Principles of Preoperative and Operative Surgery

Patient–Surgeon Relationship

Clear, precise, and unambiguous communication to establish an understanding of mutual expectations and trust is at the pinnacle of the patient–surgeon relationship. A surgeon’s initial encounter with a patient most commonly is in the context of a new diagnosis and is initiated by either a professional or self-referral. A history and physical examination, whether in an urgent/emergent or elective setting, initially should focus on confirmation or rebuttal of the suspected diagnosis. Inquiries regarding patients’ personal interests, as well as their relationships with their community and society, help create a common bond between the patient and surgeon. In addition to direct communication with the patient, knowledge of their situation is augmented by a thorough review of relevant diagnostic laboratory and imaging results. Through this process, an experienced surgeon effectively recreates the clinical context of a patient’s situation during the period preceding evaluation for the illness in question.

Further patient management is directed by the differential diagnosis generated during the initial evaluation. If the differential diagnosis contains items of equipoise that require distinct treatments, further investigations may be necessary to distinguish between these options. In general, the principle of Occam razor or parsimony applies to surgical diagnosis and management; it is important to pursue only those diagnostic studies that have a high likelihood of producing actionable results. Tests with a near-perfect pretest probability, and those unlikely to alter treatment, should be avoided. Once a diagnosis is secured, the objective and urgency of potential surgical therapy are considered.

Surgical Objectives

Achieving a joint understanding of surgical objectives and expectations between patient and surgeon is paramount to improve patient satisfaction and outcomes. There are three broad potential objectives of surgical intervention: disease prevention, disease control, and symptom palliation. Examples of operations aimed at disease prevention include prophylactic mastectomy, colectomy, pancreatectomy or thyroidectomy for heritable cancer syndromes, endarterectomy for asymptomatic carotid stenosis, or appendectomy in the setting of Ladd procedure for intestinal malrotation. These operations are aimed at preempting a disease process. Operations for disease control address a process that is ongoing. Examples include resections for malignancy, cholecystectomy for acute cholecystitis, enterolysis for bowel obstruction, bypass for vascular occlusive disease, or knee replacement for arthritis. With these operations, patients may expect partial or complete resolution of the targeted disease process. Finally, palliative operations are aimed at improving quality of life, rather than curing a disease. Examples include proximal decompression for malignant bowel obstruction or gastrojejunostomy for unresectable pancreatic cancer with gastric outlet obstruction. Inadequate communication of an operation’s objectives precludes informed consent and can have dramatic, negative implications for a patient’s perioperative decision-making.

Elective, Urgent, and Emergent Indications

Appropriate triage of surgical therapy is important for patient outcomes as well as resource distribution. Accurate categorization of surgical urgency also has implications for quality reporting. The American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) differentiates between emergency and elective operations and reports different levels of accuracy for patient risk estimates, with emergency cases having both superior predictive accuracy for mortality and significantly higher observed-to-expected ratios. Within ACS NSQIP, emergency surgery is characterized by an ongoing acute process that can result in rapid deterioration in a patient’s condition for which unnecessary delay can potentially threaten the clinical outcome. On the other hand, elective operations generally involve a patient who has completed preoperative surgical evaluation during a separate patient–urgeon encounter and is subsequently scheduled for operation. Inpatients, referrals from the emergency department, and direct transfers from clinic are excluded from elective patient categorization. Urgent operations are a relatively ill-defined category and have an acuity level in between those of elective and emergent cases. The World Society of Emergency Surgery created the Timing of Acute Care Surgery classification in 2013, which subdivides urgent cases into those with an ideal time-to-surgery falling between immediate to within 48 hours from diagnosis. Recognition of urgency should be one of the first steps of the preoperative surgical evaluation, as it affects subsequent patient decision-making, counseling, investigatory testing, and perioperative management.

Risk Assessment

Perioperative risk assessment has an impact on all aspects of surgical planning, including the decision to operate, choice of operation, perioperative management, and goals of care discussions. Communication with patients regarding the risks of a proposed operation must be coupled with a thorough review of patient comorbidities and functional status. Hence, patients with prohibitively high operative risk can be protected from inappropriate operations, while those with borderline risk can be medically optimized preoperatively. From the viewpoint of perioperative management, appropriate risk stratification facilitates resource allocation, including intraoperative monitoring, use of intensive care unit (ICU) services after the operation, and potential use of medical consultation. Finally, transparent categorization of patient risk factors improves institutional reporting and allows for multi-institutional comparisons of risk-adjusted patient outcomes.

