Preoperative Evaluation in the 21st Century

Since publication of the American College of Cardiology/American Heart Association guideline on the perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery in 2014, preoperative assessment of functional capacity has expanded beyond a subjective assessment of metabolic equivalents to include the use of a standardized patient-reported questionnaire.⁶  The Duke Activity Status Index was found to be superior to subjective assessment, biomarkers, and formal cardiopulmonary exercise testing for the prediction of postoperative mortality and myocardial infarction within 30 days.¹⁵  The Duke Activity Status Index is a 12-question patient-reported survey of functional capacity that provides a score between 0 and 58.2.¹⁶  The Duke Activity Status Index should be interpreted based on the overall score and not translated into metabolic equivalents. A Duke Activity Status Index score of 34 or more represents the lowest risk category, while scores of less than 25 represent a higher risk of major adverse cardiac events, moderate to severe postoperative complications, and new disability.¹⁷  Although the use of the Duke Activity Status Index enhances preoperative functional capacity assessment, limitations exist due to its length, lack of broad applicability due to the inclusion of culturally specific activity questions, and a question about sexual activity. Shortened four- and five-question versions of the Duke Activity Status Index have been demonstrated to perform as well as the full-length Duke Activity Status Index score,¹⁸  and the MET-REPAIR questionnaire offers additional options for activity questions to assess functional capacity.¹⁹  Another significant change in the approach to cardiac risk assessment is the addition of cardiac biomarkers such as N-terminal pro-B-type natriuretic peptide or brain natriuretic peptide to preoperative risk assessment. Canadian and European guidelines recommend obtaining a brain natriuretic peptide on patients aged more than 65 yr or more than 45 yr with risk factors for cardiovascular disease to enhance risk prediction. A pro-B-type natriuretic peptide level of 300 ng/L or more or a brain natriuretic peptide level of 92 ml/L or more has been suggested as the threshold at which routine postoperative monitoring for major adverse cardiac events be performed.^20,21  This recommendation is based on a meta-analysis that demonstrated an increase in mortality and nonfatal myocardial infarction in patients with levels above these thresholds.^20,22  A preoperative N-terminal pro-B-type natriuretic peptide is also recommended as part of risk assessment in patients with pulmonary hypertension in combination with functional status assessment, World Health Organization class of pulmonary hypertension, and review of an echocardiogram for the presence of right ventricular dysfunction.²³ 

Symptomatology (and the absence of symptoms) is an important factor in the risk assessment process that informs the need for additional preoperative testing in patients with severe valvular disease, coronary artery disease, and heart failure. Eliciting new or worsening symptoms of chest pain or pressure, syncope or presyncope, orthopnea, and dyspnea on exertion are all concerns that warrant further assessment.⁶  Furthermore, heightened attention is given to heart failure–associated risk. The condition portends a greater risk of postoperative morbidity and mortality than coronary artery disease or poor functional capacity alone.²⁴  Symptomatic heart failure with reduced ejection fraction poses the highest risk and asymptomatic diastolic dysfunction poses less but still increased risk compared with patients without the diagnosis.⁶  Appropriate use criteria for a preoperative transthoracic echocardiogram include the detection of a new murmur, symptoms suggestive of valvular heart disease, known moderate or greater degrees of valvular disease without the presence of an echocardiogram within 1 yr, and new or increased exertional dyspnea of unknown etiology.^6,21,25 

The use of a validated risk calculator is recommended to assess risk in patients preparing for noncardiac surgery.⁶  Patients with an estimated risk of a major adverse cardiac event of less than 1% will not benefit from additional testing before surgery.⁶  Examples of recommended risk prediction models include the American College of Surgeons National Surgical Quality Improvement Program Risk Calculator,²⁶  the National Surgical Quality Improvement Program Myocardial Infarction or Cardiac Arrest calculator,²⁷  and the Revised Cardiac Risk Index.²⁹  However, these calculators exhibit intercalculator discordance, with Revised Cardiac Risk Index displaying fair to poor agreement with the other tools.²⁹   The development of myocardial injury after noncardiac surgery is now a perioperative cardiac complication of interest due to its association with increased incidence of 30-day mortality.^22,30  Inclusion of myocardial injury after noncardiac surgery as a prognostically relevant event has led to a significant underestimation of risk with the Revised Cardiac Risk Index score. Thus, the utility of the Revised Cardiac Risk Index score when used alone is reduced for preoperative decision- making.³¹  Surgery-specific risk scores, such as the Vascular Study Group of New England Cardiac Risk Index also outperform the Revised Cardiac Risk Index in specific patient populations.³²  In patients at increased risk of major adverse cardiac event, preoperative cardiac testing should only be performed if it is warranted regardless of the need for noncardiac surgery, due to the lack of evidence that further diagnostic or prognostic evaluation and intervention improves surgical outcomes.³³  These conclusions cannot be extrapolated to significant left main coronary artery disease due to exclusion from previous studies and the low prevalence (less than 6%) of this finding in patients who undergo preoperative cardiac testing.³⁴ 

