Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

PART ONE – Basic Principles in Pediatric Anesthesia

Chapter 5 – Thermoregulation: Physiology and Perioperative Disturbances

Igor Luginbuehl,Bruno Bissonnette,
Peter J. Davis

  

 

Temperature Monitoring, 154

  

 

Physiology of Thermal Regulation, 155

  

 

Afferent Thermal Sensing, 156

  

 

Central Regulation, 156

  

 

Efferent Response, 157

  

 

Thermal Regulation in the Newborn,157

  

 

Heat Loss Mechanisms, 158

  

 

Conduction, 159

  

 

Radiation, 159

  

 

Convection, 160

  

 

Evaporation, 160

  

 

Heat Generation, 161

  

 

Nonshivering Thermogenesis,161

  

 

Shivering Thermogenesis, 162

  

 

Dietary Thermogenesis, 162

  

 

Effect of Anesthesia on Thermoregulation, 163

  

 

Thermoregulation and General Anesthesia, 163

  

 

Thermoregulation and Regional Anesthesia,166

  

 

Anesthesia and Hypothermia, 166

  

 

Internal Redistribution, 166

  

 

Thermal Imbalance, 167

  

 

Thermal Steady State (Plateau or Rewarming),168

  

 

Anesthesia and Hyperthermia, 169

  

 

Prevention of Hypothermia, 170

  

 

Operating Room Temperature,171

  

 

Radiant Heaters, 171

  

 

Reflecting Blankets, 171

  

 

Skin Warming Devices, 171

  

 

Warming Mattresses, 172

  

 

Warming Fluids, 172

  

 

Humidified and Heated Gases,172

  

 

Transportation, 173

  

 

Summary, 173

One of many physiologic adaptations required for human survival is the ability to establish and maintain a constant core body temperature. The significance of thermal regulation for neonates was appreciated as early as the 1900s when Budin noted a significant difference in the mortality rate among infants with different body temperatures (1907). Other investigators have confirmed the importance of thermal stability in the adaptive process and further elucidated the mechanisms by which infants and children behave as homeotherms ( Silverman and Blanc, 1957 ; Cross et al., 1958 ; Silverman et al., 1958 ; Bruck, 1961 ). By definition, a homeothermic organism maintains a constant core temperature despite changes in the ambient temperature. Only a few other physiologic parameters are as vigorously and effectively controlled as the central temperature. The central body temperature in humans refers to the temperature of the vessel-rich group organs (brain, heart, lungs, liver, and kidneys) and under normal circumstances is maintained within ±0.2°C of its set point of 37.0°C. This so-called interthreshold range defines the limits, within which no thermoregulatory effector responses are triggered and the human organism in fact behaves in a poikilothermic manner. The musculoskeletal system makes up the major part of the peripheral compartment, which can be seen as a dynamic buffer in the thermoregulatory system, whereas the skin represents the shell compartment, which acts as a barrier to the environment.

Although temperature control is subjected to a circadian rhythm, which starts in the first months of life, nocturnal oscillations in infants ( Brown et al., 1992 ) and a monthly rhythm in fertile women (due to a higher set point temperature in the luteal phase of the menstrual cycle) ( Hardy, 1961 ) are found. The control within these rhythms is very tight, however, and accomplished by a delicate and effective system that balances heat production and heat loss. Despite this effective regulatory system, the ability of the organism to dissipate or generate heat by regulating skin blood flow, sweat production, minute ventilation, and metabolism is often overwhelmed by external factors. Anesthesia and surgery have a powerful effect on this sophisticated thermoregulatory system, and minor changes in body temperature may result in cellular and tissue dysfunction, explaining the need not only for a regulation within so narrow limits but also for perioperative temperature monitoring.

Accidental hypothermia is a frequent occurrence in patients of any age undergoing anesthesia and surgery. Hence, it is unfortunately often accepted as a consequence of the surgical procedure. This common occurrence led Pickering to comment that “The most effective means of cooling a man is to give him an anesthetic” (1958).

This chapter discusses the principles of thermoregulation in infants and children and the relative merits of different anatomic sites of temperature monitoring. In addition, we review the influence of anesthetic agents on the child's thermoregulatory system, the physiologic consequences of hypothermia and hyperthermia, and the techniques to prevent perioperative hypothermia.

▪ TEMPERATURE MONITORING

The unit for temperature in the Système Internationale is Kelvin (K), where 0 K = -273.15°C. Most countries measure temperature in Celsius degrees and a few measure in Fahrenheit degrees. The following formulas can be used to convert them from one unit into the other:(5.1)  (5.2)  

To detect perioperative changes in temperature, an appropriate measurement of temperature is mandatory. Guidelines of the American Society of Anesthesiologists require that one method for measuring body temperature during anesthesia be available.

Mercury-in-glass thermometers were standard in earlier times; currently, the most common thermometers are thermocouples and thermistors. A thermocouple consists of two different metals, often copper and constantan (a copper-nickel-manganese-iron alloy). The principle is based on the Seebeck effect, which states a small current is produced at any junction of two different metals from the thermoelectric series. The magnitude of this voltage is temperature dependent and therefore can be used for temperature measuring.

An exponential change in electrical resistance with temperature is the principle of the thermistor type of thermometer, which is a semiconductor resistor that consists of a tiny piece of metal (copper, nickel, manganese, or cobalt). The change in resistance is used to measure temperature. Both thermocouple and thermistor probes are inexpensive and sufficiently accurate for clinical purposes.

Temperature-sensitive liquid crystals have been used to measure skin temperature. Although these devices are easy and convenient to handle, they generally do not meet the accuracy criteria for clinical use, because they can be influenced by changes in body temperature and skin blood flow ( Leon et al., 1990 ; MacKenzie and Asbury, 1994 ). Simply adding a constant correcting value (e.g., 2.2°C) to an arbitrary skin temperature such as the forehead is highly unlikely to provide a reliable estimate of central temperature ( Burgess et al., 1978 ; Leon et al., 1990 ).

Body temperature varies widely throughout the body. On the one hand, core tissues tend to maintain a constant temperature due to high perfusion. On the other hand, peripheral tissues have much lower and less uniform temperatures ( Colin et al., 1971 ), which may differ several degrees within a short distance from each other.

Core temperature is difficult to define. Benzinger suggested that core temperature is the temperature of the hypothalamus and that tympanic membrane probes reliably indicate core temperature (1969). No physiologic evidence suggests that hypothalamic temperature precisely represents central temperature, however. Core temperature is the most important thermoregulatory controller and therefore of most clinical interest. Core temperature-measuring sites available for clinical use are tympanic membrane, nasopharynx, distal esophagus, pulmonary artery, and, with some limitations, bladder and rectum. These sites usually provide equal readings in awake and anesthetized humans undergoing noncardiac surgery ( Cork et al., 1983 ). Different monitoring sites under certain conditions may represent different temperatures. The physiologic and clinical significance of such differences may vary.

Body temperature can be monitored at numerous anatomic sites. The precision and accuracy of measurements at these sites have been studied ( Cork et al., 1983 ; Bissonnette et al., 1989a ), and each site has its advantages and disadvantages. The site of temperature monitoring should reflect core temperature and be associated with none or only minimal morbidity.

Skin temperature offers little as a reflection of core temperature ( Lacoumenta and Hall, 1984 ; Bissonnette et al., 1989a ). Skin temperature varies depending on the site of monitoring. A number of investigators have suggested monitoring between 4 and 15 sites, using both weighted and unweighted formulas to accurately describe mean skin temperature ( Ramanathan, 1964 ; Colin et al., 1971 ;Shanks, 1975 ; Puhakka et al., 1994 ). To be of clinical value, skin temperature must accurately reflect central temperature in the perioperative setting such as mild hypothermia and malignant hyperthermia. It is unlikely that skin temperature correlates well with central temperature during the early phase of malignant hyperthermia, because the circulating catecholamine concentrations may be up to 20 times higher than normal and significantly affect skin perfusion ( Sessler, 1986 ; Sessler and Moayeri, 1990 ).

Tympanic membrane temperature has been suggested to be the most ideal temperature-monitoring site. To reflect tympanic temperature, it is not necessary for the temperature probe to be in contact with the tympanic membrane, as long as the external auditory canal is sealed by the probe, thereby allowing the air column trapped between the probe and the tympanic membrane to reach a steady state. In the initial postoperative period in infants and children after cardiac reconstructive surgery, tympanic temperature does not correlate well with brain temperature ( Bissonnette et al., 2000 ) and therefore does not provide an accurate estimate of central body temperature ( Muma et al., 1991 ). Because of the difficulty associated with obtaining appropriate-sized thermistors and reports of tympanic membrane perforation, its clinical use has been discouraged.

Nasopharyngeal temperature probes can accurately reflect core temperature if placed properly (i.e., placing the tip of the temperature probe in the posterior nasopharynx close to the soft palate). This should provide a good estimate of the hypothalamic temperature. If used in combination with uncuffed tubes with a moderate to large air leak, the resulting airflow may lead to inaccurate reading. Slight and self-limiting bleeding from the nose is not uncommon (especially in children with large adenoids), and its preclusion in mask anesthesia has limited its routine use. In contrast, oral temperature is considered less adequate ( Cork et al., 1983 ) and therefore is not recommended as an accurate intraoperative temperature-monitoring site.

Esophageal temperature probes are often combined with an esophageal stethoscope, which makes this site particularly attractive in the pediatric population. In infants and children, and in cachetic patients, the thermal insulation is minimal between the tracheobronchial tree and the esophagus. The respiratory gas flow, therefore, may result in erroneous temperature readings ( Bissonnette et al., 1989a ), especially when the respiratory gas flow is high and its temperature differs significantly from the body temperature. Furthermore, central temperature is measured only if the tip of the probe is placed in the distal third of the esophagus at the point where the heart sounds are the loudest ( Bissonnette et al., 1989a ; Stoen and Sessler, 1990 ). In patients with endotracheal intubation, esophageal temperature is more reliable than rectal temperature and more practical than tympanic temperature.

Axillary temperature probes are notoriously unreliable guides to core temperature, because they are frequently malpositioned. It is not only the most commonly used but also the most convenient site for temperature monitoring. Axillary temperature has been reported to be as accurate in measuring central temperature as tympanic membrane, esophageal, and rectal temperature sites ( Bissonnette et al., 1989a ) but only when the tip of the thermometer is carefully placed over the axillary artery and the arm is closely adducted ( Bissonnette et al., 1989a ). Infusing cool solutions at high flow rates in small children on the ipsilateral side of the thermometer may result in falsely low temperature readings.

Rectal temperature monitoring bears the problems of probe insulation by feces, exposure of the probe to cooler blood returning from the legs, the influence of an open abdominal cavity during laparotomy, or bladder irrigations with either cold or warm solutions. Nonetheless, with these restrictions kept in mind, the rectal site can provide central temperature reading ( Bissonnette et al., 1989a ), and the minimal morbidity associated with its use and its ease of insertion confer major advantages. Relative contraindications for the use of a rectal probe are patients with inflammatory bowel disease, neutropenia, or thrombocytopenia and patients whose bowel or bladder is to be irrigated.

Bladder temperature is considered to be one of the most accurate sites for measuring core temperature. It has been demonstrated to be identical to pulmonary artery temperature as long as urinary output is high ( Horrow and Rosenberg, 1988 ). When urinary output is normal or less than normal, this site becomes inaccurate to represent central temperature. A pulmonary artery catheter with a distal-tip thermistor can accurately reflect pulmonary blood temperature, but due to its invasive nature, its use is limited to special situations, namely, in critically ill children.

The site or sites of temperature monitoring generally are a function of the operative procedure. For cardiac surgery patients, in whom temperatures from different body sites convey useful information, temperature is usually measured at multiple sites (e.g., rectum, bladder, esophagus, nasopharynx, tympanic membrane).

In the pediatric patient who does not have an endotracheal tube and is undergoing a short operative procedure, either rectal or axillary temperature monitoring can be used. If the child has an endotracheal tube in place, another option includes the use of a distal esophageal temperature probe.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ PHYSIOLOGY OF THERMAL REGULATION

Survival from body temperatures as low as 13.7°C has been reported ( Gilbert et al., 2000 ), whereas death resulting from protein denaturation occurs within 7°C from normality at approximately 44°C. Compared with each other, the tolerance to cold is more than three times higher than that to heat, which explains why the system to dissipate heat has to be much more effective than the cold defense system.

The thermoregulatory system is similar to other physiologic control systems in the sense that the brain uses negative feedback mechanisms to keep variations from normal values as minimal as possible (Shanks, 1975 ). The principal site of temperature regulation is the hypothalamus, which integrates afferent signals from temperature-sensitive cells found in most tissues, including other parts of the brain, spinal cord, central core tissues, respiratory tract, gastrointestinal tract, and the skin surface ( Fig. 5-1 ). The processing of thermoregulatory information occurs in three stages:

  

1.   

Afferent thermal sensing

  

2.   

Central regulation

  

3.   

Efferent response

 
 

FIGURE 5-1  Illustration of the thermoregulatory pathways with afferent information from the body being integrated in the anterior hypothalamus and triggering of efferent responses in the posterior hypothalamus.

 

 

▪ AFFERENT THERMAL SENSING

In the periphery, anatomically distinct warm and cold receptors sense the ambient temperature. The skin contains about 10 times more cold than warm receptors, acknowledging the important function of the skin in the detection of cold ( Poulos, 1981 ). Thermosensitive receptors are also found in close proximity to the great vessels, the viscera, and the abdominal wall and in the brain and the spinal cord. Each receptor type transmits the information through an afferent nerve conduction pathway. Although they originate from anatomically different nerve fibers, the speed of transmission is influenced by the intensity of the stimulus rather than the type of nerve fiber. It is well established that the rate of the change in skin temperature alters its apparent importance. Rapid changes contribute about five times as much to the central regulatory system as comparable slow changes ( Wyss et al., 1975 ). In patients not undergoing cardiopulmonary bypass, the rate of change in core temperature does not appear to substantially influence the magnitude of the provoked regulatory responses. The thermal information from cold-sensitive receptors, which have their maximal discharge rate of impulses at a temperature of 25° to 30°C, is transmitted to the preoptic area of the hypothalamus by A delta fibers.

The thermal information gathered by peripheral warm receptors, which have their maximal discharge rate at 45° to 50°C ( Pierau and Wurster, 1981 ), is carried by unmyelinated C fibers. These C fibers also detect and convey pain sensations, which explains why intense heat cannot be distinguished from severe pain ( Pierau and Wurster, 1981 ; Poulos, 1981 ). Although most ascending thermal information travels along the spinothalamic tracts in the anterolateral spinal cord, no single spinal tract is solely responsible for conveying thermal information ( Hellon, 1981 ).

▪ CENTRAL REGULATION

The anterior hypothalamus is responsible for the integration of the afferent thermal information, whereas the posterior hypothalamus controls the descending pathways to the effectors. Thermal inputs from skin surface, spinal cord, and deep body tissues are integrated in the preoptic area of the anterior hypothalamus and compared with the threshold temperatures for heat and cold. The hypothalamus then carefully regulates mechanisms for heat generation and dissipation to maintain body temperature within the narrow limits of its set point (interthreshold range).

The preoptic area of the hypothalamus contains cold- and heat-sensitive neurons, with the latter predominating by 4:1 ( Boulant and Bignall, 1973 ). This area also receives and processes nonthermic afferent information, which seems to be important in controlling the adaptive mechanisms and the behavior of the organism ( Hori and Katafuchi, 1998 ).

Local heat sensation results in increased discharge rate from these heat-sensitive neurons and leads to the activation of heat loss mechanisms. Conversely, cold-sensitive neurons in the hypothalamus respond with increased discharge rates to cooling of the preoptic area of the hypothalamus ( Boulant and Hardy, 1974 ; Boulant and Demieville, 1977 ). Thermosensitive neurons also exist in the posterior hypothalamus, the reticular formation, the medulla, the lower brainstem, and the spinal cord, although their function remains to be fully elucidated ( Guieu and Hardy, 1970 ; Simon, 1974 ; Cabanac, 1975 ;Dickenson, 1977 ).

It seems that under normal conditions, the contribution of the central thermoreceptors to thermal regulation is limited by the marked predominance of the input of peripheral receptors ( Downey et al., 1964). These central receptors take over thermoregulation if the sensory input from peripheral sensors is disrupted (e.g., central neuraxial anesthesia or spinal cord transsection), but they are less efficient compared with peripheral thermoreceptors ( Downey et al., 1967 ).

The threshold represents the central temperature for which a particular regulatory effector becomes active ( Box 5-1 ). When the integrated input from all sources exceeds the upper or falls below the lower threshold, efferent responses are initiated from the hypothalamus to maintain normal body temperature. The slope of the response intensity plotted against the difference between the thermal input temperature and the threshold temperature is called the gain of that response (i.e., the intensity of the response).

