High intensity interval training in coronary artery disease patients, is it worth the effort?

Coronary artery disease (CAD) is the leading cause of death in Europe1 and worldwide.2 Based on a large body of evidence that shows that participation in cardiac rehabilitation (CR) reduces CAD mortality and morbidity3 and improves exercise capacity4 and quality of life,3 referral to exercise-based CR is now a class I recommendation for CAD patients.5 Nevertheless, the effectiveness of CR depends largely on the intensity, duration, frequency and type (FITT) of exercise;4,6,7 although the optimal exercise prescription remains under debate. For decades, there has been a historical preference in the guidelines to recommend moderate intensity continuous training (MCT) for patients with CAD.5,6 Yet, approximately 15 years ago, high(er) intensity interval training (HIIT) alternating high intensity (80–120% of peak oxygen consumption (VO2)) bouts with periods of active or passive recovery was introduced in CR by the Norwegian group of Wisloff et al., showing superior clinical improvements in patients with CAD8 and chronic heart failure (CHF).9 Among others, these proof of concept studies reported a two to three times larger increase in exercise capacity after HIIT compared to MCT.8,9 Since then, a growing number of trials has compared the effectiveness of HIIT and MCT in CAD patients and other chronic disease populations. Although the results of these small individual studies were contradictory and inconclusive, pooling their data by means of meta-analytic techniques unambiguously showed higher clinically relevant improvements in peak VO2 in CAD patients after HIIT ranging from 1.53 to 1.78 ml/kg/min.10–13 Since the publication of the largest meta-analysis,10 involving nine studies and 206 CAD patients, the multicentre study SAINTEX-CAD14 was published almost doubling the number of CAD patients included so far. In the present issue of this journal, a systematic review and meta-analysis was performed by Gomes-Neto et al.15 to provide an updated analysis of data from randomised controlled trials comparing HIIT to MCT in patients with CAD. The primary outcomes were exercise capacity and quality of life. The overall results of this study, involving data of 11 trials and 594 patients, are in line with previous meta-analyses10–13 and confirm a significantly larger effect size for exercise capacity (+1.25 ml/min/kg) in favour of HIIT. However, the novelty of this study is to be found in a subgroup analysis based on whether the exercise protocols were or were not isocaloric. When Gomes-Neto et al.15 pooled the data of four trials that matched their exercise interventions for energy expenditure (EE), they found similar improvements in exercise capacity (+0.36 ml/kg/min (–0.14 to +0.85) P = 0.16) following HIIT and MCT. These results suggest that not the intensity nor the modality but the total EE of the exercise programmes determines the improvement in exercise capacity after training. This concords with a recent meta-analysis by Kraal et al.,4 who showed that session duration, programme length and training intensity were all related to the improvement in exercise capacity, but found no independent effect of any of these training characteristics after correction for total EE in patients with CAD. As such, one could conclude that HIIT offers a time-efficient alternative for MCT as it requires less time (approximately 30 minutes/week) to be spent exercising while providing similar health benefits compared to MCT. As a result, HIIT can overcome one of the most frequently cited barriers to physical activity, which is lack of time.16 Nevertheless, these results also imply that one can achieve the same gains with less strain: i.e. if we spend enough minutes exercising at moderate intensity we obtain the same improvements, which, from an adherence perspective, might be as important as saving 30 minutes per week. The challenging part of the present meta-analysis15 is observed when the data of seven trials that applied non-isocaloric exercise interventions were pooled for analysis. In this subgroup, the increase in exercise capacity following HIIT significantly exceeded that of MCT (+1.87 ml/kg/min (1.1 to 2.64) P < 0.001). Given the lack of a description of EE in these studies, this observation is difficult to interpret. If we speculate that EE is higher in HIIT compared to MCT, this result would be in line with the above-mentioned conclusion. Yet, from the present data it is actually unclear whether caloric expenditure was higher in the HIIT arm compared to the MCT arm. Earlier meta-analyses17,18 performed in obese patients clearly demonstrated that when training protocols were not matched, EE was usually less in HIIT compared to MCT. In patients with CAD, Pattyn et al.19 measured the EE of HIIT and MCT programmes as planned in the SAINTEX-CAD study (based on Wisloff et al.),9,20 and the actually performed training intensities.14 In a sample of 18 male patients (mean age 62.4 years) they reported that EE was less in the HIIT group compared to the MCT group in the study of Conraads et al.,14 whereas they established the isocaloricity of Wisloff’s original exercise protocols.9,20 If we now assume that caloric expenditure was lower in the HIIT arms of the current meta-analysis, this implies that intensity and/or modality are important, but not EE. This is in contrast