Core body temperature measurement: a comparison of axilla, tympanic membrane and pulmonary artery blood temperature

Abstract

This research study was undertaken to examine the relationship between pulmonary artery blood temperature (regarded as the 'gold standard' measurement for core body temperature), axilla temperature using the Tempa.DOT Ax chemical thermometer and tympanic membrane temperature using the Diatek 9000 InstaTemp thermometer. Sixty adult intensive care patients had their temperatures monitored. A single set of five simultaneous temperatures, i.e. left and right axilla, left and right tympanic membrane (TM), and pulmonary artery (PA) blood were recorded. The mean difference between left and right TM temperatures was 0.58 degree C, and although both were moderately well correlated with PA temperature (r = 0.63 and 0.78, respectively) the mean differences between the two sites were clinically significant (0.85 degree C and 0.94 degree C, respectively). The range of differences between the sites was significant. Plotting limits of agreement showed that both left and right TM temperatures may be up to 1.2 degrees C above or 1.3 degrees C below PA blood temperature: a clinically unacceptable range. In particular, large temperature differences were recorded when patients were lying with one side of their head to a pillow. Fan therapy directed to the head was not found to affect these differences significantly. The mean difference between left and right axilla temperatures was 0.36 degree C, and although both were modestly correlated with PA temperature (r = 0.48 and 0.53, respectively) the mean differences between the two sites were clinically significant (0.47 degree C and 0.50 degree C, respectively). The range of differences between the sites was particularly significant. Plotting limits of agreement showed that both left and right axilla temperatures may be up to 1.2 degrees C above or 1.6 degrees C below PA blood temperature: a clinically unacceptable range. Because the range of temperature differences found between PA blood and the other sites was so great, it is concluded that neither the chemical axilla thermometer nor the tympanic membrane thermometer used in this study are clinically reliable tools for adult intensive care patients.