Coronary artery bypass grafting (CABG) is the most common cardiac surgical procedure, with nearly 200,000 operations performed annually in the United States according to the Society of Thoracic Surgeons (STS) database. Reoperative CABG accounts for approximately 5 percent of all isolated CABG procedures and is often the result of graft failure, progression of native coronary disease, or incomplete revascularization.^1,2 Recent STS data show that there has been a decrease in the number and percentage of isolated redo-CABG operations (2000, N = 8929, or 6.1 percent; 2009, N = 8189, or 4.3 percent). There are probably many reasons for this trend, including improved medical management and interventional cardiology. Most redo-CABG procedures today are performed more than 10 years after the primary CABG, and are technically more challenging because of the difficulty of sternal reentry, the presence of patent bypass grafts, and limitations of potential bypass conduits. In addition to the technical complexities, the risk profile of reoperative candidates continues to increase. Redo-CABG patients are older, have greater extent of coronary artery disease (CAD), and have worse left ventricular (LV) function. However, these obstacles have been offset by improvements in operative techniques and perioperative care, resulting in improved survival and postoperative outcomes in redo-CABG patients.

Surgical reintervention rates and recurrent symptoms are two of the primary outcomes used to gauge the success of the initial CABG procedure. In two large cohort studies, freedom from redo-CABG at 1, 5, 10, and 15 years was 99, 97, 89, and 72 percent, respectively.^2,3 In addition, freedom from recurrent angina after initial CABG surgery has been approximately 80 percent at 10 years.⁴ In a multivariable analysis, preoperative angina, use of vein grafts only, previous myocardial infarction (MI), incomplete revascularization, female gender, smoking, and younger age were found to be independent risk factors for recurrent angina.⁵ Another large study defined diffuseness of CAD using a validated diffuseness score, and found that diffuse CAD was an independent predictor of morbidity and mortality.⁶

Reoperative candidates include patients with recurrent symptoms and jeopardized viable myocardium due to progression of CAD or graft failure. Predictors of redo-CABG include young age at initial operation, male gender, worse symptom severity, absence of internal mammary artery (IMA) grafts, incomplete revascularization, impaired functional status, and multivessel disease.^7,8 Initially, incomplete revascularization or worsening atherosclerosis of the native coronary arteries accounted for the majority of redo-CABG operations. As the redo-CABG population has evolved to include older patients with worse LV dysfunction and triple-vessel disease, graft failure has become the predominant etiologic factor. More recently, however, there has been an increase in the number of reoperative candidates with progression of their native disease distal to patent arterial graft anastomoses.⁸

Choice of graft conduits at the primary operation has been shown to be a significant factor in determining the need for reoperation. The increased use of arterial conduits, including IMA and radial artery grafts, instead of saphenous vein graft (SVG) conduits, has reduced the rate of reoperation significantly in recent years. IMA patency rates after primary CABG surgery exceed 90 percent at 10 years; in contrast, SVGs begin to deteriorate rapidly after 5 years and have a 50 percent patency rate at 10 years.⁹ Furthermore, nearly 20 percent of SVGs occlude by 1 year.¹⁰ Although some debate remains, there are several large retrospective studies that support improved survival in patients receiving total arterial revascularization, in both primary and redo-CABG procedures.¹¹ Because of improved patient survival and patency rates of arterial grafts, the IMA has become the choice conduit—when available—for primary CABG as well as for redo-CABG surgery for SVG stenosis.

Early SVG stenosis within the first year of primary CABG often is due to intimal hyperplasia. The increased intimal thickness presumably is due to exposure of the vein to systemic arterial pressures. This proliferative process results in vein graft atherosclerosis, which is the leading cause of late occlusion and accelerates dramatically after 5 years. SVG atherosclerosis has a worse prognosis than does native vessel atherosclerosis. The lesion in a vein graft is often diffuse and prone to plaque rupture, with increased rates of thromboembolic phenomena. As a result, even minimal manipulation of SVGs or antegrade cardioplegia during redo-CABG surgery can dislodge atheromatous emboli and result in MI. This represents the most common cause of operative mortality in patients undergoing redo-CABG surgery. Arterial grafts are much less prone to occlusion, although they can develop neointimal proliferation, most commonly at the distal anastomotic site.

