Perioperative Aspirin Increases Major Bleeding After Noncardiac Surgery

In patients undergoing noncardiac surgery, giving aspirin before surgery and throughout the early post-operative period fails to decrease risk of death or non-fatal myocardial infarction (MI) and is associated with a significantly increased risk of major bleeding, according to data presented March 31, 2014, at the American College of Cardiology/i2 Scientific Session and simultaneously published in The New England Journal of Medicine. The results were similar regardless of whether patients had or had not been taking aspirin before the study. 

For the international, randomized, controlled POISE-2 trial (Perioperative Ischemic Evaluation 2), researchers led by P.J. Devereaux, MD, PhD, of the Population Health Research Institute (Ontario, Canada), enrolled 10,011 patients undergoing noncardiac surgery at 135 hospitals in 23 countries between July 2010 and December 2013. Patients were grouped by aspirin use (any dose daily for 4 of the 6 weeks before surgery; n = 4,382) and those who had not been taking aspirin (n = 5,628). Of these, 4,998 were assigned to receive low-dose perioperative aspirin and 5,012 to receive placebo. In patients with prior aspirin use, aspirin was stopped a median of 7 days before surgery. 

No Protection, Increased Bleeding

The primary outcome of death or nonfatal MI at 30 days was similar for the aspirin and placebo groups, as were secondary outcomes (table 1). Use of aspirin also showed no affect on a variety of tertiary outcomes.

Table 1. Primary and Secondary Outcomes



(n = 4998)

(n = 5,012)

HR (95% CI)

P Value

Primary Composite Outcome



0.99 (0.86-1.15)


Secondary Outcomes

Death, Nonfatal MI, Stroke
Death nonfatal MI, Cardiac Revascularization, Nonfatal Pulmonary Embolism, or Nonfatal DVT








0.98 (0.85-1.13)

 0.99 (0.86-1.14)




In terms of safety, aspirin increased the risk of major bleeding compared with placebo (4.6% vs 3.8%; P = 0.04), with the most common regions of bleeding being surgical site (78.3%) and the gastrointestinal tract (9.3%). Additionally, major or life-threatening bleeding was an independent predictor of MI (HR 1.82; 95% CI 1.40-2.36; P < 0.001).

The rate of stroke was 0.3% in the aspirin group and 0.4% in the placebo group (P = 0.62), and the median length of hospital stay was 4 days in both groups (P = 0.79). There was no significant difference in length of stay in intensive care or cardiac care units and no significant effect of clonidine on the results between aspirin and placebo.

Overall, the effect of aspirin on the primary composite outcome was consistent regardless of whether patients did or did not have prior aspirin use.

Analysis of the timing of aspirin administration demonstrated an absolute increase in the risk of a composite bleeding outcome associated with aspirin of 1.2% from the day of surgery to 30 days (P = 0.01) and 0.9% from day 4 after surgery to 30 days (P < 0.001). If a patient survived without the composite bleeding outcome until day 8 after surgery, the increase in risk from day 8 to day 30 was 0.3% (P = 0.29).

Bleeding Data Require Clarification

Panelist Yochai Birnbaum, MD, of Baylor College of Medicine (Houston, TX), inquired about subgroup differences among the different types of surgery: orthopedic, general, urologic/gynecologic, vascular, thoracic, and other.

Dr. Devereaux said there will be a separate analysis published on risk across surgeries and added that “there is much more bleeding than what people appreciate, if you look for it.”

In an accompanying editorial, Prashant Vaishnava, MD, and Kim A. Eagle, MD, both of University of Michigan Health System and Medical School (Ann Arbor, MI), say it is likely that aspirin prevented some perioperative MIs through thrombus inhibition, “but this may have been at the expense of bleeding and other myocardial infarctions induced by a mismatch between the supply of and demand for oxygen.” They stress the importance of evaluating the temporal relationship between major bleeding and MI.

“Future progress in perioperative medicine may depend on the implementation of strategies that successfully address one pathophysiological mechanism of perioperative myocardial infarction without being limited by another,” they add.

Study Details

Mean patient age was 68.6 years; 52.8% of the patients were men, 32.7% had a history of vascular disease, and 4.3% had undergone previous coronary stenting. 


Source:Devereaux PJ, Mrkobrada M, Sessler DI, et al. Aspirin in patients undergoing noncardiac surgery. N Engl J Med. 2014;Epub ahead of print.


Related Story:

  • POISE-2 was funded through grants from the Canadian Institutes of Health Research, the Commonwealth Government of Australia’s National Health and Medical Research Council, and the Spanish Ministry of Health and Social Policy.
  • Dr. Devereaux reports receiving research support and/or grants from Abbott Diagnostics, Bayer, Boehringer Ingelheim, Covidien, Roche Diagnostics, and Stryker.