Residency · Residency · Cardiology

Coronary Artery Bypass Grafting: Indications and Graft Selection

Indications for CABG

Survival Benefit (Class I)

Coronary artery bypass grafting confers a demonstrated survival benefit in several well-defined clinical scenarios. Left main coronary artery disease with stenosis of 50% or greater represents the gold standard indication, with bypass grafting demonstrating superiority over percutaneous coronary intervention in patients with SYNTAX scores exceeding 32, as established by the SYNTAX trial. In patients with SYNTAX scores of 32 or below, the EXCEL trial suggested comparable outcomes between the two strategies, though these results remain debated. Three-vessel coronary artery disease represents another strong indication, particularly in patients with diabetes mellitus. The landmark FREEDOM trial demonstrated that bypass grafting reduced death, myocardial infarction, and stroke compared to percutaneous intervention in diabetic patients with multivessel disease at 5 years, with the benefit sustained at 8 years of follow-up. For three-vessel disease with a SYNTAX score exceeding 22, bypass grafting is the preferred strategy, as supported by both the SYNTAX and BEST trials.

Two-vessel disease with proximal left anterior descending artery involvement carries an established benefit from bypass grafting, particularly when a left internal mammary artery-to-left anterior descending graft is constructed. In patients with significant coronary artery disease and a left ventricular ejection fraction below 35% with viable myocardium, the STICH trial with 10-year follow-up demonstrated that bypass grafting plus optimal medical therapy reduced all-cause mortality by 16% compared to optimal medical therapy alone. For diabetic patients with significant coronary artery disease, the FREEDOM trial established bypass grafting superiority over percutaneous intervention irrespective of the SYNTAX score.

Symptom Relief

Beyond prognostic indications, bypass grafting provides effective symptom relief for patients with angina refractory to optimal medical therapy when the coronary anatomy is suitable for surgical revascularization. This includes patients with diffuse coronary artery disease not amenable to percutaneous intervention and those with failed percutaneous intervention or recurrent in-stent restenosis.

Guidelines-Based Decision Making

Contemporary guidelines mandate a heart team discussion for patients with left main disease, multivessel coronary artery disease, and complex coronary anatomy. The SYNTAX score serves as the primary anatomic complexity scoring tool, with low scores of 0 to 22 indicating that either percutaneous intervention or bypass grafting is reasonable, intermediate scores of 23 to 32 generally favoring bypass grafting, and high scores exceeding 32 strongly favoring bypass grafting. The SYNTAX II score represents an evolution that integrates anatomic complexity with clinical variables including age, ejection fraction, chronic kidney disease, chronic obstructive pulmonary disease, and peripheral vascular disease to provide individualized risk prediction.

Conduit Selection

Left Internal Mammary Artery (LIMA)

The left internal mammary artery graft to the left anterior descending artery is the single most important component of any coronary artery bypass operation, with 10-year patency exceeding 95% and a dramatic improvement in long-term survival. The superiority of this conduit derives from its robust endothelial nitric oxide production, inherent resistance to atherosclerosis, favorable size-matching to the left anterior descending artery, and preserved vasa vasorum when harvested as a pedicled graft. The left internal mammary artery should be used in virtually all bypass operations unless specific contraindications exist, such as subclavian artery stenosis, prior mediastinal radiation, or emergency situations with inadequate time for harvest. In-situ pedicle grafting is preferred over use as a free graft to maintain the internal thoracic artery blood supply.

Right Internal Mammary Artery (RIMA)

Bilateral internal mammary artery grafting has been associated with improved long-term survival compared to single internal mammary artery grafting combined with vein grafts in observational studies. However, the ART trial did not demonstrate a significant difference in 10-year survival between single and bilateral internal mammary artery strategies, though this may have been related to the study being underpowered and a high crossover rate. A significant concern with bilateral internal mammary artery grafting is the increased risk of deep sternal wound infection, occurring in 3 to 5% compared to 1 to 2% with single internal mammary artery grafting, particularly in diabetic patients, obese individuals, and those with chronic obstructive pulmonary disease. The skeletonized harvest technique reduces this risk. The right internal mammary artery can be directed to the left anterior descending artery as a free graft or in-situ to left anterior descending branches, or to the left circumflex system. A Y-graft configuration with the right internal mammary artery arising from the left internal mammary artery allows coverage of multiple arterial territories from a single inflow source.

