Which factors predict the behavior of ventricular extrasystoles in athletes over time?

Spontaneous behavior of ventricular extrasystoles (VE) was analysed. From a database containing 578 athletes with VE, 84 males and 11 females (29.9±18.1 years) having ≥100 VE or repetitive VE [ventricular couplets (VC) or ventricular tachycardias (VT)] at first 24-hour Holter electrocardiographic monitoring (24-h-HM) (baseline) and at least 1-year of follow-up (3.1±2.2 years) over the past 10 years were selected. The baseline was compared with the last 24-h-HM to establish DVE (VE reduction of at least 98%/24h in the absence of VC or VT). SDVE was calculated as standard deviation of the number of VE on serial 24-h-HMs. DVE and SDVE were considered as dependent variables. Independent variables were: age, sex, type of sport, symptoms, baseline VE rate (BVE), baseline VC and VT, VE morphology, VE behavior during the baseline training session, disqualification from competitive sports, echocardiographic abnormalities. DVE occurred in 32 athletes (34%). SDVE varied from 0 to 12658 VE/24h (1916±2649.9). Disappearance of VE during the baseline training session (DVET) correlated to DVE (P=0.0319). BVE directly correlated to SDVE (P=0.0008). Athletes' VE are highly variable over time, their variability depending on BVE, and they not infrequently tend to disappear. The only useful variable for predicting DVE is DVET.