Manual Therapy Journal - Volume 8, Issue 4, Pages 193-270 (November 2003)

Author: Editors: Ann Moore and Gwen Jull

18 downloads 782 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

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 193–194 r 2003 Elsevier Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00098-5

Editorial

Clinical Guideline Developments

when completed. There was a serious concern that duplication of guidelines might be occurring frequently throughout the world. A call was made for increasing the avenues available for international communication amongst therapists to prevent this duplication and to facilitate more collaboration and sharing of knowledge and effort in guideline development. Discussions were held as to how this sharing of information concerning guidelines could be brought about. The greater use of websites was considered amongst other suggestions. As editors of the Manual Therapy Journal we are committed to the dissemination of any information that will beneﬁt musculoskeletal therapy practice. Therefore we would welcome letters to the editors which publicize new guideline developments and also professional issues which summarize guideline developments in terms of the philosophy underpinning the development of the guideline, the methodological approach employed in their development as well as key messages emanating from the guidelines themselves. The increase in research activity within the professions practicing musculoskeletal therapy will, it is hoped, soon produce an increasing number of research papers which are methodologically sound and which explain clearly what interventions have been used in studies. There are many ambiguities present in the past literature, for example, what does conventional physiotherapy really mean? What are active mobilizations? What is active exercise for low back pain? What does education and advice really consist of? Unless we become very speciﬁc in recording exactly how interventions are carried out in research studies, what they consist of, the rationale for use as well as having the opportunity to include this information in our published articles or on associated journal websites, it will be impossible for guideline developers to distil good from poor practice and effective from ineffective therapeutic activities. It is very easy to prescribe treatments in terms of what are locally/nationally acceptable, but evidence is precious and must convey as much information as possible to ensure that good practice is disseminated as widely as possible.

The rise in the development of clinical guidelines throughout the health disciplines across the world has closely followed the evidence-based practice movement. There are a number of issues inﬂuencing the quality and quantity of guidelines that have been developed in relation to musculoskeletal therapy practice. The number of randomized clinical trials in the ﬁeld of musculoskeletal therapy is increasing yet there is still a comparatively small quantity of evidence available to support many aspects of practice and policy. The methodological quality of the clinical trials underpinning the evidence (measured using gold standard criteria) is very varied. The lack of speciﬁcity and detail in many research papers often does not allow the clear identiﬁcation of the actual intervention/modalities used within the studies and therefore there is the potential for signiﬁcant disagreement between members of guideline development panels about what the evidence is actually saying. The proliferation of guidelines in the same topic areas at local, national and international levels and across professional groups also remains an issue and the probability rather than the possibility of reinventing the wheel is ever present. There is also a need to ﬁrmly place guidelines within the context of current practice frameworks and unless this is done, clinicians will increasingly ﬁnd it more and more difﬁcult to incorporate guidelines into their practice. There is also a need for clinicians to have the opportunity to discuss with their peers, the implementation of guidelines into their practice within their own local contexts. Clinicians also require time to reﬂect on how clinical guidelines may be impacting on their practice. Not least, is the need for clinicians to be aware of the danger of clinical guidelines becoming a substitute for sound and rigorous clinical reasoning processes. At the recent World Congress of Physical Therapy in Barcelona, Spain, round table discussions were held on clinical guideline development. It was interesting to hear the similarity in the concerns raised by delegates from different countries regarding the lack of international notiﬁcation of the commencement of guideline development and/or the wide international publication of individual guidelines 193

ARTICLE IN PRESS 194 Manual Therapy

For guideline developers the next imminent challenge will be keeping up to date with the emergence of new research and new evidence so the need for 2-yearly reviews of current guidelines is highly likely. With the growth in consumer-led services, consumers are being increasingly involved in research planning as a matter of course. Consumers are now usually involved in guideline development. This is important as their input can be very valuable in helping the guideline development teams formulate guidelines which are relevant to consumers’ expectations and their expressed needs in relation to the treatment or management they may experience.

r 2003 Elsevier Ltd. All rights reserved.

Most guideline developers are required to include in published guidelines, a section which explains the guidelines in lay terms for patients which can be displayed in clinical departments, placed on relevant websites or ﬂyers can be produced for dissemination to individual patients. This will mean that our efforts in guideline development can be of positive beneﬁt to clinicians but also importantly of beneﬁt for patients who have increasing knowledge needs. Ann Moore Gwen Jull Co-editors

Manual Therapy (2003) 8(4), 193–194

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 195–206 r 2003 Elsevier Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00094-8

Masterclass

Dynamic evaluation and early management of altered motor control around the shoulder complex M.E. Magarey, M.A. Jones Discipline of Physiotherapy, School of Health Sciences, University of South Australia, Adelaide, Australia

SUMMARY. Altered dynamic control appears to be a signiﬁcant contributing factor to shoulder dysfunction. The shoulder relies primarily on the rotator cuff for dynamic stability through mid-range. Hence, any impairment in the dynamic stabilizing system is likely to have profound effects on the shoulder complex. The rotator cuff appears to function as a deep stabilizer, similar to the transversus abdominus and vastus medialis obliquus, with some evidence of disruption to its stabilizing function in the presence of pain. Similarly, serratus anterior appears to function as a dynamic stabilizer, also demonstrating altered function in painful shoulders. Examination of dynamic control begins with a detailed examination of posture, evaluation of natural movement patterns and functional movements and assessment of the speciﬁc force couples relevant to shoulder function. One useful strategy in management of altered motor control related to these force couples is that of training isolated contraction of the rotator cuff prior to introduction of loaded activity, together with facilitation and training of appropriate scapular muscle force couples – serratus anterior and trapezius, in relation to arm elevation. r 2003 Elsevier Ltd. All rights reserved.

effective in other areas of the body (Richardson & Jull 1995; O’Sullivan et al. 1997a, b, 2000; Richardson et al. 1999). They are based on research on muscle function and control (Hodges & Richardson 1996; O’Sullivan et al. 1997a, b, 2000; Richardson et al. 1999; Cowan et al. 2000, 2001; Shumway-Cook & Woollacott 2001), reports from other experienced clinicians (for example, Wilk & Arrigo 1992; Kibler & Chandler 1994; Wilk 1994; Kibler 1998a, b; Chmielewski & Snyder-Mackler 2001; McConnell 2001) coupled with extensive clinical experience within a framework of sound reﬂective reasoning (Jones et al. 2000) and knowledge of patterns of presentation of shoulder disorders (Magarey 1999). Panjabi’s (1992) now familiar concept of a ‘‘neutral zone’’ for the lumbar spine as a zone in which translatory movements are available can equally be applied to the glenohumeral joint (Hess 2000). The capsulolabral structures (passive restraints) are responsible for setting the limits of passive movement (Jobe 1990; O’Brien et al. 1990; O’Driscoll 1993; Pagnani & Warren 1994) with the muscles, inﬂuenced in their activity by their neural control, responsible for maintaining the humeral head centred in the

INTRODUCTION The focus of this paper is assessment and management of dynamic control of the shoulder complex. The shoulder is a mobile joint that relies heavily for mid-range stability on muscle control (Schenkman & Rugo de Cartaya 1987, 1994; Lippitt & Matsen 1993; Lippitt et al. 1993; Souza 2000; Ciullo 1996; Kibler 1998a; David et al. 2000). Therefore, evaluation of such control and treatment directed at its improvement should form an integral part of management of all shoulder disorders. The programmes suggested are yet to be subjected to the rigours of scientiﬁc evaluation but follow principles demonstrated to be Received: 1 November 2001 Revised: 7 March 2003 Accepted: 4 July 2003 Mary E Magarey Dip Physio, Grad Dip Advanced Manip Therapy, PhD, Mark A Jones, BS(Psychology), RPT, Grad Dip Advanced Manip Therapy, MAppSc (Manipulative Physiotherapy), Discipline of Physiotherapy, School of Health Sciences, University of South Australia, North Terrace, Adelaide, South Australia 5153, Australia. Correspondence to: Tel.: +61-8-8302-2768; Fax: +61-8-83022766. 195

ARTICLE IN PRESS 196 Manual Therapy

glenoid fossa during mid-range movements, thus stiffening the joint (Schenkman & Rugo de Cartaya 1987, 1994; Lippitt & Matsen 1993; Lippitt et al. 1993; Wilk 1994; Burkhart 1996; Ciullo 1996; David et al. 1997, 2000; Kibler 1998a). Any disruption to those mechanisms can lead to abnormal translation of the humeral head during active movement. In relation to the scapula, muscle and neural inﬂuences are very important to its stability as its ligamentous attachments are limited to those of the acromioclavicular joint (Kibler 1998b). The balance of muscle activity within force couples is often more important to normal function than isolated strength of individual muscles (Kibler 1998a, b; Kibler & Chandler 1994). Such balance is determined by the length of muscle and associated fascial tissue and the pattern of recruitment. When tested in isolation in a classic isometric manual muscle test, a muscle may test strongly, but perform poorly during functional activity. Kibler (1998a) used the term, the ‘length-dependent pattern’ of muscle activity to describe cocontraction force couples which operate locally around a joint, controlled by feedback from muscle spindle receptors and responding to perturbations of joint position. The primary function of such force couples is maintenance of joint stability. One key force couple relevant to stability of the glenohumeral joint is that between the lower elements of the rotator cuff – subscapularis anteriorly and infraspinatus/teres minor posteriorly (Saha 1971; Poppen & Walker 1978; Kapandji 1982; Soderberg 1986; Schenkman & Rugo de Cartaya 1987, 1994; Norkin & Levangie 1988; Burkhart 1994, 1996; Wilk 1994) (Fig. 1). These muscles are ideally placed to draw the humeral head into the glenoid and maintain its axis of rotation, so that they can perform their role of concavity compression (Saha 1971; Lippitt & Matsen 1993; Lippitt et al. 1993; Sharkey & Marder 1995; Wuelker et al. 1995, 1998). Failure of function of these muscles in their stabilizing role will lead to creation of an abnormal axis of rotation (Poppen &

Fig. 1—Pictorial representation of the transverse and coronal plane force couples of the rotator cuff, demonstrating the role of the infraspinatus posteriorly and subscapularis anteriorly to draw the humeral head into the glenoid fossa. Reproduced with kind permission of Williams & Wilkins, Baltimore, from Burkhart (1996). r 2003 Elsevier Ltd. All rights reserved.

Walker 1978; Howell & Galinat 1987; Schenkman & Rugo de Cartaya 1987, 1994; Howell et al. 1988; Souza 2000) and abnormal translation of the head of humerus (Burkhart 1994, 1996). In the scapulothoracic area, the force couples associated with movement overhead alter through range, as the axis of rotation changes with increasing elevation and plane of movement (Inman et al. 1944; Poppen & Walker 1978; Dvir & Berme 1978; Schenkman & Rugo de Cartaya 1987, 1994; Bagg & Forrest 1988; Culham & Laprade 2000; Abelew 2001), but the primary contributors are serratus anterior and trapezius (Inman et al. 1944; Basmajian 1963; Kapandji 1982; Bagg & Forrest 1986, 1988; Schenkman & Rugo de Cartaya 1987, 1994; Norkin & Levangie 1988; Souza 2000). In the early part of range, when the axis of rotation is at the root of the scapular spine, the principal rotators are the upper ﬁbres of both serratus anterior and trapezius (Basmajian 1963), whereas when the axis of rotation moves towards the acromioclavicular joint, the relative contribution of upper trapezius lessens while that of lower trapezius increases, together with the lower ﬁbres of serratus anterior (Basmajian 1963; Schenkman & Rugo de Cartaya 1987, 1994) (Fig. 2A, B). Serratus anterior is, therefore, a signiﬁcant component of the force couple throughout range (Bagg & Forrest 1986). David et al. (2000) demonstrated consistent activation of the rotator cuff prior to the more superﬁcial delto-pectoral muscles during isokinetic rotation in normal shoulders, conﬁrming their role as dynamic stabilizers for the glenohumeral joint. Similarly, analysis of activation during rotation in the normal shoulder revealed that at least one component of the antagonist rotator cuff was always active (David et al. 2000), providing evidence of their stabilizing role. Strong evidence is available that pain alters the timing of contraction in stabilizing muscles – transversus abdominis and multiﬁdus in relation to the lumbar spine (Hides et al. 1994, 1996; Richardson & Jull 1994, 1995; Hodges & Richardson 1996, 1998; Hodges et al. 1996; O’Sullivan et al. 1997a, b, 2000), vastus medialis obliquus in relation to the knee (Cowan et al. 2001). Preliminary continuation of our shoulder stabilization research (David et al. 2000) with unstable shoulders has shown widely differing patterns of onset of muscle activity, with failure of the rotator cuff and biceps to be activated prior to the delto-pectoral group and, in some instances, failure to ﬁre until after the onset of movement – thus demonstrating a similar disruption to stabilizing function as found in the knee and lumbar spine. Kibler (1998b) considered that serratus anterior and lower trapezius are susceptible to inhibition in painful shoulders. This inhibition is seen early as a non-speciﬁc response to any painful condition in the shoulder, presenting as a disorganization of the Manual Therapy (2003) 8(4), 195–206

ARTICLE IN PRESS Altered motor control around the shoulder complex 197

Fig. 2—Force couples around the scapula relevant in arm elevation (Adapted from Bagg and Forrest 1986; Kapandji 1982). Depicted is the most common pattern of muscle recruitment reported by Bagg and Forrest (1986). (A) In the ﬁrst 601, the axis of rotation of the scapula is situated at the root of the spine of the scapula. Primary muscles involved in upward rotation of the scapula are lower ﬁbres of serratus anterior and upper trapezius, working via the clavicle, with lower and middle trapezius functioning eccentrically to control the movement. In this range, muscle function is highly variable. (B) In the next 601, the axis of rotation begins to move along the spine of the scapula towards the acromioclavicular joint. This means that the emphasis of contribution of the muscles changes, with the ﬁbres of lower trapezius now becoming more actively involved in upward rotation, along with those of lower serratus anterior and upper trapezius. (C) By the time the arm reaches 1201 of elevation, the axis of rotation is at the acromioclavicular joint. Upper trapezius is no longer positioned to be able to function to upwardly rotate the scapula, whereas lower trapezius is now ideally situated to perform this function, in conjunction with lower serratus anterior. (D) In the ﬁnal stages of elevation, lower trapezius and lower serratus anterior are the primary rotators of the scapula, with upper trapezius functioning to rotate the clavicle and middle trapezius working eccentrically to control the degree of upward rotation.

normal ﬁring pattern and a decreased ability to produce torque and to stabilize the scapula, a phenomenon Kibler (1998b) described as ‘scapular dyskinesis’. Manual Therapy (2003) 8(4), 195–206

Alteration in activity of serratus anterior in the painful shoulders of swimmers and throwers has been found when compared to that of non-painful athletes (Glousman et al. 1988; Scovazzo et al. 1991; Pink r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 198 Manual Therapy

et al. 1993). Wadsworth and Bullock-Saxton (1997) found signiﬁcant delay in activation of serratus anterior in the painful shoulders of swimmers compared with non-painful shoulders, with little change in timing of activation of trapezius. All these studies highlight serratus anterior as the primary stabilizer of the scapulothoracic region, functioning in a manner similar to other deep stabilizers. The movement of elevation has been used as an example of the need to consider force couples. Clearly, different considerations must be made if the disorder involves other movements and loads. Adduction against resistance needs to be considered in conjunction with elevation in the throwing or swimming athlete, for example. This function is well described in Schenkman and Rugo de Cartaya (1994) and Souza (2000).

PRINCIPLES OF A DYNAMIC ASSESSMENT Patients move in a variety of ways, with a wide range of what can be called ‘normal’ (Shumway-Cook & Woollacott 2001). Inﬂuences on movement patterns include avoidance of pain, general health and mood, relative length of tissues, strength and level of activity of muscles and timing of contraction of those muscles. Functional demands and habitual activities also contribute to development of particular movement patterns (Shumway-Cook & Woollacott 2001). All these factors must be considered during examination of muscle function around the shoulder. Antalgic movement patterns are familiar features of physical examination – a classic example is the arm that drifts towards the plane of the scapula during frontal plane abduction, with prevention of this movement causing pain. A patient who is unwell or depressed often demonstrates a hunched posture with slow, heavy movements. Such a posture, if habitual, could lead to learned poor scapular rotation during arm elevation, with the potential for development of a subacromial impingement. The concept of relative ﬂexibility or relative stiffness should be familiar to physiotherapists (Sahrmann 2001). Movement occurs in the path offering the least resistance, such that compensation for a tightened tissue or restricted joint occurs with movement in a different plane or of a different body part. A weakened muscle will also disrupt a normal movement pattern as its weakness must be compensated for by an altered pattern of activity in a substitute muscle capable of achieving similar action. Altered timing of contraction, as discussed above, inﬂuences movement patterns, such that either the torque producing muscles tend to be activated without pre-setting by the stabilizers (Hodges & Richardson 1996, 1998; David et al. 1997; Cowan et al. 2001) or the nature of activity of the stabilizers r 2003 Elsevier Ltd. All rights reserved.

is converted from one which is direction-independent to one which becomes direction-speciﬁc (Hodges & Richardson 1997). To achieve a successful outcome from any dynamic stabilization programme, rehabilitation must be centred on the patient’s abilities rather than impairments. The starting point for progression must be correct movement patterns and muscle recruitment. Training in an incorrect pattern will only reinforce the pattern. Therefore, the assessment of muscle function must reveal the patient’s abilities in addition to impairments. As a simple example, if lower trapezius is tested in a standard manual muscle test position (Kendall et al. 1993) and found to be weak, the position in which the muscle is tested, or the load placed on the muscle during testing, must be incrementally reduced to a stage where the contraction can be initiated and maintained successfully. The point from which to start re-training of lower trapezius, if appropriate, is that where the contraction can be successfully achieved. Each of the tests described below is based on this principle. Gentile (1992) advocated that goal-directed functional behaviour should be analysed at three different levels: the action itself, the movements that are incorporated in that action and the neuromotor processes that drive the movements – for example, the integrity of the motor and sensory systems.

DYNAMIC EXAMINATION Knowledge of, and the indications for, inclusion of speciﬁc muscle length (Evjenth & Hamberg 1980) and isolated strength tests (Kendall et al. 1993) is assumed by the reader. In this paper, those components and techniques that we have found particularly useful during dynamic examination will be discussed. Observation of posture Altered joint position such that some muscles appear tight or overactive and others lengthened or underactive provides early hypotheses in relation to muscle function. Observation of the posture of the whole body is an integral component of postural assessment of the upper quarter. This should occur in the context of the patient’s functional demands, so that it includes evaluation of routine postures used by the patient. Lower quarter muscle development and spinal posture can indicate whether whole body integrated movement patterns are adequate for normal upper quarter muscle function. Cervicothoracic posture has considerable inﬂuence on scapular position and mobility and therefore, also glenohumeral mobility (Crawford & Jull 1991; Culham & Peat 1993; Solem-Bertoff et al. 1993). Speciﬁc analysis of scapular and arm position will Manual Therapy (2003) 8(4), 195–206

ARTICLE IN PRESS Altered motor control around the shoulder complex 199

then provide initial clues to the comparative load carried by the glenohumeral and scapulothoracic joints. Finally, speciﬁc analysis of contour and tone of all relevant muscle groups should be made. Analysis of movement patterns Particular emphasis is placed on detailed visual analysis of spontaneous movement patterns. Speciﬁc attention is given to recruitment of particular muscle groups related to each movement, visual assessment of the timing of that recruitment and the relative contribution of all body parts to the movement. An important part of normal function is the ability to dissociate different body parts during movement. Our clinical experience has shown us that the inability to dissociate trunk from scapular movement, for example, is often a signiﬁcant contributor to shoulder dysfunction. In the same way, poor trunk and pelvic stability place considerable stress on the upper quarter during loaded or rapid functional activities (Kibler 1998a,b). Control of the movement, both concentric and eccentric, is also evaluated, at speeds relevant to the patient’s presentation. Similarly, if symptoms are only provoked after repetition or under load, these components are included. Repetition of arm elevation while holding a small weight may demonstrate altered movement patterns not detected with a single unloaded movement. Careful attention is placed on detection of substitution strategies and provocation of symptoms. Any asymmetries found are corrected actively, if possible, and passively to evaluate their effect on symptom production and performance of the movement. Active correction provides some insight to the patient’s awareness of postural or movement impairment and the appropriate motor strategy to correct it. Postural correction or facilitation of a more normal muscle activation that alters pain on movement provides a positive indication of a relationship between pain and movement and of the potential beneﬁt of a dynamic rehabilitation programme. Speciﬁc evaluation of relevant force couples In an initial examination, more importance is placed by the authors on evaluating the force couples relevant to stabilization of the shoulder complex than on isolated manual muscle testing as knowledge of the more subtle stabilizing capability of the shoulder complex allows better interpretation of the results of the more superﬁcial muscle strength assessments. Testing of more global muscle function – termed the ‘force–dependent’ patterns by Kibler (1998a) – tends to be addressed at later sessions. Manual Therapy (2003) 8(4), 195–206

Rotator cuff The two tests that we have developed and used over a number of years to determine dynamic control of the head of humerus in the glenoid are the dynamic rotary stability test and the dynamic relocation test. 1. The Dynamic Rotary Stability Test (DRST). The DRST is used to evaluate the ability of the rotator cuff to maintain the normal centring of the humeral head in the glenoid when loaded through rotation (Howell & Galinat 1987; Howell et al. 1988). In a frankly unstable shoulder or one in which rotator cuff dynamic control is lacking, the humeral head can be felt to translate when the rotator cuff is loaded. In more subtle situations, or where the instability is more functional than structural, provocation of symptoms, alteration in the quality of contraction, clicking/clunking and compensation by other muscle groups are often noted, without the sensation of humeral head translation. The DRST is undertaken in different parts of the range of glenohumeral ﬂexion and abduction from neutral towards the functional position(s) in which the patient has symptoms, whether pain, weakness, apprehension or instability (Fig. 3). The aim is to ﬁnd the position(s) in range where the patient has control of the head of humerus as close as possible to the position in which control is lost. The test is performed isometrically, isotonically, concentrically and eccentrically at different speeds and under different loads. The amount of resistance added is usually light to moderate, as the assessment is one of the ability to stabilize, rather than one of strength of rotation. 2. The Dynamic Relocation Test (DRT). The DRT is a test of the ability of the transverse force couple of the rotator cuff to stabilize the head of humerus in the glenoid against a de-stabilizing load. It is predicated on the knowledge that, in normal situations, the rotator cuff functions in some degree of cocontraction to stabilize the glenohumeral joint during dynamic function and this activation precedes that of the more superﬁcial torque producing muscles (David et al. 1997, 2000). Co-contraction stiffens a joint and is an important feature of early stages of skill acquisition (Shumway-Cook & Woollacott 2001). In patients with shoulder pain, the co-contraction and pre-setting appears to be lost. Once the ability to isolate the co-contraction has been determined in the optimal position (Fig. 4), maintenance of this isolated contraction can be evaluated in different positions and during different tasks. Patients with a dysfunctional shoulder may ﬁnd isolation of this contraction to the rotator cuff difﬁcult without considerable facilitation and practice. Once the patient can master the relocation contraction, the ability to maintain it during arm movement is evaluated, using any relevant functional movement, with progressively increasing difﬁculty. If r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 200 Manual Therapy

Fig. 3—(A, B) Dynamic rotary stability test, demonstrating two different positions in which humeral head control can be evaluated. The operator’s left hand is placed over the humeral head so that he is able to detect any translation that occurs during contraction of the rotators. In the example shown, he is resisting isometric lateral rotation in mid-range and near end-range, in a position functionally relevant for a thrower.

abnormality were detected in the DRST, the test position in which loss of control was found can be reevaluated with facilitation of dynamic relocation. If control is improved, dynamic rehabilitation has a good chance of success. The speciﬁc techniques for these tests are outlined elsewhere (Magarey & Jones, 2003). To date, research on the reliability and validity and on establishing normative values for these tests is incomplete. Scapular stabilization and movement

Fig. 4—Dynamic relocation test. The patient is positioned such that his upper arm is in approximately 60–801 of abduction in the scapular plane, as this position optimizes the line of pull of the lower elements of the rotator cuff. The middle ﬁnger of the operator’s left hand is placed over the belly of subscapularis, so that he is able to detect activity in this muscle during the test. He is also able to feel activation of the more superﬁcial muscles at the same time. Alternatively, the operator may palpate subscapularis from a posterior approach – particularly useful if the patient has a tendency to over-activate the pectorals. With his right hand, the operator applies a very gentle longitudinal distraction force to the arm and asks the patient to draw his arm into the socket, while he feels for activation of subscapularis, remembering that this occurs in conjunction with activation of infraspinatus/teres minor. r 2003 Elsevier Ltd. All rights reserved.

Scapular stabilizing and movement function is evaluated with two methods, using weightbearing assessment of scapular control and with modiﬁed PNF diagonal patterns in isolation from and in conjunction with arm movements. 1. Weightbearing assessment of scapular control. Weightbearing assessment allows evaluation of a number of factors. In particular, it is a useful position for testing dissociation of spinal movement from scapular movement and lumbar from thoracic movement, in addition to scapular control. Although dissociation can be evaluated in many different Manual Therapy (2003) 8(4), 195–206

ARTICLE IN PRESS Altered motor control around the shoulder complex 201

Fig. 5—Assessment of scapular function in four point kneeling. (A) The standard starting position, emphasizing spinal neutral position. (B) A more challenging position for the subject’s right scapular region, in one arm weightbearing.

positions, the steps to evaluate this ability are integral to the scapular assessment and are therefore included. The standard starting position for weightbearing assessment is four point kneeling, although assessment should be undertaken in multiple different positions, as a wide variation appears to exist in people’s ability to function in weightbearing. The positions used include leaning against a wall or table, four point kneeling, prone on elbows and weightbearing in the frontal and scapular plane. In the frontal plane, the contribution of the trapezius components of the force couple may be stressed more than serratus anterior as a result of the lesser protraction component (Inman et al. 1944; Schenkman & Rugo de Cartaya 1994). In four point kneeling, the patient’s ability to dissociate pelvic from lumbar, lumbar from thoracic movement and thoracic from cervical movement is evaluated ﬁrst. Spinal dissociation and awareness of natural posture will facilitate success in scapular control re-training. The next step is to determine whether the patient can protract and retract the scapulae without concurrent spinal movement (Fig. 5). If this can be achieved, the scapula’s holding ability in neutral (mid-range) protraction is then assessed through different stages and types of Manual Therapy (2003) 8(4), 195–206

loading. If loading in this position fails to demonstrate any impairment, the assessment can be progressed to more challenging positions or demands. Equally, if scapular control is not adequate, positions which require less weightbearing or cognitive load should be evaluated. 2. Scapular diagonal patterns. One method in which to assess functional muscle performance is through the use of the PNF patterns from glenohumeral extension/abduction/medial rotation to elevation/adduction/lateral rotation (D1) and from extension/adduction/medial rotation to elevation/ abduction/lateral rotation (D2) (Voss et al. 1953; Knott & Voss 1968; Engle 1994). During these movements, the scapula moves from retraction/ depression/downward rotation to protraction/elevation/upward rotation and from protraction/depression/downward rotation to retraction/elevation/ upward rotation, respectively. Having ﬁrst determined that the relevant range is available passively, an initial assessment of the patient’s ability to perform these scapular patterns independent of arm movement is undertaken. Without inclusion of the arm, the rotation component of the scapular movement is minimal, but the resultant diagonal movements are regularly dysfunctional with a painful shoulder. This may simply relate to lack of familiarity with the movement, so inclusion of stimulation with passive, active assisted and resisted movement through the patterns is used to determine whether this is the case (Fig. 6A,B). If so, repeat assessment of unassisted active scapular diagonal movement is signiﬁcantly improved, whereas in the impaired shoulder, such improvement is not immediately evident (Fig. 7). Often, patients and non-symptomatic individuals will be biased in their un-loaded scapular diagonal patterns, possibly reﬂecting learned movements. For example, physiotherapists and keyboard operators frequently present with an exaggerated protraction component at the expense of shoulder elevation in the D1 pattern. Finally, the scapula’s ability to rotate upwardly during arm elevation is evaluated, using similar principles to those described for the scapular patterns. With this assessment, the emphasis is on the retraction/downward rotation to protraction/ upward rotation component with less emphasis on depression/elevation (Fig. 8). In most instances, the authors have found that, at initial dynamic assessment, these tests, coupled with appropriate muscle length assessment, are all that need to be included for the upper quarter. Speciﬁc evaluation of isolated muscle strength may be appropriate, particularly for the athletic population, but frequently, because of impairment in the stabilizing force couples, such evaluation is withheld until stabilization is improved. However, evaluation of r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 202 Manual Therapy

Fig. 6—(A) D1 scapular PNF pattern – depression/retraction component. (B) D1 scapular PNF pattern – elevation/protraction component.

Fig. 7—(A) D2 scapular PNF pattern – depression/protraction component. (B) D2 scapular PNF pattern – elevation/retraction component.

trunk control, either the ability to isolate the deep stabilizers, as described by Richardson et al. (1999) and control of pelvic and hip muscle function, or through control of neutral and later out-of-neutral postures and movements should be included, as upper quarter stability requires a strong stable base on which to work (Fleisig et al. 1994; Kibler 1998a,b).

MANAGEMENT OF MUSCLE CONTROL DYSFUNCTION OF THE SHOULDER COMPLEX From the examination ﬁndings, a management plan can be made, addressing those aspects of each part of the examination found to be impaired and ensuring maintenance of an appropriate balance between function of the scapulothoracic and scapulohumeral muscles. Training for control of one region must not occur at the expense of the other. Similarly, training for either the glenohumeral joint or scapulothoracic region must be implemented in positions of total body control and stability. In this paper, the dominant features associated with the early stages of rehabilitation are addressed. Our approach to management of muscle control impairment of the shoulder complex follows similar r 2003 Elsevier Ltd. All rights reserved.

Fig. 8—Facilitation of scapular upward rotation. Resistance is provided to protraction through the patient’s hand between the therapist’s upper arm and trunk and to upward rotation through the lateral aspect of the scapula and through resisted extension of the arm. The technique may be performed with the resistance to scapular movement only if resisted arm movement provokes pain.

principles to those outlined in other dynamic stabilization programmes (for example, O’Sullivan et al. 1997a, b, 2000; Richardson et al. 1999; Comerford & Mottram 2001; Sahrmann 2001). Progression through the programmes is considered Manual Therapy (2003) 8(4), 195–206

ARTICLE IN PRESS Altered motor control around the shoulder complex 203

in terms of attainment of speciﬁc skills and control – a criteria-based protocol (Chmielewski & SnyderMackler 2001) – rather than length of time, as in many shoulder rehabilitation programmes (for example, Kunkel & Hawkins 1994; Loeb et al. 1994; Souza 2000; O’Brien et al. 1994; Scarpinato & Andrews 1994; Timmerman et al. 1994; Ciullo 1996). Our approach also works on the principle of speciﬁcity of muscle function and the importance of functional relevance for transfer of training (Kibler & Chandler 1994; Kibler 1998a; Shumway-Cook & Woollacott 2001) so that, as soon as possible, any re-training occurs in positions relevant to the patient’s habitual activities. The authors use the concept of re-training by breaking function into interim steps (Schmidt 1991; Winstein 1991). One goal of any motor control rehabilitation is to gain awareness of, and the ability to, activate the deep stabilizers of the region prior to activation of the, usually, more superﬁcial torque producing muscles and to maintain that activation during activity. Another is retraining of optimal movement patterns. Both involve motor programme retraining and therefore, reﬁned, controlled activation of the deep stabilizing force couples, using either strategies of isolation or controlled posture or movement facilitated by imagery. Activation of isolated muscles is often difﬁcult to conceptualize. Therefore, patient explanation of the reasons for the programme and the processes required become important aspects of the management. Similarly, imagery can facilitate understanding of the action required. Without the patient’s understanding and collaboration in the process, it is doomed to failure, as perseverance, even when there is little obvious initial change, is essential to success. Pain inhibition appears to have a powerful inﬂuence on the motor system (Richardson 1987; Hodges & Richardson 1996, 1998; Cowan et al 2001), so palliative treatment may be necessary to decrease pain in the early stages. However, there is no need to wait for pain to settle before starting a motor control programme within a pain-free range at a load that does not provoke symptoms, as often, restoration of control acts as a potent pain inhibitor. Retraining motor programming, or the neural control in Panjabi’s (1992) model, is dependent on motor learning. Motor learning involves learning new strategies for sensing as well as moving, arising from a complex of perception— cognition—action processes (Shumway-Cook & Woollacott 2001). Motor learning can be enhanced by the use of mental imagery, tactile, verbal, visual, taping, weightbearing and movement oriented cues – different cues are effective with different people. Initially, facilitation is undertaken in an optimal position for the relevant muscles, usually mid-range. Manual Therapy (2003) 8(4), 195–206

Frequent stimulation and repetition improve awareness and the ability to activate far more than an isolated exercise session once a day (Catalano & Kleiner 1984; Shumway-Cook & Woollacott 2001). Therefore, while learning the activation technique, we encourage patients to practise it for a few minutes several times a day. Initially, each region is trained in isolation – that is, the rotator cuff is worked with the scapula in an unchallenged position and vice versa. The muscles are worked in co-contraction in their relevant force couples, with the contractions initially isometric and isotonic with low load, with a gradual build up and release. Once dynamic stability is established, the positions in which control is emphasized are determined by the examination ﬁndings and functional needs of the patient. During the DRST, for example, the patient may have demonstrated good control up to 601 of abduction in isotonic external rotation, while isometric control may have been satisfactory to 1201. If these positions are re-tested with pre-setting of the rotator cuff via the DST manoeuvre, better control may be found. Training should then be started in those positions in which the patient has control, but as close to the position where that control is lost as possible. Isometric and isotonic training can be undertaken concurrently, as long as the patient is aware of the different sensations associated with control and lack of control. Teaching this difference in feel may be time consuming initially, but is essential to success of the programme, as training in a position in which control is lacking may reinforce poor movement patterns. As control is mastered, the load can be increased cognitively by asking the patient to maintain control in one area and work on the other. Once activation can be achieved in an isometric, stable situation, we encourage the patient to incorporate the activation into simple tasks of daily living. Deliberate activation of the rotator cuff in the DRT manoeuvre while waiting at trafﬁc lights in a car, or prior to reaching to answer the telephone; setting of the scapulae in an optimal position while at a computer or before reaching into a cupboard are examples of cognitive challenge. When such tasks are mastered, physical load, speed and more complicated, integrated tasks can be progressively added. Progression is made with any particular exercise only when the step before is mastered. The more times the technique is repeated and the more different situations in which it is repeated, the quicker the patient is likely to master it. While this intensive training is underway for the upper quarter, any more general impairment in motor control should also be addressed. Issues such as poor dissociation may need to be improved before scapular work can be initiated and transversus abdominis and gluteal control can be incorporated and progressed from the ﬁrst day of treatment. If r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 204 Manual Therapy

control is good, but general strengthening is indicated, this can be undertaken in conjunction with the upper quarter control programme. Progression in each region may be at different rates, depending on the degree of impairment, so each component must be regularly checked and progressed individually. Once this level is mastered, function is re-evaluated to determine the need for further progression. For many non-athletic patients, we have found that addressing the motor control issues and teaching the patient appropriate strategies to continue monitoring and stimulating the control are sufﬁcient for a return to normal function. However, in the manual labourer or overhead athlete, further progression is needed. Whether rehabilitation is ceased at this point or continued, the authors have found that the patient who has had pain in the shoulder needs to continually monitor their dynamic control and practise it regularly, or the control tends to be lost. A brief self-directed activation session once or twice a week or deliberate pre-activation during normal function is sufﬁcient to maintain the control once mastered to this level.

