Immediate posttraumatic seizures: Is routine hospitalization necessary?

Objective: A recent Internet survey of pediatric neurosurgeons showed that 86% routinely admitted children with immediate posttraumatic seizures (PTS) for a brief period of observation. We wished to determine whether certain children meeting predefined criteria could instead be safely discharged fro m the emergency room. Methods: We reviewed the records of children admitted during the past 5 years with a diagnosis of seizure and head injury. Child ren with a minor head injury, a PTS occurring within 24 h of injury and no intracranial abnormalities on admission CT scan were included. Children wit h previous neurological conditions, a history of prior seizures (other than PTS or febrile seizures), a prior history of anticonvulsant use, or intrac ranial abnormalities on the admission CT scan were excluded. Records were a bstracted for child's age, gender, length of admission, previous history of PTS or febrile seizures, mechanism of injury, location of impact, time bet ween impact and PTS, the number, length and type of PTS, Glasgow Coma Score (GCS) on admission, subsequent complications and hospital costs. Results: Seventy-one children met the inclusion criteria. Eleven children presented to the emergency room with prolonged seizures, transient apnea or persisten tly low GCS and required admission to the intensive care unit (ICU). Among the 60 remaining children with simple PTS, none had further seizures during the follow-up period, and none had significant complications. The average cost of hospitalization was known for 58 children; after excluding the cost s for 5 patients who were admitted to the ICU, the average hospital cost am ounted to USD 1,615 per patient. Conclusions: Our data suggest that childre n with isolated minor head injuries and simple PTS who recover fully in the emergency room, whose CT scans show no intracranial abnormalities and who have no prior history of neurological disease, epilepsy or anticonvulsant u se are at low risk for recurrent seizures or neurological complications, an d could potentially be sent home to a reliable caretaker and a stable home situation. However, because of the limited sample size in this study, the s tatistical risk of a bad outcome may be as high as 9%; we therefore suggest that much larger studies are potentially needed before this becomes a stan dard policy.