Pediatric neurosurgical patients present a unique set of anesthetic challenges that include treatment of increased intracranial pressure (ICP), the frequent use of lateral and prone positions, and the anesthetic implications of neurophysiologic monitoring. These and other considerations for the neurosurgical patient are addressed in the first part of this chapter, followed by a discussion of anesthetic techniques for common pediatric neurosurgical procedures.
Pathophysiology and Treatment of Increased ICP
It is important to understand the pathophysiology of increased ICP in the child. The rigid cranial vault contains 80% brain tissue, 10% cerebrospinal fluid (CSF), and 10% blood. The Monro–Kellie doctrine states that, for the ICP to remain normal, alterations in the volume of one compartment must be compensated for by opposite changes in another compartment. Once these compensatory measures have been maximized, the ICP will rise dramatically with only a small increase in blood volume ( Fig. 26.1 ). Clinical manifestations include hypertension, bradycardia, and irregular respirations with alterations in mental status.
Autoregulation is the brain’s ability to maintain cerebral blood flow (CBF) despite changes in blood pressure ( Fig. 26.2 ).
In the adult uninjured brain, CBF remains relatively constant when arterial pressures are between 50 and 150 mm Hg. In the traumatized brain, this regulatory ability may be lost, and CBF may increase or decrease as the blood pressure increases or decreases, respectively. CBF remains constant when the Pa o 2 is above 60 mm Hg but increases dramatically with hypoxemia. A linear relationship exists between CBF and Pa CO 2 in that blood flow increases as carbon dioxide increases. Furthermore, CBF is closely linked to the cerebral metabolic rate of oxygen (CMRO 2 ) consumption; it will increase if CMRO 2 increases, as may occur with seizures, pain, fever, or agitation.
Intracranial hypertension is usually defined as an ICP above 20 mm Hg for more than 5 minutes, and in children is most commonly caused by traumatic brain injury. Neuroresuscitative goals are the prevention of secondary brain injury and limitation of further damage to surrounding neurons.
Structural and functional differences in cranial anatomy between children and adults influence cerebral physiology and management of increased ICP. Normal ICP in small children ranges from 2 to 4 mm Hg compared with the normal adult ICP between 8 and 18 mm Hg. The newborn skull does not completely fuse until the latter part of the first year of life. The intracranial space is relatively compliant and the dura is able to expand when the brain becomes edematous as a result of trauma, hemorrhage, or space-occupying lesions. Thus neonates and small infants may not develop signs and symptoms of the early stages of pathologic processes that increase brain mass. An important clinical correlate is that signs and symptoms of increased ICP in a small infant or neonate may be evidence of already advanced disease. When increased ICP is suspected in these patients, clinical findings of irritability, tense fontanels, psychomotor delay, papilledema, seizures, or characteristic thumb-printing on a plain skull film may aid in the diagnosis. Later in childhood, after fusion of the cranial sutures, children may exhibit less intracranial compliance than adults. This may be caused by a relatively higher percentage of brain tissue to CSF and blood vessels in children. Thus children may be at higher risk for dangerous increases in ICP with relatively less edema, hemorrhage, CSF accumulation, or tumor mass.
The major goal in the management of intracranial hypertension is to ensure adequate oxygen and substrate delivery to the brain to prevent tissue metabolic crisis and neuronal injury. This is achieved at the macrocellular level by ensuring adequate oxygenation and ventilation, lowering the intracranial pressure, and maintaining an adequate cerebral perfusion pressure by optimizing systemic blood pressure.
In cases of increased ICP, the general management goal is to achieve an ICP <20 mm Hg in all ages. The cerebral perfusion pressure (CPP) is the difference between mean arterial pressure (MAP) and the ICP (or central venous pressure [CVP], whichever is greater). Numerous studies have verified that CPP varies directly with age. In children less than 8 years old, a CPP >40 mm Hg is recommended, whereas in older children the CPP should be >60 mm Hg. There is insufficient evidence as to whether it is more important to decrease ICP or optimize CPP during conditions of acutely increased ICP, and whether such manipulation affects neurologic outcome. Nevertheless, in pediatric patients, CPP values <40 mm Hg are strongly correlated with worse outcomes at any ICP.
Intracranial pressure can be monitored using extradural, intraparenchymal, or intraventricular catheters. Intraventricular catheters have the additional therapeutic benefit of CSF drainage for ICP elevation management. The main indication for placement of an ICP monitor is a depressed Glasgow Coma Scale in the setting of acute brain injury.
