Striatal and thalamic automatic segmentation, morphology, and clinical correlates in Parkinsonism.

161 

Abstract Submission 

May Erlinger¹, Jeffrey Looi², Rosa Molina-Ruiz³, Eva Lopes Valdes³

¹Australian National University, Canberra, Australia, ²Research Centre for the Neurosciences of Ageing, Academic Unit of Psychiatry and Addiction Medicine, Canberra, ACT, ³Hospital Clinico San Carlos, Madrid, Spain

May Erlinger  
Australian National University
Canberra, Australia

Jeffrey Looi  
Research Centre for the Neurosciences of Ageing, Academic Unit of Psychiatry and Addiction Medicine
Canberra, ACT

Rosa Molina-Ruiz, MD, PhD  
Hospital Clinico San Carlos
Madrid, Spain

Eva Lopes Valdes, MD PhD  
Hospital Clinico San Carlos
Madrid, Spain

Parkinson's disease (PD), multisystem atrophy (MSA), and progressive supranuclear palsy (PSP) are each considered progressively degenerative neurological movement disorders, characterised by variable, but similar, presentations of movement impairment and accompanying cognitive and behavioural neuropsychiatric symptoms such as sleep disturbances, depression, and autonomic dysfunction (1). MSA-parkinsonian type is most significantly clinically differentiated from Parkinson's disease through its widespread impacts on autonomic functions (2). PSP can be clinically differentiated from PD through its supranuclear gaze palsy and postural instability (3) which can also occasionally present in MSA. However, these diseases do not always classically present with their characteristic symptom, and instead have significant overlap in their clinical features of bradykinesia, rigidity, tremors, rendering them difficult to discriminate and diagnose in-vivo. The thalamus and striatum have been previously identified as having potential as biomarkers for neurodegenerative disorders, especially as related to disease onset (4,5), progression (6,7) and severity (8) and therefore may serve as potent and reliable disease-specific biomarkers for differentiating atypical parkinsonism. This poster investigates striatal and thalamic volume and morphology as distinguishing biomarkers, and their relationship to neuropsychiatric symptoms.

Automatic segmentation to calculate volume and shape analysis of the caudate nucleus, putamen, and thalamus were performed in 18 PD patients, 12 MSA, 15 PSP, and 21 healthy controls. T1-weighted MRI data were acquired for patients and healthy controls by a 3.0T Phillips MR scanner, with pre-processing of the images conducted through FSL, and automatic bilateral subcortical structure segmentations of the relevant regions of interest (ROI) (bilateral caudate nucleus, putamen, and thalamus) created using FSL-FIRST. The segmentations created were included in a shape analysis using the SPHARM-PDM module in 3D Slicer (9). Multivariate analysis of covariance was conducted to determine significant differences between PD, MSA, PSP, and controls in the ROI volumes. The covariates included in analysis were age, sex, and total ICV to account for head size. For shape analyses, we used the Covariate Significance Testing module of SlicerSALT, with a family-wise error correction for multiple comparisons, with age and sex included as covariates, to compare PD, MSA, PSP and controls. Volume and shape of ROIs were then correlated clinical measures of parkinsonism and neuropsychiatric function.

The volumetric analyses determined there was a significant difference in all the ROIs between groups, with the bilateral thalami (L: p=0.017, R: p=0.011) and right putamen (p=0.032) of PSP patients found to be significantly smaller than controls. The left caudate (p=0.025) and left putamen (p=0.025) significantly correlated with the Neuropsychiatric Inventory total score. Bilateral thalamus, caudate, and left putamen had significantly different morphology between groups. The antero-ventral medial portion of the bilateral caudate head varied significantly (L: p=0.004, R: p=0.012) between groups, driven by differences between PSP and healthy controls (p=0.004). The ventro-posterior aspect of the left putamen differed significantly (p=0.028), and was driven by PSP (p<0.001) and PD (p<0.001) compared to controls. The left antero-ventral thalamus (p=0.004), and the right dorsolateral thalamus (p=0.012) were significantly different between groups, which was bilaterally driven by differences in PSP patients compared to controls (L: p=0.002, R: p<0.001).

This study demonstrated that PSP striatal and thalamic volumes and shapes are significantly different when compared with controls. Parkinsonian disorders could not be differentiated on volumetry or morphology, however there are trends for volumetric and morphological changes associated with PD, MSA, and PSP.

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) ¹

Higher Cognitive Functions Other

Segmentation and Parcellation

Subcortical Structures ²

Morphometrics

Movement Disorder

MRI

STRUCTURAL MRI

Structures

Sub-Cortical

Thalamus

Other - Striatal