Differential Cognitive Effects of Unilateral Subthalamic Nucleus Deep Brain Stimulation for Parkinson's Disease

Abstract

Objective: The aim of this study was to investigate the cognitive effects of unilateral directional versus ring subthalamic nucleus deep brain stimulation (STN DBS) in patients with advanced Parkinson's disease.

Methods: We examined 31 participants who underwent unilateral STN DBS (left n = 17; right n = 14) as part of an National Institutes of Health (NIH)-sponsored randomized, double-blind, crossover study contrasting directional versus ring stimulation. All participants received unilateral DBS implants in the hemisphere more severely affected by motor parkinsonism. Measures of cognition included verbal fluency, auditory-verbal memory, and response inhibition. We used mixed linear models to contrast the effects of directional versus ring stimulation and implant hemisphere on longitudinal cognitive function.

Results: Crossover analyses showed no evidence for group-level changes in cognitive performance related to directional versus ring stimulation. Implant hemisphere, however, impacted cognition in several ways. Left STN participants had lower baseline verbal fluency than patients with right implants (t [20.66 = -2.50, p = 0.02]). Verbal fluency declined after left (p = 0.013) but increased after right STN DBS (p < 0.001), and response inhibition was faster following right STN DBS (p = 0.031). Regardless of hemisphere, delayed recall declined modestly over time versus baseline (p = 0.001), and immediate recall was unchanged.

Interpretation: Directional versus ring STN DBS did not differentially affect cognition. Similar to prior bilateral DBS studies, unilateral left stimulation worsened verbal fluency performance. In contrast, unilateral right STN surgery increased performance on verbal fluency and response inhibition tasks. Our findings raise the hypothesis that unilateral right STN DBS in selected patients with predominant right brain motor parkinsonism could mitigate declines in verbal fluency associated with the bilateral intervention. ANN NEUROL 2024;95:1205-1219.

Figure 1.. Participant CONSORT Diagram.

Figure 2.. Cognitive performance following directional versus ring unilateral STN stimulation.

(A) Gray lines indicate within-participant changes in cognitive performance over time after unilateral STN DBS, bold lines indicate sample means, arrows indicate direction of improved function, and p-values indicate the significance of the slope over time. (B) Unity plots contrast change in cognitive performance versus pre-op baseline during ring versus directional stimulation. Open circles indicate participants who were put on ring stimulation first during the double-blind stimulation randomization, and closed circles indicate participants who were put on directional stimulation first. p-values indicate the significance of the slope. None of the y-intercepts were significant (verbal fluency: p=0.750; response inhibition: p=0.874; immediate recall: p=0.421; delayed recall: p=0.273)

Figure 3.. Baseline and longitudinal cognitive function in unilateral STN DBS patients.

(A) Boxplots of baseline multi-domain cognitive performance by implant hemisphere. p-values indicate significance between groups at baseline. (B, C) Line segments show within-participant changes in cognitive performance over time by implant hemisphere, and bold lines show group means. On raw scores of tests of verbal fluency and memory, higher scores reflect increased output, while a lower score on a test of inhibition reflects a faster performance. On percent change of all tests, a positive change reflects an increase while a negative change reflects a decline. p-values indicate significant difference between groups in slopes over time.

Figure 4.

(A) Relationships between the percent change from baseline in verbal fluency and in UPDRS part III scores. There was no relationship between change in verbal fluency and change in UPDRS III, at the group level or by hemisphere. (B) Relationships between the percent change from baseline in response inhibition and in UPDRS part III scores. Faster response inhibition was significantly related to improvements in UPDRS III scores, at the group level and within each hemisphere. (C) Quality of life, measured by the Parkinson’s disease 8 questionnaire, improved after DBS. (D) BrainLab rendered schematic of lead locations relative to the geometric center of each participant’s STN in AC-PC space. Pink indicates the overlaid individual STNs, and the color scale reflects improvements, declines, or stability in verbal fluency performance for the active contact at 12 months. (E) Verbal fluency performance as a function of whether the DBS lead passes through the caudate, touches the caudate, or does not touch the caudate. In this figure, there are multiple observations per participant, which is how boxplots and error bars were generated for the small number of cases where the DBS lead passes through the caudate. As can be seen, passing through the caudate in either hemisphere is detrimental to verbal fluency. Also, after removing the cases where the caudate is traversed, the same pattern of left STN decline and R STN increase at the group levels is observed.