly DH, Hall AJ. Intra-rater and inter-rater reliability of a weight-bearing lunge measure of ankle dorsiﬂexion. Australian Journal of Physiotherapy 1998;44(3):175–80. Carlsson AM. Assessment of chronic pain: aspects of the reliability and validitiy of the visual analogue scale. Pain 1983;16(1): 87–101.

Cook JL, Khan MK, Purdam C. Conservative treatment of patellar tendinopathy. Physical Therapy in Sports 2001;2:1–12. Cook JL, Khan MK, Purdam C. Achilles tendinopathy. Manual Therapy 2002;7(3):121–30. Davidson CJ, Ganion LR, Gehlsen GM, Verhoestra B, Roepke JE, Sevier TL. Rat tendon morphologic and functional changes resulting from soft tissue mobilization. Medicine & Science in Sports & Exercise 1997;29(3):313–9. Elveru RA, Rothstein JM, Lamb RL. Goniometric reliability in a clinical setting. Physical Therapy 1988;68(5):672–7. Eriksen HA, Pajala A, Leppilahti J, Risteli J. Increased content of type III collagen at the rupture site of Achilles tendon. Journal of Orthopaedic Research 2002;20(6):1352–7. Galloway MT, Jokl P, Dayton OW. Achilles tendon overuse injuries. Clinical Sports Medicine 1992;11(4):771–82. Gehlsen GM, Ganion LR, Helfst R. Fibroblast responses to variation in soft tissue mobilization pressure. Medicine & Science in Sports & Exercise 1999;31(4):531–5. Gomez MA, Woo SL, Amiel D, Harwood F, Kitabayashi L, Matyas JR. The effects of increased tension on healing medial collateral ligaments. American journal of Sports Medicine 1991;19(4):347–54. Hayem G. Tenology: a new fontier. Joint Bone Spine 2001;68:19–25.

ARTICLE IN PRESS R.E. Christenson / Manual Therapy 12 (2007) 63–71 Humble RN, Nugent LL. Achilles’ tendonitis. An overview and reconditioning model. Clinics in Podiatric Medicine and Surgery 2001;18(2):233–54. Hunter G. Speciﬁc soft tissue mobilisations in the treatment of soft tissue lesions. Physiotherapy 1994;80(1):15–21. Hunter G. Speciﬁc soft tissue mobilisation in the management of soft tissue dysfunction. Manual Therapy 1998;3(1):2–11. Hunter G. The conservative management of Achilles tendinopathy. Physical Therapy in Sports 2000;1:6–14. Ja¨rvinen M, Jo´zsa L, Kannus P, Ja¨rvinen TL, Kvist M, Leadbetter W. Histopathological ﬁndings in chronic tendon disorders. Scandinavian Journal of Medicine & Science in Sports 1997;7(2):86–95. Ja¨rvinen TA, Kannus P, Paavola M, Ja¨rvinen TLN, Jo´zsa L, Ja¨rvinen M. Achilles tendon injuries. Current Opinion in Rheumatology 2001;13(2):150–5. Kader D, Saxena A, Movin T, Maffulli N. Achilles tendinopathy: some aspects of basic science and clinical management. British Journal of Sports Medicine 2002;36(4):239–49. Kannus P. Etiology and pathophysiology of chronic tendon disorders in sports. Scandinavian Journal of Medicine & Science in Sports 1997;7(2):78–85. Kannus P, Jo´zsa L, Natri A, Ja¨rinen M. Effects of training, immobilization and remobilization on tendons. Scandinavian Journal of Medicine & Science in Sports 1997;7(2):67–71. Khan KM, Maffulli N. Tendinopathy: an Achilles’ heel for athletes and clinicians. Clinical Journal of Sports Medicine 1998;8(3):151–4. Khan KM, Cook JL, Bonar F, Harcourt P, A˚stro¨m M. Histopathology of common tendinopathies. Update and implications for clinical management. Sports Medicine 1999;27(6):393–408. Kvist M. Achilles tendon injuries in athletes. Sports Medicine 1994;18(3):173–201. Maffulli N, Kader D. Tendinopathy tendo achillis. The Journal of Bone and Joint Surgery 2002;84-B(1):1–8. Maffulli N, Ewen SW, Waterston SW, Reaper J, Barrass V. Tenocytes from ruptured and tendinopathic Achilles tendons produce greater quantities of type III collagen than tenocytes from normal Achilles tendons: an in vitro model of human tendon healing. American Journal of Sports Medicine 2000;28(4):499–505. McCrory JL, Martin DF, Lowery RB, Cannon WD, Curl WW, Read HM, et al. Etiologic factors associated with Achilles tendinitis in

71

runners. Medicine & Science in Sports & Exercise 1999;31(10):1374–90. Movin T, Gad A, Reinholt FP, Rolf C. Tendon pathology in longstanding achillodynia. Biopsy ﬁndings in 40 patients. Acta Orthopaedica Scandanavia 1997;68:170–5. Nourbakhsh MR, Ottenbacher KJ. The statistical analysis of single subject data: a comparative examination. Physical Therapy 1994;74(8):768–76. Ottenbacher K. Evaluating clinical changes: strategies for occupational and physical therapists. Baltimore: Williams and Wilkins; 1986. p. 137–40. Paavola M, Kannus P, Ja¨rvinen TA, Khan K, Jo´zsa L, Ja¨rvinen M. Achilles tendinopathy. The Journal of Bone and Joint Surgery 2002;84-A(11):2062–76. Palastanga N, Field D, Soames R. Anatomy and human movement: structure and function. London: Butterworth Heinemann; 1993 [p. 538]. Price DD, McGrath PA, Raﬁ A, Buckingham B. The validation of visual analogue scales as ratio scale measures for chronic and experimental pain. Pain 1983;17(1):45–56. Robinson JM, Cook JL, Purdam C, Visentini PJ, Ross J, Maffulli N, et al. The VISA-A questionnaire: a valid and reliable index of the clinical severity of Achilles tendinopathy. British Journal of Sports Medicine 2001;35:335–41. Silbernagel KG, Thomee´ R, Thomee´ P, Karlsson J. Eccentric overload training for patients with chronic Achilles tendon pain—a randomised controlled study with reliability testing of the evaluation methods. Scandinavian Journal of Medicine & Science in Sports 2001;11(4):197–206. Singer B, Dunne J, Singer KP, Allison G. Evaluation of triceps surae muscle length and resistance to passive lengthening in patients with acquired brain injury. Clinical Biomechanics 2002;17(2):152–61. Wren T, Beaupre G, Carter D. Tendon and ligament adaptation to exercise, immobilisation and remobilisation. Journal of Rehabilitation Research and Development 2000;37(2):217–24. Youdas JW, Bogard CL, Suman VJ. Reliability of goniometric measurements and visual estimates of ankle joint active range of motion obtained in a clinical setting. Archives of Physical Medicine & Rehabilitation 1993;74(10):1113–8.

ARTICLE IN PRESS

Manual Therapy 12 (2007) 72–79 www.elsevier.com/locate/math

Technical and measurement report

The reliability of selected motion- and pain provocation tests for the sacroiliac joint Hilde Stendal Robinsona,, Jens Ivar Broxb, Roar Robinsonc, Elisabeth Bjellandc, Stein Solemc, Trym Teljec a

Section for Health Science, University of Oslo, P.O. Box 1153, Blindern, N-0318 Oslo, Norway b The National Hospital, Orthopaedic Department, N-0027 Oslo, Norway c Hans & Olaf Physiotherapy Clinic, Oslo, Norway Received 17 March 2004; received in revised form 3 June 2005; accepted 20 September 2005

Abstract The objective of the study was to assess inter-rater reliability of one palpation and six pain provocation tests for pain of sacroiliac origin. The sacroiliac joint (SIJ) is a potential source of low back and pelvic girdle pain. Diagnosis is made primarily by physical examination using palpation and pain provocation tests. Previous studies on the reliability of such tests have reported inconclusive and conﬂicting results. Fifty-six women and ﬁve men aged 18–50 years old were included in the study. Fifteen patients had ankylosing spondylitis; 30 women had post partum pelvic girdle pain for more than 6 weeks; and 16 people had no low back or pelvic girdle pain. All participants were examined twice on the same day by experienced manual therapists. Percentage agreement and kappa statistic were used to evaluate the tests reliability. Results showed percentage agreement and kappa values ranged from 67% to 97% and 0.43 to 0.84 for the pain provocation tests. For the palpation test the percent agreement was 48% and the kappa value was 0.06. Clusters of pain provocation tests were found to have good percentage agreement, and kappa values ranged from 0.51 to 0.75. In conclusion this study has shown the reliability of the pain provocation tests employed were moderate to good, and for the palpation test, reliability was poor. Clusters out of three and ﬁve pain provocation tests were found to be reliable. The cluster of tests should now be validated for assessment of diagnostic power. r 2005 Elsevier Ltd. All rights reserved. Keywords: Sacroiliac joint; Pelvic girdle pain; Clinical tests; Reliability

1. Introduction The sacroiliac joint (SIJ) as a source of back pain is a recurrent subject of controversy (Walker, 1992), but several authors state that the SIJ is a potential source for pain in the lumbar spine and buttock area (Potter and Rothstein, 1985; Shaw, 1992; Schwarzer et al., 1995; Mooney, 1997). The prevalence for SIJ dysfunction as a primary source of low back pain is reported from 0.4% (Cyriax, 1978) to 35% (Schwarzer et al., 1995) to 98% (Shaw, Corresponding author. Tel.: +47 22 85 84 21; fax: +47 22 85 84 11.

E-mail address: [email protected] (H.S. Robinson). 1356-689X/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2005.09.004

1992). This disparity is partly explained by the lack of valid criteria that prevalence can be judged by van der Wurff et al. (2000a). Although the SIJ is accepted as a source of pain, there is no general agreement concerning the different diagnostic tests and their reliability and validity. The primary diagnosis of sacroiliac pain is made by clinical history and physical examination. A wide variety of SIJ tests are available for detecting dysfunction, but no test seems to be superior to another. Many of the tests are inﬂuenced by various structures in the lower back, the hip joint and the soft tissues overlying the SIJ and consequently the tests lose their precision (Maigne et al., 1996). Furthermore, assessment and interpretation of the tests are often not

ARTICLE IN PRESS H.S. Robinson et al. / Manual Therapy 12 (2007) 72–79

standardized. However, it is necessary for test results to be both valid and reliable, since reliability alone is not sufﬁcient to support the quality of a diagnostic test (van der Wurff et al., 2000b). Several studies have assessed inter-examiner reliability of tests for SIJ pain and dysfunction. These tests are divided into those that assess movement or position by palpation (palpation tests) and those that stress the structure to reproduce the patient’s symptoms (pain provocation tests) (Laslett and Williams, 1994). Previous studies have reported that reliability is low for palpation tests and from poor to excellent for pain provocation tests (Potter and Rothstein, 1985; Laslett and Williams, 1994; Strender et al., 1997). A single test might not be sufﬁcient for diagnosing SIJ pain and some authors have suggested the use of a cluster of tests (Haas, 1991; Cibulka and Koldehoff, 1999; Kokmeyer et al., 2002; Riddle and Freburger, 2002). Others doubt this will be better because it will be possible for two observers to disagree on single tests, yet be in agreement on the ﬁnal conclusion (van der Wurff et al., 2000a; Freburger and Riddle, 2001). There must be an assumption that all the tests used in a cluster should have acceptable reliability. Van der Wurff et al. (2000a, b) published a review article on reliability studies. They found that the majority of studies (six out of 11) reported that the reliability of SIJ tests for mobility and pain provocation was low. There was a tendency for ‘positive’ conclusions to be inversely proportional to the methodological score (van der Wurff et al., 2000a). Their recommendation for future studies was that the population should include a control group without SIJ dysfunction together with a group of patients with presumed SIJ dysfunction (van der Wurff et al., 2000b). Because of the inconclusive and conﬂicting results in previous studies of tests for SIJ pain, our study aimed to assess the inter-examiner reliability for seven commonly used clinical tests. We also included tests not previously evaluated. We included only one palpation test because previous studies have reported their poor reliability. To avoid confusion and discrepancies in the testing methods, we used standardized operational deﬁnitions and the tests were performed by experienced physiotherapists who are specialized in manual therapy (MT).

