Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Supratentorial Masses: Anesthetic Considerations
Epidemiology
According to the Central Brain Tumor Registry of the United States (CBTRUS) for the years 2007–2011, the incidence of primary brain and central nervous system (CNS) tumors is 21.42/100,000 per year. For children and adolescents 0–19 years of age, the incidence is 5.42/100,000. The incidence is 27.85/100,000 for adults (> 20 years). For the year 2015, the estimate is 68,470 primary brain and CNS new cases. The overall mortality rate is 4.26/100,000. Approximately 34% of the tumors are malignant. The most common tumor is meningioma (36%) followed by glioblastoma (15%). The broad category glioma represents 28% of all tumors. The 5-years survival rate for malignant brain and CNS tumors is 34%, but only 5% for glioblastoma. The majority of tumors (> 80%) are supratentorial. For all primary brain tumors, the median age of diagnosis is 59 years. From 1985 to 1999, the incidence of primary brain tumors rose modestly (1.1% per year). The exact incidence of brain metastases is unknown but certainly underestimated. In about 25% of patients who die from cancer, central nervous system metastases are detected at autopsy. For the five most common sources of brain metastases (breast, colorectal, kidney, lung, and melanoma), 6% of the patients suffer this complication within 1 year of diagnosis of the primary cancer. Thus, these five cancers probably cause approximately 37,000 cases of brain metastases per year in the United States. Conversely, about 10% of patients with lung cancer present to the physician with symptoms from a brain metastasis.
General Considerations
For patients the problems associated with supratentorial tumors result from local and generalized pressure, whereas for surgeons the difficulties arise during surgical exposure because the brain is particularly susceptible to damage from retraction and mobilization. Anesthesia for supratentorial tumors thus requires an understanding of the pathophysiology of localized or generalized rising intracranial pressure (ICP); the regulation and maintenance of intracerebral perfusion; how to avoid secondary systemic insults to the brain ( Box 11.1 ); the effects of anesthesia on ICP, perfusion, and metabolism; and the therapeutic options available for decreasing ICP, brain bulk, and tension perioperatively. Specific problems include massive intraoperative hemorrhage, seizures, and air embolism in the head-elevated or sitting position or if venous sinuses are traversed. Further questions are how to monitor the brain’s function and environment, and whether to aim for rapid anesthesia emergence or for prolonged postoperative sedation and ventilation. Finally, the concurrence of various intracranial and extracranial pathologic conditions should not be forgotten, such as the presence of cardiovascular or pulmonary disease or—in the case of metastases—the existence of paraneoplastic phenomena and the effects of chemotherapy or radiotherapy. This concept can be summarized as follows:
| The anesthetic goal: | To preserve brain from secondary insult |
| The anesthetic risk factors: | Hypoxemia, hypercapnia, anemia, hypotension |
| The anesthetic actions: | Conserve cerebral autoregulation and CO 2 responsiveness Maximize brain elastance to decrease retractor pressure |
Box 11.1
Secondary Insults to the Already Injured Brain
Intracranial
-
Increased intracranial pressure
-
Midline shift: tearing of the cerebral vessels
-
Herniation: falx, transtentorial, trans-foramen magnum, transcraniotomy
-
Epilepsy
-
Vasospasm
Systemic
-
Hypercapnia
-
Hypoxemia
-
Hypotension or hypertension
-
Hypo-osmolality or hyperosmolality
-
Hypoglycemia
-
Hyperglycemia
-
Low cardiac output
-
Hyperthermia
Pathophysiology of Rising Intracranial Pressure
The main normal intracranial components of the brain (tissue, intravascular blood, cerebrospinal fluid [CSF]) are contained in an unyielding skull. Hence any increase in their volume—or the addition of an abnormal mass—must be compensated by a concurrent reduction in volume of one or more of these components, mainly CSF or blood (the brain is largely incompressible) ( Fig. 11.1 ). The ability of these homeostatic mechanisms to compensate depends not only on the volume of the mass but also on the speed at which it arises: for rapidly expanding masses, the ICP volume curve shifts markedly to the left. Early but limited homeostasis is provided by extracranial shifts of intracranial blood, followed by larger-capacity displacement of CSF—which is ineffective if CSF flow is obstructed. Once these compensatory mechanisms are exhausted, ICP rises rapidly, which is followed by impairment of cerebral circulation and, ultimately, by brain herniation; generally subfalcine (“midline shift”) or transtentorial, the end stage of compensation. This concept can be summarized as follows:
| The cornerstone of neuroanesthesia: | Intracranial pressure–volume relationship |
