Hypoxic-ischemic brain injury is a diagnostic term that encompasses a complex constellation of pathophysiological and molecular injuries to the brain induced by hypoxia, ischemia, cytotoxicity, or combinations of these conditions (Busl and Greer 2010). The typical causes of hypoxic-ischemic brain injury – cardiac arrest, respiratory arrest, near-drowning, near-hanging, and other forms of incomplete suffocation, carbon monoxide and other poisonous gas exposures, and perinatal asphyxia – expose the entire brain to potentially injurious reductions of oxygen (i.e., hypoxia) and/or diminished blood supply (ischemia).
Although the concept of hypoxic-ischemic brain injury is generally well accepted in clinical medicine, there remains a lack of consistency with respect to the terms used to denote this type of injury, particularly in the neurorehabilitation literature (Arciniegas 2010). ‘Anoxic brain injury,’ ‘anoxic brain damage,’ and ‘anoxic encephalopathy’ are the most commonly used clinical and research descriptors of this condition, and generally are used as synonyms for hypoxic-ischemic brain injury. However, these terms overstate the severity of one pathophysiologic contributor to injury – decreased delivery of oxygen to the brain, most accurately described as hypoxia rather than as anoxia – and ignore entirely the often concurrent and more injurious decrease in perfusion of the brain (i.e., ischemia). With respect to the use of ‘anoxia’ in these diagnostic terms, true anoxia (frank absence of oxygen in the blood) is a rare and debatably survivable event: although complete cessation of respiratory function eliminates introduction of new oxygen into the circulatory system, oxygen remains available, albeit in rapidly diminishing quantities, in the blood for extraction and use by brain tissue for at least several minutes thereafter. With respect to the omission of ‘ischemic’ from these diagnostic terms, purely hypoxic injuries (i.e., those associated with respiratory arrest but preserved circulation, as in many near-drownings, near-hangings, or sedative intoxications) produce brain dysfunction that tends to be transient and to result in less severe and permanent brain damage than those produced by combined hypoxia-ischemia (i.e., cessation of respiration and circulation, as in cardiac arrest) (Greer 2006; Busl and Greer 2010). The term ‘anoxic encephalopathy’ is problematic for these reasons as well as the fact that it describes a consequence of injury, i.e., encephalopathy, and not the injury itself. For these reasons, hypoxic-ischemic brain injury (HI-BI) more accurately and completely describes the pathophysiology of this subcategory of acquired brain injuries. HI-BI therefore is recommended for use over ‘anoxic brain injury’ or other variants of it as the term with which to deescribe such injuries in clinical and research contexts.
Pathophysiology of Hypoxic-Ischemic Brain Injury
Diminutions in circulating oxygen levels may result from a failure of gas exchange in the lungs, decreases in blood-oxygen saturation from pulmonary dysfunction or interference by other gases (e.g., carbon monoxide), or insufficient levels of hemoglobin (as in profound anemia); in such circumstances, hypoxic brain injury may occur. Decreased perfusion of the brain occurs when blood flow to it is partially or completely restricted (e.g., compression of arterial flow in near-hanging or near-strangulation), when blood pressure is very low (e.g., hypotensive or hemorrhagic shock), or when circulation ceases entirely (e.g., cardiac arrest). These conditions deprive the brain not only of oxygen but also glucose and all other nutrients as well as the nutrient/waste exchange process required to support brain metabolism, resulting in the development of a hypoxic-ischemic state. This state is characterized by cellular energy failure, membrane depolarization, brain edema, excess neurotransmitter release (particularly the excitatory amino acid neurotransmitters) and uptake inhibition, increases in intracellular calcium, production of oxygen-free radicals, lipid peroxidation, and disturbances in autoregulation of cerebral blood flow at the micro- and macroscopic levels (Calvert and Zhang 2005; Busl and Greer 2010). Hypoxic and hypoxic-ischemic states at least transiently disrupt brain function and, if sufficiently severe and/or prolonged, may lead to neuronal death and irreversible brain injury.
