A Novel Model to Estimate Key Obstructive Sleep Apnea Endotypes from Standard Polysomnography and Clinical Data and Their Contribution to Obstructive Sleep Apnea Severity

OSApredictor: A tool for prediction of moderate to severe obstructive sleep apnea-hypopnea using readily available patient characteristics

Sleep apnea is a common sleep disorder. The availability of an easy-to-use sleep apnea predictor would provide a public health benefit by promoting early diagnosis and treatment. Our goal was to develop a prediction tool that used commonly available variables and was accessible to the public through a web site. Using data from polysomnography (PSG) studies that measured the apnea-hypopnea index (AHI), we built a machine learning tool to predict the presence of moderate to severe obstructive sleep apnea (OSA) (defined as AHI ≥15). Our tool employs only seven widely available predictor variables: age, sex, weight, height, pulse oxygen saturation, heart rate and respiratory rate. As a preliminary step, we used 16,958 PSG studies to examine eight machine learning algorithms via five-fold cross validation and determined that XGBoost exhibited superior predictive performance. We then refined the XGBoost predictor by randomly partitioning the data into a training and a test set (13,566 and 3392 PSGs, respectively) and repeatedly subsampling from the training set to construct 1000 training subsets. We evaluated each of the resulting 1000 XGBoost models on the single set-aside test set. The resulting classification tool correctly identified 72.5 % of those with moderate to severe OSA as having the condition (sensitivity) and 62.8 % of those without moderate to-severe OSA as not having it (specificity); overall accuracy was 66 %. We developed a user-friendly publicly available website (https://manticore.niehs.nih.gov/OSApredictor). We hope that our easy-to-use tool will serve as a screening vehicle that enables more patients to be clinically diagnosed and treated for OSA.

Strengths, weaknesses, opportunities, and threats of using AI-enabled technology in sleep medicine: a commentary

Over the past few years, artificial intelligence (AI) has emerged as a powerful tool used to efficiently automate several tasks across multiple domains. Sleep medicine is perfectly positioned to leverage this tool due to the wealth of physiological signals obtained through sleep studies or sleep tracking devices and abundance of accessible clinical data through electronic medical records. However, caution must be applied when utilizing AI, due to intrinsic challenges associated with novel technology. The Artificial Intelligence in Sleep Medicine Committee of the American Academy of Sleep Medicine reviews advancements in AI within the sleep medicine field. In this article, the Artificial Intelligence in Sleep Medicine committee members provide a commentary on the scope of AI technology in sleep medicine. The commentary identifies 3 pivotal areas in sleep medicine that can benefit from AI technologies: clinical care, lifestyle management, and population health management. This article provides a detailed analysis of the strengths, weaknesses, opportunities, and threats associated with using AI-enabled technologies in each pivotal area. Finally, the article broadly reviews barriers and challenges associated with using AI-enabled technologies and offers possible solutions.

CITATION

Bandyopadhyay A, Oks M, Sun H, et al. Strengths, weaknesses, opportunities, and threats of using AI-enabled technology in sleep medicine: a commentary. J Clin Sleep Med. 2024;20(7):1183-1191.

Paediatric sleep diagnostics in the 21st century: the era of "sleep-omics"?

Paediatric sleep diagnostics is performed using complex multichannel tests in specialised centres, limiting access and availability and resulting in delayed diagnosis and management. Such investigations are often challenging due to patient size (prematurity), tolerability, and compliance with "gold standard" equipment. Children with sensory/behavioural issues, at increased risk of sleep disordered breathing (SDB), often find standard diagnostic equipment difficult.SDB can have implications for a child both in terms of physical health and neurocognitive development. Potential sequelae of untreated SDB includes failure to thrive, cardiopulmonary disease, impaired learning and behavioural issues. Prompt and accurate diagnosis of SDB is important to facilitate early intervention and improve outcomes.The current gold-standard diagnostic test for SDB is polysomnography (PSG), which is expensive, requiring the interpretation of a highly specialised physiologist. PSG is not feasible in low-income countries or outwith specialist sleep centres. During the coronavirus disease 2019 pandemic, efforts were made to improve remote monitoring and diagnostics in paediatric sleep medicine, resulting in a paradigm shift in SDB technology with a focus on automated diagnosis harnessing artificial intelligence (AI). AI enables interrogation of large datasets, setting the scene for an era of "sleep-omics", characterising the endotypic and phenotypic bedrock of SDB by drawing on genetic, lifestyle and demographic information. The National Institute for Health and Care Excellence recently announced a programme for the development of automated home-testing devices for SDB. Scorer-independent scalable diagnostic approaches for paediatric SDB have potential to improve diagnostic accuracy, accessibility and patient tolerability; reduce health inequalities; and yield downstream economic and environmental benefits.

