Importance: Evidence is lacking from randomized clinical trials of hypoglossal nerve stimulation in obstructive sleep apnea (OSA).
Objective: To evaluate the safety and effectiveness of targeted hypoglossal nerve stimulation (THN) of the proximal hypoglossal nerve in patients with OSA.
Design, setting, and participants: This randomized clinical trial (THN3) was conducted at 20 centers and included 138 patients with moderate to severe OSA with an apnea-hypopnea index (AHI) of 20 to 65 events per hour and body mass index (calculated as weight in kilograms divided by height in meters squared) of 35 or less. The trial was conducted from May 2015 through June 2018. Data were analyzed from January 2022 through January 2023.
Intervention: Implant with THN system; randomized 2:1 to activation at month 1 (treatment) or month 4 (control). All received 11 months of THN with follow-up at months 12 and 15, respectively.
Main outcomes and measures: Primary effectiveness end points comprised AHI and oxygen desaturation index (ODI) responder rates (RRs). Treatment responses at months 4 and 12/15 were defined as a 50% or greater reduction in AHI to 20 or less per hour and an ODI decrease of 25% or greater. Coprimary end points comprised (1) month 4 AHI and ODI RR in the treatment greater than the control group and (2) month 12/15 AHI and ODI RR in the entire cohort exceeding 50%. Secondary end points included sleep apnea severity (AHI and ODI) and patient-reported outcomes (Epworth Sleepiness Scale, Functional Outcomes of Sleep Questionnaire, and EQ-5D visual analog scale).
Results: Among 138 participants, the mean (SD) age was 56 (9) years, and 19 (13.8%) were women. Month 4 THN RRs were substantially greater in those in the treatment vs control group (AHI, 52.3% vs 19.6%; ODI, 62.5% vs 41.3%, respectively) with treatment-control standardized mean differences of 0.725 (95% CI, 0.360-1.163) and 0.434 (95% CI, 0.070-0.843) for AHI and ODI RRs, respectively. Months 12/15 RRs were 42.5% and 60.4% for AHI and ODI, respectively. Improvements in AHI, ODI, Epworth Sleepiness Scale, Functional Outcomes of Sleep Questionnaire, and EQ-5D visual analog scale scores were all clinically meaningful (medium to large effect size). Two serious adverse events and 100 nonserious related adverse events were observed from the implant procedure or study protocol.
Conclusions and relevance: This randomized clinical trial found that THN demonstrated improvements in sleep apnea, sleepiness, and quality of life in patients with OSAs over an extended AHI and body mass index range without prior knowledge of pharyngeal collapse pattern. Clinically meaningful improvements in AHI and patient-reported responses compared favorably with those of distal hypoglossal nerve stimulation trials, although clinically meaningful differences were not definitive for ODI.
Trial registration: ClinicalTrials.gov Identifier: NCT02263859.