Surgical risk and the preoperative evaluation and management of adults with obstructive sleep apnea

INTRODUCTION — Obstructive sleep apnea (OSA) is increasing in prevalence [1-3], and the number of perioperative patients with OSA is likely to continue to rise in parallel with the increase in obesity [1,4,5]. OSA increases the risks of perioperative complications and should be suspected, recognized, and managed in the perioperative period to minimize postoperative morbidity and mortality [6].

The preoperative evaluation and management of patients with known or suspected OSA are reviewed here. The intraoperative and postoperative management of patients with OSA and preoperative evaluation of patients with obesity are discussed separately. (See "Intraoperative management of adults with obstructive sleep apnea" and "Postoperative management of adults with obstructive sleep apnea" and "Preanesthesia medical evaluation of the patient with obesity".)

EPIDEMIOLOGY — The prevalence of OSA in surgical patients is unclear. Using International Classification of Disease-9 codes, the prevalence has been estimated to be as high as 6 to 17 percent [3,7-9]. However, rates may be as high as 68 to 70 percent in high-risk populations such as patients undergoing bariatric or other major noncardiac surgery [10-12].

Many patients with OSA are undiagnosed at the time of surgery [12-14]. As examples:

●In one 2012 study, over 60 percent of surgical patients with moderate to severe OSA confirmed by polysomnography were not identified as having OSA during preoperative clinical evaluation by surgeons or anesthesiologists [13].

●In another study, adult surgical patients with at least one risk factor for postoperative cardiovascular events and no preexisting OSA diagnosis were assessed preoperatively with type 3 diagnostic portable sleep apnea testing devices [12]. Sixty-eight percent of patients were discovered to have OSA, among whom 31 percent had at least moderate OSA and 11 percent had severe OSA. (See "Home sleep apnea testing for obstructive sleep apnea in adults", section on 'Type 3 devices (portable devices)'.)

Similarly, the diagnosis of obesity hypoventilation syndrome (OHS) is frequently unknown at the time of surgery [15]. Consequently, a high level of suspicion for OSA and OHS should be maintained, particularly in high-risk patients. (See "Clinical manifestations and diagnosis of obesity hypoventilation syndrome", section on 'Risk factors'.)

PERIOPERATIVE COMPLICATIONS — Patients with OSA have a two- to fourfold higher risk for perioperative complications compared with patients without OSA [6-8,12,15-29]. Respiratory complications (oxygen desaturation, respiratory failure) are the most common, followed by cardiovascular complications [12,19]. Others include difficulty with airway management and postoperative delirium, all leading to higher resource utilization.

Risk factors for perioperative complications — The invasiveness of the surgical procedure is the most important determinant of perioperative risk in patients with OSA. Severity of OSA may also affect perioperative risk.

●Type of surgery – Major surgery, airway surgery, and surgery that requires general anesthesia are associated with multiple factors that exacerbate airway obstruction and increase the risk of perioperative complications in these patients [30,31]. These factors are shown in a table (table 1), and are discussed separately. (See "The effects of medications on sleep quality and sleep architecture" and "Obstructive sleep apnea: Overview of management in adults", section on 'Nonsupine sleep position' and "Postoperative management of adults with obstructive sleep apnea", section on 'Upright positioning' and "Postoperative management of adults with obstructive sleep apnea", section on 'Floor management' and "Pathophysiology of upper airway obstruction in obstructive sleep apnea in adults", section on 'Rostral fluid displacement' and "Intraoperative management of adults with obstructive sleep apnea", section on 'Intravenous fluid management'.)

●Severity of OSA – The literature on the correlation between severity of OSA and perioperative risk is conflicting. Examples include the following:

•In a prospective international cohort study of at-risk adults undergoing major noncardiac surgery, unrecognized severe OSA (but not mild or moderate OSA) was associated with increased risk of 30-day postoperative cardiovascular complications [12].

•Similarly, one large retrospective cohort study of surgical patients compared perioperative complications in patients with a known diagnosis of OSA, patients who were diagnosed with OSA preoperatively, and controls without OSA [32]. Cardiac and respiratory complications were twice as likely in both groups of patients with OSA as in controls, and risk increased with increasing severity of OSA.

•In contrast, two other retrospective population-based studies did not find an association between severity of OSA and perioperative complications [33,34].

•A 2022 systematic review of the literature found four observational studies that assessed the effect of mild, moderate, or severe OSA on postoperative cardiopulmonary complications [29]. Pooled analysis found an association between severe (but not mild or moderate) OSA and pulmonary complications with an odds ratio (OR) of 2.34 (95% CI 1.42-3.87), and no association between any level of OSA and cardiac complications.

Other risk factors include the following:

●Patients in whom OSA is newly diagnosed at the time of surgery may be at increased risk of perioperative complications, compared with patients with known OSA [35].

●The need for postoperative opioids is an important risk factor for oxygen desaturation and apnea in patients with OSA. (See "Postoperative management of adults with obstructive sleep apnea", section on 'Pain control'.)