BOX 5-1 

Definition of Temperature Regulation Terms

Threshold temperature

Central temperature that elicits a regulating effect, e.g., vasoconstriction, vasodilatation, shivering, sweating, nonshivering thermogenesis

Interthreshold range

Temperature range over which no regulatory response occurs

Gain

Intensity of regulatory response

Mean body temperature

Physiologically weighted average temperature from various tissues

Nonshivering thermogenesis

Heat production (above basal metabolism) not associated with muscle activity

Shivering thermogenesis

Heat production through voluntary muscle activity

Dietary thermogenesis

Heat production through metabolism of nutrients

 

The difference between the lowest temperature at which warm responses are triggered and the highest temperature at which cold responses are triggered indicates the thermal sensitivity of the system. As previously stated, the interthreshold range is the temperature range over which no regulatory responses occur (although the brain presumably detects these temperature changes). It changes from approximately 0.4°C in the awake state to approximately 3.5°C during anesthesia. Compared with normal human body temperature (37.0 ± 0.2°C), the interthreshold range is wider in the hypothermic state than in the hyperthermic state. This physiologic system acts as an “all-or-none” phenomenon. The mechanism by which the body determines the absolute threshold temperatures is not known, but it appears that the thresholds are influenced by several factors such as sodium, calcium, thyroid hormones, tryptophan, general anesthetics and other drugs, circadian rhythm, exercise, pyrogens, food intake, and cold and warm adaptation. Central regulation is fully functional in infancy but may be impaired in the premature, elderly, or extremely ill patient. It is now known that regulatory responses are based on mean body temperature.

▪ EFFERENT RESPONSE

Mean body temperature (MBT) is a physiologically weighted average temperature that reflects the thermoregulatory importance of various tissues but in particular that of the central compartment. In unanesthetized subjects, the MBT can be calculated as follows:(5.3)  

where T denotes temperature measured in °C or as follows (several different formulas exist: Ramanathan, 1964 ; Colin et al., 1971 ; Shanks, 1975 ; Puhakka et al., 1994 ):(5.4)  

where MSK reflects the mean skin temperature (in °C), which then equals(5.5)  

From that it follows:(5.6)  

Although skin temperature is the most important parameter in triggering behavioral changes, for the thermoregulatory autonomic response, the thermal input from the skin contributes only about 20% (Cheung and Mekjavic, 1995 ; Lenhardt et al., 1999 ). The main part of this autonomic response depends on the afferent information from the central core ( Jessen and Mayer, 1971 ; Simon, 1974 ; Jessen and Feistkorn, 1984 ), which includes the brain (parts other than the hypothalamus), the spinal cord, and deep abdominal and thoracic tissues, with each of them contributing about 20% to the central thermoregulatory control ( Jessen and Mayer, 1971 ; Mercer and Jessen, 1978 ).

Although the most commonly described thermoregulatory model is a set point system in which hypothalamic temperatures above or below a predetermined level trigger warm or cold defenses, respectively, temperature regulation may also be described by a system of thresholds and gains for each particular thermoregulatory response. Efferent responses (behavioral changes, cutaneous vasoconstriction or vasodilatation, nonshivering thermogenesis, shivering, and sweating) appear to be mediated according to the central interpretation of the afferent input.

The thermal steady state is actively defended when the hypothalamus responds to thermal changes, that is, temperatures exceeding the interthreshold range. Thus, thermal deviations from the threshold temperature initiate efferent responses that either increase metabolic heat production (nonshivering thermogenesis and shivering) and decrease environmental heat loss (active vasoconstriction and behavioral changes), or increase heat loss (active vasodilatation, sweating, and behavioral maneuvers). These thermoregulatory effectors work by adjusting their own threshold and gain according to the physiologic needs, and they do so by selecting the order of responses and by regulating the intensities.

Behavioral responses to environmental temperatures outside the thermoneutral range (approximately 28°C for an unclothed adult) are quantitatively the most important thermoregulatory effectors in humans (e.g., heating the home, looking for shelter, putting on a jacket, etc.) and are much more efficient than all of the autonomic responses combined. Cutaneous vasoconstriction is not only the first thermoregulatory response to hypothermia, but also the most consistent one. Total digital skin blood flow can be categorized into a nutritional (capillaries) and a thermoregulatory (arteriovenous shunts) component. Cold-mediated decreases in cutaneous blood flow are most pronounced (down to 1% of the normal blood flow in an environment with neutral temperature) in arteriovenous shunts of the hands, feet, ears, lips, and nose ( Grant and Bland, 1931 ; Hillman et al., 1982 ). These shunts are typically 100 μm in diameter, which means that one can divert 10,000 times as much blood as a capillary with a 10-μm diameter under otherwise unchanged conditions (i.e., same length and pressure gradient). Shunt flow is primarily regulated by norepinephrine (released from presynaptic adrenergic terminals), which binds to peripheral α2-receptors that are sensitized by local cooling and inhibited by temperatures equal to or higher than 35°C ( Sessler and Ponte, 1982 ). Flow decreases not only in the arteriovenous shunts but also in the far more numerous capillaries ( Coffman and Cohen, 1971 ). Despite the impressive decrease in cutaneous perfusion secondary to thermoregulatory vasoconstriction, the resulting reduction of heat loss from the hands and feet decrease by 50%, but only by 17% from the trunk, resulting in an overall heat loss reduction of only 25% ( Sessler et al., 1991 ).

In contrast, warm exposure initially results in sweating, which results in the triggering of massive precapillary vasodilatation with marked increase in skin blood flow. This allows for huge amounts of heat to be transported to the skin, which then dissipate to the environment, mainly by evaporation due to preconditioning by sweat. Voluntary muscle activity, nonshivering thermogenesis, and shivering are efferent mechanisms, which lead to heat generation.

▪ THERMAL REGULATION IN THE NEWBORN

Premature infants or infants small for gestational age, but also full-term neonates, have an exceptionally large skin surface area compared with their body mass (normal ratio: term neonate, 1; adult, about 0.40) and an increased thermal conductance (thin layer of subcutaneous fat). Furthermore, evaporative heat loss is increased due to a reduced keratin content in the infant's skin. Therefore, neonates lose proportionately more heat through skin than do adults in a similar environment. In contrast to the adult, the capabilities and the functional range of the neonate's thermoregulatory system are significantly limited and easily overwhelmed by environmental factors. The lower temperature limit of thermal regulation in adults is 0°C, whereas that in newborns is 22°C. The combination of increased heat loss and a diminished efficacy of the thermoregulatory response with a reduced ability to generate heat makes these infants prone to hypothermia. The same anatomic properties that are responsible for the increased risk of hypothermia also allow for three to four times faster rewarming in infants and children compared with adults ( Szmuk et al., 2001 ).

The neutral temperature is defined as the ambient temperature, at which the oxygen demand (as a reflection of metabolic heat production) is minimal and temperature regulation is achieved through nonevaporative physical processes alone. For unclothed adults, the neutral temperature is about 28°C; for neonates, 32°C; and for preterm infants, 34°C (see later). In a thermoneutral environment, the cutaneous arteriovenous shunts are open and skin blood flow is maximal.

In general, the maintenance of the core temperature in a cool environment results in an increased oxygen consumption and metabolic acidosis. It was demonstrated long ago that in full-term neonates, oxygen consumption does not correlate with rectal temperature but rather increases directly with the skin surface-to-environment temperature gradient ( Adamson et al., 1965 ). Oxygen consumption was minimal at gradients of 2° to 4°C. Thus, at environmental temperatures of 32° to 34°C and abdominal skin temperature of 36°C, the resting newborn infant is in a state of minimal oxygen consumption (i.e., the neutral thermal state). Normal rectal temperature, therefore, does not imply a state of minimal oxygen consumption in this age group ( Fig. 5-2 ).

 
 

FIGURE 5-2  A, Relation of oxygen consumption (   o2) to rectal temperature. Note complete lack of correlation. B, Relation between oxygen consumption and temperature gradient between skin temperature and environment (ΔTS-E) in full-term human newborns with varying deep body and skin temperatures.  (From Adamson KJ Jr, Gandy GM, James LS: The influence of thermal factors upon oxygen consumption of the newborn human infant.J Pediatr 66:495, 1965.)

 



Of particular concern in view of thermoregulation in the newborn is the head, which comprises up to 20% of the total skin surface area and shows the highest regional heat flux ability ( Anttonen et al., 1995). In addition, the thin skull bone and the usually sparse scalp hair in combination with the close proximity to the well-perfused brain (core temperature) further favor heat loss from the head. Facial cooling may increase oxygen requirements by up to 23% in the term infant and up to 36% in the premature infant ( Sinclair, 1972 ), thereby demonstrating the effectiveness of protecting the infant from heat loss by covering the head.

Thermoregulatory vasoconstriction and vasodilatation most likely establish during the first day of life ( Lyons et al., 1996 ) and can occur in both the premature and the full-term infant ( Bruck, 1961 ). With vasoconstriction, cutaneous blood flow decreases and the effect of tissue insulation increases, which results in an overall reduction in conductive and convective heat losses.

Heat dissipation in the premature or small-for-gestational-age infant represents the extreme of thermal regulation in the neonate. Small size and decreased insulation tissue result in increased volume index and thermal transfer coefficients, thereby challenging the thermoregulating capacity of these infants. These disadvantages narrow the temperature range for thermoregulatory stability. In small-for-gestational-age infants, a slightly lower skin surface area-to-volume ratio and an increased motor tone offer some protection compared with the premature infant in regard to heat loss or transfer. In addition to the physical limitations of heat conservation in infants and children, surgery can further increase heat loss and fluid requirements by exposing the visceral surfaces of the abdomen and thorax, thereby exacerbating evaporative heat and water losses.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ HEAT LOSS MECHANISMS

For an organism to be homeothermic, it must be able both to dissipate and to produce heat. Controlled heat loss is fundamental in homeotherms and is accomplished in two stages ( Swyer, 1973 ), both governed by the physical laws of conduction, radiation, convection, and evaporation ( Box 5-2 ). The second law of thermodynamic states that heat transfer is only possible from a warmer to a cooler object, never from a cooler to a warmer object. This means that the warmer object (in the operating room setting, this is almost exclusively the patient) is used to warm up the surrounding cooler objects (operating room walls, tables, etc.). Although most anesthesiologists consider heat loss to be a nuisance, we have to keep in mind that without any heat loss to the environment (i.e., perfect insulation), the body of an awake adult at rest (assuming a metabolic rate of 75 W) would warm up by at least 1°C per hour! (Keep in mind that during exercise, the metabolic heat generation can increase 10-fold.) This can be calculated with the following formula:(5.7)  

BOX 5-2 

Mechanisms (by Percent) of Heat Loss in Neutral Thermal Environments

Radiation

39%

Convection

34%

Evaporation

24%

Conduction

3%

 

where HSR is the heat storage rate (W), m is the body mass (kg), κ is the specific heat coefficient of the human body (3.5 · 103 J/°C) ( Burton, 1935 ), dTB is the change in body temperature (°C), and dt is the time interval (seconds).

The first stage of heat loss is transfer of heat from the body core (central compartment) to the periphery and the skin surface, which is referred to as the concept of internal redistribution of heat. The second stage is dissipation of heat from the skin surface to the environment. Physiologic manipulations of regional blood flow and changes in the thermal conductance properties of the insulation tissue can influence both gradients. Most studies of thermal regulation in infants and children have quantified the relative contributions of radiation, convection, evaporation, and conduction to heat loss. A study in newborns in an environment at neutral temperature found radiation, convection, evaporation, and conduction to account for 39%, 34%, 24%, and 3%, respectively, of total heat loss ( Hey, 1973 ). Changes in the environment (e.g., the operating room) can alter not only the magnitude of heat loss but also the relative contribution of each of these four physical components. Figure 5-3 gives an overview of the heat loss mechanisms involved in the operating room setting.

 
 

FIGURE 5-3  Schematic illustration of the mechanisms contributing to perioperative hypothermia: (1) conduction, (2) evaporation, (3) convection, and (4) radiation.  (Modified from Gurtner C, Paul O, Bissonnette B: Temperature regulation: physiology and pharmacology. In Bissonnette B, Daleus B, editors: Pediatric anesthesia. Principles and Practice. New York, 2002, McGraw-Hill, p173.)

 



▪ CONDUCTION

Conduction is heat transfer between two surfaces in direct contact. The amount of heat transferred (C) depends on the temperature difference between two objects in contact (T1 - T2), the surface area in contact (A), and the conductive heat transfer coefficient h k of the materials. It can be calculated as follows:(5.8)  

The coefficient h k is a property of the material or interface between objects that determines the rate of heat transfer per unit area per unit temperature difference (W/m2 · °C). It seems convenient to think of conductance as the inverse of resistance. During surgery, relatively little heat is lost to the environment via conduction, because patients are well insulated from surrounding objects ( Allen, 1987 ). Conduction is also represented in the energy needed to warm cool irrigation solutions and intravenous fluids, which have the potential to significantly reduce body temperature. Care should also be taken to ensure that the patient's skin is not in contact with metallic surfaces. (In addition, contact with metallic surfaces during surgery must be avoided to prevent skin burns from electrocautery.) The physiologic factors controlling conductive heat loss in newborns are cutaneous blood flow and the thickness of the subcutaneous tissue (insulation).

▪ RADIATION

Radiant heat loss is the transfer of heat between two objects of different temperature but not in contact with each other (e.g., radiation is the mechanism by which the sun warms the earth). The emitted radiation carries energy from the warmer to the cooler object, therefore causing the warmer object to cool and the cooler object to warm. This heat transfer occurs in the infrared light spectrum. Heat exchange by radiation depends on the difference of the fourth power of the absolute temperatures of the two objects:(5.9)  

where R is heat transfer by radiation (W), e is the emissivity (a material property with a value between 0 and 1), σ is the Stefan Boltzmann constant (5.67 · 10-8 J/sec · m2 · K 4), A denotes the surface area of the object (m2), Tsk is the skin temperature, and Tr is the temperature of the second object (both temperature values in K).

To calculate radiation heat exchange for clinical purposes, where Tsk and Tr are close to each other, it is acceptable to use a first-power relationship as follows:(5.10)  

where h rdenotes the radiation coefficient. The value of h r depends on the temperatures of the two surfaces and such surface characteristics as emittance and reflectance (reflected in the emissivity value). It is obvious that heat transfer by radiation principally depends on the temperature of the two surfaces concerned but is unaffected by air movement or the distance between the surfaces. It can take place across a vacuum ( Allen, 1987 ).

As previously stated, newborns and infants have a large surface area-to-volume ratio, thus radiant heat loss is proportionally greater the smaller the infant. In both the awake and the anesthetized infant, radiation is the major factor for heat loss under normal conditions. The human body is an excellent emitter of energy at wavelengths relevant for heat transfer, and the probability of photon reflection in the standard operating room is near zero. Radiant heat loss in the operating room is therefore a function of the difference between the patient's body temperature and the objects in the room, such as the walls and solid objects. Radiant heat loss is diminished by increasing the ambient temperature of the room, thereby reducing the temperature gradient between the patient and the operating room walls and its content. As long as a temperature gradient exists, the patient continues to warm up walls and solid objects by exchanging radiant energy. At a room temperature of 22°C, about 70% of the total heat loss is due to radiation ( Hardy et al., 1941 ). A single-layer covering dramatically reduces the losses by convection and radiation; thus, a thin shirt (e.g., a silk blouse providing negligible insulation) results in considerably increased thermal comfort.

▪ CONVECTION

Convective heat loss is the transfer of heat to moving molecules such as air and liquid. The thin air layer adjacent to the skin is warmed by conduction from the body. Although changes in body posture and minute ventilation may affect convective heat loss, convection plays only a minor role in heat loss. In the case of a naked individual exposed to air, the rate and direction of convective heat exchange depend on the airflow velocity and the temperature difference between the air and the skin surface. The situation is more complicated in a clothed individual. Convective heat transfer may be calculated as follows:(5.11)  

where Q is the heat exchange by convection (W), A is the surface area (m2), h qis the convective heat transfer coefficient (W/m2 · °C), Tsk is the mean skin temperature (°C), and Tais the ambient temperature (°C). The convective heat exchange coefficient is not a constant and depends on the rate of air movement, the shape of the body, and the medium. It may be calculated from h c= 8.3 V0.5(W/m2 · °C), where V is the air movement (m/sec) ( Allen, 1987 ). Thus, convective heat losses increase in proportion to the temperature difference between the body surface and the surrounding fluid (liquid or gas) and the square root of the flow velocity of the fluid (liquid or gas) in contact with the patient. Convective losses are experienced outdoors as the “wind chill factor.”

▪ EVAPORATION

Evaporative heat loss occurs through the skin and the respiratory system. Under conditions of thermal neutrality, evaporation accounts for 10% to 20% of heat loss. Physical factors that govern evaporative heat loss include relative humidity of the ambient air, velocity of airflow, and lung minute ventilation. The driving force for evaporation is the vapor pressure difference between the body surface and the environment. Evaporative losses include several components: (1) sweat (sensible water loss), (2) insensible water loss from the skin, respiratory tract, and open surgical wounds, and (3) evaporation of liquids applied to the skin such as antibacterial solutions. The evaporation of water from a surface is an energy-dependent process, energy that is absorbed from the surface during transition from the liquid to the gaseous state. This energy is called the latent heat of vaporization, and in the case of sweat, it has a value of 2.5 ×106 J/kg. This figure emphasizes the extraordinary power of the human sweating mechanism as a means of dissipating heat, especially when it is considered that an adult in good physical condition can produce up to 2 liters of sweat per hour. In an environment where the air temperature is equal to or higher than the skin temperature, sweating is the only mechanism available for the dissipation of heat from metabolic production. In this situation, anything that limits evaporation, such as high ambient humidity or impermeable clothing, easily leads to heat storage and a potentially fatal rise in body temperature.

