Search Results (121)

Background/Objectives: Obstructive sleep apnea (OSA) is a prevalent chronic disorder whose first-line treatment, continuous positive airway pressure (CPAP), is effective only if the patient maintains sufficient adherence. Early predictions of the risk of low adherence would make it possible to personalize follow-up and optimize healthcare resources. The aim of this study was to develop and evaluate a machine-learning-based clinical decision support system to predict CPAP adherence after one year of treatment. Methods: A cohort of 200 patients with OSA from the Sleep-Disordered Breathing Unit of Hospital Álvaro Cunqueiro in Vigo was used. The cohort was split into a training set (n = 160) and an independent test set (n = 40). Two scenarios were defined: Scenario A, with pre-treatment variables, and Scenario B, which also includes early adherence metrics. In each scenario, variables were selected through recursive feature elimination. The selected variables were apnea-hypopnea index (AHI), 3% oxygen desaturation index (ODI3%), chronic obstructive pulmonary disease and neck circumference in Scenario A, and first-month adherence, ODI3% and AHI in Scenario B. Once these subsets were defined, several classifiers were analyzed. Results: Random Forest was the model selected in both scenarios. On the test set, Scenario A reached an area under the curve (AUC) of 0.71 (sensitivity 0.83; specificity 0.45) and Scenario B an AUC of 0.91 (sensitivity 0.90; specificity 0.82). Conclusions: Early prediction of CPAP adherence using machine learning is feasible; incorporating real first-month use markedly improves discriminative ability. The system was integrated into a web prototype as a proof of concept. Given the modest sample size and the absence of external validation, these results should be interpreted as preliminary, corresponding to an exploratory, feasibility study. For future implementation, an extensive clinical validation process will be required, along with the expansion of the database, which will likely contribute to improving the system’s robustness and generalization capacity. Full article

Obstructive sleep apnea (OSA) is associated with increased cardiovascular, respiratory, and all-cause mortality, yet the long-term survival impact of continuous positive airway pressure (CPAP) remains contested, and treatment is usually analysed as a binary exposure rather than by the quality of the response achieved. In a single-centre cohort of 4368 adults referred for polysomnography and followed for 8–20 years (prespecified subgroup with apnea–hypopnea index [AHI] ≥ 15, n = 2304), we applied cause-specific Cox and Fine–Gray competing-risks models, together with machine-learning classifiers, to characterise all-cause, cardiovascular, and pulmonary mortality. CPAP was associated with reduced all-cause (hazard ratio [HR] 0.75), cardiovascular (HR 0.75), and pulmonary mortality (HR 0.54), with the benefit confined to severe OSA (HR 0.66) and absent in moderate disease (HR 0.95). Good responders showed a significant 34% reduction in all-cause mortality (HR 0.66), whereas poor responders showed no significant reduction. Nocturnal desaturation (time with oxygen saturation < 90%) was the dominant independent predictor of pulmonary mortality, and baseline features predicted response category only moderately (macro-averaged AUC 0.76). Long-term CPAP confers a severity- and response-dependent survival benefit; nocturnal hypoxaemia is a key, modifiable driver of respiratory death. Full article

Background/Objectives: Obstructivesleep apnea (OSA) is one of the most widespread forms of sleep disease, affecting over 936 million adults globally. The health consequences of obstructive sleep apnea (OSA) are well documented; however, it remains largely underdiagnosed because the current gold-standard diagnostic method, polysomnography (PSG), is often costly, time-consuming, and unavailable in many healthcare settings. To address these challenges, this study presents a novel explainable deep learning (DL) framework for automated multi-class OSA severity classification using single-lead electrocardiogram (ECG) signals. Methods: The proposed framework integrates a hybrid CNN–BiLSTM architecture with explainable artificial intelligence (XAI) techniques to generate clinically meaningful predictions and explanations across four OSA severity classes: Normal, Mild, Moderate, and Severe. The framework was evaluated using the publicly available PhysioNet Apnea-ECG dataset (70 recordings) together with an institutional ECG dataset (150 recordings), resulting in a combined cohort of 220 recordings. Results: The proposed framework achieved an overall classification accuracy of 94.7%, with sensitivity and specificity values of 92.3% and 96.1%, respectively. Furthermore, the proposed model consistently outperformed conventional machine learning algorithms, including Support Vector Machine (SVM), Random Forest, and XGBoost, by 5.5%, 4.2%, and 2.9%, respectively. To enhance transparency and clinical trust, SHAP (SHapley Additive exPlanations) was employed to identify the most influential physiological predictors driving model decisions. Heart rate variability features, particularly RMSSD and pNN50, emerged as the strongest indicators of OSA severity. Moreover, computational efficiency analysis revealed that the model required only 0.23 s to process a 60 s ECG epoch on a standard computing platform, supporting its suitability for real-time deployment. Conclusions: The findings demonstrate that explainable deep learning applied to ECG signals can provide accurate, interpretable, and computationally efficient assessment of OSA severity. The proposed framework may support OSA screening, clinical triage, and early intervention, particularly in resource-constrained healthcare environments. Full article

