The study, led by Liz E. Lezama Oropeza, with colleagues Elisa Basora-Rovira and Aarti Shakkottai, reviewed polysomnographic data from fifteen infants between zero and twelve months of age who were born at term and had no identified syndromes. The mean gestational age of the cohort was 38.4 weeks, with a narrow spread of plus or minus 0.6 weeks, and eighty percent of the infants were male. The average age at the first sleep study was 7.5 months, with a standard deviation of 3.6 months, meaning the group spanned early infancy through the first birthday.

Polysomnography, the gold-standard overnight sleep test, records brain waves, breathing effort, airflow, oxygen saturation, carbon dioxide levels, and body movements to quantify how often breathing is disrupted during sleep. In this cohort, the average apnea-hypopnea index, or AHI, which counts the number of apneas and hypopneas per hour of sleep, was 23.1 events per hour. The mean obstructive AHI, which isolates events caused by physical blockage of the upper airway, was 20 events per hour, while the mean central apnea index, reflecting pauses in breathing driven by immature or irregular brainstem signaling rather than airway obstruction, was 3.1 events per hour. For context, an obstructive AHI above 5 events per hour is generally considered moderate sleep apnea in pediatric populations, so the average infant in this series fell well into the moderate-to-severe range.

Oxygenation data added another layer of concern. Mean oxygen saturation during sleep was 97.5 percent, but the mean nadir, the lowest saturation reached during any single desaturation event, was 81.0 percent, with a wide standard deviation of 9.9 percent. Six of the fifteen infants met formal criteria for sleep-related hypoxemia, defined in this study as oxygen saturation below 90 percent for at least five minutes during the recording. Carbon dioxide measurements told a different story: mean and peak carbon dioxide levels were 35 and 48.6 millimeters of mercury, respectively, and none of the patients met criteria for sleep-related hypoventilation, which would require elevated carbon dioxide above 50 millimeters of mercury for more than a quarter of total sleep time. The pattern suggests that in these otherwise healthy infants, the dominant problem was intermittent airway obstruction causing repeated oxygen dips rather than sustained underventilation.

The clinical picture preceding referral was often subtle. Only a third of the infants had documented snoring, the symptom most parents and clinicians associate with sleep apnea, and nearly half, forty-seven percent, carried a diagnosis of gastroesophageal reflux, a condition that has been linked to airway inflammation and laryngeal hypersensitivity in infants. Seventy-three percent of the infants underwent some form of upper airway evaluation, such as flexible laryngoscopy, and seven of those evaluations revealed abnormal findings. Laryngomalacia, a condition in which immature, floppy supraglottic tissue collapses inward during inspiration, is the most common congenital cause of stridor and obstructive events in infants and featured among the abnormal findings in this series.

Follow-up polysomnography, obtained in ten of the fifteen infants at an average age of around 13.4 months, provided the study’s most striking finding. Four of the ten infants showed complete resolution of their sleep-disordered breathing without any surgical or device-based intervention. One infant achieved resolution while using continuous positive airway pressure, or CPAP, a machine that delivers a steady stream of pressurized air through a mask to splint the airway open during sleep, but was subsequently lost to follow-up. In the remaining five infants, sleep-disordered breathing persisted but improved. The researchers also examined whether disease severity tracked with age, finding a weak negative correlation between AHI and age, with a Spearman rho of minus 0.31 and a p-value of 0.26, which did not reach statistical significance in this small sample but pointed in the direction of decreasing severity as infants matured.

The authors emphasize that the frequency of moderate-to-severe obstructive sleep apnea in this population is high, but that the clinical implications remain unclear. Diagnostic thresholds for sleep-disordered breathing in infants are poorly standardized, and much of the pediatric literature derives normative values from older children or from high-risk infant populations whose airway physiology differs substantially. The infant upper airway is anatomically distinct: narrow cartilaginous structures, predominantly nasal breathing, high airway compliance, and immature neuromuscular control of the pharyngeal dilator muscles all predispose to collapse during sleep, particularly during active or rapid eye movement sleep when muscle tone falls. As infants grow, the airway enlarges, adenoid and tonsillar tissue relative to the airway changes, and neuromuscular control matures, which may explain why many infants in this series appeared to outgrow their disease.

The study has clear limitations that the authors acknowledge. As a single-center case series of fifteen infants, it cannot establish prevalence in the broader population of healthy term infants, since all participants were referred for suspected sleep-disordered breathing at a tertiary care center. The weak correlation between age and AHI did not reach statistical significance, and the follow-up cohort was small, with five of fifteen infants lacking repeat sleep studies. The authors also note that there is limited data on effective treatment strategies for this group, and that the natural history of untreated moderate-to-severe disease in otherwise healthy infants remains poorly characterized.

Nevertheless, the findings carry practical weight for parents and clinicians. Snoring in a full-term baby without a syndrome is not a reliable screening sign, appearing in only a third of this cohort, and reflux may be a more common companion diagnosis than previously appreciated. The authors conclude that further research is warranted to refine diagnostic thresholds and standardize the diagnostic evaluation and interventions for sleep-disordered breathing in this unique population. Until larger, prospective studies establish normative infant polysomnography values and clarify which infants need intervention versus watchful waiting, the study suggests that clinicians evaluating otherwise healthy infants for suspected sleep-disordered breathing should maintain a low threshold for overnight testing, while recognizing that a substantial share of these babies may simply grow out of it.

Subject of Research: A case series of fifteen term, non-syndromic infants referred for suspected sleep-disordered breathing found high rates of moderate-to-severe obstructive sleep apnea on polysomnography, with many infants outgrowing the condition by follow-up.

Article Title: Correlation between phenotype and polysomnography findings of sleep-disordered breathing in term, non-syndromic infants: a case series study

Article References: Lezama, L., Basora-Rovira, E., & Shakkottai, A. (2026). Correlation between phenotype and polysomnography findings of sleep-disordered breathing in term, non-syndromic infants: a case series study. Frontiers in Sleep, 5, Article 1877709. https://doi.org/10.3389/frsle.2026.1877709

Image Credits: AI Generated

DOI: 10.3389/frsle.2026.1877709

Keywords: infant sleep apnea, sleep-disordered breathing, polysomnography, apnea-hypopnea index, laryngomalacia, gastroesophageal reflux, oxygen saturation, hypoxemia, CPAP, term infants, Frontiers in Sleep, pediatric sleep

Obesity hypoventilation syndrome is defined by the combination of obesity and chronically elevated carbon dioxide levels in the blood, a state known as chronic alveolar hypoventilation. In affected individuals, the mechanical burden of excess body weight on the chest wall and diaphragm, together with central respiratory drive abnormalities frequently intertwined with obstructive sleep apnea, prevents the lungs from clearing carbon dioxide adequately during wakefulness and sleep. The result is chronically elevated arterial or bicarbonate-derived carbon dioxide, often accompanied by low oxygen levels that worsen at night. Left untreated, the syndrome drives pulmonary hypertension, cor pulmonale, and substantially elevated mortality, and it multiplies healthcare utilization as patients cycle through emergency departments and hospital wards.

The scale of the problem in hospitalized populations is well documented in the literature the Colorado team cites. A widely referenced 2004 study by Nowbar and colleagues found that obesity-associated hypoventilation was common among hospitalized patients and was associated with worse outcomes, including higher rates of intensive care admission, prolonged stays, and increased mortality. Later work by Povitz and colleagues showed that individuals with obesity and chronic hypoxemia who were treated for sleep-disordered breathing still consumed considerable healthcare resources, underscoring how often the condition is recognized late or managed incompletely. Budweiser and colleagues demonstrated that mortality in patients with obesity hypoventilation syndrome receiving noninvasive ventilation depends heavily on prognostic factors, reinforcing that early identification and appropriate therapy matter enormously for survival.

Despite this evidence base, diagnosis rates remain dismal. The American Thoracic Society’s 2019 clinical practice guideline, authored by Mokhlesi, Masa, Brozek and colleagues, laid out an official evaluation and management framework for the syndrome, yet adherence in real-world hospital settings has lagged. The guideline recommends screening at-risk hospitalized patients, confirming chronic hypoventilation with appropriate testing, and initiating positive airway pressure therapy, but translating those recommendations into bedside practice requires systems that prompt clinicians at the right moment with the right information. That is precisely the gap the electronic clinical pathway is designed to bridge.

The Colorado team’s pathway operates within the electronic health record, the same digital environment where hospitalists already spend much of their day. By embedding screening logic directly into admission and rounding workflows, the tool is intended to flag patients whose clinical profile suggests obesity hypoventilation syndrome, prompt confirmatory evaluation such as arterial blood gas or serum bicarbonate assessment followed by sleep testing, and guide clinicians toward guideline-concordant management including initiation of noninvasive ventilation when indicated. The approach reflects a broader movement in implementation science: rather than relying on clinicians to remember guideline details amid the cognitive load of inpatient medicine, decision support systems deliver the guideline at the point of care, where it can actually change behavior.

