Magnetic resonance imaging of saccadic eye movements in Parkinson’s disease

Abstract

Parkinson’s disease (PD) is a progressive neurodegenerative disorder with cardinal signs of bradykinesia, tremor, rigidity and postural instability. Current PD treatment only alleviate symptoms of PD, however, there is an intense research focus on identifying disease-modifying therapies. No biochemical or laboratory marker exists for PD progression. An objective and repeatable biomarker would allow assessment of the efficacy of novel disease-modifying therapies, as well as allowing precisely tailored treatment.

Saccadic eye movements are an accurately measured and easily repeatable performance measure. Saccades show characteristic impairments in PD and show promise as the basis of a novel biomarker. Our current understanding of the causes of saccade performance deficits in PD is speculative. This thesis will examine saccades in PD using MRI, aiming to develop a better understanding of PD processes affecting eye movements and assessing MRI analyses for future studies into this area.

Eye movement and MRI data from 96 PD and 33 controls from the New Zealand Brain Research Institute was analysed in the first study using voxel-based whole-brain comparisons of structural, perfusion and diffusion MRI, with visually-guided and volitional saccade tasks. The results largely indicated no significant group differences; however subtle structural and perfusion group-differences were detected in the temporal lobe and precuneus. A characteristic PD-related perfusion pattern has been previously identified for 68 PD and 24 controls of this study, the extent of which is expressed by a network score in a given individual. This network score was compared with saccadic eye movement performance of these participants. Every eye movement measure significantly correlated with the network score, supporting the use of this score as a biomarker of general PD status. Furthermore, increased hypoperfusion in the areas described by the network score may contribute, in part, to eye movement deficits in PD.

In the next study, 16 PD and 16 controls underwent functional MRI scanning while performing reflexive and predictive saccades. Group-level activity differences were not detected, however, the medial frontal eye field (FEF) subregions demonstrated greater activity in the reflexive task compared to the predictive. This finding is novel and FEF subregions have not been explicitly described to relate to reflexive or predictive saccade tasks. This is relevant in the context of our overall understanding of eye movement control.

The final study investigated memory-guided saccades in PD using functional MRI. Blood-oxygen level dependent (BOLD) activity in the bilateral posterior parietal cortex was altered in PD patients compared to controls during this task. Additional principal component (PCA) and K-means clustering analyses were able to identify regions exhibiting similar BOLD time courses during task performance. This supports the ability of PCA to identify key components patterns related to function, from BOLD time course data.

This study of eye movements in PD using MRI has identified several areas for investigation in future studies. This study was cross-sectional but has allowed us to establish initial benchmark findings for future comparison, and has helped validate analysis methods for future longitudinal studies into eye movements and PD using MRI.