Despite Some Good News, Dedicated Stents Fail to Surpass Standard Care for Bifurcation Lesions

PARIS, France—Two studies presented May 20 on the opening day of EuroPCR 2014 showed mixed results for stents designed specifically to treat bifurcation lesions.

For the COBRA (COmplex coronary Bifurcation lesions: RAndomized comparison of a strategy using dedicated self-expanding biolimus A9-eluting stent vs a culotte strategy using everolimus-eluting stents) trial, Christophe Dubois, MD, PhD, of Gasthuisberg University Hospitals Leuven (Leuven, Belgium), and colleagues randomized 40 patients to 1 of 2 strategies:

  • The dedicated bifurcation stent Axxess (n = 20; Biosensors International, Singapore) plus 2 Biomatrix DES (Biosensors), all of which elute biolimus A9 from an abluminal biodegradable polymer
  • Optimal culotte technique using the everolimus-eluting Xience Prime (n =20; Abbott Vascular, Abbott Park, IL)

Baseline characteristics were similar between the 2 groups apart from lesion length in the proximal main vessel, which was longer for patients randomized to culotte/Xience (10.11 ± 3.67 mm vs 7.35 ± 3.66 mm; P = .02). However, cumulative lesion length, including the proximal main vessel and distal branches, was similar in both groups. Due to study protocol, the size and number of stents were greater in the Axxess/Biomatrix group (P < .001 for both).

At 9 months, the percent of uncovered stent struts per bifurcation segment on OCT (primary endpoint) was similar for both strategies but numerically favored culotte/Xience (table 1).

Table 1. Percent Uncovered Stent Struts on OCT at 9 Months


(n = 17)

(n = 18)

P Value

Proximal Main Vessel




Bifurcation Core




Distal Main Vessel




Side Branch




However, 9-month OCT also demonstrated that the Axxess/Biomatrix strategy results in greater mean stent area compared with culotte/Xience (11 vs 7.9 mm2; P < .0001) and greater mean lumen area (10 vs 7.1 mm2; P = .0003) in the proximal main vessel. Stent and lumen areas were equivalent between the 2 groups within the distal main vessel and side branch.

In-stent late lumen loss, meanwhile, was lower in the bifurcation core with Axxess/Biomatrix (0.04 vs 0.39 mm; P = .002), with the strategy showing positive trends for that endpoint in the proximal main vessel, ‘Axxess segment,’ distal main vessel, and side branch.

Clinical outcomes at 1 year were similar between the 2 groups. No cardiac death or stent thrombosis occurred. There were 2 periprocedural MIs (without clinical consequences) in the Axxess/Biomatrix group and 1 remote MI in the culotte/Xience group, 1 TLR with culotte/Xience, and 1 TVR in each of the 2 groups (P = NS for all).

“Both strategies resulted in excellent clinical outcomes at 1 year,” Dr. Dubois concluded.

While there was no difference in the primary endpoint, culotte/Xience produced “delayed neointimal coverage in the proximal main vessel as compared with more distal segments, related to overlapping stent segments proximally,” he said. “In contrast, we observed with Axxess and Biomatrix a trend towards a higher percentage of uncovered struts in all bifurcation segments, despite avoiding overlapping stent segments with this strategy. This finding is most likely related to differences in strut thickness, polymer, and antiproliferative drug.”

The larger lumina seen with Axxess likely stem from more anatomically correct sizing as well as the conical shape and self-expanding properties of the device, Dr. Dubois added.

Panel member Helge Möllmann, MD, of Kerckhoff Heart Centre (Bad Nauheim, Germany), questioned the comparison groups, asking, “Isn’t it a little bit difficult to say if these results are mainly due to the strategies [used] or to the different stents?”

Dr. Dubois replied that the idea was to compare a fully biolimus A9 strategy with a novel device. But he acknowledged that interpretation could be difficult, particularly in terms of late lumen loss, which “does not result from differences in neointimal tissue coverage but actually from the growth of the proximal main vessel as a consequence of the [novel stent’s] ongoing self-expanding properties.”

