Management of Intracranial Hypertension and Mass Effect

For several decades, the management of intracranial hypertension has been a major focus of neurological and neurosurgical critical care. The concepts of intracranial pressure (ICP) monitoring and treatment were first developed in patients with head injuries and later applied to a wider range of disorders. The focus of ICP management has evolved over time. Initially, attention centered only on lowering ICP. Later ICP was treated in relation to blood pressure (BP) and cerebral perfusion pressure (CPP). In addition, the importance of pressure gradients within the intracranial vault that produce the tissue shifts and clinical deterioration discussed in Chapter 2 has been widely recognized.

The rationale for treating increased ICP or reduced CPP is based on the concept that either one eventually leads to cerebral hypoperfusion and global cerebral ischemia. Although this may be the case in the final stages of many disease processes, this concept leaves many questions unanswered. If critical reductions in CPP are the cause of ischemic damage why not raise BP to counteract the effects of elevated ICP? How can hyperventilation, which directly reduces cerebral blood flow (CBF), be useful in treating cerebral hypoperfusion? Why do treatments designed to lower ICP also reverse herniation syndromes that occur while ICP and CPP are normal? In the final analysis, although not questioning the deleterious effects of raised ICP, it has been difficult to prove that aggressive medical treatment improves outcome. Despite these polemic issues, few doubt that treatment of raised ICP in many circumstances is appropriate; it is the timing, method of monitoring, and specific treatments that are constantly under discussion.

The medical treatment of elevated ICP employs strategies that are designed to reduce the contents of the intracranial vault by reducing CSF, blood, and brain volume (largely water content). Surgical approaches remove the mass that is the proximate cause of raised ICP or decompress the cranium by removing a portion of the skull. These interventions treat both pressure gradients and tissue shifts. None of these interventions is used in isolation but rather in an integrated fashion designed to address the individual circumstances.

The physiology of the intracranial vault is reviewed in the previous chapter; this discussion of treatment builds on those principles. Each intervention is discussed individually and then an approach to combining them into an integrated treatment plan is presented. It is clear that the clinician is often faced with decisions for which there is insufficient evidence to indicate a clear course of action. The ultimate treatment plan often must be developed based on a combination of data, institutional bias, individual experience, and most importantly ongoing assessment of how the patient responds to any given intervention.

INDICATIONS FOR INTRACRANIAL PRESSURE MONITORING AND TREATMENT

The goal of intervention may be to reduce ICP, improve CPP, or mitigate the effects of tissue shifts. To meet the first two goals ICP monitoring is necessary in order to guide therapy. In the treatment of the parallel problem of intracranial tissue shifts, however, ICP monitoring is not necessary and may be falsely reassuring because clinical signs of herniation can develop while ICP remains normal (1,2). The response to treatment when dealing with tissue shifts only can be monitored with serial clinical examinations and imaging studies.

The value of ICP monitoring lies in its ability to reflect compromised cerebral perfusion, monitor for deterioration in difficult-to-assess patients, prevent iatrogenic elevations in ICP, provide prognostic information, and assess the response to treatment. Monitoring should be considered for any patient in whom it can be surmised that intracranial compliance is reduced and ICP might be elevated, and in processes known to be associated with delayed or progressive intracranial hypertension. Indications for ICP monitoring (Table 3.1) have been most studied in severe traumatic brain injury (TBI) and in which evidence-based guidelines have been published (3) and later revised (4, 5, 6, 7, 8 and 9).

TABLE 3.1. Indications for intracranial pressure monitoring in severe head injury (Glasgow Coma Scale <9)

Abnormal CT scan (hematoma, contusion, edema, compressed basal cisterns
Normal CT scan and two or more of the following:
	Age >40
	Unilateral or bilateral motor posturing
	Systolic blood pressure <90 mm Hg
From McDowell ME, Wolf AV, Steer A. Osmotic volumes of distribution. Idiogenic changes in osmotic pressure associated with administration of hypertonic solutions. Am J Physiol 1955;180:545-558, with permission.

