A, The apnea-hypopnea index decreased from baseline by a mean of 25.2 events/h (95% CI, 23.6-26.7 events/h) at 6 months and 16.5 events/h (95% CI, 14.3-18.7 events/h) at 12 months. B, Mean difference in the Epworth Sleepiness Scale score was 4.6 points (95% CI, 4.0-5.1 points) at 6 months and 5.2 points (95% CI, 4.4-6.0 points) at 12 months. C, Mean difference in the Functional Outcomes of Sleep Questionnaire score was 3.7 (95% CI, 2.8-4.7) at 6 months and 3.4 (95% CI, 2.9-3.9) at 12 months. D, Mean difference in the oxygen desaturation nadir was 7.0% (95% CI, 5.5%-8.5%) at 6 months and 5.5% (95% CI, 4.2%-6.8%) at 12 months. Boxes represent 25th to 75th quartiles separated by the median value. Whiskers capture data within the 1.5-interquartile range. Each dot represents an individual patient.
eTable 1. Change in Clinical Outcomes at 6 Months Postoperatively
eTable 2. Change in Clinical Outcomes at 12 Months Postoperatively
eTable 3. Linear Regression Analysis of Factors Related to Apnea-Hypopnea Index Outcomes 6 Months After Hypoglossal Nerve Stimulator Implantation
eTable 4. Linear Regression Analysis of Factors Related to Epworth Sleepiness Scale (ESS) Outcomes 6 Months After Hypoglossal Nerve Stimulator Implantation
eTable 5. Linear Regression Analysis of Factors Related to Oxygen Desaturation Nadir (O2N) Outcomes 6 Months After Hypoglossal Nerve Stimulator Implantation
eTable 6. Linear Regression Analysis of Factors Related to Apnea-Hypopnea Index (AHI) Outcomes 12 Months After Hypoglossal Nerve Stimulator Implantation
eTable 7. Linear Regression Analysis of Factors Related to Oxygen Desaturation Nadir (O2N) Outcomes 12 Months After Hypoglossal Nerve Stimulator Implantation
eTable 8. Linear Regression Analysis of Factors Related to Epworth Sleepiness Scale (ESS) Outcomes 12 Months After Hypoglossal Nerve Stimulator Implantation
eTable 9. Linear Regression Analysis of Factors Related to Functional Outcomes of Sleep Questionnaire (FOSQ) Outcomes 12 Months After Hypoglossal Nerve Stimulator Implantation
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Kent DT, Carden KA, Wang L, Lindsell CJ, Ishman SL. Evaluation of Hypoglossal Nerve Stimulation Treatment in Obstructive Sleep Apnea. JAMA Otolaryngol Head Neck Surg. 2019;145(11):1044–1052. doi:10.1001/jamaoto.2019.2723
What factors are associated with response to hypoglossal nerve stimulation treatment of obstructive sleep apnea across multiple clinical cohorts?
In this pooled cohort analysis of 4 observational cohorts comprising 584 patients, greater postoperative improvement in the apnea-hypopnea index was found to be associated with higher preoperative apnea-hypopnea index, older patient age, and lower body mass index.
Hypoglossal nerve stimulation was associated with clinically significant improvements in obstructive sleep apnea severity, daytime sleepiness, and sleep-related quality of life, and patient characteristics were associated with degree of improvement.
Hypoglossal nerve stimulation is a treatment option for patients with obstructive sleep apnea unable to tolerate continuous positive airway pressure. This study evaluates demographic factors that may be associated with greater improvements in postoperative outcomes of interest.
To examine the association of hypoglossal nerve stimulation with obstructive sleep apnea severity, daytime sleepiness, and sleep-related quality of life.
Design, Setting, and Participants
Patient-level data were pooled from 3 prospective cohorts and 1 retrospective observational cohort comprising 584 adults with moderate to severe obstructive sleep apnea unable to tolerate or benefit from continuous positive airway pressure. The data were gathered from the Stimulation Therapy for Apnea Reduction Trial; a postmarket approval study conducted in Germany; the multicenter, international Adherence and Outcome of Upper Airway Stimulation for OSA Registry; and a retrospective cohort study from 2 sites in the United States.
Hypoglossal nerve stimulation.
