Systematic review Effectiveness osteoarthritis 4.pdf · To evaluate the effectiveness of physiotherapy exercise after discharge from hospital on function, walking, range of motion,

P

A

BOoDDTcEfOSRtbLCd©

K

pO

h0

ARTICLE IN PRESSHYST-801; No. of Pages 14

Physiotherapy xxx (2015) xxx–xxx

Systematic review

Effectiveness of land-based physiotherapy exercise followinghospital discharge following hip arthroplasty for

osteoarthritis: an updated systematic review

Catherine J. Minns Lowe a,∗, Linda Davies a, Catherine M. Sackley b,Karen L. Barker a,c

a Physiotherapy Research Unit, Oxford University Hospitals NHS Trust, Nuffield Orthopaedic Hospital, Windmill Road,Headington, Oxford, UK

b School of Rehabilitation Sciences, University of East Anglia, Norwich Research Park, Norwich, UKc Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, Oxford University Hospitals NHS Trust,

Oxford, UK

bstract

ackground Existing review required updating.bjective To evaluate the effectiveness of physiotherapy exercise after discharge from hospital on function, walking, range of motion, quality

f life and muscle strength, for patients following elective primary total hip arthroplasty for osteoarthritis.esign Systematic review from January 2007 to November 2013.ata sources AMED, CINAHL, EMBASE, MEDLINE, Kingsfund Database, and PEDro. Cochrane CENTRAL, BioMed Central (BMC),he Department of Health National Research Register and Clinical Trials.gov register. Searches were overseen by a librarian. Authors wereontacted for missing information. No language restrictions were applied.ligibility criteria Trials comparing physiotherapy exercise vs usual/standard care, or comparing two types of relevant exercise physiotherapy,

ollowing discharge from hospital after elective primary total hip replacement for osteoarthritis were reviewed.utcomes Functional activities of daily living, walking, quality of life, muscle strength and joint range of motion.tudy appraisal Quality and risk of bias for studies were evaluated. Data were extracted and meta-analyses considered.esults 11 trials are included in the review. Trial quality was mixed. Newly included studies were assessed as having lower risk of bias

han previous studies. Narrative review indicates that physiotherapy exercise after discharge following total hip replacement may potentiallyenefit patients in terms of function, walking and muscle strengthening.imitations The overall quality and quantity of trials, and their diversity, prevented meta-analyses.onclusions Disappointingly, insufficient evidence still prevents the effectiveness of physiotherapy exercise following discharge to be
etermined for this patient group. High quality, adequately powered, trials with long term follow up are required.
2015 Chartered Society of Physiotherapy. Published by Elsevier Ltd. All rights reserved.

eywords: Physiotherapy; Exercise; Hip arthroplasty; Systematic review

Please cite this article in press as: Minns Lowe CJ, et al.

ing hospital discharge following hip arthroplasty for osteoarthrhttp://dx.doi.org/10.1016/j.physio.2014.12.003

∗ Correspondence: Physiotherapy Research Unit, Oxford University Hos-itals NHS Trust, Nuffield Orthopaedic Centre, Windmill Road, Headington,xford OX3 7HE, UK. Tel.: +44 1865 737526.

E-mail address: [email protected] (C.J. Minns Lowe).

I

tdEt

ttp://dx.doi.org/10.1016/j.physio.2014.12.003031-9406/© 2015 Chartered Society of Physiotherapy. Published by Elsevier Ltd.

ntroduction

Osteoarthritis is a leading cause of pain and disability inhe UK and worldwide: estimates of age-standardised inci-

Effectiveness of land-based physiotherapy exercise follow-itis: an updated systematic review. Physiotherapy (2015),

ence rates of hip osteoarthritis amongst women and men inurope are approximately 53.2 and 38.1 per 100,000 respec-

ively, with prevalence and incidence increasing with age [1].

All rights reserved.

dx.doi.org/10.1016/j.physio.2014.12.003


mailto:[email protected]


ARTICLE IN PRESSPHYST-801; No. of Pages 14

2 hysioth

Nssttel

epWdtJrdrarswimtprrdtwts

ehlectofewtrtp[basiyblc

aq

optmttm

M

S

AaffcMCHrma

S

pwvWcprssMaispiaomh

C.J. Minns Lowe et al. / P

ICE guidelines state that joint arthroplasty should be con-idered for people with osteoarthritis whose joint symptomsubstantially impact upon their quality of life and are refrac-ory to conservative treatment [2]. These guidelines also statehat referral for arthroplasty should be made before patientsxperience prolonged and established pain and functionalimitation.

The number of hip arthroplasties taking place in the UKach year is substantial. Latest figures indicate that 76,448rimary total hip arthroplasties were carried out in England,ales and Northern Ireland in 2012 (92% of which were

ue to osteoarthritis) [3] and 6956 primary hip arthroplas-ies were undertaken in Scotland in 2011 [4]. The Nationaloint Registry for England, Wales and Northern Ireland alsoeported 10,040 revisions performed in 2012, a revision bur-en now reaching 12% [3]. A similar hip arthroplasty revisionate of 11% was reported for 2011 in Scotland [4]. Tradition-lly, physiotherapy has been a routine component of patientehabilitation following hip joint replacement. The length oftay following joint arthroplasty continues to decrease [5],ith the duration period for post operative in-patient phys-

otherapy being reduced. It is known that impairments inuscle strength and postural stability and functional limita-

ions remain a year after surgery [6] and the effectiveness ofost discharge physiotherapy upon functional ability after hipeplacement is therefore a valid question. Current uncertaintyegarding effectiveness makes it difficult for patients makingecisions about their own health, for health care providerso advise patients and for service providers to determinehether to provide a post discharge physiotherapy service

o patients. It is also unclear what should be included in anyervice provision.

In 2009 we published a systematic review evaluating theffectiveness of physiotherapy exercise after discharge fromospital on function, walking, range of motion, quality ofife and muscle strength, for osteoarthritic patients followinglective primary total hip arthroplasty [7]. The review con-luded that existing trials were generally poor in quality andhat insufficient evidence existed to establish the effectivenessf physiotherapy exercise following primary hip replacementor osteoarthritis. The need for further high quality trials wasmphasised. The last trial included in our original reviewas from 2004 and it therefore seemed timely to consider

he value of repeating the review for 2014. There have beenecent systematic reviews assessing the efficacy of resis-ance training following hip arthroplasty [8], pre operativehysiotherapy and post operative hip movement restrictions9] and a review of predominantly early (inpatient) land-ased and aquatic-based exercise therapy [10]. There haslso been a systematic review of five outpatient or homeetting physiotherapist-directed rehabilitation exercise stud-es [11]. But there has been no published update similar



et to our review. We therefore considered that it woulde useful to update our original review of post dischargeand-based physiotherapy exercise and place it in this widerontext of rehabilitation evidence now provided by these

pntc

erapy xxx (2015) xxx–xxx

dditional reviews which have asked different, but related,uestions.

This review therefore aimed to provide an update of ourriginal review and to explore to what extent is post dischargehysiotherapy exercise effective, in terms of improving func-ion, quality of life, mobility, range of hip joint motion and

uscle strength, for osteoarthritic patients following elec-ive primary unilateral total hip arthroplasty? We also aimedo see if additional outcomes, such as pain or performanceeasures, could now be included in the review.

ethods

earching

8 trials were identified from the previous review (up topril 2007); please see Minns Lowe et al. [7] for search

nd exclusion details. The following databases were searchedrom 1st January 2007 until 14th November 2013 (by LD)or randomised controlled trials relating to land-based exer-ise post hip arthroplasty: AMED, CINAHL, EMBASE,EDLINE, Kingsfund Database, and PEDro. CochraneENTRAL, BioMed Central (BMC), the Department ofealth National Research Register and Clinical Trials.gov

egister were also searched. The search strategy was for-ulated by a health librarian. No language restrictions were

pplied.

