Measured and predicted resting metabolic rate in patients with inflammatory bowel disease

Objective: The present study aimed to compare measured and estimated resting metabolic rate (RMR) predicted by selected equations in patients with nonactive inflammatory bowel disease (IBD) on an outpatient university clinic regimen. Research Methods & Procedures: Seventy-two adult (>= 20 years) IBD patients (45 with Crohn's disease-CD) had RMR measured (mRMR) m RMR) by indirect calorimetry and also estimated by predictive equations (Cunningham, Henry, Anjos et al., and Marra et al.). Body composition was assessed by DXA. Absolute Bias (estimated- m RMR) and % Bias (Bias/mRMR) m RMR) were calculated. Agreement was assessed as the limit of agreement (LoA) in the Bland & Altman approach. Results: There was no difference in age, body composition and m RMR between individuals with CD (5414.2 f 1023.7 kJ/day) and ulcerative colitis (5443.9 f 1008.9 kJ/day). Among the equations, only the Anjos et al.'s population-specific equation (-52.1 [642.0] kJ/day, P = 0.493; LoA:-1311; 1206 kJ/d) accurately estimated RMR. The equations of Marra et al. produced the highest % Bias (24.1 f 14.8%). The Bland & Altman plots showed that the range of the LoA was relatively similar for all equations. In the simple regression analysis, the model with FFM resulted in a higher coefficient of determination (R2 2 = 0.51 for DC 0.74 for UC) compared to the model that included BM (R2 2 = 0.35 for DC and 0.65 for UC). Conclusions: Among the equations analyzed, only Anjos et al.'s accurately estimated RMR in outpatients with nonactive IBD. However, caution is advised when applying it at the individual level, due to the wide observed LoA. (c) 2024 Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.