Author + information
- Received November 2, 2004
- Revision received March 8, 2005
- Accepted March 22, 2005
- Published online August 16, 2005.
- Thierry Lefèvre, MD⁎,⁎ (, )
- John Ormiston, MD†,
- Giulio Guagliumi, MD‡,
- Heinz-Peter Schultheiss, MD§,
- Laurent Quilliet, MD∥,
- Bernhard Reimers, MD¶,
- Philippe Brunel, MD#,
- Williams Wijns, MD**,
- H.J. Buettner, MD††,
- F. Hartmann, MD‡‡,
- Susan Veldhof, RN§§,
- Karin Miquel, PhD§§,
- Xiaolu Su, MS§§ and
- Willem J. van der Giessen, MD∥∥
- ↵⁎Reprint request and correspondence:
Dr. Thierry Lefèvre, Institut Hospitalier Jacques Cartier, Rue du Noyer Lambert, 91300 Massy, France
Objectives The goal of this study was to evaluate the safety and performance of the Multi-Link Frontier coronary bifurcation stent system (Guidant Corp., Santa Clara, California), a novel dedicated device designed for permanent side branch (SB) access, stent delivery by simultaneous kissing balloon inflation, and optimal main branch (MB) and SB ostium scaffolding.
Background The treatment of coronary bifurcation lesions remains challenging, and various approaches using stents have been proposed.
Methods The primary end point was the 180-day incidence of major adverse cardiac events (MACE) per intent-to-treat analysis. Secondary end points included device success, 30-day MACE, angiographic restenosis, and target lesion revascularization (TLR) rates at 180 days.
Results After a learning phase of two cases per center, 105 patients were prospectively included in 11 centers. The left anterior descending coronary artery/diagonal bifurcation was the target in 80% of cases. The Frontier stent was successfully implanted in 96 patients (91%), and procedural success was obtained in 93%. Two patients suffered in-hospital myocardial infarction (MI) secondary to SB occlusion, and one patient underwent elective coronary artery bypass grafting. At 30 days and 6 months, the MACE rates were 2.9% and 17.1% (no death, no subacute stent thrombosis, Q-wave MI 1.0% and 1.9%, non–Q-wave MI 1.0% and 1.9%, TLR 1.0% and 13.3%). The MB in-stent restenosis was 25.3%, in-segment 29.9%. The SB restenosis was 29.1%. The overall restenosis rate for any branch was 44.8%.
Conclusions The results of this Frontier registry demonstrate the safety and performance of this dedicated stent system for the treatment of bifurcation lesions. The device can be successfully implanted in more than 90% of all cases, with a high procedural success rate and low 30-day and 6-month MACE rates.
Percutaneous treatment of coronary bifurcation lesions remains challenging (1–19), and a variety of approaches using kissing, skirt, trousers, T stenting, and others have been published (7,12,14–18). However, coronary stenting in bifurcation lesions is still associated with a lower procedural success and higher rate of restenosis compared with nonbifurcation lesions.
The Multi-Link (ML) Frontier coronary bifurcation stent system (Guidant Corp., Santa Clara, California) is a novel dedicated device specifically designed for the treatment of bifurcation lesions. The Frontier registry was set up to investigate the safety and performance of this device for the treatment of de novo or restenotic bifurcation lesions.
Main inclusion criteria were stable, unstable angina (Braunwald classification 1 to 2 A, B), or silent ischemia in patients with de novo or restenotic (no in-stent restenosis) bifurcation lesion in a main branch (MB) vessel of a reference diameter ≥2.5 mm and ≤4.0 mm (visually assessed) that could be covered by a total stent length of ≤31 mm. Side branch (SB) diameter needed to be ≥2.0 mm. Patients with two-vessel disease were included if the nonbifurcated lesion located in a different vessel was successfully treated before the bifurcation procedure. Major exclusion criteria were the presence of thrombus in the target lesion, a bifurcation angle >75° per visual estimation, severe tortuosities or calcifications proximal to the target lesion, and left main lesions.