2. Methods After approval by the Committee for Medical Research Ethics, Health Region I in Norway, subjects for participation were recruited consecutively from family physicians, rheumatologists, physiotherapists, obstetricians and midwives. The study included 61 participants; women and men aged 18–50 years old.

73

The ﬁrst group of participants were recruited among patients with ankylosing spondylitis (AS). They could have no ankylosis in the pelvic area and be without obvious kyphosis; the second group of patients were recruited among women with post partum pelvic girdle pain for more than 6 weeks; and the control group were healthy subjects with no low back pain, pelvic girdle pain or hip pain for the previous 3 months. The ﬁrst and second groups were selected to increase the chance that the origin of their pain could be the SIJ. The following data was collected for each participant: age, gender, diagnosis, number of childbirths (groups 2 and 3), duration of symptoms, former back pain, pain location using a pain drawing and pain intensity by using a visual analogue scale (VAS). The physiotherapists in this study, specialized in MT, had an average of 5.8 years of work experience after completing their MT education. 2.1. Testing procedure Every participant was examined by two physiotherapists, with a break of 1 h between examinations. One of the therapists examined all the participants; the second therapist was randomly selected from a pool of three therapists based on their availability for a given participant. The second assessor examined the participant ﬁrst or second according to randomisation. Both assessors examined each participant and both performed all the included tests. The therapists were blinded for the participants’ diagnosis, their pain history and pain drawing. The two assessors were also blinded to the results of the other. In clinical practice it is important to reproduce the patients’ pain in an examination in order to determine the pain origin. In this study we have tried to take this into account, by asking the participant not only if they felt pain during testing but also if they could recognize the pain. During the examination we did not ask about pain localisation. For the pain provocation tests, the results were classiﬁed using a specially designed examination form recording pain in three categories. They were: concordant pain (reproduction of the patients’ pain), discordant pain (pain different from patients’ pain) and no pain. The results for the mobility test were also classiﬁed in three categories: (1) movement in right SIJ greater than movement in left SIJ; (2) movement in right SIJ less than movement in left SIJ and; (3) movement in right SIJ equal to movement in left SIJ. After each examination the form was completed, placed in an envelope and sealed to avoid bias. 2.2. Tests Tests 1–5 are passive pain provocation tests. Test 6 is a palpation test. Test 7 is an active pain provocation

ARTICLE IN PRESS 74

H.S. Robinson et al. / Manual Therapy 12 (2007) 72–79

test. The latter means that the provocation procedure is generated by the participants themselves: (1) Compression test: The subject lies on their side and the examiner stands in front of the subject. Pressure is applied into the pelvis by the examiner leaning his chest against the uppermost iliac crest. The test is assumed to stretch the posterior sacroiliac ligaments or compress the anterior part of the SIJ (Fig. 1). (2) Distraction or ‘‘gapping’’ test: The subject lies supine close to the side of the table. The examiner stands at the side of the table facing the subject. The examiner applies cross-armed pressure to the anterior superior iliac spines (ASIS) directed laterally. This procedure is assumed to stretch the anterior sacroiliac ligaments. (3) Posterior pelvic pain provocation test (P4): The subject lies supine on the table. Standing at the side of the table, the examiner ﬂexes the ipsilateral leg to approximately 901 hip ﬂexion while knee remains relaxed. The examiner applies a graded force through the long axis of femur, there by causing a posterior shearing stress in the SIJ. Excessive adduction of the hip is avoided, because the combination of ﬂexion and adduction normally is uncomfortable or painful to the patient. (4) Patrick– Faber test: The subject lies supine on the table, and the examiner stands next to the subject.

The examiner brings the subjects ipsilateral hip and knee into ﬂexion and puts the heel against the knee of the other leg and then ﬁxates the contralateral ASIS to ensure the lower back stays in a neutral position. The ipsilateral leg is then lowered against the table and the examiner applies a light overpressure to the subject’s knee. It is assumed that both the anterior sacroiliac ligament and the hip joint are stressed. (5) Bilateral internal rotation of the hip and unilateral internal rotation of the hip (Fig. 2): The subject lies prone on the table. In both tests the examiner internally rotates the femur at the hip joint and when reaching the end of the movement it is assumed that the SIJ is stressed. (6) Test of joint-play (palpation test): The subject lies prone. When testing the right SIJ the examiner stands by the left side of the table facing the subject. The examiners right hand lifts the right ilium (the hand under anterior spina iliaca superior) and the left index ﬁnger palpates the movement between the sacrum and ilium with the heel of that hand stabilising the sacrum. This test is assumed to test the passive range of motion (ROM) between the ilium and the sacrum (Magee, 1997). (7) Drop-test: This is an active pain provocation test. The subject performs a standardized motion which

Fig. 1. Compression test.

ARTICLE IN PRESS H.S. Robinson et al. / Manual Therapy 12 (2007) 72–79

75

Fig. 2. Bilateral internal rotation of the hip joint.

may provoke the pain. Standing on one foot, the subject lifts the heel from the ﬂoor and drops down on the heel again (Fig. 3). Before the study started, the physiotherapists had training sessions to make sure that the tests were assessed and interpreted in the same manner.

positive tests). Agreement on a case deﬁnition was based on at least three/four positive tests out of ﬁve. Cluster 2 included 3 tests (P4, Patrick–Faber and internal rotation) and case deﬁnition was two positive tests out of three. The sum score was calculated as for cluster 1, with a maximum score 3.

2.3. Statistical analysis 3. Results Statistical analysis of the data was performed using the SPSS statistical package version 11.0. For each SIJ test on each side, the percentage agreement between the two therapists and the kappa agreement coefﬁcient (k) with 95% conﬁdence interval were calculated. The kappa coefﬁcient effectively discounts the proportion of agreement that is expected by chance. K ranges in values between 1 and +1. Positive values signify agreement better than chance; 0 denotes agreement no better than chance; and negative value signify agreement worse than chance (Altman, 1991). We also looked at two clusters including the pain provocation tests with the best kappa values, and calculated percentage agreement and kappa for the clusters for each side separately. The following ﬁve tests were included in cluster 1: the distraction test, P4, Patrick–Faber, bilateral internal rotation and one-sided internal rotation. We calculated a sum score for examination 1 and for examination 2 for each side separately. Maximum score was 5 on each side (ﬁve

Sixty-one people (56 women and ﬁve men) with a mean age 31.6 years participated in the study. Fifteen patients had AS and low back pain. They had no ankylosis in the pelvic area and were without obvious kyphosis; 30 women had post partum pelvic girdle pain for more than 6 weeks; and 16 subjects (the control group) did not report low back pain, pelvic girdle pain or hip pain for the previous 3 months. Table 1 presents the demographic data for the participants. The patients in the AS group (15) and post partum group (30) had experienced their symptoms for a median period of 82 and 11 months, respectively. All the patients had reported pain in the pelvic and low back region on pain drawings. All participants were examined twice, and there were no dropouts. One participant was excluded due to language problems. The percentage agreement for each pain provocation test was between 67% and 97% and the kappa coefﬁcient from moderate to good with k¼ 0:4320:84.

ARTICLE IN PRESS H.S. Robinson et al. / Manual Therapy 12 (2007) 72–79

76

The percentage agreement was 48% and the kappa was 0.06 for the joint-play test. When we merged discordant pain and no pain, and repeated the analysis, the percentage agreement increased to between 74% and 97% and kappa also

increased for all tests to k ¼ 0:4820:88. The exceptions were compression test right side and the drop test (Table 2). For the case deﬁnition (agreement on at least three out of ﬁve tests) in cluster 1, it was 76–90% agreement and kappa was between 0.51 and 0.75. For cluster 2 it was 89–91% agreement and kappa between 0.69 and 0.75 (Table 3). The healthy controls were correctly identiﬁed by both clusters.

4. Discussion The reliability of the pain provocation tests in this study is acceptable. Reliability can be inﬂuenced by three factors: the participants, the therapists and the clinical tests. The participants were recruited from patients seen by doctors, physiotherapists and midwives, and can be seen as a consecutive sample. The observed difference concerning sex and age in the groups is considered of little importance. We attempted to have only women in the AS group, but we were not able to recruit the desired number of women with AS. The overall prevalence of AS is reported from 0.1% to 1.4%. The ratio between men and women is reported 2–3:1 (Gran and Husby, 1993, 1998). According to a recent study there is no important clinical or radiological gender difference reported (Ryall and Helliwell, 1998). The control group has a matched number of male participants. The AS patients selected had no ankylosis in the pelvic area and they had no obvious kyphosis to disturb the blinding of the therapists. While preparing the study we discussed how to interpret the participants’ responses to the tests. Did the participant recognize the pain (concordant), or was it another pain (discordant)? This is similar to what we are doing in clinical practice. We are attempting to reproduce the patients’ actual pain on examination. Still the subjective nature of pain may confound the

Fig. 3. Drop-test.

