Clinician's Commentary on Moncion et al.

Exercise-based cardiac rehabilitation (CR) is a multidisciplinary approach that is associated with significant reductions in morbidity and mortality as well as improvements in cardiorespiratory fitness (CRF) and thus is a Class 1A recommendation for people with cardiovascular disease.^1,2 The studies that this recommendation was based upon were largely conducted in the outpatient phases (phase II and III) where CR is predominantly delivered. Following the outpatient phases of CR is the less commonly available maintenance-phase (phase IV).³ Phase IV programs are usually voluntary and paid for out-of-pocket, although this may vary by country and program.³ Data on maintenance CR is scant, with a recent meta-analysis identifying only 10 randomized controlled trials since database inception.³ How well these phases are coordinated and the continuity of care of delivery is unknown. Having a seamless transition through these phases may be important, as an emerging system-based factor that has been shown to affect CR outcomes is the amount of elapsed time from the cardiac event to start of outpatient CR (wait time). In general, longer wait time to initiate phase II and III CR programs have been associated with mitigated improvements in CRF and other important outcomes.^4–7 There has only been one study that has examined the effect of wait time on outcomes in phase IV maintenance programs.⁸

In this issue of Physiotherapy Canada, Moncion et al., add to the limited literature by presenting a retrospective analysis of data to determine if wait time to enrolment in phase IV CR has an effect on change in CRF trajectories over five years in men and women. CRF is a key outcome measure given that it is a strong independent predictor of cardiovascular and all-cause mortality.^9–15 In this study, retrospective patient data were collected from a 31-year period (January 1985 to December 2016) of 599 patients who had participated in maintenance CR. Of these patients, 20.8% of women (n=32) and 33.9% of men (n=151) satisfied the study inclusion criteria with most being excluded for not performing ≥2 cardiopulmonary exercise tests (CPET). The results revealed that there was no sex difference in wait time to start of the maintenance program (1.8±3.0 years for men and 1.3±1.7 years for women; p=.2). The primary analysis demonstrated no association between change in CRF and elapsed time to enrollment after the cardiac event x enrolment duration over a period of up to 5 years of CR participation in men or women. In addition, predicted trajectories suggested a modest increase in CRF for women who enrolled early (6 months) and late (1.5 years) of 1.3 and 1.6 mL·kg⁻¹·min⁻¹ respectively with males improving over the first 2 years (1.3 and 1.5 mL·kg⁻¹·min⁻¹ respectively) and then declining to below baseline CRF levels at 5 years. The number of observations for early and late groups or by sex were not reported. In mixed-model analysis, baseline age and CRF were included as covariates. The authors conclude that “Early enrolment in CR is recommended and encouraged, but the benefits of long-term CR are possible despite delays.”

No Effect of Wait Time on Cardiorespiratory Fitness

The authors are to be commended for an examination of a phase of CR that is understudied and for disaggregating their results by sex. The authors cite only one other study that had examined the effect of wait time to maintenance CR enrolment on CRF and it was conducted by their group in the same CR program but without disaggregating results by sex.⁸ The study examined data from 62 of 486 patients who had enrolled over a 10-year period (2004–2013) for a duration of 18 months. The result was that a delay of ≥114 days (the median delay) resulted in mitigated improvement in CRF compared to earlier enrolment. This result was in contrast to result of the current study but aligned with results from earlier phase studies.^4–7 The authors explain that the contrasting result in their earlier study may have been owing to their pre-post design that did not allow for identification of non-linear changes in CRF over time, and the smaller sample size that did not allow for a mixed model analyses.

