Chapter 9. Preoperative Evaluation for Cardiac Surgery

Improved outcomes and advances in surgical techniques have enabled the application of cardiac surgery in patient populations previously considered ineligible for an intervention of this magnitude. During the previous decade, the in-hospital mortality rate for coronary artery bypass surgery (CABG), the most common surgical procedure in the world, declined from 2.8 to 1.6% (43% relative reduction) despite an older and sicker patient population.1,2 Thorough preoperative risk assessment by the medical and surgical team is of critical importance in minimizing perioperative and long-term morbidity and mortality. It also enables physicians to counsel patients and families on what they can expect postoperatively, thus allowing them to make informed decisions regarding the treatment options available to them. This chapter reviews the essential information that the cardiologist and surgeon must collect and review to evaluate a patient for cardiac surgery. This information includes patient and disease characteristics, medications, and surgical considerations that can be integrated into scoring systems to provide a semiquantitative risk assessment (Table 9-1).

Patient Characteristics and Conditions

Age

The volume of cardiac surgical procedures in elderly people continues to increase as life expectancy improves and benefits outweigh risks. Although perioperative mortality does not vary significantly by age, 1-year mortality is greater in patients greater than 75 years of age.3 Octogenarians have nearly double the mortality rate compared with younger patients (4.1 versus 2.3%) and more than 60% have at least one nonfatal postoperative complication.4,5 The most prevalent complications include prolonged ventilation in intensive care units, reoperation for bleeding, and pneumonia—all of which result in longer hospital stays.5 A higher proportion of complications occur in elderly patients with low body weight (BMI <23).6 There is evidence that off-pump (OP) surgery using only arterial conduits may confer a survival benefit and improve long-term quality of life.7,8 With improved surgical techniques and careful patient selection, nonagenarians can safely undergo cardiac surgery with a 95% 30-day survival and 93% survival to hospital discharge.9

Gender

Some but not all epidemiologic studies suggest that female gender is an independent predictor of postoperative morbidity and mortality.10–12 Gender differences are present in both traditional CABG and off-pump surgery.13 Several large retrospective cohort studies of patients undergoing CABG found that women had higher mortality rates than men, even after adjusting for comorbidities and confounding factors, including BSA.10,11 Possible explanations for worse outcomes in women include smaller coronary arteries (which might enhance the difficulty of performing anastomoses and limit graft flow), differences in referral for surgery (ie, women being referred at later disease stages), and gender differences in self-reported outcomes.14 Data suggest that benefit in terms of quality of life after cardiac surgery is similar for men and women.15

Race

Although crude post-CABG mortality rates differ significantly by race, data suggest that after controlling for patient and hospital variables, these differences are small.16,17 However, in the United States, self-described black race is associated with an increased risk of postoperative complications, including prolonged ventilatory support, length of stay, reoperation for bleeding, and postoperative renal failure.18

Diabetes

Patients with diabetes have significantly worse outcomes following cardiac surgery.19–21 Studies have shown diabetes to be an independent predictor of in-hospital mortality after CABG although emerging evidence indicates that the severity of diabetes, specifically target organ damage, may be important in risk stratification.22,23 Postoperative mortality does not differ significantly between nondiabetic and diabetic patients without diabetic sequelae, although diabetic patients with vascular disease and/or renal failure have an increased risk of mortality.23 Insulin-dependent type II diabetics in particular are at increased risk of major postoperative complications including renal failure, deep sternal wound infection, and prolonged hospital stay.24,25 Strict perioperative glucose control has been shown to lower operative mortality and the incidence of mediastinits.26,27 Off-pump surgery also appears to decrease postoperative morbidity in diabetic patients.28

Renal Function

Renal dysfunction is common in patients undergoing cardiac surgery. Approximately half of patients undergoing CABG have at least mild renal dysfunction and one-fourth have at least moderate renal dysfunction.29 There is a graded increase in operative mortality and morbidity with worsening preoperative renal function.29–31 Renal insufficiency is associated with greater risk of both 30-day (OR = 3.7) and 1-year mortality (OR = 4.6).32 Even mild renal dysfunction (serum creatinine 1.47 to 2.25 mg/dL) is associated with increased rates of operative and long-term mortality, need for postoperative dialysis, and postoperative stroke.33

