Coma, Vegetative State, Brain Death, and Increased Intracranial Pressure

Published on 03/03/2015 by admin

Filed under Neurology

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 9984 times

16 Coma, Vegetative State, Brain Death, and Increased Intracranial Pressure

Coma

Clinical Vignette

A 76-year-old man was found unconscious in bed at home. His wife informed the emergency physician that he had a history of prostate cancer but no risk factors for vascular disease. She denied knowledge of any recent head injury. He was totally unresponsive to verbal and painful stimuli. Neurologic examination showed he was comatose with pinpoint pupils. Eye movements to doll’s-eyes maneuvers were full and conjugate, and cold caloric stimulation of the ear canals produced ipsilateral tonic deviation of the eyes without nystagmus. There were bilateral withdrawal responses of his extremities to noxious stimuli, and bilateral Babinski signs.

Intravenous administration of 0.4 mg naloxone produced dramatic change, with full awakening within a few minutes. Results of a subsequent brain computed tomography (CT) were normal, confirming that there was no evidence of intracerebral, subarachnoid, or subdural hemorrhage, cerebral infarction, or mass lesions. When asked about narcotic use, the patient stated that he was no longer able to tolerate the pain of metastatic prostate cancer; he had taken an overdose of an opioid analgesic.

The first vignette reflects a common scenario for anoxic–ischemic brain injury secondary to cardiac arrest. The degree of recovery usually depends on a number of factors that include age, extent, or duration of the ischemic insult and the initial presenting neurologic examination. A detailed history, review of in-field records, and serial examinations is an essential first step in guiding prognostic predications and determining the need for further workup. The second vignette illustrates the classic case of “toxic-metabolic” induced coma. Despite profound unresponsiveness and very miotic pupils, neurologic examination demonstrated retained brainstem reflexes and CT results were normal. Although pontine hemorrhage is often suspected in a comatose septuagenarian with pinpoint pupils, intact reflexive eye movement strongly indicated a metabolic cause. The prognosis in such cases, in the absence of secondary hypoperfusion or anoxia, is overwhelmingly favorable.

Consciousness is the state of awareness of internal and external stimuli and is manifested by the ability to react to these stimuli through thought or by directed physical movement. Coma is the loss of awareness of stimuli and the ability to react voluntarily to them. Full consciousness of one’s self and the surroundings can be disrupted partially without total loss of arousal or wakefulness and is sometimes referred to as a state of stupor. Varying degrees of consciousness can therefore be roughly delineated by the specificity or accuracy of the response to a particular stimulus, and terms such as confusion, obtundation, or stuporous are used to reflect this. However, the most useful tool in following patients with altered mentation remains the exact description of patient behavior and reactions to specific stimuli. For example, it is preferable to indicate that a patient stays alert without stimulation but is unable to give the exact date and location or follow two sequential commands than to only say that the patient is confused or clouded. Nevertheless, defining terms may provide uniformity in meaning when encountered in patient records. Obtundation describes a condition in which repeated stimuli are needed to draw the patients’ attention back to a task. Stupor is a state of extreme inattention in which wakefulness and minimal interaction with the examiner can be achieved only by repeated or constant stimulation. Delirium is an acute confusional state often involving sympathetic nervous system overactivity with attention marred by hyperexcitability. Tachycardia, perspiration, hypertension, and hallucinations may all be features of delirium.

Because the examining physician can only infer thought from patients’ actions (e.g., speech or movement), a reliable and reproducible physical examination is essential in evaluating the comatose or stuporous patient. The neurologist must make every effort to establish the presence or absence of a directed nonreflexive response and judge its quality. For example, in basilar occlusion when the lesion is restricted to the basis pontis a “locked-in” syndrome occurs with the patient seemingly having no directed response to stimuli, yet on close examination may blink in an exact fashion to instructions and indirectly answer questions appropriately. Partially preserved voluntary vertical eye movements may also be present. Similarly, in patients with severe acute polyneuropathies such as Guillain–Barré syndrome, consciousness is preserved but difficult to assess and quantify secondary to profound peripheral weakness.

The Glasgow Coma Scale assesses and quantifies the degree of consciousness across three measures: response to verbal commands, response of eye opening, and the nature of motor movements in response to verbal or physical stimuli. Those not responding to verbal commands or opening their eyes with a Glasgow Coma Scale score of 8 or less are defined as being in a coma (Fig. 16-1). The Glasgow Coma Scale is one of the primary predictors of long-term outcomes, especially in cases of head trauma.

Prevalence of the different etiologies of coma varies depending on the population surveyed. For example, head trauma and intoxicants are the major causes in registries based on densely populated high-crime areas. Stroke and cardiac events are the leading etiologies in suburban areas with retirement communities. Overall, trauma, stroke, diffuse anoxic–ischemic brain insult (secondary to cardiorespiratory arrest), and intoxicants are the leading mechanisms for coma. Infections, seizures, and metabolic–endocrine disorders account for the remaining cases (Fig. 16-2).

