Twenty-Five Years of No-Touch Saphenous Vein Harvesting for Coronary Artery Bypass Grafting: Structural Observations and Impact on Graft Performance

Abstract

The saphenous vein is the most common conduit used in coronary artery bypass grafting (CABG) yet its failure rate is higher compared to arterial grafts. An improvement in saphenous vein graft performance is therefore a major priority in CABG. No-touch harvesting of the saphenous vein is one of the few interventions that has shown improved patency rates, comparable to that of the left internal thoracic artery. After more than two decades of no-touch research, this technique is now recognized as a Class IIa recommendation in the 2018 European Society of Cardiology and the European Association for Cardio-Thoracic Surgery guidelines on myocardial revascularization. In this review, we describe the structural alterations that occur in conventional versus no-touch saphenous vein grafts and how these changes affect graft patency. In addition, we discuss various strategies aimed at repairing saphenous vein grafts prepared at conventional CABG.

Keywords: Cardiology; Coronary Artery Bypass; Guidelines; Mammary Arteries; Myocardial Revascularization; Saphenous Vein.

Fig. 1

Differences in patency, left internal thoracic artery (LITA) vs. no-touch (NT) saphenous vein (SV) and LITA vs. conventional (C) SV at 1.5, 8.5, and 16 postoperative years. The six confidence intervals (a-f ) and the margins of 10 and 15 percentage units are the basis for comparing LITA with the SV with respect to potential equivalence and non-inferiority (Samano et al., 2015)^[17]. CI=confidence interval.

Fig. 2

Examples of no-touch (NT) and conventional (C) saphenous vein (SV) grafts. The left panels show representative SV explants using both harvesting techniques. NT SV has its surrounding cushion of fat intact and has not been distended. C SV has the fat removed and has been distended to overcome spasm. The transverse section of NT SV shows an intact surrounding cushion of perivascular fat (PVF), an undamaged adventitia (ADV) and thick media (TM). As this vessel has not been distended, the lumen (L) is thrown into folds. The section of C SV exhibits various forms of damage. Much of the ADV has been stripped off, almost to the level of the external elastic lamina (small arrows). The media is thinner than that of the NT SV and L is grossly dilated, both due to high pressure intraluminal distension (Kopjar et al., 2016)^[22].

Fig. 3

Histological, cellular, and ultrastructural comparison of notouch (NT) vs. conventional (C) saphenous vein (SV) grafts. Top panels are representative examples of part transverse sections through the wall of NT and C SVs stained with Elastic van Gieson and prepared for coronary artery bypass grafting. The intima surrounding lumen (L) of NT SV is thrown into folds. The vein ‘wall’ is thick with smooth muscle cells of the media (M) separated by the internal and external elastic laminae and with an intact adventitia (ADV) in which the vasa vasorum is located (red arrow). The intimal folds are absent in C SV, the vessel wall is thinner than that from NT SV (due to distension), and the ADV is mostly removed and damaged. The white arrows near vein L indicate the internal elastic lamina. The lower panels show damage caused to C SV when compared to NT SV. Panel A shows an intact endothelial lining of the NT vein L when compared to the C vein shown in panel D. Only a proportion of the endothelial cells in D stain red using the antibody CD31 (arrow). Panel B is a transmission electron micrograph showing the uniform shape and distribution of vascular smooth muscle cells in the M of a NT SV. The appearance of smooth muscle cells in the M of C SV is very different. Panel C shows the wall of a NT SV with an intact vasa vasorum (arrows) within the ADV. The ADV of C SV (F) is mostly removed with remnants indicated by the arrows. Endothelial cells of L and vasa vasorum stain dark brown in these sections with an intact layer lining the NT vein, but regions of denudation present in the C vein. The endothelial cells of the vasa vasorum at the M/ADV border are evident in panel C, but absent in panel F (Ahmed et al., 2004)^[25].

Fig. 4

Unsupported and supported saphenous vein (SV) grafts for coronary artery bypass grafting. A) An example of a conventionally harvested SV with outermost tissue removed (Rueda et al.^[36], 2008). Insert, examples of DacronTM ‘EXTENT’ and ‘VEST’ external supports (Mawhinney et al.^[58], 2017). B) An example of a no-touch (NT) harvested SV with external tissue intact (Rueda et al.^[36], 2008). C) Angiogram of an SV with VEST at 12-month follow-up (Mawhinney et al.^[58], 2017). D) NT SV harvesting prevents kinking of excessively long graft (Rueda et al.^[36], 2008).

Fig. 5

Saphenous vein (SV) histology of endoscopic vein harvesting (EVH) vs. no-touch (NT) techniques. Left panel is a transverse section of an SV prepared with EVH. The perivascular layers and endothelium are damaged. The arrow shows intimal tearing (Kiani et al.^[69], 2011). Middle panel to the left shows a diagram with a small incision above the knee used for insertion of EVH instruments. To the right, it shows a long incision in thigh and calf used for both open vein harvesting and NT harvesting. Right panel is a transverse section of a NT harvested SV with perivascular fat and adventitia intact and endothelium undamaged (Kopjar et al.^[22], 2016).