Outcome of CPAP Titration for Moderate-to-Severe OSA Under Drug-Induced Sleep Endoscopy: A Randomized Controlled Crossover Trial

Background: The titration pressure of continuous positive airway pressure (CPAP) is important in patients with obstructive sleep apnea (OSA). This study aimed to understand the difference between drug-induced sleep endoscopy (DISE)-guided CPAP titration and conventional sleep center (CSC) CPAP titration in patients with OSA. Methods: In this randomized, controlled, and single-blind crossover trial, we compared the effects of 1-month CPAP treatment in patients with OSA with either DISE-guided CPAP titration or CSC CPAP titration. Twenty-four patients with OSA were recruited for the study. All patients underwent polysomnography, DISE-guided CPAP titration, and accommodation. Initially, patients were randomly assigned to receive either DISE-guided CPAP titration or CSC CPAP treatment for the first month. They were then switched to other treatments in the second month. The Epworth sleepiness scale (ESS) score was recorded at baseline, 1 and 2 months. Results: The upper limit of the pressure of DISE-guided titration and CSC CPAP titration was not significantly different (13.9 ± 0.7 vs. 13.5 ± 0.5 cm H₂O; P = 0.92). The residual apnea-hypopnea index and compliance were also not significantly different between the groups. ESS score significantly improved from baseline to 1 month after CPAP treatment in both groups. Both epiglottis (anterior-posterior collapse) and tongue base collapse were significantly associated with 95% CPAP pressure (P = 0.031 and 0.038, respectively). After multivariate regression analyses, the epiglottis (anterior-posterior collapse) was an independent factor for 95% CPAP pressure. The incidence rate of bradycardia was 58.3%, which is a safety concern for DISE. Despite the high incidence of bradycardia, all patients with bradycardia recovered with proper management. Conclusion: Both modalities were comparable in terms of establishing the pressure settings required to treat patients. Further large-scale studies are required to confirm these results. Trial registration: https://clinicaltrials.gov/, NCT03523013.