●OSA-related comorbidities (eg, obesity [36], systemic hypertension, pulmonary hypertension, cardiac arrhythmias, coronary artery disease, and heart failure) may also increase perioperative risk. (See "Obstructive sleep apnea and cardiovascular disease in adults".)

●Compared with patients with OSA alone, patients with obesity hypoventilation syndrome (OHS; characterized by persistent daytime hypercapnia with PaCO₂ >45 mmHg) and patients with overlap syndrome (chronic obstructive pulmonary disease [COPD] plus OSA) have an additional increased risk of pulmonary and cardiac complications, intensive care transfer, and length of stay [15,22]. (See "Sleep-related breathing disorders in COPD".)

Respiratory complications — Multiple studies have reported increased rates of pulmonary complications in patients with OSA, including worsening of OSA, acute respiratory failure requiring use of noninvasive ventilation or emergent intubation, complicated post-extubation course, postobstructive pulmonary edema (from breathing against an obstructed upper airway), respiratory arrest, acute respiratory distress syndrome (ARDS), and oxyhemoglobin desaturation [7,16-19,30,37-42].

●One qualitative systematic review of surgical patients with OSA found that 9 of the 15 included studies reported an increased rate of pulmonary complications (OR ranging from 1.07 to 4.4) including oxygen desaturation, respiratory failure, ARDS, pneumonia, and difficult reintubation [19].

●A large prospective study reported that OSA was associated with unplanned tracheal intubation or postoperative invasive or noninvasive ventilation (adjusted hazard ratio [HR] 6.16 [95% CI 2.51-15.16] for severe OSA; 6.26 [95% CI 2.85-13.75] for moderate OSA; 2.28 [95% CI 1.04-5.03] for mild OSA) [12].

●In a database study of over 6 million surgical procedures [17], patients with OSA undergoing general surgery had higher rates of intubation/mechanical ventilation (11 versus 6 percent), aspiration pneumonia (3 versus 2 percent), and ARDS (4 versus 2 percent), compared with patients without OSA. Patients with OSA having orthopedic surgery had a similar increase in risk of complications, although at lower absolute rates, suggesting that risk varies with the type of surgery.

●A meta-analysis of 19 studies found that OSA was associated with increased risks of respiratory complications postoperatively (OR ranging from 1.54 to 2.36) [29].

Cardiovascular complications — OSA increases the risk of perioperative cardiovascular complications. A number of reports including one large prospective study, 2012 and 2020 meta-analyses, and a large qualitative-systematic review report increased rates of cardiovascular complications (OR ranging from 1.06 to 2.6) in patients with OSA, including large blood pressure fluctuations, atrial fibrillation, myocardial infarction, cardiac arrest, congestive heart failure, stroke, thromboembolism and shock [6,7,12,16,18,19,32,43-45]. However, there is significant heterogeneity in the characteristics and findings of these reports [44].

Increased cardiovascular risk in OSA patients was illustrated by a prospective study of 1218 patients 45 years or older without a known diagnosis of OSA who had at least one cardiovascular risk factor and were assessed with a portable sleep study prior to major noncardiac surgery (eg, general, orthopedic, or vascular surgery) [12]. Unrecognized OSA was discovered in two-thirds of patients, and about 1 in 9 had severe sleep apnea. Severe OSA was found to be an independent risk factor for postoperative cardiovascular events. Compared with patients who did not have OSA, patients with OSA had a higher risk of a composite outcome of myocardial injury, cardiac death, congestive heart failure, thromboembolism, new atrial fibrillation, and stroke within 30 days of surgery (adjusted HR 1.49, 95% CI 1.19-2.01; however, the association was only significant in those with severe OSA (adjusted HR 2.23, 95% CI 1.49-3.34). Whether interventions can affect this risk remains unknown. In addition, the mean cumulative duration of oxyhemoglobin desaturation less than 80 percent during the first three postoperative nights for patients with cardiovascular complications) was longer than for patients with no cardiovascular complications (23.1 versus 10.2 minutes). In a post hoc analysis of preoperative nocturnal pulse oximetry in the same patient cohort, independent predictors of cardiovascular events included oxygen desaturation index (ODI) ≥30 events per hour (adjusted hazard ratio [aHR] 1.63 [95% CI 1.1-2.5]), and cumulative time spent during sleep with oxygen saturation below 80% (CT80) ≥10 minutes (aHR 1.79 [95% CI 1.3-2.5]) [46].

Several cohort studies comparing treated versus untreated OSA patients also reported that untreated patients have an increased risk of postoperative cardiac arrest and shock (OR 2.2) [32] and myocardial infarction (OR 2.6) [6]. (See "Obstructive sleep apnea and cardiovascular disease in adults".)

Mortality — The association between OSA and perioperative mortality is unclear [19], with various studies showing increased [12,23] or comparable [14] mortality. While decreased mortality [7,18,45] has been reported in retrospective studies, this may be due to unrecognized or undiagnosed OSA in the control or non-OSA groups, and/or the possibility that patients with diagnosed OSA received monitoring and treatment of their OSA.

Others — Other adverse perioperative events associated with OSA include acute renal failure (OR 2.43), wound hematomas or seromas (OR 1.36), postoperative intensive care unit transfer (OR 2.81), and longer hospital stay (6.8 versus 5.1 days) [16,23,47].

