Residency · Residency · Critical Care

Traumatic Brain Injury Management

Classification and Initial Assessment

GCS Classification

The Glasgow Coma Scale remains the foundational tool for classifying the severity of traumatic brain injury and guiding initial management decisions. Mild TBI, corresponding to GCS scores of 13 to 15, encompasses the concussion spectrum and typically does not require ICU-level care. Moderate TBI, with GCS scores of 9 to 12, represents an intermediate severity requiring close neurological monitoring and potential escalation. Severe TBI, defined by GCS scores of 3 to 8, mandates ICU admission and endotracheal intubation for airway protection, as these patients lack the capacity to protect their airway and are at risk for aspiration and respiratory failure. Among the individual components of the GCS, the motor score has been consistently shown to be the most predictive of outcomes, carrying more prognostic weight than the verbal or eye-opening components.

Primary vs. Secondary Brain Injury

The conceptual framework for TBI management rests upon the distinction between primary and secondary brain injury. Primary injury refers to the direct mechanical damage sustained at the moment of impact and includes cortical contusions, diffuse axonal injury, epidural and subdural hematomas, and skull fractures. This damage is instantaneous and irreversible by the time the patient reaches medical attention. Secondary injury, in contrast, refers to the cascade of subsequent insults that compound the initial damage and worsen outcomes. These include hypotension, hypoxemia, hyperthermia, hyperglycemia, seizures, elevated intracranial pressure, and cerebral edema. The entire focus of ICU management is directed at preventing and treating secondary injury, as this represents the only modifiable component of the disease process.

Imaging

Non-contrast computed tomography of the head is the first-line imaging modality in TBI, providing rapid identification of hemorrhage, mass effect, midline shift, hydrocephalus, and skull fractures. CT angiography should be obtained for suspected vascular injury, including dissection or pseudoaneurysm, in all patients with skull base fractures or cervical spine fractures traversing the transverse foramen. Magnetic resonance imaging is superior to CT for detecting diffuse axonal injury and brainstem injury and provides valuable prognostic information, but should be obtained only when the patient is sufficiently stable for transport to the scanner. The Marshall CT classification system categorizes injuries into diffuse injury grades I through IV and evacuated versus non-evacuated mass lesions, providing a framework that correlates with intracranial pressure and clinical outcomes.

Specific Lesion Types

Epidural hematoma presents as a lens-shaped (biconvex) collection, most commonly from rupture of the middle meningeal artery. The classic "lucid interval," in which a patient initially appears well before rapidly deteriorating, occurs in 20 to 50 percent of cases. Surgical evacuation is indicated when the hematoma exceeds 30 mL in volume or 15 mm in thickness, or when midline shift exceeds 5 mm. Subdural hematoma appears as a crescent-shaped collection from rupture of bridging veins and requires surgical evacuation when thickness exceeds 10 mm, midline shift exceeds 5 mm, or the GCS score drops by 2 or more points. Traumatic subarachnoid hemorrhage is common in TBI and is associated with delayed cerebral vasospasm, which should be monitored with transcranial Doppler ultrasonography. Cerebral contusions represent hemorrhagic parenchymal injuries that may expand significantly in the first 24 to 48 hours, a phenomenon known as "blossoming contusion" that necessitates serial imaging. Diffuse axonal injury results from rotational shearing forces at the gray-white matter junction, with characteristic involvement of the corpus callosum, dorsolateral brainstem, and subcortical white matter.

ICP Management

ICP Monitoring

Intracranial pressure monitoring is indicated in patients with severe TBI (GCS 3-8) who have an abnormal CT scan, or in those with a normal CT scan who have two or more of the following risk factors: age greater than 40 years, motor posturing, or systolic blood pressure below 90 mmHg. The external ventricular drain (EVD) is the gold standard for ICP monitoring because it provides both continuous pressure measurement and the ability to perform therapeutic CSF drainage. EVDs are typically placed at Kocher's point, located 1 cm anterior to the coronal suture in the mid-pupillary line. The infection risk of EVDs ranges from 5 to 15 percent and can be reduced through the use of antibiotic-impregnated catheters and strict aseptic technique. Intraparenchymal monitors, such as the Codman or Camino devices, use fiber-optic or strain gauge technology and provide accurate ICP measurement, though they lack the capability for therapeutic CSF drainage and cannot be recalibrated once placed.

