Introduction

Coronary artery disease (CAD) is the most common cause of death worldwide making it one of the most important public health issues [1]. Treatment options include medical management, percutaneous coronary intervention (PCI), and CABG surgery. For patients with complex multivessel CAD, CABG is the treatment of choice, making it one of the most commonly performed surgeries in the world, with about 400,000 procedures performed in USA alone [2, 3]. The effectiveness of coronary surgery is directly related to bypass graft patency. Arterial grafts have better patency than vein grafts and thus are preferred conduits. However, in certain situations or patient populations, arterial grafting may not be the best option and saphenous vein grafting should be considered instead.

In this review, we have examined the role of saphenous vein as a bypass conduit, described mechanism and risk factors for its failure, identified strategies to improve its durability, and discussed its current role in coronary artery bypass grafting.

Saphenous vein as bypass conduit

The commonly used grafts in surgical revascularization include internal thoracic artery (ITA), radial artery, gastroepiploic artery, and the saphenous vein [4]. Saphenous vein was the conduit used in the first series of CABG and except for the revascularization of the left anterior descending coronary artery (LAD), it remains the most commonly used bypass conduit [5]. This is because it is easily harvested and is technically easy to use due to its wall characteristics and large diameter. Also, because it is long and plentiful, it can reach any coronary artery and can be used to graft multiple vessels. However, its durability and longevity are not ideal. One year after CABG, 10 to 20% of saphenous vein grafts fail [6,7,8]. From 1 to 5 years, an additional 5 to 10% fail, and from 6 to 10 years, an additional 20 to 25% fail [9]. Ten years after surgery, only about half of saphenous vein grafts are patent, and of those, only half are free of stenosis [10].

Mechanism of saphenous vein graft failure

Early saphenous vein graft failure, occurring during first year of surgery, occurs due to technical errors, thrombosis, and intimal hyperplasia [11, 12]. During initial harvesting and preparation, venous conduits undergo a period of ischemia and distention from reperfusion resulting in increases in oxidation stress and damage to the endothelial lining of the lumen. Furthermore, initial exposure to arterial pressure also results in endothelial damage [13, 14]. This leads to platelet adherence which may result in graft thrombosis and acute occlusion. The adherence of platelets to the intima also serves as the initial event in the development of intimal hyperplasia. Platelets, after adhering to the intimal surface, release mitogenic proteins leading to smooth muscle cell migration which results in intimal proliferation and hyperplasia [15,16,17,18].

Late failure, occurring after a year from surgery, occurs due to arteriosclerosis [19, 20]. Mural thrombi and intimal hyperplasia are the early stages of arteriosclerosis [8, 19]. With time, lipid accumulates in the areas of intimal hyperplasia, resulting in arteriosclerotic plaque, and eventual graft stenosis or occlusion [16, 17, 20, 21].

Risk factors of saphenous vein graft failure

Saphenous vein graft failure can be due to both, patient-related and surgical factors. Patient-related factors include demographic and clinical factors such as Hispanic race, hyperlipidemia, hypertension, renal insufficiency, and current tobacco use [22, 23]. Female gender is also a risk factor and is largely attributable to smaller coronary vessel diameter [23, 24]. Surprisingly, diabetes is not a risk factor for saphenous vein graft occlusion. A recent study assessed 20,066 saphenous grafts and 7903 ITA grafts and found no influence of diabetes on ITA or saphenous vein graft patency over more than 20 years after CABG. ITA graft patency remained stable over time, whereas saphenous vein graft patency declined progressively for patients with and without diabetes. Because patients with diabetes have more severe coronary artery stenosis, one would expect that stenosis in their bypass grafts would also be more severe than in patients without diabetes, resulting in lower graft patency. However, this does not appear to be the case [25]. This is important because the proportion of patients undergoing CABG who have diabetes have increased over the last few decades and today, diabetic patients represent nearly 40 to 50% of all patients undergoing CABG [26, 27].

Surgical factors associated with vein graft failure injury include injury to the graft during harvesting, preparation, and initial exposure to arterial pressures. Small target vessel diameter (≤ 1.5 mm) also plays a role [28]. In the Reduction in Graft Occlusion Rates (RIGOR) study population, this was a strong predictor of saphenous vein graft occlusion and was thought to be due to impaired distal run-off [23]. Low mean graft blood flow also plays a role in saphenous vein graft failure. A study by McLean et al. showed a 14% increase in the odds of saphenous vein graft occlusion for every 10 ml/min decrease in flow [23].