Risk assessment begins by considering the nature of the disease and patient comorbidities while weighing the risks of possible surgical interventions. For a number of diagnoses, multiple surgical approaches may be available, each with advantages and disadvantages in terms of the morbidity profile, quality outcomes, and durability of the therapy. These must be weighed carefully in the context of each patient’s presenting condition and baseline level of health. The American Society of Anesthesiologists (ASA) physical status categorization has frequently served as a simple, initial rubric to summarize baseline patient comorbidity. Patient categorization into ASA 1 to 5 ( Table 10.1 ) helps stratify patients as low risk (1–2), intermediate risk (3), and high risk (4–5). The addition of an “E” designation signifies emergent operation, hence indicating a higher risk. First introduced in 1941, increasing ASA class was shown in a number of landmark studies to be associated with early postoperative mortality. This relationship remains true in the modern era for both emergent and elective operations, with class 5E associated with nearly 20% likelihood of early postoperative mortality.

Next, operations can be categorized into low-, intermediate-, and high-risk groups. This categorization is most commonly approached through expert opinion and consensus guidelines, such as those proposed by the European Society of Cardiology and European Society of Anesthesiology, which stratifies patients based on estimated 30-day risk of cardiac events ( Table 10.2 ). Combination surgical risk models including patient comorbidity and operative risk have been developed using logistic regression with mortality or major complication as the dependent variable. For very low-risk, outpatient procedures, the risk of death is less than 1 in 50,000; conversely, high-risk operations performed for life-threatening conditions in critically ill patients can have expected mortality rates routinely exceeding 20% ( Fig. 10.1 ).

There are a wide variety of tools to help quantify operative risk in the preoperative setting. These tools can predict risks of broad outcomes such as death or length of stay or specific events such as reoperation, intraoperative blood loss, or specific surgical complications. From a methodology standpoint, these tools can be grouped into categorical scales, risk scores, or prediction models. Categorical scales are easy to calculate and are frequently subjective. The most classic example of a categorical scale is the ASA Classification of Physical Status, which is frequently used in preoperative risk estimation. Surgical risk scores combine several predictors, usually chosen using multivariable predictive modeling, to estimate risk of a specific outcome. An example of a risk score is the model for end-stage liver disease (MELD) score used to predict short-term prognoses in patients with end-stage liver disease. More recently, the adoption of advanced statistics to analyze large, multi-institutional datasets has created numerous risk prediction models that account for patient-level risk factors to generate estimated likelihoods of multiple surgical outcomes. The most commonly used and cited example of a surgical risk score is the ACS NSQIP Surgical Risk Calculator (SRC).

The ACS NSQIP universal SRC was developed in 2013 using standardized clinical data from more than 500 NSQIP participant hospitals. This online tool predicts adverse postoperative outcomes based on 20 preoperative patient-level characteristics, including demographics and comorbidities ( Table 10.3 ). Risks associated with procedure type are incorporated using Current Procedural Terminology (CPT) codes. Updated in 2016, the database has been calibrated to predict more accurately outcomes for lowest- and highest-risk patients. It currently contains data from more than 3.8 million cases across 740 hospitals and is publicly accessible at http://riskcalculator.facs.org . Originally designed to predict eight postoperative adverse outcomes, the tool has evolved to currently report likelihoods of 13 specific or composite outcomes within 30 days of surgery ( Box 10.1 ).

Levels of discrimination for these outcomes are generally strong, with c-statistics higher than 0.75 for all predicted outcomes. In particular, discrimination for 30-day postoperative mortality is excellent, with c-statistic exceeding 0.9. Recent investigations have evaluated combining ACS NSQIP SRC models with preoperative biologic markers such as hypoalbuminemia or with organ-specific metrics such as chronic liver disease in patients selected for liver resection. Within the last 5 years, the SRC model has also been adapted for the pediatric population. The Pediatric SRC incorporates nearly 200,000 cases across 67 hospitals and accounts for 382 CPT codes; it has demonstrated excellent predictive accuracy for mortality and morbidity following operations in children. The ACS NSQIP SRC is specifically designed to facilitate patient counseling and consent prior to surgery; as such, it does not take into account intraoperative findings. Despite its excellent calibration within the broad ACS NSQIP dataset, recent studies have identified lapses in predictive accuracy within smaller homogenous patient populations. Therefore, it cannot replace familiarity with institution- and surgeon-specific performance.

