International Journal of Mechanical Diagnosis and ... - McKenzie

International Journal of 

Mechanical Diagnosis and Therapy ® 

2009 

Volume 4, No. 1 

Udvidet version 

The McKenzie Institute ® International 

Center for Postgraduate Study in Mechanical Diagnosis and Therapy

Institut for Mekanisk Diagnose og Terapi/ 

McKenzie Institut Danmark� 

�� 

Formand 

Eva Hauge, Dip.MDT 

eva.hauge@nal-net.dk 

kursus@mckenzie.dk 

� 

�� 

Sekretær 

Anne Juul Sørensen, Dip. MDT 

info@mckenzie.dk 

� 

�� 

Kasserer 

Merethe Fehrend, Cert.MDT 

merethe@fehrend.dk 

� 

�� 

Bestyrelsesmedlemmer 

Charlotte Krog, Dip. MDT 

ck@mckenzie.dk 

Simon Simonsen, Dip.MDT 

simon@mckenzie.dk 

Michael Rømer, Cert. MDT 

nsm@webspeed.dk 

�� 

Suppleant 

Anne Mette Anthonsen 

boevlingbjergfysio@netfyssen.dk 

� 

Webmaster 

Martin Nielsen 

fys.martin@gmail.dk 

� 

Ansvarshavende redactor 

Martin Melbye, Dip.MDT 

melbye@mckenzie.dk 

www.mckenzie.dk 

©2007 The McKenzie Institute® 



� 

�� 

Editor 

Helen Clare, PT, PhD, Dip. MDT 

� 

�� 

Editorial Review Board 

Helen Clare, PT, PhD, Dip. MDT 

clare.ha@bigpond.com 

Stuart Horton,PT, Dip. MDT 

stuart.horton@otago.ac.nz 

Sinikka Kilpikoski, MSc, PT, Dip. MDT 

sinikka.kilpikoski@kolumbus.fi 

Stephen May, MSc, Dip. MDT 

s.may@shu.ac.uk 

Mark Werneke, PT, Dip. MDT 

mwerneke@centrastate.com 

�� 

Production 

The McKenzie Institute USA 

Nancy Morden 

Executive Asst. 

nancy@mckenziemdt.org 

� 

The McKenzie Institute International 

Headquarters 

1 Alexander Road 

Raumati Beach 5255 

New Zealand 

� 

Postal Address 

PO Box 2026 



Telephone: +64-4 299-6645 

Facsimile: +64-4 299-7010 

Email: mckinst@xtra.co.nz 

www.mckenziemdt.org 

©2007 The McKenzie Institute®

I n d h o l d 

Volume 4, No. 1 April 2009 

Editorials 

3 

4 

5 

6 

Fra formanden-Mckenzie Institut Danmark 

Eva Hauge, Dip MDT 

Redaktørens 

Brian Sørensen, fysioterapeut 

Guest Editorial 

Colin Davies, PT. Dip. MDT 

Abstracts oversat til dansk 

Simon Simonsen, Dip MDT 

Original Research 

10 Sacro-iliac joint pain: How much of it is there and what might be the problem? 

Hans van Helviort, MA, PT, Dip. MDT, Dip. MT (Neth) and Stephen May, MA FSCP, Dip. MDT, MSc (UK) 

15 Measuring and modifying the frequency and duration of lumbar flexion 

Ronald F. Bybee, PT, DPT, OCS, Dip.MDT, Allyn Byars, PhD, Laura B. Bearden, MPT, 

Tina M. Logan, MPT, Amanda P. Moeller, MPT, Susan J.Hall, PhD (US) 

23 Prognosis of outcome in patients who demonstrate a lumbar direction of preference 

Caralynn Felege, PT, DPT, Amy Urbank, PT, DPT and Ron Schenk, PT, PhD (US) 

31 Surgery versus conservative care for patients with sciatica - which is better? A review and 

consideration of methodological issues 

Stephen May, MA FCSP, Dip. MDT, MSc (UK), Carol Dionne, PT, Cert. MDT (US), 

Richard Rosedale, PT, Dip. MDT (CAN) 

37 The reliability of the McKenzie classification system using extremity 

McKenzie assessment forms 

Elizabeth Kelly E, BSc, Stephen May, MA FCSP, Dip. MDT, MSc, Jenny Ross, MCSP, BSC (UK) 

41 

45 

How long does it take to treat a Derangement? 

A prospective study of 26 cervical and 66 lumbar spine patients 

Martin Melbye, PT, Dip. MDT, Camilla Nymand, PT, Dip. MDT, Troels Balskilde, PT, Dip. MDT (DNK) 

Is manual therapy targeted at specific spinal segments possible or necessary for treatment? 

Stephen May, MA FCSP, Dip. MDT, MSc (UK) 

Literature Reviews 

53 

Stephen May, MA FCSP, Dip. MDT, MSc (UK) 

Case Presentations 

61 Neck pain and headache caused by spontaneous intracranial hypotension 

Megan Kelly, BPhty and Stuart Horton, MPhty, Dip. MDT (NZ) 

65 

A role for Mechanical Diagnosis and Therapy in pre-surgical selection and post-operative 

rehabilitation after lumbar microdiscectomy 

Christie Downing, PT, DPT, Cert. MDT (US) 

Volume 4, No. 1 March 2009 International Journal of Mechanical Diagnosis and Therapy ® - 1


I n d h o l d 

70 The modified Broström repair 

A reflective case report of an intentional non-adherer 

Chris Littlewood, MSc, Dip. MDT (UK) 

74 Restless leg syndrome vs. lumbar derangement 

Highlighting MDT clinical reasoning and outcome documentation 

Mark Werneke, MPT, Dip. MDT (US) 

81 

Differential considerations for a patient with anterior knee pain: 

A case report using Mechanical Diagnosis & Therapy (MDT) 

Karrie Giger, PT, DPT, Cert. MDT (US) 

Educational Updates 

87 

MDT Fellowship Program 

Ron Schenk, PT, PhD, OCS, FAAOMPT, Cert. MDT (US) 

Research Foundation 

90 

International Mechanical Diagnosis and Therapy Research Foundation 

Upcoming Events 

91 

Future MDT conferences 

Author Submission Guidelines 

92 

Guidelines and editorial calendar 

Kursusoversigt 2009 

94 


Fra formanden McKenzie Institut Danmark 

Eva Hauge, Dip.MDT 

Fagligheden var i høj kurs på den netop afholdte Fagfestival, 

hvor over 1500 fysioterapeuter deltog i oplæg 

der dækkede alle facetter af det fysioterapeutiske fagområde. 

En Fagfestival som Danske Fysioterapeuter 

bestemt kan være stolt af og herunder selvfølgelig også 

alle de oplægsholdere som bidrog til vidensdeling. 

Astrid Læssøe fra Mandag Morgen som gennemgik undersøgelsen 

”Mønsterbrydere” – et debatoplæg om fysioterapeuter 

i fremtidens sundhedsvæsen – var ikke i 

tvivl om fysioterapeuters store faginteresse – i interview 

med fysioterapeuter var et gennemgående ord – fag- 

fag- fag. 

Høj faglighed skal også være et fortsat indsatsområde 

på det muskuloskeletale felt, hvor vi i IMDT fortsat skal 

være en seriøs medspiller. Vi skal fortsat udbrede kendskabet 

til MDT, gerne stimulere til forskning og kvalitetsudvikling 

og samarbejde med relevante parter i forhold 

til at gøre sundhedsvæsenet endnu bedre. Der er fortsat 

et stort behov for dygtige fysioterapeuter som seriøse 

medspillere på det muskuloskeltale felt i primær såvel 

som i sekundærsektoren, og jo mere velfunderede vi 

er i en evidensbaseret praksis, jo bedre kan vi agere i 

den tværfaglighed som er essentiel for gode patientforløb 

og rehabilitering. 

Fag - Fag - Fag 

Editorials 

Der ligger mange spændende udfordringer foran os – 

kan vi i primærsektoren få mulighed for at oprette elektroniske 

databaser, hvor vi kan samle data og dokumentation 

om vores kliniske arbejde? Kan vi opnå en 

dansk masteruddannelse indenfor det muskuloskeletale 

felt? Kan vi fordoble vores antal af fysioterapeuter med 

en Diploma i MDT indenfor få år? Hvis du har lyst til at 

være med i en debat om faglig udvikling indenfor MDT 

har du mulighed for at søge optagelse i vores nystiftede 

Faglige udvalg – se annoncering andet sted i dette blad. 

Med denne ekstra fyldige udgave af IJMDT vil jeg gerne 

sige velkommen til vores nye Redaktør Brian Sørensen 

og samtidig takke vores tidligere redaktør Camilla Nymand 

for en stor indsats gennem 5 år. 

Vi har netop afholdt generalforsamling og kan byde velkommen 

til et nyt bestyrelsesmedlem, Dorte Knudsen 

samt to nye suppleanter Jon Grønset og Katrine Fehrend 

– jeg ser frem til et spændende samarbejde med 

de nye og vil samtidig sige tak til de afgåede medlemmer 

Martin Melbye, Simon Simonsen og Michael Rømer. 

Med ønsket om et rigtig godt forår til alle 

De bedste hilsner 

Eva Hauge 

DipMDT 

Formand IMDT 


Fra redaktøren : Fagre ny verden. 

Brian Sørensen, fysioterapeut 

Så er årets 1. udgave af International Journal of MDT 

udsendt, og jeg vil gerne byde dig velkommen ombord. 

Som ny redaktør er det med en vis spænding i maven, 

at udsende dette årets første blad, som også er mit første 

blad, på denne, for mig, noget uvante post i fysioterapiens 

verden. At varetage en post som redaktør er, jo 

umiddelbart noget anderledes end den praktiske verden 

jeg, i mit virke som fysioterapeut, normalt befinder mig i. 

Men så alligevel er der visse paralleller. Fysioterapi er 

jo både teori- og praksisverden. Bladets indhold er jo 

udover case præsentationer, informationer fra bestyrelsen 

etc. også præget af original forskning og masser af 

teoretisk viden. Den teoretiske viden som gerne skulle 

integreres i den daglige praktiske verden. Taler vi om 

evidensbaseret fysioterapi, taler vi bl.a. om integrationen 

af nyeste teori i daglig praksis. En overgang / proces 

som ikke altid er lige nem at praktisere bl.a. pga. 

manglende tid og måske manglende adgang og viden 

om, hvorledes vi opdateres på, hvad der rør sig. 

Personligt har jeg anvendt de tidligere udgivelser af bladet 

som informations-, inspirations- og ikke mindst refleksionskilde 

i den daglige praksis. Processen med 

teori til praksis lettes da også betydeligt, når den leveres 

via egen postkasse, nærmest som en flaskepost 

med modtageradresse der kommer fast x antal gange 

årligt. ”Så bliver det da ikke nemmere at holde sig bare 

en smule opdateret”. 

Et stort tillykke til : 

14 fysioterapeuter har bestået 

Credential evalueringen 28. 

februar 2009 i København 

Jeg håber derfor, at jeg, i det der må betragtes som min 

jomfrurejse i den redaktionelle verden, kan bidrage med 

et spændende og læsevenligt blad. Præcist som det er 

min oplevelse at den tidligere redaktør i allerhøjeste 

grad formåede. 

Bladet vil i denne omgang være fyldt med en del ekstra 

materiale. Til den ivrige læser vil dette være helt fantastisk. 

Omvendt kan det for andre virke noget tungt at 

komme igennem alt den litteratur på engelsk. Vi har 

derfor oversat Original research abstracts til dansk i et 

forsøg på at gøre bladet mere tilgængeligt. 

Så slå benene op og nyd bladet med en kop kaffe mens 

du venter på at foråret for alvor sætter ind. 

Brian Sørensen 

Fysioterapeut, redaktør 

Jacob Birkenfeldt Kierstein, Rødovre Hanne Lise Christensen, Frederikssund 

Sidsel Kristina Flader Skov, Svendborg 

Nicolaj Hove, Holstebro Peter Bo Jørgensen, Hammel 

Sanne Kiltinge, Grevinge Asker Lau Kristensen, Holbæk 

Trine Muckert, Ulfborg Tina Mørck Kappel, Aalborg 

Mia Nørgaard Edrich, København Morten Pallisgaard Støve, Aalborg 

Pernille Rask, Lystrup Jacob Thomsen, Hillerød 

Thomas Thoustrup, København 

Editorials 


Editorial 

Colin Davies, PT, Dip. MDT 

The waves roll in from Africa and crash onto the golden 

sands of Ipanema. Crowds gather along the beach, but 

over in the next bay an even bigger crowd is gathering. 

People from all over the world are assembling in the 

Sheraton Hotel, Rio de Janeiro, for the 11th McKenzie 

Institute International Conference. Sound enticing? 

Well it should, because the biennial conference is the 

very pinnacle of the McKenzie Institute educational 

programme. Here’s why. 

We have attracted some of the most prominent names 

in orthopaedic medicine to our conferences; Alf 

Nachemson, Nik Bogduk, Gordon Waddell, Michael 

Adams, Barrie Vernon-Roberts and Vert Mooney are 

just some of our former Conference Faculty. They have 

shared the stage with many of our own faculty; people 

such as Ron Donelson, Stephen May, Helen Clare and 

Audrey Long have given equally learned presentations. 

Robin McKenzie’s clinical demonstrations have opened 

the eyes of many invited speakers to the potential of 

MDT as a diagnostic and therapeutic system. This 

exposure to MDT has helped stimulate important 

research amongst some of our guests, such as Charles 

Aprill, Michael Adams and Susan Mercer. 

The conference affords a wonderful opportunity for 

clinicians from around the world to meet in a 

stimulating, academic forum with the opportunity of 

making new friends and clinical contacts. Audrey 

Long’s landmark study was an international multicentre 

study where many of the contacts were forged 

through the international conference. Sometimes the 

social events have proved as important as the platform 

presentations. Dinner by torchlight in the Trastevere in 

Rome and the conga line snaking through the 

restaurant high above Queenstown spring to mind. The 

social events in Rio promise to scale new heights of 

sheer pleasure. Oh, and did I mention the caipirinha? 

An amazing drink made in Rio, but designed in heaven. 

The holding of a respected conference helps to 

stimulate interest from clinicians and researchers in the 

area it is being held. Europe has hosted the event four 

times, North America on five occasions and, 

memorably, New Zealand in 2007. 

Now in 2009, it is the turn of Brasil. We are hoping that 

our Rio conference will act as a focal point for the 

further development of MDT in South America. We 

have chosen as our theme “Challenging the Structures 

in Musculoskeletal Medicine”. Some of our speakers, in 

particular, Karim Khan, Heather McKay and Susan 

Mercer, will address the challenge of developing 

therapeutic load strategies for rehabilitation. Stephen 

May, Susan Mercer and Karim Khan will challenge 

some of the popular treatment protocols. For those who 

wonder why the various guideline authorities have not 

given more prominence to MDT research, you will not 

want to miss Kevin Spratt’s challenge to Messer’s 

Cochrane and Co, nor the reply from Maurits van 

Tulder. Gary Jacob, Hans van Helvoirt and Robin 

McKenzie will offer a refreshing perspective of the 

psychosocial model. 

One of the most keenly anticipated presenters will be 

Robin McKenzie himself. Robin has graciously agreed 

to a very prominent role as moderator, panelist, 

speaker and most importantly clinician. Here is the 

chance for those who have never seen Robin treating 

patients, to observe him in action. You will not be 

disappointed. 

Rio 2009 has one further challenge for us all. At the 

10th International Conference in Queenstown, Robin 

observed that despite the quality of the programmes 

offered, McKenzie conferences always seem to attract 

between 350-450 participants. The Scientific and 

Organizing Committees have accepted the challenge to 

fill the auditorium (it holds 700) in Rio de Janeiro, but 

we need your help. Come to Rio, tell your medical 

colleagues to come to Rio and make Rio 2009 the best 

attended McKenzie conference ever. 

See you there! 

. 

Ny fælles pris uddelt for første gang på Fagfestivalen 2009 

Editorials 

Den fælles pris ”Rygraden”, som er stiftet af Danske Fysioterapeuters Fagforum for Muskuloskeletal Fysioterapi 

(DFFMF) og Institut for Mekanisk Diagnostik og Terapi, McKenzie Institut Danmark (IMDT), blev på Fagfestivalen 

2009 givet til Nils-Bo de Vos Andersen for hans store arbejde med den udvidede lænderygundersøgelse. Prisen er 

på kr 30.000,- og vil fremover uddeles hver 3. år. 


Abstracts oversat til dansk. (”Der tages forbehold for trykfejl”). 

Simon Simonsen Dip. MDT 


Sacro-iliac joint pain– how much of it is there and what might be the problem? 

Hans van Helvoirt, MA, PT, Dip. MDT, Dip. MT and Stephen May, MA FCSP, Dip. MDT, MSc 

(”For hele artiklen på engelsk se side 10”). 

Abstract. 

SI-led (SIJ) smerter menes at være årsag til rygsmerter hos et mindretal af patienter med rygsmerter. Patofysiologien 

af SIJ er ukendt, men det er ment, at den kan have både mekanisk og kemisk oprindelse. SIJ smerte kan kun identificeres 

når lumbale og hofteleds problemer er blevet udelukket. Formålet med denne retrospektiv gennemgang af 

kliniske data fra en columna specialist klinik var at kortlægge forekomsten af SIJ, kliniske kendetegn, og om en mekanisk 

eller kemisk patofysiologi syntes at være årsag til symptomerne. Ud af 278 patienter; 12% syntes at have 

smerter stammer fra SIJ. Disse patienter var overvejende kvindelige og årsagen til debut var overvejende traumer. 

De fleste af disse patienter viste sig at have en kemisk kilde af symptomer og responderede på kortison indsprøjtning. 

Et mindretal havde en mekanisk årsag til symptomerne. SIJ synes at være årsag til symptomer i et mindretal af 

patienter, og kan identificeres ved hjælp af en særlig algoritme. For at kunne valideres bør disse fund bekræftes i 

yderligere prospektive undersøgelser. 

Measuring and modifying the frequency and duration of lumbar flexion 

Ronald F. Bybee, PT, DPT, OCS, Dip. MDT, Allyn Byars, PhD, Laura B. Bearden, MPT, 

Tina M. Logan, MPT, Amanda P. Moeller, MPT, Susan J.Hall, PhD (USA) 


Abstract. 

Formål. At undersøge hyppighed og varighed af lumbal flexion i løbet af en fire timers arbejde / skole dagen og teste 

hypotesen, at bæres “PAS-1000” reduceres hyppighed og varighed af flexion. 

Deltagere. En ni ugers undersøgelse blev udført involverer 60 volontører: 30 sundheds studerende og 30 sundhedspersonale, 

alderen 20 til 54 år. Enhver deltager, der havde oplevet lændesmerter inden for de foregående tre måneder, 

planlagde graviditet eller havde sensoriske udfald blev eksluderet. 

Metoder. Flexions hyppighed og varighed blev indsamlet ved hjælp af PAS-1000. Det er en enhed bestående af et 

indstilleligt pres-aktiveret enhed holdes på plads mod spinosus med et bælte. Det er i stand til enten: 1) at yde postural 

vibrationsstimuli [feedback / PASV eller ingen feedback / PASNV]; 2) registrering af hyppighed og varighed af 

lumbal flexion over en 4-timers periode [med en datalogger, når 20 ° (± 5 º ) af flexion blev nået eller overskredet], 

eller 3) begge samtidigt. 

Eksperiment 1 - Deltagerne havde de PAS-1000 for 2 dage. På dag 1, vibrationer blev aktiveret, og på dag 2, vibrationer 

blev deaktiveret med hyppighed og varighed af lumbal flexion registreres på begge dage. 

Experiment 2 - Nitten deltagerne valgte at bære PASV 1 time dagligt i 30 dage. På dag 30, de havde den PASNV 

med datalogger, som registreres hyppigheden og varigheden af flexion. 

Resultater. Procentdel af lumbal flexion på dag 1, 2 og 30, varierede fra 0,11% -64,32%, 0,01% -15,33% og 0,00% - 

50,34%, respektivt. Antal flexioner på dag 1, 2 og 30, varierede fra 4-579, 1-546 og 0-292, hhv. Procentdel af flexion 

og antallet af flexioner blev reduceret betydeligt mellem dag 1 og dag 2, (p = 0,0001, p = 0,0001) og mellem dag 1 

og dag 30 (p = 0,0097, p = 0,0336). 

Diskussion og Konklusion. Data understøtter brugen af PASV at forbedre arbejdsstillinger under brugen, og som 

en effektiv træningsanordning, ved at begrænse lumbal flexion og dermed et muligt fald i risikoen for lændesmerter. 

Samlet set PASV var effektivt til at reducere hyppigheden og varigheden af lumbal flexion i raske personer og yderligere 

forskning bør gennemføres som supplement til disse resultater. 





Prognosis of outcome in patients who demonstrate a lumbar direction of preference 

Caralynn Felege, PT, DPT, Amy Urbank, PT, DPT and Ron Schenk, PT, PhD 


Abstract. 

Formålet med dette eksperimentelle randomiseret test-retest design var at afgøre, om en score baseret på svar på 

gentagne end-range lumbale bevægelser havde relation til opfattelsen af funktion og smerte hos patienter. 

Denne undersøgelse blev foretaget som en del af et forskningsprojekt, der undersøger effekten af Mekanisk Diagnose 

og Terapi (MDT) og Spinal Thrust Manipulation (STM) hos patienter, der mødtes tre ud af de fem kriterier for 

“Clinical prediction rule” for spinal manipulation (CPR). Et “Repeated Movement Scoring system” (RMSS) blev udformet 

af forfatterne til at vurdere scoringer af gentagne lumbal spinal bevægelser som bestemt ved baseline. 

Efter en indledende undersøgelse, blev otte patienter med akut LBP, der opfyldte CPR randomiseret til enten en 

MDT gruppe eller et STM gruppen. Fordi en elektronisk Random Number Generator blev brugt som en del af forskningsprojektet, 

blev syv patienter randomiseret til MDT og en til STM gruppen. 

Oswestry Disability Questionnaire (OSW), Fear-avoidance belief Questionnarie (FABQ), og den numeriske Pain Rating 

Scale (NPRS) blev administreret på første undersøgelse og på et to-ugers opfølgning fra datoen for den første 

undersøgelse. 

På grund af den lille sample size, fokuserede denne undersøgelse på forholdet mellem de gentagne bevægelser 

score (RMSS) og resultater. Sample t-test og Pearsons Coefficient of Correlation blev brugt i analysen. 

Signifikant korrelationer blev observeret mellem NPRS og RMSS gentagne bevægelser score (r = 88, p



The reliability of the McKenzie classification system using extremity 


Elizabeth Kelly E, BSc, Stephen May, MA FCSP, Dip. MDT, MSc, Jenny Ross, MCSP, BSC 



Abstract 

Formål: At undersøge reliabiliteten af erfarne fysioterapeuter i at klassificere patienter i McKenzie's mekaniske syndromer 

fra McKenzie undersøgelsesskemaer. 

Design: Inter-tester reliabilitet pilotundersøgelse ved hjælp af undersøgelsesskemaer. lokation: NHS ambulante 

patienter, UK. 

Metode: Tre McKenzie 'credentialed' terapeuter, som havde taget et kursus i Mekanisk Diagnostik og Terapi for ekstremiteter, 

undersøgte uafhængigt 11 udfyldte McKenzie undersøgelsesskemaer med klassificering konklusion udeladt. 

Disse var blevet udført af en fjerde terapeut, som er underviser i McKenzie Institut, og hvis klassificering repræsenterede 

en "gold standard" i forhold til de andre terapeuters 'konklusioner. 

Resultat: Reliabiliteten blev udtrykt i procent aftale og kappa. 

Resultater: Reliabilitet i syndrom klassificering var »godt«, 82% enighed, kappa 0.7. Imod "gold standard" var terapeuter 

med mere uddannelse mere pålidelige. 

Konklusion: I denne pilotundersøgelse var reliabilitetnen af McKenzie's klassificering i ekstremitets patienter god 

og yderligere undersøgelser i en større kohorte af terapeuter og blandt virkelige patienter, snarere end på undersøgelsesskemaer 

bør udføres . 

Nøgleord: pålidelighed, McKenzie, undersøglesesskema, ekstremiteter 



Martin Melbye, PT, Dip.MDT, Camilla Nymand, PT, Dip. MDT & Troels Balskilde, PT, Dip.MDT (DNK) 


Abstract. 

Videnskabelig litteratur indberetter ikke varigheden af succesfulde behandlingsprogrammer for nakke og lænde patienter, 

selv om dette er et af de vigtigste hensyn til patienterne. Baseret på empirisk viden, vil patienter klassificeret i 

derangement syndromet, som beskrevet af McKenzie, reagere hurtigt på konservativ behandling. Tidligere randomiserede 

kontrollerede forsøg har rapporteret klinisk relevante resultater for lænde patienter efter to og otte ugers behandling. 

Formålet med denne undersøgelse er at følge behandlingsvarighed for cervikale og lumbale patienter klassificeret i 

derangement syndromet. Patienternes selv rapporterede smertescore, funktionelle niveau og selvhjulpenhed ved 

start og slut blev brugt til at vurdere, om varigheden af behandlingen var tilstrækkelige til at fremkalde klinisk relevante 

ændringer i sundhedsstatus. Patienterne blev behandlet med hjemme øvelser, terapeut teknikker, ergonomi, funktionelle 

øvelser og profylaktiske anvisninger. I alt 66 lumbale og 26 cervikale columna patienter blev inkluderet og 

behandlet i gennemsnit henholdvis 2,4 og 4,3 uger. Smertescore, funktionelt niveau og selvhjulpenheds niveau var 

væsentligt bedre, samt klinisk relevant. Forbedringer for lumbale patienter var i overensstemmelse med resultaterne 

fra to randomiserede kontrollerede forsøg. Den eksterne validitet af denne undersøgelse kan være begrænset på 

grund af manglende data fra ekskluderede patienter, samt eksistensen af neurologiske fund og andre indikatorer for 

sværhedsgrad. 

Nøgleord: outcome measure, klinisk database, derangement syndrom, prognose 





Is manual therapy targeted at specific spinal segments possible 

or necessary for treatment? 

Stephen May, MA FCSP, Dip. MDT, MSc 


Abstract 

Segmental specifik behandlingen er blevet anbefalet til at optimere resultaterne i manuel terapi. Dette review har 

ønsket at undersøge nogle åbenlyse beslægtede begreber: 

1) Er en posterior-anterior mobilisering specifikke for et bevæge segment? 

2) Er en specifik mobilisering bedre end standard (uspecifik) mobilisering? 

3) Er der evidens der støtter reliabiliteten i at fastslå specifikke problemer? 

4) Er der evidens der støtter validiteten i at fastslå specifikke problemer? 

Relevante undersøgelser blev fundet ved søgning i MEDLINE og søgning i referencer fra disse studier. Over tredive 

undersøgelser blev gennemgået som omhandlede disse forskellige emner. Posterior-anterior mobilisering berører 

ikke kun den enkelte segmenter niveau som det bliver anvendt på, men rygsøjlen lokalt (som helhed) . Der er intet, 

der tyder på, at specifikke manuelle teknikker er overlegen i forhold til ikke-specifikke eller standardiserede teknikker. 

Det er usandsynligt, at specifikke funktionsnedsættelser, såsom niveau, fikseringer eller stivhed, kan påvise reliabelt 

mellem terapeuter. Dokumentationen vedrørende forbindelsen mellem sådanne tilstande og rygsmerter er selvmodsigende. 

Samlet set tyder det på, at manuel terapi rettet mod specifikke segmenter hverken er muligt eller nødvendigt 

for at optimere resultaterne. 


Sacro-iliac joint pain– how much of it is there and what might be the problem? 

Hans van Helvoirt, MA, PT, Dip. MDT, Dip. MT and Stephen May, MA FCSP, Dip. MDT, MSc 

Introduction 

Sacro-iliac joint (SIJ) pain is a possible source of lumbo 

-pelvic disorder in some patients with non-specific low 

back pain. In a systematic review into the diagnostic 

validity of procedures to identify patho-anatomical 

sources of back pain against reference standards, 

these reference standards diagnosed SIJ pain in 28- 

61% of patient populations investigated (Hancock et al 

2007). It was suggested that only clusters of pain 

provocation tests have potential validity for identifying 

SIJ pain. Furthermore, as SIJ tests are associated with 

a very high rate of false positive responses in patients 

with chronic low back pain (Laslett 1997), it is good 

clinical practice to exclude lumbar spine and hip joint 

problems before considering the SIJ as a source of 

symptoms (McKenzie and May 2003, Laslett et al 2003, 

2005a). Exclusion of lumbar problems is done by the 

identification of centralisation, directional preference or 

a lateral shift, and exclusion of hip problems by pain 

provocation testing. Compared to an injection reference 

standard after excluding lumbar and hip problems and 

then needing at least three positive SIJ pain 

provocation tests, good levels of diagnostic accuracy 

were achieved using this algorithm: sensitivity 91%, 

specificity 87%, positive likelihood ratio 7.0 (Laslett et al 

2003). 

It has been suggested that SIJ pain may be primarily 

mechanical and amenable to mechanical therapy or 

inflammatory and non-responsive to conservative 

treatment (McKenzie and May 2003). A case study of a 

mechanical responder has been published (Horton and 

Franz 2007). The aim of this retrospective review of a 

clinical database is to explore rates of SIJ pain in a 

consecutive patient population as a proportion of all low 

back pain, and to explore their clinical characteristics 

and possible source of pathophysiology. 

Methods 

Consecutive patients, referred by local GPs, 

neurologists and neurosurgeons to two specialist spinal 

clinics, and assessed by a highly skilled physical 

therapist with a Diploma in Mechanical Diagnosis and 

Treatment and with 20 years of experience in clinical 

practice, were included in this report. This includes presurgery 

screening for local neurosurgeons, which may 

bias the patient population to those with more severe 

symptoms. Diagnostic classification was done by the 


Abstract 

Sacro-iliac joint (SIJ) pain is thought to be the source of back pain in a minority of the back pain population. The 

pathophysiology of SIJ is unknown, but it is thought it may have both mechanical and chemical origins. SIJ pain can 

only be identified once lumbar and hip joint problems have been excluded. The aim of this retrospective review of 

clinical data from a specialist spine clinic was to identify the prevalence of SIJ, clinical characteristics and whether a 

mechanical or chemical pathophysiology appeared to be the cause of symptoms. Out of 278 patients; 12% appeared 

to have pain originating from the SIJ. These patients were predominantly female and cause of onset was 

predominantly trauma. The majority of these patients appeared to have a chemical source of symptoms and responded 

to corticosteroid injections. A minority had a mechanical source of symptoms. SIJ appears to be the 

source of symptoms in a minority of patients, and can be identified using a specific algorithm. To be validated, these 

findings need to be confirmed in further prospective studies. 

criteria in Table 1. Patient data is related to 

consecutive patients seen in the first four months of 

2008. 

To be classified as SIJ patients, they had to be 

negative to lumbar and hip repeated movement testing 

and be concordant pain positive to three of five SIJ 

tests (distraction, thigh thrust, compression, sacral 

thrust, Gaenslen’s). Patients who had been clinically 

classified as SIJ were then given a xylocaine 

anaesthetic block under radiographic guidance and 

then were re-tested ten minutes later. If SIJ tests were 

then negative and there was at least 80% pain relief at 

rest, the SIJ classification was confirmed. SIJ patients 

were then classified as chemical, mechanical, 

stabilisation or manual therapy according to the criteria 

in Table 2. Treatments included multiple corticosteroid 

injection and/or mechanical interventions as 

appropriate. Patients were asked to rate their 

percentage improvement from intake when they were 

discharged; these were rated as follows: 

�� ≥ 90% improvement = resolved/better 

�� > 60% improvement = moderate improvement 

�� > 30% improvement = minor improvement 

�� ≤ 30% improvement = no change 

Results 

Data was available for 278 patients regarding 

classification (Figure 1): 48% were classified as 

derangement, a further 20% as irreducible 

derangement, and 34 (12%) were diagnosed to have 

SIJ pain, and most other classifications were present in 

4% or less of the population. 

Details about the 34 patients classified with SIJ 

problems are listed in Table 3. Regarding the 34 

patients classified as SIJ, 35% had three positive SIJ 

tests, 39% had four positive tests and 24% had five 

positive tests. The tests used and their positive rate in 

those deemed to be SIJ positive were as follows: 

distraction test (73.5%), thigh thrust test (91%), 

compression test (41%), sacral thrust test (79%), 

Gaenslen’s (88%). Symptoms were quite variable and 

included lumbar/pelvic, buttock, groin, and leg pain 

referred to the calf and foot. The majority were 

diagnosed to have a chemical pathology (Figure 2). A 

smaller proportion had a mechanical problem, and a 


Sacro-iliac joint pain– how much of it is there and what might be the problems? 

Table 1. Classification labels and criteria 

Classification Criteria Reference 

Derangement-reducible Centralisation, directional preference in response 

to repeated movements (RM) 


McKenzie & May 2003, 

Long et al 2004 

Derangement-irreducible Peripheralisation or worsening distal pain in 

response to RM 

McKenzie & May 2003 

Dysfunction Intermittent end-range pain only in response to RM McKenzie & May 2003 

Hip No response to lumbar RM, pain groin/anterior thigh, 

positive pain provocation hip tests 

McKenzie & May 2003 

SIJ No response to lumbar RM, negative to hip tests, 

pain provocation SIJ tests positive 

Laslett et al 2003 

Post-operative Recent spinal surgery Referred after spine 

surgery in last 6 months 

Spinal stenosis Leg pain on walking and worsening leg pain with 

sustained extension 

Red flags Indicators for cauda equina, cancer or fracture McKenzie & May 2006 

ZJ Negative to all above, unilateral concordant pain 

on palpation and positive on medial branch blocks 

Table 2. SIJ classification algorithm and criteria once lumbar and hip problems excluded 

Classification Criteria 

Manchikanti et al 2004 

1. Mechanical �� 2 or 3 out of 5 SIJ pain provocation tests positive 

�� Intermittent or constant pain 

�� Directional preference with anterior or posterior pelvic rotation (derangement) 

�� Produced/not worse consistent pain response with anterior or posterior pelvic rotation 

(dysfunction) 

2. Chemical �� 3 out of 5 SIJ pain provocation tests positive 

�� Pain constant and irritable 

�� All repeated movements increase and worsen 

3. Stabilisation �� 2 or 3 out of 5 SIJ pain provocation tests positive 

�� Additionally positive to 3 out of 4 of following tests: 

active straight leg raise 

Trendelenburg 

thigh thrust 

long ligament provocation test 

4. Manual therapy �� 2 or 3 out of 5 SIJ pain provocation tests positive 

�� All criteria from above negative 

�� Obvious pelvic torsion-patient stands with shoulders in frontal plane, 

but pelvis is rotated away from frontal plane 



smaller proportion had a mixed picture, in which once 

an inflammatory component had been dealt with, a 

mechanical classification was apparent. 

Treatments were predominantly injections: one 

injection in one patient, two injections in six patients, 

three injections in six patients; four injections in one 

patient; 12 patients had several injections followed by 

MDT; in four patients, injections followed several MDT 

sessions; and only in three patients was treatment 

solely MDT or manual therapy. There was at least a 

moderate improvement or resolution of symptoms in 

over 60% of patients and no change in 23% (Table 4). 

Discussion 

In this consecutive cohort of 278 patients, 12% were 

diagnosed with SIJ problems after lumbar spine and 

hip joint problems were excluded. Certain clinical 

characteristics of the SIJ sub-group stand out, namely 

the preponderance of females (76.5%), and the onset, 

which was commonly related to trauma (56%) and 

pregnancy to a lesser extent (20.5%). Constant pain 

was common (68%) and even more common if those 

with missing data about this issue were excluded 

(77%). 

