Skull fractures

Overview 
Theory 
Diagnosis 
Management 
Follow up 
Resources 

Please note that formulations/routes and doses may differ between drug names and brands, drug formularies, or locations. Treatment recommendations are specific to patient groups: see disclaimer

INITIAL

suspected skull fracture (any type)

1st line – stabilisation and supportive care

Back

1st line –

stabilisation and supportive care

Take an Airway, Breathing, Circulation ( ABC) approach to assessing and stabilising any patient who has sustained a head injury.[28] See Assessment of traumatic brain injury, acute.

If you suspect cervical spine injury, ensure full cervical spine immobilisation.[28] See Acute cervical spine trauma.

Manage pain effectively because it can lead to a rise in intracranial pressure.[28] Follow local protocols for pain management.

Provide reassurance, splint limb fractures, and catheterise a full bladder when needed.[28]

Do not routinely give other medical interventions (e.g., anticonvulsant and antibiotic prophylaxis) for isolated skull fractures.

Anticonvulsants, when given, are usually on the recommendation of a neurosurgeon for associated underlying intracranial injury such as subarachnoid haemorrhage, subdural/epidural haemorrhage, or intra-parenchymal haemorrhage. In such patients, anticonvulsants are given for the first 7 days post-injury to prevent early traumatic brain injury-associated seizures. There are no data to support prolonged antiseizure prophylaxis in the absence of documented seizures post-injury.[65]
There is little definitive evidence of benefit for antibiotics in decreasing the risk of subsequent meningitis or other infections in fractures, with or without cerebrospinal fluid (CSF) leak.[66][67][68][69] In practice, antibiotics are frequently used if there is an open skull fracture.
Pneumococcal vaccination is recommended for patients with a basilar skull fracture and a CSF leak.[70]

Children rarely require surgery. However, those with frontal skull fractures may be more likely to require operative repair.[60]

ACUTE

confirmed closed non-depressed fracture

1st line – observation and monitoring

Back

1st line –

observation and monitoring

Treat most non-depressed (linear) fractures, including basilar skull fractures, conservatively as long as:

There is no suspicion or evidence of intracranial pathology
Neurological status is normal
There is no evidence of cranial nerve damage or cerebrospinal fluid (CSF) leak.

Conservative treatment consists of observation to rule out any ongoing complications such as CSF leak, seizure, or infection.

Practical tip

Observations should be made every 30 minutes until the Glasgow Coma Scale (GCS) score has returned to 15, including the following as minimum:[28]

GCS plus pupil size and reactivity
Limb movements
Respiratory rate
Heart rate
Blood pressure
Temperature
Blood oxygen saturation.

For patients with a GCS of 15, the frequency of observations (starting after the initial assessment) should be:[28]

Half-hourly for 2 hours, then
1-hourly for 4 hours, then
2-hourly thereafter.

Revert to every 30 minutes if the patient deteriorates at any time after the initial 2-hour period.[28]

Observation of admitted infants and children aged <5 years should only be performed by experienced staff.[28]

Reassess the patient and consider an immediate computed tomography (CT) scan if there are any signs of neurological deterioration such as:[28]

Development of agitation or abnormal behaviour
A sustained (for at least 30 minutes) drop of 1 point in GCS score (give greater weight to a drop of 1 point in the motor response score of the GCS)
Any drop of ≥3 points in the eye-opening or verbal response scores of the GCS, or ≥2 points in the motor response score
Development of severe or increasing headache or persistent vomiting
New or evolving neurological symptoms or signs, such as pupil inequality or asymmetry of limb or facial movement.

A supervising doctor should conduct the appraisal.[28]

If after 24 hours of observation the patient has not achieved a GCS score of 15 and has had a normal CT scan, consider further CT scan or MRI scanning and discuss with the radiology department.[28]

Consider – tranexamic acid

Back

Consider –

tranexamic acid

Additional treatment recommended for SOME patients in selected patient group

Tranexamic acid should be used as soon as possible in patients with major trauma and active or suspected active bleeding.[59]

In the absence of major trauma and active or suspected active bleeding, consider tranexamic acid as soon as possible within 2 hours of the injury, in the pre-hospital or hospital setting and before imaging, in patients with a head injury and a GCS score of 12 or less.[28]

Bear in mind that this is standard practice in some countries, including the UK, but not a routine approach worldwide.

