This study tested the hypothesis that the threshold for thermoregulatory vasoconstriction is lowered as isoflurane concentration increases, but that the intensity of vasoconstriction, once triggered, is well preserved during isoflurane anesthesia. The thermoregulatory threshold was prospectively defined as the central temperature at which vasoconstriction occurred, and significant vasoconstriction was defined as a skin-surface temperature gradient (forearm–fingertip) ≥ 4° C. The threshold for thermoregulatory vasoconstriction and the intensity of vasoconstriction, measured as maximum skin-temperature gradient, was determined in six unpremedicated patients electively donating a kidney during isoflurane anesthesia, and in four healthy, awake volunteers. All anesthetized patients were deliberately cooled and became hypothermia. Vasoconstriction occurred in five of six at central temperatures between 35.3 and 32.4° C, at end-tidal isoflurane concentrations between 0.74 and 1.65%. The patient who did not vasoconstrict received the highest isoflurane concentration (≈2.5%) and reached a central temperature of 31° C. Unanesthetized volunteers also were exposed to cold and each vasoconstricted at a temperature near 37° C. The threshold for thermoregulatory cutaneous vasoconstriction was inversely correlated with anesthetic dose, the thermoregulatory threshold decreasing ≈3 ° C/% isoflurane concentration. There were no statistically significant differences between maximum skin-surface temperature gradients in awake volunteers and patients given isoflurane, or between any of the groups when patients from previous studies given halothane or nitrous oxide/fentanyl anesthesia were included in the comparison. These data indicate that the intensity of vasoconstriction, once triggered, is similar during several different types of anesthesia. A high correlation between calf–toe and forearm–fingertip temperature gradients, and between esophageal and tympanic membrane temperatures, also was demonstrated.