Evaluation of perioperative pulmonary risk

INTRODUCTION — Postoperative pulmonary complications contribute significantly to overall perioperative morbidity and mortality.

The definition of postoperative pulmonary complications, pertinent anesthetic physiology, identification of patient and procedure-related risk factors, and the role of preoperative pulmonary function testing and pulmonary risk indices will be reviewed here. Strategies for reducing postoperative pulmonary complications, the management of postoperative pulmonary complications, and the evaluation of candidates for lung resection surgery are discussed separately.

●(See "Overview of the management of postoperative pulmonary complications".)

●(See "Strategies to reduce postoperative pulmonary complications in adults".)

●(See "Preoperative physiologic pulmonary evaluation for lung resection".)

DEFINITION OF POSTOPERATIVE PULMONARY COMPLICATIONS — The reported frequency of postoperative pulmonary complications in the literature varies from 2 to 70 percent. This wide range is due in part to patient selection and procedure-related risk factors, although differing definitions for postoperative complications account for much of the variability and make comparison of reported incidences across different studies difficult.

The preferred definition is a postoperative pulmonary abnormality that produces identifiable disease or dysfunction that is clinically significant and adversely affects the clinical course [1]. This would include several major categories of clinically significant complications, including [2,3]:

●Atelectasis

●Infection, including bronchitis and pneumonia

●Respiratory failure (mechanical ventilation for >48 hours after surgery or unplanned reintubation)

●Hypoxemia

●Exacerbation of underlying chronic obstructive pulmonary disease (COPD) or asthma

The rate of postoperative pulmonary complications across all types of surgery was 6.8 percent in a systematic review of studies that provided explicit outcome definitions [2].

IMPACT OF POSTOPERATIVE PULMONARY COMPLICATIONS — The National Surgical Quality Improvement Program (NSQIP) found that postoperative pulmonary complications occurred in 6 percent of 165,196 patients who underwent major abdominal surgery [4]. In a separate report of 1202 patients undergoing non-cardiothoracic surgery with general anesthesia, patients with at least one pulmonary complication had higher rates of mortality, ICU admission, and length of stay [5]. All patients were American Society of Anesthesiologist (ASA) physical status class 3 or greater.

In another study that used the NSQIP database, investigators examined the impact of particular postoperative complications on 30-day postoperative mortality among 2.82 million patients undergoing outpatient surgery [6]. Ten and one-quarter percent of patients who developed a postoperative pulmonary complication died within 30 days after surgery. Among eight types of postoperative complications studied, only cardiac complications conferred higher rates of 30-day mortality.

Given the frequency and morbidity of pulmonary complications following surgery, estimation of their risk should be a standard element of all preoperative medical evaluations.

PERIOPERATIVE PULMONARY PHYSIOLOGY — Postoperative pulmonary complications follow logically as an extension of normal perioperative pulmonary physiology. Reduced lung volume after surgery is a major factor contributing to the development of postoperative pulmonary complications.

Thoracic and upper-abdominal surgery are associated with a reduction in lung volumes in a restrictive pattern as follows [7,8]:

●Vital capacity (VC) is reduced by 50 to 60 percent and may remain decreased for up to one week.

●Functional residual capacity (FRC) is reduced by about 30 percent.

Diaphragmatic dysfunction appears to play the most important role in these changes; postoperative pain and splinting are also factors [9]. Reduction of the FRC below closing volumes contributes to the risk of atelectasis, pneumonia, and ventilation/perfusion (V/Q) mismatching. Microatelectasis results in areas of the lung that are perfused but not ventilated, leading to impaired gas exchange with consequent postoperative hypoxemia [10].

A decrease in tidal volume, loss of sighing breaths, and increase in respiratory rate occur after abdominal and thoracic surgery and contribute to the risk of complications. In addition, residual effects of anesthetic agents, neuromuscular blockers, and postoperative opioids depress the respiratory drive. Inhibition of cough and impairment of mucociliary clearance of respiratory secretions are factors that contribute to the risk of postoperative infection [11].

Lower abdominal surgery is associated with similar changes but to a lesser degree. Reductions in lung volumes are generally not seen with surgery on the extremities [12].

PATIENT-RELATED RISK FACTORS — Risk factors for pulmonary complications can be grouped into patient-related and procedure-related risks [2]. The potential patient-related factors that have been studied include the following:

●Age

●Chronic obstructive pulmonary disease

●Asthma

●Smoking

●General health status

●Obesity

●Obstructive sleep apnea

●Pulmonary hypertension

●Heart failure

●Upper-respiratory infection

●Coronavirus disease 2019 (COVID-19) infection

●Metabolic and nutritional factors

Age — Data regarding the influence of age as a predictor of postoperative pulmonary complications have consistently shown that age has an independent effect even after adjustment for comorbidities that increase with older age [2,5,13,14].

A systematic review prepared for the American College of Physicians estimated the impact of age on postoperative pulmonary complications among studies that used multivariable analysis to adjust for age-related comorbidities [2]. This review made the novel observation that age >50 years was an important independent predictor of risk. When compared with patients <50 years old, patients aged 50 to 59 years, 60 to 69 years, 70 to 79 years, and ≥80 years had odds ratios (OR) of 1.5 (CI 1.31-1.71), 2.28 (CI 1.86-2.80), 3.90 (CI 2.70-5.65), and 5.63 (CI 4.63-6.85), respectively, of developing pulmonary complications.