election

We sought prospective, comparative clinical trials ofatients undergoing total hip replacement for osteoarthritisho received a physiotherapy exercise rehabilitation inter-ention following discharge from hospital post operatively.e used broad definitions of ‘physiotherapy’ and ‘exer-

ise’ to include exercises/exercise programmes advised orrovided by physiotherapists/physical therapists during theehabilitative period following discharge from hospital afterurgery occurring in the out patient, community or homeetting. In the light of the early rehabilitation review by Dionaco and Castiglioni [10] we did not include recent tri-

ls commencing during in-patient stay. Trials were includedf they compared a physiotherapy intervention vs usual ortandard care or compared two different types of relevanthysiotherapy intervention. We excluded trials in which thentervention consisted of an electrical adjunct to physiother-py. Effectiveness outcomes included in trials were measuresf functional activities of daily living, walking, self-reporteasures of quality of life, muscle strength and range of

ip joint motion. It was not considered possible to include


ain as a separate effectiveness outcome since most trials doot include specific measures for pain; most trials use func-ional measures, which include pain as one component whichannot be separated out from the score as a whole. Study



hysioth

ea

D

LiCbmcdcd‘woamodortb‘p

D

o(rtdabfirnwamladt

R

S

J

NrsAi[rTolr

S

atctthpisa

od[cevt[gwmacoca[4yrtTe[u


ligibility was assessed and agreed by two reviewers (CMLnd LD).

ata abstraction, quality and risk of bias assessment

Two reviewers independently extracted the data (CML andD) and each recorded it on a data extraction form contain-

ng quality indicators from the CONSORT statement and theASP guidelines as previously described [7]. Items coulde marked as yes, no, unclear or partial. Items were onlyarked as yes if they fully and explicitly met the detailed

riteria laid out in the CONSORT standards [12]. Any initialisagreements regarding study quality were discussed untilonsensus was reached. Major disagreement was rare, usuallyisagreement between reviewers was the more minor ‘yes’ topartial/unclear’ or ‘no’ to ‘partial/unclear’. 100% agreementas obtained. A third reviewer (KB) was available in the eventf consensus not being reached. Where key study details werebsent or unclear the authors were emailed for further infor-ation. Both reviewers assessed studies for potential sources

f bias according to the Cochrane risk of bias table [13]. Ran-om sequence generation, allocation concealment, blindingf outcome assessment, incomplete outcome data, selectiveeporting, and other sources were assessed. It was acceptedhat participants and treatment providers are usually unable toe blinded in physiotherapy trials. Each study was judged asgood’, ‘reasonable’ or ‘possible bias’ and a brief explanationrovided.

ata synthesis

In our previous review we presented formal summariesf data for walking speed and hip abductor muscle strengthlack of data and data diversity prevented this for function,ange of joint motion or quality of life). We emphasised thathese were data summaries and ‘that the mixed quality andiversity of trials in this review prevented explanatory meta-nalyses from being undertaken, since the results would riskeing misleading or erroneous, and we do not intend thesegures to be interpreted in this way’ [7]. However, in recenteviews these data summaries have been presented as sig-ificant and statistically significant findings. In consequencee decided that, unless new trials were of sufficient quality

nd quantity to enable meta-analyses we would not carry outeta-analytic summaries for this systematic review. Simi-

arly, unless the quantity of trials sufficiently increased, thessessment of publication bias was felt to be inappropriateue to the small number of trials available for inclusion inhe review.

esults



tudy selection

58 potentially relevant studies were identified fromanuary 2007 and their abstracts were screened for retrieval.

umf2

erapy xxx (2015) xxx–xxx 3

ine full papers were subsequently assessed against theeview’s inclusion criteria. Of these, a total of three newtudies [14–16] were included in the systematic review.

summary is provided in the PRISMA flow diagramn Fig. S1. Table S2 contains a list of excluded studies39–66]. The characteristics of the studies included in theeview and their risk of bias assessments are summarised inables 1 and 2. The number of studies with potential riskf bias plus the diversity in trial interventions and outcomesed to our decision not to undertake meta-analyses for thiseview.

ummary of the interventions and comparisons

The interventions and comparisons included in the reviewre summarised in Table 3. The interventions provided to par-icipants in the trials showed variation. One trial interventiononsisted of a home exercise programme [17], one interven-ion was a home exercise programme with follow up visitso check and progress exercises [18], another also included aome exercise programme but with additional visits and tele-hone calls where necessary [19]. The exercises incorporatednto trials also varied from addressing range of motion andtrengthening [17,19] to targeting strength, postural stabilitynd functional exercises [18].

The majority of interventions included a wide variety ofutpatient physiotherapy approaches; these included aerobicance routines [20], individualised physiotherapy treatment21], supervised strengthening sessions [22], supervised exer-ising sessions plus home exercises [23] and supervisedrgometer cycling sessions [15]. The most frequent inter-ention was group training [14,16,24]. The majority ofrials allocated participants to intervention or control groups15–17,19–21,23]. Three trials compared the interventionroup against an intervention based on usual care. Theseere described as traditional isometric and active range ofovement exercises [18], a home exercise programme [24]

nd standard rehabilitation [22]. A further trial comparedlinic vs home based exercise programmes [14]. The timingf the intervention varied from soon after surgery or dis-harge from hospital to several years; more specifically, soonfter surgery [22], at 2 weeks post operatively [15], 5 weeks21], 8 weeks [14,23,24], 3 months [16] and 6 months [20],

to 12 months [18] while others took place up to severalears post operatively [17,19]. The duration of interventionsanged from at least 3 weeks [15], 5 to 8 weeks [16–18,21]o programmes lasting around 3 to 4 months [19,20,22,24].he frequency of physiotherapy ranged from daily homexercise [19] to, most commonly, 1 to 4 times per week14–16,18,20,21,23,24]. Participants were usually followedp immediately post intervention with no long term follow


p, except for Kaae et al. [21] who additionally included a 6onth follow up, Heiberg et al. [16] who included a one year

ollow up and Liebs et al. [15] who followed participants up to4 months.


Please

cite

this

article

in

press

as:

Minns

L

owe

C

J,

et

al.

Effectiveness

of

land-based

physiotherapy

exercise

follow

-ing

hospital

discharge

follow

ing

hip

arthroplasty

for

osteoarthritis:

an

updated

systematic

review

.

Physiotherapy

(2015),http://dx.doi.org/10.1016/j.physio.2014.12.003

AR

TIC

LE

IN P

RE

SS

PHY

ST-801;

No.

of Pages

14

4

C.J.

Minns

Low

e et

al. /

Physiotherapy

xxx (2015)

xxx–xxx

Table 1Study characteristics of the trials evaluated in the systematic review (n = 11). This table provides information regarding the participants, the interventions, the main outcomes and findings.

Paper Participants (sample size) Intervention (and time ofintervention)

Main outcome measures (andtime/second of follow up)

Results

Johnsson et al., 1988 Unilateral primary THR(n = 30) patients

Intervention group: organised OPDPT exercises for 1 to 2 months vscontrol group (2 months postoperatively)

Passive hip mobility. Maximumisometric muscle strength. Limb lengthdiscrepancy. Walking speed.Trendelenberg. Limp. Walking supports.Stair climbing. Sitting and rising fromchair. ADL. (Follow up: 6 to 8 weeks,and 6/12 post operatively)

No significant differences between groups for anyoutcomes

Kaae et al., 1989 Unilateral THR patients(n = 26)

5 weeks intervention of OPD PT(mobility, gait and posture trainingplus any individual treatment needs)three times a week (5 weeks postoperatively)

Joint mobility. Hip Strength. Posturalstability. Stamina in walking. Gait. BorgScale. ADL questionnaire. (Follow up:2½ and 6 months post operatively)

Differences between gps for stamina, gait, securitywithout sticks, ADL and Borg values, favouringintervention. No differences for joint mobility,strength and postural stability

Patterson et al., 1995 Unilateral or bilateral primaryfemale THR patients (n = 20)

Comparison of intervention (exerciseprogramme, including aerobic-danceroutines, twice weekly for 3 months)vs control (6 months post operatively)

Exercise intensity. Treadmill test (peakVO2 in ml/kg/minute, peak VO2 in peakheart rate, peak respiratory exchangeratio, exercise duration, VO2 atanaerobic threshold, peak lactate, VO2 atlactate threshold. Blood pressure. Heartrate. ST segment displacement, end tidalCO2 VCO2, VE anaerobic threshold.Borg scale. Walking speed. Body mass.Skin folds. Body fat. (Follow up: postintervention at 3 months)

