Comparative Analysis of Perioperative Myocardial Infarction Criteria in Predicting Outcomes After Coronary Artery Bypass Surgery

Abstract

Background There have been various definitions for perioperative myocardial infarction (pMI) following coronary artery bypass grafting, but it is unclear which definition is most relevant. The purpose of this study is to (i) investigate the association between the various pMI definitions and postoperative mortality and (ii) identify the definition that has the greatest incremental value in predicting postoperative mortality.

Methods The study examined a cohort of 3,211 patients undergoing elective coronary artery bypass grafting at a tertiary hospital in Korea. The prognostic value of perioperative myocardial infarction was assessed according to six different pMI definitions: (i) the 4th Universal Definition of Myocardial Infarction (4UD), (ii) the Academic Research Consortium (ARC), (iii) the Society for Cardiovascular Angiography and Interventions (SCAI), (iv) the Valve Academic Research Consortium (VARC), (v) the European Association of Cardio-Thoracic Surgery (EACTS), and (vi) the myocardial injury criteria proposed by the Vascular Events in Surgery Patients Cohort Evaluation (VISION) Cardiac Surgery study. The co-primary endpoints of the study were operative mortality (all-cause mortality within 30 days of surgery or before discharge) and postoperative 1-year all-cause mortality. The incremental prognostic value of each definition was evaluated using the area under the receiver operating characteristic curve (AUC), likelihood ratio tests, net reclassification improvement (NRI), and integrated discrimination improvement (IDI).

Results The operative mortality rate was 1.3% (42 of 3,211), and the 1-year mortality rate was 3.4% (109 of 3,211). The VISION myocardial injury criteria showed the strongest independent association with operative mortality [odds ratio (OR) 5.27; 95% confidence interval (CI) 2.84 to 9.79; 4UD: OR 3.49; 95% CI 1.47 to 5.74; ARC: OR 3.49; 95% CI 1.70 to 7.20; SCAI: OR 2.54; 95% CI 1.38 to 4.67; VARC: OR 2.43; 95% CI 1.32 to 4.49; EACTS: OR 3.29; 95% CI 1.75 to 6.19]. The VISION criteria had the highest incremental value in predicting operative mortality (AUC 0.78; Chi-square statistic 24.19; NRI 0.325; IDI 0.0116). On the other hand, 4UD showed the strongest association with 1-year mortality [hazard ratio (HR) 3.29; 95% confidence interval (CI) 2.18 to 4.97; ARC: HR 2.92; 95% CI 1.83 to 4.67; SCAI: HR 2.16; 95% CI 1.48 to 3.15; VARC: HR 2.32; 95% CI 1.60 to 3.38; EACTS: HR 3.09; 95% CI 2.08 to 4.58; VISION: HR 2.25; 95% CI 1.47 to 3.44]. The 4UD and EACTS criteria demonstrated the most significant incremental value for predicting 1-year mortality (4UD: AUC 0.77, Chi-square statistic 26.91, NRI 0.324, IDI 0.0063; EACTS: AUC 0.76, Chi-square statistic 27.37, NRI 0.349, IDI 0.0055).

Conclusions Among various definitions for pMI following coronary artery bypass grafting, the VISION myocardial injury criteria provide the greatest incremental value in predicting operative mortality, whereas the 4UD and EACTS definitions of Myocardial Infarction are the most effective predictors of 1-year all-cause mortality.