METHODOLOGICAL CONSIDERATIONS OF BODY COMPOSITION ASSESSMENTS AND PREDICTING ATHLETIC PERFORMANCE: THE DA KINE STUDY

Abstract

Assessment of body composition provides clinically relevant data on fat and fat-free mass (FM, FFM) that deepens the knowledge of body mass alone, to enhance human performance and identify links between disease/cancer risk. Total body water (TBW), typically 0.73% of FFM is the body's largest compartment; therefore, errors in its measure can impact body composition estimates the greatest, especially in special populations like athletes who tend to be outside of normal hydration, leading to deleterious effects on performance. However, the accuracy and association of various body composition methods, including emerging and standard techniques remain unclear. The central hypothesis of this dissertation was that body composition measures that adjusted for hydration status were more accurate to muscle performance than those insensitive to hydration. The objective of this dissertation was to determine how to derive accurate and precise body composition in athletes with varying states of hydration by exploring specific TBW methodologies, and finally how these measures are associated with muscular strength. The Da Kine Study of 80(40male) athletes aged 21.8 (SD-4.2) years was conducted to achieve this goal by introducing three central study aims. Aim I compared the accuracy using Lin’s (CCC) and precision by root-mean-square coefficient of variation (RMS-CV%) of differing body composition models/methods to the criterion five-compartment model (5CM). Laboratory methods included: air-displacement plethysmography (ADP), deuterium (D2O) dilution [using Fourier-transform infrared spectrophotometry (FTIR), and isotope-ratio mass-spectrometry (IRMS)], dual-energy x-ray absorptiometry (DXA), underwater-weighing (UWW), and field-methods: three bioelectrical impedance (BIA) devices (S10/SFB7/SOZO), and 3-Dimensional Optical (3DO) scans. Leg and trunk strength was measured via dynamometry. Athletes' hydration status was significantly (p=0.001) outside of the normal range in both males (63-73%) and females (58-78%). While no significant mean differences were present between methods, significant individual accuracy and precision estimates were observed. ADP and DXA produced the highest agreement of FM and FFM to the 5CM (CCC=0.90-0.95) in both sexes. All measures of FFM produced excellent precision (<1%), whereas only ADP, DXA, D2O, and S10 had <2%. More accurate body composition (ADP/DXA) determination produced better associations with muscle strength. Aim II assessed the impact of accuracy and precision on TBW estimates using various specimen types (saliva/urine), analytical methods (FTIR/IRMS/BIA), and across three laboratories. These measurements were then substituted into both the 5CM and the 2-compartment model (2CM) for FM. TBW precision varied between techniques (0.3-1.2%) with D2O-IRMS-saliva as the criterion. All TBW measures produced substantial-perfect agreement with the criterion (CCC>0.90), except for SFB7/SOZO (CCC<0.90). FM from nearly all 2CM-D2O/BIA produced poor agreement, however, the 5CM-D2O produced substantial agreement and the BIA devices S10/SOZO had a moderate agreement. Aim III compared a novel approach to measure skeletal muscle mass by deuterated creatine (D3Cr) dilution and its relationship to muscle strength and similar measures using DXA. All measures of leg/trunk strength were more highly associated with DXA, with female trunk strength as the highest (R2=0.61, RMSE=19.4) when compared to D3Cr (R2=0.54, RMSE=17.4). Additionally, whole body values of body composition produced higher associations to muscle strength over normalized values to body weight or height. Concluding that accurate and precise estimates of body composition can be determined in athletes of varying states of hydration and suggest that a more accurate body composition produces better strength estimates. The study emphasizes the importance of using criterion modeling, specific TBW methodologies, and various devices such as DXA, 3DO, and BIA, to serve as reliable approaches for assessing body composition and enhancing human performance.