CHAPTER 98: PERIOPERATIVE EVALUATION FOR NONCARDIAC SURGERY

Over time, the perioperative evaluation of cardiac disease in patients undergoing noncardiac surgery has been the object of much research and debate. The desire to optimally manage these patients is often confused with “cardiac clearance,” but this term fails to capture the complexity of care. As with most clinical scenarios, the cornerstone of good perioperative management is a conscientious history and physical exam. Understanding a patient's functional status, cardiac symptoms and signs, and whether they are acutely worsening are of paramount importance in guiding management decisions. As in the American College of Cardiology (ACC)/American Heart Association (AHA) and the European Society of Cardiology (ESC)/European Society of Anesthesiology (ESA) guidelines,^1,2 we believe the best clinical approach to evaluating cardiac disease in patients prior to surgery is in the framework of whether the cardiac disease is active or stable while understanding the urgency and risk of the operation needed. A summary of our global approach to perioperative evaluation and management may be found in Fig. 98–1.¹

The time period in which surgery must take place is of critical importance in deciding how to evaluate patients preoperatively. This temporal urgency dictates the extent of clinical evaluation and potential changes in management prior to surgery. This is clearly reflected in the ACC/AHA clinical guideline writing group's classification scheme of emergent, urgent, time-sensitive, and elective for temporal necessity of surgery, which is summarized in Table 98–1.¹

Active cardiac disease typically requires delay of nonemergent, nonurgent surgery. In elective surgery, management of acute cardiac conditions is similar to outside the context of imminent surgery. Active cardiac disease can exist in the form of active ischemia, arrhythmia, heart failure (HF), and/or severe valve disease.

Active Ischemia and Revascularization

In acute coronary syndromes, cardiac catheterization is generally recommended for defining coronary anatomy, risk stratification, and devising the optimal treatment and revascularization strategy if appropriate. The optimal revascularization strategy is based upon the weight of existing evidence and recent clinical practice guidelines.^1,2 Patients with ST-segment elevation myocardial infarction (STEMI) may benefit from immediate reperfusion, and non–ST-segment elevation myocardial infarction (NSTEMI) may benefit from an early invasive strategy within 72 hours in appropriately selected patients.

If percutaneous coronary intervention (PCI) is pursued, the timing of surgery must be considered. If noncardiac surgery must be performed within a manner of weeks, balloon angioplasty without stenting may be considered. If surgery is not essential, but still time sensitive, bare metal stents (BMS) can be used if lesion parameters permit. Although using BMS in this setting ideally portends 1 year of dual antiplatelet therapy (DAPT), one can instead consider a shorter course of P2Y₁₂ platelet inhibition, which may help reduce operative bleeding. If drug-eluting stents (DES) are implanted, surgery should also ideally be delayed for 1 year. However, surgery may be considered at 180 days as the risk of stent thrombosis may have stabilized in settings of noncardiac surgery, particularly in the era of second- and third-generation DES.^3,4,5

Management of Atrial Fibrillation

Atrial fibrillation (AF) is the most commonly encountered perioperative cardiac arrhythmia. A known history of clinically stable AF does not require modified management or special evaluation other than a consideration of how to manage perioperative anticoagulation. In determining a perioperative anticoagulation strategy, the risks of bleeding associated with a procedure must be weighed against that of bleeding. In cases where operative bleeding is minimal, such as cataract surgery, or implantation of cardiac rhythm devices, anticoagulation with vitamin K antagonists may continue uninterrupted. Observational data indicates that when pursued, bridging anticoagulation can be done safely.^6,7,8 However, when concerns over operative bleeding are significant, recent randomized data indicate brief interruptions in vitamin K–factor dependent anticoagulation may be superior to bridging anticoagulation in low-risk patients. Published in 2015, the Perioperative Bridging Anticoagulation in Patients with Atrial Fibrillation (BRIDGE) trial randomized 1884 patients to perioperative interruption of anticoagulation versus bridging anticoagulation with dalteparin. At 30-day follow-up, there was no difference in the rate of transient ischemia or stroke between the two groups, but there were higher rates of major and minor bleeding in the group that received bridging anticoagulation.⁹ These randomized data inform the discussion very well, but limitations of the study included poor representation of patients with CHADSVASC scores of 5 and above, as well as a lack of patients undergoing carotid endarterectomy, major cancer surgery, and other surgeries with high thromboembolic risk. All told, we believe the best strategy for patients and surgeries with AF and low thromboembolic risk is interruption of anticoagulation. In cases of high thromboembolic risk either resulting from patient comorbidities or prosthetic heart valves, an informed discussion should take place with patients with consideration of perioperative anticoagulation bridging. With regard to target-specific oral anticoagulants, stopping these anticoagulants 48 hours prior to surgery and pursuing either an interruption or bridging approach based upon thromboembolic risk is reasonable. It is worth noting that dabigatran is the only anticoagulant currently with a reversal agent, although others agents are in development for the other anticoagulants.¹⁰