Table 1 Demographic data Demographic data

Ankylosing spondylitis ðn ¼ 15Þ

Post partum pelvic pain ðn ¼ 30Þ

No back pain ðn ¼ 16Þ

Age, mean (range) Women (%) Parity

34.2 (25–50) 12 (80) 1 2 3

31.2 (25–43) 30 (100) 17 11 2 6.3 (1.5–20) 6 (20) 15 (50) 22 (73) 30 (100) 11 (4–24) 36 (21.4) 7 (1–18)

29.6 (24–34) 14 (87) 10 1 0 5.7 (2.5–12) 1 (6.25) 6 (37.5) 0 (0) 1 (6.25) 0 (0) 0 (0) 1.5 (0–9)

Age in months, youngest child, mean (range) Symptoms in earlier pregnancies (%) Previous back pain (%) Pain drawing in SIJ area (%) Low back/pelvic area on pain drawing (%) Period of pain in months, mean (range) Mean of VAS in mm (SD) Number of areas on pain drawing, mean (range)

14 (93) 12 (80) 15 (100) 82 (6–252) 47 (20.1) 14.6 (6–38)

ARTICLE IN PRESS H.S. Robinson et al. / Manual Therapy 12 (2007) 72–79

side none of the participants reported discordant pain. This most likely explains the difference for the two tests. Our ﬁndings are in agreement with other studies reporting acceptable reliability for the P4 test (Laslett and Williams, 1994; Dreyfuss et al., 1996) and we conﬁrm the high reliability reported by Ostgaard et al. (1994). We also found acceptable reliability for the compression and distraction test. Previous studies have reported divergent results ranging from poor to acceptable for these tests. McCombe et al. (1989) reported poor reliability for both compression and distraction test, most likely because of differences in examination technique (McCombe et al., 1989). The side differences and the wide conﬁdence intervals reported for the compression test and drop test in the present study suggest that the estimated kappa values of the best side for these tests are not good indicators of reliability. Therefore, an alternative interpretation of the variation reported is that these tests do not have acceptable reliability. Strender et al. (1997) conclude that the compression test is not reliable (Strender et al., 1997), while Laslett and Williams (1994) report good reliability for both tests, with an agreement of 88% and k between 0.69 and 0.73 (Laslett and Williams, 1994). In former studies most researchers have used a population of back patients when evaluating SIJ tests (Potter and Rothstein, 1985; Herzog et al., 1989; Dreyfuss et al., 1994; Laslett and Williams, 1994; Schwarzer et al., 1995; Maigne et al., 1996; Laslett, 1997; Broadhurst and Bond, 1998; Slipman et al., 1998; Cibulka and Koldehoff, 1999; Freburger and Riddle, 1999; Levangie, 1999; Meijne et al., 1999). Some researchers have studied the tests in pregnant or postpartum women (Ostgaard et al., 1994; Wormslev et al., 1994; Kristiansson and Svardsudd, 1996; Albert et al., 2000; Mens et al., 2001, 2002). In the present study we included two patient groups (patients with AS and women with post partum pelvic girdle pain) together with a healthy control group, to increase the chance that SIJ might be the origin for pain. This is in accordance with former recommendations for studies on SIJ tests (van der Wurff et al., 2000b).

interpretation of the test results (Saal, 2002). The research setting, with its restrictions, made it difﬁcult to communicate about the patients’ pain. This may also have inﬂuenced the interpretation and thereby the reliability. The order of the therapists on examination was randomized in the present study and therefore did not inﬂuence the results. None of the pairings of therapists did better than the others, but one pair had inferior results on two tests. They disagreed more on the Patrick–Faber test and on the compression test right side. These disagreements most likely contributed to a lower kappa on these tests compared with the others. During the training sessions prior to the study, there was no disagreement on performance and interpretation of the tests. The most obvious reason for this difference is that this couple examined fewer participants than the other pairings. When discordant pain and no pain were merged into no pain, the percentage agreement and kappa increased markedly for every test except for the compression test right side and the drop test. For the drop test the number of patients reporting pain was small compared with the other tests. For the compression test on right Table 2 Inter-examiner agreement on SIJ tests when no pain and discordant pain is merged n ¼ 61

% Agreement

Kappa

95% CI

Compression test right side Compression test left side Distraction test P4 right side P4 left side Patrick–Faber test right side Patrick–Faber test left side Bilateral internal rotation Internal rotation right side Internal rotation left side Drop test right side Drop test left side

82 88 82 84 87 80 74 79 90 89 97 88

0.48 0.67 0.63 0.76 0.74 0.60 0.48 0.56 0.78 0.88 0.84 0.47

0.18–0.78 0.43–0.91 0.43–0.83 0.48–0.86 0.57–0.91 0.39–0.81 0.27–0.69 0.33–0.79 0.60–0.94 0.75–1.01 0.61–1.06 0.11–0.83

77

Table 3 Cluster reliability results Case deﬁnition

Cluster Cluster Cluster Cluster Cluster Cluster

1 1 1 1 2 2

right side left side right side left side right side left side

3 3 4 4 2 2

of of of of of of

5 5 5 5 3 3

n ¼ 61

n ¼ 45 (only patients)

% Agreement

Kappa

95% CI

% Agreement

Kappa

95% CI

80 85 90 90 91 91

0.60 0.69 0.71 0.75 0.71 0.75

0.40–0.80 0.51–0.87 0.49–0.93 0.56–0.94 0.47–0.95 0.54–0.96

76 80 87 87 89 89

0.51 0.60 0.68 0.71 0.69 0.72

0.26–0.76 0.36–0.84 0.44–0.92 0.49–0.93 0.44–0.94 0.49–0.95

Cluster 1: Distraction test, P4 left or right, Patrick–Faber left or right, bilateral internal rotation, internal rotation left or right. Cluster 2: P4, Patrick–Faber and internal rotation left or right, respectively.

ARTICLE IN PRESS 78

H.S. Robinson et al. / Manual Therapy 12 (2007) 72–79

We have chosen to present test results for the left and right side separately, in contrast to what other authors have done. We have used the statistically recommended method because the tests on the left and right sides are not independent in an individual patient (Altman, 1991). Although some authors have suggested that motion and position palpation tests are reliable, (Herzog et al., 1989; Cibulka and Koldehoff, 1999) there is little evidence to support this view. In our study kappa was negative for the joint play test (0.06). This is in agreement with previous studies from Potter and Rothstein (1985) and van Deursen et al. (1990) demonstrating poor reliability for palpation tests, although they did not use the same test (Potter and Rothstein, 1985; van Deursen et al., 1990). In recent years several authors have recommended a multi-test regimen for evaluation of SIJ pain (Haas, 1991; Cibulka and Koldehoff, 1999; Kokmeyer et al., 2002; Riddle and Freburger, 2002). Cibulka and Koldehoff (1999) used only palpation tests and classiﬁed the tests as positive and negative, regardless of what side the SIJ dysfunction was (Cibulka and Koldehoff, 1999). As previously discussed it is recommended to calculate kappa for each side (Altman, 1991). Riddle and Freburger (2002) used the same four tests as Cibulka, but concluded differently. They found poor reliability for the cluster, with a slightly higher kappa on single tests (Riddle and Freburger, 2002). Palpation is considered an important tool in MT, but it seems difﬁcult to gain reliable inter-examiner agreement. A valid diagnostic test procedure requires reliable tests. (Herzog et al., 1989; van der Wurff et al., 2000a, b). Other factors inﬂuence validity, but the aim of the present study was to evaluate reliability. Kokmeyer et al. (2002) used a regimen of ﬁve SIJ pain provocation tests, and the threshold for a positive selection was set at three positive tests. Using this deﬁnition, only seven out of 78 subjects had positive tests identiﬁed by both examiners. Weighted kappa was 0.70 (Kokmeyer et al., 2002). They also calculated kappa for each individual test (distraction test, compression test, Gaenslen test, Patrick test and Thigh thrust (P4) test) and reported kappa between 0.46 and 0.67, which is in agreement with our results on single tests. Kokmeyer et al. (2002) used a cluster to make a diagnosis in each patient (Kokmeyer et al., 2002). In the present study the main purpose was to examine the reliability for single tests and no diagnoses were made by the examiners. When we calculated kappa for a cluster of tests, we looked at each side separately and the agreement on case deﬁnition was based on three positive tests out of ﬁve in cluster 1. This is in accordance with what clinicians do in their daily practice when diagnostic decisions are judged from the results of several tests. Kappa was good (0.60–0.75) (Altman, 1991), which is in accordance with the results from Kokmeyer et al.

(2002). The good kappa results should be linked to the case deﬁnition and the number of cases. When three positive tests out of ﬁve were required we found agreement on 18 and 20 cases for the right and left sides, respectively. This is a higher number of cases than in former studies, probably because of the selected population. We also looked at a case deﬁnition of four positive tests out of ﬁve for cluster 1, and found kappa to be good also for this deﬁnition (0.68–0.75) (Altman, 1991). The difference of clusters 1 and 2 is that the latter consists of three one-sided tests only (P4, Patrick–Faber and internal rotation) and case deﬁnition was set on two positive tests out of three in cluster 2. Kappa was also good for this cluster (0.69–0.75) (Altman, 1991). We found agreement on case deﬁnition for 8 and 10 cases on right and left sides, respectively, when using cluster 2. This is fewer than when using cluster 1. The asymptomatic individuals were correctly identiﬁed by both clusters. The main difference when using the two clusters is on the number of cases identiﬁed. The validity of several tests included some of the tests included in this present study have been reported previously (Dreyfuss et al., 1996; Slipman et al., 1998). There is still no agreement about the gold standard for SIJ pain. Anaesthetic blocks have been used, but seem to investigate intra-articular sources of pain, thus neglecting the structures surrounding the SIJ (van der Wurff et al., 2000b). The pain provocation tests claim to load the whole SIJ complex including the surrounding structures. If this is the case such tests should be classiﬁed as provocation tests for pelvic girdle pain, but a gold standard which takes this into consideration is still not available (van der Wurff et al., 2000b; Kokmeyer et al., 2002). Our aim was to assess reliability. Results were good for a few individual tests and for a medium and a small cluster of tests, but poor for the commonly used palpation tests. Future studies should assess the validity of individual tests and especially clusters of tests. In the present study the examiners were physiotherapists experienced in MT. They were working in the same setting, attempting to optimize agreement by practicing the skills of the required procedures. Thus, less agreement is expected in an ordinary clinical setting or between various medical specialists and procedures.

5. Conclusion Among experienced therapists, reliability was moderate to good for all the pain provocation tests and poor for the palpation test (joint play). For the compression test and the drop test the conﬁdence intervals were wider and the side difference larger than for the other tests. In clinical practice we usually make conclusions based on the results of several tests. The clusters of three and ﬁve pain provocation tests used in this study showed good

ARTICLE IN PRESS H.S. Robinson et al. / Manual Therapy 12 (2007) 72–79

reliability, although further studies are needed to assess their validity.

