This article has a correction. Please see:
Author + information
- Received September 23, 2004
- Revision received October 10, 2004
- Accepted October 25, 2004
- Published online January 18, 2005.
- Jean-Jacques Goy, MD, FESC* (, )
- Jean-Christophe Stauffer, MD, FACC,
- Manon Siegenthaler,
- Alain Benoît and
- Charles Seydoux, MD
- ↵*Reprint requests and correspondence:
Dr. Jean-Jacques Goy, Chemin des Croix-Rouges 7, 1007 Lausanne, Switzerland
Objectives We conducted this trial to assess whether a sirolimus-eluting stent (SES) produces similar results to a paclitaxel-eluting stent (PES) when used in the real world of interventional cardiology.
Background Several drug-eluting stents have been shown to exert a beneficial effect on restenosis when used in the treatment of coronary artery disease. Any potential superiority of one drug over the others, however, is still unknown.
Methods To evaluate whether a PES or an SES is superior in daily practice, we randomized all patients suitable to receive a drug-eluting stent in our institution. Clinical follow-up was obtained after at least six months.
Results A total of 202 patients were included in this trial. One hundred patients received a PES and 102 received an SES. Procedural success was 99% in both groups. Incidence of major adverse cardiac events at follow-up (mean 7 ± 2 months) was 4% with the PES and 6% with the SES (p = 0.8). The need for target lesion revascularization was very low in both groups (1% with the PES and 3% with the SES).
Conclusions Our results confirm that the high success rate obtained with both stents in randomized trials can be replicated in routine clinical practice. In this small group of patients we were unable to show any advantage of one stent over the other.
The use of stents has significantly improved the outcome of percutaneous coronary intervention (PCI) (1,2). However, despite major advances in angioplasty and stenting, in-stent restenosis has remained a major limitation until recently. Since 2000, drug-eluting stents have emerged as a very promising approach in preventing restenosis, and several different compounds have been shown to have a major impact on both the angiographic and the clinical outcome (3–9). However, even if the randomized trials show similar clinical benefit with these drug-eluting stents when compared with bare metal stents, no direct comparison between various devices is available, and prospective randomized data are strongly needed. Thus, we prospectively collected all data pertaining to the use of the sirolimus-eluting stent (SES) and paclitaxel-eluting stent (PES) in a randomized prospective fashion for a series of consecutive patients treated in our institution. The trial was started as soon as the PES became commercially available in Switzerland, with the purpose to compare the two stents in the real world of interventional cardiology.
Patients and methods
Between April 2003 and January 2004, all patients requiring PCI and selected to receive a drug-eluting stent were randomized in a prospective trial comparing a PES (Taxus, Boston Scientific, Maple Grove, Minnesota) with an SES (Cypher, Cordis Corp., Miami, Florida). There was no restriction to inclusion except the preference of the patient or our uncertainty regarding obtaining follow-up information.
The procedure was performed via the right femoral artery through a 6-F guiding catheter. The revascularization was nearly always done during the same session as the diagnostic angiography, except for those patients with left main stenosis. Intravascular ultrasound was not used. Direct stenting was allowed but was restricted to low-risk lesions and left to the operator's discretion. All patients were pretreated with aspirin 100 mg/day, and a minority received clopidogrel 75 mg/day for five to seven days before the procedure. Intravenous heparin (70 U/kg) was given at the beginning of the procedure. If not given earlier, a 300 mg loading dose of clopidogrel was administered at the end of the procedure. A successful procedure was defined as a residual stenosis <20% without a major adverse cardiac event (MACE) during the in-hospital stay.
Creatine kinase, creatine kinase-MB fraction, and/or troponin values were measured at least once on the morning of the day following the procedure. A diagnosis of non–Q-wave myocardial infarction (MI) was made if a value above twice the upper limit of normal was measured. A 12-lead electrocardiogram was recorded at the end of the procedure, and further tracings were obtained if indicated. Patients remained in hospital until the next day. Long-term aspirin 100 mg/day was prescribed at discharge, together with 2 to 12 months of clopidogrel 75 mg/day.
Quantitative coronary angiography evaluation was obtained in multiple views. For patients with angiographic follow-up, restenosis was defined as a 50% or more reduction of the luminal diameter occurring within the stented segment or the 5 mm proximal and distal to the stent.
Clinical follow-up was obtained at one and six months, either by a visit or by telephone contact with the patient or his/her referring physician. Information was collected on vital status, occurrence of MI, additional revascularization procedures, coronary angiography, clinical angina status, and current medication. As a rule, control angiography was performed only when clinically required (clinical or silent ischemia). Death, MI, additional PCI or coronary artery bypass grafting to the target lesion, documented target lesion occlusion, or stent thrombosis were considered to be a MACE.