Main Outcomes and Measures
Severity of obstructive sleep apnea was the primary outcome. The apnea-hypopnea index (AHI) (<5, normal; 5-15, mild; 15-30, moderate, and >30, severe) and Epworth Sleepiness Scale (range, 0-24; score >10 indicates pathologic sleepiness) outcomes were available at 2 to 6 months from 2 cohorts (n = 398), at 12 months from 1 cohort (n = 126), and at both times from 1 cohort (n = 60). Sleep-related quality of life and oxygen saturation nadir data were collected where available. Linear mixed-effects models were constructed to examine associations between clinical variables and reported postoperative outcomes at 6 and 12 months with study included as a random effect.
Of the 584 patients included in the study, 472 were men (80.8%); mean (SD) age was 58.5 (11.0) years. Greater improvement in the postoperative AHI was associated with a higher preoperative AHI (−0.74 events/h; 95% CI, −0.82 to −0.67), older patient age (−0.10 events/h; 95% CI, −0.20 to −0.00), and lower body mass index (0.52; 95% CI, 0.22-0.83). After adjusting for these variables and considering all patients in the analysis, the AHI was statistically higher at 12 months than at 6 months (3.24 events/h; 95% CI, 1.67-4.82 events/h).
Conclusions and Relevance
Hypoglossal nerve stimulation demonstrated clinically significant improvements in obstructive sleep apnea severity, daytime sleepiness, and sleep-related quality of life in this pooled cohort of patient-level results. Age, body mass index, and preoperative AHI appeared to be associated with treatment outcomes, and these variables may explain some of the difference between 2- to 6-month and 12-month outcomes.
Obstructive sleep apnea is a common disorder that has been independently associated with a variety of health conditions.1-4 Continuous positive airway pressure (CPAP) is the first-line therapeutic choice in adults, but patient tolerance and adherence to therapy remain problematic.5 Alternative treatment modalities have been developed, including mandibular advancement devices, positional therapy devices, soft tissue and craniofacial surgeries, as well as hypoglossal nerve stimulation.
Animal research in the 1990s confirmed that the genioglossus muscle is the primary pharyngeal dilator muscle during sleep, with secondary support from the palatal musculature.6-10 Results from subsequent human trials of transcutaneous and direct genioglossus muscle stimulation were inconsistent, showing that stimulation improved airflow but also variably caused patient arousal and discomfort.11-16 Further efforts demonstrated that recruitment of the genioglossus muscle via hypoglossal nerve stimulation yielded more consistent improvements in airflow and better patient tolerance.17
An initial hypoglossal nerve stimulation prototype (Inspire I; Inspire Medical Systems Inc) was implanted into 8 patients in 2001. Initial results suggested improvements in obstructive sleep apnea disease burden, but multiple device failures occurred within the first year.18 A second-generation device (Inspire II; Inspire Medical Systems Inc) was implanted into 30 patients as part of a staged feasibility study published in 2011.19 Factors associated with success included an apnea-hypopnea index (AHI) less than or equal to 50 events/h, body mass index (BMI) less than or equal to 32 (calculated as weight in kilograms divided by height in meters squared), and lack of circumferential palatal collapse during drug-induced sleep endoscopy. These factors were used to create enrollment criteria for 126 patients in the Stimulation Therapy for Apnea Reduction (STAR) Trial.20
The STAR trial was a prospective, multicenter, international study published in 2014.20 Based on the initial results, the US Food and Drug Administration (FDA) approved the second-generation device in 2014. Subsequently, a prospective cohort study reported outcomes from 60 patients who underwent implantation in Germany21,22 and a retrospective cohort study reported outcomes from 2 sites in the United States (N = 97).23 More recently, results from the first 301 patients enrolled in the prospective, multicenter, international Adherence and Outcome of Upper Airway Stimulation for OSA (ADHERE) Registry have been published.24 None of the patients in the US study were enrolled in the ADHERE Registry.
The second-generation device is currently the only FDA-approved hypoglossal nerve stimulation device. Other hypoglossal nerve stimulation devices have either closed experimental trials or have FDA trials currently under way.25-27 As such, the STAR trial and the US, German, and ADHERE Registry cohorts reflect the available experience with hypoglossal nerve stimulation therapy outcomes. Herein, we present a pooled analysis of patient-level data from all 4 studies to describe outcomes and identify possible preoperative factors associated with patient outcomes.
This was a pooled analysis of all available patient-level data from the 4 published studies using a single type of hypoglossal nerve stimulator (Inspire II) for obstructive sleep apnea.20-24 The study was approved by the Vanderbilt University Medical Center Institutional Review Board, with waiver of informed consent. Individual patient demographic information, as well as preimplant and postimplant polysomnographic data and questionnaire responses, were obtained from Inspire Medical Systems for the STAR, German, and ADHERE Registry cohorts and from authors of the US cohort study report.
