Can drug-induced sleep endoscopy (DISE) predict compliance with positive airway pressure therapy? A pilot study.

Abstract

Drug-induced sleep endoscopy (DISE) has been poorly exploredas an examination to assess positive airway pressure (PAP) therapy in patients with obstructive sleep apnea (OSA). The present study aimed to identify by DISE possible characteristics related to low compliance with PAP therapy due to respiratory complaints. Patients using PAP for OSA underwent DISE in two conditions: (1) baseline (without PAP) and (2) PAP (with the same mask and airway pressure used at home). We compared patients reporting low compliance to PAP due to respiratory complaints to those well-adapted to therapy. VOTE classification (assessment of velopharynx, oropharynx, tongue base, and epiglottis)and TOTAL VOTE score (the sum of VOTE scores at each anatomical site) were assessed. ROC curve analyzedthe accuracy of TOTAL VOTE to predict low compliance due to persistent pharyngeal obstruction in both conditions. Of 19 patients enrolled, all presented multilevel pharyngeal obstruction at baseline condition, with no difference between groups at this study point. When PAP was added, the median VOTE value was higher in the epiglottis (P value=0.02) and tended to be higher at the velum and tongue base in the poorly adapted group; TOTAL VOTE score was also significantly increased in patients withlow compliance (P value<0.001). ROC curve demonstrated that patients with TOTAL VOTE scored 2.5 or more during DISE with PAP presented a 4.6-fold higher risk for low compliance with PAP therapy due to pharyngeal obstruction (AUC: 0.88±0.07; P value<0.01; sensitivity: 77%; specificity: 83%). Adding PAP during aDISE examination mayhelp to predict persistent pharyngeal obstruction during PAP therapy.