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


PART THREE – Clinical Management of Special Surgical Problems

Chapter 18 – Anesthesia for Pediatric Neurosurgery

Elliot J. Krane,Bridget M. Philip,
Kelly K. Yeh,
Karen B. Domino



Neurophysiology, 651



Cerebral Blood Flow,651



Intracranial Pressure, 652



Effects of Drugs on Neurophysiology, 652



Intravenous Anesthetics,653



Inhaled Anesthetics, 654



Muscle Relaxants, 656



Vasodilators, 656



Principles of Pediatric Neuroanesthesia, 657



Preoperative Evaluation, 657



Premedication, 658



Induction of Anesthesia, 658



Positioning of the Patient, 659



Intraoperative Monitoring, 660



Neurophysiologic Monitoring, 664



Selection of Anesthetic Agents,664



Temperature Regulation, 665



Fluid Management, 666



Anesthetic Management for Neurosurgical Procedures, 668



The Newborn, 668



Infant and Child, 670



Summary, 678

The practice of anesthesia for pediatric neurosurgery encompasses the skills and knowledge of pediatric anesthesiology as well as the knowledge of cerebral pathophysiology. To provide optimal neuroanesthetic management and neurologic intensive care, the anesthesiologist must understand the physiology of the central nervous system, the effects of drugs on cerebral hemodynamics, and the physiology of the developing child.



Specific data on cerebral blood flow (CBF) in human infants and children are sparse, and many anesthetic principles must be inferred from the data in animals and adult humans. Total CBF in the adult is about 50 mL/100 g of brain tissue per min ( Lassen and Christensen, 1976 ). The amount of flow varies in different anatomic areas. CBF is higher in gray matter (80 mL/100 g brain per min) than in white matter (20 mL/100 g brain per min). Lower CBF probably occurs in the developing brain of premature and newborn infants (30 to 40 mL/100 g per min), but in infants and older children, global CBF is higher than in adults (65 to 85 mL/100 g per min) ( Settergren et al., 1980 ; Kreisman et al., 1989 ; Chiron et al., 1992 ).

Regional cerebral metabolic rate for oxygen (CMRo2) is an important determinant of regional CBF because supply is closely linked to demand in both adults and children. CBF therefore increases with increased CMRo2, such as that associated with seizures or fever. CBF also decreases with reduced CMRo2, such as that caused by hypothermia or barbiturates. CMRo2 is higher in children (5 mL/100 g per min) than in adults (3 to 4 mL/100 g per min) ( Kennedy and Sokoloff, 1967 ).

Besides metabolic demand, CBF depends on cerebral perfusion pressure (CPP) and on arterial oxygen (Pao2) and carbon dioxide (Paco2) tensions. CPP is equal to the mean arterial pressure (MAP) minus the cerebral venous pressure, which is well approximated by intracranial pressure (ICP) when the cranium is intact. The CBF is autoregulated to changes in MAP; that is, it remains constant when MAP is between 60 and 150 mm Hg in adults ( Lassen and Christensen, 1976 ). Within this range, the cerebral blood vessels dilate at lower blood pressures and constrict at higher blood pressures to maintain a constant CBF. However, the autoregulatory response may take up to 2 minutes to occur. Because the MAP in infants is less than 60 mm Hg in the first year of life, the autoregulatory limits are probably lower. In neonatal lambs and dogs, the lower limit of autoregulation is 40 mm Hg ( Purves and James, 1969 ; Hernandez et al., 1980 ; Rogers et al., 1980 ). One author extrapolated from animal data to describe the autoregulatory range as between 20 and 60 mm Hg in high-risk neonates ( Pryds, 1991 ). In a study of 17 extremely low-birth-weight infants, Munro and others (2004) noted that cerebral autoregulation appears to be functional in normotensive but not hypotensive infants and that the lower limit of autoregulation appears to be a mean blood pressure of 30 mm Hg.

The true autoregulatory range in human premature infants, full-term newborns, infants, and children is undefined at the present, and the practitioner must do his or her best to extrapolate from newborn animal studies to humans and to interpolate between these and what is known to be the autoregulatory range in adults. Above and below the limits of autoregulation, CBF is passively dependent on CPP; inadequate CBF produces cerebral ischemia, and excessive CBF leads to edema formation and possibly intracranial hemorrhage.

The lower limit of autoregulation is lower with drug-induced hypotension than during hypovolemic hypotension ( Fitch et al., 1973 ). Signs of cerebral ischemia generally do not occur when MAP is reduced to less than 60 mm Hg in adults during deliberate hypotension and general anesthesia. When blood pressure is increased above the autoregulatory range, large increases in CBF, disruption of the blood-brain barrier, and edema formation occur ( Hatashita et al., 1986 ). Chronic hypertension raises the upper limit of autoregulation, although long-term antihypertensive therapy may restore the limit to normal ( Hoffman et al., 1983 ). Autoregulation is attenuated by hypercapnia, hypoxia, high concentrations of volatile anesthetics, nitroprusside, and trauma. In local areas around brain tumors and in areas of focal cerebral ischemia, autoregulation is lost and perfusion is pressure dependent ( Dong et al., 1996 ; Schmieder et al., 2000 ). Autoregulation is also impaired in premature infants with respiratory distress ( Lou et al., 1979 ; Milligan, 1980 ; Daven et al., 1983 ; Ong et al., 1986 ).

In the adult, normal cerebral vessels dilate in response to increases in Paco2 and constrict in response to decreases in Paco2 ( Lassen and Christensen, 1976 ). CBF varies linearly with Paco2 between 20 and 80 mm Hg, so that a 4% change in CBF occurs for each 1-mm Hg change in Paco2 within this range. At Paco2 greater than 80 mm Hg the cerebral vasculature is maximally dilated, and sensitivity to further increases in Paco2 decreases. As Paco2 decreases below 20 mm Hg, CBF does not decrease further, presumably because ischemia-induced metabolic changes override the response to CO2. The mechanism of the response of the cerebral circulation to CO2 involves changes in cerebrospinal fluid (CSF) and in periarteriolar pH. CO2 crosses the blood-brain barrier more freely than do bicarbonate ions. Decreases in Paco2 acutely increase periarteriolar pH, which gradually normalizes owing to the subsequent movement of bicarbonate out of the CSF over the next 8 to 24 hours ( Lassen and Christensen, 1976 ). CBF also returns to normal within 24 hours. With chronic hyperventilation, as occurs in patients in the intensive care unit, CSF pH is near normal despite low Paco2. If the Paco2 acutely rises to normal levels, the periarteriolar pH decreases, and CBF and ICP increase.

The cerebrovascular response to changes in Paco2 is not completely developed at birth ( Rogers et al., 1980 ; Shapiro et al., 1980 ; Hansen et al., 1984 ). Hypercapnia increases CBF in newborn animals. The newborn brain is relatively insensitive to moderate degrees of hypocapnia. Brain blood flow does not decrease significantly from normocapnia until extreme degrees of hypocapnia (Paco2 <15 mm Hg) occur ( Hansen et al., 1984 ).

The cerebral vasculature is less sensitive to changes in Pao2. In adults, CBF does not increase until Pao2 falls below 50 mm Hg ( Lassen and Christensen, 1976 ). CBF then increases exponentially, with a fivefold increase occurring when Pao2 is 25 mm Hg. Hyperoxia reduces adult CBF by approximately 10%. The fetal and neonatal circulation responds to small changes in Pao2 ( Rogers et al., 1980 ), perhaps because of the greater oxygen affinity of fetal hemoglobin. The age at which this heightened responsiveness decreases is unknown.

Other factors that affect CBF include hematocrit, body temperature, and autonomic tone. In the adult, increasing the hematocrit to greater than 50% reduces CBF by increasing blood viscosity and reducing the hematocrit to less than 30% increases CBF by decreasing viscosity ( Wood et al., 1983 ). CMRo2 and CBF are reduced by decreasing body temperature. Immaturity of the sympathetic nervous system in neonates may contribute to unusual CBF responses to the release of catecholamines ( Rogers et al., 1980 ).

In summary, little is known about the physiology of the cerebral circulation in the newborn and infant human, although some studies have shown that in healthy full-term neonates, dynamic regulation of the cerebral circulation may occur on the first day of life ( Mochalova et al., 1983 ). Data must be extrapolated from current knowledge of the physiology of human adults and animals. While normal newborn infants probably autoregulate CBF in response to changes in MAP, the autoregulatory limits are undefined; their cerebral circulation is less responsive to hypocapnia than is that of adults. In addition, vasomotor paralysis, in which the cerebral vessels normally do not autoregulate or respond to changes in Paco2, may occur in newborns with respiratory distress syndrome and in children with brain tumors, trauma, or brain ischemia.


In the adult the cranium is rigid. Brain tissue is noncompressible, and its volume is relatively constant. Increases in the volume of brain, CSF, or blood compartments must be compensated for by decreases in another compartment; this movement of blood or CSF from the cranial compartment occurs at the expense of an increase in ICP, creating a pressure gradient favoring such movement. Subsequently, further compensatory translocation of blood or fluid requires still greater gradients—that is, compliance diminishes. The compensatory mechanisms may fail when acute increases in intracranial volume (caused by hematoma or trauma, for example) occur rapidly. As a space-occupying lesion expands and the compensatory mechanisms fail, small increases in intracranial volume cause large increases in ICP. This phenomenon generates a typical pressure-volume, or compliance, curve ( Fig. 18-1A ).


FIGURE 18-1  A, Ideal intracranial compliance curve. Note that if intra-cranial pressure (ICP) lies within the shaded area, ICP is normal, but intracranial compliance is limited. Relatively small changes in intracranial volume lead to relatively large increases in ICP. B, The intracranial compliance curve changes after a reduction in cerebral tissue volume, such as might occur after a dose of mannitol or after treatment of cerebral edema around a tumor with dexamethasone. Intracranial compliance increases in these circumstances.



In the infant, cranial decompression can also occur through expansion of the skull size. The anterior fontanel remains open until around 1 year of age, and the cranial sutures of a child's skull do not fuse until as late as the tenth year. Slow increases in intracranial volume can be offset by a slow increase in skull size before ICP increases. In contrast, skull diameter cannot increase with rapid changes in intracranial volume, such as after head injury. In these cases, ICP changes occur as in the adult.

The absence of intracranial hypertension says nothing about intracranial compliance, which may be severely impaired even if ICP is normal ( Bruce et al., 1977 ; Wilkinson, 1981 ). Figure 18-1A illustrates this concept. During periods of diminished compliance but normal ICP (shaded area), small perturbations in intracranial volume are poorly tolerated and produce marked elevation of ICP and clinical deterioration. When intracranial compliance is known to be limited, patients should be treated with the appropriate techniques and safeguards as if intracranial hypertension were indeed present. Measures that reduce cerebral volume by dehydrating cerebral tissue (administration of mannitol or furosemide) or by reducing cerebral edema (as does dexamethasone) should also improve cerebral compliance (Fig. 18-1B ).

Because a patient's intracranial compliance is seldom known, any patient with a space-occupying intracranial mass is assumed to have reduced intracranial compliance (meaning that small changes in intracranial volume result in large changes in pressure). The goal of neuroanesthetic management is to control CBF and cerebral blood volume (CBV) and, therefore, ICP.

Copyright © 2008 Elsevier Inc. All rights reserved. -

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

Copyright © 2005 Mosby, An Imprint of Elsevier


The effects of the commonly used anesthetic agents on CBF, CMRo2, and CSF dynamics are discussed in this section and are summarized in Table 18-1 . There are no data on these effects in infants and children. The responses in children are assumed to be the same as those in adults.

TABLE 18-1   -- Effects of anesthetic agents on cerebral metabolic rate for oxygen (CMRO2), intracranial pressure (ICP), and cerebrospinal fluid (CSF) dynamics

Anesthetic Agent






↓ ↓

↓ ↓


↓ ↓


↓ ↓

↓ ↓


↓ ↓





Lidocaine (nontoxic)

↓ ↓



↓ ↓






Nitrous oxide



↓ ↓








↓ ↓




↓ ↓




Increase in ICP dose is related and blunted by hyperventilation.



Intravenous anesthetics generally reduce or do not alter CBF and ICP, with the exception of ketamine, which increases both CBF and ICP.


Propofol is an intravenous induction agent with cerebrovascular properties similar to thiopental: both may depress systemic blood pressure but both potently decrease CMRo2, CBF, and ICP ( Van Hemelrijck et al., 1990 ; Warner et al., 1990 ; Fischer et al., 1992 ). Cerebral autoregulation and cerebral responsiveness to changes in arterial CO2 tension ( Fox et al., 1992 ; Petersen et al., 2003 ; Karsli et al., 2004 ) are well preserved during propofol anesthesia. Propofol is a good alternative to thiopental for induction of anesthesia for neurosurgery. In the presence of nitrous oxide and hypocapnia, propofol results in lower ICP and higher cerebral perfusion pressure (CPP) and in reduced cerebral cortical edema than isoflurane or sevoflurane. In patients with or at risk for intracranial hypertension and decreased cerebral perfusion, maintenance of anesthesia with propofol is superior to inhaled halogenated anesthetics, at least until the dura has been opened.


The barbiturates decrease CBF, CBV, and CMRo2 in a dose-dependent manner ( Pierce et al., 1962 ; Michenfelder, 1974 ; Albrecht et al., 1977 ) and therefore reduce ICP. Neither CBF nor cerebral metabolism is significantly altered by subanesthetic doses of barbiturates. When the electroencephalogram (EEG) becomes isoelectric, CBF and CMRo2 decrease to about 50% of normal, and additional doses of barbiturates have little further effect. Barbiturates may also be used to prevent increases in ICP that can occur with laryngoscopy and endotracheal intubation.

Autoregulation and the cerebrovascular response to changes in Paco2 remain intact during barbiturate anesthesia. The rate of CSF formation and the resistance to reabsorption of CSF are not altered by barbiturates ( Mann et al., 1979 ). In doses that suppress the EEG, barbiturates reduce cerebral damage in animal and human models of focal cerebral ischemia ( Smith et al., 1974 ; Nussmeir et al., 1986 ;Nehls et al., 1987 ). In animals, barbiturates also reduce the extent of cerebral edema after a cortical freeze injury. This decrease in edema is in contrast to the response observed with the volatile anesthetics (Smith and Macque, 1976 ).


Opioids have little effect on CBF, CBV, or ICP unless respiration is depressed and Paco2 is increased ( Miller et al., 1975 ; Misfeldt et al., 1976 ; Jobes et al., 1977 ; Moss et al., 1978a ).


When combined with nitrous oxide, fentanyl decreases CBF by 47% and CMRo2 by 18% ( Michenfelder, 1974 ). Autoregulation and CO2 responsiveness of the cerebral circulation are not altered. Fentanyl does not alter the rate of CSF formation, but it reduces the resistance to CSF reabsorption by 50% (Artru, 1983a, 1984b [22] [23]), the effect of which is to decrease CSF volume to a degree of unknown clinical significance. The neonatal cerebral circulation is also unaffected by fentanyl ( Yaster et al., 1987 ).

Sufentanil and Alfentanil

Alfentanil in high doses (10 to 20 mcg/kg) reduces both CBF and CMRo2 by 25% to 30% ( Stephan et al., 1991 ; Warner et al., 1991 ), whereas at exceedingly high doses (10 to 200 mcg/kg), sufentanil may transiently increase CBF while reducing CMRo2 ( Milde et al., 1990 ). However, at conventional doses, sufentanil ( Herrick et al., 1991 ; Weinstabl et al., 1991 ) and alfentanil ( Mayberg et al., 1993 ) do not appear to have adverse effects on the cerebral vasculature or upon ICP in most patients. In a subset of patients with severe head injuries and very poor intracranial compliance, sufentanil may cause a small (e.g., <10 mm Hg) and transient increase in ICP that may be clinically significant in some settings ( Sperry et al., 1992 ; Weinstabl et al., 1992 ; Albanese et al., 1993 ).


Remifentanil is an ultra-short-acting opioid that is rapidly metabolized by plasma cholinesterases. The very short clinical duration of effect of remifentanil and its context-sensitive half-life that is independent of the duration of infusion (Minto et al., 1997, 2003 [219] [220]) make it an appealing opioid for lengthy neurosurgical procedures after which rapid return of consciousness is desirable. As is the case with other opioids that have been studied, remifentanil does not increase CBF or ICP ( Hoffman et al., 1993 ; Warner et al., 1996 ; Baker et al., 1997 ; Guy et al., 1997 ; Ostapkovich et al., 1998 ;Paris et al., 1998 ; Rizzi, 1998 ; Wagner et al., 2001 ; Lorenz et al., 2002 ; Gerlach et al., 2003 ; Klimscha et al., 2003 ; Lorenz et al., 2003 ; Engelhard et al., 2004 ; Lagace et al., 2004 ). Remifentanil, like other opioids, preserves cerebral autoregulation and CO2 reactivity ( Baker et al., 1997 ; Ostapkovich et al., 1998 ; Klimscha et al., 2003 ; Engelhard et al., 2004 ).

Return of consciousness is very rapid after remifentanil is discontinued, and the frequency of administration of naloxone to permit neurologic assessment is decreased ( Guy et al., 1997 ). However, because remifentanil analgesia is very brief after its discontinuation, a long-acting opioid analgesic must be administered to prevent severe pain and rebound hypertension before or soon after remifentanil is discontinued ( Domaingue, 2001 ; Gelb et al., 2003 ; Gerlach et al., 2003 ; Cafiero et al., 2004 ).

Other Intravenous Anesthetics

Etomidate reduces ICP by decreasing CBF and CMRo2 by 34% and 45%, respectively. It, too, preserves the CO2 responsiveness of the cerebral circulation ( Renou et al., 1978 ; Moss et al., 1979 ). A side effect of etomidate administration is myoclonus. Myoclonus has been reported after prolonged continuous infusion of etomidate ( Laughlin and Newberg, 1985 ). Lidocaine in clinical doses decreases CBF and reduces the increase in ICP associated with endotracheal intubation ( Sakabe et al., 1974 ; Donegan and Bedford, 1980 ) and suctioning.

Thebenzodiazepines (diazepam, lorazepam, and midazolam) decrease CBF and CMRo2 approximately 25% ( Cotev and Shalit, 1975 ; Rockoff et al., 1980 ; Forster et al., 1982 ; Nugent et al., 1982 ;Nakahashi et al., 1991 ). Clinical doses of midazolam and diazepam do not alter ICP ( Tateishi et al., 1981 ; Giffin et al., 1984 ).


In contrast to the other intravenous anesthetic agents, ketamine is a potent cerebrovasodilator. Ketamine increases CBF by 60% with little change in CMRo2 ( Dawson et al., 1971 ; Takeshita et al., 1972 ;Schwedler et al., 1982 ). The cerebrovascular response to administration of ketamine is thought to be the result of regional cerebral activation induced by the drug ( Hougaard et al., 1974 ). Ketamine produces a marked increase in ICP, which can be reduced, but not prevented, by hyperventilation ( Gardner et al., 1972 ; Sari et al., 1972 ; Shapiro et al., 1972 ). The increase in CBF, and presumably in ICP, can be blocked by previous administration of thiopental ( Dawson et al., 1971 ). Ketamine has been associated with sudden elevation of ICP and clinical deterioration when used in patients with hydrocephalus and other intracranial pathology ( Lockhart and Jenkins, 1972 ; Shapiro et al., 1972 ; Crumrine et al., 1975 ). Ketamine is not used in patients with reduced intracranial compliance.


All currently used inhaled anesthetics, including nitrous oxide, are cerebrovasodilators to various degrees. The vasodilation can be minimized by the use of low concentrations and concomitant hyperventilation.

Nitrous Oxide

Nitrous oxide is a weak cerebrovasodilator whose effects on CBF are offset by hyperventilation and barbiturate anesthesia ( Algotsson et al., 1992 ). The variability of effects that nitrous oxide has on CBF and ICP in different reports results from differences in experimental species and background anesthesia. In many animals, nitrous oxide in subanesthetic doses (60% to 70%) causes excitement and cerebral metabolic stimulation, with an accompanying increase in CBF ( Theye and Michenfelder, 1968 ; Sakabe et al., 1978 ; Pelligrino et al., 1984 ; Todd, 1987 ).

Because nitrous oxide is not an adequate anesthetic in the absence of other inhalation or intravenous anesthetics, the modification of the cerebral effects of nitrous oxide by additional anesthetic drugs is particularly important. Seventy percent nitrous oxide does not, for example, cause a change in CBF, but it does reduce CMRo2 by 15% to 20% during barbiturate and narcotic anesthesia ( Sakabe et al., 1978). However, when nitrous oxide is added to a volatile anesthetic such as isoflurane ( Cucchiara et al., 1974 ; Manohar and Parks, 1984 ) or halothane ( Sakabe et al., 1976 ), both CBF and CMRo2 increase. When nitrous oxide is added to sevoflurane, cerebral hyperemia increases and autoregulation is impaired ( Iacopino et al., 2003 ). The cerebrovascular responses to changes in Paco2 and MAP are preserved during nitrous oxide anesthesia.

ICP may increase in response to nitrous oxide in patients with intracranial mass lesions and reduced intracranial compliance ( Henriksen and Jorgensen, 1973 ; Moss and McDowall, 1979 ; Iacopino et al., 2003 ). The increase in ICP with nitrous oxide, however, is readily reversible by diazepam and barbiturate anesthesia and simultaneously initiated hyperventilation ( Phirman and Shapiro, 1977 ).

The use of nitrous oxide for pediatric neuroanesthesia remains controversial. Some anesthesiologists prefer to avoid nitrous oxide because of its ability to increase CMRo2 and reduce the cerebral protective effects of barbiturates. Others are concerned because nitrous oxide readily diffuses into collections of intracranial air and may increase ICP in the presence of pneumocephalus ( Saidman and Eger, 1965 ;Artru, 1982 ; Skahen et al., 1986 ). Asymptomatic accumulation of intracranial air occurs commonly during craniotomies, especially those associated with posterior fossa surgery and drainage of CSF (Toung et al., 1986 ).

Some anesthesiologists discontinue nitrous oxide before closure of the dura to reduce the incidence of tension pneumocephalus. Others administer nitrous oxide throughout the procedure without any obvious detrimental effects, and indeed one randomized control trial comparing anesthetic techniques with and without nitrous oxide in patients undergoing sitting craniotomies showed no difference in the incidence of size of pneumocephalus between the three groups ( Hernandez-Palazon et al., 2003 ). It may be that nitrous oxide equilibrates with intracranial air before the dura is closed. If so, ICP would not increase during craniodural closure because air pockets would already contain nitrous oxide. In addition, the discontinuance of nitrous oxide would decrease ICP, as nitrous oxide diffused back into the bloodstream ( Skahen et al., 1986 ). Maintenance with nitrous oxide until the end of the surgery may be advantageous because it permits rapid awakening and may reduce the intracranial gas volume and the likelihood of delayed tension pneumocephalus.

Nitrous oxide is generally not contraindicated during sitting craniotomies despite the fact that the volume of a venous air embolus (VAE) expands in the presence of nitrous oxide. In fact, this phenomenon actually increases the sensitivity of monitoring for VAE by capnography ( Losasso et al., 1992a ), while at the same time nitrous oxide neither increases the risk of VAE ( Losasso et al., 1992b ) nor increases the hemodynamic consequences of VAE provided that nitrous oxide is discontinued when VAE is first detected ( Losasso et al., 1992b ).


Isoflurane is the most popular of the volatile anesthetics for neuroanesthesia. Its popularity is based on the fact that it affects CBF less than does halothane at equivalent minimum alveolar concentration (MAC) doses ( Todd and Drummond, 1984 ; Drummond and Todd, 1985 ; Algottson et al., 1988 ), the fact that 1 MAC isoflurane preserves cerebral autoregulation ( McPherson and Traystman, 1988 ) and CO2 responsiveness ( McPherson et al., 1989 ), and the belief that it may provide cerebral protection ( Newberg and Michenfelder, 1983 ; Newberg et al., 1983 ; Verhaegen et al., 1992 ). Cerebral autoregulation is less affected by isoflurane than by halothane ( Todd and Drummond, 1984 ). In addition, isoflurane does not change CSF production, and it reduces the resistance to reabsorption of CSF (Artru, 1984a, 1984b [20] [21]). During hypocapnia, CBF is lower with 1.0 MAC isoflurane (with 75% nitrous oxide) than with nitrous oxide alone ( Cucchiara et al., 1974 ; Drummond and Todd, 1985 ;Scheller et al., 1986 ). In contrast, 1.0 MAC halothane (with 75% nitrous oxide) increases CBF.

Despite their dissimilar effects on CBF, isoflurane and halothane increase ICP equally in an animal model of brain injury ( Scheller et al., 1987 ). This is probably because isoflurane and halothane increase CBV to a similar degree ( Artru, 1984d ; Archer et al., 1987 ). In patients with reduced intracranial compliance, isoflurane increases ICP. The increase in ICP in these patients can be attenuated by simultaneous initiation of hyperventilation ( Adams et al., 1981 ). Isoflurane may be safely used in patients with small supratentorial brain tumors ( Madsen et al., 1987 ) but may cause dangerous increases in ICP in patients with large intracranial mass lesions that are associated with a midline shift evident on the computed tomography (CT) scan ( Grosslight et al., 1985 ). As with halothane, isoflurane should be avoided in patients with reduced intracranial compliance until the dura is open, if ICP is not being monitored.

Isoflurane decreases CMRo2 by 30%, and it causes an isoelectric EEG at concentrations above 2.0 MAC ( Newberg et al., 1983 ). It is unique among the volatile agents in that it preserves normal cerebral energy states and aerobic metabolism at very low blood pressure (40 mm Hg), in contrast to the findings observed with hypotension induced by halothane, trimethaphan, or sodium nitroprusside ( Newberg et al., 1984 ). In studies of mice exposed to 5% oxygen, isoflurane increased survival time and thus may have provided some degree of cerebral protection. In studies of incomplete global ischemia in isoflurane-anesthetized dogs, cerebral energy stores were increased, presumably through depression of cortical electrical activity and cerebral metabolism ( Newberg and Michenfelder, 1983 ). Protective effects of isoflurane, however, were not observed in a primate model of regional cerebral ischemia ( Nehls et al., 1987 ).


Sevoflurane is a fluorinated ether with a low blood-gas solubility. Studies in rabbits suggest that sevoflurane does not increase CBF at 0.5 to 1.0 MAC. Sevoflurane does cause increases in cerebral blood flow and ICP and a decrease in cerebral oxygen consumption ( Scheller et al., 1988 ) and CPP similar to isoflurane ( Petersen et al., 2003 ). Compared with isoflurane, sevoflurane allows more rapid emergence after lengthy neurosurgery, allowing more rapid neurologic assessment ( Gauthier et al., 2002 ).

The increase in CBF with sevoflurane in normocapnic children is less than that with halothane ( Monkhoff et al., 2001 ). Taken together, the available evidence suggests that sevoflurane is a more appropriate inhaled anesthetic than halothane during craniotomy and is equivalent in its cerebrovascular effects to isoflurane while allowing more rapid recovery after long anesthesia. As is the case with other halogenated agents, if ICP or intracranial compliance is compromised, sevoflurane should be withheld until the dura has been opened ( Petersen et al., 2003 ).