For all studies, participants demonstrating a history of difficulty accepting or adhering to CPAP therapy were eligible for hypoglossal nerve stimulation. In the STAR trial, exclusion criteria included compromising pharyngeal anatomy (eg, 3-4+ tonsil size), complete concentric collapse of the soft palate observed during drug-induced sleep endoscopy, an AHI less than 20 or greater than 50 events/h, a central or mixed apnea index comprising greater than 25% of the total AHI, or a nonsupine AHI less than 10 events/h (AHI<5, normal; 5-15, mild; 15-30, moderate, and >30, severe).20 The German cohort excluded patients with an AHI of less than 20 or greater than 65 events/h as determined by a home sleep apnea test and those with a BMI greater than 35.21 Patients in the US and ADHERE cohorts were eligible for implantation if they had an AHI of 20 to 65 events/h, in accordance with post-FDA approval criteria.23,24 In 2017, the FDA revised the AHI indications for implantation such that some patients in the ADHERE cohort were eligible for implantation with an AHI of 15 to 65 events/h.
In each study cohort, approved patients underwent hypoglossal nerve stimulation implantation. Individual surgeon technique varied, but the core elements of the procedure were identical: a stimulation electrode was placed on select branches of the hypoglossal nerve to facilitate tongue protrusion, a respiratory sensing lead was placed between the internal and external intercostal muscles, and an implanted neurostimulator was placed over the pectoralis muscle on the ipsilateral side. The device was activated for each patient 1 month postoperatively and patients were instructed in the use of a controller to initiate and adjust stimulation. At approximately 2 months postoperatively, an in-laboratory polysomnographic study was performed in which device variables were manipulated to optimally treat obstructive sleep apnea while minimizing sleep disturbances.
The STAR trial clinical protocol required an additional overnight polysomnographic study at 6 months with further adjustment of device variables prior to an observational efficacy polysomnographic study at 12 months in which the outcome of current device settings on obstructive sleep apnea was observed overnight. The German cohort underwent postoperative home sleep apnea testing to assess therapy efficacy at 6 and 12 months after the in-laboratory titration study at 2 months. The US and ADHERE cohorts followed the standard hypoglossal nerve stimulation clinical protocol in the United States, where a 2-month postoperative titration study was conducted for device adjustment with further evaluations for efficacy conducted at the discretion of the patient’s care team.
In all 4 cohorts, the primary outcome measure was the severity of obstructive sleep apnea as measured by the AHI from in-laboratory polysomnography or the respiratory event index from home sleep apnea testing. To aid cohort comparison in this pooled analysis, respiratory event index outcomes were considered equivalent to AHI and are referred to as such. All patients completed preoperative polysomnographic testing to establish candidacy, and all patients underwent device titration at approximately 2 months. Further polysomnographic study collection varied between cohorts. In the US cohort, postoperative AHI values were recorded from the portion of the night during the 2-month titration study in which optimum stimulation factors were observed. Participants in the ADHERE registry also had AHI values recorded from the optimum portion of the night from the 2-month titration study. A portion of the participants in ADHERE also completed a 6- or 12-month postoperative home sleep apnea test as part of routine clinical care to assess full-night efficacy. The AHI values from the home sleep apnea test were recorded instead of the previous titration study AHI if available. The German cohort collected postoperative AHI values from 6- and 12-month measures in all patients.
Several additional outcomes were reported for each cohort. The STAR, US, and German cohorts recorded oxygen desaturation nadir outcomes. All cohorts measured subjective sleepiness using the Epworth Sleepiness Scale (ESS). With a range of 0 to 24, a score greater than 10 on the ESS is considered to indicate pathologic sleepiness. In the STAR and German cohorts, sleep-related quality of life was measured with the Functional Outcomes of Sleep Questionnaire (FOSQ). With a range of 5 to 20, a score less than 17.9 indicates sleep-related, disease-specific functional impairment, with lower scores indicating greater dysfunction. In the STAR and German cohorts, the ESS and FOSQ scores were collected at the time of the polysomnographic study used for AHI outcome measurement. In the US and ADHERE cohorts, the ESS score was determined at the first postoperative clinic visit after optimal device titration, typically between 2 and 6 months after implantation.
