Original Contributions Effectiveness of Forced Air Warming After Pediatric Cardiac Surgery Employing Hypothermic Circulatory Arrest Without Cardiopulmonary Bypass Dmitri V. Guvakov, MD,* Albert T. Cheung, MD,† Stuart J. Weiss, PhD, MD,† Nikolai B. Kalinin, MD,‡ Nikita O. Fedorenko, MD,‡ Anatoli V. Shunkin, PhD, MD,† Vladimir N. Lomivorotov, PhD, MD,§ Alexander M. Karaskov, PhD, MD§ Departments of Anesthesiology and Surgery of the Novosibirsk Institute of Circulatory Pathology, Novosibirsk, Russia, and the Department of Anesthesiology of the University of Pennsylvania, Philadelphia, PA
*NIH Research Fellow in Anesthesiology †Associate Professor ‡Attending Anesthesiologist §Professor Address correspondence to Dr. Dmitri Guvakov, Department of Anesthesia, Dulles 7, University of Pennsylvania, 3400 Spruce Street, Philadelphia, PA 19104-4823, USA. E-mail:
[email protected] The study was supported in part by an equipment grant from Augustine Medical, Inc., Eden Prairie, MN, and a grant from the Office of International Medical Programs at the University of Pennsylvania, Philadelphia, PA. Received March 16, 2000; revised manuscript accepted for publication September 6, 2000.
Study Objective: To evaluate the effectiveness of forced-air warming compared to radiant warming in pediatric cardiac surgical patients recovering from moderate hypothermia after perfusionless deep hypothermic circulatory arrest. Design: Prospective unblinded study. Setting: Teaching hospitals. Patients: 24 pediatric cardiac surgical patients. Intervention: Noncyanotic patients undergoing repair of atrial or ventricular septal defects were cooled by topical application of ice and rewarmed initially in the operating room by warm saline lavage of the pleural cavities. On arrival at the intensive care unit (ICU), patients were warmed by forced air (n ⫽ 13) or radiant heat (n ⫽ 11). The time, heart rate, and blood pressure at each 0.5°C increase in rectal temperature were measured until normothermia (36.5°C) to determine the instantaneous rewarming rate. Measurements and Main Results: Baseline characteristics were not different in the two groups. The mean (⫾ SD) age was 5.6 ⫾ 3.4 years, weight was 20 ⫾ 8 kg, esophageal temperature for circulatory arrest was 25.7 ⫾ 1.3°C, and duration of circulatory arrest was 25 ⫾ 11 minutes. The mean core temperature on arrival at the ICU was 29.9 ⫾ 1.3°C and ranged from 26.1 to 31.5°C. The mean rewarming rate for each 0.5°C was greater (p ⬍ 0.05) for forced-air (2.43 ⫾ 1.14°C/hr) than radiant heat (2.16 ⫾ 1.02°C/hr). At core temperatures ⬍33°C, the rewarming rate for forced-air was 2.04 ⫾ 0.84°C/hr and radiant heat was 1.68 ⫾ 0.84°C/hr (p ⬍ 0.05). At core temperatures
Journal of Clinical Anesthesia 12:519 –524, 2000 © 2000 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
0952-8180/00/$–see front matter PII S9052-8180(00)00210-5
Original Contributions
ⱖ33°C, the rewarming rate for forced air was 2.76 ⫾ 1.20°C/hr and radiant heat was 2.46 ⫾ 1.08°C/min (p ⫽ 0.07). Significant determinants of the rewarming rate in a multivariate regression model were age (p ⬍ 0.001), temperature (p ⬍ 0.05), time after arrival to the intensive care unit (p ⬍ 0.05), pulse pressure (p ⬍ 0.05) and warming device (p ⬍ 0.001). The duration of ventilatory support and ICU length of stay was not different in the two groups. Conclusions: Both forced-air and radiant heat were effective for rewarming moderately hypothermic pediatric patients. When core temperature was less than 33°C, the instantaneous rewarming rate by forced air was 21% faster than by radiant heat. © 2000 by Elsevier Science Inc. Keywords: congenital heart defects, deep hypothermic circulatory arrest, forced air rewarming, hypothermia, pediatric cardiac surgery, radiant infrared heating, temperature.
