Chapter 4: Neurotrauma

This is a typical case of severe TBI with bilateral hemorrhagic contusions in the temporal lobes. Its proximity to the bony structures makes it a frequent location in the brain to be contused in trauma. Resuscitation of TBI patients varies widely, however, due to the heterogeneity of the disease itself. The aim of all good early resuscitation efforts is to begin as early as possible, with many efforts beginning in the prehospital setting, with an attention to airway, breathing, and circulation.

Three specific end points have been found to be independent predictors of poor outcome in the prehospital/emergency department setting: hypothermia, hypoxia, and hypotension.¹ Hypothermia is likely a marker of poor resuscitation, and most would agree that core body temperature should be passively supported during the resuscitation phase rather than actively warmed with a device. Aggressive volume resuscitation for hypotension and adequate ventilation are the primary focus of initial resuscitation efforts. Prehospital resuscitation with hypertonic saline in TBI has failed to demonstrate a long-term benefit,² and in a post-hoc analysis of the Saline versus Albumin Fluid Evaluation trial,³ fluid resuscitation with albumin was associated with higher mortality rates than was resuscitation with saline. Therefore, the administration of isotonic crystalloids is the preferred method by which to volume resuscitate. All TBI patients should be ventilated to a goal of normal Pco₂ and be given supplemental oxygen to achieve Spo₂ greater than 90%. During the early resuscitation phase, it is important to realize that simple measures such as elevation of the head of the bed (30 degrees), midline positioning of the head (relieving any blockage of jugular venous drainage), and adequate pain control and sedation are very simple and effective methods to reduce intracranial pressure (ICP).

Reported risk factors for seizures include GCS score less than 10, cortical contusions, depressed skull fractures, wounds with dural penetration, prolonged length (longer than 24 hours) of coma, and posttraumatic amnesia. The majority of early posttraumatic seizures occur within the initial 48 hours of injury.⁴ However, some seizures may escape clinical detection, and may be unnoticed in intubated sedated patients in the absence of electroencephalographic (EEG) monitoring. The presence of convulsive SE is associated with high mortality rate. Effective prophylaxis of early posttraumatic seizures reduces brain metabolic demands, thereby reducing intracranial pressure and neurotransmitter release. This in turn minimizes secondary brain injury. Furthermore, anticonvulsant treatment can minimize cognitive and behavioral sequelae.

Phenytoin is an established standard antiepileptic drug (AED) in the setting of acute TBI. The American Academy of Neurology suggests using phenytoin for seizure prevention only in the first 7 days after TBI.⁵ If AEDs are given, it is important to remember that there is no added benefit of continuing beyond 7 days if it was given as a prophylaxis in the setting of trauma. The availability of newer AEDs questions the use of phenytoin as the first-line AED in this setting. Levetiracetam has gained favor in acute brain injury setting due to its tolerability, ease of use without the need to follow a drug level, and minimal drug-to-drug interactions. It has been used in the neurocritical care setting for several years now and numerous studies have reported on oral as well as IV use of this medication for the treatment or prevention of seizures in TBI. However, existing studies of levetiracetam regarding the safety and efficacy are not definitive, and do not provide enough evidence that levetiracetam has better short- and long-term outcomes when compared to phenytoin.⁶

High doses of steroids are greatly beneficial in experimental models, reducing lipid peroxidation and improving tissue recovery. Clinical studies with glucocorticoids have not shown similar benefit. The results from the large Corticosteroid Randomization After Significant Head injury (CRASH) trial demonstrated no benefit and increased mortality rate in TBI patients randomized to 3 g of methylprednisolone in the first 72 hours after injury. Currently, no data exist to support the use of glucocorticoid steroids acutely, and given the increased mortality rates seen in the CRASH trial, they are contraindicated acutely after brain injury.^7,8