This observation was confirmed in the ARISCAT study of 2644 patients that derived a multifactorial risk index to predict pulmonary complications [15]. After adjusting for confounders more common with older age, advanced age was one of seven independent predictors of postoperative pulmonary complications. When compared with patients ≤50 years old, the odds ratio for development of pulmonary complications in patients >80 years old was 5.1 (CI 1.9-13.3).

In addition, even among elderly patients undergoing surgery, the risk is highest among the oldest patients. For example, in a systematic review of the risk factors for postoperative pneumonia after hip fracture surgery in the elderly (defined as age ≥65 years), subjects who developed pneumonia were on average five years older than those who did not develop pneumonia [16].

Therefore, even healthy older patients carry a substantial risk of pulmonary complications after surgery. This is in sharp contrast to perioperative cardiac risk, where age is not an independent risk factor after adjusting for confounders. (See "Evaluation of cardiac risk prior to noncardiac surgery".)

Chronic obstructive pulmonary disease (COPD) — COPD is an important patient-related risk factor for postoperative pulmonary complications. In an observational study using the National Surgical Quality Improvement Program (NSQIP) database including over 450,000 patients, unadjusted rates of postoperative pneumonia, prolonged ventilation, and reintubation among patients with COPD were 6.5 percent, 8.8 percent, and 5.5 percent, respectively [17].

A systematic review found that among studies that used multivariable analysis to adjust for patient-related confounders, the odds ratio for postoperative pulmonary complications attributable to COPD was 2.36 (CI 1.90-2.93) [2]. In the NSQIP observational study described above, after adjusting for confounders, COPD was an independent predictor for postoperative pneumonia (OR 1.71, 95% CI 1.59-1.83), reintubation (OR 1.54, 95% CI 1.42-1.66), and failure to wean from the ventilator (OR 1.45, 95% CI 1.35-1.56), in addition to non-pulmonary adverse outcomes [17].

Despite the increased risk of postoperative pulmonary complications in patients with obstructive lung disease, there appears to be no prohibitive level of pulmonary function below which surgery is absolutely contraindicated. For example, in a report of surgery in patients with severe COPD (FEV₁ <50 percent predicted), mortality was 5.6 percent (primarily related to a high mortality rate after cardiac surgery) and severe postoperative pulmonary complications occurred in 6.5 percent [18].

The impact of COPD on various postoperative outcomes after total hip arthroplasty was examined in a study of the National Surgical Quality Improvement Program (NSQIP) database [19]. After multivariable adjustment (n = 2426), complications, such as pneumonia, unplanned intubation, and prolonged ventilation, were all more common among patients with COPD. In addition, perioperative mortality (OR 2.21, CI 1.08-4.55) and 30-day readmission rates (OR 1.49, CI 1.24-1.79) were both more common.

While COPD is an important risk factor for postoperative complications, it confers less risk than other patient-related risk factors including age and heart failure [2].

The benefit of surgery must be weighed against the known risks; even very high-risk patients may proceed to surgery if the indication is sufficiently compelling. (See "Strategies to reduce postoperative pulmonary complications in adults", section on 'Chronic obstructive lung disease'.)

Asthma — Despite early reports indicating that patients with asthma had higher than expected rates of postoperative pulmonary complications, more recent studies have found no link for patients with well-controlled asthma. (See "Anesthesia for adult patients with asthma", section on 'Preoperative evaluation'.)

Smoking — Current cigarette smokers have an increased risk for postoperative pulmonary complications; smoking cessation prior to surgery reduces the risk of postoperative complications; smoking cessation for at least four to eight weeks may be even more effective. (See "Strategies to reduce postoperative pulmonary complications in adults", section on 'Smoking cessation'.)

In a 2014 meta-analysis of 107 cohort and case-control studies, preoperative smoking was associated with an increased risk of postoperative complications, including pulmonary complications (RR 1.73, 95% CI 1.35-2.23) [20]. Smokers with a greater than 20 pack-year smoking history have a higher incidence of postoperative pulmonary complications than those with a lesser pack-year history [21]. The risk attributable to cigarette smoking is procedure dependent and is higher for surgeries that carry intrinsically greater risk for pulmonary complications (eg, cardiovascular and oncologic procedures) [22].

Obesity — Physiologic changes that accompany class 3 or higher obesity (table 1) may include reduction in lung volumes, ventilation/perfusion mismatch, and relative hypoxemia [23]. These findings might be expected to accentuate similar changes seen with anesthesia and increase the risk of pulmonary complications. (See "Chest wall diseases and restrictive physiology", section on 'Obesity'.)

However, obesity has consistently been shown not to be an independent risk factor for postoperative pulmonary complications. A review of 10 series of obese gastric bypass patients, as an example, showed a 3.9 percent incidence of pneumonia and atelectasis, a rate similar to that seen in the general population [23]. Similarly, in a report of 602 patients undergoing bariatric surgery (mean body mass index [BMI] 42 kg/m²), clinically significant postoperative pulmonary complications (atelectasis and pneumonia) occurred in 1.8 percent of patients [24].

In a study that used the NSQIP database (n = 141,802), pulmonary complications were no more common among obese adults (BMI >30 kg/m²) than among those with a healthy weight (BMI 18.5 to 24.9 kg/m²) [25]. Unexpectedly, underweight patients actually sustained more postoperative pulmonary complications. This finding is persistent even after multivariable adjustment, for some, but not all, major surgical procedures.

