Endurance training in young athletes: What happens in childhood, stays in childhood?

Transition from the fetus to adulthood involves a series of structural, functional and metabolic changes to reach full maturity of all organs. Growth, sexual maturation and a change in body composition are central players in this process. The heart adapts to the increasing metabolic demands by increasing myocardial mass and chamber size, accompanied by an imbalance between the left (LV) and the right ventricle (RV) that translates in the ECG as a positivization of precordial T-waves.1 Puberty is a period of rapid changes, with large variability between individuals.1 Cardiac growth depends on a complex interplay of individual and environmental factors such as sex, body mass, and race. Non-fat body mass governs metabolic demands and drive cardiac growth during puberty.2 Superimposing an additional stress such as regular endurance training further increases heterogeneity and complicates the characterization of the heart of children athletes. In adults, regular physical activity is well-known to promote cardiac changes, so-called “athlete’s heart”, but whether these also develop in children and preadolescents, and if so how, remain much less explored. Indeed, there is a lack of well-calibrated normality values for trained children.3 Recently, Age- and body surface area-adjusted nomograms were derived from a large Italian registry of peri-pubertal football players.4 In this report, left ventricular diameters did not correlate to training load and were similar to those of non-athletic populations, suggesting little impact of regular physical activity on cardiac remodelling in children. Conversely, in a meta-analysis including data from 40 studies and over 14,000 young athletes competing in 30 different sports, McClean and collaborators found a typical “athlete’s heart” pattern, including left ventricle and atrial dilation and, interestingly, concentric hypertrophy.5 With the quest for professionalism on the rise since young ages, and an increasing pressure to succeed, a detailed knowledge of the heart of heavily trained young athletes, and their consequences in the long-term, is warranted. The Norwegian group led by Sebastian Sarvari recently characterized the cardiac remodeling ocurring in endurance-trained children by mean of 3D echocardiography and strain measurements. They studied 76 young athletes engaged in cross-country training, a highly demanding activity: at the age of 12, they had been training an average of 7 hours per week, for 5.4 years.6 In comparison with non-athletes, athletes presented with RV dilation and dysfunction and, interestingly, LV dilation and concentric hypertrophy, similar to McClean and colleagues data.5 In the current issue of the European Journal of Preventive Cardiology,7 the same group report follow-up data on those 48 individuals attending a second evaluation three years after baseline. Echocardiographic and fitness data are compared between those who kept training and those who were not involved in endurance training anymore. At 12-year-old baseline, both groups were comparable in terms of exercise capacity and cardiac structure and function. In this new evaluation, those who remained active improved their physical fitness to a larger degree, decreased heart rate at rest, and presented with LV dilation in comparison to those who abandoned regular endurance training. No differences between groups were noted in any other parameter, including left atrial size or systolic and diastolic function. Altogether, both publications6,7 add valuable data to the field, with two strengths that are particularly relevant and informative. First, follow-up data is provided through puberty. The authors elegantly reduce variability associated with puberty by recruiting a homogeneous cohort of athletes with a very narrow age range. These athletes are assessed at a right pre-pubertal stage (12 years old) and a right post-pubertal stage (15 years old). A second strength, not without drawbacks, is that not only current athletes were scanned, but also those who were training intensively at age 12 and had discontinued thereafter. This last group may be informative on the “sequels” of heavy training during childhood. The study design enables drawing conclusions on the effect of both puberty and endurance exercise discontinuation on cardiac remodeling. On the one hand, according to the authors data, puberty is characterized by the regression of RV dominance,1 as denoted by a reduction of indexed RV volume in both current and former athletes, and an increase in absolute cardiac mass and in left chambers size, as expected. On the other hand, those who quit training have a lower maximum oxygen uptake, a remarkably shorter time to exhaustion, a higher heart rate and lower indexed LV volumes. All these are consistent with a regression of an athlete’s heart features. However, findings showing a lack of regression of some athlete's heart characteristics are challenging. The fact that all athletes were already well-trained at 12 years old is of utmost importance. At baseline, LV mass was higher in athletes than in controls,6 and remained similar between those who remained training and those who had quit at follow-up.7 Similarly, at baseline athletes presented with a dilated and dysfunctional RV.6 At follow-up, while there was a “trend” for RV dilation regression in former athletes, the RV function measured with global longitudinal strain (GLS) remained similar to those who remained competitive. Do these results mean that, to some extent, exercise-induced remodeling during childhood persist over time, even after ceasing physical activity? Unfortunately, a sedentary group, free from an endurance training history, is not provided as a comparator. However, data on retired, long-term elite athletes suggest that athlete’s heart features do not completely regress after deconditioning.8 Further, it is becoming increasingly accepted that insults during fetal life have a remarkable impact in adulthood.9 It is therefore plausible that changes occurring during early life (childhood) or at stages of rapid growth (puberty) could persist for long. Consequently, the main issue to address is the long-term impact of these persistent changes – are they adaptative and provide an advantage, or do they flag subclinical damage that is not fully resolved after discontinuing endurance training? On the one hand, it is reasonable to support their benignity – systolic or diastolic function are not impaired, and values remain within the normal range. However, in non-athlete adults, a higher LV mass associates with worse cardiovascular outcomes.10 On the other hand, the RV is an increasingly recognized player and a powerful predictor of cardiac fitness.11 It has been claimed that in athletes, stroke volume is maintained at rest despite reduced GLS, on the grounds of dilated RV; decreased GLS might serve as a contractile reserve to increase performance during exercise.12 Conversely, the characteristics of the RV make it particularly vulnerable to pressure and volume overload which may progress, in few athletes, to RV maladaptative remodeling and the induction of ventricular arrhythmias.13 While these results are challenging, some may be just explained by methodological limitations attributable to its observational nature and sample size. At least two groups may bias conclusions. One third of individuals, on average less fit at baseline, did not attend follow-up scanning.7 Similarly, amongst those attending, one third had discontinued regular endurance training, but reasons were not disclosed: one may argue that insufficient physical capacity and inability to succeed could have discouraged them from remaining competitive. Overall, it is reasonable to speculate that those who were still participating in high-level endurance training are biased towards a group with better innate conditions for training, and differences amongst groups solely reflect those differences. Unfortunately, this is inherent to research in this field. Only randomizing children to high level endurance training or to sedentarism would overcome this issue, but this is obviously an unfeasible approach. The large number of missing individuals finally accounts for the low sample size, comprising only 17 and 30 individuals per group. A larger sample size could move some of the p values in Table 2 beyond the 0.05 threshold (or vice versa). This highlights the frailty of the current trust in the p value research, and substantiates some opposing positions.14 In conclusion, while data from Bjerring and collaborators shed some light into the progressive remodelling occurring in those children who remain competitive during puberty, it raises some additional issues. Whether cardiac changes persist and have any clinical implication in the long-term is particularly relevant and needs to be assessed in larger studies. The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was partially supported by grants from the Instituto de Salud Carlos III (PI16/00703, PI16/00435; PI19/00443); Fundació La Marató de TV3 (20152730); CERCA program/Generalitat de Catalunya.