States that affect cognition and attention without affecting wakefulness such as the various degenerative dementias (characterized by progressive cognitive deterioration) and focal brain lesions (which cause restricted cortical dysfunction) do not fit the definition of coma. Sleep is a normal patterned physiologic disconnection of the cortex from external stimuli and is discussed elsewhere (Chapter 15).

Evaluation and Treatment of the Comatose Patient

The initial evaluation of a patient in coma must occur simultaneously with its management. Any delay in treatment while waiting to determine the exact cause is not acceptable. Clearing the airway and ensuring adequate ventilation and oxygenation with a bag mask or intubation, if needed, must be addressed immediately. Management of hypotension must be prompt, especially in suspected cases of increased intracranial pressure (ICP). Hemodynamic collapse should never be attributed to an intracranial process, and cardiac or circulatory causes need to be sought. These form the “ABCs of coma management”: airway, breathing, and circulation (Fig. 16-3). Immobilizing the neck until a cervical spine injury is excluded is also important in cases of suspected trauma.

Emergent evaluation of comatose patients requires the following blood studies: a full blood count, glucose level, serum chemistry, toxicology screen, liver profile, thyroid function tests, arterial blood gases, and cultures. Creatine kinase and troponin measurements, in conjunction with electrocardiography, are important for excluding myocardial infarction and transient cardiac arrest. Anticonvulsant drug levels and an electroencephalograph (EEG) can help identify patients with nonconvulsive status epilepticus.

The immediately treatable causes of coma are hypoglycemia and narcotic intoxication. These can be managed promptly, once oxygenation and hemodynamic status are stable. Infusion of 100 mg thiamine must precede the infusion of 50 mL of 50% dextrose in water as a precaution against Wernicke encephalopathy. This is postulated to be due to osmotic or metabolic damage to the mammillary bodies and the medial thalamus exerted by glucose, which, in the absence of thiamine, cannot be transported and metabolized in the tissue. When narcotics overdose is suspected, such as in comatose patients with miotic pupils, 0.4 mg intravenous (IV) naloxone, a central opioid antagonist, improves the level of consciousness within minutes. Repeated doses may be needed to maintain wakefulness and reverse respiratory depression. Caution should be exercised for known or suspected opioid dependency as abrupt or complete reversal of opioid effects by repeated doses may precipitate an acute withdrawal state. In such instances, only supportive care should be provided once the diagnosis is made. Administration of flumazenil, a pure benzodiazepine antagonist (0.2 mg IV), given three to four times, can improve the mental state and reverse respiratory depression in benzodiazepine overdose. As with naloxone, it should be used cautiously in those with a history of long-term benzodiazepine use or dependency as it can precipitate seizures. It should generally be avoided in patients with epilepsy and those at risk of seizures.

Urgent intravenous antibiotic coverage is indicated for febrile patients because time is crucial in treating meningitis and septicemia (Chapter 48). Lumbar puncture should be performed only after brain imaging has excluded mass lesions that could lead to herniation.

Assessment of the comatose patient should include examination of the skin. Rashes may indicate streptococcal or staphylococcal meningitis, bacterial endocarditis, or systemic lupus erythematosus. Purpura may indicate meningococcal meningitis, a bleeding diathesis, or aspirin intoxication. Skin dryness suggests anticholinergic or barbiturate overdose, whereas excessive perspiration indicates cholinergic poisoning, hypoglycemia, and other causes of sympathetic overactivity. Dark pigmentary changes in the axillary and genital areas suggest adrenal insufficiency, whereas doughy pale skin is typical of myxedema. Renal failure may present with urea salt crystal skin condensations or “urea frost.” Facial or basal skull fractures often cause ecchymosis around the eyes (raccoon eyes or panda bear sign) or in the mastoid area (Battle sign). Extremities must be examined for needle and track marks that indicate intravenous drug abuse.

The patient’s breath may be uremic, fruity as in ketoacidosis, or have the musty fishy odor of hepatic failure. Fever may indicate meningitis or encephalitis but also occurs with sympathomimetic or tricyclic (anticholinergic) overdose and drug or alcohol withdrawal. Occasionally a low-grade fever occurs with subarachnoid hemorrhage or brainstem lesions.

Focal neurologic signs on initial examination may implicate a structural lesion as the cause of coma and should be followed closely for signs of evolving herniation until brain imaging can be performed. Other causes of focal presentation are compensated old brain injuries clinically reemerging as a result of seizures, toxins, or metabolic derangements. However, metabolic disorders including nonketotic hyperosmolar hyperglycemia, hypoglycemia, and hepatic coma may cause focal seizures or lateralizing neurologic signs without focal brain lesions. Evolving signs of increased intracranial pressure or herniation must be treated promptly regardless of cause; there is no use in waiting for brain CT results or other tests.

Electroencephalography is often helpful in evaluating patients with altered consciousness or coma. An abnormal tracing makes psychogenic coma unlikely. EEG detects nonconvulsive or absence status, which can present de novo without a history of epilepsy. Although nonspecific, diffuse EEG background slowing correlates with metabolic derangements and focal slowing with localized structural brain disease. Hepatic and other metabolic encephalopathies may show triphasic waves. In herpes simplex encephalitis, periodic lateralized epileptiform temporal lobe discharges are often seen and support the clinical diagnosis. Finally, when a basis pontis lesion with the “locked-in syndrome” is suspected, a normal EEG shows that the patient is alert despite limited or no obvious response to stimuli.

