Clinical ResearchFive-Year Follow-up of the Plaque Sealing With Paclitaxel-Eluting Stents vs Medical Therapy for the Treatment of

Intermediate Nonobstructive Saphenous Vein Graft Lesions(VELETI) Trial

Josep Rod!es-Cabau, MD, Olivier F. Bertrand, MD, PhD, Eric Larose, MD, Jean-Pierre D!ery, MD,St!ephane Rinfret, MD, Marina Urena, MD, Miguel Jerez, MD, Luis Nombela-Franco, MD,Henrique B. Ribeiro, MD, Ricardo Allende, MD, Guy Proulx, MD, Can M. Nguyen, MD,Jean-Rock Boudreault, MD, Jacques Rouleau, MD, Louis Roy, MD, Onil Gleeton, MD,G!erald Barbeau, MD, Bernard Noël, MD, M!elanie Côt!e, MSc, Jean-Pierre Despr!es, PhD,

Gilles R. Dagenais, MD, and Robert DeLarochellière, MDQu!ebec Heart and Lung Institute, Qu!ebec City, Qu!ebec, Canada

ABSTRACTBackground: Very few data exist on the long-term follow-up of patientswith intermediate nonobstructive saphenous vein graft (SVG) lesions.The purpose of this study was to evaluate the 5-year clinical outcomesof the patients enrolled in the Moderate Vein Graft Lesion StentingWith the Taxus Stent and Intravascular Ultrasound (VELETI) and thefactors associated with SVG disease progression and outcomes.Methods: Patients with ! 1 intermediate SVG lesion (30%-60% diam-eter stenosis) were randomized to either stenting the SVG lesion witha paclitaxel-eluting stent (PES group, n " 30) or to medical treatmentalone (MT group, n" 27). All patients were followed yearly up to 5 years.Results: Major adverse cardiac events (MACEs) (cardiac death,myocardial infarction [MI], revascularization) related to the target SVGlesion tended to be lower in the PES group (17% vs 33%; P " 0.146)due to a lower lesion revascularization rate (13% vs 33%; P " 0.072),with no difference in cardiac death or MI between groups. MACEs

R!ESUM!EIntroduction : Très peu de donn!ees sur le suivi à long terme despatients ayant des l!esions interm!ediaires non obstructives de la greffede veine saphène (GVS) existent. Le but de cette !etude !etait d’!evaluerles r!esultats cliniques après 5 ans des patients inscrits à l’essai VELETI(Moderate VEin graft LEsion Stenting With the Taxus Stent and Intra-vascular Ultrasound), et les facteurs associ!es à la progression et auxr!esultats des l!esions li!ees à la GVS.M!ethodes : Les patients ayant ! 1 l!esion interm!ediaire de la GVS(st!enose de 30 % à 60 % en diamètre) ont !et!e r!epartis au hasard soitpour l’implantation d’une endoprothèse à !elution de paclitaxel (n " 30,groupe EEP) ou pour le traitement m!edical seul (n " 27, groupe TM) dela l!esion de la GVS. Tous les patients ont !et!e suivis chaque ann!eedurant 5 ans.R!esultats : Les !ev!enements cardiaques ind!esirables majeurs (ECIM;mort cardiaque, infarctus du myocarde [IM], revascularisation) li!es à la

The appearance and progression of saphenous vein graft(SVG) atherosclerosis remains the first cause of graft failure,and about half of the SVGs are already occluded 10 years aftercoronary artery bypass grafting (CABG).1 Several studies haveidentified the presence of intermediate nonobstructive SVGlesions as the most important predictor of SVG atherosclerosisprogression and cardiac events.2-9 The VELETI (Moderate

Vein Graft Lesion Stenting With the Taxus Stent andIntravascular Ultrasound) trial was a randomized prospectivepilot trial assessing the effects of stenting intermediate non-obstructive SVG lesions on the prevention of atherosclerosisprogression as evaluated by angiography and intravascularultrasonography at 1-year follow-up.8 The study showed theefficacy and safety of sealing moderate nonobstructive SVGlesions with a drug-eluting stent (DES) to prevent SVGatherosclerosis progression. The VELETI trial also confirmedthe high rate of rapid progression and clinical events related tointermediate nonobstructive SVG lesions at 1-year follow-up.The purpose of our study was to evaluate the 5-year outcomesof the patients enrolled in the VELETI trial and to determinethe factors associated with cardiac events in these patients.

