Written by Graham Van Schaik
Spoon Feed
In pediatric patients with blunt head injury, PECARN traumatic brain injury (TBI) low-risk criteria accurately identified infants <3 months of age at low-risk for clinically important TBI (ciTBI), BUT these low-risk infants had a higher rate of findings on CT (TBI-CT). Use caution in these young ones.
Why does this matter?
Blunt head trauma is common in children < 1 year, accounting for nearly 81,000 ED visits annually. Avoiding unnecessary CTs in the pediatric population makes all of us happy, but it is often difficult to determine for which patient the benefit of a scan will outweigh the risk. PECARN decision tools are nearly universal in Pediatric EDs, but prediction rules for TBI among children with minor head trauma have not specifically been evaluated in infants younger than 3 months.
They relly hit it on the head with this one…
Design: This is a secondary analysis of the public use data set from the PECARN prospective observational study, limiting the scope to only infants younger than 3 months. Outcomes were: 1) ciTBI* 2) TBI-CT** 3) Skull Fracture on CT***
Results: Within the PECARN dataset there were 10,904 children <2 years old with minor blunt head trauma, which dropped to 1,147 (10.5%) who were < 3 months. After additional exclusion for missing data, 1,081 infants were analyzed. CTs were obtained in 58.6% of cases. Among patients who did not meet PECARN low-risk criteria the rate of ciTBI was 4.2%. TBI-CT within this group was 21.3%, and skull fractures 28%. Among PECARN low-risk patients, only 1 had a ciTBI (1/514, 0.2%). However, ten patients, or 5.1%, had TBI-CT and 9 patients, or 4.6%, had skull fractures. Of those who were low-risk using PECARN criteria but had TBI-CT, the most common mechanism was fall <3 feet, with the most common exam findings being abrasions and lacerations or hematomas above the clavicles (excluding hematomas of the scalp).
Implications: This study really gets you thinking – while the PECARN prediction rules are highly sensitive to identify those at low risk of a ciTBI, the rate of TBI-CT in this group still remained quite high (5.1%). This is enough to make me think twice about rigidly applying the PECARN decision tool in this age group. I am also left wondering how this decision rule will be applied if rapid MRI becomes more prevalent.
Limitations: This was a retrospective analysis without granular details on the mechanisms of injury other than the general category and if it was PECARN high risk or not. Also, we know that PECARN shouldn’t be applied in suspected non-accidental trauma, and it is unknown if some of the CT scans were obtained for that purpose. Also, as providers are (rightly) cautious with infants, it is unknown if some patients were hospitalized for 2 or more nights based only on age alone rather than objective findings. Lastly, CTs weren’t obtained for all patients, so the true prevalence of TBI-CT may be higher or lower.
Overall, it seems the PECARN decision rule still does a remarkable job of identifying ciTBI in infants <3 months, though it does not as accurately predict TBI-CT. Given that there is little data on how TBI-CT sustained at this age affects long term behavioral and cognitive development, caution should be applied in this group and a generous dose of shared decision making.
Source
Risk of Traumatic Brain Injuries in Infants Younger than 3 Months With Minor Blunt Head Trauma. Ann Emerg Med. 2021 Sep;78(3):321-330.e1. doi: 10.1016/j.annemergmed.2021.04.015. Epub 2021 Jun 17.
*ciTBI defined as death from the TBI, TBI requiring neurosurgical procedures, intubation for at least 24 hrs for the TBI, or hospitalization for 2 or more nights because of head trauma and ongoing signs and symptoms in association with the TBI on CT.
**TBI-CT defined as any acute traumatic intracranial findings or skull fractures depressed by at least the width of the table of the skull.
***Defined as ANY skull fracture seen on CT including basilar, depressed, and other fractures meeting the definition of TBI on CT or requiring neurosurgical intervention.
