Manual Therapy Journal - Volume 13, Issue 1, Pages 1-90 (February 2008)

VOLUME 13 NUMBER 1 PAGES 1–90 February 2008 Editors International Advisory Board Ann Moore PhD, GradDipPhys, FCSP, Ce...

Author: Editors: Ann Moore and Gwen Jull

226 downloads 962 Views 6MB Size Report

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

Report copyright / DMCA form

DOWNLOAD PDF

VOLUME 13 NUMBER 1 PAGES 1–90 February 2008

Editors

International Advisory Board

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

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

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

Visit the journal website at http://www.intl.elsevierhealth.com/journals/math doi:10.1016/S1356-689X(07)00134-8

ARTICLE IN PRESS

Manual Therapy 13 (2008) 1 www.elsevier.com/locate/math

Editorial

Less is More Manual Therapy has developed into a leading journal in the ﬁeld of musculoskeletal therapy internationally which is reﬂected in its readership, its contributors and the construct of its Editorial and International Advisory Boards which is representative of 18 countries around the world. The international perspective of Manual Therapy has recently been enhanced by the welcoming of Dr Tim Flynn a leading orthopaedic manual therapy researcher and clinician from the United States of America onto the Editorial Committee of Manual Therapy. In this editorial Dr Flynn responds to an invitation to introduce himself to the Manual Therapy community and highlight from his perspective a professional issue in manual therapy in the USA. Editors Ann Moore Gwendolen Jull I was asked to write this editorial to introduce myself as the newest member of the Editorial Committee of Manual Therapy. It is an honor to collaborate with the members of the editorial team of this prestigious journal and I look forward to assisting the journal in bringing musculoskeletal research and best practice knowledge to musculoskeletal practitioners. Writing a personal introduction would seem like a straight forward process. I am Tim Flynn, PT. I am a physical therapist. I am board certiﬁed in Orthopaedic Physical Therapy and fellowship trained in Orthopaedic Manual Physical Therapy. I am the President of the American Academy of Orthopaedic Manual Physical Therapists and the faculty of Regis University in Denver, CO, USA. I am actively involved in a wide range of musculoskeletal research topics. I maintain a private practice in Fort Collins, Colorado, where I manage primarily spinal disorders including many patients with ‘‘failed back syndrome.’’ I tend to think of that term as a nice way to describe what

1356-689X/$ - see front matter r 2008 Published by Elsevier Ltd. doi:10.1016/j.math.2008.01.001

the statistics in the US continues to tell us—we have unacceptably high rates of spinal surgery without optimal or even marginal outcomes achieved. An introduction would seem simple enough if it was not for the fact that physical therapists in the area of the world that I practice (US) have a professional identity crisis. Physical therapist education in the US (like the rest of the world) has undergone a rapid transition in the requirements for initial professional level-training as well as increasing standardization of post-professional residency and fellowship training particularly in the areas of musculoskeletal care and manual physical therapy. However, the average patient/client in our country has little idea of what a physical therapist does nor are they aware of the large volume of evidence that now supports manual physical therapy interventions for a wide range of musculoskeletal disorders. We, as physical therapists, are largely responsible for this identity crisis. It is not unusual to see a wide array of alphabet soup behind our names (for example—Cert. MDT, CHT, CMP, COMT, CSCS, DPT, FACSM, FAAOMPT, FAPTA, FCAMT, FFCFMT, MTC, MPT, OCS, OMT, SCS). This dilutes who we are. Most of our professional colleagues are recognized by their professional title designator (MD, DO) and their specialized training is described to the consumer (i.e. board-certiﬁed Orthopaedic Surgeon; fellowship trained in Foot & Ankle). The patient/client in our culture understands this; they do not understand all of the initials that we put with our names. We are not deﬁned by our techniques or our numerous certiﬁcations, rather we are deﬁned by our unique body of knowledge and our perspective on managing the musculoskeletal system. I believe clarity in our identity will come with less initials and more description. I look forward to bringing this perspective to manual therapy readership. Timothy W. Flynn, PT, PhD

ARTICLE IN PRESS

Manual Therapy 13 (2008) 2–11 www.elsevier.com/locate/math

Masterclass

Sensorimotor disturbances in neck disorders affecting postural stability, head and eye movement control Julia Treleaven Neck Pain and Whiplash Research Unit, Division of Physiotherapy, University of Queensland, Brisbane, Qld 4072, Australia Accepted 8 June 2007

Abstract The receptors in the cervical spine have important connections to the vestibular and visual apparatus as well as several areas of the central nervous system. Dysfunction of the cervical receptors in neck disorders can alter afferent input subsequently changing the integration, timing and tuning of sensorimotor control. Measurable changes in cervical joint position sense, eye movement control and postural stability and reports of dizziness and unsteadiness by patients with neck disorders can be related to such alterations to sensorimotor control. It is advocated that assessment and management of abnormal cervical somatosensory input and sensorimotor control in neck pain patients is as important as considering lower limb proprioceptive retraining following an ankle or knee injury. Afferent information from the cervical receptors can be altered via a number of mechanisms such as trauma, functional impairment of the receptors, changes in muscle spindle sensitivity and the vast effects of pain at many levels of the nervous system. Recommendations for clinical assessment and management of such sensorimotor control disturbances in neck disorders are presented based on the evidence available to date. r 2007 Elsevier Ltd. All rights reserved. Keywords: Sensorimotor; Eye; Head; Postural stability; Cervical; Management

1. Introduction Sensorimotor control of stable upright posture and head and eye movement relies on afferent information from the vestibular, visual and proprioceptive systems, which converge in several areas throughout the central nervous system. The cervical spine has an important role in providing the proprioceptive input and this is reﬂected in the abundance of cervical mechanoreceptors and their central and reﬂex connections to the vestibular, visual and central nervous systems. Muscle spindles in the cervical region are found in high densities especially in the suboccipital region where there are up to 200 muscle spindles per gram of muscle. This number is considerable when compared to the ﬁrst lumbrical in the thumb where there are 16 muscle Tel.: +617 3365 2275; fax: +617 3365 2775.

E-mail address: [email protected] 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.06.003

spindles per gram (Kulkarni et al., 2001; Boyd Clark et al., 2002; Liu et al., 2003). The cervical muscles, especially the suboccipital muscles, relay information to and receive information from the central nervous system and there are speciﬁc connections between the cervical receptors and the visual and vestibular apparatus and the sympathetic nervous system (Selbie et al., 1993; Bolton et al., 1998; Corneil et al., 2002; Hellstrom et al., 2005). The cervical afferents are also involved in three reﬂexes inﬂuencing head, eye and postural stability: the cervico-collic reﬂex (CCR), the cervico-ocular reﬂex (COR) and the tonic neck reﬂex (TNR). These reﬂexes work in conjunction with other reﬂexes, which are inﬂuenced by vestibular and visual input for coordinated stability of the head, eyes and posture. The CCR activates neck muscles in response to stretch to assist in the maintenance of head position (Peterson, 2004). The COR works with the vestibuloocular reﬂex and optokinetic reﬂex, acting on the extraocular muscles, to assist

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 2–11

3

Cerebral cortex Thalamus

Superior colliculous Medial and lateral vestibular nucleii Central cervical nucleus Dorsal column nucleii

Hypothalamus Reticular formation

Spinothalamic Spinocerebellar Propriospinal Tracts

Cerebellum

Ventral Dorsal horn

COR

Cervical

SNS

CCR

Ocular

VCR

OKR

Vestibular

VOR

VSR TNR

Lower limb Fig. 1. Central and reﬂex connections associated with the cervical spine afferents.

clear vision with movement (Mergner et al., 1998). The TNR is integrated with the vestibulospinal reﬂex to achieve postural stability (Yamagata et al., 1991) (Fig. 1). The importance of the cervical central and reﬂex connections can be realized from response to artiﬁcial disturbances to the cervical afferents in asymptomatic individuals. Sectioning of the cervical nerves or anaesthetic injections into the neck causes nystagmus, disequilibrium and severe ataxia (DeJong and DeJong, 1977; Ishikawa et al., 1998). Vibration of neck muscles, which is thought to stimulate muscle spindle afferents, induces several disturbances in asymptomatic individuals including changes to eye and head position, alterations to body sway and the velocity and direction of gait and running (Lennerstrand et al., 1996; Bove et al., 2002; Courtine et al., 2003). Similar effects have been demonstrated by either simple isometric neck muscle contractions or induced neck muscle fatigue (Gosselin et al., 2004; Schmid and Schieppati, 2005; Vuillerme et al., 2005). Such disturbances are thought to result from a mismatch between abnormal information from the cervical spine and normal information from the vestibular and visual systems.

2. Disturbances in sensorimotor control in neck disorders Considering the experimental evidence, it is not surprising that disturbances in cervical joint position sense (JPS) (Revel et al., 1991; Heikkila and Astrom,

1996; Treleaven et al., 2003), postural stability (Karlberg et al., 1996; Sjostrom et al., 2003; Treleaven et al., 2005a, b; Field et al., 2007) and oculomotor control, such as altered smooth pursuit and saccadic eye movement (Tjell et al., 2003; Treleaven et al., 2005a, b; Storaci et al., 2006), can present in patients with neck disorders of either an insidious or traumatic nature as a result of cervical somatosensory dysfunction. An increased gain of the cervico-ocular reﬂex has also been demonstrated in patients with whiplash (Montfoort et al., 2006). Altered smooth pursuit neck torsion control occurs in neck pain subjects (both idiopathic and traumatic) but not in those with vestibular disorders and central nervous system dysfunction. Smooth pursuit neck torsion control is a difference in eye movement control when measured with the neck in torsioned (i.e., trunk turned 451 but the head remains neutral compared to a neutral head, neck and trunk position) and supports the premise of neck afferent dysfunction as the cause of the smooth pursuit eye movement disturbances (Tjell and Rosenhall, 1998; Tjell et al., 2003; Treleaven et al., 2005a, b). Complaints of dizziness and or unsteadiness can also, but not necessarily, occur in patients with chronic cervical headache and persistent whiplash-associated disorders (Treleaven et al., 2003; Jull et al., 2007). Some neck pain patients also report visual complaints, loss of balance and actual falls (Hulse and Holzl, 2000; Treleaven et al., 2003). Cervical vertigo is considered a principle cause when associated with a whiplash injury (Wenngren et al., 2002; Treleaven et al., 2006) but other

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 2–11

4

(Le Pera et al., 2001; Thunberg et al., 2001; Flor, 2003). Psychosocial stresses might also alter muscle spindle activity via activation of the sympathetic nervous system (Passatore and Roatta, 2006). It is likely that several processes combine to cause an immediate and sustained alteration in somatosensory function originating from the cervical spine, which inﬂuence the tuning and integration of input within the sensorimotor control system. Secondary impairment of vestibular system functioning might also occur (Fischer et al., 1997) and this has implications for management of such disturbances in neck disorders. The majority of the research into sensorimotor control disturbances has been undertaken in patients with persistent neck pain, however there is evidence that deﬁcits could occur soon after the onset of pain (Sterling et al., 2003) and may have an inﬂuence on prognosis (Hildingsson et al., 1993). Thus, routine assessment of head and eye movement control and postural stability in neck disorders is recommended. Evidence to date would suggest that management of disturbed sensorimotor control due to cervical somatosensory dysfunction might need to address the primary causes of the altered somatosensory activity as well as secondary effects. Speciﬁc treatments to the neck such as acupuncture, manual therapy and cranio-cervical ﬂexion training have improved cervical joint position error, vertigo and/or standing balance in patients with neck pain (Fattori et al., 1996; Heikkila et al., 2000; Reid and Rivett, 2005; Palmgren et al., 2006; Jull et al., 2007). Alternatively, programs that emphasize gaze stability, eye/head co-ordination and cervical position sense without local cervical spine treatment have resulted in decreased medication intake, improved neck pain and disability and cervical joint position sense (Revel et al., 1994; Humphreys and Irgens, 2002; Jull et al., 2007).

causes of dizziness and unsteadiness should be considered, such as damage to the vertebral artery, vestibular receptors or central nervous system, elevated anxiety or medication intake (Baloh and Halmagyi, 1996; Ernst et al., 2005; Sturzenegger et al., 1994). Greater deﬁcits in tests of head and eye movement control and postural stability have been measured in patients with neck disorders of traumatic origin, in association with the complaint of dizziness (Tjell et al., 2003; Treleaven et al., 2003, 2005a, b), although these deﬁcits can present in the absence of dizziness as well as in patients with idiopathic neck pain (Kristjansson et al., 2003; Tjell et al., 2003; Field et al., 2007). Although the causes of the disturbances are similar it has been shown that an individual patient may present with dysfunction in either one or several systems. For example, an individual patient may have moderate disturbances to eye movement control but not necessarily have the same degree or any disturbance in cervical JPS or balance (Treleaven et al., 2006). It would appear that either decreased or increased cervical somatosensory activity can result in disturbances of sensorimotor function (Hinoki and Niki, 1975; DeJong and DeJong, 1977). This occurs via a number of mechanisms (Fig. 2). Cervical mechanoreceptor function could be altered as a result of direct trauma to mechanoreceptors (Loescher et al., 1993; Chen et al., 2006), functional impairment of muscles such as increased fatigability (Falla, 2004) or degenerative changes in the muscles such as ﬁbre transformation, fatty inﬁltration and muscle inhibition or atrophy (Uhlig et al., 1995; McPartland et al., 1997; Kristjansson et al., 2004; Elliott et al., 2006). In addition, the effects of pain at many levels of the nervous system can change muscle spindle sensitivity and alter the cortical representation and modulation of cervical afferent input

CNS Represenatation cortex Somatic reorganisation

Visual system

Altered reflexes

Altered cervical afferent input

Altered reflexes

Vestibular system

Altered mechano receptor

Altered muscle spindle activity Trauma Ischaemia Inflammation Altered neuromuscular control Morphological changes

Pain Stress-SNS activation

Fig. 2. Mechanisms of disturbances of cervical somatosensory input in neck disorders.

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 2–11

5

Improvements in balance and symptoms of dizziness have also been observed following a vestibular or oculomotor rehabilitation program in patients with persistent whiplash (Hansson et al., 2006; Storaci et al., 2006). Currently local treatment to the cervical spine in conjunction with tailored programs for sensorimotor control is recommended for patients with neck disorders. The tailored sensorimotor program is similar to that used in vestibular rehabilitation (Herdman, 2000). This combined approach will address the local causes of abnormal cervical afferent input and consider the important links between the cervical, vestibular and ocular systems and any secondary adaptive changes in the sensorimotor control system.

3. Clinical assessment of sensorimotor control disturbances in neck disorders Based on the evidence to date, the assessment of sensorimotor control in the neck pain patient should include: investigation of the symptom of dizziness and unsteadiness and measurement of cervical joint position error, postural stability and oculomotor control. The clinician though should be aware of other possible causes of the disturbances and interview the patient and choose tests accordingly to determine the most likely cause. With respect to dizziness, the description, temporal pattern, associated symptoms and history of the complaint of dizziness can be useful in assisting differential diagnosis of cervicogenic dizziness. Other symptoms such as visual complaints (blurred vision, words jumping and unclear contours of objects) loss of balance, actual falls, difﬁculty walking in the dark, on stairs or negotiating doorways should be noted. The Dizziness Handicap Inventory Short Form Questionnaire (Tesio et al., 1999) can be used to quantify the functional impact of the dizziness. The physical examination may include measures of cervical joint position sense, postural stability and oculomotor control as indicated. In research, precise measurements are taken of these features. However, they can be assessed satisfactorily in the clinical situation albeit with lesser precision. 3.1. Cervical joint position sense (JPS) A simple measure for cervical JPS is the use of a small laser pointer or torch mounted onto a lightweight headband as used by Revel et al. (1991) (Fig. 3). The patient is seated 90 cm from a wall and the starting point projected by the laser is marked. The patient (blindfolded or eyes closed) performs an active neck movement and then returns as accurately as possible to the starting position. The ﬁnal laser position is measured against the starting position in centimeters. This method

Fig. 3. Test of joint position sense using a laser pointer. A measure of joint relocation error is gained from measuring the distance between the return and starting positions.

provides a quantitative assessment tool as errors as little as 3–41 (4–5 cm) can indicate a deﬁcit in cervical JPS (Revel et al., 1991; Treleaven et al., 2003). Errors are measured following active return from cervical extension, ﬂexion and rotation. Relocating to selected points in range (Loudon et al., 1997) and accuracy in retracing patterns (Kristjansson et al., 2004) can also be used to assess cervical kinaesthesia. Jerky movements, searching or overshooting the initial position, reproduction of dizziness and/or a noticeable difference of cervical movement patterns in the test with eyes closed may also indicate impaired cervical kinaesthetic sense. 3.2. Oculomotor Oculomotor assessment incorporates assessment of gaze stability (the ability to maintain gaze while moving the head), smooth pursuit (eye follow while keeping the head still), saccadic eye movement (rapid eye movements to change a point of ﬁxation) and eye/head coordination (maintaining gaze when both the eyes and head are moving). The tests are usually performed with the patient in a sitting position but can be tested initially in supine lying if necessary. The starting position will depend on the severity of the patient’s complaints and overall physical ﬁndings. 3.2.1. Gaze stability Gaze stability is assessed by asking the patient to keep the eyes focused on a target while actively moving the head into ﬂexion, extension and rotation (Fig. 4). Inability to maintain focus on the target, reduced or

ARTICLE IN PRESS 6

J. Treleaven / Manual Therapy 13 (2008) 2–11

awkward cervical motion or reproduction of symptoms such as dizziness, blurring of vision or nausea are signs of an abnormal response. 3.2.2. Eye follow The patient keeps the head still while following a moving target with the eyes as closely as possible. The target is moved slowly side to side (201/s through a visual angle of 401). The test is repeated with the head still but with the trunk rotated up to 451 (Fig. 5). The test is repeated on the opposite side. Any difference is noted in smooth eye follow or symptom reproduction in

these neck torsion positions compared to the neutral position. Often patients with neck disorders will be unable to keep up with the target and demonstrate quick catch up eye movements when the neck is in torsion particularly when the target is crossing the midline. Saccadic eye movements at the extremes of the visual angle and with the change of movement direction are not considered abnormal. 3.2.3. Saccadic eye movement The patient follows and ﬁxes their gaze on a target that is quickly moved and then held still momentarily. The target is moved in several different directions. The patient’s ability to quickly move to the target and ﬁxate on it is noted. 3.2.4. Eye/head co-ordination The patient moves the eyes and head in the same direction to focus on a point, leading with the eyes ﬁrst to a target and then the head ensuring the eyes keep focused on the target. This can be performed to the left, right and up and down. 3.3. Postural stability

Fig. 4. Testing gaze stability during head rotation to the right.

A modiﬁed sensory organization test is used to assess postural stability. Visual and proprioceptive input from the lower limbs is altered as the tests progress (Shumway-Cook and Horak, 1986). Balance in comfortable and narrow stance can be assessed with the patient standing on a ﬁrm and then a soft surface such as a piece of 10 cm dense foam. The tests should be performed with both eyes open and closed. Inability to maintain stance for 30 s, noticeably large increases in sway, slower responses to correct sway or rigidity to prevent sway are

Fig. 5. Eye follow keeping the head still is tested (a) with the neck in a neutral position and repeated (b) with the neck torsioned via trunk rotation. The neck torsion is tested to both the left and right. A deterioration in eye follow in the neck torsion positions, when compared to the neutral position suggests a cervical origin of deﬁcits in eye movement control.

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 2–11

considered abnormal responses. It is thought that people with neck disorders rely more on vision and other somatosensory inputs for balance and thus deﬁcits will be greatest when these inputs (e.g., vision) are reduced. For increasing challenge, the patient can be tested in tandem and then single leg stance on a ﬁrm surface with eyes open and closed. It is reasonable to expect that a person under the age of 60 years can maintain stability for up to 30 s in the comfortable and narrow stance tests. Subjects under 45 years should also be able to complete tandem and single leg stance tests. (Treleaven et al., 2005a, b) Assessment of the ability to complete the 30 s of the tandem stance can be a useful screening test (Treleaven et al., 2005a, b; Field et al., 2007). 3.4. Additional tests There is some evidence that the neck may directly inﬂuence vestibular functioning (Fischer et al., 1997). In addition, primary vestibular pathology is also possible following a whiplash injury or in the middle aged/elderly population. Thus, assessment of postural control disturbances may need to address secondary vestibular inﬂuences on postural control. While there is certainly overlap between cervical and vestibular systems in a number of the testing procedures already described, additional tests to investigate vestibular function more closely may be necessary in some neck pain patients, especially those complaining of loss of balance and falls. Factors such as ageing, pre-existing vestibular pathology and medical conditions might increase the degree of disturbances in those with neck disorders (Poole et al., 2007). Additional tests might include tests of dynamic balance and functional ambulation (Herdman, 2000; Alpini et al., 2005), as well as testing of the vestibuloocular reﬂex (Herdman, 2000). In cases where Benign Paroxysmal Positional Vertigo is suspected the Hallpike-Dix manoeuvre should be included (Herdman, 2000). Referral for a more thorough investigation of the vestibular or central nervous system may be warranted in those where cervical causes of the disturbances cannot be substantiated.

4. Management of sensorimotor control disturbances in neck disorders The ﬁndings of the assessment will direct and tailor the most appropriate management of sensorimotor control disturbances in the individual patient with a neck disorder. It is suggested that management include both local treatment to the neck to decrease pain and improve neuromuscular function in combination with tailored sensorimotor exercises to improve any deﬁcits in cervical JPS, oculomotor control and postural stability. This addresses the causes of abnormal cervical

7

mechanoreceptor input as well as the effects resulting from the potential conﬂict arising from abnormal cervical afferent input and normal vestibular and ocular input. Exercises for each system should be performed two to ﬁve times per day. Temporary reproduction of dizziness is acceptable however exacerbation of neck pain or headache is not acceptable. If this occurs the exercises should be modiﬁed by decreasing the number of repetitions or altering the patient position to a more supported position such as supine lying. Progression of each exercise set can be achieved by altering the duration, repetitions and the degree of difﬁculty of the task. Exercises can also be progressed by performing activities such as an eye task or cervical JPS practice while sitting on an unstable surface or while standing with the feet in an unstable base of support for example, heel toe stance, or while walking. Some examples follow. 4.1. Cervical joint position sense The patient practices relocating the head back to the natural head posture and to pre-determined positions in range from the movement directions assessed to be abnormal. The patient may practice ﬁrst with the eyes open and then with eyes closed, lining up their natural head posture and target positions with points on the wall to check their accuracy on return. The exercise can be made more precise for home use with the use of a pencil torch or laser attached to a headband. Higher level skills training could include tracing intricate patterns, such as a ﬁgure of 8 placed on the wall, with the head using the head torch or laser for feedback. 4.2. Oculomotor exercises The exercise level is set on the basis of the oculomotor assessment. The degree of difﬁculty for oculomotor exercises can be increased by increasing the speed of the task, range of motion, changing the patient’s position and the focus point or background (Fig. 6). Table 1 outlines how each of the various variables can be altered to progress the exercises. 4.2.1. Eye follow with a stationary head The patient eyes follow a target moving side to side and up and down keeping the head still. For home use, the patient could practice tracking a tennis ball tossed sideways or up and down whilst keeping the head still. 4.2.2. Saccadic eye movements Saccadic movements of a target are performed at randomised eye positions. Progression can include increasing the speed of the movements, position of the patient and the visual background or focus point (Table 1).

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 2–11

8

Focus

Focus on these words

Focus

Focus on these words

Focus

Focus on these words

Fig. 6. Progression of eye exercises by altering either the focus point or the background of the focus point.

Table 1 Various methods of progression of the oculomotor exercises Variable

Start

Progress

Further progression

Target Background to target Patient position Neck position Speed Vision Range of motion Duration of exercise Frequency of exercise

Single spot Plain Supine—sitting Neutral Slow Unrestricted Small 30 s 2 day

Word Stripes Standing—vary stance Torsion 301 Medium Restricted peripheral Medium 1–2 min 3 day

Business card Checker board Walking Torsion 451 Fast Restricted peripheral Large 5 min 5 day

4.2.3. Gaze stability Training may commence with the clinician performing slow passive neck movements or the patient performing active neck movements, while they ﬁxate their gaze on a point on the ceiling. The clinician can also passively move the trunk while the patient maintains gaze on a target. The patient can be asked to ﬁx their gaze on a point, close their eyes and perform a neck movement and open their eyes after the head movement to check for stable gaze on the target. Any task may be progressed by restricting peripheral vision, using a pair of swimming goggles that have been blackened out except for a small area in the center of each side. 4.2.4. Eye/head co-ordination The exercises commence with rotating the eyes and head to the same side, in both left and right and up

and down directions. As a progression, the eyes are moved ﬁrst, then the head, but the patient continues to maintain focus on two targets which can be positioned horizontally or vertically. Further progression could include the patient performing active neck rotation to follow a slow moving target while their peripheral vision is restricted. The patient can also practice rotating the eyes and head to the opposite side in both directions. For home practice, the patient could hold a target in front of them and move the target and their head in opposite directions. The patient should also practice moving the eyes, head, neck and arm following their thumb or moving the eyes, head, neck and trunk to look as far behind as possible. Active movements of the head to follow a moving target can be performed both with unrestricted and peripheral restricted movement.

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 2–11

9

as a tailored sensorimotor exercise program to improve identiﬁed deﬁcits in postural stability and head and eye movement control. The recommended clinical assessment and management is based on the available evidence to date, however this is an emerging area and more extensive research is needed to reﬁne and identify methods for assessment and determine the optimal strategies for management of such disturbances in patients with neck disorders. There will be a case study published online to support this Masterclass at a later date. References

Fig. 7. Home practice of tandem stance balance exercise.

4.3. Postural stability The starting level for balance retraining will depend on which tests the patient failed or had difﬁculty with in the clinical assessment. This could be comfortable, narrow, tandem or single leg stance. The conditions for training progress from eyes open, eyes closed to different supporting surfaces. Patients practice the exercise at home perhaps in a corner area, such that they are able to regain balance easily if necessary (Fig. 7). They gradually increase the time of maintaining stability to reach 30 s. Challenges to the system can be increased with the addition of relocation practice or oculomotor exercises to the balance exercise. Walking forwards, backwards and sideways while actively moving the head into different directions, maintaining direction and velocity of gait, also challenge cervical afferent input for balance mechanisms. 5. Conclusion Given the importance of the neck for postural stability, head and eye movement control, as well as the nature of the changes in sensorimotor control seen in those with neck disorders, assessment and management of such disturbances should form an important part of the multimodal approach to neck disorders. This should include addressing the causes of the altered cervical somatosensory input such as improving neuromuscular function and decreasing pain and inﬂammation as well

Alpini D, Ciavarro GL, et al. Evaluation of head-to-trunk control in whiplash patients using digital CranioCorpoGraphy during a stepping test. Gait and Posture 2005;22(4):308–16. Baloh R, Halmagyi G. Disorders of the vestibular system. New York: Oxford University Press; 1996. Bolton PS, Kerman IA, et al. Inﬂuences of neck afferents on sympathetic and respiratory nerve activity. Brain Research Bulletin 1998;47(5):413–9. Bove M, Courtine G, et al. Neck muscle vibration and spatial orientation during stepping in place in humans. Journal of Neurophysiology 2002;88(5):2232–41. Boyd Clark L, Briggs C, et al. Muscle spindle distribution, morphology, and density in the longis colli and multiﬁdus muscles of the cervical spine. Spine 2002;27(7):694–701. Chen CY, Lu Y, et al. Distribution of A-delta and C-ﬁber receptors in the cervical facet joint capsule and their response to stretch. Journal of Bone and Joint Surgery. (American Volume) 2006; 88A(8):1807–16. Corneil BD, Olivier E, et al. Neck muscle responses to stimulation of monkey superior colliculus. I. Topography and manipulation of stimulation parameters. Journal of Neurophysiology 2002;88(4): 1980–99. Courtine G, Papaxanthis C, et al. Gait-dependent integration of neck muscle afferent input. Neuroreport 2003;14(18):2365–8. DeJong PI, DeJong JM. Ataxia and nystagmus induced by injection of local anaesthetics in the neck. Annals of Neurology 1977;(1): 240–6. Elliott J, Jull G, et al. Fatty inﬁltration in the cervical extensor muscles in persistent whiplash-associated disorders—a magnetic resonance imaging analysis. Spine 2006;31(22):E847–55. Ernst, et al. Management of Posttraumatic vertigo. Otolaryngology Head and Neck Surgery 2005;132(4):554–8. Falla D. Unravelling the complexity of muscle impairment in chronic neck pain. Manual Therapy 2004;9(3):125–33. Fattori B, Borsari C, et al. Acupuncture treatment for balance disorders following whiplash injury. Acupuncture and ElectroTherapeutics Research 1996;21(3–4):207–17. Field S, Treleaven J, et al. Standing balance: a comparison between idiopathic and whiplash induced neck pain. Manual Therapy 2007, in press, doi:10.1016/jmath.2006.12.005. Fischer A, Verhagen WIM, et al. Whiplash injury. A clinical review with emphasis on neurootological aspects. Clinical Otolaryngology 1997;22(3):192–201. Flor H. Cortical reorganisation and chronic pain: implications for rehabilitation. Journal of Rehabilitation Medicine 2003;35:66–72. Gosselin G, Rassoulian H, et al. Effects of neck extensor muscles fatigue on balance. Clinical Biomechanics 2004;19(5):473–9. Hansson EE, Mansson NO, et al. Dizziness among patients with whiplash-associated disorder: a randomized controlled trial. Journal of Rehabilitation Medicine 2006;38(6):387–90.

ARTICLE IN PRESS 10

J. Treleaven / Manual Therapy 13 (2008) 2–11

Heikkila H, Astrom PG. Cervicocephalic kinesthetic sensibility in patients with whiplash injury. Scandinavian Journal of Rehabilitation Medicine 1996;28(3):133–8. Heikkila H, Johansson M, et al. Effects of acupuncture, cervical manipulation and NSAID therapy on dizziness and impaired head repositioning of suspected cervical origin: a pilot study. Manual Therapy 2000;5(3):151–7. Hellstrom F, Roatta S, et al. Responses of muscle spindles in feline dorsal neck muscles to electrical stimulation of the cervical sympathetic nerve. Experimental Brain Research 2005;165(3): 328–42. Herdman S. Vestibular rehabilitation. Philadelphia: Davis; 2000. Hildingsson C, Wenngren BI, et al. Eye motility dysfunction after softtissue injury of the cervical-spine—a controlled, prospective-study of 38 patients. Acta Orthopaedica Scandinavica 1993;64(2): 129–32. Hinoki M, Niki H. Neurotological studies on the role of the sympathetic nervous system in the formation of traumatic vertigo of cervical origin. Acta Otolaryngology Supplement 1975;330:185– 96. Hulse M, Holzl M. Vestibulospinal reﬂexes in patients with cervical disequilibrium (‘‘the cervical staggering’’). HNO 2000;48(4): 295–301. Humphreys B, Irgens P. The effect of a rehabilitation exercise program on head repositioning accuracy and reported levels of pain in chronic neck pain subjects. Journal of Whiplash and Related Disorders 2002;1(1):99–112. Ishikawa K, Matsuzaki Z, et al. Effect of unilateral section of cervical afferent nerve upon optokinetic response and vestibular nystagmus induced by sinusoidal rotation in guinea pigs. Acta Otolaryngology Supplement 1998;537:6–10. Jull G, Falla D, et al. Retraining cervical joint position sense: the effect of two exercise regimes. Journal of Orthopaedic Research 2007; 25(3):404–12. Karlberg M, Magnusson M, et al. Postural and symptomatic improvement after physiotherapy in patients with dizziness of suspected cervical origin. Archives of Physical Medicine and Rehabilitation 1996;77(9):874–82. Kristjansson E, Dall’Alba P, et al. A study of ﬁve cervicocephalic relocation tests in three different subject groups. Clinical Rehabilitation 2003;17(7):768–74. Kristjansson E, Hardardottir L, et al. A new clinical test for cervicocephalic kinesthetic sensibility: ‘‘the ﬂy’’. Archives of Physical Medicine and Rehabilitation 2004;85(3):490–5. Kulkarni V, Chandy M, et al. Quantitative study of muscle spindles in suboccipital muscles of human foetuses. Neurology India 2001;49: 355–9. Le Pera D, Graven-Nielsen T, et al. Inhibition of motor system excitability at cortical and spinal level by tonic muscle pain. Clinical Neurophysiology 2001;112(9):1633–41. Lennerstrand G, Han Y, et al. Properties of eye movements induced by activation of neck muscle proprioceptors. Graefes Archive for Clinical and Experimental Ophthalmology 1996;234(11):703–9. Liu J, Thornell L, et al. Muscle spindles in the deep muscles of the human neck: a morphological and immunocytochemical study. Journal of Histochemistry and Cytochemistry 2003;51(2): 175–86. Loescher AR, Holland GR, et al. The distribution and morphological characteristics of axons innervating the periodontal ligament of reimplanted teeth in cats. Archives of Oral Biology 1993;38(9): 813–22. Loudon JK, Ruhl M, et al. Ability to reproduce head position after whiplash injury. Spine 1997;22(8):865–8. McPartland JM, Brodeur RR, et al. Chronic neck pain, standing balance, and suboccipital muscle atrophy—a pilot study. Journal of Manipulative Physiological Theraputics 1997;20(1):24–9.

Mergner T, Schweigart G, et al. Eye movements evoked by proprioceptive stimulation along the body axis in humans. Experimental Brain Research 1998;120(4):450–60. Montfoort I, Kelders WPA, et al. Interaction between ocular stabilization reﬂexes in patients with whiplash injury. Investigative Ophthalmology and Visual Science 2006;47(7):2881–4. Palmgren PJ, Sandstrom PJ, et al. Improvement after chiropractic care in cervicocephalic kinesthetic sensibility and subjective pain intensity in patients with nontraumatic chronic neck pain. Journal of Manipulative and Physiological Therapeutics 2006;29(2):100–6. Passatore M, Roatta S. Inﬂuence of sympathetic nervous system on sensorimotor function: whiplash associated disorders (WAD) as a model. European Journal of Applied Physiology 2006;98:423–49. Peterson BW. Current approaches and future directions to understanding control of head movement. Brain Mechanisms for the Integration of Posture and Movement 2004;143:369–81. Poole E, Treleaven J, et al.The inﬂuence of neck pain on balance and gait parameters in community dwelling elders. Manual Therapy 2007 (epub ahead of print). Reid SA, Rivett DA. Manual therapy treatment of cervicogenic dizziness: a systematic review. Manual Therapy 2005;10(1):4–13. Revel M, Andre-Deshays C, et al. Cervicocephalic kinesthetic sensibility in patients with cervical pain. Archives of Physical Medicine and Rehabilitation 1991;72:288–91. Revel M, Minguet M, et al. Changes in cervicocephalic kinesthesia after a proprioceptive rehabilitation program in patients with neck pain: a randomized controlled study. Archives of Physical Medicine and Rehabilitation 1994;75:895–9. Schmid M, Schieppati M. Neck muscle fatigue and spatial orientation during stepping in place in humans. Journal of Applied Physiology 2005;99(1):141–53. Selbie WS, Thomson DB, et al. Suboccipital muscles in the cat neck: morphometry and histochemistry of the rectus capitis muscle complex. Journal of Morphology 1993;216(1):47–63. Shumway-Cook A, Horak F. Assessing the inﬂuence of sensory integration on balance. Physical Therapy 1986;66:1548–50. Sjostrom H, JA, JH, et al. Trunk sway measures of postural stability during clinical balance tests in patients with chronic whiplash injury symptoms. Spine 2003; 28(15): 1725–34 Sterling M, Jull G, et al. Development of motor system dysfunction following whiplash injury. Pain 2003;103(1–2):65–73. Storaci R, Manelli A, et al. Whiplash injury and oculomotor dysfunctions: clinical-posturographic correlations. European Spine Journal 2006;15(12):1811–6. Sturzenegger M, Stefano GD, et al. Presenting symptoms and signs after whiplash injury: the inﬂuence of accident mechanisms. Neurology 1994;44(4):688–93. Tesio L, Alpini D, et al. Short form of the dizziness handicap inventory. American Journal of Physical Medicine and Rehabilitation 1999;78(3):233–41. Thunberg J, Hellstrom F, et al. Inﬂuences on the fusimotor-muscle spindle system from chemosensitive nerve endings in the cervical facet joints in the cat; possible implications for whiplash induced disorders. Pain 2001;91:15–22. Tjell C, Rosenhall U. Smooth pursuit neck torsion test: a speciﬁc test for cervical dizziness. American Journal of Otology 1998;19(1):76–81. Tjell C, Tenenbaum A, et al. Smooth pursuit neck torsion test— a speciﬁc test for whiplash associated disorders? Journal of Whiplash and Associated Disorders 2003;1(2):9–24. Treleaven J, Jull G, et al. Dizziness and unsteadiness following whiplash injury: characteristic features and relationship with cervical joint position error. Journal of Rehabilitation Medicine 2003;35(1):36–43. Treleaven J, Jull G, et al. Smooth pursuit neck torsion test in whiplash associated disorders-relationship to self reports of neck pain and disability, dizziness and anxiety. Journal of Rehabilitation Medicine 2005a;37(4):219–23.

ARTICLE IN PRESS J. Treleaven / Manual Therapy 13 (2008) 2–11 Treleaven J, Jull G, et al. Standing balance in persistent WAD— comparison between subjects with and without dizziness. Journal of Rehabilitation Medicine 2005b;37(4):224–9. Treleaven J, Jull G, et al. The relationship of cervical joint position error to balance and eye movement disturbances in persistent whiplash. Manual Therapy 2006;11(2):99–106. Uhlig Y, Weber BR, et al. Fiber composition and ﬁber transformation in neck muscles of patients with dysfunction of the cervical spine. Journal of Orthopaedic Research 1995;13:240–9.

11

Vuillerme N, Pinsault N, et al. Postural control during quiet standing following cervical muscular fatigue: effects of changes in sensory inputs. Neuroscience Letters 2005;378(3):135–9. Wenngren B, Pettersson K, et al. Eye motility and auditory brainstem response dysfunction after whiplash injury. Acta Orthopaedica Scandinavica 2002;122(3):276–83. Yamagata Y, Yates BJ, et al. Participation of Ia reciprocal inhibitory neurons in the spinal circuitry of the tonic neck reﬂex. Experimental Brain Research 1991;84(2):461–4.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 12–28 www.elsevier.com/locate/math

Review

Can we predict poor recovery from recent-onset nonspeciﬁc low back pain? A systematic review Peter M. Kenta,b,, Jennifer L. Keatingc a

Monash Department of Clinical Epidemiology at Cabrini Hospital, Vic., Australia Department of Epidemiology & Preventive Medicine, Faculty of Medicine, Nursing & Health Sciences, Monash University, Vic., Australia c Department of Physiotherapy, School of Primary Health Care, Faculty of Medicine, Nursing & Health Sciences, Monash University, Vic., Australia b

Received 16 August 2005; received in revised form 12 April 2007; accepted 23 May 2007

Abstract This systematic review of prospective cohort studies investigated the evidence for prognostic factors for poor recovery in recentonset nonspeciﬁc low back pain (NSLBP). Medline, Cinahl, Embase, PsychINFO, and AMED databases were searched and citation tracking was performed. Fifty studies met the inclusion criteria. Bivariate and multivariable prognostic factor/outcome associations were extracted. Two reviewers independently performed data extraction and method quality assessment. Where data were available, odds ratios for bivariate associations were calculated and meta-analysis was performed on comparable prognostic factor/outcome associations. Despite the number of studies that have investigated these prognostic factors, uncertainty remains regarding which factors are associated with particular outcomes, the strength of those associations and the extent of confounding between prognostic factors. This uncertainty is the result of the disparate methods that have been used in these investigations, incomplete and contradictory ﬁndings, and an inverse relationship between study quality and the reported strength of these associations. The clinical implication is that the formation of clinically useful predictive models remains dependent on further high-quality research. The research implications are that subsequent studies can use the ﬁndings of this review to inform prognostic factor selection, and that prognostic studies would ideally be designed to enhance the capacity for ﬁndings to be pooled with those of other studies. r 2007 Elsevier Ltd. All rights reserved. Keywords: Low back pain; Outcome assessment; Prognosis; Recovery of function

1. Introduction Low back pain (LBP) is common, costly, diagnostically challenging, and has a variable clinical course. Rapid improvements occur in the ﬁrst 3 months postonset, but improvements are gradual thereafter (Hestbaek et al., 2003; Pengel, 2003). At 6 months following onset, 16% (range 3–40%) of patients initially off-work remain off-work, and at 12 months post-onset, 62% of all patients (range 42–75%) still have pain. Within 12 months of onset, recurrences of pain (60%, range Corresponding author. Monash Department of Clinical Epidemiology at Cabrini Hospital, 183 Wattletree Road, Malvern, Victoria, 3144 Australia. Tel.: +61 39508 1652; fax: +61 39508 1368. E-mail address: [email protected] (P.M. Kent).

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.05.009

44–73%) and of work absence (33%, range 26–37%) are common (Hestbaek et al., 2003). Patients who do not recover early are at much greater risk of long-term poor recovery and utilize the bulk of compensable health-care resources (Torstensen et al., 1998; Van Tulder et al., 2000). If the capacity existed to identify LBP patients at risk of poor recovery early in the clinical course of the condition, appropriate intervention might inﬂuence outcomes. Frustration experienced by patients and clinicians might be reduced by advanced warning of a potentially modest recovery, with subsequent recalibration of expectations, rehabilitation strategies, and lifestyle adjustments. Furthermore, identiﬁcation of important prognostic factors for LBP patients could facilitate baseline comparability of subjects recruited

ARTICLE IN PRESS P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

into research studies. The Cochrane Back Review Group reported a need for systematic reviews to determine the most important prognostic factors for poor recovery, and referred to this as a ‘Holy Grail’ question (Bouter et al., 2003). Several systematic reviews have examined prognostic factors for poor recovery in back pain (Rohling et al., 1995; Linton, 2000; Kuch, 2001; Teasell, 2001; Teasell and Bombardier, 2001), and for LBP speciﬁcally (Crook et al., 1998; McIntosh et al., 2000b; Truchon, 2001; Pincus et al., 2002). However, many reviews have included only workers and have measured only occupational outcomes. Some prognostic studies have used return-to-work as the sole outcome measure, and as return-to-work is inﬂuenced by many factors, it is unlikely to mirror all the dimensions of the clinical course of LBP. Previous LBP prognostic reviews have also included cohorts with speciﬁc LBP and LBP with neurocompressive signs. Speciﬁc LBP is caused by serious pathology (cancer, spinal osteomyelitis, fracture, spinal stenosis, cauda equina, ankylosing spondylitis, visceral-referred pain). Neurocompressive signs (pins & needles, numbness and weakness) in LBP are usually caused by nerve root irritation. Speciﬁc LBP and LBP with neurocompressive signs represent approximately 20% of LBP in primary care. The remaining 80% cannot be deﬁnitively linked to a speciﬁc cause and is most accurately described as nonspeciﬁc LBP (NSLBP) (Spengler and David, 1985; Deyo et al., 1992). As there is evidence that neurocompressive signs are risk factors for chronicity (Burton et al., 1995; Hunt, 2002), it is possible that NSLBP has both a clinical course (Di Fabio, 1995) and prognostic factors that are different from neurocompressive LBP. Similarly, NSLBP may have a clinical course and prognostic factors that are different from speciﬁc LBP. No review of NSLBP has quantitatively investigated the prognostic utility of factors from the broad range of clinical assessments (physical impairment, pain, activity limitation, psychosocial function, and participation restriction) and included a comprehensive range of outcomes that are relevant to people with this condition. Many prognostic studies have been performed in LBP but predictive models derived in these studies do not appear to be commonly used in clinical practice. Potential barriers to the clinical application of previous prognostic research may include an inadequate strength of predictive models, prognostic associations being inconsistently reproduced across studies and predictive models presented in formats not readily interpretable by clinicians. Authorities encourage prognostic studies to be built on theoretical models and the results of previous studies (Altman and Lyman, 1998). In NSLBP there is an absence of theoretical models that traverse assessment

13

domains. However, as there is no shortage of previous studies, a preliminary step towards optimising further prognostic studies would be to synthesize previous results. The validity of prognostic factors would be enhanced if the association with a particular outcome were reproduced across independent samples. Greater conﬁdence in the utility of a prognostic factor might also be warranted if measures of the strength of its relationship with an outcome (such as odds ratios) could be pooled through meta-analysis. The aim of this systematic review was to investigate the prognostic factors of poor-recovery that have been studied in recent-onset NSLBP (o12 weeks) using the outcomes of pain, activity limitation, and participation restriction. The review aimed to identify the signiﬁcant prognostic factor/outcome associations that have been found, to report the predictive strength of these associations, to investigate the inﬂuence of methodological quality on these associations, and to synthesize these results or report barriers inhibiting this synthesis.

