Nitrous Oxide Decreases the Threshold for Vasoconstriction Less Than Sevoflurane Or Isoflurane

Nitrous Oxide Decreases the Threshold for Vasoconstriction Less Than Sevoflurane or Isoflurane

Makoto Ozaki, MD*, Daniel I. Sessler, Mm, Hidehiro Suzuki, MD*, Kyoko Ozaki, MD*, Chiharu Tsunoda, MD*, and Kenji Atarashi, MD* *Department of Anesthesia, Tokyo Women’s Medical College, Tokyo, Japan, and the tThermoregulation Research Laboratory, Department of Anesthesia, University of California, San Francisco, California

The core temperature triggering thermoregulatory (1%); 2) sevoflurane alone (2%); 3) N,O (60%) and 0.5 arteriovenous shunt constriction is designated the MAC isoflurane (0.6%); and, 4) isoflurane alone threshold for vasoconstriction. High thresholds are (1.2%). A forearm minus fingertip, skin temperature generally desirable because vasoconstriction helps gradient 20°C was considered significant vasocon- prevent further core hypothermia by decreasing cu- striction; the esophageal temperature triggering va- taneous heat loss and constraining metabolic heat to soconstriction identified the threshold. Morphomet- the core thermal compartment. Previous studies sug- ric characteristics were comparable in each group. gest that nitrous oxide (N,O) may inhibit thermoreg- The threshold for vasoconstriction was 35.8 + 0.3”C ulatory vasoconstriction less than comparable doses in the patients given 50% N,O combined with 0.5 of volatile anesthetics. To confirm this impression, MAC sevoflurane, which was significantly greater we tested the hypothesis that 0.5 minimum alveolar than that in those given 1.0 MAC sevoflurane: 35.1 2 anesthetic concentration (MAC) N,O combined with 0.4”C. Similarly, the threshold for vasoconstriction 0.5 MAC sevoflurane or isoflurane would reduce the was 35.9 + 0.3”C in the patients given 60% N,O com- vasoconstriction threshold less than 1.0 MAC bined with 0.5 MAC isoflurane, which was signifi- sevoflurane or isoflurane. With institutional review cantly greater than that in those given 1.0 MAC board approval, we studied 40 patients, aged 20-60 isoflurane: 35.0 + 0.5”C. We thus conclude that N,O yr, undergoing open abdominal surgery. No premed- impairs thermoregulation less than sevoflurane or ication was given. Ten patients each were anesthe- isoflurane. tized with: 1) N,O (50%) and 0.5 MAC sevoflurane (Anesth Analg 1995;80:1212-6)

ntraoperative hypothermia is common, and causes All general anesthetics so far tested significantly de- potentially serious complications (l-7). The reduc- crease the thermoregulatory threshold for vasoconstic- I tion in core temperature immediately after in- tion (i.e., the core temperature triggering vasoconstric- duction of general anesthesia results largely from tion) (8-11). Because vasoconstriction is effective in anesthetic-induced inhibition of tonic thermoregula- minimizing additional core hypothermia (15), anesthetic tory vasoconstriction (8 -1 l), and subsequent core-to- combinations causing less-than-typical thermoregula- peripheral redistribution of body heat (12). Heat loss tory inhibition may be preferable in some patients. exceeding metabolic heat production then typically Nitrous oxide (N,O) (60%) combined with fentanyl continues to reduce body temperature for several ad- (4 pg -kg-’ *h-l) decreases the threshold to 34.2 + ditional hours (13). Finally, reemergence of thermo- 0.5”C (16). Similarly, 30% N,O decreases the shivering regulatory vasoconstriction (in patients becoming threshold -1°C (17), thus nearly doubling the normal sufficiently hypothermic) moderates further core hy- sweating-to-shivering range (18). However, adding pothermia by decreasing cutaneous heat loss (14) N20 to enflurane decreases the vasoconstriction and constraining metabolic heat to the core thermal threshold less than expected (191, suggesting that N,O compartment (15). may impair thermoregulation less than comparable doses of other anesthetics. To confirm this impression, This study was supported by the Joseph Drown Foundation and we tested the hypothesis that 0.5 minimum alveolar National Institutes of Health Grant GM49670. anesthetic concentration (MAC) N20 combined with Accepted for publication January 31, 1995. 0.5 MAC sevoflurane or isoflurane would reduce the Address correspondence and reprint requests to Daniel I. Sessler, MD, Thermoregulation Research Laboratory, University of Califor- vasoconstriction threshold less than 1 .O MAC sevoflu- nia, San Francisco, CA 94143-0648. rane or isoflurane alone.

