Impact of Adiposity Status on the Association of Regular Statin Use and Increased Number of Aberrant Crypt Foci

INTRODUCTION: While animal studies consistently show a preventive role of statin use in colon cancer development, evidence from human studies is conflicting with recent reports suggesting an increased risk of aberrant crypt foci and adenoma. We hypothesize that insufficient control of confounding due to indication bias for prescribing of statins, particularly obesity-induced physiologic dysregulation, might partly explain discrepant results. METHODS: We analyzed data from patients receiving standard colonoscopy at the Colon Cancer Prevention Program of the University of Connecticut Health Center, which was followed by chromoendoscopy for ACF detection. ACF number was categorized at low (RESULTS: Participants who reported regular use of statins had a higher mean ACF number than participants who reported taking (17.20 vs. 9.02, respectively, p=0.001). The mean age for the high ACF group was greater than the mean age for the low ACF group (59.40 versus 54.17 years, respectively, p=0.002). Never and past smokers were more likely to have low ACF than current smokers (58.9% vs. 60.0% vs. 18.2%, respectively, p=0.037). Univariate logistic regression showed that patients who regularly took statins were 3.9 (95%CI=1.43-10.74) times more likely to have a high ACF count than those not taking statins regularly, which was reduced to 2.67 (95%CI=0.89-7.97) in the age-adjusted analysis. In the basic multivariate model, the OR for high ACF among regular statin users was 1.47 (95%CI=0.28-7.74) in comparison to patients who consumed 1 or fewer pills per week. When adiposity measures were added into the basic multivariate model, the ORs for high ACF were: 1.24 for BMI (95%CI=0.21-7.47; p=0.017); 0.68 for WHR (95%CI=0.10-5.40; p=0.702); and 1.20 for WC (95%CI=0.20-1.20; p=0.842). Compared to taking neither aspirin nor statins (referent group), univariate ORs were: 15.0 (95%CI=1.55-145.23) for taking statins only on a regular basis; 1.83 (95%CI=0.59-5.68) for regular use of aspirin only; and 4.29 (95%CI=1.24-14.83) for regular use of both statins and aspirin. OR estimates were slightly attenuated in age-adjusted analyses and were substantially reduced in multivariate analyses, and no longer statistically significant. CONCLUSIONS: While regular statin use in our study population was associated with a statistically significant higher ACF count in univariate analyses, this effect did not remain after controlling for key risk factors for colon cancer. It is possible that prior evidence of an adverse role of statin use in colorectal neoplasia in human studies may be explained in part by confounding. Our results must be interpreted with caution due to small numbers in study groups