Studies on eye movements in Parkinson's disease

Abstract

Heterogeneity in Parkinson’s Disease (PD) phenotype and genotype is probably the main reason why, despite the abundance of biomarkers, we still lack a robust method for diagnosis and prognosis, besides clinical evaluation. Subjective changes in vision and objective measures in eye movements have been extensively studied, but the results are mainly used to better understand the pathophysiology of PD and are not integrated into the clinical praxis.

The aim of this doctoral project was to examine if eye movements could serve as useful biomarkers for PD diagnosis and prognosis, and investigate their association with motor function, cognition, and medication effect. In addition, we aimed to examine cognition in a group of patients with a rare metabolic disorder and prominent eye-movement difficulties, the Norrbottnian Gaucher Disease 3 (GD3).

Saccades, reading, and sustained fixation were examined in PD patients and healthy controls (HC) in the first three studies. Recruitment took place at Karolinska University Hospital Huddinge for the first two studies, and for the third study at Academic Specialist Center in Stockholm. Three different eye trackers were used, a head-mounted and two screen based, and the assessments were performed in a clinical setting. In the first two studies patients were examined in ON and OFF medication status, in order to evaluate the role of levodopa. In study 1, we examined saccadic parameters in 20 HC and 40 PD patients; study 2 involved reading assessments for 13 HC and 19 PD patients; in study 3 we examined sustained fixation in 43 HC and 50 PD patients. Recruitment for study 4 took place at Sunderby Regional Hospital, in Luleå, and we examined 10 patients with the Norrbottnian type of GD3. Cognitive evaluation was done with the Repeatable Battery for Assessment of Neuropsychological Status (RBANS).

PD participants had worse saccadic performance, a slower reading speed, and deficient fixation control. Saccadic gain was associated with motor performance, while latency was related to cognition. Levodopa had no effect on saccadic gain, it worsened latency for the horizontal visually guided saccades and ameliorated the latency of antisaccades, but not the error rate or reading performance. We assumed that reading difficulties were attributed to cognitive, rather than oculomotor deficits. Fixation was more easily interrupted in PD compared to HC, and PD participants’ pupils did not dilate to the same extent as HC, in response to the cognitive effort put during sustained fixation. In study 4 we found that patients with the Norrbottnian type of GD3 have an overall worse cognitive performance compared to that of healthy population, scoring worse in memory and attention tests, present however with preserved language and visuospatial skills.

The eye-tracking studies led to the conclusion that this method could be integrated into the clinical praxis as part of the clinical evaluation. It is easy to perform and provides reliable results that enable the understanding of motor, cognitive, and behavioral changes in PD. In order to do so, we would need a common protocol of assessment, so that the results would be comparable between different populations. The last study identified RBANS as a useful and easy-to-use tool for the cognitive examination of Norrbottnian GD3 patients.