Recently, brain tissue oxygen tension monitors have been used more frequently to guide neuroresuscitative strategies. These probes are placed in the white matter of the brain parenchyma, often in vulnerable brain tissue at risk for tissue hypoxia, cellular injury, and neuronal death. Brain tissue oxygen tension (PbtO 2 ) levels of 10 to 15 mm Hg correlate with poor outcome, while values of 25 to 26 mm Hg are observed in an uninjured brain. Augmenting CBF to target the PbtO 2 allows for additional goal-directed targets for cerebral resuscitation.
Fundamental methods to lower ICP are the same in children as in adults. These include elevating the head >30 degrees, avoiding jugular kinking by keeping the head positioned in the midline, ensuring adequate analgesia and sedation (caution in spontaneously ventilating patients with a natural airway), controlling ventilation with avoidance of excessive hyper or hypocapnea (i.e., which may result in unintended cerebral vasodilation or vasoconstriction and thus unintended effects on CBF), avoiding excessive ventilating pressures if positive-pressure ventilation and positive end-expiratory pressure (PEEP) are used, and optimizing arterial oxygenation.
Tracheal intubation is indicated if airway protective reflexes are absent or if there is a depressed level of consciousness Glasgow Coma Scale (GCS ≤8). Aside from risks of aspiration and worsening hypoxemia, acute hypoventilation can result in cerebral vasodilation, which can increase ICP and decrease CPP. Short-term hyperventilation is one of the most potent therapies to lower elevated ICP. However, prolonged hyperventilation or excessive hypocapnea is associated with worsening cerebral ischemia and is not recommended.
Recent studies indicate that application of PEEP does not increase intracranial pressure. Therefore, PEEP should be used to minimize alveolar collapse and optimize adequate oxygenation if clinically indicated. The current cardiovascular management strategy consists of volume resuscitation and vasopressor support as needed to maintain CPP. Administration of inhaled anesthetics will cause generalized cerebral vasodilation. In adults, an acute increase in ICP will render the patient more susceptible to the deleterious effects of volatile anesthetics. This is probably true in the pediatric population as well. Furthermore, neonates and infants appear to maintain the same degree of cerebral vasoconstriction in response to decreased P co 2 , as seen in adults. However, there is evidence that in anesthetized infants and children, cerebral vasodilation may occur at lower levels of P co 2 than observed in adults.
Additional principles of ICP management include avoidance of hyperthermia, maintenance of normoglycemia, and seizure prophylaxis if indicated. Hyperthermia increases CMRO 2 and may cause a mismatch between metabolic demand and supply. Hyperglycemia will exacerbate neuronal injury and therefore should be aggressively treated with insulin. Additionally, seizure activity (especially if uncontrolled) can greatly increase CMRO 2 and cerebral metabolic demand, potentially worsening neuronal injury.
Hyperosmolar therapy continues to be an important component of ICP management. Mannitol (0.25–1 g/kg infused intravenously over 10–30 minutes) creates an osmotic gradient that draws extracellular brain water across the blood–brain barrier into the intravascular space. It is also thought to decrease blood viscosity, thereby enhancing oxygen delivery. The prophylactic use of mannitol is not recommended because of its volume-depleting diuretic effect, which can increase risk for systemic hypotension as well as acute kidney injury in susceptible patients. Mannitol should be reserved for critical patients with increased ICP and signs of transtentorial herniation, or intraoperatively, to assist the neurosurgeon with brain relaxation. Hypertonic saline is being used more frequently to control ICP. It reduces cerebral swelling by producing an osmolar gradient to decrease cerebral water content. Continuous infusion of 3% NaCl through a central line will prevent swings in osmolality and maintain intravascular volume. It is not unusual to have serum sodium concentrations between 150 and 160 mEq/L. Side effects of hyperosmolar therapy include renal failure (especially if the serum osmolality is >320 mEq/L), hemolysis, and subarachnoid hemorrhage related to tearing of bridging vessels caused by rapid brain shrinkage. Rapid increases in serum sodium are also associated with development of central pontine myelinolysis, so 3% saline boluses should be carefully administered.