Several comments can be made about the value of the 

suggested algorithm for identifying SIJ problems. As 

long as lumbar spine and hip joint problems were first 

excluded, clusters of three to five or more positive pain 

provocation tests were frequently found; the distraction 

test (73.5%), thigh thrust test (91%), sacral thrust test 

(79%) and Gaenslen’s test (88%) appeared to be most 

useful. The nature of pathophysiology was most 

frequently thought to be chemical and patients 

frequently responded to anaesthetic and corticosteroid 

injections. These findings would appear to validate this 

process for determining SIJ pain. 

According to Laslett (2008), it is now ‘generally 

accepted’ that the prevalence of SIJ pain in the low 

back pain population is about 13% (95% confidence 

intervals: 9-26%); a figure very close to the 12% 

diagnosed in this data base. Furthermore, Laslett 

suggests that the key SIJ pain provocation tests are 

distraction, thigh thrust, Gaenslen’s, sacral thrust and 

compression. These were the five tests used in this 

study and, all except the latter, were positive in 73% or 

more of the patients with SIJ problems. Three positive 

pain provocation tests is the optimal number of tests, 

with the highest sensitivity (91% or 85%) and specificity 

(78% or 79%), and the highest positive likelihood ratios 

(4.3 or 4.0) (Laslett et al 2005b, van der Wurff et al 

2006). 

The hypotheses regarding the causes of SIJ pain 

remains largely speculative at this point in time and 

requires additional research to confirm. However, 

Laslett (2008) suggests that there is some evidence to 

support the notion of an inflammatory condition and of 

ligamentous laxity causing instability, and that these 

pathologies may be caused or perpetuated by trauma 


or pregnancy. Several factors would support an 

inflammatory cause in this population – the high 

prevalence of constant pain, the common traumatic 

onset, and the frequent positive response to the 

corticosteroid injection. If the high prevalence rate of 

inflammatory pain at the SIJ is confirmed in further 

studies, and the need for articular injections is 

confirmed as the optimal treatment, it does identify this 

group as being one that will be impossible to treat from 

a standard physical therapy approach. Although 

‘mechanical’ problems exist at the SIJ from this data, 

they would appear to be a minority. It was noted that 

the mechanical problems were only positive to three 

tests, whereas the chemical problems were frequently 

positive to more than three, which may be a way of 

helping to determine this group clinically. 

A number of significant weaknesses exist in this data 

set, such as its retrospective nature, some missing 

data, issues of lack of blinding, lack of standardisation 

and lack of use of validated outcome measures. To 

validate these findings requires a prospective data 

collection with clearly established and pre-defined 

operational definitions, interventions and outcomes. In 

a clinical data set, blinding is not feasible, but the use 

of patient-reported outcome measures would help to 

minimise bias. 

Conclusions 

In this retrospective review of a standard clinical data 

set, the prevalence rate of patients with SIJ pain was 

estimated to be 12%, after lumbar spine and hip joint 

problems had been eliminated. Certain clinical features 

stood out about this group: persistent symptoms, 

constant symptoms, traumatic onset and female 

gender. The patients with SIJ pain all had multiple 

positive SIJ pain provocation tests. Some of these 

patients responded positively to corticosteroid 

injections, a smaller number to MDT, and some to a 

mixed intervention. These findings need to be 

replicated in a prospective study with more rigorous 

methodology to be validated. 

References 

Hancock MJ, Maher CG, Latimer J, Spindler MF, 

McAuley JH, Laslett M, Bogduk N (2007). Systematic 

review of tests to identify the disc, SIJ or facet joint as 

the source of low back pain. Eur Spine J doi:10.1007/ 

s00586-007-0391-1 

Horton SJ, Franz A (2007). Mechanical Diagnosis and 

Therapy approach to assessment and treatment of 

derangement of the sacro-iliac joint. Manual Therapy 

12.126-132. 

Laslett M (1997). Pain provocation sacroiliac joint tests: 

reliability and prevalence. In: Vleeming A, Mooney V, 

Dorman T, Snijders C, Stoeckart R (eds). Movement, 

Stability & Low Back Pain. The essential role of the 

pelvis. Churchill Livingstone, New York. 



Laslett M, Young SB, Aprill CN, McDonald B (2003). 

Diagnosing painful sacroiliac joints: a validity study of a 

McKenzie evaluation and sacroiliac provocations tests. 

Aus J Physio 49.89-97. 

Laslett M, McDonald B, Tropp H, Aprill CN, Oberg B 

(2005a). Agreement between diagnosis reached by 

clinical examination and available reference standards: 

a prospective study of 216 patients with lumbopelvic 

pain. BMC Musculoskeletal Disorders 6:28. Available 

at: http://www.biomedcentral.com/1471-2474/6/28 

Laslett M, Aprill CN, McDonald B, Young SB (2005b). 

Diagnosis of sacroiliac joint pain: validity of individual 

provocation tests and composites of tests. Manual 

Therapy 10.207-218. 

Table 3. Demographic details about 34 SIJ patients 

Item N (%) 

Gender: 

Female 26 (76.5%) 

Male 8 (23.5%) 

Age - mean (range) 

Work status: 

43 (21 - 78) 

Working 12 

Part-time working 9 

Off work due to problem 5 

Student 2 

Housewife 3 

Pension 1 

Missing data 2 

Duration of symptoms months - mean (range) 60 (3 - 312) 

Onset: 

Pregnancy 7 (20.5%) 

Trauma 19 (56%) 

Insidious 8 (23.5%) 

Symptom frequency: 

Constant 23 (68%) 

Intermittent 7 (21%) 

Missing data 4 


Laslett M (2008). Evidence-based diagnosis and 

treatment of the painful sacroiliac joint. J Manual Manip 

Ther 16.142-152. 

Manchikanti L, Boswell MV, Singh V, Pampati V, 

Damron KS, Beyer CD (2004). Prevalence of facet joint 

pain in chronic spinal pain of cervical, thoracic, and 

lumbar regions. BMC Musculoskeletal Disorders 6:28. 

Available at: http://www.biomedcentral.com/1471- 

2474/5/15 

Van der Wurff P, Buijs EJ, Groen GJ (2006). A multitest 

regimen of pain provocation tests as an aid to reduce 

unnecessary minimally invasive sacroiliac joint 

problems. Arch Phys Med Rehabil 87.10-14. 



, 

Table 4. Outcomes for 34 patients with SIJ problems 

Outcome* N (%) 

Resolved/better 14 (41%) 

Moderate improvement 7 (21%) 

Minor improvement 1 (3%) 

No change 8 (23%) 

Missing data 4 (12%) 

*see Methods for definition of outcomes 

Figure 1. Classification or diagnoses in 278 consecutive patients 

Figure 2. Sub-classification in 34 patients with SIJ pain 

chem 

der 

dys 

ins 

mt 

chem/der 


Key: 

chem = inflammatory or chemical pain 

der = derangement 

dys = dysfunction 

ins = instability 

mt = responded to manual therapy 

chem/der = mixed inflammatory and 

derangement component 

Volume 4, No. 1 March 2009 International Journal of Mechanical Diagnosis and Therapy ® - 14 

rd 

id 

dys 

hip 

sij 

zj 

po 

ss 

red 

mt 

in 

Key: 

rd = reducible derangement 

id = irreducible derangement 

dys = dysfunction 

sij = sacroiliac joint pain 

zj = zygapophyseal joint pain 

po = postoperative 

ss = spinal stenosis 

red = red flag pathology 

mt = manual therapy sub-group 

in = inconclusive


Ronald F. Bybee, PT, DPT, OCS, Dip. MDT, Allyn Byars, PhD, Laura B. Bearden, MPT, 

Tina M. Logan, MPT, Amanda P. Moeller, MPT, Susan J.Hall, PhD (USA) 

Approximately 80% of all adult individuals experience 

low back pain sometime in life (Kelsey and White 

1990). Despite this high percentage of individuals with 

low back pain, acute trauma accounts for only 4% of 

low back pain complaints (Rowe 1971). The majority of 

low back pain complaints arise from the repetitive 

movements that individuals perform daily, with onset 

occurring in a gradual, transient manner (Weber 1983). 

Low back pain can be caused by a variety of factors, 

one of these being mechanical stress imposed on the 

spine during flexion. The body is designed to withstand 

flexion in the lumbar region, being protected by 

intervertebral discs, posterior ligaments, zygapophyseal 

joints, and back musculature. However, large forces 

and repetitive movements can impose stress on one or 

more of these structures that may result in damage. 

During flexion of the lumbar spine, the vertebrae rotate 

from a posteriorly inclined position to a neutral position. 

Greater motion can be achieved from this flexed 

position by anterior displacement of one vertebrae over 

the next. This position places most of the compressive 

stress on the anterior portion of the intervertebral disc, 

resulting in posterior movement of the nucleus 

pulposus and increased pressure at the posterior and 

lateral edges of the disc (Bogduk 2005). 

Nachemson et al (1964) measured intradiscal 

pressures of the lower lumbar spine in various 

positions. They found that pressure in the discs during 

sitting was an average of 30% higher than in standing. 

Proper sitting posture helps maintain the normal spinal 

curves found in a correct standing posture, however, 

after a few minutes of sitting the lumbar spine assumes 

a flexed position. In this position, the back musculature 

has relaxed and stress is placed on ligamentous 


Abstract 

Purpose. To study the frequency and duration of lumbar flexion during a four hour work/school day and test the 

hypothesis that wearing the PAS-1000 decreases the frequency and duration of flexion. 

Subjects. A nine week study was conducted involving 60 volunteers: 30 allied health students and 30 allied health 

professionals, ages 20 to 54 years. Any participant who had experienced low back pain within the previous three 

months, was currently pregnant, or had sensory deficits was excluded. 

Methods. Flexion frequency and duration data were collected using the PAS-1000. It is a device consisting of an 

adjustable pressure-activated device held in place against the spinous processes by a belt. It is capable of either: 1) 

providing postural vibratory [feedback/PASV or no feedback/PASNV]; 2) recording the frequency and duration of 

lumbar flexion over a 4-hour period [with a Data Logger, when 20° (± 5º) of flexion was reached or exceeded]; or 3) 

both simultaneously. Experiment 1 – Participants wore the PAS-1000 for 2 days. On day 1, the vibration was 

activated and on day 2, the vibration was deactivated with frequency and duration of lumbar flexion recorded on both 

days. Experiment 2 – Nineteen participants chose to wear the PASV 1 hour daily for 30 days. On day 30, they wore 

the PASNV with Data Logger, which recorded the frequency and duration of flexion. Results. Percentage of lumbar 

flexion on days 1, 2, and 30, ranged from 0.11%-64.32%, 0.01%-15.33%, and 0.00%-50.34%, respectively. Number 

of flexion events on days 1, 2, and 30, ranged from 4-579, 1-546, and 0-292, respectively. Percentage of flexion and 

number of flexion events were significantly reduced between day 1 and day 2, (p = 0.0001, p = 0.0001) and between 

day 1 and day 30 (p = 0.0097, p = 0.0336). 

Discussion and Conclusion. The data supports the use of PASV to improve posture during use and as an effective 

training device, by limiting lumbar flexion and thus possible decrease of the risk for low back pain. Overall, the PASV 

was effective in reducing the frequency and duration of lumbar flexion in healthy individuals and further research 

should be conducted to supplement these findings. 

structures (McKenzie and May 2003). Sitting in this 

position for extended periods of time results in 

increased intradiscal pressure and excessive stress on 

the lumbar spine, which may predispose a person to 

low back pain (Nachemson et al 1964). Williams et al 

(1991) found that there was a 21% decrease in back 

pain intensity when patients with low back pain 

maintained a lordotic posture and 14.5% increase in 

back pain intensity for patients sitting with a kyphotic 

posture. 

Spinal flexion stresses the posterior elements of the 

intervertebral disc, as well as the low back muscles. 

Kippers and Parker (1984) determined, with the use of 

EMG, that erector spinae activity increased from the 

beginning of lumbar flexion in standing and remained 

constant to 80°, then became inactive beyond 80°, 

which resulted in increased stress to the surrounding 

supporting structures. The addition of a hand held, 10pound 

weight further increased the duration and 

intensity of the erector spinae activity with lumbar 

flexion (Kippers and Parker 1984). Tan et al (1993) 

determined that at 15° of spinal flexion, the erector 

spinae and latissimus dorsi showed increased 

myoelectric activity. As flexion increased, the erector 

spinae fired linearly, whereas, the latissumus dorsi fired 

non-linearly at submaximal and maximal exertion (tan 

et al 1993). It has been inferred from these studies 

(Kippers and Parker 1984) (Tan et al 1993) that 

activities requiring repetitive flexion force the back 

muscles to work harder, therefore compressing 

intervertebral structures which may contribute to low 

back pain. 

Snook et al (1998) conducted a study to determine if 

decreasing early morning lumbar flexion would reduce 



chronic, non-specific low back pain. Their results 

revealed an 18% to 29% reduction in back pain 

intensity on the treatment group as compared with a 

6% to 9% decrease in the control group. Based on their 

results, chronic, non-specific low back pain was 

reduced by controlling early morning flexion. A small 

percentage of the subjects achieved this reduction 

without any medical intervention beyond the advice to 

control morning flexion. The researchers concluded 

that the subjects’ change in behavior was a major 

contributing factor (Snook et al 1998). 

Some occupations require repetitive spinal flexion, 

which has been associated with a higher incidence of 

low back pain. According to Ekes et al (1995), improper 

body mechanics during lifting causes the majority of 

work-related low back pain. For example, Tanii and 

Masuda (1985) documented that intradiscal pressure 

during back lifting was three times higher than that 

measured during leg lifting (Tanii and Masuda 1985). 

With back lifting, the spine is in a flexed position, 

whereas, leg lifting maintains the lumbar lordosis, thus 

decreasing stress on the back. 

It has been well documented that spinal flexion 

stresses the structures and muscles of the low back 

and increases the risk for low back pain (Williams et al 

1991) (Kippers and Parker 1984) (Tan et al 1993) 

(Snook et al 1998) (Ekes et al 1995) (McClure et al 

1997) (Jayasinghe et al 1978) (Hoogendoorn et al 

2000) (Simunic et al 2001) (Jackson et al 2001) (Olson 

et al 2004) (Rogers and Granata 2006). Pynt et al 

(2001) advocate the lordotic posture in sitting, “In 

summary, then, a sustained lordosed sitting posture 

decreases disc pressure and thereby disc 

degeneration, exhibits less injurious levels of ligament 

tension, and although it increases zygapophyseal 

loading, this is not of itself considered hazardous to 

spinal health. A sustained kyphosed sitting posture, on 

the other hand, increases intradiscal pressure leading 

to increased fluid loss, decreased nutrition, and altered 

cell synthesis and biomechanics of the disc, appearing 

to culminate ultimately in disc degeneration that is a 

cause of low back pain”. 

The purpose of our research was to document the 

frequency and duration of lumbar flexion during a four 

hour work/school day among students and healthcare 

professionals. We also investigated the efficacy of the 

Posture Analysis System-1000 (PAS) in decreasing the 

frequency and duration of lumbar flexion when worn 

and if any improvement would exist after wearing the 

PAS for 30 days. Our hypothesis was that the 

biofeedback of the PAS would decrease frequency and 

duration of lumbar flexion that occurs among students 

and healthcare professionals. 

Methods 

Institutional Review Board (IRB) approval from the 

University of Texas at El Paso was obtained prior to 

data collection. 


Participants 

Sixty participants, including 23 males and 37 females 

between the ages of 20 and 54 years volunteered for 

Experiment 1. The subject sample consisted of 30 

University of Texas and El Paso allied health students 

and 30 allied health professionals employed in El Paso, 

Texas. See Table 1 for subject characteristics. The 

allied health students were enrolled in physical therapy 

and occupational therapy programs. The allied health 

professionals included physical therapists, physical 

therapy assistants, and occupational therapists. 

Nineteen participants volunteered for Experiment 2. 

This group consisted of 10 students and nine 

healthcare professionals between the ages of 21 and 

54 years. Individuals who were pregnant, had low back 

pain (LBP) requiring medical care within the previous 

three months, or had sensory deficits were excluded 

from the study. Prior to participation, each participant 

received and signed an informed consent form and 

filled out questionnaires pertaining to the following 

areas: medical history, activity level, and history of low 

back pain. 

Apparatus 

The PAS-1000 has two modes of operation. The 

Posture Analysis System with Vibration Feedback 

(PASV), is a small, mechanical, pressure activated, 

sensing device, that is housed in a belt so that it rests 

against the spine. It produces a 1.5V output signal and 

a 100Hz vibration when activated. During flexion of the 

spine, the spinous processes place pressure on the 

sensor and activate a vibratory mechanism similar to 

that produced by an electronic pager or cellular 

telephone. Maintaining lumbar lordosis or movement 

into extension does not apply pressure to the sensor 

and, therefore, the vibratory mechanism remains 

inactive. The PAS-1000 with no Vibration (PASNV) is 

the same as PASV except that the vibratory element is 

blocked by a 1k resistor that produces an electrical 

signal but no audible or tactile signal. Therefore, when 

it is activated by the subject’s flexion, the signals for 

flexion events and duration can be stored without 

triggering the vibratory element. 

The Data Logger (Clear Sky Products, P.O. Box 

188236, Carlsbad, California 92009) is a stand alone, 

32k data recording device, which can be configured to 

collect 32k voltage data points at intervals in multiples 

of 0.5 seconds. When used with the PASV or PASNV 

devices, it records either 0V or 1.5V at the programmed 

sampling interval. An IBM compatible PC was used to 

program the Data Logger, store the data, and perform 

calculations on data collected. The Spine Logger 

Software (Clear Sky Products) was used for 

programming and downloading the Data Logger. A 

universal inclinometer was used to measure the degree 

of lumbar flexion that would activate the vibratory 

mechanism. 



The PAS-1000, developed by Clear Sky Products, was 

previously found to have a reliability of 98.1% and an 

overall validity of 95.6% using Cohen’s Kappa analysis 

(Clear Sky Products 1996). Clear Sky Products 

provided five devices with data loggers and 20 devices 

without Data Loggers for use in the study. 

Instruments 

Instruments included an informed consent form, and 

questionnaires on medical history, and activity level at 

work/school. A modified Oswestry Low Back Pain 

Disability Questionnaire, reworded in past tense, was 

also administered. 

Pilot Study 

Three investigators participated in the pilot study to 

learn to operate the equipment properly; to fit the 

equipment to the subjects; and to test the 

completeness, understandability, and relevance of the 

questionnaires. This also helped determine the 

necessary wearing schedule, efficacy of the 

biofeedback, and the overall comfort and ease of 

equipment application. 

Main Study 

The main research design was a nine-week study, 

using two groups with 30 participants each. Each 

participant was screened for sensation and vibratory 

sense in the trunk region from T12-L5. Light touch was 

tested by stroking with cotton balls, and vibratory sense 

was tested using the PASV. All individuals meeting the 

inclusion requirements were invited to participate in the 

study. All subjects were informed that the objective of 

the study was to improve posture by decreasing the 

lumbar flexion in their lifestyle. 

The PAS-1000 was custom fitted to each participant by 

the same researcher and positioned over the L2-L3 

spinous processes. The exact position of the PAS-1000 

was marked on the skin with a permanent marker to 

ensure that proper placement was maintained with 

subsequent fittings. The markings were above and 

below the belt at the 12 o’clock, 3 o’clock, and 9 o’clock 

positions. The vibratory sensor was placed in the 6 

o’clock position (over the spine) and also marked. The 

researcher verified proper placement of the device and 

noted any problems on days one, and two, and then, 

once weekly during the 30 day portion of the study. The 

sensitivity of the PAS-1000 was adjusted with a 

universal inclinometer to activate when 20° (±5°) of 

lumbar flexion was reached or exceeded. Sensitivity 

adjustments were made using the adjustment screw 

and foam pads provided with the PAS-1000. The 

adjustment screw is located on the posterior aspect of 

the PAS-1000; a clockwise rotation decreases the 

sensitivity and counterclockwise rotation increases the 

sensitivity. When necessary, foam pads were used to 

increase the sensitivity of the PAS-1000 by decreasing 

the distance between the PAS-1000 and the L2-L3 

spinous processes. The experiment was conducted on 

two consecutive work/school days. 


Experiment 1 

Day 1: The investigators met with the participants to fit 

the belt, to mark the proper position of the belt, 

to provide instructions, and to fill out the 

medical history, activity level, and history of low 

back pain questionnaires. One researcher 

activated the PASNV with Data Logger to begin 

data collection. After a four hour wearing period 

the data were downloaded. 

Day 2: The PASV with Data Logger was re-fitted, and 

then activated. The Data Logger was set to 

record data over four hours. Day two data were 

downloaded that evening. 

This procedure was repeated with the next group of 

participants until the goal of 60 participants was met. 

Experiment 2 

The participants in Experiment 2 were given 

instructions on the 30-day wear schedule, placement of 

the device, and asked to keep a daily log of the exact 

wearing times of the PAS-1000. They were required to 

demonstrate independence with the device through 

reverse demonstration to ensure adequate cognitive 

ability for proper use of the device. Participants 

involved in the 30-day study were fitted with the PASV 

without the Data Logger and wore the belt for 1-hour 

per day for 30 consecutive days. 

Day 30: The participants wore the PASNV with Data 

Logger for four hours and data were 

downloaded that evening. This procedure was 

repeated until data collection for all 19 

participants had been completed. 

Data Analysis 

The independent variable under study was the 

biofeedback provided by the PAS-1000. Two 

dependent variables were measured: frequency and 

duration of lumbar flexion. The frequency and duration 

data for Experiment 1 were collected on days 1 and 2, 

while Experiment 2 data were collected on day 30. The 

Data Logger is capable of collecting data over 4.5 

hours. The trial period was four hours. The frequency of 

flexion data were converted to frequency in a four hour 

period by taking the middle four hours of the test. 

Duration of flexion was converted to percent of lumbar 

flexion (flexion time (sec.) ÷ total time (sec.) x 100). A 

flexion event was recorded by the data logger when 

20° (±5°) of lumbar flexion was reached or exceeded. 

Then, the statistical analyses of all data were run on 

The Statistical Analysis System (SAS), version 6.12. 

Descriptive statistics, including means and standard 

deviations, were calculated for all quantitative 

variables. Simultaneous multivariate Hotellings T 2 test 

(α = 0.05) revealed at least one mean difference in 

duration and frequency of flexion with a significant 

difference for day 1 versus day 2 (T2 - value = 83.4, Fvalue 

= 40.99, df = (2, 58), p-value = 0.0001) and for 

day 1 versus day 30 (T2 –value = 9.60, F-value = 4.53, 



df = (2, 17), p-value = 0.0264). Univariate paired t-tests 

(α = 0.05) were used for overall group differences in the 

frequency and duration of lumbar flexion. These were 

separated into day 1 versus day 2, and day 1 versus 

day 30. The Shapiro-Wilk test was used to verify the 

assumption of normality. Two-sample t-tests were used 

to determine if the differences between the student 

group and the healthcare professional group were 

significant (α = 0.05). Differences across occupations 

were also evaluated using a paired t-test (α = 0.025 

with Bonferroni correction). A chi-square test (α = 0.05) 

was used to determine any association between each 

questionnaire variable and past history of back injury. 

Results 

Tables 2 and 3 display the mean differences and 

standard deviations for the frequency and duration of 

lumbar flexion. The questionnaire data on the location 

of pain with extended sitting revealed that 65% of 

participants who experienced pain reported its location 

in the low back, while 18% reported the pain in the low 

back and between the shoulder blades (Table 1). Of 

the 65% that reported low back pain with prolonged 

sitting, 20% had reported no previous back injury. 

Therefore, the complaints might merely be related to 

poor posture. The percent lumbar flexion on day 1 

ranged from 0.11% to 64.32% and the number of 

flexion events ranged from 4 to 579 events. Day 2 data 

revealed a range in the percent of lumbar flexion from 

0.01% to 15.33% and a range in flexion events from 1 

to 546 events. On day 30, the percent of lumbar flexion 

ranged from 0.00% to 50.34% and the number of 

flexion events ranged from 0 to 292 events. 

Overall, there was a significant difference between 

mean percent of lumbar flexion for day 1 versus day 2 

(p = 0.0001, Table 2, Figure 1) and for day 1 versus 

day 30 (P = 0.0097, Table 2, Figure 1). The mean 

number of lumbar flexion events for day 1 versus day 2 

(P = 0.0001, Table 3, Figure 2), and day 1 versus day 

30 (p = 0.0336, Table 3, Figure 2) also decreased 

significantly. 

In addition, there was a significant mean difference 

between students and healthcare professionals in the 

percent of lumbar flexion on day 1 versus day 2 (P – 

0.0001, Figure 3) and day 1 versus day 30 (P = 

0.0045, Figure 3). However, differences between 

students and healthcare professionals in number of 

lumbar flexion events for the 1 versus day 2 (P = 0.99, 

Figure 4) and day 1 versus day 30 (P = 0.35, Figure 4) 

were not significant. The average wear time for 

experiment 2 was 66.48 minutes per day with a mean 

of only 2.43 days missed. 

Data pertaining to the percent of lumbar flexion and the 

number of flexion events separated by occupation can 

be found in Tables 2-3. All variables presented in the 

questionnaires, were tested for association with 

previous back injury and no significant associations 

were found based on chi-square tests. Since no 


significant associations were found between prior low 

back pain and our questionnaire data, data obtained 

from the Modified Owestry Low Back Pain 

Questionnaire were not analyzed. 

Discussion 

The purpose of Experiment 1 was to determine the 

frequency and duration of lumbar flexion during a four 

hour work/school day among students and healthcare 

professionals. In Experiment 2, we set out to determine 

the effectiveness of the PAS-1000 in decreasing the 

frequency and duration of lumbar flexion over a 30-day 

period. 

The data were analyzed grouping the students and 

professionals together and then, separating them by 

occupation, several significant differences were 

revealed. When reviewing the range of the number of 

flexion events during day 1 as compared to day 2, 

there appears to be only a slight difference. This is 

attributed to the fact that four participants had 

abnormally high number of events (> 200) on day 2, 

while all other participants (56) stayed within a range of 

1 to 200 events. 

The overall analysis of the percentage of lumbar flexion 

from day 1 as compared to day 2 showed a significant 

mean decrease. After separating the data, significant 

decreases were also found in both students and 

healthcare professionals. The students were found to 

have a greater decrease in percentage of lumbar 

flexion from day 1 as compared to day 2, than the 

healthcare professionals. This may have resulted from 

the students sitting for extended periods of time in 

class, which increased their day 1 flexion values 

compared to the professionals. Therefore, the 

professionals began with a lower percentage for day 1, 

which left a smaller margin for improvement (Figure 3). 

Overall, the number of flexion events also showed a 

significant mean decrease between day 1 as compared 

to day 2 (Figure 2). When the group was analyzed by 

occupation, the number of flexion events on day 1 

compared to day 2 again showed a significant 

decrease (Figure 4). It was expected that the 

frequency and duration of flexion would decrease from 

day 1 to day 2 due to the vibratory feedback from the 

belt, which signaled the participant when he/she flexed 

the lumbar spine. From the data, one can conclude that 

the PASV is effective at decreasing the frequency and 

duration of lumbar flexion. 

The purpose of the 30-day study was to determine the 

training effects of the PASV. The wear time was set at 

≥1 hour (not to exceed 2 hours), to increase 

compliance during the extended phase of the study. It 

is unknown whether this is sufficient time to obtain a full 

training effect. Participants may well need to wear the 

belt more than one hour per day. 

Only 19 participants were able to commit to the 30-day 

study. An analysis of the data from day 1 to day 30 



revealed a significant decrease in the percent of lumbar 

flexion. However, when the data were further analyzed 

to account for occupation (students and healthcare 

professionals), only the percent of lumbar flexion 

among students decreased significantly, while 

healthcare professionals decreased, but not enough to 

be statistically significant. Data were then re-analyzed 

removing one professional whose percent of lumbar 

flexion data on day 30 were extreme. This resulted in a 

significant decrease in the remaining healthcare 

professionals from day 1 to day 30 with a P-value of 

0.0033. 

When the numbers of flexion events in Experiment 2 

were analyzed for the entire group, the difference was 

found to be significant (Figure 2). A decrease in the 

number of flexion events was noted, for both students 

and healthcare professionals, however, the decreases 

were not significant (Figure 4). This lack of significance 

could be due to the small sample size in Experiment 2. 

The questionnaire data were analyzed in hopes of 

finding a significant association with prior history of low 

back injuries. The results revealed no significant 

associations between variables included in the 

questionnaire and prior history of back injury. The lack 

of significance could be attributed to small sample size 

and because relatively few participants reported low 

back pain. 

Recently, a study (Geldhof et al 2006) was conducted 

in Belgium with elementary school children. One of the 

objectives of the multifactorial education program was 

to decrease flexion postures. The investigators were 

successful in significantly decreasing the duration of 

trunk flexion. This was a two-year study involving many 

educational tools and many educators (Geldhof et al 

2006). It would be of interest to compare the success of 

a relatively simple device such as the PAS to these 

more complex, time consuming methods. 

This study had both strengths and limitations. Selected 

limitations of the PAS-1000 that might have influenced 

the outcome of the data have been noted. The 

limitations encountered were more of a hindrance to 

research data gathering than they would be with use of 

the PAS-1000 in a clinical setting. These limitations 

included human error in positioning of the PAS-1000, 

movement of the PAS-1000 away from the original 

placement during wear time, and compliance with the 

30-day wear schedule. Several participants reported 

that the PAS-1000 was comfortable to wear, but that it 

slid upward during the 4-hour wear period with sitting or 

standing. All 60 participants reported experiencing 

movement of the PAS-1000, but only 6% of the 

participants reported that they might have been 

affected by the limitations. 

The limitations of experiment 2 included the limited 

number of PAS-1000’s with Data Loggers and the 

allotted research time frame. A larger number of 


devices would have allowed for more participants and 

might have resulted in a significant difference between 

students and healthcare professionals. 

Several control measures were implemented to 

strengthen the study design. One researcher custom fit 

all participants with the belt on days 1, 2 and 30 and 

checked once weekly during the 30-day study, for 

problems and adjustments to the PAS-1000. 

Participants also kept a daily log recording minutes the 

PAS-1000 was worn. This indicated their compliance 

with the daily wear schedule. The above stated control 

measures may have also resulted in retention of all 

participants. 

Clinical Significance 

The results of this study showed promise in reducing 

flexion in our lifestyle. This needs to be correlated with 

actual decrease in risk of low back pain. We 

recommend that future research be conducted using 

the PAS-1000 as a tool for providing biofeedback to 

monitor and decrease early morning flexion, which has 

been shown to be related to decreasing chronic, nonspecific 

low back pain by Snook et al (1998). An 

analysis of the PAS-1000 with a wear time greater than 

1 hour per day for 30 days should be conducted to 

determine the full training effect of the device. Further 

research should also incorporate a longer follow-up 

study to measure the retention of the training effects, 

as well as, the duration of re-training necessary to 

maintain the training effect. Research should also be 

conducted on participants currently suffering from low 

back pain to determine significance in a clinical setting. 

The data supports the use of PASV to improve posture 

during use and as an effective training device, by 

limiting lumbar flexion and thus possible decrease of 

the risk for low back pain. Overall, the PASV was 

effective in reducing the frequency and duration of 

lumbar flexion in healthy individuals and further 

research should be conducted to supplement these 

findings. 

Acknowledgements 

Partially funded by the Texas Physical Therapy 

Education and Research Foundation. 

References 

Bogduk N. Clinical Anatomy of the Lumbar Spine and 

Sacrum. 4th ed. Sydney, Australia: Churchill 

Livingstone; 2005:11-26. 

Clear Sky products. Clinical research involving the 

Spine Tuner Posture Control Device. Carlsbad, CA 

June 1996. 

Ekes AM, Keister JD, Loseth AE, McKenzie CL. 

Reliability of Lift Alert as a feedback devise for 

detecting changes in body position. J Occupational 

Rehab. 1995;5(1):17-25. 



Geldhof E, Cardon G, De Bourdeaudhuij I, De Clercq 

D. Effects of a two-year mulifactorial back education 

program in elementary schoolchildren. Spine. 2006;31 

(17):1965-1973. 

Hoogendoorn WE, Bongers PM, de Vet HCW, Douwes 

M, Koes BW, Miedema MC, Ariens, GAM, Boulter LM. 

Flexion and rotation of the trunk and lifting at work are 

risk factors for low back pain. Spine. 2000;25(23):3087- 

3092. 

Jackson M, Solomonow M, Zhou B, Baratta RV, Harris 

M. Mulitifidus EMG and tension-relaxation recovery 

after prolonged static lumbar flexion. Spine. 2001;26 

(7):715-723. 

Jayasinghe WJ, Harding RH, Anderson JAD, 

Sweetman BJ. An electromyographic investigation of 

postural fatigue in low back pain – a preliminary study. 

Electromyography Clin Neurophysiology. 1978;18:191- 

198. 

Kelsey JL, White AA. Epidemiology and impact of low 

back pain. Spine. 1980;5:133-142 

Kippers V, Parker AW. Posture related to myoelectric 

silence of erectors spinae during trunk flexion. Spine. 

1984;9(7):740-745. 

McClure PW, Esola M, Rachel S, Siegler S. Kinematic 

analysis of lumbar and hip motion while rising from a 

forward, flexed position in patients with and without a 

history of low back pain. Spine. 1997;22(5);552-558. 

McKenzie RA, May S. The Lumbar Spine: Mechanical 

Diagnosis and Therapy. Vol 1 Waikanae, New Zealand: 

Spinal Publications Ltd; 2003:143-146. 

Nachemson A, Goteborg, Sweden, Morris, JM. In vivo 

measurement of intradiscal pressure: Discometry, a 

method for the determination of pressure in the lower 

lumbar discs. J Bone Joint Surg. 1964;46A:1077-1080. 

Olson MW, Li L, Solomonow M. Flexion-relaxation 

response to cyclic lumbar flexion. Clin Biomech. 

2004;19:769-776. 

Pynt J, Higgs J, Mackey M. Seeking the optimal 

posture of the seated lumbar spine. Physio Theory 

Prac. 2001;17:14. 

Rogers EL, Granata KP. Disturbed paraspinal reflex 

following prolonged flexion-relaxation and recovery. 

Spine. 2006;31(7):839-845. 

Rowe ML. Low back pain in industry: an updated 

position. J Occup Med. 1971;13:476-478. 

Simunic DI, Broom ND, Robertson PA. Biomechanical 

factors influencing nuclear disruption of the 

intervertebral disc. Spine. 2001;26(11):1223-1230. 

Snook SH, Webster BS, McGorry RW, Fogleman MT, 

McCann KB. The reduction of chronic nonspecific low 

back pain through the control of early morning lumbar 

flexion. Spine. 1998;23(23):2601-2607. 


Tan JC, Parnianpour M, Nordin M, Hofer H, Willems B. 

Isometric maximal submaximal trunk extension at 

different flexed positions in standing: Triaxial torque 

output and EMG. Spine. 1993;18(16):2480-2490. 

Tanii K, Masuda T. A kinesiologic study of the erector 

spinae activity during trunk flexion and extension. 

Ergonomics. 1985;28:883-893. 

Weber H. Lumbar disc herniation: A controlled, 

prospective study ten years of observation. Spine. 

1983;8:131-140. 

Williams MM, Hawley JA, McKenzie RA, van Wijmen 

PM. A comparison of the effects of two sitting postures 

on back and referred pain. Spine. 1991:16(10):1185- 

1191. 

