Functional connectivity fingerprints of hallucinations in Parkinson’s Disease Dementia
Poster No:
294
Submission Type:
Abstract Submission
Authors:
Sara Stampacchia1, Fosco Bernasconi1, Alice Albrecht1, Konstantin Toussas2, John Paul Taylor3, Alan Thomas4, Enrico Amico5, Olaf Blanke6
Institutions:
1École polytechnique fédérale de Lausanne (EPFL), Geneve, Switzerland, 2University of Geneva, Geneve, Switzerland, 3Newcastle University, Newcastle, United Kingdom, 4Newcastle University, Newcastle, United Kingdom, 5Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland, 6EPFL, Geneva, Switzerland
First Author:
Co-Author(s):
Introduction:
Hallucinations in Parkinson's Disease are linked to the development of dementia in Parkinson's Disease (PDD)1. Minor hallucinations (MH) emerge early in the disease, and previous work has shown that early-stages PD patients with and without MH are distinguishable in terms of functional connectivity (FC) 2,3. Recently, we switched the focus from group differences to inter-subject variability, and found distinguishable FC fingerprints in PD patients with and without MH. 4. The current study extends the investigation to the latest disease stages (PDD) to explore if identifiability based on FC-fingerprints persists, and whether different patterns drive inter-individual variability in patients with and without later and more structured forms of hallucinations, i.e., visual hallucinations (VH).
Methods:
Forty-four patients with PDD were included (PD-VH, 27; PD-nH, N=17). The presence of VH was evaluated clinically by expert neurologists. As expected, PDD-VH had higher severity and distress of hallucinations and delusions (Neuropsychiatric Inventory – NPI: p > .010). There were no differences across groups for sex assigned at birth (p=.780) and age (p=.144). In line with what has been previously described in the literature, patients with PDD-VH had lower level of cognitive functioning (Cambridge Cognition Examination – CAMCOG: p = .016) and higher level of motor impairment (Unified Parkinson's Disease Rating Scale – UPDRS: p<.001) relative to PDD-nH.
We estimated individual FC matrices using Pearson's correlation between the averaged BOLD signals of 278 cortical and subcortical nodes5. We estimated individual variability in FC using the following metrics introduced and estimated in healthy subjects and neurological patients 6,7. First, we calculated within (ISelf) and between (IOthers) subjects test-retest FC similarity across the first 180s and second 180s of the same scan. A second metric, IDiff, provided a group-level estimate of distance between ISelf and IOthers. Then, we explored the spatial specificity of differences across patients using edgewise intra-class correlation (ICC). ICC quantifies the within-subject similarity between test and retest for each edge (FC between 2 regions), such that the higher the ICC of an edge the more the two regions show stable patterns of functional connectivity across test and retest within-subjects, as well as variable patterns of functional connectivity between-subjects of their group.
We estimated individual FC matrices using Pearson's correlation between the averaged BOLD signals of 278 cortical and subcortical nodes5. We estimated individual variability in FC using the following metrics introduced and estimated in healthy subjects and neurological patients 6,7. First, we calculated within (ISelf) and between (IOthers) subjects test-retest FC similarity across the first 180s and second 180s of the same scan. A second metric, IDiff, provided a group-level estimate of distance between ISelf and IOthers. Then, we explored the spatial specificity of differences across patients using edgewise intra-class correlation (ICC). ICC quantifies the within-subject similarity between test and retest for each edge (FC between 2 regions), such that the higher the ICC of an edge the more the two regions show stable patterns of functional connectivity across test and retest within-subjects, as well as variable patterns of functional connectivity between-subjects of their group.
Results:
First, data showed that individual FC profiles were highly distinguishable within each group, i.e., among patients with the same diagnosis, but also specific clinical symptoms, i.e., visual hallucinations. At the whole-brain level, we found that in each group, patients were always distinguishable from other patients (ISelf>IOthers in all cases). IDiff was comparably high in both groups, there were no significant differences in ISelf across groups, and IDiff was different from null distribution at p<.001 in all groups (Fig.1A and 1B). Second, we found that FC patterns that identified patients differed between PDD-VH and PDD-nH (Fig.2A).
Conclusions:
This study extends previous work showing group-differences in FC 2,3 and in FC-fingerprints between PD patients with and without MH, by finding that the functional connections remain individual-specific also in the most advanced stages of the disease and distinguishable between patients with later forms of hallucinations7. This work confirms that the individual variability – well-known in clinical practice – is reflected by differences in FC even in the more advanced stages of the disease and during dementia (PDD). These findings help characterizing the neural basis of late hallucinations in PD (i.e., VH) and may pave the way for a personalised understanding of the altered brain mechanisms in hallucinations as well as for their early detection.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 1
Lifespan Development:
Aging
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling 2
Novel Imaging Acquisition Methods:
BOLD fMRI
Keywords:
Computational Neuroscience
Degenerative Disease
FUNCTIONAL MRI
Neurological
Other - Functional connectivity fingerprints
1|2Indicates the priority used for review
Provide references using author date format
2. Bejr-kasem, H. et al. Disruption of the default mode network and its intrinsic functional connectivity underlies minor hallucinations in Parkinson’s disease. Movement Disorders 34, 78–86 (2019).
3. Bernasconi, F. et al. Robot-induced hallucinations in Parkinson’s disease depend on altered sensorimotor processing in fronto-temporal network. Sci. Transl. Med. 13, eabc8362 (2021).
4. Stampacchia, S. et al. Functional connectivity fingerprints in Parkinson’s disease patients with hallucinations. in (in preparation).
5. Shen, X., Tokoglu, F., Papademetris, X. & Constable, R. T. Groupwise whole-brain parcellation from resting-state fMRI data for network node identification. NeuroImage 82, 403–415 (2013).
6. Amico, E. & Goñi, J. The quest for identifiability in human functional connectomes. Sci Rep 8, 8254 (2018).
7. Stampacchia, S. et al. Fingerprinting of brain disease: Connectome identifiability in cognitive decline and neurodegeneration. 2022.02.04.479112 Preprint at https://doi.org/10.1101/2022.02.04.479112 (2022).
8. McGraw, K. O. & Wong, S. P. Forming Inferences About Some Intraclass Correlation Coefficients. Psychological Methods 17 (1996) doi:1082-989X/96/S3.00.
9. Shrout, P. E. & Fleiss, J. L. Intraclass Correlations: Uses in Assessing Rater Reliability. Psychological Bulletin 86, 420–428 (1979).