Introduction Deliberate hypothermia is required to decrease the risk of brain injury in cardiac operations that employ circulatory arrest. At the Novosibirsk Institute of Circulatory Pathology in Novosibirsk, Russia, a technique for cardiac operations that employ hypothermic circulatory arrest without using cardiopulmonary bypass (CPB) based on the technique originally described by Lewis and Taufic has been refined and is still used routinely.1,2 Although operations can be performed using CPB at the Novosibirsk Institue of Circulatory Pathology, the technique of perfusionless hypothermic circulatory arrest is considered the standard of care for the operative repair of noncyanotic congenital cardiac defects in selected patients.3 The technique involves topical cooling of anesthetized patients to a core temperature in the range of 25°C followed by the initiation of circulatory arrest by inflow occlusion of the superior vena cava (SVC) and inferior vena cava (IVC) to permit operation. Accumulated clinical experience and technical refinements in the conduct of perfusionless deep hypothermic circulatory arrest for the operative repair of noncyanotic congenital cardiac defects at the Novosibirsk Institute of Circulatory Pathology have demonstrated clinical outcomes that were comparable to operations performed using CPB.2,3 The technique of perfusionless deep hypothermic circulatory arrest continues to be used at the Novosibirsk Institute of Circulatory Pathology because it is less expensive than techniques that employ extracorporeal perfusion, the operation is simplified, and there is a decreased need for allogeneic blood products. In contrast to techniques employing CPB that incorporate a heat exchanger, rewarming patients after perfusionless hypothermic circulatory arrest has been problematic. Initial rewarming in the operating room (OR) by warm saline lavage instilled into the pleural cavities was sufficient to restore circulatory function, but patients typically were moderately hypothermic on arrival at the postoperative intensive care unit (ICU). Final rewarming to normothermia must be completed in the ICU. Devices that 520
J. Clin. Anesth., vol. 12, November 2000
Table 1. Patient Characteristics by Group
Parameter Age (yrs) Weight (kg) Height (cm) Diagnosis
Group 1 Forced-Air (n ⴝ 13)
Group 2 Radiant Heating (n ⴝ 11)
6.4 ⫾ 4.2 22.2 ⫾ 9.8 114 ⫾ 24 ASD-II (n ⫽ 5) VSD (n ⫽ 7) ASD⫹VSD (n ⫽ 1)
4.8 ⫾ 2 16.8 ⫾ 4.6 105 ⫾ 17 ASD-II (n ⫽ 6) VSD (n ⫽ 4) CoAo ⫹ASD-II (n ⫽ 1)
p-value* 0.22 0.09 0.3
Note: Values are means ⫾ SD. ASD-II ⫽ atrial septal defect secundum, VSD ⫽ ventricular septal defect, CoAo ⫽ coarctation of aorta. * p-values are for Group 1 versus Group 2.
have been tried included wool blankets, electric blankets, inhalation of warmed humidified gases, and radiant infrared heaters. At present, radiant infrared heating is the predominant method being used. Forced-air warming devices have been demonstrated to be effective for the treatment and prevention of perioperative hypothermia in noncardiac surgical patients, but have not been tested for use in this specific patient population.4 –7 The objective of this study was to compare the effectiveness of forced-air warming to radiant heating for restoring moderately hypothermic pediatric cardiac surgical patients to normothermia after perfusionless hypothermic circulatory arrest. The hypothesis that rewarming rate was faster using forced-air warming was tested.