Data on the possible association between OSA and postoperative delirium are conflicting [48-50]. All of these possible adverse events lead to greater resource utilization. The pathophysiology of delirium in patients with OSA could involve compromised regional cerebral blood flow during airway obstruction, oxidative stress from hypoxia, sleep disruption, surgical inflammation, increased cortisol levels, perioperative medication, and postoperative pain.

Patients with OSA are also more likely to present with difficulty in airway management for anesthesiologists. In a meta-analysis of 16 observational studies of surgical patients, OSA was associated with a fourfold increase in the incidence of difficult intubation or difficult mask ventilation during anesthesia [51]. In a surgical cohort study of patients who underwent major surgery, moderate to severe OSA was associated with a three- to fourfold increased risk of difficult intubation [52]. (See "Intraoperative management of adults with obstructive sleep apnea".)

INITIAL ASSESSMENT — We suggest making OSA screening part of standard preanesthetic evaluation, using a sleep-focused history and physical examination with findings processed via screening questionnaires. The objective of preoperative assessment is to identify patients with known OSA or at high risk for OSA before surgery to allow targeted perioperative interventions and implementation of precautions that may help to reduce perioperative complications. When feasible, initial preoperative assessment for patients with known or suspected OSA should occur far enough in advance of surgery to allow optimization of therapy. Our assessment strategy is consistent with guidelines and practice statements from the American Society of Anesthesiologists, the American Academy of Sleep Medicine, the Society of Anesthesia and Sleep Medicine, and others [30,31,53-55].

Preoperative evaluation of healthy and patients with obesity are discussed separately. (See "Preoperative medical evaluation of the healthy adult patient" and "Preanesthesia medical evaluation of the patient with obesity".)

Patients with known OSA — Preoperative evaluation of patients with OSA should specifically include assessment of OSA severity and adequacy of management [30,31,54-56]. (See 'Proceeding with further evaluation or surgery' below.)

●Severity of the OSA – Documentation should include current symptoms and signs of OSA and recent sleep studies (on and off therapy). (See "Clinical presentation and diagnosis of obstructive sleep apnea in adults", section on 'Classification of severity'.)

●Documentation of current treatment – OSA therapy may include positive airway pressure (PAP) therapy (eg, continuous PAP [CPAP] or bi-level [BPAP]), oral appliances, or less commonly, hypoglossal nerve stimulation device. The settings and specifics of existing OSA therapy should be documented, and patients should be instructed to bring their PAP equipment or oral appliances on the day of surgery, to facilitate reinstitution of therapy postoperatively. (See "Postoperative management of adults with obstructive sleep apnea", section on 'Positive airway pressure therapy'.)

●Adequacy of current treatment – Patients who are using therapy should be asked about residual symptoms, problems with adherence with treatment or technical difficulties, significant changes in weight since therapy was initiated, and reasons for discontinuation if therapy is no longer used. Adherence to treatment can also be confirmed using downloads from the patient's device, if feasible. (See "Evaluation and management of residual excessive sleepiness in adults with obstructive sleep apnea" and "Assessing and managing nonadherence with continuous positive airway pressure (CPAP) for adults with obstructive sleep apnea" and "Downloading data from positive airway pressure devices in adults".)

Patients without a known diagnosis — OSA can be suspected on the basis of clinical history and physical examination, or results of preoperative screening, starting with questionnaires.

Clinical evaluation — Patients considered at increased risk of OSA include those with excessive daytime sleepiness and at least two of the following: habitual loud snoring; witnessed apnea, gasping, or choking; or diagnosed hypertension [57]. However, use of these parameters is unable to stratify the risk as high or low, compared with the use of a screening questionnaire. Other clinical features of OSA are shown in a table (table 2). (See "Quantifying sleepiness" and "Clinical presentation and diagnosis of obstructive sleep apnea in adults", section on 'Diagnostic evaluation'.)

Screening with a questionnaire — The gold standard for the diagnosis of OSA is the polysomnogram (PSG) (see "Polysomnography in the evaluation of sleep-disordered breathing in adults"). However, PSG is impractical as a routine preoperative screening tool. A number of questionnaires have been developed to identify high-risk patients more simply.

Whom to screen — Given the increasing prevalence of OSA, the ease of screening with a questionnaire, and increased perioperative risks of OSA, we agree with the Society of Anesthesia and Sleep Medicine guidelines that recommend screening all preoperative patients for OSA using clinical parameters and screening questionnaires [55]. However, there is a lack of data to support improved outcomes with this practice, and screening all patients confers an operational and economic burden. Therefore, some experts screen only patients at high risk of OSA, including the following:

●Patients with obesity (body mass index [BMI] ≥35 kg/m² (calculator 1)), especially those scheduled for bariatric surgery [33,58].

OSA may be present in up to 94 percent of bariatric surgery patients [59].

●Patients with medical conditions in which OSA is prevalent including hypertension, atrial fibrillation, type 2 diabetes, uncontrolled hypothyroidism, congestive heart failure, and stroke.

