Corticosteroids and Traumatic Brain Injury: Status at the End of the Decade of the Brain.
Of the 2 million patients with traumatic brain injuries (TBI) that are evaluated annually (US), 10% are considered severe. Additionally, 50,000 persons with severe TBI die at the scene. In spite of advances in resuscitation and intensive care, prognosis remains as dismal as it was 20 years ago, and mortality has remained constant at between 20-50%. Glucocorticosteroids have been used in TBI for over 30 years. Ghajar et al surveyed American trauma centers (N=277) and reported that in 64% of those centers, steroids were used in more than half of their patients. A similar survey from the United Kingdom and Ireland reported that 14% of responding centers (n= 35) identified steroids as their "treatment of choice" for intracranial hypertension - in the absence of strong endorsements of efficacy and the risks associated with steroid use. This article reviews the neuropathology of severe TBI and evidence for steroid efficacy in this population, and addresses implications for the future.
Physiopathology of Brain injury
Primary or "impact" injury refers to the initial physical forces (kinetic energy) applied to bony, glial, neuronal and vascular elements, producing structural and functional changes. Translational and rotational forces distort skull, brain and vessel architecture, producing, at the macroscopic level, contusions, lacerations and diffuse axonal injuries, including axotomy following severe shear. Large and small vessel injuries are expressed as hemorrhage, dissection, thrombosis or microhemorrhages; all can create areas of focal ischemia.[50,59,65,70,71,75] The blood-brain barrier (BBB) may also be physically disrupted. It is an anatomic-physiologic barrier characterized by tight junctions between cerebrovascular endothelial cells and limited transcellular vesicular transport, and normally restricts the egress of ionized and water soluble substances and large proteins out of the vascular compartment. Additive shear effects and percussion waves transiently open the BBB and increase permeability to large molecules.[2,32,35,42,64]
Functionally, waves of kinetic energy disturb hypothalamic and brainstem vasomotor center function, producing initial transient vasomotor paralysis, leading to reduced cerebrovascular tone, increased cerebral blood volume (CBV) and a congestive hyperemia.[45,73] Increases in brain blood and water volumes elevate intracranial pressure (ICP), reduce cerebral perfusion pressure (CPP) and lead to perfusion failures. The violence of the impact to the head alone can change ion channel permeability, initiate extensive depolarization, with subsequent increases in extracellular (EC) potassium ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]), increased intracellular (IC) concentrations of calcium ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]), sodium ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]), chloride, obligatory water and the loss of free magnesium.[20,60,65,71]
As early as 1911, Allen realized that multiple events - interdependent and additive in their effects - could occur after primary injury and destroy the substrate required for neurological recovery. Today, a multitude of cascading biochemical and cellular processes have been implicated as possibly underlying late clinical deterioration that may affect as many as one third of survivors. Mechanisms of secondary injury include ionic ([Ca.sup.++], [Na.sup.+]), water and neurotransmitter derangements, excessive activation of excitatory amino acids (EAA), failure of cellular energetics with free radical formation and destruction of CNS membranes by lipid hydrolysis/peroxidation and a destructive inflammatory response. The expression, temporal course, interrelationships and susceptibility to correction of secondary injury mechanisms across injury severity and type have not been elucidated; the key to improving clinical outcome appears to be understanding, then avoiding or reducing the sequelae of secondary insults.[13,23,50,65,67] The interaction of systemic (eg, hypotension, ischemia, abnormalities of temperature, electrolytes and osmolality) and intracranial events (eg, intracranial hypertension, brain edema, hydrocephalus, seizures and alterations in cerebral blood flow), may amplify the destruction of secondary injury, complicate clinical management and adversely affect outcome.[13,20,70,75] What follows is a review of selected secondary injury processes thought to be important in the pathobiology of TBI.
Glutamate, NMDA Receptors and Calcium
Severe impact injury that deforms brain substance, alters ion channel permeability and dysregulates ion movement induces widespread depolarization that involves glutamate, the most abundant excitatory amino acid in the brain. Glutamate binds to at least four classes of receptors: NMDA (N-methyl-D-aspartate), kainate, AMPA (alpha-amino-3-hydroxy-5-methyl-4-isoxax-zoleproprionic acid) and a family of Q (quisqualate) receptors. Although the receptor groups may not participate equally in secondary injury processes, and different cellular populations likely differ in their respective vulnerabilities to glutamate, glutamate release is linked to subsequent receptor mediated injury.[15,60,67] What signals the increased production and release of the neuromodulator is unclear; some have suggested an initiating hypothalamic signal. Excess EC glutamate overstimulates postsynaptic action potential generation:[15,47] AMPA activation facilitates IC movement of [Na.sup.+], [K.sup.+] and hydrogen ([H.sup.+]); [Na.sup.+] influxes drive the ensuing NMDA depolarization. In turn, NMDA activity stimulates [Ca.sup.++] conductance into the cell.[65,75] These ionic fluxes further reverse membrane polarity, which continues to drive depolarization. Reversal of the [Na.sup.++] - [Ca.sup.++] pump amplifies intracellular [Ca.sup.++] accumulation; ischemia opens receptor-operated [Ca.sup.++] channels and contributes to the breakdown of ionic homeostasis.[47,60,67] Potassium shifts to the EC compartment produce rapid astrocytic swelling as they attempt to absorb [K.sup.+] to preserve ionic equilibrium, and this displacement produces its own set of events, including changes in depolarization and membrane, metabolic and synaptic function. It can also increase monoamine release, which increases glial permeability to [Na.sup.+], leading to more swelling.[7,9,23,47,50,67]
Injury-induced ionic perturbations require that energy be diverted to restore equilibrium through ATP-dependent membrane pumps, shifting available energy from normal cellular functions. For example, NMDA activity controls the [Ca.sup.++]-[K.sup.+] exchange, and [Ca.sup.++] influx activate the pump and consumes energy. However, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] irreversibly binds to mitochondrial membranes and reduces ATP production, disrupting energetics even further. Postsynaptic oxygen is diverted to a sequence of reactions with EAA which induce hypermetabolism that outstrips the energy supply required to sustain normal cell function. Energy needs are satisfied by hyperglycolysis, with IC accumulation of lactate and increasing acidity ([pH.sub.IC]).[20,47,67] NMDA receptor overactivity appears to link synaptic overactivity, reduced glutamate re-uptake in neurons and glia and intracellular [Ca.sup.++] overload.[9.59.67,70]
Calcium is a key regulator of cell growth and differentiation, cytoskeletal integrity, membrane excitability and exocytosis and synaptic activity and ubiquitous as a co-factor in IC processes. Its focal role is dependent upon tight control of IC and EC concentrations. Influx occurs through voltage-dependent and ligand-gated channels. Intracellular sequestration and buffering is energy dependent, and efflux is reliant upon normal pump function. Although multiple ions participate in early depolarization, calcium, because of its central role in normal cell function, dominates the changes that occur in secondary injury. Intracellular [Ca.sup.++] accumulation is associated with the activation of lipolytic and proteolytic enzymes, the breakdown of the microtubular structures, and enhanced catabolism. Reductions in pH and falling magnesium both amplify [Ca.sup.++] mediated havoc.[65,67]
