Perioperative myocardial infarction is the most common cause of morbidity and mortality in patients who have had noncardiac surgery, but its diagnosis can be difficult. The present study was designed to determine whether the measurement of serum levels of cardiac troponin I, a highly sensitive and specific marker for cardiac injury, would help establish the diagnosis of myocardial infarction. We obtained preoperative measurements of MB creatine kinase, total creatine kinase, and cardiac troponin I, in addition to base-line electrocardiograms and two-dimensional echocardiograms, in 96 patients undergoing vascular surgery and 12 undergoing spinal surgery. Blood samples were obtained every 6 hours for at least the first 36 hours after surgery, and electrocardiograms were obtained daily; a second echocardiogram was obtained approximately three days after surgery. The appearance of a new abnormality in segmental-wall motion on the postoperative echocardiogram (that is, an abnormality that had not been seen on the preoperative echocardiogram) was considered to be indicative of perioperative infarction. Eight patients who underwent vascular surgery had new abnormalities in segmental-wall motion and received a diagnosis of perioperative infarction. All eight had elevations of cardiac troponin I, and six had elevations of MB creatine kinase. Of the 100 patients without perioperative infarction detected by echocardiography, 19 had elevations of MB creatine kinase, and 1 had a slight elevation of cardiac troponin I. The measurement of cardiac troponin I is a sensitive and specific method for the diagnosis of perioperative myocardial infarction. It avoids the high incidence of false diagnoses associated with the use of MB creatine kinase as a diagnostic marker.