Obstructive Sleep Apnea Severity Levels Research Articles

Obstructive Sleep Apnea Characterization: A Multimodal Cross-Recurrence-Based Approach for Investigating Atrial Fibrillation.

Obstructive sleep apnea (OSA) is believed to contribute significantly to atrial fibrillation (AF) development in certain patients. Recent studies indicate a rising risk of AF with increasing OSA severity. However, the commonly used apnea-hypopnea index in clinical practice may not adequately account for the potential cardiovascular risks associated with OSA. (1) Objective: to propose and explore a novel method for assessing OSA severity considering potential connection to cardiac arrhythmias. (2) Method: the approach utilizes cross-recurrence features to characterize OSA and AF by considering the relationships among oxygen desaturation, pulse arrival time, and heart-beat intervals. Multinomial logistic regression models were trained to predict four levels of OSA severity and four groups related to heart rhythm issues. The rank biserial correlation coefficient, \boldmath rrb, was used to estimate effect size for statistical analysis. The investigation was conducted using the MESA database, which includes polysomnography data from 2055 subjects. (3) Results: a derived cross-recurrence-based index showed a significant association with a higher OSA severity (\boldmath p 0.01) and the presence of AF (\boldmath p 0.01). Additionally, the proposed index had a significantly larger effect, \boldmath rrb, than the conventional apnea-hypopnea index in differentiating increasingly severe heart rhythm issue groups: 0.14 0.06, 0.33 0.10, and 0.41 0.07. (4) Significance: the proposed method holds relevance as a supplementary diagnostic tool for assessing the authentic state of sleep apnea in clinical practice.

Combining a wireless radar sleep monitoring device with deep machine learning techniques to assess obstructive sleep apnea severity.

The gold standard for diagnosing obstructive sleep apnea (OSA) is polysomnography (PSG). However, PSG is a time-consuming method with clinical limitations. This study aimed to create a wireless radar framework to screen the likelihood of 2 levels of OSA severity (ie, moderate-to-severe and severe OSA) in accordance with clinical practice standards. We conducted a prospective, simultaneous study using a wireless radar system and PSG in a Northern Taiwan sleep center, involving 196 patients. The wireless radar sleep monitor, incorporating hybrid models such as deep neural decision trees, estimated the respiratory disturbance index relative to the total sleep time established by PSG (RDIPSG_TST), by analyzing continuous-wave signals indicative of breathing patterns. Analyses were performed to examine the correlation and agreement between the RDIPSG_TST and apnea-hypopnea index, results obtained through PSG. Cut-off thresholds for RDIPSG_TST were determined using Youden's index, and multiclass classification was performed, after which the results were compared. A strong correlation (ρ = 0.91) and agreement (average difference of 0.59 events/h) between apnea-hypopnea index and RDIPSG_TST were identified. In terms of the agreement between the 2 devices, the average difference between PSG-based apnea-hypopnea index and radar-based RDIPSG_TST was 0.59 events/h, and 187 out of 196 cases (95.41%) fell within the 95% confidence interval of differences. A moderate-to-severe OSA model achieved an accuracy of 90.3% (cut-off threshold for RDIPSG_TST: 19.2 events/h). A severe OSA model achieved an accuracy of 92.4% (cut-off threshold for RDIPSG_TST: 28.86 events/h). The mean accuracy of multiclass classification performance using these cut-off thresholds was 83.7%. The wireless-radar-based sleep monitoring device, with cut-off thresholds, can provide rapid OSA screening with acceptable accuracy and also alleviate the burden on PSG capacity. However, to independently apply this framework, the function of determining the radar-based total sleep time requires further optimizations and verification in future work. Lin S-Y, Tsai C-Y, Majumdar A, etal. Combining a wireless radar sleep monitoring device with deep machine learning techniques to assess obstructive sleep apnea severity. J Clin Sleep Med. 2024;20(8):1267-1277.

An explainable deep-learning architecture for pediatric sleep apnea identification from overnight airflow and oximetry signals

Deep-learning algorithms have been proposed to analyze overnight airflow (AF) and oximetry (SpO2) signals to simplify the diagnosis of pediatric obstructive sleep apnea (OSA), but current algorithms are hardly interpretable. Explainable artificial intelligence (XAI) algorithms can clarify the models-derived predictions on these signals, enhancing their diagnostic trustworthiness. Here, we assess an explainable architecture that combines convolutional and recurrent neural networks (CNN + RNN) to detect pediatric OSA and its severity. AF and SpO2 were obtained from the Childhood Adenotonsillectomy Trial (CHAT) public database (n = 1,638) and a proprietary database (n = 974). These signals were arranged in 30-min segments and processed by the CNN + RNN architecture to derive the number of apneic events per segment. The apnea-hypopnea index (AHI) was computed from the CNN + RNN-derived estimates and grouped into four OSA severity levels. The Gradient-weighted Class Activation Mapping (Grad-CAM) XAI algorithm was used to identify and interpret novel OSA-related patterns of interest. The AHI regression reached very high agreement (intraclass correlation coefficient > 0.9), while OSA severity classification achieved 4-class accuracies 74.51% and 62.31%, and 4-class Cohen’s Kappa 0.6231 and 0.4495, in CHAT and the private datasets, respectively. All diagnostic accuracies on increasing AHI cutoffs (1, 5 and 10 events/h) surpassed 84%. The Grad-CAM heatmaps revealed that the model focuses on sudden AF cessations and SpO2 drops to detect apneas and hypopneas with desaturations, and often discards patterns of hypopneas linked to arousals. Therefore, an interpretable CNN + RNN model to analyze AF and SpO2 can be helpful as a diagnostic alternative in symptomatic children at risk of OSA.