●Patients with a history of difficult intubation or upper airway characteristics that predict a difficult intubation (table 3) [38,39,60].

In a study that included 137 adults undergoing evaluation for OSA, the Mallampati score (figure 1) was independently associated with an increased risk of OSA [61]. For every one-point increase in the Mallampati score, the odds of having OSA increased more than twofold. (See "Clinical presentation and diagnosis of obstructive sleep apnea in adults".)

Screening questionnaires — There are multiple screening tools available to help identify individuals who may have OSA. The sensitivity of existing OSA screening tools is generally higher than their specificity, which means that the false positive rate is high [62-64]. When the score is high, patients do not always have OSA, but when the score is low, patients are unlikely to have OSA. Thus, questionnaires are most helpful when they are negative (ie, low score), and patients with positive screening need further evaluation to diagnose OSA. Preoperative screening is discussed here. Screening in the general population is discussed separately. (See "Clinical presentation and diagnosis of obstructive sleep apnea in adults", section on 'Screening questionnaires'.)

Screening tools used in the preoperative setting include the following:

●STOP-Bang questionnaire – The STOP-Bang questionnaire is the best validated tool for preoperative screening for OSA, and is the screening tool that we use [65-67]. The STOP-Bang questionnaire requires "yes," or "no," responses to eight questions about snoring, tiredness, observed apneas, and blood pressure; BMI >35kg/m², age >50 years, neck circumference >40 cm, and male gender (biologic sex), assigned one point each (table 4) [63,68-73]. Patients with zero to two positive responses are considered "low risk," those with three or four are at "intermediate risk," and those with ≥5 positive responses are at "high risk" of having OSA [69,71,72].

The predictive value of STOP-Bang screening has been studied in surgical patients. A meta-analysis of three studies in surgical patients reported that the probability of moderate to severe OSA (confirmed by polysomnography) increased from 15 percent at a STOP-Bang score of 3, to 65 percent at a STOP-Bang score of 7 or 8 [74]. A STOP-Bang score of ≤2 effectively excludes a diagnosis of OSA, and a score of ≥5 makes OSA highly likely (specificity approximately 80 percent).

The finding of an elevated serum bicarbonate may increase the specificity of intermediate STOP-Bang scores (ie, 3 or 4). We do not routinely order a serum bicarbonate solely for the diagnosis of OSA, but bicarbonate is often measured preoperatively for patients with significant comorbidities or who undergo major surgery. In a prospective study of 384 surgical patients who were screened for OSA, the addition of a serum bicarbonate level ≥28 mmol/L to a STOP-Bang score ≥3 increased the specificity for predicting moderate to severe OSA from 37 to 85 percent [75].

Although the STOP-Bang questionnaire gives all items equal weight, not all items have equal predictive value for OSA. A two-step scoring system may improve performance, though it adds complexity (table 5) (calculator 2) [76]. The first step is completion of the STOP-Bang score, and designation of patients at low (≤2), intermediate (3 or 4), or high (≥5) risk. The second step is applied to intermediate-risk patients (ie, score 3 or 4), and involves analysis of the specific items that have been scored positive. Patients who answer yes to two or more of the STOP (snoring, tiredness, observed apnea, high blood pressure) questions, and positive responses to BMI >35 kg/m², male gender (biologic sex), or neck circumference >40 cm (16 in) are designated as high risk of OSA.

BMI thresholds of >35 kg/m² are suitable for white individuals. Lower BMI cutoffs (>27.5 kg/m²) for Chinese and Indian populations have been shown to increase sensitivity without compromising the predictive ability for moderate-to-severe OSA [65,77].

An intermediate- to high-risk STOP-Bang score may predict increased risk of postoperative pulmonary and cardiac complications [12,35,78].

•A 2017 meta-analysis of 10 prospective and retrospective studies including approximately 23,600 patients who underwent a variety of surgical procedures reported a fourfold higher risk of significant postoperative complications and longer hospital stay (on average two days) in patients with intermediate- to high-risk STOP-Bang scores (≥3) than in patients with low-risk scores (0 to 2) [78].

•In a 2022 meta-analysis of cohort studies of patients who were screened preoperatively for OSA, a STOP-Bang score of ≥5 was associated with a fivefold higher risk of cardiovascular complications, and a twofold increased risk of respiratory complications compared with STOP-Bang scores <5 [79]. The quality of evidence for all outcomes was very low.

•In a prospective study of patients undergoing noncardiac surgery without a known diagnosis of OSA who underwent a preoperative portable sleep study, the STOP-Bang questionnaire result was high risk in 26 percent of patients, moderate risk in 53 percent, and low risk in 21 percent [12]. A high-risk score of 5 to 8 was associated with an increased cardiovascular risk following surgery (hazard ratio [HR] 1.68, 95% CI 1.11-2.54) and intensive care unit (ICU) readmission, while moderate-risk score of 3 or 4) was associated with an increased risk of ICU readmission and wound infection.

