TRAUMATIC BRAIN INJURY
The acute care surgeon is certain to encounter neurologic disease while treating the acutely ill patient. Trauma will clearly be the source of most neurologic pathology encountered by the acute care surgeon. The serious nature of many neurologic emergencies, their dramatic forms of presentation, and their complexity can distract providers in the early stages of resuscitation. A preplanned, methodical, but efficient approach to these patients provides the best hope for a quality outcome.
INITIAL DECISIONS—EMERGENCY DEPARTMENT MANAGEMENT
The Advance Trauma Life Support Course has ensured that adequate resuscitation is provided to the victims of trauma. It is important to remember that other neurologic emergencies also require prompt primary resuscitation. In the case of trauma, management of airway, breathing, and life-threatening bleeding is essential.
Airway and Breathing
Epidemiology has demonstrated that patients with traumatic brain injury (TBI) who are allowed oxygen saturations <90% have poorer outcomes. Work with brain tissue monitors has demonstrated that hypoxic insults are additive. This means that multiple brief hypoxic insults add up to a total time of hypoxic insult. Studies have shown that a total of 30 minutes of hypoxia time can result in significantly poorer outcomes.–13
It would seem that limiting such small insults via a well-secured airway would be best for the brain-injured patient. For this reason, orotracheal intubation has been advocated as part of prehospital care for all patients with a Glasgow Coma Score (GCS) < 9. Interestingly, when the impact of prehospital intubation on patients with severe head injuries was studied, patients who were intubated actually had worse outcomes.4,5
The factors contributing the increased morbidity in patients intubated in the field appear to be hyperventilation and poorly performed intubation. Hyperventilation has been implicated for some time as a source of secondary brain injury due to its capacity to cause cerebral vasoconstriction and cerebral ischemia.6 Work in the last few years has demonstrated the tendency for patients intubated in the field to be hyperventilated and has demonstrated inferior outcomes in this same patient group. However, hypoventilated patients also appear to have inferior outcomes. Further complicating the issue of airway management in patients with TBI are concerns about the increased morbidity of poorly performed intubation. Many of the early providers of care to severely brain-injured patients will have poor intubation skills, only performing intubations on patients one to two times in 1–2 years.
Optimum care appears to be intubation at the earliest, safest time. This involves balancing the skill of the intubator with the time to intubation. In an urban setting, this might mean delaying intubation for the 10 minutes required to transport to an emergency department (ED) with an anesthesiologist available, while in a rural setting immediate intubation by a less skilled provider might be required.
Once intubated, pCO2 should be maintained in the 35–40 mm Hg range per the Brain Trauma Foundation Guidelines, thereby avoiding both hypoventilation and hyperventilation. Recent data suggest that best outcomes may result from slightly lower levels, in the 30–35 mm Hg range, but this will require further validation prior to changing the recommendation.7,8
There should be no confusion that hyperventilation in the presence of signs of herniation is appropriate. In the prehospital or ED environments, before intracranial pressure (ICP) monitoring has been instituted, patients who manifest clinical signs of herniation such as a unilaterally dilated pupil, an asymmetric motor examination, or a declining GCS should be hyperventilated in an attempt to blunt the impact of herniation. What is to be avoided is prophylactic or inadvertent hyperventilation.
While end-tidal CO2 (ETCO2) has many uses in the early management of injured patients, it lacks the accuracy, sensitivity, and specificity to manage the pCO2 parameters for ventilation in TBI. This is because concurrent conditions such as hypotension, cardiac failure, pulmonary contusion, and even frequent patient movement severely confound the correlation between Arterial and ETCO2. A recent study confirmed this poor correlation and found that patients presenting with ETCO2 in the 35–40 mm Hg range were likely to be underventilated (pCO2 > 40 mm Hg) 80% of the time and severely underventilated (pCO2 > 50 mm Hg) 30% of the time.9
Circulation
Even a single episode of systolic blood pressure below 90 mm Hg can result in poor outcomes for the victims of TBI.3,10 For this reason, victims of TBI require vigorous resuscitation of their systolic blood pressure to >90 mm Hg. Ninety millimeters of mercury has traditionally been the threshold used in studies of outcome after head injury. Its basis lies in historical precedent, and it may be that sharp changes in mortality are actually observed at a different threshold.11
There is a trend in trauma surgery to set lower resuscitation thresholds and to limit crystalloid resuscitation, especially in penetrating abdominal trauma, to prevent exacerbation of physiologically staunched severe bleeding and to prevent dilution of oxygen carrying capacity. Both of these concerns argue that lower systolic blood pressure resuscitation end points may be appropriate. Without denying the validity of either argument, the fact remains that epidemiologically, TBI patients with systolic blood pressure <90 have poorer outcomes. While further research is needed to determine if another cutoff might make more sense from a physiologic, mortality, or outcome point of view, for now the data support resuscitation of patients with suspected brain injury to 90 mm Hg and all efforts should be expended to assure that patient’s systolic blood pressures are kept at this level.
