Chapter 73 Impairment of Consciousness and Coma
The study of consciousness and causes of alteration of consciousness represents one of the oldest areas of neuroscience. The Edwin Smith Surgical Papyrus (circa 1700 b.c.) is the oldest historical document to use the term “brain” and contains the first description of the meninges, cerebrospinal fluid, and the convolutions of the brain surface. It contains vivid analytical descriptions of injury to the nervous system and deficits in neurological function caused by the injuries [Sanchez and Burridge, 2007]. However, in Ancient Egyptian medicine, the heart and the mind were synonymous [Mohamed, 2008]. Therefore, from the earliest recorded neuroscience document until current times, the location of the mind and the nature of consciousness have been the subject of study and debate based on the state of philosophy and science of the time.
The French mathematician, physicist, and philosopher, René Descartes (1596–1650), associated the mind with consciousness and self-awareness, and distinguished the mind from the brain. He postulated that consciousness and self-awareness were separate from the physical brain and sought to identify the “seat” of their convergence. He postulated that the mind and the brain were integrated at the pineal gland, a theory based more on basic knowledge of the gross anatomy of the brain than an understanding of neurophysiology. His writings perpetuated the concept of dualism: that the mind and mental processes are not physical and are separate from the brain. His is one of the earliest and most articulate descriptions of the mind–body problem as it exists today. Stated differently, how can consciousness and awareness be resolved within the structure and function of the brain [Gierer, 2008]? Dualism continues to have scientific, philosophical, and practical implications in medical practice today [Demertzi et al., 2009].
Multiple philosophical, metaphysical, and psychologic theories of consciousness have been proposed [Zeman, 2001]. William James, in 1890, wrote that, when the word consciousness is limited to the personal self of the individual, the cortex is the sole organ of consciousness in man [James, 1890]. He also emphasized attention (internal and external) and the ability to direct the focus of attention as fundamental to consciousness. None the less, reconciliation of the concept of mind (consciousness and awareness) and the structure and function of the brain has long been elusive [Sperry, 1952]. Evidence that there is a neural correlate of consciousness, however, is clearly established [Brain, 1950; Crick and Koch, 1998; Crick and Koch, 2003; Damasio, 2003; Evans, 2003; Negrao and Viljoen, 2009].
The modern understanding of wakefulness and consciousness begins with the work of von Economo, who first characterized encephalitis lethargica in 1917 and subsequently studied the clinical course and neuropathology of its victims for the remainder of his career. He did not think that encephalitis lethargica was caused by influenza, which was pandemic at the time, but acknowledged that influenza might predispose to its development. His intuition about etiology of encephalitis lethargica has been subsequently borne out [McCall et al., 2008]. In 1929, von Economo addressed the College of Physicians and Surgeons at Columbia University in New York and proposed discreet subcortical centers that control wakefulness and sleep, based on his studies [von Economo, 1930]. He showed that wakefulness arose from the upper brainstem and that hypothalamic lesions predispose to sleep. The extraordinary accuracy of his observations has been recently reviewed and confirmed [Saper et al., 2005].
Also in 1929, Hans Berger published his electroencephalographic (EEG) study in humans [Berger, 1929; Gloor, 1969]. The convergence of neuropathology and neurophysiology with the newer ability to stimulate discreet areas of the brain electrically led to Moruzzi and Magoun’s discovery that stimulation of the brainstem reticular formation produced EEG desynchronization in the cerebral cortex and behavioral alerting [Moruzzi and Magoun, 1949]. Subsequently, many studies have shown the importance of the ascending arousal system and other closely allied systems in the control of wakefulness and sleep and in the pathophysiology of coma [Saper et al., 2001, 2005; Parvizi and Damasio, 2003; Hobson, 2005; Schwartz and Roth, 2008; Perlis et al., 2009].
The ascending arousal system (reticular activating system, ascending reticular activating system) has been the subject of research concerning its role in attention and consciousness for more than 50 years. However, advances in understanding of neuroanatomy restimulated interest in the role of the thalamus and neocortex in attention and consciousness [Brodal, 1964]. Mesulam wrote, in 1985: “Today, even though vestiges of predominately subcortical theory of attention can occasionally still be detected, there is an emerging sense that neocortex, thalamus and brainstem are inextricably linked in the modulation of attention and more complex aspects of this function are executed predominately by neocortical mechanism” [Mesulam, 1985].
