Neural Correlates of Effects of Internally versus Externally Guided Partnered Rehabilitative Tango for People with Parkinson’s Disease

Abstract

ABSTRACTRestoring function to damaged neural pathways, or promoting compensatory strategies to overcome dysfunctional neural pathways have been topics of inquiry within motor rehabilitation. This study considers these topics in Parkinson’s disease (PD), where disruption within the striatal-thalamic-cortical (STC) circuits can cause impairment in internally guided (IG) movements. A related, but separate externally guided (EG) movement network, recruits the intact cerebellar-thalamic-cortical (CTC) loop to facilitate movement in response to sensory cues, and is effective in remediating motor function. Partnered dance with leading and following roles may be used as proxies for training IG and EG strategies respectively, in PD and can test which strategy is more effective in remediating effects of PD. Leaders determine variables associated with IG and communicate step amplitude, timing, and direction to the follower. Followers use an EG strategy to sense and interpret directional pressure cues from the leader and then enact an appropriate movement response. This study examines how IG and EG training strategy affects STC and CTC circuits and their behavioral outcomes to examine whether compensatory or direct entrainment of neural pathways is more effective in PD. Fifty-eight participants were recruited with mild-moderate PD (stages 1-3) and randomly assigned to EG, IG or non-dance education control group and assessed before and after 12 weeks of biweekly interventions. Participants were assessed with standard cognitive and motor behavioral measures and lay in a Magnetic Resonance Imaging (MRI) scanner while they tapped their foot under two conditions: internal (tapping a learned rhythm: IT) and external (participant taps in response to an assistant’s felt tap on the participant’s hand: ET) guidance. The foot-tapping data collected with an accelerometer were evaluated by analyzing the frequency spectrum to calculate amplitude and timing of the foot taps. The functional (fMRI) data were pre-processed (AFNI), registered (MNI), and analyzed for changes in activation using a general linear model in SAS and AFNI. Postintervention, both the EG and the IG groups showed clinically significant changes on disease severity, but the EG group showed improvements on cognitive, motor, and mood variables. The EG group also outperformed the IG group in the in-scanner task performance measured by the foot accelerometer. Imaging data revealed a significant increase in the EG group in the primary motor cortex lower limb region and the parts of the cerebellar circuits, particularly right Cerebellar Lobule VIIIa. The control group showed an increase in activity in the putamen compared to the IG and EG groups that could be due to a different compensatory pathway. All findings were corroborated using a region of interest approach examining the same pathways according to an atlas. Our results indicate that the most effective strategy for the PD participants involved external cues that increased activity in the compensatory CTC pathway, the primary motor region, and significant improvements on almost all behavioral measurements.