GRU-powered sleep stage classification with permutation-based EEG channel selection.

We present a new approach to classifying the sleep stage that incorporates a computationally inexpensive method based on permutations for channel selection and takes advantage of deep learning power, specifically the gated recurrent unit (GRU) model, along with other deep learning methods. By systematically permuting the electroencephalographic (EEG) channels, different combinations of EEG channels are evaluated to identify the most informative subset for the classification of the 5-class sleep stage. For analysis, we used an EEG dataset that was collected at the International Institute for Integrative Sleep Medicine (WPI-IIIS) at the University of Tsukuba in Japan. The results of these explorations provide many new insights such as the (1) drastic decrease in performance when channels are fewer than 3, (2) 3-random channels selected by permutation provide the same or better prediction than the 3 channels recommended by the American Academy of Sleep Medicine (AASM), (3) N1 class suffers the most in prediction accuracy as the channels drop from 128 to 3 random or 3 AASM, and (4) no single channel provides acceptable levels of accuracy in the prediction of 5 classes. The results obtained show the GRU's ability to retain essential temporal information from EEG data, which allows capturing the underlying patterns associated with each sleep stage effectively. Using permutation-based channel selection, we enhance or at least maintain as high model efficiency as when using high-density EEG, incorporating only the most informative EEG channels.

Low arousal threshold is associated with altered functional connectivity of the ascending reticular activating system in patients with obstructive sleep apnea.

A low arousal threshold (LAT) is a pathophysiological trait of obstructive sleep apnea (OSA) that may be associated with brainstem ascending reticular activating system-cortical functional connectivity changes. We evaluated resting-state connectivity between the brainstem nuclei and 105 cortical/subcortical regions in OSA patients with or without a LAT and healthy controls. Twenty-five patients with moderate to severe OSA with an apnea-hypopnea index between 20 and 40/hr (15 with and 10 without a LAT) and 15 age- and sex-matched controls were evaluated. Participants underwent functional magnetic resonance imaging after overnight polysomnography. Three brainstem nuclei-the locus coeruleus (LC), laterodorsal tegmental nucleus (LDTg), and ventral tegmental area (VTA)-associated with OSA in our previous study were used as seeds. Functional connectivity values of the two brainstem nuclei (LC and LDTg) significantly differed among the three groups. The connectivity of the LC with the precuneus was stronger in OSA patients than in controls regardless of the concomitant LAT. The connectivity between the LDTg and the posterior cingulate cortex was also stronger in OSA patients regardless of the LAT. Moreover, OSA patients without a LAT showed stronger LDTg-posterior cingulate cortex connectivity than those with a LAT (post hoc p = 0.013), and this connectivity strength was negatively correlated with the minimum oxygen saturation in OSA patients (r = - 0.463, p = 0.023). The LAT in OSA patients was associated with altered LDTg-posterior cingulate cortex connectivity. This result may suggested that cholinergic activity may play a role in the LAT in OSA patients.

Toward Interpretable Sleep Stage Classification Using Cross-Modal Transformers.

Accurate sleep stage classification is significant for sleep health assessment. In recent years, several machine-learning based sleep staging algorithms have been developed, and in particular, deep-learning based algorithms have achieved performance on par with human annotation. Despite improved performance, a limitation of most deep-learning based algorithms is their black-box behavior, which have limited their use in clinical settings. Here, we propose a cross-modal transformer, which is a transformer-based method for sleep stage classification. The proposed cross-modal transformer consists of a cross-modal transformer encoder architecture along with a multi-scale one-dimensional convolutional neural network for automatic representation learning. The performance of our method is on-par with the state-of-the-art methods and eliminates the black-box behavior of deep-learning models by utilizing the interpretability aspect of the attention modules. Furthermore, our method provides considerable reductions in the number of parameters and training time compared to the state-of-the-art methods. Our code is available at https://github.com/Jathurshan0330/Cross-Modal-Transformer. A demo of our work can be found at https://bit.ly/Cross_modal_transformer_demo.

The Complex Relationship between Neuromodulators, Circadian Rhythms, and Insomnia in Patients with Obstructive Sleep Apnea.

