Development and validation of a phase-sensitive bioelectrical equation for estimating skeletal muscle mass using DXA as reference

Background and aim: The development of appropriate bioimpedance-based equations for assessing body composition requires high-standard reference methods. Conversion factors have been proposed to derive skeletal muscle mass (SMM) from dual-energy x-ray absorptiometry (DXA) outputs. The objective of this study was to present a predictive equation for estimating SMM starting from bioelectrical impedance analysis (BIA) and using DXA as criterion. Methods and results: A total of 228 men and 125 women, aged 18–82 years underwent foot-to-hand BIA at a single frequency, alongside DXA scanning. To develop the new equation, bioelectrical parameters were recorded for 2/3 of the total sample and computed in a stepwise regression analysis. Agreement analyses were conducted on the remaining 1/3 of the sample assessing the performance of the new and a previously published equation (Janssen). The best developed model was: SMM (kg) = −17.612 + (sex × 3.397) + (stature2/reactance × 0.037) + (phase angle × 3.509) – (age × 0.052) with a coefficient of determination (R2) of 0.88 and standard error of estimation of 2.25 kg, where sex is 0 for women and 1 for men. Agreement analysis revealed no significant mean bias (−0.01) for the new model, whereas the Janssen equation overestimated SMM (+3.65 kg), with 95 % limits of agreement of −4.2 to 4.2 and −0.1 to 8.0, respectively. Lin's concordance correlation coefficient indicated higher accuracy for the new equations compared to the Janssen equation. Conclusions: This study presents a novel BIA-based predictive equation for assessing SMM in male and female individuals, based on sex, reactance, phase angle, and age.