CONCLUSION In this paper, we have presented an approach to dynamic evaluation and management of the shoulder complex that we use in conjunction with detailed passive examination and management as indicated for each patient. The ideas presented here represent one set of strategies that we have found to improve shoulder function. A deliberate setting in neutral rather than speciﬁc pre-activation can also be effective with different patients. Interestingly, a number of researchers have demonstrated that stable movement patterns become more unstable just prior to a transition to a new movement pattern in both adults and children (Kelso & Tuller 1984; Gordon 1987; Woollacott & Shumway-Cook 1990). Therapists should be aware of this possibility when reassessing patients undergoing a dynamic rehabilitation programme such as the one suggested. It is important that patients be made aware of this possibility so that they do not become discouraged through this phase of their training. While the dynamic component of our management approach has not been challenged through clinical trial, it is based on similar principles to those demonstrated as effective in the cervical and lumbar spine regions (O’Sullivan et al. 1997a,b, 2000; Jull 2000; Jull et al. 2002). If the whole process is linked by sound reﬂective clinical reasoning (Jones et al. 2000), the optimal balance of passive and dynamic management will be included. r 2003 Elsevier Ltd. All rights reserved.

References Abelew T 2001 Kinesiology of the shoulder. In: Tovin BJ, Greenﬁeld BH (eds) Evaluation and Treatment of the Shoulder: An Integration of the Guide to Physical Therapist Practice. FA Davis Co, Philadelphia, Ch 2, pp 25–44 Bagg SD, Forrest WJ 1986 Electromyographic study of the scapular rotators during arm abduction in the scapular plane. American Journal of Physical Medicine & Rehabilitation 65(3): 111–124 Bagg SD, Forrest WJ 1988 A biomechanical analysis of scapular rotation during arm abduction in the scapular plane. American Journal of Physical Medicine and Rehabilitation 67(6): 238–245 Basmajian JV 1963 The surgical anatomy and function of the armtrunk mechanism. Surgical Clinics of North America 43: 1471 Burkhart SS 1994 Reconciling the paradox of rotator cuff repair versus debridement: A uniﬁed biomechanical rationale for the treatment of rotator cuff tears. Journal of Arthroscopic and Related Surgery 10(1): 4–19 Burkhart SS 1996 A uniﬁed biomechanical rationale for the treatment of rotator cuff tears: Debridement versus repair. In: Burkhead WZ (ed.) Rotator Cuff Disorders. Williams and Wilkins, Baltimore, Ch 21, pp 293–312 Catalano JF, Kleiner BM (1984) Distant transfer and practice variability. Perceptual and Motor Skills 58: 851–856 Chmielewski TL, Snyder-Mackler L 2001 Therapeutic exercise and functional progression of the shoulder. In: Tovin BJ, Greenﬁeld BH (eds) Evaluation and treatment of the shoulder. An integration of the guide to physical therapist practice. PFA Davis Co, Philadelphia, Ch 16, p 379–398 Ciullo JV 1996. Shoulder Injuries in Sport. Human Kinetics, Champaign Comerford MJ, Mottram SL 2001 Functional stability re-training: principles and strategies for managing mechanical dysfunction. Manual Therapy 6(1): 3–14, doi:10.1054/math.2000.0389 Cowan SM, Bennell KL, Hodges PW 2000 The test-retest reliability of the onset of concentric and eccentric vastus medialis obliquus and vastus lateralis electromyographic activity in a stair stepping task. Physical Therapy in Sport 1: 129–136 Cowan SM, Bennell KL, Hodges PW, Crossley KM, McConnell J 2001 Delayed onset of electromyographic activity of vastus medialis obliquus relative to vastus lateralis in subjects with patellofemoral pain syndrome. Archives of Physical Medicine and Rehabilitation 82: 183–189 Crawford HJ, Jull GA 1991 The inﬂuence of thoracic form and movement on range of shoulder ﬂexion. Physiotherapy, Theory and Practice 9: 143–148 Culham E, Laprade J 2000 Biomechanics of the shoulder complex. In: Dvir Z (ed.) Clinical Biomechanics. Churchill Livingstone, London, Ch 6, pp 141–164 Culham E, Peat M 1993 Functional anatomy of the shoulder complex. Journal of Orthopaedic and Sports Physical Therapy 18(1): 342–350 David G, Jones MA, Magarey ME, Sharpe MH, Dvir Z 1997 Rotator cuff muscle performance during glenohumeral joint rotations: an isokinetic, electromyographic and ultrasonographic study. Tenth Biennial Conference, Manipulative Physiotherapists Association of Australia, Melbourne David G, Magarey M, Jones M, Turker . K, Sharpe M, Dvir Z 2000 EMG and strength correlates of selected shoulder muscles during rotations of the glenohumeral joint. Journal of Clinical Biomechanics 2: 95–102 Dvir Z, Berme N 1978 The shoulder complex in elevation of the arm: A mechanism approach. Journal of Biomechanics 11:219 Engle RP 1994 Proprioceptive neuromuscular facilitation for the shoulder. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, Ch 38, p 451 Evjenth O, Hamberg J 1980 Muscle Stretching in Manual Therapy. A Clinical Manual. Vol. 1: The Extremeties. Alfta Rehab Frlag, Alfta, Part 2, pp 13–48 Fleisig GS, Dillman CJ, Andrews JR 1994 Biomechanics of the shoulder during throwing. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, Ch 31, p 355–368 Manual Therapy (2003) 8(4), 195–206

ARTICLE IN PRESS Altered motor control around the shoulder complex 205

Gentile A 1992 The nature of skill acquisition: Therapeutic implications for children with movement disorders. In: Forssberg H, Hirschfeld H (eds) Movement Disorders in Children. Medical Sport Science, Karger, Basel, pp 31–40 Glousman R, Jobe F, Tibone J, Moynes D, Antonelli D, Perry J. 1988 Dynamic electromyographic analysis of the throwing shoulder with glenohumeral instability. Journal of Bone and Joint Surgery 70A(2): 220–226 Gordon J (1987) Assumptions underlying physical therapy intervention: Theoretical and historical perspectives. In: Carr JH, Shepherd RB, Gordon J, Gentile AM, Held JM (eds) Movement Sciences: Foundations for Physical Therapy in Rehabilitation. Aspen, Rockville, MD, pp 1–30 Hess SA 2000 Functional stability of the glenohumeral joint. Manual Therapy 5(2): 63–71, doi:10.1054/math.2000.0241 Hides JA, Richardson CA, Jull GA 1996 Multiﬁdus muscle recovery is not automatic following resolution of acute ﬁrst episode low back pain. Spine 21: 2763–2769 Hides JA, Stokes MJ, Saide M, Jull GA, Cooper DH 1994 Evidence of lumbar multiﬁdus muscle wasting ipsilateral to symptoms in patients with acute/subacute low back pain. Spine 19: 165–172 Hodges PW, Richardson CA 1996 Inefﬁcient muscular stabilization of the lumbar spine associated with low back pain: A motor control evaluation of transversus abdominis. Spine 21: 2640–2650 Hodges PW, Richardson CA 1997 Feedforward contraction of transversus abdominis is not inﬂuenced by the direction of arm movement. Experimental Brain Research 114: 362–370 Hodges PW, Richardson CA 1998 Delayed postural contraction of transversus abdominis in low back pain associated with movement of the lower limbs. Journal of Spinal Disorders 11: 46–56 Hodges PW, Richardson CA, Jull GA 1996 Evaluation of the relationship between the ﬁndings of a laboratory and clinical test of transversus abdominis function. Physiotherapy Research International 1: 30–40 Howell SM, Galinat BJ 1987 The containment mechanism: The primary stabilizer of the glenohumeral joint. Paper read at Annual Meeting of American Academy of Orthopaedic Surgeons, San Francisco, January 23 Howell SM, Galinat BJ, Renzi AJ, Marone PJ 1988 Normal and abnormal mechanics of the glenohumeral joint in the horizontal plane. Journal of Bone and Joint Surgery 70A(2): 227–232 Inman VT, Saunders JB de CM, Abbott LC 1944 Observations on the function of the shoulder joint. Journal of Bone and Joint Surgery 26A: 1–30 Jobe CM 1990 Gross anatomy of the shoulder. In: Rockwood CA, Matsen FA (eds) The Shoulder. WB Saunders, Philadelphia, Ch 2, pp 34–98 Jones MA, Jensen G, Edwards I 2000 Clinical reasoning in physiotherapy. In: Higgs J, Jones MA (eds) Clinical Reasoning in the Health Professions, 2nd edn. Butterworth Heinemann, Oxford, Ch 12, pp 117–127 Jull GA 2000 The physiotherapy management of cervicogenic headache: A randomised controlled trial. In: Singer K (ed.) Abstracts of the 7th Scientiﬁc Conference of the International Federation of Orthopaedic Manipulative Therapists in conjunction with the Biennial Conference of the Manipulative Physiotherapists Association of Australia, Perth, p 87 Jull G, Trott P, Potter H, Zito G, Niere K, Shirley D, Emberson J, Marschner I, Richardson C 2002 A randomised controlled trial of exercise and manipulative therapy for cervicogenic headache. Spine 27(17): 1835–1843 Kapandji IA 1982 The Physiology of Joints. Annotated Diagrams of the Mechanics of the Human Joints. Vol 1, Upper Limb, 5th edn. Churchill Livingstone, Edinburgh Kelso JAS, Tuller B 1984 A dynamical basis for actions systems. In: Gazanniga MS (ed.) Handbook of Cognitive Neuroscience. Plenum, New York, pp 321–356 Kendall FP, McCreary EK, Provance PG 1993 Muscles: Testing and Function, 4th edn. Williams and Wilkins, Baltimore, Ch 8, pp 235–298 Kibler B 1998a Shoulder rehabilitation: Principles and practice. Medicine and Science in Sports and Exercise S40–S50, http:// www.wwilkins.com/MSSE Manual Therapy (2003) 8(4), 195–206

Kibler B 1998b The role of the scapula in athletic shoulder function. American Journal of Sports Medicine 26 (2); 325–339 Kibler WB, Chandler TJ 1994 Sport-speciﬁc conditioning. The American Journal of Sports Medicine 22(3): 121–432 Knott M, Voss D 1968 Proprioceptive Neuromuscular Facilitation. Harper and Row, New York Kunkel SS, Hawkins RJ 1994 Open repair of the rotator cuff. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, Ch 15, p 141–152 Lippitt S, Matsen F 1993 Mechanisms of glenohumeral joint stability. Clinical Orthopaedics and Related Research 291: 20–28 Lippitt SB, Vanderhooft E, Harris SL, Sidles JA, Harryman DT, Matsen FA 1993 Glenohumeral stability from concavitycompression: A quantitative analysis. Journal of Shoulder and Elbow Surgery 2(1): 27–35 Loeb PE, Andrews JR, Wilk KE 1994 Arthroscopic debridement of rotator cuff injuries. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, Ch 17, p 165 Magarey ME 1999 The shoulder complex: Differentiation of different diagnostic procedures. Correlation of diagnoses from clinical orthopaedic, physiotherapy and arthroscopic examination. PhD Thesis, University of South Australia Magarey ME, Jones MA (2003) Evaluation of dynamic stability of the glenohumeral joint. Manual Therapy. This issue McConnell J 2001 Neuromuscular re-education strategies for the shoulder girdle. In: Tovin BJ, Greenﬁeld BH (eds) Evaluation and Treatment of the Shoulder. An Integration of the Guide to Physical Therapist Practice. FA Davis Co, Philadelphia, Ch 14, pp 347–365 Norkin CC, Levangie PK 1988 Joint Structure and Function, A Comprehensive Analysis. FA Davis Company, Philadelphia. O’Brien SJ, Arnoczky SP, Warren RF, Rozbruch SR 1990 Developmental anatomy of the shoulder and anatomy of the glenohumeral joint. In: Rockwood CA, Matsen FA (eds) The Shoulder. WB Saunders, Philadelphia, Ch 1, pp 1–33 O’Brien SJ, Pagnani MJ, Panariello RA, O’Flynn HM, Fealy S 1994 Anterior instability of the shoulder. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, Ch 18, p 177–203 O’Driscoll SW 1993 Atraumatic instability: Pathology and pathogenesis. In: Matsen FA, Fu FH, Hawkins RJ (eds) The shoulder: A Balance of Mobility and Stability. American Academy of Orthopaedic Surgeons, Rosemont, IL, Ch 18, pp 305–316 O’Sullivan PB 2000 Lumbar segmental ‘instability’: Clinical presentation and speciﬁc stabilizing exercise management. Manual Therapy 5(1): 2–12, DOI:10.1054/math.1999.0213 O’Sullivan PB, Twomey LT, Allison GT 1997a Evaluation of speciﬁc stabilizing exercise in the treatment of chronic low back pain with radiologic diagnosis of spondylolysis or spondylolisthesis. Spine 22: 2959–2967 O’Sullivan PB, Twomey LT, Allison GT, Taylor J 1997b Speciﬁc stabilising exercise in the treatment of chronic low back pain with a clinical and radiological diagnosis of lumbar segmental ‘instability’. In: Proceedings, 10th Biennial Conference, Manipulative Physiotherapists Association of Australia, Melbourne, MPAA, Melbourne, p 139 Pagnani MJ, Warren RF 1994 Stabilisers of the glenohumeral joint. Journal of Shoulder and Elbow Surgery 3(3): 173–190 Panjabi M 1992 The stabilising system of the spine. Part II: Neutral zone and stability hypothesis. Journal of Spinal Disorders 5: 390–397 Pink M, Jobe FW, Perry J, Browne A, Scovazzo ML, Kerrigan J 1993 The painful shoulder during the butterﬂy stroke. An electromyographic and cinematographic analysis of twelve muscles. Clinical Orthopaedics and Related Research 288: 60–72 Poppen NK, Walker PS 1978 Forces at the glenohumeral joint in abduction. Clinical Orthopaedics and Related Research 135: 165–170 Richardson C 1987 Atrophy of vastus medialis in patello-femoral pain syndrome. In: Proceedings Tenth International Congress of the World Confederation of Physical Therapy, Sydney, p 400 Richardson C, Jull G 1994 Concepts of rehabilitation for spinal stability. In Boyling JD, Palastanga N (eds) Grieve’s Modern r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 206 Manual Therapy

Manual Therapy, 2nd edn. Churchill Livingstone, Edinburgh, pp 705–720 Richardson CA, Jull GA 1995 Muscle control – pain control. What exercises would you prescribe? Manual Therapy 1: 2–10 Richardson C, Jull G, Hodges P, Hides J 1999 Therapeutic Exercise for Spinal Segmental Stabilization in Low Back Pain: Scientiﬁc Basis and Clinical Application. Churchill Livingstone, Edinburgh Saha AK 1971 Dynamic stability of the glenohumeral joint. Acta Orthopaedica Scandinavica 42: 491–505 Sahrmann SA 2001 Diagnosis and Treatment of Movement Impairment Syndromes. Mosby, St Louis Scarpinato DF, Andrews JR 1994 Posterior instability of the shoulder. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, Ch 19, p 205–214 Schenkman M, Rugo de Cartaya V 1987 Kinesiology of the shoulder complex. Journal of Orthopaedic and Sports Physical Therapy 8: 438–449 Schenkman M, Rugo de Cartaya V 1994 Kinesiology of the shoulder complex. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, Ch 2, p 15–33 Schmidt RA (1991) Motor learning principles for physical therapy. Contemporary management of motor control problems. Proceedings of the II Step Conference, APTA, Alexandria, VA Scovazzo ML, Browne A, Pink M, Jobe FW, Kerrigan J 1991 The painful shoulder during freestyle swimming. The American Journal of Sports Medicine 19(6): 577–582 Sharkey NA, Marder RA 1995 The rotator cuff opposes superior migration of the humeral head. American Journal of Sports Medicine 23: 270–275 Shumway-Cook A, Woollacott MJ 2001 Motor Control: Theory and Practical Applications. Lippincott Williams and Wilkins, Philadelphia Soderberg GL 1986 Kinesiology. Application to Pathological Motion. Williams and Wilkins, Baltimore

r 2003 Elsevier Ltd. All rights reserved.

Solem-Bertoft E, Thuomas K-A, Westerberg C-E 1993 The inﬂuence of scapular retraction and protraction on the width of the subacromial space. An MRI study. Clinical Orthopaedics and Related Research 296: 99–103 Souza TA (ed.) (2000) Sports injuries of the shoulder. Churchill Livingstone, New York; Culham & Laprade Timmerman LA, Andrews JR, Wilk KE 1994 Mini-open repair of the rotator cuff. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, Ch 16, p 153–163 Voss DE, Knott M, Kabat M 1953 Application of neuromuscular facilitation in the treatment of shoulder disabilities. Physical Therapy Review 33: 536–541 Wadsworth DJS, Bullock-Saxton JE 1997 Recruitment patterns of the scapular rotator muscles in freestyle swimmers with subacromial impingement. International Journal of Sports Medicine 18: 618–624 Wilk KE 1994 Current concepts in the rehabilitation of athletic shoulder injuries. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, Ch 30, p 335–354 Wilk KE, Arrigo CA 1992 An integrated approach to upper extremity exercises. In: Timm KE (ed.) Exercise Principles. Orthopedic Physical Therapy Clinics of North America 9: 337–378 Winstein CJ (1991) Designing practice for motor learning: clinical implications. Contemporary management of motor control problems. Proceedings of the II Step Conference, APTA, Alexandria, VA Woollacott M, Shumway-Cook A 1990 Changes in posture control across the life span: A systems approach. Physical Therapy 70: 799–807 Wuelker N, Korell M, Thren K 1998 Dynamic glenohumeral joint stability. Journal of Shoulder and Elbow Surgery 7(1): 43–52 Wuelker N, Roetman B, Roessig S 1995 Coracoacromial pressure recordings in a cadaveric model. Journal of Shoulder and Elbow Surgery 4(6): 462–467

Manual Therapy (2003) 8(4), 195–206

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 207–213 r 2003 Elsevier Science Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00012-2

Original article

Static innominate asymmetry and leg length discrepancy in asymptomatic collegiate athletes C.J. Krawiec, C.R. Denegar, J. Hertel, G.F. Salvaterra, W.E. Buckley Athletic Training Research Laboratory, Department of Kinesiology, Pennsylvania State University, USA

SUMMARY. The objectives of the study were to assess: (1) static innominate asymmetry in the sagittal plane, (2) leg length discrepancy (LLD), and (3) the relationship between static innominate rotation and LLD in asymptomatic collegiate athletes. The study was an observational study by design which took place in a University athletic training research laboratory. The participants were twenty-four male and 20 female asymptomatic intercollegiate athletes who volunteered to take part in the study. Static innominate asymmetry was assessed with a caliper/inclinometer tool and LLD was measured with a tape measure using standard clinical methods. Results showed that forty-two subjects (95%) demonstrated some degree of static innominate asymmetry. In 32 subjects (73%), the right innominate was more anteriorly rotated than the left. Nearly all subjects were determined to have unequal leg lengths with a majority, 30 subjects (68%), showing a slightly longer left leg. Weak correlations (r = 0.33 – 0.44) were identiﬁed between static innominate asymmetry and LLD. In conclusion static innominate asymmetry and LLD are common among asymptomatic collegiate athletes. This information provides clinicians with normative data of common clinical measures in a physically active population. r 2003 Elsevier Science Ltd. All rights reserved.

motion between the sacrum and the innominate bones exists such that motion ipsilaterally is dependent on and relative to motion and position contralaterally (Bemis & Daniel 1995). Clinical assessment of innominate position and motion is subsequently made by describing motion or position of one side in relation to the other (Beal 1982; Erhard & Bowling 1977; Cibulka et al. 1988; Crowell et al. 1994; Bemis & Daniel 1995). The literature contains very little documentation of the incidence of speciﬁc innominate asymmetries as determined under controlled investigation. This is in large part due to the difﬁculty in assessing pelvic asymmetry, arising from the low reliability of common clinical tests and the lack of a gold standard objective measure (Potter & Rothstein 1985; Cummings & Crowell 1988; Dreyfuss et al. 1994). The most common pelvic asymmetry that has been studied is that of innominate rotation in the sagittal plane (Beal 1982; Cibulka et al. 1988; Crowell et al. 1994; Bemis & Daniel 1995). However, research related to the symmetry between the innominates in healthy and symptomatic populations is limited.

INTRODUCTION Physical medicine and rehabilitation clinicians regularly examine the pelvic girdle when evaluating patients with lumbosacral and lower extremity pain and dysfunction. Faulty biomechanics of the lower extremity, pelvis, and spine are thought to contribute to musculoskeletal pathology in these regions. Alignment and motion in the pelvic region is particularly complex making clinical assessment difﬁcult. The bony pelvis is comprised of the right and left innominate bones, which are each composed of the fused segments of the ilium, ischium and pubis. The Received: 10 July 2002 Revised: 31 October 2002 Accepted: 15 January 2003 Chad J. Krawiec, MS, ATC, Craig R. Denegar, PhD, ATC, PT, Jay Hertel, Phd, ATC, George F. Salvaterra, Phd, ATC, W.E. Buckley, PhD, ATC, Athletic Training Research Laboratory, Department of Kinesiology, Pennsylvania State University, USA. Correspondence to: JH, Athletic Training Research Laboratory, Department of Kinesiology, Pennsylvania State University, 269 Recreation Building, University Park, PA 16802, USA. Tel.: +1 814 865 8816; Fax: +1 814 865 1275; E-mail: [email protected] 207

ARTICLE IN PRESS 208 Manual Therapy

Leg length discrepancy (LLD) has long been implicated as an etiological factor of pain and dysfunction throughout the lower quarter (Beal 1977; 1982; Woerman & Binder 1984; Gogia & Braatz 1986; Aspergren et al. 1987; Danbert 1988; Schuit et al. 1989; Beattie et al. 1990; Don Tigny 1990; Hoyle et al. 1991; Mannello 1992; Cummings et al. 1993; Gross et al. 1998). Pelvic asymmetry and LLD are interrelated because the innominates will typically adapt in either an anteriorly or posteriorly rotated position in order to lengthen or shorten the extremity relative to the contralateral side (Kuchera & Kuchera 1997). The extent of the relationship between LLD and pelvic asymmetry has been investigated and described in the literature (Pitkin & Pheasant 1936; Cummings et al. 1993), however, the natural occurrence of this relationship has not been previously documented in a healthy athletic population. The purposes of our study were to assess: (1) static innominate asymmetry in the sagittal plane, (2) leg length discrepancy (LLD), and (3) the relationship between static innominate rotation and LLD in asymptomatic collegiate athletes. Innominate position asymmetry differences between males and females were also examined.

landmarks, namely the anterior superior (ASIS) and posterior superior (PSIS) iliac spines, and measurement of the sagittal plane rotation of the innominate bone. A standard clinical tape measure with 1 mm units was utilized to measure leg length. A large paper clip was ﬁxed to the tape in a manner that allowed the tape to slide through the paper clip and the tip of the free arm of the paper clip was bent over the edge of the tape, creating a pointer in line with the increments on the tape (see Fig. 2). The proximal end of the tape was placed at the ASIS. The distal measurement was made utilizing the paper clip. The clip was slid up the tape until the bent tip butted up beneath the distal end of the medial or lateral malleoli, and the measurement was read where the adjacent arm crossed the tape. Procedures One investigator (CJK) made all palpations for measurement placements of leg length and innominate

METHODS Subjects Twenty-four male (age=19.771.2 years, h=185.17 7.3 cm, mass=81.479.9 kg) and 20 female (age= 19.471.2 years, h=168.676.9 cm, mass=64.276.2 kg) healthy intercollegiate athletes volunteered to participate. All subjects read and signed an informed consent form approved by the Institutional Review Board at the Pennsylvania State University. All subjects were free of self-reported pain in the lumbosacral spine, hip, groin, thigh, or buttocks that caused the subject to seek medical attention or caused interruption or signiﬁcant limitation of normal physical activity or athletic participation for a period of 1 week or more during the 6 months prior to the study.

Fig. 1—Measurement of static innominate rotation using the PALM. Caliper ends are place on the ipsilateral ASIS and PSIS.

Measurement tools Sagittal plane innominate position was measured using the Palpation Meter (PALM), a commercially available caliper—inclinometer instrument (Performance Attainment Associates, St. Paul, MN). The PALM, shown in Fig. 1, consists of an inclinometer and two caliper arms. The bubble inclinometer is a semi-circular arc with one-degree gradations that range from 01 to 301 on either side of the midline. The caliper tips allow for direct palpation of bony r 2003 Elsevier Science Ltd. All rights reserved.

Fig. 2—A large paper clip was ﬁxed to the tape measure in a manner that allowed the tape to slide through the paper clip and the tip of the free arm of the paper clip was bent over the edge of the tape, creating a pointer in line with the increments on the tape. Manual Therapy (2003) 8(4), 207–213

ARTICLE IN PRESS Asymptomatic collegiate athletes 209

Statistical analysis Intraclass correlation coefﬁcients (ICC) and standard errors of measurement (SEM) were calculated to estimate the intratester reliability and precision of measurement for the static innominate rotation and LLD measurements. Means of the three innominate rotation measurements were calculated for the right and left sides. Positive angular measurements indicated anterior Manual Therapy (2003) 8(4), 207–213

innominate rotation, while negative values indicated posterior rotation. The difference in innominate rotation between the right and left sides was calculated by subtracting the left side mean from the right side mean. Descriptive statistics and frequency distributions were calculated. Means were calculated for the left and right leg length measurements. LLD was calculated from the means of the three repeated measures by subtracting the value for the left leg from the value for the right leg. A positive value indicated that the right leg was longer, and a negative value indicated that the left leg was longer. Descriptive statistics and frequency distributions were calculated for the LLD measures. Pearson product moment correlations were calculated between: (1) MM and LM LLD measures, (2) MM LLD and innominate rotation measures, and (3) LM LLD and innominate rotation measures. Independent t-test were conducted to assess gender differences in each measure. The alpha level was set at Po0.05 for the statistical comparisons. All data were analysed using the SPSS 8.0 statistical package (SPSS Inc., Chicago, IL).

RESULTS ICCs for the innominate rotation measures were 0.99 and SEMs ranged from 0.441 to 0.471. ICCs for the LLD measures were 0.99 and SEMs ranged from 0.14 to 0.22 cm. Forty-two subjects (95%) had some degree of innominate position asymmetry and in 32 subjects (73%) the right innominate was in a more anteriorly rotated position (1.9172.61). The distribution of innominate rotation asymmetry is illustrated in Fig. 3. A majority (n=24, 54.5%) of subjects

40

13

Percentage of Subjects

position, while an assistant recorded the numerical measurements, thus blinding the palpating investigator from the results. Measurement of static innominate asymmetry was made in standing, with the subject’s anterior thigh snug against a treatment table and looking at a ﬁxed point on the wall to reduce postural sway. Initial palpation of the ASIS was made by bringing the thumbs from inferior to superior, coming underneath the solid ledge created by the inferior slope of the ASIS and then to the most prominent protrusion of the ASIS, where an adhesive marker was placed. The tester then palpated the PSIS by tracing the iliac crest posteriorly to the most posterior part and moving the thumbs superiorly and laterally from the sacrum to the ledge created by the PSIS. After butting the thumbs underneath the PSIS, the tester moved his thumb superiorly to the most prominent protrusion of the PSIS, and placed an adhesive marker over it. To derive a kinesthetic sense of the location of the landmarks, the tester then palpated the ipsilateral ASIS and PSIS prominences as before, except this time with the index ﬁngers. Using the caliper tips of the PALM, the investigator repeated the initial palpation, established the position on the ipsilateral ASIS and PSIS, and stabilized the PALM (see Fig. 1). An assistant then read the inclinometer and recorded the measurement in degrees. The investigator positioning the PALM was not informed of the measurement recordings. Positive angular measurements indicated anterior innominate rotation, while negative values indicated posterior rotation. Three measurements were made on both the right and left innominates. Leg length difference was measured with subjects lying supine on a plinth. Leg length was measured by two methods; by measuring from the ASIS to the medial malleolus (MM) and from the ASIS to the lateral malleolus (LM) (Hoppenfeld 1976). The tip of the tape measure was held at the ASIS. The tape coursed across the lower extremity to the respective malleolus where the measurement was made by butting the paper clip tip against the distal ledge of the malleolus. The assistant lifted the tape, read, and recorded the measurement without informing the tester who laid the tape of the results. Three measurements of each measurement were made on both the right and left limbs.

30 10 8

20 7

10

3

3

0 <-2º

0 to 2.0º -2.0 to -0.01º

4.01 to 6º 2.01 to 4º

>6º

Degrees of innominate asymmetry

Fig. 3—Percentage of subjects and absolute number of subjects (above bars) with innominate rotation differences between the right and left sides. A positive (+) value indicated that the right innominate was in a more anteriorly rotated position than the left, while a negative ( ) value indicated that the left innominate was in a more anteriorly rotated position than the right. r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 210 Manual Therapy

presented with innominate position differences greater than 21, irrespective of direction. Across our sample, the right innominate was in a more anteriorly rotated position than the left by a mean of 1.911. Only two subjects were measured as having no difference between the sagittal plane rotation of the right and left innominates. Forty-two subjects (95%) exhibited a LLD with 30 subjects (68%) demonstrating a longer left leg. The mean MM LLD assessment was 0.2270.73 cm. The mean LM LLD was –0.1470.62 cm.The frequency distribution of the LLD measurements are illustrated in Figs. 4 and 5. There was a high frequency of measurable LLD across our sample with only two subjects showing equal leg lengths by the LM method

and no subjects showing equal leg lengths by the MM method. The measured differences were small as most subjects had less than a 1 cm difference. Only six subjects (13.7%) displayed a LLD greater than 1 cm by the MM method, and only ﬁve subjects (11.3%) displayed a LLD greater than 1 cm by the LM method. A relatively strong correlation was found between the MM and LM LLD measures (r=0.75, Po0.01). Weak negative correlations were found between the measures of innominate rotation and MM LLD (r=0.33, P=0.026) and LM LLD (r=0.44, P=0.003). A non-signiﬁcant (t=1.09, P=0.28) difference in innominate rotation was found between males and females with females demonstrating a slightly greater (0.91) right anterior innominate rotation.

40

DISCUSSION

Percentage of Subjects

14

30

11

20

7 6 5

10 1

0 -2.0 to -1.01 -0.5 to -0.01 -1.0 to -0.51 0 to 0.5

0.51 to 1.0 1.01 to 2.0

Centimeters of leg length discrepancy

Fig. 4—Percentage of subjects and absolute number of subjects (above bars) with leg length differences between the right and left limbs as measured from the ASIS to medial malleolus. A positive (+) value of leg length difference indicated that the right leg was the longer leg, and a negative ( ) value indicated that the left leg was longer.

40 15

Percentage of Subjects

13

30

8

20

10

3

3 2

0 -2.0 to -1.01

-0.5 to -0.01 0.51 to 1.0 1.01 to 2.0 -1.0 to -0.51 0 to 0.5 Centimeters of leg length discrepancy

Fig. 5—Percentage of subjects and absolute number of subjects (above bars) with leg length differences between the right and left limbs as measured from the ASIS to lateral malleolus. A positive (+) value of leg length difference indicated that the right leg was the longer leg, and a negative ( ) value indicated that the left leg was longer. r 2003 Elsevier Science Ltd. All rights reserved.

The primary ﬁndings of this investigation were: (1) asymmetry of the innominate bones was demonstrated frequently with the majority of subjects demonstrating a more anteriorly rotated right innominate, (2) LLD was also frequently observed with a longer left leg being more common, and (3) there were relatively weak correlations between the magnitude of LLD and innominate rotation asymmetry. Innominate rotation asymmetry The ICCs for our measures of innominate position are higher than other studies utilizing a caliper and inclinometer measurement tool (Walker et al. 1987; Cibulka et al. 1988; Crowell et al. 1994; Gilliam et al. 1994; Hagins et al. 1998). Hagins et al. (1998) previously investigated the measurement properties of the PALM and reported an ICC of 0.98 and an SEM of 3.661, which was much higher than our mean SEM of 0.451. Our estimates of reliability and precision of measurement lend credibility to our ﬁndings. Using a 95% conﬁdence interval, measures of innominate position would be 70.911 from the true measurement. With conﬁdence that measurement could repeatedly be made within 11 of precision, the 24 subjects (54.5%) showing an innominate rotation difference of greater than 21 and the 32 subjects (73%) showing a difference of at least 11 can be considered to be recorded with great accuracy. Thus we conclude that a substantial proportion of our sample of healthy intercollegiate athletes presented with a quantiﬁable sagittal plane innominate rotation asymmetry. Barakatt et al. (1996) previously reported a mean innominate rotation difference of 2.3171.71 in their study of gymnasts. Their results indicated the right Manual Therapy (2003) 8(4), 207–213

ARTICLE IN PRESS Asymptomatic collegiate athletes 211

innominate to be in a more posteriorly rotated position than the left. This ﬁnding is similar to our results in magnitude but not direction. The unique physical characteristics of gymnasts, who may tend to possess greater overall range of motion than other athletes, may affect the comparisons that can be made between these two studies. Hagins et al. (1998) reported that 43% of their subjects showed a difference in innominate position of greater than 21, irrespective of direction. They also reported that 52% of their subjects were in a more anteriorly rotated position on the right and 43% were in a more anteriorly rotated position on the left. (Hagins et al. 1998) Variations in palpation methods or control of subject postural sway may account for the differences in the results between their study and our current study. The ranges of innominate rotation measurements we identiﬁed were quite small. The largest innominate rotation asymmetry we recorded was 71. The clinical relevance of asymmetries of these magnitudes remains unknown. Currently, the signiﬁcance of magnitude of asymmetry is left to the discretion of clinicians and must be placed in the context of the entire presentation of signs and symptoms of individual patients. Leg length discrepancy A tape measure was used to detect LLD in this study because of the common use in clinical settings as well as previous reports of validity and reliability for these methods (Woerman & Binder 1984; Gogia & Braatz 1986; Beattie et al. 1990; Hoyle et al 1991; Mannello 1992). The high intratester reliability (ICC=0.99) found in our study was consistent with or higher than previous studies. Using a 95% conﬁdence interval, measures would be 70.36 cm for the MM method and 70.31 cm for the LM method. Even though the SEMs we reported were small, error was still an issue. In 11 out of 44 subjects (25%), the estimated long leg was different between the MM and LM methods. Potential sources of error for both methods include palpation variations, tissue contour and deformation, and body positioning. The LM method may be considered the better of the two methods based on its slightly greater precision of measurement although ﬁrm conclusions are difﬁcult to draw. Differences between the two measures support previous research that shows the LM method to be more accurate and precise than the MM method (Woerman & Binder 1984). This may be due factors such as having a more direct line of measurement with less torsion to the tape measure, and the lateral malleolus landmark being more prominent. The high frequency of occurrence of LLD we found was consistent with previous ﬁndings (Beal Manual Therapy (2003) 8(4), 207–213

1977; Mannello 1992). By the MM method, 30 subjects (68%) demonstrated a longer left leg. The LM measurements were more centrally distributed than the MM measurements with 24 subjects (54.5%) exhibiting a longer left leg. Our ﬁndings agree with Beal’s report of the more frequent left long leg (Beal, 1977). The more frequent ﬁnding of a long left leg is interesting because of the possible link to innominate position asymmetry. It has been theorized and demonstrated in research that the innominates will typically adapt to an anteriorly or posteriorly rotated position in order to lengthen or shorten one extremity relative to the other (Cummings et al. 1993; Kuchera & Kuchera 1997). Leg length discrepancy and innominate rotation asymmetry Our results demonstrate a frequent occurrence of both anterior innominate rotation asymmetry and LLD, but failed to demonstrate a strong correlation between these two measures. There is evidence of a weak association between the ﬁndings of a longer left leg and an anteriorly rotated innominate on the right side, however, the variation in LLD measures accounts for less than 19% of the variation seen in innominate rotation asymmetry (r= 0.33 to 0.44). The bivariate correlations between MM LLD and innominate rotation, and LM LLD and innominate rotation were negative correlations. At ﬁrst glance, this might be interpreted to mean that as LLD increases innominate position asymmetry decreases. This would not seem sensible because it is theorized that as LLD increases the amount of asymmetry between innominates also increases. However when interpreting our results, the negative sign in the correlation coefﬁcient can be attributed to our subtracting the measurements of the left side from the measurements of the right side. If the equation were changed to subtracting the right side measures from the left side, the correlation would then be positive. As the amount of LLD increases, one would expect that innominate position asymmetry would also increase due to the antagonistic mechanical movement of the two innominates. The innominate on the side opposite to the longer leg would anteriorly rotate in order to lengthen the short leg, and the long leg innominate would rotate posteriorly to shorten the long leg. The ﬁndings across our sample support this association. There was a greater frequency of a more anteriorly rotated right innominate and a majority of subjects were estimated to have a longer left leg. These ﬁndings are consistent with those of Pitkin and Pheasant (1936) and Cummings et al. (1993). Previously it was mentioned that the innominate on the contralateral side to the longer leg would tend r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 212 Manual Therapy

to be found in a more anteriorly rotated position in order to lengthen the short leg. This seems to be mechanically reasonable, however, this is a phenomenon best seen in standing, weight-bearing positions. Dispute arises when, as in our study, leg length is measured in a supine, non-weight bearing position. In supine positions, an anteriorly rotated innominate is associated with the longer leg, and a posterior rotated innominate is associated with the shorter leg (Bemis & Daniel 1987). Any correlation between supine measures of leg length, particularly clinical measures utilizing pelvic landmarks, and standing measures of innominate position may be called into question. The supine measure of leg length was used in our study, despite this inevitable conﬂict, because we felt that this method provided the most commonly used clinical method of measuring leg length without expensive radiographic imaging techniques. It is difﬁcult to make an accurate assessment of leg length while standing, because any weight bearing measure could be considered a measure of functional leg length. The muscular action, joint compression, and overall compensation that occur in standing are factors that are troublesome when trying to uncover the impact of true LLD on innominate asymmetry. The presence of LLD and innominate asymmetry may also be a compensatory result of repetitive motor patterns related to limb dominance. Our subjects were predominantly right hand dominant and the right side innominate was most often in an anteriorly rotated position along with a higher frequency of a longer left leg. LLD may also be related to side dominance and motor control as a result of developmental patterns. Further research is needed to investigate this possibility.