Associations of deep sleep and obstructive sleep apnea with family relationships, life satisfaction, and physical stress experience in children: a caregiver perspective.

Obstructive sleep apnea (OSA) and poor quality of sleep negatively impacts health-related quality of life in adults, but few studies have evaluated the association between sleep disturbance (eg, OSA, inadequate sleep) and health-related quality of life domains (eg, family relations, life satisfaction) in children. Children ages 8-17 years referred to a sleep center for routine polysomnography from April 2022 to August 2022 were approached to participate in the study, and children visiting the department of pediatrics for their wellness visit were recruited for comparisons. Statistical analysis was conducted using R 3.6.0. Ninety-nine children were recruited from the sleep clinic, and 23 children were recruited from the primary care clinic. Of these children, 62 were diagnosed with obstructive sleep apnea (31 mild, 12 moderate, 19 severe), and 37 did not meet criteria for a diagnosis. Health-related quality of life domains did not differ across OSA severity levels. Children in general had lower life satisfaction and higher physical stress experience compared to children visiting for their wellness examination (well-child visitors, P = .05 and P = .005, respectively). Children with severe OSA had significantly lower life satisfaction and significantly higher physical stress experience when compared with well-child visitors (P = .008 and P = .009, respectively). Correlation analysis showed that N3 (deep) sleep was positively associated with family relations and life satisfaction, while it was negatively associated with anger. Based on caregiver response, N3 sleep is positively associated with family relations and life satisfaction and negatively associated with anger. Severe OSA is associated with lower life satisfaction and higher physical stress experience. Bhushan B, Zee PC, Grandner MA, etal. Associations of deep sleep and obstructive sleep apnea with family relationships, life satisfaction, and physical stress experience in children: a caregiver perspective. J Clin Sleep Med. 2023;19(12):2087-2095.

0568 Trends in OSA Disease Severity Over Nearly Two Decades: Update on the VA San Diego Experience

Abstract Introduction The Veterans Health Administration (VA) is one of the largest integrated healthcare systems in the United States. Key risk factors for obstructive sleep apnea (OSA) include aging, weight, and male gender, all of which are highly prevalent in the VA population and help to explain the higher rates of OSA in the VA. The Pulmonary Sleep Program at the VA San Diego Healthcare System (VASDHS) started a patient database over twenty years ago for its home sleep apnea testing (HSAT) program. An analysis of ten years of diagnostic HSAT data was reported on over 12,500 patients in 2014. This analysis focuses on an update on the last four years and a comparison to 2014. Methods A retrospective review of the local clinical database of HSAT diagnostic testing was conducted. The database contained 8,928 sleep studies from 2018 to 2022. Invalid studies were removed. Only initial diagnostic studies were included. Results Of the 8,928 diagnostic studies, 2,564 tested negative for OSA (28.7%). Of the 6,364 positive cases, 43.3% were mild, 28.8% were moderate, and 27.9% were severe. The percentages of each OSA severity level were stable across these years (mild: 42%-45%; moderate: 27%-32%; and severe: 26%-29%). The mean AHI did not significantly change from 2018 (24.1+/-20.6) to 2022 (25.1+/-21.2) (overall mean 24.4+/-20.3; range: 5-140). Conclusion Based on this extensive retrospective review of diagnostic testing at a single, high-volume HSAT program, it appears that OSA severity levels have stabilized over these past five years, with the highest percentage of mild OSA cases (~43%) followed by moderate (~29%) and severe (~28%). Our facility saw a significant decrease in severe cases (60% to 30%) and increase in mild OSA (7% to 35%) from 2004 to 2013. Today, over 4 out of every 10 new OSA diagnoses are mild, which has significant treatment implications given that adherence to the two primary medical device options is suboptimal. Identifying patient-centered treatment options for these patients is a priority. Support (if any) VA RR&D D2651-R and the Sleep Medicine Section of the VA San Diego Healthcare System supported this project.

