Agreement in the scoring of respiratory events and sleep among international sleep centers

Temporal and sleep stage-dependent agreement in manual scoring of respiratory events

Obstructive sleep apnea diagnosis is based on the manual scoring of respiratory events. The agreement in the manual scoring of the respiratory events lacks an in-depth investigation as most of the previous studies reported only the apnea-hypopnea index or overall agreement, and not temporal, second-by-second or event subtype agreement. We hypothesized the temporal and subtype agreement to be low because the event duration or subtypes are not generally considered in current clinical practice. The data comprised 50 polysomnography recordings scored by 10 experts. The respiratory event agreement between the scorers was calculated using kappa statistics in a second-by-second manner. Obstructive sleep apnea severity categories (no obstructive sleep apnea/mild/moderate/severe) were compared between scorers. The Fleiss' kappa value for binary (event/no event) respiratory event scorings was 0.32. When calculated separately within N1, N2, N3 and R, the Fleiss' kappa values were 0.12, 0.23, 0.22 and 0.23, respectively. Binary analysis conducted separately for the event subtypes showed the highest Fleiss' kappa for hypopneas to be 0.26. In 34% of the participants, the obstructive sleep apnea severity category was the same regardless of the scorer, whereas in the rest of the participants the category changed depending on the scorer. Our findings indicate that the agreement of manual scoring of respiratory events depends on the event type and sleep stage. The manual scoring has discrepancies, and these differences affect the obstructive sleep apnea diagnosis. This is an alarming finding, as ultimately these differences in the scorings affect treatment decisions.

Multidiagnostic chest-worn patch to detect obstructive sleep apnea and cardiac arrhythmias

STUDY OBJECTIVES

Evaluate the performance of the SANSA device to simultaneously assess obstructive sleep apnea and cardiac arrhythmias.

METHODS

Participants suspected or known to have obstructive sleep apnea underwent polysomnography while wearing SANSA. SANSA's algorithm was trained using 86 records and tested on 67 to evaluate training bias. SANSA performance was evaluated against ground truth polysomnography scored by the consensus of 3 technologists. Polysomnography scoring from individual testing sites was also evaluated against consensus. Diagnostic performance was evaluated using standard apnea-hypopnea index cutoffs. Apnea-hypopnea index and total sleep time agreement was analyzed using correlation and Bland-Altman plots. Electrocardiogram was reviewed for presence of significant arrhythmias (frequent premature atrial/ventricular complexes and atrial fibrillation).

RESULTS

SANSA's sensitivity and specificity to detect obstructive sleep apnea ranged from 91-97% and 78-97%, respectively, across all severity levels. SANSA total sleep time correlation with consensus polysomnography total sleep time was 0.83 with a mean difference of 3.8 minutes (limits of agreement: -91.1 to 98.7). Significant arrhythmias were detected in 32% of participants. These participants had a greater apnea-hypopnea index (27.5 vs 15.8 events/h, P = .003) and spent nearly twice as long at reduced oxygenation levels (47.5 vs 20.5 minutes under 88% oxygen saturation, P = .009).

CONCLUSIONS

SANSA is a promising tool for comprehensive obstructive sleep apnea evaluation, offering the unique advantage of concurrent arrhythmia detection. This dual functionality may improve patient outcomes through early diagnosis and management of both conditions.

CITATION

Goldstein C, Ghanbari H, Sharma S, et al. Multidiagnostic chest-worn patch to detect obstructive sleep apnea and cardiac arrhythmias. J Clin Sleep Med. 2025;21(5):855-866.

Detecting arousals and sleep from respiratory inductance plethysmography

PURPOSE

Accurately identifying sleep states (REM, NREM, and Wake) and brief awakenings (arousals) is essential for diagnosing sleep disorders. Polysomnography (PSG) is the gold standard for such assessments but is costly and requires overnight monitoring in a lab. Home sleep testing (HST) offers a more accessible alternative, relying primarily on breathing measurements but lacks electroencephalography, limiting its ability to evaluate sleep and arousals directly. This study evaluates a deep learning algorithm which determines sleep states and arousals from breathing signals.

METHODS

A novel deep learning algorithm was developed to classify sleep states and detect arousals from respiratory inductance plethysmography signals. Sleep states were predicted for 30-s intervals (one sleep epoch), while arousal probabilities were calculated at 1-s resolution. Validation was conducted on a clinical dataset of 1,299 adults with suspected sleep disorders. Performance was assessed at the epoch level for sensitivity and specificity, with agreement analyses for arousal index (ArI) and total sleep time (TST).

RESULTS

The algorithm achieved sensitivity and specificity of 77.9% and 96.2% for Wake, 93.9% and 80.4% for NREM, 80.5% and 98.2% for REM, and 66.1% and 86.7% for arousals. Bland-Altman analysis showed ArI limits of agreement ranging from - 32 to 24 events/hour (bias: - 4.4) and TST limits from - 47 to 64 min (bias: 8.0). Intraclass correlation was 0.74 for ArI and 0.91 for TST.

CONCLUSION

The algorithm identifies sleep states and arousals from breathing signals with agreement comparable to established variability in manual scoring. These results highlight its potential to advance HST capabilities, enabling more accessible, cost-effective and reliable sleep diagnostics.

Continuous sleep depth index annotation with deep learning yields novel digital biomarkers for sleep health

Traditional sleep staging categorizes sleep and wakefulness into five coarse-grained classes, overlooking subtle variations within each stage. We propose a deep learning method to annotate continuous sleep depth index (SDI) with existing discrete sleep staging labels, using polysomnography from over 10,000 recordings across four large-scale cohorts. The results showcased a strong correlation between the decrease in sleep depth index and the increase in duration of arousal. Case studies indicated that SDI captured more nuanced sleep structures than conventional sleep staging. Clustering based on the digital biomarkers extracted from the SDI identified two subtypes of sleep, where participants in the disturbed subtype had a higher prevalence of several poor health conditions and were associated with a 33% increased risk of mortality and a 38% increased risk of fatal coronary heart disease. Our study underscores the utility of SDI in revealing more detailed sleep structures and yielding novel digital biomarkers for sleep medicine.

Arousal burden is highest in supine sleeping position and during light sleep

STUDY OBJECTIVES

Arousal burden (AB) is defined as the cumulative duration of arousals during sleep divided by the total sleep time. However, in-depth analysis of AB related to sleep characteristics is lacking. Based on previous studies addressing the arousal index, we hypothesized that the AB would peak in the supine sleeping position and during non-rapid eye movement stage 1 and show high variability between scorers.

METHODS

Nine expert scorers analyzed polysomnography recordings of 50 participants, the majority with an increased risk for obstructive sleep apnea. AB was calculated in different sleeping positions and sleep stages. A generalized estimating equation was used to test the association between AB and sleeping positions, sleep stages, and scorers. The correlation between AB and arousal index was tested with Spearman's rank-order correlation.

RESULTS

AB significantly differed between sleeping positions (P < .001). The median AB in the supine sleeping position was 47-62% higher than in the left and right positions. The AB significantly differed between the sleep stages (P < .001); the median AB was more than 200% higher during non-rapid eye movement stage 1 than during other sleep stages. In addition, the AB differed significantly between scorers (P < .001) and correlated strongly with arousal index (r = .935, P < .001).

CONCLUSIONS

AB depends on the sleeping position, sleep stage, and scorer, as hypothesized. AB behaved similarly to the arousal index, but the high variability in the ABs between scorers indicates a potential limitation caused by subjective manual scoring. Thus, the development of more accurate techniques for scoring arousals is required before AB can be reliably used.

CITATION

Pitkänen M, Huovinen J, Rissanen M, et al. Arousal burden is highest in supine sleeping position and during light sleep. J Clin Sleep Med. 2025;21(2):337-344.

What radio waves tell us about sleep!

The ability to assess sleep at home, capture sleep stages, and detect the occurrence of apnea (without on-body sensors) simply by analyzing the radio waves bouncing off people's bodies while they sleep is quite powerful. Such a capability would allow for longitudinal data collection in patients' homes, informing our understanding of sleep and its interaction with various diseases and their therapeutic responses, both in clinical trials and routine care. In this article, we develop an advanced machine-learning algorithm for passively monitoring sleep and nocturnal breathing from radio waves reflected off people while asleep. Validation results in comparison with the gold standard (i.e. polysomnography; n = 880) demonstrate that the model captures the sleep hypnogram (with an accuracy of 80.5% for 30-second epochs categorized into wake, light sleep, deep sleep, or REM), detects sleep apnea (AUROC = 0.89), and measures the patient's Apnea-Hypopnea Index (ICC = 0.90; 95% CI = [0.88, 0.91]). Notably, the model exhibits equitable performance across race, sex, and age. Moreover, the model uncovers informative interactions between sleep stages and a range of diseases including neurological, psychiatric, cardiovascular, and immunological disorders. These findings not only hold promise for clinical practice and interventional trials but also underscore the significance of sleep as a fundamental component in understanding and managing various diseases.

Are scoring respiratory effort-related arousals worth the effort? --A study comparing outcomes between 4 % vs 3 % hypopnea scoring rules

STUDY OBJECTIVES

The respiratory effort-related arousal (RERA) has been combined with apneas and hypopneas into the respiratory disturbance index (RDI). RERAs are characterized by ≥ 10 s of increasing upper airway effort terminating in arousal without meeting hypopnea criteria. The recent change to hypopnea definitions now includes a ≥30 % reduction in airflow for 10 s with EITHER a 3 % oxygen desaturation OR an arousal. Consequently, many events previously categorized as RERAs will now be included in the 3 % hypopneas, likely reducing the number of events scored as RERAs. We hypothesized that the 3 % apnea-hypopnea index (3%AHI) would approximate the 4%RDI, with the number of 3 % RERAs being negligible.