CONCLUSIONS Our results demonstrated the frequent incidence of innominate rotation and LLD in asymptomatic intercollegiate athletes, however there were weak statistical correlations between measures of LLD and innominate rotation. Subjects most frequently demonstrated greater anterior innominate rotation of the right side and a longer left leg. Innominate rotations and LLD appear to occur normally in asymptomatic athletes without history of lumbosacral symptoms. Thus, a physical examination ﬁnding of static asymmetry must be put in context of a complete examination including a thorough history and tests of joint mobility and stability. Further research is needed to explore objective methods of quantifying pelvic asymmetries among healthy and symptomatic populations and the relationships between LLD, innominate rotation, and lower quarter dysfunction. r 2003 Elsevier Science Ltd. All rights reserved.

References Aspegren DD, Cox JM, Trier KK 1987 Short leg correction: A clinical trial of radiographic vs. non-radiographic procedures. Journal of Manipulative and Physiological Therapeutics 10(5): 232–238 Barakatt E, Smidt GL, Dawson JD, Wei SH, Heiss DG 1996 Interinnominate motion and symmetry: Comparison between gymnasts and nongymnasts. Journal of Orthopaedic and Sports Physical Therapy 23(5): 309–319 Beal MC 1977 The short leg problem. Journal of the American Osteopathic Association 76: 745–751 Beal M 1982 The sacroiliac problem: Review of anatomy, mechanics, and diagnosis. Journal of the American Osteopathic Association 81(10): 667–679 Beattie P, Isaacson K, Riddle DL, Rothstein JM 1990 Validity of derived measurements of leg-length differences obtained by use of a tape measure. Physical Therapy 70(3): 150–157 Bemis T, Daniel M 1987 Validation of the long sitting test on subjects with iliosacral dysfunction. Journal of Orthopaedic and Sports Physical Therapy 8(7): 336–345 Cibulka MT, Delitto A, Koldehoff RM 1988 Changes in innominate tilt after manipulation of the sacroiliac joint in patients with low back pain: an experimental study. Physical Therapy 68(9): 1359–1363 Crowell RD, Cummings GS, Walker JR, Tillman LJ 1994 Intratester and intertester reliability and validity of measures of innominate bone inclination. Journal of Orthopaedic and Sports Physical Therapy 20(2): 88–97 Cummings G, Crowell RD 1988 Source of error in clinical assessment of innominate rotation. Physical Therapy 68(1): 77–79 Cummings G, Scholz JP, Barnes K 1993 The effect of imposed leg length difference on pelvic bone symmetry. Spine 18(3): 368–373. Danbert RJ 1988 Clinical assessment and treatment of leg length inequalities. Journal of Manipulative and Physiological Therapeutics 11(4): 290–295. Don Tigny RL 1990 Anterior dysfunction of the sacroiliac joint as a major factor in the etiology of idiopathic low back pain syndrome. Physical Therapy 70(4): 250–265 Dreyfuss P, Dryer S, Grifﬁn J, Hoffman J, Walsh N 1994 Positive sacroiliac screening tests in asymptomatic adults. Spine 19(10): 1138–1143 Erhard R, Bowling R 1977 The recognition and management of the pelvic component of lowback and sciatic pain. Bulletin of the Orthopaedic Section, APTA 2(3): 4–15 Gilliam J, Brunt D, MacMillan M, Kinard RE, Montgomery WJ 1994 Relationship of the pelvic angle to the sacral angle: measurement of clinical reliability and validity. Journal of Orthopaedic and Sports Physical Therapy 20(4): 193–199 Gogia PP, Braatz JH 1986 Validity and reliability of leg length measurements. Journal of Orthopaedic and Sports Physical Therapy 8(4): 185–188 Gross MT, Burns CB, Chapman SW, Hudson CJ, Curtis HS, Lehmann JR, Renner JB 1998 Reliability and validity of rigid lift and pelvic leveling device method in assessing functional leg length inequality. Journal of Orthopaedic and Sports Physical Therapy 27(4): 285–294 Hagins M, Brown M, Cook C 1998 Intratester and intertester reliability of the Palpation Meter (PALM) in measuring pelvic position. Journal of Manual and Manipulative Therapy 6(3): 130–136 Hoppenfeld S 1976 Physical Examination of the Spine and Extremities. Appleton-Century-Crofts, Norwalk, pp 165–166 Hoyle DA, Latour M, Bohannon RW 1991 Intraexaminer, interexaminer, and interdevice comparability of leg length measurements obtained with measuring tape and Metrecom. Journal of Orthopaedic and Sports Physical Therapy 14(6): 263–268 Kuchera ML, Kuchera WA 1997 Postural considerations in coronal and horizontal planes. In: Ward RC (ed.) Foundations for Osteopathic Medicine. Williams and Wilkins, Baltimore, pp 983–986 Mannello DM 1992 Leg length inequality. Journal of Manipulative and Physiological Therapeutics 15(9): 576–590 Manual Therapy (2003) 8(4), 207–213

ARTICLE IN PRESS Asymptomatic collegiate athletes 213

Pitkin HC, Pheasant HC 1936 Sacrarthrogenetic telalgia. Journal of Bone and Joint Surgery 56: 365–374 Potter NA, Rothstein JM 1985 Intertester reliability for selected clinical tests of the sacroiliac joint. Physical Therapy 65(11): 1671–1675 Schuit D, McPoil TG, Mulesa P 1989 Incidence of sacroiliac malalignment in leg length discrepancies. Journal of the American Podiatric Medical Association 79(8): 380–383

Manual Therapy (2003) 8(4), 207–213

Walker ML, Rothstein JM, Finucane SD, Lamb RL 1987 Relationships between lumbar lordosis, pelvic tilt, and abdominal muscle performance. Physical Therapy 67(4): 512–516 Woerman AL, Binder-MacLeod SA 1984 Leg length discrepancy assessment: Accuracy and precision in ﬁve clinical methods of evaluation. Journal of Orthopaedic and Sports Physical Therapy 5(5): 230–239

r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 214–222 r 2003 Elsevier Science Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00013-4

Original article

Health care provider’s attitudes and beliefs towards chronic low back pain: the development of a questionnaire R.W.J.G. Ostelon,y, S.G.M. Stomp-van den Bergw, J.W.S. Vlaeyenz, P.M.J.C. Woltersy, H.C.W. de Vet8 Department of Epidemiology, Maastricht University, The Netherlands, w Netherlands Institute for Health Services Research, The Netherlands, z Department of Clinical & Experimental Psychology, Maastricht University, The Netherlands, y Hogeschool Zuyd, Department of Physiotherapy, Heerlen, The Netherlands, 8 Institute for Research in Extramural Medicine, VU University Medical Center, The Netherlands

n

SUMMARY. Attitudes and beliefs, or the treatment orientation, of health care providers appear to be important in the management of non-speciﬁc chronic low back pain (CLBP). The aims of the current study were two-fold: First of all, the physiotherapists’ opinion towards various aspects of the management of CLBP was surveyed. Secondly, in a principal factor analysis, it was investigated whether underlying dimensions could be identiﬁed in order to develop the Pain Attitudes and Beliefs Scale for Physiotherapists (PABS PT). In total, 421 physiotherapists (response rate 62.3%) participated in this study. The results suggested that the majority of physiotherapists hold the opinion that CLBP is not a dangerous condition, that sport should not be discouraged and that patients should not refrain from all physical activity. Moreover physiotherapists seem to hold the opinion that the way patients view their pain inﬂuences the progress of symptoms. Finally, physiotherapists seem to hold the opinion that therapy can completely alleviate the functional symptoms and that therapy may have been successful even if pain remains. The principal factor analysis (PAF) yielded an interpretable 2-factor model. Based on highest loading items, factor 1 was labelled ‘biomedical orientation’, whereas factor 2 was labelled ‘behavioural orientation’. The internal consistency (Cronbach’s Alpha) of factor 1 was 0.84 and for factor 2, 0.54 explaining 25.2% and 8.2%, respectively, of the total variance. Assessment of the effect of the physiotherapists’ characteristics on scores on the different scales was encouraging as results pointed in the directions one would expect. Physiotherapists who attended biopsychosocial education courses had statistically signiﬁcantly higher scores on the ‘behavioural orientation’ factor and vice versa. Biomedical specialists scored statistically signiﬁcantly higher on the ‘biomedical orientation’ factor. Furthermore, the ﬁndings suggested that the PABS PT discriminates between physiotherapists with a ‘behavioural orientation’ vs those with a ‘biomedical orientation’. To examine the inﬂuence of these different treatment orientations with regard to CLBP on patient outcome is a challenge for the near future. r 2003 Elsevier Science Ltd. All rights reserved.

INTRODUCTION Received: 6 September 2002 Accepted: 18 January 2003

Non-speciﬁc chronic low back pain is widely recognized as a major medical, social and economic problem and represents major challenges to health care (Waddell 1998). Important factors that determine how patients deal with non-speciﬁc chronic low back pain (CLBP) are often not related to physical pathology or pain severity. Attitudes and beliefs about the relationship between pain and function appear to explain partially the variation in disability reported by individuals with chronic pain (Rainville et al. 1995). Attitudes and beliefs of patients towards CLBP are inﬂuenced by many factors, including past

Raymond W.J.G. Ostelo, PhD, RPT, Department of Epidemiology, Maastricht University, The Netherlands, Suzanne G.M. Stomp-van den Berg, MSc, RPT, Netherlands Institute for Health Services Research, The Netherlands, John W.S. Vlaeyen, PhD, Department of Clinical & Experimental Psychology, Maastricht University, The Netherlands, Pieter M.J.C. Wolters, RPT, Hogeschool Zuyd, Department of Physiotherapy, Heerlen, The Netherlands, Henrica C.W. de Vet PhD, Institute for Research in Extramural Medicine, VU University Medical Center, The Netherlands Correspondence to: RWJGO, EMGO Institute, VU University Medical Center, Van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands. Tel.: +31 20 4448180; Fax: +31 20 4448181; E-mail: [email protected] 214

ARTICLE IN PRESS Health-care provider’s attitudes and beliefs 215

pain experience (Fordyce et al. 1973), culture (Sanders et al. 1992), and social and economic factors (Lee et al. 1989; Volinn, 1989). The pain attitudes and beliefs of patients might be strengthened if patients ﬁnd that they share the same attitudes and beliefs with health care providers. In a study by Rainville et al. (2000) the attitudes and beliefs of physicians correlated with their rating of severity of patients’ symptoms irrespective of their expertise in CLBP. Furthermore, physicians’ attitudes and beliefs (or treatment orientation) appear to inﬂuence their recommendations regarding activities and work for patients with CLBP (Rainville et al. 2000). In another study the physiotherapist’s opinion about the importance of exercise in prevention appeared to be an important predictor of the total number of instructions given to a patient, indicating that a greater belief in the importance of exercise was reﬂected in more instructions (Kerssens et al. 1999). It appears that, in general, attitudes and beliefs (or treatment orientation) of health-care providers are important in the management of CLBP. Various disciplines, with possibly differing treatment orientations concerning the management of CLBP, are involved in its management. In the Netherlands, patients with CLBP often consult physiotherapists: 27% of all patients consulting physiotherapists suffer from CLBP (Ravensberg van et al. 1995). This study, therefore, speciﬁcally focused on physiotherapists. Two possible important physiotherapists’ attitudes (or treatment orientations) can be extracted from literature regarding CLBP. First of all, a physiotherapist can derive his or her treatment orientation from the biomechanical model of disease. This model is based upon the notion that pain and disability are a consequence of physical pathology. Diagnosis of the pathology provides the basis for physical treatment of the illness. Since pain is a signal of pathology or tissue damage, a physiotherapist with a predominantly biomechanical treatment orientation towards CLBP will very likely adapt his or her treatment to the pain level of the patient (i.e. use a pain-contingent treatment approach). Furthermore, treatment will primarily be aimed at ﬁnding the physical impairment that is the cause of the pain and treating this impairment. The second source of physiotherapists’ treatment orientation comes from the biopsychosocial model of CLBP. This model emphasises the role of psychological and social factors in the development and maintenance of complaints. Therefore, pain does not have to be a sign of pathology or tissue damage, but is also inﬂuenced by social and psychological factors. Because of these factors, disability due to pain can be maintained long after the initial pathology has healed. According to this model, it is not necessarily beneﬁcial to adapt the treatment to the pain level of the patient. The treatment should rather focus on an increase in activity according to a previously deﬁned Manual Therapy (2003) 8(4), 214–222

timeframe. Physiotherapists with a predominantly biopsychosocial treatment orientation generally hold . a time-contingent treatment approach (Lindstrom et al. 1992). The aims of the current study were two-fold: First of all, the physiotherapists’ opinion towards the various aspects of the management of CLBP was surveyed. Secondly, in a principal factor analysis, it was assessed whether underlying dimensions could be identiﬁed in order to develop the Pain Attitudes and Beliefs Scale for Physiotherapists (PABS PT). This scale facilitates the assessment of the role attitudes and beliefs (or treatment orientation) on the development and persistence of CLBP.

METHODS Item collection and expert reviewing We reviewed existing questionnaires concerning patients’ attitudes and beliefs towards chronic pain and rephrased items to a therapist’s point of view. An example is item number 1 of the Tampa Scale for Kinesiophobia (TSK): ‘I’m afraid that I might injure myself if I exercise’ was rephrased as ‘If patients complain of pain during exercise, I worry that damage is being caused’. The following questionnaires were screened: (1) Tampa Scale for Kinesiophobia (TSK) (Miller et al. 1991), a measure of excessive, irrational and debilitating fear of physical movement and activity. The Dutch version (TSKDV) has a fair and consistent internal validity (KoleSnijders, 1993; Vlaeyen et al. 1995). (2) Pain Catastrophizing Scale (PCS) (Sullivan et al. 1995) that measures an exaggerated negative orientation toward pain. (3) The Back Beliefs Questionnaire (BBQ) (Symonds et al. 1996) measures the attitudes and beliefs of patients regarding the inevitability of his/her back pain. (4) The Fear Avoidance Beliefs Questionnaire (FABQ) focuses on patients’ beliefs about how physical activity and work affected their low back pain (Waddell et al. 1993). In addition we added items that we considered relevant in the management of CLBP. Then an expert review procedure was performed: several experienced physiotherapists involved in educating physiotherapists in cognitive behavioural approaches and researchers in the ﬁeld of chronic pain were consulted. Two important criteria for this expert-validity procedure were: (1) items should be unambiguous; (2) items should be able to discriminate between different treatment orientations. Initially the questionnaire consisted of 37 items, six of which were excluded after the expert procedure, yielding the ﬁnal 31-item questionnaire. The items are presented in Tables 1 and 2. Items 1,11,13,15,19,20,21 and 30 are based on the TSK-DV. Items 23 and 27 are based on r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 216 Manual Therapy Table 1. Results per item from survey (for excluded items) No. 15 21 1 16 17 18 8 19 29 2 28

Item

Mean (SD)

Back pain indicates that there is something dangerously wrong with the back Sport should not be recommended for patients with back pain Back pain sufferers should refrain from all physical activity in order to avoid injury The way patients view their pain inﬂuences the progress of the symptoms Therapy may have been successful even if pain remains Therapy can completely alleviate the functional symptoms caused by back pain Unilateral physical stress is not a cause of back pain If ADL activities cause more back pain, this is not dangerous Even if the pain has worsened, the intensity of the next treatment can be increased Good posture prevents back pain TENS and/or back braces support functional recovery

1.6 1.8 1.8 5.3 4.7 4.7 2.6 3.4 2.9 4.3 3.4

Reasons for exclusion

(0.8) (0.9) (1.0) (0.7) (1.1) (1.2) (1.3) (1.3) (1.2) (1.2) (1.3)

A A A A A A B C C D D

These items were excluded from the factor analysis, reasons are stated in column ‘Reasons for Exclusion’: A=skewness, B=minimal loading criteria, C=loading on both factors, D=raise of alpha (F1) if item deleted,). No.=number of item on questionnaire as administered, for clarity of presentation the order of presentation of the items was changed. ‘Mean (SD)’=Mean and standard deviation: 1=‘totally disagree’, 2=‘largely disagree’, 3=‘disagree to some extent’, 4=‘agree to some extent’, 5=‘largely agree’, and 6=‘totally agree’.

Table 2. Results per item from (1) survey and (2) factor analysis No. 25 24 31 26 22 30 10 13 14 9 20 5 23 4 7 12 6 3 27 11

Item

Mean (SD)

Increased pain indicates new tissue damage or the spread of existing damage Pain reduction is a precondition for the restoration of normal functioning The severity of tissue damage determines the level of pain It is the task of the physiotherapist to remove the cause of back pain If back pain increases in severity, I immediately adjust the intensity of my treatment accordingly If patients complain of pain during exercise, I worry that damage is being caused Pain is a nociceptive stimulus, indicating tissue damage The best advice for back pain is: ‘Take care’ and ‘Make no unnecessary movements’ Patients with back pain should preferably practice only pain free movements Patients who have suffered back pain should avoid activities that stress the back Back pain indicates the presence of organic injury Not enough effort is made to ﬁnd the underlying organic causes of back pain If therapy does not result in a reduction in back pain, there is a high risk of severe restrictions in the long term Reduction of daily physical exertion is a signiﬁcant factor in treating back pain The cause of back pain is unknown Functional limitations associated with back pain are the result of psychosocial factors Mental stress can cause back pain even in the absence of tissue damage Knowledge of the tissue damage is not necessary for effective therapy There is no effective treatment to eliminate back pain A patient suffering from severe back pain will beneﬁt from physical exercise

IC

F1

2.6 3.3 2.4 3.0 3.8

(1.1) (1.5) (1.3) (1.6) (1.4)

0.461 0.422 0.403 0.340 0.389

0.671 0.657 0.626 0.555 0.544

2.5 3.2 2.2 3.1 2.2 2.4 3.4 2.6

(1.1) (1.5) (1.7) (1.4) (1.1) (1.2) (1.4) (1.2)

0.364 0.337 0.286 0.299 0.182 0.374 0.239 0.193

0.541 0.526 0.526 0.495 0.444 0.431 0.407 0.396

3.5 3.0 3.0 4.4 2.6 2.8 4.2

(1.3) (1.4) (1.3) (1.3) (1.4) (1.4) (1.2)

0.170 0.222 0.174 0.179 0.133 0.113 0.195

0.329

F2

0.528 0.476 0.450 0.321 0.305 0.294

No.=number of item on questionnaire as administered, for purpose of clarity of presentation the order of presentation of the items was changed. ‘Mean (SD)’= Mean and standard deviation: 1=‘totally disagree’, 2=‘largely disagree’, 3=‘disagree to some extent’, 4=‘agree to some extent’, 5=‘largely agree’, and 6=‘totally agree’. ‘IC’=Initial Communality from Factor analysis, ‘F1’=Loading on factor 1, ‘F2’=Loading on factor 2.

the BBQ. Items 9 and 14 are based on FABQ and the researchers added items 2,3,4,5,6,7,8,10,12,16, 17,18,22,24,25,26,28,29 and 31. The questionnaire and a pre-paid return envelope were mailed to a random sample of physiotherapists (n=540) belonging to the Royal Dutch Association of Physiotherapy (KNGF). In addition, we speciﬁcally included all therapists (n=137) who had attended speciﬁc education courses (over the last 3 years) in the management of chronic pain provided by The Knowledge Center (NPi) for allied health professionals. After 1 month non-responders received a reminder by mail. The physiotherapists were instructed that the purpose of this questionnaire was to analyse how therapists approach the most common forms of back pain which does not include back pain resulting from a r 2003 Elsevier Science Ltd. All rights reserved.

radicular syndrome, cauda equina syndrome, fractures, infections, inﬂammation, a tumour or metastasis. Items are scored on a six-point Likert scale: ‘totally disagree’ (score:1), ‘largely disagree’ (score:2), ‘disagree to some extent’ (score:3), ‘agree to some extent’ (score:4), ‘largely agree’ (score:5) and ‘totally agree’ (score:6). ‘Extreme’ was deﬁned as score 1 or 2 for ‘disagreement’ and score 5 or 6 for ‘agreement’. Statistical analysis For demographic variables of participants, distribution frequencies were calculated. To assess the physiotherapists’ opinion towards the various aspects of the management of CLBP the mean score and standard deviation of each item were calculated. To Manual Therapy (2003) 8(4), 214–222

ARTICLE IN PRESS Health-care provider’s attitudes and beliefs 217

explore underlying dimensions a principal factor analysis (PAF) with an oblimin rotation was performed; a statistical approach aimed at clustering of items and item reduction. Prior to this procedure, each item was individually analysed because factor analysis can be biased if items are skewed (Bernstein & Teng, 1989). Items were excluded if skewness was not between 1 and +1 or if less than 10% of responders scored in each of the ‘extreme’ categories. On the remaining items the principal factor analysis (PAF) was performed. For the extraction of the number of factors, the eigenvalue >1 rule and the scree test were used. Also the content of the factors was taken into account. Items with loadings r0.25 on any factor were excluded. If loading on one factor exceeded 0.25 but differences between loadings was less then 0.1, items were also excluded. Cronbach’s alpha was calculated as a measure of internal consistency. For all analyses SPSS10.0 for Windows (SPSS Inc. North Michigan Avenue, Chicago, IL, 60611) was used. Characteristics of physiotherapists Various characteristics of physiotherapists might have an effect on the treatment orientation toward CLBP. First we assessed how gender, age and whether physiotherapists themselves had an experience of low back pain affected the scores of the different dimensions of the questionnaire. Also we explored the effect of the following variables: Differences in education courses attended: Education courses might affect the scores on the questionnaire. Education courses were coded either as ‘biomedical’ (e.g. manual therapy, McKenzie) or ‘biopsychosocial’ (e.g. cognitive behavioural management such as graded activity, operant treatment approaches or respondent techniques). Subjects reporting education courses in both ﬁelds were excluded from these analyses, in order to establish maximal contrast on this variable. Specialization: It is quite possible that education courses do not necessarily overlap with self-reported specialization. One might have attended education courses as an introduction without being fully committed to such an approach, whereas self-reported specialization probably better reﬂects the true approach to which subjects feel committed. Therefore, we asked respondents to indicate their primary specialization. Subjects reporting ‘Manual Therapist’ or ‘McKenzie therapist’ were coded as a ‘biomedical’ specialist; subjects reporting ‘Chronic Pain Therapist’ or similar expressions (‘Chronic Recurrent Conditions’ or ‘Pain Rehabilitation’) were coded as ‘biopsychosocial’. Work setting: We hypothesized that in rehabilitation centres, physiotherapists are used to working in a more multidisciplinary setting as opposed to Manual Therapy (2003) 8(4), 214–222

physiotherapists in private clinics and may, therefore, hold a more biopsychosocial orientation. Furthermore, in the Netherlands pain clinics are typically situated within rehabilitation centres (or specialized hospitals) and therefore physiotherapists are more exposed to severe, chronic pain patients, which might be associated with a more biopsychosocial orientation.

RESULTS Six hundred and seventy-seven questionnaires were mailed. After one reminder 421 (62.3%) questionnaires were returned (37% women). The response rates of the random sample of KNGF-physiotherapists and the therapists who had attended the NPieducation courses were 62.5% and 62.1%, respectively. The average age was 41.7 years (SD: 7). The majority (86%) were working in private clinics while 4% worked in a hospital, 3% in rehabilitation centers and 7% had a combination or another work setting. Physiotherapists in the private clinics had on average 17 years (SD: 7) of work experience. Data examination Three hundred and seventy-three responders (88.6%) completed all 31 items. One responder returned a blank questionnaire (no reason stated) and was, therefore, excluded. Furthermore, the sample included three responders with 6, 5 and 4 missing values, respectively. Two responders reported 3 missing values, 3 responders showed 2 missing values and 39 responders were identiﬁed with 1 missing value. Missing items were excluded from analysis but other items of those responders were taken into account because there was no pattern to the missing items. For clarity of presentation the items are presented in two tables, with Table 1 presenting the items that were eliminated for the factor analysis while Table 2 presents the results of the factor analysis. Both tables also present the mean score and standard deviation of every item, which was used to survey the physiotherapists’ opinion towards the various aspects of the management of CLBP. Physiotherapists’ opinion Table 1 shows three items producing mean scores of less than 2 with small standard deviations. These scores indicate that physiotherapists strongly disagree with the fact that back pain indicates that there is something dangerously wrong (item 15). Furthermore they disagree strongly with the fact that sport should not be recommended for patients with back pain (item 21) and that they should refrain from all physical activity (item 1). On the other hand three items showed consistently high mean scores. In other r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 218 Manual Therapy

words, physiotherapists agree with the fact that the way patients view their pain inﬂuences the progress of the symptoms (item 16). Moreover they agree with the fact that therapy can completely alleviate the functional symptoms (item 18) and that therapy may have been successful even if pain remains (item 17). All other items revealed mean scores that can be considered to lie in the middle range with a wider standard deviation, indicating that for these items neither strong agreements nor strong disagreements existed. Exploring underlying dimensions A principal factor analysis (PAF) with an oblimin rotation was performed. Initially 6 items were excluded because they were too skewed (see Table 1). The PAF was performed in order to cluster the different items into factors and to reduce items of the total questionnaire. The Kaiser–Meyer–Olkin Measure appeared to be 0.869. Barlett’s Test of Sphericity with an approx. Chi-square of 2089.7, was highly signiﬁcant. Both measures justiﬁed the continuation of analysis. The eigenvalue >1 criterion initially yielded 7 factors. According to the ‘scree test’ 2 factors remained. The subsequent factor analysis also yielded 2 factors. Analysing the results of the loadings led to the exclusion of 3 more items. Item 8 had a loading r0.250. Item 19 and item 29 had loadings on both factors with a difference not exceeding the preset threshold of 0.1. A high score on factor 1 (16 items) could be interpreted as a ‘Biomedical Orientation’. Examples of items loading highly on this factor are: ‘Increased pain indicates new tissue damage or the spread of existing damage’ (item 25); ‘Pain reduction is a precondition for the restoration of normal functioning’ (item 24); and ‘The severity of tissue damage determines the level of pain’ (item 31). Based on the highest loading items, factor 2 (6 items) could be labelled as ‘Behavioural Orientation’. Examples of items loading high on this factor are: ‘The cause of low back pain is unknown’ (item 7), ‘Psychosocial factors are associated with functional limitations’ (item 12), and ‘Mental stress can cause back pain even in the absence of tissue damage’ (item 6). The ﬁnal results of the factor analysis are given in Table 2. The correlation between the ‘Biomedical Orientation’ factor and the ‘Behavioural Orientation’ factor was 0.30. Internal consistency To determine the internal consistency of both factors we conducted a reliability analysis on each factor by calculating Cronbach’s alpha. The ‘Biomedical Orientation’ factor (16 items) appeared to have an Alpha of 0.83. After deleting 2 items that were r 2003 Elsevier Science Ltd. All rights reserved.

indicated as giving a raise of Alpha (item 2 and item 28), Alpha increased somewhat to 0.84. This resulted in a 14-item factor. The ‘Behavioural Orientation’ factor (6 items) appeared to have an Alpha of 0.54. Total variance explained by factor 1=25.2% and 8.2% by factor 2. Relation between characteristics of physiotherapists and scores on the questionnaire To explore the effect of age (cut-off point 42= median), gender (male, female), and whether subjects had experienced an episode of low back pain themselves (present or past) on the scores on both factors, mean scores and standard deviations were computed for the different categories of each variable. These mean scores of both categories were compared using independent sample t-tests for every factor. Table 3 summarizes the results, the theoretical range of scores and the mean scores with the standard deviations, and the range of the entire sample. Women tended to score higher compared to men on both the ‘biomedical’ factor (not statistically signiﬁcant) and the ‘behavioural’ factor (statistically signiﬁcant P ¼ 0:007). The variable age revealed a difference on factor 1: age Z42 statistically signiﬁcantly higher scores. Previous personal low back pain experience did not show any differences. The results for the variable ‘Education Courses’ pointed in the direction expected. Scores for the biomedical factor, though not statistically signiﬁcant, were higher for physiotherapists who followed biomedical education courses. For the ‘behavioural’ factor, physiotherapists who took biopsychosocial education courses scored statistically signiﬁcantly higher (P ¼ 0:002). Moreover, biomedical specialists scored statistically signiﬁcantly higher on the ‘biomedical’ factor as opposed to the biopsychosocial specialists (P ¼ 0:002), whereas biopsychosocial specialists scored higher (although not statistically signiﬁcant) on the ‘behavioural’ factor. Although scores for the variable ‘Work Setting’ pointed in the direction expected (physiotherapists working in a private clinic score higher on the ‘biomedical’ factor and physiotherapists working in a rehabilitation centre score higher on the behavioural factor) no statistically signiﬁcant differences were found.

DISCUSSION A national survey among Dutch physiotherapists regarding their opinion towards the management of CLBP suggests that there is a reasonable consensus with regard to the fact that back pain is not a dangerous condition. Also there seems to be consensus that sport should not be discouraged for patients with CLBP and that they should not refrain from all Manual Therapy (2003) 8(4), 214–222

ARTICLE IN PRESS Health-care provider’s attitudes and beliefs 219 Table 3. Characteristics of physiotherapists in relation to scores on Factors 1 and 2 Factor 1 biomedical Theoretical min–max Mean(SD) Range

Factor 2 biopsychosocial

14–84 40.2 (10.5) 16–75

6–36 20.0 (4.4) 7–34

Mean (SD)

t-test F1

P-value

Mean (SD)

t-test F2

P-value

Gender Male (n=255) Female (n=147)

39.7 (10.8) 41.2 (10.0)

1.32

0.185

19.5 (4.6) 20.7 (3.8)

2.73

0.007n

Age o42 years (n=208) Z42 years (n=212)

38.9 (10.6) 41.6 (10.3)

2.81

0.005n

20.1 (4.1) 19.8 (4.6)

0.83

0.407

Low back pain experience Yes (n=214) No (n=201)

40.1 (10.9) 40.4 (10.1)

0.26

0.796

19.9 (4.6) 19.9 (4.2)

0.10

0.917

Education courses Biomedical (n= 61) Biopsychosocial (n=34)

40.1 (10.3) 36.2 (10.9)

1.68

0.095

18.4 (3.9) 21.4 (5.2)

3.18

0.002n

Specialization Biomedical (n=89) Biopsychosocial (n=15)

41.2 (10.3) 31.9 (9.7)

3.16

0.002n

18.6 (4.4) 20.4 (4.3)

1.50

0.134

Work setting Private clinic (n=361) Rehabilitation centre (n=14)

40.6 (10.3) 36.8 (8.7)

3.16

0.187

19.7 (4.4) 21.5 (4.2)

1.40

0.153

Mean score and standard deviations within the brackets. n Statistically signiﬁcant at alpha=0.05 level.

physical activity. Moreover physiotherapists seem to hold the opinion that the way patients view their pain inﬂuences the progress of the symptoms. Finally, physiotherapists seem to hold the opinion that therapy can completely alleviate the functional symptoms and that therapy may have been successful even if pain remains. Factor analysis yielded an interpretable 2-factor model. The procured questionnaire was named: Pain Attitude and Beliefs Scale for Physiotherapists (PABS PT). High scores on factor 1 were labelled ‘Biomedical Orientation’, whereas high scores on factor 2 were labelled ‘Behavioural Orientation’. Cronbachs’ alpha of factor 1 was satisfactory whereas factor 2 had poor internal consistency. The variables ‘Education Courses’, ‘Specialisation’ and ‘Work Setting’ affected the scores on the factors as anticipated. In terms of usefulness the PBAS PT could be used to assess the treatment orientation of pain-therapists, e.g. a physiotherapist. Survey The survey yielded 6 items on which the vast majority agreed or disagreed. Typically, these items are addressed in Dutch (General Practitioners’ (Faas et al. 1996), as well as physiotherapy (Bekkering et al. 2001)) guidelines for CLBP. Dutch physiotherapists seem to acknowledge the pivotal role of patients’ ideas regarding the development and maintaining of CLBP. Furthermore, the results suggested that Manual Therapy (2003) 8(4), 214–222

physiotherapists focus their treatment on functional limitations rather than on pain. This was in concordance with the strong disagreement that back pain indicates that there is something dangerously wrong with the back, and that patients with CLBP should refrain from physical activity, which might reﬂect the knowledge of the Dutch guidelines (as well as other international guidelines). The guidelines stress the importance of reassuring patients with CLBP that nothing is dangerously wrong and explaining patients that usual activities are not harmful. In general, we concluded, that the physiotherapists’ opinions toward CLBP are in line with these recommendations. However, most other items that reﬂect on diagnosis or treatment in more detail (e.g. ‘Knowledge of tissue damage is not necessary for effective treatment’ (item 3) or, ‘Not enough effort is made to ﬁnd underlying organic causes of CLBP’ (item 5), or ‘Patients suffering from severe back pain will beneﬁt from physical exercise’ (item 11) showed scores that were neither in strong agreement nor strong disagreement and there was a wide variation. A reason for this might be that on the one hand from their training, physiotherapists are more biomedically oriented, whereas on the other hand, the biopsychosocial model is advocated (Faas et al. 1996; Bekkering et al. 2001), thereby probably shifting the treatment orientation towards a more behavioural perspective. The wide variation indicates lack of r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 220 Manual Therapy

consistency with regard to exact details in the management of CLBP. Analyses The communalities as presented in Table 2 were low, which made the interpretation of the factor solution yielded by the factor analytical approach somewhat problematic. Because of these low communalities, the values of the loadings were not particularly high. Furthermore the internal consistency and explained variance of factor 1 was acceptable, but both were disappointing for factor 2. However, as we considered this as an initial impetus to developing a questionnaire and to cluster items in order to be able to assess whether there were underlying dimensions in the treatment orientation of physiotherapists, we continued this analysis. Because of these critical notes, the 2-factor model should be interpreted with caution, but we still think that because 2 factors emerged, these data suggested that both factors had to be taken into account. Furthermore, because of the correlation between the 2 factors, the results showed that a biomedical treatment orientation (factor 1) is not diametrically opposed to behavioural treatment orientation (factor 2). Despite the somewhat disappointing psychometric results, we considered it encouraging that the variables ‘Education Courses’, ‘Specialisation’ and ‘Work setting’ affected scores on both factors as anticipated. Women scored slightly higher on both factors, but although there was a statistically signiﬁcant difference on factor 2, we considered this small difference to be marginal. Previous low back pain experience revealed no differences in sum scores on the two factors. Physiotherapists over 42 years of age scored statistically signiﬁcantly higher on the biomedical orientation. This might possibly reﬂect the fact that the behavioural orientation is a relatively new development. Older physiotherapists may have had less opportunity to get acquainted with this approach. Comparison with other questionnaires To our knowledge, there is one other questionnaire concerning health care providers’ beliefs and attitudes. This ‘Health Care Providers’ Pain and Impairment Relationship Scale (HC-PAIRS)’ was developed by Rainville (Rainville et al. 1995). In both studies by Rainville (Rainville et al. 1995, 2000) in which attitudes and beliefs of health care providers were assessed, the HC-PAIRS was used. This scale was validated for community health care and rehabilitation providers. Compared to the HCPAIRS (Rainville et al. 1995) this study focuses more on diagnostic issues and issues that speciﬁcally address circumstances regarding treatment or advice r 2003 Elsevier Science Ltd. All rights reserved.

given to CLBP-patients by physiotherapists. Furthermore, the PABS PT has two subscales whereas the HC-PAIRS contains one scale only. Therefore, in our opinion, the PABS PT is more informative because it gives more information on how sum scores on the two subscales are achieved. In the case of one scale only, more combinations of items might lead to a similar sum score, thereby disguising how this score was achieved. Limitations The current study might have missed some important items because initially we screened patient questionnaires and attitudes and beliefs of patients are not necessarily the same as physiotherapists’ attitudes and beliefs. But as experienced physiotherapists in the ﬁeld of the management of chronic pain, who are also involved in training physiotherapists in cognitive behavioural approaches were involved in the development of the PABS PT, the possibility of missing important items was reduced. Moreover, Houben (Houben et al. 2001) showed that HC-PAIRS and PABS PT correlated highly, suggesting that there were no substantial omissions in the PABS PT as compared to the HC PAIRS. In the current study selection bias might have occurred because physiotherapists interested in cognitive behavioural approaches were more likely to return the questionnaire. However, the response rates of the random sample of KNGF-physiotherapists and the therapists who had attended the NPieducation courses were comparable. Moreover the demographic variables of the current study such as gender, age and work setting were comparable to the results of the study among Dutch physiotherapists by de Groot (de Groot et al. 1999). Therefore we think that the chance that selection bias affected the results of the current study is minimal. Another limitation might be the country in which this survey was performed. Transferring these results to other countries should be carried out with caution because deﬁnitions of specialization and education courses might differ greatly between countries. Before studying treatment orientations of physiotherapists in other countries by means of the PABS PT the characteristics of physiotherapists as used in the current study should be adapted. Future Although factor 1 showed satisfactory psychometric results, there certainly is room for improvement, especially for factor 2. Because of this further research is needed to improve the quality of this questionnaire. The reason that factor 2 needs further improvement is mainly due to the fact that some items that were initially formulated from the Manual Therapy (2003) 8(4), 214–222

ARTICLE IN PRESS Health-care provider’s attitudes and beliefs 221

biopsychosocial perspective were excluded from the factor analysis because most physiotherapists (dis)agreed to a great extent. To improve factor 2 in terms of internal consistency we suggest that future research should focus on items addressing biopsychosocial issues. To develop the PABS PT further we suggest that a variety of disciplines including for example chiropractors, orthopaedic surgeons or therapists highly specialized in treatment of chronic pain are incorporated, as they are more likely to achieve extreme scores. Amongst physiotherapists, the biopsychosocial model has recently gained much attention that could have modiﬁed treatment orientation. Probably there were no subjects holding extreme treatment orientations towards the management of CLBP. By including the above-mentioned disciplines one can assess whether the PABS PT can discriminate between different (more extreme) treatment orientations. For further investigation with regard to validity we suggest the use of external criteria. A possible external criterion could be the Photograph series of Daily Activities (PHODA) (Kugler et al. 1999). This instrument was developed in order to create a hierarchy of back-stressing activities reﬂecting the full range of situations that patients avoid, starting with those that provoke only mild discomfort, and ending with activities or situations well beyond that patient’s present abilities. Furthermore the impact of these different treatment orientations on the management of CLBP should be investigated by, for example, by assessing recommendations regarding treatment options for hypothetical patients, as described in patient vignettes. To examine the inﬂuence of these different treatment orientations with regard to chronic low back pain on patient outcome is also a challenge for the near future.

