Heart Rate Variability and Pulse Rate Variability: Do Anatomical Location and Sampling Rate Matter?

Area of Science:

• Biomedical Engineering
• Physiological Measurement
• Neuroscience

Background:

• Photoplethysmography (PPG) and wearable sensors are increasingly used for physiological monitoring.
• Existing studies show bias between pulse rate variability (PRV) from PPG and heart rate variability (HRV) from ECG.
• The minimum sampling frequency for valid PRV and its comparability between cerebral and peripheral vasculature remain unclear.

Purpose of the Study:

• To determine the minimum sampling frequency for valid PRV metrics.
• To compare PRV and HRV metrics across different anatomical locations.
• To assess the validity and bias of PRV compared to HRV at various sampling rates.

Main Methods:

• Concurrent electrocardiography (ECG) and PPG data collected from 54 participants in an upright position.
• PPG data acquired at the left third phalanx, middle cerebral artery, and posterior cerebral artery at 1000 Hz.
• Data downsampled (20-500 Hz) and analyzed for HRV and PRV using Bland-Altman plots and coefficient of variation.

Main Results:

• A sampling frequency of approximately 100-200 Hz yielded PRV metrics with <2% bias.
• A sampling rate of 40-50 Hz resulted in a bias of <20% for PRV metrics.
• PRV measures showed a slight elevation (1-8% bias) compared to HRV at 1000 Hz, with greater validity at lower sampling rates.

Conclusions:

• PRV and HRV are not interchangeable due to waveform differences, despite correlations.
• Optimal sampling rates for PPG-derived PRV are identified for minimizing bias.
• PPG-derived PRV demonstrates higher validity at lower sampling frequencies compared to ECG-derived HRV.