Thalamic Functional Connectivity Gradients in Children with Temporal Lobe Epilepsy

Poster No:

347 

Submission Type:

Abstract Submission 

Authors:

Xiyu Feng1, Hua Xie2, Rory Piper1,3, Freya Prentice1, Priyanka Illapani2, Lauren Reppert2, Seok-Jun Hong4, Mohamad Koubeissi5, Torsten Baldeweg1, Leigh Sepeta2

Institutions:

1UCL Great Ormond Street Institute of Child Health, London, United Kingdom, 2Children’s National Hospital, Washington, D.C., USA, 3Great Ormond Street Hospital, London, United Kingdom, 4Sungkyunkwan University, Seoul, South Korea, 5The George Washington University Medical Faculty Associates, Washington, D.C., USA

First Author:

Xiyu Feng  
UCL Great Ormond Street Institute of Child Health
London, United Kingdom

Co-Author(s):

Hua Xie  
Children’s National Hospital
Washington, D.C., USA
Rory Piper  
UCL Great Ormond Street Institute of Child Health|Great Ormond Street Hospital
London, United Kingdom|London, United Kingdom
Freya Prentice  
UCL Great Ormond Street Institute of Child Health
London, United Kingdom
Priyanka Illapani  
Children’s National Hospital
Washington, D.C., USA
Lauren Reppert  
Children’s National Hospital
Washington, D.C., USA
Seok-Jun Hong  
Sungkyunkwan University
Seoul, South Korea
Mohamad Koubeissi  
The George Washington University Medical Faculty Associates
Washington, D.C., USA
Torsten Baldeweg  
UCL Great Ormond Street Institute of Child Health
London, United Kingdom
Leigh Sepeta  
Children’s National Hospital
Washington, D.C., USA

Introduction:

The thalamus can participate in spread of epileptic activity (Wu et al., 2023) and is a target for therapeutic neuromodulation (Piper et al., 2022). Thalamic functional connectivity alterations exist in adult patients with temporal lobe epilepsy (TLE) and may affect seizure freedom after surgery (He et al., 2017). Here we used data-driven 'connectopic mapping' (Haak et al., 2018) to investigate the spatial organization (gradients) of thalamic cortical and subcortical connections in pediatric TLE. We aimed to uncover the functional gradients within the thalamus and investigate differences in thalamus to whole brain connectivity between children with TLE and healthy controls.

Methods:

64 children with TLE (5-18 years, left TLE n=51, right n=13) and 61 healthy controls (6-20 years) underwent language fMRI using a covert verb generation task at Great Ormond Street Hospital, London, UK. For children with right TLE, images were flipped so the seizure focus is the left hemisphere for all patients. fMRI data were preprocessed using fMRIprep software. 1) Connectopic mapping: We generated within-thalamus FC similarity matrices for each hemisphere and applied non-linear manifold learning to this matrix, yielding gradients for each side of the thalamus. We also created projection maps depicting changes in thalamic-to-whole brain connectivity along this gradient. 2) A SurfStat linear model (Worsley et al., 2009) assessed variations in projection maps related to disease status and duration, age and gender.

Results:

The primary thalamic gradient followed an anterior-to-posterior axis (Fig. 1) for both children with TLE and controls. In the cortex, there were no differences in the projection maps between patients and controls (FDR-corrected p>0.05). The anterior thalamus displayed greater connectivity with prefrontal and orbitofrontal cortices, as well as the basal ganglia. The mid-thalamic gradient zone was preferentially connected to the somatosensory cortex and anterior hippocampus. The posterior thalamus was more connected with the visual cortex and posterior hippocampus. However, in the subcortex, projection maps exhibited differences between patients and controls (FDR-corrected p<0.05). Compared to controls, patients showed stronger connectivity of the anterior thalamus to subcortical areas (Fig. 2), including the ventral striatum, subthalamic nuclei, substantia nigra as well as amygdala and hippocampus in both ipsi- and contralesional hemispheres. Duration of epilepsy was not correlated with projection map variations in patients (FDR-corrected p>0.05). Age and gender did not affect the cortical or subcortical projection maps for patients or controls (FDR-corrected p>0.05).
Supporting Image: Figure1.png
   ·Fig. 1
Supporting Image: Figure2.png
   ·Fig. 2
 

Conclusions:

A primary anterior-to-posterior functional gradient was observed within the thalami in both children with TLE and their healthy peers, revealing a gradual shift of thalamic connectivity across the entire brain. While there were no group differences in connectivity between thalamus and cortical regions, there was heightened connectivity in children with TLE between the anterior thalamus and the basal ganglia, with a predominance for the ipsilesional side. It has been suggested that within the loop involving the limbic system and thalamus, the basal ganglia play a significant seizure modulating role (Bröer, 2020; He et al., 2019). Further investigations are needed to explore the clinical potential of connectopic mapping.

Disorders of the Nervous System:

Neurodevelopmental/ Early Life (eg. ADHD, autism) 1

Modeling and Analysis Methods:

fMRI Connectivity and Network Modeling 2

Neuroanatomy, Physiology, Metabolism and Neurotransmission:

Subcortical Structures

Novel Imaging Acquisition Methods:

BOLD fMRI

Keywords:

Development
Epilepsy
FUNCTIONAL MRI
Pediatric Disorders
Sub-Cortical
Thalamus

1|2Indicates the priority used for review

Provide references using author date format

Bröer, S. (2020). Not Part of the Temporal Lobe, but Still of Importance? Substantia Nigra and Subthalamic Nucleus in Epilepsy. Frontiers in Systems Neuroscience, 14.
Haak, K. V. (2018). Connectopic mapping with resting-state fMRI. NeuroImage, 170, 83–94.
He, X. (2019). Disrupted basal ganglia–thalamocortical loops in focal to bilateral tonic-clonic seizures. Brain, 143(1), 175–190.
He, X. (2017). Presurgical thalamic “hubness” predicts surgical outcome in temporal lobe epilepsy. Neurology, 88(24), 2285–2293.
Piper, R. J. (2022). Towards network-guided neuromodulation for epilepsy. Brain: A Journal of Neurology, awac234.
Worsley, K. (2009). SurfStat: A Matlab toolbox for the statistical analysis of univariate and multivariate surface and volumetric data using linear mixed effects models and random field theory. NeuroImage, 47, S102.
Wu, T. Q. (2023). Multisite thalamic recordings to characterize seizure propagation in the human brain. Brain, 146(7), 2792–2802.