Renoprotective drugs, such as fenoldopam and N-acetylcysteine, have no effect on the deterioration of renal function in high-risk patients.34,35 Off-pump CABG is associated with a lower prevalence of the need for postoperative renal replacement therapy; larger studies are needed to determine if this correlates with improved outcomes.36

Pulmonary Function

It is well established that patients with compromised pulmonary function, predominantly owing to chronic obstructive pulmonary disease (COPD), have a higher mortality and increased incidence of postoperative complications including arrhythmias, reintubation, pneumonia, prolonged ICU duration, and increased hospital length of stay.37,38 Postoperative respiratory failure is a common complication (14.8% in a New York State database) with a higher incidence (14.8%) in combined CABG and valve operations.39 Optimizing respiratory status prior to surgery, including smoking cessation, antibiotics for pneumonia, and treatment of COPD flares with bronchodilator therapy and steroids is a critical part of preoperative management.40 There is evidence that intensive inspiratory muscle training prevents postoperative pulmonary complications in high-risk patients.41

Hematologic

The most important hematologic consideration in patients undergoing cardiac surgery is heparin-induced thrombocytopenia type II (HIT). HIT is an immune-mediated, potentially life-threatening thrombotic condition characterized by the formation of antibodies against the heparin-platelet factor 4 (PF4) complex.42 A diagnosis of HIT should be suspected in patients begun on heparin therapy within the preceding 5 to 10 days who have unexplained thrombocytopenia, venous or arterial thrombosis, or a platelet count that has fallen 50% or more from baseline even if absolute thrombocytopenia is not present.42 HIT can occur earlier in patients recently exposed to heparin. HIT can also be caused by exposure to low molecular weight heparin (LMWH), but the incidence is much lower.43 Although the prevalence of HIT antibodies is relatively common preoperatively (4.3%) and postoperatively (22.4%), the occurrence of thrombotic events occurs only in a minority of patients (6.3%) who have a positive antibody.42 CABG patients with HIT have a dramatically increased risk of vein graft occlusion (68% versus 20%); however, they do not appear to have an increased rate of internal mammary graft occlusion.44

There is emerging consensus regarding anticoagulant management of patients with HIT.42 Repeat heparin exposure during CBP is an option for patients with a previous history of HIT if repeat antibody testing is negative on two consecutive daily samples, although it is prudent to use alternative antithrombotic agents for preoperative and postoperative anticoagulation. Patients with prior HIT and detectable antibodies are at higher risk with reexposure and should only be administered heparin if platelet activation assays (ie, serotonin release assay) are negative. For patients with acute HIT, it is recommended to delay surgery if possible. If surgery must proceed, bivalirudin is an alternative, although attention must be given to its several unique pharmacologic properties.42

It is difficult to manage the competing risks of thrombosis and bleeding in patients with hypercoagulable disorders. Aggressive prophylaxis with unfractionated heparin or LMWH is indicated for patients not on chronic anticoagulation.45,46 Special considerations should be given to patients with antiphospholipid syndrome owing to their high risk of thrombotic events and the difficulty in interpreting coagulation parameters.47,48

Atrial Fibrillation

Preexisting AF in patients undergoing CABG is not associated with increased in-hospital mortality or major morbidity but is a risk factor for reduced 5-year survival.49 However, postoperative AF (POAF) is associated with increased in-hospital and long-term mortality, stroke, hemodynamic compromise, renal failure, longer length of stay, and greater hospitalization costs.50,51 AF is common after cardiac surgery, occurring in up to 40% of CABG patients, 50% of patients undergoing valve surgery and as many of 60% of patients after combined CABG and valve operation.50,52–54 The pathophysiology of postoperative AF is likely related to the age-related degenerative changes in the atrial myocardium coupled with inflammation, bleeding into the pericardial space, and perioperative electrophysiologic abnormalities in atrial refractoriness, conduction velocity, and transmembrane potentials.55 Evaluation of the risk for POAF is an important part of preoperative evaluation because prophylactic therapy can substantially reduce the incidence of AF.