These guidelines by their specific reference to traumatic lesions are of limited value in other diseases. The natural history of ICP changes in other conditions is not as well understood owing to a lack of data from routine ICP monitoring. In general, a Glasgow Coma Scale (GCS) score of 8 or less in the context of a large intracranial mass and radiographically observed tissue shifts has been used as an indication to instituting ICP monitoring because it is presumed that the degree of tissue shift reflected by a poor clinical state is likely be associated with increased ICP. Computed tomography (CT) signs that suggest (but by no means assure) the presence of elevated ICP include: obliteration of sulci of the cerebral hemisphere and, more sensitively, of the basal cisterns (10); tissue shifts across the midline; and hydrocephalus.

Another, albeit indirect indication for monitoring, arises when it becomes necessary to sedate a patient to a degree that limits neurological assessment or where an intervention may itself increase ICP (e.g., weaning of hyperventilation, general anesthesia for systemic surgery, etc.). In some circumstances the insertion of a ventricular catheter has therapeutic benefit by draining CSF while it is used for monitoring. Finally, ICP measurement can provide important prognostic information in certain conditions (11,12).

Medical and surgical measures to lower ICP are also effective in reducing tissue shifts and herniation. This statement is supported by the empiric observation that the clinical findings of herniation may improve, even if transiently, with these interventions and indeed, rapid improvement in these findings is the strongest indication of survival (13). These clinical improvements may be substantial and sustained as a recent observational study demonstrated (14). The mechanism responsible for clinical improvement is not entirely understood but is presumed to be the obvious one; namely, that the degree of tissue displacement and distortion is reduced.

For clinical purposes, ICP is usually measured in the supratentorial compartment. It also can be assessed from the lumbar subarachnoid space if free communication exists among brain compartments, but lumbar recordings may be inaccurate because of the distensibility of the spinal subarachnoid space. Continuous lumbar recording has found its main use in cases of pseudotumor cerebri. If intracranial mass lesions or noncommunicating hydrocephalus are present, this procedure, or course, can be dangerous and may result in the downward herniation of brain structures.

Practical Aspects of Intracranial Pressure Monitoring

Location

Invasive ICP monitoring devices can be placed in extradural, subdural, intraparenchymal, intraventricular, or intraspinal locations. Extradural and subdural devices are now infrequently used since the readings are not always reliable, primarily as a result of technical limitations of the devices. When the ventricles are of sufficient size many clinicians prefer the use of ventricular catheters because they provide the added ability to treat elevated ICP.

The decision about where to place an intraparenchymal device can be problematic. All other issues aside the preferred location is the right frontal lobe because it is felt that placing the catheter tip in that region does the least clinical ascertainable damage. However, when a mass lesion is present the issue is more complicated. It is not uncommon to measure a pressure gradient between the two hemispheres if monitors are placed bilaterally (15,16). There is little guidance in the literature and in most studies of ICP there was no consistent placement of monitors ipsilateral or contralateral to mass lesions. When using a monitor to define the ICP and CPP goals of therapy it is important to take into account whether the monitor is ipsilateral or contralateral to the mass lesion, and consider modifying treatment thresholds accordingly.

Types of Monitoring Devices, Their Advantages and Disadvantages

The gold standard for ICP measurement remains a ventricular catheter coupled to an external transducer through a continuous column of fluid in low compliance tubing. The setup is identical to those used for arterial and pulmonary artery pressure monitoring. (A similar apparatus also can be used to monitor pressure from the spinal subarachnoid space.) In the late 1970s and early 1980s a hollow bolt (“Richmond bolt” or screw) that was screwed into the skull and communicated directly with the subarachnoid space was popular but not always reliable. A number of more modern devices have been introduced in which the transducer is placed inside the skull and the pressure measurement transmitted to an electronic interface via a fiberoptic or electrical cable.

