ABSTRACT
BACKGROUND: The diagnosis of cardiac complications is particularly challenging in the postoperative course of non-cardiac surgery. Follow-up of patients suggests that silent or symptomatic postoperative myocardial infarction have similar short-term outcomes. Cardiac troponin I (cTnI) has been reported as being a sensitive and specific marker of these complications. METHODS: We conducted a prospective study to determine the cut-off values of cTnI which may predict cardiac complications, i: in the postoperative period until discharge, and ii: during a 1-year period after aortic surgery. Three hundred and twenty-nine consecutive patients undergoing infrarenal aortic surgery were included over a 2-year period in a single center. cTnI was measured at recovery and on the 1st, 2nd and 3rd postoperative days. The presence or absence of cardiac complications was classified by reviewers who had no knowledge of cTnI. For evaluation of the ideal discrimination value of cTnI between the complicated and uncomplicated patient groups, we calculated receiver-operator characteristics for the mean values of the peak of cTnI. RESULTS: Thirteen patients (4%) developed 19 postoperative cardiac complications. Thirteen patients (4%) died in the postoperative period. Nine patients (3%) developed 10 cardiac complications during the 1-year follow-up in 316 patients. In 280 patients, cTnI was below 0.5 ng/ml, in 22 patients between 0.5 and 1.5 ng/ml and the 27 remaining patients had a cTnI higher than 1.5 ng/ml. The area under the curve for postoperative cardiac complications was 0.84 (SD=0.21). A limit value of 0.54 ng/ml yielded a sensitivity of 75% and a specificity of 89%. The area under the curve for late cardiac complications was 0.45 (SD= 0.13). CONCLUSION: A cTnI level greater than 0.54 ng/ml appears to be correlated with the occurrence of cardiac complications in the period until discharge, but no value of cTnI is predictive of late cardiac complications occurring in the 1st year after aortic surgery.
