Purpose: The gradient between temperatures measured at different body sites is not constant; one factor which will change this gradient is rapid changes in body temperature. Measurement of this gradient was done in patients undergoing rapid changes in body temperature to establish the best site to measure temperature and to compare two brands of commercial tympanic thermometers.
Method: A total of 228 sets of temperatures were measured from probes in the oesophagus, rectum, and axilla and from two brands of tympanic thermometer and compared with pulmonary artery (PA) temperature in 18 adults during cardiac surgery.
Results: Measurements from the oesophageal site was closest to PA readings (mean difference 0.0 +/- 0.5 degree C) compared with IVAC tympanic thermometer (mean difference -0.3 +/- 0.5 degree C), Genius tympanic thermometer (mean difference -0.4 +/- 0.5 degree C), axillary (mean difference 0.2 +/- 1.0 degrees C) and rectal (mean difference -0.4 +/- 1.0 degree C) readings. When data during cooling were analysed separately, all sites had similar gradients from PA except for rectal, which was larger. On rewarming, oesophageal readings were closest to PA readings; tympanic readings were closer to PA than were rectal or axillary readings. Readings from the two brands of tympanic thermometer were equivalent.
Conclusion: Oesophageal temperature is more accurate and will reflect rapid changes in body temperature better than tympanic, axillary, or rectal temperature. When oesophageal temperature cannot be measured, tympanic temperature done by a trained operator should become the reading of choice.