Influence of Diagnostic Testing on Surgical Outcomes in Non-Lesional Focal Pediatric Epilepsy: Insights From PERC Cohort

Document Type

Conference Proceeding

Publication Date

12-9-2024

Abstract

Rationale: Presurgical evaluations are crucial in determining the eligibility for surgery especially in non-lesional focal epilepsy (NLFE) cases. However, the lack of standardized evaluation methods creates uncertainties with regard to the influence of diagnostic tests on surgical eligibility and seizure outcomes. We analyzed the impact of MEG, PET, and SPECT evaluation on both surgical decisions and long-term outcomes in a multicenter pediatric NLFE cohort.
Methods: The Pediatric Epilepsy Research Consortium (PERC) Surgery Database is a prospective, observational study collecting data between Nov 2017 and Feb 2024 on children 0-18 years referred for epilepsy surgery across 23 US pediatric epilepsy centers. We assessed the relationship between presurgical evaluations (total number and type of tests) and the decision to offer surgery, the type of surgery offered (1-stage vs. 2-stage using invasive EEG), and seizure-free outcomes using multivariate logistic regression analysis.

Results: 296 NLFE patients from the PERC database (43% females, 74% white), with a mean age of 6.2 years at seizure onset were included. Patients failed 3.7 anti-seizure medications on average before undergoing phase I evaluation (mean age 12.1 years), followed by surgery at an average age of 13.2 years. In addition to MRI and non-invasive EEG, patients underwent MEG (113, 38%), SPECT (85, 29%), and PET (249, 84%) alone or in combination, according to individual center preferences (Table 1). Epilepsy surgery was offered to 221 individuals (75%): 149 (67%) were recommended invasive EEG evaluation, and 72 (33%) for single-stage surgery. At the time of analysis, 163 patients (55%) had undergone epilepsy surgery (101 after a 2-stage procedure and 62 after a 1-stage procedure).

A higher total number of presurgical tests was associated with the decision to offer invasive EEG monitoring (p=0.004) and surgery (p=0.02). Non-invasive evaluation increased the odds of offering invasive EEG monitoring to various degrees (Table 2). The presence of divergent results increased the likelihood of additional tests (p< 0.001), while congruent results increased the odds of offering surgery. No individual test or combination of tests predicted the decision to offer surgery. Despite the varied approaches to presurgical evaluation across centers, when stratifying surgeries by temporal/extratemporal or grouping by dominant versus nondominant hemispheres, seizure freedom at 2 years post-op was not impacted by the cumulative number of tests, their combinations, or congruent results (Table 2).

Comments

American Epilepsy Society Annual Meeting, December 6-10, 2024, Los Angeles, CA

Link to Full Text

Share

COinS