Aspirin for Primary Prevention of Cardiovascular Events in People with Diabetes

Updated:May 19,2014

Darren K. McGuire, MD, MHSc
Pub Date: Thursday, May 27, 2010
Author: M. Odette Gore, MD


Pignone M, Alberts MJ, Colwell JA, et al. Aspirin for primary prevention of cardiovascular events in people with diabetes: a position statement of the American Diabetes Association, a scientific statement of the American Heart Association, and an expert consensus document of the American College of Cardiology Foundation. Circulation 2010. Published online before print, May 27, 2010. 10.1161/CIR.0b013e3181e3b133.

Article Text

The worldwide prevalence of diabetes mellitus (DM) exceeded 220 million in 2009 and is expected to top 360 million over the next two decades [1,2], increasingly including younger adults, adolescents, and even children [3], and exposing an ever-expanding population to markedly elevated lifetime risks of incident cardiovascular disease (CVD) complications. Thus, the imperative to optimize strategies with regard to efficacy, safety, benefit/risk ratio, and cost efficiency for primary CVD risk prevention in patients with DM has never been more acute. In that light, the recently published position paper developed jointly by the American Diabetes Association (ADA), American Heart Association (AHA), and American College of Cardiology Foundation (ACCF) provides a systematic review of the published data and concise guidance for the most appropriate, evidence-based use of aspirin (ASA) for primary CVD risk modification in patients with DM.[4]

Ideally, guidance with regard to the application of preventive therapies and strategies should be based on robust evidence derived from prospective, randomized trials specifically designed and adequately powered to assess their effects on important clinical outcomes. Across the spectrum of primary CVD prevention, including but not limited to the population of patients with DM, the paucity of such clear-cut data for the majority of individual primary-risk interventions often leaves ample room for controversy, as is the case for ASA for primary CVD prevention in DM. As summarized in the position paper, a number of recently published clinical trials and meta-analyses have demonstrated less efficacy than previously believed with regard to the treatment effects of ASA on risk for primary coronary heart disease complications and stroke. The observations have resulted in modifications of treatment recommendations toward more restricted subgroups of patients with diabetes at particularly high risk for CVD events, both in updated guidelines previously published [5,6], and in the present position paper [4], while at the same time recognizing the need for ongoing investigation to resolve remaining uncertainties.

These new summary recommendations are largely consistent with the most recent ADA guidelines [5], but are more explicit in discouraging the use of ASA for primary prevention in patients with DM and low CVD risk. Likewise, this new position paper is in line with, but slightly more restrictive than, the recommendations of the U.S. Preventive Services Task Force [6], that advised against the use of ASA for primary prevention in women younger than 55 years old and in men younger than 45 years old (regardless of DM status). And in light of the recent data available over the past year, all of these recommendations are much more restrictive than previous iterations of the ADA guidelines [7] and of the joint ADA/AHA scientific statement from 2007 [8], which recommended ASA for primary prevention in all DM patients more than 40 years of age, or in diabetic patients with additional CVD risk factors regardless of age. This reversal in recommendations consistently endorsing use in a more restrictive, higher-risk DM patient population across professional societies reflects a sobering recognition of the present uncertainty regarding the utility of ASA in primary CVD prevention in the DM population, and of the evolution of expert opinion finally being supplanted by clinical trial data. Although this recent paper is well organized, concise, and culminates with very helpful specific clinical recommendations, two points warrant further consideration: (a) expanded consideration of the benefit/risk ratio with regard to bleeding complication of ASA; and (b) expanded discussion of topics of future investigation.

Benefit/Risk Considerations

In most recent study publications, guideline iterations, and the present position paper, most attention has focused on the modest magnitude of absolute and relative risk reductions observed attributable to ASA therapy. However, a critical counterpoint to consider in the overall determination of benefit/risk ratio is the very low placebo-subtracted incremental risk of severe bleeding complications associated with ASA therapy in the primary prevention cohorts studied--estimates from the ADA/AHA/ACCF meta-analyses of one intracerebral hemorrhage (ICH) and three serious gastrointestinal bleeds (GIB) per 10,000 patient-years of ASA exposure. As bleeding risk is clearly associated with a number of factors related to CVD risk [9], these bleeding complication rates are in stark contrast with an incremental risk of severe bleeding associated with ASA use in secondary prevention populations estimated in the range of 1 per 500 to 1000 patient-years of exposure.[9,10]