Each of these systems has inherent advantages and special problems. Fluid coupled systems allow the transducer to be rezeroed periodically after placement, thus avoiding inaccurate measurements owing to drift of the electromechanical transducer. This is not possible with implantable transducers, the only means of rezeroing and calibration being removal and reinsertion. (Drift was an important concern with earlier versions of implantable transducers and it was recommended that the devices be changed every few days; as the technology has improved this has become less of an issue.) Fluid coupled systems also require that the integrity of the fluid column be maintained in order to have accurate readings. The column must be contained within semirigid noncompliant tubing. Air bubbles within the column dampen the waveform and result in underestimation of the pressure. In addition, the catheter can become clogged with clot or debris, resulting in inaccurate readings. Implantable transducers do not suffer from these limitations but the older fiberoptic cables were
subject to breakage and signal loss. The greatest drawback to fluid-coupled systems is the potential for infection that gains access to the cranium through the inner and outer surfaces of the connecting tubes.

Additionally, when using an external transducer it must be physically level with the site of the desired pressure measurement. Common locations include the vertex, brow, or tragus. The tragus is frequently chosen as an indicator of midventricular level. Any change in the height of the patient’s head with respect to the transducer requires a repetition of the leveling process. Affixing the transducer to the side of the patient’s head may avoid inaccuracies caused by changes in head height; however, head rotation then can introduce a similar error. With implantable systems, the transducer is zeroed to atmospheric pressure once, just prior to insertion and it is then placed in the compartment of interest. This arrangement allows for accurate pressure readings regardless of head position.

Complications

The two major complications of ICP monitoring are intracranial hemorrhage and infection. Both are extremely rare when using implantable transducers. The incidence of major hemorrhage with intraventricular catheters is on the order of 1% to 2% and has been attributed to coagulopathy or difficulty with catheter placement.

Intraventricular catheters carry a small but significant risk of ventriculitis, the reported incidence ranging from 4% to 10% (17,18). Many centers routinely administer antibiotics prior to placement of an intraventricular catheter (IVC) and continue to administer them in an attempt to prevent infection. The choice of agents varies widely, ranging from agents specifically directed at skin flora to broad-spectrum antibiotics. The efficacy of prophylactic antibiotics has not been established and recent nonrandomized controlled studies have failed to demonstrate any such benefit (19,20). Prophylactic antibiotics are usually not administered when implantable devices are inserted.

Several factors are thought to increase the risk of infection with the use of IVCs. Early on, it was identified that the duration of monitoring was the major risk factor and the infection rate was seen to rise exponentially after 5 to 7 days (17). More recent experience with use of catheters tunneled under the scalp has called this observation into question. Now it is common to leave catheters in place up to 2 to 3 weeks if they are carefully handled by experienced personnel. Another important risk factor for infection appears to be the number of times that the integrity of the system is violated, either inadvertently, or in order to sample the CSF or flush debris or clot through the tubing. Because the best means of monitoring for and identifying infection is through sampling CSF, this observation presents a dilemma. A compromise is reached in some centers by routinely sampling CSF every other day to monitor for infection, whereas others do so only in the presence of fever or if the catheter is to be left in place for a prolonged period of time. Interpretation of laboratory analysis of CSF samples to look for signs of ventriculitis can be difficult. Many patients have blood mixed in the CSF. In addition, there is often an inflammatory reaction to the catheter, blood in the CSF, or as part of the disease process that is difficult to distinguish from an inflammatory response to an infection. Therefore, the most useful components of the CSF analysis when evaluating for ventriculitis are the glucose concentration, Gram stain, and culture.

An IVC easily can become obstructed with clot in the presence of intraventricular hemorrhage. This situation is evident as a steady rise in ICP over hours, loss of the ballistic waveform, and an inability to drain CSF out of the catheter. Flushing the catheter with preservative-free saline frequently solves the problem, but infection risk is raised and reobstruction is common, especially in the presence of substantial amounts of intraventricular blood. The instillation of thrombolytic agents has been suggested in this situation but obviously
carries the risk of recurrent hemorrhage. The safety and efficacy of this maneuver is currently under investigation (21,22). It is important to note that failure of CSF drainage in the setting of rising ICP also can result from poor catheter positioning, displacement of the catheter from the ventricular system because of tissue shifts or collapse of the ventricles from overdrainage. These possibilities can be evaluated through the use of imaging studies.

