Traumatic Brain Injury - Management in Africa

An African Perspective  
Pathophysiology of Brain Injury  
  Normal Homeostasis - Autoregulation  
  Intracranial Hypertension  
  Physiology of Injury  
Classification of Head Injury  
  Classification by Severity  
Classification by Pathology  
    Diffuse Injury  
    Focal Injury  
Initial Management
  The Neurological Examination
Brain Death
Imaging in Head Trauma
  Alternate Imaging Modalities
Managment of Brain Injury
  Invasive Monitoring
  Measures to Treat Intracranial Pressure
  Basic Neurosurgical Procedures in Trauma
    External Ventricular Drainage
    Exploratory/"Woodpecker" Surgery
    Craniotomy / Craniectomy
  General Measures
Guidelines for the Management of Head Injury
Management Algorithm

The brain has monumental functions, mediating consciousness as well as neural and hormonal control of the body. Brain injury thus has a special place amongst traumatic insults due to its uniquely intimate and profound impact on the person. Patients with brain injury are often a challenge to manage, requiring difficult decisions and an infrequently used skill-set for most physicians. This task is made even more difficult when one is practicing in an environment without resources such as CT scanners which have become instrumental to the management of such patients in neurosurgical centers over the past 30 years.
This review is designed to assist physicians managing traumatic brain injury (TBI) while working in environments with limited resources. We will present details on the current guidelines and management paradigms as they exist in the literature. Though craniotomies have been performed in Africa for centuries (1) we anticipate that the majority of our readers will have insufficient resources to provide the type of care described in the current literature. This review will therefore try to help practitioners to provide the highest level of care possible with the resources at their disposal.

An African Perspective
In Africa, the epidemiology of traumatic brain injury is very similar to that seen in the West, though this may be because Africa’s publishing neurosurgical centres are located in urban, industrialized centres. As in the West, young males are disproportionately affected and traffic accidents are a predominant mechanism of injury (2, 3). Furthermore, like North America, TBI is the most common cause of traumatic death in the young (2, 4-6) and age is one of the strongest predictors of outcome (2, 7).
Different from the West, however, is the fact that injured Africans face a median distance of 60 km to a hospital and a corresponding nine hour delay in reaching such care (4). Police, good samaritans or relatives are frequently responsible for transport to healthcare facilities (8). Though problems in neurotrauma care are not unique to Africa (6), efforts to improve transport, centralize trauma care (6), and improve trauma management within individual hospitals will almost certainly result in improved patient outcomes in Africa (9, 10). The development of neurosurgical ‘Centres of Excellence’ in African countries, partnered with similar centres in developed countries, has been proposed as a means of improving African neurotrauma care, as has the development of an Advanced Brain Life Support course which would supplement the Advanced Trauma Life Support course (6).

Pathophysiology of Brain Injury

Normal Homeostasis - Autoregulation:
Cerebral blood flow is normally maintained at a constant level thanks to autoregulatory processes which match flow to the metabolic needs of the brain (11, 12). There are numerous forms of autoregulation. Pressure autoregulation holds that between systolic blood pressures of 50 and 160 mmHg (higher in those with chronic hypertension), the diameter of cerebral arterioles is altered to maintain a constant blood supply to the brain (13). Viscosity autoregulation alters vessel diameter and thus blood flow in response to changes in viscosity (14). The cerebral vasculature is also highly responsive to metabolites allowing for metabolic autoregulation (15). Perhaps best understood, however, is the relationship between Pa O2, Pa CO2 and cerebral blood flow. Hypoxia is a potent vasodilator (16) and Pa CO2 levels have a similar and more profound effect on cerebrovascular tone (17).
Understanding autoregulation is essential in managing traumatic brain injury. Autoregulation may be lost following a head injury, thus making the brain prone to ischemia. Avoidance of hypotension is thus crucial as the ability to compensate is lost. Some forms of autoregulation are generally preserved, however, and are essential for treatment (18). Reactivity to CO2 explains the efficacy of hyperventilation in head injury management (19) as well as mannitol’s effects which are largely due to reduction in viscosity and resulting vasoconstriction (20). Cerebral blood volume can change up to 40cc’s as a result of these processes, effecting a 29-fold change in intracranial pressure (ICP) (11).

Intracranial Hypertension:
Intracranial hypertension is the end result of multiple intracranial processes that can be seen in trauma (21). The Kellie-Monro Doctrine describes that the volume within the skull is fixed, and that an exponential increase in pressure occurs secondary to processes such as hemorrhage or swelling which add to that volume. Such pressure elevation can impair blood flow to the brain. As a compensatory response, the body produces hypertension and bradycardia; combined with irregular breathing, these comprise Cushing’s triad (22) which is seen in its classic form only a third of the time. When treating head injured patients, it is important to follow the cerebral blood flow. This is not easily measured directly so clinically we use a surrogate, the cerebral perfusion pressure (CPP), which is calculated as follows: CPP=ICP-MAP where MAP is the mean arterial pressure, calculated as 2/3 the diastolic pressure + 1/3 the systolic pressure.

Physiology of Injury:
The damage to the brain that occurs at the time of initial injury is known as primary injury. Following this the brain is sensitive to additional insults which have become known as secondary injury (23). These are surprisingly common, seen in perhaps 90% of patients (24). Fortunately, however, many of these factors can be ameliorated by physicians. The following are important, frequent causes of secondary injury:
Cerebral Edema: Brain swelling, or edema, is very common following brain injury. It peaks several days post-insult and can cause focal deficits and elevated ICP. Though effective for brain tumor-associated edema, steroids have shown a 20% higher mortality rate (25) when given to the head injured, and should not be administered.
Hypotension: Hypotension, defined as one or more observation of a systolic blood pressure less than 90 mmHg, is associated with a two-fold increased risk of mortality (26). Hypotensive or under-resuscitated patients have been shown to have higher ICP because of compensatory cerebral vasodilation (27, 28). Hypotension must thus be avoided at all costs.
Hypoxemia: Unfortunately the brain is more sensitive to ischemia after trauma than in the normal state, and furthermore hypoxia results in compensatory cerebral vasodilation, increasing intracranial pressure (28). Hypoxia, defined as the observation of a single PaO2 less than 60 mmHg, is thus correlated with a much worse prognosis (26). Aggressive treatment is mandated.
Pyrexia: Metabolism increases 10-13% per degree centigrade and fever thus increases the needs of a brain already stretching to meets its metabolic requirements. It can also increase cerebral blood flow by as much as 97%, raising ICP (11).

Head Injury Classification
Head injury is classified in many ways: by the severity, resulting pathology (focal or diffuse injury) or mechanism of injury (blunt or penetrating).