RESEARCH QUESTION

How does the transition from the 4 % to the 3 % hypopnea rules impact the significance of RERAs in clinical practice, and how we should relate the AHI and RDI using the different hypopnea rules?

METHODS

We prospectively collected 76 consecutive polysomnography results in 4 adult age groups. We re-scored the respiratory events utilizing both the 3 % and the 4 % hypopnea rules and compared the outcomes.

RESULTS

Among 76 diagnostic studies (mean age 47.5 years, males 47.4 %), the 3 % RERA index [0.8 (0.0, 3.1)] [median (Q1, Q3)] was significantly lower than the 4 % RERA index [3.5 (1.0, 7.3)]. The 3%AHI was 3.07 ± 9.23 (mean ± SD) higher than the 4%RDI (p = 0.005). The 3%AHI was 8.63 ± 8.86 higher than the 4%AHI in all age groups (p < 0.001). This was mainly due to an increased hypopnea index (+8.51 ± 9.03, p < 0.001). In patients with obstructive sleep apnea (OSA), the 3%RERA contributes 4.3 % to the 3%RDI, while the 4%RERA contributes 27.7 % to the 4%RDI.

INTERPRETATIONS

Both 3%RDI and 3%AHI are higher than the 4%RDI, primarily due to identification of more hypopnea events, resulting in more patients being classified as having OSA. This change in criteria complicates the comparison of hypopnea and RERA contributions between sleep studies scored using the different hypopnea rules.

Comparison of automated deep neural network against manual sleep stage scoring in clinical data

OBJECTIVE

To compare the accuracy and generalizability of an automated deep neural network and the Philip Sleepware G3™ Somnolyzer system (Somnolyzer) for sleep stage scoring using American Academy of Sleep Medicine (AASM) guidelines.

METHODS

Sleep recordings from 104 participants were analyzed by a convolutional neural network (CNN), the Somnolyzer and skillful technicians. Evaluation metrics were derived for different combinations of sleep stages. A further comparison between the Somnolyzer and the CNN model using a single-channel signal as input was also performed. Sleep recordings from 263 participants with a lower prevalence of OSA served as a cross-validation dataset to validate the generalizability of the CNN model.

RESULTS

The overall agreement between automated and manual scoring for sleep staging in 104 participants outperformed that of the Somnolyzer according to various metrics (accuracy: 81.81 % vs. 77.07 %; F1: 76.36 % vs. 73.80 %; Cohen's kappa: 0.7403 vs. 0.6848). The results showed that the left electrooculography (EOG) single-channel model had minor advantages over the Somnolyzer. In terms of consistency with manual sleep staging, the CNN model demonstrated superior performance in identifying more pronounced sleep transitions, particularly in the N2 stage and sleep latency metrics. Conversely, the Somnolyzer showed enhanced proficiency in the analysis of REM stages, notably in measuring REM latency. The accuracy in the cross-validation set of 263 participants was also above 80 %.

CONCLUSIONS

The CNN-based automated deep neural network outperformed the Somnolyzer and is sufficiently accurate for sleep study analyses using the AASM classification criteria.

Multi-centre arousal scoring agreement in the Sleep Revolution

We investigated arousal scoring agreement within full-night polysomnography in a multi-centre setting. Ten expert scorers from seven centres annotated 50 polysomnograms using the American Academy of Sleep Medicine guidelines. The agreement between arousal indexes (ArIs) was investigated using intraclass correlation coefficients (ICCs). Moreover, kappa statistics were used to evaluate the second-by-second agreement in whole recordings and in different sleep stages. Finally, arousal clusters, that is, periods with overlapping arousals by multiple scorers, were extracted. The overall similarity of the ArIs was fair (ICC = 0.41), varying from poor to excellent between the scorer pairs (ICC = 0.04-0.88). The ArI similarity was better in respiratory (ICC = 0.65) compared with spontaneous (ICC = 0.23) arousals. The overall second-by-second agreement was fair (Fleiss' kappa = 0.40), varying from poor to substantial depending on the scorer pair (Cohen's kappa = 0.07-0.68). Fleiss' kappa increased from light to deep sleep (0.45, 0.45, and 0.53 for stages N1, N2, and N3, respectively), was moderate in the rapid eye movement stage (0.48), and the lowest in the wake stage (0.25). Over a half of the arousal clusters were scored by one or two scorers, and less than a third by at least five scorers. In conclusion, the scoring agreement varied depending on the arousal type, sleep stage, and scorer pair, but was overall relatively low. The most uncertain areas were related to spontaneous arousals and arousals scored in the wake stage. These results indicate that manual arousal scoring is generally not reliable, and that changes are needed in the assessment of sleep fragmentation for clinical and research purposes.

State-of-the-art sleep arousal detection evaluated on a comprehensive clinical dataset

Aiming to apply automatic arousal detection to support sleep laboratories, we evaluated an optimized, state-of-the-art approach using data from daily work in our university hospital sleep laboratory. Therefore, a machine learning algorithm was trained and evaluated on 3423 polysomnograms of people with various sleep disorders. The model architecture is a U-net that accepts 50 Hz signals as input. We compared this algorithm with models trained on publicly available datasets, and evaluated these models using our clinical dataset, particularly with regard to the effects of different sleep disorders. In an effort to evaluate clinical relevance, we designed a metric based on the error of the predicted arousal index. Our models achieve an area under the precision recall curve (AUPRC) of up to 0.83 and F1 scores of up to 0.81. The model trained on our data showed no age or gender bias and no significant negative effect regarding sleep disorders on model performance compared to healthy sleep. In contrast, models trained on public datasets showed a small to moderate negative effect (calculated using Cohen's d) of sleep disorders on model performance. Therefore, we conclude that state-of-the-art arousal detection on our clinical data is possible with our model architecture. Thus, our results support the general recommendation to use a clinical dataset for training if the model is to be applied to clinical data.

Definition of diaphragmatic sleep disordered breathing and clinical meaning in Duchenne muscular dystrophy

BACKGROUND

Diaphragmatic sleep disordered breathing (dSDB) has been recently identified as sleep dysfunction secondary to diaphragmatic weakness in Duchenne muscular dystrophy (DMD). However, scoring criteria for the identification of dSDB are missing.This study aimed to define and validate dSDB scoring criteria and to evaluate whether dSDB severity correlates with respiratory progression in DMD.

METHODS

Scoring criteria for diaphragmatic apnoea (dA) and hypopnoeas (dH) have been defined by the authors considering the pattern observed on cardiorespiratory polygraphy (CR) and the dSDB pathophysiology.10 sleep professionals (physiologists, consultants) blinded to each other were involved in a two-round Delphi survey to rate each item of the proposed dSDB criteria (Likert scale 1-5) and to recognise dSDB among other SDB. The scorers' accuracy was tested against the authors' panel.Finally, CR previously conducted in DMD in clinical setting were rescored and diaphragmatic Apnoea-Hypopnoea Index (dAHI) was derived. Pulmonary function (forced vital capacity per cent of predicted, FVC%pred), overnight oxygen saturation (SpO2) and transcutaneous carbon dioxide (tcCO2) were correlated with dAHI.

RESULTS

After the second round of Delphi, raters deemed each item of dA and dH criteria as relevant as 4 or 5. The agreement with the panel in recognising dSDB was 81%, kappa 0.71, sensitivity 77% and specificity 85%.32 CRs from DMD patients were reviewed. dSDB was previously scored as obstructive. The dAHI negatively correlated with FVC%pred (r=-0.4; p<0.05). The total number of dA correlated with mean overnight tcCO2 (r 0.4; p<0.05).

CONCLUSIONS

dSDB is a newly defined sleep disorder that correlates with DMD progression. A prospective study to evaluate dSDB as a respiratory measure for DMD in clinical and research settings is planned.

Characterisation of Symptom and Polysomnographic Profiles Associated with Cardiovascular Risk in a Sleep Clinic Population with Obstructive Sleep Apnoea

Aim

Recent data have identified specific symptom and polysomnographic profiles associated with cardiovascular disease (CVD) in patients with obstructive sleep apnoea (OSA). Our aim was to determine whether these profiles were present at diagnosis of OSA in patients with established CVD and in those with high cardiovascular risk. Participants in the Sydney Sleep Biobank (SSB) database, aged 30-74 years, self-reported presence of CVD (coronary artery disease, cerebrovascular disease, or heart failure). In those without established CVD, the Framingham Risk Score (FRS) estimated 10-year absolute CVD risk, categorised as "low" (<6%), "intermediate" (6-20%), or "high" (>20%). Groups were compared on symptom and polysomnographic variables.

Results

629 patients (68% male; mean age 54.3 years, SD 11.6; mean BMI 32.3 kg/m2, SD 8.2) were included. CVD was reported in 12.2%. A further 14.3% had a low risk FRS, 38.8% had an intermediate risk FRS, and 34.7% had a high risk FRS. Groups differed with respect to age, sex and BMI. OSA severity increased with established CVD and increasing FRS. The symptom of waking too early was more prevalent in the higher FRS groups (p=0.004). CVD and FRS groups differed on multiple polysomnographic variables; however, none of these differences remained significant after adjusting for age, sex, and BMI.