Table 1. Subject characteristics (n=60) 

Gender 

23 males 

37 females 

Age 

Students 

Mean = 26.25 

SD = 6.43 

Professionals 

Mean = 33.82 

SD = 7.15 

Occupation 

30 Students 

30 Professionals 

Location of pain reported with extended sitting 

Students 

3 Low back 

1 Between shoulder blades and low back 

Professionals 

7 Low back 

2 Between shoulder blades and low back 



Table 2. Mean differences and standard deviations for percentage of time in lumbar flexion 

n Mean SD P value 

Day 1 to Day 2 

Students 30 29.78 21.52 0.0001 

Professionals 30 11.27 10.06 0.0001 

Total 60 20.52 19.09 0.0001 


Students 10 31.21 23.10 0.0021 


Total 19 16.74 25.23 0.0097 

Table 3. Mean differences and standard deviations for number lumbar flexion events 

n Mean SD P value 


Students 30 110.97 156.02 0.0005 


Total 60 110.71 145.84 0.0001 


Students 10 125.45 202.70 0.0820 


Total 19 89.89 170.33 0.0336 

Figure 1. Overall Mean Percentage of Time in Lumbar Flexion. 

45 

40 

35 

30 

25 

20 

15 

10 

5 

0 

Day 1 Day 2 Day 30 




Figure 2. Overall Mean Number of Lumbar Flexion Events. 

250 

200 

15 0 

10 0 

50 

0 


Figure 3. Mean Percentage of Time in Lumbar Flexion by Occupation. 

35 

30 

25 

20 

15 

10 

5 

0 


Figure 4. Mean Number of Lumbar Flexion Events by Occupation. 

250 

200 

15 0 

10 0 

50 

0 


Students 

P rofessionals 

Students 

P rofessionals 





Caralynn Felege, PT, DPT, Amy Urbank, PT, DPT and Ron Schenk, PT, PhD 

Abstract 

The purpose of this experimental randomized test-retest design was to determine if a score based on responses to 

repeated end-range lumbar movements related to perceived level of function and pain levels in patients. This study 

was conducted as part of a research project investigating the efficacy of Mechanical Diagnosis and Therapy (MDT) 

and Spinal Thrust Manipulation (STM) in patients who met three out of the five criteria for the clinical prediction rule 

for spinal manipulation (CPR). A Repeated Movement Scoring System (RMSS) was designed by the authors to rate 

the scores of repeated lumbar spinal movements as determined at baseline. 

Following an initial examination, eight subjects with acute LBP who met the CPR were randomly assigned to either 

a MDT group or a STM group. Because a computerized random number generator was used as part of the 

research project, seven subjects comprised the MDT group and one subject was in the STM group. The Oswestry 

Disability Questionnaire (OSW), Fear-Avoidance Beliefs Questionnaire (FABQ), and the Numerical Pain Rating 

Scale (NPRS) were administered at the initial examination and at a two-week follow-up from the date of the initial 

examination. 

Due to the small sample size, this investigation focused on the relationship of the repeated movement score 

(RMSS) and outcomes. Sample t--Tests and the Pearson Product-Moment Coefficient of Correlation were used in 

analysis. 

Significant correlations were observed between NPRS and RMSS repeated movement score (r = .88, p



The study examined this subgroup of people with LBP 

as part of a research project investigating the effects of 

Mechanical Diagnosis and Therapy (MDT) in patients 

who meet the CPR for spinal manipulation. 

Methods 

Design 

A randomized trial was conducted to assess whether 

the spinal thrust technique or MDT was more effective 

in the management of individuals who met the CPR for 

spinal manipulation. A secondary purpose, and the 

focus of this particular study, was to investigate 

whether a relationship existed between the results of 

the repeated movement examination at the intake initial 

examination and the patient’s outcome at two weeks 

following the initial examination, as measured by the 

Oswestry Low Back Pain Disability Questionnaire 

(OSW) (Fairbank and Pynsent 2000) and the Numerical 

Pain Rating Scale (NPRS) (Childs et al 2005). The 

FABQw was used to screen patients to determine the 

individual’s fear of engaging in their previous work 

activities. This study was reviewed and approved by 

the Daemen College Human Subjects Research 

Review Committee and the Catholic Heath System of 

Buffalo, NY Human Subjects Committee. 

Subjects and Setting 

The sample population for this study was comprised of 

individuals with LBP being referred to outpatient 

physical therapy at a Catholic Health System clinic in 

western New York State. All potential subjects for this 

study were referred to physical therapy by a physician. 

Inclusion criteria for participation in the study were 

based on the Clinical Prediction Rule (CPR) for spinal 

manipulation. Due to the limited nature of the CPR, 

inclusion required meeting only three of the five CPR 

criteria. Exclusion criteria from the study included 

pregnancy, age of less than 18 years, Worker’s 

Compensation recipients, pending litigation associated 

with the current injury, history of lumbar surgery, history 

of psychological illness, inability to understand English, 

or history of a systemic inflammatory disease. 

Examiners 

Data were collected by eight physical therapists who 

were in-serviced in spinal manipulation through two two 

hour sessions, and who were certified in Mechanical 

Diagnosis and Treatment (MDT). All participating 

physical therapists had a minimum of seven years of 

clinical experience. 

Instruments 

The MDT Repeated Movement Scoring System 

(RMSS) (Table 1) was developed by the researchers 

as a means of determining if the repeated movement 

examination, in combination with the neurological 

examination, could be used to establish a prognosis for 

the patient’s outcome. The RMSS was based on the 

work of Robin McKenzie (McKenzie 1981), who 

developed the concept of testing repeated end range 

spinal movements to arrive at a classification for people 

with spinal pain. The testing of repeated end range 

spinal movements may result in symptoms being 

reproduced, increased, or decreased during; and 

centralized, peripheralized, worsened, or made better 

after. Based on the findings of the repeated end-range 

spinal movements and the patient’s response to these 

movements, a direction of preference for therapeutic 

exercise may be determined (McKenzie 1981). The 

direction of preference is the direction of spinal 

movement which causes symptoms to be abolished, 

decreased, or centralized. Centralization refers to a 

rapid change in the perceived location of pain from a 

distal or peripheral location to a more proximal or 

central one. Centralization has been found to be a valid 

predictor of outcome in several studies (Donelson et al 

1990; Long et al 1995; Sufka et al 1998; Werneke et al 

1999) and as such, may be effective in establishing a 

prognosis for the patient. The opposite of centralization 

is referred to as peripheralization. This occurs when 

symptoms move from a more central location to a more 

peripheral or distal location. 

The Oswestry Disability Questionnaire (OSW) 

(Fairbank and Pynsent 2000) is a 10-category 

questionnaire intended to provide the physical therapist 

with information relating the patient’s pain and 

perceived level of disability. Categories of the 

questionnaire include pain intensity, personal care, 

lifting, walking, sitting, bending, sleeping, social life, 

traveling, and employment. The test is graded on a 0- 

100 point scale, with a higher score indicating greater 

disability related to LBP. 

The Fear-Avoidance Beliefs Questionnaire (FABQ) 

(Fritz and George 2002) is intended to determine the 

patient’s level of fear associated with activities which 

may trigger continued or additional LBP. Utilization of 

the tool may in part explain the transition between the 

acute and chronic phases of LBP as well as the 

sustainability of chronic low back syndromes. Sixteen 

questions comprise the FABQ, which is divided into two 

subscales, one for work (questions 6, 7, 9, 10, 11, 12, 15) 

and one for physical activity (questions 2-5). Scoring is 

based out of 96 possible points; each item can earn a 

maximum of six points. Higher FABQw scores indicate 

decreased likelihood of returning to work (George et al 

2003). 

The Numerical Pain Rating Scale (NPRS) (Childs et al 

2005) is a self-reported measurement tool used to 

indicate the average pain experienced with LBP. The 

scale is either verbally or graphically administered to 

the patient, either on a 0-10 or 0-100 scale (for this 

study, the verbal 0-10 scale was used). The scale is 

anchored at two opposing extremes of “0” indicating 

“no pain” and “10” indicating “most extreme pain.” 

The passive Straight Leg Raise (SLR) is a diagnostic 

test used to determine the presence of adverse neural 

tension. The patient is positioned in supine as the 

examiner passively dorsiflexes the foot. The examiner 




then raises the lower extremity until the patient notes 

burning, tension, or a shooting sensation down the 

posterior aspect of the limb. The examiner then 

plantarflexes the foot without changing the hip joint 

angle, predictably resulting in a sudden alleviation in 

distal radicular symptoms. The hip joint angle at 

symptom onset is noted, and the process is performed 

bilaterally. Angles documented at



expiration. The thrust maneuver was then performed in 

an anterior-to-posterior direction through the contact on 

the patient’s ASIS. 

Mechanical Diagnosis and Therapy (MDT): seven 

subjects assigned to this group received treatment 

based on their history and response to repeated endrange 

lumbar movements. Depending on the syndrome 

classification, a treatment program was prescribed 

which included postural correction and performance of 

repeated end-range lumbar movements intended to 

reduce and/or abolish the patient’s signs and 

symptoms. 

Data Analysis 

Data were analyzed using the Pearson Product- 

Moment Coefficient of Correlation. This statistical 

measure is useful in assessing internal consistency 

within a data set, when more than two responses are 

possible, as in ranking on the Clinical Prediction Rule, 

or patient response to repeated end-range lumbar 

movements (Werneke et al 2005). This test was used 

to quantify the strength and direction of the relationship 

between the treatment group receiving the MDT 

intervention and patient outcome as scored on the 

OSW (Werneke et al 2005). Sample t-Tests were 

performed on the changes to the Oswestry Disability 

Questionnaire (OSW), the Numerical Pain Rating Scale 

(NPRS), and the Repeated Movement Scoring System 

(RMSS) at baseline and follow-up. A Pearson Product- 

Moment Coefficient of Correlation was performed on 

NPRS and RMSS scores, as well as OSW and RMSS 

scores, to identify correlation within the two sets of 

data. 

Results 

A total of eight subjects participated in this study. The 

sample consisted of five females and three males 

between the ages of 32 and 49 years, with a mean age 

of 41 years. A complete summary of subject 

demographic data can be referenced in Table 2. 

Analysis of the OSW revealed statistically significant 

improvements in patient function from initial evaluation 

to the follow-up two weeks later (t = 2.565, p



Despite direct access, patients still tend to be referred 

to physical therapy following the acute stage of injury. 

The CPR is highly specific in that four out of the five 

criteria must be met on the subject’s initial evaluation to 

achieve an outcome of an estimated 95% improvement 

in function (Flynn et al 2002). An additional limitation is 

that the reliability of the RMSS proposed by the authors 

has not been established. 

Future Research 

Future research using the RMSS should be conducted 

with a larger subject population and with other potential 

patient classifications. The tool may then be assessed 

in regards to its reliability and clinical utility. 

Conclusions 

The purpose of this study was to determine if the MDT 

repeated movement score correlates with patient 

outcome following treatment with repeated end-range 

lumbar spinal movements. This study demonstrates a 

correlation between the Oswestry Disability 

Questionnaire and the patients’ RMSS score when 

three out of five CPR criteria are met prior to treatment. 


The authors wish to thank the International Mechanical 

Diagnosis and Therapy Research Foundation for 

funding the main research project. 

References 

Childs JD, Fritz JM, Flynn TW, Irrgang JJ, Johnson KK, 

Majkowski GR, Delitto A (2004) A clinical prediction 

rule to identify patients with LBP most likely to benefit 

from spinal manipulation: A validation study. Annals 

Internal Medicine 141(12):920-928. 

Childs JD, Piva SR, Fritz JM (2005) Responsiveness of 

the numeric pain rating scale in patients with low back 

pain. Spine 30(11):1331-1343. 

Cook C, Hegedus EJ, Ramey K (2005) Physical 

therapy exercise intervention based on classification 

using the patient response method: A systematic 

review of the literature. Journal of Manipulative 

Physiological Therapeutics 13(3):152-162. 

Delitto A, Erhard RE, Bowling RW (1995) A treatmentbased 

classification approach to low back syndrome: 

Identifying and staging patients for conservative 

treatment Physical Therapy 75(6):470-489. 

Donelson R, Silva G, Murphy K (1990) Centralization 

phenomenon: its usefulness in evaluating and treating 

referred pain. Spine 15:211-213. 

Erhard RE, Delitto A, Cibulka MT (1994) Relative 

effectiveness of an extension program and a combined 

program of manipulation and flexion and extension 

exercises in patients with acute low back syndrome. 

Physical Therapy 74(12):1093-1100. 

Fairbank JT, Pynsent PB (2000) The Oswestry 

disability index. Spine 25 (22):2940-2953. 

Flynn T, Fritz J, Whitman J, Wainner R, Magel J, 

Rendeiro D, Butler B, Garber M, Allison S (2002) A 

Clinical Prediction Rule for classifying patients with LBP 

who demonstrate short-term improvement with spinal 

manipulation. Spine 27:2835-2843. 

Fritz JM, George S (2002) Identifying psychosocial 

variables in patients with acute work-related LBP: the 

importance of fear-avoidance beliefs. Physical Therapy 

82:973-983. 

George SZ, Delitto A (2005) Clinical examination 

variables discriminate among treatmentbased 

classification groups: A study of construct validity in 

patients with acute LBP. Physical Therapy 85:306-314. 

Hicks GE, Fritz JM, Delitto A, McGill SM (2005) 

Preliminary development of a clinical prediction rule for 

determining which patients with LBP will respond to a 

stabilization exercise program. Archives of Physical 

Medicine Rehabilitation 86:1753-1762. 

Lenth, R. V. (2006). Java Applets for Power and 

Sample Size [Computer software]. Retrieved 

11/12/2008 from http://www.stat.uiowa.edu/~rlenth/ 

Power. 

Long AL. (1995) The centralization phenomenon: Its 

usefulness as a predictor of outcome in conservative 

treatment of chronic LBP (A Pilot Study). Spine 

20:2513-2521 

McKenzie RA (1989) The lumbar spine: Mechanical 

Diagnosis and Therapy Spinal Publications, Walikanae, 

New Zealand. 

Miller ER, Schenk RJ, Karnes JL, Rousselle JG (2005) 

A comparison of the McKenzie approach to a specific 

spine stabilization program for chronic LBP. Journal of 

Manipulative Physiological Therapeutics 13(2):103- 

112. 

Sufka A, Hauger B, Trenary M et. al (1998) 

Centralization of low back pain and perceived 

functional outcome. Journal of Orthopaedic and Sports 

Physical Therapy. 27:205-212. 

Werneke M, Hart D, Cook D (1999) A descriptive study 

of the centralization phenomenon. Spine 24:676-683. 

Werneke MW, Hart DL (2005) Centralization: 

association between repeated end-range Pain 

responses and behavioral signs in patients with acute 

non-specific LBP. Journal of Rehabilitative Medicine 

37:286-290. 




Table 2. Subject demographic data 

Subject Gender Age 

Pre 

OSW 

Post 

Table 1. Repeated Movement Scoring Scale (RMSS) 

Pre 

FABQw 

Post 

Scoring of Repeated Lumbar Movements 

5 No Direction of Preference 

(At least 2 neurological signs) 


(At least 1 neurological sign) 


(No neurological signs) 

2 Direction of Preference 

(s/s decreased, but no better) 

1 Direction of Preference 

(s/s centralized) 

0 

Direction of Preference 

(s/s abolished) 

NPRS 

(out of 10) 

Pre 

Post 

Pre 

RMSS 

Score 

Post 


Hip IR 

SLR 

Pain Below 

Knee 

PIVM 

R L R L Pre Post Pre Post 

Pre Post Pre Post Pre Post Pre Post 

1 F 49 74 36 10 5 6 4 2 1 0-30 0-30 0-53 0-53 0-90 0-90 0-95 0-80 No No normal normal 

2 F 32 46 26 12 10 6 2 1 0 0-30 0-30 0-33 0-30 0-75 0-80 0-60 0-70 No No normal normal 

3 M 48 26 6



Table 3. Subject pre– vs. post-test data 

Analyzed Outcome 

Measure 

Pre-test Post-test P value difference 

(2-tailed) 

X SD X SD 

OSW Scores 42.0 18.0 29.3 13.2 0.03 

NPRS Scores 6.75 1.28 3.38 3.07 0.04 

MDT Scores 1.50 0.54 0.63 0.74 0.06 

Figure 1: OSW analysis using sample t-Test. 

The x-axis represents each subject, while the y-axis represents OSW score. Series One (in dark) 

indicates score upon initial examination, while Series Two (in light) indicates score at the two week 

follow-up. Statistical analysis reveals that 75% of subjects improved function by an average of 63% 

when treated with MDT. 



Figure 2. NPRS analysis using sample t-Test 


The x-axis represents each subject, while the y-axis represents NPRS score. Series One (in dark) 

indicates score upon initial examination, while Series Two (in light) indicates score at the two week 

follow-up. Statistical analysis reveals that 88% of subjects experienced pain levels decreased by an 

average of 55% when treated with MDT. 

Figure 3. Pearson Product-Moment Coefficient of Correlation (NPRS/RMSS score correlation) 

The x-axis begins at initial evaluation (pre-test) and ends at the two week follow-up (post-test). The yaxis 

represents NPRS score and RMSS score. Statistical analysis reveals an 88% correlation. Correlation 

is significant at the 0.01 alpha level. 


Surgery versus conservative care for patients with sciatica - which is better? 

A review and consideration of methodological issues 

Stephen May, MA FCSP, Dip. MDT, MSc, Carol Dionne, PT, PhD, OCS, Cert. MDT, 

Richard Rosedale, PT, Dip. MDT 

Introduction 

Disc herniation causing sciatica or lumbar 

radiculopathy exists in about 5-10% of the back pain 

population (McKenzie and May 2003, page 90). It is a 

clearly defined and recognisable clinical entity, 

characterised by a dermatomal pain pattern, 

neurological signs and symptoms, restricted straight leg 

raise, and confirmed by imaging evidence at the 

appropriate side and segmental level (Kramer 1990). 

The natural history is considered to be positive, if rather 

protracted, with recovery from severe leg pain and 

functional disability over the first three months from 

onset (McKenzie and May 2003, pages 97-101). Thus, 

it is recommended that conservative treatment is used 

first, but if this fails then surgery is recommended. 

However, it is still unclear if surgery confers better 

outcomes than non-operative treatment, and 

determining the optimal conservative treatment also 

remains unclear. 

Several reviews have been conducted regarding 

surgery for disc herniations. Hoffman et al (1993) 

included 81 studies, most of which had substantial 

methodological flaws, but 65% to 85% of patients 

reported no sciatica one year after surgery compared 

with 36% of conservatively treated patients. Long-term 

effects were similar. Memmo et al (2000) found 

evidence to support immediate pain relief with surgical 

intervention, but that recent studies had shown 

comparable short-term results with non-operative 

treatments, and that long-term results showed no 

statistical differences between surgical and 

conservative treatments. They suggested that 

‘aggressive rehabilitation with appropriate pain control 

may actually be the superior treatment option when 

one considers the cost, complications and morbidity 

associated with surgery.’ (Memmo et al 2000, page 79). 

Similarly conclusions were made in an updated 

Cochrane review (Gibson and Waddell 2007). ‘Surgical 

discectomy for carefully selected patients with sciatica 

due to lumbar disc prolapse provides faster relief from 

the acute attack than conservative management, 

although any positive or negative effects on the natural 

history of the underlying disc disorder are still 

unclear.’ (Gibson and Waddell 2007, page 1735). 

The aim of this review is to consider three questions: 

1. Is it better to recommend a surgical or conservative 

treatment option for patients with sciatica? 


Abstract 

It is unclear if patients with disc herniations should be treated with surgery or conservatively. This review attempts to 

summarise the limited literature in this area, and make some conclusions. There have been a number of 

randomised controlled trials and observational studies that have attempted to address this issue. On the whole, but 

not consistently, these suggested that surgery resulted in quicker alleviation of pain and disability, but no difference 

long-term. However, a number of methodological limitations should be considered when interpreting these studies. 

Finally, we suggest a number of different questions and study designs that might be used to address some the 

issues relevant to this topic. 

2. Are there any prognostic indicators for these 

different routes? 

3. What is the optimal conservative intervention for 

these patients? 

The review will start with the description of a recent four 

year follow-up of a randomised trial and observational 

study, as this study demonstrates a number of 

problems about the literature in this area. Previous 

work that compared surgical and non-operative 

treatment for sciatica will then be described. Finally, 

some conclusions will be provided to address the 

questions posed above. 

Weinstein JN, Lurie JD, Tosteson TD, Tosteson 

ANA, Blood EA, Abdu WA, Herkowitz H, Hilibrand 

A, Albert T, Fischgrund J (2008). Surgical versus 

nonoperative treatment for lumbar disc herniation. 

Four-year results for the spine patient outcomes 

research trial (SPORT). Spine 33.2789-2800. 

Objectives 

To assess the four year outcomes of surgery and nonoperative 

care for patients with disc herniations. 

Design 

Prospective randomised trial and observational study 

depending on patient’s preferences. 

Setting 

Thirteen multidisciplinary spine centres in 11 states in 

the United States. 

Patients 

Of 2,720 patients who were screened, 1,991 were 

eligible and 747 declined consent. Of the 1,244 (46%) 

enrolled, 501 were involved in the randomised trial and 

743 in the observational study, in which patients chose 

their treatment. To be included, patients had to have 

signs and symptoms of lumbar radiculopathy persisting 

for at least six weeks with a disc herniation at the 

appropriate side and level on imaging, and be 

considered surgical candidates. 

Interventions 

The surgery was standard, open discectomy with 

examination of the nerve root. The non-operative 

intervention was not standardised, but deemed to be 

‘usual care’ and included physical therapy, education 

and home exercises, anti-inflammatory drugs, epidural 

injections, chiropractic, and opioid analgesics. By four 

years, 59% of those randomised to surgery had 

received surgery, compared to 45% of those 




randomised to non-operative care, 95% of those who 

chose surgery and 24% of those who chose nonoperative 

care. 

Outcome measure 

Primary outcomes were the bodily pain scale and the 

physical functional scale of the SF36 and the Oswestry 

disability index at six weeks, three and six months, and 

one, two and four years. Secondary outcome included 

self-reported improvement, work status, satisfaction 

and sciatica bothersomeness. Over the four years, 96% 

of the original cohorts provided at least one follow-up 

visit and 65% to 87% provided data at each time point. 

Main results 

Due to the high levels of crossovers from initial 

intervention assignment or choice data analysis was 

conducted both on an intention-to-treat analysis and on 

an as-treated basis. On an intention-to-treat analysis 

primary outcomes favoured surgery, but there were no 

significant differences at any time point. Sciatica 

bothersomeness index was significantly more improved 

in the surgery group at one and four years. In the astreated 

analysis there were no significant differences 

between the randomised and observational groups, 

and the cohorts were combined for analysis. There 

were significant differences in all primary and 

secondary outcomes, except work status, favouring the 

surgery intervention at all time points. In most 

outcomes, the surgery intervention group displayed 

significantly better outcomes at six weeks, which had 

improved somewhat further at three to six months, at 

which point similar levels of improvement were 

maintained at all time points through to four years. 

Conclusions 

Non-adherence to treatment protocols meant that 

analysis by intention-to-treat underestimated the 

treatment effect. Analysis as-treated demonstrated 

significantly better primary and secondary outcomes, 

except work status, for surgically treated patients over 

non-operatively treated patients at all follow-up points, 

with maintenance at four years. 

Comments 

This is the four-year follow-up of a previously published 

randomised controlled trial (Weinstein, Tosteson, Lurie 

et al 2006a) and observational study (Weinstein, Lurie, 

Tosteson et al 2006b) with two years follow-up. These 

earlier reports showed major improvements in both 

surgically and non-operatively treated arms of the 

studies, with greater improvements reported in the 

surgically treated groups, whether randomised or 

elected. However, the differences were not significant 

in an intention-to-treat analysis, except sciatica 

bothersomeness and self-rated improvement. 

Furthermore, adherence to treatment assignment was 

limited; at three months, 50% of patients randomised to 

surgery had actually received it, but so had 30% of 

those randomised to non-operative care (Weinstein, 

Tosteson, Lurie et al 2006a). 


The present four-year report addresses a number of 

concerns about the external validity of randomised 

controlled trials involving surgery. Are patients who are 

happy to be randomised to either surgery or 

conservative treatment representative of all patients? In 

addition, protocol failure is common with patients 

randomised to conservative care receiving surgery and 

vice versa. With these concerns in mind, this study 

maximised recruitment by having both a randomised 

and an observational cohort and attempted two 

analyses, so that the as-per-treatment analysis could 

offer a possibly truer treatment effect than the intentionto-treat 

analysis, which was associated with 

considerable non-adherence to treatment assignment. 

With an intention-to-treat analysis, there were few 

significant differences between the surgically treated 

and conservatively treated patients, but cross-over into 

the alternate interventions was very common, and so 

an alternative analysis was also offered. Intention-totreat 

analysis is the preferred way that trials should be 

evaluated, by which is meant patients are analysed by 

the groups to which they were randomised whatever 

treatment they actually received. With the analysis by 

treatment actually received, there were significant 

differences favouring the surgically treated groups at all 

time points. It should be noted that the conservatively 

treated group also recorded substantial improvements 

in all outcomes, though these were not as great as in 

the surgically treated group. For instance, Oswestry 

scores had improved by 25% and 38% respectively at 

four years. Improvements in both surgically treated and 

conservatively treated patients reached a maximum at 

six months and then persisted over four years. 

The study contained both a randomised and an 

observational cohort, depending on patient 

preferences, with patients in the latter able to choose 

their intervention. The two cohorts were very similar at 

baseline, with the observational group being slightly 

more symptomatic and functionally impaired, but there 

were no significant differences in any of the primary 

outcomes so the combined analysis of the two cohorts 

seemed justified. 

The non-operative treatment intervention was nonstandardised 

and poorly described, but was a mixture 

of medical interventions (NSAIDs, opioids, analgesics 

and corticosteroid injections) and manual therapy. 

Although ‘physical therapy’ was the commonest 

ingredient, this is not described in any detail. This lack 

of description makes it difficult to suggest which among 

the various forms of conservative care has the better 

effect. Over 70% of the disc herniations were described 

as extrusions and sequestrations, so it seems unlikely 

that any conservative treatment was able to influence 

the herniation in any way. However, the optimal 

conservative treatment for disc herniation is not clear. 

It should be acknowledged that certain design features 

incorporated in these studies make a less robust 




design than a randomised placebo-controlled trial using 

intention-to-treat analysis, which offers greater 

protection from confounding or bias. In a commentary 

on the earlier studies, Flum (2006) noted a number of 

surgical interventions (ligatation of the internal 

mammary artery for coronary disease, arthroscopic 

debridement and joint irrigation for osteoarthritis of the 

knee, and transmyocardial laser revascularisation for 

heart failure), that were assumed to be beneficial until 

tested against placebo surgery and were then found to 

be no better than the placebo interventions. He argued 

that such sham-controlled trials are only justified when 

genuine equipoise exists about the merits of an 

intervention, which was the situation at the heart of the 

present trial. The authors argued that a shamcontrolled 

trial was impractical and unethical. It could 

be argued that given the power of patient expectations 

clearly demonstrated in previous placebo-controlled 

surgical trials, the only definitive way of proving the 

efficacy of surgery over non-operative care is through a 

placebo-controlled trial. 

The patients who received surgery differed in a number 

of ways from those who did not. They tended to be 

more bothered by their pain and disability, felt their 

clinical situation was deteriorating, were younger, and 

had lower income and educational levels. Furthermore, 

patients could have been considerably influenced by 

the referring physician’s bias to direct to surgery or to 

conservative care, and thus their expectations parallel 

their physician’s bias. 

Previous studies 

Weber (1983) described a randomised controlled trial 

of 126 patients with radicular pain still present after 14 

days and neurological signs and symptoms who were 

randomised to surgery or conservative treatment. The 

conservative treatment is vaguely described as 

involving a gradual increase in activity and exercises 

and Back School. Again, the problem of crossover was 

common with 17 of the conservatively treated group 

receiving operation. Outcomes statistically favoured the 

surgery group at one year, but not at four and ten 

years. There were good outcomes in 33% of nonoperative 

group, 47% of non-operative group receiving 

surgery and 66% of surgery group a one year. At four 

and ten years there were good outcomes in 51% and 

55% of non-operative group; 53% and 59% of nonoperative 

group receiving surgery; and 66% and 58% 

of surgery group. This is a classic study and the first 

attempt to compare operative and non-operative care 

for sciatica, however potentially critical flaws have been 

noted in the methodology (Bessette et al 1996). 

Namely, the large number of crossovers, the 

inadequate sample size and the insensitive outcome 

measurements. 

The Maine Lumbar Spine Study was purely an 

observational study of operative and non-operative 

care for sciatica as selected by physician and patient 


with no attempt at randomisation (Atlas et al 1996, 

2001, 2005). Two hundred and seventy-five patients 

were treated surgically and 232 conservatively; 

symptoms had been present for more than three 

months in over 50% of patients. The conservative 

treatment was non-standardised, but included bed rest, 

exercises, physical therapy, manipulation TENS, back 

brace, narcotic analgesics and epidural steroids. In the 

patients initially to be treated non-operatively 23% 

received surgery in the year following. Both groups 

improved but improvements were statistically better in 

the surgery group, for instance, the predominant 

symptoms was much better or gone in 71% of surgery 

group and in 43% of non-operative group (P < 0.001). 

Employment and workers’ compensation status was 

equal in both groups. There were still some significant 

differences favouring surgery over conservative care at 

five and at ten years, such as more resolution of 

symptoms and greater satisfaction, but work and 

disability status was the same in both groups at ten 

years (Atlas et al 2001, 2005). 

The outcomes however tended to reflect the severity of 

baseline symptoms. The surgery group was much 

worse at baseline, with significantly more severe 

symptoms and higher levels of disability, and thus 

greater improvements would be expected from their 

worse start point. Most of the patients with mild 

baseline symptom severity were treated nonoperatively, 

but in this group there were no significant 

differences between groups. Patients with moderate 

baseline symptom severity were treated roughly equally 

operatively and non-operatively, with significant 

differences favouring the former. The majority of 

patients with severe baseline symptom severity were 

treated surgically. The non-randomised study design 

and the lack of baseline similarity, are serious flaws in 

the Maine Lumbar Spine Study, which increase the risk 

of confounding and bias affecting the results. 

Osterman et al (2006) compared microdiscectomy with 

a control group that received ‘similar physiotherapeutic 

instructions initially’ and isometric exercises in a 

randomised controlled trial with 56 patients with six to 

twelve weeks of radicular pain, neurological signs and 

symptoms and imaging studies confirming a relevant 

disc extrusion or sequestration. Eleven patients in the 

control group (40%) crossed over to receive surgery, 

but both an intention-to-treat analysis and an as-treated 

analysis was conducted. Both groups improved over 

the two-year follow-up, with greater improvements in 

the surgery group in most outcomes at most time 

points, but only differences in leg pain at six weeks and 

satisfaction with treatment significantly favoured the 

surgery group. The as-treated analysis showed no 

significant differences between those who actually 

received surgery and those who actually received nonoperative 

care. The small sample size and large 

number of crossovers in this trial mean that the results 

should be interpreted with caution. 




Peul et al (2007) conducted a randomised controlled 

trial with 283 patients with severe sciatica of six to 

twelve weeks duration and corresponding imaging 

studies who received either early surgery or prolonged 

conservative care, which involved GP advice, 

encouragement to resume daily activities, and referral 

to a physiotherapist if reluctant to move. Fifty-five 

patients (39%) in the conservative care group received 

surgery, but mean time to surgery was 19 weeks, 

compared to two weeks in those assigned to surgery; 

all analyses were by intention-to-treat. Both groups 

made substantial improvements over time, with greater 

improvements in the surgery group in most outcomes 

initially, though differences between the groups 

became minimal by about six months. There were 

significant differences favouring the surgery group in 

leg pain severity, but not in functional disability or back 

pain. A two-year follow-up found that 89% and 44% of 

those assigned to surgical and conservative care 

actually received surgery (Peul et al 2009). Again, the 

earlier short-term benefit of improvement in leg pain 

was no longer significant by six months and continued 

to narrow after this. Three percent of the conservative 

cohort and 5% of the early surgery group had recurrent 

sciatica and further surgical intervention. Irrespective of 

groups, 20% of patients had an unsatisfactory outcome 

at two years. They conclude ‘since the treatment 

effects of early surgery are gone after six months, wellinformed 

patients, rather than physicians should decide 

whether and when to have surgery’ (Peul et al 2009). 

Conclusions and implications 

With these randomised controlled trials and 

observational studies comparing operational and nonoperational 

interventions for sciatica due to disc 

herniation, the conclusions are very similar. Very often 

surgery resulted in earlier improvement, especially in 

leg pain, but long-term results tend to equalise. 

However, these results should not be accepted at face 

value given the methodological flaws discussed above. 

The randomised controlled trials have a number of 

major weaknesses, such as limited sample size, large 

numbers of cross-over, poor outcome measurement, 

and the use of an as-treated analysis in place of the 

more rigorous analysis by intention-to-treat to generate 

statistically significant results. The observational 

studies also suffer from critical flaws, principally lack of 

randomisation resulting in unequal baseline measures 

and other confounding factors. In other words, although 

it looks like the surgery intervention offers short-term 

better improvement, there are a number of reasons 

why these outcomes should be interpreted with 

caution. The non-operative care groups improved also, 

though not always as much initially, and the longerterm 

improvement is very similar between operative 

and non-operative groups. The inclusion criteria for 

these studies were generally similar regarding radicular 

symptoms, neurological signs and symptoms and 

imaging study confirmation of a disc herniation. 


Exclusion criteria were often similar and related to 

requirement for urgent surgery, such as cauda equina, 

or previous surgery. Duration of symptoms did vary 

between studies however. 

Duration of symptoms is a crucial issue, rather difficult 

to disentangle and presents something of a Catch-22 in 

this area of research. If patients are recruited too soon 

and randomised to surgery then this could include a 

large number who would have gotten better with 

conservative care or spontaneously recovered given 

more time. However, if recruitment is left too long then 

patients will mostly already have failed conservative 

treatment, and so to randomise to yet more 

conservative care is likely to have a negative effect on 

patient expectations, is likely to encourage noncompliance 

with conservative group randomisation and 

create a bias for the surgery group. 

The "bottom line" for your patient with sciatica is 

probably that if they are desperate to return to work or 

have young children to attend to or other desperate life 

choice issues go for surgery; especially with more 

severe leg pain and disability. Otherwise, if they are 

coping, and especially with milder symptomatology, 

non-operative management seems most appropriate. 

Interestingly despite greater improvements in 

symptoms and disability in the surgery groups, several 

studies found no differences in work or work 

compensation status. This is obviously an important 

outcome given the greater cost of surgery, but a health 

economics analysis would be needed to address the 

question of which treatment is most cost-effective. 

Several studies have suggested that those with more 

severe symptoms, especially leg pain are more likely to 

benefit from surgery; whilst those with mild symptoms 

are less likely to see a clear difference in favour of 

surgery. Apart from this factor of symptom severity, 

there was limited comment on any other prognostic 

factors favouring either intervention. One study found 

that age older than 37 years and disc herniation at L4- 

L5, rather than L5-S1, were factors favouring surgery, 

but these are hypotheses that need verification 

(Osterman et al 2006). 

Is there an optimal conservative treatment strategy 

provided by these studies? Unfortunately not; in 

general, treatments are non-standardised, eclectic and 

often described as ‘usual care’. The descriptions 

sometimes give a brief overview of the type of things 

that might be included, but are mostly so vague, 

heterogeneous and eclectic that it is impossible to 

know what was really done and replicate it. In most of 

these studies, it seems that the non-operative care that 

was given received scant consideration of its contents, 

and certainly no ‘best course’ for non-operative care is 

apparent. The description of conservative care as 

‘physical therapy’ is particularly unhelpful and displays 

an additional ignorance and contempt for the 




conservative treatment arm. It would be helpful in future 

studies if the actual intervention was detailed and 

physical therapy not considered a "black box". The 

inclusion/exclusion criteria of these studies frequently 

delineate characteristics of patients who are likely to 

respond to surgery, another bias to favour this group. 

Nothing in the inclusion/exclusion criteria makes any 

suggestion about selection for conservative care, such 

as directional preference/centralisation. This is not an 

even playing field, and not a true comparison of 

surgery versus conservative care. 