01995 by the International Anesthesia Research Society 1212 An&h Analg 1995;80:12124 0003-2999/95/$5.00 ANESTH ANALG OZAKI ET AL. 1213 1995;80:1212-6 NITROUS OXIDE AND THERMOREGULATION

Methods inserted until the patients felt the thermocouple touch the tympanic membrane; appropriate placement was With institutional review board approval and in- confirmed when they easily detected a gentle rubbing formed consent, we studied 40 ASA physical status of the attached wire. The probe was then taped in I-II patients. They were aged 20-60 yr, and undergo- place, the aural canal occluded with cotton, and the ing elective abdominal surgery. None of the partici- external ear covered with a gauze pad. Tympanic pants had a history of thyroid disease, dysautonomia, membrane temperatures correlate well with distal Raynaud’s syndrome, or malignant hyperthermia. esophageal temperatures in the perioperative period Surgery started near 8:30 AM in nearly all of the pa- (24,25). After induction of anesthesia, core tempera- tients, and typically finished near noon. Twenty initial ture was recorded from the distal esophagus. patients were given sevoflurane; subsequently, 20 oth- Mean skin temperature was calculated from four sites: ers were studied during isoflurane anesthesia. 0.3&m + Tam) + O.p(Tthigh + Tcaif) (26). Fingertip Upon arrival to the operating suite, 10 mL/kg of perfusion was evaluated using forearm minus fingertip unwarmed intravenous (IV) fluid was administered. skin temperature gradients. Perfusion was recorded Without any premeditation, anesthesia was induced from an arm exposed to the operating room environ- by inhalation of 70% N,O and progressively greater ment and not having a blood pressure cuff or IV catheter. concentrations of sevoflurane or isoflurane. Intubation There is an excellent correlation between skin tempera- of the trachea was facilitated by administration of ture gradients and volume plethysmography (27). vecuronium bromide (0.1 mg/kg). Mechanical venti- Temperatures were recorded from Mon-a-Therm@ lation was maintained with a circle system having a thermocouples (Mallinckrodt Anesthesia Products, fresh gas flow of 6 L/min, and adjusted to maintain Inc., St. Louis, MO). They were connected to a cali- end-tidal Pco, near 35 mm Hg. No airway heating or brated Iso-Thermex@ 16-channel electronic thermom- humidification was provided. No thiopental or opioid eter (Columbus Instruments International, Corp., Co- was administered. lumbus, OH) having an accuracy of O.l”C and a Patients given sevoflurane were then randomly as- precision of O.Ol”C. signed to maintenance anesthesia consisting of: 1) 50% Heart rate was monitored continuously using three- N,O (-0.5 MAC) and 0.5 MAC sevoflurane (1%); or 2) lead electrocardiography. Blood pressure was deter- 1.0 MAC sevoflurane (2%) (20,21). Patients given mined oscillometrically at 5-min intervals. We used isoflurane were then randomly assigned to mainte- oscillometric rather than direct arterial blood pressure nance anesthesia consisting of: 1) 60% N,O (eO.5 measurements to minimize the artifact induced by MAC) and 0.5 MAC isoflurane (0.6%); or 2) 1.0 MAC thermoregulatory vasoconstriction (28,29). Respira- isoflurane (1.2%) (20,22). The concentration of N,O tory gas concentrations were quantified using a cali- was increased in the second series because there is still brated end-tidal gas analyzer (Datex Medical Instru- some controversy about the MAC of this drug in mentation, Inc., Tewksbury, MA). Data were recorded humans. at lo-min intervals, starting immediately before induc- Supplemental vecuronium was administered