Halothane is a cerebral vasodilator that decreases cerebrovascular resistance (CVR) and increases CBF in a dose-dependent fashion ( Wollman et al., 1964 ; Albrecht et al., 1977 ; Todd and Drummond, 1984 ; Brussel et al., 1991 ). The increase in CBF is transient; CBF decreases to baseline levels after 150 minutes of halothane anesthesia ( Albrecht et al., 1983 ). CBV, however, remains elevated by 11% to 12% over a 3-hour period of halothane administration ( Artru, 1983b ). Halothane reduces CMRo2 by 17% to 33% ( Albrecht et al., 1977 ). The cerebral vasculature remains responsive to changes in arterial Paco2 ( Alexander et al., 1964 ; Wollman et al., 1964 ; Drummond and Todd, 1985 ). Halothane in high concentrations (2.0 MAC) abolishes autoregulation of the cerebral circulation in response to changes in MAP in both adults ( Miletich et al., 1976 ; Todd and Drummond, 1984 ) and infants ( Messer et al., 1989 ). Halothane alters blood-brain barrier permeability, promoting the extravasation of plasma proteins into normal brain during periods of acute hypertension ( Forster et al., 1978 ). Halothane reduces CSF formation by 30% in dogs and increases the resistance of reabsorption of CSF (Artru, 1983a, 1984b [22] [23]).

Because ICP is determined by CBV, CSF volume, and brain tissue volume, it is not surprising that ICP increases with halothane ( Jennet et al., 1969 ; DiGiovanni et al., 1974 ). Peak increases are observed in 3 to 13 minutes, although the increase persists over 3 hours of halothane exposure ( Artru, 1983b ). The increase in ICP in patients with intracranial mass lesions can be attenuated, but not totally prevented, by establishing hyperventilation for 10 minutes before the introduction of halothane ( Adams et al., 1972 ). If ICP is not being monitored, halothane should not be used in patients with reduced intracranial compliance until the dura is open and its effects on the brain can be seen.


Desflurane is an inhalation agent that is chemically similar to isoflurane. Its physicochemical properties are remarkable for a blood-gas partition coefficient even lower than that of nitrous oxide, permitting rapid uptake and washout of the gas. Desflurane effects on cerebral metabolism and hemodynamics are not as well studied as the effects of the other inhalation agents, but the existing animal studies suggest that its effects are not unique in any way. Desflurane at clinical concentrations is a potent cerebral vasodilator, increasing ICP ( Artru, 1994 ; Artru et al., 1994 ), increasing CBF by 50% at 1.5 MAC, and reducing autoregulation of CBF ( Lutz et al., 1990 ). However, cerebrovascular responsiveness to hypocapnia is preserved during desflurane anesthesia in laboratory animals, protecting the animal from increases in ICP if hyperventilation occurs during desflurane anesthesia ( Lutz et al., 1991 ; Young, 1992 ). In patients with mass lesions, equivalent MAC doses of desflurane and isoflurane are similar in terms of absolute CBF, the response to increasing doses, and the preservation of CO2 reactivity ( Ornstein et al., 1993 ; Fraga et al., 2003 ). Desflurane effects on the EEG are also similar to those of isoflurane. At increasing concentrations, the electroencephalographic frequency decreases and the amplitude increases. Burst suppression appears at about 1.24 MAC ( Rampil et al., 1991 ).



Succinylcholine is now very infrequently used in pediatric anesthesia, because of its association with life-threatening hyperkalemia and cardiac arrest in children with undiagnosed myopathies. This has led to a “black box warning” by the U.S. Food and Drug Administration, reserving its use in children for emergency intubation where securing the airway is necessary. Most often, the use of high-dose nondepolarizing neuromuscular blockers is appropriate even for emergently securing the airway in most children, but there are cases in which the most rapid and immediate pharmacologic paralysis is required, and succinylcholine remains the drug of choice in these circumstances.

Life-threatening hyperkalemia has also been associated with the administration of succinylcholine after many types of central nervous system disorders, including closed-head injury, even without motor deficits ( Mazze et al., 1969 ; Thomas, 1969 ; Smith and Grenvik, 1970 ; Stone et al., 1970 ; Stevenson and Birch, 1979 ; Frankville and Drummond, 1987 ), cerebral hypoxia caused by near-drowning (Tong, 1987 ), subarachnoid hemorrhage ( Iwatsuki et al., 1980 ), encephalitis ( Cowgill et al., 1974 ), cerebrovascular accidents ( Cooperman, 1970 ), and paraplegia ( Cooperman et al., 1970 ; Tobey, 1970). The onset of the period of vulnerability is not well defined. It may begin as early as 24 to 48 hours after injury and may last up to 1 to 2 years after injury ( Cooperman, 1970 ). Because the period of risk for succinylcholine-induced hyperkalemia after cerebral injury is undefined, succinylcholine should be avoided in these patients, except in the period immediately after injury.

Succinylcholine can increase CBF and ICP in patients with reduced intracranial compliance ( Cottrell et al., 1983 ; Thiagarajah et al., 1985 ; Minton et al., 1986 ; Stirt et al., 1987a ), probably because of cerebral stimulation from succinylcholine-induced increases in afferent muscle spindle activity ( Lanier et al., 1986 ). The increases in CBF and ICP can be blunted by deep general anesthesia or by previous paralyzing or “defasciculating” doses of nondepolarizing muscle relaxants ( Minton et al., 1986 ; Stirt et al., 1987a ). In contrast, most nondepolarizing relaxants have little effect on CBV and ICP ( Lanier et al., 1985 ; Minton et al., 1985 ; Rosa et al., 1986 ; Stirt et al., 1987b ), unless associated with histamine release (d-tubocurarine, atracurium), which causes transient cerebrovasodilation and increased ICP (Tarkkanen et al., 1974 ; Vesely et al., 1987 ). Atracurium, in doses that do not release histamine, does not increase ICP, despite the accumulation of laudanosine, a major metabolic product of atracurium and a potential central nervous system arousal agent.

Nondepolarizing Muscle Relaxants

The presence of motor deficits or the administration of anticonvulsants may affect the dose of nondepolarizing muscle relaxant necessary in neurosurgical patients. Hemiplegia from an upper motor neuron lesion (such as a stroke or a brain tumor) is associated with resistance to nondepolarizing relaxants on the paretic side ( Graham, 1980 ; Moorthy and Hilgenberg, 1980 ; Azar, 1984 ; Shayevitz and Matteo, 1985 ). Excessive doses of muscle relaxants may be given if dosage is guided by a nerve stimulator monitoring a hemiplegic extremity. In contrast, an increased response to nondepolarizing muscle relaxants is observed in paretic muscle lower motor neuron lesions (e.g., paraplegia and quadriplegia) ( Brown and Charlton, 1975 ; Azar, 1984 ).

Acute administration of several anticonvulsants, including phenytoin, phenobarbital, trimethadione, and ethosuximide, enhances nondepolarizing neuromuscular blockade or delays its reversal ( Ghandi et al., 1976 ; Spacek et al., 1999 ). Patients receiving chronic phenytoin or carbamazepine therapy are resistant to the effects of nondepolarizing relaxants, including pancuronium ( Chen et al., 1983 ; Roth and Ebrahim, 1987 ), metocurine ( Ornstein et al., 1985 ), vecuronium ( Ong et al., 1986 ; Ornstein et al., 1987 ; Alloul et al., 1996 ), and rocuronium ( Spacek et al., 1999 ; Hernandez-Palazon et al., 2001 ) but interestingly, not mivacurium or atracurium ( Ornstein et al., 1987 ; Spacek et al., 1996 , 1997). The cause of phenytoin-induced resistance to nondepolarizing muscle relaxants and the lack of the same effect with mivacurium or atracurium are unclear. Finally, no data have yet been published describing the interactions of the anticonvulsant felbamate, gabapentin, levetiracetam, tiagabine, topiramate, sodium valproate, or valproic acid with nondepolarizing neuromuscular blocking drugs.


The direct-acting vasodilators, including sodium nitroprusside, adenosine triphosphate (ATP), adenosine, nitroglycerin, diazoxide, and hydralazine, are cerebrovasodilators and may increase CBF and ICP (Stoyka and Schultz, 1975 ; Turner et al., 1977 ; Cottrell et al., 1978, 1980 [65] [66]; Marsh et al., 1979 ; Ghani et al., 1989 ; McDowall, 1985 ). The calcium channel blockers also raise CBF and ICP (Cottrell et al., 1984 ; Mazzoni et al., 1985 ). These drugs should therefore be avoided in patients with reduced intracranial compliance, unless the dura is open or ICP is monitored. Indirect-acting antihypertensives, including trimethaphan (a ganglionic blocker), propranolol and esmolol (β-adrenergic blockers), and labetalol (a combined α/β-blocker), do not increase CBF or ICP ( Magness et al., 1973 ; Turner et al., 1977 ; VonAken et al., 1982 ; McDowall, 1985 ). These agents are useful for the control of blood pressure in patients with elevated ICP. Trimethaphan may interfere with the neurologic examination by causing mydriasis, cycloplegia, or anisocoria.

Sodium nitroprusside lowers the range of cerebral autoregulation. Brain-surface oxygen tension is greater ( Maekawa et al., 1979 ) and metabolic disturbances in brain biochemistry (e.g., lactate, pyruvate, and phosphocreatine levels) are less during nitroprusside-induced hypotension than with trimethaphan-induced or hemorrhage-induced hypotension ( Michenfelder and Theye, 1977 ). In addition, cortical blood flow and electrical activity are better preserved at lower MAP with sodium nitroprusside ( Ishikawa and McDowall, 1980 ). Nitroprusside, however, induces more pronounced blood-brain barrier dysfunction than does trimethaphan ( Ishikawa et al., 1983 ). Because CBF, brain oxygen tension, neuronal function, and brain metabolism are better maintained with sodium nitroprusside than with trimethaphan at a MAP of 50 mm Hg, sodium nitroprusside is the preferred agent for deliberate hypotension.

Copyright © 2008 Elsevier Inc. All rights reserved. -

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

Copyright © 2005 Mosby, An Imprint of Elsevier



The preoperative evaluation of the pediatric patient is discussed in Chapter 8 , Preoperative Preparation. The general principles of preoperative assessment of children should be applied to the child with neurosurgical disease. In addition to these basic issues, the assessment and documentation of neurologic impairment and risk of intraoperative complications must be addressed. Of central importance in the preoperative assessment is for the anesthesiologist to determine whether elevated ICP is present, to assess whether a risk exists for regurgitation and aspiration, and to anticipate what surgical positioning will be required and to know its impact on anesthetic management.

The medical history may provide evidence of reduced intracranial compliance or elevated ICP. A history of headaches, especially postural headaches that are worse in a recumbent position or in the morning, is highly suggestive of intracranial hypertension. Although the child who is less than 3 years of age is not able to communicate the existence of headache, the parents frequently describe irritability, which suggests that headache may exist. Feeding intolerance or vomiting, especially in the morning and without attendant nausea, is also evidence of intracranial hypertension and should lead the anesthesiologist to check the serum electrolyte and blood urea nitrogen levels preoperatively. A history of protracted vomiting may also suggest that a significant risk of aspiration exists during the anesthetic induction and should prompt an appropriate modification of the induction technique (discussed later in this chapter). Table 18-2 outlines the clinical signs associated with intracranial hypertension in infants and children.

TABLE 18-2   -- Signs of intracranial hypertension in infants and children



Infants and Children



Decreased consciousness

Full fontanel


Cranial nerve (III and VI) palsies

Widely separated cranial sutures


Loss of upward gaze (setting sun sign)

Cranial enlargement


Signs of herniation, Cushing's triad, pupillary changes



The preoperative history should also include medications, which frequently include anticonvulsant agents and steroids. In patients taking steroids, intraoperative and postoperative steroid administration is necessary both to treat cerebral edema and as replacement therapy for adrenal axis suppression.

The physical examination includes an assessment of the patient's airway, cardiovascular status, state of hydration, and neurologic status including an age-specific Glasgow Coma Scale (GCS) score ( Table 18-3 ) ( Reilly et al., 1988 ; Tatman et al., 1997 ), if consciousness is obtunded. If a cardiac murmur is detected during auscultation of the heart in a patient for whom a suboccipital craniotomy in the sitting position is planned, a formal cardiologic evaluation is indicated to rule out the existence of intracardiac shunting, which might lead to paradoxical air embolism (discussed later in this chapter), therefore relatively contraindicating use of the sitting position. Indeed, because approximately 20% to 25% of children without heart murmurs have a patent foramen ovale that would permit paradoxical air embolism in the event of a VAE ( Schwarz et al., 1994 ; Fuchs et al., 1998 ), a cogent argument could be made for routine echocardiographic evaluation of all children in whom sitting craniotomies are planned to identify those individuals who should not be operated on in the sitting position ( Fuchs et al., 1998 ).

TABLE 18-3   -- The Glasgow Coma Scale and Pediatric Glasgow Coma Scale

Glasgow Coma Scale (3 to 15 possible points)

Best Motor Response


Obeys commands


Localizes pain


Withdraws from pain


Abnormal flexion


Abnormal extension



Best Verbal Response











Eye Opening


Spontaneous eye opening


Opens eyes to voices


Opens eyes to pain



Pediatric Glasgow Coma Scale (3 to 15 possible points) ( Tatman et al., 1997 )

Best Motor Response

>5 years

<5 years


Obeys commands

Normal spontaneous movements


Localizes supraorbital pain

Withdraws to touch


Withdraws from nail bed pain



Flexion to supraorbital pain



Extension to supraorbital pain





Best Verbal Response

>5 years

<5 years



Alert, babbles, coos, words to usual ability



Less than usual ability, irritable cry


Inappropriate words

Cries to pain


Incomprehensible sounds

Moans to pain


No response to pain





Eye Opening




To voices


To pain




Closed (swelling, dressing)

Pain should be made by pressing hard on the supraorbital notch (beneath medial end of eyebrow) with your thumb, except for M4, which is tested by pressing hard on the flat nail surface with the barrel of a pencil.




The presence of intracranial hypertension may be deduced from the history, but when the medical history is brief, as in the case of head trauma, the existence of intracranial hypertension is sometimes more difficult to assess. The GCS (see Table 18-3 ) is very useful in the acute setting; a coma score of less than 6 in adults suggests the presence of acute intracranial hypertension ( Bruce et al., 1977 ), and the GCS on presentation is well correlated with clinical outcome ( Reilly et al., 1988 ; Grewal and Sutcliffe, 1991 ; Ong et al., 1996 ; Servadei et al., 1998 ; Coughlan et al., 2003 ) and the intraoperative development of a coagulopathy in children ( Keller et al., 2001 ). In general, a patient who is comatose and has an abnormal response (decorticate posturing, decerebrate posturing, or flaccidity) to a painful stimulus (such as squeezing a digit) probably has elevated ICP, whereas the child who withdraws appropriately from a painful stimulus or who has a higher response to pain, such as eye opening, probably does not have elevated ICP at that time.

The elements of Cushing's triad—hyperventilation, bradycardia, and hypertension—are late sequelae of intracranial hypertension and portend impending cerebral herniation. Other late signs of intracranial hypertension associated with herniation syndromes include pupillary asymmetry; pupillary dilation or eccentricity; cranial nerve palsies, particularly third and sixth nerve paralysis; irregular respirations, particularly Cheyne-Stokes respirations; and hypotension ( Bell and McCormick, 1978 ). These findings indicate a premortal condition and require immediate and decisive therapy, including endotracheal intubation and hyperventilation and the intravenous infusion of mannitol.

The remainder of the neurologic examination includes the assessment of cranial nerve function, peripheral tone and strength, deep tendon reflexes, plantar reflexes, and coordination. If a hemiparesis is present, it should be noted. The anesthesiologist need not perform a highly detailed neurologic examination, a task left for the attending neurosurgeon and the neurologist, but should focus on the neurologic functions important in planning the anesthetic management and postoperative care.

The laboratory studies of the neurosurgical patient include determination of the hematocrit and serum electrolyte levels, the latter to identify abnormalities of serum sodium and potassium associated with vomiting and dehydration, mannitol, or diuretic therapy or to detect the syndrome of inappropriate secretion of antidiuretic hormone, which may complicate intracranial pathology (see Chapter 4 , Regulation of Body Fluids and Electrolytes). A blood sample should be sent to the blood bank for typing and cross-matching. Additional studies, such as an electrocardiogram, an echocardiogram, a coagulation profile, renal or hepatic function studies, blood gas analysis, or radiographic studies, may be indicated as well. Patients with suprasellar tumors such as craniopharyngiomas frequently have pituitary dysfunction and therefore should have a complete endocrine evaluation, including thyroid and adrenal function studies. Adrenal or thyroid replacement therapy can then be prescribed as indicated from the results of the studies.


Premedication is often withheld from children undergoing neurosurgery. Sedatives and opioids should never be administered to an unobserved and unmonitored patient with elevated ICP, hypotonia, or central nervous system depression. Premedication in such patients may produce airway obstruction or respiratory depression and hypercapnia, with subsequent elevation of ICP. On the other hand, premedication, with a benzodiazepine such as midazolam, of children with diminished intracranial compliance is not associated with respiratory depression and eliminates anxiety and crying with their attendant cardiovascular changes, which themselves elevate ICP.

Children with intracranial aneurysms or arteriovenous malformations, who do not have increased ICP and who are at risk for intracranial bleeding if arterial hypertension occurs, should be sedated before transport to the operating room. Various premedications that may be used are discussed in Chapter 8 , Preoperative Preparation. In older children with vascular malformations, preoperative β-adrenergic blockade with oral propranolol the night before and the morning of surgery helps to prevent tachycardia and hypertension during induction of anesthesia and laryngoscopy and is a useful adjunct when controlled hypotension is used. A total oral propranolol dose of 1 mg/kg divided into three or four doses at 6-hour intervals is usually effective. Alternatively, rapid intravenous β-blockade can be achieved before the induction of anesthesia with an intravenous infusion of esmolol (500-mcg/kg bolus followed by 50 to 200 mcg/kg per min). Although it is a drug to be avoided in patients with reduced intracranial compliance, intramuscularly delivered ketamine can be considered in patients who have no intravenous access, who will not tolerate an oral premedication, and who will benefit from premedication.


Induction of general anesthesia should be planned to minimize the risk of inducing sustained, life-threatening intracranial hypertension. As a rule, acute neurosurgical patients have increased ICP and are at risk for gastric aspiration; they should have an intravenous cannula inserted preoperatively. Anesthesia in these children should be induced with an intravenous agent to minimize the risk of aspiration. Induction with a short-acting barbiturate or propofol and a muscle relaxant diminishes ICP and allows the anesthesiologist to induce hyperventilation early in the beginning of anesthesia. It usually is not feasible preoperatively to insert devices to monitor ICP using local anesthesia in children. Thus, ICP usually cannot be monitored before anesthetic induction.

If ICP is believed to be severely elevated, adjunctive measures should be instituted before induction. These include treatment with mannitol (0.25 to 0.5 g/kg) ( Marshall et al., 1978 ) or with potent loop diuretics (furosemide, 0.25 to 1.0 mg/kg), steroid therapy in children with cerebral neoplasms (dexamethasone, 0.5 mg/kg), and drainage of CSF when possible in children with an indwelling ventriculoperitoneal or ventriculoatrial shunt (discussed later in this chapter).

Propofol (2 to 4 mg/kg) is also for induction of anesthesia. Sodium thiopental (4 to 7 mg/kg), sodium thiamylal (4 to 7 mg/kg), and methohexital (2 to 3 mg/kg) are short-acting barbiturates that are also appropriate for intravenous induction, and each markedly lowers ICP. Ketamine has no rational use as an induction agent in neurosurgical anesthesia (see earlier discussion).

It is not rare in the practice of pediatric anesthesia to have difficulty in establishing preoperative venous access. Occasionally, vascular access is extraordinarily difficult, and attempts to place an intravenous catheter result only in further crying and struggling, thereby exacerbating ICP. If no clinical deterioration ensues (that is, the child is still conscious), the intracranial compliance is probably adequate to allow modification of the anesthetic technique.

In the situation described, two alternatives exist. The first is to induce sleep with barbiturates administered by rectum. Thiopental (20 to 25 mg/kg as a 10% solution) may be used (see Chapter 10 , Induction of Anesthesia). After about 5 to 10 minutes, when the child is well sedated, a vein is cannulated with the use of a subcutaneous infiltration of lidocaine or the preapplication of local anesthetic cream (EMLA [lidocaine/prilocaine] or Lidoderm [lidocaine]). Induction of general anesthesia then may proceed, with the rectal barbiturate augmented with an intravenous induction dose of propofol or barbiturate before relaxation and laryngoscopy.

The second alternative is to perform an inhalation induction. While there are some theoretical reasons why isoflurane may be preferable to sevoflurane or halothane as a neuroanesthetic (see earlier discussion), there is little question that inhalation inductions usually proceed more smoothly with sevoflurane or halothane, and the advantages of isoflurane are meager against the deleterious effects on ICP of coughing, breath holding, laryngospasm, and subsequent hypercapnia. These effects are more likely to occur during isoflurane inhalation inductions than during sevoflurane or halothane inductions. When consciousness is lost, ventilation should be controlled by the anesthesiologist, with hyperventilation initiated. The end-tidal concentration of the halogenated agent should not exceed 1 MAC, to minimize the risk of inducing intracranial hypertension. After intravenous access is secured, the anesthetic technique may be changed to a more conventional propofol/barbiturate-nitrous oxide-opioid-relaxant neuroanesthetic.

Neuromuscular blocking agents usually are administered to facilitate laryngoscopy and endotracheal intubation. If the planned procedure is a lengthy one and there is no concern of gastric aspiration, a nondepolarizing relaxant may be administered. Because histamine release transiently dilates the cerebral vasculature and elevates ICP, while simultaneously depressing systemic arterial pressure and CPP, drugs that do not release histamine are preferable. Vecuronium or pancuronium (0.08 to 0.15 mg/kg) and rocuronium (0.8 to 1.5 mg/kg) are nondepolarizing muscle relaxants that cause no histamine release.

When a significant risk of gastric aspiration exists, the most rapid intubation of the trachea possible is desirable. Succinylcholine may be used in this circumstance. Although succinylcholine may elevate ICP in patients with diminished intracranial compliance ( Minton et al., 1985 ), the elevation of ICP is usually small, transient, and overshadowed by the clear adverse effects on ICP of retching, coughing, aspiration, and hypoventilation, which may occur with a more prolonged induction ( Marsh et al., 1980 ). An alternative to the use of succinylcholine is high-dose rocuronium. At doses of 0.8 mg/kg, intubating conditions can be achieved in less than 1 minute (see Chapter 6 , Pharmacology of Pediatric Anesthesia) and results in satisfactory conditions for intubation in about 1 minute. The duration of neuromuscular blockade after this dose is about 1 hour ( Lennon et al., 1986 ).

Laryngoscopy is a potent stimulant and nearly always results in systemic and intracranial hypertension if anesthetic depth is inadequate ( Moss et al., 1978b ). Adjuvant drugs useful to blunt this response after induction of anesthesia are the rapid-acting opioid fentanyl (1 to 5 mcg/kg), sufentanil (0.2 to 1 mcg/kg), alfentanil (10 to 50 mcg/kg), and remifentanil (1 mcg/kg), or lidocaine (1.5 mg/kg), given intravenously 1 or 2 minutes before laryngoscopy ( Bedford et al., 1980 ; Namill et al., 1981 ).

When a risk of aspiration exists, a modification of the rapid sequence induction technique is indicated. This modification calls for intravenous induction with propofol or thiopental, and a neuromuscular blocking drug, which may be either succinylcholine (preceded by a defasciculating drug), rocuronium, or high-dose vecuronium. Immediately after consciousness is lost, the patient is gently hyperventilated with oxygen while cricoid pressure is applied by the assistant. In so doing, ICP may be controlled while complete neuromuscular blockade and good conditions for intubation are awaited.


Certain considerations are common to all positions of the patient used during neurosurgery. If neurosurgery is known for one characteristic, it is its lengthy duration. This feature requires the anesthesiologist to carefully pad bony prominences such as the condyles of the elbows, the sacrum, and the ankles; to ensure that the limbs are in neutral positions; and to avoid stretching or compressing peripheral nerves. Eyes should be lubricated with a wetting agent or an antibiotic ointment to prevent corneal injury and should be carefully taped shut or padded. Urinary catheters are advisable for long surgical procedures to prevent bladder distention and to aid in fluid management.

As for most other pediatric surgical procedures, positioning a child for neurosurgery presents certain challenges. Although patients frequently disappear under surgical drapes and equipment, the anesthesiologist must make efforts to guarantee an unobstructed view of part of the patient, to have access to an extremity in case additional vascular catheters are needed, and, most important, to have access to the patient's airway and the anesthesia circuit in order to permit visual inspection of the airway and intermittent suctioning of the trachea. These goals may be achieved by suspending the drapes on bars or by tunneling the drapes.

Supratentorial procedures (such as parietal craniotomy or placement of a ventriculoperitoneal shunt) are generally performed with the patient lying supine and the head turned to one side. The anesthesiologist should secure the endotracheal tube to the nondependent side of the mouth to prevent oral secretions from loosening the adhesive tape used to fasten the endotracheal tube.

The prone position is used in pediatric neurosurgery for spine surgery and encephalocele repair and by some neurosurgeons for suboccipital craniotomy. In establishing the prone position, it is important to ensure that the patient's weight does not rest on the abdomen but rather is supported by bolsters under the pelvis and chest. Excessive abdominal pressure impedes ventilation of the lungs and leads to compression of the inferior vena cava and distention of the epidural venous plexus, either of which may increase surgical blood loss during spinal surgery. Flexion of the neck in the prone position can change the position of the endotracheal tube and cause intubation of a main stem bronchus.

In children older than about 4 years, explorations of the posterior fossa frequently are performed with the child in the sitting or semisitting position, to minimize intraoperative bleeding and tissue swelling and to facilitate surgical exposure. In younger children, these procedures are done with the child lying prone. Positioning in either the prone or the sitting position requires extreme flexion of the neck to expose the suboccipital skull, which frequently leads to intraoral kinking of the endotracheal tube. The use of armored endotracheal tubes minimizes this problem. Extreme neck flexion may also compromise venous or lymphatic drainage from the tongue and result in postoperative macroglossia. For this reason, oropharyngeal airways should not be in place while the patient is positioned ( Albin et al., 1976 ).

The sitting position for suboccipital craniotomy is still preferred by some neurosurgeons for several reasons. This position facilitates exposure and use of the operating microscope; places the head above the heart, which favors venous drainage away from the surgical field; and reduces blood loss and edema formation. Although the seated position is advantageous to the neurosurgeon, it presents the anesthesiologist with two problems. The first is that cardiovascular stability is impaired during general anesthesia, and movement of the patient from the supine to the sitting position frequently results in postural hypotension. This can be prevented by wrapping the lower extremities with elastic bandages to minimize venous pooling, using a nitrous oxide-propofol-opioid technique, or limiting halogenated inhalation agents to less than 1 MAC before positioning ( Marshall et al., 1984 ). Then the patient can be positioned slowly, in stages, while the anesthesiologist carefully monitors the patient's hemodynamics. The second problem for the anesthesiologist is more difficult to solve—that is, the occurrence of VAE ( Bedford, 1983 ).


Routine monitors for all pediatric anesthesia include precordial or esophageal stethoscope, pulse oximeter, continuous electrocardiogram, temperature thermistor, and noninvasive and/or intra-arterial measurement of blood pressure. Capnography for measurement of end-tidal CO2 tension provides confirmation of endotracheal intubation, continuous monitoring of the patency of the breathing circuit, and a reliable trend monitor for guidance of hyperventilation of the neurosurgical patient.