Prognosis

Determining the prognosis of an individual comatose patient is a difficult task. Statistical numbers given to patients’ families as measures of outcome probabilities often are difficult to apply in relation to their loved one. The focus usually shifts to the chances of recovery, no matter how limited, rather than the likelihood of severe disability. A statistical grid or flow chart cannot be relied on to decide each individual case, and numerous factors, including cause of the coma, the evolution of the neurologic examination, age, comorbidities, and the religious or philosophical beliefs of the patient and the family must be considered.

Recovery from drug intoxication, barring ischemic brain injury from secondary hypoxemia or circulatory collapse, is usually good, with rare mortality or instances of severe disability. In hepatic and likely other metabolic comas, only brainstem dysfunction, with disruption of oculocephalic reflexes and loss of pupillary reactivity, increases the likelihood of poor prognosis or death. The duration of the coma and absent localizing motor responses do not exclude a good recovery and probably only reflect persistent metabolic derangement. Hepatic encephalopathy, to a large extent, is caused by the accumulation in the portal system of ammonia derived primarily from enzymatic activity of intestinal bacteria upon nitrogenous material and amino acids. Liver failure leads to shunting of portal vein ammonia into the systemic circulation and the brain not having proper detoxification. The effects of ammonia and other toxic elements on the brain include astrocyte swelling with cytotoxic brain edema, altered cerebral blood flow (CBF), and the accumulation of inhibitory neurosteroids and inflammatory cytokines. Numerous precipitants have been identified (anemia, constipation, dehydration, excessive dietary protein, gastrointestinal bleeding, metabolic alkalosis, hypoglycemia, hypothyroidism, hypoxia, infection, sedatives) and aggressive treatment is necessary to reverse the encephalopathy. Removal of intestinal ammonia with nonabsorbable disaccharides (lactulose) and antibiotics such as neomycin or metronidazole are commonly used and relatively effective. The mechanism of coma and brain edema in acute fulminate hepatic failure is less understood but likely involves some of the same pathophysiologic mechanism already described. However, a large proportion of these patients do not respond to treatment and have poor outcomes. Treatment consists of the above outlined measures and the preservation of cerebral perfusion pressure (CPP) by close monitoring and control of ICP. Liver transplantation, however, remains the most effective and immediate treatment to control brain edema and ICP and to reverse coma.

In most instances, coma from head trauma has a better outcome than that from nontraumatic mechanisms or cardiac arrest. Although severe head trauma has a mortality of approximately 50% within the first 48 hours, few surviving patients remain in a permanent vegetative state and most progress toward some degree of functional improvement. Those who remain vegetative usually succumb within 3–5 years. There are rare reports of patients who awaken after a prolonged vegetative period and show some return of functionality. None, however, return to their premorbid status or even an independent state. Signs that correlate with a poor prognosis after head trauma are age older than 60, bilateral pupillary abnormalities or absent oculocephalic reflexes at initial examination in a relatively stable patient. Large volumes of contused brain, large intra- or extra-axial hematomas, and lack of intracranial pressure response to conventional medical treatment (usually associated with compression of basal cisterns on CT) also betoken a poorer prognosis (Fig. 16-4).

Anoxic–ischemic causes of coma have a mortality rate of up to 60–70%, with generally only 10–15% of patients returning to a good functional status. The lack of bilateral pupillary responses for more than 6–12 hours correlates highly with poor functional outcome and death. Absent oculocephalic (vestibule–ocular or caloric) responses after 24 hours likely have the same prognostic value. In patients who do retain or regain pupillary reactivity, the absence of at least reflexive flexor motor movements on day 1, or some withdrawal movement on day 3, also holds a poor prognosis, with less than 10% chance of recovery to a state of even moderate disability. Lack of spontaneous eye opening or of localizing motor movements on day 7 holds the same grim prognostic significance. Myoclonus status epilepticus (generalized multifocal unrelenting myoclonus) correlates with severe ischemic damage to the cortex, brainstem, and spinal cord and is strongly associated with in-hospital mortality or a vegetative state.

Other laboratory findings have been shown to reliably predict a poor prognosis and can be used to assist in the evaluation. These include EEG tracings (without sedatives or metabolic abnormalities) showing patterns of complete suppression, burst suppression or periodic discharges upon a generalized flat background, absent N20 somatosensory evoked responses after 24 hours, and neuron-specific enolase >33 µg/L beyond the first day.

Vocalizations or any verbal response early within the first day of the causative event indicates a relatively good chance of functional improvement within a year.

These observations can guide families and staff toward the best course of action for each patient. Often the examination is changing or unclear. Consequently, further waiting and repeated evaluations, although stressful for the family, result in more certainty in the appropriateness of the eventual decisions taken. Those showing unfavorable prognostic signs on day 1 and who show no improvement or evolution in their neurologic examination are not likely to do well. However, for individuals who exhibit evolving neurologic function, the duration of observation needs to be extended and the final determination of outcome delayed, even if the initial examination shows no major interactive or directed function.

