Comparison of Epilepsy Surgery Conference Predicted Outcomes to Actual Outcomes
Abstract number :
2.451
Submission category :
9. Surgery / 9B. Pediatrics
Year :
2024
Submission ID :
403
Source :
www.aesnet.org
Presentation date :
12/8/2024 12:00:00 AM
Published date :
Authors :
Presenting Author: Purva Choudhari, MD – Nationwide Children's Hospital
Katie Bonds, DO – Children's Hospital Colorado
Krista Eschbach, MD – University of Colorado Anschutz Medical Campus – Children’s Hospital Colorado
Kevin Chapman, MD – Phoenix Children's Hospital
Allyson Alexander, MD, PhD – Children's Hospital Colorado
Angela Price, MD – University of Texas Southwestern
Wynne Bird, RN – Children's Hospital Colorado
Charuta Joshi, MD – University of Texas Southwestern
Rationale: Drug resistant epilepsy (DRE) affects 30% of patients with epilepsy leading to significant morbidity and increased mortality. It is standard practice to present DRE patients at a multidisciplinary epilepsy surgery conference (ESC) to discuss the optimal surgical interventions for such patients.1 To reduce associated biases, a REDCap survey was established in 2018 at Children’s Hospital Colorado (CHC) and thereafter, at Children’s Health Dallas (CHD), for team members to anonymously evaluate the next best management for each patient during ESC. There is a paucity of data assessing the objectivity and value of pediatric ESC. This study’s aim was to compare actual seizure outcome post-operatively as predicted by a panel of experts during ESC.
Methods: Patients who had undergone epilepsy surgery with a 12-month (+/- 3 months) follow-up, with corresponding completion of REDCap ESC surveys by at least 50% of epileptologists, were included in this study. A review of 88 patients at CHC from 12/2021-12/2022 and 52 patients at CHD from 6/2022-12/2023 yielded 20 and 10 patients, respectively, who met the inclusion criteria. Seizure outcome was measured by percent seizure reduction at time of last follow-up. A consensus variable was defined to measure the strength of the team’s agreement for predicted seizure outcome: strong for 71-100% agreement, fair for 50-70% agreement, and low for < 50% agreement. The predicted seizure outcome for each patient was determined by the predicted outcome with the highest consensus. If there was no consensus, then the predicted outcome was considered uncertain (5 patients). A seizure freedom score was calculated for each patient.2
Results: Of the 30 patients who met inclusion criteria, a resection/ablation or disconnection was completed in 77% of patients while 23% underwent neuromodulation. The epilepsy etiology of 77% of patients was focal while 23% either had multifocal etiology or Lennox Gastaut syndrome (LGS). The consensus for most predicted outcomes was fair (47%) and strong (40%). The majority (76%, 19/25) of predicted outcomes were underpredicted compared to the actual seizure outcomes. The ESC members accurately predicted 8% (2/25) of the actual seizure outcomes. Underestimation of predicted outcomes was not associated with palliative surgery/neuromodulation or a resective surgery. Predicted outcomes were overestimated in 2 patients who had lesionectomy/ablation and in 2 patients who had lobectomies. Notably, 4 patients who received neuromodulation for LGS/multifocal epilepsy were predicted for < 65% seizure reduction and had a low seizure freedom score (0-2) but had > 65% seizure reduction post-operatively.
Conclusions: Frequently, an expert panel of epileptologists underestimated patients’ seizure outcomes without clear association with epilepsy etiology or surgery type.
References
1. Englot DJ, Birk H and Chang EF. Seizure outcomes in nonresective epilepsy surgery: an update. Neurosurg Rev 2017;40(2):181-194.
2. Garcia Gracia C, Yardi R, Kattan MW, et al. Seizure freedom score: a new simple method to predict success of epilepsy surgery. Epilepsia. 2015;56(3):359-365. doi:10.1111/epi.12892
Funding: No funding.
Surgery