Mild head trauma: who needs a CT scan

A 22-year-old student arrives at the emergency department after hitting her head in a bicycle fall. She was wearing a helmet, remembers everything, has a bump on the occipital region and a mild headache. She is awake, oriented and with a normal neurological exam. In the next room, a 79-year-old man hit his head when he tripped on the rug and also looks well. The team needs to decide who goes for the CT scan.

Mild head trauma is one of the most frequent emergency complaints. Most patients have no intracranial injury, but a minority have bleeds that need to be identified in time. Clinical decision rules were created to separate these groups. This text shows how they work, what studies from several countries say about their performance and why older adults and children require their own reasoning.

Key points

• Most mild head traumas have a normal CT scan.
• Rules such as the Canadian, the New Orleans, the CHIP and the NICE rule trade sensitivity for specificity in different ways.
• CT scans are still used more than necessary in many services.
• Older adults have more injuries and fewer signs, and the rules perform worse in them.
• In children, the PECARN rule and shared decision-making with the parents help choose.

The CT scan problem in mild trauma

Mild head trauma is usually defined by a Glasgow coma scale of 13 to 15 on arrival at the service (Foks et al., 2018). In this group, the CT scan finds an injury in a small portion of patients, and an even smaller portion needs neurosurgery. Foks et al. studied patients aged 16 or older with mild head trauma in hospitals in the Netherlands, from March 2015 to December 2016 (Foks et al., 2018). Among 4557 eligible patients, 3742 had a CT scan, 384 had a traumatic intracranial finding and 74 had an injury with neurosurgical potential (Foks et al., 2018).

These numbers show the size of the challenge. The team needs to find a few important injuries among many normal exams. Ordering a CT scan for everyone exposes many people to radiation and congests the service, and ordering it for few risks missing a hematoma that grows.

Use beyond what is necessary

Saran et al. did a systematic review with meta-analysis on the overuse of CT scans in mild head trauma and included 8 studies (Saran et al., 2024). The pooled overuse rate was 27 per 100 exams (Saran et al., 2024). The rate varied according to the criterion used to judge the indication, and the authors warned about the high heterogeneity among the studies (Saran et al., 2024).

These results remind us that a guideline on paper changes practice little if it does not come with changes in the flow. When the exam is easy to order, quick to do and reassuring for everyone, it tends to become routine. The fear of missing an injury, the pressure of time, the family's expectation and the lack of a place to observe the patient push the decision toward the CT scan, even when the rule would say the risk is very low.

Changing this picture requires more than publicizing the rule. It helps to have the rule's form in the medical record, periodic feedback to the team on the exam rate and an honest conversation about the risk with the patient. The text on overdiagnosis discusses the harms of investigating too much, and the text on clinical decision-making shows how to choose exams by what they change in management.

Decision rules in adults

The most studied rules in adults are the Canadian head CT rule, the New Orleans criteria, the CHIP rule, developed in the Netherlands, and the NICE guideline, from the United Kingdom (Foks et al., 2018). All of them combine items such as age, vomiting, amnesia, signs of skull fracture, dangerous mechanism, anticoagulant use and the evolution of the Glasgow scale. They differ in the choice of items and in the weight of each one.

This difference changes how the rule behaves. A rule with many items orders more CT scans and misses fewer injuries, and a rule with few items orders fewer scans and misses more. The choice depends on what the service is willing to accept.

Why the rules diverge

The rules were born in different populations, with different definitions of outcome. Some were designed to find any injury on CT, and others to find only the injuries that lead to neurosurgery or death. A rule made for the second goal looks poorly sensitive when it is assessed by the first, and the reverse also happens.

The inclusion criteria also change. Some rules exclude anticoagulated patients, children or people without loss of consciousness, and others include everyone. Before adopting a rule, it is worth checking whether the service's population resembles the population in which it was validated. The text on pretest probability shows how local prevalence changes the value of any test.

External validation of four rules

In the study by Foks and colleagues, sensitivity for any traumatic finding on CT ranged from 73 in 100 with NICE to 99 in 100 with the New Orleans criteria (Foks and colleagues, 2018). Specificity ranged from 4 in 100 with New Orleans to 61 in 100 with NICE (Foks and colleagues, 2018). The authors recommended the CHIP rule, because it greatly reduces CT scans and misses few injuries with neurosurgical potential (Foks and colleagues, 2018).

The study shows the trade-off every rule makes. New Orleans almost never misses an injury, but orders CT for almost everyone. NICE orders fewer scans, but lets more findings pass. The ideal rule depends on the risk threshold accepted by the service: the authors showed that New Orleans was preferable at a low threshold, NICE at a higher threshold and CHIP at an intermediate threshold (Foks and colleagues, 2018).