Peak VO2 increased in the intervention gp comparedto baseline (P < 0.05) but did not differ significantlyfrom control. Intervention gp significantly increasedtheir walking speed by 10.1% compared withbaseline and control (P < 0.05), and showedincreased VO2 at lactate threshold (compared tobaseline and control)

Sashika et al., 1996 THR patients includingrevisions and re-revisions(n = 23)

6 week home programme. 3 gps. Gp1 = range of motion and lowresistance isometric musclestrengthening. Gp 2 = all gp 1 ex pluseccentric hip abductor musclestrengthening. Gp 3 = control (mean26.4 months post operatively)

Range of hip motion. Muscle strength.Hip abductors Maximal isometric torque.Gait speed. Cadence. Japaneseorthopaedic score. Compliance. (Followup: post intervention at 6 weeks)

Significant improvements within groups formaximum isometric torque (gp 1, P < 0.01 THRside; gp 3, P < 0.05 THR side and both sides for gp2, P < 0.01). Gait speed and cadence also improvedwithin gps 1 and 2

Jan et al., 2004 Unilateral primary THRpatients (n = 58)

Comparison of a 12 weeks dailyhome exercise programme (withphone calls and extra visits ifnecessary) vs control group (at least1.5 years post operatively)

Muscle strength. Walking speed.Functional ability. Compliance. (Followup: post intervention at 12 weeks)

Exercise group showed significant within grouppre–post intervention difference (P < 0.5). ‘Highcompliance exercisers’ showed significantly greater(P < 0.05) improvement in outcome measures than‘low compliance exercisers’ and control

Nyberg and Kreuter,2002

Primary THR patients(n = 55)

All taught home training programme.Treatment groups: additional grouptraining (movement training, muscletraining and walking exercises) twicea week for 15 weeks. (8 weeks postoperatively)

MACTAR instrument. Hip joint range ofmotion. Walking speed. Pain on activity.Quality of life. Opinion on training.(Follow up: post intervention at 6 monthspost operatively)

No significant differences observed between thegroups


Please

cite

this

article

in

press

as:

Minns

L

owe

C

J,

et

al.

Effectiveness

of

land-based

physiotherapy

exercise

follow

-ing

hospital

discharge

follow

ing

hip

arthroplasty

for

osteoarthritis:

an

updated

systematic

review

.

Physiotherapy


AR

TIC

LE

IN P

RE

SS

PHY

ST-801;

No.

of Pages

14

C.J.

Minns

Low

e et

al. /

Physiotherapy

xxx (2015)

xxx–xxx

5

Suetta et al., 2004 Unilateral primary THRpatients (n = 36).

3 group comparison.1. Standard rehabilitation (SR)1 hour/day for12 weeks2. Standard rehabilitation plusunilateral strength training 3/weekfor 12 weeks (ST)3. SR plus neuromuscular electricalstimulation (NMES) – arm excludedfrom systematic review) (postoperatively 1 hour/day for 12 weeks)

Quadriceps muscle cross sectional area.Maximal voluntary isometric quadricepsstrength. Rapid muscle forcedevelopmentEMG to evaluate changes in muscleactivation. (Follow up: 5 and 12 weekspost operatively)

ST gp showed significant differences in maximalisometric muscle strength (24%, P < 0.01);contractile rapid force development (26 to 45%,P < 0.05) and contractile impulse (27 to 32%,P < 0.05). Mean EMG signal amplitude for vastuslateralis was larger in ST gp than SR gp at 5 and 12weeks after surgery (P < 0.05). No within gpsignificant differences for other gps

Trudelle-Jackson andSmith, 2004

THR patients (n = 34) (4 to 12months post operatively)

2 groups for 8 week programme 4 to12 months post surgery. Intervention:strength and postural stabilityexercises vs traditional isometric andactive range of movement ex (4 to 12months post operatively)

Oxford Hip Scores. Muscle strength.Postural stability. Fear of falling.Compliance. (Follow up: postintervention at 8 weeks)

Statistically significant results observed for alloutcomes in intervention gp, except fear of falling.No significant differences were observed within thetraditional gp

Galea et al., 2008 Unilateral primary THRpatients (n = 23) (8 weekspost operatively)

Comparison of 8 weeks clinic-basedgroup exercise programme twice aweek with home based programme

Timed up and go test. Stair climbing test.6 minute walk test. WOMAC. AQoL.Gait (GAITRite for speed, cadence, steplength, step time, stance time percentage,single-support time). (Follow up pre andpost intervention)

Between group differences for TUG (P = 0.042). Noother significant differences between groups. Withingroup significant differences for WOMAC function,AQoL, TUG, stair climbing and 6 minute walk test,walking speed, cadence, and step length,single-support time. For the clinic group withingroup significant differences observed for step time,stance time, double-support time

Liebs et al., 2010 Unilateral primary THRpatients (n = 203; 193 OA and10 osteonecrosis) (2 weekspost operatively)

2 groups. Intervention = ergometercycling 3 times per week for at leastthree weeks) vs control

WOMAC. Physical Component ofSF-36, Lequesne hip and knee score,Satisfaction question. (Follow up 3, 6,12, 24 months)

Significant differences observed for WOMACfunction at 3 months (P = 0.046), 24 months(P = 0.019); WOMAC pain at 3 months (P = 0.049);WOMAC stiffness at 24 months (P = 0.047); SF-36at 6 months (P = 0.011) and 24 months (P = 0.004);Lequesne at 24 months (P = 0.043)

Heiberg et al., 2012a Unilateral primary THRpatients (n = 68) (3 monthspost operatively)

Two groups. Intervention: 12sessions group led walking skillprogramme, twice a week for70 minutes compared with control

6 minute walk test. Stair climbing test.Figure of 8 test. Index of muscle function(IMF). Range of motion. Self-efficacyquestions developed for this study. Harriship score. HOOS. (Follow up postintervention and 1 year)

No significant differences groups in any of themeasured variables. Within group scores, theintervention group showed greater improvement in 6minute walk test, stair climbing, IMF, extension,HHS and self-efficacy. By 1 year, 9 interventionpatients reported 1 to 1 falls, 13 control patientsreported 1 to 2 falls and 2 control patients 3 to 5times

Key – THR = denotes total hip replacement; OPD PT = out patient department physiotherapy; Gp = group; ADL = activities of daily living; WOMAC = Western Ontario and McMaster Universities OsteoarthritisIndex; SF-36 = Short Form General Health Survey; HOOS = Hip dysfunction and osteoarthritis outcome score.

a Additional information from author.


Please

cite

this

article

in

press

as:

Minns

L

owe

C

J,

et

al.

Effectiveness

of

land-based

physiotherapy

exercise

follow

-ing

hospital

discharge

follow

ing

hip

arthroplasty

for

osteoarthritis:

an

updated

systematic

review

.

Physiotherapy


AR

TIC

LE

IN P

RE

SS

PHY

ST-801;

No.

of Pages

14

6

C.J.

Minns

Low

e et

al. /

Physiotherapy

xxx (2015)

xxx–xxx

Table 2Assessment of risk of bias for 10 studies (Kaae et al., 1989 excludeda); graded as possible risk of bias or reasonable or good.