This consideration of absolute incremental risk for bleeding complications highlights the limitations of interpreting relative risk estimates when absolute risk is so small, especially when the relative risk estimate is reported in the absence of the absolute risk--a common situation that tends to inappropriately exaggerate and sensationalize the differences observed. For example, as per Table 2 in the meta-analysis recently reported by De Berardis et al. [11], a two- to fivefold increased risk of major bleeding complications seems alarming and prohibitive, but the clinical relevance can only be deduced by knowing the absolute event rates, which are not reported. Carrying this example further, if using the combined incremental risk estimates for ICH/serious GIB from the ADA/AHA/ACCF meta-analysis of 1 per 2,500 patient-years (0.04% annualized risk) the number-needed-to-treat for a year to cause harm would be 2,500 patients. In contrast, using the summary estimate from the position paper of 10% relative risk reduction (RRR) with ASA for CVD events in a population with as low as a 10% CVD risk, this would yield a 1% absolute CVD risk reduction (number needed to treat = 100); therefore, in this data-based example, 25 CVD events would be prevented for every ICH/serious GIB caused--an undoubtedly favorable risk/benefit balance.

In light of these considerations, the bleeding risk of ASA in the setting of primary prevention may be much less of a concern than most have previously perceived (and the writers of the recent meta-analyses, guidelines, and position papers have amplified). However, in this context, even extremely small increments of efficacy may achieve a favorable risk/benefit balance among DM patients with intermediate and higher CVD risk. Consequently, given the very low cost of ASA and the population attributable CVD risk associated with DM, net clinical benefit may be achieved with cost efficiency even among DM patients at only intermediate CVD risk, including those with risks well below the 10% threshold endorsed in the present paper. Therefore, in the hypothetical example provided in the position paper where a patient at 20% 10-year CVD risk could have risk reduced to 13% with a statin, and then to 10% with blood pressure control, we contend the premise presented that it then might not be worth treating with ASA to achieve an additional 10% relative reduction--the data presented simply do not support this conclusion.

Ongoing Uncertainty and Future Research
In the position paper, the authors comment on needs for future research, which we will expand to more extensively frame the myriad questions and extensive uncertainty that remain regarding the optimal use of ASA for CVD risk mitigation in patients with DM. Extending beyond the commentary about ongoing primary prevention trials of ASA use in DM cohorts, future research will also be required to establish the optimal ASA dosage for both primary and secondary prevention in DM. Underscoring this uncertainty is the observation in the present meta-analysis that for the CHD endpoint, the ETDRS trial contributed almost one-half of the analytic weight--a study that compared ASA 650 mg daily versus placebo, contrasted by the fact that every contemporary primary prevention guideline endorses lower-dose ASA (ranging from 75 to 162 mg daily). However, no clinical outcomes trials have been executed comparing various ASA dosing, and these dosing recommendations remain grounded in "expert opinion." Whether higher ASA dosing would result in more robust CVD risk reduction and at what cost regarding bleeding and safety remains uncertain.Another area of emerging interest and uncertainty centers on the most appropriate dosing frequency of ASA, especially among patients with diabetes with numerous perturbations of platelet morphology, activity, and turnover.[12] Based on biochemical profiling, data are emerging that some patients may require twice daily dosing (or perhaps more frequently) to yield consistent pharmacodynamic effects on cyclooxygenase inhibition and thromboxane generation.[13]

Further investigation is also required to clarify whether there are differences in efficacy or safety according to DM type--type 1 or type 2--and what influence primary versus secondary prevention status has on these differences, if any exist. Presently, due to virtually nonexistent data specifically regarding ASA effects in patients with type 1 diabetes, the recommendations tend to be extrapolated from the dataset deriving almost exclusively from the study of patients with type 2 DM. Furthermore, given the similarity between the Preventive Services Task Force recommendations for primary CVD risk prevention with ASA for the overall population [6], and those specific to DM [4,5], whether DM status differentially influences such treatment decisions remains less clear than we have previously believed.