Evidence: Tranexamic acid

In the UK, the National Institute for Health and Care Excellence (NICE) does not make a recommendation for the use of tranexamic acid in people with mild traumatic brain injury, stating that more research is required to assess whether the benefits outweigh the risk of blood clots in this population. NICE does make a recommendation to consider tranexamic acid in patients with a head injury and a GCS score of 12 or less who are not thought to have active extracranial bleeding.

As part of the 2023 update of the NICE head injury guideline, two randomised controlled trials (RCTs) were identified assessing the use of tranexamic acid in adults with traumatic brain injury and no suspicion of extracranial bleeding.[28][55][61] NICE did not find any evidence for children or infants.

Adults

After reviewing the evidence from these two RCTs, the NICE committee made a weak recommendation to consider giving these patients tranexamic acid as soon as possible within 2 hours of injury, but only if the GCS score is ≤12.[28]

The first RCT was the CRASH-3 trial, a large, pragmatic RCT of tranexamic acid versus placebo in adults (aged 16 years and over) with traumatic brain injury and a GCS score of ≤12 or any intra-cranial bleed on CT scan and no extra-cranial bleeding (n=12,737, mean age 42 years, 175 centres in 29 countries).[62][63]

A regimen of initial tranexamic acid intravenous infusion given over 10 minutes followed by a second over 8 hours was compared with a saline placebo.
According to the trial protocol, patients had to be treated within 8 hours of injury; however, this was changed to less than 3 hours from injury part way through the trial.
Overall 9202 (72%) of participants were enrolled within 3 hours.
28% of participants had a mild traumatic brain injury (33% moderate and 38% severe).
There was no significant difference in the primary outcome of head injury-related death at 28 days (18.5% tranexamic acid vs. 19.8% placebo; relative risk [RR] 0.94, 95% CI 0.86 to 1.02, moderate quality as assessed by GRADE).
There was also no significant difference when patients with severe injury (GCS score 3 or bilateral unreactive pupils) were excluded from the analysis, or in all-cause mortality for the whole population.
There were, however, fewer head injury-related deaths within the first 24 hours (RR 0.81, 95% CI 0.69 to 0.95).
There were also fewer head-injury deaths at 28 days in the subgroup of patients with less severe injuries (GCS score 9-15: 5.8% tranexamic acid vs. 7.5% placebo; RR 0.78, 95% CI 0.62 to 0.95; patients with bilateral reactive pupils: 11.5% tranexamic acid versus 13.2% placebo; RR 0.87, 95% CI 0.77 to 0.98).
There was no evidence of increased complications or adverse events including vascular occlusive events, seizures, sepsis, or renal failure.
CRASH-3 did not report outcomes separately for patients with mild and moderate traumatic brain injury. Correspondence from the study authors, requested by the NICE committee, suggested significant uncertainty about the effectiveness of tranexamic acid in people with mild traumatic brain injury, although no additional data was shared.[28]

The second RCT (n=1280, 20 centres and 39 emergency medical services agencies in the US and Canada) compared a single bolus of tranexamic acid with placebo within 2 hours of traumatic brain injury given in a pre-hospital setting.[61]

4% of participants had a mild traumatic brain injury (39% moderate and 57% severe). Subgroup results for outcomes by severity of injury were not reported.
The evidence suggested a reduced all-cause mortality with tranexamic acid compared to placebo at 28 days (absolute effect 49 fewer per 1000, 95% CI 89 fewer to 9 more) and at 6 months (absolute effect 40 fewer per 1000, 95% CI 87 fewer to 30 more).
There was no clinically important difference in hospital-free days at 28 days, neurosurgical intervention at 28 days, degree of disability at discharge and 6 months (measured by Glasgow Outcome Scale-Extended >4) or serious adverse events (myocardial infarction, pulmonary embolism, deep vein thrombosis, or thrombotic stroke).
The NICE 2023 guideline did not make a separate recommendation for older adults as they were awaiting results from the CRASH-4 trial looking specifically at this population.[28][64]