Pupillary Response. Pupillary asymmetry, the clinical manifestation of temporal lobe herniation, has high diagnostic and prognostic utility. Pupillary asymmetry is defined as a difference of >1 mm between pupils. A dilated pupil is >4 mm. A fixed pupil shows no response to bright light. Pupillary asymmetry and its duration should be carefully documented.
It should be remembered that multiple factors can create this finding. Hypotension, hypoxia, and direct orbital trauma are common causes of pupillary dilation. One iatrogenic cause is the belladonna alkaloids commonly used by ophthalmologists for detailed ophthalmologic examinations. Hypoxia and hypotension should be corrected as herniation is being excluded as the cause for pupillary dilatation. Orbital trauma can be excluded using a swinging light test that assesses the direct and consensual response of each pupil.
Until mass effect has been ruled out, pupillary dilatation should be assumed to be due to mass effect.
Glasgow Coma Score. An important part of the primary survey is to obtain an accurate GCS. The GCS12 is critical in classifying the severity of head injury and determining its subsequent management. Patients with a GCS of 14–15 are classified as having mild head injury; they have a 2% chance of elevated ICP, a 2% chance of any lesion on computed tomography (CT), and <0.1% chance of that lesion being surgically significant. Moderate head injuries have a GCS of 9–13, a 20% chance of elevated ICP, and an approximately 10% chance of having a lesion on CT scan. Severe head injuries need to be intubated and have an approximately 50% chance of having elevated ICP. Severely head-injured patients with a normal head CT do not need ICP monitoring unless they are in a high-risk group defined as having two of the following three characteristics: age >40, a history of hypotension (systolic blood pressure <90), or unilateral or bilateral motor posturing. Severe head injuries have a GCS of 3–8.–1214 Unfortunately, in as many as 44% of patients, a full GCS cannot be obtained, especially early in the course of care. Patients who are hypoxic, hypotensive, hypothermic, or hypoglycemic have depressed mental status due to a poor environment for the brain and not due to brain pathology. These conditions should be corrected prior to relying on the GCS for management decisions. Similarly, the common use of paralytics and sedatives in rapid sequence intubation introduces confounding factors that must clear prior to relying on the GCS.
Recent work has questioned the utility of the GCS for certain applications, suggesting other methods for classifying TBI. There also continues to be discussion about whether patients with GCS 13 should be treated as mild or moderate head injuries. It is the author’s practice to treat them as moderate.15
Penetrating Injuries
Penetrating Brain Injuries (PBI) to the head, particularly gunshot wounds to the head, can carry as high as 90% mortality. Decisions on who should be resuscitated can often be particularly difficult.
Important factors to assess in making this decision are the age of the patient, the circumstance of the injury, and the caliber of weapon. In addition, it is useful to classify Penetrating Brain Injury (PBI) as tangential, penetrating, or perforating injuries. Tangential injuries, which strike but do not enter the calvarium, have a lower mortality rate.16 A penetrating injury occurs when the projectile enters the calvarium, often driving bone before it into the brain, but remains lodged within the calvarium. A perforating injury occurs when the projectile also exits the brain, creating a tract completely across the head. Traditional PBI teaching has been that injuries that cross the midline are the most lethal, and some class III data support this.16,17
Victims of penetrating head trauma who present with a GCS of 3–5 have only a small chance of an acceptable outcome. At the same time, several studies have shown a reasonable prognosis for patients with PBI and GCS 13–15.16,18–20 It should be remembered that these assumptions are based on postresuscitation GCS.
Patients with a depressed respiratory rate or hypotension on presentation after penetrating trauma are likely near death and are at greater risk for a poorer outcome.
