Comparison of bioelectrical impedance analysis and dual-energy X-ray absorptiometry for the assessment of appendicular body composition in anorexic women

Objective: To establish the accuracy of bioelectrical impedance analysis (BIA) for the assessment of appendicular body composition in anorexic women. Design: Cross-sectional study. Setting: Outpatient University Clinic. Subjects: A total of 39 anorexic and 25 control women with a mean (s.d.) age of 21 (3)y. Methods: Total, arm and leg fat-free mass (FFM) were measured by dual-energy X-ray absorptiometry and predicted from total and segmental BIA at 50 kHz. The predictor variable was the resistance index (RI), that is, the ratio of height2 to body resistance for the whole body and the ratio of length2/limb resistance for the arm and leg. Results: Predictive equations developed on controls overestimated total, arm and leg FFM in anorexics (P<0.0001). Population-specific equations gave a satisfactory estimate of total and appendicular FFM in anorexics (P = NS) but had higher percent root mean square errors (RMSEs%) as compared to those developed on controls (8% vs 5% for whole body, 12% vs 10% for arm and 10% vs 8% for leg). The accuracy of the estimate of total and leg FFM in anorexics was improved by adding body weight (Wt) as a predictor with RI (RMSE% = 5% vs 8% and 7% vs 10%, respectively). However, the same accuracy was obtained using Wt alone, suggesting that in anorexics, BIA at 50 kHz is not superior to Wt for assessing total and leg FFM. Conclusion: BIA shows some potential for the assessment of appendicular body composition in anorexic women. However, Wt is preferable to BIA at 50 kHz on practical grounds. Further studies should consider whether frequencies >50 kHz give better estimates of appendicular composition in anorexics as compared to Wt.