Comparison of body temperature measurement with temporal artery thermometer and other fever measurement methods in children aged 3-9 years

Background: Fever is defined as a rise in body temperature above normal and is the most common reason for children to apply to the emergency department in many countries of the world with different levels of development. For over a century, fever could be measured rectally, axillary and orally with mercury thermometers. Nowadays, infrared and electronic thermometers have started to measure body temperature from the rectum, axilla, oral cavity, tympanic membrane, skin and finally the temporal artery. In this study, it was aimed to compare the temporal artery thermometer, which has been used in recent years, with oral electronic, infrared non-contact skin and tympanic thermometers and to determine the sensitivity and specificity values of each method in different fever estimation values. Materials and methods: This prospective study included 104 children aged 3 to 9 years, who met the exclusion criteria, admitted to the general pediatrics and pediatric emergency outpatient clinics of a large city hospital. The body temperature of each participant was measured with an oral, tympanic, cutaneous and temporal artery thermometer, following the application instructions of the devices. The reference temperature measurement method was oral thermometer. Sensitivity, specificity, positive and negative predictive values were calculated using appropriate statistical methods with the data. Results: All thermometers correlated with oral thermometer measurements. The strongest correlation was found in tympanic thermometer measurements. The significance of the difference between the median values of the reference thermometer and other body temperature measurement methods was evaluated. Oral body temperature measurements were significantly higher than cutaneous measurements. The body temperature value obtained with the temporal measurement was found to be significantly higher than the oral measurement. There was no statistical difference between oral and tympanic measurement. Different cut-off values were determined for all measurements, and sensitivity, specificity, positive predictive value and negative predictive value were calculated for each value. In the cutaneous measurement, the sensitivity decreased as the fever predictive value increased. The highest sensitivity was 97.2% at 37 °C, while the lowest sensitivity was 16.7% at 38 °C. The specificity was quite high for all cutoff values and increased to 100% at 37.4 °C and above measurements. Tympanic measurement showed high sensitivity and specificity for all cutoff values. While it was 100% sensitive at 37.5 °C and below, the sensitivity decreased at 37.7 °C and above. Its specificity increased with the measured value, reaching 100% at the 38 °C cutoff value. Temporal artery thermometer showed high sensitivity for all cut-off values. The lowest sensitivity was determined as 91.7% at 38 °C. As the predictive value decreased, the specificity decreased and the lowest was determined as 48.5% at 37 °C. Finally, sensitivity, specificity, positive predictive value and negative predictive values calculated for parental sentiment were 97.2%, 88.2%, 81.4%, and 98.4%, respectively. Conclusions: In our study, we interpreted the method that best reflects the reference thermometer as tympanic thermometer, second temporal and third cutaneous thermometer. Becouse of the measurement of the tympanic thermometer in clogged with cerumen and otitis ears is not safe, we can place the TAT in the second rank in fever screening, behind the reference thermometer. In our study, TAT preceded cutaneous thermometer due to the higher rate of detecting patients. 1. Keywords: Body temperature measurement, Fever, Thermometer, Temporal artery, Oral