Obstructive sleep apnea (OSA) is currently the most common sleep-disordered breathing disorder. It is estimated that 936 million people aged 30-69 worldwide suffer from the disease[1]. At present, all-night PSG with manual on-duty is the gold standard for diagnosing OSA and judging its severity. However, all-night PSG with manual on-duty has defects such as high equipment and environmental requirements, complex analysis techniques, high costs, which are difficult to meet the screening of a large population of OSA diagnostic clinical needs. With the innovation of technology, the diagnosis and monitoring of OSA are gradually developing in the direction of lightweight, portable, and remote. Portable monitoring (PM) devices such as oximeters can accurately record hypoxic events at night and effectively identify sleep Fragmentation is becoming a new trend.
New Diagnostic Indicators of OSA May Replace AHI
Although AHI is still used as an index to evaluate the severity of sleep apnea, more and more physicians report that the AHI index does not match the actual clinical symptoms of patients, and has substantial limitations in predicting the risk of complications such as cardiovascular disease. At Sleep 2023, Zoll held a seminar called Looking Beyond the AHI, and more scholars began to explore whether there is a better indicator than AHI, which can evaluate sleep-disordered breathing more scientifically and objectively.
Most of the new indicators are still derived from PSG. As the gold standard for diagnosing OSA, PSG still has much content waiting to be discovered and applied. Sleep experts are trying to split new indicators from PSG, relying on a broader range of case data and more cutting-edge algorithms to make the diagnosis of OSA lightweight and popular. Some indicators, such as blood oxygen saturation, have clearly shown advantages over AHI, and have been put into large-scale application to guide clinical treatment.
A.Hypoxic Burden
Experiments have shown that the index of nocturnal hypoxia is more predictive of CVD and postoperative complications than AHI. Olaf Oldenburg et al conducted more than 10 years of sleep follow-up and obtained data from 963 heart failure patients with reduced left ventricular function (HF-REF).[2] Fifty-eight percent of these patients had severe-to-severe sleep-disordered breathing (SDB).
Analysis of the above data concluded that the potential number of apnea and hypopnea in a patient’s hour is limited, so the AHI cannot fully describe moderate to severe SDB. The team argues that hypoxemia and hypoxaemic burden, as a composite outcome of apnea and hypopnea, better account for the adverse effects of nocturnal respiratory events and sleep arousal.
B. SBII
OSA has a high specificity, and there are individual differences in the clinical manifestations, pathophysiological mechanisms, and response to treatment of patients. Thus, although the AHI remains the gold standard for classifying OSA severity and guiding treatment, it has limitations in capturing the complete clinical picture of OSA.
As a new indicator, SBII has attracted more and more attention, which is calculated as the sum of the product of the duration of each blocking event and the relevant desaturation area, and then divided by the total insomnia time, the unit is (%min 2 )/h. Lu, Cao et conducted a cluster analysis on 147 OSA patients through 20 OSA symptoms and SBII indicators.[3] The results showed that SBII has higher sensitivity in evaluating the severity of OSA and has a better predictive ability for cardiovascular disease outcomes, may be able to replace AHI for OSA assessment.
C. EEG
Scholars such as Shashank Manjunath believe that EEG signals can be used as new indicators to judge sleep stages through new signal processing techniques.[4] In this experiment, the sleep data of 2881 users were divided into units of 30s, and OSA negative or positive was used as the grouping standard, and a permutation test was performed on each frequency band. The results showed that EEG was statistically significantly different in identifying OSA-positive and OSA-negative patients. Therefore, the team believes that EEG could serve as an emerging indicator to assist in the identification of OSA before respiratory disturbances.
Complications of OSA are becoming a diagnostic focus
A. Hypertension
The incidence of hypertension in patients with sleep apnea is 50% to 90%, while the incidence of sleep apnea in hypertensive patients is 20% to 45%. Sleep-disordered breathing is highly associated with hypertension because sympathetic activation causes vasoconstriction and raises blood pressure through increased vascular resistance. The latest study found that the rate of oxygen saturation desaturation at night was positively correlated with the severity of hypertension.[5] In addition, ambulatory blood pressure monitoring (ABPM) through wearable devices can greatly improve the detection rate of masked hypertension (m-HTM). Therefore, the early diagnosis and treatment of OSA has important positive significance for the management of hypertension.