as tion of anesthesia (“initial” values). needed to maintain one to two twitches in response to As in previous studies (151, we considered a gradi- supramaximal stimulation of the ulnar nerve at the ent of 0°C to indicate significant thermoregulatory wrist. At least 10 mL * kg-i . h-i fluid was given IV vasoconstriction. The distal esophageal temperature and blood products were replaced to maintain the triggering significant vasoconstriction was considered hematocrit between 25%-32%. Administered fluids the thermoregulatory threshold. The preinduction were warmed to 37°C. fluid bolus was not considered part of the intraoper- The patients were covered with a single layer of ative fluid balance. Fluid administered to the time of surgical draping (23); no other warming measures vasoconstriction was divided by the time to vasocon- were taken during the study period. Twenty min- striction to produce the fluid administration rate. Am- utes after significant vasoconstriction was observed, bient temperature in each case was averaged over time patients were actively rewarmed using a forced-air from induction of anesthesia until vasoconstriction. system (Bair Hugger@; Augustine Medical, Inc., Morphometric data, initial core temperatures, am- Eden Prairie, MN). Subsequent anesthetic manage- bient temperatures, and fluid administration rates in ment was left to the discretion of the responsible the patients given each volatile anesthetic were com- anesthesiologist. pared using Mann-Whitney U-tests. Anesthetic con- Ambient temperature was measured using a ther- centrations, vasoconstriction thresholds, core cooling mocouple positioned at the level of the patient, well rates, times to constriction, mean skin temperatures, away from any heat-producing equipment. Core tem- heart rates, and arterial blood pressures at the time of perature, before induction of anesthesia, was meas- constriction were similarly compared using Mann- ured at the tympanic membrane. The aural probe was Whitney &tests. All values are expressed as means 1214 OZAKI ET AL. ANESTH ANALG NITROUS OXIDE AND THERMOREGULATION 1995;80:1212-6

Table 1. Morphometric Characteristics, Anesthetic Management, and Results in Patients Given 0.5 MAC Sevoflurane and 50% N,O (Sevoflurane/N,O), 1.0 MAC Sevoflurane, 0.5 MAC Isoflurane and 60% N,O (Isoflurane/N,O), and 1.0 MAC Isoflurane Sevoflurane/N,O Sevoflurane Isoflurane/N,O Isoflurane Age (yr) 45 t 11 51 26 50 2 12 51 28 Weight (kg) 62 ? 9 59 t 6 54 k 10 55 + 9 Height (cm) 164 2 10 162 ? 7 158 ? 10 160 2 7 Gender (M/F) 5/5 5/5 4/6 5/5 Initial core temperature PC) 37.1 + 0.2 3.72 t- 0.3 37.3 f 0.4 37.3 IT 0.1 Ambient temperature (“C) 23.6 + 1.5 23.1 ” 1.4 23.9 + 0.6 23.8 t 1.0 End-tidal (anesthetic) (%) 1.0 + 0.1 *2.0 f 0.1 0.6 2 0.1 *1.2 * 0.1 End-tidal (N,O) (%) 50 r 1 60 -c 1 Fluid administration rate 14.2 2 0.7 14.3 + 2.3 14.0 t 0.3 13.6 + 4.7 (mL * kg-i * h-i) Vasoconstriction threshold (“C) 35.8 2 0.3 *35.1 + 0.4 35.9 ? 0.3 *35.0 ? 0.5 Vasoconstriction time (h) 1.2 + 0.4 1.9 + 0.7 1.1 ? 0.4 $1.7 2 0.4 Rate of core cooling (“C/h) 1.2 t 0.3 1.2 2 0.3 1.3 2 0.4 1.2 + 0.4 Skin temperature (“C) 32.6 ? 0.6 32.0 +- 1.1 32.9 t 0.6 32.5 + 1.1 Heart rate (bpm) 65 2 6 73 ? 8 75 + 10 71 + 9 Mean blood pressure (mm Hg) 92 + 8 84 +- 11 93 2 9 95 2 12 All data except the morphometric values, initial (preinduction) core temperatures, ambient temperatures, and fluid administration rates were recorded at the time of vasoconstriction. Values are expressed as means 2 SD; * statistically significant differences between the groups (P < 0.01). MAC = minimum alveolar anesthetic concentration.

