BMI, Surgical Intervention and Methotrexate use impact Mucosal Disease in African Americans with Crohnʼs Disease

Abstract

To identify predictors of clinical and mucosal disease activity in African Americans (AA) with Crohn's Disease (CD). Single-center, retrospective study. We reviewed records of 99 AA patients with CD who were seen at the outpatient clinic between 2000 and 2009. Demographic, clinical and outcome variables were examined using descriptive statistics. These included body mass index (BMI), gender, age at diagnosis and last observation, duration, location and behavior of disease, perianal disease, surgical intervention, number of surgeries and time to first surgery, and smoking. Medication history was also recorded. BMI was based on the first observation visit with the following categories; 18.5-24.9 mg/kg2 (normal weight), 25.0-29.9mg/kg2 (overweight), equal to or > 30mg/kg2 (obese). Clinical outcome was based on the average American College of Gastroenterology (ACG) clinical severity score during last year of observation. Mucosal outcome was based on simplified endoscopic score for CD (SES-CD) during the same period. Bivariate analysis was completed using chi square or Fisher's exact test. Multivariable analysis was performed using step-wise logistic regression. Non-parametric correlation analysis between clinical and mucosal disease was also done. The level of significance was assessed at p<0.05. Mean ages at diagnosis and last observation were 31 and 42 years respectively. Mean duration of disease was 11 years. Mean observation period was 4 years (range1-9 years). Male: female ratio was 26 (26%): 73 (74%). Mean BMI was 27 mg/kg2. Thirty three (42%) patients were normal weight, 32 (32%) were overweight and 20 ((26%) were obese. Sixty (60%) patients had undergone surgery of which 71% had had two or greater surgeries. Mean time to first surgery was 4 years. Twenty eight (28%) patients were on a traditional immune modulator with the same number on a biologic. Five (5%) were on methotrexate (MTX). In regard to clinical outcome, 34 (34%) patients were in remission, 33 (33%) had mild disease and 32 (32%) had moderate to severe disease. In terms of mucosal disease, twenty four (24%) patients' disease was inactive, 29 (32%) had mild mucosal disease whereas 39 (42%) patients had moderate to severe mucosal disease. Bivariate analysis revealed that surgery was a statistically significant predictor of increased mucosal disease (P= 0.0314 OR 2.413 95%CI 1.082-5.385). Undergoing two or more surgeries was also a predictor of increased mucosal disease (P= 0.0353 OR 2.413 95%CI 1.089-10.796). Controlling for age at last observation, gender and duration of disease demonstrated that being obese vs. normal weight was a statistically significant predictor of increased mucosal disease activity (P= 0.0022 OR 8.325 95% CI 2.159-32.107). History of surgery remained statistically significant (P= 0.0127 OR 3.873 95% CI 1.336-11.231). Despite a small patient cohort on MTX, it was also a predictor of increased mucosal disease (P= 0.0151 OR= 6.526 05% CI 1.437-29.632).Non-parametric correlation between clinical and mucosal disease was positive and statistically significant (Correlation =0.3852 P= 0.0001) but we did not observe any statistically significant predictors of clinical disease activity. Higher BMI, history of surgical intervention and methotrexate use predict increased mucosal inflammation in AA patients with CD.