The purpose of this study was to identify indices of coronary artery reperfusion in patients treated with thrombolytic therapy for acute myocardial infarction (AMI) by means of characteristics from the serum creatine kinase (CK) isoenzyme MB time-activity curve. Frequent blood sampling as performed in three groups with a first AMI: 29 patients treated with intravenous thrombolytic therapy who had a patent infarct-related artery with normal flow (TIMI-3) at acute catheterization (reperfusion group); four patients with a persistently closed infarct-related artery (no reperfusion group); and 44 patients who did not receive any therapy aimed at coronary reperfusion (no thrombolytic therapy group). In the latter group we prospectively estimated that 25% would have spontaneous reperfusion. A physiologically based computer-calculated multi-compartment method was used to determine the characteristics of the serum CK-MB time-activity curve. In addition to demonstrating an earlier increase, a shorter time to peak of serum CK-MB and a lower estimated infarct size in the reperfusion group (p = 0.025 to 0.00001), the appearance rate constant (k1) and time from estimated initial increase to peak of CK-MB in the blood stream (tRP) were significantly different from those values in the no thrombolytic therapy group (p < 00001). A cutoff level indicating reperfusion if k1 was >0.185 or tRP was <16.5 hours demonstrated overlapping values between these two groups in only four patients (k1), two patients (tRP), and six patients with a combination. This corresponds to a sensitivity of 86% (k1), 93% (tRP), or 100% (combination) and an estimated specificity of 100% (k1), 100% (tRP), or 89% (combination). The k1 index can be reliably estimated as early as 4 hours after the onset of CK-MB release into the blood stream. Thus these indices reflect the presence or absence of coronary reperfusion and may be useful for monitoring thrombolytic therapy in patients with AMI.