•In a prospective international study of difficulty with airway management in 869 patients without prior diagnosis of OSA who had general anesthesia for noncardiac surgery, patients were assessed preoperatively with STOP-Bang scores, bedside airway examination, and portable sleep monitoring devices for OSA [52]. A STOP-Bang score of ≥3 was associated with a three- to fourfold increased odds of difficult intubation relative to STOP-Bang score of 0 to 2 [52].

●Other screening tools – Other screening tools that can be used in the surgical population:

•Berlin Questionnaire – This uses questions on snoring, excessive daytime sleepiness, sleepiness while driving, apneas during sleep, hypertension, and BMI to stratify patients as having a high or low risk for OSA (table 6 and table 7) [64]. It identifies preoperative patients with OSA (apnea-hypopnea index [AHI] >5/hour) with a sensitivity of 69 percent and a specificity of 56 percent, and those with severe OSA (AHI >30/hour) with a sensitivity of 87 percent and a specificity of 46 percent [62]. In the meta-analysis of cohort studies of patients who were screened for OSA preoperatively, a Berlin score of ≥2 (high risk) versus <2 was not associated with increased postoperative complications, ICU admission, or mortality [79]. The quality of the data was very low.

•Others – Other tools (eg, sleep apnea clinical score [SACs], NoSAS, multivariable apnea prediction [MVAP]) are described separately. The Epworth Sleepiness Scale is neither sensitive nor specific for screening for moderate to severe OSA [80,81]. (See "Clinical presentation and diagnosis of obstructive sleep apnea in adults", section on 'Screening questionnaires'.)

Associated conditions — Preoperative evaluation should include assessment of OSA-associated conditions that may require preoperative optimization, affect the decision to operate, or change the management of anesthesia. These include:

●Difficulty with airway management – OSA itself and a number of associated patient characteristics (eg, obesity, large neck circumference, snoring) are risk factors for difficulty with airway management for anesthesia (table 8 and table 9 and table 3), and for airway complications after extubation (table 10). Assessment and management of the difficult airway for anesthesia are discussed separately. (See "Management of the difficult airway for general anesthesia in adults", section on 'Recognition of the difficult airway' and "Intraoperative management of adults with obstructive sleep apnea", section on 'Airway management'.)

●Comorbidities – Patients with obesity may require specific preoperative evaluation for comorbidities that are more prevalent in this population, including hypertension, heart disease, diabetes, metabolic syndrome, and kidney disease. This evaluation is discussed separately. (See "Evaluation of cardiac risk prior to noncardiac surgery", section on 'Initial evaluation' and "Preanesthesia medical evaluation of the patient with obesity" and "Preoperative evaluation for anesthesia for noncardiac surgery".)

●Obesity hypoventilation syndrome (OHS) – OHS is defined as awake alveolar hypoventilation (PaCO₂ >45 mm Hg) in an individual with obesity which cannot be attributed to other conditions associated with alveolar hypoventilation. (See "Clinical manifestations and diagnosis of obesity hypoventilation syndrome".)

OHS occurs in 11 percent of patients with known OSA; in turn, 90 percent of patients with OHS have OSA [82]. The presence of concurrent OHS further increases the perioperative risk for patients with OSA and changes the preoperative management (testing and PAP therapy) as well as the management of anesthesia (airway, ventilation, and sedation) [22].

Thus, in patients with obesity and OSA, it is appropriate to measure a serum bicarbonate; a level >28 mmol/L may suggest OHS and should prompt measurement of arterial blood gases, along with further evaluation for the etiology of hypoventilation. [82]. (See "Clinical manifestations and diagnosis of obesity hypoventilation syndrome" and "Preanesthesia medical evaluation of the patient with obesity", section on 'Obesity hypoventilation syndrome'.)

●Overlap syndrome – In patients with known or suspected OSA, the presence of overlap syndrome (ie, chronic obstructive lung disease [COPD] plus OSA) also further increases the perioperative risk [22]. The suspicion for COPD should be raised in a smoker with dyspnea and chronic cough, and the clinician should obtain evidence from previous imaging or pulmonary function tests, if available, to confirm the diagnosis.

●Pulmonary hypertension (PH) – PH may complicate OSA (and OHS [82], and overlap syndrome) [83]. Because the diagnosis of PH impacts perioperative management and prognosis, a preoperative echocardiogram should be performed when there is a high index of suspicion for severe pulmonary hypertension, and either signs/symptoms of right heart dysfunction, sleep-related hypoxemia despite PAP, or severe obesity. (See "Obstructive sleep apnea and cardiovascular disease in adults", section on 'Pulmonary hypertension' and "Treatment and prognosis of pulmonary arterial hypertension in adults (group 1)".)

●Gastroesophageal reflux disease (GERD) – GERD may be more common in patients with OSA; it is not clear whether this is an independent association or due to common risk factors (eg, obesity) [84-86]. While this diagnosis may be elicited on history, it is not routine to perform formal testing preoperatively in patients with OSA. (See "Clinical manifestations and diagnosis of gastroesophageal reflux in adults".)

