INTRODUCTION — Many patients undergoing major noncardiac surgery are at risk for a cardiovascular event. The risk is related to patient- and surgery-specific characteristics. Identification of increased risk provides the patient (and surgeon) with information that helps them better understand the benefit-to-risk ratio of a procedure and may lead to interventions that decrease risk.
This topic will review the initial preoperative cardiac evaluation, which includes an attempt to quantify risk. The management of cardiac risk (in an attempt to reduce morbidity and mortality) and issues related to the perioperative evaluation and management of heart failure or myocardial infarction (MI) are discussed separately. (See "Management of cardiac risk for noncardiac surgery" and "Perioperative myocardial infarction or injury after noncardiac surgery" and "Perioperative management of heart failure in patients undergoing noncardiac surgery".)
INCIDENCE — The incidence of an adverse cardiovascular outcome is related to the baseline risk.
A 1995 review of major published series found that the pooled average rates of selective outcome of myocardial infarction (MI) and cardiac death varied with the population studied [1]:
●Among unselected surgical patients over age 40 – Perioperative MI in 1.4 percent and cardiac death in 1.0 percent.
●Among consecutive surgical patients with some selection criteria – Perioperative MI in 3.2 percent and cardiac death in 1.7 percent.
In a retrospective study of 663,635 adults not taking beta blockers who underwent major noncardiac surgery in 2000 and 2001, in-hospital mortality increased progressively from 1.4 to 7.4 percent according to a preoperative assessment of risk using the revised cardiac risk index (RCRI) described below (table 1) [2]. (See 'Revised cardiac risk index' below.)
A 2016 study, using information in a large administrative database of United States hospital admissions (2004 to 2013), found a 3 percent incidence of major adverse cardiovascular and cerebrovascular events (in-hospital, all-cause death, acute MI, or acute ischemic stroke) [3]. These events were most common after vascular, thoracic, and transplant surgery.
Patients with underlying cardiovascular disease, including peripheral artery disease or stroke, have an increased risk of perioperative cardiac complications compared with patients without extant atherosclerosis for two reasons:
●They constitute a selected population with a high incidence of significant coronary artery disease [4,5]. In addition, left ventricular systolic dysfunction (left ventricular ejection fraction ≤40 percent) is five times more common in patients with cerebrovascular disease or peripheral artery disease compared with matched controls [6].
●Physiologic factors associated with surgery predisposed to myocardial ischemia, which is more pronounced in patients with underlying coronary disease. These include volume shifts and blood loss, enhanced myocardial oxygen demand from elevations in heart rate and blood pressure secondary to stress from surgery, and an increase in postoperative platelet reactivity [7].
Despite the increased risk in the population with vascular disease, the rates of MI and death have decreased over time. For example, perioperative mortality after carotid endarterectomy is approximately 1 percent and for abdominal aortic aneurysm repair is <3 percent [8].
OUR APPROACH — All patients scheduled to undergo noncardiac surgery should have an assessment of the risk of a cardiovascular perioperative cardiac event [9,10]. An algorithm developed for use in patients with known coronary artery disease or at high risk can be used for this purpose (algorithm 1). The rationale for this recommendation and a detailed discussion of risk assessment tools (models) are presented below. (See 'Risk assessment' below.)
Very high-risk patients and those undergoing emergency or urgent surgery are approached somewhat differently.
Very high-risk patients — Patients with recent MI (60 days) [11] or unstable angina, decompensated heart failure, high-grade arrhythmias, or hemodynamically important valvular heart disease (aortic stenosis in particular [12]) are at very high risk for perioperative MI, heart failure, ventricular fibrillation or primary cardiac arrest, complete heart block, and cardiac death. All such patients should be optimally treated, with possible referral to a cardiologist for further evaluation and management. (See "Noncardiac surgery in adults with aortic stenosis" and "Perioperative management of heart failure in patients undergoing noncardiac surgery", section on 'Risk classification by heart failure syndrome'.)
Emergency or urgent surgery — Patients who require emergency or urgent surgery are at increased risk of a perioperative cardiovascular event at any level of baseline risk. In these cases, risk indices derived from elective surgery cohorts are not accurate, although they may provide an estimate of the minimal risk.