Surgery within the cranial cavity, surgery expected to result in significant blood loss (loss of 20% of the estimated blood volume), or surgery with the potential for rapid blood loss (loss of 10% of the estimated blood volume in less than 15 minutes) requires the use of an arterial cannula to permit continuous monitoring of arterial blood pressure and episodic determination of arterial blood gases, electrolytes, hematocrit, etc.

Central venous cannulation via the internal jugular, external jugular, subclavian, or femoral vein is indicated for a craniotomy in the sitting position to permit aspiration of air from the right atrium and for other procedures in which major blood loss is anticipated. The use of the internal jugular vein is typically avoided in children with established, or at risk for, intracranial hypertension because of the potential to compromise cerebral venous drainage and further elevate ICP. Alternative sites are the external jugular, subclavian, and femoral veins.

Air can be successfully aspirated from central venous cannulas in 33% of episodes of VAE in children ( Cucchiara and Bowers, 1982 ). Multiorifice central venous catheters specifically engineered for maximal aspiration of air, when properly positioned at the junction of the right atrium and superior vena cava, are used during sitting craniotomies in adults. These catheters are not commercially available in pediatric sizes or lengths, and their usefulness has not been demonstrated in pediatric neuroanesthesia.

Venous Air Embolism

Veins are thin-walled structures that collapse when incised if their pressure is subatmospheric. However, diploic veins, which bridge the scalp and the dura, are tethered open by bony connections in the skull and do not collapse when incised. The dural sinuses of the brain are also tethered open by dural and bony connections.

In the sitting position, there is a significant hydrostatic gradient between the head and the heart, and air may enter these open vessels. Venous air travels first to the heart and then to the lungs. Small amounts of air result in no physiologic change as long as there is no anatomic cardiac shunt (see later). Larger amounts of air become trapped in the lungs and result in pulmonary embolism: increased alveolar dead space and subsequent retention of CO2, ventilation-perfusion mismatch and hypoxemia, and pulmonary hypertension ( Adornato et al., 1978 ). Still larger amounts of venous air may become air-locked in the superior vena cava ( Martin et al., 1984 ) or at the junction of the superior vena cava and the right atrium ( Bunegin et al. 1981 ; Cucchiara et al., 1985 ), reduce cardiac output and blood pressure, and, in the extreme, produce pulseless electrical activity and cardiac arrest.

In the sitting position, the incidence of Doppler-detectable VAE in adults during craniotomies ranges from 20% to 40%, and roughly 40% of detectable air emboli are associated with hypotension (Michenfelder et al., 1972 ; Bithal et al., 2003 ). In children, the incidence of Doppler-detectable air emboli is about the same as in adults ( Cucchiara and Bowers, 1982 ; Bithal et al., 2004 ), whereas VAE sufficient to decrease pulmonary blood flow and end-tidal CO2 occurs in about 9% of cases, of which 21% are associated with hypotension ( Harrison et al., 2002 ). In a pediatric patient, because a child's head makes up a greater surface area of the body, the incidence of VAEs during a craniotomy is at least the same ( Cucchiara and Bowers, 1982 ; Bithal et al., 2004 ) or higher, approximately 66% to 80% (Soriano et al., 2002 ). In addition, in children, the frequency with which detectable air emboli are physiologically significant and produce hypotension may be much higher ( Michenfelder et al., 1972 ).

The incidence of VAE during craniosynostosis repair in supine infants is as high as 67% to 83%, with hemodynamic changes occurring in one third of those in whom intravascular air is detected ( Harris et al., 1987 ; Faberowski et al., 2000 ). The pediatric neuroanesthesiologist must monitor for venous air in children undergoing craniotomy in the sitting position, or craniosynostosis repair, and treat air entrainment promptly (see discussion of treatment of VAE later in this chapter).

If a cardiac shunt exists—that is, an atrial septal defect, a ventricular septal defect, a patent foramen ovale, or a patent ductus arteriosus, air may enter the left side of the heart and systemic circulation, embolizing vital organs ( Perkins-Pearson et al., 1982 ; Albin et al., 1984 ; Mehta et al., 1984 ; Cucchiara et al., 1985 ; Soriano et al., 2002 ). This phenomenon is referred to as paradoxical air embolism (PAE) and, in the extreme, may result in cerebral or myocardial infarction. This may be likely if a significant bolus of venous air results in pulmonary arterial hypertension and elevation of right atrial pressure to levels exceeding left atrial pressure ( Adornato et al., 1978 ; Perkins-Pearson et al., 1982 ). The sitting position is therefore relatively contraindicated when such an intracardiac septal defect is known to exist, unless the neurosurgeon has a compelling need to perform the surgery with the patient seated. Children with cardiac murmurs must be evaluated by echocardiography preoperatively to rule out septal defects. Routine preoperative contrast echocardiography before all sitting craniotomies in individuals with normal physical examinations is of unproved value in preventing PAE ( Black et al., 1990 ).

A positive right-to-left atrial pressure gradient also may occur in the absence of a large air embolus; such a pressure gradient develops in many patients during prolonged anesthesia in the sitting position (Perkins-Pearson et al., 1982 ). This would allow even small volumes of air to move across a probe-patent foramen ovale into the systemic circulation. Because approximately 20% of the adult population have a probe-patent foramen ovale ( Goss, 1973 ), the risk of PAE may be relatively great. This fact has prompted some to ask whether the sitting position for suboccipital craniotomy should not be abandoned altogether ( Albin, 1984 ). Although this conclusion is not widely shared, certainly the risk of PAE mitigates against the sitting position if a septal defect is known to exist preoperatively.

VAE is not confined to neurosurgical procedures in the sitting position. It has been reported in children during craniosynostosis repair in either the supine (Harris et al., 1986, 1987 [118] [119]; Phillips and Millikan, 1988) or the lateral ( Joseph et al., 1985 ) position and in children undergoing posterior fossa explorations in the prone position ( Meridy et al., 1974 ). It is, however, an unusual complication with these surgical positions and one that may be avoided by controlling ventilation, preventing negative intrathoracic pressure, maintaining adequate intravascular volume, and monitoring for venous air during craniosynostosis repair, particularly if a large osteotomy is anticipated (during craniofacial surgery, for example).

Air may enter the venous system slowly and insidiously, not altering physiology for a period of many minutes. On the contrary, although uncommon, air may enter rapidly in large volumes, producing immediate symptoms or even cardiac arrest if a large dural sinus has inadvertently been opened by the neurosurgeon. The goal of monitoring for VAE is to detect air entry well before physiologic changes occur, giving the anesthesiologist the time to inform the neurosurgeon early.

Several techniques and indicators may be used to detect VAE. In order of sensitivity and specificity ( Fig. 18-2 ), they are transesophageal echocardiography (TEE), precordial Doppler ultrasonography, end-tidal capnography, end-tidal nitrogen detection, pulmonary artery pressure, central venous pressure, esophageal stethoscope, and measurement of systemic blood pressure ( Gildenberg et al., 1981 ;Cucchiara et al., 1985 ; Black et al., 1990 ).


FIGURE 18-2  Relative sensitivities of different methods of monitoring for venous air embolism. The precordial Doppler is approximately 40 times more sensitive than the capnograph or pulmonary artery catheter in detecting vascular air. It will detect 1/100th of the amount of air associated with a change in blood pressure.  (Data from Gildenberg PL, O—Brien RP, Britt WJ, et al.: J Neurosurg 54:75, 1981.)




Of the various techniques for monitoring VAE, central venous pressure measurement, the esophageal stethoscope, and systemic blood pressure measurement cannot detect venous air until clinical deterioration is well established ( Adornato et al., 1978 ). According to Bithal and others (2004) , of 96 children scheduled to undergo a posterior craniotomy, capnometry detected 22% of air emboli (decrease in CO2 tension of >0.7 kPa) and hypotension occurred in 33% of children (>20% change from baseline). Hence, more sensitive monitoring is required to detect smaller amounts of venous air before continuing embolization reaches physiologic significance.

Transesophageal echocardiography (TEE) has come into favor as the most sensitive, invasive, intraoperative instrument for detecting air emboli. In addition, it has the added benefit of identifying and localizing left atrial air.

In 1998, Mammoto and others (1998) described using TEE in 21 adult patients undergoing craniotomies in the sitting position to detect venous and paradoxical air emboli. VAEs were detected in every patient, whereas paradoxical air emboli appeared in only three patients with the most severe VAEs causing a reduction in end-tidal CO2 and an increase in pulmonary artery pressures. Although TEE may be an excellent monitor for such intraoperative complications, especially in children, anesthesiologists must be aware of the potential for even bigger hazards with the use of this probe.

Particularly in children less than 10 kg, the placement of the TEE probe may be difficult and high resistance may be met with its passage into the esophagus. The anesthesiologist must take care to use a pediatric probe, lubricated well, and must never force the probe against resistance, as children are prone to esophageal perforation. In addition, the TEE probe may cause difficulty with ventilation and maintenance of adequate tidal volumes in smaller infants. Thus, although highly sensitive, the use of a TEE probe to detect air emboli in children during neurosurgical procedures may not be optimum and, to date, has not become the standard of practice.

Doppler ultrasound is the most sensitive noninvasive monitoring technique; it can detect minute quantities of venous air ( Edmonds-Seal and Maroon, 1969 ; Maroon et al., 1969 ; Edmonds-Seal et al., 1970; Michenfelder et al., 1972 ; Maroon and Albin, 1974 ). In fact, the introduction of Doppler ultrasound into clinical practice transformed the once-held notion that VAE was a rare and catastrophic event to the present understanding that VAE is a common phenomenon. The Doppler detector should be placed in the second to fourth intercostal space to the right of the sternum; placement over the right atrium should be confirmed by rapid injection of saline solution into a peripheral intravenous catheter or into a central venous catheter. For those patients placed in the prone position and at risk for VAE, the posterior Doppler probe placement between the right scapula and spine can be effective for infants weighing less than 6 kg ( Soriano et al., 1994 ). Because of the nearly continuous use of electrocautery, the Doppler apparatus is of limited use when air is most likely to embolize, that is, during the surgeon's entry into the cranial cavity. A second monitor is therefore necessary, one that is not affected by electrical interference in the operating room and one that is less sensitive than the Doppler apparatus, the capnograph. When a significant volume of air embolizes to the pulmonary vascular bed, alveolar dead space is created and the end-tidal CO2 level decreases. A sudden decline in end-tidal CO2 level is highly suggestive of VAE. If, however, the decreasing end-tidal CO2 concentration is accompanied by a decrease in arterial blood pressure, the diagnosis may be obscured, because a reduction in cardiac output of any cause may diminish end-tidal CO2 tension. If the anesthesiologist remains cognizant of this limitation, capnography is a valuable adjunctive technique for monitoring VAE.

When air embolizes to the pulmonary vascular bed, the nitrogen within the air embolism is exhaled by the lungs and may be detectable as end-tidal nitrogen by exhaled gas analysis ( Drummond et al., 1985; Matjasko et al., 1985 ). The appearance of nitrogen in the end-tidal gases in association with decreasing end-tidal CO2 tension is virtually pathognomonic of VAE ( Losee et al., 1982 ).

Some authors have recommended placement of a pulmonary artery catheter both to detect and to treat VAE in adults. This practice is not used in children for three reasons. Pulmonary artery pressure monitoring is no more sensitive than capnography for detecting air embolism, pulmonary artery catheters are associated with significant risk, and the proximal lumen of a pulmonary artery catheter is too small for effective and rapid aspiration of cardiac air. Artru (1992) compared the volume of air embolism recovery using a multiorifice catheter and a 7F pulmonary artery catheter; he found a more than fivefold increase in gas recovery using the former. One can readily see how the volume of gas recovery would be further diminished through a pediatric 5F pulmonary artery catheter. If, however, a pulmonary artery catheter is inserted for other indications, the anesthesiologist should consider placing an additional central venous catheter specifically to treat air embolism or use a multiorifice introducer sheath for pulmonary artery catheters ( Bowdle and Artru, 1988 ) that extends to the superior vena cava-right atrial junction and obviates the need for an additional central catheter.

In summary, all patients undergoing surgery that is associated with a risk of VAE should be monitored using a device to detect the entry of venous air, in addition to capnography and end-tidal gas analysis. The pulmonary artery catheter should not be used routinely for this purpose. Consideration should be given to placing an appropriately positioned central venous catheter to allow aspiration of venous air as well as to guide fluid therapy.

Confirming Placement of a Central Venous Cannula

The ability to aspirate air from the right atrium of a patient who has had a VAE was first recognized in 1965 ( Michenfelder et al., 1966 ). After initial reports of successful air aspiration from a central venous catheter, the decision to place the central venous catheter in the mid portion of the right atrium was made intuitively ( Michenfelder et al., 1969 ). Later, Bunegin and others (1981) showed that in an in vitro model of air aspiration, air was more likely to become trapped at the junction of the superior vena cava and the right atrium, which in adults is 3 cm above the sinoatrial node. This observation was confirmed in vivo several years later with the use of two-dimensional echocardiography during craniotomies performed in the sitting position ( Cucchiara et al., 1985 ). The ideal position for the tip of the central venous catheter is, therefore, the junction of the superior vena cava and the right atrium. Catheter placement may be confirmed in one of three ways.

The first method of confirming correct catheter placement is fluoroscopy or a chest radiograph, both readily available in the operating room. The second method is to attach the catheter (via a metal hub or a stopcock) to the V lead of an electrocardiograph during placement ( Martin, 1970 ). When the catheter tip is at the level of the sinoatrial node, large P waves have a characteristic biphasic conformation (Fig. 18-3 ). The catheter may then be withdrawn 1 to 3 cm (depending on the patient's age and size) to the level of the vena caval-right atrial junction. For the catheter to conduct electrocardiographic currents, it must be filled with an electrolyte solution. The solution may be either hypertonic saline solution or 8.4% sodium bicarbonate; the latter is more readily available in the operating room suite (Colley and Artru, 1984 ). The third method is to use pressure waveform monitoring while inserting the central venous catheter. In this technique, the catheter is inserted until pressure monitoring demonstrates the tip to lie within the right ventricle. The catheter is withdrawn until the ventricular waveform disappears; then it is withdrawn an additional 4.3 cm in adult-sized patients. The catheter orifice then lies in the ideal position at the vena caval-right atrial junction ( Mongan et al., 1992 ). Unfortunately, there are no data guiding this technique for pediatric patients.


FIGURE 18-3  Continuous electrocardiographic monitoring of central venous catheter placement for craniotomy in the sitting position. The catheter is filled with hypertonic saline solution or 8.4% sodium bicarbonate and attached to a metal stopcock, which serves as the V lead of the electrocardiogram. When the catheter tip is above the sinoatrial (SA) node, the P wave has a negative deflection. When the catheter tip lies at the level of the SA node, the P wave is biphasic, with equal positive and negative voltages. When the catheter tip lies below the SA node, the P wave voltage is predominantly positive. In the adult, the correct position for the catheter tip is probably 1 to 3 cm above the SA node.



Treatment of VAE

When a VAE is detected, treatment is initiated immediately; the treatment is tailored to the severity of the embolus. Figure 18-4 presents a treatment algorithm.


FIGURE 18-4  Decision algorithm for the management of venous air embolism during a craniotomy in the sitting position.



On detection of venous air, the surgeon is notified that air entrainment is occurring. The surgeon responds by identifying the points of air entry and sealing them or by flooding the surgical field with saline solution. Inspiratory hold after lung inflation (Valsalva maneuver) is useful both in preventing further air entry and in assisting the surgeon in identifying the source of air entry ( Sharma and Tripathi, 1994 ). If the air embolus is physiologically significant, that is, it decreases end-tidal CO2 or affects hemodynamics, nitrous oxide should be discontinued and the patient ventilated with 100% oxygen. Because nitrous oxide is less diffusible in blood than in gas, it diffuses into a pulmonary air embolus and enlarges its volume severalfold ( Munson, 1971 ), causing further physiologic compromise ( Mehta et al., 1984 ). Simultaneously, the anesthesiologist may compress the jugular veins under the surgical drapes to retard the rate of air entry into the central circulation until entry has been occluded. In cases of very severe VAE resulting in hypotension, the head of the operating table is lowered to increase venous return to the heart from the lower extremities, to decrease the rate of air entrainment, and to permit cardiopulmonary resuscitation in severe cases. When venous air is detected during a craniotomy performed in a sitting child, the success with which air may be aspirated varies from 38% to 60% (Cucchiara and Bowers, 1982 ).

Positive end-expiratory pressure (PEEP) is not indicated in the treatment of VAE. Theoretically, PEEP might decrease the rate of air entry by raising venous pressure but, in fact, Bedford and Perkins-Pearson (1982) found that PEEP (10 cm H2O) did not elevate venous pressure sufficiently to stop air entry. It further impaired cardiac output and also raised right atrial pressures to levels exceeding left atrial pressures, thus potentially causing PAE in patients with patent foramen ovalae. The latter observation has been challenged by Pearl and Larson (1986) , who demonstrated that PEEP (8 cm H2O) did not increase right atrial pressures in excess of left atrial pressures in an animal model of VAE. Using a similar animal model, however, Toung and others (1988) found that PEEP (15 cm H2O) did not raise cerebral venous pressure to above atmospheric pressure, whereas tourniquet compression of the veins of the neck did. The treatment of VAE, therefore, should include jugular compression but omit the use of PEEP. The possibility of PAE through an unsuspected patent foramen ovale mandates that aspiration of air be attempted during all episodes of air embolism. A central venous catheter should be placed in children undergoing a craniotomy in the sitting position.

After VAE has been treated, end-tidal CO2 tension returns to baseline. At this point, nitrous oxide may be reintroduced if desired, whereas end-tidal CO2 is carefully monitored for further changes. If the reintroduction of nitrous oxide causes end-tidal CO2 to decrease, air persists in the pulmonary circulation, and the nitrous oxide should be discontinued and an alternative anesthetic technique adopted (Shapiro et al., 1982 ).


The purpose of neurophysiologic monitoring is to detect neurosurgical trauma or ischemia at a time at which neurologic injury is reversible, thus improving functional outcome. It is used in certain cases in which the risk of mechanical trauma to vital structures, or ischemia, is significant. When alterations in physiologic function occur, it alerts the neurosurgeon and anesthesiologist to alter the neurosurgical approach, or to restore hemodynamics (also see Chapter 9 , Pediatric Anesthesia Equipment and Monitoring).

Although internationally accepted guidelines for intraoperative neurophysiologic monitoring (INM) are still far from being established, an article provides the following guidelines for such monitoring ( Sala et al., 2002 ):



INM is mandatory whenever neurologic complications are expected on a known pathophysiologic mechanism. INM should always be performed when any of the following are involved: supratentorial lesions in the central region and language-related cortex, brainstem tumors, intramedullary spinal cord tumors, conus-cauda equina tumors, rhizotomy for relief of spasticity, or spina bifida with tethered cord.



Monitoring of motor evoked potentials (MEPs) is most appropriate to access functional integrity of descending motor pathways in the brainstem.



Somatosensory evoked potential (SSEP) monitoring is of value in assessment of the functional integrity of sensory pathways leading to the sensory cortex.



Monitoring of brainstem auditory evoked potentials remains a standard technique during surgery in the brainstem, cerebellopontine angle, and posterior fossa.



Mapping of the motor nuclei of cranial nerves VII, IX, X, and XII on the floor of the fourth ventricle is valuable in identifying “safe entry zones” into the brainstem.


Electroencephalography is the most reliable intraoperative monitor for focal cerebral ischemia characterized by an attenuation of high-frequency activity and the appearance of slow delta waves ( Soriano et al., 2002 ). In addition to monitoring the integrity of at-risk cerebral cortex, an intraoperative EEG is used to identify epileptiform foci for surgical excision. Cerebral hypoxia or ischemia causes the appearance of large-amplitude slow waves in the 1- to 4-Hz range in the regions affected. When electroencephalographic monitoring suggests the presence of ischemia, the cause must be identified and corrected, and differentiated from other benign sources of electroencephalographic changes such as increased anesthetic depth, hypocarbia, or hypothermia.

Somatosensory Evoked Potentials

SSEP monitoring is used most commonly during surgery on the spine, but it is also used during neurosurgery around the foramen magnum, or brainstem. During SSEP monitoring, electrical stimulation is made to the median or ulnar nerves at the wrists and the common peroneal or posterior tibial nerves at the ankles. Evoked electrical potentials can be measured overlying the spinal cord and the cerebral cortex. SSEPs are dependent on intact large myelinated peripheral sensory fibers, the posterior (dorsal) columns of the spinal cord, and the spinothalamic track. An absence, decreased amplitude, or increased latency of evoked potentials is highly suggestive of neurologic compromise ( Sala et al., 2002 ).

SSEPs are extremely sensitive to volatile anesthetics and nitrous oxide. Total intravenous anesthesia is generally associated with better preservation of SSEPs than are low-dose volatile anesthetic techniques. Hypothermia also degrades the quality of SSEPs. The application of SSEP monitoring and the effects of anesthetics on SSEP monitoring are discussed in greater detail in Chapter 9 , Pediatric Anesthesia Equipment and Monitoring.

Motor Evoked Potentials

MEPs are a top-down monitoring method of measuring brainstem and spinal cord integrity, whereas SSEPs are a bottom-up system. In other words, in MEP monitoring, the motor strips of the cerebral cortex or the motor fibers of the spinal cord are electrically or magnetically stimulated, whereas the resultant motor nerve action potentials or muscle action potentials that occur distally are recorded. MEPs demonstrate the integrity of the descending/anterior columns of the spinal cord, whereas SSEPs primarily monitor the integrity of the ascending/dorsal columns ( Sala et al., 2004 ).

MEPs are even more sensitive to anesthetic agents than are SSEPs, and, of course, motor action potentials would disappear altogether if neuromuscular blocking agents were to be used. Even low-dose volatile anesthesia can obliterate MEPs ( Kalkman et al., 1991 ). The role of MEP monitoring in pediatric neurosurgery is as yet undefined, as is the ideal anesthetic technique. In adults, an anesthetic technique using ketamine, alfentanil, and etomidate preserves MEPs better than volatile agents, nitrous oxide, or propofol ( Kalkman et al., 1994 ; Ubags et al., 1997 ; Sihle-Wissel et al., 2000 ).

Transcranial Doppler

Transcranial Doppler measurement of middle cerebral artery blood velocity and, by extension, estimation of CBF is a technique that remains experimental in its clinical neurosurgical application. It has been clinically used by one investigator in adult patients during clipping of cerebral aneurysms, but it has no routine use in pediatric neurosurgery at the present time. It does have utility, however, as an experimental tool for the determination of the effects of anesthetic agents and physiologic changes on CBF and has emerged in the last decade as the best noninvasive method for the estimation of CBF in humans.

Transcranial Cerebral Oximetry

In the future, near infrared spectroscopy (NIRS), which measures cerebral oxygenation through an adhesive probe attached to the forehead, may become customary for neurosurgical procedures. It is a very sensitive monitor for cerebrovascular changes and assesses oxygen delivery and extraction, as well as CBV and CBF ( Soriano et al., 2002 ). At the present time, this monitoring technique is experimental.


Infiltration of the scalp with lidocaine or bupivacaine by the neurosurgeon is a simple and very useful technique for blunting the hemodynamic response to incision and craniotomy and for reducing the anesthetic requirement ( Hillman et al., 1987 ).

Although nitrous oxide is a mild stimulant of cerebral metabolism and blood flow ( Theye and Michenfelder, 1968 ; Dahlgren et al., 1981 ; Fitzpatrick and Gilboe, 1982 ), its effects on cerebral hemodynamics are overshadowed by previous administration of thiobarbiturates and concomitant hyperventilation ( Phirman and Shapiro, 1977 ). Nitrous oxide may be used safely in patients with intracranial hypertension. Nitrous oxide has two disadvantages during neurosurgery: its effect on VAE and its effect on pneumocephalus—both are discussed in previous sections.

The effects of halogenated anesthetic agents on cerebral metabolism, blood flow, and ICP were discussed earlier. Halogenated inhalation anesthetics increase CBF, CBV, and ICP in a dose-dependent fashion, an effect that may be blunted or avoided by hyperventilation of the patient during their administration. In cases of preoperative intracranial hypertension or intracranial mass lesions, however, halogenated agents generally should be withheld until the dura is open unless ICP is monitored, so that the cerebral hemodynamic response may be observed.

In this circumstance, most neuroanesthesiologists would choose an anesthetic technique combining propofol or a barbiturate, nitrous oxide, opioids, and a nondepolarizing muscle relaxant. Such a “balanced” anesthesia technique is also associated with more rapid awakening after prolonged procedures than is typically seen with halogenated anesthetics, allowing earlier neurologic assessment of the neurosurgical patient.

Thiobarbiturates or propofol is usually administered intermittently during balanced anesthesia, a practice that reduces cerebral metabolism and blood flow and decreases ICP. Fentanyl is a commonly chosen opioid for neurosurgery because no histamine release is associated with this drug. Fentanyl analgesia requires 5 to 10 mcg/kg and is maintained with 1 to 3 mcg/kg per hr, either by intermittent bolus administration or by continuous infusion. For lengthy neurosurgical procedures, remifentanil, alfentanil, and sufentanil lead to more rapid awakening because of the more rapid metabolism and smaller volumes of distribution of the drugs ( Shafer and Varvel, 1991 ; Youngs and Shafer, 1994 ). Of these short-acting opioids, remifentanil confers no residual analgesia whatever within a few minutes of the termination of its infusion; it is imperative that the anesthesiologist administer a long-acting opioid analgesic when, or soon after, discontinuing a remifentanil infusion to avoid the sudden onset of severe surgical pain, hypertension, etc.

Nondepolarizing muscle relaxants that do not release histamine (e.g., rocuronium, pancuronium, and vecuronium) are preferred for neurosurgery.

Adjunctive measures to control intracranial hypertension during neurosurgery include hyperventilation to a Paco2 of approximately 30 mm Hg, administration of mannitol or diuretics to reduce brain volume, and control of arterial blood pressure. Cerebral autoregulation is usually impaired in the presence of cerebral pathology and is blunted further by inhalation anesthetics. CBF, therefore, may fluctuate with alteration of arterial blood pressure. Maintenance of normotension is important in preserving CBF and preventing increases in ICP.

Antihypertensives that act by direct vasodilation (hydralazine, adenosine, sodium nitroprusside, nitroglycerin, and nitric oxide) are potent cerebral vasodilators and increase both CBF and ICP. Preferable are indirect-acting antihypertensives (fenoldopam, propranolol, esmolol, and labetalol), which can control hypertension in neurosurgical patients without increasing ICP.

Controlled, or deliberate, hypotension is used in procedures associated with the risk of significant intraoperative hemorrhage, such as craniofacial surgery, and cerebrovascular surgery of arterial aneurysms and arteriovenous malformations. Deliberate hypotension is contraindicated in the presence of intracranial hypertension or diminished intracranial compliance. Because the lower limit of cerebral autoregulation in children is not known, the lower limit of acceptable blood pressure in children during deliberate hypotension is also not known. In the absence of data to guide clinical decisions, the adopted practice is to decrease MAP by no more than 30% to 40% below the “awake” baseline but no less than 40 mm Hg for infants younger than 6 months and 50 mm Hg for older children. Deliberate hypotension and its complications are discussed in greater detail in Chapter 12 , Blood Conservation in Infants and Children.


The maintenance of normal body temperature in children, once very challenging for pediatric anesthesiologists, has become second nature with the use of forced hot air warming mattresses and blankets, in concert with warming lights, heated humidification of inspired gases, and adjustment of ambient room temperature (indeed, gone are the days of working in rooms warmed to near-tropical degrees to maintain the temperatures of small infants undergoing lengthy procedures). The maintenance of normothermia and the avoidance of both hypothermia and hyperthermia are discussed elsewhere in this text (see Chapter 5 , Thermoregulation: Physiology and Perioperative Disturbances).