Children and infants

Due to the lack of evidence in children and infants, the NICE guideline committee extrapolated from the evidence in adults to people aged <16 years in the same recommendation, just with separate advice regarding dosage.[28]

Primary options

tranexamic acid: children ≥1 month of age: 15-30 mg/kg intravenously as a single dose given as soon as possible within 2 hours of the injury, maximum 2 g/dose; adolescents ≥16 years of age and adults: 2 g intravenously as a single dose given as soon as possible within 2 hours of the injury

confirmed closed depressed fracture

1st line – observation and monitoring

Back

1st line –

observation and monitoring

Treat conservatively first-line.

Operative elevation and repair offer little benefit in terms of reduction in risk of seizure, infection, or neurological deficit.

Conservative treatment consists of observation to rule out any ongoing complications such as CSF leak, seizure, or infection.

Practical tip

Observations should be made every 30 minutes until the Glasgow Coma Scale (GCS) score has returned to 15, including the following as minimum:[28]

GCS plus pupil size and reactivity
Limb movements
Respiratory rate
Heart rate
Blood pressure
Temperature
Blood oxygen saturation.

For patients with a GCS of 15, the frequency of observations (starting after the initial assessment) should be:[28]

Half-hourly for 2 hours, then
1-hourly for 4 hours, then
2-hourly thereafter.

Revert to every 30 minutes if the patient deteriorates at any time after the initial 2-hour period.[28]

Observation of admitted infants and children aged <5 years should only be performed by experienced staff.[28]

Reassess the patient and consider an immediate computed tomography (CT) scan if there are any signs of neurological deterioration such as:[28]

Development of agitation or abnormal behaviour
A sustained (for at least 30 minutes) drop of 1 point in GCS score (give greater weight to a drop of 1 point in the motor response score of the GCS)
Any drop of ≥3 points in the eye-opening or verbal response scores of the GCS, or ≥2 points in the motor response score
Development of severe or increasing headache or persistent vomiting
New or evolving neurological symptoms or signs, such as pupil inequality or asymmetry of limb or facial movement.

A supervising doctor should conduct the appraisal.[28]

If after 24 hours of observation the patient has not achieved a GCS score of 15 and has had a normal CT scan, consider further CT scan or MRI scanning and discuss with the radiology department.[28]

Consider – tranexamic acid

Back

Consider –

tranexamic acid

Additional treatment recommended for SOME patients in selected patient group

Tranexamic acid should be used as soon as possible in patients with major trauma and active or suspected active bleeding.[59]

Bear in mind that this is standard practice in some countries, including the UK, but not a routine approach worldwide.

Evidence: Tranexamic acid

Adults

A regimen of initial tranexamic acid intravenous infusion given over 10 minutes followed by a second over 8 hours was compared with a saline placebo.
According to the trial protocol, patients had to be treated within 8 hours of injury; however, this was changed to less than 3 hours from injury part way through the trial.
Overall 9202 (72%) of participants were enrolled within 3 hours.
28% of participants had a mild traumatic brain injury (33% moderate and 38% severe).
There was no significant difference in the primary outcome of head injury-related death at 28 days (18.5% tranexamic acid vs. 19.8% placebo; relative risk [RR] 0.94, 95% CI 0.86 to 1.02, moderate quality as assessed by GRADE).
There was also no significant difference when patients with severe injury (GCS score 3 or bilateral unreactive pupils) were excluded from the analysis, or in all-cause mortality for the whole population.
There were, however, fewer head injury-related deaths within the first 24 hours (RR 0.81, 95% CI 0.69 to 0.95).
There were also fewer head-injury deaths at 28 days in the subgroup of patients with less severe injuries (GCS score 9-15: 5.8% tranexamic acid vs. 7.5% placebo; RR 0.78, 95% CI 0.62 to 0.95; patients with bilateral reactive pupils: 11.5% tranexamic acid versus 13.2% placebo; RR 0.87, 95% CI 0.77 to 0.98).
There was no evidence of increased complications or adverse events including vascular occlusive events, seizures, sepsis, or renal failure.
CRASH-3 did not report outcomes separately for patients with mild and moderate traumatic brain injury. Correspondence from the study authors, requested by the NICE committee, suggested significant uncertainty about the effectiveness of tranexamic acid in people with mild traumatic brain injury, although no additional data was shared.[28]

