Relationship between gray matter volume and SEEG biomarkers of epileptogenicity
Poster No:
1704
Submission Type:
Abstract Submission
Authors:
Jacob Bunyamin1, Thanomporn Wittayacharoenpong1, Mohammad-Reza Nazem-Zadeh1, Parveen Sagar1, Noam Bosak1, Zhibin Chen1, Joshua Laing1, Matthew Gutman1, Martin Hunn1, Patrick Kwan1, Terence O'Brien2, Meng Law2, Ben Sinclair2, Andrew Neal1
Institutions:
1Monash University, Melbourne, VIC, 2Monash University, Melbourne, Victoria
First Author:
Co-Author(s):
Introduction:
The success of epilepsy surgery relies on accurately identifying the epileptogenic zone (EZ). There is a lack of imaging biomarkers that correlate with epileptogenicity. We aim to explore whether gray matter volume (GMV) is associated with stereo-electroencephalography (SEEG) epileptogenicity biomarkers.
Methods:
We included 3T non-contrast T1 scans from MRI-positive and MRI-negative drug-resistant focal epilepsy patients admitted for SEEG and age-sex-matched non-epileptic controls. MRI end: We segmented and modulated the T1 scans (pre-SEEG for epilepsy patients) to obtain a GMV map for each subject using CAT12. We then performed a non-parametric voxel-based comparison on SnPM between each patient's scan and the controls, resulting in pseudo-t-maps reflecting GMV differences between patients and controls, adjusted for age, sex, and total intracranial volumes. SEEG end: We classified SEEG contacts based on their involvement in (i) the EZ and (ii) interictal epileptogenicity biomarkers. The EZ was defined as contacts proposed for primary EZ radio-frequency thermocoagulation (RF-THC) after the SEEG procedure. Interictal epileptogenicity biomarkers were defined as contacts generating the top 10% of spikes, fast ripples, and cross-rates of HFO*Spikes for each patient. Statistical analysis: We calculated the mean GMV value within a 3 mm radius at each SEEG contact coordinate from the pseudo-t-maps. We then performed mixed-effect logistic regression and performance analysis to assess the ability of GMV to classify epileptogenic vs non-epileptogenic contacts within this dataset.
·Figure 1. Study Methodology
Results:
We included 50 patients (mean age 33.7±10.1 years, female 52.0%, MRI-negative 76.0%) and 68 controls (mean age 34.90±9.86 years, female 55.9%). Epileptogenic zone: From 4,878 contacts, increased GMV was associated with contacts involved in the EZ across the whole cohort (OR=1.19, 95% CI: 1.10-1.29, p<0.001), MRI-positive (OR=1.24, 95% CI: 1.09-1.40, p<0.001), and MRI-negative subgroup (OR=1.16, 95% CI: 1.04-1.28, p=0.005). GMV had a fair discriminative ability in the whole cohort (AUC 0.72), and MRI-negative subgroup (AUC 0.72), but was poor for the MRI-positive subgroup (AUC 0.69). At their optimal cut-off point, the sensitivity and specificity of GMV were 64% and 68% (whole cohort), 55% and 76% (MRI-positive), and 74% and 60% (MRI-negative subgroup). Interictal epileptogenicity biomarkers: From 5,346 contacts, there was no association between GMV and contacts displaying the top 10% of spikes (p=0.527), fast ripples (p=0.827), and cross-rates of HFO*Spikes (p=0.911) in the whole cohort. GMV had a poor ability to discriminate contacts within the top 10% of spikes (AUC 0.50), fast ripples (AUC 0.51), and cross-rates of HFO*Spikes (AUC 0.52) in the whole cohort.
Conclusions:
In our cohort, increased local GMV can discriminate epileptogenic vs non-epileptogenic SEEG contacts with a fair ability. Further assessment with unseen datasets is required to assess the generalizability of its performance.
Disorders of the Nervous System:
Neurodevelopmental/ Early Life (eg. ADHD, autism)
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Anatomy and Functional Systems 1
Physiology, Metabolism and Neurotransmission:
Neurophysiology of Imaging Signals 2
Keywords:
ELECTROPHYSIOLOGY
Epilepsy
MRI
STRUCTURAL MRI
Other - Stereo-EEG
1|2Indicates the priority used for review