Author + information
- Received June 21, 2004
- Revision received September 1, 2004
- Accepted September 13, 2004
- Published online January 4, 2005.
- Michael Schlüter, PhD*,
- Joachim Schofer, MD*,* (, )
- Anthony H. Gershlick, MD†,
- Erick Schampaert, MD‡,
- William Wijns, MD§,
- Günter Breithardt, MD, FACC∥,
- for the E- and C-SIRIUS Investigators
- ↵*Reprint requests and correspondence:
Dr. Joachim Schofer, Othmarscher Kirchenweg 168, D-22763 Hamburg, Germany
Objectives We sought to assess the impact of direct stenting (DS) using the sirolimus-eluting stent (SES) on angiographic and clinical outcomes.
Background The SES is superior to bare-metal stents in the treatment of native de novo coronary artery lesions in randomized, controlled trials.
Methods A post hoc analysis was performed on 225 patients (158 men; 62 ± 11 years old) who received SES in the pooled cohorts of the European and Canadian Sirolimus-Eluting Stent in Coronary Lesions (E-SIRIUS and C-SIRIUS, respectively) trials. Of these patients, 57 (25%) had undergone DS at the investigator's discretion. Lesion predilation preceded SES implantation in the remaining 168 patients.
Results Patient and lesion characteristics were no different between the two subgroups, except for a lower prevalence of moderate to severe lesion calcification (5% vs. predilation 19%, p = 0.017) and a lower baseline diameter stenosis (61.6% vs. predilation 68.1%, p < 0.001) in the DS subgroup. At eight months, in-lesion late loss (0.10 vs. 0.19 mm at predilation, p = 0.14) and in-lesion binary restenosis (2.0% vs. 6.1% at predilation, p = 0.46) tended to be lower after DS. Clinical follow-up at one year revealed non-significantly reduced incidences of target lesion revascularization (1.8% vs. 5.4% at predilation, p = 0.46) and major adverse cardiac events (5.3% vs. 8.9% at predilation, p = 0.57).
Conclusions Direct SES deployment performed at the investigator's discretion was as safe and efficacious at mid-term follow-up as stenting preceded by lesion predilation.
Improvements in stent design now allow for reliable, safe, and cost-effective direct stenting (DS) (1–4). Outcomes with sirolimus-eluting stents (SES) are superior to bare-metal stents (5–8). Only the European Sirolimus-Eluting Stent in Coronary Lesions (E-SIRIUS) trial (7) and its “sister” Canadian trial, C-SIRIUS (8), allowed direct implantation of the SES. Concern has been expressed about the potential drug and/or polymer loss from the SES when directly implanted, which may reduce efficacy. This report assesses the feasibility and safety of DS in the pooled subgroup of patients from the E- and C-SIRIUS trials.
The E- and C-SIRIUS are parallel trials designed to assess long-term clinical outcomes to five years, conducted at 35 centers in Europe and Israel and 8 centers in Canada. Patient and angiographic inclusion and exclusion criteria were identical for both trials (7,8).
Of 452 patients enrolled, 225 (158 men; mean [±SD] age 62 ± 11 years) received one or more SES and formed the basis of the present prespecified post hoc analysis. Direct stenting was performed at 30 of 43 centers in a total of 57 patients, representing a median of 40% (range 10% to 60%) of patients receiving SES in these centers and 25.3% of the total SES patient population. Patient and lesion characteristics in the two subgroups are given in Table 1.
Late loss = the minimum lumen diameter (MLD) after the intervention minus MLD at eight months; binary restenosis = ≥50% diameter stenosis at follow-up; major adverse cardiac event (MACE) = cumulative incidence of death, myocardial infarction (MI), emergency coronary artery bypass graft surgery, and ischemia-driven target lesion revascularization (TLR).
Continuous variables are presented as the mean value ± SD, except where noted, with subgroup differences assessed by the Student ttest. Categorical variables are presented as counts and/or percentages, with differences between subgroups assessed by the Fisher exact test (2 × 2 table) or chi-square test (3 × 2 table). Event-free survival was estimated by the Kaplan-Meier method. Statistical significance was assumed for a two-sided p value of <5%.
Baseline patient and lesion characteristics
Baseline patient and lesion characteristics in the DS and predilation subgroups were well matched (Table 1). However, the prevalence of moderate-to-severe lesion calcification was higher in predilated lesions (19% vs. 5%, p = 0.017). Angulation ≥45° in the proximal vessel was more prevalent in DS lesions (18% vs. 7%, p = 0.036).
Procedure duration and total fluoroscopy time were significantly reduced by DS (Table 2).Stent placement using the chosen approach was successful in all patients, with no crossovers from DS. Mean total stent lengths were 23.6 and 24.1 mm, and multiple stents were implanted in 40% and 51% of patients in the DS and predilation groups, respectively (Table 2). The investigators tended to post-dilate DS less often (28% vs. 43%, p = 0.06).
Quantitative coronary angiography
Reference vessel diameter was similar in both subgroups, but MLD at baseline was significantly greater (1.00 vs. 0.83 mm, p < 0.001), and consequently, baseline diameter stenosis was significantly less (61.6% vs. 68.1%, p < 0.001) in the DS subgroup (Table 3).Stent implantation resulted in identical immediate in-stent and in-lesion angiographic outcomes (Table 3).