Association of Backscattered Ultrasonographic Imaging of the Tongue With Severity of Obstructive Sleep Apnea in Adults

Determining interventions to manage obstructive sleep apnea (OSA) depends on clinical examination, polysomnography (PSG) results, and imaging analysis. There remains the need of a noninvasive and cost-effective way to correlate relevant upper airway anatomy with severity of OSA to direct treatment and optimize outcome. To determine whether backscattered ultrasonographic imaging (BUI) analysis of the tongue is associated with severity of OSA in adults. In this prospective, single-center, diagnostic study of a consecutive series of patients (aged ≥18 years) at a sleep surgery clinic, the 89 included patients had a PSG within 3 years at the time of ultrasonography and BUI analysis between July 2020 and March 2022. Patients were excluded if body mass index had changed more than 10% since time of PSG. A standardized submental ultrasonographic scan with laser alignment was used with B-mode and BUI analysis applied to the tongue. The B-mode and BUI intensity were associated with the apnea-hypopnea index (AHI), a measure of severity of apnea from normal (no OSA) to severe OSA. Ultrasonography and PSG. The main outcomes were BUI parameters and their association with AHI value. Eighty-nine patients were included between July 2020 and March 2022. A total of 70 (78.7%) male patients were included; and distribution by race and ethnicity was 46 (52%) White participants, 22 (25%) Asian participants, and 2 (2%) African American participants, and 19 (21%) others. Median (IQR) age was 37.0 (29.0-48.3) years; median (IQR) BMI was 25.3 (23.2-29.8); and median (IQR) AHI was 11.1 (5.6-23.1) events per hour. At the middle to posterior tongue region, the 4 OSA severity levels explained a significant portion of the BUI variance (η2 = 0.153-0.236), and a significant difference in BUI values was found between the subgroups with AHI values of less than 15 (no OSA and mild OSA) and greater than or equal to 15 (moderate OSA and severe OSA) events per hour. The echo intensity showed no significant differences. The BUI values showed a positive association with AHI, with a Spearman correlation coefficient of up to 0.43. Higher BUI values remained associated with higher AHI after correction for the covariates of BMI and age. In this prospective diagnostic study, standardized BUI analysis of the tongue was associated with OSA severity. With the practicality of ultrasonography, this analysis is pivotal in connecting anatomy with physiology in treatment planning for patients with OSA.

Left ventricular function assessed with two-dimensional speckle tracking and three-dimensional echocardiography in patients with obstructive sleep apnea

Objective: To investigate left ventricular function by two-dimensional (2D) speckle tracking and three-dimensional (3D) echocardiography in obstructive sleep apnea (OSA) patients. Subject and method: Patients who were diagnosed with OSA based on the results of full polysomnography and had apnea- hypopnea index values > 5 events/hour of sleep were included in the study. All patient underwent polysomnography recording. Comprehensive 2-dimensional echocardiography were performed to all subjects according to American Society of Echocardiography guidelines [5], followed by real-time 3D echocardiography. Result: From June 2021 to June 2022, there were 46 OSA patients were included in the study. Mean age of OSA patients 51.3 ± 11.7 years. Men 63% (29/46), women 37% (15/46). The apnea-hypopnea index, percentage of total sleep time with an oxygen saturation level < 90%, arousal index increased in OSA patients. E/E’ increased (12.6 ± 4.3), GLS mildly reduced (-16.5 ± 3.8%) while other echocardiographic parameters were in normal range. Results of ANOVA showed an increase in the following LA parameters on 3D echocardiography according to the levels of OSA severity: LV mass index, 3D maximum LA volume, 3D maximum LA volume index, E/E’ were significantly higher in the severe OSA group compared with the moderate OSA and the mild OSA (p<0.05 for all). There was a decrease in the E’ velocity and global longitudinal strain (GLS) (p<0.05 for all). Conclusion: In patients with obstructive sleep apnea, the severity of OSA was associated with lower E′ velocity, greater E/E′ ratio, larger 3D LV volume, greater LV mass index, and poorer left ventricular global longitudinal strain.

Assessment of left atrial volume and function by real-time three-dimensional echocardiography in obstructive sleep apnea

Aims: To evaluate the volumetric and functional abnormalities of the left atrium (LA) using real-time three-dimensional echocardiography (RT3DE) in patients with obstructive sleep apnea (OSA) compared with a control group.
 Methods: Patients with OSA and age-and sex-matched control subjects were recruited to the study in Bach Mai hospital. All patients underwent polysomnography recording. Comprehensive 2-dimensional echocardiographic (2DE) was performed to all subjects according to American Society of Echocardiography guidelines, followed by RT3DE. Maximum LA volume, maximum LA volume indexed for body surface area (BSA), minimum LA volume, and LA volume before contraction were obtained.
 Results: From June 2021 to June 2022, there were 46 OSA patients and 30 controls were included in the study. Mean age of OSA patients 51.3 ± 11.7. Men accounted for 63% (29/46). The apnea-hypopnea index, percentage of total sleep time with an oxygen saturation level < 90%, arousal index were higher in OSA patients compared with controls. There was a significantly higher 2D LA volume index and E/E’ ratio in the OSA patients compared with the controls (p<0.05). RT3D echocardiography results showed a larger maximum LA volume indexed for BSA, LA volume before atrial contraction, total LA systolic volume, and active LA systolic volume and a higher active LA ejection fraction in the OSA patients compared with controls (p<0.05 for all). There was an increase in the following LA parameters on RT3DE according to the levels of OSA severity: 3D maximum LA volume, 3D maximum LA volume index, 3D LA volume before LA contraction; 3D LA total systolic volume, 3D LA active systolic volume, 3D LA active systolic ejection fraction (p<0.05). On the other hand, 2D maximum LA volume and 2D maximum LA volume index did not differ according to the severity of OSA (p>0.05)
 Conclusion: There was a significant increase in LA volume and augmentation of LA active systolic function in patients with OSA. The severity of OSA was directly linked with the enlargement of 3D LA volume while this was not seen with 2D LA volume.

