Delayed Surgical Evaluation After Drug-Resistant Epilepsy Diagnosis Worsens Outcomes in Children: A Multicenter Study.
TL;DR
Delays from DRE diagnosis were independently associated with reduced seizure freedom, supporting presurgical evaluation within 1 year as an evidence-based quality benchmark, and DRE-to-evaluation interval, not total epilepsy duration, predicted outcomes.
Abstract
Background
AND
Objectives
Despite its proven effectiveness, epilepsy surgery for drug-resistant epilepsy (DRE) remains underutilized and frequently delayed. Previous studies of epilepsy duration before surgery-using variable delay thresholds (2-20 years)-were small, single-center cohorts focused mainly on temporal/frontal lobe epilepsy, showed better seizure freedom with earlier surgery, but did not distinguish total epilepsy duration from DRE duration. As contemporary epilepsy surgery now includes broader indications and emphasizes faster evaluation, the timing and impact of evaluation across this wider population remain unclear. We examined factors associated with evaluation timing from DRE diagnosis and its effect on surgical outcomes in a large multicenter cohort.
Methods
Using a prospective database across 29 US centers, we analyzed associations between patient and epilepsy factors and DRE-to-evaluation interval-defined as the interval from DRE diagnosis to phase 1 video-EEG admission, categorized as shorter (<1 year) or longer (≥1 year)-and compared seizure freedom between groups using multivariable logistic regression adjusted for etiology, seizure type, neuroimaging, and surgical factors.
Results
Among 1,310 children, 720 (55%) had shorter and 590 (45%) longer DRE-to-evaluation intervals. Shorter interval was associated with lesional epilepsy (OR 1.65, 95% CI 1.30-2.08), focal seizures (2.80, 2.13-3.70), and normal neurologic exams (1.86, 1.49-2.33). Structural congenital and acquired etiologies were linked to shorter interval, while genetic etiologies (1.73, 1.30-2.32) were linked to longer interval. Among 624 surgical patients (357 shorter, 267 longer), seizure freedom occurred in 53% vs 27% (3.04, 2.17-4.29; p < 0.01). After adjustment, longer interval remained independently associated with lower seizure freedom (0.59, 0.35-1.00; p = 0.0497). DRE-to-evaluation interval, not total epilepsy duration, predicted outcomes. DISCUSION In this first large multicenter study across diverse epilepsy types applying a 1-year benchmark, nearly half of pediatric patients experienced delays, particularly those with MRI-negative, generalized, or genetic epilepsies. Delays from DRE diagnosis were independently associated with reduced seizure freedom, supporting presurgical evaluation within 1 year as an evidence-based quality benchmark.