Apnea-hypopnea Index Thresholds Research Articles

Deep learning approaches for assessing pediatric sleep apnea severity through SpO2 signals

Pediatric Sleep Apnea–Hypopnea (SAH) presents a significant health challenge, particularly in diagnostic contexts, where conventional Polysomnography (PSG) testing, although effective, can be distressing for children. Addressing this, our research proposes a less invasive method to assess pediatric SAH severity by analyzing blood oxygen saturation (SpO2) signals. We adopted two advanced deep learning architectures, namely ResNet-based and attention-augmented hybrid CNN-BiGRU models, to process SpO2 signals in a one-dimensional (1D) format for Apnea–Hypopnea Index (AHI) estimation in pediatric subjects. Employing the CHAT dataset, which includes 844 SpO2 signals, the data was partitioned into training (60%), testing (30%), and validation (10%) sets. A predefined validation subset was randomly selected to ensure the models' robustness via a threefold cross-validation approach. Comparative analysis revealed that while the ResNet model attained an average accuracy of 72.9% across four SAH severity categories with a kappa score of 0.57, the CNN-BiGRU-Attention model demonstrated superior performance, achieving an average accuracy of 75.95% and a kappa score of 0.63. This distinction underscores our method's efficacy in both estimating AHI and categorizing SAH severity levels with notable precision. Further, to evaluate diagnostic capabilities, the models were benchmarked against common AHI thresholds (1, 5, and 10 events/hour) in each test fold, affirming their effectiveness in identifying pediatric SAH. This study marks a significant advance in the field, offering a non-invasive, child-friendly alternative for pediatric SAH diagnosis. Although challenges persist in accurately estimating AHI, particularly in severe cases, our findings represent a critical stride towards improving diagnostic processes in pediatric SAH.

Machine Learning Models to Enhance the Berlin Questionnaire Detection of Obstructive Sleep Apnea in at-Risk Patients

The Berlin questionnaire (BQ), with its ten questions, stands out as one of the simplest and most widely implemented non-invasive screening tools for detecting individuals at a high risk of Obstructive Sleep Apnea (OSA), a still underdiagnosed syndrome characterized by the partial or complete obstruction of the upper airways during sleep. The main aim of this study was to enhance the diagnostic accuracy of the BQ through Machine Learning (ML) techniques. A ML classifier (hereafter, ML-10) was trained using the ten questions of the standard BQ. Another ML model (ML-2) was trained using a simplified variant of the BQ, BQ-2, which comprises only two questions out of the total ten. A 10-fold cross validation scheme was employed. Ground truth was provided by the Apnea–Hypopnea Index (AHI) measured by Home Sleep Apnea Testing. The model performance was determined by comparing ML-10 and ML-2 with the standard BQ in the Receiver Operating Characteristic (ROC) space and using metrics such as the Area Under the Curve (AUC), sensitivity, specificity, and accuracy. Both ML-10 and ML-2 demonstrated superior performance in predicting the risk of OSA compared to the standard BQ and were also capable of classifying OSA with two different AHI thresholds (AHI ≥ 15, AHI ≥ 30) that are typically used in clinical practice. This study underscores the importance of integrating ML techniques for early OSA detection, suggesting a direction for future research to improve diagnostic processes and patient outcomes in sleep medicine with minimal effort.

Home monitoring for clinically suspected obstructive sleep apnea in pregnancy.

To determine if a home sleep apnea test (HSAT) using a type III portable monitor (PM), Nox-T3 (Nox Medical, Inc., Reykjavik, Iceland), detects obstructive sleep apnea in pregnant women. Ninety-two pregnant women (34.5 ± 4.3 years; gestational age 25.4 ± 8.9 weeks; body mass index 29.9 ± 4.7 kg/m2) with suspected obstructive sleep apnea underwent HSAT with the Nox-T3 PM followed by overnight polysomnography (PSG) and PM recording simultaneously in the laboratory within 1 week. PMs were scored automatically and manually using a 3% criteria and compared with PSGs scored by following guidelines. Apnea-hypopnea indexes were 8.56 ± 10.42, 8.19 ± 13.79, and 8.71 ± 14.19 events/h on HSAT, in-laboratory PM recording, and PSG (P = .955), respectively. Bland-Altman analysis of the apnea-hypopnea index on PSG vs HSAT showed a mean difference (95% confidence interval) of -0.15 (-1.83, 1.53); limits of agreement (± 2 SD) were -16.26 to 16.56 events/h. Based on a threshold apnea-hypopnea index ≥ 5 events/h, HSAT had 91% sensitivity, 85% specificity, 84% positive-predictive value, and 92% negative-predictive value compared with PSG. When comparing the simultaneous recordings, closer agreement was observed. Automated vs manual analysis of PM showed no significant difference. A type III PM had an acceptable failure rate and high diagnostic performance operating as a reasonable alternative for in-laboratory PSG in pregnant women. Wang J, Zhang C, Xu L, etal. Home monitoring for clinically suspected obstructive sleep apnea in pregnancy. J Clin Sleep Med. 2023;19(11):1951-1960.

