Category: Ataxia
Objective: To assess the differences of HRV parameters between MSA-C and SCA and their age-matched normal controls and also to use the supervised ML to construct the distinctive model.
Background: Cardiac autonomic function assessment in multiple system atrophy-cerebellar subtype (MSA-C) and spinocerebellar ataxia (SCA) using heart rate variability (HRV) during both at rest and deep breathing (DB) has been rarely systematically studied. Also, none has used machine learning (ML) process to construct the distinctive model to help differentiate these conditions.
Method: Patients with either MSA-C or SCA, along with their respective 1:1 age- matched normal controls were recruited from the Neurological Institute of Thailand (NIT) during Dec 2023 to Jun 2024. Recording with Polar® H10 chest strap, 5-minute each during at rest and DB, was performed in all participants. Time and frequency domain HRV parameters were compared between the diseases and their controls, and also between the two disease conditions. ML with different methods was then employed to find best distinctive model.
Results: Forty-four patients including 22 MSA-C and 22 SCA, and 44 normal controls were included for analysis. Most HRV parameters in both MSA-C and SCA were significantly lower than controls. Percentage change of the HRV values during DB as compared with at rest was comparable between SCA and their controls, but significantly different for MSA-C. Top-performing model was constructed to differentiate MSA-C from SCA with a recall (sensitivity) of 1.00 and 0.57 for MSA-C and SCA, respectively. Corresponding precision (positive predictive value) was 0.7 and 1.00 and F1 score was 0.82 and 0.73
Conclusion: Cardiac autonomic functions, both sympathetic and parasympathetic systems, were impaired in MSA-C and SCA, more pronounced in MSA-C. Baseline HRV at rest disclosed a preferential parasympathetic loss in SCA, whereas a less efficient sympathetic activity was demonstrated during DB in MSA-C. Distinctive model using HRV data is promising but required further valid
To cite this abstract in AMA style:
P. Sathirapanya, N. Limotai, C. Limotai, N. Suanprasert, S. Rujirussawarawong, C. Kongkamol, N. Cheetanom. Cardiac Autonomic Dysfunction in Patients with Multiple System Atrophy and Spinocerebellar Ataxia: A Comparative Study and Distinctive Machine Learning Model [abstract]. Mov Disord. 2025; 40 (suppl 1). https://www.mdsabstracts.org/abstract/cardiac-autonomic-dysfunction-in-patients-with-multiple-system-atrophy-and-spinocerebellar-ataxia-a-comparative-study-and-distinctive-machine-learning-model/. Accessed October 5, 2025.« Back to 2025 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/cardiac-autonomic-dysfunction-in-patients-with-multiple-system-atrophy-and-spinocerebellar-ataxia-a-comparative-study-and-distinctive-machine-learning-model/