Infrared Ear Thermometry in Water-Related Accidents—Not a Good Choice

Abstract

Hypothermia in near-drowning victims is a serious problem that impacts clinical decision-making. The purpose of this trial was to determine the reliability of tympanic temperature measurements compared to oral temperature measurements after immersion in water. After ethical approval was obtained, we studied oral and tympanic temperature in 25 volunteer swimmers (aged 18-49 years). Sublingual (Fixotherm; Tradesell Europe, Eglharting, Germany) and tympanic (First Temp Genius; Sherwood Medical, Sulzbach, Germany) temperature measurements were performed before entering the water, after 45 min of immersion in water, and 15 min after leaving the water. During the immersion phase, the ears were temporarily immersed. A control group (the same 25 volunteers) had to swim for the same amount of time without ever immersing their heads in the water. The trial was performed in an indoor swimming pool at 28 degrees C water and 30 degrees C air temperature. The oral temperature did not change over time in either group. The tympanic temperature was significantly lower after immersion compared to baseline in the "immersed" group (33.7 degrees C vs. 37.5 degrees C, p < 0.001), increased significantly in the recovery period, but remained significantly lower than baseline (36.0 degrees C vs. 37.5 degrees C, p < 0.001). At baseline, the oral temperature was lower compared to the tympanic temperature. This relationship reversed after immersion and remained reversed until the end of the trial in the immersion group. The control group maintained oral temperatures lower than tympanic throughout the study; furthermore, the control group had no clinically relevant change in oral or tympanic temperature over the time (tympanic temperature: 37.4 degrees C vs. 37.2 degrees C, p = 0.06). Our data suggest that in water-related accidents such as near drowning, the values of body (core) temperature obtained via use of infrared ear thermometry are unreliable, and should not be used for clinical decision-making.