Page 86 - Read Online

Page 2 of 18 Browne et al. Vessel Plus 2024;8:19 https://dx.doi.org/10.20517/2574-1209.2023.126

INTRODUCTION
Coronary artery bypass grafting (CABG) surgery is one of the most common cardiac surgical procedures
[1]
that improves clinical outcomes for patients with coronary artery disease . The left internal mammary
artery (LIMA) is often the conduit of choice in CABG surgery, specifically when grafted to the left anterior
[2,3]
descending artery due to superior long-term patency rates . While the use of saphenous veins remains
common for supplementary grafts, surgeons often consider using supplementary arterial grafts instead,
primarily the right internal mammary artery (RIMA) or the radial artery. This choice is primarily based on
the arterial grafting hypothesis, which posits that arterial conduits should yield better patient outcomes than
venous conduits. Recent meta-analyses of trials with angiographic data suggest that the radial artery might
be the optimal conduit to the left circumflex or right coronary artery territories. This contradicts the
widespread belief that RIMA, due to its biological equivalence to the LIMA, would be the logical choice .
[4,5]
However, recent reports of suboptimal rates of angiographic RIMA failure , and adverse clinical outcomes
[6,7]
related to skeletonized RIMA harvesting are worrisome . The latest surgical guidelines have given a
[8,9]
Class 1 (Level of evidence B-R) recommendation to the use of the radial artery over the saphenous vein [10,11] .
However, determining the best conduits for the left circumflex or right coronary territories remains
challenging due to varying surgical characteristics such as conduit availability and quality, grafting
[12]
configurations, and severity of coronary artery disease . In this narrative review, we aim to provide a
comprehensive summary of the angiographic and clinical outcomes of the most used conduits for coronary
artery bypass grafting.

RELATIVE EFFICACY OF CABG CONDUITS
Individual patient meta-analyses
A recent individual patient meta-analysis (Gaudino et al. ) of seven randomized trials (ACTIVE ,
[14]
[13]
[15]
[18]
[6]
[17]
[16]
CASCADE , COMPASS CABG , DACAB , POPular CABG , PREVENT IV , TARGET ) that
[19]
included 4,413 patients and 13,163 grafts, revealed that graft failure remains common 1 year after CABG
surgery (33.7% of patients had at least 1 failed graft and 16.6% of all grafts had failed). Graft failure rates by
conduit were 9.7% (387/4,006) for LIMA, 13.8% (21/152) for radial artery, 19.7% (172/8,740) for saphenous
vein, and 23.0% (61/265) for RIMA [Table 1]. Unexpectedly, the RIMA had the highest failure rate of the
conduits. With regard to target vessel location, RIMA graft failure rates were 17.6% (10/56) when grafted to
the left anterior descending artery, 24.1% (39/162) to the left circumflex artery, and 25.5% to the right
coronary artery. Graft failure was strongly associated with myocardial infarction, repeat revascularization,
and an increased risk of mortality, indicating graft status may be a useful surrogate for these clinical
outcomes.

A previous individual patient data meta-analysis (Gaudino et al.) of six other randomized trials (RAPCO ,
[20]
[25]
RAPS , RSVP , Petrovic , Stand-in-Y , Song ) from the Radial Artery Database International
[22]
[21]
[23]
[24]
Alliance (RADIAL) database compared long-term rates of graft failure in 1,091 patients and 2,281 grafts .
[4]
After a lengthy angiographic follow-up of 5.4 years, graft failure rates were 2.3% (21/921) for LIMA, 9.4%
(67/710) for radial artery, 13.5% (10/74) for RIMA, and 17.5% (101/576) for saphenous vein grafts . When
[4]
grafted specifically to the left circumflex artery, failure rates were 8.7% (38/439) for radial artery, 14.5%
(8/55) for RIMA, and 16.4% (49/315) for saphenous veins. The combined rank order from both meta-
analyses was LIMA (1st) > radial artery (2nd) > saphenous vein and RIMA (3rd or 4th), suggesting overall
relative rates of RIMA graft failure are generally suboptimal (i.e., comparable to vein grafts).
It is worth noting that the rates of graft failure are much higher for all conduits in the more recent Gaudino
2023 meta-analysis compared with the earlier Gaudino 2020 meta-analysis despite a shorter follow-up
[Table 1]. A key difference between these meta-analyses is that 5 of the 6 trials included in the 2020 analysis