Key result
Radial artery grafting for coronary artery bypass surgery appears to improve late survival compared to saphenous vein grafting in observational studies, though definitive randomized evidence is lacking.
Why the study?
Does radial artery grafting improve late survival compared to saphenous vein grafting in patients undergoing CABG?
Does radial artery grafting improve late survival compared to saphenous vein grafting in patients undergoing CABG?
Answers are grounded in this paper and cite their sources.
Retrospective studies suggest a survival benefit for radial artery over saphenous vein grafts in CABG, but definitive randomized trials powered for clinical outcomes are still needed.
Unequivocal evidence showed that the use of the left internal thoracic artery (ITA) to the left anterior descending coronary artery improves survival when compared with the saphenous vein (SV) [ 1 , 2 ]. Owing to its histological and biochemical characteristics, the ITA is almost immune to atherosclerosis and seems to protect the downstream native coronary system from disease progression. Furthermore, numerous observational studies and a recent meta-analysis have shown that using the bilateral ITA (BITA) improves long-term survival [ 3 ]. However, <5% of US surgeons in 2005 used BITA grafting although in Europe this rate is probably higher [ 4 ]. Greater technical complexity, increased risk of sternal wound complications associated with BITA grafting, particularly in diabetic and obese women, and perhaps the absence of long-term results from a randomized controlled trial (RCT) comparing single ITA with BITA grafting, can explain this low rate of BITA grafting. ITA properties, good matching size and production of endothelial modulators were assumed to extend to other arteries. The radial artery (RA) was initially explored in the 1970s with disappointing results. After modifications in harvesting and preparation, and use of vasodilators, Acar et al . [ 5 ] reinvestigated RA grafting, with a revival of its use as a second arterial conduit. The RA, owing to its length, ease of use and low harvesting-related morbidity, became a good alternative to the SV and its associated late degenerative disease. Despite a higher incidence of atherosclerosis, intimal hyperplasia and medial calcification and spasm when compared with ITA, observational studies, several RCTs and meta-analysis demonstrated improved RA patency after 4 years, when compared with the SV [ 6 , 7 ]. However, some observational studies reported a high rate of functional RA failure, closely associated with the degree of stenosis of native target coronary artery [ 8 , 9 ]. Two RCTs comparing RA with SV graft patency showed no statistical difference. While one RCT reported only 1-year results, the Radial Artery Patency Comparative Outcomes (RAPCO) trial, conducted by highly experienced surgeons in Australia, showed equivalent survival and event-free survival and graft patency in both arms at a 6-year follow-up [ 10 , 11 ]. Importantly, a recent network meta-analysis showed significantly better complete and functional RA patency compared with SV patency in studies with more than 4 years of follow-up [ 12 ]. However, proving that better late RA patency translates into improved survival and event-free survival is difficult to demonstrate. Indeed, there were only three RCTs comparing RA and SVG that studied clinical outcomes and only one demonstrated superior clinical outcomes with RA grafts [ 9 , 11 , 13 ]. In this issue of the Journal , three retrospective studies addressed the issue of RA use in coronary artery bypass grafting, and its impact on late survival [ 14–16 ]. Shi et al. [ 14 ] reported a study cohort of 3220 patients receiving at least 1 RA to supplement a single in situ ITA whereas 786 received only veins to supplement a single ITA. At 15 years, after propensity-score matching of 507 patient pairs, the RA group showed significantly better survival. Among 2149 patients ≥70 years, the RA protected from late death and, in 327 matched pairs, survival was significantly better in the RA versus SV group. Schwann et al. [ 15 ] reported a large retrospective analysis of 8220 patients in 3 unequal groups: 4484 receiving left ITA (LITA) with additional vein grafts (single arterial bypass grafting or SABG), 3095 LITA with one or more RA grafts (RA group) and 641 LITA with one or more RITA grafts (RITA group). After propensity-score matching, 551 RA, BITA and SABG triplets were obtained. Despite propensity-score adjustment, there remained significant differences between groups, SABG group patients being older and having a significantly lower completeness of revascularization index, leading to further multivariate regression analysis. Late survival was equivalent in the RA and BITA cohorts, and both significantly better than for SABG. Grau et al. [ 16 ] assessed the impact of adding a third arterial graft (RA) in the setting of BITA grafting. From an initial cohort of 4370 patients with 3-vessel disease, 2 groups (BITA + RA vs BITA + SV) of 183 patients could be matched. When the study period was divided into three eras, there was a huge variation in the incidence of BITA + RA grafting (19, 73 and 8%, respectively). There was no significant difference in long-term survival between the two groups over the 14-year period. However, before year 10, the BITA + SV group had a trend towards higher survival, whereas follow-up after year 10 showed a trend favouring the BITA + RA ± SV patients. All three studies were retrospective, two being multicentric, encompassing a very large period, from 13 to 24 years, and used propensity-score analysis to adjust for baseline imbalances. Matching minimizes but does not eliminate differences between groups. No amount of statistical adjustment can compensate for unaccounted or unquantifiable factors such as frailty, conduit selection, target artery quality, harvesting technique, vasodilators use, pre-existing RA atherosclerosis or clinical judgement. Furthermore, in studies encompassing a wide time-span, more recent patients tended to receive more arterial grafts and benefit from improved management techniques. Also, the matching process, aiming to eliminate differences in preoperative characteristics, significantly decreased the number of patients from the initial cohort, greatly diminishing generalizability. In summary, the majority of randomized studies comparing RA and SV graft patency, in selected patients, suggest that the RA is superior. For surgeons familiar with RA use, late survival seems improved but the level of evidence is low. Do these three studies increase the level of evidence and will they have an impact on clinical practice? As it is impossible to eliminate with certainty that patient selection, technical issues and surgical experience influence the results of observational studies, even using adjustment methods like propensity-score analysis, retrospective observational studies will unlikely change the level of evidence, but clinical decision-making is often based on incomplete evidence. Till late results of randomized studies, powered to test survival and major adverse events are known, selection of the second (and third) graft will be guided by patient characteristics, coronary disease patterns and surgeon preference. The 2014 ESC/EACTS guidelines on myocardial revascularization state that total arterial revascularization should be considered in patients with reasonable life expectancy (level of evidence B) [ 17 ].
No takes yet. Share an insight, caveat, or question.
Sousa‐Uva et al. (2015) conducted an editorial in Coronary artery disease. Radial artery grafting vs. Saphenous vein grafting was evaluated. Radial artery grafting for coronary artery bypass surgery appears to improve late survival compared to saphenous vein grafting in observational studies, though definitive randomized evidence is lacking.
Synapse has enriched 10 related papers. Consider them for comparative context: