Long term surgical outcome and its predictors in lesional posterior cortex epilepsy

Posterior cortex epilepsy (PCE) is associated with high drug resistance and variable surgical outcomes. Data on long-term outcomes and predictors, particularly from low- and middle-income countries, are limited. We evaluated long-term seizure outcomes and their predictors following resective surgery for lesional PCE.

We retrospectively analyzed a prospectively maintained database from a tertiary epilepsy center in India. Of 3560 patients evaluated for drug-resistant PCE, 191 underwent surgery (2001-2019); 184 with ≥ 1-year follow-up were included. Clinical, imaging, and electrophysiological variables were assessed. Seizure outcomes were classified using Engel criteria. Predictors were evaluated using bivariate and multivariate logistic regression, along with Cox proportional hazards analysis.

At a mean follow-up of 7 years (range 1-19), 63% achieved Engel class I. Gliosis was the most common etiology (∼51%). Antiseizure medications were discontinued in 9.2%, and 7.3% required redo surgery.Bivariate analysis showed favorable outcome with older age at onset (OR 1.11, p = 0.011), concordant ictal EEG (OR 2.31, p = 0.037), and complete resection (OR 0.105, p = 0.0006), while multilobar MRI involvement (OR 3.4, p = 0.0024), specifically temporo parietal and temporoparieto occipital patterns, acute postoperative seizures (OR 4.01, p = 0.0024), and abnormal postoperative EEG (OR 4.41-6.97, p < 0.01) predicted unfavorable outcome. On multivariate analysis, incomplete resection (adjusted OR 5.62, p = 0.001) and abnormal EEG at 1 year (adjusted OR 6.72, p < 0.001) remained independent predictors. Seizure freedom declined over time (92% at 1 year to 32% at 15 years), with better outcomes following complete resection.

Resective surgery in selected lesional PCE yields meaningful long-term seizure control. Completeness of resection and postoperative EEG are the strongest predictors of outcome.