Classification by Severity:
The Glasgow Coma Scale (29) (GCS, Table 1) is generally used to indicate the severity of a head injury. In general patients with a GCS of 14 or 15 are considered to have a mild injury, those with GCS 9-13 a moderate injury, and those with GCS 8 or less a severe head injury. The Head Injury Severity Scale (Table 2) is a slight variation on this system.

Classification by Pathology:
While head injury can be classified as focal or diffuse, the reality is that both generally exist to varying degrees. One must also remember as well that up to 20% of patients who present with a diffuse injury will develop a mass lesion, most commonly in the first day following the injury (30). In severe head injury, diffuse injury predominates but focal lesions carry a higher mortality rate (31).

Diffuse Injury:
The hallmark of diffuse brain injury is alteration of mental status out of proportion to CT findings. Rotational acceleration/deceleration forces are key, causing the cortex to move at a different speed than deeper brain structures, damaging axons. Concussion and diffuse axonal injury (DAI) exist at opposite ends of a gradient, but a dividing line has been established: traumatic coma of less than six hours is classified as concussion, while coma of longer duration is considered DAI.
Concussion: Concussion is defined as any alteration in mental status following head injury. Clinical hallmarks are confusion, amnesia, and loss of consciousness. CT is typically normal, however, MRI may detect lesions. Grading schemes for concussion have fallen out of favour (32), but the one perhaps most widely used is that of the American Academy of Neurology (Table 3). Full recovery from a concussion is not always seen (33).
Diffuse Axonal Injury: DAI is seen in 80-100% of those with fatal head injuries (34). The absence of a lucid interval is said to be a hallmark. While most lesions in this disorder are non-hemorrhagic and microscopic, punctate hemorrhages can be seen at grey-white matter interfaces, and in the upper pons, dorsolateral midbrain and corpus callosum. Severe DAI has a mortality rate as high as 50% (35). Interestingly, associated axon disruption is seen predominantly in a delayed fashion and preventing this is a subject of intensive research.

Focal Injury:
Scalp Laceration: Scalp lacerations can lead to significant hemorrhage if not managed appropriately. It must be remembered that managing circulatory compromise is a higher priority than neurological disability! Scalp lacerations are of special significance when they are associated with a skull fracture, (making it an open fracture), putting patients at risk for meningitis. Some scalp defects may require flaps for reconstruction (36).
Skull Fractures: The skull is a complex bony structure and can be affected by a myriad of fractures. Basal skull fractures should be suspected in the face of raccoon eyes, Battle’s sign, hemotympanum, otorrhea or rhinorrhea. The numerous important vascular and neurological structures that traverse the skull base can be compromised such as the optic nerve, and the seventh and eight cranial nerves. A basal skull fracture is a contraindication to the insertion of nasal devices (ie. nasogastic tubes) as there is a risk of intracranial placement through a bony defect. Blows with a small area of impact are generally responsible for depressed skull fractures (Figure 1a,b). Depressed skull fractures should generally be elevated if (11):

  1. the depth of the depressed fragments is equal to or greater than the width of the surrounding bone
  2. conservative management would result in a cosmetic deficit
  3. the fracture is open
  4. associated with a deficit from compression of underlying brain

Open skull fractures deserve special consideration. While many would prefer operative debridement and closure of such wounds within 6 hours of injury, the recently published surgical guidelines for head trauma are more conservative in their recommendations (37). They state that conservative management of open skull fractures is acceptable so long as the above indications for surgical management of a depressed skull fractures are absent, and the following additional criteria are met:

  1. there is no evidence of dural penetration
  2. absence of a surgical intracranial hematoma
  3. the fracture does not involve the frontal sinus
  4. there is no evidence of infection
  5. no pneumocephalus is apparent
  6. absence of gross wound contamination

They recommend that all individuals with open skull fractures should be treated prophylactically with antibiotics, but this too is controversial and different from the practice of the authors, who prefer use of antibiotics only in the face of established infection.

Intracranial Hematoma
Intracranial hematomas can be life-threatening and may require surgical intervention. They are seen more frequently as the severity of injury increases, and it must be remembered that they can develop in a delayed fashion.
Cerebral Contusion: Cerebral contusions make up 45% of primary intra-axial traumatic lesions and are present in nearly half of moderate and severe closed head injuries (31). They occur when the brain impacts against bony prominences or dural folds (Figure 2a,b). Coup contusions lie below the region of cranial impact. Contre-coup contusions are seen at a site distant and often (but not necessarily) contralateral to the impact can be seen. A third of these progress in size and they are responsible for less mass effect than would be anticipated, though associated swelling can mandate resection or decompression.
Intracerebral hematomas: Intracerebral hematomas account for 10% of all post-traumatic intracranial hematomas (31). Intracerebral hematomas may form from contusions, but are classified as the former if blood makes up greater than 2/3 of the lesion or if there is a well-defined margin. Delayed intracerebral hematomas (Figure 3b) are not seen on the initial scan, but appear within 24-48h.
Epidural Hematoma: Epidural hematomas (Figure 4a,b) occur in about 6% of patients with severe head injury (31) and are more common in younger patients. They often arise from skull fractures over the pterion which disrupt the middle meningeal artery. The clinical course classically involves a loss of consciousness followed by a lucid interval, then a subsequent decline within 6 hours of injury. A lucid interval occurs in 10-50% of cases and is a good prognostic sign; mortality is doubled when it is not seen (38). Radiographically these hemorrhages tend to be lens-shaped and they do not cross suture lines but can cross the falx cerebri and tentorium cerebelli, as well as the dural sinuses. Fortunately 70% of patients requiring surgery for epidural hematomas have a good outcome and with rapid treatment mortality may be as low as 5% (39).
Subdural Hematoma: Acute subdural hematomas are the most common post-traumatic focal intracranial lesion (Figure 3a) (31). They are associated with a more significant impact force than epidural hematomas and DAI frequently co-exists. Subdural hematomas classically arise from disruption of bridging veins as they cross from the cerebral cortex to the venous sinuses. Cerebral atrophy puts bridging veins on a stretch, and these hemorrhages are thus seen more commonly in older patients and alcoholics. They can also arise from cortical lacerations. They tend to spread diffusely over the hemisphere in a crescent shape. Unlike epidural hematomas they can cross sutures but not dural attachments, and they are sometimes seen along the falx and tentorium. Patients with acute subdural hematomas fare poorly compared to those with epidural hematomas, with an associated mortality rate of 50-90% (40). In children subdural hematomas must raise suspicion of physical abuse.
Subarachnoid Hemorrhage: The prevalence of traumatic subarachnoid hemorrhage (SAH) in severely head injured patients is probably greater than 40% (41) and, in fact, trauma is the most frequent cause of subarachnoid hemorrhage (Figure 5). It may be important to rule out aneurysmal hemorrhage in these cases. Here history is critical, and furthermore traumatic SAH is generally seen over the cerebral convexity as opposed to the basal cisterns as is typically seen with aneurysms.
Intraventricular Hemorrhage: Post-traumatic intraventricular hemorrhage is seen in about 25% of severe head injuries (11), and is most frequently an extension of intracerebral hematoma in periventricular areas. Temporary ventriculostomy is often indicated.
Penetrating Head Injury: Penetrating head injury is frequently seen as a result of knife or gunshot wounds, with the latter being classified as ‘missile’ injuries (Figure 6a,b). Because these wounds are open, they are inherently contaminated. The role for debridement is controversial. We are aware that many textbooks advocate aggressive debridement of debris and devitalized tissue for all patients. Furthermore, some have advocated radical or repeated surgery to remove as much foreign body as possible (with resulting destruction of much functional brain tissue) in hopes of preventing infection. This strategy, as mandated by the United States’ military during the Vietnam War, has been associated with significant morbidity and mortality (42). The penetrating head injury guidelines currently recommend a more conservative approach (42). An extensive review of published data on this topic led to the recommendation that those patients with ‘significant’ intracranial pathology should undergo debridement in addition to the other brain surgery they require. Level III evidence indicates, however, that those without mass effect may not benefit from debridement of the missile tract, as patients that are not debrided “are not measurably worse” than those who are (42).

