Can a bout of exercise harm the human heart?

Regular exercise provides a number of health benefits, including favourable effects in the cardiovascular system. The American College of Sports Medicine recommends healthy adults perform 150 min of moderate‐to‐vigorous aerobic exercise per week (Garber et al. 2011). However, the health benefit of exercise doses beyond current recommendations is debatable. As ultra‐endurance events (e.g. marathons and triathlons) have become increasingly popular, it is critical to explore the effects of long‐lasting, high‐intensity exercise on the cardiovascular system. Recent studies have led to speculation that prolonged, high‐intensity exercise can provoke adverse cardiac and vascular remodelling and transient cardiac dysfunction. Therefore, further research is warranted to determine the relationship between exercise dose and cardiovascular function in order to analyse the risk‐to‐benefit ratio.

In the recent study by Stewart et al. (2016) in the Journal of Physiology, cardiac function and cardiac injury biomarkers were examined in ‘recreationally active’ men (27 ± 4 years old) before, during and following acute bouts of endurance exercise to determine the effects of exercise intensity on cardiac performance. It was hypothesized that transient cardiac dysfunction and biomarker release would be significantly greater following heavy‐intensity exercise compared to moderate‐intensity exercise. Participants completed two cycling trials of 90 min (heavy intensity) and 120 min (moderate intensity) on separate visits. Total mechanical work performed during each trial was matched within individuals, thereby enabling comparisons of functional changes between exercise trials. Twelve‐lead electrocardiograph and two‐dimensional echocardiograph recordings were taken along with blood samples at baseline, during exercise (50%, 75% and 100% of cycling time) and in recovery (1, 3, 6 and 24 h). Furthermore, at each time point, electrocardiograph and echocardiograph measurements were taken at rest and during a standardized low‐intensity exercise challenge on a semi‐recumbent cycle ergometer. The inclusion of measurements during low‐intensity exercise was a strength of the study, as it controlled for any acute changes in blood pressure which may occur post‐exercise, providing stable loading conditions on the heart during exercise and recovery periods.

Stewart et al. (2016) observed reductions in left ventricular (LV) global longitudinal strain and right ventricular (RV) global longitudinal strain following the 90 min trial, peaking at 1 h post‐exercise and persisting for > 24 h into recovery. Interestingly, only RV global longitudinal strain was reduced in the 120 min trial; however, it returned to baseline after 24 h of recovery. Stroke volume was significantly decreased half‐way through both cycling trials, and persisted for 1 h following the 120 min trial and 3 h following the 90 min trial. Cardiac injury biomarkers, such as serum cardiac troponin (cTn), increased during both trials; however, the elevations were significantly greater in the 90 min trial.

The current study demonstrates that transient reductions in cardiac function and increases in cardiac injury biomarkers are present following endurance exercise, and can persist for 24 h following exercise. However, the magnitude and time course of such responses depend on the exercise intensity–duration stimulus. Nonetheless, several factors should be considered when interpreting study findings, including (i) the training history of these ‘recreationally active’ participants, (ii) the exercise study stimulus, and (iii) the cardiac structure and function outcome measures reported.

Stewart and colleague's (2016) study recruited ‘recreationally active’ men who were currently engaged in endurance sports; however, the authors loosely quantified the exercise training history of their sample. Prior training history, including exercise intensity and duration, may influence cardiac structure, function and rhythm, which in turn may impact their responses to a single bout of exercise. While there is no established criterion, those who meet physical activity guidelines (i.e. 150 min of moderate‐to‐vigorous exercise per week) could be identified as ‘recreationally active’. The cohort examined in this study exceeded this criterion by three‐fold with an average of 480 ± 180 min of cycling per week. Their average maximal oxygen intake of 54.1 ± 5.9 ml kg⁻¹ min⁻¹ would be considered superior, and they were habitually active with an average exercise training history of 10 ± 8 years. Collectively these observations suggest the cohort examined in this study were highly trained ‘sub‐elite athletes’ and thus their characterization as ‘recreationally active’ may not be suitable. Detailed information on current and long‐standing training history may be important in interpreting the study findings, and therefore future physical activity and exercise training studies should include detailed reporting of exercise training history with variables, like METS‐mins per week or TRIMPS (training impulse), where intensity and duration are considered. While a longitudinal study design would provide the greatest insight into the effects of regular and chronic exercise training on cardiovascular function, the time commitment required may pose a challenge. Thus, perhaps a cross‐sectional study examining a truly ‘recreational’ cohort, sub‐elite and elite athlete cohorts would be sufficient for exploring the effects of training.