Three-dimensional orthodontic imaging in children across the age spectrum and correlations with obstructive sleep apnea.

To determine baseline facial convexity measurements in children with obstructive sleep apnea (OSA) across the age spectrum. Polysomnogram, stereophotogrammetry, and biometric data were collected from children aged 0-18 years who were being investigated for OSA. Analyses evaluated differences in facial convexity according to OSA severity and other sleep parameters, while adjusting for age, ethnicity, and sex. Ninety-one children, aged 0.05-16.02 years, met the inclusion criteria for this study. Initial analysis showed that the logarithm of age had a significant effect on facial convexity (P = 8.3·10-7) with significant effects of sex (P = 1.3·10-2), while excluding OSA. Ordinal logistic regression taking into consideration age, sex, weight, height, and ethnicity with OSA severity categorized as obstructive apnea-hypopnea index negative, mild, moderate, or severe showed that facial convexity was associated with OSA severity (P = 2.2·10-3); an increasing obtuse angle of convexity increased the tendency to be classified as having severe OSA. Using three-dimensional imaging, we found an added impact of infancy on changes of facial convexity with age. While modeling could describe facial convexity without any OSA-associated sleep parameters, differences in facial convexity were present among groups with different levels of OSA severity adjusted for growth (age, weight, and height), sex, and ethnicity. The method provides a safer and cheaper alternative to other medical imaging techniques in children and holds potential for future use in studies of craniofacial structure. Tyler G, Machaalani R, Waters KA. Three-dimensional orthodontic imaging in children across the age spectrum and correlations with obstructive sleep apnea. J Clin Sleep Med. 2023;19(2):275-282.

A 2D convolutional neural network to detect sleep apnea in children using airflow and oximetry

The gold standard approach to diagnose obstructive sleep apnea (OSA) in children is overnight in-lab polysomnography (PSG), which is labor-intensive for clinicians and onerous to healthcare systems and families. Simplification of PSG should enhance availability and comfort, and reduce complexity and waitlists. Airflow (AF) and oximetry (SpO2) signals summarize most of the information needed to detect apneas and hypopneas, but automatic analysis of these signals using deep-learning algorithms has not been extensively investigated in the pediatric context. The aim of this study was to evaluate a convolutional neural network (CNN) architecture based on these two signals to estimate the severity of pediatric OSA. PSG-derived AF and SpO2 signals from the Childhood Adenotonsillectomy Trial (CHAT) database (1638 recordings), as well as from a clinical database (974 recordings), were analyzed. A 2D CNN fed with AF and SpO2 signals was implemented to estimate the number of apneic events, and the total apnea-hypopnea index (AHI) was estimated. A training-validation-test strategy was used to train the CNN, adjust the hyperparameters, and assess the diagnostic ability of the algorithm, respectively. Classification into four OSA severity levels (no OSA, mild, moderate, or severe) reached 4-class accuracy and Cohen's Kappa of 72.55% and 0.6011 in the CHAT test set, and 61.79% and 0.4469 in the clinical dataset, respectively. Binary classification accuracy using AHI cutoffs 1, 5 and 10 events/h ranged between 84.64% and 94.44% in CHAT, and 84.10%–90.26% in the clinical database. The proposed CNN-based architecture achieved high diagnostic ability in two independent databases, outperforming previous approaches that employed SpO2 signals alone, or other classical feature-engineering approaches. Therefore, analysis of AF and SpO2 signals using deep learning can be useful to deploy reliable computer-aided diagnostic tools for childhood OSA.

