Impaired topographic organization in Parkinson's disease with mild cognitive impairment

Background Mild cognitive impairment (MCI) is common in Parkinson's disease (PD), and graph theory approaches can be performed to investigate the topographic organization in newly diagnosed drug-naïve PD patients with MCI. Method We recruited PD patients with MCI (PD-MCI), PD patients with cognitive unimpaired (PD-CU), and age- and sex-matched healthy controls (HCs). Resting-state functional MRI (fMRI) whole-brain connectivity was examined, and topographic properties were measured with age, sex and education as covariates. Correlation analyses were performed between topographic features and cognitive scores. Results Newly diagnosed drug-naïve PD patients and HCs presented small-world properties, and PD patients had increasing random organizations of brain networks, especially in PD patients with MCI. We also found a descending trend (HC > PD-CU > PD-MCI) in the clustering coefficient (Cp), characteristic path length (Lp) and local efficiency (Eloc), and a rising trend (HC < PD-CU < PD-MCI) in the global efficiency (Eglob). Only PD patients with MCI showed decreased nodal centralities in nodes of the sensorimotor network (SMN), default mode network (DMN), and the ventral anterior prefrontal cortex (vent aPFC), and increased nodal centralities in nodes of the cingulo-opercular network (CON), occipital network, and the ventral lateral prefrontal cortex (vlPFC). The increased nodal centralities in the parietal node of CON negatively correlated with cognitive scores in all PD patients. Conclusion Our results suggested that newly diagnosed drug-naïve PD patients had increasing random organizations of brain networks, especially in PD-MCI patients. Nodal changes were mainly observed in PD-MCI patients.