Achieving Better Understanding of Obstructive Sleep Apnea Treatment Effects on Cardiovascular Disease Outcomes through Machine Learning Approaches: A Narrative Review

Obstructive sleep apnea (OSA) affects almost a billion people worldwide and is associated with a myriad of adverse health outcomes. Among the most prevalent and morbid are cardiovascular diseases (CVDs). Nonetheless, randomized controlled trials (RCTs) of OSA treatment have failed to show improvements in CVD outcomes. A major limitation in our field is the lack of precision in defining OSA and specifically subgroups with the potential to benefit from therapy. Further, this has called into question the validity of using the time-honored apnea-hypopnea index as the ultimate defining criteria for OSA. Recent applications of advanced statistical methods and machine learning have brought to light a variety of OSA endotypes and phenotypes. These methods also provide an opportunity to understand the interaction between OSA and comorbid diseases for better CVD risk stratification. Lastly, machine learning and specifically heterogeneous treatment effects modeling can help uncover subgroups with differential outcomes after treatment initiation. In an era of data sharing and big data, these techniques will be at the forefront of OSA research. Advanced data science methods, such as machine-learning analyses and artificial intelligence, will improve our ability to determine the unique influence of OSA on CVD outcomes and ultimately allow us to better determine precision medicine approaches in OSA patients for CVD risk reduction. In this narrative review, we will highlight how team science via machine learning and artificial intelligence applied to existing clinical data, polysomnography, proteomics, and imaging can do just that.

Obstructive sleep apnea heterogeneity and autonomic function: a role for heart rate variability in therapy selection and efficacy monitoring

Obstructive sleep apnea is a highly prevalent sleep-related breathing disorder, resulting in a disturbed breathing pattern, changes in blood gases, abnormal autonomic regulation, metabolic fluctuation, poor neurocognitive performance, and increased cardiovascular risk. With broad inter-individual differences recognised in risk factors, clinical symptoms, gene expression, physiological characteristics, and health outcomes, various obstructive sleep apnea subtypes have been identified. Therapeutic efficacy and its impact on outcomes, particularly for cardiovascular consequences, may also vary depending on these features in obstructive sleep apnea. A number of interventions such as positive airway pressure therapies, oral appliance, surgical treatment, and pharmaceutical options are available in clinical practice. Selecting an effective obstructive sleep apnea treatment and therapy is a challenging medical decision due to obstructive sleep apnea heterogeneity and numerous treatment modalities. Thus, an objective marker for clinical evaluation is warranted to estimate the treatment response in patients with obstructive sleep apnea. Currently, while the Apnea-Hypopnea Index is used for severity assessment of obstructive sleep apnea and still considered a major guide to diagnosis and managements of obstructive sleep apnea, the Apnea-Hypopnea Index is not a robust marker of symptoms, function, or outcome improvement. Abnormal cardiac autonomic modulation can provide additional insight to better understand obstructive sleep apnea phenotyping. Heart rate variability is a reliable neurocardiac tool to assess altered autonomic function and can also provide cardiovascular information in obstructive sleep apnea. Beyond the Apnea-Hypopnea Index, this review aims to discuss the role of heart rate variability as an indicator and predictor of therapeutic efficacy to different modalities in order to optimise tailored treatment for obstructive sleep apnea.

Multi-night measurement for diagnosis and simplified monitoring of obstructive sleep apnoea

Substantial night-to-night variability in obstructive sleep apnoea (OSA) severity has raised misdiagnosis and misdirected treatment concerns with the current prevailing single-night diagnostic approach. In-home, multi-night sleep monitoring technology may provide a feasible complimentary diagnostic pathway to improve both the speed and accuracy of OSA diagnosis and monitor treatment efficacy. This review describes the latest evidence on night-to-night variability in OSA severity, and its impact on OSA diagnostic misclassification. Emerging evidence for the potential impact of night-to-night variability in OSA severity to influence important health risk outcomes associated with OSA is considered. This review also characterises emerging diagnostic applications of wearable and non-wearable technologies that may provide an alternative, or complimentary, approach to traditional OSA diagnostic pathways. The required evidence to translate these devices into clinical care is also discussed. Appropriately sized randomised controlled trials are needed to determine the most appropriate and effective technologies for OSA diagnosis, as well as the optimal number of nights needed for accurate diagnosis and management. Potential risks versus benefits, patient perspectives, and cost-effectiveness of these novel approaches should be carefully considered in future trials.

Update on the genetic basis of obstructive sleep apnoea - hype or hope?

PURPOSE OF REVIEW

The obstructive sleep apnoea syndrome (OSAS) is a chronic, common condition in western societies which can lead to adverse cardiometabolic effects if left untreated and is one of the commonest causes of excessive daytime somnolence.

RECENT FINDINGS

The presentation of OSAS is diverse and is thought to comprise of different intermediate phenotypes and endotypes in varying proportions in each individual. Unfortunately, due to its heterogeneity and the changing definitions of the disorder by workers in the field, attempts at revealing the genetic basis of OSAS has been fraught with difficulty.

SUMMARY

This brief review presents a short update on the achievements of the past three decades in this understudied and underfunded area of endeavour in respiratory sleep medicine. The genetic underpinnings of OSAS remain elusive.