Materials and Methods The study was designed as a prospective unblinded comparison of forced-air warming vs. radiant infrared heating for the treatment of moderate postoperative hypothermia in pediatric cardiac surgical patients operated on using perfusionless hypothermic circulatory arrest. After obtaining written informed consent, 24 patients were enrolled into a protocol approved by the Hospital of the University of Pennsylvania and the Novosibirsk Institute of Circulatory Pathology Institutional Review Boards. Patients entered into the study were assigned to either forced-air rewarming (Group 1, n ⫽ 13) or radiant infrared heating (Group 2, n ⫽ 11). Patients were eligible for enrollment into the study if they were scheduled to undergo a first-time elective surgical repair of a noncyanotic congenital cardiac defect using perfusionless hypothermic circulatory arrest. In addition, patients had to be between 1 and 20 years of age, between 12 kg and 65 kg in body weight, have a preoperative left ventricular ejection fraction greater than 50%, and have a hypothermic circulatory arrest period that was anticipated to be less than 60 minutes (Table 1).
Intraoperative Management All operations were performed using the method of perfusionless hypothermic circulatory arrest.2,3 After in-
External rewarming after perfusionless HCA: Guvakov et al.
duction of general anesthesia and tracheal intubation, anesthesia was maintained with diethyl ether (1– 6 vol %), morphine 0.5 mg/kg, and nondepolarizing neuromuscular blockade with pipecuronium bromide 0.1 mg/kg (Organon Inc, West Orange, NJ). During operation, core temperature was measured using a temperature probe with its tip positioned in the midesophagus. Before cooling, low molecular weight dextran 10 mL/kg intravenously (IV) (Reopoliglukin OAO “Kraspharma” Krsanoyarsk, Russia) was administered for hemodilution and heparin 50 IU/kg IV was administered for anticoagulation. Cooling was initiated by the topical application of shaved ice to the head and body. At an esophageal temperature of 30°C, ice was removed from the body, but left on the head. At an esophageal temperature of 24°C to 26°C, sodium bicarbonate 2 mEq/kg and heparin 200 IU/kg was injected directly into the left ventricle. Two minutes after heparin administration, hypothermic circulatory arrest was initiated by sequentially occluding the SVC, IVC, and aorta. After completion of the surgical repair, ice was removed from the head and the aorta was unclamped. Circulatory function was restored by manual cardiac massage, defibrillation, and electrical cardiac pacing (EKS01ELJ, 210766 PS, Tomsk, Russia) when necessary. All patients were initially rewarmed by the repeated instillation of sterile warm (38°C to 40°C) normal saline lavage into the pleural cavities until the esophageal temperature remained ⱖ 33°C after pleural lavage was discontinued.1,3,8 Inspired gases and IV fluids administered in the OR were not warmed. After the neutralization of heparin with protamine, and closure of the surgical incision, patients were immediately transported to the cardiothoracic intensive care unit (ICU).
Postoperative Management Immediately on arrival at the ICU, postoperative active rewarming was initiated by the application of a forced-air warming blanket attached to a forced-air warming unit (500/OR, Augustine Medical Inc., Eden Prairie, MN) set at 42°C (Group 1) or the use of a radiant infrared wave heater (Ameda 8835, Ameda AG, Huenenberg, Switzerland or Drager 4200, Telford, PA) positioned 80 cm above the patient (Group 2) set for maximum heat delivery (skin temperature assumed to be ⱕ32°C throughout the period of active warming). All patients were mechanically ventilated (Babylog or Evita, Drager) with warming of the inspired gases to 38°C (Aquapor, Drager). Core temperature was measured with a rectal temperature probe. During active rewarming, patients were administered acetaminophen 200 mg or aspirin 300 mg when the rectal temperature reached 35.2°C to 35.5°C. Diazepam 0.1 to 0.5 mg/kg was administered for shivering or sedation. The rectal temperature was monitored continuously from the time of arrival at the ICU (t ⫽ 0 min). The instantaneous rewarming rate was determined from the elapsed time it took for each incremental 0.5°C increase in rectal temperature and expressed as dT/dt (°C/h). Heart rate (HR), noninvasive blood pressure (BP), and urine
output were also recorded at each 0.5°C increase in rectal temperature. Measurements were continued until the rectal temperature reached 36.5°C, when active warming was discontinued. Patients were monitored to assess if core temperature decreased after the discontinuation of active warming, the duration of postoperative mechanical ventilatory support, need for vasopressor or inotropic support, ICU length of stay, and presence of adverse events or complications.