Background: Hypoglossal nerve stimulation (HNS) is an established surgical therapy for adults with moderate-to-severe obstructive sleep apnea (OSA) who are intolerant to positive airway pressure. Although aggregate response rates of ~70–80% have been reported, substantial variability across clinical settings remains poorly understood. Prior meta-analyses have largely emphasized pooled continuous outcomes, limiting interpretation of responder-based endpoints and drivers of between-study heterogeneity. Methods: A PRISMA-compliant systematic review and meta-analysis was performed. MEDLINE, Embase, and Cochrane CENTRAL were searched from inception through 31 December 2025. Eligible studies enrolled adults with OSA treated with implantable HNS, reported Sher-defined response (≥50% AHI reduction and residual AHI < 20 events/hour), and/or continuous outcomes, and included ≥20 patients. Random-effects models (REML) were applied. Heterogeneity was quantified using I² and τ², with prediction intervals. Meta-regression assessed baseline AHI, BMI, and follow-up duration. Subgroup analyses examined device laterality, stimulation modality, sleep assessment method, and follow-up. Results: Thirty-eight studies (39 cohorts; n = 3220) were included. The pooled Sher response rate was 74.0% (95% CI 67.6–79.5%). Heterogeneity was substantial. HNS significantly improved all continuous outcomes (AHI −23.3 events/hour; ESS −4.5 points; ODI −14.5 events/hour). Comparative analyses favored HNS over surgical comparators, inactive stimulation, and delayed treatment. Revision and explantation rates were 5% and 4%, respectively. Meta-regression showed no significant effects of baseline AHI, BMI, or follow-up, explaining negligible variance. Subgroups suggested numerically higher response with breathing-synchronized stimulation, but heterogeneity remained high. Conclusions: HNS achieves Sher response in approximately three-quarters of appropriately selected CPAP-intolerant OSA patients, with durable clinical benefits and a favorable safety profile. Persistent unexplained heterogeneity highlights limitations of conventional predictors and underscores the need for more granular response determinants. Full article

Background/Objectives: Obstructive sleep apnea (OSA) is associated with an increased risk of motor vehicle accidents, traditionally attributed to excessive daytime sleepiness (EDS). However, subjective sleepiness may be underreported and does not consistently reflect functional impairment. We aimed to examine the association between objectively measured drowsiness during simulated driving and traffic accidents in OSA, and to compare the predictive roles of objective drowsiness, subjective sleepiness, and OSA severity. Methods: Fifty-one male drivers underwent overnight polysomnography followed by a 50-min driving simulation. OSA severity was categorized as moderate-to-severe (AHI ≥ 15 events/h) or no/mild (AHI < 15 events/h). A frontal camera captured facial expressions. Drowsiness was quantified using eye-closure-based metrics: PERCLOS, the ratio of frames with closed eyes to total observable frames, and CLOSDUR, representing eye-closure duration. Drowsiness was defined as PERCLOS ≥ 0.3 or CLOSDUR ≥ 2 s. Drowsiness-related traffic accidents were recorded. Multivariable logistic regression models were adjusted for age, body mass index, Epworth Sleepiness Scale (ESS), and total sleep time. Results: Drowsiness-related traffic accidents occurred more frequently in participants with moderate-to-severe OSA than in those with no or mild OSA (72.2% vs. 40.0%, p = 0.030). Drowsiness duration, but not ESS, was positively associated with the number of traffic accidents (r = 0.46, p = 0.001). While continuous AHI was not associated with accidents, moderate-to-severe OSA was independently associated with higher accident risk (adjusted OR 6.45, 95% CI 1.43–29.13; p = 0.015). Conclusions: Driving-related accident risk in OSA was associated with objectively measured drowsiness and moderate-to-severe disease, whereas subjective sleepiness assessed by the ESS showed no significant association. These findings suggest that functional impairment, rather than self-reported symptoms, may be more relevant for identifying high-risk drivers. Full article