Many studies have confirmed that the midbrain reticular formation provides much of its cortical activation through the thalamus, and that reciprocal connections between the thalamus and cortex play a pivotal role in attention and consciousness [Crick, 1984; Newman, 1995; Steriade, 1997; Van der Werf, 2002; Evans, 2003; Llinás and Steriade, 2006; Pinault, 2004].
The historical understanding of the thalamus as the major synapse point for environmental sensory information on the way to the cortex is unchallenged. However, the larger role of the thalamus is that of the major source of extracortical activation for the cortex. The thalamic reticular nuclei provide connections with the cortex that allow cortical regulation of noncortical output. Thus the action of the midbrain reticular formation is primarily excitatory, causing increased but relatively nonspecific arousal, while the thalamic reticular nuclei have a more inhibiting role, allowing focus [Crick, 1984; Newman, 1995].
The complex interplay among the brainstem reticular core, the thalamus, and the cortex form the basis of attention, awareness, and consciousness [Boveroux et al., 2008]. The existence of cortical activation from subcortical centers and cortical regulation of input across hundreds of thousands of reciprocal networks (in parallel activated through the thalamus and within the cortex) are the basis for cognitive function, including attention and arousal, awareness of self and environment, and consciousness. Several convergent lines of evidence suggest that this complicated interaction is, to some degree, conducted by rhythmic fast activity (40 Hz) originating in the thalamus [Min, 2010; Llinas, et al., 1998]. How large numbers of neuron networks are orchestrated to provide sufficient degree of information integration necessary for consciousness is a field of on-going study [Lundervold, 2010].
Functional neuroimaging of brain activity using positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) has substantially contributed to our understanding of the brain’s functional networks [Monti et al., 2009; Owen, 2008]. PET is a form of in vivo autoradiography that utilizes water radiolabeled 15oxygen as the positron emitter to measure brain blood flow. Once produced, positrons almost immediately combine with electrons and, by destroying one another, they emit two high-energy gamma rays that travel in opposite directions. Radiation detectors, arranged around the head, detect these paired gamma rays and, utilizing computers, localize the source with great accuracy.
fMRI measures a complex function of blood flow related to the observation that the amount of oxygen in a certain area of the brain influences its magnetic properties (see Chapter 11). This change, the blood oxygen level dependent or BOLD effect, can be detected by fMRI [Raichle, 2009]. fMRI measures blood oxygenation, which indirectly reflects neuronal activity. Multiple studies have shown that the fMRI signal is primarily related to synaptic activity from the input and processing of information locally, and not related to signal transmission to other areas of the brain. Thus, focal changes in the fMRI signal reflect focal neuronal processing and are not a byproduct of a broader or more diffuse process [Raichle and Mitun, 2006; Khader et al., 2008; He and Raichle, 2009; Logothetis, 2002; Lippert, 2010].
The brain is not, however, simply driven reflexively, quietly waiting to react to stimuli. Rather, the brain has substantial on-going metabolic needs that account for approximately 20 percent of all energy consumed by the body, while representing only about 2 percent of total body weight in the average adult. Moreover, 60–80 percent of overall brain energy use is related to neuronal signaling [Raichle and Mintun, 2006]. None the less, changes in brain activity produce relatively small fluctuations of energy utilization, often less than 5 percent, compared to the high energy consumption of the resting state [Raichle, 2010].
PET and fMRI studies have shown the existence of areas of high resting baseline activity wherein metabolic activity actually decreases during certain goal-directed tasks. This is called the resting state or default mode of brain function [Raichle et al., 2001]. This leads to the conclusion that most brain function is intrinsic and involves integration of information for interpreting, responding to, and predicting environmental demands [Raichle, 2010].
Coordination of the default mode network and other brain functional systems is a subject of intense study. Spontaneous fluctuations in the BOLD signals of fMRI demonstrate consistent temporal correlations within brain networks that are associated with different functions [He et al., 2008]. In different situations and conditions, local field potentials across a wide frequency range (5–90 Hz) have been correlated with the spontaneous fluctuation of the fMRI signal. Recently, however, lower-frequency local field potentials (slow cortical potentials) have been studied and correlated with fMRI signal fluctuations. Because their temporal scale (<4 Hz) overlaps with the fMRI signal (<0.5 Hz) they seem to be a likely source for broadly distributed information integration in the brain [He and Raichle, 2009].
