Abstracts

Rationale: Premenstrual dysphoric disorder (PMDD) may be unusually common among women with epilepsy (Shuman et al. Epilepsia 2003;44(S9):A292). The roles of epileptic brain substrates, antiepileptic drugs (AEDs) and hormones, however, remain to be determined. The purpose of this investigation was to determine the relationship of PMDD to epileptic, AED and reproductive endocrine features. Methods: This prospective, multicenter investigation used the standardized and validated Endicott Daily Record of Severity of Problems (DRSP) to designate PMDD (Endicott et al. Arch Women's Mental Health 2006;9:41-49). The 14 item rating scale was completed daily for two consecutive menstrual cycles by 43 women with epilepsy (WWE) on various AED monotherapies (carbamazepine -14, lamotrigine - 20, levetiracetam - 8 and topiramate - 1) and 16 normal controls (NC) at the time of this interim report. DRSP scores were then tallied to determine if premenstrual symptoms were 1) characteristic of the PMDD syndrome, 2) sufficiently severe, 3) more severe than mid follicular symptoms and 4) resulted in functional impairment. Endicott scoring requires that these 4 criteria be met in each of two consecutive cycles for a total score of 8 in order to meet PMDD designation. All of the women provided a mid-luteal blood sample each cycle for batched assays to measure serum estradiol (E) and progesterone (P) as well as AED levels. Results: None of the 59 women fully met the 4 Endicott criteria for the designation of PMDD in each of two consecutive cycles. Seven women met the less strict threshold of 6 of 8 DRSP criteria over two cycles (WWE: 6/43 [14.0%] and NC: 1/16 [6.3%]; Fisher exact test p = .661). DRSP scores were significantly greater with R than L unilateral epileptic foci (R:L [mean ± SD] - 4.00 ± 1.73 v 2.54 ± 1.45; t = 2.252, p = .035). DRSP scores did not differ significantly among the AED groups but DRSP scores showed a significant inverse correlation with LTG levels (Pearson r = -.3660, p = .023) whereas they showed a direct correlation with CBZ levels (r = .3661, p = .072). DRSP scores were greater with anovulatory (midluteal progesterone <5 ng/ml) than ovulatory cycles (1.80 ± .96 v 1.33 ± .91; t = 2.083, p = .040) and changes in depression, anxiety and mood swing scores between the premenstrual and mid follicular phases correlated significantly (Spearman correlations p<.05) with mid-luteal E/P ratios. DRSP scores did not differ significantly between WWE and NC. There was no significant difference between women with primary generalized epilepsy and partial onset seizures. There was no significant correlation with age of seizure onset, duration of epilepsy or seizure frequency. Conclusions: These preliminary results suggest that epilepsy, AED levels and hormone ratios may all influence PMDD in WWE. PMDD scores may be greater with R than L epileptic foci and with anovulatory than ovulatory menstrual cycles. PMDD scores may correlate directly with mid-luteal E/P ratios and inversely with LTG levels. Supported by an investigator-initiated grant from GlaxoSmithKline