Medication adversely impacts visually-guided eye movements in Parkinson's disease

Abstract

Objective: We examined whether previous inconsistent findings about the effect of anti-Parkinsonian medication on visually-guided saccades (VGS) were due to the use of different paradigms, which change the timing of fixation offset and target onset, or different target eccentricities.

Methods: Thirty-three participants with Parkinson's disease (PD) completed the VGS tasks OFF and ON medication, along with 13 healthy controls. Performance on 3 paradigms (gap, step, and overlap) and 2 target eccentricities was recorded. We used mixed models to determine the effect of medication, paradigm, and target eccentricity on saccade latency, gain, and peak velocity.

Results: First, we confirmed known paradigm effects on latency, and target eccentricity effects on gain and peak velocity in participants with PD. Second, latency was positively associated with OFF medication Movement Disorders Society - Unified Parkinson's Disease Rating Scale (MDS-UPDRS) motor score in PD. Third, medication prolonged latency for the larger target eccentricity across the 3 paradigms, while decreasing gain and peak velocity in the step paradigm across target eccentricities.

Conclusions: Medication adversely affected and was not therapeutically beneficial for VGS. Previous inconsistencies may have resulted from chosen target eccentricity.

Significance: The negative medication effect on VGS may be clinically significant, as many activities in daily life require oculomotor control, inhibitory control, and visually-guided shifts of attention.

Keywords: Levodopa; Parkinson’s disease; Response latency; Saccades.

Fig. 1.

Experimental procedure for the 3 visually-guided saccade task paradigms. Trials begin with central fixation, then a peripheral target appears, and participants make a saccade to the target. Inset: The timing of target onset in relation to fixation offset differed in each paradigm: gap (top), step (middle), and overlap (bottom).

Fig. 2.

Primary saccade latency. (A) Observed mean latencies and standard errors for healthy controls (HC) and participants with Parkinson’s disease (PD) OFF and ON medication. Means are shown for each target eccentricity, 10° (left panel) or 15° (right panel), and for each paradigm, gap (light gray), step (medium gray), or overlap (dark gray). (B) The interaction between medication status (PD OFF vs PD ON) and target eccentricity (10° (dark gray) or 15° (light gray)) averaged over paradigm. (C) OFF medication Movement Disorders Society – Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) Part III motor score was a significant covariate in the mixed model, which was positively associated with mean latency (gray line). The observed latency means averaged over medication status, paradigm, and target eccentricity (black dots) are overlaid on the estimated relationship.

Fig. 3.

Primary saccade gain. Observed mean gain and standard errors for healthy controls (HC) and participants with Parkinson’s disease (PD) OFF and ON medication. Means are shown for each target eccentricity, 10° (left panel) or 15° (right panel), and for each paradigm, gap (light gray), step (medium gray), or overlap (dark gray).

Fig. 4.

Primary saccade peak velocity. (A) Observed mean peak velocity and standard errors for healthy controls (HC) and participants with Parkinson’s disease (PD) OFF and ON medication. Means are shown for each target eccentricity, 10° (left panel) or 15° (right panel), and for each paradigm, gap (light gray), step (medium gray), or overlap (dark gray). (B) Levodopa equivalent daily dose (LEDD) was a significant covariate in the mixed model, which was negatively associated with mean peak velocity (gray line). The observed peak velocity means averaged over medication status, paradigm, and target eccentricity (black dots) are overlaid on the estimated relationship.