P406 Malnutrition during hospitalisation for acute severe IBD is associated with increased risk of relapse

Abstract

Background: Current data suggest that malnutrition during IBD flare predicts in-hospital mortality, surgery, and length of admission. However, specific effects of nutrition during IBD flare on resolution of inflammation are unknown. We hypothesised that nutritional factors during hospitalisation for acute severe IBD are associate with relapse of inflammation in the year following discharge. Methods: This retrospective study was a part of a care-quality project. Patients admitted to the Karolinska Hospital Gastroenterology ward with IBD flare from 01.01.2015 to 29.02.2016 were identified from hospital records and the National Swedish IBD Register (SWIBREG). Patients who underwent complete resection of inflamed tissue during the admission were excluded. Data on nutritional intake, disease factors, inflammatory markers, and daily calorie requirement (calculated during the admission: 30 kcal/kg, adjusted in case of extreme values of age, BMI or physical activity) were extracted. Relapse was defined as the following outcomes: requirement of new steroid prescription, intensification of biological therapy, readmission, surgery, calprotectin. Adjustments for age and gender were made, and for multiple comparisons the false discovery rate (FDR) control was used. Results: In total, 100 patients were included. Steroid prescriptions during the follow-up year was significantly associated with the number of days with calorie intake <30, <50, and <70% of calculated need (regression coefficients: 0.32, 0.24, and 0.18, showing a dose-response relationship; FDR control gave q-values 0.0760, 0.0738, 0.119, respectively). Overall, 23%, 50%, and 74% of patients had calorie intake <30, <50, and <70% of calculated need. Markers of inflammation severity (CRP, calprotectin and stool frequency on admission) were not associated with steroid-requiring relapse. Readmissions were significantly associated with a rise in weight/BMI during the inpatient period and with ongoing immunosuppressive treatment at admission (q-values: 0.000771, 0.0230, and 0.112, respectively). Other outcomes were not associated with any of the factors tested. Hence, six discoveries with significance at the 0.05 level with acceptance of 12% FDR were found. Conclusions: Days with suboptimal calorie intake during hospitalisation for IBD flare are associated with steroid-requiring relapse in the subsequent year. We speculate that suboptimal nutrition may adversely affect attainment of durable remission by impeding either epithelial cell regeneration or microbial restitution. The association between readmission and immunosuppressant use and weight gain during the admission may be, respectively, a marker of disease severity or be caused by low serum albumin leading to oedema.