Established risk factors for the development of POAF include history of AF, male gender, decreased left ventricular ejection fraction (LVEF), left atrial enlargement, valvular heart surgery, COPD, diabetes, chronic renal failure, rheumatic heart disease, and withdrawal from beta blocker and angiotensin 1 converting enzyme inhibitor (ACE inhibitor) therapy.50,56

A large Cochrane Database meta-analysis reviewed 58 randomized trials of more than 8500 patients and determined that beta blockers (OR 0.35), sotalol (OR 0.36), amiodarone (0.54), and atrial pacing (OR 0.57) were all highly effective in reducing postoperative AF.57 Overall, prophylactic therapy was associated with a nonsignificant 24% reduction in stroke. In a meta-analysis of randomized trials and risk-adjusted observational studies, off-pump (OP) CABG was associated with significant reductions in postoperative AF (41 and 22%, respectively).58

Practical recommendations include the continued use of beta blockers through the perioperative period because they are effective and safe in most patients.56 Amiodarone can be added to high-risk patients but should be a second-line agent because of its short- and long-term safety profile. There are limited but emerging data suggesting that the addition of magnesium, statins, N-3 polyunsaturated fatty acids, nitroprusside, and corticosteroids might be of additional benefit.56 Although prophylaxis is effective, it is still controversial whether preventing POAF translates into a reduction in stroke and postoperative complications.

Surgical Considerations

Reoperation

Hospital mortality among patients undergoing reoperative cardiac surgery has traditionally been higher than among patients undergoing primary operation.59–61 This is likely due to the higher risk profile of patients undergoing reoperation (older, more extensive vascular and coronary disease, multiple comorbidities) and the demanding surgical aspects including sternal reentry, pericardial adhesions, in situ arterial grafts, and diseased saphenous vein grafts.62

Despite these factors, hospital mortality associated with coronary reoperation has decreased with greater surgical experience and now approaches that of primary CABG.62,63 With careful preoperative risk evaluation and surgical management, reoperation can be performed safely.

Prior Radiation

Patients receiving thoracic radiation for treatment of malignancies before cardiac surgery have poorer short- and long-term outcomes.64,65 Thoracic radiation exposure is heterogeneous with respect to different malignancies and there is a gradient of risk.

A study dividing patients undergoing cardiac surgery into three levels of radiation exposure: extensive (Hodgkin's disease, thymoma, and testicular cancer), variable (non-Hodgkin's lymphoma and lung cancer), and tangential (breast cancer), demonstrated that patients with extensive radiation exposure had longer radiation-to-operation interval, poorer pulmonary function and more severe aortic regurgitation, diastolic dysfunction, and left main coronary stenosis.66 Hospital deaths (13 versus 8.6 versus 2.4%) and respiratory complications (24 versus 20 versus 9.6%) were higher after more extensive radiation, and 4-year survival was poorer (64 versus 57 versus 80%).

Surgical Complexity and Technique

It is important to consider the type of cardiac surgical procedure (CABG, valve, or combined) as well as the surgical technique (on-pump, off-pump, minimally invasive, robotic, and hybrid) when providing preoperative risk assessment as mortality and morbidity risk may vary.

Valve surgeries generally have a higher complication rate than isolated CABG and combined surgeries have the highest risk (Fig. 9-1). Participants in the Society of Thoracic Surgeons National Adult Cardiac Surgery Database (STS NCD) collected data from 2002 and 2006 on more than 3.6 million procedures.67–69 They reported 30-day mortality and a composite end point of mortality and major in-hospital morbidity, including stroke, renal failure, prolonged ventilation, deep sternal wound infection, and reoperation. CABG mortality was 2.3%, with a 14.4% rate for the combined end point of mortality and major morbidity.67 Isolated valve procedures had a higher mortality rate of 3.4% (aortic valve replacement [AVR] 3.2%, mitral valve replacement [MVR] 5.7%, and mitral valve prolapse [MVP] 1.6%).68 Combined mortality and major morbidity was also higher at 18.3% (AVR 17.4%, MVR 26.7%, MVP 12.7%). Combined CABG and valve procedures had the highest mortality rate at 6.8% (AVR + CABG 5.6%, MVR + CABG 11.6%, and MVP + CABG 7.4%) with combined mortality and major morbidity of 30.1% (AVR + CABG 26.3%, MVR + CABG 43.2%, and MVP + CABG 33.5%).69