| The main goal of neuroanesthesia: | Avoiding intracranial compartment volume increase, especially for cerebral blood volume (anesthetics, mean arterial pressure autoregulation, CO 2 ) |
| Anesthetic risk factor: | Administration of hypotonic fluidsMedications that affect cerebral autoregulation |
Volume Effects of Intracranial Tumors
The intracranial volume effects of tumors are due not only to the mass of the tumor itself but also to the surrounding vasogenic brain edema. Such edema, commonly seen on preoperative computed tomography (CT) or magnetic resonance imaging, apparently results from secretory factors that increase vascular permeability in the nearby brain. Peritumoral edema is particularly marked around fast-growing tumors, generally responds well to corticosteroid therapy, and can persist or even rebound after surgery for excision of the tumor. Thus the areas surrounding large tumors suffer from ischemia resulting from compression (cerebral blood flow [CBF]) in peritumoral tissue may be decreased by up to one-third compared with normal tissue). Treatment with steroids such as dexamethasone usually results in dramatic decreases in surrounding brain edema. The emergency and preoperative treatment of peritumoral vasogenic edema is the only good indication for steroid therapy in this context.
The Blood–Brain Barrier and Edema
The blood–brain barrier is also affected by intracranial pathologic conditions. Normally the blood–brain barrier is impermeable to large or polar molecules and variably permeable to ions and small hydrophilic nonelectrolytes. Thus any disruption of the blood–brain barrier permits water, electrolytes, and large hydrophilic molecules to enter perivascular brain tissues, leading to vasogenic brain edema. In this case, leakage—and the resulting brain edema—is directly proportional to the cerebral perfusion pressure (CPP). Vasogenic edema should be differentiated from osmotic edema (caused by a drop in serum osmolality) and cytotoxic edema (secondary to ischemia). Blood osmolality is a critical determinant of cerebral edema because a 19-mmHg pressure gradient across the blood–brain barrier is generated for every milliosmole. In contrast, oncotic pressure plays a minor role. Neuroimaging shows disruption of the blood–brain barrier in many tumors. New strategies are being investigated to improve drug delivery to brain tumors. In the future, it is possible that new treatments to augment blood–brain barrier permeability (osmotic blood–brain barrier disruption, intra-arterial chemotherapy) will interfere with perioperative management.
Intracerebral Perfusion and Cerebral Blood Flow
CBF is regulated at the level of the cerebral arteriole. It depends on the pressure gradient across the vessel wall (which in turn is the result of CPP) and PaCO 2 value (which depends on ventilation) ( Fig. 11.2 ). CBF autoregulation, dominant to ICP homeostasis, keeps CBF constant in the face of changes in CPP or mean arterial pressure (MAP). It does this through alterations in cerebral vasomotor tone (i.e., cerebrovascular resistance [CVR]). Autoregulation is normally functional for CPP values of 50 to 150 mmHg and is impaired by many intracranial (e.g., blood in CSF, trauma, tumors) and extracranial (e.g., chronic systemic hypertension) pathologic conditions. It is also affected by drugs used in anesthesia.
If CPP is inadequate, tissue perfusion will decrease when the lower limit of autoregulation is less than 50 mmHg (if autoregulation is intact). Ischemia results at levels of CBF below 20 mL/100 g/min unless CPP is restored (by increasing MAP or decreasing ICP) or cerebral metabolic demand is reduced (through deepened anesthesia or hypothermia). Increased ICP resulting in reduced CPP is met by cerebral arteriolar relaxation; in parallel, MAP is increased via the systemic autonomic response. As a result, a vicious cycle can be established, particularly in the presence of impaired intracranial homeostasis, as cerebral vessel relaxation increases cerebral blood volume (CBV), thus further raising ICP. In addition, an acute reduction in CPP or MAP tends to acutely increase ICP (the so-called vasodilatory cascade ). Reductions in PaCO 2 induce vasoconstriction, reducing CBF, CBV, and thus ICP. Conversely, hypercapnia increases ICP and should be prevented in the perioperative period. This makes hyperventilation a useful tool for the acute control of intracerebral hyperemia and elevated ICP ( Fig. 11.3 ), as summarized here:
| The anesthetic goal: | Hemodynamic stability |
| The reason: | Autoregulation takes 30 to 120 seconds to be established; thus sharp MAP fluctuations entrain undesirable CBF, CBV, and ICP changes |
| The formulas: | CBF = CPP/CVRCPP = MAP − ICPNormally, ICP < CVP |
Anesthesia and Intracerebral Pressure, Perfusion, and Metabolism
Anesthesia exerts major effects on the intracranial environment through a variety of drug and nondrug effects. These effects are sensitive to the state of the intracranial and extracranial environment (e.g., cerebral compliance, presence or absence of intracranial pathologic condition, general volemic state).