In many cases, there is not sufficient time for an extensive evaluation of the severity of a patient’s cardiovascular problem, and in most cases the benefit of proceeding with surgery outweighs the risk of waiting to perform additional testing. In the absence of preoperative assessment because of the minimal time available before surgery, clinicians must be available postoperatively to help manage the possible cardiovascular complications in at-risk patients.
INITIAL PREOPERATIVE EVALUATION — Once a determination is made that noncardiac surgery will be considered, the patient should be evaluated for the risk of a cardiovascular complication. This evaluation is generally performed by a primary care clinician. The information obtained is used to assess risk. In patients assessed to be at elevated (intermediate or high) cardiovascular risk, a referral to a cardiologist for further evaluation may be indicated.
At the time of the initial preoperative evaluation, the clinician should inquire about symptoms such as angina, dyspnea, syncope, and palpitations as well as a history of heart disease, including ischemic, valvular, or cardiomyopathic disease, and a history of hypertension, diabetes, chronic kidney disease, and cerebrovascular or peripheral artery disease. The role of functional status as traditionally assessed by the clinician is discussed below. (See 'Functional status/capacity' below.)
The physical examination should focus on the cardiovascular system and include blood pressure measurements, auscultation of the heart and lungs, abdominal palpation, and examination of the extremities for edema and vascular integrity. Important findings include evidence of heart failure or a murmur suspicious for hemodynamically significant valvular heart disease. (See 'Risk factors' below.)
Similar to recommendations made in the 2014 American College of Cardiology/American Heart Association (ACC/AHA) and European Society of Cardiology/European Society of Anesthesiology (ESC/ESA) guidelines on noncardiac surgery, we obtain an electrocardiogram (ECG) in many patients with known cardiovascular disease, significant arrhythmia, or significant structural heart disease unless the patient is undergoing low-risk surgery (surgery associated with less than 1 percent morbidity/mortality such as ambulatory surgery) [9,10,13]. A preoperative ECG can be obtained in asymptomatic patients without known cardiovascular disease, but it is rarely helpful. Some ECG abnormalities seem to be associated with a worse prognosis in observational studies, but the association is inconsistent across studies. ECG abnormalities are not part of either the revised cardiac risk index (RCRI) or the National Surgical Quality Improvement Plan (NSQIP) because of the lack of prognostic specificity associated with these findings. (See 'Risk assessment' below.)
The rationale for obtaining a preoperative ECG comes from the utility of having a baseline ECG should a postoperative ECG be abnormal.
For those patients who receive a preoperative ECG, it should be evaluated for the presence of Q waves or significant ST-segment elevation or depression, which raises the possibility of myocardial ischemia or infarction, left ventricular hypertrophy, QTc prolongation, bundle-branch block, or arrhythmia [14].
Functional status/capacity — Cardiac functional status or capacity, as determined by doctors assessing patients with a brief set of questions, has been thought to be positively associated with postoperative outcomes. This assessment has been included in many risk models. None of the risk models discussed below are optimally constructed. We use the more standardized Duke Activity Status Index (DASI) standardized questionnaire for many patients. (See 'Risk assessment' below.)
Functional status can be expressed in metabolic equivalents (1 MET is defined as 3.5 mL O2 uptake/kg per min, which is the resting oxygen uptake in a sitting position). The ability to achieve four METs of activity without symptoms is thought to be a good prognostic indicator [15].