In neurosurgical procedures, the induction of mild to moderate hypothermia (32° to 35°C) is used for cerebral protection during procedures in which there is a risk of intraoperative cerebral ischemia. For every 1°C below normal body temperature, the cerebral metabolic rate of oxygen consumption decreases 7% ( Rosomoff and Holaday, 1954 ), which increases the ability of the cerebral tissue to tolerate periods of ischemia, and therefore is thought to improve surgical outcome. Whereas this has been experimentally demonstrated in animal models, it has not been so demonstrated in the human neurosurgical patient, although mild hypothermia has been shown to improve outcome in head-injured patients ( Gal et al., 2002 ).

Deliberate Hypothermia

Biagas and Gaeta (1998) proposed that hypothermia can attenuate injury due to inflammation, excitotoxic amino acids (glutamate), nitric oxide, and free radicals. Marion (2002) showed a significant decrease in CSF glutamate and interleukin-1β levels in patients with a depressed GCS score maintained between 32° and 33°C for 12 hours compared with those maintained with a similar GCS score at a normothermic state. In addition, studies have shown that hypothermia is more effective than barbiturate therapy in lowering ICP in pediatric patients refractory to conventional therapy ( Hayashi, 2000 ). Hypothermia may help to control elevated ICP in children, but it fails to arrest progression of infarction, such as that from an acute, subdural hematoma in a child ( Inamasu et al., 2002 ).

Deliberate hypothermia may be induced in the child by not using the usual warming devices, leaving the child undraped in a cold operating room after induction of anesthesia and while inserting vascular catheters, positioning, and draping, and by using a cooling mattress, until the target temperature is achieved. Endovascular cooling of the brain has been described and offers superior cerebral cooling and rewarming compared with whole body cooling ( Steinberg et al., 2004 ). Rewarming is achieved by using forced hot air warming mattresses and blankets and by raising the ambient room temperature. It is of great importance to achieve adequate rewarming to greater than 36°C before emergence and awakening to avoid the vasoconstriction, hypertension, and shivering that would occur in an emerging hypothermic child, followed by the hemodynamic instability that would be seen with the vasodilatation that accompanies rewarming later during the recovery period. Maintenance of general anesthesia is best continued until body temperature is sufficient for the child to tolerate emergence and awakening.

Complications of hypothermia and its reversal must be understood by the anesthesiologist; these include coagulopathy, immunodeficiency, and hyperglycemia. Children develop a coagulopathy during hypothermia, until normothermia is restored. Severe immunodeficiency and secondary infections can arise. For example, Lo and others (2002) cited a case of pancytopenia postoperatively in an 8-year-old child who had undergone resection of a craniopharyngioma under hypothermia. This condition resolved on further rewarming in the intensive care unit.

Moreover, it is critical to lower a high glucose value because glucose quickly penetrates the blood-brain barrier and increases pyruvate and lactate by inhibiting the tricarboxylic acid (TCA) cycle metabolism ( Hayashi, 2000 ). In addition, children, like adults, are more difficult to awaken from general anesthesia if they are not normothermic and neuromuscular blockade is more difficult to reverse under hypothermic conditions. Finally, residual hypothermia on emergence causes shivering, and this increases basal, metabolic oxygen consumption dramatically. If this occurs, small doses of intravenous meperidine (0.1 to 0.2 mg/kg) may be effective in inhibiting shivering.


The management of fluids in the neurosurgical patient is a clinical challenge because blood loss is difficult to measure. Inadequate fluid replacement leads to cardiovascular instability, and overhydration with hypo-osmotic solutions increases cerebral edema. Furthermore, diuretics used to reduce brain bulk cause intravascular volume shifts with electrolyte disturbances.

Blood-Brain Barrier Effects

Fluid and solute flux across the cerebral circulation is restricted by the blood-brain barrier. The blood-brain barrier is formed by cerebral capillary endothelial cells connected by a continuous, tight intercellular junction. Because of this tight junction, the cerebral capillary markedly restricts passage of most polar, hydrophilic molecules, except those with a specific carrier-mediated transport system (for example, glucose, essential amino acids). Small, polar molecules such as water move rapidly across the blood-brain barrier, with a 3-minute half-time for equilibration ( Bering, 1952 ). In contrast, ions such as sodium move more slowly across the blood-brain barrier (half-time for equilibration is 2 to 4 hours) ( Bakay, 1960 ); larger hydrophilic molecules such as albumin and mannitol are excluded from the brain by an intact blood-brain barrier.

Water movement across the blood-brain barrier depends on the osmotic gradient between plasma and brain. Water is removed from the brain's interstitial space when plasma osmotic pressure is increased, as with administration of mannitol or hypertonic saline solution. Water may move into the relatively hyperosmolar brain with rapid correction of hyperosmolar states (uremia, hyperglycemia) and with hyponatremia. Intravenous administration of free water causes a marked and prolonged increase in CSF pressure ( Weed and McKibben, 1916 ). Administration of 5% dextrose solution is similar to giving free water because the uptake and metabolism of glucose are so rapid.

The blood-brain barrier becomes disrupted with many types of brain injury, including head trauma, subarachnoid hemorrhage, stroke, brain tumors, hypertension, hypercapnia, status epilepticus, and sodium nitroprusside administration ( Pollay and Roberts, 1980 ). Movement of water across the blood-brain barrier then becomes a function more of hydrostatic pressure gradients than of osmotic gradients. The permeability of sodium, albumin, and mannitol is increased markedly. Fortunately, in most types of brain injury, large areas of intact blood-brain barrier remain.

Fluid Therapy Administration

The traditional approach to fluid management in neurosurgical patients is to restrict fluid intake in an attempt to prevent cerebral edema and subsequent increases in ICP ( Shenkin et al., 1976 ). Like all postsurgical patients, neurosurgical patients transiently retain water and sodium postoperatively because of increased aldosterone and antidiuretic hormone (ADH) secretion, which may lead to hyponatremia and intravascular volume overload. Hypervolemia may cause hypertension, and a rebound increase in ICP may occur after administration of mannitol. Although hypervolemia should be avoided, there is little evidence to support the benefits of fluid restriction in neurosurgical patients. Even severe water restriction is only modestly effective in reducing brain water content in animals ( Jelsma and McQueen, 1967 ). Fluid restriction has several detrimental effects, including hypovolemia and hypotension, especially in response to anesthetic agents and positive-pressure ventilation, inadequate renal perfusion, electrolyte and acid-base disturbances, lability while using vasodilators for deliberate hypotension, hypoxemia, and undesired reductions in CBF. Reductions in cardiac output increase pulmonary shunting and may cause hypoxemia in the presence of regional lung disease (e.g., atelectasis, pneumonia, pulmonary contusion) ( Cheney and Colley, 1980 ). Decreases in Pao2 have been observed in neurosurgical patients who are hypovolemic from osmotic diuresis and excessive fluid restriction. Clinical experience has shown that Pao2 increases dramatically with volume administration in these patients.

Colloid-containing solutions often have been used during neurosurgery because albumin is excluded from the extracellular fluid of the brain in the presence of an intact blood-brain barrier. Many clinicians believe that the administration of large amounts of crystalloid itself can cause cerebral edema. This belief was supported by a study that noted an increase in brain water content and ICP when lactated Ringer's solution, but not hydroxyethyl starch, was used for fluid replacement during isovolemic hemodilution of normal rabbits ( Todd et al., 1984 ). These results later were ascribed to differences in osmolality, rather than the colloid osmotic pressures, of the two solutions ( Zornow et al., 1987 ). Brain water content in normal rabbits does not differ when volume is replaced with isotonic lactated Ringer's solution or hydroxyethyl starch of the same osmolality. Moreover, in a study involving patients 1 to 38 months, Paul and others showed a larger decrease in hemoglobin levels postoperatively after 20 mL/kg of 6% hydroxyethyl starch (a synthetic colloid) compared with 20 mL/kg of lactated Ringer's solution (a crystalloid) ( Paul et al., 2003 ). Such studies demonstrate the superiority of colloids for plasma expansion in children but do not analyze the effects on the brain.

In contrast, brain water is increased when intravascular volume is replaced with a hypo-osmotic colloid solution. Osmolality, rather than colloid osmotic pressure, determines water movement across the intact blood-brain barrier. In the presence of a disrupted blood-brain barrier induced experimentally by a cortical freeze injury, however, albumin therapy is associated with less cerebral edema than is iso-osmotic crystalloid therapy ( Albright et al., 1984 ).

In summary, it is unclear whether the best solution for use in patients with brain injury is isotonic colloid or crystalloid. Proponents of colloid emphasize that with most intracranial lesions, large areas of the brain still have an intact blood-brain barrier. In addition, the effect on brain water of the greater volume of crystalloid needed to restore intravascular volume is unknown. Proponents of crystalloid hypothesize that colloid might increase cerebral edema by holding additional fluid in the damaged area when it crosses the disrupted blood-brain barrier.

Because of the uncertainties just described, there is no clear advantage to using colloid or isotonic crystalloid for routine fluid therapy in neurosurgical patients. Hypertonic saline solutions may have some use in the resuscitation of head-injured patients; their use, however, is still experimental in the neuro-surgical setting. Hypotonic solutions are not used during neurosurgery because they increase brain edema: 5% dextrose, 5% dextrose in 0.2% saline solution, and 5% dextrose in 0.45% saline solution are all functionally hypotonic despite the tonicity out of the bottle, because the glucose is rapidly metabolized and is also transported into the brain, with the net effect of the hypotonic electrolyte solution remaining; administration of these solutions increases brain water content.

Furthermore, several studies in animals have demonstrated that dextrose administration, with or without hyperglycemia, augments brain damage in global ( Lanier et al., 1987 ) and regional ( Pulsinelli et al., 1982 ) cerebral ischemia. Increased blood glucose also appears to enhance postischemic neurologic damage in humans ( Pulsinelli et al., 1983 ). Dextrose administration increases brain glucose and anaerobic metabolism of glucose in the presence of ischemia increases lactic acid, which causes neurologic damage. Because decreased cerebral perfusion and brain ischemia may occur with retraction in any intracranial procedure, routine glucose administration is not used in patients undergoing craniotomy who are not at risk for hypoglycemia.

Acceptable isotonic fluids are lactated Ringer's solution, normal saline, or any of the commercially available multiple-electrolyte balanced salt solutions that are designed to emulate the plasma water composition (e.g., Normosol, Abbott Pharmaceuticals). Normal saline is probably the most common crystalloid administered during pediatric craniotomies as it is slightly hyperosmolar (308 mOsm/kg) compared with serum osmolarity (285 to 290 mOsm/L) and therefore helps to prevent cerebral edema. However, anesthesiologists must use caution; large quantities of normal saline produce a hyperchloremic metabolic acidosis and hypernatremia in children ( Constable, 2003 ). Lactated Ringer's (273 mOsm/L), on the other hand, is slightly hypo-osmolar and large quantities of intravenous infusion can increase cerebral edema formation.



Mannitol decreases brain water content by about 1% to 2% by increasing plasma osmolality, which creates an osmotic gradient across the intact blood-brain barrier. The amount of water that can be withdrawn from the brain depends on the magnitude of the osmotic gradient and the integrity of the blood-brain barrier. Mannitol is less effective with larger lesions. With a damaged blood-brain barrier, the concentration gradient moves mannitol into the brain, and this may account for the rebound increase in ICP occasionally seen after mannitol administration.

Mannitol causes a triphasic hemodynamic response. Transient (1 to 2 minutes) hypotension may occur after rapid administration of mannitol because of vasodilation ( Coté et al., 1979 ). Subsequently, blood volume, cardiac index, and pulmonary capillary wedge pressure increase, reaching a maximum shortly after the termination of infusion ( Rudehill et al., 1983 ). ICP may increase transiently because of increases in CBV and CBF. By 30 minutes after mannitol, blood volume returns to normal after diuresis, and pulmonary capillary wedge pressure and cardiac index decrease to less than normal levels.

Mannitol decreases blood viscosity and red blood cell rigidity, which may enhance perfusion of the brain's microcirculation. Mannitol transiently reduces hematocrit, increases serum osmolality, and causes hyponatremia, hypochloremia, acidosis resulting from bicarbonate ion dilution, and hyperkalemia after large (2 g/kg) doses. Prolonged and marked hyperosmolality with hyponatremia can occur in patients with acute and chronic renal failure.

Mannitol is the most efficient therapy for normalization of elevated cerebral extraction of oxygen representing oligemic ischemia in the presence of an elevated ICP. Mannitol is usually given in doses from 0.25 to 1.0 g/kg, which raises serum osmolality by 10 to 20 mOsm/kg ( Soriano et al., 1996 ). Lower doses reduce ICP acutely and cause fewer electrolyte abnormalities but must be given more frequently (Marshall et al., 1978 ). The benefits and disadvantages of speed of administration and dose need to be weighed carefully in each patient.


Furosemide has been reported to lower ICP and brain water content when used alone in large (1 mg/kg) doses ( Cottrell et al., 1977 ) or combined with mannitol in smaller doses ( Pollay et al., 1983 ;Roberts et al., 1987 ). In contrast to mannitol, however, furosemide has been shown by some studies in normal and damaged brains to have little effect on ICP ( Roberts et al., 1987 ). Furosemide may be preferred to mannitol in patients with cardiac or renal disease because it does not increase blood volume or ICP, nor does it cause electrolyte abnormalities as severe as those caused by mannitol. In large doses, furosemide reduces CSF formation and may reduce water and ion penetration across the blood-brain barrier ( Pollay et al., 1983 ). Furosemide prolongs the effectiveness of mannitol by sustaining the induced increase in serum osmolality. Reductions in ICP and brain volume are consistently greater and last longer with mannitol plus furosemide than with either agent alone. However, hyponatremia, hypokalemia, hypochloremia, hyperosmolality, and a significantly greater rate of water and electrolyte excretion occur with the combination of diuretics.

Copyright © 2008 Elsevier Inc. All rights reserved. -

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

Copyright © 2005 Mosby, An Imprint of Elsevier




Myelodysplasia is a congenital failure of the neural tube to close, a process that normally occurs by 28 days—gestation. In its most common form, fusion fails in the middle or caudal neural groove, resulting in thoracic or lumbosacral meningomyelocele. When the site of failure is more cephalad, encephaloceles result. Anencephaly represents a defect in anterior closure of the neural groove. Figure 18-5 illustrates a newborn infant with a lumbosacral meningomyelocele with a partially epithelialized neural sac and exposed neuroplaque. With a meningomyelocele, the defective neural tissue is in open communication with the environment. Neurologic function very frequently is severely impaired distal to the defect. An encephalocele is a similar phenomenon usually involving the occipital skull and upper cervical spine ( Fig. 18-6 ).


FIGURE 18-5  A, Newborn infant with a lumbosacral meningomyelocele. Note the partial epithelialization of the neural tissue, which is exposed to air centrally. Neurologic function was severely impaired distal to the defect. B, Schematic of meningomyelocele.  (B from Carter S, Gold AP: Nervous system. In Rudolph AM, Barnett HL, Einhorn AH, editors: Pediatrics, 16th ed. Norwalk, CT, 1977, Appleton-Century-Crofts. Copyright © 1977 The McGraw-Hill Companies, Inc.)





FIGURE 18-6  A newborn infant with an occipital encephalocele. Neuro-logic function was intact. The anesthesiologist's primary challenge is to control and intubate the airway in the lateral position to avoid traumatizing the lesion.



After closure of the neural tube, mesodermal and ectodermal structures complete the formation of the spinal column, skeletal muscles, and skin of the back. A defect in this process may result in herniation of dural elements posteriorly during morphogenesis, producing a meningocele ( Fig. 18-7 ), which, unlike a meningomyelocele, contains no neural tissue. Although neurologic function is normal in children with meningoceles, the spinal cord is often tethered caudally by the sacral nerve roots. The tethered cord results in orthopedic or urologic symptoms in later childhood if it is not surgically corrected.


FIGURE 18-7  A, Newborn infant with a lumbar meningocele. No neural elements were within the sac, and neurologic function was normal distal to the lesion. B, Schematic of meningocele.  (B from Carter S, Gold AP: Nervous system. In Rudolph AM, Barnett HL, Einhorn AH, editors: Pediatrics, 16th ed. Norwalk, CT, 1977, Appleton-Century-Crofts. Copyright © 1977 The McGraw-Hill Companies, Inc.)




Congenital lesions of the central nervous system, unlike other types of congenital anomalies such as omphaloceles, are generally not associated with other anomalies or congenital heart disease; therefore, routine preoperative screening need not include a cardiologic evaluation. Most children with meningomyelocele also have an associated Arnold-Chiari malformation (see discussion later in this chapter), and hydrocephalus ultimately develops, necessitating a CSF shunt. The Chiari malformationsinvolve the hindbrain and are classified into four types. Three of the four types involve hindbrain herniation, and the fourth type consists of cerebellar hypoplasia or aplasia ( Box 18-1 ).

BOX 18-1 

The Chiari Malformations

Chiari Type I

Tonsillar herniation >5 mm below the plane of the foramen magnum

No associated brainstem herniation or supratentorial anomalies

Low frequency of hydrocephalus

Chiari Type II

Caudal herniation of the vermis, brainstem, and fourth ventricle

Associated with myelomeningocele and multiple brain anomalies

High frequency of hydrocephalus and syringohydromyelia

Chiari Type III

Occipital encephalocele containing dysmorphic cerebellar and brainstem tissue

Chiari Type IV

Hypoplasia or aplasia of the cerebellum

Anesthetic Considerations

Meningoceles and myelomeningoceles usually are repaired within the first day of life to minimize bacterial contamination of the exposed spinal cord and subsequent sepsis, which is the most common cause of death in this population. The anesthetic considerations relate to the concerns of anesthesia for newborns (described in Chapter 16 , Anesthesia for Neonates and Premature Infants) and management of the airway.

As Figures 18-6 and 18-7 [6] [7] illustrate, intubation of the trachea may pose a challenge to the anesthesiologist. Endotracheal intubation may be performed with the infant in the left lateral decubitus position, with the head held in the midline by an assistant, or in the supine position, with the weight supported on the pelvis and portions of the spine not involved with the defect. An assistant must be available to ensure that no physical trauma to the neuroplaque occurs. In newborns in which a difficult airway is suspected, an awake intubation may be performed after atropine premedication (20 mcg/kg, minimum dose 0.1 mg) and preoxygenation. Otherwise, the induction may proceed by mask inhalation or by intravenous administration of propofol or thiopental and relaxant. If pancuronium is not the muscle relaxant of choice, atropine should precede intubation of the trachea. Future advancements in the management of myelomeningoceles lead to early correction through fetal neurosurgery, which involves a general anesthetic for a cesarean section for the mother ( Sutton et al., 2001 ) (also see Chapter 15 , Anesthesia for Fetal Surgery).

Blood loss during meningomyelocele repair usually is not excessive, averaging about 30 mL, or 10% of blood volume. If the lesion is extensive, however, the surgeon may find it necessary to undermine large areas of skin and fascia or to make fasciotomy incisions at the flanks to achieve primary closure of the defect. These practices will necessitate skin grafting over the fasciotomies and increase blood loss significantly.

Conservation of body heat is important for infants with a myelomeningocele, particularly because autonomic control below the level of the defect is abnormal. The operating room should be warmed to 27°C (80°F) before surgery and until the infant is draped. Radiant heat lamps are used during positioning and skin preparation until the infant is draped and again at the end of surgery, a forced hot air warmer is used to maintain body temperature, and humidification of inspired gases further prevents heat loss and minimizes pulmonary complications by humidification of the airway.

For poorly understood reasons, children with myelodysplasia have a significantly increased prevalence of allergy to latex in later childhood and adulthood ( Kurup and Fink, 2001 ; Hepner and Castells, 2003 ). Thought at one time to be the consequence of frequent catheterization of neurogenic bladders by latex catheters or multiple previous surgical exposure to latex, it is now clear that latex allergy develops even in children without a history of multiple catheterizations or surgery ( Hochleitner et al., 2001 ), although multiple previous surgeries have been identified as one risk factor, as well as a history of atopy, food allergies, and nonwhite race ( Kelly et al., 1994 ). To prevent early sensitization to latex, newborns undergoing repair of myelomeningoceles and meningoceles should be treated as if they were latex allergic, using a latex-free surgical environment ( Nieto et al., 2002 ) (see Chapter 32 , Systemic Disorders).

Neonatal Hydrocephalus

Hydrocephalus itself is not a disease but rather is the consequence of many disease processes. In the newborn period, the two most common causes are anatomic anomalies associated with myelodysplasia and prematurity.

The Arnold-Chiari malformation is a collection of anatomic abnormalities that includes displacement of the cerebellar vermis through the foramen magnum, elongation of the brainstem and fourth ventricle, and noncommunicating hydrocephalus. Most children with meningomyelocele have an associated Arnold-Chiari malformation and ultimately develop hydrocephalus, usually in the first month of life.

Hydrocephalus frequently develops in premature infants who have had an intraventricular hemorrhage; this results in noncommunicating hydrocephalus from scarring or deposition of fibrinous deposits around the aqueduct of Sylvius. More often, hydrocephalus is the consequence of diminished CSF resorption from the arachnoid granulations from scar or fibrin deposition, and it is referred to as communicating hydrocephalus. Neonatal meningitis produces hydrocephalus by a similar mechanism.

Anesthetic management depends on whether ICP is elevated. Slow development of hydrocephalus in newborns is accompanied by increasing skull diameter to accommodate the increase in CSF volume, and, thus, normal ICP is maintained. Rapidly developing hydrocephalus outpaces gradual skull growth and results in rapid elevation of ICP and cerebral herniation unless it is treated. A tense anterior fontanel, irritability, vomiting, and ophthalmoplegia suggest intracranial hypertension.

Surgical management involves the shunting of CSF from a lateral ventricle to either the peritoneum (ventriculoperitoneal shunt) or sometimes into the right atrium (ventriculoatrial shunt). In the absence of abdominal pathology, the ventriculoperitoneal shunt is preferred. This allows the neurosurgeon to insert a redundant length of shunt tubing, which will accommodate the child's growth.

Anesthetic Considerations

Anesthetic management includes the usual considerations and depends on the presence or absence of intracranial hypertension. In the premature infant undergoing ventriculoperitoneal shunt placement, the inspired oxygen fraction should be limited to maintain Pao2 and arterial oxygen tensions at 70 mm Hg and 95% to 97%, respectively. These levels minimize the risk of retinopathy of the premature infant. Ventilation should be controlled, with careful attention to inflation pressures to minimize the risk of pulmonary trauma, and measures should be taken to conserve body heat. In addition, serious dysrhythmias, especially bradycardias, may occur during ventriculostomies in children with obstructive hydrocephalus ( El-Dawlatly et al., 1999 ). Besides using atropine to treat such bradyarrhythmias, simply alerting the surgeon of the bradycardia and pausing may resolve the problem without medications. The postoperative care of the premature or ex-premature infant who is less than 50 weeks' postconceptual age requires the use of a cardiorespiratory monitor or oximeter to detect postoperative apnea. Even if ICP is well controlled and a reservoir is placed, infants with a history of hydrocephalus may develop postoperative apneic spells, retractions, and/or vocal chord paralysis ( Nishino et al., 1998 ).

Depressed Skull Fractures

Depressed skull fractures occur in the newborn when the infant's head descends through a narrow birth canal and is fractured by the mother's ischium; they usually are not the result of obstetric application of forceps ( Bruce, 1980 ). A depressed skull fracture is a greenstick fracture, frequently resembling an indented table tennis or Ping-Pong ball, and is rarely associated with neurologic injury. Because of early concern regarding the administration of anesthetics to newborns, at one time such fractures were repaired using only local anesthesia, but modern anesthetic techniques make general anesthesia safe in newborns. Because the child is neurologically normal, no anesthetic considerations exist apart from those pertaining to all newborns. The surgeon introduces a periosteal elevator through a small incision in the coronal suture, advancing the instrument until it lies under the indentation in the skull. The elevator is then pushed up, and the scalp is sutured. More extensive or comminuted depressed skull fractures occasionally require a more extensive scalp flap and the use of titanium fixators to hold together the bone fragments.



Craniosynostosis, or craniostenosis, represents premature intrauterine fusion of one or more cranial sutures and causes an abnormal skull shape. Most often, it involves only the sagittal suture and results in a deformity that is primarily cosmetic. Development, intellect, and ICP are normal. However, if left uncorrected, many children experience cortex-associated retardation of intelligence. Occasionally, more than one suture is stenosed. Without treatment, this may result in intracranial hypertension as the brain grows and the skull does not. Multiple suture craniosynostosis is most commonly seen in association with craniofacial anomalies, particularly Apert's and Crouzon's syndromes, which are also associated with hypoplasia of the orbits and mid portion of the face.

Children with single-suture craniosynostosis are usually healthy. Surgery is most often performed between 2 and 6 months of life, a period that corresponds to the physiologic nadir of hemoglobin. The acceptable blood loss is very small, and blood transfusion is frequently needed. At least one intravenous catheter for volume infusion is necessary. Arterial cannulation for continuous monitoring is not routinely required for single-suture craniectomies but should be used for multiple-suture procedures.

Several reports have appeared describing VAE during craniosynostosis repair in infants (Harris et al., 1986, 1987 [118] [119]). Most recently, using a precordial Doppler probe, Faberowski and coworkers demonstrated that of 23 patients undergoing craniectomies for craniosynostosis, 19 demonstrated 64 episodes of VAE without cardiovascular collapse ( Faberowski et al., 2000 ). The incidence of clinically important air embolism during craniectomies is still undefined but is probably small. It may nevertheless be prudent to monitor patients with a precordial Doppler ultrasound device and to attempt central venous catheter placement to allow aspiration of venous air from the heart.

The surgery for sagittal synostosis is extradural and entails craniectomies on both sides of the sagittal suture. Blood loss begins with scalp incision early in surgery and is exceedingly difficult to quantify; thus, it is important for the anesthesiologist to begin transfusion early, before hypovolemia occurs. If an arterial or a central venous cannula has been placed, serial measurements of hematocrit guide transfusion and fluid therapy. A new surgical technique for sagittal suture craniosynostosis repair (spring-mediated cranial expansion) has been described that is associated with significantly less blood loss (Ririe et al., 2003 ). The place of this technique in the surgical armamentarium has yet to be defined. It is crucial for the anesthesiologist to anticipate which technique is used, as the former causes considerably more blood loss and, consequently, the patient is more likely to need a blood transfusion ( Ririe et al., 2003 ).

Multiple-suture craniectomies are often performed in association with reconstruction of the midportion of the face and orbital advancement (see Chapter 20 , Anesthesia for Pediatric Plastic Surgery). The associated blood loss is large, averaging 50% to 150% of total blood volume ( Davies and Munro, 1975 ). This consideration necessitates the placement of two volume infusion catheters before the surgical incision, or one peripheral volume infusion catheter and one central venous catheter. An arterial cannula is necessary to monitor arterial pressure and for serial blood gas and hematocrit determinations. Central venous pressure monitoring aids fluid therapy and allows aspiration of venous air in the unusual event of air embolism.

