Validity of bioelectrical impedance analysis to estimate body composition changes after bariatric surgery in premenopausal morbidly women.

Research paper by Silvia S Savastano, Annamaria A Belfiore, Carolina C Di Somma, Concetta C Mauriello, Annalisa A Rossi, Genoveffa G Pizza, Annalba A De Rosa, Giovanni G Prestieri, Luigi L Angrisani, Annamaria A Colao

Indexed on: 29 Oct '09Published on: 29 Oct '09Published in: Obesity Surgery


In obese patients, subtle variations of the hydration of soft tissues can propagate errors in bioelectrical impedance analysis (BIA) measures of body composition. Bioelectrical impedance vector analysis (BIVA) is a useful method to evaluate tissue hydration. Laparoscopic adjustable gastric banding (LAGB) is a purely restrictive bariatric surgical procedure resulting in lower fat-free mass (FFM) loss than other malabsorptive or mixed intervention. The aim of this study was to evaluate the 6- and 12-month changes in body composition in a homogeneous group of premenopausal morbidly obese women treated by LAGB by comparing the results of conventional BIA and BIVA with dual-energy X-ray absorptiometry (DXA) method. Forty-five consecutive morbidly obese patients (mean age, 35.3 +/- 9.1 years; body mass index, 34.5-48.7 kg/m(2)) were prospectively evaluated at the Endocrinology Unit of the Department of Molecular and Clinical Endocrinology and Oncology. The LAGB device (Lap-Band System; Inamed Health, Santa Barbara, CA, USA) was inserted laparoscopically. Soft tissue hydration was evaluated by BIVA; fat mass (FM) and FFM were evaluated by BIA (BIA 101 RJL, Akern Bioresearch, Firenze, Italy) and by DXA (Hologic QDR 4500A S/N 45622; Hologic Inc., Bedford, MA, USA). Pre- and postoperative BIVA vectors indicated a normal hydration in all patients. Postoperatively, the excess of body weight loss was mainly due to a decrease in FM. The regression analysis of BIA and DXA methods at baseline and at the 6- and 12-month follow-up for FM r (2) values were 0.98, 0.94, and 0.99, respectively (p < 0.001); FM% r (2) values were 0.91, 0.89, and 0.98, respectively (p < 0.001); and FFM r (2) values were 0.87, 0.82, 0.99, respectively (p < 0.001). BIA and DXA measurements of body composition exhibit a high relative agreement in the study group of normo-hydrated obese subjects. BIA tends to overestimate FFM, but this effect is reduced along with the weight loss during the follow-up. Under the stable hydration, the BIA method may be useful as an alternative to DXA in a selected clinical setting when repeated comparisons of body composition are required.