Acknowledgements The authors would like to thank Cobie Martens for her assistance with regard to data management. Furthermore we would like to thank Michael Schermer from The Knowledge Centre (NPi) for allied health professionals for his efforts with regard to the survey.

chronic pain. Archives of Physical Medicine and Rehabilitation 54: 399–408 Houben RMA, Ostelo RWJG, Vlaeyen JWS, Wolters PMJC, Peters M, Stomp-van den Berg SGM. 2001. Internal Report de Groot AWM, Hop JH, Meulenbeek AJG, Poot AA, van Praag BMS 1999 De betekenis van Fysiotherapie (The signiﬁcane of physiotherapy). Report for the Royal Dutch Society of Physiotherapy, Amsterdam (October 1999) Kerssens JJ, Sluijs EM, Verhaak PF, Knibbe HJ, Hermans IM 1999. Back care instructions in physical therapy: A trend analysis of individualized back care programs. Physical Therapy 79: 286–295 Kole-Snijders AMJ 1993 Validity of the Tampa Scale for Kinisifobia—Dutch Version (TSK—DV) for chronic low back pain patients. In: 7th World Congress on Pain, Paris Kugler K, Wijn J, Geilen M, de Jong J, Vlaeyen JWS 1999 The photograph series of daily activities (PHODA). Institute for Rehabilitation research and Academy for Physiotherapy, Heerlen, The Netherlands Lee PW, Chow SP, Lieh Mak F, Chan KC, Wong S 1989 Psychosocial factors inﬂuencing outcome in patients with low-back pain. Spine 14: 838–843 . I, Ohlund C, Eek C, Wallin L, Peterson LE, Fordyce Lindstrom WE, Nachemson AL 1992 The effect of graded activity on patients with subacute low back pain. Physical Therapy 72: 279–290 Miller RP, Kori SH, Todd DD 1991 The Tampa Scale. Tampa, FL Rainville J, Bagnall D, Phalen L 1995 Health care providers’ attitudes and beliefs about functional impairments and chronic back pain. The Clinical Journal of Pain 11: 287–295 Rainville J, Carlson N, Polatin P, Gatchel RJ, Indahl A 2000 Exploration of physicians’ recommendations for activities in chronic low back pain. Spine 25: 2210–2220 Ravensberg van CD, Klaveren van AJJ, Wams HWA, Elvers JHW, Oostendorp RAB, Hendriks HJM 1995 Variabelen in samenhang met het aantal behandelingen fysiotherapie. NPi, Amersfoort Sanders SH, Brena SF, Spier CJ, Beltrutti D, McConnell H, Quintero O 1992 Chronic low back pain patients around the world: cross-cultural similarities and differences. The Clinical Journal of Pain 8: 317–323 Sullivan MJL, Bishop SR, Pivik J 1995 The pain catastrophizing scale: Development and validation. Psychological Assessment 7: 524–532 Symonds TL, Burton AK, Tillotson KM, Main CJ 1996 Do attitudes and beliefs inﬂuence work loss due to low back trouble? Occupational Medicine 46: 25–32 Vlaeyen JWS, Kole Snijders AM, Boeren RG, Eek van H 1995 Fear of movement/(re)injury in chronic low back pain and its relation to behavioral performance. Pain 62: 363–372 Volinn IJ, 1989. Issues of deﬁnitions and their implications: AIDS and leprosy. Social Science & Medicine 29: 1157–1162 Waddell G. 1998. The Back Pain Revolution. Churchill Livingstone, Edinburgh, London, New York Waddell G, Newton M, Henderson I, Somerville D, Main CJ 1993 A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain 52: 157–168

References

APPENDIX

Bekkering GE, Hendriks HJM, Koes BW, Oostendorp RAB, Ostelo RWJG, Thomassen J, Tulder van MW 2001 KNGFrichtlijn Lage-rugpijn. Supplement bij Nederlands Tijdschrift voor Fysiotherapie 111: 1–24 Bernstein IH, Teng G 1989 Factoring items and factoring scales are different: Spurious evidence for multidimensionality due to item categerization. Psychological Bulletin 105: 476–477 Faas A, Chavannes AW, Koes BW, Van den Hoogen JMM, Mens JMA, Smeele LJM, Romeijnders ACM, Van der Laan JR 1996 NHG-Standaard Lage-Rugpijn. Huisarts en wetenschap 39: 18–31 Fordyce WE, Fowler Jr, RS, Lehmann JF, Delateur BJ, Sand PL, Trieschmann RB 1973 operant conditioning in the treatment of

In this appendix the Pain Attitude and Beliefs Scale for Physiotherapists (PABS PT) is presented as administered. Because the questionnaire needs further research we advocate administering all 31 items as listed below. Possibly the order of the items could inﬂuence the score. Instructions: The purpose of this list is to help us analyse how you, the therapist, approach the most common forms of back pain. We do not mean back pain resulting from a radicular syndrome, cauda

Manual Therapy (2003) 8(4), 214–222

r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 222 Manual Therapy

equina syndrome, fractures, infections, inﬂammation, a tumour or metastasis. It is not our intention to test your knowledge of back pain. We would simply like to know how you approach the treatment of back pain. We are looking for your opinion; the opinions of others are not relevant. Scoring method: We would like you to indicate the level to which you agree or disagree with each

statement. 1=‘totally disagree’, 2=‘largely disagree’, 3=‘disagree to some extent’, 4=‘agree to some extent’, 5=‘largely agree’, and 6=‘totally agree’. For factor 1 the range is from 14 through 84, and from 6 through 36 for factor 2. To calculate the score of factor 1 add the scores of items 4,5,9,10,13, 14,20,22,23,24,25,26,30 and 31. For factor 2 add the scores of items 3,6,7,11,12, and 27.

A.1. THE PAIN ATTITUDE & BELIEFS SCALE FOR PHYSIOTHERAPISTS (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31)

Back pain sufferers should refrain from all physical activity in order to avoid injury 1 2 3 4 5 6 Good posture prevents back pain 1 2 3 4 5 6 Knowledge of the tissue damage is not necessary for effective therapy 1 2 3 4 5 6 Reduction of daily physical exertion is a signiﬁcant factor in treating back pain 1 2 3 4 5 6 Not enough effort is made to ﬁnd the underlying organic causes of back pain 1 2 3 4 5 6 Mental stress can cause back pain even in the absence of tissue damage 1 2 3 4 5 6 The cause of back pain is unknown 1 2 3 4 5 6 Unilateral physical stress is not a cause of back pain 1 2 3 4 5 6 Patients who have suffered back pain should avoid activities that stress the back 1 2 3 4 5 6 Pain is a nociceptive stimulus, indicating tissue damage 1 2 3 4 5 6 A patient suffering from severe back pain will beneﬁt from physical exercise 1 2 3 4 5 6 Functional limitations associated with back pain are the result of psychosocial factors 1 2 3 4 5 6 The best advice for back pain is: ‘Take care’ and ‘Make no unnecessary movements’ 1 2 3 4 5 6 Patients with back pain should preferably practice only pain free movements 1 2 3 4 5 6 Back pain indicates that there is something dangerously wrong with the back 1 2 3 4 5 6 The way patients view their pain inﬂuences the progress of the symptoms 1 2 3 4 5 6 Therapy may have been successful even if pain remains 1 2 3 4 5 6 Therapy can completely alleviate the functional symptoms caused by back pain 1 2 3 4 5 6 If ADL activities cause more back pain, this is not dangerous 1 2 3 4 5 6 Back pain indicates the presence of organic injury 1 2 3 4 5 6 Sport should not be recommended for patients with back pain 1 2 3 4 5 6 If back pain increases in severity, I immediately adjust the intensity of my treatment accordingly 1 2 3 4 5 6 If therapy does not result in a reduction in back pain, there is a high risk of severe restrictions in the long term 1 2 3 4 5 6 Pain reduction is a precondition for the restoration of normal functioning 1 2 3 4 5 6 Increased pain indicates new tissue damage or the spread of existing damage 1 2 3 4 5 6 It is the task of the physiotherapist to remove the cause of back pain 1 2 3 4 5 6 There is no effective treatment to eliminate back pain 1 2 3 4 5 6 TENS and/or back braces support functional recovery 1 2 3 4 5 6 Even if the pain has worsened, the intensity of the next treatment can be increased 1 2 3 4 5 6 If patients complain of pain during exercise, I worry that damage is being caused 1 2 3 4 5 6 The severity of tissue damage determines the level of pain 1 2 3 4 5 6

r 2003 Elsevier Science Ltd. All rights reserved.

Manual Therapy (2003) 8(4), 214–222

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 223–232 r 2003 Elsevier Science Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00017-1

Original article

Reproducibility and repeatability: errors of three groups of physiotherapists in locating spinal levels by palpation E.V. Billis, N.E. Foster, C.C. Wright School of Health and Social Sciences, Coventry University, Priory Street, Coventry, UK

SUMMARY. Location of spinal levels by palpation is an essential part of physiotherapy assessment and treatment of spinal musculoskeletal conditions. This study aimed to (i) investigate the reproducibility and repeatability of three groups of physiotherapists in locating three randomly selected spinal levels by palpation, (ii) explore whether these groups were palpating similar points at each level, and (iii) investigate whether there were differences in palpation across the spinal levels tested. Reproducibility was estimated using a convenience sample of 30 physiotherapists: 13 undergraduate students, 10 clinicians, and seven manual therapists (MTs). Repeatability was investigated using six physiotherapists (two students, two clinicians and two MTs). Each therapist palpated the spinous processes (C5, T6, L5) once for reproducibility, and 10 times for repeatability, on each asymptomatic model. The skin overlying the spinal levels was marked with an invisible pen. Using an ultra-violet light, marks were transcribed onto transparent plastic strips for analysis, and the distances were measured with an electronic caliper. Repeated measures analyses of variance (ANOVA) indicated poor reproducibility across therapists at all spinal levels (F=18.43, P=0.001), but good repeatability within therapists (F=2.09, P=0.161). Students produced different mean locations of their palpatory marks from the other two groups in two spinal levels. Clinicians and MTs were more reproducible than students, and located similar levels. Palpation of L5 spinous process presented the most difﬁculty, for all groups. Further research is needed to compare different methods of palpation and explore whether reproducibility can be improved. r 2003 Elsevier Science Ltd. All rights reserved.

tion and location of symptomatic spinal levels, evaluation of patient progress and selection of appropriate treatment techniques (Jull 1994; Magarey 1994; Petty & Moore 1998). Over many decades, spinal palpation has developed in terms of both theory and practice (Russell 1983; Evans 1994). Physiotherapists must be able to locate individual vertebral levels accurately and consistently in order to perform a localized manual diagnostic test and compare palpatory ﬁndings between spinal levels or across treatment sessions (Simmonds & Kumar 1993; Binkley et al. 1995; McKenzie & Taylor 1997). Therefore, research which investigates physiotherapists’ precision in locating speciﬁc spinous processes is necessary. The reproducibility (inter-tester) and repeatability (intra-tester) of spinal (landmark) palpation has been explored in four studies over the last decade (Simmonds & Kumar 1993; Binkley et al. 1995; McKenzie & Taylor, 1997; Downey et al. 1999). These have shown acceptable levels of repeatability

INTRODUCTION An essential component of the physiotherapy assessment and treatment of spinal musculoskeletal problems is palpation (Simmonds & Kumar 1993). This ancient manual skill is extensively used for identiﬁcaReceived: 6 September 2002 Revised: 24 January 2003 Accepted: 11 February 2003 Evdokia V. Billis, BHSc, MSc, MCSP, MMACP, Clinical Physiotherapist (Athens, Greece), and part-time lecturer, Department of Physiotherapy, Technological Educational Institute (TEI), Lamia, Greece, Nadine E. Foster, BSc (Hons), DPhil, MCSP, Lecturer in the Department of Physiotherapy Studies, Faculty of Health, Keele University, Staffordshire, UK, Chris C. Wright, BSc (Hons), C. Math MIMA, FSS, Principal Lecturer in Research Methods, School of Health and Social Sciences, Coventry University, UK. Corresponding author: NEF, Department of Physiotherapy Studies, Keele University, Staffordshire ST5 5BG, UK. Tel: +44 1782 584 195; Fax: +144 01782 584 255; E-mail: [email protected] 223

ARTICLE IN PRESS 224 Manual Therapy

within each physiotherapist, whereas, reproducibility across therapists was poor in three of the four studies. Similar ﬁndings have also been published in osteopathic (Burton et al. 1990; O’Haire & Gibbons 2000) and chiropractic literature (Byﬁeld et al. 1992). However, when examining the studies available, several issues require consideration. Research exploring the effects of clinical experience and postgraduate training in palpation has produced conﬂicting results. Some studies have demonstrated a higher level of consistency by experienced manual therapists (MTs) (Jull & Bullock 1987; Jull et al. 1988; Trott et al. 1989; Jull 1994), while others have not found clinical experience or training to affect palpatory precision (Potter & Rothstein 1985; Maher & Adams 1994). Simmonds and Kumar (1993) utilized a convenience sample of 20 physiotherapists (undergraduate students and clinicians) whose experience in palpation was not clearly deﬁned. In McKenzie and Taylor’s (1997), study the 14 experienced therapists participating in lumbar palpation did not have any postgraduate manual therapy qualiﬁcations. Both of these studies demonstrated poor consistency between therapists, which was partly attributed to the lack of therapists’ clinical experience and postgraduate training. Experience and training however, did not appear to be the reason for the poor results in Binkley et al.’s (1995) study in which inter-tester reliability was explored by six experienced physiotherapists, three of whom had postgraduate training in manual therapy. Therapists were requested to identify, by palpation, lumbar spinous processes of 18 low back pain patients. Their poor results (51% agreement, a generalized kappa value of 0.30 and an intraclass correlation coefﬁcient of 0.69) were mostly attributed to the lack of standardization of testing procedures. In the most recent study of physiotherapists, Downey et al. (1999) investigated the inter-tester reliability of three pairs of specialist MTs in locating randomly selected lumbar spinal levels on 20 low back pain patients. All MTs had more than 7 years of clinical experience, and at least 3 years since completing their postgraduate manual therapy qualiﬁcations. In contrast with previous studies, Downey et al. (1999) concluded good inter-tester reliability for each pair (weighted kappa values ranged from 0.86–0.98), which was attributed to the therapists’ postgraduate training. However, their sample was relatively small and consisted of MTs from the same practice (possibly utilizing similar palpatory methods), thus, limiting the generalizability of their results. In addition, they did not compare their sample with other groups of different experience and training. Therefore, the role which clinical experience and postgraduate training plays in the precision of spinal palpation requires further research. A second issue requiring consideration is that of the palpation itself. Previous studies have included r 2003 Elsevier Science Ltd. All rights reserved.

mostly lumbar and pelvic landmark palpation, while precision in cervical and thoracic landmark palpation has not been investigated. In trying to establish repeatability, the number of repeated trials is important. Only two studies have explored the repeatability of intra-tester measurements in locating spinal landmarks for palpation. Simmonds and Kumar (1993) demonstrated good intra-tester results, yet only two repeated trials were performed by each therapist. McKenzie and Taylor (1997) performed ﬁve repeated trials on each model, however, only three physiotherapists took part. In addition, the statistical analyses of available studies require further consideration. Most studies have assessed ‘reliability’ using different statistical procedures. Several authors suggest the intraclass correlation coefﬁcient (ICC) to be the preferred index of reliability (Shrout & Fleiss 1979; Haas 1991a, b; 1995). However, values of the ICC do not reﬂect the expected magnitude of measurement error and can, therefore, be difﬁcult to interpret for clinical measurements. Such information is provided by the precision of measures or, as in the case of spinal palpation, through reproducibility and repeatability limits, respectively (Mason et al. 1989; Altman 1991) where reproducibility limits express the limits of variability that can be expected on the difference in independent measurements made on a subject by different testers, and repeatability limits express the limits of variability that can be expected on the difference between repeated, independent measurements made on a subject by the same tester (Mason et al. 1989). Therefore, for the purposes of this study, reproducibility and repeatability limits were selected to describe the consistency of identifying spinal levels. The aims of this study were to investigate reproducibility and repeatability of physiotherapists in locating three spinal levels (one cervical, one thoracic and one lumbar) by palpation, and to take into account two factors that have not been sufﬁciently investigated in the previous literature; (i) comparison between clinical experience and postgraduate manual therapy training of therapists, and (ii) consistency of palpation across the different spinal regions tested. Therefore, the study explored: *

*

*

whether there were any differences in the intertester and intra-tester variability of palpatory marks across three physiotherapy groups (with varying degrees of experience) or across the three spinal levels, whether there were any differences in mean location of the palpatory marks across the three therapist groups, at each spinal level (which would reﬂect whether the therapist groups were locating a different point, at each spinal level), and reproducibility and repeatability limits of a given spinal level. Manual Therapy (2003) 8(4), 223–232

ARTICLE IN PRESS Locating spinal levels by palpation 225

METHODS Subjects Physiotherapists Following ethical approval from Coventry University’s Health and Social Sciences Ethics Committee and written consent, 30 physiotherapists participated in the reproducibility (inter-tester component) of the study, as follows: (i) Group 1: Thirteen ﬁnal year undergraduate students from the University who were all familiar with spinal palpation, (ii) Group 2: Ten clinicians, who had a minimum of 2 years experience in spinal musculoskeletal physiotherapy (range: 2–12 years, mean: 4.2 years), and had no accredited postgraduate manual therapy qualiﬁcations, and (iii) Group 3: Seven MTs who had accredited postgraduate training in manual therapy (their qualiﬁcations were recognized by the Manipulation Association of Chartered Physiotherapists), and a minimum of 5 years experience in spinal physiotherapy (range: 5–8 years, mean: 6.1 years). For the repeatability (intra-tester component), six physiotherapists participated as raters; two ﬁnal year students, two clinicians with no accredited training in manual therapy (mean clinical experience: 3.7 years), and two MTs (mean clinical experience: 5.5 years). The reasons for having two different groups for the repeatability and reproducibility elements of the study were largely related to practical and resource issues. Subjects involved in the reproducibility element could not attend on the scheduled date for the repeatability element, for example, some of the qualiﬁed staff were at work and some of the students had to attend classes. Models Nine undergraduate students from the University participated as models; ﬁve participated in the reproducibility and four in the repeatability component of the study. All were young (range: 18–23 years, mean age: 20.2 years), healthy individuals with no known spinal abnormality or history of spinal problems. Each model provided written consent to participate in this study. Testing procedure Three spinous processes, C5, T6 and L5, were randomly selected for palpation, by taking numbers from three boxes, each box corresponding to a spinal region (cervical, thoracic and lumbar). Manual Therapy (2003) 8(4), 223–232

Reproducibility Each of the ﬁve models was positioned in prone lying on a plinth, the position which most closely resembles spinal palpation in the clinical setting (Magarey 1994; Maitland et al. 2001). Therapists were randomly allocated to each experimental cubicle, and requested to palpate C5, T6 and L5 spinous processes using their own palpatory methods1 and to mark the middle of each spinous process with a small dot, using a pen (WH Smith Security Marker), which was visible only under ultra-violet (UV) light. This method has been utilized in previous studies (Burton et al. 1990; Simmonds & Kumar 1993; McKenzie & Taylor 1997; Downey et al. 1999). All therapists had the opportunity to practice invisible skin marking on a volunteer model, prior to performing the palpations. No feedback was permitted between testers and models and, on completion of each palpation, therapists remained in their cubicles until permission to move onto the next cubicle and model was given. Repeatability The procedure was identical to that of the reproducibility investigation. Each therapist had to palpate and mark each spinous process 10 times, with a 30 min break between two consecutive trials (palpation and marking) on the same model and spinal level. It therefore took 5–6 h for each therapist to complete the 10 repeated palpation trials on each model. Recording procedure Palpatory marks (or dots) were identiﬁed on each model’s back using a small low emission UV light (Model Number 23300, Ring Lighting, Leeds, England). To allow analysis, the palpatory marks were transcribed from each model’s back onto a transparent plastic strip (Melinex Clear with a thickness of 50 mm; Number 61082). Each model’s prone position was standardized (arms resting either side of their body, face at the aperture of the plinth and head position standardized by placing a 400gr cylindrical tin of diameter 7.5 cm between the model’s chin and chest) and the accurate alignment of the plastic strip on each model’s back was established in the pilot study. None of the models reported problems in maintaining or resuming this position during the study. To transfer the dots onto a strip, the investigator identiﬁed and marked with permanent skin ink, six reference marks (two cervical, two thoracic, and two lumbar) on each model’s back (each two marks lying 1

There were three different methods of palpation used to locate C5 and T6; these were via initial palpation of the occiput or C7 and T1 spinous processes or C6 spinous process. There were four different palpation methods used to locate L5; these were via initial palpation of the iliac crests, the sacrum, the 12th rib or the occiput. r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 226 Manual Therapy

approximately 10 cm bilaterally from the spine). The plastic strip was then applied and taped along the middle of each model’s back from the sacrum to the occiput, and the six reference marks were replicated onto the strip. Using the UV light, all dots were identiﬁed and traced from the models’ backs onto the strip with a permanent pen. The strip was then removed and the invisible skin markers were erased with isopropanol, leaving only the reference marks to enable accurate relocation of the strip. This method of application and alignment of the strip was repeated (n ¼ 20) in the pilot study by the investigator; the strip was aligned accurately, on all repeated trials without problems, thus showing the researcher’s consistency in application of the strip, and in reproducing the standardized, recording position. For the reproducibility component, one strip was used for each model, and a different colour permanent pen was used for each of the therapist group’s marks. For the repeatability component, one strip was used for each therapist’s repeated trials on each model. Once all dots were traced onto a strip for each model, distances were measured using an electronic caliper with a measurement error of 70.01 mm (Kennedy 331-133 Digimatic Caliper, Kennedy Tools, England). To calculate the distance between the marks, a parallelogram was plotted on the plastic strip (for each spinal level), along the most distal palpatory points, and the distance of each palpatory mark was calculated (see Fig. 1). Thus, this ‘central line’ for measurements was an external point of reference situated somewhere in the centre of all palpatory marks. For the reproducibility component, 15 parallelograms were made; three for each model, each one corresponding to a different spinal level. For the repeatability component, as each of the six therapists were investigated separately, one parallelogram was plotted for each therapist’s repeated trials at each spinal level. Data analysis All analyses were performed using the Statistical Package for the Social Sciences (SPSS, Version 9.0 for Windows) and Excel (97). All statistical tests were performed at the 5% level of signiﬁcance. Data were summarized using boxplots, means, standard deviations and ranges of measurements. Spinous processes are large structures with considerable surface area, especially in the lumbar region, and they vary much in size, shape and direction (Williams et al. 1989). No comparative measures of the length or surface area of the tips of the spinous processes, which can actually be palpated through the skin, could be found in the literature. Key texts on palpation (e.g. Field 1994) explain that, for example, the spines of C3, 4 and 5 are closely packed together due to the curvature of r 2003 Elsevier Science Ltd. All rights reserved.

Fig. 1—The parallelogram method of calculating the distances between palpatory marks. The plotted parallelogram enclosed all the marks and had its boundaries lying along the most peripherally marked points. A line was drawn parallel to the line joining the PSISs of the model, which passed through the middle of the parallelogram (central line). A negative value was given for points situated below the central line and a positive value for points above it. Each mark was measured by its vertical distance to the central line; see distances A and B in ﬁgure. The different colour palpatory marks represent the different therapist groups (black marks for the students, blue for the clinicians, and red for the MTs).

the spine, but do not go so far as to explain how much of the tip of the spinous process should be palpable. The only available data are the anatomical cadaveric measurements provided by Panjabi et al. (1991a, b) and McKenzie and Taylor (1997)2 but these cannot be equated to the length of the tips of the spinous processes that can be palpated through the skin during a manual therapy examination. A two-factor repeated measures ANOVA was used to test for differences in variability in location across groups of therapists and across spinal levels. The analysis was performed on the natural logarithms of variances (Winer 1971, Montgomery 2001) computed for each combination of spinal level, therapist group and model. Further analyses were performed using within-model contrasts to identify which therapist group(s) and spinal level(s) were associated with signiﬁcantly different variances. The same analyses were performed on the intra-tester variances across therapist groups and spinal levels. A repeated measures ANOVA was used to test for inter-therapist group differences in mean location of the palpatory marks at each spinal level. The analysis was performed on mean distances computed for each 2

The length of the spinous process of C5 is 28.5mm, measured from the lower posterior tip to the vertebral body centroid, T6 is 53.8mm, measured from the centroid of the superior end-plate to the most inferoposterior tip and L5 is 16.4mm (no details are provided of how this was measured) Manual Therapy (2003) 8(4), 223–232

ARTICLE IN PRESS Locating spinal levels by palpation 227

combination of therapist group and model, at each spinous process. At each spinal level, further analyses were performed using within-model contrasts to identify the group(s) associated with signiﬁcantly different mean palpatory locations. Based on the results from the ANOVA on variances, 95% reproducibility limits were computed for each spinal level and each therapist group, using formulae derived from Mason et al. (1989) and Haas (1995). These limits provide an interval estimate for the true population reproducibility. Where zero appears within the interval, this means that the true difference in independent palpations could be zero, which would indicate good reproducibility. Similarly, 95% repeatability limits were computed for each spinal level.

RESULTS Reproducibility Tables 1 and 2 summarize the inter-tester palpatory marks and the boxplot in Fig. 2 provides a visual illustration of differences in palpatory marks across therapist groups and spinal levels. Statistically signiﬁcant differences were found between the variances of palpatory marks across therapist groups (F ¼ 18:43; P ¼ 0:001) and spinal levels (F ¼ 11:43; P ¼ 0:005). The variance of palpatory marks was signiﬁcantly larger for students than for clinicians (P ¼ 0:043) and MTs (P ¼ 0:001), and for spinous process T6 compared with C5 (P ¼ 0:009). Statistically signiﬁcant differences were also found between the mean palpatory marks across therapist groups for spinous processes C5 (F ¼ 10:53; P ¼ 0:006), T6 (F ¼ 5:37; P ¼ 0:033) and L5 (F ¼ 6:09; P ¼ 0:025). The mean palpatory mark for students was signiﬁcantly different from that for clinicians (P ¼ 0:028) and MTs (P ¼ 0:031) at C5, and from MTs (P ¼ 0:021) at L5. Whilst no other signiﬁcant differences were found at the 5% level, differences between the mean palpatory mark for students and clinicians at L5 (P ¼ 0:057) and at T6

Table 2. Inter-tester mean palpatory marks for spinous processes according to physiotherapy group Spinous process Therapist group

C5 Mean

T6 Mean

L5 Mean

Students (n ¼ 65) Clinicians (n ¼ 50) Manual therapists (n ¼ 35)

4.4n 26.0 23.9

13.2 37.8 33.6

9.5w 2.5 4.8

* Signiﬁcantly different mean at spinous process C5 for students compared with clinicians (P ¼ 0:028) and manual therapists (P ¼ 0:031). w Signiﬁcantly different mean at spinous process L5 for students compared with manual therapists (P ¼ 0:021). This value is a minus number because it was below the central line of the parallelogram. All other values are positive and were therefore above the central line. All measurements are in mm.

(P ¼ 0:059) are worth a mention. Hence, there was no evidence of a difference in mean location of C5, T6 or L5 for clinicians compared with MTs, but there was some evidence that students produced different mean locations than clinicians and MTs. Approximate 95% reproducibility limits, for each therapist group, at each spinal level, are given in Table 3. These are based on the smallest differences in palpatory marks on any of the ﬁve models, between any two testers within the same therapist group. It can be seen that the limits of variability expected on the difference in measurements made independently on a model by two testers (from the same therapist group) is relatively large. However, the actual difference might be zero (since zero is contained within the limits). In general, clinicians and MTs were more reproducible in the cervical and thoracic regions than students. No clear difference in performance was found in the lumbar region, indicating greatest difﬁculty in palpating L5 spinous process.

Repeatability Table 4 provides an example of each tester’s repeated palpatory marks on one of the four models (Model 4), and the boxplot in Fig. 3 gives a visual illustration of repeatability. Statistically signiﬁcant differences were

Table 1. Inter-tester standard deviation (SD) and range of palpatory marks for spinous processes according to physiotherapy group Spinous process T6n

C5

L5

Therapist group

SD

(Range)

SD

(Range)

SD

(Range)

Studentsw (n ¼ 65) Clinicians (n ¼ 50) Manual therapists (n ¼ 35)

22.5 15.1 15.2

(103.0) (30.5) (29.9)

44.1 32.4 28.2

(177.1) (109.5) (119.1)

16.3 16.9 16.0

67.2 (60.6) (52.5)

*Signiﬁcantly larger variance (P ¼ 0:009) at spinous process T6 compared with C5. w Signiﬁcantly larger variance than clinicians (P ¼ 0:043) and manual therapists (P ¼ 0:001). All measurements are in mm. Manual Therapy (2003) 8(4), 223–232

r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 228 Manual Therapy

Fig. 2—Boxplot to show inter-therapist group palpation marks at each spinal level. O denote marks which were ‘deviant’ based on the middle 50% of marks (represented by the box); where ‘deviant’ represents more than 1.5 box lengths from the middle 50% of marks.

Table 3. 95% reproducibility limits for each therapist group at each spinal level and 95% repeatability limits at each spinal level Spinous process C5

T6

L5

57.33 to 57.41 23.30 to 23.32 20.81 to 22.13

121.22 to 121.44 78.02 to 78.04 74.76 to 74.76

42.77 to 42.77 44.90 to 44.90 40.14 to 40.14

12.71 to 12.71

19.55 to 19.55

25.28 to 25.28

n

95% reproducibility limits Therapist group Students Clinicians Manual therapists 95% Repeatability limitsw *

Reproducibility limits based on the smallest difference between any two testers in the same therapist group, for each spinal level. Repeatability limits based on the smallest difference between any two repeat measurements for the same tester, for each spinal level. All measurements are in mm.

w

found between the intra-tester variances of palpatory marks across spinal levels (F ¼ 6:98; P ¼ 0:008), but not across therapists (F ¼ 2:09; P ¼ 0:161). No signiﬁcant interaction between spinal level and tester was found (F ¼ 2:63; P ¼ 0:104). The variance of palpatory marks for L5 was signiﬁcantly larger than for C5 (P ¼ 0:014), and arguably for T6 (P ¼ 0:051). There was no signiﬁcant difference between the intratester variances for palpatory marks for C5 and T6 (P ¼ 0:125). No consistent difference in performance of the six therapists was found on repeated measurements in the cervical, thoracic or lumbar regions, across the four models. Approximate 95% repeatability limits, at each spinal level, are given in Table 3. These are based on the smallest differences in palpatory marks on any of the four models, for the same therapist. The limits of variability expected on the difference between two repeat, independent measurements by the same r 2003 Elsevier Science Ltd. All rights reserved.

therapist are relatively small for C5 and T6, but larger for L5. This indicates that the greatest errors occurred in therapists’ repeated palpations of the L5 spinous process.