Initial Management
The airway is the highest management priority in neurotrauma, as in all trauma. Patients with a GCS score of 8 or less (those not localizing) require intubation. Airway protection helps to prevent secondary injury from hypoxia and hypercarbia. Cervical spine precautions are paramount during airway interventions. Breathing, circulation and the stabilization of vital signs must be the next priority. A hypoperfused brain does not function normally, so a Glasgow Coma Scale score has little value unless obtained post-resuscitation.
The neurological examination may also be obscured iatrogenically, as when patients are sedated by physicians who inappropriately feel that movement is exacerbating their head injury. Unfortunately this often occurs at the time of first assessment when monumental decisions must be made: whether a patient should go to surgery, or is beyond help. These agents are often necessary, especially to facilitate imaging; short acting, reversible agents (such as short-acting benzodiazepines) should be used whenever possible.
The Neurological Examination:
The top priorities in the neurological examination are: determining if there is evidence for life-threatening intracranial pathology and assessing the patient’s baseline level of neurological function such that change can be detected. The two ‘neurological vital signs’- the pupillary examination and the GCS – are most important for this purpose and should be assessed first.
The initial pupillary exam assesses only pupil size and reactivity to light. A single dilated pupil (at least >1mm larger than the other, as differentiated from anisocoria which is normal) that does not react to light suggests ipsilateral uncal herniation and mandates aggressive measures to reduce ICP as such a patient is potentially ‘salvageable.’ Remember, however that traumatic mydriasis is important to consider as a cause; it is often associated with hyphema. Bilateral fixed and dilated pupils suggest a very poor prognosis and a patient who may not benefit from intervention.
The GCS is then scored. All physicians should be able to score a GCS rapidly and accurately. A declining exam, defined as a 2 or more point drop in GCS, is also an indication for initiation of therapies to treat ICP.
In a patient who is completely unresponsive, the next priority is assessing brainstem integrity by means of the corneal and gag reflexes. Vestibulocaloric testing also helps to assess the integrity of the brainstem, following confirmation of an intact tympanic membrane. Doll’s eyes testing is generally not used in trauma patients as it involves manipulation of the cervical spine.
The head must then be examined in detail. Hair removal may be needed to facilitate proper examination. If a CT scanner is not available, wounds should inspected to determine if a skull fracture underlies them or if there is egress of CSF. However these wounds should be probed with extreme caution if at all. Signs consistent with basal skull fracture should be sought (raccoon eyes, Battle’s sign, hemotympanum, otorrhea or rhinorrhea). To determine the presence of cerebrospinal fluid (CSF) within bloody fluid, one can look for a halo sign when a drop of fluid is placed on tissue paper.
Attention should then turn to the face for a detailed examination. Where possible, a full cranial nerve examination should be performed. If the eyes are markedly deviated to one side it could indicate ongoing seizure involving the frontal eye fields contralateral to the direction of gaze. Tongue-biting or incontinence may also suggest seizure. Impaired light reaction (relative afferent papillary defect) can suggest optic nerve compromise which may need treatment. The absence of papilledema should not be reassuring in trauma patients as they rarely develop this finding. Though impractical to assess initially, the olfactory nerve is the most frequently injured cranial nerve and is very debilitating because of its role in taste. An abducens palsy may represent a ‘false-localizing’ sign. Because of its long intra-cranial course, elevated ICP can compromise the function of this nerve. The side involved may not correspond to that of intracranial pathology so physicians should not be misled by this finding.
The neck often receives little attention in a trauma. Clearing the cervical spine generally involves clinical exam in a lucid patient and often imaging. This is not important initially and it is rarely feasible because of impaired mental status. As such, the cervical spine must be protected until this can be completed.
Brain Death: Brain death is considered legally equivalent to cardiac death. Local protocols generally define how brain death should be diagnosed. These involve demonstrating an absence of brain function on exam in addition to ensuring that conditions that may depress consciousness such as medications, metabolic disturbances, and hypothermia are absent. Absent blood flow to the brain on imaging and an isoelectric EEG may be adjuncts. More than one physician must typically participate in a declaration of brain death.

Imaging in Head Trauma
Computed Tomography:
CT scanning is by far the most important imaging modality in head trauma patients. All patients with moderate or severe head injuries should be scanned as soon as possible with repeat imaging at 24 hours, if a CT scanner is available. The Canadian CT Head Rule helps to determine which patients with minor head injury should be imaged (43). Though intended for CT scanning these rules could be extrapolated for imaging with whatever modalities are available. Magnetic resonance imaging has little utility in acute head injuries, though it may have a role later in diagnosing DAI.

