Authors :
Presenting Author: Fidelia Gaba, BS – University of Pittsburgh, School of Medicine
Jasmine Hect, BS – University of Pittsburgh, School of Medicine
Emily Harford, BS – 2. Children’s Hospital of Pittsburgh, University of Pittsburgh Medical Center
Taylor Abel, MD, FAES, FAANS – Children’s Hospital of Pittsburgh, University of Pittsburgh Medical Center
Rationale:
Anywhere from 7 to 20% of children with epilepsy have drug resistant epilepsy (DRE). A major goal of neurosurgical treatment for pediatric patients with DRE is the cessation of drop attacks using disconnective surgeries such as corpus callosotomy (CC) or MRI-guided laser interstitial thermal therapy corpus callosum ablation (CCA). It is unclear, however, what patient characteristics are predictors for complete drop attack freedom after disconnective surgery. Previous works have found that the thickness of the isthmus and genu in the corpus callosum were predictive factors for first drug efficiency in self-limited epilepsy with centrotemporal spikes.
We hypothesize that the thickness of the corpus callosum will be significantly associated with seizure freedom or lack thereof following disconnective surgery in pediatric patients with DRE. We also explore the application of correlational and differential tractography methods to identify what diffusion MRI metrics may be correlated with response to surgery and identify variations in microstructure between responders and nonresponders to surgery.
Methods:
The study cohort consists of 23 pediatric patients with DRE who underwent CC or CCA as a treatment for atonic seizures.The width of the genu, spenium, and trunk of the corpus callosum as well as the total apical to posterior length was measured on T1-weighted images. Mann-Whiney tests were used to compare mean lengths in patients that were drop attack free (responders), and not drop attack free (nonresponders) postoperatively. Data were analyzed in GraphPad Prism (Version 10.2.3 (347), April 21, 2024).
DSI studios was used to field correct preoperative images and reconstruct the diffusion data using a Q-Space Diffeomorphic Reconstruction (QSDR). After reconstruction, correlational tractography was applied to map tracts that have anisotropy correlated with response/nonresponse to surgery and differential tractography was applied to visualize tracts showing degeneration in corpus callosum compared to controls (age- and sex-matched regressed images generated from the Human Connectome Project Developmental database).
Results:
At longest follow up responders to surgery had a total length of the corpus callosum that was 9.19% shorter than nonresponders (p = 0.029). Correlational tractography results suggested a higher mean diffusivity (FDR = 0.089), higher fractional anisotropy (FDR = 0.214) and lower quantitative anisotropy (FDR = 0.190) correlated with being a responder to surgery. Differential tractography showed that both responders and nonresponders had tracks showing 20% or more degeneration in the periphery of the corpus callosum but nonresponders to surgery had more medial, crossing tracks that showed degeneration when compared to age-sex-matched controls.
Conclusions:
Our study suggests that patients with longer apical to posterior lengths of their corpus callosum may not respond as well to corpus callosotomy. It also lays the methodological framework for use of diffusion MRI metrics to identify patients who may be nonresponders to surgery.
Funding: This study was supported by the Children’s Hospital of Pittsburgh Foundation and National Institutes of Health grant R21DC019217.