DISCUSSION Reproducibility Results demonstrated statistically signiﬁcant differences in the variances of palpatory marks across the three groups of therapists (P ¼ 0:001) and across the three spinal levels (P ¼ 0:005), indicating fairly poor reproducibility at all spinal levels. The ﬁndings are consistent with most previous studies (Simmonds & Kumar 1993; Binkley et al. 1995; McKenzie & Taylor 1997). Manual Therapy (2003) 8(4), 223–232

ARTICLE IN PRESS Locating spinal levels by palpation 229 Table 4. Intra-tester standard deviation (SD) and range of palpatory marks for the repeated measurements by each tester, at each spinal level, for one model (Model 4)n Spinous process Tester

C5 SD (Range)

T6 SD (Range)

L5w SD (Range)

Student 1 Student 2 Clinician 1 Clinician 2 Manual therapist 1 Manual therapist 2

3.97 7.16 2.25 2.71 0.97 2.27

2.05 12.72 5.45 3.22 3.69 2.30

16.41 15.47 18.74 4.52 7.12 4.39

(11.99) (19.42) (7.90) (9.06) (3.07) (6.02)

(6.30) (30.48) (16.32) (9.77) (11.12) (7.11)

(46.47) (39.12) (53.73) (16.26) (22.58) (15.33)

n

The above ﬁgures are provided for one of the four models only as an example. w signiﬁcantly larger variance at spinous process L5 compared to C5 (P ¼ 0:014) and T6 (P ¼ 0:051). All measurements are in mm.

The differences between physiotherapy groups appeared to be consistent across the spinal levels. The variance of palpatory marks for the students was signiﬁcantly larger than for clinicians (P ¼ 0:043) and MTs (P ¼ 0:001), whereas no signiﬁcant difference was found between clinicians and MTs (P ¼ 0:056). This indicates that clinicians and MTs were more similar in palpation than students and their spinal marks were more reproducible. Perhaps the results reﬂect what would have been expected (i.e. that students’ palpation would be more variable). Clinicians and MTs however, were found to have no statistically signiﬁcant differences between the variance of their palpatory marks. Some authors argue

that the level of experience and training does not affect palpatory precision (Potter & Rothstein 1985; Maher & Adams 1994; Binkley et al. 1995), whilst others suggest that students receiving the same palpatory training become consistent in their manual skills (Hardy & Napier 1991). Experienced clinicians and MTs, on the other hand, are used to incorporating other palpatory cues i.e. pain provocation, resistance to movement (Gonnella et al. 1982; Hardy & Napier 1991), which were not investigated in this study. Perhaps pain and other palpatory cues are more important in guiding therapists during clinical examination, than speciﬁc landmark palpation (Jull et al. 1994, 1997; Christensen et al. 2002). However, despite less variation amongst clinicians and MTs (compared to the students), the ranges (and SDs) of all groups were large, with maximum ranges of 103.0 mm in the cervical spine (students), 177.1 mm in the thoracic spine (MTs), and 60.6 mm in the lumbar spine (clinicians). These ﬁgures are larger that those of previous studies; for example, the largest mean inter-tester distance between marks was 37.3 mm in the McKenzie and Taylor (1997) study, and 44 mm in Downey et al.’s (1999) study. Large discrepancies between the palpatory marks are also evident with the 95% reproducibility limits (Table 3). Although the exact reasons for the results cannot be deﬁnitively established, the use of ‘free palpation’ may have contributed as suggested in previous studies (Simmonds & Kumar 1993; McKenzie & Taylor

Fig. 3—Boxplot to show intra-therapist repeatability at each spinal level, for each therapist. O denote marks which were ‘deviant’ based on the middle 50% of marks (represented by the box); where ‘deviant’ represents more than 1.5 box lengths from the middle 50% of marks. *denote marks which were ‘very deviant’ based on the middle 50% of marks; where ‘very deviant’ represents more than 3 box lengths from the middle 50% of marks. Manual Therapy (2003) 8(4), 223–232

r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 230 Manual Therapy

1997; O’Haire & Gibbons 2000). Therapists were permitted to use their own methods of palpation, using their own reference points to identify each landmark. For example, L5 palpation was initiated by some testers from the iliac crests and by others from the sacrum. Therefore, performance depended upon the anatomical integrity of these other landmarks. Since it is known that anatomical variations exist within asymptomatic populations (MacLean et al. 1990; Grieve 1994), it is possible that different methods of palpation resulted in different landmark identiﬁcation. However, as there is no ’gold standard’ method for spinal palpation, the palpatory methods utilized in the current study, reﬂect those used in current physiotherapy practice. The differences between the current study and previous ones could be attributed, to some extent, to differences in methodology. Firstly, the ranges in most previous studies are not reported; only mean distances between marks are given. Secondly, the methods of measuring the distances were different; previous studies arbitrarily took one palpatory mark as the zero point without providing any information on its location. The current study utilized a point (the ‘central line’), which was an external point and not a therapist’s mark. Thirdly, all previous studies used smaller physiotherapy sample sizes than the current one. While there were no statistically signiﬁcant differences in mean location of C5, T6 or L5 for clinicians compared with MTs (P > 0:05), there was some evidence that students produced different mean locations than clinicians and MTs in the cervical and lumbar region. In the clinician and MT groups, no statistically signiﬁcant results were found. This provides some evidence that these two groups were indeed locating similar spinal levels, on average. It was evident that the L5 spinous process was the most difﬁcult to palpate. Although close to prominent landmarks such as the posterior superior iliac spines and the iliac crests, L5 has a deep location (Ebraheim et al. 1999) and is the smallest of the three studied (Grieve 1988; Oliver & Middleditch 1991). The ability to palpate a spinous process is considered a basic skill and a prerequisite for other manual therapy techniques (Jull 1994; Magarey 1994; Petty & Moore 1998). If therapists are unable to reproduce the same spinous process by palpation, it is not unreasonable to assume that other spinal manual therapy techniques (based on palpation) would lack reproducibility, too. In addition, other spinal structures which are believed to be more difﬁcult to palpate (e.g. transverse processes), are likely not to be palpated consistently. This lack of agreement among physiotherapists highlights the need for investigation of methods of palpation in order to establish which are the most reproducible. r 2003 Elsevier Science Ltd. All rights reserved.

Repeatability There were no statistically signiﬁcant differences in the variability of palpatory marks within therapists (P ¼ 0:161), indicating good repeatability. This is also evident by the smaller ranges and SDs of the palpatory marks (Table 4). The 95% repeatability limits at each spinal level illustrated that the limits of variability expected on the difference between two repeat, independent measurements by the same therapist tended to be small for C5 and T6 but larger for L5. This indicates inconsistency in the therapists’ repeated palpations for L5 spinous process (possibly due to the reasons discussed earlier). This inconsistency is also evidenced by the statistically signiﬁcant differences found in the intra-tester variances of palpatory marks across L5 spinous process. Although there were no differences between C5 and T6, the palpation of L5 spinous process produced signiﬁcantly larger variation than C5 (P ¼ 0:014) and T6 (P ¼ 0:051). In general the ﬁndings are in agreement with previous studies (Simmonds & Kumar 1993; McKenzie & Taylor 1997). Although the distances between marks in the current study appear to be of a larger magnitude, none of the previous studies performed as many repeated trials. The good repeatability indicates that physiotherapists are able to return to the same spinal level within treatment sessions or during reassessment procedures, which is of importance since assessment and treatment is frequently based on reevaluation of palpatory ﬁndings (Magarey 1994; Petty & Moore 1998). It must be acknowledged, however, that the current study focused on reliability. In the clinical setting, it would be ideal to be able to regard certain palpation procedures as accurate. It was not possible to assess validity in the current study, for example through comparison with radiological ﬁndings, so it is not known whether physiotherapists did indeed palpate the correct spinal levels or if the models had the same number of vertebrae within each spinal region. To determine if therapists were locating the correct spinal level, procedures such as X-ray would have been required, which was not within the scope of this study.

CONCLUSION Taking into account the limitations of the study, the research indicated that physiotherapists have fairly poor reproducibility and good repeatability in locating C5, T6 and L5 spinous processes by palpation. The reproducibility data demonstrated that students were less consistent than clinicians and MTs in palpation, whereas there were no differences between clinicians and MTs. Students produced different Manual Therapy (2003) 8(4), 223–232

ARTICLE IN PRESS Locating spinal levels by palpation 231

mean locations from clinicians and MTs in C5 and L5 spinous processes and were thus locating a different point at these spinal levels. In addition, for both the reproducibility and repeatability component of the study, for all groups, the L5 spinous process presented most difﬁculty in palpation. Further research is needed to explore different methods of spinal palpation.

Acknowledgements We would like to thank all physiotherapists and subjects who participated in this study. The ﬁrst author would also like to acknowledge the State Scholarships Foundation (IKY), Republic of Greece, for ﬁnancing this study in partial fulﬁllment of an MSc in Manipulative Therapy.

References Altman DG (1991) Some common problems in medical research. In: Altman DG (ed). Practical Statistics for Medical Research. Chapman & Hall, London, Ch 14 pp 396–409 Binkley J, Stratford PW, Gill C 1995 Interrater Reliability of Lumbar Accessory Motion Mobility Testing. Physical Therapy 75(9): 786–795 Burton KA, Edwards VA, Sykes DA 1990 ‘Invisible’ skin marking for testing palpatory reliability. Journal of Manual Medicine 5: 27–29 Byﬁeld DC, Mathiasen J, Sangren C 1992 The reliability of osseous landmark palpation in the lumbar spine and pelvis. European Journal of Chiropractice 40: 83–88 Christensen HW, Vach W, Vach K, Manniche C, Haghfelt T, Hartvigsen L, Hoilund-Carlsen PF 2002 Palpation of the upper thoracic spine: An observer reliability study. Journal of Manipulative and Physiological Therapeutics 25(5): 285–292 Downey BJ, Taylor NF, Niere KR 1999 Manipulative physiotherapists can reliably palpate nominated lumbar spinal levels. Manual Therapy 4(3): 151–156 Ebraheim NA, Inzerillo C, Xu R 1999 Are anatomic landmarks reliable in determination of fusion level in posterolateral lumbar fusion? Spine 24(10): 973–974 Evans DH 1994 The reliability of assessment parameters: Accuracy and palpation technique. In: Boyling JD, Palastanga N (eds) Grieve’s Modern Manual Therapy, 2nd edn. Churchill Livingstone, Edinburgh, Ch 49, pp 539–546 Field D 1994 Anatomy Palpation and Surface Markings. Butterworth Heinemann Ltd, London, Ch 4 p 128 Gonnella C, Paris SV, Kutner M 1982 Reliability in evaluating passive intervertebral motion. Physical Therapy 62(4): 436–444 Grieve GP 1988 Applied anatomy-regional. In: Grieve GP (ed) Common Vertebral Joint Problems. Churchill Livingstone, Edinburgh, Ch 1, pp 1–77 Grieve GP 1994 Bony and soft-tissue anomalies of the vertebral column. In: Boyling JD, Palastanga N (eds) Grieve’s Modern Manual Therapy, 2nd edn. Churchill Livingstone, Edinburgh, Ch 17 pp 227–250 Haas M 1991a Statistical methodology for reliability studies. Journal of Manipulative and Physiological Therapeutics 14(2): 119–132 Haas M 1991b The reliability of reliability. Journal of Manipulative and Physiological Therapeutics 14(3): 199–208 Haas M 1995 How to evaluate interexaminer reliability using an interexaminer reliability study design. Journal of Manipulative and Physiological Therapeutics 18(1): 10–15 Hardy GL, Napier JK 1991 Inter and intratherapist reliability of passive accessory movement technique. New Zealand Journal of Physiotherapy 19(3): 22–24 Manual Therapy (2003) 8(4), 223–232

Jull GA 1994 Examination of the articular system. In: Boyling JD, Palastanga N (eds) Grieve’s Modern Manual Therapy, 2nd edn. Churchill Livingstone, Edinburgh, Ch 37 pp 511–527 Jull GA, Bogduk N, Marsland A 1988 The accuracy of manual diagnosis for cervical zygapophyseal joint syndromes. The Medical Journal of Australia 148(7): 233–236 Jull GA, Bullock M 1987 A motion proﬁle of the lumbar spine in an ageing population assessed by manual examination. Physiotherapy Practice 3: 70–81 Jull GA, Treleaven J, Versace G 1994 Manual examination: Is pain provocation a major cue for spinal dysfunction? Australian Journal of Physiotherapy 40(3): 159–165 Jull GA, Zito G, Trott P, Potter H, Shirley D, Richardson C 1997 Inter-examiner reliability to detect painful upper cervical joint dysfunction. Australian Journal of Physiotherapy 43(2): 125–129 MacLean JGB, Tucker JK, Latham JB 1990 Radiographic appearances in lumbar disc prolapse. Journal of Bone and Joint Surgery 72B(5): 917–920 Magarey ME 1994 Examination of the cervical and thoracic spine. In: Grant R (ed) Physical Therapy of the Cervical and Thoracic Spine, 2nd edn. Churchill Livingstone, New York, Ch 7 pp 109–144 Maher C, Adams R 1994 Reliability of pain and stiffness assessments in clinical manual lumbar spine examination. Physical Therapy 74(9): 801–811 Maitland GD, Hengeveld E, Banks K, English K 2001 Assessment. Cervical spine. Thoracic spine. Lumbar spine. In: Maitland GD, Hengeveld E, Banks K, English K (eds) Maitlands Vertebral Manipulation, 6th edn. Butterworth-Heinemann, Oxford, Chs 10–12 pp 227–283 Mason RL, Gurst RF, Hess JL 1989 Combinations of ﬁxed and random effects. In: Mason RL, Gurst RF, Hess JL (eds) Statistical Design and Analysis of Experiments: With Applications to Engineering and Science. John Wiley & Sons, New York, Ch 17 pp 360–364 McKenzie AM, Taylor NF 1997 Can physiotherapists locate lumbar spinal levels by palpation? Physiotherapy 83(5): 235–239 Montgomery DC 2001 Discovering dispersion effects. In: Montgomery DC (ed). Design and Analysis of Experiments, 5th edn. John Wiley, New york, Section 3.8, p 111 O’Haire C, Gibbons P 2000 Inter-examiner and intra-examiner agreement for assessing sacroiliac anatomical landmarks using palpation and observation: Pilot study. Manual Therapy 5(1): 13–20 Oliver J, Middleditch A 1991 Structure of the vertebral column. In: Oliver J, Middleditch A (eds) Functional Anatomy of the Spine. Butterworth-Heinmann, London, Ch 1, p 2 Panjabi MM, Duranceau J, Goel V, Oxland T, Takata K 1991a Cervical human vertebrae. Quantitative three-dimensional anatomy of the middle and lower regions. Spine 16(8): 861–869 Panjabi MM, Takata K, Goel V, Federico D, Oxland K, Duranceau J, Krag M 1991b Thoracic human vertebrae. Quantitative three-dimensional anatomy. Spine 16(8): 888–901 Petty NJ, Moore AP 1998 Physical examination. In: Petty NJ, Moore AP (eds) Neuromusculoskeletal Examination and Assessment. Churchill Livingstone, Edinburgh, Ch 3 pp 31–97 Potter NA, Rothstein JM 1985 Intertester reliability for selected clinical tests of the sacroiliac joint. Physical Therapy 65(11): 1671–1675 Russell R 1983 Diagnostic palpation of the spine: A review of procedures and assessment of their reliability. Journal of Manipulative and Physiological Therapeutics 6(4): 181–183 Shrout PE, Fleiss JL 1979 Interclass correlations and uses in assessing rater reliability. Psychological Bulletin 86: 420–428 Simmonds MJ, Kumar S 1993 Health care ergonomics. Part II: Location of bony structures by palpation – a reliability r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 232 Manual Therapy

study. International Journal of Industrial Ergonomics 11: 145–151 Trott PH, Evans DH, Frick R 1989 Accuracy of manual palpation skills: The ability of manipulative physiotherapists to detect changes in force and displacement on a palpation simulator. Proceedings of Manipulative Therapist Association of Australia, Adelaide, pp 207–214

r 2003 Elsevier Science Ltd. All rights reserved.

Williams PL, Warwick R, Dyson M, Bannister LH (eds) 1989. Gray’s Anatomy, 37th edn. Churchill Livingston, London, Ch 3, p 316 Winer BJ 1971 The choice of a scale of measurement and transformation. In: Winer BJ (ed) Statistical Principles in Experimental Design, 2nd edn. McGraw-Hill, Kogagusha, Ch 5, pp 397–400

Manual Therapy (2003) 8(4), 223–232

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 233–241 r 2003 Elsevier Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00053-5

Original article

Stabilizing training compared with manual treatment in sub-acute and chronic low-back pain E. Rasmussen-Barr*,w, L. Nilsson-Wikmarn, I. Arvidssonn Neurotec Department, Division of Physiotherapy, Karolinska Institute, Stockholm, Sweden, w Ortoped Medicinskt Center, Stockholm, Sweden

n

SUMMARY. The present aim was to compare the effects of stabilizing training with those of manual treatment in patients with sub-acute or chronic low-back pain (LBP). Forty-seven patients were randomized to a stabilizing training group (ST group) or a manual treatment group (MT group). The patients underwent a 6-week treatment programme on a weekly basis. Pain, health and functional disability level at the start of treatment, after treatment, and at 3- and 12-month follow-ups were assessed. In the ST group all assessed variables improved signiﬁcantly (Po0.05) after the treatment period and were maintained long term. After the treatment period there was a signiﬁcant difference between the groups in assessed function (Po0.05). More individuals in the ST group had improved than in the MT group. At the 3-month follow-up signiﬁcantly more improved individuals were evident in the ST group regarding pain, general health and functional disability levels. In the long term, signiﬁcantly more (Po0.05) patients in the MT group reported recurrent treatment periods. The study did not indicate any clear short-term differences between the groups in the accessed outcome measures. In the long term, however, stabilizing training seemed to be more effective than manual treatment in terms of improvement of individuals and the reduced need for recurrent treatment periods. r 2003 Elsevier Ltd. All rights reserved.

within 3 months: 75% of these exhibited LBP and related disability 1 year after the consultation. Physiotherapy for patients with LBP varies. Manual therapy might have an effect on assessed pain and functional disability levels (Blomberg et al. 1992; Di Fabio 1992; Koes et al. 1992; Faas et al. 1993; Hansen et al. 1993; van Tulder et al. 1997; Skargren 1998). However, when comparing different treatment approaches, outcome measures show no signiﬁcant short- or long-term differences (Cherkin et al. 1983; van Tulder et al. 1997; Seferlis et al. 1998; Torstensen et al. 1998; Petersen et al. 2002). These studies do, however, show that physiotherapy is better for the patient than no treatment at all or only contact with a general practitioner. A recent review of various LBP treatments concluded that an active approach is indeed effective (van Tulder 2001). There is, however, still no evidence as to what exercises or what training is best for various subgroups (Campello et al. 1996; Faas 1996, Nordin & Campello 1999; Mannion et al. 2001). In clinical reality, manual treatment (e.g. stretching, traction, mobilization and manipulation) and training are

INTRODUCTION Low-back pain (LBP) is a major economic problem (SBU 2000). Some 60–80% of the population will at some time exhibit LBP (Waddell 1987). Several studies report persisting LBP and disability at 1-year follow-up, as well as recurrence within and after 1 year (van der Hoogen et al. 1997; Croft et al. 1998; Skargren 1998; Hides 2001). Croft et al. (1998) reported that 90% of a group of patients with LBP in primary care stopped consulting their doctor

Received: 16 July 2002 Revised: 31 January 2003 Accepted: 7 April 2003 Eva Rasmussen-Barr, RPT, MSc, Ortoped Medicinskt Center, Stockholm Sweden, Lena Nilsson-Wikmar, RPT, Neurotec Department, Division of Physiotherapy, Karolinska Institute, Stockholm, Sweden, Inga Arvidsson, RPT, PhD, Neurotec Department, Division of Physiotherapy, Karolinska Institute, Stockholm, Sweden. Correspondence to: ERB, Ortoped Medicinskt Center, Scheelegatan 28, Stockholm SE-112 28, Sweden. Tel.: +46 8 6175605; Fax: +46 8 6526030; E-mail: [email protected] 233

ARTICLE IN PRESS 234 Manual Therapy

often used in combination, to give the patient pain relief and better function. Patients often get better, but the pain frequently recurs and many patients seek treatment over and over again. Stabilizing training is reportedly effective in the management of LBP (Saal & Saal 1989; Robinson 1992). The aim is to attain adequate dynamic control of lumbar spine forces, thus eliminating repetitive injury to the structures of the spinal segments and related structures (Saal & Saal 1989; Saal 1990). Richardson and Jull (1995, 1999) have described speciﬁc stabilizing exercises with co-contraction of the deep abdominal (m. Transversus Abdominus, m. Obliquus Internus) and lumbar multiﬁdus muscles. In recent clinical trials, these exercises have proved effective in the management of LBP in the short term as well as in the long term (Hides et al. 1996, 2001; O’Sullivan et al. 1997). The objective of the present study was to evaluate speciﬁc stabilizing training for a group of patients with sub-acute or chronic LBP, with diagnosed segmental dysfunction of the lower back, and to compare this group with a control group treated with a more passive approach — manual treatment. The hypothesis was that stabilizing training is more effective in the long term than manual treatment. PATIENTS AND METHODS Patients Patients with low-back pain (sub-acute, chronic or recurrent) seeking care at a physiotherapy clinic in Stockholm in 1999–2000 were asked to participate in the study. Forty-seven patients (12 men, 35 women) volunteered to take part. They received verbal and written information about the trial. Inclusion criteria: Men and women aged 18–60 years with LBP (pain >6 weeks) with or without radiation to the knee and pain provoked by provocation tests of lower lumbar segments (Strender et al. 1997; Van Dillen et al. 1998).

Exclusion criteria: Prior segmental stabilizing training, manual treatment in the previous 3 months, prior spinal surgery, radiation to the leg or legs with overt neurological signs, pregnancy, known lumbar disc hernia, diagnosed inﬂammatory joint disease, known severe osteoporosis, or known malignant disease. After a physical examination by physiotherapists with an international diploma in orthopaedic manipulative therapy (MT) the patients were randomized into either of two treatment groups: a stabilizing training group (ST group) (n=24) and a control group treated with manual methods (MT group) (n=23) (Table 1). The ﬁrst woman and ﬁrst man included in the study were randomized to one of the groups by lot (25 ST cards and 25 MT cards in a box). The men and the women were then separately and consistently randomized to either group. At randomization the patients were assigned a unique code. Five patients (two in the ST group and three in the MT group) dropped out of the trial immediately after inclusion due to preference for other treatment or training outside the trial (Table 1). The groups were thus: ST n=22, MT: n=20. One MT-group patient was diagnosed with lumbar hernia (MRI) during the intervention (Table 1). All dropouts are included in the baseline data (Table 2). The Ethics Committee at the Karolinska Institutet in Stockholm approved the study. Procedure Before and after the intervention and at 3- and 12month follow-ups, assessments were made. After the examination and before the intervention the patients ﬁlled out a questionnaire regarding age, gender, sick leave, pain duration, medication, earlier treatment, and exercise habits (Skargren 1998). Pre-coded envelopes with questionnaires and assessment instruments were mailed to the patients before the ﬁrst treatment, after the last treatment and for the 3- and

Table 1. Flowchart showing the study ﬂow, number of subjects and number of dropouts

Inclusion

Treatment After treatment

12 months

5

2* 24

3 month

ST 22

ST 22

17

17

MT 20

MT 19

16

14

47 23 3*

1

3

2

The shaded boxes represent subjects who answered questionnaires and the open boxes represent dropouts. *Subjects that dropped out before treatment. r 2003 Elsevier Ltd. All rights reserved.

Manual Therapy (2003) 8(4), 233–241

ARTICLE IN PRESS Stabilizing training with manual treatment 235 Table 2. Subjects’ characteristics (n=47)

Age (year) Median (SD) Gender (female/male) Medication (every day/seldom/never) Exercise habits (seldom/weekly) Previous treatment for LBPn Pain duration 6–12 weeksn >12 weeksn Pain (recurrent/daily) Sickleave (yes/no) Assessments Pain (VAS) median (25th/75th) Pain when sitting (VAS>30) Pain when bending forward (VAS>30) Disability Oswestry median (25th/75th) DRI median (25th/75th) General health (VAS) median (25th/75th)

ST-group (n=24)

MT-group (n=23)

39(712) 17/7 1/13/10 7/17 17(71%)n 3(12%)n 21(88%)n 22/2 2/22

37(710) 18/5 1/14/8 8/15 18(78%)n 2(9%)n 21(91%)n 22/1 3/20

33(27/49) 13(54%)n 14(60%)n

32(21/49) 14(61%)n 16(70%)n

18(9/25) 32(15/47) 35(22/47)

14(10/21) 33(20/46) 30(18/41)

ST=Stabilizing training group, MT=manual treatment group. n Number of individuals (% of individuals in the group).

12-month follow-ups. All pre-paid envelopes were returned by mail. Pain assessment: Visual analogue scales (VAS) (Huskisson 1983) were used to assess pain. The VAS was a 10-cm vertical line, anchored from ‘no pain’ to ‘unbearable pain’. The patient registered the magnitude of pain with a mark on the line. Health assessment: To assess general health, another visual analogue scale (VAS) was used. The 10-cm vertical VAS was anchored by ‘best health imaginable’ and ‘worst health imaginable’ (Skargren 1998). Functional disability: The Oswestry Low Back-Pain Questionnaire (OSW) was used to assess physical function. The instrument is designed to assess how pain affects various activities of daily living and gives a percentage score of level of function (Fairbank et al. 1980). Through the Disability Rating Index (DRI), perceived difficulty in performing 12 daily activities (e.g. sitting, bending forward, and climbing stairs) was examined. Each question is scored on a 10-cm visual analogue scale from ‘no difﬁculty’ (0) to ‘totally unable’ (10). The DRI index is the mean score of all items (Sale! n et al. 1994). After the intervention, all the patients assessed their satisfaction using a third 10-cm visual analogue scale. The scale was anchored with ‘not satisﬁed at all’ and ‘very satisﬁed’ with the treatment. At the 3and 12-month follow-ups, the patients were asked if they had had to seek physiotherapy treatment between treatment and follow-up (yes/no). Intervention ST group The ST-group patients (n=22) underwent a 6-week treatment programme, meeting individually with a Manual Therapy (2003) 8(4), 233–241

physiotherapist (MT) once a week for 45 min. The patients were told how to activate and control their deep abdominal and lumbar multiﬁdus (MF) muscles (Richardson & Jull 1995; Richardson et al. 1999). The ﬁrst phase was cognitive and the patients were taught how these muscles act as stabilizers for the lumbar spine. The importance of re-learning motor control of these muscles was underlined. The patients were taught how to activate the deep abdominal muscles together with relaxed breathing in different positions (e.g. supine crooked-lying, four-point kneeling, prone, sitting and standing). The activation of MF together with the deep abdominal muscles was also trained. The physiotherapist monitored the patient by palpating the lower abdominal quadrant for deep tensioning of the abdominal muscles and by palpating the MF at the painful level. A biopressure unit (Chattanooga Paciﬁc P/L, Australia) was used in the learning process. The exercises were gradually developed by applying low load to the muscles through the limbs in different positions. The patients were instructed in how to use contraction of the muscles during activities of daily living, and in situations that set off pain. They were encouraged to perform a training programme, designed to take 10–15 min, at home each day. They kept a training diary to control compliance. During each session the physiotherapist monitored how well the patient was able to control the muscle activity and to perform the exercises. The patients were also taught basic ergonomics. MT group The MT-group patients (n=20) underwent a 6-week programme, being treated individually once a week by a physiotherapist (MT) for 45 min. Manual techniques (Evjenth & Hamberg 1988; Grieve 1988; Kaltenborn 1989) were used, based on ﬁndings from r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 236 Manual Therapy

the physical examination. They could include a combination of muscle stretching, segmental traction, and soft tissue mobilization and, if needed mobilization of stiff thoracic and upper lumbar segments. No manipulation was done. The patients were encouraged to go on with their usual activities or exercises (not controlled). None of these exercises included speciﬁc stabilizing exercises. The patients were also taught basic ergonomics.

RESULTS Forty-one patients completed the 6-week intervention. Thirty-three patients replied to the 3-month follow-up questionnaire and thirty-one to the 12month follow-up (Table 1). At baseline there was no signiﬁcant difference between the groups regarding age, gender, sick leave, pain duration, medication, exercise habits or earlier treatment (Table 2). Within-group analysis

ANALYSIS OF DATA As the data were not normally distributed, nonparametric statistics were used for the analysis. *

*

*

Friedman’s ANOVA (analysis of variance) was used to assess difference in outcome measures within each group after the treatment period and at the 3- and 12-month follow-ups. The Mann–Whitney U-test was used: (a) on baseline data, to account for group differences on entry to the trial, and (b) on ‘change’ scores (difference between individual baseline and 3- and 12-month follow-up scores) for each measure. Mann–Whitney test was also used to assess differences between the two groups after treatment and after the 3- and 12-month follow-ups. The chi square test was used to compare the number of ‘improved’ individuals in the two groups after the treatment period and at the 3 and 12-month follow-ups and recurrent treatment periods in the follow-ups. This test was also used to assess differences in recurrent treatment periods between the groups.

The level of statistical signiﬁcance was set at Po0.05. Box plots were used to visualise the results.

Analysis of difference within each group after the treatment period revealed signiﬁcant improvement in the ST group regarding assessed pain (Po0.001), functional disability levels (DRI; Po0.001, OSW; Po0.001) and assessed health (Po0.05). At the 3-and 12-month follow-ups the ST group had maintained its improvement with regard to the assessed variables (Table 3) (Figs 1–4). The MT group showed no signiﬁcant improvement either after the treatment period (pain P=0.15, general health P=0.63, DRI P=0.09) or at the 3 or 12-month follow-ups. Disability levels assessed with the OSW showed a signiﬁcant improvement (P=0.01) after the treatment period, though this was not maintained in the long term (Table 3) (Figs 1–4). Between-groups analysis When comparing change-scores of outcome measures between the groups, there was a tendency towards signiﬁcant difference with regard to functional disability (DRI) in the ST group after the treatment (P=0.04) and maintained in the long term (Table 3). No signiﬁcant difference was seen between the groups after the treatment period regarding pain (P=0.33), disability levels assessed with OSW (P=0.08), or

Table 3. Median values (25th/75th) for the ST and MT groups

Pain (VAS) St group (N) Mt group (N) Oswestry St group Mt group Dri St group Mt group Health (VAS) St group Mt group

Inclusion

After

3 months

12 months

P (within)

P (between)

33(27/49) 24 32(21/49) 23

20(8/32) 22 24(11/47) 19

14(3/22) 17 22(7/45) 16

13(5/23) 17 18(9/38) 14

o0.001n

ns

18(8/25) 14(10/21)

9(6/16) 12(8/14)

6(4/10) 13(3/20)

8(2/10) 8(6/19)

o0.001n ns

ns

32(15/47) 33(20/46)

22(7/29) 28(16/43)

12(7/23) 28(8/39)

13(6/29) 23(11/33)

o0.001n ns

0.042n

35(22/47) 30(18/41)

21(14/31) 28(17/49)

20(12/27) 21(13/32)

18(12/28) 32(22/43)

o0.05n ns

ns

ns

P values for differences within and between groups on inclusion, after treatment and at follow-ups. ST=Stabilizing training group; MT=manual treatment group; ns=not signiﬁcant. n After treatment period. r 2003 Elsevier Ltd. All rights reserved.

Manual Therapy (2003) 8(4), 233–241

ARTICLE IN PRESS Stabilizing training with manual treatment 237

Fig 1—Assessed pain (VAS), at baseline, after the treatment period and at the 3- and 12-month follow-ups. ST=Stabilizing training group; MT=manual treatment group.

Fig 2—Assessed general health (VAS), at baseline, after the treatment period and at the 3- and 12-month follow-ups. ST=Stabilizing training group; MT=manual treatment group.

Fig 3—Functional disability level (DRI) at baseline, after the treatment period and at the 3- and 12-month follow-ups. ST=Stabilizing training group; MT=manual treatment group.

health (P=0.06). At the 3-month follow-up, no difference was seen between the groups (pain P=0.15, OSW P=0.10, health P=0.53); nor at 12-month follow-up (pain P=0.69, OSW P=0.24, health P=0.06). Manual Therapy (2003) 8(4), 233–241

Numbers of improved individuals in each group were compared with regard to assessed variables after the treatment period and at the 3- and 12-month follow-ups. Cut-off scores were established to categorize clinically signiﬁcant changes. Minimal clinical r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 238 Manual Therapy

Fig 4—Functional disability level (OSW) at baseline, after the treatment period and at the 3- and 12-month follow-ups. ST=Stabilizing training group; MT=manual treatment group.

Table 4. Improved individuals in the ST and MT groups after treatment and at the 3- and 12-month follow-ups regarding assessed variables: pain (VAS o10), general health (VAS o10), DRI (score o10) and Oswestry (scoreo10) After treatment

Pain Health DRY OSW

ST

MT

(n=22)

(n=19)

15 10 12 10

10 5 5 3

3-month follow-up P

ns ns ns ns

ST

MT

(n=17)

(n=16)

16 13 14 12

7 6 7 7

12-month follow-up P

0.002 0.02 0.03 ns

ST

MT

(n=17)

(n=14)

14 10 12 10

7 4 7 4

P

ns ns ns ns

ST=Stabilizing training group, MT=manual treatment group; ns=not signiﬁcant.

improvement was set at o10 mm (VAS, DRI) and o10% (OSW). After the treatment period there were more improved individuals in the ST group than in the MT group according to the assessed variables (Table 4). At the 3-month follow-up there were signiﬁcantly more improved individuals in the ST group regarding pain (P=0.002), general health (P=0.02) and functional disability (DRI; P=0.03) (Table 4). At the 3- and 12-month follow-ups more (Po0.05) MT-group patients reported recurrent treatment periods than ST-group patients did. At the 3-month follow-up, eight (50%) vs two (11%) had undergone new treatment periods; and at the 12-month follow-up seven (50%) vs two (11%). After the treatment period there was no difference (P=0.37) between the groups regarding assessed satisfaction with treatment or training.

