NASHVILLE, TENN. – Axillary thermometry outperformed both rectal and temporal artery thermometry in 205 newborns aged 12-72 hours in a study performed at the University of North Carolina at Chapel Hill.
The infants had two temperatures taken by each method over a period of 15 minutes, for a total of six readings per child and 1,230 measurements overall. Axillary thermometry proved both accurate and reliable. Rectal thermometry was accurate but less reliable, and temporal thermometry was reliable but less accurate.
Lead investigator Ketan Nadkarni, MD, a 3rd-year pediatrics resident, and his colleagues wanted to compare the three methods head-to-head to make sure axillary thermometers were okay to use in the nursery, and to see if it really was necessary to tell parents to use rectal thermometers; many are reluctant to use them. Plus, “there’s been a lot of controversy” in pediatrics “over the best way to measure temperature,” Dr. Nadkarni said at the Pediatric Hospital Medicine annual meeting.
“With our data, we think axillary is what we should continue to use in the newborn nursery,” he said. Some attending physicians still are hesitant to recommend axillary thermometers to new parents, but “all of the nurses are aware of” the study findings “and a lot of the residents are, too, so I think we are starting to move” in that direction.
The study had some unexpected findings as well: “The biggest surprise was how wide the distribution of rectal temperatures was. The distribution” around the mean “was way larger than we had thought, so [rectal thermometry was] not very reliable at all. Our study surprisingly exhibited suboptimal performance in terms of reliability,” for rectal thermometry, he said at the meeting, which was sponsored by the Society of Hospital Medicine, the American Academy of Pediatrics, and the Academic Pediatric Association.
Specifically, the average distance of any given rectal measurement from the mean rectal temperature of 98.3º F was 0.45º F. The second rectal temperature in the study sometimes varied a half a degree or more from the first taken shortly before, in the same infant.
The average distance of an axillary temperature from the axillary mean of 98.32º F was 0.32º F; for temporal thermometry it was 0.34º F from a mean of 98.55º F.
Another surprise was that temporal thermometry overestimated temperature by an average of about a quarter of a degree, compared with rectal readings. Even small overestimates could lead to unnecessary sepsis work-ups; “the last thing we want is to hospitalize these kids when they don’t need to be,” Dr. Nadkarni said.
The mean axillary and rectal temperatures, meanwhile, were only 0.02º F apart, which was not statistically significant. “Axillary was absolutely interchangeable with rectal in terms of accuracy,” he said.
The children were born at 37 weeks’ gestation or later, and were excluded if they had a temperature of 100.4º F or higher by any method. Rectal and axillary temperatures were taken with a Welch Allyn. Temple temperatures were taken with an Exergen .
The investigators plan to run a similar trial in the ED with children up to 3 months old.
There was no external funding for the work, and Dr. Nadkarni had no relevant financial disclosures.