Evaluation of the limits to accurate sweat loss prediction during prolonged exercise

Eur J Appl Physiol. 2007 Sep;101(2):215-24. doi: 10.1007/s00421-007-0492-x. Epub 2007 May 30.

Abstract

Sweat prediction equations are often used outside their boundaries to estimate fluid requirements and generate guidance. The limitations associated with these generalized predictions have not been characterized. The purposes of this study were to: (1) evaluate the accuracy of a widely used sweat prediction equation (SHAP) when widening it's boundaries to include cooler environments (2 h) and very prolonged exercise (8 h), (2) determine the independent impact of holding skin temperature constant (SHAP36), and (3) describe how adjustments for non-sweat losses (NSL) and clothing saturation dynamics affect prediction accuracy. Water balance was measured in 39 volunteers during 15 trials that included intermittent treadmill walking for 2 h (300-600 W, 15-30 degrees C; n=21) or 8 h (300-420 W, 20-40 degrees C; n=18). Equation accuracy was assessed by comparing actual and predicted sweating rates (211 observations) using least-squares regression. Mean and 95% confidence intervals for group differences were compared against a zone of indifference (+/-0.125 l/h). Sweating rate variance accounted for by SHAP and SHAP36 was always high (r2>0.70), while the standard error of the estimate was small and uniform around the line of best fit. SHAP errors were >0.125 l/h during 2 and 8 h of exercise. SHAP36 errors were <0.125 l/h for 2 h conditions but were higher at 8 h in three of the six warmest trials. Adjustments for NSL and clothing saturation dynamics help explain SHAP errors at 2 and 8 h, respectively. These results provide a basis for future development of accurate algorithms with broader utility.

Publication types

  • Clinical Trial
  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Adult
  • Algorithms*
  • Confidence Intervals
  • Exercise / physiology*
  • Female
  • Forecasting
  • Humans
  • Male
  • Models, Biological
  • Physical Endurance / physiology*
  • Predictive Value of Tests
  • Regression Analysis
  • Skin Temperature / physiology
  • Sweating / physiology*
  • Water-Electrolyte Balance / physiology