Evaporative heat loss depends on the water vapor pressure gradient between the skin surface and the ambient air and may be calculated from the following:(5.12)  

where E is the evaporative heat loss (W); h e is the evaporative heat transfer coefficient (W/m2 · kPa), Psk is the water vapor pressure at the skin surface, and Pa is the ambient water vapor pressure (both pressure values in kPa). The coefficient for evaporative heat exchange (h e) incorporates the latent heat of vaporization of water and the paramount effect of air movement on evaporation. For practical purposes, h e may be calculated from h e= 124 V0.5 (W/m2 · kPa), where V is the airflow velocity (m/sec). Similar to convection, the important point to note is that evaporation is determined by the vapor pressure gradient between the exposed body surface and the ambient air and the rate of airflow across the surface ( Allen, 1987 ).

Physiologic factors that affect evaporative losses relate to the infant's ability to sweat and to increase the minute ventilation. Although the physical characteristics of the newborn predispose him or her to heat loss, it has been demonstrated that neonates are capable of sweating in a warm environment ( Bruck, 1961 ). Full-term neonates begin to sweat when rectal temperature is between 37.5° and 37.9°C and ambient temperature exceeds 35°C. Although the onset of sweat production in infants small for gestational age is slower than in full-term infants, the maximum rates of sweat production are comparable (Sulyok et al., 1973 ). Premature infants with a gestational age of less than 30 weeks show no response because their sweat glands are not yet fully developed.

A small amount of heat is lost when dry inspired respiratory gases are humidified by water evaporating from the tracheobronchial epithelium. In adults, respiratory losses account for only 5% to 10% of total heat loss during anesthesia and surgery ( Bickler and Sessler, 1990 ), and total insensible loss accounts for approximately 25% of the total heat dissipated. Because minute ventilation on a per-kilogram basis is significantly higher in infants and children than in adults, respiratory losses represent about one third of the total heat loss. Heat loss from the respiratory tract increases if one breathes cool, dry gas as opposed to warm, moisturized gas (Bissonnette et al., 1989a, 1989b [24] [25]).

Heat loss from evaporation inside a large surgical incision may equal all other sources of intraoperative heat loss combined ( Roe, 1971 ). Because of increased evaporative heat loss, hypothermia is also more likely to occur if the skin of the patient is wet or in contact with wet drapes.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ HEAT GENERATION

The ability to produce heat by increasing the metabolic rate and oxygen consumption is the other component of thermal regulation in the homeotherm ( Hull and Smales, 1978 ). Although three of the physical mechanisms that lead to heat loss (i.e., conduction, radiation, and convection) can theoretically also be used to passively warm up a patient, the body has the ability to actively produce heat. Heat generation is achieved via four mechanisms:

  

1.   

Voluntary muscle activity

  

2.   

Nonshivering thermogenesis

  

3.   

Involuntary muscle activity (shivering)

  

4.   

Dietary thermogenesis

The behavioral aspect of heat production (voluntary muscle activity) is absent in the perioperative period and therefore its role in heat production is not discussed further. Of the two remaining mechanisms for heat production, nonshivering thermogenesis is the major component in the newborn, whereas shivering thermogenesis is the main mechanism for heat production in the adult. The contribution of nonshivering thermogenesis in adults is debatable ( Jessen, 1980a ).

Although the time course and relation between nonshivering thermogenesis and shivering thermogenesis in infants have been described ( Hull and Smales, 1978 ), the exact time sequence and factors involved in the developmental aspects of switching on and off the nonshivering thermogenesis mechanism remain to be elucidated. It seems that the importance of nonshivering thermogenesis is rapidly decreasing after the first year of life, while at the same time shivering thermogenesis is becoming more and more effective.

▪ NONSHIVERING THERMOGENESIS

Nonshivering thermogenesis is defined as an increase in metabolic heat production (above the basal metabolism) not associated with muscle activity. It occurs mainly through metabolism of brown fat, but to a lesser degree also in skeletal muscle, liver, brain, and white fat.

Brown fat differentiates in the human fetus at 26 to 30 weeks of gestational age. It comprises only 2% to 6% of the infant's total body weight and is located in six main areas: between the scapulae, in small masses around blood vessels in the neck, in large deposits in the axilla, in medium-size masses in the mediastinum, around the internal mammary vessels in the mediastinum, and around the adrenal glands or kidneys.

Brown fat is a highly specialized tissue whose brown color is secondary to the abundant content of mitochondria in the cytoplasm of its multinucleated cells. These mitochondria appear to be densely packed with cristae and have an increased content of respiratory chain components ( Himms-Hagen, 1976 ). They are unique in their ability to uncouple oxidative phosphorylation, resulting in heat production instead of generating adenosine triphosphate. This uncoupling is mediated by the presence of the uncoupling protein (UCP), also termed thermogenin, which is located in the inner mitochondrial membrane.

Brown fat is highly vascularized and has a rich innervation, which appears to be primarily β-sympathetic in origin and responsible for the uncoupling of oxidative phosphorylation (Karlberg et al., 1962, 1965 [123] [124]). Cold stress increases sympathetic nerve system activity and norepinephrine production, leading to increased lipase activity in the brown fat tissue ( Schiff et al., 1966 ). As a consequence, hydroxylation of triglycerides and release of free fatty acids occur. These free fatty acids appear to act on the UCP and thereby to increase the protein conductance across the mitochondrial membrane. In addition to norepinephrine, glucocorticoids and thyroxin have been implicated as factors triggering nonshivering thermogenesis ( Gale, 1973 ; Jessen, 1980b, 1980c [115] [116]). The heat produced by nonshivering thermogenesis is mainly a byproduct of fatty acid metabolism, but to a minor degree it can also be a product of glucose metabolism. The activation of brown fat metabolism results in an increased proportion of the cardiac output being diverted through the brown fat. This proportion may reach as much as 25% of the cardiac output, thereby facilitating the direct warming of the blood.

Pharmacologic inhibition of nonshivering thermogenesis is possible with ganglionic and β-receptor blockade ( Silverman et al., 1964 ), inhalational anesthetics ( Ohlson et al., 1994 ), and surgically by sympathectomy ( Stern et al., 1965 ). The inhibition of nonshivering thermogenesis by inhalational anesthetics begins as early as 5 minutes after turning on the vapor and starts to wean off within approximately 15 minutes after discontinuation of the anesthetic ( Ohlson et al., 1994 ). Nonshivering thermogenesis is also inhibited in infants anesthetized with fentanyl and propofol ( Plattner et al., 1997).

Nonshivering thermogenesis seems to be quite variable in adults but most often does not appear to be functional or relevant ( van Marken Lichtenbelt and Daanen, 2003 ). This is supported by the fact that oxygen consumption does not increase significantly when patients are vasoconstricted ( Mestyan et al., 1964 ; Dawkins and Scopes, 1965 ; Sessler et al., 1988 ). It seems that adults have the potential to regenerate brown fat tissue under certain pathologic conditions, such as high and sustained sympathetic stimulation (pheochromocytoma), Chagas disease, hibernoma (benign brown fat tumor), and marked cold acclimatization ( Lean et al., 1986 ; Garruti and Ricquier, 1992 ; Vybiral et al., 2000 ).

In contrast, prematures, full-term neonates, and infants are able to double their metabolic heat production during cold exposure ( Mestyan et al., 1964 ; Dawkins and Scopes, 1965 ; Hey and Katz, 1969 ). Clinically significant nonshivering thermogenesis is possible within hours after birth ( Oya et al., 1997 ) and may persist up to the age of 2 years. Although it is the main source of thermoregulatory heat production in infants, it should be kept in mind that its effect is limited and does not compensate for the decreased ability of newborns and infants to effectively reduce heat loss through cutaneous vasoconstriction or to increase heat production through shivering.

During general anesthesia in children, nonshivering thermogenesis is not triggered by core hypothermia or cold exposure and therefore does not seem to be functional ( Plattner et al., 1997 ). Halothane anesthesia has been shown to block nonshivering thermogenesis in children (Ohlson et al., 1994, 1997 [166] [165]; Dicker et al., 1995 ). It has been demonstrated in animal studies that pharmacologic inhibition of nonshivering thermogenesis by β-blockade does not prevent shivering thermogenesis ( Bruck and Wunnenberg, 1965 ). In the animals studied, shivering did not fully compensate for the lack of heat produced by nonshivering thermogenesis. The magnitude of nonshivering thermogenesis in animals varies among species, but it appears that in newborn versus adult animals and in cold-adapted versus warm-adapted animals, nonshivering thermogenesis is significant ( Himms-Hagen, 1976 ).

▪ SHIVERING THERMOGENESIS

Although there appears to be some developmental sequence in human thermal regulation to the onset of shivering thermogenesis, the precise mechanisms and/or factors that govern this development are unclear. With increasing age, shivering thermogenesis assumes a more prominent role in thermoregulation. It has been demonstrated that shivering occurs only after all of the other mechanisms, such as behavioral responses, nonshivering thermogenesis (both ineffective under anesthesia), and maximal vasoconstriction, have failed to maintain body temperature within the interthreshold range ( Hemingway, 1963 ; Hemingway and Price, 1968 ). Until recently, newborns and infants were considered not to be able to shiver, presumably because of the general immaturity of the musculoskeletal system on the one hand and the limited muscle mass on the other hand, which would render muscle activity ineffective in cold defense. A few reports exist about shivering in neonates with shivering occurring at rectal temperatures of 35.0° to 35.3°C ( Brück, 1992 ; Petrikovsky et al., 1997 ). It is debatable whether this shivering was indeed thermoregulatory in origin or whether drugs and other factors (all mothers received an intrapartum amnioinfusion) were to blame. In general, neonates do not shiver, and if they do, it is of minor importance in thermoregulation.

Shivering can briefly result in an up to sixfold increase in metabolic heat production ( Giesbrecht et al., 1994 ), but only a twofold increase is sustained ( Horvath et al., 1956 ). Shivering is characterized by involuntary, irregular muscular activity that begins in the muscles of the upper body. The basal frequency of the shivering tremor in the electromyogram is typically around 200 Hz ( Israel and Pozos, 1989 ). Superimposed slow tremor spasms also occur, producing a “waxing and waning” electromyographic pattern at a rate of two to eight cycles per minute ( Stuart et al., 1966 ; Sessler et al., 1991 ).

The impulses from cold receptors impinge at the motor center for shivering, which is located in the dorsomedial part of the posterior hypothalamus adjacent to the wall of the third ventricle. Under warm conditions, this center is inhibited by impulses from the heat-sensitive area in the preoptic region of the anterior hypothalamus; however, a predominance of cold impulses from the skin and the spinal cord activates the shivering center, which results in stimulation of the anterior motor neurons of the spinal cord. Initially, this results in an increased skeletal muscle tone throughout the body. These signals do not cause the actual shivering. Only when this tone is raised over a certain level does shivering become visible ( Guyton, 2000 ).

Because of this increased muscle activity, oxygen consumption (   o2) and CO2 production proportionally increase by up to 400% to 600% for a short period of time ( Horvath et al., 1956 ; Benzinger, 1969; Ciofolo et al., 1989 ; Just et al., 1992 ; Giesbrecht et al., 1994 ).

In healthy patients, this increase in    o2 is met by an increase in cardiac output without any hemodynamic compromise. In patients with already limited hemodynamic and coronary reserves, this increase in    o2 can lead to a decreased mixed venous oxygen content that, under a less-than-perfect ventilation/perfusion ratio, may result in a decreased arterial oxygen content and, consequently, decreased tissue oxygen delivery. An inverse correlation has been shown between intraoperative temperature and postoperative    o2 ( Roe et al., 1966 ), as well as between different anesthetics (see “Thermoregulation and General Anesthesia”) and postoperative    o2. Shivering not only is an unpleasant experience for the patient in the postoperative period but it can also increase intraocular and intracranial pressure (Mahajan et al., 1987 ; Rosa et al., 1995 ).

The incidence of postoperative shivering is inversely related to the core temperature; shivering was also found in patients kept strictly normothermic during anesthesia with isoflurane or desflurane, indicating that a substantial fraction of shivering is nonthermoregulatory ( Horn et al., 1998 ) with pain being a potential trigger ( Horn et al., 1999 ). Inhibition of shivering with meperidine in unanesthetized, actively cooled volunteers results in a more than threefold higher and more than fourfold prolonged core temperature afterdrop and a 37% decreased rewarming rate compared with the shivering control group ( Giesbrecht et al., 1997 ).

▪ DIETARY THERMOGENESIS

Stimulation of energy expenditure and thermogenesis by certain nutrients (i.e., proteins and amino acids) is well known. Despite muscle paralysis and decreased metabolism during general anesthesia, the infusion of small amounts of amino acids resulted in an up to fivefold increased heat generation during anesthesia compared with awake adults ( Sellden et al., 1994 ). Using preoperative and intraoperative amino acid infusions, the same researchers were able to prove this advantage clinically in achieving a core temperature of 36.5 ± 0.1°C at the end of surgery, whereas the temperature dropped to 35.7 ± 0.1°C in the control group ( Sellden and Lindahl, 1999 ). Similar findings have been reported for preoperative amino acid infusion in patients undergoing spinal anesthesia ( Kasai et al., 2003 ). Although effective, the exact mechanism behind this form of thermogenesis has not been fully elucidated. It seems that stimulation of cellular amino acid oxidation is crucial. Furthermore, protein synthesis and breakdown in extrasplanchnic tissues, requiring extra synthesis of ATP, could be a contributing factor as well. It has been concluded that about half of the heat generated in association with amino acid infusion is splanchnic in origin and that blood flow in extrasplanchnic (but not splanchnic) tissues is increased significantly, reflected by an increase in cardiac output of almost 20% ( Brundin and Wahren, 1994 ). Except for a different time course, the average whole body thermic effect of intravenous amino acid administration was not different from the one seen with oral protein ingestion.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ EFFECT OF ANESTHESIA ON THERMOREGULATION

▪ THERMOREGULATION AND GENERAL ANESTHESIA

Anesthetics may interfere with thermal regulation at both peripheral and central receptor sites. In adults, general anesthesia has been shown to decrease the thermoregulatory threshold temperature triggering a response to hypothermia on average by approximately 2.5°C and increases the threshold temperature initiating a response to hyperthermia to a lesser degree (approximately 1.3°C) (Sessler et al., 1988a and b [200] [201]) (Boxes 5-3 and 5-4 [3] [4]). This anesthesia-induced expansion of the interthreshold range ( Fig. 5-4 ) results in a wider temperature range over which active thermoregulatory responses are absent. Within this range, patients are poikilothermic, and body temperature is changing passively in proportion to the difference between metabolic heat production and heat loss to the environment. Vasoconstriction and nonshivering thermogenesis are the only thermoregulatory responses available to anesthetized, paralyzed, hypothermic infants and children. Patients with mild hypothermia during surgery (e.g., a central body temperature of about 34.5°C) demonstrate profound peripheral vasoconstriction, which can easily be verified using skin surface temperature gradients (e.g., forearm versus fingertip skin temperature), laser Doppler flowmeter, volume plethysmography, or other techniques ( Stuart et al., 1966 ; Sessler et al., 1988a and b [200] [201]).

BOX 5-3 

General Effects of Anesthetics on Thermoregulation

  

1.   

Lower hypothermia threshold

  

2.   

Increase hyperthermia threshold

  

3.   

Widen interthreshold range

  

4.   

No effect on gain response (except for desflurane)

BOX 5-4 

Specific Effects of Anesthetics on Thermoregulation

  

1.   

Opioids reduce threshold for vasoconstriction and shivering as a linear function of dose.

  

2.   

Propofol reduces threshold for vasoconstriction and shivering as a linear function of dose.

  

3.   

Volatile anesthetics produce nonlinear inhibition, the threshold being proportionally greater at higher end-tidal concentrations.

  

4.   

Volatile anesthetics at comparable minimum alveolar concentration values produce similar amounts of inhibition.

  

5.   

N2O decreases vasoconstriction threshold less than other volatile agents.

 
 

FIGURE 5-4  Schematic illustration of the thermoregulatory thresholds and gains in awake and anesthetized adults, children, and infants. The vertical height/depth of the lines represents the maximal intensity of each effector response, whereas the slope (negative values for anesthetized patients in this graph) represents the gain of the response. The horizontal axis corresponds to the core temperature. The threshold is defined as the core temperature triggering a response. The sensitivity of the thermoregulatory system is the range between the first cold response (vasoconstriction) and the first warm response (sweating) and is called interthreshold range. NST, nonshivering thermogenesis.

 

 

Under anesthesia, the maximal intensity of peripheral vasoconstriction is similar to that in awake volunteers, indicating that the gain of the response is preserved, although at a markedly lowered threshold. The only exemption seems to be desflurane, which lowers not only the threshold temperature for vasoconstriction but also the gain ( Kurz et al., 1995 ). The temperature at which vasoconstriction and nonshivering thermogenesis occur identifies the corresponding lower thermoregulatory threshold for the anesthetic agent at any given concentration or dose.