To be part of consciousness, information has to be integrated because conscious experience is a unitary and undivided whole [Tononi, 2008]. Slow cortical potentials, already studied in cognitive neuroscience as event-related potentials, may be the ideal neural mechanism for information integration across large areas of cortex. Slow cortical potentials take their origin from the more superficial cortical layers where far-reaching intracortical and corticocortical connections preferentially terminate, allowing influence of the slow cortical potentials across large areas of cortex. The relatively slow time scale of the slow cortical potentials would facilitate synchronization across wide areas independent of axonal conduction delay [He and Raichle, 2009]. Slow cortical potentials may subserve information integration across large cortical areas contributing to conscious awareness.
Because neurobiological processes that are realized in brain structures are responsible for consciousness [Baars et al., 2003; Neylan, 1995; Searle, 2000], neurologic disorders of the cortex and its brainstem activators result in impairment of consciousness and coma. Understanding neurologic disease depends on the clinical assessment and interpretation of consciousness, the content of consciousness, and alterations of consciousness. The evaluation of all disorders of higher cortical function fundamentally begins with consciousness assessment, and all signs and symptoms of neurologic impairment must be interpreted based on the state of consciousness and awareness [Giacino, 1997].
Consciousness refers to the state of awareness of self and environment [James, 1890; Plum and Posner, 1982]. Evaluation of consciousness in the pediatric patient must take into account age and the appropriate developmental level. Conscious individuals may have abnormal content of consciousness, such as hallucinations, delirium, or dementia. Therefore, normal consciousness must be distinguished from abnormal consciousness as objectively as possible. Unconsciousness is unawareness of self and environment, and may be physiologic (sleep) or pathologic (coma or the vegetative state). The fundamental difference between sleep and coma is that, with appropriate stimulus intensity and duration, a sleeping person can be aroused to a normal state of consciousness, whereas a comatose patient can not. Thus, the diagnosis of coma and other impairments of consciousness involves state and reactivity. The physiologic differences between sleep and coma are actually more profound. Sleep is a biologically active state with identifiable behavioral and EEG stages, whereas coma is a state of reduced neuronal activity [Moruzzi, 1972].
Along the continuum from normal consciousness to coma or unarousable unconsciousness, many terms are used to describe mental state and reactivity. When there is doubt about the appropriate use of one of these terms, it is far better to describe the state and reactivity of the individual rather than label it. Definitions of consciousness, impairment of consciousness, coma, and related states have been proposed throughout the history of medicine and have been reviewed, refined, and stated systematically during the past several decades [Ashwal, 1996; Ashwal and Cranford, 2002; Bates, 1993; Bozza Marrubini, 1984; Medical Research Council Brain Injuries Committee, 1941; Michelson and Ashwal, 2004; Plum and Posner, 1982; Owen, 2008].
Definitions
Consciousness is the spontaneously occurring state of awareness of self and environment. Consciousness has two dimensions – wakefulness and awareness [Multi-Society Task Force on the Persistent Vegetative State, 1994]. Normal consciousness requires arousal, an independent, autonomic-vegetative brain function subserved by ascending stimuli – emanating from pontine tegmentum, posterior hypothalamus, and thalamus – that activate mechanisms inducing wakefulness. Awareness is subserved by cerebral cortical neurons and their reciprocal projections to and from the major subcortical nuclei. Awareness requires wakefulness, but wakefulness can be present without awareness.
Impairment of Consciousness with Reduced Mental State
Coma is a state of deep, unarousable, sustained pathologic unconsciousness with the eyes closed that results from dysfunction of the ascending reticular-activating system in the brainstem or in both cerebral hemispheres [Multi-Society Task Force on Persistent Vegetative State, 1994; Young, 2009]. Coma usually requires the period of unconsciousness to persist for at least 1 hour to distinguish coma from syncope, concussion, or other states of transient unconsciousness. The term unconsciousness implies global or total unawareness, and applies equally to patients in coma or vegetative states. Patients in coma are unconscious because they lack both wakefulness and awareness. In contrast, patients in vegetative states are unconscious because, although they have retained wakefulness, they lack awareness. The depth of coma may be further specified by assessment of brainstem reflexes, breathing pattern, change of pulse or respiratory rate to stimulation, or stimulus-induced nonspecific movement.
Vegetative State, Minimally Conscious State, and Related Conditions
Table 73-1 lists several of the major neurologic conditions that the clinician must be capable of differentiating from coma. This table was modified from the report issued by the Multi-Society Task Force on the Persistent Vegetative State [1994] and more recent work that is defining the concept of the minimally conscious state [American Congress of Rehabilitation Medicine, 1995; Giacino et al., 1997]. The principal neurologic conditions in which there may be overlap of clinical findings on examination that should be differentiated from coma are discussed in the next sections.