STOP-Bang and Smartwatch’s Two-Step Approach for Obstructive Sleep Apnea Screening

Background and Objectives Obstructive sleep apnea (OSA) is associated with various health risks, including hypertension, cerebrovascular disease, myocardial infarction, diabetes, cognitive impairment, and all-cause mortality. While overnight polysomnography (PSG) is the gold standard for diagnosing OSA, it is costly and time-consuming. The STOP-Bang questionnaire is a convenient tool for OSA screening, but its high sensitivity comes at the expense of low specificity. The purpose of this study was to investigate the usefulness of combining the STOP-Bang questionnaire and a smartwatch capable of measuring oxygen saturation in screening for OSA.Subjects and Method Of the patients scheduled for PSG due to OSA, 109 patients voluntarily participated in the study by filling out a STOP-Bang questionnaire during their first visit and wearing a smartwatch during PSG.Results There were 80 males and 29 females, with the patients’ mean age of 45±13.3 years. Based on the apnea-hypopnea index (AHI), 19 patients (17.4%) were normal, 28 (25.7%) had mild OSA, 23 (21.1%) had moderate OSA, and 39 (35.8%) had severe OSA. When using the AHI threshold of AHI ≥15/h, the STOP-Bang alone showed sensitivity of 85.5% and specificity of 61.7%. Combining the STOP-Bang questionnaire with a smartwatch resulted in a slight decrease in sensitivity and a significant increase in specificity, yielding the values of 80.5% and 84.4%, respectively.Conclusion A two-step approach using the STOP-Bang and a smartwatch was implemented to enhance the diagnostic accuracy of screening OSA.

0475 Optimal Thresholds for Split-Night Polysomnography with Transitions in Laboratory Hypopnea Scoring Criteria

Abstract Introduction Performance of split-night polysomnography, the application and titration of positive airway pressure (PAP) therapy after an initial diagnostic interval, is a common practice in clinical sleep medicine. There is no currently recommended apnea-hypopnea index (AHI) threshold at which PAP should be applied during polysomnography. Differences in American Academy of Sleep Medicine (AASM) and Centers for Medicare and Medicaid Services (CMS) hypopnea scoring criteria and limitations on real-time hypopnea classification complicates identification of an optimal split-night threshold. Of particular concern is that patients may be “split” under AASM hypopnea rules but have insufficiently severe sleep disordered breathing under CMS rules to be diagnosed or treated appropriately. This study aimed to clarify optimal AHI thresholds for split night polysomnography in the context of a laboratory-wide transition from CMS to AASM hypopnea scoring criteria. Methods Effective October 1, 2021, our laboratory transitioned solely to scoring AASM-defined hypopneas, with additional post hoc identification of CMS hypopneas for reporting purposes. With this change, the split-night threshold was changed from 15 to 30/hr. All diagnostic polysomnograms (without PAP therapy) performed on adult patients through October 31, 2022, were retrospectively analyzed to clarify the effect of changing hypopnea scoring criteria in this context. Results 634 diagnostic polysomnograms were analyzed. An AHI threshold of 15/hr (using AASM hypopnea criteria) in the first two hours of sleep with at least 3 hours of time remaining for PAP titration would have resulted in 96 additional patients receiving treatment with PAP therapy. Among these, only one (1.04%) had CMS AHI below 5/hr. Sixty-eight of these patients had CMS AHI 5-15/hr during the first two hours of sleep. Among these, only eight patients did not have an additional comorbidity or symptom profile under CMS guidelines making them eligible for PAP therapy, however, none had CMS AHI >15/hr when the full night of sleep was analyzed. Conclusion Our study suggests there is little need to raise the split-night threshold when transitioning from CMS to AASM hypopnea criteria, and doing so substantially reduces the number of patients treated with PAP therapy in the sleep laboratory. Support (if any) None

Utilizing Envelope Analysis of a Nasal Pressure Signal for Sleep Apnea Severity Estimation.