For halothane administered in a concentration of 1.0% in oxygen to healthy adults undergoing donor nephrectomy, the thermoregulatory threshold is 34.4 ± 0.2°C ( Sessler et al., 1988a ). In infants and children anesthetized with 0.6% halothane and a caudal epidural block with bupivacaine, the threshold for vasoconstriction was 35.7°C ( Bissonnette and Sessler, 1992 ). Of note is that in children weighing more than 30 kg, the central temperature continued to drop after the vasoconstriction threshold had been reached, whereas in children and infants below 30 kg, the central temperature remained constant or even slightly increased. These data show that thermoregulatory defense in infants and smaller children is more effective than in older children and adults.

The administration of subanesthetic concentrations of nitrous oxide (10% to 25% in a normoxic mixture) to healthy adult volunteers resulted in a significant and dose-independent reduction of shivering thermogenesis ( Cheung and Mekjavic, 1995 ). In a concentration of 0.6 (63%) minimum alveolar concentration (MAC), nitrous oxide resulted in a calculated reduction of the vasoconstriction threshold to 35.7 ± 0.6°C ( Goto et al., 1999 ). Overall, nitrous oxide decreased the vasoconstriction threshold less than equipotent concentrations of the volatile anesthetic agents sevoflurane and isoflurane.

In a small study in adults anesthetized with isoflurane, the decrease in the thermoregulatory threshold for vasoconstriction was found to be inversely correlated to the isoflurane concentration, and the threshold temperature decreased by approximately 3°C/% end-tidal isoflurane concentration ( Stoen and Sessler, 1990 ). In a more recent study, the same group found that the dose dependence was not linear, with isoflurane reducing the threshold temperature disproportionately at higher anesthetic concentrations ( Xiong et al., 1996 ). In adults anesthetized with 0.7% isoflurane, the shivering temperature threshold was decreased as was the maximum intensity of shivering. The gain of shivering increased significantly and was associated with a clonic muscular activity that was not a component of regular shivering ( Ikeda et al., 1998 ).

Inhaled Anesthetic Agents

The vasoconstriction threshold in infants and children anesthetized with isoflurane differs only slightly from that in adults ( Bissonnette and Sessler, 1990 ). The thermoregulatory threshold for vasoconstriction in pediatric patients anesthetized with 1 MAC halothane in 70% nitrous oxide is higher (35.8 ± 0.5°C) ( Bissonnette and Sessler, 1992 ) than reported in an adult study (34.4°C) ( Sessler et al., 1988a ). In the adult study, the patients were anesthetized without nitrous oxide and the administered halothane concentration (1.3 MAC) was significantly greater than the halothane concentration used in the pediatric study (1.0 MAC). In a separate study, Nebbia and others (1996) noted a similar thermoregulatory threshold (35.8 ± 0.3°C) in pediatric patients who were anesthetized with a caudal epidural block with bupivacaine and 1 MAC halothane in oxygen.

Thermoregulatory inhibition under general anesthesia with equipotent doses of halothane is therefore likely to affect adults and children similarly. The high surface area-to-mass ratio in infants, which allows for a rapid loss of heat to the environment, is largely offset by a high intrinsic metabolic rate. Environmental heat loss is further reduced by a well-developed thermoregulatory vasoconstriction (Bissonnette and Sessler, 1990 ). Although there was a trend toward increased threshold temperatures in smaller infants and children anesthetized with similar (age-corrected) concentrations of isoflurane, differences between the groups were not statistically significant and spanned only about 0.3°C. This indicates that inhibition of thermoregulatory vasoconstriction is similar in anesthetized infants and children and relatively independent of the body weight ( Bissonnette and Sessler, 1990 ). This relatively constant degree of thermoregulatory inhibition in infants and children of different ages is in marked contrast to the age-related changes in MAC of isoflurane. In infants 1 to 6 months of age, the MAC for isoflurane is approximately 1.5 times higher than the MAC for adults. Because of the physical and physiognomic properties of infants and children, the speed of cooling is higher than in adults.

Sevoflurane was found to be similar to isoflurane in regard to decreasing the thermoregulatory threshold for vasoconstriction, although at a slower rate ( Ozaki et al., 1997 ; Saito, 1997 ). Desflurane has been shown to increase the sweating threshold temperature in a concentration-dependent, linear way. The threshold temperatures for vasoconstriction and shivering at 0.8 MAC were in the same range as expected for other volatile anesthetics; however, at 0.5 MAC, the vasoconstrictive threshold was reduced less. Thus, for desflurane there may be a nonlinear concentration-response relationship ( Annadata et al., 1995 ). Desflurane also decreases the gain of thermoregulatory vasoconstriction ( Kurz et al., 1995 ).

Enflurane is an interesting inhalational agent with respect to thermoregulatory effects. In healthy adult volunteers anesthetized with 1.3% enflurane (equivalent to approximately 0.77 MAC), the thermoregulatory threshold for vasoconstriction was found to be 35.1 ± 0.6°C without any stimulation and 35.5 ± 0.8°C during painful electrical stimulation, demonstrating a slight, although clinically insignificant, effect of nociception in offsetting the anesthesia-induced thermoregulatory inhibition ( Washington et al., 1992 ). Caudal or lumbar epidural blockade is useful in eliminating this effect during abdominal and peripheral surgical procedures. Thermoregulatory studies with enflurane in the pediatric population are hampered by the lack of enflurane MAC studies for this age group. In pediatric patients aged 1 to 12 years, 1.67% enflurane (equals 1 MAC in adults, which is estimated to be the equivalent of 0.75 to 1.0 MAC for the population studied) with caudal bupivacaine caused a profound depression of the thermoregulatory threshold temperature for vasoconstriction. Most patients in this study failed to achieve vasoconstriction despite reaching mean temperatures of 33.9 ± 0.9°C ( Nebbia et al., 1996 ). These researchers therefore concluded that the risk of hypothermia is significantly higher compared with isoflurane or halothane.

The effects of different inhalational anesthetics on the threshold for thermoregulatory vasoconstriction are summarized in Figure 5-5 . Furthermore, hypothermia can affect not only the physical characteristics of inhalational anesthetics but also the pharmacokinetics and pharmacodynamics of intravenous agents. For inhalational agents, hypothermia lowers the MAC (for isoflurane there is a linear decrease of 5.1%/°C) and increases the tissue solubility ( Vitez et al., 1974 ; Eger and Johnson, 1987 ; Antognini, 1993 ; Antognini et al., 1994 ; Liu et al., 2001 ). Thus, for any inspired concentration of an inhalational agent in a hypothermic patient, an increased amount of the anesthetic agent is delivered to the tissues when in fact the anesthetic requirements are decreased. The pharmacokinetics of barbiturates ( Kadar et al., 1982 ) and narcotics are also affected by hypothermia ( Koren et al., 1987 ).

 
 

FIGURE 5-5  Threshold temperature (in °C) for thermoregulatory vasoconstriction in adults at different concentrations of inhalational anesthetics alone or in combination with nitrous oxide. The x-axis denotes the minimum alveolar concentration (MAC) equivalents of the different inhalational anesthetics.  (Data compiled from Sessler and Ponte, 1982 ; Washington et al., 1992 ; Kurz et al., 1995 ; Ozaki et al., 1995 ; Annadata et al., 1995 ; Goto et al., 1999 .)

 



Intravenous Agents

The effect of opioids on thermoregulation remained unclear until recently. Alfentanil has been demonstrated to significantly reduce the thermoregulatory threshold temperature for vasoconstriction. This reduction appears to be linear and in proportion to the plasma drug concentration ( Kurz et al., 1995 ). Meperidine (pethidine) and sufentanil linearly reduce the shivering threshold temperature ( Alfonsi et al., 1998 ). This side effect can be used to treat postoperative shivering. Neither meperidine nor alfentanil reduces the gain and the maximum shivering intensity ( Ikeda et al., 1998 ). Tramadol slightly decreases the threshold temperature for sweating, whereas the thresholds for vasoconstriction and shivering decrease linearly with the tramadol plasma concentration. Overall, with a doubling of the interthreshold range, its effects on thermoregulation can be considered mild ( De Witte et al., 1998 ).

A comparison between the temperature effects in children anesthetized with either ketamine or halothane showed that halothane decreases rectal temperature more than ketamine and that children with the highest surface area-to-body weight ratio had the greatest decrease in body temperature regardless of the agent used ( Engelman and Lockhart, 1972 ). In adults, ketamine causes less thermoregulatory suppression than other anesthetic agents ( Hunter et al., 1981 ).

In the case of propofol, a small study in adult volunteers showed a significant and linear decrease in the threshold temperatures for vasoconstriction and shivering, whereas the sweating threshold temperature increased only slightly ( Leslie et al., 1994 ; Matsukawa et al., 1995 ). Furthermore, the induction of anesthesia with a single bolus dose of propofol (2.5 mg/kg) in adults and maintenance of anesthesia with sevoflurane in 60% nitrous oxide resulted in lower core temperatures (35.5 ± 0.3°C) compared with patients where sevoflurane and nitrous oxide only were used for induction and maintenance of anesthesia (36.2 ± 0.2°C) ( Ikeda et al., 1999 ). This led to the suggestion that the brief propofol-induced vasodilatation is sufficient to facilitate the core-to-peripheral redistribution of body heat resulting in nonrecoverable heat loss to the environment.

Midazolam only slightly decreases the threshold temperature for sweating, but more so for vasoconstriction with a tripling of the interthreshold range, which is quite similar to the results found for central neuraxial nerve blockade ( Kurz et al., 1995 ). These results contrast with the findings for volatile anesthetics, propofol or opioids, where the interthreshold range increases by a factor of 10 to 15 ( Kurz et al., 1995 ).

A bolus dose of clonidine followed by an infusion results in a dose-independent increase in the threshold temperature for sweating, but the gain remains unchanged ( Delaunay et al., 1996 ). Its use for premedication neither affects redistribution hypothermia nor worsens hypothermia during general anesthesia ( Bernard et al., 1998 ). Atropine not only blocks sympathetic cholinergic-mediated sweating but also increases the threshold temperatures for sweating and therefore may lead to hyperthermia in children ( Fraser, 1978 ).

▪ THERMOREGULATION AND REGIONAL ANESTHESIA

During regional anesthesia, central thermoregulation remains intact and therefore provides some protection against hypothermia. Anesthetic interference with regional thermal sensation (afferent and efferent pathways) with inhibition of cutaneous vasoconstriction and shivering in the anesthetized area, internal redistribution of body heat, and increased heat loss to the environment may contribute to intraoperative hypothermia. In many aspects, the factors causing intraoperative hypothermia under neuraxial anesthesia are similar to those under general anesthesia. Like during general anesthesia, redistribution of body heat from core to peripheral compartments accounts for 89% of the initial (first hour) drop in core temperature. In the subsequent 2 hours, redistribution contributed 62% to the core temperature decrease (Matsukawa et al., 1995a, 1995b [149] [150]). The extent of this redistribution, and thus also the decrease in core temperature, depends on inhibition of peripheral vasoconstriction rather than on centrally mediated effects. Neuraxial anesthesia usually affects a major part of the body mass; hence, the decrease in core temperature can be quite pronounced. Heat production during regional anesthesia is only minimally decreased ( Hynson et al., 1991 ).

In contrast to general anesthesia, patients under central neuraxial anesthesia may fail to reach an equilibrium state where heat loss and heat generation are equal, because peripheral vasoconstriction is completely inhibited by neuraxial blockade. In addition, extensive regional anesthesia may alter or even block the thermal input to the hypothalamus from a major part of the body (the information from the more active cold sensors at a normal leg skin temperature of about 33°C seems not to reach the hypothalamus, which could potentially be interpreted as relative leg warming), with the number of dermatomes blocked being directly proportional to the inhibition of central thermoregulation ( Ozaki et al., 1994 ; Leslie and Sessler, 1996 ; Frank et al., 2000 ). Heat loss may therefore be an ongoing issue until sympathetic function and consequently vasoconstriction have been restored. Under these circumstances, hypothermia may become even more severe than under general anesthesia. Although peripheral vessels not affected by regional anesthesia are maximally vasoconstricted, a further drop in core temperature may not be prevented, because the body mass cephalad to the block is usually much smaller.

Once the patient's core temperature reaches the shivering threshold, shivering is initiated; however, neuraxial blockade reduces the gain of shivering by more than 60%, mainly due to a failure of the upper body muscles to compensate for lower body paralysis ( Kim et al., 1998 ). It is unlikely that the thermoregulatory changes seen under regional anesthesia are influenced by systemic absorption of local anesthetics, because a study aiming to generate equal plasma drug concentrations without regional anesthesia failed to reproduce these effects ( Glosten et al., 1991 ).

Compared with general anesthesia, the administration of a regional anesthetic technique reduces the risk of hypothermia, especially during surgery in which a small incision is made and the patient is kept well insulated. In contrast, with large surgical incisions, hypothermia can be profound and even more severe than with general anesthesia, and recovery to normal body temperature may be prolonged (Cattaneo et al., 2000 ).

In adults, threshold temperatures for sweating, vasoconstriction, and shivering during spinal anesthesia and epidural anesthesia seem to be comparable with a doubling of the interthreshold range ( Ozaki et al., 1994 ). It has been demonstrated in adults that the combination of general anesthesia with epidural anesthesia further reduces the threshold temperature for thermoregulatory vasoconstriction and thus significantly aggravates hypothermia compared with general anesthesia alone ( Joris et al., 1994 ). It is interesting that in this context, diabetic patients with autonomic neuropathy showed lower core temperatures and delayed thermoregulatory vasoconstriction during general anesthesia than diabetic patients without autonomic dysfunction ( Kitamura et al., 2000 ).

In contrast to the thermoregulatory effects of regional anesthesia in adults, the presence of a caudal block in children anesthetized with halothane has been shown not to significantly affect the threshold temperature for vasoconstriction in children (35.7°C without versus 35.9°C with caudal block) ( Bissonnette and Sessler, 1992 ). A survey revealed that only a third of clinicians is monitoring temperature during regional anesthesia ( Frank et al., 1999 ). From the aforementioned, it is obvious that temperature should also be monitored in these patients, as significant hypothermia is common, which otherwise remains undetected and therefore also untreated.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ ANESTHESIA AND HYPOTHERMIA

General anesthesia decreases the temperature threshold at which the body initiates a thermoregulatory response to cold stress. Mild intraoperative hypothermia (1° to 3°C below normal) is common and results from a combination of events:

  

1.   

An approximately 30% reduction in metabolic heat generation during anesthesia ( Brismar, 1982 )

  

2.   

Increased environmental exposure

  

3.   

Anesthetic-induced central inhibition of thermoregulation ( Sessler and Ponte, 1982 ; Sessler, 1991 )

  

4.   

Redistribution of heat within the body ( Hynson et al., 1991 )

Hypothermia has a typical profile during general anesthesia and usually develops in three phases ( Fig. 5-6 ).

  

1.   

Internal redistribution of heat

  

2.   

Thermal imbalance

  

3.   

Thermal steady state (plateau or rewarming)

 
 

FIGURE 5-6  The body of a child demonstrating the dynamic changes within the three temperature compartments before induction of anesthesia, during the first hour of anesthesia, and after thermoregulatory vasoconstriction. In the awake human, temperature and size of the compartments are well preserved and considered normal. Following induction of anesthesia, the internal redistribution of heat results in a rapid decrease in core temperature and an increase in peripheral temperature, together resulting in enlargement of the central and shrinkage of the peripheral compartments. The drop in central temperature is mainly caused by the distribution of its heat to a larger volume and actual heat loss is minimal at this stage. Once thermoregulatory vasoconstriction has been initiated (after about 3 hours of anesthesia), the central compartment is shrinking in favor of the peripheral compartment. The raise in central temperature is now due to generated heat being contained in a smaller volume.

 

 

▪ INTERNAL REDISTRIBUTION

To simplify the understanding of the internal redistribution concept, it is useful to divide the human body into three compartments: central, peripheral, and skin (or “shell”). The core temperature represents the central compartment temperature. The vessel-rich group organs are part of this central compartment and represent about 10% of the body weight in adults but up to 22% in neonates, and they receive about 75% of the cardiac output. At rest, the central compartment in an awake adult accounts for approximately 66% of the body mass and extends to about 71% during general anesthesia ( Deakin, 1998 ).

The peripheral compartment comprises the remaining part of the body mass and acts as a dynamic buffer to accommodate any changes in core temperature by vasodilatation or vasoconstriction. Its estimated buffer capacity of more than 600 kJ allows the body to maintain a constant core temperature with a minimal amount of energy spent for thermoregulation despite absorption or dissipation of significant amounts of heat.

The skin compartment (or “shell”) is almost virtual and represents the barrier between the previous two compartments and the environment. After induction of anesthesia, the peripheral vasodilatation causes an increase in the size of the central compartment, forcing it to redistribute its heat within a larger volume. Furthermore, the decrease in metabolic heat production caused by anesthesia reduces the amount of energy available to compensate for the enlargement of this compartment. The concept of internal redistribution of heat therefore consists not of heat loss to the environment but of a measurable decrease in central temperature and an increase in the peripheral and skin compartment temperatures due to redistribution of heat.

