Symptomatic hip osteoarthritis is associated with poor general health status and surgical intervention (total hip replacement; THR) is the most effective treatment for end stage disease. Thisprocedure generally resolves pain, but function usually remains substantially sub-optimal. This protracted disability has detrimental economic, social and health consequences. ‘Standard rehabilitation’ (SR), (i.e. low intensity exercise, not involving progressive resistance training (PRT)), typically permits patients to regain basic levels of function but fails to resolve thesignificant muscle wastingand subsequent strength deficits associated with the condition. Supervised PRT following THRproduces good results in terms of muscle strength and function. However, deliveryof this type of program is expensive due to the high costs associated with supervision, facility provision, and transport of patients. A home-based program featuring relatively high intensityPRTbut not requiring a high degree of supervision,would potentially overcome theseproblems. Before commencing this study, evidence was lacking regarding whether home-based PRT regimes with weekly supervision in the early postoperative period were effectivein restoring muscle mass and physical function in THR patients.
Chapter 2describes the results of a pilot randomised controlled trial comparing a home-based PRT program with weekly supervision in the early postoperative period after THR against SR (control) in terms of improving muscle strength and physical function at up to 1 year follow up. A prospective single blinded study (assessor blinded to results of randomisation) was conductedover a 2 year period (April 2010 to March 2012). Of 50 patients initially recruited (home-based PRT n =26, SR (control) n=24) after informed consent, 26 completed 9-12 monthsfollow up(home-based PRT n=13, SR (control) n=13). There was no effect for treatment (home-based PRT or SR(control) ) in terms of the primary outcome measure assessed, i.e. maximal voluntary contraction of the operated leg quadriceps(MVCOLQ in Newtons (N)over theperiod of follow up. As anticipated, there was a significant effect of time (i.e. improvement)in the primary outcome, with improvements in the secondary outcomes as well. The exceptionwas the leanmass of the operated leg,which showed no improvementover the
5period of follow up. Being in the SR(control) group as opposed to the home-based PRT group led to significant improvements in three of the secondary outcomes assessed; GS (estimated effect 0.185m/s; p=0.009), SCP (estimated effect -5.665s, p=0.038), and 6MWT (estimated effect 86.393m, p=0.004) at 9 to 12 month follow up. This study led to the conclusion that early home-based PRT is deliverable and well tolerated but is not successful in providing functionalgain beyond that achievable by SR in this population. Due to the large loss to follow up (30%), the results need to be interpreted with caution. As a pilot study the results indicate that there is no need to perform a definitive trial of the home-based PRT intervention due its lack of effect on the primary outcome variable assessed. Centre-based regimes appear to be the only modality that is able to provideadditional functional benefits in the early period following total hip replacement surgery, perhaps due to the supervision afforded to patients and hence compliance to higher intensity training.
Although SR was found to confer some benefits over home-based PRT in terms of the secondary outcomes assessed in Chapter 2,SR post-THR is not well defined in the existing literature. Thus, Chapter 3aimed to investigate the nature of standard care that exists in the UK post-THR. Questionnaire item development about standard rehabilitationpractice was guided by an initial focus group interview (after informed consent) with practising physiotherapists.Anonline questionnaire was thensent via email to physiotherapists working in the UK from January to May 2011. 106 responses were obtained from a total of 130 physiotherapists’ contacted (81.5% response rate), with the physiotherapists consideringthat the most important muscles to target in all phases of rehabilitation being:the hip abductors (62.2%), followed by the quadriceps (16.9%), and other muscles (21%). Exercise type prescribed revealed no consensus, with weight bearing (42%), functional (45%) and bed-based/bridging/postural exercises (13%) favoured. 83.7% were able to define the basis of progressive resistance training (PRT), but only 33% prescribed it. The study concluded that standard rehabilitation in the UK after THR is variable, and appearsto rarely include PRT. This could be a factor contributing to the prolonged poor function in some patients.An attempt was then made in Chapter 4to assess on a molecular level, the changes that occur in the vastus lateralis(VL) of patients with end-stage hip osteoarthritisand duringthe early phase of rehabilitationfollowing THR. mRNA expression was assessed using reverse-transcriptase polymerase chain reaction (RT-PCR) from VL muscle biopsy samples obtained from patients intraoperatively and at intervals up to 9-12 months postoperatively(6 weeks,6 months and 9-12 months). The gene panel for RT-PCR was chosen in terms of anticipated activity with regards to the metabolic processes of muscle hypertrophy, atrophy, lipid metabolism and inflammation. It was hypothesised that there would be no effect of biopsy site(proximal versus distal)on gene expression of the VL muscle in patients at the time of THR surgery,and that in patients undergoing early home-based PRT as opposed to SR(control), there would be an increase in genetic markers of hypertrophy and lipid metabolism, with a decrease in genetic markers of atrophy and inflammation.The results of the former analyses showed that hip joint inflammation appeared to have no effecton gene expressionon samples taken from 2 sitesin the VL, suggesting that for these sorts of analyses, single site muscle sampling is appropriate. With regard tothe latter hypothesis;muscle inflammation in the VL of the operated leg at the 6 week time point was reduced. Despite increases in markers of hypertrophyover the period of follow up, these did not reach significance. Significant reductions in markers of lipid metabolism were found (at 6 weeks) and this perhaps warrants further investigation with regards to metabolic efficiency in this group of patients. Participation in the home-based PRT regime did not demonstrate an objective difference in mRNA expression of the genetic panel chosen, confirming at a cellularlevel, the lack of effect on leg lean mass andobjectively assessed function observed (Chapter 2).