Vegetative State
The vegetative state can be described as a condition of complete unawareness of the self and the environment, accompanied by sleep–wake cycles with either complete or partial preservation of hypothalamic and brainstem autonomic (vegetative) functions [Zeman, 1997; Monti et al., 2010]. Criteria to diagnose the vegetative state have been recommended for adults and children by the Multi-Society Task Force on Persistent Vegetative State and are listed in Box 73-1. Children in a vegetative state lack evidence of self-awareness or recognition of external stimuli. Rather than being in a state of “eyes-closed” coma, they remain unconscious but have irregular periods of wakefulness alternating with periods of sleeping. Vegetative patients demonstrate a variety of sounds, emotional expressions, and body movements, and they may smile or shed tears. They have inconsistent head- and eye-turning movements to sounds and inconsistent, nonpurposeful trunk and limb movements. The most objective sign is lack of sustained visual fixation or visual tracking.
It is estimated that there are about 4000–10,000 children in a vegetative state in the United States [Ashwal, 1996] and about 100,000 worldwide [Ashwal, 2005]. The etiology of the vegetative state in children can be classified into the following three broad groups of disorders:
The percentage of children having these etiologies of the vegetative state is as follows: acute traumatic and nontraumatic injuries (30 percent), perinatal insults (17.7 percent), chromosomal disorders or congenital malformations (13.0 percent), infections (10.3 percent), and unknown causes (28 percent) [Ashwal et al., 1994].
The most common causes of acute brain injury leading to the vegetative state in children are head trauma and hypoxic-ischemic encephalopathies [Adams et al., 2000]. Severe traumatic brain injury in children is often due to nonaccidental trauma but also occurs after a motor vehicle crash, particularly when infant restraints have not been used or when a pedestrian versus motor vehicle crash occurs. Hypoxic-ischemic injuries after cardiorespiratory arrest occur at birth, after episodes of near-miss sudden infant death syndrome or near-drowning, and after other acute life-threatening episodes, including shock due to systemic trauma or infection.
The clinical course of evolution to a vegetative state after an acute injury usually begins with eyes-closed coma for several days to weeks, followed by the appearance of sleep–wake cycles. Gillies and Seshia retrospectively determined that the transition from eyes-closed coma to the vegetative state averaged 8.6 ± 1.7 days in 17 children aged 1 month to 6 years [Gillies and Seshia, 1980]. Other responses, such as decorticate and decerebrate posturing (1.7 days), roving eye movements (1.8 days), and eye blinking (3.3 days), appeared earlier than sleep–wake cycles [Gillies and Seshia, 1980].
Diagnosis of the vegetative state is made clinically [AAN Quality Standards Subcommittee, 1995]. There are no confirmatory laboratory tests, in contrast to establishing the diagnosis of brain death. However, the absence of cortical (scalp-recorded) potentials of the somatosensory-evoked response to median nerve stimulation has been associated with the vegetative state in some studies [Beca et al., 1995; Frank et al., 1985]. Neuroimaging studies usually demonstrate diffuse or multifocal cerebral disease involving the gray and white matter or major CNS malformations [Ashwal, 1996]. In patients with traumatic and nontraumatic brain injury, serial imaging studies may demonstrate progressive atrophy. PET studies may demonstrate significant global reductions of cerebral glucose metabolism in infants [Larsen et al., 1993]. Proton magnetic resonance spectroscopy measurement of increased cerebral lactate in newborns, infants, and children with nervous system disease is associated with severe disability, survival in the vegetative state, or death [Ashwal et al., 1997].
Issues have been raised for several of the disorders of consciousness with regard to the diagnostic accuracy of the clinical assessment versus the use of advanced imaging or neurophysiological methods to assess such severely affected patients [McKinney and Giacino, 2008; Owen and Coleman, 2007; Giacino et al., 2009; Gawryluk et al., 2010]. In addition, several recent studies in adult patients thought to be in a vegetative state have raised considerable interest, as investigations with fMRI have shown evidence of awareness and presumably some level of higher cortical function [Owen and Coleman, 2008]. In addition, deep brain stimulation in minimally conscious-state patients [Schiff et al., 2009] has resulted in clinical improvement and has raised additional issues as to the role of functional imaging in evaluating vegetative state patients and also in potentially new approaches to treatment.
