The era of single-strategy revascularization surgery appears to be a thing of the past. Techniques such as the systematic use of double internal mammary artery without combining other grafts, or the exclusive use of the left internal mammary artery with multiple venous grafts, have yielded significant outcomes in revascularization surgery. However, although this is a highly standardized procedure, where systematic repetition is key to achieving consistently reproducible results, pursuing the same strategy without adjusting to different patient profiles seems to be a mistake as of today.

Various studies, previously discussed on this blog and led by Gaudino, the first author of this consensus document, have suggested this. However, the significance of this work lies in its nature as an intersocietal consensus document, involving what might be termed the “NATO of cardiac surgery”: the EACTS and the STS together. Such works lend consistency and authority to the statements they contain, underscoring the principle that unity strengthens the argument. In a highly competitive field, such unity becomes particularly valuable, especially in the absence of a new update to revascularization guidelines. Additionally, documents like this provide greater depth to the treatment of a topic that, although mentioned in the latest clinical guidelines, this is addressed superficially.

The document comprises a systematic review of studies focused on clinical outcomes and patency of various coronary graft types, each analyzed individually:

1. Radial Artery: Recent randomized studies have shown excellent patency rates for the radial artery, with a comparison to the saphenous vein yielding an HR = 0.44, p < 0.001. In terms of survival, observational studies using the radial artery as a second arterial graft demonstrated a 26% increase in 6-year survival, compared to the same revascularization strategy using the saphenous vein. The radial artery was also associated with an HR = 0.67 and 0.73 at 5 and 10 years, respectively, in the composite event of recurrent myocardial infarction and need for revascularization. In the RAPCO study, radial artery outcomes were superior to those of the right mammary artery as a second arterial graft, showing better patency at 10 years (HR = 0.45) and a lower rate of myocardial infarction and/or revascularization procedures at 15 years (HR = 0.74). It is worth noting that in most studies, the radial artery was anastomosed to the severely diseased circumflex territory, with limited experience in the right territory.

In terms of harvesting technique, it is generally well tolerated in terms of hand function, although performing an Allen test beforehand is recommended, given the variability in test outcomes between operators. Some studies have suggested the benefits of prior echocardiographic assessment to evaluate development, patency, and preexisting calcification before harvesting. In the absence of comparative studies, a common-sense approach prevails, selecting grafts of appropriate quality, with preserved ulnar collateral circulation and without formal contraindications for use (patients with Dupuytren’s disease, previous or untreated carpal tunnel syndrome, or any forearm trauma; prior catheterization; potential use for vascular access for dialysis; or vasospastic disorders such as Raynaud’s phenomenon). The literature reports a complication rate of approximately 9%, all minor (paresthesias, local pain, less frequently, grip strength loss or distal claudication), although publication bias is likely.

Endoscopic harvesting has been suggested as feasible in expert hands, although it is associated with a higher incidence of endothelial damage, spasm, and challenges in controlling side branches. Given that the radial artery has less collateralization and a more predictable course than the saphenous vein, it may potentially benefit from this technique more than venous grafts. However, solid evidence is scarce and subject to publication biases, so no conclusions are drawn on this aspect.

As for its use, there is insufficient evidence regarding proximal anastomosis. It is typically configured in T or Y anastomosed to the left internal mammary artery, although direct aortic configurations are also common. This strategy may protect the graft from competitive flow but exposes the proximal anastomosis to increased stress. Therefore, using a smaller punch size than that typically employed with the saphenous vein is recommended. The authors do not mention configurations yielding favorable results, such as direct anastomosis immediately over the anastomotic head of a previously anastomosed venous graft, creating a V-configuration. This approach mitigates direct stress on the radial artery and minimizes dependence on the donor vein segment, which might otherwise compromise patency. For distal vessels, it is preferable to use them in vessels with severe proximal stenosis, >70% in the left territory and >90% in the right, to prevent competitive flow. These criteria were established in studies such as RAPS and RAPCO, which included them in their selection criteria. Regarding functional assessment, the IMPAG study set a cutoff of FFR >0.78, although most arterial grafts used were left internal mammary arteries rather than radial arteries, leaving limited data on this specific point. Arguably, a functional study likely suggests intermediate lesions (50-70%) for which this graft might not be advised, regardless of the FFR outcome.

Finally, vasodilator use is recommended for at least one year post-revascularization surgery with dihydropyridine calcium channel blockers (nifedipine, amlodipine). These agents should be carefully titrated for clinical tolerance (20% incidence of headache) and to not to interfere with the use of other pharmacological agents with proven prognostic value in these patients (beta-blockers in the presence of prior infarction, ACE inhibitors/ARBs/ARNI in the presence of ventricular dysfunction).

1. Right Internal Mammary Artery (RIMA): For many years, RIMA was seen as a promising second arterial graft that could demonstrate benefits over the strategy of a single internal mammary artery graft. A meta-analysis by Benedetto et al., encompassing 9 clinical trials with follow-up periods of more than 4 years, indicated a fourfold increase in patency for RIMA compared to saphenous vein grafts (OR = 0.25). Another meta-analysis by Gaudino et al. found a patency rate of 90.9% for RIMA at over 5 years. Clinically, the ART study showed no difference at 10 years between the use of one or two internal mammary arteries. However, a crossover rate of approximately 20% between groups and the inclusion of patients with additional radial artery grafts in both groups compromised the ability to draw solid conclusions regarding the primary objective.