DETERMINING THE NEED FOR EMERGENCY SURGERY
Emergency Radiologic Studies
Cervical Spine Management. Patient who have sustained TBI are at a higher risk for cervical spine injury. Identification of patients who have sustained cervical trauma is crucial since they are at a greater risk of further, possibly, catastrophic injury. Conversely, the morbidity of a prolonged time in a cervical collar, such as decubiti and infection, and the impediments to care created by cervical collars make identification of patients who are at acceptable low risk for further injury to the cervical spine also important, since in these patients the collar can be removed. The goal is to have a process that is very sensitive to detecting high-risk cervical spine injuries and very specific, able to exclude patients who are not at high risk for a major cervical spine injury.
Traditionally, screening was done via imaging. The plain cervical x-rays have been estimated to be 92%–96% sensitive and 78%–97% specific, with some estimates running lower.21 CT has been estimated to be 96%–100% sensitive and 90%–100% specific for anatomical abnormalities.21,22 Imaging modalities, however, detect anatomical abnormalities. Judgment is still required to determine which of these abnormalities are substantive or constitute an increased risk to the patient. Judgment is required to determine which patients have an acceptable risk, that is, which patients are “cleared.”
Criteria other than imaging have been developed to identify low-risk patients. The National Emergency X-Radiography Utilization Study (NEXUS) criteria use characteristics of the history and physical to screen for low-risk patients. Per the NEXUS criteria, patients who can reliably answer questions and who are neurologically intact, without pain at rest, without pain with palpation of the neck and without pain or other symptoms with motion of the neck have a 99.8% chance of being free of cervical injury (negative predictive value) and may have their collar removed without imaging.23 Unfortunately, only 12% of presenting patients meet these criteria.
The Canadian C-Spine Rule adds features of the mechanism of injury to the examination and history to screen for high-risk patients. The Canadian C-Spine Rule has a 100% negative predictive value.24
The goal of all these techniques is to identify to the practitioner both patients at an acceptably low risk for further neurologic injury, that is, those who are “cleared,” and those who remain at increased risk and who warrant further protection and management of their cervical spine. Until this determination is made, all victims of TBI should have their cervical spine protected in a rigid cervical collar.
Computed Tomography of the Head
MILD HEAD INJURY
Since the overwhelming majority of head injuries that present are mild and insignificant, multiple organizations have released guidelines on which brain injury patients should undergo CT scanning, in an effort to limit its unnecessary utilization. The summary of these guidelines is that it is not necessary to obtain a head CT in patients who have no loss of consciousness and are neurologically normal. Problems arise, however, in defining what is meant by “neurologically normal.” All guidelines define this as the absence of posttraumatic amnesia (PTA), confusion, or impaired alertness.–2527 Some of these features may be present with a GCS of 15, depending on the method used for obtaining the GCS. While tests for PTA and screening tests for mild head injury are available, they are not in wide use in emergency rooms today and their utility in this busy environment is questionable.28 It is therefore difficult to reliably identify and document patients who may not require CT scanning. While it is possible to omit CT scanning in some patients, in most cases, it appears to be cost-effective and safer to triage head-injury patients, including mild head-injury patients, with CT.
TIMING OF CT
CT scanning should be obtained as early as is safely possible in the patient’s care. Patients should be adequately resuscitated prior to being taken to the CT scanner.26 In many urban trauma centers, CT scans can often be obtained within minutes of arrival in the ED, indeed within minutes of the injury. These “ultra early” CT scans can be obtained prior to substantial accumulation of intracranial blood or swelling. Note should be made of patients who receive “ultra early” scanning and subsequently decline in mental status. Such patients may warrant repeat scanning.
FEATURES ON CT
The focus of the CT examination is to identify intracranial hematomas. There are several other features of the scan that are important.
Compression of the basal cisterns is important to note. Basal cistern effacement is the anatomic correlate for progressing temporal lobe herniation. Effaced or compressed basal cisterns are a warning of progressing herniation. Absent basal cisterns are a grim marker of well-advanced herniation.
Midline shift is also important; its use as a criterion for removal of various hematomas is discussed elsewhere. It is important to note that midline shift is caused not only by hematomas but also by cerebral edema.