Barbiturate coma (burst suppression on EEG) is indicated when intracranial hypertension is not controlled with conventional therapies. Barbiturates lower ICP by reducing CMRO 2 , which then results in cerebral vasoconstriction and a reduction in CBF and volume. Barbiturates produce respiratory depression and a dose-dependent decrease in arterial blood pressure and cardiac output that often requires significant volume resuscitation and inotropic therapy. When the above measures fail, decompressive craniectomy can be considered in patients with reversible injury. The long-term neurologic benefit in patients with severe traumatic brain injury managed with decompressive craniectomy in pediatrics is unclear.
There has been a major shift in the way P co 2 is managed during adult neurosurgery, and this has extended into pediatric practice. Adult guidelines conclude that prophylactic hyperventilation (Pa co 2 <35 mm Hg) can theoretically compromise cerebral perfusion during a time when CBF is already reduced and may thus promote cerebral ischemia. Moreover, studies have shown that hyperventilation does not consistently lower CBF and may cause loss of autoregulation. Studies on hyperventilation in the pediatric population have yielded similar findings. Thus in pediatric patients, prophylactic hyperventilation is not recommended for fear of vasoconstriction-induced aggravation of cerebral ischemia, especially in situations of compromised regional CBF. Rather, mild hyperventilation to the lower end of eucapnia (P co 2 in mid 30 s) is most often used, the major reason being to offset the possible vasodilatory effects of general anesthetics. In the face of acute, dangerously increased ICP (i.e., sudden decline in mental status in patient with suspected increased ICP), lowering the P co 2 below the previous level will usually result in cerebral vasoconstriction and, at least temporarily, decrease ICP. This principle may extend to P co 2 levels below 30 mm Hg. When managing increased ICP, the fractional inspired oxygen concentration should always be set to 1.0 to maximize oxygen delivery to the brain cells.
Preoperative Assessment
The preoperative evaluation of the neurosurgical patient is critical because it can directly impact intraoperative planning, choice of anesthetic technique, physiologic goals, and ultimately disposition. Many children presenting for neurosurgery may have been hospitalized for evaluation of a recently diagnosed brain mass, vascular malformation, intracranial bleed, or malfunctioning shunt. During the course of this hospitalization, baseline laboratory values and trends, abnormal physical exam findings, and intracranial imaging should be available and evaluated. Other children with known brain pathology may have had their evaluation and imaging as an outpatient and may be admitted to the hospital on the day of the surgery. Required blood tests include a hemoglobin, and in most tumor or neurovascular cases, a type-and-screen or crossmatch if significant blood loss is anticipated. Electrolytes should be obtained if there is a possibility of hormonal alterations of sodium homeostasis, such as cerebral salt wasting, diabetes insipidus, acid–base disturbances, or syndrome of inappropriate antidiuretic hormone secretion (SIADH). These lab values may require intervention in critical patients, whether encountered preoperatively or during the course of the anesthetic. Many neurosurgical patients are often on an antiepileptic drug regimen, whether prophylactically (i.e., brain mass) or therapeutically. Unless the drug regimen is changing, or the child’s seizures are uncontrolled, preoperative anticonvulsant levels are typically not indicated. In addition to consultation with the neurosurgical team, prior computed tomography (CT), magnetic resonance imaging (MRI), or angiography imaging should also be evaluated preoperatively, as these can aid in determining the surgical approach, patient positioning, and preoperative planning.
Documentation of a neurologic examination that focuses on function of the cranial nerves, brachial plexus and lumbar/sciatic plexus (especially in patients with focal neurologic deficits) is essential in the preoperative period, as these values can be compared with postoperative exams, if there is concern for an intraoperative complication. In patients with concern for severe ICP and risk for impending herniation, a focused cranial nerve evaluation with emphasis on the pupillary light reflex should be considered and routinely assessed.
The majority of children presenting for neurosurgery will benefit from preoperative anxiolysis. Because most of these children already have indwelling intravenous (IV) catheters, midazolam can be titrated to effect in the preoperative holding area. Oral or intranasal midazolam may also be administered in children without IV access. In patients presenting with ongoing pain, opioids such as fentanyl can be administered, but should be done cautiously as excessive sedation will result in hypoventilation and hypercarbia, which contribute to potential worsening of increased ICP; this situation, however, is unusual in children. Those children with acutely increased ICP will present emergently and are often obtunded or too ill to benefit from preoperative anxiolysis. In patients with acute presentation of neurosurgical pathology (acute subdural hematoma, sudden change in mental status or neurologic examination), airway protection and ventilation must be frequently assessed, and if necessary, secured before arrival in the operating room (OR) to optimize cerebral perfusion and minimize secondary insult.