B.Cardiovascular Disease
As the link between hypoxia burden and cardiovascular disease emerges, studies have demonstrated an overall prevalence of sleep-disordered breathing in patients with symptomatic heart failure of 40% to 60%, with OSA accounting for approximately one-third of cases. In addition, some patients with heart failure who are obese and have reduced ejection fraction have mixed OSA and CSA. [6]
OSA has been generally recognized as an independent risk factor for heart failure (HF) and arrhythmia, but the treatment and management of OSA has not received enough attention in the practical application of cardiovascular diseases. Currently, several teams have attempted to use low-flow nocturnal oxygen therapy to improve hospital admissions and mortality for heart failure and central sleep-disordered breathing.
New methods of collecting physiological information are emerging as a new direction for accurate diagnosis of OSA
Screening tools for OSA are mainly scales and objective sleep monitoring equipment. The sleep scale is Berlin Spa, ESS, etc., but the sensitivity and specificity are relatively low, and it is easy to produce false negative results, and the final score is not positively correlated with the severity of OSA, so it is omitted here.
Sleep monitoring is mainly divided into two types: PSG and HAST. Below we will explain these two monitoring methods according to the more detailed classification methods of I, II, III, and IV.
1. PSG and Type I & Type II Sleep Monitoring Devices
In 1994, the AASM divided the monitoring for OSA assessment and diagnosis into 4 grades: Type I, Type II, Type III or Type IV, and type II~IV were uniformly classified as PM. Both type I and type II equipment require no less than 7 measurement parameters, mainly including nasal and/or oral airflow; thoracoabdominal movement; snoring; EEG; electrooculogram (EOG); electromyogram (EMG); oxygen saturation and occasionally transcutaneous CO2. The main difference between Type I and Type II is the need for on-call and intervention. Type I has always been known as the gold standard for sleep apnea monitoring, and is the core benchmark for comparison of various new methods of monitoring.
At present, the new development direction of PSG has turned to AI. The amount of sleep monitoring data is huge, and the parameters involved are complex and intertwined, which determines that artificial intelligence-assisted analysis will become a powerful helper for PSG.
AI may help Optimize PSG Interpreted and Managed.
In terms of interpretation, AI is learning the specifications of the collection and analysis of sleep monitoring devices, trying to automatically generate sleep analysis reports, including sleep stages, AHI and other indicators. Experiments have proved that the accuracy of artificial intelligence analysis and interpretation of sleep analysis can reach 80%- 85%, and the speed is far beyond the artificial map. However, there is still no grown system in terms of data preprocessing, artificial intelligence judgment standards, and standard data set specifications, which also need continuous exploration.
In terms of sleep management, AI can provide timely feedback on the patient’s physiological state through intelligent interactive functions, which can help to discover the characterization of disease subtypes, thereby more accurately diagnosing and classifying diseases, and gradually realizing the long-term management of OSA subtypes and complications, for more sensitive early intervention and timely warning.
NOX is developing AI algorithms to help sleep technologists deliver diagnostic care faster while ensuring diagnostic accuracy. NOX’s AI algorithm includes sleep staging, AHI and periodic limb movement, and the team verified the accuracy and global adaptability of the AI algorithm by collecting sleep data from different regions. NOX believes that by reducing the cost of data collection and analysis, physicians will have the time to make more rational assessments of treatment effects.
NOx also launched NOXA1S, a PSG system that can be used at home, which can be used to assist in the diagnosis of SDB and monitor features such as insomnia and sleep fragmentation. The combination of electrode simplification, data transmission methods, and AI is making home PSG a reality and gradually promoting the advancement of remote sleep management.
2. HST or HSAT with Type III Sleep Monitoring Devices
Type III sleep monitoring devices require no less than four monitoring parameters: airflow, respiratory effort, oxygen saturation and heart rate, and the devices needs to be easily moved outside the sleep center to evaluate OSA. The most widely used in home sleep apnea test (HSAT) is class III sleep monitoring devices.
The AASM guideline points out that HSAT can be used to clinically assess the diagnosis of high-risk moderate-to-severe OSA patients. It has high sensitivity, specificity, and accuracy for the diagnosis of OSA, and is more friendly to patients with limited mobility or who are not suitable for sleep center monitoring.
Using portable and easy-to-use wearable devices as acquisition devices, and using innovative technologies and algorithms to analyze and calculate AHI and sleep structure has become the main mode of family sleep monitoring. Type III sleep detection equipment is not bound by monitoring methods and monitoring objects. The form and usage of the equipment are based on innovative concepts, and at the same time have advantages in portability and ease of use.