Traumatic subarachnoid blood is, in fact. While tSAH does not create significant mass effect, it is a prognostic marker for increased ICP and poorer outcome.136
CT scanning also allows good imaging of the skull and the skull base. Many fractures can be identified on CT. Particular attention should be paid to skull fractures in “ultra early” scans since they may portend delayed development of a hematoma.29
In penetrating head injury, perforating lesions carry a higher mortality with perforating lesions that cross the midline being the most lethal; these lesions can be seen on CT. One exception worth noting is bilateral frontal lobe involvement. Kaufman noted a mortality rate of 12% in this group and good outcomes in 30%, considerably better than the outcomes for bihemispheric lesions in general.30 Conversely, if the tract is further posterior in the brain, more critical structures will be damaged. Such a posterior tract is likely to traverse the ventricles and ventricular penetration by the tract has been shown to have a strong association with increased mortality.18,31,32
Because of the great reliance on CT scanning by many therapies for TBI, comparison of various therapies during research trials requires a standardized description of CT scans to allow classification and comparison of these scans. Marshall proposed such a classification in 1991.33
CT findings were divided into mass lesions and diffuse injuries. Mass lesions were further divided into evacuated (any lesion surgically evacuated) and not evacuated (high or mixed density lesion >25cc; not surgically evacuated).
Diffuse injury II, cisterns present with midline shift of 0-5 mm and no high or mixed density lesions >25cc but may include bone fragments and foreign bodies; Diffuse injury III, cisterns compressed or absent with midline shift of 0-5 mm and no high or mixed density lesions >25cc; Diffuse injury IV, midline shift >5 mm; no high or mixed density lesion >25cc.
The Marshall Scale has become a standard for classifying CT scans for research and clinical work. In addition, it has also been used as a tool to study the prognostic value of CT scans. This concept of performing risk analysis of various CT findings has been carried forward by several researchers.34 Knowledge of a particular injuries natural history via quantitative risk analysis provides a more precise tool for determining who might require surgery for mass lesion evacuation, ICP monitoring or other interventions.
Magnetic Resonance Imaging. As new magnetic resonance imaging (MRI) technologies evolve, the role of MRI in the evaluation of the head-injured patient is changing. While there is much interesting research in this area, there is little that is ready for routine clinical use.
SURGICAL MANAGEMENT
Removal of Mass Lesions
After resuscitation, the most important decision is that of surgical management. One important factor in patients with some form of intracranial bleeding is the volume of the hematoma. Many CT scanners will estimate the hematoma volume. If such an estimate is not available, hematoma volume can be estimated by a technique described by Kothari.35
Acute Epidural Hematoma. All epidural hematomas with a volume >30 cm2 need to be evacuated, regardless of the patient’s GCS. The criteria for nonoperative management are a volume on CT < 30 cm2, a thickness of <15 mm, and a midline shift <0.5 mm in a patient with a GCS > 8 and no focal deficit. All of these criteria should be met for the patient to be managed nonoperatively.36
Patients with an acute epidural hematoma, anisocoria, and a GCS < 9 should undergo craniotomy as soon as possible, regardless of the size of the hematoma.36
Acute Subdural Hematomas. For subdural hematomas, those with a thickness >10 mm or a midline shift >5 mm should be evacuated regardless of the patient’s GCS. A patient with an acute subdural hematoma that is <10 mm thick and midline shift <5 mm but with fixed and dilated or asymmetric pupils, an ICP > 20 mmHg, or a decline in GCS of 2 or more points from the time of injury to hospital admission should also have their hematoma removed. Patients with acute subdural hematomas also need to have their clots removed as soon as possible.37 Subdural hematomas should be removed using craniotomy. All patients with a GCS < 9 and an acute subdural hematoma should be monitored with an ICP monitor.37
Parenchymal Lesions. Parenchymal lesions consist of intraparenchymal clots and contusions. Their management has always been less clearly defined than the management of epidural and subdural hematomas
Focal parenchymal lesions should be removed in three circumstances. Any patient with a parenchymal mass lesion and signs of progressive neurologic deterioration due to the lesion, medically refractory intracranial hypertension, or signs of mass effect on CT scan should be treated operatively. Any patient with any lesion >50 cm3 in volume should be treated operatively. Patients with GCS scores of 6–8 with frontal or temporal contusions >20 cm3 in volume with midline shift of at least 5 mm and/or cisternal compression on CT scan should be treated operatively.38 Craniotomy with evacuation of mass lesion is recommended for these patients.38
Patients with parenchymal mass lesions who do not show evidence for neurologic compromise, have controlled ICP, and have no major signs of mass effect on CT scan may be managed nonoperatively with intensive monitoring and serial imaging.38
Posterior Fossa Lesions. Posterior fossa lesions are particularly dangerous. These lesions often do not manifest their mass effect by mental status change but rather by vital sign changes. These changes are often subtle and missed, with the ensuing tonsillar herniation often presenting as cardiopulmonary collapse.
