NEW YORK (Reuters Health) – In a large population-based cohort from Italy, aspirin therapy for primary prevention of cardiovascular events was associated with a significantly increased risk of major bleeding.

“Our study suggests that in the real world the incidence of major bleeding (with aspirin) is markedly higher than expected, based on results of randomized trials,” Dr. Antonio Nicolucci, from Consorzio Mario Negri Sud in Maria Imbaro, Italy, told Reuters Health by email.

“Therefore, before prescribing aspirin to a patient with moderate cardiovascular risk, physicians should carefully evaluate the risk of bleeding, which is particularly elevated in elderly people, in those with a previous episode of major bleeding, in smokers, in those with hypertension, and in those taking other antiplatelet agents or anticoagulants,” he advised.

In the study, published June 6 in JAMA, patients with diabetes had high rates of bleeding, independent of aspirin use.

Diabetes “might represent a different population in terms of both expected benefits and risks associated with antiplatelet therapy,” the study team writes. Dr. Nicolucci told Reuters Health, “The role of aspirin in individuals with diabetes remains to be clarified.”

In a linked editorial in JAMA entitled, Hemorrhagic Complications Associated With Aspirin: An Underestimated Hazard in Clinical Practice?, Dr. Jolanta M. Siller-Matula of the Medical University of Vienna says this study “underscores that the potential risk of bleeding should be carefully considered in decision making.”

The findings also “reinforce the European recommendations for aspirin use (which) do not recommend aspirin for primary prevention because of its unfavorable risk-to-benefit profile,” she notes.

Dr. Siller-Matula added in an email to Reuters Health: “I believe that the benefits of aspirin in primary prevention were somehow overblown. Not in terms of the relative risk reduction (which is still a favorable 12%) but in (terms) of number needed to treat, which is very high (1,430), six-fold higher than for the secondary prevention.”

Using administrative health data from 4.1 million citizens of Puglia, Dr. Nicolucci and colleagues identified 186,425 adults who began taking low-dose aspirin (up to 300 mg) during a six-year index period (2003-2008). These individuals were propensity-matched on a 1-on-1 basis with individuals who did not take aspirin during this period. The main outcome of interest was hospital admission for major bleeding (GI or cerebral) occurring after the start of aspirin therapy.

During six years, 6,907 first episodes of major bleeding requiring hospitalization were registered, including 4,487 episodes of GI bleeding and 2,464 episodes of intracranial bleeding. The overall incidence rate of bleeding events per 1000 person-years was 5.58 for aspirin users compared with 3.60 for non-users, yielding an incidence rate ratio (IRR) of 1.55.

Use of aspirin for primary prevention, Dr. Siller-Matula points out, was associated with “a 55% relative risk increase in major hemorrhagic events (both intracranial and gastrointestinal; absolute risk increase 0.2%), which corresponds to an excess of 20 major bleeding events among 10,000 treated patients (number needed to harm, 500). This is comparable with the relative risk estimates derived from meta-analyses of randomized trials.”

The increased risk of bleeding with aspirin was evident in most subgroups with the exception of patients with diabetes (IRR 1.09). Irrespective of aspirin use, diabetes was independently associated with a 36% increased risk of major bleeding events (IRR 1.36). Dr. Siller-Matula says this observation highlights two “important issues: decreased antiplatelet efficacy of aspirin in patients with diabetes and the overlap of risk factors that predict vascular and bleeding events.”

The effect of aspirin for primary prevention of cardiovascular disease in patients with diabetes remains “unclear,” Dr. Siller-Matula says, adding that two ongoing studies (ACCEPT-D and ASCEND) should provide additional data on the safety and benefit profile of aspirin in diabetic patients.

“While waiting for the results of ongoing trials, it seems prudent to continue to use aspirin in individuals with a 10-year risk of cardiovascular events over 15%,” Dr. Nicolucci said.

Dr. Siller-Matula added, “The recommendation for aspirin use based on the individual risk profile might seem not appropriate in each patient, as the risk factors for ischemic and bleeding events overlap. We need a new risk score for aspirin use in primary prevention, which incorporates both aspects: bleeding and thrombosis. Such a tool would be very helpful.”

In her editorial, she writes that with aspirin as well as the more potent platelet inhibitors and anticoagulants, “there is only a thin line between efficacy and safety, and the reduction in ischemic events comes at the cost of increased major bleedings. Therefore, future studies investigating the risks and benefits for individual patients appear to be mandatory to help physicians appropriately make recommendations about aspirin use for primary prevention.”

The study also found that statin therapy was associated with a 33% lower risk of major bleeding. “This can represent an additional positive effect of statins,” Dr. Nicolucci said. The authors note in their paper, however, that the effect of statins on intracranial bleeding is “controversial” and the suggestion in this study of a protective effect “should be studied further.”

Dr. Siller-Matula reports receiving speaker fees from Eli Lilly and AstraZeneca. Dr. Nicolucci reports a research grant and lecture fees including service on speakers bureaus from Bayer.

SOURCE:

Association of Aspirin Use With Major Bleeding in Patients With and Without Diabetes

JAMA 2012;307:2286-2294,2318-2320.