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

remain motionless with robust preservation of visual tracking in the form of smooth pursuit movements (or optokinetic responses). Limited preservation of brief visual fixation can be accepted in VS, but robust and consistent visual tracking as seen in akinetic mutism is absent in VS.66

Patient 9–2

A 47-year-old right-handed man was brought to the ICU with progressive somnolence and unresponsiveness. Neurologic examination revealed bilateral third nerve palsy, fluctuating bradycardia with hypertension, and extensor posturing to pain. The initial CT scan (Figure 9–5A) revealed a large mass lesion centered on the mesencephalon with surrounding edema. Intracranial lymphoma was suspected and confirmed by biopsy. The patient

received cranial irradiation, IV steroids, and chemotherapy. A posttreatment MRI (Figure 9–5B) demonstrated resolution of mass effect with high signal abnormalities within the upper mesencephalon and hypothalamus. The patient appeared alert but did not initiate communication. He occasionally displayed sudden periods of agitated behavior. Responses to simple questions were markedly delayed, but correct using yes and no answers. Physical examination was notable for waxy flexibility as well as rigidity, and spontaneous movements were minimal and limited to the left upper extremity.

EEG showed periods of frontal intermittent rhythmic delta activity and mild generalized slowing. An HmPAO single photon emission computed tomography (SPECT) scan revealed diffuse profound frontal bihemispheric hypoperfusion (left greater than right, see Figure 9–5C). The patient’s clinical state did not improve prior to death from a systemic infection.

Autopsy of brain was normal except for the midbrain, hypothalamus, and left paramedian thalamus, which showed infiltration of lymphoma cells and necrosis in the midline of the midbrain extending rostrally into the left thalamus to involve the intralaminar nuclei and surrounding tissue.

Late Recoveries From the

Minimally Conscious State

Word-of-mouth stories and news reports sometimes claim dramatic recovery from prolonged coma or VS. Invariably, these reports generate wide public interest and much confusion concerning the difference between coma and VS,

as well as between diagnosis and prognosis. The Multisociety Task Force64,65 examined 14

cases from the media and found that the majority of these ‘‘late’’ recoveries from VS fell within their guidelines (i.e., less than 3 months following an anoxic injury or 12 months following a traumatic brain injury in an adult). Nonetheless, as noted above, a few rare, welldocumented late recoveries underscore the statistical nature of the guidelines for prognosis of permanent VS. However, most reports of late recovery from ‘‘coma’’ involve very late transition of MCS patients to emergence (see page 373). There are no data to allow guidelines for the expected duration of MCS. Some MCS patients harbor significant residual capacities as demonstrated by wide fluctuation of cognitive function.91 The term minimally conscious state seems most appropriate; alternatives include minimal responsive state and minimal awareness state.92 Minimal responsiveness as assessed at the bedside may belie considerable cognitive capacities without further evaluation of etiologic mechanisms, including normal cognitive function as present in the locked-in state, discussed below.

LOCKED-IN STATE

A related and important issue is late recovery of consciousness in patients with severe motor and sensory impairment leading to the lockedin or partial locked-in state (condition with severe motor disability approximating the traditional definition). The locked-in state is not

a disorder of consciousness, as reviewed in Chapter 1. Nonetheless, because most cases of the locked-in state are due to a pontine injury, it is common for patients to experience an initial coma (see 93 for an example) or to respond inconsistently during the initial period of the injury similar to MCS. In a survey of 44 lockedin patients, the mean time of diagnosis was 2.5 months after onset; in more than one-half of these cases, a family member and not a physician first recognized the condition.94 Furthermore, investigators working with locked-in patients often report early counseling of withdrawal of care either because of an incorrect diagnosis or based on physician attitudes without a careful and vetted informed consent process that includes a review of the available

information on quality of life obtained from surveys of patients in this condition.94,95 While

it is quite reasonable to doubt that most people would want to trade a normal existence for that of a locked-in patient, the important question is whether a locked-in patient would rather live or die. Quality-of-life assessments administered to locked-in patients provide a source of information for patients and families as do written first-person accounts, several of which have become well known.96 Doble and colleagues95 reported on 5-, 10-, and 20-year survival (83%, 83%, and 40%, respectively) and quality of life in 29 patients. Among several notable findings, these investigators found that 12 patients remained living 11 years after the study onset; seven of these patients described ‘‘satisfaction with life,’’ five were noted to exhibit occasional depressive symptoms, but none held a DNR order. Leon-Carrion and associates94 described quality-of-life measures in more detail in their survey of 44 locked-in patients (Table 9–13). The majority of these patients (86%) described a good capacity to maintain attention, nearly half (47%) described their mood as ‘‘good,’’ most (81%) met with friends at least twice a month, and 30% could maintain sexual relations (Table 9–13).

Quality of life was also assessed in 17 chronic (i.e., more than 1 year) locked-in patients who used eye movements or blinking as a principal mode of communication, lived at home,

and had a mean duration of locked-in state of 6 years (range 2 to 16 years).97,98 Self-scored

perception of mental health (evaluating mental well-being and psychologic distress) and personal general health were not significantly

Table 9–13 Functional Measurements

in a Cohort of Locked-in Patients

(N¼ 44)

lower than values from age-matched French control subjects. Importantly, perception of mental health and the presence of physical pain correlated with the frequency of suicidal thoughts (r ¼ –0.67 and 0.56, respectively, p < 0.05), indicating the importance of proper pain management in chronic locked-in patients who are frequently undertreated. At present,

there are three European societies for locked- in-patients with a membership exceeding 300 persons (http://alis-asso.fr/).