Direct clinical evidence for the presence of glutamate in humans after severe TBI is very recent. Bullock et al, using microdialysis cortical probes, recorded elevations in EAA 10-20 times normal in patients with contusions. Secondary ischemic events which elevated ICP and lowered CPP were associated with EAA spikes 50X normal. Elevations, which may reflect limited ability of injured astrocytes to take up the EAA, persisted through study duration (4 days) and were associated with poorer outcomes. It was hypothesized that initial primary injury may cause micropores in membranes, with glutamate leakage contributing to edema and ischemia by opening ion channels. High levels have also been documented in human cerebrospinal fluid (CSF) after injury.[23,50,62] Animals also have very high glutamate levels after experimental injury. These data support the importance of an excitotoxic process that is linked to acute posttraumatic cell swelling and amplification of [Ca.sup.++] mediated injury in at least some types of injury.[9,50,62]
Metabolites of Arachidonic Acid
Accumulating intracellular [Ca.sup.++], augmented by reduced endoplasmic reticulum, mitochondrial and protein buffering, activates proteases, lipases and peroxidases which attack cell membranes, mitochondria and intracellular fatty acids, releasing phosphates and lysosome contents which exacerbate autodestruction. Phospholipase A2 and C, activated by [Ca.sup.++] and thrombin, split arachidonic acid from membrane phospholipids. Active prostaglandins (PG), prostanoids (thromboxanes (TXAs), prostacyclin) leukotrienes and other lipid mediators result from its partial oxygenation. The amount and type of products vary by substrates, enzyme systems and co-factors present and the initiating stimulus. Phospholipase A may also increase platelet activating factor, which can worsen ischemia and disrupt BBB integrity
The metabolic products, particularly the prostanoids, influence cerebrovascular tone and flow directly and by activation of inflammatory mediators.[23,44,47,50,59,65,67] Large extraparenchymal intracranial arteries tend to produce prostacylin and [PGF.sub.2], while smaller surface pial arteries generate prostacyclin. In areas or states of incomplete ischemia, thromboxane may dominate and reduce flow. The cumulative effect of arachidonic acid product activity is to exacerbate ischemia by enhancing vasospasm and activation of potent inflammatory mediators.[23,44]
Free Radical-Induced Lipid Peroxidation
Free radicals are short lived but highly reactive compounds normally generated in cellular metabolism and destroyed by endogenous anti-oxidants such as Vitamin E and a variety of enzyme systems. The reaction between a free radical and membrane fatty acid leads to a chain of events known as lipid peroxidation (LP), the "signature process of oxygen-radical-induced injury in the brain"(p, 118). Adjacent molecules interact to pair a free electron and lower its energy. Free radicals remove a H+ from a polyunsaturated fatty acid, forming a lipoxy radical (L), a carbon-centered free radical located within the lipid membrane of neurons, glial elements or vascular cells. Lipoxy radicals react with membrane oxygen, forming LOO??, which in turn attacks structural proteins, enzymes, nucleic acids, then a second fatty acid. If unopposed, the process spreads over cell membrane surfaces, causing conformational (structural) changes, impairment in enzymatic function, disruption of ion gradients, and, if severe enough, destruction of the lipid and protein constituents of the cell, and ultimately, cellular disintegration until it is scavenged by an electron acceptor. LP reaction products - aldehydes; hydrocarbon gases - are also toxic, altering energy functions, increasing vascular permeability and causing edema.[33,47,59] Oxygen free radicals also produce vascular dilatation and impair autoregulation, which may further compromise per fusion.
Loss of [Ca.sup.++] homeostasis and the generation of free radicals are coupled events.65 Enhanced formation related to increased [Ca.sup.++] IC occurs by two mechanisms; the first is believed to be the predominant source of free radicals. Intermediates of arachidonic acid metabolism generate and release lactate and O2?? during restoration of perfusion.[59,65,67,71] The second mechanism proposes that in ischemic brain there is enhanced [Ca.sup.++] catalyzed conversion of xanthine dehydrogenase to xanthine oxidase, which, with restoration of perfusion, produces a burst of radicals.[57,65] Increased [Ca.sup.++] may also deactivate endogenous free radical scavengers.
The central nervous system is the most supportive environment for free radical generation, with a ravenous appetite for oxygen and replete with polyunsaturated membrane lipids; large numbers of lysosomes and reduced scavengers relative to other tissues.[40,59,70] Sulfur containing amino acids, neuronal and glial membranes are vulnerable to oxidant injury. The microvasculature, with a high Pa[O.sub.2], high levels of xanthine oxidase and exposed to platelets, neutrophils and other participants in the inflammatory response, is especially susceptible. Transient episodes of hypotension or hypoxia, common after severe TBI, can increase xanthine oxidase activity and produce a damaging burst of radicals. When insufficient oxygen is available to accept electrons in the mitochondria, oxygen auto-oxidizes, generating free radicals.[59,67] In contused and hemorrhagic areas, free radical mediated injury may be very severe. Dislocation of endogenous iron from hemoglobin, transferrin, ferritin and other proteins during hemorrhage occurs and acts as a metal catalyst, donating free electrons for free radical formation, further catalyzing radical-initiated membrane disruption. Acidosis may drive injury by dislocating protein-bound iron.[23,59,65,71]
There is strong experimental evidence for the significance of LP. In humans, evidence for the importance of oxygen free radical induced LP is circumstantial. Survivors manifested elevated CSF levels of LP byproducts which didn't correlated with admission GCS scores. Other indirect evidence includes increased levels of LP products in jugular plasma after injury, positive correlations of increased free radicals and loss of vascular regulation, vasogenic edema and brain swelling, as well as some experimental support for a neuroprotective antioxidant effect.[32,33,59,71] LP is probably a dominant posttraumatic cause of secondary injury, associated with microvascular injury, loss of axons, BBB dysfunction, membrane permeability changes and edema, increased ionic fluxes which increase energy consumption.[16,20,22,71] Presently, although free radical mediated LP can be extrapolated, its pattern, temporal course and variability across injury type or severity have not been characterized.
The brain, with a high constant need for flow, high [O.sub.2] extraction ratio and minimal compensatory reserves, is exquisitely sensitive and vulnerable to failures of perfusion, whether they are autoregulatory responses to acute injury or produced by systemic events.[23,69,70] Histologic evidence for ischemic lesions in fatal TBI exceeds 85%.[13,39,50] In survivors, evidence for an "ischemic penumbra," regions of marginally perfused tissue which is viable if temporarily dysfunctional, support the criticality of guaranteeing adequate flow. Focal or generalized loss of autoregulatory capacity, the intrinsic ability of the brain to sustain its own perfusion over a wide range of pressures, is a response to brain injury. When upper or lower limits are exceeded, blood flow becomes pressure dependent. At low CPPs, the pressure dependency of cerebral circulation makes the brain vulnerable to hypotensive insult; in areas of ischemia, the BBB may open. Excessive pressures can render the BBB incompetent, resulting in edema, increased ICP and hemorrhage. Cerebral blood flow is determined by local regulation (eg, Pa[CO.sub.2], Pa[O.sub.2], neurogenic mechanisms), CPP, the gradient driving forward flow and cerebrovascular resistance (CVR), which is defined by Poiseuille's law. There is evidence in human injury of blood flow abnormalities, ranging from hyperemia to ischemia to infarct.[1,57] Very high and very low flow states reflect an uncoupling of metabolism from flow and are associated with poor outcome.
