Injury to the Brain

Injury to the brain can occur from both endogenous and exogenous causes. Exogenous injury most commonly results from automobile-related trauma, although external blows to the head and face, penetrating wounds, gunshot injury, and impact from falling are also possible. At initial impact, primary mechanical disruption of intracranial tissues may occur. The force of impact can rupture axons, blood vessels, and other neural elements. Primary mechanical injury may then initiate numerous secondary pathophysiologic sequelae, such as metabolic alterations in neuronal or glial cells, impairment of vascular supply to normal tissue (ischemia), impairment of cerebrovascular autoregulation, hemorrhage (intraparenchymal, intraventricular, extradural, or subdural), irritation (seizure generation), obstruction of the ventricular system, edema formation, production of physiologically active products, and, most often, increased intracranial pressure (ICP). As ICP reaches life-threatening levels, shifts of brain parenchyma or brain herniation can occur. Differing types of brain herniation include subfascial, rostral and caudal transtentorial, and foramen magnum herniation. Although some forms of herniation are treatable, brain herniation through the foramen magnum is often deadly.

The intracranial nervous system resides in a unique physiologic environment. Protected by, but also confined within, the bony limits of the cranium, nervous tissue exists with cerebrospinal fluid (CSF) and blood. Disease of the brain parenchyma or other intracranial tissues alters this physiologic equilibrium. As primary injury can be prevented but not treated per se, treatment of brain injury is directed toward treatment of secondary pathophysiologic events and basic life support. Important secondary pathophysiologic consequences of intracranial injury are discussed further so that a fundamental understanding of the intracranial environment following injury can be appreciated during the treatment process.

Cerebral Edema

Many intracranial diseases result in or are associated with brain edema. With acute injury, brain edema becomes maximal between 24 and 48 hours after injury, but may persist for a week or more [1]. Brain edema has been categorized as vasogenic, cytotoxic, or interstitial based on cause and anatomic areas of involvement [2]. Any or all of these types of edema may be present in an animal with brain disease.

Cytotoxic edema (intracellular edema) results from failure of cellular energy, with resultant failure to extrude sodium from within the cell. Intracellular water increases and cells swell. This edema most often results from disease processes such as toxicity, ischemia, or hypoxia.

Interstitial edema is defined as increased water content of the periventricular white matter owing to movement of CSF across the ventricular walls in instances of hydrocephalus. Periventricular white matter is reduced owing to the disappearance of myelin lipids secondary to increases in hydrostatic pressure or decreases in periventricular blood flow in the white matter [3].

Vasogenic edema is the most common form of edema associated with CNS neoplasia. This type of edema results from vascular injury secondary to physical disruption of the vascular endothelium or functional alterations in endothelial tight junctions. Differences in transmural pressure gradients result in extravasation of fluid from cerebral vessels to the extracellular fluid (ECF) spaces of the brain [4]. These abnormalities allow for fluid to move from the vascular to perivascular spaces. Areas of the brain where ECF is normally enlarged provide a natural conduit for fluid movement. Increases in intravascular pressure owing to loss of autoregulation, vascular obstruction, or hypertension (e.g., Cushing's response and cerebral ischemic response) may perpetuate edema formation.

Vasogenic edema migrates away from areas of vascular disruption via bulk flow [4]. Fluid movement depends on a balance between the opposing forces of capillary hydrostatic pressure and tissue-resistance pressure. Vasogenic edema usually spreads readily through the white matter, possibly because of the orderly arrangement of nerve fibers found there. Also, the movement of this type of edema may be related to the low capillary density and blood flow in normal white matter. Deep white matter of the involved cerebral hemisphere is preferentially affected.

Hemorrhage

Hemorrhage, either within or around the brain may result in rapid cerebral dysfunction often by alterations in cerebral volume (mass effect). In comparison with humans, subdural hematomas are uncommon in dogs [5]. Systemic coagulation abnormalities (e.g., thrombocytopenia) and hypertension may potentiate hemorrhagic potentials. Vascular damage from therapeutic intervention (e.g., radiation therapy) may also influence the incidence of tumor-related hemorrhage. Also, loss of cerebral vascular autoregulation may predispose to hemorrhage or infarction.

