Abusive head trauma is a common cause of death and traumatic brain injury in young children. Computed tomographic (CT) scanning and magnetic resonance imaging (MRI) can identify children who have suffered abusive head trauma, but identifying children who need these studies is often difficult. When clear signs of inflicted head trauma are present, investigation is certainly indicated. In the absence of clear evidence, however, it is more difficult to determine when imaging is appropriate. Laskey and associates reviewed the current standards of evaluation for abusive head trauma among children with suspected physical abuse as evidenced by an order for a skeletal survey.

The authors reviewed the medical records of 51 children up to to 48 months of age who had skeletal surveys for suspected abuse and who had no evidence of head trauma. Outcomes examined included the following: (1) imaging by CT or MRI; (2) dilated fundoscopic examination performed by an ophthalmologist; and (3) any positive results from these diagnostic procedures.

Twenty-four of the 51 children evaluated (47 percent) had a fracture identified on the skeletal survey, most commonly an arm fracture. Thirty-eight (75 percent) had neuroimaging; 11 of these (29 percent) demonstrated an intracranial injury such as a subdural hematoma, cerebral hematoma, or epidural hematoma. Opthalmology consults were performed on 35 patients (69 percent), with retinal hemorrhage identified in one child (3 percent). The only factor significantly related to ordering neuroimaging or ophthalmologic examination was an age younger than 12 months. Children in this age group had more frequent evaluations for intracranial trauma than children older than 12 months.

The authors conclude that evidence is still lacking about the most appropriate use of neuroimaging in many asymptomatic younger children with suspected physical abuse. They recommend that physicians consider MRI as the better imaging choice because of greater accuracy for the variety of potential traumatic injuries and because of the attention focused on the potential danger of cancer later in life among children undergoing CT. They add that dilated fundoscopic examination is a safe alternative, but that reliance on retinal hemorrhage as a criterion for head trauma would miss a large number of children with occult intracranial injuries. Further studies are needed to guide the most appropriate use of neuroimaging in the evaluation of asymptomatic young children who may have been physically abused.