0629 Characterizing and Predicting Hypoglossal Nerve Stimulator Outcomes: Updated Analysis from an Urban Sleep Clinic

Abstract Introduction Obstructive sleep apnea (OSA) is a common condition linked to significant morbidity. Hypoglossal nerve stimulation (HGNS) is an important treatment option for selected patients with moderate to severe OSA who cannot tolerate positive airway pressure therapy. However, predictors of successful HGNS outcomes remain unclear. This pilot study sought to identify demographic and clinical factors associated with HGNS treatment success in patients at a Southwestern urban outpatient sleep clinic. Methods 72 patients who underwent HGNS implantation between 2019 and 2025 were reviewed; 30 met inclusion criteria by completing a follow-up sleep study and alive at analysis. Most excluded patients are pending follow-up. Demographics, BMI, polysomnographic data, and comorbidities were collected. Treatment success was defined as ≥50% reduction in apnea–hypopnea index (AHI) and/or post-treatment AHI < 15 events/hour at optimal settings, combined with adequate adherence. Univariable linear regression assessed continuous variables, chi-square testing assessed categorical variables, and logistic regression generated odds ratios with 95% confidence intervals. Results Of the 30 included patients, 25 achieved treatment success and 5 did not. The cohort included 19 males and 11 females; 25 patients identified as White and 5 as non-White. Three patients were using GLP-1 receptor agonists. The non-successful group had a higher mean pre-treatment AHI (43.0 ± 13.2) than the successful group (31.3 ± 18.0). Mean age at implantation was 59.7 ± 8.8 years in the non-successful group and 62.8 ± 10.4 years in the successful group. Pre- and post-treatment BMI values were similar between groups. Mean oxygen nadirs were 83% in the non-successful group and 81% ± 3.2% in the successful group. Time to device activation differed, averaging 35.8 ± 6.7 days in the non-successful group versus 56.2 ± 7.2 days in the successful group. Odds of non-success were modestly higher among patients with insomnia (OR 1.42) and among females (OR 1.18), though confidence intervals were wide. Conclusion In this pilot sample, higher baseline AHI appeared to be associated with HGNS treatment failure. Female sex and comorbid insomnia showed mild trends toward non-success. Larger studies are needed to validate these preliminary findings and to assess additional factors, including GLP-1 agonist use, that may influence HGNS outcomes. Support (if any)