Abstracts

Rationale: There is little evidence-based guidance for the treatment of medically refractory epilepsy patients. We previously showed (Poolos et al., Neurology, 2012) that only the antiepileptic drug (AED) combination of lamotrigine (LTG) and valproate (VPA) had superior efficacy compared to other combinations of eight frequently-used AEDs in a refractory, developmentally disabled population. We now expand our prior analysis to include four additional AEDs and an additional five years of patient records.Methods: We retrospectively analyzed medical records, consisting of monthly seizure frequency and AED dosages, for 164 developmentally disabled adults with refractory epilepsy housed in two Washington State institutions. On average, each patient had 206 months (17.2 years) of records, and had been exposed to six different AED regimens of up to three concurrent AEDs. We quantified the ratio of average seizure frequency observed during treatment with each AED regimen, to the seizure frequency observed in the aggregate average of all other AED regimens to which the patient had been exposed. This seizure frequency ratio (SFR) gave a within-patient measure of response to each AED regimen that normalized for variation in baseline seizure frequency among patients.Results: We found that patients exposed to two concurrent AEDs had a seizure frequency that was 81% of that in patients exposed to monotherapy (n=127, p=.017), demonstrating a benefit of duotherapy compared to monotherapy. However patients exposed to three AEDs at once showed no improvement compared to those on duotherapy (SFR=102%, n=110, p>.05). Only two AED regimens showed significantly better efficacy compared to the aggregate average: LTG/VPA (SFR=51%, n=40, p=1.7 x10-6) and LTG/VPA/topiramate (TPM; SFR=46%, n=15, p=.048). Of note, neither LTG in monotherapy (SFR=82%, n=32, p=.144) or VPA in monotherapy (SFR=128%, n=54, p=.070) yielded significantly better efficacy when used alone. Two AED regimens showed significantly worse efficacy compared to the aggregate average: carbamazepine (CBZ)/VPA (SFR=137%, n=57, p=.011) and LTG/CBZ (SFR=134%, n=25, p=.042). Notably, CBZ/VPA was the most frequently utilized AED combination in this sample, followed by VPA/phenytoin (PHT) and LTG/VPA.Conclusions: This expanded analysis of the 12 most frequently-used AEDs in a refractory patient population spanning over 17 years of follow-up per patient reinforces our previously published finding that only LTG/VPA, or a combination containing LTG/VPA, exerts significantly better efficacy. We also find that use of more than two concurrent AEDs yields no improvement in seizure control. Examining additional relationships between some newer AED combinations was limited by small sample sizes, but the significance of responses specific to duotherapy (versus mono- or tri-therapy) deserves further investigation in larger samples. These findings suggest that LTG and VPA in combination yield a synergistic increase in efficacy compared to the sum of the efficacies of each drug individually.