These are just a few of the key remaining questions with regard to ASA use for CVD risk mitigation, overall and specifically in the setting of DM. Because very large and very long (i.e., very expensive) randomized clinical trials will be required to address most of these questions, the clinical and research community will be challenged to develop creative methods to potentially address the issues. Among other possibilities, given the proliferation of CVD clinical outcome trials assessing glucose-lowering drugs in development (and ultimately those on the market as well) in response to modifications of the regulatory landscape [14], these programs provide a rich opportunity to consider factorial randomized comparisons of various ASA strategies added to the primary study comparisons.

Finally, on a practical note, the authors endorse the use of three online risk prediction tools. Of these, we found the UKPDS risk engine to be most useful for clinical purposes as proposed ( takes you directly to the risk engine download). This risk prediction engine is specific to those with prevalent diabetes, is quite fast, and outputs individual predictions for stroke and coronary heart disease. The Atherosclerosis Risk in Communities risk engine includes only CHD risk prediction and not stroke, and also can be used for individuals older than 45 years old--the ADA tool, although comprehensive and useful for individuals, is much too complicated and slow to be of use in a busy clinical practice; waiting time of several minutes with option to have the results e-mailed later.

Overall, this position paper and updated meta-analysis succinctly and quantitatively distills our current knowledge of ASA therapy for primary CVD prevention in DM, and provides a set of practical clinical practice recommendations that may in time prove imperfect, but for now represent the state-of-the-art and provide substantial consistency across professional societies and organizations.


  1. The World Health Organization. World health organization: diabetes, fact sheet no. 312, November 2009.
  2. Wild S, Roglic G, Green A, et al. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004;27:1047-1053.
  3. Rosenbloom AL, Joe JR, Young RS, Winter WE. Emerging epidemic of type 2 diabetes in youth. Diabetes Care 1999;22:345-354.
  4. Pignone M, Alberts MJ, Colwell JA, et al. Aspirin for primary prevention of cardiovascular events in people with diabetes: a position statement of the American Diabetes Association, a scientific statement of the American Heart Association, and an expert consensus document of the American College of Cardiology Foundation. Circulation 2010. Published online before print, May 27, 2010. 10.1161/CIR.0b013e3181e3b133,
  5. American Diabetes Association. Standards of medical care in diabetes - 2010. Diabetes Care 2010;33[Suppl 1]:S11-61.
  6. US Preventive Services Task Force. Aspirin for the prevention of cardiovascular disease: U.S. Preventive services task force recommendation statement. Ann Intern Med 2009;150:396-404.
  7. American Diabetes Association. Standards of medical care in diabetes - 2009. Diabetes Care 2009;32[Suppl 1]:S13-61.
  8. Buse JB, Ginsberg HN, Bakris GL, et al. Primary prevention of cardiovascular diseases in people with diabetes mellitus: a scientific statement from the American Heart Association and the American Diabetes Association. Circulation 2007;115:114-126.
  9. Baigent C, Blackwell L, Collins R, et al. Aspirin in the primary and secondary prevention of vascular disease: Collaborative meta-analysis of individual participant data from randomised trials. Lancet 2009;373:1849-1860.
  10. Hernandez-Diaz S, Garcia Rodriguez LA. Cardioprotective aspirin users and their excess risk of upper gastrointestinal complications. BMC Med 2006;4:22.
  11. De Berardis G, Sacco M, Strippoli GF, et al. Aspirin for primary prevention of cardiovascular events in people with diabetes: meta-analysis of randomised controlled trials. BMJ 2009;339:b4531.
  12. Mathewkutty S, McGuire DK. Platelet perturbations in diabetes: implications for cardiovascular disease risk and treatment. Expert Rev Cardiovasc Ther 2009;7:541-549.
  13. Dragani A, Pascale S, Recchiuti A, et al. The contribution of cyclooxygenase-1 and -2 to persistent thromboxane biosynthesis in aspirin-treated essential thrombocythemia: implications for antiplatelet therapy. Blood 2010;115:1054-1061.
  14. Gore MO, McGuire DK. Cardiovascular disease and type 2 diabetes mellitus: regulating glucose and regulating drugs. Curr Cardiol Rep 2009;11:258-263.

-- The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association

AHA Scientific Journals

AHA Scientific Journals