Smartwatch-Based Blood Pressure Measurement Demonstrates Insufficient Accuracy

Area of Science:

• Biomedical Engineering
• Cardiovascular Technology
• Wearable Health Devices

Background:

• Novel cuffless blood pressure (BP) measuring smartwatches are emerging with regulatory approval.
• These devices often lack sufficient clinical validation.
• This study evaluates a CE-cleared smartwatch for BP measurement in diverse patient groups.

Purpose of the Study:

• To investigate the accuracy and reliability of a specific CE-cleared smartwatch for cuffless BP measurement.
• To assess BP measurement bias and precision in normotensive and hypertensive individuals.
• To compare BP variability between smartwatch and traditional ambulatory monitoring.

Main Methods:

• Simultaneous BP measurements using a Samsung Galaxy Watch Active 2 and 24-h ambulatory BP monitoring (ABPM).
• Calibration of the smartwatch followed by 24-h monitoring.
• Accuracy assessment via sensitivity, specificity, predictive values, ROC curves, Bland-Altman, and Taffé methods.
• BP variability analysis using standard deviation, coefficient of variation, and average real variability.

Main Results:

• Forty patients participated; proportional bias was observed in systolic and diastolic BP measurements.
• Low systolic BPs were overestimated, high BPs underestimated; diastolic BPs were consistently overestimated.
• Sensitivity and specificity for hypertension detection were 83% and 41%; AUCs were 0.78 (systolic) and 0.93 (diastolic).
• BP variability was systematically higher with ABPM than the smartwatch.

Conclusions:

• The Samsung Galaxy Watch Active 2 exhibits systematic bias, rendering it unsuitable for clinical use currently.
• Existing standards for non-invasive sphygmomanometers are not met and may not apply to cuffless devices.
• New validation standards are urgently needed for wearable BP monitors, alongside further research into non-invasive, high-frequency BP monitoring.