Role of upper airway evaluation in the multidisciplinary management of obstructive sleep apnea in children below two years of age

BACKGROUND

Diagnosis and treatment of obstructive sleep apnea (OSA) in infants and young children is challenging because of its clinical heterogeneity and lack of age-specific guidelines.

AIM

We report the management and treatment outcome of OSA in children below 2 years of age. Treatment decisions were based upon the pattern of upper airway (UA) obstruction, clinical presentation and OSA severity.

METHODS

Retrospective, non-randomized observational cohort study at a tertiary center. Children with OSA who underwent an UA evaluation (drug-induced sleep endoscopy or direct laryngoscopy) were included.

RESULTS

We studied 100 patients, 57 boys and 43 girls, age 0.72 years (0.0-2.0) and OSA confirmed by polysomnography. Multilevel UA collapse was present in 26%, (adeno)tonsillar hypertrophy in 31% and 21% had laryngomalacia. Laryngomalacia was more common in children below 6 months of age and adenotonsillar hypertrophy was observed mainly in children >1.5 year of age. Surgical and nonsurgical treatment guided by UA findings, improved OSA severity at group level with a significant reduction (p < 0.001) in obstructive apnea/hypopnea index from 10.8/h (2.1-99.1) to 1.7/h (0.0-73.0), an improvement in mean oxygen saturation from 96.9% (88.9-98.4) to 97.4% (92.3-99.0), in minimal oxygen saturation from 85.4% (37.0-96.0) to 88.8% (51.0-95.5) and oxygen desaturation index from 5.1/h (0.2-52.0) to 1.3/h (0.0-47.8).

CONCLUSION

Multidisciplinary management of young children with OSA guided by the pattern of UA obstruction and OSA severity, reduces OSA severity. The pattern of UA obstruction changes in the first 2 years of life from a dynamic collapse to structural abnormalities.

A Comparison of Drug-Induced Sleep Endoscopy Findings in Infants and Toddlers

OBJECTIVE

To compare patterns of obstruction in infants and toddlers with obstructive sleep apnea (OSA) by analyzing drug-induced sleep endoscopy (DISE) examinations.

STUDY DESIGN

Retrospective chart review.

SETTING

An academic tertiary care pediatric hospital at a single institution.

METHODS

Children aged ≤3 years who underwent DISE with propofol titration from 2020 to 2022 were included in the study. Data extracted included age, body mass index, comorbidities, and polysomnography parameters. The sites of anatomic obstruction during DISE were recorded using a validated 10-site scoring system.

RESULTS

A total of 86 children underwent DISE, including 15 infants (<1 year) and 71 toddlers (>1≤3 years). Sixty-two percent were male and 37% were female. The mean age of the infant group was 9.6 months (95% confidence interval [CI]: 7.92, 11.28 months), and the mean age of the toddler group was 2.3 years (95% CI: 2.06, 2.50 years). The mean total number of sites of obstruction was not significantly different between infants (N = 2.5) and toddlers (N = 3.2, P = .086). The most common subsite of significant obstruction in both infants and toddlers was the aryepiglottic folds (46.7% and 36.8%, respectively; P = .578). The sites of obstruction and frequency of their occurrence were not significantly different between the 2 groups.

CONCLUSION

Multisite airway obstruction is common among children with OSA, especially at the epiglottis, aryepiglottic folds, and tongue base. There is no difference in sites of obstruction and their frequency between toddlers and infants.

The Prevalence of Sentinel Central Events (SCent) During Drug-Induced Sleep Endoscopy

OBJECTIVE

With the recent addition of airflow and respiratory effort channels, our group has observed central and mixed apnea events during drug-induced sleep endoscopy (DISE). We measured the frequency and timing of sentinel central and/or mixed events (SCents), as well as assessed for differences in velum, oropharynx, tongue, and epiglottis (VOTE) classification compared to obstructive events.

STUDY DESIGN

Prospective single-cohort study.

SETTING

Tertiary Care Academic Medical Center.

METHODS

Patients underwent DISE between June 2020 and November 2022. Nasal airflow, thoracoabdominal effort belt signals, and videoendoscopy were simultaneously captured. Demographics, sleep study, and DISE data were compared among patients with and without SCents using Student's T tests or χ2 tests.