2. Methods A systematic review was conducted using MOOSE systematic review principles (Stroup et al., 2000). 2.1. Searching Searches were performed of Medline (from 1966), Cinahl (from 1982), Embase (from 1966), PsychINFO (from 1967), and AMED (from 1985) databases from inception to February 2007. The electronic search strategy is available from the corresponding author. Searching was also performed of the reference lists of relevant reviews and retrieved papers, and citation tracking of authors of relevant studies. 2.2. Selection 2.2.1. Inclusion criteria For studies to be included they had to prospectively investigate the relationship between prognostic factors at the inception of a recent-onset episode of NSLBP (but not necessarily the ﬁrst episode), and at least one of the outcomes of pain, activity limitation, or participation restriction. Studies needed to report results testing the statistical association between prognostic factors and outcomes or the raw data that enabled such tests to be performed. Only studies reported in English were included due to an absence of translation resources. 2.2.2. Exclusion criteria Studies were excluded if NSLBP could not be isolated from other conditions, or if participants were pregnant, or had speciﬁc diseases such as inﬂammatory arthritis,

ARTICLE IN PRESS 14

P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

tumour, or fracture. They were also excluded if more than 15% of the participants had either experienced this episode for more than 3 months or had neurocompressive symptoms. The threshold of 15% was arbitrarily chosen for pragmatic reasons. A number of otherwise acceptable studies may have been excluded with a lower threshold for episode duration, and potentially useful prognostic factors may have been overlooked. Some studies use the term sciatica without deﬁnition, and in this review sciatica was interpreted as indicating the presence of neurocompressive symptoms. However, as the descriptor ‘sciatica’ has sometimes been used to describe any LBP-associated leg pain, these patients may have had NSLBP. Due to this ambiguity, studies with less that 15% of participants with neurocompressive symptoms were included. This review excluded studies that collected only crosssectional data, descriptions of natural history or clinical course without data on prognostic factors, and descriptions of incidence or prevalence of NSLBP. 2.3. Methodological quality Method quality of included studies was assessed using six criteria, based on those recommended by Hudak et al. (1996): loss to follow-up—at least 80% of subjects assessed at presentation were also assessed at outcome; inception cohort—all patients eligible for admission into the study were admitted into the study (random selection or consecutive cases); prognostic variables— assessment of the prognostic variables was deﬁned in a way that enabled replication; outcome variables—outcome variables were measured using deﬁned methods; blind ﬁnal assessor—assessors measured prognostic variables and outcomes independently of each other, or, if not, it was likely that knowledge of the prognostic variable did not bias measurements of outcome; analysis—data were reported that allowed estimates of odds of poor recovery (odds ratios) and their 95% CI. The percentage agreement obtained in this review by the two authors using these quality assessment items was calculated. 2.4. Data extraction Identiﬁcation of suitable studies, data extraction, data coding, and quality assessment were performed completely independently and in duplicate by the two authors. Differences of opinion were resolved by discussion. The following data were extracted from each prognostic study: cohort details (authors, publication title, journal, year of publication, setting, country, subject numbers at inception, proportion at follow-up), predictor and outcome variables (name, deﬁnition, measurement scales used), how the relationship between predictor and outcome was quantiﬁed, statistical sig-

niﬁcance of the relationship, and data for construction of a contingency table from which odds of poor recovery could be calculated. 2.4.1. Classification of predictors A coding taxonomy was created to allow identiﬁcation of predictors that appeared to be comparable, despite being described using various labels. This taxonomy (on-line Additional File 1) allowed predictor variables across studies to be sensibly clustered, and for an assessment domain (physical impairment, pain, activity limitation, participation restriction, history, clinician factors, or therapeutic response) to be assigned to each variable. A simple taxonomy of self-explanatory descriptive codes was developed and applied without further operational deﬁnition. The percentage agreement obtained in this review by the two authors using the coding taxonomy was calculated. 2.4.2. Classification of outcomes Outcomes were classiﬁed using the same taxonomy into one of three domains: pain, activity limitation, and participation restriction. Within each outcome domain, poor recovery was also deﬁned across different studies in a variety of ways, even when the same outcome measure was used. Many authors arbitrarily dichotomized outcome scales and deﬁned poor recovery as persistence of scores in the ‘worst’ part of the scale. Some authors dichotomized outcome scales at their midpoint, while others used higher or lower points. Some authors dichotomized outcome scales using formulae such as ‘‘less than a 50% change in initial score is considered to represent poor recovery’’. Pain outcomes reported in the included studies were duration, improvement, intensity, satisfaction, and recovery (either complete or incomplete). Activity limitation outcomes were bed rest, domestic activity limitation, Graded Chronic Pain Scale scores, Oswestry score, recovery (either complete or incomplete), recovery rate, Roland-Morris score, SF-36 score, Sickness Impact Proﬁle score and ‘Activity limitation’ (miscellaneous measures of activity limitation not classiﬁed by other items). Participation restriction outcomes were duration of compensation, longer time-off-work, return-to-fullwork duties, return-to-work duties, and time-off work. The outcomes were divided into short-term (o3 months) and long-term. This temporal dichotomisation was performed due to the considerable evidence that most recovery from pain and reduction in activity limitation has occurred by 3 months, and recovery is minor thereafter (Pengel, 2003). 2.4.3. Data extraction for the frequency of an association between an individual predictor and outcome Data were extracted that allowed the evaluation of the relationship between individual predictors and

ARTICLE IN PRESS P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

outcomes. Relationships between prognostic factors and outcomes in bivariate and/or multivariable associations were extracted. The number of studies that examined each prognostic factor and outcome association and found this association to be statistically signiﬁcant (using any statistical technique) were tallied. Bivariate relationships between factors and outcomes were reported in the included studies using a variety of methods: differences between presenting mean scores for those who did or did not recover, w2 tests for association between presence of a factor and outcome, and odds ratios for poor recovery. In the included studies, multivariable relationships were also determined using a variety of methods: multiple linear regression, multiple logistic regression, discriminant analysis, Kaplan–Meier survival curves, principle components analysis, and Cox proportional hazards models. 2.4.4. Data extraction for the strength of an association (odds ratio) between an individual predictor and outcome Contingency tables (has/doesn’t have prognostic factor, improved/did not improve) were constructed from available data wherever possible, and odds ratios and their 95% CIs were calculated. Where a paper dichotomized a prognostic factor using more than one cut-off-point, arbitrarily only the dichotomisation associated with the largest odds ratio was used. Where a zero in one cell of a contingency table would otherwise preclude the calculation of an odds ratio, 0.5 was added to each cell (Cooper and Hedges, 1994). Where more than one study reported comparable prognostic factor/ outcome associations and odds ratios were provided or could be calculated, random-effects meta-analyses were performed (Fleiss, 1993). The relationship between the quality score of the studies that contributed prognostic factors with signiﬁcant odds ratios and those that did not was investigated using a paired t-test. The relationship between study quality score and the size of signiﬁcant odds ratios was investigated using linear regression. All statistical tests were performed using SPSS 11.0 (SPSS Inc., Chicago, USA).

15

Jordan et al., 1998; Haldorsen, 1998; Indahl et al., 1998; Smucker et al., 1998; Williams et al., 1998; De Gagne, 1999; Gaines and Hegmann, 1999; Macfarlane et al., 1999; Reis et al., 1999; Schiottz-Christensen et al., 1999; Tate, 1999; Thomas et al., 1999; Van der Weide et al., 1999; Dasinger, 2000; Mahmud et al., 2000; McIntosh et al., 2000a, 2006; Seferlis et al., 2000; Werneke and Hart, 2001, 2004; Fritz et al., 2001; Fransen et al., 2002; Flynn et al., 2002; Fritz and George, 2002; Hunt, 2002; Nordin et al., 2002; Carey, 2003; Schultz, 2004; Steenstra et al., 2005; Sieben et al., 2005; Faber et al., 2006; Heneweer et al., 2007. 3.1. Methodological quality The two authors scored the quality assessment criteria for the included studies with an initial complete agreement of 78%. The mean quality score for all of the included studies was 4.6/6 (SD 0.9, range 2–6). The methodological quality scores for the included studies are shown in Table 1. For 38 studies (76%), greater than 80% follow-up was reported. Reports of 36 studies (72%) described random or consecutive case selection for participant recruitment. For all but two studies (96%) the assessment of prognostic factors was reported in a way that would enable replication, and the same number of studies also measured outcomes using deﬁned methods. The protocols for 41 studies (82%) rendered it improbable that knowledge of the prognostic factor could bias measurement of outcome. For 20 studies (40%) data were reported in a form that allowed the odds of poor recovery to be determined. 3.2. Prognostic factors of poor recovery

3. Results

In total, data for 1501 prognostic factors and outcome associations were extracted and entered into a database (Filemaker Pro, Filemaker Inc., Santa Clara, USA). The initial coding of these associations using the taxonomy resulted in 63% complete agreement and differences in initial coding were minor. Associations between prognostic factors and outcomes (both short and long term) for each study are available from the corresponding author.

The ﬂowchart of the review is shown in Fig. 1. Fifty studies (54 papers) met the inclusion criteria (Pedersen, 1981; Roland and Morris, 1983; Roland et al., 1983; Singer et al., 1987; Lanier and Stockton, 1988; Goertz, 1990; Burton and Tillotson, 1991; Lehmann et al., 1993; Coste et al., 1994; Lindstrom et al., 1994; Burton et al., 1995; Gatchel et al., 1995a, b; Hadler, 1995; Indahl, 1995; Cherkin et al., 1996; Oleinick et al., 1996, 2000; Hazard et al., 1997; Reid et al., 1997; Valat et al., 1997; Wahlgren, 1997; Epping-

3.2.1. Results for the frequency and strength of bivariate association Table 2 describes both the number of studies that found each prognostic factor and outcome association to be statistically signiﬁcant (using any statistical technique), and the number of studies that examined each association (reported in brackets). For each prognostic factor, it also lists the largest statistically signiﬁcant odds ratio for a long-term association with any outcome (if available). In addition, it lists the pooled odds ratio for each prognostic

ARTICLE IN PRESS 16

P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

Fig. 1. Review ﬂow chart.

factor for which meta-analysis was performed. The entire set of prognostic factor and outcome associations for which odds ratios were available can be requested from the corresponding author. Signiﬁcant associations with poor recovery were found for many prognostic factors. Given the large number of factors that have been investigated and the relatively small size of some cohorts, some prognostic associations are likely to be spurious. For example, being currently employed (Macfarlane et al., 1999) and being currently unemployed (Reis et al., 1999) were both found in different studies to be prognostic factors. Greater weight can be put on factors that have been examined in a number of studies and reported in a form

that allowed meta-analytic pooling. However, these were few (approximately 8%) in this systematic review (identiﬁed in Table 2). 3.2.2. Results for the frequency and strength of multivariable association The number of prognostic factors retained in multivariable models, a description of any factors retained in more than one multivariable model, the variance explained by these models and the number of models tested in independent samples are shown in Table 3. Many prognostic factors were retained in multivariable models but few were retained across two or more multivariable models. Many multivariable

Table 1 Characteristics and methodological quality of included prognostic factor studies

1

Secondary care Orthopaedic outpatient department, and orthopaedic clinic Primary-care Osteopathic clinics

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

Burton and Tillotson (1991), United Kingdom Burton et al. (1995), United Kingdom Carey and Garrett (2003), USA Cherkin et al. (1996), USA Coste et al. (1994), France Dasinger et al. (2000), USA De Gagne (1999), Canada Epping-Jordan et al. (1998), USA Faber et al. (2006), Holland Flynn et al. (2002), USA Fransen et al. (2002), New Zealand Fritz et al. (2000, 2001), USA Gaines and Hegmann (1999), USA Gatchel et al. (1995a, b), USA Goertz (1990), USA Hadler (1995), USA Haldorsen et al. (1998), Norway Hazard et al. (1997), USA Heneweer et al. (2007), Holland Hunt et al. (2002), Canada Indahl et al. (1995, 1998), Norway Lanier and Stockton (1988), USA Lehmann et al. (1993), USA Lindstrom et al. (1994), Sweden Macfarlane et al. (1999), United Kingdom Mahmud et al. (2000), USA McIntosh et al. (2000a, b), Canada

n

Followup (%)

480% followup

Inception study

Prognostic factor

Outcome method

Blinded outcome

Analysis

Quality score (0–6)

47

No

Yes

Yes

Yes

Yes

No

4

105

74

No

Yes

Yes

Yes

Yes

No

4

Primary-care Health Maintenance Organisation Primary-care Health Maintenance Organisation Primary-care Medical clinics Workers’ compensation database Primary-care Physiotherapy clinic Primary-care Military clinic Primary-care Medical clinics Secondary-care Military Physiotherapy clinic Workers’ compensation database Secondary-care Physiotherapy clinic Secondary-care Occupational medicine clinic Secondary-care Occupational medicine and Orthopaedic clinics Primary-care Occupational medicine clinic Primary-care Medical clinics Primary-care Medical and Physiotherapy clinics Workers’ compensation database Secondary-care Physiotherapy clinics Workers’ compensation database Secondary-care Spine clinic Primary-care Medical clinics Secondary-care Spine clinic Secondary-care Spine clinic Primary-care Medical clinics

1633 219 103 433 41 140 103 75 854 78 55 421

96 90 89 100 49 56 87 95 100 100 100 91

Yes Yes Yes Yes No No Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes No Yes No Yes No

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes No Yes Yes Yes Yes Yes No Yes

No Yes No No Yes No No Yes Yes No No No

5 6 5 5 3 4 4 5 6 4 4 4

207 1366 260 304 66 192 975 121 60 103 294

100 96 100 88 85 83 100 96 78 95 100

Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes

Yes Yes Yes Yes Yes No Yes No No Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes No Yes Yes Yes Yes No Yes Yes

No No Yes Yes No No No No No No Yes

5 5 6 5 5 4 5 4 2 5 6

Workers’ compensation database Secondary-care Physiotherapy clinics, and a Workers’ compensation database

98 1984

100 88

Yes Yes

Yes Yes

Yes No

Yes Yes

Yes Yes

No No

5 4 17

190

ARTICLE IN PRESS

Setting

P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

Study

18

Table 1 (continued ) n

Followup (%)

480% followup

Inception study

Prognostic factor

Outcome method

Blinded outcome

Analysis

Quality score (0–6)

28 29 30 31 32 33 34 35

Secondary-care Physiotherapy clinics Workers’ compensation database Workers’ compensation database Workers’ compensation database Primary-care Medical clinic Workers’ compensation database Primary-care Medical clinics Primary-care Medical clinic

7077 1952 8652 115 78 210 238 230

100 100 100 77 100 99 92 84

Yes Yes Yes No Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes No Yes Yes Yes

Yes Yes Yes Yes No Yes Yes Yes

Yes Yes Yes No No Yes Yes Yes

Yes No No No Yes No Yes Yes

6 5 5 3 3 5 6 6

Primary-care Medical clinics

524

96

Yes

No

Yes

Yes

Yes

Yes

5

Workers’ compensation database Secondary-care Orthopaedic clinic Primary-care Medical clinics Primary-care Medical clinics Primary-care Medicine and Chiropractic clinics, and a Health Maintenance Organisation Primary-care Occupational Physicians Workers’ compensation database Primary-care Medical clinics

192 180 222 252 757

83 68 77 87 100

Yes No No Yes Yes

Yes No No No Yes

Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes

Yes Yes Yes No Yes

No Yes Yes No No

5 4 4 3 5

615 44 294

100 100 61

No Yes No

Yes Yes Yes

Yes Yes Yes

Yes Yes Yes

Yes Yes No

Yes No Yes

5 5 4

2493

100

Yes

No

Yes

Yes

No

Yes

4

120

90

Yes

No

Yes

Yes

Yes

Yes

5

138 223 300 136

55 84 57 60

No Yes No No

No Yes No Yes

Yes Yes Yes Yes

Yes Yes Yes Yes

Yes Yes Yes No

Yes No Yes No

4 5 4 3

36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

McIntosh et al. (2006), Canada Nordin et al. (2002), USA Oleinick et al. (1996) USA Oleske et al. (2000), USA Pedersen (1981), Denmark Reid et al. (1997), USA Reis et al. (1999), Israel Roland and Morris (1983), Roland et al. (1983), United Kingdom Schiottz-Christensen et al. (1999), Denmark Schultz et al. (2004), Canada Seferlis et al. (2000), Sweden Sieben et al. (2005), Holland Singer et al. (1987), Canada Smucker et al. (1998), USA Steenstra et al. (2005), Holland Tate et al. (1999), Canada Thomas et al. (1999), United Kingdom Valat (2000), France Van der Weide et al. (1999), Holland Wahlgren et al. (1997), USA Werneke and Hart (2001), USA Werneke and Hart (2004), USA Williams et al. (1998), USA

Primary-care Medical and secondary-care Rheumatology clinics Secondary-care Occupational medicine clinics Primary-care Military clinic Secondary-care Physiotherapy clinic Secondary-care Physiotherapy clinics Primary-care Military clinic

No ¼ criterion not met or unclear if met. Inception cohort ¼ all patients eligible for admission into the study were admitted into the study (random selection or as consecutive cases). Prognostic variable ¼ assessment of the prognostic variables was deﬁned in a way that enabled replication. Outcome method ¼ outcome variables were measured using deﬁned methods. Blinded outcome ¼ assessors measured prognostic variables and outcomes independently of each other, or it was likely that knowledge of the prognostic variable did not bias measurements of outcome. Analysis ¼ data are reported that allows estimates of risk of poor prognosis (e.g. odds or risk ratios) and the 95% CI.

ARTICLE IN PRESS

Setting

P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

Study

ARTICLE IN PRESS P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

19

Table 2 Studies providing evidence of statistical associations between factors and poor recovery in nonspeciﬁc low back pain Prognostic factor

Pooled odds ratio Largest signiﬁcant Number of studies reporting a signiﬁcant association (95% CI) for odds ratio (95% (number of studies investigating this association) long-term results CI) for long-term results Pain Activity Participation limitation restriction ST

History Older agea

Activity 2.23 Participation 1.61 1(3) (0.35–14.07) (1.05–2.47)a Participation 1.56 (1.05–2.30)

LT

ST

LT

ST

LT

3(7)

0(3)

5(11)

2(5)

4(15)

Cause-known Duration of episodes usually 430 days First onset Mechanism of injurya

1(1) 1(1) 1(1) 1(1)

0(1)

Onset-sudden Onset-gradual Previous therapy

0(2) 0(1) 2(2)

1(1)

2(3)

2(5)

Prior episodesa

Activity 2.98 (1.42–6.23) Participation. 0.99 (0.39–2.53)

Previous X-ray Pain Analgesia use Catch pain Hip ROM pain-free Hip ROM pain-prone knee bend Impulsion pain (Valsalva) Leg paina

Activity 2.12 (1.30–3.50) Activity 3.60 (1.20–11.00)

Pain latency Pain location Pain quality Prone knee bend

0(1) 0(1) 1(2)

1(2) 0(1)

Activity 2.53 (1.60–4.10)

0(1) 1(2) 0(1) 1(2)

0(1) 0(2)

3(10)

1(2)

2(5)

0(1) 3(11)

1(1)

0(1) 0(1)

1(1) Participation. 2.10 Pain 2.45 (0.96–4.62) (1.20–4.99)a Activity 3.30 (1.10–9.60)a

Participation 1.45 Activity 2.84 (1.10–1.91) (1.70–4.80)a Participation 10.98 (1.05–188.7)a

3(4)

0(1) 1(1) 1(1) 2(4)

1(3)

2(4)

1(3)

1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 21, 22, 23, 25, 26, 27, 29, 30, 31, 33, 36, 38, 39, 40, 42, 44, 46, 48, 49 5 5 4, 45 10, 15, 20, 29, 30, 34, 42, 43 1, 5, 10, 25, 36, 39 10, 15, 35, 39 1, 4, 20, 34, 36, 39 1, 2, 5, 6, 9, 10, 11, 15, 23, 27, 31, 34, 36, 38, 39, 40, 42, 43, 44, 45, 46, 48, 49 36

2 5 1 1

1(1) 1(1) 0(1)

McGill Pain Questionnaire score McKenzie classiﬁcation Multiple sites of pain Night pain Pain-movement related Pain-standing worst Longer pain duration Activity 2.27 (0.53–9.65) Pain frequency Greater pain intensitya

1(1)

Studies (see Supplementary Appendix 1 for code)

0(2)

1(1)

0(1)

5, 36, 38

3(6)

1(3)

4(5)

1, 2, 4, 10, 11, 15, 16, 22, 25, 26, 27, 31, 35, 36, 44, 45, 46, 48, 49

0(1)

0(2)

0(1)

2, 37, 40

1(1) 0(1)

1(1) 0(1)

1(1) 1(1)

48 48, 49 5 1, 5, 10, 36 5, 10 1, 2, 4, 5, 7, 10, 16, 25, 31, 34, 35, 36, 38, 40, 44, 48 1, 10, 27 2, 4, 5, 6, 9, 10, 14, 12, 20, 25, 27, 36, 37, 38, 39, 40, 43, 46, 48, 49, 50

1(1) 2(6)

0(1) 1(1) 2(2)

1(2) 2(5)

0(1) 1(2)

0(1) 0(1) 0(1)

0(1)

2(7)

0(1) 1(1) 1(1) 0(2)

0(3)

5(9)

0(1)

7(11)

0(1)

0(2) 0(2)

0(1) 0(2) 0(1)

20 1, 10, 36, 37, 39, 42 23, 36, 50 1

ARTICLE IN PRESS P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

20 Table 2 (continued ) Prognostic factor

Pooled odds ratio Largest signiﬁcant Number of studies reporting a signiﬁcant association (number of studies investigating this association) (95% CI) for odds ratio (95% long-term results CI) for long-term results Pain Activity Participation limitation restriction ST

Resisted contraction pain Sleep disturbance Unilateral pain

Activity 2.80 (1.40–5.60)

1(1)

ST

LT 1

Compression/ distraction test Limited extension Limited ﬂexion

0(1)

1(1)

27 32

1(1)

25, 44

0(1) 1(1)

1(1)

Participation 1.85 1(1) (1.17–2.90)a Participation 8.83 (1.94–40.11)a

Centralisationa

1(1) 1(2)

0(1)

0(1) 0(3)

0(1)

0(1)

1(2)

2 11, 21, 25, 44

1(3)

1, 2 9, 11, 22, 25, 31, 36

1(1)

10, 49

0(1) 0(1) Activity 4.49 (2.10–9.50)

Height Hip ROM Knee extension ROM

High Roland-Morris score Quebec Disability Scale score Severity of activity limitationa SF 36 score (physical health component) Sickness Impact Proﬁle (physical component)

LT

0(1) Activity 4.91 (1.62–14.83) Activity 6.42 (2.70–15.00)

Physical impairment Abdominal strength Greater aerobic capacity Antalgic list Greater body massa

Activity limitation Activity limitation (nonspeciﬁc) High Oswestry scorea

ST

1(1)

Widespread pain

Limited lateral ﬂexion McKenzie side-shift Muscle spasm Palpatory hypomobility Palpatory tenderness Percussion Physical Impairment Index Posture Red ﬂags Restricted movement Positive SLRa

LT

1(1) 0(2)

0(2) 1(2)

0(1)

10 1(1) 1(4)

0(1) 1(1)

0(1)

1(1) 1(3)

1, 10, 21, 35, 44 1, 10, 20, 35, 44

0(1)

31 01 44

0(1)

2, 10, 20, 35, 44 10 15 10

1(1) Activity. 2.39 (1.10–5.10)

1(1) 0(1)

0(2) 0(1)

1(2)

0(1) 1(1)

1(1) 0(1) 1(1) 0(1)

Activity 4.24 (0.40–44.84)

Activity 18.11 (2.11–155.10)

0(1) 0(2)

2(3)

0(1) Participation 2.69 Activity 3.35 (1.01–7.15) (1.42–2.37) Participation 5.89 (2.84–12.20)a

1(2)

1(2)

0(1)

10 36 12, 50

0(1) 0(1)

1(1) 0(1) 1(2)

0(1) 0(1) 2(5)

0(1) 2(3)

0(1) 0(1) 0(4)

10, 35 20 1, 5, 36, 44 1, 2, 5, 10, 20, 22, 27, 32, 35, 36, 38, 40, 49

1(3)

1(4)

1(1)

2(3)

14, 19, 20, 37, 39, 40, 42, 46 10, 11, 12, 38, 49

1(4)

2, 4, 5,9, 22, 23, 35, 46

0(3)

1(3)

0(1)

39

0(1)

4 0(1)

0(1)

Studies (see Supplementary Appendix 1 for code)

1(2)

37 50

ARTICLE IN PRESS P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

21

Table 2 (continued ) Prognostic factor

Pooled odds ratio Largest signiﬁcant Number of studies reporting a signiﬁcant association (number of studies investigating this association) (95% CI) for odds ratio (95% long-term results CI) for long-term results Pain Activity Participation limitation restriction ST

Sit to stand Sit-up test Walking/stair climbing

ST

LT 45 1 45

1(1)

2(2)

Participation 1.99 (1.39–2.86) Pain 2.68 2(2) (1.28–5.58) Participation 2.08 (1.50–2.89)

Comorbiditya Company size Compensable case management Compensable condition

1(2)

0(1)

1(4)

1(1) 1(1)

1(1) 1(1)

1(2) 0(1)

0(1)

2(4) 1(4) 1(1)

14, 21, 27, 49 1, 20, 22, 32, 34, 38, 45, 46, 48, 49 18

1(1)

11, 15

1(3)

4, 11, 14, 37, 45

1(1)

0(1)

1(1)

22, 27, 28, 29 6, 30 34

1(4)

0(1)

3(4) 0(2)

4, 5, 14, 16, 23, 25, 34, 36, 40, 45, 46, 48 2, 19, 36, 39, 46

2(3)

6, 21, 33, 46

1(1) 2(4)

Poor coping strategiesa Activity 3.24 (1.46–7.22) Current employment length Currently employed Currently unemployed

Activity 2.37 (1.30–4.40)

3(4)

1(2) 1(1) 2(2)

Pain 28.70 (3.52–233.91) Participation 1.82 (1.03–3.22)

Longer driving time

Participation 0.99 (0.63–1.55)

Pain 1.97 (0.98–3.97) Activity 1.38 (0.64–2.99) Participation 0.61 (0.30–1.24)

1(3)

1(2)

2(2)

0(1) 1(1)

0(1)

0(1)

0(1) 1(1)

0(1)

0(4)

0(1)

1(2)

5, 25, 36 34, 44 1(7) 0(1)

2, 4, 11, 12, 14, 22, 23, 31, 34, 37, 39, 48, 49 2, 23, 47, 50

2(2)

11, 21 5, 14, 23 4, 7, 11, 14, 16, 17, 21, 22, 23, 34, 39 4, 7, 15, 21, 22, 23, 27, 29, 30, 38, 42, 46, 48, 49 3, 4, 5, 8, 14, 22, 50 23, 37, 42

0(2)

0(1)

1(1)

0(1) 0(1)

0(2) 1(5)

0(3)

0(1)

1(3)

1(2)

4(9)

0(1)

1(2)

0(1)

2(3)

1(2)

1(1) 4(5)

1(1) Participation 2.77 (1.02–7.55) Pain 2.23 0(2) (1.20–4.16)

0(1) 2(2)

2(2)

0(2) 1(4)

2(6)

0(3)

2(9)

Employment type Ethnicity Expectations of recoverya Family support Fear-avoidance beliefsa

0(3)

Activity 4.60 (1.40–14.00)a

Participation 1.50 (0.48–4.71)

Distress

Gendera

LT

1(1)

Psychosocial Alternative duties available Anxiety

DSM diagnosisb Lower education level

ST

1(1)

Participation restriction Duration of episode Longer time-off-work Participation 1.47 Activity 15.1 (0.74–2.92) (4.64–49.17) Vermont Disability Prediction Questionnaire

Depressiona

LT

Studies (see Supplementary Appendix 1 for code)

0(3) 1(4) 1(6)

3(16)

11, 31, 34, 46 2, 9, 10, 12, 19, 24, 39, 48, 49 2, 4, 5, 6, 7, 9, 10, 11, 14, 15, 17, 21, 22, 23, 26, 27, 29, 30, 31, 33, 35, 36, 38, 40, 42, 46, 48, 49

ARTICLE IN PRESS P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

22 Table 2 (continued ) Prognostic factor

Pooled odds ratio Largest signiﬁcant Number of studies reporting a signiﬁcant association (number of studies investigating this association) (95% CI) for odds ratio (95% long-term results CI) for long-term results Pain Activity Participation limitation restriction ST

a

General Health

Activity 4.56 (2.19–9.49)

General Health Questionnaire score

Income level Less job controla

Participation 1.95 (1.02–3.72)a

Job difﬁculties Job physically demandinga

Job psychological demandsa Job satisfactiona

Pain 2.22 0(1) (1.14–4.34) Activity 7.90 (2.50–25.00)a Activity 3.34 1(1) (1.50–7.20) Participation 2.78 (2.00–3.87)

Activity 4.00 0(2) (1.10–14.00) Participation 2.04 (1.41–2.96)a

LT

ST

LT

ST

Litigation External locus of control Marital status

Activity 3.20 (1.20–9.00)a

3(3)

Activity 4.30 (1.60–12.00) Participation 0.31 (0.15–0.65)

Medication use Negative thoughts Non-compensable condition Non-organic signs

3(3)

2(3)

4, 25, 31, 37, 44, 46

0(1)

1(1)

1(1)

4, 11, 25, 44

1(2) 0(1)

1(2)

2(2) 0(2)

1(3) 2(6)

7, 8, 16, 21, 30, 33 6, 9, 11, 33, 37, 45, 46

0(1) 2(3)

2(3) 2(3)

1(2) 2(11)

5, 34, 42, 45, 46 1, 4, 5, 6, 9, 11, 15, 22, 24, 26, 27, 29, 37, 42, 45, 46, 48

2(6)

9, 10, 23, 37, 42, 46

1(6)

4, 5, 11, 25, 31, 38, 44, 46, 48, 49, 50

1(1) 2(3)

23 2, 11, 21, 46

3(1)

4, 7, 14, 17, 21, 23, 30, 36 20, 40 20, 39 38, 45

1(4)

1(1)

1(4)

3(6)

0(2)

0(1)

0(1)

0(1) 0(1) 0(1)

0(1) 0(1) 0(1)

1(1)

1(6)

1(1) 1(1)

1(2)

0(2)

1(1)

(2)

Participation 2.69 (1.09–6.66)

Participation in an early return-to-work program Less patient satisfaction Personality SF 36 score (mental health component) Sickness Impact Proﬁle (psychosocial component) Signiﬁcant life events Smoking

0(1)

1(2)

0(1)

Pain behaviours Being a parent

LT

2(2)

0(2) Activity 2.65 (1.27–5.50) Participation 0.82 (0.42–1.62)

1(1)

1(2) 2(2)

10, 12, 13, 20, 23, 27, 48, 49 1, 2, 34, 48, 49 17, 21, 30

1(1)

43

1(1)

34 1(1)

0(1)

0(1)

Less social activity

Activity 2.31 (1.12–4.76)

Somatic anxiety

Participation 1.41 (0.82–2.44)

1(2)

Activity 2.26 (1.30–4.40) Participation 2.79 (1.98–3.93) 1(1)

Studies (see Supplementary Appendix 1 for code)

0(1)

1(1) 0(1)

0(1)

0(1)

0(1) 1(4)

14, 46 37 50

0(1) 1(1)

1(1) 0(4)

2(1)

2(2)

11, 22 4, 11, 22, 25, 31, 38, 44, 46 11, 44, 46

1(2)

1(2)

2, 11, 48, 49

ARTICLE IN PRESS P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

23

Table 2 (continued ) Prognostic factor

Pooled odds ratio Largest signiﬁcant Number of studies reporting a signiﬁcant association (number of studies investigating this association) (95% CI) for odds ratio (95% long-term results CI) for long-term results Pain Activity Participation limitation restriction

Sporting activity Stress Substance abuse Well-being

ST

LT

1(1)

1(2) 0(1)

1(2) 1(1)

0(2)

1(2)

Therapeutic response Traction helps Treatment type Clinician factors Clinician conﬁdence in managing LBP Diagnosis

ST

LT

ST

LT 1(2) 0(1) 0(1) 1(1)

1, 9, 21, 45, 46 31 14 9, 50

1(2)

1 14, 42

1(1)

0(1) Activity 11.6 (2.90–46.90)

1(1)

Diagnostic testing

0(1) 0(1)

1(2)

Studies (see Supplementary Appendix 1 for code)

41 1(2) 0(1)

0(1)

15, 22, 27, 34, 36, 46 22

Bold odds ratios are statistically signiﬁcant. Italicised odds ratios have a point estimate less than 1.0, and indicate a reduced risk of poor recovery. a Retained in two or more multivariate models. ST ¼ short-term outcome (o3 months), LT ¼ long-term outcome (43 months). b DSM ¼ the Diagnostic and Statistical Manual of the American Psychiatric Association. Activity ¼ activity limitation, participation ¼ participation restriction.

predictive models were reported in the included studies but very few were tested in independent samples. On average, the outcome variance explained by these models was less that 50%.

odds ratios should be interpreted cautiously as only a preliminary guide to the strength of the association between prognostic factor and outcome.

3.2.3. Characteristics of the available data Table 4 reports the number of prognostic factors investigated within each assessment domain, the proportion of these for which odds ratios were available, and the number of prognostic factor/outcome associations reported with signiﬁcant odds ratios. The included studies provided a dataset of prognostic factor investigations that are not complete, as most factors have not been investigated for all outcome domains. Therefore, it is not currently possible to comment on whether some prognostic factors are signiﬁcantly associated with some outcomes but not others. The most common prognostic factors with signiﬁcant odds ratios were in the pain and psychosocial domains, but these were also the domains most commonly investigated and reported in ways that facilitated estimation of odds ratios. There was no signiﬁcant difference between the quality score of the studies that contributed prognostic factors with signiﬁcant odds ratios and those that did not (t ¼ 1.48, p ¼ 0.13). However, there was a systematic inverse relationship between study quality score and the size of signiﬁcant odds ratios. Signiﬁcant odds ratios with higher values tended to be derived from studies with lower quality scores (r ¼ 0.37, p ¼ 0.03). Therefore, the reported

4. Discussion This considerable body of prognostic research has resulted in little certainty regarding important NSLBP prognostic factors, due to contradictory, inconsistent and incomplete ﬁndings. Studies have investigated different factors in varying cohorts (settings, interventions), used diverse investigative strategies, measured factors and outcomes in unstandardized ways, and evaluated relationships to an incomplete set of outcomes. Our intention had been to synthesize data on the bivariate associations by ranking these on their strength of association (as measured by odds ratios). Our conﬁdence in this method was eroded by ﬁnding a signiﬁcant inverse relationship between quality score and odds ratio size, and also by only a third of studies reporting data in a form that allowed the odds of poor recovery to be determined. We do not believe that ‘vote counting’ (summing the number of studies ﬁnding an association, with or without regard to the number of studies that have investigated this association) to be a valid alternative method by which to synthesize these results. When examined in isolation, a number of presenting characteristics have been found to be associated with

ARTICLE IN PRESS P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

24 Table 3

Summary of the multivariable models Pain

Activity limitation

Number of prognostic factors retained in any multivariable model

28

27

(1, 4, 5, 7, 25, 28, 45, 47, 48, 50)a,b

(2, 10, 12, 22, 28, 31, 36, 38, 44, 46, 50)

Number of prognostic factors retained in 2 multivariable models

6

5

Older age, prior episodes, leg pain, unable to remain at work, depression, job dissatisfaction

Prior episodes, leg pain, high fear-avoidance beliefs, higher pain intensity, poor coping strategies

Number of prognostic factors retained in 3 or more multivariable models

0

2

Number of multivariable models tested in independent samples (total number of multivariable models)

1 (12)c

1 (18)

Outcome variance explained (r2)b

46% (SD7%, range 26–52%) (8, 31, 50)a

43% (SD 17%, range 23–69%) (2, 8, 12, 22, 31, 39, 50)

Participation restriction 35 (5, 6, 14, 15, 16, 20, 21, 22, 26, 27, 29, 37, 38, 43, 46, 48, 49) 10 Older age, prior episodes, expectations of recovery, female gender, mechanism of injury, leg pain, higher activity limitation, higher pain intensity, size of company, higher job physical demands, higher job psychological demands 6

Leg pain, higher pain intensity

Older age, prior episodes, expectations of recovery, leg pain, higher activity limitation, higher pain intensity, higher job physical demands 1 (30)

Time-off-work 29% (SD 9%, range 22–35%) (15, 42)

Return to work 32% (42)

a

Studies with multivariable models contributing to these summaries are listed in parentheses. Not all multivariable models were reported in ways that allowed identiﬁcation of the predictors that were retained in the model, or allowed determination of the amount of variance explained by the model. c Some studies contributed more than one multivariable model (e.g., different models for short-term and long-term outcomes). b

Table 4 Prognostic factors for poor recovery in recent-onset nonspeciﬁc low back pain. Analysis of prognostic factors by assessment domaina Outcome Prognostic domain

Psychosocial History Pain Physical impairment Activity limitation Participation restriction Clinician factors Therapeutic response

Pain

Activity limitation

Participation restriction

Sort term

Long term

Sort term

Long term

Sort term

Long term

20 9 9 8 3 1 1 0

27 5 14 8 4 1 0 1

11 5 10 12 1 1 1 0

34 5 13 14 4 1 2 0

18 6 8 9 1 1 1 0

40 4 13 10 4 3 1 1

(95%), 16 (78%), 3 (44%), 2 (25%), 1 (0%), 0 (0%), 0 (0%), 0 (0%), 0

(19%), 3 (0%), 0 (7%), 1 (0%), 0 (0%), 0 (0%), 0 (0%), 0 (0%), 0

(22%), 1 (20%), 0 (80%), 2 (42%), 4 (0%), 0 (0%), 0 (0%), 0 (0%), 0

(44%), 8 (100%), 3 (69%), 4 (71%), 3 (25%), 1 (100%), 1 (50%), 1 (0%), 0

(6%), (0%), (0%), (0%), (0%), (0%), (0%), (0%),

1 0 0 0 0 0 0 0

(58%), 0 (50%), 1 (31%), 4 (30%), 0 (25%), 1 (67%), 1 (0%), 0 (0%), 0

a The number of prognostic factors investigated within each domain (and the proportion reported with odds ratios), and number of factors that reported signiﬁcant odds ratios.

poor recovery from recent-onset NSLBP, but few factors are retained across studies in multivariable modelling. This is understandable, as the multivariable analyses used in these studies varied with respect to the

statistical technique used, the factor entry order, and factor/subject ratio, all of which can affect the capacity for an association to be displayed and the conﬁdence in that association (Bradburn et al., 2003; Clark et al.,

ARTICLE IN PRESS P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

2003). In addition, very few multivariable models were validated in independent samples despite the importance of this method for conferring credibility to a model. It is typical for the predictive strength of a multivariable model to degrade when tested in an independent sample (Altman and Lyman, 1998). Most multivariable models explained only moderate amounts of variance in outcomes. However, most primary care clinicians believe that NSLBP is heterogeneous and describes a number of conditions (sub-groups) (Kent and Keating, 2004). Therefore, if sub-groups with differing prognoses do exist within the NSLBP population, a single predictive model might poorly predict outcomes for all and subgroup-speciﬁc predictive modelling may be required to increase model strength. These ﬁndings illustrate the methodological constraints to synthesising these data. Although this systematic review does describe the signiﬁcant prognostic factor/outcome associations that have been found for recent-onset NSLBP, it is unable to conﬁdently either report the predictive strength of these associations or synthesize the ﬁndings beyond that presented in Tables 2 and 3. If we are to progress the investigation of prognostic factors of poor recovery from recent-onset NSLBP, the ﬁndings of this review could be used to assist the design of subsequent prognostic studies. If researchers are to design prognostic studies based on theoretical models and the results of previous studies, the associations described in this review can be used to inform those processes. It would also be ideal if researchers undertaking prognostic studies could reach consensus regarding standardized methods for measuring a core set of prognostic variables and outcomes (Deyo et al., 1998). Study design and the selection of multivariable technique could be informed by best practice in other ﬁelds of prognostic research (Altman, 2001, 1998), and this would improve the capacity for pooling of results across studies. Strengths of this systematic review are its comprehensiveness, the inclusion of non-occupational outcomes, the mapping of prognostic factors across assessment domains, and the description of both bivariate and multivariable ﬁndings. It provides empirical data to inform decisions regarding predictor/outcome relationships to investigate in future prognostic studies; it presents a broad picture of the current state of knowledge and signals methods for advancing knowledge in this ﬁeld. There are also a number of limitations to this systematic review. For each study, we rated the predictors and outcomes as being measured in ways that allowed reproducibility, but did not rate the validity of measurement used for each predictor and outcome. As there were 1501 prognostic factor and outcome associations, the analysis and reporting of this were beyond the scope of the current review. As only English

25

language published papers were included in this review, there is potential for the ﬁndings to contain cultural and publication bias. Finally, across the included studies there is potential for differential treatment effects to bias estimates of prognostic ability; however, as the consensus of systematic reviews is that common treatments are equally effective in NSLBP (Ernst and Harkness, 2001; Ferreira et al., 2002, 2003; Assendelft et al., 2003; Cherkin et al., 2003), this element may not have affected the results.

5. Conclusions Despite the number of studies that have investigated prognostic factors for their association with poor outcome for people with recent-onset NSLBP, uncertainty remains regarding which factors are associated with particular outcomes, the strength of those associations and the extent of confounding between prognostic factors. This uncertainty is the result of the disparate methods that have been used in these investigations, and an inverse relationship between study quality and the reported strength of these associations. Both of these characteristics of the available data currently inhibit meaningful synthesis of prognostic ﬁndings. The clinical implication is that the formation of clinically useful predictive models remains dependent on further high-quality research. The research implications are that cohort or controlled trial studies of recent-onset NSLBP can use the ﬁndings of this review to inform prognostic or covariate factor selection, and that prognostic studies would ideally be designed to enhance the capacity for ﬁndings to be pooled with those of other studies. Acknowledgements This work was supported by Joint Coal Board Health & Safety Trust (Australia) and the National Health and Medical Research Council (Grant 384366). No beneﬁts in any form have been, or will be, received from a commercial party related directly or indirectly to the subject of this manuscript. Appendix A. Supplementary materials Supplementary data associated with this article can be found in the online version at doi:10.1016/j.math. 2007.05.009. References Altman DG. Systematic reviews of evaluations of prognostic variables. British Medical Journal 2001;323:224–8.

ARTICLE IN PRESS 26

P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

Altman DG, Lyman GH. Methodological challenges in the evaluation of prognostic factors in breast cancer. Breast Cancer Research and Treatment 1998;52:289–303. Assendelft WJ, Morton SC, Yu EI, Suttorp MJ, Shekelle PG. Spinal manipulative therapy for low back pain: a meta-analysis of effectiveness relative to other therapies. Annuls of Internal Medicine 2003;138(11):871–81. Bouter LM, Pennick V, Bombardier C. Cochrane back review group. Spine 2003;28(12):1215–8. Bradburn MJ, Clark TG, Love SB, Altman DG. Survival analysis Part 3: multi-variate data analysis-choosing a model and assessing its adequacy and ﬁt. British Journal of Cancer 2003;89:605–11. Burton AK, Tillotson KM. Prediction of the clinical course of low-back trouble using multivariable models. Spine 1991;16(1): 7–14. Burton AK, Tillotson KM, Main CJ, Hollis S. Psychosocial predictors of outcome in acute and subchronic low back trouble. Spine 1995;20(6):722–8. Carey TS. The relation of race to outcomes and the use of health care services for acute low back pain. Spine 2003;28(4):390–4. Cherkin DC, Deyo RA, Street JH, Barlow W. Predicting poor outcomes for back pain seen in primary care using patients’ own criteria. Spine 1996;21(24):2900–7. Cherkin DC, Sherman KJ, Deyo RA, Shekelle PG. A review of the evidence for the effectiveness, safety, and cost of acupuncture, massage therapy, and spinal manipulation for back pain. Annuls of Internal Medicine 2003;138(11):898–906. Clark TG, Bradburn MJ, Love SB, Altman DG. Survival analysis Part 4: further concepts and methods in survival analysis. British Journal of Cancer 2003;89:781–6. Cooper H, Hedges LV. The handbook of research synthesis. New York, NY, USA: Russel Sage Foundation; 1994. Coste J, Delecoeuillerie G, Cohen de Lara A, Le Parc JM, Paolaggi JB. Clinical course and prognostic factors in acute low back pain: an inception cohort study in primary care practice. British Medical Journal 1994;308(6928):577–80. Crook J, Moldofsky H, Shannon H. Determinants of disability after a work related musculoskeletal injury. Journal of Rheumatology 1998;25:1570–7. Dasinger LK. Physical workplace factors and return to work after compensated low back injury: a disability phase-speciﬁc analysis. Journal of Occupational and Environmental Medicine 2000;42(3): 323–33. De Gagne TA. The evolution of chronic pain: adjustment status following treatment for acute low back pain. Dissertation Abstracts International: Section B: the Sciences and Engineering 1999; 60(3-B). Deyo R, Rainville J, Kent D. What can the history and physical examination tell us about low back pain. Journal of the American Medical Association 1992;268(6):760–5. Deyo RA, Battie M, Beurskens AJ, Bombardier C, Croft P, Koes B, et al. Outcome measures for low back pain research. A proposal for standardized use. Spine 1998;23(18):2003–13. Di Fabio RP. Disability and functional status in patients with low back pain receiving workers’ compensation: a descriptive study with implications for the efﬁcacy of physical therapy. Physical Therapy 1995;75(3):180–93. Epping-Jordan J, Wahlgren D, Williams R, Pruitt S, Slater M, Patterson T, et al. Transition to chronic pain in men with low-back pain: predictive relationships among pain intensity, disability and depressive symptoms. Health Psychology 1998;17:421–7. Ernst E, Harkness E. Spinal manipulation: a systematic review of sham-controlled, double-blind, randomized clinical trials. Journal of Pain & Symptom Management 2001;22(4):879–89. Faber E, Burdorf A, Bierma-Zeinstra SMA, Miedema HS, Koes BW. Determinants for improvement in different back pain measures and

their inﬂuence on the duration of sickness absence. Spine 2006;31(13):1477–83. Ferreira M, Ferreira P, Latimer J, Herbert R, Maher C. Does spinal manipulative therapy help people with chronic low back pain? Australian Journal of Physiotherapy 2002;48:277–83. Ferreira M, Ferreira P, Latimer J, Herbert R, Maher C. Efﬁcacy of spinal manipulative therapy for low back pain of less than three months’ duration. Journal of Manipulative & Physiological Therapeutics 2003;26(9):593–601. Fleiss JL. The statistical basis of meta-analysis. Statistical Methods in Medical Research 1993;2:121–45. Flynn T, Fritz JW, Whitman M, Wainner RS, Magel J, Rendeiro D, et al. A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improvement with spinal manipulation. Spine 2002;27(24):2835–43. Fransen M, Woodward M, Norton R, Coggan C, Dawe M, Sheridan N. Risk factors associated with the transition from acute to chronic occupational back pain. Spine 2002;27(1):92–8. Fritz J, George S. Identifying psychosocial variables in patients with acute work-related low back pain: the importance of fearavoidance beliefs. Physical Therapy 2002;82(10):973–84. Fritz JM, George SZ, Delitto A. The role of fear-avoidance beliefs in acute low back pain: relationships with current and future disability and work status. Pain 2001;94(1):7–15. Gaines WGJ, Hegmann KT. Effectiveness of Waddell’s nonorganic signs in predicting a delayed return to regular work in patients experiencing acute occupational low back pain. Spine 1999;24(4): 396–400. Gatchel RJ, Polatin PB, Kinney RK. Predicting outcome of chronic back pain using clinical predictors of psychopathology: a prospective analysis. Health Psychology 1995a;14(5):415–20. Gatchel RJ, Polatin PB, Mayer TG. The dominant role of psychosocial risk factors in the development of chronic low back pain disability. Spine 1995b;20(24):2702–9. Goertz MN. Prognostic indicators for acute low-back pain. Spine 1990;15(12):1307–10. Hadler NM. The inﬂuence of indemniﬁcation by workers’ compensation insurance on recovery from acute backache. North Carolina Back Pain Project. Spine 1995;20(24):2710–5. Haldorsen EMH. Predictors for outcome of a functional restoration program for low back pain patients-a 12-month follow-up study. European Journal of Physical Medicine & Rehabilitation 1998;8(4):103–9. Hazard RG, Haugh LD, Reid S, McFarlane G, MacDonald L. Early physician notiﬁcation of patient disability risk and clinical guidelines after low back injury: a randomized, controlled trial. Spine 1997;22(24):2951–8. Heneweer H, Aufdemkampe G, van Tulder MW, Kiers H, Stappaerts KH, Vanhees L. Psychosocial variables in patients with (sub)acute low back pain an inception cohort in primary care physical therapy in the Netherlands. Spine 2007;32(5):586–92. Hestbaek L, Leboeuf C, Manniche C. Low back pain: what is the longterm course? A review of studies of general patient populations. European Spine Journal 2003;12(2):149–65. Hudak PL, Cole DC, Haines AT. Understanding prognosis to improve rehabilitation: the example of lateral elbow pain. Archives of Physical Medicine and Rehabilitation 1996;77:586–93. Hunt DG. Are components of a comprehensive medical assessment predictive of work disability after an episode of occupational low back trouble? Spine 2002;27(23):2715–9. Indahl A. Good prognosis for low back pain when left untampered: a randomized clinical trial. Spine 1995;20(4):473–7. Indahl A, Kaigle A, Reikera˚s O, Holm S. Five-year follow-up study of a controlled clinical trial using light mobilization and an informative approach to low back pain. Spine 1998;23(23): 2625–30.

