BACKGROUND: Early detection of sleep apnea, the health condition where airflow either ceases or decreases episodically during sleep, is crucial to initiate timely interventions and avoid complications. Wearable artificial intelligence (AI), the integration of AI algorithms into wearable devices to collect and analyze data to offer various functionalities and insights, can efficiently detect sleep apnea due to its convenience, accessibility, affordability, objectivity, and real-time monitoring capabilities, thereby addressing the limitations of traditional approaches such as polysomnography.
OBJECTIVE: The objective of this systematic review was to examine the effectiveness of wearable AI in detecting sleep apnea, its type, and its severity.
METHODS: Our search was conducted in 6 electronic databases. This review included English research articles evaluating wearable AI\'s performance in identifying sleep apnea, distinguishing its type, and gauging its severity. Two researchers independently conducted study selection, extracted data, and assessed the risk of bias using an adapted Quality Assessment of Studies of Diagnostic Accuracy-Revised tool. We used both narrative and statistical techniques for evidence synthesis.
RESULTS: Among 615 studies, 38 (6.2%) met the eligibility criteria for this review. The pooled mean accuracy, sensitivity, and specificity of wearable AI in detecting apnea events in respiration (apnea and nonapnea events) were 0.893, 0.793, and 0.947, respectively. The pooled mean accuracy of wearable AI in differentiating types of apnea events in respiration (normal, obstructive sleep apnea, central sleep apnea, mixed apnea, and hypopnea) was 0.815. The pooled mean accuracy, sensitivity, and specificity of wearable AI in detecting sleep apnea were 0.869, 0.938, and 0.752, respectively. The pooled mean accuracy of wearable AI in identifying the severity level of sleep apnea (normal, mild, moderate, and severe) and estimating the severity score (Apnea-Hypopnea Index) was 0.651 and 0.877, respectively. Subgroup analyses found different moderators of wearable AI performance for different outcomes, such as the type of algorithm, type of data, type of sleep apnea, and placement of wearable devices.
CONCLUSIONS: Wearable AI shows potential in identifying and classifying sleep apnea, but its current performance is suboptimal for routine clinical use. We recommend concurrent use with traditional assessments until improved evidence supports its reliability. Certified commercial wearables are needed for effectively detecting sleep apnea, predicting its occurrence, and delivering proactive interventions. Researchers should conduct further studies on detecting central sleep apnea, prioritize deep learning algorithms, incorporate self-reported and nonwearable data, evaluate performance across different device placements, and provide detailed findings for effective meta-analyses.

BACKGROUND: Over the past three decades, our understanding of sleep apnea in women has advanced, revealing disparities in pathophysiology, diagnosis, and treatment compared to men. However, no real-life study to date has explored the relationship between mask-related side effects (MRSEs) and gender in the context of long-term CPAP.
METHODS: The InterfaceVent-CPAP study is a prospective real-life cross-sectional study conducted in an apneic adult cohort undergoing at least 3 months of CPAP with unrestricted mask-access (34 different masks, no gender specific mask series). MRSE were assessed by the patient using visual analog scales (VAS). CPAP-non-adherence was defined as a mean CPAP-usage of less than 4 h per day. The primary objective of this ancillary study was to investigate the impact of gender on the prevalence of MRSEs reported by the patient. Secondary analyses assessed the impact of MRSEs on CPAP-usage and CPAP-non-adherence depending on the gender.
RESULTS: A total of 1484 patients treated for a median duration of 4.4 years (IQ25-75: 2.0-9.7) were included in the cohort, with women accounting for 27.8%. The prevalence of patient-reported mask injury, defined as a VAS score ≥ 5 (p = 0.021), was higher in women than in men (9.6% versus 5.3%). For nasal pillow masks, the median MRSE VAS score for dry mouth was higher in women (p = 0.039). For oronasal masks, the median MRSE VAS score for runny nose was higher in men (p = 0.039). Multivariable regression analyses revealed that, for both women and men, dry mouth was independently and negatively associated with CPAP-usage, and positively associated with CPAP-non-adherence.
CONCLUSIONS: In real-life patients treated with long-term CPAP, there are gender differences in patient reported MRSEs. In the context of personalized medicine, these results suggest that the design of future masks should consider these gender differences if masks specifically for women are developed. However, only dry mouth, a side effect not related to mask design, impacts CPAP-usage and non-adherence.
BACKGROUND: INTERFACEVENT IS REGISTERED WITH CLINICALTRIALS.GOV (NCT03013283).FIRST REGISTRATION DATE IS 2016-12-23.

