Prophylactic Anticonvulsants After Neurosurgery

Abstract

Six prospective, controlled trials have examined the effects of antiepileptic drugs (AEDs) given to prevent the occurrence of seizures following neurosurgery. Some studies have concentrated on specific reasons for the neurosurgery (brain tumor) while others have included people with a variety of indications for surgery. Phenytoin (PHT) has been studied most, but carbamazepine (CBZ) and phenobarbital (PB) have also been evaluated to some extent. Studies of people with traumatic brain injury (some of whom were operated on) provide some, but less direct, evidence of the prophylactic effects of AEDs after neurosurgery. Despite considerable variation in reasons for the neurosurgery, AEDs given, and study design, the overall conclusions are remarkably consistent. The seizure risk is reduced about 40%-50% for the first week after neurosurgery in those given the older AEDs compared with those given placebo or no treatment. After the first few weeks, none of the drugs has been proven to reduce the incidence of seizures and in most situations the best estimate is essentially no effect, but effects on the order of a 25%-50% reduction in late (epileptic) seizures cannot be ruled out. The new generation of AEDs have not been tested as prophylactic agents after neurosurgery. Although there are no guidelines for prophylaxis following neurosurgery in general, these results are consistent with the guidelines of professional organizations for subsets of neurosurgery cases. Those guidelines consider prophylaxis, especially using PHT, to be an option for the first week after surgery but that the routine use of prophylactic anticonvulsants after the first week is not warranted.

FIGURE 1

Meta-analysis results for seizure prophylaxis clinical trials following craniotomy. The relative risk for each study is marked by a square on a line that indicates the 95% confidence interval (CI) for that relative risk. Meta-analysis results for a time period and drug are marked by a diamond with the center at the overall relative risk estimate and the points extending to the ends of the 95% CI. The author's name is preceded by an asterisk (*) if the study did not use a placebo control. The seizure rate for the active treatment and control arms, the relative risk estimate and its confidence interval, and, for the meta-analyses the p-value, accompany each study or combination of studies. A relative risk of one, representing no treatment effect, is marked by the dashed vertical line. Adapted from Temkin NR. Antiepileptogenesis and seizure prevention trials with antiepileptic drugs: Meta-analysis of controlled trials. Epilepsia 2001;42:515–524, 2001 with permission of Blackwell Publishing, Inc.