It is commonly thought that increased ICP slows gastric emptying and renders the patient at risk for aspiration of gastric contents. Yet very few data exist, especially in children, to substantiate this belief. In addition, clinical experience has shown that aspiration is rare in this population. Therefore, premedication with H 2 -antagonists and metoclopramide is not indicated unless the procedure is emergent and the child had a recent meal, or if there are other reasons to suspect that the child has abnormally increased gastric contents.
Anesthetic Techniques
A modified rapid sequence induction of general anesthesia is the preferred technique because of the small possibility of increased gastric contents in patients with increased ICP. Any IV induction agent can be used, along with a nondepolarizing neuromuscular blocker, while an assistant provides cricoid pressure during gentle bag-mask ventilation to minimize excessive hypercarbia. Succinylcholine is usually avoided because of its propensity to increase ICP, yet, if a reasonable risk for pulmonary aspiration of gastric contents exists, a full rapid sequence induction using succinylcholine or high-dose rocuronium is indicated. In the subset of patients with prior history of upper or lower motor neuron deficits (i.e., stroke, paraplegia, etc.), succinylcholine is contraindicated out of concern for uncontrolled hyperkalemic response because of potential upregulation of acetylcholine receptors throughout the muscle membrane. With the recent introduction of sugammadex, which has been shown to have a good safety profile in the pediatric population, rapid sequence induction with rocuronium is a good alternative. If the child does not have an indwelling IV catheter, and there is no reasonable risk for pulmonary aspiration of gastric contents, then an inhaled induction can be performed. Cricoid pressure is applied as soon as the child loses consciousness, and IV access is rapidly attained to permit continuation of a modified rapid sequence technique. The benefit of cricoid pressure here is questionable, and some practitioners may choose to omit it.
The mechanism by which succinylcholine causes transient increases in ICP is unknown. Some evidence suggests that it is caused by afferent neuronal muscle spindle activity that results from muscle fasciculations. In adults, pretreatment with a small dose of a nondepolarizing muscle relaxant may prevent this increase in ICP from succinylcholine. Because small children tend not to exhibit muscle fasciculations, this effect of succinylcholine may not be observed. Nevertheless, unless succinylcholine is indicated based on the child’s risk for aspiration or airway status, it is best avoided during induction of general anesthesia.
When avoidance of potential acute increases in ICP is of concern during induction of anesthesia, additional therapies are indicated. These include IV opioids, lidocaine, and even dexmedetomidine, which will help blunt the hemodynamic response to laryngoscopy and tracheal intubation, and thus help minimize dangerous increases in ICP. In addition, scalp infiltration with a local anesthetic will limit the hemodynamic response to the surgical incision. Traditionally, the bulk of the opioid dose is given toward the beginning of the neurosurgical case, because tracheal intubation, cranial pin application for head immobilization, positioning, scalp incision, and craniotomy are the most painful and stimulating events. Furthermore, residual opioid effect is undesirable at the end of the procedure when the goals are the rapid attainment of consciousness and tracheal extubation to facilitate an immediate neurologic exam. Fentanyl, 2 to 6 μg/kg, is commonly used during induction and can be used intraoperatively as an infusion with careful consideration for context-sensitive half-time. Remifentanil is a reasonable alternative and can be used as a continuous infusion throughout the procedure. Some adult studies have suggested that alfentanil and sufentanil may increase ICP by either increasing CSF volume or increasing CBF. A study in children examined the effect of alfentanil (10–40 μg/kg) on ICP in children with moderately elevated ICP presenting for revision of VP shunts. Alfentanil consistently produced a decrease in CPP that was largely accounted for by decreases in blood pressure. Increases in ICP were not observed. Ketamine is usually avoided in high-risk situations because of its propensity to increase ICP.