4% of participants had a mild traumatic brain injury (39% moderate and 57% severe). Subgroup results for outcomes by severity of injury were not reported.
The evidence suggested a reduced all-cause mortality with tranexamic acid compared to placebo at 28 days (absolute effect 49 fewer per 1000, 95% CI 89 fewer to 9 more) and at 6 months (absolute effect 40 fewer per 1000, 95% CI 87 fewer to 30 more).
There was no clinically important difference in hospital-free days at 28 days, neurosurgical intervention at 28 days, degree of disability at discharge and 6 months (measured by Glasgow Outcome Scale-Extended >4) or serious adverse events (myocardial infarction, pulmonary embolism, deep vein thrombosis, or thrombotic stroke).
The NICE 2023 guideline did not make a separate recommendation for older adults as they were awaiting results from the CRASH-4 trial looking specifically at this population.[28][64]

Children and infants

Primary options

Consider – dural repair and cranioplasty

Back

Consider –

dural repair and cranioplasty

Additional treatment recommended for SOME patients in selected patient group

Consider operative elevation and repair of dura and cranioplasty if the patient has any one of:[5][11][55][71][72][73][74]

A depression >1 cm or more than the thickness of the skull
Gross cosmetic deformity
An associated operable intracranial lesion
Persistent CSF leakage despite conservative management.
- See persistent cranial nerve injury or CSF leakage section below.

Consider – prophylactic anticonvulsant therapy

Back

Consider –

prophylactic anticonvulsant therapy

Additional treatment recommended for SOME patients in selected patient group

Post-traumatic seizures (PTS) are common following severe traumatic brain injury, and risk of PTS is significantly increased even following mild and moderate brain injury.[11][82][83]

Consider short-term anticonvulsants for prevention of PTS. There is some evidence to support the use of anticonvulsants in preventing PTS in the first 7 days following a head injury with associated underlying intracranial injury.[82][84][85] [ ] Evidence particularly supports phenytoin, but levetiracetam may also be used. The decision is usually made by a neurosurgeon.

Anticonvulsants have not been shown to have any effect on decreasing the risk of late PTS (≥8 days) or post-traumatic epilepsy, and their use beyond the first week post-injury is not supported or recommended.[11][82][83][84][85]

Do not use anticonvulsants for either early or late PTS prevention in isolated skull fracture in the absence of underlying brain injury and outside of severe depressed skull fracture.

For patients who continue to have seizures and carry the diagnosis of post-traumatic epilepsy, treat as you would epilepsy of non-traumatic origin.[11][84][85] See Focal seizures and Generalised seizures.

Only consider prophylactic anticonvulsant therapy for open depressed skull fractures or fractures associated with an underlying brain injury. It is not indicated or recommended for simple isolated skull fractures. If a seizure occurs, treat therapeutically (as with any non-traumatic seizure) with benzodiazepines and subsequent anticonvulsant medication.

Primary options

phenytoin: children and adults: consult specialist for guidance on dose

levetiracetam: children and adults: consult specialist for guidance on dose

confirmed open fracture

1st line – observation and monitoring

Back

1st line –

observation and monitoring

Treat conservatively if there is no intracranial haemorrhage, cerebrospinal fluid (CSF) leak, gross cosmetic deformity, or gross contamination.[11][72][73][74] Conservative treatment consists of observation to rule out any ongoing complications such as CSF leak, seizure, or infection.

Practical tip

Observations should be made every 30 minutes until the Glasgow Coma Scale (GCS) score has returned to 15, including the following as minimum:[28]

GCS plus pupil size and reactivity
Limb movements
Respiratory rate
Heart rate
Blood pressure
Temperature
Blood oxygen saturation.

For patients with a GCS of 15, the frequency of observations (starting after the initial assessment) should be:[28]

Half-hourly for 2 hours, then
1-hourly for 4 hours, then
2-hourly thereafter.