With induction of anesthesia, the central core temperature starts to decrease rapidly by approximately 0.5° to 1.5°C during the first 30 to 45 minutes of anesthesia ( Fig. 5-7 ). Although this process results in reduced core temperature, the total body heat decreases only slightly. Heat is—as the name implies—mainly redistributed and not dissipated. This redistribution of heat accounts for 81% of the core temperature decrease in the first hour of anesthesia, whereas the remainder is the result of the anesthesia-induced reduction in metabolism and increased heat loss. For the subsequent 2 hours of anesthesia, the impact of redistribution on total heat loss decreases to approximately 43% ( Matsukawa et al., 1995 ). Accordingly, by using a vasoconstrictor such as phenylephrine, the magnitude of hypothermia caused by redistribution can be decreased ( Ikeda et al., 1999 ).

 
 

FIGURE 5-7  This graph represents the three phases typical for the course of hypothermia during general anesthesia. A represents the internal redistribution of heat; B, the phase with ongoing net loss of heat to the environment; and C, the thermal steady state, which in children and infants is in fact a rewarming phase. The slope of each phase varies as a function of the age group.  (Modified from Bissonnette B: Thermoregulation and paediatric anaesthesia. Curr Opin Anaesthesia 6:537, 1993.)

 



The internal redistribution results in shrinkage of the peripheral and enlargement of the central compartment, which explains not only the decreased core temperature (the same amount of heat is now distributed to a larger volume) but also the increased temperature in the peripheral and skin compartments. This is reflected by a more than fourfold increase in the perfusion of forearms and particularly legs after induction of anesthesia, and a forearm-fingertip or calf-toe temperature gradient that may exceed 8°C ( Matsukawa et al., 1995 ).

▪ THERMAL IMBALANCE

This is the result of a combination of reduced heat production and increased heat loss to the environment. During this second phase, which lasts about 2 to 3 hours, the heat loss to the environment leads to an approximately linear decrease in mean body temperature (typically 0.5 to 1.0°C/hr). Anesthesia contributes to the decreased heat production by limiting muscular activity, reducing the metabolic rate, and eliminating the work of breathing ( Stoen and Sessler, 1990 ; Washington et al., 1992 ). Heat loss to the environment is a function of the temperature difference between body surface and ambient structures (concept of patient warming up the environment). Heat loss, therefore, decreases passively as patients become more hypothermic. Radiation, convection, evaporation, and conduction all contribute to heat loss from the patient to the environment during anesthesia and surgery.

▪ THERMAL STEADY STATE (PLATEAU OR REWARMING)

The third phase of the hypothermic response to anesthesia consists of a thermal steady state, where metabolic heat production equals heat dissipation to the environment and the core temperature therefore remains constant. This plateau occurs between 34.5° and 35.5°C. Thus, the patient must increase the heat production, decrease the heat loss, or both to prevent further hypothermia.

A study in adults undergoing isoflurane anesthesia showed that the effect of cutaneous vasoconstriction reduces heat loss by a maximum of 25%, which is relatively small compared with the fall in metabolic rate and the increase in evaporative heat losses from the surgical incision ( Sessler et al., 1992 ). It is presumed that this happens because heat loss to the environment is determined principally by the capillary blood flow in large areas of the skin covering the limbs and the trunk. These capillaries cannot constrict as effectively as the arteriovenous shunts but markedly outnumber the arteriovenous shunts. Thus, it is possible that vasoconstriction contributes to the thermal plateau by reestablishing the temperature gradient between the central and the peripheral compartments and thereby preventing metabolic heat from being transported to the periphery, from where it would dissipate. The metabolic heat produced in the body core is distributed to a now smaller central compartment, allowing the temperature of this compartment to be maintained at a constant level.

To reinforce this theory of compartment size, it should be noted that the use of a limb tourniquet during surgical procedures influences the thermoregulatory response in children and adults ( Bloch et al., 1992 ; Estebe et al., 1996 ). The tourniquet-induced hyperthermia is most likely due to reduction in the size of the peripheral compartment and the containment of metabolic heat within the central thermal compartment. Accordingly, the core temperature drops after deflation of the tourniquet ( Estebe et al., 1996 ; Sanders et al., 1996 ; Akata et al., 1998 ). Despite a now constant core temperature, total body heat content is diminishing as heat loss to the environment continues.

In contrast to that in adults, the third phase in infants and children is a rewarming rather than a plateau phase (see Fig. 5-7 ). As mentioned earlier, general anesthesia decreases heat production by inhibiting muscular activity and nonshivering thermogenesis and by reducing metabolic rate production. Thus, the only possible explanation for this rewarming phase must be the occurrence of marked vasoconstriction within the peripheral and central compartments, leading to shrinkage of the central compartment. Thus, the amount of metabolic heat produced is distributed within a smaller central compartment volume, resulting in raised core temperature. Furthermore, this is also associated with a simultaneous increase in oxygen consumption, CO2 production, and systemic norepinephrine levels, which have been observed in infants anesthetized with isoflurane and paralyzed with vecuronium (B. Bissonnette, unpublished data).

Infants differ from adults in that intraoperative thermoregulatory responses are sufficiently effective to significantly increase the core temperature despite constant ambient temperatures. A clinical study in children undergoing general anesthesia for surgery found a twofold increase in    o2during mild hypothermia ( Ryan, 1982 ). With either active or passive rewarming, significant physiologic stress is imposed on the infant. Passive surface rewarming (with the use of warm blankets, bundling, or other measures) turns off the skin cold receptors. If the normal core temperature is not reached or maintained with passive surface rewarming, hypothermia may result in hypoventilation or even apnea, relative anesthetic overdose (reduced MAC at lower temperatures), and finally metabolic acidosis ( Fig. 5-8 ). The increased oxygen demand to maintain the normal core temperature in the anesthetized infant may create or exacerbate a preexisting cardiopulmonary insufficiency. The release of norepinephrine to trigger vasoconstriction may contribute to the development of acidosis and hypoxia, thereby increasing right-to-left pulmonary shunting. Sustained pulmonary artery hypertension and right-to-left pulmonary shunting may lead to the formation of a vicious cycle.

 
 

FIGURE 5-8  Vicious cycle resulting from hypothermia in neonates.  (Modified from Klaus M, Faranoff A: Care of the high-risk neonate. Philadelphia, 1986, WB Saunders.)

 



In adult patients, a correlation of intraoperative hypothermia with an early increase in postoperative    o2 has been demonstrated ( Roe et al., 1966 ). In addition, this study evaluated the effects of various anesthetic agents on postoperative    o2. Although halothane anesthesia in adults was associated with the largest increase in    o2, no anesthetic agent or combination of agents in pediatric patients has been shown to clearly offer more protection from the adverse effects of hypothermia. It has been demonstrated in adults that although anesthetics can modify the thermoregulatory response to hypothermia, anesthetized patients are not poikilothermic ( Sessler et al., 1987 ).

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ ANESTHESIA AND HYPERTHERMIA

Similar to hypothermia, hyperthermia triggers important physiologic thermoregulatory responses using threshold and gains. The threshold represents the central temperature for which a particular regulatory effector becomes active, whereas the gain quantifies the intensity of the response (see Fig. 5-4 ). The effector mechanisms during hyperthermia are well preserved during anesthesia when central temperature increases. Regarding controlled hyperthermia (i.e., increased central temperature), it has been demonstrated that the efferent responses seen in awake individuals were also preserved in anesthetized subjects ( Lopez et al., 1993 ). The efferent response threshold was shifted to higher temperatures, thereby creating an expansion of the interthreshold range, which corresponds to the difference between the normal central temperature and the first efferent response triggered by the hypothalamus. In a healthy, awake person, the variability of the system (interthreshold range) is only 0.4°C. Within this range, it is said that the individual is poikilothermic; that is, the change in central temperature does not trigger any thermoregulatory effector responses. The interesting observation in regard to the interthreshold range resides in the difference between the shift observed in hypothermia and the shift in hyperthermia.

The poikilothermic range to the hypothermic side in the anesthetized patient may be expanded up to 2.5° to 3.5°C. Clinical studies in human volunteers have suggested that the threshold for active vasodilatation and sweating was only 1.0° to 1.4°C higher in anesthetized than in awake individuals ( Lopez et al., 1993 ). This observation suggests that the human physiology responds more aggressively to the threats of hyperthermia than it does for hypothermia. Thus, the speculative explanation is that hyperthermia is far more dangerous than a comparable degree of hypothermia ( Lopez et al., 1993 ).

The efferent responses during hyperthermic stress in anesthesia are limited to two mechanisms: active vasodilatation and sweating. The vasodilatation triggered in response to warm stress is not simply the absence of vasoconstriction but rather an active and effective vasodilatation resulting in increased dissipation of heat ( Detry et al., 1972 ; Rübsamen and Hales, 1984 ). It has been demonstrated that the effect of hyperthermia on the peripheral vasculature causes a significant increase in blood flow ( Tankersley et al., 1991 ; Matsukawa et al., 1995 ). The observation of active cutaneous vasodilatation in infants under anesthesia, although difficult to quantify (skin flushing), suggests that the thermoregulatory response to hyperthermia may be preserved.

Sweating represents an increase in evaporative cutaneous heat loss during episodes of heat stress. The relatively high heat of vaporization of sweat (2.5 · 106 J/kg) makes sweating an extremely effective process. It allows an up to fivefold increase in heat loss to the environment, making it proportionally more effective than all of the defense mechanisms against cold combined ( Fusi et al., 1989 ).

A study in adult volunteers showed that sweating remains functional during isoflurane anesthesia ( Sessler, 1991 ). It has also been demonstrated that men sweat more than women. In infants and children who weigh less than 15 kg, scientific evidence suggests that sweating under anesthesia in this age group is less effective than in older children and adults (B. Bissonnette, unpublished data).

The benefits provided by induced hyperthermia may be desirable during peripheral microvascular surgery when an increase in regional blood flow is important. The physiologic relationship between maximal vasodilatation and sweating is not fully understood, and the possibility that maximal vascular dilatation may not occur until the central temperature increases even further remains possible. One of the clinical limitations of the use of induced hyperthermia in increasing cutaneous blood flow is the efficiency of the sweating mechanism. Despite active transfer of about 50 W across the patient's skin via convection and radiation ( Sessler, 1993 ), it was possible to show that the central temperature remains relatively constant or even decreases. Although shivering can easily double the heat production, sweating can result in the dissipation of more than 10 times the amount of normal basal heat production ( Guyton, 2000 ).

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ PREVENTION OF HYPOTHERMIA

Heat loss in an infant can occur for a variety of reasons. Exposure of body cavities to low environmental temperatures and humidity, infusion of cold fluids, and ventilation with cold and dry gases, in combination with the infant's physical characteristics of the large body surface area-to-volume ratio and the minimal insulating tissue layer, all increase the potential for an infant or a child to become hypothermic during anesthesia. Nevertheless, hypothermia must not be viewed as an inevitable consequence of surgery. Although hypothermia may be protective in a small subgroup of patients with certain ischemic conditions ( Illievich et al., 1994 ), for the majority of our patients the adverse effects outweigh the benefits and inadvertent core hypothermia must be avoided.

A study in 200 adult patients scheduled for colorectal surgery showed that patients who were allowed to become hypothermic (34.7 ± 0.6°C) during the procedure showed a more than three-fold higher rate of surgical wound infection than did the group who was actively warmed to keep the body normothermic (36.6 ± 0.5°C) ( Kurz et al., 1996 ). Furthermore, the times to suture removal and discharge from hospital in the hypothermia group were prolonged by 1 and 2.6 days, respectively. It has been suggested by these and other researchers ( Jonsson et al., 1991 ) that the vasoconstriction triggered by hypothermia may result in a decreased partial pressure of oxygen in the tissues, which leads to increased wound infection and, finally, delayed wound healing, even in the absence of an infection ( Fig. 5-9 ). A shorter hospital stay for normothermic versus hypothermic patients has also been reported by others ( Sellden and Lindahl, 1999 ). In addition, hypothermia per se has been shown to reduce chemotaxis and phagocytosis of granulocytes, natural killer cell cytotoxicity, migration of macrophages, and synthesis of immunoglobulins and thereby directly affect the immune response ( Leijh et al., 1979 ; van Oss et al., 1980 ; Wenisch et al., 1996 ; Beilin et al., 1998 ).

 
 

FIGURE 5-9  Hypothermia and its consequences on immune function and wound healing (↓ represents a decrease; ↑, an increase in the corresponding response).

 

 

Hypothermia significantly delays the reactions of the coagulation cascade (most likely due to a direct effect on the activity of the coagulation factors) with prolongation of prothrombin time and partial thromboplastin time ( Rohrer and Natale, 1992 ). An inhibition of platelet function with prolonged bleeding time was found during hypothermia, secondary to inhibited upregulation of platelet surface protein GMP-140 and downregulation of the glycoprotein GP Ib-IX complex, reduced platelet aggregation, and thromboxane B2 (the stable metabolite of thromboxane A2) generation ( Michelson et al., 1994 ). This platelet dysfunction is fully reversible with rewarming. Thromboelastography confirms these findings with a prolongation of the reaction and coagulation times as well as a reduction in the clot formation rate ( Douning et al., 1995 ). It is therefore not surprising that blood loss in patients undergoing hip arthroplasty was found to be significantly higher in the hypothermia group (core temperature, 35 ± 0.5°C) than in the normothermia group (core temperature, 36.6 ± 0.4°C) ( Schmied et al., 1996 ). Similar results have been confirmed by other researchers ( Bock et al., 1998 ).

A significantly greater incidence of myocardial ischemia and Pao2 values below 80 mm Hg has been reported in hypothermic patients (defined by sublingual temperature measured on arrival to the recovery room) compared with normothermic patients during the first 24 hours after lower extremity vascular surgery with either epidural or general anesthesia ( Frank et al., 1993 ). Although these data are most likely not of relevance to most pediatric patients due to the absence of coronary heart disease, they nevertheless demonstrate the widespread and potential impact of hypothermia on the body in surgical patients.

Hypothermia affects drug metabolism and results in diminished metabolism and prolonged action. Although in adults intraoperative hypothermia resulted in delayed recovery from anesthesia compared with normothermic patients ( Lenhardt et al., 1997 ), no such differences could be demonstrated in children ( Bissonnette and Sessler, 1993 ). Studies on the effects of hypothermia and muscle relaxants have demonstrated that hypothermia decreases the requirements for nondepolarizing muscle relaxants, due to an increased sensitivity of the neuromuscular junction and to diminished biliary (indicating a reduced affinity for the drug substrate to microsomal enzymes) and renal elimination of the drug ( Ham et al., 1978 ; Miller et al., 1978 ). The same has been confirmed for other medications ( McAllister and Tan, 1980 ).

Even in the absence of similar studies for the pediatric population, there are no reasons to believe that the results would be significantly different from those in adults. Avoiding hypothermia in the infant and child is therefore crucial and requires not only an understanding of thermal physiology but also meticulous attention to detail in anesthetic care. The following recommendations are intended to help minimize intraoperative heat loss.

▪ OPERATING ROOM TEMPERATURE

Because evaporative heat losses from the respiratory tract account for only about 5% to 10% of total heat loss during anesthesia, it is obvious that keeping the operating room temperature at an optimal level is crucial in the prevention of hypothermia. The major source of heat loss in the anesthetized patient is radiation. As mentioned previously, radiant heat loss is a function of the temperature difference between that of the patient and that of the environment. The effectiveness of controlling ambient operating room temperature in regulating the temperature of newborns during surgery has been demonstrated ( Bennett et al., 1977 ). In adults, 21°C is reported as the critical ambient temperature for maintaining normal (36° to 37.5°C) nasopharyngeal or esophageal temperatures ( Morris and Wilkey, 1970 ; Morris, 1971a, 1971b [158] [159]). Operating room temperatures of 27° and 29°C are recommended for full-term and premature newborns, respectively. It is essential that every operating room be equipped with an individual thermistor control unit so that the temperature in each operating room can be controlled individually to meet the needs of each child.

▪ RADIANT HEATERS

Radiant heaters are used during induction of anesthesia and insertion of catheters, until the patient is prepared and draped. Prolonged use of radiant heaters may result in increased insensible water losses. Also, if radiant heaters are too close to the patient, they can cause skin burns.

▪ REFLECTING BLANKETS

The use of reflective blankets in adults has produced conflicting results, and data on their use in infants and children are sparse. In adults, it has been reported that normothermia can be maintained with reflective blankets if 60% or more of the patient's body surface area is covered ( Bourke et al., 1984 ). Although wrapping infants in reflective blankets to cover 60% of the body surface area may be cumbersome or impossible, we recommend that uninvolved skin areas be covered. Of particular concern in this regard is the head, which comprises up to 20% of the total skin surface area in a neonate and shows the highest regional heat flux ability ( Anttonen et al., 1995 ). Facial cooling increases oxygen requirements by 23% in the term infant and by 36% in the premature infant ( Sinclair, 1972 ). The practice of covering the head with a plastic bag easily and significantly reduces radiant, convective, and evaporative heat loss. Passive insulators are commonly used to prevent cutaneous heat loss. Insulating covers may be chosen on the basis of cost and convenience. Most likely, the percentage of skin surface area covered is more important than the choice of insulating material or the skin region covered ( Sessler et al., 1991 ) ( Fig. 5-10 ).

 
 

FIGURE 5-10  Because the head of an infant constitutes a large fraction of its body size, covering the infant's head and uninvolved areas of the child's body with plastic wrap can greatly affect evaporative heat loss.