OBJECTIVE: To determine the impact and best management sequence between adenotonsillectomy (AT) and rapid palatal expansion (RPE) on the apnea-hypopnea index (AHI) and minimum oxygen saturation (MinSaO2) in nonobese pediatric obstructive sleep apnea (OSA) patients presenting balanced maxillomandibular relationship.
METHODS: Thirty-two nonobese children with balanced maxillomandibular relationship and a mean age of 8.8 years, with a graded III/IV tonsillar hypertrophy and maxillary constriction, participated in a cross-over randomized controlled trial. As the first intervention, one group underwent AT while the other underwent RPE. After 6 months, interventions were switched in those groups, but only to participants with an AHI > 1 after the first intervention. OSA medical diagnosis with the support of Polysomnography (PSG) was conducted before (T0), 6 months after the first (T1) and the second (T2) intervention. The influence of sex, adenotonsillar hypertrophy degree, initial AHI and MinSaO2 severity, and intervention sequence were evaluated using linear regression analysis. Intra- and intergroup comparisons for AHI and MinSaO2 were performed using ANOVA and Tukey\'s test.
RESULTS: The initial AHI severity and intervention sequence (AT first) explained 94.9% of AHI improvement. The initial MinSaO2 severity accounted for 83.1% of MinSaO2 improvement changes. Most AHI reductions and MinSaO2 improvements were due to AT.
CONCLUSIONS: Initial AHI severity and AT as the first intervention accounted for most of the AHI improvement. The initial MinSaO2 severity alone accounted for the most changes in MinSaO2 increase. In most cases, RPE had a marginal effect on AHI and MinSaO2 when adjusted for confounders.

OBJECTIVE: Advanced signal processing of photoplethysmographic data enables novel analyses which may improve the understanding of the pathogenesis of dysglycemia associated with sleep disorders. We aimed to identify sleep-related pulse wave characteristics in diabetic patients compared to normoglycemic individuals, independent of cardiovascular-related comorbidities.
METHODS: This cross-sectional evaluation of the population-based Swedish CArdioPulmonary bioImage Study (SCAPIS) included overnight oximetry-derived pulse wave data from 3997 subjects (45 % males, age 50-64 years). Metabolic status was classified as normoglycemic (n = 3220), pre-diabetic (n = 544), or diabetic (n = 233). Nine validated pulse wave features proposed to influence cardiovascular risk were derived and compared between metabolic status groups. Logistic prediction models and genetic matching were applied to capture diabetes-related pulse wave characteristics during sleep. The model was controlled for anthropometrics, lifestyle, sleep apnea, and in the final adjustment even for cardiometabolic factors like dyslipidaemia, hypertension, and coronary artery calcification.
RESULTS: Pulse wave-derived parameters differed between normoglycemic and diabetic individuals in eight dimensions in unadjusted as well as in the partially adjusted model (anthropometric factors and sleep apnea, p ≤ 0.001). All covariates confirmed significant differences between normoglycemic and diabetic subjects (all p ≤ 0.001). Reduced cardio-respiratory coupling (respiratory-related pulse oscillations) (β = -0.010, p = 0.012), as well as increased vascular stiffness (shortened pulse propagation time (β = -0.015, p = 0.001), were independently associated with diabetes even when controlled for cardiometabolic factors. These results were confirmed through a matched cohort comparative analysis.
CONCLUSIONS: Photoplethysmographic pulse wave analysis during sleep can be utilized to capture multiple features of modified autonomic regulation and cardiovascular consequences in diabetic subjects. Dampened heart rate variability and increased vascular stiffness during sleep showed the strongest associations with diabetes.

