Systolic time intervals (STI) and the apexcardiographic A wave were measured to evaluate their predictive value with respect to angina pectoris (AP) in a group of 1231 men aged from 17 to 64 years. At entry 55 men had typical AP and during a 4-year follow-up 87 men developed AP. The mean STI values were significantly lengthened and the A wave height increased in subjects with manifest old AP compared with the other subjects. Analysis of the incidence of new AP by quintiles of STI showed that the AP incidence was related to the duration of electromechanical systole (QS2) and left ventricular ejection time (LVET). The lowest AP incidence was found in the second quintile of QS2 and increased gradually towards the highest QS2 quintile. The highest AP incidence was, however, found in subjects with the shortest QS2. The AP incidence analyzed by the quintiles of LVET showed a gradual decrease from the lowest to the fourth quintile, but the highest AP incidence was observed in the highest LVET quintile. These relationship between the STI and AP incidence persisted after age-adjustment. The A wave height was significantly increased in subjects who developed new AP. The quintile approach, however, failed to show an increase of age-adjusted AP incidence in the highest A wave quintile. Our data thus show that typical angina is associated with abnormal STI measurements and that in asymptomatic men a lengthened as well as shortened QS2 and LVET and a high apexcardiographic A wave are associated with an increased risk of angina pectoris.