Alternate Imaging Modalities:
Generations of neurosurgeons practiced without the luxury of CT scanning, and the following are some techniques which they utilized (44). Some of the imaging techniques can be very time-consuming, especially in inexperienced hands; in many situations it may be prudent to proceed with surgery before imaging as a significant delay could prove fatal.
Skull x-rays: Skull x-rays can detect skull fractures, gross pneumocephalus, and may be able to detect midline shift based on displacement of a calcified pineal gland (45). The presence of a skull fracture mandates close observation as it is associated with a 20 fold increased risk of hematoma in an unconscious patient, and a 400 fold increase in those that are conscious (46). Guidelines have been devised to help determine when skull x-rays should be employed, and following them can help to reduce costs (3).
Echo-Encephalography: Though highly operator-dependent, ultrasound can be used to image the cranium. Displacement of the midline greater than 2mm detected in this fashion is considered pathological, and indicates the side of the lesion (44). Absence of shift does not rule out a hematoma: lesions may be bilateral, located subfrontal or in the posterior fossa. Generalized brain edema also remains a possibility. This technique has been demonstrated to be 95.3% accurate in diagnosing midline shift (47).
Cerebral Angiography: A mainstay in the pre-CT era (48), this imaging technique can assist with the diagnosis of most supra-tentorial hematomas. Displacement of vessels from the inner table of the skull, or shift of the peri-callosal or callosomarginal arteries can assist with localization (Figure7). These arteries can be used to diagnose hydrocephalus (as may occur with a posterior fossa hematoma) on a lateral film when elevated with a broadened curvature. The literature, as well as the authors, have found that the requisite carotid puncture is associated with a low rate of complications.
Ventriculography: Instillation of contrast via an external ventricular drain (placement of which is to be described shortly) has utility in demonstrating ventricular compression that could indicate a deeply-located hematoma or hydrocephalus secondary to compression of the cerebral aqueduct. This procedure was considered inferior to angiography and was rarely performed in the pre-CT era.
Air-Encephalography: Though performing this procedure is discouraged by Northfield (44), in the absence of contrast one can inject a few cc’s of air intraventricularly. This is obviously relatively contraindicated in a patient with elevated ICP. The injected air can be imaged readily by x-ray, providing similar information to ventriculography. An important difference, however, comes in understanding that air will rise to the top of any fluid column, making positioning key. Generally supine positioning is used for such imaging.

Management of Brain Injury
In those with mild or minor head injury, decision-making centers around deciding if a patient should be admitted for observation or discharged home, based on their risk for a life-threatening intracranial lesion. Such lesions are seen in about 2% of these patients (11) and generally present in the first 12-24 hours (49). We recommend the Canadian CT Head Rule for assistance in determining which patients should be imaged (43) (Table 4). All patients with an anomaly on CT or x-ray should be admitted for 24h of observation with a CT repeated prior to discharge if possible. Those meeting criteria for observation at home (Table 5) may be discharged if the attending physician feels it is prudent to do so. Patients with moderate and severe head injuries must be admitted and are generally monitored in an intensive care setting.
General Measures: Avoid hypotonic intravenous fluids as these agents may exacerbate cerebral edema. Prophylaxis for gastric stress ulceration – Cushing’s ulcers – should generally be prescribed (50). Despite numerous studies, the optimal agent for this purpose remains controversial; proton pump inhibitors, H2 antagonist, and sucralfate are all acceptable based on current literature (51). Patients with brain injury are at high risk for coagulopathy such as disseminated intravascular coagulation and this must be watched for. Metabolic disorders are common, especially hyponatremia and may contribute to poor mental status and seizures. Blood pressure is frequently and inappropriately lowered in neurotrauma patients with disastrous consequences. Many fear that hypertension will lead to edema or exacerbation of hemorrhage, but the reality is that this is an adaptive response directed at perfusing the brain in the face of elevated ICP. Seizures occur in 14% of head injury patients and can raise the metabolic rate to 150-250% of normal (52), as well as the ICP (53). Anticonvulsants decrease the incidence of early but not late seizures from 14 to 4% (54). Current recommendations are to treat all patients with traumatic hemorrhage on more than one CT cut with 7 days of anticonvulsant, typically dilantin (55). Proper nutrition has been correlated with decreased mortality in head trauma patients and is thus encouraged (56).
Invasive Monitoring: Ventriculostomy, which is capable of both monitoring ICP and draining CSF, is recommended for all trauma patients with GCS less than 9 with abnormal CT scans, or those with normal CT scans and two of the following: age older than 40, systolic blood pressure less than 90 or posturing on exam (57). Normal ICP values are age-dependent (Table 6). The treatment threshold for intracranial hypertension is accepted to be 20-25 mmHg, even in children (58). Data correlating outcome with the length of time above an ICP of 20 mmHg is suggestive of the need for aggressive treatment when this level is reached (59). A CPP target of 60 mmHg is now recognized as a guideline by the American Association of Neurological Surgeons Subsection on Neurotrauma and Critical Care.

Measures to Treat Intracranial Pressure:
Positioning: Elevation of the head of the bed to 30-45 degrees (if no contraindication) has been shown to optimize arterial inflow and venous drainage (60). Keeping the neck midline prevents kinking of the jugular veins. Central line placement into the jugular veins should be avoided. Cervical spine collars can be removed with care if patients are sedated/paralyzed enough that they are not moving. They should be replaced later when elevated ICP resolves and remain in place until the cervical spine can be cleared.
Mannitol: Mannitol is very effective and can be life-saving. It decreases blood viscosity by reducing erythrocyte volume, decreasing hematocrit as well as increasing erythrocyte deformability (20). In the face of intact viscosity autoregulation, the result is cerebral vasoconstriction, reduced cerebral blood volume and ICP (11). It also works as an osmotic diuretic and draws free water out of the brain, reducing its volume. The dosage of mannitol is generally 1 g/kg, followed by repeat 0.25 g/kg doses given every 6 hours as needed. Administration should be followed by infusion of saline to maintain intravascular volume. Mannitol has maximal effect in 10 minutes and continues to work for up to 6 hours. Intermittent bolus dosing is more effective than a continuous infusion (55). A serum sodium value of 155 or greater, or a serum osmolality of 320 or greater are contraindications to administration because of the risk of acute tubular necrosis. Caution is required in those with congestive heart failure or hyperemia because of intravascular volume expansion. In fact, this property of mannitol has led to it being recommended as a rescuscitation fluid in the head injured, contrary to what many would think given its diuretic effects (58).
Furosemide has a similar duration of action and enhances free water clearance. It can be administered every six hours at the midway point between mannitol doses. It appears to have a synergistic effect and may slow CSF production(61, 62).
Hyperventilation: Hyperventilation may be the most effective measure in an acutely deteriorating patient as it has a very rapid effect, significantly reduces intracranial volume and because cerebrovascular response to CO2 is almost universally preserved. When used for chronic ICP control, however, worse outcomes are seen, perhaps by lowering cerebral blood flow excessively (63). A low-normal pCO2 is now the recommended target. Prolonged hyperventilation should never be used in the first 24h following head injury as it can further compromise reduced cerebral blood flow during this period though it remains listed in the guidelines as a possible treatment for ICP elevation refractory to other treatments (55).
Sedation and Paralysis: These agents may be needed to facilitate ventilation in intubated patients as well as in the imaging of restless patients. They also decrease cerebral metabolism, and blunt ICP elevations associated with noxious stimuli. Barbiturates have shown mixed results in traumatic brain injury. In patients with elevated ICP refractory to other measures, they achieve a 2 fold greater chance of control. However, hypotension refractory to fluid and pressors develops in 50% of patients (64). Increased risk of pneumonia and the risk of myopathy and neuropathy are major drawbacks.
Hypothermia: Hypothermia works by reducing cerebral metabolism and blood flow proportionately. Mild hypothermia (temperature 32-33°C) started within 6 hours of injury and maintained for 24-48 hours may be of benefit (65-67). Cardiac arrhythmia is a concern, especially below 28°C; coagulopathy, decreased cardiac index, reduced renal function, pancreatitis and masking infections are also concerns.
Surgery: Recently guidelines were published that outline which patients are best brought to surgery, based on clinical features and imaging findings (37). Asymptomatic patients and/or those with intracranial hematomas with thickness less than 15mm (though many recommend 10mm), volume less that 30cc’s, and midline shift less than 5mm can generally be watched. Posterior fossa lesions and hematomas in children should be treated more aggressively. Generally we discourage non-trained individuals from attempting burrhole drainage. Burrholes provide insufficient access and thick clot is very difficult to drain through such a hole even when blood is located in the epidural space. They can be used for diagnostic purposes, and to temporize a patient as will be described.
Northfield, in his pre-CT era text, provided recommendations for taking action (imaging or immediate exploratory surgery) and we feel his recommendations are still prudent. The following were suggested indications (44):

