Abstracts

Rationale: Relative seizure reduction is the commonly used outcome measure in epilepsy treatment trial designs. Many of these trials specify the required baseline seizure frequency of three or more seizures/month to observe a statistically relevant change after treatment. We aimed to determine the proportion of epilepsy patients that would be eligible for enrollment into trials based on seizure frequencyMethods: We collected longitudinal seizure frequency data from a carefully-defined epilepsy population. Included patients were ≥17 years old and had ≥12 month follow-up and ≥2 clinic visits. Three groups were selected corresponding to the syndromes most commonly selected for epilepsy treatment trials: 1) Focal Epilepsy: All patients with seizures consisting of any combination of simple partial, complex partial, and secondary generalized seizures. 2) Lennox-Gastaut Syndrome: All patients diagnosed with either symptomatic or cryptogenic Lennox-Gastaut Syndrome 3) Primary Generalized Epilepsy: All patients in whom seizures are confined to primary generalized tonic-clonic seizures. The seizure frequency is defined in seizures per month, and remission as seizure frequency =0 for at least 12 months. We calculated log10 seizure frequency to determine the cumulative distribution of seizure frequency.Results: From a population of 1639 patients, 725 met the inclusion criteria. The seizure frequency and remission rates of three groups including all patients and patients who have not reached remission for at least 12 months are seen in Table 1 and Table 2. The cumulative seizure frequency distribution is represented in Figure. Table 2 shows the seizure frequency rates of patients with ongoing seizures and who did not reach remission for at least 12 months.Conclusions: About 50% of patients with medically intractable epilepsy in these often studied syndromes will not qualify for epilepsy trials given a minimum baseline seizure frequency requirement of ≥3 seizures per 4-8 weeks. Only about 45 % of partial epilepsy, 55% of LGS patients, and only about 40% of IGE patients will be represented in typical epilepsy trials. Cumulative seizure rate distributions can be used to estimate number of patients available for recruitment into epilepsy clinical trials. Trials based on a high seizure frequency and frequency change may be limiting generalizability of trials to a small proportion of medically intractable patients.