Background: Annually, ≈80,000 Americans receive guideline-based primary prevention implantable cardioverter-defibrillators (ICDs), but appropriate firing rates are low. Current selection criteria for ICDs rely on left ventricular ejection fraction, which lacks sensitivity and specificity. Because scar-related myocardial tissue heterogeneity is a substrate for life-threatening arrhythmias, we hypothesized that cardiac magnetic resonance identification of myocardial heterogeneity improves risk stratification through (1) its association with adverse cardiac events independent of clinical factors and biomarker levels and (2) its ability to identify particularly high- and low-risk subgroups.
Methods and results: In 235 patients with chronic ischemic and nonischemic cardiomyopathy with a left ventricular ejection fraction of ≤35% undergoing clinically indicated primary prevention ICD implantation, gadolinium-enhanced cardiac magnetic resonance was prospectively performed to quantify the amount of heterogeneous myocardial tissue (gray zone [GZ]) and dense core scar. Serum high-sensitivity C-reactive protein (hsCRP) and other biomarkers were assayed. The primary end point was appropriate ICD shock for ventricular tachycardia/fibrillation or cardiac death, which occurred in 45 (19%) patients at a 3.6-year median follow-up. On univariable analysis, only diuretics, hsCRP, GZ, and core scar were associated with outcome. After multivariable adjustment, GZ and hsCRP remained independently associated with outcome (P<0.001). Patients in the lowest tertile for both GZ and hsCRP (n=42) were at particularly low risk (0.7% per year event rate), whereas those in the highest tertile for both GZ and hsCRP (n=32) had an event rate of 16.1% per year, P<0.001.
Conclusions: In a cohort of primary prevention ICD candidates, combining a myocardial heterogeneity index with an inflammatory biomarker identified a subgroup with a very low risk for adverse cardiac events, including ventricular arrhythmias. This novel approach warrants further investigation to confirm its value as a clinical risk stratification tool. Clinical Trial Registration- URL: http://www.clinicaltrials.gov. Unique identifier: NCT00181233.