Canadian versus New Orleans

Alzuhairy and colleagues gathered 14 articles with 21 140 samples in a meta-analysis that compared the Canadian rule and the New Orleans criteria (Alzuhairy and colleagues, 2020). To predict a positive finding on CT, the Canadian rule had sensitivity of 89.8 in 100 and specificity of 38.3 in 100, and New Orleans had sensitivity of 97.2 in 100 and specificity of 12.3 in 100 (Alzuhairy and colleagues, 2020). The authors concluded that both have good predictive value and can be used one in place of the other when there is a contraindication (Alzuhairy and colleagues, 2020).

Smaller studies repeat the pattern. Kavalci and colleagues included 175 patients in Turkey and found sensitivity of 76.4 in 100 and specificity of 41.7 in 100 for the Canadian rule, against 88.2 and 6.9 in 100 for New Orleans (Kavalci and colleagues, 2014). Mata-Mbemba and colleagues assessed 142 patients at a tertiary hospital in Japan and found age to be the strongest item for predicting an important finding on CT (Mata-Mbemba and colleagues, 2016).

Study	Population	Main finding	Source
Validation of four rules	4557 adults in the Netherlands	Sensitivity of 73 to 99 in 100, depending on the rule	(Foks and colleagues, 2018)
Meta-analysis	14 articles, 21 140 samples	Canadian with sensitivity of 89.8 and New Orleans with 97.2 in 100	(Alzuhairy and colleagues, 2020)
Overuse	8 studies	27 in 100 CT scans beyond what is indicated	(Saran and colleagues, 2024)
Older adults	2310 patients aged 60 or over	Intracranial injury in 13 in 100	(Coffeng and colleagues, 2023)
Children in France	1499 children	Sensitivity of 100 in 100 for the PECARN rule	(Lorton and colleagues, 2016)

The older adult who looks well

Age appears in almost every rule, and for good reasons. Coffeng and colleagues followed 5517 patients with mild head trauma, of whom 2310 were 60 or older (Coffeng and colleagues, 2023). Older adults had loss of consciousness and post-traumatic amnesia less often, but had more intracranial injuries, 13 versus 10 per 100, and 30-day mortality of 2.0 versus 0.1 per 100 (Coffeng and colleagues, 2023).

Older adults, therefore, have more injuries and fewer signs. When age was removed as a risk factor, the sensitivity of the rules in older adults ranged from 60 per 100 with the Canadian rule to 97 per 100 with New Orleans (Coffeng and colleagues, 2023). Among older adults whose only risk factor was age, intracranial injury appeared in 5 to 8 per 100, depending on the rule (Coffeng and colleagues, 2023). The authors suggest performing a CT scan on every older adult with mild head trauma until age-specific rules exist (Coffeng and colleagues, 2023).

Anticoagulants and bleeding

Anticoagulant use also weighs heavily. Albers and colleagues reviewed 3088 patients with mild head trauma and a Glasgow score of 14 or 15 who underwent CT between 2002 and 2011, with the indication given by the Canadian rule (Albers and colleagues, 2013). Of these, 149 had intracranial bleeding (Albers and colleagues, 2013). The incidence increased with age and with anticoagulant use, although the correlation was weak, and no patient with bleeding died or worsened neurologically (Albers and colleagues, 2013).

This result invites thinking in two steps. The first is finding the bleeding, and the second is deciding what it means for that patient. The text on prognostic reasoning discusses how to move from the finding to predicting what comes next.

The gentleman in the next room

Returning to the opening scene, the 79-year-old man who tripped on the rug enters almost every rule through age. Even if he seems well, the evidence indicates that older adults have more injuries and fewer clinical signs (Coffeng and colleagues, 2023). Asking about anticoagulant use, usual cognition and previous falls completes the assessment. The text on baseline rate in diagnosis shows why the same symptom weighs more in groups with a higher prevalence of disease.

Children and the PECARN rule

In children, concern about radiation is greater, because lifetime cancer risk increases with early exposures. The PECARN network developed rules that classify the child with mild head trauma as very low, intermediate or high risk and recommend skipping the CT scan in the very low risk group (Hess and colleagues, 2018).

Lorton and colleagues validated the PECARN rule in three emergency departments in France, with children under 16 years old and a Glasgow score of 14 or 15 who sought care within 24 hours of the trauma (Lorton and colleagues, 2016). Among 1499 children, 76 underwent CT and 9 had clinically important traumatic brain injury (Lorton and colleagues, 2016). All 9 were outside the very low risk group, and sensitivity was 100 per 100 with specificity of 69.9 per 100 (Lorton and colleagues, 2016).

