Authors :
Presenting Author: Yasuo Nakai, MD, PhD – Wakayama Medical University
Hiroki Nishibayashi, MD, PhD – Department of Neurological Surgery – Wakayama Medical University; Tomohiro Donishi, MD, PhD – Department of System Neurophysiology – Wakayama Medical University; Sunao Oe, MD – Department of Neurological Surgery – Wakayama Medical University; Yoshiki Kaneoke, MD, PhD – Professor, Department of System Neurophysiology, Wakayama Medical University; Naoyuki Nakao, MD, PhD – Professor, Department of Neurological Surgery, Wakayama Medical University
Rationale: Resting-state functional magnetic resonance imaging (rs-fMRI)has contributed to the understanding of large-scale brain networks in the patients with epilepsy. Conventionally, cross-correlation analysis has been performed to determine the functional connectivity, which provides a metric of functional associations that reflects a temporal average over an entire scan. More recently, as dynamic functional connectivity analysis, co-activation pattern (CAP) analysis which reflects instantaneous brain configurations at single time points (Liu Xiao, et al. NeuroImage 2017). In this study, we aim to investigate whether the CAP analysis can detect the network alterations in temporal lobe epilepsy (TLE) and extra-temporal lobe epilepsy (extra-TLE).
Methods: Rs-fMRI (TR=3.0sec, 5 min×3) was performed in 13 patients with TLE (age; 35±15 years old [mean±SD], five male and eight females, Engel class I; 92%), 10 patients with extra-TLE (age; 33.4±5 years old [mean±SD], six males and six females, seizure onset zone; frontal seven, parietal four, one occipital 1, Engel class I; 40%), and age and sex matched healthy controls. All patients underwent resective surgery. Here we report the results obtained with K = 6 CAPs taking into consideration clustering stability and previous studies. Thus, the occurrence of each CAP was compared by using Mann-Whitney U tests with false discovery rate correction at significant level of 0.05.
Results: The occurrences of CAP1 and 2 which showed synchronization of the default mode network (DMN) and cerebellum were not different between HC and TLE (p > 0.05). The occurrence of CAP3 and 4 which showed complementary activation of the DMN and sensorimotor network (SMN) increased in TLE (CAP3; p = 0.016, CAP4; p = 0.024). The occurrence of CAP5 which showed activation in the central executive network (CEN) decreased in TLE (p = 0.029). On the other hand, the occurrence of CAP2 and 4, showing synchronization of cerebellum, the DMN and CEN, did not differ between HC and extra-TLE (p > 0.05). The occurrence of CAP3 and 4, which showed synchronization of DMN and the sensorimotor network (SMN), increased in extra-TLE. (CAP1; p < 0.001, CAP3; p = 0.009). The occurrence of CAP5, which showed activation in the salience network (SN), decreased in extra TLE (CAP5; p < 0.001).