In the setting of AF with rapid uncontrolled rates (above a heart rate of 110 bpm), surgery may be delayed to determine a rhythm or rate control strategy. Before selecting an appropriate strategy, secondary causes of rapid rates such as sepsis, metabolic disturbance, and volume derangements should be thoroughly investigated. Once the management strategy is adequately achieved, surgery may again be considered. If rapid AF results in hemodynamic instability, prompt rhythm control by electrical cardioversion is indicated. With other forms of supraventricular tachycardia (SVT), arrhythmia termination by vagotonic maneuvers, atrioventricular (AV) nodal or antiarrhythmic medications may be beneficial depending on the specific rhythm.

Uncontrolled Arrhythmias and High-Degree Conduction Disease

In the setting of sustained ventricular tachycardia (VT), prompt arrhythmia termination by chemical or electric means is critical. Cases of polymorphic and monomorphic VT necessitate investigation of ischemia as a cause. As a general rule, monomorphic VTs are typically scar mediated while polymorphic VTs are more likely to have ischemic, metabolic, or iatrogenic etiologies. These rhythms have the potential to be life threatening and prompt stabilization must occur before undertaking elective noncardiac surgery.

High-degree conduction disease (symptomatic bradycardia, sick sinus syndrome, second-degree Mobitz type II heart block, and complete heart block) can significantly complicate the perioperative period. Temporary transvenous pacing or permanent pacemaker implantation may be necessary before proceeding with surgery.¹¹

In patients with implanted cardiac rhythm devices, special attention should be paid to the use of monopolar electrocautery during surgery and its potential interaction with device programming and function. In cases where only bipolar cautery or a harmonic scalpel will be used, disruption of implanted rhythm devices is extremely unlikely. However, cases where monopolar electrocautery will be used above the umbilicus pose the highest risk of interaction with implanted rhythm devices. In pacemaker-dependent patients, reprogramming to the DOO or VOO setting can minimize oversensing and failure to pace. In patients with implanted cardioverter-defibrillators (ICDs), turning off tachytherapies is helpful to avoid unnecessary patient shocks. These changes can be accomplished by magnet placement or device reprogramming. If ICD tachytherapies are disabled, connection to an external defibrillator until device reprogramming is highly advised. A dedicated discussion about rhythm device management should occur prior to surgery with a plan for immediate postoperative reprogramming to the original settings.¹²

Heart Failure

As the prevalence of HF increases, the need for perioperative management of HF becomes ever more common. It is important to discern whether HF is occurring with a reduced or preserved left ventricular ejection fraction (LVEF), or whether it is in the setting of hypertrophic obstructive cardiomyopathy (HOCM). Special care must be taken in HOCM to avoid arterial dilation and overdiuresis, as such conditions can exacerbate left ventricular intracavitary obstructive gradients and potentially lead to perioperative complications. Active HF is associated with a high risk of perioperative cardiac complications.^13,14,15 In prior iterations of the cardiac risk index, active HF had the strongest association with perioperative major adverse cardiac events (MACE) of the factors considered.¹⁴ Natriuretic peptide measurement can help manage and predict perioperative events.^{16,17,18,19,20,21} In the setting of active nonobstructive HF, it is optimal to delay surgery for diuresis until euvolemia is achieved. Prior investigation has found this approach to help mitigate the risk of MACE.²²

Valvular Heart Disease

If significant valvular heart disease is suspected based on symptoms of dyspnea or a new or changing murmur on exam, echocardiography is critical for quantification of degree of stenosis or regurgitation and determination of surgical risk.

