Файл: PLUM AND POSNER S DIAGNOSIS OF STUPOR AND COM-1.pdf

the effect of each stream of cortical processing commanding its own plan of action.

The brain requires a funnel to narrow down the choices from all of the possible modes of action to the single plan of motor behavior that will be pursued. The physical substrate of this process is the basal ganglia. All cortical regions provide input to the striatum (caudate, putamen, nucleus accumbens, and olfactory tubercle). The output from the striatum is predominantly to the globus pallidus, which it inhibits by using the neurotransmitter GABA.100,101 The pallidal output pathways, in turn, also are GABAergic and constitutively inhibit the motor thalamus, so that when the striatal inhibitory input to the pallidum is activated, movement is disinhibited. By constricting all motor responses that are not specifically activated by this system, the basal ganglia ensure a smooth and steady, unitary stream of action. Basal ganglia disorders that permit too much striatal disinhibition of movement (hyperkinetic movement disorders) result in the emergence of disconnected movements that are outside this unitary stream (e.g., tics, chorea, athetosis).

Similarly, the brain is capable of following only one line of thought at a time. The conscious self is prohibited even from seeing two equally likely versions of an optical illusion simultaneously (e.g., the classic case of the ugly woman vs. the beautiful woman illusion) (Figure 1–7). Rather, the self is aware of the two alternative visual interpretations alternately. Similarly, if it is necessary to pursue two different tasks at the same time, they are pursued alternately rather than simultaneously, until they become so automatic that they can be performed with little conscious thought. The striatal control of thought processes is implemented by the outflow from the ventral striatum to the ventral pallidum, which in turn inhibits the

mediodorsal thalamic nucleus, the relay nucleus for the prefrontal cortex.100,101 By dis-

inhibiting prefrontal thought processes, the striatum ensures that a single line of thought and a unitary view of self will be expressed from the multipath network of the cerebral cortex.

An interesting philosophic question is raised by the hyperkinetic movement disorders, in which the tics, chorea, and athetosis are thought to represent ‘‘involuntary movements.’’ But the use of the term ‘‘involuntary’’ again presupposes

Figure 1–7. A classic optical illusion, illustrating the inability of the brain to view the same scene simultaneously in two different ways. The image of the ugly, older woman or the pretty younger woman may be seen alternately, but not at the same time, as the same visual elements are used in two different percepts. (From W.E. Hill, ‘‘My Wife and My Mother-in-Law,’’ 1915, for Puck magazine. Used by permission. All rights reserved.)

a homunculus that is in control and making decisions. Instead, the interrelationship of involuntary movements, which the self feels ‘‘compelled’’ to make, with self-willed movements is complex. Patients with movement disorders often can inhibit the unwanted movements for a while, but feel uncomfortable doing so, and often report pleasurable release when they can carry out the action. Again, the conscious state is best considered as an emergent property of brain function, rather than directing it.

Similarly, hyperkinetic movement disorders may be associated with disinhibition of larger scale behaviors and even thought processes. In this view, thought disorders can be conceived as chorea (derailing) and dystonia (fixed delusions) of thought. Release of prefrontal cortex inhibition may even permit it to drive mental imagery, producing hallucinations. Under such conditions, we have a tendency to believe that somehow the conscious self is a homunculus that is being tricked by hallucinatory sensory experiences or is unable to command thought processes. In fact, it may be more accurate to view the sensory experience and the behavior as manifestations of an altered consciousness due to malfunction of the brain’s machinery for maintaining a unitary flow of thought and action.

Neurologists tend to take the mechanistic perspective that all that we observe is due to ac-

tion of the nervous system behaving according to fundamental principles. Hence, the evaluation of the comatose patient becomes an exercise in applying those principles to the evaluation of a human with brain failure.

Structural Lesions That Cause

Altered Consciousness in Humans

To produce stupor or coma in humans, a disorder must damage or depress the function of

either extensive areas of both cerebral hemispheres or the ascending arousal system, including the paramedian region of the upper brainstem or the diencephalon on both sides of the brain. Figure 1–8 illustrates examples of such lesions that may cause coma. Conversely, unilateral hemispheric lesions, or lesions of the brainstem at the level of the midpons or below, do not cause coma. Figure 1–9 illustrates several such cases that may cause profound sensory and motor deficits but do not impair consciousness.

