Total and segmental body composition examination in collegiate football players using multifrequency bioelectrical impedance analysis and dual x-ray absorptiometry

The current study examined the influence of player position on the agreement between multifrequency bioelectrical impedance analysis (MfBIA) and dual X-ray absorptiometry (DXA) when assessing total and segmental percent body fat (BF%), fat mass (FM), and fat-free mass (FFM) in National Collegiate Athletic Association Division I collegiate football athletes. Forty-four male collegiate athletes (age = 19 ± 1 year; height = 1.9 ± 1.0 m; and body mass = 106.4 ± 18.8 kg) participated. Player positions included: offensive linemen (OL; n = 7), tight ends (TE; n = 4), wide receivers (WR; n = 9), defensive linemen (DL; n = 6), defensive backs (DB; n = 8), linebackers (LB; n = 6), and running backs (RB; n = 4). Total and segmental body composition measured using MfBIA were compared with values obtained using DXA. Compared with DXA, MfBIA underestimated BF% (3.0 ± 3.8%), total FM (2.5 ± 4.3 kg), arm FM (0.4 ± 0.8 kg), arm FFM (1.4 ± 0.9 kg), leg FM (2.8 ± 2.0 kg), and leg FFM (5.4 ± 2.4 kg) (all p < 0.001; arm FM p = 0.002) and overestimated total FFM (-2.4 ± 4.5 kg) (p < 0.001). Limits of agreement (LOAs) were: ±7.39% (BF%), ±8.50 kg (total FM), ±1.50 kg (arm FM), ±1.83 kg (arm FFM), ±3.83 kg (leg FM), ±4.62 kg (leg FFM), and ±8.83 kg (total FFM). No significant differences were observed between devices for trunk FM (-0.3 ± 3.0 kg; p = 0.565) and trunk FFM (0.4 ± 2.4 kg; p = 0.278), with LOAs of ±5.92 and ±4.69 kg, respectively. Player position significantly affected all between-device mean body composition measurement differences (adjusted p ≤ 0.05), with OL demonstrating the greatest effect on each variable. Therefore, MfBIA does not seem accurate in examining between-player body composition in college football players.