Abstract
INTRODUCTION: Heart rate variability (HRV) reflects cardiac autonomic regulation, and reduced HRV is associated with Parkinson's disease (PD). However, studies regarding the implications of HRV measures for the clinical manifestations of PD have shown inconclusive results. We examined the relationship between HRV measures obtained via long-term monitoring using a wearable electrocardiogram (ECG) device and the diagnosis and clinical characteristics of PD. METHODS: Seventeen controls and 20 patients with PD were prospectively enrolled. The HRV measures were recorded using a wearable ECG device for up to 72 h. Time- and frequency-domain measures were derived from the HRV analysis, and their association with PD diagnosis and clinical features was investigated. We investigated their association with neuroimaging biomarkers using magnetic resonance imaging to explore the underlying neural correlates. RESULTS: The diagnosis of PD was associated with several HRV measures, including a decreased standard deviation of N-N intervals, standard deviation of all heart rates, and low-frequency (LF) power. Among these HRV measures, only LF power was associated with clinical features of PD. LF power was positively correlated with the tremor sub-score (r = 0.500; p = 0.035) and negatively associated with the left (r = -0.598; p = 0.024) and right (r = -0.693; p = 0.006) cerebellar hemispheres in patients with PD. CONCLUSION: Low-frequency power may be used as a biomarker for tremor-associated pathophysiology of PD. Moreover, a wearable ECG device with its capability for long-term monitoring might be a promising tool for diagnosing PD.