ARTICLE IN PRESS P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28 Kent PM, Keating J. Do primary-care clinicians think that nonspeciﬁc low back pain is one condition? Spine 2004;29:1022–31. Kuch K. Psychological factors and the development of chronic pain. Clinical Journal of Pain 2001;17(4 Suppl):S33–8. Lanier DC, Stockton P. Clinical predictors of outcome of acute episodes of low back pain. Journal of Family Practice 1988;27(5): 483–9. Lehmann TR, Spratt KF, Lehmann KK. Predicting long-term disability in low back injured workers presenting to a spine consultant. Spine 1993;18(8):1103–12. Lindstrom I, Ohlund C, Nachemson A. Validity of patient reporting and predictive value of industrial physical work demands. Spine 1994;19(8):888–93. Linton S. A review of psychological risk factors in back and neck pain. Spine 2000;25:1148–56. Macfarlane GJ, Thomas E, Croft PR, Papageorgiou AC, Jayson MI, Silman AJ. Predictors of early improvement in low back pain amongst consulters to general practice: the inﬂuence of pre-morbid and episode-related factors. Pain 1999;80(1–2):113–9. Mahmud MA, Webster BS, Courtney TK, Matz S, Tacci JA, Christiani DC. Clinical management and the duration of disability for work-related low back pain. Journal of Occupational and Environmental Medicine 2000;42(12):1178–87. McIntosh G, Frank J, Hogg-Johnson S, Bombardier C, Hall H. Prognostic factors for time receiving worker’s compensation beneﬁts in a cohort of patients with low back pain. Spine 2000a;25(2):147–57. McIntosh G, Frank J, Hogg-Johnson S, Hall H, Bombardier C. Low back pain prognosis: structured review of the literature. Journal of Occupational Rehabilitation 2000b;10(2):101–15. McIntosh G, Hall H, Boyle C. Contribution of nonspinal comorbidity to low back pain outcomes. Clinical Journal of Pain 2006;22(9):765–9. Nordin M, Hiebert R, Pietrek M, Alexander M, Crane M, Lewis S. Association of comorbidity and outcome in episodes of nonspeciﬁc low back pain in occupational populations. Journal of Occupational & Environmental Medicine 2002;44(7):677–84. Oleinick A, Gluck J, Guire K. Factors affecting ﬁrst return to work following a compensable occupational back injury. American Journal of Industrial Medicine 1996;30:540–55. Oleske D, Andersson G, Lavender S, Hahn J. Association between recovery outcomes for work-related low back disorders and personal, family and work factors. Spine 2000;25(10):1259–65. Pedersen PA. Prognostic indicators in low back pain. Journal of the Royal College of General Practitioners 1981;31(225):209–16. Pengel LH. Acute low back pain: systematic review of its prognosis. British Medical Journal 2003;327(7410):1–5. Pincus T, Burton A, Vogel S, Field A. A systematic review of psychological factors as predictors of chronicity/disability in prospective cohorts of low back pain. Spine 2002;27(5):E109–20. Reid S, Haugh LD, Hazard RG, Tripathi M. Occupational low-back pain: recovery curves and factors associated with disability. Journal of Occupational Rehabilitation 1997;7(1):1–14. Reis S, Hermoni D, Borkan JM, Biderman A, Tabenkin C, Porat A. A new look at low back complaints in primary care: a RAMBAM Israeli family practice network study. Journal of Family Practice 1999;48(4):299–303. Rohling ML, Binder LM, Langhinrichsen Rohling J. Money matters: a meta-analytic review of the association between ﬁnancial compensation and the experience and treatment of chronic pain. Health Psychology 1995;14(6):537–47. Roland M, Morris R. A study of the natural history of low-back pain. Part II: development of guidelines for trials of treatment in primary care. Spine 1983;8(2):145–50. Roland MO, Morrell DC, Morris RW. Can general practitioners predict the outcome of episodes of back pain? British Medical Journal Clinical Research Ed 1983;286(6364):523–5.

27

Schiottz-Christensen B, Nielsen GL, Hansen TS, Sorensen HT, Olesen F. Long-term prognosis of acute low back pain in patients seen in general practice: a 1-year prospective follow-up study. Family Practice 1999;16(3):223–32. Schultz IZ. Psychosocial factors predictive of occupational low back disability: towards development of a return-to-work model. Pain 2004;107(1–2):77–85. Seferlis T, Nemeth G, Carlsson AM. Prediction of functional disability, recurrences, and chronicity after 1 year in 180 patients who required sick leave for acute low-back pain. Journal of Spinal Disorders 2000;13(6):470–7. Sieben JM, Vlaeyen JWS, Portegijs PJM, Verbunt JA, van RietRutgers S, Kester ADM, et al. A longitudinal study on the predictive validity of the fear-avoidance model in low back pain. Pain 2005;117(1–2):162–70. Singer J, Gilbert JR, Hutton T, Taylor DW. Predicting outcome in acute low-back pain. Canadian Family Physician 1987;33: 655–9. Smucker DR, Konrad TR, Curtis P, Carey TS. Practitioner selfconﬁdence and patient outcomes in acute back pain. Archives of Family Medicine 1998;7(3):223–8. Spengler DM, David DP. Industrial low back pain: a practical approach. In: Wiesel SW, editor. Industrial low back pain: a comprehensive approach. Charlottesville, VA, USA: The Michie Company; 1985. p. 869–71. Steenstra IA, Koopman FS, Knol DL, Kat E, Bongers PM, de Vet HC, et al. Prognostic factors for duration of sick leave due to lowback pain in Dutch health care professionals. Journal of Occupational Rehabilitation 2005;15(4):591–605. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, et al. Meta-analysis of observational studies in epidemiology— a proposal for reporting. Journal of the American Medical Association 2000;283(15):2008–12. Tate RB. Predictors of time loss after back injury in nurses. Spine 1999;24(18):1930–5. Teasell RW. Compensation and chronic pain. Clinical Journal of Pain 2001;17(4 Suppl):S46–51. Teasell RW, Bombardier C. Employment-related factors in chronic pain and chronic pain disability. Clinical Journal of Pain 2001; 17(4 Suppl):S39–45. Thomas E, Silman AJ, Croft PR, Papageorgiou AC, Jayson MI, Macfarlane GJ. Predicting who develops chronic low back pain in primary care: a prospective study. British Medical Journal 1999;318(7199):1662–7. Torstensen TA, Ljunggren AE, Meen HD, Odland E, Mowinckel P, Geijerstam S. Efﬁciency and costs of medical exercise therapy, conventional physiotherapy, and self-exercise in patients with chronic low back pain. A pragmatic, randomized, single-blinded, controlled trial with 1-year follow-up. Spine 1998;23(23): 2616–24. Truchon M. Determinants of chronic disability related to low back pain: towards an integrative biopsychosocial model. Disability and Rehabilitation 2001;23(17):758–67. Valat JP, Goupille P, Vedere V. Low back pain: risk factors for chronicity. Revue du Rhumatisme (English Edition) 1997;64(3): 189–94. Van der Weide WE, Verbeek JH, Salle HJ, Van Dijk FJ. Prognostic factors for chronic disability from acute low-back pain in occupational health care. Scandinavian Journal of Work, Environment & Health 1999;25(1):50–6. Van Tulder M, Malmivaara A, Esmail R, Koes B. Exercise therapy for low back pain-a systematic review within the framework of the Cochrane Collaboration Back Review Group. Spine 2000;25(21): 2784–96. Wahlgren DR. One-year follow-up of ﬁrst onset low back pain. Pain 1997;73(2):213–21.

ARTICLE IN PRESS 28

P.M. Kent, J.L. Keating / Manual Therapy 13 (2008) 12–28

Werneke M, Hart DL. Centralization phenomenon as a prognostic factor for chronic low back pain and disability. Spine 2001;26(7): 758–65. Werneke MW, Hart DL. Categorizing patients with occupational low back pain by use of the Quebec Task Force Classiﬁcation system versus pain pattern classiﬁcation procedures:

discriminant and predictive validity. Physical Therapy 2004;84(3): 243–54. Williams RA, Pruitt SD, Doctor JN, Epping-Jordan JE, Wahlgren DR, Grant I, et al. The contribution of job satisfaction to the transition from acute to chronic low back pain. Archives of Physical Medicine and Rehabilitation 1998;79(4):366–74.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 29–36 www.elsevier.com/locate/math

Review

Pre-manipulative testing and the use of the velocimeter Lucy C. Thomasa,, Darren A. Rivetta, Philip S. Boltonb,c a

Discipline of Physiotherapy, School of Health Sciences, Faculty of Health, The University of Newcastle, University Drive, Callaghan, NSW 2308, Australia b Discipline of Human Physiology, School of Biomedical Sciences, Faculty of Health, The University of Newcastle, NSW, Australia c Hunter Medical Research Institute, New Lambton, NSW, Australia Received 25 May 2004; received in revised form 22 June 2006; accepted 30 November 2006

Abstract Manipulation of the cervical spine remains a common intervention for neck pain and dysfunction, despite the well-documented associated risk of vertebrobasilar stroke. The currently advocated pre-manipulative risk assessment protocols include the use of provocative positional tests to challenge the integrity of the vascular supply to the brain. This paper critically evaluates the validity of these pre-manipulative provocative tests in the light of ultrasonographic blood ﬂow studies of the vertebral arteries. It also critically evaluates the evidence concerning the clinical utility of a portable continuous wave Doppler device (or velocimeter) to examine vertebral artery blood ﬂow prior to neck manipulation. There is clear evidence that the provocative tests may produce both false positive and false negative ﬁndings. Initial research regarding the clinical use of a velocimeter suggests this device may provide a more objective assessment of vertebral artery blood ﬂow than the provocative tests. However, the sensitivity, speciﬁcity and reliability of the use of the velocimeter in identifying abnormal vascular ﬂow in the vertebral arteries, and therefore its clinical utility, has not yet been fully established. r 2007 Elsevier Ltd. All rights reserved. Keywords: Cervical manipulation; Vertebral artery; Doppler ultrasonography; Cervical vertebrae

1. Introduction Cervical spine manipulation is a procedure commonly used by a number of health care professions, including physiotherapists, osteopaths, chiropractors and medical practitioners, primarily for the treatment of neck pain and headache. It is important for the clinician to recognise that while it may provide an effective means of treating neck pain and headache (Costello and Jull, 1999), cervical spine manipulation may result in adverse events and complications (Senstad et al., 1997; Di Fabio, 1999). Minor events are usually temporary, ranging from localised discomfort and stiffness to dizziness and nausea. More serious events, such as transient ischaemic attacks, stroke and even death Corresponding author. Tel.: +612 49217821; fax: +612 49217902.

E-mail address: [email protected] (L.C. Thomas). 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.11.003

have been reported subsequent to neck manipulation (Kleynhans, 1980; Klougart et al., 1996). Concerns about the more serious neurovascular complications have led both the Australian Physiotherapy Association (APA) and the Chartered Society of Physiotherapy (CSP) to develop and recommend guidelines to assist the clinician to identify patients in whom neck manipulation may be contraindicated (Grant, 1988; Barker et al., 2000; Magarey et al., 2000, 2004; Kerry et al., 2005; Rivett et al., 2006). The medical, osteopathic and chiropractic professions have adopted similar guidelines and protocols (International Federation of Manual Medicine, 1979; Kleynhans, 1980; George et al., 1981). In recent years the validity of such protocols has been the subject of much controversy, particularly with respect to the provocative test procedures (Reggars, 2005; Thiel and Rix, 2005). In response, ultrasound examination has lately been proposed as a more

ARTICLE IN PRESS 30

L.C. Thomas et al. / Manual Therapy 13 (2008) 29–36

objective method of pre-manipulatively assessing the integrity of blood supply to the hindbrain and the effects of cervical spine position on vertebral artery blood ﬂow (Rivett, 2001; Haynes, 2002). Despite promising preliminary research regarding the clinical utility of continuous wave ultrasound, there are a number of technical and operator issues that impact on its use and some limitations to this research. Thus, the aims of this paper are to review current premanipulative provocative testing procedures in the light of in vivo ultrasonographic ﬂow studies and to critically evaluate the utility of a handheld ultrasound device (the continuous wave Doppler velocimeter) which has recently been proposed by Haynes (2000, 2002) and others for use as a pre-manipulative clinical tool to assess vertebral artery blood ﬂow.

2. Serious adverse events following neck manipulation The more devastating complications associated with cervical spine manipulation are thought to arise from changes in blood ﬂow and/or the integrity of vascular structures of the neck (Terrett, 1987). It has been shown that neck rotation, common to many manipulative procedures, may alter blood ﬂow in the vertebral arteries in some individuals (Toole and Tucker, 1960; Selecki, 1969). It has also been argued that mechanical insult to one or more of the arteries in the neck during the manipulative procedure may account for some of the neurovascular events (Terrett, 1987). Despite a recent biomechanical study which suggests that neck manipulation is unlikely to produce sufﬁcient force to cause mechanical vertebral artery trauma (Symons et al., 2002), the most commonly accepted mechanism of injury is that of intimal dissection of the vertebral and/ or basilar arteries (Frisoni and Anzola, 1991). It has been proposed that evaluation of vertebral artery blood ﬂow status during cervical spine rotation may provide an indirect indication of the biomechanical stress placed on the vertebral artery by this position, and which the thrust component of the manipulation might be expected to increase (Haynes, 2004).

history, physical examination of the cervical spine, and provocative tests of the vertebral arteries (see for example (Terrett, 1987; Magarey et al., 2000; Magarey et al., 2004). The provocative test component of the protocol is intended to challenge the vascular supply to the brain and almost all of these tests involve rotation of the head and neck (Fig. 1). The presence of signs or symptoms of cerebrovascular ischaemia during or soon after this procedure is presumptive evidence of compromised vascular supply to the brain (International Federation of Manual Medicine, 1979; Kleynhans, 1980; George et al., 1981; Magarey et al., 2000, 2004). The rationale for the use of head and neck rotation as a provocative test comes from cadaveric (Toole and Tucker, 1960; Selecki, 1969), modelling (Haynes et al., 2002) and in vivo studies (Faris et al., 1963; Haynes and Milne, 2001), which show that head and neck rotation can reduce ﬂow in the vertebral artery contralateral to the direction of rotation. This position also simulates the head and neck posture of many manipulative procedures. It is important to recognise however, that the test procedure does not reproduce the mechanics associated with the manipulative thrust itself, which may also affect blood ﬂow volume (Licht et al., 1999). If signs or symptoms of cerebrovascular ischaemia become evident during the provocative manœuver, it is assumed that the manoeuvre has temporarily ‘provoked’ a neurovascular event, which is a contraindication for neck manipulation in itself and potentially a medical emergency. It is important to note that some signs and symptoms of cerebrovascular ischaemia, such as dizziness may also arise from non-ischaemic pathologies, for example, vestibular disorders. Arguably the provocative tests may produce a transient cerebrovascular ischaemic event because the manoeuvre has temporarily compromised blood ﬂow to

3. Pre-manipulative risk assessment protocols and provocative tests In principle, the risk assessment protocols aim to determine if the patient is presenting with a neurovascular event, has any pre-existing or current signs or symptoms that suggest they may be at risk of stroke or, more particularly, if the patient is at risk of a cerebrovascular event during a manipulative or other manual procedure to the cervical spine. Assessment protocols typically consist of three components: case

Fig. 1. Provocative testing using combined end-range rotation and extension of the cervical spine.

ARTICLE IN PRESS L.C. Thomas et al. / Manual Therapy 13 (2008) 29–36

31

4. Assessing blood ﬂow in vivo and the Doppler velocimeter

scanned. A digital display may provide graphical information concerning blood ﬂow frequency (i.e. velocity) and direction, with respect to the probe’s position. In some models, quantitative values of blood ﬂow, including the peak and mean velocities, are also provided. However, in contrast to duplex ultrasound, the velocimeter has no real-time imaging capabilities to assist in the location of vessels. Successful scanning of a vertebral artery results in a characteristic audible pulsatile signal which helps distinguish it from veins and other arteries in the area (Haynes et al., 2000; Zwiebel, 2000). The velocimeter has an advantage over duplex ultrasound of being portable and equipped with a small probe which can easily be applied to the upper cervical and sub-occipital regions (Fig. 2), as well as being relatively inexpensive and simple to operate. Potentially a velocimeter examination could be performed and the results interpreted by the manipulative practitioner in a typical clinical environment and in a timeframe that would facilitate appropriate referral and/ or treatment without delay. Haynes (1996) has described a protocol for ultrasonographic examination of the vertebral artery at the suboccipital and C1–C2 levels. More recently, Haynes (2002) and Rivett (2001) have proposed that the velocimeter could be used in pre-manipulative screening

Ultrasonographic studies suggest provocative tests have low sensitivity and speciﬁcity for detecting reduced vertebral artery ﬂow. However, they also suggest that duplex ultrasound may offer a more valid method of assessing blood ﬂow in the vertebral arteries prior to neck manipulation (Arnetoli et al., 1989; Haynes, 1996; Licht et al., 1998; Rivett et al., 2000). Duplex ultrasound provides a direct method of both visualising the boundaries of the cervical vessels and measuring blood ﬂow dynamics. It also has the advantage of being noninvasive and relatively quick to perform. However, duplex ultrasound is not a practical option for use in the typical private clinical practice of a manipulative therapist, as it is a procedure requiring considerable specialised training and costly equipment. Instead, Haynes (1996) and Rivett (2001) have proposed that a not too dissimilar device, the Doppler velocimeter, may be a more suitable tool to assess the status of vertebral artery blood ﬂow during pre-manipulative risk assessment protocols. The Doppler velocimeter is a basic continuous wave Doppler ultrasound device, currently used in medical applications for peripheral arterial and venous ﬂow studies for the detection of stenoses, reﬂux and alterations in ﬂow direction (Huntleigh, 2002). It has a transducer probe the size of a pen attached to the main handheld unit. The device produces an audible signal that varies in pitch to reﬂect blood ﬂow changes, as a result of varying resistance to ﬂow in the vessel being

Fig. 2. Examination of the right vertebral artery in contralateral cervical spine rotation using the velocimeter at the sub-occipital level.

the hindbrain. Alternatively, the head and neck movement may have initiated a neurovascular event due to the release of thrombo-embolic material or changed the vascular integrity sufﬁciently to produce a neurovascular event. If signs or symptoms of cerebrovascular ischaemia are not evident during or immediately subsequent to the provocative test procedure, it may be concluded that the vascular supply to the brain is sufﬁcient in these circumstances to maintain adequate perfusion. It is important to be aware that the provocative tests, in and of themselves, cannot identify anatomical anomalies or pathologies that may be associated with a cerebrovascular ischaemic event. Such information can only be obtained with specialised imaging techniques such as digital subtraction or magnetic resonance angiography. In particular, over the last decade or so duplex ultrasound has gained increasing acceptance as a valid tool for determining some vascular pathologies and for measuring vertebral artery blood ﬂow (Freed et al., 1998). It has the advantage of being considerably less hazardous than angiography.

ARTICLE IN PRESS 32

L.C. Thomas et al. / Manual Therapy 13 (2008) 29–36

to determine the status of vertebral artery blood ﬂow at rest, and to examine ﬂow velocity changes when the head and neck have been rotated. As changes in ﬂow velocity may occur with vessel stenosis and can be identiﬁed by changes in pitch of the audible signal (or amplitude of the digital display), Haynes (2000) has proposed that a change in audible signal may give an indication of the biomechanical forces imparted to the artery by cervical spine positioning. Such changes could be considered when assessing the risk of vertebral artery compromise associated with neck manipulation. Although all this remains to be demonstrated, it is not unreasonable to assume that absence of ﬂow, or marked alteration in ﬂow during head and neck rotation would be evidence for the need for a more comprehensive vascular evaluation of the patient prior to undertaking manipulative therapy (Rivett et al., 2005). Haynes (1996) initially used the velocimeter in a study to determine the effects of neck movement on vertebral artery blood ﬂow velocity. He examined the vertebral arteries of 140 patients (280 arteries) and found complete cessation of the Doppler signal on contralateral passive neck rotation in 5% of patients, indicating there was no blood ﬂow in those arteries. Of particular note was that there were no symptoms of vertebrobasilar ischaemia elicited during pre-manipulative provocative testing of these same persons, suggesting the presence of adequate collateral circulation. Haynes proposed that the velocimeter might be useful in detecting altered vascular integrity in the vertebral arteries which is currently not identiﬁable by the provocative tests recommended in pre-manipulative risk assessment protocols (Kleynhans, 1980; George et al., 1981; Grant, 1988; Magarey et al., 2000, 2004). However, in Haynes’ study, only one patient’s velocimeter results were veriﬁed by follow-up duplex scanning.

5. Merit of a clinical test and validity of the provocative tests Ideally, a diagnostic test should be able to demonstrate acceptable levels of sensitivity, speciﬁcity and reliability, as well as demonstrate clinical utility (Sackett et al., 1991). The ability of a test to detect a clinical condition when it is really present, that is, the true positives, is its sensitivity. A sensitivity of 100% means the test will always ﬁnd those with the condition. In clinical terms this is a rare scenario, but the closer to 100%, the more accurate the test is considered. In contrast, speciﬁcity is the ability of the test to identify people who do not have the condition, that is, the true negatives, and is also expressed as a percentage (Sackett et al., 1991; Davidson, 2002). Reliability or repeatability is the ability of the test to give consistent ﬁndings each

time it is performed by the same operator (intra-rater reliability) and between a number of operators (interrater reliability) when measuring the same variable. Reliability is commonly scored by the kappa (k) statistic, the closer the value is to 1, the closer the agreement between measurements. The k statistic is used for nominal data, while continuous data may be scored using intra-class correlation coefﬁcients (ICCs) (Domholdt, 2000). The clinical utility or usefulness of a diagnostic test can be determined by its ability to satisfy certain criteria. These include how the test compares with a gold or criterion standard. That is, whether it actually predicts risk and whether the patient will be better off, in terms of better clinical outcome, for having had the test performed (Sackett et al., 1991). The initial physiotherapy pre-manipulative testing protocol and more recent guidelines advocated by the APA (Grant, 1988; Magarey et al., 2000) have been subjected to critical review in recent years, particularly with respect to questions being raised about the validity of the provocative tests to detect vertebrobasilar insufﬁciency (see for example (Bolton et al., 1989; Dunne, 2001; Gross and Kay, 2001; Mann and Refshauge, 2001; Rivett, 2001). Furthermore, the provocative tests provide only an indirect assessment of blood ﬂow dynamics. Thiel and Rix (2005) have challenged the utility of the provocative tests in clinical scenarios involving vertebral artery dissection, spasm or stenosis. The possibility of false negative results was ﬁrst raised by Bolton et al. (1989) in a single case report involving the use of digital subtraction angiography. More recently, duplex ultrasound studies comparing vertebral artery blood ﬂow changes in subjects with positive and negative pre-manipulative test ﬁndings have led to concerns about the validity and appropriate interpretation of the provocative tests. False positive provocative test results have been suggested in some studies (Thiel et al., 1994; Licht et al., 1998), in which individuals with putative signs or symptoms of vertebrobasilar insufﬁciency on provocative testing were found to have normal vertebral artery blood ﬂow when examined in the same positions. Cote et al. (1996) in a secondary analysis of Thiel et al.’s (1994) study calculated a sensitivity of 0% and speciﬁcity between 67% and 90% for the provocative test used. Others have reported both false positive and false negative ﬁndings (Rivett, 2000). Rivett et al. (2000) found there were no meaningful signiﬁcant differences in blood ﬂow in various head positions, when comparing patients exhibiting positive and those with negative responses to pre-manipulative testing. Eight of the 200 vertebral arteries (4%) ultrasonographically examined in this study (Rivett, 2000) demonstrated total occlusion at either end-range or 451 contralateral rotation but no symptoms of vertebrobasilar insufﬁciency were exhibited. The authors concluded that collateral blood supply

ARTICLE IN PRESS L.C. Thomas et al. / Manual Therapy 13 (2008) 29–36

is usually sufﬁcient to maintain constant ﬂow to the hindbrain even if one vertebral artery is compromised. However, it could be argued that those individuals who show marked changes in blood ﬂow on ultrasound evaluation may be at risk of a neurovascular event following neck manipulation due to the additional stress of the manipulative thrust on an already compromised artery. Nevertheless, the ability to be able to detect abnormal vertebral artery blood ﬂow could assist in identifying a putatively ‘at risk’ patient and provide a more timely referral to specialist care.

6. Validity of continuous wave Doppler ultrasound Haynes (2002) advocated the merit of the Doppler velocimeter for vertebral artery assessment on the basis of a number of studies comparing continuous wave ultrasound to the criterion or gold standard of angiography (Kaneda et al., 1977; Hennerici et al., 1981; Ringelstein and Zeumer, 1985; White, 1986; Karnik et al., 1987; De Bray et al., 1997; Amadori et al., 1998). These studies showed high sensitivity values (between 85% and 91%) for the detection of marked stenosis (460%) in the vertebral arteries, and high speciﬁcity values (between 94% and 100%). There are, however, some problems with extrapolating the ﬁndings of these studies to both the velocimeter and the type of patients in whom neck manipulation might be considered. The continuous wave ultrasound machines used in these studies supplied more quantitative data than a velocimeter can normally provide. Furthermore, the reported high sensitivity and speciﬁcity values may have largely been due to the fact that many of the patients fell into groups with severe vascular disease whose history alone would have been a contraindication for cervical manipulation. In a review of early continuous wave ultrasound studies, White (1986) suggests that the high speciﬁcity levels obtained could have been subject to bias due to the rare incidence of the conditions being investigated, thus giving a high probability of an examiner ‘guessing’ a negative ﬁnding and being correct. Even in these advanced vascular disease groups, continuous wave Doppler could only detect those individuals with severe occlusion (460%) of the vertebral arteries. Furthermore in these early studies, comparison between ultrasound and angiography was generally unblinded and uncontrolled.

7. Limitations of Doppler ultrasound examination of vertebral artery ﬂow While Doppler ultrasound has advantages in terms of patient comfort, non-invasiveness and relative time of performance in comparison to angiography, there are a

33

number of potential problems associated with its use. These include high operator dependence, poor reliability and difﬁculties with localisation of the target vessel. Reliability can be improved if a number of blood ﬂow measurements are taken and averaged (Haynes et al., 2000). Examination of the vertebral artery is particularly subject to operator error (Zwiebel, 2000) because it is a small vessel, with a mean diameter of 4 mm (Haynes, 2000), and can thus be difﬁcult to detect without realtime imaging. The velocimeter utilises continuous wave Doppler ultrasound but does not have imaging capabilities and relies solely on the accurate identiﬁcation of a characteristic audio signal for the artery. Identiﬁcation errors can occur due to signals being picked up from more superﬁcial vessels rather than the vertebral artery itself. When using the velocimeter it is also not possible to determine the Doppler angle accurately or direct the Doppler beam to sample from the centre of the vessel, both factors of which are desirable to minimise error. The velocimeter technique therefore necessitates adjustment to the position of the probe once the Doppler signal is identiﬁed to obtain the clearest possible signal in each neck position. There is the potential for large variations in the audio signal output due to slight movement of the probe, particularly during neck rotation. The use of too much pressure with the probe may compress the artery and if insufﬁcient coupling gel is used there may be poor conduction through the skin. Technical difﬁculties such as these may signiﬁcantly distort the signal and be mistaken for alterations in blood ﬂow velocity related to changes in neck position (Haynes, 2000). The examination technique of the operator is therefore an important factor in obtaining meaningful blood ﬂow information. Furthermore, in contrast with duplex ultrasound, the velocimeter does not provide quantitative information on which diagnostic decisions can be based. Interpretation of ﬁndings is generally based on the qualitative information obtained from the auditory signal. While the human ear is considered to be highly sensitive in its ability to distinguish quite subtle alterations in pitch (Zwiebel, 2000), analysis of an auditory signal is subjective and dependent on the experience of the operator. A further consideration is that patients presenting for examination in primary contact practice with marked vertebral artery blood ﬂow changes are relatively uncommon, probably less than 5% of this population (Haynes et al., 2000), and some authors suggest that ﬂow changes on neck rotation may represent a normal variant and not be clinically important (Zaina et al., 2003). This presents some difﬁculty in determining the accurate interpretation of velocimeter examination to distinguish between normal variation of vertebral artery blood ﬂow in neutral and during contralateral neck rotation and poor technique or true abnormal blood ﬂow.

ARTICLE IN PRESS 34

L.C. Thomas et al. / Manual Therapy 13 (2008) 29–36

8. Validity and reliability of the velocimeter Notwithstanding such limitations, Haynes (2000) reported surprisingly high sensitivity and speciﬁcity values of 100% each, when comparing velocimeter ﬁndings to those of duplex ultrasound, which he considered the criterion standard. Haynes (2000) examined 20 patients presenting to a private chiropractic practice with the velocimeter directed at the suboccipital and C2 levels of the vertebral artery. He categorised the blood ﬂow signals as either unchanged, major decrease in ﬂow, or major increase in ﬂow during head and neck rotation. The patients were then examined with a duplex ultrasound scanner at C3–C4 and C1–C2 levels, and peak systolic velocity measured. The study found good concordance (k ¼ 1.00 at p ¼ 0.01) between velocimeter and duplex ultrasound ﬁndings in respect of increased and decreased blood ﬂow velocity, despite the duplex ultrasound and velocimeter examinations being performed at different levels of the vertebral artery. The concordance was interpreted by the investigators as indicating that the velocimeter is a valid tool for detecting marked changes in vertebral artery ﬂow in the hands of an experienced operator. However, it should be noted that all the patients were pre-selected by velocimeter examination, and not the criterion standard (duplex ultrasound). To investigate reliability, Haynes et al. (2000) compared the results of vertebral artery velocimeter examinations between an experienced and an inexperienced operator. A naı¨ ve operator was trained in a 2-h session in the technique of insonating the vertebral artery at the sub-occipital and C2 levels. Twenty patients attending a chiropractic practice served as subjects for the study and were examined by the experienced operator and then by the inexperienced operator who was blinded to the ﬁndings of the initial examination. The study showed good inter-rater reliability (k ¼ 0.78 at p ¼ 0.05), which suggests that the device could be successfully used by a novice following a short training session. However, a particular limitation of both these velocimeter studies (Haynes, 2000; Haynes et al., 2000) is that subjects were pre-selected by velocimeter examination by a highly experienced examiner for comparative assessment. Pre-selection of subjects on the basis of velocimeter examination may have meant that the full range of possible blood ﬂow changes were not represented and arguably those who were selected for the study may have been those with more profound changes in blood ﬂow on rotation. While these studies suggest that some people with altered blood ﬂow in the vertebral arteries during head and neck rotation may be identiﬁed with the velocimeter, the sensitivity and speciﬁcity of the use of the velocimeter to identify the full range of patients who

present to manipulative therapists, remain to be established. Further research should address this issue and use duplex ultrasound examination of a group of neck pain patients to determine what degree of change in vertebral artery blood ﬂow between neutral and endrange contralateral neck rotation can actually be detected by velocimeter examination. While the velocimeter appears promising as a premanipulative risk assessment tool, the limitations identiﬁed need to be considered and it remains to be shown if it will provide more objective information than is currently available to the manipulative therapist in clinical practice. It is important to recognise that the velocimeter, like the currently used provocative tests, will not predict cerebrovascular events due to inappropriate or poor performance of neck manipulation, or provide detailed diagnostic information. Nevertheless, if with further research the sensitivity, speciﬁcity and reliability of the use of the velocimeter is demonstrated to be clinically acceptable, it may yet serve as a clinically useful instrument to examine the vascular integrity of the vertebral artery and its blood ﬂow response to head and neck positioning in patients being considered for neck manipulation.

9. Conclusion In the light of recent studies showing poor correlation between vertebral artery blood ﬂow changes and signs and symptoms of vertebrobasilar ischaemia during provocative testing, it would seem prudent to search for more objective means of evaluating blood ﬂow during pre-manipulative risk assessment. Doppler ultrasound in the form of a handheld velocimeter may provide a more objective screening tool for manipulative therapists, although this has not yet been clearly demonstrated. If the velocimeter is found to be a valid and reliable means of detecting vertebral artery blood ﬂow changes it would give manual therapy practitioners greater ability to identify patients presenting with a putative risk factor for a vertebrobasilar incident. Such people could then be referred to a neurovascular clinic for more deﬁnitive examination. References Amadori A, Arnetoli G, Nuzzaci G, Stephani P. Continuous wave Doppler of vertebral arteries in non-invasive diagnosis and management of vertebrobasilar TIAs. Angiology 1998;39:365–70. Arnetoli G, Amadori A, Stephani P, Nuzzaci G. Sonography of vertebral arteries in De Kleyn’s position in subjects and in patients with vertebrobasilar transient ischemic attacks. Angiology 1989;40: 716–20. Barker S, Kesson J, Turner G, Conway J, Stevens D. Guidance for pre-manipulative testing of the cervical spine. Manual Therapy 2000;5(1):37–40.

ARTICLE IN PRESS L.C. Thomas et al. / Manual Therapy 13 (2008) 29–36 Bolton PS, Stick PE, Lord RSA. Failure of clinical tests to predict cerebral ischemia before neck manipulation. Journal of Manipulative and Physiological Therapeutics 1989;12(4):304–7. Costello J, Jull G. Neck position statement. Australian Physiotherapy Association: Melbourne; 1999. pp. 1–30. Cote P, Kreitz BG, Cassidy D, Thiel H. The validity of the extensionrotation test as a clinical screening procedure before neck manipulation: a secondary analysis. Journal of Manipulative and Physiological Therapeutics 1996;19(3):159–64. Davidson M. The interpretation of diagnostic tests: a primer for physiotherapists. Australian Journal of Physiotherapy 2002;48: 227–32. De Bray JM, Penisson-Besnier I, Dubas E, Emille J. Extracranial and intracranial vertebrobasilar dissections; diagnosis and prognosis. Journal of Neurology, Neurosugery and Psychiatry 1997;63: 46–51. Di Fabio RP. Manipulation of the cervical spine: risks and beneﬁts. Physical Therapy 1999;79:50–65. Domholdt E. Physical therapy research: principles and applications. Philadelphia, PA: W.B Saunders Company; 2000. Dunne J. Pre-manipulative testing: predicting risk or pretending to? Australian Journal of Physiotherapy 2001;47:165. Faris A, Poser CM, Wilmore DW, Agnew CH. Radiologic evaluation of neck vessels in healthy men. Neurology 1963;13:386–96. Freed KS, Brown LB, Carroll MD. The extracranial cerebral vessels. In: Rumack CM, Wilson SR, Charboneau JW, editors. Diagnostic ultrasound, vol. 1. St. Louis, MO: Mosby Year Book Inc.; 1998. p. 885–919. Frisoni G, Anzola G. Vertebrobasilar ischaemia after neck motion. Stroke 1991;22:1452–60. George PE, Silverstein HT, Wallace H, Marshall M. Identiﬁcation of the high risk pre-stroke patient. Journal of Chiropractic 1981; 1(Suppl):26–8. Grant R. Dizziness testing and manipulation of the cervical spine. In: Grant R, editor. Clinics in physical therapy: physical therapy of the cervical and thoracic spine, vol. 17. New York: Churchill Livingstone; 1988. p. 111–24. Gross AR, Kay TM. Guidelines for pre-manipulative testing of the cervical spine—an appraisal. Australian Journal of Physiotherapy 2001;47:166–7. Haynes MJ. Doppler studies comparing the effects of cervical rotation and lateral ﬂexion on vertebral artery blood ﬂow. Journal of Manipulative and Physiological Therapeutics 1996;19:378–84. Haynes M. Vertebral arteries and neck rotation: Doppler velocimeter and duplex results compared. Ultrasound in Medicine and Biology 2000;26(1):57–62. Haynes M. Vertebral arteries and cervical movement: Doppler ultrasound velocimetry for screening before manipulation. Journal of Manipulative and Physiological Therapeutics 2002;25(9):556–67. Haynes M. Internal forces sustained by the vertebral artery during spinal manipulative therapy: letter to the, editor. Journal of Manipulative and Physiological Therapeutics 2004;27(1):67–8. Haynes MJ, Milne N. Color duplex sonographic ﬁndings in human vertebral arteries during cervical rotation. Journal of Clinical Ultrasound 2001;29:14–24. Haynes M, Hart R, McGeachie J. Vertebral arteries and neck rotation: Doppler velocimeter interexaminer reliability. Ultrasound in Medicine and Biology 2000;26(8):1363–7. Haynes M, Cala LA, Melsom A, Mastaglia F, Milne N, McGeachie J. Vertebral arteries and cervical rotation: modelling and magnetic resonance angiography studies. Journal of Manipulative and Physiological Therapeutics 2002;25:370–83. Hennerici M, Aulich A, Sandmann W, Freud H. Incidence of asymptomatic extracranial arterial disease. Stroke 1981;12(6): 750–8. Huntleigh. Huntleigh Diagnostics. Perth, WA; 2002.

35

International Federation of Manual Medicine. On the prevention of accidents arising from manipulative therapy of the cervical spine. Declaration of the presidium of the German Association of Manual Medicine, 1979. Kaneda H, Irino T, Minami T, Taneda M. Diagnostic reliability of the percutaneous ultrasonic Doppler technique for vertebral artery occlusive disease. Stroke 1977;8(5):571–9. Karnik R, Sto¨llberger C, Ammerer H, Perneczky G, Slany J, Brennner H. Validity of the continuous-wave sonography of the vertebrobasilar system. Angiology 1987;38:556–61. Kerry R, Taylor A, Mitchell JA, Brew J, Kiely R, McCarthy C. Cervical artery dysfunction assessment framework. Manipulative Association of Chartered Physiotherapists, 2005. Kleynhans AM. Complications and contraindications to spinal manipulative therapy. In: Haldeman S, editor. Modern developments in the principle and practice of chiropractic. New York: Apple-Century-Crofts; 1980. p. 359–82. Klougart N, Leboeuf-Yde C, Ramsmussen RL. Safety in chiropractic practice. Part 1: the occurrence of cerebrovascular accidents after manipulation to the neck in Denmark from 1978–1988. Journal of Manipulative and Physiological Therapeutics 1996;19:371–7. Licht PB, Christensen HW, Hojgaard P, Hiolund-Carlson PF. Triplex ultrasound of vertebral artery ﬂow during cervical rotation. Journal of Manipulative and Physiological Therapeutics 1998; 21:27–31. Licht PB, Christensen HW, Svendsen P, Hoilund-Carlsen PF. Vertebral artery ﬂow and cervical manipulation: an experimental study. Journal of Manipulative and Physiological Therapeutics 1999;22(7):431–5. Magarey M, Coughlan B, Rebbeck T. Clinical guidelines for premanipulative procedures for the cervical spine. Australian Physiotherapy Association and Musculoskeletal Physiotherapy Australia 2000:1–7. Magarey M, Rebbeck T, Coughlan B, Grimmer K, Rivett D, Refshauge KM. Pre-manipulative testing of the cervical spine review, revision and new clinical guidelines. Manual Therapy 2004; 9:95–108. Mann T, Refshauge KM. Causes of complications from cervical manipulation. Australian Journal of Physiotherapy 2001;47: 255–66. Reggars J. Pre-manipulative VBI testing worthless. Chiropractic and Osteopathic College of Australasia Newsletter (COCA) 2005 June. Ringelstein EB, Zeumer HKP. Non-invasive diagnosis of intracranial lesions in the vertebrobasilar system. A comparison of Doppler sonographic and angiographic ﬁndings. Stroke 1985;16(5): 848–55. Rivett D. Vertebral artery blood ﬂow during pre-manipulative testing of the cervical spine. In: School of physiotherapy. Dunedin: University of Otago; 2000. p. 321. Rivett D. A valid pre-manipulative screening tool is needed. Australian Journal of Physiotherapy 2001;47:165. Rivett D, Sharples K, Milburn P. Vertebral artery blood ﬂow during pre-manipulative testing of the cervical spine. In: The 7th scientiﬁc conference of the International Federation of Orthopaedic and Manipulative Therapists, Perth, Australia, 2000. Rivett D, Thomas L, Bolton PS. Pre-manipulative testing: where do we go from here? New Zealand Journal of Physiotherapy 2005;33(3): 78–84. Rivett D, Shirley D, Magarey M, Refshauge KM. APA clinical guidelines for assessing vertebrobasilar insufﬁciency in the management of cervical spine disorders—2006. Melbourne: Australian Physiotherapy Association, Muculo-skeletal Physiotherapy Australia; 2006. Sackett DL, Haynes RB, Guyatt GH, Tugwell P. Clinical epidemiology a basic science for clinical medicine. Boston: Little, Brown and Company; 1991.

ARTICLE IN PRESS 36

L.C. Thomas et al. / Manual Therapy 13 (2008) 29–36

Selecki BR. The effects of rotation of the atlas on the axis: experimental work. Medical Journal of Australia 1969;56:1012–5. Senstad O, Leboeuf-Yde C, Borchgrevink C. Frequency and characteristics of side effects of spinal manipulative therapy. Spine 1997;22(4):435–41. Symons B, Leonard T, Herzog W. Internal forces sustained by the vertebral artery during spinal manipulative therapy. Journal of Manipulative and Physiological Therapeutics 2002;25(8):504. Terrett AGJ. Vascular accidents from cervical spine manipulation: report on 107 cases. Journal of the Australian Chiropractic Association 1987;17(4):75–82. Thiel H, Rix G. Is it time to stop functional pre-manipulative testing of the cervical spine? Manual Therapy 2005;10:154–8.

Thiel H, Wallace K, Donat J, Yong-Hing K. Effect of various head and neck positions on vertebral artery blood ﬂow. Clinical Biomechanics 1994;9:105–10. Toole JF, Tucker SH. Inﬂuence of head position upon cerebral circulation. Archives of Neurology 1960;2:616–23. White DN. Vertebral ultrasonography. In: Zwiebel WJ, editor. Introduction to vascular ultrasonography. Philadelphia: W.B. Saunders Company; 1986. p. 217–43. Zaina C, Grant R, Johnson C, Dansie B, Taylor J, Spyropolous P. The effect of cervical rotation on blood ﬂow in the contralateral artery. Manual Therapy 2003;8(2):103–9. Zwiebel WJ. Introduction to vascular ultrasonography. Philadelphia, PA: Harcourt Health Sciences; 2000.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 37–42 www.elsevier.com/locate/math

The initial effects of a Mulligan’s mobilization with movement technique on range of movement and pressure pain threshold in pain-limited shoulders Pamela Teys, Leanne Bisset, Bill Vicenzino Division of Physiotherapy, School of Health and Rehabilitation Sciences, Therapies Building 84A, The University of Queensland, St Lucia Qld 4072, Australia Received 15 December 2005; received in revised form 10 May 2006; accepted 21 July 2006

Abstract There is little known about the speciﬁc manual therapy techniques used to treat painfully limited shoulders and their effects on range of movement (ROM) and pressure pain threshold (PPT). The objective of this study was to investigate the initial effects of a Mulligan’s mobilization with movement (MWM) technique on shoulder ROM in the plane of the scapula and PPT in participants with anterior shoulder pain. A repeated measures, double-blind randomized-controlled trial with a crossover design was conducted with 24 subjects (11 males and 13 females). ROM and PPT were measured before and after the application of MWM, sham and control conditions. Signiﬁcant and clinically meaningful improvements in both ROM (15.3%, F (2,46) ¼ 16.31 P ¼ 0:00) and PPT (20.2%, F ð2; 46Þ ¼ 3:44, P ¼ 0:04) occurred immediately after post treatment. The results indicate that this speciﬁc manual therapy treatment has an immediate positive effect on both ROM and pain in subjects with painful limitation of shoulder movement. Further study is needed to evaluate the duration of such effects and the mechanism by which this occurs. r 2006 Elsevier Ltd. All rights reserved. Keywords: Shoulder pain; MWM; ROM; Pain

1. Introduction Shoulder pain with a subsequent restriction of movement is a common problem in both the sporting and working population. Approximately 1% of adults consult a general medical practitioner with an episode of shoulder pain each year (Bridges-Wegg, 1992; Pope et al., 1997). There is a dearth of high-quality trials that support or refute the use of physiotherapy in shoulder pain (Green et al., 2004), but there is some support for individualized programmes of manual therapy and exercises in the treatment of shoulder impingement syndrome (Michener et al., 2004). Two trials conducted by Bang Corresponding author. Tel.: +617 33652781; fax: +617 33652775.

E-mail address: [email protected] (B. Vicenzino). 1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.07.011

and Deyle (2000) and Nicholson (1985), which rated six and ﬁve out of ten, respectively, on the PEDro quality rating scale (www.pedro.fhs.usyd.edu.au), reported that supervised exercise combined with manual therapy was better than supervised exercise alone in the treatment of shoulder impingement. Mobilization with movement (MWM) is a class of manual therapy techniques that is widely used in the management of musculoskeletal pain. It involves the manual application of a sustained glide by a therapist to a joint while a concurrent movement of the joint is actively performed by the patient (Mulligan, 1999). Studies using MWM techniques on the elbow and ankle have shown them to be effective in reducing pain as measured by visual analogue scale (VAS) and pressure pain threshold (PPT) and increasing joint range of movement (ROM) (Vicenzino and Wright, 1995;

ARTICLE IN PRESS 38

P. Teys et al. / Manual Therapy 13 (2008) 37–42

O’Brien and Vicenzino, 1998; Chen et al., 1999; Abbott, 2001; Abbott et al., 2001; Paungmali et al., 2003a; Collins et al., 2004). During shoulder movement in participants with no pathology the humeral head remains relatively centered in the glenoid, predominantly through small translatory glides in the glenoid (Harryman et al., 1990). Earlier studies have identiﬁed that altered shoulder kinematics are associated with shoulder pain (Howell et al., 1988; Ludewig and Cook, 2000, 2002; Halder et al., 2001). Kinematic studies of patients with impingement, rotator cuff tears, loss of capsuloligamentous integrity or neuromuscular fatigue, have demonstrated abnormal or excessive superior and/or anterior translation of the humeral head in the glenoid fossa (Fu et al., 1991; Kamkar and Irrgang, 1993; Flatow et al., 1994). It would appear that excessive translation of the humeral head along the glenoid results in pain and functional impairment (Matsen et al., 1993). It has been suggested that the application of a posterior glide MWM to the shoulder may correct this fault and allow optimal pain-free motion to occur (Mulligan, 1999). Hsu et al. (2000) in a study of 11 cadavers, found the application of an anterior– posterior glide towards the end of range of abduction was effective in improving the range of glenohumeral abduction. To date, no studies have investigated the effects of the MWM in people with shoulder pain and reduced ROM. The aim of our study was to evaluate the effect of a MWM on shoulder ROM and PPT.