NOX displayed the latest product HAT T3 at the Sleep2023 conference, which includes two customizable bipolar channels (ECG, EMG, EOG, EEG), and the available parameters include snoring signal, SpO2, pulse plethysmography, etc., and the wearing method is RIP chest fixed.
SleepImage also launched a new generation of wearable home sleep testing devices at sleep2023, which has been approved by the FDA and can collect basic blood oxygen data and provide in-depth data on sleep stages, sleep fragments and blood oxygen saturation drops. The ring-shaped design is more in line with the daily living habits of modern people, the silicone material can accommodate fingers of different sizes, which greatly simplifies the equipment and conditions required for the diagnosis of sleep-disordered breathing, whether it is a good choice for people with breathing difficulties or nasal intubation.
New OSA Diagnostic Models are Emerging
OSA is having a huge impact on the medical and health market. In 2015, the medical expenses caused by OSA and its related results in the United States were as high as 12.4 billion US dollars, and it is estimated that it will increase to 49.5 billion US dollars. Early recognition and treatment of OSA will improve patient outcomes, reduce healthcare costs, and reduce the burden of treatment for other complications.
With the development of technologies such as sensors, wireless transmission, and artificial intelligence analysis, it is a new development direction to build a convenient OSAHA diagnostic screening and service model. Comfortable, fast and efficient diagnostic screening is the key to future sleep disease management. Let’s take a look at the new changes in the market.
1. Longitudinal Sleep Monitoring
Additional pathophysiological events triggered by obstructive respiratory events may trigger other complications such as sympathetic activation, oxidative stress, and sleep fragmentation. Therefore, when the physiological state changes at night, the use of AHI index to evaluate OSA in a single night has great limitations. Studies show that the chance of misdiagnosing a patient with OSA based on a single night’s sleep data is about 43%.[7]
Longitudinal sleep monitoring has become a major trend in home sleep testing. Traditional data sleep monitoring often only acquires data for one or several nights. When the physiological state of the tester changes, the sleep center staff can obtain very limited information.
Longitudinal sleep monitoring can more fully identify a patient’s changing sleep patterns over time, helping to optimize treatment options in a timely manner. The collection of longitudinal data could not only improve the feasibility of home sleep testing, but also improve the effectiveness of the HST as a tool for diagnosing and managing sleep. Sonia Ancoli-Israel, professor emeritus of psychiatry at UC San Diego, said “to more personalized treatment and a greater probability of long-term treatment success”
The existing traditional medical model is difficult to meet the huge diagnosis volume of OSA and the need of chronic disease management. Longitudinal sleep monitoring will help more patients with sleep-disordered breathing obtain the chronic disease management services they deserve. Sleep apnea requires ongoing monitoring to assess treatment response. When doctors can obtain objective treatment basis on daily or monthly actual physiological sleep data, more timely intervention can be quickly achieved. Teixeira, president of the European Society of Sleep Technology, said: Multi-night data collection can help improve patient compliance, which will not only help patients get more comfortable and personalized treatment plans, but also help reverse patient non-compliance.
Longitudinal data can help doctors identify more subtle trend changes and respond quickly to provide more timely and targeted treatment options in a short period of time. Sleepimage’s CEO said that when doctors see an objective improvement in treatment based on the test results, coupled with patients’ supervisory feelings about improved quality of life and possibly reduced need for medication, perceptions of HST will become like those currently seen with hypertension or diabetes.
2. Remote Monitoring Platforms are Simplifying Sleep Management
Remote sleep monitoring uses remote network technology to quickly complete sleep data analysis to ensure the accuracy of remote sleep monitoring outside the sleep center. The American Academy of Sleep Medicine (AASM) first released guidelines on telemedicine in the diagnosis and treatment of sleep disorders in 2015, recommending well-equipped sleep centers to carry out sleep telemedicine, through this new model to improve sleep quality. Diagnosis and treatment efficiency and patient treatment compliance, so as to better improve patients’ daytime symptoms and quality of life, and reduce the resulting disease burden.
The growing use of HSAT or OCST provides the technical foundation for a remote sleep management platform, which is the best way to integrate medical systems, doctors, users and sleep service providers.
Most remote systems allow users to directly extract the raw data monitored by wearable sleep devices from the data cloud, and the user can superimpose multiple data into a single grid to achieve detailed data visualization at a single precise time, and utilize the system’s built-in A variety of data templates explore the correlation between parameters, and users can also adjust the time dimension to evaluate the treatment effect. These data will be automatically and quickly displayed in the data panel.