There is some support in human injury for the importance of early ischemia which is defined in terms of flow ([is less than] 18 ml/100G brain/min) and metabolism (oxygen extraction). Transcranial Dopplers of middle cerebral artery flow performed within 60 minutes of severe blunt TBI disclosed marked reduction in flow velocity that was strongly associated with severe injury. In the first 4-6 hours after injury, another group found that extremes of flow, when associated with low [AVDO.sub.2], suggests true early ischemia, and was associated with poor outcome. Early ischemia has also been linked to vasospasm in large arteries, affecting as many as 40% of patients.[5,20,24,50] Delayed, generalized reduction in CBF, that is, postischemic hypoperfusion, has been documented after and associated with severe injury[23,50,53,57,62] There is suggestion that early changes in CBF establish a foundation for cellular ischemic injury, and may initiate or stimulate the secondary biochemical and cellular cascades which further damage the injured brain. Traumatized brain is hypersensitive to episodes of late hypoxia/ ischemia.[7,50,60] Focal ischemia causes a cellular "power failure;" clinical consequences seem proportional to duration and degree of perfusion compromise. Unchecked ischemia leads to depletion of high energy compounds, causing electrical and membrane abnormalities. Failure of energy dependent ionic pumps destroys the transmembrane gradients necessary to protect normal IC function. Ionic shifts abolish normal membrane potentials and support membrane depolarization which in turn accelerates [Ca.sup.++] entry and propagation of the injury cascades described above.[7,60,65] Reperfusion is associated with increased generation of free radical species as oxygen is reduced by xanthine oxidase, which can participate in LP and contribute to vascular paralysis.
Brain injury induces a generalized inflammatory response, which begins very early after injury, and peaks several days later. It is characterized by gliosis, recruitment of white cells and platelets, cytokine and chemokine (chemoattractant cytokines) release, increases in vascular permeability and endothelial injury.[23,25,28] Systemically generated factors can cross leaky BBB in to brain parenchyma. Neurons, astrocytes, microglia and vascular endothelium are capable of cytokine and chemokines synthesis, producing TGF, IL-1 IL-6, TNF and growth factors which may go on to participate in secondary injury events.[23,25,28,42,50.51.72]
Experimental evidence supports the increased presence of multiple inflammatory mediators after injury.[25,72] In humans, posttraumatic cytokine assays showed elevated systemic and intracranial (CSF) levels of IL-6, suggesting it may have a role of some sort. Serum and CSF levels peaked in days 1 and 2 after-trauma; levels correlated strongly when BBB was disrupted. IL-1, IL-8 and TNF were undetectable, but may have been assayed too late or present in lower than detectable amounts. Evidence is contradictory for their presence and significance at this time. IL-1 and TNF prime the endothelium for white cell adhesion and subsequent vascular plugging, generation and release of free radicals. IL-6 may impede or prevent tissue oxygen utilization, thus contributing to poor outcome and at least partly explain why optimizing Pa[O.sub.2] hasn't been successful in improving outcomes.
The role of immune cells in TBI is largely unknown. In animal and human injury, activated white cells adhered to vascular endothelium and accumulated in the microvasculature, especially during the first 4-48 hours. Presumably, their accretion is dependent on the increased expression of adhesion molecules on activated endothelium. Microglial (CNS macrophage equivalents) cells have also been observed in injured brain. However, at this time, there is no compelling evidence for white cell-mediated injury; their relationship to secondary injury cascades and posttraumatic cerebrovascular /BBB function is not known.[35,71]
Cerebral edema is "a condition of increased water per unit volume of cerebral parenchymal tissues (p.3), and exists in three varieties which are classified by the site of increased brain water. Vasogenic edema is excess third space water and linked to increased BBB permeability; cytotoxic edema is excess endothelial and neuroglial water and linked pathologically to disorganized osmoregulation of the cell. Interstitial edema refers to an expansion of the CSF compartment. Edema of any type becomes problematic when it causes compression or displacement of adjacent structures or impairs perfusion.[26,64] The Monro-Kellie model describes the volume pressure relationship of intracranial constituents. Blood, water and cellular elements, deformable but incompressible, generate an intracranial pressure (ICP). Increases in the volume of any one or more elements, if able to overwhelm the compensatory, reduction in the remaining elements, elevate ICP, reduce CPP and CBF and lead to ischemia. Effective management depends on the cause.
Intractable intracranial hypertension is the most common cause of death in severe TBI. Pathologically, it is related to overwhelming increases in any one or more of the intracranial compartments; the affected compartment(s) and compensation may change over time, and vary with injury type, severity and intervention. Initially, it may be primarily vascular in etiology - that is, an initial hyperemia associated with the violence of the impact injury. Vasodilatation may be direct vessel response to trauma.[1,23,45,62,71]
Posttraumatic edema is vasogenic and cytotoxic and evolves over time; in human head injury, CT evidence of edema is most pronounced 3-4 days after injury.[2,23] Patterns may vary by injury type and severity. Microvascular or BBB disruption by primary impact and inflammatory response changes result in altered vascular endothelial permeability and early vasogenic edema. Excessive glutamate-NMDA activation, [Ca.sup.++] mediated free radical and LP injury, inflammation and acidosis dominate the mechanisms associated with IC and EC accumulation of brain water. Reperfusion injury, causing diminished capacity to protect ionic gradients and IC volume, is associated with cytotoxic edema formation; inflammatory mediators, especially where BBB is open, have been linked to vasogenic edema formation. Virtually every secondary injury process has been linked to development or perpetuation of either cytotoxic and vasogenic edema, and reflects collapse of cellular energetics and osmoregulation.
Cerebral edema, however, may not be a key factor after TBI and has been called "an epiphenomenon of the more basic injury events" (p.971). Cerebral edema, possibly a normal concomitant of injury, appears to be a gross, clinical marker for the composite deleterious effects of multiple biochemical and cellular derangements that occur after severe TBI, culminating in intracractable intracranial hypertension and death. Yoshino et al studied brain (contrast) CT-intracranial hypertension and circulatory changes in 42 adults with isolated TBI. Of the 25 fatalities, 16 had hematomas and 9 had brain bulk increases or malignant brain swelling. Those with brain bulk increases had increased CBF (and volume) confirmed by Xenon uptake. In contrast, survivors with brain bulk increases manifested density changes suggestive of increased brain water rather than blood. He concluded that the relative roles of edema and hyperemia are unclear and that effective intervention must be mechanism-specific, and felt that the "ineffectiveness of steroids in controlling increased ICP in HI may support the irrelevance of brain edema"(p.837) in that group.
