Tu1409 Body Mass Index (BMI) is Associated With Gastrointestinal Symptoms in a Racially and Ethnically Diverse IBS Patient Population

Abstract

Background: Obesity is an increasingly prevalent medical problem that has been linked to variousmedical disorders, including gastroesophageal reflux disease andmetabolic syndrome. Obesity and increased body mass index (BMI) have also been associated with upper and/ or lower gastrointestinal (GI) symptoms in community or managed care populations, which have been primarily Caucasian. There is limited data on whether BMI is associated with irritable bowel syndrome (IBS) and if the association is affected by race. Aims: 1) To compare BMI between IBS patients and healthy controls (HCs) while controlling for demographic variables: age, gender, race/ethnicity and education; 2) To determine if BMI is associated with bowel habit subtypes, GI and somatic symptom ratings and quality of life (QOL) in IBS. Methods: Rome III positive IBS patients and HCs were recruited primarily from advertisement in a racially and ethnically diverse metropolitan community. Subjects completed questionnaires regarding demographics, GI and psychological symptoms including GI symptom-specific anxiety (Visceral Sensitivity Index [VSI]), QOL (SF-12), and non-GI somatic symptom severity (PHQ-12). T tests, Spearman's correlation, Kruskal-Wallis and chi-squared tests were used for bivariate comparisons and linear regression was used to control for demographic variables. Significance was assessed at 0.05. Results: A total of 871 subjects (374 IBS) participated in the study. Demographic and BMI data are shown in the Table 1. Rome III bowel habit subtypes in the IBS patients were: 23% IBS-C, 22% IBS-D and 55% IBS-M. Median BMI significantly differed by race/ethnicity AA: 26.7, Hispanic: 24.7, White: 23.7, Asian: 21.2, Other: 23.9 (p<0.001). Mean BMI was significantly higher in IBS than in HCs (P=0.022), but this difference became non-significant after controlling for demographics (P=0.715). Within the IBS group, BMI significantly differed by bowel habit subtype (P=0.010). After controlling for demographics, BMI in IBS-M was significantly higher than IBS-C with a 2.04 unit increase in BMI in IBS-M relative to IBS-C (P=0.005). In IBS, BMI negatively correlated with physical composite score of QOL (r=-0.27, P<0.001) and positively correlated with abdominal pain ratings (r=0.15, P=0.004) and VSI (r=0.12, P=0.027) but did not correlate with non-GI somatic symptom severity. Significance was maintained after controlling for demographics. Conclusion: BMI differences related to race/ ethnicity can account for BMI differences between IBS and HCs. However in IBS, BMI is significantly associated with bowel habit predominance, abdominal pain, and GI symptomspecific anxiety even after controlling for demographic variables. BMI is not associated with non-GI somatic symptom severity suggesting that BMI has an independent relationship to GI symptoms that is not related to presence of comorbidities. Demographics and BMI of Study Subjects