2. Methods A repeated measures, crossover, double-blinded randomized, placebo-controlled trial was conducted to evaluate the initial effects of a shoulder MWM on ROM and PPT. This design was used to reduce the effects of individual variation and strengthen internal validity. 2.1. Participants Twenty-four participants (11 males and 13 females) aged between 20 and 64 years (mean 46.1 years SD+9.86) were recruited from the general population in southeast Queensland. The primary inclusion criterion was the inability to elevate the arm greater than 1001 in the plane of the scapula because of the presence of pain over the anterior aspect of either shoulder. The duration of the pain had to be greater than one month to ensure that there was an established shoulder condition and for less than one year so as to limit the study population to those whose pain was not likely to be a result of such conditions as recalcitrant frozen shoulder. The main exclusion criterion was shoulder pain that was deemed not to be musculoskeletal in origin. Other exclusion criteria were any medical

condition that would exclude the patient from physiotherapy treatment, active inﬂammatory disease, infection, cancer, neuromuscular disorders and fractures around the shoulder. The participants were also screened for involvement of the cervical spine that may have contributed to the shoulder condition and excluded if there was evidence of cervical spine referral of pain to the shoulder. A physiotherapist who holds a post-graduate Sports physiotherapy degree and has greater than 15 years clinical experience performed all screening examinations. Ethical clearance was obtained from the University of Queensland’s Medical Research Ethics Committee and signed informed consent was gained from all participants prior to their inclusion in the study. 2.2. Outcome measures (dependent variables) The outcome measures were taken by an investigator skilled in their application and who remained blind to the allotted treatment condition. The outcome measures used were range of glenohumeral elevation in the plane of the scapula and PPT over the anterior shoulder. 2.2.1. Pain-free range of movement in the scapular plane A universal goniometer was used to measure the ROM in the plane of the scapula. This has been shown to demonstrate good intra-tester reliability if consistent landmarks are used (Hayes et al., 2001). The plane of the scapula is deﬁned as 301 anterior to the coronal plane. This was calculated by aligning the axis of the goniometer along the superior aspect of the shoulder and moving one arm of the goniometer 301 forward from that frontal plane whilst the other arm of the goniometer remained in the frontal plane. The patient was then asked to move the affected arm in that plane through a small arc of movement short of pain, by aligning the arm movement to a vertical line drawn up the wall. The line on the wall was used to aid test–retest repeatability. Goniometric measurement of elevation in the plane of the scapula was achieved by aligning the centre of the goniometer with the centre of the glenohumeral joint, one arm of the instrument along the lateral border of the scapula and the other along the humerus in line with the lateral epicondyle aided by skin markers. A measure of active ROM was taken. The participant was asked to move the arm into elevation along the plane of the scapula just to the onset of pain and this process was repeated three times. This technique was in accordance with guidelines of goniometric measurement as outlined by Moore (Gerhardt, 1992). 2.2.2. Pressure pain threshold A quantitative measure of pain was obtained by the use of pressure pain algometry, which has demonstrated

ARTICLE IN PRESS P. Teys et al. / Manual Therapy 13 (2008) 37–42

good inter- and intra-rater correlation and reliability in other studies (Pontinen, 1998). The most sensitive point was located over the anterior aspect of the shoulder by manual palpation and marked with a permanent marker so that the same point could be used for pre- and post-condition application measures. As in previous work carried out in this laboratory, (Sterling et al., 2001; Paungmali et al., 2003a; Collins et al., 2004) pressure was applied via a digital pressure algometer (Somedic AB, Farsta, Sweden) applied perpendicular to the skin at a rate of 40 kPa/s through a rubber-tipped probe (area 1 cm2). The patient was instructed to activate a button as soon as a change of sensation from one of pressure to one of pain was experienced (threshold of pain). This process was repeated three times with a 30-s rest period between each measurement. 2.3. Experimental conditions (independent variables) There were two independent variables in the research design; treatment condition and time (pre-, postapplication). Treatment condition had three levels, which included the MWM, a sham and a control condition. A physiotherapist who was blind to the preand post-outcome measures (i.e. played no part in taking the outcome measures) applied all conditions. This physiotherapist held both musculoskeletal and sports post-graduate degrees with more than 10 years clinical experience. The treatment condition consisted of the application of a postero-lateral glide (MWM) to the affected shoulder (Fig. 1). The participant was seated and the therapist stood beside the participant on the opposite side to the affected shoulder. One hand was placed over the scapula posteriorly while the thenar eminence of the

Fig. 1. The MWM technique in which the therapist applies a posterolateral glide to the humeral head along the plane of the glenohumeral joint while stabilizing the scapula with the other hand.

39

other hand was placed over the anterior aspect of the head of the humerus. A posterior gliding force was applied to the humeral head. The participant was then asked to raise the affected arm in the plane of the scapula to the point of pain onset while the therapist sustained the gliding force to the humeral head, with care to avoid the sensitive coracoid process. Three sets of 10 repetitions were applied with a rest interval of 30 s between each set. The therapist endeavored to maintain the glide at right angles to the plane of movement throughout the entire range. The participant was instructed that the MWM procedure, including arm elevation, was to be pain free, and must be ceased immediately if any pain was experienced during the application (Mulligan, 1999; Exelby, 2002). The sham condition replicated the treatment condition except for the hand positioning. The therapist stood on the opposite side of the participant and placed one hand along the clavicle and sternum and the other on the posterior aspect of the humeral head of the affected shoulder. A simulated anterior glide was performed but with minimal pressure actually applied. The participant was asked to elevate the affected shoulder in the plane of the scapula through half of their available pain-free range to minimize the likelihood of pain provocation. The number of repetitions and sets were as per the treatment group. In the control condition the participant was seated for the same length of time but no manual contact between the therapist and the participant took place. 2.4. Procedure Participants were initially assessed for their suitability for inclusion in the study and underwent a physical screening of the affected shoulder and cervical spine by an experienced post-graduate Sports physiotherapist with more than 15 years of clinical experience. This session was also used to familiarize the participant with the testing procedures, laboratory environment and investigators. Participants attended three sessions at approximately the same time each day to prevent any diurnal variations in joint range and pain potentially confounding results and with at least an intervening 24 h interval to reduce the inﬂuence of any carry-over effect. Testing was conducted in a temperature and humidity controlled laboratory. The participants were requested to avoid factors that may inﬂuence their shoulder pain, such as analgesics and/or anti-inﬂammatory medication during the week of testing. At each experimental session, following the recording of baseline measures, each participant received one of the three treatment conditions (MWM, sham, control), in a randomized order known only to the treating therapist. The treatment allocation sequence was block

ARTICLE IN PRESS P. Teys et al. / Manual Therapy 13 (2008) 37–42

40

randomized using the drawing of lots and concealed from the investigator who took the outcome measurements. Following the application, outcome measures were again taken. Participant blinding was facilitated by recruitment of people who had no experience of the manipulative therapy techniques applied to the shoulder and by careful instruction that did not refer to the study’s aims of evaluation of a treatment technique. Subjects were informed that the study was investigating the effects of manual handling on shoulder pain. An exit questionnaire assessed the adequacy of patient blinding. Results of the exit questionnaire showed that three participants (12%) correctly guessed they had only received active treatment and none had correctly guessed that they had received either a sham or control.

3. Reliability Acceptable intra-rater reliability was determined through analysis of pre- to post-control measures of ROM and PPT. For this study the intra-class correlation coefﬁcient (ICC 2,1) and standard error of the measurement (SEM) for ROM were estimated to be 0.98 and 1.331 respectively. The ICC 2,1 and the SEM for PPT were estimated to be 0.96 and 10.7 kPa respectively. This indicates that both the size of the error (SEM) and the ICC are indicative of reliable measures.

4. Data management and analysis Two independent variables were incorporated into the research design: treatment (MWM, sham, control) and time (pre- and post-application). Dependent variables included ROM and PPT. Prior to analysis, the average of triplicate measures of ROM and PPT were calculated. A two-factor analysis of variance (ANOVA) and appropriate post-hoc tests of simple effects were then performed on each of the two dependent variables to

test the hypothesis that MWM produced changes in excess of sham and control from pre- to post-application (P ¼ 0:05).

5. Results 5.1. Range of movement There was a signiﬁcant Time by Condition interaction effect for ROM (F ð2;46Þ ¼ 16:3, P ¼ 0:000) with a signiﬁcant mean improvement of 161 (P ¼ 0:000) preto post-treatment after the application of the MWM compared with 4 1 (P ¼ 0:06) for the sham application and no change (P ¼ 0:84) for the control condition (Table 1). The mean differences between the MWM and Sham (101) and MWM and Control (111) were statistically different after application; Po:02; where they were not different at baseline. 5.2. Pressure pain threshold There was a signiﬁcant Time by Condition interaction for PPT (F ð2;46Þ ¼ 3:4, P ¼ 0:04), which demonstrated a mean improvement of 63 kPa following the application of the MWM (P ¼ 0:000) pre- to post-treatment application compared with 26 kPa (P ¼ 0:05) for the sham application and 20 kPa (P ¼ 0:07) for the Control application. The mean differences between the MWM and Sham (45 kPa; P ¼ 0:04 and between MWM and Control (46 kPa; P ¼ 0:02 were statistically signiﬁcant. There were no signiﬁcant differences pre-application. 5.3. Methodological considerations There was no loss to follow-up and no adverse effects reported. There was no carry-over effect when the preapplication data for all experiment sessions (i.e. before each intervention was applied) were evaluated.

Table 1 The mean (95% CI) for range of movement (ROM) in degrees and pressure pain threshold (PPT) in kPa for the mobilization with movement treatment technique (MWM), Sham (S) and Control (C). Also included are the mean differences (95% CI) between pre- and post-intervention, as well as the differences between MWM-S and MWM-C MWM

Mean (95% CI)

Condition mean differences (95% CI)

Sham

Control

MWM-S

MWM-C

ROM

Pre Post Diff

102.2 (94.5 to 109.9) 117.8 (110.2 to 125.5) 15.6 (10.1 to 21.1)

103.9 (96.4 to 111.5) 107.9 (98.7 to 117.1) 3.9 (0.1 to 7.9)

106.2 (96.9 to 115.5) 106.4 (96.7 to 116.2) 0.27 (2.4 to 3)

1.8 (8.7 to 5.2) 9.9 (4.3 to 15.6) N/A

3.9 (11.4 to 3.5) 11.4 (2.3 to 20.5) N/A

PPT

Pre Post Diff

310.8 (258.8 to 362.9) 373.4 (313.6 to 433.1) 62.6 (33.6 to 91.5)

302.5 (252.3 to 352.6) 328.3 (275.6 to 381.0) 25.9 (0.2 to 51.6)

307.1 (254.7 to 359.5) 327.1 (271.1 to 383.0) 20 (1.5 to 41.5)

8.4 (27.7 to 44.6) 45.1 (1.7 to 88.4) N/A

3.8 (36.7 to 44.2) 46.3 (9.1 to 83.6) N/A

Denotes a statistical signiﬁcant difference Po0:05.

ARTICLE IN PRESS P. Teys et al. / Manual Therapy 13 (2008) 37–42

6. Discussion This study demonstrated that the application of the Mulligan’s MWM technique to participants with a painful restriction of shoulder movement produced an immediate and signiﬁcant improvement in ROM and PPT pre- to post-intervention when compared to sham or control conditions. There are no other published studies of the effects of this technique on participants with shoulder pain. However, these ﬁndings are consistent with studies conducted in other joints of the body that have shown similar effects with the MWM techniques (O’Brien and Vicenzino, 1998; Abbott et al., 2001; Paungmali et al., 2003b; Collins et al., 2004). The clinical relevance of the magnitude of improvement in ROM gained following the MWM compared to the Sham (101) after only one treatment session is arguably comparable to 421 improvement in abduction following four sessions of intensive massage (van den Dolden and Roberts, 2003) and 221 improvement after 4–10 sessions of individualized shoulder treatment (mainly exercises) over a month (Ginn et al., 1997). Wright (1995) has postulated that the mechanisms responsible for manual therapy treatment effects (e.g. as in the increases in ROM and PPT in our study) may feasibly involve changes in the joint, muscle, pain and motor control systems. In our study the standardized mean difference (SMD) for ROM (1.2) was greater than the SMD for PPT (0.9). The change in ROM was not related to the change in PPT (Pearson’s correlation coefﬁcient R ¼ 0:29 P ¼ 0:17) possibly indicating that the underlying mechanisms of the MWM may be related to local joint or muscle structures rather than the pain system. The technical difference between the MWM and sham application was that the MWM involved the application of a postero-lateral joint glide while the patient performed an active movement compared with the sham that involved no glide. This data, when considered along with studies showing that forward translation of the humeral head painfully limits shoulder movement, (Ludewig and Cook, 2000) leads us to speculate that Mulligan’s proposed mechanism of action for MWM’s as a reduction of a positional fault may have some credence. The application of the shoulder MWM also resulted in small but positive changes in PPT pre- to postintervention. The mean differences between the MWM and the sham and the MWM and the control condition post-intervention were 45 kPa (95% CI: 2–88) and 45 kPa (95% CI: 9–84), respectively. Other studies of the upper limb have demonstrated similar effects in PPT following the application of a Mulligan’s MWM (Vicenzino et al., 2001, Paungmali et al., 2003b). These studies, along with others (Sterling et al., 2001; Paungmali et al., 2003a; Souvlis et al., 2005) have

41

proposed that manipulative therapy may provide sufﬁcient sensory input to activate the endogenous pain inhibitory systems. Further studies need to be conducted in the shoulder to determine if endogenous pain inhibitory systems are involved in manipulation-induced changes of PPT in the shoulder. The comparison between the MWM and Sham conditions should also take into account that the latter limited abduction to half available range: that is, some of the difference between MWM and Sham may be attributable to the MWM utilizing a greater range of abduction. Certainly, ethically it was undesirable to ask participants to experience repeated pain and pragmatically it is difﬁcult to ensure compliance with return visits to the experiment if the subject was experiencing repeated painful movements at these visits. A limitation of this study was that only the initial effects of the MWM were measured and the time-course of these effects is as yet unknown. Therefore inferences drawn from this study should be limited to those seen in a single treatment session. Another limitation is that only measures of impairment (ROM, PPT) were made, but no measures of function or disability. Several case studies/series have shown that continued treatment with a MWM coincided with a resolution of the condition on function and disability measures (Vicenzino and Wright, 1995; Hseih et al., 2002; Kochar and Dogra, 2002). Further studies to evaluate such issues as the timecourse of the effect of this particular MWM, and the outcome on disability and function after a course of treatment are warranted.

7. Conclusion The results from this study indicate that the shoulder MWM may be a useful manual therapy technique to apply to participants with a painful limitation of shoulder elevation in order to predominantly gain an initial improvement in ROM and PPT.

References Abbott J. Mobilization with movement applied to the elbow affects shoulder range of movement in subjects with lateral epicondylalgia. Manual Therapy 2001;6:170–7. Abbott J, Patta C, Jensen R. The initial effects of an elbow mobilization with movement technique on grip strength in subjects with lateral epicondylalgia. Manual Therapy 2001;6:163–9. Bang M, Deyle G. Comparison of supervised exercise with and without manual physical therapy for patients with shoulder impingement syndrome. Journal of Orthopaedic & Sports Physical Therapy 2000;30:126–37. Bridges-Wegg C Treatment in general practice in Australia. Medical Journal of Australia, 1992 (Suppl S1–S56). Chen SK, Simonian PT, Wickiewicz TL, Otis JC, Warren RF. Radiographic evaluation of glenohumeral kinematics: a muscle

ARTICLE IN PRESS 42

P. Teys et al. / Manual Therapy 13 (2008) 37–42

fatigue model. Journal of Shoulder and Elbow Surgery 1999;8: 49–52. Collins N, Teys P, Vicenzino B. The initial effects of a Mulligan’s mobilization with movement technique on dorsiﬂexion and pain in subacute ankle sprains. Manual Therapy 2004;9:77–82. Exelby L. The Mulligan concept: its application in the management of spinal conditions. Manual Therapy 2002;7:64–70. Flatow E, Soslowsky J, Ticker J, Pawluk R, Hepler M, Ark J, Mow V, Bigliani L. Excursion of the rotator cuff under the acromion. Patterns of subacromial contact. American Journal of Sports Medicine 1994;22:7779–88. Fu F, Harner C, KLein A. Shoulder impingement syndrome: a critical review. Clinical Orthopaedics and Related Research 1991;269: 162–73. Gerhardt J. Documentation of Joint Motion. Portland, Oregon: Isomed Inc; 1992. Ginn K, Herbert R, Khouw W, Lee R. A randomized controlled trial of a treatment for shoulder pain. Physical Therapy 1997;77:802–9. Green S, Buchbinder R, Hetrick S. Physiotherpay interventions for the treatment of shoulder pain. Cochrane Library 2004;2. Halder AM, Zhao KD, O’Driscoll SW, Morrey BF, An KN. Dynamic contributions to superior shoulder stability. Journal of Orthopaedic Research 2001;19:206–12. Harryman D, Sidles J, Clark J, K. M, Gibb T, Matsen F. Translation of the humeral head on the glenoid with passive glenohumeral motion. The Journal of Bone and Joint Surgery 1990;72-A:1334–43. Hayes K, Walton J, Szomor Z, Murrell GAC. Reliability of 5 methods for assessing shoulder range of motion. Australian Journal of Physiotherapy 2001;47:289–94. Howell S, Galinat M, Renzi A, Marone P. Normal and abnormal mechanics of the glenohumeral joint in the horizontal plane. The Journal of Bone and Joint Surgery 1988;70-A:227–32. Hseih C, Vicenzino B, Yang CH, HU M, C Y. Mulligan’s mobilization with movement for the thumb: a single case report using magnetic resonance imaging to evaluate positional fault hypothesis. Manual Therapy 2002;7:44–9. Hsu AT, Ho L, Ho S, Hedman T. Joint position during anterior– posterior glide mobilization: its effect on glenohumeral abduction range of motion. Archives of Physical Medicine and Rehabilitation 2000;81:210–4. Kamkar A, Irrgang J. Nonoperative management of secondary shoulder impingement syndrome. Journal of Orthopaedic & Sports Physical Therapy 1993;17:212–24. Kochar M, Dogra A. Effectiveness of a speciﬁc physiotherapy regimen on patients with tennis elbow: clinical study. Physiotherapy 2002;88:333–41. Ludewig PM, Cook TA. Alterations in shoulder kinematics and associated muscle activity in people with symptoms of shoulder impingement. Physical Therapy 2000;80:276–91.

Ludewig PM, Cook TA. Translations of the humerus in persons with shoulder impingement symptoms. Journal of Orthopaedic & Sports Physical Therapy 2002;32:248–59. Matsen F, Fu F, Hawkins R. The shoulder: a balance of mobility and stability. Rosemont, IL: American Academy of Orthopaedic Surgeons; 1993. Michener LA, Walsworth M, Burnet E. Effectiveness of rehabilitation for patients with subacromial impingement syndrome: a systematic review. Journal of Hand Therapy 2004:152–64. Mulligan B, Manual Therapy ‘‘NAGS’’, SNAGS’’, ‘‘MWMS’’ etc, Plane view Services Ltd., 1999. Nicholson G. The effects of passive joint mobilizations on pain and hypomobility associated with adhesive capsulitis of the shoulder. Journal of Orthopaedic & Sports Physical Therapy 1985;6:238–46. O’Brien T, Vicenzino B. A study of the effects of Mulligan’s mobilization with movement treatment of lateral ankle pain using a case study design. Manual Therapy 1998;3:78–84. Paungmali A, O’Leary S, Souvlis T, Vicenzino B. Hypoalgesic and sympathoexcitatory effects of mobilization with movement for lateral epicondylalgia. Physical Therapy 2003a;83:374–83. Paungmali A, Vicenzino B, Smith M. Hypoalgesia induced by elbow manipulation in lateral epicondylalgia does not exhibit tolerance. The Journal of Pain 2003b;4:448–54. Pontinen PJ. Reliability, validity, reproducibility of algometry in diagnosis of active and latent tender spots and trigger points. Journal of Musculoskeletal Pain 1998;6:61–71. Pope D, Croft P, Pritchard C, Silman A. Prevalence of shoulder pain in the community: the inﬂuence of case deﬁnition. Annals of Rheumatic Diseases 1997;56:308–12. Souvlis T, Vicenzino B, Wright A. Neurophysiological effects of spinal manual therapy. In: Boyling J, Jull G, editors. Grieve’s modern manual therapy the vertebral column. Sydney: Churchill Livingstone; 2005. p. 367–79. Sterling M, Jull G, Wright A. Cervical mobilisation: concurrent effects on pain, sympathetic nervous system activity and motor activity. Manual Therapy 2001;6:72–81. van den Dolden P, Roberts D. A trial into the effectiveness of soft tissue massage in the treatment of shoulder pain. Australian Journal of Physiotherapy 2003;49:183–8. Vicenzino B, Paungmali A, Buratowski S, Wright A. Speciﬁc manipulative therapy treatment for chronic lateral epicondylalgia produces uniquely characteristic hypoalgesia. Manual Therapy 2001;6:205–12. Vicenzino B, Wright A. Effects of a novel manipulative therapy technique on tennis elbow: a single case study. Manual Therapy 1995;1:30–5. Wright A. Hypoalgesia post manipulative therapy: a review of a potential neurophysiological mechanism. Manual Therapy 1995;1: 11–6.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 43–49 www.elsevier.com/locate/math

Multiﬁdus size and symmetry among chronic LBP and healthy asymptomatic subjects Julie Hidesa,b,, Craig Gilmorea, Warren Stantona,b, Emma Bohlscheida a

Division of Physiotherapy, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane 4072, Australia b The UQ/ Mater Back Stability Clinic, Mater Misericordiae Hospital, Raymond Terrace, South Brisbane, Qld 4101, Australia Received 6 June 2006; accepted 13 July 2006

Abstract Previous studies have provided evidence of multiﬁdus muscle atrophy in people with low back pain (LBP). In cases of acute LBP, these studies have shown that the pattern of atrophy is both vertebral level and side speciﬁc. For chronic LBP, there are conﬂicting reports about the extent and location of muscle atrophy. The purpose of this study was to compare chronic LBP patients and asymptomatic subjects on measures of multiﬁdus size (cross-sectional area; CSA) and symmetry (proportional difference of relatively larger side to smaller side). Data were obtained from 40 asymptomatic subjects without a prior history of LBP (13 females, 27 males), and a retrospective audit was undertaken of records from 50 chronic low back pain patients (27 females, 23 males) presenting to a back pain clinic. Results of the analysis showed that chronic LBP patients had signiﬁcantly smaller multiﬁdus CSAs than asymptomatic subjects at the lowest two vertebral levels. Males were found to have signiﬁcantly larger multiﬁdus CSAs than females at all vertebral levels except L5, the most common symptomatic level as determined by manual examination. The greatest asymmetry between sides was seen at the L5 vertebral level in patients with unilateral pain presentations. The smaller multiﬁdus CSA was ipsilateral to the reported side of pain in all cases. The results of this study support previous ﬁndings that the pattern of multiﬁdus muscle atrophy in chronic LBP patients is localized rather than generalized. Furthermore, between side asymmetry may be seen in chronic LBP patients presenting with a unilateral pain distribution. r 2006 Elsevier Ltd. All rights reserved. Keywords: Chronic low back pain (LBP); Multiﬁdus; Manual examination; Physiotherapy

1. Introduction The lumbar multiﬁdus muscle has been the subject of considerable research using imaging techniques. Data pertaining to multiﬁdus size and symmetry in normal subjects have been obtained using tracings of transverse sections in a cadaver study (Amonoo-Kuoﬁ, 1983), Magnetic Resonance Imaging (MRI) (Millerchip et al., 1988; Tracy et al., 1989) and real-time ultrasound Corresponding author. Division of Physiotherapy, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Queensland 4072 Australia. Tel.: +61 7 3365 2718; fax: +61 7 3365 2775. E-mail address: [email protected] (J. Hides).

1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.07.017

imaging (Hides et al., 1992, 1994, 1995; Stokes et al., 2005). These techniques allow the normal morphology of the muscle to be compared with conditions of acute and chronic low back pain (LBP). Several studies have examined multiﬁdus size among asymptomatic, healthy persons. A study of healthy young females using ultrasound imaging and MRI showed that the multiﬁdus was symmetrical between sides at each vertebral level studied (L2-S1) and increased incrementally in size from L2 to L5 (Hides et al., 1995). This was in agreement with an earlier study conducted on a small population of normal healthy adults (Hides et al., 1994). Fifty-one normal subjects were measured using real-time ultrasound imaging at the L4 vertebral level, and in 10 subjects measurements were

ARTICLE IN PRESS 44

J. Hides et al. / Manual Therapy 13 (2008) 43–49

made at each vertebral level from L2-S1. The magnitude of the between side difference was only 374%. A recent study of 120 asymptomatic subjects provided reference ranges for the lumbar multiﬁdus at the L4 and L5 vertebral levels across various age ranges using real-time ultrasound imaging (Stokes et al., 2005). Results showed that multiﬁdus CSA was larger in males, and age had no effect on multiﬁdus size. In agreement with previous studies, between-side symmetry was high for multiﬁdus muscle size (CSAo10% between-side difference). In cases of acute LBP, atrophy of the multiﬁdus muscle has been demonstrated. A study of acute unilateral LBP patients using real-time ultrasound imaging demonstrated a unilateral decrease in multiﬁdus CSA ipsilateral to painful symptoms, i.e. marked asymmetry between the symptomatic and asymptomatic side. (Hides et al., 1994, 1996). The wasting was predominantly isolated to one vertebral level (L5) and the level identiﬁed using real-time ultrasound imaging corresponded with the ‘most affected’ based on clinical examination. Similar ﬁndings were reported by Barker et al. (2004) in a recent MRI study of 50 LBP patients with unilateral pain of greater than 12 weeks duration. Selective atrophy of multiﬁdus was found ipsilateral to the side of symptoms at the vertebral level determined by clinical examination. Changes were found to be less at the vertebral levels above and below the clinically determined levels. Similarly, localized multiﬁdus atrophy has been demonstrated among patients with chronic LBP. In a study using CT scanning, Danneels et al. (2000) showed that patients with chronic LBP displayed reduced multiﬁdus CSA at the L4 vertebral endplate when compared with control subjects. Other vertebral levels were not signiﬁcantly different in size. A recent MRI investigation of chronic LBP also found evidence of multiﬁdus atrophy at the L4 and L5 vertebral levels, but not at the other vertebral levels. Eighty-ﬁve percent of the subjects imaged displayed disc degeneration at the L4 and L5 vertebral levels. Therefore, the primary aim of this study was to compare multiﬁdus muscle size and between side symmetry in chronic LBP patients and asymptomatic subjects without a history of LBP. A secondary aim was to determine if patients with chronic LBP with different pain presentations (e.g. central/ bilateral LBP, unilateral LBP) displayed different patterns of multiﬁdus muscle atrophy.

2. Methods 2.1. Subjects Records from 50 patients (23 males and 27 females) presenting to a hospital based back pain clinic between 1998 and 2002 for assessment and management of chronic LBP were used in this study. The mean age and

standard deviation of the ages for this sample was 46.8713.2 years. The subjects had a history of LBP in excess of three months, with LBP deﬁned as pain localized between T12 and the gluteal fold. Individual cases were excluded from review due to the following reported criteria: previous lumbar surgery, pregnancy, systemic disease plus any reﬂex and or motor signs of nerve root or cauda equina compression. Ultrasound data (multiﬁdus CSA) were also obtained from a convenience sample of 40 individuals with no history of LBP. The mean age for this sample of 27 males and 13 females was 28.475.7 years. Exclusion criteria included a signiﬁcant history of LBP, previous lumbar surgery, previous lumbar injury, signiﬁcant spinal abnormality (detected by clinical examination), history of neuromuscular disease, history of signiﬁcant joint disease, and involvement in speciﬁc training of the low back muscles within the previous 3 months. The project was approved by the Mater Hospital Ethics Committee, South Brisbane, Queensland, Australia. Written, informed consent was obtained from all subjects. 2.2. Procedure Previous clinical trials have established the validity of using ultrasound imaging to assess multiﬁdus size, showing it to be a repeatable and reliable imaging technique in the hands of a trained assessor (Hides et al., 1992, 1994, 1995). Validity of the measure has also been demonstrated by comparison with MRI (Hides et al., 1995). Multiﬁdus CSAs were measured from L2 to L5 vertebral segments. The ultrasound imaging apparatus (Diasonics-Synergy, Japan) was equipped with a 5-MHz convex array transducer. While the subjects were lying in a prone position with pillows under the hips to eliminate the lumbar lordosis, the spinous processes of L2–L5 were palpated and marked with a pen prior to imaging. The subjects were instructed to relax the paraspinal musculature, electro-conductive gel was applied, and the transducer placed transversely over the spinous process of L2. The resultant image was displayed on the screen (Fig. 1) from which multiﬁdus CSAs were determined using electronic calipers. This process was repeated for each subsequent vertebral segment. Bilateral images of the multiﬁdus muscles were obtained except in the case of larger subjects where left and right sides were imaged separately. After the assessment of muscle size, symptomatic vertebral levels for LBP patients were determined by manual examination. This involved central lumbar postero-anterior intervertebral movements as described by Maitland (1986). Manual examination results were based on the patient providing a verbal response in relation to pain provocation to determine the most ‘‘affected’’ vertebral

ARTICLE IN PRESS J. Hides et al. / Manual Therapy 13 (2008) 43–49

45

pain (left/right back or leg pain, as nominated by the patient and drawn on a body chart); symptomatic vertebral level determined by manual examination (L2– L5); and multiﬁdus CSA (left/right for L2–L5). Fewer patients (n ¼ 22) had CSA at L2 recorded in their charts. Files from only one clinician were used in this study to ensure consistency of the measures. The clinician selected for the study had previously demonstrated reliability for the measurement of the multiﬁdus muscle CSA using real-time ultrasound imaging (coefﬁcient of variation ¼ 3.58%; root-mean-square error ¼ 0.17 cm2). Subject questionnaires were routinely administered, collected and scored by a research assistant at the time of assessment, so that the clinician conducting the measurements of the muscle system was blind to the results of the questionnaires. The grouping of cases as ‘bilateral’ or ‘unilateral’ pain was based on body chart reports of leg pain and/or low back pain. For the majority of cases, unilateral leg pain or unilateral back pain were coded as ‘unilateral’ and bilateral leg pain or bilateral or central back pain were coded as ‘bilateral’. Among the chronic LBP patients, 30 were coded as bilateral cases and 12 coded as unilateral cases. In cases of ‘‘mixed’’ grouping (different code for leg pain vs. LBP) the code was based on leg pain. Consequently, three cases of unilateral leg pain and bilateral or central back pain were coded as ‘unilateral’, and four cases of bilateral leg pain and unilateral back pain coded as ‘bilateral’. Fig. 1. Bilateral transverse image at the L5 vertebral level showing atrophy of the right multiﬁdus muscle in a subject with right sided low back pain, with and without CSA tracings.

level/s. This has been shown to be the most reliable method of performing a manual examination (Phillips and Twomey, 1996). The clinician who performed the ultrasound measurements was not blinded to group allocation, as the patient data were collected in hospital based clinic. The data collected on multiﬁdus muscle size in normal subjects were collected by the same clinician as part of a separate study. Complete blinding of the clinician to different aspects of the examination of the LBP patients was not possible in this study due to data being collected in a hospital based clinic with an emphasis on patient care. The clinician performed ultrasound imaging of the multiﬁdus muscle prior to manual examination. Chart audits were conducted by a research assistant with physiotherapist qualiﬁcations, but with no afﬁliation to either the research clinic or research team. Audit information was entered onto a spreadsheet from the following self-reported measures: general level of pain on day of assessment [visual analogue scale (VAS) rated 0–10]; level of disability [Roland Morris Questionnaire (RMQ) rated 0–24]; symptom duration (months); side of

2.3. Statistical analysis Information from eight patients was excluded due to missing data, leaving 42 LBP cases (21 males, 21 females) and 40 asymptomatic cases for the analysis. An initial analysis of variance was used to compare the unilateral and bilateral groups in terms of their age, duration of LBP symptoms, general level of pain (VAS) and level of disability (RMQ). 2.3.1. Multifidus size The data for multiﬁdus CSA were averaged across the left and right sides. The effect of the variables ‘pain group’ and ‘gender’ on multiﬁdus muscle size (including the interaction between these factors) was analyzed using analysis of covariance in the Statistical Package for Social Sciences (SPSS). The variable ‘pain group’ was coded to represent three groups (no pain/asymptomatic, bilateral or central pain, unilateral pain). The variable of ‘age’ was entered as a covariate in the analyses, resulting in effects being adjusted for age. As patients may be symptomatic at more than one level, separate analysis was conducted for the vertebral levels of L2, L3, L4 and L5. Between group contrasts were used to test for differences of the unilateral pain group versus the bilateral pain group and asymptomatic

ARTICLE IN PRESS J. Hides et al. / Manual Therapy 13 (2008) 43–49

46

3.2. Manual examination

group. Mean values (adjusted for age) and 95% conﬁdence intervals were obtained in order to graph the differences between the groups across segmental levels.

Twenty-nine cases reported pain provocation upon manual examination of the L5 vertebral level, 11 at the L4 vertebral level and one at the L3 vertebral level. One LBP subject reported pain provocation on manual examination of L2, L3 and L4 vertebral levels.

2.3.2. Multifidus symmetry The degree of symmetry of multiﬁdus CSA was calculated as a percentage difference between sides, relative to the larger side [% difference ¼ (largest side – smallest side/largest side) 100)] (Hides et al., 1996). Analysis of covariance with between group contrasts was used to compare the degree of symmetry of the unilateral pain group with the bilateral pain group and asymptomatic ‘normal’ group. ‘Gender’ was also included as a factor and ‘age’ was entered as a covariate in the analyses. Separate analyses were conducted for the vertebral levels of L2, L3, L4 and L5.

3.3. Multifidus size (CSA)—chronic LBP vs. asymptomatic subjects Results of the analysis (Table 1) showed that at the L4 and L5 vertebral levels, asymptomatic subjects had signiﬁcantly larger multiﬁdus muscles compared with LBP subjects. Mean values for the unilateral, bilateral and healthy asymptomatic groups are shown in Fig. 2. Although the asymptomatic subjects were younger than the subjects with LBP, this had little effect on the results as the factor of ‘age’ was not associated with CSA at any level of the lumbar spine. In asymptomatic subjects, a clear increase in multiﬁdus CSA with caudal progression was demonstrated (Fig. 2). At levels L2–L4, male

3. Results 3.1. Self-reported questionnaires

8

Unilateral

7

As there were no differences between the unilateral and bilateral groups in terms of their age, duration of symptoms, level of pain and level of disability (P40.05), means and standard deviations for the combined LBP patients are reported. Chronic LBP subjects reported a mean pain VAS score of 4.4372.65, with individual pain ratings ranging between zero and maximum (10/10) at the time of examination. Baseline RMQ disability scores ranged between 0 and 20 (out of a possible 24) with a mean score of 7.0075.58. The mean duration of symptoms was 62.19790.58 months, with a range of 3 months to 40 years (480 months), reﬂecting the prolonged chronicity of the LBP population presenting to the clinic.

Bilateral

CSA (cm2)

6

Asymptomatic

5 4 3 2 1 0 L2

L3

L4

L5

Fig. 2. Size of multiﬁdus muscle in transverse section determined by ultrasound imaging at vertebral levels L2–L5 for asymptomatic and chronic LBP subjects (bars represent 95% conﬁdence intervals).

Table 1 Size (cm2) of multiﬁdus CSA for asymptomatic subjects and LBP patientsa Variables

Pain group Asymptomatic Bilateral or central Unilateral (reference category for p values) Gender Females Males Age Pain Group Gender a

L2 (n ¼ 63)

L3 (n ¼ 79)

L4 (n ¼ 82)

L5 (n ¼ 82)

Mean (SD)

P

Mean (SD)

P

Mean (SD)

P

Mean (SD)

P

2.48 (0.98) 2.27 (1.48) 2.44 (2.28)

0.97 0.56

3.66 (1.43) 3.31 (1.59) 3.34 (2.43)

0.37 0.29

5.42 (1.88) 4.07 (1.88) 3.40 (3.11)

0.001 0.065

6.48 (1.72) 3.78 (1.73) 3.22 (2.84)

0.001 0.093

2.01 (1.37) 2.78 (1.13)

0.001

3.02 (1.40) 3.85 (1.36)

0.001

3.80 (1.74) 4.79 (1.70)

0.001

4.35 (1.59) 4.64 (1.56)

0.22

Marginal means (adjusted for age); SD ¼ standard deviation.

0.15 0.73

0.08 0.31

0.70 0.39

0.63 0.67

ARTICLE IN PRESS J. Hides et al. / Manual Therapy 13 (2008) 43–49

subjects were found to have signiﬁcantly larger multiﬁdus CSA than their female counterparts. This relationship was not evident at L5. 3.4. Symmetry—unilateral chronic LBP vs. bilateral chronic LBP and asymptomatic subjects Results of the analysis (Table 2) showed that at L4 and L5 vertebral levels, unilateral LBP patients had signiﬁcantly more asymmetry in the size of their multiﬁdus muscles than bilateral LBP patients and asymptomatic subjects. At L2 and L3 there were no differences found. ‘Gender’ was not related to degree of asymmetry at any vertebral level. Although the asymptomatic subjects were younger than the subjects with LBP, ‘age’ was not signiﬁcantly related to the degree of asymmetry at any level of the lumbar spine. The results for subjects with unilateral pain presentations showed that in all cases, the smaller side was ipsilateral to the reported side of symptoms.

47

the L4/L5 and L5/S1 vertebral levels. Eighty-ﬁve percent of the cases studied also had disc degeneration at the L4 and L5 vertebral levels. In the current investigation, no differences were found between the normal subjects and the chronic LBP patients at the L2 and L3 vertebral levels. The results from the asymptomatic subjects (Fig. 2) show that multiﬁdus size increased on progression from L2 to L5 (Hides et al., 1995). Stokes et al. (2005), who measured L4 and L5 vertebral levels, also showed that multiﬁdus size increased proportionally from L4 to L5 in both genders (Stokes et al., 2005). The mean CSAs of the multiﬁdus muscles were comparable with results previously reported by Hides et al. (1992, 1994), and were similar to the results of Stokes et al. (2005) for the L4 vertebral level. At the L5 vertebral level, mean CSA’s reported by Stokes et al. (2005) were larger than those reported previously and larger than the mean CSA’s reported in this investigation. One explanation may be that the subjects studied by Stokes et al. (2005) had larger mean body masses and were more active.

4. Discussion 4.2. Symmetry of multifidus size—unilateral LBP vs. bilateral LBP and asymptomatic subjects

4.1. Multifidus size—chronic LBP vs. asymptomatic subjects Several previous imaging studies have reported evidence of multiﬁdus muscle atrophy in patients with LBP. Studies have investigated post-operative patients (Sihvonen et al., 1993), patients with acute/ subacute LBP (Hides et al., 1994, 1996) and patients with chronic LBP (Barker et al., 2004; Danneels et al., 2000, 2001; Kader et al., 2000). In agreement with the work of Danneels et al. (2000), Barker et al. (2004) and Kader et al. (2000), the pattern of atrophy seen in the chronic LBP patients investigated appeared to be speciﬁc and localized. Atrophy was greatest at the two lowest vertebral levels. Kader et al. (2000) performed a retrospective review of 75 LBP patients who presented for MRI, and found multiﬁdus atrophy was greatest at

Between-side comparisons are performed in the clinical situation to examine for unilateral abnormalities. Results of studies performed on normal subjects have previously shown that the multiﬁdus muscle is symmetrical between sides. Hides et al. (1992, 1994) reported a 374% between side difference in normal subjects (relative to the larger side), and Stokes et al. (2005) recently reported between-side differences that ranged from 7.2% to 9.6% (relative to the smaller side) at L4 and L5 vertebral levels. Between side differences for normal subjects ranged from 2.9% to 5.8% in the current study, suggesting that the multiﬁdus muscle is relatively symmetrical. Based on these results, future studies could regard asymmetry of the multiﬁdus (410%) as a potential abnormality.

Table 2 Asymmetry (percentage difference between sides) of multiﬁdus CSA for asymptomatic subjects and LBP patientsa Variables

Pain group Asymptomatic Bilateral or central Unilateral (reference category for contrasts) Gender Females Males Age Pain group gender a

L2 (n ¼ 63)

L3 (n ¼ 77)

L4 (n ¼ 82)

L5 (n ¼ 82)

Mean% (SD)

P

Mean% (SD)

P

Mean% (SD)

P

Mean% (SD)

P

5.2 (6.7) 2.8 (10.2) 1.8 (15.7)

0.17

3.8 (8.3) 3.8 (9.2) 4.0 (14.1)

0.92

3.4 (12.0) 5.1 (12.2) 11.8 (19.1)

0.003

1.9 (15.2) 10.5 (15.5) 17.5 (24.2)

0.001

3.6 (9.4) 2.9 (7.8)

0.64 0.60 0.37 0.06

Marginal means (adjusted for age); SD ¼ standard deviation.

4.3 (8.13) 3.4 (7.9)

0.93 0.46 0.15 0.48

7.1 (11.3) 6.5 (10.6)

0.004 0.68 0.44 0.77

11.5 (14.3) 8.4 (13.4)

0.016 0.15 0.65 0.46

ARTICLE IN PRESS 48

J. Hides et al. / Manual Therapy 13 (2008) 43–49

Hides et al. (1994) documented multiﬁdus asymmetry in subjects with acute unilateral ﬁrst episode LBP patients, with 3178% asymmetry between the symptomatic and asymptomatic sides. Similar results were found in LBP patients with a unilateral pain presentation of greater than 12 weeks duration (mean 15.675.8 weeks) (Barker et al., 2004). Barker et al. (2004) reported a between-side difference of 21.7% in multiﬁdus CSA at the symptomatic vertebral level. Although the mean duration of symptoms of the LBP patients in the present study was much longer than that reported by Barker et al. (2004), our results for patients with unilateral pain presentations support the previous ﬁnding of localized multiﬁdus asymmetry with atrophy ipsilateral to symptoms at speciﬁc vertebral levels. The results of the current study were very similar (between-side differences of 17.5% at L5, 11.8% at L4, 4.0% at L3 and 1.8% at the L2 vertebral level in subjects with a unilateral pain presentation). The between-side differences were lower in patients with bilateral or central pain presentations and ranged from 2.8% to 10.5%. However, the ﬁnding of some asymmetry (410%) at the L5 vertebral level in chronic LBP patients presenting with bilateral or central pain could possibly be related to previous episodes of LBP. It is not uncommon in the clinical situation for patients to report central or bilateral LBP on presentation, but report that previous episodes were unilateral in distribution. The documented asymmetry in some cases with bilateral or central pain may therefore represent residual atrophy from previous episodes of unilateral LBP. This is the ﬁrst study to our knowledge to evaluate the pattern of multiﬁdus atrophy in chronic LBP patients and to relate its’ ﬁndings to different pain presentations.

assessor. Furthermore, lumbar muscle atrophy can be characterized by a decrease in muscle size and by alterations in consistency, which can be due to ﬁbrotic changes, scar tissue or fatty inﬁltration (Mayer et al., 1989; Parkkola et al., 1993; Sihvonen et al., 1993; Stokes et al., 1992). These alterations have been measured in subjects with LBP in CT studies (Danneels et al., 2000). The fat content of skeletal muscle may also be inﬂuenced by age (Tsubahara et al., 1995). The consistency of the muscles was not assessed in this study.

5. Conclusion Patients with chronic LBP had signiﬁcantly smaller multiﬁdus CSAs than asymptomatic subjects at the lowest two vertebral levels. The greatest asymmetry between sides was seen at the L5 vertebral level in patients with unilateral pain presentations. The results of this study support previous ﬁndings that the pattern of multiﬁdus muscle atrophy in chronic LBP patients is localized rather than generalized. This supports the clinical assumption that exercise therapy needs to be speciﬁc and tailored to address speciﬁc localized impairments present in patients with chronic LBP. Furthermore, between side asymmetry may be seen in chronic LBP patients presenting with a unilateral pain distribution. Clinical assessment of the multiﬁdus muscle in LBP patients should include a comparison between vertebral levels and between sides to detect impairments and guide exercise intervention.

Acknowledgements 4.3. Limitations and future directions This study has a number of limitations. The study sample size was small, though comparable with other similar investigations (Barker et al., 2004; Kader et al., 2000). The subjects in the asymptomatic group were not matched to the subjects in the LBP group. In this study, the clinician who measured multiﬁdus muscle size was not blinded to group allocation (asymptomatic vs. LBP) as the LBP subjects were seen in a clinical situation. Ideally, the person conducting the measurements would be blinded to group allocation. The assessor was blinded to the results of the pain and disability questionnaires. Despite the assessor undergoing repeatability and reliability trials prior to this study, there was the potential for measurement error. Among subjects with chronic LBP, changes in muscle consistency commonly occur in the muscles, which appear hyper-echoic (increased brightness of the image). This can increase the difﬁculty of conducting accurate measurements, but was minimized in this study by using an experienced

The authors would like to acknowledge the subjects studied, the support of Ms. Linda Blackwell (Director of Physiotherapy Services at the Mater Misericordiae Hospital, South Brisbane, and Manager of the UQ/ Mater Back Stability Clinic), Megan Bamberry (physiotherapist) for chart audit and data entry, Heidi Keto, Virginia Brooks and Jack Simpson for assistance with this manuscript and Associate Professor Carolyn Richardson (Division of Physiotherapy, The University of Queensland and Director of the UQ/ Mater Back Stability Clinic, Mater Hospital South Brisbane).

References Amonoo-Kuoﬁ HS. The density of muscle spindles in the medial, intermediate and lateral columns of human intrinsic post-vertebral muscles. Journal of Anatomy 1983;136:509–19. Barker KL, Shamley DR, Jackson D. Changes in the cross-sectional area of multiﬁdus and psoas in patients with unilateral back pain— the relationship to pain and disability. Spine 2004;29(22):E515–9.

