Low-dose aspirin increases risk of major bleeding, but not in diabetes

Low-dose aspirin for primary prevention of cardiovascular events is generally considered a bad idea because the risks of bleeding seem to outweigh the benefits. However, things may be different for people with diabetes.

A claims study used data from more than 4 million residents of Perugia, Italy, over a period of six years. Of these, 186 425 people were taking low-dose aspirin. Propensity score matching was used to assign one control, without long-term low-dose aspirin use, to each participant taking aspirin. Propensity scoring took into account several potential confounders, including age, sex, diabetes, previous admission to hospital for cardiovascular disease, and use of various drugs. However, behavioural risk factors such as smoking, poor diet, excess alcohol consumption, obesity, and use of over-the-counter drugs (including aspirin) could not be considered in this study.

Overall, 5.58 (95% CI 5.39 - 5.77) hospital admissions for major bleeding in the gastrointestinal tract or brain occurred per 1 000 person years in users of aspirin; the incidence in non-users was 3.60 (3.48 - 3.72) per 1 000 person years (incidence rate ratio 1.55, 1.48 - 1.63). In other words, for 1 000 people treated over a year, two extra cases of major bleeding will occur – about the number of major cardiovascular events prevented in people with a 10-year risk of 10 - 20%.

In people with diabetes, however, no increased risk was seen (1.09, 0.97 - 1.22); the risk of admission to hospital for major bleeding was increased by about one-third in all patients with diabetes, irrespective of aspirin use. This could be due to the high platelet reactivity seen in diabetes. The clinical implications that this might have remain to be explored.

Of note, the incidence of major bleeding found in this study is about 5 times higher than that reported in randomised trials.

De Berardis G, et al. JAMA 2012;307(21):2286-2294. doi:10.1001/jama.2012.5034.