In a separate NSQIP analysis, authors evaluated 30-day mortality, morbidity, and different potential postoperative complications among 5.57 million patients undergoing procedures from nine different surgical specialties [26]. While obesity (BMI ≥30 kg/m²) conferred higher rates of infection, venous thromboembolism, and renal complications, there was no difference in rates of mortality or pulmonary complications between patients who had normal weights and those in each of three different categories of obesity severity. Adjusted odds ratios for obesity categories I, II, and III (stratified by BMI) for pulmonary complications were 0.90 (CI 0.88-0.92), 0.94 (CI 0.91-0.96), and 0.97 (CI 0.95-0.99).

In a systematic review, among eight studies that adjusted for confounders more common among obese patients, only one reported obesity to be a predictor of postoperative pulmonary complication rates [2]. Obesity is not a significant risk factor for postoperative pulmonary complications and should not affect patient selection for otherwise high-risk procedures.

Patients with obesity hypoventilation syndrome (ie, body mass index >30 kg/[mg]² and an arterial tension of carbon dioxide [PaCO₂] >45 mmHg [6 kPa]) may be more likely to need postoperative positive airway pressure, although their specific perioperative pulmonary risks have not been formally studied. (See "Clinical manifestations and diagnosis of obesity hypoventilation syndrome" and "Treatment and prognosis of the obesity hypoventilation syndrome".)

Obstructive sleep apnea — Obstructive sleep apnea (OSA) is a risk factor for pulmonary (and other) complications after surgery. This is discussed in detail elsewhere. (See "Surgical risk and the preoperative evaluation and management of adults with obstructive sleep apnea", section on 'Perioperative complications' and "Clinical presentation and diagnosis of obstructive sleep apnea in adults".)

Many patients who are preparing for surgery have undiagnosed OSA. Screening for OSA prior to surgery is discussed in detail elsewhere. (See "Surgical risk and the preoperative evaluation and management of adults with obstructive sleep apnea", section on 'Initial assessment'.)

Interstitial lung disease — Interstitial lung disease, particularly idiopathic pulmonary fibrosis (IPF), is associated with an increased risk of postoperative pulmonary complications [27-29]. Among 282 patients with IPF who underwent elective surgical procedures at a single center, 14 percent experienced acute respiratory worsening, 5 percent had an acute exacerbation of IPF, and 9 percent developed pneumonia [30]. Baseline home oxygen use (relative risk [RR] 2.70, 95% CI 1.50-4.86) and increasing surgical time (per 60 minutes; RR 1.03, 95% CI 1.02-1.05) were identified as independent risk factors for postoperative acute respiratory worsening. (See "Anesthesia for patients with interstitial lung disease or other restrictive disorders".)

Pulmonary hypertension — Pulmonary hypertension (PH) increases complication rates after surgery, including in patients with mild to moderate pulmonary hypertension [31,32]. Potential complications include hemodynamic instability resulting in severe hypoxemia, acute right heart failure/circulatory collapse, cardiac dysrhythmias, and death. The increased risk warrants careful consideration of indications for surgery and discussion of potential risks with patients with pulmonary hypertension. (See "Treatment and prognosis of pulmonary arterial hypertension in adults (group 1)", section on 'Surgical or periprocedural care'.)

●A large database analysis of over 17 million hospitalizations for major noncardiac surgery and 143,846 patients with a diagnosis of PH demonstrated a significantly higher in-hospital mortality in patients with PH (4.4 versus 1.1 percent, adjusted odds ratio 1.51, 95% CI 1.47-1.55) [33]. Risk of cardiogenic shock was also highly increased (0.6 versus 0.1 percent, adjusted OR 2.37, 95% CI 2.20-2.55).

●Similarly, in a separate administrative database, perioperative mortality for orthopedic patients with PH (n = 3543) was markedly increased compared with matched controls (OR 3.72, 95% CI 2.13-6.39 for hip replacement; OR 4.55, 95% CI 2.16-9.39 for knee replacement) [34]. A limitation to both of these studies was the use of ICD-9 coding for comorbidities, which underestimates the prevalence of pulmonary hypertension.

●Data are more limited in terms of pulmonary comorbidities. One prospective observational study comparing 62 patients with PH of any etiology with matched controls showed a significantly increased risk for delayed extubation (21 versus 3 percent, OR 7.98, 95% CI 1.7-37.0) [35]. In a systematic review, four of five studies that reported respiratory failure or a composite pulmonary complication rate found a statistically significant increase in patients with PH [36].

The increased risk for postoperative complications appears to be true regardless of the underlying etiology of the pulmonary hypertension. As an illustration, authors studied 145 surgical patients with pulmonary hypertension, excluding those where the condition was due to left heart disease [37]. Complications included respiratory failure (n = 41), cardiac dysrhythmias (n = 17), congestive heart failure (n = 16), renal insufficiency (n = 10), and sepsis (n = 10). Risk predictors included a history of pulmonary embolus, New York Heart Association functional class ≥2, intermediate or high-risk surgery, and duration of anesthesia >3 hours.

Specific features of patients with pulmonary arterial hypertension that increase the risk of perioperative complications and mortality are discussed separately. (See "Treatment and prognosis of pulmonary arterial hypertension in adults (group 1)", section on 'Surgical or periprocedural care'.)