+ SD; differences were considered significant when isoflurane than in those given only sevoflurane or P < 0.01. isoflurane, although anesthetic potency was comparable in each of the groups. N20, thus, impairs thermo- Results regulation less than sevoflurane or isoflurane. Painful stimulation increases the vasoconstriction Morphometric characteristics were comparable in the threshold during enflurane anesthesia (lo), presum- four groups. The initial core temperatures, ambient ably by increasing the activation of the sympathetic operating room temperatures, core cooling rates, fluid nervous system. The observed relatively high vaso- administration rates, heart rates, and blood pressures constriction threshold during N,O administration did not differ significantly. End-tidal sevoflurane, may similarly result from the drug’s “sympathetic isoflurane, and N,O concentrations did differ, per pro- activation” (32). tocol (Table 1). Thermoregulatory vasoconstriction, once initiated, is The threshold for vasoconstriction was 35.8 2 0.3”C remarkably effective in minimizing further intraopera- in the patients given 50% N20 combined with 0.5 tive core hypothermia (15). [The clinical importance of MAC sevoflurane. In contrast, the threshold was 35.1 thermoregulation is illustrated by exaggerated hypother- ? 0.4”C in those given 1.0 MAC sevoflurane. Simi- mia when leg vasoconstriction is prevented by combin- larly, the threshold for vasoconstriction was 35.9 + ing regional and general anesthesia (33).] Consequently, 0.3”C in the patients given 60% N,O combined with anesthetics producing relatively little inhibition of ther- 0.5 MAC isoflurane, which was significantly greater moregulatory vasoconstriction provide some protection than that in those given 1.0 MAC isoflurane: 35.0 2 against intraoperative core hypothermia. Our data sug- 0.5”C (Fig. 1). gest that combining N,O with volatile anesthetics reduces the vasoconstriction threshold less than a com- Discussion parable potency of volatile anesthetics alone; this combination may thus be preferable in some patients. N,O is known to impair thermoregulatory control in The lower limit of “normothermia” is generally con- rodents, causing an autonomic (30) and behavioral sidered to be 36°C. However, core temperatures this (31) decrease in core temperature. Similarly, N,O re- low are unusual in unanesthetized individuals, and duces both the vasoconstriction (19) and shivering typically observed-if at all-near 3:00 AM (34). Mean (17,18) thresholds in humans. However, this report is core temperature in humans is near 37”C, and the the first quantifying the relative potency of N,O and a range of temperatures tolerated without provoking volatile anesthetic. Our results indicate that the vaso- thermoregulatory responses is only 0.2”C (35). Conse- constriction threshold was significantly greater in pa- quently, a more physiologic definition of normother- tients given a combination of N,O and sevoflurane or mia might be 37°C. Even 0.5 MAC N,O combined ANESTH ANALG OZAKI ET AL. 1215 1995;80:1212-6 NITROUS OXIDE AND THERMOREGULATION

37 core temperature plateau resulting from intraoperative thermoregulatory vasoconstriction (15). Accord- 1 ingly, we considered a gradient of 0°C [corresponding 36 1 to a finger blood flow of ~1 mL/min (2711 as significant vasoconstriction in this study. Threshold 35 _ 0 A consequence of accepting less vasoconstriction as (“Cl 0 f “significant,” is that thresholds defined in this fashion 34- Q will be systematically higher than when a more re- strictive definition of constriction is used. Conse- quently, thresholds defined in terms of a 0°C skin “V temperature gradient should not be directly