Persistent Vegetative State

Clinical Vignette

A 23-year-old woman was an unrestrained driver in a “head-on” automobile accident. She was ejected 30 feet through the windshield and sustained major head trauma. On arrival in the emergency department, she was totally unresponsive, hypotensive, and tachycardic. Brain CT demonstrated generalized cerebral edema and diffuse subarachnoid hemorrhage. Neurologic examination showed her to be unresponsive even to painful stimuli, other than some rare nonpurposeful right leg movements. Pupils were minimally and inconsistently reactive, and she had a dense left hemiplegia. Subsequent magnetic resonance imaging (MRI) demonstrated bilateral focal contusions of the cerebral hemispheres, shear injury of the splenium of the corpus callosum and brainstem edema. Four months later, after no improvement in her clinical state, she was diagnosed with persistent vegetative state (PVS).

Cases of PVS frequently involve young individuals with healthy cardiovascular and pulmonary systems. Before the recognition of these patients’ hopeless outcomes after a few months of no improvement, some were maintained in chronic care facilities or their parents’ homes for years, with the unrealistic hope that they might someday regain the ability to meaningfully interact.

The vegetative state, minimally conscious states, or post-coma unawareness are terms that describe a state of preserved brainstem and hypothalamic functions with absent or insufficient cortical function to sustain awareness of environment and self. Wakefulness is by definition preserved, and patients may cycle through sleep stages. There is no behavioral evidence of even the simplest reproducible response. Patients may startle, look about, occasionally move a limb, shift position, or yawn, but none of these actions are consistently in response to a specific stimulus (Fig. 16-5). Even the most basic voluntary actions, such as chewing and swallowing, are lost. Once reversible metabolic or exogenous causes have been eliminated, the condition is called persistent when it lasts without change for more than 1 month. It is considered permanent when lasting more than 12 months for traumatic brain injury and more than 3 months for nontraumatic causes. After these time limits, the chance of recovery is exceedingly low and at best progresses to severe disability.

As with coma, individuals in a posttraumatic PVS have better chances of recovery than cases due to medical causes. Nevertheless, one third of all these patients die within the first year. Of patients with head trauma, one third regains consciousness after 3 months and approximately one half in a year. Overall, one fourth of all patients with traumatic PVS recover to a level of moderate disability, mostly those who regain awareness within 3 months.

Of patients with nontraumatic PVS, more than 50% die within a year and only approximately 10–15% regain consciousness by the third month. Most remain severely disabled, with rare improvement in functional status. If the condition persists longer, there is minimal chance of any significant functional recovery. Neither age nor cause seems to correlate with eventual recovery, but of those recovering, the younger patients show somewhat better outcomes than older patients, at least in locomotion and self-care. After 3 months, once PVS is considered permanent, withdrawal of nutritional support and hydration can be discussed with the family. Many physicians consider such withdrawal acceptable, based on the contention that nutritional support in such cases constitutes medical treatment that is neither alleviating suffering nor improving the overall condition. When viewed as a human or legal rights issue, such reasoning becomes more complex and less applicable as a general principle.

Increased Intracranial Pressure and Cerebral Herniation

Rostrocaudal Signs of Brain Compromise

As pressure from a hemispheric lesion increases, patients gradually move from being easily roused, but inattentive, to sleepy and unable to maintain wakefulness, then to coma—a state of absent voluntary or directed response to external or internal stimuli.

The ascending reticular formation, excited by sensory input, mediates arousal and consciousness to the cortex via the thalamic nuclei. Lesions that cause coma are at one of three levels along the neuraxis: bilateral cerebral cortex, the thalami, or the upper brainstem. The classic concept of herniation and coma produced by brain mass lesions pertains to a hemispheric process that ultimately causes “rostrocaudal” deterioration of function, gradually coursing down the hemispheres into the medulla. Although these “stages” rarely manifest symmetrically in a strict and clearly delineated sequential pattern, this paradigm remains useful for evaluating deteriorating patients with evolving neurologic signs. In addition to the level of consciousness, important physical examination elements include pupillary size and reactivity, reflexive eye movements, limb posturing, and breathing pattern (Fig. 16-4, Table 16-1).

Pupillary Reactivity and Eye Movements

When pressure onto or across the diencephalon exists, loss of wakefulness results, but patients may transiently continue to withdraw appropriately from uncomfortable stimuli and to resist passive limb movements. Pupils are small and retain reactivity, although at times blunted and subtle to detect. Although there is no visual fixation, eye movements are conjugate and full. As pressure mounts across the thalami onto the mesencephalon, pupillary and eye movement abnormalities appear. Involvement of CN-III or its nucleus initially causes irregular and poorly reactive pupils (corectopia). Eventually, eye movements are disrupted by CN-III or CN-VI lesions or from involvement of the medial longitudinal fasciculus (MLF). The MLF, a paracentral dorsally located tract coursing up the vestibular nuclei to the CN-III nucleus, maintains conjugate eye movements either initiated voluntarily in the waking state or induced reflexively from cervical or vestibular inputs in comatose patients. This pathway provides the basis of doll’s-eyes testing or caloric stimulation testing of the semicircular canals. An intact MLF system keeps the eyes from moving passively when the examiner rolls the head to one side. The eyes remain in their primary position in relation to the examiner or seem to move to the opposite side in relation to the head rolling. With unilateral caloric stimulation of the ears, the eyes deviate conjugately to one side or another, depending on the water temperature used for irrigation. The direction of the convection current induced in the semicircular canals by different temperatures determines the direction of eye movement. With the head maintained in the neutral position, cold water causes the eye to deviate to the side of the stimulated ear while warm water causes deviation away from the stimulated ear. Disruption of the MLF system causes an abnormal or absent responses of these reflexive eye movements (Fig. 16-6). Therefore oculocephalic testing checks the integrity of a large portion of the brainstem from the vestibular nuclei to the mesencephalic third-nerve nucleus.