Obstructive sleep apnea (OSA) has been linked to disruptions in circadian rhythm and neurotrophin (NFT) signaling. This study explored the link between neuromodulators, chronotype, and insomnia in OSA. The participants (n = 166) underwent polysomnography (PSG) before being categorized into either the control or the OSA group. The following questionnaires were completed: Insomnia Severity Index (ISI), Epworth Sleepiness Scale, Chronotype Questionnaire (morningness-eveningness (ME), and subjective amplitude (AM). Blood samples were collected post-PSG for protein level assessment using ELISA kits for brain-derived neurotrophic factor (BDNF), proBDNF, glial-cell-line-derived neurotrophic factor, NFT3, and NFT4. Gene expression was analyzed utilizing qRT-PCR. No significant differences were found in neuromodulator levels between OSA patients and controls. The controls with insomnia exhibited elevated neuromodulator gene expression (p < 0.05). In the non-insomnia individuals, BDNF and NTF3 expression was increased in the OSA group compared to controls (p = 0.007 for both); there were no significant differences between the insomnia groups. The ISI scores positively correlated with all gene expressions in both groups, except for NTF4 in OSA (R = 0.127, p = 0.172). AM and ME were predicting factors for the ISI score and clinically significant insomnia (p < 0.05 for both groups). Compromised compensatory mechanisms in OSA may exacerbate insomnia. The correlation between chronotype and NFT expression highlights the role of circadian misalignments in sleep disruptions.

[Analysis of the difference of therapeutic effect and predictive factors between postural and non-postural OSA after modified uvulopalatopharyngoplasty].

Objective:To analyze the factors influencing the outcome of uvulopalatopharyngoplasty in positional obstructive sleep apnea（POSA） and non-positional OSA（NPOSA） patients, and to explore the differences between the two groups. Methods:The data of 101 patients with obstructive sleep apnea who received treatment from November 2020 to November 2023 were retrospectively analyzed. Among them, 45 positional patients（POSA group） and 56 non-positional patients（NPOSA group）, who underwent overnight polysomnography were included. The upper airway（UA） anatomy was evaluated by three-dimensional computer tomography（3D-CT）. All patients received revised uvulopalatopharyngoplasty with uvula preservation and were followed using polysomnography for at least three months postoperatively. Results:The overall effective rate was 55.45%. The surgical success rate in POSA undergoing UPPP was higher than NPOSA（POSA 30/45, 66.7% versus NPOSA 26/56, 46.4%, P=0.042）. The H-UPPP effect of POSA was negatively correlated with the minimum lateral airway of the Velopharyngeal airway（r=-0.505, P<0.001）, the minimum lateral airway of the glossopharyngeal airway（r=-0.474, P=0.001） and the minimum cross-sectional area（r=-0.394, P=0.007）. Logistic analysis showed that minimal lateral airway of the glossopharynxgeum（mLAT）（OR 0.873; 95%CI 0.798-0.955, P=0.003） was a significant predictor for surgical outcomes among POSA patients. In NPOSA, age（OR 0.936; 95%CI 0.879-0.998, P=0.042） was a significant predictor for surgical outcomes. Conclusion:The effect of H-UPPP was higher in POSA than in NPOSA. The width of glossopharyngeal mLAT was an important predictor of POSA efficacy. Age was a predictor of NPOSA efficacy.

The effect of central sleep apnea on sleep architecture in children with obstructive sleep apnea.

OBJECTIVE: This study aimed to investigate how central sleep apnea (CSA) impacts sleep patterns in children with obstructive sleep apnea (OSA). METHODS: Children undergoing polysomnography (PSG) were enrolled and sorted into two groups: those with OSA alone (Group A) and those with both OSA and CSA (CAI <1 nd: children with 10 % CSA or more and less than 50 %, Group B). Statistical analysis was conducted to compare sleep structure and clinical features between Group A and Group B. RESULTS: Group B exhibited significantly higher respiratory events, apnea hypoventilation index, apnea index and oxygen desaturation index (ODI) compared to Group A (p < 0.05). Group B also showed higher total sleep time and arousal index than Group A (P < 0.05). The proportion of time spent in stage N3 was lower in Group B than in Group A (P < 0.05). Moreover, mean heart rate and minimum heart rate were higher in Group B compared to Group A (P < 0.05).Minimum oxygenation levels (including non-rapid eye movement (NREM) stages) were lowe in Group B than in Group A (P < 0.05). Additionally, the prevalence of positional obstructive sleep apnea (P-OSA) was greater in Group B than in Group A (P < 0.05). CONCLUSION: In comparison to those with OSA alone, children with OSA and concurrent CSA exhibited distinct sleep patterns, including reduced N3uration, higher arousal index, longer respiratory events, higher ODI, and lower oxygen saturation, higher heart rate.