PROCEEDING WITH FURTHER EVALUATION OR SURGERY — The majority of patients with known or suspected OSA may proceed to surgery without additional testing or treatment for OSA. However, select groups may need to delay surgery so that a diagnosis can be made, therapy started, or therapy adjusted before proceeding with surgery. In those who need further testing or optimization of therapy, we prefer to schedule surgery no sooner than one week after changes are made, to receive the benefit of treatment and in particular to accustom the patient to the positive pressure airway (PAP) device or settings (in the event it is new). This time frame may be modified by the urgency of the surgery and preferences of the patient and surgeon. Patients who are on PAP therapy for OSA should continue treatment up to the day of surgery and should preferably bring their PAP device (or other OSA treatment device, such as a mandibular repositioning device) with them on the day of surgery.

Factors affecting the decision — The decision to defer surgery for further evaluation and/or treatment must be individualized and based upon a number of factors including whether patients are treated or untreated, the suspected severity of OSA, comorbidity burden, and the risk level of surgery. The threshold to defer surgery is higher in patients undergoing low-risk procedures including noninvasive surgery, endoscopy, arthroscopy, and cataract surgery, and surgeries that are not likely to require significant intra-or post-procedural opioids. In contrast, we have a lower threshold to defer surgery in patients with uncontrolled OSA undergoing high-risk procedures including major invasive surgery that impacts airway or cardiopulmonary function and surgeries requiring substantial postoperative opioid medication. Other factors that affect the decision include variations in institutional practice, level of surgeon support for completion of sleep testing prior to surgery, the availability of preoperative evaluation clinics and sleep study evaluation, monitoring capabilities of the facility where the surgery is to be performed, likelihood of PAP use in the postoperative period, and clinician and patient preference. The risks and benefits should be discussed with the patient.

Regardless of whether surgery is deferred for further evaluation, all patients should be managed with intraoperative and postoperative risk mitigation measures and precautions. (See "Intraoperative management of adults with obstructive sleep apnea".)

Sleep studies — If further evaluation for OSA is deemed necessary, options include in-laboratory or at home sleep apnea testing (HSAT). These options are discussed in detail separately. (See "Clinical presentation and diagnosis of obstructive sleep apnea in adults", section on 'Selecting home or in-laboratory testing'.)

Data describing HSAT as a preoperative tool for the diagnosis of OSA are limited. (See "Home sleep apnea testing for obstructive sleep apnea in adults".)

●In a prospective observational study of 475 adult patients scheduled for inpatient surgery, nocturnal oximetry and oxygen desaturation index (ODI) ≥15 events per hour was able to predict moderate-to-severe OSA with high sensitivity and specificity [87].

●In a post-hoc analysis of a multicenter prospective study of over 1200 surgical patients who were evaluated preoperatively with overnight pulse oximetry and HSAT, the sensitivity and specificity of ODI ≥15 events per hour for predicting moderate to severe OSA were 88.4 (95% CI: 85.7-90.6) and 95.4 percent (95%CI, 94.2-96.4), respectively [88]. The area under the receiver operator curve for moderate and severe OSA was 0.983 (95% CI, 0.977-0.988) and 0.979 (95% CI, 0.97-0.909) respectively. In a separate post hoc analysis of pulse oximetry data in the same cohort, oxygen desaturation index (ODI) ≥30 events per hour and cumulative time spent during sleep with oxygen saturation below 80% (CT80) ≥10 min predicted 30 day postoperative cardiovascular events [46]. (See 'Cardiovascular complications' above.)

Elective surgery

Well-controlled OSA — Patients with well-controlled OSA are those who are asymptomatic and adherent with therapy who have not experienced ≥ 10 percent weight gain since their last sleep evaluation. In this population, no further evaluation prior to surgery is required regardless of whether patients are undergoing low-risk or high-risk surgical procedures.

Presumably these patients are already on PAP and will likely benefit from continued PAP use. Data to support this observation are derived from retrospective studies that demonstrate a reduction in complications in those OSA patients who are treated with PAP compared with untreated OSA. As an example, a prospective cohort study of 26,842 surgical patients, of whom approximately 10 percent had OSA, reported that untreated OSA was associated with an increase in cardiopulmonary complications, compared with patients with OSA treated with PAP therapy (adjusted odds ratio [OR] 1.8) [6]. Indirect evidence in nonsurgical OSA patients also suggests that PAP therapy for more than two weeks improves physiologic outcomes (eg, hypertension, oxygen saturation, and apnea-hypopnea index [AHI]) [89].

Uncontrolled or suspected OSA — The preoperative management of patients with uncontrolled known OSA (ie, patients who are symptomatic despite therapy, nonadherent or intolerant of therapy) and patients with suspected OSA is controversial. There is considerable variation in practice and limited guidance on how best to approach these patients. We suggest the following:

●Patients WITH evidence of an associated significant or uncontrolled systemic disease or additional problems with ventilation or gas exchange – We agree with guidelines that advocate that in patients with partially treated/untreated OSA, surgery should be deferred for evaluation and treatment when there is evidence of an associated significant or uncontrolled systemic disease or additional problems with ventilation or gas exchange such as [55]:

•Hypoventilation syndrome

•Severe pulmonary hypertension

•Resting hypoxemia in the absence of other cardiopulmonary disease

In such cases, evaluation for further management in conjunction with a sleep specialist, if possible, is advised. (See "Obstructive sleep apnea: Overview of management in adults".)