  1. Increasing headache, drowsiness, or vomiting in a conscious patient.
  2. A patient who arrives to your centre unconscious with a report of earlier consciousness subsequent to the injury.
  3. A patient unconscious since the injury, a deepening level of unconsciousness or increasing restlessness.
  4. Progressive papillary change, hemiparesis, posturing, and changes in the vital signs suggestive of a Cushing’s response. Surgery is recommended as early as possible in the evolution of these findings to optimize outcome.
  5. Failure to improve in the first few days following injury.

Northfield recommended special caution in patients presenting with post-traumatic cerebrospinal fluid leak, as this is increases the likelihood of a hematoma, and the loss in intracranial volume can prevent clinical diagnosis of a hematoma until a stage where decline can be precipitous.

Basic Neurosurgical Procedures in Trauma
In the following section we will describe several surgical procedures which can play a life-saving role in head injured patients. Ideally this section would complement proper surgical instruction, however we recognize that even in inexperienced hands these have the potential to be life-saving. We trust that readers will exercise appropriate judgement in deciding how to implement the techniques we will describe, and seek proper surgical training if they would likely be required to perform such procedures. It may be necessary to improvise surgical instruments in a facility not equipped for such procedures (68).
External Ventricular Drainage (EVD): This surgical technique involves placement of a tube into the lateral ventricle and is also known as ventriculostomy (Figure 8a,b). This is the gold standard technique for measuring ICP (55), and very importantly, this can allow CSF drainage, which can be life-saving in individuals with elevated ICP. Indications for EVD insertion were previously discussed.
Ventriculostomy need not be performed in an operating room, and is analogous to placement of a central line. Local anaesthetic alone can be adequate for placement. Typically ventriculostomies are placed on the right side, which is non-dominant even in the majority of left-handed individuals. Placement may be a challenge following brain trauma as swelling can cause ventricular effacement. Nonetheless, with experience even apparently effaced ventricles can be cannulated.
The entry point for a ventriculostomy is generally made in the mid-pupillary line (about 4cm lateral to the midline), and 2cm anterior to the coronal suture, which can generally be palpated. A line drawn directly superior to a point 1cm anterior to the tragus can also be used to approximate the antero-posterior entry point. Hair in the surgical area is shaved, and a generous area is then ‘prepped and draped.’ An incision 2-3cm long is made, generally in the saggital plane. A self-retaining retractor holds the skin edges apart, and the surgeon can then drill a hole through the skull. If the hole is small (as is commonly fashioned with commercial kits) an 11-blade can be used to incise the dura. In a larger burrhole this is accomplished by incising the dura in cruciate fashion. The leaflets of dura are then cauterized with a bipolar to cause them to shrink and retract. Bipolar cautery is then used to create an opening in the pia on the surface of the brain through which the ventriculostomy tube can be passed.
The tubing is then passed through the brain into the ventricle after careful consideration of the trajectory. Some feel that the tubing should simply be directed as perpendicular to the surface of the brain as possible. Others aim for the medial canthus of the ipsilateral eye and a point 1cm anterior to the tragus of the ipsilateral ear. The distance of insertion must be monitored meticulously as this tubing must not be advanced more than 7cm from the surface of the skull or damage to the brainstem can (and frequently does) occur. Entry into the ventricle is generally associated with a ‘popping’ sensation. The tubing is then tunnelled subcutaneously and brought out through a separate stab incision. It is then sutured in place, knowing that dislodgement of the tubing can otherwise occur.
If ventriculostomy is unsuccessful after 3 passes or if the requisite equipment is not available, placement of an alternate measuring device is recommended. Numerous monitors are available and are too numerous to describe here; insertion techniques are generally unique to the device.
Exploratory or “Woodpecker Surgery”: In the days prior to CT scanning a form of exploratory surgery was generally performed on patients (Figure 9). This was done predominantly for diagnostic purposes as evacuation of a hematoma through a burrhole is nearly impossible even in experienced hands (69). Fortunately this technique facilitates rapid conversion to craniotomy when an extra-axial hematoma is detected. Many referred to this colloquially as “woodpecker surgery”.
When performing this surgery, access to both sides of the head is required, so supine positioning is used. The entire head is then shaved, sterilized and draped. A series of burrholes are then placed through separate, non-contiguous incisions. Burrholes are placed in descending order of probability of locating a clot. The first burrhole is performed on the side considered most likely to harbour a clot (ie. ipsilateral to a blown pupil or contralateral to hemiparesis) recognizing that this may not be the appropriate side in a significant proportion of patients. (A blown pupil denotes the correct side of a hematoma only 83% of the time!(11)) The first burrhole is placed temporally, 2.5cm above the zygomatic arch and just anterior to the ear. If no hematoma is found, then the surgeon should fashion an identical burrhole on the contralateral side. If both temporal burrholes are negative, the surgeon should then proceed to place frontal burrholes located antero-superiorly, just behind the hairline and a few centimetres off of the midline. If both of these are negative, burrholes are then fashioned over the parietal lobes.
In general, for each pair of burrholes the one on the side of highest suspicion should be placed first. The incisions should be aligned such that they can be connected into a complete craniotomy incision. The burrholes should also be placed far apart so as to facilitate a large craniotomy which is capable of exposing the majority of the hemisphere allowing the surgeon to surgically manage the majority of pathologies that may be encountered. Care must be taken, however, to stay away from the venous sinuses.
An epidural hematoma may be readily visible just below the skull; a subdural hematoma may be suggested by a bluish hue visible through the dural, though dural opening should routinely be conducted to rule out such a clot. If a clot is located, the procedure is then converted to a craniotomy.
In extreme circumstances, such as herniation in the face of an expected delay in surgical management, one can enlarge the temporal burrhole with a rongeur and open the dura to decompress the brain and a portion of a hematoma if present, temporizing until more definitive surgery can be performed.
Craniotomy/Craniectomy: A large incision is generally made to facilitate a generous craniotomy for the reasons already described. Such an incision is generally in the shape of a “question mark” (Figure10a,b). It is started anterior to the ear, above the tragus so as to avoid the facial nerve. After extending rostrally just above the pinna it is turned posteriorly, circling around the occipitoparietal area, thusly turning anteriorly to finish in the anterior frontal region behind the hairline and a couple of centimetres off of the midline. Generally the incision is made down to bone at the time of the first cut; no special attention is paid to dissection of the temporalis muscle as might be done in other procedures. Hemostasis is achieved with Rainey clips or cautery and the incision is made in a staged process to prevent significant blood loss. The temporal aspect of the incision and a generous temporal burrhole are made before completing the remainder of the incision. The dura is opened to achieve some degree of decompression while the remainder of the craniotomy is completed. The flap is then retracted forwards and secured in place. Monopolar cautery is of great utility in dissecting the superior attachment of the temporalis muscle off of the superior temporal line.
A variable number of burrholes are then made. Three are generally placed in the same locations described for “woodpecker surgery”, with additional locations being in the low posterior occipitotemporal region, and another in the ‘keyhole’ which is a bony depression just posterolateral to the orbit. With this latter burrhole, care must be taken to drill at the posterior aspect of the depression so as to avoid entry into the orbit.
An appropriate instrument (which is blunt and curved such as a Penfield 3) is then passed into the burrhole in the plane between the dura and the skull so as to separate the dura from the inner table to prevent laceration when the bone flap is cut. The burrholes are then connected with a series of saw cuts. A Gigli saw is hand-held and can be used if powered drills are not available. Care is taken not to drop the slippery bone flap while it is delicately dissected off of the underlying dura and removed.
An epidural hematoma should be immediately visible, if present. Even when it is the only hematoma presumed present, one should make a small dural incision to ensure there is not evidence of subdural hemorrhage. A cruciate dural incision can be made with a combination of scalpel and scissors to remove a subdural hematoma. Clot should be removed carefully with suction and irrigation. Gentle brain retraction is used along with appropriate lighting (preferably a headlight) to remove hematoma beyond the margins of the craniotomy, taking care not to damage bridging veins. Any identified sources of hemorrhage should be cauterized or otherwise dealt with.
Once the evacuation is completed and hemostasis is achieved, a decision must be made as to whether or not the bone flap should be replaced. Some prefer to leave the bone flap out (ie. perform a craniectomy) following evacuation of an acute subdural hematoma which is very likely to be followed by significant brain swelling. This is rarely required for an epidural hematoma and in all types of brain injury little is lost if this is performed in a delayed fashion so long as it can be performed expediently. A flap can be preserved in a very cold freezer, or in a pocket fashioned in the patient’s abdominal adipose tissue. When the bone flap is replaced, this is optimally done with small plates and screws designed for the purpose, though holes can be drilled to facilitate passage of wire or heavy suture than can also be used to hold the skull in place.

