Adjunctive Radial Artery Grafts No Better Than SVGs in Multivessel CABG

New long-term data from the CSP 474 trial offers a challenge to some of the conventional wisdom around graft choice in CABG surgery and may further the debate over recent guideline recommendations.

Presenting long-term mortality numbers from the VA study at the European Association for Cardio-Thoracic Surgery (EACTS) 2022 conference last week, researchers showed that when an adjunctive graft was needed in patients with multivessel coronary artery disease, use of the radial artery instead of a saphenous vein did not confer a survival advantage.

Even with the new results, though, surgeons should, when possible, be aiming to perform multiarterial grafts in patients with multivessel coronary disease, say researchers, although surgical skills should drive decisions as to the optimal conduit.

“Yes, multiarterial grafting should be encouraged, but it should be used in select patients, in select coronary targets, and by experienced surgeons,” lead researcher Faisal Bakaeen, MD (Cleveland Clinic, OH), told TCTMD. “At your center, if you’re comfortable using radial grafts or any arterial grafts, such as the right internal thoracic artery, please continue to use them because surgeon experience is an important determinant of outcomes.”

But if surgeons instead go with a saphenous vein graft, for whatever reason, these late CSP 474 data suggest that survival is not compromised.

Guideline Implications

The most recent European guidelines on myocardial revascularization recommend a strategy that incorporates multiple arterial grafts in appropriately selected patients, with the additional arterial graft, typically the radial artery, preferred over the saphenous vein following an initial left internal thoracic artery (LITA) to the LAD (class I, level of evidence B). Bilateral internal thoracic artery grafting is also recommended if patients don’t have a high risk for sternal wound infection (class IIa, level of evidence B).

Graft choice was more contentious in the United States guidelines, where the Society of Thoracic Surgeons and American Association for Thoracic Surgery (AATS) recently declined to endorse the 2021 guidelines for coronary artery revascularization put forward by the American College of Cardiology (ACC), American Heart Association (AHA), and the Society for Cardiovascular Angiography and Interventions (SCAI), in part due to the class 1 recommendation that the radial artery should be preferentially used as the adjunctive CABG conduit in multivessel CAD. The STS position is that when a second arterial graft is needed beyond the LITA-to-LAD graft, it should be either the right internal thoracic artery or radial artery, although the strength of that recommendation is tempered (class IIa, level of evidence B).

As Bakaeen explained to TCTMD, “when you give something a class I recommendation, it becomes the standard of care, but it’s not a generalizable standard of care.”

No Mortality Difference at 17 Years

CSP 474 originally included 757 patients (mean age 61 years, 99% male, 90% white) undergoing elective CABG surgery between 2003 and 2009 at 11 Veterans Affairs (VA) medical centers in the United States. For all surgeries, the LITA was used to graft the LAD when possible and the other vessels grafted with either the radial artery or saphenous vein. Nearly 60% of secondary grafts involved either the radial artery or saphenous vein to the circumflex artery, while a little less than one-third were to the right coronary artery.

The primary endpoint of the original study—angiographic graft patency at 1 year—showed there was no difference between the two procedures. For this follow-up analysis, researchers evaluated long-term mortality in 733 patients showing that, over more than 17 years, there was no significant difference in the risk of all-cause mortality between those who received a radial artery graft and those who received a saphenous vein graft (HR 0.99; 95% 0.8-1.2).

In an analysis of graft patency at 5 years, assessed with angiography in a subset of 196 patients whose grafts were patent at 1 year, occlusions were seen in three patients treated with a radial artery and two with a saphenous vein graft.  

Bakaeen said the VA database only allowed them a look at all-cause mortality, which is a both a strength and weakness of the current study.

“We can’t rule out a potential benefit of the radial artery in the long run when it comes to major adverse cardiovascular events, because we didn’t have reliable data,” he said. “Neither did we have reliable data on patency. The incomplete patency results that we do have seem to indicate an absence of signal with respect to the superiority of the radial graft, but we can’t say with certainty that we can rule out a radial benefit.”