Minimally invasive surgical techniques can be divided by approach and use of cardiopulmonary bypass (CBP). The use of alternatives to standard median sternotomy has become increasingly adopted in surgical centers for both CABG and valve surgeries. The potential benefits of minimally invasive surgery are earlier extubation, reduced incisional and thoracic discomfort, lower rates of wound infection, less blood loss, and shorter recovery times.70

Off-pump CABG (OP CABG) is performed with small skin incisions and stabilization devices to reduce motion of target vessels while anastomoses are performed without CBP. Current surgical approaches to valve surgery require cardiopulmonary bypass and cardioplegic arrest. Meta-analyses of observational and randomized trials of OP CABG versus CBP have not demonstrated a clear advantage of OP CABG over CBP with respect to mortality or morbidity.58,71 Compared with on-pump surgery, OP CABG generally shows consistent reductions in postoperative AF, blood loss, wound infections, and myocardial injury, with nonsignificant trends toward lower death, MI, and stroke.58,72,73 The benefits of OP CABG are particularly notable in elderly patients and those with heavily calcified (porcelain) aortas.72

There have been steady technologic advancements in optics, instrumentation, and perfusion technology that have facilitated use of totally endoscopic robotic cardiac surgery.74 This technology has been applied to many cardiac surgical procedures, in particular mitral valve repair (MVP) and totally endoscopic coronary artery bypass (TECAB) grafting. Short-term results are promising but long-term studies are lacking.

Simultaneous “hybrid” percutaneous coronary intervention and minimally invasive surgical bypass grafting procedures in a single, specially designed operating suite are also gaining more widespread acceptance. Hybrid procedures require close cooperation between surgical and interventional teams. Although there are limited data available, hybrid patients have similar angiographic vessel patency and major adverse cardiac events (MACE) at 6 months with shorter hospital length of stay, intubation times, and less blood loss despite aggressive antiplatelet therapy.75

Patients undergoing cardiac surgery are frequently on several medications for their preexisting cardiovascular conditions including aspirin, clopidogrel, beta blockers, ACE inhibitors, and statins. Perioperative medical therapy has been demonstrated to improve outcomes in patients undergoing cardiac surgery; however, these medications are probably underutilized because of concern for postoperative complications. Careful review of a patient's medication record with an assessment of the potential risks and benefits of continuing each medication is an important component of preoperative evaluation.

Antiplatelet Therapy

Aspirin

Aspirin (ASA) is a common medication in patients undergoing cardiac surgery. There are conflicting data regarding the benefits and risks of preoperative (predominantly CABG) ASA. Preoperative ASA has been demonstrated to reduce in-hospital mortality in several observational and case control studies.76–78 A large retrospective study in patients undergoing CABG found a 66% mortality reduction.76 The concern for excess bleeding, however, limits the use of preoperative ASA. Early studies indicated that preoperative ASA increased the rates of postoperative bleeding and need for transfusions, although this was not seen in subsequent studies.76,77,79,80

There are several guideline statements that provide slightly different recommendations regarding the use of preoperative ASA. The 2007 focused update of the 2004 ACC/AHA guidelines for the management of patients with ST-elevation MI (STEMI) recommends that ASA should not be withheld before either elective or nonelective CABG.81 The 2008 American College of Chest Physicians (ACCP) guideline on the primary and secondary prevention of CAD and the 2005 Society of Thoracic Surgeons practice guideline on antiplatelet drugs during CABG in stable patients recommend that it is reasonable to withhold ASA prior to elective CABG but that it should be resumed in the early postoperative period if there are no contraindications.82,83