In patients with aortic stenosis (AS), anesthesia can cause hypotension, tachycardia, and decreased coronary perfusion leading to MACE.¹⁴ Symptomatic severe AS patients should undergo valve replacement prior to surgery when possible, as retrospective data confirm higher risk of MACE, specifically in the form of HF in patients with severe AS who undergo surgery.²³ In poor surgical candidates, valvuloplasty has had good perioperative results and relatively low complication rates, although the results have limited durability and long-term mortality is high without definitive management with valve replacement.^24,25 Although not specifically studied, transcatheter aortic valve replacement (TAVR) may be considered for patients with severe symptomatic disease for whom open surgery is not an option. Results of the Placement of Aortic Transcatheter Valves (PARTNER) trials have shown favorable outcomes in appropriate candidates.^26,27 If surgery is attempted without treatment of severe AS, meticulous cardiac anesthesia and perioperative monitoring are critical in the operative and postoperative setting.

Patients with severe mitral stenosis (MS) are prone to tachycardia, hypotension, and reduced cardiac output in the perioperative setting. It is critical to recognize the presence of rheumatic or nonrheumatic MS, and the characteristic physical exam finding in severe MS of an opening snap of the mitral valve and a long duration diastolic rumble. Symptomatic MS patients who are good candidates for balloon valvulotomy or surgical commisurotomy should undergo treatment before elective surgery.²⁸

As a general rule, left-sided regurgitant valve lesions are better tolerated than stenotic lesions.^29,30 The main pathology of mitral regurgitation stems from left ventricular volume overload, and is best managed by maintaining preload and avoiding excessive afterload. Prior investigations have revealed that patients with moderate-to-severe and severe aortic regurgitation (AR) had higher in-hospital mortality than case-matched patients without AR (9.0% vs 1.7%, P = .008). Predictors of mortality were a reduced LVEF (< 55%) and renal dysfunction (creatinine > 2.0).²⁹ In cases of asymptomatic moderate-to-severe or severe MR, noncardiac surgery may be considered with echocardiographic and selected invasive hemodynamic monitoring.¹ Postoperatively, patients should be monitored in the intensive care unit. Prior investigation from a single tertiary care center of patients with moderate-to-severe or severe MR revealed worse outcomes in propensity matched cases (driven by 30-day mortality, myocardial infarction [MI], HF, and stroke) in the group with MR.³⁰ Important predictors of poor outcome were ischemic source of MR, diabetes mellitus (DM), reduced LVEF < 35%, and a history of carotid endarterectomy.

Understanding how best to manage stable patients involves a more nuanced and predictive approach. In this circumstance, management has become more conservative over time, with fewer patients needing further cardiac testing or revascularization than in times past, largely because neither invasive or medical therapies in otherwise stable patients have shown substantial benefit. Prior randomized investigations in stable coronary artery disease (CAD) patients have not proven that coronary revascularization is effective in patients without left main (LM) coronary disease for modifying the risk of perioperative events, even for patients undergoing high-risk surgery.³¹ Risk-stratification methodologies, approaches to perioperative testing, and medical management are all important components of the perioperative evaluation of stable cardiac patients.

Perioperative Risk Stratification

Perioperative risk can generally be divided into two components: the inherent risk associated with an operation and the risk attributed to patient specific conditions. Three main risk-stratification algorithms exist: the revised cardiac risk index (RCRI), the American College of Surgeons National Surgical Quality Improvement Program Myocardial Infarction and Cardiac Arrest (NSQIP MICA), and the NSQIP Surgical Risk Calculator. We believe that the RCRI is most easily applied clinically, and the NSQIP tools are best for research. A comparison of the various tools is given in Table 98–2.