Various activity scales provide the clinician with a set of questions to determine a patient's functional capacity [16]. Indicators of functional status include the following:
●Can take care of self, such as eat, dress, or use the toilet (1 MET)
●Can walk up a flight of steps or a hill or walk on level ground at 3 to 4 mph (4 METs)
●Can do heavy work around the house, such as scrubbing floors or lifting or moving heavy furniture, or climb two flights of stairs (between 4 and 10 METs)
●Can participate in strenuous sports such as swimming, singles tennis, football, basketball, and skiing (>10 METs)
The 2018 METS prospective cohort study concluded that subjectively assessed preoperative functional capacity did not accurately identify patients with poor cardiopulmonary fitness or predict postoperative morbidity or mortality [17]. In this study of 1401 patients scheduled for major noncardiac surgery and who had one or more risk factors for cardiac complications, the predictive ability of the subjective assessment of functional capacity (in METS) was compared with the Duke Activity Status Index (DASI) standardized questionnaire(table 2), formal cardiopulmonary exercise testing (CPET), and measurement of N-Terminal pro-brain natriuretic peptide concentrations. These three comparators have been previously validated as capable of predicting postoperative cardiovascular events, although CPET may offer some advantages for predicting all-cause complications as opposed to cardiovascular events alone (see "Natriuretic peptide measurement in non-heart failure settings", section on 'Postoperative complications'). The primary outcome was death or myocardial infarction within 30 days after surgery. Subjective assessment of functional capacity had a 19.2 percent sensitivity and a 94.7 percent specificity for predicting the inability to attain four metabolic equivalents during CPET. Stated another way, subjective assessment resulted in a substantial misclassification of high-risk patients as low risk [18]. Only DASI scores were associated with successfully predicting the primary outcome (adjusted odds ratio 0.96, 95% CI 0.83-0.99). A DASI score of <34 was associated with an increased risk of 30-day death, myocardial infarction, and moderate to severe complications [19]. If a 50 percent increased rate of events is considered clinically (although not statistically) significant, then a DASI of <25 points (approximately 4 METS) would be a second cutoff to identify the subgroup in whom testing should be considered if it would change management. Peak oxygen consumption was associated with moderate to severe complications, but neither it nor anaerobic threshold was predictive of the primary outcome.
The results of the METS study make us less confident that risk assessment models/tools/algorithms that include subjective assessment of functional capacity are optimally constructed (see 'Risk assessment' below). However, in a study of patients at high cardiovascular risk undergoing noncardiac surgery, self-reported functional capacity of less than two flights of stairs was independently associated with major adverse cardiac events and all-cause mortality at 30 days and one year. The addition of self-reported functional capacity to surgical and clinical risk improved risk classification [20].
RISK FACTORS — The following clinical and surgery-specific factors have been associated with an increase in perioperative risk of a cardiovascular event and are used in one or both of the models discussed below (revised cardiac risk index [RCRI] or the myocardial infarction or cardiac arrest [MICA] calculator derived from the American College of Surgeons surgical risk calculator [ACS-SRC]). The newer ACS-SRC includes 20 patient risk factors in addition to the surgical procedure (see 'Risk assessment' below):
●Surgery-specific risk (RCRI and ACS-SRC) – The reported rate of cardiac death or nonfatal myocardial infarction (MI) is more than 5 percent in high-risk procedures, between 1 and 5 percent in intermediate-risk procedures, and less than 1 percent in low-risk procedures (table 3). Institutional and/or individual surgeon experience with the procedure may increase or lower the risk. Emergency surgery is associated with particularly high risk, as cardiac complications are two to five times more likely than with elective procedures (table 4). This issue is discussed in greater detail separately. (See "Preoperative evaluation for anesthesia for noncardiac surgery", section on 'Surgical risk'.)
●History of ischemic heart disease (RCRI).
●History of heart failure (RCRI).
●History of cerebrovascular disease (RCRI).
●Insulin dependent diabetes mellitus (RCRI).
●Preoperative serum creatinine ≥2.0 mg/dL (RCRI) or >1.5 mg/dL (ACS-SRC).
●Increasing age (ACS-SRC).
●American Society of Anesthesiologist class (ACS-SRC).
●Preoperative functional status (ACS-SRC). (See 'Functional status/capacity' above.)
While not included in the risk factors above, the following patient characteristics have been associated with increased risk:
●Atrial fibrillation – A retrospective, administrative database study demonstrated an association between a history of prior admission for atrial fibrillation and postoperative complications [21]. The risk associated with atrial fibrillation was higher than that associated with a diagnosis of coronary artery disease.
●Obesity – Patients with obesity are at increased risk for adverse cardiovascular events at the time of noncardiac surgery. However, obesity has not been shown to be a predictor independent of end-organ damage. (See "Obesity: Association with cardiovascular disease".)
The issue of whether the preoperative approach to patients with obesity should differ from that in the general population is uncertain [22].
RISK ASSESSMENT
Overview — All patients scheduled to undergo noncardiac surgery should have an initial assessment of the risk (in percent) of a cardiovascular perioperative cardiac event using validated models that typically include information from the history, physical examination, electrocardiogram, and type of surgery [9,10] (see 'Risk factors' above). The purpose of this assessment is to help the patient and health care providers weigh the benefits and risks of the surgery and optimize the timing of the surgery. On occasion, risk assessment will uncover undiagnosed problems or suboptimally treated prior conditions that need attention.