Conclusion

Higher CVD risk was associated with waking too early in patients with OSA. Polysomnographic variations between groups were explained by demographic differences. Further work is required to explore the influence of OSA phenotypic characteristics on susceptibility to CVD.

Polysomnographic characteristics of excessive daytime sleepiness phenotypes in obstructive sleep apnea: results from the international sleep apnea global interdisciplinary consortium

STUDY OBJECTIVES

Excessive daytime sleepiness (EDS) is a major symptom of obstructive sleep apnea (OSA). Traditional polysomnographic (PSG) measures only partially explain EDS in OSA. This study analyzed traditional and novel PSG characteristics of two different measures of EDS among patients with OSA.

METHODS

Sleepiness was assessed using the Epworth Sleepiness Scale (>10 points defined as "risk of dozing") and a measure of general sleepiness (feeling sleepy ≥ 3 times/week defined as "feeling sleepy"). Four sleepiness phenotypes were identified: "non-sleepy," "risk of dozing only," "feeling sleepy only," and "both at risk of dozing and feeling sleepy."

RESULTS

Altogether, 2083 patients with OSA (69% male) with an apnea-hypopnea index (AHI) ≥ 5 events/hour were studied; 46% were "non-sleepy," 26% at "risk of dozing only," 7% were "feeling sleepy only," and 21% reported both. The two phenotypes at "risk of dozing" had higher AHI, more severe hypoxemia (as measured by oxygen desaturation index, minimum and average oxygen saturation [SpO2], time spent < 90% SpO2, and hypoxic impacts) and they spent less time awake, had shorter sleep latency, and higher heart rate response to arousals than "non-sleepy" and "feeling sleepy only" phenotypes. While statistically significant, effect sizes were small. Sleep stages, frequency of arousals, wake after sleep onset and limb movement did not differ between sleepiness phenotypes after adjusting for confounders.

CONCLUSIONS

In a large international group of patients with OSA, PSG characteristics were weakly associated with EDS. The physiological measures differed among individuals characterized as "risk of dozing" or "non-sleepy," while "feeling sleepy only" did not differ from "non-sleepy" individuals.

Can intraoral and facial photos predict obstructive sleep apnea in the general and clinical population?

STUDY OBJECTIVES

This study aimed to evaluate and compare measurements of standardized craniofacial and intraoral photographs between clinical and general population samples, between groups of individuals with an apnea-hypopnea index (AHI) ≥ 15 and AHI < 15, and their interaction, as well as the relationship with the presence and severity of obstructive sleep apnea (OSA).

METHODS

We used data from 929 participants from Sleep Apnea Global Interdisciplinary Consortium, in which 309 patients from a clinical setting and 620 volunteers from a general population.

RESULTS

AHI ≥ 15 were observed in 30.3% of the total sample and there were some interactions between facial/intraoral measures with OSA and both samples. Mandibular volume (p < 0.01) and lateral face height (p = 0.04) were higher in the AHI ≥ 15 group in the clinical sample compared to the AHI ≥ 15 group in the general population and AHI < 15 group in the clinical sample. When adjusted for sex and age, greater mandible width (p < 0.01) differed both in the clinical and in the general population samples, reflecting AHI severity and the likelihood of OSA. The measure of smaller tongue curvature (p < 0.01) reflected the severity and probability of OSA in the clinical sample and the higher posterior mandibular height (p = 0.04) showed a relationship with higher AHI and higher risk of OSA in the general population. When adjusted for sex, age, and body mass index, only smaller tongue curvature (p < 0.01) was associated with moderate/severe OSA.

CONCLUSIONS

Measures of greater tongue and mandible were associated with increased OSA risk in the clinical sample and craniofacial measurement was associated in the general population sample.

Review and perspective on sleep-disordered breathing research and translation to clinics

Sleep-disordered breathing, ranging from habitual snoring to severe obstructive sleep apnea, is a prevalent public health issue. Despite rising interest in sleep and awareness of sleep disorders, sleep research and diagnostic practices still rely on outdated metrics and laborious methods reducing the diagnostic capacity and preventing timely diagnosis and treatment. Consequently, a significant portion of individuals affected by sleep-disordered breathing remain undiagnosed or are misdiagnosed. Taking advantage of state-of-the-art scientific, technological, and computational advances could be an effective way to optimize the diagnostic and treatment pathways. We discuss state-of-the-art multidisciplinary research, review the shortcomings in the current practices of SDB diagnosis and management in adult populations, and provide possible future directions. We critically review the opportunities for modern data analysis methods and machine learning to combine multimodal information, provide a perspective on the pitfalls of big data analysis, and discuss approaches for developing analysis strategies that overcome current limitations. We argue that large-scale and multidisciplinary collaborative efforts based on clinical, scientific, and technical knowledge and rigorous clinical validation and implementation of the outcomes in practice are needed to move the research of sleep-disordered breathing forward, thus increasing the quality of diagnostics and treatment.

Multicentre sleep-stage scoring agreement in the Sleep Revolution project

Determining sleep stages accurately is an important part of the diagnostic process for numerous sleep disorders. However, as the sleep stage scoring is done manually following visual scoring rules there can be considerable variation in the sleep staging between different scorers. Thus, this study aimed to comprehensively evaluate the inter-rater agreement in sleep staging. A total of 50 polysomnography recordings were manually scored by 10 independent scorers from seven different sleep centres. We used the 10 scorings to calculate a majority score by taking the sleep stage that was the most scored stage for each epoch. The overall agreement for sleep staging was κ = 0.71 and the mean agreement with the majority score was 0.86. The scorers were in perfect agreement in 48% of all scored epochs. The agreement was highest in rapid eye movement sleep (κ = 0.86) and lowest in N1 sleep (κ = 0.41). The agreement with the majority scoring varied between the scorers from 81% to 91%, with large variations between the scorers in sleep stage-specific agreements. Scorers from the same sleep centres had the highest pairwise agreements at κ = 0.79, κ = 0.85, and κ = 0.78, while the lowest pairwise agreement between the scorers was κ = 0.58. We also found a moderate negative correlation between sleep staging agreement and the apnea-hypopnea index, as well as the rate of sleep stage transitions. In conclusion, although the overall agreement was high, several areas of low agreement were also found, mainly between non-rapid eye movement stages.

Polysomnography scoring-related training and quantitative assessment for improving interscorer agreement

STUDY OBJECTIVES

To efficiently improve the scoring competency of scorers with varying levels of experience across regions in Taiwan, we developed a training program with a cloud-based polysomnography scoring platform to evaluate and improve interscorer agreement.

METHODS

A total of 70 scorers from 34 sleep centers in Taiwan (job tenure: 0.5-39.0 years) completed a scoring test. All scorers scored a 742-epoch (30 s/epoch) overnight polysomnography recording of a patient with a moderate apnea-hypopnea index. Subsequently, 8 scoring experts delivered 8 interactive online lectures (each lasting 30 minutes). The training program included identifying scoring weaknesses, highlighting the latest scoring rules, and providing physicians' perspectives. Afterward, the scorers completed the second scoring test on the same participant. Changes in agreement from the first to second scoring test were identified. Sleep staging, sleep parameters, and respiratory events were considered for evaluating scoring agreement.

RESULTS

The scorers' agreement in overall sleep stage scoring significantly increased from 74.6 to 82.3% (median score). The proportion of scorers with an agreement of ≥ 80% increased from 20.0% (14/70) to 58.6% (41/70) after the online training program. In addition, the scorers' agreement in overall respiratory-event scoring increased to 88.8% (median score) after training. The scorers with a job tenure of 2.0-4.9 years exhibited the highest level of improvement in overall sleep staging (their median agreement increased from 72.8 to 84.9%; P < .001).

CONCLUSIONS

Our interactive online training program efficiently targeted the scorers' scoring weaknesses identified in the first scoring test, leading to substantial improvements in scoring proficiency.

CITATION

Liao Y-S, Wu M-C, Li C-X, Lin W-K, Lin C-Y, Liang S-F. Polysomnography scoring-related training and quantitative assessment for improving interscorer agreement. J Clin Sleep Med. 2024;20(2):271-278.

Sleep Apnea Physiological Burdens and Cardiovascular Morbidity and Mortality

Rationale: Obstructive sleep apnea is characterized by frequent reductions in ventilation, leading to oxygen desaturations and/or arousals. Objectives: In this study, association of hypoxic burden with incident cardiovascular disease (CVD) was examined and compared with that of "ventilatory burden" and "arousal burden." Finally, we assessed the extent to which the ventilatory burden, visceral obesity, and lung function explain variations in hypoxic burden. Methods: Hypoxic, ventilatory, and arousal burdens were measured from baseline polysomnograms in the Multi-Ethnic Study of Atherosclerosis (MESA) and the Osteoporotic Fractures in Men (MrOS) studies. Ventilatory burden was defined as event-specific area under ventilation signal (mean normalized, area under the mean), and arousal burden was defined as the normalized cumulative duration of all arousals. The adjusted hazard ratios for incident CVD and mortality were calculated. Exploratory analyses quantified contributions to hypoxic burden of ventilatory burden, baseline oxygen saturation as measured by pulse oximetry, visceral obesity, and spirometry parameters. Measurements and Main Results: Hypoxic and ventilatory burdens were significantly associated with incident CVD (adjusted hazard ratio [95% confidence interval] per 1 SD increase in hypoxic burden: MESA, 1.45 [1.14, 1.84]; MrOS, 1.13 [1.02, 1.26]; ventilatory burden: MESA, 1.38 [1.11, 1.72]; MrOS, 1.12 [1.01, 1.25]), whereas arousal burden was not. Similar associations with mortality were also observed. Finally, 78% of variation in hypoxic burden was explained by ventilatory burden, whereas other factors explained only <2% of variation. Conclusions: Hypoxic and ventilatory burden predicted CVD morbidity and mortality in two population-based studies. Hypoxic burden is minimally affected by measures of adiposity and captures the risk attributable to ventilatory burden of obstructive sleep apnea rather than a tendency to desaturate.

