Cuffless estimation of blood pressure: importance of variability in blood pressure dependence of arterial stiffness across individuals and measurement sites

Objective: Measuring arterial pulse transit time (PTT) to estimate blood pressure (BP) without conventional brachial cuff-based measurement is not new, but is a focus of current wearable technologies research. Much research pertains to efficient, accurate sensing of artery-related waveforms, yet the relationship between PTT and BP receives less attention despite being key for accurate BP estimation. This study investigated BP/PTT calibration by quantifying anatomical site variability (n=10, 3 female, age 30±9 years) and individual variability (n=103, 50 female, age 53±22 years). Methods: BP and pulse wave velocity (PWV) were measured both seated and supine. Carotid-femoral PWV (cfPWV), carotid-radial PWV (crPWV) and carotid-finger-volume PWV (cvPWV) were measured with the wrist and hand positioned at the level of the upper thigh to achieve the same hydrostatic pressure effect across all measurements. Results: The postural change invoked a small (4±7 mmHg) change in brachial diastolic BP with an additional 27±2 mmHg change in hydrostatic pressure. cfPWV decreased in the supine position (-1.75±0.17 m•s^-1, p<0.001) but crPWV and cvPWV were more variable. The calibration term (ΔBP/ΔPWV) across the sample population varied from 6.6 to 98.3 mmHg•s•m^-1 (mean 22±14 mmHg•s•m^-1) and was correlated with age, heart rate, diastolic and pulse pressure, and weight. These variables did not explain the majority of the variability (R²=0.248). Conclusion: There is anatomical site and between-individual variability in the calibration term for BP estimation from PTT. Significance: Using and accurately calculating hydrostatic changes in BP within the individual may be one method to increase accuracy of this calibration term.