There is wide variability in the predicted risk of cardiac complications using different risk-prediction tools as each was developed in different populations undergoing various surgical procedures and had different definitions of risk factors, postoperative complications, and timeframes for follow-up, so there cannot be a valid comparison [23,24]. Additionally, subsequent studies evaluating these tools also often used them in a different setting than originally derived (specific patient or procedure groups, eg, older adults, cardiac history or risk factors, vascular or orthopedic surgery) or outcomes not in the original models (eg, myocardial injury, stroke, or overall mortality, urgent/emergency surgery). We caution users of commercially available online calculators as they may overestimate predicted risk compared with one of our recommended calculators using original study data.
We use either the revised cardiac risk index (RCRI), also referred to as the Lee index (table 1) [25], or the American College of Surgeons surgical risk calculator (ACS-SRC) [26]. The RCRI is simpler and has been widely used and validated over the past 20 years. The ACS-SRC is more complex, requiring calculation through an online tool, and has yet to be validated in other populations. A simpler tool also derived from the National Surgical Quality Improvement Program (NSQIP) database is the myocardial infarction or cardiac arrest (MICA) calculator. The MICA calculator outperformed the RCRI in some circumstances, and the newer ACS-SRC is more comprehensive and procedure-specific. However, as mentioned previously, neither of these has yet to be prospectively validated. Furthermore, direct comparison of the various tools is difficult due to their different definitions of risk factors, complications, and outcomes [27]. Therefore, the optimal tool incorporates local factors into the preoperative workflow. In addition, based on the 2018 METS study, we have concerns regarding the predictive ability of these tools (see 'Functional status/capacity' above). Practitioners should become familiar with one model and use it regularly.
These models provide the user with the risk of a cardiac complication in percent (table 5).
We do not recommend using older models such as the original Goldman cardiac risk index [28], the Detsky modified risk index, or the Eagle criteria [29-34]. Risk assessment should include information from the chosen scoring system with the inherent risk of the surgery.
The value of all these risk indicators (models) may be diminishing over time, as the cardiovascular risk of surgery is declining [35]. This may result from the changing nature of postoperative myocardial infarction (MI), from a type 1, plaque rupture, MI to a type 2, hemodynamic MI. In the POISE trial of 8351 patients at high risk for or with atherosclerosis undergoing noncardiac surgery, only 35 (0.4 percent) required coronary revascularization postoperatively [36]. Thus, the value of risk prediction models may be waning as the original end point of interest decreases and newer studies are focusing on the endpoint of myocardial injury after noncardiac surgery. (See "Perioperative myocardial infarction or injury after noncardiac surgery", section on 'Definitions of myocardial infarction and myocardial injury'.)
MICA NSQIP database risk model — The NSQIP database was used to determine risk factors associated with intraoperative/postoperative MI or cardiac arrest [37]. Among over 200,000 patients who underwent surgery in 2007, 0.65 percent developed perioperative MI or cardiac arrest. On multivariate logistic regression analysis, five factors were identified as predictors of MI or cardiac arrest:
●Type of surgery
●Dependent functional status
●Abnormal creatinine
●American Society of Anesthesiologists’ class (table 6)
●Increased age
A risk model was developed using these five factors and subsequently validated on a 2008 data set (n = 257,385). The risk model had a relatively high predictive accuracy (C statistic of 0.874) and outperformed the RCRI (C statistic of 0.747). An easy-to-use calculator was developed from this model.
Revised cardiac risk index — The RCRI was published in 1999 and has been used worldwide since then [25]. In the derivation of the index, 2893 patients (mean age 66) undergoing elective major noncardiac procedures (with an expected length of stay >2 days) were monitored for major cardiac complications (cardiac death, acute MI, pulmonary edema, ventricular fibrillation/cardiac arrest, and complete heart block) (table 1). The index was validated in a cohort of 1422 similar individuals. The predictive value was significant in all types of elective major noncardiac surgery except for abdominal aortic aneurysm surgery (figure 1).
