December 15, 2004

Implantable Defibrillators for the Prevention of Mortality in Patients With Nonischemic CardiomyopathyA Meta-analysis of Randomized Controlled Trials

Author Affiliations

Author Affiliations: Division of Cardiovascular Medicine, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, Mass.

JAMA. 2004;292(23):2874-2879. doi:10.1001/jama.292.23.2874

Context Implantable cardioverter defibrillator (ICD) therapy is effective in primary and secondary prevention of sudden cardiac death among patients with prior myocardial infarction and depressed ejection fraction. However, conclusive evidence of survival benefit in patients with nonischemic cardiomyopathy (NICM) is still lacking.

Objective To determine whether ICD therapy reduces all-cause mortality in patients with NICM.

Data Sources MEDLINE (1966-2004), EMBASE (1991-2004), the Cochrane Central Register of Controlled Trials (through first quarter, 2004), reports presented at scientific meetings (2003-2004), and bibliographic review of secondary sources. Search terms included defibrillator, randomized controlled trials, clinical trials, andsudden death.

Study Selection Eligible studies were prospective randomized controlled trials of ICD or combined cardiac resynchronization therapy and defibrillator (CRT-D) vs medical therapy enrolling at least some individuals with NICM and reporting all-cause mortality as an outcome. Of 675 potentially relevant articles screened initially, 8 reports of randomized trials enrolling a total of 2146 patients with NICM were included.

Data Extraction Included studies were reviewed to determine the number of patients randomized, mean duration of follow-up, primary end point, mortality of ICD cohort, and mortality of control cohort.

Data Synthesis Five primary prevention trials enrolling 1854 patients with NICM were identified; pooled analysis suggested a significant reduction in total mortality among patients randomized to ICD or CRT-D vs medical therapy (risk ratio [RR], 0.69; 95% confidence interval [CI], 0.55-0.87; P = .002). Mortality reduction remained significant even after elimination of CRT-D trials. Two of the 3 secondary prevention trials presented subgroup estimates for ICD efficacy in NICM. Pooled analysis of these secondary prevention trials (n = 256 patients with NICM) indicated an equivalent but nonsignificant mortality reduction with ICD therapy (RR, 0.69; 95% CI, 0.39-1.24; P = .22). Analysis of all 7 trials combined demonstrated a statistically significant 31% overall reduction in mortality with ICD therapy (RR, 0.69; 95% CI, 0.56-0.86; P = .002).

Conclusion ICD therapy appears to significantly reduce mortality in selected patients with NICM.