Informed Consent

Surgeons have an ethical obligation to discuss and pursue informed consent with any patient considering an operation. Comprehensive, transparent, and clear communication of perioperative risks and potential benefits is mandatory. To successfully guide the patient through the consent process, a surgeon must possess a thorough technical understanding of the proposed operation, the most probable perioperative course, and potential pitfalls and complications. A clear and precise communication of risks and expectations is paramount; technical jargon should be avoided. The consent process must take into account all of the preceding facets of surgical objectives, urgency, and patient risk assessment. Systematic reviews indicate that common components of a consent discussion should include (1) the disease diagnosis, (2) the proposed procedure, (3) procedure-related risks, (4) likelihood of success of the procedure, (5) mental capacity of the patient, and (6) alternative treatment options. One of the most challenging impediments to informed consent is the knowledge gap between surgeon and patient. To overcome this, consent processes can be augmented for specific diseases and procedures using decision aids, visual materials, specialized written materials, and previously discussed risk calculators. In general, these supplementary instruments have been shown to improve patient knowledge and satisfaction with decision-making.

Comprehensive Geriatric Assessment

Chronologic demarcation of the “geriatric” patient is elusive. A combination of advanced age, comorbidities, and functional and/or cognitive decline contributes to the definition of a geriatric patient. Due to the clinical and social complexities of the geriatric population, an appropriate preoperative evaluation is multifaceted. Collaboration between ACS NSQIP and the American Geriatrics Society (AGS) produced a set of guidelines for multidomain assessment of geriatric patients. This approach was not novel; the concept of comprehensive geriatric assessment (CGA) was first implemented in the 1980s and 1990s in medical inpatient and long-term outpatient settings. Geriatric assessments were shown to improve independent living, physical function, and long-term mortality. Studies reporting the implementation of CGA within surgical populations are rare; a recent systematic review found positive impacts of CGA use on procedural cancellation rate, surgical complications, and length of stay. Importantly, historic data demonstrate that the most effective CGA programs are those that impact direct medical care recommendations. The CGA is comprised of medical, mental health, functional capacity, social, and environmental domains. The ACS NSQIP/AGS collaborative framework ( Box 10.2 ) refocuses the CGA toward themes more relevant to operative and perioperative care. While all aspects of the framework are vital, several are more pertinent within the geriatric population considered for surgery.

Cognitive Impairment and Delirium

Nearly one in five elderly patients has dementia or cognitive impairment. The ACS NSQIP/AGS framework recommends routine neurocognitive assessment for the elderly in the preoperative setting. Specifically, the guidelines recommend cognitive evaluation for any geriatric patient without a preexisting diagnosis of cognitive impairment. Methods include cognitive assessments such as the Mini-Cog ( Box 10.3 ) and/or interviews with the patient’s support structure (spouse or family) or affiliated healthcare providers. The Mini-Cog’s advantages include the large body of evidence supporting its usefulness, ease of implementation (3 minutes to complete), and focus on attention and executive function. Any findings suggestive of cognitive impairment should prompt referral to a primary care, mental health, or geriatric specialist. Establishing cognitive impairment early in the preoperative setting has direct implications for patient–physician communication, decision-making capacity, and informed consent.

In the postoperative setting, cognitive impairment strongly predicts delirium, which has an incidence of nearly 50% among geriatric patients. Documentation of preexisting cognitive impairment improves interpretation of perioperative mental status and encourages avoidance of medications that may precipitate delirium. In the geriatric population, postoperative delirium has a profound impact on hospital length of stay, long-term postoperative cognition, cost of care, and mortality. Because evidence-based treatments for delirium are few, the majority of studies focus on the identification of risk factors and prevention. The AGS best practice guidelines for delirium recommend preoperative risk factor screening for all geriatric surgical patients ( Box 10.4 ). Identification of these risk factors should raise awareness to avoid second-hit insults (such as high-risk medication administration, sleep cycle disturbance) and to implement simple measures that improve the patient’s orientation. Such measures include holding high-risk medications preoperatively, providing hearing aids, encouraging sleep hygiene with preservation of day/night cycle, and employing the help of the patient’s family for reorientation during postoperative in-hospital recovery.

