Sleep Apnea and Perioperative Complications After Noncardiac Surgery

Abstract & Commentary

By David J. Pierson, MD, Professor, Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington, Seattle. Dr. Pierson reports no financial relationships relevant to this field of study.

This article originally appeared in the July issue of Critical Care Alert. At that time it was peer reviewed by William Thompson, MD, Associate Professor of Medicine, University of Washington, Seattle, and Leslie Hoffman, PhD, RN, Department of Acute/Tertiary Care, School of Nursing, University of Pittsburgh. Dr. Thompson and Dr. Hoffman report no financial relationships relevant to this field of study.

Synopsis: In a population-based study using hospital discharge diagnosis codes, patients with sleep apnea who underwent knee arthroplasty or open abdominal procedures were more likely to require invasive mechanical ventilation and to be diagnosed with aspiration pneumonia or ARDS than were matched patients without sleep apnea. Knee-replacement patients, but not those undergoing laparotomy, also were more likely to be diagnosed with pulmonary embolism.

Source: Memtsoudis S, et al. Perioperative pulmonary outcomes in patients with sleep apnea after noncardiac surgery. Anesth Analg 2011;112:113-121.

In this study of a large administrative database, the incidence of selected complications in patients diagnosed with sleep apnea (SA) was compared to that in patients undergoing similar surgical procedures who were not diagnosed with SA. The authors examined data from the National Inpatient Sample, a database sponsored by the Agency for Healthcare Research and Quality, for each year from 1998 through 2007, for patients who underwent total knee arthroplasty or an open abdominal surgical procedure. This yielded 117,283 patients coded as having SA and 5,934,420 patients without this diagnosis. A subset of the latter population was matched to those with SA using demographic variables via the propensity scoring method. The perioperative complications examined in these two groups were aspiration pneumonia, pulmonary embolism, the adult respiratory distress syndrome (ARDS), and the need for perioperative intubation and mechanical ventilation.

Of the 2,610,441 knee arthroplasties, 65,774 patients (2.52%) were coded for SA; the corresponding numbers for abdominal surgical procedures were 51,509 of 3,389,753 (1.40%). Patients with SA were more likely to be male and were five times more likely to be obese than patients without SA; additionally, their average comorbidity indices were higher, and they were more likely to be admitted on an emergent rather than an elective basis.

Aspiration pneumonia occurred more frequently in SA patients after both knee arthroplasty (1.18% vs 0.84%) and laparotomy (2.79% vs 2.05%). The same was found for ARDS (1.06% vs 0.45% for knee procedures; 3.79% vs 2.44% for abdominal procedures) and also for perioperative invasive mechanical ventilation (3.99% vs 0.79%; 10.8% vs 5.94%, respectively), with all these differences being statistically significant. Pulmonary embolism occurred more often after knee arthroplasty in patients with SA (0.51% vs 0.42%, P = 0.0038), but its incidence was not increased with SA following abdominal surgery. From these findings, the authors conclude that SA is an independent risk factor for perioperative pulmonary complications in noncardiac surgery.


This study adds incrementally to our appreciation of the importance of SA as a comorbidity in hospitalized patients and as a risk factor for perioperative complications. However, the study has some major limitations, the most important of which for the purposes of this newsletter are shortcomings of the administrative-database approach to studies of disease prevalence and complication incidence, and the article's broad generalizations about both the study population and the implications of the results.

Because of its design, this study grossly underestimates the prevalence of SA among the patients in the database. Only patients assigned one of two SA diagnosis codes were included, and the SA prevalence found (2.51% for patients undergoing knee arthroplasty and 1.49% for those who had laparotomies) is probably a 10-fold underestimation of the true prevalence. A recent review of the management of SA in surgical patients1 cites several epidemiologic studies indicating that sleep-disordered breathing occurs in approximately 20% of adults in this population, with nearly 7% exhibiting moderate-to-severe obstructive SA. For example, one recent study found that 22% of the adult general surgical population had obstructive SA.2 In that study, 70% of the SA patients were undiagnosed before presentation for perioperative evaluation; other sources estimate that as many as 80%-90% of surgical patients with SA are undiagnosed.1,2

The present study examined SA prevalence and associated complications in two specific populations of surgical patients: those undergoing total knee arthroplasty and those having an open abdominal procedure. The fact that the study included only these specific procedures is mentioned only once, in the methods section. However, throughout the paper the authors refer not to these operations but to "orthopedic procedures" and "general surgical procedures," and the title further broadens the population to "noncardiac surgery." The potential implications of the study's results as summarized in the abstract and elaborated in the discussion section of the paper should thus be interpreted cautiously to avoid unwarranted over-generalization.

These problems with the design of the study and the interpretation of its findings notwithstanding, the article emphasizes the fact that patients with SA are particularly predisposed to perioperative pulmonary problems. Most such patients have not been diagnosed, and they are increasingly encountered by intensivists following elective surgery or when admitted for other reasons. Complete evaluation for suspected SA is not feasible during acute hospitalization, but heightened awareness of this condition and its associated risks for a variety of complications during the perioperative period should aid in prevention, early diagnosis, and appropriate management.


1. Adesanya AO, et al. Perioperative management of obstructive sleep apnea. Chest 2010;138:1489-1498.

2. Finkel KJ, et al. Prevalence of undiagnosed obstructive sleep apnea among adult surgical patients in an academic medical center. Sleep Med 2009;10:753-758.