OBJECTIVE Several combinations of risk factors for death or cardiac events after coronary artery bypass grafting have been described. We studied the prognostic value of the preoperative serum levels of cardiac troponin T. METHODS We studied 468 patients who underwent elective coronary artery bypass grafting. Preoperative and postoperative levels of cardiac troponin T and creatine kinase MB, electrocardiograms, clinical data, and events were recorded prospectively. No acute ischemic changes were present on the electrocardiogram before the operations, and preoperative creatine kinase MB serum levels were within normal limits in all patients. RESULTS Ninety-seven (97/468, 21%) patients had serum levels of troponin T greater than 0.02 microg/L within 24 hours before coronary artery bypass grafting. Hospital mortality was similar in this group and in the patients with preoperative levels less than 0.02 microg/L (1% in each group). Nine patients (9/97, 9%) with elevated levels of troponin T before the operation had a perioperative myocardial infarction compared with 12 patients (12/371, 3%) among the group with lower troponin T levels (p = 0.015, RR = 2.9). Congestive heart failure occurred in 10 (10/97, 10%) and 8 (8/371,2%) patients, respectively (p = 0.0009, RR = 4.8). Intensive care unit (p = 0.002) and postoperative hospital length of stay (p = 0.09) were all longer in patients with the elevated preoperative troponin T level. In a logistic regression analysis, troponin T level before the operation was the variable most strongly correlated with postoperative myocardial infarction (p = 0.003). CONCLUSIONS Preoperative troponin T stratification before coronary artery bypass grafting identifies a subgroup of patients with increased risk of postoperative cardiac complications.