Ventricular Arrhythmias and Sudden Death in Nonischemic Dilated Cardiomyopathy: Matter of Sex or Scar?

BACKGROUND:To evaluate the association between sex and ventricular arrhythmias (VA) or sudden death (SD) in nonischemic dilated cardiomyopathy, including analysis of potential confounders. METHODS AND RESULTS:Retrospective cohort study of consecutive patients with DCM referred for cardiac magnetic resonance at 2 tertiary hospitals. The primary combined end point encompassed sustained VA, appropriate implantable cardioverter defibrillator therapies, resuscitated cardiac arrest, and SD. We included 1165 patients with median follow-up of 36 months (interquartile range 20-58 months). The majority of patients (66%) were males. Males and females had similar left ventricular ejection fraction, but the prevalence of late gadolinium enhancement (LGE) at cardiac magnetic resonance was significantly higher among males (48% vs 30%, P < .001). Males had higher cumulative incidence of the primary end point (8% vs 4%, P = .02), and male sex was a significant predictor of the primary end point at univariate analysis (hazard ratio 1.93, P = .02). However, LGE had a major confounding effect in the association between sex and the primary outcome: the hazard ratio of male sex adjusted for LGE was 1.29 (P = .37). LGE+ females had significantly higher cumulative incidence of the primary end point than LGE- males (13% vs 1.8%, P < .001). CONCLUSIONS:In patients with DCM, the prevalence of LGE is significantly higher among males, implying a major confounding effect in the association between male sex and VA or SD. LGE+ females have significantly higher risk than LGE- males. These data do not support the inclusion of sex into risk stratification algorithms for VA or SD in DCM.