Intravenous Anesthetics
Intravenous anesthetics include barbiturates, propofol, etomidate and ketamine. Apart from anesthesia induction, propofol is being increasingly used for maintenance as a continuous intravenous infusion (often computer controlled). All the intravenous drugs mentioned are cerebral vasoconstrictors that act by depression of cerebral metabolic rate (CMR), except ketamine. Ketamine increases whole brain CBF without changing CMR in healthy volunteers. At subanesthetic doses, ketamine increases regional glucose metabolic rate and CBF. The other agents decrease CBF, CBV, and ICP while leaving autoregulation and vessel reactivity to PaCO 2 intact (see Fig. 11.3 ). , CMR reduction reflects brain activity and is mediated through the electrical but not the basal metabolic activity of the neurons. Hence there is a ceiling effect for CMR reduction at electroencephalogram (EEG) burst suppression ( Fig. 11.4 ). In contrast to volatile anesthetics, propofol has been shown capable of suppressing the cerebrostimulatory effects of nitrous oxide. Etomidate directly inhibits adrenal cortisol secretion for 24 to 48 hours even after a single injection, and its use is often associated with myoclonic (not epileptic) movements.
Volatile Anesthetics
All volatile anesthetics are cerebral vasodilators, but isoflurane, sevoflurane, and desflurane also reduce CMR. A flat EEG is obtained with these three agents at around 2 minimum alveolar concentrations (2 MAC), a concentration at which maximum metabolic depression is achieved. The response of cerebral metabolism to rising concentrations of volatile anesthetics is not linear. The decrease in CMR is steep from 0 to 0.5 MAC and then more gradual up to 2 MAC. The effect of volatile anesthetics on CBF is the result of their vasodilatory properties and flow-metabolism coupling. At low concentrations (< 1 MAC), CBF is lower than in the awake person. But CBV is unchanged with isoflurane and decreased with propofol at comparable concentrations. Among the volatile anesthetics, sevoflurane is the least vasodilating and desflurane the most. The effects of xenon are more complex. This agent decreases CBF in gray matter, particularly in specific brain areas such as the thalamus, the cerebellum, the cingulated gyrus, and the hippocampus, and increases CBF in white matter. It does not impair flow-metabolism coupling.
For the normal brain and volatile concentrations below 1 MAC, PaCO 2 reactivity remains intact, permitting control of vasodilation by hypocapnia. (However, the presence of a pathologic brain condition or use of a high-MAC volatile anesthetic may impair or even abolish PaCO 2 reactivity and autoregulation.)
Nitrous Oxide
Nitrous oxide is cerebrostimulatory, increasing CBF, CMR, and sometimes ICP. Its effect is not uniform throughout the brain but is limited to selected brain regions (basal ganglia, thalamus, insula), changing the regional distribution of CBF. If substituted for an equipotent concentration of a volatile anesthetic agent, nitrous oxide increases CBF. For the normal brain, the resulting cerebral vasodilation can be controlled by hypocapnia or the addition of an intravenous anesthetic. However, volatile agents have no such attenuating effect; CMR and CBF are higher during 1 MAC anesthesia produced by a nitrous oxide-volatile anesthetic combination than that produced only by a volatile anesthetic. This effect is especially deleterious in the actual or potential presence of brain ischemia. Particularly for repeat craniotomy, the potential of nitrous oxide, which is poorly soluble, to diffuse into and hence expand hollow spaces must be remembered as it could cause tension pneumocephalus in patients with intracranial air (repeat neurosurgery or head trauma).
