Author + information
- Received February 19, 2007
- Revision received July 3, 2007
- Accepted July 8, 2007
- Published online October 23, 2007.
- Dietrich Baumgart, MD⁎ (, )
- Volker Klauss, MD,
- Frank Baer, MD,
- Franz Hartmann, MD,
- Helmut Drexler, MD,
- Wolfgang Motz, MD,
- Heinrich Klues, MD,
- Stefan Hofmann, MD,
- Wolfgang Völker, MD,
- Thomas Pfannebecker,
- Hans-Peter Stoll, MD,
- Georg Nickenig, MD,
- SCORPIUS Study Investigators
- ↵⁎Reprint requests and correspondence:
Dr. Dietrich Baumgart, Preventicum—Center of Diagnostics Theodor-Althoff-Str. 47, 45133 Essen, Germany.
Objectives This study sought to analyze the effectiveness of drug-eluting stents in a high-risk group of diabetic patients. Previously, this had been analyzed only in substudies of larger trials or in clinical investigations enrolling a small number of patients.
Background Drug-eluting stents are highly effective in reducing the rate of in-stent restenosis.
Methods Two hundred patients with diabetes and de novo coronary artery lesions were enrolled in 16 centers: 98 were randomly assigned to sirolimus-eluting stents (SES) and 102 received bare-metal stents (BMS). The primary end point was in-segment late luminal loss. Major adverse cardiac events (MACE) rate was analyzed at 30 days and 8 and 12 months.
Results The extent of in-segment late luminal loss in the SES group was 0.18 mm compared with 0.74 mm in the BMS group. In-segment restenosis was identified on follow-up angiography in 8.8% of the patients in SES and in 42.1% in BMS (p < 0.0001). Target lesion revascularization was performed in 5.3% of the patients in SES and in 21.1% of the patients in BMS (p = 0.002). The SES was effective in the treatment group with oral diabetic medication as well as in the insulin-dependent treatment group (3.6% SES vs. 38.8% BMS). There was no subacute stent thrombosis in the SES group up to 1 year. The MACE rate was not significantly different at 30 days. At 12 months, MACE rate was 14.7% in SES versus 35.8% in BMS.
Conclusions The SES is safe and highly effective in patients with diabetes mellitus and coronary artery disease and associated with a significant decrease in the extent of late luminal loss.
Diabetic patients with coronary artery disease constitute a high-risk group for future cardiovascular events (1). Percutaneous interventional treatment of coronary stenosis in this patient cohort is associated with an increased risk of repetitive restenosis (2–4). Therefore coronary bypass surgery has been considered as a primary treatment for diabetic patients (5,6).
The effectiveness of drug-eluting stents, especially of the sirolimus-eluting stent (SES), has been demonstrated in several studies (7–9). The effectiveness of SES in diabetic patients has so far been shown only in substudies of large multicenter trials (10), in single center studies, or smaller studies employing just a few selected study sites (11–13). Furthermore, diabetic patients are known for increased thrombocyte aggregation and thrombotic events (14). This has raised specific safety issues for the implantation of drug-eluting stents in diabetic patients. The SCORPIUS (German Multicenter Randomized Single Blind Study of the CYPHER Sirolimus-Eluting Stent in the Treatment of Diabetic Patients with De Novo Native Coronary Artery Lesions) study sought to investigate the impact of SES on neointimal proliferation in diabetic patients and to analyze morbidity and mortality in this particular patient subset.
Study design and eligibility
This randomized, open-label study was initiated and designed by the investigators, complied with the provisions of the Declaration of Helsinki regarding investigation in humans, and was approved by local ethic committees for all investigational sites, and written informed consent was obtained from all patients. The study was supported by a grant from the Cordis Company, Langenfeld, Germany.
Eligible patients had a history of stable or unstable angina and signs of myocardial ischemia and a manifest diabetes mellitus, proven by fasting glucose (12 h) >127 mg/dl or oral glucose challenge: ≥200 mg/dl after 2 h or diabetes mellitus currently treated with oral antidiabetics or insulin. A single newly diagnosed lesion in a native coronary artery resulting in stenosis of 51% to 99% of the luminal diameter, given a reference diameter between ≥2.5 and ≤3.5 mm and a lesion length measuring ≤42 mm (as estimated visually on angiography) was targeted for treatment. The major criteria for exclusion were recent myocardial infarction (within the previous 24 h); an ejection fraction of <30%; a target lesion located in an ostium, at a bifurcation, within an “unprotected” left main coronary artery, or in a vessel with thrombus or severe calcification; and the need to treat non-target lesions in the same or a different coronary vessel during the index procedure.