The technical rationale for embedding such pathways is grounded in the physiology of the disease itself. Obesity hypoventilation syndrome is frequently confused with simple obesity or with obstructive sleep apnea alone, yet its management differs in important ways. Patients with the syndrome often require higher levels of positive airway pressure support than patients with uncomplicated sleep apnea, and a subset needs noninvasive ventilation with pressure support to augment ventilation rather than simple continuous positive airway pressure. The 2021 technical expert panel report from the American College of Chest Physicians, the American Association for Respiratory Care, the American Academy of Sleep Medicine, and the American Thoracic Society, published in Chest by Mokhlesi and colleagues, detailed optimal noninvasive ventilation strategies for hypoventilation syndromes, including titration approaches and follow-up structures. An electronic pathway can encode these nuances, ensuring that a hospitalist encountering an at-risk patient receives not just a flag but a coherent, evidence-based management sequence.

The reimbursement landscape adds another layer of urgency to the project. The Centers for Medicare and Medicaid Services maintains compliance guidance for respiratory assist devices, the equipment category covering noninvasive ventilators used for chronic hypoventilation, and documentation requirements for these devices are stringent. Hospitals that fail to document the diagnostic workup and clinical rationale for noninvasive ventilation risk claim denials, while patients whose documentation is incomplete may face delays in obtaining home equipment. An electronic pathway that structures the diagnostic evaluation and generates the necessary documentation serves a dual purpose: it improves clinical care and it aligns the medical record with payer requirements, a combination that hospital administrators and clinicians alike have reason to welcome.

The Colorado group’s report is explicitly framed as a concise research report rather than a full trial, and the authors are careful about what it does and does not establish. The manuscript, which received no external financial support and was approved as exempt by the Colorado Multiple Institutions Review Board under protocol number 25-0674, describes the pathway’s design and its intended role in inpatient care. The authors, who include Camacho, Andani, Welsh, Burden, and Green, report no relevant conflicts of interest, and they note that the views expressed are their own rather than official positions of their affiliated institutions, which span the University of Colorado School of Medicine, University of Colorado Hospital at the Anschutz Medical Campus, the University of South Florida, and the Eastern Colorado Veterans Affairs Medical Center.

What makes the work notable for the broader clinical community is its positioning at the intersection of several converging pressures. Hospitalized patients with obesity hypoventilation syndrome represent a high-risk, high-cost population whose condition is systematically missed, the guideline infrastructure to treat them already exists, and the technology to deliver that guidance at scale is now standard in hospital information systems. The remaining challenge, which implementation research has repeatedly shown to be the hardest one, is changing clinician behavior reliably and sustainably. Electronic pathways succeed when they are trusted, unobtrusive, and tightly integrated into existing workflows, and they fail when they generate alert fatigue or conflict with clinical judgment. The Colorado team’s design choices will therefore be scrutinized as much for their usability as for their clinical content.

For patients, the stakes of closing the diagnosis gap are concrete. Untreated obesity hypoventilation syndrome shortens lives, worsens quality of life, and drives avoidable hospitalizations, while treated patients experience meaningful improvements in daytime function, sleep quality, and survival. If an electronic pathway embedded in the hospital record can convert a guideline that most hospitalists know exists into action that reliably happens at the bedside, the study’s implications extend well beyond a single institution. The full article, published by the Society of General Internal Medicine under exclusive license to Springer Nature, is available to readers via its digital object identifier, and the underlying data are available from the corresponding author upon reasonable request.

Subject of Research: Health & Medicine

Article Title: Electronic Clinical Pathway for Enhanced Diagnosis and Management of Obesity Hypoventilation Syndrome Among Hospitalized Patients

Article References: Camacho, J., Andani, S., Welsh, C., Burden, M., & Green, K. K. (2026). Electronic Clinical Pathway for Enhanced Diagnosis and Management of Obesity Hypoventilation Syndrome Among Hospitalized Patients. Journal of General Internal Medicine. https://doi.org/10.1007/s11606-026-10750-7

Image Credits: AI Generated

DOI: 10.1007/s11606-026-10750-7

Keywords: obesity hypoventilation syndrome, electronic clinical pathway, clinical decision support, hospital medicine, sleep-disordered breathing, noninvasive ventilation, American Thoracic Society guideline, electronic health records, Electronic, Clinical, Pathway, Enhanced

The clinical stakes are considerable. Obstructive sleep apnea is associated with fragmented sleep, intermittent hypoxemia, excessive daytime sleepiness, neurocognitive impairment, and increased cardiometabolic risk, yet many affected individuals remain undiagnosed or wait years for a diagnosis. The gold standard, in-laboratory polysomnography, captures a rich set of physiological signals including EEG, ECG, airflow, respiratory effort, oxygen saturation, and body position, but it is expensive, labor-intensive, and difficult to scale to population-level screening. Home sleep apnea testing has improved access, but it creates new demands for automated, robust, and interpretable analysis. The review argues that this is precisely where machine learning and deep learning can transform the field, provided the technology is validated rigorously enough for real-world deployment.

To bring order to a rapidly expanding literature, the authors propose a multidimensional taxonomy organized around four complementary dimensions: acquisition platforms, data modalities, AI methods, and clinical applications. Six primary research domains emerge from this framework: polysomnography and home sleep apnea testing signal analysis; wearable, smartphone, and contactless screening; audio and reduced-channel physiological-signal diagnosis; imaging and craniofacial anatomical risk prediction; phenotyping, endotyping, and risk stratification; and treatment management, CPAP adherence, and remote monitoring. A seventh, cross-cutting theme covers explainable, trustworthy, and clinically deployable AI, reflecting the growing recognition that interpretability, robustness, and deployment feasibility matter across every modality rather than belonging to a single one.

The historical arc the review traces is striking. The earliest phase focused on automating diagnosis from conventional physiological recordings, relying on handcrafted signal-processing features and classical machine learning classifiers for apnea detection, severity classification, and apnea-hypopnea index estimation. A second phase expanded into reduced-channel and home-oriented screening, exploring ECG and heart rate variability signals and wearable multimodal sensing as lower-burden alternatives to the sleep laboratory. The third phase marked the deep learning transition, with convolutional, recurrent, transformer-style, and multimodal architectures becoming common for PSG, HSAT, ECG, PPG, and audio signals. The fourth phase broadened the field into wearable, contactless, and multimodal assessment, incorporating wrist-worn consumer devices, sleep-sound analysis, radar-based sensing, and bed-integrated tactile sensors. Most recently, the field has moved toward trustworthy, clinically deployable AI and, at the frontier, large language models that support patient education, report summarization, clinical documentation, and decision-support workflows.

Technically, the reviewed corpus reveals a methodological landscape in transition. Deep learning, convolutional neural networks, multimodal fusion, transformer-based models, and attention mechanisms dominate recent work, while classical machine learning, feature engineering, support vector machines, and random forests remain widely used, particularly for clinical-variable and risk-stratification studies. A visualization-based synthesis of the 218 core papers, using term co-occurrence mapping and tag-based evidence maps, confirms that the literature remains strongly diagnosis-centered: OSA detection, respiratory-event detection, severity classification, screening, and AHI estimation are the dominant tasks. By contrast, treatment management, CPAP adherence support, and remote monitoring are strikingly underrepresented, with only six multi-label assignments in the taxonomy tables, suggesting a major opportunity for future work beyond diagnostic automation.

The data landscape is equally heterogeneous. PSG and HSAT studies center on airflow, EEG, SpO2, and respiratory-effort signals; wearable and contactless studies rely on PPG, actigraphy, sleep audio, accelerometry, and radar or radio-frequency sensing; imaging studies use craniofacial photographs, MRI radiomics, endoscopy, and airway morphology; and phenotyping studies depend on clinical variables, PSG-derived summaries, and comorbidity features. Representative examples span the full spectrum: pediatric OSA detection from airflow and oximetry, apnea and hypopnea localization in PSG using multi-scale object detection, transformer-based audio segmentation for fine-grained non-invasive apnea detection, radar-based apnea-hypopnea detection with hybrid CNN-LSTM networks, contactless breathing-vibration monitoring, and multimodal diagnostic prediction models combining two-dimensional facial images with clinical data.

Yet the review’s most sobering findings concern validation. Only 31 of 218 studies, or 14.2 percent, evaluated a locked model in a strict external or cross-dataset test, and only four used a prospective validation cohort. The unit of data partitioning emerged as a critical source of optimistic bias: when epochs, windows, or events from the same participant are randomly allocated to both training and test sets, performance estimates can be inflated. One audio-based study illustrates the point dramatically, achieving 88.32 percent accuracy with a subject-dependent segment split but only 68.55 percent accuracy with a subject-independent split. A cross-database analysis of heart rate variability-based apnea detection found that external-validation accuracy, sensitivity, and specificity were 20 to 40 percent lower than epoch-wise cross-validation results. These findings support participant-level or recording-level partitioning, explicit reporting of allocation units, and external testing whenever the intended use involves previously unseen patients or devices.