Acknowledgment The authors would like to thank Professor Nina Vollestad at Section for Health Science, University of Oslo for valuable advice on the statistics in this study. References Altman DG. Practical statistics for medical research. London: Chapman & Hall; 1991. Albert H, Godskesen M, Westergaard J. Evaluation of clinical tests used in classiﬁcation procedures in pregnancy-related pelvic joint pain. European Spine Journal 2000;9(2):161–6. Broadhurst NA, Bond MJ. Pain provocation tests for the assessment of sacroiliac joint dysfunction [see comments]. Journal of Spinal Disorders 1998;11(4):341–5. Cibulka MT, Koldehoff R. Clinical usefulness of a cluster of sacroiliac joint tests in patients with and without low back pain. Journal of Orthopaedic and Sports Physical Therapy 1999;29(2):83–9. Cyriax J. Textbook of orthopaedic medicine, 7th ed. London: Cassell Ltd; 1978. Dreyfuss P, Dryer S, Grifﬁn J, Hoffman J, Walsh N. Positive sacroiliac screening tests in asymptomatic adults. Spine 1994;19(10):1138–43. Dreyfuss P, Michaelsen M, Pauza K, McLarty J, Bogduk N. The value of medical history and physical examination in diagnosing sacroiliac joint pain [see comments]. Spine 1996;21(22):2594–602. Freburger JK, Riddle DL. Measurement of sacroiliac joint dysfunction: a multicenter intertester reliability study. Physical Therapy 1999;79(12):1134–41. Freburger JK, Riddle DL. Using published evidence to guide the examination of the sacroiliac joint region. Physical Therapy 2001;81(5):1135–43. Gran JT, Husby G. The epidemiology of ankylosing spondylitis. Seminars in Arthritis and Rheumatism 1993;22(5):319–34. Gran JT, Husby G. Clinical, epidemiologic, and therapeutic aspects of ankylosing spondylitis. Current Opinion in Rheumatology 1998;10(4):292–8. Haas M. Interexaminer reliability for multiple diagnostic test regimens. Journal of Manipulative and Physiological Therapeutics 1991;14(2):95–103. Herzog W, Read LJ, Conway PJ, Shaw LD, McEwen MC. Reliability of motion palpation procedures to detect sacroiliac joint ﬁxations. Journal of Manipulative and Physiological Therapeutics 1989;12(2):86–92. Kokmeyer DJ, van der Wurff P, Aufdemkampe G, Fickenscher TC. The reliability of multitest regimens with sacroiliac pain provocation tests. Journal of Manipulative and Physiological Therapeutics 2002;25(1):42–8. Kristiansson P, Svardsudd K. Discriminatory power of tests applied in back pain during pregnancy. Spine 1996;21(20):2337–43. Laslett M. Pain provocation sacroiliac tests: reliability and prevalence. In: Vleeming A, et al., editors. Movement, stability & low back pain. The essential role of the pelvis. 1st ed. Edinburgh: Churchill Livingstone; 1997. p. 287–95. Laslett M, Williams M. The reliability of selected pain provocation tests for sacroiliac joint pathology. Spine 1994;19(11):1243–9. Levangie PK. Four clinical tests of sacroiliac joint dysfunction: the association of test results with innominate torsion among patients with and without low back pain. Physical Therapy 1999;79(11): 1043–57.

79

Magee DJ. Orthopaedic physical assessment. 3rd ed. Philadelphia: W.B. Saunders Company; 1997. Maigne JY, Aivaliklis A, Pfefer F. Results of sacroiliac joint double block and value of sacroiliac pain provocation tests in 54 patients with low back pain. Spine 1996;21(16):1889–92. McCombe PF, Fairbank JC, Cockersole BC, Pynsent PB. Volvo Award in clinical sciences. Reproducibility of physical signs in lowback pain. Spine 1989;14(9):908–18. Meijne W, van Neerbos K, Aufdemkampe G, van der WP. Intraexaminer and interexaminer reliability of the Gillet test. Journal of Manipulative and Physiological Therapeutics 1999;22(1):4–9. Mens JM, Vleeming A, Snijders CJ, Koes BW, Stam HJ. Reliability and validity of the active straight leg raise test in posterior pelvic pain since pregnancy. Spine 2001;26(10):1167–71. Mens JM, Vleeming A, Snijders CJ, Ronchetti I, Stam HJ. Reliability and validity of hip adduction strength to measure disease severity in posterior pelvic pain since pregnancy. Spine 2002;27(15):1674–9. Mooney V. Sacroiliac joint dysfunction. In: Vleeming A, Mooney V, Dorman T, Snijders C, Stoeckart R, editors. Movement, stability & low back pain. The essential role of the pelvis. 2nd ed. Edinburgh: Churchill Livingstone; 1997. p. 37–52. Ostgaard HC, Zetherstrom G, Roos-Hansson E. The posterior pelvic pain provocation test in pregnant women. European Spine Journal 1994;3(5):258–60. Potter NA, Rothstein JM. Intertester reliability for selected clinical tests of the sacroiliac joint. Physical Therapy 1985;65(11):1671–5. Riddle DL, Freburger JK. Evaluation of the presence of sacroiliac joint region dysfunction using a combination of tests: a multicenter intertester reliability study. Physical Therapy 2002;82(8):772–81. Ryall NH, Helliwell PS. A critical review of ankylosing spondylitis. Critical Reviews in Physical and Rehabilitation Medicine 1998;10(3):265–301. Saal JS. General principles of diagnostic testing as related to painful lumbar spine disorders: a critical appraisal of current diagnostic techniques. Spine 2002;27(22):2538–45. Schwarzer AC, Aprill CN, Bogduk N. The sacroiliac joint in chronic low back pain. Spine 1995;20(1):31–7. Shaw JL. The role of the sacroiliac joint as a cause of low back pain and dysfunction. In: Vleeming A, Mooney V, Snijders C, Dorman T, editors. First interdisciplinary world congress on low back pain and its relation to the sacroiliac joint, San Diego, CA, 1992. p. 67–80. Slipman CW, Sterenfeld EB, Chou LH, Herzog R, Vresilovic E. The predictive value of provocative sacroiliac joint stress maneuvers in the diagnosis of sacroiliac joint syndrome. Archives of Physical Medicine and Rehabilitation 1998;79(3):288–92. Strender LE, Sjoblom A, Sundell K, Ludwig R, Taube A. Interexaminer reliability in physical examination of patients with low back pain. Spine 1997;22(7):814–20. van der Wurff P, Hagmeijer RH, Meyne W. Clinical tests of the sacroiliac joint. A systemic methodological review. Part 1: reliability. Manual Therapy 2000a;5(1):30–6. van der Wurff P, Meyne W, Hagmeijer RH. Clinical tests of the sacroiliac joint. Manual Therapy 2000b;5(2):89–96. van Deursen LLJM, Patijn AL, Ockhuysen AL, Vortman BJ. The Value of some clinical tests of the sacroiliac joint. Journal of Manual Medicine 1990;5:96–9. Walker JM. The sacroiliac joint: a critical review. Physical Therapy 1992;72(12):903–16. Wormslev M, Juul AM, Marques B, Minck H, Bentzen L, Hansen TM. Clinical examination of pelvic insufﬁciency during pregnancy. An evaluation of the interobserver variation, the relation between clinical signs and pain and the relation between clinical signs and physical disability. Scandinavian Journal of Rheumatology 1994;23(2):96–102.

ARTICLE IN PRESS

Manual Therapy 12 (2007) 80–83 www.elsevier.com/locate/math

Case report

A contractile dysfunction of the shoulder Chris Littlewooda,, Stephen Mayb a

Trinity Medical Centre, Musculoskeletal Services, Wakefield West PCT/Mid Yorkshire Hospitals NHS Trust, Thornhill Street, Wakefield WF1 1PG, UK b Faculty of Health and Wellbeing, Sheffield Hallam University, UK Received 17 June 2005; received in revised form 7 September 2005; accepted 16 November 2005

1. Introduction

2. History

The McKenzie method of mechanical diagnosis and therapy (McKenzie, 1981, 1990; McKenzie and May, 2000, 2003), is commonly used in the management of spinal disorders (Foster et al., 1999; Gracey et al., 2002; Jackson, 2001). The method involves a mechanical evaluation, in which end-range repeated movements are performed and symptom and mechanical responses are monitored. According to the response patients are then classiﬁed in one of three mechanical syndromes: derangement, dysfunction, and postural syndrome. The mechanical evaluation when used with spinal patients has demonstrated reliability amongst trained clinicians (Razmjou et al., 2000; Fritz et al., 2000; Kilpikoski et al., 2002), and prognostic validity (Long, 1995; Sufka et al., 1998; Werneke et al., 1999; Werneke and Hart, 2001). The classiﬁcation is used to determine management, as the different mechanical syndromes are treated in different ways. In McKenzie’s original description (McKenzie, 1981) he stated that the system of diagnosis and treatment could equally well be applied to extremity problems, and a description of how this could be done was published more recently (McKenzie and May, 2000). The purpose of this case report is to describe this method of diagnosis and treatment as applied to a patient with a shoulder problem. The patient was classiﬁed as having a contractile dysfunction and treated (see Table 1) according to the principles of mechanical diagnosis and therapy.

A 57-year-old gentleman presented with a 12-month history of pain affecting the left shoulder and upper arm (Fig. 1). The onset was insidious and the problem was now unchanging over time. The symptoms were intermittent, and consistently produced with lifting, elevating the arm, reaching behind the back or lying on the left side. Symptoms eased off with rest. He was generally ﬁt and well and he reported no previous problems relating to the shoulder or neck. At the time of initial consultation he was not currently taking medication, except analgesics as needed. Previous non-steroidal anti-inﬂammatory drugs (NSAIDs) had been ineffective and there had been no further treatment to date. Recent ultrasound evaluation revealed a partial tear of the supraspinatus tendon. The gentleman was continuing to work as a printer. This job requires prolonged neutral standing positions interspersed with regular transfer of print screens of various weights and dimensions. Repetitive lifting is required at both above and below shoulder level.

Corresponding author. Tel./fax: +44 (0)1924 398 063.

E-mail address: [email protected] (C. Littlewood). 1356-689X/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2005.11.002

3. Examination Observation of the patient revealed normal appearance and normal alignment of the neck and shoulder. There was no pain at rest. A baseline examination of the shoulder revealed a full active and passive range of movement of the left shoulder. There was an arc of pain provoked between 801 and 1201 with active abduction and end range pain produced with lateral rotation and reaching behind the back. Scapulo-humeral rhythm was

ARTICLE IN PRESS C. Littlewood, S. May / Manual Therapy 12 (2007) 80–83 Table 1 Characteristics of contractile dysfunction (McKenzie and May, 2000) Traumatic or insidious onset Symptoms present for several months Range of movement largely preserved Consistent response to initial testing Symptoms provoked by resisted movements, particularly resisted midrange movements

81

The patient was instructed to perform active abduction in the ‘target zone’ or where the painful arc was felt. Advice was given that the exercises should be performed 10–15 times, or until fatigue, at least twice per day. Education related to the expected response was given, namely that the exercises should provoke the symptoms which should not remain worse as a result. However, reassurance was offered that initially this treatment regimen may provoke a temporary mild exacerbation of pain which should not last greater than 3 weeks from commencing the programme. This temporary exacerbation of symptoms is thought to represent reinvigoration of the remodelling process (McKenzie and May, 2000).

4. Review appointment 1 month later

Fig. 1. Body chart illustrating pain presentation.

deﬁned as normal during active movement testing. Pain was also provoked during isometric abduction and lateral rotation. A repeated movement examination of the cervical spine did not reveal any local abnormality nor did these examination procedures provoke a symptomatic response or create any change in the presenting shoulder problem. Further examination of the shoulder focused upon the most provocative movements, and repeated movements revealed a consistent response. Pain was consistently provoked during the movement of active abduction, and an end range pain response was provoked consistently with lateral rotation and reaching the hand behind the back. These responses did not alter with repetition, or with increased load using a small weight. Thus, these tests produced pain with movement but were no worse as a result of that movement. Repeated isometric testing of abduction and lateral rotation similarly provoked pain at the shoulder that was no worse following the test. The resisted tests were the most symptomatic. Following this examination process, it was felt that the problem related to the shoulder and a provisional mechanical diagnosis of contractile dysfunction was made.