BILATERAL HEMISPHERIC

DAMAGE

Bilateral and extensive damage to the cerebral cortex occurs most often in the context of hypoxic-ischemic insult. The initial response to loss of cerebral blood flow (CBF) or insufficient oxygenation of the blood includes loss of consciousness. Even if blood flow or oxygenation is restored after 5 or more minutes, there may be widespread cortical injury and neuro-

nal loss even in the absence of frank infarction.102,103 The typical appearance pathologi-

cally is that neurons in layers III and V (which receive the most glutamatergic input from other cortical areas) and in the CA1 region of the hippocampus (which receives extensive glutamatergic input from both the CA3 fields and the entorhinal cortex) demonstrate eosinophilia in the first few days after the injury. Later, the neurons undergo pyknosis and apoptotic cell death (Figure 1–10). The net result is pseudolaminar necrosis, in which the cerebral cortex and the CA1 region both are depopulated of pyramidal cells.

Alternatively, in some patients with less extreme cortical hypoxia, there may be a lucid interval in which the patient appears to recover, followed by a subsequent deterioration. Such a patient is described in the historical vignette on this and the following page. (Throughout this book we will use historical vignettes to describe cases that occurred before the modern era of neurologic diagnosis and treatment, in which

Figure 1–9. Lesions of the brainstem may be very large without causing coma if they do not involve the ascending arousal system bilaterally. (A) Even an extensive infarction at the mesopontine level that does not include the dorsolateral pons on one side and leaves intact the paramedian midbrain can result in preservation of consciousness. (B) Lesions at

a low pontine and medullary level, even if they involve a hemorrhage, do not impair consciousness. (Patient 1–2, p. 33)

the natural history of a disorder unfolded in a way in which it would seldom do today. Fortunately, most such cases included pathologic assessment, which is also all too infrequent in modern cases.)

HISTORICAL VIGNETTE

Patient 1–1

A 59-year-old man was found unconscious in a room filled with natural gas. A companion already had died, apparently the result of an attempted double suicide. On admission the man was unresponsive. His blood pressure was 120/80 mm Hg, pulse 120, and respirations 18 and regular. His rectal temperature was 1028F. His stretch reflexes were hypoactive, and plantar responses were absent. Coarse rhonchi were heard throughout both lung fields.

He was treated with nasal oxygen and began to awaken in 30 hours. On the second hospital day he was alert and oriented. On the fourth day he was afebrile, his chest was clear, and he ambulated. The neurologic examination was normal, and an evaluation by a psychiatrist revealed a clear sensorium with ‘‘no evidence of organic brain damage.’’ He was discharged to his relatives’ care 9 days after the anoxic event.

At home he remained well for 2 days but then became quiet, speaking only when spoken to. The following day he merely shuffled about and res-

ponded in monosyllables. The next day (13 days after the anoxia) he became incontinent and unable to walk, swallow, or chew. He neither spoke to nor recognized his family. He was admitted to a private psychiatric hospital with the diagnosis of depression. Deterioration continued, and 28 days after the initial anoxia he was readmitted to the hospital. His blood pressure was 170/100 mm Hg, pulse 100, respirations 24, and temperature 1018F. There were coarse rales at both lung bases. He perspired profusely and constantly. He did not respond to pain, but would open his eyes momentarily to loud sounds. His extremities were flexed and rigid, his deep tendon reflexes were hyperactive, and his plantar responses extensor. Laboratory studies, including examination of the spinal fluid, were normal. He died 3 days later.

An autopsy examination showed diffuse bronchopneumonia. The brain was grossly normal. There was no cerebral swelling. Coronal sections appeared normal with no evidence of pallidal necrosis. Histologically, neurons in the motor cortex,

hippocampus, cerebellum, and occipital lobes appeared generally well preserved, although a few sections showed minimal cytodegenerative changes and reduction of neurons. There was occasional perivascular lymphocytic infiltration. Pathologic changes were not present in blood vessels, nor was there any interstitial edema. The striking alteration was diffuse demyelination involving all lobes of the cerebral hemispheres and sparing only the arcuate fibers (the immediately subcortical portion of the cerebral white matter). Axons were also reduced in number but were better preserved than was the myelin. Oligodendroglia were preserved in demyelinated areas. Reactive astrocytes were considerably increased. The brainstem and cerebellum were histologically intact. The condition of delayed postanoxic cerebral demyelination observed in this patient is discussed at greater length in Chapter 5.

Another major class of patients with bilateral hemispheric damage causing coma is of those

who suffer from brain trauma.104 These cases usually do not present a diagnostic dilemma, as there is usually history or external evidence of trauma to suggest the cause of the impaired consciousness.

