OBJECTIVE: To evaluate the applicability and validity of bioelectrical impedance analysis (BIA) compared with dual-energy X-ray absorptiometry (DXA) in a population of girls with restrictive anorexia nervosa (AN). METHODS: A total of 30 girls (11-19 years old) with AN were enrolled. DXA and BIA (BIA software and the Deurenberg equations) were used to estimate the body composition. The correlation between the methods was assessed by Pearson's correlation coefficient and the Bland-Altman method. RESULTS: The mean FFM estimates were 33.2 kg (BIA software), 32.8 kg (BIA, Deurenberg equation) and 33.1 kg (DXA). The mean FM values were 5.6 kg (BIA software), 6.2 kg (BIA, Deurenberg equation) and 6.4 kg (DXA). There was a high correlation between the FFM values estimated with the two methods (BIA software vs DXA r=0.917, p<0.001; Deurenberg equation vs DXA r=0.931, p<0.001). For the FFM, the limits of agreement were equal to ±3.34 kg for the BIA software and ±2.96 kg for the Deurenberg equation. For the FM, the limits of agreement were equal to ±4.60 kg for the BIA software and ±3.82 kg for the Deurenberg equation. CONCLUSION: The results show a good correlation between DXA and BIA. BIA seems to be a valid alternative for epidemiological and clinical evaluations.
OBJECTIVE: To evaluate the applicability and validity of bioelectrical impedance analysis (BIA) compared with dual-energy X-ray absorptiometry (DXA) in a population of girls with restrictive anorexia nervosa (AN). METHODS: A total of 30 girls (11-19 years old) with AN were enrolled. DXA and BIA (BIA software and the Deurenberg equations) were used to estimate the body composition. The correlation between the methods was assessed by Pearson's correlation coefficient and the Bland-Altman method. RESULTS: The mean FFM estimates were 33.2 kg (BIA software), 32.8 kg (BIA, Deurenberg equation) and 33.1 kg (DXA). The mean FM values were 5.6 kg (BIA software), 6.2 kg (BIA, Deurenberg equation) and 6.4 kg (DXA). There was a high correlation between the FFM values estimated with the two methods (BIA software vs DXA r=0.917, p<0.001; Deurenberg equation vs DXA r=0.931, p<0.001). For the FFM, the limits of agreement were equal to ±3.34 kg for the BIA software and ±2.96 kg for the Deurenberg equation. For the FM, the limits of agreement were equal to ±4.60 kg for the BIA software and ±3.82 kg for the Deurenberg equation. CONCLUSION: The results show a good correlation between DXA and BIA. BIA seems to be a valid alternative for epidemiological and clinical evaluations.
Authors: E Cacciari; S Milani; A Balsamo; E Spada; G Bona; L Cavallo; F Cerutti; L Gargantini; N Greggio; G Tonini; A Cicognani Journal: J Endocrinol Invest Date: 2006 Jul-Aug Impact factor: 4.256
Authors: M Marra; A Caldara; C Montagnese; E De Filippo; F Pasanisi; F Contaldo; L Scalfi Journal: Eur J Clin Nutr Date: 2008-11-12 Impact factor: 4.016