Background/Objectives: Obstructive sleep apnea (OSA) and related sleep disorders affect a substantial proportion of hospitalized patients, with an estimated 48% pooled prevalence of undiagnosed OSA in cardiac inpatients and up to 80% of moderate-to-severe community OSA cases carrying no formal diagnosis at the time of hospital admission. In parallel, frailty—a state of heightened physiological vulnerability arising from cumulative multi-system biological decline—is present in 40–80% of inpatients and shares deep, bidirectional neurobiological pathways with sleep-disordered breathing through circadian dysregulation, intermittent hypoxia, hypothalamic–pituitary–adrenal axis activation, and chronic low-grade inflammation. Despite this convergence, no prior study has integrated validated, administratively computable frailty phenotyping with a machine learning framework specifically designed to predict inpatient sleep disorder diagnosis—and OSA in particular—at the point of hospital admission. The present study addresses this gap by developing an admission-time, explainable machine learning framework for the prediction of inpatient sleep disorder diagnoses (ICD-10 G47.x, encompassing OSA G47.3, insomnia G47.0, hypersomnia, and circadian rhythm disorders) and of insomnia specifically (ICD-10 G47.00). Methods: We developed and evaluated a suite of five binary classification models—XGBoost, Random Forest, LightGBM, CatBoost, and Decision Tree—using 9682 balanced hospitalization episodes from the MIMIC-IV (version 2.2) database. The predictor set comprised 23 admission-time structured features across three domains: (i) frailty and comorbidity burden, including the Hospital Frailty Risk Score (HFRS) derived from ICD-10 codes, the Elixhauser comorbidity index, prior admission history, and six binary disease flags (obesity, hypertension, type 2 diabetes, heart failure, COPD, and depression/anxiety); (ii) physiological and laboratory biomarkers from the first 24 h of care, including minimum SpO₂, heart rate variability, hemoglobin, creatinine, albumin, and arterial blood gas parameters; and (iii) sociodemographic and administrative variables encompassing age, sex, ethnicity, insurance type, and admission acuity. Model performance was assessed through five-fold stratified cross-validation and bootstrap confidence intervals (n = 1000 iterations), with predictor importance quantified using SHapley Additive exPlanations (SHAP). Results: XGBoost achieved the strongest aggregate performance across all evaluation metrics, attaining an area under the receiver operating characteristic curve (AUC) of 0.871 (95% CI: 0.856–0.887), accuracy of 79.6%, F1-score of 0.820, and sensitivity of 94.9%, correctly identifying 903 of 952 true positive cases in the held-out test set; all gradient boosting frameworks substantially outperformed the Decision Tree baseline (AUC 0.836). SHAP analysis identified the HFRS and Elixhauser index as the two dominant predictors, followed by depression/anxiety, obesity, hypertension, and minimum SpO₂—a hierarchy that recapitulates the canonical clinical phenotype of obstructive sleep apnea in frail inpatients rather than that of primary insomnia, indicating that the model is preferentially capturing the OSA–frailty axis within the broader G47.x outcome. The predicted probability outputs were well-calibrated across all risk deciles. Conclusions: Frailty-derived features, in combination with admission-time clinical and physiological data, can predict inpatient sleep disorder diagnoses—predominantly OSA—with high sensitivity and well-calibrated risk estimates. The deployable, interpretable nature of the XGBoost model makes it directly suitable for integration into clinical decision support systems, offering a screening tool that requires no dedicated instrumentation beyond routine admission data. By flagging high-risk patients at the moment of admission, the framework provides a concrete mechanism for accelerating referral for definitive diagnostic confirmation (overnight oximetry, polysomnography) and earlier initiation of CPAP and related therapies, with direct implications for reducing the persistent diagnostic gap, perioperative risk, and preventable adverse outcomes in frail hospitalized populations. Full article

