Randomized Controlled Trial of the Medtronic Endeavor Drug-Eluting Coronary Stent System Versus the Cypher Sirolimus-Eluting Coronary Stent System in De Novo Native Coronary Artery Lesions - ENDEAVOR III: Follow-Up to 5 Years


The goal of the trial was to evaluate treatment with the Endeavor zotarolimus-eluting stent (ZES) compared with treatment with the Cypher sirolimus-eluting stent (SES) among patients with de novo coronary lesions.


Treatment with the ZES would be noninferior compared with the SES for assessment of in-segment late lumen loss at angiographic follow-up.

Study Design

Patients Enrolled: 436
Mean Follow Up: 9 months; follow-up to 5 years
Mean Patient Age: 61 years

Patient Populations:

  • Single de novo native coronary lesion with vessel diameter 2.5-3.5 mm, lesion length 14-27 mm, and stent length 18-33 mm


  • Serum creatinine >2.0 mg/dl
  • Acute MI within 72 hours
  • Stroke or transient ischemic attack within 6 months
  • Left ventricular ejection fraction <30% 
  • Prior drug-eluting stent in any vessel
  • Unprotected left main disease

Primary Endpoints:

  • In-segment late lumen loss at 8 months, assessed for noninferiority (0.20 mm noninferiority margin of difference)

Secondary Endpoints:

  • Target lesion revascularization, target vessel revascularization, and target vessel failure at 9 months
  • Angiographic binary restenosis at 8 months

Drug/Procedures Used:

Patients at 30 US sites were randomized in a 3:1 manner to use of the ZES (n = 323) or use of the SES (n = 113). Patients underwent angiographic and intravascular ultrasound (IVUS) follow-up at 8 months. Clinical follow-up continued to 5 years.

Concomitant Medications:

Antiplatelet therapy for ≥3 months

Principal Findings:

Baseline clinical characteristics were well-balanced between groups, with the exception of gender (65% males in the ZES group vs. 81% in the SES group, p < 0.01). Diabetes was present in 29% of patients, and prior myocardial infarction (MI) in 20%. Device success was higher in the Endeavor group (98.8% vs. 94.7%, p = 0.02), as was procedure success (98.1% vs. 91.2%, p = 0.002). At 1 year, dual antiplatelet therapy was used in 18% versus 17% of participants, and at 5 years, it was used in 7.0% versus 6.4% of participants, respectively, for ZES versus SES.

Angiographic follow-up was completed in 87% of the ZES group versus 83% of the SES group. The primary endpoint of in-segment late loss did not meet the noninferiority margin of difference of 0.20 mm, with an observed difference of 0.21 mm greater late loss in the ZES group (0.34 mm for ZES vs. 0.13 mm for SES, p = NS for noninferiority, p < 0.001 for superiority). In-stent late loss was also larger in the ZES group (0.60 mm vs. 0.15 mm, p < 0.001), as was binary restenosis both in-stent (9.2% vs. 2.1%, p = 0.02) and in-segment (11.7% vs. 4.3%, p = 0.04). In the IVUS cohort, volume obstruction was higher with the ZES (16.1% vs. 2.7%, p < 0.001), and acquired late incomplete apposition was lower (0.5% vs. 5.9%, p = 0.02).

At 9-month follow-up, the composite of target vessel failure did not differ by treatment group (12.0% vs. 11.5%, p = 1.0), nor did the individual components of death (0.6% vs. 0%) or target lesion revascularization (6.3% vs. 3.5%, p = 0.34), but non-Q-wave MI was lower in the ZES group (0.6% vs. 3.5%, p = 0.04).

At 5 years, death occurred in 5.2% versus 13.0% (p = 0.02), cardiac death occurred in 0.3% versus 2.8% (p = 0.06), Q-wave MI occurred in 0.3% versus 0.9% (p = 0.45), non-Q-wave MI occurred in 0.7% versus 3.7% (p = 0.04), definite/probable stent thrombosis occurred in 0.7% versus 0.9% (p > 0.99), and target lesion revascularization occurred in 8.1% versus 6.5% (p = 0.68), respectively, for ZES versus SES.


Among patients with single de novo coronary lesions, use of the ZES was associated with larger in-segment late lumen loss at 8-month angiographic follow-up compared with the SES.

While angiographic parameters were better in the SES group, there was no difference in target lesion revascularization or target vessel failure between groups; however, non-Q-wave MI was actually lower in the ZES stent group at intermediate follow-up.

At 5 years, when very few patients remained on dual antiplatelet therapy, death, cardiac death, and non-Q-wave MI were all significantly lower in the ZES group. This must be interpreted cautiously since the actual number of events was small. Tthe trial was powered for a continuous angiographic endpoint, not clinical events, and there was no difference in stent thrombosis or target lesion revascularization at intermediate- or long-term follow-up.

Moreover, in a broad patient population, the SORT OUT III trial documented superior clinical efficacy of SES over ZES at 18 months, while the ZEST trial documented similar clinical efficacy at 12 months. The ongoing PROTECT trial will also provide addition data in over 8,000 patients regarding the comparative efficacy and safety of these two stents.


Kandzari DE, Mauri L, Popma JJ, et al. Late-Term Clinical Outcomes With Zotarolimus- and Sirolimus-Eluting Stents: 5-Year Follow-Up of the ENDEAVOR III (A Randomized Controlled Trial of the Medtronic Endeavor Drug [ABT-578] Eluting Coronary Stent System Versus the Cypher Sirolimus-Eluting Coronary Stent System in De Novo Native Coronary Artery Lesions). JACC Cardiovasc Interv 2011;4:543-550.

Kandzari DE, et al. Comparison of Zotarolimus-Eluting and Sirolimus-Eluting Stents in Patients With Native Coronary Artery Disease. J Am Coll Cardiol 2006;48:2440-7.

Presented by Dr. David E. Kandzari at TCT 2005, Washington, DC.

Clinical Topics: Invasive Cardiovascular Angiography and Intervention, Stable Ischemic Heart Disease, Chronic Angina

Keywords: Myocardial Infarction, Coronary Restenosis, Coronary Stenosis, Thrombosis, Coronary Vessels, Sirolimus, Diabetes Mellitus, Stents

< Back to Listings