Real-world Patients in VA Study

In a patient-level meta-analysis published in 2018, known as the Radial Artery Database International Alliance (RADIAL), investigators showed that use of the radial artery over the saphenous vein was associated with a 33% lower risk of adverse cardiac events. In follow-up angiography, which was performed after a mean of 50 months, the radial artery was associated with significantly lower rates of occlusion compared with the saphenous vein.

Mario Gaudino, MD (NewYork-Presbyterian/Weill Cornell Medicine, NY), who led that analysis, told TCTMD that CSP 474 is the only trial that has not shown any difference in graft patency at 1 year, so the absence of a mortality benefit in long-term follow-up isn’t surprising.

He pointed out that using the radial artery is associated with a lower risk of MI and repeat revascularization (28% and 50% reductions in the RADIAL meta-analysis, respectively), but wasn’t associated with any difference in survival. Any potential mortality effect would be quite small or negligible, said Gaudino in an email, “as the status of a non-LAD target is unlikely to affect survival.” MI and repeat revascularization are important outcomes, he said, and “mortality alone does not tell the whole story.”

These CSP 474 data notwithstanding, Gaudino stressed the overall risk-benefit of radial artery grafting is very favorable, noting that the incidence of harvesting-site complications is very low as opposed to using the right internal thoracic artery.

Cardiac surgeon Stephen Fremes, MD (University of Toronto/Sunnybrook Health Sciences Centre, Canada), said the radial artery should be routinely preferred over the saphenous vein “when used to treat a moderate-sized, left-sided target vessel with a high-grade lesion or right-sided vessel with subocclusive disease.”

Patients also must be good candidates for the radial grafting, such as by having adequate ulnar collateral circulation and a radial artery at least 2-3 mm in diameter on noninvasive imaging. Obesity, diabetes, and sex don’t impact the decision to use the radial artery, but advanced chronic kidney disease and peripheral vascular disease are both contraindications. It’s also relatively contraindicated in patients with poor LV function, he said.

“Anatomically, the [radial artery] is usually constructed as an aortocoronary graft,” said Fremes in an email. “In that configuration, it can typically reach to all coronary targets but may not reach to targets in the posterolateral territory, on the opposite site of the heart from the ascending aorta. When used as a composite graft, it can reach all territories.”

In 2004, Fremes, along with first author Nimesh Desai, MD (Sunnybrook Health Sciences Centre), published a study showing that use of the radial artery was associated with a lower rate of graft occlusion compared with saphenous vein grafts. Fremes also noted that he was involved in developing the ACC/AHA/SCAI guidelines for coronary revascularization. While the surgical societies chose not to endorse the guidelines, Fremes, who remained a listed co-author on them, believes they ultimately got it right with respect to the radial artery class I recommendation.   

VA Study Achieved Excellent Outcomes

To TCTMD, Bakaeen said studies supporting the radial artery over the saphenous vein for secondary grafts are largely observational, and that randomized controlled trials conducted to date have been small and performed at centers where operators are experienced in performing radial artery grafting. Those results, said Bakaeen, are not really generalizable to all operators, leaving some uncertainty about the relative merits of radial grafting. In addition to those concerns, most studies lacked robust follow-up and event adjudication, with only the RAPCO study designed to assess outcomes long-term.

The CSP 474 study is the largest study to date and includes surgeons working at VA centers, making it more generalizable with respect to US practice patterns, he said. He pointed out that the 1-year patency rate with either the radial artery or saphenous vein grafts was 89%, which is comparable to outcomes achieved by the very best hospitals.

“Nobody can argue that these were suboptimal procedures or a lower-standard surgery,” said Bakaeen. “In fact, I think they received excellent operations.”

For Bakaeen, an ideal patient for a radial graft is one who is younger, or who has a life expectancy beyond 10 years, has a severely diseased and critical non-LAD target, and anatomy favoring harvesting of the radial artery. When using the radial artery to bypass a severe blockage, this is less likely to result in competitive flow, the phenomenon where there is an equilibrium between the residual flow through the native coronary artery and flow through the bypass graft at the anastomosis. Competitive flow, he said, is an important predictor of radial graft patency.