The Brain Trauma Foundation guidelines state that ICP should be monitored in those with a postresuscitation GCS of 3 to 8 and an abnormal CT scan, and further for those with a similar severity and a normal CT scan if two of the following are present: age older than 40 years, posturing, or hypotension. The first choice for monitoring should always be an external ventricular drain because it can be recalibrated after placement and also offers therapy in the form of cerebrospinal fluid (CSF) diversion. Parenchymal monitors provide continuous monitoring of ICP but can only be calibrated prior to placement and have an associated drift in their measurement.⁹

The ability to monitor brain tissue oxygen tension (Pbto₂) has added a dimension of monitoring that provides insight to cerebral metabolism. The preponderance of case series data indicate that a Pbto₂ level less than 15 mm Hg is associated with poor outcome, and may respond to either augmentation of mean arterial blood pressure (MAP), alteration in ventilation strategy, or transfusion of red blood cells. However, many questions regarding the utility of this monitoring technique have yet to be answered, such as timing, location, and pertinent treatment thresholds. Until such data become available, it is important to remember that the Pbto₂ value is a composite of various components of the content of oxygen in the arteries as well as an unknown quantity of passive diffusion. This value is also a very focal/regional measure and should be put into context of the entire clinical picture prior to embarking on therapeutic interventions.¹⁰

There are sufficient data to support the use of periprocedural antibiotics, but the infusion must begin prior to the skin incision. The choice of antibiotic should be either cefazolin (1 g IV) or nafcillin (1 g IV). Vancomycin (1 g IV) can be utilized for those patients with a known penicillin allergy. The continued use of antibiotics for prophylaxis is controversial and has not been proven to reduce the rate of ventriculitis and may increase the incidence of drug-resistant bacterial infections.¹¹

The 2007 Brain Trauma Foundation (BTF) guidelines have indicated that CPP thresholds should be maintained within a range of greater than 50 mm Hg and less than 70 mm Hg. This threshold is based upon data that support CPP values lower than 50 mm Hg are associated with more cerebral insults, while achieving CPP more than 70 mm Hg on every TBI patient results in a fourfold increase in lung injury likely due to aggressive volume resuscitation and excessive use of pressors.¹² The use of Pbto₂ monitoring may help guide targeted CPP management so as to avoid tissue hypoxia while minimizing the risk of complications related to pressure augmentation.

There have been two multicenter randomized controlled trials (RCTs) that have not demonstrated any benefit for the use of therapeutic hypothermia as a neuroprotectant.^13-15 However, therapeutic hypothermia can still be considered as an option in the treatment of raised intracranial pressure.¹³ Owing to the lack of data and associated management complexities, therapeutic hypothermia has traditionally been reserved as a treatment option for patients who are either refractory or unable to receive osmotic therapy. The advent of modern temperature-modulating devices has allowed for a more safe and efficient delivery of therapeutic hypothermia, and as a result, this therapy may become more widely utilized in TBI patients. Since ICP is strongly dependent on core body temperature, any decrease in temperature less than 37°C will likely result in a reduction in ICP. However, this therapy has been traditionally targeted to lower core body temperature to 32^oC to 34°C. The risk for infectious complications with therapeutic hypothermia is duration-dependent, with the rate of infectious complications rising sharply above 72 hours. Hypothermia will also result in a coagulopathy and increased risk of bleeding, although it is important to realize that there has not been any significant increase in intracranial bleeding due to hypothermia in any of the RCTs.

If hypothermia is utilized to control ICP, careful consideration should be also given to shivering, electrolyte imbalances, ventilator management, and rewarming. A commonly overlooked aspect of hypothermia's effect on normal physiology occurs in relation to the management of ventilation and blood-gas results. As a result of decreased metabolic activity, there is decreased production of carbon dioxide at lower body temperatures. If mechanical ventilation parameters are left unchanged during cooling, hyperventilation will occur, which may cause cerebral vasoconstriction and reduced cerebral blood flow. At lower temperatures, the solubility of gases in a liquid mixture increases. Most blood-gas analyzers warm the sample to 37°C prior to determining the values for pH, Pco₂, and Po₂. Without correcting for the patient's actual body temperature, this analysis will overestimate the Pco₂ during hypothermia because portions of the dissolved gases come out of solution on warming and contribute to the partial pressure detected. Two methods of blood-gas analysis exist with regard to changing patient body temperature: (1) alpha-stat management and (2) pH-stat management. Using alpha-stat ventilator management, the Pco₂ detected by the blood-gas analyzer at 37°C is targeted to 40 mm Hg regardless of body temperature. This method of blood-gas management may lead to hyperventilation due to the overestimation of Pco₂. In contrast, pH-stat management targets a temperature-corrected Pco₂ and pH. Using this method, as body temperature decreases, the total amount of Co₂ in blood will increase as solubility increases. This poses the risk for inducing hypercapnia, which may result in cerebral vasodilation and increased intracranial pressure.¹⁶

