Abstracts

Approximately 50% patients will achieve early seizure control on the first drug regimen. There is little information on predictors of success. We have evaluated factors associated with seizure control on the first AED (defined as [ge]12 months seizure free) in an unselected cohort of 890 adolescent and adult patients with newly diagnosed epilepsy of whom 814 were followed longitudinally. We examined demographic characteristics, prior medical history, family history of epilepsy, seizure characteristics (including, duration of seizures, number of seizures prior to first AED, seizure semiology, seizure clusters and status epilepticus), and results of neurodiagnostic studies (imaging and EEG) to identify predictors of seizure control. Multivariate Cox proportional hazard models adjusting for age were constructed. There were 178 adolescents (21.9%) and 92 (11%) people age 65 and older. There were 422 males (51%). Generalized tonic clonic seizures occurred in 670 cases (82%), partial seizures in 135 (16%), and other seizure types in 9 (2%).
Etiology was identified in 247 (30.3%). Only number of pretreatment seizures and a history of seizure predicted seizure control. The likelihood of seizure control increased with decreasing number of pretreatment seizures and when there was no history of seizure clusters.[table1] Frequency and pattern of seizures were the only predictor of seizure control with the first anticonvulsant regime in this large cohort of newly diagnosed patients. Etiology, seizure type, and demographic characteristics were not predictive. This suggests that seizure severity and response are determined prior to initial therapy.