LAUSR

Objectives Greater serum levels and dietary intake of vitamin D have been inversely associated with the risk of multiple cancers including colon cancer. Most colorectal cancers are thought to arise from adenomatous polyps, which become dysplastic under the influence of numerous factors. Prospective data are needed to distinguish between association or a causative role of vitamin D in the pathogenesis of colorectal cancer. Methods A prospective cohort study was designed, located at a hospital-based screening colonoscopy referral center, including Department of Defense beneficiaries aged 18 yr or older. A serum 25-hydroxyvitamin D level was drawn, and colonoscopy findings were recorded. A power calculation using p = 0.8, alpha = 0.05 generated a necessary sample size of n = 224 to detect an association between vitamin D level and adenomatous polyp. Unconditional multivariable logistic regression modeling was used to evaluate associations between outcomes, adjusted for known risk factors. Results Of final population (n) = 228, 155 (68%) were diagnosed with vitamin D insufficiency (<30 ng/mL) and 104 (46%) were found to have adenomatous polyps. Adjusted odds ratio for adenomatous colon polyp given vitamin D <30 ng/mL = 1.16 (95% CI 0.57-2.36). Conclusions Serum vitamin D level was not associated with increased adenomatous polyp detection at screening colonoscopy in this population.