RESULTS

On average, the cohort (n = 103) was middle-aged (53.5 ± 12.1 years), overweight (body mass index of 29.7 ± 5.3 kg/m2), and had severe obstructive sleep apnea (apnea-hypopnea index of 30.7 ± 18.7 events/h). Forty-seven patients (46%) were found to have at least 1 SCent. Among those with SCent, 45 (95.7%) transitioned to obstructive pathology after an average of 7.91 ± 2.74 minutes, with at least 95% of patients expected to do so within 12.57 minutes. Twenty-nine out of 47 patients (61.2% [95% confidence interval: 46.4.9%, 75.5%]) with SCent had meaningful differences between central/mixed and obstructive VOTE scores.

CONCLUSION

Central events were present in almost half of our cohort. At least 95% of patients were expected to transition to obstructive events within 12 to 13 minutes of propofol initiation. In addition, over half of patients demonstrate significantly different VOTE scores between central and obstructive events. These factors should raise awareness of central events and scoring passive apneas during DISE and consider delaying VOTE scoring.

The Whimsical Nature of Airway Obstruction during Drug-Induced Sleep Endoscopy

A 47-year-old male patient diagnosed with severe obstructive sleep apnea (OSA) sought alternatives to positive airway pressure, prompting evaluation with drug-induced sleep endoscopy (DISE). He underwent a specialized DISE with nasal airflow and pharyngeal pressure monitoring. During obstructive apneas, airflow and pressure signals demonstrated dynamic, multilevel upper airway collapse, with shifting sites of airflow obstruction as respiratory effort increased. This case report illustrates how quantitative airflow and pressure measurements can complement the standard DISE exam and aid in surgical decision-making. Laryngoscope, 2024.

The Repeatability of Pharyngeal Opening Pressure Under Drug-Induced Sleep Endoscopy

OBJECTIVE

Pharyngeal opening pressure (PhOP) is a measure of upper airway collapsibility that can be obtained during drug-induced sleep endoscopy (DISE) using a continuous positive airway pressure (CPAP) titration. However, the stability of PhOP over the course of sedation during DISE remains unclear. This study aims to compare repeat measures of PhOP over the course of DISE.

STUDY DESIGN

Single arm prospective study.

SETTING

Single tertiary care institution.

METHODS

Patients had 2 CPAP titrations while undergoing DISE. Collected data included patient demographics, PhOP, patient sedation index (PSI), and duration of and between CPAP titrations. t Tests, test-retest coefficient analysis, and repeated measures correlation were performed.

RESULTS

Twenty-five patients completed the study between 2022 and 2023 with 22 patients having sedation depth (PSI) recording. Most were male (76%), obese (average body mass index: 30.24 kg/m2), with severe obstructive sleep apnea (average apnea-hypopnea index: 39.8 events/hr). Test-retest analysis showed good-excellent correlation between PhOP values (intraclass correlation coefficient = 0.892, P < .0001, n = 25). Average time between CPAP titrations was 15 minutes to 6 seconds. Over that time, PhOP increased by an average of 0.72 cmH2O (P = .06, n = 25) and PSI decreased by 9.5 units (P = .01, n = 22). Repeated measures correlation showed a weak negative correlation between PhOP and PSI (r = -.45, P = .03, n = 22).

CONCLUSION

The results showed repeatability of PhOP values over the course of DISE. When adjusted for sedation depth (PSI), deeper sedation was weakly associated with greater PHOP. However, the magnitude of this change was small and we conclude that PhOP remains relatively stable over the course of DISE (Effects of Lung Volume on Upper Airway Patency During DISE [DISE-Pulm], NCT05350332, clinicaltrials.gov).

Association Between Soft Tissue Measures From Computed Tomography and Upper Airway Collapsibility on Drug-Induced Sleep Endoscopy

OBJECTIVE

Positive airway pressure (PAP) titration during drug-induced sleep endoscopy (DISE) provides objective measures of upper airway collapsibility. While skeletal measurements relate to collapsibility measures on DISE, the influence of soft tissue dimensions on upper airway collapsibility is not known. We analyzed the relationship of measures of upper airway soft tissue volumes, specifically soft palate, pharyngeal lateral walls, and tongue, with metrics of collapsibility.

STUDY DESIGN

Cross-sectional analysis from a prospective cohort.

SETTING

Academic medical center.

METHODS

Patients seeking PAP alternative therapies for obstructive sleep apnea (OSA) underwent standardized supine computed tomography (CT) acquisition and DISE protocols. The CT analysis primarily focused on soft tissue volumes and, secondarily, on airway and skeletal volumetric measures. DISE with PAP administration (DISE-PAP) enabled the determination of the pressure at which inspiratory airflow first commenced (pharyngeal critical pressure, PcritA) and the pressure at which inspiratory flow limitation was abolished (pharyngeal opening pressure, PhOP). Both unadjusted and adjusted correlation analyses were performed to understand the relationship between upper airway anatomy and either PcritA or PhOP.

