The use of long-term oral anticoagulant treatment after myocardial infarction remains controversial because of conflicting findings on mortality in previous trials and the increased risk of bleeding associated with anticoagulants.
We have carried out a randomised, placebo-controlled, double-blind, multicentre trial in 3404 hospital survivors of myocardial infarction. Eligible patients were randomly assigned to anticoagulant (nicoumalone or phenprocoumon) or placebo treatment within 6 weeks of discharge. The target prothrombin time was 2.8-4.8 international normalised ratio. During mean follow-up of 37 (range 6-76) months there were 170 deaths among 1700 anticoagulant-treated patients and 189 in 1704 placebo-treated patients (hazard ratio 0.90 (95% CI 0.73-1.11)). Anticoagulant treatment led to significant reductions by comparison with placebo treatment in recurrent myocardial infarction (114 vs 242 patients; hazard ratio 0.47 (0.38-0.59)) and cerebrovascular events (37 vs 62; 0.60 (0.40-0.90)). Major bleeding complications were seen in 73 patients who received anticoagulants and 19 who received placebo.
We conclude that long-term oral anticoagulant treatment after myocardial infarction in low-risk patients has a limited effect on mortality but achieves substantial benefit by reducing the risk of cerebrovascular events and recurrent myocardial infarction.