Sleep disorder and apnea events detection framework with high performance using two-tier learning model design

Sleep apnea is defined as a breathing disorder that affects sleep. Early detection of sleep apnea helps doctors to take intervention for patients to prevent sleep apnea. Manually making this determination is a time-consuming and subjectivity problem. Therefore, many different methods based on polysomnography (PSG) have been proposed and applied to detect this disorder. In this study, a unique two-layer method is proposed, in which there are four different deep learning models in the deep neural network (DNN), gated recurrent unit (GRU), recurrent neural network (RNN), RNN-based-long term short term memory (LSTM) architecture in the first layer, and a machine learning-based meta-learner (decision-layer) in the second layer. The strategy of making a preliminary decision in the first layer and verifying/correcting the results in the second layer is adopted. In the training of this architecture, a vector consisting of 23 features consisting of snore, oxygen saturation, arousal and sleep score data is used together with PSG data. A dataset consisting of 50 patients, both children and adults, is prepared. A number of pre-processing and under-sampling applications have been made to eliminate the problem of unbalanced classes. Proposed method has an accuracy of 95.74% and 99.4% in accuracy of apnea detection (apnea, hypopnea and normal) and apnea types detection (central, mixed and obstructive), respectively. Experimental results demonstrate that patient-independent consistent results can be produced with high accuracy. This robust model can be considered as a system that will help in the decisions of sleep clinics where it is expected to detect sleep disorders in detail with high performance.

Stepwise Add-On and Endotype-informed Targeted Combination Therapy to Treat Obstructive Sleep Apnea: A Proof-of-Concept Study

Rationale: Oral appliance therapy (OAT) is an effective treatment for many people with obstructive sleep apnea (OSA). However, OSA pathogenesis is heterogeneous, and, in ∼50% of cases, OAT does not fully control OSA. Objectives: This study aimed to control OSA in individuals with an incomplete response to OAT alone by using additional targeted therapies informed by OSA endotype characterization. Methods: Twenty-three people with OSA (apnea-hypopnea index [AHI], 41 ± 19 events/h) not fully resolved (AHI, >10 events/h) with OAT alone were prospectively recruited. OSA endotypes were characterized pretherapy during a detailed physiology study night. Initially, an expiratory positive airway pressure (EPAP) valve and supine avoidance device therapy were added to target the impaired anatomical endotype. Those with residual OSA (AHI, >10 events/h) then received one or more nonanatomical interventions based on endotype characterization. This included O2 (4 L/min) to reduce high loop gain (unstable respiratory control) and 80/5 mg atomoxetine-oxybutynin to increase pharyngeal muscle activity. Finally, if required, OAT was combined with EPAP and continuous positive airway pressure (CPAP) therapy. Results: Twenty participants completed the study. OSA was successfully controlled (AHI, <10 events/h) with combination therapy in all but one participant (17 of 20 without CPAP). OAT plus EPAP and supine avoidance therapy treated OSA in 10 (50%) participants. OSA was controlled in five (25%) participants with the addition of O2 therapy, one with atomoxetine-oxybutynin, and one required O2 plus atomoxetine-oxybutynin. Two participants required CPAP for their OSA, and another was CPAP intolerant. Conclusions: These novel prospective findings highlight the potential of precision medicine to inform targeted combination therapy to treat OSA. Clinical trial registered with the Australian New Zealand Clinical Trials Registry (ACTRN12618001995268).

Domiciliary transcutaneous electrical stimulation in patients with obstructive sleep apnoea and limited adherence to continuous positive airway pressure therapy: a single-centre, open-label, randomised, controlled phase III trial

Background

Hypoglossal nerve stimulation (HNS) for obstructive sleep apnoea (OSA) is a novel way to manage the condition. We hypothesised that in patients with OSA and limited adherence to continuous positive airway pressure (CPAP) therapy, domiciliary transcutaneous electrical stimulation (TESLA) would control sleep apnoea and provide health benefits.

Methods

We undertook a single-centre, open-label, randomised, controlled phase III trial in patients with OSA (apnoea-hypopnoea-index [AHI] 5-35 h-1), a BMI of 18.5-32 kg∗m-2, and a documented lack of adherence to CPAP therapy (<4 h∗night-1) at Guy's & St Thomas' NHS Foundation Trust (hospital), UK. Patients were randomly assigned (1:1) using minimisation (gender and OSA severity) to receive TESLA or usual care (CPAP) for at least 3 months; sleep study analysis was provided without knowledge of the assignment arm. The primary outcome was change in AHI at 3-months. The primary outcome and safety were analysed in the intention-to-treat population. Data are reported as median (interquartile range), unless otherwise explained. This trial is registered at ClinicalTrials.gov, NCT03160456.

Findings

Between 6 June 2018 and 7 February 2023, 56 participants were enrolled and randomly assigned (29 patients in the intervention group and 27 in the usual care group). Patients were followed up for a median of 3.0 months (IQR 3.0; 10.0). The groups were similar in terms of age (55.8 (48.2; 66.0) vs 59.3 (47.8; 64.4) years), gender (male:female, 19:10 vs 18:9) and BMI (28.7 (26.4; 31.9) vs 28.4 (24.4; 31.9) kg∗m-2). The unadjusted group difference in the Δ AHI was -11.5 (95% CI -20.7; -2.3) h-1 (p = 0.016). Adjusted for the baseline value, the difference was Δ AHI -7.0 (-15.7; 1.8) h-1 (p = 0.12), in favour of the intervention. Minor adverse events were found in one of the participants who developed mild headaches related to the intervention.