Any inhaled volatile agent can be used for maintenance of general anesthesia; in adults, this choice does not affect the outcome of neurosurgical procedures. All volatile general anesthetic agents can cause cerebral vasodilation and increase ICP by increasing CBF and volume, yet there is some reduction in CMRO 2 that may offset this augmentation. Studies in children have used the transcranial Doppler technique to measure the blood flow velocity of the middle cerebral artery, which in turn may represent overall CBF and volume. In children, isoflurane appears to have minimal effects on CBF and cerebrovascular reactivity to CO 2 between 0.5 and 1.5 minimum alveolar concentration (MAC). Furthermore, administration of a constant concentration of isoflurane over time does not affect cerebral hemodynamic variables. Sevoflurane appears to have similar effects as isoflurane, although it has not been well studied in children. Desflurane has been shown to increase ICP in adults despite application of hypocarbia but has not been directly studied in children. Nitrous oxide (N 2 O) increases CBF when used alone and in combination with propofol or sevoflurane. However, it does not appear to increase CBF when combined with desflurane. This lack of effect may be explained by the potent baseline cerebrovascular dilation effect of desflurane. Overall, the vasodilatory cerebrovascular effects of all inhaled anesthetics in children are similar to that of adults. With normal levels of ICP, there is probably no clinical difference between agents. In conditions of increased ICP, although effects on CBF are mitigated by moderate hyperventilation for all volatile anesthetic agents, a balanced anesthetic technique should be employed with minimal concentrations of inhaled agents combined with an opioid-based technique.
Normally, fentanyl 1 to 5 μg/kg/h or remifentanil 0.1 to 0.3 μg/kg/min is continued throughout the procedure and targeted to desired hemodynamic parameters. As part of the preanesthetic planning, one should consider the gradual increase in the context sensitive half-time of fentanyl if used as an infusion during a prolonged procedure, especially when extubation and prompt neurologic evaluation is planned at the end of the case. In adults, studies that have examined the role of remifentanil as the primary anesthetic agent to facilitate early extubation have not demonstrated any differences in short- or long-term outcome. Similar studies have not been performed in children. Neuromuscular blockade throughout the procedure is encouraged to facilitate positioning, especially if Mayfield pins secured to the cranium are used, and to assure lack of patient movement during brain dissection. Children on chronic anticonvulsant therapy will require more frequent dosing of aminosteroidal muscle relaxants. If the child has a preexisting hemiparesis, the twitch monitor should be placed on the nonhemiparetic side.
Use of N 2 O is controversial in neurosurgery because there is evidence that it raises CMRO 2 and may increase CBF and ICP. No outcome studies exist that influence the decision whether to include it in the anesthetic management of children. However, because N 2 O is not essential for any reason in neuroanesthesia, it should not be used when increased ICP is a possibility. In addition, N 2 O is contraindicated in any child who returns for a repeat craniotomy within 1 month because of the possible development of pneumocephalus secondary to air remaining within the ventricles or cisternal system.
A general rule in pediatric neuroanesthesia is that if a child’s trachea was not intubated on arrival to the OR, then he or she should be awakened at the completion of the procedure with the intent of tracheal extubation and immediate neurologic evaluation. Exceptions include cases where adverse intraoperative events occurred that would likely cause postoperative cardiorespiratory depression, need for immediate postoperative imaging, or if there is concern with the ability of the child to protect their airway from obstruction or aspiration. Children with acute head trauma who underwent tracheal intubation in the field or the emergency department are also potential candidates for tracheal extubation after a successful evacuation of a blood clot or hemorrhage, assuming all cardiorespiratory parameters have normalized. Children in whom life-threatening increased ICP is expected to persist into the postoperative period should receive continuous sedation and neuromuscular blockade, and should be managed postoperatively in the intensive care unit (ICU) with controlled mechanical ventilation.
Positioning Children for Neurosurgery
Positioning pediatric patients for neurosurgery entails proactive attention to detail that will prevent complications or problems during the procedure. This includes careful fixation of the endotracheal tube and securing of circuit, atraumatic placement of an orogastric tube and esophageal temperature probe, administration of petroleum-based eye lubrication, careful securing of arterial and IV catheters/lines, and careful positioning of an indwelling urinary catheter. Most often the patient will be positioned supine with the head turned to the side, but many procedures require prone, lateral, head-up (i.e., semi-sitting, reverse trendelenburg) or even sitting (rarely) positioning. All anesthetic monitors and access lines should be placed and their adequacy confirmed before draping. In addition, the anesthesiologist must plan the workspace so as to include adequate access to the patient during the surgery. Padding pressure points will help prevent compression injuries. If the patient is prone, free and easy abdominal movement with ventilatory movements should be confirmed by inspection of the abdomen and confirmation that ventilatory compliance is unchanged from baseline.