Clopidogrel

Preoperative clopidogrel is associated with increased postoperative bleeding, transfusions, and reoperation.84,85 The 2004 ACC/AHA CABG guideline and the 2008 American College of Chest Physicians (ACCP) guideline on the primary and secondary prevention of CAD recommend stopping clopidogrel 5 to 7 days before surgery.19,83 This approach is reasonable for patients with planned elective stenting or in patients with STEMI (less than 1% proceed to CABG), but is problematic for individuals presenting with non-ST elevation ACS where the decision between PCI and CABG is not known until angiography. Because early clopidogrel therapy is critical to reduce the devastating complication of stent thrombosis, it is reasonable to give clopidogrel in this setting and accept the risk of bleeding in the minority of patients who proceed to surgical revascularization. Similar considerations are likely needed for new third-generation thienopyridines.86,87 One such agent, prasugrel, has recently become available and has demonstrated more potent platelet inhibition than clopidogrel at a cost of increased bleeding.86

Beta Blockers

Beta blockers are a cornerstone of treatment for patients with ischemic heart disease and heart failure.88,89 Beta blockade has been shown to reduce the rate of cardiovascular events in high-risk patients undergoing noncardiac surgery.90,91 Only 60% of cardiac surgical patients receive preoperative beta blockers; however, because of the concern for postoperative vasoplegia and exacerbation of reactive airway disease.92,93 A large North American observational analysis demonstrated preoperative administration of beta blockers was associated with a survival benefit with the exception of patients with a severely depressed left ventricular ejection fraction (<30%).94 Beta blockers are also highly effective in reducing the incidence and duration of postoperative AF (see atrial fibrillation section).

ACE Inhibitors

ACE inhibitors decrease mortality and cardiovascular events in patients with hypertensive and ischemic heart disease.95 Preoperative treatment with ACE inhibitors is controversial. ACE inhibitors and angiotensin II receptor blockers (ARBs) have been shown to reduce the incidence of postoperative AF.50,96 The withdrawal of chronic ACE inhibition can cause rebound hypertension and an increase in ischemia-related events.97,98 However, preoperative administration of ACE inhibitors or ARBs has been associated with postoperative vasoplegia and an increased requirement for vasopressor drug administration.92,99

Statins

Preoperative statin therapy is associated with a 38 to 43% reduction in mortality.100,101 Beneficial reductions are also observed with stroke and the incidence of postoperative AF.101 Statin pretreatment (atorvastatin 40 mg daily) initiated 7 days prior to surgery is associated with shorter hospital length of stay.102 The effects are lipid-independent and are likely mediated through anti-inflammatory and other pleiotropic effects.103,104 It is recommended that all patients undergoing cardiac surgery (especially CABG) receive preoperative statin therapy unless contraindicated.19

Preoperative risk assessment has important implications not only for individual patient well-being, but also as a qualitative tool to serve as a reference standard to compare outcomes among surgeons, institutions, or assessment of new procedures and techniques. There are numerous risk stratification scores and systems that have been developed from large databases to quantity the mortality and morbidity risks of cardiac surgery. Both patient and surgical factors are considered preoperatively and assessed for their ability to predict postoperative complications. This section focuses on three of the most widely used risk scoring systems: EuroSCORE, the Society of Thoracic Surgeons Score, and the Northern New England Score (Table 9-2).

All of these risk models share important limitations. Preoperative risk factors included in the model can change significantly over time, leading to substantial underestimation or overestimation of postoperative risk.105 Caution should also be used when interpreting the results of a model with respect to an individual patient's risk. The models are derived from large databases and do not include important considerations for an individual's specific risk profile such as surgeon experience and comorbidities not incorporated in the model's derivation.

The European System for Cardiac Operative Risk Evaluation (EuroSCORE)

The EuroSCORE, initially published in 1999, is the most rigorously evaluated scoring system in cardiac surgery.106 The score is calculated by assessing 17 risk factors (patient, cardiac, and operation) known to affect outcome. There are two available methods: the original additive model and more recent logistic model.107,108 Studies have indicated that the additive model overestimates mortality in low-risk patients and underestimates mortality in high-risk patients.108–110 The logistic model was designed to address these issues but there is still concern that this model overestimates mortality in many risk groups.108 The logistic EuroSCORE is more accurate at predicting mortality in combined CABG and valve surgery.111 The EuroSCORE can be calculated online (www.euroscore.org).