Angiographic follow-up at eight months was obtained from 148 patients in the predilation (88.1%) and 51 patients in the DS subgroup (89.5%). Slightly higher mean values in in-stent and in-lesion MLD were noted in the DS subgroup (in-stent: 2.36 vs. 2.24 mm, p = 0.12; in-lesion: 2.08 vs. 1.99 mm, p = 0.26). In-stent late loss was >50% lower in the DS subgroup (0.10 vs. 0.21 mm, p = 0.073; Fig. 1)(in-lesion: 0.10 mm for DS vs. 0.19 mm for predilation) (Table 3). In-lesion binary restenosis was 67% less in the DS subgroup (2.0% vs. 6.1%, p = 0.46; relative risk 0.32, 95% confidence interval [CI] 0.04 to 2.48).
A multivariate model of in-lesion binary restenosis yielded no statistically significant impact of DS.
In-hospital non–Q-wave MIs occurred in three predilation group patients (1.8%); one patient in the DS group (1.8%) developed stent thrombosis inside two abutting stents that evolved into a Q-wave MI and necessitated TLR (Table 4).
At one year, MACE and TLR rates in the DS group were 40% and 67% less than that in the predilation group: relative risks 0.59 (95% CI 0.18 to 1.96) and 0.33 (95% CI 0.04 to 2.53), respectively (Table 4, Fig. 2).One-year event-free survival was 94.7% (DS) versus 91.0% for MACE (predilation, p = 0.37) and 98.2% versus 94.6% for TLR (p = 0.26).
The pooled SES data from the E- and C-SIRIUS trials were used to assess the impact of DS on angiographic and clinical outcomes. The analysis was prespecified and acceptable because the trials had identical patient and lesion inclusion criteria.
Our analysis revealed that DS of the SES was as feasible and efficacious as balloon-facilitated stenting, with 100% technical success achieved in both subgroups and no crossovers from DS. Multiple stents were implanted in 40% and 51% of patients, respectively, and in-lesion late loss and in-lesion binary restenosis at eight months were not statistically different. Thus, anxieties regarding a loss of polymer or drug appear to be unfounded. There were no safety concerns at one month or one year associated with DS of the SES.
This analysis needs cautious interpretation. Assignment to either balloon-facilitated stenting or DS was not randomized, with a significantly lower prevalence of moderately to severely calcified lesions and a significantly lower baseline diameter stenosis in DS. However, coronary artery angulation more often exceeded 45°, and operators tended to post-dilate less frequently in this group.
When SES implantation is considered, DS appears as safe and efficacious as predilation. Provided the targeted coronary artery lesion lends itself to DS, this approach is associated with good mid-term angiographic and clinical outcomes. Preprocedural MLD is a predictor of restenosis after stenting, as are reference vessel diameter and post-procedural MLD (9). The larger preprocedural MLD may have accounted for the trend toward lower eight-month angiographic and improved one-year clinical outcome in DS. The decrease by >50% in in-stent late loss and the lack of restenosis outside the stent associated with DS are noteworthy. The lack of statistical significance of the differences in outcomes after DS and predilation strategy may be due to a type II statistical error, but may indicate there are no true differences. A randomized trial powered to detect any such difference is warranted.
These results are in concordance with a recent post hoc analysis of the TAXUS-II trial (10) and are supported by the Direct Stenting Using the Sirolimus-Eluting Stent (DIRECT) trial (11), which compared a group of 225 patients who underwent DS with a historical predilated SES group from the SIRIUS trial (6). The latter investigation concluded that “direct stenting was non-inferior to pre-dilation for all (clinical and angiographic) endpoints assessed.”
The DS of SES performed at the investigator's discretion proved feasible, safe, and efficacious in terms of angiographic and mid-term clinical outcomes. Concerns that drug-eluting stents may be less effective when deployed without predilation because of polymer and/or drug loss appear unfounded.
This study was sponsored by Cordis, a Johnson & Johnson Company.
- Abbreviations and acronyms
- confidence interval
- Canadian Sirolimus-Eluting Stent in Coronary Lesions trial
- direct stenting
- European Sirolimus-Eluting Stent in Coronary Lesions trial
- major adverse cardiac events
- myocardial infarction
- minimum lumen diameter
- sirolimus-eluting stent(s)
- Sirolimus-Eluting Stent in Coronary Lesions trial
- target lesion revascularization
- Received June 21, 2004.
- Revision received September 1, 2004.
- Accepted September 13, 2004.
- American College of Cardiology Foundation
- Wilson S.H.,
- Berger P.B.,
- Mathew V.,
- et al.
- Martínez-Elbal L.,
- Ruiz-Nodar J.M.,
- Zueco J.,
- et al.
- Ijsselmuiden A.J.,
- Serruys P.W.,
- Scholte A.,
- et al.
- Morice M.C.,
- Serruys P.W.,
- Sousa J.E.,
- et al.
- Schampaert E.,
- Cohen E.A.,
- Schlüter M.,
- et al.
- Mercado N.,
- Boersma E.,
- Wijns W.,
- et al.
- ↵Moses JW, Leon MB, Popma JJ, Cohen SA, Kuntz RE. DIRECT: Direct stenting using the sirolimus-eluting Bx Velocity stent: procedural, clinical, and angiographic outcomes compared to a predilatation strategy. Presented at the American College of Cardiology 53rd Annual Scientific Session, New Orleans, LA, March 7–10, 2004.