Depression

Increased comorbidity and medical burden, conditions highly prevalent in the geriatric population, can exacerbate depressive symptoms. Additional risk factors include sleep disturbance, low functional status, low education level, and heavy alcohol or polysubstance use. Elderly patients with preoperative depressive symptoms can experience postoperative delirium at a significantly higher rate and longer duration. Depression also may lower the threshold for pain and is a predictor of chronic postoperative pain. In the intensive care setting, depression is associated with increased mortality and reduced quality of life following discharge. Following cardiac surgery, depression and anxiety may increase the likelihood for coronary disease recurrence and mortality. The ACS NSQIP/AGS guidelines recommend the Patient Health Questionnaire-2 as a pragmatic preoperative screening tool for elderly patients; positive findings should be followed by appropriate referral. Optimal management of depression requires a multidisciplinary approach frequently involving both psychiatric medications and cognitive behavioral therapy.

Medication Management

The ACS NSQIP/AGS framework emphasizes the importance of obtaining a comprehensive medication history for all older patients, including over-the-counter medications, eye drops, vitamins, and herbal products. Adverse drug reactions, inappropriate dosing, and polypharmacy can be avoided by considering potential changes in drug metabolism and clearance in the perioperative setting.

The American College of Cardiology (ACC) and American Heart Association (AHA) guidelines for perioperative beta blockade supports continued administration of beta blockers for patients already on this medication. Patients undergoing intermediate-risk surgery with known coronary artery disease or risk factors for ischemic heart disease may be candidates for perioperative beta blockade. If initiation of beta blockers is indicated in the preoperative setting, treatment ideally should start weeks before elective surgery and should be titrated to a target heart rate of 60 to 80 beats/minute. Adverse effects of beta blocker initiation too close to the time of surgery include risk of stroke, hypotension, and death. For patients with limited cardiac risk factors, rapid initiation of beta blockers in the acute preoperative setting is not indicated.

The AGS Beers Criteria for Potentially Inappropriate Medication Use are particularly relevant to elderly patients at risk for polypharmacy. The latest guidelines (updated in 2015) serve as a reference to check for medications with high-risk adverse-effect profiles, common drug-drug interactions, impaired renal and/or hepatic clearance, perioperative sedation, and a predisposition to delirium. Some of the medications, such as benzodiazepines, are categorically contraindicated as they have been demonstrated to increase risk of cognitive impairment, delirium, falls, and other adverse outcomes in older adults.

Functional Status and Frailty

Elderly patients can be impaired in their performance of tasks necessary for independent living. These functional limitations are associated with perioperative complications, discharge to facilities other than home, and postoperative mortality. The association between functional dependence and postoperative mortality is present in patients over 60 but is magnified in patients over 80. A simple way to obtain a broad sense of functional dependence is to inquire about a history of falls. More detailed instruments for scoring functional status include the activities of daily living and instrumental activities of daily living, which describe the ability to perform basic and higher-level functions, respectively ( Box 10.5 ).

The social and family support systems surrounding the patient are intimately interwoven with the functional level of the geriatric patient. The living situation of the patient—independent, with family, in assisted living, or with a neighboring support structure—has far-reaching implications not only for their overall health but also as indicators of postoperative disposition and recovery. Identification and incorporation of the patients’ living situation and support system into perioperative decision-making is vital to successful patient-centered recovery processes and management of expectations.

A series reporting on surgical outcomes of octogenarians and nonagenarians have shown that age is often a poor independent marker for surgical risk. A more accurate predictor is frailty. While no single definition of frailty exists, the AGS suggests that frailty is a syndrome comprised of a combination of weakness, fatigue, weight loss, decreased balance, low physical activity, slowed motor processing, social withdrawal, cognitive changes, and vulnerability to stressors. The impact of frailty on postoperative outcomes cannot be overstated. Frailty is associated with major complications and early postoperative mortality for cardiothoracic, orthopedic, otolaryngologic, and elective cancer operations. In a large ACS NSQIP study, preoperative frailty index was more strongly associated with postoperative cardiac arrest and death than ASA class or history of myocardial infarction (MI).

