Summary of main findings
The aim of the PEDAL trial was to assess clinical effectiveness and cost-effectiveness of a 6-month intradialytic exercise programme on QoL, compared with usual care for HD in the UK. The PEDAL trial is novel in that it is the first, to our knowledge, to evaluate intradialytic exercise as would most likely be delivered nationally, should NHS commissioning include exercise training as part of the service specification for in-centre HD. Exercise training was delivered by physiotherapy assistants (band 4 technical instructors), supported by regional co-ordinators who were physiotherapists or exercise physiologists. Thus, the PEDAL trial evaluated the most realistic yet safe exercise rehabilitation intervention possible in this population.
The baseline demographic data of the PEDAL trial cohort broadly aligns with the ‘real-world’ dialysis population in the UK as detailed in the latest data from the UK Renal Registry43 and also the PIVOTAL (Proactive IV iron Therapy in haemodiALysis patients) trial, a recent, large interventional study conducted in UK renal units.44 Baseline age, sex, BMI, blood pressure, diabetes prevalence and the proportion of current smokers are all similar to UK Renal Registry data.43 However, the current study included a larger proportion of black African and black Caribbean participants, and a lower proportion of white participants, than is representative of the broader prevalent dialysis population in the UK. The baseline data are similar to most dialysis populations worldwide.45–48 There are some differences in rates of diabetes prevalence, with lower rates in our patient cohort than in dialysis populations in the USA and Asia, which may limit application of the results to all other dialysis populations.
As delivered, the PEDAL programme did not improve the QoL of the whole sample, as assessed by the KDQoL-SF PCS (p = 0.055), nor did it improve QoL as assessed by the secondary outcomes of the EuroQol-5 Dimensions, five-level version (EQ-5D-5L) or KDQoL-SF multi-item scales. In the whole sample there were no statistical or absolute changes in functional capacity, physiological variables, anthropometric measures or clinical measures. More than half (58%) of patients temporarily halted receiving their exercise intervention for a continuous period of > 2 weeks during the study; in total, patients completed less than half [47%, interquartile range 28–77%] of exercise sessions that were prescribed. These data suggest that the HD patients had poor compliance with the exercise intervention. The nested qualitative study revealed that patient views relating to change in health status, having comorbidities and having days where there was an increase in symptoms such as tiredness were possible reasons for not engaging with the exercise intervention. Because predominant themes were the ‘exercise culture’ and camaraderie that results from exercising with others, a cluster randomisation design may perhaps have fostered this camaraderie on a whole renal unit level, preventing the perception of participants feeling ‘left out’ if they were allocated to the control group.
We chose the KDQoL-SF instrument because of its validity in CKD patients49 and inclusion of a generic core that has been widely used in CKD and other populations. The lack of effectiveness of the PEDAL programme in the whole sample on PCS score warrants comparison with previous studies. A Cochrane review9 completed in 2011 concluded that exercise was beneficial for QoL in CKD patients but, unfortunately, no meta-analysis or risk-of-bias assessment was performed on this outcome measure. Furthermore, many of the included studies were not representative of the HD population because they investigated transplant,50 pre dialysis51 or ambulatory peritoneal dialysis52 patients. More recent meta-analyses have concluded positive effects of exercise but have shown significant effects on the role physical subscale only,53 showing no effect on PCS score,18 or did not analyse this outcome owing to insufficient studies.13,54 Other meta-analyses have included non-randomised studies53 or rated studies as at high risk of bias and showing considerable heterogeneity.14,17,19,54 Previous meta-analyses have also included extradialytic exercise programmes19,54,55 or studies that have delivered progressive resistance training as opposed to aerobic cycling alone.10,15,16,19,20,56 In this regard, one meta-analysis22 usefully compared aerobic exercise, progressive resistance training and combined exercise; only progressive resistance training increased PCS score. Detailed analysis of the very few empirical studies included in reviews that do show positive effects of aerobic intradialytic exercise on QoL reveals that they have often utilised unrealistic interventions that would be difficult to implement in routine care. For example, Ouzouni et al.57 investigated an exercise programme similar to the PEDAL programme [intradialytic cycling plus TheraBands (TheraBand, Akron, OH, USA) and ankle weights] but three physiologists were employed to intensely supervise only 10 patients for 10 months to obtain a significant effect. Taken together with the results reported herein, it is highly unlikely that clinically implementable intradialytic aerobic exercise training per se increases QoL, particularly at a whole-population level.
