A true bifurcated lesion in coronary atherosclerotic disease is present in approximately 15% of patients referred for percutaneous coronary intervention (PCI).1 The percutaneous approach to this especially complex scenario has been faced in many different ways depending on the techniques available at any particular time.2–4 In the last few years, the introduction of drug-eluting stents (DES) has remarkably improved the outcome in bifurcation lesions compared with using bare-metal stents (BMS), resulting in fewer adverse events and lower main branch (MB) restenosis rates.5–7 However, the most suitable approach to the side branch (SB) remains uncertain. Although the initial ‘provisional’ stenting technique (i.e. stenting of the SB after MB stenting only in cases of suboptimal or inadequate result) is probably the prevailing approach, the four stent techniques that allow the stenting of both branches (i.e. crush, V, T, culottes) are appealing. However, even if the strategy of stenting both branches when the SB stenosis is suitable for stenting is promising, data in the literature indicate that clinical outcomes are better if routine usage of the two-stent strategy is avoided.8
The introduction of dedicated stents may be supposed to be an attractive strategy for approaching the different types of bifurcation lesion. These stents are specifically designed to provide good deliverability, secured access to the SB and complete coverage of the lesion site without double/triple layers of stent struts. They also incorporate the benefits of drug elution and ensure drug availability to all diseased surfaces. The objective of the observational multicentre Nile Croco Registry was to assess the angiographic and six-month clinical results of a cohort of patients with bifurcation lesions who underwent PCI with dedicated stents.
Patients and Lesions
From May 2005 to December 2007, 103 consecutive patients were enrolled in the Nile Croco International registry at two French sites and on Spanish sites (Centre Cardiologique d’Evecquemont, Centre Cardiologique du Nord and University Hospital Vall d’Hebron). The main inclusion criteria were:
- stable, unstable angina or silent ischaemia;
- presumed new bifurcation lesion in a main vessel with a reference diameter ≥2.5mm (visually assessed); and
- an SB diameter ≥2.0mm.
Patients with two- or three-vessel disease were included if the other vessel lesions could presumably be successfully treated by PCI or were not tributary for either PCI or coronary bypass graft (CABG). In addition, patients with other lesions in the same vessel were included if these were proximal to the bifurcation lesion.
The major exclusion criteria were:
- presence of non-cardiac severe pathology or life expectancy of less than six months;
- bifurcation lesion type 0,0,1 according to the Medina Classification9 (type 4b according to Massy class);
- presence of thrombus in the target lesion;
- severely tortuous coronary vessel;
- calcifications proximal to the target lesion; and
- left main coronary lesion or left anterior descending artery (LAD) lesion involving the ostia.
The Nile Croco® intracoronary stent system consists of an 18mm balloon-expandable chromium–cobalt BMS pre-mounted on a dedicated delivery system with two independent balloons, each with a rapid-exchange lumen for the two guidewires required. The MB balloon is available in diameters of 2.5, 3.0 and 3.5mm, and the SB balloon in diameters of 2.0, 2.5 and 3.0mm. There are five devices available using different combinations: MB 2.5 SB 2.0, MB 3.0 SB 2.0, MB 3.0 SB 2.5, MB 3.5 SB 2.5 and MB 3.5 SB 3.0. The Nile Croco intracoronary stent is compatible with 6F guiding catheters (inner diameter ≥0.70 inches), and consists of a dedicated MB stent proximally crimped over the tip of the SB catheter. The stent consists of three segments: the distal segment includes six to eight cells (depending on the size), the medial segment includes eight to 10 cells and the proximal segment includes seven to nine cells. This design ensures the same metal–artery ratio along the bifurcation and avoids cell overstretching, bearing in mind that the morphology of the artery at the bifurcation sites is not cylindrical and thus should accommodate the SB portal once implanted. The delivery system is based on two separate rapid-exchange balloons: one for stent deployment in the MB and the other for opening the stent strut towards the SB. It allows final kissing-balloon inflation and a strategy of MB stenting with provisional SB stenting. The design incorporates an auto-release sheath that wraps both catheter shafts to avoid distal dislodgement of the tip of the balloons while pushing the entire system to position the stent at the lesion site.