From the perspective of Mechanical Diagnosis and 

Therapy, several questions remain unanswered. In 

some of the studies extrusions and sequestration were 

differentiated from disc protrusions, and made to be 

inclusion criteria or used as descriptors of the disc 

herniations. However, no attempt was made to 

differentiate the various types of disc herniation and 

their response to non-operative care. It did not appear 

to make any difference, but was this because the nonoperative 

care made no attempt to look for 

centralisation or directional preference? Is there a subgroup 

with sciatica who may respond as rapidly to 

conservative care as to surgery? Is the group with 

sequestration and extrusions the one most likely to 

benefit from early surgery? 

We suggest the wrong questions are being asked; it 

should not be, “is surgery or conservative care better 

for patients with disc herniations?” It should be “who 

will respond best to conservative care, and who will 

respond best to surgery?” These are probably two 

exclusive groups – those that respond conservatively 

do not need surgery, and those that need surgery will 

not respond conservatively. Again, it relates to the 

assumption that even ‘specific’ sub-groups of back pain 

patients are homogeneous, but we would suggest that 

not all patients with disc herniations are the same. The 

way these are sub-classified could be related to 

pathology (protrusions/extrusions/sequestrations), but 

we suggest a more useful classification would be 

based on directional preference/centralisers versus non 

-responders. 

A number of study designs could begin to address 

these questions, for instance: 

�� Evaluation of MDT as a screening tool to distinguish 

responders (with directional preference/ 

centralisation) from non-responders that need 

surgery. 

�� Cohort/observational study of patients with sciatica 

with baseline data about symptoms, neurological 

signs and symptoms, imaging studies and directional 

preference/centralisation status, etc. to determine 

which factors are most important predictors of 

outcomes, which might include surgery, as well as 

symptoms and function. 


�� Randomised trial comparing different specific 

conservative interventions in responders. 

�� Randomised trial comparing surgery or ‘wait and see 

policy’ in non-responders. 

�� Randomised trial in specific sub-groups of disc 

herniation patients (protrusion, extrusion, 

sequestration) comparing surgery and specific nonoperative 

care. 

References 

Atlas SJ, Deyo RA, Keller RB, Chapin AM, Patrick DL, 

Long JM, Singer DE (1996). The Maine lumbar spine 

study, part II. 1-year outcomes of surgical and 

nonsurgical management of sciatica. Spine 21.1777- 

1786. 

Atlas SJ, Keller RB, Chang Y, Deyo RA, Singer DE 

(2001). Surgical and non-surgical management of 

sciatica secondary to a lumbar disc herniation: five-year 

outcomes from the Maine Lumbar Spine Study. Spine 

26.1179-1187. 

Atlas SJ, Keller RB, Yen AW, Deyo RA, Singer DE 

(2005). Long-term outcomes of surgical and 

nonsurgical management of sciatica secondary to a 

lumbar disc herniation: 10 year results from the Maine 

Lumbar spine Study. Spine 30.927-935. 

Bessette L, Liang MH, Lew RA, Weinstein JN (1996). 

Classics in Spine. Surgery literature revisited. Spine 

21.259-263. 

Flum DR (2006). Interpreting surgical trials with 

subjective outcomes. Avoiding unsportsmanlike 

conduct. JAMA 296.2483-2485. 

Gibson JNA, Waddell G (2007). Surgical interventions 

for lumbar disc prolapse. Updated Cochrane review. 

Spine 32.1735-1747. 

Hoffman RM, Wheler KJ, Deyo RA (1993). Surgery for 

herniated lumbar discs: a literature synthesis. J Gen 

Intern Med 8.487-496. 

Kramer J (1990). Intervertebral Disk Diseases. Causes, 

Diagnosis, Treatment and Prophylaxis. (2nd Ed.) 

Thieme Medical Publishers, New York. 

McKenzie R, May S (2003). The Lumbar Spine: 

Mechanical Diagnosis and Therapy. Spinal 

Publications New Zealand Ltd, Waikanae, New 

Zealand. 

Memmo PA, Nadler S, Malanga G (2000). Lumbar disc 

herniations: a review of surgical and non-surgical 

indications and outcomes. J Back Musculoskeletal 

Rehab 14.79-88. 

Osterman H, Seitsalo S, Karppinen J, Malmivaara A 

(2006). Effectiveness of microdiscectomy for lumbar 

disc herniation. A randomized controlled trial with 2 

years of follow-up. Spine 31.2409-2414. 




Peul WC, van Houwelingen HC, van den Hout WB et al 

(2007). Surgery versus prolonged conservative 

treatment fro sciatica. New England J Med 356.2245- 

2256. 

Peul WC, van den Hout WB, Brand, R, Thormeer 

TWM, Koes BW, et al (2009). Prolonged conservative 

care versus early surgery in patients with sciatica 

caused by lumbar disc herniation: two year results of a 

randomised controlled trial. Br Med J doi:10.1136/ 

bmja143 

Weber H (1983). Lumbar disc herniation. A controlled 

prospective study with ten years of observation. Spine 

8.131-140. 

Medlem søges til udvalget 


Weinstein JN, Tosteson TD, Lurie JD et al (2006a). 

Surgical versus nonoperative treatment for lumbar disc 

herniation. The spine patient outcomes research trial 

(SPORT). A randomized trial. JAMA 296.2441-2450. 

Weinstein JN, Lurie JD, Tosteson TD et al (2006a). 

Surgical versus nonoperative treatment for lumbar disc 

herniation. The spine patient outcomes research trial 

(SPORT). An observational cohort. JAMA 296.2451- 

245. 

Vinderne af konkurrencen på den netop afholdte Fagfestival blev : 

1. præmie ; et McKenzie kursus ; Tanya Nielsen, København 

2. præmie; "The Cervical and Thoracic Spine MDT af Robin McKenzie" ; Morten Markvard Aalborg 

3. præmie; "The Lumbar Spine MDT af Robin McKenzie" ; Bo Hangen, Ry 

4. Præmie; "Treat Your Own Neck og Treat Your Own Neck" samt en Lumbar Roll ; Merethe Schultz, Aalborg 

alle præmier er sponseret af ProTerapi 

Fagligt udvalg IMDT 

Bestyrelsen har, som led i de mere langsigtede visioner for instituttet, besluttet at nedsætte et Fagligt Udvalg indenfor 

IMDT. 

Udvalget nedsættes foreløbig for en 2-årig periode, hvor der er budgetteret med 2 årlige heldagsmøder (honoreres 

med en dagsdiæt kr 2315,-). 

Det faglige udvalg skal være en ”tænketank” med reference til bestyrelsen i forhold til den fortsatte faglige udvikling 

indenfor Mekanisk Diagnostik og Terapi. 

Det er således tanken at det faglige udvalg fungerer som en idébank og ikke som udførende kraft. 

Udvalgets sammensætning: 1 bestyrelsesmedlem, 1 underviser, 1 Diploma samt et medlem af IMDT som ikke aktuelt 

er engageret i undervisningsgruppen eller bestyrelsen. 

Se yderligere informationer på hjemmesiden: www.Mckenzie.dk 

Ansøgningsfrist d. 1. maj 2009 





Elizabeth Kelly E, BSc, Stephen May, MA FCSP, Dip. MDT, MSc, Jenny Ross, MCSP, BSC 

Abstract 

Objectives: To investigate the reliability of experienced physiotherapists in classifying patients in McKenzie’s nonspecific 

mechanical syndromes from extremity McKenzie assessment forms. 

Design: Inter-therapists reliability pilot study using patient vignettes. 

Setting: NHS outpatients department, UK. 

Method: Three McKenzie ‘credentialed’ therapists, who had taken a course in Mechanical Diagnosis and Therapy 

for the Extremities, independently examined 11 completed McKenzie assessment forms with the classification conclusion 

omitted. These had been completed by a fourth therapist who was a faculty member of the McKenzie Institute 

teaching programme, and whose classification conclusions represented a ‘gold standard’ against which to judge 

the other therapists’ conclusions. 

Outcomes: Reliability was expressed as percentage agreement and kappa. 

Results: The reliability of syndrome classification was ‘good’, 82% agreement, kappa 0.7. Against the ‘gold standard’ 

the therapist with more training was more reliable. 

Conclusion: In this pilot study, the reliability of McKenzie’s mechanical syndrome classification in extremity patients 

was good and warrants further investigation in a larger cohort of therapists and amongst real patients rather than 

paper vignettes. 

Key words: reliability, McKenzie, assessment form, extremity 

Introduction 

McKenzie first described a system of Mechanical Diagnosis 

and Therapy (MDT) that involved the classification 

and management of back pain (McKenzie 1981). The 

system is based on non-specific classification into mechanical 

syndromes that are derived from an assessment 

that uses repeated movements whilst symptoms 

are monitored, which then directs management. In the 

first textbook, McKenzie stated that the principles could 

equally be applied to any musculoskeletal problem, including 

extremity problems; and the application of the 

system to extremity problems was later described 

(McKenzie and May 2000). 

The mechanical syndromes are non-specific and do not 

seek to apply specific pathological labels to musculoskeletal 

symptoms. In extremity problems the syndromes 

are: 

�� derangement; identified by the abolition or decrease 

of symptoms, and an increase in restricted range of 

movement in response to repeated movements 

�� articular dysfunction, which is identified by intermittent 

pain consistently produced at a restricted endrange 

with no rapid change of symptoms or range 

�� contractile dysfunction, which is identified by intermittent 

pain, consistently produced by loading the 

musculotendinous unit, for instance, with an isometric 

contraction against resistance 

�� postural syndrome is only produced by sustained 

loading, which once avoided, the rest of the physical 

examination should be normal 

Each syndrome requires a different management approach. 

If the patient cannot be classified in one of 

the mechanical syndromes, they are classified as nonmechanical 

or ‘other’, which includes categories such 

as, trauma, post-surgery or chronic pain state 

(McKenzie and May 2000, 2003). 

Such non-pathology based classifications are in contrast 

to much literature about extremity musculoskeletal 

problems, which are based on a diagnose-and-treat 

medical model (Cyriax 1981; Corrigan and Maitland 

1983; Kesson and Atkins 1998). Several problems exist 

with such a paradigm. Of the myriad tests that are used 

to make such diagnoses, only some have been scientifically 

validated as being able to identify, the specific lesion 

that it is claimed they identify or where validation 

has been attempted problems exist about diagnostic 

accuracy. For instance, diagnostic tests for subacromial 

impingement syndrome have demonstrated reasonable 

sensitivity or specificity, but respectively poor specificity 

or sensitivity (Calis et al 2000). Or a review of tests for 

meniscal or ligament injuries at the knee generally 

found moderate levels of sensitivity and specificity for 

tests for anterior and posterior cruciates and meniscus, 

but no reports at all about medial or lateral collateral 

ligaments (Solomon et al 2001). 

Furthermore, the reliability of many tests is weak to 

moderate at best (Cooperman et al 1990; Fitzgerald 

and McClure 1995; McClure et al 1989; Cushnagan et 

al 1990; Hayes et al 1994; Jones et al 1992; Liesdek et 

al 1997; de Winter et al 1999), which means that clinicians 

are coming to different conclusions from the same 

tests. This does not make for a sound basis on which 

management strategies should be constructed. Furthermore 

Jones and Rivett (2004) suggested that a structurally 

based assessment process alone does not provide 

sufficient information to understand the problem, nor to 

justify the course of management; and that structureoriented 

clinical reasoning tends to promote inflexible 

management strategies. 




There is very limited scientific literature about the 

application of MDT principles to extremity 

musculoskeletal problems. Two case studies of a 

shoulder derangement and a shoulder contractile 

dysfunction have been published (Aina and May 2005, 

Littlewood and May 2007). Amongst McKenzie trained 

therapists the classification system has demonstrated 

reasonable clinical utility with 69% of 242 non-spinal 

patients being classified in one of the mechanical 

syndromes (May 2006). 

A key aspect of any classification system is reliability – 

do different clinicians reach the same conclusion about 

a patient in front of them? A reasonable level of 

reliability has been demonstrated with back and neck 

pain patients by therapists who are trained in the 

McKenzie approach (Kilpikoski et al 2002; Ramjou et al 

2000; Clare et al 2004, 2005). Reliability studies have 

been conducted either using real patients with one 

clinician observing and one assessing (Ramjou et al 

2000; Clare et al 2005), or two adjacent examinations 

(Kilpikoski et al 2002), or using paper vignettes on 

standard assessment forms (Clare et al 2004), or using 

videotaped examinations (Dionne et al 2006). There is 

no published study that has evaluated the reliability of 

the MDT classification system as it applies to extremity 

musculoskeletal problems. The aim of this study was, 

therefore, to investigate the inter-examiner reliability of 

experienced MDT clinicians in classifying patients from 

information presented on McKenzie extremity 

assessment forms. 

Method 

Ethics approval for this study was gained from Robert 

Gordon University Student Project Ethical Review 

board and from Aberdeen NHS Research Ethics 

Committee, UK. 

The examiners were three physical therapists with a 

high level of training in the McKenzie method; they had 

all taken the Credentialing exam (1997) and the 

extremity MDT course (2001), and one had attained a 

Diploma in MDT, which is the highest level of education 

in the McKenzie Institute educational programme. The 

therapists had worked together for a number of years, 

and commonly assessed patients together and 

discussed findings and definitions that were to be used 

in the study. They were all familiar with the definitions 

and operational definitions provided for MDT 

mechanical classifications (McKenzie and May 2000, 

2003). 

Eleven MDT extremity patient assessment forms were 

used; these were real patient records with patient 

details and classification conclusion omitted. These 

were recorded by a Senior Faculty member of the 

McKenzie Institute education programme, who had 

previously classified these patients, thus providing a 

‘gold standard’ diagnostic classification against which 

the other three classifications could be judged. 


The three therapists independently analysed the 11 assessment 

sheets and following this, classified each patient 

in one of the mechanical syndromes (McKenzie 

and May 2000). Forms were completed in separate 

rooms and no discussion about the patient vignettes 

was permitted. The therapists also completed a ‘clinical 

decision making form’, which detailed provisional classifications 

and reasons for decisions as the form was being 

analysed. The aim of this data is that it would provide 

some understanding of the rationale for decisions 

made. 

Kappa coefficient and percentage agreement was calculated 

across the three therapists. Kappa was also 

calculated for each therapist compared to the ‘gold 

standard’ of the Senior Faculty member of the 

McKenzie Institute. Kappa gives an interpretation of 

agreement that takes chance into account, with kappa 

of 0.00 equal to no better than chance agreement, and 

kappa of 1.00 meaning perfect agreement. Interpretation 

of kappa was according to Altman (1991): < 0.20 

poor; 0.21-0.40 fair; 0.41-0.60 moderate; 0.61-0.80 

good; 0.81-1.00 very good. Data was analysed using 

the MKAPPASC.SPS macro in SPSS version 13.0 for 

Windows. 

The interpretation of the data on the decision making 

process was by content analysis, with recurring issues 

between therapists forming themes (Green and 

Thorogood 2004). Rigour of data analysis was provided 

by a second researcher identifying themes blinded to 

previous analysis. Data will be presented as themes 

with examples. 

Results 

Multi-rater agreement between the three therapists was 

82%, kappa value 0.70. Agreement between the three 

therapists and the ‘gold standard’ was 91%, kappa 0.86; 

73%, kappa 0.55; 73% kappa, 0.57. 

The clinical reasoning process involved the following: 

Theoretical knowledge: 

The timeframe excluded dysfunction but could implicate 

inflammatory. 

Pain at rest may indicate derangement, inflammatory, 

chronic pain state, serious pathology. 

Pattern recognition: 

There was an increase in range of movement therefore 

a derangement. 

Pattern of end range pain, no worse with repeated 

movements, no directional preference, so articular dysfunction. 

Directional preference for extension, so a derangement. 




Use of the classification system: 

The therapists agreed that a patellofemoral problem 

was unlikely to be classified in the McKenzie system, 

and should be classified as ‘other’. 

Limitations to assessment form: 

Timeframe does not suggest contractile dysfunction 

although contractile unit implicated following resisted 

tests...would like further testing. 

Discussion 

In this reliability study, three McKenzie trained clinicians 

classified 11 patient vignettes on McKenzie extremity 

assessment forms and reached a ‘good’ level of agreement 

(Altman 1991), with a kappa of 0.70. The most 

highly trained of the three therapists with a Diploma in 

MDT had ‘very good’ reliability, with a kappa of 0.86, 

compared to the ‘gold standard’; compared to the Credentialed 

therapists with ‘moderate’ levels of agreement 

and kappa at 0.55 and 0.57. 

As stated initially, this is the first study into the reliability 

of the McKenzie extremity assessment and classification 

system. Reliability studies have been conducted 

previously into back and neck pain patients (Kilby et al 

1990, Riddle and Rothstein 1993, Kilpikoski et al 2002; 

Ramjou et al 2000; Clare et al 2004, 2005). These generally 

found ‘good’ and ‘very good’ levels of agreement, 

except when therapists had minimal or no training in the 

McKenzie system (Kilby et al 1990, Riddle and Rothstein 

1993), when kappa values indicated only ‘fair’ 

agreement. Only one other study has used paper vignettes 

to test the reliability of the McKenzie classification 

system with 25 cervical and 25 lumbar spine assessment 

forms (Clare et al 2004); this produced kappa 

values of 0.56 and 0.68 for syndromes and subsyndromes. 

The present study suggests that the classification 

system has similar levels of reliability when 

used with extremity patients, and even better amongst 

highly trained therapists. This latter point that better 

trained therapists will have better levels of reliability has 

not been formally tested before with the MDT classification 

system, but is suggested by a comparison of the 

data from Kilpikoski et al (2002), Ramjou et al (2000) 

and Clare et al (2004, 2005) with the data from Riddle 

and Rothstein (1993). 

Clinical reasoning involved in the therapists’ decision 

making involved theoretical background knowledge, 

pattern recognition, and understanding of the presence 

and absence of the classifications’ definitions; but also 

included recognition of the difficulties and limitations of 

making classification decisions based on paper case 

studies. Clinical reasoning processes have been studied 

quite widely (Jones and Rivett 2004), but the process 

involved in how therapists make classification decisions 

has not previously been studied. 

The lack of interaction with the patient that is inherent 

with using paper cases may have hindered the level of 

agreement between therapists lacking the benefit of the 


wealth of data that personal interaction brings. However 

these were real case studies. Besides the limited applicability 

with a real clinical interaction of using paper 

cases, several other limitations with the present study 

are apparent. The sample size of therapists and paper 

cases was small; operational definitions were not predefined 

for the study, but were available (McKenzie and 

May 2000, 2003) and known to the participants; furthermore, 

these results cannot be generalised to other 

therapists who do not have the level of training of these 

clinicians. 

This pilot study would suggest that the McKenzie extremity 

assessment and classification process has good 

reliability amongst trained therapists, and thus meets 

one aspect of clinical utility necessary in any classification 

system. Further research is needed to confirm the 

reliability of the system, and it is suggested that if a 

study with paper vignettes is repeated that this involves 

larger numbers of trained therapists. The alternative is a 

reliability study with real patients in select groups of patients, 

such as with shoulder or knee pain. 

In conclusion, this study with three McKenzie trained 

therapists found a good level of reliability, kappa of 0.7, 

when they classified 11 paper vignettes on McKenzie 

extremity assessment forms. 

References 

Aina A, May S. A shoulder derangement. Manual Therapy 

2005;10:159-163. 

Altman DG. Practical Statistics for Medical Research. 

Chapman & Hall, London, 1991, p404. 

Calis M, Akgun K, Birtane M, Karacan I, Calis H, Tuzun 

F. Diagnostic values of clinical diagnostic tests in 

subacromial impingement syndrome. Ann Rhem Dis 

2000;59:44-47. 

Clare HA, Adams R, Maher CG. Reliability of the 

McKenzie spinal pain classification using patient assessment 

forms. Physiotherapy 2004;90:114-119. 

Clare HA, Adams R, Maher CG. Reliability of McKenzie 

classification of patients with cervical or lumbar pain. J 

Manipulative Physiol Ther 2005;28:122-127. 

Cooperman JM, Riddle DL, Rothstein JM (1990). Reliability 

and validity of judgements of the integrity of the 

anterior cruciate ligament of the knee using the 

Lachman’s test. Physical Therapy 70.225-233. 

Corrigan B, Maitland GD. Practical Orthopaedic Medicine. 

Butterworth Heinemann, Oxford, 1983. 

Cushnaghan J, Cooper C, Dieppe P, Kirwan J, McAlindon 

T, McCrae F (1990). Clinical assessment of osteoarthritis 

of the knee. Ann Rheum Dis 49.768-770. 

Cyriax J. Textbook of Orthopaedic Medicine. Diagnosis 

of Soft Tissue Lesions (8th edition).Bailliere Tindall, 

London, 1982. 




Dionne CP, Bybee RF, Tomaka J. Inter-rater reliability 

of McKenzie assessment in patients with neck pain. 

Physiotherapy 2006;92:75-82. 

De Winter AF, Jans MP, Scholten RJPM, Deville W, van 

Schaardenburg D, Bouter LM. Diagnostic classification 

of shoulder disorders: interobserver agreement and determinants 

of disagreement. Ann Rheum Dis 

1999;58:272-277. 

Fitzgerald GK, McClure PW (1995). Reliability of measurements 

obtained with four tests for patellofemoral 

alignment. Physical Therapy 75.84-92. 

Green J, Thorogood N. Qualitative Methods for Health 

Research. Sage, London. 

Hayes KW, Petersen C, Falconer J (1994). An examination 

of Cyriax’s passive motion tests with patients having 

osteoarthritis of the knee. Physical Therapy 74.697- 

709. 

Jones A, Hopkinson N, Pattrick , Berman P, Doherty M 

(1992). Evaluation of a method for clinically assessing 

osteoarthritis of the knee. Ann Rheum Dis 51.243-245. 

Jones M, Rivett DA. Introduction to clinical reasoning. 

In: Eds, Jones M, Rivett DA. Clinical Reasoning for 

Manual Therapists. Butterworth Heinemann, Edinburgh, 

2004, p16-17. 

Kesson M, Atkins E. Orthopaedic Medicine. A Practical 

Approach. Butterworth Heinemann, Oxford, 1998. 

Kilby J, Stigant M, Roberts A. The reliability of back pain 

assessment by physiotherapists, using a “McKenzie 

algorithm”. Physiotherapy 1990;76:579-583. 

Kilpikoski S, Airaksinen O, Kankaanpaa M, Leminen P, 

Videman T, Alen M. Interexaminer reliability of low back 

pain assessment using the McKenzie method. Spine 

2002;27:E207-E214. 

Liesdek C, van der Windt DAWM, Koes BW, Bouter LM. 

Soft-tissue disorders of the shoulder. A study of interobserver 

agreement between general practitioners and 

physiotherapists and an overview of physiotherapeutic 

treatment. Physiotherapy 1997;83:12-17. 

Littlewood C, May S. A contractile dysfunction of the 

shoulder. Manual Therapy 2007;12:80-83 


May S. Classification by McKenzie’s mechanical syndromes: 

report on directional preference and extremity 

patients. International J Mechanical Diagnosis and 

Therapy 2006;1:3:7-11. 

McKenzie RA. The Lumbar Spine. Mechanical Diagnosis 

and Therapy. Spinal Publications, New Zealand, 

1981. 

McKenzie R, May S. The Human Extremities Mechanical 

Diagnosis and Therapy. Spinal Publications New 

Zealand Ltd, 2000. 

McKenzie R, May S. The Lumbar Spine Mechanical 

Diagnosis and Therapy. Spinal Publications New Zealand 

Ltd, 2003. 

McClure PW, Rothstein JM, Riddle DL. Intertester reliability 

of clinical judgements of medial knee ligament 

integrity. Physical Therapy 1989;69:268-275. 

Razmjou H, Kramer JF, Yamada R. Intertester reliability 

of the McKenzie evaluation in assessing patients with 

mechanical low-back pain. J Orthop Sports Phys Ther 

2000;30:368-389. 

Riddle D, Rothstein J. Intertester reliability of 

McKenzie’s classifications of the syndrome types present 

in patients with lumbar pain. Spine 1993;18:1333- 

1344. 

Solomon DH, Simel DL, Bates DW, Katz JN, Schaffer 

JL. Does this patient have a torn meniscus or ligament 

of the knee? Value of the physical examination. JAMA 

2001;286:1610-1620. 

Authors: 

Kelly E, BSc* 

May S, FCSP, MSc** 

Ross J, MCSP, BSc */~~ 

*Formerly Robert Gordon University, Aberdeen, Scotland 

** Faculty of Health and Wellbeing, Sheffield Hallam 

University, Broomhall Road, Sheffield, S10 2BP, UK 

00 44 (0)114 225 2370 

00 44 (0)114 221 7303 (fax) 


** corresponding author 

~~College of Medicine, Dentistry and Nursing, 

Dundee University, Scotland 





Martin Melbye, PT, Dip.MDT, Camilla Nymand, PT, Dip. MDT & Troels Balskilde, PT, Dip.MDT (DNK) 

Abstract 

Scientific literature does not report duration of successful treatment programmes for neck and low back patients, 

although this is one of the key interests of patients. Based on empirical knowledge, patients classified into the derangement 

syndrome, as described by McKenzie, respond rapidly to conservative management. Previous randomised 

controlled trials have reported clinically relevant outcomes for low back patients, following a two and eight week 

treatment programme. The purpose of this study is to track treatment duration for cervical and lumbar patients classified 

into the derangement syndrome. Patients self reported scores for pain, functional level and self efficacy at 

intake and discharge to evaluate whether the duration of treatment was sufficient to allow for clinical relevant changes 

in health status. Patients were treated with home exercise, therapist generated forces, posture correction, functional 

restoration exercises and prophylactic instructions. A total of 66 lumbar and 26 cervical spine patients were 

included and treated for an average of 2.4 and 4.3 weeks, respectively. Scores for pain, functional level and self 

efficacy improved significantly and to a clinically relevant extent. Effect sizes for lumbar patients were consistent 

with findings from two randomized controlled trials. The external validity of this study may be limited due to missing 

data from excluded patients, as well as existence of neurological findings and other indicators of severity. 

Keywords: Outcome measure, clinical database, derangement syndrome, prognosis 

Background 

Predicting the future is a difficult task in general and for 

musculoskeletal therapy clinicians as well. A variety of 

prognostic indicators are being reported in the scientific 

literature, such as personal, clinical, biomechanical and 

psychosocial variables. Few of these studies report an 

actual timeframe for succesful treatment of patients 

though. However, information about prognosis is one of 

patients’ key interests (May 2001), and it is obviously 

predictable that people will want to know when they 

might be able to return to work and normal activities. 

Research also show that well informed patients are more 

compliant, more satisfied with treatment and achieve 

better treatment results (Robinson and Thomson 2001). 

In a study of 260 low back patients, Petersen (Petersen 

et al 2002) showed improvement in outcome measures 

after an eight week intervention of either McKenzieoriented 

treatment or intensive strengthening. There 

was no reporting on differences between patients classified 

into various subgroups. In another study by Long 

(Long et al 2004), a total of 80 low back patients matching 

the derangement subgroup, improved significantly 

following a two week treatment programme. Systematic 

data collection from clinical practice, could assist the 

clinician when informing the patient about prognosis 

and setting realistic treatment goals. 

It is suggested that low back patients classified into the 

derangement-syndrome, as described by Robin McKenzie 

(McKenzie and May 2003; McKenzie and May 

2006), respond rapidly to mechanical treatment, but 

how rapidly? The prevalence of derangement syndrome 

in low back pain patients has been reported to be between 

75% and 80% (Hefford 2008). The prevalence of 

derangement syndrome in neck patients has been reported 

to be between 81% and 79% (Hefford 2008). 

The aim of this study was to track treatment duration for 

cervical and lumbar spine patients classified into the 

derangement syndrome, as described by McKenzie. 

Secondly, the aim was to evaluate whether this 

timeframe was sufficient for patients to improve. Data 

was collected during the development and pilot test period 

of a clinical quality of care database. 

Methods 

Inclusion criteria were: Symptoms related to lumbar or 

cervical spine and a completed patient questionnaire as 

distributed by a clinic secretary. Patients were excluded 

if they did not show up for the final visit or for some 

other reason did not fill in the self report questionnaire 

for clinical outcomes. During the pilot test the number of 

excluded patients was not tracked. 

Clinicians: All data was collected from the patients by 

three clinicians working in three different primary care 

clinics in Denmark. All three were Diplomaed in Mechanical 

Diagnosis & Therapy. Clinical experience ranged 

from 6 to 9 years. 

Treatment time: Patients filled in a questionnaire at the 

arrival at the clinic immediately prior to the initial evaluation. 

Another questionnaire was completed immediately 

after the last visit, before the patient left the clinic. The 

date was registered on both questionnaires and treatment 

duration was calculated. The decision to end the 

treatment was a mutual decision between the patient 

and clinician. Usually the decision to end treatment was 

made when the outcome was satisfactory, when it was 

apparent that no further improvement was possible or 

the patient decided that they were able to manage on 

their own. 

Outcome measures: Pain level was measured on a 0 

to 10 point numerical pain rating scale (NPRS), where 0 

points meant no pain and 10 points meant the worst 

imagineable pain. Patients were asked to score their 

maximum pain level for the two previous weeks. Functional 

level was scored on the Neck Disability Index 

(NDI) (Vernon and Mior 1991) for cervical patients and 

the 23-item Roland Morris Disability Questionnaire 

(RMDQ) (Roland and Morris 1983) for lumbar patients. 

Both the NDI and RMDQ are disease specific questionnaires 

related to disability by means of activities of daily 

living. The NDI relates to pain and leisure activities as 

well. The Danish translation of 




RMDQ has been validated scientifically (Albert et al 

2003). The NDI was translated into Danish by the 

author of this paper. This translation has not been validated. 

For self efficacy, a scale from the Acute Low 

Back Screening Questionnaire (Hurley et al 2001) was 

used. The scale is a numerical rank from 1 to 10 points. 

For the purpose of this project the results were transformed 

to a percentage. Thus, a score of 0% means the 

patients is completely unable to reduce their pain and 

100% means that they are able to abolish pain completely. 

Data analysis: Outcome measures were analysed and 

students t-test was used to calcutate p-values for statistical 

significance. 

Type of treatment 

All patients were evaluated and treated in accordance 

with the McKenzie method, including directional specific 

exercises, overpressure and mobilization as needed, 

posture correction, individually tailored exercises aiming 

to restore normal functional capacity and instruction in 

prophylactic strategies. 

Population characteristics 

Patient population demographic characteristics are 

shown in Table 1. For both cervical and lumbar patients 

the cohort was middle aged, mainly subacute to chronic, 

according to Québec Task Force definitions (QTF). 

According to QTF definitions, symptom duration from 0- 

7 days is labeled as acute, more than seven days and 

less than seven weeks as subacute and duration more 

than seven weeks is labeled as chronic. A minority of 

the patients were on sick leave at intake. 

Results 

Treatment duration for cervical derangements was 2.4 

weeks average; 95% confidence interval for cervical 

patients treatment duration was 1.7-3.0 weeks. Number 

of treatment sessions for cervical patients was 3.7 sessions 

on average; 95% confidence intervals was 3.3- 

4.2. 

Treatment duration for lumbar patients was 4.3 weeks 

average; 95% confidence interval for lumbar patients 

treatment duration was 3.2-5.3 weeks. Number of treatment 

sessions for lumbar patients was 4.3 sessions on 

average, 95% confidence intervals was 3.9. 

Table 1 - demographic characteristics 

Tables 2 and 3 show the cohort status related to pain, 

functional level and self efficacy at intake and discharge 

for cervical and lumbar patients respectively. All changes 

were statistically significant. 

Discussion 

A change of 30% for pain and functional scores has 

been suggested as a clinically meaningful improvement 

(Ostelo et al 2008). Another research group suggested 

that a 10.5 points change on NDI as the minimal detectable 

change (Pool et al 2007). For lumbar patients, a 

minimal change of 2.5 points on RMDQ is required 

(Kovacs et al 2007). Based on these criteria the changes 

in pain and functional scores were both clinically 

relevant and statistically significant. For the self efficacy 

measure used there has not been any report on cut-off 

point for a clinically relevant change. However, if the 

30% change rule is applied to this measure, patient self 

efficacy improved as well. 

The pain and functional outcomes found in this study 

was somewhat comparable to those reported for lumbar 

patients by Petersen (Petersen et al 2002) and Long 

(Long et al 2004) (Appendix A). In Petersen (Petersen 

et al 2002), a longer treatment duration of eight weeks 

was reported and a larger majority of patients were on 

sick leave at intake. Also, outcomes from this study were 

not reported for those patients classified into the derangement 

syndrome exclusively. In Long (Long et al 

2004), treatment duration was two weeks only, although 

a larger majority of patients were sick listed at intake in 

this study as well. Furthermore, our study was not a 

randomised controlled trial, as were the other two. These 

differences may explain the slight differences in 

measured outcomes between the three study populations. 

The results of this project has a value in relation to informing 

patients and setting realistic treatment goals. However, 

a considerable limitation to these results is the 

fact the amount of excluded patients were not tracked. 

Therefore, the ability to generalize the results are limited. 

Further to this point, no discrimination was made 

between patients with or without neurologic signs of root 

compression. For future projects, data would be even 

more valuable were it risk adjusted for clinical 

Cervical (n=26) Lumbar (n=66) 

Acute* 4 (15%) 8 (12%) 

Subacute* 13 (50%) 36 (55%) 

Chronic 9 (35%) 22 (33%) 

Age - average (range) 37.7 yrs (21-70) 40. 8 yrs (21-77) 

Women 21 (81%) 34 (52%) 

Sick listed at intake 4 (15%) 12 (18%) 


*Based on QTF definitions 




characteristics, such as, chronicity, neurological deficit, 

presence of deformity and symptom distribution. It 

would also be relevant to investigate the treatment duration 

of patients classified into other mechanical syndromes 

in comparison to the derangement syndrome. 

Conclusion 

In a cohort of patients classified into the derangement 

syndrome, treatment duration was on average 2.7 

weeks [95% CI: 1.7-3.0] for 26 cervical and 4.3 weeks 

[95% CI: 3.2-5.3] for 66 lumbar patients. The timeframe 

was sufficient for patients to gain a clinically relevant 

and statistically significant improvement as measured 

by a pain score, disease specific functional scores and 

self efficacy. Based on the 95% confidence intervals, 

there seems to be only a few weeks difference in treatment 

duration between the shortest and longest course. 

References 

Albert HB, Jensen AM et al (2003). "Kriterievalidering af 

Roland Morris-spørgeskemaet." Ugeskrift for læger 165 

(18): 1875-1880. 

Hefford C (2008). "McKenzie classification of mechanical 

spinal pain: profile of syndromes and directions of 

preference." Man Ther 13(1): 75-81. 

Hurley DA, Dusoir TE et al (2001). "How effective is the 

acute low back pain screening questionnaire for predicting 

1-year follow-up in patients with low back pain?" 

Clin J Pain 17(3): 256-63. 

Kovacs FM, Abraira V et al (2007). "Minimal Clinically 

Important Change for Pain Intensity and Disability in 

Patients With Nonspecific Low Back Pain." Spine 32 

(25): 2915-2920. 

Long A, Donelson R et al (2004). "Does it Matter Which 

Exercise? A Randomized Control Trial of Exercise for 

Low Back Pain." Spine 29(23): 2593-2602. 

