Abstracts

In multiple prospective studies of severe traumatic brain injury, posttraumatic epilepsy emerges within 2 years in approximately 20% of patients. Antiepileptic agents used to date have suppressed early posttraumatic seizures but not late ones. Multiple experimental models of epileptogenesis suggest sodium valproate (VPA) may be effective in preventing epileptogenesis if given soon after the insult. A pilot trial of rapid delivery of VPA to persons sustaining severe TBI was performed to assess the feasibility of full scale testing of this hypothesis.
18 patients were enrolled over 12 months (3/00 - 2/01) at Kings County Hospital, Bklyn, NY. Consecutive patients with severe head trauma received either VPA (25 mg/kg IV over 4 min then 15 mg/kg/d adjusted to trough levels of 80-100 mcg/ml for 1 week) or phenytoin (IV 20 mg/kg at 150 PE/min then 300 mg qd adjusted to trough levels of 10-20 mcg/ml) on alternating weeks. Deferred consent was obtained. Patients were assesssed for injury severity and complications of trauma and treatment, and followed for two years for seizure occurrence.
Of the18 patients enrolled,15 received follow up. 7 were in the VPA arm and 8 in the phenytoin arm. Median age for the VPA group was 28 yrs (range 20-56) and 29 yrs (4-50) for the non-VPA group. 3 of 7 were women in the VPA group and 2/8 in the non-VPA group. Blood on head CT, was 6/7 and 6/8 respectively.
Initial enrollment rates of 10% of eligible patients rose to 90% over the course of the study. In the treatment group, VPA was administered within 1 hr of TBI 86% of the time. 76% of patients arrived in the ER within 40 minutes of TBI.
Of the 7 patients assigned to the treatment group, 6 received the VPA according to protocol in the ED and 1 received it according to protocol for one week. Due to inexperience with VPA dosing, 2 patients had elevated VPA levels (140) and decreased platelet counts (80 and 116) which rose to prior levels with discontinuation of VPA. 1 patient sustained a temporary depression of mental status which responded to VPA discontinuation. No hemorrhages were noted in either group. 1 of the patients in the VPA group died for reasons not related to VPA. The follow up rate after 2 years was 83%.
Of the 6 surviving patients in the VPA group, 1 (17%) developed seizures in the two years following TBI. Of the 8 patients in the control group, 1 (12.5%) developed seizures.
This study shows that delivering a full IV VPA load within 1 hr of TBI is possible in the inner city population served by Kings County Hospital. Low enrollment precludes demonstrating efficacy of early VPA administration in preventing posttraumatic epilepsy. In addition the limited side effects in supra-therapeutic dosing, suggest rapid VPA protocols are safe in severe TBI.
[Supported by: Abbott Laboratories]