Prediction of sleep apnea severity using a textile-based electrocardiogram monitoring device

Abstract

Introduction: Sleep apnea syndrome (SAS) is diagnosed and graded by polysomnography (PSG) based on the number of apnea-hypopnea events per hour of sleep (apnea-hypopnea index, AHI). PSG is expensive and technically complex, therefore, simpler and cost-effective devices for screening and follow-up are desirable. In this context, a wearable electrocardiogram (ECG) acquisition system was developed by the Swiss Federal Laboratories for Materials Science and Technology. The aim of this study was to evaluate the value of overnight ECG recordings to assess apnea severity using an ECG-belt in a population of patients with suspicion of SAS. Methods: The ECG-belt consists of a semi-elastic polyester backbone with directly embroidered Ag/Ti-coated PET yarn forming the electrodes. Patients with suspicion of SAS were recruited. ECG signal was recorded in every patient using both the belt and PSG. Heart rate variability (HRV) analysis was performed and prediction accuracy was assessed for both acquisition systems using support vector machines. Results: Whole night measurements were performed in 241 patients. The median AHI was 21 [IQR: 7-40]/h and 65% of the patients suffered from SAS (AHI >10/h). The ECG-belt had an accuracy, sensitivity and specificity of 72%, 70% and 74%, respectively, to discriminate individuals with no/mild SAS vs. moderate/severe SAS. The prediction accuracy was comparable to PSG. Conclusion: In individuals suspected for SAS a novel textile-based ECG-belt was able to detect HRV-based disturbances of the vegetative nervous system, which allowed for the assessment of sleep apnea severity. HRV gives additional insights into the cardio-vascular stress level during sleep.