Challenges of Applying Automated Polysomnography Scoring at Scale

Automatic polysomnography analysis can be leveraged to shorten scoring times, reduce associated costs, and ultimately improve the overall diagnosis of sleep disorders. Multiple and diverse strategies have been attempted for implementation of this technology at scale in the routine workflow of sleep centers. The field, however, is complex and presents unsolved challenges in a number of areas. Recent developments in computer science and artificial intelligence are nevertheless closing the gap. Technological advances are also opening new pathways for expanding our current understanding of the domain and its analysis.

Creating an Optimal Approach for Diagnosing Sleep Apnea

Sleep apnea is nowadays recognized as a treatable chronic disease and awareness of it has increased, leading to an upsurge in demand for diagnostic testing. Conventionally, diagnosis depends on overnight polysomnography in a sleep clinic, which is highly human-resource intensive and ignores the night-to-night variability in classical sleep apnea markers, such as the apnea-hypopnea index. In this review, the authors summarize the main improvements that could be made in the sleep apnea diagnosis strategy; how technological innovations and multi-night home testing could be used to simplify, increase access, and reduce costs of diagnostic testing while avoiding misclassification of severity.

MSED: A Multi-Modal Sleep Event Detection Model for Clinical Sleep Analysis

Clinical sleep analysis require manual analysis of sleep patterns for correct diagnosis of sleep disorders. However, several studies have shown significant variability in manual scoring of clinically relevant discrete sleep events, such as arousals, leg movements, and sleep disordered breathing (apneas and hypopneas). We investigated whether an automatic method could be used for event detection and if a model trained on all events (joint model) performed better than corresponding event-specific models (single-event models). We trained a deep neural network event detection model on 1653 individual recordings and tested the optimized model on 1000 separate hold-out recordings. F1 scores for the optimized joint detection model were 0.70, 0.63, and 0.62 for arousals, leg movements, and sleep disordered breathing, respectively, compared to 0.65, 0.61, and 0.60 for the optimized single-event models. Index values computed from detected events correlated positively with manual annotations (r2 = 0.73, r2 = 0.77, r2 = 0.78, respectively). We furthermore quantified model accuracy based on temporal difference metrics, which improved overall by using the joint model compared to single-event models. Our automatic model jointly detects arousals, leg movements and sleep disordered breathing events with high correlation with human annotations. Finally, we benchmark against previous state-of-the-art multi-event detection models and found an overall increase in F1 score with our proposed model despite a 97.5% reduction in model size.

Autonomic arousal detection and cardio-respiratory sleep staging improve the accuracy of home sleep apnea tests

Introduction: The apnea-hypopnea index (AHI), defined as the number of apneas and hypopneas per hour of sleep, is still used as an important index to assess sleep disordered breathing (SDB) severity, where hypopneas are confirmed by the presence of an oxygen desaturation or an arousal. Ambulatory polygraphy without neurological signals, often referred to as home sleep apnea testing (HSAT), can potentially underestimate the severity of sleep disordered breathing (SDB) as sleep and arousals are not assessed. We aim to improve the diagnostic accuracy of HSATs by extracting surrogate sleep and arousal information derived from autonomic nervous system activity with artificial intelligence. Methods: We used polysomnographic (PSG) recordings from 245 subjects (148 with simultaneously recorded HSATs) to develop and validate a new algorithm to detect autonomic arousals using artificial intelligence. A clinically validated auto-scoring algorithm (Somnolyzer) scored respiratory events, cortical arousals, and sleep stages in PSGs, and provided respiratory events and sleep stages from cardio-respiratory signals in HSATs. In a four-fold cross validation of the newly developed algorithm, we evaluated the accuracy of the estimated arousal index and HSAT-derived surrogates for the AHI. Results: The agreement between the autonomic and cortical arousal index was moderate to good with an intraclass correlation coefficient of 0.73. When using thresholds of 5, 15, and 30 to categorize SDB into none, mild, moderate, and severe, the addition of sleep and arousal information significantly improved the classification accuracy from 70.2% (Cohen's κ = 0.58) to 80.4% (κ = 0.72), with a significant reduction of patients where the severity category was underestimated from 18.8% to 7.3%. Discussion: Extracting sleep and arousal information from autonomic nervous system activity can improve the diagnostic accuracy of HSATs by significantly reducing the probability of underestimating SDB severity without compromising specificity.

Translation of obstructive sleep apnea pathophysiology and phenotypes to personalized treatment: a narrative review

Obstructive Sleep Apnea (OSA) arises due to periodic blockage of the upper airway (UA) during sleep, as negative pressure generated during inspiration overcomes the force exerted by the UA dilator muscles to maintain patency. This imbalance is primarily seen in individuals with a narrowed UA, attributable to factors such as inherent craniofacial anatomy, neck fat accumulation, and rostral fluid shifts in the supine posture. Sleep-induced attenuation of UA dilating muscle responsiveness, respiratory instability, and high loop gain further exacerbate UA obstruction. The widespread comorbidity profile of OSA, encompassing cardiovascular, metabolic, and neuropsychiatric domains, suggests complex bidirectional relationships with conditions like heart failure, stroke, and metabolic syndrome. Recent advances have delineated distinct OSA phenotypes beyond mere obstruction frequency, showing links with specific symptomatic manifestations. It is vital to bridge the gap between measurable patient characteristics, phenotypes, and underlying pathophysiological traits to enhance our understanding of OSA and its interplay with related outcomes. This knowledge could stimulate the development of tailored therapies targeting specific phenotypic and pathophysiological endotypes. This review aims to elucidate the multifaceted pathophysiology of OSA, focusing on the relationships between UA anatomy, functional traits, clinical manifestations, and comorbidities. The ultimate objective is to pave the way for a more personalized treatment paradigm in OSA, offering alternatives to continuous positive airway pressure therapy for selected patients and thereby optimizing treatment efficacy and adherence. There is an urgent need for personalized treatment strategies in the ever-evolving field of sleep medicine, as we progress from a 'one-size-fits-all' to a 'tailored-therapy' approach.

International Consensus Statement on Obstructive Sleep Apnea

BACKGROUND

Evaluation and interpretation of the literature on obstructive sleep apnea (OSA) allows for consolidation and determination of the key factors important for clinical management of the adult OSA patient. Toward this goal, an international collaborative of multidisciplinary experts in sleep apnea evaluation and treatment have produced the International Consensus statement on Obstructive Sleep Apnea (ICS:OSA).

METHODS

Using previously defined methodology, focal topics in OSA were assigned as literature review (LR), evidence-based review (EBR), or evidence-based review with recommendations (EBR-R) formats. Each topic incorporated the available and relevant evidence which was summarized and graded on study quality. Each topic and section underwent iterative review and the ICS:OSA was created and reviewed by all authors for consensus.

RESULTS

The ICS:OSA addresses OSA syndrome definitions, pathophysiology, epidemiology, risk factors for disease, screening methods, diagnostic testing types, multiple treatment modalities, and effects of OSA treatment on multiple OSA-associated comorbidities. Specific focus on outcomes with positive airway pressure (PAP) and surgical treatments were evaluated.

CONCLUSION

This review of the literature consolidates the available knowledge and identifies the limitations of the current evidence on OSA. This effort aims to create a resource for OSA evidence-based practice and identify future research needs. Knowledge gaps and research opportunities include improving the metrics of OSA disease, determining the optimal OSA screening paradigms, developing strategies for PAP adherence and longitudinal care, enhancing selection of PAP alternatives and surgery, understanding health risk outcomes, and translating evidence into individualized approaches to therapy.

Overview of a Sleep Monitoring Protocol for a Large Natural Population

A standard operating procedure for studying the sleep phenotypes in a large population cohort is proposed. It is intended for academic researchers in investigating the sleep phenotypes in conjunction with the clinical sleep disorders assessment guidelines. The protocol refers to the definitive American Academy of Sleep Medicine (AASM) manual for setting polysomnography (PSG) technical specifications, scoring of sleep and associated events, etc. On this basis, it not only provides a standardized procedure of sleep interview, sleep-relevant questionnaires, and laboratory-based PSG test, but also offers a comprehensive process of sleep data analysis, phenotype extraction, and data storage. Both the objective sleep data recorded by PSG test and subjective sleep information obtained by the sleep interview and sleep questionnaires are involved in the data acquisition procedure. Subsequently, sleep phenotypes can be characterized by observable/inconspicuous physiological patterns during sleep from PSG test or can be marked by sleeping habits like sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, daytime dysfunction, etc., from sleep interview or questionnaires derived. In addition, solutions to the problems that may be encountered during the protocol are summarized and addressed. With the protocol, it can significantly improve scientific research efficiency and reduce unnecessary workload in large population cohort studies. Moreover, it is also expected to provide a valuable reference for researchers to conduct systematic sleep research.