Guidelines for the Management of Head Injury
Perhaps the biggest recent development in head injury has been the establishment of numerous guidelines to assist clinicians. Unfortunately, the level of evidence upon which the recommendations are based is poor, however the available evidence has been meticulously gathered and analyzed. The Brain Trauma Foundation has made these available on their website. Separate evidence-based guidelines are available for:

Treatment Algorithm

As a summation of the literature and our clinical experience we present the following algorithm for African physicians facing the difficult decisions inherent in initial patient management. Every clinical circumstance is unique and based on patient factors, the skills and experience of the physician, and the resources at hand. Physician judgement should supersede at all times. Patient age and comorbidities should also factor highly in decision-making.


Though traumatic brain injury patients can be amongst the most intellectually and emotionally challenging patients to manage, it is important to avoid nihilism and to remember that aggressive management and adherence to guidelines is leading to improved outcomes. Further, hope is justified because research is bringing forth new treatments that may help these often devastated patients.

Gregory W. J. Hawryluk, MD and Mark Bernstein MD, FRCSC
Division of Neurosurgery
University of Toronto
Toronto, Canada

Save review as .PDF

Reference List

  1. Furnas DW, Sheikh MA, van den Hombergh P, Froeling F, Nunda IM. Traditional craniotomies of the Kisii tribe of Kenya. Ann Plast Surg. 1985;15(6):538-56. Abstract only. (18 kb)
  2. Bahloul M, Chelly H, Ben Hmida M, Ben Hamida C, Ksibi H, Kallel H, Chaari A, Kassis M, Rekik N, Bouaziz M. Prognosis of traumatic head injury in South Tunisia: a multivariate analysis of 437 cases. J Trauma. 2004;57(2):255-61. (185 kb)
  3. Thanni L. Evaluation of guidelines for skull radiography in head injury. Niger Postgrad Med J. 2003;10(4):231-3. Abstract only. (16 kb)
  4. Otieno T, Woodfield JC, Bird P, Hill AG. Trauma in rural Kenya. Injury. 2004;35(12):1228-33. (91 kb)
  5. Akang EE, Kuti MA, Osunkoya AO, Komolafe EO, Malomo AO, Shokunbi MT, Amutta SB. Pattern of fatal head injuries in Ibadan--a 10 year review. Med Sci Law. 2002;42(2):160-6. Abstract only. (20 kb)
  6. Basso A, Previgliano I, Duarte JM, Ferrari N. Advances in management of neurosurgical trauma in different continents. World J Surg. 2001;25(9):1174-8. (122 kb)
  7. Mwang'ombe NJ, Kiboi J. Factors influencing the outcome of severe head injury at Kenyatta National Hospital. East Afr Med J. 2001;78(5):238-41. Abstract only. ( 17 kb)
  8. Solagberu BA, Adekanye AO, Ofoegbu CP, Udoffa US, Abdur-Rahman LO, Taiwo JO. Epidemiology of trauma deaths. West Afr J Med. 2003;22(2):177-81. Abstract only. (21 kb)
  9. Shokunbi MT, Solagberu BA. Mortality in childhood head injury in Ibadan. Afr J Med Med Sci. 1995;24(2):159-63. Abstract only. (17 kb)
  10. Bullock MR, du Trevou MD, van Dellen JR, Nel JP, McKeown CP. Prevention of death from head injury in Natal. S Afr Med J. 1988;73(9):523-7. Abstract only. (18 kb)
  11. Winn HR, editor. Youman's Neurological Surgery. 5 ed. Philadelphia: Saunders; 2004.
  12. Lassen N. Cerebral blood flow and oxygen consumption in man. Physiol Rev. 1959;39(2):183-238. (5 540 kb)
  13. Kontos HA, Wei EP, Navari RM, Levasseur JE, Rosenblum WI, Patterson JL Jr. Responses of cerebral arteries and arterioles to acute hypotension and hypertension. Am J Physiol. 1978;234(4):H371-83. (2 263 kb)
  14. Muizelaar JP, Wei EP, Kontos HA, Becker DP. Cerebral blood flow is regulated by changes in blood pressure and in blood viscosity alike. Stroke. 1986;17(1):44-8. (480 kb)
  15. Siesjo B. Cerebral circulation and metabolism. J Neurosurg. 1984;60(5):883-908. Abstract only. (16 kb)
  16. Cohen PJ, Alexander SC, Smith TC, Reivich M, Wollman H. Effects of hypoxia and normocarbia on cerebral blood flow and metabolism in conscious man. J Appl Physiol. Aug;():183-9. 1967;23(2). (1 497 kb)
  17. Kontos HA, Raper AJ, Patterson JL. Analysis of vasoactivity of local pH, PCO2 and bicarbonate on pial vessels. Stroke. 1977;8(3):358-60. (229 kb)
  18. Bouma, G. Cerebral Circlation after Severe Head Injury: A Clinical Study. [Thesis], Amsterdam, University of Amsterdam 1993.
  19. Enevoldsen EM, Jensen FT. Autoregulation and CO2 responses of cerebral blood flow in patients with acute severe head injury. J Neurosurg. 1978;48(5):689-703. Abstract only. (17 kb)