Random sequencegeneration(selection bias)

Allocationconcealment(selection bias)

Blinding of outcomeassessment (detection bias)(patient-reported outcomes)

Incomplete outcomedata addressed(attrition bias)

Lack of selectivereporting (reportingbias)

Lack of other sourcesof bias

Our evaluation

Johnsson et al.,1988

Statesrandomised. Nodetails

Unclear Unclear Unclear. 30 patientsbegan study. ‘resultsof another six patientsfor the study could notbe evaluated’

No apparent problems Unclear determinationof sample size. Smallsample with somelarge standarddeviations

Possible bias due to smallsample size and lack ofdetail provided for keyindicators of quality

Patterson et al.,1995

Assigned bylocation: whetherparticipants livedwithin taxi driveof hospital

Unclear Yes No apparent drop outbut not all assessmentdata collected for allparticipants at bothtimepoints

No apparent problems Unclear determinationof sample size. Smallsample, analysesbased upon normallydistributed data butthis is not reported

Possible bias due torandomisation approach,and lack of detailprovided for keyindicators of quality

Sashika et al., 1996 Not randomised No Unclear Partial No apparent problems Unclear determinationof sample size. Smallsample includingprimary, revision andre-revisions

Possible bias due tonon-randomisedapproach, and lack ofdetail provided for keyindicators of quality

Nyberg and Kreuter,2002b

Alternatelyassigned

Patients did notknow there wasanother group

Blind outcome assessment Loss to follow up:intervention = 27.3%(n = 6/22)Control = 26%(n = 6/23)

Hip flexion dataobtained from authors

Convenience sample Possible bias due torandomisation approach,sample size and loss tofollow up

Jan et al., 2004b Alternatelyassigned

No concealment Blind outcome assessment n = 53/58 completed.8.6% loss to follow up

Paper reported resultsby high/lowcompliance. Betweengroup comparisondata obtained

Between groupcomparisons data andsample sizecalculations obtainedfrom authors. Targetsample size not metbut power at least 0.8for all measures

Possible bias due torandomisation approachand lack of allocationconcealment

Suetta et al., 2004b Computerisedrandomisation

Unclear Not all blinded Loss to follow up:standard rehabilitation(SR) = 25% (n = 9/12)SR+ strengthtraining = 18%(n = 11/13)

No apparent problems Authors stated powercalculations wereconducted

Possible bias due to lackof full blind outcomeassessment and drop out


Please

cite

this

article

in

press

as:

Minns

L

owe

C

J,

et

al.

Effectiveness

of

land-based

physiotherapy

exercise

follow

-ing

hospital

discharge

follow

ing

hip

arthroplasty

for

osteoarthritis:

an

updated

systematic

review

.

Physiotherapy


AR

TIC

LE

IN P

RE

SS

PHY

ST-801;

No.

of Pages

14

C.J.

Minns

Low

e et

al. /

Physiotherapy

xxx (2015)

xxx–xxx

7

Trudelle-Jacksonand Smith, 2004b

Random numberstable

Unclear Not blinded Loss to follow up:intervention = 22%(n = 14/18 completed),control 12.5%(n = 14/16 completed

No apparent problems Convenience sample.No sample sizecalculation

Possible bias due tosample size, drop out andlack of blinding

Galea et al.,2008 + authorc

Independentcomputer-generated randomnumbers list

Concealed usingopaque envelopes

Blind outcome assessment Small loss to followup (n = 23/25completed)

No apparent problems Unclear determinationof sample size

Reasonable

Liebs et al., 2010 Adequate Adequate Patient-reported outcomes Loss to follow up:intervention = 15.4%(n = 88/104),control = 25.2%(n = 74/99)

No apparent problems No apparent problems Well written up study.Possible bias due toamount and uneven lossto follow up

Heiberg et al.,2012 + authorc

Drew envelopefrom 70 prepreparedenvelopes

Concealed usingclosed, opaque,sealed and mixedenvelopes

Blind outcome assessment Loss to follow up:intervention = 8.6%(n = 3/35),control = 3%(n = 1/32). Lastobservation carriedforward to obtaincomplete dataset

No apparent problems Non-consecutiveapproaching ofpatients, participantsnot included if too farfrom the hospital(≈30 km), worsepreoperative scoresamongst thosedeclining toparticipate limitgeneralizability

Reasonable

a This trial was written up in summary form rather than as a journal paper.b Judgement based on additional information and data from authors from our original review.c Additional information provided by lead author of study.


Please cite this article in press as: Minns Lowe CJ, et al. Effectiveness of land-based physiotherapy exercise follow-ing hospital discharge following hip arthroplasty for osteoarthritis: an updated systematic review. Physiotherapy (2015),http://dx.doi.org/10.1016/j.physio.2014.12.003


8 C.J. Minns Lowe et al. / Physiotherapy xxx (2015) xxx–xxx

Table 3Summary of trial interventions and comparisons included in the hip replacement trials (n = 8). Details of the intervention and comparison groups were availablefrom the papers and authors and these are summarised.

Early programme provided to all participants Intervention Comparison

Johnsson et al., 1988 20 minutes physiotherapy session days 7 to12 post operation. Supine: straight leg raise,hip extension against resistance, hipabduction. Sitting: knee flexion, extension.Gait: stick/second. Advice: regardingsitting/bending; aids, raised toilet seat, seatcushion. Drive, cycle, swim from 6 to 8weeks post operation

Outpatient physiotherapy. Supine: activatedabdominals, hamstrings and quads by (1) liftpelvis with lower limb resting on cushion.(2) Lift pelvis holding a ball between flexedknees keeping feet on floor. Repeat 1 and 2weightbearing on 1 leg alternately. Fourpoint kneeling: alternate leg extension.Standing: weight shift side-to-side indifferent directions to activate gluts, hams,hip abductors, tenscia fascia lata, iliopsoas,quadriceps and gastrocnemius, standing ontoes: whilst flexing hips and knees. Singleleg stand: swing other leg slow backwardsand forwards. St: step up/down alternate leg.Sitting: sit to stand (weight bearing mostly 1leg at a time to activate gluteal, hamstrings,tenscia fascia lata, iliopsoas and quadriceps).Walking programme: 2× weeks for 1/12.Then 1× week for 1/12 OR 1× week for2/12. Duration 45 minutes. Do until 6/12follow up

Kaae et al., 1989 Inpatient exercise programme plus homeexercise programme after discharge

Outpatient physiotherapy: mobility training,gait and posture correction and to addressindividual needs as appropriate. Treatmentfrom around 5/52 post operation, around 45minutes 3× a week for 5 weeks

Patterson et al., 1995 Warm up 5 to 10 minutes (walking,flexibility). Aerobic dance to music (15 to 20minutes). Cool down: stretches and balanceexercises (10 minutes). Short relaxationperiods (decreased as fitness improved). 2×per week for 12/52 at least 6/12 postoperation

No programme – normalactivities only

Sashika et al., 1996 Group A. Range of movement, supine: hipflexion. All fours: hip flexion. Sitting: hipflexion. Strength exercises (using ankleweight 20 to 30% of maximum isometricmuscle torque), Supine: for hip flexors. Sidelying: for hip abductors. Prone: for hipextensors. Sitting: for knee extensors. GroupB. as for Gp A plus 1 strength exercise instanding (no ankle weight): 1 leg stand:elevating pelvis for hip abductors. 2× dayfor 6/52

No programme

Nyberg and Kreuter,2002

Inpatient: gait training with 2 crutches,weight bearing as pain allows, range ofmovement exercises. Visit to physiotherapist8 weeks post op for exercises. Standing:unilateral hip extension, abduction plusankle dorsiflexion. Repeat alternate leg.Bilateral heel raises. Standing at wall: gentlesquats. Sitting: knee extension, knee lifts

Movement training in standing and lying andmuscle training against the limb’s ownweight and walking in standing and lying.Walking exercises.Group training 2× week for 15 weeks. Eachsession approximately 45 minutes

Continued with hometraining until follow up at6 months post op

Jan et al., 2004 Range of motion: hip and knee flexion andextension. Strength (with 1 kg ankle weightfor women, 2 kg for men) Supine: hip flexionwith (a) knee flexed, (b) knee extended.Prone: hip extension. Side lying: hipabduction. Standing: alternate single legstance (5 seconds hold) 10 reps, 2× daily. 30minutes walk

Nil else added



C.J. Minns Lowe et al. / Physiotherapy xxx (2015) xxx–xxx 9

Table 3 (Continued)

Early programme provided to all participants Intervention Comparison

Suetta et al., 2004 Standard rehabilitation booklet. Bedexercises: ankle dorsiflexion, plantarflexion,isometric gluts, pelvic and thigh muscles.Sitting: knee extension. Standing: hipabduction, knee flexion, step training andcalf stretches. 2× daily

Progressive quadriceps training. In patient,sitting: daily unilateral knee extension (3×10 reps) with sandbags strapped to ankle.From day 7, 10 minutes static cycling warmup, supine: leg press and knee extensor gymmachine training 3× week. Trainingintensity increased from 20 RM max (∼50%if 1 RM) in week 1; 15 RM (∼65% of 1RM)in weeks 2 to 4; 12RM (∼70% of 1RM) inweeks 5 to 6; 8RM (∼80% of 1RM) for last6 weeks. Weeks 1 to 6: 3 to 5 sets of 10 reps.Weeks 6 to 12: 5 to 8 sets of 10 reps.Exercise done quickly in concentric phaseand slowly during eccentric phase. Trainingload adjusted weekly

