ABSTRACT
PURPOSE: This study aimed to evaluate the accuracy of a Bluetooth position monitor called NaTu sensor and its mobile phone application for detecting sleep position among patients with obstructive sleep apnea (OSA) during polysomnography (PSG). METHODS: A cross-sectional study was conducted on adults with suspected of having OSA who underwent PSG. Sleep positions were recorded simultaneously using a video-validated PSG position sensor and the NaTu sensor. The area under the Receiver Operator Characteristic curve (ROC AUC), sensitivity, and specificity values were calculated to evaluate the validity of the NaTu sensor. RESULTS: Ninety participants (56.7% male) were included, with median age of 40.0 years and body mass index of 29.4 kg/m2. The mean AHI was 58.4 ± 31.2 events/hour, categorizing the severity of OSA as mild (5.6%), moderate (18.9%), and severe (75.5%). Sleep positions (supine, lateral right, lateral left) identified by the NaTu sensor closely agreed with the video-validated PSG. The kappa statistic demonstrated almost perfect agreement (k = 0.95, P < 0.001) for overall position recording. The ROC AUC for identifying supine, lateral right, and lateral left positions ranged from 0.974 to 0.981, with sensitivity ranging from 95.1% to 99.1% and specificity from 96.5% to 99.6%. CONCLUSION: Our wearable sensor monitoring significantly agrees with video-validated PSG in identifying sleep positions. This device is reliable and accurate for position monitoring and could be an alternative tool for monitoring positions in in-lab PSG, home sleep apnea testing, or tracking positional treatment at home. REGISTRATION: Thaiclinicaltrials.org with number TCTR20210701008.