The surgeon often ligates the endotracheal tube with silk suture or wire to the alveolar ridge to prevent its being dislodged under the drapes during surgery. If a nasotracheal tube is inserted in anticipation of prolonged postoperative ventilation in the intensive care unit, it should be sutured in place to the nares with silk suture. To prevent a corneal injury, the eyes should be lubricated with an antibiotic ointment or wetting agent and may be sutured closed with a tarsorrhaphy stitch or covered with a corneal shield.

Anesthetic Considerations

Either inhalation or balanced anesthesia is appropriate for craniofacial repairs. Intravenous anesthesia results in less cardiovascular depression and better postoperative analgesia. Moderate deliberate hypotension is usually used to reduce intraoperative bleeding (see Chapter 12 , Blood Conservation).

Complicated craniofacial surgery involves intracranial surgery; a reduction of brain volume aids the surgical technique. This may be accomplished with hyperventilation, osmotic dehydration of the brain with mannitol, diuretic therapy with furosemide, or continuous CSF drainage with a lumbar, subarachnoid malleable needle or an epidural catheter placed through a Tuohy needle into the subarachnoid space.

The postoperative care depends on the extent of surgery and subsequent facial and airway edema. In surgery below the orbital ridge, extensive facial edema is common, and the endotracheal tube can be left in place for 48 hours after surgery until resolution of facial and airway edema. To ensure the security of the endotracheal tube, the child can be heavily sedated during this time to maintain neuromuscular blockade and ventilate the lungs mechanically until edema resolves and permits extubation of the trachea.

Vascular Malformations

Arteriovenous Malformations

Arteriovenous malformations (AVMs) are congenital nests of abnormal blood vessels. They may occur anywhere within the body, but when they occur in the brain, they present in four different manners: in the newborn as congestive heart failure (CHF), and in the older child as seizures, hydrocephalus, or, most commonly and tragically, intracranial hemorrhage ( Millar et al., 1994 ; Newfield and Hamid, 2001).

AVMs, unless they are very large, are usually occult in newborns. According to Millar and others, only 18% of AVMs become symptomatic before the age of 15 and symptoms vary from hemorrhage (50%) to seizures and hydrocephalus (36%) in infants and children, whereas newborns may present with high-output, congestive heart failure (18%) ( Millar et al., 1994 ). AVMs are associated with a mortality rate of greater than 90% when CHF is present, and the anesthesiologist should assess signs and symptoms of CHF preoperatively. Although newborn malformations often are not amenable to surgical intervention, the anesthesiologist will nonetheless be called on to anesthetize the child for cerebral diagnostic angiography, embolization, microsurgery, or stereotactic radiosurgery. In addition, the role of endovascular therapy for intracranial aneurysms is rapidly evolving and, in the near future, thrombolytic intra-arterial therapy in children will become more common.

Anesthetic Considerations

Anesthetic management should minimize cardiovascular depression and wide fluctuations in arterial blood pressure to decrease the risk of spontaneous intracranial hemorrhage. Intracranial hypertension usually is not present, but many lesions obstruct CSF pathways and cause hydrocephalus. Local anesthesia, with an infusion of propofol, usually suffices for angiography. More invasive procedures, such as attempts at embolization, require a deeper general anesthesia, endotracheal intubation, and neuromuscular blockade to ensure immobility. Invasive cardiovascular monitoring and controlled hypotension are requirements. Arteriovenous malformations may pose a risk of VAE if a venous structure is opened by the surgeon, and therefore devices to monitor for an embolism should be in place.

As discussed previously, it is often useful to establish preanesthetic β-adrenergic blockade before inducing anesthesia. Anesthetic induction often uses a technique of graded stimuli. An intravenous induction with thiopental and/or propofol and opioid is followed by the introduction of nitrous oxide with or without a potent inhaled anesthetic by mask. Then stimuli of increasing intensity are introduced, and the cardiovascular response to each is judged. If a stimulus produces an increase in blood pressure or heart rate, the anesthetic depth is increased. A potent and rapidly acting hypotensive agent, such as sodium nitroprusside, esmolol, or fenoldopam, should be at hand for the rapid treatment of resultant hypertension. After several milder stimuli, the trachea finally is intubated after anesthesia has been augmented with a further dose of propofol and intravenous lidocaine. A suitable sequence of graded stimuli might be vascular cannulas, oral airway, bladder catheter, laryngoscopy with lidocaine spray, intubation, Mayfield head tongs, and, finally, surgical incision. One technique combines opioid-based anesthetics with neuromuscular blockade and low-dose inhalation anesthesia to control blood pressure.

High inspired concentrations of halogenated anesthetic have an unpredictable effect on arterial blood pressure in the newborn, particularly in the presence of CHF. An alternative and very satisfactory technique is total intravenous anesthesia (TIVA) using an infusion of propofol and remifentanil, with a neuromuscular blocking agent to provide absolute immobility. This technique allows the most rapid emergence with the ability to perform an early postembolization neurologic examination.

Moyamoya Syndrome

During the past decade, major, tertiary centers are seeing an increase in the number of cases of moyamoya disease in adults and children alike, and the etiology is still unknown. Moyamoya is an idiopathic, chronic vaso-occlusive disorder of the distal, internal carotid arteries and circle of Willis that presents as transient ischemic attacks or recurrent strokes in children. Moyamoya means “hazy puff of smoke” in Japanese and refers to the angiographic appearance of the abnormal network of vessels that develop at the base of the brain and basal ganglia to supply a collateral route of blood flow ( Suzuki and Takaku, 1969 ). An increase in elastin gene expression has been identified, suggesting the importance of overproduction of elastin in the pathophysiology ( Yamamoto et al., 1997 ).

Moyamoya syndrome is not a distinct entity but rather the syndromic consequence of a number of individual disease processes, and hence may be seen with previous cranial radiation, Alagille syndrome and other causes of hypercholesterolemia, neurofibromatosis, and trisomy 21, to name some of the more common associations ( Jacob and Kausalya, 1990 ; Horn et al., 2004 ; Kamath et al., 2004 ; Kim et al., 2004 ; Spetzler, 2004 ).

Interestingly, patients with Alagille syndrome develop vascular lesions such as moyamoya, and anesthesiologists need to be aware of the comorbidities associated with this syndrome before formulating an anesthetic plan for neurosurgical procedures. Alagille syndrome is classically composed of five characteristics: typical peculiar facies, chronic cholestasis, posterior embryotoxon, butterfly-like vertebral-arch defects, and cardiovascular malformations (most commonly peripheral or branch pulmonary artery stenosis) ( Alagille, 1996 ). The 20-year mortality is 75%, and factors contributing to mortality are complex congenital heart disease (15%), intracranial bleeding (25%), and hepatic disease or hepatic transplantation ( Emerick et al., 1999 ). Hyperlipidemia is common in patients with Alagille syndrome. However, hyperlipidemia alone is not adequate to explain the severe vascular abnormalities and unidentified genetic factors that predispose Alagille patients to develop vasculopathy ( Woolfenden et al., 1999 ). It is important for anesthesiologists to keep in mind that patients with Alagille syndrome who present with focal cerebral ischemic symptoms in their preoperative history should be evaluated for moyamoya syndrome or more proximal carotid lesions with magnetic resonance imaging or angiography.

The anesthetic management for revascularization is not complicated by considerations of ICP, the effects of anesthetics on CBF, and so on, but requires careful maintenance of normal physiologic blood hemodynamics and of normocarbia, because both hypocarbia and hypercarbia can produce vascular steal phenomena and result in ischemia to marginally perfused cerebral tissue ( Sumikawa and Nagai, 1983 ; Bingham and Wilkinson, 1985 ; Chadha et al., 1990 ; Martino and Werner, 1991 ; Kurehara et al., 1993 ; Petty, 1993 ; Soriano et al., 1993 ; Henderson and Irwin, 1995 ; Kansha et al., 1997 ; Sato et al., 1999 ). As for other cerebrovascular surgical procedures in which cerebral ischemia is a risk, the induction of mild to moderate hypothermia is frequently used.

Epilepsy Surgery

According to the Epilepsy Foundation, 300,000 American children under the age of 14 have epilepsy, and for those with seizures not controlled by medications there are now therapeutic options. When medication fails to adequately control seizures, neurologists may place children on ketogenic diets or implant vagus nerve stimulators. In refractory cases, one option is surgical resection of the epileptic cortex. 20% to 30% of patients with intractable epilepsy may benefit from a surgical procedure.

Epilepsy surgery is most beneficial in patients with partial epilepsy due to a structural lesion, which most frequently lies within the temporal lobe. The standard surgical procedures for patients with epilepsy include focal resection, corpuscallostomy, hemispherectomy, and vagal nerve stimulation. The most common operation for patients with intractable seizures is a partial temporal lobectomy. When the epileptic focus lies in the left hemisphere, a Wada test (an intracarotid injection of a barbiturate in the awake or sedated patient) may be performed to determine if the site of the surgery is indeed dominant and contains the speech center. If in fact the side of the lobectomy is within the speech center and the child is old enough to follow directions appropriately, the craniotomy may be performed under deep sedation with electrocorticographic monitoring ( Tobias and Jimenez, 1997 ; Domaingue, 2001 ; Ard et al., 2003 ). With propofol and low-dose remifentanil, as well as appropriate local wound infiltration by the surgeon, the patient can be awakened intraoperatively to interact with the surgeon once the seizure focus is exposed. This minimizes the risk of surgical damage to the speech center. More often, however, children are not appropriate subjects for an awake craniotomy, and the epileptic focus is electrically mapped preoperatively and then subsequently resected.

In this instance, children undergo magnetic resonance imaging before a craniotomy for placement of an electrode grid followed at some time later by a craniotomy and surgical placement of surface and/or depth grids marking the epileptic foci. The earlier discussions regarding craniotomies may be used as a guideline for anesthetic management; there are generally no concerns regarding the management of ICP or CBF, the primary considerations being the effects of chronic administration of anticonvulsants on the pharmacodynamics and pharmacokinetics of neuromuscular blockers, and the relative resistance of the patient to these agents ( Alloul et al., 1996 ; Hernandez-Palazon et al., 2001 ). In addition, patients may experience a seizure at any time before, during, or after the anesthesia, and provisions must be at hand for the rapid treatment of the seizure with a short-acting barbiturate, benzodiazepine, or propofol. After the surgical placement of electrical cortical grids and recovery from the anesthetic, patients are monitored for several days in the hospital, using continuous electroencephalography and other neurologic testing, to map the epileptic foci.

After cortical mapping and identification of the seizure focus, the patient returns to the operating room for the definitive cortical resection. The anesthetic management includes the principles involved with craniotomies. In addition, epilepsy surgery frequently requires manipulation and stimulation of structures that can cause acute and severe bradycardia, sinus arrest, and hypotension ( Sato et al., 2001 ; Sinha et al., 2004 ). The structures most sensitive to stimulation, especially by temperature changes from irrigation, include the amygdala, insular cortex, and brainstem. Despite surgical cortical resection of the epileptic focus, patients remain at risk during emergence and recovery from anesthesia for seizures. Provisions must always be at hand for the rapid treatment of a seizure and for control of the patient's airway should a seizure compromise respiration.

Vagal nerve stimulation (VNS) is another surgical alternative to treatment of medically intractable seizures. Because seizures are highly synchronized patterns on the EEG, the thought is that appropriately timed stimulation of the vagus nerve can blunt the paroxysmal epileptiform activity. The device is programmed to provide baseline stimulation of the left vagus nerve. Implantation of the device involves creating an infraclavicular subcutaneous pocket to house the impulse generator, tunneling the electrodes subcutaneously to a left anterior cervical incision and connecting the electrodes to the left vagus nerve. The principles of the VNS are similar to those of cardiac pacemakers, and consequently, for patients returning to surgery for other reasons, the VNS unit needs to be turned off so there will be no interference with the electrocautery device. Side effects of VNS involve vocal cord paralysis, bronchoconstriction, bradycardia, and asystole ( Patwardhan et al., 2000 ; Smyth et al., 2003 ; Bijwadia et al., 2005 ).

Finally, for those children who present for anesthesia who have been managed with a ketogenic diet, the anesthetic management requires the use of glucose-free intravenous solutions, and periodic measurement of serum pH and bicarbonate to detect and allow the early treatment of metabolic acidosis.

Spinal Cord Surgery

Spinal surgery in children is necessary to correct various congenital or acquired conditions such as tethered spinal cords, myelomeningoceles, primary or metastatic tumors, Chiari malformations, and herniated discs. Presenting symptoms usually include progressive gait abnormalities, paresis, neuropathic pain, and/or changes in bowel or bladder function. A laminectomy may be performed before the onset of the neurologic symptoms or to alleviate existing symptoms.

Preoperative anesthetic considerations for laminectomies include a detailed history and physical examination of the patient, including accurate documentation of any sensory or motor deficits and other neurologic symptoms prior to surgery. As discussed in the section on myelodysplasias, patients with myelomeningoceles have a high incidence of allergy to latex, so appropriate precautions should be taken ( Kurup and Fink, 2001 ; Hepner and Castells, 2003 ).

The focus of intraoperative management is minimizing spinal cord ischemia and compression on the spinal cord. These are accomplished by maintenance of spinal cord perfusion pressure through control of blood pressure and minimizing venous congestion through careful positioning of the patient to prevent compression of the abdomen.

Blood loss during most neurosurgical laminectomies is minor; however, highly vascular tumors may cause substantial bleeding, which can be especially significant over several hours of surgical time. In addition to routine monitoring and vascular access, placement of an intra-arterial line should be strongly considered in surgical cases that are expected to last several hours, especially if for tumor resection, because the blood loss can be insidious, significant, and difficult to assess. Placement of a urinary catheter should also be used in these cases as an additional measure of fluid status and to avoid postoperative bladder distention.

New neurologic deficits occur as a complication of spinal cord surgery. Intraoperative neuromonitoring of spinal cord function is often used by surgeons for early detection of spinal cord compromise. The intraoperative wake-up test is the traditional method for assessing the integrity of the spinal cord, but this can be impossible in an infant or a young child who is not able to follow commands appropriately (Soriano et al., 2002 ). SSEPs and MEPs provide a moment-to-moment assessment of spinal cord function, but use of the two has implications for anesthetic management as both are sensitive to anesthetic agents ( Sala et al., 2002, 2004 ; Soriano et al., 2002 ; Strahm et al., 2003 ). Abrupt changes in volatile anesthetic concentration can affect the signals for SSEPs and MEPs ( Soriano et al., 2002 ). Mapping MEPs precludes the use of neuromuscular blocking drugs. The anesthetic management of patients undergoing surgery under SSEP or MEP monitoring is discussed in Chapters 7 , 9 , and 21 .

Positioning of the child for spinal cord surgery carries risks discussed previously in the section on prone positioning. Of great importance is the prevention of abdominal compression, which has the effect of raising inferior caval pressure and causing the shunting of venous blood into the epidural venous plexus from the cava. The chest and pelvis must be placed on bolsters to elevate the abdomen from the operating room table. The head may be turned to one side or maintained on a headrest in the midline, with care being taken to avoid overflexion or overextension of the neck. After final positioning has been achieved, breath sounds must be auscultated to confirm repositioning has not changed the endotracheal tube position. Pressure points must all be padded, and assurance must be made that there is no compression of the male genitalia or the female breasts, the orbits, and the auricles of the ears. The eyes and ears should be rechecked routinely and periodically as part of the intraoperative monitoring.

Lesions Usually Associated with Elevated Intracranial Pressure


Hydrocephalic children with ventriculoperitoneal or ventriculoatrial shunts frequently experience shunt malfunction or failure and come to medical attention when intracranial pressure becomes elevated. Early symptoms of shunt malfunction are headache and irritability, and later symptoms include lethargy, seizure, vomiting, and ophthalmoplegia ( Fig. 18-8 ). Most commonly, the shunt malfunction occurs in the distal shunt tubing within the atrium or peritoneum, or in the valve in the scalp. If the malfunction is not in the proximal intraventricular portion of the shunt, the neurosurgeon, pediatrician, or anesthesiologist can place a needle in the shunt reservoir (which is easily palpable under the scalp) and withdraw an aliquot of CSF, thereby lowering ICP. This maneuver may be lifesaving. Placing a needle in the reservoir also provides the anesthesiologist the ability to monitor ICP ( Fig. 18-9 ). In a select group of patients with noncommunicating hydrocephalus and preservation of the pathway between the subarachnoid space and the venous system, a third ventriculostomy can be performed and alleviate the need for a shunt. Complications of the third ventriculostomy include third cranial nerve paresis, hemiparesis, and bradycardia ( El-Dawlatly et al., 2000 ). Because the fenestration of the third ventricular floor is in close proximity to the basilar artery, traumatic hemorrhage can also occur.


FIGURE 18-8  An infant with hydrocephalus. Note enlarged head and the downward gaze (the setting sun sign). The latter suggests the presence of intracranial hypertension.




FIGURE 18-9  Technique for inserting a needle in the reservoir of a ventriculoatrial or ventriculoperitoneal shunt, to remove cerebrospinal fluid or to monitor intracranial pressure.  (From Wilkinson HA: Intracranial pressure monitoring: Techniques and pitfalls. In Cooper PR, editor, Head injury. Baltimore, 1981, Williams & Wilkins.)




Anesthetic Considerations.

The anesthetic technique depends on whether intracranial hypertension exists. If so, no opioid premeditation should be given, and an intravenous induction with propofol or a thiobarbiturate and nondepolarizing agent is appropriate, as discussed earlier. Hyperventilation and avoidance of potent inhalation agents help to control ICP. Once the ventricles have been decompressed, a halogenated agent may be introduced and hyperventilation discontinued.

Brain Tumors

Cancer is the most common nontraumatic cause of death in children, and brain tumors are the most common solid tumor of childhood. In adults, two thirds of brain tumors are supratentorial; the opposite is true in children, in whom two thirds of brain tumors occur in the posterior fossa ( Table 18-4 ). Astrocytomas (including glioblastomas) of various degrees of malignancy, and medulloblastomas together account for more than half of pediatric central nervous system tumors ( Walker, 1976 ).

TABLE 18-4   -- Distribution of common brain tumors in children, according to location and histologic appearance

Location and Type of Tumor

Percentage of All Brain Tumors[*]


45 to 60

Primitive neuroectodermal tumor (medulloblastoma)

20 to 25

Low-grade cerebellar astrocytoma

12 to 18


4 to 8

Malignant brainstem glioma

3 to 9

Low-grade brainstem astrocytoma

3 to 6


2 to 5

Supratentorial Hemispheric

25 to 40

Low-grade astrocytoma

8 to 20

Malignant glioma

6 to 12


2 to 5

Mixed glioma

1 to 5


1 to 5


1 to 2

Choroid-plexus tumor

1 to 2

Primitive neuroectodermal tumor

1 to 2


0.5 to 2


1 to 3

Supratentorial Midline

15 to 20


6 to 9


4 to 8

Low-grade chiasmatic-hypothalamic glioma

1 to 2

Germ-cell tumor

1 to 2

Pituitary adenoma

0.5 to 2.5


2 to 6

Low-grade glioma

1 to 2

Germ-cell tumor

0.5 to 2

Pineal parenchymal tumor

0.5 to 2

With permission from Pollack IF: N Engl J Med 331:1500, 1994.


Percentages are derived from reviews of population-based and institutional tumor registries (and Pollack IF, et al., unpublished data).



Some children with brain tumors have significant elevation of ICP because of the mass effect of the tumor, cerebral edema, or secondary hydrocephalus if the tumor obstructs CSF pathways. Others have minimal alteration in cerebral dynamics but have come to medical attention while the tumor is small and causing symptoms as a result of destruction or compression of neural structures.

Anesthetic Considerations.

Planning the anesthetic care hinges on whether ICP is elevated. Supratentorial tumor resections require invasive monitoring and an anesthetic technique designed to control ICP ( Fig. 18-10 ). Suboccipital craniotomies for explorations of the posterior fossa or brainstem present the anesthesiologist with several unique problems associated with positioning and maintenance of the airway (discussed earlier) and with changes in respiratory and cardiovascular function associated with brainstem compression ( Allan et al., 1970 ) ( Fig. 18-11 ). Neurosurgeons commonly place the child's head in a Mayfield head frame for tumor resection and, thus, the anesthesiologist must keep in mind the potential for skull fractures, dural tears, and intracranial hematomas from the pins in pediatric patients ( Soriano et al., 2002 ).


FIGURE 18-10  Magnetic resonance image of supratentorial tumor. High-grade gliomas are generally less well circumscribed with irregular enhancement.




FIGURE 18-11  Magnetic resonance image of infratentorial tumors. A, Brainstem glioma. B, Ependymoma displacing the pons and medulla.



A common intraoperative complication to posterior fossa exploration is associated with surgical trauma of the brainstem, especially when surgery is within the fourth ventricle. Brainstem compression or retraction may alter the respiratory pattern or cause hiccoughs or apnea. To facilitate monitoring for these events, spontaneous breathing was once advocated. The risk of spontaneous ventilation and negative thoracic pressure in promoting VAE, however, outweighs its benefit ( Michenfelder et al., 1969 ). Brainstem compression also may be detected by its effects on cardiovascular function. Arrhythmias occur simultaneously with alterations in ventilatory control, particularly sudden tachycardia, premature ventricular beats, nodal arrhythmias, and bradyarrhythmias ( Michenfelder et al., 1969 ;Millar, 1972 ; Davies and Munro, 1975 ; Albin et al., 1976 ; Pollack, 1994 ) and occur in as many as 14% of pediatric patients ( Meridy et al., 1974 ). There also may be a sudden alteration in vascular tone, resulting in sudden hypotension or hypertension. Notifying the neurosurgeon to discontinue the manipulation is generally the only intervention needed, although postoperative brainstem dysfunction may follow.

Occasionally, intraoperative surgical trauma to the brainstem during tumor resection results in postoperative dysfunctions. These include delayed awakening or prolonged unresponsiveness, impaired respiratory drive and central hypoventilation, loss of protective airway reflexes, and vocal cord paralysis. For these reasons, postoperative extubation of the trachea should follow spontaneous ventilation and return of consciousness, although some authors advocate extubation during spontaneous ventilation and deep anesthesia to minimize coughing ( Allan et al., 1970 ). Shortly after extubation, the integrity of airway reflexes should be assessed. Immediate stridor after extubation suggests vocal cord paralysis, whereas delayed stridor is more consistent with postextubation croup. Fiberoptic transnasal laryngoscopy is useful for assessment of vocal cord function when the diagnosis is unclear.

Postoperative craniotomy patients are observed at least overnight in an intensive care setting. Latent complications, such as intracranial bleeding, must be detected early, evaluated, and treated expeditiously to minimize neurologic sequelae. Bleeding in the posterior fossa usually produces dramatic alterations in consciousness and rapid deterioration. Seizures and inappropriate secretion of antidiuretic hormone, with subsequent hyponatremia, also are occasional delayed complications of craniotomies.


Craniopharyngiomas are histologically benign suprasellar tumors, the morbidity of which result from local destruction or compression of nearby important structures, notably the hypothalamus, the optic chiasm, and the pituitary gland. They are the third most common intracranial tumor in children and are treated by resection and decompression followed by radiation therapy ( Gonc et al., 2004 ).

These patients often have endocrine abnormalities, and thyroid and adrenal function studies are obtained preoperatively. In a study review of 66 children with craniopharyngiomas, growth retardation or pubertal delay seems to be one of the first findings. Moreover, there is commonly a delayed diagnosis as evidenced by the presence of nausea and vomiting due to elevated ICP from tumor size ( Gonc et al., 2004 ). If resection of the tumor is urgent, that is, if compression of the optic chiasm threatens vision, adrenal insufficiency is assumed to be present and is treated accordingly. Hydrocephalus and intracranial hypertension also may be present, necessitating ventriculostomy placement before resection of the tumor. In view of the hazards of surgical resection, predominantly cystic tumors can be treated with intracavity irradiation (phosphorus 32). Perioperative morbidity is low, and the cyst involutes in the majority of patients ( Pollack, 1994 ) ( Fig. 18-12 ).


FIGURE 18-12  Magnetic resonance image of craniopharyngiomas demonstrating both cystic and solid components.



The usual surgical approach to craniopharyngiomas in children is a frontal craniotomy, with dissection under a frontal lobe through an olfactory nerve to reach the optic chiasm. Microscopic technique is usually used, and surgery is often long and laborious.

Postoperative problems are common. Diabetes insipidus occurs frequently, usually within hours of surgery. Occasionally, diabetes insipidus may become evident intraoperatively; its treatment in either event is the same. The diagnosis of diabetes insipidus may be made if there are large urinary losses with associated euvolemia or hypovolemia, increasing serum sodium and serum osmolality, and inappropriately dilute urine osmolality (urine osmolality is usually less than 200 mOsm/L during diabetes insipidus). The diuresis of diabetes insipidus must be replaced on an hourly basis with an appropriately dilute intravenous fluid; 2.5% dextrose in 0.2 normal saline solution is appropriate, but even this low concentration of glucose frequently produces hyperglycemia. If fluid replacement of urine loss alone is inadequate to maintain euvolemia or results in unacceptable elevation of the serum glucose level, as is usually the case, diabetes insipidus must be treated pharmacologically with an aqueous vasopressin infusion, intranasal desmopressin (1-desamino-8-D-arginine vasopressin [DDAVP]), or intravenous DDAVP. Aqueous vasopressin often produces hypertension and decreased splanchnic blood flow because of the vasopressor effect of vasopressin, whereas DDAVP is free of cardiovascular side effects but its duration of action is longer than that of vasopressin. Vasopressin is started at an infusion rate of 0.5 mU/kg per hr. The rate may be successively doubled until the desired antidiuresis is achieved. The pediatric intranasal dose of DDAVP is 0.05 to 0.3 mL/day divided into two doses (5 to 30 mcg/day). The intravenous dose is 0.5 to 3 mcg/day, also divided into two doses.

Neurosurgery around the hypothalamus is also associated with cerebral salt wasting, which also causes massive diuresis. Easily confused with diabetes insipidus, cerebral salt wasting causes natriuresis, hyponatremia (unlike diabetes insipidus), and markedly elevated urine sodium concentration (50 to 150 mEq/L). The syndrome resolves spontaneously, but it may persist for weeks. Therapy is supportive, consisting of salt replacement to compensate for the natriuresis.

Anesthetic Considerations.

In addition to the concerns regarding diabetes insipidus and cerebral salt wasting, an ongoing controversy among neurosurgeons and neuroanesthesiologists is whether the trachea should be extubated with the patient awake or deeply anesthetized after craniotomy. Extubation when the patient is awake ensures the presence of intact airway reflexes and allows rapid evaluation of neurologic function by the neurosurgeon. Deep extubation in the presence of halogenated agents is often associated with less coughing and bucking on emergence but leaves the patient at risk for airway obstruction, reflux and aspiration of gastric contents, and delays in awakening and neurologic assessment. The anesthesiologist must reach his or her own accord with the neurosurgeon. However, extubation when the patient is awake generally is preferred. It may be accomplished with little or no coughing and bucking if (1) an adequate level of narcosis exists at the end of surgery, (2) neuromuscular relaxation is maintained until just before awakening is desired, (3) intravenous lidocaine is given just before nitrous oxide inhalation is terminated, and (4) no halogenated inhalation agents (which delay emergence) are present during awakening. If these conditions are met, patients frequently respond to commands shortly after nitrous oxide is terminated and will not cough. If coughing does occur, it may be treated with intravenous lidocaine and a short-acting synthetic opioid.