Eligible patients underwent pre-dilation of the target lesion after wiring both branches of the bifurcation lesion. A Nile Croco intracoronary stent system of an appropriate size was selected according to the MB and SB distal reference diameters. The device was then loaded onto both wires and pushed up to the lesion (see Figures 1a and 2). It was placed where the central marker of the MB balloon matched the origin of the SB. If there was no wire wrapping, the stent became in phase with the bifurcation by means of a self-rotation, which was stopped at the level of the carina; at this point resistance was usually felt. The correct positioning was checked by angiography. If there were difficulties in reaching the desired position, the SB wire was pulled back proximal to the carina and the SB was re-wired in order to avoid or solve potential wire wrapping.
Once the stent was placed in the desired position, it was deployed by inflation of the MB balloon (see Figues 1b and 2). The long tip of the SB balloon usually came off from the mid-part of the stent to the SB. The SB balloon was then pushed to the level of the carina required to perform the kissing balloon inflation at the position(s) desired (see Figures 1c, 1d and 2). The special shape of the proximal part of the SB balloon is designed to avoid overexpansion of the proximal segment of the bifurcation while inflating the kissing balloon, thus permitting an anatomical final deployment of the stent at the carina. Thereafter, the delivery system was removed (see Figures 1e and 2). The SB may or may not be post-treated with additional balloon inflation or stent implantation, at the investigator’s discretion.
Patients received aspirin (>100mg) and a loading dose of clopidogrel according to each centre’s clinical practice. Anticoagulation therapy and the use of glycoprotein IIb/IIIa agents were left to the discretion of the investigators. After the procedure, all patients received aspirin ≥75mg/day and clopidogrel ≥75mg/day for at least one month.
End-points and Definitions
The primary end-points were stent placement success (the stent with the delivery system reaching the desired position at the carina of the bifurcation) and angiographic success of the MB and SB. Success was considered complete if MB residual stenosis was <30% and SB residual stenosis was <50%, and partial if MB residual stenosis was <30% but SB residual stenosis was >50%.
Secondary end-points were in-hospital and six-month events, including major adverse cardiovascular events (MACEs) and other in-hospital severe adverse events, such as stent occlusion or haemorrhagic complications requiring blood transfusion or surgery. MACEs were defined as cardiac death, Q-wave and non-Q-wave myocardial infarction and target lesion revascularisation (TLR). TLR was defined as MB and/or SB target site revascularisation by either PCI or CABG. All patients were followed up by a phone call after previous informed consent.
Between May 2005 and December 2007, the subset ‘Spanish–French experience’ enrolled 103 patients at three sites. Ten of the 103 patients were not eligible to enter into the registry, four of them due to left main disease, four due to life expectancy <6 months, one due to very severe calcification of the target vessel and one due to scheduled mitral valve repair in <6 months. Eighty-five of the 93 remaining patients (91%) were followed for months. In one case the device never reached the lesion and seven cases were included too recently for the follow-up (<6 months). Patient demographics in the study group are summarised in Table 1, and bifurcation and lesion types are described in Table 2 and Figure 3.
Most of the devices employed had diameters ranging between 2.5/2.0mm and 3.5/2.5mm (see Table 3). Device placement success was achieved in 96% of cases and procedural success in 98% (see Table 2). The device could not be delivered in a bifurcated lesion located at the crux cordis of a severely tortuous coronary right artery. In addition, the lesion had an important calcification component.
Clinical End-point Analysis
There was one confirmed in-hospital MACE in the 93 recruited patients (1%): this was a death due to refractory shock in an 81-year-old patient with three-vessel disease and severe mitral regurgitation who had been rejected for cardiac surgery. The MACE rate at six months in the 85 patients with complete follow-up was 12%. The clinical and/or ischaemia-driven TLR rate was 10.6% (see Table 4). One patient died due to unknown reasons during follow-up, 10 days after the stent implantation.
The first Nile Croco Registry shows a high rate of device success (96%) and favourable acute and long-term clinical results, as measured by MACE at 180 days. In addition, the procedural success rate achieved (98%) provides a very good index of safety, especially considering the particularly complex lesions that were treated with this novel and dedicated device.