DIENCEPHALIC INJURY

The relay nuclei of the thalamus provide the largest ascending source of input to the cerebral cortex. As a result, it is no exaggeration to

Figure 1–11. A series of drawings illustrating levels through the brainstem at which lesions caused impairment of consciousness. For each case, the extent of the injury at each level was plotted, and the colors indicate the number of cases that involved injury to that area. The overlay illustrates the importance of damage to the dorsolateral pontine tegmentum or the paramedian midbrain in causing coma. (From Parvizi and Damasio,110 with permission.)

say that virtually any deficit due to injury of a discrete cortical area can be mimicked by injury to its thalamic relay nucleus. Hence, thalamic lesions that are sufficiently extensive can produce the same result as bilateral cortical injury. The most common cause of such lesions is the ‘‘tip of the basilar’’ syndrome, in which vascular occlusion of the perforating arteries that arise from the basilar apex or the first segment of the posterior cerebral arteries can produce bilateral thalamic infarction.105

However, careful examination of the MRI scans of such patients, or their brains postmortem, usually shows some damage as well to the paramedian midbrain reticular formation and often in the posterior hypothalamus. Other causes of primarily thalamic damage include thalamic hemorrhage, local infiltrating tumors, and rare cases of diencephalic inflammatory lesions (e.g., Behc¸et’s syndrome).106,107

Another example of severe thalamic injury causing coma was reported by Kinney and colleagues108 in the brain of Karen Anne Quinlan, a famous medicolegal case of a woman who remained in a persistent vegetative state (Chapter 9) for many years after a hypoxic brain injury. Examination of her brain at the time of death disclosed unexpectedly widespread thalamic neuronal loss. However, there was also extensive damage to other brain areas, including the cerebral cortex, so that the thalamic damage alone may not have caused the clinical loss of consciousness. On the other hand, thalamic injury is frequently found in patients with brain injuries who eventually enter a persistent vegetative state (Chapter 9).104

Ischemic lesions of the hypothalamus are rare, because the hypothalamus is literally encircled by the main vessels of the circle of Willis and is fed by local penetrating vessels from all the major arteries. However, the location of the hypothalamus above the pituitary gland results in localized hypothalamic damage in cases of pituitary tumors.109 The hypothalamus also may harbor primary lymphomas of brain, gliomas, or sarcoid granulomas. Patients with hypothalamic lesions often appear to be hypersomnolent rather than comatose. They may yawn, stretch, or sigh, features that are usually lacking in patients with coma due to brainstem lesions.

UPPER BRAINSTEM INJURY

Evidence from clinicopathologic analyses firmly establishes that the midbrain and pontine area critical to consciousness in humans includes the paramedian tegmental zone immediately ventral to the periaqueductal gray matter, from the caudal diencephalon through the rostral pons.110 Numerous cases are on record of small lesions involving this territory bilaterally in which there was profound loss of consciousness (see Figure 1–11). On the other hand, we have not seen loss of consciousness with lesions confined to the medulla or the caudal pons. This principle is

illustrated by the historical vignettes on pages 30 and below.

HISTORICAL VIGNETTES

Patient 1–2

A 62-year-old woman was examined through the courtesy of Dr. Walter Camp. Twenty-five years earlier she had developed weakness and severely impaired position and vibration sense of the right arm and leg. Two years before we saw her, she developed paralysis of the right vocal cord and wasting of the right side of the tongue, followed by insidiously progressing disability with an unsteady gait and more weakness of the right limbs. Four days before coming to the hospital, she became much weaker on the right side, and 2 days later she lost the ability to swallow.

When she entered the hospital she was alert and in full possession of her faculties. She had no difficulty breathing and her blood pressure was 162/ 110 mm Hg. She had upbeat nystagmus on upward gaze and decreased appreciation of pinprick on the left side of the face. The right sides of the pharynx, palate, and tongue were paralyzed. The right arm and leg were weak and atrophic, consistent with disuse. Stretch reflexes below the neck were bilaterally brisk, and the right plantar response was extensor. Position and vibratory sensations were reduced on the right side of the body and the appreciation of pinprick was reduced on the left.

The next day she was still alert and responsive, but she developed difficulty in coughing and speaking and finally she ceased breathing. An endotracheal tube was placed and mechanical ventilation was begun. Later, on that third hospital day, she was still bright and alert and quickly and accurately answered questions by nodding or shaking her head. The opening pressure of cerebrospinal fluid (CSF) at lumbar puncture was 180 mm of water, and the xanthochromic fluid contained 8,500 red blood cells/mm3 and 14 white blood cells/mm3.