BENEFICIAL EFFECTS OF WEIGHT LOSS ON CENTRAL PRESSURE, ARTERIAL STIFFNESS AND SYMPATHETIC TONE IN OBESE SUBJECTS WITH OBSTRUCTIVE SLEEP APNEA

Objective: Obstructive sleep apnea (OSA) is known to be an independent cardiovascular risk factor. The presence of OSA and obesity may have synergistic effects on the progression of cardiovascular disease. To evaluate the effect of weight loss on vascular disease in obese patients with different OSA severity levels. Design and method: Individuals of both sexes, age between 40 and 70 years and body mass index (BMI) > = 30 and < 40 kg/m2 were submitted to nutritional intervention for 8 weeks with restriction of 800 kcal/day. Pulse wave velocity (PWV) and central hemodynamic parameters were evaluated by oscillometry (Mobil-O-Graph) and sympathetic tone by heart rate variability (HRV). The sleep study was performed using a portable home device (WatchPAT). Results: Patients (n = 70) were divided into two groups based on the apnea-hypopnea index (AHI): mild-absent (MA) group (AHI< 15 events/h, n = 25) with mean age 50 ± 6 years, BMI = 34.5 ± 2.8 kg/m2 and moderate-severe (MS) group (AHI> = 15 events/h, n = 45) with mean age 54 ± 8 years, BMI = 33.8 ± 2.6 kg/m2. After the intervention, weight loss was similar in both groups (-3.20 vs -3.32 kg). Both groups showed a significant reduction in anthropometric parameters. The MS group showed a significant reduction in peripheral systolic blood pressure (BP) (127 ± 15 vs 122 ± 15 mmHg, p = 0.007) and the MA group demonstrated a significant reduction in diastolic BP (78 ± 6 vs 75 ± 6 mmHg, p = 0.046). Central systolic pressure (116 ± 13 vs 107 ± 11 mmHg, p = 0.007), PWV (6.9 ± 0.6 vs 6.7 ± 0.6 m/s, p = 0.002) and vascular age (48 ± 6 vs 46 ± 6 years, p = 0.020) were significantly reduced only in the MA group. This group also demonstrated a significant increase in the RR interval (838 ± 225 vs 948 ± 130 ms, p = 0.032) with a reduction in heart rate (69 ± 5 vs 64 ± 8 bpm, p = 0.015). Conclusions: A small weight loss in a short period of time was able to promote more beneficial effects in obese individuals with non-severe sleep apnea including favorable outcomes on central hemodynamic parameters, arterial stiffness and autonomic activity.

0707 Effects of obstructive sleep apnea on Anti-SARS-COV-2 igg levels in older adults vaccinated against covid-19

Abstract Introduction Sleep disorders and sleep deprivation induces decreased antibody response following vaccination for different viral diseases (including H1N1, influenza and hepatitis A). The same has been speculated for COVID-19. This study aimed to assess whether obstructive sleep apnea (OSA) reduces antibody levels after COVID-19 vaccination among older adults. Methods This was a convenience-sample study composed of older adults (≥60 years old). Those who underwent polysomnography at the Sleep Institute (São Paulo, Brazil) and received complete COVID-19 vaccination schedule were considered eligible. Individuals with previous diagnosis of COVID-19, less than 15 days between vaccination and IgG testing, or CPAP use in the last 3 months were excluded. Anti-SARS-CoV-2 IgG levels were measured using a chemiluminescence assay. The participants were distributed in the following groups, according to their apnea-hypopnea index (AHI): no/mild OSA (AHI<15), moderate OSA (AHI≥15 and <30) and severe OSA (AHI≥30). The effects of OSA on IgG levels (ANOVA), the correlation between IgG levels and AHI (Spearman’s correlation test) and the association between serostatus (positive vs. negative) and OSA severity levels (X2 test) were analysed. Results were considered as statistically significant when p<0.05. Results The sample included 122 older adults (median age 72.0 - IQR: 5.7), of whom 35 (28.6%) had AHI no/mild OSA; 31 (25.4%) had moderate OSA, and 56 (45.9%) had severe OSA. Oxford/AstraZeneca was the most referred vaccine (n=111, 91.0%), followed by CoronaVac (n=9, 9.0%). Seropositive status (IgG count ≥50.0 AU/mL) was observed in 90.2% of the participants and the median IgG levels in the complete sample was 273.0 AU/ML (IQR: 744.0). No/mild, moderate and severe OSA groups presented IgG levels of 482.0 (IQR: 677.0), 285 (IQR: 884.0) and 181.0 (IQR: 598.0), respectively, with no statistical difference them (p=0.606). There was no statistically significant correlation between AHI index and IgG levels (Spearman’s rho=-0.169, p=0.063) and no significant association between serostatus and OSA severity groups (X2=0.912; p=0.634). Conclusion Anti-SARS-CoV-2 IgG levels after vaccination are not significantly affected by OSA among older adults. Thus, despite being at higher risk for severe cases, OSA does not decrease the antibody response following vaccination against COVID-19. Support (If Any) AFIP, CNPq, CAPES

Influence of nocturnal insomnia symptoms on obstructive sleep apnea diagnosis in a clinical referral cohort.

