Smartwatch for OSA screening scores high in pilot trial

A smartwatch that captures photoplethysmography (PPG) signals for screening obstructive sleep apnoea (OSA) performs as good as polysomnography (PSG) or home sleep apnoea test (HSAT), as shown in a pilot trial.

A total of 119 patients (median age 49 years, 23.5 percent female, mean body mass index 26.6 kg/m²) were recruited from the Chinese People’s Liberty Army General Hospital. Of these, 20 patients were assessed for a whole-night sleep study by a smartwatch and PSG simultaneously, 82 by a smartwatch and HSAT simultaneously, and the remaining 17 were excluded due to the poor quality of PPG signals.

A total of 83 patients were diagnosed with OSA (mild 36.1 percent; moderate 21.7 percent; severe 42.2 percent), as confirmed by physicians based on patients’ medical history, physical examinations, and the results of HSAT or PSG test.

“The results of the screening algorithm used in the smartwatch were consistent with those results obtained from the testing of medical devices used to diagnose sleep apnoea. In addition, we concluded that the diagnostic efficiency of the smartwatch for moderate-to-severe apnoea was higher than that for the mild condition,” the investigators noted.

Indeed, when compared with HSAT, the smartwatch had an accuracy of 87.9 percent for predicting moderate-to-severe OSA (AHI ≥15); the corresponding sensitivity and specificity were 89.7 percent and 86.0 percent, respectively. Meanwhile, compared with PSG, the smartwatch showed an accuracy of 92.9 percent, sensitivity of 85.7 percent, and specificity of 100 percent. [Nat Sci Sleep 2021;13:1533-1544]

The predictive ability of the smartwatch for moderate-to-severe OSA patients (AHI ≥15) was comparable with that of HSAT (p=0.75) and PSG (p=0.52).

PRV as surrogate for HRV

The investigators used a broadly used wearable device that integrates seven sensors, including an optical heart rate sensor and a wrist pulse oximeter, which were specifically used to detect blood volume changes in the microvascular bed of the tissue and record PPG signals.

A proprietary algorithm facilitated generation of clinically relevant respiratory waveforms and pulse rate variability (PRV) as well as detection of the sleep-wake states from the PPG data, the investigators noted. “Thus, sleep time, apnoea-hypopnea index (AHI), and oxygen saturation could be estimated.”

Whereas OSA has shown a good correlation with heart rate variability (HRV) and blood oxygen saturation, the investigators used PRV as a surrogate for HRV, in view of the good agreement between PRV and HRV, as well as the portability and continuity of PRV monitoring. [Sleep Sci 2019;12:214-221; Eur J Anaesthesiol 2011;28:34-38]

Large population screening

“A smart device, such as a PPG-based smartwatch, is used to monitor health status during sleep,” according to the investigators who pointed out the advantages of using the device for OSA screening.

“Firstly, the smartwatch makes testing at home extremely convenient. Secondly, the low load reduces disturbance during sleep monitoring. Finally, the repeatability and low cost can be conducive to screening OSA in a large population,” they said.

The investigators also underscored a need for the screening of OSA and the potential benefit of PPG-based approach among high-risk populations, given that most of the patients who participated in the pilot trial study had comorbid conditions, such as hypertension, hyperlipidaemia, arrhythmia, and coronary heart disease. Only 13 (12.7 percent) patients had no other complications except for snoring.