Obstructive sleep apnea (OSA) severity assessment is based on manually scored respiratory events and their arbitrary definitions. Thus, we present an alternative method to objectively evaluate OSA severity independently of the manual scorings and scoring rules. A retrospective envelope analysis was conducted on 847 suspected OSA patients. Four parameters were calculated from the difference between the nasal pressure signal's upper and lower envelopes: average (AV), median (MD), standard deviation (SD), and coefficient of variation (CoV). We computed the parameters from the entirety of the recorded signals to perform binary classifications of patients using three different apnea-hypopnea index (AHI) thresholds (5-15-30). Additionally, the calculations were undertaken in 30-second epochs to estimate the ability of the parameters to detect manually scored respiratory events. Classification performances were assessed with areas under the curves (AUCs). As a result, the SD (AUCs ≥ 0.86) and CoV (AUCs ≥ 0.82) were the best classifiers for all AHI thresholds. Furthermore, non-OSA and severe OSA patients were separated well with SD (AUC = 0.97) and CoV (AUC = 0.95). Respiratory events within the epochs were identified moderately with MD (AUC = 0.76) and CoV (AUC = 0.82). In conclusion, envelope analysis is a promising alternative method by which to assess OSA severity without relying on manual scoring or the scoring rules of respiratory events.

Measuring Sleep Stages and Screening for Obstructive Sleep Apnea with a Wearable Multi-Sensor System in Comparison to Polysomnography.

To assess the performance of a wearable multi-sensor system (SensEcho) in comparison to polysomnography (PSG) in measuring sleep stages and searching for obstructive sleep apnea (OSA). Participants underwent overnight simultaneous monitoring using SensEcho and PSG in a sleep laboratory. SensEcho analyzed the recordings spontaneously, and PSG was assessed as per standard guidelines. The degree of snoring was evaluated according to the guidelines for the diagnosis and treatment of OSA hypopnea syndrome (2011 revision). The Epworth Sleepiness Scale (ESS) was used to assess general daytime sleepiness. This study included 103 Han Chinese, 91 of whom (age 39.02 ± 13.84 years, body mass index 27.28 ± 5.12 kg/m2, 61.54% male) completed the assessments. The measures of total sleep time (P = 0.198); total wake time (P = 0.182); shallow sleep (P = 0.297), deep sleep (P = 0.422), rapid eye movement sleep (P = 0.570), and awake (P = 0.336) proportions were similar between SensEcho and PSG. Using an apnea-hypopnea index (AHI) cutoff of ≥ 5 events/h, the SensEcho had 82.69% sensitivity and 89.74% specificity. Almost the same results were obtained at an AHI threshold of ≥ 15 events/h. Although the specificity increased to 94.67%, it decreased to 43.75% at an AHI cutoff of ≥ 30 events/h. This study demonstrated that SensEcho can be used to evaluate sleep status and screen for OSA. Nevertheless, improving the accuracy of its assessment of severe OSA and further testing its effectiveness in community and home environments is necessary.

Screening for obstructive sleep apnea: comparing the American Academy of Sleep Medicine proposed criteria with the STOP-Bang, NoSAS, and GOAL instruments.

We evaluated the performance of the 2017 American Academy of Sleep Medicine criteria (AASM2017) in screening obstructive sleep apnea (OSA) and compared them with 3 other validated instruments: NoSAS score, STOP-Bang, and GOAL questionnaires. From July 2019 to December 2021, 4,499 adults undergoing overnight polysomnography were included. The AASM2017 instrument considers an increased high risk for moderate-to-severe OSA in the presence of excessive daytime sleepiness and at least 2 of the following 3 criteria: loud snoring; observed apnea, gasping, or choking; and hypertension. OSA severity was based on polysomnography-derived apnea-hypopnea index cutoffs: 5.0 events/h, 15.0 events/h, and 30.0 events/h. Predictive performance was evaluated by the area under the curve and contingency tables. When screening for any OSA severity, AASM2017 displayed a sensitivity of 31.0-40.6% and a specificity of 80.8-89.6%. For all apnea-hypopnea index thresholds, AASM2017, unlike the GOAL, STOP-Bang, and NoSAS, exhibited superior specificity but markedly lower sensitivity. GOAL, STOP-Bang, and NoSAS, but not AASM2017 criteria, emerged as an adequate screening tool for any OSA severity (all areas under the curve > 0.7) and performed significantly better than AASM2017 in predicting any OSA severity (all P < .001). For all severity OSA levels, GOAL, STOP-Bang, and NoSAS displayed similar performance when compared to each other (all P > .05). GOAL, STOP-Bang, and NoSAS instruments, but not AASM2017 criteria, emerge as useful OSA screening tools in a large referral single-center clinical cohort. Duarte RLM, Magalhães-da-Silveira FJ, Gozal D. Screening for obstructive sleep apnea: comparing the American Academy of Sleep Medicine proposed criteria with the STOP-Bang, NoSAS, and GOAL instruments. J Clin Sleep Med. 2023;19(7):1239-1246.