compared 1.2 % Isoflurane 0.6 % Isoflurane with those defined by other criteria. However, ob- 37 and 60 % N20 served differences between the groups in this study remain valid. A major assumption of this investigation is that the 36 1 anesthetic potencies of N,O and sevoflurane or isoflu- Threshold rane are additive. Although N,O and three volatile (“Cl 8 0 anesthetics were reported to contribute nonlinearly to 35 total MAC in rats (39), the apparent nonadditivity : resulted from an incorrect MAC for N,O (40). Current data thus suggest that MAC fractions of N,O and 34 ’ volatile anesthetics are additive. 2% Sevoflurane 1% Sevoflurane A second major assumption of this investigation is and 50% N20 that the reported anesthetic potencies of N,O, sevoflu- Figure 1. The vasoconstriction threshold was significantly greater rane, and isoflurane are accurate. The MAC of N,O in in patients given 0.5 minimum alveolar anesthetic concentration humans remains to be absolutely established. The (MAC) sevoflurane combined with 50% nitrous oxide (35.8 t 0.4”C) than in those given 1.0 MAC sevoflurane (35.1 ? 0.3”C). Similarly, most commonly cited value is 1.04% (21), although the vasoconstriction threshold was significantly greater in patients some controversy remains because a suitable test un- given 0.5 MAC isoflurane combined with 60% nitrous oxide (35.9 2 der hyperbaric conditions has yet to be reported. Sim- 0.3”C) than in those given 1.0 MAC isoflurane (35.0 k 0.5”C) (*P < 0.01). Open circle indicate the vasoconstriction threshold in each ilarly, the MAC of sevoflurane is generally considered patient; the closed circles show the mean and SD in each group. to be 2.0% (20,41), although 1.8% has also been reported (42). The MAC of isoflurane, in contrast, is with 0.5 MAC sevoflurane or isoflurane reduced the clearly established at 1.15% (20,22). Available data vasoconstriction threshold to <36”C. Clinicians there- thus suggests that total anesthetic potency was com- fore should not depend on thermoregulatory re- parable in our four study groups. sponses alone to prevent intraoperative hypothermia. Because the results in our first set of patients might Instead, patients should be given sufficient passive have been compromised by uncertainty about the ex- insulation (23,361 and active warming (13,37,38) to act MAC values of N20 and sevoflurane, we also maintain intraoperative core temperatures well above evaluated patients given isoflurane where there is no 36°C. doubt about the MAC. To further assure that our In previous studies we usually considered skin tem- initial conclusions were valid, we used a MAC of N,O perature gradients exceeding 4°C as significant vaso- equaling 1.2 atm, a value certainly exceeding the ac- constriction (9-11). We initially chose this value [cor- tual partial pressure required to prevent movement. responding to a finger blood flow of -0.2 mL/min Once again, the data indicated that the threshold for (2711 to minimize the chance of inappropriately attrib- vasoconstriction was significantly greater in patients uting vasoconstriction resulting from inadequate given the combination of N,O and isoflurane than anesthetic depth or vascular volume depletion to ther- isoflurane alone. moregulation. However, thermoregulatory thresholds In summary, the threshold for vasoconstriction was are defined by the core temperatures triggering re- 35.8 + 0.3”C in the patients given 50% N,O combined sponses. An increase in the gradient, from typical with 0.5 MAC sevoflurane, which was significantly vasodilated values of --2°C to 0°C better indicates greater than that in those given 1.0 MAC sevoflurane: initiation of vasoconstriction, and is thus more consis- 35.1 + 0.4”C. Similarly, the threshold for vasoconstric- tent with the physiologic definition of “threshold.” tion was 35.9 + 0.3”C in the patients given 60% N,O Furthermore, we have previously demonstrated that a combined with 0.5 MAC isoflurane, which was signif- (calf-to-toe) gradient of 0°C best indicates onset of the icantly greater than that in those given 1.0 MAC 1216 OZAKI ET AL. ANESTH ANALG NITROUS OXIDE AND THERMOREGULATION 1995:80:12124