Breathing

Respiratory patterns also change with worsening levels of consciousness in coma (Fig. 16-7). The earliest breathing alterations are Cheyne–Stokes respirations. Hemispheric forebrain structures serve to regulate breathing by mechanisms independent of CO2 accumulation. With bilateral cerebral cortex damage, this breathing control is lost, and CO2-driven breathing is accentuated with only modest CO2 accumulations, thus inducing an increased rate and depth of respiration. This reactive hyperpnea leads to an eventual decrease in arterial CO2 and, again without forebrain control, loss of respiratory drive. The ensuing apnea then allows CO2 to reaccumulate and the cycle to repeat itself, resulting in hyperpnea of a crescendo–decrescendo pattern, alternating with intervening episodes of brief apnea.

Midbrain and upper pons lesions cause hyperventilation with a constant rate and amplitude, without periods of apnea. The reasons for the so-called “central neurogenic hyperventilation” are unclear but are unlikely to be of purely neuronal origin. Lung congestion caused by immobility and poor airway protection likely play a major role. Hypothalamic and midbrain lesions engender increased sympathetic activity, which in turn promotes capillary fluid seepage, worsening lung congestion and, in extreme cases, pulmonary edema.

Injury to the lower half of the pons damages the respiratory control system, possibly generating apneustic breathing; a pattern of prolonged end inspiratory pauses alternating with end-expiratory pauses of several seconds, without the crescendo–decrescendo pattern of Cheyne–Stokes breathing. Further damage causes this pattern to fragment into an irregular, unpredictable rhythm of varying amplitude, intermixed with pauses of variable length. Ultimately, destruction of the centrally located dorsomedial medullary respiratory center causes total cessation of breathing, even before circulatory collapse occurs.

Variation from the Classic Rostrocaudal Paradigm

Unilateral cerebral lesions can cause asymmetric pressures leading to medial temporal lobe (uncal) herniation through the tentorial incisure, with direct compression of the midbrain (Fig. 16-8). In this case, the diencephalic features described above are not seen; instead, rapid loss of consciousness is immediately followed by a decerebrate posture. This is usually heralded by a compressive palsy of CN-III as it exits the ventral aspect of the midbrain and runs between the superior cerebellar and posterior cerebral arteries across the top of the tentorium. The initial sign is pupillary dilation, followed by ophthalmoplegia and ptosis with downward displacement of the abducted globe. Hemiplegia ipsilateral to the lesion may result from compression of the contralateral anteriorly located pyramidal tract against the anterior edge of the tentorium (Kernohan notch phenomenon). Further increase in pressure causes stretching of the pontine penetrators off the basilar artery or venous congestion with paramedian hemorrhages and usually irreversible worsening (Fig. 16-8). A sudden worsening and increase in ICP may also result from posterior cerebral artery compression with occipital lobe infarction. Finally, the cingulate gyrus may herniate beneath the falx cerebri, compressing the ipsilateral or contralateral anterior cerebral artery and causing infarction in its distribution (Fig. 16-8).

Infratentorial lesions within the brainstem tegmentum or secondary to brainstem compression, such as with cerebellar mass lesions, cause coma abruptly as a result of the almost immediate involvement of the reticular activating system. Lesions involving the midbrain result in oculomotor- or infranuclear-type ophthalmoplegia or both, with fixed irregular or midposition dilated pupils. If the lesion involves the pons without the midbrain, sympathetic fibers running up to the CN-III nucleus are destroyed, and pupils are pinpoint in size but still reactive to light. An internuclear ophthalmoplegia occurs from bilateral MLF lesions, but vertical oculocephalic movements, controlled by the tectal midbrain, are preserved. Cerebellar mass lesions can cause forward brainstem displacement and may produce findings similar to those described for isolated pontine lesions or may cause cerebellar tissue crowding and herniation upward around the midbrain through the tentorium or downward around the medulla through the foramen magnum. Cerebellar tonsillar herniation down through the foramen magnum causes sudden respiratory and circulatory arrest, without gradual signs of evolving brainstem dysfunction.