Opioids
Opioids have been associated with short-term increases in ICP, particularly sufentanil or alfentanil. Reflex cerebral vasodilation after decreases in MAP and hence in CPP is the underlying mechanism for the transient increases in ICP, , although a direct modest cerebral vasodilator effect has been demonstrated. This effect demonstrates the sensitivity of intracerebral drug effects to the intracranial and extracranial environment and the importance of maintaining normovolemia for ICP stability. Generally, opioids modestly reduce CMR and do not affect flow-metabolism coupling, autoregulation, or the carbon dioxide sensitivity of the cerebral vessels. Remifentanil has been extensively studied. Its cerebral effects are comparable to those of other opioids, and its use in neuroanesthesia has been validated in clinical trials. ,
Other Drugs
Vasodilating antihypertensive agents such as nitroglycerine, nitroprusside, and nicardipine increase ICP and should be avoided. Cerebral vasodilation may result from a normal autoregulation response or direct arterial vasodilation. For example, sodium nitroprusside increases ICP, but intracarotid injection of nitroprusside does not change CBF. Conversely, verapamil decreases cerebrovascular resistance in humans by inducing direct cerebral vasodilation. Theophylline constricts cerebral vessels but increases CSF production and is a potent central nervous system (CNS) stimulant, raising the risk of convulsions. Most β-adrenergic blockers, especially esmolol, do not interfere with cerebral blood flow or metabolism.
Reducing Intracranial Pressure, Brain Bulk, and Tension
The anesthesiologist possesses a number of instruments to achieve ICP reduction and brain relaxation ( Box 11.2 ), and thus to improve the quality of surgical exposure and to reduce retractor pressure. The effectiveness of these instruments depends on intact intracerebral homeostatic mechanisms.
Box 11.2
Management of Intracranial Hypertension and Brain Bulging
Prevention
-
Euvolemia
-
Sedation, analgesia, anxiolysis
-
No noxious stimulus applied without sedation and local anesthesia
-
Head-up position, no compression of the jugular veins, head straight
-
Osmotic agents: mannitol, hypertonic saline
-
β-Blockers or clonidine or lidocaine
-
Steroids, if a tumor is present
-
Adequate hemodynamics: mean arterial blood pressure, central venous pressure, pulmonary capillary wedge pressure, heart rate
-
Adequate ventilation: Pao 2 > 100 mmHg, PaCO 2 35 mmHg
-
Intrathoracic pressure as low as possible
-
Hyperventilation on demand before induction
-
Use of intravenous anesthetic agents for induction and maintenance in case of tensed brain
Treatment
-
Cerebrospinal fluid drainage if ventricular or lumbar catheter in situ
-
Osmotic agents
-
Hyperventilation
-
Augmentation of anesthesia with intravenous anesthetic agents: propofol, thiopentone, etomidate
-
Muscle relaxants
-
Venous drainage: head up, no positive end-expiratory pressure, reduction of inspiratory time
-
Mild controlled hypertension if autoregulation present
Intravenous Anesthetics
Intravenous anesthetics reduce CMR, CBF, and hence CBV and ICP, leading to a diminution of brain bulk, as discussed previously. Cerebral vasoconstriction depends on intact flow-metabolism coupling ( Figs. 11.4 and 11.5 ) and is dose related up to neuronal electrical silence (EEG burst suppression). Like autoregulation, flow-metabolism coupling is impaired by brain contusion and other intracerebral pathologic conditions.
Hyperventilation
Hyperventilation results in hypocapnia and subsequent cerebral vasoconstriction. In the context of intact autoregulation, CBF is roughly linearly related to PaCO 2 between 20 and 70 mmHg. However, the carbon dioxide reactivity of cerebral vessels may be impaired or abolished in the presence of head injury or other intracerebral pathologic conditions, by high inspired concentrations of volatile anesthetics, or, particularly if the vessels are already dilated, by nitrous oxide. The CBF-, CBV-, and ICP-reducing effects of hypocapnia are acute and apparent for less than 24 hours. A typical value to aim for is a PaCO 2 of 30 to 35 mmHg; arterial blood gas analysis rather than end-tidal CO 2 ( etco 2 ) should be used as a controlling variable because of the possibility of large arterioalveolar CO 2 gradients in neurosurgical patients. The effectiveness of hyperventilation (PaCO 2 at 25 ± 2 mmHg) for controlling brain bulk in the patient under either isoflurane or propofol anesthesia has been demonstrated.