isoflurane: 35.0 + 0.5”C. N,O thus impairs thermoreg- 19. Cochrane R, Belani K, Sessler DI. Nitrous oxide decreases the ulation less than sevoflurane or isoflurane. thermoregulatory threshold for vasoconstriction [abstract]. Anesth Analg 1993;76S48. 20. Frink EJ Jr, Malan TP, Atlas M, et al. Clinical comparison of sevoflurane and isoflurane in healthy patients. Anesth Analg The authors thank Mallinckrodt Anesthesiology Products, Inc., for 1992;74:241-5. providing the thermocouples used in this study. 21. Hornbein T, Eger EI II, Winter I’, et al. The minimum alveolar concentration of nitrous oxide in man. Anesth Analg 1982;61: 553-6. References 22. Stevens WC, Dolan WM, Gibbons RT, et al. Minimum alveolar concentrations (MAC) of isoflurane with and without nitrous 1. Frank SM, Beattie C, Christopherson R, et al. Unintentional oxide in patients of various ages. Anesthesiology 1975;42: hypothermia is associated with postoperative myocardial ische- 197-200. mia. Anesthesiology 1993;78:468-76. 23. Sessler DI, McGuire J, Sessler AM. Perioperative thermal insu- 2. Carli F, Emery PW, Freemantle CAJ. Effect of peroperative lation. Anesthesiology 1991;74:875-9. normothermia on postoperative protein metabolism in elderly 24. Steen R, Sessler DI. The thermoregulatory threshold is inversely patients undergoing hip arthroplasty. Br J Anaesth 1989;63: proportional to isoflurane concentration. Anesthesiology 1990; 276-82. 72~822-7. 3. Heier T, Caldwell JE, Sessler DI, et al. Mild intraoperative 25. Cork RC, Vaughan RW, Humphrey LS. Precision and accuracy hypothermia increases duration of action and spontaneous re- of intraoperative temperature monitoring. Anesth Analg 1983; covery of vecuronium blockade during nitrous oxide-isoflurane 62:211-4. anesthesia in humans. Anesthesiology 1991;74:815-9. 26. Ramanathan NL. A new weighting system for mean surface 4. Just B, Delva E, Camus Y, et al. Oxygen uptake during recovery temperature of the human body. J Appl Physiol 1964;19:531-3. following naloxone. Anesthesiology 1992;76:60-4. 27. Rubinstein EH, Sessler DI. Skin-surface temperature gradients 5. Valeri CR, Khabbaz K, Khuri SF, et al. Effect of skin temperature correlate with fingertip blood flow in humans. Anesthesiology on platelet function in patients undergoing extracorporeal by- 1990;73:541-5. pass. J Thorac Cardiovasc Surg 1992;104:108-16. 28. Hynson JM, Sessler DI, Moayeri A, et al. The effects of pre- 6. Sessler DI, Rubinstein EH, Moayeri A. Physiological responses induction warming on temperature and blood pressure during to mild perianesthetic hypothermia in humans. Anesthesiology propofol/nitrous oxide anesthesia. Anesthesiology 1993;79: 1991;75:594-610. 219-28. 7. Sheffield CW, Sessler DI, Hunt TK. Mild hypothermia during 29. Hynson JM, Sessler DI, Moayeri A, et al. Thermoregulatory and isoflurane anesthesia decreases resistance to E. coli dermal in- anesthetic-induced alterations in the differences between femo- fection in guinea pigs. Acta Anaesthesiol Stand 1994;38:201-5. ral, radial, and oscillometric blood pressure measurements. An- 8. Leslie K, Sessler DI, Bjorksten A, et al. Propofol causes a dose- esthesiology 1994;81:1411-21. dependent decrease in the thermoregulatory threshold for va- 30. Quack RM, Panek RW, Kouchich FJ, et al. Nitrous oxide-in- soconstriction, but has little effect on sweating. Anesthesiology duced hypothermia in the rat. Life Sci 1987;41:683-90. 