Dipyridamole echocardiography test (DET) has gained acceptance due to its safety, feasibility, diagnostic accuracy and prognostic power. The main limitation of the test is a less than ideal sensitivity in some patient subsets, such as those with limited coronary artery disease. Atropine with dipyridamole might theoretically combine to become a synergistic ischaemic stress test, by increasing myocardial oxygen demand through chronotropic stress and by reducing flow supply through a shortening of the diastolic interval under maximal coronary vasodilation. The aim of this study was to assess the effects of the addition of atropine to DET. Three hundred and twenty-one patients (age=58±9 years), referred for testing in the echo lab, were initially studied by DET. Of these, 151 were stopped during or within the 2 min following dipyridamole infusion because of achievement of a predetermined end-point: obvious echocardiographic positivity (n = 137), severe chest pain (n = 3), diagnostic ST segment changes (n = 7) or limited side effects (n = 4). In another three cases, atropine was not given due to a history of glaucoma or severe prostatic hypertrophy. In the remaining 167 patients with a negative DET test, atropine (0.25 mg intravenously, repeated every min up to a maximum of 1 mg, if necessary) was added, starting 3 min after the end of the dipyridamole infusion. The dipyridamole-atropine echo test (DETA) was positive in 32 and negative in 135 patients, and no major side effects occurred in any patient. The peak heart rate was significantly higher during DETA than with DET alone (108±16 vs 86±14 beats . min-1; P<0.0001). In the subset of 178 patients who were studied while not taking antianginal therapy, who had no prior myocardial infarction or revascularization procedure and who underwent coronary catheterization, independently of the test results, coronary angiography showed normal coronary arteries in 48 patients and significant coronary artery disease (CAD) (≥ 50% luminal reduction in at least one major coronary vessel by quantitative coronary arteriography) in 130 patients-with single-, double- and triple-vessel disease in 56, 47 and 27 patients, respectively. The sensitivity was 96/130 for DET and 110/130 for DETA (72 vs 85%, P<0.01) while the specflcity was 96% and 92% (P=ns), respectively. The addition of atropine to dipyridamole, which causes further chronotropic stress to the myocardium already challenged by flow maldistribution, is well tolerated and safe, and increases the sensitivity of the test for the detection of coronary artery disease with no loss in specificity. © 1993 The Europen Society of Cardiology.