Aspirin for Primary Prevention of Cardiovascular Events in Diabetes: Title and subTitle BreakStill an Open Question

The use of aspirin for primary prevention of cardiovascular events in individuals with diabetes is widely recommended by existing guidelines, but the evidence supporting its efficacy is surprisingly scarce.¹ Recommendations seem based mainly on extrapolations from data from other high-risk groups, rather than on solid data derived from studies conducted specifically in patients with diabetes. Indeed, an increasing amount of evidence suggests that the efficacy of antiplatelet therapy in patients with diabetes may be lower than in individuals without diabetes.²

In this issue of JAMA, Ogawa and colleagues³ report the results of the JPAD trial (Japanese Primary Prevention of Atherosclerosis With Aspirin for Diabetes), specifically designed to address the issue of antiplatelet therapy for the primary prevention of cardiovascular events in diabetes. The trial included 2539 patients with type 2 diabetes and without a history of atherosclerotic disease randomly assigned to low-dose aspirin (80-100 mg per day) or a nonaspirin control group.³ After a median follow-up of 4.37 years, aspirin therapy was associated with a nonsignificant reduction (hazard ratio [HR], 0.80; 95% confidence interval [CI], 0.58-1.10) in the risk of the primary composite end point, including fatal or nonfatal ischemic heart disease, fatal or nonfatal stroke, transient ischemic attack, and peripheral arterial disease. No differences in total mortality were detected. In a subgroup analysis restricted to individuals aged 65 years or older, a marginally significant reduction in the incidence of the primary end point was documented in patients treated with aspirin as compared with controls (HR, 0.68; 95% CI, 0.46-0.99). Aspirin therapy was associated with an increased risk of gastrointestinal bleeding and retinal hemorrhage. Four patients in the aspirin group and none in the control group required blood transfusion.

What does this study add to current knowledge of the efficacy of aspirin for the primary prevention of cardiovascular events in patients with diabetes? Should the study results be considered as definite proof that aspirin is less effective in individuals with diabetes than in other high-risk groups?

The last meta-analysis on the efficacy of antiplatelet therapy in the prevention of major cardiovascular events showed a clear benefit for the entire population of more than 140 000 patients (22% reduction in the risk of major cardiovascular events), but no statistically significant benefit was documented in the subgroup of about 5000 diabetic patients (7% risk reduction).⁴ The results of the JPAD trial (ie, a nonsignificant reduction of 20% in the primary end point) are compatible with the overall results of the meta-analysis, and a benefit of treatment, at least for the primary study end point, cannot be ruled out. Nevertheless, a much lower effect, of the same magnitude as that documented in the diabetes subgroup of the meta-analysis, is also compatible with the wide CIs of the relative risk estimate reported by Ogawa et al.

The lack of precision and the low statistical power in the JPAD trial are the consequence of the substantially lower-than-expected event rate in the trial population. Sample-size estimation was based on the assumption that 52 events per 1000 Japanese diabetic patients would have occurred every year. However, the observed incidence rate for events included in the primary end point was less than one-third of that expected—ie, 13.6 per 1000 person-years in the aspirin group and 17.0 per 1000 person-years in the control group. The study was also underpowered to detect differences in the incidence of rare events, such as hemorrhagic stroke or severe bleeding. As a consequence, much longer follow-up, or a larger sample, would have been needed to reliably estimate benefits and risks related to aspirin therapy.

The trial also poses some problems in terms of generalizability of results. The low event rate reported in the JPAD trial documents that the study population had a very low baseline risk of cardiovascular events; would the same results apply to a population with a substantially higher cardiovascular risk? Would the benefits clearly outweigh the risks in individuals with a risk of cardiovascular events similar to that usually found in everyday practice? Furthermore, the epidemiology of cardiovascular disease in Japan is substantially different from that of non-Japanese, Western populations, as documented by the high incidence of cerebrovascular events, particularly hemorrhagic stroke, and the relatively low incidence of coronary heart disease.⁵ A meta-analysis of primary prevention randomized controlled trials suggested that aspirin reduced the risk of myocardial infarction, but not the risk of ischemic stroke, in men, whereas in women aspirin predominantly reduced the risk of stroke with no effect on myocardial infarction.⁶ The overall effect of antiplatelet therapy in women and men could therefore substantially differ in Japanese patients as compared with other patient populations. Given these findings, the role of antiplatelet therapy in the context of the overall approach to cardiovascular risk reduction in individuals with diabetes remains to be elucidated.

Recently, the results of 2 large-scale trials have called into question the role of aggressive metabolic control in reducing the risk of major cardiovascular events, particularly in high-risk populations.^{7 - 8} These observations, coupled with the consistent documentation of the difficulty in reaching the desired targets for blood pressure and lipid profile in a large proportion of individuals with diabetes,⁹ underscore the need for effective preventive strategies, particularly among elderly patients. In this respect, the findings of the JPAD trial relative to patients older than 65 years are encouraging. Nevertheless, these results should be interpreted with caution, considering the inherent limitations of subgroup analyses and the difficulty in obtaining a reliable estimate of risks for rare adverse events. The risk of major bleeding sharply increases in individuals older than 70 years,¹⁰ making the balance between benefits and harms uncertain.

In addition to JPAD, 3 large-scale randomized clinical trials are currently investigating the role of aspirin in patients with diabetes. Two of these trials, ASCEND (A Study of Cardiovascular Events in Diabetes, International Standard Randomized Controlled Trial Number ISRCTN60635500) and ACCEPT-D (Aspirin and Simvastatin Combination for Cardiovascular Events Prevention Trial in Diabetes),¹¹ are expected to enroll more than 15 000 patients overall and will help clarify the role of aspirin for primary prevention of cardiovascular disease. The results of the third trial, the Prevention of Progression of Arterial Disease and Diabetes (POPADAD) study, have been recently published. The study, involving 1276 patients with asymptomatic peripheral arterial disease, found no evidence of benefit of aspirin on cardiovascular events and mortality.¹²

While waiting for the results of ongoing studies, a meta-analysis of individual patient data from the most recent trials would provide the opportunity to explore, using relevant clinical end points, the incremental effect of antiplatelet therapy over standard care, usually including antihypertensive agents and statins, and its role in specific populations such as elderly patients or women. In the meantime, the decision to prescribe aspirin should be made on an individual patient basis after careful evaluation of the balance between the expected benefits and the risk of major bleeding. The issue of aspirin therapy for patients with diabetes is an example of how, in the presence of a long-lasting uncertainty, scientific organizations or governmental bodies should provide the foundation for answering this question by promoting pragmatic, large-scale clinical trials. Considering all diabetic patients with no history of cardiovascular disease (except those with documented contraindications or perceived indications) as candidates for randomized clinical trials would represent a major contribution to the credibility of scientific methods in guiding practice.