Revert to every 30 minutes if the patient deteriorates at any time after the initial 2-hour period.[28]

Observation of admitted infants and children aged <5 years should only be performed by experienced staff.[28]

Reassess the patient and consider an immediate computed tomography (CT) scan if there are any signs of neurological deterioration such as:[28]

Development of agitation or abnormal behaviour
A sustained (for at least 30 minutes) drop of 1 point in GCS score (give greater weight to a drop of 1 point in the motor response score of the GCS)
Any drop of ≥3 points in the eye-opening or verbal response scores of the GCS, or ≥2 points in the motor response score
Development of severe or increasing headache or persistent vomiting
New or evolving neurological symptoms or signs, such as pupil inequality or asymmetry of limb or facial movement.

A supervising doctor should conduct the appraisal.[28]

If after 24 hours of observation the patient has not achieved a GCS score of 15 and has had a normal CT scan, consider further CT scan or MRI scanning and discuss with the radiology department.[28]

Consider – tranexamic acid

Back

Consider –

tranexamic acid

Additional treatment recommended for SOME patients in selected patient group

Tranexamic acid should be used as soon as possible in patients with major trauma and active or suspected active bleeding.[59]

Bear in mind that this is standard practice in some countries, including the UK, but not a routine approach worldwide.

Evidence: Tranexamic acid

Adults

A regimen of initial tranexamic acid intravenous infusion given over 10 minutes followed by a second over 8 hours was compared with a saline placebo.
According to the trial protocol, patients had to be treated within 8 hours of injury; however, this was changed to less than 3 hours from injury part way through the trial.
Overall 9202 (72%) of participants were enrolled within 3 hours.
28% of participants had a mild traumatic brain injury (33% moderate and 38% severe).
There was no significant difference in the primary outcome of head injury-related death at 28 days (18.5% tranexamic acid vs. 19.8% placebo; relative risk [RR] 0.94, 95% CI 0.86 to 1.02, moderate quality as assessed by GRADE).
There was also no significant difference when patients with severe injury (GCS score 3 or bilateral unreactive pupils) were excluded from the analysis, or in all-cause mortality for the whole population.
There were, however, fewer head injury-related deaths within the first 24 hours (RR 0.81, 95% CI 0.69 to 0.95).
There were also fewer head-injury deaths at 28 days in the subgroup of patients with less severe injuries (GCS score 9-15: 5.8% tranexamic acid vs. 7.5% placebo; RR 0.78, 95% CI 0.62 to 0.95; patients with bilateral reactive pupils: 11.5% tranexamic acid versus 13.2% placebo; RR 0.87, 95% CI 0.77 to 0.98).
There was no evidence of increased complications or adverse events including vascular occlusive events, seizures, sepsis, or renal failure.
CRASH-3 did not report outcomes separately for patients with mild and moderate traumatic brain injury. Correspondence from the study authors, requested by the NICE committee, suggested significant uncertainty about the effectiveness of tranexamic acid in people with mild traumatic brain injury, although no additional data was shared.[28]

4% of participants had a mild traumatic brain injury (39% moderate and 57% severe). Subgroup results for outcomes by severity of injury were not reported.
The evidence suggested a reduced all-cause mortality with tranexamic acid compared to placebo at 28 days (absolute effect 49 fewer per 1000, 95% CI 89 fewer to 9 more) and at 6 months (absolute effect 40 fewer per 1000, 95% CI 87 fewer to 30 more).
There was no clinically important difference in hospital-free days at 28 days, neurosurgical intervention at 28 days, degree of disability at discharge and 6 months (measured by Glasgow Outcome Scale-Extended >4) or serious adverse events (myocardial infarction, pulmonary embolism, deep vein thrombosis, or thrombotic stroke).
The NICE 2023 guideline did not make a separate recommendation for older adults as they were awaiting results from the CRASH-4 trial looking specifically at this population.[28][64]

Children and infants

Primary options

Consider – prompt debridement plus dural repair and cranioplasty

Back

Consider –

prompt debridement plus dural repair and cranioplasty

Additional treatment recommended for SOME patients in selected patient group

Prompt surgery is indicated if the patient has:[11][73][74]

Intracranial haemorrhage
CSF leak
Gross contamination
Gross cosmetic deformity.

