A COMPARISON OF ASPIRIN WITH PLACEBO IN PATIENTS TREATED WITH WARFARIN AFTER HEART-VALVE REPLACEMENT

Background. Despite the use of warfarin, major systemic embolism remai ns an important complication in patients with heart-valve replacement. Although the addition of antiplatelet agents has the potential to red uce this complication, their efficacy and safety when given in combina tion with warfarin are uncertain. Methods. In a randomized, double-bli nd, placebo-controlled trial, we assessed the efficacy and safety of a dding aspirin (100 mg per day) to warfarin treatment (target internati onal normalized ratio, 3.0 to 4.5) in 370 patients with mechanical hea rt valves or with tissue valves plus atrial fibrillation or a history of thromboembolism. Results. A total of 186 patients were randomly ass igned to aspirin and 184 to placebo, and they were followed for up to 4 years (average, 2.5). Major systemic embolism or death from vascular causes occurred in 6 aspirin-treated patients (1.9 percent per year) and 24 placebo-treated patients (8.5 percent per year) (risk reduction with aspirin, 77 percent; 95 percent confidence interval, 44 to 91 pe rcent; P<0.001). Major systemic embolism, nonfatal intracranial hemorr hage, or death from hemorrhage or vascular causes occurred in 12 patie nts assigned to aspirin (3.9 percent per year) and 28 patients assigne d to placebo (9.9 percent per year) (risk reduction, 61 per cent; 95 p ercent confidence interval, 24 to 80 percent; P = 0.005); major system ic embolism or death from any cause occurred in 13 patients (4.2 perce nt) and 33 patients (11.7 percent), respectively (risk reduction, 65 p ercent; 95 percent confidence interval, 33 to 82 percent; P<0.001); an d death from all causes occurred in 9 patients (2.8 percent) and 22 pa tients (7.4 percent), respectively (risk reduction, 63 percent; 95 per cent confidence interval, 19 to 83 percent; P = 0.01). Bleeding occurr ed in 71 patients in the aspirin group (35.0 percent), as compared wit h 49 patients in the placebo group (22.0 percent) (increase in risk, 5 5 percent; 95 percent confidence interval, 8 to 124 percent; P = 0.02) ; major bleeding occurred in 24 and 19 patients, respectively (increas e in risk, 27 percent; 95 percent confidence interval, -30 to 132 perc ent; P = 0.43). Conclusions. In patients with mechanical heart valves and high-risk patients with prosthetic tissue valves, the addition of aspirin to warfarin therapy reduced mortality, particularly mortality from vascular causes, together with major systemic embolism. Although there was some increase in bleeding, the risk of the combined treatmen t was more than offset by the considerable benefit.