Long-term seizure outcome after resective surgery in patients evaluated with intracranial electrodes.

PURPOSE: Despite advances in "noninvasive" localization techniques, many patients with medically intractable epilepsy require the placement of subdural (subdural grid electrode, SDE) and/or depth electrodes for the identification and definition of extent of the epileptic region. This study investigates the trends in longitudinal seizure outcome and its predictors in this group.

METHODS: We reviewed the medical records, and electroencephalography (EEG) data of 414 consecutive patients who underwent intracranial electrode placement (SDE and/or depth electrodes) at Cleveland Clinic Epilepsy Center between 1998 and 2008. A favorable outcome was defined as complete seizure freedom, discounting any auras or seizures that occurred within the first postoperative week. Survival curves were constructed, and Cox proportional hazard modeling was used to identify outcome predictors.

KEY FINDINGS: The estimated probability of complete seizure freedom was 61% (95% confidence interval [CI] 58-64%) at one postoperative year, 47% (95% CI 44-50%) at 3 years, 42% (95% CI 39-45%) at 5 years, and 33% (95% CI 28-38%) at 10 years. Half of all seizure recurrences occurred within the first two postoperative months. Subsequently, the rate of seizure freedom declined by 4-5% every 2-3 years. After multivariate analysis, two independent predictors of seizure recurrence were identified: (1) prior resective surgery (p ≤ 0.002), mostly in patients with temporal lobe resections, and (2) sublobar or multilobar resection (p ≤ 0.02), mostly in patients following frontal lobe resections.

SIGNIFICANCE: Favorable seizure outcomes are possible in the complex epilepsy population requiring invasive EEG studies. We propose that mislocalization of the epileptogenic zone or its incomplete resection account for early postoperative recurrences, whereas epileptogenesis may lead to later relapses.