The effects of preinduction warming on temperature and blood pressure during propofol/nitrous oxide anesthesia

Forced-air surface warming versus oesophageal heat exchanger in the prevention of peroperative hypothermia

BACKGROUND

In a prospective, randomized, placebo-controlled study we investigated the efficacy of 2 different heating methods in 24 patients undergoing abdominal surgery of at least 2 h expected duration.

METHODS

Group I: control, no active warming. Group II: forced-air surface warming on upper extremities and upper thorax. Group III: warming with oesophageal heat exchanger. All patients had a standardized, combined general and epidural anaesthesia. Core and skin temperatures were measured at induction of general anaesthesia, and subsequently every 30 min, and changes in total body temperature were calculated.

RESULTS

There were no statistically significant differences between the 3 groups regarding demographic data. Patients in groups I and III developed hypothermia, while this was not the case with patients in group II. When using analysis of variance with repeated measurements, there was no significant difference in core temperature, comparing group I and group III (P = 0.299) or the interaction between time and treatment of these groups (P = 0.373). As a consequence, data from groups I and III were pooled and regarded as an internal group on the one hand, and group II as an external group on the other hand. Core temperature, the mean skin temperature and total body temperature were significantly different comparing the internal group and the external group. The interaction between time and treatment was likewise found to be significantly different.

CONCLUSION

We conclude that in major abdominal procedures lasting 2 h or more, serious hypothermia develops unless effective measures to prevent hypothermia are used. Forced-air warming of the upper part of the body is effective in maintaining normothermia in these patients, while central heating with an oesophageal heat exchanger, at least in its present form, does not suffice to prevent hypothermia.

Oxygen metabolism during liver transplantation: the effect of dichloroacetate

Dichloroacetate (DCA) stimulates pyruvate dehydrogenase (PDH), accelerating recovery of the postischemic heart. Because DCA also stimulates hepatic PDH, it may facilitate graft recovery during liver transplantation (OLT). Hepatic removal and replacement during OLT produce major changes in O2 consumption (VO2), and return of baseline VO2 has been used to index early graft function. We examined the effect of DCA on O2 metabolism during OLT. Forty patients received DCA 80 mg/kg intravenously in divided doses, and 40 served as controls. Serial measurements were made for body temperature, hemodynamics, O2 metabolic indices, and plasma substrate and hormonal concentrations. Oxygen delivery (DO2I) and consumption (VO2I) indices were calculated. Patients exhibited stable hemodynamics, with similar fluid and blood product requirements. Compared with the dissection stage, DO2I and VO2I were decreased during the anhepatic stage (31% and 36%, respectively), then returned to dissection stage values soon after portal vein unclamping. Temperature decreased during the anhepatic stage and returned toward dissection stage value after graft perfusion. DCA reduced lactic acidosis and NaHCO3 use but did not alter hemodynamics or measures of O2 metabolism or body temperature. VO2 is decreased during the anhepatic stage largely due to loss of hepatic metabolism. Restoration of VO2 by 30 min after portal vein unclamping reflects rapid recovery of O2 metabolism by the graft liver, but DCA does not accelerate recovery of VO2. DCA does not seem to facilitate early graft hepatic function as indexed by VO2.

IMPLICATIONS

We evaluated whether dichloroacetate, which stimulates pyruvate dehydrogenase, can accelerate recovery of graft liver hepatic function during liver transplantation, as indexed by oxygen consumption. We found that despite evidence that it activated pyruvate dehydrogenase, dichloroacetate did not affect recovery of transplanted liver function.

Sympathetic blockade does not enhance tissue warming during isolated heated limb perfusion

Isolated, heated limb perfusion is used for the treatment of locally recurrent melanoma, intransit metastases, and acral lentiginous melanomas. Tissue warming during this procedure requires adequate perfusion within the isolated extremity. At our institution, spinal or epidural anesthesia was used to produce sympathetic blockade and vasodilation for lower extremity procedures. More recently, we began using mild systemic hyperthermia to produce active thermoregulatory vasodilation. In the presence of heat stress, sympathetic blockade may actually decrease skin blood flow because active cutaneous vasodilation, which is associated with sweating, is dependent on intact sympathetic innervation. We therefore investigated whether the continued use of neuraxial blockade was justified. Twenty patients undergoing lower extremity perfusions were alternately assigned to receive either combined general and spinal anesthesia or general anesthesia alone. All were aggressively warmed using forced air and circulating water. There were no significant differences in tissue temperatures (measured at four sites in the isolated limb) between groups at any time before or after the start of perfusion. Similarly, pump flow (715 +/- 211 mL/min versus 965 +/- 514 mL/min) and the time required to achieve an average tissue temperature of 39 degrees C (43 +/- 16 vs 34 +/- 13 min) were not different between groups (spinal versus no spinal). Sweating was observed in all but three patients at esophageal temperatures of 37.9 +/- 0.6 degrees C. We conclude that sympathetic blockade confers no added benefit for tissue warming during isolated limb perfusions in the presence of induced mild systemic hyperthermia.