Statistical Analysis The independent or two-sample t-test was used to test for differences in the instantaneous rewarming rate, mean arterial pressures (MAPs), and HR in the two groups of patients. Because the risk of hemodynamic instability and cardiac dysrhythmias was believed to be greater at core temperature ⬍33°C, the mean instantaneous rewarming rate for rectal temperatures ⬍33°C and for rectal temperatures ⱖ 33°C were compared for the two groups of patients. Linear regression analysis was used to determine the relationship between individual patient variables and the instantaneous rewarming rate. Variables with a p-value ⬍ 0.1 in the univariate regression analysis were entered in a multiple stepwise linear regression analysis to determine the effect of factors such as patient age, height, weight, body surface area, and type of active warming device on the instantaneous rewarming rate. In the regression models, use of a forced-air warming unit was assigned a numerical value of 1, and use of a radiant warmer was assigned a numerical value of 2. A p-value less than 0.05 was considered significant.
Results All 24 patients entered into the study completed the protocol. Baseline characteristics of patients assigned to forced-air warming, (Group 1, n ⫽ 13) and radiant heat (Group 2, n ⫽ 11) were not different (Table 1). The patients studied had an age range of 1.6 yr to 13 yr and a weight range of 12.0 kg to 42.5 kg. The esophageal temperature at the initiation of hypothermic circulatory arrest ranged from 23.0°C to 29.3°C, and duration of hypothermic circulatory arrest ranged from 11 min to 49 min. The initial rectal temperature of patients on arrival at the ICU ranged from 28.5°C to 31.5°C and was not different in the two groups (Table 2). There was no difference in total dose of diazepam administered to either group. All patients were cared for in the same ICU and with an ambient temperature set at 24 ⫾ 1°C. The mean (⫾SD) instantaneous rewarming rate averaged over the entire rewarming period for patients in Group 1 who underwent forced-air warming was 2.43 ⫾ 1.14°C/hr and was greater (p ⬍ 0.05) that the mean instantaneous rewarming rate of 2.16 ⫾ 1.02°C/hr measured in Group 2 who underwent radiant heating (Table 3). For patients with rectal temperatures ⬍33°C, the mean instantaneous rewarming rate for Group 1 was 2.04 ⫾ 0.84°C/hr and 1.68 ⫾ 0.84°C/hr for Group 2 (p ⬍ 0.05). For patients with rectal temperature ⱖ33°C, the mean J. Clin. Anesth., vol. 12, November 2000
521
Original Contributions
Table 4. Factors that Determined the Instantaneous Rewarming Rate in Univariate Regression Analysis
Table 2. Intraoperative Parameters by Group
Parameter Duration of circulatory arrest (min) Esophageal temperature at start HCA (°C) Rectal temperature at arrival to ICU (°C)
Group 1 Forced-Air (n ⴝ 13)
Group 2 Radiant Heating (n ⴝ 11)
p-value*
28.1 ⫾ 10.4
21.7 ⫾ 10.4
⬎0.14
25.4 ⫾ 1.2
26.2 ⫾ 1.5
⬎0.16
29.6 ⫾ 1.5
30.4 ⫾ 0.7
⬎0.1
Factor
Note: Values are means ⫾ SD. HCA ⫽ hypothermic circulatory arrest, ICU ⫽ intensive care unit. * p-values are for Group 1 versus Group 2.
