Paediatric sleep apnea event prediction using nasal air pressure and machine learning.

Sleep-disordered breathing is an important health issue for children. The objective of this study was to develop a machine learning classifier model for the identification of sleep apnea events taken exclusively from nasal air pressure measurements acquired during overnight polysomnography for paediatric patients. A secondary objective of this study was to differentiate site of obstruction exclusively from hypopnea event data using the model. Computer vision classifiers were developed via transfer learning to either normal breathing while asleep, obstructive hypopnea, obstructive apnea or central apnea. A separate model was trained to identify site of obstruction as either adeno-tonsillar or tongue base. In addition, a survey of board-certified and board-eligible sleep physicians was completed to compare clinician versus model classification performance of sleep events, and indicated very good performance of our model relative to human raters. The nasal air pressure sample database available for modelling comprised 417 normal, 266 obstructive hypopnea, 122 obstructive apnea and 131 central apnea events derived from 28 paediatric patients. The four-way classifier achieved a mean prediction accuracy of 70.0% (95% confidence interval [67.1-72.9]). Clinician raters correctly identified sleep events from nasal air pressure tracings 53.8% of the time, whereas the local model was 77.5% accurate. The site of obstruction classifier achieved a mean prediction accuracy of 75.0% (95% confidence interval [68.7-81.3]). Machine learning applied to nasal air pressure tracings is feasible and may exceed the diagnostic performance of expert clinicians. Nasal air pressure tracings of obstructive hypopneas may "encode" information regarding the site of obstruction, which may only be discernable by machine learning.

Hypoglossal Nerve Stimulation in Adolescents With Down Syndrome and Obstructive Sleep Apnea.

IMPORTANCE: Obstructive sleep apnea (OSA) affects up to 60% of children with Down syndrome (DS) and may persist in half of patients after adenotonsillectomy. Children with DS who have persistent OSA often do not tolerate treatment with positive pressure airway support devices or tracheotomy for their residual moderate to severe OSA. The hypoglossal nerve stimulator is an implantable device that delivers an electrical impulse to anterior branches of the hypoglossal nerve in response to respiratory variation, resulting in tongue base protrusion that alleviates upper airway obstruction in adults. OBJECTIVE: To determine whether hypoglossal nerve stimulation is safe and effective in children with DS. DESIGN, SETTING, AND PARTICIPANTS: Case series of the first 6 adolescents with DS to undergo hypoglossal nerve stimulator implantation. Participants were 6 children and adolescents (12-18 years) with DS and severe OSA (apnea hypopnea index [AHI] > 10 events/h) despite prior adenotonsillectomy. INTERVENTION: Inspire hypoglossal nerve stimulator placement. MAIN OUTCOMES AND MEASURES: Patients were monitored for adverse events. Adherence to therapy was measured by hours of use recorded by the device. Efficacy was evaluated by comparing AHI and OSA-18, a validated quality-of-life instrument, scores at baseline and follow-up. RESULTS: In 6 patients (4 male, 2 female; aged 12-18 years), hypoglossal nerve stimulator therapy was well tolerated (mean use, 5.6-10.0 h/night) and effective, resulting in significant improvement in OSA. At 6- to 12-month follow-up, patients demonstrated a 56% to 85% reduction in AHI, with an overall AHI of less than 5 events/h in 4 children and less than 10 events/h in 2 children. Children also demonstrated a clinically significant improvement (mean [SD] overall change score, 1.5 [0.6]; range, 0.9-2.3) on the OSA-18, a validated quality-of-life instrument. CONCLUSIONS AND RELEVANCE: Hypoglossal nerve stimulation was well tolerated and effective in the study population, representing a potential therapeutic option for patients with DS and refractory OSA after adenotonsillectomy who are unable to tolerate positive pressure airway devices. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT2344108.

Hypoglossal Nerve Stimulator Implantation in an Adolescent With Down Syndrome and Sleep Apnea.

Obstructive sleep apnea (OSA) is more common in children with Down syndrome, affecting up to 60% of patients, and may persist in up to 50% of patients after adenotonsillectomy. These children with persistent moderate to severe OSA require continuous positive airway pressure, which is often poorly tolerated, or even tracheotomy for severe cases. The hypoglossal nerve stimulator is an implantable device that produces an electrical impulse to the anterior branches of the hypoglossal nerve, resulting in tongue protrusion in response to respiratory variation. It is an effective treatment of sleep apnea in select adult patients because it allows for alleviation of tongue base collapse, improving airway obstruction. Herein we describe the first pediatric hypoglossal nerve stimulator implantation, which was performed in an adolescent with Down syndrome and refractory severe OSA (apnea hypopnea index [AHI]: 48.5 events/hour). The patient would not tolerate continuous positive airway pressure and required a long-standing tracheotomy. Hypoglossal nerve stimulator therapy was well tolerated and effective, resulting in significant improvement in the patient's OSA (overall AHI: 3.4 events/hour; AHI: 2.5-9.7 events/hour at optimal voltage settings depending on sleep stage and body position). Five months after implantation, the patient's tracheotomy was successfully removed and he continues to do well with nightly therapy.