Obstructive sleep apnea (OSA) is increasingly recognized not merely as a localized anatomical airway disorder, but as a complex systemic condition. While the apnea–hypopnea index (AHI) remains the traditional standard for diagnosis, it possesses inherent limitations in adequately predicting downstream adverse outcomes, necessitating the adoption of novel, comprehensive markers for refined risk stratification. This narrative review summarizes recent evidence on circulating biomarkers and device-based metrics that may complement the AHI for risk stratification in OSA. Beyond traditional metrics, emerging biomarkers in obstructive sleep apnea offer a multifaceted view of the disease, utilizing inflammatory, vascular, metabolic and novel molecular indicators to enhance clinical characterization. These biomarkers reflect key pathophysiological mechanisms, including systemic inflammation, endothelial dysfunction, metabolic imbalance and altered gut microbiota. Commonly studied markers show associations with disease severity and adverse health outcomes. In addition, novel molecular markers, such as microRNAs and advances in sleep study metrics, provide further insight into disease burden, prognosis and cardiovascular comorbidities. A multidimensional framework integrating accessible laboratory markers with device-based metrics may improve identification of patients with OSA who are at highest risk for cardiovascular and metabolic complications. Full article

Background: Ultrasonography (US) has emerged as a non-invasive method for anatomical and functional evaluation of upper airway structures in adult obstructive sleep apnea (OSA). However, its role in severity stratification, dynamic assessment, elastographic characterization, and therapeutic monitoring remain to be investigated. Background/Objectives: The goal herein is thus to systematically review and synthesize available evidence on US assessment in adults with OSA, including structural parameters, dynamic measurements, correlation with the apnea–hypopnea index (AHI), integration with artificial intelligence, and evaluation of myofunctional therapy outcomes. Methods: A PRISMA-compliant systematic review of 19 studies (2007–2025) was conducted, evaluating US in adult patients with polysomnography-diagnosed OSA. Observational, pilot, case–control, and exploratory studies were included. Risk of bias was assessed using the National Institutes of Health Quality Assessment Tool for observational studies. Due to methodological heterogeneity, a structured qualitative meta-analytic synthesis was performed. Results: The tongue base was the most frequently studied structure. Increased tongue thickness, area, and stiffness were consistently associated with higher AHI. Elastography revealed increased intrinsic rigidity in patients with OSA. Dynamic US correlated with drug-induced sleep endoscopy findings and hyoid displacement. Machine learning integration improved severity prediction. A single study evaluated anatomical changes following myofunctional therapy, representing a nascent research area. US may become a complementary, non-invasive tool for anatomical and functional assessment of upper airway structures in adult OSA. Conclusions: Further standardization of acquisition protocols and well-designed longitudinal studies are needed to clarify the clinical role of US in phenotyping and therapeutic monitoring. Full article

Background/Objectives: This study aimed to investigate the impact of allergic diseases (AD) or obstructive sleep apnea (OSA) risk, as a host factor, on the development of severe Mycoplasma pneumoniae Pneumonia (SMPP) in children by analyzing the clinical data of pediatric patients with Mycoplasma pneumoniae Pneumonia (MPP). Methods: This retrospective study enrolled children hospitalized with Mycoplasma pneumoniae pneumonia (MPP) at Shanghai Ninth People’s Hospital from November 2024 to November 2025. Patients were classified into severe (SMPP) and mild (MMPP) groups. Demographic, clinical, laboratory, and questionnaire data were collected and compared between groups. Univariate and multivariate logistic regression analyses were performed to identify independent predictors of SMPP and construct a nomogram. The model was validated for discrimination, calibration, and clinical utility using ROC curves, calibration plots, and decision curve analysis, with internal validation by bootstrap resampling. Results: Among the 150 enrolled children with MPP, 35 (23.3%) were classified as severe (SMPP) and 115 (76.7%) as mild (MMPP). Patients with SMPP exhibited significantly higher frequencies of allergic diseases, prolonged fever and steroid use, elevated inflammatory markers (CRP, LDH, D-dimer, ferritin, ALT), and higher PSQ and RQLQ scores (all p < 0.05). Disease severity was positively correlated with these clinical, laboratory, and questionnaire-based parameters. Multivariate logistic regression identified allergic diseases, PSQ score, LDH, and ferritin as independent predictors of SMPP. A nomogram incorporating these four factors demonstrated good predictive performance, with an internally validated C-index of 0.827, satisfactory calibration (Hosmer–Lemeshow p = 0.116), and clinical utility within a 0–25% threshold probability range on decision curve analysis. Conclusions: Children with MPP and comorbid AD or OSA risk are more likely to develop SMPP. Among children aged 6–12 years, RQLQ score is positively correlated with the severity of MPP. AD, PSQ score, LDH, and ferritin are independent risk factors for SMPP. Clinicians should be alert to the development of SMPP when children with MPP present with a history of AD, PSQ score >3.5, LDH >327.50 U/L, or ferritin >120.05 ng/mL. The visual nomogram model constructed by combining these risk factors demonstrates improved predictive performance for SMPP, with high predictive efficacy and accuracy. It has great clinical value and can be used for individualized risk assessment and early intervention. However, our proposed nomogram requires external validation prior to broader implementation. Full article

