To determine which children who had sustained head injuries would not benefit from CT scans, Kuppermann and colleagues3 conducted a prospective cohort study of more than 42,000 children from 25 North American emergency departments (EDs).
Head trauma is a common occurrence in childhood; injuries range from mild bumps to severe traumatic brain injury. Because it is imperative to identify, diagnose, and treat clinically important traumatic brain injury in a timely fashion, physicians frequently feel that they must obtain CT scans in children with anything but the most innocent of head injuries to avoid missing a potentially serious situation. However, CT scans increase a child’s risk of malignancy later in life,1,2 and many of the abnormalities identified on CT do not require acute intervention.
An analysis of records of children with trauma.
To determine which children who had sustained head injuries would not benefit from CT scans, Kuppermann and colleagues3 conducted a prospective cohort study of more than 42,000 children from 25 North American emergency departments (EDs). The authors included children younger than 18 years who presented less than 24 hours after blunt head trauma with Glasgow Coma Scale scores of 14 or 15. Patients with trivial injuries, such as running into stationary objects or falling at ground level with no signs of trauma other than abrasions or lacerations, were excluded, as were children with underlying neurological conditions, penetrating trauma, or imaging at an outside hospital before transfer.
Investigators at each site used standardized forms to record patient histories, mechanism of injury, signs, and symptoms; this information was recorded before imaging results were known. Whether a CT scan was obtained and whether a child was admitted were at the ED physician’s discretion. Patients discharged from the ED were contacted by telephone 7 to 90 days later to see whether any injuries were missed.
The authors defined clinically important traumatic brain injury as that which resulted in death, neurosurgery, intubation for more than 24 hours, or hospital admission for 2 or more nights. Brief intubations that might have been used for reasons other than brain injury (such as for sedation for imaging) were not included in the outcome definition. The authors also excluded from the outcome measure single-night hospitalizations for observation during which no intervention was necessary.
Non-radiological predictors of traumatic brain injury. Validated predictors of clinically important traumatic brain injury in children younger than 2 years include altered mental status, non-frontal scalp hematoma, loss of consciousness for at least 5 seconds, severe mechanism of injury, palpable skull fracture, and not acting normally according to the parent. The risk of clinically important traumatic brain injury in a child with none of these 6 predictors was found to be 0.02%.
For children 2 years and older, the validated predictors include abnormal mental status, any loss of consciousness, history of vomiting, severe mechanism of injury, signs of basilar skull fracture, and severe headache. The risk of clinically important traumatic brain injury in a child with none of these 6 predictors was less than 0.05%.
The authors concluded that children who did not have any of the 6 predictors for their age-group did not need CT scans. If the predictors had been used in this study population to determine whether to order CT imaging, 25% of the CT scans done in children younger than 2 years and 20% of the CT scans done in children 2 years and older would have been obviated.
How the study results might cut down on unnecessary CT scans.
Despite the strong findings and large sample size, there are some limitations to the Kuppermann study. Other clinicians might have preferred a different definition of what constitutes clinically important traumatic brain injury. Also, injuries that did not meet the criteria for being “clinically important” were not studied.
In addition, not all study participants had CT scans, and patients were mainly from pediatric centers where clinicians may have had more experience with pediatric head trauma than clinicians at adult centers. These limitations should underscore the point that these predictors are meant to aid clinical decision making, not replace it.
While it has its limitations, the Kuppermann study nonetheless has the potential to significantly reduce the number of unnecessary CT scans in children with head injuries. It could do so by obviating scans in children who do not have any of the 6 age-appropriate predictors of clinically important traumatic brain injury. For children with 1 or more predictors, either CT scans or observation may be appropriate, depending on the situation.
Brenner DJ. Estimating cancer risks from pediatric CT: going from the qualitativeto the quantitative.
. 2002;32:228-233, 242-244.
2. Brenner DJ, Hall EJ. Computed tomography-an increasing source of radiation exposure. N Engl J Med. 2007;357:2277-2284.
3. Kuppermann N, Holmes JF, Dayan PS, et al; Pediatric Emergency Care Applied Research Network (PECARN). Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study. Lancet. 2009;374:1160-1170.