It is important to acknowledge that the pathophysiologic processes occurring in HI-BI also are characteristic of the non-hemorrhagic forms of stroke. However, the term ‘stroke’ is generally used to denote injury resulting from focal or multifocal ischemia (i.e., that occurring in one or a few specific vascular territories) whereas HI-BI denotes global (i.e., whole brain) exposure to and injury from hypoxia and/or hypoxia-ischemia. Having said that, not all areas of the brain are equally vulnerable to the injurious effects of hypoxia and hypoxia-ischemia: injury from these processes tends to be most pronounced in the superior brainstem, cerebellum, white matter and subcortical structures supplied by the distal branches of deep and superficial penetrating blood vessels, cerebral white matter at the zones between the major cerebral artery territories (so called ‘border zones’ or ‘watershed areas’), CA1 region of the hippocampus, and neocortical layers 3, 5, and 6 (injury to which produces ‘laminar cortical necrosis,’ referring to the death of cells in these layers, or lamina, in the cortex) (Arbelaez, Castillo et al. 1999; Chalela, Wolf et al. 2001; Busl and Greer 2010).
Neurological and Neurobehavioral Consequences of Hypoxic-Ischemic Brain Injury
The consequences of HI-BI commonly include seizures (event-related and recurrent), disturbances of sensorimotor function, and a broad array of cognitive, emotional, and behavioral disturbances (Anderson and Arciniegas 2010; Lu-Emerson and Khot 2010).
Seizures and Myoclonus
As many as one-third of individuals sustaining an HI-BI develop seizures in the immediate post-injury period, typically beginning within 24 hours of injury but occurring or recurring over the first two weeks thereafter. The development of such seizures most likely reflects the effects of injury-induced excitotoxic processes on cortical neurons. Most post-hypoxic seizures usually are partial complex or myoclonic in character and occur intermittently. The occurrence of early seizures does not necessarily portend the development of post-hypoxic epilepsy or persistent post-hypoxic myoclonus nor does it invariably predict poor neurological or functional outcome. However, the post-hypoxic status epilepticus (SE) or myoclonic SE is associated, almost invariably, with a fatal outcome from the HI-BI (Rossetti, Logroscino et al. 2007). It is likely that mortality associated with SE is a reflection of the severity of underlying injury rather than the development of SE per se, although the possibility of SE aggravating the underlying neurological injury has not been definitively excluded. The frequency of late seizures after HI-BI is not well established, although common clinical experience suggests that a nontrivial minority of individuals experience this problem. Treatment of post-hypoxic epilepsy and/or myoclonus follows that of other secondary epilepsies and myoclonus, and generally is similarly effective.
Post-hypoxic parkinsonism, dystonia, chorea, athetosis, and tremor are rare but potentially disabling consequences of HI-BI. Among these problems, post-hypoxic parkinsonism and dystonia are most common. Post-hypoxic parkinsonism is generally symmetric and predominantly akinetic-rigid (i.e., not tremor-predominant) but may sometimes include resting or postural tremor as well. The development of this condition most likely reflects the vulnerability of the globus pallidus and the substantia nigra - pars reticularis to the adverse effects of hypoxia and/or ischemia. Post-hypoxic dystonia, which tends to reflect injury to the putamen, is often asymmetric initially but over time may progress to a more symmetric and generalized form. Although these conditions may develop in the early post-injury period, they often are delayed sequelae of HI-BI, developing months or years after injury. Unfortunately, these conditions appear less responsive to pharmacologic treatment than primary parkinsonism (i.e., Parkinson’s disease) and idiopathic dystonia, perhaps reflecting hypoxic-ischemic-induced damage and/or destruction of the neurons in these structures that ordinarily are the targets of these pharmacotherapies.
Disorders of Elementary Sensorimotor Function
Injury to descending corticospinal tracts, whether in the deep white matter of the cerebral hemispheres, in the crus cerebri at the level of the midbrain, or in the spinal cord, may produce impairments of elementary motor function. Involvement of the corticospinal tracts at the level of the cerebral hemispheres or upper brainstem may produce variable patterns and severities of motor weakness, up to and including quadriplegia. An uncommon but remarkable post-hypoxic motor syndrome is the ‘man in a barrel’ syndrome, or bibrachial paresis; this condition is characterized by bilateral proximal upper extremity paresis with preservation of lower extremity function, and reflects hypoxic-ischemic injury to the ‘watershed’ zone of white matter between the anterior and middle cerebral artery territories. Similarly, paraparesis and quadriparesis are potential consequences of hypoxic-ischemic watershed infarctions in the upper and lower thoracic and lumbar regions of the spinal cord. Rehabilitative interventions for these problems, and complications of them such as spasticity, contractures, gait and mobility impairments, follow those applied for similar motor impairments due to other causes. The effectiveness of these motor-specific rehabilitative interventions in this population is not well established, but common clinical experience and several rehabilitation outcome studies (Shah, Al-Adawi et al. 2004; Burke, Shah et al. 2005; Shah, Carayannopoulos et al. 2007) suggests that typical rehabilitation programs may improve the functional status of many HI-BI survivors with such problems.
