Long-term cardiovascular function and cardiopulmonary performance in athletes after COVID-19: results from the COSMO study.

INTRODUCTION

Most studies in athletes have focused on short-term outcomes after SARS-CoV-2 infection, whereas long-term data on cardiac function and exercise performance remain limited. This prospective study evaluated 12-month changes in cardiac structure, myocardial deformation, and cardiopulmonary performance in elite and recreational athletes after mild COVID-19.

METHODS

Fifty-two athletes (median age 32.5 (24.8-46.3) years; 27 females) from endurance, power, and mixed sports underwent echocardiography, cardiopulmonary exercise testing (CPET), and bioimpedance analysis after infection (T0; median 88 days post-infection) and again at 12-month follow-up (T1). Measurements included left and right ventricular global longitudinal strain (LV/RV GLS), RV free wall strain (FWS), left atrial strain, CPET performance and ventilatory parameters, and anthropometric variables. Longitudinal changes were evaluated using paired comparisons and linear mixed-effects models. Univariable and multivariable regressions were performed to identify predictors of follow-up VO2peak and its longitudinal change.

RESULTS

Cardiac structure, LV and RV systolic function, and myocardial deformation remained stable over 12 months, with no evidence of adverse remodeling. A significant reduction in E/E' medial was observed at follow-up (p = 0.008), while all other diastolic indices did not change significantly. Cardiopulmonary performance was preserved, with no significant changes in maximal workload, absolute or relative VO2peak, ventilatory efficiency, or oxygen pulse (all p > 0.05). CPET differences were limited to a higher peak RER (p = 0.045) and a lower resting heart rate at T1 (p = 0.010). Body composition changed modestly, with higher BMI (p = 0.035), while pulmonary function parameters remained stable. In multivariable linear regression, baseline VO₂peak was the only independent predictor of follow-up VO₂peak (β = 0.92, 95% CI 0.68-1.15; p < 0.001; R² = 0.64), whereas sex, sport type, and baseline LV GLS were not associated with performance at T1.

CONCLUSION

Athletes recovering from mild SARS-CoV-2 infection show no evidence of impaired myocardial structure, strain, or cardiopulmonary performance at 12 months post-infection. Minor variations in diastolic indices and body composition appeared physiological, and baseline aerobic fitness was the strongest predictor of follow-up performance. These findings indicate a stable 1-year cardiopulmonary outcome and support the safety of return to sport in asymptomatic athletes.