A close inspection of temporal changes in previously published earlier phase studies (II and III) help to shed some light on why no association was demonstrated in the current study. These studies demonstrated mitigated change in CRF and exercise capacity the longer the delay in enrolment.^4–7 In one study, our group examined 6497 post coronary artery bypass graft surgery patients and the effect that wait time to enrolment had on CR outcomes after adjusting for socioeconomic status, driving distance to CR, baseline CRF, medical history/complications, sex, age and many other covariates.⁴ There was a progressively greater decline in the relative improvement in CRF from a peak of 23.7% after a wait time of up to 8 weeks post event to ≤9.2% improvement after a wait time of ≥6 months. Therefore, as the mean wait time in the current study was beyond this, at 1.8 years for men and 1.3 years for women, there would not be much more room for wait time to further mitigate improvements in CRF. Indeed, in the current study, early enrollment was defined as 6 months, which was where a threshold occurred in our previous study beyond which little further decline in CRF occurred. Furthermore, in contrast to the current study, our study demonstrated for the first time that female sex was an independent and significant predictor of longer delay to CR, even after we accounted statistically for factors that predict delay to CR entry.⁴

Predicted Change in Cardiorespiratory Fitness over 5 years of Participation

Regarding the secondary outcome, the current study reported modest predicted changes over 5 years in CRF in women and an increase in the first 2 years then a subsequent decline in men to below baseline levels. The reason for these modest changes, if any, may be multifactorial. First, it is possible that these patients had participated in CR previously so that in the maintenance phase their exercise prescriptions were minimally changed, if at all. However, information on previous CR participation or the change in exercise prescription over the 5-year period was not available. Second, despite having conducted a graded exercise test with gas exchange analysis, the method of prescribing intensity was based on a percentage of heart rate reserve. While this is a guideline-based approach, it does not take advantage of a more tailored method of exercise prescription using the anaerobic threshold, especially when most patients do not reach a true physiological maximum on the exercise test.^16,17 Indeed, for an individualized and metabolically uniform prescription, exercise at the ventilatory anaerobic threshold should be prescribed to men and women when the data is available.^18–21 It was unclear why respiratory gas exchange was measured during exercise testing in the current study's CR program but not used for exercise prescription. Finally, in meta-regression analysis, earlier training post myocardial infarction has been associated with greater beneficial effects on remodelling (increased ejection fraction and decreased end diastolic and systolic volumes).²² Therefore, as the authors of the current study point out, the patients may have been beyond the window of time to CR entry where larger changes in left ventricular remodeling occur.

While this single group study conducted by Moncion et al., adds to the literature, change in CRF relative to a control group is of clinical importance as one might expect a decay in CRF over time in maintenance CR non-attenders. A recent meta-analysis examined the effects of phase IV studies that had included a control group.³ Randomized controlled studies were extracted from databases from inception to January 2020. There were 10 trials of 21 publications identified, all were conducted in North America, Europe, or Asia, with only 3 trials that included data on sex.³ Pooled analysis revealed that there was no significant effect of maintenance CR on change in CRF or exercise capacity (METS) compared to usual care. Further, there was no change in functional capacity in women compared to usual care or in men compared to active comparison. Despite the lack of improvement in CRF, there were improvements in quality of life and plasma lipids compared to usual care.

Maintenance Cardiac Rehabilitation

Regarding the main outcome of this study, there was no evidence that earlier entry to maintenance CR augments change in CRF in men or women. Therefore, while benchmarks for timely enrollment in the early phases of CR have been published,²³ there is currently insufficient evidence of the need for benchmarks for entry to maintenance CR. Furthermore, there is limited data demonstrating the benefits of maintenance CR. However, maintenance CR may provide an opportunity to participate in CR for patients that were not able to participate in the earlier phases for unavoidable reasons or because they were not at the stage where they were ready to make behavior changes etc. In this respect, maintenance CR may favour enrolment of a greater proportion of women, given that they are less likely than men to attend phase II and III CR but this should be examined in future studies.^24,25

Maintenance programs are generally paid for out-of-pocket, as in the current study, but programs often do not have the resources to support them and thus they are less commonly available.³ However, with the success of virtual models of CR delivery developed during the COVID-19 pandemic,²⁶ use of these models alone or combined with on-site visits (hybrid models) may provide a lower resource option for graduated patients who do not require close monitoring.²⁷ Future studies are needed to examine the effect of these program delivery models as they may make it feasible for phase IV programming to be delivered at the same site as earlier phase CR. Integration of these models may also help to improve the continuity of care, increase the number enrolled, and ultimately provide long-term support for women and men with cardiovascular disease.