Despite a temporary initial exacerbation of symptoms, the patient reported good exercise compliance. Furthermore, he had noticed a reduction in pain and increased freedom of movement. Examination revealed a reduced painful arc during abduction from 901 to 1201 only. Furthermore, the previous end of range pain provoked with lateral rotation and reaching behind the back had now been abolished. At this stage isometric testing of abduction and lateral rotation was negative. The initial mechanical diagnosis was conﬁrmed and the patient was instructed to continue loaded active exercise within the target zone as previously described. Instruction regarding the potential need to increase the load of the exercise, i.e. through the use of a hand weight, was given to maintain the desired response, namely exercise which produces symptoms that are no worse upon cessation of the exercise.

5. Review appointment 6 weeks later (10 weeks following initial consultation) At this stage the patient reported that he no longer had any signiﬁcant problems relating to the shoulder. Examination conﬁrmed his report with no abnormality detected at the shoulder. The contractile dysfunction had improved without complication over the expected time frame. Based upon this, the patient was discharged from the service.

6. Discussion Patients with shoulder problems are frequently encountered in health care (Van der Windt et al., 1995; May, 2003). The natural history is not always good with only about 50% of new cases in primary care recovering completely in 6 months (Van der Windt et al.,

ARTICLE IN PRESS 82

C. Littlewood, S. May / Manual Therapy 12 (2007) 80–83

1995; Croft et al., 1996; Kuijpers et al., 2004). Rotator cuff pathology is one of the most commonly cited causes of shoulder pain (Van der Windt et al., 1995; Herberts et al., 1981; Chard et al., 1988, 1991), with the long-term outcome frequently poor despite treatment (Chard et al., 1988). The reliability of making a speciﬁc structural diagnosis at the shoulder is weak (Liesdek et al., 1997; De Winter et al., 1999), and the validity of tests only moderate (Calis et al., 2000). A recent systematic review found that the best method of treating rotator cuff pathology is unclear, in part due to the low methodological quality of the studies. However, exercise appeared to have the most consistent evidence in its favour (Grant et al., 2004). Part of the problem in determining the best management strategy lies in a commonly used synonym, tendonitis, with the implication of chronic inﬂammation in the sufﬁx itis. In fact inﬂammatory cells have never been found in chronic tendon problems (Khan et al., 1999; Kraushaar and Nirschl, 1999; Chard et al., 1994). What is found is a loss of collagen continuity and a failed attempt at repair with ground substance and ﬁbroblasts (Kraushaar and Nirschl, 1999; Chard et al., 1994; Khan et al., 2000). This misnomer means that treatment based on the assumption of inﬂammation, i.e. NSAIDs, rest and ultrasound, is misplaced and inappropriate. Tendinosis, not tendonitis, is the appropriate term for the pathology. The modern emphasis is now on exercises, especially eccentric exercise, which will load the affected tissue, stimulate the repair process and promote tendon healing (Khan et al., 1999, 2000; Kraushaar and Nirschl, 1999). The mechanical diagnosis and therapy classiﬁcation of contractile dysfunction avoids a structurally speciﬁc diagnosis, but ascribes pain to ‘structurally impaired soft tissues’ within the muscle or tendon recognized chieﬂy by pain on loading or a resisted test (McKenzie and May, 2000). The ‘contractile dysfunction’ model may represent rotator cuff tendinopathy/tendinosis, but rather than implicate speciﬁc tissues, the mechanical diagnosis represents a functional diagnosis. Furthermore, this recognizes that musculoskeletal systems do not act in isolation but rather ‘functional units.’ Pain will persist until the tissue is remodelled, which is done by loading the impairment with active or resisted movements. Loading must be sufﬁcient to produce a degree of pain that settles once the repeated movements have ceased. This active exercise based management strategy for dysfunction of the contractile unit seems in line with contemporary knowledge about tendon pathology.

7. Conclusion This case study details the history and physical examination of a patient with a shoulder problem that

according to the principles of mechanical diagnosis and therapy was classiﬁed as having a contractile dysfunction. The problem was treated with loading the structurally impaired soft tissue, and over several weeks it resolved. Such exercises that stress the affected tissue would appear to be highly relevant in the rehabilitation of chronic tendon problems. The mechanical diagnosis and therapy classiﬁcation system would appear to have relevance in extremity as well as spinal problems.

References Calis M, Akgun K, Birtane M, Karacan I, Calis H, Tuzun F. Diagnostic values of clinical diagnostic tests in subacromial impingement syndrome. Annals of Rheumatic Disease 2000;59:44–7. Chard MD, Sattelle LM, Hazleman BL. The long-term outcome of rotator cuff tendonitis—a review study. British Journal Rheumatology 1988;27:385–9. Chard MD, Hazleman R, Hazleman BL, King RH, Reiss BB. Shoulder disorders in the elderly: a community study. Arthritis Rheumatism 1991;34:766–9. Chard MD, Cawston TE, Riley GP, Gresham GA, Hazleman BL. Rotator cuff degeneration and lateral epicondylitis: a comparative histological study. Annals of Rheumatic Disease 1994;53:3034. Croft P, Pope D, Silman AJ, et al. The clinical course of shoulder pain: prospective cohort study in primary care. British Medical Journal 1996;313:601–2. De Winter AF, Jans MP, Scholten RJPM, Deville W, van Schaardenburg D, Bouter LM. Diagnostic classiﬁcation of shoulder disorders: interobserver agreement and determinants of disagreement. Annals of Rheumatic Disease 1999;58:272–7. Foster NE, Thompson KA, Baxter GD, Allen JM. Management of nonspeciﬁc low back pain by physiotherapists in Britain and Ireland, a descriptive questionnaire of current clinical practice. Spine 1999;24:1332–42. Fritz JM, Delitto A, Vignovic M, Busse RG. Interrater reliability of judgements of the centralisation phenomenon and status change during movement testing in patients with low back pain. Archives of Physical Medicine and Rehabilitation 2000;81:57–61. Gracey JH, McDonough SM, Baxter GD. Physiotherapy management of low back pain. A survey of current practice in Northern Ireland. Spine 2002;27:406–11. Grant HJ, Arthur A, Pichora DR. Evaluations of intervention for rotator cuff pathology: a systematic review. Journal of Hand Therapy 2004;17:274–99. Herberts P, Kadefors R, Andersson G, Petersen I. Shoulder pain in industry: an epidemiological study on welders. Acta Orthopaedica Scandinavia 1981;52:299–306. Jackson DA. How is low back pain managed? Retrospective study of the ﬁrst 200 patients with low back pain referred to a newly established community-based physiotherapy department. Physiotherapy 2001;87:573–81. Khan KM, Cook JL, Bonar F, Harcourt P, Astrom M. Histopathology of common tendinopathies. Sports Medicine 1999;27:93–408. Khan KM, Cook JL, Taunton JE, Bonar F. Overuse tendinosis, not tendonitis: Part 1: a new paradigm for a difﬁcult clinical problem. Physician Sportsmedicine 2000;28:38–46. Kilpikoski S, Airaksinen O, Kankaanpaa M, Leminen P, Videman T, Alen M. Interexaminer reliability of low back pain assessment using the McKenzie method. Spine 2002;27:E207–14. Kraushaar BS, Nirschl RP. Tendinosis of the elbow (tennis elbow). Journal of Bone and Joint Surgery 1999;81A:259–78.

ARTICLE IN PRESS C. Littlewood, S. May / Manual Therapy 12 (2007) 80–83 Kuijpers T, van der Windt DAWM, van der Heijden GJMG, Bouter LM. Systematic review of prognostic cohort studies on shoulder disorders. Pain 2004;109:420–31. Liesdek C, van der Windt DA, Koes BW, Bouter LM. Soft-tissue disorders of the shoulder. Physiotherapy 1997;83:12–7. Long A. The centralisation phenomenon. Its usefulness as a predictor of outcome in conservative treatment of chronic low back pain. Spine 1995;20:2513–21. May S. An outcome audit for musculoskeletal patients in primary care. Physiotherapy Theory and Practice 2003;19:189–98. McKenzie RA. The lumbar spine. Mechanical diagnosis and therapy. New Zealand: Spinal Publications; 1981. McKenzie RA. The cervical and thoracic spine. Mechanical diagnosis and therapy. New Zealand Ltd.: Spinal Publications; 1990. McKenzie R, May S. The human extremities mechanical diagnosis and therapy. New Zealand Ltd.: Spinal Publications; 2000.

83

McKenzie R, May S. The lumbar spine mechanical diagnosis and therapy. 2nd ed. New Zealand Ltd: Spinal Publications; 2003. Razmjou H, Kramer JF, Yamada R. Intertester reliability of the McKenzie evaluation in assessing patients with mechanical lowback pain. Journal of Orthopaedic and Sports Physical Therapy 2000;30:368–89. Sufka A, Hauger B, Trenary M, Bishop B, Hagen A, Lozon R, et al. Centralisation of low back pain and perceived functional outcome. Journal of Orthopaedic and Sports Physical Therapy 1998;27:205–12. Van der Windt DA, Koes BW, de Jong BA, Bouter LM. Shoulder disorders in general practice: incidence, patient characteristics, and management. Annals of Rheumatic Disease 1995;54:959–64. Werneke M, Hart DL, Cook D. A descriptive study of the centralisation phenomenon. A prospective analysis. Spine 1999;24:676–83. Werneke M, Hart DL. Centralization phenomenon as a prognostic factor for chronic pain or disability. Spine 2001;26:758–65.