Individual patient data from all cohorts were pooled. The change in AHI, ESS score, FOSQ score, and oxygen desaturation nadir at 2 to 6 months and at 12 months were estimated with 95% CIs. The association between outcomes and age, sex, BMI, neck circumference, preoperative AHI, oxygen desaturation nadir, hypertensive status, diabetes status, and originating cohort were explored using linear regression. The ADHERE registry additionally measured preimplantation depression, and the association between depression and outcomes was evaluated separately for this cohort. To take advantage of all of the data, we then used linear mixed-effects models with an adjusted random intercept for participant by study to evaluate associations of preoperative characteristic variables and postoperative time with overall outcomes. Statistical analyses were completed using the statistical package R, version 3.3.0 (R Core Team) with add-on packages Hmisc, rms, lme4, and ggplot2.28 Values were considered significant at P < .05.
There were 584 patients in the pooled analysis, with 126 in the STAR trial, 60 in the German cohort, 97 in the US cohort, and 301 in the ADHERE Registry (Table 1). Mean (SD) age was 58.5 (11.0) years, 472 were men (80.8%), 97% were white, and mean BMI was 28.9 (3.6). Preoperatively, the mean AHI was 33.8 (15.5) events/h, mean ESS score was 11.8 (5.3), mean FOSQ score was 13.9 (3.3), and mean oxygen desaturation nadir was 78.4 (8.9). We observed no characteristic differences between cohorts (P > .75). After combining final time point data available from each cohort, the mean AHI had decreased to 11.0 (13.6) events/h (P < .001) and mean ESS score had decreased to 7.1 (4.5) points (P < .001). The FOSQ score increased to 17.3 (2.9) (P < .001) and the oxygen desaturation nadir increased to 84.3 (5.5) (P < .001).
The AHI decreased by a mean of 25.2 events/h (95% CI, 23.6-26.7) at 6 months and by 16.5 events/h (95% CI, 14.3-18.7) at 12 months (Figure, eTable 7, and eTable 8 in the Supplement). Surgical success, defined as a decrease in AHI by at least 50% and to below 20 events/h,29 was observed in 450 patients (77.1%) at their last evaluation. The AHI decreased to less than 15 in 444 patients (76.0%) and less than 5 in 244 patients (41.8%).
The ESS score decreased by a mean of 4.6 points (95% CI, 4.0-5.1 points) at 6 months and by 5.2 points (95% CI, 4.4-6.0 points) at 12 months. At their last evaluation, 437 of 580 patients (75.3%) reported an ESS score less than 10. The FOSQ scores increased by a mean of 3.7 points (95% CI, 2.8-4.7 points) at 6 months and 3.4 points (95% CI, 2.9-3.9 points) at 12 months. At the final evaluation, 100 of 179 patients (55.9%) reported an FOSQ score greater than 17.9. In addition, the oxygen desaturation nadir increased by a mean of 7.0% (95% CI, 5.5%-8.5%) at 6 months and 5.5% (95% CI, 4.2%-6.8%) at 12 months.
Exploratory regression models for each outcome adjusted for other available variables were completed (Table 2; eTables 1-9 in the Supplement). Briefly, we found that the degree of change in outcomes at 2 to 6 and at 12 months appeared to be associated with age, BMI, and preoperative obstructive sleep apnea burden. Statistical significance testing and effect size magnitudes suggest that patients with hypertension or diabetes experienced less improvement in AHI at 12 months. There were no associations with depression or neck circumference. Linear mixed-effects model findings are reported in Table 3 for the AHI, ESS score, FOSQ score, and oxygen desaturation nadir using characteristics available across all studies: age, sex, BMI, baseline objective and subjective disease measures, and time after surgery (2-6 months vs 12 months). A random intercept was included for the participant by study combination to assess the association between demographic variables and overall outcomes. As in the exploratory analyses, patients with greater preoperative disease burden tended to experience greater improvements, while younger and heavier patients experienced less improvement in disease. Greater improvement in the postoperative AHI was associated with a higher preoperative AHI (−0.74 events/h; 95% CI, −0.82 to −0.67), older patient age (−0.10 events/h; 95% CI, −0.20 to −0.00), and lower body mass index (0.52; 95% CI, 0.22-0.83). When controlling for other underlying population characteristics in the combined analysis, time since surgery resulted in an AHI statistically higher at 12 months than at 6 months with difference of only 3.24 events/h (95% CI, 1.67-4.82 events/h).