instantaneous rewarming rate was 2.76 ⫾ 1.20°C/hr for Group 1 and 2.46 ⫾ 1.08°C/hr for Group 2 (p ⫽ 0.08). Factors that were significant determinants of the rewarming rate in the univariate regression analysis were the type of warming device (p ⬍ 0.05), height (p ⬍ 0.01), weight (p ⬍ 0.002), body surface area (p ⬍ 0.01), age (p ⬍ 0.01), time after arrival at the ICU (p ⬍ 0.02), and the rectal temperature (p ⬍ 0.01) (Table 4). Factors that were significant determinants of the instantaneous rewarming rate in the multiple linear regression model were patient age (p ⬍ 0.001), the type of device used for active warming (p ⬍ 0.001), the rectal temperature of the patient at the time of measurement (p ⬍ 0.04), the elapsed time after arrival to ICU when the measurement was obtained (p ⬍ 0.05), and the pulse pressure (p ⬍ 0.05) (Table 5). No patients in either group exhibited hemodynamic instability, suffered injuries, or had adverse effects related to the active rewarming protocol. There were no significant differences in HR or MAP during the rewarming period in the two groups (Figure 1). Core temperature afterdrop following the discontinuation of active warming was not was detected in either group. The mean (⫾SD) duration of postoperative mechanical ventilatory support was 9.7 ⫾ 2.2 hours and was not different in the two
Warming device Height Weight Body surface area Pulse pressure Age Time after arrival at ICU Rectal temperature
Regression coefficient
Standard Error
p-value*
0.004 °C/min 0.2 cm/°C/ min 0.409 kg/°C/ min 0.016 m2/°C/ min 0.18 mm Hg/ °C/min 0.001 months/ °C/mins 0.37 min/°C/ min 0.002 (°C/°C/ min)
0.002 0.49
0.048 ⬍0.01
0.127
0.002
0.05
⬍0.01
0.119
0.12
0.312
⬍0.01
0.163 0.479
0.023 ⬍0.01
Note: Forced-air rewarming was assigned a numerical value of 1 and radiant heating was assigned a numerical value of 2 in the regression model. * p-values are for the linear regression.
groups. The ICU length of stay was also not different for the two groups. No patients in either group had postoperative neurologic deficits detected by routine postoperative neurologic assessment. All patients were discharged from the hospital without complications.
Discussion Pediatric patients undergoing repair of noncyanotic congential heart lesions using the technique of perfusionless hypothermic circulatory arrest were moderately hypother-
Table 3. Mean Instantaneous Rewarming Rate by Group
Parameter Rewarming rate (°C/hr) Rewarming rate T ⬍ 33 (°C/hr) Rewarming rate T ⱖ 33 (°C/hr)
Group 1 Forced-Air (n ⴝ 13)
Group 2 Radiant Heating (n ⴝ 11)
p-value*
2.43 ⫾ 1.14
2.16 ⫾ 1.02
⬍0.05
2.04 ⫾ 0.84
1.68 ⫾ 0.84
⬍0.05
2.76 ⫾ 1.20
2.46 ⫾ 1.08
0.08
Note: Values are means ⫾ SD. T ⫽ rectal temperature. * p-values are for Group 1 versus Group 2. 522
J. Clin. Anesth., vol. 12, November 2000
Figure 1. Mean arterial pressure (MAP) displayed by circles and heart rate (HR) displayed by triangles, according to rectal temperature (Trectal) during active rewarming in the intensive care unit. Values for group 1 are represented by the open symbols (forced-air rewarming, n ⫽ 13) and values for group 2 are represented by the closed symbols (radiant heat rewarming, n ⫽ 11). All values are means ⫾ standard error of the mean.
External rewarming after perfusionless HCA: Guvakov et al.
Table 5. Results of Multiple Regression Analysis to Determine the Factors that Predict the Instantaneous Rewarming Rate
Factor Age
*Warming device Time after arrival at ICU Pulse pressure T rectal
Regression coefficient 0.002 (months/ °C/ min) 0.008 (C/min) 0.41 (min/°C/ min) 0.23 (mmHg/ °C/min) 0.96 (T°C/°C/ min)
SE
pvalue*
0.32
⬍0.001
0.002 0.20
⬍0.001 ⬍0.004
0.11
0.04
0.46
0.038
SE ⫽ standard error, ICU ⫽ intensive care unit, T rectal ⫽ rectal temperature. * p-values are for the model: rewarming rate (°CC/min) ⫽ constant ⫹ a1x1 ⫹ a2x2 ⫹ anxn, where x1, x2 . . . xn are clinical variables and a1, a2 . . . an are the corresponding coefficients. Forced air warming was assigned a numerical value of 1 and radiant heating was assigned a numerical value of 2 in the regression model.
