Two-dimensional echocardiography with intravenous injection of dipyridamole (0.56 mg/kg) was performed in 33 consecutive patients with acute (2 +/- 2 days) postero-inferior myocardial infarction for semiquantitative segmental wall motion analysis. The results were compared with those of coronary angiography which was carried out during the hospital period. After a second evaluation of the recordings the following results were obtained: feasibility: 94 per cent with 85 per cent of segments analysed. Residual ischaemia in the first days of myocardial necrosis was common (70%). The ischaemia was often clinically silent including during the investigation (61%). When pain occurred, it always followed changes in regional wall motion. The dipyridamole test suggested multivessel disease with a sensitivity of 72 per cent and a specificity of 90 per cent, and residual arterial stenosis with a sensitivity of 75 per cent ans specificity of 80 per cent the secondary effects were minor. The main limitation of the test is related to the distinction between pharmacological and physiologic ischaemia. A positive test was associated with lesions justifying myocardial revascularisation (coronary bypass or angioplasty) in 19 out of 23 cases but with a very poor correlation with the topography of the coronary lesions. A negative test indicated arterial occlusion, residual stenosis with extensive myocardial damage or a normal coronary angiogram. Therefore, the dipyridamole echocardiography test may help identify a group of patients with little or no myocardial ischaemia in whom invasive investigations could be deferred; these patients contrast with the group with a positive test indicating residual ischaemia in which the coronary lesions should be documented by coronary angiography.