To decide whether or not to CT for mild head trauma, look to the rules

There is sufficient existing literature on which to base clinical decision rules separating patients with apparently minor but clinically serious head trauma—and thus in need of CT neuroimaging and prompt clinical intervention—from similarly alert and responsive head-trauma patients who would be better off avoiding the radiation exposure and the costs.

That’s among the key findings in a study published online Dec. 22 in JAMA.

Joshua Easter, MD, MSc, of the University of Virginia, and colleagues reviewed 14 studies ranging in scope from 431 patients to 7,955 patients, encompassing a total of 23,079, all of whom had Glasgow Coma Scale (GCS) scores between 13 and 15.

Such scores represent the healthier end of the scale, meaning these patients presented with mild to zero impairment in alertness as gauged by motor response, verbal response and eye opening.  

The team found that 7.1 percent of the apparently mild cases did, in fact, have severe intracranial injury.

The prevalence of injuries leading to death or requiring neurosurgical intervention was 0.9 percent.

A number of findings at physical exam correlated with severe intracranial injury showing up in CT scans:

• Skull fracture (likelihood ratio, 16);
• GCS score of 13 (likelihood ratio, 4.9);
• Two or more vomiting episodes (likelihood radio, 3.6);
• Any decline in GCS score (likelihood ratio ranging from 3.4 to 16); and
• Pedestrians struck by motor vehicles (likelihood ratio ranging from 3.0 to 4.3).

Meanwhile, the absence of any of the features of the Canadian CT head rule (65 or older, two or more vomiting episodes, amnesia for more than 30 minutes, pedestrian struck by car, driver or passenger ejected from vehicle, on the ground for longer than one minute, GCS under 15 at two hours) had a likelihood ratio of 0.04, cutting the probability of severe injury to 0.31 percent.

In addition, the absence of all the New Orleans criteria (older than 60, intoxication, headache, vomiting, amnesia, seizure or trauma above the clavicle) had a likelihood ratio of 0.08, reducing the probability of severe intracranial injury to 0.61 percent.

Weeding out low-riskers

The rules taken up in the Easter study are high in sensitivity and low-to-intermediate in specificity—and are designed, at least in part, to weed out patients who are at low risk of serious injury and so do not require CT imaging, the authors note in their study discussion.

Often these rules are misinterpreted as the presence of one of the features of the rule calling for CT, they add.

“The decision to discharge, observe or recommend CT to the patient with at least one feature of a rule depends on the setting, clinician’s judgment about the likelihood of injury, patient preference, number of features present, and the particular features present,” they write.

Easter et al. also point out that, while they’re called rules, these “instruments” should augment and “not replace clinical judgment.”

Rules nearly always right  

In a supplemental podcast, Ed Livingston, MD, JAMA’s deputy editor for clinical reviews and education, stresses that the Canadian rule and New Orleans criteria are “really good at identifying patients who have a very low risk for ever needing an intervention.”

“If you go about doing a careful assessment of the patient and appropriately document what you did and what you were thinking when you made your decisions,” adds Livingston, “there is very little likelihood that a malpractice action brought against you is going to succeed.”

What did Easter learn from reviewing the copious evidence reported in the rational clinical examination article?

“I think you have to be sure that you are applying the rules in the fashion that they were intended,” Easter says in the podcast.

There are certain populations that have not been studied extensively in the context of the rules, he adds.

“For example, patients who are anti-coagulated or patients who are intoxicated or the elderly—they were incorporated into some of these studies, but there weren’t large studies looking at how the rules perform in those specific cohorts,” says Easter. “So you have to be cautious applying the rules to those particular patient populations.”

A ‘great need’ identified

In an associated JAMA editorial, Frederick Rivara, MD, MPH, of the University of Washington and editor of JAMA Pediatrics, and colleagues state that, in most minor head-trauma cases, CT scans don’t do much to guide care or optimize long-term patient-centered outcomes.

“It is misleading to use the mental justification that clinicians should apply decision rules for performing a CT scan for patients presenting to the ED with minor blunt head trauma because doing so can help reassure patients and their families that normal CT findings rule out brain injury,” they write.

“There is a great need,” write Rivara and colleagues, “for the development of better diagnostic tools to help clinicians identify the 10 percent to 20 percent of patients in this group who are at risk of persistent symptoms and longer-term consequences or require more detailed evaluation.”