Cross-sectional interrelationships between chronotype, obstructive sleep apnea and blood pressure in a middle-aged community cohort.

Chronotype is linked to adverse health measures and may have important associations with obstructive sleep apnea and blood pressure, but data are limited. This study aimed to determine the separate and combined associations of chronotype with obstructive sleep apnea and blood pressure in a middle-aged community population. Adults (n= 811) from the Raine Study (female=59.2%; age mean [range]=56.6 [42.1-76.6] years) were assessed for chronotype (Morningness-Eveningness Questionnaire), blood pressure and hypertension (doctor diagnosed or systolic blood pressure ≥ 140 mmHg and/or diastolic ≥ 90 mmHg), and obstructive sleep apnea at different in-laboratory apnea-hypopnea index thresholds (5, 10, 15 events per hr). Linear and logistic regression models examined relationships between chronotype and the presence and severity of obstructive sleep apnea, blood pressure, hypertension, and blood pressure stratified by obstructive sleep apnea severity at above-mentioned apnea-hypopnea index thresholds. Covariates included age, sex, body mass index, alcohol consumption, smoking, physical activity, sleep duration, anti-hypertensive medication, insomnia, and depressive symptoms. Most participants were categorised as morning (40%) or intermediate (43%), with 17% meeting criteria for evening chronotypes. Participants with apnea-hypopnea index ≥ 15 events per hr and morning chronotype had higher systolic (9.9 mmHg, p< 0.001) and a trend for higher diastolic blood pressure (3.4 mmHg, p= 0.07) compared with those with an evening chronotype, and higher systolic blood pressure compared with those with an intermediate chronotype (4.8 mmHg, p= 0.03). Across chronotype categories, no differences in systolic or diastolic blood pressure or odds of hypertension were found at apnea-hypopnea index thresholds of ≥ 5 or ≥ 10 events per hr. Among participants with apnea-hypopnea index ≥ 15 events per hr, systolic blood pressure is higher in those with a morning chronotype than evening and intermediate chronotypes. Assessment for morning chronotype may improve risk stratification for hypertension in patients with obstructive sleep apnea.

Robust Method for Screening Sleep Apnea With Single-Lead ECG Using Deep Residual Network: Evaluation With Open Database and Patch-Type Wearable Device Data

This paper proposes a robust method to screen patients with sleep apnea syndrome (SAS) using a single-lead electrocardiogram (ECG). This method consists of minute-by-minute abnormal breathing detection and apnea-hypopnea index (AHI) estimation. Heartbeat interval and ECG-derived respiration (EDR) are calculated using the single-lead ECG and used to train the models, including ResNet18, ResNet34, and ResNet50. The proposed method, using data from 1232 subjects, was developed with two open datasets and experimental data and evaluated using two additional open datasets and data acquired from an abdomen-attached wearable device (in total, data from 189 subjects). ResNet18 showed the best results, having an average Cohen's kappa coefficient of 0.57, in the abnormal breathing detection. Moreover, SAS patient classification, with 15 as the AHI threshold, yielded an average Cohen's kappa coefficient of 0.71. The results of patient classification were biased toward data from the wearable patch-type device, which may be influenced by different ECG waveforms. The proposed method is tuned with a sample of the data from the device, and the performance result of Cohen's kappa increased from 0.54 to 0.91 for SAS patient classification. Our method, proposed in this paper, achieved equivalent performance results with data recorded using an abdomen-attached wearable device and two open datasets used in previous studies, although the method had not used those data during model training. The proposed method could reduce the development costs of commercial software, as it was developed using open datasets, has robust performance throughout all datasets.