To assess whether nocturnal symptoms of insomnia influence the prevalence of obstructive sleep apnea (OSA) in a clinical referral cohort. Insomnia was defined by the presence of 1 or more of the following complaints: difficulty initiating sleep, difficulty maintaining sleep, and/or early morning awakenings. OSA severity was based on an apnea-hypopnea index: ≥ 5.0 events/h (any OSA), ≥ 15.0 events/h (moderate/severe OSA), and ≥ 30 events/h (severe OSA). Multivariate logistic regression analysis was used to evaluate predictive factors for OSA diagnosis and severity. Overall, 12,021 outpatients referred for polysomnography were grouped into 2 cohorts: without insomnia (58.2%) and with insomnia (41.8%). Individuals without insomnia had a higher prevalence of OSA than those with insomnia (P < .001, for all OSA severity levels). The presence of insomnia was negatively associated with diagnosis of any OSA (adjusted odds ratio: 0.852; 95% confidence interval: 0.769-0.944), moderate/severe OSA (adjusted odds ratio: 0.819; 95% confidence interval: 0.751-0.892), and severe OSA (adjusted odds ratio: 0.816; 95% confidence interval: 0.746-0.892). Moreover, the number of nocturnal symptoms of insomnia was associated with a lower likelihood of OSA, even when adjusted for other confounders such as sex, age, body mass index, neck circumference, excessive daytime sleepiness, hypertension, and type 2 diabetes mellitus. In this present study that included a large sample of consecutive outpatients, there was an inverse relationship between the number of nocturnal symptoms of insomnia and OSA diagnosis. Duarte RLM, Magalhães-da-Silveira FJ, Gozal D. Influence of nocturnal insomnia symptoms on obstructive sleep apnea diagnosis in a clinical referral cohort. J Clin Sleep Med. 2022;18(5):1271-1278.

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

Rationale: There are at least four key pathophysiological endotypes that contribute to obstructive sleep apnea (OSA) pathophysiology. These include 1) upper-airway collapsibility (Pcrit); 2) arousal threshold; 3) loop gain; and 4) pharyngeal muscle responsiveness. However, an easily interpretable model to examine the different ways and the extent to which these OSA endotypes contribute to conventional polysomnography-defined OSA severity (i.e., the apnea-hypopnea index) has not been investigated. In addition, clinically deployable approaches to estimate OSA endotypes to advance knowledge on OSA pathogenesis and targeted therapy at scale are not currently available.Objectives: Develop an interpretable data-driven model to 1) determine the different ways and the extent to which the four key OSA endotypes contribute to polysomnography-defined OSA severity and 2) gain insight into how standard polysomnographic and clinical variables contribute to OSA endotypes and whether they can be used to predict OSA endotypes.Methods: Age, body mass index, and eight polysomnography parameters from a standard diagnostic study were collected. OSA endotypes were also quantified in 52 participants (43 participants with OSA and nine control subjects) using gold-standard physiologic methodology on a separate night. Unsupervised multivariate principal component analyses and data-driven supervised machine learning (decision tree learner) were used to develop a predictive algorithm to address the study objectives.Results: Maximum predictive performance accuracy of the trained model to identify standard polysomnography-defined OSA severity levels (no OSA, mild to moderate, or severe) using the using the four OSA endotypes was approximately twice that of chance. Similarly, performance accuracy to predict OSA endotype categories ("good," "moderate," or "bad") from standard polysomnographic and clinical variables was approximately twice that of chance for Pcrit and slightly lower for arousal threshold.Conclusions: This novel approach provides new insights into the different ways in which OSA endotypes can contribute to polysomnography-defined OSA severity. Although further validation work is required, these findings also highlight the potential for routine sleep study and clinical data to estimate at least two of the key OSA endotypes using data-driven predictive analysis methodology as part of a clinical decision support system to inform scalable research studies to advance OSA pathophysiology and targeted therapy for OSA.

Validation of the GOAL Questionnaire as an Obstructive Sleep Apnea Screening Instrument in Bariatric Surgery Candidates: a Brazilian Single-Center Study.

To validate the GOAL questionnaire as obstructive sleep apnea (OSA) screening tool in bariatric surgery (BS) patients and compare it with other existing instruments. Before performing full polysomnography (PSG), all participants were screened for OSA with the following instruments: GOAL, STOP, STOP-Bang, No-Apnea, NoSAS, and Epworth Sleepiness Scale (ESS). Discrimination was assessed by area under the curve (AUC), while predictive parameters were calculated by contingency tables. Correlation was evaluated by the Spearman correlation coefficient (r). OSA severity was classified based on the apnea/hypopnea index(AHI): ≥ 5.0/h (OSA≥ 5), ≥ 15.0/h (OSA≥ 15), and ≥ 30.0/h (OSA≥ 30). Overall, 814 BS individuals (70.8% of females) were enrolled. We found a high prevalence of OSA≥ 5 (82.6%), OSA≥ 15 (60.0%), and OSA≥ 30 (38.8%). GOAL questionnaire was positively correlated with the AHI (r = 0.570, p < 0.001). Using a score ≥ 2 to classify patients at high risk of OSA, GOAL questionnaire had sensitivities ranging from 73.7 to 89.2% and specificities ranging from 78.2 to 51.0% for predicting OSA≥ 5, OSA≥ 15, and OSA≥ 30. At all OSA severity levels, GOAL exhibited similar and non-inferior discrimination when compared with STOP-Bang, No-Apnea, and NoSAS (all p values > 0.05), and performed significantly better than STOP and ESS (all p values < 0.001). In a cohort of BS patients, GOAL, No-Apnea, STOP-Bang, and NoSAS, but not ESS and STOP, enable satisfactory discrimination as OSA screening instruments at all disease severity levels.

