Insights into body composition adaptation: Should we reconsider the use of Body Mass Index in some sports?

Abstract

BACKGROUND: The purposes of this study were to indentify the under/overweight/obese frequencies by Body Mass Index (BMI) and body fat percentage (BF%) in athletes within groups of sport and to investigate the accuracy of the BMI as a measure of BF%. METHODS: Cross-sectional design study on elite male athletes (N.=2234, aged 22±4 years) from 51 sports disciplines who were classified according to two different sport classifications: predominant characteristic of training (four group model) and type and intensity of exercise (nine group model). All athletes underwent full anthropometric testing. RESULTS: After stratification, the majority of athletes were in normal weight category. According to 4 group model, BMI is showed as statistically significant, reliable and independent predictor of BF% in all groups of sports. In nine groups model all correlated parameters were positive for athletes being statistically significant (P<0.001) with exception of group LSMD, MSMD and HSMD (P>0.05). The highest positive correlation between BMI and BF% was in group MSLD (r=0.53; P<0.001) and in power sports group (r=0.24; P<0.001). CONCLUSIONS: BMI could be an accurate predictor of BF% in athletes but that depends on group of sport. Our results suggest the BMI could use only in power and MSLD groups of sport. © 2016 Edizioni Minerva Medica.