Central body temperature, which usually is well controlled, typically decreases more than 1 degree C during the 1st h of general anesthesia. This hypothermia has been attributed partially to an anesthetic-induced peripheral vasodilation, which increases cutaneous heat loss to the environment. Based on the specific heat of humans, heat loss would have to increase more than 70 W for 1 h (in a 70-kg person) to explain hypothermia after induction of general anesthesia. However, during epidural anesthesia, sympathetic blockade increases heat loss only slightly. Furthermore, thermoregulatory vasoconstriction in unanesthetized humans decreases heat loss to the environment only 15 W. Therefore, we tested the hypothesis that the hypothermia that follows induction of general anesthesia does not result from increased cutaneous heat loss. Heat loss and skin-surface and tympanic membrane temperatures, before and after induction of isoflurane anesthesia, were measured in five minimally clothed volunteers. Peripheral skin blood flow was evaluated with venous-occlusion volume plethysmography and skin-surface temperature gradients. Cutaneous heat losses in watts were summed from ten area-weighted thermal flux transducers. Tympanic membrane temperature, which was stable during the 30-min control period preceding induction, decreased 1.2 +/- 0.2 degrees C in the 50 min after induction. Isoflurane anesthesia decreased mean arterial blood pressure approximately 20%. Average skin-surface temperature increased over 15 min to 0.5 degree C above control. Heat loss from the trunk, head, arms, and legs decreased slightly, whereas loss from the hands and feet (10.5% of the body surface area) doubled (P less than 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)