Surgical decompression limits the damage caused by secondary injury (delayed brain injury) by reducing increased ICP with subsequent improvement in brain oxygenation. As with any surgical procedure, there are inherent risks. The decision to perform decompressive techniques is primarily based on the evacuation of a mass lesion, with temporal and frontal lesions more likely to result in decompression. There are no standard clinical criteria, although many institutions reserve this procedure for younger patients in coma who have failed at least one ICP-lowering measure. The key is not to delay the procedure since the primary benefit is as a result of reducing sequelae from secondary injury. Once decompression is decided upon, resection of a larger bone fragment is performed to allow for greater dural expansion with less risk of herniation. All clinical studies of decompression have demonstrated immediate reductions in ICP; however, there is conflicting evidence regarding the effects of craniectomy on long-term outcomes. As a result, this technique is not as widely utilized as medical interventions for lowering ICP.^17,18

The identification of any cervical spine injuries is part of the standard evaluation after the patient has been hemodynamically stabilized. The algorithm is straightforward in patients who are awake and alert. In this population, the need for additional radiographic evaluations is based upon the bedside examination. Those without pain (distracting, midline neck), no neurologic deficits localizing to the cervical spine, and not intoxicated are very unlikely to have significant cervical spinal injury and can be cleared without radiographs. If any of these signs or symptoms is present, then patients should undergo full C-T1 spine radiographs (including anteroposterior, lateral, and odontoid films) in addition to CT images from the occiput to T1. In patients with an alteration in mental status with a negative CT and gross motor function of extremities, flexion/extension radiography should not be performed. Many institutions have begun relying upon magnetic resonance (MR) images to look for ligamentous injury; however, the risk- to-benefit ratio of obtaining MR images in addition to CT is not clear, and its use must be individualized in each institution. If a decision is made to perform MRI, it must be done within 72 hours of injury in order to have reliable results. Most modern CT scanners in addition to plain radiographs likely identify nearly all significant cervical spine instabilities; however, there are no evidenced-based guidelines to support the practice of clearing cervical spines with only CT images in obtunded patients.¹⁹

There is a high rate of DVT after TBI, indicating that this patient population should have prophylactic therapy. However, there is considerable concern that administering heparin and low-molecular-weight heparin will increase the risk for hematoma expansion.²⁰ In the absence of acquired coagulopathy, the risk for hematoma expansion decreases significantly after 48 hours. In addition, there is a very low risk for hematoma expansion with prophylactic doses of heparin or low-molecular-weight heparin resulting in hematoma expansion in this patient population. For these reasons, a reasonable, safe approach toward this patient population is to begin prophylaxis 24 hours after radiographic demonstration of hematoma stability. Others have advocated surveillance with weekly venous duplex ultrasound in addition to prophylaxis, although there is no evidence that this practice reduces the complications related to DVT after TBI.²¹

There are instances, such as the need for repeated surgeries or ongoing bleeding, where the use of DVT prophylaxis is not initiated for a prolonged period of time. In these cases, the use of temporary inferior vena cava (IVC) filters should be considered. However, it must be recognized that an IVC filter only provides incomplete prevention for only lower-extremity DVT and can pose a risk for future complications related to filter thrombosis, migration, and/or fracturing. Therefore, removing IVC filters once a patient is able to receive pharmacologic prophylaxis is an important step in the prevention of future complications.