Control session 1× week:physiotherapist checkedhome exercises

Trudelle-Jackson andSmith, 2004

Sitting: sit to stand. Standing: unilateral heelraises, partial knee flexion, single leg stand,knee raises with alternate arm raises, sideand back leg raises, unilateral pelvic raisingand lowering. Repetition rate (RR) = 15, 3 to4× week for 8 weeks. If able RR increasedto 20 at 1st follow up (2 weeks) and 2× 20 at2nd follow up (8 weeks). Control and qualityof movement emphasised

Gluts, hamstrings andquadriceps sets, anklepumps, heel slides. Sup:hip abduction, internalrotation and externalrotation. RR as forintervention group

Galea et al., 2008 Supervised, centre-based exercise grouptwice a week (45 minutes) for 8 weeks. 7exercises: figure of 8 path walk (5 minutes);Sit to stand (5 minutes); Single leg stance(5 minutes); Climbing steps; Standing: Hipabduction, Heel raises, side stepping.Repetitions and difficulty levels adjusted andprogressed

Home-based groupreceived illustratedinstruction guide of thesame exercises

Liebs et al., 2010 Ergometer cycling sessions (supervised byphysiotherapist) three times a week for atleast three weeks. Resistance of theergometer was set to a minimum (e.g. 30 W).Saddle height set to allow knee extension

No ergometer cycling

Heiberg et al., 2012 No restrictions on weightbearing. Movementrestrictions 3 months (avoid hip flexion past90◦, hip adduction and internal rotationbeyond neutral). Daily in-patientphysiotherapy 30 minutes (self-careinstructions, joint mobility musclestrengthening exercises (bed or bench),learning to walk with an aid. 73% ofparticipants had post dischargephysiotherapy (mainly flexibility andstrengthening exercises on bed/bench)

Groups n = 2 to 8 participants supervised byphysiotherapist. 12 sessions twice weekly.70 minutes of exercise. Exercises,repetitions, difficulty levels were progressedto suit individuals. Programme included:warm up with music (10 minutes, weighttransfers, sidesteps with arm swing, walkingat different speeds and step lengths); sit tostand (5 minutes); lunges (5 minutes);single-leg stance (5 minutes); standing onfoam balance pad (10 minutes); stepup/down (5 minutes); Stair climbing(5 minutes); obstacle course (10 minutes);throwing ball (5 minutes); walking(5 mi

Encouraged to continuewith exercises and keepgenerally active

Sp

mt

T(H

elf-report measures of function (8 trials, 484articipants)



Nine studies reported results for functional activityeasures [14–19,21,23,24]. The measures used included

he following: The Oxford Hip Score [18], The McMaster

oOu

nutes); stretching 5 minutes

oronto Arthritis Patient Preference Disability QuestionnaireMACTAR) [24], the functional component of the Harrisip Score [19], Unspecified activities of daily living/patterns


f activity self-report measures [21,23] and the Japaneserthopaedic Score [17]. The newly identified studies havesed the Western Ontario and McMaster Universities



1 hysioth

Aasp

d[bsTebwgis

W

ewwmi[sft[eswiwIi

Ntiifgaaw

R

fih[re

dtgjti

M

uAcd[aastami

O

obwsntwgsbn

D

sdijf

etmd

0 C.J. Minns Lowe et al. / P

rthritis Index (WOMAC) [14,15] and the Hip disabilitynd Osteoarthritis Outcome Score HOOS [16]; however thetudy interventions differed substantially from each other,reventing meaningful meta-analysis.

Within the individual trials, one demonstrated significantifferences between groups in favour of the intervention15] five demonstrated no observed significant differencesetween groups [14,16,21,23,24], while one trial used thecore to describe the group characteristics at baseline [17].wo trials [18,19] showed significant within group differ-nces for the treatment arm only, indicating a treatmentenefit within the treatment groups. Galea et al. [14] showedithin group differences for both home and clinic basedroups and Heiberg et al. [16] reported greater within groupmprovements for the intervention group. The results areummarised in Tables 1 and 2.

alking (8 trials, 303 participants)

Some form of walking outcome measurement was used inight trials [14,16,17,19–21,23,24] although the means byhich walking was measured varied. Mean ‘comfortable’alking speed over an unspecified time/distance measured inetre/second was measured in one trial [20]. Walking speeds

n metre/minute were provided in two trials [17,19]. Jan et al.19] measured fast and free walking speeds on hard and grassurfaces plus free walking speed measured on a spongy sur-ace, whilst Sashika et al. [17] provided no further details ofhe test. Maximum walking speeds were provided in two trials23,24]. Few procedural details were provided by Johnssont al. [23] while Nyberg and Kreuter [24] measured self-elected maximum walking speed in seconds over a 30 malkway. Six minute walk tests were included in two stud-

es [14,16]. A twelve minute stamina walking test, withoutalking aids, on a treadmill was used by Kaae et al. [21].

n addition, cadence [17] and subjective gait analyses werencluded in one trial each [21].

The results from these trials are mixed (see Table 1).o significant differences were observed between groups in

hree trials [16,23,24]. One trial showed one significant find-ng for timed up and go, which includes walking as one ofts elements, but no between group significant differencesor 6 minute walk tests [14]. Observed differences betweenroups were noted in walking stamina by Kaae et al. [21]nd found to be significant in another trial (P < 0.05) [20]. Inddition, significant differences within interventions groupsithin a trial were observed within two trials [17,19].

ange of joint motion (5 trials, 202 participants)

Range of motion was used as an outcome measure inve trials [16,17,21,23,24]. Measures included passive



ip flexion, extension defect, abduction and adduction23], passive hip flexion, extension, abduction and internalotation using a goniometer [24] and active hip flexion,xtension and abduction [16]. Kaae et al. [21] provide few

tdfs


etails regarding measurement. Sashika et al. [17] reportedhat hip flexion did not improve significantly within anyroups. No significant differences between groups for hipoint range of motion were reported in the remaining fourrials [16,21,23,24]; one study reported greater improvementn extension within the intervention group [16].

uscle strength (6 trials, 207 participants)

No new information was obtained. Muscle strength wassed as an outcome measure in six trials [17–19,21–23].

variety of measures included isometric quadriceps mus-le force measurements (maximal voluntary contraction) viaynamometry [22], hip abductor maximal isometric torque17], isometric muscle strength, of hip flexors, extensors,bductors and adductors and knee extensors and flexors via

strain gauge [23]; isokinetic hip abductor, flexor and exten-or muscle strength via dynamometry [19]; a BEP-IIIa forceransducer to measure hip flexor, hip extensor, hip abductornd knee extensor muscle strength via a ‘make test’ [18] andanual muscle testing [17,21]. The results are summarised

n Table 1 and presented in depth in the earlier review [7].

ther self-report measures

A 0 to 100 mm visual analogue scale to measure qualityf life was used in one trial [24]. No significant differencesetween the groups were demonstrated. Patient satisfactionas included in one study [15]; the authors report more ‘very

atisfied’ responses in the intervention group but there wereo between group differences. Self-efficacy was included inhe trial by Heiberg et al. [16]; no between group differencesere observed, but the intervention group showed withinroup improvement (P = 0.04). Heiberg et al. [16] also mea-ured self-report fall rates, numbers were too small to analyseut more control group participants reported falls and a higherumber of falls than those in the intervention group.

iscussion

This systematic review finds that, disappointingly, it istill not yet possible to establish the extent to which postischarge physiotherapy exercise is effective, in terms ofmproving function, quality of life, mobility, range of hipoint motion and muscle strength, for osteoarthritic patientsollowing elective primary unilateral total hip arthroplasty.

Lack of efficacy also cannot be assumed/concluded. Thexisting data indicate the potential for post discharge physio-herapy to benefit patients in terms of self-reported function,

easures of functional performance and muscle strength. Theiversity, lack of trials, generally small sample sizes plus


he unsatisfactory quality of many existing trials prevent aefinitive answer at this time and also prevent meta-analysesrom being performed. However, on a positive note, the newtudies included in this review were assessed as providing



hysioth

mtaiatmiscfc

etpaillndsbttsapw[itadw

pcriacnrgmiurrfiorfi

fiOwqimfit

C

namtaisrwit[oaotmnaa[

L

pbtawabfrand the third reviewer was not required to achieve consensus.

