Several studies have confirmed that utilization of two or more arterial grafts is associated with better long-term survival compared to conventional single-arterial coronary artery bypass grafting (CABG) based on the left internal thoracic artery (LITA) and saphenous vein graft (SVG) [1, 2, 3]. Given its relatively easy harvest and lower risk of deep sternal infection compared to bilateral internal thoracic artery (ITA) [4], the radial artery (RA) has emerged as the predominant choice for a second arterial graft in both the United States and Europe [5, 6]. To improve long-term cardiac outcomes, the most recent guidelines have increased their recommendation level for employing RA grafts to class I [7].

Frequent use of the right RA for angiography prior to CABG has raised concerns about intimal tears and impending graft failure [8]. Therefore, unilateral RA grafting is often used more frequently than bilateral RA. Sequential grafting of the RA is a safe method of choice and helps to maximize the number of constructed arterial grafts. A number of studies have investigated the long-term outcomes of CABG with sequential RA grafting. These reported superior 10-year survival compared with conventional LITA and SVG [1, 9]. The RA conduit is known to be sensitive to competitive flow from the native coronary artery, which may be a key factor in premature graft loss [10]. However, multiple distal anastomoses in sequential grafts can increase the risk of coronary steal and graft flow reversal [11]. Previous studies did not account for significant heterogeneity in the target vessels of RA grafts, and hence the efficacy of sequential RA grafting remains unclear.

We previously reported that the preoperative quantitative flow ratio (QFR) of target vessels at the site of RA anastomoses was a robust predictor of occlusion [12]. Indeed, QFR proved to be a more accurate indicator of RA graft occlusion than the degree of stenosis. There is currently a lack of published literature on the different risk factors for RA graft patency. Therefore, the aim of this study was to conduct a risk-adjusted comparison of long-term patency between individual and sequential RA grafting.

This single-center retrospective cohort study investigated the long-term patency of RA bypass grafts to identify risk factors associated with graft occlusion. The sequential method of anastomoses was found to be a significant predictive factor for RA graft occlusion, as well as target vessels with a QFR $>$0.71.

Following revival of the RA conduit in the 1990s [14], its utilization in CABG has been supported by solid evidence demonstrating its clinical benefits. The RADIAL project found that patients with RA grafts as the second conduit in addition to the left ITA had significantly reduced incidence of MACCEs at both 5- and 10-year follow-up compared to SVG [2, 3]. The RA also showed advantages over the right ITA, with better long-term patency and fewer sternal-related complications [15]. Consequently, utilization of the RA was recommended in the guidelines and is frequently utilized in clinical practice as the primary choice for second arterial graft material [7, 16].

Sequential RA grafting helps to achieve complete revascularization and maximize the number of arterial grafts. However, there have been only a few studies on the patency of sequential RA grafting. Schwann et al. [9] assessed the angiographic results of sequential RA grafting in 122 patients with a median follow-up of 1.8 years, finding an overall patency of 71%. Furthermore, no significant difference in patency was observed between sequential and individual RA grafts. However, the conclusion reached by these authors is open to question, first because the angiographic follow-up was symptom-driven, and second because the sample size of nonsequential grafts in their study was limited. There are also ongoing concerns about the long-term patency of sequential RA grafts due to the coronary steal phenomenon.

Clinical and demographic characteristics were taken into account in the present study, and specifically the functional characteristics of target vessels. We observed a significant association between the sequential grafting method and the occlusion of RA grafts following CABG surgery. Several pathophysiological factors may be responsible for the superior patency observed with single conduits. First, sequential grafts may be more susceptible to blood flow reversal compared to individual grafts. In the context of RA to non-LAD bypass, Nakajima et al. [17] reported that 86.3% of competitive flow occurred at the distal end of sequential grafts. Graft occlusion in our cohort was lowest when the QFR of both the middle and distal anastomoses was $\le$0.71. However, the occlusion of sequential grafts increased when the QFR of any anastomoses was $>$0.71. Similar to the SVG sequential bypass technique [18], our strategy was aimed at targeting the largest lesion with the highest degree of stenosis, and with good run-off as the most distal anastomoses. Selection of the intermediate target may affect long-term patency of the entire sequential grafts. Second, incisions made in the middle of RA conduits may lead to additional endothelial damage and the occurrence of RA spasm. Subsequent focal neointimal hyperplasia at anastomotic sites could potentially affect the patency of the graft. Third, sequential bypass grafts carry a greater risk of kinking and positioning complications. In the present study, the majority (82.9%) of sequential grafts were anastomosed using the off-pump method, thereby mitigating potential errors associated with the measurement of graft length. Moreover, we adjusted for the use of cardiopulmonary bypass and observed no significant effect of this procedure on graft patency. Further research is warranted to address this issue in more depth.