When assessing the inherent risk of an operation, typically low-stress surgeries with minimal fluid shifts are considered low risk (eg, cataracts, biopsies) and are associated with a < 1% risk of MACE.³¹ Vascular surgeries are considered the highest risk, but up-front risk can be tempered if an endovascular approach is taken, although conversion to open surgery remains a possibility. The intrinsic risk of an operation is multifactorial and somewhat difficult to predict; the ACC/AHA guidelines divide operations into low- and elevated-risk procedures.¹ In the ESC/ESA guidelines, operative risk is divided into low (< 1%), intermediate (1%-5%), and high risk (> 5%).²

RCRI

The RCRI is a simple and validated tool that assesses the risk of a major cardiac perioperative event defined as MI, pulmonary edema, ventricular fibrillation or primary cardiac arrest, or complete heart block.¹³ The RCRI uses six factors (surgical risk, history of ischemic heart disease, HF, cerebrovascular disease, insulin usage, and preoperative creatinine above 2), only one of which is directly attributable to the inherent risk of the operation (whether the surgery is suprainguinal vascular, intraperitoneal, and intravascular). A patient with no predictors or one predictor is considered low risk, and those with two or more predictors are considered high risk for MACE.

Biomarker testing with natriuretic peptide levels appears to independently predict the risk of cardiovascular events up to the first 30 days after vascular surgery and significantly improves the performance of the RCRI.^{9,10,11,12,13,14, 16,17,18,19,20,21} Further study is needed to validate whether measuring biomarkers leads to management change that ultimately improves patient outcomes.

NSQIP and NSQIP MICA

The NSQIP MICA and NSQIP risk calculators are based on data from over 1 million operations collected from 525 participating centers.³² The NSQIP MICA score was created in 2011 in a large derivation and validation study. The inherent risk of surgery was derived from odds ratios based on surgical site, with inguinal hernias used as the reference. The primary outcomes were cardiac arrest (defined as a chaotic rhythm requiring basic or advanced life support) or MI. The NSQIP MICA appears to outperform the RCRI, particularly in cases of vascular surgery.

The NSQIP uses procedure specific terminology, including whether surgery is emergent, to assess the perioperative risk of MACE, death, or eight other outcomes. It requires 21 patient-specific variables for effective calculation and is likely the closest estimate of surgery specific risk of any of the risk calculators. The drawbacks are that it has not been validated in an external population, and it relies upon the American Society of Anesthesiologists classification, which is subject to considerable inter observer variability.^33,34,35

Perioperative Testing

Perioperative testing should be limited to circumstances that would change management independent of the surgery planned. Testing is not necessary for low-risk procedures, and in the current environment of overuse of perioperative testing, specialty societies have undertaken efforts to reduce unnecessary testing.^36,37 As has been noted previously, revascularization in stable coronary disease prior to surgery has not been associated with improved outcomes.³¹ The cornerstone of determining the indication for further cardiac testing is assessment of a patient's symptoms, exercise status, and functional capacity. The risk of perioperative complications is indirectly proportional to functional status.³⁸ The conventional measurement standard for assessing functional status is metabolic equivalents (METs), where 1 MET is the basal oxygen consumption of a 40-year-old 70-kg man. If a patient has not had a recent exercise test, functional status can be estimated by performance during activities of daily living.³⁹ As an easy reference for functional status categories, METs and their examples may be found in Table 98–3.

Generally, formal functional testing is not indicated when a patient can easily and repetitively perform over 4 METs during activities of daily living, and an even stronger argument against testing can be made above 10 METs. Obtaining a 12-lead electrocardiogram (ECG) should generally be reserved for patients with known arrhythmia, CAD, peripheral vascular disease, and structural heart disease. Preoperative left ventricular function assessment is appropriate in patients with dyspnea of unknown origin, and patients with known left ventricular dysfunction and worsening or poorly controlled symptoms.