Conventional doses of the glucocorticosteroids bind to cytoplasmic receptors, form steroid-receptor complexes which migrate to the nucleus, where they induce or suppress expression of cell-specific genes. Antiinflammatory activity is related to glucocorticosteroids ability to inhibit phospholipase A2, cyclooxygenase and lipooxygenase pathways, thereby limiting the release of arachidonic acid and its metabolites. Other actions are summarized on Table 1. The net effects of reducing vasoactive prostanoids, PG, and vascular sensitivity are to reduce edema and ischemia.[14,29,31,44,54]
* Inhibit phospholipase A2, cyclooxygenase and lipooxygenase pathways [right arrow] limit arachidonic acid release, including TXAs.(44)
* Downregulate pro-inflammatory cytokines - notably 1[Alpha] 1[Beta], IL-6 and TNF [right arrow] dampen the inflammatory response
* Reduce vascular permeability (endothelial response), plasma and protein leak into the EC space
* Inhibit phagocytic response [right arrow] possible reduction in oxygen radical generation
* Improve blood flow through direct dilatation and reduced sensitivity to vasoconstrictors
* May reduce mRNA stability which mediates gene transcription, including those involved in the synthesis of cytokines and chemokines.
In the mid to late 1980s, methylprednisolone (MP), at doses 10-1000 times the level required to saturate receptors, was found to attenuate cellular injury. By a receptor independent means, high dose MP could exert a direct antioxidant effect, achieving membrane concentrations sufficient to defend susceptible fatty acids and cholesterol from attack. It may also have had beneficial effects on calcium homeostasis and prevention of progressive posttraumatic ischemia by multiple means. Doses of 30 mg/kg dexamethasone may be required to inhibit lipid peroxidation. That is, conventional dosing may achieve antiinflammatory effects and preserve or support perfusion; much higher doses may confer some protection against LP by a different mechanism.
Evidence for Corticosteroid Efficacy in TBI
Steroid use in TBI was stimulated by the clinical observation in 1961 that steroids reduced central nervous system peritumoral edema[41,59] and that this effect was often associated with dramatic functional improvement. Development of peritumoural edema was attributed to blood brain barrier (BBB) dysfunction, permitting water and solute entry into the extracellular spaces. Rapid and severe white matter swelling was the result. This vasogenic edema was remarkably amenable to reduction by steroids, and often occurred without changes in ICP or reduction in CT-evident brain edema. Presumption of efficacy in TBI injury clearly rested on the assumption that posttraumatic edema was also vasogenic. Clinicians assumed that steroids reduced edema, which should theoretically reduce intracranial pressure and improve outcome).[17,30]
Introduction of pathology specific drug therapy in humans is usually preceded by unequivocal evidence of efficacy and safety in preclinical research, using injury models that reflect the pathophysiology as understood in humans. In TBI, small animal studies used different species, models of injury, anesthesia, different steroids in different dosages for variable periods of time; provided poor definitions of dose response characteristics, different follow-up periods, using different outcome measures and weak control of extraneous variables. Findings were equivocal in suggesting steroid benefit.[23,31,55,59] Though there existed little in the way of powerful rationale to extrapolate to humans, clinical use was introduced by desperate clinicians lacking any effective therapy and persisting high morbidity and mortality. Preliminary human studies were characterized by small samples, with weak control of extraneous variables (eg, age, treatment protocols, ICU management, drug, posology, treatment duration, difficulties with injury severity stratification and meaningful outcome measures. In short, the first efforts did not support a beneficial effect for steroids in severe TBI.
Equivocal early results led to 5 prospective randomized clinical trials (RCT) between 1979 and 1983 involving almost 600 patients (Table 2), and they be assessed in their historical context. Early work was hobbled by absence of standardized systems to describe injury severity and outcome. Computerization lacked present day sophistication, and the methodological approaches to answering questions of efficacy were rudimentary. Popularization of the Glasgow Coma Scale (GCS), Glasgow Outcome Scale (GOS), computed tomography (1973) and intracranial pressure monitoring (1960s) revolutionized patient care and facilitated cross center comparisons with universal injury severity and outcome languages. However, until the past decade there has been a dearth of basic science driving clinical neurosurgical management of TBI. In hindsight, study discussions were remarkable for their lack of discourse about pathology specific interventions. In large part, clinical observations and logic drove the study of steroids in TBI.
Table 2 Major Studies of Steroid Use in Traumatic Brain Injury Author (y) N Design Gudeman et al 20 prospective; (1979) US severe HI descriptive Time to treat: 6 hours 80% GCS [is less 1. 40 mg Q6h; 2G Duration Rx: unspec than or equal bolus, 4 x 500 mgm to] 8 Q6h; 4x 250mg Q6h, 20/20 ICPM 40 mgm Q6h 2. Retrospective comparison Methylprednisolone Cooper et al 76 RCT prospective, (1979) US severe HI double blind Time to treat: 6 hours (Grady Scale) Duration Rx: 11d 51/76 ICPM 1. Placebo 2. Bolus: 10 mgm IV;16mg/d x 6d IM; 5 day taper 3. Bolus: 60 mgm IV; 98 mg/d x 6d IM; 5 day taper Dexamethasone Saul et al 100 RCT prospective, (1981) US isolated HI - GCS double blind Time to admission: score [is less 6 hours than or equal 1. Placebo Duration: to] 7 non-responders 72 hours 100/100 ICPM 2 Bolus: 5 mg/ responders 7-10d kg/day Methylprednisolone (or dexamethasone equivalent) Braakman et al 161 RCT prospective, (1983) UK GCS score [is double blind Time to treat: 6 hours less than or Duration: 10d equal to] 8 1. Placebo 161/161 CT some ICP 2. 100 mg/d x4d, 16 mg/d IM/IV x 3d, 12, 8, 4 mgm/d IV/IM, then DC Dexamethasone Giannotta S1 et al 88 RCT prospective, (1984) US GCS score [is double blind Time to Rx: 6 hours less than or Duration: 11 d equal to] 8 1. 30mg/k Q6h x 2, 250 mg Q6 x 8, 8 d taper 2. 1.5 mg/k Q6h x2, 25 mg Q6 x 8, 8 d taper 3. placebo Methylprednisolone Dearden et al 130 RCT prospective, (1986) UK moderate and double blind Time to treat: 8 hours severe HI Duration Rx: 5 d 20% multiple 1. Placebo trauma 2. 50 mg IV boluls; 100 mg/d x 3d; 2 d tape (50, 25 mg) Dexamethasone Author (y) Outcome measures Findings Gudeman et al 3 mth GOS NS GOS (1979) US ICP NS ICP Time to treat: 6 hours Volume pressure NS VPR Duration Rx: unspec response (VPR) 85% hyperglycemia CT 72 hours S GI bleeding (50% adverse drug require transfusion) effects Cooper et al 6 mth GOS NS 6 month mortality (1979) US (dichotomized) NS GOS Time to treat: 6 hours adverse drug NS infection rate Duration Rx: 11d effects NS ICP GCS days NS increased 1,3,7,14 complications in high dose Saul et al 72 hour response NS GOS (dichotomous) (1981) US to steroids NS # responders Time to admission: GCS Early responders may 6 hours GOS benefit from steroid Duration: ICP, NS non-responders 72 hours responders 7-10d Braakman et al 1 and 6 month NS survival (1983) UK GOS ICP not analyzed Time to treat: 6 hours NS increase in Duration: 10d pulmonary infections treatment group Giannotta Sl et al ICP NS GOS (1984) US GOS 6 months NS adverse effects Time to Rx: 6 hours adverse drug High dose MP < 40 y Duration: 11 d effects had significantly lower mortality (6% vs. 43% p<.05) Dearden et al 6 month GOS NS GOS (1986) UK ICP NS ICP % peak Time to treat: 8 hours adverse drug sustained ICP Duration Rx: 5 d effects S hyperglycemia - steroid gr; NS ICU LOS S Steroid with ICP peak > 30 mm Hg worse; excluded protocol violators from analysis