ARTICLE IN PRESS J. Hides et al. / Manual Therapy 13 (2008) 43–49 Danneels LA, Vanderstraeten GG, Cambier DC, Witvrouw EE, De Cuyper HJ. CT imaging of trunk muscles in chronic low back pain patients and healthy control subjects. European Spine Journal 2000;9:266–72. Danneels LA, Vanderstraeten GG, Cambier DC, Witvrouw E, Bourgois J, Dankaerts W, De Cuyper HJ. Effects of three different training modalities on the cross sectional area of the lumbar multiﬁdus muscle in patients with chronic low back pain. British Journal of Sports Medicine 2001;35(3):186–91. Hides JA, Cooper DH, Stokes MJ. Diagnostic ultrasound imaging for measurement of the lumbar multiﬁdus muscle in normal young adults. Physiotherapy Theory and Practice 1992;8:19–26. Hides JA, Stokes MJ, Saide M, Jull GA, Cooper DH. Evidence of lumbar multiﬁdus wasting ipsilateral to symptoms in patients with acute/subacute low back pain. Spine 1994;19(20):165–72. Hides JA, Richardson CA, Jull GA. Magnetic resonance imaging and ultrasonography of the lumbar multiﬁdus muscle—comparison of two different modalities. Spine 1995;20(1):54–8. Hides JA, Richardson CA, Jull GA. Multiﬁdus muscle recovery is not automatic after resolution of acute, ﬁrst-episode low back pain. Spine 1996;21(23):2763–9. Kader DF, Wardlaw D, Smith FW. Correlation between MRI changes in the lumbar multiﬁdus muscles and leg pain. Clinical Radiology 2000;55(2):145–9. Maitland GD. Vertebral manipulation, 5th ed. London: ButterworthHeinemann; 1986. pp. 282–5. Mayer TG, Vanharanta H, Gatchel RJ, Mooney V, Barnes D, Judge L, Smith S, Terry A. Comparison of CT scan muscle measurements

49

and isokinetic trunk strength in postoperative patients. Spine 1989;14(1):33–6. Millerchip R, Savage A, Edwards RHT. Magnetic resonance anthropometry of muscles stabilising the lumbar spine. Clinical Science 1988;75(suppl 19):34. Parkkola R, Rytokoski U, Kormano M. Magnetic Resonance Imaging of the discs and trunk muscles in patients with chronic low back pain and healthy control subjects. Spine 1993;18(7):830–6. Phillips DR, Twomey LT. A comparison of manual diagnosis with a diagnosis established by a uni-level lumbar spinal block procedure. Manual Therapy 1996;1(2):82–7. Sihvonen T, Herno A, Paljarvi L, Airaksinen O, Partanen J, Tapaninaho A. Local denervation atrophy of paraspinal muscles in postoperative failed back syndrome. Spine 1993;18(5):575–81. Stokes MJ, Cooper RG, Morris G, Jayson MIV. Selective changes in multiﬁdus dimensions in patients with chronic low back pain. European Spine Journal 1992;1:38–42. Stokes M, Rankin G, Newham DJ. Ultrasound imaging of lumbar multiﬁdus muscle: normal reference ranges for measurements and practical guidance on the technique. Manual Therapy 2005;10: 116–26. Tsubahara A, Chino N, Akoboshi K, Okajima Y, Takahashi H. Age related changes of water and fat in muscles estimated by magnetic resonance (MR) imaging. Disability and Rehabilitation 1995;17(6): 298–304. Tracy MF, Gibson MJ, Szypryt EP, Rutherford A, Corlett EN. The geometry of the muscle of the lumbar spine determined by magnetic resonance imaging. Spine 1989;14:186–93.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 50–56 www.elsevier.com/locate/math

Original article

Inter-examiner reliability of four static palpation tests used for assessing pelvic dysfunction Ulrika Holmgren, Kerstin Waling Department of Community Medicine and Rehabilitation, Physiotherapy Umea˚ University, Umea˚, Sweden Accepted 30 August 2006

Abstract In muscle energy technique (MET), palpation is an important tool aimed at detecting asymmetry and selecting interventions. The aim of this study was to test the inter-examiner reliability of static palpation of the transverse processes of L5 (L5), sacral sulci (SS), inferior lateral angles of the sacrum (ILA), and the medial malleoli (MM) in a clinical setting. Twenty-ﬁve participants, aged 18–78 years, with low back pain and/or sacroiliac pain with or without radiating pain to the knee, were independently examined by two experienced physiotherapists. For L5, SS, ILA, the proportion of observed agreement was 40–44% and the k coefﬁcient 0.11 (SE ¼ 0.12) to 0.17 (SE ¼ 0.10). For MM, the observed agreement was 52% and the k coefﬁcient 0.28 (SE ¼ 0.15). Differences in palpation technique seem to be the most likely source of the low inter-examiner reliability in this study. For clinical practise, continued use of these tests as methods for detecting asymmetry and selecting interventions is of doubtful utility. r 2006 Elsevier Ltd. All rights reserved. Keywords: Inter-tester reliability; Physical therapy; Reproducibility; Sacroiliac joint

1. Introduction In the last two decades, researchers have concluded that the sacroiliac joint may be one cause of low back pain (Walker, 1992; Daum, 1995; Schwarzer et al., 1995; Dreyfuss et al., 1996; Maigne et al., 1996; Slipman et al., 1996; Laslett et al., 2003; Laslett et al., 2005; van der Wurff et al., 2006). The prevalence of the sacroiliac joint as a source of low back pain is reported to be between 13% and 48% in different studies (Daum, 1995; Schwarzer et al., 1995; Maigne et al., 1996; Touissant et al., 1999a, b). Manual examination is a standard method of examining the sacroiliac joint (Touissant et al., 1999a, b). Some manual tests are designed to assess the location of and relative symmetry between the left and right side of the joint (static palpation), or to assess Corresponding author. Ersboda Va˚rdcentral, Ha¨lsogra¨nd 3, S-906 25 Umea˚, Sweden. E-mail address: [email protected] (U. Holmgren).

1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.09.009

movement of bony landmarks associated with the sacroiliac joint (motion palpation). Application of force to the sacroiliac joint or related structures in an attempt to reproduce pain is another kind of test (pain provocation) (Walker, 1992; Simmonds and Kumar, 1993; Laslett et al., 2003, 2005; Meijne et al., 1999; van der Wurff et al., 2000, 2006; Freburger and Riddle, 2001). Tests that fall into these three categories are typically performed without assistance of instrumentation (van der Wurff et al., 2000; Freburger and Riddle, 2001). Attempts to use X-ray, bone scan or laboratory tests in investigating the sacroiliac joint have not been successful (Dreyfuss et al., 1996; Slipman et al., 1996; Tullberg et al., 1998). Since tests of the sacroiliac joint are often based on palpation, accurate and reliable palpation skills are a prerequisite for them to be considered potentially valuable. Both static palpation tests (Potter and Rothstein, 1985; Lindsay et al., 1995; O’Haire and Gibbons, 2000) and motion palpation tests (Potter and Rothstein, 1985; Meijne et al., 1999; Vincent-Smith and Gibbons, 1999) in the

ARTICLE IN PRESS U. Holmgren, K. Waling / Manual Therapy 13 (2008) 50–56

pelvic region have to date failed to demonstrate acceptable levels of both inter- and intra-examiner reliability. In muscle energy technique (MET) palpation is an important diagnostic tool (Chaitow, 2003). Bilateral palpation of the transverse processes of L5 (L5), the sacral sulci (SS), and the inferior lateral angles of the sacrum (ILA) are tests used to detect asymmetry and dysfunction in the pelvic region. Studies examining the reliability of pelvic tests used in MET are scarce, and exclusively performed on asymptomatic subjects (O’Haire and Gibbons, 2000). O’Haire and Gibbons (2000) evaluated the inter- and intra-examiner reliability in static palpation of the posterior superior iliac spine (PSIS), the SS, and the ILA. For all three landmarks, the results indicated an inter-examiner agreement only slightly better than what was expected by chance. The intra-examiner reliability was somewhat higher than the inter-examiner reliability, a result consistent with other studies (Meijne et al., 1999; Vincent-Smith and Gibbons, 1999). The authors suggested that difﬁculties in locating anatomical landmarks and the use of asymptomatic subjects might have reduced the reliability (O’Haire and Gibbons, 2000). In a study by Downey et al. (1999), the ability to palpate lumbar spinous processes was evaluated. Three pairs of manipulative physiotherapists presented almost perfect inter-examiner reliability in locating the nominated spinal levels. Studies evaluating the inter-examiner agreement in bilateral palpation of the transverse processes of L5 have to our knowledge not been published. As functional leg length discrepancy is believed to appear due to for example pelvic distortions (Walker, 1992; Chaitow, 2003), leg length evaluation often accompanies the pelvic examination in MET (Chaitow, 2003). Leg length can be assessed by comparing the positions of the medial malleoli (MM) in a prone position. To our knowledge, no study has been presented assessing the inter-examiner reliability of static palpation of the MM in a prone position. Since the results of the four mentioned tests (L5, SS, ILA, MM) are important in the selection of intervention techniques (Weiselﬁsh-Giammatteo and Giammatteo, 2003), we felt it important to study their reliability in a clinical setting. Consequently, the aim of this study was to test the inter-examiner reliability of three static palpation tests used to detect asymmetry in the pelvic region (L5, SS, ILA), and one static palpation test to evaluate leg length (MM) in a symptomatic population.

2. Material and method 2.1. Participants For 4 months, 27 patients with symptoms of low back pain and/or sacroiliac pain at two primary health care

51

clinics were consecutively recruited to participate in the study. To be included they should not have radiating pain below the knee or have been subject to lumbar surgery during the last year. They had to be 18 years or older, and be able to lay prone for 10 min. Two potential participants (a woman at one clinic and a man at the other clinic) failed to come at the time appointed leaving totally 25 patients to the study. Of these 18 were women and seven men. The mean age was 46 years (range 18–78 years), mean height 170 cm (range 156–187 cm) and mean weight 76 kg (range 56–99 kg). During the 12 months preceding the study, 14 of the participants had experienced daily pain. Nine had not been able to work full time or fulﬁl their domestic duties, and 13 had reduced their spare time activities. Twenty participants reported pain during the last 7 days. On the day of the assessment, the participants were asked to rate pain intensity on a visual analogue scale (VAS) ranging from 0 ¼ no pain to 100 ¼ worst possible pain (Huskisson, 1974; Carlsson, 1983). The ratings ranged between 0 and 70 with a group average of 29. The participants were given both verbal and written information about the study and gave their written consent to participate. All participants were informed that they were free to withdraw from the study at any time. The managers of the two primary health care clinics where the study took place and the Ethics Committee of the Medical Faculty of Umea˚ University (Ref. no. 03–351) approved the study. 2.2. Examiners The examiners consisted of three female physiotherapists (PT) (A, B and C), who were working in primary health care. All three PTs had around 15 years of experience of patients with low back pain and sacroiliac joint dysfunction. They were all trained in MET and had practised the technique for more than 10 years. 2.3. Procedure The test order, the performance of the tests, and the position of patients and examiners were determined prior to the study. The tests were performed in the following order: L5, SS, ILA, MM. PT A, also one of the authors (UH), was one of the examiners during all test sessions. Depending on which of the two clinics the patient belonged to, one of the other two PTs (B or C) was the second examiner. Tossing a coin randomized the order of which PT ﬁrst examined each new participant. The test procedure was performed in a true clinical setting on the participant’s initial visit to the physiotherapy clinic. Each participant underwent two examination sessions with a different examiner each time. The participants were instructed to lie in a prone position

ARTICLE IN PRESS 52

U. Holmgren, K. Waling / Manual Therapy 13 (2008) 50–56

Fig. 1. The position of the participant and the examiner when assessing the transverse processes of L5 (L5), the sacral sulci (SS) and the inferior lateral angles of the sacrum (ILA).

with lowered arms, looking down through a hole in the examination table (Fig. 1). They were told not to move until both PTs had completed their examinations. In addition, they were instructed not to inform the second examiner of the results of the ﬁrst examination. The ﬁrst PT palpated and observed the stated anatomical landmarks, ﬁlled in an assessment form, put the form in an envelope, sealed the envelope and left the examination room. The second PT entered the room within 2 min and performed the same procedure as the ﬁrst PT. The two envelopes remained sealed until all participants were examined. 2.4. Assessments During all examinations the PTs were instructed to stand on the right-hand side of the examination table while performing static palpation of L5, SS, and ILA. While performing static palpation of MM, the PTs stood at the end of the examination table. The PTs were instructed to palpate gently in order to feel the beginning of a sense of resistance in the structure, and to assess if the anatomical landmarks were symmetrical or not (Chaitow, 2003; Weiselﬁsh-Giammatteo and Giammatteo, 2003). If necessary, they used their dominant eye to evaluate the results (Weiselﬁsh-Giammatteo and Giammatteo, 2003). To palpate the transverse processes of the L5, the PTs ﬁrst located the L5 spinous process. From the L5 spinous process the thumbs were moved laterally over the area of the transverse processes to assess if one side appeared to be more posterior (Figs. 2 and 3). The SS

were examined by placing the pads of the thumbs on both PSIS, followed by moving the thumbs medially into the SS to assess any differences in depth of the right and left side (Fig. 3). To examine the ILA, the pads of the thumbs were placed laterally to the sacral hiatus on the posterior aspects of the ILAs to assess whether one side appeared more posterior compared to the other side (Weiselﬁsh-Giammatteo and Giammatteo, 2003) (Fig. 3). The leg length was examined by placing the participant’s legs side by side, followed by the examiner placing their thumbs at the medial border of the MM. The PTs compared visually if one leg appeared to be longer (Fig. 4). The possible ﬁndings for each test are summarized in Table 1. 2.5. Statistical methods The Statistical Package for the Social Sciences (SPSS) version 9.0 was used to calculate the percentage agreement between therapists (%), the Kappa agreement coefﬁcient (k), its standard error (SE), its 95% conﬁdence interval (CI), and its signiﬁcance (p). The observed proportion of positive agreement (Ppos) and the observed proportion of negative agreement (Pneg) between the therapists were calculated using Excel 6.0. The k coefﬁcient evaluates inter-examiner agreement of categorical data and determines the proportion of agreements, which is expected by chance contrary to the percentage agreement, which is not chance corrected. The k coefﬁcient ranges in value from 1 to +1 where positive values signify agreement better than chance, a

ARTICLE IN PRESS U. Holmgren, K. Waling / Manual Therapy 13 (2008) 50–56

53

Fig. 2. The examiner evaluating the position of the transverse processes of the L5 (L5).

Fig. 3. Anatomical illustration of the three lumbo-pelvic palpation points: A, the transverse processes of L5 (L5), B, the sacral sulci (SS), C, the inferior lateral angles of the sacrum (ILA).

value 0 denotes agreement no better than chance and a negative value signiﬁes agreement worse than chance. According to the guidelines proposed by Landis and Koch (1977), the strength of agreement between 0 and 1 can be interpreted as follows: 0.00–0.20 slight agreement, 0.21–0.40 fair agreement, 0.41–0.60 moderate agreement, 0.61–0.80 substantial agreement, 0.81–1.00 almost perfect agreement. A k value that

reaches 0.40 or better is considered to be acceptable for clinical use. 3. Results The inter-examiner reliability of examination of the four static palpation tests in all 25 participants is presented in Table 2. The proportion of observed

ARTICLE IN PRESS U. Holmgren, K. Waling / Manual Therapy 13 (2008) 50–56

54

Fig. 4. The examiner evaluates the position of the medial malleoli by placing the thumbs on the medial border followed by a visual comparison if one leg appears to be longer.

Table 1 Description of the possible ﬁndings for each diagnostic test Diagnostic test

Possible ﬁndings

Transverses processes of L5 (L5)

Symmetrical Right side more posterior than left side Left side more posterior than right side

Sacral sulci (SS)

Symmetrical Right side deeper than left side Left side deeper than right side

Inferior lateral angles of the sacrum (ILA)

Symmetrical Right side more posterior than left side Left side more posterior than right side

Medial malleoli (MM)

Symmetrical Right side more superior than left side Left side more superior than right side

Table 2 Inter-examiner reliability of static palpation of four anatomical landmarks (L5, SS, ILA, MM) on 25 participants L5

SS

ILA

MM

% Ppos right Ppos left Pneg

40 15 34 8

44 33 0 28

40 6 18 36

52 28 19 31

k SE 95% CI P

0.17 0.1 0.03, 0.37 0.08

0.11 0.13 0.14, 0.36 0.37

0.11 0.12 0.12, 0.34 0.28

0.28 0.15 0.01, 0.57 0.04

% ¼ proportion of observed agreement, Ppos ¼ proportion of positive ﬁndings (%), Pneg ¼ proportion of negative ﬁndings (%), k ¼ kappa, SE ¼ standard error (kappa), CI ¼ conﬁdence interval (kappa), P ¼ signiﬁcance beyond chance agreement (kappa).L5, tranverse processes of L5; SS, sacral sulcus; ILA, inferior lateral angles of the sacrum; MM, medial malleoli.

4. Discussion agreement for L5, SS, and ILA, varied from 40% to 44%, and the k coefﬁcients for the same tests varied from 0.11 (SE ¼ 0.12) to 0.17 (SE ¼ 0.10) indicating an agreement slightly better than what was expected by chance. For the MM, the proportion of observed agreement was somewhat higher (52%), and the k coefﬁcient was also higher (0.28, SE ¼ 0.15) indicating fair agreement. The proportion of positive agreements varied from 0% to 34%, and the proportion of negative agreements varied from 8% to 36%, indicating that the examiners disagreed frequently on both positive and negative ﬁndings.

The results indicated poor reliability for all four tests, a result consistent with other static palpation studies (Potter and Rothstein, 1985; Lindsay et al., 1995; O’Haire and Gibbons, 2000). In the study we included all subjects that for a period of 4 months, consulted the PTs at two primary health clinics for low back pain and/ or sacroiliac pain, with or without radiating pain to the knee. A broad inclusion criterion of reported low back and/or sacroiliac pain was chosen to make sure that the tests were evaluated on participants on whom the tests normally would be conducted routinely as part of their

ARTICLE IN PRESS U. Holmgren, K. Waling / Manual Therapy 13 (2008) 50–56

clinical evaluation (Domholdt, 2000; Fritz and Wainner, 2001). The majority of women only reﬂect the present gender distribution in primary health care. The study was performed in a clinical setting and designed to mirror the clinical situation where a physiotherapist at times uses colleagues for a second opinion or where different PTs meet the patient from one occasion to another. This study showed that the inter-examiner reliability in a symptomatic population was just as low as in studies of asymptomatic populations (Lindsay et al., 1995; Vincent-Smith and Gibbons, 1999; O’Haire and Gibbons, 2000). The proportion of percentage agreement was in most cases less than 50%, and the k values for L5, SS and ILA indicated a result only slightly better than what was expected by chance. For the MM, the result was somewhat better but still not acceptable for clinical use. Regarding inter-examiner reliability the result of the present study must be considered unacceptable, especially in the light of the Ppos and Pneg values. Standardization of measurement techniques is claimed to be the best way to improve reliability (Domholdt, 2000; Fritz and Wainner, 2001). In the present study, efforts were made to have a highly standardized but at the same time clinically practical and realistic test situation. Thereby the risk of bias due to the testing conditions was minimized. The distinct description of the palpation technique in MET possibly contributed to improved reliability in contrast to other palpation studies where therapists have been permitted to use their own methods of palpation (Lindsay et al., 1995, O’Haire and Gibbons, 2000). However, during the examination some of the participants spontaneously reported that the force used differed between therapists. Variation in examination technique, such as applied force and timing, has been shown to occur both within and between therapists irrespective of experience (Levin et al., 2005), and could inﬂuence both the test outcome and the reliability. Since all three PTs had similar experiences of the method MET and we wanted to resemble a clinical situation, no attempt to synchronize the examiners by training sessions was performed. According to Simmonds and Kumar (1993), not synchronizing the examiners may result in poorer reliability, but perhaps it gives a more clear indication of clinical practise. Keating et al. (1993), have shown that it is possible to train therapists to recognize and communicate the amount of force during palpation, but studies utilizing training programmes to synchronize examiners prior to the investigation have reported both high (Laslett and Williams, 1994; Kokmeyer et al., 2002) and low levels of reliability (Vincent-Smith and Gibbons, 1999; O’Haire and Gibbons, 2000; Riddle and Freburger, 2002). Previously it has been suggested that ﬁndings of pelvic tests could be more stable in symptomatic participants

55

compared to asymptomatic ones (Vincent-Smith and Gibbons, 1999). The time between the two examinations in our study was less than 2 min, which was considered to be sufﬁcient to prevent the participants from moving. However, the second examiner in our study more often reported positive ﬁndings. Small movements when examining the MM, and/or increasing pain due to the prone position, or from the ﬁrst examination, may have led to changes in the soft tissue texture and contributed to different assessments by the examiners. The low reliability could also be a result of a systematic difference between examiners in their evaluation, but perhaps the most likely explanation is that the structures palpated lay too far below the surface to be palpated with any kind of accuracy. It is important to note that the results of our study are not applicable to other conditions than those presented here, and that the results should be interpreted with caution. In this study, the inter-examiner reliability is evaluated, and reliability is only a measure of reproducibility, not a gauge of accuracy. However, the results of this study clearly show that the four studied tests performed by experienced examiners and in a clinical setting have poor or limited reliability. Consequently, continued use of these tests as methods for detecting asymmetry and selecting interventions is doubtful and should be reconsidered, especially since a selection of pain provocation tests (distraction test, thigh thrust test, Gaenslen’s test, compression test and sacral thrust test) has been able to show both acceptable reliability and validity and can also be used to identify the sacroiliac joint as a source of low back pain (Laslett and Williams, 1994; Kokmeyer et al., 2002; Laslett et al., 2003; Young et al., 2003, Laslett et al., 2005; van der Wurff et al., 2006).

5. Conclusion The static palpation of L5, SS, ILA, and MM performed by experienced examiners in symptomatic subjects demonstrated an inter-examiner reliability only slightly better than what was expected by chance. Differences in palpation technique seem to be the most likely source of the low inter-examiner reliability in this study. For clinical practise, continued use of these tests as methods for detecting asymmetry and selecting interventions is of doubtful utility.

Acknowledgements We thank the men and women who consented to participate in the study and the physiotherapists Berith Joelsson and Anette Persson for their valuable help with the study.

ARTICLE IN PRESS 56

U. Holmgren, K. Waling / Manual Therapy 13 (2008) 50–56

The study was approved by the Ethical Committee of the Medical Faculty of Umea˚ University, Umea˚, Sweden (y437/03, ref. no. 03-351).

References Carlsson AM. Assessment of chronic pain. I. Aspects of the reliability and validity of the visual analogue scale. Pain 1983;16:87–101. Chaitow L. Muscle energy techniques. 2nd ed. London: Churchill Livingstone; 2003. Daum WJ. The sacroiliac joint: an underappreciated pain generator. The American Journal of Orthopedics 1995;6:475–8. Domholdt E. Physical therapy research: principles and applications. 2nd ed. Philadelphia, W.B: Saunders Company; 2000. Downey BJ, Taylor NF, Niere KR. Manipulative physiotherapists can reliably palpate nominated lumbar spine levels. Man Ther 1999;4:151–6. Dreyfuss P, Michaelsen M, Pauza K, McLarty J, Bogduk N. The value of medical history and physical examination in diagnosing sacroiliac joint pain. Spine 1996;21:2594–602. Freburger JK, Riddle DL. Using published evidence to guide the examination of the sacroiliac joint region. Phys Ther 2001;81:1135–43. Fritz JM, Wainner RS. Examining diagnostic tests: an evidence-based perspective. Phys Ther 2001;81:1546–64. Huskisson EC. Measurement of pain. Lancet 1974;2:1127–31. Keating J, Matyas TA, Bach TM. The effect of training on physical therapists’ ability to apply speciﬁed forces of palpation. Phys Ther 1993;73:38–46. Kokmeyer DJ, van der Wurff P, Aufdemkampe G, Fickenscher TCM. The reliability of multitest regimens with sacroiliac pain provocation tests. J Manip Physiol Ther 2002;25:42–8. Landis JR, Koch GG. The measurement of observer agreement of categorical data. Biometrics 1977;33:159–74. Laslett M, Williams M. The reliability of selected pain provocation tests for sacroiliac joint pathology. Spine 1994;19:1243–9. Laslett M, Young SB, Aprill CN, McDonald B. Diagnosing sacroiliac joints: a validity study of a McKenzie evaluation and sacroiliac provocation tests. The Australian Journal of Physiotherapy 2003;49:89–97. Laslett M, Aprill CN, McDonald B, Young SB. Diagnosis of sacroiliac joint pain: validity of individual provocation tests and composites of tests. Man Ther 2005;10:207–18. Levin U, Nilsson-Wikmar L, Stenstro¨m CH. Variability within and between evaluations of sacroiliac pain with the use of distraction testing. J Manip Physiol Ther 2005;28(9):688–95. Lindsay DM, Meeuwisse WH, Monney ME, Summersides J. Interrater reliability of manual therapy assessment techniques. Phys Ther Can 1995;43(3):173–81.

Maigne J-Y, Aivaliklis A, Pfefer F. Results of sacroiliac joint double block and value of sacroiliac pain provocation tests in 54 patients with low back pain. Spine 1996;21:1889–92. Meijne W, van Neerbos K, Aufdemkampe G, van der Wurff P. Intraexaminer and interexaminer reliability of the Gillets test. J Manip Physiol Ther 1999;22:4–9. O’Haire C, Gibbons P. Inter-examiner and intra-examiner agreement for assessing sacroiliac anatomical landmarks using palpation and observation: pilot study. Man Ther 2000;5:13–20. Potter NA, Rothstein JM. Intertester reliability for selected clinical tests of the sacroiliac joint. Phys Ther 1985;65:1671–5. Riddle DL, Freburger JK. Evaluation of the presence of sacroiliac joint region dysfunction using a combination of tests: a multicenter intertester reliability study. Phys Ther 2002;82:772–81. Schwarzer AC, Aprill CN, Bogduk N. The sacroiliac joint in chronic low back pain. Spine 1995;20:31–7. Simmonds MJ, Kumar S. Health care ergonomics. Part 1: the fundamental skill of palpation—a review and critique. International Journal of Industrial Ergonomics 1993;11:135–43. Slipman CW, Sterenfeld EB, Chou LH, Herzog R, Vresilovic E. The value of radionuclide imaging in the diagnosis of sacroiliac joint syndrome. Spine 1996;21:2251–4. Touissant R, Gawlik CS, Rehder U, Ru¨ther W. Sacroiliac dysfunction in construction workers. J Manip Physiol Ther 1999a;22: 134–8. Touissant R, Gawlik CS, Rehder U, Ru¨ther W. Sacroiliac joint diagnostics in the Hamburg construction workers study. J Manip Physiol Ther 1999b;22:139–43. Tullberg T, Blomberg S, Branth B, Johnsson R. Manipulation does not alter the position of the sacroiliac joint. Spine 1998;23: 1124–9. Walker JM. The sacroiliac joint: a critical review. Phys Ther 1992;72:903–16. van der Wurff P, Hagmeijer RHM, Meyne W. Clinical tests of the sacroiliac joint: a systematic methodological review. Part 1: reliability. Man Ther 2000;5:30–6. van der Wurff P, Buijs EJ, Groen GJ. A multitest regimen of pain provocation tests as an aid to reduce unnecessary minimally invasive sacroiliac joint procedures. Arch Phys Med Rehab 2006;87:10–4. Vincent-Smith B, Gibbons P. Inter-examiner and intra-examiner reliability of the standing ﬂexion test. Man Ther 1999;4:87–93. Weiselﬁsh-Giammatteo S, Giammatteo T. Integrative manual therapy for biomechanics: application of muscle energy and ‘‘beyond’’ technique: treatment of the spine, ribs, and extremities/by Sharon Weiselﬁsh Giammatteo and Thomas Giammatteo, vol. III. California: North Atlantic Books; 2003. Young S, Aprill C, Laslett M. Correlation of clinical examination characteristics with three sources of chronic low back pain. The Spine Journal 2003;3:460–5.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 57–62 www.elsevier.com/locate/math

Original article

What is the effect of taping along or across a muscle on motoneurone excitability? A study using Triceps Surae Caroline M. Alexandera,b,, Marie McMullana, Philip J. Harrisona a

Department of Physiology, University College London, Gower St, London WC1E 6BT, UK Department of Physiotherapy, Hammersmith Hospitals NHS Trust, Charing Cross Hospital, Fulham Palace Rd, London W6 8RF, UK

b

Received 9 August 2005; received in revised form 17 July 2006; accepted 27 August 2006

Abstract Taping along the skin overlying lower trapezius reduces motoneurone excitability in healthy subjects [Alexander, C.M., Stynes, S., Thomas, A., Lewis, J., Harison, P.J., 2003. Does tape facilitate or inhibit the lower ﬁbres of trapezius? Manual Therapy 8, 37–41]. It remains unclear whether this effect is: (a) speciﬁc to trapezius and (b) speciﬁc to the direction of application of the tape. In light of this, the excitability of another muscle was measured in order to see if these results were repeatable and independent of the muscle taped. Thus, the excitability of the medial and lateral gastrocnemius (MG and LG) and soleus (Sol) motoneurone pool was assessed using the Hoffman reﬂex (H reﬂex). The amplitude of this reﬂex was measured with the tape aligned across and then along the direction of the MG muscle ﬁbres. Tape aligned across the ﬁbres failed to affect motoneurone excitability (MG P ¼ 0.61, LG P ¼ 0.69, Sol P ¼ 0.17). Under tape and sports tape applied together aligned along the MG muscle reduced the excitability of both MG and LG (19% (P ¼ 0.01) and 13% (P ¼ 0.01), respectively). These observations suggest that any change to movement patterns with tape application cannot be explained by facilitation of the motoneurone excitability. r 2006 Elsevier Ltd. All rights reserved. Keywords: Tape; Gastrocnemius; Muscle

1. Introduction The application of tape has been widely used for many years by physiotherapists as a useful adjunct to prophylactic or rehabilitation programmes in the management of a wide variety of neuromusculoskeletal disorders. The effects of taping which are most commonly described include: inhibition of overactive synergists or antagonists, facilitation of underactive movement synergists, promotion of proprioception, optimisation of joint alignment, pain reduction and unloading of irritable neural tissue (Host, 1995; Morrisey, 2000). The technique of tape application is widely anecdotal and varies depending Corresponding author. Hammersmith Hospitals NHS Trust, Charing Cross Hospital, Fulham Palace Rd, London W6 8RF, UK. Tel.: +44 0207 679 0855; fax: +44 0208 846 7783. E-mail address: [email protected] (C.M. Alexander).

1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.08.003

upon the therapist applying it and the desired effect. The technique normally involves preparation of the skin, the application of a protective undertape followed by a nonelastic, rigid overtape which is used to apply tension over the underlying soft tissues. One common presumption, based in the main upon anecdotal evidence, is that tape facilitates and inhibits muscles depending upon the way it is applied. For example, tape applied under tension in the direction of the muscle ﬁbres, is thought to facilitate the underlying muscle (Morrisey, 2000). However, tape applied across the belly of the muscle, is thought to inhibit the muscle (Tobin and Robinson, 2000). The mechanisms by which these taping effects are achieved have limited scientiﬁc support and as yet, are inconclusive though most theories highlight the proprioceptive and mechanical effects with lesser effects hypothesized to be placebo (Simoneau, 1997; Hume and Gerrard, 1998, Alexander et al., 2003).

ARTICLE IN PRESS 58

C.M. Alexander et al. / Manual Therapy 13 (2008) 57–62

If tape does act by causing a change to afferent input, then motoneurone excitability must be affected. One way in which changes to motoneurone excitability can be explored is by measuring changes in spinal reﬂex amplitude. Indeed, if reﬂex amplitude were reduced with application of tape, then this would reﬂect a reduction of motoneurone excitability. Conversely, if reﬂex amplitude increases this would reﬂect a facilitation of motoneurone excitability. A spinal reﬂex that is commonly used to investigate motoneurone excitability is the H reﬂex. Indeed, this reﬂex has been previously used in healthy subjects to show that taping along the direction of the ﬁbres of lower trapezius decreases the amplitude of the lower trapezius H reﬂex (Alexander et al., 2003) and therefore produces an inhibition of motoneurone excitability. These investigations have been extended here to examine the generality of these ﬁndings. Thus, can these observations be extended to other muscles? To be speciﬁc, is this a rule that is independent of the muscle but dependant upon the direction of application of tape? To explore these questions, a muscle that is not commonly taped in this way was chosen—the triceps surae—in order to focus upon (a) the effect of the direction of the tape and (b) a muscle where no expectations of this application of tape pre-exist. The triceps surae are a muscle group for which the H reﬂex has been extensively used in neurophysiological testing and its behaviour is well understood (Pierrot-Deseilligny and Mazevet, 2000). To evoke its H reﬂex, the nerve supply to gastrocnemius is electrically stimulated. This has a number of effects (see Fig. 1). One effect is that the efferents to the muscle are stimulated. This efferent volley causes the muscle to contract resulting in a short latency motor response or M response. This is not a reﬂex. Another effect of the stimulus is that the afferents from the muscle are stimulated. This results in an impulse that monosynaptically connects to the motoneurones in the spinal cord. When these motoneurones ﬁre, the muscle will contract a second time, resulting in the H reﬂex. Thus, the aim of this study was to examine whether tape would affect the triceps surae H reﬂex in the same way as it affects the trapezius H reﬂex. Finally, the effect of taping along the muscle ﬁbres has been contrasted with those effects of taping across the muscle ﬁbres.

2. Method Subjects were recruited from the staff and students of the Department of Physiology, UCL. However, those who had suffered pain from any of their joints or muscles during the previous 2 years or those who had a rheumatological or neurological problem were excluded. With local ethical approval (University College London Ethics Committee) and informed consent, recordings were made from a group of 14 healthy subjects. 11

Fig. 1. A diagrammatic representation of the H reﬂex pathway (top) and the Soleus H reﬂex (bottom). This is a modiﬁcation of the diagram seen in Alexander et al. (2003): (A) electrical stimulation of a mixed peripheral nerve evokes two responses from the muscle it supplies. The ﬁrst response is the motor response (M response). This is from the direct stimulation of the motor axons. The second response (or H reﬂex) is evoked by stimulation of the afferent supply, which monosynaptically activates the efferent axons. (B) The soleus responses to electrical stimulation of the tibial nerve. This average is the result of 10 stimuli.

women and three men were recruited (age range: between 24 and 48 years). The excitability of the gastrocnemius motoneurone pool was measured using the gastrocnemius H reﬂex. Electrical 1 ms square wave pulses were delivered percutaneously every ﬁve seconds (Digitimer DS7A stimulator) to the tibial nerve. The anode (a metal plate smeared with electrode gel) was strapped over the distal surface of the quadriceps. The cathode (a Medicotest disc, pre-gelled electrode) was adhered to the skin over the tibial nerve in the popliteal fossa. Whilst sitting, surface EMG was recorded using adhesive electrodes placed edge to edge with the recording area 3 cm apart and positioned over the medial gastrocnemius (MG), lateral gastrocnemius (LG) and soleus (Sol) muscles (Williams et al., 1989; Simonsen and Dyhre-Poulsen, 1999) The EMG was ampliﬁed (Digitimer NL824) and ﬁltered (Neurolog NL125) with a bandwidth of 30 Hz–3 kHz. The data were converted from an analogue to a digital signal at a sampling frequency of 4 kHz (CED 1401) and stored for later analysis by CED Signal software (see Fig. 2; for details of the H reﬂex as an investigative tool see PierrotDeseilligny and Mazevet, 2000).

ARTICLE IN PRESS C.M. Alexander et al. / Manual Therapy 13 (2008) 57–62

59

Fig. 2. A diagrammatic representation of the recording (dashed arrows) and stimulating (full arrows) methods.

Using a single investigator who was not blinded during the procedure or investigation of the data, the tape was attached to the skin overlying MG in two ways. In one way the tape was positioned so that it was aligned to go across the direction of the muscle ﬁbres of MG. In the other way the tape was positioned so that it was aligned to go along the direction of these muscle ﬁbres. As the MG muscle ﬁbres descend with a lateral inclination, the tape was always at an angle to the horizontal or vertical (see Fig. 3). Two different types of tape were applied. Firstly, a protective elastic under tape was carefully applied to the skin overlying MG without creating tension or distortion of the skin. Secondly, a rigid non-elastic zinc oxide tape (‘sports’ tape) was applied directly over the under tape with enough tension to create crimping or puckering of the skin. The investigator was a senior physiotherapist experienced in the use of tape. In both variations of tape placement, the stimulus intensity was varied until a mid-amplitude H reﬂex was evoked with its accompanying M response. At this stimulus intensity a minimum of 40 stimuli were evoked: (i) before taping, (ii) with under tape aligned across or along the direction of the MG ﬁbres, (iii) with the addition of rigid tape positioned over the under tape and ﬁnally (iv) with both tapes removed. The peak to peak amplitude of all the M responses and the H reﬂexes were measured. The amplitude of these responses can be variable. One cause of this variability is that the stimulating probe can move in relation to the nerve. However, the M response is a useful measure of the stability of this effective stimulus. In particular, any variability of the amplitude of the M response reﬂects a change in the position of the stimulating electrode upon the nerve rather than a change to motoneurone excitability. Conversely, if the stimulating electrode is in a steady position to the nerve and the amplitude of the M response is steady, then any changes to the H reﬂex amplitude are due to the changing motoneurone excitability (Alexander et al., 2003). Thus, to ensure that any alteration in reﬂex amplitude was in response to the tape and not to ﬂuctuations in the effectiveness of the stimulus, only series of recordings in which the

Fig. 3. The location and orientation of tape applied to the skin overlying medial gastrocnemius. The picture to the left shows tape applied along the direction of muscle ﬁbres, while the picture to the right shows tape applied across the direction of the muscle ﬁbres.

M response was consistent were used. As the M responses were inconsistent in four subjects, the data from these subjects was discarded. The data from the remaining 10 subjects were analysed further. The amplitude of the M responses across all four conditions was measured and a one way analysis of variance was used to assess whether the M response was consistent throughout the recording period. If the M response varied such that the amplitude was signiﬁcantly different between tape conditions, the variable M responses along with the accompanying H reﬂex were ﬁltered out until the ANOVA demonstrated nonsigniﬁcant ﬂuctuations in amplitude of the M response. After ﬁltering out any erroneous M responses and their accompanying H reﬂexes, the EMG from the remaining subjects was averaged. This provided an average H reﬂex amplitude collected: (i) before application of tape; (ii) with under tape aligned along or across the muscle ﬁbres; (iii) with sports tape overlying the undertape; and (iv) after both tapes were removed. These amplitudes were then compared across the different experimental conditions both whilst taping along MG and then whilst taping across MG using a two-way ANOVA without replication. Where a paired Student’s t-test revealed no signiﬁcant difference between pre and post tape results, the amplitudes were averaged and a two-way ANOVA without replication was used to assess for differences between the combined pre/post taping condition, the under tape condition and with both tapes in place. A post hoc Tukey test revealed any signiﬁcant differences across these different experimental regimes. Finally, power calculations were performed (signiﬁcance set at 5%) to ensure that the data set was sufﬁciently extensive to support the conclusions being made.

ARTICLE IN PRESS 60

C.M. Alexander et al. / Manual Therapy 13 (2008) 57–62

3. Results 3.1. Taping along the fibres of MG Fig. 4 shows the effects of taping along the ﬁbres of MG. As can be seen in Fig. 4A the dominant feature of the recording is the large M response, which is then followed by the H reﬂex. The amplitude of the H reﬂex is indicated using the dashed horizontal lines. Upon application of the under tape the M response remains the same amplitude but, in this record, the H reﬂex appears to get smaller (Fig. 4B). Application of the sports tape reduces the amplitude of the H reﬂex substantially (75%, Fig. 4C). Removal of both tapes then results in the H reﬂex reverting back to its control amplitude (Fig. 4D). Note the M response remained constant throughout indicating that the effective stimulus remained stable throughout the recording period (Alexander et al., 2003). Thus, any change in reﬂex excitability was dependant on the application of tape. Results from another experiment can be seen graphically in Fig. 5 where the amplitude of each individual H

reﬂex has been plotted sequentially during the four conditions of taping. As can be seen, the application of sports tape along the muscle systematically inhibits the H reﬂex (open symbols). Thus, as was expected from previous investigations, taping along the ﬁbres of MG inhibited the H reﬂexes of MG. Fig. 6 summarizes the data from all 10 subjects. As indicated in Fig. 6, the application of under tape reduced the H reﬂex amplitude by 7%. However, this is not statistically signiﬁcant. On the other hand, application of the rigid sports tape produced a much more dramatic reduction in reﬂex amplitude which is statistically signiﬁcant (P ¼ 0.01). On average, the application of both under tape and rigid sports tape inhibited the reﬂex by 19714% (mean7SEM). Removal of both tapes produced a reversal of this inhibition. The amplitude of the H reﬂex was then not signiﬁcantly different when compared before and after application of the tape (P ¼ 0.25). Recordings were also made of LG and Sol H reﬂexes whilst taping MG. The LG H reﬂex reacted in a very similar way to the MG reﬂex, albeit in a reduced

Fig. 4. Averaged electromyograms from one individual illustrating the M response and H reﬂex evoked under four conditions: (A) the M response and H reﬂexes evoked before the application of tape; (B) with the under tape applied along the ﬁbres of medial gastrocnemius; (C) with the addition of sports tape; and (D) with both tapes removed. The downward arrows mark the H reﬂexes. The upward arrows mark the stimulus artefact. Note that the average H reﬂex amplitude varies while the M response amplitude remains constant.

Fig. 5. The effect of tape aligned along the ﬁbres of medial gastrocnemius in one individual. The amplitude of the M responses (closed symbols) remain stable whilst the H reﬂex amplitudes (open symbols) vary.

ARTICLE IN PRESS C.M. Alexander et al. / Manual Therapy 13 (2008) 57–62

61

Fig. 6. The mean amplitude of the medial gastrocnemius H reﬂex across four different conditions (7the average SEM for each condition, n ¼ 10). The amplitude of the H reﬂex was signiﬁcantly different with the application of the sports tape aligned along the ﬁbres of medial gastrocnemius compared to the pre and post tape conditions. The * indicates signiﬁcant results (where Po0.05).

Fig. 7. The effect of tape aligned across the ﬁbres of medial gastrocnemius upon the LG H reﬂex in one individual. The amplitude of the M responses (closed symbols) and H reﬂexes (open symbols) remain constant.

manner. That is, the reﬂex was inhibited by the application of the under tape and sports tape by 13712% (mean7SEM, P ¼ 0.01). The H reﬂex of Sol showed a similar trend but did not reach signiﬁcance (P ¼ 0.06). 3.2. Taping across the fibres of MG When taping across the ﬁbres of MG the tape failed to have an effect upon H reﬂex amplitude. Fig. 7 illustrates a typical result of taping across the MG upon the LG H reﬂex amplitude. The population data emphatically support this result. Thus, taping across the ﬁbres of

MG did not affect motoneurone pool excitability (MG P ¼ 0.61, LG P ¼ 0.69, Sol P ¼ 0.17).

4. Discussion The results of this study conﬁrm and extend previous observations that taping along the length of a muscle inhibits that muscle’s reﬂex excitability (Alexander et al., 2003). While the results of previous work are based on reﬂexes evoked in the trapezius muscle, the present work has extended this to a completely different group of muscles, triceps surae. In addition, the effects of taping

ARTICLE IN PRESS 62

C.M. Alexander et al. / Manual Therapy 13 (2008) 57–62

along and across the direction of the ﬁbres of MG have been assessed. In assessing the excitability of the motoneurones of triceps surae, changes in the amplitude of H reﬂexes of MG, LG and Sol were monitored. It has been assumed previously that the taping along or across the direction of the muscle ﬁbres would facilitate or inhibit H reﬂexes, respectively (Morrisey, 2000; Tobin and Robinson, 2000), as reﬂected by an increase or a decrease in H reﬂex amplitude. Contrary to these expectations, the results of this study indicate that taping along the direction of the ﬁbres of MG actually reduces the amplitude of the H reﬂex and thus inhibits triceps surae. However, taping across the direction of the ﬁbres of MG produces no change in excitability as reﬂected by no change in H reﬂex amplitude. On removal of both tapes, the H reﬂex reverts back to its original amplitude. Taping along the direction of the ﬁbres of a muscle has now been explored in both trapezius and gastrocnemius. In both studies taping inhibits both trapezius (Alexander et al., 2003) and gastrocnemius. While it is true that this has been found in only two muscles it cannot be claimed to be universally applicable. However, it is likely to be more wide spread in occurrence than merely the two muscles studied. As taping across the gastrocnemius muscle fails to alter reﬂex amplitude it seems reasonable to suggest that taping across the muscle does not have an effect upon motoneurone excitability in this muscle in healthy subjects. However, this does not suggest that the application of tape is not a useful technique in physiotherapy. Indeed, recently Lewis et al. (2005) demonstrated that taping altered posture, increased the range of movement of the shoulder and in patients with shoulder impingements, pain was experienced further into range. The results of the present investigations do however suggest that previous explanations of the mechanisms involved may need to be revised. Just how tape affects its inﬂuence upon motoneurone excitability remains open to speculation. Biomechanical and various neural mechanisms have been discussed (Host, 1995; Robichaud and Agostinucci, 1996; Morrisey, 2000). Indeed, in a previous investigation of the effect of tape upon trapezius excitability, possible mechanisms were also discussed (Alexander et al., 2003). One suggested mechanism is that tape along the muscle may shorten it, thus unloading the intrafusal muscle ﬁbres of the spindle. If tape does indeed shorten the intrafusal ﬁbres, this might result in a decrease in the tonic discharge rate of the spindle leading to a reduced facilitation of the motoneurone pool and subsequent reduction in the H reﬂex amplitude seen here. In addition in this case, as both LG and Sol share common Ia input from MG (Eccles et al., 1957), a reduction in Ia activity from the MG spindle should and did result in a decrease in H reﬂex amplitude of LG and Sol. Moreover, taping

across the muscle is unlikely to signiﬁcantly change muscle length and therefore is unlikely to alter muscle spindle ﬁring rates. In accordance with this, taping across the muscle does not change H reﬂex amplitude. The activation of cutaneous receptors has also been cited as another possible mechanism driving the effect of tape (Robichaud and Agostinucci, 1996). However, with the application of tape, one might expect that the inhibition seen here might also be present irrespective of the direction in which the tape is positioned. In addition, while some cutaneous receptors may be sensitive to the direction of pull of the tape, it seems unlikely that this directional sensitivity would be associated with the direction of the underlying muscle ﬁbres. In conclusion, while these experiments were not designed to study the mechanisms underlying the results described here, it is clear that the results are consistent with the hypothesis that the inhibition observed may be related to the unloading of muscle spindles during the muscle shortening that may occur during taping along the length of the muscle. However, this is not to deny that other factors may be involved, particularly in patients with different pathologies. It would therefore clearly be of interest to extend this work to different patient groups. References Alexander CM, Stynes S, Thomas A, Lewis J, Harrison PJ. Does tape facilitate or inhibit the lower ﬁbres of trapezius? Manual therapy 2003;8:37–41. Eccles JC, Eccles RM, Lundberg A. The convergence of monosynaptic excitatory afferents on to many different species of alpha motoneurones. The Journal of Physiology 1957;137:22–50. Host HH. Scapular taping in the treatment of anterior shoulder impingement. Physical Therapy 1995;75:803–12. Hume P, Gerrard D. Effectiveness of external ankle support. Bracing and taping in Rugby Union. Sports Medicine 1998;25:285–312. Lewis JS, Wright C, Green A. Subacromial impingement syndrome: the effect of changing posture on shoulder range of movement. Journal of Orthopaedic Sports Physical Therapy 2005;35(2):72–87. Morrisey D. Proprioceptive shoulder taping. Journal of Bodywork and Movement Therapies 2000;4:189–94. Pierrot-Deseilligny E, Mazevet D. The monosynaptic reﬂex: a tool to investigate motor control in humans. Interest and limits. Neurophysiologie Clinique 2000;30:67–80. Robichaud J, Agostinucci J. Air-splint pressure effect on soleus muscle alpha motorneuron reﬂex excitability in subjects with spinal cord injury. Archives of Physical Medicine and Rehabilitation 1996;77:778–82. Simoneau JA. Changes in ankle joint proprioception resulting from strips of athletic tape applied over the skin. Journal of Athletic Training 1997;32:141–7. Simonsen EB, Dyhre-Poulsen P. Amplitude of the human soleus H reﬂex during walking and running. The Journal of Physiology 1999;515(3):929–39. Tobin S, Robinson G. The effect of McConnell’s vastus lateralis inhibition taping tecnhique on vastus lateralis and vastus medialis obiquus activity. Physiotherapy 2000;86(4):173–83. Williams P, Warwick R, Dyson M, Bannister L. Myology. In: Williams P, Warwick R, Dyson M, Bannister L, editors. Gray’s anatomy. 37th ed. London: Churchill Livingstone; 1989. p. 648–9.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 63–67 www.elsevier.com/locate/math

Original article

Anterior positional fault of the ﬁbula after sub-acute lateral ankle sprains Tricia J. Hubbarda,, Jay Hertelb a

Department of Kinesiology, University of North Carolina, 9201 University City Blvd., Charlotte, NC 28223, USA b Kinesiology Program, University of Virginia, Charlottesville, VA 22904, USA Received in revised form 5 July 2006; accepted 26 August 2006

Abstract Recent evidence has suggested a positional fault of the ﬁbula exists in chronically unstable ankles. However, there has been little research examining positional faults after a sub-acute lateral ankle sprain (LAS). Our purpose was to measure the position of the distal ﬁbula in relation to the distal tibia in subjects with sub-acute LASs and to determine if there is a relationship between the amount of swelling and ﬁbular position. Eleven subjects with a unilateral sub-acute LAS and 11 healthy controls participated. The Wilcoxon signed rank test revealed a signiﬁcant ðp ¼ 0:008Þ difference within the ankles of the injured group. The sprained ankles had a mean ﬁbular position of 14.2+3.4 mm and were positioned signiﬁcantly more anteriorly than the contralateral uninjured ankles (17.0+3.2 mm). The Mann–Whitney test revealed a signiﬁcant difference ðp ¼ 0:045Þ between the sprained ankle and the side-matched limbs of the controls (16.8+2.3) Pearson product moment correlations revealed a strong positive correlation between ﬁbular position and swelling (r ¼ 0.793, p ¼ 0:004). Those ankles with more swelling had the most anteriorly positioned ﬁbulae. The ﬁbulae in sub-acutely sprained ankles appear to be positioned more anteriorly compared to the contralateral ankles. This positional fault may be maintained acutely by swelling. Published by Elsevier Ltd. Keywords: Ankle sprain; Fibula; Fluoroscopy; Tibioﬁbular joint

1. Introduction Lateral ankle sprains (LASs) are among the most common injuries suffered during athletic activities (Garrick, 1977). The anterior taloﬁbular ligament (ATFL) is reported to be the weakest and ﬁrst ligament injured in an ankle sprain (Brostrom, 1964). This is followed by the calcaneoﬁbular ligament (CFL) and the posterior taloﬁbular ligament (PTFL). Rupture of the ATFL occurs as an isolated injury in 66% of all ruptures of the ankle ligaments and in combination with a rupture of the CFL in another 20% (Brostrom, 1964).

