To test the sensitivity and specificity of noninvasive tests for detecting myocardial ischemia, 78 patients undergoing cardiac catheterisation for suspected coronary artery disease (CAD) were studied in prospective blinded manner with exercise echocardiography (EE), ecg stress test (ET) and 24-hour Holter monitoring for analysis of ST-segment changes (HM). All noninvasive tests were repeated after one year follow-up period. Exercise was performed in a supine position using a bicycle ergometer. Real time 2-DE examinations were performed in the left lateral position: at rest, at the peak of exercise and immediately after exercise testing. The sensitivity and specificity of EE, ET and HM for detection of CAD were: 76%, 92%; 90%, 48%; and 80%, 71% respectively. In a group of 48 pts with the narrowing of coronary artery (CA) < 50%, the follow-up EE was negative in 45. In a group of 30 pts with significant narrowing of CA, 10 were asymptomatic within one year after PTCA or CABG; in 9 of them the follow-up EE as well as ET were negative. HM less correlated with clinical course in asymptomatic subjects: in 4 patients it was negative and in 6 - positive. The reminding 20 patients with CAD treated medically, with PTCA or CABG were symptomatic after one year. The follow-up EE, ET and HM studies were positive in 17, 19 and 17 cases respectively. We conclude that sensitivity of EE in diagnosis of CAD is comparable to ET and HM, but its specificity is higher.(ABSTRACT TRUNCATED AT 250 WORDS)