Manual Therapy (2007) 12(1), 84

Diary of events

5th International Course on the Hand October 21–25, 2007 Target audience: colleagues of the following disciplines; physical medicine and rehabilitation, plastic- and hand surgery, physical- and occupational therapy and other health care professionals, interested in the topic of the hand Lectures include: Prof. Dr. S.E.R Hovius, Ton A.R. Schreuders PT, PhD and G. Van Strein MSc Accreditation applied for at the EACCME (Accreditation Council) of the European Union of Medical Specialists (UEMS) More information and registration: website: www.vitalmedbodrum.com E-mail: [email protected]

10th International Conference in Mechanical Diagnosis and Therapy — The Evidence Mounts 23–25 March 2007 Queenstown, New Zealand Honorary Chairman: Robin McKenzie Presented by: The McKenzie Institute International For more information visit: www.mckenziemdt.org 4th Low Back Pain Symposium April 30–May 3, 2007 Target audience: colleagues of the following disciplines; physical medicine and rehabilitation, orthopaedic surgery, neurosurgery, physical-, occupational-, manual- and Mensendieck therapy. Moreover, company doctors, medical advisors of insurance companies and other health care professionals interested in the topic of low back pain. Chairmen: Prof. Dr. Henk J. Stam, Prof. David Niv M.D. FIPP, Prof. Dr. Mehmet Zileli Accreditation applied for at the EACCME (Accreditation Council) of the European Union of Medical Specialists (UEMS) More information and registration: website: www.vitalmedbodrum.com E-mail: [email protected]

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

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

First international Fascia Research Congress Basic Science and Implication for Conventional and Complementary Health Care 4–5 October 2007 The Conference Center, Harvard Medical School Boston MA http://www.fascia2007.com

84

ARTICLE IN PRESS

Manual Therapy 12 (2007) e1 www.elsevier.com/locate/math

Letter to the Editor Immediate effects of thoracic manipulation in patients with neck pain: A randomized clinical trial I read with interest the paper by Cleland et al. on the immediate effects of thoracic manipulation in patients with neck pain (Manual Therapy 2005, 10: 127–35), and wish to share some comments with your readers, in light of our practice. In 1998, the French Society of Manual Medicine (SOFMMOO) issued guidelines on cervical manipulation (Maigne, 1998; Vautravers and Maigne, 2000 and at www.sofmmoo.com, in English). One of them, the most important, advised against the use of a manoeuvre with rotational thrust in females under the age of 50 years. Instead of them, (non-thrust cervical techniques, and/or) thrust manipulations of the upper thoracic spine were recommended. Actually, it had been noticed empirically by the author that these latter could efﬁciently relieve some forms of common neck pain. The hypothetic mechanism was that they could act on

1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.02.005

the cervicothoracic muscles (semispinalis, splenius) and relax them. I am pleased to see that, thanks to the study by Cleland et al. there is now a serious scientiﬁc basis to this recommendation. References Maigne JY. les recommandations de la SOFMMOO. Reviews in Medical Orthopedics 1998;52:16–7. Vautravers P, Maigne JY. Cervical spine manipulation and the precautionary principle. Joint Bone Spine 2000;67:272–6.

(President) Jean-Yves Maigne, MD Medecine Physique, Hopital Hotel-Dieu 75181 Paris Cedex 04, France E-mail address: [email protected]

ARTICLE IN PRESS

Manual Therapy 12 (2007) e2 www.elsevier.com/locate/math

Letter to the Editor Re: Improving application of neurodynamic (neural tension) and treatments: A message to researchers and clinicians M. Shacklock, 10(3) (August 2005) 175–179 As researchers and clinicians we welcome the timely reminder of the part that neural physiology plays in the symptoms experienced by many of our patients. It is only with an understanding of the physiological effects of nerve strain and nerve inﬂammation that clinicians can fully appreciate the potential undesirable effects that sustained and symptom producing nerve ‘‘stretching’’ techniques can have. However, other than this we are not sure what else this editorial offers to current thinking around ‘‘adverse nerve tension’’ or the physiological changes that produce abnormal responses in patients when applying strain on neural tissue. Recent scientiﬁc publications appear to have been overlooked by the author, surprising in an editorial that prompts researchers and clinicians to investigate the literature fully. Nerve movement and strain, contrary to the author assertions, have been measured in vivo, using ultrasound imaging, in both patients and controls (Hough et al., 2000; Dilley et al., 2001, 2003; Greening et al., 2001, 2005). In fact in what is considered to be the commonest nerve entrapment ‘‘carpal tunnel syndrome’’ no reduction in longitudinal sliding through the carpal tunnel was observed leading to the suggestion that mechanisms other than increased nerve strain are responsible for symptoms (Erel et al., 2003). Much more is now known regarding the effects of nerve inﬂammation. Both nerve stretching (within the physiological range) and nerve pressure produce ectopic ﬁring in nerves that are inﬂamed (Bove et al., 2003; Dilley and Lynn, 2004; Greening, 2005). This can lead to a cascade of changes within the central nervous system some of which may result in chronic pain. Thus, neural stretching techniques that produce painful responses are unlikely to be in the patient’s best interest. More up to date thinking in the treatment of these conditions is one that proposes an integrated multi system approach, one less devoted to just neural tissue. In conclusion less debate over nomenclature and more grounding in scientiﬁc fact would better serve manual

1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.01.004

therapists as they seek to communicate with their medical colleagues and to effectively treat patients with these problems. References Bove GM, Ransil BJ, Lin HC, Leem JG. Inﬂammation induces ectopic mechanical sensitivity in axons of nociceptors innervating deep tissues. Journal of Neurophysiology 2003;90:1949–55. Dilley A, Lynn B. Stretch responses of axons in regions of local inﬂammation in rat peripheral nerves. Comparative Biochemistry and Physiology A-Molecular and Intergrative Physiology 2004;137(Suppl 1):S111–52. Dilley A, Greening J, Lynn B, Leary R, Morris V. The use of crosscorrelation analysis between high frequency ultrasound images to measure longitudinal median nerve movement. Ultrasound in Medicine and Biology 2001;27:1211–8. Dilley A, Lynn B, Greening J, Deleon N. Quantitative in vivo studies of median nerve sliding in response to wrist, elbow, shoulder and neck movements. Clinical Biomechanics (Bristol, Avon) 2003; 18(10):899–907. Erel E, Dilley A, Greening J, Morris V, Lynn B. Longitudinal sliding of the median nerve in the forearm during ﬁnger movements in normal subjects and in patients with carpal tunnel syndrome. Journal of Hand Surgery (Br) 2003;28(5):439–44. Greening J. How inﬂammation and minor nerve injury contribute to pain in nerve root and peripheral neuropathies. In: Boyling J, Jull G, editors, Modern manual therapy of the vertebral column. 3rd ed. 2005. p. 205–15. Greening J, Dilley A, Lynn B. In vivo study of nerve movement and mechanosensitivity of the median nerve in whiplash and nonspeciﬁc arm pain patients. Pain 2005;115:248–53. Greening J, Lynn B, Leary R, Warren L, O’Higgins P, Hall-Craggs M. The use of ultrasound imaging to demonstrate reduced movement of the median nerve during wrist ﬂexion in patients with non-speciﬁc arm pain. Journal of Hand Surgery 2001;26B(5): 401–6. Hough AD, Moore AP, Jones MP. Measuring longitudinal nerve motion using ultrasonography. Manual Therapy 2000;3: 173–80.

Jane Greening Department of Physiology, University College London, London, UK Rachel Leary R.L. Physiotherapy Clinics Ltd., London, UK

ARTICLE IN PRESS

Manual Therapy 12 (2007) e3–e6 www.elsevier.com/locate/math

Reply to Letter to the Editor Dear Editors, I thank Dr. Greening and Dr. Leary for offering their opinions about my Editorial in Manual Therapy (Shacklock, 2005a) and the Editors, Prof. Moore and Prof. Jull, for the opportunity to respond. Based on the issues raised by Greening and Leary, it is quite clear that ongoing discussion of this important aspect of health science is much needed. I apologize for the lengthy response but I believe it necessary to do the discussion justice. Greening and Leary are certainly correct in that to quote them could have been quite justiﬁable. However, the Editorial was not speciﬁcally focused on the content of their work nor was it a review of the literature or discussion on the causes and treatment of neurogenic pain. If it were, it would have extended to hundreds of references on aspects such as neuritis, neuropathology, the nature of nerve pain in the absence of nerve conduction changes, immune, inﬂammatory and nociceptive mechanisms within nerve tissue, neurobiomechanics, central pain mechanisms and so on, which was well beyond the scope of the Editorial. An aim of the Editorial was to highlight some of the key issues faced by many clinicians and some researchers in relation to neurodynamic testing and treatment. The text under the section on Review of the Literature was a case in point and did not mean that the Editorial was also a literature review. The background was that, in reviewing research submissions for journals and organizations such as Manual Therapy, Pain and the Physiotherapy Research Foundation, patterns had emerged and these were what prompted the Editorial. Another of the objectives was to illustrate generally what problems exist and offer some solutions. In relation to the comments by Greening and Leary on neurogenic pain being a multisystem problem that requires such treatment, we are in complete agreement. But, again, the Editorial was not so much about neurogenic pain and its treatment as it was about ‘neurodynamic (formerly neural tension) tests’. With that, the Editorial at no time suggested that neurodynamic tests were a panacea for neurogenic pain, or that treatment should be localized to this modality, nor should it have. In addition, a focal aspect of the Editorial was not so much whether alterations in physiology occur with 1356-689X/$ - see front matter r 2006 Published by Elsevier Ltd. doi:10.1016/j.math.2006.01.005

neuropathic pain but rather that they most likely happen in our patients and from then on the discussion was intended to be about how we might consider neurodynamic testing. As mentioned earlier, to discuss the many speciﬁc mechanisms of mechanosensitivity and other forms of neuropathophysiology was well beyond the scope of the Editorial and there was certainly no intention to omit speciﬁc studies on the subject. In fact, many authors in addition to Greening and Leary would have had good justiﬁcation for complaint had physiological aspects been a focus and I agree that the Editorial would have been wholly unsatisfactory. In relation to my statement that we do not have direct measurements of neural tension in our patients, clariﬁcation is necessary. I did not say that we have no measurements of ‘movement’ and ‘strain’. What I did say is: ‘‘... in the patients in whom an abnormal test is manifest, we have no direct measurements of the tension in the nervous systemy’’. This is based on the fact that tension is different from movement and strain. Erel et al. (2003) measured movement and strain but they did not measure tension. In another study, Dilley et al. (2005) showed that neural tissues become active more easily in the inﬂamed state than in the non-inﬂamed state when they are elongated. This was in animals and my discussion at that point was about humans. For us to conclude soundly that our patients hold ‘adverse neural tension’, we really need measurements that are as direct, and as speciﬁc to the location in question, as possible. I know of no studies in which neural tension in patients has been measured, neither did Greening and Leary provide any. Hence, I see no conﬂict between the work of Greening and Leary and the content of my Editorial in this particular aspect. In taking into account the general message of my Editorial and what Greening and Leary state in their letter, we are actually often in agreement about the issue of neural tension—that stretching of nerves may not provide the promise that many had originally hoped—but we say it in a different way. This aspect of our dialogue actually highlights the importance of nomenclature and a common understanding of the terms we use. The other part of my comment on there not being available direct measurements of neural tension in our patients relates to the performance of neurodynamic tests clinically. Again, therapists currently have no