Scoring sleep with artificial intelligence enables quantification of sleep stage ambiguity: hypnodensity based on multiple expert scorers and auto-scoring

STUDY OBJECTIVES

To quantify the amount of sleep stage ambiguity across expert scorers and to validate a new auto-scoring platform against sleep staging performed by multiple scorers.

METHODS

We applied a new auto-scoring system to three datasets containing 95 PSGs scored by 6-12 scorers, to compare sleep stage probabilities (hypnodensity; i.e. the probability of each sleep stage being assigned to a given epoch) as the primary output, as well as a single sleep stage per epoch assigned by hierarchical majority rule.

RESULTS

The percentage of epochs with 100% agreement across scorers was 46 ± 9%, 38 ± 10% and 32 ± 9% for the datasets with 6, 9, and 12 scorers, respectively. The mean intra-class correlation coefficient between sleep stage probabilities from auto- and manual-scoring was 0.91, representing excellent reliability. Within each dataset, agreement between auto-scoring and consensus manual-scoring was significantly higher than agreement between manual-scoring and consensus manual-scoring (0.78 vs. 0.69; 0.74 vs. 0.67; and 0.75 vs. 0.67; all p < 0.01).

CONCLUSIONS

Analysis of scoring performed by multiple scorers reveals that sleep stage ambiguity is the rule rather than the exception. Probabilities of the sleep stages determined by artificial intelligence auto-scoring provide an excellent estimate of this ambiguity. Compared to consensus manual-scoring, sleep staging derived from auto-scoring is for each individual PSG noninferior to manual-scoring meaning that auto-scoring output is ready for interpretation without the need for manual adjustment.

Deep learning-based algorithm accurately classifies sleep stages in preadolescent children with sleep-disordered breathing symptoms and age-matched controls

Introduction

Visual sleep scoring has several shortcomings, including inter-scorer inconsistency, which may adversely affect diagnostic decision-making. Although automatic sleep staging in adults has been extensively studied, it is uncertain whether such sophisticated algorithms generalize well to different pediatric age groups due to distinctive EEG characteristics. The preadolescent age group (10-13-year-olds) is relatively understudied, and thus, we aimed to develop an automatic deep learning-based sleep stage classifier specifically targeting this cohort.

Methods

A dataset (n = 115) containing polysomnographic recordings of Icelandic preadolescent children with sleep-disordered breathing (SDB) symptoms, and age and sex-matched controls was utilized. We developed a combined convolutional and long short-term memory neural network architecture relying on electroencephalography (F4-M1), electrooculography (E1-M2), and chin electromyography signals. Performance relative to human scoring was further evaluated by analyzing intra- and inter-rater agreements in a subset (n = 10) of data with repeat scoring from two manual scorers.

Results

The deep learning-based model achieved an overall cross-validated accuracy of 84.1% (Cohen's kappa κ = 0.78). There was no meaningful performance difference between SDB-symptomatic (n = 53) and control subgroups (n = 52) [83.9% (κ = 0.78) vs. 84.2% (κ = 0.78)]. The inter-rater reliability between manual scorers was 84.6% (κ = 0.78), and the automatic method reached similar agreements with scorers, 83.4% (κ = 0.76) and 82.7% (κ = 0.75).

Conclusion

The developed algorithm achieved high classification accuracy and substantial agreements with two manual scorers; the performance metrics compared favorably with typical inter-rater reliability between manual scorers and performance reported in previous studies. These suggest that our algorithm may facilitate less labor-intensive and reliable automatic sleep scoring in preadolescent children.

Benchmarking performance of an automatic polysomnography scoring system in a population with suspected sleep disorders

Aim

The current gold standard for measuring sleep disorders is polysomnography (PSG), which is manually scored by a sleep technologist. Scoring a PSG is time-consuming and tedious, with substantial inter-rater variability. A deep-learning-based sleep analysis software module can perform autoscoring of PSG. The primary objective of the study is to validate the accuracy and reliability of the autoscoring software. The secondary objective is to measure workflow improvements in terms of time and cost via a time motion study.

Methodology

The performance of an automatic PSG scoring software was benchmarked against the performance of two independent sleep technologists on PSG data collected from patients with suspected sleep disorders. The technologists at the hospital clinic and a third-party scoring company scored the PSG records independently. The scores were then compared between the technologists and the automatic scoring system. An observational study was also performed where the time taken for sleep technologists at the hospital clinic to manually score PSGs was tracked, along with the time taken by the automatic scoring software to assess for potential time savings.

Results

Pearson's correlation between the manually scored apnea-hypopnea index (AHI) and the automatically scored AHI was 0.962, demonstrating a near-perfect agreement. The autoscoring system demonstrated similar results in sleep staging. The agreement between automatic staging and manual scoring was higher in terms of accuracy and Cohen's kappa than the agreement between experts. The autoscoring system took an average of 42.7 s to score each record compared with 4,243 s for manual scoring. Following a manual review of the auto scores, an average time savings of 38.6 min per PSG was observed, amounting to 0.25 full-time equivalent (FTE) savings per year.

Conclusion

The findings indicate a potential for a reduction in the burden of manual scoring of PSGs by sleep technologists and may be of operational significance for sleep laboratories in the healthcare setting.

Preventive pharmacological treatment in subjects at risk for fatal familial insomnia: science and public engagement

Engaging patients as partners in biomedical research has gradually gained consensus over the last two decades. They provide a different perspective on health priorities and help to improve design and outcomes of clinical studies. This paper describes the relationship established between scientists and members of a large family at genetic risk of very rare lethal disease, fatal familial insomnia (FFI). This interaction led to a clinical trial based on the repurposing of doxycycline - an antibiotic with a known safety profile and optimal blood-brain barrier passage - which in numerous preclinical and clinical studies had given evidence of its potential therapeutic effect in neurodegenerative disorders, including prion diseases like FFI. The design of this trial posed several challenges, which were addressed jointly by the scientists and the FFI family. Potential participants excluded the possibility of being informed of their own FFI genotype; thus, the trial design had to include both carriers of the FFI mutation (10 subjects), and non-carriers (15 subjects), who were given placebo. Periodic clinical controls were performed on both groups by blinded examiners. The lack of surrogate outcome measures of treatment efficacy has required to compare the incidence of the disease in the treated group with a historical dataset during 10 years of observation. The trial is expected to end in 2023. Regardless of the clinical outcome, it will provide worthwhile knowledge on the disease. It also offers an important example of public engagement and collaboration to improve the quality of clinical science.

Influence of social jetlag on daytime sleepiness in obstructive sleep apnea

Social jetlag is the discrepancy between socially determined sleep timing on workdays and biologically determined sleep timing on days free of social obligation. Poor circadian timing of sleep may worsen sleep quality and increase daytime sleepiness in obstructive sleep apnea (OSA). We analysed de-identified data from 2,061 participants (75.2% male, mean [SD] age 48.6 [13.4] years) who completed Sleep Apnea Global Interdisciplinary Consortium (SAGIC) research questionnaires and underwent polysomnography at 11 international sleep clinic sites. Social jetlag was calculated as the absolute difference in the midpoints of sleep between weekdays and weekends. Daytime sleepiness was assessed using the Epworth Sleepiness Scale (ESS). Linear regression analyses were performed to estimate the association between social jetlag and daytime sleepiness, with consideration of age, sex, body mass index, ethnicity, insomnia, alcohol consumption, and habitual sleep duration as confounders. Of the participants, 61.5% had <1 h of social jetlag, 27.5% had 1 to <2 h, and 11.1% had ≥2 h. Compared to those with <1 h of social jetlag, those with ≥2 h of social jetlag had 2.07 points higher ESS (95% confidence interval [CI] 0.77-3.38, p = 0.002), and those with 1 to <2 h of social jetlag had 0.80 points higher ESS (95% CI 0.04-1.55, p = 0.04) after adjustment for potential confounding. Interaction with OSA severity was observed; social jetlag appeared to have the greatest effect on daytime sleepiness in mild OSA. As social jetlag exacerbates daytime sleepiness in OSA, improving sleep timing may be a simple but novel therapeutic target for reducing the impact of OSA.

Computer-assisted analysis of polysomnographic recordings improves inter-scorer associated agreement and scoring times

STUDY OBJECTIVES

To investigate inter-scorer agreement and scoring time differences associated with visual and computer-assisted analysis of polysomnographic (PSG) recordings.

METHODS

A group of 12 expert scorers reviewed 5 PSGs that were independently selected in the context of each of the following tasks: (i) sleep staging, (ii) scoring of leg movements, (iii) detection of respiratory (apneic-related) events, and (iv) of electroencephalographic (EEG) arousals. All scorers independently reviewed the same recordings, hence resulting in 20 scoring exercises per scorer from an equal amount of different subjects. The procedure was repeated, separately, using the classical visual manual approach and a computer-assisted (semi-automatic) procedure. Resulting inter-scorer agreement and scoring times were examined and compared among the two methods.

RESULTS

Computer-assisted sleep scoring showed a consistent and statistically relevant effect toward less time required for the completion of each of the PSG scoring tasks. Gain factors ranged from 1.26 (EEG arousals) to 2.41 (leg movements). Inter-scorer kappa agreement was also consistently increased with the use of supervised semi-automatic scoring. Specifically, agreement increased from Κ = 0.76 to K = 0.80 (sleep stages), Κ = 0.72 to K = 0.91 (leg movements), Κ = 0.55 to K = 0.66 (respiratory events), and Κ = 0.58 to Κ = 0.65 (EEG arousals). Inter-scorer agreement on the examined set of diagnostic indices did also show a trend toward higher Interclass Correlation Coefficient scores when using the semi-automatic scoring approach.

