Predictors of Hypoglossal Nerve Stimulator Usage

Introduction: Hypoglossal nerve stimulator (HGNS) is an effective therapy for eligible patients with obstructive sleep apnea (OSA) and continuous positive airway pressure (CPAP) intolerance. A recent study evaluating usage trends showed that 85% of patients were excellent or good users with mean therapy use over 6 hours per night. The current study aims to describe usage over time and identify predicators of usage.

Methods: A single institution retrospective chart review was merged with Inspire SleepSync usage data. Data for 50 patients is presented and analysis of a larger cohort is underway. Of the patients, 66% were male and 34% were female with a mean age of 59.9 years (±11.4). Usage data was available for a mean of 582 days following device activation. Regression analysis was performed to assess the change over time with cross-sectional sampling of usage data for mean nightly usage and days used per month. Multivariate linear regression analysis was performed to determine predictors related to patient demographics, OSA severity, and comorbidities.

Results: There was a statistically significant decrease in total days used per month (r = -0.36, p < 0.0001). A decrease in mean nightly usage was not significant (r = -0.12, p = 0.12). Insomnia was negatively associated with days used per month at 3 months (p = 0.02). At 6 months, hypertension (p = 0.008) and anxiety (p = 0.044) were negatively associated with days used per month. At 1 year, male sex was negatively associated with days used per month (p = 0.026). At 1 year, “never-smoker” status was positively associated with mean nightly usage (p = 0.035) and insomnia was negatively associated with mean nightly usage (p = 0.007).

Conclusions: There is an overall decrease in HGNS usage over time. Male sex, hypertension, anxiety, and insomnia may be associated with lower therapy usage over time. Never-smoker status may be positively associated with therapy usage. Next steps include conducting studies with a larger cohort to verify these findings. Identifying predictors of usage will enable the development of targeted interventions to improve patient adherence.