A 2009 systematic review evaluated the ability of the RCRI to predict cardiac complications and mortality after major noncardiac surgery in various populations and settings [38]. The RCRI performed moderately well in distinguishing patients at low compared with high risk for all types of noncardiac surgery but was somewhat less accurate in patients undergoing only vascular noncardiac surgery. In addition, RCRI did not predict all-cause mortality well. However, this is expected, as it does not capture risk factors for noncardiac causes of perioperative mortality and only one-third of perioperative deaths are due to cardiac causes.
The risk of major cardiac complications (cardiac death, nonfatal MI, nonfatal cardiac arrest, postoperative cardiogenic pulmonary edema, complete heart block) varied according to the number of risk factors. The following combined rates of nonfatal MI, nonfatal cardiac arrest, and cardiac death were seen in various studies [39]:
●No risk factors – 0.4 percent
●One risk factor – 1 percent
●Two risk factors – 2.4 percent
●Three or more risk factors – 5.4 percent
The percentages presented above may underestimate a risk that includes other cardiovascular outcomes such as complete heart block or heart failure.
●In a large retrospective analysis on mortality, the perioperative risk was evaluated in 663,665 adults with no contraindications to beta blockers who underwent major noncardiac surgery in 2000 and 2001 at 329 hospitals in the United States [2]. In-hospital mortality in patients not treated with beta blockers increased progressively from 1.4 percent at a score of 0 to 7.4 percent at a score ≥4. The rate of mortality in this study was higher at the same RCRI than the combined endpoint of cardiac death, nonfatal MI, and nonfatal cardiac arrest in the earlier RCRI (Goldman) population [39].
●In the PeriOperative ISchemic Evaluation (POISE) randomized trial of over 8000 patients undergoing noncardiac surgery between 2002 and 2007, the combined rate of cardiovascular death, nonfatal MI, and nonfatal cardiac arrest was 6.9 percent in the placebo group [40]. The majority of these individuals were RCRI 1 or 2. (See "Management of cardiac risk for noncardiac surgery", section on 'Beta blockers'.)
There are several factors that probably contribute to the higher event rate in these two later studies:
●The original RCRI risk prediction model did not take all-cause mortality into account [25]
●RCRI only included in-patient complications, not 30-day event rates.
●In earlier studies, creatine kinase-MB fraction was used to diagnose MI, rather than troponins, which are more sensitive. (See "Troponin testing: Analytical considerations".)
●The type of MI after surgery is changing. The incidence of a type 1, plaque rupture, MI is decreasing while type 2, hemodynamic MI is increasing.
One study reexamined the original six risk factors to confirm their validity in a large modern prospective database, including 9519 patients aged ≥50 undergoing elective non-cardiac surgery with an expected length of stay ≥2 days at two major tertiary-care teaching hospitals [41]. Compared with the RCRI, a simplified five-factor model ("reconstructed RCRI") using high-risk type of surgery, history of ischemic heart disease, congestive heart failure, cerebrovascular disease, and preoperative glomerular filtration rate (GFR) <30 mL/minute instead of serum creatinine >2 mg/dL (but not including diabetes or insulin treatment) resulted in superior prediction of major cardiac complications following elective noncardiac surgery.
AUB-POCES index (renamed AUB-HAS2) — The 2019 American University of Beirut-Pre-Operative Cardiovascular Evaluation Study (AUB-POCES) prospectively derived and validated a new preoperative cardiovascular risk index (CVRI) [42]. It was subsequently renamed AUB-HAS2 based on the six predictors of risk identified by multivariate logistic regression analysis in the derivation cohort: history of heart disease, heart symptoms of angina or dyspnea, age ≥75 years, anemia with hemoglobin <12 mg/dL, vascular surgery, and emergency surgery. Patients were assigned a score of 0, 1, 2, 3, and >3 based on the number of predictors. The incidence of the primary outcome of death, MI, or stroke at 30 days increased steadily across the increasing scores, in both the derivation (0, 0.5, 2.0. 5.6, and 15.7 percent, respectively) and validation (0.3, 1.6, 5.6, 11.0, and 17.5 percent, respectively) cohorts. The area under the receiver operator characteristic curve was significantly better than that for RCRI (0.9 versus 0.79) discussed above, although outcomes differed (see 'Revised cardiac risk index' above). CVRI used 30-day rather than in-hospital outcomes, used all-cause rather than cardiac mortality, and included stroke. Patients with emergency surgery had been excluded for derivation of RCRI.
