DKK Ng, JCS Ho, RKN Yuen

The purpose of this study was to evaluate the accuracy and precision of tympanic temperature taken in full-term neonates.

Methodology

Neonates who were admitted during the period from January to February, 1995 were recruited into the study. The tympanic temperature was recorded by Thermoscan PRO-1 tympanic infrared thermometer (Thermoscan, Inc., San Diego, CA) The rectal temperature was recorded by mercury thermometer. At each temperature taking setting, the left and right tympanic temperature was taken twice on each side and rectal temperature was taken once. Normothermia was defined as rectal temperature between 35.0°C and 37.8°C.

Results

Forty-nine patients were recruited in this study. The total number of rectal readings was 1181. Tympanic temperature readings corresponded to rectal temperature more than 37.8°C or less than 35.0°C were excluded from analysis. The mean left tympanic temperature was 36.64°C +/- 0.35°C. The mean right tympanic temperature was 36.64°C +/- 0.37°C. The mean difference of right and left tympanic temperature was 0.00°C +/- 0.21°C. The repeatability of left and right tympanic temperature was the same at 0.17°C (acceptable if < 0.2°C, American Society for Testing and Materials Standard). In 59 % of tympanic temperature taking, the first reading was higher than the second reading. Survey of nursing staff (n=20) was conducted after the study was finished. Thermoscan was rated as excellent in terms of speed of use, cleanliness and safety. However, eight out of twenty (40%) rated the thermoscan difficult to use because they were concerned they could not point the scan at the tympanic membrane accurately. Twelve out of twenty (60%) were concerned about the cost as a new probe cover was used at every setting. Nine out of twenty (45%) were concerned about the accuracy of thermoscan.

Conclusion

This study showed that use of tympanic infrared thermometer produce highly reproducible result in neonates. The reference range is 35.9°C - 37.4°C. The accuracy would be increased if the higher figure of two readings are taken as the true tympanic temperature.