Recovery from the vegetative state in children depends on the etiology and is reviewed in more detail in the Multi-Society Task Force on the Persistent Vegetative State report [1994]. In children with severe closed-head injuries who were vegetative 1 month after injury, follow-up at 1 year found that 29 percent remained in a vegetative state, 9 percent had died, and 62 percent recovered consciousness. Recoveries after 12 months were not reported, although one study found that 2 of 40 children with traumatic brain injury began to recover after 1 year in a vegetative state [Kriel et al., 1993]. In contrast, children in a nontraumatic vegetative state have a much poorer potential for recovery than those in a traumatic vegetative state. At 1 year, most children remained in a vegetative state (65 percent) or died (22 percent); only 13 percent demonstrated recovery, usually to a severe disability. It was based on these data and consensus opinion that the Task Force came to the conclusions that, in children (as well as in adults), the vegetative state could be judged permanent 12 months after traumatic brain injury and 3 months after nontraumatic injury. The chance for recovery after these time periods was exceedingly low, and the patient almost always progressed to a severe disability. This perspective has received general acceptance, although concerns have been raised about the certainty of diagnosis and whether there is a greater potential for patients to recover than previously realized [Giacino et al., 1997; Jennett, 1997].
It is important to identify children who are in a vegetative state correctly because of the implications for continued care, family expectations, and the need for rehabilitation. Specialty programs for coma recovery are usually directed at patients who are severely disabled or are in a minimally conscious or vegetative state. Management issues are confounded by a lack of understanding of expected recovery patterns, terminology inconsistency, lack of or inappropriate use of outcome measures, and inadequate knowledge concerning medical management based on scientifically sound practices for this special patient population [Zasler et al., 1991]. Children in a vegetative state have been reported to have a considerably shorter than normal life expectancy [Ashwal et al., 1994]. Children in a vegetative state who are younger than 1 year of age had a median survival of 2.6 years, whereas the median survival of children aged 2–6 years was 5.2 years. In children 7–18 years of age, the reported median survival was 7 years.
Minimally Conscious State
The minimally conscious state has been defined as a condition of severely altered consciousness in which the person demonstrates minimal but definite behavioral evidence of self- or environmental awareness [Giacino et al., 1997]. Recovery from coma depends on the cause of coma, extent of brain injury, and ability of injured areas to recover or of uninjured areas to assume the functions of injured areas. Not all comatose patients recover rapidly or completely. A variety of terms have been used to define these states, and there is often confusion between recovery states and prognostic categories. The term minimally conscious state has been proposed to describe those patients who were in coma or a vegetative state and who begin to demonstrate minimal signs of awareness. The term minimally conscious state may also be used to describe patients with degenerative disorders who are no longer functionally interactive but are not in the vegetative state. Patients in a minimally conscious state are able to do the following:
Criteria to diagnose the minimally conscious state are shown in Box 73-2 [Giacino et al., 2002].
Box 73-2 Diagnostic Criteria for the Minimally Conscious State
Purposeful behavior, including movements or effective behaviors that occur in contingent relation to relevant environmental stimuli and are not due to reflexive activity. Some behavioral examples of qualifying purposeful behaviors include the following:
(Adapted from Giacino JT et al. The minimally conscious state. Definition and diagnostic criteria. Neurology 2002;58:349.)
Children in the minimally conscious state are usually victims of acquired brain injury (traumatic and nontraumatic), neurodegenerative and neurometabolic disorders, or congenital or developmental disorders [Ashwal, 2003]. Based on limited literature on this disorder in adults [Giacino and Kalmar, 2005], it is possible that such children, depending on the etiology of the insult, may have a better potential for neurologic recovery and a longer life expectancy than children in the vegetative state. In a study of 5075 children aged 3–15 years in a vegetative state or minimally conscious state, 564 were in the vegetative state, 705 in the minimally conscious state and immobile, and 3806 in the minimally conscious state and mobile. The 8-year survival rates were 63 percent for vegetative state, 65 percent for immobile minimally conscious state, and 81 percent for mobile minimally conscious state. Thus, presence of consciousness was not a critical variable in determining life expectancy in this study, and mobility was more important than consciousness in predicting survival [Strauss et al., 2000].
Locked-In Syndrome
The locked-in syndrome refers to a condition in which patients retain consciousness and cognition but are unable to move or communicate because of severe paralysis. This condition is a result of diseases involving the descending corticospinal and corticobulbar pathways at or below the pons or of severe involvement of the peripheral nervous system. By definition, patients in coma or a vegetative state are unconscious and differ from patients in a locked-in syndrome, who retain awareness, although this difference may be difficult to determine. The locked-in syndrome is quite rare in children [Habre et al., 1996; Scott et al., 1997]. Some patients with this condition can establish limited communication using eye movements.
