Objective: To identify cognitive phenotypes within Parkinson’s disease (PD) and compare them to the current Movement Disorder Society (MDS) diagnostic standards for Mild Cognitive Impairment in PD (PD-MCI).

Background: The concept of MCI in PD research has shown potential for identifying at-risk dementia patients. However, the Dual Syndrome Hypothesis suggests that there are clinically heterogenous cognitive phenotypes within PD-MCI. The present study therefore aimed to identify these subtypes using a data-driven cluster approach and define their clinical features.

Method: Eighty-five non-demented PD out-patients underwent physical examination, psychological evaluation, and cognitive assessment at level II of the MDS diagnostic criteria for PD-MCI. An exploratory K-means cluster analysis was performed on 10 variables covering each of the five conventional cognitive domains, reclassified into either the executive or memory domain. The final cluster structure was determined on the basis of its quantitative strength and theoretical relevance, and clusters were compared on demographic  and clinical measures.

Results: The resulting cluster structure revealed a progressive gradient of four distinct cognitive phenotypes: Cognitively-Intact; Executive-Impaired; Memory-Impaired; and Globally-Impaired. Demographic profiling revealed significant differences in the age, gender split, and estimated premorbid IQ of the phenotypes, as well as the severity of cognitive (MoCA, PD-CRS) and motor (UP-DRS III) symptoms. However, differences between the clusters on measures of levodopa intake, disease duration, depression, apathy, and anxiety were non-significant.

Conclusion: These results validate the existence of distinct cognitive phenotypes within PD-MCI and establish the unique clinical characteristics at each stage of impairment. Future research into the cognitive progression and clinical trajectory of each phenotype is recommended moving forward.

« Back to 2019 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/identifying-subtypes-of-mild-cognitive-impairment-in-parkinsons-disease-using-cluster-analysis/