Though declining in frequency, the sitting position may occasionally be used in pediatric patients for surgical access to the posterior fossa; it entails the same monitoring and safety considerations as for adults in the sitting position. Similarly, in children there is also the risk for venous air embolism from entrainment of air into open vessels. A retrospective audit of complications associated with the sitting position in children reported the incidence of venous air embolism to be about 9%. All were detected and treated appropriately, and none directly caused morbidity or mortality. Other pediatric studies have reported an incidence of venous air embolism in the sitting position as high as 37%. Some evidence exists that children have higher dural sinus pressures than adults, and may account for the generally lower incidence of venous air embolism in children compared with adults. However, some studies suggest that venous air embolism is more likely to result in hypotension in pediatric patients because, in theory, the same-sized air bubble would be larger relative to the smaller blood volume of children and, therefore, cause greater hemodynamic instability. Overall, outcome studies in children in the sitting position do not show greater risk than in adults. If one is planning to use the sitting position in a child, it is strongly recommended to obtain a preoperative echocardiograph to rule out any interatrial or interventricular communications. Even when the standard sitting position is not used, the risk of venous air embolism via open venous channels in the bone and dural sinuses still exists, especially when the head of the table is elevated to improve surgical access and cerebral venous drainage. In addition to invasive blood pressure monitoring transduced at the level of the brain and vigilant hemodynamic monitoring, the standard use of a well-positioned and tested precordial Doppler for these situations is recommended to rapidly detect the presence of a venous air embolism. Although some institutions may place central venous catheters for the purpose of theoretical management of a venous air embolism, there is little data in children to suggest that the benefits of this indication outweigh the potential risks of its placement and use.
Manipulation of the head and cervical spine is frequent during neurosurgical positioning and complications of such positioning (i.e., brain ischemia, paraplegia) can be disastrous. Rotation of the cervical spine can result in some decrease in ipsilateral carotid blood flow. Extreme flexion of the neck can compromise vertebral and carotid arterial blood flow as well venous drainage, and positioning should be checked to ensure that there is sufficient thyromental distance (2–3 fingerbreadth). Given the inherent risks, patient positioning should be a joint effort between surgical and anesthesia teams, and the final position should be agreed upon before incision.
When a child is positioned for neurosurgery in a head frame fixation device (e.g., Mayfield or Sugita) there is often flexion of the neck. This maneuver may result in downward displacement of the endotracheal tube within the trachea and cause the endotracheal tube to enter the right main bronchus. Therefore in anticipation of neck flexion, the endotracheal tube should be positioned on the higher side within the trachea, to compensate for its descent. Once positioning is finalized, bilateral breath sounds should be confirmed; in children with healthy lungs, any unexplained oxygen saturation below 96% (especially if accompanied with increased inspiratory pressures and decreased breath sounds to left lung fields) should prompt an exploration for a right main bronchial intubation.
When an invasive head fixation device is used to secure the patient’s head, one must take into consideration the stimulation from placement of the device throughout the anesthetic. Pin placement may serve as a highly stimulating part of the anesthetic, especially if lower levels of maintenance are used during line placement and initial positioning at the start of the case. At the end of the case, the anesthetic should be planned to maintain adequate depth until the pins are removed, while planning for a rapid emergence thereafter for extubation and neurologic evaluation.
Neurophysiologic Monitoring
Many neurosurgeons employ the use of intraoperative neurophysiologic monitoring to evaluate the function of neurologic pathways and to prevent potential cerebral or spinal cord ischemia. This may include electrocorticography (ECoG), electroencephalography (EEG), somatosensory evoked potentials (SSEP), motor evoked potentials (MEP), and brainstem auditory evoked potentials (BAEP). When these modalities are used, the anesthesiologist should proactively discuss with the neurophysiologist the implications of anesthetic management on the accuracy of the monitoring. Because many anesthetic agents (i.e., clinically required concentrations of volatile agents) have impact on the sensitivity and quality of neurophysiologic signals, the anesthetic goal in these situations is to maintain steady-state levels of anesthetic agents. The technique most frequently employed is total intravenous anesthesia without neuromuscular blockade, often with a titratable analgesic (i.e., remifentanil, fentanyl, sufentanil) and use of a propofol infusion or low-dose volatile agent for hypnosis. Use of remifentanil is advantageous in these situations, as it will be rapidly eliminated at the end of the procedure to allow prompt return of spontaneous ventilation and emergence. Additionally, significant changes in physiologic variables (temperature, hemodynamics, ventilation, electrolytes, glucose levels) have been associated with changes in evoked potentials.