Heart failure — The risk of pulmonary complications may be even higher in patients with heart failure than in those with COPD. This is suggested by data from the systematic review that formed the basis of the American College of Physicians guideline, in which the pooled adjusted odds ratio for pulmonary complications was 2.93 (95% CI 1.02-8.43) for heart failure patients and 2.36 (1.90-2.93) for patients with COPD [2].

The original Goldman cardiac risk index has been shown to predict postoperative pulmonary as well as cardiac complications [38]. Although the Revised Cardiac Risk Index is now more commonly used to estimate risk for cardiovascular complications, validation studies of the revised index in predicting pulmonary complications have not been done. (See "Evaluation of cardiac risk prior to noncardiac surgery".)

General health status — Overall health status is an important determinant of pulmonary risk. Functional dependence (ie, need for assistance with activities of daily living) and impaired sensorium each increase postoperative pulmonary risk [2]. The commonly used American Society of Anesthesiologists (ASA) physical status classification correlates well with pulmonary risk (table 2). (See "Overview of anesthesia", section on 'Risk assessment'.)

The criteria for assigning ASA class include the presence of a systemic disease that affects activity or is a threat to life. Thus, patients with significant preexisting lung disease would be classified in a higher ASA class. ASA class >2 confers a 4.87 fold increase in risk (95% CI 3.34-7/10) [2].

●ASA 1 – Healthy.

●ASA 2 – Mild systemic disease (eg, well-controlled hypertension, stable, asthma, diabetes mellitus).

●ASA 3 – Severe systemic disease (eg, history of angina, COPD, poorly controlled hypertension, class 3 or higher obesity (table 1)).

●ASA 4 – Severe systemic disease with a constant threat to life (eg, history of unstable angina, uncontrolled diabetes or hypertension, advanced renal, pulmonary, or hepatic dysfunction).

●ASA 5 – Moribund patient not expected to survive without operation (eg, ruptured aortic aneurysm).

●ASA 6 – A declared brain-dead patient whose organs are being removed for donor purposes.

Upper respiratory infection — While data regarding the risk of pulmonary complications among adults undergoing high-risk surgical procedures with current or recent upper respiratory infection (URI) are limited, it appears prudent to defer elective surgery in this setting. Most of the data regarding the impact of URIs on perioperative morbidity derive from older studies of children undergoing surgery. Children with active URIs have more minor postoperative respiratory events such as oxygen desaturation, but no apparent increase in major morbidity or long-term sequelae [39]. (See "Anesthesia for the child with a recent upper respiratory infection".)

COVID-19 infection

Both active and recent COVID-19 infection markedly increase the risk for postoperative mortality and morbidity. For example, in a prospective cohort study of 1581 patients undergoing surgery and a confirmed diagnosis of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) within seven days before or 30 days after surgery, rates of mortality and PPC were 11.0 and 39.5 percent, respectively [40]. Risk exists for up to seven to eight weeks after COVID-19 infection. All patients preparing for surgery should be asked about symptoms that may suggest current or recent COVID-19 infection. The approach to routine versus selective testing of patients before elective surgery has evolved over the course of the pandemic. This is discussed elsewhere. (See "COVID-19: Perioperative risk assessment, preoperative screening and testing, and timing of surgery after infection".)

Metabolic and nutritional factors — A multifactorial risk index for postoperative respiratory failure identified two metabolic risk factors [41]. Albumin less than 3 g/dL and blood urea nitrogen (BUN) greater than 30 mg/dL each predicted risk; odds ratios were 2.53 and 2.29, respectively. A systematic review found that the magnitude of risk associated with a low serum albumin was similar to the degree of the most important patient-related risk factors and a stronger predictor of risk than an elevated BUN [2]. In a large database analysis of patients undergoing redo bariatric surgery (n = 58,777), serum albumin <3.5 g/dl was associated with higher rates of pneumonia, unplanned reintubation, and death [42].

PROCEDURE-RELATED RISK FACTORS — Surgical factors that may potentially affect pulmonary risk include the following:

●Surgical site

●Duration of surgery

●Type of anesthesia

●Type of neuromuscular blockade

Additionally, emergency surgery increases the risk for pulmonary complications [43].

Surgical site — Surgical site is the single most important factor in predicting the overall risk of postoperative pulmonary complications; the incidence of complications is inversely related to the distance of the surgical incision from the diaphragm. Thus, the complication rate is significantly higher for thoracic and upper abdominal surgery than for lower abdominal and all other procedures [44,45].

In a systematic review of 83 univariate studies, complication rates for upper abdominal surgery, lower abdominal surgery, and esophagectomy were 19.7, 7.7, and 18.9 percent, respectively [2]. The higher rates of complications in upper versus lower abdominal surgery relate to the effect upon respiratory muscles and diaphragmatic function. Abdominal aortic aneurysm repair is also associated with a high risk of postoperative pulmonary complications [41]. Other high risk procedures include head and neck surgery and neurosurgery [46,47].

In the ARISCAT study described above (see 'Age' above), surgical site was one of seven independent predictors of pulmonary complications [15]. Risk was highest for upper abdominal surgery (OR 4.4, CI 2.3-8.5) and intrathoracic surgery (OR 11.4, CI 4.9-26.0).