Ventricular Obstruction

Space-occupying intracranial disease may impinge on the ventricular system, resulting in obstruction of CSF flow. Common areas of obstruction, owing to the inherently small diameter of the ventricular system, include the interventricular foramen (connection between the lateral and third ventricles) and the mesencephalic aqueduct. Knowledge of normal CSF flow allows for determination of the location of the obstruction. Sequestration of CSF in an obstructed component of the ventricular system may result in increases in intracranial pressure and interstitial brain edema.

Increases in Intracranial Pressure

Intracranial pressure is the pressure exerted between the skull and the intracranial tissues. As the skull is relatively inelastic compared with the other intracranial tissues, ICP is determined primarily by changes in intracranial tissue volume and the compensatory ability of these tissues to accommodate for volume changes [1,6-8].

The three major components of volume residing within the intracranial space are the brain tissue (intracranial cellular elements), the cerebrospinal fluid, and blood [7]. Intracranial disease often increases the volume of one of these components. For ICP to remain normal, an increase in volume of one of these three intracranial components must be compensated for by a decrease in volume of one or both of the other components (intracranial compliance). Although compliance can initially aid in stabilizing ICP, it has finite limits. When compliance is exhausted, ICP will increase. This relationship may vary among individual patients and may be influenced by the extent and location of an intracranial mass as well as cerebral blood flow [9]. As ICP increases, the pressure within the intracranial space decreases cerebral blood flow. With decreased cerebral perfusion, neuronal ischemia, hypoxia, dysfunction, and death ultimately result.

Cerebral blood flow and CSF are in constant flux. ICP is dynamic and pulse-like, with a small pressure wave occurring almost simultaneously with each heartbeat [10,11]. This waveform is referred to as the intracranial pulse pressure wave (ICPPW). Periodic fluctuations may occur in response to normal body functions, such as coughing or abdominal straining, via increases in intrathoracic pressure and venous pressure impairing venous return from the intracranial space. With intracranial disease, ICP may be elevated persistently or episodically, resulting in persistent or episodic clinical abnormalities, respectively. As exemplified by Lundberg, periodic elevations in ICP (plateau waves) with ensuing return of ICP to normal levels can occur in humans with intracranial disease [12]. These periodic elevations are thought to originate from cerebral vasodilation secondary to decreased cerebral perfusion in patients with intact cerebrovascular autoregulatory capacity [13]. In this situation, single-time measurements of ICP may not accurately reveal these episodic elevations, and critical periods of raised ICP will be overlooked.

Numerous physiologic and nonphysiologic variables influence ICP measurements. These include anesthetic agents [14-18] and body weight [19]. The effects of anesthetic agents on ICP are many and varied, and the need for anesthesia to measure ICP can potentially alter values obtained. Specific effects of anesthetic agents on ICP have been reviewed elsewhere [14-18].

An absolute level wherein ICP is elevated has not been established. This may be the result of the inherent inaccuracies of the monitoring equipment and the normal variation in baseline ICPs. Most information from humans suggests that ICP higher than 15 to 20 mm Hg is elevated [20]. Cerebral blood flow may not be significantly decreased until ICP reaches about 30 mm Hg [21]. Experimentally, however, ICP was raised to very high levels (> 100 mm Hg) before brain death occurred [22]. We have seen ICPs as high as 30 to 40 mm Hg in dogs that subsequently have recovered from brain injury. Whereas some have suggested that the degree and persistence of ICP elevation is associated with outcome after head injury [20], others have suggested that ICP monitoring does not influence overall prognosis [23].

