Clinical predictors of post-traumatic seizures in children with head trauma.

STUDY OBJECTIVES

To determine the clinical characteristics associated with early post-traumatic seizures in children with head trauma. DESIGN, SETTING, AND TYPE OF PARTICIPANTS: Retrospective chart review; urban trauma center/pediatric emergency department. Trauma patients aged 3 months to 15 years given discharge diagnosis ICD-9-CM codes indicating head trauma and seen from 1988 to 1990 were eligible for the study.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Of 194 patients, 96% suffered blunt trauma and 53% had a loss of consciousness. Fifty-one percent of 141 computed tomography (CT) scans of the head were abnormal, most frequently demonstrating skull fractures (34%), subdural hematomas (15%), and cerebral contusions (14%). Eighteen patients (9.3%) suffered post-traumatic seizures. A loss of consciousness, a low Glasgow Coma Scale (GCS) score (3 to 8), and an abnormal CT scan were associated with post-traumatic seizures (P < .02, .001, and .02, respectively). However, only a low GCS score was predictive of post-traumatic seizures when these factors were considered simultaneously (P < .001), with 38.7% of patients with low GCS scores suffering post-traumatic seizures and 3.8% of patients with high GCS scores suffering post-traumatic seizures. In children with low GCS scores, treatment with phenytoin was associated with a decrease in post-traumatic seizures.

CONCLUSION

In the pediatric head trauma patient, a GCS score of 3 to 8 appears to be predictive of post-traumatic seizures. The data from this retrospective study are consistent with the hypothesis that prophylactic phenytoin reduces post-traumatic seizures in the pediatric head trauma patient with a low GCS score.