Serum ALT is an Independent Predictor for Polypectomy During Screening Colonoscopy

Background: Diabetes and its antecedent condition, metabolic syndrome, have been associated with an increased risk of development of colon polyps and colorectal carcinoma. Nonalcoholic fatty liver disease (NAFLD) frequently coexists with these conditions and is associated with elevation in serum alanine aminotransferase (ALT). Our aim was to test our apriori hypothesis that after adjusting for currently established predictors of colon polyps, elevated serum ALT will be associated with increased rate of polypectomy. Methods: We conducted retrospective analysis on ten Veterans affairs (VA) VISN representing half of all VA medical centers, obtained from Veterans Affairs Medical SAS Data sets and Decision Support System from 2002 to 2009. We included adult patients without diabetes mellitus between the ages of 45 and 65 receiving first VA documented outpatient screening colonoscopy using validated methods. We excluded patients with human immunodeficiency virus infection, and those with liver disease from hepatitis C or alcohol abuse as well as those with missing values for ALT. Baseline ALT was defined as the ALT level closest to date of colonoscopy and within 12 months of colonoscopy. We categorized ALT levels to 0 to <30 IU/L (referent), 30 to <40, 40 to <50, 50 to <60 and ≥ 60. Based on literature review, we identified the following as established predictors of polyp occurrence: age, male gender, tobacco use, obesity, and use of nonsteroidal anti-inflammatory drugs and conducted logistic regression analysis with these variables in addition to ALT. Results: A total of 43,533 patients met our study criteria, of this 18591 (42.7%) had polypectomy. On adjusted analysis, ALT remained a significant additional predictor of polypectomy (Table). When compared to patients with ALT < 30, we found a linear increase in the rates of polypectomy with adjusted Odds Ratio (OR) of 1.03, 1.13, 1.19 and 1.26 among patient with ALT of 30 to <40, 40 to <50, 50 to <60 and≥ 60 respectively. Conclusions: In this well powered national Veteran cohort, ALT was found to be an independent additional predictor of polypectomy. More importantly, we found a linear relationship between ALT levels and polypectomy. Utilizing ALT as a surrogate for NAFLD, our data provides further support for NAFLD being an independent risk factor for polyp occurrence.