Functional Testing

Functional testing is potentially useful in patients with poor functional status when results will change management. Functional testing can be used to diagnose CAD, assess angina, and decide how to treat left ventricular dysfunction, dyspnea, or poor functional status. The evidence base for functional testing is fraught with limitations, including a heavy focus on vascular surgical patients, outdated stress protocols, and a lack of contemporary CAD management techniques. Nonetheless, the data confirm that patients able to achieve 7 to 10 METs are at low risk of perioperative cardiac events, and those that are unable to achieve 4 to 5 METs are at elevated risk.^40,41,42,43

The available evidence regarding the role for preoperative functional testing is primarily from single centers using dobutamine stress echocardiography (DSE) or radionuclide myocardial perfusion imaging (MPI).^{44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61} The predominant findings are summarized below:

1. Moderate to large areas of ischemia are associated with higher perioperative risk of MI and/or death.

2. Functional testing has a high negative predictive value.

3. Evidence of prior MI without inducible ischemia has little predictive value for perioperative events.

4. Prior studies have demonstrated clinical value and safety of either type of stress test.

The data are quite equivocal as to which technique performs best. We believe these techniques should be viewed equally given variable institutional expertise and experience with either technique.

Medical Management

After a thorough history, physical examination, and risk stratification, titration or addition of cardiac medications can be helpful in perioperative cardiac management. Most investigations and debate have centered on the role of β-blockers prior to surgery; there is still equipoise about the role of this therapy in the perioperative setting. Our discussion of perioperative medical management will focus on the role of β-blockade, statins, aspirin, DAPT, and clonidine.

β-Blockers

The debate around the use of perioperative β-blockade has nearly come full circle. Earlier, small randomized studies indicated possible benefit of β-blockers surrounding vascular surgery that was subsequently confirmed in the Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography (DECREASE) studies.^47,62,63 However, later publications questioned β-blocker utility, which was the impetus for the large randomized Perioperative Ischemic Evaluation (POISE) trial.⁶⁴ POISE revealed that perioperative β-blockade improved perioperative cardiac events, but at the expense of more bradycardic and hypotensive episodes, strokes, and most importantly, deaths. After critical investigation, all but the initial DECREASE studies were withdrawn from the published literature as a result of concern for scientific misconduct.⁶⁵ Although some have criticized POISE for choices in β-blocker dosing and frequency that may exceed those used in clinical practice, further investigations have also confirmed equivocal findings with perioperative β-blockade.^66,67

Based on the current literature, patients on longstanding stable β-blocker doses should have the medication continued at the time of surgery because of the risk of discontinuing β-blockade leading to subsequent withdrawal and related surges in heart rate and blood pressure.^66,67,68,69 β-Blockers may help mitigate cardiac risk in moderate to high-risk patients, and specifically in patients with RCRI scores of ≥ 3.⁷⁰ Otherwise, β-blockers should be used only for appropriate clinical circumstances irrespective of upcoming surgery, as it is still unclear if they have any positive effects on perioperative outcomes for most patients. If β-blockade use is pursued, initiation and titration should occur several days or weeks prior to surgery to avoid adverse events.

Statins

Statin therapy is effective for both primary and secondary prevention of cardiac events in patients with CAD, and statin indications are continuing to expand.^71,72 With regard to statin therapy's relationship to perioperative events, the randomized data are somewhat limited. A small trial of 100 patients randomized to either statin or placebo before vascular surgery found significantly lower MACE with statin use.⁷³ A Cochrane analysis revealed similar results, but again with a small sample size and most events attributed to a single study.⁷⁴ Statins may be helpful if started before surgery and present a great opportunity to improve long-term outcomes after surgery, but should be reserved for patients with established atherosclerosis. Both the ACC/AHA and ESC/ESA guidelines recommend statin therapy be continued at the time of surgery and that they may be of benefit in patients with peripheral arterial disease undergoing vascular surgery.^1,2 The ESC/ESA stipulate that statins are best initiated at least 2 weeks prior to vascular surgery.² The ACC/AHA guidelines go a step further and indicate that it might be reasonable to initiate statin therapy in patients without known atherosclerosis who have an indication for statins undergoing elevated risk surgery.¹