Background: Obstructive sleep apnoea (OSA) is a highly prevalent yet underdiagnosed disorder in patients with cardiovascular disease. Growing evidence suggests a pathophysiological link between OSA and coronary artery disease (CAD); however, the relationship between OSA severity and anatomical complexity of coronary lesions remains incompletely understood. Aim: The aim of this study is to assess the prevalence of OSA in patients undergoing coronary angiography and to evaluate the association between sleep-disordered breathing parameters and the severity of CAD expressed by the SYNTAX score. Methods: This prospective study enrolled 103 consecutive patients referred for invasive coronary angiography. All participants underwent overnight type III cardiorespiratory polygraphy. OSA severity was classified according to the Apnea–Hypopnea Index (AHI). The anatomical complexity of CAD was assessed using the SYNTAX score. Linear regression analyses were performed to determine associations between polysomnographic parameters and SYNTAX score. Results: Significant CAD was diagnosed in 74.8% of patients. OSA was highly prevalent, with severe OSA observed in 36.4% of patients with significant CAD compared to 3.8% in those without significant stenoses (p = 0.003). Patients with significant CAD had higher AHI (18.8 vs. 13.5 events/h; p = 0.003), higher oxygen desaturation index (ODI) (19.3 vs. 12.9 events/h; p = 0.003), and greater mean oxygen desaturation (4.1% vs. 3.8%; p = 0.008). In multivariable regression analysis, AHI (B = 0.329; 95% CI [0.083, 0.576]; p = 0.009) and nicotinism (B = 8.693; 95% CI [2.573, 14.814]; p = 0.006) independently predicted higher SYNTAX scores. Interestingly, each 1% increase in snoring percentage was associated with a 0.203-point reduction in SYNTAX score (95% CI [−0.339, −0.068]; p = 0.004). Conclusions: OSA is highly prevalent in patients undergoing coronary angiography and is independently associated with greater anatomical complexity of CAD. Sleep-disordered breathing, particularly AHI and nocturnal hypoxemia, may represent important non-traditional risk markers of advanced coronary atherosclerosis. Systematic screening for OSA should be considered in patients with suspected or confirmed CAD. Full article

Background: The clinical value of lateral cephalograms for obstructive sleep apnea (OSA) risk assessment remains controversial, largely because previous case–control studies often lacked objective exclusion of OSA in control subjects and insufficiently controlled for confounding. This age-matched case–control study evaluated whether craniofacial characteristics differ between individuals with and without OSA and whether these craniofacial measurements independently predict OSA-related outcomes after adjustment for relevant confounders. Methods: A total of 54 adults were included (27 with OSA and 27 without OSA). OSA was defined by poly(somno)graphy (apnea–hypopnea index [AHI] ≥ 5). Control subjects were prospectively recruited, and OSA was excluded through polygraphy (AHI < 5). Lateral cephalograms were used to assess six PAS levels (P1–P6), 16 hyoid- and soft palate-related parameters, and sagittal/vertical skeletal characteristics. Potential confounders were controlled for by adjustment for BMI and craniofacial skeletal pattern. The PAS measurements were defined as the primary endpoint; soft palate and hyoid-related variables were considered secondary exploratory endpoints. Statistical analyses included independent samples t-tests, multiple linear regression models, and sensitivity analyses adjusted for sex. Results: Craniofacial skeletal characteristics did not differ between groups. PAS dimensions showed no significant intergroup differences and were not independently associated with AHI after adjustment, whereas BMI consistently emerged as the strongest predictor. Uvula length and thickness were significantly greater in the OSA group; however, neither parameter independently predicted AHI in regression models. In contrast, subjects with OSA exhibited a significantly more inferior/anterior hyoid position across multiple models. In the primary regression models, several hyoid-related variables were associated with AHI. However, these associations were attenuated in additional sensitivity analyses after adjustment for sex and were no longer consistently statistically significant. Sex was a relevant covariate in several models. Conclusions: Static PAS measurements derived from lateral cephalograms provide no clinically meaningful information for OSA screening or risk stratification. Although several hyoid-related variables were associated with AHI in primary models, these associations were attenuated after adjustment for sex and should therefore be interpreted as exploratory. When lateral cephalograms are already clinically indicated, hyoid position may provide complementary anatomical information, but its independent predictive value remains uncertain. Full article