Head Trauma

Accidents are the most common cause of death in childhood, and most children dying of traumatic injuries have head trauma. The outcome of childhood head trauma is superior to its outcome in adults (Bruce et al., 1978 ; Kraus et al., 1987 ). The Glasgow Coma Scale (see Table 18-3 ) is a useful predictor of outcome: mortality associated with a score of less than 8 is 59% ( Bruce et al., 1981 ; Kraus et al., 1987 ). Global cerebral reactivity to CO2 is preserved in many children with a Glasgow Coma Scale score of greater than 4 ( Meyer et al., 1999 ).

Children with head trauma are different from adults with head trauma in three important ways. The first is the epidemiology of the lesions. The second is the phenomenon of “malignant brain edema” seen commonly in children but rarely in adults. The third is the hemodynamic response to cranial hemorrhage in children.

Only a minority of children with head trauma require surgical intervention to remove an intracranial hematoma ( Kraus et al., 1987 ). Most head-injured children have cerebral concussions or contusions. Epidural hematoma accounts for fewer than 10% of pediatric head injuries, and subdural hematomas are equally uncommon. More common is diffuse axonal cerebral injury accompanied by cerebral swelling, or “malignant brain edema.” This is a notable problem in children from birth to 16 years of age who seem to be prone to the development of acute, diffuse brain swelling, even in association with a seemingly minor closed head injury ( Bruce, 1980 ; Bruce et al., 1981 ). At the cellular level, brain injury initiates an excitotoxic cascade, which increases CSF glutamate. This cascade also activates the N-methyl-D-aspartate (NMDA) receptors, which are involved in the modulation of intracellular secondary messengers. The end result is an increase in intracellular calcium and a cascade of intracellular destructive reactions that include proteolysis, lipid peroxidation, free radical formation, and degeneration of neurons.

Brain swelling in this setting is caused by dramatic increases in cerebral metabolism, CBF, and CBV rather than to primary edema formation, which follows secondarily ( Kasoff et al., 1972 ; Bruce et al., 1981 ). One might intuitively conclude that therapy designed to minimize cerebral metabolism and control arterial blood pressure would be most effective, but experimental confirmation of the best treatment is lacking.

Arterial hypertension is a common sequel to head trauma. Hypertension may be caused by pain, agitation, or intracranial hypertension. Arterial hypertension may in turn contribute to cerebral edema formation, particularly in regions of the brain with vasomotor paralysis and increased CBF ( Durward et al., 1983 ), in which blood flow is passively pressure dependent. Control of arterial blood pressure then becomes an important consideration. If arterial hypertension is secondary to acute elevation of ICP, therapy should first be directed toward lowering ICP. If acute intracranial hypertension has been ruled out as the cause, analgesics and sedatives may suffice. Arterial hypertension unresponsive to these initial measures may, however, be a nonspecific response to intracranial injury. As such it is frequently associated with increased cardiac stimulation and increased serum catecholamine levels. In this setting, propranolol and labetalol are useful agents for control of hypertension and tachycardia and do not elevate ICP ( Feibel et al., 1981 ; Robertson et al., 1983 ).

When hypotension is associated with a head injury in a child, the first consideration is acute blood loss from associated lesions as small as scalp lacerations, as more than 50% of the severely head-injured children are multiple trauma patients ( Orliaguet et al., 1998 ). In the absence of blood losses, “neurogenic hypotension” could be the cause of hypotension. Its precise mechanism has not been clearly elucidated, but exhaustion of endogenous catecholamines after a massive release following trauma has been suggested as an explanation ( Chesnut et al., 1998 ). As is the case in adults, when hypotension is associated with a head injury in a child, one should search for an associated injury in the thorax and abdomen, but unlike the adult patient, a cranial hematoma may contain one third to one half of an infant's blood volume. Small children with skull fractures and associated epidural or subgaleal hematomas can have significant hemorrhage into the hematoma and subsequent hypovolemia and hypotension, which may be further aggravated by preoperative mannitol or diuretic therapy. Head injuries in adolescents and adults are not associated with blood loss of such a relative magnitude, and hypotension in this population is caused most often by a second injury.

Spinal cord injuries in children may exist concurrently with head injuries and must be ruled out. Plain radiographs of the spine in preschool-aged children are difficult to interpret because of incomplete ossification of the spine, and occult cervical fractures may be present. The best way to make this determination is during the cranial CT scan, by scanning the cervical spine as well. If a spine injury is suspected or known to be present, the trachea must be intubated only while an assistant, preferably a neurosurgeon, holds and stabilizes the head and neck, preventing excessive flexion or extension of the spine. If the head- and spinal cord-injured patient also has an airway compromised by facial trauma, a tracheostomy under local anesthesia is the safest way to avoid loss of the airway and further injury to the spinal cord.

Anesthetic Considerations.

The anesthetic care of the child with a head injury often begins in the emergency department or the trauma ward. Evaluation of the airway and endotracheal intubation are of course the first priorities. The awake or arousable child may be watched closely. The unresponsive child should have an endotracheal tube inserted and be hyperventilated. ICP rarely is monitored at this point, but intracranial hypertension should be assumed to be present and the effects of laryngoscopy and intubation on ICP should be borne in mind. These potent stimuli should follow a sleep dose of a short-acting propofol (2 to 3 mg/kg) or thiobarbiturate (4 to 7 mg/kg) and succinylcholine (after defasciculation). Intravenous lidocaine (1 to 2 mg/kg) and a potent synthetic opioid (e.g., fentanyl, sufentanil, or remifentanil) are useful adjunctive drugs to blunt the cerebral hemodynamic response to laryngoscopy and intubation. If hypovolemia exists or is suspected, the dose of propofol or barbiturate should be reduced, or midazolam (0.1 to 0.6 mg/kg) may be substituted for hypnosis and amnesia. Midazolam should be avoided in the presence of overt hypovolemia. With facial or airway trauma, a difficult intubation is possible, and neuromuscular blockade usually is withheld. Intubation may be performed after topical anesthesia, a laryngeal nerve block, and sedation with lorazepam or midazolam. Nasotracheal intubation should not be performed when a basilar skull fracture is suspected. The management of the difficult airway is discussed in Chapter 10 , Induction of Anesthesia.

The child with head trauma who requires neurosurgery should be fully monitored, with the use of invasive cardiovascular monitors (arterial cannula and a central venous catheter). Adequate quantities of banked blood must be available in the operating room, and an intravenous cannula of sufficient size for fluid resuscitation should be placed. The choice of anesthetic agents should take into account the likelihood of intracranial hypertension, and this is discussed in preceding sections. Sharples and others noted that almost 30% of the children dying within the first hours following head trauma could have been saved if adequate evaluation and prompt treatment of hypoventilation and hypotension had been initiated earlier ( Cruz, 1996 ). Hypotension alone and hypotension with associated hypoxemia triple and quadruple the mortality of pediatric brain injuries ( Pigula et al., 1993 ).

Copyright © 2008 Elsevier Inc. All rights reserved. -

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

Copyright © 2005 Mosby, An Imprint of Elsevier


Anesthesia for the pediatric neurosurgical patient requires a knowledge of cerebral pathophysiology and an understanding of the interaction between anesthetic agents and cerebral physiology. It is crucial for the anesthesiologist, neurosurgeon, and radiologist to work together to provide a team-structured approach toward diagnosis, resection, and treatment for pediatric patients, especially because children gain 80% of their brain weight between 6 and 12 months and this time is so critical for a good, positive outcome ( Kang and Lee, 1999 ). For the anesthesiologist caring for neurosurgical patients, careful attention to detail is important in ensuring optimal outcome for each child.

Copyright © 2008 Elsevier Inc. All rights reserved. -

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

Copyright © 2005 Mosby, An Imprint of Elsevier


Adams et al., 1981. Adams RW, Cucchiara RF, Gronert GA, et al: Isoflurane and cerebrospinal fluid pressures in neurosurgical patients.  Anesthesiology  1981; 54:97.

Adams et al., 1972. Adams RW, Gronert GA, Sundt Jr TM, et al: Halothane, hypocapnia, and cerebrospinal fluid pressure in neurosurgery.  Anesthesiology  1972; 37:510.

Adornato et al., 1978. Adornato DC, Gildenberg PL, Ferrario CM, et al: Pathophysiology of intravenous air embolism in dogs.  Anesthesiology  1978; 49:127.

Alagille, 1996. Alagille D: Alagille syndrome today.  Clin Invest Med  1996; 19:325.

Albanese et al., 1993. Albanese J, Durbec O, Viviand X, et al: Sufentanil increases intracranial pressure in patients with head trauma.  Anesthesiology  1993; 79:493.

Albin, 1984. Albin M: The paradox of paradoxic air embolism.  Anesthesiology  1984; 61:222.

Albin et al., 1976. Albin MS, Babinski M, Maroon JC, et al: Anesthetic management of posterior fossa surgery in the sitting position.  Acta Anaesthesiol Scand  1976; 20:117.

Albrecht et al., 1983. Albrecht RF, Miletich DJ, Madala JR: Normalization of cerebral blood flow during prolonged halothane anesthesia.  Anesthesiology  1983; 58:26.

Albrecht et al., 1977. Albrecht RF, Miletich DJ, Rosenberg R, et al: Cerebral blood flow and metabolic changes from induction to onset of anesthesia with halothane or pentobarbital.  Anesthesiology  1977; 47:252.

Albright et al., 1984. Albright AL, Latchaw RE, Robinson AG: Intracranial and systemic effects of osmotic and oncotic therapy in experimental cerebral edema.  J Neurosurg  1984; 60:481.

Alexander et al., 1964. Alexander SC, Wollman H, Cohen PJ, et al: Cerebrovascular response to PaCO2 during halothane anesthesia in man.  J Appl Physiol  1964; 19:561.

Algotsson et al., 1988. Algotsson L, Messeter K, Nordstrom CH, et al: Cerebral blood flow and oxygen consumption during isoflurane and halothane anesthesia in man.  Acta Anaesthesiol Scand  1988; 32:15.

Algotsson et al., 1992. Algotsson L, Messeter K, Rosen I, et al: Effects of nitrous oxide on cerebral haemodynamics and metabolism during isoflurane anaesthesia in man.  Acta Anaesthesiol Scand  1992; 36:46.

Allan et al., 1970. Allan D, Kim H, Cox JM: The anaesthetic management of posterior fossa explorations in infants.  Can Anaesth Soc J  1970; 17:227.

Alloul et al., 1996. Alloul K, Whalley DG, Shutway F, et al: Pharmacokinetic origin of carbamazepine-induced resistance to vecuronium neuromuscular blockade in anesthetized patients.  Anesthesiology  1996; 84:330.

Archer et al., 1987. Archer DP, Labrecque P, Tyler JL, et al: Cerebral blood volume is increased in dogs with administration of nitrous oxide or isoflurane.  Anesthesiology  1987; 67:642.

Ard et al., 2003. Ard J, Doyle W, Bekker A: Awake craniotomy with dexmedetomidine in pediatric patients.  J Neurosurg Anesthesiol  2003; 15:263.

Artru, 1984d. Artru AA: Relationship between cerebral blood volume and CSF pressure during anesthesia with isoflurane or fentanyl in dogs.  Anesthesiology  1984; 60:575.

Artru et al., 1994. Artru AA, Powers K, Doepfner P: CSF, sagittal sinus, and jugular venous pressures during desflurane or isoflurane anesthesia in dogs.  J Neurosurg Anesthesiol  1994; 6:239.

Artru, 1984a. Artru AA: Effects of enflurane and isoflurane on resistance to reabsorption of cerebrospinal fluid in dogs.  Anesthesiology  1984; 61:529.

Artru, 1984b. Artru AA: Effects of halothane and fentanyl anesthesia on resistance to reabsorption of CSF.  J Neurosurg  1984; 60:252.

Artru, 1983a. Artru AA: Effects of halothane and fentanyl on the rate of CSF production in dogs.  Anesth Analg  1983; 62:581.

Artru, 1994. Artru AA: Intracranial volume/pressure relationship during desflurane anesthesia in dogs: Comparison with isoflurane and thiopental/halothane.  Anesth Analg  1994; 79:751.

Artru, 1984c. Artru AA: Isoflurane does not increase the rate of CSF production in the dog.  Anesthesiology  1984; 60:193.

Artru, 1982. Artru AA: Nitrous oxide plays a direct role in the development of tension pneumocephalus intraoperatively.  Anesthesiology  1982; 57:59.

Artru, 1983b. Artru AA: Relationship between cerebral blood volume and CSF pressure during anesthesia with halothane or enflurane in dogs.  Anesthesiology  1983; 58:533.

Artru, 1992. Artru AA: Venous air embolism in prone dogs positioned with the abdomen hanging freely: Percentage of gas retrieved and success rate of resuscitation.  Anesth Analg  1992; 75:715.

Azar, 1984. Azar I: The response of patients with neuromuscular disorders to muscle relaxants: A review.  Anesthesiology  1984; 61:173.

Bakay, 1960. Bakay L: Studies in sodium exchange.  Neurology  1960; 10:564.

Baker et al., 1997. Baker KZ, Ostapkovich N, Sisti MB, et al: Intact cerebral blood flow reactivity during remifentanil/nitrous oxide anesthesia.  J Neurosurg Anesthesiol  1997; 9:134.

Bedford et al., 1980. Bedford RF, Parsing JA, Pobereskin L, et al: Lidocaine or thiopental for rapid control of intracranial hypertension.  Anesth Analg  1980; 59:435.

Bedford and Perkins-Pearson, 1982. Bedford RF, Perkins-Pearson NAK: PEEP for treatment of venous air embolism.  Anesthesiology  1982; 57:A379.

Bedford, 1983. Bedford RF: Venous air embolism: A historical perspective.  Semin Anesth  1983; 2:169.

Bell and McCormick, 1978. Bell WE, McCormick WF: Transtentorial and cerebellar herniations.   In: Schaffer AJ, Markowitz M, ed. Major problems in clinical pediatrics, vol 8, Increased intracranial pressure in children: Diagnosis and treatment,  Philadelphia: WB Saunders; 1978.

Bering, 1952. Bering Jr EA: Water exchange of central nervous system and cerebrospinal fluid.  J Neurosurg  1952; 9:275.

Biagas and Gaeta, 1998. Biagas KV, Gaeta ML: Treatment of traumatic brain injury with hypothermia.  Curr Opin Pediatr  1998; 10:271.

Bijwadia et al., 2005. Bijwadia JS, Hoch RC, Dexter DD: Identification and treatment of bronchoconstriction induced by a vagus nerve stimulator employed for management of seizure disorder.  Chest  2005; 127:401.

Bingham and Wilkinson, 1985. Bingham RM, Wilkinson DJ: Anaesthetic management in moya-moya disease.  Anaesthesia  1985; 40:1198.

Bithal et al., 2003. Bithal P, Dash HH, Vishnoi N, et al: Venous air embolism: Does the site of embolism influence the hemodynamic changes?.  Neurol India  2003; 51:370.

Bithal et al., 2004. Bithal PK, Pandia MP, Dash HH, et al: Comparative incidence of venous air embolism and associated hypotension in adults and children operated for neurosurgery in the sitting position.  Eur J Anaesthesiol  2004; 21:517.

Black et al., 1990. Black S, Muzzi DA, Nishimura RA, et al: Preoperative and intraoperative echocardiography to detect right-to-left shunt in patients undergoing neurosurgical procedures in the sitting position.  Anesthesiology  1990; 72:436.

Bowdle and Artru, 1988. Bowdle TA, Artru AA: Treatment of air embolism with a special pulmonary artery catheter introducer sheath in sitting dogs.  Anesthesiology  1988; 68:107.

Brown and Charlton, 1975. Brown JC, Charlton JE: Study of sensitivity to curare in certain neurological disorders using a regional technique.  J Neural Psychiatry  1975; 38:34.

Bruce et al., 1981. Bruce DA, Alavi A, Bilaniuk L, et al: Diffuse cerebral swelling following head injuries in children: The syndrome of “malignant brain edema.”.  J Neurosurg  1981; 54:170.

Bruce et al., 1977. Bruce DA, Berman WA, Schut L: Cerebrospinal fluid pressure monitoring in children: Physiology, pathology and clinical usefulness.  Adv Pediatr  1977; 24:233.

Bruce et al., 1978. Bruce DA, Schut L, Bruno LA, et al: Outcome following severe head injury in children.  J Neurosurg  1978; 48:679.

Bruce, 1980. Bruce DA: Special considerations in the pediatric age group.   In: Cooper PR, ed. Head injury,  Baltimore: Williams & Wilkins; 1980.

Brussel et al., 1991. Brussel T, Fitch W, Brodner G, et al: Effects of halothane in low concentrations on cerebral blood flow, cerebral metabolism, and cerebrovascular autoregulation in the baboon.  Anesth Analg  1991; 73:758.

Bunegin et al., 1981. Bunegin L, Albin MS, Helsel PE, et al: Positioning the right atrial catheter: A model for reappraisal.  Anesthesiology  1981; 55:343.

Cafiero et al., 2004. Cafiero T, Burrelli R, Latina P, et al: Analgesic transition after remifentanil-based anesthesia in neurosurgery. A comparison of sufentanil and tramadol.  Minerva Anesthesiol  2004; 70:45.

Chadha et al., 1990. Chadha R, Singh S, Padmanabhan V: Anaesthetic management in moyamoya disease.  Anaesth Intensive Care  1990; 18:120.

Chen et al., 1983. Chen J, Kim YD, Dubois M, et al: The increased requirement of pancuronium in neurosurgical patients receiving Dilantin chronically.  Anesthesiology  1983; 59:A288.

Cheney and Colley, 1980. Cheney FW, Colley PS: The effect of cardiac output on arterial Wood oxygenation.  Anesthesiology  1980; 52:496.

Chesnut et al., 1998. Chesnut RM, Gautille T, Blunt BA, et al: Neurogenic hypotension in patients with severe head injuries.  J Trauma  1998; 44:958.

Chiron et al., 1992. Chiron C, Raynaud C, Maziere B, et al: Changes in cerebral blood flow during brain maturation in children and adolescents.  J Nucl Med  1992; 33:696.

Colley and Artru, 1984. Colley PS, Artru AA: ECG-guided placement of Sorenson CVP catheters via arm veins.  Anesth Analg  1984; 63:953.

Constable, 2003. Constable PD: Hyperchloremic acidosis: The classic example of strong ion acidosis.  Anesth Analg  2003; 96:919-922.

Cooperman et al., 1970. Cooperman LH, Strobel Jr GE, Kennell EM: Massive hyperkalemia after administration of succinylcholine.  Anesthesiology  1970; 32:161.

Cooperman, 1970. Cooperman LH: Succinylcholine-induced hyperkalemia in neuromuscular disease.  JAMA  1970; 213:1867.

Coté et al., 1979. Coté CJ, Greenhow E, Marshall BE: The hypotensive response to rapid intravenous administration of hypertonic solutions in man and rabbit.  Anesthesiology  1979; 50:30.

Cotev and Shalit, 1975. Cotev S, Shalit MN: Effects of diazepam on cerebral blood flow and oxygen uptake after head injury.  Anesthesiology  1975; 43:117.

Cottrell et al., 1984. Cottrell JE, Giffin JP, Hartung J, et al: Intracranial pressure during nifedipine-induced hypotension.  Anesth Analg  1984; 62:1078.

Cottrell et al., 1980. Cottrell JE, Gupta B, Rappaport H, et al: Intracranial pressure during nitroglycerin-induced hypotension.  J Neurosurg  1980; 53:309.

Cottrell et al., 1983. Cottrell JE, Hartung J, Giffin JP, et al: Intracranial and hemodynamic changes after succinylcholine administration in cats.  Anesth Analg  1983; 62:1006.

Cottrell et al., 1978. Cottrell JE, Patel K, Turndorf H, et al: Intracranial pressure changes induced by sodium nitroprusside in patients with intracranial mass lesions.  J Neurosurg  1978; 48:329.

Cottrell et al., 1977. Cottrell JE, Robustelli A, Post K, et al: Furosemide- and mannitol induced changes in intracranial pressure and serum osmolality and electrolytes.  Anesthesiology  1977; 47:28.

Coughlan et al., 2003. Coughlan MD, Fieggen AG, Semple PL, et al: Craniocerebral gunshot injuries in children.  Childs Nerv Syst  2003; 19:348.

Cowgill et al., 1974. Cowgill DB, Mostello LA, Shapiro HM: Encephalitis and a hyperkalemic response to succinylcholine.  Anesthesiology  1974; 40:409.

Crumrine et al., 1975. Crumrine RS, Nulsen FE, Weiss MH: Alterations in ventricular fluid pressure during ketamine anesthesia in hydrocephalic children.  Anesthesiology  1975; 42:758.

Cruz, 1996. Cruz J: Adverse effects of pentobarbital on cerebral venous oxygenation of comatose patients with acute traumatic brain swelling: Relationship to outcome.  J Neurosurg  1996; 85:758.

Cucchiara and Bowers, 1982. Cucchiara RF, Bowers B: Air embolism in children undergoing suboccipital craniotomy.  Anesthesiology  1982; 57:338.

Cucchiara et al., 1985. Cucchiara RF, Seward JB, Nishimura RA, et al: Identification of patent foramen ovale during sitting position craniotomy by transesophageal echocardiography with positive airway pressure.  Anesthesiology  1985; 63:107.

Cucchiara et al., 1974. Cucchiara RF, Theye RA, Michenfelder JD: The effects of isoflurane on canine cerebral metabolism and blood flow.  Anesthesiology  1974; 40:571.

Dahlgren et al., 1981. Dahlgren N, Ingvar M, Yokoyama H, et al: Influence of nitrous oxide on local cerebral blood flow in awake, minimally restrained rats.  J Cereb Blood Flow Metab  1981; 1:211.

Daven et al., 1983. Daven JR, Milstein JM, Guthrie RD: Cerebral vascular resistance in premature infants.  Am J Dis Child  1983; 137:328.

Davies and Munro, 1975. Davies DW, Munro IR: The anesthetic management and intraoperative care of patients undergoing major facial osteotomies.  Plast Reconstr Surg  1975; 55:50.

Dawson et al., 1971. Dawson B, Michenfelder JD, Theye A: Effects of ketamine on canine cerebral blood flow and metabolism: Modification by prior administration of thiopental.  Anesth Analg  1971; 50:443.

DiGiovanni et al., 1974. DiGiovanni AJ, Goodnick J, Neign JL, et al: The effect of halothane anesthesia on intracranial pressure in the presence of intracranial hypertension.  Anesth Analg  1974; 53:823.

Domaingue, 2001. Domaingue CM: Propofol/remifentanil in neurosurgery.  Anaesth Intensive Care  2001; 29:205.

Donegan and Bedford, 1980. Donegan MF, Bedford RF: Intravenously administered lidocaine prevents intracranial hypertension during endotracheal suctioning.  Anesthesiology  1980; 52:516.

Dong et al., 1996. Dong ML, Kofke WA, Policare RS, et al: Transcranial Doppler ultrasonography in neurosurgery: Effects of intracranial tumour on right middle cerebral artery flow velocity during induction of anaesthesia.  Ultrasound Med Biol  1996; 22:1163.

Drummond et al., 1985. Drummond JC, Prutow RJ, Scheller MS: A comparison of the sensitivity of pulmonary artery pressure, end-tidal carbon dioxide, and end-tidal nitrogen in the detection of venous air embolism in the dog.  Anesthesiology  1985; 64:688.

Drummond and Todd, 1985. Drummond JC, Todd MM: The response of the feline cerebral circulation to PaCO2 during anesthesia with isoflurane and halothane and during sedation with nitrous oxide.  Anesthesiology  1985; 62:268.

Durward et al., 1983. Durward QJ, Del Maestro RF, Amacher AL, et al: The influence of systemic arterial pressure and intracranial pressure on the development of cerebral vasogenic edema.  J Neurosurg  1983; 59:803.

Edmonds-Seal and Maroon, 1969. Edmonds-Seal J, Maroon JC: Air embolism diagnosed with ultrasound.  Anaesthesia  1969; 24:438.

Edmonds-Seal et al., 1970. Edmonds-Seal J, Prys-Roberts C, Adams AP: Transcutaneous Doppler ultrasound flow detectors for diagnosis of air embolism.  Proc R Soc Med  1970; 63:831.

El-Dawlatly et al., 1999. El-Dawlatly AA, Murshid W, El-Khwsky F: Endoscopic third ventriculostomy: A study of intracranial pressure vs. haemodynamic changes.  Minim Invasive Neurosurg  1999; 42:198.

El-Dawlatly et al., 2000. El-Dawlatly AA, Murshid WR, Elshimy A, et al: The incidence of bradycardia during endoscopic third ventriculostomy.  Anesth Analg  2000; 91:1142-1144.

Emerick et al., 1999. Emerick KM, Rand EB, Goldmuntz E, et al: Features of Alagille syndrome in 92 patients: Frequency and relation to prognosis.  Hepatology  1999; 29:822.

Engelhard et al., 2004. Engelhard K, Reeker W, Kochs E, et al: Effect of remifentanil on intracranial pressure and cerebral blood flow velocity in patients with head trauma.  Acta Anaesthesiol Scand  2004; 48:396.

Faberowski et al., 2000. Faberowski LW, Black S, Mickle JP: Incidence of venous air embolism during craniectomy for craniosynostosis repair.  Anesthesiology  2000; 92:20.

Feibel et al., 1981. Feibel JH, Baldwin CA, Joynt RJ: Catecholamine-associated refractory hypertension following acute intracranial hemorrhage: Control with propranolol.  Ann Neurol  1981; 9:340.

Fischer et al., 1992. Fischer M, Moskopp D, Nadstawek J, et al: Totale intravenose Anaesthesia mit Propofol und Alfentanil verringert im Gegensatz zu einer kombinierten Inhalationsanaesthesie fie Fließgeschwindigkeit inder A. cerebri media.  Anaesthetist  1992; 41:15.

Fitch et al., 1973. Fitch W, Ferguson GG, Sengupta D, et al: Autoregulation of cerebral blood flow during controlled hypotension.  Stroke  1973; 4:324.

Fitzpatrick and Gilboe, 1982. Fitzpatrick JH, Gilboe DD: Effects of nitrous oxide on the cerebrovascular tone, oxygen metabolism, and EEG of the isolated perfused canine brain.  Anesthesiology  1982; 57:480.

Forster et al., 1982. Forster A, Judge O, Morel D: Effects of midazolam on cerebral blood flow in human volunteers.  Anesthesiology  1982; 56:453.

Forster et al., 1978. Forster A, VanHorn K, Marshall LF, et al: Anesthetic effects on blood-brain barrier function during acute arterial hypertension.  Anesthesiology  1978; 49:26.

Fox et al., 1992. Fox J, Gelb AW, Enns J, et al: The responsiveness of cerebral blood flow to changes in arterial carbon dioxide is maintained during propofol-nitrous oxide anesthesia in humans.  Anesthesiology  1992; 77:453.

Fraga et al., 2003. Fraga M, Rama-Maceiras P, Rodino S, et al: The effects of isoflurane and desflurane on intracranial pressure, cerebral perfusion pressure, and cerebral arteriovenous oxygen content difference in normocapnic patients with supratentorial brain tumors.  Anesthesiology  2003; 98:1085.

Frankville and Drummond, 1987. Frankville DD, Drummond JC: Hyperkalemia after succinylcholine administration in a patient with closed head injury without paresis.  Anesthesiology  1987; 67:264.

Fuchs et al., 1998. Fuchs G, Schwarz G, Stein J, et al: Doppler color-flow imaging: Screening of a patent foramen ovale in children scheduled for neurosurgery in the sitting position.  J Neurosurg Anesthesiol  1998; 10:5.

