# Differentiating effects of levodopa and subthalamic nucleus deep brain stimulation on motor features in Parkinson disease

## TL;DR

The study compares how levodopa and subthalamic nucleus deep brain stimulation affect different motor symptoms in Parkinson's disease, finding each treatment has unique strengths.

## Contribution

The study identifies six distinct motor domains and shows that treatment responses to levodopa and STN-DBS are only weakly correlated, suggesting symptom-specific treatment selection.

## Key findings

- STN-DBS is more effective for upper body tremor compared to levodopa.
- Levodopa shows greater improvement in axial symptoms and lower body movement issues.
- Weak correlations between treatment responses suggest limited predictability of one treatment's success based on another.

## Abstract

•Factor analysis identified six distinct parkinsonian motor domains across conditions.•STN-DBS superior for upper body tremor; levodopa better for axial symptoms and legs.•Weak correlations between levodopa and STN-DBS challenge use of levodopa response in candidacy.•Treatment selection should target specific symptoms rather than global scores.

Factor analysis identified six distinct parkinsonian motor domains across conditions.

STN-DBS superior for upper body tremor; levodopa better for axial symptoms and legs.

Weak correlations between levodopa and STN-DBS challenge use of levodopa response in candidacy.

Treatment selection should target specific symptoms rather than global scores.

While deep brain stimulation of the subthalamic nucleus (STN DBS) is traditionally used to treat motor fluctuations in Parkinson disease (PD), recent progress in levodopa delivery systems may offer non-invasive alternatives to surgical options. To gain insight into symptoms more likely to be treated effectively by STN DBS or levodopa, we systematically compared the differential effects of levodopa and STN-DBS on specific parkinsonian motor components.

We retrospectively analyzed 395 PD patients who received bilateral STN-DBS at Washington University (1999–2020). UPDRS-III was obtained in three conditions: preoperative OFF-medication, preoperative ON-medication, and postoperative ON-DBS/OFF-medication. Exploratory factor analysis identified a consensus structure across conditions. Treatment responses by factor were compared using Wilcoxon rank-sum tests and correlations were assessed with Kendall’s tau.

Both treatments significantly improved all motor features. STN-DBS showed slightly superior efficacy for upper body tremor (mean change: −3.84 vs −3.43, p < 0.001), while levodopa demonstrated much greater effectiveness for axial symptoms (−5.41 vs −2.65, p < 0.001), and slightly greater effectiveness for lower body tremor (−1.25 vs −1.06, p < 0.001), and lower body bradykinesia (−1.18 vs −0.91, p = 0.005). Correlations between treatment responses were weak but statistically significant for rigidity (τ = 0.158), bradykinesia (hand: τ = 0.118; leg: τ = 0.179), and axial symptoms (τ = 0.207).

Levodopa and STN-DBS demonstrate similar but distinct therapeutic profiles across parkinsonian motor domains. However, weak correlations between treatment modalities for all motor features challenge the utility of levodopa responsiveness for DBS candidacy and support individualization of treatment selection based on symptom profile.