In cases where a DVT has been diagnosed, the general recommendation is to wait for 7 days after intracranial surgery, although there are no clinical trial data to support any specific timeline. Decisions regarding timing should balance the risk for intracranial as well as systemic bleeding with the urgency for therapy.

Unfortunately, there is no level 1 evidence to support one therapy over another. In fact, owing to a paucity of prospective clinical trial data, there is a lack of level 1 data to support the use of any osmotic therapy after TBI. However, each therapy does have a slightly different physiologic impact, which may lead to a preferential use of one therapy over another.²²

Mannitol

Mannitol is an osmotic agent that draws excess fluid from the cranial cavity, thereby decreasing ICP, and has also been associated with significant diuresis, acute renal failure, hyperkalemia, hypotension, and rebound increments in ICP. For these reasons, it has been recommended that mannitol only be used when signs of elevated ICP or deteriorating neurologic status suggest the benefits of mannitol outweigh potential complications or adverse effects. There remains considerable uncertainty regarding how and when it should be used but the recommended dose is 0.25 to 1.0 g/kg of body weight, with a goal to avoid hypotension due to intravascular volume depletion. Some clinicians have advocated replacement of urinary losses to avoid intravascular volume depletion. While improved outcomes may be obtained with higher doses, decisions regarding higher doses (higher than 1 g/kg) of mannitol administration should be made on a case-by-case basis.

Hypertonic Saline

Hypertonic saline is an osmotic agent that has traditionally been used as an adjunct to mannitol or in individuals who have become tolerant to mannitol. However, recent studies have examined hypertonic saline as a primary measure for ICP control. Hypertonic saline exerts its effect primarily by increasing serum sodium and osmolarity, thereby establishing an osmotic gradient. Water diffuses passively from cerebral intracellular and interstitial spaces into capillaries resulting in a reduction in ICP. Although mannitol works similarly, sodium chloride has a better reflection coefficient (1.0) than mannitol (0.9), making it a better osmotic agent. Hypertonic saline may also normalize resting membrane potential and cell volume by restoring normal intracellular electrolyte balance in injured cells. The dose and administration varies greatly, with boluses ranging between 30 mL of 23.4% NaCl and 150 mL of 3% NaCl, whereas others have advocated the use of a continuous infusion of either 2% or 3% NaCl to reach a goal Na of 150 mmol/L. Regardless of the choice of administration, therapy should be targeted to a specific ICP/CPP goal.

Based on this evidence, some general clinical guidelines regarding the use of mannitol and hypertonic saline:

1. Mannitol may be of added benefit when there is intravascular volume overload, but careful attention should be paid to urine output so that patients do not become intravascular volume depleted. The goal is euvolemic intravascular volume status.

2. Mannitol dosing should be done as a weight-based schedule ranging between 0.25 and 1.0 g/kg.

3. While there is no ceiling as to when osmotic therapies should be discontinued, the likelihood of developing renal insufficiency does increase with supranormal serum sodium levels.

4. Hypertonic saline should be administered with caution in patients with renal insufficiency or congestive heart failure.

5. Hypertonic saline may have an additional benefit of raising the MAP and therefore have a dual impact on improving CPP.

Regulation of blood carbon dioxide levels has a significant impact on cerebral blood flow and therefore intracranial volume and intracranial pressure. During mild hyperventilation, increased oxygen extraction can compensate for decreased blood flow and volume, allowing normal cellular metabolism to continue; however, prolonged hyperventilation may increase metabolic acidosis. In the short term, hyperventilation decreases cerebral blood Co₂, leading to an increase in pH, which may diminish the detrimental effects of acidosis. However, this process depends on the availability of bicarbonate in the cerebrospinal fluid. Prolonged hyperventilation may deplete bicarbonate levels, which may in turn result in ischemia and poorer outcomes. There have been four studies examining hyperventilation after TBI. The only RCT examining prolonged hyperventilation demonstrated poorer clinical outcomes, which were likely due to a depletion of cerebral bicarbonate supplies.²³ As a result, ventilation goals should be used to maintain arterial Co₂ within the range of normal (35 to 45 mm Hg). The use of hyperventilation should be reserved for rescue therapy (sudden surges in ICP), temporary reduction of ICP during procedures, and/or as an intermediate intervention until a more durable therapy can be initiated.

