The efficacy and safety of aspirin in individuals with and without diabetes: A collaborative meta-analysis of individual participant data from randomised trials

Background: Whether aspirin is of net clinical benefit for the primary prevention of vascular events in those at high risk, such as some people with diabetes, is uncertain. Methods: Individual participant data from 26 randomised trials (123,361 individuals, 760,000 person-years) assessing aspirin versus control in people with and without diabetes were included. Major outcomes included serious vascular events (SVE: myocardial infarction, stroke or vascular death) and major extracranial bleeding (MEB). Findings: In the primary prevention setting, aspirin reduced SVE (0.65% vs 0.71% per year; relative risk (RR) 0.90 [95% confidence interval (CI) 0.85-0.96]; p=0.0006) and increased MEB (0.10% vs 0.07% per year; RR 1.52 [95% CI 1.30-1.78]; p<0.0001). These proportional effects were similar in people with and without diabetes (χ21=0.74, p=0.39 and χ21=1.00, p=0.32 respectively). Estimated absolute effects of aspirin according to diabetes status and vascular risk suggested that some people, particularly those with diabetes and 5-year risk >5%, may derive net clinical benefit in the primary prevention setting. At a 5-year risk level of 5-10%, for example, people without diabetes might have a 1.6% (95% CI 1.1-2.1) reduction in 5-year SVE balanced against a 0.5% (95% CI 0.3-0.8) increase in 5-year MEB. In comparison, people with diabetes and 5-year risk of 5-10% might have a 1.7% (95% CI 1.2-2.2) reduction in 5-year SVE balanced against a 0.3% (95% CI 0.1-0.4) increase in 5-year MEB. Even if an approximate halving of baseline vascular risk by modern preventative measures such as statins is factored in before aspirin therapy is considered, aspirin may still have a worthwhile net clinical benefit in such people at least at moderate vascular risk. Interpretation: Some people without previous disease but at higher vascular risk, such as people with diabetes and 5-year risk >5%, may derive net clinical benefit from aspirin. Further direct randomised evidence in these individuals will help clarify the balance of risks and benefits.