Does drug-induced sleep endoscopy predict surgical success in transoral robotic multilevel surgery in obstructive sleep apnea?

Abstract

The aim of this study was to determine if drug-induced sleep endoscopy (DISE) was predictive of success for patients undergoing transoral robotic surgery (TORS) and multilevel procedures for sleep apnea. Retrospective case series of patients who underwent TORS surgery for sleep apnea METHODS: Before and after polysomnograms were analyzed to assess improvement, success, and cure. Improvement was defined as any decrease in apnea-hypopnea index (AHI), success as an AHI <20 with a decrease >50%, and cure as an AHI <5. DISE videos were scored using the NOHL (nose, oropharynx, hypopharynx, larynx) and VOTE (velum, oropharynx, tongue, epiglottis) classification systems. One hundred one patients were available for analysis. Eighty-seven percent of patients had an improvement in their AHI. Fifty-one percent met criteria for success, whereas 17% were cured. The degree of collapse at individual NOHL and VOTE subsites as well as total additive scores did not predict improvement, success, or cure. Patients with no oropharyngeal lateral collapse in the VOTE classification system were more likely to improve following surgery (P = .001); however, this effect did not hold for success or cure. Multivariate analysis of DISE variables was not predictive of success. In obstructive sleep apnea patients, there is a 51% success rate and a 17% cure rate. DISE, as scored by the NOHL and VOTE system, did not readily identify patients who would benefit most from surgery. Patients with lateral oropharyngeal collapse may be poorer candidates. Prospective, larger studies are required to further evaluate the use of DISE in predicting success following TORS. 4 Laryngoscope, 127:971-976, 2017.