Corresponding author. Tel.: +1 704 687 6202; fax: +1 704 687 3350. E-mail address: [email protected] (T.J. Hubbard).

1356-689X/$ - see front matter Published by Elsevier Ltd. doi:10.1016/j.math.2006.09.008

After an acute LAS, attention is typically paid to the lateral ligamentous structures. However, in addition to injury of the ligamentous structures of the ankle, hypomobility may develop after injury (Denegar and Miller, 2002; Hertel, 2002). This hypomobility has been often been demonstrated in the form of a positional fault (Mulligan, 1995). The relative void in the literature regarding positional faults gives the impression they do not contribute to residual symptoms or to increasing the risk for reinjury. Mulligan (1995) has proposed that some individuals diagnosed with LASs experience an anterior positional fault of the distal ﬁbula on the tibia. Since the original hypothesis proposed by Mulligan (1995), several studies have examined ﬁbular position in subjects after an acute ankle sprain or with chronic ankle instability (Kavanagh, 1999; Mavi et al., 2002; Eren et al., 2003;

ARTICLE IN PRESS 64

T.J. Hubbard, J. Hertel / Manual Therapy 13 (2008) 63–67

Berkowitz and Kim, 2004; Hubbard et al., 2006). Results have suggested both anterior (Kavanagh, 1999; Mavi et al., 2002; Hubbard et al., 2006) and posterior (Eren et al., 2003; Berkowitz and Kim, 2004) positional faults although measurement difference likely contribute to those contradicting ﬁndings. Our purpose was to examine the position of the distal ﬁbula in relation to the distal tibia in subjects with sub-acute LASs and to determine if a relationship exists between the amount of swelling and ﬁbular position.

2. Methods 2.1. Subjects Eleven recreationally active subjects with sub-acute LASs (5 males and 6 females, age ¼ 20.170.94 yr, mass ¼ 75.5721.5 kg, ht ¼ 172.1714.1 cm) and 11 subjects with no previous history of ankle injury (ﬁve males and six females, age ¼ 20.070.89 yr, mass ¼ 71.47 11.2 kg, ht ¼ 170.678.6 cm), participated in this study. A certiﬁed athletic trainer initially examined the injured subjects. Subjects with a LAS were then referred to the researchers and offered the opportunity to participate in the study. The mean grade of ankle sprain suffered by subjects was 1.9+0.53 on a 3-category scale. Fluoroscopic images were taken between 2 and 10 days after injury (mean ¼ 4.7+2.5 days). All healthy subjects were recruited from undergraduate courses until we had an equal number of males and females in both the injured and healthy groups. The sides were assigned so that there were an equal proportion of right and left involved ankles in the sub-acute LAS and control groups.

abduction or adduction in the frontal plane. The subject’s knee was in full extension. Towels were then placed under the leg to maintain neutral position. The foot of the top leg was then placed on the image intensiﬁer of the ﬂuoroscope. The foot was passively positioned in maximum dorsiﬂexion for testing (Hubbard et al., 2006). Subjects were observed to ensure that no rotations of the lower extremity occurred during testing. The order of which ankle was tested ﬁrst was randomly assigned by 1 examiner. The same examiner positioned and made measurements on all subjects. A lateral image was then recorded. The same procedure was then followed for the opposite leg. A radiographic marker (4.5 mm long) was placed on all ankles to correct for variances in magniﬁcation. After the images were printed, measurements were made to determine the position of the ﬁbula (Fig. 1). A tape measure was used to measure the distance between the anterior margin of the ﬁbula and the anterior margin of the tibia in millimetres. Measurements were made perpendicular to a line drawn vertically from the most anterior point of the tibia. To measure swelling, the ﬁgure-of eight method was used. High intratester and intertester reliability have been reported (ICC ¼ 0.99 for both) (Tatro-Adams et al., 1995). Subject’s ankles were positioned in neutral. The beginning of the tape measure was placed midway between the anterior tibialis tendon and the lateral malleolus. The tape measure was then placed around the foot, distal to the navicular tuberosity, across the arch proximal to the base of the ﬁfth metatarsal, across the

2.2. Instrumentation The position of the distal ﬁbula in relationship to the distal tibia was determined by taking a lateral image of both ankles. A Mini 6600 Fluoroscope with a digital mobile C-arm (OEC Medical Systems Inc., Salt Lake City, UT) recorded images. 2.3. Procedures The position of the ﬁbula was measured with ﬂuoroscopy. Using previously reported methods (Hubbard et al., 2006). Test–retest reliability (ICC (3, 1) ¼ 0.98, SEM ¼ 0.64 mm) and intratester reliability (ICC (3, 1) ¼ 0.92, SEM ¼ 0.72 mm) were also previously reported (Hubbard et al., 2006). Subjects were positioned side lying on the treatment table. The subject’s posterior thigh was positioned against a bolster to ensure the hip was maintained in a neutral position in the sagittal plane. Additionally, a ﬂuid inclinometer was placed on the lateral joint line of the knee to ensure the leg was kept in neutral and not in

Fig. 1. Fibular position was measured by the distance between the anterior edge of the distal ﬁbula and the anterior edge of the distal tibia was measured in centimeters and then converted to millimetres. Reprinted from Hubbard TJ, Hertel J, Sherbondy P. Fibular position in Individuals With Self-Reported Chronic Ankle Instability. J Orthop Sports Phys Ther. 2006; 36(1):3–9, with permission of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association.

ARTICLE IN PRESS T.J. Hubbard, J. Hertel / Manual Therapy 13 (2008) 63–67

anterior tibialis tendon, and back around the heel. The measurement was recorded in centimeters for both the right and left ankles. 2.4. Data analysis

65

ﬁbular position difference was explained by variance in ankle girth difference.

4. Discussion

3. Results The Wilcoxon signed rank test revealed signiﬁcant differences within the ankles of the sub-acute LAS group ðp ¼ 0:008Þ. The injured ankle of the sub-acute LAS group had a mean ﬁbular position of 14.2+3.4 mm posterior to the anterior edge of the tibia compared to 17.0+3.2 mm for the contralateral uninjured ankle. There were no signiﬁcant side-to-side differences within the healthy group ðp ¼ 0:563Þ. The Mann–Whitney test revealed a signiﬁcant difference between the injured ankle of the sub-acute LAS group and the matched side of the control group ðp ¼ 0:045Þ, there were no signiﬁcant differences between the uninjured ankle of the sub-acute LAS group and the matched side of the control group ðp ¼ 0:438Þ (Table 1). Fibular position for all 22 subjects is listed in Table 2. Pearson product moment calculations revealed a statistically signiﬁcant positive correlation between the side-to-side differences in ﬁbular position and swelling ðr ¼ 0:793; p ¼ 0:004Þ. (Fig. 2) 63% of variance in the

There was a statistically signiﬁcant difference in ﬁbular position for the subjects with sub-acute LAS. Our ﬁndings suggest an anterior positional fault was Table 2 Fibular position (mm) for all 22 subjects Sub-acute LAS group

Control group

Sub-acute ankle Opposite ankle

Matched ankle

Opposite ankle

12.00 9.50 17.00 19.00 17.20 15.00 13.30 8.10 16.70 15.00 13.00

15.00 14.20 17.60 13.50 15.10 20.20 15.00 19.30 17.60 18.90 18.60

14.40 15.90 15.00 13.30 13.80 20.70 14.50 20.70 17.20 16.50 19.40

12.00 20.50 20.80 19.00 21.10 17.00 14.40 13.50 19.40 15.30 14.70

8 7 Ankle Girth Difference

The data did not ﬁt a normal distribution so nonparametric statistics were calculated. The Wilcoxon signed rank test was used to test for side-to-side differences within both the sub-acute LAS group and the control group. Mann–Whitney tests were used to test for differences between the injured ankle of the sub-acute LAS group and the side matched ankle of the control group, and the uninjured ankle of the sub-acute LAS group and the side-matched ankle of the control group. Bivariate correlations using Pearson product moment calculations were made between the amount of side-toside difference in swelling and the corresponding difference in ﬁbular position for the sub-acute LAS group. The level of signiﬁcance was set a priori at po0:05 for all analyses.

6 5 4 3 2 1 0 0

2

4 6 8 Fibular Position Difference

10

12

Fig. 2. Pearson product moment calculations* between the difference in ankle circumference and ﬁbular position difference for the sub-acute lateral ankle sprain group ðn ¼ 11Þ. Girth measures are in cm, and ﬁbular position measures are in mm.

Table 1 Means, standard deviations, and range of ﬁbular positiona for a group with a sub-acute lateral ankle sprain (LAS) ðn ¼ 11Þ and a control group ðn ¼ 11Þ Measurement

Fibular position (mm) a

Sub-acute LAS group

Control group

Sub-acute ankle

Opposite ankle

Matched ankle

Opposite ankle

14.273.4 (8.1–19.0)

17.073.2 (12.0–21.1)

16.872.3 (13.5–20.2)

16.572.7 (13.3–20.7)

The position of the ﬁbula is deﬁned as the distance between the anterior edge of the distal tibia and anterior edge of the distal ﬁbula measured with a ﬂuoroscopic image.

ARTICLE IN PRESS 66

T.J. Hubbard, J. Hertel / Manual Therapy 13 (2008) 63–67

present in those with sub-acute LAS when compared to their contralateral ankles and the side-matched controls. Additionally our effect size was high (0.91) within the ankles of the sub-acute LAS group, and high (1.15) between the involved ankles of the sub-acute and matched control ankles. Based on our effect sizes, we believe the altered position of the ﬁbula was enough for clinical meaningful effects. It is important to note that not all subjects had an anteriorly displaced ﬁbula. Two of the 11 subjects (18%) with a sub-acute ankle sprain did not have an anteriorly positioned ﬁbula. Anterior positional fault of the distal ﬁbula may be present in some but not all subjects after an ankle sprain (Table 2). Our data also suggests that those participants with greater differences in ﬁbular position had more swelling as measured by differences in ankle girth. As the difference in ankle girth increased the difference in ﬁbular position between the involved and uninvolved ankles also increased. Additionally, 63% of the variance in ﬁbular position difference is explained by the variance in ankle girth. Swelling immediately after injury may maintain ﬁbular displacement. However, further research is necessary to examine the relationship of swelling and ﬁbular position. Those with more swelling may have had a more severe ankle injury which may also account for the ﬁbular position measures. Both anterior and posterior positional faults have been reported after acute ankle sprains and in those with CAI. Methodological differences may account for differences in ﬁndings. Kavanagh (1999) examined this hypothesis in a series of cases. She hypothesized there would be a greater range of anterior-posterior movement possible at the distal ﬁbula if a positional fault of the ﬁbula occurred after an ankle sprain. She reported a signiﬁcantly greater amount of movement occurred in one third of the subjects with acute ankle sprains (Kavanagh, 1999). More objective methods were used by Mavi et al. (2002) and Hubbard et al. (2006). Both measured the distance between the anterior margin of the ﬁbula and the anterior margin of the tibia on radiographic images. They both reported the ﬁbula was positioned more anteriorly in the injured group than in the control group. Despite this evidence of anteriorly positioned ﬁbula, three recent studies reported the ﬁbula to be posteriorly positioned in those with lateral ankle instability. Scranton et al. (2000), Eren et al. (2003), and Berkowitz and Kim (2004) examined the position of the ﬁbula in subjects with acute ankle sprains and ankle instability. The axial malleolar index was computed from CT scans. The results from these studies suggest the ﬁbula may be positioned posteriorly in relation to the medial malleolus after ankle sprain (Scranton et al., 2000; Eren et al., 2003; Berkowitz and Kim, 2004). The authors measured the relationship in a transverse plane at the talocrural joint. The major limitation is that the measurement of ﬁbular position is based on the position

of the talus. Previous research has reported decreased posterior mobility of the talus after an ankle sprain (Denegar et al., 2002). This decreased mobility could have been caused by an anteriorly positioned talus. If the talus is positioned anteriorly, the ﬁbula would have appeared to be positioned posteriorly. Studies (Mavi et al., 2002 and Hubbard et al., 2006) that reported an anteriorly positioned ﬁbula have measured ﬁbular position in relation to the tibia without consideration of talar position. This may explain the different results and comparison of these different methods of assessing distal ﬁbula position is certainly warranted. Additionally, a limitation of using ﬂuoroscopy is we do not currently know the validity of the measure. The ﬂuoroscope records a 2-dimentional image, which limits the ability to identify and measure any rotations of the ﬁbula that may occur (Hubbard et al., 2006). Distal ﬁbular mobility and position should be examined after acute ankle injury. If the distal ﬁbula is positioned anteriorly, manual therapy techniques such as ﬁbular mobilization should be utilized. Since the goal is to increase the range of motion, a grade III or IV technique is used (Mulligan, 1995). The patient is positioned supine on a treatment table. The clinician places the palm of their hand on the lateral malleolus. A posterior mobilization is then applied to the lateral malleolus (Mulligan, 1995). Typically 15 impulses are applied to the joint. Additionally, the posterior ﬁbular glide can be combined with active ankle inversion (Greenman, 1996). Treatments typically last until full physiologic and arthrokinematic motion are regained. Previous research has demonstrated positive outcomes with distal ﬁbula and talar mobilization after ankle sprains. O’Brien and Vicenzino (1998), applied posterior ﬁbular mobilizations in two subjects after an acute ankle sprain. They reported immediate reduction in pain, increases in range of inversion, improved outcome, and improvements in function. Green et al. (2001) demonstrated that subjects treated with anterior to posterior mobilizations of the talus received fewer treatments to obtain pain-free dorsiﬂexion range of motion. These results were later supported by Collins et al. (2004) who also reported signiﬁcant immediate improvement in dorsiﬂexion range of motion after mobilization. Further research examining the effect of mobilization techniques at the distal tibioﬁbular joint is needed in larger samples.

5. Conclusion We identiﬁed an anteriorly positioned distal ﬁbula in individuals with sub-acute LAS. We do not currently know if altered ﬁbular position was a predisposing factor to injury. It appears that swelling maintains ﬁbular displacement acutely.

ARTICLE IN PRESS T.J. Hubbard, J. Hertel / Manual Therapy 13 (2008) 63–67

References Berkowitz CM, Kim DH. Fibular position in relation to lateral ankle instability. Foot and Ankle International 2004;25:318–21. Brostrom L. Sprained ankles: I, Anatomic lesions on recent sprains. Acta Chirurgica Scandinavica 1964;128:483–95. Collins N, Teys P, Vicenzino B. The initial effects of a Mulligan’s mobilization with movement technique on dorsiﬂexion and pain in subacute ankle sprains. Manual Therapy 2004;9:77–82. Denegar CR, Miller SJ. Can chronic ankle instability be prevented? Rethinking management of lateral ankle sprains. Journal of Athletic Training 2002;37:430–5. Denegar CR, Hertel J, Fonseca J. The effect of lateral ankle sprain on dorsiﬂexion range of motion, posterior talar glide, and joint laxity. Journal of Orthopaedic and Sport Physical Therapy 2002;32:166–73. Eren OT, Kucukkaya M, Kabukcuoglu Y, Kuzgun U. The role of a posteriorly positioned ﬁbula in ankle sprain. American Journal of Sports Medicine 2003;31:995–8. Garrick JG. The frequency of injury, mechanism of injury, and epidemiology of ankle sprains. American Journal of Sports Medicine 1977;5(2):241–2. Green T, Refshauge K, Crosbie J, Adams R. A randomized controlled trial of a passive accessory joint mobilization on acute ankle inversion sprains. Physical Therapy 2001;81:984–94. Greenman PE. Principles of manual medicine. 2nd ed. Baltimore MD: Williams & Wilkins; 1996. p. 434–9.

67

Hertel J. Functional anatomy, pathomechanics, and pathophysiology of lateral ankle instability. Journal of Athletic Training 2002;37: 364–75. Hubbard TJ, Hertel J, Sherbondy P. Anterior ﬁbular displacement with chronic ankle instability. Journal of Orthopaedic and Sport Physical Therapy 2006;1:3–9. Kavanagh J. Is there a positional fault at the inferior tibioﬁbular joint patients with acute or chronic ankle sprains compared to normals? Manual Therapy 1999;4:19–24. Mavi A, Yildirim H, Gunes H, Pestamalci T, Gumusburun E. The ﬁbular incisura of the tibia with recurrent sprained ankle on magnetic resonance imagining. Saudi Medical Journal 2002;23: 845–9. Mulligan BR. Manual therapy: ‘‘NAGS,’’ ‘‘SNAGS,’’ ‘‘MWMS,’’ etc. 3rd ed. Wellington, New Zealand: Plane View Services LTD; 1995. p. 95–101. O’Brien T, Vicenzino B. A study of the effects of Mulligan’s mobilization with movement treatment of lateral ankle pain using a case study design. Manual Therapy 1998;3:78–84. Scranton Jr PE, McDermott JE, Rogers JV. The relationship between chronic ankle instability and variations in mortise anatomy and impingement spurs. Foot and Ankle International 2000;21: 657–64. Tatro-Adams D, McGann SF, Carbone W. Reliability of the ﬁgure-ofeight method of ankle measurement. Journal of Orthopedic and Sports Physical Therapy 1995;22:161–3.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 68–74 www.elsevier.com/locate/math

Original article

The active straight leg raising test (ASLR) in pregnant women: Differences in muscle activity and force between patients and healthy subjects M. de Groota,, A.L. Pool-Goudzwaarda,b,c, C.W. Spoora, C.J. Snijdersa a

Department of Biomedical Physics and Technology, University Medical Center Rotterdam, Erasmus MC, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands b Medical Center Impact Zoetermeer, The Netherlands c Cobra, Centre for Research on Musckuloskeletal System, Amersfoort, The Netherlands Received 9 July 2005; received in revised form 10 July 2006; accepted 30 August 2006

Abstract Pregnancy-related low back and pelvic pain (PLBP) is a frequent complication of pregnancy. Although pathological mechanisms underlying PLBP are obscure, dysfunction of the sacroiliac joints (SI-joints) seems to play an important role. A cross-sectional study was performed on 24 pregnant women with and without PLBP. The objective was to determine muscle activation patterns of trunk and leg muscles during the active straight leg raising test (ASLR) and static hip ﬂexion, and to determine maximal hip ﬂexion force at 0 and 20 cm leg raise height. Moreover, the effort to raise the leg was scored. The measurements resulted in several signiﬁcant differences between the patients and healthy controls; among others (a) patients scored subjectively more effort during ASLR, (b) at both 0 and 20 cm leg raise height patients had less hip ﬂexion force, and (c) patients developed more muscle activity during ASLR. Since pregnant women with PLBP developed a higher muscle activity during ASLR with a signiﬁcantly lower output at 0 and 20 cm than healthy pregnant women, it could be proposed that the ASLR demonstrates a disturbed load transfer across the SI-joints in this population. r 2006 Elsevier Ltd. All rights reserved. Keywords: Active straight leg raising test; Pregnancy; Low back and pelvic pain; Sacroiliac joint

1. Introduction Pain in the lumbar spine and pelvic region is a frequent complication of pregnancy and delivery. The prevalence of pregnancy-related low back and pelvic pain (PLBP) varies widely from 14.2% to 56% (Mantle et al., 1977; Fast et al., 1987; Berg et al., 1988; O¨stgaard et al., 1991, 1994, 1996; Orvieto et al., 1994; Heiberg-Endresen, 1995; Wergeland and Strand, 1998; Bjo¨rklund et al., 1999; Larsen et al., 1999; Albert et al., 2000, 2001; Wu et al., 2004). The pain is mainly located in the sacral area and Corresponding author. Tel.: +31 10 4087389; fax: +31 10 4089463.

E-mail address: [email protected] (M. de Groot). 1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.08.006

the area of the symphysis pubis with or without radiation to the groins, thighs, buttocks and coccygeus region (Fast et al., 1987; Kristiansson et al., 1996; Mens et al., 1996; O¨stgaard et al., 1996; Perkins et al., 1998; Ro¨st et al., 2004). Several daily activities, like standing, sitting, forward bending, lifting, climbing stairs and walking, tend to increase the pain (Fast et al., 1987; Kristiansson et al., 1996; Mens et al., 1996; Ro¨st et al., 2004). The pain is often quite mild but in 6–15% it is considered to be severe, interfering with daily life activities (Mantle et al., 1977; Berg et al., 1988; Heiberg-Endresen, 1995; Bjo¨rklund et al., 1999; Wu et al., 2004). Pathological mechanisms underlying PLBP are a matter of debate. According to several authors, dysfunction

ARTICLE IN PRESS M. de Groot et al. / Manual Therapy 13 (2008) 68–74

of the sacroiliac joints (SI-joints) plays an important role in PLBP (Sands, 1958; Berg et al., 1988; Snijders et al., 1995b). The primary function of these joints is to transfer the loads from the upper part of the body to the legs and vice versa (Snijders et al., 1993a). SI-joint dysfunction is ascribed to instability, hyper- or hypolaxity, hyper- or hypomobility or altered stiffness of the joint (Walker 1992; Harrison et al., 1997; O’Sullivan et al., 2002; Bussey et al., 2004; Hungerford et al., 2004). Mechanical stability, the ability of a joint to bear loading without uncontrolled displacements (PoolGoudzwaard et al., 2003), is very important in the functioning of the SI-joint. The mechanical stability of the SI-joints depends on speciﬁc anatomic features (form closure) (Vleeming et al., 1990a, b) and on tension of ligaments and muscles crossing the pelvic joints (force closure) (Snijders et al., 1993a, b). Muscles with a transverse orientation can produce forces that cross the SI-joints in the appropriate direction to produce force closure. These especially include the gluteus maximus, the internal oblique abdominal and the transverse abdominal muscles (Snijders et al., 1995a; Richardson et al., 2002; Hungerford et al., 2003). The role of proprioception and motor control in the stability of the lumbar spine and pelvic region has been recognized (Hides et al., 1996; Hodges and Richardson, 1996; O’Sullivan et al., 2002; Wu et al., 2002; Hungerford et al., 2003). It is important to diagnose the SI-joint (dys)function properly in order to treat the problem in an appropriate way. Diagnosing SI-joint function is very difﬁcult because the joint is complex, as it forms a functional unity with the symphysis pubis and the ﬁfth lumbar vertebra. Traditionally, diagnosis of SI-joint function is based on a quality history and manual examination (Dreyfuss et al., 1994). It was hypothized that symptoms in PLBP are caused by overloading of ligaments of the pelvic ring and/or lumbopelvic junction during activities in which loads have to be transferred between legs and trunk (Mens et al., 1999; Snijders et al., 1993a). The active straight leg raising test (ASLR) is a check for this system (Mens et al., 2001). In this test, performed in supine position, a subject has to raise one leg, with the knee extended, 20 cm above the couch (Mens et al., 1999). The test is scored, only by the patient, rating the impairment on a 6-point Likert scale. (Mens et al., 1999). The ASLR is a valid and reliable test to discriminate between patients with PLBP and healthy subjects and to test the severity of PLBP (Mens et al., 2001, 2002). However, objective measurements are lacking. There is some evidence that an impaired ASLR is associated with laxity of the pelvic joints. In PLBP, a signiﬁcant correlation is found between an impaired ASLR and radiographically measured laxity (the

69

amount of motion in a joint that results from forces or moments, giving an indication of joint compression) in the pubic symphyses during the Chamberlain test (Chamberlain, 1930). During this test radiographs are made of the pubic symphysis while standing on one leg, alternating left and right, on a small bench with the other leg hanging passively beside the bench (Chamberlain, 1930). During a positive Chamberlain test both a widening of the pubic symphyses as a forward rotation of the innominate on the side of the passively hanging leg is visible (Mens et al., 1999). This is in line with the ﬁnding of Hungerford et al. (2004) that during single leg loading, at the side of support, in subjects with SI-joint pain, the innominate rotated forwards in contrast to the backwards rotation in healthy control subjects (Hungerford et al., 2004). Forwards rotation of the iliac bone, during load transfer could be indicative of failure of the force closure mechanism and a hampered load transfer through the pelvis. The aim of the present study was to determine the muscle activities of the rectus femoris (RF), abdominal obliques, psoas major (PM) and adductor longus (AL) muscle during the ASLR as well as during static hip ﬂexion. These muscles are involved in force closure and hip ﬂexion. Unfortunately, it was not possible to measure the muscle activity of the internal oblique abdominal and the transverse abdominal muscles with surface electrodes neither was it possible to measure the gluteus maximus muscle as the subject was lying supine. Also, the effort to raise the leg and the maximal muscle force that can be statically developed for hip ﬂexion were measured. These measures can lead to further understanding of the ASLR in pregnant women. By comparing women with and without PLBP, this can give more insight into management of the problem of PLBP.

2. Material and methods 2.1. Participants The study was performed on 24 pregnant women with an age between 20 and 40 years and a gestational age of 12–40 weeks. A classiﬁcation was made in two groups: the ﬁrst group comprised 11 patients with PLBP, the second group comprised 13 healthy controls without PLBP. The exclusion criteria for both groups were: a history of low back and pelvic pain before pregnancy; fracture, neoplasm or previous surgery of the lumbar spine, the pelvic girdle, the hip joint or the femur; or a systemic disease of the locomotor system. The Medical Ethical Committee of the Erasmus MC approved the protocol. All subjects gave written informed consent.

ARTICLE IN PRESS 70

M. de Groot et al. / Manual Therapy 13 (2008) 68–74

2.2. Procedure and instrumentation

2.5. External hip flexion force

In this cross-sectional study, every woman performed the ASLR in supine position with straight legs and feet 20 cm apart. The instruction to the women was ‘‘Try to raise your legs, one after the other, 20 cm above the couch without bending the knees’’ (Mens et al., 1999). The velocity of raising the leg was not prescribed. The raising height was deﬁned by placing the force gauge 20 cm above the ankle joint (Fig. 1). Before performing the ASLR in all women body weight, BMI and leg length on both sides (from top of trochantor major of the femur to the caudal side of the lateral malleolus of the ankle) were measured.

The maximal external force the woman can statically develop during at least 0.5 s for hip ﬂexion with a straight knee was measured just above the ankle joint with the leg still lying on the examination table (0 cm position) and at the end of the ASLR (20 cm position) (Fig. 1). For the recording of the maximal external force a digital force gauge was used (model 9200, Aikoh Engineering CO., LTD, Osaka, Japan), which was connected with a Porti system (Twente Medical System International BV, The Netherlands). The read-out was done with LabView 7.1. 2.6. Muscle activity

2.3. Questionnaires All women completed a questionnaire to assess several sociodemographic data and the Dutch version of the Quebec Back Pain Disability Scale (QBPDS) (Kopec et al., 1995; Schoppink et al., 1996). The QBPDS is a 20-item self-administered instrument designed to assess the functional disability. It asks the subject to rate her degree of difﬁculty in performing each activity from 0 (not difﬁcult at all) to 5 (unable to do). Scores for the 20 items are summed; a higher rating indicates greater functional disability. 2.4. Effort Effort during the ASLR was scored by all women on a six-point Likert scale: 0 ¼ not difﬁcult at all, 1 ¼ minimally difﬁcult, 2 ¼ somewhat difﬁcult, 3 ¼ fairly difﬁcult, 4 ¼ very difﬁcult, 5 ¼ unable to perform. The scores of both legs were added, so that the summed score ranged from 0 to 10 (Mens et al., 2001).

Disposable pre-gelled, self-adhesive surface EMG electrodes (Ag/AgCl discs) were placed, as advised by Delagi et al. (1994), at the left and right side of the body at the following positions: RF: on the anterior aspect of the thigh, midway between the superior border of the patella and the anterior superior iliac spine; AL: 5 cm distal to the pubic bone; external oblique abdominal (EO): midway between the highest point in the iliac crest and the anterior superior iliac spine; PM: 2 cm lateral to the femoral artery and 1 cm below the inguinal ligament. A reference electrode was placed over the right lateral malleolus of the ﬁbula. All electrodes were placed with an interelectrode distance of 20 mm and aligned parallel to the underlying muscle ﬁbres. EMG values at maximal voluntary contraction (MVC) were obtained with manually applied resistance to each muscle (Kendall and Kendall Mc Creary, 1986). The resistance given by the examiner was built up and the highest muscle contraction that was held for at least 0.5 s was taken to the MVC for each muscle. EMG recordings were made during the ASLR as well as during the determination of the maximal external force at 0 and 20 cm raising height and normalized to the MVC. The recordings were done with the Porti data acquisition system. All EMG signals were band-pass ﬁltered at 10–500 Hz and sampled at 1000 Hz by using a 22 bit analogue-digital converter. The digitized signals were full wave rectiﬁed and low-pass ﬁltered using a linear envelope ﬁlter. The data were stored on a computer for later analysis in LabView 7.1. 2.7. Data analysis

Fig. 1. Equipment to deﬁne 20 cm raising height during the ASLR and to measure the hip ﬂexion force.

Statistical analyses were performed with the SPSS software package (SPSS Inc., 233 S. Wacker Drive, Chicago, IL 60606, Version 11.0). Within both groups, the paired-samples t-test was used to measure if there was any difference in results between the left and right sides for the control subjects or between the asymptomatic and symptomatic sides for the women with PLBP.

ARTICLE IN PRESS M. de Groot et al. / Manual Therapy 13 (2008) 68–74

To test differences between the groups, the independent sample t-test was used. For all tests, the alpha level was set at 0.05.

Table 2 Clinical ﬁndings

Table 1 Sociodemographic data of the subjects participating in the study

Age (years) Length (m) Weight before pregnancy (kg) BMI before pregnancy (kg/m2) Weight at the moment (kg) BMI at the moment (kg/ m2) Leg length (cm) Number of previous pregnancies Gestational age (weeks)

Non-PLBP (n ¼ 13) Mean (SD)

PLBP (n ¼ 11) Mean (SD)

P-value

31.7 (4.9) 1.70 (0.08) 70.1 (10.2)

30.0 (3.8) 1.68 (0.06) 68.1 (10.1)

0.361 0.517 0.642

24.4 (4.1)

23.8 (3.1)

0.746

79.2 (12.8)

72.4 (19.1)

0.318

27.5 (4.6)

27.2 (3.0)

0.821

88.8 (5.4) 2.1 (1.3)

88.9 (3.1) 1.9 (0.9)

0.967 0.728

26.8 (6.3)

26.6 (7.3)

0.949

PLBP (n ¼ 11) Mean (SD)

P-value

20.0 (14.3) 0.9 (1.1)

50.2 (17.7) 3.9 (2.0)

0.000a 0.000a

129.0 (26.3)

83.5 (31.8)

0.000a

84.7 (23.0)

42.4 (19.9)

0.000a

34.7 (9.5)

50.6 (7.0)

0.000a

Non-PLBP (n ¼ 13) Mean (SD)

3. Results The study concerned 24 pregnant women, comprising 11 women with PLBP and 13 women without. Sociodemographic data of both groups are given in Table 1, with no signiﬁcant differences between the two groups. Women with PLBP scored signiﬁcantly higher (mean ¼ 3.9, SD ¼ 2.0) at the subjective score on the ASLR test than the healthy controls (mean ¼ 0.9, SD ¼ 1.1). The QBPDS score in women with PLBP ranged from 18 to 78, the mean score was 50.2 (SD ¼ 17.7). For the healthy controls, the QBPDS score ranged from 0 to 46, with a mean score of 20.0 (SD ¼ 14.3). In both groups, no statistical differences were found in muscle activity and hip ﬂexion force between the left and right sides or between the asymptomatic and symptomatic sides, so the results were averaged. The clinical ﬁndings are summarized in Table 2. Healthy controls achieved signiﬁcantly more hip ﬂexion force at both 0 and 20 cm than the women with PLBP. Both groups achieved less hip ﬂexion force at 20 cm than at 0 cm. Subjects without PLBP delivered 129.0 N at 0 cm leg raise height and 84.7 N at 20 cm; this is a decrease of 34.7%. Subjects with PLBP showed with 50.6% a signiﬁcantly greater decrease in force (Table 2). During the ASLR and the maximal external force measurements, the activity of the muscles was measured and normalized to the MVC. The muscles are divided in the homolateral and heterolateral side. Homolateral means at the side of the raised leg and heterolateral is the opposite side. During the ASLR the women with

71

Quebec score Subjective score of ASLR test Hip ﬂection force at 0 cm (N) Hip ﬂexion force at 20 cm (N) Decrease in hip ﬂexion force (%) a

Signiﬁcant difference at a ¼ 0:05.

PLBP used more muscle activity compared to the healthy controls. The differences of the homolateral RF (P ¼ 0.001), PM (Po0.001) and EO (P ¼ 0.023) muscles and the heterolateral PM (P ¼ 0.029) and EO (P ¼ 0.005) muscles were signiﬁcant (Fig. 2A). At 0 cm hip ﬂexion, the women with PLBP used signiﬁcantly less muscle activity for the heterolateral RF (P ¼ 0.022) compared to the women without PLBP (Fig. 2B). Women with PLBP, at 20 cm hip ﬂexion, used signiﬁcantly less muscle activity of the homolateral PM (P ¼ 0.039) than the healthy controls did (Fig. 2C).

4. Discussion The present study reveals changes in the activation patterns of trunk and leg muscles in pregnant women with PLBP compared to healthy controls, during the ASLR test. These changes, like a signiﬁcantly increased EMG, as % of MVC activity, of the hip ﬂexors during lifting of the leg in combination with loss of effort to rise the leg, seem to indicate that PLBP patients try hard to lift a leg but are not or less able to do so. This is in line with the subjective score on the ASLR in the PLBP group. In our population, women without PLBP had a mean score of 0.9 (SD ¼ 1.1) compared to 3.9 (SD ¼ 2.0) for the women with PLBP. These results indicate that women with PLBP scored positive on the ASLR and experienced signiﬁcantly more difﬁculty in raising their legs compared to the pregnant women without PLBP. Since no differences were found in body weight, BMI and leg length between PLBP patients and healthy controls we can conclude that the load of the leg was the same in both groups and cannot be the explanation of the increased EMG activity of the hip ﬂexors with poor results. An explanation for increased EMG activity can be that women with PLBP try to stabilize the pelvis (force closure). After all, as mentioned in the introduction,

ARTICLE IN PRESS M. de Groot et al. / Manual Therapy 13 (2008) 68–74

72 ASLR

Leg raising height 0 cm

1.000

1.000 PLBP (n=11) Non PLBP (n=13)

0.900

0.800

0.700 0.600 0.500 0.400 0.300

0.700 0.600 0.500 0.400 0.300

0.200

0.200

0.100

0.100

0.000

0.000 homolat homolat homolat homolat heterolat heterolat heterolat heterolat RF*

(A)

ALP

M*

EO*

RF

ALP

M*

PLBP (n=11) Non PLBP (n=13)

0.900

Normalised EMG

Normalised EMG

0.800

EO*

homolat homolat homolat homolat heterolat heterolat heterolat heterolat RF

AL

PM

OE

(B)

Muscles

RF*

AL

PM

OE

Muscles

Leg raising height 20 cm 1.000 PLBP (n=11) Non PLBP (n=13)

0.900

Normalised EMG

0.800 0.700 0.600 0.500 0.400 0.300 0.200 0.100 0.000

homolat homolat homolat homolat heterolat heterolat heterolat heterolat RF

(C)

AL

PM*

OE

RF

AL

PM

OE

Muscles * Significant difference at α = 0.05

Fig. 2. EMG as a fraction of MVC during A the ASLR B constrained hip ﬂexion at 0 cm raising height and C constrained hip ﬂexion at 20 cm raising height for the rectus femoris (RF), adductor longus (AL), psoas major (PM) and the external oblique abdominal (EO) muscles.

a correlation was found between impairment of the ASLR and increased laxity of the pelvic ring in women with PLBP (Mens et al., 1999) resulting in a forward rotation of the innominate bone during lifting of a leg. As a result, women with PLBP may need more muscle action to stabilize the pelvis and demonstrate more effort to raise the leg. Another explanation of changes in activation patterns of trunk and leg muscles and loss of hip ﬂexion force in PLBP patients can be a disturbed proprioception and decrease in function of muscles surrounding the pelvis because of pain and fatigue. This is in line with the theory stated by Mens et al. (2001). The results of this study seem to indicate that the ASLR test is related to muscle function for load transfer over the pelvic region. Besides a signiﬁcant increase of EMG activity of the hip ﬂexors, signiﬁcant loss of hip ﬂexion force could be demonstrated in PLBP patients. So, women without PLBP deliver more hip ﬂexion force with the same muscle activity than women with PLBP. A possible explanation is that women without PLBP stabilize their spinal column and pelvic joints more effectively, whereas women with PLBP need more muscle force to

reach for the same goal. These results are in line with previous studies involving the ASLR. Subjects with SIjoint pain or PLBP displayed a decrease in diaphragmatic motion during the ASLR compared to control subjects, which suggests a bracing or splinting action of the diaphragm in conjunction with increased production of intra-abdominal pressure (IAP) (O’Sullivan et al., 2002; Pool-Goudzwaard et al., 2005). Subjects with SIjoint pain demonstrated during the ASLR also a signiﬁcant drop of the pelvic ﬂoor as compared with little movement in the control group (O’Sullivan et al., 2002). The drop of the pelvic ﬂoor indicates a decrease in tension in the pelvic ﬂoor muscles leading to a decrease of SI-joint stiffness (Pool-Goudzwaard et al., 2004). A pelvic belt reduced the impairment of ASLR (Mens et al., 1999), which is in line with the ﬁnding that manual pelvic compression through the iliac bones during ASLR resulted in normal diaphragmatic motion and pelvic ﬂoor descent (O’Sullivan et al., 2002). Pelvic compression could increase stiffness in the pelvic joints, which unloads sensitized ligamentous structures, allowing normalized motor responses during ASLR. These ﬁndings agree with the theoretical model of force closure

ARTICLE IN PRESS M. de Groot et al. / Manual Therapy 13 (2008) 68–74

of the SI-joints (Snijders et al., 1993a). Loss of force, as demonstrated by the signiﬁcant difference between force at 0 cm and hip ﬂexion force at 20 cm can also be induced by pain and fear for pain. However, pain or fear was not measured, so the inﬂuence of these factors is not clear. A signiﬁcant difference between the groups is present on the score of the QBPDS. This scale is developed to measure the grade of disability in non-speciﬁc low back pain; however, the scale appeared also suitable in patients with PLBP (Mens et al., 2001). The current study indicates that women with PLBP experience more functional impairment than the women without PLBP. We assume that if changes in muscle activation and loss of hip ﬂexion force are present in PLBP patients during the ASLR test these changes also occur during activities of daily life. Hence, a higher score on functional impairment on the QBPDS seems logical. A possibly disturbed load transfer across the SI-joints caused through proprioceptive motor control deﬁcits associated with PLBP is very difﬁcult to measure. This study on patients with PLBP used the clinical ASLR test, and demonstrated that these subjects used signiﬁcantly more muscle activity, but could produce less force compared to the control group. Modiﬁcation of the ASLR, with the addition of EMG and force measures, provides a new method of assessing the dysfunctions present in PLBP.

Acknowledgements This research was supported by the Technology Foundation STW, applied science division of NWO and the technology programme of the Ministry of Economic Affairs of The Netherlands.

References Albert H, Godskesen M, Westergaard J. Evaluation of clinical tests used in classiﬁcation procedures in pregnancy-related pelvic joint pain. European Spine Journal 2000;9(2):161–6. Albert H, Godskesen M, Westergaard J. Prognosis in four syndromes of pregnancy-related pelvic pain. Acta Obstetricia et Gynecologica Scandinavica 2001;80(6):505–10. Berg G, Hammar M, Mo¨ller-Nielsen J, Linde´n U, Thorblad J. Low back pain during pregnancy. Obstetrics and Gynecology 1988;71(1):71–5. Bjo¨rklund K, Nordstro¨m ML, Bergstro¨m S. Sonographic assesment of symphyseal joint distention during pregnancy and post partum with special reference to pelvic pain. Acta Obstetricia et Gynecologica Scandinavica 1999;78(2):125–30. Bussey MD, Yanai T, Milburn P. A non-invasive technique for assessing innominate bone motion. Clinical Biomechanics 2004;19(1):85–90. Chamberlain WE. The symphysis pubis in the roentgenexamination of the sacro-iliac joint. American Journal of Roentgenology and Radium therapy 1930;24:621–5.

73

Delagi EF, Iazzetti J, Perotto AO. Anatomical guide for the electromyographer—the limbs and trunk. Springﬁeld: Charles Thomas; 1994. Dreyfuss P, Dreyer S, Grifﬁn J, Hoffman J, Walsh N. Positive sacroiliac screening tests in asymptomatic adults. Spine 1994; 19(10):1138–43. Fast A, Shapiro D, Ducommun EJ, Friedmann LW, Bouklas T, Flowman Y. Low-back pain in pregnancy. Spine 1987;12(4): 3671–83. Harrison DE, Harrison DD, Troyanovich SJ. The sacroiliac joint: a review of anatomy and biomechanics with clinical implications. Journal of Manipulative and Physiological Therapeutics 1997; 20(9):607–17. Heiberg-Endresen E. Pelvic pain and low back pain in pregnant women—an epidemiological study. Scandinavian Journal of Rheumatology 1995;24(4):135–41. Hides JA, Richardson CA, Jull GA. Multiﬁdus muscle recovery is not automatic after resolution of acute, ﬁrst-episode low back pain. Spine 1996;21(23):2763–9. Hodges PW, Richardson CA. Inefﬁcient muscular stabilization of the lumbar spine associated with low back pain: a motor control evaluation of transversus abdominis. Spine 1996;21(22): 2640–50. Hungerford B, Gilleard W, Hodges P. Evidence of altered lumbopelvic muscle recruitment in the presence of sacroiliac joint pain. Spine 2003;28(14):1593–600. Hungerford B, Gilleard W, Lee D. Altered patterns of pelvic bone motion determined in subjects with posterior pelvic pain using skin markers. Clinical Biomechanics 2004;19(5):456–64. Kendall FC, Kendall Mc Creary E. Spieren, tests en functies [Muscles, tests and functions]. Houten: Bohn Staﬂeu Van Loghum; 1986. Kopec J, Esdaile JM, Abrahamowicz M, Abenhaim L, WoodDauphinee S, Lamping DL, et al. The Quebec back pain disability scale. Measurement properties. Spine 1995;20(3):341–52. Kristiansson P, Sva¨rdsudd K, von Schoultz B. Back pain during pregnancy: a prospective study. Spine 1996;21(6):70270–8. Larsen EC, Wilken-Jensen C, Hansen A, Vendelbo Jensen D, Johansen S, Minck H, et al. Symptom-giving pelvic girdle relaxation in pregnancy. Acta Obstetricia et Gynecologica Scandinavica 1999;78(2):105–10. Mantle MJ, Greenwood RM, Currey HL. Backache in pregnancy. Rheumatology and Rehabilitation 1977;16(2):95–101. Mens JM, Vleeming A, Stoeckart R, Stam HJ, Snijders CJ. Understanding peripartum pelvic pain—implications of a patient survey. Spine 1996;21(11):1363–70. Mens JM, Vleeming A, Snijders CJ, Stam HJ, Ginai AZ. The active straight leg raising test and mobility of the pelvic joints. European Spine Journal 1999;8(6):468–74. Mens JM, Vleeming A, Snijders CJ, Koes BW, Stam HJ. Reliability and validity of the active straight leg raise test in posterior pelvic pain since pregnancy. Spine 2001;26(10):1167–71. Mens JM, Vleeming A, Snijders CJ, Koes BW, Stam HJ. Validity of the active straight leg raise test for measuring disease severity in patients with posterior pelvic pain after pregnancy. Spine 2002; 27(2):196–200. Orvieto R, Achiron A, Ben Rafael Z, Gelernter I, Achiron R. Lowback pain of pregnancy. Acta Obstetricia et Gynecologica Scandinavica 1994;73(3):209–14. O¨stgaard HC, Andersson GB, Karlsson K. Prevalence of back pain in pregnancy. Spine 1991;16(5):549–52. O¨stgaard HC, Zetherstro¨m G, Roos-Hansson E, et al. Reduction of back and posterior pelvic pain in pregnancy. Spine 1994;19(8): 894–900. O¨stgaard HC, Roos-Hansson E, Zetherstro¨m G. Regression of back and posterior pelvic pain after pregnancy. Spine 1996;21(23): 2777–80.