NB: CT - computerized tomography - brains; Glasgow coma scale; GOS - Glasgow outcome scale; HI head injury; ICP - intracranial pressure; ICU - intensive care unit; ICPM intracranial pressure monitoring; IV intravenous route of administration; IM intramuscular route of drug administration; NS statistically nonsignificant; S statistically significant RCT randomized clinical trial
Studies of the past decade were RCT, a design recognized as the gold standard for testing efficacy of a new treatment or application. It has been estimated that minimal sample size required to demonstrate efficacy was 300;[29,39] unfortunately, none of the studies is large enough to reliably detect small differences and answer research questions convincingly. Researchers tended to operationalize injury severity similarly as GCS [is less than or equal to] 8. However, the samples were heterogeneous. One group included 20% with moderate injury; others mixed persons with isolated focal and diffuse injuries, hypotension and multisystem trauma. Others included children[6,18,30] who tend to have different pathology and outcomes. Within some individual studies, comparison and treatment groups differed prior to interventions. For example, Dearden noted that 22% of the treatment group had GCS scores [is greater than or equal to] 8 versus 13% his placebo group. Potentially, this bias in favor of the healthier treated group, rather than the intervention, may have accounted for the findings. In another instance, a retrospective comparison group differed from the treatment group in the number of mass lesions, and patients with abnormal motor responses and brainstem dysfunction. These two characteristics carry significant negative prognostic implications and may have had an effect on outcome independent of treatment. Finally, one group excluded diabetics and those with peptic ulcer disease, clearly biasing their report of adverse drug (steroid) effects.
There was weak control of extraneous variables. Supportive interventions were not standardized. For example, in a study involving two centers, the authors reported significant differences between treatment groups in the number of patients mechanically ventilated and those with ICP monitoring. Current guidelines may be valuable in reducing practice variation, greater control of confounding variables will strengthen confidence in RCT findings. The intervention (posology), time to admission and time to treat varied across studies. A therapeutic window was never defined for MP or dexamethasone. We know today that spinal tissue uptake falls rapidly with passage of time due to cellular loss or deteriorating perfusion, and early therapy appears to be critical to drug efficacy. If brain tissue behaves similarly, it is possible that many if not most patients got "too little, too late." In light of some animal evidence that suggests that 30 mg/kg MP or 60 mg/kg prednisolone administered within 5 minutes of injury improved outcome after nervous system injury, few trials even approached effective dosing. Gudeman and Giannotta came close but subjects were treated over 6 hours after injury.[29,30] Assumptions of antiinflammatory potency tied to the traditional glucocorticoid profile are likely not reliable; studies using "comparable" MP or dexamethasone dosages may have lacked true equivalency. Attainment of therapeutic blood - to say nothing of tissue - levels was never validated; this may have been of greatest importance with intramuscular dosing.[6,71,29] Indeed Giannotta suggested that Q 3 hourly dosing may have been more appropriate than the 6 hourly schedule used because of tissue half-life. Cumulative doses and treatment duration varied widely. In one instance, researchers treated only improving subjects beyond three days (responders improved to a GCS score [is greater than] 8 within 72 hours of admission). Seventeen subjects were treated and died and were also excluded from analysis. Failure of "intention to treat" analysis skewed results as well.[10,63]
The Glasgow Outcome Scale, with five scale steps (dead, persistent vegetative state, severe or moderate disability and good recovery) was widely used across studies, and facilitated communication about outcomes and monitoring as intended by its designers. Unfortunately, in many instances, it was collapsed to yield verdicts of dead or alive and lost the sensitivity to illuminate small but consequential treatment effects. Statistical power also suffered from the paucity of functional and morphological endpoints.[10,62] Several groups reported steroid adverse effects. Once again, evidence conflicted, likely reflecting differences in subjects and their co-morbidities, drug, dose and duration of treatment, as well as variability in gastroprotection. Retrospectively defined endpoints, such as "pulmonary infection" are difficult to meaningfully interpret.
In summary, the methodological weaknesses reviewed above were not bad science, but rather reflected the "black box" nature of TBI. There is contradictory evidence for steroid efficacy in improving morbidity and mortality after severe TBI. Given the small sample sizes, heterogeneity of injuries and methodological limitations, the non-significance might simply reflect inability to detect difference, incorrectly suggesting no effect. Although findings generally failed to reach statistical significance; the value of subgroup analysis was supported in some cases. Cooper noted the importance of injury type (focal versus diffuse) and concluded that some injuries were beyond treatment by virtue of the severity of the primary injury; Dearden found that steroid treated patients wih ICP spikes greater than 30 mm Hg had a significantly poorer outcome than non-treated individuals Saul et al noted that early responders who received steroids had better, and steroid treated non-responders had worse outcomes than controls. Prophetically, one researcher noted that "human head injury is a more complex process than that produced in most experimental models."[17,p.314] In today's light, not only is that accurate; but conclusions of "no benefit" may have been premature.
Of the myriad of adverse effects associated with glucocorticoid use, steroid treated survivors of TBI must be monitored carefully for vascular, infectious and metabolic complications and screened periodically for drug-drug interactions.
