Authors :
Presenting Author: Shilpi Modi, PhD – Thomas Jefferson University
Ankeeta A., PhD – Post-Doc Fellow, Department of Neurology, Thomas Jefferson University; Walter Hind, PhD – Post-Doc Fellow, Department of Neurology, Thomas Jefferson University; Michael Sperling, MD – Professor, Department of Neurology, Thomas Jefferson University; Joseph Tracy, PhD – Professor, Department of Neurology, Thomas Jefferson University
Rationale:
Temporal lobe epilepsy (TLE) is the most common focal epilepsy syndrome in adults, with focal to bilateral tonic-clonic seizures (FBTCS+) its most severe form. Static resting state (RS) connectivity studies have indicated that thalamo-cortical connectivity is an important feature of FBTCS+. In this project, we focus on dynamic functional connectivity (dFC) and pursue a comprehensive characterization of RS dFC in FBTCS+ examining four distinct set of subcortico-cortico communication/propagation pathways.
Methods:
Fifty-five left TLE patients and 56 age matched healthy controls underwent a 8 min resting state fMRI scan. Patients were grouped based on FBTCS history: (i) no history of FBTCS (FBTCS-, n=15); (ii) recurrent FBTCS+ within the year prior to scanning (current, n=27); and(iii) FBTCS+ beyond a year prior to scanning (remote, n=13). After pre-processing using fMRIPrep 20.1.0, 7 intrinsic cortical networks (Visual, Somatomotor, Dorsal Attention [DAN], Ventral Attention [VAN], Limbic, Control, Default Mode [DMN]; Schaefer atlas), and four subcortical systems (mesial-temporal/striatum/thalamus-Melbourne atlas and 7-network cerebellar parcellations- Buckner atlas) were extracted. Adjacency matrices were built (sliding-window approach, width: 40s, 50% overlap), community detection conducted, module allegiance calculated and dFC measures- recruitment, integration and flexibility computed. To determine the atypicality in the dFC matrices, the deviation scores were calculated,
Zpat = (dFCpat - µcon)/scon
where µcon and scon were the mean and standard deviation of the same dFC matrix from the healthy controls, for each patient. Multivariate ANOVAs on the three dFC measures were run (within-subject factor: left/right hemisphere; dFC measures (run separate for each of the subcortico-cortico combinations; between-subject factor: group - FBTCS-, FBTCS+ remote, current). Results:
In thal. /cort. analysis, FBTCS+ curr. showed increased recruitment in the DAN and the anterior thalamus (versus FBTCS- and FBTCS+ rem.). In the striatum/cortical analysis, FBTCS+ curr. demonstrated greater flexibility in the lt. compared to the rt. hemisphere, and had greater recruitment in the limbic network (versusFBTCS+ rem.). FBTCS+ groups (rem. p=0.008; curr., p=0.021) showed a greater recruitment in the DAN (versus FBTCS- ).
There was an increased integration between caudate and lt.-rt. VAN; cereb-5 and lt.-rt. VAN in FBTCS+ curr. (versus FBTCS-). As compared to FBTCS+ rem., FBTCS+ curr. showed greater integration between caudate/lt. control and cereb-5/lt. DMN systems. FBTCS+ rem. also showed an increased cereb-5/rt. VAN integration.
Conclusions:
Our data showed dFC abnormalities in FBTCS+ curr. (more than FBTCS+rem.) across several subcortico-cortio systems thereby extending the FBTCS literature beyond the thalamus.. These data indicate a complex set of subcortio-cortico propagation pathways may support the secondary generalization in this group.
Funding:
Joseph Tracy (PI), NIH/NINDS, R01 NS112816-01