The ML Frontier coronary stent system consists of an 18-mm balloon-expandable 316L stainless steel stent premounted on a dedicated delivery system with two balloons and two guidewire lumens. The MB balloon diameter is 2.5, 3, 3.5, or 4.0 mm, and the SB balloon 2.0 mm for the 2.5- and 3-mm device and 2.5 for the 3.5- and 4-mm device. The ML Frontier is compatible with 7-F guides (inner diameter ≥0.78 inch). The stent is made of a series of 15 rings. The eight distal rings are made of six crests, and the ninth ring of eight crests in order to accommodate the SB portal; the proximal six rings are made of seven crests to accommodate the larger proximal MB diameter. The delivery system is based upon a dual guidewire design: an over-the-wire inner lumen for the SB balloon and a rapid exchange lumen for the MB balloon. The design incorporates a common inflation lumen for simultaneous inflation of the two balloons. In order to assist with tracking and avoid guidewire crossing, the MB balloon tip includes a pocket on the distal sleeve for joining the MB and SB balloon tips with a mandrel.
Eligible patients had predilation of the target lesion after wiring both branches of the bifurcation lesion. An ML Frontier stent of an appropriate size was selected according to the distal reference of the MB. The device was back-loaded onto the MB wire and pushed up to the lesion (Fig. 1A).After removing the joining mandrel and unjoining the balloon tips, a 300-cm wire was inserted into the SB balloon lumen and pushed into the MB. Then the device was pulled back proximal to the bifurcation. The long wire was inserted into the SB (Fig. 1B), and the previously used SB wire for predilatation was removed. Then the stent was advanced into position (Fig. 1C) by simply pushing the device. The stent becomes in phase with the bifurcation by a self rotation and is stopped at the level of the carena. A resistance is felt, and correct positioning is checked by angiography. Then the stent is deployed by simultaneous kissing inflation (Fig. 1D), and the delivery system is removed (Fig. 1E). The SB may or may not be posttreated by balloon or kissing balloon dilation, or stent implantation. Patients received aspirin (250 mg) and a loading dose of clopidogrel (300 mg) before angioplasty. A conventional bolus of 10,000 IU of unfractionated heparin was given at the start of the procedure. The activated clotting time was monitored and kept at a therapeutic level ≥250 s for the entire procedure. The use of glycoprotein IIb/IIIa agents was left to the discretion of the investigators. After the procedure, all patients received aspirin (≥75 mg/day for six months) and clopidogrel (75 mg/day for one month).
End points and definitions
The aim of the Frontier registry was to investigate the safety and feasibility of the ML Frontier stent implantation. The primary end point was major adverse cardiac events (MACE) at 180 days, defined as death, Q-wave or non–Q-wave myocardial infarction (MI), and target lesion revascularization (TLR). Non–Q-wave MI was defined as creatinine kinase ≥3 times the upper limit of normal with elevated MB fraction. Q-wave MI was defined as new pathological Q waves in two or more leads in addition to elevated enzymes as above. Target lesion revascularization was defined as MB and/or SB target site revascularization by percutaneous coronary intervention (PCI) or coronary artery bypass grafting. Target vessel revascularization (TVR) was defined as any vessel revascularization.
Secondary end points were device success; procedural success (defined as <50% residual diameter stenosis in both branches by quantitative coronary angiography [QCA] using the assigned device and without MACE at discharge); MACE at 30 days; binary restenosis; and target vessel failure at 6 months (composite of death, Q-waveor non–Q-wave MI, and TVR).
Quantitative coronary angiography was performed at baseline, after procedure, and at 180-day follow-up. Binary restenosis was defined as ≥50% diameter stenosis in the MB or in the SB using an interpolated method with validated edge detection algorithm (CAAS II Analysis System, Pie Medical BV, Maastricht, the Netherlands). In-segment analysis was defined as SB to SB for the MB proximal and distal to the bifurcation lesion and bifurcation to SB for SB analysis.
Sample size and statistical analysis
After a roll-in phase of two patients per center, 100 patients were planned to be included in the registry. Although roll-in patients complied fully with all the enrollment criteria, they were reported in a separate analysis.