Frailty can be measured using exhaustive, multidimensional scales such as the CGA or more pragmatic tests such as the Timed Up and Go (TUG) tool, which measures functional mobility. In a comparison of four frailty scales to predict postoperative outcomes of cardiac surgery, gait speed outperformed several more extensive scales in predicting mortality or major morbidity. As the most pervasive gait speed measure, TUG time is calculated by measuring the time it takes for a patient to rise from a chair, walk 3 m, turn, and return to a sitting position in the same chair ( Fig. 10.2 ). In a multi institutional study of patients undergoing minor and major elective operations for solid malignancies, a TUG time >20 seconds was associated with a 50% risk of major complications among patients older than 70, compared to 14% for patients with a TUG time ≤20 seconds.

Although the physical function domain of frailty may be the easiest and most objective to measure, more comprehensive scales such as CGA may produce more actionable results that can improve optimization in the preoperative setting for elective procedures.

Patient Counseling

Integrated into the process of informed consent is clear communication between physician and patient regarding the patient’s individualized goals of treatment. It is critical for the patient to have clear and realistic expectations regarding the likely treatment course and any potential complications. In the case of older patients, conducting this discussion in the presence of the patient’s anticipated postoperative support system—including spouse, adult children, or home nurse—can help ensure that both the patient and the support system are informed regarding necessary postoperative care. This is even more vital if the patient has any cognitive impairment. Regardless of baseline cognitive or functional state, it is strongly recommended that older patients anticipating elective surgery should arrange for an advanced directive and designate a health care proxy or surrogate decision maker. These documents should be prominently featured in the medical chart.

Cardiovascular System

As the US population continues to age, patients with heart disease undergoing elective, noncardiac surgery will increase. Perioperative cardiac complications are associated with morbidity, mortality, and cost. However, preoperative cardiac intervention to reduce the risk of noncardiac surgery is rarely needed, except when such an intervention is indicated for management of the patients’ baseline condition. In the preoperative setting, the goal of cardiology evaluation (if indicated) is not to provide medical “clearance,” but rather to provide information regarding the patient’s cardiac risk profile and the management options for this risk. The overarching principle of preoperative cardiovascular evaluation is to obtain supplemental testing only when these tests have a reasonable likelihood of changing management. These changes may involve delaying the operation, preoperative revascularization, medical optimization, modifying perioperative monitoring, or referral to a specialty care center.

The ACC and AHA published collaborative guidelines regarding perioperative cardiac risk in 2007 that included identification of three surgery-specific risk categories. In general, patients undergoing low-risk operations do not require any preoperative cardiac testing.

• 1.

  Vascular surgery: Highest risk category. Associated with cardiac morbidity rates greater than 5%. Examples include aortic and peripheral vascular surgery. Endovascular surgery is included in this category.

• 2.

  Intermediate-risk surgery: Cardiac morbidity rates range from 1% to 5%. Examples include abdominal and thoracic procedures, carotid endarterectomy, orthopedic surgery, and head and neck surgery.

• 3.

  Low-risk surgery: Cardiac morbidity rates are generally less than 1%. Examples include endoscopic, superficial soft tissue, cataract, breast, and ambulatory operations.

Preoperative Testing (Electrocardiogram, Echocardiography, Stress Test, Angiography)