Normal ICP is below 15 mmHg. The Brain Trauma Foundation (BTF) guidelines, in their fourth edition published in 2016, established a treatment threshold of ICP greater than 22 mmHg, updated from the previous threshold of 20 mmHg based on additional data correlating this threshold with outcomes.

Cerebral Perfusion Pressure (CPP)

Cerebral perfusion pressure, calculated as the difference between mean arterial pressure and intracranial pressure (CPP = MAP - ICP), is a critical determinant of cerebral blood flow and oxygen delivery. The BTF guidelines recommend targeting a CPP of 60 to 70 mmHg. CPP values below 50 mmHg are consistently associated with cerebral ischemia and worse outcomes. However, aggressive augmentation of CPP above 70 mmHg through fluid loading and vasopressors should be avoided, as Robertson et al. demonstrated that this approach increases the risk of acute respiratory distress syndrome without improving neurological outcomes. In severe TBI, cerebral autoregulation is frequently impaired, meaning that cerebral blood flow becomes linearly dependent on CPP rather than being maintained across a range of perfusion pressures, making CPP management even more critical.

Multimodal Monitoring

The era of single-parameter ICP-guided management is evolving toward multimodal neuromonitoring that integrates multiple physiological variables to provide a more comprehensive picture of brain physiology. Brain tissue oxygenation monitoring using the Licox probe measures partial pressure of oxygen in brain tissue (PbtO2), with normal values of 20 to 35 mmHg and a treatment target above 20 mmHg. The BOOST-3 trial (2024) compared PbtO2-guided therapy to ICP-guided therapy alone and demonstrated a trend toward improved functional outcomes, a finding that is likely to be practice-changing in the field of neurocritical care.

Cerebral microdialysis measures interstitial concentrations of lactate, pyruvate, glucose, glutamate, and glycerol, with a lactate-to-pyruvate ratio exceeding 40 indicating cerebral metabolic crisis. Although primarily a research tool, microdialysis is trending toward broader clinical application. Jugular venous oximetry measures oxygen saturation in the jugular bulb (SjvO2), with normal values of 55 to 75 percent; values below 50 percent indicate cerebral ischemia, while values above 75 percent suggest hyperemia. Transcranial Doppler provides information about pulsatility index, mean flow velocity, and vasospasm detection. Near-infrared spectroscopy (NIRS) enables non-invasive trending of regional cerebral oxygenation, though its utility is limited by shallow tissue penetration.

<image>Tiered ICP management algorithm. Top: "ICP >22 mmHg confirmed." Tier 0 (baseline measures): HOB 30 degrees, head midline, avoid jugular compression, sedation/analgesia, temperature control, seizure prophylaxis. Tier 1: CSF drainage via EVD (10-20 mL increments), osmotherapy (mannitol or hypertonic saline), optimize CPP 60-70 mmHg. Tier 2: if refractory — mild hyperventilation (PaCO2 30-35 mmHg, temporary only), deepen sedation (propofol/midazolam), neuromuscular blockade. Tier 3: if still refractory — decompressive craniectomy or barbiturate coma (pentobarbital infusion to burst-suppression). Each tier shows specific interventions with doses, monitoring parameters, and criteria for escalation. Side panel showing CPP calculation, PbtO2 targets, and warning signs requiring immediate intervention. Include red flags for herniation: unilateral pupil dilation, posturing, Cushing triad.</image>

Tiered ICP Management

General Measures (Tier 0)

The foundation of ICP management consists of general measures that should be applied to all patients with severe TBI. Head of bed elevation to 30 degrees reduces ICP by promoting venous drainage, and the head should be maintained in a neutral midline position to prevent compression of the jugular veins. Cervical collars that compress the jugular veins should be replaced with towel rolls for cervical stabilization if the C-spine has not been cleared. Adequate sedation and analgesia, typically with propofol combined with fentanyl or remifentanil, reduce metabolic demand and prevent ICP spikes from agitation or pain.

