We compared the prevalence and complexity of exercise-induced arrhythmias in BLSA subjects on chronic diuretic monotherapy of hypertension with that in normotensive control group. Although the prevalence of exercise induced arrhythmias was higher in the diuretic treated group (D) that in controls (C) 57% vs 38%, p less than .05, this difference was due entirely to the high prevalence of simple ventricular ectopic beats (VEB): 44% versus 26%, p less than .05. No difference between the groups was found in the prevalence of frequent or complex ectopic beats. Between 1974 and 1984, 80 BLSA subjects (6.9% of those tested), developed frequent (more than 10% of beats in any minute) ventricular ectopic beats (VEB) or runs of 3 or more VEB on at least one maximal treadmill exercise test. Those developing frequent or repetitive VEB were older than the remaining subjects 63.8 plus/minus 12.5 vs 50.0 plus/minus 16.1 yr, p less than .0001. Within the former group, an ischemic ST segment response to exercise was observed in only 11%. Over a mean follow-up period of 4.6 years without anti-arrhythmic drug therapy, only one cardiac death and one non-fatal myocardial infarction have occurred. To investigate the mechanisms for exercise-induced complex ventricular arrhythmias in apparently healthy subjects, we compared repolarization time, i.e QT intervals, in 20 BLSA subjects with exercise induced nonsustained ventricular tachycardia (VT) with those of 20 normal BLSA individuals matched for age and sex, who were free of exercise-induced arrhythmias. Although no difference in heart rate was present at rest or at maximal effort between groups, those with exercise- induced VT had longer QT intervals then the controls (278 plus/minus 9 vs 259 plus/minus 6 msec, p less than .08). Whereas 12 of 20 VT subjects failed to shorten their QT interval by 25% only 5 of 20 controls demonstrated this finding, p less than .05. (Formerly: Complication of Maximal treadmill Exercise in Apparently Normal Subjects)