RESULTS

One hundred thirty-nine subjects completed both CT and DISE-PAP. On average, patients were male (70.5%), white (84.2%), middle-aged (56.6 ± 13.5 years), and overweight (29.6 ± 4.7 kg/m2), with moderate-severe apnea-hypopnea index (29.7 ± 21.3 events/h). Adjusted for age, sex, body mass index, and skeletal volumes, soft palate, and lateral pharyngeal wall volumes were not associated with PhOP or PcritA, but a larger tongue was associated with more positive PhOP (⍴ = 0.20, P = .02), and more positive PcritA (⍴ = 0.16, P = .07). Exploratory analyses revealed smaller minimum cross-sectional retropalatal area and intramandibular volume were also associated with increased collapsibility measures.

CONCLUSION

After controlling for clinical factors and skeletal volume, greater tongue volume was associated with more severe collapsibility during DISE. These results, in concert with previous work, suggest that greater tongue volume in a smaller skeletal dimensions contribute to the severity of airway collapsibility, a key driver of OSA pathogenesis.

Multilevel Airway Obstruction Phenotypes in Adult OSA

Objective

To describe multilevel phenotypes of airway obstruction identified on drug-induced sleep endoscopy (DISE) in adults.

Study Design

Retrospective chart review.

Setting

Tertiary care center.

Methods

Video recordings of DISE on adult patients were retrospectively scored. A cross-correlation matrix was created to detect significant correlations between DISE findings at anatomical subsites. Three multilevel phenotypes resulted from the matrix: complete collapse at the tongue base with complete collapse at the epiglottis (T2-E2), complete circumferential obstruction at the velum with complete lateral pharyngeal wall collapse at the oropharynx (V2C-O2LPW), and incomplete collapse at the velum with complete collapse due to tonsillar hypertrophy (V0/1-O2T). The mean difference (MD) and 95% confidence interval (CI) were calculated for demographic and polysomnogram metrics of each phenotype compared to all other subjects.

Results

Phenotype 1 (T2-E2) (n = 88) had older ages (MD 5.784 years, CI [1.992, 9.576]), lower body mass index (BMI) (MD -1.666 kg/m², CI [02.570, -0.762]), and smaller neck circumferences (MD -0.448 in., CI [-9.14, -0.009]) than the other phenotypes. Phenotype 2 (V2C-O2LPW) (n = 25) had higher BMIs (MD 2.813 kg/m², CI [1.362, 4.263]), higher neck circumference (MD 0.714 in., CI [0.004, 1.424]), and higher apnea-hypopnea index (MD 8.252, CI [0.463, 16.041]). Phenotype 3 (V0/1-O2T) (n = 20) had younger ages (MD -17.697, CI [-25.215, -11.179]).

Conclusion

Three distinct multilevel phenotypes of obstruction were identified on DISE, suggesting different anatomic subsites collapse in a nonrandom pattern. The phenotypes appear to represent distinct patient groups and their identification may have implications in terms of pathophysiology and treatment modalities.

An updated review of pediatric drug-induced sleep endoscopy

Objectives

Drug-induced sleep endoscopy (DISE) involves assessment of the upper airway using a flexible endoscope while patients are in a pharmacologically-induced sleep-like state. The aim of this article is to review the current literature regarding the role of DISE in children with obstructive sleep apnea (OSA). The indications, typical anesthetic protocol, comparison to other diagnostic modalities, scoring systems, and outcomes are discussed.

Methods

A comprehensive review of literature regarding pediatric DISE up through May 2017 was performed.

Results

DISE provides a thorough evaluation of sites of obstruction during sedation. It is typically indicated for children with persistent OSA after tonsillectomy, those with OSA without tonsillar hypertrophy, children with risk factors predisposing then to multiple sites of obstruction, or when sleep-state dependent laryngomalacia is suspected. The dexmedotomidine and ketamine protocol, which replicates non-REM sleep, appears to be safe and is often used for pediatric DISE, although propofol is the most commonly employed agent for DISE in adults. Six different scoring systems (VOTE, SERS, Chan, Bachar, Fishman, Boudewyns) have been used to report pediatric DISE findings, but none is universally accepted.

Conclusions

DISE is a safe and useful technique to assess levels of obstruction in children. There is currently no universally-accepted anesthetic protocol or scoring system for pediatric DISE, but both will be necessary in order to provide a consistent method to report findings, enhance communication among providers and optimize surgical outcomes.

Level of Evidence

N/A.