Abstract 11027: Short-and Long-Term Prognostic Value of Multiple Cardiac Biomarker Measurement in Patients Undergoing Elective Coronary Artery Bypass Graft Surgery

Abstract

Introduction: Coronary artery bypass grafting (CABG) is associated with defined degree of myocardial injury reflected by perioperative rise of cardiac biomarkers. The present study aimed to evaluate multiple cardiac biomarkers and their association with short- and long-term outcome following CABG. Methods: In this prospective study, a total of 1030 patients with coronary artery disease (CAD) were enrolled between 07/2008 and 11/2018, undergoing elective, isolated, on-pump CABG. Perioperative cardiac troponin I and T (cTnI, cTnT) or ultra-sensitive (us-) cTnI were measured over 72hrs after surgery. Mean Follow-up was 5.5±2.7years with 100% completeness. Primary study endpoint was in-hospital mortality (IHM) and major adverese cardiac and cerebrovascular event (MACCE) rate at 30 days, as well as mortality in the long-term, defined as death from any cause. Results: Patients were 67.6±9.4years of age, 81.8% male, presenting 3-vessel CAD in 82% and/or left-main disease in 38%. Log EuroSCORE-I was 4.3±3.7%. Perioperative serum concentrations of cTnI ( P =0.05), us-cTnI ( P =0.02), but not cTnT ( P =0.79) differed significantly between survivors and non-survivors and cTnI ( P <0.001) and us-cTnI ( P <0.001) between patients with and without MACCE within the first 30 days after surgery (2-way ANOVA). Kaplan-Meier analysis for time to event variables, clearly showed poor long-term prognosis in patients with high perioperative cTnI, having a 1-, 5- and 10-years survival of 93.5%, 81.3% and 54.5%, as compared to patients with medium or low cTnI ( P =0.03). Multivariable logistic regression analyses revealed cTnI, but not cTnT and us-cTnI, as a strong and independent predictor for IHM ( P <0.001) and MACCE ( P =0.02). In a receiver operating curve (ROC) analysis, cTnI of 9.2ng/mL with an area under curve (AUC) of 0.72 ( P =0.02) was identified as an optimal cut-off value for predicting IHM and a cut-off value of 6.38ng/mL with an AUC of 0.85 ( P <0.001) for the prediction of MACCE. Conclusion: Serial perioperative cTnI measurment and/or its ultra-sensitive isoform allows for identifying patients with worse clinical short- and long-term outcomes following CABG. CTnT however, failed to identify patients at risk and was inferior in predicting short- and long-term outcomes.