Laparoscopic cholecystectomy is associated with shorter recovery times, less postoperative pain, and less reduction in postoperative lung volumes [2]. In a systematic review, authors evaluated outcomes of 1248 patients undergoing cholecystectomy for acute cholecystitis [48]. Laparoscopic surgery was associated with lower rates of morbidity (OR 0.46, CI 0.34-0.61) and mortality (OR 0.2, CI 0.04-0.89), and a trend towards lower rates of pneumonia (OR 0.51, CI 0.25-1.01). However, only 4 of 10 eligible studies were randomized trials, so confounding may have been present.

In a pooled analysis of 12 studies of laparoscopic versus open colon cancer surgery, there was a nonsignificant trend towards reduced pulmonary complications (OR 0.65, CI 0.28-1.49) [2].

Duration of surgery — Surgical procedures lasting more than three to four hours are associated with a higher risk of pulmonary complications [45,49]. As an example, a study of risk factors for postoperative pneumonia in 520 patients found an incidence of 8 percent for surgeries lasting less than two hours versus 40 percent for procedures lasting more than four hours [44]. This observation suggests that, when available, a less ambitious, briefer procedure should be considered in a very high-risk patient.

General anesthesia versus neuraxial or regional anesthesia — There are conflicting data with regard to the pulmonary risk of spinal or epidural anesthesia when compared with general anesthesia [50]. Overall, it appears likely that general anesthesia leads to a higher risk of clinically important pulmonary complications than does epidural or spinal anesthesia, although further studies are required to confirm this observation. (See "Strategies to reduce postoperative pulmonary complications in adults", section on 'Anesthetic technique'.)

●One study, as an example, found no difference in mortality or in the rate of pulmonary infection or thromboembolic disease between patients undergoing total hip or total knee arthroplasty with neuraxial anesthesia compared with general anesthesia [51].

●In a study of 2644 patients with severe COPD (based on National Surgical Quality Improvement Program [NSQIP] definitions), who were matched to controls, rates of pneumonia (3.3 versus 2.3 percent), prolonged ventilator dependence (2.1 versus 0.9 percent), and unplanned postoperative intubation (2.6 versus 1.8 percent) were all modestly more common among patients receiving general anesthesia than among those receiving either neuraxial or regional anesthesia [52].

●The impact of anesthesia type was examined in a separate review of matched subjects from the NSQIP database (n = 328,540) [53]. When compared with general anesthesia, regional anesthesia (spinal, epidural, or peripheral) was associated with fewer respiratory complications (OR 0.76, CI 0.69-0.84), but no difference was observed in mortality or myocardial infarction.

●A large systematic review that predated the above studies evaluated the results of 141 trials that included 9559 patients who received neuraxial blockade (either epidural or spinal anesthesia) with or without general anesthesia or general anesthesia alone [54]. Patients receiving neuraxial blockade had an overall 39 percent reduction in the risk of pneumonia and a 59 percent decrease in the risk of respiratory depression. Potential limitations of the review include study heterogeneity and small sample size.

While general anesthesia confers higher rates of pulmonary complications than other anesthetic types, other anesthetic considerations are more important in determining anesthetic type than the risk of pulmonary complications. The decision regarding anesthetic type is usually best left to the anesthesiologist.

Regional nerve block is associated with lower risk and should be considered when possible for high risk patients [12]. As an example, an axillary block with conscious sedation could be used for an upper extremity procedure.

Type of neuromuscular blockade — Residual neuromuscular blockade can cause diaphragmatic dysfunction, impaired mucociliary clearance, hypoventilation, and ultimately contribute to postoperative pulmonary complications. Residual neuromuscular blockade is also an important risk factor for critical respiratory events in the immediate postoperative period [55]. (See "Strategies to reduce postoperative pulmonary complications in adults", section on 'Reversal of neuromuscular blockade'.)

PREOPERATIVE RISK ASSESSMENT — The preoperative risk assessment is largely based on the clinical evaluation, supplemented by selected testing for respiratory impairment. Such testing may include pulmonary function tests (PFTs), gas exchange evaluation (eg, ambulatory oximetry, six-minute walk test, arterial blood gas analysis), chest radiograph, and cardiopulmonary exercise testing. (See "Overview of pulmonary function testing in adults" and "Cardiopulmonary exercise testing in the evaluation of unexplained dyspnea".)

Clinical evaluation — A complete history and physical examination are the most important elements of preoperative risk assessment [38,45]. Significant risk factors, as defined above, should be identified. Any history suggesting unrecognized chronic lung disease or heart failure, such as exercise intolerance, unexplained dyspnea, or cough, requires further consideration [56]. We advise questioning all patients prior to major surgery about symptoms that may suggest obstructive sleep apnea, in particular, through the use of the STOP-Bang questionnaire (table 3). (See "Surgical risk and the preoperative evaluation and management of adults with obstructive sleep apnea", section on 'Screening with a questionnaire'.)

Physical examination should be directed toward evidence for obstructive lung disease, especially noting decreased breath sounds, wheezes, rhonchi, or prolonged expiratory phase [45].

