A Field-based Three-Compartment Model Derived from Ultrasonography and Bioimpedance for Estimating Body Composition Changes

PURPOSEThe purpose of this study was to assess the agreement of a field-based 3-compartment model (3CFIELD) as compared to a laboratory-based 3-compartment model (3CLAB) for tracking body composition changes over time. METHODSResistance-trained males completed a supervised nutrition and resistance training intervention. Before and after the intervention, assessments were performed via air displacement plethysmography (ADP), bioimpedance spectroscopy (BIS), portable ultrasonography (US), and bioelectrical impedance analysis (BIA). ADP body density and BIS body water were utilized within the reference 3CLAB, while US-derived body density and BIA body water were utilized within 3CFIELD. Two-compartment model body composition estimates provided by US and BIA were also examined. Changes in fat-free mass (FFM) and fat mass (FM) were analyzed using ANOVA with repeated measures, equivalence testing, Bland-Altman analysis, linear regression, and related validity analyses. RESULTSSignificant increases in FFM (3CLAB4.0±4.5 kg; 3CFIELD3.9±4.2 kg; US3.2±4.3 kg; BIA3.9±4.2 kg) and FM (3CLAB1.3±2.2 kg; 3CFIELD1.4±2.2 kg; US2.1±2.6 kg; BIA1.4±2.9 kg) were detected by all methods. However, only 3CFIELD demonstrated equivalence with 3CLAB. Additionally, 3CFIELD exhibited superior performance to US and BIA individually, as indicated by the total error (3CFIELD1.0 kg; US1.8 kg; BIA1.6 kg), 95% limits of agreement (3CFIELD±2.1 kg; US±3.3 kg; BIA±3.1 kg), correlation coefficients (3CFIELD0.79-0.82; US0.49-0.55; BIA0.61-0.72), and additional metrics. CONCLUSIONSThe present study demonstrated the potential usefulness of a field-based 3C model incorporating US and BIA data for tracking body composition changes over time, as well as its superiority to US or BIA individually. As such, this accessible multi-compartment model may be suitable for implementation in field or limited-resource settings.