Abstract 9769: Measurement of Myocardial Necrosis Biomarkers After Surgical or Percutaneous Intervention Assessed by Cardiac Magnetic Resonance for Diagnosis of Procedure-Related Myocardial Infarction. The MASS-5 Trial
Introduction: The correlation between the elevation of cardiac biomarkers and diagnosis of procedure-related myocardial infarction (MI) continues to be debated, especially since the emergence of high-sensitive troponin assays (cTnI). However, late gadolinium enhancement (LGE) in cardiac magnetic resonance (CMR) has been established as an alternative for diagnosing AMI.
Methods: We prospectively studied 219 patients with stable multivessel coronary artery disease and preserved ventricular function formally indicated for percutaneous or surgical revascularization. Measurements of biomarkers like high-sensitive troponin I (cTnI) and creatine kinase (CKMB) were systematically obtained before and after the procedure, 6, 12, 24, and 36 hours after PCI and up to 72 hours after CABG. Furthermore, CMR was performed in all patients before and after the procedures. The diagnosis of MI was defined as up to 5 times and 10 times the 99th percentile, for PCI and CABG, respectively.
Results: Of 219 patients included in the study, 202 completed it; 69 (34.1%) underwent on-pump CABG (ONCAB), 67 (33.9%) underwent off-pump CAGB (OPCAB), and 66 (32.0%) underwent PCI. Using the Third Universal Definition of Myocardial Infarction and based on the receiver operating characteristic (ROC) curve, for ONCAB patients, the best cutoff of TnI was 6.5ng/mL (162 times the 99th percentile) with a sensitivity of 100% and specificity of 3.6%, and 37.5ng/mL (8.5 times the 99th percentile) for CKMB with a sensitivity of 69% and specificity of 73%. For OPCAB patients, the best cutoff of TnI was 4.5ng/mL (112 times the 99th percentile) with a sensitivity of 100% and specificity of 9%, and 24ng/mL (5.4 times the 99th percentile) for CKMB with a sensitivity of 61% and specificity of 84%. For PCI patients, the best cutoff of TnI was 7.5ng/mL (187 times the 99th percentile) with a sensitivity of 100% and specificity of 45.6%, and 19ng/mL (4.3 times the 99th percentile) for CKMB with a sensitivity of 57% and specificity of 98%.
Conclusion: In this sample, compared with LGE in CMR, CKMB was more accurate than cTnl for the diagnosis of myocardial infarction in the 3 groups studied. These data suggest a higher troponin cutoff for the diagnosis of procedure-related MI.
Author Disclosures: W. Hueb: None. P.C. Rezende: None. L.M. Costa: None. R.M. Vieira de Melo: None. F.T. Oikawa: None. C.H. Nomura: None. A.V. Villa: None. C.M. Strunz: None. A.C. Hueb: None. E.E. Silva: None. J.A. Ramires: None. R. Kalil Filho: None.
- © 2014 by American Heart Association, Inc.