Pulmonary function testing — PFTs are not needed in the majority of patients undergoing extrathoracic surgery [43,57]. In most cases, PFTs confirm the clinical impression without adding to the estimation of risk. However, all candidates for lung resection should have preoperative pulmonary function tests performed (see "Preoperative physiologic pulmonary evaluation for lung resection"). Additionally, PFTs may be useful in patients with known or suspected respiratory disease (eg, reduced exercise tolerance, unexplained dyspnea, cigarette smoking >20 years, chronic obstructive pulmonary disease [COPD], interstitial lung disease).

Based on published guidelines, a reasonable approach to patient selection for preoperative pulmonary function testing follows [43,58]:

●PFTs should not be used as the primary factor to deny surgery and should not be ordered routinely (in the absence of respiratory symptoms) prior to abdominal surgery or other high-risk surgeries (with the exception of lung resection surgery, which is addressed separately). (See "Preoperative physiologic pulmonary evaluation for lung resection".)

●Obtain PFTs for patients with COPD or asthma if clinical evaluation cannot determine if the patient is at their best baseline and that airflow obstruction is optimally reduced. In this case, PFTs may identify patients who will benefit from more aggressive preoperative management.

●Obtain PFTs for patients with dyspnea or exercise intolerance that remains unexplained after clinical evaluation. In this case, the differential diagnosis may include cardiac disease or deconditioning. The results of PFTs may change preoperative management.

●Obtain PFTs prior to cardiac surgery in patients with respiratory symptoms that may be attributable to cardiac disease or undiagnosed respiratory disease.

A number of measures of pulmonary function have been evaluated. Spirometry is widely available, and measures of the forced expiratory volume in one second (FEV₁) and forced vital capacity (FVC) have been frequently reported. Early reviews suggested criteria for increased risk that included the following [59]:

●FEV₁ <70 percent predicted

●FVC <70 percent predicted

●FEV₁/FVC ratio <0.65

Studies examining the role of PFTs in preoperative testing include the following:

●In a report of 11,217 patients undergoing coronary bypass surgery, authors stratified patients based on spirometric values [60]. After adjustment, while severe COPD (as defined by spirometric values) conferred higher rates of early mortality, there was no impact on rates of prolonged mechanical ventilation. Similarly, in a case control study of 164 patients undergoing abdominal surgery, no component of spirometry predicted pulmonary complications [45].

●A critical review of preoperative pulmonary function testing evaluated 14 studies that met strict methodologic criteria [2]. Spirometric values were significant risk predictors in three of four studies that used multivariable analysis. However, other factors conferred higher odds ratios for pulmonary complications than did abnormal spirometry in two of these studies: ASA class >3 [61] and chronic mucous hypersecretion [62].

●Two well-designed case-control studies have evaluated the benefit of PFTs as risk predictors. In a study of patients undergoing abdominal surgery, there was no difference in FEV₁, FVC, or FEV₁/FVC between patients who had a pulmonary complication and those who did not [45]. In contrast, factors from the physical examination did predict risk. In a second report, investigators matched smokers with severe airflow obstruction and an FEV₁ of less than 40 percent of predicted to smokers with a normal FEV₁ [63]. Only bronchospasm was more common among those patients with abnormal spirometry. The incidence of postoperative pneumonia, prolonged intubation, prolonged intensive care unit stay, and death were not significantly different between the two groups.

Few studies have compared the incremental value obtained by spirometry with the risk estimate based on clinical evaluation. Most patients at risk for complications due to moderate to severe COPD can likely be identified based on history and physical examination.

One group assessed the incremental benefit of PFTs in 320 patients with possible COPD based on respiratory symptoms (240 with confirmed COPD by spirometry) [64]. The authors compared three models of preoperative assessment: "standard" risk assessment including comorbidities, sociodemographic, and surgical data; a "COPD assessment model" that also included GOLD key indicators, pack-years of smoking, and exercise capacity; and a "PFT model" that included additional PFT parameters. There were minimal differences between the models in terms of ability to predict postoperative pulmonary complications; the strongest predictors were procedural (thoracic or upper abdominal surgery).

Assessment of oxygenation and hypercapnia — Measurement of pulse oxygen saturation (SpO₂) by oximetry is included in the ARISCAT Risk Index and the Gupta calculator for postoperative pneumonia (see 'ARISCAT' below and 'Gupta calculator for postoperative pneumonia' below). Assessment of SpO₂ can help stratify risk, particularly before high-risk surgeries [15,65]. Arterial blood gas (ABG) analysis is rarely needed as part of preoperative assessment. Based on clinical experience, indicators that an ABG might be useful include a resting SpO₂ <93 percent, an abnormal serum bicarbonate, and severe abnormalities on PFTs (eg, FEV₁ <1 liter).

In general, patients with hypercapnia can be identified based on established clinical risk factors, such as severe COPD or neuromuscular disease, rather than by routine ABGs. Several small case series have suggested a high risk of postoperative pulmonary complications among patients with a PaCO₂ >45 mmHg, a finding usually seen only in patients with severe chronic obstructive lung disease [12,66]. The risk associated with this degree of PaCO₂ elevation is not necessarily prohibitive, although it should lead to a reassessment of the indication for the proposed procedure and aggressive preoperative preparation.

One study reported an association between preoperative hypoxemia and postoperative pulmonary complications among 102 patients undergoing surgery for gastric or esophageal cancer [67].

Current data do not support the use of preoperative arterial blood gas analyses to stratify risk for postoperative pulmonary complications.

