Brain functional connectivity and neuropsychiatric symptoms in patients with mild cognitive impairment, cerebrovascular disease and Parkinson disease
Neda Rashidi‐Ranjbar, Nathan W Churchill, Sandra E Black, Sanjeev Kumar, Maria Carmela Tartaglia, Morris Freedman, Anthony E Lang, Thomas Steeves, Richard H Swartz, Gustavo Saposnik, Paula M McLaughlin, Sean Symons, Stephen C. Strother, Bruce G. Pollock, Tarek K. Rajji, Miracle Ozzoude, Brian Tan, Stephen R. Arnott, Robert Bartha, Michael Borrie, Mario Masellis, Stephen H Pasternak, Andrew R Frank, Dallas Seitz, Zahinoor Ismail, David F. Tang‐Wai, Leanne Casaubon, Jennifer Mandzia, Christopher J. M. Scott, Dar Dowlatshahi, David A. Grimes, Connie Marras, David G. Munoz, Joel Ramirez, Courtney Berezuk, Melissa F. Holmes, Corinne E. Fischer, Tom A. SchweizerAbstract
Background
Mild Behavioral Impairment (MBI) is a condition characterized by neuropsychiatric symptoms (NPS) in older adults without dementia, serving as a precursor to various forms of dementia. This study explores the association between NPS and functional connectivity (FC) within the default mode network (DMN), executive control network (ECN), and salience network (SN) across three high‐risk cohorts: mild cognitive impairment (due to Alzheimer’s) (MCI, n = 79), cerebrovascular disease (CVD, n = 144), and Parkinson’s disease (PD, n = 132).
Method
A total of 367 participants were recruited from the Ontario Neurodegenerative Disease Research Initiative (ONDRI). The assessment of NPS utilized the Neuropsychiatric Inventory Questionnaire (NPI‐Q), with symptom severity rated on a scale from 1 to 3 (mild, moderate, severe). Resting‐state FC was analyzed for the DMN, ECN, and SN, using dual regression analysis to generate subject‐specific whole‐brain FC maps for each network. The association between FC maps and NPS scores was examined using FSL’s randomise with 5,000 permutations, while controlling for age, sex, and education. Results are presented following cluster False Discovery Rate (FDR) correction.
Result
The study revealed significant associations between NPS and FC specific to each cohort. In the MCI group, disturbed appetite and nighttime behaviors were correlated with increased FC of the dorsal DMN (p<0.05, R = 0.47, and p = 0.01, R = 0.47). The CVD group exhibited correlations between higher levels of anxiety and decreased FC of the dorsal DMN (p<0.05, R = ‐0.4), ventral DMN (p<0.05, R = ‐0.33), and bilateral ECN (p<0.05, R = ‐0.35 and R = ‐0.33). The PD group showed disturbed nighttime behavior associated with increased FC in ventral DMN (p<0.05, R = 0.35) and bilateral ECN (p<0.05, R = 0.43 and R = 0.37).
Conclusion
This research underscores disorder‐specific correlations between specific NPS domains and FC in MCI, CVD, and PD, emphasizing the unique neural underpinnings of symptomatology in each group. Furthermore, it is essential to note the inherent heterogeneity in all groups. Overall, the pathological substrates of neurodegenerative disorders likely play a pivotal role in shaping the neural correlates of MBI within each disorder. These findings provide valuable insights into targeted interventions and avenues for future research in neurodegenerative disorders.