Frequency-specific functional connectivity related to the rehabilitation task of stroke patients

Purposes: Stroke survivors suffering from deficits in motor control typically show persistent motor symptoms and limited functional abilities, which affect their functional independence in daily life. Active rehabilitation training is commonly applied for stroke patients to recover from motor dysfunction. The global connectivity reflects the synchronization of cardiac and respiratory activities in the cerebral regions. However, the understanding of the patterns of frequency-specific global connectivity (GC) and functional connectivity (FC) when performing active rehabilitation training is still far from comprehensive. This study was conducted to investigate the brain network patterns of stroke patients while performing a four-limb linkage rehabilitation training using the functional near-infrared spectroscopy (fNIRS) method.

Methods: Two groups of stroke patients (LH, left hemiplegia; RH, right hemiplegia) and one healthy group were recruited to participate in this study. The wavelet phase coherence (WPCO) method was used to calculate the frequency-specific GC and FC of the brain network in four frequency bands: I, 0.6-2 Hz; II, 0.145-0.6 Hz; III, 0.052-0.145 Hz; and IV, 0.021-0.052 Hz.

Results: Results showed that the healthy group exhibited lower WPCO in the four frequency bands during the task state than during the resting state (P < 0.05). Interestingly, the stroke groups showed increased WPCO in the frequency band II during the task state than during the resting state (P < 0.05). Moreover, significantly lower WPCO values in the frequency bands III (P < 0.05) were found during task state in the RH and LH groups compared with the healthy group. The RH group showed increased WPCO values in the frequency band II during the task state compared with the healthy group (P < 0.05). In addition, the RH group showed increased WPCO in the frequency bands I (P < 0.05) and II (P < 0.05) than the LH group. Notably, the rehabilitation task did not induce significant changes in stroke groups in the frequency band IV, which implied the neurogenic activity.

Conclusions: The reductions in FC suggested that the brain impairments caused a disturbed neurovascular coupling regulation in stroke patients. Results in frequency band IV suggested that the limb movement rehabilitation task intrinsically may not produce remarkable effect on the neurogenic activity of stroke patients. Significant differences in WPCO between the LH and RH groups suggested that the rehabilitation task should be specifically designed for individual rehabilitation. The frequency-specific FC methods based on WPCO would provide a potential approach to quantitatively assess the effect of rehabilitation tasks.