Résumé
PURPOSE: This study compared the temporal artery temperature (TAT) measured by infrared temporal artery thermometers to the axillary temperature (AT) measured by standard mercury-in-glass thermometers, and evaluated accuracy of the TAT measurement for clinical practice. METHODS: A total of 247 adult inpatients in general wards in a tertiary medical center located in Seoul participated in the study. The TAT was measured within one minute after the AT measurement. Data were analyzed using descriptive statistics, paired t-test, Pearson correlation coefficient, linear regression, and the Bland-Altman plot. RESULTS: There was a significant difference in mean temperature between AT and TAT, 36.89℃ (SD=0.70) versus 37.35℃ (SD=0.72). The Bland-Altman plots demonstrated the difference between the AT and TAT as −1.29 to +0.33. The specificity and sensitivity of the TAT in detecting fever were high. The positive predictive values were 57.5% and 71.0% when the AT were higher than 38.0℃ and the TAT fever cutoff levels were 38.0℃ and 38.3℃ respectively. CONCLUSION: TAT and AT were highly correlated and agreeable, indicating that TAT is as accurate as AT. The findings suggested that TAT measurement can be used in clinical practice. For accurate communication between medical personnel, medical institutions need to provide guidelines for temperature measurement, especially for the use of thermometer and measurement sites.