Table 2 - Outcome measures cervical patients (n=26) 

Measure First visit Last visits P 

Pain 5.7 ± 1.7 3.4 ± 1.9



Appendix A - comparison of outcomes for lumbar patients 

Study Function Pain Population 

characteristics 

Petersen 

et al 

(2002) 

*) 

Long 

et al 

(2004) 

Melbye 

et al 

(2008) 

LBPRS 

Start average: 36.7% 

End average: 19.7% 

Ave reduction: 17% 

RMDQ24 

Start: 17.9 ± 5.7 

End: 11.4 ± 7.6 


RMDQ23 

Start average: 11.4 ± 5.0 

End average: 2.7 ± 3.3 


BOX6 

Start average: 19 of 60 

End average: 7 of 60 


BOX 

Back pain start: 5.9 ± 2.4 

Back pain end: 2.5 ± 2.0 

Leg pain start: 4.6 ± 2.5 

Leg pain end: 1.6 ± 1.8 

Ave reduction back pain: 34% 

Ave reduction leg pain: 30% 

BOX 

Start average: 6.1 ± 1.7 

End average: 2.6 ± 2.0 


Lumbar (n=128) 

Age: 34.5 yrs (range 23-52) 

Females: 62 of 132 (47%) 

Sick listed at intake: 42 of 132 (32%) 

Median symptom duration: 8 mos 

Lumbar (n=80) 

Age: 42.9 ± 9.6 yrs 


Acute: 14.1% 

Subacute: 39.7% 

Chronic: 46.2% 

QTF - criteriae 

Sick listed at intake: 32.5% 

Ave symptom duration: 13.7 wks 

Lumbar (n=66) 

Age: 40.8 yrs (range 21-77) 


Acute: 12% 

Subacute: 55% 

Chronic: 33% 

QTF - criteriae 

Sick listed at intake: 18% 

*) Patients in this study were not sub-grouped in relation to outcome measures 

RMDQ23: 23-item disease specific functional scale for low back patients in Danish 

RMDQ24: 24-item disease specific functional scale for low back patients 

BOX: 11-point box scale for worst pain 

BOX6: 11-point box scale for pain at the moment, worst pain and average pain. 

All three aspects were scores for back and leg pain respectively 


LBPRS: Low back pain rating scale 

Danish 15-item functional scale 

Results are expressed on a 0-100% (0% no difficulties and 100% highest score on difficulties on all items) 

Treatment 

duration 

and visits 

6 visits ave 

over 8 wks 

3-6 visits 

over 2 wks 

4.3 wks 

(95%CI: 3.2-5.3) 

4.3 visits 

(95%CI: 3.9-4.8) 






Abstract 

Segmental specificity of treatment has been recommended for optimising outcomes in manual therapy. This review 

wanted to examine some necessary related concepts: 

1) Is a posterior-anterior mobilisation specific to a motion segment? 

2) Is a specific mobilisation better than a standardised mobilisation? 

3) Is there evidence to support the reliability of determining specific problems? 

4) Is there evidence to support the validity of determining specific problems? 

Relevant studies were located by search of Medline and search of references from located studies. Over thirty studies 

were reviewed that addressed these different issues. A posterior-anterior mobilisation does not only affect the 

segmental level to which it is applied, but the local spine as a whole. There is no evidence to suggest that specific 

manual therapy techniques are superior to non-specific or standardised ones. It is unlikely that specific impairments, 

such as the comparable level, fixations or stiffness, can be reliably detected between therapists. The evidence regarding 

the link between such specific impairments and back pain is contradictory. Overall, the evidence suggests 

that manual therapy aimed at specific segments is neither possible nor necessary to optimise outcomes. 

Introduction 

Treatment specificity has been advocated by manual 

therapists. This means that forces provided by manual 

therapy techniques should be applied at specific spinal 

levels and in specific directions to be of optimal 

effectiveness. For instance, ‘significant comparable 

signs will be evident on palpation at the appropriate 

intervertebral level’ (Maitland 1986, page 73). 'Manual 

diagnosis can consistently and accurately determine 

the offending level in cases of spinal pain' in the 

cervical spine (Jull et al 1988). 'A patient's physiological 

movements may appear normal, yet the palpation tests 

for intervertebral movement will reveal joint 

signs' (Maitland et al 2005, page 150). However, the 

belief that manual therapy treatment should be directed 

at specific segmental levels, although it appears to be 

logical, has not been explicitly proven. The specificity of 

manual therapy techniques has been advocated, but is 

it necessary? 

If this assumption is true, then certain criteria should 

pertain. If specificity of manual therapy was important 

then it should be attainable and techniques should be 

shown to be able to target specific motion segments. 

Furthermore, it should be demonstrated that such 

specifically selected manual therapy procedures are 

more effective than standardised or randomly selected 

ones. A further underlying assumption is that treatment 

is aimed at specific impairments, which are diagnosed 

by the physical examination. For instance, that a 

hypomobile spinal segment, which is determined by 

palpation examination, requires mobilisation to improve 

the patient’s condition. This assumption requires 

reliable and valid methods to make the diagnosis. 

The aims of this review were to explore these issues by 

exploring the following questions: 

1. Is the effect of specific techniques actually 

specific? As posterior-anterior (PA) or extension 

mobilisations have been most commonly studied; 

this question will be addressed by understanding 

what PA mobilisations achieve physiologically. 

2. If clinicians can choose specific manual therapy 

techniques, are these more effective than stan 

dardised or randomly chosen techniques? 

3. Can specific impairments be diagnosed reliably? 

4. Is there a validated correlation between specific im 

pairments and spinal pain? 

Methods 

To find material for this review, Medline was searched 

up to June 2008 with the following terms combined with 

Boolean operators: manual therapy, mobilisation/ 

mobilization, manipulation, spine, effect, palpation, reliability, 

validity. As the results of this search were not 

very effective, as it missed papers that were already in 

the author’s library, particular emphasis was placed on 

hand searches. These included reference lists of all 

retained articles, and hand searching of relevant appropriate 

journals: Manual Therapy, Journal of Manipulative 

and Physiological Therapeutics, Journal of Manual 

and Manipulative Therapy, Clinical Biomechanics, 

Physiotherapy, Physical Therapy, and Australian Journal 

of Physiotherapy for the last two years. As the nature 

of study design varied considerably, a formal 

evaluation of study quality was not attempted. 

Results 

The effect of extension or posterior-anterior 

(PA) mobilisations 

Snodgrass et al (2006) conducted a systematic review 

into what forces were applied during a PA spinal mobilisation. 

They retrieved 20 papers investigating the quantitative 

measurement of applied force during a PA mobilisation, 

with most focusing on the lumbar spine. Techniques, 

measurement and reporting procedures were 

performed using a range of methodologies. When defined 

by magnitude, frequency, amplitude and displacement, 

PA mobilisations were found to be extremely variable 

among clinicians applying the same technique. For 

instance, when applying grade I mobilisations to the 

lumbar spine, average peak force varied from 10 to 50 

Newtons; grade II from 15 to 120 Newtons; grade III 

from 120 to 225 Newtons; grade IV 




from 90 to 240 Newtons. Although there is a general 

pattern of increasing force with higher grades, there is 

considerable overlap between grades and marked 

variation between practitioners applying the same 

grade. It is reasonable to conclude that grades of mobilisation 

applied by different therapists are not consistent. 

A number of studies have evaluated the effects of PA 

mobilisations in terms of motion achieved at the lumbar 

and cervical spines (Table 1). These consistently show 

that the effect is not localised to the single segment 

where the force is applied, but affects the whole spine. 

In general, forces applied at lower lumbar segments 

produce extension of the whole lumbar spine; whereas 

forces applied at upper lumbar segments produce flexion 

at the lower segments. Clearly, it cannot be stated 

that the effect of a PA mobilisation is localised to the 

vertebral motion to which it is applied. However, in contrast 

to these findings, two studies actually found that 

there was no or minimal translation or intervertebral motion 

in response to PA mobilisations to the cervical 

spine in controls and in subjects with neck pain 

(McGregor et al 2001, 2005). PA mobilisations were 

simply shown to cause soft tissue compression. 

Another way researchers have looked at this issue is by 

analysing the source of cavitation sounds compared to 

the segmental level supposedly being manipulated 

(Beffa and Matthews 2004, Ross et al 2004). No meaningful 

relationship was found between the segment being 

manipulated and the source of cavitation sound. So, 

the sound of the manipulation ‘pop’ is neither related to 

therapeutic effectiveness (Flynn et al 2003, 2006) nor to 

the segment being manipulated (Beffa and Matthews 

2004, Ross et al 2004). 

Does specificity improve outcomes? 

The next question is whether specific techniques chosen 

by therapists are more effective than standardised 

techniques over which the therapist has no choice? Two 

randomised controlled trials (Chiradejnant et al 2003, 

Haas et al 2003), and a meta-analysis (Kent et al 2005) 

have examined the effect of clinician choice in manual 

therapy treatment; the trials were not included in the 

meta-analysis. 

Chiradejnant et al (2003) conducted a randomised controlled 

trial in which 140 patients with non-specific low 

back pain were randomly allocated to receive either the 

therapist-chosen mobilisation technique or a randomly 

assigned mobilisation technique. Mobilisation techniques 

in both groups were applied at the segmental 

level and at the grade nominated by the examining 

therapist; only the technique was selected by the therapist 

or randomly applied. Follow-up measures were 

taken immediately after the interventions. Both interventions 

had an immediate effect in relieving back pain, but 

the choice of mobilisation had no effect on any of the 

outcomes 


investigated. Mobilisations applied to the lower lumbar 

spine had a greater effect than those applied to the upper 

lumbar spine, but the specific technique used 

seemed unimportant (Chiradejnant et al 2003). 

Haas et al (2003) randomly assigned a group of patients 

with neck pain to either a segmentally targeted 

manipulation chosen by the examining clinician according 

to endplay restriction or to a randomly chosen manipulation. 

There were significant improvements in both 

groups short-term, but no significant differences between 

groups. 

Kent et al (2005) included manual therapy interventions 

selected at the discretion of the treating clinician compared 

to no, sham or other interventions in ten trials. 

When trials using manual therapy were compared to no 

or sham intervention there was a significant difference 

in effect size for short-term activity limitation in favour of 

the no/sham intervention. There were no significant differences 

in any other short or long-term outcomes. 

When chosen manual therapy techniques were compared 

to other interventions there were significant differences 

in short-term pain and activity limitations in favour 

of the other interventions, but no significant differences 

long-term. The authors were cautious about making 

emphatic conclusions due to the limited available data. 

However, they were clear that the present data does not 

support the premise that therapist-selected manual therapy 

techniques have a greater therapeutic effect than 

no, sham or other treatments. All of the pooled estimates 

of effect size favoured the groups where clinicians 

did not have choice over their manual therapy 

technique. 

The evidence from these three papers makes for sobering 

reading for believers in specificity of manual therapy 

in non-selected non-specific low back pain or neck pain 

patients. In essence, the data suggests that not only do 

specific techniques appear to confer no additional benefits 

to randomly chosen ones (Chiradejnant et al 2003, 

Has et al 2003), but also that treatment effect size appears 

to be greater in no/sham or other treatment compared 

to selected manual therapy procedures (Kent et 

al 2005). It should be noted, as well, that in the recent 

development of clinical prediction rules for identifying 

successful responders to manipulation (Flynn et al 

2002, Childs et al 2004), the technique used was a 

standardised, routine manipulation purportedly aimed at 

the sacro-iliac joint and not related to examination findings, 

but still successful. 

Reliability of palpation findings 

Although there are a few studies that suggest that palpation 

can be reliable, for instance in detection of congenital 

fusion in the cervical spine (Humphreys et al 

2004), the overwhelming weight of evidence points in 

the opposite direction. A number of recent systematic 

reviews have examined the reliability and validity of spinal 

examination procedures (Hestboek and Leboeuf- 




Yde 2000, van der Wurff et al 2000a, 2000b, Seffinger 

et al 2004, van Trijffel et al 2005, Hollerwoger 2006, 

May et al 2006, Stochkendahl et al 2006). A consistent 

finding has been the poor reliability of palpation based 

assessment, compared to the moderate reliability of 

some examination procedures based on symptom response 

(Table 2). From the available evidence, there is 

little doubt that decisions based upon palpation findings 

are of dubious consistency and, therefore, do not make 

a reliable basis upon which to base management decisions. 

Two reviews (May et al 2006, Stochkendahl et al 

2006) produced levels of evidence based on the quality 

of the literature, with the following conclusions. There 

was conflicting evidence for identifying spinal level, 

muscle spasm and instability; and moderate evidence of 

low reliability of passive accessory and passive physiological 

movements, the comparable level, and 

‘fixations’ (May et al 2006). There was strong evidence 

for unacceptable reliability for motion palpation (mean 

kappa 0.17) and soft-tissue changes (mean kappa 

0.03); and conflicting evidence for static palpation and 

global assessment (Stochkendahl et al 2006). Consistently, 

pain responses have shown stronger reliability 

than findings made by palpation (Table 2). 

Validity of palpation findings 

We might also ask if PA mobilisations are valid at detecting 

movement impairments or diagnostic classifications. 

Manual PA tests were unable to detect the most 

and least mobile segments as identified by the ‘gold 

standard’ of MRI investigation (Landel et al 2008). Previously, 

it was suggested that an experienced manual 

therapist could accurately identify symptomatic vertebral 

segments with neck pain, with 100% sensitivity and 

100% specificity (Jull et al 1988). However, a replication 

study using contemporary criteria demonstrated poor 

specificity (47%), but reasonable sensitivity (89%) in 

identifying cervical zygapophyseal joint pain (King et al 

2007). In general then, it must be concluded that palpation 

is neither reliable nor valid at identifying pathology. 

Is there a correlation between specific impairments 

and spinal pain? 

Regarding the relationship between impairments and 

back pain, a number of studies have considered this 

with a variety of methodologies (Table 3). Generally, 

findings have been contradictory. Some studies have 

identified an association between hypermobility (Kulig et 

al 2007, Abbott et al 2006) or hypomobility (Lundberg 

and Gerdle 2000, McGregor et al 2002, Abbott and Mercer 

2003, Abbott et al 2005, 2006) and back pain. However, 

the findings have not been consistently found and 

other studies have found no association between hypermobility 

(Lundberg and Gerdle 2000, Beneck et al 2005, 

Abbott et al 2006) or hypomobility (Beneck et al 2005, 

Kulig et al 2007, Owens et al 2007) and back pain or 

the most painful segment. 


Discussion 

In response to the questions posed above, the following 

answers arise from the material reviewed in this article. 

A PA mobilisation does not only affect the segmental 

level at which it is applied, but the local spine as a 

whole. However, there is also some evidence that they 

have no effect on inter-segmental motion at all, they 

simply cause soft tissue compression. Specific manual 

therapy techniques are not superior to non-specific or 

standardised ones. It is unlikely that specific impairments, 

such as the comparable level, fixations or 

amount of accessory movement, can be reliably detected 

between therapists. Nor does it appear that palpation 

findings offer valid methods of identifying impairment 

or pathology. The evidence regarding the link between 

such specific impairments and back pain is contradictory. 

Only the evidence regarding the link between specific 

impairments and back pain was contradictory and might 

alter in light of new evidence, especially as the technology 

in this area is changing and developing. In the past, 

however, the cut-off points for diagnosis of hypomobility 

and hypermobility have been largely arbitrary, and it is 

clear that there is wide variability and a wide range of 

translation in asymptomatic individuals. Earlier studies 

reported a high prevalence of instability in 23-69% of 

chronic LBP subjects from flexion-extension radiographs 

(Abbott et al 2006). But these prevalence rates 

were deemed to be suspect as arbitrary definitions of 

'abnormal' would have lead to high rates of falsepositives 

(Abbott et al 2006). Abbott et al (2006) used a 

statistically defensible method, but their method only 

allowed for statistical not clinical identification. They reported 

hypomobility rates of 18-35% and hypermobility/ 

instability rates of 5-32% depending on the method 

used and direction of abnormality (rotation or translation). 

The overall contradictory findings suggest there may 

well be an association between these impairments and 

LBP, but this does not mean there is a causal link, nor 

that treatment directed at these impairments will lead to 

improved levels of pain and function. To prove a causal 

link, a prospective cohort study is necessary and such a 

study was not located. As it is it cannot be known if hypomobility 

and hypermobility cause back pain, are consequences 

of it, or are incidental findings rather than 

‘impairments’, within the range of normal and nothing to 

with spinal pain. Studies that have sought to address 

the link between such impairments and treatment have 

produced contradictory findings. Fritz et al (2005) categorised 

patients as having either hypomobility (71%) or 

hypermobility (11.5%) and then randomised them to 

either spinal manipulation or stabilisation exercises. Patients 

with hypomobility who received manipulation and 

those with hypermobility who received stabilisation exercises 

showed greater improvements than those 

treated with the other 




intervention, which logically would be expected if these 

impairments were relevant to symptoms. Failure rates in 

patients with hypomobility were 26% in those receiving 

manipulation and 74% in those receiving stabilisation 

exercises. Failure rates in patients with hypermobility 

were 83% in those receiving manipulation and 22% in 

those receiving stabilisation exercises. So, this study 

would suggest a link between these impairments and 

the type of treatment; however a similar study failed to 

demonstrate such a link. Ferreira et al (2008) also assessed 

spinal stiffness before and after randomisation 

and treatment by manipulation, stabilisation exercises 

or a general exercise programme. All groups showed a 

significant decrease in stiffness following treatment, but 

the decrease was not dependent on the treatment. 

There was a significant relationship (P=0.02) between 

changes in stiffness and pain and function, but the correlation 

was weak (r=0.18 and -0.28 respectively). So, 

this study does not support a link between impairment 

and specific treatment – if there was such a link it might 

be expected that manipulation would produce the most 

significant decreases in stiffness. 

The nature of this review has meant that it has included 

a large variety of different study methodologies, and as 

mentioned in the introduction, the initial search was of 

limited value in selecting appropriate evidence. Most 

papers were found from reference lists of the included 

papers that were already in the author's library. It is difficult 

to know if other relevant evidence has been missed; 

however as most of the conclusions were generally consistent 

from a number of sources, any fundamental 

change in the study conclusions would be unlikely in the 

face of new material. But the difficulty of knowing if this 

review has been truly comprehensive is a significant 

weakness. Likewise, the review included multiple study 

designs, but has not been able to review study quality 

nor to determine the optimal study methods for addressing 

these questions. 

Conclusions 

From the evidence reviewed here, and taking into account 

the fact that some evidence is contradictory or 

that some additional evidence may not have been located, 

some conclusions can be drawn. Mobilisation 

forces are not consistently applied by different therapists. 

Although there is increasing force with increasing 

grades there is considerable overlap between the different 

grades. The effect of a PA mobilisation is not limited 

to the segment to which it is applied, but affects the 

whole local spine. Some evidence suggests that PA 

mobilisation only produces soft tissue compression and 

does not affect spinal motion at all. Regarding specific 

manual therapy techniques, these appear to confer no 

additional benefit to randomly chosen ones, but also the 

treatment effect size appears to be less in selected 

manual therapy procedures compared to no, sham or 


other treatments. Studies are consistent in demonstrating 

the poor reliability of findings based on palpation, so 

if these are being used to make management decisions 

it is clear that these are based on inconsistent decisions 

made by therapists. There is contradictory evidence 

about the validity of manual therapy examination to determine 

impairment or pathology. There is contradictory 

evidence relating impairments and back pain, and contradictory 

evidence relating specific impairments to specific 

treatment interventions. 

References 

Abbott JH, Mercer SR (2003). Lumbar segmental hypomobility: 

criterion-related validity of clinical examination items 

(a pilot study). NZ J Physio 31.3-9. 

Abbott JH, McCane B, Herbison P, Moginie G, Chapple C, 

Hogarty T (2003). Lumbar segmental instability: a criterionrelated 

validity study of manual therapy assessment. BMC 

Musculoskeletal Dis 6:56 http:// 

www.biomedcentral.com/1471-2474-6-56 

Abbott JH, Fritz JM, McCane B, Shultz B, Herbison P, Lyons 

B, Stefanko G, Walsh RM (2006). Lumbar segmental 

mobility disorders: comparison of two methods of defining 

abnormal displacement kinematics in a cohort of patients 

with non-specific mechanical low back pain. BMC Musculoskeletal 

Dis 7.45 http://www.biomedcentral.com/1471- 

2474/7/45 

Beffa R, Matthews R. Does the adjustment cavitate the 

targeted joint? An investigation into the location of cavitation 

sounds. Journal of Manipulative and Physiological 

Therapeutics 2004;27:e2. 

Beneck GJ, Kulig K, Landel RF, Powers CM. The relationship 

between lumbar segmental motion and pain response 

produced by a posterior-to-anterior force in persons with 

non-specific low back pain. Journal of Orthopaedic and 

Sports Physical Therapy. 2005;35:203-209. 

Caling B, Lee M. Effect of direction of applied mobilization 

force on the posteroanterior response in the lumbar spine. 

Journal of Manipulative and Physiological Therapeutics 

2001;24:71-78. 


Majkowski GR, Delitto A (2004). A clinical prediction rule to 

identify patients with low back pain most likely to benefit 

from spinal manipulations: a validation study. Ann Intern 

Med 141:920-928. 

Chiradejnant A, Maher CG, Latimer J, Stepkovitch N. Efficacy 

of “therapist-selected” versus “randomly selected” 

mobilisation techniques for the treatment of low back pain: 

A randomised controlled trail. Australian Journal of Physiotherapy 

2003;49:233-241. 

Ferreira ML, Ferreira PH, Latimer J, Herbert RD, Maher C, 

Refshauge K (2008). Relationship between spinal stiffness 

and outcome in patients with chronic low back pain. Manual 

Therapy (In Press). 

Flynn T, Fritz J, Whitman J, Wainner R, Magel J, Rendeiro 

D (2002). A clinical prediction rule for classifying patients 




with low back pain who demonstrate short-term improvement 

with spinal manipulation. Spine 27:2835-2843. 

Flynn TW, Fritz JM, Wainner RS, Whitman JM (2003). The 

audible pop is not necessary for successful spinal highvelocity 

thrust manipulation in individuals with low back 

pain. Arch Phys Med Rehabil 84;1057-1060. 

Flynn TW, Childs JD, Fritz JM (2006). The audible pop 

from high-velocity thrust manipulation and outcome in individuals 

with low back pain. J Manip Physiol Ther 29.40-45. 

Fritz JM, Whitman JM, Childs JD (2005). Lumbar spine 

segmental mobility assessment: an examination of validity 

for determining intervention strategies in patients with low 

back pain. Arch Phys Med Rehab 86.1745-1752. 

Haas M, Groupp E, Panzer D Partna L, Lumsden S, Aickin 

M. Efficacy of cervical endplay assessment as an indicator 

for spinal manipulation. Spine 2003;28:1091-1096. 

Hestboek L, Leboeuf-Yde C (2000): Are chiropractic test 

for the lumbo-pelvic spine reliable and valid? A systematic 

critical literature review. Journal Manipulative Physiological 

Therapeutics. 23:258-275. 

Hollerwoger D (2006). Methodological quality and outcomes 

of studies addressing manual cervical spine examinations: 

A review. Manual Therapy 11.93-98. 

Humphreys BK, Delahaye M, Peterson CK (2004). An investigation 

into the validity of cervical spine motion palpation 

using subjects with congenital block vertebrae as a 

'gold standard'. BMC Musculoskeletal Disorders 5.19 http:// 

www.biomedcentral.com/1471-2474/5/19 

Iguchi T, Kanemura A, Kasahara K, Kurihara A, Doita M, 

Yoshiya S (2003). Age distribution of three radiological 

factors for lumbar instability: probable aging process of the 

instability with disc degeneration. Spine 28.2628-2633. 

Jull G, Bogduk N, Marsland A (1988). The accuracy of 

manual diagnosis for cervical zygapophyseal joint pain 

syndrome. Med J Aus 148.233-236. 

Kent P, Marks D, Pearson W, Keating J. Does clinician 

treatment choice improve the outcomes of manual therapy 

for non-specific low back pain? A metaanlysis. Journal of 

Manipulative and Physiological Therapeutics 2005;28:312- 

322. 

King W, Lau P, Lees R, Bogduk N (2007). The validity of 

manual examination in assessing patients with neck pain. 

Spine J 7.22-26. 

Kulig K, Landel R, Powers CM. Assessment of lumbar 

spine kinematics using dynamic MRI: A proposed mechanism 

of sagittal plane motion induced by manual posteriorto-anterior 

mobilization. Journal of Orthopaedic and Sports 

Physical Therapy. 2004;34:57-64. 

Kulig K, Powers CM, Landel RF, Chen H, Fredericson M, 

Guillet M, Butts K. Segmental lumbar mobility inindividuals 

with low back pain: in vivo assessment during manual and 

self-imposed motion using dynamic MRI. BMC Musculoskeletal 

Dis 20078:8. 

http://www.biomedcentral.com/1471-2474-8-8 


Landel R, Kulig K, Fredericson M, Li B, Powers CM. Intertester 

reliability and validity of motion assessments during 

lumbar spine accessory motion testing. Physical Therapy 

2008;8843-49. 

Lee R, Evans J. Towards a better understanding of spinal 

mobilisation. Physiotherapy 1994;80:68-73. 

Lee R, Evans J. An in vivo study of the intervertebral 

movements produced by posteroanterior mobilization. 

Clinical Biomechanics 1997;12:400-408. 

Lee RYW, McGregor AH, Bull AMJ, Wragg P. Dynamic 

response of the cervical spine to posteroanterior mobilisations. 

Clinical Biomechanics 2005;20:228-231. 

Lundberg G, Gerdle B. Correlations between joint and spinal 

mobility, spinal configuration, segmental mobility, segmental 

pain. Symptoms and disabilities in female health 

care personnel. Scandinavian Journal Rehabilitation Medicine 

2000;32:124-133. 

Maitland G (1986). Vertebral Manipulation (5th Edition). 

Butterworths, London. 

Maitland G, Hengeveld E, Banks K, English K (2005). Vertebral 

Manipulation (7th Edition). Elsevier, Edinburgh. 

May S, Littlewood C, Bishop A (2006). Reliability of procedures 

used in the physical examination of non-specific low 

back pain: a systematic review. Aus J Physio 52.91-102. 

McGregor AH, Wragg P, Gedroyc W (2001). Can interventional 

MRI provide an insight into the mechanics 0f a posterior-anterior 

mobilisation? Clin Biomechanics 16.926- 

929. 

McGregor A, Anderton L, Gedroyc W. The assessment of 

intersegmental motion and pelvic tilt in elite oarsmen. 

Medicine and Science in Sports & Exercise. 2002;34:1143- 

1149. 

McGregor AH, Wragg P, Bull AMJ, Gedroyc W (2005). 

Cervical spine mobilisations in subjects with chronic neck 

problems: an interventional MRI study. J Back Musculo 

Rehab 18.21-28. 

Owens EF, DeVocht JW, Gudavalli MR, Wilder DG, 

Meeker WC. Comparison to posteroanterior spinal stiffness 

measures to clinical and demographic findings at 

baseline in patients enrolled in a clinical study of spinal 

manipulation for low back pain. Journal of Manipulative 

and Physiological Therapeutics 2007;30:493-500. 

Powers CM, Kulig K, Harrison J, Bergman G. Segmental 

mobility of the lumbar spine during a posterior to anterior 

mobilization: assessment using dynamic MRI. Clinical Biomechanics 

2003;18:80-83. 

Ross JK, Bereznick DE, McGill SM. Determining cavitation 

location during lumbar and thoracic spinal manipulation. Is 

spinal manipulation accurate and specific? Spine 

2004;29:142-1457. 

Seffinger MA, Najm WI, Mishra SI et al (2004): Reliability 

of spinal palpation for diagnosis of back and neck pain. A 

systematic review of the literature. Spine 29:E413-E425. 




Snodgrass SJ, Rivett DA, RobertsonVJ. Manual forces 

applied during posterior-to-anterior spinal mobilization: A 

review of the evidence. Journal of Manipulative and 

Physiological Therapeutics 2006;29:316-329. 

Stochkendahl MJ, Christensen HW, Hartvigsen J, Vach 

W et al (2006). Manual examination of the spine: a systematic 

critical literature review of reproducibility. Journal 

of Manipulative and Physiological Therapeutics 29.475- 

485. 

Van der Wurff P, Hagmeijer RHM, Meyne W (2000a): 

Clinical tests of the sacroiliac joint. A systematic methodological 

review. Part 1: Reliability. Manual Therapy 

5:30-36. 

Table 1. Effects of extension or posterior-anterior (PA) mobilizations 


Van der Wurff P, Meyne W, Hagmeijer RHM (2000b). Clinical 

tests of the sacroiliac joint. A systematic methodological 

review. Part 2: Validity. Manual Therapy 5.89-96. 

Van Trijffel E, Anderegg Q, Bossuyt PMM, Lucas C (2005). 

Inter-examiner reliability of passive assessment of intervertebral 

motion in the cervical and lumbar spine: a systematic 

review. Manual Therapy 10: 256-269. 

Ref Methods Subjects Findings Conclusions 

Lee and 

Evans 

1994 

Lee and 

Evans 

1997 

Caling 

and Lee 

2001 

McGregor 

et al 2001 

Powers 

et al 2003 

Kulig 

et al 2004 

Lee 

et al 2005 

McGregor 

et al 2005 

Landel 

et al 2008 

Review of work 

using a biomechanical 

model - 

lumbar 

X-ray using loading 

frame to apply 

static force at L4 

Assessment of 

lumbar stiffness 

using a simulator 

PA within MRI to 

cervical spine 

Manual application 

of lumbar PA 

within MRI 


of lumbar PA 

within MRI 


of PA within MRI 

machine at C5 

PA within MRI to 

cervical spine 

Is lumbar PA 

exam reliable between 

raters and 

valid compared to 

MRI? 

Model Maximal extension force at segment 

where force was applied, 

but affects all segments; posterior 

shear above and anterior 

shear below 

12 healthy 

volunteers 

24 healthy 

volunteers 

5 healthy 

volunteers 

11 healthy 

volunteers 

20 healthy 

volunteers 

19 healthy 

volunteers 

5 subjects 

with NP 

29 subjects 

with LBP 

Lumbar segments extended, L5 

-S1 flexed; L1/2 to L3/4 translated 

posteriorly, 

L5-S1 anteriorly 

Stiffness varied with angle of 

force when applied at L3, but 

not at L5 

No significant intersegmental 

motion or translation with grade 

I or IV, but significant soft tissue 

compression 

Force applied at L3/4 to L5-S1 

caused extension at adjacent 2 

segments. Force at L1/2 and 

L2/3 caused flexion at lower 

segments 

Force applied at L3/4 to L5-S1 

caused extension at adjacent 2 

segments. Force at L1/2 and 

L2/3 caused flexion at lower 

segments 

Produced extension of upper 

motion segments and flexion at 

C7-T1, middle segments were 

inconsistent 

No significant intersegmental 

motion or translation with grade 

I or IV, but significant soft tissue 

compression 

Overall reliability kappa 0.71; 

poor for most mobile segment 

(kappa 0.29); 

Validity was poor for least and 

most mobile segments (kappa 

0.04 and 0.00) 

Amount of local movement very 

small, highly improbable can be felt; 

whereas bending of the whole 

spine is much greater 

Limited mobility, unlikely that movement 

can be assessed with reliability 

Not clear if clinicians could detect 

changes in stiffness of 5-10% 

PA to cervical spine have no effect 

on intervertebral motion 

PA caused extension locally, but 

affect on lordosis depended on 

where force was applied 

PA at single segment caused motion 

of all segments; direction depended 

on where force was applied 

PA at single segment caused motion 

of all segments and overall 

increase in lordosis 

PA to cervical spine have no effect 

on intervertebral motion 

PA could not detect least and most 

mobile segments compared to ‘gold 

standard’ of MRI 


Reference Review topic N Conclusions regarding palpation 

Hestboek & 

Leboeuf-Yde 

2000 

Van der Wurff 

et al 2000a 

Van der Wurff 

et al 2000b 

Seffinger 

et al 2004 

Van Trijffel 

et al 2005 

Hollerwoger 

2006 

May 

et al 2006 

Stochkendahl 

et al 2006 



Table 2. Conclusions about palpation findings from systematic reviews 

Chiropractic tests for lumbar 

and pelvis to determine 

need for manipulation – 

reliability and validity 

Tests for sacroiliac joint - 

reliability 

Tests for sacroiliac joint - 

validity 

Spinal palpation for back 

and neck pain - reliability 

Passive assessment of 

intervertebral motion in 

cervical and lumbar - reliability 

Manual cervical spine examination 

– reliability and 

validity 

Physical exam procedures 

lumbar spine - reliability 

Manual exam of spine - 

reliability 


30 Only tests for palpation for pain had acceptable results. Motion palpation 

for lumbar spine showed poor reliability, but might be valid. Motion 

palpation of pelvis seemed to be slightly reliable, but was not 

valid. Presence of a manipulative lesion remains hypothetical 

11 Mobility tests showed consistently poor reliability; whereas pain provocation 

tests reliability was contradictory with good reliability in some 

studies 

11 No evidence to support the validity of mobility or pain provocation 

tests in identifying SIJ dysfunction 

49 12 highest quality studies found pain provocation, motion and landmark 

tests to have acceptable reliability (kappa > 0.40). The majority 

of palpatory diagnostic tests demonstrated low reliability. Pain provocation 

tests are most reliable 

19 Overall reliability for both areas of spine was poor to fair; range poor 

to substantial. Assessment of motion at C1/2 and C2/3 consistently 

fair reliability 

15 Detection of segmental cervical ‘dysfunction’ on manual assessment 

alone is questionable 

48 Conflicting evidence for identifying spinal level, muscle spasm and 

instability. Moderate evidence of low reliability of passive accessory 

and passive physiological movements, the comparable level, and 

‘fixations’. 

48 Strong evidence for reliability of osseous pain (mean kappa 0.53) and 

soft tissue pain (mean kappa 0.42). Strong evidence for unacceptable 

reliability for motion palpation (mean kappa 0.17) and soft-tissue 

changes (mean kappa 0.03). Conflicting evidence for static palpation 

and global assessment 


Table 3. Correlation between lumbar impairments and spine pain 

Reference Aims / Methods Subjects Findings 

Lundberg and 

Gerdle 2000 

McGregor 

et al 2002 

Abbott and 

Mercer 2003 

Iguchi 

et al 2003 

Abbott 

et al 2005 

Beneck 

et al 2005 

Abbott 

et al 2006 

Kulig 

et al 2007 

Owens 

et al 2007 



Do manual segmental 

mobility tests correlate 

with disability? 

Does mobility obtained 

by MRI predict LBP 

history? 

Is hypomobility associated 

with LBP / flex-ext 

x-rays? 

Prevalence of radiological 

factors for instability 

in chronic LBP population? 

Correlation between 

PAIVM and PPIVM / 

flex-ext x-rays 

Any association between 

MRI-segmental 

motion / pain response 

during a PA? 

Prevalence of lumbar 

segmental rigidity and 

instability using flex-ext 

x-rays using different 

criteria? 

Is lumbar segmental 

mobility with PA on MRI 

altered in LBP subjects? 

Relationship between 

PA stiffness, as measured 

by electronic sensors, 

and clinical factors? 

607 women;48% LBP 

during last week 

20 men; 4 with present 

LBP, 7 with h/o LBP 

12 LBP patients versus 

20 controls 

880 LBP patients with 

flex-ext x-rays 

138 patients with episodic 

or chronic LBP 

35 LBP subjects: 27 

reported most painful 

segment 

138 subjects with 

chronic LBP compared 

to 30 pain-free sample 

45 individuals with LBP, 

20 without LBP 

Hypomobility // higher disability (P

Review of Studies 


Shoulder pathology cannot be diagnosed by 

clinical examination 

Just after the publication of another systematic review 

about the lack of evidence to support shoulder 

diagnosis through clinical examination (Hegedus et al 

2008), which was reviewed in the last International 

Journal of Mechanical Diagnosis and Therapy, comes 

another review that draws the same conclusions. 

Hughes PC, Taylor NF, Green RA (2008). Most 

clinical tests cannot accurately diagnose rotator 

cuff pathology: a systematic review. Aus J Physio 

54.159-170. 

Objective 

To address the research question, do clinical tests 

accurately diagnose rotator cuff pathology? 

