Nonsteroidal anti-inflammatory drug use after 3 years of aspirin use and colorectal adenoma risk: observational follow-up of a randomized study

Abstract

Background: Frequent use of nonsteroidal anti-inflammatory drugs (NSAIDs) has been shown to reduce the risk of colorectal adenomas in randomized trials. We examined the persistence of the protective effect after the cessation of randomized aspirin treatment and whether it is affected by the duration and frequency of subsequent NSAID use.

Methods: We used data from the Aspirin/Folate Polyp Prevention Study (AFPPS), in which 1121 subjects were randomly assigned to receive placebo or aspirin (81 or 325 mg/d) for 3 years. After the end of treatment and a follow-up colonoscopy, AFPPS participants were invited to remain under follow-up until their next surveillance colonoscopies, scheduled 3-5 years later. Information regarding use of NSAIDs during posttreatment follow-up was gathered periodically via questionnaires. Average weekly NSAID use was classified as sporadic (<2 days per week), moderate (2 to <4 days per week), or frequent (>or=4 days per week). The analysis was stratified according to randomized aspirin groups and posttreatment NSAID use; placebo subjects who later were sporadic NSAID users formed the reference group. The primary outcomes were all adenomas and advanced lesions. Adjusted relative risks and 95% confidence intervals were computed with generalized linear models. All statistical tests were two-sided.

Results: A total of 850 subjects underwent a posttreatment colonoscopy, on average 4 years after the end of study treatment. The protective effect of 81 mg of aspirin for colorectal adenomas persisted with continued posttreatment NSAID use. The risk of any adenoma among frequent NSAID users was 26.8% vs 39.9% among placebo subjects who later used NSAIDs sporadically (adjusted relative risk = 0.62, 95% confidence interval [CI] = 0.39 to 0.98; P(trend) with NSAID use frequency = .03). The unadjusted absolute risk reduction was 13.1 percentage points (95% CI = -0.3 to 26.5 percentage points) (P = .07). Results for 325 mg of aspirin were similar, although not statistically significant. For advanced lesions, small numbers of endpoints limited the analysis, but findings among subjects randomly assigned to 81 mg of aspirin suggested a protective association regardless of posttreatment NSAID use.

Conclusion: Long-term and frequent use of NSAIDs may enhance the chemopreventive effect of aspirin against colorectal neoplasia.

Figure 1

Analytic design. We focused on two exposures: ASA treatment assignment during the randomized study period (subjects were randomly assigned to placebo, 81 mg of ASA per day [ASA 81], or 325 mg of ASA per day [ASA 325]) and self-reported use of NSAIDs (<2, 2–4, or ≥4 days per week) during the posttreatment F/U period. The primary outcome was defined as the occurrence of adenoma in the first posttreatment F/U colonoscopy. *Results have been published (1). ASA = aspirin; F/U = follow-up; NSAIDs = nonsteroidal anti-inflammatory drugs.

Figure 2

Study flow diagram for the randomized study period and the posttreatment observational follow-up period. *Other reasons for exclusion before randomization: one died, 47 ineligible for reasons related to the folate component (eg, anemia), 28 concurrent illnesses, 26 declined to continue, and 17 other reasons. ASA = aspirin; NSAID = nonsteroidal anti-inflammatory drug.