Nevertheless, there was some evidence of benefits to KDQoL-SF-assessed QoL and aerobic capacity in patients who completed > 30% of prescribed exercise sessions. Participants, clinicians and researchers continue to advocate intradialytic aerobic exercise. The PEDAL trial uniquely assessed the cost of delivery of its intervention using careful recording of harms and using health economic methods. The number of hospitalisations, all-cause mortality and cardiovascular mortality were not different between the groups. Although these results should be interpreted cautiously because of the lack of events for these secondary outcomes, there was no obvious increase in SAEs in the intervention group. The economic cost of delivering the PEDAL intervention ranged from £882 to £1866 per patient per year (depending on pay band of the physiotherapy assistant, whether or not London weighting was applied and staff–patient ratio). This calculation assumed that intradialytic exercise would be offered as part of a general physiotherapy service (with enough capacity to provide absence cover), with costs profiled throughout the programme (e.g. exercise equipment cost is shared throughout its service life), with patients who are 100% compliant and with physiotherapy assistants supervising between 6 and 10 patients per dialysis session without incurring any travel costs. For context, the cost of delivery of HD in the UK is approximately £35,023.58 Because the PEDAL trial was not clinically effective at a whole-sample level, whether or not the cost of delivery of intradialytic exercise is justified to enhance patient choice remains a matter for debate. Because the observed benefits to KDQoL-SF-assessed QoL may be due to an experimenter effect, a similar improvement in QoL to that achieved herein could arguably be achieved with less cost and less effort through one-to-one patient interactions and provision of patient education resources on physical activity.
The PEDAL trial aimed to address a criticism of previous exercise studies that recruited only patients with high physical health status by having more inclusive inclusion criteria. Perhaps including patients with lower physical health status prevented benefits of the exercise intervention being realised. However, this explanation is unlikely because our sample is like other pragmatic exercise studies in CKD patients (e.g. in the seminal study of Painter,59 baseline PCS score was 36 AUs). Besides, poor baseline physical health status is not a reason for lack of response to exercise intervention, because patients with poorer PCS scores respond better to intervention than those with higher PCS scores.7
Health economic evaluation
The economics evaluation suggests that, over 6 months of implementation, the PEDAL intervention resulted in higher NHS costs with no gain in QoL as measured by the EQ-5D-5L. A number of factors should be borne in mind; for example, the benefit from the PEDAL intervention could be avoiding cardiovascular events and other illnesses in future years, and the 6-month time horizon would not capture this. In addition, the PEDAL protocol could be refined to target different patients on dialysis and/or to include a time point at which the benefit of the intervention was assessed with only those reporting a good QoL gain continuing. Both of these could potentially improve the cost-effectiveness of the intervention.
Qualitative analyses of patient views
The explanatory theory developed through the qualitative substudy of the PEDAL trial has provided in-depth insights into the influences on people’s decision-making relating to participation both in the trial and in the intervention over time. Themes are supported by the few previous qualitative studies; however, our data are more extensive, and the resulting theoretical explanation goes further to enable suggestions for future implementation and intervention sustainability. A novel finding relates to the importance of culture, and leadership to enable this, in the renal unit. Providers interviewed felt that the trial did not impinge much on their jobs, which supported trial implementation; however, this may not be conducive to sustainability of the intervention beyond the trial. Moving forward beyond the PEDAL trial, it will be most important to influence the attitudes, beliefs and behaviours of providers. Where providers are engaged and empowered, they are much more likely to engage and empower patients in participating. Once a patient is participating (where capable), the numerous physical, functional, psychological and social benefits are likely to support their ongoing participation. This in itself may then feed forward into a positive, proactive culture in the renal unit. Further research is needed to explore the complexities of intervention implementation in the longer term, with a view to aspects of behaviour change and insights from implementation science.
Limitations
The study was designed to assess a pragmatic, clinically implementable intradialytic exercise intervention. By design, the study relied on a patient-reported outcome measure for its primary outcome; it is recognised that the primary limitation of this study was the lack of an attention control group. To alleviate this concern, multiple outcomes were assessed, and objective measures were obtained whenever possible; only objective measures of habitual physical activity were not of sufficient quality to be presented in this manuscript (owing to poor patient compliance with wearing accelerometers). In addition, many patients withdrew from the study (following randomisation but before baseline, and also during follow-up) or were too unwell to complete all functional capacity and physiological outcome assessments. This particularly affected the assessments of peak aerobic capacity but is to be expected in a study with inclusive inclusion criteria and arduous outcome assessments. Finally, the study was not powered to detect differences in some secondary outcomes, including mortality. Nevertheless, we reported these data to allow a balance of benefits and harms to be assessed. Future studies should address these concerns by including attention control groups and being adequately powered to investigate minimal clinically important differences of all outcome measures. Based on the findings presented herein, studies should address poor compliance by investigating exercise counselling and behaviour change interventions (such as motivational interviewing and self-determination theory) and should specifically target ways to overcome complications associated with concurrent medical events (such as fatigue). Investigators should consider alternative mechanisms of action of exercise interventions, including self-efficacy and stage of change.43 Finally, the benefits of including progressive resistance training should be confirmed, even if extradialytic delivery is required.
Conclusion
A pragmatic intradialytic aerobic exercise programme, which could realistically be commissioned as part of routine care, did not improve QoL of this deconditioned sample of HD patients. The lack of clinical effectiveness and cost-effectiveness was probably due to poor compliance and only modest changes in aerobic capacity, particularly in a subgroup of patients with concurrent medical events and associated fatigue.
Future work
The benefits of longer interventions, including progressive resistance training, should be confirmed, even if extradialytic delivery is required. Future studies need to evaluate whether or not there are subgroups of patients that may benefit from this type of intervention, and also whether or not there is scope to optimise the exercise intervention to improve compliance and effectiveness. This should include further qualitative work.