Not all patients who report chest pain warrant preoperative stress testing.^35,36  The pretest probability of coronary artery disease is lowest in female patients who are less than 40 yr of age with nonanginal chest pain and highest in male patients more than 60 yr of age with typical angina.³⁷  Although the coronary artery calcium score has been demonstrated to add additional discrimination for general cardiac risk assessment, the benefits are often outweighed by the costs, rate of incidental noncardiac findings, and risk of radiation.³⁸  Although coronary computed tomographic angiography may improve risk estimation of cardiac mortality and myocardial infarction when combined with the Revised Cardiac Risk Index score, the test has also been demonstrated to overestimate risk by 5-fold.³⁹  Thus, the role of coronary computed tomographic angiography remains uncertain. In patients without left main or three-vessel coronary artery disease who are undergoing noncardiac surgery, routine coronary revascularization is not recommended solely to reduce perioperative cardiac risk, because it has not demonstrated a reduction in postoperative complications.^33,40 

Hypertension is prevalent in preoperative patients and is associated with other cardiovascular and renal risk factors. Patients should be screened for sequelae of high blood pressure (stroke, heart failure, chronic kidney disease, or peripheral vascular disease). Patients with inadequately controlled hypertension exhibit greater hemodynamic lability intraoperatively than patients with well-controlled hypertension.⁴¹  Definitive preoperative blood pressure targets have not yet been established. However, in patients with systolic blood pressure of 180 mmHg or more or diastolic blood pressure of 110 mmHg or more, consideration should be given to deferring surgery until blood pressure has been addressed due to evidence of increased risk of perioperative myocardial infarction, neurologic complications, renal failure, and 90-day mortality compared with patients with better preoperative blood pressure control.^39,41 

Perioperative risk assessment and decision-making related to cerebrovascular accidents have evolved. Previous studies suggested that perioperative risk of stroke and major adverse cardiac events remained elevated for up to 9 months after stroke⁴² ; however, current American College of Cardiology/American Heart Association guidelines suggest that it may be reasonable to proceed with planned surgery at 6 months after cerebrovascular accident.⁴³  By contrast, results of a recent cohort study of more than 5 million patients suggested that the risk of postoperative stroke and death in patients after cardiovascular accident plateau at 3 months, making it reasonable to proceed with surgery any time thereafter.⁴⁴  Risk of major adverse cardiac events was not assessed in this study. Postoperative stroke risk may be predicted with high accuracy using a web-based risk prediction tool that includes factors such as age, history of cerebrovascular or cardiovascular disease, the surgery type and emergency status, American Society for Anesthesiologists (ASA) Physical Status score, hematocrit, serum creatinine, and sodium.⁴⁵  Anticoagulants and antiplatelet agents should be discontinued for as brief a period as possible (if at all) to reduce the risk of perioperative stroke. Attention to glycemic control and avoidance of hypotension are other important, potentially modifiable factors associated with perioperative stroke risk.³⁹ 

Postoperative pulmonary complications are associated with an additional $25,000 in cost per episode and are responsible for 25% of postoperative deaths within 1 week after bowel surgery.⁴⁶  The definition of postoperative pulmonary complications spans a wide range of diagnoses, including bronchospasm, atelectasis, pneumonia, exacerbation of underlying lung conditions, prolonged mechanical ventilation, and respiratory failure. The incidence is highest in upper abdominal and thoracic surgery.⁴⁷ 