Gal et al., 2002. Gal R, Cundrle I, Zimova I, et al: Mild hypothermia therapy for patients with severe brain injury.  Clin Neurol Neurosurg  2002; 104:318.

Gardner et al., 1972. Gardner AE, Dannemiller FJ, Dean D: Intracranial cerebrospinal fluid pressure in man during ketamine anesthesia.  Anesth Analg  1972; 51:741.

Gauthier et al., 2002. Gauthier A, Girard F, Boudreault D, et al: Sevoflurane provides faster recovery and postoperative neurological assessment than isoflurane in long-duration neurosurgical cases.  Anesth Analg  2002; 95:1384.

Gelb et al., 2003. Gelb AW, Salevsky F, Chung F, et al: Remifentanil with morphine transitional analgesia shortens neurological recovery compared to fentanyl for supratentorial craniotomy.  Can J Anaesth  2003; 50:946.

Gerlach et al., 2003. Gerlach K, Uhlig T, Huppe M, et al: Remifentanil-propofol versus sufentanil-propofol anaesthesia for supratentorial craniotomy: A randomized trial.  Eur J Anaesthesiol  2003; 20:813.

Ghandi et al., 1976. Ghandi IC, Jindal MN, Patel VK: Mechanism of neuromuscular blockade with some antiepileptic drugs.  Arzneimittelforschung  1976; 26:258.

Ghani et al., 1989. Ghani GA, Sung YF, Weinstein MS, et al: Effects of intravenous nitroglycerin on the intracranial pressure and volume pressure response.  J Neurosurg  1989; 58:562.

Giffin et al., 1984. Giffin JP, Cottrell JE, Shwiry B, et al: Intracranial pressure, mean arterial pressure, and heart rate following midazolam or thiopental in humans with brain tumors.  Anesthesiology  1984; 60:491.

Gildenberg et al., 1981. Gildenberg PL, O'Brien RP, Britt WJ, et al: The efficacy of Doppler monitoring for the detection of venous air embolism.  J Neurosurg  1981; 54:75.

Gonc et al., 2004. Gonc EN, Yordam N, Ozon A, et al: Endocrinological outcome of different treatment options in children with craniopharyngioma: A retrospective analysis of 66 cases.  Pediatr Neurosurg  2004; 40:112.

Goss, 1973. Goss CM: Gray's anatomy,  29th ed.. Philadelphia, Lea & Febiger, 1973.

Graham, 1980. Graham DH: Monitoring neuromuscular block may be unreliable in patients with upper-motor-neuron lesions.  Anesthesiology  1980; 52:74.

Grewal and Sutcliffe, 1991. Grewal M, Sutcliffe AJ: Early prediction of outcome following head injury in children: An assessment of the value of Glasgow Coma Scale score trend and abnormal plantar and pupillary light reflexes.  J Pediatr Surg  1991; 26:1161.

Grosslight et al., 1985. Grosslight K, Foster R, Colohan AR, et al: Isoflurane for neuro-anesthesia: Risk factors for increases in intracranial pressure.  Anesthesiology  1985; 63:533.

Guy et al., 1997. Guy J, Hindman BJ, Baker KZ, et al: Comparison of remifentanil and fentanyl in patients undergoing craniotomy for supratentorial space-occupying lesions.  Anesthesiology  1997; 86:514.

Hansen et al., 1984. Hansen NB, Brubakk AM, Brattid D, et al: The effect of variations in PaCO2 on brain blood flow and cardiac output in the newborn piglet.  Pediatr Res  1984; 18:1132.

Harris et al., 1986. Harris MH, Strafford MA, Rowe RW, et al: Venous air embolism and cardiac arrest during craniectomy in a supine infant.  Anesthesiology  1986; 65:547.

Harris et al., 1987. Harris MH, Yemen TA, Strafford MA, et al: Venous embolism during craniectomy in supine infants.  Anesthesiology  1987; 67:816.

Harrison et al., 2002. Harrison EA, Mackersie A, McEwan A, et al: The sitting position for neurosurgery in children: A review of 16 years' experience.  Br J Anaesth  2002; 88:12.

Hatashita et al., 1986. Hatashita S, Hoff JT, Isbii S: Focal brain edema associated with acute hypertension.  J Neurosurg  1986; 64:643.

Hayashi, 2000. Hayashi N: [Brain hypothermia treatment for the management of severe pediatric brain injury].  No To Hattatsu  2000; 32:122.

Henderson and Irwin, 1995. Henderson MA, Irwin MG: Anaesthesia and moyamoya disease.  Anaesth Intensive Care  1995; 23:503.

Henriksen and Jorgensen, 1973. Henriksen HT, Jorgensen PB: The effect of nitrous oxide on intracranial pressure in patients with intracranial disorders.  Br J Anaesth  1973; 45:486.

Hepner and Castells, 2003. Hepner DL, Castells MC: Latex allergy: An update.  Anesth Analg  2003; 96:1219.

Hernandez et al., 1980. Hernandez MJ, Brennan RW, Bowman GS: Autoregulation of cerebral blood flow in the newborn dog.  Brain Res  1980; 184:199.

Hernandez-Palazon et al., 2001. Hernandez-Palazon J, Tortosa JA, Martinez-Lage JF, et al: Rocuronium-induced neuromuscular blockade is affected by chronic phenytoin therapy.  J Neurosurg Anesthesiol  2001; 13:79.

Hernandez-Palazon et al., 2003. Hernandez-Palazon J, Martinez-Lage JF, de la Rosa-Carrillo VN, et al: Anesthetic technique and development of pneumocephalus after posterior fossa surgery in the sitting position.  Neurocirugia (Astur)  2003; 14:216.

Herrick et al., 1991. Herrick IA, Gelb AW, Manninen PH, et al: Effects of fentanyl, sufentanil, and alfentanil on brain retractor pressure.  Anesth Analg  1991; 72:359.

Hillman et al., 1987. Hillman DR, Rung GW, Thompson WR, et al: The effect of bupivacaine scalp infiltration on the hemodynamic response to craniotomy under general anesthesia.  Anesthesiology  1987; 67:1001.

Hochleitner et al., 2001. Hochleitner BW, Menardi G, Haussler B, et al: Spina bifida as an independent risk factor for sensitization to latex.  J Urol  2001; 166:2370.

Hoffman et al., 1993. Hoffman WE, Cunningham F, James MK, et al: Effects of remifentanil, a new short-acting opioid, on cerebral blood flow, brain electrical activity, and intracranial pressure in dogs anesthetized with isoflurane and nitrous oxide.  Anesthesiology  1993; 79:107.

Hoffman et al., 1983. Hoffman WE, Miletich DJ, Albrecht RF: Cerebrovascular response to hypotension in hypertensive rats: Effect of antihypertensive therapy.  Anesthesiology  1983; 58:326.

Horn et al., 2004. Horn P, Pfister S, Bueltmann E, et al: Moyamoya-like vasculopathy (moyamoya syndrome) in children.  Childs Nerv Syst  2004; 20:382.

Hougaard et al., 1974. Hougaard K, Hansen A, Brodersen P: The effect of ketamine on regional cerebral blood flow in man.  Anesthesiology  1974; 41:562.

Iacopino et al., 2003. Iacopino DG, Conti A, Battaglia C, et al: Transcranial Doppler ultrasound study of the effects of nitrous oxide on cerebral autoregulation during neurosurgical anesthesia: A randomized controlled trial.  J Neurosurg  2003; 99:58.

Inamasu et al., 2002. Inamasu J, Ichikizaki K, Matsumoto S, et al: Mild hypothermia for hemispheric cerebral infarction after evacuation of an acute subdural hematoma in an infant.  Childs Nerv Syst  2002; 18:175.

Ishikawa et al., 1983. Ishikawa T, Funatsu N, Okamoto K, et al: Blood-brain barrier dysfunction following drug-induced hypotension in the dog.  Anesthesiology  1983; 59:526.

Ishikawa and McDowall, 1980. Ishikawa T, McDowall DG: Electrical activity of the cerebral cortex during induced hypotension with sodium nitroprusside and trimethaphan in the cat.  Br J Anaesth  1980; 53:605.

Iwatsuki et al., 1980. Iwatsuki N, Kuroda N, Keisuke A, et al: Succinylcholine-induced hyperkalemia in patients with ruptured cerebral aneurysms.  Anesthesiology  1980; 53:64.

Jacob and Kausalya, 1990. Jacob R, Kausalya R: Moyamoya disease.  Anaesth Intensive Care  1990; 18:582.

Jelsma and McQueen, 1967. Jelsma LF, McQueen JD: Effect of experimental water restriction on brain water.  J Neurosurg  1967; 26:35.

Jennett et al., 1969. Jennett WB, Barker J, Fitch W, et al: Effect of anaesthesia on intracranial pressure in patients with space-occupying lesions.  Lancet  1969; 1:61.

Jobes et al., 1977. Jobes DR, Kennell EM, Bush GL, et al: Cerebral blood flow and metabolism during morphine-nitrous oxide anesthesia in man.  Anesthesiology  1977; 47:16.

Joseph et al., 1985. Joseph MM, Leopold GA, Carillo JF: Venous air embolism during repair of craniosynostosis in the lateral position.  Anesth Rev  1985; 12:46.

Kalkman et al., 1994. Kalkman CJ, Drummond JC, Patel PM, et al: Effects of droperidol, pentobarbital, and ketamine on myogenic transcranial magnetic motor-evoked responses in humans.  Neurosurgery  1994; 35:1066.

Kalkman et al., 1991. Kalkman CJ, Drummond JC, Ribberink AA: Low concentrations of isoflurane abolish motor evoked responses to transcranial electrical stimulation during nitrous oxide/opioid anesthesia in humans.  Anesth Analg  1991; 73:410.

Kamath et al., 2004. Kamath BM, Spinner NB, Emerick KM, et al: Vascular anomalies in Alagille syndrome: A significant cause of morbidity and mortality.  Circulation  2004; 109:1354.

Kang and Lee, 1999. Kang JK, Lee KS: Comparison between pediatric and adult neurosurgery: Management and future perspectives. Tethered cord syndrome, hydrocephalus, craniosynostosis.  Childs Nerv Syst  1999; 15:795.

Kansha et al., 1997. Kansha M, Irita K, Takahashi S, et al: Anesthetic management of children with moyamoya disease.  Clin Neurol Neurosurg  1997; 99(suppl 2):S110.

Karsli et al., 2004. Karsli C, Luginbuehl I, Bissonnette B: The cerebrovascular response to hypocapnia in children receiving propofol.  Anesth Analg  2004; 99:1049.

Kasoff et al., 1972. Kasoff SS, Zingesser LH, Shulman K: Compartmental abnormalities of regional cerebral blood flow in children with head trauma.  J Neurosurg  1972; 36:463.

Keller et al., 2001. Keller MS, Fendya DG, Weber TR: Glasgow Coma Scale predicts coagulopathy in pediatric trauma patients.  Semin Pediatr Surg  2001; 10:12.

Kelly et al., 1994. Kelly KJ, Pearson ML, Kurup VP, et al: A cluster of anaphylactic reactions in children with spina bifida during general anesthesia: Epidemiologic features, risk factors, and latex hypersensitivity.  J Allergy Clin Immunol  1994; 94:53.

Kennedy and Sokoloff, 1967. Kennedy C, Sokoloff L: An adaptation of the nitrous oxide method to the study of the circulation in children; normal values for cerebral blood flow and metabolic rate in childhood.  J Clin Invest  1967; 36:1130.

Kim et al., 2004. Kim SK, Seol HJ, Cho BK, et al: Moyamoya disease among young patients: Its aggressive clinical course and the role of active surgical treatment.  Neurosurgery  2004; 54:840.

Klimscha et al., 2003. Klimscha W, Ullrich R, Nasel C, et al: High-dose remifentanil does not impair cerebrovascular carbon dioxide reactivity in healthy male volunteers.  Anesthesiology  2003; 99:834.

Kraus et al., 1987. Kraus JF, Fife D, Conroy C: Pediatric brain injuries: The nature, clinical course, and early outcomes in a defined United States population.  Pediatrics  1987; 79:501.

Kreisman et al., 1989. Kreisman NR, Olson JE, Home DS, et al: Cerebral oxygenation and blood flow in infant and young adult rats.  Am J Physiol  1989; 256:R78.

Kurehara et al., 1993. Kurehara K, Ohnishi H, Touho H, et al: Cortical blood flow response to hypercapnia during anaesthesia in moyamoya disease.  Can J Anaesth  1993; 40:709.

Kurup and Fink, 2001. Kurup VP, Fink JN: The spectrum of immunologic sensitization in latex allergy.  Allergy  2001; 56:2.

Lagace et al., 2004. Lagace A, Karsli C, Luginbuehl I, et al: The effect of remifentanil on cerebral blood flow velocity in children anesthetized with propofol.  Paediatr Anaesth  2004; 14:861.

Lanier et al., 1987. Lanier W, Stangland KJ, Scheithauer BW, et al: The effects of dextrose infusion and head position on neurologic outcome after complete cerebral ischemia in primates: Examination of a model.  Anesthesiology  1987; 66:39.

Lanier et al., 1986. Lanier WL, Milde JH, Michenfelder JD: Cerebral stimulation following succinylcholine in dogs.  Anesthesiology  1986; 64:551.

Lanier et al., 1985. Lanier WL, Milde JH, Michenfelder JD: The cerebral effects of pancuronium and atracurium in halothane-anesthetized dogs.  Anesthesiology  1985; 63:589.

Lassen and Christensen, 1976. Lassen NA, Christensen MS: Physiology of cerebral blood flow.  Br J Anaesth  1976; 48:719.

Laughlin and Newberg, 1985. Laughlin TP, Newberg LA: Prolonged myoclonus after etomidate anesthesia.  Anesth Analg  1985; 64:80.

Lennon et al., 1986. Lennon RL, Olson RA, Gronert GA: Atracurium or vecuronium for rapid sequence endotracheal intubation.  Anesthesiology  1986; 64:510.

Lo et al., 2002. Lo L, Singer ST, Vichinsky E: Pancytopenia induced by hypothermia.  J Pediatr Hematol Oncol  2002; 24:681.

Lockhart and Jenkins, 1972. Lockhart CH, Jenkins JJ: Ketamine-induced apnea in patients with increased intracranial pressure.  Anesthesiology  1972; 376:92.

Lorenz et al., 2003. Lorenz IH, Kolbitsch C, Hinteregger M, et al: Remifentanil and nitrous oxide reduce changes in cerebral blood flow velocity in the middle cerebral artery caused by pain.  Br J Anaesth  2003; 90:296.

Lorenz et al., 2002. Lorenz IH, Kolbitsch C, Hormann C, et al: The influence of nitrous oxide and remifentanil on cerebral hemodynamics in conscious human volunteers.  Neuroimage  2002; 17:1056.

Losasso et al., 1992a. Losasso TJ, Black S, Muzzi DA, et al: Detection and hemodynamic consequences of venous air embolism. Does nitrous oxide make a difference?.  Anesthesiology  1992; 77:148.

Losasso et al., 1992b. Losasso TJ, Muzzi DA, Dietz NM, et al: Fifty percent nitrous oxide does not increase the risk of venous air embolism in neurosurgical patients operated upon in the sitting position.  Anesthesiology  1992; 77:21.

Losee et al., 1982. Losee JM, Sherrill D, Virtue RW, et al: Quantitative detection of venous air embolism in the dog by mass spectrometry measurement of end tidal nitrogen.  Anesthesiology  1982; 57:A146.

Lou et al., 1979. Lou HC, Lassen NA, Friis-Hansen B: Impaired autoregulation of cerebral blood flow in the distressed newborn infant.  J Pediatr  1979; 94:118.

Lutz et al., 1990. Lutz LJ, Milde JH, Milde LN: The cerebral functional, metabolic, and hemodynamic effects of desflurane in dogs.  Anesthesiology  1990; 73:125.

Lutz et al., 1991. Lutz LJ, Milde JH, Milde LN: The response of the canine cerebral circulation to hyperventilation during anesthesia with desflurane.  Anesthesiology  1991; 74:504.

Madsen et al., 1987. Madsen JB, Cold GE, Hansen ES, et al: The effect of isoflurane on cerebral blood flow and metabolism in humans during craniotomy for small supratentorial cerebral tumors.  Anesthesiology  1987; 66:332.

Maekawa et al., 1979. Maekawa T, McDowall DG, Okuda Y: Brain-surface oxygen tension and cerebral cortical blood flow during hemorrhage and drug-induced hypotension in the cat.  Anesthesiology  1979; 51:313.

Magness et al., 1973. Magness A, Yashon D, Locke G, et al: Cerebral function during trimethaphan-induced hypotension.  Neurology  1973; 23:506.

Mammoto et al., 1998. Mammoto T, Hayashi Y, Ohnishi Y, et al: Incidence of venous and paradoxical air embolism in neurosurgical patients in the sitting position: Detection by transesophageal echocardiography.  Acta Anaesthesiol Scand  1998; 42:643.

Mann et al., 1979. Mann JD, Mann ES, Cookson SL: Differential effects of pentobarbital, ketamine hydrochloride, and enflurane anesthesia on CSF formation rate and outflow resistance in the rat.  Neurosurgery  1979; 4:482.

Manohar and Parks, 1984. Manohar M, Parks C: Regional distribution of brain and myocardial perfusion in swine while awake and during 1.0 and 1.5 MAC isoflurane anesthesia produced without and with 50% nitrous oxide.  Cardiovasc Res  1984; 18:344.

Marion, 2002. Marion DW: Moderate hypothermia in severe head injuries: The present and the future.  Curr Opin Crit Care  2002; 8:111.

Maroon and Albin, 1974. Maroon JC, Albin MS: Air embolism diagnosed by Doppler ultrasound.  Anesth Analg  1974; 53:399.

Maroon et al., 1969. Maroon JC, Edmonds-Seal J, Campbell RL: An ultrasonic method for detecting air embolism.  J Neurosurg  1969; 31:196.

Marsh et al., 1980. Marsh ML, Dunlop BJ, Shapiro HM, et al: Succinylcholine intracranial pressure effects in neurosurgical patients.  Anesth Analg  1980; 59:550.

Marsh et al., 1979. Marsh ML, Shapiro HM, Smith RW, et al: Changes in neurological status and intracranial pressure associated with sodium nitroprusside administration.  Anesthesiology  1979; 51:336.

Marshall et al., 1978. Marshall LF, Smith RW, Rauscher LA, et al: Mannitol dose requirements in brain-injured patients.  J Neurosurg  1978; 48:169.

Marshall et al., 1984. Marshall WK, Bedford RF, Miller ED: Cardiovascular responses in the seated position-impact of four anesthetic techniques.  Anesth Analg  1984; 62:648.

Martin, 1970. Martin JT: Neuroanesthetic adjuncts for patients in the sitting position. III. Intravascular electrocardiography.  Anesth Analg  1970; 49:793.

Martin et al., 1984. Martin RW, Ashleman B, Colley PS: Effects of cardiac output on the clearance of air emboli from the superior vena cava.  Anesthesiology  1984; 60:580.

Martino and Werner, 1991. Martino JD, Werner LO: Hypocarbia during anaesthesia in children with moyamoya disease.  Can J Anaesth  1991; 38:942.

Matjasko et al., 1985. Matjasko J, Petrozza P, Mackenzie CF: Sensitivity of end-tidal nitrogen in venous air embolism detection in dogs.  Anesthesiology  1985; 63:418.

Mayberg et al., 1993. Mayberg TS, Lam AM, Eng CC, et al: The effect of alfentanil on cerebral blood flow velocity and intracranial pressure during isoflurane-nitrous oxide anesthesia in humans.  Anesthesiology  1993; 78:288.

Mazze et al., 1969. Mazze RI, Escue HM, Houston JB: Hyperkalemia and cardiovascular collapse following administration of succinylcholine to the traumatized patient.  Anesthesiology  1969; 31:540.

Mazzoni et al., 1985. Mazzoni P, Giffin JP, Cottrell JL, et al: Intracranial pressure during diltiazem-induced hypotension in anesthetized dogs.  Anesth Analg  1985; 64:1001.

McDowall, 1985. McDowall DG: Induced hypotension and brain ischaemia.  Br J Anaesth  1985; 57:110.

McPherson et al., 1989. McPherson RW, Briar JE, Traystman RJ: Cerebrovascular responsiveness to carbon dioxide in dogs with 1.4% and 2.8% isoflurane.  Anesthesiology  1989; 70:843.

McPherson and Traystman, 1988. McPherson RW, Traystman RJ: Effects of isoflurane on cerebral autoregulation in dogs.  Anesthesiology  1988; 69:493.

Mehta et al., 1984. Mehta MP, Sokoll MD, Gergis DS: Effects of venous air embolism on the cardiovascular system and acid base balance in the presence and absence of nitrous oxide.  Acta Anaesthesiol Scand  1984; 28:226.

Meridy et al., 1974. Meridy HW, Creighton RE, Humphreys RP: Complications during neurosurgery in the prone position in children.  Can Anaesth Soc J  1974; 21:445.

Messer et al., 1989. Messer J, Haddad J, Bientz J, et al: Influence of anesthetics on cerebral blood flow velocity in infancy. Effects of halothane versus thiopental-fentanyl.  Dev Pharmacol Ther  1989; 13:145.

Meyer et al., 1999. Meyer P, Legros C, Orliaguet G: Critical care management of neurotrauma in children: New trends and perspectives.  Childs Nerv Syst  1999; 15:732.

Michenfelder et al., 1969. Michenfelder JD, Martin JT, Altenburg BM, et al: Air embolism during neurosurgery: An evaluation of right-atrial catheters for diagnosis and treatment.  JAMA  1969; 208:1353.

Michenfelder et al., 1972. Michenfelder JD, Miller RH, Gronert GA: Evaluation of an ultrasonic device (Doppler) for the diagnosis of venous air embolism.  Anesthesiology  1972; 36:164.

Michenfelder et al., 1966. Michenfelder JD, Terry HR, Daw EF, et al: Air embolism during neurosurgery: A new method of treatment.  Anesth Analg  1966; 45:390.

Michenfelder and Theye, 1977. Michenfelder JD, Theye RA: Canine systemic and cerebral effects of hypotension induced by hemorrhage, trimethaphan, halothane, or nitroprusside.  Anesthesiology  1977; 46:188.

Michenfelder and Theye, 1971. Michenfelder JD, Theye RA: Effect of fentanyl, droperidol and Innovar on canine cerebral metabolism and blood flow.  Br J Anaesth  1971; 43:630.

Michenfelder, 1974. Michenfelder JD: The interdependency of cerebral function and metabolic effects following massive doses of thiopental in the dog.  Anesthesiolog y  1974; 41:231.

Milde et al., 1990. Milde LN, Milde JH, Gallagher WJ: Effects of sufentanil on cerebral circulation and metabolism in dogs.  Anesth Analg  1990; 70:138.

Miletich et al., 1976. Miletich DJ, Ivankovich AD, Albrecht RF, et al: Absence of auto-regulation of cerebral blood flow during halothane and enflurane anesthesia.  Anesth Analg  1976; 55:100.

Millar et al., 1994. Millar C, Bissonnette B, Humphreys RP: Cerebral arteriovenous malformations in children.  Can J Anaesth  1994; 41:321.

Millar, 1972. Millar RA: Neuroanaesthesia in the sitting position.  Br J Anaesth  1972; 44:495.

Miller et al., 1975. Miller R, Tausk HC, Stark DCC: Effect of Innovar, fentanyl and droperidol on the cerebrospinal fluid pressure in neurosurgical patients.  Can Anaesth Soc J  1975; 22:502.

Milligan, 1980. Milligan DWA: Failure of autoregulation and intraventricular haemorrhage in preterm infants.  Lancet  1980; 1:896.

Minto et al., 2003. Minto CF, Schnider TW, Gregg KM, et al: Using the time of maximum effect site concentration to combine pharmacokinetics and pharmacodynamics.  Anesthesiology  2003; 99:324.

Minto et al., 1997. Minto CF, Schnider TW, Shafer SL: Pharmacokinetics and pharmacodynamics of remifentanil. II. Model application.  Anesthesiology  1997; 86:24.

Minton et al., 1986. Minton MD, Grosslight K, Stirt JA, et al: Increases in intracranial pressure from succinylcholine: Prevention by prior nondepolarizing blockade.  Anesthesiology  1986; 65:165.

Minton et al., 1985. Minton MD, Stirt JA, Bedford RF: Increased intracranial pressure from; succinylcholine: Modification by prior nondepolarizing blockade.  Anesthesiology  1985; 63:A391.

Misfeldt et al., 1976. Misfeldt BB, Jorgensen PB, Spotoff H, et al: The effects of droperidol and fentanyl on intracranial pressure and cerebral perfusion pressure in neurosurgical patients.  Br J Anaesth  1976; 48:963.

Mochalova et al., 1983. Mochalova LD, Khodov DA, Zhukova TP: Cerebral circulation control in healthy full-term neonates.  Acta Paediatr Scand Suppl  1983; 311:20.

Mongan et al., 1992. Mongan P, Peterson RE, Culling RD: Pressure monitoring can accurately position catheters for air embolism aspiration.  J Clin Monit  1992; 8:121.

Monkhoff et al., 2001. Monkhoff M, Schwarz U, Gerber A, et al: The effects of sevoflurane and halothane anesthesia on cerebral blood flow velocity in children.  Anesth Analg  2001; 92:891.

Moorthy and Hilgenberg, 1980. Moorthy SS, Hilgenberg JC: Resistance to nondepolarizing muscle relaxants in paretic upper extremities of patient with residual hemiplegia.  Anesth Analg  1980; 59:624.

Moss and McDowall, 1979. Moss E, McDowall DG: ICP increases with 50% nitrous oxide in oxygen in severe head injuries during controlled ventilation.  Br J Anaesth  1979; 51:757.

Moss et al., 1979. Moss E, Powell D, Gibson RM, et al: Effect of etomidate on intracranial pressure and cerebral perfusion pressure.  Br J Anaesth  1979; 51:347.

Moss et al., 1978a. Moss E, Powell D, Gibson RM, et al: Effects of fentanyl on intracranial pressure and cerebral perfusion pressure during hypocapnia.  Br J Anaesth  1978; 50:779.

Moss et al., 1978b. Moss E, Powell D, Gibson RM, et al: Effects of tracheal intubation on intracranial pressure following induction of anaesthesia with thiopentone or althesin in patients undergoing neurosurgery.  Br J Anaesth  1978; 50:353.

Munro et al., 2004. Munro MJ, Walker AM, Parfield CP: Hypotensive extremely low birth weight infants have reduced cerebral blood flow.  Pediatrics  2004; 114:1591-1596.

Munson, 1971. Munson ES: Effect of nitrous oxide on the pulmonary circulation during venous air embolism.  Anesth Analg  1971; 50:785.

Nakahashi et al., 1991. Nakahashi K, Yomosa H, Matsuzawa N, et al: Effect on cerebral blood flow of midazolam during modified neurolept-anesthesia.  Masui  1991; 40:1787.

Namill et al., 1981. Namill JF, Bedford RF, Weaver DC, et al: Lidocaine before endotracheal intubation: Intravenous or laryngotracheal?.  Anesthesiology  1981; 55:578.

Nehls et al., 1987. Nehls DG, Todd MM, Spetzler RF, et al: A comparison of the cerebral protective effects of isoflurane and barbiturates during temporary focal ischemia in primates.  Anesthesiology  1987; 66:453.

Newberg and Michenfelder, 1983. Newberg LA, Michenfelder JD: Cerebral protection by isoflurane during hypoxemia or ischemia.  Anesthesiology  1983; 59:29.

