A Two-Year Comparative Analysis of Upper Airway Stimulation and Continuous Positive Airway Pressure Therapy Outcomes

Introduction: Upper airway stimulation (UAS) is an effective treatment for obstructive sleep apnea (OSA) patients who fail continuous positive airway pressure (CPAP) therapy. Previous studies have demonstrated UAS’s efficacy in reducing disease severity with high compliance rates. This study directly compares functional, behavioral, and physiological treatment outcomes in patients treated with CPAP and UAS over the course of two years.

Methods: A prospective cohort study was conducted at a single institution. Eligible subjects had moderate to severe OSA (AHI 15-65), had BMI ≤ 35, and were medically eligible to receive UAS or CPAP therapy. Demographics, Epworth sleepiness scale (ESS), Functional Outcomes of Sleep Questionnaire (FOSQ), WatchPAT home sleep test metrics, and device usage data were collected at baseline and every 6 months for 2 years. Adherence was defined as device usage ≥4 hrs/night for ≥70% of nights. Mean disease alleviation (MDA) was calculated to assess overall disease control. Mann-Whitney U tests and linear regressions were performed in Python.

Results: 30 CPAP (57.8 ± 10.0 years old, 83% male) and 20 UAS (63.0 ± 18.0 years old, 85% male) subjects were enrolled. At baseline, CPAP patients had worse ESS, FOSQ, AHI, RDI, and oxygen (O2) nadir) (p < 0.05). Using pairwise comparisons, at 6 months, UAS showed greater adjusted compliance (p < 0.01), and at 12 months, CPAP showed better treatment efficacy (p < 0.05). At 18+ months, differences between groups (n = 8 CPAP, 8 UAS) were not significant. Findings were consistent among the subjects who used their devices on the nights of their sleep studies.

Using a mixed effects model, CPAP demonstrated greater improvements in RDI (β = -0.891 vs 0.002), AHI (β = -0.903 vs -0.066), and O2 nadir (β = 0.491 vs -0.136) per month at 0.05 significance level. MDA for CPAP also increased by ~0.024/month (p < 0.05).

Conclusions: CPAP patients showed greater improvements in RDI, AHI, and O2 nadir, but started with worse baseline metrics, suggesting more room for improvement. UAS appeared less effective, but better tolerated. More data is needed to more accurately assess longitudinal effects.