MECHANISMS UNDERLYING OUTCOMES OF SEVERE BRAIN INJURY: NEUROIMAGING STUDIES AND CONCEPTUAL FRAMEWORKS

The above discussion details the problems of diagnostic accuracy and prognosis for disorders of consciousness. At present, careful clinical evaluations combined in some instances with structural imaging criteria, or measurements of early cortical sensory responses, remain the foundation for decision making. Available guidelines invariably indicate likelihoods of death or VS as outcomes rather than providing reliable indices of potential for functional recoveries with or without persistent disabilities. In large part this is a consequence of the fact that preserved brainstem function may only herald PVS. Moreover, it is clear that in the aggregate, the clinical neurologic examination and assessments of structural brain integrity provide only limited insight into the neurophysiologic mechanisms of coma, VS, or MCS. This is because the functional impairment of distributed neuronal populations of the cerebral cortex, basal ganglia, and thalamus underlying the conditions often cannot be adequately assessed by these methods. Neuroimaging techniques that can directly assess functional changes within these cerebral networks hold significant promise to ultimately improve diagnostic accuracy and understanding of the pathophysiology of the severely injured brain (see 99 for review).

Expanded use of neuroimaging techniques for evaluating functional outcomes of patients recovering from coma will likely have the greatest impact on the category of severe disability. This broad category includes within its limits patients who, while not permanently unconscious, as in the chronic VS, may nonetheless never regain a capacity to communicate, as well as other patients near the functional borderline of independence in activities of daily living. More than 20 years ago, the third edition of Stupor and Coma commented that the overly broad definition of severe disability needed sig-

nificant refinement. As discussed above, recent efforts to define MCS are a step in this direction. The significance of identifying the physiologic mechanisms underlying different functional outcomes within the category of severe disability is that this knowledge will lead to a better understanding of the necessary and sufficient neurologic substrates to recover consciousness and varying levels of cognitive capacity. Just as the concept of brain death clarified the concept of death, MCS and other future subdivisions of the category of severe disability will force us to consider the concept of consciousness more precisely.

FUNCTIONAL IMAGING OF THE PERSISTENT VEGETATIVE STATE

Levy and associates100 provided the first experimental evidence supporting the clinical hypothesis that patients in VS were unconscious. Using FDG-PET, seven patients in PVS were compared to three patients in the locked-in state and 18 normal subjects. In PVS patients, cerebral metabolic rates were globally reduced by 50% or more. Regional cerebral blood flow measurements showed a similar but more variable pattern of global reduction. Subsequent studies have confirmed these findings, with an average of less than 50% of normal metabolic rates in most VS patients studied (reduced further to 30% to 40% in cases of hypoxicischemic etiology).101–105 Comparable reduc-

tions are identified during generalized anesthesia106,107 and in stage IV sleep in normal

individuals.108 The small number of patients in the locked-in state (three) in the Levy study had a low average metabolic rate, but recent quantitative FDG-PET studies have demonstrated essentially normal resting metabolic rates in the cerebrum, even acutely.99 Cerebellar metabolic rates were low, consistent with the lack of motor outflow in the locked-in state.98

More sensitive imaging techniques have recently been applied to the evaluation of PVS

patients. They reveal a marked loss of distributed network processing in VS.99,104,109 Ele-

mentary auditory and somatosensory stimuli fail to produce brain activation outside of primary sensory cortices (Figure 9–6). The data suggest multiple functional disconnections along the auditory or somatosensory cortical pathways and support the inference that the

residual cortical activity seen in PVS patients does not reflect awareness. The findings are consistent with evidence of early sensory processing in PVS patients as measured by evoked potential studies, but loss of later components39; they suggest that VS/PVS correlates with failure of sensory information to propagate beyond the earliest stages of cortical processing. Preliminary studies discussed below indicate that MCS patients show wider activation of cortical networks, findings that may help ultimately distinguish the conditions among patients with severe sensory and motor impairments limiting behavioral assessments (e.g., spastic contractions and blindness).

Atypical Behavioral Features in

the Persistent Vegetative State

Stereotyped behavior, typically limbic displays of crying, smiling, or other emotional patterns that are not related to environmental stimuli, occur in some VS patients. Occasionally, other fragments of behavior that may appear semipurposeful, or inconsistently related to environmental stimuli, arise in VS/PVS patients. Neuroimaging studies, including FDG-PET, magnetoencephalography (MEG), and functional MRI (fMRI), have identified residual cerebral circuits underlying such isolated behavioral fragments.105,110,117 One remarkable patient studied had remained in the PVS for 20 years but infrequently expressed single words (typically epithets) not related to environmental stimulation (Figure 9–7C). Two other patients in this group revealed similar isolated metabolic activity that could be correlated with unusual behavioral patterns.105 These data provide novel evidence for the modular organization of the brain and suggest that preservation of residual cerebral activity following severe brain injuries is not random. Regional preservation of cerebral metabolic activity likely reflects both preservation of anatomic connectivity and endogenous neuronal firing patterns of remnant but incomplete networks.

Further study of this patient showed that islands of higher resting brain metabolism included Heschl’s gyrus (Figure 9–8), Broca’s area, Wernicke’s area, and the left anterior basal ganglia (caudate nucleus, possibly putamen). Despite limited amounts of remaining left thalamus identified by MRI that expressed