Before the index procedure, eligible patients were randomly allocated to treatment with an SES (Cypher Stent, Cordis, Miami, Florida) or a standard bare metal stainless steel stent (Bx-Velocity, Cordis) in a 1:1 ratio by single blinded telephone randomization at each site by an external provider.
Coronary stent procedure
Before and after the index procedure, all patients received oral aspirin (at least 100 mg daily starting at least 12 h before procedure) and oral clopidogrel (a loading dose of 300 to 375 mg 24 h before the procedure and then 75 mg daily for 6 months). During the procedure, intravenous heparin boluses were administered. The use of intravenous glycoprotein IIb/IIIa inhibitors was at the discretion of the physician. Lesions were treated with standard interventional techniques, including mandated balloon dilation before placement of the stent. One or 2 stents of the assigned type were used to treat the target lesion. The SES contained 140 μg of sirolimus/cm2of stent-surface area within a copolymer matrix that was 5- to 10-μm thick and was designed to release approximately 80% of the total dose of sirolimus in 30 days.
Data collection, follow-up, and core laboratory analyses
All data were submitted to a third-party data coordinating center (EuroNet CTTF, Aachen, Germany) independent of the sponsor, and the investigators had full access to the data. The investigators also initiated, performed, and reviewed all analyses and made the decisions about publication. Clinical follow-up information was obtained for all patients by the research coordinators at each site at 30, 240, and 365 days. All clinical end points were adjudicated by an independent clinical-events committee that was unaware of the treatment-group assignments. A separate independent data and safety monitoring board that was not affiliated with the study sponsor or the investigators reviewed all data periodically to identify potential safety issues (all complications, including death, stent thrombosis, and myocardial infarction) and to review the conduct of the study (the pace of enrollment, patients’ eligibility, and compliance with data collection). The monitoring board did not perform an interim analysis with regard to the primary efficacy end point at 8 months, because enrollment was completed before the 9-month primary end point was reached in the first patient.
Coronary angiograms, obtained at baseline, at the completion of the stenting procedure, and at 240 days of follow-up, were submitted to an independent angiographic core laboratory (Department of Cardiology, University Clinic Frankfurt, Germany, not a study site) and were analyzed with the use of a computer-based system (Medis, Leiden, the Netherlands). “Binary” restenosis was defined as stenosis of at least 50% of the luminal diameter in the target lesion. Late luminal loss was defined as the difference between the minimal luminal diameter (MLD) at the completion of the stenting procedure and that measured during follow-up. Quantitative angiographic measurements of the target lesion were obtained in the “in-stent” zone (including only the stented segment) and in the “in-segment” zone (including the stented segment as well as the margins 5 mm proximal and distal to the stent).
Study end points
The primary end point of this study was the late luminal loss within the in-segment zone at 8 months.
The secondary clinical end points included the late luminal loss both within the stent and within the proximal and distal margins, revascularization of the target lesion or the target vessel (clinically driven coronary artery bypass graft [CABG] or repeated percutaneous transluminal coronary angioplasty [PTCA] due to restenosis or closure of the target lesion), the rate of in-stent restenosis, lesion, as well as procedural success and all major adverse events determined for the first 30 days and cumulatively at 8 and 12 months after the placement of the stent. Stent thrombosis was retrospectively analyzed with the new definition of stent thrombosis generated by the Academic Research Consortium (ARC) (15). Consequently, stent thrombosis was classified as “definite,” “probable,” or “possible” on the basis of the major adverse cardiac events (MACE) narratives written by an independent reviewer.