Participants in this study were required to complete two separate cycling trials after abstaining from exercise for a minimum of 48 h. For the purposes of testing the study hypothesis, the use of a controlled intervention was a methodological strength and was necessary to ensure results were not influenced by any preceding stimulus. However, this design poses a limitation in the ability to generalize the results to the cohort in the present study, given that they partake in an average of 8 ± 3 h of cycling per week and are not likely to abstain from exercise for two consecutive days. As a result, there is a loss of ability to understand how heavy and moderate intensity exercise within a typical training regime affects the heart. Given that the present study observed cardiac decrements to persist for 24 h following heavy‐intensity exercise, it is arguably more important to understand the effect of cumulative bouts of exercise with limited recovery (e.g. < 48 h as was given in this study) on cardiac function. Previous work by Oosthuyse et al. (2012) suggests that diastolic function is reduced and fails to return to baseline each day following four consecutive days of cycling for 3 h per day. However, the authors noted the magnitude of diastolic dysfunction was not cumulative, and instead the repetitive strenuous bouts primed the cardiovascular system, and compensatory mechanisms were observed. Thus, cardiac function varies depending on the preceding stimulus to which an individual is exposed. The results of the current study should be taken with caution given that only one bout of exercise following 48 h of rest was explored. Therefore, while the exercise stimulus was appropriate for addressing the study hypothesis, it was not representative of typical training for this cohort. Future studies should explore the effects of exercise in ‘real‐life’ settings, in which individuals complete protocols within their typical training schedule.

The echocardiographic assessment of cardiac function in athletes has been reported extensively. There are profound differences in the cardiac structure and function of athletes following both acute and chronic exercise training. In the current study, Stewart et al. (2016) evaluated LV and RV systolic function by reporting on the average of echocardiographic‐based global longitudinal strain from multiple wall segments. Therefore, while this study is fairly novel because of the repeated cardiac function measurement over a 24 h post‐exercise period, an opportunity to report on the integration of a closed‐loop cardiovascular system was missed. It is arguably important to consider the entire cardiac cycle (including both systole and diastole), cardiovascular system (including atria, ventricles and connecting vasculature), and potential regional wall differences (e.g. septum versus the lateral wall segments in the ventricle). Previous work would suggest the following about cardiac function and prolonged exercise: (i) impairments in diastolic function may exist and precede impairments in systolic function, (ii) differences in regional wall function may exist and cannot be distinguished when only reporting global longitudinal strain, and (iii) atria are also subject to changes in structure and function with exercise, including increased wall tension and transient conduction delays in the atria and, therefore, are not merely volume reservoirs (Wilhelm et al. 2014). Interestingly, Stewart et al. (2016) have recently published additional work on these young athletes highlighting that changes in ventricular systolic strain following high‐intensity exercise are more profound in the right ventricle than in the left ventricle, albeit still negating the role of atrial function and connecting vasculature. As well, the authors acknowledge that reductions in LV longitudinal strain may be unique to the septal myocardium and may result from ventricular interactions secondary to exercise‐induced RV dysfunction.

In conclusion, Stewart et al. (2016) provide some novel insight into the magnitude and time course development of the effects of endurance exercise on cardiac systolic function. It appears that exercise‐induced reductions in systolic function and increases in injury biomarkers persist for 24 h following exercise; however, manipulating the preceding stimulus can alter this magnitude and time course development. While the results of this study suggest prolonged high‐intensity exercise may provoke unfavourable transient cardiac dysfunction, it is important to consider whether these findings are clinically relevant. Future work should explore whether these transient impairments in cardiac function accumulate over time to cause pathological adaptations, or if they are instead a necessary physiological stimulus for eliciting beneficial training adaptations.

Additional information

Competing interests

None declared.

Author contributions

All authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

No funding was obtained for this review.