A scoping review highlighting the need for outcomes research in hypoglossal nerve stimulation for patients with Down syndrome

INTRODUCTION

Obstructive sleep apnea (OSA) is common in individuals with Down syndrome (DS).¹ Hypoglossal nerve stimulation (HNS) has been shown to be effective in positive airway pressure (PAP)–intolerant adults with OSA,² and in March 2023, the Food and Drug Administration approved HNS in pediatric patients with DS. A current appraisal of the literature investigating HNS in adults and children with DS and OSA has not been reported.

DATA SOURCES

Data sources used were Ovid Medline, EMBASE, Scopus, and ClinicalTrials.gov from January 2014 to August 2023.

REVIEW METHODS

Manuscripts were systematically reviewed (2 reviewers) with a priori selection criteria (Figure 1). Inclusion criteria were as follows: (1) DS and OSA and (2) ≥ 1 outcome measure (apnea-hypopnea index [AHI]), O₂ Saturation, Quality-of-Life questionnaires, and/or complications). AHI type was clarified based on sleep study type: (1) titration AHI (t-AHI) represents optimal AHI identified during postimplant titration polysomnogram (T-PSG), (2) overall AHI (o-AHI) represents average AHI during a T-PSG (includes many voltage levels), and (3) nontitration AHI (NT-AHI) represents respiratory events measured during a nontitration polysomnogram (NT-PSG) or home sleep apnea test (HSAT), reflecting both reduction in AHI at consistent voltage and device usage. Duration of therapy usage was a secondary outcome. The exclusion criterion was overlapping patient populations (determined by study evaluation and author contact). Risk of bias was evaluated using the Newcastle-Ottawa Quality Assessment Scale.^3,4

Figure 1. Systematic review flowchart.

RESULTS

Forty-two identified studies led to 17 fully reviewed and 5 included (2 pediatric and 3 adult studies (Table 1). Four retrospective studies (n = 1–3)^5,7–9 and 1 prospective study (n = 42) are included.⁶ Risk-of-bias assessment demonstrated poor quality for the case reports^5,7–9 and fair quality for the prospective study.⁶

Table 1.

Included studies.

Study	Patient Population	Total Cases	Mean Age, y	Mean Baseline NT-AHI (SD)	Mean Follow-up NT-AHI (SD)	QOL Baseline (SD)	QOL Follow-up (SD)	Usage Hours/Night (SD)	Quality Score
Kay et al (2021–retrospective)5	Pediatric	1	13	44.9	12.2	N/A	N/A	10.5	4
Yu et al (2022–prospective)6	Pediatric	42	15 (3.0)	23.5 (9.7)	11.0 (13.4)	66 (19.8)	31.3 (10.8)	9.0 (1.8)	6
Li et al (2019–retrospective)7	Adult	3	37.3	48.6 (13.2)	7.3 (4.5)	N/A	N/A	8.2	5
Van de Perck et al (2018–retrospective)8	Adult	1	23	61.5	23	N/A	N/A	7.45	5
Scheffler et al (2023–retrospective)9	Adult	1	32	58.2	13.5	N/A	N/A	10	5

N/A = not applicable, NT-AHI = apnea-hypopnea index (events/h) on nontitration polysomnogram or home sleep apnea test, QOL = quality of life.

Two pediatrics studies (n = 43) were evaluated. Kay et al⁵ (retrospective, n = 1) found an NT-AHI reduction of 73% at 8 months by HSAT and mean usage of 10.5 hours per night. Yu et al⁶ (prospective, n = 42) did not report T-PSG measures but reported a mean NT-AHI reduction of 53% at 12 months, an average OSA-18 quality-of-life score improvement, few minor complications, and mean usage of 9.0 hours per night.

Three retrospective adult case series (n = 5) were evaluated. T-PSG in 2 studies (n = 4)^7,8 demonstrated t-AHI reduction from 81% to 100%. Li et al (n = 3)⁷ reported a second study for each patient (T-PSG in 2; HSAT in 1). In T-PSG patients, 1 demonstrated an o-AHI of 40 events/h and t-AHI of 12 events/h at 9 months, the other o-AHI of 6.8 events/h and a t-AHI of 0 events/h at 11 months. In the HSAT patient, an NT-AHI of 38.5 events/h at 28 months was reported. Van de Perck et al⁸ (n = 1) demonstrated an NT-AHI reduction to 23.0 events/h at 6 months, a 63% reduction.⁸ Scheffler et al⁹ (n = 1) demonstrated a t-AHI of 7.2 events/h (all centrals), then an NT-AHI of 13.5 events/h (45% centrals) on NT-PSG.⁹ One minor, self-resolving facial neuropraxia was reported.⁷ Mean usage was 8.6 hours per night.^7–9

DISCUSSION

With the pathophysiology of OSA in patients with DS including relative macroglossia, HNS is a logical treatment option for those who are PAP-intolerant. The only review on HNS in patients with DS reported on children and suggested improved AHI,¹⁰ but 7 of the 9 studies included overlapping patient populations and AHI type was not delineated. Thus, we sought to identify all studies reporting patients with DS and OSA following HNS provision, as well as to report outcomes measures and to clarify AHI type.

Based on our systematic search, as expected, AHI was the most reported polysomnographic outcome, showing consistent reduction in both pediatric and adult populations. One report demonstrated quality-of-life improvement,⁶ while excellent usage and safety profiles appeared to be consistent.^5–9 Overall, the studies demonstrate that HNS appears to be safe and potentially effective. However, the existing body of literature is scarce. Aside from Lu et al,⁶ studies are exclusively retrospective and include 1–3 patients, limiting quality, conclusions, and external validity. One area of clarification was type of sleep study, titration vs nontitration, and the value of AHI. t-AHI can be reported as an outcome, but it really is a target, while o-AHI is of questionable utility. The better capture of efficacy is NT-AHI, measured once the patient reaches what are clinically felt to be optimal settings and incorporating usage, which is often higher than the other measures.

The limitations herein include the dearth of reports, small populations, absence of controls, and limited overlap of polysomnographic outcomes. This systematic search suggests a reduction in AHI and a modest complication profile, but the limitations are significant. Thus, we primarily highlight the significant need for reports with greater numbers, more rigorous study criteria, and clear outcome measures to improve our understanding in this important population and to help guide clinical care.

DISCLOSURE STATEMENT

The co-authors N.P.O. and S.M. were members of the Physicians Advisory Council for Inspire Medical Systems, Inc, and, as such, received intermittent consultatory payments; the predetermined conclusion of their term was May 2023. S.M. is also on the faculty of the Fellowship training program for Inspire Medical Systems, Inc, and receives an honorarium and travel funding. The other authors report no conflicts of interest.