Colorectal polyp prevention by daily aspirin use is abrogated among active smokers.

Abstract

Based on suggestive findings from a recent study of high-risk Japanese patients, we sought to determine whether the risk of colorectal polyps associated with smoking may be modified by daily use of aspirin in an analysis of a large US screening population. This is a cross-sectional study of 2,918 consecutive colonoscopy patients at a university hospital over a 30-month period. Data were abstracted from electronic medical records. Multivariate models of polyp counts were used to examine the competing risks of smoking and aspirin use. Models were further stratified by polyp location (proximal vs. distal) and pathologic subtype (dysplastic vs. serrated). Incidental rate of polyps was higher among active smokers [incidence rate ratio (IRR) 1.72; 95 % confidence interval (CI) 1.46-2.02] and lower among daily aspirin users (IRR 0.73; 95 % CI 0.61-0.86) compared to those who used neither. Smoking interacts significantly with aspirin use resulting in loss of aspirin protection (IRR 1.69; 95 % CI 1.28-2.24). Stratified analyses demonstrate that aspirin specifically reduces the risk of traditional dysplastic adenomas (IRR 0.72; 95 % CI 0.61-0.86) not serrated/hyperplastic polyps (IRR 0.92; 95 % CI 0.72-1.17) and that the modification of aspirin protection by smoking is primarily observed within the distal colorectum (p < 0.03). We report for the first time, in a typical risk US clinical population, a lack of protective association of aspirin for polyps among active smokers. Future prospective studies are recommended to confirm this mitigating effect in order to improve the precision of the growing evidence base about the chemopreventive benefit of aspirin in colorectal cancer.