Chest radiographs — We suggest not obtaining routine preoperative chest radiographs unless there is evidence for undiagnosed or unstable cardiopulmonary disease such as new or escalating exertional dyspnea, wheezing, angina, orthopnea, lower extremity edema, hypoxemia, rales/rhonchi, or heart murmur (algorithm 1). This approach is consistent with the recommendations of multiple independent advisory organizations [43,68-70]. (See "Preoperative medical evaluation of the healthy adult patient", section on 'Chest radiographs'.)

Preoperative chest radiographs add little to the clinical evaluation in identifying patients at risk for perioperative pulmonary complications [71]. Abnormal findings on chest radiograph occur frequently and are more prevalent in older patients, but there is little evidence to support that these findings are helpful in perioperative assessment [72,73].

As an example, one study screened 905 surgical admissions for the presence of clinical factors that were thought to be risk factors for an abnormal preoperative chest radiograph [74]. These risk factors were age over 60 years or clinical findings consistent with cardiac or pulmonary disease. No risk factors were evident in 368 patients; of these, only one (0.3 percent) had an abnormal chest radiograph, which did not affect the surgery. On the other hand, 504 patients had identifiable risk factors; of these, 114 (22 percent) had significant abnormalities on preoperative chest radiograph.

The primary purpose of the preoperative chest radiograph in patients with risk factors is to facilitate evaluation and treatment of undiagnosed or worsening cardiopulmonary disease. For elective surgeries, this requires obtaining the chest radiograph early enough to facilitate evaluation and management of these conditions. We suggest chest imaging at least two weeks prior to surgery, but some findings may require additional delay to allow for preoperative optimization. (See "Strategies to reduce postoperative pulmonary complications in adults", section on 'Preoperative strategies'.)

Exercise testing — Cardiopulmonary exercise testing (CPET), which includes calculation of maximum oxygen uptake and ventilatory anaerobic threshold, is used to assess patients with abnormal PFTs to determine the safety of planned lung resection surgery. (See "Preoperative physiologic pulmonary evaluation for lung resection" and "Cardiopulmonary exercise testing in cardiovascular disease".)

●Cardiopulmonary exercise testing – CPET may also have a role in the evaluation of patients with unexplained dyspnea who are undergoing non-cardiopulmonary surgery. In one systematic review, a majority of nine eligible studies found that both maximum oxygen uptake and ventilatory anaerobic threshold predicted survival and overall postoperative complications in patients undergoing non-cardiopulmonary thoraco-abdominal surgery [75]. The studies did not measure postoperative pulmonary complication as a separate outcome. It is unknown to what extent this test adds value to the risk estimate obtained after a careful history and physical examination.

●Six-minute walk test – A simplified form of exercise testing that can be accomplished in an office setting is the six-minute walk test. In this test, patients are asked to walk as far as possible in six minutes and are allowed to rest if needed. The outcome is total distance walked. In a study of 117 patients undergoing thoracic or upper abdominal surgery, six-minute walk distance correlated with both the rates and severity of postoperative complications [76]. This study included all postoperative medical complications and was not limited to pulmonary complications. Similar results have been reported by other investigators [77,78]. (See "Overview of pulmonary function testing in adults", section on 'Six-minute walk test'.)

ESTIMATING POSTOPERATIVE PULMONARY RISK — Risk prediction tools use preoperative factors to estimate the risk of postoperative pulmonary complications. These tools are useful to stratify risk when advising patients before surgery and, in some cases, to identify patients most likely to benefit from risk-reduction interventions. For example, patients predicted to be at high risk of postoperative respiratory failure may be scheduled for postoperative care in a more intensive care location. All four risk indices below offer the advantage of providing a numerical estimate of risk, rather than a qualitative category of risk. (See "Strategies to reduce postoperative pulmonary complications in adults".)

The ARISCAT tool offers the advantage of the use of readily available clinical information and provides an estimate of the risk of any postoperative pulmonary complications, but the disadvantage of the inclusion of minor complications that may not impact outcome or contribute importantly to morbidity. The two Gupta risk calculators will be particularly helpful to establish the risk of a single complication, either pneumonia or respiratory failure, although they need to be downloaded to a personal device to perform the calculations. The Arozullah index will be of use primarily in research settings, as the tool is too complicated to use in clinical practice.

ARISCAT — The ARISCAT Risk Index (calculator 1) predicts the overall incidence of postoperative pulmonary complications (of any severity), by assigning a weighted point score to seven independent risk factors (table 4) [15]:

●Advanced age

●Low preoperative oxygen saturation

●Respiratory infection within the past month

●Preoperative anemia

●Upper abdominal or thoracic surgery

●Surgery lasting more than two hours

●Emergency surgery

The incidence of pulmonary complications in patients with scores stratified as low-, intermediate-, and high-risk is 1.6, 13.3, and 42.2 percent, respectively.

This index has the advantage of being simple to calculate manually at the bedside with readily available clinical information, but the disadvantage of inclusion of minor complications of little clinical significance (eg, new wheezing treated with bronchodilators). It was derived from a prospective evaluation of 2464 surgical patients and validated in a cohort of 5099 patients in Europe [65]. It has since been externally validated with similar test characteristics [65,79].

Arozullah respiratory failure index — The Arozullah respiratory failure index predicts the incidence of postoperative respiratory failure (mechanical ventilation for ≥48 hours) based on several factors, including type of surgery, laboratory results, functional status, history of chronic obstructive pulmonary disease (COPD), and age (table 5) [41]. Point scores are stratified into five classes with risk of respiratory failure ranging from 0.5 to 26.6 percent (calculator 2).

