Changes in Hypoxemia Metrics After Upper Airway Surgery for Obstructive Sleep Apnea

Abstract

AbstractObjectiveThe apnea–hypopnea index (AHI) quantifies obstructive sleep apnea (OSA) severity but has limited correlation with subjective symptoms and cardiovascular risk. Sleep parameters related to oxygenation status and hypoxic burden may offer utility as additional predictors of cardiovascular risk and the physiologic consequences of OSA. This study aims to assess the change in measures of oxygenation status following sleep surgery for OSA.Study DesignA retrospective cohort study.SettingA single‐institution tertiary care center.MethodsPatients who underwent hypoglossal nerve stimulator (HGNS) placement, expansion sphincter pharyngoplasty (ESP), or maxillomandibular advancement (MMA) for continuous positive airway pressure‐intolerant OSA were included. The percentage of sleep time below SaO2 90% (pT90) was used to characterize oxygenation status. Wilcoxon signed‐rank test compared changes in sleep metrics from preoperatively to postoperatively. Linear regression was performed to assess whether changes in pT90 were correlated with changes in AHI and Epworth Sleepiness Scale (ESS), respectively.ResultsA total of 219 patients were included (65% HGNS, 24% ESP, 11% MMA). The average decline in AHI and ESS was 11.9 events/h (standard deviation [SD] = 19.9, P < .001) and 3.2 points (SD = 4.7, P < .001), respectively. Among measures of sleep oxygenation, pT90 (11.5 → 8.9, P < .001), oxygen desaturation index (27.1 → 15.1, P < .001), and SpO2 nadir (79.9 → 81.2, P = .03) improved significantly following surgery. On linear regression, changes in pT90 were significantly correlated with changes in AHI (β = 0.31, 95% confidence interval [CI] 0.18‐0.45, P < .001) and ESS (β = .05, 95% CI 0.00, 0.09, P = .038), respectively.ConclusionUpper airway surgery improves metrics of oxygen status in addition to the AHI. Changes in pT90 correlate with subjective symptoms of OSA per ESS, suggesting utility in capturing disease burden.