Performance reporting also depends heavily on the clinical endpoint and evaluation metric. A single-effort-belt model trained on MGH data yielded an external SHHS AUROC of 0.92 but an AUPRC of only 0.56, showing how class prevalence shapes interpretation. A model trained and validated on SHHS and tested on unseen MESA and MrOS cohorts reported an event-level AUROC of 0.87, an AUPRC of 0.90, and an F1-score of roughly 0.72 for AHI-severity classification. A prospectively validated commercial smartwatch algorithm achieved an AUC of 0.931 for moderate-or-greater OSA against physician-scored PSG in 53 participants, while a questionnaire-based model retained an AUC of 0.91 in a later prospective temporal cohort of 189 participants. Within-study comparisons, such as ANN and random-forest models achieving AUCs of 0.70 in a prospective SAGIC testing cohort, similar to the STOP-BANG questionnaire at 0.71, show that comparative conclusions are most reliable when models are evaluated on the same cohort, target, reference standard, and participant-separated test set.

The challenges the review identifies extend well beyond statistics. Data heterogeneity across devices, sleep laboratories, scoring protocols, and patient populations makes generalization difficult. Interpretability is essential because sleep clinicians must understand which respiratory events, oxygen desaturation patterns, sleep stages, or anatomical features influenced a prediction before acting on it; explainable ECG models and airflow-based models using Grad-CAM, LIME, SHAP, relevance maps, and attention mechanisms demonstrate how this can be achieved. Privacy and data governance concerns are amplified when data flow through home devices, mobile applications, and cloud platforms, making lightweight, on-device, energy-efficient AI particularly relevant. Large language models introduce their own risks, including hallucination, guideline non-adherence, and unclear accountability, requiring source grounding, clinical validation, and clinician oversight before deployment.

The opportunities, however, are equally compelling. Scalable screening through wearables, smartphones, and contactless sensors could identify at-risk individuals before formal sleep testing and triage patients for PSG, HSAT, or specialist evaluation. Precision sleep medicine, powered by multimodal AI that integrates physiological signals, symptoms, comorbidities, demographics, imaging, and treatment data, could move the field beyond the apnea-hypopnea index alone toward treatable traits and personalized management. Remote monitoring and adaptive management, drawing on CPAP device logs, residual events, usage patterns, and patient feedback, could reduce clinic burden while improving continuity of care. The authors conclude that future progress will depend on rigorous external validation, transparent reporting, reproducible benchmarking, privacy-preserving data governance, fairness across demographic groups, and safe integration into sleep-medicine workflows, positioning trustworthy AI and carefully validated language models as key enablers of more accessible, scalable, and personalized sleep apnea care.

Subject of Research: Artificial intelligence applications for the diagnosis and management of sleep-disordered breathing and obstructive sleep apnea

Article Title: Artificial intelligence for sleep-disordered breathing: a taxonomy of methods, data modalities, and clinical applications

Article References: Dinh, T., Lisik, D., Fournier-Viger, P., Tran, D., Yu, P. S., Van Hong, H., & Zou, D. (2026). Artificial intelligence for sleep-disordered breathing: a taxonomy of methods, data modalities, and clinical applications. Frontiers in Sleep, 5, Article 1927453. https://doi.org/10.3389/frsle.2026.1927453

Image Credits: AI Generated

DOI: 10.3389/frsle.2026.1927453

Keywords: obstructive sleep apnea, sleep-disordered breathing, artificial intelligence, machine learning, deep learning, polysomnography, wearable sensors, large language models, precision sleep medicine, explainable AI, CPAP adherence, external validation

The original article, titled “The importance of classifying central hypopneas: a call to action,” argued that hypopneas—shallow or reduced breathing events during sleep—should not all be lumped together. Hypopneas can arise through obstructive mechanisms, in which the upper airway narrows or collapses despite ongoing respiratory effort, or through central mechanisms, in which the brain’s drive to breathe transiently wanes. The distinction matters in principle because the two phenomena reflect fundamentally different physiology: one is a mechanical problem of airway patency, the other a problem of neural control of ventilation. Standard polysomnography, however, scores both events under a single apnea-hypopnea index, or AHI, and most clinical practice proceeds as though the difference were inconsequential.

Dr. Kumari’s letter, “Central hypopnea classification: bridging the gap between event identification and clinical utility,” pressed on precisely the point where the call to action was most vulnerable. She noted that the original manuscript proposed no quantitative threshold to guide clinical decision-making and offered no evidence that classifying hypopneas changes disease categorization, treatment response, or patient-centered outcomes. In other words, even if central hypopneas can be identified, no one knows how many of them constitute a meaningful disease burden, and no one has demonstrated that knowing the proportion of central events improves any measurable outcome. These are not pedantic objections; they are the standard demands of evidence-based medicine, and the reply concedes them almost entirely.

Ahn and Malhotra’s response is notable for its candor. They write that research in this area is lacking, that the field has simply not focused on classifying hypopneas, and that very little data exists to understand how hypopnea classification may impact disease categorization and treatment outcomes. The motivation behind the original paper, they explain, was exactly that—to highlight a need to better understand and classify these events. Rather than presenting a finished clinical framework, the authors describe their manuscript as a deliberate act of awareness-raising, intended to provoke research, reflection, and discussion rather than to prescribe thresholds or algorithms.

What gives the argument its urgency is the arrival of mechanistically specific treatments for sleep-disordered breathing. For most of the modern history of the field, the therapeutic arsenal was relatively blunt: continuous positive airway pressure, oral appliances, weight loss, and positional therapy, all aimed primarily at keeping the upper airway open. That era is ending. Pharmacotherapy for obstructive sleep apnea has entered clinical practice, and neurostimulation—whether hypoglossal nerve stimulation for airway control or emerging approaches targeting respiratory drive—offers interventions that act on specific physiological pathways. A drug or a stimulator designed to stabilize airway mechanics may do little for a patient whose events are predominantly central, and vice versa. If treatments are becoming mechanism-specific, the authors argue, then proper classification of hypopneas may be critical to guide interventions while optimizing health outcomes for patients with various conditions.

The reply also makes a broader and arguably more provocative point: sleep medicine cannot even agree on thresholds for the disease it knows best. The authors observe that the field struggles to identify an exact threshold for treating obstructive sleep apnea using the well-established AHI. Treatment decisions are usually based on a combination of AHI, hypoxemia, clinical symptoms, and comorbidities rather than on the index alone. The conventional severity categories—mild, moderate, and severe, corresponding to cutoffs of fewer than 5, 5 to 15, 15 to 30, and more than 30 events per hour—are explicitly described as arbitrary or consensus-based rather than biological. There is no physiological discontinuity at 15 events per hour that justifies calling one patient’s disease moderate and another’s severe; the boundaries exist because committees drew them.

The limitations of the AHI run deeper than arbitrary cutoffs. As the authors note, AHI is a poor predictor of OSA complications in general, and a clinically important improvement in AHI is challenging to define. A patient whose event count drops from 30 to 15 per hour has halved their index, but whether that change translates into reduced cardiovascular risk, improved cognition, better daytime function, or longer life is far from clear. This weakness has driven interest in alternative metrics, including the sleep apnea–specific hypoxic burden, or SASHB, a measure that quantifies the cumulative oxygen desaturation a patient experiences rather than simply counting discrete events. The authors report that SASHB may do a better job at predicting mortality and cardiovascular disease than AHI, though they immediately temper that promise: minimal data are available to establish SASHB as a validated surrogate measure. The lesson they draw is that the optimum metrics for studies in sleep-disordered breathing likely vary with the outcome of interest, and that treatment decisions are often guided, in part, by patient preferences.

This framing reframes Kumari’s critique in an instructive way. Her demand for validated thresholds and outcome data is exactly right by the standards of clinical science, but applying that standard uniquely to central hypopnea classification would hold the new question to a bar the rest of the field has never cleared. The AHI itself, the foundation of sleep medicine for decades, rests on consensus cutoffs with weak predictive power for the complications it is supposed to forecast. If the field waits for perfect validation before distinguishing central from obstructive hypopneas, it may repeat the pattern of the AHI: adopting a metric on convention, using it for decades, and only later discovering its predictive limitations. The authors’ implicit argument is that the time to start generating outcome data on hypopnea classification is now, before mechanistically specific treatments make the question unavoidable.

The scientific process, as the reply acknowledges, will take time. Building the evidence base requires studies that score hypopneas by mechanism, track patients with different central-versus-obstructive event proportions, and measure whether classification changes disease categorization, treatment selection, or patient-centered outcomes such as symptoms, quality of life, and cardiovascular risk. It requires agreement on how to identify central hypopneas reliably in the first place, since scoring rules for hypopneas have historically been contentious even before mechanism is considered. And it requires the kind of broad multidisciplinary consensus that the original call to action assembled, with authors spanning multiple institutions and subspecialties. Ahn and Malhotra report that there was broad agreement among the authors of their manuscript that investigation into hypopnea classification and how it changes management is important to pursue, and they credit Dr. Kumari with helping raise awareness of these proposed next steps.