The primary efficacy end point was the late loss in the analyzed segment at month 8. The segment encompassed the proximal peri-stent area (5 mm proximal to the stent), the stent, and the distal peri-stent area (5 mm distal to the stent). The planned sample size of 190 patients provided 80% power to detect a difference of 0.3 mm in mean late loss between the 2 treatment arms at a significance level of 5%, provided the SD of late loss did not exceed 0.7 mm. Treatment groups were compared with analysis of covariance, with late loss as dependent variable, treatment (type of stent) and center as factors, and the post-procedure MLD within segment as covariate. Secondary end points were evaluated with the following methods: analysis of covariance for the late loss within stent and within proximal and distal peri-stent area; Kaplan-Meier estimates and log-rank tests for the time to first target lesion revascularization (TLR), first target vessel revascularization, and first target vessel failure (within 274 days post-intervention); and Cochran-Mantel-Haenszel tests, adjusting for center effects and categorical variables. Analyses of secondary end points were exploratory, examined the hypothesis of equality of the 2 stents, and used a 2-sided significance level of 5%.
All analyses, except those of angiographic data, were based on the safety population, which consisted of all patients in whom the intervention for study stent implantation was started. The analyses of angiographic data used all randomized patients in the safety population in whom angiographic data were available both immediately post-procedure and at the month 8 follow-up. The follow-up angiography was only accepted if it had been performed at least 214 days after intervention and no new intervention had been carried out during this time or if it had been performed within 14 days before re-PTCA of target lesion or CABG of target vessel or if it showed a significant restenosis (≥50% diameter stenosis). Patients were analyzed as treated in the safety analyses and as randomized in the analyses of angiographic data.
The statistical analyses were described in the study protocol and detailed in a statistical analysis plan before database lock. Analyses were performed with SAS (version 8.2, SAS Institute, Cary, North Carolina). The statistical analyses were performed independently by Datamap GmbH (Freiburg, Germany). All reported p values are 2-sided.
Characteristics of the patients and the lesions
Between December 2002 and September 2004, 200 patients gave written informed consent and 98 patients were randomly assigned to the SES group, whereas 102 patients were assigned to the bare-metal stent (BMS) group. (Table 1).
After randomization, 7 patients (3 in the SES group and 4 in the BMS group) withdrew their written consent and were excluded from the analysis; 193 patients entered the safety analysis. Three patients (1 in the SES group and 2 in the BMS group) did not receive the correct study stent The groups were well matched, with no significant differences in the frequency of cardiac risk factors, demographic data, or lesion characteristics (Table 2).A total of 230 stents were used, 118 in the SES and 112 in the BMS group.
The average balloon-inflation pressure after stenting was 15 atm, and the mean (± SD) balloon/artery ratio 1.15. An average of 1.2 stents were implanted per target lesion, with a mean stent length of 18.3 ± 8.3 mm. The mean ratio of the stent length to the lesion length was 1.6 ± 0.5 (Table 3).A total of 230 stents were used, 118 in the SES and 112 in the BMS group. The distribution according to stent size for the SES and BMS groups was as follows: stents of 2.5 mm in diameter: 35 in the SES and 24 in the BMS group; stents of 3.0 mm in diameter: 83 in the SES and 88 in the BMS group. There were no stents used with 3.5 mm in diameter.
Quantitative coronary angiography
The dimensions of the lesion at baseline were similar in the 2 groups (Table 3). Follow-up angiographic data were available in 68 patients in the SES group (72% of the patients assigned to undergo angiographic follow-up) and 76 in the BMS group (80% of the patients assigned to undergo angiographic follow-up). Table 4shows that at follow-up the MLD, stenosis as a percentage of the luminal diameter, and the late luminal loss in both the in-stent zone and the in-segment zone were all improved with the SES as compared with the standard stent (p < 0.001 for all comparisons). The frequency of binary in-segment restenosis (stenosis of at least 50% of the luminal diameter) was 8.8% in the SES group and 42.1% in the standard-stent group (p < 0.001).
Major adverse cardiac events are listed in Table 5.In-hospital events occurred at a similar frequency in the 2 groups (including death, myocardial infarction, and repeated revascularization); the proportion of patients with any in-hospital major adverse event was 1.1% in the SES group and 3.2% in the standard-stent group (p = 0.32), including 1 death of unknown cause in the SES group and 3 myocardial infarctions as well as 1 TLR in the BMS group.
