93 patients with acute myocardial infarction entered into a multicentre, randomised fibrinolytic therapy study underwent coronary angiography prior to treatment with intracoronary streptokinase or intravenous anisoylated plasminogen streptokinase activator complex (APSAC). Subsequent to administration of fibrinolytic therapy, coronary arteriography of the infarct-related artery was also performed at 15, 30, 45, 60, 75 and 90 minutes. Angiographically defined coronary perfusion was graded as follows: grade 0--no perfusion; grade 1--vessel penetration by contrast without perfusion; grade 2--partial perfusion with delayed flow and/or clearance; grade 3--normal flow and clearance. Two independent readers at separate sites reviewed all serial angiograms with consensus achieved with a third reader. Disagreement potentially affecting therapeutic outcome (grades 1 vs 2, 0 vs 2, 0 vs 3, and 1 vs 3) occurred for only 15 angiographic views. Evaluation of agreement by the K index demonstrated a reasonable level of agreement for all grades (K = 0.73). In order to assess the effect of angiographic classification and timing upon reperfusion percentage rates, 4 reperfusion criteria were applied to these serial angiograms: group A = grade 2 or 3 flow at 90 minutes; group B = grade 2 or 3 flow at any time; group C = grade 1, 2 or 3 flow at any time in patients with control grade 0 or 1 flow; group D = grade 1, 2 or 3 flow at any time in patients with only grade 0 flow at control. Percentage reperfusion varied widely depending upon reperfusion criteria: group A: streptokinase 49%, APSAC 44%; group B: streptokinase 70%, APSAC 50%; group C: streptokinase 86%, APSAC 67%; group D: streptokinase 79%, APSAC 59%.(ABSTRACT TRUNCATED AT 250 WORDS)