Clinical Effects of Intracranial Pressure Alterations

The major intracranial effect of increasing ICP is alteration of cerebral perfusion pressure (CPP) (Fig. 40-1) [9,21,24,25]. Cerebral perfusion is dependent on systemic blood flow and intracranial pressure expressed via the formula CPP = MABP - ICP (MABP = mean arterial blood pressure) [19]. For CPP to remain constant, the effects of increased ICP on blood flow to the brain must be reciprocated by increases in systemic blood pressure. Cerebral perfusion pressure is a determinant of cerebral blood flow (CBF) but is not always equivalent; in many instances, however, as CPP decreases so does CBF. Intracranial pressure and CPP, however, have an inverse relationship (Fig. 40-1). When CPP falls below 60 to 70 mm Hg, this creates physiologic abnormalities within the brain.

Figure 40.1. Schematic relationship between intracranial pressure (ICP) and cerebral perfusion pressure (CPP). As ICP increases, CPP ultimately decreases to result in clinical signs.

Blood flow to the brain is coupled to the cerebral metabolic rate. When CBF decreases, the brain recognizes the ischemia and evokes a spectrum of physiologic alterations known as the cerebral ischemic response [26]. These physiologic alterations are thought to emanate from vasomotor centers in the lower brain stem. Failure of adequate blood flow to remove CO2 from receptors in these centers increases local CO2 concentrations, which in turn, stimulates the sympathetic nervous system to increase systemic blood pressure. Systemic hypertension results as an attempt to maintain cerebral blood flow. Baroreceptors within the systemic vascular system recognize this hypertensive situation, and send this information to vagal centers also found in the lower brain stem. The resultant increase in systemic vagal tone reflexively causes bradycardia. The systemic hypertension and associated reflex bradycardia is commonly referred to as Cushing's reflex and may explain why many animals with intracranial disease have bradycardia.

As ICP increases and CPP subsequently decreases to a significantly low level, a large catecholamine release occurs [22]. This catecholamine release may result in myocardial ischemia known as brain-heart syndrome [11,27]. Clinically, arrhythmias are noted. Focal myocardial ischemia is the pathologic lesion, with focal white streaks seen grossly in the myocardium [27]. Histologically, myocardial degeneration is common. Brain-heart syndrome has been described with numerous intracranial and spinal pathologic lesions and in different species including dogs, sheep, cattle, horses, pigs, goats, and humans.

Intracranial pressure alterations are often responsible for clinical decline in many animals with brain disease. As stated previously, owing to the confining nature of the skull, changes in intracranial volume increase ICP when compensation is exhausted. With structural brain disease, the brain component usually enlarges as a result of neoplastic cell infiltration, edema, or inflammation. As the brain compartment volume increases, the CSF and blood compartments must decrease or ICP will increase. Compensation for increased brain tissue volume initially involves shifting of CSF out of the skull, decreased CSF production, and eventually decreased cerebral blood flow.

These compensatory mechanisms prevent increases in ICP for some undetermined period of time. In general, the more slowly the volume increase occurs, the more readily this volume increase is compensated. When compensatory mechanisms are exhausted, relatively small increases in intracranial volume result in dramatic elevations of ICP. At this time, clinical signs become apparent. Initial signs of increased ICP may be nonspecific and limited to alterations in mental status (progression to stupor and coma), cranial nerve dysfunction, and paresis. Unfortunately, clinical signs of increased ICP often become most obvious too late in the disease to allow for effective therapy.

Alterations in Cerebral Blood Flow (Ischemia and Hypoxia)

Because normal function of neurons depends on an adequate supply of oxygen, physical disruption of blood flow can significantly impair normal brain function. Cerebral vessels are directly responsive to PaCO2 concentrations, with cerebral blood flow coupled to cerebral metabolic rate [1]. As PaCO2 concentrations increase, cerebral vessels dilate to increase blood flow to the brain. The opposite effect is seen with decreased PaCO2. This effect of PaCO2 is a component of cerebral blood flow autoregulation. The cerebral vessels have the ability to change diameter in response to PaCO2 (chemical autoregulation) as well as blood pressure (pressure autoregulation) in order to maintain a relatively constant cerebral blood flow. Cerebral vessels change diameter through perivascular changes in pH as a direct result of PaCO2 concentrations similar to what occurs in the chemosensitive area of the medulla oblongata for stimulation of respiration.