Interpretation

Domiciliary TESLA can be used safely and effectively in OSA patients with poor adherence to CPAP, with favourable impact on sleepiness and sleep fragmentation. Despite pandemic-related limitations of the amended protocol this trial provides the evidence that TESLA improves clinically meaningful outcomes over the observed follow up period, and the transcutaneous approach is likely to offer an affordable alternative for responders to electrical stimulation in clinical practice.

Funding

British Lung Foundation, United Kingdom Clinical Research Collaboration-registered King's Clinical Trials Unit at King's Health Partners.

Endotyping, phenotyping and personalised therapy in obstructive sleep apnoea: are we there yet?

Obstructive sleep apnoea (OSA) was traditionally thought to be mainly caused by obesity and upper airway crowding, and hence OSA management was not personalised according to particular characteristics, with most symptomatic patients receiving continuous positive airway pressure therapy. Recent advances in our understanding have identified additional potential and distinct causes of OSA (endotypes), and subgroups of patients (phenotypes) with increased risk of cardiovascular complications. In this review, we discuss the evidence to date as to whether there are distinct clinically useful endotypes and phenotypes of OSA, and the challenges to the field in moving towards delivering personalised therapy in OSA.

Clinical applications of artificial intelligence in sleep medicine: a sleep clinician's perspective

BACKGROUND

The past few years have seen a rapid emergence of artificial intelligence (AI)-enabled technology in the field of sleep medicine. AI refers to the capability of computer systems to perform tasks conventionally considered to require human intelligence, such as speech recognition, decision-making, and visual recognition of patterns and objects. The practice of sleep tracking and measuring physiological signals in sleep is widely practiced. Therefore, sleep monitoring in both the laboratory and ambulatory environments results in the accrual of massive amounts of data that uniquely positions the field of sleep medicine to gain from AI.

METHOD

The purpose of this article is to provide a concise overview of relevant terminology, definitions, and use cases of AI in sleep medicine. This was supplemented by a thorough review of relevant published literature.

RESULTS

Artificial intelligence has several applications in sleep medicine including sleep and respiratory event scoring in the sleep laboratory, diagnosing and managing sleep disorders, and population health. While still in its nascent stage, there are several challenges which preclude AI's generalizability and wide-reaching clinical applications. Overcoming these challenges will help integrate AI seamlessly within sleep medicine and augment clinical practice.

CONCLUSION

Artificial intelligence is a powerful tool in healthcare that may improve patient care, enhance diagnostic abilities, and augment the management of sleep disorders. However, there is a need to regulate and standardize existing machine learning algorithms prior to its inclusion in the sleep clinic.

The role of artificial intelligence in the treatment of obstructive sleep apnea

BACKGROUND

The first-line and most common treatment for obstructive sleep apnea is nasal continuous positive airway pressure, which serves as a pneumatic splint to stabilize the upper airway and is effective when used with appropriate adherence. Continuous positive airway pressure compliance rates remain significantly low despite machine improvements and compliance intervention. Other treatment options include oral appliances, myofunctional therapy, and surgery. The aim of this project is to elucidate the role of artificial intelligence within improving the treatment of obstructive sleep apnea.

METHODS

Related publications between 1999 and 2022 were reviewed from PubMed and Embase databases utilizing search terms "artificial intelligence," "machine learning," "obstructive sleep apnea," and "treatment." Both authors independently screened the results by title/abstract then by full text review. 126 non-duplicate articles were screened, 38 articles were included after title and abstract screen and 30 articles were included after full text review. The inclusion criteria are outline in the PICO framework and involved studies focused on artificial intelligence application in guiding and evaluating obstructive sleep apnea treatment. Non-English articles were excluded.

RESULTS

The role of artificial intelligence in the treatment of OSA was categorized into the following sections: Predicting treatment outcomes of various treatment options, Improving/Evaluating treatment, and Personalizing treatment with improving understanding of underlying mechanisms of OSA.

CONCLUSIONS

Artificial intelligence has the capacity to improve the treatment of OSA through predicting outcomes of treatment options, evaluating the treatment the patient is currently utilizing and increasing understanding of the mechanisms that contribute to OSA disease process and physiology. Implementing AI in guiding treatment decisions allows patients to connect with treatment methods that would be most effective on an individual basis.

Opioid Use Disorder, Sleep Deficiency, and Ventilatory Control: Bidirectional Mechanisms and Therapeutic Targets

Opioid use continues to rise globally. So too do the associated adverse consequences. Opioid use disorder (OUD) is a chronic and relapsing brain disease characterized by loss of control over opioid use and impairments in cognitive function, mood, pain perception, and autonomic activity. Sleep deficiency, a term that encompasses insufficient or disrupted sleep due to multiple potential causes, including sleep disorders, circadian disruption, and poor sleep quality or structure due to other medical conditions and pain, is present in 75% of patients with OUD. Sleep deficiency accompanies OUD across the spectrum of this addiction. The focus of this concise clinical review is to highlight the bidirectional mechanisms between OUD and sleep deficiency and the potential to target sleep deficiency with therapeutic interventions to promote long-term, healthy recovery among patients in OUD treatment. In addition, current knowledge on the effects of opioids on sleep quality, sleep architecture, sleep-disordered breathing, sleep apnea endotypes, ventilatory control, and implications for therapy and clinical practice are highlighted. Finally, an actionable research agenda is provided to evaluate the basic mechanisms of the relationship between sleep deficiency and OUD and the potential for behavioral, pharmacologic, and positive airway pressure treatments targeting sleep deficiency to improve OUD treatment outcomes.