Received for publication October 9, 2013. Accepted November 5, 2013.

Corresponding author: Dr Josep Rod!es-Cabau, Qu!ebec Heart and LungInstitute, 2725, chemin Sainte-Foy, Qu!ebec G1V 4G5, Canada. Tel.: #1-418-656-8711; fax: #1-418-656-4544.

E-mail: josep.rodes@criucpq.ulaval.caSee page 144 for disclosure information.

0828-282X/$ - see front matter ! 2014 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.cjca.2013.11.002

Canadian Journal of Cardiology 30 (2014) 138e145

Methods

Study design and patients

Details of the VELETI study design have been providedpreviously.8 Briefly, patients who had undergone previousCABG involving SVG implantation and who required coronaryand SVG angiography by clinical indication were screened, andthose with at least 1 intermediate SVG lesion (30%-60%diameter stenosis) that was not the culprit lesion responsible forthe clinical syndrome of the patient were eligible. A total of 57patients were included in the study, and patients wererandomized to either stenting the moderate SVG lesion witha paclitaxel-eluting stent (PES) (Taxus Express2 stent, BostonScientific Corp, Natick, MA [30 patients]) or to medicaltreatment alone (27 patients).

Informed consent was obtained from each patient, and thestudy protocol conforms to the ethical guidelines of the 1975Declaration of Helsinki as reflected in a priori approval by theinstitution’s human research committee.

Clinical follow-up and outcomes

Patients had clinical visits at 1-, 6-, and 12-month follow-upand yearly thereafter up to 5 years. The follow-up was completein 100% of the patients. At each clinical visit, the occurrence ofclinical events was systematically evaluated; this included theoccurrence of coronary or SVG revascularization, myocardialinfarction (MI), and death. Any death was considered of cardiacorigin unless proved otherwise. Occurrence of stent thrombosiswas defined and classified according to the Academic ResearchConsortium criteria.10 In the case of repeated coronary/SVGangiography (with or without revascularization) during thefollow-up period, the angiogram was systematically reviewedand compared with the baseline angiogram. Quantitative coro-nary angiographic measurements of SVG lesions were obtainedin all cases, as previously described.8 All major adverse cardio-vascular events (MACEsdcardiac death, MI, revascularization)were classified as either related or unrelated to the target SVGlesion and to the target SVG independently by 2 investigators.In case of disagreement, a third investigator also evaluated thecase for final assignment. The occurrence of SVG obstruction

was considered to be related to the intermediate SVG lesionunless proved otherwise.

Statistical methods

Comparison of numerical variables was performed usingthe Student t test or Wilcoxon rank-sum test depending onvariable distribution. The c2 test (if > 5 events) or the Fisherexact test (if $ 5 events) was used to compare qualitativevariables. Differences in clinical events between groups wereevaluated using the Cox proportional hazard model. Univar-iate and multivariate Cox proportional hazard models wereused to determine both differences in clinical events betweengroups and the predictive factors of MACE at 5-year follow-up. The variables exhibiting a P value < 0.05 in the univar-iate analysis were included in the multivariate model. Thevariables included in the model for the predictors of MACErelated to the target SVG were total cholesterol level and thenumber of lesions per SVG. The variables included in themodel for the predictors of global MACE were female sex andtotal cholesterol level. Kaplan-Meier curves were used topresent the cumulative incidence of MACE over timeaccording to randomization group. The data were analyzedusing the statistical package SAS, version 9.1.3 (SAS Institute,Cary, NC).

ResultsBaseline clinical and angiographic characteristics of the

study population according to target SVG randomizationgroup (medical treatment vs PES) are shown in Table 1.

