Abstracts

RATIONALE: After attending this presentation, clinicians will learn to monitor anticonvulsant levels of older drugs more frequently than recommended by current guidelines to avoid the occurrence of breakthrough seizures in women with epilepsy during pregnancy.
METHODS: Thirty-five women with epilepsy were followed during 44 pregnancies as part of the Women[ssquote]s Health in Epilepsy program from 7/1994-4/2002. All women were scheduled to see an epilepsy physician and an epilepsy nurse specialist at monthly intervals from the discovery of the pregnancy until the 36th week of gestation. Trough anticonvulsant levels were ordered monthly from the beginning of pregnancy until week 24, biweekly from week 25-35, and weekly from week 36 until delivery. Data analysis was limited to those pregnancies which occurred while the mother was taking any one of the anticonvulsants under study - namely phenobarbital, phenytoin, carbamazepine, or valproic acid - in monotherapy for the duration of her pregnancy. Anticonvulsant levels were examined along with requirements for adjustments of doses and the occurrence of seizures.
RESULTS: Of the forty-four pregnancies reviewed, four were prematurely terminated, eight were excluded due to polypharmacy, three were in women taking no anticonvulsant agents, and five were in women using newer anticonvulsant agents, and one patient was lost to follow-up. Of the remaining twenty-three pregnancies, six were in women taking phenobarbital, six were in women taking phenytoin, eight were in women taking carbamazepine, and three were in women taking valproic acid. Of these, 4/6 pregnancies on phenytoin and 4/6 pregnancies on phenobarbital required more than four adjustments of dose during the pregnancy to maintain levels in the patient[ssquote]s optimal range, i.e., in the range of levels shown to be associated with the patient remaining seizure-free. Half of all pregnancies on these two agents were associated with breakthrough seizures when anticonvulsant levels fell below the patient[ssquote]s therapeutic range. 5/8 pregnancies in six women on carbamazepine similarly required adjustment of anticonvulsant doses more than four times during the pregnancy; two of these women (accounting for four pregnancies) had breakthrough seizures when levels dropped. By contrast, only one of the pregnancies occurring in women taking valproic acid monotherapy was associated with dramatic alterations in blood level.
CONCLUSIONS: Women with epilepsy taking older anticonvulsant agents are at risk for breakthrough seizures due to alterations in anticonvulsant levels. Anticonvulsant levels must be monitored more frequently than current recommendations would suggest in order to avoid seizures during pregnancy and the attendant risks of adverse pregnancy outcome. Reasons for this might include the tendency for older anticonvulsants to be dependent upon hepatic P-450 enzymatic system for metabolism; activation of this system occurs during pregnancy.