There was a lower rate of out-of-hospital adverse events during the 240 days of follow-up in the SES group than in the standard-stent group (14.7% vs. 29.5%, p = 0.02); reductions were mainly attributable to the decrease in TLRs (from 21.1% to 5.3%). The number of deaths and myocardial infarctions were not different between the groups. In the SES group there were 3 cardiac deaths (2 of which occurred suddenly), 1 of unknown origin and 4 myocardial infarctions (2 of which occurred in the target vessel territory). In the BMS group there were 2 noncardiac deaths, 1 death of unknown cause and 2 myocardial infarctions in nontarget vessel territories. During the 8 months’ follow-up coronary artery bypass operations were performed in 3 patients in the SES group (all involving the target segment) and in 5 patients in the BMS group (in 2 not involving the target segment). Target lesions were revascularized by percutaneous interventions (PCIs) in 2 patients in the SES group and in 17 patients in the BMS group.
Of 28 (39%) patients in the BMS group with a significant restenosis, 20 (21%) received a TLR, of which 13 (14%) had a positive ischemic test. In the SES group, 5 (5.3%) of 6 (8.8%) patients with a restenosis received a repeat revascularization, of whom all had a positive ischemic test.
The rate of survival free of target vessel failure for 240 days increased from 77.4% with the standard stent to 94.3% with a SES (p < 0.001) (Fig. 1).
During the additional 4 months of clinical follow-up there was 1 cardiac death in the BMS group, whereas no additional death or myocardial infarction occurred in the SES group. One patient in the SES group and 7 patients in the BMS group had to undergo TLR.
In the course of 12 months’ follow-up, 25 patients in the SES group and 15 in the BMS group received an additional PCI outside the target vessel. Seven patients in the SES group and 3 patients in the BMS group received a PCI within the target vessel but outside the target segment.
There was no subacute stent thrombosis in any group (occurring <24 h after placement of the stent or during the first 30 days after placement). There was 1 late stent thrombosis (occurring between 31 and 270 days after placement)—1 in the SES group after early discontinuation of the antiplatelet combination therapy with aspirin and clopidogrel. In most patients, the dual platelet therapy was given longer than recommended. At 8 months, 90% of the patients in the SES group and 87% of the patients in the BMS group were still taking clopidogrel, whereas almost all patients were taking aspirin (99% vs. 98%). At 12 months 35% of the patients in the SES group and 33% of the patients in the BMS group were still taking clopidogrel, whereas the rate of patients taking aspirin was 81% vs. 80%.
According to the newly defined ARC criteria for stent thrombosis (see previous text), there was no “definite” stent thrombosis in the SES group as opposed to 2 “definite” stent thromboses in the BMS group. In the latter group, an adverse event in 1 additional patient was classified as a “possible” stent thrombosis. Furthermore, in the SES group 2 adverse events were defined as “probable” and 2 adverse events were defined as “possible” stent thromboses (Table 6).
This first German multicenter study is the largest randomized study in diabetic patients comparing drug-eluting stents versus BMS. At the same time it is the second multicenter study in this particular patient cohort. It demonstrates an excellent result for SES in diabetic patients. This is of considerable importance, because to date clinical studies have not been adequately designed nor have they been large enough or sufficiently powered to reveal detailed results with respect to the safety and effectiveness of SES in the high-risk group of diabetic patients.
The first data on the effectiveness of the SES in diabetic patients were revealed by a sub-analysis of the SIRIUS (Sirolimus-Eluting Stent In De Novo Native Coronary Lesions) trial in 279 patients. At 270 days, TLR rate was reduced in diabetic patients from 22.3% in the BMS group to 6.9% in the SES group (10). Likewise, MACE were reduced in diabetic patients from 25% to 9.2% with SES (10).
Later trials specifically designed to answer detailed questions on SES in diabetic patients confirmed the superiority of the SES in this high-risk population for recurrent restenosis. Both the smaller nonrandomized Porto I trial (13) as well as the larger mono-centric ISAR-DIABETES (Paclitaxel-Eluting or Sirolimus-Eluting Stents to Prevent Restenosis in Diabetic Patients) study (11) demonstrated TLR rates of 1.7% and 6.4%, respectively. In the latter study, in segment late luminal loss was 0.24 mm greater in the paclitaxel-stent group than in the SES group (p = 0.002) (11).
