Abstracts

Rationale: To investigate seizure outcome and its predictors in patients with profound intellectual disability (ID) managed in an academic epilepsy center. Methods: We retrospectively analyzed the data from institutionalized patients with ID and history of seizures seen in the epilepsy clinic over a 5-year period. Pertinent information extracted at the first visit (FV) and last visit (LV) included demographic data, epilepsy syndrome, seizure frequency, changes in antiepileptic drugs (AEDs) and vagus nerve stimulator (VNS) implants. We classified the seizure frequency into: extremely frequent (?5/month); frequent (1-4/month); occasional (4-11/year); rare (1-3/year); seizure free (no seizures in 1 year). Based on the change in seizure frequency, we defined the seizure control between FV and LV as: improved; worsened; unchanged seizure free; unchanged seizure frequency. We then defined 2 seizure outcome groups: favorable (i.e., patients who remained seizure free or improved); unfavorable (i.e., patients who continued to have seizures at the same frequency or worsened). For consistency, the patients with rare or occasional seizures at FV who continued to have seizures at the respective frequency at LV were included in the unchanged seizure frequency group. Based on EEG and imaging, we classified the epilepsy syndrome as: focal; multifocal; generalized; unknown. Results: The patient and seizure outcome data are shown in Table 1. There were 95 patients (51 males), aged 21-74 years, followed for 5-60 months (mean 42). The epilepsy syndrome was defined in 56 patients. Seizure outcome was favorable in 61 (64.2%) and unfavorable in 34 (35.8%) patients. Between FV and LV, the number of AEDs increased (mean 1.78 vs. 2.18), AEDs were completely discontinued in 2 patients, and VNS was implanted in 10 patients. At LV, lamotrigine, levetiracetam and valproate were the most commonly used AEDs in 27-45% of patients while ten other AEDs were used in 1-15% of patients. Univariate analysis (Table 2) showed that the epilepsy syndrome, seizure frequency at FV or VNS implantation were not associated with outcome; however, increase in the number of AEDs, addition of lamotrigine and addition of levetiracetam were significantly associated with unfavorable outcome. Multivariate logistic regression analysis using stepwise variable selection showed that increase in the number of AEDs [odds ratio (OR) 4.70, 95% confidence interval (CI) 1.75-12.61, p=0.002] and addition of lamotrigine in lamotrigine-na ve pts (OR 3.82, 95% CI 1.31-11.13, p=0.014) to be significantly associated with unfavorable outcome. Conclusions: Nearly two-thirds of patients with profound ID and epilepsy achieved favorable seizure outcome. The epilepsy syndrome, seizure frequency at first visit and VNS did not predict outcome. However, the need for increased number of AEDs to control seizures, and the addition of broad-spectrum AEDs, particularly lamotrigine, predicted unfavorable outcome. These results highlight the refractory nature of epilepsy and the challenges of management in this population. Acknowledgement: The study was supported in part by a grant from the Commonwealth of Kentucky.