Postoperative Complications in Patients With Obstructive Sleep Apnea

Background: Unrecognized obstructive sleep apnea (OSA) is associated with unfavorable perioperative outcomes among patients undergoing noncardiac surgery (NCS). Methods: The study population was chosen from 39,771 patients who underwent internal medicine preoperative assessment between January 2002 and December 2006. Patients undergoing NCS within 3 years of polysomnography (PSG) were considered for the study, whereas those <18 years of age, with a history of upper airway surgery, or who had had minor surgery under local or regional anesthesia were excluded. Patients with an apnea-hypopnea index (Au) >= 5 were defined as OSA and those with an AHI <5 as control subjects. For adjusting baseline differences in age, sex, race, BMI, type of anesthesia, American Society of Anesthesiology class, and medical comorbidities, the patients were classified into five quintiles according to a propensity score. Results: Out of a total of 1,759 patients who underwent both PSG and NCS, 471 met the study criteria. Of these, 282 patients had OSA, and the remaining 189 served as control subjects. The presence of OSA was associated with a higher incidence of postoperative hypoxemia (OR, 7.9; P = .009), overall complications (OR, 6.9; P = .003), and ICU transfer (OR, 4.43; P = .069), and a longer hospital length of stay (LOS), (OR, 1.65; P = .049). Neither an AHI nor use of continuous positive airway pressure at home before surgery was associated with postoperative complications (P = .3 and P = .75, respectively) or LOS (P = .97 and P = .21, respectively). Conclusions: Patients with OSA are at higher risk of postoperative hypoxemia, ICU transfers, and longer hospital stay. CHEST 2012; 141(2):436-441