To investigate the network connectivity patterns that occur in the dopamine D2 receptors (D2R) network in Parkinson’s disease (PD) patients with mild cognitive impairment (PD-MCI) and normal cognition (PD-NC) using graph theory analysis methodology. 

Cognitive decline in PD is a common sequela of the disorder, occurring in up to 60% of PD patients. Previous research found that there is a gradual loss of D2R in the cortex of PD patients as parkinsonism advances. Brain network analysis using graph theory methods is a new way of investigating the whole-brain communication patterns between different brain regions.

43 participants (healthy control=13, PD-NC=13, PD-MCI=17) underwent 90 minute PET scans with the radioligand [11C]FLB 457, which binds with high affinity to D2 receptors in the cortex. PET scans were acquired using a high-resolution PET/CT SiemensBiograph HiRez XVI PET scanner. An MRI scan for co-registration was also acquired using a 3.0 T GE Discovery MR750 MRI system. Imaging data was processed using the PMOD v4.0 PET analysis software package to calculate the partial volume error corrected binding potentials for each region of interest (ROI) in the AAL atlas. Graph theory analysis was performed using the Graph Theory Analysis Toolbox for MATLAB.

Two brain networks were examined: the dorsal dopamine network with nigrostrial dopamine projections comprised of 16 bilateral ROIs, and the mesolimbic dopamine network with ventral tegmental area projections comprised of 28 bilateral ROI. The dorsal dopamine brain network was found to be more efficient in PD-CU patients compared to PD-MCI patients, with reduced small-worldness and clustering (p<0.05). The mesolimbic dopamine network had reduced network efficiency in PD-CU compared to healthy controls, but no differences were found between the PD groups (p<0.05).

Dopaminergic dysregulation in PD results in less efficient brain network architecture in PD-MCI patients compared to PD-CU.