Aspirin and Dual Antiplatelet Therapy

There appears to be uncertain benefit of aspirin for preventing cardiac events in the perioperative period in nonstented patients. Observational data have suggested an increased risk of thrombotic events associated with discontinuation of aspirin.⁷⁵ Large randomized studies have not shown benefit of taking aspirin. The Pulmonary Embolism Prevention (PEP) trial randomized 13,356 patients to either 160 mg aspirin or placebo with no difference in outcome.⁷⁶ The POISE-2 trial was a 2×2 factorial trial studying the effects of aspirin and clonidine in perioperative events. It randomized 10,010 patients undergoing orthopedic, general, urologic, and gynecological surgery to 200 mg of aspirin or placebo, with no benefit of aspirin in this cohort with relatively low rates of CAD (23%) and with endarterectomy patients excluded. Further study is needed to understand aspirin's impact on high-risk patient groups. It is worth noting that POISE-2 also included relatively few patients with stents. In patients with coronary stents, aspirin is best continued during surgery and should be reinitiated as soon can be done safely if stopped.⁷⁷ Consensus should be reached between the surgeon, anesthesiologist, and cardiologist regarding the risk benefit profile (thrombosis vs bleeding) of continuing aspirin therapy in individual patients.

Several important factors must be considered regarding the timing of noncardiac surgery after coronary stent implantation. These factors are the risk of stent thrombosis, the consequences of delaying surgery, and the risk of procedural associated bleeding related to DAPT. DAPT is known to significantly reduce the risk of stent thrombosis, and these affects are most profoundly noted in the first 4 to 6 weeks after stent implantation.^{78,79,80,81,82} Mechanistically, stent thrombosis at the time of surgery is believed to result from endothelial disruption and the proinflammatory and prothrombotic effects throughout the coronary.^83,84 Newer-generation DES are associated with a lower risk of stent thrombosis and often require a shorter duration of DAPT. Investigations of treating patients with newer-generation DES with 3 and 6 months of DAPT compared with longer durations after implantation did not reveal any significant difference in MACE, although these studies were underpowered.^{85,86,87,88,89} Pooled analyses of these trials as well as data from the Patterns of Nonadherence to Antiplatelet Regimens in Stented Patients (PARIS) registry involving physician-initiated DAPT interruptions are all reassuring for shorter DAPT durations.^90,91 For patients treated with any type of stent for any indication, elective surgery should be delayed and DAPT should not be interrupted within 30 days of stent implantation. In cases of patients treated for acute coronary syndromes, truly elective cases should ideally be delayed for 12 months to allow a full DAPT duration. Surgery can be considered after 30 days with BMS and 6 months after DES with interruption in DAPT. However, in cases of emergent, urgent, and time-sensitive surgery, interruption of DAPT can be considered 3 to 6 months after DES implantation with newer-generation DES.⁵

Clonidine

Initial investigations involving α-2 agonism at the time of surgery appeared to indicate that clonidine was beneficial. In 2004, randomized clinical data of 200 patients undergoing noncardiac surgery demonstrated better mortality up to 2 years in patients who took clonidine (15% vs 29%, P = .035).⁹² Meta-analysis of 31 α-2 agonist trials also revealed promising effects, notably in vascular surgical patients.⁹³ However, the POISE-2 trial, which was orders of magnitude larger than prior investigations, failed to demonstrate these benefits and rather showed that clonidine increased the rate of nonfatal cardiac arrest and clinically important hypotension.⁷⁷ Therefore, clonidine is not recommended for mitigating cardiac risk at the time of noncardiac surgery.

With all the intricacies of perioperative evaluation, the best preoperative cardiac test remains a careful history and physical directed at finding and grading CAD, left ventricular dysfunction, valve disease, rhythm disorders, or previously unsuspected vascular disease. Any testing or treatment should be justified by the patient's disease alone, because there is little evidence that treatments for the sole purpose of improving perioperative outcomes have any substantial benefit.