The Frontier registry was powered to compare the primary end point of MACE at 180 days with previously published studies (6,7,19–21) that used a strategy of MB stenting with SB provisional T stenting. They reported a mean MACE rate at six months of 39% (upper 95% confidence limit 51%). The statistical test used for this analysis was the exact Clopper-Pearson method (22). Assuming the Frontier device had a 180-day MACE rate of 34%, 89 patients were required to be included in the study to ensure a 95% statistical power. The primary end point was reported on an intent-to-treat basis. The secondary end points are presented by a per-protocol analysis except for device and procedural success. For time-to-event variables (such as time to MI, TLR, MACE), survival curves were constructed using Kaplan-Meier estimates.
The Data and Safety Monitoring Board and the Clinical Events Committee were separately designated groups of qualified individuals not involved in the study, which monitored safety and adjudicated clinical end points respectively. The QCA analysis was performed by an independent core laboratory (Cardialysis, Rotterdam, the Netherlands). The trial sponsor was Guidant Corp.
Between May and October 2002, 25 patients were included in the roll-in phase in 13 sites, and 105 patients were enrolled in the study at 11 sites. Patient demographics in the study group are summarized in Table 1and bifurcation lesion types are described in Figure 2.
Procedural characteristics of the study patients are summarized in Table 2.Device success was 91%, and procedural success 93%. The device could not be delivered in eight cases because of vessel calcification and in one case because of wire wrap. In one patient, there was a final in-stent diameter stenosis >50%. The SB was stented in 43% of cases.
The clinical, angiographic, and procedural characteristics of the 25 lead-in cases were comparable to the registry. Device success was obtained in 100% of cases and procedural success in 96%.
The results of QCA are reported in Table 3.Baseline reference vessel diameter of the MB and SB were relatively small (2.77 ± 0.51 mm and 2.10 ± 0.67 mm, respectively). Main branch lesion length was 12 ± 6 mm. At 180 days' follow-up, the MB late lumen loss was 0.84 ± 0.55 mm, in-stent binary restenosis 25.3%, and in-segment 29.9%. The in-segment SB binary restenosis rate (Table 4)was 29.1%.
Clinical end point analysis
The in-hospital and 30-day hierarchical MACE rates were both 2.9% (Table 5).At 180 days, the MACE rate (the primary end point) was 17.1%, and the MB TLR rate was 15.2% (9.5% ischemia driven). Overall, ischemia-driven TLR (MB and/or SB) was 11.4%. Figure 3shows the Kaplan-Meier survival curves for freedom from MI, freedom from TLR, and freedom from MACE, which were 96.2%, 83.6%, and 82.7% at 190 days, respectively.
The main findings of the ML Frontier registry are the following: 1) despite the fact that this device is the first generation of a dedicated stent for treating bifurcation lesions, device success is relatively high (91%); 2) acute patency of the SB is high (98% Thrombolysis In Myocardial Infarction flow grade 3 after procedure); 3) safety is excellent with a 2.9% MACE rate at one month, no cases of subacute or late thrombosis, and no death at 180 days; 4) late loss (0.84 ± 0.55 mm) is comparable to late loss observed in nonbifurcation lesions treated with bare metal stents and is associated with an acceptable ischemia-driven TLR of 11.4%.
The era of dedicated stents
Bifurcation lesions have been treated for many years with a large variety of stenting approaches ranging from the simplest to the most complex. Because stents were not designed for treating bifurcation lesions and also because the main technical problem is the risk of SB compromise during the procedure, a new paradigm of dedicated stents appeared in the 1990s. Bifurcated stents were, however, associated with a relatively low rate of device success (23) due to their rigidity and poor profile. As a consequence, the idea of treating the MB and protecting the SB ostium of the bifurcation with a single dedicated stent and delivery system emerged. This is the first controlled study with angiographic follow-up on the use of a dedicated stent. Other devices developed by various companies are currently being investigated.
Implantation of the ML Frontier with access to the SB when required was obtained in 100% of cases in the roll-in phase and 91% in the registry. The study results were obtained in quite complex lesions (77% type B2, 29% moderate-to-heavy calcification, and 29% had an angulation >75°). It is interesting to note that, despite similar clinical and angiographic characteristics, the rate of device success in the roll-in phase was 100%, suggesting that the learning phase is relatively short with this device, but also that easier cases were probably selected.