IMPLICATIONS

Sympathetic blockade prevents adrenergic vasoconstriction, but also inhibits active, neurally mediated cutaneous vasodilation (a normal thermoregulatory response to heat). In slightly hyperthermic patients, we demonstrated that spinal anesthesia does not improve convective tissue warming during isolated, heated limb perfusion. Mild systemic hyperthermia may promote greater vasodilation than sympathetic blockade.

Perioperative thermoregulation and heat balance

In summary, both regional and general anesthesia markedly impair the normal precise regulation of core body temperature. Consequently, inadvertent perioperative hypothermia is common. Hypothermia develops because the typical operating room environment is cold; however, it is anesthetic-induced impairment of thermoregulatory responses that contributes most. Internal redistribution of body heat is a surprisingly important factor, contributing more to core hypothermia than net heat loss in most patients. There is now convincing evidence that a typical amount of intraoperative hypothermia, say 2 degrees C, predisposes numerous complications and alters patient outcome. Fortunately, effective methods are available for preventing hypothermia.

Enflurane decreases the threshold for vasoconstriction more than isoflurane or halothane

Intraoperative hypothermia results largely from anesthetic-induced inhibition of tonic thermoregulatory vasoconstriction. Sufficient hypothermia, however, triggers peripheral vasoconstriction, which usually prevents further decrease in core temperature. The thermoregulatory effects of all volatile anesthetics have been tested in adults and/or children, but different anesthetics have not been directly compared. We therefore evaluated thermoregulatory responses during enflurane, isoflurane, and halothane administration. Anesthesia was maintained with 1 minimum alveolar anesthetic concentration (MAC) of halothane, isoflurane, or enflurane in 27 patients undergoing intraabdominal surgery. Patients were maintained normovolemic and normocapnic but were allowed to cool passively. A forearm minus fingertip, skin-temperature gradient of 4 degrees C identified significant vasoconstriction; the core temperature triggering vasoconstriction identified the threshold. Morphometric characteristics, initial core temperatures, ambient operating room temperatures, blood pressures, and anesthetic potencies were similar in each group. All eight patients given halothane vasoconstricted at a core temperature of 35.5 +/- 0.6 degrees C. Eight of the patients given isoflurane vasoconstricted at a core temperature of 35.2 +/- 0.5 degrees C. However, two others did not at minimum core temperatures of 34.0 and 33.8 degrees C. Only one patient given enflurane vasoconstricted at a core temperature of 34.6 degrees C. The other six patients never vasoconstricted, at minimum core temperatures of 33.6 +/- 0.4 degrees C. Our data indicate that enflurane profoundly inhibits thermoregulatory responses in children. The mechanism for this extraordinary inhibition remains unknown but does not result from any obvious anesthetic pharmacology or thermoregulatory physiology. We conclude that unwarmed pediatric patients will become colder when anesthetized with enflurane than with halothane or isoflurane.

Rapid core-to-peripheral tissue heat transfer during cutaneous cooling

Perioperative thermal manipulations are usually directed at the skin surface because methods of directly warming the core are invasive or ineffective. However, inadequate heat flow between peripheral and core compartments will decrease the rate at which core temperature changes. We therefore determined whether core hypothermia is delayed after initiation of surface cooling. Six volunteers were anesthetized with propofol and midazolam, and maintained under three layers of passive insulation for 2.5-4 h. Subsequently, the skin surface was cooled using forced air, 1000 L/min, at 10 degrees C. Isoflurane was added as necessary to maintain arteriovenous shunt vasodilation. Overall heat balance was determined from the difference between cutaneous heat loss (thermal flux transducers) and metabolic heat production (oxygen consumption). Average arm and leg (peripheral) tissue temperatures were determined from 19 intramuscular needle thermocouples, 10 skin temperatures, and "deep" foot temperature. Overall body heat content decreased approximately 234 kcal during 2.5 h of active cooling. Core temperature, which was nearly constant before active cooling, decreased approximately 1.3 degrees C/h. There was no delay between initiation of active cooling and the decrease in core temperature. Furthermore, peripheral (arm and leg) and core (trunk and head) tissue heat contents decreased at virtually the same rates: approximately 50 kcal/h and approximately 47 kcal/h, respectively. These data indicate that there is little restriction of heat flow between peripheral and core tissues in vasodilated, anesthetized subjects.