Newberg et al., 1984. Newberg LA, Milde JH, Michenfelder JD: Systemic and cerebral effects of isoflurane-induced hypotension in dogs.  Anesthesiology  1984; 60:541.

Newberg et al., 1983. Newberg LA, Milde JH, Michenfelder JD: The cerebral metabolic effects of isoflurane at and above concentrations that suppress cortical electrical activity.  Anesthesiology  1983; 59:23.

Newfield and Hamid, 2001. Newfield P, Hamid RK: Pediatric neuro-anesthesia. Arteriovenous malformations.  Anesthesiol Clin North Am  2001; 19:229.

Nieto et al., 2002. Nieto A, Mazon A, Pamies R, et al: Efficacy of latex avoidance for primary prevention of latex sensitization in children with spina bifida.  J Pediatr  2002; 140:370.

Nishino et al., 1998. Nishino H, Kinouchi K, Fukumitsu K, et al: [Anesthesia and perioperative management in infants with Chiari type ii malformation].  Masui  1998; 47:982.

Nugent et al., 1982. Nugent M, Artru AA, Michenfelder JD: Cerebral metabolic, vascular and protective effects of midazolam maleate.  Anesthesiology  1982; 56:172.

Nussmeier et al., 1986. Nussmeier NA, Arlund C, Slogoff S: Neuropsychiatric complications after cardiopulmonary bypass. Cerebral protection by a barbiturate.  Anesthesiology  1986; 64:165.

Ong et al., 1986. Ong BY, Bose D, Palahniuk RJ: Acidemia impairs autoregulation of cerebral blood flow in newborn lambs.  Can Anaesth Soc J  1986; 33:5.

Ong et al., 1996. Ong L, Selladurai BM, Dhillon MK, et al: The prognostic value of the Glasgow Coma Scale, hypoxia and computerised tomography in outcome prediction of pediatric head injury.  Pediatr Neurosurg  1996; 24:285.

Orliaguet et al., 1998. Orliaguet GA, Meyer PG, Blanot S, et al: Predictive factors of outcome in severely traumatized children.  Anesth Analg  1998; 87:537.

Ornstein et al., 1987. Ornstein E, Matteo RS, Schwartz AE, et al: The effect of phenytoin on the magnitude and duration of neuromuscular block following atracurium or vecuronium.  Anesthesiology  1987; 67:191.

Ornstein et al., 1985. Ornstein E, Matteo RS, Young WL, et al: Resistance to metocurine-induced neuromuscular blockade in patients receiving phenytoin.  Anesthesiology  1985; 63:294.

Ornstein et al., 1993. Ornstein E, Young WL, Fleischer LH, et al: Desflurane and isoflurane have similar effects on cerebral blood flow in patients with intracranial mass lesions.  Anesthesiology  1993; 79:498.

Ostapkovich et al., 1998. Ostapkovich ND, Baker KZ, Fogarty-Mack P, et al: Cerebral blood flow and CO2 reactivity is similar during remifentanil/N2O and fentanyl/N2O anesthesia.  Anesthesiology  1998; 89:358.

Paris et al., 1998. Paris A, Scholz J, von Knobelsdorff G, et al: The effect of remifentanil on cerebral blood flow velocity.  Anesth Analg  1998; 87:569.

Patwardhan et al., 2000. Patwardhan RV, Stong B, Begin EM, et al: Efficacy of vagal nerve stimulation in children with medically refractory epilepsy.  Neurosurgery  2000; 47:1353.

Paul et al., 2003. Paul M, Dueck M, Joachim Herrmann H, et al: A randomized, controlled study of fluid management in infants and toddlers during surgery: Hydroxyethyl starch 6% (HES 70/0.5) vs lactated Ringer's solution.  Paediatr Anaesth  2003; 13:603.

Pearl and Larson, 1986. Pearl RG, Larson CP: Hemodynamic effects of positive end-expiratory pressure during continuous venous air embolism in the dog.  Anesthesiology  1986; 64:724.

Pelligrino et al., 1984. Pelligrino DA, Miletich DJ, Hoffman WE, et al: Nitrous oxide markedly increases cerebral cortical metabolic rate and blood flow in the goat.  Anesthesiology  1984; 60:405.

Perkins-Pearson et al., 1982. Perkins-Pearson NAK, Marshall WK, Bedford RF: Atrial pressures in the seated position: Implication for paradoxical air embolism.  Anesthesiology  1982; 57:493.

Petersen et al., 2003. Petersen KD, Landsfeldt U, Cold GE, et al: Intracranial pressure and cerebral hemodynamic in patients with cerebral tumors: A randomized prospective study of patients subjected to craniotomy in propofol-fentanyl, isoflurane-fentanyl, or sevoflurane-fentanyl anesthesia.  Anesthesiology  2003; 98:329.

Petty, 1993. Petty LA: Anesthetic management of a patient with moyamoya disease: A case report.  Aana J  1993; 61:277.

Phillips and Mulliken, 1988. Phillips RJ, Mulliken JB: Venous air embolism during a craniofacial procedure.  Plast Reconstr Surg  1988; 82:155.

Phirman and Shapiro, 1977. Phirman JR, Shapiro HM: Modification of nitrous oxide-induced intracranial hypertension by prior induction of anesthesia.  Anesthesiology  1977; 46:150.

Pierce et al., 1962. Pierce EC, Lambertsen CJ, Deutsch S, et al: Cerebral circulation and metabolism during thiopental anesthesia and hyperventilation in man.  J Clin Invest  1962; 41:1664.

Pigula et al., 1993. Pigula FA, Wald SL, Shackford SR, et al: The effect of hypotension and hypoxia on children with severe head injuries.  J Pediatr Surg  1993; 28:310.

Pollack, 1994. Pollack IF: Brain tumors in children (review).  N Engl J Med  1994; 331:1500.

Pollay et al., 1983. Pollay M, Fullenwider C, Roberts PA, et al: Effect of mannitol and furosemide on blood-brain osmotic gradient and intracranial pressure.  J Neurosurg  1983; 59:945.

Pollay and Roberts, 1980. Pollay M, Roberts PA: Blood-brain barrier: A definition of normal and altered function.  Neurosurgery  1980; 6:675.

Pryds, 1991. Pryds O: Control of cerebral circulation in the high-risk neonate.  Ann Neurol  1991; 30:321.

Pulsinelli et al., 1983. Pulsinelli WA, Levy DE, Sigsbee B, et al: Increased damage after ischemic stroke in patients with hyperglycemia with or without established diabetes mellitus.  Am J Med  1983; 74:540.

Pulsinelli et al., 1982. Pulsinelli WA, Waldman S, Rawlinson D, et al: Moderate hyperglycemia augments ischemia brain damage: A neuropathologic study in the rat.  Neurology  1982; 32:1239.

Purves and James, 1969. Purves MJ, James IM: Observations on the control of cerebral blood flow in sheep fetus and newborn lamb.  Circ Res  1969; 25:651.

Rampil et al., 1991. Rampil IJ, Lockhart SH, Eger EI, et al: The electroencephalographic effects of desflurane in humans.  Anesthesiology  1991; 74:434.

Reilly et al., 1988. Reilly PL, Simpson DA, Sprod R, et al: Assessing the conscious level in infants and young children: A paediatric version of the Glasgow Coma Scale.  Childs Nerv Syst  1988; 4:30.

Renou et al., 1978. Renou AM, Vernhiet J, Macrez P, et al: Cerebral blood flow and metabolism during etomidate anaesthesia in man.  Br J Anaesth  1978; 50:1047.

Ririe et al., 2003. Ririe DG, David LR, Glazier SS, et al: Surgical advancement influences perioperative care: A comparison of two surgical techniques for sagittal craniosynostosis repair.  Anesth Analg  2003; 97:699.

Rizzi, 1998. Rizzi RR: No difference between remifentanil and fentanyl in patients undergoing craniotomy.  Anesthesiology  1998; 88:271.

Roberts et al., 1987. Roberts PA, Pollay M, Engles C, et al: Effect on intracranial pressure of furosemide combined with varying doses and administration rates of mannitol.  J Neurosurg  1987; 66:440.

Robertson et al., 1983. Robertson CS, Cliftonn GL, Taylor AA, et al: Treatment of hypertension associated with head injury.  J Neurosurg  1983; 59:455.

Rockoff et al., 1980. Rockoff MA, Naughton KVH, Shapiro HM, et al: Cerebral, circulatory, and metabolic responses to intravenously administered lorazepam.  Anesthesiology  1980; 53:215.

Rogers et al., 1980. Rogers MC, Nugent SK, Traystman RJ: Control of cerebral circulation in the neonate and infant.  Crit Care Med  1980; 8:570.

Rosa et al., 1986. Rosa G, Orfei P, Sanfilippo M, et al: The effects of atracurium besylate (Tracrium) on intracranial pressure and cerebral perfusion pressure.  Anesth Analg  1986; 65:381.

Rosomoff and Holaday, 1954. Rosomoff HL, Holaday DA: Cerebral blood flow and cerebral oxygen consumption during hypothermia.  Am J Physiol  1954; 179:85.

Roth and Ebrahim, 1987. Roth S, Ebrahim ZY: Resistance to pancuronium in patients receiving carbamazepine.  Anesthesiology  1987; 66:691.

Rudehill et al., 1983. Rudehill A, Lagerkransen M, Lindquist C, et al: Effects of mannitol on blood volume and central hemodynamics in patients undergoing cerebral aneurysm surgery.  Anesth Analg  1983; 62:875.

Saidman and Eger, 1965. Saidman LJ, Eger II EI: Changes in cerebrospinal fluid pressure during pneumoencephalography under nitrous oxide anesthesia.  Anesthesiology  1965; 26:67.

Sakabe et al., 1978. Sakabe T, Kuramoto T, Inoue S, et al: Cerebral effects of nitrous oxide in the dog.  Anesthesiology  1978; 48:195.

Sakabe et al., 1976. Sakabe T, Kuramoto T, Kumagae S, et al: Cerebral responses to the addition of nitrous oxide to halothane in man.  Br J Anaesth  1976; 48:957.

Sakabe et al., 1974. Sakabe T, Maekawa T, Ishikawa T, et al: The effects of lidocaine on canine cerebral metabolism and circulation related to the electroencephalogram.  Anesthesiology  1974; 40:433.

Sala et al., 2002. Sala F, Krzan MJ, Deletis V: Intraoperative neurophysiological monitoring in pediatric neurosurgery: Why, when, how?.  Childs Nerv Syst  2002; 18:264.

Sala et al., 2004. Sala F, Lanteri P, Bricolo A: Motor evoked potential monitoring for spinal cord and brain stem surgery.  Adv Tech Stand Neurosurg  2004; 29:133.

Sari et al., 1972. Sari A, Okuda Y, Takeshita H: The effects of thalamonal on cerebral circulation and oxygen consumption in man.  Br J Anaesth  1972; 44:330.

Sato et al., 1999. Sato K, Shirane R, Kato M, et al: Effect of inhalational anesthesia on cerebral circulation in moyamoya disease.  J Neurosurg Anesthesiol  1999; 11:25.

Sato et al., 2001. Sato K, Shamoto H, Yoshimoto T: Severe bradycardia during epilepsy surgery.  J Neurosurg Anesthesiol  2001; 13:329.

Scheller et al., 1988. Scheller MS, Tateishi A, Drummond JC, Zornow MH: The effects of sevoflurane on cerebral blood flow, cerebral metabolic rate for oxygen, intracranial pressure, and the electroencephalogram are similar to those of isoflurane in the rabbit.  Anesthesiology  1988; 68:548.

Scheller et al., 1987. Scheller MS, Todd MM, Drummond JC, et al: The intracranial pressure effects of isoflurane and halothane administered following cryogenic brain injury in rabbits.  Anesthesiology  1987; 67:507.

Scheller et al., 1986. Scheller MS, Todd MM, Drummond JC: Isoflurane, halothane and regional cerebral blood flow at various levels of PaCO2 in rabbits.  Anesthesiology  1986; 64:598.

Schmieder et al., 2000. Schmieder K, Schregel W, Harders A, et al: Dynamic cerebral autoregulation in patients undergoing surgery for intracranial tumors.  Eur J Ultrasound  2000; 12:1.

Schwarz et al., 1994. Schwarz G, Fuchs G, Weihs W, et al: Sitting position for neurosurgery: Experience with preoperative contrast echocardiography in 301 patients.  J Neurosurg Anesthesiol  1994; 6:83.

Schwedler et al., 1982. Schwedler M, Miletich DJ, Albrecht RF: Cerebral blood flow and metabolism following ketamine administration.  Can Anaesth Soc J  1982; 29:222.

Servadei et al., 1998. Servadei F, Nasi MT, Cremonini AM, et al: Importance of a reliable admission Glasgow Coma Scale score for determining the need for evacuation of posttraumatic subdural hematomas: A prospective study of 65 patients.  J Trauma  1998; 44:868.

Settergren et al., 1980. Settergren G, Lindblad BS, Persson B: Cerebral blood flow and exchange on oxygen, glucose, ketone bodies, lactate, pyruvate and amino acids in anesthetized children.  Acta Paediatr Scand  1980; 69:457.

Shafer and Varvel, 1991. Shafer SL, Varvel JR: Pharmacokinetics, pharmacodynamics, and rational opioid selection.  Anesthesiology  1991; 74:53.

Shapiro et al., 1980. Shapiro HM, Greenberg JH, Van Horn Naughton K, et al: Heterogeneity of local cerebral blood flow-PaCO2 sensitivity in neonatal dogs.  J Appl Physiol  1980; 49:113.

Shapiro et al., 1972. Shapiro HM, Wyte SR, Harris AB: Ketamine anaesthesia in patients with intracranial pathology.  Br J Anaesth  1972; 44:1200.

Shapiro et al., 1982. Shapiro HM, Yoachim J, Marshall LF: Nitrous oxide challenge for detection of residual intravascular pulmonary gas following venous air embolism.  Anesth Analg  1982; 61:304.

Sharma and Tripathi, 1994. Sharma K, Tripathi M: Detection of site of air entry in venous air embolism: Role of Valsalva maneuver.  J Neurosurg Anesthesiol  1994; 6:209.

Sharples and Eyre, 1991. Sharples PM, Eyre JA: Head injury—how community paediatricians can help.  Arch Dis Child  1991; 66:908.

Shayevitz and Matteo, 1985. Shayevitz JR, Matteo RS: Decreased sensitivity to metocurine in patients with upper motoneuron disease.  Anesth Analg  1985; 64:767.

Shenkin et al., 1976. Shenkin HA, Bezier HS, Bouzarth WF: Restricted fluid intake: Rational management of the neurosurgical patient.  J Neurosurg  1976; 45:432.

Sihle-Wissel et al., 2000. Sihle-Wissel M, Scholz M, Cunitz G: Transcranial magnetic-evoked potentials under total intravenous anaesthesia and nitrous oxide.  Br J Anaesth  2000; 85:465.

Sinha et al., 2004. Sinha PK, Neema PK, Manikandan S, et al: Bradycardia and sinus arrest following saline irrigation of the brain during epilepsy surgery.  J Neurosurg Anesthesiol  2004; 16:160.

Skahen et al., 1986. Skahen S, Shapiro HM, Drummond JC, et al: Nitrous oxide withdrawal reduces intracranial pressure in the presence of pneumocephalus.  Anesthesiology  1986; 65:192.

Smith et al., 1974. Smith A, Hoff J, Nielsen S, et al: Barbiturate protection in acute focal cerebral ischemia.  Stroke  1974; 5:1.

Smith and Macque, 1976. Smith AL, Macque JJ: Anesthetics and cerebral edema.  Anesthesiology  1976; 45:64.

Smith and Grenvik, 1970. Smith RB, Grenvik A: Cardiac arrest following succinylcholine in patients with central nervous system injuries.  Anesthesiology  1970; 33:558.

Smyth et al., 2003. Smyth MD, Tubbs RS, Bebin EM, et al: Complications of chronic vagus nerve stimulation for epilepsy in children.  J Neurosurg  2003; 99:500.

Soriano et al., 2002. Soriano SG, Eldredge EA, Rockoff MA: Pediatric neuro-anesthesia.  Anesthesiol Clin North Am  2002; 20:389.

Soriano et al., 2002. Soriano SG, McCann ME, Laussen PC: Neuro-anesthesia. Innovative techniques and monitoring.  Anesthesiol Clin North Am  2002; 20:137.

Soriano et al., 1996. Soriano SG, McManus ML, Sullivan LJ, et al: Cerebral blood flow velocity after mannitol infusion in children.  Can J Anaesth  1996; 43:461.

Soriano et al., 1994. Soriano SG, McManus ML, Sullivan LJ, et al: Doppler sensor placement during neurosurgical procedures for children in the prone position.  J Neurosurg Anesthesiol  1994; 6:153.

Soriano et al., 1993. Soriano SG, Sethna NF, Scott RM: Anesthetic management of children with moyamoya syndrome.  Anesth Analg  1993; 77:1066.

Spacek et al., 1997. Spacek A, Neiger FX, Spiss CK, et al: Atracurium-induced neuromuscular block is not affected by chronic anticonvulsant therapy with carbamazepine.  Acta Anaesthesiol Scand  1997; 41:1308.

Spacek et al., 1996. Spacek A, Neiger FX, Spiss CK, et al: Chronic carbamazepine therapy does not influence mivacurium-induced neuromuscular block.  Br J Anaesth  1996; 77:500.

Spacek et al., 1999. Spacek A, Nickl S, Neiger FX, et al: Augmentation of the rocuronium-induced neuromuscular block by the acutely administered phenytoin.  Anesthesiology  1999; 90:1551.

Sperry et al., 1992. Sperry RJ, Bailey PL, Reichman MV, et al: Fentanyl and sufentanil increase intracranial pressure in head trauma patients.  Anesthesiology  1992; 77:416.

Spetzler, 2004. Spetzler RF: Moyamoya.  J Neurosurg Spine  2004; reply 541

Steinberg et al., 2004. Steinberg GK, Ogilvy CS, Shuer LM, et al: Comparison of endovascular and surface cooling during unruptured cerebral aneurysm repair.  Neurosurgery  2004; 55:307.

Stephan et al., 1991. Stephan H, Groger P, Weyland A, et al: Einflu von Sufentanil aug Hirndurchblutung, Hirnstoffwechsel und die CO2-Reaktivitat der menschlichen Hirngefaße.  Anaesthesist  1991; 40:153.

Stevenson and Birch, 1979. Stevenson PH, Birch AA: Succinylcholine-induced hyperkalemia in a patient with a closed head injury.  Anesthesiology  1979; 51:89.

Stirt et al., 1987a. Stirt JA, Grosslight KR, Bedford RF, et al: “Defasciculation” with metocurine prevents succinylcholine-induced increases in intracranial pressure.  Anesthesiology  1987; 67:50.

Stirt et al., 1987b. Stirt JA, Maggio W, Haworth C, et al: Vecuronium: Effect on intracranial pressure and hemodynamics in neurosurgical patients.  Anesthesiology  1987; 67:570.

Stone et al., 1970. Stone WA, Beach TP, Hamelberg W: Succinylcholine: Danger in the spinal-cord-injured patient.  Anesthesiology  1970; 32:168.

Stoyka and Schultz, 1975. Stoyka WW, Schultz H: The cerebral response to sodium nitroprusside and trimethaphan controlled hypotension.  Can Anaesth Soc J  1975; 22:275.

Strahm et al., 2003. Strahm C, Min K, Boos N, et al: Reliability of perioperative SSEP recordings in spine surgery.  Spinal Cord  2003; 41:483.

Sumikawa and Nagai, 1983. Sumikawa K, Nagai H: Moyamoya disease and anesthesia.  Anesthesiology  1983; 58:204.

Sutton et al., 2001. Sutton LN, Sun P, Adzick NS: Fetal neurosurgery.  Neurosurgery  2001; 48:124.

Suzuki and Takaku, 1969. Suzuki J, Takaku A: Cerebrovascular “moyamoya” disease. Disease showing abnormal net-like vessels in base of brain.  Arch Neurol  1969; 20:288.

Takeshita et al., 1972. Takeshita H, Okuda Y, Sari A: The effects of ketamine on cerebral circulation and metabolism in man.  Anesthesiology  1972; 36:69.

Tarkkanen et al., 1974. Tarkkanen L, Laitinen L, Johanssen G: Effects of d-tubocurarine on intracranial pressure and thalamic electrical impedance.  Anesthesiology  1974; 40:247.

Tateishi et al., 1981. Tateishi A, Maekawa T, Takeshita H, et al: Diazepam and intracranial pressure.  Anesthesiology  1981; 54:335.

Tatman et al., 1997. Tatman A, Warren A, Williams A, et al: Development of a modified paediatric coma scale in intensive care clinical practice.  Arch Dis Child  1997; 77:519.

Theye and Michenfelder, 1968. Theye RA, Michenfelder JD: The effect of nitrous oxide on canine cerebral metabolism.  Anesthesiology  1968; 29:1113.

Thiagarajah et al., 1985. Thiagarajah S, Sophie S, Azar I, et al: Effect of succinylcholine on the ICP of cats with and without thiopental pretreatment.  Anesthesiology  1985; 63:A392.

Thomas, 1969. Thomas ET: Circulatory collapse following succinylcholine: Report of a case.  Anesth Analg  1969; 48:333.

Tobey, 1970. Tobey RE: Paraplegia, succinylcholine, and cardiac arrest.  Anesthesiology  1970; 32:359.

Tobias and Jimenez, 1997. Tobias JD, Jimenez DF: Anaesthetic management during awake craniotomy in a 12-year-old boy.  Paediatr Anaesth  1997; 7:341.

Todd and Drummond, 1984. Todd MM, Drummond JC: A comparison of the cerebrovascular and meta-bolic effects of halothane and isoflurane in the cat.  Anesthesiology  1984; 60:276.

Todd et al., 1984. Todd MM, Tommasino C, Moore S, et al: The effects of acute isovolemic hemodilution on the brain: A comparison of crystalloid and colloid solutions.  Anesthesiology  1984; 61:A122.

Todd, 1987. Todd MM: The effects of PaCO2 on the cerebrovascular response to nitrous oxide in the halothane-anesthetized rabbit.  Anesth Analg  1987; 66:1090.

Tong, 1987. Tong TK: Succinylcholine-induced hyperkalemia in near-drowning.  Anesthesiology  1987; 66:720.

Toung et al., 1988. Toung TJK, Miyabe M, McShane AJ, et al: Effect of PEEP and jugular venous compression on canine cerebral blood flow and oxygen consumption in the head elevated position.  Anesthesiology  1988; 68:53.

Toung et al., 1986. Toung TSK, McPherson RW, Donham RT, et al: Pneumocephalus: Effects of patient position on the incidence and location of aerocele after posterior fossa and upper cervical cord surgery.  Anesth Analg  1986; 65:65.

Turner et al., 1977. Turner JM, Powell D, Gibson RM, et al: Intracranial pressure changes in neurosurgical patients during hypotension induced with sodium nitroprusside or trimethaphan.  Br J Anaesth  1977; 49:419.

Ubags et al., 1997. Ubags LH, Kalkman CJ, Been HD, et al: The use of ketamine or etomidate to supplement sufentanil/N2O anesthesia does not disrupt monitoring of myogenic transcranial motor evoked responses.  J Neurosurg Anesthesiol  1997; 9:228.

Van Hemelrijck et al., 1990. Van Hemelrijck J, Fitch W, Mattheussen M, et al: Effect of propofol on cerebral circulation and autoregulation in the baboon.  Anesth Analg  1990; 71:49.

Verhaegen et al., 1992. Verhaegen MJ, Todd MM, Warner DS: A comparison of cerebral ischemic flow thresholds during halothane/N2O and isoflurane/N2O anesthesia in rats.  Anesthesiology  1992; 76:743.

Vesely et al., 1987. Vesely R, Hoffman WE, Gil KSL, et al: The cerebrovascular effects of curare and histamine in the rat.  Anesthesiology  1987; 66:519.

VonAken et al., 1982. VonAken H, Puchstein C, Schweppe ML, et al: Effect of labetalol on intracranial pressure in dogs with and without intracranial hypertension.  Acta Anaesthesiol Scand  1982; 26:615.

Wagner et al., 2001. Wagner KJ, Willoch F, Kochs EF, et al: Dose-dependent regional cerebral blood flow changes during remifentanil infusion in humans: A positron emission tomography study.  Anesthesiology  2001; 94:732.

Walker, 1976. Walker MD: Diagnosis and treatment of brain tumors.  Pediatr Clin North Am  1976; 23:131.

Warner et al., 1991. Warner C, Hoffman WE, Baughman VL, et al: Effects of sufentanil on cerebral blood flow, cerebral blood flow velocity, and metabolism in dogs.  Anesth Analg  1991; 72:177.

Warner et al., 1990. Warner C, Hoffman WE, Segil IJ, et al: Propofol decreases cerebral and spinal cord blood flow and maintains autoregulation in rats.  J Neurosurg Anesthesiol  1990; 2:220.

Warner et al., 1996. Warner DS, Hindman BJ, Todd MM, et al: Intracranial pressure and hemodynamic effects of remifentanil versus alfentanil in patients undergoing supratentorial craniotomy.  Anesth Analg  1996; 83:348.

Weed and McKibben, 1916. Weed LH, McKibben PS: Pressure changes in the cerebrospinal fluid following intravenous injection of solutions of various concentrations.  Am J Physiol  1916; 48:512.

Weinstabl et al., 1991. Weinstabl C, Mayer N, Richling B, et al: Effect of sufentanil on intracranial pressure in neurosurgical patients.  Anaesthesia  1991; 46:837.

Weinstabl et al., 1992. Weinstabl C, Mayer N, Spiss CK: Sufentanil decreases cerebral blood flow velocity in patients with elevated intracranial pressure.  Eur J Anaesthesiol  1992; 9:481.

Wilkinson, 1981. Wilkinson HA: Intracranial pressure monitoring: Techniques and pitfalls.   In: Cooper PR, ed. Head injury,  Baltimore: Williams & Wilkins; 1981.

Wollman et al., 1964. Wollman H, Alexander SC, Cohen PJ, et al: Cerebral circulation of man during halothane anesthesia.  Anesthesiology  1964; 25:180.

Wood et al., 1983. Wood JH, Simeone FA, Fink EA, et al: Hypervolemic hemodilution in experimental focal cerebral ischemia.  J Neurosurg  1983; 59:500.

Woolfenden et al., 1999. Woolfenden AR, Albers GW, Steinberg GK, et al: Moyamoya syndrome in children with Alagille syndrome: Additional evidence of a vasculopathy.  Pediatrics  1999; 103:505.

Yamamoto et al., 1997. Yamamoto M, Aoyagi M, Tajima S, et al: Increase in elastin gene expression and protein synthesis in arterial smooth muscle cells derived from patients with moyamoya disease.  Stroke  1997; 28:1733.

Yaster et al., 1987. Yaster M, Koehler RC, Traystman RJ: Effects of fentanyl on peripheral and cerebral hemodynamics in neonatal lambs.  Anesthesiology  1987; 66:524.

Young, 1992. Young WL: Effects of desflurane on the central nervous system.  Anesth Analg  1992; 75:S32.

Youngs and Shafer, 1994. Youngs EJ, Shafer SL: Pharmacokinetic parameters relevant to recovery from opioids.  Anesthesiology  1994; 81:833.

Zornow et al., 1987. Zornow MH, Todd MM, Moore SS: The acute cerebral effects of changes in plasma osmolality and oncotic pressure.  Anesthesiology  1987; 67:936.