Patterns of Intraoperative Hypothermia

Hypothermia during general anesthesia develops with a characteristic pattern. An initial rapid decrease in core temperature is followed by a slow, linear reduction in core temperature. Finally, core temperature stabilizes and subsequently remains virtually unchanged ( Fig. 40-7 ). Each portion of this typical pattern has a different etiology.

Volatile anesthetics cause vasodilation through a direct peripheral action.^[54] More importantly, they also inhibit tonic thermoregulatory vasoconstriction, which results in arteriovenous shunt dilation. ^{[21] [22] [23] [24] [25]} Nonetheless, anesthetic-induced vasodilation increases cutaneous heat loss only slightly.^[55] Anesthetics reduce the metabolic rate 20% to 30%.^[56] However, even the combination of increased heat loss and reduced heat production is insufficient to explain the 0.5°C to 1.5°C decrease in core temperature usually observed during the first hour of anesthesia.

The key to understanding the initial decrease in core temperature is to appreciate that body heat is not normally evenly distributed. Core temperature represents only about half the body mass (mostly the trunk and head); the remaining mass is typically 2°C to 4°C cooler than the core. This core-to-peripheral tissue temperature gradient is normally maintained by tonic thermoregulatory vasoconstriction. Anesthetic-induced vasodilation, however, allows core heat to flow peripherally. This heat redistribution warms the arms and legs, but does so at the expense of the core ( Fig. 40-8 and Fig. 40-9 ).^[57]

1577

Figure 40-7 Hypothermia during general anesthesia develops with a characteristic pattern. An initial rapid decrease in core temperature results from a core-to-peripheral redistribution of body heat. This redistribution is followed by a slow, linear reduction in core temperature that results simply from heat loss exceeding heat production. Finally, core temperature stabilizes and subsequently remains virtually unchanged. This plateau phase may be a passive thermal steady state or might result when sufficient hypothermia triggers thermoregulatory vasoconstriction. Results are presented as means ± SD.

After the initial redistribution hypothermia, core temperature usually decreases in a slow, linear fashion for 2 to 4 hours. This reduction results simply from heat loss exceeding metabolic heat production.^[58] After 3 to 4 hours of anesthesia, core temperature generally reaches a plateau and remains virtually constant for the duration of surgery.^[31] The core temperature plateau may simply represent a thermal steady state (heat production equaling heat loss) in patients remaining relatively warm.^[59] In others, however, the plateau phase is associated with peripheral thermoregulatory vasoconstriction triggered by core temperatures of 33°C to 35°C.^[60]

Thermoregulatory vasoconstriction during anesthesia significantly decreases cutaneous heat loss,^[45] but this decrease alone is usually insufficient to produce a thermal steady state. Furthermore, neither adults^[41] nor infants^[43]

Figure 40-8 Cartoon illustrating internal redistribution of body heat after induction of general anesthesia. Hypothermia after induction of spinal or epidural anesthesia results similarly, but redistribution is restricted to the legs.

Figure 40-9 Changes in body heat content and distribution of heat within the body during induction of general anesthesia (at elapsed time zero). Subtraction of the change in mean body temperature from the change in core (tympanic membrane) temperature results in the core hypothermia specifically attributable to redistribution. Redistribution hypothermia was thus not a measured value; instead, it is defined by the decrease in core temperature not explained by the relatively small decrease in systemic heat content. After 1 hour of anesthesia, core temperature had decreased 1.6°C ± 0.3°C, with redistribution contributing 81% to the decrease. Even after 3 hours of anesthesia, redistribution contributed 65% to the entire 2.8°C ± 0.5°C decrease in core temperature. Results are presented as means ± SD. (Redrawn with modification from Matsukawa T, Sessler DI, Sessler AM, et al: Heat flow and distribution during induction of general anesthesia. Anesthesiology 82:662–673, 1995.)