  20. Muizelaar JP, Wei EP, Kontos HA, Becker DP. Mannitol causes compensatory cerebral vasoconstriction and vasodilation in response to blood viscosity changes. J Neurosurg. 1983;59(5):822-8. Abstract only. (18 kb)
  21. Marmarou A, Maset AL, Ward JD, Choi S, Brooks D, Lutz HA, Moulton RJ, Muizelaar JP, DeSalles A, Young HF. Contribution of CSF and vascular factors to elevation of ICP in severely head-injured patients. J Neurosurg. 1987;66(6):883-90. Abstract only. (21 kb)
  22. Cushing H. Concerning a definite regulatory mechanism of the vasomotor center which controls blood pressure during cerebral compression. Johns Hopkins Hosp Bull 1901;12:290-292.
  23. Nagy KK, Joseph KT, Krosner SM, et al. The utility of head computed tomography after minimal head injury. J Trauma 1999;46:268-270. (177 kb)
  24. Jones PA, Andrews PJ, Midgley S, Anderson SI, Piper IR, Tocher JL, Housley AM, Corrie JA, Slattery J, Dearden NM, et al. Measuring the burden of secondary insults in head-injured patients during intensive care. J Neurosurg Anesthesiol. 1994;6(1):4-14. Abstract only. (23 kb)
  25. Collaborators CT. Effect of intravenous corticosteroids on death within 14 days in 10008 adults with clinically significant head injury (MRC CRASH trial): randomised placebo-controlled trial. Lancet 2004;364:1321-1328. (206 kb)
  26. Chesnut RM, Marshall LF, Klauber MR, Blunt BA, Baldwin N, Eisenberg HM, Jane JA, Marmarou A, Foulkes MA. The role of secondary brain injury in determining outcome from severe head injury. J Trauma. 1993;34(2):216-22. Abstract only. (20 kb)
  27. Rosner MJ, Daughton S. Nonsurgical management of increased intracranial pressure. J Trauma. 1990;30(8):933-40. Abstract only. (17 kb)
  28. Siesjo BK, Carlsson C, Hagerdal M, et al. Brain metabolism in the critically ill. Crit Care Med 1976;4:283-294. Abstract only. (18 kb)
  29. Teasdale G, Jennett B. Assessment of coma and impaired consciousness: A practical scale. Lancet 1974;2:81-84. (563 kb)
  30. Clifton GL, Grossman RG, Makela ME, Miner ME, Handel S, Sadhu V. Neurological course and correlated computerized tomography findings after severe closed head injury. J Neurosurg. 1980;52(5):611-24. Abstract only. (21 kb)
  31. Foulkes M, Eisenberg HM, Jane JA, et al. The Traumatic Coma Data Bank: Design, methods and baseline characteristics. J Neurosurg 1991;75:S8-S13.
  32. McCrory P, Johnston K, Meeuwisse W, Aubry M, Cantu R, Dvorak J, Graf-Baumann T, Kelly J, Lovell M, Schamasch P. Summary and agreement statement of the 2nd International Conference on Concussion in Sport, Prague 2004. Clin J Sport Med. 2005;15(2):48-55. (97 kb)
  33. Hawryluk G, Bernstein M. Post-concussion syndrome. Parkhurst Exchange. 2005;13:107-112.
  34. Gentleman SM, Roberts GW, Gennarelli TA, Maxwell WL, Adams JH, Kerr S, Graham DI. Axonal injury: a universal consequence of fatal closed head injury? Acta Neuropathol (Berl). 1995;89(6):537-43. (19 kb)
  35. Gennarelli TA, Spielman GM, Langfitt TW, Gildenberg PL, Harrington T, Jane JA, Marshall LF, Miller JD, Pitts LH. Influence of the type of intracranial lesion on outcome from severe head injury. J Neurosurg. 1982;56(1):26-32.
  36. Legbo JN, Shehu BB. Managing scalp defects in sub-Saharan Africa. East Afr Med J. 2004;81(2):87-91. Abstract only. (17 kb)
  37. Bullock MR, Chesnut R, Ghajar J, Gordon D, Hartl R, Newell DW, Servadei F, Walters B, Wilberger JE. Guidelines for the Surgical Management of Traumatic Brain Injury. Neurosurgery 2006;53. (2 327 kb)
  38. Jamieson KG, Yelland J. Extradural hematoma: Report of 167 cases. J Neurosurg 1968;29:13-23.
  39. Greenburg M. Handbook of Neurosurgery, Chapter on Head Injury. 5 ed. New York: Thieme Medical Publishers; 2005.
  40. Marshall LF, Gautille T, Klauber MR, et al. The outcome of severe closed head injury. J Neurosurg 1991;75(Suppl):S8-S13.
  41. Eisenberg HM, Gary HE Jr, Aldrich EF, Saydjari C, Turner B, Foulkes MA, Jane JA, Marmarou A, Marshall LF, Young HF. Initial CT findings in 753 patients with severe head injury. A report from the NIH Traumatic Coma Data Bank. J Neurosurg. 1990;73(5):688-98. (874 kb)
  42. Pruitt B. Management and Prognosis of Penetrating Brain Injury. Journal of Trauma 2001;51(2):S1-S86. (11 kb)
  43. Stiell IG, Wells GA, Vandemheen K, Clement C, Lesiuk H, Laupacis A, McKnight RD, Verbeek R, Brison R, Cass D, Eisenhauer ME, Greenberg G, Worthington J. The Canadian CT Head Rule for patients with minor head injury. Lancet. 2001;357(9266):1391-6. (95 kb)