What emerges from the exchange is a rare public acknowledgment of how much foundational work remains in a mature field. Sleep medicine can detect breathing events with exquisite precision, stage sleep across multiple physiological channels, and generate indices to many decimal places—yet the meaning of those numbers, and the boundaries of the disease they describe, rest on conventions that were never biologically derived. The central hypopnea debate is a microcosm of that larger condition. As pharmacological and neurostimulation therapies multiply, the pressure to match patients to mechanisms will only intensify, and the field will need thresholds, surrogate markers, and outcome data that currently do not exist. The reply’s authors do not pretend otherwise. Their contribution is to insist, in print and with unusual candor, that the gap between identifying an event and knowing what to do about it is the most important problem sleep medicine is not yet studying—and that filling it will take the kind of patient, unglamorous research that begins with a call to action and ends, years later, with evidence.

Subject of Research: Classification of central hypopneas in sleep-disordered breathing and its clinical utility

Article Title: Reply to “Central hypopnea classification: bridging the gap between event identification and clinical utility”

Article References: Ahn, A. P., & Malhotra, A. (2026). Reply to “Central hypopnea classification: bridging the gap between event identification and clinical utility”. Journal of Clinical Sleep Medicine, 22(1), Article 194. https://doi.org/10.1007/s44470-026-00211-7

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00211-7

Keywords: central hypopnea, sleep apnea, apnea-hypopnea index, polysomnography, sleep-disordered breathing, hypoxic burden, neurostimulation, pharmacotherapy, obstructive sleep apnea, clinical thresholds, outcomes research, sleep medicine

Sleep-disordered breathing, or SDB, encompasses a spectrum of conditions in which airflow during sleep is partially or completely obstructed. At the milder end lies habitual snoring, where the soft tissues of the upper airway vibrate noisily but breathing remains largely intact. At the severe end sits obstructive sleep apnea, or OSA, in which the airway collapses repeatedly, causing pauses in breathing, drops in blood oxygen, and abrupt micro-awakenings that fragment sleep into shards. Children with OSA often appear exhausted during the day, struggle with attention and learning, and in some cases develop behavioral problems that mimic attention-deficit hyperactivity disorder. The condition is far from rare, and its prevalence is believed to be rising alongside childhood obesity.

Allergic rhinitis, meanwhile, is one of the most common chronic conditions of childhood. It inflames the nasal mucosa, produces congestion and postnasal drip, and has long been associated with poor sleep quality in both children and adults. Previous studies have established that children with allergic rhinitis are more likely to snore and more likely to be diagnosed with sleep-disordered breathing. What remained murky was precisely how the allergy affects the architecture of sleep itself—particularly the delicate machinery of arousals, the brief awakenings that the sleeping brain deploys in response to internal disturbances. The new research, led by Tzu-Yuan Kang and Wan-Yu Lai of China Medical University Hospital in Taichung, set out to illuminate that relationship with the help of polysomnography, the gold-standard technique for measuring sleep.

Polysomnography is an elaborate overnight recording in which electrodes track brain waves, eye movements, muscle tone, heart rhythm, airflow, and blood oxygen. From these signals, sleep technicians can identify every respiratory event, every limb movement, and every arousal, and can compute indices that quantify how frequently each occurs per hour of sleep. The respiratory arousal index, or RAI, counts the awakenings triggered specifically by breathing disturbances, while the spontaneous arousal index captures awakenings with no identifiable trigger, and the periodic limb movement arousal index tallies those caused by involuntary leg jerks. Together, these measures offer a granular portrait of how fragmented a child’s sleep truly is.

The Taiwanese team conducted a retrospective analysis of polysomnography data collected between 2009 and 2020 at a tertiary hospital, focusing on children under twelve years of age who had been evaluated for sleep-disordered breathing. The participants were sorted into four groups according to their polysomnographic findings and clinical diagnoses, distinguishing children with obstructive sleep apnea from those with simple snoring, and accounting for whether allergic rhinitis was present. In total, 643 children were included in the analysis. The researchers then applied multivariable regression models to tease apart the independent contributions of allergy, obesity, age, sex, asthma, and other comorbidities to the various arousal indices.

The results were unambiguous. Children with obstructive sleep apnea who also suffered from allergic rhinitis exhibited significantly more respiratory events and a significantly higher respiratory arousal index than their counterparts without the allergy, with both differences reaching statistical significance at P less than 0.001. In other words, the combination of a blocked, inflamed nose and an already compromised airway produced measurably more fragmented sleep than either condition alone. This finding aligns with the mechanistic intuition that nasal inflammation narrows the upper airway, increases negative pressure during inspiration, and thereby lowers the threshold at which each breathing effort triggers a cortical awakening.

But the study also produced a genuinely surprising result. Using multivariable negative binomial regression—a statistical approach suited to count data such as arousal counts—the team found that allergic rhinitis was independently associated with a lower periodic limb movement arousal rate, with an adjusted incidence rate ratio of 0.393 and a 95 percent confidence interval spanning 0.183 to 0.848. This association held even after the researchers adjusted for comorbid asthma and OSA status. The finding, reported at P equal to 0.017, suggests that allergic rhinitis does not simply amplify every form of sleep disruption; rather, it appears to reshape the pattern of arousals in a selective way, boosting respiratory-driven awakenings while somehow dampening those driven by limb movements. The authors describe this as the first evidence of a significant link between allergic rhinitis and impaired arousal regulation in pediatric sleep-disordered breathing.

The regression analysis also mapped the broader landscape of factors that drive respiratory arousals in children. Older age was associated with a modestly lower respiratory arousal rate, with an adjusted incidence rate ratio of 0.914 at P equal to 0.011, hinting that the airways of older children may be more resilient to collapse. Male sex emerged as a risk factor, with a ratio of 1.643 at P equal to 0.006, consistent with the well-documented male predominance in sleep apnea. Higher body mass index carried a ratio of 1.048 per unit at P equal to 0.018, underscoring the metabolic dimension of the disease. Most dramatically, the presence of obstructive sleep apnea itself carried a ratio of 12.382 at P less than 0.0001, confirming that a formal apnea diagnosis is by far the strongest predictor of respiratory-driven sleep fragmentation. Enuresis, or bedwetting, was also independently associated with higher respiratory arousal rates, with a ratio of 2.065 at P equal to 0.029—a connection that may reflect the way sleep fragmentation disrupts the hormonal regulation of nighttime urine production.

The clinical implications of the study are considerable. The authors argue that their findings underscore the necessity of screening for sleep disturbances in children with allergic rhinitis, a population that pediatricians frequently encounter but whose sleep problems may go unexamined. A child brought in for chronic sneezing and congestion may, in fact, be harboring significant sleep fragmentation that affects cognition, mood, growth, and cardiovascular health. Conversely, the results suggest that managing comorbid allergic rhinitis is a critical component of pediatric sleep disorder therapy. For a child undergoing evaluation or treatment for snoring or obstructive sleep apnea, leaving the underlying allergy untreated may undermine the effectiveness of interventions ranging from nasal surgery to adenotonsillectomy, the most common surgical treatment for pediatric OSA.

The study is not without limitations inherent to its design. As a retrospective, single-center analysis, it relied on children who had already been referred for sleep evaluation, which may limit generalizability to the broader population of allergic children who snore. The authors also note that the manuscript was prepared with assistance from the generative AI tool Gemini to improve grammatical accuracy and readability, with the authors reviewing and editing the content and taking full responsibility for it. The research was supported by the Holistic Health care Foundation of China Medical University Hospital, and the ethics review board of the hospital approved the study with a waiver of informed consent given its retrospective nature. Even so, the work opens a compelling new window onto the physiology of childhood sleep. It suggests that the brain’s arousal system, far from being a passive victim of airway obstruction, is dynamically modulated by allergic inflammation—and that the humble stuffy nose may hold more power over a child’s sleeping brain than anyone had previously measured.

Subject of Research: The association between allergic rhinitis and polysomnographic arousal indices in children with sleep-disordered breathing

Article Title: Association of allergic rhinitis with polysomnographic arousal indices in children with sleep-disordered breathing

Article References: Kang, T.-Y., Wei, C.-C., Hang, L.-W., Li, Y.-H., Yen, H.-R., Wang, J.-Y., Lin, C.-H., & Lai, W.-Y. (2026). Association of allergic rhinitis with polysomnographic arousal indices in children with sleep-disordered breathing. BMC Pediatrics. https://doi.org/10.1186/s12887-026-07684-8

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07684-8

Keywords: allergic rhinitis, sleep-disordered breathing, obstructive sleep apnea, polysomnography, arousal index, pediatrics, snoring, respiratory arousal index, sleep architecture, asthma, enuresis, children's sleep

The commentary, published in October 2026, draws attention to research examining geographic and racial and ethnic patterns of polysomnography use among Medicaid-enrolled children between 2017 and 2019, a body of work led by investigators including C. Christine McLaughlin, Jamie L. Hawke, Emily F. Boss, and their colleagues. Medicaid covers roughly forty percent of American children, making it the single largest payer of pediatric care in the country, and an ideal lens through which to observe how diagnostic practice varies when financial barriers at the point of care are nominally removed. What those data reveal is not a smooth national distribution of sleep testing but stark regional and demographic gradients that persist even among children who share the same insurance coverage.