For patients at low risk for perioperative cardiac complications based on surgical and clinical risk factors, no further testing is indicated prior to surgery. For patients with known risk factors for coronary artery disease, the ACC/AHA 2014 guidelines provide a useful stepwise approach to further preoperative testing ( Fig. 10.3 ). First, the surgeon determines the urgency of the operation and identifies patient cardiac risk factors or known coronary artery disease. Any emergent operation should proceed, using patient risk factors to guide perioperative monitoring and management. Second, in cases of urgent or elective surgery, the patient should be assessed for acute coronary syndrome and, if suspected, referred for cardiology evaluation as appropriate. An important component of this assessment is an estimation of patient functional capacity, classically measured in metabolic equivalents of task (METs). An extensive collection of METs for common activities has been compiled by Ainsworth and colleagues. Representative examples are listed in Table 10.4 . ACC/ACH guidelines recommend that patients with METs ≥4 without symptoms of cardiac disease proceed with elective or urgent operation. Third, in the absence of acute coronary syndrome, additional testing is pursued on the basis of the combined clinical and surgical risk factors listed above, taking into account baseline functional capacity. Any patient undergoing a low-risk operation—regardless of clinical risk factors even with functional capacity <4 METs—has a low risk for cardiac complication and does not require further testing.

Table 10.4

Estimated MET requirements for various activities.

Modified from Hlatky MA, Boineau RE, Higginbotham MB, et al. A brief self-administered questionnaire to determine functional capacity (the Duke Activity Status Index). Am J Cardiol . 1989;64:651–654, copyright 1989 with permission from Elsevier; and adapted from Fletcher GF, Balady G, Froelicher VF, et al. Exercise standards. A statement for healthcare professionals from the American Heart Association. Writing Group. Circulation . 1995;91:580–615.)

	Can you:		Can you:
1 MET 4 METs	Take care of yourself? Eat, dress, or use the toilet?	4 METs Greater than 10 METs	Climb a flight of stairs or walk up a hill? Walk on level ground at 4 mph (6.4 kph)?
	Walk indoors around the house?		Run a short distance?
	Walk a block or two on level ground at 2 to 3 mph (3.2–4.8 kph)?		Do heavy work around the house like scrubbing floors or lifting or moving heavy furniture?
	Do light work around the house like dusting or washing dishes?		Participate in moderate recreational activities like golf, bowling, dancing, doubles tennis, or throwing a baseball or football?
			Participate in strenuous sports like swimming, singles tennis, football, basketball, or skiing?

kph , Kilometers per hour; MET , metabolic equivalent of the task; mph , miles per hour.

For patients undergoing an operation that is not low risk, a 12-lead electrocardiogram (ECG) is indicated for patients with known coronary disease, arrhythmia, peripheral artery disease, and cerebrovascular disease. Even for asymptomatic patients, ECG may be considered, except for those undergoing a low-risk operation. Assessment of left ventricular function through echocardiography is reasonable for patients with dyspnea of unknown origin or progressive heart failure. For patients with known left ventricular dysfunction, a preoperative echocardiogram should be considered if there has not been an assessment within 1 year preceding surgery or they have a decrement in functional status or change in symptoms. Exercise testing with cardiac imaging may be indicated for patients with elevated risk and poor (<4 METs) or unknown functional capacity if patients have three or more risk factors. Patients fitting these criteria who are unable to complete exercise testing may be referred for pharmacologic stress testing, either through dobutamine stress echocardiography or stress myocardial perfusion imaging. Importantly, all of the aforementioned tests should be pursued only if there is a realistic likelihood that the obtained data could change management.

Pulmonary System

Postoperative pulmonary complications occur in approximately 6% of patients after major abdominal operations and are associated with increased mortality, ICU admission, and a greater length of stay. While the exact definition of pulmonary complication varies, the major categories include pneumonia/infection, respiratory failure requiring prolonged ventilation, exacerbation of chronic obstructive pulmonary disease (COPD), and lobar/parenchymal collapse with or without associated effusion. The American College of Physicians (ACP) provided guidelines for pulmonary complication risk assessment in 2006 based on a systematic review of patient- and procedure-related preoperative risk factors. The Assess Respiratory Risk in Surgical Patients in Catalonia (ARISCAT) study, one of the largest prospective multi-institutional studies on pulmonary complications, supplemented these guidelines in 2010 and proposed an objective scale for risk stratification ( Table 10.5 ).