In conclusion, despite recent research with less risk of


uch less risk of bias and it is believed that, following fur-her studies, meta-analyses will soon be both possible andppropriate. Also, follow up times are improving and themportance of following up patients for at least one year postrthroplasty, and the need to carefully consider follow upimepoints [25], is being recognised. Additionally, a recent

eta-analysis assessing the recovery of physical function-ng following total hip arthroplasty, of 24 prospective cohorttudies and seven trials (different to those in our review)oncluded that at 6 to 8 months post operatively, physicalunctioning had generally recovered to about 80% to that ofontrols, dipping just below 80% at one year [26].

Our previous review identified the need for research toxplore the optimum time-point at which to offer any addi-ional post discharge physiotherapy exercise intervention toatients and to determine effective treatment type, contentnd dosage of treatment via adequately powered trials whichncorporate long term follow up. The last decade has addedimited improvements to this knowledge base, but has high-ighted the growing challenges surrounding arthroplasty. Theumber of arthroplasties carried out each year has risenuring this time and the UK population shows increasingenescence [3,4]. We also believe that the increasing num-ers of people undergoing revision hip arthroplasty [3,4] raisehe need for more research into exercise rehabilitation forhis, now sizeable, group of patients. The rising rates of obe-ity and associated lower hip function scores at follow upfter total hip arthroplasty provide further challenges; obeseatients fear pain induced by physical activity/movementhich corresponds to lower self-reported physical mobility

27]. Additionally, although rehabilitation trials are needed,t is recognised that these would be difficult to undertake inhe UK; many funding bodies do not fund treatment costsnd, since many hospital Trusts do not routinely provide postischarge physiotherapy [28], research study treatment costsould be non-reimbursable.We concur with Di Monaco that the sparse evidence

revents the production of a detailed evidence-based exer-ise protocol following hip arthroplasty [10]. Their updatedeview of exercise following hip arthroplasty (n = 9 stud-es) includes two of the 11 studies included in our reviewnd provides a detailed overview regarding earlier exer-ise studies, including in-patient, plus aquatic exercise. Noew trials provided data regarding muscle strength in oureview. However research begun during in-patient stay sug-ests that early maximal strength training is feasible anday be effective in regaining muscle strength [29,30]. This,

n addition to the research presented in this review, leadss to concur with previous narrative reviews [8,10,31] thatesistance training appears to have a beneficial role inehabilitation following arthroplasty and is recommendedor inclusion into programmes. However, recent research



ndicates that this is not the case for the UK [28]. Bothur review and that of Di Monaco et al. differ from theecent review by Coulter et al. [11]. Coulter et al. includedve studies in their review and meta-analyses of exercise

btfa


ollowing hip arthroplasty, concluding that physiotherapymproves hip abductor strength, gait speed and cadence.f the four studies included in both their review and ours,e believe only one study [14] demonstrated sufficientuality to include in meta-analyses. A search of trials reg-sters reveal that research evaluating the effect of an 8-week

ulticomponent rehabilitation programme on strength andunction (NCT01817010) and a home based resistance train-ng programme (ISRCTN13019951) are presently underwayo provide additional information.

linical implications

There is no current consensus regarding the optimal andecessary outcomes/measurements required following hiprthroplasty, where even the definition of function varies,aking it problematic to integrate findings from multiple

rials [27,32]. Self-report measures, such as the WOMACnd Oxford Hip Score, have been demonstrated to have ceil-ng effects post hip arthroplasty [33] and there is no clearlyuperior outcome measure of self-reported function. Withegard to performance, this review indicates that the 6 minutealk test is being used more frequently and, if future stud-

es include this outcome, a meaningful meta-analysis forhis measure of walking will soon be possible. One trial16] has included self-efficacy in their study. The influencef psychological factors upon outcomes following total hiprthroplasty is currently unclear due to lack of informationr conflicting evidence [34] and future studies may wisho consider addressing this need. Heiberg et al. [16] also

easured self-reported falls. This seems to be a highly perti-ent outcome since falls are a significant cause of morbiditynd mortality in the elderly, can cause psychological distressnd it is known that certain types of exercise prevent falls35].

imitations

Whilst the reviewers believe the search strategy to be com-rehensive, inclusive (no language restrictions) and to haveeen successful in locating relevant trials for inclusion inhe review, it remains a possibility that other studies existnd have been missed in the review. Additional informationas sought from authors which improved data completeness

nd evaluation of the newly included studies. Assessment ofias, study quality and data extraction was generally similarrom both independent assessors. Following discussion botheviewers were in full agreement for all items for all papers


ias, further research is still required to establish the extento which post discharge physiotherapy exercise is effectiveor patients following elective primary unilateral total hiprthroplasty for osteoarthritis.



1 hysioth

A

owrbrEfcw

u

c

A

fj

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[


cknowledgements

We would like to gratefully acknowledge the assistancef Mary Galea and Kristi Heiberg who kindly provided usith additional information to include in the update of this

eview. We also thank all those who provided assistancey providing additional information included in the earliereview: Margareta Kreuter, Birgitta Nyberg, Charlotte Suetta,laine Trudelle-Jackson and Pei-Fang Tang. We also grate-

ully acknowledge the assistance of Dr Michael Dewey whoarried out the meta-analytic summaries in the earlier reviewhich are referred to in this update.Ethical approval: Ethical approval was not required to

pdate this review.Funding: None declared.Conflict of interest: The author’s declare that they have no

onflicting interests.

ppendix A. Supplementary data

Supplementary data associated with this article can beound, in the online version, at http://dx.doi.org/10.1016/.physio.2014.12.003.

eferences

[1] National Institute for Health and Care Excellence. Overview – arthri-tis of the hip (end stage)-hip replacement (total) and resurfacingarthroplasty (Rev TA2, TA44); September 2013. http://www.nice.org.uk/nicemedia/live/13690/65488/65488.pdf [accessed 21.01.14].

[2] NICE Clinical Guideline 59. Osteoarthritis: the care and managementof osteoarthritis in adults; 2008. http://www.nice.org.uk/nicemedia/pdf/CG59NICEguideline.pdf [accessed 21.01.14].

[3] National Joint registry for England, Wales and Northern Ireland. 10thannual report; 2013. http://www.njrcentre.org.uk/njrcentre/Portals/0/Documents/England/Reports/10th annual report/NJR%2010th%20Annual%20Report%202013%20B.pdf [accessed 21.01.14].

[4] Scottish Arthroplasty Project. Biennial report 2012: a summary ofprocedures and outcomes for patients undergoing arthroplasty oper-ations during 2010 & 2011, http://www.arthro.scot.nhs.uk/Reports/sap national report 2012.pdf [accessed 21.01.14].

[5] Health Service Journal. Length of stay analysis for hip replacements byTrust – United Kingdom, http://www.performance-healthcheck.co.uk/hip-replacement-los/national/#national-data [accessed 21.01.14].

[6] Trudelle-Jackson E, Emerson R, Smith S. Outcomes of total hip arthro-plasty: a study of patients one year postsurgery. J Orthop Sports PhysTher 2002;32(6):260–7 [Epub 2002/06/14].

[7] Minns Lowe C, Barker K, Dewey M, Sackley C. Effectiveness ofphysiotherapy exercise following hip arthroplasty for osteoarthritis:a systematic review of clinical trials. BMC Musculoskelet Disord2009;10:98.

[8] Kristensen J, Franklyn-Miller A. Resistance training in muscu-loskeletal rehabilitation: a systematic review. Br J Sports Med2012;46(10):719–26 [Epub 2011/07/28].

[9] Sharma V, Morgan P, Cheng E. Factors influencing early reha-bilitation after THA: a systematic review. Clin Orthop Relat Res2009;467(6):1400–11.



10] Di Monaco M, Castiglioni C. Which type of exercise therapy is effectiveafter hip arthroplasty? A systematic review of randomized controlledtrials. Eur J Phys Rehabil Med 2013;49(6):893–907.


