Journal of Clinical Sleep Medicine最新文献

Study objectives: Obstructive sleep apnea (OSA) is considered a risk factor for sleepiness at the wheel (SW) and near-miss accidents (NMA). To date, there are subjective and objective methods such as the Maintenance of Wakefulness Test (MWT) to investigate sleepiness. However, these methods have limitations. Therefore, a new analysis of the MWT was introduced based on the identification of microsleeps (MS). So, we tested MS analysis to improve the discriminative ability of MWT in recognizing individuals at risk for SW and NMA in a population with OSA.

Methods: The study was conducted on 100 naïve patients with suspected OSA referred to our Sleep Medicine Unit. All patients performed a full-standard polysomnography and MWT. The MWT was analyzed according to standard criteria, and the presence of MS episodes, the mean MS latency and the MS density (the mean absolute or relative number of MS) were assessed.

Results: Microsleeps were observed in 100% of alert or sleepy patients and 47% of the fully alert (P <0.0001). Almost 90% of patients reporting NMA showed episodes of MS during MWT. The occurrence of NMA was related to EDS, MS latency and MS density (P <0.001). The discriminative power for the NMA of MS density measures was higher than that derived from latency analysis, particularly in patients without EDS and with a simultaneous mean sleep latency >33 minutes.

Conclusions: MS analysis provides objective evidence of sleepiness and, therefore, could improve the discriminative ability of the MWT in recognizing individuals at high risk for accidents.

Study objectives: Sleep-disordered breathing (SDB) is prevalent in children with sickle cell disease (SCD) and is associated with worse outcomes. This study aims to compare the outcomes of polysomnography (PSG) performed for pediatric patients with SCD at three United States centers.

Methods: We included children with SCD aged 0-21 who underwent PSG at three American Academy of Sleep Medicine accredited centers: the University of Alabama at Birmingham (UAB), the University of Florida (UF), and Duke University Hospital (DUH), between 2012 and 2022. Descriptive statistics were used as appropriate to compare the baseline characters and PSG outcomes among the different centers.

Results: A total of 210 children with SCD from the three centers were included, with comparable sex, SCD genotypes, hemoglobin, hematocrit levels, and chronic transfusion. Children from the different centers exhibited variations in age (P < .001), BMI (P < .05), mean corpuscular volume (P < .05), and hydroxyurea use (P < .05) at the time of the PSG. Overall, the three centers showed significantly different PSG outcomes. Patients from UF had worse obstructive sleep apnea, oxygenation, and periodic leg movement events, together with lower hydroxyurea usage. While those from DUH showed higher hypoventilation and arousal indices.

Conclusions: This multicenter study underscores variations in PSG outcomes among pediatric SCD patients at different centers in the Southeast United States. These findings emphasize the need for standardized approaches to screen for SDB, refer to PSG, and interpret the results in children with SCD. These conclusions may apply to other genetic disorders associated with an increased risk of sleep-disordered breathing.

Study objectives: While various wearable EEG devices have been developed, performance evaluation of the devices and their associated automated sleep stage classification models is mostly limited to healthy subjects. A major barrier for applying automated wearable EEG sleep staging in clinical populations is the need for large-scale data for model training. We therefore investigated transfer learning as strategy to overcome limited data availability and optimize automated single-channel EEG sleep staging in people with sleep disorders.

Methods: We acquired 52 single-channel frontopolar headband EEG recordings from a heterogeneous sleep-disordered population with concurrent PSG. We compared three model training strategies: 'pre-training' (i.e., training on a larger dataset of 901 conventional PSGs), 'training-from-scratch' (i.e., training on wearable headband recordings), and 'fine-tuning' (i.e., training on conventional PSGs, followed by training on headband recordings). Performance was evaluated on all headband recordings using 10-fold cross-validation.

Results: Highest performance for 5-stage classification was achieved with fine-tuning (κ = .778), significantly higher than with pre-training (κ = .769) and with training-from-scratch (κ = .733). No significant differences or systematic biases were observed with clinically relevant sleep parameters derived from PSG. All sleep disorder categories showed comparable performance.