  44. Northfield D. Head Injury. In: Surgery of the Central Nervous System. London: Blackwell Scientific Publications; 1973. p. 753-830. (99 kb)
  45. Obisesan AA, Bohrer SP. The uses and abuses of skull X-rays in head injury. Niger Med J. 1979;9(1):65-9.
  46. Trauma ACoSCo. Advanced Trauma Life Support Program for Doctors. 7th ed: First Impression; 2004.
  47. Chen B. Reappraisal of the value of mode-A-echoencephalography. Dis Nerv Syst. 1977;38(2):83-7. Abstract only. (15 kb)
  48. Umerah B. Cerebral angiography in Central Africa. Diagn Imaging. 1981;50(4):225-8. Abstract only. (15 kb)
  49. Yamaki T, Hirakawa K, Ueguchi T, Tenjin H, Kuboyama T, Nakagawa Y. Chronological evaluation of acute traumatic intracerebral haematoma. Acta Neurochir (Wien). 1990;103(3-4):112-5. Abstract only. (18 kb)
  50. Kamada T, Fusamoto H, Kawano S, Noguchi M, Hiramatsu K. Gastrointestinal bleeding following head injury: a clinical study of 433 cases. J Trauma. 1977;17(1):44-7. Abstract only. (18 kb)
  51. Mutlu GM, Mutlu EA, Factor P. Prevention and treatment of gastrointestinal complications in patients on mechanical ventilation. Am J Respir Med. 2003;2(5):395-411. Abstract only. (19 kb)
  52. Meldrum BS, Nilsson B. Cerebral blood flow and metabolic rate early and late in prolonged epileptic seizures induced in rats by bicuculline. Brain. 1976;99(3):523-42. Abstract only. (17 kb)
  53. Goitein KJ, Shohami E. Intracranial pressure during prolonged experimental convulsions in cats. J Neurol. 1983;230(4):259-66. Abstract only. (16 kb)
  54. Temkin NR, Dikmen SS, Wilensky AJ, Keihm J, Chabal S, Winn HR. A randomized, double-blind study of phenytoin for the prevention of post-traumatic seizures. N Engl J Med. 1990;323(8):497-502. (211 kb)
  55. Brain Trauma Foundation AAoN, Surgeons JSoNaCC. Guidelines for the management of severe traumatic brain injury. J Neurotrauma. 2000;17:451–627. (1 257 kb)
  56. Rapp RP, Young B, Twyman D, Bivins BA, Haack D, Tibbs PA, Bean JR. The favorable effect of early parenteral feeding on survival in head-injured patients. J Neurosurg. 1983;58(6):906-12. Abstract only. (19 kb)
  57. Narayan RK, Kishore PR, Becker DP, Ward JD, Enas GG, Greenberg RP, Domingues Da Silva A, Lipper MH, Choi SC, Mayhall CG, Lutz HA 3rd, Young HF. Intracranial pressure: to monitor or not to monitor? A review of our experience with severe head injury. J Neurosurg. 1982;56(5):650-9.
  58. Brain Trauma Foundation AAoN. Guidelines for the Acute Medical Management of Severe Traumatic Brain Injury in Infants, Children and Adolescents. Pediatric Critical Care Medicine 2003;4(3):S1-S75. (5 639 kb)
  59. Marmarou A, Anderson RL, Ward JD, et al. Impact of ICP and hypotension on outcome in patients with severe head injury. J Neurosurg 1991;75:S59-S64. (712 kb)
  60. Feldman Z, Kanter MJ, Robertson CS, et al. Effect of head elevation on intracranial pressure, cerebral perfusion pressure, and cerebral blood flow in head-injured patients. J Neurosurg 1992;76:207-11. Abstract only. (23 kb)
  61. Buhrley LE, Reed DJ. The effect of furosemide on sodium-22 uptake into cerbrospinal fluid and brain. Exp Brain Res 1972;14:503-10.
  62. Marion DW, Letarte PB. Management of intracranial hypertension. Contemp Neurosurg 1997;19(3):1-6. (817 kb)
  63. Muizelaar JP, Wei EP, Kontos HA, Becker DP. Adverse effects of prolonged hyperventilation in patients with severe head injury: A randomized clinical trial. J Neurosurg 1983;75:822-828. Abstract only. (26 kb)
  64. Eisenberg HM, Frankowski RF, Contant CF, Marshall LF, Walker MD. High-dose barbiturate control of elevated intracranial pressure in patients with severe head injury. J Neurosurg. 1988;69(1):15-23. Abstract only. (18 kb)
  65. Clifton GL, Allen S, Barrodale P, et al. A Phase II Study of moderate hypothermia in severe brain injury. J Neurotrauma 1993;10:263-271. Abstract only. (21 kb)
  66. Marion DW, Obrist WD, Carlier PM, et al. The use of moderate therapeutic hypothermia for patients with severe head injuries: A preliminary report. J Neurosurg 1993;79:354-362. Abstract only. (21 kb)
  67. Clifton GL, Miller ER, Choi SC, et al. Lack of effect of induction of hypothermia after acute brain injury. N Engl J Med 2001;344:556-563. (102 kb)
  68. Wiggin T. Emergency craniotomy using improvised instruments. Trop Doct. 2001;31(3):174. (356 kb)
  69. Muhammad I. Management of head injuries at the Abu Hospital, Zaria. East Afr Med J. 1990;67(6):447-51. Abstract only. (15 kb)
  70. Foundation BT. Guidelines for Prehospital Management of Traumatic Brain Injury. 2000. (935 kb)



Previous Review
Surgery in Africa Home
Next Review