Effects of different treatments for obstructive sleep apnea on temporomandibular joint: a randomized clinical trial.

BACKGROUND: In recent years, obstructive sleep apnea (OSA) has been increasingly recognized as a significant health concern. No previous studies assessed the effect of recommended treatment modalities of patients with OSA on the temporomandibular joint (TMJ). The current study aimed to evaluate the effect of different treatment modalities of OSA, such as continuous positive airway pressure (CPAP), mandibular advancement device (MAD), and oral myofunctional therapy (OMT) on subjective symptoms, clinical, and radiographic signs of temporomandibular disorders. PATIENTS & METHODS: This hospital-based prospective randomized controlled clinical trial study was approved by the institutional review board and formal patient consent, 39 OSA patients, ranging in age from 19 to 56 after confirmation with full night Polysomnography (PSG) with healthy TMJ confirmed clinically and radiographically with magnetic resonance imaging (MRI) were randomly allocated into three treatment groups. Group 1: 13 patients were managed with CPAP after titration, group 2: 13 patients were managed with digitally fabricated MAD, and group 3: 13 patients were managed with OMT. The following parameters were evaluated before and 3 months after the intervention. Pain using a visual analogue scale (VAS), maximum inter-incisal opening (MIO), lateral movements, and clicking sound of TMJ. MRI was done before and 3 months after the intervention. RESULTS: Out of the 83 patients enrolled, 39 patients completed the treatment. There were no statistically significant differences in lateral jaw movements or clicking, and no significant difference in MRI findings between the three studied groups before and after the intervention. The OMT group showed a statistically significant difference in pain (p = 0.001), and MIO (p = 0.043) where patients experienced mild pain and slight limitation in mouth opening after 3 months of follow-up in comparison to MAD and CPAP groups. CONCLUSION: CPAP and MAD are better for preserving the health of TMJ in the treatment of OSA patients. While OMT showed mild pain and slight limitation of MIO (that is still within the normal range of mouth opening) compared to CPAP and MAD. TRIAL REGISTRATION: The study was listed on www. CLINICALTRIALS: gov with registration number (NCT05510882) on 22/08/2022.

Smart Sleep Monitoring: Sparse Sensor-Based Spatiotemporal CNN for Sleep Posture Detection.

Sleep quality is heavily influenced by sleep posture, with research indicating that a supine posture can worsen obstructive sleep apnea (OSA) while lateral postures promote better sleep. For patients confined to beds, regular changes in posture are crucial to prevent the development of ulcers and bedsores. This study presents a novel sparse sensor-based spatiotemporal convolutional neural network (S3CNN) for detecting sleep posture. This S3CNN holistically incorporates a pair of spatial convolution neural networks to capture cardiorespiratory activity maps and a pair of temporal convolution neural networks to capture the heart rate and respiratory rate. Sleep data were collected in actual sleep conditions from 22 subjects using a sparse sensor array. The S3CNN was then trained to capture the spatial pressure distribution from the cardiorespiratory activity and temporal cardiopulmonary variability from the heart and respiratory data. Its performance was evaluated using three rounds of 10 fold cross-validation on the 8583 data samples collected from the subjects. The results yielded 91.96% recall, 92.65% precision, and 93.02% accuracy, which are comparable to the state-of-the-art methods that use significantly more sensors for marginally enhanced accuracy. Hence, the proposed S3CNN shows promise for sleep posture monitoring using sparse sensors, demonstrating potential for a more cost-effective approach.

[Idiopathic hypersomnia: a kind of sleep disorder that cannot be ignored].

Idiopathic hypersomnia(IH) is a chronic central disorders of hypersomnolence that manifests as excessive daytime sleepiness occurring despite normal or prolonged sleep time. Due to the individual heterogeneity of disease, the high overlap of clinical, poor repeatability of polysomnography monitoring results and the lack of clear disease biomarkers, clinical diagnosis and differential diagnosis are still difficult. This article summarizes the update of diagnostic criteria, clinical manifestations, diagnosis and treatment strategies of IH, in order to receive attention, increase the recognition rate of clinical diagnosis, reduce the misdiagnosis rate and missed diagnosis rate.

[The relationship between rapid eye movement electromyogram activity and sleep stability in rapid eye movement sleep behavior disorder].

Objective: To investigate the relationship between the types of electromyogram (EMG) activity and sleep stability during rapid eye movement (REM) in patients with rapid eye movement sleep behavior disorder(RBD). Methods: One hundred and three patients with RBD who met the inclusion and exclusion criteria at the Second Affiliated Hospital of Air Force Military Medical University from January 2017 to December 2019 were retrospectively analyzed. The general situation, clinical symptoms, sleep and emotion questionnaires and nocturnal PSG data were collected. According to the different proportions of tonic and phasic EMG activity, the group with a higher proportion of tonic EMG than phasic EMG was defined as the tonic dominant group, and the group with a higher proportion of phasic EMG than tonic was defined as the phasic dominant group. The sleep instability index was calculated according to the ratio of the number of transitions from sleep to wakefulness to the total sleep time of each sleep stage. Multiple linear regression was used to explore the relationship between REM EMG activity and sleep instability index. Results: A total of 35 idiopathic RBD (iRBD) patients were included, aged(54.5±18.2)years, with 17 males and 18 females. There were 27 RBD with Parkinson's disease (PD), with an average age of (59.4±7.9)years, including 17 males and 10 females. Additionally, there were 41 RBD patients with narcolepsy, aged (21.2±13.2)years, consisting of 22 males and 19 females. Both iRBD and RBD patients with PD had lower objective total sleep time, sleep latency, sleep efficiency, wake time after sleep onset and the percentage of N3 sleep compared to RBD with episodic sleep disorder (all P<0.05). N1-W index[M(Q1, Q3),10.6 (6.5, 16.9)/h vs 7.3 (4.7, 10.5)/h], N2-W index [4.0 (2.2, 5.6)/h vs 2.3 (1.5, 3.9)/h], NREM-W index [ (5.8±2.9)/h vs (4.5±3.2)/h] and REM-W index[ 3.9 (1.9, 7.3)/h vs 2.7 (1.0, 4.0)/h] in the phasic dominant group were higher than those in the tonic dominant group. After adjusting for confounding factors, the effect of phasic EMG dominant group on REM-W was higher than that in the tonic dominant group (ß=2.05, 95%CI: 0.09-3.26, P=0.012). Conclusion: In RBD patients, the phasic EMG activity has a significant impact on sleep stability, especially on REM sleep stability.

Association between obstructive sleep apnea and arrhythmia and heart rate variability among hypertensive patients.

BACKGROUND: The relationship between obstructive sleep apnea (OSA) and the occurrence of arrhythmias and heart rate variability (HRV) in hypertensive patients is not elucidated. Our study investigates the association between OSA, arrhythmias, and HRV in hypertensive patients. METHODS: We conducted a cross-sectional analysis involving hypertensive patients divided based on their apnea-hypopnea index (AHI) into two groups: the AHI ≤ 15 and the AHI > 15. All participants underwent polysomnography (PSG), 24-hour dynamic electrocardiography (DCG), cardiac Doppler ultrasound, and other relevant evaluations. RESULTS: The AHI > 15 group showed a significantly higher prevalence of frequent atrial premature beats and atrial tachycardia (P = 0.030 and P = 0.035, respectively) than the AHI ≤ 15 group. Time-domain analysis indicated that the standard deviation of normal-to-normal R-R intervals (SDNN) and the standard deviation of every 5-minute normal-to-normal R-R intervals (SDANN) were significantly higher in the AHI > 15 group (P = 0.020 and P = 0.033, respectively). Frequency domain analysis revealed that the low-frequency (LF), high-frequency (HF) components, and the LF/HF ratio were also significantly elevated in the AHI > 15 group (P < 0.001, P = 0.031, and P = 0.028, respectively). Furthermore, left atrial diameter (LAD) was significantly larger in the AHI > 15 group (P < 0.001). Both univariate and multivariable linear regression analyses confirmed a significant association between PSG-derived independent variables and the dependent HRV parameters SDNN, LF, and LF/HF ratio (F = 8.929, P < 0.001; F = 14.832, P < 0.001; F = 5.917, P = 0.016, respectively). CONCLUSIONS: Hypertensive patients with AHI > 15 are at an increased risk for atrial arrhythmias and left atrial dilation, with HRV significantly correlating with OSA severity.

Melatonin mediates intestinal barrier dysfunction and systemic inflammation in moderate-severe OSA patients.

