Article Text

Download PDFPDF
Systematic review with meta analysis
Aspirin reduces cardiovascular events in primary prevention of cardiovascular disease but at a near equivalent risk of increased bleeding
  1. Michael D Miedema1,
  2. Salim S Virani2
  1. 1Minneapolis Heart Institute, Minneapolis, Minnesota, USA
  2. 2Michael E. DeBakey Veterans Affairs Medical Center and Baylor College of Medicine, Houston, Texas, USA
  1. Correspondence to : Professor Salim S Virani, Health Services Research and Development (152), 2002 Holcombe Boulevard, Houston, TX 77030, USA; virani{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Commentary on: OpenUrlCrossRefPubMed.


The use of aspirin in primary prevention of cardiovascular disease (CVD) remains controversial, as randomised controlled trials (RCTs) have produced mixed results. Recently, the United States Preventive Services Task Force (USPSTF) published updated recommendations on the use of aspirin for primary prevention of CVD and colorectal cancer.1 This systematic review and meta-analysis served as the basis for these USPSTF recommendations for primary CVD prevention.


All trials from the previous 2009 USPSTF guideline on the use of aspirin for primary CVD prevention (minimum of 75 mg every other day for 1 year or more) in adults aged 40 or more were included, and a comprehensive literature search for additional trials from January 2008 to June of 2014 was performed. Outcomes evaluated included non-fatal myocardial infarction (MI), non-fatal stroke (all types), CVD mortality (composite of death due to MI, stroke and CVD) and all-cause mortality. A Mantel-Haenszel fixed effects model was used as the primary statistical analysis method. Forest plots were used to show adjusted relative risk (RR). Several prespecified subpopulations were analysed, including analyses according to sex, age and diabetic status.


A total of 11 eligible RCTs (N=118 445) were included. There was significant heterogeneity in the sample size, inclusion criteria, duration of follow-up and aspirin dosage among the trials. Of 11 trials, 8 studied aspirin doses of 100 mg or less per day or 100 mg or less every other day. Aspirin was associated with a significant 22% risk reduction in non-fatal MI (RR 0.78 (95% CI 0.71 to 0.87)) with significant heterogeneity among included trials. Analyses comparing aspirin with the control group showed no difference in non-fatal stroke (RR 0.95 (95% CI 0.85 to 1.06)) or CVD mortality (RR 0.94 (95% CI 0.86 to 1.03)) with little or no all-cause mortality benefit (RR 0.94 (95% CI 0.89 to 0.99)). Restricting analyses to eight trials involving 100 mg or less per day of aspirin showed similar and a statistically significant 17% reduction in non-fatal MI, and a statistically significant reduction in non-fatal stroke (RR 0.86 (95% CI 0.76 to 0.98) with no benefit for CVD or all-cause mortality. Benefits began within the first 1–5 years. Older individuals seemed to derive a greater RR reduction for non-fatal MI from aspirin therapy, but no other effect modification by sex or diabetic status was noted.


Consistent with prior studies, this current meta-analysis demonstrated a modest benefit of aspirin therapy in CVD prevention largely driven by a reduction in non-fatal MI events. Importantly, this benefit persisted with 100 mg per day or less of aspirin therapy. No clinically significant reduction in stroke events, CVD mortality or all-cause mortality was seen. In general, the RCTs studying aspirin for primary prevention have trended from efficacy in the earlier studies towards more neutral results in the four more recent RCTs.2 Rates of statin use and blood pressure control were much lower in earlier aspirin RCTs compared with the most recent aspirin RCTs. Having patient populations with well-controlled risk factors may have removed the low-hanging fruit that was present in previous aspirin trials.

Another important issue to take into account is the increased risk of bleeding associated with aspirin therapy. As shown in an accompanying article by the same authors, even low-dose aspirin therapy increased risk of major gastrointestinal bleeding by 58% (OR 1.58 (95% CI 1.29 to 1.95)) and possibly haemorrhagic stroke (OR 1.27 (95% CI 0.96 to 1.68)). Therefore, any net benefit in reduction of non-fatal MI events associated with aspirin therapy could be negated by an increase in bleeding events. Taken together, low-dose aspirin therapy should be reserved for patients with high 10-year CVD risk and acceptable bleeding risk. Healthy lifestyle, risk factor modification (blood pressure control, smoking cessation) and possibly statin use (associated with a much better risk–benefit profile compared with aspirin therapy) where appropriate should be cornerstones of primary CVD prevention. Fortunately, the results of four large ongoing RCTs will provide guidance on the optimal use of aspirin in primary CVD prevention in the modern era.

Implications for practice

Although ambiguity remains, this meta-analysis provides a rationale for the risk-based approach limiting aspirin use to middle-aged individuals with high (≥10%) 10-year CVD risk that was endorsed by the USPSTF while taking into account their bleeding risk. Given the relatively modest benefit which seems to be diminishing when recent RCTs are analysed coupled with definite harms, a thorough patient-provider discussion concerning risks and benefits is warranted before aspirin initiation for primary CVD prevention.


View Abstract


  • Competing interests None declared.

  • Provenance and peer review Commissioned; internally peer reviewed.