Objective: To disclose the neural mechanism of paroxysmal kinesigenic dyskinesia (PKD) based on the findings from high-density electroencephalogram (hd-EEG) and to explore a neural biomarker for PKD.

Background: PKD is believed related with the disturbance of neurologic electrophysiology due to its clinical features. However, the neuropathological mechanism of PKD is far from being revealed.

Method: 98 PKD patients and 96 healthy controls (HC) were enrolled. Resting-state hd-EEG were recorded and activity of 113 brain regions (ROIs) were evaluated by source localization. For each region, 6 canonical frequency band oscillations were defined their power spectral density (PSD) were calculated. To estimating functional connectivity (FC) between regions, a power envelope connectivity (PEC) analysis was employed at each oscillation.  Both PSD and PEC features were compared between HC and PKD, subtypes of PKD. Prediction models established based on lasso verified the robust of features from PSD and PEC for identify PKD and its subtypes.

Results: Both the difference in PSD and PEC features between HC and PKD is mainly from theta oscillation, mainly involving regions of cerebellum, subcortical, and prefrontal cortex. While the difference between features from gamma oscillation dominantly between patients in remission and symptomatic patients,  cerebellum, sensory-motor, occipital, and parietal cortex are most related. In term of PSD, motor-type PKD patients (only triggered by motor factors) have less significant features in theta oscillation than HC compared to mixed-type PKD patients. While in term of PEC, motor-type subgroup has more significant feature in gamma oscillation than HC compared to mixed type subgroup. Results of prediction models showed that features from PSD and PEC can accurately distinguish PKD from HC, and remission patients from symptomatic patients. Moreover, its effectiveness has been verified in external data.

Conclusion: The key abnormal neural activity in PKD patients are theta and gamma oscillations. The results indicate that theta oscillation seems to be related to the abnormal psychological state of PKD patients, while gamma oscillation mainly comes from abnormal motor symptoms. Almost all the brain regions are involved in PKD, and cerebellum, sensory-motor cortex, pre-frontal cortex as well as subcortical regions are the most related areas. Our results suggest that markers from EEG could achieve accurate diagnosis of PKD.

« Back to 2023 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/neuropathic-mechanism-of-paroxysmal-kinesigenic-dyskinesia-based-on-eeg/