Utility of bioimpedance methods for the assessment of fat-free mass in adult outpatients with inflammatory bowel disease.

Abstract

Clinically accessible body composition assessment is required to identify fat-free mass (FFM) depletion, which is common in inflammatory bowel disease (IBD) and poorly correlated with body mass index (BMI). The aim of this study was to compare FFM assessed using bioimpedance spectroscopy (BIS) and multifrequency bioelectrical impedance analysis (MFBIA) with dual energy x-ray absorptiometry (DXA) in adult outpatients with IBD. FFM was measured using DXA, BIS, and MFBIA in 40 outpatients with IBD. Paired t tests, Lin and Pearson's correlations, and limits of agreement (LOA) analysis were used to compare FFMDXA with FFMBIS and FFMMFBIA. Participants (26 men, 24 with Crohn's disease) had a median (IQR) age 39 y (32-50 y) and median (IQR) BMI 24.2 kg/m2 (21.9-26.4 kg/m2). Mean FFMDXA was 51.6 ± 12 kg. FFMBIS was highly correlated with FFMDXA (rp=0.97; P < 0.001); however, BIS significantly overestimated FFM compared with DXA by a mean 3.4 ± 2.6 kg (P < 0.001) and showed wide 95% LOA (-1.7 to 8.4 kg). FFMBIS estimations improved when FFM was adjusted for by BMI using Moissl's predictive algorithm, reducing mean bias to 0.1 ± 2 kg (P=0.858; 95% LOA -3.9 to 4 kg). The bias between FFMMFBIA and FFMDXA was 1.3 ± 2 kg (P=0.002) with 95% LOA -2.6 to 5 kg (n=28). Unadjusted BIS overestimated FFM in IBD outpatients compared with DXA with poor agreement at an individual level. Adjusting for BMI ameliorated the overestimation. It is suggested for the estimation of FFM in outpatients with IBD that MFBIA or the Moissl algorithm with BIS be used if DXA is unavailable.