Our results may provide an explanation for the unsuitability of extensive constructions involving more than two arterial conduits using sequential RA grafting. In a meta-analysis of 10,287 patients, Gaudino et al. [19] found that the use of a third individual arterial graft was associated with a 24% relative survival benefit. In contrast, Schwann et al. [20] did not observe a statistically significant advantage in 15-year survival in patients with three-artery grafts compared to those with two-artery grafts. The latter study consisted almost entirely of patients who received a left ITA and supplemental sequential RA grafts (bilateral, sequential, or both) rather than bilateral ITA. The construction of an extensive artery bypass based on sequential RA may be less effective clinically compared to the use of bilateral ITA and an individual RA. The higher attrition rate of sequential RA grafts observed in our study may be due to one of these underlying factors.

We also found that prior myocardial infarction was a protective factor for RA graft patency. This observation concurs with the results of subgroup analysis in the RADIAL project, which favored RA over SVG [2]. We speculate that patients who had collateral flow were prone to limited previous infarct size, and also had enhanced native competitive flow, thereby leading to decreased patency of RA grafts after CABG [21].

These findings underscore the potential clinical utility of QFR in guiding arterial graft planning during CABG. Preoperative QFR assessment may help identify target vessels with high competitive flow risk, particularly in sequential grafting, and thus assist in selecting optimal anastomosis sites to reduce the likelihood of graft occlusion. Incorporating QFR into routine surgical planning may enhance long-term graft patency and improve patient outcomes, although prospective validation is warranted.

This study has several limitations. First, its retrospective design may have included unmeasured confounding factors. Baseline differences, such as less use of cardiopulmonary bypass in the individual group, may also have introduced some bias in patency comparisons, as the choice of on-pump or off-pump grafting was based on coronary anatomy and surgical judgment. However, all procedures were performed by highly experienced surgeons who adhered to standardized protocols. Additionally, sequential anastomosis remained as an independent risk factor for occlusion after the incorporation of variables into the multivariate model. Second, some patient selection bias for RA grafting may have occurred. However, our imaging follow-up was planned and not symptom-driven, and up to 75.4% of patients successfully completed postoperative imaging follow-up. Follow-up of incomplete and non-uniform imaging may introduce bias, despite the use of time-to-event analyses that account for censoring. Third, the limited sample size and relatively short follow-up gave rise to a low incidence of MACCEs, particularly in the sequential group, making it difficult to assess the impact of RA occlusion on clinical outcomes.

Compared to individual RA bypass grafting, sequential RA was associated with inferior graft patency after CABG. Multiple target options exist for sequential grafts, and inappropriate selection may result in increased competition of blood flow and subsequent failure of the RA graft.

CABG, coronary artery bypass grafting; RA, radial artery; ITA, internal thoracic artery; SVG, saphenous vein graft; QFR, quantitative flow ratio; CCTA, coronary computed tomography angiography; MACCEs, major adverse cardiac or cerebrovascular events; IQR, interquartile range; LAD, left anterior descending; 95% CI, 95% confidence interval; aHR, adjusted hazard ratio.

The data underlying this article cannot be shared publicly due to the policy of data sharing of our centre. The data will be shared on reasonable request to the corresponding author.

SZ, ZH, and ZZ contributed to the conception and design of the study. SZ and CW performed the data analysis. SZ drafted the manuscript. ZH, YZ, and ZZ provided critical revisions to the manuscript. All authors contributed to the conception and editorial changes in the manuscript. All authors participated in the editing process and approved the final version of the manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was carried out in accordance with the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Fuwai Hospital (No. 2021-1554). Due to the retrospective nature of the study, written informed consent was waived.

This study was supported by grants from National High Level Hospital Clinical Research Funding (2024-GSP-TJ-01), Beijing Natural Science Foundation - Fengtai Innovation Joint Fund (L251022), and the high-level hospital clinical research of Fuwai Hospital, Chinese Academy of Medical Sciences (2023-GSP-RC-14,2023-GSP-QN-27).

The authors declare no conflict of interest.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/RCM39689.