The etiology and management of inadvertent perioperative hypothermia

Mild perioperative hypothermia is a frequent complication of anesthesia and surgery. Core temperature should be monitored during general anesthesia and during regional anesthesia for large operations. Reliable sites of core temperature monitoring include the tympanic membrane, nasopharynx, esophagus, bladder, rectum, and pulmonary artery. The skin surface is not an acceptable site for monitoring core temperature. Anesthetic-induced vasodilation initially rapidly decreases core temperature secondary to an internal redistribution of heat rather than an increased heat loss to the environment. Both general and regional anesthetics impair thermoregulation, increasing the interthreshold range; that is, the range of core temperatures over which no autonomic response to cold or warmth occurs. Preinduction skin surface warming is the only means to prevent this initial redistribution hypothermia. Forced-air warming is the most effective method of rewarming hypothermic patients intraoperatively.

Propofol blood concentration and the Bispectral Index predict suppression of learning during propofol/epidural anesthesia in volunteers

Propofol is often used for sedation during regional anesthesia. We tested the hypothesis that propofol blood concentration, the Bispectral Index and the 95% spectral edge frequency predict suppression of learning during propofol/epidural anesthesia in volunteers. In addition, we tested the hypothesis that the Bispectral Index is linearly related to propofol blood concentration. Fourteen healthy, male volunteers were studied on three randomly ordered days: no propofol, target propofol blood concentration 1 microgram/mL, and target propofol blood concentration 2 micrograms/mL. Each day, epidural anesthesia (approximately T11 level) was induced using 2% 2-chloroprocaine. Propofol was infused by a computer-controlled pump, and propofol concentration measured in central venous blood. We administered a Trivial Pursuit-type question task on all 3 days. The electroencephalogram was monitored continuously (Fp1, Fp2; reference, Cz; ground, mastoid). Propofol caused concentration-related impairment of learning. The propofol blood concentration suppressing learning by 50% was 0.66 +/- 0.1 microgram/mL. The Bispectral Index value when learning was suppressed by 50% was 91 +/- 1. In contrast, the 95% spectral edge frequency did not correlate well with learning. The Bispectral Index decreased linearly as propofol blood concentration increased (Bispectral Index = -7.4.[propofol] + 90; r2 = 0.47, n = 278). There was no significant correlation between the 95% spectral edge frequency and propofol concentration. In order to suppress learning, propofol blood concentrations reported to produce amnesia may be targeted. Alternatively, the Bispectral Index may be used to predict anesthetic effect during propofol sedation.

Pancuronium does not decrease oxygen consumption during hypothermic or normothermic cardiopulmonary bypass

Muscle relaxation reportedly reduces systemic oxygen consumption 30% during hypothermic cardiopulmonary bypass. A potential mechanism is inhibition of hypothermia-induced subclinical muscle tone ("microshivering"). Accordingly, we tested the hypothesis that pancuronium administration would decrease systemic oxygen consumption during hypothermic cardiopulmonary bypass, but not during normothermic bypass. We studied 20 patients undergoing normothermic cardiopulmonary bypass at a nasopharyngeal temperature of 37.6 +/- 0.8 degrees C (mean +/- SD) and 15 patients undergoing hypothermic bypass at a nasopharyngeal temperature of 28.2 +/- 1.2 degrees C. Oxygen consumption during bypass was determined, using the Fick principle, before and after administration of pancuronium 0.15 mg/kg. Morphometric characteristics and anesthetic management were similar in the two groups. Oxygen consumption averaged 93 +/- 28 mL.min-1.m-2 in the normothermic patients and 43 +/- 10 mL.min-1.m-2 in the hypothermic group. Administration of pancuronium decreased oxygen consumption only 2% in the normothermic patients. However, muscle relaxation also decreased oxygen consumption only 2% in the hypothermic patients. We were thus unable to confirm our hypothesis because muscle relaxation induced by administration of pancuronium did not significantly reduce oxygen consumption in either group.