Evaluating Prediction Models of Sleep Apnea From Smartphone-Recorded Sleep Breathing Sounds

Breathing sounds during sleep are an important characteristic feature of obstructive sleep apnea (OSA) and have been regarded as a potential biomarker. Breathing sounds during sleep can be easily recorded using a microphone, which is found in most smartphone devices. Therefore, it may be easy to implement an evaluation tool for prescreening purposes. To evaluate OSA prediction models using smartphone-recorded sounds and identify optimal settings with regard to noise processing and sound feature selection. A cross-sectional study was performed among patients who visited the sleep center of Seoul National University Bundang Hospital for snoring or sleep apnea from August 2015 to August 2019. Audio recordings during sleep were performed using a smartphone during routine, full-night, in-laboratory polysomnography. Using a random forest algorithm, binary classifications were separately conducted for 3 different threshold criteria according to an apnea hypopnea index (AHI) threshold of 5, 15, or 30 events/h. Four regression models were created according to noise reduction and feature selection from the input sound to predict actual AHI: (1) noise reduction without feature selection, (2) noise reduction with feature selection, (3) neither noise reduction nor feature selection, and (4) feature selection without noise reduction. Clinical and polysomnographic parameters that may have been associated with errors were assessed. Data were analyzed from September 2019 to September 2020. Accuracy of OSA prediction models. A total of 423 patients (mean [SD] age, 48.1 [12.8] years; 356 [84.1%] male) were analyzed. Data were split into training (n = 256 [60.5%]) and test data sets (n = 167 [39.5%]). Accuracies were 88.2%, 82.3%, and 81.7%, and the areas under curve were 0.90, 0.89, and 0.90 for an AHI threshold of 5, 15, and 30 events/h, respectively. In the regression analysis, using recorded sounds that had not been denoised and had only selected attributes resulted in the highest correlation coefficient (r = 0.78; 95% CI, 0.69-0.88). The AHI (β = 0.33; 95% CI, 0.24-0.42) and sleep efficiency (β = -0.20; 95% CI, -0.35 to -0.05) were found to be associated with estimation error. In this cross-sectional study, recorded sleep breathing sounds using a smartphone were used to create reasonably accurate OSA prediction models. Future research should focus on real-life recordings using various smartphone devices.

Detecting obstructive sleep apnea by craniofacial image-based deep learning.

This study aimed to develop a deep learning-based model to detect obstructive sleep apnea (OSA) using craniofacial photographs. Participants referred for polysomnography (PSG) were recruited consecutively and randomly divided into the training, validation, and test groups for model development and evaluation. Craniofacial photographs were taken from five different angles (front, right 90° profile, left 90° profile, right 45° profile, and left 45° profile) and inputted to the convolutional neural networks. The neural networks extracted features from photographs and outputted the probabilities of the presence of the disease. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated using PSG diagnosis as the reference standard. These analyses were repeated using two apnea-hypopnea index thresholds (≥ 5 and ≥ 15events/h). A total of 393 participants were enrolled. Using the operating point with maximum sum of sensitivity and specificity, the model of the photographs exhibited an AUC of 0.916 (95% confidence interval [CI], 0.847-0.960) with a sensitivity of 0.95 and a specificity of 0.80 at an AHI threshold of 5 events/h; an AUC of 0.812 (95% CI, 0.729-0.878) with a sensitivity of 0.91 and a specificity of 0.73 at an AHI threshold of 15 events/h. The results suggest that combining craniofacial photographs and deep learning techniques can help detect OSAautomatically. The model may have potential utility as a tool to assess OSA probability in clinics or screen forOSA in the community.

A meta-analysis of diagnostic test performance of peripheral arterial tonometry studies.

The objective of this meta-analysis was to analyze agreement in apnea-hypopnea index (AHI) determination between peripheral arterial tonometry (PAT) and polysomnography (PSG) studies. Mean AHI bias and standard deviation extracted from Bland-Altman plots reported in studies were pooled in a meta-analysis, which was then used to calculate percentage errors of limit agreement in AHI determination by PAT using PSG AHI as the reference. Individual participant data (where reported in studies) were used to compute Cohen's kappa to assess agreement between PSG and PAT on sleep apnea severity and for computing the sensitivity and specificity of PAT at different AHI thresholds using PSG AHI as the reference. From 17 studies and 1,318 participants (all underwent simultaneous PSG and use of the WatchPAT device), a pooled mean AHI bias of 0.30 (standard error [SE], 0.74) and a WatchPAT AHI percentage error of 230% was calculated. The meta-analysis of Cohen's kappa for agreement between PSG and WatchPAT studies for classifying patients with no sleep apnea, mild, moderate, or severe sleep apnea severity was 0.45 (SE, 0.06), 0.29 (SE, 0.05), 0.25 (SE, 0.07), and 0.64 (SE, 0.05), respectively. At AHI thresholds of 5, 15 and 30 events/h, WatchPAT studies showed pooled sensitivities and specificities of 94.11% and 43.47%, 92.21% and 72.39%, and 74.11% and 87.10%, respectively. Likelihood ratios were not significant at any AHI threshold. The results of this meta-analysis suggest clinically significant discordance between WatchPAT and PSG measurements of AHI, significant sleep apnea severity misclassification by PAT studies, and poor diagnostic test performance. Iftikhar IH, Finch CE, Shah AS, Augunstein CA, Ioachimescu OC. A meta-analysis of diagnostic test performance of peripheral arterial tonometry studies. J Clin Sleep Med. 2022;18(4):1093-1102.