The SB was stented in 43% of cases, and failure of SB stenting was not observed, suggesting an easy SB access through the portal. In addition, the procedural success was 93%, at least in the same range as that of prior reported observational noncontrolled studies using bare-metal stents (7,20,21).
The MB in-stent late loss of 0.84 ± 0.55 mm is comparable to late loss observed in nonbifurcation lesions and associated with an acceptable binary restenosis rate of 25.3%. The restenosis rate for both branches (44.8%) compares also favorably with the data of Yamashita et al. (7) who reported binary restenosis rates of 62% and 48% in their study comparing different techniques to treat bifurcation lesions, although the MB reference vessel size was smaller in the Frontier study (2.77 ± 0.51 mm compared to 3.09 ± 0.57 mm in the study by Yamashita et al. ).
The lower in-segment restenosis rate (not significant) observed when SB was posttreated by balloon angioplasty (19.4%) compared to 35.9% when stented and 31.3% in cases of no posttreatment shows similar trends toward better outcome when using one stent compared to two stents as observed in previous studies (7,20,21,24–27).
The ML Frontier stent provided favorable acute and long-term clinical results, as measured by MACE at hospital discharge and at 30 and 180 days when compared to the literature (Table 6)(7,20,21,28). The 30-day MACE rate of 2.9% demonstrated no additional risk of subacute MB or SB occlusions when using this device. The primary end point result, MACE at 180 days, was 17.1% (95% upper confidence limit 25.7%), which was well below the 34% (95% upper confidence limit 51%) calculated as an objective performance criterion when designing the study (6,7,19–21). Ischemia-driven TLR was 11.4%, which compares favorably with previous uncontrolled published studies using a provisional T stenting approach (5,6,19–21,25).
Limitations of the ML Frontier stents are that they are larger in profile and less flexible than conventional stents and, consequently, more difficult to deliver in tortuous or calcified arteries, and require a guiding catheter ≥7-F. A further limitation is that there are currently no dedicated devices that elute an antirestenotic drug. In a recent study (28) in which patients were randomized to receive either one or two sirolimus-eluting stents, restenosis rates were lower than historical bare metal stent controls (Table 6). It is expected that in the future, dedicated bifurcation stents such as the ML Frontier will be available coated with an antirestenotic medication and a lower profile, which should broaden their applicability.
The Frontier registry evaluated the first clinical use of a novel dedicated stent system for the treatment of coronary bifurcation lesions. The safety and efficacy of this specifically designed stent allowing continuous SB access were demonstrated with low in-hospital and 180-day MACE.
For a list of the recruitment authors, countries, and patient totals, as well as the Steering Committee, Data Safety Monitoring Board, Clinical Event Committee, and sponsors, please see the online version of this article.
This study was supported by Guidant Corporation, Santa Clara, California. Susan Veldhof, Dr. Miquel, and Xiaolu Su are employees of Guidant Corporation. Dr. Guagliumi has a consultant agreement with Guidant.
- Abbreviations and Acronyms
- major adverse cardiac events
- main branch
- myocardial infarction
- percutaneous coronary intervention
- quantitative coronary angiography
- side branch
- target lesion revascularization
- target vessel revascularization
- Received November 2, 2004.
- Revision received March 8, 2005.
- Accepted March 22, 2005.
- American College of Cardiology Foundation
- Haase J.
- Yamashita T.,
- Nishida T.,
- Adamian M.G.,
- et al.
- Fishman D.L.,
- Savage M.P.,
- Leon M.B.,
- et al.
- Iniguez A.,
- Macaya C.,
- Alfonso F.,
- Goicolea J.,
- Hernandez R.,
- Zarco P.
- Choi S.,
- Tahk S.,
- Kim H.,
- et al.
- Al Suwaidi J.,
- Berger P.B.,
- Rihal C.S.,
- et al.
- Clopper C.J.,
- Pearson E.
- Colombo A.,
- Moses J.W.,
- Morice M.C.,
- et al.