DISCUSSION The hypothesis of this study was that an active approach, i.e. stabilizing training, is more effective than a passive, i.e. manual treatment, in the long term in a group of patients with sub-acute or chronic lowback pain. The ST group showed signiﬁcant improvement in all the assessed variables after the treatment period r 2003 Elsevier Ltd. All rights reserved.

maintained in the long term. There was a signiﬁcant difference between the groups regarding change scores in assessed disability (DRI), which is in line with recent research (O‘Sullivan et al. 1997). In assessed pain and general health, no signiﬁcant difference was seen between the groups. Two different instruments were used to assess disability. Firstly, the Oswestry low-back pain questionaire, is designed to assess how pain affects ten activities of daily living and with the second, the DRI, perceived difficulty in performing 12 daily activities was examined (e.g. sitting, bending forward, and climbing stairs). The Disability Rating Index (DRI) reﬂects function and disability rather than pain and each question is scored on a 10-cm visual analogue scale from no difﬁculty (0) to ‘totally unable’ (10). The DRI was perhaps more sensitive to clinically important changes in this group of patients. To detect a minimal clinically important change compared to a statistical difference in outcome measures between the groups, individuals who had improved on the assessed variables were counted. Signiﬁcantly more improved individuals were reported in the ST group than in the MT group on assessed pain, health, and disability (DRI) at the 3-month follow-up. Fritz and Irrgang (2001), reported a minimal clinically important change of 7o points in total score, using the Oswestry Manual Therapy (2003) 8(4), 233–241

ARTICLE IN PRESS Stabilizing training with manual treatment 239

Low-Back-Pain Questionnaire with patients with chronic LBP. Another study deﬁned improved pain assessment (VAS) with a cut-off score of o2.8 mm (Mannion et al. 2001). Use of assessment scales in clinical trials may pose a problem as the patient’s assessment is subjective and the researcher has to interpret what improvement actually means. However, assessment scales are considered to be an appropriate tool for evaluating outcome measures in the treatment of patients with LBP (Deyo et al. 1998). Low-back-pain is difﬁcult to diagnose. Only about 15% of patients have an X-ray-veriﬁed diagnosis, e.g. spondylolisthesis or lumbar disc hernia (Nachemsson 1985; SBU 2000). Trauma, bad posture, muscle fatigue or degeneration leading to a zygapophysial joint strain (Kirkaldy-Willis 1992; Panjabi 1992a, b; Gardner-Morse et al. 1995) may lead to ‘self-injury’ and trigger the onset of pain. Seventy-four per cent of the present patients reported recurrent pain and a majority had aggravated pain provoked by active or sustained positions. More than 50% reported pain (VAS >30) when sitting or with slight forward bending. These patients could ﬁt into clinical patterns described by O’Sullivan (2000) as lumbar segmental instability (LSI). This group of patients seldom gets better in the long term with manual treatment. Manual treatment may be beneﬁcial in diminishing pain and thus restoring function in the short term. The long-term effect of manual treatment, however, remains unknown. Both treatment groups in the present study had weekly individual contact with a physiotherapist and part of the short-term improvement might be attributed to the speciﬁc attention the patient enjoyed. A majority of the patients in this study had undergone manual treatment at an earlier stage and reported poor results. The patients in the MT group seemed to achieve a short-term beneﬁt after the treatment period but in the long-term perspective 50% reported a recurrent need for treatment. This makes it difﬁcult, in the long term, to compare the groups with regard to change scores. The recurrent pattern demonstrated in Figures 1–4 is well recognized in clinical reality. The MT group was, however, as satisﬁed with the intervention as the ST group was. The present training method differs from others, e.g. strength training, by performing the exercises graded and under low-load conditions (Richardson & Jull 1995; Richardson et al. 1999). The cognitive approach probably helps the patient to understand the mechanisms of LBP better. Our patients were urged do the exercises every day and most important of all, to activate the stabilizing muscles in all activities of daily living. The lesser need for recurrent physiotherapy visits in the long term in the ST group could suggest that the intervention provided the necessary impetus for continuing with the exercises Manual Therapy (2003) 8(4), 233–241

even in the long term. This was not assessed but could have been one way to detect why some of the ST group patients did not improve their pain and disability scores. Subjects with low-back pain may have impaired motor control of the stabilizing muscles of the spine (Hodges & Richardson 1996). The present study does not show that the improvement in the ST group was due to re-learning of motor control of the deep abdominal and MF muscles, as no objective measurement was made. This could be of interest for future studies. O’Sullivan et al. (1997) studied a group of patients with X-ray-veriﬁed spondylolisthesis who trained their deep abdominal and lumbar MF muscles. The group signiﬁcantly improved on both short- and long-term outcome measures. The present results do not show such a distinct difference in between-group measurements. One reason might be that our training (6 weeks) was shorter than O’Sullivan’s ten week period. Another reason could be that O’Sullivan’s patients were more clinically deﬁned than those in this study, who lacked X-ray-veriﬁed diagnoses. The treatment period was considered sufﬁcient in the light of a 4-week pilot trial, and a weekly visit was considered the most appropriate as most of the patients were working. Hides et al. (1996) reported a training period of 4 weeks for patients with acute unilateral LBP. These patients suffered signiﬁcantly less recurrent pain in the long term (Hides et al. 2001). We did not compare cost-effectiveness between the groups. However, 74% of our patients (n=47) had earlier been treated by physiotherapists (65%) or naprapaths and chiropractors (35%). More than 50% of these reported poor outcomes of earlier treatments. These ﬁgures indicate that many patients with LBP seek treatment recurrently, incurring huge costs for society. In the long term signiﬁcantly (Po0.05) more patients reported recurrent need of treatment in our MT group (50%) than in our ST group (11%). Due to the unexpected number of dropouts from the study (ﬁve initially, ten later), the authors have elected to consider it a pilot study. All subjects gave oral consent to participate: written consent would perhaps have been more binding and should be considered in a future study.

CONCLUSION The dropout rate precludes any deﬁnite conclusion and the results are therefore preliminary. In the long term, training of the stabilizing muscles of the spine, however, seems to be more effective judged by alleviated symptoms and less recurrent need for treatment. The effectiveness of an intervention in recurrent treatment periods perhaps offers the most r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 240 Manual Therapy

interesting and sensitive outcome measure, as a great many patients with LBP seek treatment over and over again. The present study provides sufﬁcient trends to justify another study, which should include more subjects. A future study could also evaluate characteristics of the subjects who improve and the ones who do not. Acknowledgements The authors wish to thank Staffan Ekblom at the Statistical Research Group, University of Stockholm, for statistical advice, and Lennart Blomqvist, MD, for helpful suggestions. The study was ﬁnancially supported by the Anne-Marie and Ragnar Hemborg Foundation.

References Blomberg S, Sv.ardsudd K, Mildenberger FD 1992 A controlled multicentre trial of manual therapy in low-back pain. Scandinavian Journal of Health Care 10: 170–178 Campello M, Nordin M, Weiser S 1996 Physical exercise and low back pain. Scandinavian Journal of Medical Science and Sports 6: 63–72 Cherkin DC, Deyo RA, Batti!e M, Street MN, Barlow W 1983 A comparison of physical therapy, chiropractic manipulation and provision of an educational booklet for the treatment of patients with low back pain. New England Journal of Medicine 339: 1021–1029 Croft PR, Macfarlane GJ, Papageorgiou AC, Thomas E, Silman A 1998 Outcome of low back pain in general practice: A prospective study. British Medicine Journal 316: 1356–1359 Deyo RA, Battie M, Beurskens AJ, Bombardier C, Croft P, Koes B, Malmivaara A, Roland M, von Korff, Wadell G 1998 Outcome measures for low back pain research. Spine 18: 2003–2013 Di Fabio RP 1992 Efﬁcacy of manual therapy. Physical Therapy 12: 853–864 Evjenth O, Hamberg J 1988 Muscle Stretching in Manual Therapy . II, 2nd edn. Alfta Rehab Forlag, Alfta, Sweden Faas A, Chavannes AW, van Eijk JT, Gubbels J 1993 A randomised, placebo-controlled trial of exercise therapy in patients with acute low back pain. Spine 20: 1388–1395 Faas A 1996 Exercises: Which ones are worth trying for which patients and when? Spine 24: 2874–2879 Fairbank JC, Couper J, Davies JB, O’Brian J 1980 The Oswestry low back pain disability questionnaire. Physiotherapy 3: 271–273 Fritz JM, Irrgang JJ 2001 A comparison of a modiﬁed Oswestry low back pain disability questionnaire and the Quebec back pain disability scale. Physical Therapy 2: 776–788 Gardner Morse M, Stokes IA, Laible JP 1995 Role of muscles in lumbar spine stability in maximum extension efforts. Journal of Orthopaedic Research 13: 812–808 Grieve PG 1988 Common Vertebral Joint Problems, 2nd edn. Churchill Livingstone, New York, p 300–301, 313–319, 403–433, 450–500 Hansen FR, Bendix T, Skov P, Jensen CV, Kristensen JH, Krohn L, Schioler H 1993 Intensive, dynamic back-muscle exercises, conventional physiotherapy or placebo-control treatment of low-back pain. A randomised observer-blind trial. Spine 1: 98–108 Hides JA, Richardson CA, Jull GA 1996 Multiﬁdus recovery is not automatic following resolution of acute ﬁrst episode low back pain. Spine 20: 2763–2769 Hides JA, Jull GA, Richardson CA 2001 Long-term effects of speciﬁc stabilizing exercises for ﬁrst-episode low back pain. Spine 11: 243–248 Hodges PW, Richardson CA 1996 Inefﬁcient muscular stabilization of the lumbar spine associated with low back pain. Spine 21: 2640–2650 r 2003 Elsevier Ltd. All rights reserved.

Huskisson EC 1983 Visual Analogue Scales. Pain measurement and Assessment. Raven Press, New York pp 33–37 Kaltenborn FM 1989 Manuell mobilisering av ryggraden. Oslo, Norway Kirkaldy-Willis WH, Burton CV 1992 Managing Low Back Pain, 3rd edn. Churchill Livingstone, New York, pp 105–119 Koes BW, Bouter LM, van Mameren H, Essers AH, Verstegen GM, Hofhuizen DM, Houben JP, Knipschild PG 1992 The effectiveness of manual therapy, physiotherapy, and treatment by general practitioner for non-speciﬁc back and neck complaints. Spine 1: 28–35 Mannion AF, Muntener M, Taimela S, Dvorak J 2001 Comparison of three active therapies for chronic low back pain: Results of a randomized clinical trial with one year follow–up. Rheumatology 40: 772–778 Nachemsson A 1985 Lumbar spine instability. A critical update and symposium summary. Spine 3: 290–291 Nordin M, Campello M 1999 Physical therapy. Exercises and the modalities: When, what and why? Neurological Clinics of North America 1: 75–89 O’Sullivan PB, Twomey LT, Allison G 1997 Evaluations of speciﬁc stabilising exercise in the treatment of chronic low back pain with radiological diagnosis of spondylolysis or spondylolistheses. Spine 24: 2959–2967 O’Sullivan PB 2000 Lumbar segmental ‘instability’: Clinical presentation and speciﬁc stabilizing exercise management. Manual Therapy 1: 2–12 Panjabi MM 1992a The stabilising system of the spine. Part I. Function, dysfunction, adaptation and enhancement. Journal of Spinal Disorder 4: 383–389 Panjabi MM 1992b The stabilising system of the spine. Part II. Neutral zone and instability Hypothesis. Journal of Spinal Disorder 4: 390–397 Petersen T, Kryger P, Ekdahl C, Olsen S, Jacobsen S 2002 The effect of McKenzie therapy as compared with that of intensive strengthening training for the treatment of patients with subacute or chronic low back pain. Spine 16: 1702–1709 Richardson CA, Jull GA 1995 Muscle control — pain control. What exercise would you prescribe? Manual Therapy 1: 2–10 Richardson CA, Jull GA, Hodges PW, Hides J 1999 Therapeutic Exercise for Spinal Segmental Stabilisation in Low Back Pain. Scientiﬁc Basis and Clinical Approach. 1st edn. Churchill Livingstone, London Robinson R 1992 The new back school prescription: Stabilization training. part I. Occupational Medicine 1: 17–27 Saal JA, Saal JS 1989 Nonoperative treatment of herniated lumbar intvertebral disc with radiculopathy: An outcome study. Spine 14: 431–437 Saal JA 1990 Dynamic muscular stabilization in the non-operative treatment of lumbar syndromes. Orthopedic Review 19: 691–700 ( Nordemar R 1994 The Sal!en BA, Spangfort EV, Nygren AL, disability rating index: An instrument for assessment of disability in clinical setting. Journal of Clinical Epidermiology 12: 1423–1434 SBU — The Swedish Council on Technology Assessment in Health Care 2000 Back pain, Neck pain — An evidence based review . P 1998 ConservaSeferlis T, N!emeth G, Carlsson AM, Gillstrom tive treatment in patients sick-listed for acute low-back pain: A prospective randomised study with 12 months follow-up. European Spine Journal 7: 461–470 Skargren E 1998 Evaluation of physical exercise, physiotherapy and chiropractic in the management of back pain. . Thesis. Faculty of Health Sciences, Linkoping University, Sweden . Strender LE, Sjoblom A, Sundell K, Ludwig R, Taube A 1997 Interexaminer reliability in physical examination of patients with low back pain. Spine 22: 814–820 Torstensen TA, Ljunggre, AE, Meen HD, Odland E, Monwinckel P, Geijerstam S 1998 Efﬁciency and costs of medical exercise therapy, conventional physiotherapy and self-exercise in patients with chronic low back pain. Spine 23: 2616–2624 van der Hoogen HJ, Koes BW, Deville W, van Eijk J, Bouter LM 1997 The prognosis of low back pain in general practices. Spine13: 1515–1521 Manual Therapy (2003) 8(4), 233–241

ARTICLE IN PRESS Stabilizing training with manual treatment 241

van Dillen LR, Sahrman SA, Norton BJ, Caldwell CA, Flemming DA, McDonell MK, Wolsey NB 1998 Reliability of physical examination items used for classiﬁcation of patients with low back pain. Physical Therapy 9: 979–988 van Tulder MW, Koes BW, Bouter LM 1997 Conservative treatment of acute and chronic non-speciﬁc low back pain. Spine 18: 2128–2156

Manual Therapy (2003) 8(4), 233–241

van Tulder MW 2001 Treatment of low back pain: myths and facts. Schmertz 6: 499–503 Waddell G 1987 A new clinical model for the treatment of lowback pain. Spine 7: 632–644

r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 242–246 r 2003 Elsevier Science Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00016-X

Technical & measurement report

Can the Internet be used as a medium to evaluate knee angle? T. G. Russell*, G. A. Jull, R. Woottonw Department of physiotherapy, The University of Queensland, Australia, wCentre for Online Health, The University of Queensland, Australia

n

SUMMARY. Telemedicine promises to revolutionize medical care delivery in rural and remote areas. The ability to accurately evaluate physical impairment via the Internet is important to the possible future provision of Internetbased physiotherapy. This study evaluated the reliability and validity of assessing knee range of motion via the Internet. Two therapists evaluated knee angle on a single subject via two methods of assessment: the Internet and the traditional method (face-to-face). Nine random positions of the knee were chosen with the principal examiner performing 20 face-to-face and two sets of 20 Internet measures in each position (n=540). The secondary therapist performed Internet assessments only. The Internet connection was established at a readily available speed of 17 kbit/s. The Internet-based goniometer was found to be a valid tool for measuring both knee ﬂexion and extension angles. It was shown to possess both high intra and inter-rater reliability. Difference average plots of the scores veriﬁed the consistency of measurement between both modes of assessment. The successful evaluation of the physical outcome measure of knee range of motion via the Internet assists the further development of Internetbased physiotherapy applications. r 2003 Elsevier Science Ltd. All rights reserved.

assessment tools challenge the performance of accurate physical assessments. Physiotherapists frequently measure joint range of motion to determine the presence or absence of dysfunction, establish treatment goals and to evaluate treatment effects. Historically, the most common tool used to perform goniometry is the standard universal goniometer (UG). The UG is a valid tool for measuring joint angles (Gogia et al. 1987), and possesses appropriate inter- and intra-rater reliability (Boone et al. 1978; Rothstein et al. 1983; Watkins et al. 1991). A few attempts have been made to measure joint angle at a distance, i.e. via telemedical applications. Charlebois et al. (2000) investigated the intra- and inter-rater reliability of a videoconference-based goniometer. They used an Integrated Services Digital Network (ISDN) communication link, with two personal computers equipped with video cameras and hardware CODECs, to videoconference two sites. Images were captured from the live teleconference link and later analysed using intraclass correlation coefﬁcients (ICCs). Both intra-rater reliability (all ICCs >0.984) and inter-rater reliability (all ICCs

INTRODUCTION The rapid advancement of technology in the area of telemedicine has heralded a new age in the delivery of medical care. The Internet, with its insensitivity to distance and permeation into society, promises to revolutionize medical care delivery, particularly in rural and remote sectors of the country. Despite the many obvious beneﬁts of telemedicine, the lack of proximity to the client and the futility of conventional

Received: 5 August 2002 Revised: 21 January 2003 Accepted: 11 February 2003 Trevor G. Russell, BPhty, PhD Candidate, Gwendolen A. Jull, MPhty, PhD, FACP, Associate Professor, Department of Physiotherapy, The University of Queensland, Australia, Richard Wootton, PhD, DSc, Professor of Online Health & Director of Research, Centre for Online Health, The University of Queensland, Australia Corresponding author: TGR, Department of Physiotherapy, The University of Queensland, St Lucia QLD 4072, Australia. Tel: +61-33-65-46-91; Fax: +61-33-65-2-775; E-mail: [email protected] 242

ARTICLE IN PRESS Internet as a medium to evaluate knee angle 243

>0.983) were high. Shaw et al. (1999) conducted a study where an Orthotech knee brace with a potentiometer was afﬁxed to a subject’s knee. The measurements from the potentiometer were simultaneously transmitted via the Internet and analysed with a software package developed at Syracuse University. The authors concluded that a high consistency of measurements existed between the transmitted values and measurements taken with a UG. The cost of the equipment required for these existing applications is inconsistent with the increasing focus on cost-containment in health care (Mair et al. 2000). Telemedical applications that are inexpensive to purchase, require minimal additional equipment and are inexpensive to operate are paramount for the timely uptake of this technology. The authors have developed a lowcost telemedical application for the delivery of physiotherapy intervention to the home (Fig. 1). One tool incorporated into this application is an Internet-based goniometer, (IBG, available from the Authors). This application operates on a conventional Internet enabled personal computer over an inexpensive low bandwidth Internet connection via a standard modem. A real-time video and audio connection is facilitated with the use of lowcost web cameras. The aims of this study were to examine the intraand inter-rater reliability of the new IBG for the physical outcome measure of knee angle; and to determine the criterion validity of the IBG by comparison with the UG.

METHODS Subjects Each measurement performed in this study was dependent on the angle set on a knee board, rather than any performance by a subject. Therefore, one subject was sufﬁcient for the performance of all measures. The subject was a 24-year-old female with no history of knee dysfunction. Ethical clearance was gained from The School of Health and Rehabilitation Sciences, The University of Queensland and informed consent obtained. Raters Two physiotherapists at The University of Queensland participated in this study. The principal therapist performed measurement via the UG (n ¼ 180) and two separate measurements with the IBG (total n ¼ 360). The second therapist performed assessments via the IBG only (n ¼ 180). A 15-min training session was conducted in which both therapists were familiarized with the IBG. Physical measures The angle of the knee joint was examined via two methods: face-to-face with a UG and via the Internet with the IBG. Face-to-face measurements were made using a standard 30 cm Baselines universal goniometer. The following landmarks (Norkin & White 1995) were

Fig. 1—Physiotherapy telemedical software. Manual Therapy (2003) 8(4), 242–246

r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 244 Manual Therapy

used to calculate the range of knee motion: (1) axis of the goniometer over the lateral femoral epicondyle; (2) the proximal arm aligned with the femur, with the greater trochanter used as a reference; and (3) the distal arm aligned with the ﬁbular, using the lateral malleolus and the ﬁbular head for reference. The Internet application was conducted via a standard Internet connection at 17 kbit/s between two personal computers. Real-time video and audio connections facilitated communication between the therapist and subject. Still pictures of the video images (320 240 pixels) were taken by the application once the required knee position was achieved. Landmarks identical to the face-to-face measures were identiﬁed and marked on the computer images. From these reference points, the IBG application determined the knee angle. To facilitate the identiﬁcation of the greater trochanter on the images, the subject was requested to palpate for the bony prominence and place their index ﬁnger on what they considered to be its centre. Procedure The subject was positioned in supine with the knee supported on a kneeboard. The kneeboard was designed to hold the knee in a ﬁxed position at any point through range to facilitate accurate measurements. The web camera was aligned visually with the joint line of the knee. Nine random positions of the kneeboard were arbitrarily selected and the principal assessor initially took 40 measurements of knee angle via both the UG (n ¼ 20) and the IBG (n ¼ 20). Measurements by both methods of assessment were completed before the kneeboard was adjusted to a new angle. The UG and IBG measurements were made alternatively with a new image captured via the Internet application for each IBG measurement. To control for bias, the rater was blinded to the measurements made via both methods of assessment. An assistant recorded all measurements.

The principal and secondary assessors repeated the IBG measurements on the captured images to obtain data for both the intra- and inter-rater components of the study. The assessors were blinded to the measurements. Statistical analysis All data were analysed using the SPSS software package release 11.0. Intraclass Correlation Coefﬁcients (ICC), repeated measures analysis of variance (ANOVA) and Bland and Altman’s (1986; 1999) method of assessing reliability were chosen for statistical analysis. ICCs are commonly used in reliability papers and are presented here primarily to allow for comparison with these studies. An ICC >0.9 was considered to represent ‘good reliability’ in this study (Portney & Watkins 2000). The repeated measures ANOVA statistic was used to compare the magnitude of the means of the two measurement conditions. A level of signiﬁcance of 0.05 was set for this analysis. Limits of agreement calculations and plots of the difference between the ﬁrst and second measurement conditions vs the average of the measurement of the two conditions were constructed as described by Bland and Altman (1986) for validity, inter- and intra-studies. The means of each knee angle, rather than the individual scores, were used for statistical calculations in accordance with the recommendations of Low (1976).

RESULTS The means, standard deviation (SD) and limits of agreement (Bland & Altman 1986) for the UG and IBG measurements for each of the nine random knee angles are presented in Table 1. Knee angles with an average under 101 were grouped as knee extension and over 101 as knee ﬂexion.

Table 1. Means, standard deviations and limits of agreement for the universal goniometer and Internet-based goniometer Angle

UG

IBGA1

LA

LA

LA

Mean

SD

Mean

SD

Mean

IBGB SD

Mean

IBGA2 SD

UG–IBGA1

IBGA1–IBGB

IBGA1–IBGA2

1 2 3 4 5 6 7 8 9

52.1 58.1 67.2 78.8 81.3 96.9 9.2 6.6 2.7

0.8 1.0 0.8 0.8 0.7 0.8 0.9 1.0 1.0

52.0 58.3 67.0 78.7 81.0 97.2 8.7 6.9 3.0

0.5 0.4 0.6 0.5 0.4 0.4 0.6 0.4 0.7

51.1 58.3 67.1 79.3 81.8 97.0 7.7 6.1 2.6

0.5 0.4 0.4 0.4 0.3 0.4 0.5 0.5 0.4

51.6 59.1 67.9 78.9 81.9 97.1 9.0 6.9 3.0

0.4 0.4 0.3 0.6 0.4 0.4 0.4 0.4 0.6

1.7 2.5 1.6 1.7 1.3 2.1 1.5 2.7 2.0

0.2 1.1 1.1 1.6 1.8 0.9 0.8 0.6 1.2

1.7 0.2 0.1 1.3 0.2 1.1 1.0 1.3 2.0

Averages

50.3

0.9

50.3

0.5

50.1

0.4

50.5

0.4

1.9 to 1.8

to to to to to to to to to

1.1 2.1 1.9 1.9 1.9 1.4 2.4 2.0 1.5

to to to to to to to to to

2.1 1.1 0.9 0.2 0.1 1.2 2.7 2.3 2.0

1.0 to 1.4

to to to to to to to to to

0.9 1.9 2.0 1.7 2.0 0.8 1.6 1.4 2.0

1.0 to 1.6

Abbreviations: SD=standard deviation; UG=universal goniometer; IBG=Internet-based goniometer; IBGA1,2=primary assessor initial and subsequent measures; IBGB=second assessor measures; LA=limits of agreement. r 2003 Elsevier Science Ltd. All rights reserved.

Manual Therapy (2003) 8(4), 242–246

ARTICLE IN PRESS Internet as a medium to evaluate knee angle 245

Validity study: The ICC(2,1) for the UG and ﬁrst IBG measurements by the principal assessor (IBGA1) was 1.00 for ﬂexion and 0.99 for extension. A nonsigniﬁcant difference in mean scores between the two methods of assessment was demonstrated for ﬂexion (Fð1;5Þ ¼ 0:02; P ¼ 0:887) and extension (Fð1;2Þ ¼ 0:04; P ¼ 0:859). Intra-rater reliability study: The ICC(2,1) for the ﬁrst (IBGA1) and second measurements (IBGA2) made via the IBG by the principal assessor was 1.00 for ﬂexion and 1.00 for extension. Repeated measures ANOVA demonstrated a non-signiﬁcant difference in the mean scores for ﬂexion (Fð1;5Þ ¼ 2:53; P ¼ 0:173) and extension (Fð1;2Þ ¼ 1:81; P ¼ 0:311). Inter-rater reliability study: The ICC(2,2) for the measurements made by the principal (IBGA1) and secondary assessors (IBGB) via the IBG were 1.00 for ﬂexion and 0.96 for extension. A non-signiﬁcant difference in mean scores between the two assessors was demonstrated for ﬂexion (Fð1;5Þ ¼ 0:11; P ¼ 0:758) and extension (Fð1;2Þ ¼ 14:71; P ¼ 0:062).

DISCUSSION The IBG was found to be a highly reliable tool for measuring knee range of motion. Intra-tester reliability was higher than inter-tester reliability, a phenomenon consistent with the UG (Rothstein et al. 1983; Watkins et al. 1991; Brosseau et al. 1997). The high ICC (all >0.9) and lack of signiﬁcant difference in mean angular measurements (all P >0.05) between the observers suggest that therapists can use this tool conﬁdently and interchangeably. This was veriﬁed with a similar distribution of difference scores around the mean with no one assessor producing consistently larger scores in the difference vs average plots. The IBG thus has a considerable advantage over the UG which possesses relatively poor inter-rater reliability (Boone et al. 1978; Watkins et al. 1991). Criterion-related validity of a measurement instrument is judged by the comparison of measurements made by a new instrument with a ‘gold standard’ (Norkin & White 1995). If the magnitude of disagreement between the UG and the IBG is small enough that clinical interpretation of the assessment remains constant, the IBG may be used instead of, or interchangeably with, the UG (Bland & Altman 1999). The mean difference between the IBG and the UG was 0.031 for both ﬂexion and extension ranges of the knee. The limits of agreement between the two methods was 1.91 to 1.81, indicating that 95% of the difference scores between the methods lie in this range. The small difference in mean angular measurement between the assessment methods is neither clinically nor statistically signiﬁcant (P ¼ 0:944). The extremely high ICC (1.00) between the UG and the Manual Therapy (2003) 8(4), 242–246

IBG exceeds that of many other alternative goniometric devices (Clapper & Wolf 1988; Brosseau et al. 1997; Karkouti & Marks 1997; Charlebois et al. 2000). The reasons might be three fold. Means instead of individual observations of the 20 measurements for each angle were used in the calculation. This process has been recommended by Low (1976) to increase reliability. Secondly, highly accurate measures can be made with the IBG as evident in the magnitude of the standard deviation of the IBG (0.51) compared with the UG (0.91). This smaller variance may be due to the properties of the IBG. The therapist can accurately mark the measurement landmarks on the captured image. This would be comparable to a UG with arms that reach from the hip to the ankle. The use of long arm goniometers may reduce measurement error (Clarkson & Gilewich 1989). In addition, the software includes an axisﬁnding tool that has been found by the authors to remove much of the subjectivity of estimating the lateral femoral epicondyle for the placement of the goniometer axis. Finally, the kneeboard used in this study ensured a static subject position for the repeated measures which has been suggested as important in achieving good reliability (Watkins et al. 1991). The present study indicates the potential of the IBG. Future studies will involve multiple assessors and subjects of different genders, ages and knee pathologies to establish the generalizability of results.

CONCLUSION A low-speed Internet connection can be utilized as a medium over which reliable and valid goniometric measurements can be performed. This study demonstrated that the intra- and inter-rater reliability of the IBG is comparable to that of the UG. Unlike the universal goniometer, the high inter-rater reliability of the Internet-based goniometer indicates that the measurement of subjects can be completed by more than one therapist. Furthermore, the low cost and ease of use of this tool makes it suitable for the ﬁeld of Internet-based clinical practice.

References Bland JM, Altman DG 1986 Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1: 307–310 Bland JM, Altman DG 1999 Measuring agreement in method comparison studies. Statistical Methods in Medical Research 8: 135–160 Boone DC, Azen SP, Lin C, Spence C, Baron C, Lee L 1978 Reliability of goniometric measurements. Physical Therapy 58(11): 1355–1360 Brosseau L, Tousignant M, Budd J, Chartier N, Duciaume L, Plamondon S, O’Sullivan J, O’Donoghue S, Balmer S 1997 Intratester and intertester reliability and criterion validity of the r 2003 Elsevier Science Ltd. All rights reserved.

ARTICLE IN PRESS 246 Manual Therapy

parallelogram and universal goniometers for active knee ﬂexion in healthy subjects. Physiotherapy Research International 2(3): 150–166 Charlebois R, Cote N, O’Rourke M, Verville D, McComas J, Fruhwirth J, Bernard M, Brosseau L 2000 Intra- and interrater reliability of the video conference-based goniometer for active knee ﬂexion and extension in healthy subjects. Journal of Rehabilitation Outcomes Measures 4(3): 23–33 Clapper MP, Wolf SL 1988 Comparison of the reliability of the Orthoranger and standard goniometer for assessing active lower extremity range of motion. Physical Therapy 68(2): 214–218 Clarkson HM, Gilewich GB 1989 Musculoskeletal Assessment: Joint Range of Motion and Manual Muscle Strength, Williams & Wilkins, Baltimore Gogia PP, Braatz JH, Rose SJ, Norton BJ 1987 Reliability and validity of goniometric measurements of the knee. Physical Therapy 67(2): 192–195 Karkouti E, Marks R 1997 Reliability of photographic range of motion measurements in a healthy sample: Knee and ankle joint measurements. Physiotherapy Canada 1: 24–31

r 2003 Elsevier Science Ltd. All rights reserved.

Low JL 1976 The reliability of joint measurement. Physiotherapy 62(7): 227–229 Mair F, Haycox A, May C, Williams T 2000 A review of telemedicine cost-effectiveness studies. Journal of Telemedicine and Telecare 6(Suppl 1): 38–40 Norkin CC, White DJ 1995 Measurement of Joint Motion: A Guide to Goniometry, 2nd Ed. F.A. Davis Company, Philadelphia Portney LG, Watkins MP 2000 Foundations of Clinical Research: Applications to Practice, 2nd ed. Prentice-Hall, Englewood Cliffs, NJ Rothstein JM, Miller JP, Roettger RF 1983 Goniometric reliability in the clinical setting: Elbow and knee measurements. Physical Therapy 63: 1611–1615 Shaw D, Balch D, Gabriel D, Jenkins W 1999 Goniometry via the internet. Physical Therapy Case Reports 2(5): 215–217 Watkins MA, Riddle DL, Lamb RL, Personius WJ 1991 Reliability of goniometric measurements and visual estimates of knee range of motion obtained in a clinical setting. Physical Therapy 71(2): 90–97

Manual Therapy (2003) 8(4), 242–246

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 247–253 r 2003 Elsevier Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00095-X

Clinical note

Speciﬁc evaluation of the function of force couples relevant for stabilization of the glenohumeral joint M.E. Magarey, M.A. Jones Discipline of Physiotherapy, School of Health Sciences, University of South Australia, Adelaide, Australia

SUMMARY. Two clinical evaluation procedures for assessment of dynamic stability of the glenohumeral joint are presented in this paper, together with the biomedical evidence on which they are based. They are the Dynamic Rotary Stability Test (DRST) and the Dynamic Relocation Test (DRT). The purpose of the tests is presented and the technique required to perform the tests are described in detail. r 2003 Elsevier Ltd. All rights reserved.

The transverse force couple is responsible for maintaining the humeral head centred in the glenoid during dynamic activity. A relative balance must be maintained between the activity in subscapularis and that in infraspinatus and/or teres minor for the centring function to be achieved. In relation to elevation, failure within the transverse force couple will lead to failure of the coronal plane force couple, as the compression component of the function will be missing, allowing the pull of deltoid to translate the humeral head superiorly under the acromion (Saha 1971; Poppen & Walker 1978; Kapandji 1982; Soderberg 1986; Schenkman & Rugo de Cartaya 1987, 1994; Norkin & Levangie 1988; Wilk 1994; Burkhart 1994, 1996). On the basis of this understanding of the force couples, the authors have developed two tests to evaluate the dynamic stability of the glenohumeral joint. We have used these tests in the clinic over a number of years, both as evaluation tools and as the basis for dynamic rehabilitation of the glenohumeral joint.

INTRODUCTION The glenohumeral joint is highly mobile, with little bony or ligamentous stability provided in mid-range. Mid-range stability is provided by the rotator cuff in a function termed ‘concavity compression’ (Lippitt et al. 1993; Lippitt & Matsen 1993), where the stabilizing force of the rotator cuff contraction compresses the convex humeral head into the concavity of the glenoid. The rotator cuff can also be considered the ‘ﬁne tuner’ of the glenohumeral joint (Wilk 1994). The force couples associated with elevation of the arm have two components – a coronal plane force couple between deltoid and supraspinatus superiorly and the lower elements of the rotator cuff inferiorly, and a transverse force couple between subscapularis anteriorly and infraspinatus/teres minor posteriorly (Saha 1971; Poppen & Walker 1978; Kapandji 1982; Soderberg 1986; Schenkman & Rugo de Cartaya 1987, 1994; Norkin and Levangie 1988; Wilk 1994; Burkhart 1994, 1996). Received: 1 July 2001 Revised: 7 March 2003 Accepted: 4 July 2003 Mary E Magarey Dip Physio, Grad Dip Advanced Manip Therapy, PhD, Mark A Jones, BS (Psychology), RPT, Grad Dip Advanced Manip Therapy, MAppSc (Manipulative Physiotherapy), Discipline of Physiotherapy, School of Health Sciences, University of South Australia, North Terrace, Adelaide, South Australia 5153, Australia. Correspondence to: MEM. Tel.: +61-8-8302-2768; Fax: +61-88302-2766.