11] Coulter CL, Scarvell JM, Neeman TM, Smith PN. Physiotherapist-directed rehabilitation exercises in the outpatient or home settingimprove strength, gait speed and cadence after elective total hip replace-ment: a systematic review. J Physiother 2013;59(4):219–26 [Epub2013/11/30].

12] Schulz KF, Altman DA, Moher D, for the CONSORT Group. CON-SORT 2010 statement: updated guidelines for reporting parallel grouprandomized trials. Ann Intern Med 2010;152:726–32.

13] Higgins JPT, Green S. Cochrane handbook for systematic reviews ofinterventions version 5.1.0: the Cochrane Collaboration; 2011. Avail-able from: http://www.cochrane-handbook.org [updated March 2011].

14] Galea MP, Levinger P, Lythgo N, Cimoli C, Weller R, Tully E, et al. Atargeted home- and center-based exercise program for people after totalhip replacement: a randomized clinical trial. Arch Phys Med Rehabil2008;89(8):1442–7 [Epub 2008/07/01].

15] Liebs TR, Herzberg W, Ruther W, Haasters J, Russlies M, HassenpflugJ. Ergometer cycling after hip or knee replacement surgery: a random-ized controlled trial. J Bone Joint Surg Am 2010;92(4):814–22 [Epub2010/04/03].

16] Heiberg KE, Bruun-Olsen V, Ekeland A, Mengshoel AM. Effect of awalking skill training program in patients who have undergone totalhip arthroplasty: followup one year after surgery. Arthritis Care Res2012;64(3):415–23 [Epub 2011/12/16].

17] Sashika H, Matsuba Y, Watanabe Y. Home program of physical therapy:effect on disabilities of patients with total hip arthroplasty. Arch PhysMed Rehabil 1996;77(3):273–7 [Epub 1996/03/01].

18] Trudelle-Jackson E, Smith SS. Effects of a late-phase exercise programafter total hip arthroplasty: a randomized controlled trial. Arch PhysMed Rehabil 2004;85(7):1056–62 [Epub 2004/07/09].

19] Jan MH, Hung JY, Lin JC, Wang SF, Liu TK, Tang PF. Effects ofa home program on strength, walking speed, and function after totalhip replacement. Arch Phys Med Rehabil 2004;85(12):1943–51 [Epub2004/12/18].

20] Patterson AJ, Murphy NM, Nugent AM, Finlay OE, Nicholls DP, Bore-ham CA, et al. The effect of minimal exercise on fitness in elderlywomen after hip surgery. Ulster Med J 1995;64(2):118–25 [Epub1995/10/01].

21] Kaae K, Kristianson B, Clausen L. Hofteprojeckt [Hip project]. DanskeFysioterapeuten 1989;3:4–9.

22] Suetta C, Magnusson SP, Rosted A, Aagaard P, Jakobsen AK,Larsen LH, et al. Resistance training in the early postoperative phasereduces hospitalization and leads to muscle hypertrophy in elderly hipsurgery patients – a controlled, randomized study. J Am Geriatr Soc2004;52(12):2016–22 [Epub 2004/12/02].

23] Johnsson R, Melander A, Önnerfält R. Physiotherapy after total hipreplacement for primary arthrosis. Scand J Rehabil Med 1988;20:43–5.

24] Nyberg B, Kreuter M. Sjukgymnastick gruppträning jämförtmed hemträning för patienter opererade med total höftledsplas-tik [Physiotherapist-led group training compared to home trainingfor patients receiving total hip replacements]. Nordisk Fysioterapi2002;6:82–8.

25] Davis AM, Perruccio AV, Ibrahim S, Hogg-Johnson S, Wong R, StreinerDL, et al. The trajectory of recovery and the inter-relationships of symp-toms, activity and participation in the first year following total hip andknee replacement. Osteoarthr Cartil 2011;19(12):1413–21.

26] Vissers M, Bussman J, Verhaar J, Arends L, Furlan A, ReijmanM. Recovery of physical functioning after total hip arthroplasty:systematic review and meta-analysis of the literature. Phys Ther2011;91(5):615–29.

27] Vincent HK, Horodyski M, Gearen P, Vlasak R, Seay AN, Conrad BP,et al. Obesity and long term functional outcomes following elective totalhip replacement. J Orthop Surg Res 2012;7:16 [Epub 2012/04/27].

28] Okoro T, Ramavath A, Howarth J, Jenkinson J, Maddison P, Andrew


JG, et al. What does standard rehabilitation practice after total hipreplacement in the UK entail? Results of a mixed methods study. BMCMusculoskelet Disord 2013;14:91 [Epub 2013/03/19].


http://dx.doi.org/10.1016/j.physio.2014.12.003

http://dx.doi.org/10.1016/j.physio.2014.12.003

http://www.nice.org.uk/nicemedia/live/13690/65488/65488.pdf

http://www.nice.org.uk/nicemedia/live/13690/65488/65488.pdf

http://www.nice.org.uk/nicemedia/pdf/CG59NICEguideline.pdf

http://www.nice.org.uk/nicemedia/pdf/CG59NICEguideline.pdf

http://www.njrcentre.org.uk/njrcentre/Portals/0/Documents/England/Reports/10th_annual_report/NJR 10th Annual Report 2013 B.pdf



http://www.arthro.scot.nhs.uk/Reports/sap_national_report_2012.pdf

http://www.arthro.scot.nhs.uk/Reports/sap_national_report_2012.pdf

http://www.performance-healthcheck.co.uk/hip-replacement-los/national/

http://www.performance-healthcheck.co.uk/hip-replacement-los/national/

http://refhub.elsevier.com/S0031-9406(15)00006-1/sbref0355


































































































































































































































































http://www.cochrane-handbook.org/





































































































































































































































































































































































































































































































































































































































hysioth

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[


29] Husby VS, Helgerud J, Bjorgen S, Husby OS, Benum P, HoffJ. Early maximal strength training is an efficient treatment forpatients operated with total hip arthroplasty. Arch Phys Med Rehabil2009;90(10):1658–67 [Epub 2009/10/06].

30] Husby VS, Helgerud J, Bjorgen S, Husby OS, Benum P, Hoff J. Earlypostoperative maximal strength training improves work efficiency 6–12months after osteoarthritis-induced total hip arthroplasty in patientsyounger than 60 years. Am J Phys Med Rehabil 2010;89(4):304–14[Epub 2010/02/06].

31] Westby MD. Rehabilitation and total joint arthroplasty. Clin GeriatrMed 2012;28(3):489–508 [Epub 2012/07/31].

32] Riddle DL, Stratford PW, Singh JA, Strand CV. Variation in out-come measures in hip and knee arthroplasty clinical trials: a proposedapproach to achieving consensus. J Rheumatol 2009;36(9):2050–6[Epub 2009/09/10].

33] Marx RG, Jones EC, Atwan NC, Closkey RF, Salvati EA, SculcoTP. Measuring improvement following total hip and knee arthroplastyusing patient-based measures of outcome. J Bone Joint Surg Am2005;87(9):1999–2005 [Epub 2005/09/06].

34] Vissers MM, Bussmann JB, Verhaar JA, Busschbach JJ, Bierma-Zeinstra SM, Reijman M. Psychological factors affecting the outcomeof total hip and knee arthroplasty: a systematic review. Semin ArthritisRheum 2012;41(4):576–88 [Epub 2011/11/01].

35] Gillespie LD, Robertson MC, Gillespie WJ, Sherrington C, Gates S,Clemson LM, et al. Interventions for preventing falls in older people liv-ing in the community. Cochrane Database Syst Rev 2012;9:CD007146[Epub 2012/09/14].

36] Mikkelsen LR, Mikkelsen SS, Christensen FB. Early, intensified home-based exercise after total hip replacement – a pilot study. PhysiotherRes Int 2012;17(4):214–26 [Epub 2012/03/28].

37] Krummenauer F, Gunther KP, Witzlebf WC. The incremental costeffectiveness of in-patient versus out-patient rehabilitation after totalhip arthroplasty – results of a pilot investigation. Eur J Med Res2008;13(6):267–74 [Epub 2008/06/19].

