Predictive value of exercise testing in athletes with ventricular ectopy evaluated by cardiac magnetic resonance

BACKGROUND Exercise-induced ventricular arrhythmias (EIVA) in young athletes raise the suspicion of an underlying heart disease at risk of sudden death. OBJECTIVE We aimed to assess the prevalence and determinants of abnormal cardiac magnetic resonance (CMR) findings in athletes referred for EIVA vs non-EIVA with negative or inconclusive echo-cardiography. METHODS We performed CMR in a consecutive series of athletes aged 15-50 years referred for frequent (>500 per day) or repetitive premature ventricular beats. Clinical and CMR findings were compared between athletes with EIVA and those with non-EIVA, and predictors of abnormal CMR were assessed. RESULTS We included 36 athletes with EIVA (median age 25 years; 27 (75%) males) and 24 with non-EIVA (median age 17 years; 18 (75%) males). CMR revealed cardiac abnormalities in 20 athletes with EIVA (56%) and in 5 with non-EIVA (21%) (P = .004). In particular, left ventricular late gadolinium enhancement was identified in 17 athletes with EIVA (47%) and in 3 with non-EIVA (13%) (P=.006), mostly with a nonischemic pattern. Predictors of abnormal CMR were T-wave inversion on electrocardiography (ECG) (odds ratio [OR] 5.2; 95% confidence interval [CI] 1.0-27.1; P =.05), complex ventricular arrhythmias on 24-hour ambulatory ECG monitoring (OR 4.5; 95% CI 1.1-18.7; P=.04), and complex EIVA with a right bundle branch block or polymorphic morphology on exercise testing (OR 5.3; 95% CI 1.4-19.4; P=.01). CONCLUSION Pathological myocardial substrates on CMR were observed significantly more often in athletes with EIVA than in those with non-EIVA. Repolarization abnormalities on baseline ECG and complex EIVA with a right bundle branch block or polymorphic morphology identified the subgroup of athletes with the highest probability of CMR abnormalities.