Aspirin and/or antioxidants did not prevent CV events in diabetes and peripheral arterial disease

Question In patients with diabetes and asymptomatic peripheral arterial disease, do aspirin or antioxidants reduce risk for cardiovascular (CV) events? Methods Design Randomized, 2 2 factorial, placebo-controlled trial. Current Controlled Trials ISRCTN53295293. Allocation Concealed.* Blinding Blinded (patients, clinicians, data collectors, and outcome adjudication committee).* Follow-up period Median 7 years. Setting 16 hospitals and 188 primary care groups in Scotland, UK. Patients 1276 patients 40 years of age (mean age 60 y, 56% women) who had type 1 or 2 diabetes, asymptomatic peripheral arterial disease (anklebrachial pressure index 0.99), and no symptomatic CV disease. Exclusion criteria included regular use of study medications, contraindication to aspirin, and serious physical or psychiatric illness. Intervention Aspirin, 100 mg, plus antioxidant capsule (n =320); aspirin plus placebo (n =318); placebo plus antioxidant capsule (n =320); or placebo plus placebo (n =318). Additional CV risk reduction therapy, as appropriate, was encouraged. Outcomes A composite of death from coronary heart disease or stroke, nonfatal myocardial infarction (MI) or stroke, or amputation for critical limb ischemia; a composite of death from coronary heart disease or stroke; and all-cause death. The trial had 73% power to detect a 25% relative reduction in event rate. Patient follow-up 99% (intention-to-treat analysis). Main results Neither aspirin nor antioxidants reduced risk for any outcome compared with the corresponding placebo group (Table). Risk for all-cause death was increased in the antioxidant group (Table). No evidence of interaction between drugs was found. Conclusion In patients with diabetes and asymptomatic peripheral arterial disease, aspirin and antioxidants (separately or combined) did not reduce risk for cardiovascular events. Aspirin or antioxidants vs placebo in patients with diabetes and asymptomatic peripheral arterial disease Comparisons Outcomes at median 7 y Active drug Placebo RRR (95% CI) NNT Aspirin vs placebo Composite endpoint 18% 18% 2% (23 to 22) Not significant All-cause death 15% 16% 6% (22 to 27) Not significant RRI (CI) NNH (CI) Death from CHD or stroke 6.7% 5.5% 22% (21 to 88) Not significant Antioxidants vs placebo Composite endpoint 18% 18% 3% (19 to 29) Not significant All-cause death 18% 13% 44% (11 to 87) 18 (10 to 72) Death from CHD or stroke 6.6% 5.7% 20% (21 to 84) Not significant CHD = coronary heart disease; other abbreviations defined in Glossary. RRR, RRI, NNT, NNH, and CI calculated from hazard ratios in article. Death from CHD or stroke, nonfatal myocardial infarction or stroke, or amputation for critical limb ischemia. Commentary The American Diabetes Association (ADA) recommends aspirin for primary prevention of CV events in higher-risk patients with diabetes (1), based largely on the US Physicians Health Study, which showed a 60% reduction in MI. However, a recent meta-analysis of 5000 patients with diabetes showed no benefit for aspirin (2). The POPADAD trial found no clear benefit for either aspirin or antioxidants in preventing CV events in this population. 2 large ongoing trials of primary prevention in diabetes should help clarify the situation. The POPADAD trial enriched the primary prevention cohort by including patients with an abnormal anklebrachial index or suboptimal control of blood pressure and cholesterol levels. Nevertheless, the trials real limitation is low statistical power. Because of slow recruitment, the study was terminated at 1276 patients, far short of the original projection of 1600. Another recently published trial enrolled almost twice as many patients and also showed no difference in the same composite primary endpoint (hazard ratio 0.80, 95% CI 0.58 to 1.10) (3). Large differences can be ruled out with these sample sizes, but both trials were underpowered to detect small, potentially important, benefits. In POPADAD, the lower 95% CI of the hazard ratio was just short of a 25% benefit for aspirin. Although the recent evidence brings into question routine use of low-dose aspirin for primary prevention of CV events in diabetes, the real answer will probably come from an updated meta-analysis of individual patient data from all trials. In the meantime, we should be prepared to weigh the risks and benefits of long-term aspirin therapy with our patients on a case-by-case basis and not abandon the ADA recommendation. Further evaluation of the role of platelet aggregation testing to optimize aspirin dosing deserves consideration. Antioxidants do not seem to have a role in primary prevention of CV events.