Pulmonary function testing is rarely indicated as part of the preoperative evaluation, except in the context of major lung surgery or new, unexplained dyspnea. The most commonly used tool to predict the risk of postoperative pulmonary complications is the Assess Respiratory Risk in Surgical Patients in Catalonia score.⁴⁸  The value of the scoring tool lies in its brevity and use of only preoperative variables to assign patients to either low (1.6%), intermediate (13.3%), or high risk (42.1%). Factors include duration of surgery, location of the surgical incision, age, preoperative oxygen saturation, respiratory infection in the past month, presence of anemia, and whether or not the procedure is emergent. Other factors associated with increased risk of postoperative pulmonary complications include obesity, ASA Physical Status score 3 or more, poor preoperative functional status, low serum albumin, and a diagnosis of heart failure, chronic kidney disease, chronic respiratory disease, pulmonary hypertension, or obstructive sleep apnea.⁴⁷ 

Obstructive sleep apnea (OSA) is a common, yet frequently underrecognized condition that results in sleep fragmentation, endothelial dysfunction, hypercoagulability, and increased sympathetic activity. Ninety percent of people with moderate to severe OSA remain undiagnosed.⁴⁹  This condition increases the risk of experiencing cardiovascular complications in the perioperative period⁵⁰  as well as the risk of developing chronic medical conditions that impact long-term health, such as hypertension, atrial fibrillation, myocardial infarction, heart failure, and pulmonary hypertension.⁵¹  A formal diagnosis of OSA is determined based on the apnea/hypopnea index on a polysomnogram, with more than 30 events per hour indicating severe OSA.⁴⁹ 

A commonly used screening tool for OSA in the perioperative period is the STOP-Bang score—where scores of 5 or more indicate that the patient is at increased risk of having moderate to severe OSA.⁵²  The STOP-Bang includes eight equally weighted factors: loud snoring, reported tiredness during the day, observed apnea, hypertension, body mass index, age, neck circumference, and sex. The Society for Anesthesia and Sleep Medicine guidelines recommend routine screening for OSA as a standard component of the preanesthesia evaluation; however, the society recommends against delaying surgery for diagnosis or treatment due to the lack of definitive evidence of benefit.⁴⁹  Perioperative adverse events, including opioid-related ventilatory depression, are more common in the undiagnosed (suspected severe) and patients nonadherent with continuous positive airway pressure therapy compared with those using continuous positive airway pressure.⁵³  Therefore, patients should be encouraged to bring their home continuous positive airway pressure device to the hospital when possible, adhere to continuous positive airway pressure therapy in the perioperative period, and be considered for assignment to a postoperative ward with continuous pulse oximetry.⁵⁴  Concomitant uncontrolled clinical conditions should be addressed and optimized before proceeding with surgery. Interestingly, no association has been noted between OSA (even severe) and postoperative delirium.^50,55 

The preoperative period represents a unique window of opportunity for patients who smoke tobacco to achieve abstinence for both the perioperative period and long term.⁵⁶  Surgical patients who are current smokers are at significantly higher risk for complications including 20% higher perioperative mortality and a 40% increase in complications, including cardiac events, wound infections, pulmonary complications, and venous thromboembolism compared with never-smokers and ex-smokers.^57,58  Patients should be encouraged to quit at any time before surgery. The benefits of even ultra-short-term abstinence (the morning of surgery) are associated with a reduction in pulmonary complications compared with active smokers.^59,60  A reduction in carbon monoxide concentrations is apparent within 4 h, with an improvement in immune system function at 4 weeks and improved wound healing at 12 weeks. Most effective programs include a combination of both pharmacotherapy and behavioral modification/stress management.⁵⁶  Nicotine replacement therapy has been established as both safe and effective in the perioperative period despite previous concerns about decreased tissue perfusion and wound healing.⁶¹  Furthermore, engagement with a smoking cessation program while hospitalized is associated with a reduction in readmission to the hospital, regardless of the success of the quit attempt.⁶² 

Perioperative mortality and morbidity are associated with both dysglycemia in the immediate perioperative period and poor long-term glycemic control.⁶³  Improved glycemic control may lead to mitigation of both cost^64,65  and postoperative complications.⁶⁵ 

