Skip navigation

Aspirin and colorectal adenoma


Clinical bottom line

Aspirin 325 mg a day delayed the appearance of new adenomas in patients with previous colorectal cancer.


RS Sandler et al. A randomized trial of aspirin to prevent colorectal adenomas in patients with previous colorectal cancer. New England Journal of Medicine 2003 348: 883-890.


Patients between the ages of 30 and 80 years with colon or rectal cancer with a low risk of recurrence were recruited. Those with Dukes' stage A or B1 who had undergone curative resection were immediately eligible, and those with B2 or C if they were free of the disease after five years.

Exclusions were familial polyposis, invasive cancer within five years, had cardiovascular disease, possible contraindications to aspirin use, a high likelihood of needing NSAIDs, plus other sensible exclusions. Eligible patients underwent a three month aspirin run in period to assess adherence to therapy and toxicity. To be eligible for the trial patients must have taken at least five tablets a week. They were then randomly assigned to have 325 mg aspirin or identical placebo a day, with tablets sent from a central pharmacy.

The primary endpoint was detection of adenomas in the large bowel by colonoscopy or sigmoidoscopy after randomisation. Analysis was based on patients randomised and who had at least one bowel investigation after randomisation.


Of 635 patients randomised, 10 were subsequently found to be ineligible, eight withdrew and two never started treatment. patients were well matched at baseline. Most were over 50 years, and 60% were Dukes' stage A or B1. Most (81%) had at least one colonoscopic examination after randomisation. The median time to first colonoscopy was 11 months for the placebo group and 16 months for the aspirin group. Mean follow up was 31 months.

There were 10% fewer patients with no adenomas in the aspirin group than in the placebo group (Figure 1). This statistically significant difference yielded a number needed to treat of 10 (6-29) for treatment with aspirin 325 mg a day for an average of 31 months to prevent the occurrence of at least one adenoma.

Figure 1: Number of polyps detected at follow up colonoscopy

During the run in and treatment period there were a number of severe adverse events recorded, but no obvious difference between aspirin and placebo for these, or for all cause deaths.


Aspirin 325 mg a day delayed the appearance of new adenomas in patients with previous colorectal cancer. Over an average of 31 months 10 people would need to be treated for one to not have had an adenoma detected, compared with placebo. Aspirin seemed to delay appearance of adenomas by about six months.