A true bifurcated lesion in coronary atherosclerotic disease is present in approximately 15–25% of patients. The percutaneous approach to this particularly complex scenario has been addressed in many ways depending on the techniques available at the time. Thus, the angiographic result achieved using the initial simple balloon technique1,2 was improved with the advent of the kissing-balloon technique, which minimises the plaque shift to the SB.3,4 Thereafter, the athero-ablative techniques (rotational and directional atherectomy) reduced the TLR rate, but increased the procedural complications after the percutaneous approach.5–7 Next, the use of stents for the scaffolding of the plaque became the treatment of choice. Although the use of one or two BMS for the treatment of bifurcation lesions was initially successful, this method is now associated with high restenosis rates regardless of the technique used.8,9
The first bifurcated stents appeared in the 1990s, but their rigidity and poor profile were associated with a low rate of success.10 As a consequence, the idea of treating the MB while protecting the SB with a single dedicated stent and delivery system emerged. The first controlled study with angiographic control during follow-up was carried out with the Multi-Link Frontier coronary bifurcation stent system (Guidant Corp., Santa Clara, CA, US).11 Using the first-generation stent, this study reported a 91% device success rate and a very good safety profile (2.9% MACE at one month and 17.1% at six months; 11.4% ischaemia-driven TLR at six months). Other devices developed by various companies have also been studied, showing similar results in the initial feasibility studies.12–15
The patients included in the Nile Croco Registry were older than initially expected (mean age 65±12 years; 11% of patients were over 80 years of age). One possible explanation for this is a possible selection bias during recruitment. Thus, the patients included would likely be those who were not optimal candidates for a DES. Although we did not record all prognostic factors to accurately determine the population and arteries risk profile, the prevalence of diabetic patients (22%) as well as the mean diameter of the vessels treated indicate a low risk and thus a favourable population for angiographic success and good clinical outcomes. However, the prevalence of bifurcation lesions involving the LAD (67%) is clearly lower than that reported in previous studies (for example, in the study by Lèfevre et al. the corresponding figure was 80%11) or initial experiences. This should have led to poorer results in our series, since bifurcation lesions involving the LAD have been shown in other studies to be an independent predictive factor for angiographic and clinical success; however, this was not the case in our study.
The SB was stented in only 13% of cases despite the fact that 54% of the lesions involved the SB (on the Medina classification, 43 lesions were 1,1,1 + 1,0,1 and seven were 0,1,1). The rate of final kissing-balloon usage (97%) suggests easy SB access, which would have encouraged the operator to stent the SB if the angiographic result had been poor. However, the residual stenosis at the SB was visually estimated at below 30% in 80% of cases. This remarkably beneficial effect of the device in the SB may be related to the design of the dedicated stents. Therefore, the shape and structure of the Nile Croco may avoid some spurious angiographic visual effects that are occasionally seen with tubular single stents. For instance, a stenosis involving the SB ostia may be misleadingly interpreted as struts along the SB ostia when using tubular single stents. Another additional reason for this outcome may be the small protrusion of the struts that face the SB before the kissing balloon, which could scaffold the very proximal part of the SB after the kissing balloon.
The Nile Croco stent provided favourable acute and long-term clinical results, as measured by MACE at 180 days, compared with other series.11–13,16–18 The rate of the secondary end-point, MACE at six months, was 11.8%. Ischaemia-driven TLR was necessary in only 9.4% of cases, which is a better outcome rate than that reported in previous studies undertaken with the provisional T stenting approach.16,17,19–22
The univariate analysis performed has not shown demographic, lesion and procedure differences between those patients who presented MACE during follow-up and those who did not (see Table 5), although the small sample size prevents any firm conclusion being made.
Stent thrombosis has recently emerged as a major drawback of DES, and it appears that treatment of bifurcation lesions with a DES is associated with a higher stent thrombosis rate compared with DES treatment of non-bifurcations,23–25 and treatment of bifurcation lesions with tubular DES is not a US Food and Drug Administration (FDA)- approved indication and should therefore be considered with caution. Some dedicated devices with DES for bifurcation have been studied with preliminary results that look promising,26 but what seems obvious is that the great hope for dedicated stents that will make bifurcation easy for everybody entails looking for the best platform with which to deliver the drug. The Nile Croco stent shows promising results in the acute and mid-term that should encourage its use as a platform for dedicated DES.
The Nile Croco Registry is a study designed to examine the procedural safety and clinical feasibility of one particular dedicated stent. It does not compare the Nile Croco with other bifurcation lesion treatments and does not have the statistical power to address issues such as predictive factors for MACE during follow-up.
Bifurcation lesions have long been treated with a large variety of stenting approaches, from the simplest to the most complex. Since tubular or conventional stents are not designed to treat bifurcation lesions, these devices frequently jeopardise the SB during and after the procedure, thus increasing the risk of MACE compared with non-bifurcation lesions. Therefore, the dedicated stents may constitute, in the era of DES, a remarkable improvement in the management of these complex lesions. The design of the Nile Croco stent may both improve initial angiographic success and ensure the homogeneous delivery of drug(s).