Abstract des Autors

Several authors have described the relationship between body composition and sport performance, and underlined that both body weight and composition are related to success in sport. Body composition analysis may help to assess individuals' potential for result in a given sport and allows to develop recommendation to the athletes on the amount of weight which can be lost without sacrificing important portion of fat-free mass; for these reasons the evaluation of body composition is now widely practised in sport. The number of methods for assessment of human body composition, that is the determination of fat mass (FM) and fat-free mass (FFM), has expanded significantly in recent years. Among the techniques that have been developed for field use there are: Bioelectrical Impedance Analysis (BIA), Near-infrared Interactance (NIR) and anthropometric methods. BIA involves the measurements of bioelectrical resistive impedance for the determination of body composition, and it is based upon the principle that the electrical conducivity of the fat-free mass is far greater than that of fat mass; NIR applies the principles of light absorption, reflection, and near-infrared spectroscopy, to estimate percentage of body fat. The anthropometric methods, based generally on measurements of skinfold thickness, are probably the most widely used field methods of assessing body composition, but are usually specific to the population which they are derived from and, must be used with caution to estimate body fat in athletes, owing to methodological and biological factors. The aim of this study is to verify the adaptability of some methods for the determination of body composition in a sample of male athletes. Bioelectrical Impedance Analysis, an accurate technique recently developed, is utilized as reference method and compared with Near-infrared Interactance and five different athropometric methods, represented by the equations of Jackson and Pollock, Lohman, Norgan and Ferro-Luzzi, Sloan and Durnin and Womersley. Bioelectrical impedance analysis results highly correlated with near-infrared interactance (r=0.79; p<0.001) and anthropometric methods (r=0.55-0.89; p<0.001). Nevertheless, the average fat mass determined from BIA (FM-BIA) differs significantly from the value determined using NIR (p<0.01) and the anthropometric equations of Jackson and Pollock (p<0.001), Lohman (p<0.001), Norgan and Ferro-Luzzi (p<0.001); instead fat mass derived from Sloan and Durnin and Womersley's equations don't differ significantly from FM-BIA. These results indicate that in male athletes analized, the anthropometric equations of Sloan and Durnin and Womersly can be utilized alternatively to the more accurate but expensive BIA method. Verf.-Referat