Noncontact Sleep Monitoring With Infrared Video Data to Estimate Sleep Apnea Severity and Distinguish Between Positional and Nonpositional Sleep Apnea: Model Development and Experimental Validation.

BackgroundSleep apnea is a respiratory disorder characterized by frequent breathing cessation during sleep. Sleep apnea severity is determined by the apnea-hypopnea index (AHI), which is the hourly rate of respiratory events. In positional sleep apnea, the AHI is higher in the supine sleeping position than it is in other sleeping positions. Positional therapy is a behavioral strategy (eg, wearing an item to encourage sleeping toward the lateral position) to treat positional apnea. The gold standard of diagnosing sleep apnea and whether or not it is positional is polysomnography; however, this test is inconvenient, expensive, and has a long waiting list.ObjectiveThe objective of this study was to develop and evaluate a noncontact method to estimate sleep apnea severity and to distinguish positional versus nonpositional sleep apnea.MethodsA noncontact deep-learning algorithm was developed to analyze infrared video of sleep for estimating AHI and to distinguish patients with positional vs nonpositional sleep apnea. Specifically, a 3D convolutional neural network (CNN) architecture was used to process movements extracted by optical flow to detect respiratory events. Positional sleep apnea patients were subsequently identified by combining the AHI information provided by the 3D-CNN model with the sleeping position (supine vs lateral) detected via a previously developed CNN model.ResultsThe algorithm was validated on data of 41 participants, including 26 men and 15 women with a mean age of 53 (SD 13) years, BMI of 30 (SD 7), AHI of 27 (SD 31) events/hour, and sleep duration of 5 (SD 1) hours; 20 participants had positional sleep apnea, 15 participants had nonpositional sleep apnea, and the positional status could not be discriminated for the remaining 6 participants. AHI values estimated by the 3D-CNN model correlated strongly and significantly with the gold standard (Spearman correlation coefficient 0.79, P<.001). Individuals with positional sleep apnea (based on an AHI threshold of 15) were identified with 83% accuracy and an F1-score of 86%.ConclusionsThis study demonstrates the possibility of using a camera-based method for developing an accessible and easy-to-use device for screening sleep apnea at home, which can be provided in the form of a tablet or smartphone app.

Baseline apnea-hypopnea index threshold and adenotonsillectomy consideration in children with OSA

ObjectivesAdenotonsillectomy (AT) is the first line of treatment for pediatric obstructive sleep apnea (OSA). In some treatment guidelines, children with moderate to severe OSA, defined as apnea-hypopnea index (AHI) ≥ 5, may be recommended AT regardless of symptoms. The differences in outcomes between children randomized to watchful waiting with supportive care (WWSC) or AT were compared based on baseline OSA severity threshold of AHI≥ 5. MethodsA secondary analysis of the Childhood Adenotonsillectomy Trial, a randomized controlled trial of children with OSA aged 5–9 years who underwent AT or WWSC, was performed. The primary outcome was the change in neurocognition measured by Developmental Neuropsychological Assessment (NEPSY). Secondary outcomes included changes in behavior, symptoms of OSA, and quality of life. Outcomes were measured at baseline and the seven-month follow-up after grouping children based on whether their AHI was greater than or equal to 5. Comparisons were performed using two-way analysis of covariance (ANCOVA) while controlling for age, sex and race. Differences in treatment effect were measured using Cohen's d. ResultsOf the 397 children included, 203 received WWSC and 194 underwent AT. The treatment effects on post-randomization changes in neurocognition, measured by NEPSY in children with AHI ≥5 (Cohen's d = 0.1 [95% CI, −0.1 to 0.4]) was not significantly different from children with AHI <5 (Cohen's d = 0.1 [95% CI, −0.1 to 0.4]). Furthermore, among children in the AT group alone, the effects of AT on post-treatment changes in NEPSY did not differ based on AHI threshold (Cohen's d = −0.06 [95% CI, −0.3 to 0.2]). Additionally, the treatment effects on post-randomization changes in behavior, symptoms, and quality of life did not vary based on AHI threshold. ConclusionThe outcomes of neurocognition, behavior, symptoms, and quality of life did not differ between children with OSA randomized to WWSC or AT based on OSA severity threshold alone. Additionally, the effects of AT on post-treatment outcomes did not differ based on AHI threshold.

