Predictive value of perioperative cardiac Troponin I for adverse outcome in coronary artery bypass surgery

Objectives: Cardiac Troponin I (cTnI) is a well-known marker for myocardial damage in patients undergoing aorto-coronary bypass grafting (CABG) peaking 6-8 h after aortic declamping. The aim of this study was to evaluate cTnI release in the course of CABG procedures early, i.e. after the cessation of cardiopulmonary bypass (CPB) in order to recognize unstable cardiac function leading to hemodynamic deterioration and resulting in an adverse outcome (AO). AO is defined as the onset of myocardial infarction and/or death peri/postoperatively. Methods: Five-hundred and forty consecutive patients who underwent CABG were evaluated for cTnI release immediately prior to the induction of anesthesia (END) and after termination of CPB (END). Standard CPB with ante/retrograde cold blood cardioplegia was used. Patients with any of the following criteria were excluded: (1), CABG within 7 days of myocardial infarction; (2), emergency operation for both unstable angina and for coronary occlusion at angioplasty; (3), CABG with concomitant surgical cardiac procedures; (4), preoperative renal dysfunction requiring hemodialysis; (5), redos. Troponin I was measured with the Stratus CS(TM) fluorometric enzyme immunoassay analyzer (Dade-Behring) running on site in the operation room (OR), so values of cTnI could be obtained within 15 rein. Results: There were six deaths (1.1%) in the entire series, Q-wave myocardial infarction occurred in 19 patients (3.5%), AO was experienced by 21 patients (3.9%). The mean preoperative cTnI level was 0.04 +/-0.17 ng/l (mean standard deviation) for the entire group. The END cTnI level for the AO-group was 0.91 +/-0.5 ng/l; for all other patients, this was 0.37 +/-0.3 n/l (P<0.001). Changes in intraoperative cTnI levels relative to time course showed a marked increase for the AO-group (0.0038<plus/minus>0.0035 ng/l*min) as compared with non-AO patients (0.0019 +/-0.0015 ng/l min; P=0.028). The receiver operating characteristic curve indicates a cTnI level at CPB-end of higher than 0.495 ng/l with an area under the curve of 0.83 as the optimal cut-off point for predicting AO with a sensitivity and specificity of 76.2%. Stepwise logistic regression analysis revealed END cTnI level (odds ratio, 17.24; P<0.001), CPB time (odds ratio, 1.03; P=0.001), female sex (odds ratio, 3.8; P=0.011) as significant independent predictors for AO. Age of over 70 years (P=0.8), Cleveland Clinic risk score (P=0.65), diabetes (P=0.26), elevated preoperative creatinine level (P=0.77), severe left ventricular dysfunction (P=0.51), the number of grafts performed (P=0.15), and change of intraoperative cTnI level relative to time course (P=0.94) did not reach statistical significance. Conclusions: cTnI release as determined at the end of CABG procedures represents a strong predictor of an AO after surgery. Analyzing blood samples for cTnI with an automated device on site in the OR provides for immediate results, so specific diagnostic and therapeutic interventions can be performed before hemodynamics deteriorate. (C) 2001 Elsevier Science B.V. All rights reserved.