THE DYNAMIC ROTARY STABILITY TEST The Dynamic Rotary Stability Test (DRST) is used to evaluate the ability of the rotator cuff to maintain the humeral head centred in the glenoid when the arm is loaded through rotation. It is predicated on the 247

ARTICLE IN PRESS 248 Manual Therapy

knowledge that the head of humerus should remain centred in the glenoid throughout the range of rotation in any position of abduction or ﬂexion (Howell & Galinat 1987; Howell et al. 1988), except at the limits of range, where ligamentous tightening forces the humeral head to translate (Harryman et al. 1990, 1992). In an unstable shoulder, or one in which dynamic control by the rotator cuff is lacking, the humeral head can be felt to translate when the rotator cuff is loaded, usually anteriorly during resisted lateral rotation and posteriorly with resisted medial rotation. These directions are opposite to those associated with ligamentous tightening at end range (Harryman et al. 1990, 1992). In more subtle situations or where the instability is more functional than structural, provocation of symptoms or alteration in the quality of contraction will alert the examiner to dysfunction without the sensation of humeral head translation. In patients with subacromial impingement, for example, the humeral head may be felt to translate superiorly rather than anteriorly or posteriorly. The DRST is undertaken in different parts of the range of glenohumeral ﬂexion and abduction from neutral towards the functional position(s) in which the patient has symptoms, whether pain, weakness, apprehension or instability. The number of positions in which the test is performed depends on the irritability of the condition, the general physical status of the patient, the clarity with which the patient can identify the position(s) in which symptoms are provoked and the demands placed on the shoulder by the patient. The aim is to ﬁnd the position(s) in range where the patient has control of the head of humerus as close to the position in which control is lost as possible. The test is performed isometrically, isotonically, concentrically and eccentrically at different speeds and under different loads, depending on the demands placed on the shoulder by the patient. The amount of resistance added is usually light to moderate, as the assessment is one of the ability of the rotator cuff to stabilize, rather than one of strength of rotation. Technique The test is best performed in sitting. The therapist holds around the head of the humerus such that the middle ﬁnger is positioned longitudinally along the anterior joint line, adjacent to the coracoid process, the palm of the hand is over the top of the shoulder, adjacent to the acromioclavicular joint and the thumb is positioned longitudinally along the posterior joint line. There should be complete contact with the shoulder from anterior to posterior. The hand should be placed medial to the deltoid, if possible, to avoid contraction of deltoid masking the ability to feel humeral head movement. To examine the right r 2003 Elsevier Ltd. All rights reserved.

shoulder, the left hand is used to palpate the head of humerus. The test is ﬁrst performed isometrically, with all movements into medial rotation undertaken together, followed by all movements into lateral rotation (or vice versa). With the patient’s arm by the side, elbow ﬂexed to a right angle and in neutral pro- and supination, the therapist slowly applies manual resistance at the distal forearm to an isometric contraction up to moderate resistance into one or other rotation (with the right hand for a right shoulder). At the same time, the therapist feels for movement of the head of humerus or alteration in the quality of contraction, watches for altered scapular and trunk movement and enquires about provocation of symptoms. The arm is then moved some way into elevation, usually initially in the plane of the scapula, with the test position chosen on the basis of the principles outlined above. The isometric contraction is repeated, with the same assessment. The contraction is repeated in different positions in range until one is found in which any of the abnormal features outlined above is identiﬁed, or to full range in the symptom provoking direction (Fig. 1A and B). The test is then repeated isometrically through the same positions with the resistance towards the opposite rotation. If a thrower indicates pain in the late cocking position of a throw, for example, the ﬁnal position examined would be one of abduction/ horizontal extension and lateral rotation, whereas for a swimmer with pain at catch in the freestyle stroke, the end position would be ﬂexion combined with medial rotation. For an elderly sedentary person with a painful arc during elevation, testing may be performed below and above the range of the painful arc initially, with the ﬁnal positions approximating that found to be painful on active movement assessment. The test is then repeated isotonically through the same test positions. Initially, the patient’s arm is taken through full rotation passively, to teach the required movement, with the task then made assisted active, gradually adding resistance to the movement. The resistance added is never more than light to moderate, as the patient must be able to readily achieve the movement for the therapist to detect any abnormality. The isotonic resistance is applied slowly. As with the isometric contractions, all tests into one rotation are usually done together, followed by those in the opposite direction, as patients ﬁnd this procedure easier to master than doing both rotations in the same position before moving on to the next position. The test is then repeated in the opposite direction. Care must be taken to ensure that the movement remains one of rotation, as the patient will tend to push out into abduction with resisted lateral rotation Manual Therapy (2003) 8(4), 247–253

ARTICLE IN PRESS Stability of glenohumeral joint 249

Fig. 1—Dynamic Rotary Stability Test. (A) and (B) show the test performed in two different parts of the range of movement.

or into adduction with medial rotation, leading to a different feel of humeral head movement. This tendency becomes more obvious as the test is undertaken further into abduction range. To facilitate maintenance of a rotary movement, the therapist can ask the patient to point the tip of the elbow towards an appropriate imaginary spot on the wall in line with the upper arm and to keep it still, spinning the forearm around that point. Inappropriate resistance by the therapist, such that the patient is unable to achieve pure rotation is another cause of incorrect movement during the test. The therapist must be aware of and adjust to the normal change in strength, particularly in lateral rotation, with the arm higher in range. The therapist must also ensure that the resistance is always provided at right angles to the direction of movement, to facilitate pure rotation. The test is then progressed further, with increased speed, eccentric movement, quick changeovers from medial to lateral rotation and vice versa, concentric to eccentric changeovers, as indicated by the patient’s presentation and lifestyle/sport requirements. Sometimes, asking the patient to perform quick reversals of the movement while holding a small weight in the hand, will demonstrate the abnormal translation Manual Therapy (2003) 8(4), 247–253

more effectively than against manual resistance. Commonly the patient can detect their own lack of control or stability in different positions and therefore should always be asked how it compares to testing the other shoulder at the same angle. With each stage of the test, the aim is to ﬁnd the position where control is lost, with control identiﬁed as a smooth, controlled contraction, with no abnormal translation, pain or apprehension. Having found this position, the arm should be moved away from that position in small increments, and re-tested each time, until a position is reached where control is reestablished. Positions of control are usually found with the arm closer to the body, but this may not be the case, depending on the patient’s functional activities. Therefore, the process should be undertaken in both directions – that is, into more and less abduction/ﬂexion. Working on the principle of training overload (Kibler & Chandler 1994), the closer to the point of loss of control that can be identiﬁed for future re-training, the more quickly control is likely to be regained. Often patients have isometric control further into range than isotonic control and control at speed at a lower range than at higher ranges. If a lack of control is identiﬁed through this test, rehabilitation of the control can be r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 250 Manual Therapy

undertaken, making use of the positions of control to facilitate stabilizing activation of the rotator cuff in positions where the control is lacking. If necessary the patient’s upper arm can be supported during the testing and initial re-training, although ultimately such support must be withdrawn as the training progresses. Throughout the test, the therapist must also observe the scapula and trunk. There should be no compensation (e.g. shoulder shrug) or loss of control (e.g. altered scapular or trunk stability) elsewhere during the test. While the loss of control identiﬁed is not that of the glenohumeral joint, an inability to maintain the scapula on the chest wall or the trunk in a stable position indicates that the overall neuromuscular control at that particular point in range is lacking. Therefore, the test should be repeated in a slightly lower range until the point is found where the scapula or trunk can be maintained in position while the rotator cuff is loaded. This becomes important in management, as if the shoulder is allowed to work when either the scapula or trunk is not stable, abnormal motor programmes are encouraged and the risk of symptom aggravation is increased.

THE DYNAMIC RELOCATION TEST The Dynamic Relocation Test (DRT), initially developed by Guy David (see acknowledgement), is a test of the ability of the rotator cuff, and in particular the lower elements of the rotator cuff, to stabilize the head of humerus in the glenoid by means of co-contraction against a de-stabilizing load. It is predicated on the knowledge that, in normal situations, the rotator cuff functions in co-contraction to stabilize during dynamic activity of the more superﬁcial torque producing muscles (Kibler & Chandler 1994; Wilk 1994; Kibler 1998; David et al. 2000). In shoulders with pain, the co-contraction appears to be lost, although this has not yet been demonstrated under research conditions (David et al. 1997). The test is also based on the knowledge that activity in the normal rotator cuff precedes the onset of movement, whereas in the painful shoulder, this pre-activation appears to be disturbed and often delayed. Such muscle activity draws the humeral head into the glenoid, stiffening the joint and minimizing the amount of translation available (Wilk 1994; Kibler 1998). The best position in which to evaluate rotator cuff co-contraction is that where the ﬁbres of the relevant muscles are optimally aligned to achieve relocation of the humeral head – 60–801 of abduction in the scapular plane (Wilk 1994). Once the ability to isolate the co-contraction has been determined in this position, the ability to create the dynamic relocation r 2003 Elsevier Ltd. All rights reserved.

can be evaluated in different positions, as indicated above for the DRST, and during different tasks. Technique The starting position for the test is with the patient either sitting or supine and the arm supported in approximately 60–801 of elevation in the scapular plane, neutral rotation and 901 elbow ﬂexion. During the test, the therapist asks the patient to draw the humeral head into the glenoid by means of cocontraction of the transverse force couple (Burkhart 1994, 1996) – subscapularis anteriorly and infraspinatus/teres minor posteriorly. To evaluate this contraction, it is best if the therapist can palpate at least one component of the force couple. Palpation of a contraction of infraspinatus is difﬁcult, because overlapping muscles can cloud the assessment, but subscapularis can be palpated in isolation in the ﬂoor of the axilla. Therefore, before evaluating the relocation manoeuvre, the therapist must ﬁnd subscapularis and place the ﬂat surface of three ﬁngers over this muscle belly. The axilla can be reached from either anterior or posterior, with this handling varying between patients. To ensure that the ﬁngers are correctly placed, resist a gentle medial rotation contraction, at which time the subscapularis can be felt to tighten under the ﬁngers. If the hand is placed anteriorly, pectoralis major can also be felt to contract and the therapist must ensure that distinction can be made between the contraction of these two muscles. If the hand is placed posteriorly, distinction between contraction of latissimus dorsi and subscapularis must also be made. The remainder of the therapist’s hand is placed over the shoulder, so that the proximal phalanges are over either pectoralis major anteriorly or latissimus dorsi posteriorly. These ﬁngers can then palpate and monitor for activity in the superﬁcial muscles during the test. The therapist’s left hand is used when evaluating the patient’s right shoulder. Once the therapist is conﬁdent of hand placement over the belly of subscapularis, the patient is speciﬁcally asked to relax, as they will often develop a level of static muscle tension in anticipation of the test procedure. A description is provided to the patient of the manoeuvre, making use of visual images such as ‘draw your arm bone up into its socket’ or ‘suck the arm into its socket’. Extremely gentle resistance is applied longitudinally along the shaft of the humerus with the right hand and the patient asked to pull the humeral head back into the glenoid, as has just been explained (Fig. 2). Initially, most patients will pull in strongly with all superﬁcial muscles, with or without any contribution from the rotator cuff. Localized contraction can be facilitated by instructing the patient to reduce the effort used to produce the movement, Manual Therapy (2003) 8(4), 247–253

ARTICLE IN PRESS Stability of glenohumeral joint 251

Fig. 2—Dynamic Relocation Test.

reducing the strength of the contraction gradually more and more until it can be isolated to the rotator cuff. The patient must be encouraged to relax all superﬁcial muscles, in particular, latissimus dorsi, posterior deltoid, pectoralis major and upper trapezius, which tend to be overactive when the movement is ﬁrst attempted. Such relaxation assists the patient in relative isolation of the contraction to the rotator cuff. Light manual contact over these muscles to help the patient recognize when they are contracted can sometimes also help with relaxation. Similarly, deliberately providing some pressure on the belly of subscapularis, so that the contraction can be felt by the patient as increased pressure against the ﬁngers can stimulate a more isolated contraction. If the patient has difﬁculty mastering the task or, if the attempt involves a considerable amount of superior translation of the humeral head and overactivity of upper trapezius, relocation can be facilitated by encouraging a gentle depressive component to the drawing in movement, with the depression only of the head of humerus onto the ﬁngers in the axilla. This facilitatory technique is particularly useful for the patient whose humeral head translates superiorly during active abduction – for example, the patient with a subacromial impingement. Alternative forms of facilitation can be used, with the therapist experimenting until one is found that is useful with a particular patient. A technique that is often useful with patients with anterior instability is to provide a slight medial rotation bias to the longitudinal movement and ask the patient to draw the arm in and backwards, thus creating a slight external rotation moment in conjunction with the co-contraction. Similarly the simple imagery of ‘widening your shoulders’ or thinking of excessive protraction as a closed book and ‘opening the book’ will be sufﬁcient for some patients. The use of weightbearing can be useful with some patients, rather than the traction movement. Elastic tape applied to simulate compression of the humeral head into the glenoid can also facilitate the Manual Therapy (2003) 8(4), 247–253

Fig. 3—Application of tape to the shoulder aimed at facilitation of the stabilizing activity of the rotator cuff.

contraction. The tape is applied in gentle weightbearing through the elbow with the arm in slight scapular plane abduction and neutral rotation. The tape is spiralled from the midpoint of the anterior arm in line with the deltoid insertion with one band over the anterior deltoid and acromion and ﬁnishing along the spine of the scapula and the other in a similar fashion over posterior deltoid and the acromion, ﬁnishing along the clavicle. An initial half circle around the arm at the level of deltoid insertion is often used as an anchor, laid on with no tension, while tension is used in the tape over the anterior and posterior deltoid to draw the humeral head upwards into the glenoid. A ﬁnal anchor may be needed over the top of the shoulder girdle to ﬁx the proximal ends of the tape. Use of a non-allergenic under-tape is recommended (Fig. 3). The contraction technique is unfamiliar to most people, so interpretation of the test must be made with care. The therapist must not be too quick to assume that the patient cannot achieve the contraction, as even subjects with a normal shoulder and good kinaesthetic sense may take some time and assistance to master the task, though most can usually achieve an isolated contraction relatively readily. Patients with a dysfunctional shoulder may ﬁnd isolation of this contraction to the rotator cuff extremely difﬁcult or impossible to achieve without considerable practice and facilitation. If after such facilitation, the patient is still unable to isolate or even initiate a rotator cuff contraction, the conclusion of lack of dynamic control of the humeral head is reasonable. If the patient is able to achieve the relocation contraction and isolate it in an optimal position against the gentle resistive force, the next step is to ask them to repeat the contraction without the stimulation of the traction. The patient must then be taught to feel the contraction for themselves, with r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 252 Manual Therapy

correct placement of their ﬁngers in the axilla and the sensation of contraction in subscapularis to ensure correct positioning, if they are to practise the manoeuvre at home. The therapist must then ensure that the patient can recognize the co-contraction without external tactile or resistive facilitation as this is the ﬁrst step for home training. Once this can be achieved, progression depends on the starting position for the test. If sitting, ﬁrst wrist extension and then shoulder rotation, both performed with the elbow still supported, tend to be good initial muscular loads to add, as they provide a cognitive distraction without a signiﬁcant physical challenge. If the test were initially performed in supine, movement of the arm towards lateral rotation, still in the scapular plane, and performing the relocation manoeuvre isometrically would be an appropriate early cognitive progression. From this point, progression can be made by maintenance of the co-contraction during any relevant functional movement, with progressively increasing difﬁculty. The aim is to ﬁnd the point in range in the speciﬁc movement where the patient has control and where such control is lost. If the patient has a positive apprehension/relocation test, progression towards dynamic control of the humeral head in the apprehension position (that is, full lateral rotation in abduction) can be appropriate, in addition to development of control during more functional tasks. If abnormality of control were detected during the DRST, the test position in which loss of control was found can be re-evaluated following facilitation of dynamic relocation, to determine whether control was enhanced with this manoeuvre. If control is improved, the prognosis of beneﬁt from a dynamic rehabilitation program making use of the concepts of relocation and control is good. The test manoeuvres can be used during management of a dynamic control problem of the shoulder, as described in Magarey and Jones (2003). The authors are still working on ways of measuring the outcome of these tests objectively, so that they can be used as objective clinical and research outcome measures. Similarly, as yet, we have no ﬁgures on what could be considered normal, nor any measures of reliability or validity. At this point, manual palpation of humeral head translation or alteration of quality of contraction are the only measures of normality during the DRST, while palpation of rotator cuff contraction is the only clinical measure of normality for the DRT. This test can be measured objectively, with in-dwelling EMG electrodes, not appropriate for a clinical setting and by means of dynamic ultrasound, where the movement of the humeral head into the glenoid can be visualized, a technique also not ideal as yet for most clinicians. Research is continuing to ﬁnd ways in which more r 2003 Elsevier Ltd. All rights reserved.

objective measures, similar to those identiﬁed for the deep cervical ﬂexors (Jull 2000; Jull et al. 2002) or transversus abdominus (Richardson et al. 1999) can be applied to these test procedures. In the meantime, these tests have been found clinically useful in identiﬁcation of the patient with poor dynamic control of the glenohumeral joint and in management of these patients.

Acknowledgements Guy David, at the time of development of the DRT, was a PhD student working at the School of Physiotherapy, University of South Australia. Shortly after his return to Israel, he was tragically killed in a cycling accident. We would like to acknowledge Guy’s contribution to the development of the DRT and progression of our understanding of the function of the rotator cuff.

References Burkhart SS 1994 Reconciling the paradox of rotator cuff repair versus debridement: A uniﬁed biomechanical rationale for the treatment of rotator cuff tears. Journal of Arthroscopic and Related Surgery 10(1): 4–19 Burkhart SS 1996 A uniﬁed biomechanical rationale for the treatment of rotator cuff tears: debridement versus repair. In: Burkhead WZ (ed.) Rotator Cuff Disorders. Williams and Wilkins, Baltimore, ch 21, pp 293–312 David G, Jones MA, Magarey ME, Sharpe MH, Dvir Z 1997 Rotator cuff muscle performance during glenohumeral joint rotations: An isokinetic, electromyographic and ultrasonographic study. Tenth Biennial Conference, Manipulative Physiotherapists Association of Australia, Melbourne David G, Magarey M, Jones M, Turker . K, Sharpe M, Dvir Z 2000 EMG and strength correlates of selected shoulder muscles during rotations of the glenohumeral joint. Journal of Clinical Biomechanics 2: 95–102 Harryman DT, Sidles JA, Harris SL, Matsen FA 1992 Laxity of the normal glenohumeral joint: A quantitative in vivo assessment. Journal of Shoulder and Elbow Surgery 1(2): 66–76 Harryman DT, Sidles JA, Matsen FA 1990 The humeral head translates on the glenoid with passive motion. In: Post M, Morrey BF, Hawkins RJ (eds) Surgery of the Shoulder. Mosby Year Book, St Louis, ch 43, pp 186–190 Howell SM, Galinat BJ 1987 The containment mechanism: The primary stabilizer of the glenohumeral joint. Paper read at Annual Meeting of American Academy of Orthopaedic Surgeons, San Francisco, January 23 Howell SM, Galinat BJ, Renzi AJ, Marone PJ 1988 Normal and abnormal mechanics of the glenohumeral joint in the horizontal plane. Journal of Bone and Joint Surgery 70A(2): 227–232 Jull GA 2000 The physiotherapy management of cervicogenic headache: A randomised controlled trial. In: Singer KP (ed.) Abstracts of the 7th Scientiﬁc Conference of the International Federation of Orthopaedic Manipulative Therapists in conjunction with the Biennial Conference of the Manipulative Physiotherapists Association of Australia, p 87 Jull G, Trott P, Potter H, Zito G, Kiere K, Shirley D, Emberson J, Marschner I, Richardson C 2002 A randomised controlled trial of exercise and manipulative therapy for cervicogenic headache. Spine 27(17): 1835–1843 Kapandji IA 1982 The Physiology of Joints. Annotated Diagrams of the Mechanics of the Human Joints. Vol. 1, Upper Limb, 5th edn. Churchill Livingstone, Edinburgh Kibler B 1998 Shoulder rehabilitation: Principles and practice. Medicine and Science in Sports and Exercise S40–S50, http:// www.wwilkins.com/MSSE Kibler WB, Chandler TJ 1994 Sport-speciﬁc conditioning. The American Journal of Sports Medicine 22(3): 421–432 Manual Therapy (2003) 8(4), 247–253

ARTICLE IN PRESS Stability of glenohumeral joint 253

Lippitt S, Matsen F 1993 Mechanisms of glenohumeral joint stability. Clinical Orthopaedics and Related Research 291: 20–28 Lippitt SB, Vanderhooft E, Harris SL, Sidles JA, Harryman DT, Matsen FA 1993 Glenohumeral stability from concavitycompression: A quantitative analysis. Journal of Shoulder and Elbow Surgery 2(1): 27–35 Magarey ME, Jones MA 2003 Dynamic evaluation and early management of altered motor control around the shoulder complex. Manual Therapy. This issue Norkin CC, Levangie PK 1988 Joint Structure and Function, A Comprehensive Analysis. FA Davis Company, Philadelphia Poppen NK, Walker PS 1978 Forces at the glenohumeral joint in abduction. Clinical Orthopaedics and Related Research 135: 165–170 Richardson C, Jull G, Hodges P, Hides J 1999 Therapeutic Exercise for Spinal Segmental Stabilization in Low Back Pain:

Manual Therapy (2003) 8(4), 247–253

Scientiﬁc Basis and Clinical Application. Churchill Livingstone, Edinburgh Saha AK 1971 Dynamic stability of the glenohumeral joint. Acta Orthopaedica Scandinavica 42: 491–505 Schenkman M, Rugo de Cartaya V 1987 Kinesiology of the shoulder complex. Journal of Orthopaedic and Sports Physical Therapy 8: 438–449 Schenkman M, Rugo de Cartaya V 1994 Kinesiology of the shoulder complex. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, ch 2, p 15–33 Soderberg GL 1986 Kinesiology. Application to Pathological Motion. Williams and Wilkins, Baltimore Wilk KE 1994 Current concepts in the rehabilitation of athletic shoulder injuries. In: Andrews JR, Wilk KE (eds) The Athlete’s Shoulder. Churchill Livingstone, New York, ch 30, p 335–354

r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 254–256 r 2003 Elsevier Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00055-9

Professional issue

Justifying the on-going physiotherapy management of long-term patients T. Flanagann, P. Coburnn, P. Harcourtn, M. Zylinskin, G Jullw Transport Accident Commission, Australia, w Cervical Spine and Whiplash Research Unit, Department of Physiotherapy, The University of Queensland, Australia

n

a proportion of patients may require on-going physiotherapy treatment to maintain a functional level that is optimal but below their pre-injury status. Without this intervention, the patients’ functional condition will deteriorate. This type of treatment is referred to as maintenance physiotherapy (Flanagan & Green 2000). A clinical framework which focuses on function, rather than pain, and utilizing validated functional outcome measures to guide the objective management of long-term patients, is presented for consideration (Fig. 1). The ﬂow-chart provides the treating musculoskeletal physiotherapist with a means to assess the efﬁcacy of on-going physiotherapy management. It illustrates the distinction between rehabilitation and maintenance physiotherapy and describes a process to indicate an end point in physiotherapy management in long-term patient care.

INTRODUCTION Clinical justiﬁcation for long-term musculoskeletal patients receiving regular, on-going treatment is an important issue for the physiotherapy profession. In many cases there is no end point and there may be no rationale for continued physiotherapy treatment. The purpose of this paper is to present musculoskeletal physiotherapists and their patients with a ﬂow-chart which may help to determine when on-going physiotherapy treatment is indicated in the management of long-term patients. The common belief that musculoskeletal physiotherapists will rehabilitate all patients to pain-free independence may have contributed to the development of the situation where some patients receive what could be described as excessive physiotherapy. An aim of any physiotherapy plan should be to help patients achieve an end point with an optimum level of function. Not all patients will achieve an end point of rehabilitation to pre-injury function or have no ongoing pain. Work by Waddell on low back pain describes the need for health professionals to focus on function, rather than pain in the management of long-term conditions (Waddell et al. 1992). However,

MANAGEMENT STRATEGY The proposed strategy commences with the treating musculoskeletal physiotherapist selecting appropriate and validated functional outcome measures for that patient. The musculoskeletal physiotherapist determines when to introduce these measures and should explain the rationale for their use to the patient. The use of functional outcome measures emphasizes the pivotal role of patients in the on-going assessment and management of their condition (Waddell et al. 1992; Beattie & Maher 1997). The musculoskeletal physiotherapist’s selection of the most appropriate functional outcome measure should also make note of the responsiveness peculiar to each outcome measure (Beurskens et al. 1996; Scudds 2001). The ﬂow chart assumes the use of at least two subsequent validated functional outcome measures to assist in measuring the progress of a longterm patient. The process shows three possible

Received: 15 January 2003 Accepted: 7 April 2003 Tony Flanagan, B App Sci (Physiotherapy), Grad Dip Manipulative Physiotherapy, Private Practitioner, Fairﬁeld, Physiotherapy Consultant, Transport Accident Commission, Paul Coburn, B App Sci (Physiotherapy), B Sci (Human Movement Studies), Private Practitioner, South Morang, Physiotherapy Consultant, Transport Accident Commission, Peter Harcourt, MBBS (Melbourne), Medical Director, Victorian Institute of Sport, Medical Convenor, Transport Accident Commission, Maria Zylinski, B App Sci (Physiotherapy), Manager Medical Panel, Transport Accident Commission, Gwendolen Jull, M Phty, PhD, FACP, Associate Professor, Department of Physiotherapy, The University of Queensland, Cervical Spine and Whiplash Research Unit. Correspondence to: TF, Fairﬁeld Physiotherapy and Sports Injuries Centre, 181 Station Street, Fairﬁeld 3078, Australia. Tel.: +03 9489 7744; Fax: +03 9486 4030; E-mail: [email protected] 254

ARTICLE IN PRESS Justifying the on-going physiotherapy management of long-term patients 255

The patient with a long-term problem

Use of two Functional Outcome Measures Referral to other appropriate management

Improved

Unchanged or Deteriorated

Uncertain diagnosis and / or Management

Rehabilitation

Review diagnosis & management

Attained preinjury function.

Correct diagnosis / management Maintenance Self-management focus

Cease physiotherapy management.

Measured therapy break Test physiotherapy management Improved or Unchanged

Cease physiotherapy management

Deteriorated

Continue physiotherapy management

Regular review to monitor / justify Physiotherapy management

Fig. 1—Flow-chart for use of outcome measures and measured therapy breaks in the physiotherapy management of long-term patients.

scenarios: patient improvement, patient deterioration or patient stabilization. This enables the treating musculoskeletal physiotherapist to classify patients into rehabilitation, review or maintenance phases. Rehabilitation phase In the rehabilitation phase of management there is improvement in the patient’s functional outcome measures. The regular use of functional outcome measures is used to monitor and review the patient’s progress. If a patient recovers to their pre-injury functional state with no pain then physiotherapy management should cease. A plateauing of measures, below pre-injury status marks the completion of rehabilitation. Review of diagnosis and management If a patient’s functional outcome measures deteriorate or remain unchanged the working diagnosis and management should be reviewed. Is the diagnosis correct? Has there been a complication? Does the patient have a chronic pain syndrome? Are there nonclinical drivers? Is there non-organic pathology? Is referral to another practitioner indicated? These are the type of questions a musculoskeletal physiotherapist is likely to ask himself/herself. If the review process indicates a referral to other ‘‘appropriate management’’, then the adaptation of an alternative Manual Therapy (2003) 8(4), 254–256

management path should be subject to the same measurement criteria to determine its efﬁcacy and appropriateness. Prior to this review, the musculoskeletal physiotherapist must ensure that the patient has understood and is compliant with all self-management strategies related to his/her condition. The musculoskeletal physiotherapist’s knowledge should be adequate to manage the particular condition to give the patient the best chance of beneﬁting from physiotherapy management (Taimela et al. 2000). Furthermore, the treating musculoskeletal physiotherapist has a responsibility to ensure that their knowledge and skills are sufﬁcient to optimize a patient’s recovery beyond that which would be expected without treatment or with the effect of placebo alone (Hansen et al. 1993; Butler, 1998; Bronfort et al. 2001). Maintenance phase The third possible scenario of the ﬂow chart is one where the functional outcome measures have stabilized in a patient who has not achieved pre-injury functional status and after a clinical review there are no changes to the diagnosis and management. This phase is referred to as maintenance, and the patient requires explanation of their clinical status. This includes a focus on the patient’s role in their maintenance of function incorporating a review of r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 256 Manual Therapy

lifestyle, therapeutic exercise, general ﬁtness, work, posture, etc. A practical review of pain management to enable patients to better understand the relationship between pain and their condition should be reinforced. A patient who continues to attend physiotherapy management to prevent functional deterioration is receiving maintenance physiotherapy. The difﬁcult issue for the treating musculoskeletal physiotherapist and patient is to decide whether physiotherapy management is achieving the desired effect or if selfmanagement strategies alone are enough to maintain a patient’s optimal functional status. A measured therapy break is introduced into the decision-making process. A measured therapy break involves the serial use of functional outcome measures before and after an appropriate period of cessation of physiotherapy management. The key issue is that the practice of a measured therapy break has assisted all stakeholders to decide the necessity of physiotherapy management to maintain a patient’s condition at an optimal level. The process described in the ﬂow-chart must be ongoing. These patients require regular review utilizing functional outcome measures and measured therapy breaks. They may also require review by other health professionals to ensure that physiotherapy management continues to be justiﬁed. Therefore, physiotherapy management strategies must be challenged so that a patient in this category is given every opportunity to become independent in maintaining his/her optimal functional status.

CONCLUSION A ﬂow-chart to manage long-term patients and justify on-going physiotherapy treatment is proposed to assist both the patient and musculoskeletal physiotherapist in determining the necessity of

r 2003 Elsevier Ltd. All rights reserved.

continuing physiotherapy management. It is based on the measurement of a patient’s condition with a focus on function through the use of validated functional measures and a practice of measured therapy breaks. It highlights the need for an ongoing review of the working diagnosis and management strategy as well as the vital role of the patient in their management.

Acknowledgements Preparation of this article was made possible by the ﬁnancial and technical assistance of the Victorian Transport Accident Commission.

References Beattie P, Maher C 1997 The role of functional status questionnaires for low back pain. Australian Journal of Physiotherapy 43: 29–38 Beurskens A, De Vet H, Koke A 1996 Responsiveness of functional status in low back pain: a comparison of different instruments. Pain 65: 71–76 Bronfort G, Evans R, Nelson B, Aker P, Goldsmith C, Vernon H 2001 A randomised clinical trial of exercise and spinal manipulation for patients with chronic neck pain. Spine 26: 788–799 Butler D 1998 Integrating pain awareness into physiotherapy – wise action for the future. In: Gifford L (ed) Topical Issues in Pain. NOI Press, Cornwall; pp 1–23 Flanagan T, Green S 2000 The concept of maintenance physiotherapy. Australian Journal of Physiotherapy 46: 271–278 Hansen F, Bendix T, Skov P, Jensen C, Kristensen J, Krohn L, Schioeler H 1993 Intensive, dynamic back-muscle exercises, conventional physiotherapy, or placebo-control treatment of low-back pain. Spine 18: 98–108 Scudds R 2001 Pain outcome measures. Journal of Hand Therapy 14: 86–90 Taimela S, Takala E, Asklof T, Seppala K, Parvianen S 2000 Active treatment of chronic neck pain. Spine 25: 1021–1027 Waddell G, Somerville D, Henderson I, Newton M 1992 Objective clinical evaluation of physical impairment in chronic low back pain. Spine 17: 617–628

Manual Therapy (2003) 8(4), 254–256

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 257–260 r 2003 Elsevier Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00054-7

Case report

Treatment of osteitis pubis via the pelvic muscles A. McCarthy, B. Vicenzino Department of Physiotherapy, School of Health and Rehabilitation Sciences, University of Queensland, Queensland, Australia

lasted for 2 days, and then eased slightly. He had avoided exercise since. Pain was localized over the lower part of the rectus abdominus muscle. Running, kicking the ball, sit-up exercises and leg lifting exercises aggravated it. He was able to continue with a weight programme. Even split lunge exercises were not provocative. Only the aforementioned abdominal exercises were sore. One year previously the athlete had experienced a lower back injury, which had resolved with a physiotherapy regime concentrating on stabilizing exercises for the lumbar spine. The athlete had not sustained a maintenance routine of those exercises. Pain at that time had been central in a band across the low lumbar spine, bilateral buttocks and posterior thigh. His pain was provoked by unilateral weight bearing and hopping. For the current problem the athlete required no analgesic medication, and had not used anti-inﬂammatory drugs of his own recourse. There had been no imaging of either the lumbar spine or pelvis. Running was his most limited functional activity; therefore it was examined as the ﬁrst priority. A visual analysis of the athlete’s running gait showed the following abnormalities: markedly limited trunk motion about the waist/lumbar spine, that is, little extension or rotation; decreased hip extension, internal and external rotation; and decreased stride length. In order to discover why these movement patterns were present, an examination of the lumbar spine, sacroiliac joints and hips was undertaken. The lumbar spine and sacroiliac joints were grossly normal. At the hips, there was a ﬁxed ﬂexion deformity of 51 bilaterally on Thomas testing (Magee 1997). The athlete was unable to isolate active hip extension from lumbar extension. Hip rotations were limited to approximately 251 and this was more apparent at the limit of extension, suggesting that anterior hip structures lacked adequate extensibility. External rotation was most reduced. The Thomas test

INTRODUCTION Osteitis pubis is an entity that has been discussed greatly in the orthopaedic and sporting literature from the perspective of medical intervention (Fricker et al. 1991; Batt et al. 1995; Holt et al. 1995). Many studies recommend injection of the symphysis pubis with corticosteroid as early as possible, although the outcome reported is variable (Batt et al. 1995; Holt et al. 1995). Fricker et al. (1991) found that the average time to recovery was 9.6 months. Rest is the most common treatment prescribed, but a programme of aggressive stretching of the adductors has also been recommended (Batt et al. 1995, Holt et al. 1995). The rehabilitation of this disorder was not discussed in these papers. The purpose of this case report is to describe an alternate approach to assessment and treatment looking at the athletes motion patterns in provocative activities, in addition to focussing on the local signs.

CASE REPORT A 20-year-old male Gaelic football player attended the physiotherapy clinic having experienced low central abdominal pain, which had been present for 6 weeks, (Fig. 1). Three weeks prior to attendance, he had taken a tumble forwards after scoring a goal and experienced an immediate increase in pain, which Received: 7 April 2003 Accepted: 7 April 2003 Ann McCarthy MCSP, MPhty Studies (Sports), Head, Musculoskeletal Physiotherapy, Royal Free Hospital, London, Bill Vicenzino, PhD, MSc, Grad Dip Sports Phty, BPhty, Senior Lecturer, Director Musculoskeletal Pain and Injury Research Unit, Coordinator of Masters of Physiotherapy Studies (Sports), Department of Physiotherapy, University of Queensland, Australia. Correspondence to: B.V. Department of Physiotherapy, School of Health and Rehabilitation Sciences, University of Queensland, 4072 Queensland, Australia. Tel.: +61 7 3365 2781; Fax: +61 7 3365 2775; E-mail: [email protected] 257

ARTICLE IN PRESS 258 Manual Therapy

Fig. 1—Body chart depicting the pain presentation.

demonstrated an early anterior pelvic tilt and decreased length of iliopsoas and the quadriceps muscles. The test position became painful if the pelvis was manually stabilized by the examiner. Tests reproducing the presenting pain were: stretch and contraction of the adductor groups; an abdominal crunch exercise; palpation of the rectus abdominus insertion on to the pubis and the adjacent inguinal ligaments. In general, the athlete demonstrated a decreased ability to stabilize his lower trunk and pelvis whilst using his legs independently. Treatment was based on a progressive staged graduated therapeutic exercise programme in which isolated protective muscle function was ﬁrst established and then rapidly progressed to incorporate elements of functional tasks, ﬁrst in stable positions progressing to less stable higher demand dynamic movement patterns. Various facilitation techniques such as visual and tactile feedback of correct movement patterns and muscle activation, use of rehabilitation ball as a form of unstable base to facilitate higher level trunk stability control and repetition of staged movement patterns (kicking and running) were used. Initially, speciﬁc re-education of underactive muscles was undertaken. Gluteus medius work was developed to stabilize the hip in stance, and transversus abdominus to stabilize the low lumbar spine and pelvis. Muscles that exhibited decreased compliance on length tests (e.g., hip ﬂexor and rotator muscles) were released passively with massage techniques and inhibitory stretch techniques, but only after the client had achieved adequate activation of the gluteus medius and transversus abdominus. A home exercise programme of quadriceps, hip ﬂexor and adductor stretches was performed. In the end stages of rehabilitation, the athlete performed highlevel stability work using a large rehabilitation ball in which appropriate contraction of muscles of the pelvis and lower back was encouraged (Figs. 2a and b). r 2003 Elsevier Ltd. All rights reserved.

The athlete attended for seven sessions over a period of 5 weeks. He was able to jog after three sessions and had performed a 2 km run prior to discharge, without pain and had returned to full participation of his chosen sport. At 3 months postdischarge the athlete reported being pain-free and not limited in any way by his previous symptoms.

DISCUSSION Osteitis pubis is a condition affecting the pubic symphysis in mainly athletic groups, although it has been reported post-partum and after surgery (Fricker et al. 1991). The largest incidences are reported in the football codes, but it has been reported in several other sports (Holt et al. 1995). The most frequent symptoms reported are pubic pain, lower abdominal pain and adductor pain. Commonly reported examination signs are: pubic symphysis tenderness, adductor longus muscle insertional tenderness, decreased hip external rotation, pelvic malalignment/sacroiliac joint dysfunction, and restricted hip abduction (Fricker et al. 1991). Radiographs show periosteal reaction, sclerosis and in advanced cases demineralisation of the cortical bone (Zachezewski et al. 1996). The symphysis may appear widened in the acute case, but chronic cases may demonstrate joint narrowing. Bone scan conﬁrms the diagnosis, with increased uptake in the area of the symphysis. This must be differentiated from a pubic stress fracture (Fricker et al. 1991). Fricker et al. (1991) state that there appears to be no correlation between the bone scan or radiograph results and the severity or duration of symptoms. Although there was no imaging in the case described, the symptoms, signs and history are of a classic presentation of osteitis pubis. Manual Therapy (2003) 8(4), 257–260

ARTICLE IN PRESS Treatment of osteitis pubis 259

Fig. 2—(a and b) High level stabilizing exercises using a large rehabilitation ball.