Despite this, observational evidence has consistently shown the benefits of using RIMA as a second arterial graft in terms of reducing adverse long-term events (>10 years). However, while the rate of early occlusion for RIMA is often lower than that of venous grafts, the increased risk of mediastinitis must be weighed against the patient’s survival probability, especially in three specific clinical scenarios:

• Diabetic patients: Female, obese, and diabetic patients present the highest risk of mediastinitis, estimated at around 3%, double that of non-diabetic patients, though this may reflect publication bias.
• Low ejection fraction: Immediate arterial graft performance is not always optimal, which may lead to myocardial hypoperfusion issues in the early postoperative period. Saphenous vein grafts tend to provide better early revascularization performance, making a combined strategy of arterial and venous grafts preferable in these patients rather than pure arterial procedures.
• Advanced age: Given the lack of differential outcomes associated with the use of multiple arterial grafts within periods shorter than 10 years, using RIMA in elderly patients might increase procedural complexity and risk without a clear benefit. Studies have found similar mediastinitis rates in older and younger patients; however, while there is no consensus on an age cutoff, registries suggest diminished survival benefits from multiple arterial graft strategies in patients over 70 years.

Regarding technical utilization, the insights and solutions for RIMA usage are similar to those expressed for the radial artery. It is worth mentioning that T- or Y-composite configurations with proximal anastomoses on the aorta allow for more distal target reach. In contrast, arterial target selection with RIMA offers more flexibility than with the radial artery. Although some studies suggest better tolerance to beds with intermediate lesions, others highlight a performance compromise associated with competitive flow. The IMPAG study set a minimum FFR of >0.78 for considering RIMA implantation. However, despite histological and autoregulatory differences from the radial artery, RIMA grafts benefit from being used in vessels with adequate diameter (>1.93 mm) and severe lesions (>70%).

The authors also explore the technique of crossing RIMA in situ to the left anterior descending artery (LAD), combined with the left internal mammary artery for the rest of the left coronary territory. Limited evidence supports this configuration, with only four studies showing no difference compared to multiple arterial graft strategies while preserving LAD anastomosis with the left internal mammary artery. However, two major limitations are recognized: limited graft length to reach the LAD and potential injury risk crossing the midline during reoperation. With advances in coronary and structural intervention, reoperations may become less frequent in patients with patent grafts. Nonetheless, no recommendations are made regarding these techniques. Other approaches, such as extending RIMA with another graft (saphenous vein or radial artery) to achieve complete revascularization with sequential anastomoses (snake), are not considered.

Lastly, the authors reflect on RIMA harvesting techniques. Although skeletonization has been shown to reduce sternal complication rates, conclusions on its impact on clinical outcomes are limited due to a lack of data. Skeletonization may alter graft vasomotility through increased denervation and manipulation-related damage, though it allows for better graft length and enables exclusion of the distal segment, which tends to be more spastic and less developed. Variability in techniques, including the use of electrocautery versus harmonic scalpels or exclusion methods for collateral branches, limit the authors’ ability to reach a conclusive judgment.

1. Saphenous Vein (SV): While the left internal mammary artery remains the gold standard for revascularization, the saphenous vein remains the benchmark for comparing arterial graft outcomes. Though clearly inferior to the radial artery in optimal territories, SV grafts have shown comparable results to RIMA in a prior meta-analysis by Gaudino, reviewed previously on this blog.

With a patency rate of 82% at 5 years, the SV graft has proven successful. Especially given its versatility, which is both its primary strength and limitation, often leading to indiscriminate target selection. Cautionary measures, similar to those used with the radial artery, have yielded favorable outcomes. However, the comparable outcomes between SV and RIMA may suggest that target vessel characteristics play an equally, if not more, significant role in graft selection, with graft nature only becoming a determinant in the long term.

The authors analyze two aspects of SV graft use: endoscopic and open extraction techniques. Endoscopic harvesting substantially reduces wound complications (the most common complication in revascularization surgery) but may lower patency rates due to endothelial and structural damage, although clinical event translation is yet unclear. More evidence is needed to support endoscopic harvesting, as publication biases likely influence this inconsistency. Current clinical guidelines discourage endoscopic SV extraction. In contrast, randomized studies show better patency with the “no-touch” technique, although clinical events remain unaffected. However, this approach increases wound complication rates, as SV is harvested with surrounding fatty tissue, impacting surgical wounds.

In summary, the authors highlight that revascularization strategy should be “à la carte” rather than a “fixed menu,” tailored to each patient’s individual needs and characteristics.

COMMENTARY

This study provides an excellent update and stands as the best consensus on the topic. Some recommendations stemming from a similar analysis, such as conducting surgery with or without extracorporeal circulation support and other technical aspects, are missed. However, it is evident that, after a comprehensive review, this individualized approach will help tailor the procedure to each patient’s specific needs. And although revascularization surgery continues to involve grafting vessels beyond proximal lesions, we are transitioning from a single approach to multivessel disease revascularization to over 15 variations of the same intervention. This heterogeneity will need to be incorporated into future research, with registry-based analysis (such as the Spanish Registry of Cardiac Surgery: RECC) and big-data strategies likely offering a path to new scientific evidence. Until then, the left internal mammary artery to the left anterior descending artery remains the gold standard in myocardial revascularization.

REFERENCE

Gaudino M, Bakaeen FG, Sandner S, Aldea GS, Arai H, Chikwe J, et al. Expert Systematic Review on the Choice of Conduits for Coronary Artery Bypass Grafting: Endorsed by the European Association for Cardio-Thoracic Surgery (EACTS) and The Society of Thoracic Surgeons (STS). Ann Thorac Surg. 2023 Oct;116(4):659-674. doi: 10.1016/j.athoracsur.2023.06.010.