BACKGROUND: Intestinal barrier dysfunction and systemic inflammation are common in obstructive sleep apnoea (OSA). We aimed to investigate the role of melatonin, an anti-inflammatory mediator, in mediating the relationships between OSA, intestinal barrier dysfunction and systemic inflammation. METHODS: Two hundred and thirty-five male participants who complained with sleep problems and underwent whole night polysomnography at our sleep centre between 2017 and 2018 were enrolled. Polysomnographic data, anthropometric measurements and biochemical indicators were collected. Serum melatonin, intestinal barrier function biomarker zonula occludens-1 (ZO-1) and inflammatory biomarkers C-reactive protein (CRP) with lipopolysaccharide (LPS) were detected. Spearman's correlation analysis assessed the correlations between sleep parameters, melatonin and biomarkers (ZO-1, LPS and CRP). Mediation analysis explored the effect of OSA on intestinal barrier dysfunction and systemic inflammation in moderate-severe OSA patients. RESULTS: As OSA severity increased, serum melatonin decreased, whereas ZO-1, LPS and CRP increased. Spearman's correlation analysis showed that serum melatonin was significantly negatively correlated with ZO-1 (r = -0.19, p < .05) and LPS (r = -0.20, p < .05) in the moderate-OSA group; serum melatonin was significantly negatively correlated with ZO-1 (r = -0.46, p < .01), LPS (r = -0.35, p < .01) and CPR (r = -0.30, p < .05) in the severe-OSA group. Mediation analyses showed melatonin explain 36.12% and 35.38% of the effect of apnoea-hypopnea index (AHI) on ZO-1 and LPS in moderate to severe OSA patients. CONCLUSIONS: Our study revealed that melatonin may be involved in mediating intestinal barrier dysfunction and systemic inflammation in moderate-to-severe OSA patients.

Molecular characterization of the circadian clock in patients with Parkinson's disease-CLOCK4PD Study protocol.

INTRODUCTION: Circadian rhythms (CRs) orchestrate intrinsic 24-hour oscillations which synchronize an organism's physiology and behaviour with respect to daily cycles. CR disruptions have been linked to Parkinson's Disease (PD), the second most prevalent neurodegenerative disorder globally, and are associated to several PD-symptoms such as sleep disturbances. Studying molecular changes of CR offers a potential avenue for unravelling novel insights into the PD progression, symptoms, and can be further used for optimization of treatment strategies. Yet, a comprehensive characterization of the alterations at the molecular expression level for core-clock and clock-controlled genes in PD is still missing. METHODS AND ANALYSIS: The proposed study protocol will be used to characterize expression profiles of circadian genes obtained from saliva samples in PD patients and controls. For this purpose, 20 healthy controls and 70 PD patients will be recruited. Data from clinical assessment, questionnaires, actigraphy tracking and polysomnography will be collected and clinical evaluations will be repeated as a follow-up in one-year time. We plan to carry out sub-group analyses considering several clinical factors (e.g., biological sex, treatment dosages, or fluctuation of symptoms), and to correlate reflected changes in CR of measured genes with distinct PD phenotypes (diffuse malignant and mild/motor-predominant). Additionally, using NanoStringⓇ multiplex technology on a subset of samples, we aim to further explore potential CR alterations in hundreds of genes involved in neuropathology pathways. DISCUSSION: CLOCK4PD is a mono-centric, non-interventional observational study aiming at the molecular characterization of CR alterations in PD. We further plan to determine physiological modifications in sleep and activity patterns, and clinical factors correlating with the observed CR changes. Our study may provide valuable insights into the intricate interplay between CR and PD with a potential to be used as a predictor of circadian alterations reflecting distinct disease phenotypes, symptoms, and progression outcomes.

Advances in Sleep-Disordered Breathing in Children.

Pediatric sleep-disordered breathing disorders are a group of common conditions, from habitual snoring to obstructive sleep apnea (OSA) syndrome, affecting a significant proportion of children. The present article summarizes the current knowledge on diagnosis and treatment of pediatric OSA focusing on therapeutic and surgical advancements in the field in recent years. Advancements in OSA such as biomarkers, improving continuous pressure therapy adherence, novel pharmacotherapies, and advanced surgeries are discussed.

[Novel advances in assessment and treatment of obstructive sleep apnea].

