Am Fam Physician. 2006;73(12):2205-2208
Clinical Question
When is computed tomography (CT) indicated for patients with minor head injury?
Evidence Summary
Minor head injury is commonly seen in the primary care and emergency department settings. Minor head injury typically includes brief loss of consciousness or other symptoms of concussion. Several clinical decision rules have been developed and validated to help identify patients who need CT of the brain. These rules have the potential to help physicians identify patients with clinically important head injury (generally defined as any acute brain finding on CT that would normally require hospitalization and neurologic follow-up) while reducing the number of CT scans performed. This would save millions of dollars in unnecessary scans, reduce emergency and radiology department overcrowding, save time, and prevent unnecessary transfers from facilities without access to CT.
The New Orleans Criteria (Table 11 ) established the first successfully validated clinical decision rule for selective use of CT in minor head injury.1 The researchers enrolled 520 consecutive patients three to 97 years of age with minor head injury (defined as loss of consciousness after head trauma with a normal neurologic examination and a normal Glasgow Coma Scale [GCS] score of 15 [Table 22]). The researchers identified seven findings that independently predicted clinically significant head injury. In a prospective validation study3 that included 909 patients, only those with one or more of these findings had an abnormal head CT scan. Among the 697 patients with at least one of these findings, 57 had an abnormal CT.3
| CT is needed if the patient meets one or more of the following criteria: |
| Headache |
| Vomiting |
| Age older than 60 years |
| Drug or alcohol intoxication |
| Persistent anterograde amnesia (deficits in short-term memory) |
| Visible trauma above the clavicle |
| Seizure |
| Patient characteristics | Points |
|---|---|
| Eyes open | |
| Spontaneously | 4 |
| To speech | 3 |
| To pain | 2 |
| Never | 1 |
| Best verbal response | |
| Oriented | 5 |
| Confused | 4 |
| Inappropriate words | 3 |
| Incomprehensible sounds | 2 |
| Silent | 1 |
| Best motor response | |
| Obeys commands | 6 |
| Localizes pain | 5 |
| Flexion withdrawal | 4 |
| Decerebrate flexion | 3 |
| Decerebrate extension | 2 |
| No response | 1 |
| Total | |
A second clinical decision rule, the Canadian CT Head Rule, was developed using 3,121 patients 16 to 99 years of age who presented with minor head injury (defined as GCS score of 13 to 15 after loss of consciousness, definite amnesia, or witnessed disorientation from trauma).4 The Canadian CT Head Rule4 consists of five high-risk and two moderate-risk factors. According to the rule, CT is needed if the patient meets one or more of the following seven criteria:
GCS score lower than 15 two hours after injury
Suspected open or depressed skull fracture
Any sign of basal skull fracture (e.g., hemotympanum, “raccoon eyes,” cerebrospinal fluid, otorrhea or rhinorrhea, Battle’s sign [bluish discoloration of the postauricular region])
Two or more episodes of vomiting
Age 65 years or older
Amnesia before impact of 30 minutes or more
Dangerous mechanism (i.e., pedestrian struck by motor vehicle, occupant ejected from motor vehicle, or a fall from a height of at least 3 ft or five stairs)
The investigators found that, in the original population, the rule successfully identified all 44 patients who required neurologic intervention and only missed four out of 254 patients with clinically important injury. All of these patients had small contusions that did not require treatment and had no neurologic sequelae.4
Clinical decision rules should be validated prospectively in a new population to confirm their accuracy, particularly when the consequences of error are great, as they are with head injury. Two studies3,5 have directly compared the New Orleans Criteria with the Canadian CT Head Rule. The first study,3 conducted by the original Canadian CT Head Rule developers, identified 2,707 patients 16 to 99 years of age with minor head injury (defined as blunt head trauma within the previous 24 hours causing loss of consciousness, definite amnesia, or witnessed disorientation) and GCS scores of 13 to 15. Because the New Orleans Criteria was designed for patients with normal GCS scores, a subgroup that included only patients with a normal GCS score of 15 (n = 1,822) was used to compare the two rules. In the normal GCS subgroup, both rules were 100 percent sensitive, identifying all eight patients requiring neurosurgical intervention and all 97 patients with clinically important brain injury. However, the Canadian CT Head Rule was more specific than the New Orleans Criteria and led to a greater reduction in head CT scans (52 versus 88 percent). In patients with GCS scores of 13 to 15, the Canadian CT Head Rule also was 100 percent sensitive, identifying all 41 patients requiring neurosurgical intervention and all 231 patients with clinically important brain injury.3
The second study5 compared the rules using a group of 3,181 patients 16 to 102 years of age presenting with recent blunt trauma and a GCS score of 13 or 14 or a GCS score of 15 and one of the following risk factors: loss of consciousness, amnesia, post-traumatic seizure, vomiting, severe headache, evidence of drug use, anticoagulant use, physical evidence of injury, or neurologic deficit. The rules were adapted to apply to a broader population than those used when the rules were originally developed. Although both rules identified all patients requiring immediate neurosurgical intervention, the Canadian CT Head Rule led to a greater reduction in CT scans compared with the New Orleans Criteria (37 versus 5 percent). However, the Canadian CT Head Rule also was less sensitive at identifying patients with serious abnormal CT findings (85 versus 99.4 percent). Much of the lower sensitivity was caused by the rule’s failure to detect linear skull fractures, which it was not designed to do.
Therefore, physicians can be confident that both tools will reliably identify patients requiring neurosurgical intervention and, if limited to those with a normal GCS, they will reliably identify clinically important brain injury. However, extending either tool to patients with GCS scores of 13 or 14 is controversial, particularly because of the increased medicolegal risk to physicians in the United States.6
Applying the Evidence
A 25-year-old woman presents to your rural family practice after a fall. She struck her head, and her friend says she was “out cold” for about one minute. She does not recall the five minutes or so before falling but currently has a normal GCS score of 15 and has normal anterograde memory. She denies headache or vomiting. Should you send her to the town hospital for a head CT scan?
Answer: Because she has a normal GCS score; no evidence of intoxication, headache, vomiting, or anterograde amnesia; is younger than 60 years; and has a normal physical examination, she does not require CT according to the New Orleans Criteria and Canadian CT Head Rule. Even if she had contusions or lacerations, she would not require CT according to the Canadian CT Head Rule, because a dangerous mechanism was not involved in the fall.