Strategies to reduce SV graft failure

Surgical strategies

Harvesting

Endothelial damage during vein graft harvesting is known to be an important cause of vein graft failure. Graft injury during harvesting can serve as an inciting event leading to a cascade of endothelial pathways ultimately resulting in atherosclerosis and graft failure. Therefore, efforts should be made to harvest the vein atraumatically. Endoscopic vein harvesting is an atraumatic procedure that involves harvesting a saphenous vein without direct contact or manipulation of the vein. This technique is beneficial in reducing wound-related complications and postoperative pain and morbidity compared with open vein harvest technique. However, the Randomized On/Off Bypass (ROOBY) trial showed that endoscopic vein harvesting was associated with lower 1-year SV graft patency and higher 1-year revascularization rates compared with open harvesting [29]. More recently, Hess et al. also reported higher rates of graft failure in patients who underwent endoscopic vein harvesting 12–18 months after CABG in the PREVENT IV trial [30]. The on-going Randomized Endo-Vein Graft Prospective (REGROUP) trial aims to compare endoscopic vein harvest with open harvest in terms of major adverse cardiac events and may provide definitive evidence in this regard [31]. It is important to note, however, that whether the procedure is performed via open vein harvesting or endoscopically, the key is to take the vein atraumatically, which requires a skilled operator. Due to the improved patient comfort and reduction in leg complications—which in previous years were pervasive and common—open vein harvesting may become obsolete someday. However, the proper operator training with endoscopic vein harvesting needs to occur.

An atraumatic, “no-touch” technique for harvesting the saphenous vein has also been developed. This involves harvesting the vein with a pedicle of surrounding tissue maintaining the vasa vasorae and the nerves in the adventitia. This protects the vein from spasm and prevents the need for distension. However, wound complications remain an important issue with the use of this technique [32]. A randomized trial from a single center in Sweden showed that use of SV grafts, harvested using no-touch technique compared to conventional technique, to non-LAD targets, can provide long-term patency comparable to that of left ITA to LAD [33].

Preservation

The graft preservation solutions used after procurement and before implantation are a relatively understudied and neglected area [34]. We now know that normal saline and blood-based media are not appropriate solutions for vein preservation. Buffered saline solutions should be used instead as they have shown to have lower vein graft failure rates and trend toward better long-term clinical outcomes compared with patients in which saline- or blood-based solutions are used for graft preservation [35].

External stenting of veins

Taggart et al. have recently shown that external stents have the potential to reduce the process of diffuse intimal hyperplasia in saphenous vein grafts 1 year after CABG [36]. They prevent saphenous vein graft distension when exposed to arterial pressures and hence prevent injury. In their study, 30 patients underwent on-pump multivessel CABG including a left ITA to the LAD and saphenous vein grafts to right and circumflex territories. Each patient received one external stent device to a single saphenous vein graft to either the right or the circumflex artery. Results showed that external stenting of saphenous vein graft had a significant reduction of intimal hyperplasia by approximately 15% and a reduction in intimal thickness. When comparing stented versus non-stented saphenous vein grafts, ligating saphenous vein side branches with sutures rather than metallic clips was associated with a reduction in both plaque area and thickness. The sutures also resulted in more uniform saphenous vein graft lumen in the stented group.

Pharmacological strategy after CABG

Antiplatelet and lipid-lowering therapies are clinically proven to prevent vein graft failure. Dual antiplatelet therapy with aspirin and clopidogrel after CABG has shown to improve early saphenous vein graft patency and is traditionally used [37, 38]. However, the Platelet Inhibition and Patient Outcomes (PLATO) study showed that tricagrelor may be more effective in preventing recurrent cardiovascular events compared to clopidogrel in patients with prior coronary artery bypass graft surgery [39]. Lipid-lowering agents such as statins are also recommended. The Post Coronary Artery Bypass Graft Trial compared the effects of two lipid-lowering regimens on patients who had undergone bypass surgery and reported that aggressive lowering of low-density lipoprotein (LDL) cholesterol levels to < 100 mg/dL compared with moderate reduction to 132 to 136 mg/dL delayed atherosclerosis progression in grafts [40].

When to do saphenous vein grafting?