Lesions in the parasympathetic-vagal arc producing gastric acid hypersecretion and erosion constitute the neurogenic basis for Cushing's or stress ulcers after TBI. The incidence of erosion, with progression to significant hemorrhage in acutely ill patients with intracranial disease is approximately 17% with mortality as high as 50%. There is a dose related risk for gastrointestinal hemorrhage and perforation and steroid use. Transfusion-requiring gastric hemorrhage is a common occurrence in steroid treated TBI survivors. Peptic ulcer disease hepatorenal dysfunction, hypoalbuminemia, cumulative dose ([is greater than] 1G prednisone/equivalent) or therapy beyond 30 days and concurrent use of non-steroidal antiinflammatory agents amplify risk. Relative to other risk factors, the contribution of dexamethasone is believed to be low.[46,54]
Patients should be routinely monitored for guiaic positive stool and blood in nasogastric effluent; hematocrit should be checked regularly. Enteral nutrition may reduce risk. There is no consensus recommendation favouring prophylaxis; some suggest it for individuals with more than 2 risk factors. However, if bleeding is detected, aggressive acid suppression is necessary. The effect of drug selection on development of late onset pneumonia is unclear at this time.[46,61]
Steroid usage increases infection risk by 1.5 over controls; the prominent defect is inability to localize the site of infection, with skin and lungs particularly prone. Effects tend to be marked at doses greater than 700 mg prednisone or equivalent. Pulmonary infections were commonly observed in steroid treated TBI patients. Clinical suspicion, daily if not more frequent wound inspection and clinical respiratory exam, supplemented by review of hematologic indices are mandated in steroid treated patients. The absence of fever and leukocytosis should not allay clinical suspicion.[14,54]
Hyperglycemia normally occurs within minutes of severe TBI as part of the sympatho-adrenal stress response[3,43] and may be a proxy indicator for injury severity in humans,[11,48] Particularly in the setting of ischemia, it aggravates injury by driving anaerobic metabolism, enhancing increased glycolysation of membrane molecules, IC acidification, free radical formation, blood flow alterations and other effects,[11,43,48,65] Glucocorticosteroid use after injury amplifies the metabolic response, notably increased gluconeogenesis; proteolysis and lipolysis for gluconeogenesis, peripheral insulin resistance, hyperglycemia and subsequent osmotic diuresis.[3,14] In one study, steroid-treated survivors of TBI manifested hyperglycemia, lipolytic effects, and increases in muscle proteolysis significantly increased over controls. The net effects of TBI and increased, steroid-associated metabolic activity were significant protein wasting. The authors advised against steroid use in TBI since metabolic depletion is negatively correlated with outcome.
Frequent, scheduled blood glucose monitoring (four times daily) and avoidance of hyperglycemia may minimize secondary injury. However, definition of euglycemia in TBI is unknown and the value of tight glycemic control is not established. Preliminary work suggests that minimal reductions in level can affect outcome. Glucose free intravenous fluids have been proposed, especially during first few days when ischemic episodes may be more common; minimizing carbohydrate hyperalimentation may also be beneficial in this population. Early and aggressive nutritional support and consideration of paralytics in hypertonic patients may reduce catabolism.[3,11]
Daily review of the medication profile may reveal drug interactions. Barbiturates, which may be used for coma induction (pentobarbital) and antiepileptic drugs (phenobarbital, phenytoin, fosphenytoin and carbamazepine) may all increase steroid degradation, necessitating as much as double the original dose. Dexamethasone may antagonize neuromuscular blockade, such that the patient may require an elevated dose of paralytic. This interaction has been documented with atracuronium, mivacurium, pavulon and vecuronium. Concurrent use has also been linked to increased risk of myopathy.
Blunt trauma to the brain triggers biochemical/molecular cascades that ultimately cripple ionic and osmotic homeostasis, cellular energetics, membrane integrity and blood flow. How and by what means autocatalytic amplification occurs, and whether or not it is equally important across injury type and severity is not known; how they progress over time is not yet clear. An etiological role in the genesis of cerebral edema is unclear. Of steroids and TBI - Today, we have a molecule (steroid) in search of a mechanism, and a mechanism in search of a model. Strong evidence of a steroid-sensitive mechanism is lacking. Management guidelines concluded that steroids conferred no benefit and the recent Brain Trauma Foundation standard does "not recommended [steroids] for improving outcome or reducing ICP in patients with severe head injury"(p.15). The absence of statistically significant evidence of benefit and well known risks justify the position. But the glucocorticosteroids ought to be set aside, not discarded. Evidence suggestive of subgroup benefit and absence of any treatment for severe TBI will probably perpetuate their clinical use.
Successful treatment of TBI will only arise from full elucidation of the biology of events. The National Iinstitutes of Health-funded Multicenter Acute Spinal Cord Injury Study, a collaborative research model which is facilitating basic science advances and phase I testing across 10 spinal cord injury centers is underway and can serve as a model for head injury. Treatment effects must be interpreted in light of cross-species differences, anesthetic and injury models used, site of injury, outcome measures and the duration of follow-up. Nockels and Young admonition regarding pharmacotherapy of SCI is worth attending to in TBI:
"Researchers can no longer simply show that a drug is effective ... Laboratory studies now must define the optimal dose, duration, and timing of therapy for injuries of different severity. In order to be considered for a clinical trial, a drug must now be compared against and in combination with drugs that have been shown to be effective. The changes portend slower development of better drug therapies ..."(p. S212).
Investigators will contend with the likelihood of smaller differences across treatment arms and subsequent necessity for a greater number of subjects. Separating out the effects of different drugs used in combination protocols will certainly complicate analysis.[58,74] Given the almost infinite number of possible permutations and combinations of drugs to be tested and the complexity of that testing, animal analogues of any treatment arm offered to human subjects ought to be available. In the absence of compelling preclinical evidence of safety and efficacy, human trials should not proceed.[21,22]
Neural injury will most likely be amenable to treatment with a "drug cocktail;" it is inconceivable that a single agent could effectively block the events of secondary injury.[22,74] Future trials must be mechanism driven or targeted. Specific pathophysiological processes should be treated after the agent(s) have been shown to be effective in animals.[7,12,21] The plausibility of optimal therapy extending over time, with different agents with narrow and wide therapeutic windows side by side targeting different pathological cascades concurrently or in sequence, is high. As understanding unfolds, steroids may have a role in modulating brain inflammation by its cytokine actions; LP may represent a target for steroid use as it did in spinal cord injury. They may have some place in combination trials to come; and clinical utility in moderate and mild injury - where mechanisms of injury and outcomes are even harder to study, and preliminary work suggests heterogeneity at least within the moderately injured group. However, limitations in pharmacokinetics makes combination therapy premature.
Clinical consortia in US (American Brain Injury Consortium - ABIC) and Europe (European Brain Injury Consortium - EBIC), linking academic and pharmaceutical companies, are sharing TBI protocols and compounds. However, in light of the complexity and heterogeneity or primary and secondary injury, an even more sophisticated approach is indicated. Once again, a model from spinal cord injury exists as exemplar. The International Spinal Research Trust, founded in 1980, recently articulated a detailed and focused strategy to guide funding and coordinate researches in SCI. TBI researchers need the same coherence to drive progress, and they need it soon. Growing understanding of trauma-initiated changes in the brain ought to result in successful efforts - that may or may not include steroids - to prevent or abort damaging secondary injury processes. The black box of head injury is just now being opened.
The author acknowledges the assistance of Dr. Klein, Loyola University of Chicago, and Teri Vega-Stromberg, in the preparation of this manuscript.
[1.] Barie PS, Ghajar JBG, Firlik AD et al: Contribution of increased cerebral blood volume to posttraumatic intracranial hypertension. J Trauma 1993; 35(1):88-95.