If autoregulation is intact, hyperventilation to decrease PaCO2 will cause cerebral vasoconstriction, decreased cerebral blood volume, and subsequently, decreased ICP. Cerebrovascular autoregulatory capability is affected by a variety of intracranial processes. For example, local acidosis, common in many hypoxic and ischemic events, will disrupt local autoregulatory functions [28]. If chemical autoregulation is absent in the diseased brain, hyperventilation will not alter the vascular diameter in the affected area. In this instance, two situations are possible, both dependent on the premise that cerebrovascular autoregulatory capacity is absent owing to local disease.

As PaCO2 is decreased by hyperventilation, vessels in surrounding normal brain will constrict. Vessels in the damaged or abnormal area of the brain are already maximally dilated and, therefore, are unable to constrict. Because of the vasoconstriction in normal brain, vascular resistance increases in this area, shunting blood into the abnormal area (inverse steal or Robin Hood phenomenon) [28]. This has the positive effect of increasing cerebral blood flow to the abnormal and potentially hypoxic areas of the brain. The negative effect, however, is potentiation of hemorrhage and cerebral edema in the abnormal area owing to the increased flow.

The opposite may occur as PaCO2 increases. Vessels in normal brain surrounding the diseased area will dilate. Again, vessels in the abnormal area are already maximally dilated owing to loss of autoregulation. Vascular resistance will be decreased in surrounding normal brain, blood will be shunted away from abnormal areas, potentially decreasing hemorrhage and edema but resulting in further hypoxia (steal phenomenon) [28].

Autoregulation of cerebral blood flow can also occur in response to systemic changes in blood pressure in order to maintain relatively constant cerebral blood flow during times of both hypo- and hypertension [1]. This prevents underperfusion and resultant ischemia during times of hypotension and hemorrhage and edema at times of hypertension. In most instances, cerebral blood flow is maintained constant throughout shifts in systemic blood pressure of between 50 and 150 mm Hg [1]. Above and below these limits, cerebral blood flow depends directly on systemic blood pressure.

Whether autoregulation remains intact in brain disease is difficult to predict clinically. Global autoregulatory function of pressure has been assessed in human patients with head injury and found to be intact in about 69% [29]. It has been suggested that a crude clinical estimate of intact autoregulatory function is level of consciousness, with conscious patients thought to have intact autoregulatory capacity more often than do unconscious patients [1]. Local loss of autoregulation, however, is almost impossible to predict in the clinical setting without sophisticated testing that is not practically available.

Additionally, at least in the instance of exogenous head trauma, discrepancies can occur between the intact function of pressure versus chemical (PaCO2-responsive) autoregulatory capacities [29]. With head trauma, pressure autoregulation is often abnormal; however, PaCO2-responsive autoregulation remains intact. This is termed dissociative vasoparalysis and has important ramifications during treatment. As the cerebrovascular response to blood pressure changes are abnormal, a more intimate control of systemic blood pressure and central venous pressure is required to prevent large shifts of blood pressure and subsequent cerebral blood flow. While correction of hypotension is often necessary in the animal with concurrent head trauma and shock, hypervolemia should be avoided [25]. If pressure autoregulation is defective, significant increases in mean arterial blood pressure will increase cerebral blood flow and, ultimately, ICP [29]. If pressure autoregulation is intact, decreased blood pressure will cause cerebrovascular vasodilation, increasing cerebral blood flow, and ultimately, increasing ICP.