mic in the immediate postoperative period (Table 2). The use of forced-air warming was as effective as radiant heating to restore normothermia without risk of temperature afterdrop or paradoxical core cooling once active rewarming was initiated.9 The instantaneous rate of rewarming averaged over the period of active warming was greater with forced air warming compared to radiant warming. At core temperatures ⬍33°C, when the risks associated with hypothermia were believed to be greatest, the difference in rewarming rate with forced-air and radiant rewarming was also greater. At core temperatures ⬍33°C, the rate of rewarming using forced-air warming was 21% greater than the rate that was achieved using radiant heat. After a core temperature of 33°C was reached, the rewarming rate achieved with forced-air warming was still 12% greater than that achieved using radiant heat, but the difference did not reach statistical significance. The type of rewarming device used was also found to be a significant factor that determined the rate of rewarming in the multiple regression model. The rate of increase in core temperature observed in this study using active external rewarming with forced air was similar to the rates that have been reported when the technique was applied for the treatment of adult victims of accidental hypothermia and human volunteers subjected to cold water immersion.4,7,10,11 Experience with forcedair rewarming in hypothermic pediatric patients is limited. Initial rewarming during operation by warm saline lavage of the pleural cavities was effective for acutely increasing the core temperature by up to 8°C to enable restoration of circulatory function after circulatory arrest, but was associated with a temperature afterdrop that averaged 3°C.8,12,13 This observed afterdrop or decrease in core temperature that occurred after discontinuation of internal warming by warm saline lavage could be explained by ongoing conductive heat transfer from the core compart-
ment to the cooler peripheral tissue compartment during thermal equilibration. Temperature afterdrop has been described when the skin was exposed to colder ambient temperatures or when active external warming was not used.9,14 –16 The use of external warming by either forcedair or radiant heat was effective in preventing further afterdrop. Patient age was found to be an important determinant of the rate of rewarming. The relationship between patient age and rate of rewarming may be explained by the correlation between age and body weight, height, and surface area in the relatively heterogenous pediatric population that was studied. The relationship predicted a slower rate of rewarming in older or larger patients. The whole body heat deficit in older or larger patients compared to younger or smaller patients would be expected to be greater at similar core temperatures because of a larger peripheral compartment that needed to be rewarmed.16,17 In addition, the efficiency of heat transfer from the periphery to core during external warming may have been greater in younger patients who had a greater body surface area in relation to their body mass. The efficiency of conductive heat transfer from the peripheral compartment to the core during external rewarming also depends on skin and muscle blood flow.17 Thermoregulatory vasoconstriction is a normal response to hypothermia. Vasoconstriction of vessels supplying the skin would explain why the rewarming rate was less when the patient first arrived in the ICU and increased over time after the initiation of active rewarming. Increases in skin temperature as a consequence of external warming would induce local vasodilation and serve to enhance the efficiency of conductive heat transfer to the core.16 Once the thermoregulatory vasoconstriction threshold of approximately 35°C has been exceeded, the efficiency of heat transfer would be expected to increase further resulting in a greater rewarming rate at temperatures closer to normothermia.4 For the same reasons, the pulse pressure, which may be an indirect indicator of skin and muscle vasoconstriction,14 was found to be a relatively minor but significant factor that predicted the rate of rewarming. Studying a larger number of patients, monitoring temperature from a number of different sites, and using invasive hemodynamic monitoring may provide further insights into the importance of physiologic mechanisms that affect the efficiency and rate of rewarming. Although both forced-air and radiant heat was effective in restoring normothermia without adverse events, the initial rate of rewarming was greater when forced-air warming was used. The greater rewarming rate achieved with forced-air warming may be related to patient shielding with the disposable cover to minimize heat losses due to convection, radiation, and evaporation in combination with ensuring uniform temperatures over the entire body.4,7 Because of the limited size of the study, no differences in outcome or clinical benefit could be attributed to the faster rewarming rate achieved with the use of forced-air warming. The type of rewarming device used did not produce differences in the cardiovascular responses to rewarming. During the course of the study, the J. Clin. Anesth., vol. 12, November 2000
523
Original Contributions
forced-air warming device was assessed subjectively to be effective, convenient, easy to use, and quickly endorsed by practicing clinicians. The duration of mechanical ventilation and ICU length of stay in the absence of medical complications were not different between groups, and were believed to have been dictated more by established patterns of practice than the patient temperature or the relatively minor differences in rewarming rate. In conclusion, forced-air warming was effective for the treatment of moderate postoperative hypothermia with an average core temperature of 30°C in pediatric cardiac surgical patients. The rewarming rate achieved with forced-air warming was greater than that observed with radiant infrared heat, especially when the core temperature of the patient was less than 33°C.