PHYSIOLOGIC APPROACHES TO THE TREATMENT OF RAISED INTRACRANIAL PRESSURE

Cerebral Perfusion Pressure

The initial goal of suspected or known intracranial hypertension is to insure that CPP is not compromised during the evaluation and initiation of treatment (Chapter 2). This holds particularly for avoidance of systemic hypotension, the adverse consequences of which have been best documented in head injury where systolic blood pressures below 90 mm Hg during the first several hours are associated with substantially worse outcome (23). (See also Chapter 12 for a discussion of the management of head injury.) It follows that overzealous treatment of systemic hypertension should be avoided in the setting of suspected intracranial hypertension.

The optimal CPP in the setting of elevated ICP is not known. Cerebral autoregulation maintains a nearly constant CBF over a CCP range of approximately 50 to 150 mm Hg in normal adults. As noted in Chapter 2, autoregulation is disrupted in a number of pathologic states, including stroke, head injury, and subarachnoid hemorrhage (24, 25, 26 and 27). The loss of autoregulation causes CBF to become pressure passive (i.e., CBF varies with CPP). The best data regarding the minimum tolerable CPP come again from TBI. Several studies have suggest improved outcome when CPP is maintained above 70 mm Hg (12,28,29). However, more recent studies using intermediate physiologic measurements such as CBF and brain tissue PO₂ have indicated that adverse changes occur at a more conservative level of CPP below 50 to 60 mm Hg (3,31).

It is apparent from the preceding comments that the relationship between BP and ICP is complex. When autoregulation is preserved, systemic hypotension causes cerebrovascular vasodilation and an increase in CBV, which in turn can raise ICP if intracranial compliance is poor. Similarly, an elevation of BP produces cerebrovascular vasoconstriction, potentially reducing ICP through a decrease in blood volume. However, as noted, autoregulation is frequently disturbed following brain injuries (25, 26 and 27); in these cases ICP may vary in parallel with BP. The recommendations for the clinicians can be summarized as a need for close attention to manipulations of BP and ICP because their interaction is potentially deleterious but at the same time unpredictable. As a general rule, extremes of BP are avoided but the precise limits of these extremes cannot be stated with confidence. It may be more appropriate not to manage BP in isolation but to manipulate BP to achieve a target CPP in the range of 60 to 80 mm Hg (see the following).

Intracranial Pressure Versus Cerebral Perfusion Pressure Management

From the preceding comments (and Chapter 2), controversy exists regarding the appropriate parameter to guide the management of patients with severe TBI, ICP, or CPP. In a randomized prospective trial CBFand ICP-targeted management protocols were compared (32). The CBF-targeted protocol was designed to achieve optimal CBF, through treatment of elevated ICP and the elevation of BP to maintain a CPP greater than 70 mm Hg. The ICP-targeted protocol focused on control of ICP and did not raise BP unless CPP fell below 50 mm Hg. The CBF-targeted protocol reduced the frequency of oxygen desaturation in jugular venous blood samples (used in this study as a surrogate for cerebral ischemia), but there was no difference in neurological outcome. Interpretation of these results was confounded by a fivefold increase in the frequency
of adult respiratory distress syndrome in the CBF-targeted protocol that offset any benefit in cerebral protection.

Other data suggest that a CPP of 50 mm Hg is adequate. When monitoring brain tissue PO₂ using commercially available probes, brain tissue oxygen pressure remains fairly steady until CPP falls below a threshold of 50 to 60 mm Hg (31,33). This also suggests that the CPP threshold below which cerebral oxygen delivery is compromised is closer to 50 than 70 mm Hg.

