Effects of hydration changes on bioelectrical impedance in endurance trained individuals. Med. Sci. Sports Exerc., Vol. 30, No. 6, pp. 885-892, 1998.
Purpose: The purpose of this study was to determine how differences in hydration states and ion content of hydrating fluids affected bioelectrical impedance (BI) and hydrostatic weighing (HW) measurements.
Methods: Fifteen athletic subjects aged 19-56 yr were recruited. Relative body fat (%), fat-weight (FW), and fat-free weight (FFW) were assessed using BI and HW under normal conditions (N), hypohydration (HPO), rehydration (RHY), and superhydration (SHY) states. During the RHY and SHY trial periods, subjects were hydrated with either distilled water or an electrolyte solution (ELS). HPO and SHY levels were set at 3% of each person's normally hydrated body weight.
Results: Comparison between the distilled water and the ELS trials indicated that hydration solution had no effect on BI or HW. Thus, the results presented are the trial means of both hydration solutions combined. Both BI and HW were shown to be highly test-retest reliable (r-values: 0.96 and 0.99, respectively). The effects of exercise induced HPO followed by RHY on body composition values indicated that HW was very stable across measurement periods while BI was not. From N to the HPO state, BI %BF declined from 14.4 ± 5.3% to 12.3 ± 5.3%, respectively. After RHY, BIA %BF increased to 15.5 ± 5.8%. Similar findings occurred when subjects were superhydrated (N-BI = 13.2 ± 5.3%; SHY-BI = 15.4 ± 5.6%). With a comparison of the intercepts and slopes of HW and BIA for the N and SHY states, it was clear hydration status significantly affected the intercepts (HW: 0.37 vs. BI: 1.85) and not the slopes (HW: 1.00 vs BI: 0.99). As a result, a majority of all fluid changes were interpreted as FW by BI. During HPO, 82% of the weight loss was considered FW while during RHY or SHY, 128% and 85% of the water weight regain/gain was considered FW.
Conclusion: These results indicate that BI is not a valid technique in athletes, especially when wanting to determine body composition effects of training/detraining. This study indicates that even small fluid changes such as those that occur with endurance training may be interpreted incorrectly as changes in an athlete's body fat content.
Department of Physical Education, Exercise and Sport Sciences, East Tennessee State University, Johnson City, TN 37614
Submitted for publication February 1997.
Accepted for publication October 1997.
The authors would like to thank Bill Bartoli and the Gatorade Sport Science Institute for donating the Gatorade required for this study, and Kenyan Clark, President of Valhalla Medical Products, for donating the 1990B Bioelectrical Impedance device. In addition, we would like to acknowledge the assistance of Daryl Conant, Eric Lougas, Dave Paul, and Kristin Seigfried on this project.
Address for correspondence: Craig E. Broeder, Ph.D., East Tennessee State University, Department of Physical Education, Exercise and Sport Sciences, Johnson City, TN 37614.