Accuracy of electronic axillary temperature measurement in term and preterm neonates.

The accuracy of axillary temperature measurement using an electronic thermometer in the predictive mode was investigated in term and preterm infants. Predictive mode measurements were compared to axillary monitoring mode measurements and to rectal temperature recordings. Fifty temperature recordings were obtained in each of four groups of neonates (full-term/radiant warmer, full-term/open crib, preterm/radiant warmer, preterm/incubator). Correlation of axillary predictive and monitoring mode temperatures ranged from r = 0.83 to 0.90. Correlations between axillary monitoring mode and rectal measurements were r = 0.69 to 0.80, and between predictive mode and rectal measurements were r = 0.59 to 0.80. Small, statistically significant differences were found between predictive and monitoring mode axillary temperatures in preterm neonates (0.1 to 0.2 degrees F). Rectal temperatures were higher than axillary predictive temperatures by 0.5 degrees F and higher than axillary monitoring mode temperatures by 0.7 degrees F. Ranges of variation of +/- 0.8 to 1.1 degrees F were similar in comparisons of rectal with axillary predictive and monitoring mode measurements. The results of this study support the use of axillary measurement in the predictive mode for clinical measurement of temperature in neonates. Clinicians should be aware of the expected differences between measurements at axillary and rectal sites and the potential for variation in measurements across measurement modes and sites.