Immediate percutaneous transluminal coronary angioplasty has been advocated for patients with a residual stenosis after coronary thrombolysis because of the possibility that the residual stenosis may restrict reflow and thereby increase infarct size. Because there are few experimental data bearing on this issue, we measured left ventricular function, myocardial blood flow, and infarct size in 20 anesthetized open-chest dogs during 2 hr of left circumflex occlusion and 4 hr of reperfusion. Ten animals were reperfused through a critical stenosis of the left circumflex artery (critical stenosis group) and the remaining 10 animals underwent full reperfusion without stenosis (control group). In both groups, a comparable degree of echocardiographic systolic wall thinning was present during occlusion and partial recovery of global and regional left ventricular function in the two groups. Subendocardial blood flow was decreased in the critical stenosis group relative to the control group at 5 min after reperfusion (0.52 +/- 0.16 ml/min/g in the critical stenosis group vs 1.55 +/- 0.32 ml/min/g in the control group, p less than .05) but not at 4 hr after reperfusion, when a reduced reflow response was seen in both groups. No differences in subepicardial blood flow were seen in the two groups of animals. Infarct size was slightly greater in the critical stenosis group than the control group, but this difference was not statistically significant (infarct/risk area ratio: 55.5 +/- 7.8% in the critical stenosis group vs 39.4 +/- 9.7% in the control group, p = .21). A close inverse exponential relationship was seen between infarct size/risk area ratio and subendocardial blood flow during occlusion (r = .89, p = .001). Two control animals had high levels of subendocardial collateral flow (greater than 0.2 ml/min/g); when these animals were excluded from analysis, differences in the infarct size/risk area ratio in the control and critical stenosis groups were less striking: (55.5 +/- 7.8% in the critical stenosis group vs 48.4 +/- 9.6% in the control group). Thus, the presence of a critical stenosis results in restriction of hyperemic blood flow to the subendocardium after reperfusion but does not influence infarct size or early left ventricular functional recovery.