Radial Artery

The radial artery has emerged as the preferred second arterial conduit, with 10-year patency of approximately 85 to 90%, which is superior to saphenous vein graft patency. The RADIAL trial confirmed that the radial artery provides better outcomes than saphenous vein grafts for the composite of death, myocardial infarction, and repeat revascularization at 5 years. Optimal radial artery patency requires that the target vessel stenosis be 70% or greater, as competitive flow from lesser degrees of stenosis reduces graft patency, and the target vessel diameter should be at least 1.5 mm. Contraindications include a positive Allen test indicating inadequate ulnar collateral circulation, Raynaud disease, and the presence of an arteriovenous fistula for dialysis on the same arm. Spasm prevention requires postoperative calcium channel blocker therapy, typically amlodipine at 5 to 10 mg, for 6 to 12 months, with some centers also using papaverine and verapamil solution during harvest. The RAPCO trial demonstrated that the radial artery provides equivalent patency to free right internal mammary artery grafting at 6 years.

Saphenous Vein Graft (SVG)

The saphenous vein remains the most commonly used conduit overall due to its ready availability, ease of harvest, and versatility. However, its long-term patency of 50 to 60% at 10 years is significantly inferior to arterial conduits. Early graft failure within the first year results from thrombosis, intermediate failure from 1 to 5 years is driven by intimal hyperplasia, and late failure beyond 5 years occurs from accelerated atherosclerosis within the graft. The no-touch harvesting technique, which preserves the surrounding perivascular tissue intact, maintains the vasa vasorum and prevents spasm. The SUPERIOR-SVG trial demonstrated improved 8-year patency with this approach at 84% compared to 63% with conventional harvesting. Endoscopic harvesting reduces wound complications compared to open harvesting but may reduce graft patency, as suggested by the PREVENT-IV trial, though newer techniques and training may mitigate this concern.

Gastroepiploic Artery and Other Conduits

The right gastroepiploic artery can be used as a pedicled graft from the greater curvature of the stomach to the posterior descending artery or distal right coronary artery, with 10-year patency of approximately 60 to 70%. Its use is generally reserved for select cases when other conduits are unavailable. The inferior epigastric artery may be used as a free graft but is limited by its short length and is rarely employed in practice.

<image> An anatomical illustration of the heart showing a complete CABG operation with multiple graft types. Anterior view of the heart with the ascending aorta visible. Show four grafts: (1) LIMA (left internal mammary artery, in red) as an in-situ pedicled graft from the left subclavian artery coursing to the LAD, with the pedicle intact showing the internal thoracic vein alongside it. (2) RIMA as a Y-graft (in red) arising from the LIMA and grafted to the first obtuse marginal branch of the LCx. (3) Radial artery free graft (in orange) from the ascending aorta to the second obtuse marginal. (4) SVG (saphenous vein graft, in blue) from the ascending aorta to the PDA/distal RCA, showing the proximal anastomosis to the aorta and distal anastomosis to the coronary artery. The native coronary arteries should be visible with stenotic lesions proximal to each anastomosis site. Label each graft with its name and expected 10-year patency rate. Include a small inset showing the difference between pedicled and skeletonized IMA harvest techniques. Use anatomic coloring with arteries in red, veins in blue, myocardium in tan. </image>

Total Arterial Revascularization

Rationale

The rationale for total arterial revascularization rests on the consistently superior long-term patency of arterial grafts, including the left and right internal mammary arteries and the radial artery, compared to saphenous vein grafts. By avoiding saphenous vein grafts entirely, total arterial revascularization eliminates the problem of late vein graft failure and the need for reintervention on degenerated vein grafts. Multiple configurations are possible, including bilateral internal mammary arteries combined with a radial artery, left internal mammary artery plus radial artery with multiple sequential grafts, and left internal mammary artery plus right internal mammary artery Y-graft with sequential anastomoses.