This index was based on multivariate analysis of a cohort of 81,719 patients from the National Veterans Administration Surgical Quality Improvement Program and validated on another 99,390 patients. It is too complicated for use in clinical practice and will likely be of most value in research settings.

Gupta calculator for postoperative respiratory failure — The Gupta calculator for postoperative respiratory failure (calculator 3) uses multiple preoperative factors to predict risk of failure to wean from mechanical ventilation within 48 hours of surgery or unplanned intubation/reintubation postoperatively [80]. It is derived from the American College of Surgeons’ National Surgical Quality Improvement 2007 data set (211,410 patients for training) and 2008 data set (257,385 patients for validation) using logistic regression techniques to determine the weight of preoperative predictors.

Gupta calculator for postoperative pneumonia — The Gupta calculator for postoperative pneumonia (calculator 4) was derived in a similar manner to the respiratory failure calculator [81].

American College of Surgeons NSQIP surgical risk calculator — This calculator was derived from the large NSQIP database to estimate mortality and a variety of complications to guide both clinicians and patients when considering surgery [82]. (See "Preoperative evaluation for anesthesia for noncardiac surgery".)

It is a web-based tool that provides estimates of the risk for serious complication, any complication, pneumonia, cardiac complications, surgical site infection, urinary tract infection, venous thromboembolism, renal failure, readmission, return to operating room, death, discharge to nursing or rehab facility, or sepsis. It has the advantage of estimating risk of numerous complications, but it takes several minutes to complete and (among pulmonary complications) only estimates the risk for pneumonia.

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: Preoperative medical evaluation and risk assessment".)

SUMMARY AND RECOMMENDATIONS

●Definition – Postoperative pulmonary complications are an important source of perioperative morbidity and mortality. They represent an extension of the normal physiologic changes in the lung that occur with anesthesia. (See 'Introduction' above.)

●Patient-related risk factors – Patient-related risk factors for these complications include the following: age >50 years, chronic obstructive pulmonary disease, congestive heart failure, poor general health status (American Society of Anesthesiologists [ASA] class >2), functional dependence, obstructive sleep apnea, current cigarette use, pulmonary hypertension, recent COVID-19 infection, low oxygen saturation, and serum albumin <3.5 g/dL. (See 'Patient-related risk factors' above.)

●Procedural risk factors – Procedure-related risk factors for postoperative pulmonary complications include surgery lasting more than three hours, emergency surgery, nonreversed long-acting neuromuscular blockade, and type of surgery. Upper abdominal, thoracic, aortic, head and neck, neurosurgery, and abdominal aortic aneurysm surgeries have increased risk for pulmonary complications. (See 'Patient-related risk factors' above and 'Procedure-related risk factors' above.)

●Probable additional risk factors – Additional potential risk factors include the following: general anesthesia (compared to spinal or epidural anesthesia), hypercapnia (arterial tension of carbon dioxide [PaCO₂] >45 mmHg), abnormal chest radiograph, cigarette use within the previous eight weeks, and current upper respiratory tract infection. (See 'Patient-related risk factors' above and 'Procedure-related risk factors' above.)

●Perioperative risk assessment

•Clinical evaluation – A careful history and physical examination are the most important tools for preoperative risk assessment. Attention should be paid to known risk factors as well as symptoms that suggest underlying lung disease (eg, exercise intolerance, cough, and unexplained dyspnea). A high-risk patient may benefit from aggressive strategies to reduce pulmonary complications. (See 'Preoperative risk assessment' above and "Strategies to reduce postoperative pulmonary complications in adults".)

•Additional work-up, in selected patients – In selected patients, preoperative pulmonary function tests (PFTs), chest radiograph, and exercise testing may identify risk factors that require attention preoperatively. Current data do not support the use of preoperative arterial blood gas analyses to stratify risk for postoperative pulmonary complications. (See 'Preoperative risk assessment' above.)

-PFTs – PFTs should generally only be obtained in patients with uncharacterized dyspnea or exercise intolerance and in those with uncertain control of asthma or chronic obstructive pulmonary disease (COPD).

-Chest imaging – We suggest not obtaining routine preoperative chest radiographs unless there is evidence for undiagnosed or unstable cardiopulmonary disease such as new or escalating exertional dyspnea, wheezing, angina, orthopnea, lower extremity edema, hypoxemia, rales/rhonchi, or heart murmur (algorithm 1). (See 'Chest radiographs' above.)

-Cardiopulmonary exercise testing (CPET) – CPET is used to assess patients with abnormal PFTs to determine the safety of planned lung resection surgery. CPET may also have a role in the evaluation of patients with unexplained dyspnea who are undergoing other types of major surgery. A six-minute walk test, which can be easily performed in an office setting, may be a reasonable, simpler alternative.

●Risk calculators – Five different pulmonary risk indices are available to provide a quantitative estimate of the risk of respiratory failure, pneumonia, or general pulmonary complications. Each risk index has different strengths and weaknesses. We use the ARISCAT index (calculator 1). These tools are a useful starting point when estimating pulmonary risk before major noncardiac surgery, and they guide conversations with the patient and surgeon. (See 'Estimating postoperative pulmonary risk' above.)