The SOS (Stenting Of Saphenous vein grafts) Randomized, Controlled Trial of a Paclitaxel-Eluting Stent Vs. a Similar Bare Metal Stent in Saphenous Vein Graft Lesions Emmanouil S. Brilakis, MD, PhD Director, Cardiac Catheterization Laboratory VA North Texas Healthcare System On behalf of the SOS investigators

I, Emmanouil Brilakis DO NOT have a financial interest/arrangement or affiliation with one or more organizations that could be perceived as a real or apparent conflict of interest in the context of the subject of this presentation. Disclosure Statement of Financial Interest

Veteran Affairs, VISN-17 Clark R. Gregg fund, Harris Methodist Foundation, Fort Worth, Texas Funding

SOS design DESIGN: Prospective, randomized, multi-center trial comparing the Taxus ™ paclitaxel-eluting stent (PES) with a similar Express 2™ bare metal stent (BMS) in saphenous vein graft (SVG) lesions OBJECTIVE: To compare the 12-month angiographic and 24-month clinical outcomes between PES and BMS in SVG lesions PRINCIPAL INVESTIGATOR: Emmanouil S. Brilakis, MD, PhD. VA North Texas Healthcare System, Dallas, Texas

SOS: sample size determination 31 patients needed per group to have 80% power (2-sided alpha 0.05) to detect a 66% reduction in binary angiographic restenosis assuming 50% restenosis in BMS group and 1 lesion treated per pt Target enrollment: 40 patients per group to account for losses during follow-up

SOS: Patient flow 80 patients (112 lesions) enrolled between 2005 and 2007 in 5 clinical sites in USA and Europe Died: 1 pt Declined: 5 pts Angiographic follow-up at 12 months 33 pts 47 lesions Clinical follow-up at 24 months Angiographic follow-up at 12 months 33 pts 43 lesions BMS 39 pts 55 lesions PES 41 pts 57 lesions Died: 4 pts Emergent CABG: 1 pt Declined: 3 pts Died: 1 pt median follow-up: 18 months

Iowa City VAMC James Rossen, MD Michael E. DeBakey VAMC Biswajit Kar, MD VA North Texas HCS Coordinating Ctr Emmanouil Brilakis, MD, PhD Subhash Banerjee, MD Christopher Lichtenwalter, MD James de Lemos, MD Owen Obel, MD Michele Roesle, RN Little Rock VAMC Joe K. Bissett, MD Rajesh Sachdeva, MD SOS trial centers Collaborators: Peter Berger, MD Panayotis Fasseas, MD Onassis Cardiac Surgery Ctr Athens, Greece Vassilios Voudris, MD Panagiotis Karyofillis, MD

BMS n = 39 PES n = 41 P Value Age (yrs) 67 ± 966 ± Men (%) White (%) Diabetes (%) Current smoker (%) Hypertension (%) Hyperlipidemia (%) Baseline characteristics I

BMS n = 39 PES n = 41 P Value Years since CABG 12 ± 611 ± Presentation Stable angina (%) ACS (%) 5763 BMI (kg/m 2 ) 29 ± 430 ± Prior MI (%) Normal ejection fraction (%) Baseline characteristics II

BMS n = 39 PES n = 41 P Value SVG recipient vessel LAD/diagonal (%) Circumflex/OM (%) RCA/PDA (%) 4029 Lesion location Aortic anastomosis (%) Body (%) Distal anastomosis (%) 77 Procedural description I

BMS n = 39 PES n = 41 P Valu e Number of SVGs treated 1.15 ± ± Number of lesions treated 1.41 ± ± Number of stents/lesion 1.13 ± ± Predilatation (%) Embolic protection (%) Max balloon diameter, mm 3.20 ± ± Max inflation pressure, atm 17 ± Stent length, mm 18 ± Procedural description II

BMS n = 39 PES n = 41 P Value Heparin (%) GP IIb/IIIa inhibitors (%) Contrast, mL 257 ± ± Fluoroscopy, min 20 ± 921 ± Post PCI MI (%) Procedural success (%) Procedural description III

Late Loss mm ± 1.03 BMS (47 lesions), PES (43 lesions) Diff (95% CI) (-0.51, -1.22) P< Diff (95% CI) (-0.48, -1.46) P< ± 0.57 ± 0.98 ± 0.54

Binary angiographic restenosis % Primary study endpoint relative risk: % CI: 0.07, 0.48 p <

Minimum lumen diameter (mm) % BMS PES Before intervention Cumulative frequency distribution curves

Minimum lumen diameter (mm) % BMS PES Before intervention After intervention (in-stent minimum lumen diameter) Cumulative frequency distribution curves

BMS PES % After intervention Follow-up BMS In-stent minimum lumen diameter (mm) Follow-up PES Cumulative frequency distribution curves

Clinical outcomes median follow-up: 1.5 years

BMS n = 39 PES n = 41 P Value Death (%) Clinical outcomes median follow-up: 1.5 years

BMS n = 39 PES n = 41 P Value Death (%) Myocardial infarction (%) Clinical outcomes median follow-up: 1.5 years

BMS n = 39 PES n = 41 P Value Death (%) Myocardial infarction (%) TLR (%) TVR (%) Any revascularization (%) Clinical outcomes median follow-up: 1.5 years

BMS n = 39 PES n = 41 P Value Death (%) Myocardial infarction (%) TLR (%) TVR (%) Any revascularization (%) Target vessel failure (%) Overall MACE (%) Clinical outcomes median follow-up: 1.5 years

BMS n = 39 PES n = 41 P Value Death (%) Myocardial infarction (%) TLR (%) TVR (%) Any revascularization (%) Target vessel failure (%) Overall MACE (%) ARC definite/probable stent thrombosis (%) Clinical outcomes median follow-up: 1.5 years

Hazard ratio, 1.56 P=0.27 No. at risk BMS PES Years from stenting Death from any cause % of Patients COPD MI Lung CA SBO Stroke MI unknown BMS PES

No. at risk BMS PES Myocardial infarction % of Patients Years from stenting PES BMS Hazard ratio, 0.67 P=

Hazard ratio, 0.38 P= No. at risk BMS PES Years from stenting Target lesion revascularization % of Patients PES BMS

No. at risk BMS PES Target vessel fail ure Cardiac death, MI, TVR Years from stenting % of Patients Hazard ratio, 0.65 P=0.03 BMS PES

Clopidogrel use P=NS % of Patients

In saphenous vein graft lesions, compared to a similar bare metal stent, the Taxus ™ PES resulted in: Conclusions

In saphenous vein graft lesions, compared to a similar bare metal stent, the Taxus ™ PES resulted in: Significant reduction in 12-month binary angiographic restenosis, target lesion revascularization and target vessel failure Conclusions

In saphenous vein graft lesions, compared to a similar bare metal stent, the Taxus ™ PES resulted in: Significant reduction in 12-month binary angiographic restenosis, target lesion revascularization and target vessel failure Trends for lower target vessel revascularization, myocardial infarction Conclusions

In saphenous vein graft lesions, compared to a similar bare metal stent, the Taxus ™ PES resulted in: Significant reduction in 12-month binary angiographic restenosis, target lesion revascularization and target vessel failure Trends for lower target vessel revascularization, myocardial infarction No difference in mortality and stent thrombosis Conclusions