Sleep Deficiency and Opioid Use Disorder: Trajectory, Mechanisms, and Interventions

Opioid use disorder (OUD) is a chronic and relapsing brain disease characterized by loss of control over opioid use and impairments in cognitive function, mood, pain perception, and autonomic activity. Sleep deficiency, a term that encompasses insufficient or disrupted sleep due to multiple potential causes, including sleep disorders (eg, insomnia, sleep apnea), circadian disruption (eg, delayed sleep phase and social jet lag), and poor sleep quality (eg, sleep fragmentation, impaired sleep architecture), is present in greater than 75% of patients with OUD. This article focuses on highlighting bidirectional mechanisms between OUD and sleep deficiency and points toward promising therapeutic targets.

Association of hypoglossal nerve stimulator response with machine learning identified negative effort dependence patterns

Background

Hypoglossal nerve stimulator (HGNS) is a therapeutic option for moderate to severe obstructive sleep apnea (OSA). Improved patient selection criteria are needed to target those most likely to benefit. We hypothesized that the pattern of negative effort dependence (NED) on inspiratory flow limited waveforms recorded during sleep, which has been correlated with the site of upper airway collapse, would contribute to the prediction of HGNS outcome. We developed a machine learning (ML) algorithm to identify NED patterns in pre-treatment sleep studies. We hypothesized that the predominant NED pattern would differ between HGNS responders and non-responders.

Methods

An ML algorithm to identify NED patterns on the inspiratory portion of the nasal pressure waveform was derived from 5 development set polysomnograms. The algorithm was applied to pre-treatment sleep studies of subjects who underwent HGNS implantation to determine the percentage of each NED pattern. HGNS response was defined by STAR trial criteria for success (apnea–hypopnea index (AHI) reduced by > 50% and < 20/h) as well as by a change in AHI and oxygenation metrics. The predominant NED pattern in HGNS responders and non-responders was determined. Other variables including demographics and oxygenation metrics were also assessed between responders and non-responders.

Results

Of 45 subjects, 4 were excluded due to technically inadequate polysomnograms. In the remaining 41 subjects, ML accurately distinguished three NED patterns (minimal, non-discontinuous, and discontinuous). The percentage of NED minimal breaths was significantly greater in responders compared with non-responders (p = 0.01) when the response was defined based on STAR trial criteria, change in AHI, and oxygenation metrics.

Conclusion

ML can accurately identify NED patterns in pre-treatment sleep studies. There was a statistically significant difference in the predominant NED pattern between HGNS responders and non-responders with a greater NED minimal pattern in responders. Prospective studies incorporating NED patterns into predictive modeling of factors determining HGNS outcomes are needed.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11325-022-02641-y.

Current and Novel Treatment Options for OSA

Obstructive sleep apnoea is a challenging medical problem due to its prevalence, its impact on quality of life and performance in school and professionally, the implications for risk of accidents, and comorbidities and mortality. Current research has carved out a broad spectrum of clinical phenotypes and defined major pathophysiological components. These findings point to the concept of personalised therapy, oriented on both the distinct clinical presentation and the most relevant pathophysiology in the individual patient. This leads to questions of whether sufficient therapeutic options other than positive airway pressure (PAP) alone are available, for which patients they may be useful, if there are specific indications for single or combined treatment, and whether there is solid scientific evidence for recommendations. This review describes our knowledge on PAP and non-PAP therapies to address upper airway collapsibility, muscle responsiveness, arousability and respiratory drive. The spectrum is broad and heterogeneous, including technical and pharmaceutical options already in clinical use or at an advanced experimental stage. Although there is an obvious need for more research on single or combined therapies, the available data demonstrate the variety of effective options, which should replace the unidirectional focus on PAP therapy.

The analysis of individual pathophysiological composition opens new directions towards personalised treatment of OSA, focusing not only on pharyngeal dilation, but also on technical or pharmaceutical interventions on muscle function or breathing regulationhttps://bit.ly/3sayhkd

Point-of-care prediction model of loop gain in patients with obstructive sleep apnea: development and validation

Background

High loop gain (unstable ventilatory control) is an important—but difficult to measure—contributor to obstructive sleep apnea (OSA) pathogenesis, predicting OSA sequelae and/or treatment response. Our objective was to develop and validate a clinical prediction tool of loop gain.

