Background The goal of this prospective study was to assess whether cardiac troponin I (cTnI) could replace creatine kinase (CK)-MB mass as the serum biochemical marker for detection of acute myocardial infarction (AMI). Methods And Results Over a 3-month period, 327 nonselected, consecutive patients were evaluated for AMI with the use of modified World Health Organization criteria including serial electrocardiographs and CK-MB mass determinations at admission and 6, 12, and 24 hours after admission. cTnI measurements were also made at all time points. Sixty-two (19%) patients were diagnosed with AMI. Diagnostic sensitivity and specificity for peak concentrations were equivalent or better for cTnI (100%; 96.3%) compared with CK-MB (88.2%; 93.2%) and total CK (73.5%; 84.6%), respectively. cTnI demonstrated 100% negative predictive accuracy for ruling out AMI. Further, cTnI maintained a high diagnostic sensitivity (>94%) up to 96 hours after onset of chest pain compared with CK-MB and total CK (both 50% sensitive) in patients with AMI. However, patients with documented Q-wave infarctions had a significantly longer clearance compared with non-Q-wave infarctions (†½ 24.2 vs 7.3 hours, respectively; P <.01). There was a significant (P <.02) positive correlation (r = 0.89) between increasing CK-MB mass and increasing cTnI for AMI specimens. Conclusions These findings have strongly supported our clinical implementation of cTnI, replacing CK-MB mass as the preferred marker for detection of AMI. (Am Heart J 1999;137:332-7.)