Deciding together with the parents

The PECARN rule gives little guidance for children at intermediate risk, who may go for a CT scan or stay under home observation (Hess and colleagues, 2018). Hess and colleagues conducted a cluster randomized trial in 7 emergency departments in the United States, from April 2014 to September 2016, with children aged 2 to 18 in this intermediate group (Hess and colleagues, 2018). There were 172 professionals and 971 children (Hess and colleagues, 2018).

Parents who used a decision aid had more knowledge about their child's risk, less conflict in the decision and greater participation in the choice (Hess and colleagues, 2018). The CT rate at the department was similar, 22 versus 24 per 100, but the mean number of imaging tests in the following 7 days was lower in the decision aid group (Hess and colleagues, 2018). The trial recorded zero important injuries missed (Hess and colleagues, 2018).

This essay shows that the decision about the CT scan is also a conversation. A family that understands the risk accepts observation more readily, and the team shares responsibility for the choice with them. The text on diagnostic uncertainty offers strategies for communicating doubt honestly.

Biomarkers and what comes next

A recent line of research uses blood markers to help with the decision. Bazarian and colleagues evaluated a rapid test that measures two brain proteins, GFAP and UCH-L1, in samples from a study conducted at 22 centers in the United States and Europe (Bazarian and colleagues, 2021). Among 1901 patients with a Glasgow score of 13 to 15, the test had a sensitivity of 0.958 and a negative predictive value of 0.993 for intracranial injury on CT (Bazarian and colleagues, 2021). Specificity was 0.404 (Bazarian and colleagues, 2021).

The high negative predictive value suggests that a normal result helps reassure, but the low specificity means many abnormal results come with a normal CT. For now, the test serves as decision support, added to the clinical assessment and the rules.

What the CT scan shows beyond the decision

The CT scan also helps predict recovery. Yuh and colleagues analyzed 1935 patients with mild head trauma at 18 trauma centers in the United States and validated the findings in 2594 patients from a European cohort (Yuh and colleagues, 2021). The authors identified three groups of CT findings with different implications for recovery over one year (Yuh and colleagues, 2021). They advocate specific education and systematic follow-up for those with these findings (Yuh and colleagues, 2021).

This point broadens the reasoning. Deciding on the CT scan is one part of care, and following up with those who had an injury, even a small one, is another. Many patients with mild head trauma have symptoms for weeks, and guidance at discharge makes a difference.

Discharge with guidance

Discharge of the patient with mild head trauma is a moment for reasoning, not just paperwork. The person needs to know which signs call for an immediate return, such as a headache that worsens, repeated vomiting, unusual drowsiness, confusion, weakness on one side of the body or a seizure. They also need to know who will watch them in the first hours at home.

For older adults who live alone, this guidance changes the decision. A patient who could safely go home alongside a family member may need observation in the hospital when no one will stay with them. Recording who leaves with the patient and what was explained closes the loop of care.

How to organize the reasoning

The evidence allows building a practical sequence for mild head trauma. It starts with the correct classification and ends with discharge guidance.

• Confirm that the trauma is mild, with a Glasgow score of 13 to 15 and no focal signs.
• Choose a validated rule for adults and another for children, and apply them to everyone.
• In older adults, consider age as a risk in itself and ask about anticoagulants.
• In children at intermediate risk, talk with the parents about observation or CT.
• At discharge, give guidance on warning signs and on follow-up for those who had an injury.

The text on how to organize the reasoning in the medical record shows how to record the rule used and the decision made. This record protects the patient at shift change and protects the team when the decision is questioned.

Where the reasoning usually fails

Three failures appear often. The first is ordering a CT scan for everyone out of fear, without applying any rule. The second is applying an adult rule to a small child, or a general rule to an older adult without considering age. The third is releasing the patient without guidance on what to watch for at home.

Discussing cases of unnecessary tests and missed injuries in morbidity and mortality sessions helps calibrate the use of the rules. The text on signs, symptoms and history shows how to collect the rule items precisely, because a poorly asked amnesia question changes the decision.

Watching is also deciding

Between ordering the CT scan now and sending the patient home there is a third option: observe. A few hours of observation with repeated neurological assessment allow you to notice increasing drowsiness, worsening headache or returning vomiting. This strategy is especially useful when the rule leaves doubt, when the account of the trauma is incomplete or when the patient was intoxicated on arrival.

Observation works when it has a script. The team needs to know what to measure, how often and what triggers the CT scan. The Glasgow coma scale, the pupils, limb strength and orientation are the simplest items to repeat at the bedside. Recording each assessment with the time makes the trend visible, and it is the trend that matters more than a single value.