Sevoflurane, isoflurane, and desflurane can increase signal latency and decrease amplitude in a dose-dependent manner and can abolish MEPs to a greater degree than SSEPs. As a sole agent, N 2 O reduces amplitude and increases latency without changing wave morphology but does so to a greater effect when combined with halogenated agents. Conversely, ketamine can increase SSEPs and MEPs and can be used as a bolus or infusion, especially for surgeries at risk for significant postoperative pain (i.e., multilevel spinal fusion, scoliosis surgery); in patients with increased ICP, however, this medication should be used with caution. Similarly, etomidate has been shown to increase cortical SSEPs and a negligible MEP depression. Dexmedetomidine, an alpha-2 agonist, can be safely used as it has been shown to have clinically nonsignificant impact on neurophysiologic signals.
Fluid Management
Except for neonates or other children who may be at risk for development of hypoglycemia, glucose-containing solutions are generally avoided during neurosurgical procedures. Though definitive outcome studies are lacking, hyperglycemia has been associated with worse neurologic outcomes during episodes of brain ischemia. Isotonic solutions such as normal saline or Plasma-Lyte are typically advocated during neurosurgery to lessen the risk for excess brain water, but Lactated Ringer’s solution can also be used if ICP is not increased.
Common Pediatric Neurosurgical Procedures
Ventriculoperitoneal Shunt Insertion or Revision
Ventriculoperitoneal (VP) shunt insertion or revision is probably the most commonly performed pediatric neurosurgical procedure. A shunt is initially placed to palliate disorders that cause hydrocephalus. The proximal end of the shunt is placed within the lateral ventricle of the brain, and the shunt is tunneled underneath the scalp and skin of the neck, chest, and abdomen, where it inserts into the peritoneal cavity to drain CSF. Some children may have their shunt terminate into the atrium via one of the large veins of the neck or chest. Children with existing shunts will periodically require revisions because of a variety of reasons that include growth, obstruction, and infection. Occasionally, these revisions will be performed on an emergent basis if the child exhibits signs or symptoms of acutely increased ICP.
There are a variety of causes of hydrocephalus. The most common cause of congenital hydrocephalus is narrowing of the aqueduct of Sylvius. Acquired hydrocephalus is most commonly a result of intracranial hemorrhage in prematurely born infants. Other causes of hydrocephalus that require shunting include Arnold-Chiari compression associated with myelomeningocele, infections, tumors, and head injury.
Preoperative assessment consists of evaluation of comorbidities and the severity of increased ICP. Premedication with an anxiolytic such as midazolam is usually indicated, but when a child is deemed to have clinically important increased ICP, an IV catheter should be placed before surgery, and premedication is then carefully titrated under direct supervision. There is no routine preoperative blood work except evaluation of electrolytes if protracted vomiting is present preoperatively.
Perioperative fluids should consist of normal saline solution or Lactated Ringer’s solution. Insensible losses usually range from 2 to 4 mL/kg/h. Standard monitors are appropriate, and one IV access line is usually sufficient.
In children with preoperative signs and symptoms of increased ICP, a modified rapid sequence IV induction is indicated. A moderate amount of opioid is used to facilitate induction and tracheal intubation but should be tailored toward tracheal extubation at the completion of the procedure. Any volatile or IV anesthetic agent is appropriate for maintenance of general anesthesia. N 2 O should be avoided if the child has had a craniotomy within the past month, because of the possibility of expanding an existing pneumocephalus. After tracheal intubation, the stomach should be suctioned with an appropriate large-diameter (i.e., 14-16 French) orogastric tube, and an esophageal temperature probe is recommended.
The surgical procedure for initial insertion consists of two small incisions: one on the lateral side of the head, and the other on the skin of the abdomen, with which to retrieve the tunneled shunt and insert it into the peritoneum. Ordinarily the entire unilateral area from the head to the abdomen is prepared and draped. The child lies supine with the head turned away from the operative side. The OR table is turned 90 degrees away from the anesthesiologist ( Fig. 26.3A and B ).