Cardiac events

At 5-year follow-up, a total of 26 patients (46%) had at least1 MACE, and 14 (25%) and 16 patients (28%) had at least 1MACE related to the target SVG lesion and target SVG,respectively. Global MACEs and MACEs related to the targetSVG lesion and target SVG classified according to randomi-zation group are shown in Table 2. There were no differencesbetween groups regarding the incidence of global MACEs(MT, 48%; PES, 43%; hazard ratio [HR], 1.17; 95% confi-dence interval [CI], 0.55-2.55; P " 0.675). Up to 50% of the

related to the target SVG and global MACEs were similar betweengroups (P > 0.20 for both). A higher cholesterol level at baseline wasthe only independent predictive factor of MACEs related to the targetSVG (P " 0.016).Conclusions: Over a 5-year period, one third of intermediate lesions inold SVGs progressed, leading to a cardiac event. Stenting these lesionswith PESs tended to improve clinical outcomes by reducing lesionprogression but not SVG failure. Higher cholesterol levels were asso-ciated with SVG disease progression and clinical events. This pilotstudy provides the basis for a larger trial to determine the efficacy ofintermediate SVG lesion plaque sealing.

l!esion cible de la GVS ont g!en!eralement !et!e plus faibles (17 % vs 33 %,P " 0,146) pour le groupe EEP en raison d’un taux plus faible (13 % vs33 %, P " 0,072) de revascularisation de la l!esion, et les groupes n’ontmontr!e aucune diff!erence entre eux de mort cardiaque ou d’IM. LesECIM li!es à la l!esion cible de la GVS et les ECIM globaux ont !et!esimilaires entre les groupes (P > 0,20 pour les deux). Un taux decholest!erol plus !elev!e au d!epart !etait le seul facteur pr!edictifind!ependant des ECIM li!es à la l!esion cible de la GVS (P " 0,016).Conclusions : Sur une p!eriode de 5 ans, un tiers des l!esions inter-m!ediaires des anciennes GVS ont men!e à un !ev!enement cardiaque.L’implantation d’une EEP pour traiter ces l!esions a g!en!eralementam!elior!e les r!esultats cliniques en diminuant la progression de lal!esion, mais non l’!echec de la GVS. Les taux !elev!es de cholest!erol ont!et!e associ!es à la progression de la l!esion de la GVS et aux !ev!enementscliniques. Cette !etude pilote sert de fondement à la r!ealisation d’unessai de plus grande envergure pour d!eterminer l’efficacit!e du scelle-ment de la plaque de la l!esion interm!ediaire de la GVS.

Rod!es-Cabau et al. 139Plaque Sealing and SVGs

MACEs were related to the target SVG lesion. There wasa tendency toward a higher rate of MACEs (33% vs 17%; HR,2.25; 95% CI, 0.75-6.73; P " 0.146) and repeated revascu-larization (33% vs 13%; HR, 2.95; 95% CI, 0.91-9.60; P "0.072) related to the target SVG lesion in the MT groupcompared with the PES group. There were no differences in therate of MI or cardiac death related to the target SVG lesionbetween groups (P " 0.960 and P " 1.00 for MI and cardiacdeath, respectively). There were no differences between groupsregarding the rate of MACEs related to the target SVG (MT,33%; PES, 23%; HR, 1.57; 95% CI, 0.58-4.21; P " 0.374).

The main clinical and angiographic features of the patientswho had MACEs related to the target SVG during the follow-up period are shown in Supplemental Table S1. In the MT

group, all MACEs related to the target SVG were secondary tothe progression of the intermediate SVG lesion leading toa severe lesion or complete obstruction of the SVG. In thesepatients, the mean percent diameter stenosis of the interme-diate SVG lesion increased from 43% % 8% at baseline to75% % 14% at follow-up. SVG lesion progression led toacute coronary syndrome in 7 patients (77%; MI in 33%). Inthe PES group, MACEs related to the target SVG were relatedto late (> 1 year) restenosis of the PES in 4 patients (57%)and to progression of the disease in other segments of the SVGin 3 patients (43%). The mean percent diameter stenosis inpatients with restenosis was 76% % 16%, and 77% % 14% inpatients with SVG disease progression in nonstentedsegments. In the PES group, stent restenosis or diseaseprogression led to acute coronary syndrome in all cases (MI in29%). The cumulative incidence of MACEs related to thetarget SVG lesion, target SVG, and global MACEs over timeaccording to randomization group are shown in Figures 1, 2,and 3, respectively.