The risk factors of hospitalized COVID-19 and obstructive sleep apnea (OSA) overlap. The aim of this study is to evaluate the prevalence and associated factors of post-COVID-19 OSA in hospitalized adult patients from southeastern Romania. A follow-up study was conducted on patients hospitalized for COVID-19 at the Pneumology Hospital in Galati, Romania, between 2021 and 2022. OSA was evaluated using the Epworth and STOP-BANG questionnaires and nocturnal polygraphy monitoring. Out of 331 patients, 257 were evaluated for sleep apnea in the 12th week. The prevalence of severe OSA was 57.97%. Significant associations were found with male gender, an age over 60, obesity, and cardiovascular co-morbidities. Non-invasive ventilatory therapy (NIV) and a hygienic-dietary regimen were recommended based on severity following a control visit after a month. Developing strategies for diagnosing and monitoring sleep disorders, including home sleep apnea tests and patient education, are the next directions for post-COVID-19 management.

Fibromyalgia (FM) is a chronic non-inflammatory disorder mainly characterized by widespread musculoskeletal pain, fatigue, sleep disturbances, and a constellation of other symptoms. For this reason, delineating a clear distinction between pure FM and FM-like picture attributable to other common diseases can be extremely challenging. Physicians must identify the most significant confounders in individual patients and implement an appropriate diagnostic workflow, carefully choosing a minimal (but sufficient) set of tests to be used for identifying the most plausible diseases in the specific case. This article discusses prevalent non-rheumatological conditions commonly observed in the general population that can manifest with clinical features similar to primary FM. Given their frequent inclusion in the differential diagnosis of FM patients, the focus will be on elucidating the distinctive clinical characteristics of each condition. Additionally, the most cost-effective and efficient diagnostic methodologies for accurately discerning these conditions will be examined.

Sleep disorders are common, and largely undiagnosed in early-career workers. The combination of sleep disorders and shift work has implications for mental health, workplace safety, and productivity. Early identification and management of sleep disorders is likely to be beneficial to workers, employers and the community more broadly. We assessed the feasibility and acceptability of a tailored sleep disorder screening and management pathway for individuals with future shift work requirements. Paramedic students were invited to complete an online sleep health survey, which included validated sleep disorder screening questionnaires for insomnia, obstructive sleep apnea and restless legs syndrome. Participants were able to express interest in participating in a sleep monitoring and management study. Participants at risk for a sleep disorder were identified, contacted by the study physician (RJA), notified of their sleep disorder screening results and provided with information regarding management options. Feasibility of the screening and management pathways were determined by completion of the 12 week follow-up, and ability to engage with health services for diagnostic testing or treatment. Acceptability of these pathways was assessed with a semi-structured interview on completion of the study at 12 weeks. Screening was completed in thirty participants (mean age 22.5 ± 6.7, 63% female), 17 of whom were \'at-risk\' for a sleep disorder and offered a management pathway. All participants engaged with the study physician (RJA), with 16 completing the study (94% completion rate). Three participants with excessive daytime sleepiness received feedback from the study physician (RJA) and no further care required. Of the remaining 14 participants, 11 (78%) engaged with health services after speaking with the study physician (RJA). Those who engaged with diagnostic and management services reported that a structured pathway with online screening was convenient and easy to follow. Facilitating screening and management of sleep disorders in students with future shift work requirements is both feasible and acceptable. These findings can inform the development of a preventive strategy for sleep disorders and ideally, a health services feasibility trial for future shift workers.