Evidence

Observational data have consistently demonstrated improved 15 to 20-year survival with multiple arterial grafts compared to the combination of left internal mammary artery and saphenous vein grafts. However, the randomized trial data have been less definitive. The ART trial did not conclusively demonstrate bilateral internal mammary artery superiority, confounded by a high crossover rate and the limitations of an intention-to-treat design. The ROMA trial evaluating total arterial revascularization did not demonstrate a significant reduction in the composite of death, myocardial infarction, or repeat revascularization compared to left internal mammary artery plus saphenous vein grafts at 5 years, though the survival difference may require longer follow-up to emerge. Current practice favors total arterial revascularization in younger patients below age 70 with good life expectancy, reasonable operative risk, and absence of factors that increase the risk of sternal wound complications.

Surgical Techniques

On-Pump CABG (Conventional)

Conventional on-pump coronary artery bypass grafting utilizes cardiopulmonary bypass with cannulation of the aorta and right atrium or bicaval venous drainage, systemic hypothermia to 28 to 34 degrees Celsius, and cardioplegic arrest using antegrade and/or retrograde cold blood or crystalloid cardioplegia. The cross-clamp time represents the period of myocardial ischemia, with longer cross-clamp times associated with greater perioperative myocardial injury. A target of less than 90 minutes is appropriate for routine bypass operations. The principal advantage of on-pump surgery is the provision of a motionless, bloodless field that permits precise anastomosis construction. Disadvantages include the systemic inflammatory response triggered by the bypass circuit, hemodilution, the risk of aortic atheroemboli from cross-clamping, and potential neurologic injury.

Off-Pump CABG (OPCAB)

Off-pump coronary artery bypass grafting is performed on the beating heart without cardiopulmonary bypass, using cardiac stabilizer devices such as the Octopus and Starfish systems to immobilize the target operative area. The CORONARY trial demonstrated similar 5-year mortality and major adverse cardiovascular events for off-pump and on-pump approaches, but the off-pump group had higher rates of incomplete revascularization and more repeat revascularization procedures. The GOPCABE trial in elderly patients over 75 years of age failed to show a reduction in stroke or mortality with the off-pump technique compared to conventional surgery. The current role of off-pump surgery is primarily in experienced hands for selected patients with specific indications such as a porcelain aorta, severe aortic atheroma, or renal failure, where avoiding cardiopulmonary bypass offers a distinct advantage.

Minimally Invasive CABG (MIDCAB)

Minimally invasive direct coronary artery bypass involves constructing a left internal mammary artery-to-left anterior descending artery graft through a small left anterior thoracotomy without sternotomy, performed on the beating heart. A hybrid approach combining minimally invasive bypass for the left anterior descending artery with percutaneous intervention to non-left anterior descending targets offers the benefit of the best conduit combined with a less invasive approach for the remaining vessels. This technique is limited to the left internal mammary artery-to-left anterior descending graft, though robotic-assisted left internal mammary artery harvest is expanding its applicability.

Perioperative Management

Preoperative Assessment

The Society of Thoracic Surgeons risk calculator provides standardized prediction of 30-day mortality and morbidity outcomes including renal failure, stroke, prolonged ventilation, deep sternal wound infection, and reoperation. Frailty assessment using measures such as gait speed, grip strength, and the Katz Activities of Daily Living score independently predicts mortality and prolonged recovery, adding important prognostic information beyond traditional risk scores. Medication management requires continuation of aspirin up to the time of surgery, holding clopidogrel for at least 5 days, ticagrelor for at least 3 days, and prasugrel for at least 7 days before elective surgery. Beta-blockers should be continued perioperatively, while angiotensin-converting enzyme inhibitors and angiotensin receptor blockers are held on the morning of surgery to reduce the risk of hypotension during cardiopulmonary bypass.

Postoperative Care

Early extubation within 4 to 8 hours is targeted as part of enhanced recovery protocols that reduce intensive care unit stay. Postoperative atrial fibrillation occurs in 25 to 40% of patients and is mitigated by prophylactic perioperative beta-blockers and/or amiodarone. When atrial fibrillation persists beyond 48 hours, anticoagulation should be initiated. Significant bleeding, defined as chest tube output exceeding 200 mL per hour for 4 or more hours or exceeding 400 mL in the first hour, may necessitate re-exploration, with common causes including surgical bleeding, coagulopathy from bypass-induced platelet dysfunction, hemodilution, and hypothermia.

