Evidence suggests that physiologic pacing (dual-chamber or atrial) may be superior to single-chamber (ventricular) pacing because it is associated with lower risks of atrial fibrillation, stroke, and death. These benefits have not been evaluated in a large, randomized, controlled trial. At 32 Canadian centers, patients without chronic atrial fibrillation who were scheduled for a first implantation of a pacemaker to treat symptomatic bradycardia were eligible for enrollment. We randomly assigned patients to receive either a ventricular pacemaker or a physiologic pacemaker and followed them for an average of three years. The primary outcome was stroke or death due to cardiovascular causes. Secondary outcomes were death from any cause, atrial fibrillation, and hospitalization for heart failure. A total of 1474 patients were randomly assigned to receive a ventricular pacemaker and 1094 to receive a physiologic pacemaker. The annual rate of stroke or death due to cardiovascular causes was 5.5 percent with ventricular pacing, as compared with 4.9 percent with physiologic pacing (reduction in relative risk, 9.4 percent; 95 percent confidence interval, -10.5 to 25.7 percent [the negative value indicates an increase in risk]; P=0.33). The annual rate of atrial fibrillation was significantly lower among the patients in the physiologic-pacing group (5.3 percent) than among those in the ventricular-pacing group (6.6 percent), for a reduction in relative risk of 18.0 percent (95 percent confidence interval, 0.3 to 32.6 percent; P=0.05). The effect on the rate of atrial fibrillation was not apparent until two years after implantation. The observed annual rates of death from all causes and of hospitalization for heart failure were lower among the patients with a physiologic pacemaker than among those with a ventricular pacemaker, but not significantly so (annual rates of death, 6.6 percent with ventricular pacing and 6.3 percent with physiologic pacing; annual rates of hospitalization for heart failure, 3.5 percent and 3.1 percent, respectively). There were significantly more perioperative complications with physiologic pacing than with ventricular pacing (9.0 percent vs. 3.8 percent, P<0.001). Physiologic pacing provides little benefit over ventricular pacing for the prevention of stroke or death due to cardiovascular causes.