Watershed infarctions occurring the posterior portions of the cerebral hemispheres may produce disturbances in sensory function, and particularly impairments of visual processing. Cortical blindness and the Balint syndrome (comprised by ocular apraxia, optic ataxia, and simultanagnosia) are specific examples of disorders of sensory function that may be associated with HI-BI. Optimal rehabilitative strategies for these problems are not well developed presently.
The most extensively studied neurobehavioral sequelae of HI-BI are cognitive impairments. Most common among these are the disorders of consciousness (e.g., coma, vegetative states, minimally conscious state), impairments of attention and processing speed, memory impairment, and executive dysfunction, although disorders of language, apraxias, agnosias, visuospatial dysfunction, Balint’s syndrome (as noted above), Anton’s syndrome (anosagnosia for visual impairment), personality changes, behavioral disturbances, and disorders of mood and affect regulation have been reported as well. Discussion of the character and neuroanatomy of these problems is beyond the scope of this article, but is summarized in Anderson and Arciniegas (2010). In short, the development of this broad range of post-hypoxic cognitive impairments is consistent with the vulnerability of many cognitively salient areas to the adverse effects of global hypoxia and/or ischemia; these areas include the upper brainstem, thalamus, cerebellum, basal ganglia, medial temporal structures (especially the CA1 field of the hippocampus), cortical layers 3, 5, and 6, as well as the cerebral hemispheric deep white matter through which cognitively salient areas are connected to one another and also to motor output areas. Cognitive recovery is both common and remarkably robust in many cases, with as many as two-thirds of HI-BI survivors making substantial or complete cognitive recoveries over the first 1-2 years post-injury. Unfortunately, for those individuals in who post-hypoxic cognitive impairments persist, they are often severe and functionally disabling. The variability in cognitive outcome reflects, at least in part, reflects the effects of severity of injury, cause of injury, age of the individual affected, and interactions between these and other factors on the neuroanatomy of injury and potential for neural recovery. When interventions for post-hypoxic cognitive impairments and their functional consequences are required, nonpharmacologic and pharmacotherapeutic approaches are generally modeled after those provided to persons with posttraumatic cognitive impairments. The effectiveness of these interventions in this population is not well established, but common clinical experience suggests that they may be of benefit to some persons with HI-BI.
Delayed Post-Hypoxic Leukoencephalopathy
In rare cases, early and complete recovery from HI-BI is followed a few days to weeks later by a severedemyelinating syndrome; this syndrome, delayed post-hypoxic leukoencephalopathy, characterized by acute or subacute onset of severe and progressive neuropsychiatric problems such as delirium, psychosis, parkinsonism, and/or akinetic-mutism, and/or quadriparesis, among others. Although this condition is often described as a delayed sequelae of carbon monoxide-induced HI-BI, it has been associated with nearly all causes of HI-BI (Shprecher and Mehta 2010). The neural mechanisms of delayed post-hypoxic demyelination have not been established definitively. However, combinations of toxic exposure (e.g., carbon monoxide, inhaled heroin), genetic (e.g., pseudodeficiency of arylsulfatase A, abnormalities of other genes regulating myelin turnover), and age-associated vascular risk factors have been suggested as possible contributors to this unusual post-hypoxic condition. Regardless of mechanism, this syndrome is characterized neuropathologically by diffuse bihemispheric demyelination that generally spares the cerebellum and brainstem. Neurological and neurobehavioral improvement over the first 3 to 12 month periods following onset of this syndrome is typical, but many survivors experience persistent cognitive impairments (particularly impairments of attention, processing speed, and/or executive function), parkinsonism, and/or corticospinal tract signs. There are case reports describing symptomatic and functional improvement of the cognitive and parkinsonian sequelae of delayed post-hypoxic leukoencephalopathy during treatment with stimulants, amantadine or levodopa. The observation that these agents offer some benefit in this context despite their lack of efficacy for the same sequelae of HI-BI itself may reflect differences in the anatomy of these conditions: in HI-BI there is involvement of both gray and white matter, limiting the target of pharmacotherapies more severely than in delayed post-hypoxic leukoencephalopathy, which involves only white matter.