Acute kidney injury occurs in 10 to 20% of bypass patients, related to cardiopulmonary bypass, contrast exposure, and hemodynamic instability, and is usually transient. Cognitive dysfunction affects 30 to 50% of patients at discharge, typically resolving over weeks, though late persistent decline occurs in 10 to 20% at 1 year and is attributed to multiple factors including microemboli, hypoperfusion, and inflammation. Deep sternal wound infection occurs in 1 to 2% of cases overall, with higher rates in patients receiving bilateral internal mammary artery grafts who also have diabetes or obesity, and management includes debridement, vacuum-assisted closure, and muscle flap coverage.

Long-Term Outcomes

Thirty-day mortality for isolated bypass grafting ranges from 1 to 2% in low-risk patients to 3 to 5% in intermediate-risk patients. Graft patency at 1 year demonstrates the superiority of arterial conduits: the left internal mammary artery achieves approximately 99% patency, the radial artery approximately 95%, and saphenous vein grafts 85 to 90%. At 10 years, this advantage becomes more pronounced, with left internal mammary artery patency exceeding 95%, radial artery patency at approximately 85 to 90%, and saphenous vein graft patency declining to 50 to 60%. Long-term medical therapy after bypass grafting requires aspirin at 81 to 162 mg indefinitely, with mandatory statin therapy. While the DACAB trial suggested that ticagrelor 90 mg twice daily for 12 months improved saphenous vein graft patency, this approach has not been adopted as standard practice.

<image> A comparative bar graph illustration showing conduit patency rates at 1, 5, and 10 years for the four main CABG graft types. The x-axis shows time points (1 year, 5 years, 10 years) and the y-axis shows patency percentage (0-100%). Four grouped bars at each time point, color coded: LIMA (red): 99%, 97%, 95%. Radial artery (orange): 95%, 92%, 87%. RIMA (dark red): 98%, 96%, 93%. SVG (blue): 88%, 75%, 55%. Each bar labeled with the exact percentage. Above the bars, include icons of each conduit type. A horizontal dashed line at 80% labeled "acceptable patency threshold." Below the graph, a small table showing "Causes of Graft Failure" for each conduit: SVG early (thrombosis) / intermediate (intimal hyperplasia) / late (neoatherosclerosis); Arterial grafts: competitive flow, spasm, technical error. Title: "Long-Term Conduit Patency Comparison." </image>

Key Clinical Pearls

• LIMA-to-LAD is the single most important decision in CABG and should be performed in virtually every case -- it provides the greatest survival benefit and has the best long-term patency of any conduit
• FREEDOM trial established CABG superiority over PCI in diabetic patients with multivessel disease -- diabetic patients with multivessel CAD should be strongly counseled toward CABG
• The radial artery is the preferred second arterial conduit -- RADIAL and RAPCO trials support its use for target vessels with >= 70% stenosis; postoperative CCB for 6-12 months prevents spasm
• No-touch SVG harvesting technique significantly improves vein graft patency -- surgeons should adopt this technique when SVG is used
• Off-pump CABG has not demonstrated a clear advantage over on-pump in randomized trials and is associated with higher incomplete revascularization rates -- use should be reserved for specific clinical scenarios (porcelain aorta, renal failure) by experienced operators
• The heart team discussion is mandatory for all patients with left main or multivessel disease -- the decision between CABG and PCI must integrate anatomic complexity (SYNTAX score), patient comorbidities, conduit availability, and patient preferences

References

• Lawton JS, et al. 2021 ACC/AHA/SCAI Guideline for Coronary Artery Revascularization. Circulation. 2022;145:e4-e17.
• Farkouh ME, et al. Strategies for Multivessel Revascularization in Patients with Diabetes (FREEDOM). NEJM. 2012;367:2375-2384.
• Gaudino M, et al. Radial-Artery or Saphenous-Vein Grafts in Coronary-Artery Bypass Surgery (RADIAL). NEJM. 2018;378:2069-2077.
• Velazquez EJ, et al. Coronary-Artery Bypass Surgery in Patients with Ischemic Cardiomyopathy (STICH). NEJM. 2016;374:1511-1520.
• Taggart DP, et al. Bilateral versus Single Internal-Thoracic-Artery Grafts (ART). NEJM. 2019;380:437-446.

Read this lecture as Markdown