OBJECTIVE To determine the electroencephalographic (EEG) changes induced by mild hyperthermia during enflurane anesthesia and to test the reliability of two new infrared thermometers. METHODS Prospective laboratory evaluation. METHODS The Thermoregulation Research Laboratory at the University of California, San Francisco. METHODS 6 healthy female volunteers aged 30 +/- 8 years. METHODS Epidural anesthesia (approximately T10 dermatome) was induced and maintained using 2-chloroprocaine anesthesia. General anesthesia was induced by inhalation of nitrous oxide and enflurane and maintained with enflurane at an end-tidal concentration of 1.7%. A minimum of 2 degrees C core hyperthermia was induced by active cutaneous warming, and the volunteers subsequently were passively cooled. RESULTS EEG data were recorded from gold cup electrodes positioned at FP1 and FP2, with the reference electrode at CZ and the ground lead on the mastoid. In addition to routine EEG parameters, we evaluated the bispectral index. Bispectral analysis quantifies the phase coupling between various frequencies in the power spectrum and may be a useful measure of anesthetic depth. Core temperature was measured at the left tympanic membrane and distal esophagus. Core temperature also was determined from the right ear using two new, infrared tympanic membrane thermometers. One of these directly measures tympanic temperature, and the other extrapolates core temperature from the external ear canal. Induction of 2 degrees C core hyperthermia did not produce statistically significant or clinically important changes in beta or delta power, the 95% spectral edge frequency, or the bispectral index. Temperatures recorded from the right ear by the direct thermometer were 0.27 degrees C +/- 0.33 degrees C less than those measured in the left ear, but the values correlated well (r2 = 0.95 +/- 0.04). Temperatures recorded from the right ear by the core temperature extrapolater were 0.42 degrees C +/- 0.33 degrees C lower than those measured in the left ear, and the correlation between values was slightly worse (r2 = 0.83 +/- 0.16). CONCLUSIONS Since mild core hyperthermia does not alter routine EEG parameters or the bispectral index, typical perianesthetic thermal disturbances are unlikely to obscure EEG estimates of anesthetic depth. Both the direct thermometer and the core temperature extrapolater were found to be sufficiently accurate and precise for routine clinical use, but the direct thermometer would be preferable in the perioperative period.