Objective: To characterize ordinal patterns of tremor amplitude across rest, postural, and kinetic conditions in dystonic tremor (DT) and Parkinson’s disease (PD) using accelerometry, and to evaluate whether visual bedside assessment based on these patterns can reliably differentiate DT from PD

Background: Differentiating dystonic tremor DT from PD remains challenging. Dystonic features can be subtle, and DT is frequently misdiagnosed as PD, representing a major cause of scans without evidence of dopaminergic deficit (SWEDDs). Prior accelerometry studies suggest PD tremor predominates at rest whereas DT is more pronounced during action, but systematic comparisons in dopamine transporter imaging-confirmed cohorts remain limited.

Method: We prospectively enrolled 102 patients (51 DT, 51 PD) from a tertiary movement disorders center, all confirmed by dopamine transporter imaging. Accelerometer data were recorded during three standardized conditions: rest, posture (arms outstretched), and kinetic movement (spiral drawing). Ordinal diagnostic rules were explored stepwise, evaluating pairwise comparisons, maximal/minimal condition identification, and logical combinations. Three blinded raters visually classified tremor predominance based on the optimal rule. Inter-rater agreement was assessed using Fleiss’ kappa, and concordance with accelerometry using Cohen’s kappa.

Results: DT patients demonstrated increasing tremor amplitude across conditions (rest < postural < kinetic), whereas PD showed the opposite pattern (rest > postural > kinetic). Among all candidate rules, a simple pairwise comparison between kinetic and rest tremor amplitudes achieved the best-balanced performance (88% sensitivity, 88% specificity, 88% accuracy). Three-condition rules and logical combinations did not outperform this pairwise rule. Visual classification showed excellent inter-rater agreement (Fleiss’ kappa=0.92), high concordance with accelerometer-based classification (93.1%; Cohen’s kappa=0.86), and 89% sensitivity, 87% specificity, and 89% accuracy for identifying DT.

Conclusion: A simple ordinal comparison between kinetic and rest tremor amplitudes captures a clinically meaningful polarity differentiating DT from PD, with high inter-rater reproducibility and strong concordance with quantitative accelerometry, supporting its utility as a practical bedside adjunct.

« Back to 2026 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/kinetic-rest-tremor-polarity-for-differentiating-dystonic-tremor-and-parkinsons-disease/