Background: Obstructive sleep apnea (OSA) is a complex and diverse disorder affecting almost one billion individuals worldwide. Severity of untreated OSA, measured by the apnea–hypopnea index (AHI), is noted to be associated with an increased all-cause and cardiovascular mortality. Although widely used, AHI insufficiently captures disease variability as there is a poor correlation of symptoms with the AHI. There lies individual susceptibility to the effects of OSA and that parameter alone poorly predicts cardiovascular outcomes without considering intermittent hypoxia and the hemodynamic effects of OSA. Recognition of clinical, polysomnographic, and neurophysiological phenotypes offers an opportunity to refine diagnosis, prognosis, and management strategies. Methods: We conducted a narrative synthesis of the literature involving 70 articles, focusing on quantitative and qualitative (Q2) clinical traits, polysomnographic parameters, and mechanistic insights that enable subclassification of OSA beyond AHI. Evidence from large cohorts, animal models, and pathophysiological studies were reviewed. Results: Phenotyping based on a Q2 analysis of polysomnographic respiratory event predominance, event duration, positional and REM dependence, hypoxic burden, and arousal characteristics reveals significant heterogeneity in risk profiles and therapeutic response. Apnea-predominant OSA correlates with a higher oxygen desaturation index and Epworth sleepiness scale. Hypopnea-predominant OSA correlates with a cardiometabolic disease burden and may show a more favorable response to surgical therapies. The duration of respiratory events is related to cardiovascular risk, and REM-predominant OSA independently predicts hypertension and adverse cardiovascular outcomes. Supine-predominant OSA demonstrates treatment responsiveness to auto-positive airway pressure and positional therapy. Respiratory effort–related arousals (RERAs), RERA-predominant OSA and the broader respiratory disturbance index (RDI) provide neurophysiological insight often missed by AHI-based classifications. Hypoxic burden, rather than AHI, emerged as a superior predictor of cardiovascular events and mortality. Finally, arousal frequency and periodic limb movements independently predict cardiovascular morbidity. Conclusions: Employing Q2-based phenotyping that incorporates clinical, polysomnographic, and neurophysiological markers improves risk stratification, prognosis, and individualized management of OSA. Future investigations should prioritize integrating phenotypic subclassification into diagnostic criteria and treatment planning to advance precision medicine in sleep apnea care. Full article

