Therefore, the data was grouped into higher and lower temperature ranges. One hundred and 46 healthy men and women, aged 12 through 52 years.Īxillary temperature from both sides and oral temperature were recorded using digital electronic thermometer in supine position with due precautions.ĭifference in temperature on the two sides of axilla was found to vary by as much as 0 degree F to 3.4 degrees F. To evaluate variation of axillary temperature on the two sides and its correlation with oral temperature. In such instances an accurate temperature is necessary, and the route.Unilateral axillary temperature is recorded in unconscious patients and children due to its easy approach. For example, for immunodeficient patients and for some inpatients with a variety of severe infections, clinical interventions may be initiated when threshold levels of temperature are exceeded. In a few instances, however, a child’s exact temperature may affect the child’s subsequent management. The answer seems to be “yes and no.” In most cases, the otherwise healthy-appearing child with a minor illness (such as otitis media) is unlikely to have medical management changed despite the described differences in temperature between the axilla and the rectum.

The important question for pediatricians is whether the statistical variability identified in this meta-analysis is clinically important. While tradition has suggested that adding 1☌ to temperatures measured in the axilla approximates temperatures measured in the rectum, the data presented by Craig et al suggest this is not the case. This is a well-designed, thoroughly researched comparison of rectal and axillary temperatures. The authors conclude that the poor agreement between temperature measured in the rectum and axilla has important implications when body temperature requires precise measurement. For axillary temperatures measured with mercury thermometers there was a tendency toward improved accuracy as placement time increased. For mercury thermometers the mean difference was 0.25☌ (95% CI, −0.15☌ to 0.65☌) and for electronic thermometers the mean difference was 0.85☌ (95% CI, −0.19☌ to 1.90☌). Mean axillary temperature was always lower than mean rectal temperature, but the authors found large mean differences and sample standard deviations when rectal and axillary measurements were compared. Twenty studies (n=3201 participants) had sufficient data and met inclusion criteria for the meta-analysis.

Studies of children with hypothermia or preterm infants, and studies when the rectal thermometer was read before 3 minutes had elapsed, were rejected. Studies of children from birth to 18 years were included. They also contacted authors of studies and suppliers of clinical thermometers in order to provide details of other studies.

The authors performed an extensive search of the literature in order to find relevant studies from sources that included electronic databases for published literature and conference abstracts for unpublished literature. In this paper, Craig and colleagues from Liverpool, UK, have systematically reviewed studies comparing temperatures measured in the axilla (the test site) with temperatures measured in the rectum (the reference site) in the same child concurrently or immediately sequentially, using either mercury or electronic thermometers or indwelling thermocouple probes that had been calibrated (off-the-shelf thermometers have been shown to be inaccurate by at least 0.1☌ 2). 1 However, temperature is now routinely measured at other more convenient sites including the ear, mouth, forehead, and axilla, often with a conversion or fudge factor added to estimate the rectal temperature. Rectal temperature is considered the gold standard by many and outcome studies have often used rectal temperature to make clinical decisions.

Axillary temperature conversion chart