Head Injuries in Young Children from Motor Vehicle Collisions and Non-Accidental Trauma
Catherine M. Huber, MD1; Jodi Raymond, MPH2; Teresa M. Bell, PhD3; Shannon Thompson, MD1
1Indiana University School of Medicine, Indianapolis, IN; 2Riley Hospital for Children at Indiana University Health, Indianapolis, IN; 3Indiana University, Indianapolis, IN
Background: Pediatrics relies on history provided by caregivers and determining mechanisms of injury is critical in making a diagnosis. Caregivers may attribute injury to a mechanism like motor vehicle collision (MVC) when injuries do not match MVC. This study will describe a predictive model to assess if head injuries are caused by non-accidental trauma (NAT) or MVC.
Methods: The trauma registry was queried from January 1, 2011 to December 31, 2015. Inclusion criteria: < 6 years old at time of admission, either MVC or NAT ICD-9 E-code or ICD-10 External cause code, and hospitalized (N=570). Patients with a head injury were analyzed (n=351). Chi square analysis was utilized to see if clinical indicators were independently predictive and associated with NAT. Multivariable logistic regression analysis was done to identify independent factors related to NAT.
Results: The study population consisted of 148 and 203 children with head injury admitted for MVC or NAT, respectively. Age <2 years, male gender, Glasgow Coma Scale (GCS) <=8, subacute injury, retinal hemorrhages, subdural hemorrhages, subarachnoid hemorrhages, cerebral edema, absence of skin injury, and absence of concussion were all factors statistically significantly associated with NAT. The following variables were all significant predictors for NAT using logistic regression: age <2 years, male, subacute injury, retinal hemorrhages, subdural hemorrhages, absence of diffuse axonal injury, and absence of intraventricular hemorrhage.
Conclusions: Independent and modifiable predictors were identified to help determine if a child’s injury was due to MVC or NAT, showing that subdural hemorrhages are a strong predictor of NAT.
Back to 2018 Abstracts