Determinants of epilepsy surgery outcomes after invasive EEG: A Bayesian causal inference approach
Poster No:
1351
Submission Type:
Abstract Submission
Authors:
Chifaou Abdallah1, Zhengchen Cai2, Jean Gotman3, Christophe Grova4, Birgit Frauscher5
Institutions:
1McGill University, Montreal, Québec, 2McGill University, Montreal, Quebec, 3Montréal Neurological Institute, Montréal, Québec, 4Concordia University, Montreal, QC, 5Duke University School of Medicine, Durhan, NC
First Author:
Co-Author(s):
Introduction:
Approximately 30% of individuals with epilepsy continue to experience seizures despite multiple medications (1). For these patients, epilepsy surgery, which aims to remove the epileptogenic zone - the brain area that needs to be resected for seizure freedom- is often the only curative treatment (2). Accurate identification and mapping of this zone often require invasive electroencephalography (iEEG), where clinicians use the seizure-onset zone (SOZ) as a proxy. However, even with this advanced technique, 40% of patients still experience disabling seizures post-surgery (3), highlighting gaps in our understanding of surgical success factors. While previous studies have relied on conventional statistical methods and small sample sizes, in this study, we investigated: (i) the effects of the SOZ-resected fraction, non-SOZ-resected fraction (i.e, fraction of resection outside the SOZ), and SOZ volume on achieving a good outcome after surgery; and (ii) how factors such as epilepsy type, presence of eloquent cortex in the SOZ, and histopathology influence the probability of a good outcome.
Methods:
We analyzed data from consecutive patients who underwent iEEG-guided epilepsy surgery across three centers, with at least one year of postoperative follow-up. Since our primary outcome was binary (good versus poor surgical outcomes), we employed logistic regression models. To investigate the causal effect of our variables of interest on a good outcome, we used Bayesian causal inference. We constructed a directed acyclic graph (DAG) (4) based on causal assumptions from both literature and clinical expertise to identify relationships between variables of interest and surgical outcomes. Using a Bayesian workflow, we then estimated each parameter's posterior distribution with 10,000 Markov Chain Monte Carlo samples, providing a robust alternative to conventional point estimates (5). The SOZ and surgical cavity were quantified using volumetric masks, with key variables including SOZ volume, SOZ-resected fraction, and non-SOZ-resected fraction. To facilitate clinical interpretation, we reported marginal effects instead of log-odds. Results are presented probabilistically using the median of the marginal effect, its credible interval (CI)95%, and probability of direction, indicating effect strength ranging from 50% to 100%. Higher pd values reflect greater certainty that the effect differs from overall good outcome probability (for categorical variables) or from zero (for continuous ones).
Results:
We included 200 consecutive patients (43% temporal lobe epilepsy, 52% women, median age, 30 years [range 6-57 years]; 56% good outcome). The SOZ-resected fraction had the strongest positive impact on achieving a good outcome, with a 10% increase in SOZ-resected fraction improving the good outcome probability by 3.7% (median of the marginal effect 3.7%, CI95% [1.2%-6.1%], 99.7% certainty that this effect differs from 0). In contrast, resecting non-SOZ tissue had no clinically relevant effect. Among categorical variables, the key factors for a good outcome, using the 56% probability as a reference, were concordance between multimodal presurgical and iEEG findings (66%, [55%-78%], 92%), posterior quadrant epilepsy (69%, [55%-84%], 95%), followed by operculo-insular epilepsy (62%, [46%-76%], 72%), frontal lobe epilepsy (61%, [50%-72%], 74%), focal and regional SOZ (62%, [54%-71%], 87%; 61%, [50%-71%], 76%), and focal cortical dysplasia II (61%, [51%-73%], 78%). Conversely, a widespread SOZ negatively impacted the probability of a good outcome (33%, [18%-48%], 99%).
·Direct effect of the SOZ-resected fraction on good outcome
Conclusions:
Precise SOZ targeting is crucial for successful iEEG-guided epilepsy surgery, as resecting beyond the SOZ provides no benefit and may risk functional impairment. Widespread SOZ cases require careful evaluation, while surgeries for posterior quadrant and operculo-insular epilepsy should be reconsidered due to their favorable outcomes despite traditionally poor prognoses.
Modeling and Analysis Methods:
Bayesian Modeling 2
EEG/MEG Modeling and Analysis 1
Keywords:
Electroencephaolography (EEG)
ELECTROPHYSIOLOGY
Epilepsy
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
Provide references using APA citation style.
2. Rosenow, F. & Lüders, H. Presurgical evaluation of epilepsy. Brain vol. 124 1683–1700 Preprint at https://doi.org/10.1093/brain/124.9.1683 (2001).
3. Cardinale, F. et al. Stereoelectroencephalography: Retrospective analysis of 742 procedures in a single centre. Brain 142, 2688–2704 (2019).
4. Greenland, S. & Pearl, J. Causal Diagrams. Wiley StatsRef: Statistics Reference Online 1–10 (2017) doi:10.1002/9781118445112.STAT03732.PUB2.
5. Gelman, A. et al. Bayesian Data Analysis. Bayesian Data Analysis (2013) doi:10.1201/B16018.