It is important to operate sooner rather than later.

Treatment delay increases the risk of infectious complications.
Operative repair should concentrate on debridement of devitalised tissues, evacuation of any surgical intracranial lesions, dural closure, and cranioplasty. Bone fragment replacement does not appear to increase the risk of infectious complications.[75][11][73][74] Single-stage procedures are now routinely performed.
Grossly contaminated open skull fractures should be followed up in 2 to 3 months with CT scans to rule out intracranial infection.[5][55][74]

Consider – prophylactic anticonvulsant therapy

Back

Consider –

prophylactic anticonvulsant therapy

Additional treatment recommended for SOME patients in selected patient group

Post-traumatic seizures (PTS) are common following severe traumatic brain injury, and risk of PTS is significantly increased even following mild and moderate brain injury.[11][82][83]

Anticonvulsants have not been shown to have any effect on decreasing the risk of late PTS (≥8 days) or post-traumatic epilepsy, and their use beyond the first week post-injury is not supported or recommended.[11][82][83][84][85]

Do not use anticonvulsants for either early or late PTS prevention in isolated skull fracture in the absence of underlying brain injury and outside of severe depressed skull fracture.

Primary options

phenytoin: children and adults: consult specialist for guidance on dose

levetiracetam: children and adults: consult specialist for guidance on dose

ONGOING

persistent cranial nerve injury or CSF leakage

1st line – surgical repair

Back

1st line –

surgical repair

About 10% to 30% of skull fractures are associated with cerebrospinal fluid (CSF) leakage, although it occurs most often in patients with basilar skull fractures.[76]

Initially treat CSF leakage with lumbar drainage.[77][78] If the CSF leakage is persistent, the surgical approach depends on the site of the leak and should be decided on a case-by-case basis following discussion between an ENT specialist and a neurosurgeon.

For base of skull fractures, endonasal repair is preferable because it has better outcomes and lower morbidity than craniotomy.[56][79][77][78][76]
Uncommonly, some CSF leaks (e.g., high up the frontal sinus) may not be reachable endonasally or may require formal cranialisation of the sinus via a craniotomy.
The most common complication of intranasal surgery is anosmia.[56][77]

Also consider operative repair if there is evidence of cranial nerve injury (e.g., hearing loss persisting for >3 months, facial paralysis).[80][79]

Bear in mind, however, that there is little evidence that surgical treatment of facial paralysis is superior to conservative management.[81]

Use of this content is subject to our disclaimer

Skull fractures

Treatment algorithm

suspected skull fracture (any type)

stabilisation and supportive care

confirmed closed non-depressed fracture

observation and monitoring

Practical tip

tranexamic acid

Evidence: Tranexamic acid

Primary options

These drug options and doses relate to a patient with no comorbidities.

Primary options

Drug choice, dose and interactions may be affected by the patient's comorbidities. Check your local drug formulary.

Primary options

confirmed closed depressed fracture

observation and monitoring

Practical tip

tranexamic acid

Evidence: Tranexamic acid

Primary options

These drug options and doses relate to a patient with no comorbidities.

Primary options

Drug choice, dose and interactions may be affected by the patient's comorbidities. Check your local drug formulary.

Primary options

dural repair and cranioplasty

prophylactic anticonvulsant therapy

Primary options

These drug options and doses relate to a patient with no comorbidities.

Primary options

Drug choice, dose and interactions may be affected by the patient's comorbidities. Check your local drug formulary.

Primary options

confirmed open fracture

observation and monitoring

Practical tip

tranexamic acid

Evidence: Tranexamic acid

Primary options

These drug options and doses relate to a patient with no comorbidities.

Primary options

Drug choice, dose and interactions may be affected by the patient's comorbidities. Check your local drug formulary.

Primary options

prompt debridement plus dural repair and cranioplasty

prophylactic anticonvulsant therapy

Primary options

These drug options and doses relate to a patient with no comorbidities.

Primary options

Drug choice, dose and interactions may be affected by the patient's comorbidities. Check your local drug formulary.

Primary options

persistent cranial nerve injury or CSF leakage

surgical repair