Preoperative planning is of even greater importance in redo-CABG surgery than it is in primary procedures. For the reoperation to occur safely, adequate preoperative assessment and analysis are critical (Table 29-1). The patient’s history must be reviewed thoroughly, with close attention not only to all previous cardiac surgery, but also to any relevant procedures or comorbidities that may affect the surgeon’s approach to the reoperation. For example, a patient with a previous history of mediastinitis should be considered for a thoracotomy incision as an alternative to redo-sternotomy if the coronary anatomy is favorable for this approach. Medications should be reviewed carefully so that any drugs affecting platelet function or coagulation can be discontinued at least 1 week prior to the operation. Most importantly, it is essential to have a clear and complete understanding of the patient’s native coronary and bypass graft anatomy before undertaking a reoperation. This information can be ascertained from previous operative records and coronary angiograms. The surgeon must also determine if bypassing potentially graftable vessels will improve perfusion to viable regions of myocardium. In addition to determining the feasibility of beneficial revascularization, adequate bypass conduits must be available.

Bypass Conduits

The surgeon should have a preoperative plan for the selection of bypass conduits. Venous Doppler studies often are used to determine the adequacy of saphenous vein segments. Vein mapping is a simple, accurate, and noninvasive method to predict the course of the vein and identify venous anomalies. On occasion, venous mapping can influence the surgeon’s choice of the venectomy site. Arterial Doppler studies can assess the radial arteries. In addition, an Allen’s test can help ensure that digital perfusion will remain adequate if the radial artery is harvested as a potential conduit. Abnormalities in the IMA also can be detected with Doppler studies. IMA ultrasound studies may reduce the need for invasive preoperative assessment of patency and length, and allow for postoperative assessment of coronary artery flow reserve. However, IMA angiography provides a more complete evaluation of these arteries and should be used whenever possible.

Chest X-ray and Computed Tomography

The surgeon can evaluate the risk of sternal reentry by means of radiography. A posteroanterior and lateral chest x-ray can give the surgeon an idea of the proximity of important cardiac structures to the sternum. Steel clips that are visible on a chest radiograph also can be used to trace the path of patent IMA grafts. If deemed necessary, CT offers an enhanced view of substernal structures. CT scans can determine the size and position of cardiac chambers as well as the course of bypass conduits. This information can guide sternal reentry. In a series of 167 redo-CABG patients, high-risk findings by preoperative CT imaging led to more frequent adoption of preventive surgical strategies.¹² In patients at high risk for hemorrhage or injury to substernal structures, femoral dissection should be performed to allow for peripheral and expeditious institution of cardiopulmonary bypass (CPB). Alternative incisions should be considered if they are deemed necessary on the basis of this evaluation. In addition to standard CT scanning, intravenous angiography multislice CT and 3-D reconstruction is a recent technological advancement that has been found to be extremely accurate in detecting CABG patency and occlusion (Fig. 29-1).¹³ This imaging modality also provides clear and accurate pictures of the cardiac anatomy and the relationships of bypass grafts to mediastinal structures; however, standard angiography remains the modality of choice for assessing native coronary circulation (Fig. 29-2).

