Objective: To identify neuropsychological predictors of conversion to dementia (PDD) in our cohort of Parkinson’s disease (PD) patients.

Background: The heterogeneity of PD cognitive disorders reflects its complex neurobiological basis and it still remains unclear whether mild cognitive impairment (PD-MCI) and PDD are independent, partially overlapping syndromes or there is an obligatory continuum. The duality of cognitive deterioration has been suggested where non-dopaminergic cognitive deficits in visuospatial and memory domains could lead a way towards PDD.

Method: In this a prospective cohort study we followed 130 patients for 4 years. At each visit mild cognitive impairment (PD-MCI) and PDD were diagnosed according to Movement Disorders Society task force criteria (level 2 for PD-MCI). Depending on the change in cognitive status over time, a cognitive outcome was determined. Baseline predictors of conversion to dementia were investigated using a forward stepwise discriminant analysis.

Results: A total of 32 (29.62%) patients became demented during the follow-up period, 17 (15.74%) remained cognitively normal throughout of the study (PD-CNs), 20 (18.51%) converted from normal cognition to MCI (PD-MCIc), 18 (16.66%) were MCI during the study participation (PD-MCIs), while in 8 (7.40%) we recorded reversion from MCI to normal cognition. Cognitive deterioration was much more remarkable in patients who started the study as PD-MCI. A forward stepwise discriminant analysis was performed using a set of cognitive test variables at baseline as predictors of membership in two groups, MCI (PD-MCIc+PD-MCIs) (n=38) and PDDc (n=31). Predictors were Ray Auditory Verbal Test (recall score) (RAVLT), Stroop test (interference score), Category fluency test, Clock drawing test, Intra/Extra Dimensional Set Shift (total errors), Boston Naming Task (total score), and Digit Ordering Test (max span) (DOT). One discriminant function was calculated, with F(2, 67) = 16.79, p < .001, and Canonical R2 = .58. The structure matrix suggests that the best predictors for distinguishing between MCI and PPD patients are RAVLT and DOT (canonical loadings were .76 and .77). The classification procedure for the sample of 70 patients revealed that 74.6% were classified correctly.

Conclusion: Patients with predominantly memory and attention deficits are at most risk of PDD conversion.

« Back to MDS Virtual Congress 2021

MDS Abstracts - https://www.mdsabstracts.org/abstract/a-prospective-study-of-parkinsons-disease-dementia-defining-neuropsychological-features/