ARTICLE IN PRESS e4

Reply to Letter to the Editor / Manual Therapy 12 (2007) e3–e6

measurement whatsoever of neural tension. Manually detected resistance to movement with neurodynamic testing, which is likely to be produced by muscle contraction (Hall et al., 1998; Coppieters et al., 1999; van der Heidi et al., 2001), movement patterns and reported sensory responses can be observed but, again, these are not neural tension. With respect to previous terminology in the neurodynamics/neural tension area, I understand the frustration experienced by Greening and Leary as many others have also experienced this too, including myself. This is exactly the reason I initiated the concept of neurodynamics (Shacklock, 1995a, b). They suggest that I should place less emphasis on discussion of nomenclature. However, I believe that the way we discuss and use terms is essential to effective communication of knowledge between ourselves and other disciplines. This is evidenced by some of the world’s most respected groups adopting taxonomies and classiﬁcations for the very purpose of elucidating and communicating knowledge in a consistent and accurate way, and these came from much discussion. The Taxonomy of chronic pain terms produced by the International Association for the Study of Pain (Merskey and Bogduk, 1994) is such an example, and there are many others. I worry that to reduce discussion in this regard might have the effect of stiﬂing development of open communication and common understanding of the very things that need clariﬁcation. I agree that any terminology we use should be as scientiﬁc as possible. However, medical terms do not always cover the phenomena that we are dealing with here and, with all due respect, our medical colleagues generally know very little about clinical neurodynamics. Hence, I think manual therapists are in a fantastic position to initiate change and lead the ﬁeld. The comments by Greening and Leary about nerve stretching emphasize a key point—that nerve stretching techniques have in the past been the mainstay of the ‘neural tension’ technique. This is naturally a real problem and one that I have found burdensome since the early 1990s. Nevertheless, the approach is still widespread in manual therapy and it is my understanding (along with that of many other therapists), that ‘stretch’ suggests elongation of a nerve beyond its elastic limit so that the resting length of the nerve is altered by the manual technique. I agree with Greening and Leary—that it may not be the best thing because stretch can damage nerves (Denny-Brown and Doherty, 1945; Sunderland and Bradley, 1961) and clinically the technique often provokes symptoms. This is a key reason for me writing a book on a new system of clinical application of neurodynamics—so that therapists can use a much wider range of techniques, those that slide the nerves gently or take pressure off neural tissues by treating the structures adjacent to the nervous system, and so on (Shacklock, 2005b). To distinguish who might

or might not respond to such techniques is also a key aspect. Clearly, much more can be done to nerves than ‘stretching’ (Elvey and Hall, 1997; Butler, 2000; Shacklock, 2005b). To take the slack out of a nerve gently to the point of ﬂattening out its bands of Fontana, well within its elastic limit, may not always be contraindicated and might offer neurodynamic solutions for some patients. To suggest that nerve movements or elongation techniques should be avoided generally because of the study by Erel et al. (2003) not ﬁnding alterations in longitudinal movement of the median nerve in carpal tunnel syndrome sufferers might have the effect of preventing patients receiving potentially effective treatment, particularly because other studies show reductions in movement of the median nerve in longitudinal and transverse directions (Nakamichi and Tachibana, 1995; Valls-Solle´ et al., 1995; Greening et al., 1999; Hough et al., 2005), and restoration of this movement impairment would seem desirable. The same argument applies to the notion that the production of impulses in nociceptive axons in inﬂamed nerves with the application of tension in rats (Dilley et al., 2005) contraindicates nerve movement techniques in humans. Nociception in rats is not pain in humans. Cleland et al. (2006) showed signiﬁcant improvements in sciatica sufferers with treatment by the slump test which reproduced their clinical pain during treatment and was certainly a ﬁrm technique (held on symptoms for 30 s at a time). Those authors used the term ‘stretch’ but it is not known whether the relevant neural tissues in the subjects tested was actually stretched or the physiology changed, or both. Clinical trials on patients with evidence of upper limb neurogenic pain problems, including carpal tunnel syndrome, have shown good beneﬁts with neural mobilization (Sweeney and Harms, 1996; Rozmaryn et al., 1998). Furthermore, techniques that evoke signiﬁcant amounts of nerve tissue movement can produce improvements in physical signs and symptoms (Allison et al., 2002; Coppieters et al., 2003; Saranga et al., 2003). Also, a clinical case in support of speciﬁc neural techniques is that of a patient who came to our clinic for treatment of her sciatica. The left L4-5 nerve root was swollen (Fig. 1.) and was clearly associated with abnormal neurodynamic test responses (straight leg raise and slump tests) and reduced sensation in her foot. The problem had been present for several weeks and the patient had received various forms of physical treatment prior to attending our clinic, without a change in her symptoms. In seven visits, a complete resolution of her symptoms and neurological changes occurred with neurodynamic techniques that did not involve stretching. This once again highlights the importance of nomenclature in which I proposed the term ‘neurodynamic tests’ rather than ‘neural tension’ or ‘stretch’ in the clinical application and research of the phenomenon we are discussing (Shacklock, 1995a, b). In

ARTICLE IN PRESS Reply to Letter to the Editor / Manual Therapy 12 (2007) e3–e6

Fig. 1. Swollen L4-5 nerve root. CT Scan of the L4-5 segment showing a left swollen nerve root in the intervertebral foramen. The patient reported low back and sciatic pain and reduced sensation in her foot. Her symptoms resolved completely after seven treatments of neurodynamic techniques.

this term, both physiology and mechanical function of the nervous system are considered interdependently. If we localize our terminology to ‘tension’ and ‘stretch’, of course clinical application and research in this area may fail. Likewise, if we localize our thinking to just physiology, the same may occur. As I see it, the research by Erel et al. (2003) is important because it raises some key points: 1. Possibly, manual technique in neurodynamic testing is important in how the nerves move and respond to movement. We have demonstrated that the order and strength with which joint movements are performed affects the symptom response to neurodynamic testing of normal subjects in both upper and lower quarters (Shacklock, 1989; Zorn et al., 1995) and the movement, strain and tension in the neural tissues is affected by this phenomenon of neurodynamic sequencing (Lew et al., 1994; Tsai, 1995, discussed in detail in Shacklock (2005b)). The reason this is important is that, from a research point of view, a negative result on investigation (i.e. no loss of nerve movement) poses huge problems, not the least of which is the potential for lack of sensitivity in the investigation procedure (i.e. false negative). It is therefore vitally important that, before we conclude that the phenomenon does not exist, good evidence shows that the research method used was sufﬁciently sensitive to produce and detect the phenomenon in question. In this case, we are not talking about crosscorrelational image processing. Instead, neurodynamic technique might be a factor. Key questions raised in the study by Erel et al. (2003) are therefore: (a) What is the role of neurodynamic sequencing in neurodynamic effects? (b) All other things being equal, does prior positioning inﬂuence neurodynamics? In my opinion,

e5

it does. For instance, a loose nerve will move differently from a pre-tensioned one. Clearly, much more research on this subject is necessary before deﬁnitive answers are available. 2. Maybe disorders of neurodynamics contain subgroups and it might be necessary to delineate these with testing whose sensitivity has not been fully developed but could be improved upon. 3. The modus operandi of neural techniques may extend beyond improving movement of the nerves. This effectively leads to the statement that maybe nerves need movement for reasons other than loss of sliding capabilities, eg. the effect of nerve movement on neurophysiology. A corollary is therefore, maybe nerves whose movement is not impaired could still beneﬁt from movement due to possible effects on physiology and therefore pain. 4. Neural tissues house features common to many other types of tissue—collagen, elastin, piezoelectric characteristics (Iwasa and Tasaki, 1980; Tasaki and Iwasa, 1982a, b), viscoelasticity, blood ﬂow changes with various stresses (Zochodne and Ho, 1991), to name a few. When it came to the battle last century between resting and moving, movement won hands down and it still has the ascendancy. The next big change has been the application of speciﬁc movements for speciﬁc movement dysfunctions. I am not suggesting that mechanical treatment is appropriate for all forms of neuropathodynamics, but some may likely respond to movement, as long as it is appropriate. Once again, I thank Greening and Leary for their stimulating letter. References Allison G, Nagy B, Hall T. A randomized clinical trial of manual therapy for cervico-brachial pain syndrome—a pilot study. Manual Therapy 2002;7(2):95–102. Butler D. The sensitive nervous system. Adelaide: NOI Press; 2000. Cleland J, Childs J, Palmer J, Eberhart S. Slump stretching in the management of non-radicular low back pain: a pilot clinical trial. Manual Therapy 2006; in press. Coppieters M, Stappaerts K, Staes F. Qualitative assessment of shoulder girdle elevation during the upper limb tension test 1. Manual Therapy 1999;4:33–8. Coppieters M, Stappaerts K, Wouters L, Janssens K. The immediate effects of a cervical lateral glide treatment technique in patients with neurogenic cervicobrachial pain. Journal of Orthopedic and Sports Physical Therapy 2003;33(7):369–78. Denny-Brown D, Doherty M. Effects of transient stretching of peripheral nerve. Archives of Neurology and Psychiatry 1945;54(1):116–29. Dilley A, Lynn B, Pang S. Pressure and stretch mechanosensitivity of peripheral nerve ﬁbres following local inﬂammation of the nerve trunk. Pain 2005;117(3):462–72. Elvey R, Hall T. Neural tissue evaluation and treatment: physical therapy of the shoulder. New York: Churchill Livingstone; 1997.

ARTICLE IN PRESS e6

Reply to Letter to the Editor / Manual Therapy 12 (2007) e3–e6

Erel E, Dilley A, Greening J, Morris V, Cohen B, Lynn B. Longitudinal sliding of the median nerve in patients with carpal tunnel syndrome. Journal of Hand Surgery 2003;28B(5):439–43. Greening J, Smart S, Leary R, Hall-Craggs M, O’Higgins P, Lynn B. Reduced movement of median nerve in carpal tunnel during wrist ﬂexion in patients with non-speciﬁc arm pain. Lancet 1999;354: 217–8. Hall T, Zusman M, Elvey R. Adverse mechanical tension in the nervous system? Analysis of the straight leg raise. Manual Therapy 1998;3(3):140–6. Hough A, Moore A, Jones. M. Restricted excursion of the median nerve in carpal tunnel syndrome. Poster presentation, Manipulation Association of Chartered Physiotherapists conference. Edinburgh, 2005. Iwasa K, Tasaki I. Mechanical changes in squid giant axons. Biomedical and Biophysical Research Communications 1980;95(3): 1338–41. Lew P, Morrow C, Lew A. The effect of neck and leg ﬂexion and their sequence on the lumbar spinal cord: implications for low back pain and sciatic. Spine 1994;19(21):2421–4. Merskey H, Bogduk N. Classiﬁcation of chronic pain: deﬁnitions of chronic pain syndromes and deﬁnitions of pain terms. 2nd ed. Seattle: IASP Press; 1994. p. 212. Nakamichi K, Tachibana S. Restricted motion of the median nerve in carpal tunnel syndrome. Journal of Hand Surgery 1995;20B: 460–4. Rozmaryn L, Dovelle S, Rothman E, Gorman K, Olvey K, Bartko J. Nerve and tendon gliding exercises and the conservative management of carpal tunnel syndrome. Journal of Hand Therapy 1998;11:171–9. Saranga J, Green A, Lewis J, Worsfold C. Effect of a cervical lateral glide on the upper limb neurodynamic test 1: A blinded placebocontrolled investigation. Physiotherapy 2003;89(11):678–84. Shacklock M. The plantarﬂexion inversion straight leg raise. Master of applied science thesis. Adelaide: University of South Australia; 1989. Shacklock M. Neurodynamics. Physiotherapy 1995a;81:9–16. Shacklock M. Clinical application of neurodynamics. In: Shacklock M, editor. Moving in on pain. Sydney: Butterworth-Heinemann; 1995b. p. 123–31.