A subsequent analysis of the performance of AUB-HAS2 in nine surgical specialty groups and eight site-specific surgeries using 1,167,278 noncardiac surgeries from the NSQIP database demonstrated superior discriminatory power compared with the RCRI [43]. An advantage of AUB-HAS2 is its ease of use. Additional validation in other populations is needed.
VSGNE risk index — As the RCRI, discussed directly above, did not perform well in patients undergoing vascular surgery, the Vascular Study Group of New England (VSGNE) developed a risk index specifically for those patients [44]. However, we do not use this index in our practices.
In multivariate analysis of the VSGNE cohort, independent predictors of adverse cardiac events (MI, arrhythmia, and heart failure, but not mortality) were increasing age (odds ratio [OR] 1.7 to 2.8), smoking (OR 1.3), insulin-dependent diabetes (OR 1.4), coronary artery disease (OR 1.4), congestive heart failure (OR 1.9), abnormal cardiac stress test (OR 1.2), long-term beta-blocker therapy (OR 1.4), chronic obstructive pulmonary disease (OR 1.6), and creatinine ≥1.8 mg/dL (OR 1.7). Prior cardiac revascularization was protective (OR 0.8). This calculator is no longer available online and has been replaced by the VQI calculators discussed directly below. The RCRI substantially underestimated in-hospital cardiac event in patients undergoing elective or urgent vascular surgery, especially after lower-extremity bypass, endovascular abdominal aortic aneurysm repair, and open infrarenal abdominal aortic aneurysm. This risk index also has not been externally validated and did not include mortality as an end point.
VQI cardiac risk index — Several models were developed to predict risk of postoperative MI/myocardial injury after noncardiac surgery (MINS) during hospitalization after various vascular surgery procedures based on 88,791 nonemergency operations from the Vascular Quality Initiative (VQI) registry. These procedures included carotid endarterectomy, infrainguinal bypass, suprainguinal bypass, endovascular aneurysm repair, and open abdominal aortic aneurysm repair. An all-procedure and four procedure-specific risk calculators were created using multivariate analysis based on a derivation cohort from 2012 to 2014 (n = 61,236) and validated using a cohort (n = 27,555) from 2015 to 2016.
Predictors of MI/MINS in the all-procedure model included age, operation type, coronary artery disease, congestive heart failure, diabetes, creatinine concentration >1.8 mg/dL, stress test status, and body mass index (area under the curve [AUC] 0.75; 95% CI, 0.73-0.76).This model was less accurate than the procedure-specific calculators which included unique predictors. These calculators are available online [45,46].
ACS surgical risk calculator — Based on 1,414,006 patients encompassing 1557 unique CPT codes, a universal surgical risk calculator model [26] has been developed using a web-based tool consisting of 20 patient factors plus the surgical procedure. This model had excellent performance for mortality (C statistic = 0.944; Brier score = 0.011 [where scores approaching 0 are better]), morbidity (C statistic = 0.816, Brier score = 0.069), and six additional complications (C statistics >0.8). While more comprehensive than the other risk calculators, it is more cumbersome, which may limit its use. In addition, this calculator has not yet been externally validated. However, it is useful as a decision aid and informed consent tool for clinicians and patients. This calculator is available online.
MANAGEMENT BASED ON RISK — We use estimated risk of a major adverse cardiac event (see 'Risk assessment' above) to categorize patients into low- or higher-risk groups (see "Management of cardiac risk for noncardiac surgery"). The risk will determine whether surgery should proceed without further cardiovascular testing; be postponed pending further testing, such as stress testing or echocardiography; be changed to a lesser risk procedure (if possible) or a nonsurgical alternative (eg, radiation and/or chemotherapy or palliative care); or be cancelled so that a procedure such as coronary revascularization or heart valve replacement can take place (algorithm 1).
Low-risk patients — Patients whose estimated risk of major adverse cardiac event is less than 1 percent are labeled as being at low risk and require no additional cardiovascular testing.
Higher-risk patients — Patients whose risk of major adverse cardiac event is 1 percent or higher may require additional cardiovascular evaluation. This higher-risk group includes patients with known or suspected coronary artery or valvular heart disease. However, patients with marginally increased risk of major adverse cardiac event (1 to 2 percent) are commonly managed similarly to low-risk patients. (See 'Low-risk patients' above.)