Design 

Systematic review 

Methods 

A search was made of electronic data bases up to April 

2007. With the search strategy provided in an 

appendix, this was supplemented by a hand search of 

the included references. Title and abstracts were 

independently searched by three reviewers; included 

studies had to be full reports, involve participants with 

shoulder pain who underwent clinical diagnostic testing 

for rotator cuff pathology and a reference standard 

(operative finding or MRI), and with sensitivity and 

specificity reported. Study quality was independently 

determined by three reviewers against five explicit 

criteria. One reviewer extracted data from the studies 

which were checked by a second reviewer. Meta 

analysis was to be used if homogeneity existed across 

studies. 

Main outcome measures 

Diagnostic accuracy was determined by positive and 

negative likelihood ratios (+LR –LR respectively), and if 

these were not reported they were calculated from 

sensitivity and specificity. Likelihood ratios were 

interpreted in the following ways: 

+LR > 10 Significant shift in diagnostic probability 

-LR < 0.1 

+LR 5-10 Moderate shift in diagnostic probability 

-LR 0.1-0.2 

+LR 2-5 Small shift in diagnostic probability 

-LR 0.2-0.5 

+LR 1-2 Rarely important shifts in diagnostic 

probability 

-LR 0.5-1.0 

LR close to 1 Irrelevant shifts in diagnostic probability 


Main results 

Thirteen studies were included; four evaluated patients 

with subacromial syndrome, seven with supraspinatus 

problems; reference standards were MRI (four studies), 

arthroscopy (five studies), operative findings (three 

studies) or either of latter. In only two studies was there 

blinded measurement of the clinical tests and reference 

standard, and in five studies only those who received 

the reference standard were included. There were 89 

evaluations of 14 different clinical tests (Table 1), out of 

which only seven and three had +LR > 10 or –LR 0.2 71 4 6 

0.1 - 0.2 5 0 0 

10 (combined Hawkins/painful arc/infraspinatus 

test, Napoleon, Lift-off, belly press and drop-arm test), 

and 2 had –LR


diagnosis after a test, but only seven evaluations had 

this, and for each of these tests other studies did not 

replicate this finding or there were no other evaluations 

of that test. For instance, palpation had +LR of 29.91 in 

one study, but only 3.64 in the only other study 

evaluating palpation. Drop arm test had +LR of ∞ in 

one study, but +LR of 5.0 or less in four other studies. 

For subscapularis, the belly-press test and Napoleon 

test had +LR of 19.05 and 11.90 respectively, but only 

from single studies; and lift-off test had +LR of ∞, but 

+LR less than two in two other studies. The same was 

true of tests showing moderate shift in the probability of 

making a positive diagnosis after a test, with +LR 

between 5 and 10; either this was from a single study 

or repeat studies gave much smaller +LR values. 

Likewise, it is standard to use –LR of less than 0.1 for a 

significant shift in the probability of ruling out a 

diagnosis after a test, but only two tests showed this 

level. One study found a –LR of 0.03 for the empty can 

test using pain or weakness, but the same study using 

different criteria and five other studies, making a total of 

20 other evaluations, found –LR levels between 0.13 

and 1.00. One evaluation found –LR of 0.00 for the 

Hawkins-Kennedy test, but six other evaluations 

produced –LR of 0.16 to 0.65. Thus, even tests which 

sometimes performed well could not be used with any 

confidence as they lacked consistency or promising 

initial results had not been replicated. 

The results are not surprising and simply reflect 

previous reviews. In a review of 45 studies evaluating 

tests for all shoulder pathology, most tests performed 

poorly or if they did well this had not been replicated in 

further studies (Hegedus et al 2008). Dinnes et al 

(2003) had previously systematically reviewed 10 

studies evaluating tests for rotator cuff problems and 

found poor specificity for the overall clinical 

examination. Similarly, Mirkovic et al (2005) found that 

tests for SLAP lesions that did well in one study were 

not replicated in additional studies or no additional 

studies existed. In a more recent review, the same 

situation was still found, with the majority of articles 

reporting better rates of accuracy being of low quality 

and were contradicted by other studies (Dessaur and 

Margarey 2008). 

One reason for poor diagnostic accuracy could be the 

lack of anatomical basis for most shoulder tests found 

in one review (Green et al 2008). 

It is clearly time to move away from the phantom of 

attaching diagnostic labels to patients with shoulder 

pain; it is not possible to do this with any accuracy and 


such attempts are merely fraudulent. One review 

looked at the diagnostic labels used in clinical trials 

involving patients with shoulder pain and found no 

consistent uniform definition for any of the commonly 

used labels (Schelligerhout et al 2008). ‘The labels 

unintentionally seem to result in a Babylonian 

confusion of tongues and seem to be of little benefit for 

those with shoulder pain.’ These authors strongly 

recommend the abandonment of these diagnostic 

labels, for which there is a lack of unambiguous 

definitions, poor levels of sensitivity and specificity and 

poor reliability of diagnostic tests. Instead, they suggest 

the use of sub-groups of non-specific shoulder pain 

based on reliable clinical characteristics. Hopefully 

readers should be familiar with a non-specific subgrouping 

system based on clinical signs and symptoms 

already! 

References 

Dessaur WA, Magarey ME (2008). Diagnostic accuracy 

of clinical tests for superior labral anterior posterior 

lesions: a systematic review. J Orth Sports Phys Ther 

38.341- 352. 

Dinnes J, Loveman E, McIntyre L, Waugh N (2003). 

The effectiveness of diagnostic tests for the 

assessment of shoulder pain due to soft tissue 

disorders: a systematic review. Health Technology 

Assessment 2003;7:29. 

Green R, Shanley K, Taylor NF, Perrott M (2008). The 

anatomical basis for clinical tests assessing 

musculoskeletal function of the shoulder. Phys Ther 

Reviews 13.17-24. 

Hegedus EJ, Goode A, Campbell S, Morin A, 

Tamaddoni M, Moorman CT, Cook C (2008). Physical 

examination tests of the shoulder: a systematic review 

with meta-analysis of individual tests. Br J Sports Med 

42.80-92. 

Mirkovic M, Green R, Taylor N, Perrott M (2005). 

Accuracy of clinical tests to diagnose superior labral 

anterior and posterior (SLAP) lesions. Phys Ther Rev 

10.5-14. 

Schellingerhout JM, Verhagen AP, Thomas S, Koes 

BW (2008). Lack of uniformity in diagnostic labelling of 

shoulder pain: time for a different approach. Manual 

Therapy (In Press). 

�� 




Identification of ‘specific’ back pain unlikely on present 

evidence. 

Hancock MJ, Maher CG, Latimer J, Spindler MF, 

McAuley JH, Laslette M, Bogduk N (2007). Systematic 

review of tests to identify the disc, SIJ or facet 

joint as the source of low back pain. Eur Spine J 

DOI 10.1007/s00586-007-0391-1 

Objective 

To determine diagnostic accuracy of clinical tests to 

identify the disc, zygapophyseal (ZJ) or sacro-iliac joint 

(SIJ) as the source of low back pain. 

Design 


Methods 

Search of electronic databases to February 2006; articles 

independently selected, reviewed, and scored by 

two reviewers. The participants had to have nonspecific 

low back pain and the clinical (index) tests had 

to be compared to a ‘gold standard’ reference test. Reference 

tests were: discography for disc pain, intraarticular 

anaesthetic block for SIJ, and intra-articular 

block or medial branch block for ZJ. More strict reference 

standards for a sensitivity analysis were also prespecified: 

for disc pain this was concordant pain at one 

level and a pain-free adjacent level, for ZJ and SIJ it 

was a double control block. 


Index tests were considered clinically informative when 

positive likelihood (+LR) ratios were >2 and/or negative 

likelihood (-LR) ratios were


Manipulation no better than normal care for acute 

back pain. 

Hancock MJ, Maher CG, Latimer J, McLachlan AJ, 

Cooper CW, O’Day R, Spindler MF, McAuley JH 

(2007). Assessment of diclofenac or spinal manipulative 

therapy, or both, in addition to recommended 

first-line treatment for acute low back pain: a randomised 

controlled trial. The Lancet 370, Issue 

9599.1638-1643. 

Objective 

To investigate whether manipulation or non-steroidal 

anti-inflammatory drugs (NSAID) or both results in 

faster recovery in acute low back pain than GP advice 

and paracetamol only. 

Design 

Randomised controlled trial. 

Setting 

Primary care and private clinics, Sydney, Australia. 

Patients 

Consenting patients with acute low back pain (< 6 

weeks duration) presenting to one of 40 participating 

GPs. Exclusion criteria: serious spinal pathology, nerve 

root pathology, contraindications or presently receiving 

either treatment, spinal surgery in last six months. 320 

patients were screened and 240 were randomised to 

one of four groups; a sample size calculation recommended 

this number to detect a 20% difference in recovery 

rates between control and placebo groups with 

80% power and α level of 0.05. 

Intervention 

All patients received advice from the GP and paracetamol 

before randomisation into one of four groups: 1) 

placebo drug and placebo ‘spinal manipulation therapy’ 

(SMT); 2) NSAID and placebo SMT; 3) SMT placebo 

drug; 4) SMT and NSAID. Placebo drugs were 

identical to active NSAID in shape, colour and size. Placebo 

‘SMT’ was detuned ultrasound equal in length to 

genuine SMT session. Genuine SMT was done by 

therapists who regularly used SMT, 2-3 times a week 

over four weeks and including mobilisation (97%) and 

manipulation (5%) at the discretion of the physiotherapist. 


Primary outcomes were the first pain-free day and the 

first seven consecutive pain-free days (with pain free 

being 0 or 1 score out of 10). Secondary outcomes 

were pain score (0-10), Roland-Morris Disability Questionnaire, 

and overall perceived effect. Secondary outcomes 

were assessed at baseline and 1, 2, 4,and 12 

weeks. Adherence to treatment protocol, cointerventions 

and side-effects were monitored. 

Main results 

Paracetamol intake across the four groups was not significantly 

different. The intake of active NSAID and 


placebo drug was not significantly different. The number 

of sessions of placebo and genuine SMT was not significantly 

different. Additional interventions were also 

equal across groups. Treatment credibility was high 

across all groups. 

Patients who took active NSAID did not recover more 

quickly than those who took the placebo drug for either 

of primary outcomes (P=0.51, 0.91). Patients who received 

active SMT did not recover more quickly than 

those who received placebo SMT for either of primary 

outcomes (P=0.95, 0.87). There were no significant differences 

between active and placebo interventions in 

any of the secondary outcome measures. 

Conclusions 

Patients with acute low back pain receiving advice and 

paracetamol do not recover any more quickly with the 

additional provision of SMT or NSAID. 

Comments 

This is a very high quality RCT; proof of such that it got 

published in a journal with a very high ‘impact’ factor. 

They tested out the question whether SMT or NSAID 

would add in any additional treatment effect over and 

above standard advice and paracetamol treatment from 

the GP. To do this they used genuine treatments for 

both interventions, placebo for both interventions, and 

two groups in which one treatment was genuine and 

one was placebo. All groups improved over time, but 

there were no significant differences between placebo 

and active interventions; so the active interventions 

added no additional treatment effect to the standard GP 

care. 

However, the ‘standard GP care’ was not defined, nor 

whether additional training was given to encourage GP 

compliance with a specific package of advice; so 

whether this was really standard care or an augmented 

version is not completely clear. But even so, it is precisely 

this sub-acute group, symptoms less than 6 

weeks that SMT and NSAID are said to provide more 

effective treatment for. This high quality RCT did not 

support this position. 

A recent systematic review about manual therapy, but 

directed at back pain of more than six weeks duration, 

found that RCTs of weaker quality and with fewer patient 

numbers were more likely to produce contradictory 

findings, whereas higher quality RCTs with larger numbers 

gave more robust, if less optimistic results 

(Hettinger et al 2008). This study is high quality with 

good patient numbers, if we extrapolate the findings of 

that systematic review to patients with symptoms less 

than six weeks, then it would be suggested that these 

findings are more robust than smaller and less well constructed 

RCTs. 



Reference 

Hettinger DM, Hurley DA, Jackson A, May S, Mercer C, 

Roberts L (2008). Assessing the effect of sample size, 

methodological quality and statistical rigour on outcomes 

of randomised controlled trials on mobilisation, 

manipulation and massage for low back pain of at least 

6 weeks duration. Physiotherapy 94.97-104. 

�� 

Clinical prediction rule for manipulation not supported 

by independent study and so cannot be generally 

applied. 

Hancock MJ, Maher CG, Latimer J, Herbert RD, 

McAuley JH (2008). Independent evaluation of a 

clinical prediction rule for spinal manipulative therapy: 

a randomised controlled trial. European Spine 

Journal 2008; 17:936-943. 

Objective 

Independent evaluation of a proposed clinical prediction 

rule for responders to spinal manipulative therapy 

(SMT). 

Design 

Pre-planned secondary analysis of a previous randomised 

controlled trial that has been reviewed above, 

where it is described in more detail (Hancock et al 

2007). 

Setting 

Primary care Sydney, Australia 

Patients 

239 patients presenting to general practice with acute, 

non-specific low back pain of less than six weeks duration. 

Intervention 

Patients were randomised to one of four treatment arms 

as well as receiving GP advice: 

�� Placebo SMT and placebo diclofenac 

�� Placebo SMT and active diclofenac 

�� Active SMT and placebo diclofenac 

�� Active SMT and active diclofenac 

For the purposes of this secondary analysis the two active 

and two placebo SMT groups were respectively 

analysed as the intervention and the control groups. In 

the SMT group, patients received a range of mostly low 

velocity techniques administered at the discretion of the 

physical therapists, which was not the standardised 

SMT sacro-iliac joint technique used in the clinical prediction 

rule development. 


Patients who met four or more of the five criteria for the 

SMT clinical prediction rule (Childs et al 2004) were 

classified as positive for the rule. Pain and disability 


were the main outcome measures. Primary data analysis 

considered treatment group (active or placebo 

SMT), time (baseline to follow-up time points), and prediction 

rule status (positive or negative) in repeated 

measures analysis of variance for both pain and disability. 

Secondary data analysis involved linear regression 

to determine effect of treatment group, prediction rule 

status, interaction between treatment group and prediction 

rule status and baseline scores on pain and disability. 

Main results 

In the primary analysis there was no significant interaction 

between clinical prediction rule status, treatment 

received, and outcome for either pain (P =0.8) or disability 

(P = 0.6). In the secondary analysis, positive 

status on the rule tended to predict better outcome regardless 

of treatment, with some interactions being significant. 

Conclusions 

The clinical prediction rule performed no better than 

chance in identifying patients with acute, non-specific 

low back pain who were most likely to respond to manipulation. 

Comments 

This clinical prediction rule for manipulation responders 

was derived in one study (Flynn et al 2002) and appeared 

to work in a similar healthcare setting (Childs et 

al 2004), but showed no value in another healthcare 

setting. In the initial trial as detailed above neither diclofenac 

nor manipulation provided any additional treatment 

effect over GP advice and paracetamol. The aim 

of this secondary analysis was to see if the subgroup of 

patients identified by an earlier clinical prediction rule 

(Childs et al 2004) responded better and thus would 

validate this clinical prediction rule. This is obviously an 

important clinical question – can we identify a sub-group 

of patient who respond to manipulation? It is also important 

in terms of establishing the validity of clinical prediction 

rules, which must be validated in different clinical 

settings as part of the process of establishing their clinical 

value (Stiell and Wells 1999). Clinical prediction 

rules are notoriously bad at being transferable from different 

clinical settings (Haynes et al 2006). In this secondary 

analysis of a randomised clinical trial the clinical 

features that had earlier suggested a positive response 

to manipulation had no bearing on outcome. The clinical 

prediction rule was not helpful in identifying patients 

who responded well to manipulation. 

In terms of confirmation of a clinical prediction rule, validation 

in a different clinical setting is stage 2 of a 5stage 

validification process (McGinn et al 2004, Reilly 

and Evans 2006). Failing to achieve this independent 

evaluation means that the clinical prediction rule should 

only be used with caution and only in similar patients to 

those in which it was derived (Reilly and Evans 2006). 



It has been suggested that as the treatment in Hancock 

et al (2008) was at the discretion of the treating therapists 

and mostly consisted of mobilisations, rather than 

the standardised manipulation used in the original studies, 

this is a weakness of this evaluation (letter to editor 

by Herbert and Perle). The authors however see this as 

a strength of the present independent evaluation, which 

was to determine if the clinical prediction rule would 

generalise to SMT selected by skilled and experienced 

clinicians (letter to editor by Hancock et al). It did not, 

and the lack of generalisability could equally have been 

due to differences in patients, settings, co-interventions 

or some other unknown variable. Currently, the clinical 

prediction rule has not been shown to generalise to a 

new setting and further investigation is needed to determine 

if it is transferable to other settings. 

References 


Majkowski GR, Delitto A (2004). A clinical prediction 

rule to identify patients with low back pain most likely to 

benefit from spinal manipulations: a validation study. 

Ann Intern Med 141:920-928. 

Flynn T, Fritz J, Whitman J, Wainner R, Magel J, 

Rendeiro D (2002). A clinical prediction rule for classifying 

patients with low back pain who demonstrate shortterm 

improvement with spinal manipulation. Spine 

27:2835-2843. 

Hancock MJ, Maher CG, Latimer J, McLachlan AJ, 

Cooper CW, O’Day R, Spindler MF, McAuley JH (2007). 

Assessment of diclofenac or spinal manipulative therapy, 

or both, in addition to recommended first-line treatment 

for acute low back pain: a randomised controlled 

trial. Lancet 370.1638-1643. 

Haynes RB, Sackett DL, Guyatt GH, Tugwell P (2006). 

Clinical Epidemiology. How to do Clinical Practice Research. 

3rd edition. Lippincott Williams & Wilkins: Philadelphia. 

McGinn T, Guyatt G, Wyer P et al (2000). How to use 

articles about clinical prediction rules. J Am Med Ass 

284:79-84 

Reilly BM, Evans AT (2006). Translating clinical research 

into clinical practice: impact of using prediction 

rules to make decisions. Ann Int Med 144:201-209. 

Stiell IG, Wells GA (1999). Methodological standards for 

the development of clinical decision rules in emergency 

medicine. Ann Emerg Med 33.437-447. 


Stabilisation exercises – better than nothing, but no 

better than anything else. 

Several systematic reviews have recently been published 

about stabilisation exercises. Overall the reviews 

are consistent in their conclusions; stabilisation exercises 

are better than no or minimal treatment, but are 

no better than most other active treatments. This overview 

will describe one of these reviews and then comment 

on the general trend in the literature. 

May S, Johnson R (2008). Stabilisation exercises for 

low back pain: a systematic review. Physiotherapy 

94.179-189. 

Objective 

To systematically review the literature to determine the 

effectiveness of stabilisation exercises for the treatment 

of pain and functional disability in patients with low back 

pain. 

Design 


Methods 

Studies were identified with a search of electronic databases 

and hand search of reference lists in identified 

studies up till October 2006. To be included studies had 

to be published full reports, in English, involve adults 

with low back pain, which could be specific or nonspecific 

in nature, and be a randomised controlled trial. 

One intervention arm had to use primarily stabilisation 

exercises (SE), though possibly other interventions, as 

described by specific authors (Richardson, O’Sullivan, 

Norris); the control group had to receive an alternative 

intervention, and studies had to monitor changes in pain 

and/or functional disability. Methodological quality was 

measured using the PEDro scale, with scores 

downloaded from the PEDro site if available. Data was 

extracted independently by the 2 reviewers and disagreements 

were resolved by consensus. Both quantitative 

and qualitative methods of data analysis were 

considered depending on whether the data were sufficiently 

homogeneous or were heterogeneous according 

to back pain, outcomes, symptom duration, quality and 

control groups. 

Main results 

Twenty-one papers were included in the review, three of 

which were long-term follow-ups; thus we identified 18 

separate and relevant clinical trials. Trials involved a 

range of acute to chronic back pain, specific and nonspecific 

back pain, and most frequently the stabilisation 

exercises (SE) were combined with some other intervention. 

Twelve publications scored ≥6 on the PEDro 

scale and were considered high quality. The studies 

were sufficiently heterogeneous to disqualify a metaanalysis 

so conclusions were based on Cochrane review 

group levels of evidence. 



In the treatment of acute and sub-acute low back pain 

there was conflicting evidence in four trials about the 

effectiveness of SE, but they did not appear to be effective. 

In the treatment of chronic low back pain, SE appear 

to be effective, but the overall evidence was conflicting. 

In the short-term, five mostly high quality trials, 

showed significant differences favouring SE, one high 

quality study favoured the control group, and four, 

mostly low quality studies, showed no significant difference. 

Results were similar long-term. However, the results 

depended on the type of control group; SE were 

much more likely to generate significant differences 

against ‘inactive’ than ‘active’ control groups. 

Conclusions 

There maybe a role for specific stabilisation exercises in 

some patients with chronic back pain, but they are no 

more effectives than other active treatments. 

Comments 

This is the largest systematic review of SE conducted 

so far; it suggested that they are not effective for acute 

and sub-acute low back pain, but that they might be 

more effective than inactive control treatments for 

chronic low back pain. Compared with active treatments 

for chronic low back pain, however, the outcomes were 

very similar. Compared with ‘inactive’ controls such as, 

medical management, no treatment or standard treatment, 

the SE mostly did significantly better (5 out of 8 

did better). Whilst against active controls such as, 

McKenzie, directional preference exercises, or manual 

therapy with or without exercises, the SE were much 

less likely to do well (1 out of 8 did better). There were 

no clinically significant differences between SE and 

general strengthening, McKenzie, and directional preference 

exercises. Compared with manual therapy, three 

out of six trials favoured SE, and one favoured manual 

therapy. The other important limitation was that only 

fou,r studies used SE as the sole intervention, and more 

commonly it was combined with another intervention, 

such as manual therapy. Although this may reflect clinical 

practice it is not helpful when seeking to prove effectiveness; 

the treatment effect might have been due to 

another element of the treatment package and not the 

SE at all. 

This last point was addressed in another recent systematic 

review, which only included trials in which SE was 

the sole treatment (Standaert et al 2008). They only 

included three trials that met the criteria and concluded 

that there was moderate evidence that they were effective 

at improving pain, and strong evidence that they are 

no more effective than general exercises, and moderate 

evidence that they were more effective than manual 

therapy. 

Similar conclusions were reached by another recent 

systematic review that included 13 studies, one of which 

related to neck pain and headaches (Ferreira et 


al 2006). They reported that SE were effective for pelvic 

pain and prevention of recurrence, but not for acute low 

back pain. They also reported that SE were effective for 

chronic low back pain when compared with no or minimal 

treatment, but not when compared to other conventional 

physiotherapy. Regarding pelvic pain, the evidence 

was contradictory in our review (May and Johnson 

2008). Regarding prevention of recurrence, this 

occurred in such a specific group; they had the very first 

episode of unilateral low back pain, duration less than 3 

weeks with multifidus muscle bulk asymmetry > 11%, 

that the finding cannot be generalised. 

A slightly earlier systematic review that included seven 

trials came to similar conclusions (Rackwitz et al 2006). 

For acute back pain, they concluded that SE were 

equally effective as treatment by the GP, but more effective 

at preventing recurrences. For chronic low back 

pain, SE were more effective short and long-term than 

treatment by the GP, and may be as effective as other 

physiotherapy treatments. There was limited evidence 

that SE provide any additional benefit if combined with 

other treatment compared to other treatment alone. 

Thus, we have four contemporary systematic reviews, 

the highest level of evidence, that come to very nearly 

the same conclusions about SE, which is not surprising 

given that it is much the same literature being evaluated. 

For acute back pain, SE are no more effective 

than standard GP care; they may prevent additional 

episodes in a very specific group, but this group is 

probably so rare that this has limited clinical value. For 

chronic back pain, SE are more effective at reducing 

pain and disability than no or minimal treatment; but are 

little different than other active physiotherapy interventions. 

SE appear to affect pain more than function. SE 

do not appear to add any additional benefit to other active 

physiotherapy treatments, such as exercises or 

manual therapy. 

These reviews imply a number of research and clinical 

implications. Whilst combining interventions may reflect 

clinical practice, it makes evaluation of effectiveness 

impossible; the bottom line is, we cannot know what the 

effective part of a treatment package is if several treatments 

are combined. Trials should evaluate treatments 

singly, or use a factorial design, to truly determine their 

clinical value. The conclusion, from several reviews, 

that effectiveness is relative to the control group is similar 

to the conclusion of some recent reviews of manipulation; 

these interventions appear to be effective when 

compared to no or minimal treatments, but not when 

compared to other active treatments. This has ethical 

implications regarding control groups. Although ideally 

we would want interventions for low back pain to be 

proven better than a placebo or minimal intervention, is 

this any longer 



justifiable if all active control groups are reasonably 

equal in outcome. To really identify the best treatment 

should patients be offered at least two proven treatments 

and see which does best? 

The clinical implication of these reviews is obvious; 

there is no need to prescribe SE if physical therapists 

are providing some other active treatment. In fact, given 

the complexity of teaching and learning these exercises, 

this is probably a waste of time. Furthermore, SE do not 

appear to add anything if given in addition to general 

exercises or other physiotherapy interventions. There 

may be a subgroup of the low back pain population who 

respond more positively to SE than other interventions, 

and work has been done to define and endorse a clinical 

prediction rule based on clinical characteristics for 

identifying who will respond best to SE (Hicks et al 

2005, Brennan et a. 2006). However, a number of methodological 

problems prevent these clinical prediction 

rules from being applied to the general back pain population 

(May 2007, May and Rosedale, submitted); and 

only the prone instability test for identifying ‘instability’ 

has demonstrated reasonable inter-rater reliability (Fritz 

et al 2005, Hicks et al 2003). So, if a sub-group of the 

low back pain population exists for whom SE are the 

optimal treatment there is a lack of evidence that establishes 

a reliable and valid way of identifying this subgroup. 

References 

Brennan GP, Fritz JM, Hunter SJ, Thackeray A, Delitto 

A, Erhard RE. Identifying subgroups of patients with 

acute/sub acute “non-specific” low back pain. Spine 

2006;31:623-631. 

Cleland J, Schulte C, Durall C (2002). The role of therapeutic 

exercise in treating instability-related lumbar 

spine pain: a systematic review. J Back Musculoskeletal 

Rehab 16.105-115. 


Ferreira PH, Ferreira ML, Maher CG, Herbert RD, Refshauge 

K (2006). Specific stabilisation exercise for spinal 

and pelvic pain: a systematic review. Aust J Physio 

52.79-88. 

Fritz JM, Piva SR, Childs JD. Accuracy of the clinical 

examination to predict radiographic instability of the 

lumbar spine. Eur Spine J 2005;14:743-750. 

Hicks GE, Fritz JM, Delitto A, Mishock J. Interrater reliability 

of clinical examination measures for identification 

of lumbar segmental instability. Arch Phys Med Rehab 

2003;84:1858-1864. 

Hicks GE, Fritz JE, Delitto A, McGill SM. Preliminary 

development of a clinical prediction rule for determining 

which patients with low back pain will respond to a stabilization 

exercise program. Arch Phys Med Rehabil 

2005;86:1753-1762. 

May S (2007). Clinical prediction rules and their application 

in back pain research. Int J Mech Diagnosis Ther 

1.2. 

May S, Rosedale R (2008). Clinical prediction rules in 

back pain research – a systematic review. Submitted 

Rackwittz B, de Bie R, Ewert T, Stucki G (2006). Segmental 

stabilizing exercises and low back pain. What is 

the evidence? A systematic review of randomized controlled 

trials. Clin Rehab 20.553-67. 

Standaert CJ, Weinstein SM, Rumpeltes J (2008). Evidence-informed 

management of chronic low back pain 

with lumbar stabilization exercises. Spine J 8.114-120. 

Vinderen af konkurrencen på 

Fagfestivalen 


Neck pain and headache caused by spontaneous intracranial hypotension 

Megan Kelly, BPhty and Stuart Horton, MPhty, Dip. MDT 







Aim 

This report describes the case of a woman who was 

referred to physiotherapy by her general practitioner 

(GP) with constant neck pain and headache, and 

highlights the processes over two physiotherapy 

sessions to conclude this patient’s condition as nonmechanical. 

Clinical presentation 

A 34 year old woman attended an appointment with her 

GP to seek treatment for constant neck pain and 

headache, which came on suddenly after playing 

squash. She described her neck pain and headache as 

a constant 3-4/10 on the visual analogue scale (VAS) 

and she had a sharp stabbing pain in her head 

associated with movement at 6-7/10. This pain was at 

no time described as the worst pain and headache that 

she had ever had, but the patient was visibly in 

constant discomfort. She commented that she had 

been very fatigued since the pains began and she 

wanted to sleep much of the day. She was very 

nauseous, particularly associated with movements and 

felt the only position to help dissipate the symptoms 

was in lying. She occasionally felt dizzy, but at no point 

had lost consciousness. Her GP had referred her to 

physiotherapy to assess if her symptoms were from a 

mechanical source, because a CT scan and lumbar 

puncture showed no abnormalities or blood in the 

Cerebrospinal Fluid (CSF), which would have indicated 

a sub-arachnoid haemorrhage (SAH). Potential red 

flags noticed in this assessment included sudden onset 

of headache, no dissipation of neck pain and headache 

with the use of NSAIDS or simple analgesics, and a 

feeling of nausea and fatigue. Red flags typical of a 

serious pathology or more sinister cause, but of which 

this patient did not present with, include a lack of 

coordination, headache disturbing sleep, patient’s 

description of the headache including a similar 

description to ‘the worst headache they have ever had’ 

and a history of numbness or tingling (Graves 2006). 

Physical examination revealed the patient’s condition 

was highly irritable and she could not find comfort in 

any position other than supine or prone lying. All her 

cervical spine active range of movement (AROM) was 

normal, although painful, and caused nausea and 

increase in headache. Vertebrobasilar insufficiency 

(VBI) testing showed negative results. Repeated 

movements and sustained positions either had no 

effect or served to irritate the symptoms, so were 

abandoned. 

Classification 

At this point, a list of provisional diagnoses was made 

and then each one was considered as a source of 

symptoms, based on the findings from the history and 

physical assessments. These were as follows: 

Mechanical Headache (Derangement): Although 

possible, it was considered less likely, as repeated 

movements and sustained positions of the cervical 


spine had no effect, or worsened the symptoms 


Migraine: This was discounted because her symptoms 

had lasted for much longer than 72 hours, she was not 

sensitive to light and her symptoms were bilateral, all of 

which do not fit into the model described by the 

headache classification committee of the International 

Headache Society (IHS 2004). 

VBI: This was eliminated as the patient did not show 

positive signs with the functional positioning tests (APA 

2006). According to Coman (1986), VBI presents with 

any of the 5 D’s and 3 N’s. These are Dizziness, 

Dipoplia, Dysathria, Dysphagia, Drop attacks, 

Nystagmus, Numbness and Nausea. This patient only 

presented with two of these symptoms. 

Sub-arachnoid Haemorrhage (SAH): An original CT 

scan and lumbar puncture revealed no blood in the 

CSF fluid and no cranial signs of a sub-arachnoid 

haemorrhage. 

The patient was classified as “other”, as there was a 

strong possibility of this being a non-mechanical 

condition, although it was unclear exactly what. 

Intervention and outcome 

The majority of the first physiotherapy session 

consisted of the history and physical assessments and 

treatment was heavily dictated by the patient’s 

irritability. At this point, the treatment was kept very 

conservative. It was prudent ‘to do no harm’ as 

described by Taylor and Kerry (2006). The patient was 

advised to continue seeing her GP and to have further 

investigations if they thought this was indicated. She 

was also instructed to rest and avoid any activities that 

stirred up the symptoms. A home exercise was devised 

to try to assess further if any position or movement 

could improve symptoms. This exercise consisted of 

cervical retractions in lying x 10, as tolerated several 

times daily. This was chosen because, from the 

assessment, it was the least provocative movement 

and position to perform it. 

At the second physiotherapy session two days later, 

the patient was unchanged. Once again, she reported 

that the only time she feels better is when she lies 

down. Treatment included some manual cervical 

traction in lying with retraction, as this movement was 

comfortable for the patient, although again, there was 

no significant change. Still keeping with the very 

conservative management approach for this patient, 

she was given an upper back and neck massage, 

consisting purely of effleurage to try to decrease pain 

and any tension in muscles arisen over the previous 

couple of days. 

However, after two sessions it was strongly felt that this 

presentation was behaving as though it was nonmechanical 

and should therefore be referred back to 

the GP for further investigations. No distinct outcomes 

from treatment had become apparent. 



The patient was followed up by phone a week later. 

Her GP had made arrangements for her to receive an 

MRI scan. It was the results of this scan that revealed 

her eventual diagnosis. The patient was suffering from 

Spontaneous Intracranial Hypotension (SIH), as a 

result of a tear in the dura which was leaking CSF. The 

patient was on strict bed rest and was waiting to 

receive blood patching surgery to repair the tear in the 

dura. 

Headache in SIH 

Headache is the most common clinical symptom of 

cranial hypotension, with a distinctive feature that it is 

accentuated when the patient is upright and relieved 

when recumbent. A further distinctive feature is that the 

headache has a ‘ballistic’ component, with the pain 

being aggravated by any form of head movement, 

particularly shaking (O’Carroll and Brant-Zawadzki 

1999). Accompanying symptoms can include a 

combination of any of the following: a loss of hearing, 

tinnitus, vertigo, stiffness of the neck, nausea, and 

vomiting. On retrospective analysis of this patient’s 

presenting symptoms and the signs and symptoms of 

someone suffering from SIH, many similarities can be 

observed. Wouter et al (2005) revealed that 94% of 

patients who had SIH were initially misdiagnosed when 

they visited a doctor about their symptoms. SIH may 

even present similarly to cervical radiculopathy 

(Albayram et al 2002). SIH is most typically reported 

after certain procedures such as lumbar puncture or 

spinal surgery (Mokri et al 1997), but may be 

associated with minor trauma such as sports activities 

or severe coughing (Christoforidis et al 1998). 

Treatment of SIH Headache 

Many patients improve spontaneously or coincident 

with noninvasive measures such as bed rest, to allow 

the dural tear to heal. Epidural blood patch (EBP) is 

used in patients who fail noninvasive measures 

(Bahram and Posner 2000). It is performed by infusing 

autologous blood into the epidural space under sterile 

conditions. The presumed mechanism of action is an 

immediate gelatinous tamponade of dural leak followed 

by fibrin depositin and fibroblastic activity (Taivainen et 

al 1993). 

Conclusions 

This case study has highlighted a clinical presentation 

of SIH in a patient presenting to physiotherapy with 

headache. When assessing a patient who presents 

with signs and symptoms similar to this patient 

(constant neck pain and headaches with no resolve), 

the therapist should be proactive in recognising red 

flags and should know the appropriate treatment 

options. The therapist should always aim to ‘do no 

harm’ in this situation and be primarily focused on 

detecting whether or not the symptoms are from a 

mechanical source. Taylor and Kerry (2005) describe 

that the acute onset of a headache which results with 

or without a traumatic incident, and is unresponsive to 


any mechanical therapy, is a warning sign of an 

underlying arterial injury. It is not uncommon for a trial 

of mechanical therapy to be advised by a GP at a point 

early in the patient’s treatment; therefore the treating 

therapist must be aware of warning signs of potential 

sinister underlying conditions. 

References 

Australian Physiotherapy Association (APA 2006). 

Clinical guidelines for assessing vertebrobasilar 

insufficiency in the management of cervical spine 

disorders. 

Bahram M, Posner JB (1999) Spontaneous intracranial 

hypotension: The broadening clinical and imaging 

spectrum of CSF leaks. Neurology 55(12):1771-1772. 

Christoforidis GA, Mehta BA, Landi JL, Czarnecki EJ, 

Piaskowski RA (1998) Spontaneous intracranial 

hypotension: report of four cases and review of the 

literature. Neuroradiology 40:636-643. 