Pressure autoregulation may also falsely appear intact (false autoregulation) even when vasomotor paralysis is present [28]. In this situation, cerebral blood flow is found not to increase when systemic blood pressure is increased, suggesting intact pressure autoregulation. Because of associated cerebral swelling, however, further dilation of cerebral vessels is prevented. Therefore, no additional increase in cerebral blood flow is possible, regardless of blood pressure. All of these factors contribute to myriad intracranial blood flow alterations that can affect intracranial pressure.

Cerebral blood flow, therefore, is normally maintained through a combination of systemic blood flow (pressure) and cerebral vasculature autoregulation. Autoregulation is important to maintain a near constant cerebral blood flow over a range of systemic blood pressures. Brain injury may alter cerebral vascular autoregulatory mechanisms, thus making cerebral blood flow more dependent on systemic blood pressure. This results in inadequate perfusion of neurons during times of hypotension and overperfusion (possibly perpetuating edema formation) during times of hypertension (also referred to as luxury perfusion). Subsequent re-establishment of blood flow may also have possible detrimental effects (reperfusion injury) on spared brain through such processes as free radical formation and lactic acid accumulation.

Terminal Effects of Compartmentalized ICP Increases Brain Herniation

As intracranial volume continues to increase beyond the limits of compensation, ICP will increase so precipitously that shifts of brain parenchyma, termed brain herniation, will occur [8]. Coma and severe neurologic impairment are noted. Unfortunately, in many instances, brain herniation becomes a terminal event.

Five major types of brain herniation have been described; rostral or caudal transtentorial, subfascial or cingulate gyrus, foramen magnum, and herniation through a craniotomy defect [2,8]. Of these, caudal transtentorial, subfascial, and foramen magnum occur most commonly. Clinical signs of caudal transtentorial herniation are frequently the result of pressure distributed ventrally through the midbrain with subsequent compression of the oculomotor nerve. With unilateral herniations, an ipsilateral dilated pupil, unresponsive to light stimulation, may be seen. Monitoring for clinical signs of this type of herniation, therefore, should include periodic pupillary evaluations. If unilateral mydriasis is noted in this setting, immediate and aggressive attempts to decrease ICP should be instituted.

Foramen magnum herniation may occur quickly and results in respiratory arrest owing to associated pressure on the respiratory centers in the caudal brain stem. Foramen magnum herniation is invariably fatal, and surgical attempts at decompression after this type of herniation have not been helpful in affected dogs.

Treatment of Intracranial Pathophysiologic Sequelae

Effective treatment of the pathophysiologic sequelae of increased ICP depends on the determination and effective treatment of the primary disease. Unfortunately, in the situation of brain injury, the primary event has occurred prior to treatment. It is often necessary, therefore, to acutely treat the more immediately life-threatening secondary sequelae such as increased intracranial pressure to allow time for treatment of the primary disease. Treatment of secondary pathophysiologic events subsequent to intracranial injury is a more important determinant of acute survival than is treatment of the primary disease process. The following section provides an overview of treatment possibilities for these secondary pathophysiologic events. In-depth discussion of pros and cons of individual treatments have been discussed elsewhere [30-33].

One of the most important aspects of treatment of intracranial injury is to maintain appropriate cerebral perfusion, primarily by maintaining effective intravascular volume. This is accomplished through the administration of intravenous crystalloids and/or colloids. Mannitol has been shown to decrease blood viscosity, which may contribute to increased cerebral perfusion and decreases in ICP [34]. Decreasing blood viscosity will increase cerebral perfusion at the same level of pressure. Vasoconstriction will result, lowering cerebral blood volume and concurrently lowering ICP. Other hypertonic solutions appear to be less effective at lowering ICP as compared with mannitol, however, hypertonic saline administration has shown some benefit [35].

Treatment of intracranial hemorrhage varies depending on whether the hemorrhage has formed a discrete lesion that is increasing intracranial volume at the expense of the surrounding normal nervous tissue. In such an instance, surgical drainage may be necessary to bring about resolution. Although it is uncommon, traumatic subdural hematoma can be used as an example. As the lesion matures it becomes relatively hyperosmolar, resulting in imbibition of water and expansion of the lesion. The patient with traumatic subdural hematoma should, therefore, benefit from surgical drainage of the lesion [5]. It is also important to identify and treat any predisposing factors to hemorrhage such as hypertension or a bleeding disorder.