1177 Cost Effectiveness Of Diagnostic Approaches To Sleep Apnea Evaluation During Inpatient Rehabilitation For Moderate To Severe TBI

Abstract Introduction Recent work has highlighted prevalent obstructive sleep apnea (OSA) after traumatic brain injury (TBI) when patients are vulnerable to disruption of neural repair. The recently completed clinical trial comparing screening and diagnostic tools for OSA during inpatient rehabilitation provided the opportunity to conduct economic modeling of phased approaches using actual trial findings to address one perspective (the payor) on the value of phased testing. Methods A cost-effectiveness analysis of four phased approaches to OSA diagnosis including initial utilization of portable sleep monitoring [HSAT] regardless of pre-test probability, determination of pre-test probability using two prediction models [STOPBANG, MAPI], and initial assessment using Level 1 polysomnography was conducted. The analyses were modeled assuming all participants were considered high risk thus a negative screen or portable diagnostic test would result in a participant being referred for Level 1 polysomnography. The cost aversion used in analyses were derived from a recent white paper on the economic modeling of untreated OSA. Trial data from 214 participants were used in analyses (mean age 44 [SD 18], 82% male, 75% white, with primarily motor-vehicle related injury [44%] and falls [33%] with a sample mean emergency department Glasgow Coma Scale of 8 (SD 5). Results At AHI ≥15 (33.6%), the prediction models (STOPBANG [-$5,291], MAPI [-$5,262]) resulted in greater cost savings and effectiveness relative to the HSAT approach (-$5,210) and initial use of Level 1 PSG (-$5,011). Sensitivity analyses at AHI ≥5 (70.1%) revealed the initial use of HSAT (-$6,322.85) relative to the prediction models (MAPI [-$6,249.71], STOPBANG [-$6,237) and initial assessment with Level 1 PSG (-$5,977) resulted in greater savings and cost effectiveness. Conclusion The high rates of sleep apnea after TBI highlight the importance of accurate diagnosis and treatment of this comorbid disorder. However, financial and practical barriers exist to obtaining an earlier diagnosis during inpatient rehabilitation hospitalization. Diagnostic cost savings are demonstrated across all phased approaches and OSA severity levels with the most cost-effective approach varying by incidence of OSA. Support PCORI (CER-1511-33005), GDHS (W91YTZ-13-C-0015; HT0014-19-C-0004)) for DVBIC, NIDILRR (NSDC Grant # 90DPTB00070, #90DP0084, 90DPTB0013-01-00, 90DPTB0008, 90DPT80004-02).

Sex differences in brain oscillatory activity during sleep and wakefulness in obstructive sleep apnea.

Epidemiological studies consistently show a male predominance in obstructive sleep apnea (OSA). Hormonal differences, breathing control, upper airway anatomy and fat distribution have been proposed as causes of gender differences in OSA. Clinical manifestations are accentuated in men, although white matter structural integrity is affected in women.To the best of our knowledge, no previous studies have explored gender differences in the electrical brain activity features of OSA. Polysomnography was performed on43 patientswith untreated OSA (21 women, 22 men),and power spectral density (1-50Hz) was compared between groups across sleep and wakefulnessat two levels of OSA severity. Severe versus moderate OSA showed decreased power for fast frequencies (25-29Hz) during wakefulness. OSA men displayed decreased power of a large frequency range (sigma, beta and gamma) during sleep compared with women. Comparisons of men with severe versus moderate OSA presented significantly decreased sigma power during non-rapid eye movement (NREM) sleep, but significantly increased delta activity during REM sleep. Meanwhile, women with severe versus moderate OSA showed no significant power differences in any condition. These findings indicated a different evolution of brain oscillations between OSA men and women with significant impairment of brain activity related to cognitive processes. Our study emphasizes the importance of understanding the differential effects of sleep disorders on men and women in order to develop more precise diagnostic criteria according to gender, including quantitative electroencephalogram (EEG) analysis tools.

Obstructive sleep apnea in 2-6year old children referred for adenotonsillectomy.