A total of 7 patients died during the follow-up period, 6 ofthem of noncardiovascular causes (2 cancers, 2 bleedingcomplications, 1 case of pneumonia, and 1 car accident). Theonly cardiac death occurred in a patient of the PES group.This patient had an episode of unstable angina caused by in-stent restenosis 59 months after initial PES implantation. Thepatient had successful revascularization with implantation ofan everolimus-eluting stent but died suddenly 3 weeks afterthis intervention. This event was considered probable stentthrombosis of the everolimus-eluting stent.

Predictors of MACEs at 5-year follow-up

Clinical and angiographic characteristics of the studypopulation grouped according to the occurrence of MACEsrelated to the target SVG are shown in SupplementalTable S2. In the multivariate analysis, a higher cholesterolvalue at baseline was the only factor determining a higher riskfor target SVG disease progression leading to clinical events atfollow-up (HR of 1.20 for each increase of 10 mg/dL; 95%CI, 1.04-1.40; P " 0.016). The independent predictors ofglobal MACEs (related or unrelated to the target SVG) werefemale sex (HR, 4.51; 95% CI, 1.61-11.03; P " 0.002) andhigher cholesterol levels at baseline (HR, 1.24 for eachincrease of 10 mg/dL; 95% CI, 1.11-1.38; P < 0.001).

DiscussionSeveral studies have demonstrated the rapid progression of

intermediate nonobstructive lesions in SVGs.2-9 The VELETItrial, with systematic imaging control of the SVG at 1-yearfollow-up, showed that atherosclerotic disease progressedvery rapidly in old and moderately diseased SVGs despitelipid-lowering therapy. Such an accelerated atheroscleroticprocess was mainly related to SVG sites with higher plaqueburden leading to moderate SVG stenosis, and in up to onefifth of these patients, SVG disease progression led to a severelesion or SVG occlusion within 12 months.8 The presentstudy provides the longest follow-up data to date for thosepatients with intermediate nonobstructive SVG lesions.Significant SVG disease progression leading to severe SVGstenosis or SVG occlusion was observed in one third of thepatients at 5-year follow-up. Recently, Abdel-Karim et al.

Table 1. Baseline characteristics of the study population, according torandomization group*

Medicaltreatment(n " 27)

Paclitaxel-eluting stenting

(n " 30) P value

Baseline clinical characteristicsAge (y) 69 % 7 69 % 9 0.726Male sex 26 (96) 25 (83) 0.197Current smoker 4 (15) 1 (3) 0.180Hypertension 15 (56) 16 (53) 1.00Dyslipidemia 24 (89) 29 (97) 0.336Diabetes mellitus 10 (37) 10 (33) 0.788Previous MI 15 (56) 20 (67) 0.425Ejection fraction (%) 59 % 13 54 % 12 0.084Clinical indication for

angiographyStable angina pectoris 7 (25) 5 (17) 0.519Acute coronary syndrome 17 (63) 24 (80) 0.238Unstable angina 13 (76) 17 (71)Non-STEMI 3 (18) 5 (21)STEMI 1 (6) 0 (0)

Other 3 (11) 1 (3) 0.336Time from CABG (y) 11 % 6 11 % 6 0.540Number of grafts per patient 3.8 % 1.0 3.6 % 1.4 0.579Number of SVGs per patient 2.7 % 1.0 2.4 % 1.3 0.474Internal mammary artery to LAD 23 (85) 24 (80) 0.734PCI of a culprit lesion 12 (44) 15 (50) 0.464Baseline biological markers

Total cholesterol (mg/dL) 143 % 31 142 % 39 0.977LDL cholesterol (mg/dL) 74 % 28 73 % 34 0.911HDL cholesterol (mg/dL) 42 % 11 45 % 10 0.299Triglyceride levels (mg/dL) 136 % 53 124 % 50 0.380C-reactive protein (mg/L) 6.4 % 17.1 4.0 % 5.8 0.497

TreatmentAspirin 27 (100) 30 (100) 1.00Clopidogrel 19 (70) 26 (87) 0.200Statins 24 (89) 27 (90) 1.00