Little is known about the physiological and biomechanical factors that determine individual preferences in lying posture during sleep. This study investigated relationships between position preference and position-specific arousals, awakenings, limb movements and limb movement arousals to explore the mechanisms by which biomechanical factors influence position preference. Forty-one mature-aged adults underwent 2 nights of at-home polysomnography ~2 weeks apart, on a standardised firm foam mattress, measuring nocturnal sleep architecture and position. The lateral supine ratio and restlessness indices specific to lateral and supine positions including limb movement index, limb movement arousal index, arousal index, wake index, respiratory arousal index and apnea-hypopnea index were calculated and analysed via linear mixed-effects regression. In the supine position, all restlessness indices were significantly increased compared with the lateral position, including a 379% increase in respiratory arousals (β = 7.0, p < 0.001), 108% increase in arousal index (β = 10.3, p < 0.001) and 107% increase in wake index (β = 2.5, p < 0.001). Wake index in the supine position increased significantly with more lateral sleep (β = 1.9, p = 0.0013), and significant correlation between lateral supine ratio polysomnography 1 and lateral supine ratio polysomnography 2 (β = 0.95, p < 0.001) indicated strong consistency in sleep preference. Overall, the findings suggest that some individuals have low tolerance to supine posture, represented by a comparatively high wake index in the supine position, and that these individuals compensate by sleeping a greater proportion in the lateral position.

OBJECTIVE: This study aims to develop sleep apnea screening models with overnight SpO2 data, and to investigate the impact of the SpO2 data granularity on model performance.
METHODS: A total of 7,718 SpO2 recordings from the SHHS and MESA datasets were used. Probabilistic ensemble machine learning was employed to predict sleep apnea status at three AHI cutoff points: ≥ 5, ≥ 15, and ≥ 30 events/hour. To investigate the impact of data granularity, SpO2 data were aggregated at 30, 60, and 300 s.
RESULTS: Our models demonstrated good to excellent performance on internal test, with average area under the curve (AUC) values of 0.91, 0.93, and 0.96 for cutoffs ≥ 5, ≥ 15, and ≥ 30 at data granularity of 1 s, respectively. Both sensitivity (0.76, 0.84, 0.89) and specificity (0.87, 0.86, 0.90) ranged from good to excellent across three cutoffs. Positive predictive values (PPV) ranged from excellent to fair (0.97, 0.83, 0.66), and negative predictive values (NPV) ranged from low to excellent (0.43, 0.87, 0.98). Model performance on external test slightly dropped compared to internal test, but still achieved good to excellent AUC above 0.80 across all data granularity and all the three cutoffs. Data granularity of 300 s led to a reduction in performance metrics across all cutoffs.
CONCLUSIONS: Our models demonstrated superior performance across all three AHI cutoff thresholds compared to existing large sleep apnea screening models, even when considering varying SpO2 data granularity. However, lower data granularity was associated with decreased screening performance, indicating a need for further research in this area.

BACKGROUND: Sleep apnea (SA) and insomnia (INS) are prevalent sleep disorders among referrals to sleep clinics. People with comorbid insomnia and sleep apnea (COMISA) suffer both disorders simultaneously. The epidemiology of COMISA is not well known in the Middle East including Iran. We hypothesized that COMISA is prevalent in metropolitan sleep clinic cohorts in Iran.
METHODS: The records of patients aged < 95 years referred to sleep disorders centers in four large metropolitan areas including Kermanshah, Tabriz, Shiraz, and Ahvaz were examined. Polysomnography (PSG) was performed in all these patients in specialized centers and the results were scored by a trained technician and interpreted by a sleep specialist. SA was defined as an Apnea-Hypopnea Index (AHI ≥ 5), INS was defined by psychiatrists according to self-report and clinical interviews, and COMISA was defined if both disorders were present. Participants with neither condition were included in as comparator group. One-way ANOVAs, correlation, and linear/logistic regression analyses were used.
RESULTS: This study included 1807 patients (Mean age 49.3, SE ± 13.7; 38.8% Female). Comparator, INS, SA and COMISA made up 7.2%, 16%, 50.2% and 26.6% of the sample, respectively. Logistic regression analyses showed that male gender, older age, and increasing neck circumference, but not BMI, were associated with COMISA. Epworth Sleepiness Scale scores were lower in INS (5.39 ± 5.78) compared to the other three groups.
CONCLUSIONS: COMISA is a prevalent condition in metropolitan sleep centers in Iran among participants referred to sleep centers. The data showed that male gender and age were associated significantly with COMISA.