Adenotonsillectomy is one of the most common surgical procedures performed in children. The indications for surgery are either frequent recurrent throat infections or hypertrophy of the tonsils/adenoid vegetation, which can cause obstructive sleep apnea (OSA). There is disagreement regarding the need for sleep studies before adenotonsillectomy to confirm a diagnosis of OSA. Several studies have evaluated questionnaires and physical examination as tools to identify OSA, with conflicting results. The aim of this study was to evaluate the prevalence of OSA among children referred for adenotonsillectomy and whether questionnaires or physical examination can help identify OSA. This is a prospective cohort study of children aged 2-6 years, referred for adenotonsillectomy. Polysomnography and an otorhinological examination were performed. Tonsillar size and the oral cavity were graded using Friedman's classification and Mallampati score, respectively. The Pediatric Sleep Questionnaire (PSQ) and OSA-18 were also completed. 100 children were included. The prevalence of OSA was 87%, with 52% having moderate to severe OSA. The usefulness of the PSQ and OSA-18 for detecting OSA was evaluated using multiple cutoff points, but none yielded acceptable values for both sensitivity and specificity. In logistic regression analyses predicting different levels of OSA severity, age, Friedman tonsillar size and Mallampati score were weakly associated with OSA. The prevalence of OSA is high among children referred for adenotonsillectomy and questionnaires and clinical characteristics are not sensitive enough to detect the presence or severity of OSA.

Automated Non-Contact Detection of Head and Body Positions During Sleep

Obstructive sleep apnea (OSA) is a respiratory disorder characterized by interruption to breathing during sleep. Usually, the OSA is more severe in the supine sleeping position. Recent studies also demonstrated that the head position may play an important role in the pathophysiology of the OSA. Therefore, monitoring the sleeping body and the head position has high clinical importance to optimize the treatment of the OSA. In this paper, three machine learning approaches were used to detect the head position during sleep in infrared images. In the first two methods, supervised classifiers were trained to estimate the head position based on different feature sets extracted from infrared images. In the third method, three different convolutional neural network (CNN) structures (ResNet50, MobileNet, and Darknet19) were trained to detect the head position during sleep. To detect the body position, the same CNN architectures were trained on infrared images. Overnight sleeping data (sleep duration =5±1h) from 50 participants (age: 53 ± 15 years, BMI: 29 ± 6 kg/m2, and 30 men/20 women) with various levels of OSA severity as measured by the apnea-hypopnea index (AHI =25±29 events/h and OSA severity: 12 normal, 13 mild, 11 moderate, and 14 severe) were collected for this paper. The models were trained on the data collected in one laboratory room from half of the participants and tested on the data from the other half collected in a different laboratory room. The best performing model (Darknet19) correctly estimated the lateral versus supine head position with 92% accuracy and 94% F1-Score and the lateral versus supine body position with 87% accuracy and 87% F1-Score.

Pediatric pulse oximetry-based OSA screening at different thresholds of the apnea-hypopnea index with an expression of uncertainty for inconclusive classifications

BackgroundAssessments of pediatric obstructive sleep apnea (OSA) are underutilized across Canada due to a lack of resources. Polysomnography (PSG) measures OSA severity through the average number of apnea/hypopnea events per hour (AHI), but is resource intensive and requires a specialized sleep laboratory, which results in long waitlists and delays in OSA detection. Prompt diagnosis and treatment of OSA are crucial for children, as untreated OSA is linked to behavioral deficits, growth failure, and negative cardiovascular consequences. We aim to assess the performance of a portable pediatric OSA screening tool at different AHI cut-offs using overnight smartphone-based pulse oximetry. Material and methodsFollowing ethics approval and informed consent, children referred to British Columbia Children's Hospital for overnight PSG were recruited for two studies including 160 and 75 children, respectively. An additional smartphone-based pulse oximeter sensor was used in both studies to record overnight pulse oximetry [SpO2 and photoplethysmogram (PPG)] alongside the PSG.Features characterizing SpO2 dynamics and heart rate variability from pulse peak intervals of the PPG signal were derived from pulse oximetry recordings. Three multivariate logistic regression screening models, targeted at three different levels of OSA severity (AHI ≥ 1, 5, and 10), were developed using stepwise-selection of features using the Bayesian information criterion (BIC). The “Gray Zone” approach was also implemented for different tolerance values to allow for more precise detection of children with inconclusive classification results. ResultsThe optimal diagnostic tolerance values defining the “Gray Zone” borders (15, 10, and 5, respectively) were selected to develop the final models to screen for children at AHI cut-offs of 1, 5, and 10. The final models evaluated through cross-validation showed good accuracy (75%, 82% and 89%), sensitivity (80%, 85% and 82%) and specificity (65%, 79% and 91%) values for detecting children with AHI ≥ 1, AHI ≥ 5 and AHI ≥ 10. The percentage of children classified as inconclusive was 28%, 38% and 16% for models detecting AHI ≥ 1, AHI ≥ 5, and AHI ≥ 10, respectively. ConclusionsThe proposed pulse oximetry-based OSA screening tool at different AHI cut-offs may assist clinicians in identifying children at different OSA severity levels. Using this tool at home prior to PSG can help with optimizing the limited resources for PSG screening. Further validation with larger and more heterogeneous datasets is required before introducing in clinical practice.