1. DRUG ELUTING STENTS IN DISEASED SAPHENOUS VEIN GRAFTS: Reappraisal of the evidence Frank Van den Branden, MD Antwerp Cardiovascular Institute Middelheim Antwerp, Belgium

2. Incidence of SVG Disease I.Autologous saphenous vein graft remains the most frequently used conduit during CABG. II.The long term patency of vein grafts is limited. III.Neointimal hyperplasia leading to accelerated atherosclerosis and thrombosis is the proposed mechanism of vein graft failure. Campeau L. et al. Circulation 1979

3. Saphenous Vein Graft Disease I.Morphology : A.Diffuse, large, soft & friable atheroma B.Poorly developed or absent fibrous cap C.Little or no calcification D.Thrombus in 70% of old SVG II.Percutaneous coronary intervention is considered the preferred modality because of high morbidity and mortality of redo-CABG

4. Percutaneous coronary intervention as a treatment of diseased vein grafts (6- 15% of PCI volume) is associated with: a poor immediate outcome: 1. distal embolization a poor long term outcome: 1. Restenosis (35-60 %) 1,2 2.Progression of disease of the non-treated segments (native and vein graft) 3 DES !? Protection devices SVG intervention 1 Savage et al, NEJM 19972 Choussat et al, J Am Coll Cardiol 20003 Ellis et al, Am J Cardiol 1997

5. BMS vs. PTCA in SVG Vermeersch, Agostoni. JIntervC 2005

6. Drug Eluting Stents I.In all randomized trials SVG lesions were excluded. II.Mechanisms of in-stent restenosis are different in SVGs compared to native arteries. III.Problem of higher local prothrombotic conditions in the vein graft and the expected delay in endothelial healing after DES are claimed as possible drawbacks, as they can lead to a higher risk of acute, subacute and late thrombosis.

7. Currently Available Data I.Several single arm registries: A.Relatively small numbers B.Encouraging C.No safety issues II.3 published studies with historical BMS-control: A.No safety issues B.Conflicting data

8. 1.09 0.37 The Milan Experience Treatment of saphenous vein graft lesions with drug-eluting stents Ge, et al. JACC 2005

9. The Los Angeles Experience I.Since april 2003 : 223 consecutive patients II.139 DES, 84 BMS III.Single center, non- randomized IV.Baseline characteristics and angiographical data were comparable Drug eluting stenting is superior to bare metal stenting in saphenous vein grafts. Lee, et al. CCI 2005

10. The Washington Experience Efficacy of sirolimus- eluting stents compared with bare metal stents for saphenous vein graft intervention Chu, et al. AJC 2006

11. RRISC Trial Reduction of Restenosis In Saphenous vein grafts with Cypher stent Prospective, randomized, double-blind, non industry sponsored, single center, trial comparing SES vs. BMS in SVG lesions 75 patients with 96 lesions localized in 80 diseased SVG were included Enrollment: September 2003-November 2004 Primary endpoint : 6-month in-stent late loss Secondary endpoints (all at 6 months follow up): – Binary angiographic restenosis (in-stent/in-segment) – Clinical events (death, MI, TLR, TVR) Vermeersch, Agostoni et al. JACC 2006

12. Major Inclusion Criteria De novo lesion (stenosis >50%) in a diseased SVG Diameter ranging between 2.5 and 4.0 mm Diagnosis of angina pectoris Osial stenoses & thrombotic/calcific stenoses were allowed No maximum lesion length prespecified Major Exclusion Criteria Impaired renal function Prior stent within 5 mm of target lesion Totally occluded vein grafts Documented LV Ejection Fraction <25% Distal anastomotic stenosis Prior brachytherapy in the index vessel Recent MI (<7 days) RRISC Trial Reduction of Restenosis In Saphenous vein grafts with Cypher stent Vermeersch, Agostoni et al. JACC 2006