Many clinicians choose to modify treatment based on both ICP and CPP. In this approach, CPP is maintained above a threshold (approximately 60 mm Hg) at all times by raising BP and at the same time ICP is treated if it exceeds 20 to 25 mm Hg.

Threshold for Treatment of Raised Intracranial Pressure

The appropriate level for treatment of elevated ICP has been studied best in head injury; presumably, the same general principles apply to other conditions. Multivariate analysis of more than 1,000 severely injured patients has indicated that a high proportion of hourly ICP readings greater than 20 mm Hg is an independent predictor of poor outcome (12). It is recommended in the Guidelines for the Management of Severe TBI that active ICP treatment commence when ICP exceeds 20 to 25 mm Hg and that decisions about treatment of elevated ICP take into account serial clinical examinations and CPP (3).

The absolute value of the ICP or CPP should not be the only determinants of treatment. Mass effect with tissue shifts can compress the brainstem without necessarily elevating ICP. Indeed, clinical signs of herniation may occur in stroke patients before ICP becomes elevated (1,11). Similarly, herniation can occur with a normal ICP in TBI and ICH (2,34). Compression or obliteration of the basal cisterns is associated with poor outcome in head injury (35) and ICH (36) independent of ICP.

Patients who have a rapidly deteriorating neurological examination or clinical signs of herniation (dilated pupil and posturing) should be empirically and aggressively treated while undergoing further diagnostic studies along with placement of an ICP monitor. Continued treatment for suspected increased ICP without a monitor or for minor tissue shifts should be undertaken with caution because no intervention is without risk.

GENERAL MEASURES FOR THE TREATMENT OF RAISED INTRACRANIAL PRESSURE

A number of routine medical and nursing interventions have a substantial impact on ICP. It is important that all personnel involved in the care of critically ill neurological and neurosurgical patients be thoroughly familiar with the principles that guide patient management and develop appropriate routine practices (Table 3.2). The following discussion addresses
the management of medical problems and intensive care unit interventions that impinge on ICP. This management is a prelude to the active treatments of raised ICP that are elaborated on in the next section.

TABLE 3.2. Routine measures for patients at risk for intracranial hypertension

Mechanical ventilation
	Premedicate for intubation with ultra-short-acting intravenous anesthetic agents
	Avoid hypoxia and hypercarbia
	Use PEEP as needed to maintain oxygenation
	Watch for CO₂ retention with sedation
	If suctioning of chest physiotherapy causes plateau waves, premedicate with intravenous lidocaine (0.5-1.0 mg/kg), thiopental (˜1-3 mg/kg) or etomidate (˜0.1-0.3 mg/kg)
Patient positioning
	Elevate head of bed—optimal height variable determine for each patient
	Avoid jugular compression
		Neck in neutral position
		Avoid constricting endotracheal tube and tracheostomy tube ties
Blood pressure
	Treat CPP below 60-70 mm Hg with vasopressors and correction of hypovolemia
	Avoid CPP >90-100 mm Hg—consider antihypertensives
Treat fever aggressively
Prophylactic anticonvulsants when appropriate
Fluid and electrolytes
	Avoid excessive free water, no hypotonic fluids
	Maintain normal volume status
	Encourage hyperosmolar state with 1.25%-3% saline and/or induce free water clearance with diuretics or mannitol
CPP, cerebral perfusion pressure; PEEP, positive end-expiratory pressure.

Aspects of Respiratory Physiology and Intubation

The initiation and use of mechanical ventilation can raise ICP during intubation as a result of transiently inadequate oxygenation, CO₂ retention, coughing, use of positive end-expiratory pressure (PEEP) and during suctioning. The latter two maneuvers cause increased intrathoracic pressures to be transmitted to the cranial cavity through venous and CSF pathways.

Hypoxia and hypercarbia are potent cerebral vasodilators that can have profound effects on ICP, particularly when intracranial compliance is poor. Avoiding these detrimental circumstances requires constant assessment of respiratory status as well as careful attention to maneuvers that could raise PCO₂ or impair oxygenation, such as sedation, weaning mechanical ventilation, and suctioning. Continuous measurement of arterial oxygen saturation end expiratory carbon dioxide concentration is appropriate in virtually all circumstances of raised ICP.

