Reliability of alternative devices for postoperative patient temperature measurement: two prospective, observational studies.

Peri-operative hypothermia is associated with significant morbidity, yet limitations exist regarding non-invasive temperature assessment in the post-anaesthesia care unit (PACU). In this prospective study of 100 patients, we aimed to determine the reliability of two commonly used temperature measurement devices, forehead temporal artery temperature and tympanic measurement, in addition to an indwelling urinary catheter with temperature probe, in comparison with the final nasopharyngeal core temperature at the end of surgery. Agreement of forehead measurement with nasopharyngeal temperature showed a mean bias (±95% limits of agreement) of 0.15 °C (±1.4 °C), with a steep slope of the relationship on the Bland-Altman plot of -0.8, indicating a tendency to normalise patient temperature readings to 36.4 °C. Only 54% of hypothermic cases were correctly detected by the forehead measurement device. Agreement of tympanic measurement with nasopharyngeal core temperature measurement was marginally improved with a mean bias of 0.13 °C (95% limits of agreement ±1.15 °C). In contrast, agreement of bladder temperature with nasopharyngeal temperature showed a mean (SD) bias of 0.19 (0.28) °C (95% limits of agreement ±0.54 °C), with a relatively flat line of best fit. We demonstrated that two commonly used temperature measurement devices, forehead temporal artery temperature and tympanic measurement, compared with nasopharyngeal core temperature, were imprecise and unreliable following major surgery. However, the indwelling catheter with temperature sensor was precise and acceptable for continuous core temperature measurement in the PACU.