Further evaluation may include stress testing or echocardiography, or cardiologist consultation. We generally perform these tests if they are indicated for the patient even if they were not having surgery. Many studies of patients not at low risk have shown that performing some form of stress testing can further stratify the risk of an adverse perioperative event [22,32,33,47-56]. However, no study has shown that interventions performed consequent to the results of the test improves outcomes.
When we consider further cardiovascular evaluation for higher-risk patients, we use the approach suggested in the 2014 American College of Cardiology/American Heart Association (ACC/AHA) guideline of perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery [9,10]. In this approach, the patient’s functional capacity plays an important role (algorithm 1). In patients who can perform ≥4 METs of activity, we do not order additional tests. For those whose functional capacity is lower or unknown, additional testing may be indicated if it will influence perioperative care.
In most cases, the reason to perform additional testing will be based not on the desire to lower risk at the time of surgery but rather to lower long-term risk. That is, the patient should have additional testing done irrespective of the need for surgery. There are few circumstances in which testing should be performed solely because the patient has upcoming surgery.
FURTHER CARDIAC TESTING — In patients with known or suspected heart disease (ie, cardiovascular disease, significant valvular heart disease, symptomatic arrhythmias), we perform further cardiac evaluation (eg, stress testing) only if it is indicated in the absence of proposed surgery. There is no evidence that further diagnostic or prognostic evaluation improves surgical outcomes. Preoperative cardiac evaluation and testing may differ for patients being evaluated for liver or kidney transplant. (See "Liver transplantation in adults: Patient selection and pretransplantation evaluation", section on 'Cardiac stress testing'.)
For patients in whom a decision has been made to perform additional cardiovascular testing, its timing should be determined by the urgency of the clinical situation.
Stress testing — Stress testing is not indicated in the perioperative patient solely because of the surgery if there is no other indication. (See "Stress testing for the diagnosis of obstructive coronary heart disease" and "Stress testing in patients with left bundle branch block or a paced ventricular rhythm" and "Noninvasive testing and imaging for diagnosis in patients at low to intermediate risk for acute coronary syndrome" and "Screening for coronary heart disease in patients with diabetes mellitus" and "Prognostic features of stress testing in patients with known or suspected coronary disease".)
However, some experts routinely obtain preoperative stress imaging in patients who are scheduled for major vascular surgery.
Stress testing with exercise (with or without imaging) and pharmacologic stress testing with imaging have been well studied in patients scheduled to undergo noncardiac surgery. Although there is a clear relationship between the degree of myocardial ischemia found and prognosis, there is no evidence that prophylactic revascularization, in addition to recommended medical therapy, to prevent ischemia at the time of surgery improves outcomes [47,56-64].
Resting echocardiography — Resting echocardiography is not indicated in the perioperative patient unless there is another indication, such as to evaluate valve function in patients with a murmur or left ventricular systolic function in patients with heart failure or dyspnea of unknown cause. (See "Transesophageal echocardiography: Indications, complications, and normal views" and "Echocardiographic evaluation of prosthetic heart valves" and "Role of echocardiography in atrial fibrillation" and "Echocardiographic evaluation of the aortic valve" and "Echocardiographic evaluation of the mitral valve".)
The presence of significant left ventricular systolic dysfunction or severe valvular heart disease is associated with a worse outcome, particularly postoperative heart failure, at the time of noncardiac surgery [12,48,65-69].
24-hour ambulatory monitoring — As with echocardiography and stress testing, we do not recommend 24-hour ambulatory monitoring for perioperative diagnostic or prognostic purposes if it is not otherwise indicated. Its use has not been shown to improve outcomes in this setting [1,70,71]. The indications for 24-hour ambulatory monitoring are discussed elsewhere and are primarily for patients with syncope or significant bradycardia or tachycardia if not previously evaluated. (See "Ambulatory ECG monitoring" and "Evaluation of palpitations in adults" and "Premature ventricular complexes: Clinical presentation and diagnostic evaluation" and "Supraventricular premature beats" and "Cardiac evaluation of the survivor of sudden cardiac arrest" and "Sustained monomorphic ventricular tachycardia: Clinical manifestations, diagnosis, and evaluation" and "Syncope in adults: Risk assessment and additional diagnostic evaluation", section on 'Introduction'.)
