To assess the predictive value of coronary events reflected by changes in R-wave amplitude after exercise, 146 patients with angiographically documented coronary heart disease were studied. All patients were followed up for 6 years, during which time myocardial infarction and death of cardiovascular origin were considered endpoints. The incidence of events in patients in whom R-wave amplitude decreased (normal response) and in those in whom R-wave amplitude did not change or increase (abnormal response) were compared. The incidence of coronary events in patients with a normal response was 23% and in those with an abnormal response, 45.8% (p less than 0.01). Correlating the results with several noninvasive and angiographic variables, an abnormal R-wave response showed a significantly higher rate of events in the subsets of patients with prior myocardial infarction, absence of cardiomegaly, maximal functional capacity lower than 4 METs, maximal heart rate higher than 140 beats/min and abnormal left ventricular function. Thus, the changes in R-wave amplitude after exercise is a variable that should be taken into account when assessing the risk of future events in patients with coronary heart disease.