Methods

A retrospective cohort of consecutive adults with OSA (apnea–hypopnea index, AHI > 5/hour) based on in-laboratory polysomnography 01/2017–12/2018 was randomly split into a training and test-set (3:1-ratio). Using a customized algorithm (“reference standard”) loop gain was quantified from raw polysomnography signals on a continuous scale and additionally dichotomized (high > 0.7). Candidate predictors included general patient characteristics and routine polysomnography data. The model was developed (training-set) using linear regression with backward selection (tenfold cross-validated mean square errors); the predicted loop gain of the final linear regression model was used to predict loop gain class. More complex, alternative models including lasso regression or random forests were considered but did not meet pre-specified superiority-criteria. Final model performance was validated on the test-set.

Results

The total cohort included 1055 patients (33% high loop gain). Based on the final model, higher AHI (beta = 0.0016; P < .001) and lower hypopnea-percentage (beta = −0.0019; P < .001) predicted higher loop gain values. The predicted loop gain showed moderate-to-high correlation with the reference loop gain (r = 0.48; 95% CI 0.38–0.57) and moderate discrimination of patients with high versus low loop gain (area under the curve = 0.73; 95% CI 0.67–0.80).

Conclusion

To our knowledge this is the first prediction model of loop gain based on readily-available clinical data, which may facilitate retrospective analyses of existing datasets, better patient selection for clinical trials and eventually clinical practice.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12890-022-01950-y.

Development of a physiological-based model that uses standard polysomnography and clinical data to predict oral appliance treatment outcomes in obstructive sleep apnea

STUDY OBJECTIVES

Oral appliance (OA) therapy is a well-tolerated alternative to continuous positive airway pressure. However, it is less efficacious. A major unresolved clinical challenge is the inability to accurately predict who will respond to OA therapy. We recently developed a model to estimate obstructive sleep apnea pathophysiological endotypes. This study aimed to apply this physiological-based model to predict OA treatment responses.

METHODS

Sixty-two men and women with obstructive sleep apnea (aged 29-71 years) were studied to investigate the efficacy of a novel OA device. An in-laboratory diagnostic followed by an OA treatment efficacy polysomnography were performed. Seven polysomnography variables from the diagnostic study plus age and body mass index were included in our machine-learning-based model to predict OA therapy response according to standard apnea-hypopnea index (AHI) definitions. Initially, the model was trained on data from the first 45 participants using 10-fold cross-validation. A blinded independent validation was then performed for the remaining 17 participants.

RESULTS

Mean accuracy of the trained model to predict OA therapy responders vs nonresponders (AHI < 5 events/h) using 10-fold cross-validation was 91% ± 8%. In the independent blinded validation, 100% (AHI < 5 events/h); 59% (AHI < 10 events/h); 71% (50% reduction in AHI); and 82% (50% reduction in AHI to < 20 events/h) of the 17 participants were correctly classified for each of the treatment outcome definitions respectively.

CONCLUSIONS

While further evaluation in larger clinical data sets is required, these findings highlight the potential to use routinely collected sleep study and clinical data with machine learning-based approaches underpinned by obstructive sleep apnea endotype concepts to help predict treatment outcomes to OA therapy for people with obstructive sleep apnea.

CITATION

Dutta R, Tong BK, Eckert DJ. Development of a physiological-based model that uses standard polysomnography and clinical data to predict oral appliance treatment outcomes in obstructive sleep apnea. J Clin Sleep Med. 2022;18(3):861-870.

Impaired pharyngeal reflex responses to negative pressure: A novel cause of sleep apnea in multiple sclerosis

Obstructive sleep apnea (OSA) is common in people with multiple sclerosis (MS). However, people with MS often do not have 'typical' anatomical risk factors (i.e. non-obese and female predominance). Accordingly, non-anatomical factors such as impaired upper airway muscle function may be particularly important for OSA pathogenesis in MS. Therefore, this study aimed to investigate genioglossus (largest upper-airway dilator muscle) reflex responses to brief pulses of upper airway negative pressure in people with OSA and MS. 11 people with MS and OSA and 10 OSA controls without MS matched for age, sex and OSA severity were fitted with a nasal mask, pneumotachograph, choanal and epiglottic pressure sensors and intramuscular electrodes into genioglossus. Approximately 60 brief (250ms) negative pressure pulses (~-12cmH₂O mask pressure) were delivered every 2-6 breaths at random during quiet nasal breathing during wakefulness to determine genioglossus EMG reflex responses (timing, amplitude and morphology). Where available, recent clinical MRI brain scans were evaluated for the number, size and location of brainstem lesions in the MS group. When present, genioglossus reflex excitation responses were similar between MS participants and controls (e.g. peak excitation amplitude 229±85 vs. 282±98 % baseline, p=0.17). However, ~30% of people with MS had either an abnormal (predominantly inhibition) or no protective excitation reflex. Participants with MS without a reflex had multiple brainstem lesions including in the hypoglossal motor nucleus which may impair sensory processing and/or efferent output. Impaired pharyngeal reflex function may be an important contributor to OSA pathogenesis for a proportion of people with MS.