Sabate et al. (12) reported with the DIABETES (Diabetes and Sirolimus-Eluting Stent) trial the first randomized multicenter investigation of SES in diabetic patients. In 80 diabetic patients with 111 lesions, in segment late lumen loss in the SES group was 0.06 ± 0.4 mm as compared with 0.47 ± 0.5 mm in the BMS group. Also TLR (7.3%) and MACE (11.3%) rates at 9 months were significantly lower in the SES group.
In the present study late luminal loss within the analyzed segment was chosen as a primary end point, because it was regarded as the best surrogate end point parameter for the clinical outcome of the study at the time point of the study design in 2002 (16). This decision was based on the experience gained from the early era of drug-eluting stents where late luminal loss in the analyzed segment was thought to best reflect the degree of neointimal proliferation (17). Meanwhile, detailed analyses of large trials with SES have revealed that in-stent late loss as opposed to in-segment late loss is monotonically correlated with the probability of restenosis and yields a more efficient estimate of the restenosis process in the era of lower binary restenosis rates (18). Independent of the primary end point, in stent late loss also showed a significant reduction in the SES group as opposed to the BMS group in this study.
The late luminal loss within the analyzed segment follows basically the same pattern known from previous trials. The proximal edge showed a slight late luminal loss, whereas distally there was hardly any intimal proliferation (19). Although speculative, this difference between proximal and distal edge response has most likely been attributed to a more intense manipulation of the proximal lesion border, given the pre-dilation procedure as well as the final stent implantation. The good results might be attributed to the fact that no final redilatation that leads to additional vessel trauma was performed. Certainly, there was no positive remodeling with a negative late luminal loss as has been reported by the DIABETES study (12).
Although late luminal loss is an appropriate parameter to assess the effectiveness of a drug-eluting stent, TLR is a decisive parameter for the effectiveness of the treatment from the patient’s perspective. Target lesion revascularization, however, is a more subjective variable, and the repeat intervention might be driven by a number of factors. Clinical symptoms and/or a positive noninvasive test, however, in diabetic patients are often unreliable, owing to diabetic neuropathy (20). Whether a more subjective and operator-driven repeat intervention will be to the long-term benefit of the patient or will induce repetitive restenosis is a matter of great debate (21). The present study reveals that the oculo stenotic reflex seems to be much larger in the BMS group. Of 28 patients with a significant restenosis, 13 had a positive ischemic test in the BMS group as opposed to the patients in the SES group in which all patients with a repeat intervention had a positive ischemic test. The reasons for the increased interventional rate in the BMS group, despite a missing positive ischemic test, remain unclear and a matter of speculation but might be biased, owing to the open label design. Clearly, the clinical superiority of the SES in diabetic patients could be demonstrated independent of oculo stenotic reflex.
Although the rate of revascularization in the target segment (5.3%) was low in the SES group, a number of additional lesions at non-target sites within the target vessel or outside the target vessel had to be treated, owing to the progression of the underlying diabetic atherosclerotic disease. This rate for additional coronary interventions is wholly comparable to other studies (13) and has to be taken into account in the cost-effectiveness analysis in diabetic patients.
The effectiveness of SES especially in insulin-dependent diabetic patients has been a matter of debate. The DIABETES trial has demonstrated similar reductions in angiographic and clinical parameters of restenosis in non-insulin and insulin-requiring patients (12). On the basis of the current data, the SES reduces the rate of restenosis independent of the underlying diabetic treatment. Although the subgroups in the different treatment groups contain a rather small number of patients, the effectiveness of the SES nevertheless seems best in the insulin-dependent group of diabetic patients with a TLR rate of 3.6%. This hypothesis is underlined by the fact that the overall TLR rate is lower in our study (5.3%) as compared with the ISAR-DIABETES study (11) (6.4%), in which the rate of insulin-dependent diabetic patients was lower (28.8% vs. 43%). Likewise, the late luminal loss in the ISAR-DIABETES study was larger (0.44 ± 0.06 mm) in the group not receiving insulin than in the insulin-dependent group (0.41 ± 0.42 mm).