The clinical stakes are considerable. Obstructive sleep apnea in children affects an estimated one to five percent of the pediatric population, and its consequences extend far beyond disrupted nights. Chronic intermittent hypoxia and fragmented sleep during critical developmental windows are associated with behavioral problems, impaired attention and learning, cardiovascular strain, and metabolic dysfunction. The standard treatment for most children is adenotonsillectomy, the surgical removal of the adenoids and tonsils, which resolves airway obstruction in the majority of cases. Clinical practice guidelines from the American Academy of Pediatrics and the American Academy of Otolaryngology recommend polysomnography before surgery in most circumstances, particularly for children under three, those with obesity, craniofacial anomalies, neuromuscular disease, or other risk factors for perioperative respiratory complications.

The rationale for preoperative testing is grounded in physiology and surgical safety. The apnea-hypopnea index measured during a polysomnogram quantifies how many times per hour a child’s airflow collapses or diminishes, and children with severe obstruction face elevated risks of postoperative respiratory events, including oxygen desaturation, airway obstruction during anesthesia recovery, and the need for unplanned intensive care admission. Studies of children undergoing adenotonsillectomy have repeatedly identified preoperative polysomnographic severity as one of the strongest predictors of these complications, alongside young age, obesity, and cardiac findings. Knowing a child’s baseline severity allows anesthesiologists and surgeons to plan appropriate monitoring, choose the right postoperative setting, and anticipate trouble before it happens in the recovery room.

Yet the commentary makes clear that adherence to this testing paradigm is anything but uniform. Children enrolled in Medicaid, and children from Black, Hispanic, and other minoritized racial and ethnic groups, are consistently less likely to undergo polysomnography before adenotonsillectomy than privately insured white children, and regional variation compounds these disparities. In some parts of the country, preoperative sleep testing is routine; in others, it is the exception. The result is a system in which two children with identical symptoms and identical surgical indications may receive fundamentally different workups depending on their zip code and their family’s background.

The drivers of this variation are multifactorial, and Grigg-Damberger’s commentary situates the new utilization data within a decade of accumulated evidence. Access to pediatric sleep laboratories is geographically concentrated: accredited facilities cluster around academic medical centers and large metropolitan hospitals, leaving families in rural areas and smaller communities facing long travel distances, waiting lists that stretch for months, and the logistical burden of overnight stays far from home. For a parent working an hourly job without paid leave, an overnight sleep study two hundred miles away is not a neutral inconvenience; it can be an insurmountable barrier. Where sleep laboratories are scarce, clinicians may reasonably default to operating on clinical judgment alone, a practice the guidelines permit in uncomplicated cases but which can miss severe disease in children who look deceptively healthy by day.

Racial and ethnic disparities add a second, more troubling layer. Prior studies of tonsillectomy utilization among Medicaid-insured children have documented that Black children are less likely to receive the surgery at all despite comparable or greater disease burden, and surveys of families suggest that caregiver willingness to consider surgery varies with trust in the medical system, the quality of counseling received, and prior experiences of care. Children from minoritized groups also carry a higher baseline prevalence of severe sleep-disordered breathing, driven in part by higher rates of obesity and asthma, which makes under-testing particularly consequential: the children at greatest perioperative risk are precisely those least likely to have that risk quantified before they reach the operating room.

The commentary also engages with a genuine clinical debate: whether routine polysomnography before adenotonsillectomy is always necessary, or whether selective testing is a defensible way to conserve scarce laboratory capacity. Some investigators have argued that for the typical healthy child with clear-cut tonsillar hypertrophy and classic symptoms, surgery without a sleep study is reasonable and that mandatory testing introduces delay and cost without changing outcomes. Others counter that clinical signs correlate poorly with objective severity, that snoring frequency in young children predicts cognitive and behavioral problems better than the apnea-hypopnea index itself, and that skipping the test systematically disadvantages children whose symptoms are dismissed or whose families lack the advocacy needed to push for referral. The disagreement is not merely academic; it determines how scarce testing resources should be allocated and whether the current pattern of underuse in vulnerable populations should be read as clinical pragmatism or as inequity.

What makes the new utilization data so striking is that they emerge from a population in which insurance itself should not be the limiting factor. Medicaid removes the out-of-pocket cost barrier that private deductibles can create, yet disparities persist, which points to structural and systemic explanations: the uneven distribution of pediatric sleep medicine infrastructure, referral patterns that differ by practice setting, differences in how symptoms are perceived and communicated across cultural and linguistic lines, and the cumulative effect of fragmented care for children who move between providers. Grigg-Damberger’s framing, borrowed from the real estate mantra, is deliberate: location, location, location. A child’s address predicts their access to diagnosis in ways that clinical guidelines have not yet overcome.

The path forward, as the commentary implies, requires both measurement and infrastructure. Telemedicine-based sleep assessment, home sleep apnea testing adapted for children, and expanded training for pediatric sleep technologists could extend diagnostic capacity beyond academic centers. Quality measures that track preoperative testing rates by region and by demographic group would make the disparities visible to payers and health systems in a way that individual clinicians cannot easily see. And the underlying research, published in the Journal of Clinical Sleep Medicine, provides the kind of population-level evidence that policy interventions demand. Until then, the overnight sleep study that should precede most pediatric tonsil surgeries will remain, for too many American children, a test determined not by the severity of their disease but by the accident of where they were born.

Subject of Research: Geographic and racial/ethnic disparities in pediatric polysomnography use among US Medicaid-insured children

Article Title: Location, location, location: wide geographic and racial/ethnic variations in pediatric polysomnography in US Medicaid children

Article References: Grigg-Damberger, M. (2026). Location, location, location: wide geographic and racial/ethnic variations in pediatric polysomnography in US Medicaid children. Journal of Clinical Sleep Medicine, 22(1), Article 184. https://doi.org/10.1007/s44470-026-00170-z

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00170-z

Keywords: polysomnography, pediatric sleep apnea, Medicaid, adenotonsillectomy, health disparities, sleep-disordered breathing, tonsillectomy, geographic variation, Journal of Clinical Sleep Medicine, obstructive sleep apnea, pediatric otolaryngology, health equity

COMISA has emerged in recent years as one of the most clinically consequential phenotypes in sleep medicine. Obstructive sleep apnea, in which the upper airway repeatedly collapses during sleep and triggers intermittent drops in blood oxygen, affects an estimated 9 to 38 percent of the general population. Insomnia, characterized by difficulty falling or staying asleep with daytime consequences, is similarly widespread. What has startled researchers is how often the two overlap: roughly 30 to 50 percent of patients with sleep apnea also report chronic insomnia symptoms, and 30 to 40 percent of chronic insomnia patients show objective evidence of sleep-disordered breathing. Crucially, the combination appears to be more than the sum of its parts. Prior studies have linked COMISA to elevated risks of cardiovascular disease and incident diabetes, exceeding the risk conferred by either disorder alone, particularly when the physiological burden is measured objectively rather than by symptoms alone.

To probe the relationship between diabetes and this dual sleep disorder, investigators turned to TURKAPNE, a prospective national registry that has enrolled adults aged 18 to 80 with suspected sleep apnea from 34 centers across Türkiye since 2017. For this cross-sectional analysis, the team included 12,715 patients who underwent full-night, in-laboratory polysomnography, the gold-standard sleep test that monitors brain waves, muscle tone, eye movements, airflow, respiratory effort, and oxygen saturation. Obstructive sleep apnea was defined as an apnea-hypopnea index of at least five events per hour, while insomnia was identified through standardized questions covering difficulty initiating sleep, sleep latency of 30 minutes or longer, trouble maintaining sleep, or use of hypnotic medications. Diabetes status was based on self-reported physician diagnosis or the use of antidiabetic medications. Participants using positive airway pressure therapy or mandibular advancement devices, and those with limited life expectancy, were excluded to keep the baseline measurements clean.

The headline numbers are striking. COMISA was present in 3,275 patients, or 25.7 percent of the cohort. Diabetes was significantly more common among those with COMISA than without it, at 22.8 percent versus 15.4 percent. Flipping the perspective, roughly one in three diabetic individuals in the cohort, 33.9 percent, met criteria for COMISA, compared with 24.1 percent of non-diabetic participants. After adjusting for age, sex, body mass index, smoking status, education level, and comorbidities, diabetes remained independently associated with COMISA, with an adjusted odds ratio of 1.17. While that figure may appear modest, in a cohort of this size and in a condition as prevalent as COMISA, it translates into a substantial population-level burden, and it held firm even after accounting for obesity, the most obvious shared risk factor.

The polysomnographic comparisons within the COMISA group tell an even more compelling story about physiological severity. Diabetic patients slept shorter overall, with a median total sleep time of 353.5 minutes versus 369.0 minutes in non-diabetics, and spent more time awake after sleep onset, 50.2 minutes versus 37.8 minutes. Their arousal index, a measure of how often the brain briefly surfaces from sleep, was 18.1 events per hour compared with 14.8. Their breathing was also markedly worse: the apnea-hypopnea index reached 30.0 events per hour versus 24.0, and the oxygen desaturation index, which counts how often blood oxygen dips each hour, hit 26.0 versus 20.0. Diabetic patients also spent more time with oxygen saturation below 90 percent and reached lower nadir saturations. Sleep architecture shifted toward lighter N1 sleep with reduced rapid eye movement sleep, a pattern associated with impaired glucose metabolism in experimental studies.