Conclusions: This study emphasizes the importance of leveraging larger available datasets through deep transfer learning to optimize performance with limited data availability. Findings indicate strong similarity in data characteristics between conventional PSG and headband recordings. Altogether, results suggest the combination of the headband, classification model, and training methodology can be viable for sleep monitoring in the heterogeneous clinical population.

Study objectives: To evaluate efficacy of vibrotactile positional therapy (PT) compared to standard CPAP therapy in mild-to-moderate positional obstructive sleep apnea (pOSA).

Methods: Prospective crossover randomized controlled trial of adult patients with treatment-naïve, symptomatic, mild-to-moderate pOSA - defined as ≥5 total apnea-hypopnea index (AHI) <30 with supine-to-non-supine (s:ns)AHI ratio ≥2. Participants were randomized to in-laboratory treatment initiation polysomnography with either PT or CPAP on sequential nights before an eight-week trial of each therapy. The primary endpoint was symptomatic improvement (Epworth Sleepiness Scale; ΔESS). Secondary endpoints included patient preference, usage, sleep architecture and quality of life (QoL) measures.

Results: 52 participants were enrolled and completed both arms of the study. Participants were symptomatic with median ESS 12 (IQR 10-14). Treatment resulted in a significant (p<0.001) symptomatic improvement with both PT and CPAP (ΔESS 4; IQR 6-11) without a significant difference between treatment arms (p=0.782). PT was effective at restricting supine sleep and demonstrated improved sleep efficiency compared with CPAP, although no better than baseline. Both therapies were effective at reducing AHI, although CPAP demonstrated superior AHI reduction. There were otherwise no clinically significant differences in sleep architecture, usage, or secondary outcomes including overall patient preference.

Conclusions: In this cohort, treatment with PT or CPAP resulted in clinically significant symptomatic improvement (ΔESS) that was not significantly different between treatment arms. No real difference was seen in other secondary outcome measures. This study provides further evidence to support the use of PT as an alternative first-line therapy with CPAP in appropriately selected patients with pOSA.

Clinical trial registration: Registry: Australian New Zealand Clinic Trials Registry; Name: Prospective crossover trial of Positional and Continuous positive airway pressure Therapy for the treatment of mild-to-moderate positional obstructive sleep apnoea; Identifier: ACTRN12619000475145; URL: https://anzctr.org.au/Trial/Registration/TrialReview.aspx?id=377221&isReview=true.

Study objectives: Inflammatory and immune mechanisms are considered in restless legs syndrome (RLS) pathophysiology with several autoimmune diseases associated with RLS. There is a paucity of studies examining RLS prevalence in myasthenia gravis (MG), an autoimmune neuromuscular disease. This study investigated RLS prevalence and association with patient-reported measures in a large registry of participants with MG using a validated RLS diagnostic questionnaire.

Methods: The Myasthenia Gravis Foundation of America MG Patient Registry is used on a semi-annual basis to survey participants with MG. Patients aged ≥18 years, living in the United States, and answering "Yes" to physician diagnosed MG were invited by email to enroll in an RLS-customized web-based survey. Collection of data included demographics, disease variables, patient-reported measures with a simple depression scale, MG 15-item Quality of Life (MG-QOL15r), MG-Activities of Daily Living (MG-ADL) instruments, and 13-item short-form Cambridge-Hopkins diagnostic questionnaire for RLS (CH-RLSq13). Multivariable logistic regression models explored the association between RLS and MG variables of interest.

Results: 630 eligible participants with MG (age: 62.8±13.2; 54.9% female; 91.6% White) completed the survey. The overall prevalence of RLS was 14.8%. The prevalence of clinically significant RLS was 8.4%. The odds of having RLS were increased with higher (worse) MG-ADL, MG-QOL15r, and depression scores. History of "Thymic tumor with thymectomy" and "CPAP therapy" were also independent predictors of RLS.

