The more you train, the more ECG changes you will gain? - A systematic ECG analysis in athletes

Background
Cardiovascular-related sudden cardiac death is the leading cause of mortality in athletes during sport. Intensive training may be associated with cardiac remodeling processes and physiological ECG adaptations. Hence, 12-lead ECG interpretations in athletes require careful analyses to distinguish physiological changes to athletic training from findings suggestive of an underlying pathological condition.

Objective
This observational clinical study aimed to analyze a possible correlation between training intensity and altered ECG pattern. Beyond that, gender differences and further predictive parameters for ECG alterations were evaluated.

Methods
Data from consecutive athletes with complete data sets, that presented at our outpatient clinic between 2007 and 2022, were analyzed. High-performance athletes were compared to competitive athletes with a training intensity of > 6 hours per week and recreational athletes that trained for less than 6 hours per week. Normal 12-lead ECGs were distinguished from ECGs with physiological adaptations to exercise, borderline and abnormal ECG pattern according to the current international consensus standards for ECG interpretation in athletes.

Results
A total of 465 consecutive athletes (mean age 27.5 ± 7.7 years, 94% male) were included. Normal ECGs were documented in 34 athletes (7%). The majority (n=358, 77%) showed physiological changes related to athletic training. In 32 athletes (7%) borderline ECG-findings and in 41 athletes (9%) abnormal ECG-pattern were documented, which required further evaluation. Cramer’s V tests identified a significant correlation between physiological ECG adaptations and the athletes’ training intensity (Figure 1A). In particular “sinus bradycardia” (n=181, V-coefficient=0.320, p<0.001), “first degree atrioventricular (AV) blocks” (n=15, V=0.131, p=0.043) and QRS voltage criteria for “left and right ventricular hypertrophy” (n=46, V=0.224, p<0.001) show a mild to moderate correlation with the athletes’ training intensity (Figure 1B). With regard to the specific borderline variant “left atrial enlargement” (n=12, V=0.633, p=0.004) a  strong correlation with the athletes’ training intensity was revealed. Between exercise intensity and abnormal ECG pattern no significant correlation was observed (V=0.012, p=0.835). Concerning gender differences, exercise associated ECG variants were more frequently documented in male athletes (V=0.128, p=0.006).

Conclusion
A significant correlation between the athletes’ exercise intensity and physiological ECG adaptations was revealed. In addition, the specific borderline ECG criterion “left atrial enlargement” seems to correlate strongly with the athletes’ training intensity. Female athletes show less pronounced sport-associated ECG changes compared to male athletes. Further studies are warranted to confirm our initial observations.

Figure 1 Correlation between (A) normal, physiological, borderline and abnormal ECG pattern and (B) training related ECG variants and the athletes’ exercise intensity. A p-value < 0.05 indicates statistical significance and a Cramer’s V coefficient (V) > 0.1 indicates a mild,  V > 0.2 a moderate and V > 0.5 a strong correlation.