Defining obstructive sleep apnoea syndrome: a failure of semantic rules

Obstructive sleep apnoea syndrome (OSAS) is one of the most ubiquitous medical conditions in industrialised society. Since the recognition that symptoms of excessive daytime somnolence, problems with concentration, mood and cognitive impairment, as well as cardiometabolic abnormalities can arise as a consequence of obstructed breathing during sleep, it has been subject to variation in its definition. Over the past five decades, attempts have been made to standardise the definitions and scoring criteria used for apnoeas and hypopnoea, which are the hallmarks of obstructive sleep apnoea (OSA). However, applying these definitions in clinical and research practice has resulted in over- and under-estimation of the severity and prevalence of OSAS. Furthermore, the definitions may eventually become redundant in the context of rapid technological advances in breathing measurement and other signal acquisition. Increased efforts towards precision medicine have led to a focus on the pathophysiology of obstructed breathing during sleep. However, the same degree of effort has not been focused on how and why the latter does or does not result in diurnal symptoms, integral to the definition of OSAS. This review focuses on OSAS in adults and discusses some of the difficulties with current definitions and the possible reasons behind them.

The definition of obstructive sleep apnoea syndrome appears to be in constant flux dependent on the definitions attributed to its diagnostic componentshttps://bit.ly/3zXrWKg

Cardiopulmonary Sleep Spectrograms Open a Novel Window Into Sleep Biology-Implications for Health and Disease

The interactions of heart rate variability and respiratory rate and tidal volume fluctuations provide key information about normal and abnormal sleep. A set of metrics can be computed by analysis of coupling and coherence of these signals, cardiopulmonary coupling (CPC). There are several forms of CPC, which may provide information about normal sleep physiology, and pathological sleep states ranging from insomnia to sleep apnea and hypertension. As CPC may be computed from reduced or limited signals such as the electrocardiogram or photoplethysmogram (PPG) vs. full polysomnography, wide application including in wearable and non-contact devices is possible. When computed from PPG, which may be acquired from oximetry alone, an automated apnea hypopnea index derived from CPC-oximetry can be calculated. Sleep profiling using CPC demonstrates the impact of stable and unstable sleep on insomnia (exaggerated variability), hypertension (unstable sleep as risk factor), improved glucose handling (associated with stable sleep), drug effects (benzodiazepines increase sleep stability), sleep apnea phenotypes (obstructive vs. central sleep apnea), sleep fragmentations due to psychiatric disorders (increased unstable sleep in depression).

New and Emerging Approaches to Better Define Sleep Disruption and Its Consequences

Current approaches to quantify and diagnose sleep disorders and circadian rhythm disruption are imprecise, laborious, and often do not relate well to key clinical and health outcomes. Newer emerging approaches that aim to overcome the practical and technical constraints of current sleep metrics have considerable potential to better explain sleep disorder pathophysiology and thus to more precisely align diagnostic, treatment and management approaches to underlying pathology. These include more fine-grained and continuous EEG signal feature detection and novel oxygenation metrics to better encapsulate hypoxia duration, frequency, and magnitude readily possible via more advanced data acquisition and scoring algorithm approaches. Recent technological advances may also soon facilitate simple assessment of circadian rhythm physiology at home to enable sleep disorder diagnostics even for “non-circadian rhythm” sleep disorders, such as chronic insomnia and sleep apnea, which in many cases also include a circadian disruption component. Bringing these novel approaches into the clinic and the home settings should be a priority for the field. Modern sleep tracking technology can also further facilitate the transition of sleep diagnostics from the laboratory to the home, where environmental factors such as noise and light could usefully inform clinical decision-making. The “endpoint” of these new and emerging assessments will be better targeted therapies that directly address underlying sleep disorder pathophysiology via an individualized, precision medicine approach. This review outlines the current state-of-the-art in sleep and circadian monitoring and diagnostics and covers several new and emerging approaches to better define sleep disruption and its consequences.

The noradrenergic agent reboxetine plus the antimuscarinic hyoscine butylbromide reduces sleep apnoea severity: a double-blind, placebo-controlled, randomised crossover trial

KEY POINTS

Recent animal and human physiology studies indicate that noradrenergic and muscarinic processes are key mechanisms that mediate pharyngeal muscle control during sleep. The noradrenergic agent reboxetine combined with the anti-muscarinic hyoscine butylbromide has recently been shown to improve upper airway function during sleep in healthy individuals. However, whether these findings translate to the clinically relevant patient population of people with obstructive sleep apnoea (OSA), and the effects of the agents on OSA severity, are unknown. We found that reboxetine plus hyoscine butylbromide reduced OSA severity, including overnight hypoxaemia, via increases in pharyngeal muscle responsiveness, improvements in respiratory control and airway collapsibility without changing the respiratory arousal threshold. These findings provide mechanistic insight into the role of noradrenergic and anti-muscarinic agents on upper airway stability and breathing during sleep and are important for pharmacotherapy development for OSA.