Among patients with severe TBI, the proportion of those who experience anemia and who receive a blood transfusion during the acute postinjury phase have not been carefully described, but recent series have described 40% to 50% of patients with moderate to severe TBI having at least one hematocrit less than 30%.^24,25 Clinical guidelines have recommended that anemia not be the sole consideration in decisions regarding transfusion. Instead, the decision to transfuse should be based upon reducing tissue ischemia. The TBI patient population is generally young and includes otherwise healthy people, and the Transfusion Requirements in Critical Care (TRICC) trial suggests that this particular subgroup of critically ill patients is at risk for harm from liberal transfusion. However, there is genuine disagreement and clinical equipoise as to whether brain-injured patients would benefit from more liberal (to maintain hemoglobin level greater than 10 g/dL) or more restrictive transfusion (to maintain hemoglobin level greater than 7 g/dL).²⁶

Periodic sympathetic hyperactivity occurs up to 33% after severe TBI, although it has been reported with less frequency in other injury types as well.²⁷ These episodes are typically characterized as sudden onset of exaggerated sympathetic responses including hypertension, tachycardia, tachypnea, high fever, sweating, and papillary dilation, along with occasional dystonic posturing. Clinicians should be aware of this syndrome and understand that the first step of managing these episodes is to investigate and rule out other underlying medical conditions. DVT, pulmonary embolism, myocardial infarction, pneumothorax, and sepsis with a hyperdynamic phase can all lead to a syndrome at least partially similar to sympathetic storming. People with TBI are prone to infection such as aspiration pneumonia as well as volume overload, which often leads to total-body water overload. Pleural effusion and pulmonary congestion are not uncommon, and these conditions can mimic the syndrome as well. Bromocriptine, β blockers such as propranolol, morphine sulfate, dantrolene, and clonidine may be helpful. For severe refractory cases, continuous IV sedation and anesthetic medications are occasionally needed too. Central nervous system (CNS) storming can be challenging to manage and long-lasting. It is not uncommon to observe prolonged phases of sporadically occurring storming events even weeks after the initial injury.

Easily reproducible classification schemes are important methods by which to assess severity of injury, facilitate communication between practitioners, and, more importantly, aid in the determination of prognosis. The American Spinal Injury Association (ASIA) assessment tool utilizes both disability-specific and functional-independence measures and is considered the gold standard for the assessment of SCI (Figure 4-2).²⁸

There have been three randomized controlled trials in addressing this question: The National Acute Spinal Cord Injury Studies (NASCIS) I, II, and III.^29-31 NASCIS-I was a negative study showing no benefit with steroid. The main criticism for failure on NASCIS-I was underdosing, and as such the NASCIS-II used "high"-dose steroid, methylprednisolone 30 mg/kg IV bolus followed by 5.4 mg/kg per hour for total of 23 hours. Methylprednisolone was given to 162 patients, naloxone was given to 154 patients, and placebo was given to 171 patients.

Most (95%) of the patients who were randomized to the steroid group received the drug within the first 14 hours of the injury. There was no difference in the 6-month outcome between the groups. Therefore, NASCIS-II is considered a negative study. Only in a post-hoc subgroup analysis, which focused on the timing of the steroid administration, was there was a positive 6-month outcome for some individuals who received the drug within the first 8 hours. This post-hoc analysis has been criticized and, in general, not considered class I or II evidence, as the original study's randomized groups did not show any differences. NASCIS-III randomized patients to nonbolus, methylprednisolone 5.4 mg/kg per hour for 24 hours versus 48 hours versus tirilazad 2.5 mg/kg every 6 hours for 48 hours. There was no difference between these groups at any point. Only positive reports are again based on the post-hoc analysis showing benefit in people who received the steroid between 3 and 8 hours of injury but no benefit in those who received the drug within the first 3 hours. As a result, the routine use of steroids remains controversial and is not recommended as a standard of care. It may be considered if the administration can be performed while minimizing the associated risks of high-dose steroids,³² making it an option and not a strong recommendation.