The main complication associated with hyperventilation is reduction of CBF, which gives rise to cerebral ischemia. Thus, the anesthesiologist must balance the benefit of brain relaxation against the risk of cerebral hypoperfusion. Other side effects are linear reduction in coronary artery flow, reduced cardiac venous return, hypokalemia, and potentiation of the brain’s response to opioids.
Diuretics and Osmotic agents
Osmotic diuretics such as mannitol and hyperosmotic saline increase blood osmolality acutely, thus reducing brain water content (mainly in healthy brain tissue with an intact blood–brain barrier) and hence brain bulk and ICP. This response improves brain deformability and thereby facilitates surgical exposure. A further beneficial effect is improvement in blood rheology as a result of the reduction in edema of vascular endothelium and erythrocytes (increasing erythrocyte deformability)—the basis of mannitol’s classic “antisludge” effect. A typical regimen is to give 0.5 to 1 g/kg mannitol (150–400 mL 20% mannitol) intravenously, split between a more rapid pre-craniotomy dose and a slower infusion, until brain dissection is complete. The ICP effect is prompt, removes about 90 mL of brain water at peak effect, and lasts for 2 to 3 hours. Theoretically, equiosmolar infusions of hypertonic saline or mannitol should have the same effect for reducing brain water content. One study showed slightly better results with hypertonic saline than mannitol on intraoperative brain relaxation. Normally the aim is to keep osmolality at less than 320 mOsm/kg. Problems with the use of osmotic diuretics include hypernatremia, hypokalemia, and acute hypervolemia, which could be deleterious in patients with congestive heart failure. There is no additional benefit to using loop diuretics such as furosemide, which induces hypovolemia and does not reduce brain water content except that it may limit rebound edema formation. On the contrary, serum saline should be infused to replace urinary losses in order to avoid hypovolemia and maintain blood pressure.
Cerebrospinal Fluid Drainage
CSF drainage is achieved either by intraoperative direct puncture of the lateral ventricle or through a lumbar spinal catheter placed preoperatively. The latter is effective only if there is no caudal block to CSF outflow. Because of the risk of causing acute brain herniation, lumbar CSF drainage should be used cautiously and only when the dura is open. The patient should receive at least mild hyperventilation when CSF is drained. Normally removal of 10–20 mL of CSF is very effective in reducing brain tension. Up to 50 mL can be drained if necessary.
Other Factors
Other factors causing cerebral vasodilation and that can be corrected by the anesthesiologist include hypovolemia and hypoxia. The position of the patient (head down, extreme turning of the neck) also influences brain volume because of impaired venous drainage of the brain. This should be kept in mind when brain swelling is observed without any obvious reason after the dura is opened. Repositioning of the head to avoid excessive rotation and compression of the jugular vein may be the solution.
Vasoconstrictive Cascade
Finally, the anesthesiologist can use the vasoconstrictive cascade by mildly increasing MAP, thus increasing CPP and decreasing CBV and ICP ( Fig. 11.6 ).
General Anesthetic Management
Preoperative Assessment
Determination of anesthetic strategy for a given neurosurgical intervention depends on thorough knowledge of the neurologic and general state of the patient, the planned intervention, and holistic integration of these factors. The patient and the planned intervention should be discussed with the neurosurgeon involved.
Neurologic State of Patient
A major aim in evaluating neurologic status is to estimate how much ICP is raised, the extent of impairment of intracranial compliance and autoregulation, the localization of the tumor and how much homeostatic reserve for ICP and CBF remains before brain ischemia and neurologic impairment occur. The goal is also to assess how much permanent and reversible neurologic damage is already present. Typical pointers to these elements in the patient history, physical examination, and technical examinations are listed in Box 11.3 . The minimum examination should involve a neurologic mini-mental status assessment, comprising the patient’s ability to follow commands, the patient’s degree of orientation, the presence or absence of speech deficit, the pupil symmetry and the Glasgow Coma Scale score. Elucidating what medication the patient is receiving and for how long is important because this medication may also affect intracranial compliance, perfusion, and reserves, as well as modify the pharmacokinetics and dynamics of anesthetic drugs.
Box 11.3
Preoperative Neurologic Evaluation
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here