Coma from metabolic disease rarely conforms to the typical rostrocaudal stages and often shows concurrent findings pertaining to different nervous system levels. For example, hypoglycemia can cause unconsciousness with decerebrate posturing but preserved oculocephalic responses and pupillary reactivity. In metabolic comas from opioid overdose, pupils are tiny but reactive, even though respiratory drive may be obliterated. Also, the oculocephalics are intact, despite drug-induced pinpoint pupillary changes mimicking a pontine hemorrhage, as the vignette at the beginning of this chapter illustrates. Finally, pupillary reactivity remains relatively resistant to metabolic effects; when other brainstem signs are absent, the presence of brisk pupillary reactivity suggests a nonstructural metabolic or toxic cause.

Treatment of Increased Intracranial Pressure

When signs of increased ICP are evident, emergent treatment is needed to halt progressive obtundation or coma and to avoid herniation and irreversible brain injury. Delaying treatment to first determine the underlying pathophysiologic mechanism is of minimal benefit. An invasive, but highly effective, method to decrease ICP is CSF drainage through an external ventricular drain if applicable. Noninvasive therapeutic modalities available to acutely control increasing ICP act mainly by three mechanisms: vascular, osmotic, and metabolic. With induced vasoconstriction, cerebral blood flow (CBF) diminishes, reducing total cerebral blood volume with a subsequent decrease in ICP. Agents that create a hyperosmolar intravascular compartment in relation to brain tissue induce water movement down a gradient from cells and the interstitium into plasma, reducing total brain water content, volume, and pressure. This affects both normal and, likely to a lesser extent, damaged edematous brains. Osmotic agents may have other beneficial actions, such as preserving cerebral perfusion pressure (CPP), decreasing blood viscosity, decreasing CSF production, and the scavenging of free radicals.

Increased metabolic demand from injury or illness may increase blood flow and enhance the delivery of oxygen, which, in turn, may cause increased free radical production and cell injury. Decreasing the metabolic drive helps limit blood flow and decreases the need for tissue oxygen delivery.

Hyperventilation

The least invasive, most effective, and fastest mechanism to decrease ICP is induced hypocapnia produced by active hyperventilation. Intact brain tissue with preserved cerebrovascular autoregulation mediates this effect. Autoregulation is abolished in damaged tissue or ischemic areas, which do not respond to hyperventilation. The goal of intact cerebrovascular autoregulation is to keep CBF stable under conditions of normal slight fluctuations in systemic pressures. Factors such as fever, hypoxemia, and ischemia induce the need for increased CBF and are mediated through the vasodilatory effect of hydrogen ions from lactic and carbonic acid accumulation. The brain is therefore sensitive to CO2 levels, and increased CO2 pressure produces an almost linear increase in CBF. Similarly, decreased CO2 pressure causes vasoconstriction and a decrease in CBF. Correspondingly, intracranial blood volume and ICP decrease, provided that enough intact tissue exists to mediate this response.

The response to hyperventilation is almost immediate, with its peak effect occurring within 30 minutes. Effectiveness diminishes over the course of hours to a day, limiting the utility of hyperventilation as a long-term option to control increased ICP. Despite short-term effectiveness, cerebral vasoconstriction may eventually cause increasing ischemic brain injury, especially in vulnerable areas with previously compromised blood flow. Outcomes can worsen with its prolonged use. Therefore, hyperventilation is confined to brief intervals to urgently control sudden increases in ICP. When more permanent treatments are instituted (i.e., control of agitation, blood pressure regulation, or surgery), its use should be halted.

The relatively safe target level of PCO2 is approximately 25–35 mm Hg; lower PCO2 levels risk compromising cerebral blood perfusion with subsequent ischemia, which would eventually cause a seemingly paradoxic further increase in ICP. The usual approach is to increase respiratory rate but not tidal volume. Higher lung volumes increase intrathoracic pressure and total cerebral blood volume by compromising venous return from the brain to the heart.

Osmotic Agents and Diuretics

The osmotic agent mannitol (20–25%solution) has been the cornerstone of ICP management for decades and continues to be an effective and relatively safe treatment when used judiciously. It is usually administered rapidly over minutes (0.75–1.0 g/kg), with its initial effect usually occurring within 20 minutes and lasting approximately 6–8 hours. With repeated doses there is diminishing effectiveness and a shortening of the response duration over days. Depending on the clinical response and patient status, subsequent mannitol doses of 0.25–0.5 g/kg are administered every 4–6 hours. Concomitant measurement of serum osmolality, and sometimes ICP, is indicated. Repeated mannitol use requires extreme caution to avoid hypotension or a hyperosmolar state from too frequent or brisk diuresis. Using mannitol sparingly avoids the pitfalls of hypotension and electrolyte imbalance that can seriously undermine patient care. A staged approach in increasing the osmolarity initially to 295–300 mOsm/kg and then gradually to 310–320 mOsm/kg may be helpful. Osmolarities higher than 320 mOsm/kg are dangerous and add little to further control ICP. Standing orders for “maintenance” doses should be avoided; the clinical situation and osmolality should guide subsequent dosing. If the clinical examination, ICP and osmolarity are stable, no extra doses may be required. Serial measurement of renal function, electrolytes, and osmolarity are indicated every 4–6 hours. Potassium depletion is common, and frequent replacement is needed. Daily fluid balance and body weight measurements are essential to maintain the goal of a euvolemic hyperosmolar state. Only isotonic fluids should be used to replenish intravascular volume. Hypotonic fluids and free water should be avoided to avert recurrent brain edema.