Utility of the type 3 portable monitor for the diagnosis of sleep disordered breathing in patients with stable heart failure

Objective: To evaluate the utility of a type 3 portable monitor (PM) at home for the diagnosis of sleep disordered breathing (SDB) in patients with stable congestive heart failure (CHF). Methods: Seventy-six consecutive patients with CHF (61 males, 15 females, mean±standard deviation age (57.0±16.9) years) were enrolled from the sleep center of Peking university People's Hospital during January 2016 to January 2019, and underwent overnight, unattended home sleep apnea testing (HSAT) with a portable monitor followed by an overnight simultaneous polysomnogram (PSG) with in-laboratory portable monitor (in-lab PM) recording within one week. The consistency of apnea hypopnea index (AHI), obstructive sleep apnea index (OAI), central sleep apnea index (CAI) between HSAT and PSG as well as the in-lab PM and PSG were analyzed by Bland-Altman plot; the sensitivity and specificity of PM for the diagnosis of SDB in patients with CHF were evaluated. Results: The number of patients included in the final analysis were 65 in HSAT, 63 in in-lab PM and 65 in PSG. AHI [M(Q1,Q3)] was 26.1 (10.9,40.1) events/h by HSAT, 27.9 (11.3,43.2) events/h by in-lab PM, both were not different from AHI 29.0 (10.2,45.0) events/h by PSG (P>0.05). The AHI, OAI and CAI assessed by HSAT correlated significantly with those by PSG (r=0.892, 0.903, 0.831, P<0.05). Bland-Altman analysis of AHI, OAI, CAI by PSG versus HSAT showed a mean difference of 3.1 events/h, 0.8 events/h, 1.2 events/h; limits of consistency were -15.2 to 21.4 events/h, -9.7 to 11.3 events/h, -10.9 to 13.2 events/h, respectively. Based on a threshold of AHI ≥5 events/h, HSAT had 94.6% sensitivity, 75% specificity, compared to PSG. For detecting Cheyne-Stokes respiration (CSR), a sensitivity of 96.4%,a specificity of 97.2% were achieved, compared to PSG. Conclusion: Type 3 PM can be used to diagnose SDB in patients with CHF.

Are there sex-related differences in therapeutic CPAP levels in adults undergoing in-laboratory titration?

The first-choice therapy for adults with moderate/severe obstructive sleep apnea is continuous positive airway pressure (CPAP). However, studies evaluating whether the therapeutic CPAP level obtained from a titration is affected by sex are surprisingly scarce. Our main objective was to verify if sex influenced the optimal CPAP measurement obtained during a titration. This cross-sectional study was conducted in adults diagnosed with moderate/severe obstructive sleep apnea [baseline apnea-hypopnea index (AHI) ≥ 15.0 events/h] who underwent auto-adjusting CPAP titration (S9 or S10 AutoSet ResMed) in a sleep laboratory setting. All participants used a nasal mask during the titration. The optimal pressure, leak, and residual AHI values were registered. Multiple linear regression was used to evaluate if clinical and polysomnographic data influenced the therapeutic CPAP level setting (95th percentile pressure). A total of 1,006 adults were enrolled: 354 women and 652 men. There were no statistically significant sex-related differences in the CPAP requirements and leak values delineated during the titration; all P-values > .005. However, the median residual AHI was significantly higher in males vs females: 2.7 events/h vs 2.2 events/h (P = .008). Body mass index (β: 0.292, P < .001), baseline AHI (β: 0.167, P < .001), and age (β: 0.065, P = .035) were independent predictors of the therapeutic CPAP level settings. Sex does not significantly influence the therapeutic CPAP settings. However, age, BMI, and baseline AHI emerge as independent predictors of the 95thpercentile CPAP requirement during an auto-adjusting CPAP titration. Duarte RLM, Magalhães-da-Silveira FJ, Gozal D. Are there sex-related differences in therapeutic CPAP levels in adults undergoing in-laboratory titration? JClin Sleep Med. 2021;17(9):1815-1820.