[2.] Barzo P, Maramou A, Fatouros P et al: Magnetic resonance imaging monitored acute blood-brain barrier changes in experimental traumatic brain injury. J Neurosurg 1996; 85:1113-1121.
[3.] Bessey PQ, Downey RS, Monafo WW: Metabolic response to injury and critical ilness. Pages 325 - 335 in: Critical Care, 3rd ed, Civetta JM, Taylor RW, Kirby RR (editors). Lippincott-Raven, 1997.
[4.] Biros M: Head trauma. Pages 416-447 in: Emergency Medicine, Vol 1, 4th ed, Rosen P (editor-in-chief). Mosby, 1998.
[5.] Bouma GJ, Muizelaar P, Choi SC et al: Cerebral circulation and metabolism after severe traumatic brain injury: the elusive role of ischemia. J Neurosurg 1991; 75(11):685-693.
[6.] Braakman R, Schouten HJ, Dishoeck MB, Minderhoud JM: Megadose steroids in severe head injury. J Neurosurg 1983; 58:326-330.
[7.] Bullock R: Opportunities for neuroprotective drags in clinical management of head injury. J Emerg Med 1993;11:23-30.
[8.] Bullock R, Teasdale G: Head injuries - II. Brit Med J 1990; 300:1576-1579.
[9.] Bullock R, Zauner A, Woodward J et al: Factors affecting excitatory amino acid release following severe human head injury. J Neurosurgery 1998;89:507-518.
[10.] Buring J E, Hennekens CH: Methodology of intervention trials in individuals. Pages 585-595 in: Oxford Textbook of Public Health, Vol 2, 3rd ed, Detels R, Holland WW, McEwen J, Ammen GS (editors). Oxford University Press, 1997.
[11.] Cherian L, Goodman JC, Robertson CS: Hyperglycemia increases brain injury caused by secondary ischemia after cortical impact injury in rats. Crit Care Med 1997; 25(8):1378-1383.
[12.] Chesnut RM: Medical management of severe head injury: Present and future. New Horizons 1995; 3(3):581-593.
[13.] Chesnut RM: Secondary brain insults after head injury. Clinical perspectives. New Horizons 1995; 3(3):366-375.
[14.] Chin R, Eagerton DC, Salem M: Corticosteroids. Pages 715-740 in: The Pharmacologic Approach to the Critically Ill Patient, 3rd ed, Chernow B (editor). Williams & Wilkins, 1994.
[15.] Choi DW: Excitotoxic cell death. J Neurobiol 1992; 23(9): 1261-1276.
[16.] Complicated Head Injuries. Retrieved 6 November, 1998 from World Wide Web. Http://www.usc.edu/hsc/neurosurgery/St ... Lecture/Complicated/complic.html.
[17.] Cooper PR, Moody S, Clark K et al: Dexamethasone and severe head injury. J Neurosurg 1979; 51:307-316.
[18.] Dearden NM, Gibson JS, McDowell DG, Gibson RM, Cameron MM: Effect of high dose dexamethasone on outcome from severe head injury. J Neurosurg 1986; 64(1):81-88.
[19.] Deutschman CS, Konstantinides FN, Raup S, Cerra FB: Physiological and metabolic response to isolated closed head injury. Part 2. Effects of steroids on metabolism. J Neurosurg 1987; 66(3):388-395.
[20.] Doberstein CE, Hovda DA, Becker DP: Clinical considerations in the reduction of secondary brain injury. Ann Emerg Med 1993; 22(6):993-997.
[21.] Doppenberg EM, Bullock R: Clinical neuro-protection trials in severe traumatic brain injury: Lessons from previous studies. J Neurotrauma 1997; 14(2):71-80.
[22.] Faden AI: Pharmacologic treatment of acute traumatic brain injury. JAMA 1996; 276(7): 569-570.
[23.] Faden AI: Experimental neurobiology of central nervous system trauma. Crit Rev Neurobiol 1993; 7(3/4):175-186.
[24.] Fessler RD, Diaz FG: The management of cerebral perfusion pressure and intracranial pressure after severe head injury. Ann Emerg Med 1993; 22(6):998-1003.
[25.] Feuerstein CZ, Wang X, Barone FC: Inflammatory gene expression in cerebral ischemia and trauma. NY Acad Sci 1997; 825:179-193.
[26.] Fishman RA: Brain edema and disorders of intracranial pressure. Pages 302-310 in: Merritt's Textbook of Neurology, 9th ed, LP Rowland (editor). Williams & Wilkins, 1995.
[27.] Ghajar J, Hariri RJ, Narayan RK et al: Survey of critical care management of comatose, head injured patients in the United States. Crit Care Med 1995; 23(3):560-567.
[28.] Ghirnikar RS, Lee YL, Eng LF: Inflammation in traumatic brain injury: Role of cytokines and chemokines. Neurochem Res 1998; 23(3):329-340.
[29.] Giannotta SL, Weiss MH, Apuzzo MLJ, Martin E: High dose glucocorticoids in the management of severe head injury. Neurosurgery 1984; 15(4):497-501.
[30.] Gudeman SK, Miller JD, Becker DP: Failure of high dose steroid therapy to influence intracranial pressure in patients with severe head injury. J Neurosurg 1979; 51:301-306.
[31.] Hall ED: High-dose glucocorticoid treatment improves neurological in head injured mice. J Neurosurg 1985; 62(6):882-887.
[32.] Hall ED: Lipid antioxidants in acute central nervous system injury. Ann Emerg Med 1993; 22(6):1022-1027.
[33.] Hall ED: The role of oxygen radicals in traumatic injury: Clinical implications. J Emerg Med 1993; 11:31-36.
[34.] Harper GP, Banyard PJ, Sharpe PC: The international spinal research trust's strategic approach to the development of treatments for the repair of spinal cord injury. Spinal Cord 1996; 34:449-459.
[35.] Hartl R, Medary MB, Ruge M et al: Early white blood cell dynamics after traumatic brain injury: Effects on the cerebral microcirculation. J Cerebr Blood Flow Metab 1997; 17:1210-1220.
[36.] Integration of Neurosurgical Guidelines into the Management of Acute Brain Injury. Retrieved 9/14/98 from the World Wide Web. http://www.usc.edu/hsc/neurosurgery/ neurotrauma/guidelines.html.
[37.] Jaggi JL, Obrist WD, Gennarelli TA, Langfitt TW: Relationship of early cerebral blood flow and metabolism to outcome in acute head injury. J Neurosurg 1990: 72(2):176-182.
[38.] Jennett B, Bond M: Assessment of outcome after severe brain damage. Lancet 1975; 1:480-484.
[39.] Jennett B, Teasdale G: Management of head injuries. FA Davis, 1981.
[40.] Karakucuk EI, Pasaoglu H, Pasaoglu A, Oktem S: Endogenous neuropeptides in patients iwth acute traumatic head injury II: Changes in the levels of cerebrospinal fluid substance P, serotonin and lipid peroxidation products in patients with head trauma. Neuropeptides 1997; 31(3):259-263.