Chapters 5 and 6 describe studies performed in order to investigate the effect of psychological distress and behavioural cognitions (issues that can impacton a patient’smotivation to participate in an exercise programme) on objective and subjectively assessed physicalfunction of the patients recruited to this study. Chapter5 used the DRAM (distress and risk assessment method) to assess the impact of psychological distress on the primary outcome measure of the main study (MVCOLQ) as well as the Oxford Hip Score (OHS) and a reduced version of the Western Ontario and McMasters University Osteoarthritis personal function scale (rWOMAC PF) at 9-12 monthsfollow up (whilst controlling for randomisation into either the home-based PRT or SR(control) groups). The DRAM stratification (‘normal’ or ‘at risk/distressed’) was found to be predictive of subjectively assessedfunctionwhilst it had no impact on the MVCOLQ at 12 months follow up. Patients who were ‘at risk/distressed’had persistently lower function scores (p value range 0.001 to 0.04)), both preoperatively and at all postoperative time points,relative tothe ‘normal’ patients. This investigation wasthe first in the literature to use the DRAM tool in patients undergoing THR (it is typically administered to patients undergoing spinal surgery) and its routineuse in the screening of patients undergoing THR is indicated.However, behavioural cognitions (Recovery Locus of Control, RLOC; Theory of Planned BehaviourPerceived Behavioural Control (TPB PBC)) didnot show any impact on OHS nor rWOMAC PF in this population (again with allocationinto either home-based PRT or SR (control) groups controlled for; Chapter 6). Multiple bivariate associations were found to exist between the behaviour cognitions and the objective measure of physical function (MVCOLQ) as well assubjectively assessed function(OHS and rWOMAC PF) which warrants further investigation. The regression analysis revealed that improvement in behaviour cognitions between 6 weeks and 6 months appeared to have a negative impact on the amount of improvement in MVCOLQ at 9-12 months from pre-operative values. OHSand rWOMACPFat 9-12 month follow up, as well as the levels of functional gain over time, were best explained by the patients’ earlier functional status.
The home-based PRT program was then compared to standard rehabilitation (SR, control) from a health economics (cost consequences) analyses viewpoint(Chapter 7).Client service receipt inventories were available from 20 patients at final follow-up (9 to 12 months; Home-based PRT n=11, SR (control) n=9). The average cost per patient for physiotherapists to implement the home-based PRT programme was £313.95. Home-based PRT was £33 more expensive (bootstrapped95% confidence interval (CI)-£318, £366) than SR (control) with an incremental benefit of 13.71N in terms of the primary outcome measure(MVCOLQ; bootstrapped 95%confidence interval (CI) -54.64N, 83.83N). There was no difference between thegroups in terms of healthcare service utilisation at 12 months orQuality Adjusted Life Years (QALY). There was asignificant benefit genericallyfrom the THR operation (EQ-5D Health Utility Index (HUI) improved from 0.46 to 0.87 for whole cohort; a value compatible to data published elsewhere). In our investigation, EQ-5D HUI at final follow up was 10% better than a normal healthy population of age and sex matched individualsin the UK.
In summarythe main findings fromstudies conducted for this thesisare:Early home-based PRT is not successful in providing additional muscle strength nor objective functional gain beyond that achieved by standard rehabilitation programmes in elective THR patients. However, loss to follow up at final review (30%)for patients enrolled into this pilotstudymean that these results should be interpreted with caution Standard rehabilitation after elective THR in the UK is variable and appears to rarely include PRTCharacterisation of the molecular environment of the VL of the affected leg in patients in the early phase of post-THR rehabilitation demonstratesthat single site muscle biopsy sampling is sufficient,and processes that reflect protein breakdown (catabolism) appear to persist for 9-12 months following surgery with no obvious impact in favourof participation in the home-based PRT regimePsychologicaldistress assessed using the distress and risk assessment method (DRAM)is predictive of subjective outcome in patients undergoing THR independent of participation in an exercise rehabilitation programmeBehavioural cognitions have no impact on subjective function in this populationPreoperative functional status appears to be the most significant indicator of post-operative subjectively assessed functionParticipating in the home-based PRT regime as opposed to standard rehabilitation, SR (control),in this population of patients post-THR, costs on average£33 more per patient(bootstrapped 95% CI -£318,£366)If home-based PRT were to be reassessed in a further trial, this should be with a different regime to that assessed in this thesis; the current pilot study does not justify performing a full multi-centre randomised controlled trial.