CONCLUSIONS

Computer-assisted analysis can improve inter-scorer agreement and scoring times associated with the review of PSG studies resulting in higher efficiency and overall quality in the diagnosis sleep disorders.

An Automated Algorithm for Determining Sleep Using Single-Channel Electroencephalography to Detect Delirium: A Prospective Observational Study in Intensive Care Units

The relationship between polysomnography-based objective sleep and delirium in the intensive care unit (ICU) is inconsistent across studies, suggesting limitations in manually determining the sleep stage of critically ill patients. We objectively measured 24-h sleep using a single-channel electroencephalogram (SleepScope [SS]) and an under-mattress sleep monitor (Nemuri SCAN [NSCAN]), both of which have independent algorithms that automatically determine sleep and wakefulness. Eighteen patients (median age, 68 years) admitted to the ICU after valvular surgery or coronary artery bypass grafting were included, and their sleep time was measured one day after extubation. The median total sleep times (TSTs) measured by SS (TST-SS) and NSCAN were 548 (48−1050) and 1024 (462−1257) min, respectively. Two patients with delirium during the 24-h sleep measurement had very short TST-SS of 48 and 125 min, and the percentage of daytime sleep accounted for >80% in both SS and NSCAN. This preliminary case series showed marked sleep deprivation and increased rates of daytime sleeping in ICU patients with delirium. Although data accuracy from under-mattress sleep monitors is contentious, automated algorithmic sleep/wakefulness determination using a single-channel electroencephalogram may be useful in detecting delirium in ICU patients and could even be superior to polysomnography.

A Validation of Six Wearable Devices for Estimating Sleep, Heart Rate and Heart Rate Variability in Healthy Adults

The primary aim of this study was to examine the validity of six commonly used wearable devices, i.e., Apple Watch S6, Garmin Forerunner 245 Music, Polar Vantage V, Oura Ring Generation 2, WHOOP 3.0 and Somfit, for assessing sleep. The secondary aim was to examine the validity of the six devices for assessing heart rate and heart rate variability during, or just prior to, night-time sleep. Fifty-three adults (26 F, 27 M, aged 25.4 ± 5.9 years) spent a single night in a sleep laboratory with 9 h in bed (23:00-08:00 h). Participants were fitted with all six wearable devices-and with polysomnography and electrocardiography for gold-standard assessment of sleep and heart rate, respectively. Compared with polysomnography, agreement (and Cohen's kappa) for two-state categorisation of sleep periods (as sleep or wake) was 88% (κ = 0.30) for Apple Watch; 89% (κ = 0.35) for Garmin; 87% (κ = 0.44) for Polar; 89% (κ = 0.51) for Oura; 86% (κ = 0.44) for WHOOP and 87% (κ = 0.48) for Somfit. Compared with polysomnography, agreement (and Cohen's kappa) for multi-state categorisation of sleep periods (as a specific sleep stage or wake) was 53% (κ = 0.20) for Apple Watch; 50% (κ = 0.25) for Garmin; 51% (κ = 0.28) for Polar; 61% (κ = 0.43) for Oura; 60% (κ = 0.44) for WHOOP and 65% (κ = 0.52) for Somfit. Analyses regarding the two-state categorisation of sleep indicate that all six devices are valid for the field-based assessment of the timing and duration of sleep. However, analyses regarding the multi-state categorisation of sleep indicate that all six devices require improvement for the assessment of specific sleep stages. As the use of wearable devices that are valid for the assessment of sleep increases in the general community, so too does the potential to answer research questions that were previously impractical or impossible to address-in some way, we could consider that the whole world is becoming a sleep laboratory.

Age estimation from sleep studies using deep learning predicts life expectancy

Sleep disturbances increase with age and are predictors of mortality. Here, we present deep neural networks that estimate age and mortality risk through polysomnograms (PSGs). Aging was modeled using 2500 PSGs and tested in 10,699 PSGs from men and women in seven different cohorts aged between 20 and 90. Ages were estimated with a mean absolute error of 5.8 ± 1.6 years, while basic sleep scoring measures had an error of 14.9 ± 6.29 years. After controlling for demographics, sleep, and health covariates, each 10-year increment in age estimate error (AEE) was associated with increased all-cause mortality rate of 29% (95% confidence interval: 20-39%). An increase from -10 to +10 years in AEE translates to an estimated decreased life expectancy of 8.7 years (95% confidence interval: 6.1-11.4 years). Greater AEE was mostly reflected in increased sleep fragmentation, suggesting this is an important biomarker of future health independent of sleep apnea.

Association Between Sleep Microstructure and Incident Hypertension in a Population‐Based Sample: The HypnoLaus Study

Background

Poor sleep quality is associated with increased incident hypertension. However, few studies have investigated the impact of objective sleep structure parameters on hypertension. This study investigated the association between sleep macrostructural and microstructural parameters and incident hypertension in a middle‐ to older‐aged sample.

Methods and Results

Participants from the HypnoLaus population‐based cohort without hypertension at baseline were included. Participants had at‐home polysomnography at baseline, allowing assessment of sleep macrostructure (nonrapid eye movement sleep stages 1, 2, and 3; rapid eye movement sleep stages; and total sleep time) and microstructure including power spectral density of electroencephalogram in nonrapid eye movement sleep and spindles characteristics (density, duration, frequency, amplitude) in nonrapid eye movement sleep stage 2. Associations between sleep macrostructure and microstructure parameters at baseline and incident clinical hypertension over a mean follow‐up of 5.2 years were assessed with multiple‐adjusted logistic regression. A total of 1172 participants (42% men; age 55±10 years) were included. Of these, 198 (17%) developed hypertension. After adjustment for confounders, no sleep macrostructure features were associated with incident hypertension. However, low absolute delta and sigma power were significantly associated with incident hypertension where participants in the lowest quartile of delta and sigma had a 1.69‐fold (95% CI, 1.00–2.89) and 1.72‐fold (95% CI, 1.05–2.82) increased risk of incident hypertension, respectively, versus those in the highest quartile. Lower spindle density (odds ratio, 0.87; 95% CI, 0.76–0.99) and amplitude (odds ratio, 0.98; 95% CI, 0.95–1.00) were also associated with higher incident hypertension.

Conclusions

Sleep microstructure is associated with incident hypertension. Slow‐wave activity and sleep spindles, 2 hallmarks of objective sleep continuity and quality, were inversely and consistently associated with incident hypertension. This supports the protective role of sleep continuity in the development of hypertension.

Diagnosis of Sleep Apnoea Using a Mandibular Monitor and Machine Learning Analysis: One-Night Agreement Compared to in-Home Polysomnography

Background

The capacity to diagnose obstructive sleep apnoea (OSA) must be expanded to meet an estimated disease burden of nearly one billion people worldwide. Validated alternatives to the gold standard polysomnography (PSG) will improve access to testing and treatment. This study aimed to evaluate the diagnosis of OSA, using measurements of mandibular movement (MM) combined with automated machine learning analysis, compared to in-home PSG.

Methods

40 suspected OSA patients underwent single overnight in-home sleep testing with PSG (Nox A1, ResMed, Australia) and simultaneous MM monitoring (Sunrise, Sunrise SA, Belgium). PSG recordings were manually analysed by two expert sleep centres (Grenoble and London); MM analysis was automated. The Obstructive Respiratory Disturbance Index calculated from the MM monitoring (MM-ORDI) was compared to the PSG (PSG-ORDI) using intraclass correlation coefficient and Bland-Altman analysis. Receiver operating characteristic curves (ROC) were constructed to optimise the diagnostic performance of the MM monitor at different PSG-ORDI thresholds (5, 15, and 30 events/hour).

Results

31 patients were included in the analysis (58% men; mean (SD) age: 48 (15) years; BMI: 30.4 (7.6) kg/m2). Good agreement was observed between MM-ORDI and PSG-ORDI (median bias 0.00; 95% CI −23.25 to + 9.73 events/hour). However, for 15 patients with no or mild OSA, MM monitoring overestimated disease severity (PSG-ORDI &lt; 5: MM-ORDI mean overestimation + 5.58 (95% CI + 2.03 to + 7.46) events/hour; PSG-ORDI &gt; 5–15: MM-ORDI overestimation + 3.70 (95% CI −0.53 to + 18.32) events/hour). In 16 patients with moderate-severe OSA (n = 9 with PSG-ORDI 15–30 events/h and n = 7 with a PSG-ORD &gt; 30 events/h), there was an underestimation (PSG-ORDI &gt; 15: MM-ORDI underestimation −8.70 (95% CI −28.46 to + 4.01) events/hour). ROC optimal cut-off values for PSG-ORDI thresholds of 5, 15, 30 events/hour were: 9.53, 12.65 and 24.81 events/hour, respectively. These cut-off values yielded a sensitivity of 88, 100 and 79%, and a specificity of 100, 75, 96%. The positive predictive values were: 100, 80, 95% and the negative predictive values 89, 100, 82%, respectively.

Conclusion

The diagnosis of OSA, using MM with machine learning analysis, is comparable to manually scored in-home PSG. Therefore, this novel monitor could be a convenient diagnostic tool that can easily be used in the patients’ own home.