Shacklock M. Improving application of neurodynamic (neural tension) testing and treatments: a message to researchers and clinicians. Manual Therapy 2005a;10:175–9. Shacklock M. Clinical neurodynamics: a new system of musculoskeletal treatment. Oxford: Elsevier; 2005b. Sweeney J, Harms A. Persistent mechanical allodynia following injury of the hand: treatment through mobilization of the nervous system. Journal of Hand Therapy 1996;9(4):328–38. Sunderland S, Bradley K. Stress–strain phenomena in human peripheral nerve trunks. Brain 1961;84:102–19. Tasaki I, Iwasa K. Rapid pressure changes and surface displacements in the squid giant axon associated with production of action potentials. Japanese Journal of Physiology 1982a;32:69–81. Tasaki I, Iwasa K. Further studies of rapid mechanical changes in squid giant axon associated with action potential production. Japanese Journal of Physiology 1982b;32:505–18. Tsai Y-Y. Tension change in the ulnar nerve by different order of upper limb tension test. Master of Science thesis. Chicago: Northwestern University; 1995. Valls-Solle´ J, Alvarez R, Nun˜ez M. Limited longitudinal sliding of the median nerve in patients with carpal tunnel syndrome. Muscle and Nerve 1995;18:761–7. van der Heide B, Allison G, Zusman M. Pain and muscular responses to a neural tissue provocation test in the upper limb. Journal of Manual Therapy 2001;6(3):154–62. Zochodne D, Ho L. 1991b. Stimulation-induced peripheral nerve hyperemia: mediation by ﬁbers innervating vasa nervorum? Brain Research (12) 1991b; 546 (1): 113–8. Zorn P, Shacklock M, Trott P, Hall R. The effect of sequencing the movements of the upper limb tension test on the area of symptom production. Proceedings of the ninth biennial conference of the Manipulative Physiotherapists’ Association of Australia, 1995. p. 166–7.

(Director) Michael Shacklock NEURODYNAMIC SOLUTIONS (NDS), 6th Floor, 118 King William Street, Adelaide 5000, Australia E-mail address: [email protected]

ARTICLE IN PRESS

Manual Therapy 12 (2007) e7–e8 www.elsevier.com/locate/math

Letter to the Editor Neurodynamics in a broader perspective We read with some dismay the editorial by Shacklock (2005a), and responses by Greening and Leary (2006) and again Shacklock (2006). The Editorial on Neurodynamics (Shacklock, 2005a) was a missed opportunity to position nerve gliding exercises within the modern framework of manual therapy and pain sciences. Furthermore, some of the author’s views on research methodology (such as statements made regarding reviewing the literature, reliability and ethics) conﬂict with basic principles of research and may misguide early career researchers and clinicians. The inference in the editorial (Shacklock, 2005a) that the material by Butler (1991) and Butler and Gifford (1989) created widespread 1880s style nerve stretching is dismissed. In those publications, the term ‘‘stretch’’ was always rejected in favour of ‘‘mobilisation’’ and suggested therapy at the time was guided by Maitland’s ‘‘irritability, severity and nature’’ system. Later publications (Gifford, 1998; Butler, 2000) avoided the term ‘‘tension’’ and supported ‘‘neurodynamics’’. Still, on this note, Greening and Leary’s (2006) use of the word ‘‘stretch’’ is evocative and unfortunate. The conclusion that longitudinal nerve excursion is not restricted in patients with carpal tunnel syndrome (CTS) (Greening and Leary, 2006) based on the ﬁndings by Erel et al. (2003) seems premature. The limited research available on this topic reports conﬂicting ﬁndings (Valls-Sole et al., 1995; Erel et al., 2003; Tuzuner et al., 2004; Hough et al., 2005). More importantly however, whether longitudinal movement is restricted or not in itself should not refrain the clinician from considering nerve gliding exercises for CTS. Restoration of restricted nerve movement is unlikely to be the (main) therapeutic effect of nerve gliding exercises and alternative effects of these exercises should be considered (Coppieters and Butler, 2007). There is a wide variety of exercises available and preliminary analysis demonstrates that much gentler techniques are available than ‘‘stretching techniques’’ (Coppieters and Butler, 2007). Although clinical validation is needed, these techniques may be associated with the beneﬁcial effects of movement and nerve gliding (as

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.01.001

demonstrated by, for example, Rozmaryn et al., 1998), but without the adverse effects of excessive strain. The suggestion that techniques that cause a painful response are invariably undesired (Greening and Leary, 2006) deserves further consideration. The patient’s condition, the pain mechanism in operation and the patient’s understanding of the pain may be such that a (mildly) painful response during, or for a short duration after activity or techniques, may not be problematic and may be the most optimal path to recovery. Inﬂamed, adhered and mechanically challenged peripheral nerves undoubtedly exist. However there is surely a large group of patients where ﬁndings of altered neurodynamic tests are broadly a manifestation of a person’s inability to cope. That is, the nervous system has become peripherally and centrally unregulated due to non-mechanical inputs with resultant perturbations and contributions from the endocrine, autonomic, motor, and immune systems. The nervous system may glide reasonably well but is still sensitive to movement. The work by Erel et al. (2003) could be interpreted to support this contention. This means that responses to structural differentiation may have little to do with mechanical alterations in neural tissue, for example, ankle dorsiﬂexion increasing SLR evoked pain may be just the addition of normal, albeit threatening input to a pain neuromatrix. Responses to neurodynamic sequencing, ﬁrst proposed and practiced by Maitland (1985) though claimed by Shacklock (2005b), may simply be due to the attention placed on the ﬁrst of a combination of movements. In central sensitivity, there is still a place for neural mobilisation, as a sliding technique should allow mechanically non-aggressive large range novel movements which could be therapeutic for central mechanisms on the basis of threat reduction. The notion that education enhancing the patient’s understanding of pain may reduce the sensitivity of neurones (Moseley et al., 2004) and allow better movement has been missed, yet surely this is a part of modern neurodynamics. The role of central and homoeostatic systems in neurodynamics is the real and undiscussed issue for researchers and clinicians. Only then, will discussion on nomenclature, management, and research strategies be on ﬁrm ground.

ARTICLE IN PRESS e8

Letter to the Editor / Manual Therapy 12 (2007) e7–e8

References Butler DS. Mobilisation of the nervous system. Melbourne: Churchill Livingstone; 1991. Butler DS. The sensitive nervous system. Unley: Noigroup Publications; 2000. Butler D, Gifford L. The concept of adverse mechanical tension in the nervous system. Part 2: examination and treatment. Physiotherapy 1989;75(11):629–36. Coppieters MW, Butler DS. Do ‘sliders’ slide and ‘tensioners’ tension? An analysis of neurodynamic techniques and considerations regarding their application. Manual Therapy, 2007, accepted pending minor changes, in press [Epub ahead of print; doi:10. 1016/j.math.2006.12.008]. Erel E, Dilley A, Greening J, Morris V, Cohen B, Lynn B. Longitudinal sliding of the median nerve in patients with carpal tunnel syndrome. Journal of Hand Surgery [Br] 2003;28(5):439–43. Gifford LS. Factors inﬂuencing movement—neurodynamics. In: PittBrooke J, Reid H, Lockwood J, Kerr K, editors. Rehabilitation of movement. London: WB Saunders; 1998. p. 159–95. Greening J, Leary R. Letter-to-the-editor in response to the editorial by Shacklock (doi:10.1016/j.math.2005.03.001). Manual Therapy, 2006 [Epub ahead of print; doi:10.1016/j.math.2006.01.004]. Hough A, Moore A, Jones M. Restricted excursion of the median nerve in carpal tunnel syndrome. In: Proceedings of the second international conference on movement dysfunction, Edinburgh, 2005. Maitland G. The slump test: examination and treatment. The Australian Journal of Physiotherapy 1985;31(6):215–9. Moseley GL, Nicholas MK, Hodges PW. A randomized controlled trial of intensive neurophysiology education in chronic low back pain. Clinical Journal of Pain 2004;20(5):324–30.

Rozmaryn LM, Dovelle S, Rothman ER, Gorman K, Olvey KM, Bartko JJ. Nerve and tendon gliding exercises and the conservative management of carpal tunnel syndrome. Journal of Hand Therapy 1998;11(3):171–9. Shacklock M. Improving application of neurodynamic (neural tension) testing and treatments: a message to researchers and clinicians. Manual Therapy 2005a;10(3):175–9. Shacklock MO. Clinical neurodynamics: a new system of musculoskeletal treatment. Edinburgh: Elsevier Butterworth-Heinemann; 2005b. Shacklock M. Author’s reply to the letter-to-the-editor by Greening and Leary (doi:10.1016/j.math.2006.01.004). Manual Therapy, 2006 [Epub ahead of print; doi:10.1016/j.math.2006.01.005]. Tuzuner S, Ozkaynak S, Acikbas C, Yildirim A. Median nerve excursion during endoscopic carpal tunnel release. Neurosurgery 2004;54(5):1155–60. Valls-Sole J, Alvarez R, Nunez M. Limited longitudinal sliding of the median nerve in patients with carpal tunnel syndrome. Muscle & Nerve 1995;18(7):761–7.

David S. Butler University of South Australia, Neuro Orthopaedic Institute, Australia Michel W. Coppieters Division of Physiotherapy, The University of Queensland, Brisbane, Australia E-mail address: [email protected]

ARTICLE IN PRESS

Manual Therapy 12 (2007) e9 www.elsevier.com/locate/math

Book review Sue Greenhalgh, James Selfe, Red Flags—A Guide to Identifying Serious Pathology of the Spine, Elsevier, Churchill Livingstone, Amsterdam, New York, ISBN 044310140X, 2006 (£19.99, 200pp.). This little and handy book is very innovative, because it summarizes material in a way that has not been published before. The book starts with a general introduction of Red Flags and the way Red Flags are incorporated in several guidelines and in the literature, followed by Chapter 2, concerning clinical reasoning (an important part of the interpretation of potential red ﬂags). A very useful part of this chapter is a weighted Red Flag list, making it very easy to indicate the seriousness of ﬁndings. The structure of the book is constructed in chapters in the same way as a ﬁrst patient contact is built: subjective examination in terms of age, medical history and lifestyle questions (Chapter 3), followed by a chapter regarding questions about the current episode of pain. Then the objective examination is discussed in relation to possible Red Flags, followed by the last Chapter: ‘Conclusion’ (Chapter 6). In this Chapter, the most

doi:10.1016/j.math.2006.05.001

important Red Flags are divided per age-category (this hierarchical list of Red Flags is also available for download), followed by answering the question what to do once you have identiﬁed a number of Red Flags and how suspicious should the clinician be that this patient has a serious spinal pathology. It is illustrated with a clear ﬂow diagram of referral pathways and how to handle them. It should be kept in mind that this book only focuses on Red Flags concerning serious pathology of the spine, and not of the extremities, however, extremities may be less important from this point of view. This is a key book within physiotherapy, in particular for physiotherapists working in countries where there is direct access for patients. Of note is that the book is written for physiotherapists by physiotherapists, with clinical physiotherapy practice as starting point.

Raymond Swinkels Medical Centre Coevering, Geldrop, The Netherlands E-mail address: [email protected]