Strict normothermia with a temperature target below 38 degrees Celsius is essential, as fever increases cerebral metabolic rate and can elevate ICP by 10 to 15 percent per degree of temperature elevation, necessitating active cooling measures when needed. Glucose should be maintained at 140 to 180 mg/dL, avoiding both hypoglycemia below 80 mg/dL (which causes secondary neuronal injury) and hyperglycemia above 180 mg/dL (which worsens cerebral edema and inflammatory injury). Seizure prophylaxis with levetiracetam 1000 mg every 12 hours or phenytoin for 7 days reduces the incidence of early post-traumatic seizures, though as demonstrated by Temkin et al., prophylaxis does not prevent the development of late post-traumatic epilepsy. Ventilation should target normocarbia with a PaCO2 of 35 to 40 mmHg, avoiding hyperventilation unless the patient is acutely herniating. Normovolemia must be maintained, and hypotension vigorously prevented, as even a single episode of systolic blood pressure below 90 mmHg doubles mortality in severe TBI. Any coagulopathy should be corrected to target an INR below 1.5, platelets above 100,000/mcL, and fibrinogen above 200 mg/dL.

Osmotherapy (Tier 1)

Osmotherapy forms the first escalation step when ICP exceeds the treatment threshold despite general measures. Mannitol, administered as a 0.25 to 1 g/kg IV bolus of the 20 percent solution, acts as an osmotic diuretic that draws water from brain tissue into the intravascular space. Its onset is 15 to 30 minutes with a duration of 4 to 6 hours. Serum osmolality should be monitored and maintained below 320 mOsm/L, with the osmolar gap kept below 20. Repeated dosing with inadequate fluid replacement carries risks of hypovolemia and renal failure, and mannitol should not be used in hypotensive patients as the associated diuresis will exacerbate hypovolemia.

Hypertonic saline is available in several concentrations, with 23.4 percent (30 mL bolus via central venous catheter) and 3 percent (250 mL bolus or continuous infusion at 0.5 to 2 mL/kg/hr) being most commonly used. Serum sodium should be targeted to 145 to 155 mEq/L and should not exceed 160 mEq/L. Advantages of hypertonic saline over mannitol include the absence of diuresis (making it preferable in hypovolemic patients) and potentially superior efficacy for refractory ICP elevation. The COBI trial (2021) compared 20 percent hypertonic saline bolus to mannitol for ICP episodes and found no significant difference in ICP control, though more recent studies have suggested a modest advantage for hypertonic saline. Sodium levels must be monitored closely, as rapid fluctuations risk osmotic demyelination syndrome.

CSF Drainage (Tier 1)

External ventricular drainage provides both diagnostic and therapeutic capability. The EVD can be set to drain at a specified level (typically 10 to 15 cmH2O above the tragus), with intermittent drainage of 5 to 10 mL of CSF per ICP elevation episode, each potentially reducing ICP by 5 to 10 mmHg. Continuous drainage maintains ICP below a set threshold but precludes accurate continuous ICP monitoring when the drain is open, creating a clinical tension between therapeutic drainage and monitoring that must be managed through intermittent clamping and assessment.

Hyperventilation (Tier 2)

Hyperventilation to a PaCO2 of 30 to 35 mmHg is a temporizing measure reserved exclusively for acute herniation events. The mechanism involves hypocapnia-induced cerebral vasoconstriction, which reduces cerebral blood volume and thereby lowers ICP. However, the same vasoconstriction that reduces ICP also reduces cerebral blood flow, creating a significant risk of cerebral ischemia. This risk can be detected by monitoring PbtO2 or SjvO2, which will demonstrate falling values when vasoconstriction becomes excessive. Hyperventilation should be limited to minutes to hours while preparing for definitive intervention and should never be used as a sustained ICP management strategy.