Sustained hyperglycemia, triggered by the stress response of surgery, may impair neutrophil function and increase inflammatory cytokine production. Compared with euglycemic patients, higher rates of postoperative infection are noted in patients with mean postoperative serum glucose concentrations of 150 to 250 mg/dL and are highest in those with mean serum glucose concentrations greater than 250 mg/dL.⁶⁶  Hyperglycemia has also been associated with an increased risk of acute kidney injury, cardiovascular complications, and prolonged length of stay.⁶⁷  A definitive recommendation for the threshold at which surgery should be deferred for better glycemic control remains uncertain and may be both surgery and patient specific. The primary aim is to achieve glycemic control with a target range of 110 to 180 mg/dL to prevent adverse perioperative outcomes. However, the best predictor of intraoperative hyperglycemia may be poor longitudinal glycemic control,⁶⁸  and elevated A1c levels have also been associated with an increase in infectious complications and mortality.⁶³  Although the optimal hemoglobin A1c threshold to reduce perioperative risk remains unclear, an A1c less than 8% is a commonly recommended target across surgical populations,⁶⁹  with the Society for Ambulatory Anesthesia (Chicago, Illinois) citing a goal of A1c less than 7% for ambulatory procedures.⁷⁰ 

Because 10 to 15% of adult patients with diabetes are not formally diagnosed, and morbidity and mortality from hyperglycemia are greater when compared with patients with a diagnosis of diabetes,⁷¹  preoperative hemoglobin A1c screening in patients with risk factors for diabetes is appropriate. Patients with diabetes are more likely to have associated comorbidities such as coronary artery disease, gastroparesis, chronic kidney disease, obesity, and hypertension.⁷²  Furthermore, engagement with an endocrinologist or perioperative optimization program may lead to more frequent glucose checks, medication adjustment, or other preoperative interventions that lead to better perioperative control and the opportunity to establish long-term diabetes care.⁷³  Hypoglycemic medications that should be held on the day of surgery include α-glucosidase inhibitors, biguanides, dipeptidyl peptidase 4 inhibitors, insulin secretagogues, and thiazolidinediones.^72,74  Other classes of diabetes medications require a longer discontinuation timeframe: sodium-glucose cotransporter-2 inhibitors should be discontinued 3 days before surgery (4 days for ertugliflozin) due to the risk of euglycemic diabetic ketoacidosis.⁷⁴  Optimal perioperative management of glucagon-like peptide 1 agonists remains uncertain due to lack of evidence. Although recommendations are institution-specific, intraoperative use of the patient’s insulin pump with frequent glucose monitoring is often best, with the pump set to the patient’s lowest basal rate the morning of surgery.^74,75 

Malnutrition affects approximately 50% of patients presenting for major surgery, with the highest prevalence in patients presenting for gastrointestinal surgery.⁷⁶  Despite its association with impaired wound healing, postoperative infection, delirium, prolonged hospitalization, higher costs of care, and mortality, it remains unrecognized and untreated in all but 10% of cases.⁷⁷  All patients should be assessed for risk of malnutrition preoperatively, especially those with frailty, via a validated malnutrition screening tool. The primary aim of preoperative nutritional optimization is to address malnutrition with oral protein supplementation and to provide a specific regimen of immunonutrition in select gastrointestinal surgical populations.⁷⁶ 

Preoperative anemia is common in many surgical populations, with the highest prevalence in orthopedic, gynecologic, and colorectal surgical patients.⁷⁸  Even mild preoperative anemia is independently associated with an increased risk of 30-day mortality and morbidity after noncardiac surgery. Risks include acute kidney injury, major cardiac events, and prolonged hospitalization. The risk is not mitigated by blood transfusion, and anemia often remains unaddressed before surgery.⁷⁹ 

A common cause of preoperative anemia is iron deficiency. Diagnosis of iron deficiency anemia is made by a ferritin level less than 100 mcg/L, or transferrin saturation less than 20%, or reticulocyte hemoglobin content less than 30 pg.⁸⁰  Supplementation with intravenous iron is indicated in scenarios where oral iron is not tolerated or is ineffective due to inflammation, in situations where absorption via the gastrointestinal tract is impaired, and in blood refusal patients when supranormal hemoglobin targets are indicated. A short timeframe and large dose required for total repletion of iron stores will limit the effectiveness of oral administration due to rate limitations of gastrointestinal absorption. Many safe and effective preparations of intravenous iron exist. The main difference among them is the allowable dose per infusion. Iron deficiency may still be present with anemia of inflammation and chronic kidney disease; however, intravenous iron alone may be minimally effective unless combined with an erythropoiesis- stimulating agent. Adequate repletion of folate or B12 is the treatment for other nutritional causes of anemia.⁸⁰ 