Several factors conspire to raise ICP during intubation: hypoxia, hypercarbia, and direct tracheal stimulation (37). Etomidate is effective in blocking this reflex (38). Intravenous (i.v.) lidocaine (1.0 to 1.5 mg/kg) also has been recommended (39), although data supporting its use are lacking (40). In addition, short-acting i.v. anesthetic agents (thiopental 1 to 5 mg/kg or etomidate 0.1 to 0.5 mg/kg) also reduce brain metabolic rate and theoretically improve tolerance of a transient fall in CPP should it occur. Etomidate generally is preferred over thiopental because it is less likely to lower blood pressure.

The use of PEEP has long been a concern in patients with raised ICP. Although the pulmonary benefits of PEEP are clear, the risk, in terms of further elevation of ICP is largely theoretical for several reasons. Increases in PEEP and mean airway pressure certainly can be transmitted to the thoracic venous and CSF compartments and ultimately to the intracranial vault (41). The pressure transmitted from the airways and lungs to other thoracic structures depends on pulmonary compliance in large part. The effect on ICP is minimal in situations where high levels of PEEP are most likely to be used (a patient with stiff, poorly compliant lungs) (42,43). In most circumstances of serious respiratory failure combined with the presence of an intracranial mass, the detrimental effects of hypoxia almost always outweigh the theoretical risks of using higher levels of PEEP. Thus, it seems prudent not to forego its use simply because of raised ICP. Of course, direct measurement of ICP allows one to gauge the impact of PEEP and various other respiratory settings providing the means to optimize PEEP to achieve adequate oxygen saturation while avoiding a negative impact on ICP.

Coughing that arises spontaneously or in response to suctioning or chest physiotherapy is known to raise ICP momentarily by 30 to 40 mm Hg even in normal individuals (Fig. 3.1). In brain-injured patients coughing, or even tracheal manipulation without inducing a response, may cause a more sustained and greater rise (if compliance is poor), and may induce plateau waves that can markedly reduce CPP. One effective way to prevent stimulating a further rise in ICP is to suppress coughing with intermittent boluses of i.v. lidocaine
(0.5 to 1.0 mg/kg), thiopental (approximately 1 to 3 mg/kg), or etomidate (approximately 0.1 to 0.3 mg/kg), which are administered prior to suctioning or chest physiotherapy (44,45). Repeated doses must be used cautiously because they can lead to significant drug accumulation. Furthermore, completely blocking coughing may not be desirable because it may lead to the accumulation of pulmonary secretions, atelectasis, pneumonia, and an inability to effectively manage pulmonary infections. Brief hyperventilation with 100% oxygen before suctioning also may block the rise in ICP caused by tracheal manipulation (46, 47 and 48).

FIG. 3.1. Changes in intracranial pressure with cough.

Patient Positioning

The position of the head and neck influence ICP through a number of mechanisms. In particular, the height of the head relative to the heart influences intracranial arterial and venous pressures. Arterial pressure is reduced as a result of the work necessary to pump blood against gravity to the head and at the same time venous drainage is enhanced. It has been reported that ICP is generally lower when the head of the bed is elevated to 30 degrees compared to the horizontal position (49, 50 and 51), but these results have been inconsistent. Another analysis has suggested that the optimal degree of head elevation varies from patient to patient and therefore should be determined for each individual (51). A recent study of patients with large stroke and edema found that both ICP and CPP were maximal when the patient was supine (52). They found that the rise in ICP was more than offset by the increase in perfusion pressure caused by the brain and heart being at the same level.