Akinetic Mutism
Akinetic mutism is a rare condition consisting of pathologically slowed or nearly absent bodily movement, accompanied by a similar loss of speech [Plum and Posner, 1982]. The original description of this condition in 1941 was in an adolescent with symptoms of intermittent depressed states of consciousness secondary to a craniopharyngioma [Cairns et al., 1941]. Since then, akinetic mutism has been seen with bacterial and viral CNS infections, other tumors of the nervous system, and hydrocephalus, and occasionally as a postoperative phenomenon, especially with posterior fossa and cerebellar tumors [Lin et al., 1997; Melton et al., 1991]. Wakefulness and self-awareness are usually preserved in most patients, but the level of mental function is reduced. The condition characteristically accompanies gradually developing or subacute, bilateral damage to the paramedian mesencephalon, basal diencephalon, or inferior frontal lobes. The long-term outlook for children with akinetic mutism is unknown because few patients with this disorder have been reported. It is most likely related to the etiology and severity of the associated disease.
Brain Death
Brain death describes the permanent absence of all brain functions, including those of the brainstem [Task Force for the Determination of Brain Death in Children, 1987; Nakagawa et al., 2011]. Brain-dead patients are irreversibly comatose and apneic, and they have absent brainstem reflexes, including the loss of all cranial nerve functions. The appearance of brain death can be imitated by deep anesthesia, sedative overdose, or severe hypothermia. Patients who are brain-dead differ from patients who are in a coma, in that comatose patients usually have preserved brainstem functions and some degree of respiratory drive. Guidelines for the diagnosis of brain death in infants and children were originally published in 1987 and were recently revised in 2011; they are reviewed in greater detail in Chapter 79. A useful algorithm for distinguishing among brain death, vegetative state, locked-in syndrome, and higher-order consciousness is presented in Figure 73-1 [Ashwal and Cranford, 2002].
Consciousness Rating Scales
Several behavior rating scales have been developed because of the difficulty with imprecise use of terminology and because of the need for a truly reproducible assessment of consciousness with good inter-rater reliability [Kirkham et al., 2008]. Although typically referred to as coma scales, these scales have value in assessing consciousness alterations other than coma.
The best known and most widely used is the Glasgow Coma Scale, which yields a score of 3–15, based on best response to stimuli in the following three categories: eye opening, verbal response, and motor response (Table 73-2) [Teasdale and Jennett, 1974]. In its original form, the Glasgow Coma Scale was not developmentally suitable for assessment of newborns, infants, and younger children, and a variety of alternate scales have been proposed [Duncan et al., 1981; Hahn et al., 1988; Raimondi and Hirschauer, 1984; Reilly et al., 1988; Rubenstein, 1994; Simpson et al., 1991; Simpson and Reilly, 1982]. An ideal pediatric modification of the Glasgow Coma Scale would meet the following criteria:
The Pediatric Coma Scale [Reilly et al., 1988; Simpson et al., 1991; Simpson and Reilly, 1982] makes minor changes in the verbal scale of the Glasgow Coma Scale and redefines the “best” score based on developmental and age-appropriate norms (Table 73-3). It has the advantage of prospective evaluation and inter-rater reliability and the disadvantage of different maximum scores based on age, thus making outcome prediction and comparison difficult to study using the Glasgow Coma Scale.
| Response | Score |
|---|---|
| Eye Opening | |
| Spontaneous | 4 |
| To speech | 3 |
| To pain | 2 |
| None | 1 |
| Best Verbal Response | |
| Oriented | 5 |
| Words | 4 |
| Vocal sounds | 3 |
| Cries | 2 |
| None | 1 |
| Best Motor Response | |
| Obeys commands | 5 |
| Localizes pain | 4 |
| Flexion to pain | 3 |
| Extension to pain | 2 |
| None | 1 |
| Normal Aggregate Score | |
| Birth to 6 months | 9 |
| Less than 6–12 months | 11 |
| Less than 1–2 years | 12 |
| Less than 2–5 years | 13 |
| More than 5 years | 14 |
The Children’s Coma Scale redefines intracategory criteria and the “maximum” score, and changes the eye-opening category to “ocular response” [Raimondi and Hirschauer, 1984]. The Children’s Coma Scale meritoriously included pupillary reflexes, extraocular movements, and apnea in its categories, and is therefore substantially different from the Glasgow Coma Scale and requires different training and forms to be used (Table 73-4).