Patients with mass effect on CT scan or with neurologic dysfunction or deterioration referable to the lesion should undergo operative intervention. Mass effect on CT scan is defined as distortion, dislocation, or obliteration of the fourth ventricle; compression or loss of visualization of the basal cisterns; or the presence of obstructive hydrocephalus. The operation should take place as soon as possible. A suboccipital craniectomy is the procedure most commonly performed.39
Patients with lesions and no major mass effect on CT scan and without signs of neurologic dysfunction may be managed by close observation and serial imaging.39
Surgical Management of Diffuse Brain Swelling—Decompressive Craniectomy
As with many techniques, the term decompressive craniectomy has been used by authors to refer to several different operations. A wide hemispheric craniectomy and bifrontal craniectomy with several different methods of dural opening have all been described.–4042 Clinical experience has created the impression that decompressive craniectomy can have significant clinical value. Its true clinical efficacy is awaiting the outcome of multiple current ongoing studies (ClinicalTrials.gov identifier NCT00155987).
The quandary in decompressive craniectomy is its use in treating isolated diffuse cerebral swelling. Decompressive craniectomy may be done incidental to a craniotomy for removal of hematoma, but the decision making in this case is driven by the need to remove the hematoma. The more difficult problem is the decision and timing of surgery where the only indication is diffuse cerebral swelling.
Currently, it is felt that decompressive craniectomy should not be a “last-ditch,” salvage procedure; rather, it should be a deliberate part of the treatment protocol that is aggressively invoked early in the patient’s care when lower-tier therapies fail. Expert guidelines suggest that for patients with diffuse cerebral swelling, bifrontal decompressive craniectomy within 48 hours of injury is a treatment option. These patients should have diffuse, medically refractory posttraumatic cerebral edema and resultant intracranial hypertension.38 In addition to bifrontal decompressive craniectomy, other decompressive procedures, including subtemporal decompression, temporal lobectomy, and hemispheric decompressive craniectomy, are treatment options for patients with refractory intracranial hypertension and diffuse parenchymal injury with clinical and radiographic evidence for impending transtentorial herniation.38
Depressed Skull Fractures
Patients with open (compound) depressed cranial fractures should undergo operative intervention to prevent infection and decompress the brain if clinical or radiographic evidence of dural penetration, major intracranial hematoma, bone depression >1 cm, frontal sinus involvement, gross cosmetic deformity, wound infection, pneumocephalus, or gross wound contamination is present. Nonoperative management is appropriate for patients without any of these findings.
Elevation of the fracture and debridement of the skull, scalp, and brain followed by closure of the dura is the surgical method of choice. Replacement of the bone at the time of surgery is appropriate, although the risk of infection must be considered when one is replacing bone fragments associated with an open wound. However, such replacement can be safe if thorough irrigation and debridement have been utilized. Antibiotics can be started on all patients with open (compound) depressed fractures.43 Early operation is recommended to reduce the incidence of infection.
Closed (simple) depressed cranial fractures that are less than the width of the skull deep may be treated nonoperatively.43
In children, ping-pong fractures, or large depressions on the convexity of the skull reminiscent of the depressions commonly seen on old ping-pong (table tennis) balls, sometimes require elevation if the depression is deeper than the thickness of the skull. These fractures may be elevated by simply drilling an adjacent burr hole, carefully sliding a stout instrument underneath the fracture, and levering the fragments outward.
Fractures at the base of the skull, which involve the frontal sinuses or ethmoid sinuses, are more complex to manage and often require a collaborative approach with otolaryngology, plastics, or oromaxillofacial surgery.
Penetrating Injuries
The goals of surgery for the victim of PBI are to remove mass effect, control bleeding, control infection, to prevent cerebrospinal fluid (CSF) leak and to close the scalp. Although advocated in the past, aggressive removal of all bone and bullet fragments is no longer a goal for this surgery.–4448
Full access? Get Clinical Tree