Clinical Trial Registration

https://clinicaltrials.gov, identifier NCT04262557

CMS2-net: Semi-supervised Sleep Staging for Diverse Obstructive Sleep Apnea Severity

Although the development of computer-aided algorithms for sleep staging is integrated into automatic detection of sleep disorders, most supervised deep learning based models might suffer from insufficient labeled data. While the adoption of semi-supervised learning (SSL) can mitigate the issue, the SSL models are still limited to the lack of discriminative feature extraction for diverse obstructive sleep apnea (OSA) severity. This model deterioration might be exacerbated during the domain adaptation. Such exploration on the alleviation of domain-shift of SSL model between different OSA conditions has attracted more and more attentions from the clinic. In this work, a co-attention meta sleep staging network (CMS2-net) is proposed to simultaneously deal with two issues: the inter-class disparity problem and the intra-class selection problem. Within CMS2-net, a co-attention module and a triple-classifier are designed to explicitly refine the coarse feature representations by identifying the class boundary inconsistency. Moreover, the mutual information with meta contrastive variance is introduced to supervise the gradient stream from a multiscale view. The performance of the proposed framework is demonstrated on both public and local datasets. Furthermore, our approach achieves the state-of-the-art SSL results on both datasets.

Diagnostic Performance of Machine Learning-Derived OSA Prediction Tools in Large Clinical and Community-Based Samples

BACKGROUND

Prediction tools without patient-reported symptoms could facilitate widespread identification of OSA.

RESEARCH QUESTION

What is the diagnostic performance of OSA prediction tools derived from machine learning using readily available data without patient responses to questionnaires? Also, how do they compare with STOP-BANG, an OSA prediction tool, in clinical and community-based samples?

STUDY DESIGN AND METHODS

Logistic regression and machine learning techniques, including artificial neural network (ANN), random forests (RF), and kernel support vector machine, were used to determine the ability of age, sex, BMI, and race to predict OSA status. A retrospective cohort of 17,448 subjects from sleep clinics within the international Sleep Apnea Global Interdisciplinary Consortium (SAGIC) were randomly split into training (n = 10,469) and validation (n = 6,979) sets. Model comparisons were performed by using the area under the receiver-operating curve (AUC). Trained models were compared with the STOP-BANG questionnaire in two prospective testing datasets: an independent clinic-based sample from SAGIC (n = 1,613) and a community-based sample from the Sleep Heart Health Study (n = 5,599).

RESULTS

The AUCs (95% CI) of the machine learning models were significantly higher than logistic regression (0.61 [0.60-0.62]) in both the training and validation datasets (ANN, 0.68 [0.66-0.69]; RF, 0.68 [0.67-0.70]; and kernel support vector machine, 0.66 [0.65-0.67]). In the SAGIC testing sample, the ANN (0.70 [0.68-0.72]) and RF (0.70 [0.68-0.73]) models had AUCs similar to those of the STOP-BANG (0.71 [0.68-0.72]). In the Sleep Heart Health Study testing sample, the ANN (0.72 [0.71-0.74]) had AUCs similar to those of STOP-BANG (0.72 [0.70-0.73]).

INTERPRETATION

OSA prediction tools using machine learning without patient-reported symptoms provide better diagnostic performance than logistic regression. In clinical and community-based samples, the symptomless ANN tool has diagnostic performance similar to that of a widely used prediction tool that includes patient symptoms. Machine learning-derived algorithms may have utility for widespread identification of OSA.

Interrater reliability of sleep stage scoring: a meta-analysis

STUDY OBJECTIVES

We evaluated the interrater reliabilities of manual polysomnography sleep stage scoring. We included all studies that employed Rechtschaffen and Kales rules or American Academy of Sleep Medicine standards. We sought the overall degree of agreement and those for each stage.

METHODS

The keywords were "Polysomnography (PSG)," "sleep staging," "Rechtschaffen and Kales (R&K)," "American Academy of Sleep Medicine (AASM)," "interrater (interscorer) reliability," and "Cohen's kappa." We searched PubMed, OVID Medline, EMBASE, the Cochrane library, KoreaMed, KISS, and the MedRIC. The exclusion criteria included automatic scoring and pediatric patients. We collected data on scorer histories, scoring rules, numbers of epochs scored, and the underlying diseases of the patients.

RESULTS

A total of 101 publications were retrieved; 11 satisfied the selection criteria. The Cohen's kappa for manual, overall sleep scoring was 0.76, indicating substantial agreement (95% confidence interval, 0.71-0.81; P < .001). By sleep stage, the figures were 0.70, 0.24, 0.57, 0.57, and 0.69 for the W, N1, N2, N3, and R stages, respectively. The interrater reliabilities for stage N2 and N3 sleep were moderate, and that for stage N1 sleep was only fair.

CONCLUSIONS

We conducted a meta-analysis to generalize the variation in manual scoring of polysomnography and provide reference data for automatic sleep stage scoring systems. The reliability of manual scorers of polysomnography sleep stages was substantial. However, for certain stages, the results were poor; validity requires improvement.

CITATION

Lee YJ, Lee JY, Cho JH, Choi JH. Interrater reliability of sleep stage scoring: a meta-analysis. J Clin Sleep Med. 2022;18(1):193-202.

Automated Scoring of Sleep and Associated Events

Conventionally, sleep and associated events are scored visually by trained technologists according to the rules summarized in the American Academy of Sleep Medicine Manual. Since its first publication in 2007, the manual was continuously updated; the most recent version as of this writing was published in 2020. Human expert scoring is considered as gold standard, even though there is increasing evidence of limited interrater reliability between human scorers. Significant advances in machine learning have resulted in powerful methods for addressing complex classification problems such as automated scoring of sleep and associated events. Evidence is increasing that these autoscoring systems deliver performance comparable to manual scoring and offer several advantages to visual scoring: (1) avoidance of the rather expensive, time-consuming, and difficult visual scoring task that can be performed only by well-trained and experienced human scorers, (2) attainment of consistent scoring results, and (3) proposition of added value such as scoring in real time, sleep stage probabilities per epoch (hypnodensity), estimates of signal quality and sleep/wake-related features, identifications of periods with clinically relevant ambiguities (confidence trends), configurable sensitivity and rule settings, as well as cardiorespiratory sleep staging for home sleep apnea testing. This chapter describes the development of autoscoring systems since the first attempts in the 1970s up to the most recent solutions based on deep neural network approaches which achieve an accuracy that allows to use the autoscoring results directly for review and interpretation by a physician.

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.

Actigraphy in mechanically ventilated pediatric ICU patients: comparison to PSG and evaluation of behavioral circadian rhythmicity

Sleep disruption is common in pediatric intensive care unit (PICU) patients, but measuring sleep in this population is challenging. We aimed to evaluate the utility of actigraphy for estimating circadian rhythmicity in mechanically ventilated PICU patients and its accuracy for measuring sleep by comparing it to polysomnogram (PSG). We conducted a single-center prospective observational study of children 6 months - 17 years of age receiving mechanical ventilation and standard, protocolized sedation for acute respiratory failure, excluding children with acute or historical neurologic injury. We enrolled 16 children and monitored them with up to 14 days of actigraphy and 24 hours of simultaneous limited (10 channel) PSG. Daily actigraphy-based activity profiles demonstrated that patients had a high level of nighttime activity (30-41% of total activity), suggesting disrupted circadian activity cycles. Among n = 12 patients with sufficient actigraphy and PSG data overlap, actigraphy-based sleep estimation showed poor agreement with PSG-identified sleep states, with good sensitivity (94%) but poor specificity (28%), low accuracy (70%,) and low agreement (Cohen's kappa = 0.2, 95% CI = 0.08-0.31). Using univariate linear regression, we identified that Cornell Assessment of Pediatric Delirium scores were associated with accuracy of actigraphy but that other clinical factors including sedative medication doses, activity levels, and restraint use were not. In this population, actigraphy did not reliably discern between sleep and wake states. However, in select patients, actigraphy was able to distinguish diurnal variation in activity patterns, and therefore may be useful for evaluating patients' response to circadian-oriented interventions.

Statistical uncertainty quantification to augment clinical decision support: a first implementation in sleep medicine

Machine learning has the potential to change the practice of medicine, particularly in areas that require pattern recognition (e.g. radiology). Although automated classification is unlikely to be perfect, few modern machine learning tools have the ability to assess their own classification confidence to recognize uncertainty that might need human review. Using automated single-channel sleep staging as a first implementation, we demonstrated that uncertainty information (as quantified using Shannon entropy) can be utilized in a "human in the loop" methodology to promote targeted review of uncertain sleep stage classifications on an epoch-by-epoch basis. Across 20 sleep studies, this feedback methodology proved capable of improving scoring agreement with the gold standard over automated scoring alone (average improvement in Cohen's Kappa of 0.28), in a fraction of the scoring time compared to full manual review (60% reduction). In summary, our uncertainty-based clinician-in-the-loop framework promotes the improvement of medical classification accuracy/confidence in a cost-effective and economically resourceful manner.