Frailty is a state of diminished resilience to stress that spans multiple domains.⁸¹  This aggregate expression of low reserve may develop due to chronic illness, low physical activity, nutritional deficits, cognitive dysfunction, psychologic distress, poor social support, or low income and literacy levels.⁸²  Frailty syndrome is associated with an increased risk of mortality, discharge to a nonhome location, and delirium.⁸¹  Tools to screen for and diagnose frailty derive from either the frailty index (deficit accumulation model) or the frailty phenotype (energy depletion model).⁸²  Many screening tools exist for use as part of the preoperative evaluation. The most common frailty screening tools used in a preoperative clinic setting include the Clinical Frailty Scale and the FRAIL scale.⁸²  The Clinical Frailty Scale derives from the frailty index and requires the clinician to assign the patient a score from 1 (fittest for age) to 9 (terminally ill) based on both physical assessment and patient report of weakness and need for assistance with activities of daily living. The FRAIL scale, derived from the frailty phenotype model, is recommended by the American Geriatric Society and is in wide use in the nonsurgical setting.^81,83  Patients’ responses to questions about weight loss, walking speed, multiple illnesses, and energy level are used to determine the score. In a recent comparison of common perioperative frailty instruments, the Clinical Frailty Scale was determined to be both the most predictive of major postoperative complications and the most feasible for use in the preoperative clinic setting.⁸⁴  The goal of preoperative frailty screening is to identify patients who will benefit from formal assessment and focused interventions such as geriatric comanagement to reduce postoperative delirium and other adverse events related to frailty.

The most common screen for cognition and risk for delirium used in the preoperative evaluation is the Mini-Cog.⁸⁵  This instrument is a nondiagnostic tool derived from the Montreal Cognitive Assessment. It consists of a clock drawing and a three-word recall. Scores of 2 or less are associated with an increased risk of delirium.

Preoperative anxiety and depression have been associated with poorer postoperative outcomes, including longer length of hospitalization and mortality. Thus, the assessment of psychologic state is an important component of the preoperative evaluation.^86,87  Common screening tools in use in the preoperative clinic setting include the Hospital Anxiety and Depression Scale, the Patient Health Questionnaire, the 7-item Generalized Anxiety Disorder instrument, the Pain Catastrophizing Scales, and the distress thermometer. Patients with high degrees of psychologic distress should be offered preoperative interventions aimed at stress reduction, such as listening to pre-recorded music at a specific tempo, relaxation exercises, and cognitive behavioral therapy.⁸⁸ 

Shared decision-making is another key component for the delivery of patient-centered perioperative care. Perioperative decisions are often nuanced and complex. Taking a shared decision-making approach to align the decision to proceed with surgery (or not) and the perioperative plan with the patient’s overarching health goals is associated with greater patient satisfaction and reduced anxiety.⁸⁹ 

Multidimensional syndromes such as frailty are most successfully addressed with a patient-specific, multifaceted approach. Multimodal interventions aimed at enhancing the patient’s physical and psychologic resilience are known as prehabilitation. Traditional components of prehabilitation include nutritional optimization, physical exercise, and psychologic preparation.⁹⁰  Prehabilitation interventions are often resource and time intensive, and further research on optimal timing, duration, and combination of interventions is warranted.

Further research is also needed to identify specific interventions to successfully mitigate perioperative risk related to deficiencies in the patient’s social, economic, and physical environment. Achievement of optimal outcomes in vulnerable patient populations will likely require an emphasis on early and frequent care coordination initiatives.

The preoperative evaluation will likely continue to evolve in response to the expanding role and scope of anesthesiologists across the perioperative continuum.

The author thanks Jerrold H. Levy, M.D., F.A.H.A., F.C.C.M., Professor of Anesthesiology and Critical Care Professor of Surgery (Cardiothoracic) Duke University Hospital (Durham, North Carolina), for his review of the manuscript before submission.

Support was provided solely from institutional and/or departmental sources.

Dr. Blitz is president of the Society for Preoperative Assessment and Quality Improvement (Glenview, Illinois; www.spaqi.org). She has served on the advisory board for the Society for the Advancement of Patient Blood Management (Mt. Royal, New Jersey). She has received fees related to consultancy for Guidepoint (New York, New York) and lecturing at Providence Anesthesiology Associates’ annual conference (Charlotte, North Carolina).