Stay up-to-date in pulmonary and critical care. No spam.
Obstructive Sleep Apnea and Perioperative Complications: A Review
Obstructive sleep apnea and other sleep disorders may affect as many as 70 million U.S. adults -- 1 in 4 men and 1 in 10 women. People with obstructive sleep apnea are usually obese, have other medical conditions, and are more likely to undergo surgery than people without sleep apnea.
When they do undergo surgery, available data suggest that people with known obstructive sleep apnea are at greater risk for postoperative complications than people without sleep apnea:
- Among 471 people undergoing noncardiac surgery, those with sleep study-proven obstructive sleep apnea (AHI > 5) had a 1 in 7 (14%) incidence of complications, compared to 2.6% among their matched controls who were generally obese but OSA-negative. Hypoxemia (12% vs 2%), respiratory failure (5% vs 2%), reintubation (1.4% vs 0.5%) and ICU transfer (7% vs 2%) were more common in people with known OSA. They also stayed in the hospital days longer on average.
- A large hospitalization database study showed that people with a sleep apnea diagnosis were 2-3 times more likely to have various postoperative complications, including 4% requiring intubation/mechanical ventilation compared to 0.8% of those without sleep apnea.
These retrospective observational studies only included patients with sleep apnea that caused symptoms (most likely, daytime somnolence prompting a sleep study and the final diagnosis). Only about 4% of U.S. men and 2% of women have sleep apnea with excessive daytime somnolence, and people with symptomatic obstructive sleep apnea are probably different from people with asymptomatic sleep apnea. Because the vast majority of people with asymptomatic sleep apnea are undiagnosed, essentially no data exists about their risk for postoperative complications (or cardiovascular outcomes in general) to guide management.
Given all this uncertainty, when a surgeon, internist, or pulmonologist needs to "clear" a patient prior to surgery, the diagnosis or suspicion of obstructive sleep apnea becomes a problem. The following are not recommendations, but a statement on this uncertainty and a digest of expert opinion.
People with Known OSA Undergoing Surgery
For people with known obstructive sleep apnea, the American Society of Anesthesiologists (despite a lack of data) have recommended prolonged observation postoperatively, along with routine CPAP in the post-operative care unit (PACU) to counter respiratory suppression from anesthesia, opioid analgesia, and the surgery itself. If implemented, these measures would impose bottlenecks into busy surgical centers, with financial and logistical costs perceived as prohibitive, and most centers are simply not following the guidelines, originally published in 2006.
People With Possible OSA Undergoing Surgery
For patients with possible obstructive sleep apnea being evaluated for the possible risks of an upcoming surgery, the best approach is even less clear. Although OSA treatment with CPAP improves surrogate outcomes, no good prospective trial has examined whether screening and treatment for OSA reduces perioperative complications. Sleep specialists have recommended the use of screening and preoperative treatment anyway, with tools such as the STOP-BANG OSA questionnaire (sensitivity 84%, specificity 56% at a score of 3+) or the Flemons OSA screening tool with a score greater than 15, with referral to a sleep specialist for an expedited sleep study and CPAP titration if positive.
However, they acknowledge that screening and testing everyone with potential obstructive sleep apnea with polysomography could impose costly burdens of delay and expense, with uncertain benefit. Expert authors have advised making these decisions on a case-by-case basis, balancing the urgency of the surgery with the potential risk to the patient from untreated sleep apnea. These experts being sleep specialists, naturally they also advise that a sleep specialist should be consulted in the process, if a sleep study is seen as indicated.
If a sleep study would impose too much delay before an upcoming surgery, and a patient is believed to have a high likelihood of obstructive sleep apnea with significant potential for postoperative complications, empiric CPAP can be started preoperatively. This can be accomplished using an auto-titrating CPAP machine.
A randomized trial of 138 patients at high risk for obstructive sleep apnea (but without a diagnosis yet) undergoing joint replacement surgery in Chest July 2013, showed no benefit from empiric auto-titrating CPAP in reducing postoperative complications, although the study was not well-powered.
People with obstructive sleep apnea are at elevated risk for perioperative complications, but the true degree of risk is unknown, and the absolute risk appears low for most patients (although good data is lacking). Those with known sleep apnea should wear their CPAP at their prescribed settings, or their oral appliance, while hospitalized for their surgery (and before, and after). In the immediate postoperative period, the ASA has recommended routine CPAP administration with prolonged observation in the PACU after surgery for patients previously prescribed CPAP. However, at this writing there are no studies or outcomes data to support the ASA's recommendation.
The few studies showing increased perioperative complications among people with obstructive sleep apnea are biased by including only people with known sleep apnea, whose apnea is on average more severe and may make them more risk-prone after surgery. It's unknown whether these risks apply in the much-larger population of people with undiagnosed sleep apnea, which on average is milder.
For that reason, the proper threshold of suspicion to test and treat sleep apnea prior to surgery is likewise unclear, including whether delay of surgery is appropriate to allow for testing and treatment of sleep apnea, or for how long someone with newly diagnosed sleep apnea should be treated with CPAP or another modality prior to undergoing surgery.
Clear as mud, right? (Limitations in getting good-quality data on outcomes in obstructive sleep apnea in general include the inconvenience and cost of polysomnography -- about $3,000 -- and CPAP adherence rates of <50%.)
Sleep apnea and obesity are both so common in the general population that universal screening for sleep apnea before all surgeries requiring general anesthesia and delaying surgeries to get sleep studies and start treatment on all patients at high risk would represent a large logistical barrier to the efficiency of surgery centers. Would the reduction in risk be worth imposing this bottleneck, for an unproven benefit?
Some say no, and that's why even in its somewhat conservative guideline, the ASA only went so far as to recommend that "[a]nesthesiologists should work
with surgeons to develop a protocol whereby patients in whom the possibility of OSA is suspected ... are evaluated long enough before the day of surgery to allow preparation of a perioperative management plan."
An alternative, pragmatic approach would be to create a protocolized system in the PACU whereby CPAP would be quickly administered to all patients with any hypoxemia (especially those who are overweight or obese) and maintained throughout their hospital stay and at home until a sleep study is obtained. Nurses and respiratory therapists would be trained in the recognition and recording of apneas occurring during post-operative sleep or sedation; RTs could titrate CPAP to absence of snoring or normoxemia, and these initial settings could be used indefinitely in lieu of an outpatient sleep study, reducing costs.
Of course, all obese patients should be advised to lose weight prior to surgery; weight loss can improve obstructive sleep apnea somewhat.
For those physicians who elect to screen patients for sleep apnea prior to surgery, consider using the STOP-Bang or Flemons screening tool. There's even an online STOP-Bang calculator patients can answer themselves (but tell the Americans that 40 cm is 16 inches). Order a sleep study on anyone at high risk (STOP-Bang 3+ or Flemons 15+). And set expectations with your referring surgery colleagues that many elective surgeries may be delayed, pending the sleep apnea evaluation and initial CPAP treatment period (some experts recommend only 1 week of OSA treatment preoperatively is needed).
Memtsoudis SG et al. A rude awakening--the perioperative sleep apnea epidemic. N Engl J Med. 2013 Jun 20;368(25):2352-3.
Practice guidelines for the perioperative management of patients with obstructive sleep apnea: a report by the American Society of Anesthesiologists Task Force on Perioperative Management of patients with obstructive sleep apnea. Anesthesiology 2006 May;104(5):1081-93.
Babak Mokhlesi. Empiric Postoperative Autotitrating Positive Airway Pressure Therapy: Generating Evidence in the Perioperative Care of Patients at Risk for Obstructive Sleep Apnea. Chest 2013;144(1):5-7.