Improving the Diagnostic Ability of the Sleep Apnea Screening System Based on Oximetry by Using Physical Activity Data

Polysomnography (PSG) is the gold standard for diagnosing sleep apnea (SA), but it is costly and time-consuming. An oximeter as alternative diagnostic tool is small in size and user friendly. However, using an oximeter alone can easily underestimate SA severity. Therefore, we aimed to develop a SA screening system to solve the problem of underestimating SA severity. We developed a wireless oximetry system to record peripheral oxygen saturation, heart rate, and physical activity. Physical activity was used to remove artifacts and derive total sleep time. After artifact removal, an algorithm estimated the apnea–hypopnea index (AHI). 56 participants with different severities of SA underwent overnight inspection using home-based PSG and wireless oximetry for more than 6 h. In the four SA severity groups, the overall accuracy, sensitivity, and specificity of AHI estimated by our system (AHIEsti) were 81.25%, 69.64%, and 92.86%, respectively. AHIEsti was positively related to AHI of PSG. Receiver operating characteristic curves were plotted based on the threshold of AHI of PSG being greater than 5, 10, 15, and 30, and the corresponding areas under the curve were 0.910, 0.800, 0.794, and 0.970. This SA screening system can determine whether a patient has SA but cannot evaluate SA severity precisely. Our system is a potential screening tool for SA, supporting PSG at a lower cost.

Perception of sleep duration in adult patients with suspected obstructive sleep apnea.

Discrepancies between subjective and objective measures of total sleep time (TST) are frequent among insomnia patients, but this issue remains scarcely investigated in obstructive sleep apnea (OSA). We aimed to evaluate if sleep perception is affected by the severity of OSA. We performed a 3-month cross-sectional study of Brazilian adults undergoing overnight polysomnography (PSG). TST was objectively assessed from PSG and by a self-reported questionnaire (subjective measurement). Sleep perception index (SPI) was defined by the ratio of subjective and objective values. Diagnosis of OSA was based on an apnea/hypopnea index (AHI) ≥ 5.0/h, being its severity classified according to AHI thresholds: 5.0-14.9/h (mild OSA), 15.0-29.9/h (moderate OSA), and ≥ 30.0/h (severe OSA). Overall, 727 patients were included (58.0% males). A significant difference was found in SPI between non-OSA and OSA groups (p = 0.014). Mean SPI values significantly decreased as the OSA severity increased: without OSA (100.1 ± 40.9%), mild OSA (95.1 ± 24.6%), moderate OSA (93.5 ± 25.2%), and severe OSA (90.6 ± 28.2%), p = 0.036. Using logistic regression, increasing SPI was associated with a reduction in the likelihood of presenting any OSA (p = 0.018), moderate/severe OSA (p = 0.019), and severe OSA (p = 0.028). However, insomnia was not considered as an independent variable for the presence of any OSA, moderate/severe OSA, and severe OSA (all p-values > 0.05). In a clinical referral cohort, SPI significantly decreases with increasing OSA severity, but is not modified by the presence of insomnia symptoms.

A novel deep feature transfer-based OSA detection method using sleep sound signals

Objective: Polysomnography is typically used to evaluate the severity of obstructive sleep apnea (OSA) but the inconvenience of application and high cost considerably affect the diagnostics. In this study, sleep sound signals are used to detect OSA in patients. Approach: A deep feature transfer-based OSA detection approach is proposed. First, a deep convolutional neural network is trained on large-scale labeled audio data sets to distinguish respiration sounds from environmental noise. Second, the trained model is transferred to recognize respiration sounds in sleep sound signals. Third, the deep features of the detected respiration sounds are used to train a logistic regression classifier to identify OSA patients from potential patients. Polysomnography-based diagnosis is used as a reference. Main results: A self-collected data set of 132 potential OSA patients is applied in OSA detection experiments. The OSA detection performances are tested on four models for different apnea–hypopnea index thresholds and sexes resulting in accuracies of 80.17%, 80.21%, 81.63% and 77.22%. The corresponding areas under the receiver operating characteristic curves are 0.82, 0.80, 0.81 and 0.79. In addition, the proposed method presented a significant performance improvement compared with the state-of-the-art methods. Significance: Big data, deep learning and transfer learning can be successfully applied to improve diagnostic accuracy in OSA detection. The performance of the proposed approach is superior to that of traditional audio analysis technology. The proposed method significantly reduces difficulties in OSA detection and diagnosis, such that potential OSA patients can perform initial inspections by themselves at home.