Barbiturate Coma (Tier 3)

Pentobarbital coma, with a loading dose of 5 to 10 mg/kg followed by an infusion of 1 to 4 mg/kg/hr, represents a last-resort intervention for ICP that remains refractory to all other measures. The therapeutic target is burst-suppression on continuous EEG, typically 3 to 5 bursts per minute, which reflects maximal suppression of cerebral metabolic rate with consequent reduction in cerebral blood flow and ICP. Complications are severe and include profound hypotension requiring vasopressor support, immunosuppression increasing infection risk, paralytic ileus, and complete loss of the clinical neurological examination. EEG monitoring is mandatory during barbiturate coma, and this intervention should be used only after all other measures have been exhausted.

Decompressive Craniectomy (Tier 3)

Decompressive craniectomy has been evaluated in two major randomized trials that produced sobering results requiring careful interpretation. The DECRA trial (2011) examined early bifrontotemporoparietal decompressive craniectomy for diffuse TBI with refractory ICP and found that while craniectomy effectively reduced ICP, it was associated with worse functional outcomes, including more unfavorable Glasgow Outcome Scale-Extended scores. The RESCUEicp trial (2016) studied craniectomy for ICP refractory to medical management at a threshold of 25 mmHg and found reduced mortality (26 percent versus 49 percent) but increased rates of severe disability, with many survivors remaining in a vegetative or severely disabled state.

These trials establish that craniectomy should be considered as a last resort, with patient selection being critical. Goals of care discussions with the family should be conducted before proceeding, ensuring that the possibility of survival with severe disability is clearly communicated.

Medical Management

Blood Pressure and Hemodynamics

Prevention of hypotension is the single most impactful intervention in TBI management. A single episode of systolic blood pressure below 90 mmHg doubles mortality in severe TBI, making this the most dangerous modifiable secondary insult. MAP should be targeted to achieve CPP of 60 to 70 mmHg, using norepinephrine as the preferred vasopressor for MAP augmentation. Fluid resuscitation should use isotonic crystalloid (normal saline or balanced solutions), and hypotonic fluids must be strictly avoided as they exacerbate cerebral edema.

Coagulopathy and Anticoagulation Reversal

Trauma-induced coagulopathy is present in 30 to 40 percent of patients with severe TBI on arrival and significantly increases the risk of hemorrhage expansion. For patients on warfarin, 4-factor prothrombin complex concentrate (25 to 50 units/kg IV) combined with vitamin K 10 mg IV should be administered with the goal of achieving an INR below 1.5 within 30 minutes. For patients on direct oral anticoagulants, idarucizumab (5 g IV) reverses dabigatran, and andexanet alfa is available for factor Xa inhibitors; if specific antidotes are unavailable, 4-factor PCC should be used. The role of platelet transfusion for patients on antiplatelet agents remains controversial, though desmopressin at 0.3 mcg/kg may improve platelet function.

The CRASH-3 trial (2019) provided important guidance on tranexamic acid in TBI. Administration of TXA within 3 hours of injury reduced head injury-related death in patients with mild-to-moderate TBI (GCS 9-15) and intracranial hemorrhage. However, no benefit was demonstrated in severe TBI (GCS 3-8) or with late administration.

DVT Prophylaxis

Patients with severe TBI are at high risk for venous thromboembolism due to immobility, the injury itself, and frequently associated lower extremity fractures. Pharmacological prophylaxis should be initiated within 24 to 48 hours of injury if there is no expanding intracranial hemorrhage. A repeat CT scan at 24 hours confirming hemorrhage stability should prompt initiation of enoxaparin 40 mg subcutaneously daily, with LMWH preferred over unfractionated heparin. Mechanical prophylaxis with intermittent pneumatic compression should be applied from the time of admission until pharmacological prophylaxis is established.