Society of Thoracic Surgeons Score (STS)

The STS has an extensive database that includes more than 2 million procedures since 1997. Similar to the EuroSCORE, an online calculator is available (www.sts.org), although there is more extensive data entry required to calculate a score. Models have been developed and revised for CABG, valve surgery, and combined procedures.67–69 These models include mortality as well as multiple morbidity end points, including stroke, renal failure, prolonged ventilation, deep sternal wound infection, and prolonged postoperative stay at 30 days. Unfortunately the STS score was not included in the most comprehensive analysis comparing 19 different scoring systems, but other studies demonstrate that STS showed similar discriminative capability and predictive performance with EuroSCORE and other risk algorithms.112,113

Northern New England Score (NNE)

The Northern New England scoring system was initially published in 1992 with data collected in the United States from 1987 to 1989.114 The initial model only predicted CABG mortality, but the Northern New England Cardiovascular Disease Study Group developed a second scoring system based on data between 1996 and 1998 that included the incidence of neurologic events and mediastinitis as additional end points. This scoring system was adopted by the ACC/AHA guideline committee on CABG and is often referred to as the ACC/AHA model.19,113 The model has reasonable predictive capabilities compared with other systems.115 There are now additional models available specifically for aortic and mitral valve surgery.116 All models are available online (www.nnecdsg.org).

Stroke

Atherosclerosis is a polyvascular problem. The incidence of coexisting coronary and carotid artery disease varies between 2 and 14%.117,118 In patients undergoing CABG, 17 to 22% of patients have a carotid artery stenosis greater than 50%, and 6 to 12% have a stenosis greater than 80%.119,120 Stroke is a feared and devastating complication of cardiac surgery, with a reported incidence ranging from 0.8 to 7%.40 Approximately 30% of postoperative strokes are caused by significant carotid artery stenosis, with increasing risk correlated with the severity of the stenosis, even in asymptomatic patients.19,21 Other predictors of stroke include advanced age, type of surgery, aortic atheroma, and duration of CBP (Table 9-3).21,121

Because patients with a perioperative stroke have a threefold increased risk of death compared with patients without stroke and many are left with significant disability, carotid revascularization as a combined or staged procedure with CABG can be considered.122 Current ACC/AHA guidelines recommend (Class IIa; Level of Evidence C) that carotid endarterectomy (CEA) before or concomitant with CABG in patients with symptomatic carotid stenosis or in asymptomatic patients with a unilateral or bilateral internal carotid stenosis of 80%.123 Carotid artery stenting is a less invasive alternative.

The optimal management of these patients remains controversial. A systematic review of 97 studies involving 8972 operative outcomes after staged and synchronous CEA and CABG revealed a 10 to 12% incidence of death, stroke, and myocardial infarction.124 Carotid artery stenting has been suggested as a possible less risky alternative; however, pooled analysis of studies demonstrate a similarly elevated risk.117 These results suggest that the presence of carotid stenosis is a marker of advanced atherosclerotic disease and risk that might persist despite revascularization.117 Observational studies suggest that high-risk patients may have reduced cerebral complication rates with OP CABG.125–127 Systemic hypothermia reduces the cerebral metabolic rate and may provide protection against transient periods of cerebral ischemia, although randomized trials comparing the safety of normothermic and hypothermic CPB have produce conflicting results.128,129 As a consequence, the ACC/AHA guidelines state that selective carotid screening should only be considered in the following high-risk patient groups: older than 65 years, left main coronary artery stenosis, carotid bruit on examination, PVD, history of smoking, and history of transient ischemic attack or stroke.123

Neurocognitive Decline

Disturbances in memory, executive function, motor speed, attention, and other cognitive functions can be detected in up to 80% of patients in the first several weeks after CABG when formal neuropsychiatric testing is performed.130–132 The severity can range from subtle deficits to profound delirium and disability. Cerebral microemboli have been implicated as a potential causative mechanism.133 Postoperative hyperthermia has also been associated with greater neurocognitive dysfunction.134 Deficits usually resolve gradually in most patients.

Preoperative risk assessment is critical to minimize perioperative and long-term morbidity and mortality. Risk assessment incorporates important patient and surgical information that can be integrated into several validated scoring systems, although an individualized approach is necessary to optimize the perioperative care of the patient. Importantly, it also allows physicians to provide patients and families realistic expectations of the potential risks and benefits of surgery, enabling them to make informed decisions about their health care.