Coman WB (1986). Dizziness related to ENT 

conditions. In: Grieve GP, (ed). Grieve’s modern 

manual therapy of the vertebral column. Edinburgh: 

Churchill Livingstone. 

Graves BW (2006). Management of migraine 

headaches. Journal of Midwifery and Women’s Health. 

51: 174-184. 

International Headache Society (2004) Headache 

Classification Committee of the International Headache 

Society. Cephalalgia 24: 1–160. 

McKenzie RA, May S (2006) The cervical and thoracic 

spine. Mechanical diagnosis and therapy. (2nd ed). 

Spinal Publications, Waikanae. 

Mokri B, Piepras DG, Miller GM (1997) Syndrome of 

orthostatic headaches and diffuse pachymeningeal 

gadolinium enhancement. Mayo Clinic Procedures. 72: 

400-413. 

O’Carroll CP, Brant-Zawadzki M (1999) The syndrome 

of spontaneous intracranial hypotension. Cephalalgia 

19:80-87. 

Taivainen T, Pitkanen M, Tuominen M, Rosenberg PH 

(1993) Efficacy of epidural blood patch for post dural 

puncture headache. Acta Anesthesiology Scandinavia 

37:702-705. 

Taylor AJ and Kerry R (2006). Cervical arterial 

dysfunction assessment and manual therapy. Manual 

Therapy 11: 243-253. 

Taylor AJ and Kerry R (2005). Neck pain and headache 

as a result of internal carotid artery dissection: 

implications for manual therapists. Manual Therapy 10: 

73-77. 

Wouter I, Schievink M, Maya M and Louy C (2005). 

Cranial MRI predicts outcome of spontaneous 

intracranial hypotension. Neurology 64: 1282-1284 





Christie Downing, PT, DPT, Cert. MDT 









Abstract 

This is a case study where a 41 year old female patient with lumbar radiculopathy was classified as having an 

“irreducible derangement” via Mechanical Diagnosis and Therapy (MDT). She subsequently underwent a surgical 

microdiscectomy. She had good surgical outcomes, but redeveloped peripheral symptoms. After a short course of 

post-operative rehabilitation using MDT principles, she had excellent outcomes for pain reduction, recovery of 

motion and return to function. This case study demonstrates the use of MDT in pre-surgical screening and in postoperative 

rehabilitation. 

Introduction 

The healthcare community seeks to improve procedural 

outcomes, and therefore, patient selection and 

classification that directs treatment is a goal. This is of 

particular importance when considering low back 

surgery. It is known that the outcomes of various 

lumbar surgical procedures remain highly variable. 

Even “minimally invasive” procedures such as 

microdiscectomy are argued to not have any clear 

benefit over conservative care (Osterman et al 2006). 

Studies also reveal that patients who have been 

allocated to control groups often resort to surgical 

intervention. Good long term results for 

microdiscectomy, specifically, have been reported as 

high as 83% at 10 years (Findlay et al 1998). 

Therefore, it implies that there is perhaps a select 

group of patients for whom surgical intervention is the 

management of choice. 

Magnetic resonance imaging (MRI) has been found not 

to be a sensitive, specific or accurate (72%, 68%, 70%, 

respectively) tool to determine if surgery is appropriate 

(Weiner and Patel 2008). Therefore, the decision to 

pursue surgical intervention should not be based on the 

findings of imaging studies. 

The timing of surgical intervention has been studied in 

a randomized controlled trial of 283 patients (Peul et al 

2008) .The differences in long term outcomes between 

immediate (approximately six to twelve weeks after 

onset of symptoms) surgical intervention and delayed 

intervention with conservative care were not significant. 

However, there were short term improvements in pain 

and disability in those who underwent immediate 

surgical intervention. In the delayed intervention group, 

only 44% required surgery by the end of the study. This 

implies that whilst a proportion of patients can avoid 

surgery, there appears to be group of patients who 

would benefit from immediate surgical intervention. 

Therefore, it is imperative to identify this group early . 

The need for post surgical rehabilitation after 

microdiscectomy remains controversial. Danielsen et al 

(2006) found that rigorous stabilization exercises were 

superior to mild exercises at six and twelve months 

post-operatively after discectomy. In another study, 

where 20 patients who underwent microdiscectomy 

were randomized to either a four week exercise 

program or a control group, it was found that 

statistically different improvements in pain, disability 

and muscle function were maintained in the exercise 

group at 12 months (Dolan et al 2000). In contrast, 

other studies have demonstrated that whilst there may 

be short term benefits, no significant long term benefits 

are achieved with physical therapy after lumbar 

microdiscectomy or decompression surgery. (Erdogous 

et al 2007) (Mannion et al 2007). Manniche et al 

(1993), examined the role of “hyperextension” 

exercises in addition to stabilization in patients who 

underwent single level microdiscectomy. The study 

(n=62) found no benefit to the addition of 

hyperextension exercises. The patients were not 

assessed for a directional preference for extension 

which may have influenced the outcomes. 

The studies to date have focused on generic exercises 

and primarily on general conditioning and/or lumbar 

stabilization exercise and there have been no studies 

published that examine the role of Mechanical 

Diagnosis and Therapy (MDT) in post-surgical 

rehabilitation. 

The following case study illustrates how MDT can 

assist as a screening tool for selection of patients for 

surgery. It also demonstrates how MDT can be utilized 

for post-surgical rehabilitation after lumbar 

microdiscectomy. 

History 

This is the case of a 41 year old female physical 

therapist with a history of back and left lateral leg pain 

that radiated to the foot. She had undergone surgery 

eight weeks prior to evaluation. 

She reported a prior history of intermittent back pain 

without leg pain for approximately five years. Several 

months prior to surgery, she developed leg pain for no 

apparent reason. She did seek physical therapy from a 

therapist trained in MDT. Her treatment included both 

saggital and lateral plane forces, but both worsened her 

peripheral symptoms. A trial of mechanical traction 

produced a similar result. Following six treatment 

sessions, she was classified as having an irreducible 

derangement and was referred back to the physician. 

After undergoing a lumbar L-5/S-1 microdiscectomy, 

her leg pain was abolished and intermittent left sided 

back pain was the only remaining symptom. At follow 

up with the surgeon, she was prescribed a once daily 

dose of Celebrex and was advised to begin resuming 

activities and flexibility but without any specific 

recommendations. She commenced repeated 

extension in lying, flexion in lying, and hamstring 

stretching. She reported that she feared “scar tissue” 

build up around the nerve root and lumbar fascia; 

therefore, she felt she should perform flexion biased 

activities. She noted, however, that flexion of the 





lumbar spine and hamstring stretching would produce 

concordant leg symptoms. Although the peripheral 

symptoms never remained worse, the intensity of the 

symptoms was increasing. It was at this point she 

sought further advice for self treatment procedures. 

Treatment Day 1 

The results of the initial exam indicated that she was an 

extension responder (standing). However, it was noted 

that repeated extension failed to improve the loss of 

sidegliding and also failed to restore her ability to 

perform repeated flexion in lying without leg symptoms. 

Repeated left sidegliding in standing recovered her left 

sidegliding range of motion and restored full extension 

without pain. Furthermore, she was able to perform 

several repeated flexions in lying before leg symptoms 

were reproduced. Repeated left sidegliding appeared to 

provide her with the best symptomatic and mechanical 

response. 

Management for Day 1 included posture education, and 

avoidance of flexion in lying and hamstring stretching. 

Self treatment included repeated left sidegliding in 

standing to be performed every two hours during the 

waking hours. 


She reported 100% compliance with the home exercise 

program. She noted that upon wakening, she would 

experience pain during movement with sidegliding, but 

it was never worse. This would usually subside after 

her morning dose of Celebrex. There were no episodes 

of leg pain, but some mild left sided back pain. 

Examination revealed no observable loss of left 

sidegliding. Extension in lying produced end-range 

back pain, but did not remain worse. Flexion in lying 

produced a very faint production of leg pain to the foot 

rated 1/10 on the visual analogue scale. Left sidegliding 

produced pain during movement, but was not worse. 

Extension was combined with left sidegliding in 

standing and this subsequently resulted in abolition of 

pain during movement of left sidegliding. It also 

abolished the end-range pain with extension in lying. 

She was advised to add extension to the left 

sidegliding. 

Three days later, she reported that she was now only 

having pain during movement first thing in the morning. 

She was advised to continue the exercise described on 

Day 2. A few days after this, she reported having fallen 

twice on the ice during a winter storm. She reported a 

temporary production of peripheral symptoms, which 

resolved with the exercises. 


This was approximately 10 days after the initial 

evaluation. She had full pain-free range of motion in all 

directions. Flexion in lying produced mild symptoms to 

the thigh only, and extension produced end range pain. 

An overpressure technique was added to extension in 

lying, after which she could abolish all symptoms with 

self generated extension in lying. Since simple 

extension abolished her symptoms, she was now 

advised to perform repeated extension in lying at least 

twice a day and repeated extension in standing every 

two hours at work . She also monitored her progress by 

checking for pain during movement with left sidegliding. 


The following week, she reported being pain-free for a 

few days, and had stopped taking Celebrex. Although 

she occasionally experienced pain during movement 

with sidegliding, this usually abated quickly. On 

assessment, twenty repetitions of flexion in lying failed 

to produce any symptoms other than what she 

described as a “fascial” stretch in the lumbar spine. 

There was no subsequent pain or obstruction to 

extension. Flexion in lying was introduced to the home 

program. She was instructed to perform this twice a 

day and follow with extension in lying. Also, she was 

instructed on how to monitor for obstruction or 

worsening of symptoms after flexion and to continue 

repeated extension (lying or standing) every two hours. 

Follow up contact 

No further treatment was required. She did seek verbal 

advice on occasion, but it was clear she had an 

excellent understanding of reducing the derangement, 

maintaining reduction and prophylaxis. Three weeks 

after the last treatment, she reported that she only 

experienced occasional, momentary, pain during 

movement with left side glide. She has been able to 

resume working eight hours a day and has been able to 

resume stretching her hamstrings without any pain or 

subsequent obstruction. 

Clinical Implications 

This study has several clinical implications. First, an 

irreducible derangement was determined after six 

sessions of MDT. The treating therapist referred the 

patient back to the physician and surgery was 

performed. The patient had an immediate reduction in 

symptoms. This demonstrates how a mechanical 

assessment can assist in the selection of patients 

requiring surgery. 

Following the surgery, her symptoms indicated that a 

derangement persisted. Repeated movements in to 

flexion quickly produced symptoms that were similar to 

her pre-surgical status. However, after a period of 

management with lateral and extension principles, she 

was able to resume flexion activities without negative 

effects. Considering the rate of failure of post-surgical 

patients, MDT should be considered an important part 

of recovery for those with persistent symptoms. This 

seems to suggest that, in some cases, surgical 

resection does not “cure” a derangement, but allows it 

to be reducible. Furthermore, that treatment using MDT 

in the post surgical status is safe when precautions are 

followed. 





Finally, this study implies MDT treatment in the post 

surgical setting may not add a significant cost to the 

episode of care. Whereas the previously mentioned 

studies used between 8-24 visits (Danielsen et al 2000) 

(Dolan et al 2000) (Manion et al 2007), the patient in 

this case study only required four treatment sessions. 

There is implication here that a patient with good 

motivation may need very few treatment sessions when 

there is a focus on self directed treatment and 

education hence treatment costs are kept to a 

minimum. 

In conclusion, this case study has demonstrated that 

MDT may play an important role in pre-surgical 

screening. Furthermore, although post surgical 

rehabilitation for conditioning and stabilization 

exercises after microdiscectomy remains controversial, 

MDT may play an important role for those with 

persistent deficits and can be done so safely. Finally, 

treatment using MDT may add only minimal cost to the 

episode of care. 

References 

Danielsen J, Johnsen R, Kibsgaard S & Hellevik E 

(2000). Early aggressive exercise for postoperative 

rehabilitation after discectomy. Spine 25, 1015-20. 

Dolan P, Greenfield K, Nelson R & Nelson I (2000). 

Can exercise therapy improve the outcome after 

microdiscectomy? Spine 25, 1523-1432. 

Erdogmus C et al (2007). Physiotherapy-based 

rehabilitation following disc herniation operation: 

Results of a randomized clinical trial. Spine 32, 2041- 

2049. 

Findlay G, Hall B, Musa B, Oliveira M, & Fear S (1998). 

A 10-year follow-up of the outcome of lumbar 

microdiscectomy. Spine 23, 1168-71. 

Manniche C et al (1993).Intensive Dynamic Back 

Exercises With or Without Hyperextension in Chronic 

Back Pain After Surgery for Lumbar Disc Protrusion: A 

Clinical Trial. Spine. 18, 560-67. 

Mannion A, Denzler R, Dvorak J, Muntener M & Grob D 

(2007). A randomized controlled trial of post-operative 

rehabilitation after surgical decompression of the 

lumbar spine. European Spine Journal. 16, 1101-17. 

Osterman H, Seitsalo S, Karppinen J & Malmivaara 

(2006). Effectiveness of Microdiscectomy for Lumbar 

Disc Herniation A Randomized Controlled Trial With 2 

Years of Follow-up. Spine 31, 2409-2414. 

Peul W, von den Hout W, Brand R, Tomeer R & Koes B 

(2008). Prolonged conservative care versus early 

surgery in patients with sciatica caused by lumbar disc 

herniation: two year results of a randomized controlled 

trial. BMJ 336, 1355-1361. 

Weiner B & Patel R (2008). The accuracy of MRI in the 

detection of Lumbar Disc Containment. Journal of 

Orthopaedic Surgery and Research 3, 46. 


The modified Broström repair 


Introduction 

In the 1960's, Broström described a surgical technique 

to repair the lateral ligaments of the ankle as a treatment 

for ankle instability (Hennrikus et al 1996). This 

anatomical repair involves imbrication (shortening of the 

ligament) and direct suturing of the involved ligaments 

in an attempt to address the instability (de Vries et al 

2006, Karlsson et al 1989, Ng & Das De 2007). In the 

1980's, Gould reported a modification to this procedure 

by including the extensor retinaculum to reinforce the 

repair of the lateral ligaments (Liu & Baker 1994). The 

modified Broström repair is now a commonly used procedure 

when conservative treatment has failed and is 

widely reported and investigated in the literature (Bahr 

et al 1997, de Vries et al 2006, Fuji et al 2006, Liu & 

Baker 1994). 

Despite the common use of the modified Broström procedure, 

it is apparent that there has been little focus in 

the literature relating to the content or efficacy of postoperative 

rehabilitation programmes. Many authors 

have reported success and failure rates of the procedure 

without mention of the post-operative protocols 

utilised (Bell et al 2006). This is in contrast to the nonoperative 

population where repeated investigation has 

been undertaken and a functional approach is recommended 

in favour of immobilisation irrespective of the 

degree of injury without compromise of mechanical stability 

(de Vries et al 2006, Kannus et al 1991, Lynch & 

Renstrom 1999). 

Table 1. Description of studies 

Study Methods Participants Interventions Outcomes 

Karlsson 

et al 1995 

Karlsson 

et al 1999 

Prospective 

randomised 

controlled trial 

Prospective 

randomised 

trial 

40 participants; 18 

female, mean age 

= 24 who had undergoneanatomical 

repair/ reconstruction 

of the 

lateral ligaments 

of the ankle for 

chronic ankle instability. 

30 participants, 12 

female, median 

age – 27 who had 

undergone anatomical 

repair/ 

reconstruction of 

the lateral ligaments 

of the ankle 

for chronic ankle 

instability. 


It appears that only Karlsson et al (1995, 1999) have 

considered the benefits of a programme of early mobilisation 

versus immobilisation following surgical intervention. 

These two studies offer some preliminary evidence 

in favour of what should be regarded as cautious early 

mobilisation (see Table 1 for details of the studies). 

In a review of the literature, Lynch & Renstrom (1999) 

consider both conservative and surgical treatment options 

for ligament rupture in the ankle. They describe 

their practice including surgical procedure and postoperative 

rehabilitation, which includes immobilisation 

for 7-10 days in a plaster cast, followed by the use of a 

walking boot that allows 0 to 20 degrees of ankle plantar 

flexion. At three weeks passive plantar and dorsi 

flexion, exercises are commenced before gradually progressing 

to active movement. At six weeks post surgery, 

muscular and proprioceptive training are commenced 

(Lynch & Renstrom 1999). No scientific justification 

is offered for this protocol by these authors. Even 

though Karlsson et al (1995, 1999) have presented evidence 

in favour of early mobilisation, it appears that 

immobilisation for up to six weeks followed by a programme 

of rehabilitation remains the intervention of 

choice (Bell et al 2005). Now nine years following the 

publication by Karlsson et al, there is no further published 

evidence in the peer reviewed literature to challenge 

this persistent cautious approach to rehabilitation. 

i) Immobilisation group; plaster 

cast for 6 weeks post op. 

ii) Early ROM training in a 

Walker-Boot combined with 

a programme of supervised 

training from week 3. 

From 6 weeks both groups 

underwent the same rehabilitation 

programme 

i) Immobilisation group; plaster 

cast for 6 weeks post-op. 

ii) Mobilisation group: plaster 

cast for 7-10 days post-op, 

followed by an air-cast ankle 

brace for up to 6 weeks 

combined with controlled 

ROM training from week 3 

and strength training from 

week 5. 

Between weeks 7 to 12 both 

groups underwent the same 

rehabilitation programme. 

i) Karlsson functional ankle score: 

Rated as satisfactory in 80% of immobilisation 

group and 95% in early mobilisation 

group. 

ii) Mechanical instability: No significant 

differences between groups measured 

by anterior talar translation and talar tilt. 

iii) Sick leave: Immobilisation group = 

8.5 +/- 1.8 weeks. Early mobilisation 

group = 6.5 +/- 1.6 weeks. 

iv) Return to sport: Immobilisation group 

= 12.5 +/- 2.6 weeks. Early mobilisation 

group = 9.5 +/- 2.2 weeks. 

i) Karlsson functional ankle score: 

Rated as satisfactory in 80% of immobilisation 

group and 93% in early mobilisation 

group but this difference was not 

statistically significant. 

ii) Mechanical instability: No significant 

differences between groups measured 

by anterior talar translation and talar tilt. 




This case study is a personal account which details my 

experience of undergoing a modified Broström procedure. 

The issue of adherence is discussed before considering 

why the path of an intentional non-adherer was 

chosen. 

The Case 

A 31 year old male complains of a 13 year history of 

recurrent giving way of the right ankle following a severe 

inversion injury whilst playing football. Episodes of 

giving way were occurring with increased frequency 

over time, sometimes on a daily basis, with no worthwhile 

response to conservative care. 

Clinical examination demonstrated excessive talar tilt 

with subluxation of the ankle during testing. Stress xrays 

suggested excessive but equal levels of talar tilt 

bilaterally. MRI results were reported as rupture of the 

calcaneofibular (CCFL) ligament on the right but preservation 

of the anterior talofibular ligament (ATFL). This 

report was in conflict with a previous ultrasound scan 

which had suggested rupture of both the ATFL and 

CCFL. 

Based upon the clinical examination in combination with 

the report of frequent giving way, a modified Broström 

procedure was proposed. It was recommended that this 

be followed by six weeks of immobilisation, initially two 

weeks non-weight bearing, prior to commencing a programme 

of supervised rehabilitation. The procedure 

was carried out on the 27th August 2008 and following 

this, a plaster cast was applied to immobilise the ankle. 

The immediate post-operative period was uneventful. 

Analgesics were required until day two post-operatively, 

but at this point they were discontinued. 

During the period day two to day five following surgery, 

a progressive return to full weight bearing was achieved 

within the limits of pain. Range of movement (ROM) 

exercises were commenced within the confines of the 

plaster cast and as pain allowed. On day nine, the sutures 

were removed and the cast was replaced by an 

Air cast pneumatic walking brace. This brace maintains 

the foot/ankle in an immobilised position but is removable 

to allow washing. My belief was that this level of 

immobilisation was unnecessary and thus instead used 

a simple method of strapping to prevent further injury 

whilst I gradually increased weight bearing tolerance 

and ROM. ROM exercise was performed frequently 

throughout each day in all directions, including inversion, 

governed by the symptom response of pain produced 

with movement which was no worse upon cessation 

of that movement. This amounted to approximately 

8 to 10 repetitions repeated four to six times daily. By 

day 14, I felt sufficiently confident to continue usual activities 

of daily living without external support. By day 

21, I had resumed running and during the period day 21 

-28, managed three runs totaling 12 miles (two miles, 

four miles then 


six miles). The subjective feeling of ‘instability’ was no 

longer present during activities of daily living. By week 

six, I had resumed pre-operative levels of activity, which 

included three runs per week totalling approximately 18 

miles, and had suffered no pain or further episodes of 

giving way. At this point I returned for surgical review 

where, following examination, the surgeon felt that the 

ankle was now clinically stable. 

Discussion 

Adherence has been defined as; ‘the extent to which 

patients follow a prescribed treatment regime or lifestyle 

advice’ (Carr 2001, p.647). It has been suggested that 

non-adherence is a major problem in health care (Wroe 

2002). Two types of non-adherence are identified: 

�� Intentional non-adherence, e.g. altering the treatment 

regime to suit personal needs (Wroe 2002). 

�� Unintentional non-adherence, e.g. forgetting 

or misinterpreting treatment regimes (Carr 2001). 

Intentional non-adherence involves a rational decision 

made by the patient and is associated with the individuals’ 

beliefs and cognitions (Wroe 2002). I can accept 

that non-adherence may be a problem in health care 

when the prescribed treatment is likely to bring about 

recovery and is derived from available evidence. However, 

in my situation following a careful review and appraisal 

of the literature it was clear to me that immobilisation 

was unlikely to be the most effective means of 

recovery. 

It appeared logical when considering an anatomical repair, 

as in this situation, where existing tissue is repaired 

that theory relating to the biological process of 

soft tissue healing can be applied successfully as with 

non-operative rehabilitation. So, in this situation the rationale 

for my decision was based initially upon my understanding 

that repairing tissue passes through the 

following 3 stages: 

0-5 day’s Inflammatory phase 

5-21 day’s Repair/ Proliferation phase 

21 day’s onwards Remodelling phase 

(See McKenzie & May (2000) for a discussion of the 

implications of each of these stages.) 

This decision of early mobilisation was also grounded 

upon a personal belief that graded movement is appropriate 

and necessary in line with Wolff’s law which may 

be interpreted as suggesting: 

�� Form = Function (McKenzie & May 2000) 

Due to my background and experience, I recognise that 

this case of intentional non-adherence may only apply 

to a tiny majority of the patients we encounter. However, 

I feel that the real message here is to health 




care professionals who continue to apply post-operative 

treatment protocols without challenge. If we strive to 

become effective practitioners then we must be capable 

of offering therapy that is actually therapeutic. Again, I 

understand the dilemmas that challenging these protocols 

raises and the reasons for caution, e.g. fear of doing 

wrong, fear of litigation, ethical issues, professional 

conflict etc. However, we are all governed by one rule: 

�� Do no harm (Grieve 1991) 

Based upon the preliminary evidence presented and in 

combination with knowledge of the negative effects of 

immobilisation (see Table 2), it is imperative that we 

challenge current practice when it is clear that there is 

no adequate scientific basis. It appears logical based 

upon current evidence that there is mounting support for 

early mobilisation/loading in all situations where tissue 

repair and remodelling is required. 

Conclusion 

This case report has detailed the journey of an intentional 

non-adherer following modified Broström repair to 

treat chronic lateral ankle instability. A successful and 

accelerated outcome was achieved despite nonadherence 

to the prescribed post-operative treatment 

protocol. Although the limitations of a single case study 

should be recognised, it is hoped that this report may 

serve as a stimulus to health care professionals and 

researchers alike to challenge existing treatment pathways 

which may lack a reasoned basis and actually be 

delaying or restricting optimal recovery. 

References 

Bahr R, Pena F, Shine J, Lew W, Tyrdal S, Engebretsen 

L (1997). Biomechanics of ankle ligament reconstruction. 

The American Journal of Sports Medicine 25. 

424-432. 

Table 2 Negative effects of immobilisation 

Study Effects of immobilisation 

Bloomfield (1997) 1. Decreased bone mineral density 

2. Decreased muscle mass 

3. Decreased muscle strength 

Hayashi (1995) 1. Decreased cross-sectional area of tendon 

2. Decreased tensile strength of tendon 

Woo et al (1987) 1. Decreased cross-sectional area of ligament 

2. Decreased tensile strength of ligament 

Karlsson et al (1999) 1. Joint stiffness 

2. Increased absence from work 

3. Increased absence from sport 


Bell S, Mologne T, Sitler D, Cox J (2006). Twenty-six 

year results after Broström procedure for chronic lateral 

ankle instability. The American Journal of Sports Medicine 

34. 975-978. 

Bell S, Walthour S, Provencher M, Sitler D (2005). 

Chronic lateral ankle instability: The Broström procedure. 

Operative Techniques in Sports Medicine 13. 176- 

182. 

Bloomfield S (1997). Changes in Musculoskeletal Structure 

and Function with Prolonged Bed Rest. Medicine 

and Science in Sports and Exercise 29. 197-206. 

Brand R, Collins M, Templeton T (1981). Surgical repair 

of ruptured lateral ankle ligaments. The American Journal 

of Sports Medicine 9. 40-44. 

Carr A (2001). Barriers to the effectiveness of any intervention 

in OA. Best Practice & Research Clinical Rheumatology 

15. 645-656. 

De Vries J, Krips R, Sierevet I, Blankevoort L, van Dijk 

C (2006). Interventions for treating chronic ankle instability 

(review). Cochrane Database of Systematic Reviews 

2006, Issue 4. Art. No.: CD004124. DOI: 

10.1002/14651858.CD004124.pub2. 

Fujii T, Kitaoka H, Watanabe K, Luo Z, An K (2006). 

Comparison of Modified Broström and Evans procedures 

in simulated lateral ankle injury. Medicine & Science 

in Sports & Exercise 38. 1025-1031. 

Grieve G (1991). Mobilisation of the Spine (5th edition). 

Churchill Livingstone, London. 

Hayashi K (1996). Biomechanical Studies of the Remodelling 

of Knee Joint Tendons and Ligaments. Journal 

of Biomechanics 29. 707-716 




Hennrikus W, Mapes R, Lyons P, Lapoint J (1996). Outcomes 

of the Chrisman-Snook and Modified-Broström 

procedures for chronic lateral ankle instability. The 

American Journal of Sports Medicine 24. 400-404. 

Kannus P, Renstrom P (1991). Treatment for acute 

tears of the lateral ligaments of the ankle. The Journal 

of Bone and Joint Surgery 73. 305-312. 

Karlsson J, Bergsten T, Lansinger O, Peterson L 

(1988). Reconstruction of the lateral ligaments of the 

ankle for chronic lateral instability. The Journal of Bone 

and Joint Surgery 70. 581-588. 

Karlsson P, Bergsten T, Lansinger O, Peterson L 

(1989). Surgical treatment of chronic lateral instability of 

the ankle joint. The American Journal of Sports Medicine 

17. 268-274. 

Karlsson J, Lundin O, Lind K, Styf J (1999). Early mobilization 

versus immobilization after ankle ligament stabilization. 

Scandinavian Journal of Medicine & Science in 

Sports 9. 299-303. 

Karlsson J, Rudholm O, Bergsten T, Faxen E, Styf J 

(1995). Early range of motion training after ligament 

reconstruction of the ankle joint. Knee Surgery, Sports 

Traumatology, Arthroscopy 3. 173-177. 

Liu S, Baker C (1994). Comparison of lateral ankle ligamentous 

reconstruction procedures. The American 

Journal of Sports Medicine 22. 313-317. 


Lynch S, Renstrom P (1999). Treatment of acute lateral 

ankle ligament rupture in the athlete; conservative versus 

surgical treatment. Sports Medicine 27. 61-71. 

McKenzie R, May S (2000). The Human Extremities 

Mechanical Diagnosis and Therapy. Spinal Publications, 

New Zealand. 

Ng Z, Das De S (2007). Modified Broström-Evans- 

Gould technique for recurrent lateral ankle instability. 

Journal of Orthopaedic Surgey 15. 306-310. 

Pijnenburg A, Van Dijk C, Bossuyt P, Marti R (2000). 

Treatment of ruptures of the lateral ankle ligaments: a 

meta-analysis. The Journal of Bone and Joint Surgery 

82. 761-773. 

Woo S, Gomez M, Sites T, Newton P, Orlando C, 

Akeson WH (1987), The Biomechanical and Morphological 

Changes in the Medial Collateral Ligament of the 

Rabbit after Immobilization and Remobilization. The 

Journal of Bone and Joint Surgery 69. 1200-1211. 

Wroe, A (2002). Intentional and Unintentional nonadherence: 

A study of decision making. Journal of Behavioural 

Medicine 25. 355- 


Restless leg syndrome vs. lumbar derangement 


Mark Werneke, MPT, Dip. MDT 









Initial Physical Therapy MDT Evaluation (9/26/08) 

Therapist Clinical Assessment: 

The provisional MDT classification, based on differential 

physical therapy MDT diagnosis, was derangement 

(McKenzie and May 2003). Typically derangement can 

be identified by centralization or by lasting decreases in 

symptom intensity elicited during repeated movements. 

For this case, the patient reported no pains or concordant 

symptoms prior to, during, or after repeated movements. 

The only symptom was end range stretch complaint 

during repeated extension in lying. The patient 

was classified as a derangement based on clinical reasoning, 

utilizing both the patient’s history reporting sustained 

and frequent forward bending at work, and at 

home as well as a positive mechanical response or an 

increase in lumbar extension noted after repeated extension 

in lying movements. The principle of treatment 

was extension; directional preference was based on an 

increase in ROM (a mechanical response rather than a 

symptomatic one). 

Patient Self-Report Health Assessment: 

Patient’s function and quality of life issues were assessed 

by Focus On Therapeutic Outcome (FOTO) 

tool. (Swinkels, van den Ende et al 2007; Deutscher, 

Hart et al 2008). See Figure 1: Functional Status Intake 

Summary. Briefly, the patient’s functional status score 

was 56/100. The higher the score on the scale from 0- 

100 equates to higher levels of function. In addition, 

although not shown in Figure 1, the patient reported 

mild levels of depression and somatization (Dionne 

2005) measured by SCL-90R subscales and low levels 

of fear of physical and work activities measured by 

Waddell’s fear avoidance questionnaire (Waddell, Newton 

et al 1993). It is important to note that FOTO predicts 

both the level of functional improvement (i.e. 10 

points for this case study) and number of treatment visits 

(i.e. seven visits) expected from rehabilitation utilizing 

10 risk adjusted criteria listed on the top right hand 

corner in Figure 1. Although explaining risk adjustment 

is beyond the scope of this case report, risk adjustment 

allows the therapist the ability to compare patient’s outcomes 

with a national aggregate for more meaningful 

interpretation of clinical outcomes. 

Treatment Day 1: 

The patient was educated on the importance of posture 

using the slouch-overcorrect procedure and proper use 

of a lumbar roll. The patient was instructed to avoid prolonged 

sitting positions of >45 minutes and to take frequent 

stretch breaks at work (i.e. REIS 10 reps hourly). 

The patient was also instructed to avoid forward trunk 

bending especially after sitting or during the first few 

hours of each day. Frequent extension in standing and 

lying exercises were prescribed. In addition, the patient 

received the “Treat Your Own Back” booklet. 


Treatment 2nd Visit (10/3/08): 

The patient was reassessed one week later. The patient 

reported 75% improvement with decreased pulsating 

knee sensations when lying or sitting still. In addition, 

patient independently stopped her medications, which 

were prescribed for Restless Leg Syndrome. Patient 

reported using the lumbar roll as instructed and performed 

REIS hourly at work. Patient reported that she 

preferred to sustain prone on elbows at home vs. performing 

REIL as previously prescribed. Patient reported 

less frequent extension in lying stretches because of 

arm fatigue. Upon mechanical reassessment, the patient’s 

baseline extension ROM had improved since her 

initial evaluation. Her repeated movement reassessment 

on the 2nd visit confirmed the initial diagnosis and 

treatment principles of extension were continued. The 

patient was asked to call her physician regarding her 

change in medication use. The patient was rescheduled 

for a two week follow-up appointment and encouraged 

to call if she had any questions over that time or if her 

symptoms changed. 

Treatment 3rd visit (10/13/08): 

Patient now reported 98% better. She reported that her 

pulsating knee symptoms were no longer experienced 

when lying down after extended periods and that she no 

longer had difficulty falling asleep. The only time she felt 

concordant symptoms was when she sat too long, however 

performing REIS immediately abolished symptoms. 

Her mechanical assessment was normal; full asymptomatic 

lumbar extension ROM was observed. Prophylactic 

training was covered in detail to empower patient 

in self care principles. In addition, patient’s outcomes 

using the FOTO tool were reassessed. See Figure 

2-Functional Status Summary. The patient’s functional 

score was now 84/100; a statistically and clinically 

important improvement of 28 points. Patient achieved 

this improvement over three treatment sessions. Based 

on MDT treatment, the patient’s outcomes were more 

effective and efficient compared to the national aggregate’s 

predicted mean functional change score of 10 

points over seven visits. 

Patient contact (11/4/08): 

Patient reports her symptoms have resolved and further 

therapy appointments are not required. Patient reports 

she will call if problems arise. 

Discussion 

Restless Leg Syndrome vs. Derangement 

I found this case study interesting because not only was 

this the first patient that I evaluated and treated whose 

referring medical diagnosis was Restless Leg Syndrome, 

but also the patient’s primary presenting symptoms 

of pulsating anterior thigh and knee sensations 

were atypical. The case study highlights the importance 

of using a validated classification approach for clinical 

diagnosis and intervention to optimize patient outcomes 

and quality of care. 




Restless Leg Syndrome (RLS) is considered a neurological 

sensorimotor impairment (Oertel, Trenkwalder et 

al 2007) (Allen, Walters et al 2005). The prevalence for 

RLS varies between 5-10%. Of interest, RLS is diagnosed 

2x more commonly in women compared to men, 

and Caucasians are more prone to RLS than African- 

Americans. There is evidence that RLS can be worsened 

by surgery of any kind. 

There are four hallmark features or criteria for diagnosis 

of RLS (Allen, Walters et al 2005): 

1. an urge to move that is associated with 

abnormal sensations in the legs 

2. symptoms are worsened by rest 

3. symptoms are relieved by movement 

4. symptoms are most severe at night 

Once a patient is classified or diagnosed with RLS, the 

non-pharmacological treatment approach consists of 

sensory stimuli to the affected areas with rubbing or 

cooling etc, encouraging the patient to walk or exercise, 

and psychological training. One study conducted recently 

by the London College of Osteopathic Medicine, 

London, UK reported that specific types of lumbar manipulation 

reduces excessive sensory input and relieves 

symptoms of RLS (Peters 2001). 

After the exclusion of sleep disorders, Parkinson’s disease 

and individuals with iron deficiency, pregnancy, 

and end-stage renal disease, the clinical diagnosis of 

restless leg syndrome is primarily based on patient self 

report. Confirming a diagnosis based solely on patient’s 

self report without performing a thorough clinical examination 

is problematic. In addition, the hallmark features 

and criteria for accurate classification of patients with 

RLS have not been extensively researched and well 

designed studies are required to identify validated interventions 

for RLS as well as to determine if RLS diagnosis 

is medically warranted or necessary. 

Using validated treatment based classification systems 

to improve patient care has been recently recommended 

(Rossignol, Arsenault et al 2006) (Koes, van 

Tulder et al 2006). 

One classification method receiving increased attention 

in the literature is the McKenzie method (Clare, Adams 

et al 2004). The differential diagnosis by MDT methods 

has been reported to be advantageous because the 

approach examines patient’s self report of symptoms in 

response to repeated trunk movement tests and loading 

strategies (Wetzel and Donelson 2003) and this examination 

approach has been shown to be a reliable and 

valid method for lumbar impairment assessment and 

treatment (Aina, May et al 2004) (Donelson, Aprill et al 

1997) (Long, Donelson et al 2004). 