Coronary Angiography

Coronary angiography is a necessary step in the complete preoperative assessment of potential redo-CABG patients. To gain a better understanding of the patient’s coronary anatomy, previous angiograms and operative notes should be reviewed and are invaluable tools for the reoperative surgeon as well as the interventional cardiologist performing the angiogram. The coronary angiogram can be used to estimate the severity of atherosclerotic and thrombotic lesions. The severity of native coronary and SVG stenosis often is underestimated on the basis of a single angiographic view. Therefore, multiple projections are required to visualize the entire native coronary circulation as well as the previously constructed bypass grafts. The angiogram also can aid in determining the presence of graftable target vessels. To ensure that this is performed accurately, the angiographer must inject all relevant branches and allow enough time for the filling of the coronaries by collaterals. It is important that angiography supplement the surgeon’s intraoperative assessment of the patient’s coronary anatomy.

Assessing Viable Myocardium

Impaired LV function secondary to ischemic heart disease is not always permanent, and may improve considerably with revascularization. Restoration of blood flow to chronically underperfused and ischemic myocardium often can restore contractile function. However, revascularization of scar tissue will not result in functional improvement. Given the considerable operative morbidity and mortality for patients with LV dysfunction undergoing surgical revascularization, especially redo-CABG, it is essential to differentiate between viable and nonviable myocardium. Several noninvasive imaging modalities are used to identify physiologic markers of myocardial viability in regions with contractile dysfunction. The most commonly used techniques to ascertain viability include positron emission tomography (PET), thallium scintigraphy, dobutamine echocardiography, and cardiac magnetic resonance imaging (cMRI). PET is an established method for demonstrating preserved metabolic activity through the use of ¹⁸F-fluorodeoxyglucose as a marker of myocardial glucose utilization. Thallium scanning is used to assess myocardial perfusion and membrane integrity, whereas the myocardial contractile reserve is best evaluated by using dobutamine echocardiography or cMRI. The chief advantage of cMRI is the ability to simultaneously detect ischemic myocardium, calculate ventricular chamber size and volumes, and assess viability and scar formation.¹⁴

Nonsurgical Interventions

Redo-CABG carries with it significantly higher morbidity and mortality as compared with patients without a history of CABG. Therefore, various nonsurgical approaches to revascularization, such as percutaneous coronary interventions (PCI) with intracoronary stenting have become much more widespread in recent years. As the line begins to blur between candidates for surgical versus nonsurgical revascularization, many studies have been performed to compare these two techniques for primary revascularization. However, the data are more limited in patients in need of reintervention who already have undergone CABG. The AWESOME trial was a nationwide, prospective, multicenter effort, which demonstrated that PCI is a viable alternative for patients with medically refractory myocardial ischemia and risk factors for adverse outcome with CABG. This registry demonstrated that for the specific subset of patients with prior CABG, both the physicians and patients chose PCI by a 2:1 margin, and a survival advantage with PCI was shown in the patient-choice cohort. For the 142 randomized patients to either redo-CABG or PCI, similar survival results were found with an overall 3-year survival of 76 and 73 percent, respectively.¹⁵ However, redo-CABG is associated with greater success in providing complete revascularization as well as a reduced need for subsequent revascularization procedures. In a review of patients referred for reoperation versus PCI at the Cleveland Clinic, those undergoing redo-CABG were more likely to be male and have diabetes, hypertension, valvular disease, and severe LV systolic dysfunction. These patients generally had fewer functioning venous and arterial grafts and a larger region of myocardium in jeopardy. Although reoperation is associated with a higher risk of in-hospital complications than is PCI, it has potential long-term benefit when there is a lack of functioning bypass grafts or a lack of a patent arterial graft to the left anterior descending (LAD) coronary artery.¹⁶ SVG stenosis also can be treated with stenting; however, stenting usually is limited to grafts with focal or discrete disease. Outcomes are much poorer for diffusely degenerated SVGs. The risk of embolization from SVG atherosclerosis to the distal coronary bed is also greater compared with intracoronary or arterial graft stenting. Recent advances in percutaneous techniques include catheter-based systems designed to capture atheromatous debris. These advances are believed to decrease procedural morbidity and long-term graft restenosis rates. However, further studies are necessary to compare the long-term outcomes of these newer techniques with the outcomes from operative revascularization.