13. 75 patients (with 96 lesions) meeting the inclusion criteria 37 patients (49 lesions) randomized to BMS 38 patients (47 lesions) randomized to SES 37 patients (49 lesions) available for 6-month angiographic follow up 35 patients (44 lesions) available for 6-month angiographic follow up 1 patient died 2 patients refused angio follow up randomization No patient was lost to follow up. All patients, but 1 (dead) available for 6-month clinical follow up. 204 patients screened (September 2003-November 2004) Patients excluded (reason): 2 patients (age >85 years) 18 patients (acute MI) 7 patients (MI within the last 7 days) 3 patients (creatinine >3 mg/dL) 40 patients (vein graft with RVD >4.0 mm) 12 patients (distal anastomotic disease) 38 patients (restenotic lesions) 8 patients (enrolled in other trials) 1 patient (no informed consent ) Vermeersch, Agostoni et al. JACC 2006

14. BMS (n=37) SES (n=38) P- value Age (years)72 ± 873 ± 70.36 Men33 (89%)31 (82%)0.36 Family history29 (78%)25 (66%)0.23 Hypertension21 (57%)22 (58%)0.84 Hypercholesterolemia31 (84%)33 (87%)0.74 Current smoker4 (11%)2 (5%)0.46 Diabetes Mellitus5 (14%)6 (16%)0.78 Body mass index (Kg/m 2 )26.4 ± 3.926.4 ± 3.10.97 History of heart failure7 (19%)6 (16%)0.72 Prior myocardial infarction15 (41%)17 (45%)0.71 Prior coronary angioplasty15 (41%)12 (32%)0.42 Unstable angina pectoris19 (51%)23 (60%)0.41 Ejection Fraction (%)72 ± 1268 ± 180.37 Age of the grafts (years)12.6 ± 5.912.4 ± 4.60.92 Baseline characteristics Vermeersch, Agostoni et al. JACC 2006

15. BMS (lesions=49) SES (lesions=47) P- value Degenerated saphenous vein grafts17 (41.5%)19 (48.7%)0.51 Recipient native vessel territory 0.11 Left anterior descending/diagonal6 (12.2%)9 (19.2%) Circumflex/obtuse marginal26 (53.1%)15 (31.9%) Right coronary artery17 (34.7%)23 (48.9%) Angiographic evidence/suspect of thrombus12 (24.5%)17 (36.2%)0.21 Moderately/heavily calcified lesions9 (18.4%)8 (17%)0.86 Number of stents per patient1.46 ± 0.71.58 ± 0.70.45 Number of stents per lesion1.11 ± 0.31.28 ± 0.50.14 Total stent length per patient (mm)33.4 ± 18.236.9 ± 17.60.39 Total stent length per lesion (mm)25.2 ± 11.929.9 ± 15.60.11 Stent diameter (mm)3.36 ± 0.263.41 ± 0.190.72 Successful direct stenting44 (89.8%)44 (93.6%)0.50 Post-dilatation7 (14.3%)14 (29.8%)0.09 Maximal balloon diameter (mm)3.44 ± 0.383.56 ± 0.370.09 Maximal inflation pressure (atm)18.8 ± 2.218.7 ± 2.80.85 Vermeersch, Agostoni et al. JACC 2006

16. 0.20 0.40 0.60 0.80 1.00 0.17 0.790.380.240.19 p=0.001 p=0.9 p=0.6 0.70 0.41 p=0.01 Prox edgeDist edgeIn-stentIn-segment Late Loss Analysis Vermeersch, Agostoni et al. JACC 2006 BMS SES

17. In-segment 10 20 30 40 11.4%30.6% 13.6%32.7% p=0.024 p=0.031 In-stent Δ=19.1% RRR=0.58 Δ=19.2% RRR=0.63 BMS SES Binary Restenosis Vermeersch, Agostoni et al. JACC 2006

18. Number of lesions 20 15 10 5 0 Neointimal Volume (mm3) 1209060300 1209060300 Median: 24 mm 3 [8-34] Median: 1 mm 3 [0-13] P<0.001 39 BMS lesions 34 SES lesions BMS SES IVUS analysis BMS (lesions=39) SES (lesions=34) Stent length (mm)21.2 [17.1-30.9]23.4 [18.8-31.6]0.14 Stent volume (mm 3 )211 [143-282]214 [174-325]0.23 Lumen volume (mm 3 )175 [125-243]205 [174-310]0.023 Neointimal volume (mm 3 )24 [8-34]1 [0-13]<0.001 Agostoni et al. AJC 2007