A deep transfer learning approach for wearable sleep stage classification with photoplethysmography

Unobtrusive home sleep monitoring using wrist-worn wearable photoplethysmography (PPG) could open the way for better sleep disorder screening and health monitoring. However, PPG is rarely included in large sleep studies with gold-standard sleep annotation from polysomnography. Therefore, training data-intensive state-of-the-art deep neural networks is challenging. In this work a deep recurrent neural network is first trained using a large sleep data set with electrocardiogram (ECG) data (292 participants, 584 recordings) to perform 4-class sleep stage classification (wake, rapid-eye-movement, N1/N2, and N3). A small part of its weights is adapted to a smaller, newer PPG data set (60 healthy participants, 101 recordings) through three variations of transfer learning. Best results (Cohen's kappa of 0.65 ± 0.11, accuracy of 76.36 ± 7.57%) were achieved with the domain and decision combined transfer learning strategy, significantly outperforming the PPG-trained and ECG-trained baselines. This performance for PPG-based 4-class sleep stage classification is unprecedented in literature, bringing home sleep stage monitoring closer to clinical use. The work demonstrates the merit of transfer learning in developing reliable methods for new sensor technologies by reusing similar, older non-wearable data sets. Further study should evaluate our approach in patients with sleep disorders such as insomnia and sleep apnoea.

Inter-database validation of a deep learning approach for automatic sleep scoring

STUDY OBJECTIVES

Development of inter-database generalizable sleep staging algorithms represents a challenge due to increased data variability across different datasets. Sharing data between different centers is also a problem due to potential restrictions due to patient privacy protection. In this work, we describe a new deep learning approach for automatic sleep staging, and address its generalization capabilities on a wide range of public sleep staging databases. We also examine the suitability of a novel approach that uses an ensemble of individual local models and evaluate its impact on the resulting inter-database generalization performance.

METHODS

A general deep learning network architecture for automatic sleep staging is presented. Different preprocessing and architectural variant options are tested. The resulting prediction capabilities are evaluated and compared on a heterogeneous collection of six public sleep staging datasets. Validation is carried out in the context of independent local and external dataset generalization scenarios.

RESULTS

Best results were achieved using the CNN_LSTM_5 neural network variant. Average prediction capabilities on independent local testing sets achieved 0.80 kappa score. When individual local models predict data from external datasets, average kappa score decreases to 0.54. Using the proposed ensemble-based approach, average kappa performance on the external dataset prediction scenario increases to 0.62. To our knowledge this is the largest study by the number of datasets so far on validating the generalization capabilities of an automatic sleep staging algorithm using external databases.

CONCLUSIONS

Validation results show good general performance of our method, as compared with the expected levels of human agreement, as well as to state-of-the-art automatic sleep staging methods. The proposed ensemble-based approach enables flexible and scalable design, allowing dynamic integration of local models into the final ensemble, preserving data locality, and increasing generalization capabilities of the resulting system at the same time.

Automatic Respiratory Event Scoring in Obstructive Sleep Apnea Using a Long Short-Term Memory Neural Network

The diagnosis of obstructive sleep apnea is based on daytime symptoms and the frequency of respiratory events during the night. The respiratory events are scored manually from polysomnographic recordings, which is time-consuming and expensive. Therefore, automatic scoring methods could considerably improve the efficiency of sleep apnea diagnostics and release the resources currently needed for manual scoring to other areas of sleep medicine. In this study, we trained a long short-term memory neural network for automatic scoring of respiratory events using input signals from peripheral blood oxygen saturation, thermistor-airflow, nasal pressure -airflow, and thorax respiratory effort. The signals were extracted from 887 in-lab polysomnography recordings. 787 patients with suspected sleep apnea were used to train the neural network and 100 patients were used as an independent test set. The epoch-wise agreement between manual and automatic neural network scoring was high (88.9%, κ = 0.728). In addition, the apnea-hypopnea index (AHI) calculated from the automated scoring was close to the manually determined AHI with a mean absolute error of 3.0 events/hour and an intraclass correlation coefficient of 0.985. The neural network approach for automatic scoring of respiratory events achieved high accuracy and good agreement with manual scoring. The presented neural network could be used for analysis of large research datasets that are unfeasible to score manually, and has potential for clinical use in the future In addition, since the neural network scores individual respiratory events, the automatic scoring can be easily reviewed manually if desired.

Detailed Assessment of Sleep Architecture With Deep Learning and Shorter Epoch-to-Epoch Duration Reveals Sleep Fragmentation of Patients With Obstructive Sleep Apnea

Traditional sleep staging with non-overlapping 30-second epochs overlooks multiple sleep-wake transitions. We aimed to overcome this by analyzing the sleep architecture in more detail with deep learning methods and hypothesized that the traditional sleep staging underestimates the sleep fragmentation of obstructive sleep apnea (OSA) patients. To test this hypothesis, we applied deep learning-based sleep staging to identify sleep stages with the traditional approach and by using overlapping 30-second epochs with 15-, 5-, 1-, or 0.5-second epoch-to-epoch duration. A dataset of 446 patients referred for polysomnography due to OSA suspicion was used to assess differences in the sleep architecture between OSA severity groups. The amount of wakefulness increased while REM and N3 decreased in severe OSA with shorter epoch-to-epoch duration. In other OSA severity groups, the amount of wake and N1 decreased while N3 increased. With the traditional 30-second epoch-to-epoch duration, only small differences in sleep continuity were observed between the OSA severity groups. With 1-second epoch-to-epoch duration, the hazard ratio illustrating the risk of fragmented sleep was 1.14 (p = 0.39) for mild OSA, 1.59 (p < 0.01) for moderate OSA, and 4.13 (p < 0.01) for severe OSA. With shorter epoch-to-epoch durations, total sleep time and sleep efficiency increased in the non-OSA group and decreased in severe OSA. In conclusion, more detailed sleep analysis emphasizes the highly fragmented sleep architecture in severe OSA patients which can be underestimated with traditional sleep staging. The results highlight the need for a more detailed analysis of sleep architecture when assessing sleep disorders.

Interrater sleep stage scoring reliability between manual scoring from two European sleep centers and automatic scoring performed by the artificial intelligence-based Stanford-STAGES algorithm

STUDY OBJECTIVES

The objective of this study was to evaluate interrater reliability between manual sleep stage scoring performed in 2 European sleep centers and automatic sleep stage scoring performed by the previously validated artificial intelligence-based Stanford-STAGES algorithm.

METHODS

Full night polysomnographies of 1,066 participants were included. Sleep stages were manually scored in Berlin and Innsbruck sleep centers and automatically scored with the Stanford-STAGES algorithm. For each participant, we compared (1) Innsbruck to Berlin scorings (INN vs BER); (2) Innsbruck to automatic scorings (INN vs AUTO); (3) Berlin to automatic scorings (BER vs AUTO); (4) epochs where scorers from Innsbruck and Berlin had consensus to automatic scoring (CONS vs AUTO); and (5) both Innsbruck and Berlin manual scorings (MAN) to the automatic ones (MAN vs AUTO). Interrater reliability was evaluated with several measures, including overall and sleep stage-specific Cohen's κ.

RESULTS

Overall agreement across participants was substantial for INN vs BER (κ = 0.66 ± 0.13), INN vs AUTO (κ = 0.68 ± 0.14), CONS vs AUTO (κ = 0.73 ± 0.14), and MAN vs AUTO (κ = 0.61 ± 0.14), and moderate for BER vs AUTO (κ = 0.55 ± 0.15). Human scorers had the highest disagreement for N1 sleep (κ_N1 = 0.40 ± 0.16 for INN vs BER). Automatic scoring had lowest agreement with manual scorings for N1 and N3 sleep (κ_N1 = 0.25 ± 0.14 and κ_N3 = 0.42 ± 0.32 for MAN vs AUTO).

CONCLUSIONS

Interrater reliability for sleep stage scoring between human scorers was in line with previous findings, and the algorithm achieved an overall substantial agreement with manual scoring. In this cohort, the Stanford-STAGES algorithm showed similar performances to the ones achieved in the original study, suggesting that it is generalizable to new cohorts. Before its integration in clinical practice, future independent studies should further evaluate it in other cohorts.

Severe OSA associated with higher risk of mortality in stage III and IV lung cancer

STUDY OBJECTIVES

OSA has been associated with increased cancer incidence and mortality. The aim of this study was to investigate cancer-related mortality, overall survival, and progression-free survival in patients with suspected OSA and lung cancer.

METHODS

This was a case series analysis of lung cancer from a sleep cohort with suspected OSA between 2009 and 2014. The AHI, hypoxia index, and survival outcome were recorded. Immunohistochemistry was used to analyze hypoxia-inducible factor-1α (HIF-1α) and vascular endothelial growth factor expression in tumor pathology.

RESULTS

In the sleep cohort comprising 8,261 patients, a total of 23 patients had lung cancer. The incidence of lung cancer was significantly higher in the sleep cohort than in the entire adult population in Taiwan (crude incidence rate: 242.1 vs 51.5 per 10⁵ persons, P < .01). The 3-year cancer-related mortality was 25% in AHI < 15 events/h, 50% in AHI 15-29 events/h, and 80% in AHI ≥ 30 events/h (χ² test for trend, P = .03). In Kaplan-Meier survival analysis, patients with stage III-IV lung cancer and AHI < 30 events/h exhibited significantly better overall survival (P = .02) and progression-free survival (P = .02) than patients with severe OSA. Overexpression of HIF-1α and vascular endothelial growth factor was shown in 63% and 45% of lung tumor samples. Overexpression of HIF-1α was positively associated with AHI (P = .04).

CONCLUSIONS

In this preliminary case series, severe OSA is associated with an increased risk of cancer mortality in patients with stage III-IV lung cancer. AHI was significantly associated with HIF-1α overexpression.