There seems to be an increasing concern about safety issues in conjunction with the implantation of drug-eluting stents (22). This issue is of particular importance, because diabetic patients per se have a significantly higher complication rate compared with nondiabetic patients (23). Our analysis, however, yields no subacute stent thrombosis in the SES group. We experienced only 1 late thrombosis after discontinuation of the antiplatelet therapy on day 30 by an urologist.
Also, according to the newly defined ARC criteria for stent thrombosis, the SES group yielded no inferiority as compared with the BMS group under a comparable dual antiplatelet therapy. Our data are supported by the results from other trials. Neither the ISAR-DIABETES study nor the DIABETES trial reported any subacute stent thrombosis within 9 months in the SES group. The MACE rate remained unchanged between 8 and 12 months in the SES group, whereas it increased by 6.3% in the BMS group, mainly owing to additional percutaneous interventions. Besides its effectiveness, the SES is also a safe treatment in diabetic patients.
These findings are well in accordance with the newly published meta-analysis by Kastrati et al. (24), who concluded that diabetic patients carry chances similar to those of nondiabetic patients for having advantages and disadvantages from SES compared with BMS.
The effect of systemic additional drug treatment was not analyzed in this investigation. None of the patients received glitazones, which in other studies have been shown to beneficially affect the rate of restenosis in type 2 diabetic patients (25).
Of 200 patients randomized in the study, 156 received an angiographic follow-up. Three patients in the SES group and 2 patients in the BMS group were lost to follow-up. Twelve patients in the SES group and 13 patients in the BMS group refused a repeat angiography, owing to the lack of clinical symptoms. An almost equal number of patients in each group terminated the study prematurely (SES group: n = 12; BMS group: n = 13). Because the number of missing angiographies at follow-up is balanced between the groups, no systematic bias seems to influence the results. Conclusions about safety aspects are limited to 1 year with respect to the pre-defined follow-up period. The safety analysis, however, is continued.
Because the primary end point of the study was late luminal loss, the study was not powered to discriminate a statistical difference regarding the overall MACE and late stent thrombosis rates. Although the absolute numbers are well comparable to other studies, this limitation has to be taken into account in the interpretation of the results.
The SES is safe and highly effective up to 1 year in patients with diabetes mellitus and coronary artery disease and is also associated with a significant decrease in the extent of late luminal loss compared with a BMS.
This study was initiated by the investigators and supported by a grant from Cordis, Johnson & Johnson, Langenfeld, Germany.
- Abbreviations and Acronyms
- bare-metal stent(s)
- coronary artery bypass graft
- major adverse cardiac events
- minimal luminal diameter
- percutaneous coronary intervention
- percutaneous transluminal coronary angioplasty
- sirolimus-eluting stent(s)
- target lesion revascularization
- Received February 19, 2007.
- Revision received July 3, 2007.
- Accepted July 8, 2007.
- American College of Cardiology Foundation
- Grundy S.M.,
- D’Agostino R.B. Sr..,
- Mosca L.,
- et al.
- Abizaid A.,
- Kornowski R.,
- Mintz G.S.,
- et al.
- Elezi S.,
- Kastrati A.,
- Pache J.,
- et al.
- Frye R.L.,
- Brooks M.M.,
- Nesto R.W.
- Kuntz R.E.
- Moussa I.,
- Leon M.B.,
- Baim D.S.,
- et al.
- Sabate M.,
- Jimenez-Quevedo P.,
- Angiolillo D.J.,
- et al.
- Seabra-Gomes R.
- Moreno P.R.,
- Fuster V.
- Serruys P.W.,
- Cosortium A.R.
- Popma J.J.,
- Leon M.B.,
- Moses J.W.,
- et al.
- Regar E.,
- Serruys P.W.,
- Bode C.,
- et al.
- Mauri L.,
- Orav E.J.,
- O’Malley A.J.,
- et al.
- Mehran R.,
- Dangas G.,
- Abizaid A.S.,
- et al.
- Marchant B.,
- Umachandran V.,
- Stevenson R.,
- Kopelman P.G.,
- Timmis A.D.
- Kuchulakanti P.K.,
- Chu W.W.,
- Torguson R.,
- et al.
- Nishio K.,
- Sakurai M.,
- Kusuyama T.,
- et al.