The mainstay of ICU treatment of SCI is supportive for respiratory and cardiac systems and preventative for infections and DVTs.

Respiratory System

Cervical cord injury is often associated with changes in respiratory patterns that are important to be aware of and can impact on ventilatory strategy. High-level cervical lesions result in the recruitment of accessory muscles with inspiration resulting in expansion of the upper rib cage with concomitant ascending diaphragm. This results in a paradoxical breathing pattern, reduction in all lung volumes (except residual volumes) but increased pulmonary and chest wall compliance and work of breathing.³³ Lower cervical lesions may result in less ventilatory difficulties but a significant inability to clear secretions. As a result, this population is at significant risk for developing aspiration pneumonia.

The main goals of respiratory management are to provide aggressive pulmonary hygiene, routine bronchodilator therapy, and aggressive attempts at ventilator weaning. Attempting intermittent ventilation with high tidal volumes (10 to 15 mL/kg), sighs, or adequate positive end-expiratory pressure (PEEP) may reduce the incidence and complications related to atelectasis.

Cardiac System

Hemodynamic instability is common after SCI. Lesions involving the cervical and upper thoracic cord result in sympathetic denervation, resulting in arteriolar vasodilation, venous pooling, bradycardia, and reduced myocardial contractility. The end result of this is a shock state that is characterized by hypotension, low systemic vascular resistance, and sinus bradycardia. However, given the impairment in myocardial contractility, hypotension cannot be overcome just with volume resuscitation and often requires additional vasopressor and/or inotropic agents. Some practitioners advocate for augmentation of MAP to maintain an adequate "cord perfusion pressure" greater than 60 mm Hg, although there is no reliable method to measure cord pressure and no clinical data support this practice.³⁴

DVT Prophylaxis

SCI patients have a very high incidence of DVT, and while all agree this type of patient population should receive early prophylaxis, the regimen varies among institutions with regard to timing, dosing, and duration of pharmacologic intervention, as well as to the use of inferior IVC filters (see above). In general, there is no advantage to either subcutaneous low-molecular-weight heparin or heparin.³⁵

Infectious Complications

Infectious complications are the leading cause for morbidity and mortality after SCI.³⁶ This may be due to a combination of acquired immunodeficiency from the injury itself as well as frequent use of high-dose steroids.³⁷ For reasons outlined above, this patient population is at very high risk for developing respiratory infections. Urinary tract infection risk is elevated due to neurologic dysfunction resulting in incontinence, high bladder pressures, and reflux. There should be a low threshold for investigating for GI tract infections in patients with persistent fever given the inability of patients to mount symptoms.

Whether acute surgical intervention is needed in order to decompress the injured spinal cord continues to be a controversial topic. This is mainly due to conflicting data in the literature with an absence of large randomized prospective studies demonstrating long-term outcome benefits. In some studies, when an early surgical decompression was performed, patients did not have better outcomes, and some of them had even worse outcomes.^38-40 Despites anecdotal reports of some benefits after early decompression, there is no strong evidence at this time for performing an early surgery. From a physiologic standpoint, an early intervention in correcting the malalignment of the spine structures makes at least intuitive sense. One of the frequently pointed out criticisms for these negative studies focuses on the timing of the decompression. Like the brain, the spinal cord may not tolerate a prolonged duration of injury. So-called "early" surgical decompressions were done in days to weeks from the injury and not in the first few minutes or hours. By the time a decompression is done, typically 3 to 5 days or a week later, the injury has been completed, and no further intervention is likely to provide any long-term benefit. All of these criticisms point toward the need for a randomized controlled trial in order to better define the role and the timing of the surgical decompression.