Association Of Myocardial Biomarkers Levels And Prognosis Following Coronary Artery Bypass Graft Surgery

Background:Periprocedural myocardial infarction(PMI)is one of common complications after coronary artery bypass graft(CABG).The controversy about CABG-related PMI has focused on the type and threshold of myocardial biomarkers.To date,the threshold with prognostic significance of high-sensitivity cardiac troponin(hs-cTn)remains unclear,meanwhile compared to Creatine kinase-myocardial band(CK-MB),its advantages in prognostic significance and diagnosing PMI have not been demonstrated.Objective:To compare the prognostic significance of post-CABG CK-MB and hs-cTnI based on existing thresholds and to assess the impact of CK-MB and hs-cTnI on the PMI according to Society for Cardiovascular Angiography and Intervention(SCAI).Methods:Patients undergoing isolated elective CABG between March 2017 and December 2018 in Fuwai hospital were retrospectively included.Periprocedural CK-MB and hs-cTnI levels were collected along with other baseline characteristics.Levels of post-CABG CK-MB(<1×upper limit of normal(ULN),1-2×ULN,2-5×ULN,5-10×ULN,≥10×ULN)and hs-cTnI(<10×ULN,10-35×ULN,35-70×ULN,70-200×ULN,≥200×ULN)were turned into categorical variables by prescribed thresholds.CK-MB and hs-cTnI were respectively applied to diagnose SCAI-PMI.The primary end-points were 30-day death and 1-year all-cause death.Logistic regression and Cox regression were utilized to analyze the association of postoperative myocardial biomarkers levels and SCAI-PMI with end-points,and Kaplan-Meier method was utilized to plot the survival curves.Results:This study included 3739 patients with available postoperative CK-MB and hs-cTnI levels after isolated elective CABG,with a mean age of 61.5 years old,and the proportion of male patients was 78.0%.Compared with patients who had postoperative CK-MB<1×ULN,CK-MB≥10×ULN independently predicted 30-day death(adjusted odds ratio(OR):25.88;95%confidence interval(CI):6.31-106.13;P<0.001)and 1-year all-cause death(adjusted hazard ratio(HR):3.79;95%CI:1.35-10.59;P=0.011).Compared with patients who had postoperative hs-cTnI<10×ULN,hs-cTnI≥200×ULN independently predicted 30-day death(adjusted OR:8.09;95%CI:1.48-44.24,P=0.016),but not 1-year all-cause death(adjusted HR:1.15;95%CI:0.39-3.36;P=0.801).Using hs-cTnI to diagnose SCAI-PMI led to a much higher incidence than using CK-MB(20.4%vs 2.2%).SCAI-PMI based on CK-MB independently predicted 30-day death(adjusted OR:18.83;95%CI:5.36-66.20;P<0.001)and 1-year all-cause death(adjusted HR:4.04;95%CI:1.48-11.03;P=0.006).Instead,SCAI-PMI based on hs-cTnI only independently predicted 30-day death(adjusted for OR:4.73;95%CI:1.59-14.12;P=0.005),not 1-year all-cause deaths(adjusted HR:1.62;95%CI:0.79-3.31;P=0.185).Conclusion:In this single-center large-scale CABG cohort,postoperative CK-MB≥10×ULN could independently predict short-and medium-term prognosis,while hs-cTnI≥200×ULN still could not predict medium-term prognosis,indicating CK-MB≥10×ULN is more prognostically relevant than hs-cTnI≥200×ULN to identify high-risk patients.The SCAI-PMI based on CK-MB was associated with higher risk of death,while the SCAI-PMI based on hs-cTnI led to higher proportion of PMI but not relevant to medium-term prognosis.Therefore hs-cTnI should not be a substitute biomarker to diagnose SCAI-PMI.