Baseline angiographiccharacteristics

Number of moderateSVG lesions

30 40 0.310

Number of lesionsper patient

1.1 % 0.3 1.3 % 0.6 0.086

Reference diameter (mm) 3.65 % 0.91 3.53 % 0.51 0.555Minimal lumen

diameter (mm)2.20 % 0.57 2.15 % 0.37 0.700

Percent diameter stenosis 40 % 7 39 % 6 0.826Lesion length (mm) 9.24 % 3.29 9.95 % 4.23 0.544

CABG, coronary artery bypass grafting; HDL, high-density lipoprotein;LAD, left anterior descending artery; LDL, low-density lipoprotein; MI,myocardial infarction; PCI, percutaneous coronary intervention; STEMI, ST-elevation MI; SVG, saphenous vein graft.

* Values are reported as n (%) or mean % standard deviation.

140 Canadian Journal of CardiologyVolume 30 2014

reported a rate of intermediate SVG lesion progression as highas 84% at 3-year follow-up in a substudy of the prospectiverandomized Stenting of Saphenous Vein Grafts (SOS) trial.9

Unlike the VELETI trial, most intermediate lesions in theSOS trial were located in SVGs that had already undergonePCI because of the presence of severe lesions, and a moreaccelerated atherosclerosis progression process may occur inthese cases.

Importantly, progression of intermediate SVG lesionspresented very frequently as acute coronary syndrome. In ourstudy, all but 2 patients (82%) presented with either unstableangina or MI. Clinical events resulting from intermediateSVG lesion progression were more frequent in the first 2 yearsafter the first cardiac event (not related to the intermediatelesion); the progression rate then decreased up to 5-yearfollow-up. In a previous study, we found that one third ofthe patients with significant progression of an intermediateSVG lesion presented with MI.4 Also consistent with theseresults, Abdil-Karim et al. showed that progression of theintermediate SVG lesion led to acute coronary syndrome inmore than two thirds of patients (MI in about half of them).9

This highlights the importance of implementing preventive

strategies for reducing the progression rate of intermediateSVG lesions.

The use of a DES for SVG PCI has been shown superiorto bare-metal stents for the treatment of severe SVG lesions,with a much lower rate of repeated revascularization and noincrease in the rate of stent thrombosis.11-13 Furthermore, theuse of a first-generation DES such as a PES was associatedwith lower rates of coronary revascularization and MI at 3-year follow-up in the randomized SOS trial.14 However,longer term data on the use of DESs in SVGs are scarce, andno data exist on late outcomes after PCI for the treatment ofintermediate SVG lesions. The results of the present studyshowed that stenting intermediate lesions with a PES wasassociated with a relatively low event rate, with a clinicalrestenosis rate lower than 15% and no cases of PES throm-bosis or SVG occlusion at 5-year follow-up. This representeda reduction of > 50% in lesion revascularization comparedwith the MT group. Interestingly, no cases of stent restenosiswere observed in the first year after PES PCI, and all caseswere late at > 2 years of follow-up. This late restenosisphenomenon has also been observed in other studies usingDESs for the treatment of SVG lesions. In the Reduction of

Table 2. Cumulative rates of major adverse cardiac events according to randomization group

Medical treatment Paclitaxel-eluting stent HR (95% CI) P value

n 27 30Patients with MACE related to the target SVG lesion

In hospital 0 1 (3) e 1.00Cardiac death 0 0 e eMI 0 1 (3) e 1.00Revascularization 0 0 e e

Follow-up 9 (33) 4 (13) 2.84 (0.88-9.23) 0.082Cardiac death 0 1 (3) e 1.00MI 3 (11) 2 (7) 1.54 (0.26-9.26) 0.635Revascularization 9 (33) 4 (13) 2.95 (0.91-9.60) 0.072

Cumulative 9 (33) 5 (17) 2.25 (0.75-6.73) 0.146Cardiac death 0 1 (3) e 1.00MI 3 (11) 3 (10) 1.04 (0.21-5.17) 0.960Revascularization 9 (33) 4 (13) 2.95 (0.91-9.60) 0.072