Preoperative BNP — Brain natriuretic peptide (BNP) is a natriuretic hormone that is released primarily from the heart. It is elevated in many pathologic conditions. (See "Natriuretic peptide measurement in heart failure" and "Natriuretic peptide measurement in non-heart failure settings".)
Evidence is increasing that a N-terminal pro-B-type natriuretic peptide (NT-proBNP) level may improve preoperative risk prediction when used in conjunction with recommended risk models such as the revised cardiac risk index (RCRI) or the myocardial infarction and cardiac arrest (MICA) risk index [72,73]. (See 'Our approach' above.)
In a nested substudy within the prospective VISION cohort study, 10,402 patients having inpatient noncardiac surgery had NT-proBNP measured before surgery [74] (see "Perioperative myocardial infarction or injury after noncardiac surgery", section on 'Incidence'). In multivariable analyses, increasing NT-proBNP values were associated with an independent and incremental risk of vascular death and myocardial injury or infarction within 30 days of surgery. Adding NT-proBNP to clinical stratification using the RCRI model improved cardiac risk predication compared with RCRI alone.
We do not recommend routinely using NT-proBNP until it has been validated in other large cohorts and its use is associated with improved clinical outcomes. It may be of value in patients being considered for possible stress testing where a low value would be helpful in downgrading the estimated risk.
Troponin — The potential role of troponin testing in perioperative risk stratification is discussed elsewhere (see "Perioperative myocardial infarction or injury after noncardiac surgery", section on 'Troponin'). However, it is important to assess a preoperative troponin level since it may be elevated at baseline from preoperative ischemic events.
RECOMMENDATIONS OF OTHERS — Our approach to the evaluation of patients scheduled to undergo noncardiac surgery is generally similar to that presented in the 2014 American College of Cardiology/American Heart Association (ACC/AHA) and European Society of Cardiology/European Society of Anesthesiology (ESC/ESA) guidelines on noncardiac surgery [9,10,13]. Despite collaborative efforts of the two societies to minimize discrepancies between their guidelines, there are differences in recommendations. The Canadian Cardiovascular Society Guidelines on perioperative cardiac risk assessment and management for patients who undergo noncardiac surgery take an alternative approach and focus on using biomarker elevation such as N-terminal pro-B-type natriuretic peptide (NT-proBNP) or brain natriuretic peptide (BNP) as the key determinant of further evaluation [75]. The differences between the Societal Guidelines have been discussed in more detail elsewhere [76]. We prefer the algorithm and recommendations in the ACC/AHA guideline (algorithm 1).
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: Perioperative cardiovascular evaluation and management" and "Society guideline links: Preoperative medical evaluation and risk assessment".)
SUMMARY AND RECOMMENDATIONS
●Assessing risk in all patients – All patients scheduled to undergo noncardiac surgery should have an assessment of the risk of a cardiovascular perioperative cardiac event (algorithm 1). The patient’s functional status is an important determinant of risk. (See 'Our approach' above.)
●Initial evaluation – Identification of risk factors is derived from the history and physical examination; the type of proposed surgery influences the risk of perioperative cardiac event. (See 'Initial preoperative evaluation' above.)
•We obtain an electrocardiogram (ECG) in patients with cardiac disease (except in those undergoing low-risk surgery) in large part to have a baseline available should a postoperative test be abnormal. (See 'Initial preoperative evaluation' above.)
●Risk assessment – We use either the revised cardiac risk index (RCRI), also referred to as the Lee index, or the American College of Surgeons surgical risk calculator (ACS-SRC) to establish the patient’s risk. (See 'Risk assessment' above.)
●Further cardiac testing for some patients – For patients with known or suspected heart disease (ie, cardiovascular disease, significant valvular heart disease, symptomatic arrhythmias), we only perform further cardiac evaluation (echocardiography, stress testing, or 24-hour ambulatory monitoring) if it is indicated in the absence of proposed surgery. (See 'Further cardiac testing' above.)
ACKNOWLEDGMENT — The editorial staff at UpToDate would like to acknowledge James P Morgan, MD, PhD, and Jonathan B Shammash, MD, who contributed to an earlier version of this topic review.
The editorial staff at UpToDate would also like to acknowledge Emile Mohler III, MD, now deceased, who contributed to an earlier version of this topic review.