Background/Objectives: Obstructive sleep apnea (OSA) syndrome is characterized by recurrent upper airway obstruction during sleep and is closely associated with systemic inflammation and cardiometabolic risk. α-Klotho and fibroblast growth factor-23 (FGF-23) are emerging biomarkers with potential roles in vascular homeostasis, inflammation, and metabolic regulation. However, their relevance in OSA remains insufficiently elucidated. The aim of this study was to evaluate serum α-Klotho and FGF-23 levels in patients with OSA and to investigate their associations with disease severity. This represents a novel approach that may provide new insights into the pathophysiological mechanisms linking OSA with cardiometabolic risk. Methods: A total of 133 participants were included in this study and categorized into three groups according to apnea–hypopnea index: 1—simple snoring (n = 44); 2—non-severe OSA (n = 44); and 3—severe OSA (n = 45). Comparisons between two groups were performed using Student’s t-test for normally distributed variables. Comparisons among three or more groups were conducted using one-way ANOVA and the Kruskal–Wallis test. ANCOVA was applied to compare α-Klotho and FGF-23 levels between groups after adjustment for age, BMI, diabetes, hypertension, asthma, COPD, and thyroid disease. The predictive performance of α-Klotho and FGF-23 for severe obstructive sleep apnea was evaluated using ROC curve analysis. Results: Serum α-Klotho levels decreased significantly with increasing OSA severity (p = 0.001). Serum FGF-23 levels increased significantly across AHI groups (p = 0.001). After adjustment for age, BMI, diabetes, hypertension, asthma, thyroid disease, COPD and vitamin D levels, α-Klotho levels were lower in the severe and non-severe OSA group (p = 0.001, both) compared to the simple snoring group, whereas FGF-23 levels were higher in the severe and non-severe OSA group (p = 0.001; both) compared to the simple snoring group. In predicting the risk of severe OSA compared with non-severe OSA, an α-Klotho cut-off value of 280.3 yielded a sensitivity of 84.44% and specificity of 75%, whereas an FGF-23 cut-off value of 75.5 yielded a sensitivity of 62.2% and specificity of 72.7%. Conclusions: Serum α-Klotho levels significantly decrease while FGF-23 levels increase in correlation with OSA severity. α-Klotho exhibited superior predictive performance over FGF-23 in identifying severe OSA, suggesting its potential as a more sensitive biomarker for systemic involvement. These results indicate that the α-Klotho/FGF-23 axis is independently associated with OSA and may play a pivotal role in the pathophysiological mechanisms linking intermittent hypoxia to increased cardiometabolic risk. Full article

Obstructive sleep apnea (OSA), defined by repetitive airway obstruction during sleep, is significantly underdiagnosed, mainly due to the resource-intensive and time-consuming nature of sleep assessment technologies. Machine learning analysis of the tracheal breathing sounds (TBS) whilst awake offers an alternative approach for OSA quick screening. This study aimed to address the challenge of wakefulness OSA detection using TBS recorded with an inexpensive microphone in a noisy environment. Data of 247 individuals with various degrees of OSA severity were analyzed. Recorded data were segmented into inspiration and expiration phases, followed by acoustic features extraction, feature reduction, and classification. A two-level ensemble architecture was implemented. Nine sub-classifiers were stratified by anthropometric profiles. Each sub-classifier was constructed as an ensemble of bagged decision trees, with a final prediction via probability-based voting. The proposed algorithm achieved an accuracy of 77.1%, sensitivity of 84.3%, and specificity of 59.9%. Although these results have lower performance than those obtained previously using a high-quality microphone in a quiet room, they demonstrate that acoustic OSA detection whilst awake remains feasible, even in very noisy environments. Nevertheless, microphone quality emerged as a key determinant of classification performance. Full article

Background: Obstructive sleep apnea (OSA) is associated with slower response speed, yet conventional severity classification based on the apnea–hypopnea index (AHI) shows limited ability to predict cognitive outcomes. The AHI aggregates distinct pathophysiological processes, including intermittent hypoxemia and sleep fragmentation. Within emerging precision sleep medicine frameworks, disentangling these mechanisms is critical for improved phenotyping and personalized risk assessment. This study aimed to replicate prior findings using a Go/No-Go Continuous Visual Attention Test (CVAT) and to identify the most informative polysomnographic predictor of attentional performance in OSA. Methods: In this cross-sectional study, participants underwent full-night type I polysomnography and the CVAT. After exclusions, 84 patients with OSA and 22 polysomnographically normal controls were analyzed. The sample sizes for mean differences and correlational analyses were adequate. Attentional performance was indexed by standardized reaction time (RT), referenced to a normative database (n = 1244). Within the OSA group, linear regression with backward elimination evaluated hypoxemia and sleep fragmentation metrics. Results: Patients with OSA demonstrated significantly slower RTs than controls (p = 0.005). Within OSA, the AHI was not associated with attentional performance (p = 0.398). In the final regression model, sleep stage shifts—reflecting sleep–wake instability—emerged as the sole independent predictor of attentional slowing (β = 0.27, p = 0.013), whereas all hypoxemia indices were excluded. Conclusions: Sleep stage instability represents a cognitive vulnerability marker in OSA, independent of respiratory events. Integrating fragmentation metrics into precision sleep medicine models may enhance individualized phenotyping, identify patients at higher neurocognitive risk, and inform targeted interventions focused on stabilizing sleep architecture rather than relying solely on the AHI. Full article