Patients with MACE related to the target SVGIn hospital 0 1 (3) e 1.00

Cardiac death 0 0 e eMI 0 1 (3) e 1.00Revascularization 0 0 e e

Follow-up 9 (33) 6 (20) 1.84 (0.65-5.17) 0.248Cardiac death 0 1 (3) e 1.00MI 3 (11) 2 (7) 1.54 (0.26-9.26) 0.635Revascularization 9 (33) 7 (23) 1.63 (0.61-4.38) 0.333

Cumulative 9 (33) 7 (23) 1.57 (0.58-4.21) 0.374Cardiac death 0 1 (3) e 1.00MI 3 (11) 3 (10) 1.04 (0.21-5.17) 0.960Revascularization 9 (33) 7 (23) 1.63 (0.61-4.38) 0.333

Patients with MACE related or unrelated to the target SVGIn hospital 0 1 (3) e 1.00

Cardiac death 0 0 e eMI 0 1 (3) e 1.00Revascularization 0 1 (3) e 1.00

Follow-up 13 (48) 12 (40) 1.27 (0.58-2.78) 0.555Cardiac death 0 1 (3) e 1.00MI 3 (11) 5 (17) 0.68 (0.16-2.85) 0.592Revascularization 13 (48) 12 (40) 1.27 (0.58-2.79) 0.545

Cumulative 13 (48) 13 (43) 1.17 (0.55-2.55) 0.675Cardiac death 0 1 (3) e 1.00MI 3 (11) 6 (20) 0.56 (0.14-2.26) 0.415Revascularization 13 (48) 13 (43) 0.18 (0.55-2.55) 0.675

Values are reported as n (%).HR, hazard ratio; MACE, major adverse cardiac event; MI, myocardial infarction; SVG, saphenous vein graft.

Rod!es-Cabau et al. 141Plaque Sealing and SVGs

Restenosis in Saphenous vein Grafts with Cypher Stent(RRISC) trial, most cases of restenosis at 3-year follow-upoccurred > 1 year after sirolimus-eluting stent implanta-tion,15 and in the SOS trial, half of the cases of stent reste-nosis that occurred after a mean follow-up of 3 years afterPES implantation happened after the first year.14 This laterestenosis phenomenon contributed to reducing the efficacyof the plaque sealing strategy, and it was probably secondaryto delayed healing, late neoatherosclerosis associated with theDES, or both.16 About half of the events related to SVGfailure in the PES group resulted from SVG diseaseprogression in nonstented SVG segments. Although the SVGfailure rate was still 30% lower than in the MT group, thispilot study had a limited sample size and was not powered forclinical end points, and between-group differences in clinicalevents secondary to SVG failure were nonsignificant. Theimportance of SVG disease progression in nonstentedsegments for late clinical events after SVG PCI has been wellestablished in previous studies.14,17 SVG intervention trialsusing DESs have shown that 30%-50% of the clinical eventsoccurring within the first year after SVG DES implantationare related to nonstented SVG segments, and this rateincreases to > 50% of the events at 3-year follow-up.11-15

This rapid progression of SVG disease in nonstentedsegments highlights the importance of additional (medical)measures to prevent SVG disease progression in suchpatients.

Despite the relatively high MI rate (11%) related to SVGlesion progression in the MT group, no differences in the rateof MI were observed between the PES and MT groups.Although no cases of stent thrombosis related to the PESswere observed, 1 patient had a periprocedural MI secondary toa transient no-reflow phenomenon after PCI of the interme-diate SVG lesion, and half of the patients with PES restenosispresented with a non-ST elevation, leading to an MI rate of10% at 5-year follow-up.

The Post Coronary Artery Bypass Graft Trial showed thatintensive lipid-lowering therapy was associated with a signifi-cant reduction in SVG disease progression,18 and this translatedinto a lower rate of cardiac events at long-term follow-up.5 Themean cholesterol and low-density lipoprotein (LDL) levels inthe Post Coronary Artery Bypass Graft Trial were w 170mg/dL and 93 mg/dL, respectively. The vast majority ofpatients included in the VELETI trial received lipid-loweringtherapy, and mean cholesterol and LDL levels were muchlower (mean, 143 mg/dL and 74 mg/dL, respectively) than

Figure 1. Kaplan-Meier cumulative event curves for major adverse cardiac events (MACEs) related to the target saphenous vein graft lesionthroughout 5-year follow-up according to randomization group. (A) Death, (B) myocardial infarction (MI), (C) revascularization, (D) MACEs, cardiacdeath, MI, and revascularization.