Why would diabetes and COMISA feed into each other? The proposed mechanisms run in both directions. Sleep apnea drives metabolic dysfunction through intermittent hypoxia, sympathetic nervous system activation, oxidative stress, and systemic inflammation, all of which impair insulin sensitivity and pancreatic beta-cell function. Insomnia, meanwhile, is thought to act through chronic physiological hyperarousal and dysregulation of the hypothalamic-pituitary-adrenal axis, the body’s central stress system, which elevates cortisol and disrupts glucose homeostasis. When the two disorders coexist, these pathways may amplify one another: the hyperarousal of insomnia can intensify the hypoxic and inflammatory insult delivered by apneic events, deepening insulin resistance. Conversely, diabetes itself may sabotage sleep through nocturia, peripheral neuropathy, autonomic dysfunction, and altered thermoregulation, creating a self-perpetuating cycle in which metabolic derangement and sleep disruption each worsen the other.

One intriguing detail from the data supports this bidirectional reading. Among diabetic patients with COMISA, difficulty maintaining sleep was more prevalent than difficulty initiating sleep, at 50.8 percent versus 44.0 percent for trouble falling asleep. Sleep-maintenance insomnia, waking repeatedly in the middle of the night, is precisely the symptom pattern one would expect from nocturnal hypoglycemia risk, medication effects, and the urge to urinate that accompanies poorly controlled blood sugar. The diabetic COMISA patients were also more frequently female and more often obese, and they carried heavier loads of hypertension, coronary artery disease, obstructive airway disease, hypothyroidism, and psychiatric disorders. Beyond diabetes, the regression analysis identified female sex, psychiatric disease, obstructive airway disease, smoking, lower education level, and higher body mass index as independent correlates of COMISA, sketching a portrait of a phenotype that clusters with social and medical vulnerability.

The sex difference deserves particular attention. Women with sleep apnea are known to present atypically, complaining of insomnia, fatigue, and mood disturbance rather than the classic snoring and witnessed apneas that raise suspicion in men, a pattern that contributes to systematic underdiagnosis of sleep-disordered breathing in women. Hormonal changes around menopause, including declines in progesterone and estrogen, may simultaneously increase upper airway collapsibility and insomnia prevalence. Because diabetic patients with COMISA in this cohort were more often female, clinicians evaluating women with diabetes who report poor sleep should consider that both conditions may be present and that neither should be dismissed as mere stress or menopausal discomfort.

The clinical implications are straightforward, though implementing them will require a shift in routine practice. The authors argue that individuals with diabetes should be screened systematically for both insomnia symptoms and sleep-disordered breathing, rather than waiting for one or the other to declare itself. Early identification of COMISA opens the door to integrated treatment, pairing positive airway pressure therapy for the apnea with cognitive behavioral therapy for insomnia, the first-line behavioral treatment that avoids the pitfalls of long-term hypnotic medication. Whether treating COMISA aggressively improves glycemic control and reduces cardiometabolic risk in diabetic patients remains to be tested in intervention trials, but the severity gradient documented here makes the hypothesis plausible and urgent.

The study has limitations worth keeping in view. Diabetes and comorbidities were identified by self-report and medication use rather than laboratory confirmation, insomnia was assessed with symptom questions rather than full diagnostic criteria, and details on diabetes duration, glycemic control, and specific drug classes were unavailable. Because the design is cross-sectional, causality cannot be established; the data show association, not direction. The cohort was also exclusively Turkish, which may limit generalizability. Still, with 12,715 polysomnographically characterized patients, the statistical signal is hard to dismiss. If roughly a third of people with diabetes carry a combined sleep disorder that worsens both their nights and their metabolism, then the sleep clinic and the diabetes clinic may need to stop operating as separate worlds, and the humble question about sleep quality may become one of the cheapest diagnostic tools in metabolic medicine.

Subject of Research: The association between diabetes mellitus and comorbid insomnia and obstructive sleep apnea (COMISA)

Article Title: Diabetes mellitus is associated with increased prevalence and severity of COMISA: evidence from the nationwide TURKAPNE cohort

Article References: Gürkan, C. G., Pihtili, A., Kiyan, E., Tasbakan, M. S., Basoglu, Ö. K., Arbatli, S., Aydin, S., Cilli, A., Dursunoglu, N., Baran, B., Peker, Y., Turkapne Study Group, Ursavaş, A., Annakkaya, A. N., Aksakal, A., Balcı, A., Utkusavaş, A., Ugurlu, A. Ö., Elmalı, A. D., … Zeren Uçar, Z. (2026). Diabetes mellitus is associated with increased prevalence and severity of COMISA: evidence from the nationwide TURKAPNE cohort. Journal of Clinical Sleep Medicine, 22(1), Article 103. https://doi.org/10.1007/s44470-026-00109-4

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00109-4

Keywords: diabetes mellitus, COMISA, insomnia, obstructive sleep apnea, polysomnography, sleep fragmentation, TURKAPNE cohort, insulin resistance, sleep-disordered breathing, metabolic dysfunction, intermittent hypoxia, sleep medicine

The original study that prompted the correspondence, conducted by Hudson and colleagues and published in the same journal, examined the microbial communities of children diagnosed with obstructive sleep apnea and children with primary snoring, comparing them against the microbial signatures found in their throats and intestines. The headline finding was twofold. First, the oropharynx, the region at the back of the mouth where the soft palate, tonsils, and tongue base conspire to block airflow during apneic events, showed dysbiosis, a disturbance of the normally balanced microbial ecosystem. Second, the gastrointestinal tract displayed a milder but still detectable compositional imbalance, suggesting that the microbial consequences of disordered breathing during sleep may not be confined to the airway alone.

That dual-site finding is what makes the study scientifically provocative. The oropharynx is the obvious suspect in sleep-disordered breathing, since it is the anatomical battleground where airway collapse occurs. But the gut connection is less intuitive and more tantalizing. Microbes swallowed in saliva continuously seed the gut, and inflammatory signals generated in inflamed airway tissue can ripple systemically through the bloodstream. A growing body of research has shown that oral taxa can enrich the lung microbiome in ways associated with a Th17-skewed inflammatory phenotype, as Segal and colleagues reported in Nature Microbiology, demonstrating that microbes originating in the mouth can travel and reshape immune behavior at distant mucosal sites. If snoring children carry a subtly different gut microbial portfolio, the question becomes whether that difference is a cause, a consequence, or merely a microbial echo of the sleep disorder itself.

Min’s correspondence engages with this question from the vantage point of an otolaryngologist who has previously published on the adenotonsillar microbiome. In earlier work with Kim, published in Clinical and Experimental Otorhinolaryngology, Min examined correlations between the microbiome of the adenoids and tonsils and the clinical characteristics of pediatric patients who snore. That line of investigation matters because the adenoids and tonsils are not passive bystanders in childhood sleep apnea; they are the most common anatomical cause of airway obstruction in snoring children and the usual target of adenotonsillectomy, the standard surgical treatment. If the immune tissues of the throat harbor a distinctive microbial community in children who snore, the microbiome could plausibly influence the inflammatory state of those tissues, their tendency to enlarge, and their contribution to airway collapse.

The correspondence also arrives amid a broader scientific reckoning with how environmental factors sculpt the human microbiome. David and colleagues demonstrated in Nature in 2014 that diet rapidly and reproducibly alters the human gut microbiome, shifting the relative abundance of microbial species within days of a dietary change. Shaw and colleagues showed in mBio in 2017 that the human salivary microbiome is shaped more by shared environment than by genetics, drawing on evidence from a large family of closely related individuals. These findings cut both ways for sleep apnea research. On one hand, they suggest that household factors such as shared meals, shared bedrooms, and shared microbial exposures could confound apparent associations between sleep disorders and microbial profiles. On the other hand, they reinforce the plausibility of a genuine link, because the same environmental variables that shape a child’s microbiome, including diet, allergen exposure, and household infections, also influence airway inflammation and adenotonsillar tissue growth.

Nasal and upper airway microbiology adds further context. Yan and colleagues, writing in Cell Host and Microbe, showed that nasal microenvironments and interspecific interactions influence the complexity of nasal microbiota and the carriage of Staphylococcus aureus, underscoring how finely structured these communities are even within the small geography of the nose. Choi and colleagues reported in Allergy that patients with chronic rhinosinusitis show decreased diversity of nasal microbiota and of the extracellular vesicles those microbes secrete, linking reduced microbial diversity to chronic inflammatory disease of the airways. Reduced diversity is a recurring theme in dysbiosis research across body sites, and the oropharyngeal findings in snoring children fit that pattern, raising the possibility that a less diverse throat community might both reflect and reinforce the chronic inflammation that accompanies repeated nighttime oxygen desaturation and arousals.

