1

Scope of the problem

Saphenous vein grafts (SVGs) are routinely used in coronary artery bypass graft (CABG) surgery for non-left anterior descending artery targets. The long-term patency rates of SVGs are known to be poor with 30% occlusion in the first year and 50% at 10 years [ , ]. Thus, there has been a need for a good long term solution in patients presenting with SVG stenosis. To understand the outcomes of SVG intervention, one must be familiar with the pathogenesis of SVG disease. Early SVG failure may be related to thrombosis from platelet plug formation at the site of distal anastomosis secondary to endothelial damage from technical issues. Delayed SVG failure is related to atherosclerosis, but its progression varies from that of native coronaries. Histologically the lesion in SVG consists of large, soft, friable and thrombotic plaques [ ]. Thus, intervening on these lesions is associated with distal embolization and “no-reflow” which in turn is associated with poor short and long-term outcomes [ ].

2

Current trends in SVG intervention

The evolution of SVG intervention started similarly to native coronary disease with balloon angioplasty which subsequently was replaced by bare metal stent (BMS) as the latter was shown to be superior [ ]. Next came covered stents which eventually were found to be inferior to BMS [ ]. The search continued for a superior stent with the first and then second-generation drug-eluting stent (DES). In this meta-analysis, Nairooz et al. evaluated specifically the long-term outcomes between DES versus BMS in saphenous vein graft interventions utilizing data from the major randomized controlled trials [ ]. This encompassed 1592 subjects with a mean follow-up of 42.8 months. The authors found similar outcomes between DES and BMS in all the endpoints measured including all-cause mortality, MACE (major adverse cardiac events), cardiac death, myocardial infarction, target lesion revascularization (TLR), and target vessel revascularization (TVR). It comes as no surprise that there were no significant differences with regard to “hard” endpoints such as all-cause mortality or cardiac death since DES technologies have not been proven to improve these specific outcomes. However, DES have demonstrated lower TLR compared to BMS in native coronary artery and some studies also showed this benefit in SVG intervention at 12 months, making the equivalent outcomes of TLR and TVR unexpected [ , ]. The results from the present study suggest that even early “soft endpoint” benefits from DES are lost at long-term follow-up. This may be due to the different mechanisms of target vessel failure in SVG disease compared to native coronary disease.

The other notable finding from the study is the high overall event rates at long term follow-up with all-cause mortality of ~15% in both groups and MACE of >30%. This reflects not only current limitations in treating SVG disease but also limitations in treating progressive atherosclerosis as a systemic disease. The strengths of the current study are standardized meta-analysis methodology and inclusion of all major randomized trials with long term follow-up data. Major limitations to this study and inherent to meta-analyses are the characteristic differences between the individual studies. Except for most recently published DIVA study, all the other trials in this analysis utilized first generation DES which is no longer used in contemporary practice [ ]. This does not appear to affect the overall results as the DIVA study did not demonstrate improved outcomes despite predominantly second-generation DES usage. Follow-up protocols also varied according to the different trials. For example, in the two largest trials, the ISAR-CABG study mandated angiographic follow-up at 6 and 8 months while the DIVA study instituted clinical follow-up. This may explain the earlier advantage of TLR in ISAR-CABG with DES.

2

Current trends in SVG intervention

The evolution of SVG intervention started similarly to native coronary disease with balloon angioplasty which subsequently was replaced by bare metal stent (BMS) as the latter was shown to be superior [ ]. Next came covered stents which eventually were found to be inferior to BMS [ ]. The search continued for a superior stent with the first and then second-generation drug-eluting stent (DES). In this meta-analysis, Nairooz et al. evaluated specifically the long-term outcomes between DES versus BMS in saphenous vein graft interventions utilizing data from the major randomized controlled trials [ ]. This encompassed 1592 subjects with a mean follow-up of 42.8 months. The authors found similar outcomes between DES and BMS in all the endpoints measured including all-cause mortality, MACE (major adverse cardiac events), cardiac death, myocardial infarction, target lesion revascularization (TLR), and target vessel revascularization (TVR). It comes as no surprise that there were no significant differences with regard to “hard” endpoints such as all-cause mortality or cardiac death since DES technologies have not been proven to improve these specific outcomes. However, DES have demonstrated lower TLR compared to BMS in native coronary artery and some studies also showed this benefit in SVG intervention at 12 months, making the equivalent outcomes of TLR and TVR unexpected [ , ]. The results from the present study suggest that even early “soft endpoint” benefits from DES are lost at long-term follow-up. This may be due to the different mechanisms of target vessel failure in SVG disease compared to native coronary disease.

The other notable finding from the study is the high overall event rates at long term follow-up with all-cause mortality of ~15% in both groups and MACE of >30%. This reflects not only current limitations in treating SVG disease but also limitations in treating progressive atherosclerosis as a systemic disease. The strengths of the current study are standardized meta-analysis methodology and inclusion of all major randomized trials with long term follow-up data. Major limitations to this study and inherent to meta-analyses are the characteristic differences between the individual studies. Except for most recently published DIVA study, all the other trials in this analysis utilized first generation DES which is no longer used in contemporary practice [ ]. This does not appear to affect the overall results as the DIVA study did not demonstrate improved outcomes despite predominantly second-generation DES usage. Follow-up protocols also varied according to the different trials. For example, in the two largest trials, the ISAR-CABG study mandated angiographic follow-up at 6 and 8 months while the DIVA study instituted clinical follow-up. This may explain the earlier advantage of TLR in ISAR-CABG with DES.