Low doses of diuretics, such as furosemide (10–20 mg), may be used alone or with mannitol to enhance or hasten the response, especially when transient increase in intravascular volume, such as in congestive heart failure, may be problematic.

Hypertonic solutions have been used for more than 50 years in the treatment of ICP but, unlike mannitol, have only recently gained widespread use. Animal models and small human studies, mostly involving traumatic brain injury, show hypertonic solutions to have a more robust and longer effect in reducing ICP when compared to mannitol. They have the advantage of preserving CPP by repleting intravascular volume and enhancing microvascular circulation and well-being. Its effects on outcomes, however, have been variable and remain unestablished. There is no unified ICP protocol for hypertonic solution administration, with concentrations varying from 3% to 29.2% and given either as a bolus or as a continuous infusion to keep sodium levels about 145–155 mmol/L. Potential side effects include hypernatremia and electrolyte imbalance, non-anion gap acidosis, coagulopathies, and possibly renal failure.

General Measures

In the care of patients with or at risk for increased ICP, several other modalities help prevent escalating ICP and improve outcome.

Blood Pressure

The parameters of blood pressure control in treating ICP remain hard to define. Cerebral autoregulation is absent in damaged or ischemic brain, and perfusion of these areas is directly related to systemic pressure. Relatively high systemic pressures may increase edema in areas of a disrupted blood–brain barrier, whereas low systemic pressures may compromise perfusion and cause further tissue ischemia. Cushing response, a reaction to increasing ICP, is a sympathetically mediated increase of systemic blood pressure with reflex bradycardia, which may play a protective role in preserving CPP (ICP minus mean arterial pressure). However, if untreated, it may eventually lead to increased tissue edema in damaged areas of the brain. Conversely, overaggressive treatment may exacerbate ischemia.

The principle of preserving CPP within the range of functioning cerebral autoregulation may be the best guide, as no definite directives exist. Because the range of cerebral autoregulation shifts upward with chronic hypertension, the parameters for each patient may vary. Ideally, arterial blood pressure should be kept near its premorbid range, determined if possible from previous documented measurements. If arterial pressure is low, then increasing it with isotonic fluids or mild pressor agents is indicated. In cases in which ICP is stable, systemic hypertension is treated independently. If both systemic pressure and ICP are increased in conjunction, then an attempt at reducing ICP (e.g., with mannitol) should be initiated first and blood pressure monitored closely for resolution of reflex hypertension. If there is no response within a few minutes, a gentle attempt to bring the systemic blood pressure down is made, preferably using agents with no cerebral vasodilatory effects, that is, diuretics, angiotensin-converting enzyme inhibitors, and β-sympathetic blockers. Reliance strictly on CPP alone may not always produce favorable outcomes, and following all parameters simultaneously may provide greater benefit.

When an ICP monitor is used, the goal is to preserve CBF above the level that ensures adequate cerebral oxygen metabolic needs while keeping CPP above 60 mm Hg and approximately 70 mm Hg and avoiding persistent or recurring ICP measurements of greater than 20 mm Hg.

Brain Death

Clinical Vignette

A 56-year-old man suddenly collapsed at home after experiencing severe chest pain. His wife called the emergency technicians, who found him pulseless and cyanotic. ECG demonstrated ventricular fibrillation, but he was successfully defibrillated. After an airway was established and 100% oxygen was given, he was transported to the ED. There, neurologic evaluation showed that he was unresponsive to any form of communication. His pupils were dilated and not reactive to light stimulation. Decerebrate posturing was noted to suctioning and to noxious stimulation. Bilateral Babinski signs were present. He eventually became flaccid, and cold caloric vestibular stimulation showed no ocular response. The next day, he developed generalized myoclonus and continued to require cardiac and full respiratory support. Three days later, there was no change in his neurologic status. An apnea test showed no respiratory response to induced hypercarbia. Although he was declared brain dead, his wife asked that further testing be performed to confirm the clinical diagnosis. An EEG demonstrated electrocerebral silence, and she agreed to have life support withdrawn.

This vignette is the classic example of a patient with prolonged cardiorespiratory arrest resulting in devastating diffuse cerebral ischemic damage. Until a precise determination of brain death is established, there are many medical and legal issues to address in caring for individuals who have no effective residual brain function despite cardiopulmonary function maintained with modern intensive care therapies.

In most medical communities, a person is considered dead once there is irreversible and total cessation of all brain function, regardless of a continuing functional circulatory system. The cause of brain damage must be clearly elucidated by history, examination, or medical tests before the diagnosis of brain death is entertained. Intoxicants, sedatives, and hypothermia may present similarly to brain death but are potentially reversible and must always be considered if the cause is not clear and well documented. In many countries, including the United States, “brain death” constitutes a legal definition of death, and all life support measures can be halted.

