Abstracts

Rationale: It is not known whether the use of cyclic adjunctive progesterone supplement for the treatment of intractable seizures in women with epilepsy (WWE) affects quality of life (QOL). The purpose of this analysis is to determine if progesterone therapy in the NIH Progesterone Trial affected QOL measures differently than placebo in WWE who had intractable seizures. Methods: Two hundred ninety-four women with intractable seizures were designated to catamenial or non-catamenial strata using established criteria (Herzog et al 1997 for the Progesterone Trial Study Group) after 3 baseline months and randomized to 3 months of adjunctive treatment with progesterone or placebo in the NIH Progesterone Trial. Changes in QOL, assessed by measures of the 7 QOLIE-31 domains (dependent factors) were compared by treatment and strata (independent factors) and changes in estradiol (E2) and progesterone (P4) levels (covariates) using MANOVA. Changes in significant factors were compared by treatment and stratum using Wilcoxon rank-sum test and correlated using Spearman analysis. Results: One QOLIE-31 domain, seizure worry, showed a significant between-subjects effect (p = .003) with treatment being a significant factor (p = .008) and treatment and stratum showing significant interaction (p = .007). Changes in seizure worry were an order of magnitude more significant with progesterone treatment in the catamenial stratum than in the non-catamenial stratum. There were no significant changes with placebo in either stratum (Table). There were significant correlations between changes in seizure worry and changes in E2 (r = -.385, p = .006) and P4 (r = .318, p = .026) levels for the catamenial stratum only. There was also a significant correlation between changes in seizure worry and seizure frequency (r = -.374, p = .002) in progesterone treated subjects only. By way of pertinent negatives, there was no significant correlation between baseline to treatment changes in seizure worry scores and changes in measures of depression using Beck's Depression Inventory and Profile of Mood States questionnaires. Changes in depression scores showed no significant difference between progesterone and placebo treatment. Conclusions: Seizure worry is affected favorably by progesterone treatment in WWE, especially in women with catamenially exacerbated seizures. The changes in seizure worry scores correlate with treatment related changes in hormone levels and seizure frequency. Supported by NIH RO1 NS39466