Kathy Dujardin1, Anja J H Moonen2, Hélène Behal3, Luc Defebvre4, Alain Duhamel5, Annelien A Duits6, Lucie Plomhause7, Celine Tard8, Albert F G Leentjens9. 1. Inserm, U1171, Troubles cognitifs dégénératifs et vasculaires, Université de Lille, Lille, France; Neurology and Movement Disorders Department, Lille University Medical Center, Lille, France. Electronic address: kathy.dujardin@univ-lille2.fr. 2. Department of Psychiatry, Maastricht University Medical Center, Maastricht, The Netherlands. Electronic address: anja.moonen@maastrichtuniversity.nl. 3. Department of Biostatistics, Lille University Medical Center, Lille, France. Electronic address: hbehalchr@gmail.com. 4. Inserm, U1171, Troubles cognitifs dégénératifs et vasculaires, Université de Lille, Lille, France; Neurology and Movement Disorders Department, Lille University Medical Center, Lille, France. Electronic address: luc.defebvre@chru-lille.fr. 5. Department of Biostatistics, Lille University Medical Center, Lille, France. Electronic address: alain.duhamel@univ-lille2.fr. 6. Department of Psychiatry, Maastricht University Medical Center, Maastricht, The Netherlands. Electronic address: annelien.duits@maastrichtuniversity.nl. 7. Inserm, U1171, Troubles cognitifs dégénératifs et vasculaires, Université de Lille, Lille, France; Neurology and Movement Disorders Department, Lille University Medical Center, Lille, France. Electronic address: lucie.plomhause@chru-lille.fr. 8. Inserm, U1171, Troubles cognitifs dégénératifs et vasculaires, Université de Lille, Lille, France. Electronic address: celine.tard@chru-lille.fr. 9. Department of Psychiatry, Maastricht University Medical Center, Maastricht, The Netherlands. Electronic address: a.leentjens@maastrichtuniversity.nl.
Abstract
INTRODUCTION: Clinical presentation and progression of cognitive disorders in Parkinson's disease (PD) is heterogeneous. Our objective was to confirm prospectively a previous exploratory cluster analysis based on retrospective data that identified five cognitive phenotypes in PD. METHODS: A model-based confirmatory cluster analysis was conducted on the results of neuropsychological tests administered in 156 PD patients from two European movement disorder centers (Lille, n = 81; Maastricht, n = 75). The number of clusters was determined on the basis of statistical criteria as well as clinical plausibility. A factorial discriminant analysis assessed the quality of the clusters' separation. RESULTS: A five-cluster model was statistically superior and clinically the most relevant. These clusters can be described as follows: 1) cognitively intact patients with high level of performance in all cognitive domains (25.64%), 2) cognitively intact patients slightly slower than those in cluster 1 (26.92%), 3) patients with deficits in executive functions (37.18%), 4) patients with severe deficits in all cognitive domains, particularly executive functions (3.20%), 5) patients with severe deficits in all cognitive domains, particularly working memory and recall in verbal episodic memory (7.05%). The groups differed in terms of age, apathy and frequency of hallucinations that were all higher in the clusters with cognitive deficits, and the duration of formal education was lower in those groups. CONCLUSION: We confirm our previous exploratory analysis. Cognitive disorders in PD patients are heterogeneous and can be separated in five clusters ranging from patients with performance in the normal range to patients with severe disorders in all cognitive domains.
INTRODUCTION: Clinical presentation and progression of cognitive disorders in Parkinson's disease (PD) is heterogeneous. Our objective was to confirm prospectively a previous exploratory cluster analysis based on retrospective data that identified five cognitive phenotypes in PD. METHODS: A model-based confirmatory cluster analysis was conducted on the results of neuropsychological tests administered in 156 PDpatients from two European movement disorder centers (Lille, n = 81; Maastricht, n = 75). The number of clusters was determined on the basis of statistical criteria as well as clinical plausibility. A factorial discriminant analysis assessed the quality of the clusters' separation. RESULTS: A five-cluster model was statistically superior and clinically the most relevant. These clusters can be described as follows: 1) cognitively intact patients with high level of performance in all cognitive domains (25.64%), 2) cognitively intact patients slightly slower than those in cluster 1 (26.92%), 3) patients with deficits in executive functions (37.18%), 4) patients with severe deficits in all cognitive domains, particularly executive functions (3.20%), 5) patients with severe deficits in all cognitive domains, particularly working memory and recall in verbal episodic memory (7.05%). The groups differed in terms of age, apathy and frequency of hallucinations that were all higher in the clusters with cognitive deficits, and the duration of formal education was lower in those groups. CONCLUSION: We confirm our previous exploratory analysis. Cognitive disorders in PDpatients are heterogeneous and can be separated in five clusters ranging from patients with performance in the normal range to patients with severe disorders in all cognitive domains.
Authors: Nadeeka N W Dissanayaka; Rachael A Lawson; Alison J Yarnall; Gordon W Duncan; David P Breen; Tien K Khoo; Roger A Barker; David J Burn Journal: Parkinsonism Relat Disord Date: 2017-01-04 Impact factor: 4.891
Authors: M Hassan; L Chaton; P Benquet; A Delval; C Leroy; L Plomhause; A J H Moonen; A A Duits; A F G Leentjens; V van Kranen-Mastenbroek; L Defebvre; P Derambure; F Wendling; K Dujardin Journal: Neuroimage Clin Date: 2017-03-06 Impact factor: 4.881