On the physician side, physicians can manage multiple tasks across multiple users through the platform, including centralizing schedules for home and institutional testing, generating CPAP sequences for patient profile information, viewing prescription updates, updating patient statistics, adding insurance information, and maintaining research pre-authorization details tracking and more.
Many remote monitoring companies are expanding the scope of data integration. Remote means more room for exploration and flexibility. Somnware plans to access data from consumer wearable devices by the end of 2023. This is another significant advantage of telehealth platforms in the field of sleep medicine. Physicians will not obtain lagging information and data, and the switching of equipment will no longer be a burden on both doctors and patients.
Viatom Sleep Monitoring Solution
The change in blood oxygen saturation during sleep is a necessary indicator for judging the time of hypoventilation. At present, the oximeter is no longer limited to simply analyzing the changes in blood oxygen parameters, but also overcomes the barrier of lack of sleep staging. Through the PPG of the blood oxygen probe, the blood oxygen fluctuation range, blood oxygen change frequency, heart rate variability and respiratory rate can be obtained at the same time. On the basis of recording blood oxygen fluctuations, it is possible to disassemble the data structure of blood oxygen saturation, such as trends, slopes of ascending branches, amplitudes, etc., to better correlate blood oxygen fluctuations, sympathetic tension, and respiratory events.
The development of wearable devices has promoted the transformation of OSA diagnosis and treatment methods. Wide application conditions, long-term use strategies and high comfort are more acceptable to users, and the diagnosis of OSA has been made lightweight and popular at a lower cost.
About Us
After years of independent innovation and relying on a high-performance processing platform, Viatom has continued to make breakthroughs in wearable pulse oxygen monitors. Independently developed anti-motion algorithm, adaptive dimming technology, finger ring design more suitable for continuous wear, and optimized compression algorithm make Viatom occupy a place in the global wearable oximeter market.
Reference：
1. Benjafield AV, Ayas NT, Eastwood PR, Heinzer R, Ip MSM, Morrell MJ, Nunez CM, Patel SR, Penzel T, Pépin JL, Peppard PE, Sinha S, Tufik S, Valentine K, Malhotra A. Estimation of the global prevalence and burden of obstructive sleep apnoea: a literature-based analysis. Lancet Respir Med. 2019 Aug;7(8):687-698. doi: 10.1016/S2213-2600(19)30198-5. Epub 2019 Jul 9. PMID: 31300334; PMCID: PMC7007763.
2. Olaf Oldenburg and others, Nocturnal hypoxaemia is associated with increased mortality in stable heart failure patients, European Heart Journal, Volume 37, Issue 21, 1 June 2016, Pages 1695–1703
3. Dai L, Cao W, Luo J, Huang R, Xiao Y. The effectiveness of sleep breathing impairment index in assessing obstructive sleep apnea severity. J Clin Sleep Med. 2023 Feb 1;19(2):267-274. doi: 10.5664/jcsm.10302. PMID: 36117435; PMCID: PMC9892730.
4. Manjunath S, Perea J A, Sathyanarayana A. Topological Data Analysis of Electroencephalogram Signals for Pediatric Obstructive Sleep Apnea[J]. arXiv preprint arXiv:2304.14853, 2023.
5. Brown J, Yazdi F, Jodari-Karimi M, Owen JG, Reisin E. Obstructive Sleep Apnea and Hypertension: Updates to a Critical Relationship. Curr Hypertens Rep. 2022 Jun;24(6):173-184. doi: 10.1007/s11906-022-01181-w. Epub 2022 Mar 5. PMID: 35246797; PMCID: PMC8897114.
6. Yeghiazarians Y, Jneid H, Tietjens JR, Redline S, Brown DL, El-Sherif N, Mehra R, Bozkurt B, Ndumele CE, Somers VK. Obstructive Sleep Apnea and Cardiovascular Disease: A Scientific Statement From the American Heart Association. Circulation. 2021 Jul 20;144(3):e56-e67. doi: 10.1161/CIR.0000000000000988. Epub 2021 Jun 21. Erratum in: Circulation. 2022 Mar 22;145(12):e775. PMID: 34148375.
7. Lechat B, Naik G, Reynolds A, Aishah A, Scott H, Loffler KA, Vakulin A, Escourrou P, McEvoy RD, Adams RJ, Catcheside PG, Eckert DJ. Multinight Prevalence, Variability, and Diagnostic Misclassification of Obstructive Sleep Apnea. Am J Respir Crit Care Med. 2022 Mar 1;205(5):563-569. doi: 10.1164/rccm.202107-1761OC. PMID: 34904935; PMCID: PMC8906484.