The mechanistic stakes are considerable. Obstructive sleep apnea in children is not a benign habit but a condition associated with daytime inattention, behavioral problems, impaired growth, and cardiovascular strain. Current diagnosis relies on overnight polysomnography, a resource-intensive test, and treatment hinges on surgery or, in some cases, continuous positive airway pressure. If microbial markers in the throat or gut could help identify children at risk, or if microbial dysbiosis turned out to contribute to airway inflammation and tissue hypertrophy, the field would gain both a potential screening tool and a novel therapeutic target. Probiotic or dietary interventions that reshape the gut and oral microbiomes are already being explored for other inflammatory conditions, and the correspondence implicitly asks whether such approaches deserve systematic study in pediatric sleep medicine.

Yet the correspondence also serves as a reminder of how much caution the microbiome field demands. Associations between microbial composition and disease are notoriously vulnerable to confounding by antibiotics, diet, age, geography, and sampling technique. The salivary microbiome’s sensitivity to shared environment means that siblings in the same household can carry strikingly similar oral communities regardless of their genetics, so any study comparing snoring children with non-snoring controls must carefully account for household clustering. The mildness of the gastrointestinal imbalance reported in the original study, described explicitly as a mild compositional imbalance, suggests an effect that is real but subtle, the kind of signal that requires replication in independent cohorts before it can be translated into clinical practice. Min’s letter, by engaging critically with the findings rather than simply endorsing them, exemplifies the kind of scholarly scrutiny that determines which microbiome associations survive replication.

What emerges from this exchange is a picture of pediatric sleep medicine slowly absorbing the lessons of the microbiome era. The airway is no longer viewed as a sterile tube that occasionally gets blocked by oversized tonsils; it is a living ecological interface where bacteria, immune cells, airflow, and inflammation interact continuously. The gut is no longer seen as anatomically remote from the throat; it is connected by the constant downward flow of swallowed saliva and by circulating immune mediators. The correspondence by Min, published as volume 22, article 104 of the Journal of Clinical Sleep Medicine, does not resolve whether microbes help drive childhood sleep apnea, but it sharpens the questions that the next generation of studies must answer: which specific taxa change, whether the changes precede the disorder or follow it, whether they track disease severity, and whether correcting them could help a snoring child breathe, sleep, and grow more easily. For now, the microbial trail from throat to gut in snoring children remains an open and actively contested frontier.

Subject of Research: The association between pediatric obstructive sleep apnea, primary snoring, and oropharyngeal and gastrointestinal microbiome dysbiosis

Article Title: Correspondence regarding “Obstructive sleep apnea and primary snoring in children are associated with oropharyngeal dysbiosis and a mild compositional imbalance in the gastrointestinal tract”

Article References: Min, H. J. (2026). Correspondence regarding “Obstructive sleep apnea and primary snoring in children are associated with oropharyngeal dysbiosis and a mild compositional imbalance in the gastrointestinal tract”. Journal of Clinical Sleep Medicine, 22(1), Article 104. https://doi.org/10.1007/s44470-026-00112-9

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00112-9

Keywords: obstructive sleep apnea, primary snoring, children, microbiome, dysbiosis, oropharynx, gut microbiota, adenotonsillar microbiome, sleep-disordered breathing, pediatric sleep medicine, inflammation, Journal of Clinical Sleep Medicine

The underlying research drew on the control arm of PATS, a multicenter randomized controlled trial conducted from 2016 to 2021 at seven tertiary referral centers across the United States. The trial enrolled children aged 3.0 to 12.9 years who snored at least three nights per week and had an obstructive apnea-hypopnea index, or oAHI, below 3 events per hour, along with enlarged tonsils of Brodsky grade II or greater. Children were randomly assigned to early adenotonsillectomy or to watchful waiting with supportive care. In this secondary analysis, the researchers focused on the 234 children who did not undergo surgery, a group with a mean age of 6.2 years that included 111 girls, 65 children of Black or African American race, and 37 children of Hispanic ethnicity. The analysis was exploratory rather than prespecified, and the investigators followed observational study reporting guidelines while leveraging the rigor of the original trial’s standardized protocols.

The technical core of the study rested on two distinct definitions of progression. The first was polysomnographic: a child was considered to have progressed if the oAHI on a repeat overnight sleep study reached 3 or more events per hour at 12 months. All sleep studies were scored centrally by certified technologists blinded to each child’s clinical presentation, according to American Academy of Sleep Medicine pediatric rules. The second definition was symptomatic: a caregiver-completed Pediatric Sleep Questionnaire-Sleep-Related Breathing Disorder scale score of 0.33 or higher on the 0-to-1 range, a validated threshold indicating clinically significant symptom impact. The researchers also collected a battery of baseline measures, including the modified Epworth Sleepiness Scale for children, the OSA-18 quality-of-life instrument, behavioral inventories such as the Behavior Rating Inventory of Executive Function and the Child Behavior Checklist, physical examination findings including tonsil grade and palate position, and histories of asthma, allergies, ADHD, and environmental tobacco smoke exposure confirmed where possible by urinary cotinine assays.

The headline result on the objective side was reassuring. Of the 150 children who completed 12-month polysomnography, only 20, or 13 percent, progressed to an oAHI of 3 or greater. The mean baseline oAHI in the cohort was just 0.8 events per hour, and most children’s airway physiology remained stable or improved over the observation year. This finding aligns with earlier evidence that roughly 40 percent of children with mild-to-moderate obstructive sleep apnea show spontaneous improvement within a year without treatment, and that most children with primary snoring never develop frank apnea. For families worried that a year of observation might allow a silent slide into severe disease, the data suggest that such deterioration is the exception rather than the rule, at least over a 12-month horizon in this prepubertal, predominantly non-obese sample.

The symptomatic picture was far less comforting. Among the 192 children with follow-up questionnaire data, 110, or 57 percent, had a high symptom burden at 12 months, and the overwhelming majority of those cases, 94 percent, represented persistence of symptoms that were already elevated at baseline rather than new onset. Nearly three quarters of children whose baseline questionnaire score crossed the clinical threshold still scored above it a year later. Even more telling, among the children whose symptoms persisted or worsened and who also completed repeat sleep studies, only 18 percent had progressed objectively to an oAHI of 3 or greater. In other words, four out of five children whose breathing events remained technically mild still felt the consequences of disrupted sleep, a mismatch that echoes prior work showing that parent-reported symptom scales often mediate the behavioral benefits of adenotonsillectomy better than polysomnographic indices do.

Which children were most likely to get worse? On the objective measure, Black race more than doubled the odds of polysomnographic progression, with an unadjusted odds ratio of 2.71, and a higher baseline symptom score raised the odds by 74 percent per standard deviation. The racial disparity is consistent with findings from the earlier Childhood Adenotonsillectomy Trial, or CHAT, and with a broader literature showing that Black children carry higher rates of sleep apnea, greater severity, and more persistent disease after surgery. The authors point to multifactorial explanations, including chronic exposure to respiratory irritants and allergens common in disadvantaged neighborhoods; notably, Black children in PATS were more likely to live in such neighborhoods. On the symptomatic side, asthma nearly tripled the odds of persistence or progression, ADHD more than tripled them, and tobacco smoke exposure raised them by nearly two and a half times, while worse sleepiness, poorer disease-specific quality of life, and greater behavioral impairment at baseline all pointed in the same direction.

One result defied expectations. Children with larger tonsils, grade III or IV on the Brodsky scale, had lower odds of symptomatic persistence or progression, with an odds ratio of 0.42. Among children with big tonsils, the proportion of caregivers reporting a high symptom inventory dropped from 69 percent at baseline to 49 percent at follow-up, while among children with smaller grade II tonsils the figure barely moved, from 75 percent to 70 percent. The authors propose several mechanisms: a possible ceiling effect in which children with large tonsils are already at maximal obstruction and have room only to improve as tonsils shrink or airways grow, versus children with smaller tonsils whose obstruction may arise from adenoid hypertrophy, diffuse airway inflammation, low pharyngeal muscle tone, or other non-tonsillar sites that a simple oral examination cannot capture. PATS did not assess adenoid size, and the finding is consistent with a long-standing literature showing that tonsil size correlates poorly with objective apnea severity.

The asthma connection deserves particular attention. Asthma and sleep-disordered breathing are known to reinforce each other bidirectionally: children with asthma are roughly twice as likely to have sleep-disordered breathing, asthma severity tracks with sleep apnea severity, and disrupted sleep worsens asthma control. Within PATS, both study arms received referral for asthma management as part of supportive care, yet at follow-up only 39 percent of children with asthma in the surgery group had high symptom scores compared with 74 percent in the observation group. Although a separate analysis of surgery’s effect on asthma symptoms did not reach statistical significance, the pattern suggests that optimizing asthma treatment alone did little to relieve sleep-related symptoms over a year of watchful waiting, lending support to earlier surgery for children whose mild snoring coexists with asthma.

