Article Impact Level: HIGH Data Quality: STRONG Summary of Sports Medicine. https://doi.org/10.1007/s40279-024-02066-5 Dr. Hayden G. Dewig et al.
Points
- The study evaluated the equivalence of heart rate variability (HRV) measurements derived from photoplethysmography (PRV) versus traditional electrocardiography (ECG) among 1,084 subjects.
- A Bayesian simulation method was employed to account for pulse arrival time (PAT) variability, correlating PPG wave arrival times with QRS-wave timings from ECG data.
- The study analyzed the root mean square of successive differences (RMSSD) and the standard deviation of N-N intervals (SDNN) to compare PRV and HRV, setting a Region of Practical Equivalence (ROPE) at ±0.2% of true HRV.
- PRV matched HRV within the ROPE for SDNN in 88.4% of cases with low PAT variability (σ = 2.0), but this equivalence significantly decreased as PAT variability increased, particularly affecting RMSSD.
- The study concluded that PRV is not a reliable substitute for HRV in individuals with high PAT variability, as measurement errors increase with PAT variability, significantly impacting RMSSD.
Summary
This study investigates the equivalence of photoplethysmography-derived heart rate variability (PRV) with traditional electrocardiography (ECG)-derived heart rate variability (HRV) among 1084 subjects from the PhysioNet Autonomic Aging dataset. To address the variability in pulse arrival time (PAT)—the interval between a heartbeat and the resulting peripheral pulse—a Bayesian simulation method was used. This simulation calculated the arrival times of the PPG wave based on Gaussian prior distributions assigned to individual QRS-wave timings of each ECG series, correlating these with PAT dispersion data from earlier research.
Analyzing the root mean square of successive differences (RMSSD) and the standard deviation of N-N intervals (SDNN) in both HRV and PRV measures, the study set a priori Region of Practical Equivalence (ROPE) bounds at ±0.2% of true HRV. The results indicated that with the lowest PAT standard deviation (σ = 2.0), PRV coincided with HRV within the ROPE for SDNN in 88.4% of cases and only 21.4% for RMSSD. As PAT variability increased, the equivalence of PRV to HRV significantly decreased for both SDNN and RMSSD measures, highlighting a more considerable disparity in RMSSD.
The findings underscore that PRV cannot reliably substitute for HRV in individuals with high PAT variability. The measurement error associated with PRV increased with PAT variability, particularly affecting RMSSD more than SDNN. Thus, while PRV may offer some utility as a unique biometric measure, it exhibits significant limitations compared to HRV, especially under conditions of high physiological variability.
Link to the article: https://link.springer.com/article/10.1007/s40279-024-02066-5
References Dewig, H. G., Cohen, J. N., Renaghan, E. J., Leary, M. E., Leary, B. K., Au, J. S., & Tenan, M. S. (2024). Are Wearable Photoplethysmogram-Based Heart Rate Variability Measures Equivalent to Electrocardiogram? A Simulation Study. Sports Medicine. https://doi.org/10.1007/s40279-024-02066-5