●Patients WITHOUT evidence of an associated significant or uncontrolled systemic disease or additional problems with ventilation or gas exchange – In this population, deferring surgery for evaluation can be determined by the risk of surgery, although practice varies considerably:

•High-risk procedures – For patients with uncontrolled or suspected OSA undergoing high-risk procedures (eg, major invasive surgery that impacts airway or cardiopulmonary function and/or requires substantial postoperative opioid medication), experts disagree on whether elective surgery should be deferred. We agree with those who prefer that surgery be deferred until therapy has been optimized, if feasible. This is based upon the assumption that these patients have the greatest perioperative risk and that this risk is likely reduced when the time is taken to optimize therapy. In contrast, other experts believe that surgery should proceed with extra caution during the intra- and postoperative phases of care to avoid complications; optimization of therapy can be addressed after recovery.

•Low-risk procedures – Most patients with uncontrolled or suspected OSA who are undergoing low-risk procedures can probably proceed with surgery, provided that increased attention is paid to aggressive postoperative surveillance and treatment. However, select patients may require deferral on a case-by-case basis.

Institution of preoperative empiric positive airway pressure therapy — We agree with the Society of Anesthesia and Sleep Medicine recommendations against the routine use of empiric PAP therapy in most patients with suspected OSA just prior to surgery, because of lack of evidence of benefit [55]. Intraoperative risk mitigation strategies for OSA should be utilized, and these patients should be referred postoperatively for OSA evaluation. However, we believe that select populations might benefit, including those with obesity hypoventilation syndrome (OHS), overlap syndrome, severe OSA (AHI >30), or resting hypoxemia [71]. (See "Intraoperative management of adults with obstructive sleep apnea".)

Further data on outcomes associated with perioperative PAP are provided separately. (See "Postoperative management of adults with obstructive sleep apnea", section on 'Known OSA on treatment'.)

Urgent surgery — Urgent and emergency surgery should not be postponed to make a formal diagnosis of OSA or to institute treatment. Patients thought to be at high risk of OSA should be managed with a presumptive diagnosis of OSA, and perioperative risk mitigation strategies should be used. (See "Intraoperative management of adults with obstructive sleep apnea" and "Postoperative management of adults with obstructive sleep apnea".)

AMBULATORY VERSUS INPATIENT SURGERY — The decision to schedule surgery on an inpatient or ambulatory basis should be individualized. An algorithm for this decision is provided (algorithm 1). Practice guidelines and consensus statements have been published to address this issue and guide clinical practice [31,90]. Important considerations include the patient's age and comorbidities, type of surgery and anesthesia, severity of the OSA, the requirement for postoperative opioid analgesia, patient compliance with positive airway pressure (PAP) therapy, and capabilities of the surgical center (eg, availability of difficult airway equipment, respiratory care equipment, radiology facilities, clinical laboratory facilities, and a transfer agreement with an inpatient facility). Importantly, the plan for same-day discharge for patients with OSA should be considered tentative. These patients should be assessed after surgery and should remain in a monitored setting if they have recurrent or severe respiratory events during recovery from anesthesia. Surgeons, patients, and family members or caregivers must be made aware of the need for heightened vigilance after discharge home. Post-anesthesia care unit (PACU) and hospital discharge criteria and instructions are discussed separately. (See "Postoperative management of adults with obstructive sleep apnea", section on 'Discharge from the post-anesthesia care unit' and "Postoperative management of adults with obstructive sleep apnea", section on 'Floor management'.)

Major society guidelines (American Society of Anesthesiologists and the Society for Ambulatory Anesthesia) generally agree that the following patients are reasonable candidates for outpatient/ambulatory surgery [31,90]:

●Superficial procedures (eg, lithotripsy, arthroscopy, cataract surgery, colonoscopy, esophagogastroduodenoscopy, bronchoscopy) or procedures requiring only local or regional anesthesia can typically be performed safely on an outpatient basis in patients with OSA.

●Ambulatory surgery can be performed for patients with known or presumed OSA who will comply with recommendations for postoperative PAP, have well-controlled comorbidities, and undergo procedures that will not require predominantly opioid analgesic regimens [90]. The safety of this approach is supported by a retrospective review of 746 patients with known or suspected OSA, without uncontrolled comorbidities, who underwent ambulatory laparoscopic adjustable gastric banding for weight loss [91]. There were no cases of respiratory failure or reintubation, and the one unplanned admission was for a surgical indication. Several other studies have reported no increase in unanticipated hospital admission or other perioperative events after ambulatory surgery in well-selected patients with OSA [92-96].

Patients having other procedures should be assessed on a case-by-case basis, with the following considerations [90]:

●Painful operations that require primarily opioids for analgesia may not be appropriate to discharge home on the day of surgery, especially if these opioids are administered parenterally.