Postoperative hemodynamic and thermoregulatory consequences of intraoperative core hypothermia

STUDY OBJECTIVE

To evaluate the postoperative hemodynamic and thermoregulatory consequences of intraoperative core hypothermia.

DESIGN

Prospective, randomized clinical trial.

SETTING

Operating room and postanesthesia care unit of a university hospital.

PATIENTS

74 healthy, ASA status I, II, and III patients (average age 58 yrs) undergoing elective colon surgery.

INTERVENTIONS

Patients were randomly assigned to be kept normothermic or approximately 2.5 degrees C hypothermic during surgery. Anesthesia was maintained with isoflurane, nitrous oxide, and fentanyl. Postoperatively, surgical pain was treated with patient-controlled analgesia (PCA) opioid.

MEASUREMENTS AND MAIN RESULTS

An observer blinded to group assignment and core temperatures evaluated shivering, thermal comfort, surgical pain, heart rates (HRs), and blood pressures (BPs) during the first six postoperative hours. Morphometric characteristics, oxygen saturation, fluid balance, PCA-administered opioid, and visual analog pain scores were comparable in the two groups. Hypothermic patients felt uncomfortably cold during recovery, and their postoperative core temperatures remained significantly less than in the normothermic patients for more than four hours. Peripheral vasoconstriction and shivering were common in the hypothermic patients but rare in those kept normothermic. HRs and BPs were comparable in the two groups.

CONCLUSIONS

These data confirm that the effects of intraoperative hypothermia on postoperative HR and BP are modest in relatively young, generally healthy patients. In contrast, intraoperative hypothermia caused substantial postoperative thermal discomfort, and full recovery from hypothermia required many hours. Delayed return to care normothermia apparently resulted largely from postoperative thermoregulatory impairment.

Pre-induction skin-surface warming minimizes intraoperative core hypothermia

STUDY OBJECTIVE

To test the hypothesis that only one hour of preinduction skin-surface warming decreases the rate at which core hypothermia develops during the first hour of anesthesia.

DESIGN

Randomized, prospective study.

SETTING

Operating theater of a university hospital.

PATIENTS

16 ASA status I and II adult patients scheduled for laparoscopic cholecystectomy under general anesthesia.

INTERVENTIONS

Eight patients were assigned to receive forced-air warming for one hour before induction of anesthesia (prewarmed group); the other eight patients were covered only with a wool blanket during a similar preinduction period (control group).

MEASUREMENTS AND MAIN RESULTS

Tympanic membrane (core) and mean skin-surface temperatures were measured at 15-minutes intervals, starting one hour before induction of anesthesia. Mean skin temperature increased from 34.0 +/- 0.1 C to 37.0 +/- 0.2 degrees C in the pre-warmed group (p < 0.05), but remained unchanged at 34.7 +/- 0.3 degrees C in the control group. Core temperature during the preinduction period did not change significantly in either group. Following induction of anesthesia, core temperature decreased at a rate of 1.1 +/- 0.1 degrees C/hr in the control group, but only 0.6 +/- 0.1 degrees C/hr in the pre-warmed group (p < 0.05). After one hour of anesthesia, six of eight pre-warmed patients had core temperatures of at least 36.5 degrees C, whereas only one of the eight control patients did (p < 0.05).

CONCLUSIONS

A single hour of preoperative skin-surface warming reduced the rate at which core hypothermia developed during the first hour of anesthesia. Preoperative skin surface warming is particularly helpful during short procedures because redistribution hypothermia is otherwise difficult to treat.