The effectiveness of coronary surgery is directly related to coronary artery bypass graft patency. Therefore, for best outcomes, it is important to choose the best conduit for a given patient and a coronary target. The best conduit is the one that is most likely to remain patient. In most situations, arterial grafts have the best patency. However, one of the most important factors that influence arterial graft patency is competitive flow [6, 41]. As competitive flow increases, arterial graft patency decreases. A study evaluating the patency of ITA and saphenous vein grafts across a range of proximal stenosis by coronary system found that ITA grafts have better patency than saphenous vein grafts across all degrees of proximal stenosis in the left-sided coronary arteries and posterior descending artery [6]. For main right coronary artery (RCA), however, early (< 5 years) after operation, saphenous vein grafts were found to have equivalent or better patency than ITA grafts. At 10 years postoperatively, ITA grafts were more likely to be patent in RCA with 70% stenosis or greater (Fig. 1). Therefore, when bypassing RCA with less than 70% stenosis, an ITA graft may not be the best choice and saphenous vein grafting should be considered instead.

Fig. 1.
figure 1

Estimates of 1- and 10-year ITA and SVG patency according to coronary artery grafted and degree of proximal coronary stenosis. Solid lines are point estimates; dashed lines are 70% confidence intervals. Graphs depict a 65-year-old male non-diabetic undergoing CABG in 1974. a Grafts to the LAD 1 year after CABG. b Grafts to the LAD 10 years after CABG. c Grafts to the Dg, Cx, and PDA 1 year after CABG. d Grafts to the Dg, Cx, and PDA 10 years after CABG. e Grafts to the RCA 1 year after CABG. f Grafts to the RCA 10 years after CABG. Cx circumflex marginals, Dg diagonals, ITA internal thoracic artery, LAD left anterior descending coronary artery, PDA posterior descending artery, RCA right coronary artery, SVG saphenous vein graft. Reproduced with permission from Elsevier from reference 6

Full size image

Radial artery grafts and gastroepiploic artery too are affected by competitive flow but probably to a much greater degree than ITAs. A study by Di Mauro et al. showed that in patients undergoing bilateral ITA grafting of left-sided coronary arteries, long-term freedom from cardiac mortality was lower in patients with right gastroepiploic artery grafted onto the RCA when the stenosis was less than 80% compared to a saphenous graft. This difference disappeared when RCA stenosis was at least 80% [42]. Similarly, a study by Maniar et al. showed that the patency of radial arteries is worst if grafted to a moderately stenosed RCA [43]. Therefore, if a radial artery or gastroepiploic artery is being used to graft a RCA, it is important to make sure that the RCA is either critically stenosed or occluded, because the effect of competitive flow is more pronounced in the RCA system than the left. In patients with moderately stenosed RCA, saphenous vein grafting may result in best outcomes as its patency is not affected by competitive flow.

Another situation in which saphenous vein grafting may be a reasonable choice for revascularization of non-LAD targets is in patients who are not candidates for total arterial or bilateral ITA grafting. These include patients whose risk of deep sternal wound infection is too high or who are too old to derive benefit from multiple arterial grafting. Studies have shown that obese diabetic females with diffuse atherosclerotic burden are at the greatest risk of developing wound infections (Fig. 2) [44]. Therefore, it may be best to avoid use of bilateral ITA grafts in these patients and instead radial artery or saphenous vein grafting should be considered for revascularization of non-LAD lesions. Similarly, it may be best to avoid bilateral ITA grafting in octogenarians as they, due to their lower life expectancy, less likely derive survival benefit from use of multiple arterial grafts.

Fig. 2
figure 2

Effect of risk factors on probability of deep sternal wound infection (DSWI) across a range of body mass indexes. Simulations were based on the logistic regression model for a patient undergoing coronary artery bypass grafting with cardiopulmonary bypass, in the past decade, with complete revascularization. Other factors are listed for the individual depictions. a Effect of female sex after bilateral (BITA) and single (SITA) internal thoracic artery grafting. Simulations are based on an insulin-treated patient with no history of peripheral arterial disease (PAD) or myocardial infarction (MI). b Effect of PAD and MI after BITA grafting. Simulations are based on an insulin-treated woman. c Effect of diet-controlled versus medically treated diabetes. Simulations are based on a man with no history of PAD or MI undergoing CABG with BITA grafting. Reproduced with permission from Elsevier from reference 44

Full size image

Conclusions

Saphenous vein remains the most commonly used bypass graft for non-LAD targets. To maximize the benefits of coronary surgery, it is important to identify and adopt strategies that improve the durability and longevity of vein grafts. The choice of graft should be tailored to the patient coronary anatomy for best outcomes.