to the previous observation. So the question stays out there and more large and well-designed randomised controlled trials are needed, assessing EE and properly describing the FITT characteristics, in order to feed this intensity versus EE versus modality debate. Even without this discussion, do we need to recommend HIIT as the first contact with exercise to patients already at increased risk? Time-saving seems so far the main driver to recommend HIIT to CAD patients. Assuming that both exercise programmes are isocaloric, HIIT would save 30 minutes per week for the same increase in exercise capacity.19 Whereas 30 minutes seems little taking into account travel time of patients to supervised settings, this time difference might become interesting from a home-based perspective. However, the two large multicentre studies comparing HIIT and MCT, SAINTEX-CAD14 in CAD patients and SMARTEX21 in CHF patients, highlighted challenges around the feasibility of HIIT in daily CR practice. Indeed in these studies, the majority of participants were unable to follow the HIIT exercise protocol accurately. Moreover, active coaching seems necessary if we want our patients to train in the right heart rate zone.22 Whereas we can all agree that training intensity and probably also progression was too low in SAINTEX-CAD14 and SMARTEX,21 we have to acknowledge that the actually performed training intensity more likely reflects clinical practice of CR in which it is not possible to have a personal coach standing next to the patient during each single session. Another important barrier for physical activity and exercise is the ‘like’ factor. Therefore, patients’ preferences should be met during CR in order to optimise adherence and increase intrinsic motivation to exercise. In this context, it is imperative that patients find pleasure in exercising. Yet, Zenko et al.23 clearly showed that in sedentary and low-fitness participants, pleasure reduces with increasing exercise intensity. In this sense, HIIT might negatively influence compliance and the adoption of a lifelong physically active lifestyle. To date, none of the published comparative HIIT and MCT trials interviewed participants about the enjoyment and attractiveness of the training protocols, but the findings from Zenko et al.23 may also hold true for cardiac patients. The most important long-term outcome of CR is lifelong adherence to physical activity. Nevertheless, almost all studies included in the meta-analysis of Gomes-Neto et al.15 investigated short-term effects of HIIT and MCT. Only the SAINTEX-CAD study reported 1-year follow-up results on physical activity and exercise capacity and found similar results after both training protocols.24 Finally, ‘first do no harm’ is a popular saying that originates from the Latin phrase, ‘primum non nocere’. It is a basic principle taught in many healthcare classes. Previous reports indicated that vigorous exertion increases the likelihood of acute myocardial infarction25 and sudden cardiac death,26 especially in sedentary persons irregularly performing vigorous physical activity. Is this not the average patient referred for CR? We currently do not have much information about the safety of HIIT. Levinger et al. reported that 8% of patients with cardiometabolic diseases experienced an adverse event in the acute phase of an interval training session.27 In addition, Rognmo et al., based on studying 4846 CAD patients, reported that the risk of a cardiovascular event during HIIT was low, although five times higher compared to MCT.28 It is important to mention that study participants included in the previous studies were clinically stable and well screened.27,28 In clinical practice, even a higher rate of adverse events in response to HIIT might occur. Consequently, it is recommended that risk stratification and screening with cardiopulmonary exercise testing29 is performed in all patients willing to engage in HIIT. Following here, the question remains whether we need to implement as a first contact with exercise something which in daily practice seems unreachable for the majority of patients, of which we do not know whether patients like it, that does not seem to influence quality of life more than current practice,15 that might be somewhat less safe, and for which we do not have any evidence that it will result in an improved uptake and adherence to a lifelong physically active lifestyle. Hence, we strive for a more gradual introduction of HIIT within the CR programme and recommend first to build duration at low to moderate intensity and subsequently to increase exercise intensity over time. In conclusion, we want to congratulate the authors for addressing an interesting and highly relevant topic. However, as a result of the methodological limitations, the paucity of high quality research and the significant heterogeneity among studies as highlighted by the authors,15 caution is needed when translating these findings to the regular cardiac patient. Prior to advocating HIIT, we urgently need properly designed trials focusing on long-term physical activity engagement and effectiveness of HIIT, its effect on mortality as well as the feasibility and safety of HIIT implementation in real-life CR regimens. The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author(s) received no financial support for the research, authorship, and/or publication of this article.