[41.] Kelly DF: Steroids in head injury. New Horizons 1995; 3(3): 453-455.
[42.] Kossman T, Hans VHJ, Imhof H-G et al: Intrathecal and serum interleukin-6 and the acute-phase response in patients with severe traumatic brain injuries. Shock 1995; 4(5):311-317.
[43.] Lam JM, Winn R, Cullen BF, Sundling N: Hyperglycemia and neurological outcome in patients with head injury. J Neurosurg 1991; 75(3):545-551.
[44.] Leslie JB, Watkins WD: Eicosanoids in the central nervous system. J Neurosurg 1985; 63(11): 659-666.
[45.] Lobato RD, Sarabia R, Cordobes F et al: Posttraumatic cerebral hemisphere swelling. Analysis of 55 cases studied with computerized tomography. J Neurosurg 1988; 68(3):417-423.
[46.] Lu WY, Rhoney DH, Boling WB et al: A review of stress ulcer prophylaxis in the neurosurgical intensive care unit. Neurosurg 1997; 41(2):416-426.
[47.] Luer MS, Rhoney DH, Hughes M, Hatton J: New pharmacologic strategies for acute neuronal injury. Pharmacotherapy 1996; 16(5):830-848.
[48.] Margulies DR, Hiatt JR, Vinson D Jr, Shabot MM: Relationship of hyperglycemia and severity of illness to neurologic outcome in head injury patients. Am Surgeon 1994; 60(6), 387-390.
[49.] Matta B, Menon D: Severe head injury in the United Kingdom and Ireland: A survey of practice and implications for management. Crit Care Med 1996; 24(10):1743-1748.
[50.] McIntosh TK, Smith DH, Meaney DF et al: Neuropathological sequelae of traumatic brain Injury: Relationship of neurochemical and biomechanical mechanisms. Laboratory Investigation 1996; 74(2):315-342.
[51.] McKeating EG, Andrews PJD, Mascia L: The relationship of soluble adhesion molecule concentrations in systemic and jugular venous serum to injury severity and outcome after traumatic brain injury. Anesthesia & Analgesia 1998; 86(4): 759-765.
[52.] McKeating EG, Andrews PJD, Signorini DF, Mascia L: Transcranial cytokine gradients in patients requiring intensive care after acute brain injury. Br J Anesthesia 1997; 78(5):520-523.
[53.] McQuire JC, Sutcliffe JC, Coats TJ: Early changes in middle cerebral artery blood flow velocity after head injury. J Neurosurg 1998; 89:526-532.
[54.] Micromedex Inc [database online]. Dexamethasone. Vol. 98. Exp. 11/30/98.
[55.] Molofsky WJ: Steroids and head trauma. Neurosurg 1984; 15(3):424-426.
[56.] Multicenter animal spinal cord injury study (MASCIS). Retrieved 11 Nov, 1998 from World Wide Web. http://www.med.nyu.edu/MASCIS.html.
[57.] Muizelaar JP: Cerebral ischemia-repersuion injury after severe head injury and its possible treatment with polyethyleneglycol-superoxide dismutase. Ann Emerg Med 1993; 22(6):1014-1021.
[58.] Nockels R, Young W: Pharmacologic strategies in the treatment of experimental spinal cord. J Neurotrauma 1992; 9 Suppl 1:S211-S217.
[59.] Nguyen T, Frank E, Trunkey D: Steroids in central nervous system injury. Adv Surgery 1996; 30: 53-69.
[60.] Povlishock JT, Jenkins LW: Are the pathological changes evoked by traumatic brain injury immediate and irreversible? Brain Path 1995; 5:415-426.
[61.] Prod'hom G, Leuenberger P, Koeffer J et al: Nosocomial pneumonia in mechanically ventilated patients receiving antacid, ranitidine, or sucralfate as prophylaxis for stress ulcer. A randomized controlled trial. Ann Intern Med 1994; 120(8):653-662.
[62.] Rosomoff HL, Kochanek PM, Clark R et al: Resuscitation from severe brain trauma. Crit Care Med 1996; 24(2 Suppl):S48-S56.
[63.] Saul TG, Ducker TB, Salcman M, Carro E: Steroids in severe head injury: A prospective randomized clinical trial. J Neurosurg 1981; 54(5): 596-600.
[64.] Shapira Y, Setton D, Artru AA, Shohami E: Blood-brain barrier permeability, cerebral edema, and neurologic function after closed head injury in rats. Anesthesia & Analgesia 1993; 77(1):141-148.
[65.] Siesjo BK: Basic mechanisms of traumatic brain damage. Ann Emerg Med 1993: 22(6):959-969.
[66.] Teasdale G, Jennett B: Assessment of coma and impaired consciousness. A practical scale. Lancet 1974; 2:81-84.
[67.] Tymianski M, Tator CH: Normal and abnormal calcium homeostatsis in neurons: A basis for the pathophysiology of traumatic and ischemic central nervous system injury. Neurosurg 1996; 38(6):1176-1195.
[68.] Valadka AB, Gopinath SP, Contant CF et al: Relationship of brain tissue [PO.sub.2] to outcome after severe head injury. Crit Care Med 1998; 26(9):1576-1581.
[69.] Veremakis C, Lindner DH: Central nervous system injury: Essential physiologic and therapeutic concerns. Pages 273 - 289 in: Critical Care, 3rd ed, Civetta JM, Taylor RW, Kirby RR (editors). Lippincott-Raven, 1997.
[70.] Wald SL: Advances in the early managemaent of patients with head injury. Surg Clin North Am 1995; 75(2): 225-242.
[71.] White BC, Krause CS: Brain injury and repair mechanisms: The potential for pharmacologic therapy in closed-head trauma. Ann Emerg Med 1993; 22(6):970-979.
[72.] Woodroofe MN: Cytokine production in the central nervous system. Neurology 1995; 45(6 Suppl 6):S6-S10.
[73.] Yoshino E, Yamaki T, Higuchi T et al: Acute brain edema in fatal brain injury:Analysis by dynamic CT scanning. J Neurosurg 1985; 63(6):830-839.
[74.] Young W: Neuroprotective therapy for brain and spinal cord injury. Pages 863-874 in: The Pharmacologic Approach to the Critically Ill Patient, Chernow B (editor). Williams & Wilkins, 1994.
[75.] Zink BJ: Traumatic brain injury. Emerg Med Clin North Am 1996;14(1):115-150.
Questions or comments about this article may be directed to: Milena Segatore, RN, MScN, CNRN, St. Joseph's Hospital, 5000 W. Chambers St., Milwaukee, WI 53210-1688. She is a neurology clinical specialist.3
|Printer friendly Cite/link Email Feedback|
|Publication:||Journal of Neuroscience Nursing|
|Date:||Aug 1, 1999|
|Previous Article:||Living a Nightmare: Family Experiences of Waiting Following Neurological Crisis.|
|Next Article:||Pain Management.|