Implantable cardioverter defibrillators for primary prevention in patients with nonischemic cardiomyopathy: A systematic review and meta-analysis

BackgroundImplantable cardioverter defibrillators (ICDs) have proved their favorable outcomes on survival in selected patients with cardiomyopathy. Although previous meta-analyses have shown benefit for their use in primary prevention, the evidence remains less robust for patients with nonischemic cardiomyopathy (NICM) in comparison to patients with coronary artery disease (CAD). ObjectivesTo evaluate the effect of ICD therapy on reducing all-cause mortality and sudden cardiac death (SCD) in patients with NICM. Data sourcesPubMed (1993-2016), the Cochrane Central Register of Controlled Trials (2000-2016), reference lists of relevant articles, and previous meta-analyses. Search terms included defibrillator, heart failure, cardiomyopathy, randomized controlled trials, and clinical trials. Study selectionEligible trials were randomized controlled trials with at least an arm of ICD, an arm of medical therapy and enrolled some patients with NICM. The primary endpoint in the trials should include all-cause mortality or mortality from SCD. Data extractionHazard ratios (HRs) for all-cause mortality and mortality from SCD were either extracted or calculated along with their standard errors. Data synthesisOf the 1047 abstracts retained by the initial screen, eight randomized controlled trials were identified. Five of these trials reported relevant data regarding patients with NICM and were subsequently included in this meta-analysis. Pooled analysis of HRs suggested a statistically significant reduction in all-cause mortality among a total of 2573 patients randomized to ICD vs medical therapy (HR 0.80; 95% CI, 0.67-0.96; P=.02). Pooled analysis of HRs for mortality from SCD was also statistically significant (n=1677) (HR 0.51; 95% CI, 0.34-0.76; P=.001). ConclusionICD implantation is beneficial in terms of all-cause mortality and mortality from SCD in certain subgroups of patients with NICM.