How useful are these variables for prediction in practice? The honest answer is: modestly. The best multivariable model for polysomnographic progression, combining Black race, REM-sleep apnea index, and the baseline symptom score, achieved an area under the curve of only 0.70, barely better than the 0.62 achieved with clinical variables alone. For symptomatic persistence, the model pairing the baseline questionnaire score with tonsil grade performed better, reaching an AUC of 0.85, but the authors caution that the exploratory analysis was not powered to build a definitive risk calculator, and confidence intervals around several predictors were wide. The study also excluded children with severe obesity, recurrent tonsillitis, and major chronic illness, so the findings apply most directly to the prepubertal, otherwise healthy snorers who dominate typical clinic populations.

The clinical message that emerges is subtle but actionable. A single repeat sleep study a year later captures only the small minority of children whose breathing physiology objectively worsens, and it may miss the larger group whose symptoms, behavior, and quality of life remain impaired even as their apnea index stays low. For the symptomatic snoring child, the authors suggest, repeated symptom assessment and attention to behavioral and quality-of-life impacts may guide decisions better than polysomnography alone. Children who are Black, or who have asthma, ADHD, tobacco smoke exposure, or a heavy baseline symptom burden, may be stronger candidates for earlier adenotonsillectomy than their peers. As the largest controlled dataset yet assembled on this question, the analysis reframes watchful waiting not as a passive default but as a monitored strategy with identifiable risk profiles, and it sets the stage for longer, larger studies to determine which children truly benefit from waiting and which would be better served by operating sooner.

Subject of Research: Progression and symptom persistence of mild sleep-disordered breathing in children managed with watchful waiting instead of adenotonsillectomy

Article Title: Progression of mild sleep-disordered breathing in children managed with watchful waiting

Article References: Kirkham, E. M., Ishman, S., Garetz, S., Baldassari, C. M., Mitchell, R. B., Liu, C., Tapia, I. E., Elden, L. M., Hassan, F., Ibrahim, S., Ross, K., Cen, M., Wang, R., Redline, S., & Chervin, R. D. (2026). Progression of mild sleep-disordered breathing in children managed with watchful waiting. Journal of Clinical Sleep Medicine, 22(1), Article 106. https://doi.org/10.1007/s44470-026-00082-y

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00082-y

Keywords: pediatric sleep apnea, sleep-disordered breathing, adenotonsillectomy, watchful waiting, polysomnography, snoring, PATS trial, asthma, ADHD, health disparities, tonsil size, Journal of Clinical Sleep Medicine

The research, led by Vanessa Campes Dannenberg and Maria L. Castro-Codesal of the University of Alberta, together with colleagues at the Hospital for Sick Children in Toronto and the Alberta Children’s Hospital in Calgary, examined the medical records of 115 children and young people with neuromuscular disorders who were recommended for home non-invasive ventilation, or NIV, between 2010 and 2023. Most of the participants were male, with a median age of 11.8 years at the time of ventilation initiation. The most common diagnoses were Duchenne muscular dystrophy, accounting for 39 percent of the cohort, congenital muscular dystrophies at 19 percent, and spinal muscular atrophy at 14 percent. Nearly every child in the study carried at least one additional chronic condition, with a median of four comorbidities per patient, underscoring how medically complex this population is.

The central question was deceptively simple: when three different sets of polysomnography-based criteria are applied to the same overnight sleep study, do they identify the same children as needing breathing support? The team compared the 2023 guidelines from the American College of Chest Physicians, known as CHEST, against the traditional American Academy of Sleep Medicine, or AASM, definitions of moderate-to-severe obstructive sleep apnea and nocturnal hypoventilation, and against a new set of REM-related criteria proposed by the authors themselves. The CHEST criteria, drawn from a European Respiratory Society statement on pediatric long-term ventilation, flag children whose end-tidal or transcutaneous carbon dioxide exceeds 50 mmHg for at least 2 percent of total sleep time, or rises by 10 mmHg above their waking baseline, or whose oxygen saturation drops below 90 percent for at least 2 percent of sleep time or five continuous minutes.

Those thresholds matter because they are far more sensitive than the older AASM definitions, which require carbon dioxide to exceed 50 mmHg for a full 25 percent of total sleep time before labeling a child as hypoventilating. The results were striking. Only 9 percent of the children met the AASM-based hypoventilation criterion, while the CHEST guidelines identified 25 percent, and the REM-related criteria caught 12 percent. For nocturnal hypoxemia, for which the AASM framework offers no initiation criterion at all, the CHEST guidelines identified 22 percent of the cohort and the REM-related criteria 15 percent. In total, 71 percent of the children met at least one NIV initiation criterion across the three criteria sets, but a meaningful minority were visible only through the newer lenses.

Agreement between the criteria sets, measured with kappa statistics, told a nuanced story. Overall agreement between the CHEST and REM-related criteria was moderate, with a kappa of 0.71, and agreement on nocturnal hypoxemia between those two sets was similarly moderate at 0.69. The strongest agreement, a kappa of 0.86, emerged between the CHEST and AASM criteria on the apnea-hypopnea index, which is unsurprising given that both frameworks use similar thresholds of five or more respiratory events per hour. The divergence, in other words, was not about counting apneas and hypopneas. It was about the slow, silent buildup of carbon dioxide and the dips in oxygen that occur without any discrete respiratory event, particularly during the muscle atonia of REM sleep.

That physiology is the heart of the study’s argument. In children with neuromuscular weakness, breathing during REM sleep depends heavily on the diaphragm, because the intercostal and upper airway muscles lose their tone in that stage. When the diaphragm is also weak, carbon dioxide accumulates and oxygen levels fall, often without producing a single scored apnea. As the disease progresses, these abnormalities spread into non-REM sleep and eventually into wakefulness, culminating in chronic respiratory failure. The authors’ proposed REM-related criteria were designed to capture this earliest phase, and the analysis showed that an additional 11 percent of children met REM-related indications that the AASM-based criteria missed entirely. Thirteen percent of the cohort, twelve children, met no AASM-based criteria at all but satisfied at least one CHEST and one REM-related criterion.

The study also looked at what happened after ventilation began. Ninety-six percent of the children, 111 in total, successfully started NIV, and 73 percent of them sustained the therapy for six to twelve months, the window the researchers chose to avoid the acclimatization period when adherence is typically unstable. Adherence was respectable: children used the machine for a median of six hours per night, and on a median of 60 percent of nights they managed more than four hours of use. Clinician-documented reports from routine visits described improvements in general health in 34 percent of children, better sleep quality in 36 percent, and improved breathing during sleep in 25 percent. Most children avoided emergency department visits, ward admissions, and intensive care during the follow-up period.

Notably, the study found no significant differences in lung function before and after NIV initiation in any of the criteria groups, and no differences in adherence or clinical outcomes between groups. The authors attribute this largely to the substantial overlap among the three groups, since most children met the apnea-hypopnea thresholds common to all criteria sets. They also caution that NIV does not reverse the decline in lung function driven by the underlying disease; at best, it slows the trajectory, and changes are influenced by disease-modifying therapies and the natural history of each disorder. Only about half of the cohort could perform lung function testing that met American Thoracic Society technical standards, further limiting those comparisons.

The findings carry real weight for clinical practice. Earlier work in children with Duchenne muscular dystrophy had already suggested that disease-specific criteria could prompt ventilation for an additional 22 percent of children compared with standard AASM definitions. This study extends that insight to a broader neuromuscular population and adds a physiological rationale: many children show REM-related hypoventilation or hypoxemia with a normal apnea-hypopnea index, and the CHEST guidelines, by lowering the carbon dioxide threshold and adding an explicit hypoxemia criterion, are built to detect exactly that pattern. This matters especially for conditions such as myotonic dystrophy, in which discrete respiratory events are less common but carbon dioxide retention can still be dangerous.

The researchers acknowledge the limits of their retrospective design, their reliance on chart documentation, and the overlap that prevented subgroup comparisons. They also note that future studies may be difficult to conduct, because withholding ventilation until advanced hypoventilation develops would now be ethically questionable. Still, their conclusion is clear and forward-looking: guidelines for children with neuromuscular disease should integrate disease-specific and sleep-stage-specific polysomnographic criteria, watching REM sleep closely for the first signs of failure. Caught early, supported by a mask and a machine at home, these children may avoid the steepest part of the respiratory decline that has long defined their prognosis.

Subject of Research: Polysomnographic criteria for initiating non-invasive ventilation in children with neuromuscular diseases

Article Title: Establishing polysomnographic criteria for initiation of non-invasive ventilation in children with neuromuscular diseases

Article References: Dannenberg, V. C., Amin, R., Anagnostopoulos, G., Adeleye, A., Bendiak, G. N., Kolski, H., & Castro-Codesal, M. L. (2026). Establishing polysomnographic criteria for initiation of non-invasive ventilation in children with neuromuscular diseases. Journal of Clinical Sleep Medicine, 22(1), Article 117. https://doi.org/10.1007/s44470-026-00102-x

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00102-x

Keywords: neuromuscular disease, non-invasive ventilation, polysomnography, REM sleep, sleep-disordered breathing, nocturnal hypoventilation, Duchenne muscular dystrophy, spinal muscular atrophy, pediatric sleep medicine, CHEST guidelines, AASM criteria, nocturnal hypoxemia