Introduction: Obstructive sleep apnea (OSA) is highly prevalent in patients with cardiovascular disease (CVD), and is associated with increased mortality if left untreated. Health wearable devices capable of tracking physiologic parameters, trends, and patterns could provide convenient, early screening for OSA. This study is designed to assess (1) the accuracy of nocturnal biometric measurements collected by an investigational wrist-worn device, and (2) its utility as a portable screening device for detection of apnea.
Methods: Thirty patients with suspected OSA were recruited for polysomnography (PSG) studies and asked to wear a photoplethsmography (PPG) device (Wavelet Health, Mt. View, CA) on their wrist. Multiple physiologic parameters from the ground truth PSG device were compared to those recorded by wristbands, and included heart rate (HR), heart rate variability (HRV), oxygen saturation (SpO2) and respiration rate (RR).
Results: Comparison of HR, HRV, SpO2 and RR between PSG and wristband indicate root mean square error of 1.5±0.7 BPM, 23±10 ms, 2.9±1.4 % and 3.4±0.7 breaths per minute, respectively. There was a consistent agreement between SpO2 and HRV variations and apnea events (Figure.1). Minimum nocturnal SpO2 estimations from the wrist correlated with the finger clip pulse oximetry measurements (R2 = 0.9, p < 0.05).
Conclusion: Clinical validation of digital health wearables is the essential first step to access their applicability in real world settings. Correlation of hypoxia events between this PPG device and clinical reference PSG support further evaluation of wrist-worn health wearables for OSA screening in high risk CVD patients.
Figure 1. Blood oxygen saturation (SpO2, top) and heart rate variability (HRV, bottom) variations measured by the clinical PSG (reference) and the wrist-worn PPG device during clinical sleep study