Conclusions: RLS is common in patients with MG and is associated with worse functional status, quality of life, and depression. The thymus could play a key role in an autoimmune process associating MG with RLS.

Radiotherapy for head and neck cancer can trigger or worsen obstructive sleep apnea (OSA) due to factors such as neurological impairment, muscle atrophy, edema, and xerostomia. We present a case in which a patient developed severe OSA three months after undergoing radiochemotherapy and neck dissection for squamous cell carcinoma of the right lingual tonsil with regional metastasis. Polysomnography confirmed severe OSA. Unfortunately, the patient did not adhere to the recommended treatment. The symptoms persisted for six months but then, unexpectedly, resolved completely. A follow-up polysomnography conducted one year after radiotherapy showed no evidence of OSA. This case suggests that a recent diagnosis of OSA following head and neck radiotherapy may not be permanent and should not necessarily lead to a lifelong prognosis of sleep-related breathing disorders.

Health campaigns hold promise for promoting general awareness about OSA. In 2023, the American Academy of Sleep Medicine developed a series of messages as part of their "Count on Sleep" campaign. Three distinct messaging strategies were employed in posts disseminated on Facebook: (1) positive outcome (benefits of OSA treatment); (2) negative outcome (consequences of untreated OSA); and (3) partner-focused. We evaluated Facebook analytics to explore which, of the three strategies, was most impactful. The impressions were highest for the positive outcome message (n=120,062), followed by the negative outcome (n=12,286) and partner-focus (n=10,259) messages. Female users were more likely to engage with positive (39% v. 28%) and negative (34% v. 25%) message than males, but genders were quite balanced in engagement with the partner-focus message (36% v. 36%). There was more engagement from older adults (65 years and older) than younger adults across all message types.

Study objectives: Despite frequent sleep disruption in the paediatric intensive care unit (PICU), bedside sleep monitoring in real-time is currently not available. Supervised machine learning (ML) applied to electrocardiography (ECG) data may provide a solution, since cardiovascular dynamics are directly modulated by the autonomic nervous system (ANS) during sleep.

Methods: Retrospective study using hospital-based polysomnography (PSG) recordings obtained in non-critically ill children between 2017 and 2021. Six age categories were defined: 6-12 months, 1-3 years, 3-5 years, 5-9 years, 9-13 years and 13-18 years. Features were derived in time, frequency and non-linear domain from pre-processed ECG data. Sleep classification models were developed for two, three, four and five state using logistic regression (LR), random forest (RF) and XGBoost (XGB) classifiers during five-fold nested cross-validation. Models were additionally validated across age categories.

Results: A total of 90 non-critically ill children were included with a median (Q1, Q3) recording length of 549.0 (494.8, 601.3) minutes. The three models obtained AUROC 0.72 - 0.78 with minimal variation across classifiers and age categories. Balanced accuracies were 0.70 - 0.72, 0.59 - 0.61, 0.50 - 0.51 and 0.41 - 0.42 for two, three, four and five state. Generally, the XGB model obtained the highest balanced accuracy (p < 0.05), except for five state where LR excelled (p = 0.67).

Conclusions: ECG-based ML models are a promising and non-invasive method for automated sleep classification directly at the bedside of non-critically ill children aged 6 months to 18 years. Models obtained moderate-to-good performance for two and three state classification.

A 72-year-old patient had a severe sleep apnea syndrome well controlled for many years through CPAP therapy. When switching to a newer device with upgraded functions, therapy completely failed. A video recording performed by the patient's wife showed high frequency mask movements suggesting inability to maintain a therapeutic pressure with high frequency pressure fluctuations, confirmed afterwards during full night polysomnography and in a bench study. CPAP manufacturers may put on the market new devices with supposedly better algorithms that in fact may have escaped serious pre-marketing evaluation and that may jeopardize the efficacy of a well proven treatment. We suggest that better evaluations are necessary before marketing therapeutic devices, and that post-marketing assessment of unanticipated side effects should become the norm.