Relação da análise de impedância bioelétrica multifrequência com componentes da massa livre de gordura e a geometria de quadril-fêmur de atletas adolescentes de futebol

Abstract

Monitoring body composition in athletes is crucial for exercise and dietary prescription, particularly among adolescent soccer players—the focus of this doctoral thesis. Bioelectrical impedance analysis (BIA) has been used as a method to estimate body composition components, especially fat-free mass (FFM) and appendicular lean soft tissue (ALST), primarily through predictive equations developed using single-frequency analyzers (SF-BIA, 50 kHz). This has limited the use of multi-frequency analyzers (MF-BIA). Objectives: (1) to develop and validate predictive equations for FFM and ALST using raw bioelectrical data from MF-BIA, with dual-energy X-ray absorptiometry (DXA) as the reference method; and (2) to investigate the relationships between raw bioelectrical parameters—resistance (R), reactance (Xc), and phase angle (PhA)—and bone variables, such as bone mineral content (BMC), bone mineral density (BMD), and hip-femur bone geometry, including cross-sectional area (CSA), cross-sectional moment of inertia (CSMI), and section modulus (Z). Whole-body and segmental composition of the participants (n = 149 male soccer players, 15.6 ± 0.6 years) were assessed using DXA. For comparison purposes, bioelectrical data were also obtained using SF-BIA. Bioelectrical impedance vector analysis (BIVA) was conducted with MF-BIA data. In a subsample (n = 59), bone geometry data were also collected. The results are presented in four articles: (1) development and validation of a predictive equation for FFM using MF-BIA raw data, showing high accuracy when compared to DXA (R² = 0.95, SEE = 1.76 kg, CCC = 0.95); (2) development and validation of equations to estimate appendicular ALST (R² = 0.93, SEE = 1.04 kg, CCC = 0.96) and a specific equation for lower limbs (R² = 0.92, SEE = 0.81 kg, CCC = 0.96), both showing excellent agreement with DXA; (3) analysis of the relationship between BMC and raw bioelectrical data obtained with MF-BIA at 5, 50, and 500 kHz, for the whole body, trunk, and limbs. Whole-body (R² = 0.522) and lower-limb (R² = 0.349) BMC were best predicted at 5 kHz, and trunk BMC (R² = 0.301) at 50 kHz. Higher BMC values were associated with leftward BIVA vector shifts and higher PhA in the upper quartiles of BMC; (4) examination of the relationship between raw bioelectrical parameters, BMD, and hip-femur geometry using both SF-BIA and MF-BIA. MF-BIA (50 kHz) was more effective than SF-BIA, particularly due to stronger correlations with hip geometry and BMD. Higher BMD and CSA values were associated with lower R and higher PhA values. BIVA ellipses differed significantly according to hip-femur BMD (p = 0.0018), femoral neck BMD (p = 0.026), and CSA, independent of ALST (p = 0.024). The use of BIA to estimate BMC is discouraged, especially when bone mass is required for diagnostic purposes, since BIA cannot distinguish ALST from bone tissue.