Thrombolytic therapy for patients with acute myocardial infarction (AMI) has produced the need for an accurate early diagnostic marker. We previously developed and assessed an assay for the creatine kinase (CK)-MB subforms; assay time is 25 minutes. Plasma MB2 (tissue subform) activity, MB1 (plasma-modified subform) activity, and MB2/MB1 ratio in 56 healthy individuals were 0.61 +/- 0.33 units/l, 0.63 +/- 0.33 units/l, and 0.94 +/- 0.39, respectively. Only one individual had both an MB2 activity greater than 1.0 units/l and an MB2/MB1 ratio of more than 1.5. Similar results were obtained in 50 hospitalized patients without cardiac disease; two of these patients had both an MB2 activity and an MB2/MB1 ratio greater than the cutoff values. Among 49 patients with AMI, MB2 activity and the MB2/MB1 ratio began to increase 2 hours after AMI; the ratio reached a plateau of 3.1 by 4-6 hours. The first available plasma sample was abnormal by the subform assay in 67% of patients and by a conventional MB assay in 27% of patients. Assay sensitivities in samples collected at 2-4, 4-6, and 6-8 hours after AMI were 59%, 92%, and 100% for the subform assay and 23%, 50%, and 71% for the conventional assay (p less than 0.03 versus subform assay at each time interval). False-negative results were obtained by the subform and conventional assays in 15 and 45 samples at a mean of 2.3 and 5.8 hours, respectively. Subform assay provides rapid and reliable diagnosis of AMI within 4-6 hours after the onset of symptoms, which is 6 hours before conventional CK-MB assays are accurate.