The current indicators for assessing obstructive sleep apnea are sleep apnea-hypopnea index (AHI) and the lowest nocturnal oxygen saturation, but they do not comprehensively reflect the severity of the disease. The main treatments for OSA are continuous positive airway pressure, mandibular advancement devices and surgery, which have poor compliance and limited effectiveness. Therefore, there is a need to explore novel methods to assess and treat OSA. This article systematically summarized recent advances in this field.

Effects of Positive Airway Pressure on Cardiorespiratory Fitness in Patients with Concomitant Obstructive Sleep Apnea and Cardiovascular Disease.

Background and Objectives: Obstructive sleep apnea (OSA) is common in cardiovascular disease (CVD), although positive airway pressure (PAP) treatment has not been demonstrated to improve the cardiovascular outcome. The objective of this study is to investigate the impact of adherence to PAP therapy on cardiopulmonary exercise testing (CPET) performance in patients with concomitant OSA and CVD. Materials and Methods: This preliminary study involved symptomatic OSA patients requiring PAP treatment who had CVD. All subjects underwent polysomnography, echocardiography, and CPET at baseline. After 6 to 12 months of PAP treatment, CPET performance was re-assessed. The changes in CPET parameters before and after PAP treatment were compared between patients who were adherent to PAP and patients who were not adherent to PAP. Results: A total of 16 OSA patients with an apnea-hypopnea index of 32.0 ± 23.4 were enrolled. Patients were classified into the adherent (n = 9) and non-adherent (n = 7) groups with regard to PAP adherence. After 6 to 12 months of PAP treatment, the PAP-adherent group showed a greater increase in peak VO2 than the PAP-non-adherent group, but the difference between the two groups was not significant (p = 0.581). The decrease in ventilatory equivalent for the carbon dioxide slope (VE/VCO2) was significantly greater in the PAP-adherent group compared to the PAP-non-adherent group (p = 0.030). Conclusions: Adherence to PAP therapy for OSA is associated with an improvement in the VE/VCO2 slope, as an index of the ventilatory response to exercise, in patients with CVD. Screening for sleep apnea in CVD patients may be warranted, and strategies to optimize adherence to PAP in these patients are beneficial. Further evidence is needed to elucidate whether CPET could be routinely used to monitor treatment responses of OSA to PAP therapy in patients with CVD.

Cardio-metabolic health effects of CPAP treatment for sleep apnoea during weight loss: A randomised controlled pilot trial.

BACKGROUND AND AIMS: This study assessed whether the addition of continuous positive airway pressure (CPAP) during weight loss would enhance cardiometabolic health improvements in patients with obesity and Obstructive Sleep Apnoea (OSA). METHODS AND RESULTS: Patients with overweight or obesity, pre-diabetes and moderatesevere OSA were randomised to receive CPAP therapy with a weight loss programme (CPAP+WL) or a weight loss programme alone (WL alone). PRIMARY OUTCOME: 2-hour glucose assessed by an oral glucose tolerance test. SECONDARY OUTCOMES: 24 hr blood pressure, body composition (DEXA) and fasting blood markers. 17 patients completed 3-month follow-up assessments (8 CPAP+WL and 9 WL alone). Overall, participants in both groups lost ∼12 kg which reduced polysomnography determined OSA severity by ∼45 %. In the CPAP+WL group, CPAP use (compliance 5.29 hrs/night) did not improve any outcome above WL alone. There was no improvement in 2-hour glucose in either group. However, in the pooled (n = 17) analysis there were overall improvements in most outcomes including insulin sensitivity (.000965 units, p = .008), sleep systolic BP (- 16.2 mmHg, p = .0003), sleep diastolic BP (-9.8 mmHg, p = 0.02), wake diastolic BP (- 4.3 mmHg, p = .03) and sleepiness (Epworth Sleepiness Score -3.2, p = .0003). In addition, there were reductions in glucose area under the curve (-230 units, p = .009), total (-0.86 mmol/L, p = 0.006) and LDL cholesterol (-0.58 mmol/L, p = 0.007), triglycerides (-0.75 mmol/L, p = 0.004), fat mass (-7.6 kg, p < .0001) and abdominal fat (-310 cm3, p < .0001). CONCLUSION: Weight loss reduced OSA and improved sleepiness and cardiometabolic health. These improvements were not further enhanced by using CPAP. Results suggest weight loss should be the primary focus of treatment for patients with OSA and obesity.