 

 

▪ SKIN-WARMING DEVICES

The use of skin surface-warming devices in adults before the induction of anesthesia reduces the magnitude of the hypothermic response resulting from internal redistribution of heat ( Glosten et al., 1991 ). Aggressive skin surface warming induces peripheral vasodilatation and favorably increases the temperature of the peripheral compartment to values approaching those of the central compartment. The net result is an increased mean body temperature, because skin surface warming reduces the amount of energy transferred from the central to the peripheral compartment after induction of anesthesia. A variety of passive and active skin surface warmers are available, including circulating hot water blankets ( Stephen et al., 1960 ; Vale and Lunn, 1969 ), infrared radiant heaters ( Morris, 1971 ; Goldblat and Miller, 1972 ), and convective forced-air heaters, which blow warm air through a disposable blanket to raise the effective ambient temperature immediately surrounding the patient ( Sessler and Moayeri, 1990 ;Steele et al., 1996 ). A new system simulating water immersion with use of a special garment and feedback algorithms analyzed by computer to achieve a preset body temperature is available and seems to perform well and safe in children ( Nesher et al., 2001 ). Of all these devices, convective forced-air warmers are by far the most effective. They not only can maintain a certain body temperature but also can rewarm a hypothermic patient ( Sessler and Moayeri, 1990 ; Kurz et al., 1993 ; Ciufo et al., 1995 ; Karayan et al., 1996 ).

▪ WARMING MATTRESSES

The use of warming mattresses reduces conductive heat loss. Warming mattresses set at 40°C and covered with two layers of cotton blankets have been demonstrated to effectively conserve heat (Goudsouzian et al., 1973 ). This measure was especially significant for infants with a surface area of less than 0.5 m2. In older children and adults, only a small proportion of the skin surface area is in contact with the heating mattress, which makes it generally ineffective.

▪ WARMING FLUIDS

It is well known that the rapid infusion of chilled (1° to 6°C) intravenous fluids can be used effectively to induce hypothermia. The administration of 1 L of an ice-cold infusion in an adult is expected to drop core temperature by approximately 1.7°C, but values of up to 3°C are possible ( Baumgardner et al., 1999 ; Rajek et al., 2000 ). Intravenous fluids and blood products should thus be warmed before administration. It is particularly important to warm fluids in instances of rapid or massive fluid administration. In addition, attention should be paid to the length and the type of the infusion tubing, because significant heat loss of the infusate during transit from the warming device to the intravenous cannula may occur ( Faries et al., 1991 ; Bissonnette and Paut, 2002 ). Although a study demonstrated that conservative fluid management (1 mL/kg per hour of crystalloids warmed to 37°C) in patients aged 1 to 3 years resulted in less core hypothermia compared with the control group that received aggressive fluid replacement (10 mL/kg per hour) ( Ezri et al., 2003 ), this should not preclude pediatric patients from receiving appropriate fluid resuscitation.

Because both the peritoneal and thoracic cavities are large heat-exchanging areas, solutions for intraoperative lavage should always be warmed to body temperature. Although desirable from a thermoregulatory point of view, preparation solutions should not be warmed, because heat can cause a chemical breakdown of the iodine solution and thereby inactivate its antimicrobic properties.

▪ HUMIDIFIED AND HEATED GASES

To minimize convective and evaporative heat losses via the respiratory tract, inspiratory gases should be heated and humidified. Airway humidification in intubated patients prevents tracheal damage from dry inspired gases ( Chalon et al., 1979 ), increases tracheal mucus flow ( Forbes, 1974 ), and minimizes respiratory heat losses ( Berry et al., 1973 ; Tollofsrud et al., 1984 ). Heat and humidity can be added actively to inspired gases by evaporative or ultrasonic heated humidifiers or passively by heat and moisture exchanging filters (“artificial noses”) ( Newton, 1975 ; Chalon et al., 1984 ; Bissonnette et al., 1989 ). Furthermore, there is considerable evidence that a relative humidity of at least 50% maintains normal ciliary function in the respiratory tract and helps prevent bronchospasm ( Chalon et al., 1972 ;Forbes, 1974 ; Mercke, 1975 ). Humidification to 50% is easily obtained with heat and moisture exchanging filters and is mandatory for long procedures. In adults, heating and humidifying of gases to 37°C and 100% relative humidity not only effectively maintains normothermia but also reverses hypothermia during general surgery ( Pflug et al., 1978 ; Stone et al., 1981 ).

Because of the higher minute ventilation per kilogram of body weight in pediatric patients, airway humidification is even more effective in maintaining normothermia than in adults ( Bissonnette and Sessler, 1989 ; Bissonnette et al., 1989 ). In newborns, heat loss during general anesthesia is significantly reduced when heated, humidified gases are used instead of dry anesthetic gases ( Fonkalsrud et al., 1980 ). Although high-temperature humidification can decrease intraoperative evaporative heat loss, it may result in tracheal burns. Because there may be a large temperature gradient between the humidifier and the endotracheal tube, it is important to measure airway temperature as close to the airway as possible. This has the advantage of preventing heated gases from accidentally burning the trachea while providing the warmest possible inspiratory gas. On the other hand, it results in decreased intraoperative heat loss. If these devices are used, it is recommended to heat inspiratory gases to normal body temperature only. Although heat and moisture exchanging filters are less effective than active humidifiers (especially in the first hour after the induction of anesthesia), they seem to provide a reasonable alternative ( Bissonnette et al., 1989b ).

Additional advantages of the use of heat and moisture exchangers in small infants include the absent danger of airway burns and risk of overhumidification with the consequences of overhydration, and the reduced risk of breathing circuit disconnection ( Smith and Allen, 1986 ; Shroff and Skerman, 1988 ). A heat-moisture exchanger falsely increases the esophageal temperature by about 0.35°C above tympanic temperature or mean body temperature ( Bissonnette et al., 1989a ).

The addition of an artificial nose with a dead space of approximately 1 mL introduces trivial airway resistance and can be safely used even in the smallest infants ( Jones et al., 1988 ). Differences in the efficiency among heat and moisture exchangers are negligible ( Baumgarten, 1985 ; Bickler and Sessler, 1990 ).

▪ TRANSPORTATION

Care in the transportation of an infant cannot be overemphasized. All intraoperative efforts at maintaining intraoperative thermal stability can be lost during even a brief transport to either the post-anesthetic care room or the intensive care unit. It is therefore essential that the incubator be warmed before transport both to and from the operating room. Older infants and children should at least be covered with a warmed blanket during transport.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

▪ SUMMARY

Because of their small size with increased body surface area-to-body weight ratio and increased thermal conductance, infants and young children are at significant risk for thermal instability. This risk is even more pronounced for premature and small-for-gestational-age infants. Although unanesthetized infants are able to maintain homeothermic functions, they can do so only within a narrow range of ambient temperatures. In undergoing general anesthesia for surgery, the combination of exposure to the operating room with its usually low ambient temperature and high airflow and the use of cold infusions and dry anesthetic gases can easily overwhelm the thermal homeostatic mechanisms and, in certain instances, result in potentially serious complications. A better understanding of the physiology of the temperature-regulation system during anesthesia has improved the recognition, prevention, and management of these perioperative disturbances. The identification of the hypothermia and hyperthermia patterns in relation to the severity and the duration of the anesthetic procedure has contributed to this improvement. Furthermore, the knowledge of the different effects of each anesthetic agent on the thermoregulatory mechanism undoubtedly proves useful in providing a safe anesthesia.

Copyright © 2008 Elsevier Inc. All rights reserved. - www.mdconsult.com

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier

REFERENCES

Adamson et al., 1965. Adamson Jr KJ, Gandy GM, James LS: The influence of thermal factors upon oxygen consumption of the newborn human infant.  J Pediatr  1965; 66:495.

Akata et al., 1998. Akata T, Kanna T, Izumi K, et al: Changes in body temperature following deflation of limb pneumatic tourniquet.  J Clin Anesth  1998; 10:17.

Alfonsi et al., 1998. Alfonsi P, Sessler DI, Du Manoir B, et al: The effects of meperidine and sufentanil on the shivering threshold in postoperative patients.  Anesthesiology  1998; 89:43.

Allen, 1987. Allen JA: The thermal environment and human heat exchange.   In: Ernsting J, King P, ed. Aviation medicine,  London: Butterworths; 1987:219.

Annadata et al., 1995. Annadata R, Sessler DI, Tayefeh F, et al: Desflurane slightly increases the sweating threshold but produces marked, nonlinear decreases in the vasoconstriction and shivering thresholds.  Anesthesiology  1995; 83:1205.

Antognini, 1993. Antognini JF: Hypothermia eliminates isoflurane requirements at 20 degrees C.  Anesthesiology  1993; 78:1152.

Antognini et al., 1994. Antognini JF, Lewis BK, Reitan JA: Hypothermia minimally decreases nitrous oxide anesthetic requirements.  Anesth Analg  1994; 79:980.

Anttonen et al., 1995. Anttonen H, Puhakka K, Niskanen J, et al: Cutaneous heat loss in children during anaesthesia.  Br J Anaesth  1995; 74:306.

Baumgardner et al., 1999. Baumgardner JE, Baranov D, Smith DS, et al: The effectiveness of rapidly infused intravenous fluids for inducing moderate hypothermia in neurosurgical patients.  Anesth Analg  1999; 89:163.

Baumgarten, 1985. Baumgarten RK: Humidifiers are unjustified in adult anesthesia.  Anesth Analg  1985; 64:1224.

Beilin et al., 1998. Beilin B, Shavit Y, Razumovsky J, et al: Effects of mild perioperative hypothermia on cellular immune responses.  Anesthesiology  1998; 89:1133.

Bennett et al., 1977. Bennett EJ, Patel KP, Grundy EM: Neonatal temperature and surgery.  Anesthesiology  1977; 46:303.

Benzinger, 1969. Benzinger M: Tympanic thermometry in surgery and anesthesia.  JAMA  1969; 209:1207.

Benzinger, 1969. Benzinger TH: Heat regulation: homeostasis of central temperature in man.  Physiol Rev  1969; 49:671.

Bernard et al., 1998. Bernard JM, Fulgencio JP, Delaunay L, et al: Clonidine does not impair redistribution hypothermia after the induction of anesthesia.  Anesth Analg  1998; 87:168.

Berry et al., 1973. Berry Jr FA, Hughes-Davies DI, DiFazio CA: A system for minimizing respiratory heat loss in infants during operation.  Anesth Analg  1973; 52:170.

Bickler and Sessler, 1990. Bickler PE, Sessler DI: Efficiency of airway heat and moisture exchangers in anesthetized humans.  Anesth Analg  1990; 71:415.

Bissonnette et al., 2000. Bissonnette B, Holtby HM, Davis AJ, et al: Cerebral hyperthermia in children after cardiopulmonary bypass.  Anesthesiology  2000; 93:611.

Bissonnette and Paut, 2002. Bissonnette B, Paut O: Active warming of saline or blood is ineffective when standard infusion tubing is used: an experimental study.  Can J Anaesth  2002; 49:270.

Bissonnette and Sessler, 1993. Bissonnette B, Sessler DI: Mild hypothermia does not impair postanesthetic recovery in infants and children.  Anesth Analg  1993; 76:168.

Bissonnette and Sessler, 1989. Bissonnette B, Sessler DI: Passive or active inspired gas humidification increases thermal steady-state temperatures in anesthetized infants.  Anesth Analg  1989; 69:783.

Bissonnette and Sessler, 1990. Bissonnette B, Sessler DI: The thermoregulatory threshold in infants and children anesthetized with isoflurane and caudal bupivacaine.  Anesthesiology  1990; 73:1114.

Bissonnette and Sessler, 1992. Bissonnette B, Sessler DI: Thermoregulatory thresholds for vasoconstriction in pediatric patients anesthetized with halothane or halothane and caudal bupivacaine.  Anesthesiology  1992; 76:387.

Bissonnette et al., 1989a. Bissonnette B, Sessler DI, LaFlamme P: Intraoperative temperature monitoring sites in infants and children and the effect of inspired gas warming on esophageal temperature.  Anesth Analg  1989; 69:192.

Bissonnette et al., 1989b. Bissonnette B, Sessler DI, LaFlamme P: Passive and active inspired gas humidification in infants and children.  Anesthesiology  1989; 71:350.

Bloch et al., 1992. Bloch EC, Ginsberg B, Binner Jr RA, et al: Limb tourniquets and central temperature in anesthetized children.  Anesth Analg  1992; 74:486.

Bock et al., 1998. Bock M, Muller J, Bach A, et al: Effects of preinduction and intraoperative warming during major laparotomy.  Br J Anaesth  1998; 80:159.

Boulant and Bignall, 1973. Boulant JA, Bignall KE: Hypothalamic neuronal responses to peripheral and deep-body temperatures.  Am J Physiol  1973; 225:1371.

Boulant and Demieville, 1977. Boulant JA, Demieville HN: Responses of thermosensitive preoptic and septal neurons to hippocampal and brain stem stimulation.  J Neurophysiol  1977; 40:1356.

Boulant and Hardy, 1974. Boulant JA, Hardy JD: The effect of spinal and skin temperatures on the firing rate and thermosensitivity of preoptic neurones.  J Physiol  1974; 240:639.

Bourke et al., 1984. Bourke DL, Wurm H, Rosenberg M, et al: Intraoperative heat conservation using a reflective blanket.  Anesthesiology  1984; 60:151.

Brismar et al., 1982. Brismar B, Hedenstierna G, Lundh R, et al: Oxygen uptake, plasma catecholamines and cardiac output during neurolept-nitrous oxide and halothane anaesthesias.  Acta Anaesthesiol Scand  1982; 26:541.

Brown et al., 1992. Brown PJ, Dove RA, Tuffnell CS, et al: Oscillations of body temperature at night.  Arch Dis Child  1992; 67:1255.

Bruck, 1961. Bruck K: Temperature regulation in the newborn infant.  Biol Neonate  1961; 3:65.

Brück, 1992. Brück K: Neonatal thermal regulation.   In: Polin RA, Fox WW, ed. Fetal and neonatal physiology,  Philadelphia: WB Saunders; 1992:493.

Bruck and Wunnenberg, 1965. Bruck K, Wunnenberg B: The influence of ambient temperature in the process of replacement of non-shivering thermogenesis during postnatal development.  Fed Proc  1965; 25:1332.

Brundin and Wahren, 1994. Brundin T, Wahren J: Effects of i.v. amino acids on human splanchnic and whole body oxygen consumption, blood flow, and blood temperatures.  Am J Physiol  1994; 266:E396.

Budin, 1907. Budin P: The nursling,  London, Caxton Publishing Co., 1907.

Burgess et al., 1978. Burgess 3rd GE, Cooper JR, Marino RJ, et al: Continuous monitoring of skin temperature using a liquid-crystal thermometer during anesthesia.  South Med J  1978; 71:516.

Burton, 1935. Burton AC: Human calorimetry: The average temperature of the tissues of the body.  J Nutr  1935; 9:261.

Cabanac, 1975. Cabanac M: Temperature regulation.  Annu Rev Physiol  1975; 37:415.

Cattaneo et al., 2000. Cattaneo CG, Frank SM, Hesel TW, et al: The accuracy and precision of body temperature monitoring methods during regional and general anesthesia.  Anesth Analg  2000; 90:938.

Chalon et al., 1972. Chalon J, Loew DA, Malebranche J: Effects of dry anesthetic gases on tracheobronchial ciliated epithelium.  Anesthesiology  1972; 37:338.

Chalon et al., 1984. Chalon J, Markham JP, Ali MM, et al: The Pall Ultipor breathing circuit filter—an efficient heat and moisture exchanger.  Anesth Analg  1984; 63:566.

Chalon et al., 1979. Chalon J, Patel C, Ali M, et al: Humidity and the anesthetized patient.  Anesthesiology  1979; 50:195.

Cheung and Mekjavic, 1995. Cheung SS, Mekjavic IB: Human temperature regulation during subanesthetic levels of nitrous oxide-induced narcosis.  J Appl Physiol  1995; 78:2301.

Ciofolo et al., 1989. Ciofolo MJ, Clergue F, Devilliers C, et al: Changes in ventilation, oxygen uptake, and carbon dioxide output during recovery from isoflurane anesthesia.  Anesthesiology  1989; 70:737.

Ciufo et al., 1995. Ciufo D, Dice S, Coles C: Rewarming hypothermic postanesthesia patients: a comparison between a water coil warming blanket and a forced-air warming blanket.  J Post Anesth Nurs  1995; 10:155.

Coffman and Cohen, 1971. Coffman JD, Cohen AS: Total and capillary fingertip blood flow in Raynaud's phenomenon.  N Engl J Med  1971; 285:259.

Colin et al., 1971. Colin J, Timbal J, Houdas Y, et al: Computation of mean body temperature from rectal and skin temperatures.  J Appl Physiol  1971; 31:484.

Cork et al., 1983. Cork RC, Vaughan RW, Humphrey LS: Precision and accuracy of intraoperative temperature monitoring.  Anesth Analg  1983; 62:211.

Cross et al., 1958. Cross K, Tizard J, Trythall D: Gaseous metabolism of newborn infant breathing 15% oxygen.  Acta Paediatr  1958; 47:217.

