Biventricular (BV) pacing is under clinical investigation for the treatment of heart failure. Its impact on mortality is unknown. Patients with heart failure and ventricular tachyarrhythmias received an implantable cardioverter defibrillator with BV pacing capability. Patients were randomized 1:1 to BV pacing or no pacing, then crossed over to the alternate mode after 3 months. All-cause mortality was measured in each arm up to the point of crossover. Fifteen of 222 patients died between implant and crossover. Five patients died while programmed to BV pacing and 19 died while programmed to no pacing. Survival in the BV pacing arm was 93 +/- 4% versus 86 +/- 6% in the no pacing arm (P = 0.18). In a patient population with symptomatic heart failure and ventricular arrhythmias, BV pacing does not appear to be associated with excess mortality. Larger and longer studies will be needed to determine if BV pacing confers a survival benefit.