Abstracts

Concordance of CURRY with VEEG and PET in MRI-Negative Pediatric Epilepsy

Abstract number : 2.07
Submission category : 3. Neurophysiology / 3G. Computational Analysis & Modeling of EEG
Year : 2021
Submission ID : 1825839
Source : www.aesnet.org
Presentation date : 12/5/2021 12:00:00 PM
Published date : Nov 22, 2021, 06:50 AM

Authors :
Archana Pasupuleti, MD - Children's National Hospital; Emily Matuska, BS - Children's National Hospital; Manu Krishnamurthy, BS - Children's National Hospital; Eleanor Fanto, BS - University of Pennsylvania School of Nursing; John Schreiber, MD - Children's National Hospital; Tayyba Anwar, MD - Children's National Hospital; Dewi Depositario-Cabacar, MD - Children's National Hospital; Tesfaye Zelleke, MD - Children's National Hospital; Chima Oluigbo, MD - Children's National Hospital; Xiaozhen You, PhD - Children's National Hospital; Madison Berl, PhD - Children's National Hospital; William Gaillard, MD - Children's National Hospital

Rationale: Source localization is integral to the epilepsy surgical work-up process. There are numerous methods of source localization varying in accuracy, cost, and feasibility for young patients. We examined the utility of CURRY in identifying seizure onset zones in MRI-negative pediatric epilepsy patients. Seizure freedom rates in MRI-negative patients range from 11-60% suggesting that there is room for improving identification of the surgery target. PET, MEG and ictal SPECT localization accuracy rates range from 40-70%. CURRY is a non-invasive diagnostic tool potentially helpful for children unable to pursue PET, MEG, or SPECT. There is limited literature analyzing CURRY in the pediatric epilepsy population. We evaluated the concordance of source localization results produced by CURRY with other localization methods.

Methods: Prospective 3D source localization was conducted on 16 MRI-negative patients (5 male, mean age=13.9 years, range from 6-22 years) through Compumedics NeuroScan CURRY V.8.0. We conducted descriptive analyses comparing CURRY results to results from vEEG, PET, and final surgical plan. During the pre-surgical evaluation, CURRY was utilized in discussions of final surgery plan. We compared results at the hemisphere and lobe levels to assess lateralization and localization accuracy. Not all patients had results from each of these additional modalities and only 4 patients have undergone surgery at this time.

Results: CURRY demonstrates >60% concordance with standard pre-surgical diagnostic tests. Concordance rates for both hemisphere and lobe ranged from 63-100%. When compared to vEEG and PET, CURRY agreement ranged from 73-75% (hemisphere) and 63-73% (lobe). 3/4 patients who underwent surgery had 100% hemispheric and lobe concordance with the final surgical resection. In the remaining patient, CURRY lateralized hemisphere and lobe; however, sEEG indicated bilateral results and final surgery resulted in bilateral RNS.

Conclusions: In a majority of cases, CURRY was concordant with various diagnostic modalities in the pre-surgical evaluation. CURRY rate of accuracy was as good as vEEG and PET and can provide additional data to help decision-making with the final surgical plan. CURRY is an alternative tool to help guide surgical planning in MRI-negative patients. CURRY is a promising non-invasive source localization tool in pediatric epilepsy that demonstrates the same level of accuracy as current diagnostic tools with less of a burden on patients and a lower relative cost.

Funding: Please list any funding that was received in support of this abstract.: No funding.

Neurophysiology