Sudden cardiac death risk stratification in dilated cardiomyopathy: climbing the pyramid of knowledge.

Abstract

Nonischemic dilated cardiomyopathy (NIDCM) is the second leading cause of sudden cardiac death (SCD) in the United States. As with coronary artery disease, large-scale clinical trials evaluating implantable cardioverter defibrillator (ICD) therapy based on noninvasive risk markers have focused largely on left ventricular ejection fraction (LVEF).1,2 These studies demonstrate that ICD use in selected patients with NIDCM and depressed LVEF is associated with improved survival.3 The broader question is whether ICD use based on LVEF (and New York Heart Association class) as currently listed in the guidelines represents optimal deployment of this valuable but costly resource. As with ischemic cardiomyopathy, this question is easily answered by the observations that many SCD cases occur in patients whose LVEF is not within the guideline criteria for ICD implantation and that the majority of patients whose LVEF does fall within the guideline criteria do not experience appropriate shocks from their ICD. Although ICD shocks may overestimate the benefit of ICDs to reduce SCD, ICD shocks were reported in only 33 of 229 patients in Defibrillators in Non-Ischemic Cardiomyopathy Treatment Evaluation (DEFINITE).4 This underscores the need for improved risk stratification to better deploy ICD use for SCD prevention. In addition to providing improved clinical outcomes, better ICD deployment via improved risk stratification could dramatically enhance the cost-effectiveness of this treatment. However, the question of how to achieve improved risk stratification endures. Article see p 1101 Pezawas et al5 report on the predictive value of LVEF, heart rate variability, baroreflex sensitivity, signal averaged ECG, and microvolt T wave alternans in patients with NIDCM. Although this is a fairly small study, important strengths are the inclusion of patients with a broad array of LVEFs, repeat testing at 3 years, and the long duration of follow-up (median, 7 years). Sixty percent …