Nitrous oxide decreases the threshold for vasoconstriction less than sevoflurane or isoflurane

The core temperature triggering thermoregulatory arteriovenous shunt constriction is designated the threshold for vasoconstriction. High thresholds are generally desirable because vasoconstriction helps prevent further core hypothermia by decreasing cutaneous heat loss and constraining metabolic heat to the core thermal compartment. Previous studies suggest that nitrous oxide (N2O) may inhibit thermoregulatory vasoconstriction less than comparable doses of volatile anesthetics. To confirm this impression, we tested the hypothesis that 0.5 minimum alveolar anesthetic concentration (MAC) N2O combined with 0.5 MAC sevoflurane or isoflurane would reduce the vasoconstriction threshold less than 1.0 MAC sevoflurane or isoflurane. With institutional review board approval, we studied 40 patients, aged 20-60 yr, undergoing open abdominal surgery. No premedication was given. Ten patients each were anesthetized with: 1) N2O (50%) and 0.5 MAC sevoflurane (1%); 2) sevoflurane alone (2%); 3) N2O (60%) and 0.5 MAC isoflurane (0.6%); and, 4) isoflurane alone (1.2%). A forearm minus fingertip, skin temperature gradient > or = 0 degree C was considered significant vasoconstriction; the esophageal temperature triggering vasoconstriction identified the threshold. Morphometric characteristics were comparable in each group. The threshold for vasoconstriction was 35.8 +/- 0.3 degrees C in the patients given 50% N2O combined with 0.5 MAC sevoflurane, which was significantly greater than that in those given 1.0 MAC sevoflurane: 35.1 +/- 0.4 degrees C. Similarly, the threshold for vasoconstriction was 35.9 +/- 0.3 degrees C in the patients given 60% N2O combined with 0.5 MAC isoflurane, which was significantly greater than that in those given 1.0 MAC isoflurane: 35.0 +/- 0.5 degrees C. We thus conclude that N2O impairs thermoregulation less than sevoflurane or isoflurane.

Morphometric influences on intraoperative core temperature changes

Intraoperative core hypothermia develops in three characteristic phases: 1) core-to-peripheral redistribution of body heat that is most prominent during the first hour after induction of anesthesia; 2) subsequent slow linear decrease in core temperature resulting largely from heat loss exceeding metabolic heat production; and 3) core temperature plateau resulting when thermoregulatory vasoconstriction decreases cutaneous heat loss and constrains metabolic heat to the core thermal compartment. Accordingly, we tested the hypotheses that: 1) core cooling does not depend on body fat (BF) or the ratio of weight-to-surface area (Wt/SA) during the initial redistribution phase; 2) the core cooling phase; 2) the core cooling rate is a function of the Wt/SA ratio during the second phase; and 3) the rate of core cooling during the plateau phase (after vasoconstriction) will be determined by the percentage of BF. In 40 patients undergoing elective colon surgery, the amount of redistribution hypothermia was inversely proportional to the percentage of BF (delta TC = 0.034.BF-2.2, r2 = 0.63) and the Wt/SA ratio (delta TC = 0.052.Wt/SA-3.35, r2 = 0.66). The core cooled linearly during the second phase, and the cooling rate was inversely proportional to the Wt/SA ratio (rate = 0.035.(Wt/SA)-2.2, r2 = 0.29). Thermoregulatory vasoconstriction was effective in virtually all patients independent of their morphology, and produced a four-fold reduction in the core cooling rate. These results indicate that patient morphometric characteristics substantially influence intraoperative core temperature changes, and that the effect depends on the hypothermia phase.

Thermoregulatory vasoconstriction increases the difference between femoral and radial arterial pressures

OBJECTIVE

Thermoregulatory vasoconstriction locally increases arterial wall tension and arteriolar resistance, thereby altering physical properties of the arteries. The arterial pressure waveform is an oscillatory phenomenon related to those physical characteristics; accordingly, we studied the effects of thermoregulatory vasomotion on central and distal arterial pressures, using three hydraulic coupling systems having different dynamic responses.

METHODS

We studied 7 healthy volunteers. Central arterial pressure was measured from the femoral artery and distal pressure was measured from the radial artery, using 10.8-cm long, 20-gauge catheters. Three hydraulic coupling systems were used: (1) a 10-cm-long, 2-mm internal diameter connector; (2) a 150-cm-long, 1-mm internal diameter connector (Combidyn 520-5689, B. Braun, Melsungen, Germany); (3) a 180-cm long, 2-mm internal diameter connector (Medex MX564 and MX562, Medex Inc., Hillard, OH). Brachial artery pressure was measured oscillometrically. Core temperature was measured at the tympanic membrane. The vasomotor index, defined as finger temperature minus room temperature, divided by core temperature minus room temperature, was used to estimate the degree of vasoconstriction. Constriction was considered near maximal when the index was less than 0.1, and minimal when it exceeded 0.75. Measurements were taken every 3 min. Baseline readings were obtained when subjects were warm. They then were cooled by exposure to 20 degrees C to 22 degrees C room air and a circulating-water mattress set at 4 degrees C until index was less than 0.1. They then were rewarmed by increasing water temperature to 42 degrees C and adding a forced-air warmer until the vasomotor index exceeded 0.75. Data were analyzed by ANOVA and linear regression.