1994;81:36>0. 31. Pertwee RG, Marshall NR, MacDonald AG. Effects of subanes- 9. Kurz A, Plattner 0, Sessler DI, et al. The threshold for thermo- thetic doses of inert gases on behavioral thermoregulation in regulatory vasoconstriction during nitrous oxide/isoflurane an- mice. J Appl Physiol 1986;61:1623-33. esthesia is lower in elderly than young patients. Anesthesiology 32. Eisele JH, Smith NT. Cardiovascular effects of 40 percent nitrous 1993;79:465-9. oxide in man. Anesth Analg 1972;51:956-63. 10. Washington DE, Sessler DI, McGuire J, et al. Painful stimulation 33. Joris H, Ozaki M, Sessler DI, et al. Epidural anesthesia impairs minimally increases the thermoregulatory threshold for vaso- both central and peripheral thermoregulatory control during constriction during enflurane anesthesia in humans. Anesthesi- general anesthesia. Anesthesiology 1994;80:268-77. ology 1992;77:286-90. 34. Sessler DI, Lee KA, McGuire J. Isoflurane anesthesia and circa- 11. Bissonnette B, Sessler DI. Thermoregulatory thresholds for va- dian temperature cycles. Anesthesiology 1991;75:985-9. soconstriction in pediatric patients anesthetized with halothane 35. Lopez M, Sessler DI, Walter K, et al. Rate and gender depen- or halothane and caudal bupivacaine. Anesthesiology 1992;76: dence of the sweating, vasoconstriction, and shivering thresh- 387-92. olds in humans. Anesthesiology 1994;80:780-8. 12. Sessler DI, McGuire J, Moayeri A, et al. Isoflurane-induced 36. Sessler DI, Schroeder M. Heat loss in humans covered with vasodilation minimally increases cutaneous heat loss. Anesthe- cotton hospital blankets. Anesth Analg 1993;77:73-7. siology 1991;74:226-32. 37. Kurz A, Kurz M, Poeschl G, et al. Forced-air warming maintains 13. Hynson J, Sessler DI. Intraoperative warming therapies: a com- intraoperative normothermia better than circulating-water mat- parison of three devices. J Clin Anesth 1992;4:194-9. tresses. Anesth Analg 1993;77:89-95. 14. Sessler DI, Hynson J, McGuire J, et al. Thermoregulatory vaso- 38. Sessler DI, Moayeri A. Skin-surface warming: heat flux and constriction during isoflurane anesthesia minimally decreases central temperature. Anesthesiology 1990;73:218-24. heat loss. Anesthesiology 1992;76:670-5. 39. Cole D, Kalichman M, Shapiro H, et al. The nonlinear potency of 15. Belani K, Sessler DI, Sessler AM, et al. Leg heat content contin- sub-MAC concentrations of nitrous oxide in decreasing the ues to decrease during the core temperature plateau in humans. anesthetic requirement of enflurane, halothane, and isoflurane Anesthesiology 1993;78:856-63. in rats. Anesthesiology 1990;73:93-9. 16. Sessler DI, Olofsson CI, Rubinstein EH. The thermoregulatory 40. Gonsowski CT, Eger EI II. Nitrous oxide minimum alveolar threshold in humans during nitrous oxide-fentanyl anesthesia. anesthetic concentration in rats is greater than previously re- Anesthesiology 1988;69:357-64. ported. Anesth Analg 1994;79:710-2. 17. Passias TC, Mekjavic IB, Eiken 0. The effect of 30% nitrous 41. Nakajima R, Nakajima Y, Ikeda. Minimum alveolar concentra- oxide on thermoregulatory responses in humans during hypo- tion of sevoflurane in elderly patients. Br J Anaesth 1993;70: thermia. Anesthesiology 1992;76:550-9. 273-5. 18. Mekjavic IB, Sundberg CJ. Human temperature regulation dur- 42. Scheller MS, Saidman LJ, Partridge BL. MAC of sevoflurane in ing narcosis induced by inhalation of 30% nitrous oxide. J Appl humans and the New Zealand White rabbit. Can J Anaesth Physiol 1992;73:2246-54. 1988;35:153-6.