Taking an infant's temperature: axillary or rectal thermometer?

Abstract

Aim: to examine the validity of routine use of axillary thermometers to measure body temperature in infants.

Methods: comparison of 49 simultaneous paired axillary and rectal temperature recordings in infants between one and 11 months of age, who were considered afebrile by their clinicians. Regression analysis of the paired measurements, and a plot of the differences between the paired measurements against their means, were used to examine the limits of agreement of the two methods.

Results: the limits of agreement ranged from 0.2 to 1.6 degrees C difference between simultaneous paired measurements. As there was no consistent relationship between axillary and rectal temperatures, the use of a correction factor added to axillary values is invalid. The axillary thermometer may record almost the same as the rectal temperature or more than 1.5 degrees C lower at the same point in time.

Conclusion: we question the validity of current clinical practice in assessing infant body temperature by using axillary thermometers, and we suggest that such measurement does not reflect the rectal temperature in a reliable or consistent fashion. If infant body temperature is sought, a rectal thermometer should be used.