RESULTS

Thermoregulatory vasoconstriction was associated with marked arterial pressure waveform changes. Radial pressure showed, in lieu of a dicrotic notch, large oscillations of decreasing amplitude. Femoral pressure showed a single diastolic oscillation of smaller amplitude. The waveforms appeared different, depending on the hydraulic coupling system used, artifact being more marked with the longer connectors. On the average, radial systolic pressure exceeded femoral systolic pressure during vasoconstriction; however, during vasodilatation, femoral systolic pressure exceeded radial systolic pressure (p < 0.05). Oscillometric measurements underestimated systolic pressure, and did so more markedly during vasoconstriction. There were no differences in the values of mean and diastolic pressures.

CONCLUSION

Thermoregulatory vasoconstriction alters radial arterial pressure waveform, artifactually increasing its peak systolic pressure compared with the femoral artery. Poor dynamic responses of recording systems further distort the waveforms. Consequently, radial artery pressure may be misleading in vasoconstricted patients.

Bair hugger forced-air warming maintains normothermia more effectively than thermo-lite insulation

STUDY OBJECTIVE

To compare the ability of forced-air warming and reflective insulation to maintain intraoperative normothermia.

DESIGN

Prospective, randomized clinical trial.

SETTING

Operating rooms of a general hospital.

PATIENTS

20 ASA physical status I and II patients undergoing elective total hip arthroplasty.

INTERVENTIONS

Patients were randomly assigned to be warmed intraoperatively using forced-air or reflective insulation. Inspired gases were conditioned using a heat-and-moisture exchanger in both groups, and infused intravenous fluids were warmed to 37 degrees C.

MEASUREMENTS AND MAIN RESULTS

Distal esophageal (core) temperatures decreased approximately 0.5 degrees C in both groups during the first 45 minutes of anesthesia. Subsequently, core temperatures increased slightly in the patients given forced-air warming. In contrast, core temperatures continued to decrease in patients covered with reflective insulation. After 135 minutes of anesthesia, core temperatures were 36.4 +/- 0.6 degrees C (mean +/- SD) in the forced-air group but only 35.4 +/- 0.6 degrees C in the insulated group (p < 0.01, unpaired t-test). These data indicate that forced-air warming is superior to reflective insulation.

CONCLUSION

Reflective insulation was unable to maintain intraoperative normothermia during total hip arthroplasty. Active warming, such as that provided by forced air, was required to prevent hypothermia.

Mild hypothermia during isoflurane anesthesia decreases resistance to E. coli dermal infection in guinea pigs

Small changes in core temperature profoundly alter cutaneous blood flow, a major factor influencing resistance to wound infection. Furthermore, when measured in vitro, various immune functions are temperature dependent in the physiological range. Accordingly, we tested the hypothesis that mild hypothermia impairs and mild hyperthermia improves resistance to dermal infections. Thirty-two guinea pigs were anesthetized for 6 h using 1.5% (1.25 MAC) inspired isoflurane. Their core temperatures were maintained at either 39 degrees C (normal for guinea pigs, n = 11), 36 degrees C (n = 12), or 41 degrees C (n = 9). One h after induction of anesthesia, 2 x 10(8) E. coli were injected intradermally with a 26-g needle at eight sites on each animal's back. Core temperatures were not controlled after recovery from anesthesia, and animals in each group were maintained in the same environment. Twenty-four h after injection, the area of induration surrounding each injection site was measured. This is a standard test of resistance to wound infection. Values were compared using one-way ANOVA and Scheffé's S tests. Results are presented as means +/- standard deviations; differences were considered significant when P < 0.05. Areas of inflammation on the hypothermic animals were significantly larger (48 +/- 10 mm2) than those on normothermic (36 +/- 10 mm2) or hyperthermic (37 +/- 6 mm2) animals. These data suggest that mild hypothermia during anesthesia significantly impairs resistance to dermal infection. In contrast, mild hyperthermia does not appear to be protective.