Accuracy of Myocardial Biomarkers in the Diagnosis of Myocardial Infarction After Revascularization as Assessed by Cardiac Resonance: The Medicine, Angioplasty, Surgery Study V (MASS-V) Trial

Abstract

dial necrosis biomarkers and electrocardiographic abnormalities after revascularization procedures has resulted in a change in the myocardial infarction (MI) definition. Methods. Patients with stable multivessel disease who underwent percutaneous or surgical revascularization were included. Electrocardiograms and concentrations of high-sensitive cardiac troponin I (cTnI) and creatine kinase (CK)-MB were assessed before and after procedures. Cardiac magnetic resonance and late gadolinium enhancement were performed before and after procedures. MI was defined as more than five times the 99th percentile upper reference limit for cTnI and 10 times for CK-MB in percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG), respectively, and new late gadolinium enhancement for cardiac magnetic resonance. Results. Of the 202 patients studied, 69 (34.1%) underwent on-pump CABG, 67 (33.2%) off-pump CABG, and 66 (32.7%) PCI. The receiver operating characteristic curve showed the accuracy of cTnI for on-pump CABG, off-pump CABG, and PCI patients was 21.7%, 28.3%, and 52.4% and for CK-MB was 72.5%, 81.2%, and 90.5%, respectively. The specificity of cTnI was 3.6%, 9.4%, and 42.1% and of CK-MB was 73.2%, 86.8%, and 96.4%, respectively. Sensitivity of cTnI was 100%, 100%, and 100% and of CK-MB was 69.2%, 64.3%, and 44.4%, respectively. The best cutoff of cTnI for on-pump CABG, off-pump CABG, and PCI was 6.5 ng/mL, 4.5 ng/mL, and 4.5 ng/mL (162.5, 112.5, and 112.5 times the 99th percentile upper reference limit) and of CK-MB was 37.5 ng/mL, 22.5 ng/mL, and 11.5 ng/mL (8.5, 5.1, and 2.6 times the 99th percentile upper reference limit), respectively. Conclusions. Compared with cardiac magnetic resonance, CK-MB was more accurate than cTnI for diagnosing MI. These data suggest a higher troponin cutoff for the diagnosis of procedure-related MI.