19. 6-month MACE BMS 37 n=37 SES 38 n=38 P value In-hospital Death00 Repeat revascularization00 Periprocedural MI1 (2.7%)2 (5.3%)0.99 Between discharge and 6 months Death01 (2.6%)0.99 Myocardial infarction01 (2.6%)0.99 TLR TLR (per-patient)8 (21.6%)2 (5.3%)0.047 TVR TVR (per-patient)10 (27%)2 (5.3%)0.012 Cumulative 6-month MACE11 (29.7%)6 (15.8%)0.15 - Due to safety issues recently raised with DES (ESC/WCC 2006), we decided to further follow up our patients, in order to analyze long-term events. -In September 2006, a new approval was obtained from the local Ethics Committee to extend the follow-up. -A new informed consent was obtained from all the patients. -All patients were contacted between September and December 2006 (no lost to follow up). -Blinding was maintained for patients and referring physicians/cardiologists. Vermeersch, Agostoni et al. JACC 2006

20. BMS 37 n=37 SES 38 n=38 P value Death010 (26.3%)0.001 Myocardial infarction1 (2.7%)4 (10.5%)0.35 TLR3 (8.1%)7 (18.4%)0.30 TVR4 (10.8%)11 (28.9%)0.05 MACE after 6-month up to 32 months (median f.u.) Vermeersch, Agostoni et al. JACC 2007

21. Kaplan-Meyer Curves Vermeersch, Agostoni et al. JACC 2007

22. BMS 37 n=37 SES 38 n=38 P value Death011 (28.9%)<0.001 Myocardial infarction2 (5.4%)7 (18.4%)0.15 TLR11 (29.7%)9 (23.7%)0.55 TVR14 (37.8%)13 (34.2%)0.74 MACE15 (40.5%)22 (57.9%)0.13 Other PCI (not TLR/TVR)14 (37.8%)12 (31.6%)0.57 Double anti-platelet therapy19 (51.4%)19 (50%)0.91 Single anti-platelet therapy14 (37.8%)15 (39.5%)0.88 No anti-platelet therapy4 (10.8%)4 (10.5%)0.97 Statin therapy27 (73%)29 (76.5%)0.74 Cumulative MACE Vermeersch, Agostoni et al. JACC 2007

23. Stent Thrombosis (ARC criteria) BMS 37 n=37 SES 38 n=38 P value Definite02 (5.2%) 1 fatal at 13 months 1 non fatal at 30 months 0.49 Probable00- Possible03 (7.9%) 1 sudden death at 7.5 months 1 sudden death at 11.5 months 1 sudden death at 35 months 0.30 Total05 (13.1%) 0.054 Fisher Exact 0.022 Log Rank Vermeersch, Agostoni et al. JACC 2007

24. TimeCause of deathAnti-thrombotic therapy 5 monthsprogressive heart failureTP 7.5 monthssudden out-of-hospital deathTP, W 11.5 monthssudden out-of-hospital deathASA, TP 14.5 monthsprogressive heart failure after MI due to thrombosis of the index stent - (suspended 1 week before MI for knee surgery) 16 monthsmetastatic urothelial carcinoma- (stop 1 month before for severe anemia) 19 monthsmetastatic colon carcinoma- (stop 2 months before for anorexia) 22 monthsprogressive MOF after peri-operative (limb ischemia) MI (no stent thrombosis of the stent) - (suspended 1 week before MI for AICD change) 23.5 monthspost-operative (AVR and ReDo CABG for progression of CAD) infection ASA, TP 30 monthsprogressive Parkinson diseaseASA, TP 33 monthsprogressive MOF after ReDo CABG (documented in-stent restenosis) ASA, TP 35 monthssudden out-of-hospital deathASA, TP Causes of Death Vermeersch, Agostoni et al. JACC 2007

25. Conclusions The use of BMS was associated with lower long-term mortality than the use of SES for SVG disease. Also the 6-month reduction in repeated revascularization procedures shown with the use of SES was lost at longer- term follow-up. However: this is secondary post-hoc analysis, the play of chance should be strongly considered, “hidden” factors unrelated to stent type could have influenced the final results. Further studies are required before conclusions can be made about the safety or harm of using SES for SVG lesions.

26. For further slides on these topics please feel free to visit the metcardio.org website: http://www.metcardio.org/slides.html http://www.metcardio.org/slides.html