When caring for an individual, it is best to respect the family’s wishes, religious or personal, regarding the timing of discontinuing life support. It is important to continue to explain the situation’s finality and that circulatory collapse will invariably occur within hours to days of the onset of this clinical picture. The widespread difficulty in obtaining organs for an ever-growing list of patients awaiting transplant procedures makes it of paramount importance to broach the subject of organ donation as soon as possible as it is vital to harvest organs early. The physician who has an established relationship with the patient’s family is perhaps the one best to initiate such discussion before involving the transplant team.

Brain Death Criteria

The criteria for brain death vary among states and countries. Usually the determination is clinical, with testing used only as an ancillary or confirmatory measure. The following are generally accepted principles of brain death determination:

Mitigating Factors

When a severe cerebral insult is suspected but brain death determination cannot be confirmed because of confounding issues, the utmost should be done to correct for these specific factors before brain death assessment can proceed. For example, hypothermia is treated with a warming blanket to bring and maintain core body temperature above 36.5°C. Fluid, and at times vasopressor agents, are administered for patients with systolic blood pressures lower than 90 mm Hg.

Patients with chronic hypercapnia secondary to lung disease such as chronic obstructive pulmonary disease have a higher respiratory center PCO2 threshold, and a PCO2 of approximately 60 mm Hg may not necessarily drive the chemoreceptors, even with a functioning brainstem. The CO2 level may be allowed to climb to approximately 80 mm Hg, but such levels risk ensuing acidosis with direct cardiodepressor effects as well as arrhythmias and hypotension. Therefore, it is preferable in these instances to obtain confirmatory tests to bolster the diagnosis without risking untoward iatrogenic complications.

Although most centers in the United States uphold the general outline of the principles mentioned above, there are numerous variations and differences concerning how best to ensure diagnostic certainty. Most medical centers do not require confirmatory tests. The number of evaluations and the time span between them also differ. Many institutions require a brain death evaluation by two attending neurologists at different times and their presence at the apnea test. The specific brain death criteria and protocol for each medical center must be consulted before the evaluation is begun and a diagnosis is substantiated.

Additional Resources

Bullock R, Chestnut F, Clifton G, et al. Guidelines for the management of severe head injury. J Neurotrauma. 2000;17:471-553.

Conrad GR, Sinha P. Scintigraphy as a confirmatory test of brain death. Semin Nucl Med. 2003;33:312-323.

American Electroencephalographic Society. Guideline three: minimum technical standards for EEG recording in suspected cerebral death. J Clin Neurophysiol. 1994;11:10-13.

Wijdicks EFM, Hijdra A, Young GB, et al. Practice Parameter: Prediction of outcome in comatose survivors after cardiopulmonary resuscitation (an evidence-based review): Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology. 2006;67:203-210. A systematic review of the outcomes in coma after cardiopulmonary arrest identifying the factors that most reliably predict a poor prognosis

Hypothermia After Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. New England Journal of Medicine. 2002;346(8):549-556.

Levy DE, Bates D, Corona JJ, et al. Prognosis in non-traumatic coma. Ann Intern Med. 1981;94:293-301. Landmark paper that details the examination of postanoxic coma and the various findings that predict outcomes

Muizelaar JP, Marmarou A, Ward JD, et al. Adverse effects of prolonged hyperventilation in patients with severe head injury: a randomized clinical trial. J Neurosurg. 1991;75:731-739.

Piatt J, Schiff SJ. High dose barbiturate therapy in neurosurgery and intensive care. Neurosurgery. 1984;15:427-444.

Presidents Commission on Guidelines for the Determination of Death. Neurology. 1982;32:395.

Posner JB, Saper CB, Schiff ND, et al. The Diagnosis of Stupor and Coma. (Contemporary Neurology Series. 71). 4th ed. New York: Oxford University Press; 2007. An expanded and exhaustive edition of the classic monogram on the pathophysiology of coma and its various etiologies. Detailed description of the associated vascular and anatomic pathology. It ends with a small section on the approach to the unconscious patient and treatment

Qureshi AI, Kirmani JF, Xavier AR, et al. Computed tomographic angiography for diagnosis of brain death. Neurology. 2004;24(62):652-653.

Report of the Quality Standards Subcommittee of the American Academy of Neurology 1995;1-7.

Sazbon L, Zagreba F, Ronen J, et al. Course and outcome of patients in vegetative state of non-traumatic aetiology. J Neurol Neurosurg Psychiatry. 1993;56:407-409.

Teasdale G, Jennet B. Assessment of coma and impaired consciousness: a practical scale. Lancet. 1974;ii:81-84. The first description of the Glasgow Coma Scale that has acquired widespread use and has subsequently been shown to be a reliable tool in predicting outcome in head trauma

The Multi-Society Task Force on PVS. Medical aspects of the persistent vegetative state: parts I and II. N Engl J Med. 1994;330:1499-1508. 1572-1579

Vahedi K, Hofmeijer J, Juettler E, et alfor the DECIMAL, DESTINY, and HAMLET investigators. Early decompressive surgery in malignant infarction of the middle cerebral artery: a pooled analysis of three randomized controlled trials. Lancet Neurol. 2007;6:315-322. Data pooled from three different studies showing that decompressive hemicraniectomy more than doubles the chance of survival from “malignant” middle cerebral artery stroke, and likely improves outcomes regardless of the side affected. However, most survivors are left with significant disability