●Upper airway surgery (eg, tonsillectomy, uvulopalatopharyngoplasty, tongue base surgery) may lead to worsening of OSA from airway swelling and difficulty using PAP devices during the surgical recovery period. We monitor patients with OSA after upper airway surgery for an extended period in the PACU after the usual discharge criteria are met, and discharge them home the same day if they are otherwise healthy, and there have been no respiratory events in the PACU. In one review of patients undergoing same-day OSA surgery, the risk of severe postoperative adverse events was estimated at <2 percent, without any reports of postoperative deaths [97]. However, those with a higher apnea-hypopnea index (AHI), and patients requiring higher opioid doses for analgesia were admitted to the hospital.

●Patients with non-optimized comorbidities (eg, uncontrolled hypertension, heart failure, arrhythmias, cerebrovascular disease) may require hospital admission following surgery.

●Patients who are unwilling or unable to use PAP devices postoperatively at home may not be good candidates for ambulatory surgery.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Sleep-related breathing disorders in adults" and "Society guideline links: Preoperative medical evaluation and risk assessment".)

SUMMARY AND RECOMMENDATIONS

●Prevalence and surgical risk

•The prevalence of obstructive sleep apnea (OSA) in surgical patients has been estimated to be as high as 6 to 17 percent, perhaps higher in high-risk populations such as patients undergoing bariatric or major surgery (up to 70 percent). (See 'Epidemiology' above.)

•In patients with OSA, the incidence of perioperative complications is increased two- to fourfold; respiratory complications are most common. Patients undergoing high-risk surgical procedures are at the greatest risk. (See 'Perioperative complications' above.)

●Preoperative screening for OSA

•We suggest screening all preoperative patients for OSA (Grade 2C). If not all patients can be screened, the most critical populations to screen are patients with obesity (body mass index [BMI] ≥35 kg/m²) (calculator 1), those scheduled for bariatric surgery, and those with other medical conditions highly associated with OSA (table 2 and figure 1).

•The STOP-Bang questionnaire is the best validated tool for preoperative screening for OSA, and is the screening tool that we use (table 4). For patients who score at intermediate risk a second scoring step may improve performance of the STOP-Bang, though it adds complexity and is not widely used (table 5) (calculator 2). (See 'Screening questionnaires' above.)

●Preoperative evaluation

•Patients with known OSA should be assessed for the severity and adequacy of management. (See 'Initial assessment' above and 'Patients with known OSA' above.)

•In patients with known or suspected OSA, the clinician should look for the presence of OSA-associated medical conditions (table 2). (See 'Initial assessment' above and 'Associated conditions' above.)

●Timing of surgery – The majority of patients with known or suspected OSA may proceed to surgery without additional testing or treatment. The decision to defer surgery for further testing or treatment should be individualized based on patient, surgical, and institutional factors. (See 'Proceeding with further evaluation or surgery' above and 'Factors affecting the decision' above.)

•Emergency surgery should not be postponed to establish a formal diagnosis of OSA. (See 'Urgent surgery' above.)

•For patients who either screen at low risk for OSA or have a known diagnosis of OSA that is well controlled, no further evaluation prior to surgery is typically required. (See 'Elective surgery' above and 'Well-controlled OSA' above.)

•For patients with partially treated/untreated OSA (ie, uncontrolled OSA) or suspected OSA, with associated significant or uncontrolled systemic disease or additional problems with ventilation or gas exchange (eg, hypoventilation syndrome, severe pulmonary hypertension, or resting hypoxemia in the absence of other cardiopulmonary disease), elective surgery should be deferred for evaluation and treatment. (See 'Uncontrolled or suspected OSA' above and 'Elective surgery' above.)

•For patients with uncontrolled or suspected OSA without associated significant or uncontrolled systemic disease or additional problems with ventilation or gas exchange, the decision to defer surgery for further evaluation or treatment should be based on the level of surgical risk (see 'Risk factors for perioperative complications' above and 'Uncontrolled or suspected OSA' above):

-For high-risk surgery, we prefer to defer surgery if feasible. Others prefer to proceed with high-risk surgery without further preoperative evaluation, but with heightened intraoperative and postoperative precautions.

-For low-risk surgery, most of these patients can proceed with surgery without further evaluation, with heightened perioperative precautions.

•In patients with known or suspected OSA who need optimization of therapy or further testing, we prefer to schedule surgery no sooner than one week after changes are made. (See 'Proceeding with further evaluation or surgery' above.)

●Preoperative positive airway pressure (PAP) therapy – We do not empirically start new PAP therapy preoperatively in most patients. Patients who are on PAP therapy for OSA (or other OSA treatment device, such as a mandibular repositioning device) should continue treatment up to the day of surgery and should bring their PAP or other devices with them on the day of surgery. (See 'Patients with known OSA' above and 'Institution of preoperative empiric positive airway pressure therapy' above.)

●Ambulatory versus inpatient surgery – The decision to schedule surgery on an inpatient or outpatient basis should be based on the patient's age and comorbidities, type of surgery and anesthesia, severity of the OSA, compliance to PAP therapy, and the requirement for postoperative opioid analgesia (algorithm 1). (See 'Ambulatory versus inpatient surgery' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Robert Basner, MD, who contributed to an earlier version of this topic review.