This pooled analysis of 584 patients across research trials and regular clinical practice suggests support for hypoglossal nerve stimulation as a viable therapy option for select patients with obstructive sleep apnea who are intolerant or unaccepting of CPAP therapy. The characteristics of the patients in our study were similar to those seen in other large cohorts of patients with moderate to severe obstructive sleep apnea, comprising mostly older men.5,30,31 Patients with untreated moderate to severe obstructive sleep apnea are at increased risk of all-cause and cardiovascular death.3,5,30,32,33 Surgical therapy for CPAP-intolerant patients, including hypoglossal nerve stimulation, provides symptomatic improvement and may also provide some benefit to important cardiovascular end points associated with mortality.34 Work is currently under way to better assess the outcome of hypoglossal nerve stimulation for cardiovascular end points.35
We found that patients who underwent implantation achieved substantial postoperative reductions in AHI burden at 6 months (25.2 events/h) and 12 months (16.5 events/h). Overall, 76.0% of the pooled cohort had an AHI of less than 15 events/h, and 41.8% had an AHI of less than 5 events/h by their final evaluation. In the linear mixed-effects model, the time of postoperative assessment (6 or 12 months) was associated with AHI outcomes, with a minimally greater change observed earlier than later. The STAR trial, however, showed improvement in AHI between 6 and 12 months and stable AHI outcomes up to 5 years after implantation.36 This discrepancy may stem from differences in the timing and method of AHI outcome measurements between the different cohorts.
Twelve-month data in this pooled analysis reflect the STAR trial and German cohorts, which used sleep studies without therapy adjustment to measure outcomes.20,22 In contrast, the US and ADHERE cohorts reported AHI from a segment of the postoperative titration polysomnographic testing where optimal stimulation parameters were achieved. In ADHERE, the outcome AHI was collected from a posttitration efficacy home sleep apnea test only if ordered by the treating physician (83 of 295 patients [28.1%]).23,24 Just as in CPAP titration studies, residual AHI over the course of a hypoglossal nerve stimulation titration study may vary significantly depending on multiple factors, such as patient position, depth of sleep, and tolerance of or response to therapy. To our knowledge, only 1 patient cohort has been published reporting optimal AHI from the titration study (mean [SD], 3.2 [3.5] events/h) alongside AHI from the entire night (11.5 [14.1] events/h) at 2 months after implantation.37 Other investigators have reported a mean difference of more than 10 events/h in a single cohort of 43 patients with hypoglossal nerve stimulation when comparing the residual AHI from an optimized segment of a titration study to an all-night efficacy polysomnographic study.38
Another potential explanation for differing AHI outcomes is variation in the underlying population characteristics as there were some differences in qualifying AHI and BMI. We observed no meaningful clinical differences between cohorts, but we did observe that, after adjusting for these variables, the discrepancy between 2- to 6-month and 12-month outcomes was considerably reduced. Mean AHI change between 2 to 6 months and 12 months was significantly reduced when controlling for other underlying population characteristics in the combined analysis.
We found that age and BMI were associated with AHI outcomes, with younger and heavier patients experiencing mild decreases in surgical response. The clinical importance of these findings is uncertain at this time. A recent retrospective review by Huntley et al39 found no significant difference in surgical success rates (defined as a decrease in the postoperative AHI by 50% compared with the preoperative value and to <20 events/h) when comparing patients with hypoglossal nerve stimulation whose BMIs were above and below 32. However, prior anatomic studies have demonstrated that, with increases in BMI, there are significant changes in pharyngeal anatomy, such as an increase in tongue fat deposition.40-42 Other research suggests that nonanatomic factors, such as upper airway muscle responsiveness, may affect obstructive sleep apnea presence and severity in individuals with obesity.43 A recent post hoc analysis of the ADHERE Registry reported that increasing age was a predictor of treatment success, in agreement with the results presented here.44 However, the 2 analyses are not directly comparable because the ADHERE analysis used a binary threshold for treatment success (defined as a reduction of AHI by ≥50% and to ≤20 events/h), whereas this pooled analysis examined the association of age with degree of change in AHI. Further research is needed to determine what anatomic and nonanatomic factors best define hypoglossal nerve stimulation surgical candidates.