Dawkins and Scopes, 1965. Dawkins MJ, Scopes JW: Non-shivering thermogenesis and brown adipose tissue in the human new-born infant.  Nature  1965; 206:201.

De Witte et al., 1998. De Witte JL, Kim JS, Sessler DI, et al: Tramadol reduces the sweating, vasoconstriction, and shivering thresholds.  Anesth Analg  1998; 87:173.

Deakin, 1998. Deakin CD: Changes in core temperature compartment size on induction of general anaesthesia.  Br J Anaesth  1998; 81:861.

Delaunay et al., 1996. Delaunay L, Herail T, Sessler DI, et al: Clonidine increases the sweating threshold, but does not reduce the gain of sweating.  Anesth Analg  1996; 83:844.

Detry et al., 1972. Detry JM, Brengelmann GL, Rowell LB, et al: Skin and muscle components of forearm blood flow in directly heated resting man.  J Appl Physiol  1972; 32:506.

Dickenson, 1977. Dickenson AH: Specific responses of rat raphe neurones to skin temperature.  J Physiol  1977; 273:277.

Dicker et al., 1995. Dicker A, Ohlson KB, Johnson L, et al: Halothane selectively inhibits nonshivering thermogenesis. Possible implications for thermoregulation during anesthesia of infants.  Anesthesiology  1995; 82:491.

Douning et al., 1995. Douning LK, Ramsay MA, Swygert TH, et al: Temperature corrected thromboelastography in hypothermic patients.  Anesth Analg  1995; 81:608.

Downey et al., 1964. Downey J, Mottram R, Pickering G: The location by regional cooling of central temperature receptors in the conscious rabbit.  J Physiol (Lond)  1964; 170:415.

Downey et al., 1967. Downey JA, Chiodi HP, Darling RC: Central temperature regulation in the spinal man.  J Appl Physiol 2  1967; 2:91.

Eger and Johnson, 1987. Eger 2nd EI, Johnson BH: MAC of I-653 in rats, including a test of the effect of body temperature and anesthetic duration.  Anesth Analg  1987; 66:974.

Engelman and Lockhart, 1972. Engelman DR, Lockhart CH: Comparisons between temperature effects of ketamine and halothane anesthesia in children.  Anesth Analg  1972; 51:98.

Estebe et al., 1996. Estebe JP, Le Naoures A, Malledant Y, et al: Use of a pneumatic tourniquet induces changes in central temperature.  Br J Anaesth  1996; 77:786.

Ezri et al., 2003. Ezri T, Szmuk P, Weisenberg M, et al: The effects of hydration on core temperature in pediatric surgical patients.  Anesthesiology  2003; 98:838.

Faries et al., 1991. Faries G, Johnston C, Pruitt KM, et al: Temperature relationship to distance and flow rate of warmed i.v. fluids.  Ann Emerg Med  1991; 20:1198.

Fonkalsrud et al., 1980. Fonkalsrud EW, Calmes S, Barcliff LT, et al: Reduction of operative heat loss and pulmonary secretions in neonates by use of heated and humidified anesthetic gases.  J Thorac Cardiovasc Surg  1980; 80:718.

Forbes, 1974. Forbes AR: Temperature, humidity and mucus flow in the intubated trachea.  Br J Anaesth  1974; 46:29.

Frank et al., 1993. Frank SM, Beattie C, Christopherson R, et al: Unintentional hypothermia is associated with postoperative myocardial ischemia. The Perioperative Ischemia Randomized Anesthesia Trial Study Group.  Anesthesiology  1993; 78:468.

Frank et al., 2000. Frank SM, El-Rahmany HK, Cattaneo CG, et al: Predictors of hypothermia during spinal anesthesia.  Anesthesiology  2000; 92:1330.

Frank et al., 1999. Frank SM, Nguyen JM, Garcia CM, et al: Temperature monitoring practices during regional anesthesia.  Anesth Analg  1999; 88:373.

Fraser, 1978. Fraser JG: Iatrogenic benign hyperthermia in children.  Anesthesiology  1978; 48:375.

Fusi et al., 1989. Fusi L, Steer PJ, Maresh MJ, et al: Maternal pyrexia associated with the use of epidural analgesia in labour.  Lancet  1989; 1:1250.

Gale, 1973. Gale CC: Neuroendocrine aspects of thermoregulation.  Annu Rev Physiol  1973; 35:391.

Garruti and Ricquier, 1992. Garruti G, Ricquier D: Analysis of uncoupling protein and its mRNA in adipose tissue deposits of adult humans.  Int J Obes Relat Metab Disord  1992; 16:383.

Giesbrecht et al., 1997. Giesbrecht GG, Goheen MS, Johnston CE, et al: Inhibition of shivering increases core temperature afterdrop and attenuates rewarming in hypothermic humans.  J Appl Physiol  1997; 83:1630.

Giesbrecht et al., 1994. Giesbrecht GG, Sessler DI, Mekjavic IB, et al: Treatment of mild immersion hypothermia by direct body-to-body contact.  J Appl Physiol  1994; 76:2373.

Gilbert et al., 2000. Gilbert M, Busund R, Skagseth A, et al: Resuscitation from accidental hypothermia of 13.7 degrees C with circulatory arrest.  Lancet  2000; 355:375.

Glosten et al., 1991. Glosten B, Sessler DI, Ostman LG, et al: Intravenous lidocaine does not cause shivering-like tremor or alter thermoregulation.  Reg Anesth  1991; 16:218.

Goldblat and Miller, 1972. Goldblat A, Miller R: Prevention of incidental hypothermia in neurosurgical patients.  Anesth Analg  1972; 51:536.

Goto et al., 1999. Goto T, Matsukawa T, Sessler DI, et al: Thermoregulatory thresholds for vasoconstriction in patients anesthetized with various 1-minimum alveolar concentration combinations of xenon, nitrous oxide, and isoflurane.  Anesthesiology  1999; 91:626.

Goudsouzian et al., 1973. Goudsouzian NG, Morris RH, Ryan JF: The effects of a warming blanket on the maintenance of body temperatures in anesthetized infants and children.  Anesthesiology  1973; 39:351.

Grant and Bland, 1931. Grant RJ, Bland E: Observations on arteriovenous anastomoses in human skin and in bird's foot with special reference to the reaction to cold.  Heart  1931; 15:385.

Guieu and Hardy, 1970. Guieu JD, Hardy JD: Effects of heating and cooling of the spinal cord on preoptic unit activity.  J Appl Physiol  1970; 29:675.

Gurtner et al., 2002. Gurtner C, Paut O, Bissonnette B: Temperature regulation: physiology and pharmacology.   In: Bissonnette B, Dalens B, ed. Pediatric anesthesia. Principles and practice,  New York: McGraw-Hill; 2002:173.

Guyton, 2000. Guyton AC: Body temperature, temperature regulation and fever.   In: Guyton AC, Hall JE, ed. Textbook of medical physiology,  10th ed.. Philadelphia: WB Saunders; 2000:822.

Ham et al., 1978. Ham J, Miller RD, Benet LZ, et al: Pharmacokinetics and pharmacodynamics of d-tubocurarine during hypothermia in the cat.  Anesthesiology  1978; 49:324.

Hardy, 1961. Hardy JD: Physiology of temperature regulation.  Physiol Rev  1961; 41:521.

Hardy et al., 1941. Hardy JD, Milhorat A, DuBois A: Basal metabolism and heat loss of young women at temperatures from 22 degrees C to 35 degrees C.  J Nutr  1941; 21:38.

Hellon, 1981. Hellon RF: Neurophysiology of temperature regulation: problems and perspectives.  Fed Proc  1981; 40:2804.

Hemingway, 1963. Hemingway A: Shivering.  Physiol Rev  1963; 43:397.

Hemingway and Price, 1968. Hemingway A, Price WM: The autonomic nervous system and regulation of body temperature.  Anesthesiology  1968; 29:693.

Hey, 1973. Hey EN: Physiological principles involved in the care of the pre-term human infant.   In: Austin CR, ed. The mammalian fetus in vitro,  London: Chapman and Hall; 1973:251.

Hey and Katz, 1969. Hey EN, Katz G: Temporary loss of a metabolic response to cold stress in infants of low birthweight.  Arch Dis Child  1969; 44:323.

Hillman et al., 1982. Hillman PE, Scott NR, van Tienhoven A: Vasomotion in chicken foot: dual innervation of arteriovenous anastomoses.  Am J Physiol  1982; 242:R582.

Himms-Hagen, 1976. Himms-Hagen J: Cellular thermogenesis.  Annu Rev Physiol  1976; 38:315.

Hori and Katafuchi, 1998. Hori T, Katafuchi T: Cell biology and the functions of thermosensitive neurons in the brain.  Prog Brain Res  1998; 115:9.

Horn et al., 1999. Horn EP, Schroeder F, Wilhelm S, et al: Postoperative pain facilitates nonthermoregulatory tremor.  Anesthesiology  1999; 91:979.

Horn et al., 1998. Horn EP, Sessler DI, Standl T, et al: Non-thermoregulatory shivering in patients recovering from isoflurane or desflurane anesthesia.  Anesthesiology  1998; 89:878.

Horrow and Rosenberg, 1988. Horrow JC, Rosenberg H: Does urinary catheter temperature reflect core temperature during cardiac surgery?.  Anesthesiology  1988; 69:986.

Horvath et al., 1956. Horvath S, Spurr G, Hutt B, et al: Metabolic cost of shivering.  J Appl Physiol  1956; 8:595.

Hull and Smales, 1978. Hull D, Smales O: Heat production in the newborn.   In: Sinclair J, ed. Temperature regulation and energy metabolism in the newborn,  New York: Grune & Stratton; 1978:129.

Hunter et al., 1981. Hunter WS, Holmes KR, Elizondo RS: Thermal balance in ketamine-anesthetized rhesus monkey Macaca mulatta.  Am J Physiol  1981; 241:R301.

Hynson et al., 1991. Hynson JM, Sessler DI, Glosten B, et al: Thermal balance and tremor patterns during epidural anesthesia.  Anesthesiology  1991; 74:680.

Ikeda et al., 1998. Ikeda T, Kim JS, Sessler DI, et al: Isoflurane alters shivering patterns and reduces maximum shivering intensity.  Anesthesiology  1998; 88:866.

Ikeda et al., 1999. Ikeda T, Ozaki M, Sessler DI, et al: Intraoperative phenylephrine infusion decreases the magnitude of redistribution hypothermia.  Anesth Analg  1999; 89:462.

Ikeda et al., 1999. Ikeda T, Sessler DI, Kikura M, et al: Less core hypothermia when anesthesia is induced with inhaled sevoflurane than with intravenous propofol.  Anesth Analg  1999; 88:921.

Ikeda et al., 1998. Ikeda T, Sessler DI, Tayefeh F, et al: Meperidine and alfentanil do not reduce the gain or maximum intensity of shivering.  Anesthesiology  1998; 88:858.

Illievich et al., 1994. Illievich UM, Zornow MH, Choi KT, et al: Effects of hypothermia or anesthetics on hippocampal glutamate and glycine concentrations after repeated transient global cerebral ischemia.  Anesthesiology  1994; 80:177.

Israel and Pozos, 1989. Israel DJ, Pozos RS: Synchronized slow-amplitude modulations in the electromyograms of shivering muscles.  J Appl Physiol  1989; 66:2358.

Jessen and Feistkorn, 1984. Jessen C, Feistkorn G: Some characteristics of core temperature signals in the conscious goat.  Am J Physiol  1984; 247:R456.

Jessen and Mayer, 1971. Jessen C, Mayer ET: Spinal cord and hypothalamus as core sensors of temperature in the conscious dog. I. Equivalence of responses.  Pflugers Arch  1971; 324:189.

Jessen, 1980a. Jessen K: An assessment of human regulatory nonshivering thermogenesis.  Acta Anaesthesiol Scand  1980; 24:138.

Jessen, 1980b. Jessen K: The cortisol fluctuations in plasma in relation to human regulatory nonshivering thermogenesis.  Acta Anaesthesiol Scand  1980; 24:151.

Jessen, 1980c. Jessen K: The relation between thyroid function and human regulatory nonshivering thermogenesis.  Acta Anaesthesiol Scand  1980; 24:144.

Jones et al., 1988. Jones B, Ozaki G, Benumof J, et al: Airway resistance caused by a pediatric heat and moisture exchanger.  Anesthesiology  1988; 69:A786.

Jonsson et al., 1991. Jonsson K, Jensen JA, Goodson 3rd WH, et al: Tissue oxygenation, anemia, and perfusion in relation to wound healing in surgical patients.  Ann Surg  1991; 214:605.

Joris et al., 1994. Joris J, Ozaki M, Sessler DI, et al: Epidural anesthesia impairs both central and peripheral thermoregulatory control during general anesthesia.  Anesthesiology  1994; 80:268.

Just et al., 1992. Just B, Delva E, Camus Y, et al: Oxygen uptake during recovery following naloxone. Relationship with intraoperative heat loss.  Anesthesiology  1992; 76:60.

Kadar et al., 1982. Kadar D, Tang BK, Conn AW: The fate of phenobarbitone in children in hypothermia and at normal body temperature.  Can Anaesth Soc J  1982; 29:16.

Karayan et al., 1996. Karayan J, Thomas D, Lacoste L, et al: Delayed forced air warming prevents hypothermia during abdominal aortic surgery.  Br J Anaesth  1996; 76:459.

Karlberg et al., 1962. Karlberg P, Moore R, Oliver TJ: The thermogenic response of the newborn infant to noradrenaline.  Acta Paediatr  1962; 51:284.

Karlberg et al., 1965. Karlberg P, Moore R, Oliver TJ: Thermogenic and cardiovascular response of the newborn baby to noradrenaline.  Acta Paediatr  1965; 54:225.

Kasai et al., 2003. Kasai T, Nakajima Y, Matsukawa T, et al: Effect of preoperative amino acid infusion on thermoregulatory response during spinal anaesthesia.  Br J Anaesth  2003; 90:58.

Kim et al., 1998. Kim JS, Ikeda T, Sessler DI, et al: Epidural anesthesia reduces the gain and maximum intensity of shivering.  Anesthesiology  1998; 88:851.

Kitamura et al., 2000. Kitamura A, Hoshino T, Kon T, et al: Patients with diabetic neuropathy are at risk of a greater intraoperative reduction in core temperature.  Anesthesiology  2000; 92:1311.

Klaus and Faranoff, 1986. Klaus M, Faranoff A: Care of the high-risk neonate,  Philadelphia, WB Saunders, 1986.

Koren et al., 1987. Koren G, Barker C, Goresky G, et al: The influence of hypothermia on the disposition of fentanyl—human and animal studies.  Eur J Clin Pharmacol  1987; 32:373.

Kurz et al., 1995. Kurz A, Go JC, Sessler DI, et al: Alfentanil slightly increases the sweating threshold and markedly reduces the vasoconstriction and shivering thresholds.  Anesthesiology  1995; 83:293.

Kurz et al., 1993. Kurz A, Kurz M, Poeschl G, et al: Forced-air warming maintains intraoperative normothermia better than circulating-water mattresses.  Anesth Analg  1993; 77:89.

Kurz et al., 1995. Kurz A, Sessler DI, Annadata R, et al: Midazolam minimally impairs thermoregulatory control.  Anesth Analg  1995; 81:393.

Kurz et al., 1996. Kurz A, Sessler DI, Lenhardt R: Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. Study of Wound Infection and Temperature Group.  N Engl J Med  1996; 334:1209.

Kurz et al., 1995. Kurz A, Sessler DI, Narzt E, et al: Morphometric influences on intraoperative core temperature changes.  Anesth Analg  1995; 80:562.

Kurz et al., 1995. Kurz A, Xiong J, Sessler DI, et al: Desflurane reduces the gain of thermoregulatory arteriovenous shunt vasoconstriction in humans.  Anesthesiology  1995; 83:1212.

Lacoumenta and Hall, 1984. Lacoumenta S, Hall GM: Liquid crystal thermometry during anaesthesia.  Anaesthesia  1984; 39:54.

Lean et al., 1986. Lean ME, James WP, Jennings G, et al: Brown adipose tissue in patients with phaeochromocytoma.  Int J Obes  1986; 10:219.

Leijh et al., 1979. Leijh PC, van den Barselaar MT, van Zwet TL, et al: Kinetics of phagocytosis of Staphylococcus aureus and Escherichia coli by human granulocytes.  Immunology  1979; 37:453.

Lenhardt et al., 1999. Lenhardt R, Greif R, Sessler DI, et al: Relative contribution of skin and core temperatures to vasoconstriction and shivering thresholds during isoflurane anesthesia.  Anesthesiology  1999; 91:422.

Lenhardt et al., 1997. Lenhardt R, Marker E, Goll V, et al: Mild intraoperative hypothermia prolongs postanesthetic recovery.  Anesthesiology  1997; 87:1318.

Leon et al., 1990. Leon JE, Bissonnette B, Lerman J: Liquid crystalline temperature monitoring: does it estimate core temperature in anaesthetized paediatric patients?.  Can J Anaesth  1990; 37:S98.

Leslie and Sessler, 1996. Leslie K, Sessler DI: Reduction in the shivering threshold is proportional to spinal block height.  Anesthesiology  1996; 84:1327.

Leslie et al., 1994. Leslie K, Sessler DI, Bjorksten AR, et al: Propofol causes a dose-dependent decrease in the thermoregulatory threshold for vasoconstriction but has little effect on sweating.  Anesthesiology  1994; 81:353.

Liu et al., 2001. Liu M, Hu X, Liu J: The effect of hypothermia on isoflurane MAC in children.  Anesthesiology  2001; 94:429.

Lopez et al., 1993. Lopez M, Ozaki M, Sessler DI, et al: Physiologic responses to hyperthermia during epidural anesthesia and combined epidural/enflurane anesthesia in women.  Anesthesiology  1993; 78:1046.

Lyons et al., 1996. Lyons B, Taylor A, Power C, et al: Postanaesthetic shivering in children.  Anaesthesia  1996; 51:442.

MacKenzie and Asbury, 1994. MacKenzie R, Asbury AJ: Clinical evaluation of liquid crystal skin thermometers.  Br J Anaesth  1994; 72:246.

Mahajan et al., 1987. Mahajan RP, Grover VK, Sharma SL, et al: Intraocular pressure changes during muscular hyperactivity after general anesthesia.  Anesthesiology  1987; 66:419.

Matsukawa et al., 1995. Matsukawa T, Kurz A, Sessler DI, et al: Propofol linearly reduces the vasoconstriction and shivering thresholds.  Anesthesiology  1995; 82:1169.

Matsukawa et al., 1995. Matsukawa T, Sessler DI, Christensen R, et al: Heat flow and distribution during epidural anesthesia.  Anesthesiology  1995; 83:961.

Matsukawa et al., 1995. Matsukawa T, Sessler DI, Sessler AM, et al: Heat flow and distribution during induction of general anesthesia.  Anesthesiology  1995; 82:662.

McAllister and Tan, 1980. McAllister Jr RG, Tan TG: Effect of hypothermia on drug metabolism. In vitro studies with propranolol and verapamil.  Pharmacology  1980; 20:95.

Mercer and Jessen, 1978. Mercer JB, Jessen C: Central thermosensitivity in conscious goats: hypothalamus and spinal cord versus residual inner body.  Pflugers Arch  1978; 374:179.

Mercke, 1975. Mercke U: The influence of varying air humidity on mucociliary activity.  Acta Otolaryngol  1975; 79:133.

Mestyan et al., 1964. Mestyan J, Jarai I, Bata G, et al: The significance of facial skin temperature in the chemical heat regulation of premature infants.  Biol Neonat  1964; 7:243.

Michelson et al., 1994. Michelson AD, MacGregor H, Barnard MR, et al: Reversible inhibition of human platelet activation by hypothermia in vivo and in vitro.  Thromb Haemost  1994; 71:633.

Miller et al., 1978. Miller RD, Agoston S, van der Pol F, et al: Hypothermia and the pharmacokinetics and pharmacodynamics of pancuronium in the cat.  J Pharmacol Exp Ther  1978; 207:532.

Morris, 1971a. Morris RH: Influence of ambient temperature on patient temperature during intraabdominal surgery.  Ann Surg  1971; 173:230.

Morris, 1971b. Morris RH: Operating room temperature and the anesthetized, paralyzed patient.  Arch Surg  1971; 102:95.

Morris and Wilkey, 1970. Morris RH, Wilkey BR: The effects of ambient temperature on patient temperature during surgery not involving body cavities.  Anesthesiology  1970; 32:102.

Muma et al., 1991. Muma BK, Treloar DJ, Wurmlinger K, et al: Comparison of rectal, axillary, and tympanic membrane temperatures in infants and young children.  Ann Emerg Med  1991; 20:41.

Nebbia et al., 1996. Nebbia SP, Bissonnette B, Sessler DI: Enflurane decreases the threshold for vasoconstriction more than isoflurane or halothane.  Anesth Analg  1996; 83:595.

Nesher et al., 2001. Nesher N, Wolf T, Uretzky G, et al: A novel thermoregulatory system maintains perioperative normothermia in children undergoing elective surgery.  Paediatr Anaesth  2001; 11:555.

Newton, 1975. Newton DE: Proceedings: The effect of anaesthetic gas humidification on body temperature.  Br J Anaesth  1975; 47:1026.

Ohlson et al., 1997. Ohlson KB, Lindahl SG, Cannon B, et al: Analysis of the cellular mechanism for halothane inhibition of brown adipose tissue thermogenesis.  Ann N Y Acad Sci  1997; 813:718.

Ohlson et al., 1994. Ohlson KB, Mohell N, Cannon B, et al: Thermogenesis in brown adipocytes is inhibited by volatile anesthetic agents. A factor contributing to hypothermia in infants?.  Anesthesiology  1994; 81:176.

Oya et al., 1997. Oya A, Asakura H, Koshino T, et al: Thermographic demonstration of nonshivering thermogenesis in human newborns after birth: its relation to umbilical gases.  J Perinat Med  1997; 25:447.

Ozaki et al., 1994. Ozaki M, Kurz A, Sessler DI, et al: Thermoregulatory thresholds during epidural and spinal anesthesia.  Anesthesiology  1994; 81:282.

Ozaki et al., 1995. Ozaki M, Sessler DI, Suzuki H, et al: Nitrous oxide decreases the threshold for vasoconstriction less than sevoflurane or isoflurane.  Anesth Analg  1995; 80:1212.

Ozaki et al., 1997. Ozaki M, Sessler DI, Suzuki H, et al: The threshold for thermoregulatory vasoconstriction during nitrous oxide/sevoflurane anesthesia is lower in elderly than in young patients.  Ann N Y Acad Sci  1997; 813:789.

Petrikovsky et al., 1997. Petrikovsky B, Silverstein M, Schneider EP: Neonatal shivering and hypothermia after intrapartum amnioinfusion.  Lancet  1997; 350:1366.

Pflug et al., 1978. Pflug AE, Aasheim GM, Foster C, et al: Prevention of post-anaesthesia shivering.  Can Anaesth Soc J  1978; 25:43.

Pickering, 1958. Pickering G: Regulation of body temperature in health and disease.  Lancet  1958; 1:59.

Pierau and Wurster, 1981. Pierau FK, Wurster RD: Primary afferent input from cutaneous thermoreceptors.  Fed Proc  1981; 40:2819.

Plattner et al., 1997. Plattner O, Semsroth M, Sessler DI, et al: Lack of nonshivering thermogenesis in infants anesthetized with fentanyl and propofol.  Anesthesiology  1997; 86:772.

Poulos, 1981. Poulos DA: Central processing of cutaneous temperature information.  Fed Proc  1981; 40:2825.

Puhakka et al., 1994. Puhakka K, Anttonen H, Niskanen J, et al: Calculation of mean skin temperature and changes in body heat content during paediatric anaesthesia.  Br J Anaesth  1994; 72:548.

Rajek et al., 2000. Rajek A, Greif R, Sessler DI, et al: Core cooling by central venous infusion of ice-cold (4 degrees C and 20 degrees C) fluid: isolation of core and peripheral thermal compartments.  Anesthesiology  2000; 93:629.

Ramanathan, 1964. Ramanathan NL: A new weighting system for mean surface temperature of the human body.  J Appl Physiol  1964; 19:531.

Roe, 1971. Roe CF: Effect of bowel exposure on body temperature during surgical operations.  Am J Surg  1971; 122:13.

Roe et al., 1966. Roe CF, Santulli TV, Blair CS: Heat loss in infants during general anesthesia and operations.  J Pediatr Surg  1966; 1:266.

Rohrer and Natale, 1992. Rohrer MJ, Natale AM: Effect of hypothermia on the coagulation cascade.  Crit Care Med  1992; 20:1402.

Rosa et al., 1995. Rosa G, Pinto G, Orsi P, et al: Control of post anaesthetic shivering with nefopam hydrochloride in mildly hypothermic patients after neurosurgery.  Acta Anaesthesiol Scand  1995; 39:90.

Rübsamen and Hales, 1984. Rübsamen K, Hales JR: Role of arteriovenous anastomoses in determining heat transfer across the hindleg skin of sheep.   In: Hales JR, ed. Thermal physiology,  New York: Raven Press; 1984:259.

Ryan, 1982. Ryan JF: Altered temperature regulation section two: unintentional hypothermia.   In: Cooperman L, Orkin F, ed. Complications in anesthesiology,  Philadelphia: JB Lippincott; 1982:284.

Saito, 1997. Saito T: A comparison of the body temperature during sevoflurane anesthesia and isoflurane anesthesia.  Ann N Y Acad Sci  1997; 813:786.

Sanders et al., 1996. Sanders BJ, D'Alessio JG, Jernigan JR: Intraoperative hypothermia associated with lower extremity tourniquet deflation.  J Clin Anesth  1996; 8:504.

Schiff et al., 1966. Schiff D, Stern L, Ledue J: Chemical thermogenesis in newborn infants. Catecholamine excretion and the plasma non-esterified fatty acid-response to cold exposure.  Pediatrics  1966; 37:577.

Schmied et al., 1996. Schmied H, Kurz A, Sessler DI, et al: Mild hypothermia increases blood loss and transfusion requirements during total hip arthroplasty.  Lancet  1996; 347:289.

Sellden et al., 1994. Sellden E, Brundin T, Wahren J: Augmented thermic effect of amino acids under general anaesthesia: a mechanism useful for prevention of anaesthesia-induced hypothermia.  Clin Sci (Lond)  1994; 86:611.

Sellden and Lindahl, 1999. Sellden E, Lindahl SG: Amino acid-induced thermogenesis reduces hypothermia during anesthesia and shortens hospital stay.  Anesth Analg  1999; 89:1551.

Sessler, 1991. Sessler DI: Central thermoregulatory inhibition by general anesthesia.  Anesthesiology  1991; 75:557.

Sessler, 1986. Sessler DI: Malignant hyperthermia.  J Pediatr  1986; 109:9.

Sessler, 1993. Sessler DI: Perianesthetic thermoregulation and heat balance in humans.  FASEB J  1993; 7:638.

Sessler, 1991. Sessler DI: Sweating threshold during isoflurane anesthesia in humans.  Anesth Analg  1991; 73:300.

Sessler et al., 1992. Sessler DI, McGuire J, Hynson J, et al: Thermoregulatory vasoconstriction during isoflurane anesthesia minimally decreases cutaneous heat loss.  Anesthesiology  1992; 76:670.

Sessler et al., 1991. Sessler DI, McGuire J, Moayeri A, et al: Isoflurane-induced vasodilation minimally increases cutaneous heat loss.  Anesthesiology  1991; 74:226.

Sessler et al., 1991. Sessler DI, McGuire J, Sessler AM: Perioperative thermal insulation.  Anesthesiology  1991; 74:875.

Sessler and Moayeri, 1990. Sessler DI, Moayeri A: Skin-surface warming: heat flux and central temperature.  Anesthesiology  1990; 73:218.

Sessler et al., 1988a. Sessler DI, Olofsson CI, Rubinstein EH, et al: The thermoregulatory threshold in humans during halothane anesthesia.  Anesthesiology  1988; 68:836.

Sessler et al., 1988b. Sessler DI, Olofsson CI, Rubinstein EH: The thermoregulatory threshold in humans during nitrous oxide-fentanyl anesthesia.  Anesthesiology  1988; 69:357.

Sessler and Ponte, 1982. Sessler DI, Ponte J: Disparity between thermal comfort and psychological thermoregulatory responses during epidural anesthesia.  Anesthesiology  1982; 71:A682.

Sessler et al., 1987. Sessler DI, Rubinstein EH, Eger 2nd EI: Core temperature changes during N2O fentanyl and halothane/O2 anesthesia.  Anesthesiology  1987; 67:137.

Sessler et al., 1991. Sessler DI, Rubinstein EH, Moayeri A: Physiologic responses to mild perianesthetic hypothermia in humans.  Anesthesiology  1991; 75:594.

Shanks, 1975. Shanks CA: Heat balance during surgery involving body cavities.  Anaesth Intensive Care  1975; 3:114.

Shanks, 1975. Shanks CA: Mean skin temperature during anaesthesia: an assessment of formulae in the supine surgical patient.  Br J Anaesth  1975; 47:871.

Shroff and Skerman, 1988. Shroff PK, Skerman JH: Humidifier malfunction—a cause of anesthesia circuit occlusion.  Anesth Analg  1988; 67:710.

Silverman and Blanc, 1957. Silverman W, Blanc W: Effects of humidity on survival of newly born premature infants.  Pediatrics  1957; 22:876.

Silverman et al., 1958. Silverman W, Fertig J, Bergen A: Influence of thermal environent upon survival of newly born premature infants.  Pediatrics 2  1958; 2:876.

Silverman et al., 1964. Silverman W, Zamelis A, Sinclair J, et al: Warm nape of the newborn.  Pediatrics  1964; 33:984.

Simon, 1974. Simon E: Temperature regulation: the spinal cord as a site of extrahypothalamic thermoregulatory functions.  Rev Physiol Biochem Pharmacol  1974;1.

Sinclair, 1972. Sinclair JC: Thermal control in premature infants.  Annu Rev Med  1972; 23:129.

Sinclair, 1972. Sinclair JG: The effects of meperidine and morphine in rabbits pretreated with phenelzine.  Toxicol Appl Pharmacol  1972; 22:231.

Smith and Allen, 1986. Smith HS, Allen R: Another hazard of heated water humidifiers.  Anaesthesia  1986; 41:215.

Steele et al., 1996. Steele MT, Nelson MJ, Sessler DI, et al: Forced air speeds rewarming in accidental hypothermia.  Ann Emerg Med  1996; 27:479.

Stephen et al., 1960. Stephen CR, Dent S, Hall K: Body temperature regulation during anesthesia in infants and children.  JAMA  1960; 174:1579.

Stern et al., 1965. Stern L, Lees MH, Leduc J: Environmental temperature, oxygen consumption, and catecholamine excretion in newborn infants.  Pediatrics  1965; 36:367.

Stoen and Sessler, 1990. Stoen R, Sessler DI: The thermoregulatory threshold is inversely proportional to isoflurane concentration.  Anesthesiology  1990; 72:822.

Stone et al., 1981. Stone DR, Downs JB, Paul WL, et al: Adult body temperature and heated humidification of anesthetic gases during general anesthesia.  Anesth Analg  1981; 60:736.

Stuart et al., 1966. Stuart D, Ott K, Ishikawa K, et al: The rhythm of shivering. 3. Central contributions.  Am J Phys Med  1966; 45:91.

Sulyok et al., 1973. Sulyok E, Jequier E, Prod'hom LS: Thermal balance of the newborn infant in a heat-gaining environment.  Pediatr Res  1973; 7:888.

Swyer, 1973. Swyer P: Heat loss after birth.   In: Sinclair JC, ed. Temperature regulation and energy metabolism in the newborn,  New York: Grune & Stratton; 1973:91.

Szmuk et al., 2001. Szmuk P, Rabb MF, Baumgartner JE, et al: Body morphology and the speed of cutaneous rewarming.  Anesthesiology  2001; 95:18.

Tankersley et al., 1991. Tankersley CG, Smolander J, Kenney WL, et al: Sweating and skin blood flow during exercise: effects of age and maximal oxygen uptake.  J Appl Physiol  1991; 71:236.

Tollofsrud et al., 1984. Tollofsrud SG, Gundersen Y, Andersen R: Peroperative hypothermia.  Acta Anaesthesiol Scand  1984; 28:511.

Vale and Lunn, 1969. Vale RJ, Lunn HF: Heat balance in anaesthetized surgical patients.  Proc R Soc Med  1969; 62:1017.

van Marken Lichtenbelt and Daanen, 2003. van Marken Lichtenbelt WD, Daanen HA: Cold-induced metabolism.  Curr Opin Clin Nutr Metab Care  2003; 6:469.

van Oss et al., 1980. van Oss CJ, Absolom DR, Moore LL, et al: Effect of temperature on the chemotaxis, phagocytic engulfment, digestion and O2 consumption of human polymorphonuclear leukocytes.  J Reticuloendothel Soc  1980; 27:561.

Vitez et al., 1974. Vitez TS, White PF, Eger 2nd EI: Effects of hypothermia on halothane MAC and isoflurane MAC in the rat.  Anesthesiology  1974; 41:80.

Vybiral et al., 2000. Vybiral S, Lesna I, Jansky L, et al: Thermoregulation in winter swimmers and physiological significance of human catecholamine thermogenesis.  Exp Physiol  2000; 85:321.

Washington et al., 1992. Washington DE, Sessler DI, McGuire J, et al: Painful stimulation minimally increases the thermoregulatory threshold for vasoconstriction during enflurane anesthesia in humans.  Anesthesiology  1992; 77:286.

Wenisch et al., 1996. Wenisch C, Narzt E, Sessler DI, et al: Mild intraoperative hypothermia reduces production of reactive oxygen intermediates by polymorphonuclear leukocytes.  Anesth Analg  1996; 82:810.

Wyss et al., 1975. Wyss CR, Brengelmann GL, Johnson JM, et al: Altered control of skin blood flow at high skin and core temperatures.  J Appl Physiol  1975; 38:839.

Xiong et al., 1996. Xiong J, Kurz A, Sessler DI, et al: Isoflurane produces marked and nonlinear decreases in the vasoconstriction and shivering thresholds.  Anesthesiology  1996; 85:240.