Abstracts

Rationale: Psychogenic non-epileptic seizures (PNES), different from epilepsy, are considered to be initiated by psychogenic factors rather than electrophysiological abnormalities. Prior EEG and fMRI studies had revealed several functional networks abnormality in PNES. Nevertheless, the underlying mechanism of PNES remains unclear. The current study aimed to explore the network differences between PNES and healthy controls, as well as between PNES and idiopathic generalized epilepsy patients. Methods: Consecutive PNES patients admitted to West China Hospital between January 2010 and March 2011 were included. Patients were confirmed as PNES by two independent experienced neurologists. Patients diagnosed with idiopathic generalized epilepsy (IGE) during the same period were also included. Age-, sex-matched healthy controls (HC) were also recruited. All subjects underwent a resting-state functional MRI scan using echo plan imaging sequence following a anatomical T1 image acquisition at a 3T MRI. Group independent component analysis was applied to the preprocessed fMRI images to identify the intrinsic resting networks. The spatial maps of the selected common networks were compared between the PNES group and IGE group, between the PNES group and HC group. For those networks, the functional network connectivity was evaluated among the three groups. Results: A total of 21 PNES patients, 23 IGE patients and 24 macthed healthy controls were included. Thirty-six networks were identified. Ten common resting networks among the three groups were selected based on the neuroanatomical correspondence with the consistently identified resting-state networks described in the literatures. Compared to the IGE group and the HC group, PNES showed decreased functional integration of bilateral prefrontal cortex in the prefrontal network, left orbito-frontal cortex in the left fronto-parietal network, biltateral mesial prefontal cortex in the default mode network, anterior cingulate cortex in the self-reference network, cerebellum in the cerebellum network. PNES showed increased functional integration of bilateral caudate and insula in the salience network in contrast to the IGE patients and healthy controls. The between-group differences were larger between the PNES and the HC groups than that between the PNES and IGE groups. Compared to the HC, PNES also demonstrated decreased functional integration of inferior parietal cortex in the dorsal attention network and right fronto-parietal networks. (Fig.1) No differences within the sensori-motor network and visual network between the PNES and HC group, and between the PNES and IGE group. The functional network connectivity between these networks were statistically different among the three group, especially the connectivity with prefrontal network and fronto-parietal networks. Conclusions: PNES patients displayed abnormalities in several resting-state networks and network correlations. The prefrontal network and fronto-parietal network may play an important role in PNES.