Feasibility and performance evaluation of PPG on a Galaxy Watch in continuous central blood pressure monitoring

Abstract Aims Wearable, cuffless blood pressure (BP) monitoring systems using photoplethysmography (PPG) offer promising alternatives to traditional methods for out-of-office BP but lack rigorous validation against invasive continuous central BP measurement. This study aimed to develop and evaluate a smartwatch-based algorithm for detecting changes in mean arterial pressure (MAP) using PPG signals from the Samsung Galaxy Watch, with invasive central BP as the reference standard. Methods and results A novel algorithm was developed and calibrated using 6117 measurements from 440 participants. External evaluation was performed prospectively in 114 participants undergoing left and/or right heart catheterization, where central aortic BP was recorded at 1-min intervals using pigtail catheters. Simultaneous non-invasive PPG signals were collected from Galaxy Watch 6. The primary endpoint was the detection of a ≥15% increase in MAP from baseline, with receiver operating characteristic analysis performed to evaluate diagnostic accuracy. After calibration, the smartwatch-estimated MAP changes demonstrated strong correlation with invasive measurements (r = 0.92, P < 0.001), with a mean bias of 0.51 mmHg (Bland–Altman analysis). The algorithm showed excellent diagnostic performance for detecting ≥15% increases in MAP (AUC = 0.85; 95% CI = 0.80–0.91) and remained robust in an uncalibrated dataset. The time delay between invasive and non-invasive BP changes was minimal (median = 1.0 min), indicating near real-time tracking capability. Conclusion This study demonstrates the preliminary feasibility and accuracy of a PPG-based smartwatch algorithm for detecting meaningful BP changes in real time without the need for frequent calibration. Since our protocol did not meet the relevant validation requirements, further studies are needed to substantiate its potential clinical applicability.