Table 73-4 Children’s Coma Scale
| Sign | Score |
|---|---|
| Ocular Response | |
| Pursuit | 4 |
| Extraocular movement intact, pupils react appropriately | 3 |
| Fixed pupils or extraocular movement impaired | 2 |
| Fixed pupils and extraocular movement paralyzed | 1 |
| Verbal Response | |
| Cries | 3 |
| Spontaneous respiration | 2 |
| Apneic | 1 |
| Motor Response | |
| Flexes and extends | 4 |
| Withdraws from painful stimuli | 3 |
| Hypertonic | 2 |
| Flaccid | 1 |
| Best Total Score | 11 |
The Glasgow Coma Scale – Modified for Children maintains the same categories as the Glasgow Coma Scale and the same maximum and minimum scores, while allowing developmental and age-appropriate scoring [Hahn et al., 1988; Rubenstein, 1994]. It does, however, equate spontaneous movement with following commands to achieve 6 points in the best motor response category to allow inclusion of infants and young children (see Table 73-2). None the less, the Glasgow Coma Scale – Modified for Children allows the best direct comparison with the Glasgow Coma Scale for scoring of consciousness impairment and assessment of outcome.
The Glasgow Coma Scale and its modifications for children are useful for evaluating and quantifying the continuum from consciousness to coma more objectively, and for allowing serial reassessment that meaningfully reflects the state of the patient at a particular moment in time with good inter-rater reliability. The scores also allow a relatively large amount of information to be conveyed quickly and concisely, with minimal need to remember cumbersome definitions or to write long descriptions. Their ease of use makes them even more valuable in the critical care and emergency department settings. The Glasgow Coma Scale and its pediatric modifications do not take into account the important brainstem reflexes, such as pupillary reactivity or oculocephalic, oculovestibular, or corneal reflexes. These should be a routine part of the assessment of the patient with impaired consciousness [Narayan et al., 1981].
The sum of the Glasgow Coma Scale may convey less useful information than the ocular, motor, and verbal responses individually [Teasdale et al., 1983]. In a study of 270 children aged 2 months to 12 years with acute nontraumatic coma, multivariate analysis suggests that the ocular and motor response scores, and not the total score, were predictors of short-term survival. Additionally, the absence of one or more brainstem reflexes (oculocephalic, oculovestibular, or pupillary reactivity) predicted short-term adverse outcomes [Nayana Prabha et al., 2003].
In children, the Glasgow Coma Scale motor subscore has equal predictive value with the full scale score and is more easily and reliably obtained [Van de Voorde, 2008]. In a retrospective study of 130 children 3 months to 16 years old admitted for traumatic brain injury, the motor response of the Glasgow Coma Scale predicted outcome as well as the full Glasgow Coma Scale, had a more linear relationship to mortality, and was easier to collect accurately [Fortune and Shann, 2010]. Multiple studies have shown the value of the motor subscore of the Glasgow Coma Scale and the additive value of brainstem reflex data in predicting outcome in children with traumatic and nontraumatic coma.
The Full Outline of Unresponsiveness (FOUR) Score Coma Scale (Table 73-5) was created to address these areas of weakness in the use of the Glasgow Coma Scale and its derivatives [Wijdicks et al., 2005]. The FOUR Score is divided into four categories (eye response, motor response, brainstem reflexes, and respiration), each with a maximum score of four (range 0–4). By providing a measure of consciousness that does not require patient vocalization, assessment of motor function, evaluation of brainstem reflexes at the mesencephalic, pontine, and medullary levels, and observation of respiration, the FOUR Score improves upon the Glasgow Coma Scale and incorporates other important features of the neurologic exam. Addition of this neurologic scoring system allows recognition of brain death, the vegetative state, the locked-in syndrome, and progressive changes seen in herniation syndromes [Wijdicks, 2006].
Table 73-5 “FOUR” Score Coma Scale
| Category and Response | Score |
|---|---|
| Eye Response | |
| Eyelids open or opened, tracking or blinking to command | 4 |
| Eyelids open but not tracking | 3 |
| Eyelids closed, opens to loud voice, not tracking | 2 |
| Eyelids closed, opens to pain, not tracking | 1 |
| Eyelids remain closed with pain | 0 |
| Motor Response | |
| Thumbs up, fist, or peace sign to command | 4 |
| Localizing to pain | 3 |
| Flexion response to pain | 2 |
| Extensor posturing | 1 |
| No response to pain or generalized myoclonus status epilepticus | 0 |
| Brainstem reflexes | |
| Pupil and corneal reflexes present | 4 |
| One pupil wide and fixed | 3 |
| Pupil or corneal reflexes absent | 2 |
| Pupil and corneal reflexes absent | 1 |
| Absent pupil, corneal, and cough reflex | 0 |
| Respiration | |
| Not intubated, regular breathing pattern | 4 |
| Not intubated, Cheyne–Stokes breathing pattern | 3 |
| Not intubated, irregular breathing pattern | 2 |
| Breathes above ventilator rate | 1 |
| Breathes at ventilator rate or apnea | 0 |
(From Wijdicks E et al. Validation of a new Coma Scale: The FOUR Score. Annals of Neurology 2005;58[4]:584–593.)