38] Tsukagoshi R, Tateuchi H, Fukumoto Y, Okumura H, Ichihashi N.Stepping exercises improve muscle strength in the early postoperativephase after total hip arthroplasty: a retrospective study. Am J Phys MedRehabil 2012;91(1):43–52 [Epub 2011/12/14].

39] Artz N, Dixon S, Wylde V, Beswick A, Blom A, Gooberman-Hill R.Physiotherapy provision following discharge after total hip and totalknee replacement: a survey of current practice at high-volume NHShospitals in England and wales. Musculoskelet Care 2013;11(1):31–8[Epub 2012/07/11].

40] Bandholm T, Kehlet H. Physiotherapy exercise after fast-track totalhip and knee arthroplasty: time for reconsideration? Arch Phys MedRehabil 2012;93(7):1292–4 [Epub 2012/04/13].

41] Jacobs CA, Christensen CP, Berend ME. Sport activity after total hiparthroplasty: changes in surgical technique, implant design, and reha-bilitation. J Sport Rehabil 2009;18(1):47–59 [Epub 2009/03/27].

42] Jensen C, Roos EM, Kjaersgaard-Andersen P, Overgaard S. The effectof education and supervised exercise vs. education alone on the time tototal hip replacement in patients with severe hip osteoarthritis. A ran-domized clinical trial protocol. BMC Musculoskelet Disord 2013;14:21[Epub 2013/01/15].

43] Jensen CJ, Aagaard P, Overgaard A. Recovery in mechanicalmuscle strength following resurfacing vs standard hip arthro-plasty - a randomised controlled trial. Osteoarthr Cartil 2011;19(9):1108–16.

44] Heiberg KE, Ekeland A, Bruun-Olsen V, Mengshoel AM. Recoveryand prediction of physical functioning outcomes during the first yearafter total hip arthroplasty. Arch Phys Med Rehabil 2013;94(7):1352–9[Epub 2013/02/07].

45] Monaghan B, Grant T, Hing W, Cusack T. Functional exercise after



total hip replacement (FEATHER): a randomised control trial. BMCMusculoskelet Disord 2012;13:237 [Epub 2012/11/30].

46] Okoro T, Ramavath A, Howarth J, Jenkinson J, Maddison P, AndrewJ, et al. What does standard rehabilitation practice after total hip

[


replacement in the UK entail? Results of a mixed methods study. BMCMusculoskelet Disord 2013;14:91.

47] Aprile I, Rizzo RS, Romanini E, De Santis F, Marsan S, Rinaldi G,et al. Group rehabilitation versus individual rehabilitation followingknee and hip replacement: a pilot study with randomized, single-blind,cross-over design. Eur J Phys Rehabil Med 2011;47(4):551–9 [Epub2011/06/15].

48] Munin MC, Putman K, Hsieh CH, Smout RJ, Tian W, DeJong G, et al.Analysis of rehabilitation activities within skilled nursing and inpatientrehabilitation facilities after hip replacement for acute hip fracture. AmJ Phys Med Rehabil 2010;89(7):530–40 [Epub 2010/06/23].

49] Rahmann AE, Brauer SG, Nitz JC. A specific inpatient aquatic physio-therapy program improves strength after total hip or knee replacementsurgery: a randomized controlled trial. Arch Phys Med Rehabil2009;90(5):745–55 [Epub 2009/05/02].

50] Smith TO, Mann CJ, Clark A, Donell ST. Bed exercises followingtotal hip replacement: 1 year follow-up of a single-blinded randomisedcontrolled trial. Physiotherapy 2008;94(4):286–91.

51] Stockton KA, Mengersen KA. Effect of multiple physiotherapy ses-sions on functional outcomes in the initial postoperative period afterprimary total hip replacement: a randomized controlled trial. Arch PhysMed Rehabil 2009;90(10):1652–7 [Epub 2009/10/06].

52] Unlu E, Eksioglu E, Aydog E, Aydog ST, Atay G. The effect of exer-cise on hip muscle strength, gait speed and cadence in patients withtotal hip arthroplasty: a randomized controlled study. Clin Rehabil2007;21(8):706–11 [Epub 2007/09/12].

53] Smith TO, Mann CJ, Clark A, Donell ST. Bed exercises followingtotal hip replacement: 1 year follow-up of a single-blinded randomisedcontrolled trial. Hip Int 2009;19(3):268–73 [Epub 2009/10/31].

54] Suetta C, Andersen JL, Dalgas U, Berget J, Koskinen S, AagaardP, et al. Resistance training induces qualitative changes in musclemorphology, muscle architecture, and muscle function in elderly post-operative patients. J Appl Physiol (1985) 2008;105(1):180–6 [Epub2008/04/19].

55] Suetta C, Clemmensen C, Andersen JL, Magnusson SP, Schjerling P,Kjaer M. Coordinated increase in skeletal muscle fiber area and expres-sion of IGF-I with resistance exercise in elderly post-operative patients.Growth Horm IGF Res 2010;20(2):134–40 [Epub 2009/12/25].

56] Steinhilber B, Haupt G, Miller R, Boeer J, Grau S, Janssen P, et al.Feasibility and efficacy of an 8-week progressive home-based strength-ening exercise program in patients with osteoarthritis of the hipand/or total hip joint replacement: a preliminary trial. Clin Rheumatol2012;31(3):511–9 [Epub 2011/11/17].

57] Giaquinto S, Ciotola E, Dall’armi V, Margutti F. Hydrotherapy aftertotal hip arthroplasty: a follow-up study. Arch Gerontol Geriatr2010;50(1):92–5 [Epub 2009/03/14].

58] Liebs TR, Herzberg W, Ruther W, Haasters J, Russlies M, HassenpflugJ. Multicenter randomized controlled trial comparing early versus lateaquatic therapy after total hip or knee arthroplasty. Arch Phys MedRehabil 2012;93(2):192–9 [Epub 2011/12/27].

59] Barker KL, Newman MA, Hughes T, Sackley C, Pandit H, KiranA, et al. Recovery of function following hip resurfacing arthro-plasty: a randomized controlled trial comparing an acceleratedversus standard physiotherapy rehabilitation programme. Clin Rehabil2013;27(9):771–84 [Epub 2013/04/12].

60] Holm B, Thorborg K, Husted H, Kehlet H, Bandholm T. Surgery-induced changes and early recovery of hip-muscle strength, leg-presspower, and functional performance after fast-track total hip arthro-plasty: a prospective cohort study. PLoS ONE 2013;8(4):e62109 [Epub2013/04/25].

61] Gremeaux V, Renault J, Pardon L, Deley G, Lepers R, Casillas JM. Low-frequency electric muscle stimulation combined with physical therapyafter total hip arthroplasty for hip osteoarthritis in elderly patients: a ran-


domized controlled trial. Arch Phys Med Rehabil 2008;89(12):2265–73[Epub 2008/12/09].

62] Larsen K, Hansen TB, Thomsen PB, Christiansen T, Soballe K.Cost-effectiveness of accelerated perioperative care and rehabilitation














































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































1 hysioth

[

[

[


after total hip and knee arthroplasty. J Bone Joint Surg Am2009;91(4):761–72 [Epub 2009/04/03].

63] Larsen K, Hansen TB, Soballe K, Kehlet H. Patient-reported outcomeafter fast-track hip arthroplasty: a prospective cohort study. Health Qual



Life Outcomes 2010;8:144 [Epub 2010/12/02].64] Nankaku M, Tsuboyama T, Kakinoki R, Akiyama H, Nakamura T. Pre-

diction of ambulation ability following total hip arthroplasty. J OrthopSci 2011;16(4):359–63 [Epub 2011/05/10].

[

Available online at www.s

ScienceD


65] Schega L, Hamacher D, Wagenaar R. Related to: Poster 150 A com-parison of effects of augmented reality and verbal information basedinterventions in elderly women after hip replacement. Arch Phys MedRehabil 2011;92(October (10)):1734–5.


66] Wagenmakers R, Stevens M, Zijlstra W, Jacobs ML, van denAkker-Scheek I, Groothoff JW, et al. Habitual physical activitybehavior of patients after primary total hip arthroplasty. Phys Ther2008;88(9):1039–48 [Epub 2008/07/26].

ciencedirect.com

irect


















































































































































































http://www.sciencedirect.com/science/journal/00319406