The aim of this study was to evaluate the relation between the infarction artery status and left ventricular volumes, independently of regional ventricular dysfunction, at 4-6 weeks after a first myocardial infarction. The study group consisted of 100 patients, of whom 80 received thrombolytic treatment. Coronary and contrast left ventricular angiograms were performed at 36+/-5 days postinfarction. Left ventricular end-diastolic and end-systolic volumes were measured. The centerline chord motion method was used to calculate the extent of wall motion abnormality (percentage of chords with hypokinetic motion) and its severity (maximum units of S.D. below the normal wall motion reference). Minimum lumen diameter, patency and collateral flow in the infarction artery were also analyzed. Eight patients (group I) showed occlusion with poor collateral flow in the infarction artery, 22 patients (group II) occlusion with good collateral flow, 38 patients (group III) severe residual stenosis (minimum lumen diameter < or = 1 mm), and 32 patients (group IV) non-severe residual stenosis (minimum lumen diameter > 1 mm). Patients from group I presented greater wall motion abnormality in terms of both extent (P=0.005) and severity (P=0.007), and greater end-diastolic (P=0.07) and end-systolic (P=0.0008) volumes; there were no differences among groups II, III and IV. By stepwise multivariate regression analysis, the extent of wall motion abnormality was the main determinant of end-diastolic (P=0.0001) and end-systolic (P=0.0001) volumes; occlusion with poor collateral flow was also a significant independent factor for end-systolic volume (P=0.03). Total occlusion (including both with and without collaterals) and the minimum lumen diameter did not correlate with end-diastolic and end-systolic volumes. We concluded that (A) the extent of regional dysfunction is the primary determinant of left ventricular volumes at 4-6 weeks postinfarction. (B) The status of the infarction artery is a weak predictor of end-diastolic volume, which is the best descriptor of ventricular remodeling, although occlusion with poor collateral flow is associated to larger end-systolic volume.