Clinical Reasoning and Outcome Documentation 

In the case study presented, the patient’s demographics 

and subjective history matched the criteria for diagnosing 

Restless Leg Syndrome. The patient reported that 

exercises which were prescribed in prior non-MDT 

physical therapy treatments (June 2008) for her hip and 

back resolved her back and hip pain complaints, however 

the exercises did not affect her ongoing pulsating 

thigh and knee symptoms. In addition, patient reported 

that her RLS medication was not helpful. Based on 

physical therapy differential diagnosis and clinical reasoning 

using McKenzie methods, the patient was classified 

as a derangement and the patient’s symptoms 

quickly resolved with MDT educational principles and 

patient empowerment in self care as well as prescribing 

extension as the directional preference for movements. 

Patient stopped her RLS medications within the first 

week of MDT treatment. Although a case study design 

can not validate the efficacy of treatment rendered for 

this patient referred to therapy with Restless Leg Syndrome, 

the effectiveness and efficiency of the patient’s 

classification and subsequent treatment was supported 

by outcome assessment using a published and psychometrically 

strong outcome measurement tool (Hart, 

Mioduski et al 2006). 

References: 

Aina, A., S. May, et al (2004). "The centralization phenomenon 

of spinal symptoms--a systematic review." 

Man Ther 9(3): 134-43. 

Allen, R. P., A. S. Walters, et al (2005). "Restless legs 

syndrome prevalence and impact: REST general population 

study." Arch Intern Med 165(11): 1286-92. 

Clare, H. A., R. Adams, et al (2004). "A systematic review 

of efficacy of McKenzie therapy for spinal pain." 

Aust J Physiother 50(4): 209-16. 

Deutscher, D., D. L. Hart, et al (2008). "Implementing an 

integrated electronic outcomes and electronic health 

record process to create a foundation for clinical practice 

improvement." Phys Ther 88(2): 270-85. 

Dionne, C. E. (2005). "Psychological distress confirmed 

as predictor of long-term back-related functional limitations 

in primary care settings." J Clin Epidemiol 58(7): 

714-8. 

Donelson, R., C. Aprill, et al (1997). "A prospective 

study of centralization of lumbar and referred pain. A 

predictor of symptomatic discs and anular competence." 

Spine 22(10): 1115-22. 

Hart, D. L., Mioduski JE, Werneke MW. (2006). 

"Simulated computerized adaptive test for patients with 

lumbar spine impairments was efficient and produced 

valid measures of function." J Clin Epidemiol 59(9): 947- 

56. 




Koes, B. W., M. W. van Tulder, et al (2006). "Diagnosis 

and treatment of low back pain." BMJ 332(7555): 1430- 

4. 

Long, A., R. Donelson, et al (2004). "Does it matter 

which exercise? A randomized control trial of exercise 

for low back pain." Spine 29(23): 2593-602. 

McKenzie, R. and S. May (2003). The Lumbar Spine: 

Mechanical Doagnosis and Therapy. Waikanae, New 

Zealand, Spinal Publication, LtD. 

Oertel, W. H., C. Trenkwalder, et al (2007). "State of the 

art in restless legs syndrome therapy: practice recommendations 

for treating restless legs syndrome." Mov 

Disord 22 Suppl 18: S466-75. 

Peters, TW (2001). Restless Legs. Osteopathy Today. 

October. 


Rossignol, M., B. Arsenault et al (2006). Clinic on Low- 

Back Pain in Interdisciplinary Practice (CLIP) Guidelines. 

M. P. H. Department. Montréal, Canada, Agence 

de la sante et des services sociaux de Montréal. 

Swinkels, I. C., C. H. van den Ende, et al (2007). 

"Clinical databases in physical therapy." Physiother 

Theory Pract 23(3): 153-67. 

Waddell, G., M. Newton, et al (1993). "A Fear- 

Avoidance Beliefs Questionnaire (FABQ) and the role of 

fear-avoidance beliefs in chronic low back pain and disability." 

Pain 52(2): 157-68. 

Wetzel, F. T. and R. Donelson (2003). "The role of repeated 

end-range/pain response assessment in the 

management of symptomatic lumbar discs." Spine J 3 

(2): 146-54. 




Figure 1 

Functional Status Intake Summary 



Figure 2 

Functional Status Summary 







Karrie Giger, PT, DPT, Cert. MDT 









Introduction 

Determining the underlying cause of anterior knee pain, 

when medically diagnosed as Osgood-Schlatter Disease 

(OSD), can be difficult because of the differential 

diagnostic process that is required to identify a conclusion. 

Differential conditions to be considered could include, 

inflammation from trauma or infection, condromalacia, 

patella malalignment, or referred pain from the 

hip or back, as well as factors such as increased Qangle, 

excessive pronation of the subtalar joint, and 

tightness of the lower extremity muscles (McKenzie and 

May 2003, Calmbach et al 2003, Hartzman et al 1987). 

Therefore, a detailed history, structured physical, and 

focused neurological examination is necessary in order 

to narrow the possible causes. This will guide the 

evaluator to establish the correct physical therapy mechanical 

diagnosis and logically form a plan of care 

(Calmbach et al 2003, Visuri et al 2007). 

Although a patient’s symptom descriptors may be consistent 

with the medical diagnosis, when there is 

myotome weakness, complaints of sensory deficits, or 

symptoms in a radicular pattern, differential conditions 

should be considered and a neurological examination 

should be conducted (McKenzie and May 2003, Deyo et 

al 1992, Hampton et al 1975, Mosby 1980). When referred 

pain is suspected, the spine should be considered 

and adequately examined. Symptomatic complaints 

of pain referred from the spine are not always 

felt in the back (McKenzie and May 2003, Tippet 1994, 

Deyo 1992). Using the McKenzie standardized assessment 

process, a working hypothesis is formed during 

the history which will assist in guiding the objective examination. 

The purpose of this case report is to demonstrate the 

usefulness of the McKenzie Mechanical Diagnosis & 

Therapy examination process to assist in the differential 

diagnosis of OSD and referred lumbar pain. 

Case description 

The patient was a 21-year old male, who complained of 

bilateral knee pain. After 12 days of unrelieved symptoms, 

he visited his primary care physician who diagnosed 

him with suspected OSD. He was referred to 

physical therapy five days later with pain described as a 

“sharp throbbing ache”. The patient scored a 37% on 

the Lysholm Knee Pain Rating System indicating significant 

impairment with activities of daily living (ADL) 

(Table 1). He reported his symptoms commenced upon 

waking one morning, after working the prior evening 

and described sudden difficulty with ambulation and 

stair management. Since the patient described no 

mechanism of injury, further questioning was necessary 

to rule out the presence of any serious pathology and 

strongly suggested examining the lumbar spine for possible 

involvement. 


Tests and measures 

Neurological assessment of the patient’s myotomes and 

dermatomes were tested as described by Magee (1997) 

for L4, L5, and S1 (Table 1). Weakness and decreased 

sensation at the L4-L5 levels suggested that the lumbar 

region should be thoroughly assessed. 

Using the McKenzie Extremities assessment form, the 

lumbar spine was examined first, following the standardized 

method for the lumbar spine (McKenzie and 

May 2003). The lumbar spine was measured for movement 

loss by visual inspection to be either no movement 

loss, minimal, moderate, or major movement loss 

(McKenzie and May 2003). Baseline active range of 

motion (AROM) measurements and strength of the patient’s 

knees was also performed. A relevant spinal derangement 

with extension as the patient’s directional 

preference was concluded. This was confirmed when 

the patient was able to report abolition of pain in bilateral 

suprapatellae, decreased pain in bilateral anterior 

thighs and production of centralizing pain in his lower 

back following lumbar examination. 

Classification 

Several aspects of the patient’s history and examination 

lead to the lumbar spine as a possible cause of the patient’s 

reported knee pain. The patient’s history revealed 

an abrupt onset of constant and severe bilateral anterior 

thigh and suprapatellar pain (without trauma) that was 

worsening over the past 17 days, especially during 

work. The patient stated no previous episodes of knee 

pain. He did not complain of tenderness, display 

edema, or present with an enlarged prominence of the 

tibial tuberosity. 

The presentation of bilateral leg symptoms may be explained 

by a derangement at this level. It may be expected 

for cauda equina symptoms to be present in 

conjunction with such a derangement, but the patient 

denied having any, as noted in the history. After the 

examination was completed, it was hypothesized that 

the level of impairment was L4-L5, and the patient was 

classified as a central/symmetrical derangement with a 

directional movement preference of extension 


Prognosis 

The Guide to Physical Therapists Practice (Arena et al 

2001) suggests that patients diagnosed in the category 

of Impaired Joint Mobility, Motor Function, Muscle Performance, 

Range of Motion, and Reflex Integrity Associated 

with Spinal Disorders will demonstrate optimal improvement 

over the course of one to six months and 

that the expected range of physical therapy visits is 8 – 

24 during a single continuous episode of care. According 

to these guidelines, it is anticipated that 80% of the 

patients classified into this pattern will achieve the set 

goals and expected outcomes in this time frame. 




According to the definition of a derangement syndrome, 

movement is required to encourage reduction of the 

displaced articular tissue (McKenzie and May 2003). 

Furthermore, the process of centralization can often be 

achieved in days, rather than weeks, of exercise depending 

on patient adherence with the therapeutic intervention. 

The patient’s symptoms are expected to be 

rapidly reversible with lasting effects. The derangement 

should fully reduce, with symptoms abolished, and fullrange, 

pain-free movement regained (McKenzie and 

May 2003). The patient should have no functional limitations 

or impairments, and should be able to continue to 

work without restrictions. 

Several evidence based centralization studies report the 

ability to achieve this phenomena which is closely linked 

with a good prognosis and full recovery with appropriate 

non-operative care (Wetzel and Donelson 2003, 

Werneke et al 1999, Long et al 2004, Karas et al 1997). 

Since the patient demonstrated centralization at the 

initial examination with abolished suprapatellar pain and 

decreased bilateral anterior thigh pain, the time frame 

for the patient’s derangement to completely resolve was 

estimated to be no longer than two to four weeks. 

Intervention 

Visit 1. 

After the initial examination, the treatment included patient 

education of his mechanical diagnosis, postural 

correction, and avoidance of provocative movements/ 

activities. Using the extension principle, the patient was 

instructed to perform a specific home exercise program 

and taught how to self-modify it according to his pain 

response (McKenzie and May 2003). 

The patient was instructed to perform 10 repetitions of 

extension in lying (EIL) every two hours and as needed 

to control his pain and help continue to reduce, centralize 

or abolish his pain. The patient was issued a 

McKenzie lumbar roll to be used whenever sitting or 

driving to keep appropriate lumbar lordosis during these 

activities. 

Visit 2. 

The patient reported a pain rating VAS 7/10, but stated 

improvement, as his pain was now intermittent. He was 

no longer taking Ibuprofen because he was able to control 

his pain with the extension exercises, but was not 

able to abolish it every time. During the sessions, the 

patient reported that his attempt to further increase his 

lumbar spine extension with self-generated overpressure 

during EIL, achieved by exhaling at the end-range 

of the movement, centralized his pain. Bilateral knee 

and thigh pain was abolished and low back pain was 

produced. EIL with patient generated overpressure was 

performed again. With this attempt, his left thigh pain 

was abolished and his low back pain was better. In the 

standing position, the patient reported low back pain 

and the production of bilateral 


posterio-lateral knee pain, left greater than right, but 

decreased intensity VAS 5/10. His home program was 

EIL with self generated overpressure. 

Visit 3. 

The patient reported a significant improvement in pain. 

He reported central low back and lateral knee pain, left 

greater than right, VAS 1-2/10. His posture was notably 

improved. EIL with patient overpressure no longer had 

an effect on his pre-exercise symptoms. He reported 

bilateral lower extremity numbness to his feet. The 

head of the plinth was elevated until the patient was 

positioned at end-range lumbar extension and sustained 

for five minutes. He was able to report abolition 

of bilateral lower extremity numbness, but his back and 

lateral knees remained painful. The patient then performed 

EIL with clinician generated overpressure to L4- 

L5. This abolished his lateral knee pain and decreased 

his central low back pain and rated 1/10 on a VAS. The 

patient’s home exercises were EIL with patient generated 

overpressure or sustained EIL, if needed. 

Visit 4. 

The patient stated an overall improvement, but rated his 

central low back pain VAS 4/10 and his left anterior 

thigh/suprapatellar pain VAS 6/10. He had no pain in his 

right lower extremity. It was considered the patient may 

have a lateral component, now affecting only one lower 

extremity, as there was no further improvement of 

symptoms with force progressions in the sagittal plane. 

Consequently, lateral movement was attempted as an 

alternative. The patient was placed in the prone position 

with hips offset to the right (away from his pain). He performed 

10 repetitions of EIL in the offset position and 

reported a decrease in pain during the exercise, which 

resulted in centralization of pain to his low back VAS 

1/10. The patient was instructed to replace pure sagittal 

extension with the EIL with hips offset to the right. 

Visit 5. 

The patient reported no back or left anterior thigh/ 

suprapatellar pain. He stated that any returning pain 

was intermittent, low intensity VAS 1/10, and of short 

duration. If his left anterior thigh/suprapatellar pain recurred, 

he was able to perform EIL with hips offset, 

which abolished all of his pain after one set of 10 repetitions. 

The patient reported occasional low back pain 

only, but again stated he was able to perform EIL for 

one set of 10 and his pain would abolish. He no longer 

complained of any right anterior thigh or suprapatellar 

pain. During objective reassessment, the patient demonstrated 

good posture, no motor deficits, no sensation 

deficits, normal reflexes, negative slump test, and normal 

AROM of bilateral knees, with no report of pain and 

no movement loss of the lumbar spine (Table 1). At this 

point, it was not necessary to test any other movement 

preferences, because the 




patient demonstrated improvement with lasting effects 

from the current intervention. The patient was now 

ready to be instructed on preventative care to avoid recurrence. 

Discharge 

This patient received five physical therapy visits over a 

period of 12 days. The patient called one month after 

discharge to report that he had no pain or problems with 

any functional or work-related tasks or routine activities. 

He no longer reported back, anterior thigh or suprapatellar 

pain, and had not experienced recurrence of 

symptoms. He was verbally administered the Lysholm 

Pain Rating Test and his new score was 100 out of 100 

points indicating no impairment with ADLs (Table 1). 

Discussion 

Establishing the correct diagnosis is critical in order to 

generate appropriate management strategies to achieve 

most effective patient outcomes (Wilson et al 1999). 

Making a diagnosis using pattern recognition requires a 

well organized knowledge base and clinical experience 

that is only available to experts, but it is possible to be 

inaccurate without considering other differential diagnoses 

that may have similar complaints (McKenzie and 

May 2000). This patient’s pain was severe and continuous 

in nature, and included report of quadriceps tightness, 

possibly associated with anterior thigh discomfort, 

which increased with resisted knee extension. These 

symptoms seemed to “fit the pattern” of OSD (Visuri et 

al 2007, Gholve et al 2007, Rosenberg et al 1992). It is 

important to consider all significant information and proceed 

with the physical examination in a way that will 

allow the evaluator to either accept or refute the initial 

working hypothesis. It is not only important to consider 

the symptoms that “fit” the pattern, but also those that 

“do not fit” (Sackett 1992). 

It was hypothesized, by the author, that the level of impairment 

was L4-L5, causing the referred pain to bilateral 

anterior thighs and suprapatellae. Although this 

patient demonstrated an unusual presentation, the criteria 

of his medical diagnosis was less convincing than 

the objective display of positive neurological signs, 

which lead the author to first test the hypothesis of referred 

pain from the lumbar spine. The intervention applied, 

lumbar extension principle, allowed the patient to 

begin feeling the effects of centralization on the day of 

initial examination. The abolition of symptoms within five 

physical therapy visits, demonstrated that the correct 

therapeutic interventions were applied and therefore the 

initial hypothesis was accepted. 

The McKenzie Method of assessment has received significant 

attention in the literature and has been widely 

studied in randomized controlled trials and systematic 

reviews, of the lumbar spine (Wetzel et al 1980, Long et 

al 2004, Maher et al 1999, Koes et al 


1991). Studies need to be conducted using the 

McKenzie Method on peripheral complaints coexisting 

with spinal problems, none of which could be found during 

a search of literature by the author. Such studies 

might be useful in order to learn how many patients who 

have peripheral complaints actually have a relevant spinal 

problem. These studies may also provoke physical 

therapists to perform a thorough neurologic examination 

and spinal screening, which may elicit an immediate 

effect of the patient’s symptoms, rather than testing the 

peripheral area of complaint that is of uncertain origin 

and providing local treatment for symptomatic relief. 

Being able to confidently identify the correct source of 

symptoms at initial contact will allow the therapist to 

successfully manage the patient’s symptoms with appropriate, 

cost effective treatment that will result in 

faster recovery. 


Appreciation is extended to Dr. Michael Rabel, PT, 

MPT, DSc, my mentor, as well as my colleagues who 

took the time to peer review this case report. 

References 

Arena R, Ciccone C, Craik R, Delitto A, Crawford S, 

Fillyaw M, eds, (2001). Guide to Physical Therapist 

Practice. 2nd ed. Alexandria, VA: American Physical 

Therapy Association; No. 81. 

Calmbach W, Hutchens M (2003). Evaluation of patients 

presenting with knee pain: Part II. differential diagnosis. 

American Family Physician; 68( 5):917-922. 

Deyo R, Rainville J, Kent D (1992). What can the history 

and physical examination tell us about low back 

pain? JAMA 268(6): 760-765. 

Gholve, Purushottam, Scher, et al. (2007). Osgood- 

Schlatter syndrome. Current Opinion in Pediatrics;19 

( 144-50). 

Hampton J, Harrison M, Mitchele J, Prichard J, Sewmour 

C (1975). Relative contributions of history-taking, 

physical examination, and laboratory investigation to 

diagnosis and management of medical outpatients. BMJ 

2:486-489. 

Hartzman S, Reicher M, Bassett L, Duckwiler G, Mandelbaum 

B, Gold R (1987). MR imaging of the knee part 

II. Chronic disorders. Radiology 162:553-557. 

Karas R, McIntosh G, Hall H, Wilson L, Melles T (1997). 

The relationship between nonorganic signs and centralization 

of symptoms in the prediction of return to work 

for patients with low back pain. Physical Therapy; 77 

( 4):354-364. 

Koes B, Bouter L, Beckerman H, van der Heijden G, 

Knipschild P (1991). Physiotherapy exercises and back 

pain: a blinded review. BMJ 302.1572-1576. 




Long A, Donelson R, Fung T (2004). Does it matter 

which exercise? A randomized control trial of exercise 

for low back pain. Spine;29( 23):2593-2602. 

Magee D (1997). Physical Assessment. Third ed. W. B. 

Saunders Company, Philadelphia. 

McKenzie R, May S (2000). The Human Extremities: 

Mechanical Diagnosis & Therapy. Spinal Publications 

Ltd, Waikanae. 

McKenzie R, May S (2003). The Lumbar Spine Mechanical 

Diagnosis & Therapy. Vol two. Spinal Publications 

Ltd, Waikanae. 

Mosby C (1980). The importance of the history in the 

medical clinic and the cost of unnecessary tests. Annotations 

100(6):928-931. 

Rosenberg Z, Kawelblum M, Cheung Y, Beltran J, Lehman 

W, Grant A (1992). Osgood-Schlatter Lesion: Fracture 

or Tendinitis? Scintigraphic, CT, and MR imaging 

features. Radiology;(85):853-858. 

Table 1. Comparison of Baseline and Outcome Measures 


Sackett D (1992). A primer on the precision and accuracy 

of clinical examination. JAMA; 267(19)2638-2644. 

Tippett S (1994). Referred knee pain in a young athlete: 

a case study. J Orthop Sports Phys Ther 19(2):117-120. 

Visuri T, Pihlajamaki H, Mattila V, Kiuru M (2007). Elongated 

patellae at the final stage of Osgood-Schlatter 

Disease: A radiographic study. Science Direct;14 

( 3):198-023. 

Wilson L, Hall H, McIntosh G, Melles T (1999). Intertester 

Reliability of a Low Back Pain Classification System. 

Spine;24(3):248-253. 

Werneke M, Hart D, Cook D (1999). A descriptive study 

of the centralization phenomenon. Spine; (24):676-683. 

Wetzel F, Donelson R (2003). The role of repeated endrange/pain 

response assessment in the management of 

symptomatic lumbar discs. The Spine Journal;(3)146- 

154. 

Measure Baseline initial exam Discharge after 5 visits 

Pain intensity VAS 7/10 VAS 0/10 

Lysholm pain scale 37% significant impairment 0% no impairment 

Posture Poor Good 

Motor Bilateral quadriceps 4/5 

Left ankle dorsiflexion 3/5 

Left great toe extension 3/5 

Sensory Decreased sensation left anterior thigh, 

medial leg and great toe 

No deficits noted 

No deficits noted 

Reflexes Normal 2+ bilaterally Normal 2+ bilaterally 

Dural sign (Slump test) Positive bilaterally for reproduction 

of suprapatellar pain 

Knee ROM Right 0-125 degrees with pain 

Left 0-103 degrees with pain 

LS movement loss Flexion no movement loss, 

extension moderate movement loss, 

bilateral side gliding no movement loss 

Negative 

Bilaterally 0-130 degrees 

No movement loss with 

exception of minimal 

movement loss LS flexion 



Ron Schenk, PT, PhD, OCS, FAAOMPT, Cert. MDT 

The MDT Fellowship Program was developed in January, 

2008 to further the education of physical therapists 

in Mechanical Diagnosis and Therapy (MDT) and Orthopaedic 

Manual Physical Therapy (OMPT) practice and 

research. The program has been credentialed as a Fellowship 

program by the American Physical Therapy Association 

(APTA). This credential allows the program to 

become recognized as an approved Fellowship with the 

American Academy of Orthopaedic Manual Physical 

Therapists (AAOMPT), an organization whose mission 

is to promote manual physical therapy advocacy, research, 

and clinical practice. Completion of an approved 

Fellowship earns the designation of Fellow of the 

AAOMPT (FAAOMPT). 

The AAOMPT was developed to become the United 

States member organization within the International 

Federation of Orthopaedic Manual Therapists 

(IFOMPT). In 1992, Freddy Kaltenborn convened a 

meeting of directors of orthopaedic manual physical 

therapy residency programs. The founding members of 

the AAOMPT included Richard Erhard, Kornelia Kulig, 

Joe Farrell, Ola Grimsby, Michael Moore, Stanley Paris, 

Michael Rogers, and Bjorn Swendson. Although these 

founding members represented somewhat different 

clinical approaches, they shared a common goal of developing 

standards for evaluating competency in musculoskeletal 

care. As the AAOMPT evolves, its members 

continue to demonstrate a remarkable degree of 

professional collaboration and dissemination of information 

to promote the organizational mission. MDT, with its 

foundation in evidence based practice, deserves representation 

in such national and international organizations. 

The MDT Fellowship program’s membership and 

involvement in these organizations will allow for increased 

visibility, thereby fostering improved understanding 

and appreciation of the Method. 

The MDT Fellowship program is designed to promote 

problem solving and psychomotor skill development in 

the management of musculoskeletal disorders of the 

spine and extremities. The program builds upon the 

education provided through the Diploma program. The 

following describes how a physical therapist may proceed 

through the MDT Fellowship program. 

PATHWAY TO MDT FELLOWSHIP 

��Complete MDT Courses A-E 

��Successfully complete MDT certification examination 

(written and practical) 

��Complete PHTX 544 (300 hours) Mechanical Diagnosis 

and Therapy through Otago University 

��Complete 400 hour Clinical Residency through MII 

(USA) Clinical Sites 

��Successfully complete McKenzie Diploma MDT 

Final Examinations 

��McKenzie Diploma in MDT 

� Completion of 130 hours direct 1:1 mentorship experience 

with Fellow of American Academy of Orthopaedic 

Manual Physical Therapists (AAOMPT) 1 

��Completion of 310 hours of Clinical Problem Solving 

directed by a Fellow 2 

� Completion of thrust manipulation course at 

Daemen College taught by Ron Schenk PT, PhD, 

FAAOMPT, Cert. MDT 3 

� Apply for Fellow Status in AAOMPT 4 

1 Satisfactory proficiency in the Clinical Component of 

the OMPT residency is determined through 130 hours 

of direct 1:1 mentorship provided by a program approved 

Fellow of the AAOMPT. The mentor must be at 

least Cert. MDT to guide the candidate through this experience. 

2 The problem solving component involves completion of 

case studies and clinical decision making activities directed 

by a Fellow who holds the credentials indicated 

above. The problem solving experiences may be completed 

during the candidate’s own clinical practice time 

via distance learning communication between the candidate 

and the Fellow mentor. 

3 The three day thrust manipulation course involves instruction 

in the history of thrust manipulation, understanding 

of the terminology and evidence pertinent to 

this area of clinical practice, and the development of 

psychomotor skill in safe and appropriate delivery of the 

intervention. A practical examination will be administered 

to the candidate following completion of the 

course. The course will be held at Daemen College and 

a DVD of the techniques and access to College library 

services will be given to the course registrants who are 

MDT Fellow candidates. 

4 

Successsful completion of all components of the pathway 

will allow the candidate to earn Fellow status in the 

AAOMPT through completion of an application. 

In summary, the Mission of the MDT Fellowship Program 

is to develop and provide licensed physical therapists 

an organized and guided educational pathway to 

develop advanced skills in the understanding and application 

of the principles MDT and subsequently, OMPT. 

The program has been endorsed by Robin McKenzie 

who maintains that becoming a recognized educational 

program within the APTA, AAOMPT, and IFOMT, will 

result in increasing world-wide recognition of the 

McKenzie Method, its demonstrable clinical effectiveness, 

and its efficacy (see following letter from Robin). 

We encourage those Diplomats interested in the program 

to contact the MDT Fellowship Program Director, 

Ron Schenk at rschenk@daemen.edu or 716-839-8360. 








Research Foundation 

International Mechanical Diagnosis and Therapy Research Foundation 

In 2007, the meeting of the International Mechanical 

Diagnosis and Therapy Research Foundation (IMDTRF) 

was held in Washington, DC in the United States. This 

meeting was the second of the Foundation Board of 

Directors (Ted Dreisinger, Uffe Lindstrom, Stephen 

May, Betty Sindelar, and Mark Werneke). The origin of 

IMDTRF was the McKenzie International Research 

Committee. The Foundation was organized as an independent 

entity with a mission to fund research projects 

related to Mechanical Diagnosis and Therapy (MDT). 

During the first two years of the Foundation’s existence, 

the Board created the legal framework and the policies 

and procedures to receive, review, fund, and administer 

submitted research proposals. No monies earmarked 

for research were used during this time period. 

The funding of the first research project in 2007 was 

met with excitement by the Board. This accomplishment 

highlighted the interaction of the Foundation with the 

MDT community: a call for proposals was answered by 

a fundable research project. And all of this was completed 

in the designated time-frame! The funded project 

is titled “The effectiveness of MDT in patients who meet 

the clinical prediction rule for spinal manipulation” and 

the principal investigator is Ronald Schenk with coinvestigator 

Carol Dionne. 

An important objective from that meeting in 2007 was to 

reach out to the international community. To this end, 

Uffe, Stephen, and Ted attended the Euro meeting in 

Prague in June of this year to up-date the faculty about 

the activities of the Foundation and to seek donations to 

support the work of the Foundation to fund research 

projects. The message was well received by the faculty 

and branches and we are pleased to have received 

some monies from that effort. In addition, Ted represented 

the Foundation at the Conference of the Americas 

in Orlando, Florida in July to up-date this audience 

on Foundation activities and to seek donations here as 

well. As with the Euro meeting, the message was well 

received by this audience, and some monies have been 

received from that effort. 

The Foundation is seeking donations to increase its 

financial base so that more projects can be funded 

yearly and that the maximum grant awarded can increase. 

As in any new venture, there are costs and logistics 

that must be addressed for maintenance, growth 

and development. We understand the importance of 

stewardship of our resources. To that end, administrative 

costs for running the Foundation represent approximately 

7% of our present funds. Since our Board of Directors 

is completely voluntary, these expenses reflect 

direct costs only for accounting, mailings and our annual 

meeting. 

Growth for us requires two things: raising of more funds 

to support projects and attraction of proposals that we 

can fund. Growth also occasionally requires adjustments 

in board make up. To that end, this year, the 

Board of Directors has increased its number. We are 

pleased that Mary Sheid has joined the Board. Each 

member of the Board has specific tasks to oversee and 

Mary has come to help direct our fund-raising efforts. 

Mary is a practicing physical therapist with fund-raising 

experience. She brings to our group insight, energy and 

experience in this area. We are delighted to have her as 

a part of our Board. For more information you can see 

her bio on our website. 

November 7-9, 2008 was the third meeting of the Foundation 

Board of Directors in Copenhagen, Denmark. A 

report on this meeting will follow later. 

Submitted on behalf of the IMDTRF Board, 

Thomas E. Dreisinger, PhD - Chair 


Upcoming Institute Conferences 

Upcoming Events 



International Journal of MDT 

Information for Authors 

Please submit an electronic version of your article to 

the Editor in Chief, Helen Clare at 

clare.ha@bigpond.com. 

Your article will be reviewed by two (2) members of the 

Editorial Review Board and a decision with feedback 

will be given to you within one month. Reviewing Board 

members will act as mentors and provide advice and 

suggestions to improve submissions and will contact 

authors directly. There will be a maximum of two (2) 

review submissions. Authors whose material is included 

in the IJMDT will be provided a complimentary 

copy of the issue in which it appears. 

All work submitted to the IJMDT will be reviewed for 

scientific content, appropriateness to Mechanical Diagnosis 

and Therapy, relevance, clarity and presentation. 

The decision will be one of the following: 

�� accept 

�� revise 

�� reject - resubmission possible with major edits 

�� reject 

Word count 

Eleven pages maximum 

Presentation 

Your article should be double-spaced with 12 pt, Arial. 

Title Page 

Give title, author, author details 

Keywords 

Up to 4 keywords (if appropriate) 

Abstract 

A structured abstract of no more than 200 words 

Text 

One sort of sub-heading should be used: 

�� bolded in lower case. 

Do not use abbreviations at all. Do not use local acronyms 

unless they are fully spelled out initially. 

Types of Articles 

The IJMDT welcomes primary research papers, topical 

reviews, reviews of articles and case studies. See the 

following guidelines for submitting manuscripts. 

Primary Research Guidelines 

IJMDT is particularly interested in publishing original 

primary research. Papers should be as follows: 

�� title page (title, author, author details) 

�� abstract / key words 

�� introduction 

�� methods 

�� results 

�� discussion 

�� conclusion 

�� references 

�� tables 

Topical Review Guidelines 

Reviews on appropriate and relevant topics are also 

welcome. Papers should be as follows: 

�� title page (title, author, author details) 

�� abstract / key words 


�� methods (if a systematic review) 

�� results 

�� discussion 



�� tables 

Review of article guidelines 

Reviews of individual articles. Papers should be as follows: 

�� title 

�� objectives 

�� design 

�� setting 

�� patients 

�� intervention 

�� primary outcome measure 

�� main results 


The International Journal of Mechanical Diagnosis and Therapy ® (IJMDT) is a collaborative effort of the worldwide 

branches of The McKenzie Institute ® International (MII) emphasizing scientific study, clinical relevance and education 

related to Mechanical Diagnosis and Therapy ® (MDT). 

Submission 

deadline: 

Jan 15 May 15 Sep 15 

Issue date: Mar Jul Nov 

�� conclusions 

�� comment (include implications for MDT) 






�� 

Editor 

Helen Clare, PhD, FACP, MAppSc, 

Dip Physio, Dip. MDT 

�� 

Editorial Review Board 

Helen Clare, PhD, FACP, MAppSc, 

Dip Physio, Dip. MDT 


Stuart Horton, PT, Dip. MDT, MPhty 

stuart.horton@otago.ac.nz 

Sinikka Kilpikoski, MSc, PT, Dip. MDT 

sinikka.kilpikoski@kolumbus.fi 

Stephen May, MA, FCSP, Dip. MDT, MSc 


Mark Werneke, PT, Dip. MDT 

mwerneke@centrastate.com 

�� 

Production 


Nancy Morden 

Executive Asst. 

nancy@mckenziemdt.org 

� 

The McKenzie Institute International 

Headquarters 

1 Alexander Road 



Postal Address 

PO Box 2026 



Telephone: +64-4 299-6645 

Facsimile: +64-4 299-7010 

Email: mckinst@xtra.co.nz 

www.mckenziemdt.org 

©2009 The McKenzie Institute® 


Case Study Guidelines 

Case studies must be type written on the most current 

MII Assessment Form using the MDT standardized terminology. 

As the intent of different case studies may 

vary, a rigid structure is not indicated, but could include: 


�� history 

�� physical examination 


�� management 

�� reviews (history, physical examination, conclusion, management) 

References 

Accuracy of references is the responsibility of the author. 

In the text, state the author’s name and year of 

publication as follows: 

�� (Smith 1998) 

�� (Smith and Jones 1998) 

�� (Smith et al 1998) 

References should be typed in alphabetical order: 

author’s surname and initial (year of publication). 

Full title of paper. Name of journal in full or accepted 

abbreviation volume. First and last page. 

Examples: 

Article: 

Aina A, May S, Clare H (2004). The centralization phenomenon 

of spinal symptoms - a systematic review. 

Manual Therapy 9.134-143. 

Book: 

McKenzie RA, May S (2003). The Lumbar Spine, Mechanical 

Diagnosis and Therapy (2 nd Edition). Spinal 

Publications Ltd, Waikanae. 

Book chapter: 

Twomey LT, Taylor JR (1994). Factors influencing 

ranges of movement in the spine. In: Boyling JD, Palastanga 

N (Eds). Grieve’s Modern Manual Therapy (2 nd 

Ed). Churchill Livingstone, Edinburgh. 

All submissions via email to: 

Chief Editor, Helen Clare 


Production-MIUSA administrative office: 


126 N Salina St., Suite 403 

Syracuse, NY 13202-1059 

Ph: 1-315-471-7612 

Fx: 1-315-471-7636 

Toll-free (US and CAN) 1-800-635-8380 

Questions: email info@mckenziemdt.org 


JANUAR 

15-18, Århus 

APRIL 

2-5, Århus 

APRIL 

16-19, København 

JUNI 


JUNI 

18-21, Århus 

OKTOBER 

22-25, Århus 

NOVEMBER 


Kursusoversigt 2009 

2009 McKenzie kursus program 

JANUAR 

8-11, Århus 

MARTS 


MAJ 

14-17, Århus 

AUGUST 

13-16, Århus 

SEPTEMBER 


NOVEMBER 


JANUAR 

22-24, Århus 

APRIL 


SEPTEMBER 

14-16, Århus 

OKTOBER 


NOVEMBER 

20-22, Århus 

McKenzie Institut Danmark 2009 

Part A Part B Part C Part D Part E 

REFRESHER KURSUS 

24. Januar, København 

2. Maj, Århus 

26. September, København 

30. Oktober, Århus 

CREDENTIAL EVALUERING 

28. Februar, København 

6. Juni, Århus 

10. Oktober, København 

28. November, Århus 

JANUAR/FEBRUAR 

29-01, Århus 


MAJ 


SEPTEMBER 

10-13, Århus 

OKTOBER 


PART X 

FEBRUAR 


SEPTEMBER 

25-26, Århus 

1-2. Maj, Århus 

TILMELDING: 

Onlinetilmelding på www.mckenzie.dk eller kursussekretær Eva Hauge, Ny Kongevej 40, 5000 Odense C 

Telefon: 4228 1221 / E-mail: kursus@mckenzie.dk 

Træffetid: Man-Tir-Tor 15.30-18.00, Ons 8.00-10.00 

Der kan forekomme visse ændringer i programmet - se opdatering på www.mckenzie.dk

International Journal of Mechanical Diagnosis and ... - McKenzie

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?