0599 Effects of Proximal HGN Stimulation in HTN Subjects with Moderate to Severe OSA: A 5-Year Analysis of the THN3 RCT

Abstract Introduction Hypoglossal nerve stimulation (HGNS) improves sleep-disordered breathing (SDB), oxygenation, and quality of life (QOL) in obstructive sleep apnea (OSA), but to date, has not been shown to improve cardiovascular outcomes. The THN3 randomized, controlled trial of proximal HGNS included subjects with office systolic blood pressure (SBP) ≤160mmHg or diastolic pressure ≤100mmHg. In post-hoc analyses, we examined whether antihypertensive effects of HGNS could be identified in a hypertensive THN3 subgroup. Methods THN3 participants (AHI 20-65/hr; BMI ≤ 35 kg/m2, no drug-induced sleep endoscopy screening) were followed for 5 years after HGNS system implant. In addition to annual measurements of SDB and QOL, BP was measured at every visit. Normotensive and hypertensive subgroups were fit to a linear mixed model with quadratic terms and 2 latent classes in order to estimate group BP trajectories from the time of HGNS initiation. Differences in baseline characteristics between classes were also evaluated. Results Of 137 subjects with evaluable data, 130 (95%) belonged to a normotensive class. SBP remained stable in this group, with estimated SBP mean (95%CI) of 128.9 (127.2,130.6), 129.2 (127.7,130.8), 129.5 (127.7,131.3), 129.7 (127.8,131.6), 129.8 (127.9,131.8), and 129.9 (127.3,132.5) mmHg at Baseline followed by Years 1-5 measurements, respectively. The remaining 7 (5%) subjects belonged to a hypertensive class. SBP decreased over time in these hypertensive patients with HGNS, with the estimated SBP mean trajectory of 156.4 (145.5,167.3), 142.8 (133.8,151.8), 134.8 (125.5,144.2), 132.5 (123.2,141.7), 135.7 (126.5,144.9), and 144.5 (132.2,156.9) mmHg. Aside from SBP, the groups differed at Baseline in average oxygen desaturation index (37.5±10.1 vs. 32.6±3.3/hr, normotensive vs. hypertensive) and percentage of sleep time with oxygen saturation < 90% (11.9±9.8 vs. 4.7±3.8%). No meaningful differences were detected in Baseline age, sex or BMI. Conclusion Amelioration of OSA and nocturnal hypoxemia with proximal HGNS may improve SBP in hypertensives with moderate to severe OSA. If so, HGNS may improve long-term cardiovascular outcomes. Future analyses in the ongoing OSPREY and other trials should determine whether HGNS predicts reductions in SBP. Support (if any) LivaNova