Head trauma in children results in 600,000 emergency department visits and 95,000 hospital admissions per year. It is likely that many more such children are evaluated in physicians' offices. Predicting which children require diagnostic imaging can be difficult, and no established guidelines are in place to direct physicians who care for pediatric patients with head trauma. Published guidelines are based on limited clinical data and are not followed uniformly in practice; in addition, they generally do not specify which imaging technique is preferred. Gruskin and Schutzman performed a retrospective study to determine the incidence of skull fracture and intracranial injury in children who presented to a pediatric emergency department. They also attempted to determine which historic features and physical findings predict complications of head injury and whether clinical criteria could aid in the selection of diagnostic imaging.

Medical records were reviewed for children younger than two years of age who were discharged from a Boston children's hospital with a diagnosis of head injury, skull fracture, intracranial injury, cerebral contusion or cerebral edema. Excluded from the study were children with a history of seizures, blood dyscrasias, neurologic disorders, ventricular shunts or suspected abuse. Historic information included the estimated height of the fall, level of consciousness and presence of scalp abnormalities, and whether the child was referred by another physician or came directly to the emergency department. When skull radiographs or cranial computed tomographic (CT) scans were obtained, these results were also noted. Children were diagnosed with a “minor head injury” if they had a normal neurologic examination and were alert at discharge, and if radiologic studies were normal.

A total of 291 patients were evaluated; medical records were available for 278 patients (96 percent). Most of these children had gone directly to the emergency department. Approximately 60 percent of the children were younger than 12 months of age; 40 percent were between 13 and 24 months of age. Eighty-two percent of all children were ultimately given a diagnosis of minor head injury, and 18 percent were diagnosed with a skull fracture or an intracranial injury. However, the incidence of skull fracture/intracranial injury was 29 percent in children younger than 12 months of age and 4 percent in those older than 12 months of age. An increase in the height of the fall was associated with a higher incidence of serious injury, although low height of fall did not rule out a diagnosis of skull fracture or intracranial injury.

The incidence of seizure, emesis, behavior changes and loss of consciousness did not differ significantly between children found to have a minor head injury and children diagnosed with skull fracture/intracranial injury. Of those determined to have a minor head injury, 29 percent exhibited a behavior change, and 11 percent had emesis. There was a very high incidence (94 percent) of skull fracture/intracranial injury associated with scalp abnormality. Depressed level of consciousness and presence of skull fracture/intracranial injury were significantly correlated, but 92 percent of children with an isolated skull fracture and 75 percent with intracranial injury had a normal level of consciousness and a nonfocal neurologic examination.

The authors conclude from their data that the incidence of serious head injury from a fall is greatest in children younger than 12 months of age. Many children who apparently have minor falls may have sustained significant head injury, even in the absence of clinical signs and symptoms. Physicians should have a low threshold for ordering imaging studies in children who have fallen. A CT scan could appropriately be ordered, but a skull radiograph may be acceptable in some situations because it is easier to perform and does not require sedation. The authors report that children who fall 3 ft or less and have normal results on scalp examination and no history of neurologic symptoms do not need radiologic evaluation.

editor's note: This study seems to dispel the notion held by many physicians that if a child does not lose consciousness, the chance of a serious head injury is very small. In addition, many children with minor head injuries may exhibit behavior changes and vomiting. It appears that exercising clinical judgment and maintaining close follow-up is still prudent. In addition, there should be a very low threshold for ordering a cranial CT scan in a child under one year of age who has fallen. Obviously, more studies are needed to better define historic and clinical criteria for diagnostic imaging.—j.t.k.