142 Canadian Journal of CardiologyVolume 30 2014

those reported in the Post Coronary Artery Bypass Graft Trial.However, higher cholesterol levels were associated with a higherrate of MACEs globally and particularly with those casesassociated with SVG disease progression, suggesting that furtherlowering of cholesterol levels might be associated with a reduc-tion in cardiac events in such patients.

Study limitations

This was a single-centre pilot study. The number ofpatients included was small and the study was not powered forclinical end points. A larger randomized study is needed todetermine the efficacy of stenting intermediate SVG lesions.This is the purpose of the ongoing VELETI II trial(NCT01223443), a prospective multicentre randomized trialto evaluate the efficacy of plaque sealing with DESs in a largerstudy population. Although most patients included in thestudy had repeated coronary angiography at 1-year follow-up,no further systematic imaging studies were performed up to 5-year follow-up. Some cases of late silent SVG obstruction mayhave been missed, and the rate of SVG failure may have beenunderestimated.

ConclusionsIntermediate nonobstructive lesions in old (mean of 12

years) SVGs were associated with a high rate of rapidprogression and clinical events at 5-year follow-up. Althoughstenting these SVG lesions with PESs was safe and associatedwith a tendency toward a lower incidence of events relatedto lesion progression, it failed to significantly reduce SVGfailure because of both late stent restenosis phenomena andSVG disease progression in nonstented segments. Larger andadequately powered studies will be needed to definitely addressthe efficacy of stenting intermediate SVG lesions. Finally,despite low mean total and LDL cholesterol levels, a highercholesterol level remained the main factor determining SVGdisease progression and clinical events at follow-up, stronglysuggesting that further reductions in cholesterol levels willprobably be needed to optimize clinical outcomes in the high-risk population of post-CABG patients with old SVGs.

Funding SourcesThe VELETI trial (ClinicalTrials.gov number:

NCT00289835) was an investigator-initiated study supported

Figure 2. Kaplan-Meier cumulative event curves for major adverse cardiac events (MACEs) related to the target saphenous vein graft throughout 5-year follow-up according to the randomization group. (A) Death, (B) myocardial infarction (MI), (C) revascularization, (D) MACEs, cardiac death, MI,and revascularization.

Rod!es-Cabau et al. 143Plaque Sealing and SVGs

by an unrestricted grant from Boston Scientific Inc Canada(Toronto, Ontario, Canada) and from the Fondation del’Institut de Cardiologie de Qu!ebec, Qu!ebec, Canada.

DisclosuresThis trial was partially supported by a grant from Boston

Scientific Inc Canada.

References

1. Goldman S, Zadina K, Moritz T, et al. Long-term patency of saphenousvein and left internal mammary artery grafts after coronary artery bypasssurgery: results from a Department of Veterans Affairs Cooperative Study.J Am Coll Cardiol 2004;44:2149-56.

2. Domanski MJ, Borkowf CB, Campeau L, et al. Prognostic factors foratherosclerosis progression in saphenous vein grafts. J Am Coll Cardiol2000;36:1877-83.

3. Ellis SG, Brener SJ, DeLuca S, et al. Late myocardial ischemic events aftersaphenous vein graft intervention: importance of initially “nonsignificant”vein graft lesions. Am J Cardiol 1997;79:1460-4.

4. Rod!es-Cabau J, Facta A, Larose E, et al. Predictors of aorto-saphenousvein bypass narrowing late after coronary artery bypass grafting. Am JCardiol 2007;100:640-5.

5. Knatterud GL, White C, Geller NL, et al. Angiographic changes insaphenous vein grafts are predictors of clinical outcomes. Am Heart J2003;145:262-9.

