Name: Pediatric head injury — bulletproof documentation — ER Toolkit
Creator: Kim Trinh, MD

The hard part of the pediatric head bump isn't the scan — it's defensibly not scanning. The chart that holds up applies PECARN by age group, records the tier and the observation-vs-CT decision (with shared decision-making in the middle group), considers abusive head trauma in the young child, and gives written return precautions for delayed bleeding.

01What's at stake

About 6–8% of children with a normal GCS still have an intracranial injury on CT, though fewer than 1% need neurosurgery — so a normal GCS doesn't exclude a bleed, and a child can have a subdural without ever losing consciousness. But children are up to ten times more radiosensitive than adults and over 90% of these CTs are normal, so the goal is to scan the few who need it and safely observe the rest.

02Can't-miss diagnoses

Clinically-important TBI — epidural/subdural hematoma, contusion; can present with a normal GCS and no LOC.
Skull fracture — palpable, depressed, basilar, or open; CT has poor sensitivity for basilar fractures, so the clinical signs matter.
Abusive head trauma — in children <2, up to 20% of TBI is non-accidental; infants may show only a scalp hematoma. Red flags: an inconsistent/changing or developmentally implausible history, young age, and pattern/associated injuries (auricular, frenulum, neck, or any bruising in a pre-mobile infant). → consider NAT, involve CPS
The bleeding-disorder / anticoagulated child — give empiric factor replacement before imaging if symptomatic; symptoms predict injury better than labs.

03The PECARN rule (by age group)

High-risk → CT recommended (either age, ~4% ICI risk): GCS <15, altered mental status, or a palpable/suspected basilar skull fracture. → CT
Intermediate → CT vs observation (~1% risk): LOC, severe mechanism, vomiting, or severe headache; for age <2 only, add a non-frontal (occipital/parietal/temporal) scalp hematoma and "not acting normally per parent," and use the lower fall threshold (>3 ft / 0.9 m vs >5 ft / 1.5 m). → observe or CT
Neither → CT not recommended (very-low-risk group).
PECARN is 100% sensitive for neurosurgical injury and ~97% for clinically-important TBI in >42,000 children.

Skip the typing

Work the case in the Pediatric Head Injury Workup — it records the PECARN predictors by age group and the observation plan, and assembles an MDM that documents the tier, the CT-vs-observation decision, and the abusive-head-trauma consideration.

04Observation, CT & disposition

Observe before scanning the intermediate/lower-risk child — a 4–6 hour observation period safely reduces CT use; consider a ~6-hour watch if the child is >3 months with a single symptom and parents and clinician are comfortable, CT-ing only if symptoms persist/worsen.
Shared decision-making in the intermediate group, weighing the child's age, symptom number/severity, and family/clinician comfort against radiation risk.
Disposition: normal exam + normal CT → discharge with a responsible adult and written precautions; normal CT with ongoing/worsening symptoms → observe 12–24 h ± repeat imaging; injury on CT → neurosurgery (selected small injuries are managed per local agreements).
Return precautions (written, to a responsible third party): worsening headache, repeated vomiting, agitation/confusion, excessive drowsiness, slurred speech, weakness, seizure, or fainting — delayed bleeding can occur for up to two weeks in high-risk patients.

05What to document

▼ weak

"3-yo fell, bumped head. Acting fine. No CT needed, discharged."

▲ defensible

"3-year-old, fall from standing onto carpet, witnessed. GCS 15, acting normally per parents, no AMS, no palpable skull fracture (high-risk PECARN predictors absent). No LOC, no vomiting, no severe headache, non-severe mechanism (intermediate predictors absent) → PECARN very-low-risk: CT not recommended. History consistent with development; no pattern injuries or concerning bruising (abusive head trauma considered, not suspected). Tolerated PO, normal neuro exam after observation. Written head-injury return precautions reviewed with parents (worsening headache, repeated vomiting, drowsiness, unsteadiness, seizure) and a copy given."

06Where charts fail

Assuming a normal GCS or absent LOC excludes a bleed.
Not documenting the PECARN predictors checked, the age group, or the resulting tier.
No documented shared decision-making for the intermediate (CT-vs-observe) child.
Missing abusive head trauma — not addressing history consistency, pattern injuries, or CPS in the young child.
Verbal-only return precautions (families recall 30–50%); no written delayed-bleed warning for the anticoagulated/bleeding-disorder patient.

07Sources

Haydel MJ. Management of mild traumatic brain injury in the emergency department. Emergency Medicine Practice — Trauma EXTRA (EB Medicine). 2025 Feb.
Kuppermann N, Holmes JF, Dayan PS, et al. Identification of children at very low risk of clinically-important brain injuries after head trauma (PECARN): a prospective cohort study. Lancet. 2009;374(9696):1160-1170.
Nigrovic LE, Kuppermann N. Children with minor blunt head trauma presenting to the emergency department. Pediatrics. 2019;144(6):e20191495.
Hettler J, Greenes DS. Can the initial history predict whether a child with a head injury has been abused? Pediatrics. 2003;111(3):602-607.

Pediatric head injury: documentation that holds up