Abstract

OBJECTIVES: We sought to evaluate whether infarct size characterization by cardiac magnetic resonance imaging (MRI) is a better predictor of inducible ventricular tachycardia (VT) than left ventricular ejection fraction (LVEF). BACKGROUND: Inducibility of VT at electrophysiologic study (EPS) and low LVEF can identify patients with a substrate for VT. Magnetic resonance imaging has been shown to identify, with high precision, areas of myocardial infarction and may therefore be a better tool to evaluate for a substrate for VT. METHODS: We studied 48 patients with known coronary artery disease who were referred for EPS using cine and gadolinium-enhanced MRI. Wall motion and infarct characteristics were determined blindly and compared among patients with no inducible ventricular arrhythmias (n = 21), those with inducible monomorphic VT (MVT, n = 18), and those with either inducible polymorphic VT or ventricular fibrillation (n = 9). RESULTS: Patients with MVT had larger infarcts than patients who did not have inducible arrhythmias (mass: 49 +/- 5 g [SE] vs. 28 +/- 5 g, p < 0.005; surface area: 172 +/- 15 cm(2) vs. 93 +/- 14 cm(2), p < 0.0005). Patients with polymorphic VT/fibrillation had intermediate values (mass: 36 +/- 7 g; surface area: 115 +/- 22 cm(2)). Ejection fraction was inversely related to infarct mass and surface area, with R(2) values ranging from 0.21 to 0.27. Logistic regression and receiver-operating characteristic analysis demonstrated that infarct mass and surface area were better predictors of inducibility of MVT than LVEF. CONCLUSIONS: Infarct surface area and mass, as measured by cardiac MRI, are better identifiers of patients who have a substrate for MVT than LVEF. Further evaluation of infarct size characterization by cardiac MRI as a predictor of sudden cardiac death is warranted.