Nutrition

Early enteral nutrition within 24 to 48 hours of injury has been associated with reduced mortality in TBI patients and should be initiated as soon as the patient is hemodynamically stable. The hypermetabolic and catabolic state of severe TBI demands high protein intake at 1.5 to 2.0 g/kg/day, with a caloric target of 25 to 30 kcal/kg/day, ideally guided by indirect calorimetry when available.

Herniation Syndromes

Recognition

Rapid recognition of herniation syndromes is essential for initiating life-saving emergency treatment. Uncal (transtentorial) herniation classically presents with ipsilateral pupil dilation from compression of cranial nerve III, followed by contralateral hemiparesis, progressing to bilateral fixed and dilated pupils with decerebrate posturing. Central herniation produces bilateral pupillary constriction that progresses to bilateral dilation, with respiratory pattern changes from Cheyne-Stokes to apneustic breathing. Tonsillar (cerebellar) herniation, in which the cerebellar tonsils herniate through the foramen magnum, produces sudden apnea and cardiovascular collapse and is often rapidly fatal. The Cushing triad of hypertension, bradycardia, and irregular respirations is a late sign of critically elevated ICP and impending herniation.

Emergency Management

Emergency treatment of herniation requires simultaneous deployment of all available temporizing measures. Osmotherapy with mannitol 1 g/kg IV bolus or 23.4 percent hypertonic saline 30 mL via central venous catheter should be administered immediately. Brief hyperventilation targeting a PaCO2 of 28 to 32 mmHg provides temporary ICP reduction while definitive intervention is prepared. The head of bed should be elevated to 30 degrees, and an emergent neurosurgical consultation obtained for potential evacuation of a mass lesion or decompressive craniectomy. If an EVD is in place, CSF should be drained aggressively.

Key Clinical Pearls

• Preventing secondary brain injury is the primary goal of TBI ICU management — a single episode of hypotension (SBP <90) doubles mortality
• ICP treatment threshold is 22 mmHg (BTF 4th edition); target CPP 60-70 mmHg
• Hypertonic saline and mannitol are both effective osmotherapeutic agents — HTS may be preferred in hypovolemic patients (no diuresis)
• Hyperventilation is ONLY for acute herniation — sustained hyperventilation causes cerebral ischemia
• Decompressive craniectomy reduces mortality but increases severe disability (RESCUEicp) — goals of care discussion is essential before proceeding
• Tranexamic acid within 3 hours benefits mild-moderate TBI (CRASH-3) but not severe TBI or late administration
• Early seizure prophylaxis (7 days) reduces early seizures but does NOT prevent late post-traumatic epilepsy
• PbtO2-guided management (target >20 mmHg) is emerging as a practice-changing addition to ICP-guided therapy (BOOST-3)

References

1. Carney N, Totten AM, O'Reilly C, et al. Guidelines for the management of severe traumatic brain injury, fourth edition. Neurosurgery. 2017;80(1):6-15.
2. Hutchinson PJ, Kolias AG, Timofeev IS, et al. Trial of decompressive craniectomy for traumatic intracranial hypertension. N Engl J Med. 2016;375(12):1119-1130.
3. Cooper DJ, Rosenfeld JV, Murray L, et al. Decompressive craniectomy in diffuse traumatic brain injury. N Engl J Med. 2011;364(16):1493-1502.
4. CRASH-3 Trial Collaborators. Effects of tranexamic acid on death, disability, vascular occlusive events and other morbidities in patients with acute traumatic brain injury (CRASH-3). Lancet. 2019;394(10210):1713-1723.
5. Okonkwo DO, Shutter LA, Moore C, et al. Brain oxygen optimization in severe traumatic brain injury phase-3 (BOOST-3). N Engl J Med. 2024 (published online).

Read this lecture as Markdown