To evaluate whether the site of occlusion/stenosis in the left anterior descending coronary artery (LAD) could be diagnosed by noninvasive techniques, thallium-201 myocardial scintigraphy (TMS), 12-lead electrocardiography (EGG), and coronary arteriography were performed in 33 patients with anterior acute myocardial infarction (AMI). The subjects were divided into two groups according to the location of stenosis: ie, either proximal to the first diagonal branch (PRO, n = 18), or beyond the first diagonal branch (NON-PRO, n = 15). The location of the anterior interventricular groove was defined as 0 degrees. The extent of persistent perfusion defect was greater in the PRO group than in the NON-PRO group (0.43 +/- 0.12 vs 0.31 +/- 0.14, p < 0.01). The left margin of the defect in the basal short-axis layer was at 75 +/- 30 degrees in the PRO group and at -19 +/- 43 degrees in the NON-PRO group (p < 0.001). A defect with a left margin at >30 degrees in the basal layer was found in 94% (17/18) of the patients in the PRO group and in 6% (1/15) of the patients in the NON-PRO group (p < 0.001). An abnormal Q wave in leads a VL/I was found in 78% (14/18) of the patients in the PRO group and in 20% (3/15) of the patients in the NON-PRO group (p < 0.001). The sensitivity, specificity and total predictive accuracy of detection of proximal lesions of the LAD were 94%, 93% and 94% by TMS, and 78%, 80% and 79% by EGG, respectively. A significant difference in accuracy existed between TMS and ECG (p < 0.05). These data suggest that it is possible to diagnose the site of occlusion/stenosis of the LAD as either proximal or non-proximal in patients with anterior AMI by TMS and EGG.