Discerning whether head injuries in children are accidental or a result of physical abuse is important. Reece and Sege performed a retrospective study of cases of head injury in children to determine the relative incidence of accidental and abusive causes and to characterize the types of head injuries that are likely to be associated with abuse.

The authors reviewed the medical records of 287 cases of head injury in children one week to six years of age who were admitted to an urban pediatric hospital during a six-year period, from January 1986 through December 1991. Of these 287 cases, 233 (81.2 percent) were the result of accidents, and 54 (18.8 percent) were due to abuse.

Specific diagnoses selected for this study included linear, complex, depressed, compound, comminuted and basilar skull fractures; concussion; subgaleal, epidural and subdural hematomas; subarachnoid hemorrhage; closed head injury with loss of consciousness; and fatal head injuries. Cases suspected to result from inflicted injury were further reviewed by a multidisciplinary team to help classify them as definite abuse or unintentional injury as a result of poor supervision or neglect.

Children with head injuries related to abuse tended to be younger than those with accidental injuries. The mean age of the children in the accident group was 2.5 years, compared with a mean age of 0.7 years in the group with inflicted injuries. Boys accounted for 62 percent of the accident group and 57 percent of the abused group.

The most common diagnosis in all of the children was a linear skull fracture, and the second most common diagnosis was a complex skull fracture. Linear skull fractures occurred in 41 percent (22 patients) of the cases of abuse and in 54 percent (126 patients) of the cases of accidental injury. Complex skull fractures were present in 15 percent (eight patients) of the cases of inflicted injury and in 19 percent (44 patients) of the cases of accidental injury.

A significant difference, however, was noted between the two groups in the frequency of subdural hematoma, which accounted for 10 percent (23 patients) of the 233 cases of accidental injury compared with 46 percent (25 patients) of the 54 cases of abuse. Similarly, subarachnoid hemorrhage occurred in 8 percent (19 patients) of the accidental injury group, but in 31 percent (17 patients) of the abused group. Retinal hemorrhage was present in 2 percent (five patients) of the children who sustained accidental injuries but in 33 percent (18 patients) of those who had inflicted injuries. Lastly, old and new cutaneous injuries were noted in 50 percent of the abused children, while only 16 percent of the accidental injury group had such findings.

The most commonly reported mechanism of injury was a fall, and the second most common was a motor vehicle crash. A fall was reported as the “cause” of injury in 17 percent of the abuse cases. Falling from a height of less than 4 ft did not cause severe injury. There were 18 cases of injuries that occurred while the child was in a walker; the injuries in these 18 children were thought to be accidental. There were seven deaths in the abuse group, for a mortality rate of 13 percent. Four deaths (2 percent) occurred in the accidental injury group; three resulted from injuries sustained in a motor vehicle crash and one from injuries sustained in a two-story fall. The abused children were hospitalized for a median duration of 9.5 days (range: one to 51 days), compared with a median hospitalization of 3.0 days (range: one to 91 days) in the accidental injury group.

The authors conclude that inflicted injuries account for a substantial number of cases of head injuries in children, especially in children younger than three years of age. Subdural hematoma, subarachnoid hemorrhage and retinal hemorrhage were markedly more common in abused children. The authors believe child abuse should be strongly suspected when such injuries are present in a child without a history of a fall or with a history of a fall from a relatively low height. The authors also advocate a skeletal survey for children younger than three years when inflicted head injuries are suspected.