Aspirin Continuation or Interruption in Patients at Moderate Risk for Cardiovascular Events Undergoing Colonoscopy and/or Polypectomy; a Placebo-controlled Trial

Overview

One in 4 adults between 50 and 80 reports taking regular aspirin. The prevalence of aspirin uses increases with age as well as co-morbid vascular diseases. Patients with cardiovascular diseases are at risk of developing colorectal neoplasms. In patients undergoing screening colonoscopy, interruption of aspirin is believed to be associated with increased cardiovascular events. Continuation of aspirin can however be associated with an increased risk of post-polypectomy bleeding. International guidelines on periendoscopy management recommend the continuation of aspirin based on evidence from cohort studies, mostly retrospective, suggesting that the rate of bleeding is low. Cardiovascular complications from aspirin interruption can lead to disabilities and occasional deaths. The cardiovascular risks following aspirin continuation or interruption in endoscopy have not been well studied. There has been no randomized study to compare either strategy. Endoscopists are divided on their opinion on whether to stop or to continue aspirin. The proposed large randomized controlled trial (RCT) is powered to detect small differences in both outcomes. Findings from this RCT will address this important question and inform our clinical practice.

Aspirin is commonly prescribed to patients for protection against cardiovascular diseases. In 2019, the prevalence of aspirin use was 35.6% in the US. Patients with vascular diseases are at increased risk of having colorectal neoplasms and therefore often undergo screening colonoscopy and possible polypectomy. The management of aspirin before and after colonoscopy and polypectomy has been a contentious subject. The updated British Society of Gastroenterologists (BSG) and ESGE guidelines recommend continuing aspirin except for ESD and EMR \> 2cm. The recommendation is based on the low rate of bleeding complications in large cohort studies. In a BSG audit of 20085 colonoscopies in the UK, 52 (0.26%) cases of bleeding were reported. In the English National Bowel Cancer Screening Programme, 69,028 underwent polypectomy. The overall rate of PPB was 1.14% (4). In large series mostly retrospective (\> 1000 polypectomies), delayed PPB varied from 0.6 to 2.2 %, and the mean time to onset of bleeding was 4.0 ± 2.9 days (5). The use of aspirin did not emerge to be a factor predicting delayed bleeding. The evidence for continuing aspirin to reduce CV events is conflicting. In an early trial (6) on 220 subjects undergoing non-cardiac surgery, aspirin continuation was associated with a 7.2% risk reduction in postoperative major adverse cardiac events. In the STRATAGEM trial that enrolled 291 patients undergoing general surgery, no difference in thrombotic or bleeding events was observed between groups. The POISE-2 trial was considered the most definitive. It was a placebo-controlled trial that studied patients undergoing again noncardiac surgery, 351 of 4998 patients (7.0%) in the aspirin group and 355 of 5012 patients (7.1%) in the placebo group reached the primary outcome endpoint (death or non-fatal infarcts). Major bleeding was more common in the aspirin group (4.6% vs. 3.8%, P=0.04). Continuation of aspirin was associated with a higher risk of bleeding but did not protect against CV events. There exist variations in clinical practice. A survey of endoscopy centres in the U.S. showed that 44% of endoscopy units recommend continuing aspirin, 33% recommend stopping aspirin and 24% suggest that patients seek advice from a physician. Many endoscopists are concerned about post-polypectomy bleeding and its associated medico-legal risks. In the context of colonoscopy and polypectomy, there has been no randomized comparison between the two strategies of continuing or withholding aspirin. We therefore undertake an RCT to compare cardiovascular and bleeding outcomes in patients undergoing colonoscopy with/without polypectomy. We are planning a large RCT sufficiently powered to detect small differences in these outcomes.