National Database Analysis of Single-Level versus Multilevel Sleep Surgery

Objective Recent evidence suggests that multilevel sleep surgery improves outcomes when compared with palate surgery alone for most patients. The study objective was to compare demographic and outcomes data for palate surgery (uvulopalatopharyngoplasty [UPPP]) alone versus multilevel surgery through a national insurance claims database. Study Design Retrospective cohort study. Setting National insurance claims database. Subjects and Methods An adult cohort undergoing single-level UPPP versus UPPP with nasal and/or tongue/hypopharyngeal surgery was identified in the Truven Health Analytics MarketScan Research Databases for the years 2010 through 2012. Demographic and outcomes data were assessed at short-term (14 days), intermediate (15-60 days), and long-term (61-183 days) intervals via a multivariate regression model adjusted for age, sex, geographic region, insurance type, and the Charlson-Deyo comorbidity score. The primary long-term complication considered was positive airway pressure (PAP) equipment supply, implying possible treatment failure. Results The cohort included 14,633 patients: 7559 (51.6%), UPPP alone; 5219 (35.7%), UPPP + nasal surgery; 1164 (7.95%), UPPP + tongue/hypopharyngeal surgery; and 691 (4.7%), UPPP + nasal + tongue/hypopharyngeal surgery. Demographic data were similar among the groups. UPPP alone had lower rates of postoperative bleeding than UPPP + tongue/hypopharyngeal surgery (4.31% vs 6.19%, P = .004). Multivariate modeling indicated that the addition of either nasal surgery (odds ratio = 1.21, 95% CI = 1.10-1.34, P < .001) or tongue/hypopharyngeal surgery (odds ratio = 1.15, 95% CI = 1.00-1.32, P = .048) to UPPP was associated with increased odds of postoperative continuous positive airway pressure. Conclusions UPPP alone is currently the predominant form of sleep surgery in the United States. Multilevel surgery had greater odds of postoperative bleeding and positive airway pressure equipment supply than UPPP alone. Dedicated studies formally evaluating single- versus multilevel sleep surgery and the impact of possible surgeon/patient selection bias should be a priority.