Craniotomy/Craniectomy

A major tenet of the Monro-Kellie doctrine is that intracranial contents are confined within the cranium. It would follow, therefore, that surgical removal of the skull would potentially decrease ICP. Unilateral or bilateral craniectomy has been used as a treatment for ICP elevations which cannot be decreased by the more conventional methods stated earlier [36-40]. Although skull removal alone may be beneficial, subsequent dural incision appears significantly more effective in lowering ICP. This has been shown clinically in humans and experimentally in dogs and cats [41,42]. Ultimate functional recovery, however, remains dependent on the underlying primary brain damage caused by the initial disease process. Postoperatively, the extent and magnitude of the decrease of ICP subsequent to surgery remains uncertain.

Craniectomy and durotomy have been shown to lower ICP by 15% and 65%, respectively, in dogs and cats, and humans [38,41,]. Intracranial pressures in normal dogs approached atmospheric pressure when a bilateral or unilateral rostrotentorial craniectomy and durotomy were performed. Adequate decompression in animals with structural disease would be suggested if intraoperative ICP approached similar levels.

If large amounts of skull are removed, a cranioplasty may be necessary. Otherwise, postoperative scarring may result in brain compression similar to that which occurs because of the formation of laminectomy membrane after extensive laminectomy, resulting in spinal cord compression. Cranioplasty can be delayed until life-threatening elevations of ICP are stabilized. Some humans had improvement in persistent neurologic signs after delayed cranioplasty, suggesting that clinical signs after initial recovery may result from both the initial injury and surgical scarring [52].

Other Treatments with Questionable Benefits

Much debate remains regarding the most appropriate head position for patients with head injury. Many current recommendations from human traumatologists suggest a horizontal or neutral position to maintain cerebral perfusion. Head elevation to 30° above heart level, however, has been shown to decrease ICP primarily by facilitating venous drainage [43,44]. Opponents suggest that head elevation may decrease cerebral perfusion and, therefore, be detrimental to brain functions. Reduction of ICP with this relatively simple treatment may be beneficial in combination with other ICP-decreasing measures.

If chemical autoregulation remains intact, hyperventilation can decrease ICP owing to the established effects of PaCO2 concentrations on cerebral blood flow [1,9]. Hyperventilation is performed to maintain the PaCO2 concentration between 25 and 35 mm Hg.

Cerebrospinal fluid aspiration may be used as the initial means to decrease ICP [45]. This is most helpful in humans if a ventriculostomy is present and if ICP elevations do not exceed 30 mm Hg [9]. The use of routine CSF collection alone for this purpose is not recommended in animals with suspected increased ICP owing to the increased risk of precipitating brain herniation [46]. In one study in animals, however, no increased risk of brain herniation was found in dogs and cats with intracranial disease after CSF collection [47].

Corticosteroids have received much use in the treatment of spinal trauma, and have been recommended as a treatment for elevated ICP [48,49]. Although corticosteroids have shown to be beneficial by reducing cerebral edema in brain tumor patients, caution should be exercised when using corticosteroids for brain injury. One study in rats has suggested that corticosteroids may be advantageous in brain injury [49]; however, corticosteroids may not be efficacious in head trauma and are known to perpetuate neuronal damage if ischemia is operant [50,51]. Corticosteroid administration may increase blood glucose, a factor that may negatively influence outcome after head injury [52]. Also, the onset of beneficial effects of decreasing cerebral edema may be delayed too long to be helpful in acute elevations of ICP. Unfortunately, until further data are reported, the benefit of corticosteroid use in head injury and elevated ICP remains nebulous. Additionally, other treatments such as barbiturate-induced coma and hypothermia require further investigation [24,53-55].