The conversation with the patient and with whoever will stay with them at home is part of the decision. Explaining why the test was skipped, which signs call for an immediate return and who should stay close in the first hours makes the discharge a plan. The companion who understands the plan brings the patient back to the service at the right time.

Finally, the decision to skip the CT scan is a risk estimate, and estimates can change. The rule applied on arrival is valid for that moment. If the picture changes, the question comes back to the table and deserves a new answer.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by searching for the code in ICD-10.

The limits on using the game are described on the NEXO economy page. If you research the teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information on curation and on creators, and institutions can learn about the proposal on the partnerships page. Training with differential diagnosis and clues and searching for the right code in ICD-10 complete this training.

Frequently asked questions

Does every mild head trauma need a CT scan?

Most have a normal CT scan, and in a study in the Netherlands, 384 of the 3742 patients who had the scan had a traumatic intracranial finding (Foks et al., 2018).

Which rule is best for adults?

It depends on the accepted risk threshold, and the authors of the Dutch study recommended the CHIP rule because it greatly reduces CT scans and misses few injuries with neurosurgical potential (Foks et al., 2018).

Do older adults need special care?

They do, because older adults had more intracranial injuries, 13 versus 10 per 100, despite fewer clinical signs (Coffeng et al., 2023).

Does the PECARN rule work outside the United States?

In France, the rule had a sensitivity of 100 per 100 among 1499 children (Lorton et al., 2016).

Can a blood test replace the CT scan?

The rapid GFAP and UCH-L1 test had a negative predictive value of 0.993, and for now it works as decision support (Bazarian et al., 2021).

Take-home messages

• Most mild head traumas have a normal CT scan.
• Validated rules help order the test for those who need it.
• Older adults have more injuries and fewer signs, and age counts on its own.
• In children, the PECARN rule and the conversation with the parents guide the choice.
• Discharge needs guidance on alarm signs and follow-up.

References

1. Albers CE, von Allmen M, Evangelopoulos DS, et al. What is the incidence of intracranial bleeding in patients with mild traumatic brain injury? A retrospective study in 3088 Canadian CT head rule patients. Biomed Res Int. 2013. PMID 23936802.
2. Alzuhairy AKA. Accuracy of Canadian CT Head Rule and New Orleans Criteria for Minor Head Trauma; a Systematic Review and Meta-Analysis. Arch Acad Emerg Med. 2020. PMID 33244515.
3. Bazarian JJ, Welch RD, Caudle K, et al. Accuracy of a rapid glial fibrillary acidic protein/ubiquitin carboxyl-terminal hydrolase L1 test for the prediction of intracranial injuries on head computed tomography after mild traumatic brain injury. Acad Emerg Med. 2021. PMID 34358399.
4. Coffeng SM, Foks KA, van den Brand CL, et al. Evaluation of Clinical Characteristics and CT Decision Rules in Elderly Patients with Minor Head Injury: A Prospective Multicenter Cohort Study. J Clin Med. 2023. PMID 36769631.
5. Foks KA, van den Brand CL, Lingsma HF, et al. External validation of computed tomography decision rules for minor head injury: prospective, multicentre cohort study in the Netherlands. BMJ. 2018. PMID 30143521.
6. Hess EP, Homme JL, Kharbanda AB, et al. Effect of the Head Computed Tomography Choice Decision Aid in Parents of Children With Minor Head Trauma: A Cluster Randomized Trial. JAMA Netw Open. 2018. PMID 30646167.
7. Kavalci C, Aksel G, Salt O, et al. Comparison of the Canadian CT head rule and the new orleans criteria in patients with minor head injury. World J Emerg Surg. 2014. PMID 24742359.
8. Lorton F, Poullaouec C, Legallais E, et al. Validation of the PECARN clinical decision rule for children with minor head trauma: a French multicenter prospective study. Scand J Trauma Resusc Emerg Med. 2016. PMID 27488722.
9. Mata-Mbemba D, Mugikura S, Nakagawa A, et al. Canadian CT head rule and New Orleans Criteria in mild traumatic brain injury: comparison at a tertiary referral hospital in Japan. Springerplus. 2016. PMID 27026873.
10. Saran M, Arab-Zozani M, Behzadifar M, et al. Overuse of computed tomography for mild head injury: A systematic review and meta-analysis. PLoS One. 2024. PMID 38206917.
11. Yuh EL, Jain S, Sun X, et al. Pathological Computed Tomography Features Associated With Adverse Outcomes After Mild Traumatic Brain Injury: A TRACK-TBI Study With External Validation in CENTER-TBI. JAMA Neurol. 2021. PMID 34279565.