The aetiology of the condition is still the subject of debate. In the athletic population the condition has been described as overuse in nature. Repetitive microtrauma due to the pull of the rectus abdominus muscle has been suggested. With chronic cases, instability of the symphysis and/or sacroiliac joint’s ilial portion has been noted (Fricker et al. 1991). This instability may be an additional cause of stress at the symphysis. Bowerman (1977) suggested that repetiManual Therapy (2003) 8(4), 257–260

tive adductor muscle pull at their inferior pubic ramus insertion could produce a shearing force across the pelvis. All of the above may be partial explanations. However, they do not address the question of why the muscular over-pull or microtrauma is occurring. There is a tendency for practitioners to focus on the inciting events or mechanism of injury itself, rather than looking for causative factors, both intrinsic and extrinsic to the athlete (Meeuwisse 1994). r 2003 Elsevier Ltd. All rights reserved.

ARTICLE IN PRESS 260 Manual Therapy

Verrall et al. (1998), have used magnetic resonance imaging (MRI) scanning to examine pubic symphyses of clinically diagnosed cases of osteitis pubis. It revealed that fracture lines were present which were described as tension stress fractures (Verrall et al. 1998). These fractures may have possibly been caused by torsional stresses across the pelvis. The MRI also demonstrated bone cysts, symphyseal disc irregularity and superior pubic ligament hypertrophy. Dalstra and Huiskes (1995) have shown that the greatest load across the pubic symphysis occurs at the beginning of swing phase of gait when there is a degree of rotation across the pelvis. Using ﬁnite element analysis techniques, they have shown that muscle forces have a considerable effect on the stress patterns. Removal of muscle forces (21 of the muscles attached to the pelvis were included in the model) from their analysis showed that the load transfer from the hip joint forces is directed almost totally through an axis from the sacroiliac joints to pubic symphysis. Muscle forces across the pelvic bone, when all are acting in concert as in their model, keep the stress distributions fairly constant during the walking cycle. Dalstra and Huiskes (1995) also suggest that stress changes being kept minimal, is a favourable state with regard to fatigue failure of the bone material. It would seem reasonable to propose therefore, that an imbalance in these muscle forces might alter stresses across the pubic symphysis, reducing the motion at the symphysis and increasing the torsion of the bone. The rehabilitation approach described herein identiﬁed motor dysfunction due to altered muscle function. This motor dysfunction could have increased the susceptibility of the athlete to osteitis pubis (Meeuwisse 1994). Improving movement patterns and muscle function enables the therapist to garner an effect on the presenting signs and symptoms relatively quickly. This approach, rather than one aimed at the local pathology at the site of pain as the ﬁrst priority, enables the therapist to better direct the rehabilitation process and perhaps lessen the chances of recurrence, which is 25% in males (Hogan 1998). In the present case there was an intrinsic lack of hip extension and to a lesser extent, external rotation. We hypothesize that by improving these motions the athlete experienced a reduction in excessive strain on the pelvis by allowing a more correct balance of muscle pulls across it. Hogan (1998) describes a similar rehabilitation approach conducted in a step-wise process. Rehabilitation programmes should indeed progress from less stressful and simple early stages to more complex and demanding later stages, eventually leading to a return to sport. However, the hierarchical model suggested by Hogan (1998), appears not to allow for any interplay between the phases. Proprioceptive and sport skills need not be left until late in the process. r 2003 Elsevier Ltd. All rights reserved.

Vigorous or strenuous stretches have been recommended (Batt et al. 1995; Holt et al. 1995). In this case, it was determined that rectus abdominus and adductor over-pull was due to a dysfunction of the trunk and hip stabilizing muscles. Stretching is not likely to facilitate these muscles. There must be a prior muscular re-education programme before stretching begins in earnest. These stretches must be integrated in the re-education process by concentrating on maintaining control of the pelvic position. The stretches need not be vigorous, but should be effective. Proprioceptive neuromuscular facilitation type stretches have been shown to be the most effective (Moore and Hutton 1980; Sady et al. 1982).

CONCLUSION Osteitis pubis is a self-limiting condition that is becoming more readily identiﬁed by clinicians. Although it will eventually resolve with prolonged rest, identiﬁcation and restoration of abnormal motion patterns and muscle dysfunction about the pelvis and hip may lead to a more speedy recovery. In the general treatment of sporting injuries, betterdirected rehabilitation may be achieved when identiﬁcation of these factors is made a priority, with less focus of treating the signs locally at the site of pain.

References Batt ME, McShane JM, Dillingham MF 1995 Osteitis pubis in collegiate football players. Medicine and Science in Sport and Exercise 27(5):629–633 Bowerman JW 1977 Radiology and injury in sport. In: Radiology and Injury in Sport. New York: Appleton-Century-Crofts, pp 224–225 Dalstra M, Huiskes R 1995 Load transfer across the pelvic bone. Journal of Biomechanics 28(6):715–724 Fricker PA, Taunton JE, Ammann W 1991 Osteitis pubis in athletes: infection, inﬂammation, or injury? Sports Medicine 12(4):266–279 Hogan A 1998 A rehabilitation model for pubic symphysis injuries. In: Australian Conference of Science and Medicine in Sport, 13-16/10/1998. Sports Medicine Australia, Adelaide, p 143 Holt MA, Keene JS, Graf BK, Helwig DC 1995 Treatment of osteitis pubis in athletes. American Journal of Sports Medicine 23(5):601–606 Magee DJ 1997 Orthopedic Physical Assessment, 3rd edn. W.B. Saunders Company, Philadelphia. Meeuwisse WH 1994 Assessing causation in sport injury: a multifactorial model. Clinical Journal of Sport Medicine 4:166–170 Moore MA, Hutton RS 1980 Electromyographic investigation of muscle stretching techniques. Medicine and Science in Sport and Exercise 12(5):322–329 Sady SP, Wortman M, Blanke D 1982 Flexibility training: Ballistic, Static or Proprioceptive Neuromuscular Facilitation? Archives of Physical Medicine and Rehabilitation 63:261–263 Verrall G, Slavotinek J, Fon G 1998 Osteitis pubis in Australian rules footballers: A stress injury to the pubic bone. In: Australian Conference of Science and Medicine in Sport, 13-16/ 10/1998. Sports Medicine Australia, Adelaide, ACT, p 227 Zachezewski JE, Magee DJ, Quillen WS 1996 Athletic Injuries and Rehabilitation, 1 edn. W.B. Saunders Co., Philadelphia, PA. Manual Therapy (2003) 8(4), 257–260

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 261–262 r 2003 Published by Elsevier Science Ltd. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00041-9

Book reviews past medical history, present condition, aetiology, examination, imagery review, clinical impression, what action should be taken, treatment and results. Excellent photographs of the imagery are accompanied by clear and detailed descriptions. Comments and key points to be gained from each case are given together with references and further reading. The topics vary from the more common clinical presentations (e.g. intervertebral disc protrusion) to less common mechanical and non-mechanical pathologies. It is a useful book for all clinicians giving a wider vision on the more unusual pathologies and helps alert the practitioner for contraindications as each individual case discussed gives a clear picture of both red and yellow ﬂags. The treatment described includes manipulation, acupuncture, medication and referral to a surgeon where indicated, which may not be the treatments of choice by all clinicians. The author is a chiropractor which explains the form of treatment given. Manual therapists may differ in the choice. It is a useful book to have in one’s clinic to help in the differential diagnosis of problem patients and a good reference book for their clinical and radiographic presentations.

Stroke rehabilitation: guidelines for exercise and training to optimize motor skill. Janet Carr & Roberta Shepherd. 2003. Butterworth-Heinneman. Price: d35.00. 301 pp. ISBN: 0750647124. This new text documents much of the published evidence with respect to motor control, skill aquisition and training following stroke. It is a development of the previously published Motor Relearning Programme for Stroke (1987) by the same authors. The authors state their intention to be the provision of protocols for training the individual following stroke and as such this book provides much of the necessary information. The book is divided into three sections, the ﬁrst introduces theoretical concepts of brain reorganization, task-related training environmental issues and outcome measures which are relevant to any clinician regardless of an individual, speciﬁc therapeutic approach. Section 2 (Chapters 2–5) discusses the evidence for speciﬁc functional activities such as balance, walking, standing up and sitting down as well as reaching and manipulation. The suggested training guidelines are presented alongside each of these chapters. Much of the information in this section has been previously presented in Neurological Rehabilitation: Optimizing Motor Performance (1998) by the same authors. Section 3 discusses impairments and adaptations, strength training and physical conditioning and concludes with an overview of the rehabilitation process following stroke. Whilst this book does present relevant published evidence to support the guidelines, there is little discussion of the initial assessment process, precautions that should be considered or the intervention in the initial acute phase following stroke. In summary, this text is a useful clinical resource, however the guidelines lack speciﬁcity with respect to speciﬁc application, dosage and progression of training.

Sara Luetchford, MCSP, Dip MDT, Dip ETGOM (Physiotherapist) Private practitioner, Milan, Italy

Chronic fatigue syndrome, Christianity and culture. James M. Rotholz. 2002. The Haworth Press, Inc. Price: d17.95. 142pp. ISBN: 0789014939. Chronic fatigue syndrome, Christianity and culture by James M. Rotholz is a good read, one I would recommend for all physiotherapists and health care practitioners as well as patients with chronic disability. Rotholz’s story, told from his personal perspective of a devout Christian and anthropologist whose wife is afﬂicted with chronic fatigue and immune dysfunction syndrome (CFIDS), then acquiring CFIDS himself, provides readers with invaluable insight to what it is like living with a chronic disability. After sharing his personal story of living with CFIDS, Rotholz the anthropologist then provides an excellent overview of the inﬂuence culture and personal experiences have in shaping a person’s individual perspectives and the difﬁculty of appreciating another’s experience when viewed from your own quite different perspective. This contemporary account of the signiﬁcance of environmental and personal factors inﬂuencing individual perspectives highlights the importance of ﬁrst understanding your own perspective (e.g. on disability and what deﬁnes success) to better help you then come to understand another’s perspective without simply imposing your views and values onto them. This ideology is consistent with the modern biopsychosocial model of health and disability that encourages health practitioners to recognise and work with psychosocial as well as physical factors that contribute to a person’s disability and activity/participation restrictions.

Monica Busse, BSc (Physiotherapy) BSc (Med) Hons MSc (Med) MCSP Research Centre for Clinical Kinaesiology University of Wales College of Medicine Cardiff, UK

50 challenging spinal pain syndrome cases. L.G.F. Giles. 2003. Butterworth Heinemann Price: d29.99 247pp ISBN: 0750640081. This book is a well-documented and very readable paperback on 50 cases including the lumbar, thoracic and cervical spine. There is an excellent general introduction giving an up-to-date and well-referenced guideline for assessing the patient. Diagrams of the anatomy of the different spinal regions are included as well as tables on the causes of spinal pain. Each spinal section has an introduction giving the possible causes of spinal pain, a guideline to the physical examination and the clinical presentation. Each case is well presented with a detailed proﬁle of the 261

ARTICLE IN PRESS 262 Book reviews

By reviewing the prevailing western cultural deﬁnition of success (i.e. wealth, status, power) Rotholz convincingly illustrates how success by this deﬁnition is denied to the disabled and along with it the dignity and respect (indispensable elements of a person’s overall well-being) that society affords those who succeed. Rotholz the Christian then puts forward the ‘Biblical approach’, based on faith, hope and love, as promoting a better, fairer alternative set of values than those promoted through the traditional western deﬁnition of success. While the inclusiveness of Rotholz’s deﬁnition of success (‘yfulﬁlling one’s own personal calling according to the gifts and abilities that one possesses.’ p. 96) compared to the traditional western emphasis on wealth, status and power will sit comfortably with most, his open and at times preaching style of arguing the Biblical way will be off putting to some. However, whether you can relate to his strong Biblical perspective or not is irrelevant as the value in this book is the reﬂection it promotes on what living with a chronic disability such as CFIDS is like and how western culture’s deﬁnition of success disadvantages the disabled. Culture and personal experience contribute to shape who we are, how society views us and how we cope with our personal life circumstances. For Rotholz, Christianity has provided a means by which he can accept his disability with dignity and respect. By sharing his story Rotholz encourages us to reﬂect on our own personal/societal culture and views which will in turn help us to better understand our patients’ perspectives when disabled in a society that largely values the abled.

Mark A. Jones, BS (Psych), Cert. Phys. Ther., Grad. Dip. Advan. Manip. Ther., MAppSc (Manip. Ther.) Director, Graduate Programs in Musculoskeletal and Sports Physiotherapy, School of Health Sciences (Physiotherapy discipline) University of South, Australia

How to appraise research: a guide for chiropractic students and practitioners. C. Hagino. 2003. Churchill Livingstone. Price: d19.99. 163pp. ISBN: 0443073791

This workbook is designed to help healthcare professionals critically evaluate published research and to supplement

r 2003 Published by Elsevier Science Ltd.

other texts that comprehensively explain the concepts and theory behind critical appraisal. The strength of this book is that it provides eight articles from the manual therapy literature that are annotated with margin notes that highlight and critique different aspects of the sample articles. Drawbacks are that not all types of articles necessary for evidence-based practice are included and those that are included are not discussed equally. Each chapter includes an article that was published between 1987 and 2000 in a refereed journal. The appraisal concepts that are covered are identiﬁed in a list at the beginning of each chapter. However, one may need to read a chapter carefully to discover where some of the concepts are addressed. The ﬁrst six chapters focus on articles related to intervention and include a case report, a systematic review, and four controlled studies. Correlational study designs are addressed in chapters seven and eight with the presentation of a reliability study and a study of concurrent validity. The comments on each article, which are primarily presented in a question and answer format, assist in the application of knowledge of research and statistics to the sample articles. Explanations of different aspects of appraisal are understandable; however, in places they are brief and very few references are provided. The discussion of randomized controlled trial studies appears to be more comprehensive than the discussion of other types of intervention studies. Likewise, greater emphasis is given to the interpretation of reliability studies than to studies related to validity. As an example, the book does not include a cohort study that addresses concepts of validity such as sensitivity and speciﬁcity which would be helpful for a clinician apply research evidence to the diagnostic process. Although written for chiropractic students and practitioners, the critical appraisal skills addressed in this book are applicable to all healthcare professionals. This book appears to be appropriate for students and clinicians who want to practice critical appraisal of the literature and have other texts on research design available to consult for additional information on the concepts that are presented. David Scalzitti, PT, MS, OCS Associate Director—Research Services American Physical Therapy Association Alexandria, VA, USA

Manual Therapy (2003) 8(4), 261–262

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 263 r 2003 Published by Elsevier Ltd. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00057-2

Letter to the editor

The second point of issue is that of the therapeutic exercise programme. It is rare for authors designing exercise programmes to provide details that cover all aspects of exercise prescription process. I felt that this article did this very clearly. This can only lead to improved reliability and generalisability to clinical populations. The issue of stretching however is poorly referenced with no explanation as to why stretching helps with the conditioning process. There are also no reasons cited for multiple sets of 3–5 or a 20–30 s hold despite this often being a ‘colloquial’ prescription. Despite this I found the article informative and instructive which I think is the aim of the Masterclass article.

Dear Editors, I really enjoyed reading the recent edition of Manual Therapy and you mentioned that feedback from a consumer would be useful. I have some comments on the Masterclass article by Bill Vincenzino. The author maintains that there are no reviews of manipulative therapy for lateral epicondylalgia (LE) as it appears that the condition is one of a soft tissue disorder. I felt that the review neglected two techniques that have been previously mentioned in this journal and therefore appears to be incomplete. Hunter (1998) presented a speciﬁc soft tissue mobilization technique in a previous Masterclass aimed at the restoration of the tissue dysfunction such as this. This author also alluded to the ‘deep frictions’ technique advocated by Cyriax (1993). The article referred to a number of other authors suggesting there was accumulating evidence for the effectiveness of each technique on the basis of ‘feasible biological hypothes[e]s’. I feel that some mention of these techniques was warranted as the article aimed to present a wide musculoskeletal perspective.

References Hunter G 1998 Speciﬁc soft tissue mobilization in the management of soft tissue dysfunction. Manual Therapy. 3(1): 2–11 Cyriax J 1993 Cyriax’s Illustrated Manual of Orthopedic Medicine, 2nd edn. Butterworth Heinmann, Edinburgh, pp 19–23

Matthew Daly

263

ARTICLE IN PRESS Manual Therapy (2003) 8(4), 264–265 r 2003 Elsevier Ltd. All rights reserved. 1356-689X/03/$ - see front matter doi:10.1016/S1356-689X(03)00058-4

Letter to the editor

management of known deﬁcits or physical impairments of the condition. As outlined by Shrier (2000) stretching is in a similar situation as that outlined for the SSTM and deep frictions, that is, the evidence for its apparent clinical effectiveness has yet to be elucidated. Apart from a review of clinical studies of stretching as a preventative factor in the reduction of muscle injury, Shrier (2000) presents a synopsis of the basic science literature on the mechanisms of stretching and proposes that the conditioning response to stretching appears to be due to changes in stretch pain tolerance not actual tissue compliance (Shrier 2000). Much work needs to be done to better understand the mechanism(s) of action of many of the soft tissue techniques used in musculoskeletal physiotherapy if we are to improve upon our understanding of their place in clinic and their exact prescription parameters.

Dear Editors Re: Letter to editor from Mr Mathew Daly about the Masterclass: ‘Vicenzino B 2003 Lateral epicondylalgia: a musculoskeletal physiotherapy perspective. Manual Therapy 8(2): 66–79’ I would like to thank the editors for granting me the right of reply to the letter from Mr Daly. I focus my reply on what appears to be the predominant basis of Mr Daly’s letter: that is, the omission of two speciﬁc soft tissue mobilization techniques and insufﬁcient rationale for stretching techniques. Mr Daly refers to my statement that ‘y there are no reviews of manipulative therapyy’ for LE and claims that the Masterclass is incomplete because it neglected two techniques: the speciﬁc soft tissue mobilization (SSTM) of Hunter (1998) and the deep frictions of Cyriax (1993). There were no systematic reviews that dealt speciﬁcally with any physiotherapy treatments for lateral epicondylalgia (LE) at the time of writing. However, since publication of the Masterclass, a systematic review that speciﬁcally evaluated physical therapies, including friction massage and mobilization therapy, has been published (Smidt et al. 2003). Interestingly, it did not identify the papers that Mr Daly suggested should have been included in my Masterclass. As an update for readers of Manual Therapy and in brief, the systematic review of Smidt et al. (2003) found that although a number of studies have evaluated soft tissue mobilization techniques, none were of sufﬁcient quality to allow a valid interpretation of the data. Mr Daly proposed that the SSTM and deep friction techniques should have been included in the clinical approach that I outlined because of ‘feasible biological hypothes[e]s’(Hunter 1998). These ‘feasible biological hypothes[e]s’ refer to the importance of ‘y carefully applied movement in restoring the tensile strength and functional biomechanical properties of healing tissue’ (Hunter 1998, p. 2). However, most of the literature cited as feasible biological evidence supportive of this statement refers to relatively long duration, active, strenuous and/or exhaustive exercise with whole limb movements in animal models (e.g. treadmill running or swimming). This is not transferable to a set of frictions performed in humans, which in the main are passive techniques applied for comparatively short durations. In my opinion, there is insufﬁcient support to advocate the use of these techniques in the treatment of LE. In contrast, there is an evolving body of literature attesting to the initial and longer term ameliorative effects of therapeutic exercise, manual therapy and strapping tape in LE in humans (e.g. Pienimaki et al. 1996, 1998; Abbott et al. 2001; Svernlov and Adolfsson 2001; Vicenzino et al. 2001, 2003; Kochar and Dogra 2002; Paungmali et al. 2003a–c). LE is characterized by dysfunction in the muscle and pain systems (Vicenzino et al. 2002). In the Masterclass I developed a case that, given there was no evidence at the highest level (e.g. systematic review, meta-analysis) for manipulative physiotherapy in the treatment of LE, musculoskeletal physiotherapy should be based on the

References Abbott JH, Patla CE, Jensen RH 2001 The initial effects of an elbow mobilization with movement technique on grip strength in subjects with lateral epicondylalgia. Manual Therapy 6: 163–169 Cyriax J 1993 Cyriax’s illustrated manual of orthopaedic medicine, 2nd edn. Butterworth/Heinemann, Oxford Hunter G 1998 Speciﬁc soft tissue mobilization in the management of soft tissue dysfunction. Manual Therapy 3: 2–11 Kochar M, Dogra A 2002 Effectiveness of a speciﬁc physiotherapy regimen on patients with tennis elbow. Physiotherapy 88: 333–341 Paungmali A, O’Leary S, Souvlis T, Vicenzino B 2003a Hypoalgesia and sympathoexcitatory effects of mobilization with movement for lateral epicondylalgia. Physical Therapy 83: 374–383 Paungmali A, O’Leary S, Souvlis T, Vicenzino B 2003b Naloxone fails to antagonise initial hypoalgesic effect of a manual therapy treatment for lateral epicondylalgia. Journal of Manipulative & Physiological Therapeutics, in press Paungmali A, Vicenzino B, Smith 2003c Hypoalgesia induced by elbow manipulation in chronic lateral epicondylalgia does not exhibit tolerane. Journal of Pain, in press Pienimaki T, Karinen P, Kemila T, Koivukangas P, Vanharanta H 1998 Long-term follow-up of conservatively treated chronic tennis elbow patients. A prospective and retrospective analysis. Scandinavian Journal of Rehabilitation Medicine 30: 159–166 Pienimaki T, Tarvainen T, Siira P, Vanharanta H 1996 Progressive strengthening and stretching exercises and ultrasound for chronic lateral epicondylitis. Physiotherapy 82: 522–530 Shrier I 2000 Stretching before exercise: An evidence based approach. British Journal of Sports Medicine 34: 324–325 Smidt N, Assendelft WJJ, Arola H, Malmivaara A, Green S, Buchbinder R, van der Windt D, Bouter LM 2003 Effectiveness of physiotherapy for lateral epicondylitis: A systematic review. Annals of Medicine 35: 51–62 Svernlov B, Adolfsson L 2001 Non-operative treatment regime including eccentric training for lateral humeral epicondylalgia. Scandinavian Journal of Medicine & Science in Sports 11: 328–334 264

ARTICLE IN PRESS Letter to the editor 265

Vicenzino B, Brooksbank J, Minto J, Offord S, Paungmali A 2003 A preliminary investigation of the initial effects of elbow taping on pain free grip strength and pressure pain threshold. Journal of Orthopaedic & Sports Physical Therapy 33: 400–407 Vicenzino B, Paungmali A, Buratowski S, Wright A 2001 Speciﬁc manipulative therapy treatment for chronic lateral epicondylalgia produces uniquely characteristic hypoalgesia. Manual Therapy 6: 205–212 Vicenzino B, Souvlis T, Wright A 2002 Musculoskeletal pain. In: Strong J, Unruh A, Wright A, Baxter D (eds) Pain

Manual Therapy (2003) 8(4), 264–265

Textbook for Therapists. Churchill Livingstone, Edinburgh, pp 327–349

Bill Vicenzino Musculoskeletal Pain & Injury Research Unit, Department of Physiotherapy, The University of Queensland, Brisbane, Qld 4072, Australia

r 2003 Elsevier Ltd. All rights reserved.

Manual Therapy (2003) 8(4), 266

Diary of events

Website: Our website will be continually updated as further information becomes available – please keep watching: www.uct.ac.za/depgc/pgc/ Enquiries: If you would like further information, please send an expression of interest to: Sally Elliott, Conference Management Centre, UCT Medical School, Anzio Road, Observatory 7925, Cape Town, South Africa Tel: +27 21 406-6381; Fax: +27 21 448-6263; E-mail: [email protected]

21–22 November 2003, London, UK 10th Anniversary Symposium on Complementary Health Care at the Royal College of Physicians in London. It is an international symposium with a focus on original research and comprises two days of platform and poster presentations as well as pre-conference workshops and satellite meetings. Contact: Barbara Wider, Complementary Medicine, Peninsular Medical School, Universities of Exeter & Plymouth, 25 Victoria Park Road, Exeter EX2 4NT, UK. E-mail: [email protected] Website: www.exeter.ac.uk/FACT/sympo

21–26 August 2005, Sydney, Australia 11th World Congress on Pain, Workshop and Plenary Proposals. Please send proposals to the Chair of the Scientiﬁc Program Committee: Herta Flor, PhD, Central Institute of Mental Health, Dept of Clinical and Cognitive Neuroscience, PF 12 21 20, 68072 Mannheim, Germany. Tel: 49-621-170-3922; Fax: 49-621-170-3932; E-mail: ﬂ[email protected] Workshop and plenary suggestions should be submitted by 15 March 2003 at the latest so that they can be considered by the Scientiﬁc Program Committee. Note that announcements, deadlines, and other information relating to the 2005 Congress will be routinely updated on the IASP Web page: www.iasp-pain.org

10–13 November, 2004, Melbourne, Australia 5th Interdisciplinary World Congress on Low Back & Pelvic Pain Eﬀective Diagnosis and Treatment of Lumbopelvic Pain. Information and Call for Papers: www.worldcongresslbp.com http://www.worldcongresslbp.com Deadline submitting papers: December 15, 2003 For all further information: [email protected] mailto: [email protected] 27–30 November 2003, Sydney, Australia 13th Biennial Conference of Musculoskeletal Physiotherapy Australia—Best Evidence Better Care. Includes Multidisciplinary Whiplash Symposium. Sydney Australia, Contact: Linda Phillips 2003 Conference Secretariat, GPO Box 2609 Sydney, Australia. Tel.: 61-02-9241-1478; Fax: 61-02-9251-3552; E-mail: [email protected]; Website: www.mpa2003.com

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]

21–26 March 2004 8th IFOMT Congress (International Federation of Manipulative Therapy), International Convention Centre, Cape Town, South Africa. This promises to be one of the most exciting events in the history of physiotherapy in South Africa and you are encouraged to start your planning now. The theme for the conference is ‘‘Balancing the Outcome of Manual Therapy.’’ The programme will range from research based to clinical outcomes papers. Several speakers of international standing have indicated their willingness to participate. The programme will include sessions on Pain, Outcome Based Research, Community and Industrial Considerations, Musculo-Skeletal Spinal and Peripheral Dysfunctions.

Evidence-based manual therapy congress Further information: www.medicongress.com Intensive courses in Manual Therapy Further information: http://allserv.rug.ac.be/bvthillo If you wish to advertise a course/conference, please contact: Karen Beeton, Department of Physiotherapy, University of Hertfordshire, College Lane, Hatﬁeld, Herts AL10 9AB, UK. There is no charge for this service.

266

PII: S1356-689X(03)00105-X

Volume Contents for Vol. 8, 2003 Vol. 8, No. 1

Editorial Editorials and Scholarly Debate A. Moore, G. Jull

1

Masterclass Manipulation of the cervical spine W. A. Hing, D. A. Reid, M. Monaghan

2

Review Evidence for exercise therapy in mechanical neck disorders H. Sarig-Bahat Original articles Neuromuscular control of walking with chronic low-back pain L. Vogt, K. Pfeifer, W. Banzer Psychometric properties of the Tampa Scale for kinesiophobia and the fear-avoidance beliefs questionnaire in acute low back pain E. J. C. M. Swinkels-Meewisse, R. A. H. M. Swinkels, A. L. M. Verbeek, J. W. S. Vlaeyen, R. A. B. Oostendorp Does tape facilitate or inhibit the lower ﬁbres of trapezius? C. M. Alexander, S. Stynes, A. Thomas, J. Lewis, P. J. Harrison

10

21

29 37

Clinical note The eﬀect of postoperative ataralgesia by manual therapy after pulmonary resection F. Hirayama, Y. Kageyama, N. Urabe, H. Senjyu

42

Professional issue Spinal manipulation for low-back pain: a treatment package agreed by the UK chiropractic, osteopathy and physiotherapy professional associations E. Harvey, A. K. Burton, J. K. Moﬀett, A. Breen, in collaboration with the UK BEAM trial team

46

Letters to the editors Control subjects in whiplash studies O. Kwan Response to Drs Kwan and Friel M. Stirling, G. Jull

52

Book reviews

56

Diary of events

58

Subject Index for Volume 7 (2002)

59

54

Vol. 8, No. 2

Editorial Bridging the gap across ﬁelds of practice N. L. Choy

63

Masterclass Lateral epicondylalgia: a musculoskeletal physiotherapy perspective B. Vicenzino

66

267

268

Systematic review A systematic review of physiotherapy for spondylolysis and spondylolisthesis M. L. McNeely, G. Torrance, D. J. Magee

80

Technical and measurement note Relationship between cranio-cervical ﬂexion range of motion and pressure change during the cranio-cervical ﬂexion test D. L. Falla, C. D. Campbell, A. E. Fagan, D. C. Thompson, G. A. Jull

92

Original articles Quantitative analysis of traction in the glenohumeral joint. In vivo radiographic measurements A. Gokeler, G. H. van Paridon-Edauw, S. DeClercq, O. Matthijs, P. U. Dijkstra

97

The eﬀect of cervical rotation on blood ﬂow in the contralateral vertebral artery C. Zaina, R. Grant, C. Johnson, B. Dansie, J. Taylor, P. Spyropolous

103

Adherence to rehabilitation in patients with low back pain G. S. Kolt, J. F. McEvoy

110

Professional issue Validating clinical reasoning: a question of perspective, but whose perspective? A. M. Downing, D. G. Hunter

117

Case report False-negative extension/rotation pre-manipulative screening test on a patient with an atretic and hypoplastic vertebral artery M. D. Westaway, P. Stratford, B. Symons

120

Diary of events

128

Vol. 8, No. 3

Editorial Masterclasses Reconsidered and Revisited A. Moore, G. Jull

129

Masterclass A pain neuromatrix approach to patients with chronic pain G. L. Moseley

130

Systematic review The eﬃcacy of stretching for prevention of exercise-related injury: a systematic review of the literature S. M. Weldon, R. H. Hill

141

Original articles Psychometric properties of a generic health measure in Chinese patients with low back pain in Hong Kong A. S. L. Leung, T.-H. Lam, A. J. Hedley, L. T. Twomey

151

Lumbar multiﬁdus muscle size does not diﬀer whether ultrasound imaging is performed in prone or side lying Y. Coldron, M. Stokes, K. Cook

161

Side-to-side weight-bearing asymmetry in subjects with low back pain J. D. Childs, S. R. Piva, R. E. Erhard, G. Hicks

166

Anatomical and possible clinical relationships between the calcaneoﬁbular ligament and peroneus brevis a pilot study A. Dowling, B. Downey, R. Green, P. Reddy, J. Wickham

170

Technical measurement report Measuring range of active cervical rotation in a position of full head ﬂexion using the 3D Fastrak measurement system: an intra-tester reliability study M. Amiri, G. Jull, J. Bullock-Saxton

176

269

Case report Post-herpetic neuralgia: Possible mechanisms for pain relief with manual therapy M. I. Rabey

180

Abstracts—Manipulation Association of Chartered Physiotherapists (UK) Research Awards 2002

185

Book reviews

187

Letters to the editor

189

Diary of events

192

Vol. 8, No. 4

Editorial Clinical Guideline Developments A. Moore 193 Masterclass Dynamic evaluation and early management of altered motor control around the shoulder complex M. E. Magarey, M. A. Jones 195 Original articles Static innominate asymmetry and leg length discrepancy in asymptomatic collegiate athletes C. J. Krawiec, C. R. Denegar, J. Hertel, G. F. Salvaterra, W. E. Buckley 207 Health care provider’s attitudes and beliefs towards chronic low back pain: the development of a questionnaire R. W. J. G. Ostelo, S. G. M. Stomp-van den Berg, J. W. S. Vlaeyen, P. M. J. C. Wolters, H. C. W. de Vet 214 Reproducibility and repeatability: errors of three groups of physiotherapists in locating spinal levels by palpation E. V. Billis, N. E. Foster, C. C. Wright 223 Stabilizing training compared with manual treatment in sub-acute and chronic low-back pain E. Rasmussen-Barr, L. Nilsson-Wikmar, I. Arvidsson 233 Technical and measurement report Can the Internet be used as a medium to evaluate knee angle? T. G. Russell, G. A. Jull, R. Wootton 242 Clinical note Speciﬁc evaluation of the function of force couples relevant for stabilization of the glenohumeral joint M. E. Magarey, M. A. Jones 247 Professional issue Justifying the on-going physiotherapy management of long-term patients T. Flanagan, P. Coburn, P. Harcourt, M. Zylinski, G Jull 254 Case report Treatment of osteitis pubis via the pelvic muscles A. McCarthy, B. Vicenzino 257 Book reviews 261 Letters to the editor 263 Diary of events 266 Volume Contents and Author Index for Volume 8 (2003) 267 Subject Index and list of referees for Volume 8 (2003) will appear in Vol. 9, No. 1

Author index A ALEXANDER, C. M., 37 AMIRI, M., 176 ARVIDSSON, I., 233 B BANZER, W., 21 BARDIN, L., 189 BILLIS, E. V., 223 BREEN, A., 46 BUCKLEY, W. E., 207 BULLOCK-SAXTON, J., 176 BURTON, A. K., 46

HARRISON, P. J., 37 HARVEY, E., 46 HEDLEY, A. J., 151 HERTEL, J., 207 HICKS, G., 166 HILL, R. H., 141 HING, W. A., 2 HIRAYAMA, F., 42 HUNTER, D. G., 117 J JOHNSON, C., 103 JONES, M. A., 195, 247 JULL, G., 1, 54, 129, 176, 193, 254 JULL, G. A., 92, 242

C CAMPBELL, C. D., 92 CHILDS, J. D., 166 COBURN, P., 254 COLDRON, Y., 161 COOK, J., 190 COOK, K., 161

K KAGEYAMA, Y., 42 KOLT, G. S., 110 KRAWIEC, C. J., 207 KWAN, O., 52

R RABEY, M. I., 180 RASMUSSEN-BARR, E., 233 REDDY, P., 170 REID, D. A., 2 RUSSELL, T. G., 242 S SALVATERRA, G. F., 207 SARIG-BAHAT, H., 10 SENJYU, H., 42 SPYROPOLOUS, P., 103 STIRLING, M., 54 STOKES, M., 161 STOMP-VAN DEN BERG, S. G. M., 214 STRATFORD, P., 120 STYNES, S., 37 SWINKELS, R. A. H. M., 29 SWINKELS-MEEWISSE, E. J. C. M., 29 SYMONS, B., 120 T

L D DALY, M., 263 DANSIE, B., 103 DECLERCQ, S., 97 DENEGAR, C. R., 207 DE VET, H. C. W., 214 DIJKSTRA P. U., 97 DOWLING, A., 170 DOWNEY, B., 170 DOWNING, A. M., 117 E ERHARD, R. E., 166 F FAGAN, A. E., 92 FALLA, D. L., 92 FLANAGAN, T., 254 FOSTER, N. E., 223 G GOKELER, A., 97 GRANT, R., 103 GREEN, R., 170

LAM, T.-H., 151 LEUNG, A. S. L., 151 LEWIS, J., 37 LOW CHOY, N., 63 M

U

MAGAREY, M. E., 195, 247 MAGEE, D. J., 80 MATTHIJS, O., 97 McCARTHY, A., 257 McEVOY, J. F., 110 McNEELY, M. L., 80 MOFFETT, J. K., 46 MONAGHAN, M., 2 MOORE, A., 1, 129, 129 MOSELEY, G. L., 130

HARCOURT, P., 254

URABE, N., 42 V VAN PARIDON-EDAUW, G. H., 97 VERBEEK, A. L. M., 29 VICENZINO, B., 66, 257, 264 VLAEYEN, J. W. S., 29, 214 VOGT, L., 21

W

N NILSSON-WIKMAR, L., 233 O OOSTENDORP, R. A. B., 29 OSTELO, R. W. J. G., 214

WELDON, S. M., 141 WESTAWAY, M. D., 120 WICKHAM, J., 170 WOLTERS, P. M. J. C., 214 WOOTTON, R., 242 WRIGHT, C. C., 223 Z

P H

TAYLOR, J., 103 THOMAS, A., 37 THOMPSON, D. C., 92 TORRANCE, G., 80 TWOMEY, L. T., 151

PFEIFER, K., 21 PIVA, S. R., 166

ZAINA, C., 103 ZYLINSKI, M., 254 270