Perhaps more mundane is the problem of compression of the jugular veins by turning of the head or compressing the neck too tightly. Venous drainage can be impaired raising cerebrovenous pressure and ICP. Whether unilateral compression has an adverse effect that is more than transient is not known. On the other hand, compression of both jugular veins causes a slow and progressive rise in ICP that disappears rapidly when compression is released (Fig. 3.2). This response is so consistent that it may be used as a bedside test to assess the functioning of an intracranial ICP monitor (a variant of the Queckenstedt maneuver formerly used to detect spinal block). Therefore, it is imperative to avoid interventions that compress the jugular veins, such as securing an endotracheal or tracheostomy tube too tightly or inserting large catheters into the jugular veins.

FIG. 3.2. Changes in intracranial pressure with jugular compression.

Blood Pressure Control

Only broad guidelines can be offered here based on theoretical considerations of autoregulation, ICP, and CPP, as already discussed. If BP is “normal” but CPP is below 60 or 70 mm Hg, BP should be augmented. Raising BP in a euvolemic patient by using fluid infusion is slow, often ineffective, and can result in serious complications such as congestive heart failure, hyponatremia, and worsening of cerebral edema (32). The nature of the fluid used perhaps is of as much importance, as discussed in more detail in the following. The most rapid and consistent therapeutic hypertensive response can be achieved by the use of vasopressor agents such a phenylephrine or dopamine while administering moderate amounts of i.v. fluid (150 to 250 mL/h) if there is coexisting hypovolemia. In addition, several treatments for raised ICP can significantly lower BP through direct hypotensive (barbiturates, propofol) or diuretic (mannitol) effects. It is important to take measures to prevent these problems by replacing
urinary losses when using mannitol, and to correct BP by administering vasopressors when faced with drug-induced hypotension.

There has been concern that the use of vasopressors may lead to cerebral vasoconstriction and consequently, cerebral hypoperfusion. When the blood-brain barrier (BBB) is intact, norepinephrine and phenylephrine do not appear to alter CBF (53), but the results are less clear when the barrier is disrupted. Despite this theoretical concern, our clinical experience suggests that vasopressors can be used almost with impunity in instances of raised ICP and that hypotension is to be avoided at all costs. A potential beneficial effect of vasopressors in aborting plateau waves has been discussed already (54).

On the opposite extreme, defining a safe upper limit for BP and CPP is as important and perhaps a more frequent clinical problem in the hours after hemorrhage and stroke. Cerebral edema and hemorrhage may result when the autoregulatory range is exceeded (approximately 150 mm Hg in normotensive adults). Furthermore, autoregulation may be impaired following brain injury, thereby lowering its upper limit (i.e., CBF may rise to extreme levels at modestly elevated levels of blood pressure). Moderate degrees of hypertension are known to worsen edema following brain trauma (55); therefore, caution in allowing hypertension is warranted. Consequently, it is probably prudent to keep CPP less than 90 to 100 mm Hg when elevating BP with vasopressors. If CPP spontaneously exceeds this level, antihypertensive medications should be administered.

Fever

As temperature rises, so does cerebral metabolism and oxygen requirements (56). Hyperthermia also increases CO₂ production and, if mechanical ventilation is not adjusted appropriately in sedated patients, this can lead to hypercarbia and an increase in ICP. Elevated temperature worsens ischemic damage in experimental animals and, conversely, hypothermia is protective (57,58). In clinical work it has been found that fever is an independent predictor of poor outcome in patients with ischemic stroke (59, 60, 61, 62 and 63). Thus, although control of fever has yet to be shown to improve outcome in circumstances of raised ICP, it seems prudent to take measures to reduce it; antipyretics are routinely administered to febrile patients with CNS insults. Unfortunately, agents such as acetaminophen and surface cooling with air blankets often are ineffective in reducing fever (64). Anecdotal experience suggests that other antipyretics such as ibuprofen may be more effective, but they carry a small risk of bleeding. The use of indomethacin as an antipyretic is appealing because it independently lowers CBF and ICP in head-injured patients (65,66). Recently introduced intravascular cooling devices appear to be highly effective in controlling fever. They operate by circulating cooled saline through a central venous catheter, thereby directly cooling the blood.

Only gold members can continue reading. Log In or Register to continue