In adults, the FOUR Score has excellent inter-rater reliability when performed by nurses or physicians, and is suitable for use in neurosurgical and medical critical care and in the emergency department [Wolf, 2007; Akavipat, 2009; Stead, 2009; Fischer et al., 2010]. In a study of 60 children (ages 2–18 years) admitted for traumatic and nontraumatic injuries, the inter-rater reliability of the Glasgow Coma Scale was good and that of the FOUR Score was excellent [Cohen, 2009]. Both the FOUR Score and the Glasgow Coma Scale were predictive of in-hospital mortality and poor outcome at the conclusion of hospitalization.
The FOUR Score Coma Scale has comparable predictive value and inter-rater reliability to the Glasgow Coma Scale and is easier to use. Further studies in children will be necessary to determine the role of the FOUR Score in prediction of long-term outcome.The FOUR Score, however, may not adequately distinguish the vegetative state from the minimally conscious state when compared to the Coma Recovery Scale – Revised score [Schnakers et al., 2006]. This distinction may be more important in chronic critical care and acute and chronic rehabilitation settings than in the acute critical care setting. The Coma Recovery Scale – Revised is more complex than the FOUR Score Coma Scale, requires more training to use, and is more appropriately employed as an outcome measure [Giacino et al., 2004].
Pathophysiology
Consciousness is the result of the complex interplay between the cerebral cortex and the ascending reticular-activating system. The ascending reticular-activating system is a somewhat diffuse and widely circumscribed group of neurons that lie in the reticular formation of the brain that extends from the lower medulla to the midbrain and diencephalon [Moruzzi and Magoun, 1949; Plum and Posner, 1982]. Coma is produced by diseases or conditions that cause bilateral cerebral cortical dysfunction, ascending reticular-activating system dysfunction, or both. Pathologically, these can be subdivided as follows [Adams et al., 1982; Gennarelli et al., 1982; Plum and Posner, 1982]:
Etiologies
There may be considerable overlap between these categories, and it is always necessary to consider combined causes of impaired consciousness and coma. Disorders that produce abnormalities of osmolality, such as hyponatremia, hypernatremia, and diabetic ketoacidosis, may also lead to structural complications, such as intracranial hemorrhage or stroke as a consequence of sinovenous thrombosis. Both have been reported as rare complications of diabetic ketoacidosis [Atlluru, 1986]. Cerebral edema may complicate treatment of ketoacidosis, causing alterations of consciousness or herniation [Green et al., 1990].
Children with psychiatric disorders, such as conversion, panic, or anxiety disorders, may present with neurologic symptoms including genuine or apparent impairment of consciousness. Symptoms may include dizziness, hyperventilation, paresthesias, agitation, or restlessness. The episodic and paroxysmal nature of these episodes and the complete recovery between episodes should suggest the diagnosis [Herskowitz, 1986].
Nutritional causes of alteration of consciousness or coma are relatively uncommon in the United States but represent a significant problem in other areas of the world. These are reviewed in Chapter 103. Wernicke’s encephalopathy is related to a deficiency of thiamine. In addition to consciousness alteration, patients may have eye-movement abnormalities, ataxia, and deep tendon reflex changes. Wernicke’s encephalopathy has been reported in children and adolescents as a result of systemic malignancy and because of thiamine deficiency in total parenteral nutrition, and it may occur related to anorexia nervosa, acquired immunodeficiency syndrome (AIDS), and prolonged fasting [Bruck et al., 1991; Hahn et al., 1998; Nautiyal et al., 2004; Pihko et al., 1989; Salas-Salvado et al., 2000; Vasconcelos et al., 1999].
Niacin or nicotinic acid deficiency (pellagra), with the typical features of mental changes, dermatitis, and diarrhea, is rarely seen in the United States. However, disorders of the synthetic pathway from tryptophan to niacin or nicotinic acid or impaired intestinal absorption of tryptophan in Hartnup’s disease may produce symptoms in children or adolescents [Clayton et al., 1991; Freundlich et al., 1981].
