Characteristics Affecting Decision to Not Offer Epilepsy Surgery Following Presurgical Evaluation Across US Pediatric Epilepsy Centers

Document Type

Conference Proceeding

Publication Date

12-6-2025

Abstract

Rationale: Surgery is an effective, yet underutilized, treatment for many children with drug resistant epilepsy (DRE). While factors associated with favorable outcome following surgery are known, the decision to offer surgical treatment rests with each epilepsy center. Threshold to offer surgery differs across institutions and may be related to center experience, resources, or other factors. We aim to assess the rate at which US centers decline to offer surgery as a treatment option after evaluating a child for surgical candidacy. We describe characteristics associated with this decision and the variability in candidate selection across institutions. Methods: We queried the Pediatric Epilepsy Research Consortium Surgery database, a prospective, observational multicenter study enrolling children 0-18 years at 38 US pediatric epilepsy centers. We included data from sites enrolling all patients undergoing initial evaluation for surgical treatment of DRE. We analyzed patient and epilepsy characteristics, pre-surgical evaluation details, and compared across institutions based on the centers’ decision to offer epilepsy surgery following a phase I evaluation. We performed logistic regression on significant variables’ effect on decision to not offer surgery. Participating institutions provided program data, including annual number of evaluations for epilepsy surgery, epilepsy surgery conference participants, years of program experience, and National Association of Epilepsy Centers (NAEC) level. Correlation was tested between program data and surgery denial rates. Results: Of the 3966 patients in the database, 2480 from 29 NAEC Level 4 centers were included. 532 (21%) were not offered epilepsy surgery, with median rate of 17% (IQR 10%-28%). Multiple factors had significant effect on the likelihood of candidacy (Table 1) and all showed significant variability when evaluated across institutions (Table 2). Logistic regression showed patients with monthly seizures (OR 2.25, 95% CI [1.81, 2.79], p < 0.01), non-structural etiology (OR 2.00, 95% CI [1.55, 2.59], p < 0.01), and those without fMRI (OR 2.49, 95% CI [1.88, 3.33], p < 0.01) had the highest likelihood of being declined surgery. Across centers, these variables were present in patients not offered surgery 0%-51%, 0%-59%, and 0%-61%, respectively. 21 centers provided program details. There was a strong correlation (Pearson’s r = -0.75) between number of conference participants and the rate of not offering surgery. Other variables had weak to no correlation (r < 0.35).

Comments

American Epilepsy Society Annual Meeting, December 5-9, 2025, Atlanta, GA

Link to Full Text

Share

COinS