She lived for 23 more days. During that time she developed complete somatic motor paralysis below the face. Several hypotensive crises were treated promptly with infusions of pressor agents, but no pressor drugs were needed during the last 2 weeks of life. Intermittently during those final days, she had brief periods of unresponsiveness, but then awakened and signaled quickly and appropriately to questions

demanding a yes or no answer and opened or closed her eyes and moved them laterally when commanded to do so. There was no other voluntary movement. Four days before she died, she developed ocular bobbing when commanded to look laterally, but although she consistently responded to commands by moving her eyes, it was difficult to know whether or not her responses were appropriate. During the ensuing 3 days, evidence of wakefulness decreased. She died of gastrointestinal hemorrhage 26 days after entering the hospital.

The brain at autopsy contained a moderate amount of dark, old blood overlying the right lateral medulla adjacent to the fourth ventricle. A raspberry-appearing arteriovenous malformation, 1.4 cm in greatest diameter, protruded from the right lateral medulla, beginning with its lower border 2.5 cm caudal to the obex. On section, the vascular malformation was seen to originate in the central medulla and to extend rostrally to approximately 2 mm above the obex. From this point, a large hemorrhage extended forward to destroy the central medulla all the way to the pontine junction (Figure 1–9B). Microscopic study demonstrated that, at its most cranial end, the hemorrhage destroyed the caudal part of the right vestibular nuclei and most of the adjacent lower pontine tegmentum on the right. Caudal to this, the hemorrhage widened and destroyed the entire dorsal center of the medulla from approximately the plane of the nucleus of the glossopharyngeal nerve down to just below the plane of the nucleus ambiguus. From this latter point caudally, the hemorrhage was more restricted to the reticular formation of the medulla. The margins of this lesion contained an organizing clot with phagocytosis and reticulum formation indicating a process at least 2 weeks old. The center of the hemorrhage contained a degenerating clot estimated to be at least 72 hours old; at several places along the lateral margin of the lesion were small fresh hemorrhages estimated to have occurred within a few hours of death. It was considered unlikely that the lesion had changed substantially in size or extent of destruction in the few days before death.

Patient 1–3

A 65-year-old woman was admitted to the neurology service for ‘‘coma’’ after an anesthetic procedure. She had rheumatoid arthritis with subluxa-

tion of C1 on C2, and compression of the C2 root causing occipital neuralgia. An anesthesiologist attempted to inject the root with ethanol to eliminate the pain. Almost immediately after the injection, the patient became flaccid and experienced a respiratory arrest. On arrival in the neurology intensive care unit she was hypotensive and apneic. Mechanical ventilation was instituted and blood pressure was supported with pressors.

On examination she had spontaneous eye movements in the vertical direction only and her eyelids fluttered open and closed. There was complete flaccid paralysis of the hypoglossal, vagal, and accessory nerves, as well as all spinal motor function. Twitches of facial and jaw movement persisted. There was no response to pinprick over the face or body. CT scan showed hypodensity of the medulla and lower pons.

The patient responded to commands to open and close her eyes and learned to communicate in this way. She lived another 12 weeks in this setting, without regaining function, and rarely was observed to sleep. No postmortem examination was permitted. However, the injection of ethanol had apparently entered the C2 root sleeve and fixed the lower brainstem up through the facial and abducens nuclei without clouding the state of consciousness of the patient.

Comment: Both of these cases demonstrated the preservation of consciousness in patients with a locked-in state due to destruction of motor pathways below the critical level of the rostral pons. Chapter 2 will explore the ways in which the neurologic examination of a comatose patient can be used to differentiate these different causes of loss of consciousness.

REFERENCES

1.Scheid R, Voltz R, Guthke T, et al. Neuropsychiatric findings in anti-Ma2-positive paraneoplastic limbic encephalitis. Neurology 61, 1159–1160, 2003.

2.Mesulam MM, Waxman SG, Geschwind N, et al. Acute confusional states with right middle cerebral artery infarctions. J Neurol Neurosurg Psychiatry 39, 84–89, 1976.

3.Posner JB, Plum F. The toxic effects of carbon dioxide and acetazolamide in hepatic encephalopathy. J Clin Invest 39, 1246–1258, 1960.

4.Shimojyo S, Scheinberg P, Reinmuth O. Cerebral blood flow and metabolism in the Wernicke-Korsakoff syndrome. J Clin Invest 46, 849–854, 1967.

5.Trzepacz PT, Tarter RE, Shah A, et al. SPECT scan and cognitive findings in subclinical hepatic