ARTICLE IN PRESS 74

M. de Groot et al. / Manual Therapy 13 (2008) 68–74

O’Sullivan PB, Beales DJ, Beetham JA, et al. Altered motor control strategies in subjects with sacroiliac joint pain during the active straight leg raise test. Spine 2002;27(1):E1–8. Perkins J, Hammer RL, Loubert PV. Identiﬁcation and management of pregnancy-related low back pain. Journal of Nurse-Midwifery 1998;43(5):331–40. Pool-Goudzwaard AL, Hoek van Dijke GA, Mulder PG, Spoort CW, Snijders CJ, Stoeckart R. The iliolumbar ligament: its inﬂuence on stability of the sacroiliac joint. Clinical Biomechanics 2003; 18(2):99–105. Pool-Goudzwaard A, Hoek van Dijke G, van Gurp M, Mulder P, Snijders C, Stoeckart R. Contribution of pelvic ﬂoor muscles to stiffness of the pelvic ring. Clinical Biomechanics 2004;19(6): 564–71. Pool-Goudzwaard AL, Slieker ten Hoven MC, Vierhout ME, Mulder PH, Pool JJ, Snijders CJ, et al. Relations between pregnancyrelated low back pain, pelvic ﬂoor activity and pelvic ﬂoor dysfunction. International Urogynecology Journal and Pelvic Floor Dysfunction 2005;16(6):468–74. Richardson CA, Snijders CJ, Hides JA, Damen L, Pas MS, Storm J. The relation between the transversus abdominis muscles, sacroiliac joint mechanics and low back pain. Spine 2002;27(4):399–405. Ro¨st CC, Jaxqueline J, Kaiser A, Verhagen AP, Koes BW. Pelvic pain during pregnancy. A descriptive study of signs and symptoms of 870 patients in primary care. Spine 2004;29(22):2567–72. Sands RX. Backache of pregnancy—a method of treatment. Obstetrics and Gynecology 1958;12(6):670–6. Schoppink LE, van Tulder MW, Koes BW, Beurskens SA, Bie de RA. Reliability and validity of the Dutch adaptation of the Quebec Back Pain Disability Scale. Physical Therapy 1996;76(3):268–75. Snijders CJ, Vleeming A, Stoeckart R. Transfer of lumbosacral load to iliac bones and legs. Part 1: Biomechanics of self-bracing of the

sacroiliac joints and its signiﬁcance for treatment and exercise. Clinical Biomechanics 1993a;8(6):285–94. Snijders CJ, Vleeming A, Stoeckart R. Transfer of lumbosacral load to iliac bones and legs. Part 2: Loading of the sacroiliac joints when lifting in a stooped posture. Clinical Biomechanics 1993b;8(6): 295–301. Snijders CJ, Bakker MP, Vleeming A, et al. Oblique abdominal muscle activity in standing and in sitting on hard and soft seats. Clinical Biomechanics 1995a;10(2):73–8. Snijders CJ, Vleeming A, Stoeckart R, et al. Biomechanical modeling of sacroiliac joint stability in different postures. Spine 1995b; 9(2):419–32. Vleeming A, Stoeckart R, Volkers AC, Snijders CJ. Relation between form and function in the sacroiliac joint. Part 1: Clinical anatomical aspects. Spine 1990a;15(2):130–2. Vleeming A, Volkers AC, Snijders CJ, Stoeckart R. Relation between form and function in the sacroiliac joint. Part 2: Biomechanical aspects. Spine 1990b;15(2):133–6. Walker JM. The sacroiliac joint: a critical review. Physical Therapy 1992;72(12):903–16. Wergeland E, Strand K. Work pace control and pregnancy health in a population-based sample of employed women in Norway. Scandinavian Journal of Work, Environment and Health 1998;24(3): 206–12. Wu W, Meijer OG, Jutte PC, Uegaki K, Lamoth CJ, de Wolf S, et al. Gait in patients with pregnancy-related pain in the pelvis: an emphasis on the coordination of transverse pelvic and thoracic rotations. Clinical Biomechanics 2002;17(9):678–86. Wu W, Meijer OG, Uegaki K, Mens JM, van Dieen JH, Wuisman PI, et al. Pregnancy-related pelvic girdle pain (PPP), 1: Terminology, clinical presentation and prevalence. European Spine Journal 2004;13(7):575–89.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 75–81 www.elsevier.com/locate/math

Original article

MCKenzie classiﬁcation of mechanical spinal pain: Proﬁle of syndromes and directions of preference Cheryl Hefford New Zealand Centre for Physiotherapy Research, School of Physiotherapy, University of Otago, PO Box 56, Dunedin, New Zealand Received 14 October 2005; received in revised form 16 August 2006; accepted 30 August 2006

Abstract The purpose of this study was to develop a proﬁle of the use of McKenzie classiﬁcations of diagnosis and treatment, by physiotherapists credentialed in the McKenzie method in New Zealand. This system has been in common use for more than 20 years and the inter-rater reliability of the assessment has been previously established for therapists at this level of training. Prior studies identifying the classiﬁcation of patients according to syndrome and directional preference have been mainly for the lumbar spine. The 34 participants for this study each assessed and classiﬁed 10 consecutive spinal patients during a 10-week period. Of the 340 patients assessed, 19 were excluded. Of those with pain arising from the lumbar spine; 140/187 were classiﬁed as reducible derangement syndrome, 11/187 were classiﬁed as irreducible derangement, 11/187 as dysfunction syndrome, 1/187 as posture syndrome and 24/187 as ‘other’. For treatment in the reducible derangement syndrome; 98/140 were given extension, 8/140 were given ﬂexion and 34/140 were given lateral movements of either side gliding or rotation. Classiﬁcations and treatment for the cervical and thoracic spine groups followed similar patterns. These ﬁndings add to the external validity of the McKenzie method, and support mechanical evaluation of spinal patients according to directional preference. r 2006 Elsevier Ltd. All rights reserved. Keywords: Back pain; Mechanical classiﬁcation; Directional preference

1. Introduction In the absence of conclusive evidence for effective treatment of low back pain, researchers and clinicians have been moving towards identifying subgroups of patients in order to improve the effectiveness of speciﬁc treatments (McKenzie, 1981; Delitto et al., 1995; Wilson et al., 1999; Fritz and George, 2000; Maluf et al., 2000; Fritz et al., 2003; Kent and Keating, 2004). Several classiﬁcation systems have been described for back and neck pain (McKenzie, 1981, 1990; Spitzer et al., 1987; Delitto et al., 1995; Petersen et al., 1999; Fritz and George, 2000; Maluf et al., 2000; McKenzie and May, 2003; Petersen et al., 2003; Sterling, 2004). Because of the lack of agreement over a tissue-speciﬁc diagnosis for Tel: +64 463 5257; fax: +64 463 5257.

E-mail address: [email protected]. 1356-689X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.08.005

low back pain, the Quebec Task Force in 1987 recommended classifying patients according to symptom distribution and the existence or extent of radiating symptoms (Spitzer et al., 1987). While this was an acknowledgement of the difﬁculty of diagnosis of nonspeciﬁc low back pain, it did not address possible management options. Kent and Keating (2004) found that primary care clinicians in Australia commonly assign non-speciﬁc low back pain patients into some kind of subgroup which determines their management. However, there is no agreement among the clinicians on an acceptable classiﬁcation system. Mechanical classiﬁcation in the McKenzie system of mechanical diagnosis and therapys (MDT) was ﬁrst described in 1981(McKenzie, 1981) and continues to be in common usage in the USA, UK and New Zealand (Sullivan et al., 1996; Jackson, 2001; Gracey et al., 2002; Reid et al., 2002). It involves the assessment and

ARTICLE IN PRESS 76

C. Hefford / Manual Therapy 13 (2008) 75–81

classiﬁcation of patients into one of three mechanical syndromes (or as ‘other’), according to the symptomatic and mechanical response to repeated movements and sustained positions. Brief deﬁnitions of the three mechanical syndromes (posture, dysfunction or derangement (reducible or irreducible)), and ‘other’ are given below and described fully elsewhere (McKenzie, 1981, 1990; McKenzie and May, 2003): 1. Posture syndrome pertains to pain arising as a result of mechanical deformation of normal soft tissues from prolonged end range loading of periarticular structures. The treatment principle for posture syndrome is posture correction. 2. Dysfunction syndrome pertains to pain occurring as a result of mechanical deformation of structurally impaired tissues (such as tissue which is scarred, adhered or adaptively shortened). The treatment principle for dysfunction syndrome is to exercise into the direction of the dysfunction with the aim of remodelling the tissue. 3. Derangement syndrome pertains to pain occurring as a result of a disturbance in the normal resting position of the affected joint surfaces. Derangement may be reducible or irreducible. The treatment principle for derangement syndrome depends on the clinically induced directional preference, identiﬁed by examining the patient’s symptomatic and mechanical response to repeated movements or sustained positions. An irreducible derangement ﬁts the history criteria for derangement but no loading strategy is able to produce a lasting change on the symptoms. In the MDT system, this is conceptually thought to concern an incompetent or ruptured outer annular wall of the intervertebral disc. A reducible derangement typically demonstrates one direction of repeated movement (directional preference) which decreases or centralizes (moves towards the midline) referred symptoms, or abolishes midline symptoms (Long et al., 2004), and the opposite repeated movement which produces or increases or peripheralizes (moves more distally) the symptoms. In the lumbar spine, movements typically include ﬂexion in lying or standing; extension in lying or standing; and lateral movements of either side gliding or rotation. Similar principles of repeated movement testing and treatment are applied in the cervical and thoracic spines. They are standard movements in the MDT system and are described fully in the text books (McKenzie, 1981, 1990; McKenzie and May, 2003). 4. Other is used to describe those who do not ﬁt with the mechanical syndromes but who exhibit signs and symptoms of other known pathology such as spinal stenosis, hip disorders, sacroiliac disorders, low back pain in pregnancy, zygapophyseal disorders, spondylolysis and spondylolisthesis, and post-surgical problems. The system has demonstrated strong inter-rater reliability amongst physiotherapists trained in the

McKenzie method (k values from 0.79 to 1.0) (Werneke et al., 1999; Fritz et al., 2000; Razmjou et al., 2000; Kilpikoski et al., 2002; Clare et al., 2004, 2005). For the identiﬁcation of derangement syndrome, there is good to excellent inter-rater reliability (k values 0.7 and 0.96) (Razmjou et al., 2000; Kilpikoski et al., 2002). In one study, the inter-rater reliability for identifying directional preference was reported as excellent (90% agreement, k 0.9) (Kilpikoski et al., 2002). There has been a large amount of research into the clinical ﬁndings of centralization and directional preference, which are major aspects of MDT (Donelson et al., 1990, 1991; Long, 1995; Werneke et al., 1999; Werneke and Hart, 2001, 2003; Aina et al., 2004; Long et al., 2004). According to Clare et al. (2005), ﬁve studies have looked at the reliability of MDT classiﬁcation and have included prevalence of the syndromes (Kilby et al., 1990; Riddle and Rothstein, 1993; Razmjou et al., 2000; Kilpikoski et al., 2002; Clare et al., 2005); most of these studies have concentrated on the lumbar spine. The primary aim of this study was to establish how many of the patients with mechanical cervical, lumbar and thoracic pain were classiﬁed into each of the syndromes by speciﬁcally trained clinicians (MDT credentialed) in New Zealand clinical settings. The secondary aim was to determine, within the reducible derangement syndrome, the proportion of patients in each symptom distribution category and their given treatment direction (directional preference). It was hoped that comparing these ﬁndings with previous studies would further validate the MDT classiﬁcation system. 2. Method A survey was undertaken of all the MDT credentialed physiotherapists1 who were McKenzie Institute New Zealand Branch members in July 2004 (N ¼ 50). Packs containing instructions, information sheets, consent forms, and data sheets were distributed to all members. They were all physiotherapists thought to be working in musculoskeletal outpatient settings, either in private practice (N ¼ 47) or at a public hospital (N ¼ 3). Of these 50 potentially eligible for the study, two were unable to be contacted (no response to post, email or phone) and nine were unable to participate due to external factors such as not currently seeing patients, being overseas or working in an inappropriate environment. This meant the number conﬁrmed eligible for the study was 39. Thirty-four of the 39 eligible physiotherapists (87%) returned completed data sheets with details of 10 consecutive patients with spinal pain presenting in 1 MDT credentialed physiotherapists have completed the McKenzie Institute International education programme of a minimum of 98 h and passed a standardized examination at the conclusion of this.

ARTICLE IN PRESS C. Hefford / Manual Therapy 13 (2008) 75–81

77

Table 1 Data collection form Start date: Case/consent

1O

Finish date: Gender/age Eg. M/56

Area

Lumbar

Syndrome classiﬁcation (tick the box)

Reducible

Derangement symptom location (tick one if appropriate)

Primary principle of treatment

Central symmetrical

Extension

Unilateral asymmetrical

Flexion

derangement Cervical

Irreducible derangement

Thoracic

Dysfunction Posture

symptoms to knee/elbow Unilateral asymmetrical to below

Lateral

knee/elbow

Other

their normal clinic situation (a total of 340 patients). Patients were excluded if they failed to give consent or if they were under age 18. Data were collected over a 10-week time frame. The University of Otago Human Ethics Committee granted ethics approval for the study (Ref. 04/096). Data were collected at the initial assessment, according to the form shown in Table 1. Once therapists had completed their data sheets on 10 consecutive patients, details were posted to the author who collated the results using the Epi InfoTM (http://www.cdc.gov/ epiinfo/) epidemiological database. The participating physiotherapists were mainly older experienced physiotherapists (88% aged between 30 and 59 years), qualiﬁed for an average of 22 years (SD 9.2) and MDT credentialed for an average of 6.2 years (SD 3.6). All but two practised in a musculoskeletal private practice setting with an even mix of large city, small city and rural town practice. In their clinics, 24% of therapists stated that they saw predominantly acute and sub acute patients and 76% said they saw a mix of acute and chronic patients; no one stated they saw predominantly chronic patients. The mean data collection period was 22.8 days (SD 17.8). Of the 340 consecutive patients, 19 were excluded from the study: nine were under the age speciﬁed (18 years), and 10 did not give consent. No data was collected from the ten who denied consent (a line was drawn through the data sheet). The survey was, therefore, based upon the details of 321 patients from 34 physiotherapists; data are presented descriptively.

3. Results Demographic data and classiﬁcation of the 321 patients into the syndromes is shown for each of the lumbar, cervical and thoracic areas (Table 2). Reducible derangement was found to be the largest group in all three areas.

Symptom location patterns of the reducible derangement group, for each spinal area, are shown in Fig. 1. A clear pattern emerged across all areas with the biggest group being the asymmetrical group (above the elbow in the cervical and above the knee in the lumbar areas). The second largest group was the central or symmetrical distribution of symptoms. The smallest groups contained those with the most distal symptoms. The treatment principles given to the reducible derangement group, for each spinal area, are shown in Table 3. Extension was consistently the most common treatment principle across all spinal areas and ﬂexion the least common. The treatment principle for each symptom location pattern within the reducible derangement group for each spinal area is also shown in Table 3. When symptoms were central or symmetrical, the extension treatment principle was most commonly used; when symptoms were unilateral or asymmetrical, lateral treatment principles were used (although, even then extension principle was the most commonly used).

4. Discussion For this survey, it was found that of 321 patients assessed by 34 credentialed therapists, 92% of patients were classiﬁed into one of the three mechanical syndromes, with more than 80% classiﬁed as ‘derangement’. The classiﬁcation of patients into the MDT syndromes in this survey is consistent with previously published data from other countries (Kilby et al., 1990; Riddle and Rothstein, 1993; Razmjou et al., 2000; Kilpikoski et al., 2002; Long et al., 2004; Clare et al., 2005). The mean percentage of patients classiﬁed as derangement by these authors was 71.6%. In this study, 77.9% were classiﬁed as reducible derangement syndrome (Table 4). The treatment principle given, according to directional preference, for the majority of patients classiﬁed

ARTICLE IN PRESS C. Hefford / Manual Therapy 13 (2008) 75–81

78

Table 2 Demographic data and Syndrome Classiﬁcation for each spinal area Lumbar

Thoracic

Combined areas

Mean age

43.9

46.7

40.5

45.27 (SD 16.3)

Gender Male Female

42 (38%) 69 (62%)

102 (55%) 85 (45%)

12 (52%) 11 (48%)

% of total 156 (49%) 165 (51%)

Syndrome Posture Dysfunction Reducible derangement Irreducible derangement Other

3 (2.7%) 9 (8.1%) 90 (81.1%) 1 (0.9%) 8 (7.2%)

1 (0.5%) 11 (5.9%) 140 (74.9%) 11 (5.9%) 24 (12.8%)

0 (0%) 2 (8.7%) 20 (87.0%) 0 (0%) 1 (4.3%)

%of total 4 (1.3%) 22 (6.9%) 250 (77.9%) 12 (3.7%) 33 (10.3%)

Total

111

187

23

321 (100%)

patient numbers

Cervical

80 60 40

37%

45%

63%

18%

20

20%

25% 30%

70%

0 Lumbar

Cervical

Thoracic

Central Symmetrical Unilateral or asymmetrical above the elbow / knee Unilateral or asymmetrical below the elbow / knee Fig. 1. Symptom location patterns in the reducible derangement syndrome.

as derangement syndrome, was extension. However, it is important to note the smaller, yet highly relevant, number of patients who responded to lateral and ﬂexion movements. For the lumbar spine, the number of ﬂexion responders in this audit is consistent with what has been reported previously. A randomized controlled trial (RCT) that classiﬁed lumbar spine patients prior to randomization identiﬁed 74% with a directional preference, of which 83% were extension responders, 7% were ﬂexion responders and 10% were lateral responders (Long et al., 2004). A previous study that looked only at sagittal movements in the lumbar spine (did not include lateral movements in the assessment) found that 47% demonstrated a directional preference for either ﬂexion (7%) or extension (40%) (Donelson et al., 1991). In a study looking at both back and neck pain 77% demonstrated a directional preference (centralized or partial reduction) after multiple visits (o7 visits over a 2–3 week time frame) (Werneke and Hart, 2003). A recent systematic review of 14 studies on centralization found that on assessment, 70% of acute and sub-acute back pain patients, and 52% of chronic back pain patients demonstrated centralization. This occurred most commonly with extension exercises

or postures but also with other loading strategies (Aina et al., 2004). Although outcomes as a result of classiﬁcation and treatment direction were not addressed in this study, a recent RCT study showed that giving the ‘wrong’ direction of exercises to low back pain patients can lead to poorer outcomes (Long et al., 2004). Most of the research into MDT has been on the lumbar spine, even though the system has always included the cervical and thoracic spines (McKenzie, 1990). No studies have previously looked at MDT for the thoracic spine. The incidence of patients presenting for thoracic spine pain was small compared to those presenting with lumbar and cervical spine pain and it would be interesting to see how this might compare with studies done in other countries. In New Zealand, because of the size of the country and the wide geographical location of the participants in this study, there was a unique opportunity to get a broad cross section of the patient community. The demographics of the participants and distribution of patients was over a range of city and rural practices, with a mix of acute and chronic patients. In the MDT system, classiﬁcation is used to guide treatment. For the reducible derangement syndrome, the directional preference becomes the treatment principle. The viability of the system depends on the consistent interpretation of the classiﬁcations and treatment directions. The clinical relevance of this study is that it conﬁrms previous work on the numbers of spinal patients classiﬁed into the MDT syndromes. It adds to previous work by detailing commonly encountered speciﬁc treatment directions for spinal patients according to their differing referral patterns of pain. These consistent ﬁndings add to the stability of the MDT system and may be useful for future research designs into the prognostic and therapeutic outcomes of the speciﬁc classiﬁcation groups.

ARTICLE IN PRESS C. Hefford / Manual Therapy 13 (2008) 75–81

79

Table 3 Treatment principles for the reducible derangement syndrome according to symptom distribution

Lumbar Central or symmetrical Unilateral or asymmetrical above the knee Unilateral or asymmetrical below the knee Total treatment directions for lumbar Cervical Central or symmetrical Unilateral or asymmetrical above the elbow Unilateral or asymmetrical below the elbow Total treatment directions for Cervical Thoracic Central or symmetrical Unilateral or asymmetrical Total treatment directions for Thoracic

Extension

Flexion

Lateral movements (side bending and/or rotation)

48 (92%) 36 (57%) 14 (56%)

3 (6%) 4 (6%) 1 (4%)

1 (2%) 238 (37%) 10 (40%)

98 (70%)

8 (6%)

34 (24%)

17 (94%) 36 (63%) 12 (80%)

1 (6%) 7 (12%) 0 (0%)

0 (0%) 14 (25%) 3 (20%)

65 (72%)

8 (9%)

17 (19%)

5 (83%) 12 (86%)

0 (0%) 0 (0%)

1 (17%) 2 (14%)

17 (85%)

0 (0%)

3 (15%)

Table 4 Adapted from Clare et al (2005), Prevalence of syndromes Reference

No. of patients

% Derangement

% Dysfunction

% Postural

% Other

Kilby et al. (1990) (lumbar) Riddle and Rothstein (1993) (lumbar) Razmjou et al. (2000) (lumbar) Kilpikoski et al. (2002) (lumbar) Clare et al. (2005) (lumbar and cervical) This study Hefford (lumbar, cervical and thoracic)

41 363 45 39 50

42.7 52.9 86.7 90 86

22 34.7 4.4 2 2

2.4 9.6 2.2 Nil Nil

32.9 2.8 6.7 8 12

321

77.9 (reducible derangement) 3.7 (irreducible derangement)

6.8

1.3

10.3

5. Limitations This study has only accounted for classiﬁcations according to the ﬁrst assessment session for each patient. It has been shown that in some cases, classiﬁcation may be better judged over several visits (Werneke and Hart, 2003). This is a descriptive study of classiﬁcation at initial assessment; outcomes from the treatment have not been addressed. However, many studies have reported the prognostic and therapeutic outcomes of treatment utilising movements of directional preference (Kopp et al., 1986; Donelson et al., 1990, 1997; Long, 1995; Sufka et al., 1998; Werneke et al., 1999; Udermann et al., 2004). The data was not completely independent in that each therapist collected data on 10 consecutive spinal patients. It is possible that the previous patients they had assessed and treated could have inﬂuenced their judgement. The strengthening reporting of observational studies in epidemiology (STROBE) guidelines (www.strobe-statement.org/PDF/STROBE-Checklist-Version3.pdf) suggest

discussion of the direction and magnitude of such potential bias, but this is unknown.

6. Conclusion In a survey of 50 New Zealand physiotherapists trained in the MDT system; out of 39 potential and contactable therapists 34 provided classiﬁcation details regarding 10 consecutive spinal patients. The therapists worked in an even mix of rural and urban physiotherapy practices across the length of New Zealand. Mechanical syndromes as described by McKenzie were used for classiﬁcation purposes, with the largest single category being reducible derangement. Most commonly the extension treatment principle was used; but ﬂexion and lateral forces were used also, with the latter more often in the presence of referred symptoms. This is one of very few studies which include the cervical spine, and the ﬁrst to include classiﬁcation and treatment of the thoracic spine, even though it was described by McKenzie in 1990. More research is required to

ARTICLE IN PRESS 80

C. Hefford / Manual Therapy 13 (2008) 75–81

validate these ﬁndings as thoracic spinal pain numbers are small. The study adds to the growing body of evidence supporting the external validity of McKenzie’s mechanical syndromes. It conﬁrms baseline ﬁndings of other studies and provides new ﬁndings about the proﬁle of classiﬁcations into syndromes for all areas and the directional preference ﬁndings for the cervical and thoracic spine. These baselines are useful for both clinicians and researchers. It also supports the need for mechanical evaluation and subdivision of spinal patients according to directional preference.

Acknowledgements The author would like to thank the following: The McKenzie Institute New Zealand Branch Credentialed therapists, for their participation and encouragement in this project. Mr Stephen May, Dr Ron Donelson, Professor David Baxter and Professor S. John Sullivan for critical revision of drafts of this manuscript.

References Aina A, May S, Clare H. The centralization phenomenon of spinal symptoms—a systematic review. Manual Therapy 2004;9(3): 134–43. Clare HA, Adams R, Maher CG. A systematic review of efﬁcacy of McKenzie therapy for spinal pain. Australian Journal of Physiotherapy 2004;50(4):209–16. Clare HA, Adams R, Maher CG. Reliability of McKenzie classiﬁcation of patients with cervical or lumbar pain. Journal of Manipulative Physiological Therapy 2005;28(2):122–7. Delitto A, Erhard RE, Bowling RW. A treatment-based classiﬁcation approach to low back syndrome: identifying and staging patients for conservative treatment. Physical Therapy 1995;75(6):470–85 discussion 485–479. Donelson R, Aprill C, Medcalf R, Grant W. A prospective study of centralization of lumbar and referred pain. A predictor of symptomatic discs and anular competence. Spine 1997;22(10): 1115–22. Donelson R, Grant W, Kamps C, Medcalf R. Pain response to sagittal end-range spinal motion. A prospective, randomized, multicentered trial. Spine 1991;16(6 Suppl):S206–12. Donelson R, Silva G, Murphy K. Centralization phenomenon. Its usefulness in evaluating and treating referred pain. Spine 1990;15(3):211–3. Fritz JM, Delitto A, Erhard RE. Comparison of classiﬁcation-based physical therapy with therapy based on clinical practice guidelines for patients with acute low back pain: a randomized clinical trial. Spine 2003;28(13):1363–71 discussion 1372. Fritz JM, Delitto A, Vignovic M, Busse RG. Interrater reliability of judgments of the centralization phenomenon and status change during movement testing in patients with low back pain. Archives of Physical and Medical Rehabilitation 2000;81(1): 57–61. Fritz JM, George S. The use of a classiﬁcation approach to identify subgroups of patients with acute low back pain. Interrater

reliability and short-term treatment outcomes. Spine 2000;25(1): 106–14. Gracey JH, McDonough SM, Baxter GD. Physiotherapy management of low back pain: a survey of current practice in northern Ireland. Spine 2002;27(4):406–11. Jackson DA. How is low back pain managed? Retrospective study of the ﬁrst 200 patients with low back pain referred to a newly established community-based physiotherapy department. Physiotherapy 2001;87(11):573–81. Kent P, Keating J. Do primary-care clinicians think that nonspeciﬁc low back pain is one condition? Spine 2004;29(9):1022–31. Kilby J, Stigant M, Roberts A. The reliability of back pain assessment by physiotherapists, using a ‘‘McKenzie algorithm’’. Physiotherapy 1990;76(9):579–83. Kilpikoski S, Airaksinen O, Kankaanpaa M, Leminen P, Videman T, Alen M. Interexaminer reliability of low back pain assessment using the McKenzie method. Spine 2002;27(8):E207–14. Kopp JR, Alexander AH, Turocy RH, Levrini MG, Lichtman DM. The use of lumbar extension in the evaluation and treatment of patients with acute herniated nucleus pulposus. A preliminary report. Clinical Orthopaedics and Related Research 1986;202: 211–8. Long A. The centralization phenomenon. Its usefulness as a predictor or outcome in conservative treatment of chronic law back pain (a pilot study). Spine 1995;20(23):2513–20 discussion 2521. Long A, Donelson R, Fung T. Does it matter which exercise? A randomized control trial of exercise for low back pain. Spine 2004;29(23):2593–602. Maluf KS, Sahrmann SA, Van Dillen LR. Use of a classiﬁcation system to guide nonsurgical management of a patient with chronic low back pain. Physical Therapy 2000;80(11):1097–111. McKenzie R. Mechanical diagnosis and therapy of the lumbar spine, 1st ed. Waikanae, New Zealand: Spinal Publications Ltd.; 1981. McKenzie R. Mechanical diagnosis and therapy of the cervical and thoracic spine. Waikanae, New Zealand: Spinal Publications Ltd.; 1990. McKenzie R, May S. The lumbar spine mechanical diagnosis and therapy. 2nd ed. Waikanae, New Zealand: Spinal Publications Ltd.,; 2003. Petersen T, Laslett M, Thorsen H, Manniche C, Ekdahl C, Jacobsen S. Diagnostic classiﬁcation of non-speciﬁc low back pain. A new system integrating patho-anatomic and clinical categories. Physiotherapy—Theory Research and Practice 2003;19(4):213–37. Petersen T, Thorsen H, Manniche C, Ekdahl C. Classiﬁcation of nonspeciﬁc low back pain: a review of the literature on classiﬁcations systems relevant to physiotherapy. Physical Therapy Reviews 1999;4(4):265–81. Razmjou H, Kramer JF, Yamada R. Intertester reliability of the McKenzie evaluation in assessing patients with mechanical lowback pain. Journal of Orthopedic and Sports Physical Therapy 2000;30(7):368–83 discussion 384–369. Reid D, Hing W, McNair P, Larmer P, Robb G. Managing an acute lumbar spine condition: the ﬁndings of a vignette. In: Paper presented at the New Zealand Manipulative Physiotherapists Association biennial conference, Rotorua, New Zealand; 2002. Riddle DL, Rothstein JM. Intertester reliability of McKenzie’s classiﬁcations of the syndrome types present in patients with low back pain. Spine 1993;18(10):1333–44. Spitzer W, LeBlanc F, Dupuis M. Scientiﬁc approach to the assessment and management of activity-related spinal disorders. A monograph for clinicians. Report of the Quebec Task Force on spinal disorders. Spine 1987;12(7 Suppl):S1–S59. Sterling M. A proposed new classiﬁcation system for whiplash associated disorders—implications for assessment and management. Manual Therapy 2004;9(2):60–70.

ARTICLE IN PRESS C. Hefford / Manual Therapy 13 (2008) 75–81 Sufka A, Hauger B, Trenary M, Bishop B, Hagen A, Lozon R, et al. Centralization of low back pain and perceived functional outcome. Journal of Orthopaedic and Sports Physical Therapy 1998;27(3): 205–12. Sullivan MS, Kues JM, Mayhew TP. Treatment categories for low back pain: a methadological approach. Journal of Orthopaedic and Sports Physical Therapy 1996;24(6):359–64. Udermann BE, Spratt KF, Donelson RG, Mayer J, Graves JE, Tillotson J. Can a patient educational book change behavior and reduce pain in chronic low back pain patients? Spine Journal 2004;4(4):425–35.

81

Werneke M, Hart DL. Centralization phenomenon as a prognostic factor for chronic low back pain and disability. Spine 2001;26(7):758–64 discussion 765. Werneke M, Hart DL. Discriminant validity and relative precision for classifying patients with nonspeciﬁc neck and back pain by anatomic pain patterns. Spine 2003;28(2):161–6. Werneke M, Hart DL, Cook D. A descriptive study of the centralization phenomenon. A prospective analysis. Spine 1999;24(7): 676–83. Wilson L, Hall H, McIntosh G, Melles T. Intertester reliability of a low back pain classiﬁcation system. Spine 1999;24(3):248–54.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 82–86 www.elsevier.com/locate/math

Case report

Costochondritis: Are the symptoms and signs due to neurogenic inﬂammation. Two cases that responded to manual therapy directed towards posterior spinal structures Martin Ian Rabey St. Sampson’s Medical Centre, Grandes Maisons Road, St. Sampson’s, Guernsey, GY2 4JS Great Britain, UK Received 7 September 2006; received in revised form 16 November 2006; accepted 19 December 2006

Keywords: Thoracic spine; Chest pain; Manipulation

1. Introduction Approximately 50% of patients attending accident and emergency departments and outpatient cardiac clinics because of chest pain have a non-cardiac basis for their symptoms. They are often given a non-speciﬁc diagnosis (Mayou et al., 1997; Capewell and McMurray, 2000). One possible diagnosis these patients may receive is costochondritis. Possible causes of costochondritis will be discussed and a group of these patients delineated for whom there is no documented underlying pathophysiology. This paper then presents two such case reports and goes on to question whether costochondritis is simply localised anterior chest wall pathology. It is considered whether positive responses to manual therapy directed towards the spine and posterior rib articulations are indicative of a more widespread neuropathic contribution to the condition.

2. Pathophysiology There is no clear deﬁnition for costochondritis (Wadhwa et al., 1999). Therefore, making the diagnosis of either idiopathic costochondritis or Tietze’s syndrome is predominantly based upon clinical ﬁndings, and often occurs following the exclusion of other potentially Tel.: +44 1481 245915.

E-mail address: [email protected]. 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2006.12.007

serious or systemic pathologies (visceral pain, infection, etc.) that may give rise to chest pain. Laboratory investigations and imaging are usually normal (Wadhwa et al., 1999). Tenderness localised over the costochondral joints may be the only clinical ﬁnding in costochondritis, whereas in Tietze’s syndrome there may be associated swelling, heat and erythema (Scott and Scott, 1993; Mendelson et al., 1997). The underlying pathophysiology relating to this tenderness or swelling has not been documented. It has been suggested that the costovertebral and costotransverse joints are commonly overlooked sources of atypical chest pain and there are a number of case reports detailing manual therapy treatment for this type of presentation (Fallon, 1996; Triano et al., 1999). However, the author was unable to ﬁnd any reports of manual therapy, directed at the zygapophyseal and posterior rib articulations, being utilised as an effective treatment in cases where both localised anterior tenderness and costochondral swelling were present. 3. Presentation 3.1. Case one 3.1.1. Patient interview and history A 29-year-old female physiotherapist presented with a 10-week history of insidious onset central chest pain described as a constant dull ache or squashing sensation, with an intensity of 3 out of 10 on a visual analogue

ARTICLE IN PRESS M. Ian Rabey / Manual Therapy 13 (2008) 82–86

83

region revealed weakness in the right serratus anterior and dominance of right rhomboid activity.

Fig. 1. Body chart showing painful area—case one.

scale. There was no posterior thoracic pain. (See Fig. 1.) Aggravating factors were lifting, pushing and deep breathing. Symptoms could be eased with ice packs, taking nonsteroidal anti-inﬂammatory medication and through the maintenance of ﬂexed thoracic posture during activities. MRI scanning did not reveal any abnormality. Both the patient’s general practitioner and consultant orthopaedic surgeon had diagnosed costochondritis. There was no other relevant medical history.

3.1.2. Physical examination Observation revealed subtle swelling over the right sternocostal and upper costochondral joints. There was greater right rhomboid muscle bulk than left. Thoracic spine ﬂexion, extension and left side ﬂexion were all of full range and pain free. Right side ﬂexion showed full range of movement with provocation of chest pain at the end of range. Thoracic rotation was estimated visually as the angle made by the trunk at shoulder level when compared to the initial position and that of the ﬁxed pelvis with the patient sitting. Left rotation was 35 and right rotation 45 with both movements being described as ‘‘stiff’’ by the patient. During passive physiological motion palpation right rotation and right side ﬂexion appeared to be restricted at the T1-2 and T2-3 levels. Movement between the second and third ribs also appeared restricted on passive motion palpation. Palpation over the right articular pillar of T3 and the third rib angle provoked localised allodynia but not referred symptoms. Resisted glenohumeral lateral rotation and ﬂexion on the right were both provocative of anterior chest pain and were notably weaker than the contralateral side. Assessment of motor control in the scapula

3.1.3. Treatment During the ﬁrst treatment session, a passive oscillatory postero-anterior mobilisation was applied over the right articular pillar of T3 and over the angle of the third rib. An improvement in both pain and strength on resisted shoulder ﬂexion was gained. One week later the patient volunteered that squeezing objects was a reliable indicator of symptom severity. Range of motion of thoracic rotation remained unchanged. Further mobilisation was applied over T3 and the third rib. The patient expressed an immediate subjective 50% reduction in pain intensity whilst squeezing a large object. A week later the patient reported an overall decrease in pain intensity and greater functional use of the right upper limb. Squeezing the same large object was not provocative. However, lifting a 5 kg weight in either hand provoked chest pain. Further mobilisation was carried out as in previous sessions. Within the same treatment session a T2-3 supine high velocity thrust was administered. This was followed by a SNAG over the right third rib angle with the patient turning into right thoracic rotation as this was the most restricted movement at this stage of the treatment. This lead to a signiﬁcant improvement in lifting the 5 kg weight, and left and right rotation were both 55 and pain free. Right thoracic side ﬂexion was full range and pain free. On the fourth treatment session seven days later, the patient reported no pain during her normal work duties. Lifting 10 kg with the left hand was not provocative but was with the right hand. Further mobilisation and a repeat of the above manipulation were carried out. Left and right thoracic rotation increased to 70 degrees. Upon review six weeks later the patient stated that she was pain free, but approximately two weeks after that had a degree of recurrence due to heavy lifting at work. Over the following four weeks a further two sessions of the abovementioned mobilisation were undertaken which again cleared the symptoms. As a maintenance programme during this period exercises to restore appropriate motor control in the scapula region were prescribed and these were progressed over the next three months. During this time the patient remained symptom free. 3.2. Case two 3.2.1. Patient interview and history The second case is a 33-year-old female photographic processor. She had episodic insidious onset chest pain for 18 months. She had been diagnosed by her general practitioner as having costochondritis due to palpable swelling over the third and fourth costochondral joints on the left. She described a dull ache, 6 out of 10 on a

ARTICLE IN PRESS 84

M. Ian Rabey / Manual Therapy 13 (2008) 82–86

Fig. 2. Body chart showing painful area—case two.

visual analogue scale, at the left anterior chest wall and a corresponding point at the medial border of the left scapula. (See Fig. 2) There was also occasional formication in the anterior chest region. Aggravating activities were working at the computer for 1 hour, and running for 20 min. Non-steroidal anti-inﬂammatory medication was the only easing factor. Multiple blood tests, an electrocardiograph and a chest X-ray were all normal. She had no other relevant medical history. 3.2.2. Physical examination Thoracic ﬂexion, extension, side ﬂexion and right rotation were all pain free and of full range. Left thoracic rotation was equal in range to right rotation, but was provocative of both anterior and posterior chest wall symptoms at end of range. There appeared to be restricted passive physiological motion during testing of left rotation and side ﬂexion at the T4-5 level. Movement between the fourth and ﬁfth ribs also appeared restricted when an appreciation of the cephalad and caudad components of normal inspiratory and expiratory rib motion were tested in a passive physiological manner. Palpation over the articular pillar of T4 provoked localised allodynia, but not referred anterior symptoms. Assessment of motor control in the scapula region once again revealed dominance of the rhomboids combined with levator scapulae. There was notable wasting of the left upper trapezius leaving the scapula internally rotated and depressed at rest. 3.2.3. Treatment The initial treatment session involved passive oscillatory postero-anterior mobilisation over the left T4 articular pillar, and a supine high velocity thrust at the T45 motion segment. This had the effect of rendering left

thoracic rotation pain free. Upon her second visit two days later left thoracic rotation was full but with pain at end of range again. The previous treatment was repeated with the same effect. However, the patient was also taught to reposition her scapula using appropriate muscle activity. Six days later the patient returned and stated that repositioning of the scapula had been helpful when she had been aware of any discomfort. She had not had any anterior chest pain, but still had some intermittent scapula area pain. The formication was unchanged. A further high velocity thrust was administered to the same motion segment, and a caudally directed passive oscillatory mobilisation over the fourth rib was also undertaken to facilitate the normal respiratory rib motion that appeared restricted on passive physiological testing. Over the next month a further four similar treatments were undertaken combining the same mobilisation and manipulative techniques directed towards the zygapophyseal joint and mobilisation over the rib articulations. At this stage the patient stated that there was minimal pain on a dayto-day basis and that the formication had resolved. All thoracic movements were pain free and of full range but there was still notable tenderness when palpating over the fourth rib angle. Over the following six weeks motor control in the scapula region was progressed during which time all chest wall symptoms resolved. However, the palpable swelling over the costochondral joints remained.

4. Discussion Boyling (1999) described some possible reasons for publication of a speciﬁc case report. Apart from

ARTICLE IN PRESS M. Ian Rabey / Manual Therapy 13 (2008) 82–86

detailing potentially unexpected results of an otherwise commonplace treatment technique the case report that he was considering (Boyle, 1999) lead to the proposition of an otherwise undocumented cause for a patient’s symptoms. Boyling (1999) asked, ‘‘Were there more patients out there with similar presentations?’’ The author was able to ﬁnd ﬁve case reports where manual therapy had been utilised in the treatment of anterior chest pain. Diagnosed as ‘‘intercostal neuritis’’ Fallon (1996) reported relief of anterior chest wall pain with spinal manipulation. Triano et al. (1999) reported four cases of atypical chest pain also relieved by spinal manual therapy. However, none of these cases detailed any localised costochondral tenderness or anterior chest wall swelling that could have brought about a diagnosis of costochondritis or Tietze’s syndrome. Both cases reported by this author received a diagnosis of costochondritis from medical practitioners, although the localised swelling that both patients presented suggests that a diagnosis of Tietze’s syndrome could have been considered. This diagnosis was based upon exclusion of other pathologies and clinical ﬁndings of localised costochondral joint tenderness and swelling (Mendelson et al., 1997). Both of these patients responded to manual therapy directed solely towards posterior spinal and rib articulations. No attempt at localised anterior thoracic treatment was made. In light of this it is therefore pertinent to consider possible causes for the symptoms and signs that these patients presented with, and hypothesise why manual therapy directed at relatively distant structures may have been effective. The author was unable to ﬁnd any documented local pathophysiology of the costochondral joints in this group of patients. Costovertebral joints are highly innervated including neurones containing substance P, thus suggesting a nociceptive role. They also contain innervated meniscoid inclusions (Erwin et al., 2000). It has been postulated that an innervated meniscus may become entrapped between the articular surfaces. This would then create a situation, whereby when the subject moves traction is placed on the joint capsule through the base of meniscus and this traction is provocative of symptoms. It is then further postulated that manipulation of the affected articulation reduces this traction by producing joint gapping and hence releasing the entrapped meniscus. However, it is questionable whether the base of the meniscus could withstand the degree of tension necessary to exert this capsular traction (Bogduk and Engel, 1984). Nociceptive afferent input from the posterior spinal joints to the dorsal horn may lead to expansion of the receptive ﬁelds of second order neurones. This would potentially allow afferent input from related somatic structures, which may not necessarily be nociceptive input, to depolarise this

85

second order neurone. This may cause perception of symptoms that are incorrectly localised (Mense, 1991; Woolf and Doubell, 1994). Although this may explain anterior chest wall symptoms in the presence of posterior thoracic pathology it is not adequate to explain the visible anterior chest wall swelling noted in these two patients. In keeping with the amelioration of the two patient’s symptoms by treatment of posterior spinal structures only the most feasible possible explanation for the visible and palpable swelling in these subjects is neurogenic inﬂammation. Degenerative changes are common at the T3-4 level (44% of the study sample) although they are less frequent in subjects below 29 years of age (Boyle et al., 1998). In the thoracic spine the ventral rami at each level exit the intervertebral foraminae and pass under the rib of the same segment as the intercostal nerve, innervating the skin, and extending as far as the body’s midline anteriorly (Bogduk, 1994). Minor spinal stresses may be adequate to facilitate the release of the cytokine interleukin-1b (Okawa et al., 1998), and raised cytokine levels may be present in both pathological discs and subjects with facet joint disorders (Kang et al., 1997; Reinecke et al., 2000). In the absence of any neural compression, cytokines have been shown to induce both mechanical hypersensitivity in nerve tissue and hence pain, and structural and functional nerve injury (Olmarker and Rydevik, 2000). It is therefore possible that any degenerative change in the thoracic spine could, via chemical irritation, produce an intercostal neuropathy. Antidromic action potentials in affected neurones may cause the release of chemicals such as substance P in the innervated tissues. Antidromic chemical release has been shown to correlate with the severity of an experimental arthropathy (Levine et al., 1984). The application of an oscillatory postero-anterior mobilisation technique over the articular pillars and rib angles may have created a beneﬁcial physiological effect both to an affected articulation or the adjacent ventral ramus (Elvey, 1986). At the grades of treatment techniques used (III–V) (Maitland et al., 2005) it is likely that there would have been stimulation of both A-beta ﬁbres and nociceptors. A-beta activation may, via inhibitory interneurones, presynaptically inhibit afferent nociceptive transmission in the dorsal horn. Nociceptive stimulation may, via dorsal periaqueductal gray (dPAG) stimulation, create descending noradrenergic inhibition (Lovick, 1991; Wright, 1995; Skyba et al., 2003). Manual therapy may also activate serotonergic descending inhibitory pathways. It is postulated that the dPAG sends projections to the rostro-ventromedial medulla where the descending serotonergic ﬁbres originate (Skyba et al., 2003). Ectopic impulse generation from pathological sites along axons may be a cause of both orthodromic and antidromic action potentials.

ARTICLE IN PRESS 86

M. Ian Rabey / Manual Therapy 13 (2008) 82–86

Spontaneous activity in neuromas may be suppressed by repeated C ﬁbre stimulation at a frequency between 1 and 5 Hz (Devor and Dubner, 1998). The degree and frequency of stimulation applied during the mobilisation techniques would certainly fulﬁl these criteria. The potential physiological effects of the manual therapy techniques may therefore have led to resolution of the neuropathy postulated to underlie both the pain and swelling. The author was unable to ﬁnd any documentary evidence of the natural history of this type of disorder. It should, however, still be considered that the symptom resolution could simply have been due to this, or to a placebo effect that occurred with treatment.