Changes in oxygen desaturation nadir were also found to be significant at 2 to 6 months and at 12 months in our study. While the clinical importance of this finding for patients with hypoglossal nerve stimulation has not been elucidated, oxygen desaturation nadir has previously been established as an independent risk factor for nocturnal blood pressure surges and sudden cardiac death.45,46
Improvements in subjective daytime sleepiness were substantial in this pooled cohort, with an average decrease of approximately 5 points on the ESS at both 6 and 12 months. The improvements observed in daytime sleepiness are comparable with improvements seen in large studies of patients treated with CPAP.33,47,48 Sleep-related functional impairment showed similar improvements postoperatively. At baseline, 86.6% of the pooled cohort demonstrated sleep-related functional impairment (FOSQ score <17.9), consistent with previous data, suggesting that a large proportion of patients with untreated moderate to severe obstructive sleep apnea have significant sleep-related functional impairment.49 In the pooled cohort, the FOSQ score improved by 3.71 points at 6 months in the German cohort and 3.39 points at 12 months across the German and STAR cohorts; these changes surpassed the threshold change of 2.0, which indicates a clinically meaningful improvement in daily functioning.49 At 12 months, 55.9% of the pooled cohort patients had normalized FOSQ scores.
This pooled analysis is limited by the structure of the underlying cohorts. The AHI and ESS score were collected across all cohorts, but oxygen desaturation nadir and FOSQ scores were not collected as part of the largest cohort (ADHERE), limiting their analysis. In addition, the cohorts used varying AHI and BMI criteria for patient enrollment. Although this variability might be argued as limiting comparison between cohorts, including these variables in our modeling maximized the ability to draw generalizable conclusions about their association with outcomes. Perhaps the most substantial limitation is that, in the US and ADHERE cohorts, the timing and method of measuring AHI outcomes complicated accurate evaluation of the impact of postsurgical time on outcomes. Future hypoglossal nerve stimulation cohorts would benefit from clearly delineating postoperative AHI data as originating from an overall titration study AHI, a titration study AHI during optimal stimulation settings, or an overall efficacy study AHI without any hypoglossal nerve stimulation therapy adjustment. More recently, investigators have advocated for standardization of hypoglossal nerve stimulation outcome data, recommending 4% oxygen desaturation index from an all-night home sleep apnea test or polysomnographic efficacy studies as a primary outcome measure.38 This standardization would assist clinicians in understanding the likely effects of hypoglossal nerve stimulation therapy under usual care conditions in their own patient populations. In addition, these results are representative of only 1 hypoglossal nerve stimulation device. Other devices, using different approaches to hypoglossal nerve stimulation therapy, will need to be further studied to understand their association with clinical outcomes in patients with obstructive sleep apnea.27,50
To our knowledge, this is the first study to pool the results of multiple hypoglossal nerve stimulation cohorts to assess associations between preoperative disease measures and treatment outcomes. Hypoglossal nerve stimulation appeared to demonstrate clinically significant improvements in objective measures of obstructive sleep apnea severity and subjective measures of daytime sleepiness and sleep-related quality of life in this cohort of CPAP-intolerant patients with moderate to severe obstructive sleep apnea. Younger and heavier adults tended to have less improvement in disease.
Accepted for Publication: July 31, 2019.
Corresponding Author: David T. Kent, MD, Department of Otolaryngology, Vanderbilt University Medical Center, 1215 21st Ave S, MCE-ST, Ste 7209, Nashville, TN 37232 (firstname.lastname@example.org).
Published Online: September 26, 2019. doi:10.1001/jamaoto.2019.2723
Author Contributions: Dr Kent had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Kent, Carden, Ishman.
Acquisition, analysis, or interpretation of data: Kent, Wang, Lindsell, Ishman.
Drafting of the manuscript: Kent.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Kent, Wang, Lindsell.
Obtained funding: Kent.
Administrative, technical, or material support: Kent.
Conflict of Interest Disclosures: Dr Lindsell reported receiving grants from the National Center for Advancing Translational Sciences during the conduct of the study and grants from the National Institutes of Health (NIH), Department of Defense, Eccrine Systems, Entegrion, Baxter, and Marcus Foundation outside the submitted work. Dr Ishman reported receiving grants from Inspire Medical during the conduct of the study and grants from the NIH-National Heart, Lung, and Blood Institute and American Society of Pediatric Otolaryngology outside the submitted work. No other disclosures were reported.
Funding/Support: This study was supported by a grant from the Vanderbilt University Medical Center Institute for Clinical and Translational Research.
Role of the Funder/Sponsor: The funding organization had no role in the design and conduct of the study; analysis and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: Inspire Medical Systems, Ryan J. Soose, MD, Maurits S. Boon, MD, and Colin T. Huntley, MD, contributed cohort data for this pooled analysis; there was no financial compensation.
Create a personal account or sign in to: