Hypoglossal Nerve Stimulation and Cardiovascular Outcomes for Patients With Obstructive Sleep Apnea

Importance Sham-controlled trials are needed to characterize the effect of hypoglossal nerve stimulation (HGNS) therapy on cardiovascular end points in patients with moderate-severe obstructive sleep apnea (OSA).Objective To determine the effect of therapeutic levels of HGNS, compared to sham levels, on blood pressure, sympathetic activity, and vascular function.Design, Setting, and Participants This double-blind, sham-controlled, randomized crossover therapy trial was conducted from 2018 to 2022 at 3 separate academic medical centers. Adult patients with OSA who already had an HGNS device implanted and were adherent and clinically optimized to HGNS therapy were included. Participants who had fallen asleep while driving within 1 year prior to HGNS implantation were excluded from the trial. Data analysis was performed from January to September 2022.Interventions Participants underwent a 4-week period of active HGNS therapy and a 4-week period of sham HGNS therapy in a randomized order. Each 4-week period concluded with collection of 24-hour ambulatory blood pressure monitoring (ABPM), pre-ejection period (PEP), and flow-mediated dilation (FMD) values.Main Outcomes and Measures The change in mean 24-hour systolic blood pressure was the primary outcome, with other ABPM end points exploratory, and PEP and FMD were cosecondary end points.Results Participants (n = 60) were older (mean [SD] age, 67.3 [9.9] years), overweight (mean [SD] body mass index, calculated as weight in kilograms divided by height in meters squared, 28.7 [4.6]), predominantly male (38 [63%]), and had severe OSA at baseline (mean [SD] apnea-hypopnea index, 33.1 [14.9] events/h). There were no differences observed between active and sham therapy in 24-hour systolic blood pressure (mean change on active therapy, -0.18 [95% CI, -2.21 to 1.84] mm Hg), PEP (mean change on active therapy, 0.11 [95% CI, -5.43 to 5.66] milliseconds), or FMD (mean change on active therapy, -0.17% [95% CI, -1.88% to 1.54%]). Larger differences between active and sham therapy were observed in a per-protocol analysis set (n = 20) defined as experiencing at least a 50% reduction in apnea-hypopnea index between sham and active treatment.Conclusions and Relevance In this sham-controlled HGNS randomized clinical trial, mean 24-hour systolic blood pressure and other cardiovascular measures were not significantly different between sham and active HGNS therapy. Several methodologic lessons can be gleaned to inform future HGNS randomized clinical trials.