Broadly, the ACP guidelines indicate that patient-related risk factors for postoperative pulmonary complications include age >50 years, ASA class 2 or above, functional dependence, hypoalbuminemia (<3.5 g/dL), COPD, and heart failure. Although COPD is consistently associated with postoperative morbidity, there is no specific level of preoperative pulmonary impairment that precludes nonthoracic surgery. In fact, congestive heart failure—especially when associated with pulmonary hypertension—is a considerably stronger predictor of postoperative pulmonary complications than severe COPD. Active smoking is associated with a moderate increase in risk of postoperative complications, and smoking cessation at least 4 weeks prior to the operation reduces the risk of complications. While there is no clear evidence supporting an association between obesity and pulmonary complications per se, both obstructive sleep apnea (OSA) and obesity hypoventilation syndrome—which often complement overweight, metabolic syndrome, and morbid obesity—are associated with pulmonary complications and death.

Procedure-related risk factors that increase the risk of pulmonary complications include vascular surgery, thoracic surgery, abdominal surgery, neurosurgery, general anesthesia, head and neck surgery, procedure duration (>3 hours), and emergency surgery. Pulmonary complications increase in likelihood the closer the surgical incision is in relation to the diaphragm. Because general anesthesia conveys a higher risk of clinically relevant pulmonary complications than regional anesthesia, the latter should be considered when possible for patients with multiple patient-related risk factors.

Appropriate preoperative pulmonary evaluation begins with a thorough history and physical exam focusing on potential patient-related risk factors. Spirometry is indicated for physiologic assessment and residual lung volume estimation preceding pulmonary resection and for patients suspected of having undiagnosed COPD. Spirometry and chest radiography should be considered in patients with a preexisting diagnosis of COPD or asthma if history and physical exam cannot determine if the patient is at their optimal baseline physiology. However, these tests should not be used in routine screening for low-risk patients or if results of tests will not affect clinical decision-making. There is no prohibitive spirometric threshold below which nonthoracic surgery is strictly contraindicated. Routine chest radiography may be indicated in patients >50 years of age who are undergoing high-risk surgery. Chest radiography is used in some patients for preoperative staging in preparation for resection of abdominal and gastrointestinal neoplasms, although computed tomography (CT) has supplanted x-ray in many instances as the imaging modality of choice for staging most malignancies. Pulse oxygen saturation is a risk factor within the ARISCAT index, and its low resource utilization allows for routine screening.

There are numerous predictive indices for pulmonary complications; the most frequently cited include the Arozullah index, the ARISCAT index, and the Gupta respiratory failure calculators. The ARISCAT index is the simplest to use, featuring seven readily available preoperative predictors in a simple score system ( Table 10.5 ). The disadvantage of the ARISCAT index is that it may overestimate postoperative complication rate as its complication definition includes minor morbidities such as small radiographic effusions and bronchospasm/wheezing. The Arozullah index, derived from a veteran population, specifically targets postoperative respiratory failure. More cumbersome for routine implementation, it includes more than 12 risk factors, some of which may not be routinely available. More recently, Gupta and colleagues developed risk calculators using the ACS NSQIP dataset with primary outcomes of postoperative respiratory failure and pneumonia; these calculators are available on the web or as a downloadable mobile app.

OSA and obesity hypoventilation syndrome deserve additional consideration. Older age, obesity, and male sex are associated with a higher prevalence of OSA. A simple STOP-BANG questionnaire has been developed to screen patients for OSA and to stratify patients into risk categories based on presence of symptoms. The eight-question scoring tool includes yes/no responses to (1) snoring, (2) daytime tiredness, (3) observation of stopped breathing or interrupted breathing during sleep, (4) high blood pressure, (5) body mass index (BMI) >35, (6) age >50, (7) neck diameter >40 cm, and (8) male gender. Patients with five or more risk factors are considered at high risk for moderate to severe OSA. If an elective operation is planned, these patients should be considered for a sleep study. If positive, they should have preoperative (continuous positive airway pressure) CPAP machine fitting and optimization. Patients requiring urgent or emergent operations are managed for OSA in the postoperative setting.

Obesity hypoventilation syndrome is defined as a combination of BMI >30 with awake PaCO ₂ >45 mm Hg indicative of hypercapnia. There are no strict guidelines for arterial gas measurements in obese patients, although the highest risk for hypoventilation are patients with BMI exceeding 50. The problem of hypoventilation can be exacerbated in these patients during the perioperative period by general anesthesia and opioids. Consideration for postoperative capnography and adherence to OSA screening and postoperative CPAP use can decrease the risk of respiratory failure and death.