Comparison Between Actual and Predicted Sweat Rates in Burn Survivors Using a Predictive Model for Fluid Replacement

Sweating is a primary physiological response necessary to regulate body temperature in hot environments. Recent literature identifies a correlation between multiple variables (i.e., height and weight, environmental temperature, humidity, intensity and duration of exercise, and venue of exercise) and the rate of sweat loss during physical activity in healthy participants. These findings were then used to create a model to predict fluid loss via sweat, and thus the fluid replacement necessary to rehydrate during and following exercise (sweatratecalculator.com). Importantly, this calculator was modeled with data from uninjured, healthy individuals and therefore may not be accurate for burn survivors. Grafted skin does not produce sweat, leading to impaired thermoregulation, resulting in higher core temperatures during exercise. Thus, this calculator may over-predict the rate at which burn survivors sweat while exercising, leading to a misrepresentation of fluid replacement needs in this unique population. PURPOSE: Our study tested the hypothesis that the values obtained from sweatratecalculator.com would incorrectly predict fluid loss during exercise in the heat in burn survivors. METHODS: Data from four heat stress trials were combined, resulting in a total of 80 burn survivors (17 – 84% body surface area burned) included in the analysis. Whole body sweat loss was measured by changes in nude body mass pre- and post-exercise. Participants exercised at moderate intensity (4 to 5 METS) for 40 - 90 minutes in either air conditioned (~24ºC) or warm temperatures (~42ºC). Whole-body sweat loss was compared to the estimated sweat loss provided by the calculator based on the specific conditions the participants exercised in. RESULTS: In burn survivors with lower sweat rates (≤0.4 l/hr), the calculator over predicted sweat rates by an average of 0.19 ± 0.22 l/hr. In contrast, the calculator under predicted sweat rates by an average of 0.14 ± 0.17 l/hr in burn survivors with higher sweat rates (≥0.9 l/hr). The Bland-Altman Plot presents these data as a negative slope, reflecting a proportional bias. Despite this bias, the observed error between modelled and actual whole-body sweat rate across all burn survivors was similar to the error observed in non-burned individuals from the original publication (p