6. Le May MR, Labinaz M, Marquis JF, et al. Predictors of long-termoutcome after stent implantation in a saphenous vein graft. Am J Car-diol 1999;83:681-6.

7. Lytle BW, Loop FD, Taylor PC, et al. Vein graft disease: the clinicalimpact of stenoses in saphenous vein grafts to coronary arteries. J ThoracCardiovasc Surg 1992;103:831-40.

8. Rod!es-Cabau J, Bertrand OF, Larose E, et al. Comparison of plaquesealing with paclitaxel-eluting stents versus medical therapy for thetreatment of moderate nonsignificant saphenous vein graft lesions: themoderate vein graft lesion stenting with the taxus stent and intravascularultrasound (VELETI) pilot trial. Circulation 2009;120:1978-86.

9. Abdel-Karim AR, Da Silva M, Lichtenwalter C, et al. Prevalence andoutcomes of intermediate saphenous vein graft lesions: findings from thestenting of saphenous vein grafts randomized-controlled trial. Int J Car-diol 2013;168:2468-73.

Figure 3. Kaplan-Meier cumulative event curves for death and global (related and unrelated to the target saphenous vein graft) major adversecardiac events (MACEs) up to 5-year follow-up, according to randomization group. (A) Death, (B) myocardial infarction (MI), (C) revascularization, (D)MACEs, cardiac death, MI, and revascularization.

144 Canadian Journal of CardiologyVolume 30 2014

10. Cutlip DE, Windecker S, Mehran R, et al. Clinical end points incoronary stent trials. A case for standardized definitions. Circulation2007;115:2344-51.

11. Mehilli J, Pache J, Abdel-Wahab M, et al. Drug-eluting versus bare-metalstents in saphenous vein graft lesions (ISAR-CABG): a randomisedcontrolled superiority trial. Lancet 2011;378:1071-8.

12. Brilakis ES, Lichtenwalter C, de Lemos JA, et al. A randomizedcontrolled trial of a paclitaxel-eluting stent versus a similar bare-metalstent in saphenous vein graft lesions the SOS (Stenting of SaphenousVein Grafts) trial. J Am Coll Cardiol 2009;53:919-28.

13. Vermeersch P, Agostoni P, Verheye S, et al. Randomized double-blindcomparison of sirolimus-eluting stent versus bare-metal stent implanta-tion in diseased saphenous vein grafts: six-month angiographic, intra-vascular ultrasound, and clinical follow-up of the RRISC Trial. J Am CollCardiol 2006;48:2423-31.

14. Brilakis ES, Lichtenwalter C, Abdel-karim AR, et al. Continued benefitfrom paclitaxel-eluting compared with bare-metal stent implantation insaphenous vein graft lesions during long-term follow-up of the SOS(Stenting of Saphenous Vein Grafts) trial. JACC Cardiovasc Interv2011;4:176-82.

15. Vermeersch P, Agostoni P, Verheye S, et al. Increased late mortality aftersirolimus-eluting stents versus bare-metal stents in diseased saphenous

vein grafts: results from the randomized DELAYED RRISC Trial. J AmColl Cardiol 2007;50:261-7.

16. Habara M, Terashima M, Nasu K, et al. Morphological differences oftissue characteristics between early, late, and very late restenosis lesionsafter first generation drug-eluting stent implantation: an optical coher-ence tomography study. Eur Heart J Cardiovasc Imaging 2013;14:276-84.

17. Keeley EC, Velez CA, O’Neill WW, Safian RD. Long-term clinicaloutcome and predictors of major adverse cardiac events after percuta-neous interventions on saphenous vein grafts. J Am Coll Cardiol2001;38:659-65.

18. Post Coronary Artery Bypass Graft Trial Investigators. The effect ofaggressive lowering of low-density lipoprotein cholesterol levels and low-dose anticoagulation on obstructive changes in saphenous vein coronary-artery bypass grafts. N Engl J Med 1997;336:153-62.

Supplementary MaterialTo access the supplementary material accompanying this

article, visit the online version of the Canadian Journal ofCardiology at www.onlinecjc.ca and at http://dx.doi.org/10.1016/j.cjca.2013.11.002.

Rod!es-Cabau et al. 145Plaque Sealing and SVGs