Introduction: Deep brain stimulation(DBS) is a treatment of Parkinson’s disease(PD) treating both tremor-dominant(hyperkinesia) and akinesia-rigidity(AR) dominant phenotypes 1. Despite variability in clinical manifestations, whether DBS modulates network activity differently in these phenotypes remains unknown. A conditional change in DBS labeling allows MRI to be performed with DBS-ON. Here we analyze whether network activity varies in PD patients depending on phenotypes.

Methods: The current study involved 17 patients diagnosed with PD. Subjects, independent from DBS target, were classified into tremor/AR cohorts based on dominant PD symptoms. Subjects with clinically optimized DBS programming underwent fMRI scans with their neurostimulator cycling between ON/OFF. To ensure that fMRI and programmer were synchronized, an electronics box was integrated into the workflow2. Voxel-wise General Linear Model(GLM) using DBS-OFF data was used to determine altered regions during DBS-ON. For group analysis, previous outputs of individual subjects were used to determine voxel-wise statistics. A multiple regression connectivity analysis was performed on data from individuals to determine connections of the motor circuitry.

Results: Individual fMRI analysis revealed that 5/6 tremor subjects had anterior cerebellum and thalamic activation, 4 had motor cortex(M1) deactivation. These findings were in contrast to the AR-cohort where 5/11 subjects had M1 activation, rather than deactivation(p<0.05) and an equal amount had cerebellar activation/deactivation. Thalamic activation was similar between phenotypes. Group analysis of the tremor-cohort confirmed significant anterior cerebellar and thalamic activation, and M1 deactivation. AR group analysis confirmed M1 and thalamic activation. Connectivity studies revealed motor cortex to cerebellum connections in the tremor group, but not in the AR-cohort.

Conclusions: These data suggest that DBS modulates network activity differently in PD based on patient phenotype. An improved understanding of tremor/akinesia-rigidity differences and their functional roles in PD may allow us to exploit these connections to design more efficient personalized DBS therapy, hopefully, with better clinical results and less side effects.

Patient Care: These data suggest that DBS modulates network activity differently in PD based on patient phenotype. An improved understanding of these differences may allow us to exploit these connections to optimize DBS programming.

Learning Objectives: To examine how network activity is affected by DBS in different PD phenotypes.

References: Jankovic J, McDermott M, Carter J, et al. Variable expression of Parkinson's disease: a base-line analysis of the DATATOP cohort. The Parkinson Study Group. Neurology. 1990;40(10):1529-1534. Fiveland E, Madhavan R, Prusik J, et al. EKG-based detection of deep brain stimulation in fMRI studies. Magn Reson Med. 2017.