Conclusions and Future Directions
HI-BI is a consequence of many medical illness as well as accidental and non-accidental injuries. The neurological and neurobehavioral sequelae of HI-BI are many, often severe, and present substantial challenges to the survivors of these injuries, their families, and their healthcare providers. The basic science literature is rife with studies of the effects of hypoxia and/or ischemia on the developing and adult nervous systems. Nonetheless, HI-BI remains a relatively understudied clinical condition, particularly in the rehabilitation literature. Applying a certain measure of care-by-analogy to the management of persons with traumatic brain injury therefore is understandable and unavoidable – doing so allows those of us working with persons with HI-BI and their families to organize and deliver care that promotes neurological and functional recovery, supports adaptation to disability, and, to the greatest extent possible, facilitates re-entry into the community and workforce. Nonetheless, our efforts to provide care to and improve outcomes among persons with HI-BI will benefit from a more detailed knowledge of this condition, its neurological and neurobehavioral sequelae, and the optimal methods and systems of care needed to evaluate and rehabilitate individuals and families affected by this condition.
Toward that end, readers of the Neurotrauma Letter may find of use a series of articles published earlier this year in a special issue of the journal NeuroRehabilitation (IOS Press) focused exclusively on HI-BI. Included in this collection are reviews of the neuropathophysiology, neuroimaging assessment, and the evaluation and management of the neurological and neurobehavioral sequelae of these injuries in adults and children, as well as a discussion of the limitations of current public policy approaches to HI-BI in the United States. Also presented are reviews of two related topics, hypobaric (high-altitude) hypoxic cerebral injury and obstructive sleep apnea, both of which serve as models for the pathophysiology of HI-BI and which may inform the evaluation and management of persons with HI-BI more generally. Drs. Zasler and Kreutzer, Editors ofNeuroRehabilitation, and I hope that readers of this letter as well as others interested in HI-BI will find this issue informative and useful in their study of, neurological and neurorehabilitative care for, and advocacy on behalf of persons with HI-BI and their families.
- Anderson, C. A. and D. B. Arciniegas (2010). "Cognitive sequelae of hypoxic-ischemic brain injury: a review." NeuroRehabilitation 26(1): 47-63.
- Arbelaez, A., M. Castillo, et al. (1999). "Diffusion-weighted MR imaging of global cerebral anoxia." Ajnr: American Journal of Neuroradiology 20(6): 999-1007.
- Arciniegas, D. B. (2010). "Hypoxic-ischemic brain injury: addressing the disconnect between pathophysiology and public policy." NeuroRehabilitation 26(1): 1-4.
- Burke, D. T., M. K. Shah, et al. (2005). "Rehabilitation outcomes of cardiac and non-cardiac anoxic brain injury: a single institution experience." Brain Inj 19(9): 675-680.
- Busl, K. M. and D. M. Greer (2010). "Hypoxic-ischemic brain injury: pathophysiology, neuropathology and mechanisms." NeuroRehabilitation 26(1): 5-13.
- Calvert, J. W. and J. H. Zhang (2005). "Pathophysiology of an hypoxic-ischemic insult during the perinatal period." Neurological Research 27(3): 246-260.
- Chalela, J. A., R. L. Wolf, et al. (2001). "MRI identification of early white matter injury in anoxic-ischemic encephalopathy.[see comment]." Neurology 56(4): 481-485.
- Greer, D. M. (2006). "Mechanisms of injury in hypoxic-ischemic encephalopathy: implications to therapy."Seminars in Neurology 26(4): 373-379.
- Lu-Emerson, C. and S. Khot (2010). "Neurological sequelae of hypoxic-ischemic brain injury."NeuroRehabilitation 26(1): 35-45.
- Rossetti, A. O., G. Logroscino, et al. (2007). "Status epilepticus: an independent outcome predictor after cerebral anoxia." Neurology 69(3): 255-260.
- Shah, M. K., S. Al-Adawi, et al. (2004). "Functional outcomes following anoxic brain injury: a comparison with traumatic brain injury." Brain Inj 18(2): 111-117.
- Shah, M. K., A. G. Carayannopoulos, et al. (2007). "A comparison of functional outcomes in hypoxia and traumatic brain injury: a pilot study." J Neurol Sci 260(1-2): 95-99.
- Shprecher, D. and L. Mehta (2010). "The syndrome of delayed post-hypoxic leukoencephalopathy."NeuroRehabilitation 26(1): 65-72.