5. Conclusion These two case reports detail the successful manual therapy intervention applied to two cases of ‘‘costochondritis’’. One may postulate that neuropathic sequelae are the underlying cause of both the symptoms and signs in these two patients who ﬁtted a diagnosis that is commonly based on exclusion of other disease processes and basic clinical examination ﬁndings. This postulate may explain why both subjects responded positively to techniques administered to the posterior spinal articulations whilst no intervention was applied to the costochondral joints. In this group of patients further investigation into the underlying pathophysiological mechanisms may be warranted. Meanwhile clinicians are encouraged to consider that treatment directed towards posterior spinal structures might be indicated in cases with a similar presentation. References Bogduk N. The nervous system. In: Palastanga N, Field D, Soames R, editors. Anatomy and human movement. 2nd ed. London: Butterworth Heinemann; 1994. p. 795–878. Bogduk N, Engel R. The menisci of the lumbar zygapophyseal joints. A review of their anatomy and clinical signiﬁcance. Spine 1984;8(5):454–60. Boyle J. Is the pain and dysfunction of shoulder impingement lesion really second rib syndrome in disguise? Two case reports. Manual Therapy 1999;4(1):44–8. Boyle J, Singer K, Milne N. Pattern of intervertebral disc degeneration in the cervicothoracic junctional region. Manual Therapy 1998;3(2):72–7. Boyling J. Editorial. Manual Therapy 1999;4(2):61–2. Capewell S, McMurray J. ‘‘Chest pain-please admit’’: Is there an alternative? British Medical Journal 2000;320:951–2. Devor M, Dubner R. Centrifugal activity in afferent C-ﬁbers inﬂuences the spontaneous afferent barrage generated in nerve end neuromas. Brain Research 1998;446:396–400.

Elvey R. Treatment of arm pain associated with abnormal brachial plexus tension. Australian Journal of Physiotherapy 1986;32: 225–30. Erwin W, Jackson P, Homonko D. Innervation of the human costovertebral joint: implications for clinical back pain syndromes. Journal of Manipulative and Physiological Therapeutics 2000;23(6):395–403. Fallon J. Chiropractic manipulation in the treatment of costovertebral joint dysfunction with resultant intercostal neuralgia during pregnancy. Journal of the Neuromusculoskeletal System 1996;4(2):73–5. Kang J, Stephanovic-Racic M, McIntyre L, Georgescu H, Evans C. Toward a biochemical understanding of human intervertebral disc degeneration and herniation. Spine 1997;22:1065–73. Levine J, Clark R, Devor M, Helms C, Moskowitz M, Basbaum B. Interneuronal substance P contributes to the severity of experimental arthritis. Science 1984;222(4674):547–9. Lovick T. Interactions between descending pathways from the dorsal and ventrolateral periaquaeductal gray matter in the rat. In: Depaulis A, Bandler R, editors. The Midbrain periaquaeductal gray matter. New York: Plenum Press; 1991. p. 101–34. Maitland G, Hengeveld E, Banks K, English K. Maitland’s vertebral manipulation, 7th ed. London: Butterworth Heinnemann; 2005. Mayou R, Bryant M, Sanders D, Bass C, Klimes I, Forfar C. A controlled trial of cognitive-behavioural therapy for non-cardiac chest pain. Psychological Medicine 1997;27:1021–31. Mendelson G, Mendelson H, Horowitz S, Goldfarb C, Zumoff B. Can sup 99 m technetium methylene diphosphonate bone scans objectively document costochondritis? Chest 1997;6:1600–2. Mense S. Considerations concerning the neurobiological basis of muscle pain. Canadian Journal of Physiology and Pharmacology 1991;69:610–6. Olmarker K, Rydevik B. Selective inhibition of TNF—a prevents the nucleus pulposus-induced reduction of nerve conduction activity. Proceedings of international society for the study of the lumbar spine, Adelaide; 2000. pp. 13. Okawa A, Haro H, Takahashi M, Komori H, Shinomiya K, Yokoyama M. Biochemical response of the intervertebral disc cells induced by an acute shear force in a rat tail model. Proceedings of international society for the study of the lumbar spine, Adelaide; 1998. pp. 27. Reinecke J, Wehling P, Granrath M, Meijer H. The use of interleukin1-receptor antagonist (IL-1 ra) in the treatment of ‘‘lumbar facet syndrome’’. Proceedings of international society for the study of the lumbar spine, Adelaide; 2000. pp. 173. Scott E, Scott B. Painful rib syndrome—a review of 76 cases. Gut 1993;34:1006–8. Skyba D, Radhakrishnan R, Rohlwing J, Wright A, Sluka K. Joint manipulation reduces hyperalgesia by activation of monoamine receptors but not opioid or GABA receptors in the spinal cord. Pain 2003;106:159–68. Triano J, Erwin M, Hanson D. Costovertebral and costotransverse joint pain: a commonly overlooked pain generator. Topics in Clinical Chiropractic 1999;6(3):79–93. Wadhwa S, Phan T, Terei O. Anterior chest wall pain in postpartum costochondritis. Clinics in Nuclear Medicine 1999;24(6): 404–6. Woolf C, Doubell T. The pathophysiology of chronic pain—increased sensitivity to low threshold A-beta ﬁbre inputs. Current Opinion in Neurobiology 1994;4:525–34. Wright A. Hypoalgesia post-manipulative therapy: a review of a potential neurophysiological mechanism. Manual Therapy 1995;1: 11–6.

ARTICLE IN PRESS

Manual Therapy 13 (2008) 87–88 www.elsevier.com/locate/math

Chiropractic Osteopathic & Physiotherapy Student Conference Prize-winning Abstract & Presentation (September 2007)

An examination of pressure-pain thresholds (PPT’s) at myofascial trigger points (MTrP’s), following muscle energy technique or ischaemic compression treatment Jamie Dearing, Fiona Hamilton London School of Osteopathy (LSO), 56-60 Nelson Street, London E1 2DE, UK

Purpose: To provide quantitative data to inform technique selection when treating myofascial trigger points (MTrP’s). Relevance: Osteopathy claims pain reduction as one of its main purposes, yet there is limited quantitative research into the pain modifying effects of common osteopathic techniques. This study examines the pain modifying effects of two common techniques: ischaemic compression and muscle energy technique. Participants: For this controlled study, 50 asymptomatic subjects were selected randomly from the student base of LSO. Inclusion required the presence of a latent or active MTrP in the upper ﬁbres of the left trapezius muscle. Exclusion criteria: a known musculoskeletal pathology, clotting disorder or taking anti-coagulants. A further ﬁve subjects participated in a pilot study. Subjects were randomly allocated to one of two treatment groups (ischaemic compression or muscle energy technique) or a control group. Methods: Subjects received either a 90 s ischaemic compression technique, a muscle energy technique or listened to 3 mins of relaxing music (control). Before and after treatment their pressure-pain thresholds (PPT’s) were measured at a MTrP located in the left upper trapezius muscle. Analysis: The statistical analysis was conducted at a 95% conﬁdence level. The mean and standard deviation Corresponding author.

E-mail address: [email protected] (J. Dearing). 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.10.001

were calculated for each variable, based on the average of a pair of PPT readings taken either pre-treatment or post-treatment. Dependant t-tests were used to compare the pre-treatment pressure to the post-treatment pressure data within each group. The effect sizes of each treatment were calculated using Cohen’s d. Normal distribution was assessed using the Kolmogorov and Smirnov method. A one-way analysis of variance (ANOVA) was used to investigate the signiﬁcant difference between groups. Results: MTrP sensitivity was signiﬁcantly reduced following ischaemic compression (Po0.001) and the muscle energy treatment (Po0.001), but not in the control (P40.05). ANOVA analysis revealed no signiﬁcant difference between the two treatment groups (P40.05). However the within-group effect size was measured using Cohen’s d and found to be large for the ischaemic compression group (d ¼ 0.77), medium to large for the MET group (d ¼ 0.64), and less than small for the control group (d ¼ 0.12). Conclusion: Both ischaemic compression and muscle energy technique produce a signiﬁcant reduction in pain sensitivity at MTrP’s in the upper trapezius muscle. Ischaemic compression appeared to be more effective than muscle energy technique in reducing pain sensitivity at these trigger points in asymptomatic subjects. Implications: This data provides quantitative support for the use of both these techniques in the treatment of MTrP’s by osteopaths. Of the two treatments studied, ischaemic compression appears to have a stronger effect. This study should be extended to cover symptomatic

ARTICLE IN PRESS 88

COP Presentation Abstract / Manual Therapy 13 (2008) 87–88

patients and over a longer period to further inform clinical effectiveness. Keywords: Muscle energy technique; Ischaemic compression; Pressure-pain threshold; Myofascial trigger point

Biography: Jamie Dearing is a New Zealander who has been living and working in the UK for the last 7 years. He is a recent, prize-winning graduate of the London School of Osteopathy. Prior to this he worked in the ﬁnance industry as a business analyst. He lives with his wife and young son in North London where he works as an osteopath in private practice.

Funding acknowledgement(s): Unfunded Ethical approval: This study was approved by the London School of Osteopathy’s Research Ethics Committee.

Earlier presentations: This study was presented in September 2007 at the COP Conference held at the BSO in London for recent graduates of Chiropractic, Osteopathy and Physiotherapy.

Manual Therapy (2008) 13(1), 89

Available online at www.sciencedirect.com

Diary of events

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

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

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

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

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

89

ARTICLE IN PRESS

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

Letter to the Editor Response to: Brandt, C., Sole, G., Krause, M.W., Nel, M., 2007. An evidence-based review on the validity of the Kaltenborn rule as applied to the glenohumeral joint. Manual Therapy 12 (1), 3–11

the axis moves while the joint surfaces are rolling and gliding. These arthrokinematic movements must be considered when analysing joint motion!

Kaltenborn’s rule conﬁrmed! Brandt et al. (2007) concluded in their high quality literature review, that evidence is lacking regarding the direction of translation of the humeral head on the glenoid surface. They write that Kaltenborn’s convex– concave rule of convex–concave gliding motions may be valid for stretching the tight capsulo-ligamentous structure causing limitation of the physiological joint motion. But done by the therapist this movement may not mimic the true gliding taking place during motion. Reading this interesting paper a misinterpretation seems to have taken place, caused probably by different terminology for the description of motion. The authors are analyzing the movement of the centre of the head of humerus, while Kaltenborn’s rule describes the movements of the joint surfaces. Both are right! Based on anatomic research of MacConaill, Kaltenborn has applied the mechanical principle of the lever law to joints. Moving a bone like humerus around his axis of rotation for example in abduction, the part of the bone on one side of the axis will lift up (the shaft of humerus) and the part of the bone on the other side of the axis will go downwards (the joint surface of the humeral head). The axis of rotation is always in the convex bone.

Therefore moving the convex bone you are moving a lever of the ﬁrst order and the joint surface of the convex bone will glide into the opposite direction of the movement of the shaft. When moving the scapula, however, all the bone is on one side of the axis which remains in the head of humerus. Therefore you are moving a lever of the second order and the scapula as well as the glenoid surface will move into the same direction.

This is a description of movements of the joint surfaces with a stationary axis of rotation. But normally 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.07.003

Pure rolling occurs when the axis of motion is at the contact point of the joint surfaces—as in hypomobile joints at the end of range of motion (Jordan, 1963). In this case the whole bone is on one side of the axis and the joint surface will not glide. The centre of the joint surfaces, for example the centre of the head of humerus, will move in the direction of the movement of the shaft, i.e. in the direction of rolling. Pure gliding ( ¼ spin) occurs, when the axis of rotation passes through the contact point of the joint surfaces (MacConaill and Basmajian, 1977, p. 14). This happens for example when moving around the longitudinal axis of the neck of humerus. The centre of the head of humerus will remain stationary.

During physiological active and passive movements almost always rolling and gliding occurs simultaneously. When moving the humerus in abduction, rolling happens in an upward direction and therefore the centre of humerus will translate upwards, while the joint surface will glide downwards to a more or lesser amount. Let us analyze for example the results of Baeyens et al. (2000), who also have claimed that Kaltenborn’s rule has to be revisited. These authors have examined 3-D images of three healthy athletes moving their shoulder in 901 of abduction into more and more external rotation until the ﬁnal position of the apprehension test. They found a posterior translation of the centre of the humeral head (of about 4 mm). From an arthrokinematic point of view (and according to Kaltenborn’s rule), the humeral joint surface is rolling posteriorly while simultaneously gliding anteriorly. The posterior rolling of course moves also the centre of the head of humerus posteriorly. There is no contradiction therefore to Kaltenborn’s rule, but there are two different points of view. In another study Baeyens et al. (2001) examined three athletes with a minor anterior instability of the shoulder with a similar design. They showed that the centre of the humeral head remained centralized on the glenoid surface. This is interpreted as an insufﬁciency of the anterior part of the inferior glenohumeral ligament which does not induce the posterior motion. From an arthrokinematic

ARTICLE IN PRESS e2

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

point of view this means excessive anterior gliding. Kaltenborn’s rule predicts anterior gliding for this motion which might be too much in case of hypermobility for example when ligaments are insufﬁcient. In conclusion the direction of the (translational) movement of the head of humerus is in positive correlation with the rolling of the joint surfaces, i.e. into the same direction. The movement of the joint surfaces are following the mechanical principles of a lever. Kaltenborn’s convex rule is following the principle of a lever of the ﬁrst degree and the concave rule the principle of a lever of the second degree. Further research has to differentiate between the movement of the centre of the head of humerus and the movement of the joint surfaces when analyzing joint motions. There is scientiﬁc evidence for the movements of the centre of the head of humerus. Kaltenborn’s rule has an anatomic basic. Evidence can be seen on dynamic radiographs. It is a rule that helps the physiotherapist in making his clinical decision. Kaltenborn’s rule is a mechanical principle applied to human joints and therefore as valid as the mechanical principle itself.

References Baeyens J-P, van Roy P, Clarys JP. Intra-articular kinematics of the normal glenohumeral joint in the late preparatory phase of throwing: Kaltenborn’s rule revisited. Ergonomics 2000;43(10): 1726–37. Baeyens J-P, van Roy P, de Schepper A, Declercq G, Clarijs J-P. Glenohumeral joint kinematics related to minor anterior instability of the shoulder at the end of the late preparatory phase of throwing. Clinical Biomechanics 2001;16:752–7. Brandt C, Sole G, Krause MW, Nel M. An evidence-based review on the validity of the Kaltenborn rule as applied to the glenohumeral joint. Manual Therapy (available online at /www.sciencedirect. comS; 2007). Jordan HM. Orthopedic appliances. Springﬁeld: Charles C. Thomas Publisher; 1963. MacConaill MA, Basmajian JV. Muscles and movements, a basis for human kinesiology. Huntington, NY: Robert E. Krieger Publishing Company; 1977.

J. Schomacher Dorfstr. 24, CH-8700 Kuesnacht, Switzerland E-mail address: [email protected]

ARTICLE IN PRESS

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

Letter to the Editor The validity of clinical measures of patella position I read with interest the paper by McEwan et al. (2007) with the aim of the study to assess the validity of the clinical assessment technique of patella medio-lateral position and patella lateral tilt against the criterion measure of MRI. McEwan stated the main statistical analysis was ‘‘The degree of agreement between the two techniques (MRI and clinical) was quantiﬁed by calculation of Pearson’s product moment.’’(p. 228). However, a correlation coefﬁcient does not measure agreement. Correlation means that as one measure goes up so does the other. The other measure could go up at twice the rate of the other and the correlation would still be high. The well-referenced paper by Bland and Altman (1986) may recommend that a Bland–Altman plot, which displays the difference between methods against their mean would be more informative.

Before McConnell’s (1986) method of patella assessment could be considered valid, a more robust analysis of agreement needs to be carried out between MRI and manual measurement of patella position. References Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, 1986, i 307-10. Also available online /http://www-users.york.ac.uk/ mb55/meas/ba.htmS [accessed July 2007]. McConnell J. The management of Chondromalacia Patellae: a long term solution. Australian Journal of Physiotherapy 1986;32: 215–22. McEwan I, Herrington L, Thom J. The validity of clinical measures of patella position. Manual Therapy 2007;12(3):226–30.

Christopher Barnett Royal Newcastle Centre, Newcastle, NSW, Australia E-mail address: [email protected]

1356-689X/$ - see front matter Crown Copyright r 2007 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.09.005

ARTICLE IN PRESS

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

Letter to the Editor Re: Thomas LC, et al. Premanipulative testing and the velocimeter. Manual Therapy (2007), doi:10:1016/j.math. 2006.11.003 Dear Editor I believe that there is sufﬁcient evidence to support the replacement of the provocative tests with velocimetry, and that Thomas et al. (2007) have over-stated the limitations of Doppler and those of related validity/reliability studies. I advocated the merit of velocimetry based on the early validity trials, and also on the later validity and reliability studies that I made (Haynes, 2002). The authors stated that there are problems trying to extrapolate the ﬁndings of the earlier studies to both the type of velocimeter, and patients considered for cervical manipulation, but perhaps their arguments are ﬂawed. Thomas et al. claimed that the CW Doppler units in the earlier studies provided more quantitative data than velocimeters would normally do. However, these data would be prone to major inaccuracy, because the Doppler angle is unable to be determined without imaging the vertebral artery, and it seems that there is no evidence to support any advantage in having such data in this case. Evidence that quantitative data are unnecessary comes from the high concordance between the results of velocimeter operating on audio alone and duplex US scanning in the study by Haynes (2000). Hence, their concern about the types of velocimeters seems to be unwarranted. They explained that many of the patients in the validity trials had severe cardiovascular disease, conditions which would count as contraindications to neck manipulation. The major exceptions were the neurologically asymptomatic patients in the Hennerici et al. (1981) and Haynes (2000) studies. The authors used White’s opinion that the relative rarity of marked stenosis and occlusion of vertebral arteries would have assisted velocimeter operators to correctly guess normality for the ambiguous cases, and yielded artiﬁcially high speciﬁcity levels. However, duplex scanning results can also be uncertain (Zwiebel, 1992), and the relevant validity trials recruited vertebrobasilar patients as well (Visona et al., 1986; de Bray et al., 2001). So, on the basis of the authors’ argument, which seems illogical, these validity studies would be dismissed too, and duplex scanning rendered unsuitable as a reference standard for the type of velocimeter validity trial that they recommended. 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.09.008

Under Section 7 Thomas et al. (2006) stated that a potential problem with velocimetry is that it has poor reliability, without providing a reference to back this claim. The only subsequent mention of reliability in this section apparently referred to one of my papers, but I have never suggested that the averaging of multiple ‘‘measurements’’ is required to enhance ‘‘reliability’’. They seem to have included Zaina et al. opinion that rotational effects on vertebral artery blood velocities are a variant of normal and hence are clinically unimportant, to suggest thaty ‘‘This presents some difﬁculty in determining the accurate interpretation of velocimeter examinationy’’ Paradoxically, Thomas, et al expressed a view earlier in Section 4 where they acknowledged that it appears sensible to refer for further examination if these rotational effects were detected in case the artery is weakened by arteriopathy. This is clinically important, as is rotational stenosis induced vertebrobasilar ischaemia when collateral supply is inadequate. References de Bray JM, Pasco A, Tranquart F, Papon X, Alecu C, Giraudeau B, et al. Accuracy of colour—Doppler in the quantiﬁcation of proximal vertebral artery stenosis. Cerebrovascular Disease 2001; 11:335–40. Haynes MJ. Vertebral arteries and neck rotation: Doppler velocimeter and duplex results compared. Ultrasound in Medicine and Biology 2000;26:57–62. Haynes MJ. Vertebral arteries and cervical movement: Doppler ultrasound velocimetry for screening before manipulation. Journal of Manipulative and Physiological Therapeutics 2002;25:556–67. Hennerici M, Aulich A, Sandman W, Freund H. Incidence of asymptomatic extracranial arterial disease. Stroke 1981;12:751–7. Thomas LC, et al. Premanipulative testing and the velocimeter. Manual Therapy (2007) doi:10.1016/j.math.2006.11.003. Visona A, Luisani L, Castellani V, Ronisvalle G, Bononome A, Pagan A. The echo-Doppler (duplex) system for the detection of vertebral artery occlusive disease. Comparison with angiography. Journal of Ultrasound in Medicine. 1986;5:247–50. Zwiebel WJ. Duplex Vertebral Examination. In: Zwiebel WJ, editor. Introduction to vascular ultrasonongraphy. Philadelphia: WB Saunder Co., 1992; pp. 133–143, (Chapter 10).

M. Haynes c/o High Wycombe Chiropractic Clinic, 506 Kalamunda Road, High Wycombe 6056, Australia E-mail address: [email protected]

ARTICLE IN PRESS

Manual Therapy 13 (2008) e5–e6 www.elsevier.com/locate/math

Letter to the Editor Comments in response to letters to editor regarding article: Thomas LC, et al. Premanipulative testing and the velocimeter. Manual Therapy (2007) doi:10.1016/ j.math.2006.11.003

Comments in response to letter 1 We thank Dr. Haynes for his letter. His comments provide additional perspectives on this topic. He suggests that we are concerned with the lack of quantitative data supplied by the velocimeter, however, our problem is more to do with the lack of imaging capabilities of continuous wave ultrasound units in general. This is likely to be particularly problematic with inexperienced operators. The wide variability in size of the vertebral artery and its tortuosity in the upper cervical region make it more likely that operator error will be high in the absence of visual tracking of the vessel which is provided by duplex investigation. We do appreciate that the lack of quantitative data may perhaps be offset, somewhat, by the sensitivity of the human ear to monitor the audio output of the velocimeter. We agree with Dr. Haynes that there are difﬁculties with the accuracy of duplex scanning and appreciate his provision of additional literature, not included in our review, that supports this. Indeed, the later version of Zwiebel’s text (Zwiebel, 2000) discounts the use of duplex scanning for vertebral artery investigation altogether. The purpose of our discussion of the papers reported by Haynes (2002) with which he validates the use of the velocimeter, was to critically consider the validity of continuous wave ultrasound compared with angiography. It was apparent to us that while these papers appeared to show good concordance of continuous wave ultrasound with angiography as reported by Haynes (2002), in fact the patient groups in which the blood ﬂow was examined in these papers were quite different to those likely to attend and be assessed in a manual therapy clinic in respect of neck manipulation. In addition, some of these studies examined different segments of the vertebral artery. Therein lies our concern, in essence an extrapolation of these results to 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.09.010

a manipulation treatment population which we believe is arguably inappropriate. Our comments about reliability refer to the protocol of taking measurements at two separate locations (C0-1and C2 portals) (Haynes et al., 2000). Haynes recommends that if blood ﬂow changes are detected at the C0-1 level, the C2 level should then be explored to conﬁrm the initial ﬁndings. The suggestion of repeating and averaging blood ﬂow measures has also been raised by Johnson et al. (2000) who suggest taking a number of blood ﬂow measures and averaging them. We believe that averaging of the data may be misleading unless sufﬁcient numbers of repeat scans are undertaken and this may be quite impractical. The apparent paradox in our suggestion that rotational effects on blood ﬂow detected during provocative positional testing should be followed up but may in fact be normal variants is dependent on the observed changes on rotation being related to symptom reproduction. Obviously if signs or symptoms of vascular insufﬁciency are elicited during neck rotation it would be prudent to investigate this more fully. The clinical challenge is if blood ﬂow changes are detected on velocimeter testing but the patient is asymptomatic. Reasonable clinical acumen would suggest referral of the patient for further investigation. However, the rate of false positive ﬁndings for velocimeter examination is currently uncertain.

Comments in response to letters 2 and 3 We did not state that a clear link has been established between cervical manipulation and stroke, but rather that cervical manipulation had associated risk(s) that may result in vertebrobasilar stroke. See Senstad et al. (1997) and Di Fabio (1999) cited in our paper. We acknowledge the more recent discussion in the chiropractic literature concerning cervical manipulation and adverse events but this was not the aim of our review paper. It is established practice in the professions practicing manual therapy to include pre-manipulative testing. In this regard, we also note the comments made by the author of letter 3. It is the deﬁning and the subsequent challenge of improving the validity of such testing that is the thrust of our review.

ARTICLE IN PRESS e6

Letter to the Editor / Manual Therapy 13 (2008) e5–e6

Our concern about unblinded studies of ultrasound examination centred on the comparison of continuous wave ultrasound with angiography, not duplex scanning. We note with interest these later duplex studies referred to in letter 1 but not included in our review. We did not suggest ways in which the velocimeter technique might be enhanced. We did, however, highlight the difﬁculties we perceive may be associated with its use for testing vertebral artery blood ﬂow. Our paper was not a technical paper proposing new protocols or modifying the protocols suggested by others, but rather a review of the existing literature concerning the use of the velocimeter to examine vertebral artery blood ﬂow. We agree with the authors of letters 2 and 3 that duplex ultrasound does not represent the criterion standard for investigating vertebral artery blood ﬂow. However, if the velocimeter is to be a useful clinical screening tool it would be reasonable to expect that velocimeter usage should identify patients with blood ﬂow changes that at least duplex scanning would also identify. That is, it should perform with the same or better levels of sensitivity and speciﬁcity than duplex scanning. The more complex investigation of angiography or MRI (the criterion standard) could then be pursued based on objective evidence at least equivalent to duplex ﬁndings. If the velocimeter does not meet this duplex standard its use may in fact simply complicate matters by delivering false negative and false positive ﬁndings, as well as potentially adding an additional investigation (and cost) that may have little if any clinical utility for the patient or practitioner. We note the additional advice provided by the author of this second letter concerning the validity and reliability of the studies by Haynes which is evidenced by reference to a ‘personal communication’ and the citation of an additional paper (Arning, 2001). Nevertheless, we are not particularly convinced by this

argument. The paper by Arning (2001) was not included in our review as we only looked at English language journals. Interestingly, this study was also performed on consecutive patients attending a neurology clinic, and all patients were examined with both duplex and continuous wave ultrasound by the same operator who would therefore not have been blinded to the results. The conclusion drawn, that continuous wave ultrasound may be useful in the clinic in the hands of an experienced operator, is therefore questionable. References Arning C. Does CW ultrasound still have a value for cerebrovascular diagnosis. Nervenartzt 2001;72(8):625–8. Di Fabio RP. Manipulation of the cervical spine: risks and beneﬁts. Physical Therapy 1999;79:50–65. Haynes M. Vertebral arteries and cervical movement: Doppler ultrasound velocimetry for screening before manipulation. Journal of Manipulative and Physiological Therapeutics 2002;25(9):556–67. Haynes M, Hart R, McGeachie J. Vertebral arteries and neck rotation: Doppler velocimeter interexaminer reliability. Ultrasound in Medicine and Biology 2000;26(8):1363–7. Johnson C, Grant R, Dansie B, Spyropolous P. Measurement of blood ﬂow in the vertebral artery using colour duplex Doppler ultrasound: establishment of the reliability of selected parameters. Manual Therapy 2000;5(1):21–9. Senstad O, Leboeuf-Yde C, Borchgrevink C. Frequency and characteristics of side effects of spinal manipulative therapy. Spine 1997;22(4):435–41. Zwiebel WJ. Introduction to Vascular Ultrasonography. Philadelphia, Pensylvania: Harcourt Health Sciences; 2000.

Lucy C. Thomas, Darren A. Rivett, Philip S. Bolton Discipline of Physiotherapy, School of Health Sciences, University Drive, The University of Newcastle, Callaghan 2308, Australia E-mail address: [email protected] (L.C. Thomas)

ARTICLE IN PRESS

Manual Therapy 13 (2008) e7–e8 www.elsevier.com/locate/math

Letter to the Editor Letter to the Editor: Thomas LC, et al. Premanipulative testing and the velocimeter. Manual Therapy (2007) doi:10.1016/j.math.2006.11.003

Dear Editor, I read with interest the review by Thomas et al. (2007) who stated in their abstract that the link between stroke and cervical manipulation has been clearly demonstrated, which presumably underpinned their case for pre-manipulative screening. However, the two case control studies that addressed this issue (Rothwell et al., 2001; Smith et al., 2003) have limitations that make their results inconclusive (Rubinstein et al., 2005). Despite this, I believe that it is clinically wise to accept the evidence for the sake of patient safety, and to do likewise for a role for velocimetry in premanipulative screening despite some relatively minor limitations. They explained that early studies comparing ‘‘ultrasound’’ with arteriography were unblinded, but it needs to be stated explicitly that the term ‘‘ultrasound’’ also refers to early duplex scanning validity studies (Visona et al., 1986). Later validity studies of duplex scanning have been blinded (de Bray et al., 2001), as has a trial comparing vertebral artery velocimetry with duplex scanning (Haynes, 2000). In both instances the results, which indicated high validity, have agreed with the earlier unblinded studies, suggesting that this limitation was relatively unimportant. The authors discussed the problem of vessel misidentiﬁcation and other technical difﬁculties that can occur during vertebral artery examination using velocimetry. What they omitted acknowledging, are the methods that an examiner can use to help minimise these difﬁculties (Haynes, 2002), which appear to have assisted in obtaining good concordance in the validity and reliability studies. They suggested that the full range of ambiguous results for Haynes’ validity and reliability trials could not be tested because the patients were pre-selected through Doppler velocimetry, and that this represents a particular limitation of the studies. Instead, the authors advocated pre-selecting participants using the criterion standard, but it seems that validity 1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.09.009

trials of duplex scanning failed to satisfy this condition also. Hence, it appears inconsistent to recommend this design for studies on velocimetry, especially considering that the authors also suggested that duplex scanning be the criterion standard. Haynes ensured that 10.3% of 39 velocimeter examinations (personal communication) in the validity trial (Haynes, 2000) gave ambiguous results, and 12.5% (n ¼ 40) for the reliability trial (Haynes et al., 2000). The proportions were similar to the 12% of 500 vertebral artery velocimeter examinations that Arning and Does (2001) deemed to be unclear, which indicates that the proportions of ambiguous results in Haynes’ studies were reasonably representative. Haynes’ experimental designs were robust, largely because they were based on the correct concordance statistic, kappa (Haas, 1991). For the kappa statistic to remain stable, sufﬁcient proportions of positive and negative results are needed to prevent what is known as limited variability in the results. Given that marked rotational stenosis occurs in 5% or less of vertebral arteries, randomly selecting patients using duplex scanning, as apparently recommended by the authors, would have caused limited variability, thus instability of the kappa statistic, and presented as a major limitation to the study. References Arning C, Does CW. Doppler ultrasound still have a value for cerebrovascular diagnosis today?. Nervenarzt 2001;72:625–8. de Bray JM, Pasco A, Tranquart F, Papon X, Alecu C, Giraudeau B, et al. Accuracy of colour-Doppler in the quantiﬁcation of proximal vertebral artery stenosis. Cerebrovacsular Dis 2001;11: 335–40. Haas M. Statistical methodology for reliability studies. J Manipulative Physiol Ther 1991;14:119–31. Haynes MJ. Vertebral arteries and neck rotation: doppler velocimeter and duplex results compared. Ultrasound Med Biol 2000;26: 57–62. Haynes MJ, Hart R, McGeachie J. Vertebral arteries and neck rotation: doppler velocimeter interexaminer reliability. Ultrasound Med Biol 2000;26:1363–7. Rothwell DM, Bondy SJ, Williams JI. Chiropractic manipulation and stroke: a population-based case–control study. Stroke 2001;32: 1054–60. Rubinstein SM, Peerderman SM, van Tulder MW, Riphagen I, Haldeman S. A systematic review of the risk factors for cervical artery dissection. Stroke 2005;36:1575.

ARTICLE IN PRESS e8

Letter to the Editor / Manual Therapy 13 (2008) e7–e8

Smith WS, Johnston SC, Skalabrin EJ, Weaver M, Azari P, Albers GW, et al. Spinal manipulative therapy is an independent risk factor for vertebral artery dissection. Neurology 2003;60: 1424–8. Thomas LC, et al. Premanipulative testing and the velocimeter. Manual Theraoy, 2007, doi:10.1016/j.math.2006.11.003. Visona A, Luisani L, Castellani V, et al. The echo-Doppler (duplex) system for the detection of vertebral artery occlusive disease.

Comparison with angiography. J Ultrasound Med 1986;5: 247–50.

Vincent Karl 83160 La Valette, Var France E-mail address: [email protected]

ARTICLE IN PRESS

Manual Therapy 13 (2008) e9 www.elsevier.com/locate/math

Letter to the Editor Re: Thomas LC, et al. Premanipulative testing and the velocimeter. Manual Therapy (2007) doi: 10.1016/ j.math.2006.11.00

Dear Editor, I welcome the debate on this topic, which is very important for all practitioners of cervical manipulation, and commend Thomas et al., for their efforts in discussing it in a clear manner generally. Unfortunately, many of their criticisms of velocimetry appear to be unfounded with a number of arguments being inconsistent, and much relevant information, which is necessary to view velocimetry in a correct perspective, missing. However, I will leave the details of these concerns for others to comment on. As I understand, the authors recommended that further research on the capability of velocimetry is needed to determine whether it is superior to the current pre-manipulative screening protocol. It needs to be noted though that it is obvious from the discussions in the validity studies of duplex scanners that the scanners were used to examine vertebral arteries prior to studies demonstrating conclusively that they had greater capability than velocimetry, probably because they seemed to be promising. Currently, there is strong evidence that provocative tests lack validity, and the review authors at least accept that velocimetry appears to be promising. So, duplex scanning was seen as a potentially viable alternative to velocimetry that was considered then (and even more recently by Arning in 2000) to be useful still. Whereas, the only alternative to safe, non-invasive velocimetry that is available to practitioners of manipulation are the potentially hazardous provocative tests that are increasingly recognised as being useless! If it was decided to

1356-689X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.math.2007.09.007

wait for more studies, which could take a long time to complete, how could manipulation practitioners assess vertebral artery patency during the interim? I think that future research can be done as velocimetry is incorporated into pre-manipulative screening, which would actually assist in these studies as more practitioners become proﬁcient in the technique. After careful consideration of the available evidence, the scientiﬁc committee of the Franco-European Chiropractic Society (SOFEC) has recently recommended that members of SOFEC adopt velocimetry of vertebral arteries in the pre-manipulative screening protocol (http://www.sofec.org). There are members of the society who aim to facilitate future research on the capabilities of Doppler, and in the development of innovations that may improve the technique. Earlier this year the Institute of Franco-European Chiropractic (IFEC) decided to make it mandatory for all of its chiropractic students to be taught velocimetry, because they considered it to be the only reasonable approach, despite having some limitations. An analogy would be choosing an alarm system that is a vitally important for safety, where there are only two options; one that has been demonstrated to be ineffective, and another which has some evidence that it works, but where absolute proof is lacking. Best evidence practice relies on a mixture of common sense as well as scientiﬁc evidence.

Cyril Fischhoff University Diploma (3rd cycle) Methodology and Practice in Epidemiology, Scientific Committee of the Franco-European Chiropractic Society (SOFEC), Board of Governors of the Institut Franco-Europeen de Chiropratique (IFEC), Mauritius E-mail address: cyril.ﬁ[email protected]

ARTICLE IN PRESS

Manual Therapy 13 (2008) e10–e11 www.elsevier.com/locate/math

Book Review Science-based Rehabilitation; Theories into Practice, Kathryn Refshauge, Louise Ada, Elizabeth Ellis. 1st ed. Elsevier Butterworth Heineman (2005). 255pp., ISBN: 0-7506-5564X ‘‘Science-based Rehabilitation; theories into practice’’ is a book, edited by Kathryn Refshauge, Louise Ada and Elizabeth Ellis and published in 2005 Refshauge et al. (2005). It covers the period of 30 years of development in physiotherapy, with special emphasis of the progress in this ﬁeld made in Australia. The book is dedicated to professors Janet Carr and Roberta Sheperd, who have made important contributions to the physiotherapy profession over the last decades. The book offers a collection of works about various aspects of rehabilitation. The ﬁrst and introductory chapter is written by Roberta Sheperd and Janet Carr. It is entitled ‘‘Bridging the gap between theory and practice’’. It is well written and starts with the wonderful sentence: ‘‘Understanding the history of physiotherapy practice enables us to reﬂect on the concept of change and development in clinical practice and to feel more comfortable about the notion that clinical practice quite naturally responds and adapts as new scientiﬁc knowledge emerges.’’ The chapter offers an overview of the development of neurological physiotherapy from the 1950s on to the present. Sheperd and Carr depict—with the help of many examples—the major shift from several ‘‘schools’’ named after their originators (like Bobath, Kabat, Knott and Voss, Ayres and Brunnstrom among others) to recent experimental paradigms and contemporary scientiﬁc understandings. They illustrate how clinical research is enabling us to test the efﬁcacy of interventions. As an example they refer to the re-evaluation of the relative contributions of muscle weakness in stroke patients, requiring signiﬁcant changes in clinical practice. Another example is the ﬁnding from experimental data that strength training does not appear to result in increases in resistance to passive movement (hypertonus) or reﬂex hyperactivity (spasticity). And: experience, learning and active use of affected limbs appear to modulate the adaptive reorganization that inevitably occurs after cortical injury. Many of this kind of research examples and its consequences for physiotherapy praxis are described in this introductory chapter, setting the scene for the following chapters of the book. doi:10.1016/j.math.2007.05.008

The remaining 9 chapters of the book cover the questions, topics and themes like:

‘‘We only treat what it occurs to us to assess: the importance of knowledge-based assessment’’, by Julie Bernhardt and Keith Hill. ‘‘The quest for measurement of infant motor performance’’, by Suzann K. Campbell. ‘‘Muscle performance after stroke’’, Di J. Newham. ‘‘Changing the way we view the contribution of motor impairments to physical disability after stroke’’, Louise Ada and Colleen Canning. ‘‘How muscles respond to stretch’’, Robert Herbert. ‘‘Cardio respiratory ﬁtness after stroke’’, Sharon L. Killbreath and Glen M. Davis. ‘‘Training gait after stroke: a biomechanical perspective’’, Sandra J. Olney. ‘‘Assessment and training of locomotion after stroke: evolving concepts’’, Francine Malouin and Carol L. Richards. ‘‘Strategies to minimize impairments, activity limitations and participation restrictions in Parkinson’s disease’’, Meg Morris.

Because of limited space here I have arbitrarily chosen two chapters to say a little more about: The chapter on ‘‘Muscle performance after stroke’’ departs from the work by Janet Carr and Roberta Sheperd, which changed the approach of physiotherapists to neurological rehabilitation in a number of ways. This chapter offers an overview of recent ﬁndings on muscle strength after stroke, co-contraction during isometric and dynamic contractions, abnormal tone, voluntary activation, strength training and the relationship between muscle performance and function. Deﬁnitions are clearly described, recent empirical ﬁndings summarized and implications for practice emphasized. It has now clearly been shown that muscles are weak after stroke and that the muscle weakness is related to function. They do not get stronger during traditional rehabilitation or spontaneously with increased activity. Also it has been shown that high-intensity contractions, once viewed as something to be avoided at al costs, do not generally increase muscle tone, but in fact do improve strength and function.

ARTICLE IN PRESS Book Review / Manual Therapy 13 (2008) e10–e11

The chapter on ‘‘Changing the way we view the contribution of motor impairments to physical disability after stroke’’ is a well-structured, textbook-worthy resume of positive and negative impairments after stroke, with reference to the great 19th century neurologist Hughlings Jackson. Negative impairments are those that represent a loss of pre-existent function, such as loss of strength and dexterity, whereas positive impairments are additional, such as abnormal postures, increased proprioceptive reﬂexes (i.e. spasticity) and increased cutaneous reﬂexes. In a schematic picture the relationships are shown between motor impairments, negative and positive, and secondary impairments like adaptive behaviors, contractures and decrease in ﬁtness, resulting in several forms of disabilities. The authors have found that classifying impairments after brain damage as negative or positive, is still a useful framework for investigating the underlying causes of disability after stroke. Research ﬁndings corroborate the view that the positive impairment of spasticity is usually only of moderate intensity and has little effect on function after stroke. In addition, the prevailing view that the negative impairments of weakness and loss of dexterity has also been supported by research ﬁndings. This understanding of the nature of both weakness and the loss of dexterity as well as their interaction should prompt us to change, and further develop accurate assessment tools as well as intervention strategies. Of course I could have chosen other chapters to describe, but I rather prefer using the remaining lines for some general remarks about the strengths and limitations of this book. First and foremost, it is an interesting, well-written book covering relevant topics. The layout is all right and very well designed. Of course there is always room for improvement. I would like to read an overall deﬁnition and description of rehabilitation in the introduction. Furthermore, the title

e11

of the book should refer to physiotherapy. The way the book is designed (written by academic physiotherapists) implicates the strong emphasis is on research and practice of physiotherapists. However, rehabilitation requires more often multidisciplinary involvement. Information on the important role of several other professionals in this ﬁeld is therefore needed, since the core features of rehabilitation—as described by Wade (2001)—are:

co-ordinated, multidisciplinary team-work, by a team with expertise and an interest in disability, who actively involve the patient and family in the process, which is set within an explicitly recognized framework encompassing all aspects of illness.

So, the title of this book is not really covering the actual content, because reference to other disciplines than physiotherapy is lacking. However, I highly recommend this book: read it and study it. I really enjoyed it. References Refshauge K, Ada L, Ellis E. Science-based rehabilitation: theories into practice. Edinburgh, London, New York, Oxford, Philadelphia, St Louis, Sydney, Toronto: Elsevier Butterworth Heinemann; 2005. Wade DT. Research into rehabilitation. What is the priority?. Clin Rehab 2001;15(3):229–32.

P. Heuts Hoensbroeck Rehabilitation Centre, Hoensbroek, Netherlands E-mail address: [email protected]

ARTICLE IN PRESS

Manual Therapy 13 (2008) e12 www.elsevier.com/locate/math

Book review Muscle Energy Techniques, L. Chaitow. third ed. Elsevier, Churchill-Livingstone, New York, NY (2006). 346pp., CD included, Price £42.99, ISBN: 10 0443101140 This book aims to be a guide to the use of Muscle Energy Techniques (METs) and as is stated in the forward ‘‘you can’t learn manual skills from a book’’, this task is likely to be a difﬁcult one. But in this book and the supporting interactive DVD a very commendable attempt has been made to provide a source of supporting material to those familiar with the techniques or equally embarking on a course of study in this area. The initial chapters develop the rationale for METs and then demonstrate assessment and treatment techniques. The assessment and treatment techniques are very well described within the text and supported with excellently produced illustrations, photographs and video clips (within the DVD). Perhaps the only disappointment within these early (and subsequent) chapters is the reliance on opinion rather than evidence all too frequently within the text. Though the authors freely admit that the research in this area is in its infancy, it is disappointing that they make no attempt to

doi:10.1016/j.math.2007.05.006

distinguish between research and opinion in the citations they use, this may if the citations are not followed up mislead that there is more research support in the area than is the case. This said, chapter 9 (MET in the Physical Therapy setting) provides a very insightful overview of the research into the use of manual therapy in the treatment of low back pain. The latter chapters of the book investigate and describe the integration of MET into massage and the treatment of sports injuries, both of these chapters providing useful alternate management strategies for a variety of conditions. In summary, this book provides a very good overview of the use of METs, for those new to these techniques it will provide a clear introduction and for those more experienced practitioners an excellent point of reference.

L. Herrington Centre for Rehabilitation and Human Performance Research, Brian Blactchford Builiding, Frederick Road Capmus University, Salford, Greater Manchester M6 6PU, UK E-mail address: [email protected]

ARTICLE IN PRESS

Manual Therapy 13 (2008) e13 www.elsevier.com/locate/math

Book Review Pain in Practice: Theory and Treatment Strategies for Manual Therapists, Hubert van Griensven. Elsevier Churchill Livingstone, London (2005). 218pp., £41.99/ $83.95, ISBN: 075068884 This book aims at being a practical guide for understanding benign pain and the treatment of these pain syndromes. It was speciﬁcally written for manual therapists, but can also be used by other health practitioners in the ﬁeld of musculoskeletal disorders. The book provides practical and clinical reasoning models for explaining, assessing, and treating pain syndromes. The book contains nine chapters. The ﬁrst chapter provides a short historical overview concerning the paradigm shift from a purely biomedical model to a biopsychosocial reasoning model. It points out that the brain is the central element in the perception, processing, and response to pain. Furthermore the author points out that general, or central issues should be addressed before more speciﬁc or peripheral issues are to be treated. The mechanisms that can inﬂuence pain or the experience of pain are discussed brieﬂy in the Chapters 2–7. Chapter 2 describes pain or nociception arising from tissue damage or inﬂammation and it’s normal reaction. Chapter 3 concerns neuropathic and/or neurogenic pain resulting from larger nerve damage or minor neuropathies. Chapters 4 and 5 discuss the function and inﬂuence of the dorsal horn and the autonomic nervous system, respectively. In Chapter 6 psychological issues especially important for

doi:10.1016/j.math.2007.06.002

understanding and treating persons with persistent pain problems are described. The central role of the brain in the perception and regulation of pain is delineated in Chapter 7. Due to the complexity (of the function) of the brain this chapter was limited to models and applied clinical practice. Chapter 8 provides some practical and clinimetric aspects of measuring the consequences of pain and functioning of patients. Finally, Chapter 9 tries to provide an integration of pain models, pain mechanisms and musculoskeletal medicine into a number of assessment principles and treatment principles. This book is clearly written and easy to read. It emphasizes that, especially in case of ongoing pain, assessment and treatment solely at the biomedical level is insufﬁcient. As such the book reﬂects best up-to-date clinical practice for the musculoskeletal therapist. However, in some chapters (Chapters 3–5) a mixture of basic science and non-evidence-based theory and tests or treatments were described. Still, a lack of evidence of effect does not mean a lack of effect. This book is a practical and comprehensive guide for therapists or students who want to know more about pain and its assessment and treatment. For healthcare providers already working in this ﬁeld this book may recapitulate the main issues about pain.

Ilse Swinkels-Meewisse University of Maastricht, 6200 MD Maastricht, Netherlands

ARTICLE IN PRESS

Manual Therapy 13 (2008) e14 www.elsevier.com/locate/math

Book Review Principles of Neuromusculoskeletal Treatment and Management: A Guide for Therapists, N.J. Petty. Churchill Livingstone, New York (2004). 368pp., ISBN: 0443070628 The aim of this book is to make explicit the underlying principles behind treatment and management of patients with neuromusculoskeletal disorders. In this extensive book, the author has succeeded in her aim. It is well written, easy to follow and gives a lot of information about the state of the art in assessment and treatment. This book is divided into 9 chapters in which items as assessment, function and dysfunction of joint, muscle and nerve, principles of treatment strategies and ﬁnally a chapter on principles of patient management (A. Moore). Each chapter is completed with an extensive reference list. Although most of the content is used within curricula of education in physical therapy and/or manipulative therapy, this book brings all the knowledge together, with clear ﬁgures and illustrations. The content is based

doi:10.1016/j.math.2007.09.004

on fundamental or basic research and on that basis it tries to add knowledge to an evidence-based approach of patient management. Unfortunately it does not answer the question if a chosen approach in treatment of muscle, joint or nerve is really effective. The author states that ‘the best treatment is the one that improves the patient’s sign and symptoms in the shortest period of time.’ The reader will be curious about effectiveness and efﬁcacy of chosen treatment strategies, but that was a priori not the primary aim. However, in evidence-based management this knowledge could make the clinical reasoning process more clear. This book is highly recommended for physical therapists and manipulative therapists as a reference book and for students as a textbook.

Jan Pool SOMT Master Education in Manual Therapy, Cobra Research Institute, Amersfoort, The Netherlands E-mail address: [email protected]