Acknowledgements The authors are grateful to Dr. Elena N. Pilak, from the Novosibirsk Institute of Circulatory Pathology, for her support and helpful suggestions.
References 1. Lewis P, Taufic M: Closure of atrial septal defects with the aid of hypothermia: experimental accomplishment and our successful case. Surgery 1953;33:52–9. 2. Litasova EE, Lomivorotov VN, Gorbatich JN, et al: Deep hypothermia without extracorporeal circulation in surgery of congenital cardiac defects. J Cardiovasc Surg 1994;35:45–52. 3. Litasova EE, Lomivorotov VN: Hypothermic protection (26 – 25°C) without perfusion cooling for surgery of congenital cardiac defects using prolonged occlusion. Thorax 1988;43:206 –11. 4. Goheen MS, Duchrame MB, Kenny GP, et al: Efficacy of forcedair and inhalation rewarming by using a human model for severe hypothermia. J Appl Physiol 1997;83:1635– 40.
524
J. Clin. Anesth., vol. 12, November 2000
5. Mair P, Kornberger E: Forced hot air rewarming in hypothermic patients with cardiocirculatory arrest? Acta Anaesthesiol Scand 1998;42:1018 –20. 6. Kurz A, Sessler DI, Lenhardt R: Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. N Engl J Med 1996;334:1209 –15. 7. Steele MT, Nelson MJ, Sessler DI, et al: Forced air speeds rewarming in accidental hypothermia. Ann Emerg Med 1996;27: 479 – 84. 8. Winegard C: Successful treatment of severe hypothermia and prolonged cardiac arrest with closed thoracic cavity lavage. J Emerg Med 1997;15:629 –32. 9. Geisbrecht GG, Bristow GK: A second postcooling afterdrop: more evidence for a convective mechanism. J Appl Physiol 1992; 73:1253– 8. 10. Miller JW, Danzl DF, Thomas DM: Urban accidental hypothermia: 135 cases. Ann Emerg Med 1992;63:1065–9. 11. Geisbrecht GG, Goheen MSL, Johnston CE, et al: Inhibition of shivering increases core temperature afterdrop and attenuates rewarming in hypothermic humans. J Appl Physiol 1997;83: 1630 – 4. 12. Sanford MM: Rewarming cardiac surgical patients: warm water vs. warm air. Am J Crit Care 1997;6:39 – 45. 13. Kangas E, Niemela H, Kojo N: Treatment of hypothermic circulatory arrest with thoracotomy and pleural lavage. Ann Chir Gynaecol 1994;83:258 – 60. 14. Tveita T, Ytrehus K, Skandfer M, et al: Changes in blood flow distribution and capillary function after deep hypothermia in rat. Can J Physiol Pharm 1996;74:376 – 81. 15. Geisbrecht GG, Schroeder M, Bistrow GK: Treatment of mild immersion hypothermia by forced-air warming. Aviat Space Environ Med 1994;65:803– 8. 16. Deakin CD, Petley GW, Smith D: Pharmacological vasodilatation improves efficiency of rewarming from hypothermic cardiopulmonary bypass. Br J Anaesth 1998;81:147–51. 17. Baumgardner JE, Baranov D, Smith DS, et al.: The effectiveness of rapidly infused IV fluids for inducing moderate hypothermia in neurosurgical patients. Anesth Analg 1999;89:163–9.