EP 66. Cortico-subthalamic neural interactions: Relation to cognitive task performance and evidence for a novel interaction mode across multiple time scales in patients with Parkinson's disease. Issue 9 (September 2016)

• Record Type: Journal Article
• Authors: Hohlefeld, F.; Ewald, A.; Ehlen, F.; Tiedt, H.; Horn, A.; Kühn, A.; Curio, G.; Klostermann, F.; Nikulin, V.
• Is Part Of: Clinical neurophysiology. Volume 127:Issue 9(2016:Sep.)
• Page Start: e203
• Publication Date: 2016-09
• Subjects: Neurophysiology -- Periodicals; Electroencephalography -- Periodicals; Electromyography -- Periodicals; Neurology -- Periodicals; 612.8
• DOI: 10.1016/j.clinph.2016.05.253
• Languages: English
• ISSNs: 1388-2457

Abstract

Abstract : Introduction: Neural interactions between the cortex and the subthalamic nucleus (STN) are frequently studied in patients with Parkinson's disease and were shown to relate to movement performance and the severity of motor symptoms. However, it is unknown whether cortico-subthalamic synchronization might also relate to non-motor, rather cognitive aspects of task performance, e.g., in language context. Furthermore, we do not know whether neural interactions between cortex and STN might be characterized by additional modes of communication across different time scales and frequency bands. Objectives: Addressing these open questions we investigated (i) the relation between cortex-STN synchronization and lexical decisions in Parkinsonian patients. In contrast to previous studies we applied methods that optimize the extraction of neural synchronization from multi-channel recordings, while simultaneously excluding artifacts of volume conduction. Furthermore, we addressed (ii) for the first time the presence of cortico-subthalamic neural interactions across multiple time–frequency scales, by investigating the relation between long-range temporal correlations (across tens of seconds) in cortically dominant alpha oscillations and coherency (milliseconds) in subthalamic beta oscillations. Materials and methods: Eight patients with Parkinson's disease, operated for deep brain stimulation, performed a lexical decision task on dopaminergic medication. Simultaneous recordings were obtained from surface electroencephalography (EEG) and deep local field potentials (LFP) from the bilateral STN. In the continuous data cortico-subthalamic coherency was assessed by the maximized imaginary part of coherency (MIC), which simultaneously accounts for multi-channel recordings and is insensitive to volume conduction artifacts. Furthermore, cortical long-range temporal correlations (up to 20 s) were calculated in the amplitude envelope of EEG oscillations, and the detectability of the imaginary part of coherency (iCOHd) was calculated in STN-LFP. Results: Cortico-subthalamic coherency in 14–35 Hz oscillations estimated by MIC was positively correlated with the accuracy in lexical decisions across patients. No significant correlation was present in 7–13 Hz oscillations, and MIC was not correlated with the reaction times in any of the frequency bands. Furthermore, we showed a significant positive correlation across patients between cortical long-range correlations (8–13 Hz, whole scalp) and subcortical coherency (iCOHd) selectively in 10–20 Hz oscillations in the left STN. Conclusions: The present results suggest that the complementary usage of multivariate methods, such as MIC, might be effective in coherency studies for extracting neural synchronization from multi-channel recordings. Furthermore, cortico-subthalamic synchronization might be relevant also for cognitive aspects of task performance, here related to lexical decisions. Finally, our results provide evidence for novel, complex neural interaction modes across different time scales, reflected in the relation between neural dynamics in the most dominant rhythms in cortex and basal ganglia. Therefore, the present results contribute to further understanding of neural pathology in Parkinson's disease on macroscopic network levels.