Abstracts

Rationale: Levetiracetam (LEV) is a well tolerated antiepileptic drug shown to be effective for the adjunctive treatment of partial-onset seizure. However, little is known about prognostic factors for efficacy and tolerance of LEV. We evaluated the prognostic implications of LEV efficacy in adult refractory partial-onset epilepsy with a special interest in the efficacy at six months and long-term lack of efficacy.Methods: We retrospectively analysed 145 patients aged at least 18, presenting partial-onset epilepsy, treated with LEV as adjunctive therapy. Patients had seizure at least once a month for 3 months before commencing LEV therapy. All patients were followed up for more than six months. Outcome was measured by the mean monthly seizure frequency. Responder of LEV therapy at six months was defined as the patient with 50% or greater reduction during 4-6 months in seizure frequency from baseline. Patient with lack of efficacy was determined as responder during 4-6 months but became non-responder increasing seizure frequency again during long-term follow-up. Clinical characteristics and recent diagnostic modalities were considered as prognostic factors. Factors significant in univariate analysis were included binary logistic regression model.Results: Of 145 patients, 31 patients (21%) were completely seizure-free and 58 patients (40%) were responders (? 50% reduction of seizure frequency) at six months. Temporal lobe epilepsy (TLE) (Odds ratio=2.384, p=0.019), the presence of interictal epileptiform discharge in routine electroencephalography (OR=0.439, p=0.039) and secondary generalized tonic-clonic seizure (GTC) (OR=0.335, p=0.044) were independently related to response of LEV at six months. Mean duration of follow-up period was 21.5 13.1 months in responders at six months. Lack of LEV efficacy was observed in 19 (33 %) of responders, mainly occurred during 18 months. In patients with remission at six months, lack of LEV efficacy was significantly rare (OR=0.091, p<0.001).Conclusions: This study showed that presence of interictal epileptiform discharge and secondary GTC unfavorable influence LEV efficacy at six months, otherwise, diagnosis as TLE is a good predictor. Seizure-free patients at six months showed sustained LEV efficacy in long-term follow-up.