Mortality after abusive head trauma: Relationship to brain injury severity, age, and co-occurring bodily injuries.

BACKGROUND

Abusive head trauma (AHT), a leading cause of death from child physical abuse, disproportionately impacts the youngest children. Identifying children at highest risk following AHT has been hampered by the lack of a widely accepted measure of brain injury severity in infants and toddlers.

OBJECTIVE

Identify clinical factors associated with mortality from AHT.

PARTICIPANTS AND

SETTING

403 patients (mean age 10.8 months) treated for AHT between 2012 and 2020 at a tertiary care children's hospital.

METHODS

Single-center retrospective chart review. We tested relationships between covariates and mortality with t-tests and chi-square tests; these analyses guided development of a multivariable model using binary logistic regression.

RESULTS

Forty-two of 403 children died (10.4 %). In bivariate analyses, mortality was linked to lower Glasgow Coma Scale (GCS; d = 2.07, p < .001), older age (d = -0.55, p = .009), midline shift (Φ = 0.10, p = .04), eye injuries (V = 0.42, p < .001), fractures (Φ = 0.14, p = .004), abdominal/visceral injuries (Φ = 0.14, p = .005), neurosurgical interventions (V = 0.22, p < .001), and intubation (Φ = 0.37, p < .001). Older children were at risk for more severe brain injuries and abdominal/visceral injuries. In the multivariable model, lower GCS (p < .001) and more fractures (p = .02) were associated with mortality.

CONCLUSIONS

Our data support the use of GCS in infants and toddlers following AHT. Older children were more likely to die than younger children, with evidence for differing injury patterns by age.