Introduction: Continuous, minimally invasive monitoring of arterial health has significant potential for both improving our understanding of early progression of arteriosclerosis/atherosclerosis and enabling early treatment. Though devices exist that can report heart rate and heart rate variability in fully ambulatory settings, monitoring of pulse pressure wave morphology, including pulse wave velocity (PWV) and augmentation index (AIx) – two emerging indicators of arterial stiffness – is still currently limited to benchtop systems requiring an expert operator or cuff-based systems that are not well tolerated.

Objective: To develop and validate a device capable of continuous monitoring of hemodynamic and electrophysiological signals, including computation of derived vascular stiffness indicators (peripheral PWV and AIx) comparable to current standards.

Methods: We developed a wrist-worn device for continuous radial pulse wave capture and analysis (e.g., peripheral AIx computation) and pulse wave velocity measurement (measured from EKG landmarks to the radial pulse arrival time). The core sensors on the device are (1) a pressure sensor that rests over the radial artery, (2) a pair of dry EKG electrodes in lead-I configuration, and (3) a six-axis motion sensor. It includes a Bluetooth radio for data streaming, and sufficient on-board storage for 2 weeks of ambulatory studies. To validate the device against the current gold standard for radial tonometry, we recorded multiple 60-second sessions of tonometric data using both our device (2 sessions) and the standard Millar SPT-301 tonometer (3 sessions) in 28 subjects. We then compared AIx computed from these two devices.

Results: Using our wrist-worn device we calculated AIx values that were highly consistent with those of the Millar tonometer (r = 0.927, mean difference ± SD of 0.026 ± 0.049).

Conclusions: This platform enables continuous non-invasive hemodynamic monitoring. We intend to utilize it in future studies about how peripheral AIx and PWV behave over longer time periods, how they correlate to daily activities, including sleep, exercise, and food/medication intake, and – most importantly – whether they can serve as new, ambulatory indicators of cardiovascular risk.