Our aim was to verify whether the sensitivity of pharmachological stress echocardiography for multivessel disease after acute myocardial infarction may be improved by a more aggressive protocol, i.e. not considering the appearance of the first wall motion abnormality as the absolute end-point if it occurs in the infarcted area without clinical or instrumental markers of extensive ischemia or left ventricular dysfunction. One-hundred twenty-one consecutive patients (age 32-71 years) prospectively underwent dobutamine-atropine stress echo (dobutamine infusion up to 40 micrograms/kg/min with additional atropine 1 mg) 11.8 +/- 4.8 days after uncomplicated myocardial infarction and coronary angiography within 6 weeks. Criteria for stopping the test were: significant ST depression or elevation, typical chest pain, major arrhythmias and left ventricular dysfunction. The test was considered as positive if a deterioration of basal wall motion pattern was observed: it was defined homozonally positive (the deterioration occurred in the myocardial area fed by the culprit vessel) or heterozonally positive (the deterioration occurred in a different vascular area). A coronary stenosis > 70% of vessel lumen was defined as critical. Thirty-four patients showed a negative test result. Among the 87 patients with positive test, 65 had no further wall motion deterioration from the first-induced wall motion abnormality (WMA) to peak test (Group A), whereas nine patients showed further homozonal (Group B) and 13 further heterozonal (Group C) asynergies. Sensitivity, specificity and accuracy of dobutamine stress echocardiography for multivessel disease were, respectively, 63%, 96% and 82% using the first-induced wall motion abnormality as test end-point, whilst they were 84% (P < 0.01), 93% and 89% according to the aggressive approach previously described. Dobutamine stress time of patients with multivessel disease was higher in Groups B and C (13.1 +/- 3.6 min) than in Group A (9.8 +/- 3.7 min, P < 0.01) and, finally, the mean obstruction of non-culprit vessel was higher in Group A (62.2%) than in Group C (47.4%, P < 0.05). No major complications were found. We conclude that the sensitivity of dobutamine stress echocardiography for multivessel disease following recent myocardial infarction is critically dependent on the test end-point. It may be improved by a more aggressive approach capable to identify less severe heterozonal coronary lesions.