ABSTRACT

The noradrenergic agent reboxetine combined with the anti-muscarinic hyoscine butylbromide has recently been shown to improve upper airway function during sleep in healthy individuals. However, the effects of this drug combination on obstructive sleep apnoea (OSA) severity are unknown. Accordingly, this study aimed to determine if reboxetine plus hyoscine butylbromide reduces OSA severity. Secondary aims were to investigate the effects on key upper airway physiology and endotypic traits. Twelve people with OSA aged 52 ± 13 years, BMI = 30 ± 5 kg/m² , completed a double-blind, randomised, placebo-controlled, crossover trial (ACTRN12617001326381). Two in-laboratory sleep studies with nasal mask, pneumotachograph, epiglottic pressure sensor and bipolar fine-wire electrodes into genioglossus and tensor palatini muscles were performed separated by approximately 1 week. Each participant received either reboxetine (4 mg) plus hyoscine butylbromide (20 mg), or placebo immediately prior to sleep. Polysomnography, upper airway physiology and endotypic estimates of OSA were compared between conditions. Reboxetine plus hyoscine butylbromide reduced the apnoea/hypopnoea index by (mean ± SD) 17 ± 17 events/h from 51 ± 30 to 33 ± 22 events/h (P = 0.005) and nadir oxygen saturation increased by 6 ± 5% from 82 ± 5 to 88 ± 2% (P = 0.002). The drug combination increased tonic genioglossus muscle responsiveness during non-REM sleep (median [25th, 75th centiles]: -0.007 [-0.0004, -0.07] vs. -0.12 [-0.02, -0.40] %maxEMG/cmH₂ O, P = 0.02), lowered loop gain (0.43 ± 0.06 vs. 0.39 ± 0.07, P = 0.01), and improved airway collapsibility (90 [69, 95] vs. 93 [88, 96] %eupnoea, P = 0.02), without changing the arousal threshold (P = 0.39). These findings highlight the important role that noradrenergic and muscarinic processes have on upper airway function during sleep and the potential for pharmacotherapy to target these mechanisms to treat OSA.

Noninvasive electrical stimulation of oropharyngeal muscles in obstructive sleep apnea

Introduction: Continuous positive airway pressure (CPAP) therapy remains the standard treatment for obstructive sleep apnea. However, its proven effect is useless if the patient does not tolerate the treatment. The electrical stimulation approach has been investigated for several decades now and it seems that the implantable devices for invasive electrical stimulation of hypoglossal nerve are viewed as effective with some of them already approved for human use.Areas covered: in this review, we intent to summarize the existing records of noninvasive stimulation in sleep apnea to make the scientific community aware of the details before deciding on its future. We believe that this is a battle still to fight and more could be done bearing in mind the safety of this method.Expertopinion: noninvasive electrical stimulation has been left behind based on few, small and inconsistent studies using different stimulation parameters. These studies are difficult to compare and to draw conclusions.Electrical stimulation is a field for research in the treatment of obstructive sleep apnea, with many aspects still to be discovered, and which may become a therapeutic alternative to the use of CPAP in certain patients.

Effects of hypnotics on obstructive sleep apnea endotypes and severity: Novel insights into pathophysiology and treatment

Impaired upper airway anatomy is the main cause of obstructive sleep apnea (OSA). However, there are other important non-anatomical contributors or "endotypes" including ventilatory control instability, poor pharyngeal dilator muscle responsiveness and waking up too easily to minor respiratory events (low arousal threshold). Recent studies have focused on the potential to target specific OSA causes with novel treatments to reduce OSA severity and improve efficacy with existing non-CPAP therapies which are often suboptimal (e.g., mandibular advancement splints). One novel target is pharmacotherapy with hypnotics to increase the threshold for arousal and reduce OSA severity in the approximately 30% of patients who have a low arousal threshold endotype. This increasing body of work has produced varied and at times unexpected findings which have challenged previous knowledge on the effects of hypnotics on upper airway physiology and breathing during sleep in people with OSA. This review provides a concise overview of the latest research that has investigated the effects of common hypnotics/sedative agents on upper airway physiology and OSA severity and potential implications for OSA pathophysiology, treatment and safety. This includes a summary of the latest knowledge on the effects of hypnotics on OSA endotypes. Priorities for future research are also highlighted.

Research routes on improved sleep bruxism metrics: Toward a standardised approach

A recent report from the European Sleep Research Society's task force "Beyond AHI" discussed an issue that has been a long-term subject of debate - what are the best metrics for obstructive sleep apnoea (OSA) diagnosis and treatment outcome assessments? In a similar way, sleep bruxism (SB) metrics have also been a recurrent issue for >30 years and there is still uncertainty in dentistry regarding their optimisation and clinical relevance. SB can occur alone or with comorbidities such as OSA, gastroesophageal reflux disorder, insomnia, headache, orofacial pain, periodic limb movement, rapid eye movement behaviour disorder, and sleep epilepsy. Classically, the diagnosis of SB is based on the patient's dental and medical history and clinical manifestations; electromyography is used in research and for complex cases. The emergence of new technologies, such as sensors and artificial intelligence, has opened new opportunities. The main objective of the present review is to stimulate the creation of a collaborative taskforce on SB metrics. Several examples are available in sleep medicine. The development of more homogenised metrics could improve the accuracy and refinement of SB assessment, while moving forward toward a personalised approach. It is time to develop SB metrics that are relevant to clinical outcomes and benefit patients who suffer from one or more possible negative consequences of SB.