On the basis of previous data reported with both stents (14% of events with the PES and 6% with the SES), the number of patients initially planned was 173 per group (80% statistical power). However, because we did not observe a significant difference after the first 202 patients, we decided as clinicians that it would be acceptable to use one stent or the other indifferently in clinical practice, because 2,936 patients would have been necessary to demonstrate superiority of one device. Values are expressed as mean ± SD. All tests of significance were two-tailed (Fisher exact test, chi-square, and Mann-Whitney). Statistical significance was assigned at the p < 0.05 level.
During the study period 405 patients were treated; 202 of these (50%) were included in the trial. One hundred were randomized to a PES and 102 were randomized to a SES. The baseline data are given in Tables 1 and 2.⇓The two groups were comparable.
A total of 183 patients underwent a single procedure (89 in the PES group and 94 in the SES group). Forty-four patients (22%) had multivessel revascularization. In 25 patients, all lesions were treated in a single session. In the remaining 19, a staged procedure was performed. A mean number of 1.5 stents/patient were implanted in both groups (Table 2). Direct stenting was attempted in 35 patients in the PES group and in 32 patients in the SES group and was successful in 32 (32%) and 28 (28%) patients, respectively.
A platelet glycoprotein IIb/IIIa inhibitor was used in 11 patients (5 %): 4 in the PES group and 7 in the SES group (p = 0.6). Major in-hospital cardiac events occurred in three patients in the PES group versus three in the SES group (p = 0.9) (Table 3).Three patients in the PES group and one in the SES group had a creatine kinase rise (non–Q-wave acute MI) and one in the SES group had a Q-wave MI. There was no hospital death, and only one patient from the SES group required emergency repeat percutaneous revascularization for persistent chest pain because of a large distal dissection. Neither acute (<24 h) nor subacute (24 h to 30 days) stent thrombosis was observed in either group. The in-hospital stay was 18 ± 8 h.
A follow-up was obtained in all 202 patients (100 %) after a minimal duration of six months (mean 7 ± 2 months) (Table 4).At follow-up, six patients in the SES group and eight patients in the PES group had class II angina or greater (p = 0.6). Eighteen patients in the SES group and 15 patients in the PES group underwent control angiography because they were reporting symptoms of typical or atypical angina pectoris or had undergone PCI of a lesion of particular prognostic value (left main, last remaining vessel, and so on). The overall incidence of MACE was low, 4% in the PES group versus 6% in the SES group (including in-hospital MACE). No patient died or experienced MI in either group after leaving the hospital. Target lesion revascularization was required in three patients, one in the PES group and two in the SES group, and late stent thrombosis was documented in one patient of the SES group. The event-free survival rate at six months was 96% in the PES group and 94% in the SES group (p = 0.8).
To our knowledge this is the first randomized comparison of two drug-eluting stents in the real world of routine clinical practice. Our results confirm the excellent safety profile of both stents when used in various clinical and anatomic subsets. They compare favorably with those reported in previous randomized trials in which event-free survival rates were 94% to 95% with the SES (3–5) and 89% to 92% with the PES (8,9). Moreover, the present series is an unselected cohort of patients, some of them with high-risk lesions such as left main stenosis, in-stent restenosis, or saphenous vein graft lesions, and nearly 30% with diabetes. We were unable to show any difference between the two stents, and the incidence of a MACE was very similar in both groups at six months. Angiographic control was performed only in a minority of patients, and we may have underestimated the incidence of angiographic restenosis and silent stent thrombosis. However, even if such events did occur, there is no evidence that they were more frequent in either of the groups because the clinical outcomes were similar. On the other hand, relying on clinical follow-up only, we avoided potentially unnecessary target lesion revascularization due to the occulostenotic reflex.
Even if no difference between both drug-eluting stents available on the market today could be demonstrated in this prospective trial, a small advantage of one device over the other might exist but require a higher number of patients to reach any statistically demonstrable difference. Such a result would in any case be expected to have only limited impact on clinical practice. Our follow-up duration is relatively short, and this represents a limitation, as it implies failure to assess any potential (but unlikely) late differences between both devices. Both these issues will potentially be resolved in a definite manner when the results of larger, ongoing randomized trials become available.
Our data confirm that the excellent results obtained with drug-eluting stents in randomized trials can be replicated in the real world of routine clinical practice without any obvious detectable differences between the two devices.
We thank Drs. D. Goy-Eggenberger and P. Urban for their assistance in writing the manuscript.
This work was supported by the Clinique Cecil Heart Foundation.
- Abbreviations and acronyms
- major adverse cardiac events
- myocardial infarction
- percutaneous coronary intervention
- paclitaxel-eluting stent
- sirolimus-eluting stent
- Received September 23, 2004.
- Revision received October 10, 2004.
- Accepted October 25, 2004.
- American College of Cardiology Foundation
- Grube E.,
- Silber S.,
- Hauptmann K.E.,
- et al.
- Gershlick A.,
- De Scheerder I.,
- Chevalier B.,
- et al.