Self-reported remission of obstructive sleep apnea following bariatric surgery: cohort study.

BACKGROUND: Evidence on remission of obstructive sleep apnea (OSA) after bariatric surgery and its relation to weight loss is conflicting. We sought to identify factors associated with successful self-reported OSA remission in a large cohort of bariatric surgery patients.

METHODS: We analyzed data from the statewide, prospective clinical registry of the Michigan Bariatric Surgery Collaborative and identified 3,550 patients with OSA who underwent a primary bariatric procedure between June 2006 and October 2011 and had at least 1 year of follow-up data. We used multivariate logistic regression to identify preoperative factors associated with successful self-reported OSA remission, defined as discontinuation of continuous positive airway pressure or bilevel positive airway pressure at 1 year. Our regression model also included procedure type and weight loss at 1 year, divided into equal quintiles, as covariates.

RESULTS: The overall 1-year self-reported OSA remission rate was 60%. Significant predictors of remission included age category (per 10 yr) (OR .73, CI .69-.78), body mass index category (per 10 units) (OR .57, CI .54-.62), male gender (OR .58, CI .52-.69), hypertension (OR .83, CI .74-.99), depression (OR .78, CI .69-.88), pulmonary disease (OR .88, CI .78-.98), and baseline Health and Activities Limitations Index score (OR 1.70, CI 1.32-2.23). Relative to gastric banding, the adjusted odds of OSA remission were greater with gastric bypass (OR 2.38, CI 1.89-3.08), sleeve gastrectomy (OR 2.01, CI 1.44-2.55), and duodenal switch (OR 2.57, CI 1.02-7.26). The odds ratio of OSA remission increased stepwise through quintiles of 1-year weight loss. Relative to the lowest quintile, the odds ratios of remission in the 2(nd) through 5(th) quintiles were 1.44 (CI 1.11-1.84), 2.03 (CI 1.48-2.57), 2.47 (1.85-3.40), and 3.53 (CI 2.56-4.85).

CONCLUSIONS: Weight loss is an important predictor of self-reported OSA remission after bariatric surgery. However, independent of weight loss, there remain significant differences in the likelihood of remission between gastric banding and other bariatric procedures. This suggests that there may be metabolic, weight-independent effects of procedure type on self-reported OSA remission.