Larger Side Branches May Gain More

During the same late-breaking session, Martin B. Leon, MD, of Columbia University Medical Center (New York, NY), shared 1-year results from the Tryton Bifurcation Trial, which randomized 704 patients with true coronary bifurcations to a 2-stent strategy using the Tryton dedicated bifurcation BMS (Tryton Medical, Durham, NC) in the side branch and DES in the main vessel (n = 355) or to the accepted strategy of DES in the main vessel plus provisional side branch stenting (n = 349).

In 9-month data, presented at the 2013 Transcatheter Cardiovascular Therapeutics symposium in San Francisco, CA, the primary noninferiority endpoint of target vessel failure (TVF; cardiac death, target vessel MI, or TVR) was not met due to an increase in “small periprocedural MIs,” Dr. Leon reported.

However, the 2-stent approach reduced in-segment percent diameter stenosis of the side branch at 9 months compared with provisional stenting (31.6% vs 38.6%; P = .002) and, in post hoc subgroup analysis of cases with side branches measuring at least 2.25 mm, decreased not only diameter stenosis but also TVF.

In the 1-year findings presented at EuroPCR, TVF and its components all were similar between the 2 groups. “All of the clinical outcomes increased slightly for both groups but no important differences were observed between the 9- and 12-month outcome data,” Dr. Leon noted.

Among patients with side branches ≥ 2.25 mm, Tryton did not hold a significant advantage over provisional stenting for TVF at 1 year but results continued to numerically favor the dedicated device. There were interactions between side branch size and 12-month rates of TVF (P = .032) and target vessel MI (P = .052). Within the side branch, Tryton was associated with lower in-segment percent diameter stenosis (30.4% vs 40.6%; P = .004) and a trend toward less binary restenosis (22.2% vs 32.1%; P = .260).

Both strategies are safe and resulted in low rates of clinically driven TVR at 9 months, Dr. Leon said. “At 1 year there were slight increases in clinical outcomes but no important changes between the Tryton 2-stent and the provisional strategy.

“A post hoc analysis of larger side branches (which were 41% of enrolled patients) indicated a possible improvement in clinical and angiographic outcomes” with Tryton, he added, noting that the findings are “hypothesis-generating only.”

Apples to Oranges?

Panel member Philip Urban, MD, of Hôpital de la Tour (Meyrin, Switzerland), asked if there were plans to develop a drug-eluting Tryton stent, “because it seems a difficult game to play to put in 2 stents, 1 of which is a bare-metal stent.”

While a drug-eluting version would be desirable, Dr. Leon agreed, “for the time being, the results have been so good in the registry studies in Europe, from the standpoint of TLR and stent thrombosis, that it was not a short-term project but thought to be a longer-term project. Yes, in the future that’s one of the things that would be done.”

He predicted, however, that such changes are not likely to substantially affect outcomes, since most of the disparity in TVF arose from periprocedural MI.

In both the Tryton and COBRA trials, dedicated stents required an extra stent per case while producing largely similar results, Dr. Urban said. “We’ve got more money, more devices, same results. Where does that leave us?”

The provisional approach is “a very powerful strategy” for treating most bifurcations, Dr. Leon acknowledged, commenting that 2-stent strategies using dedicated stents may prove valuable in more complex bifurcation lesions and larger vessels.

Note: Dr. Leon is a faculty member of the Cardiovascular Research Foundation, which owns and operates TCTMD.


1. Dubois C. Complex coronary bifurcation lesions: randomised comparison of a strategy using a dedicated self-expanding biolimus A9-eluting stent versus a culotte strategy using everolimus-eluting stents: primary results of the multicenter COBRA trial. Presented at: EuroPCR; May 20, 2014; Paris, France.

2. Leon MB. TRYTON bifurcation trial 1-year results. Presented at: EuroPCR; May 20, 2014; Paris, France.

Related Stories:

  • COBRA is an investigator-initiated trial supported by grants from Biosensors International and Abbott Vascular.
  • Dr. Dubois reports serving as a consultant to Biosensors International and Edwards Lifesciences as well as receiving grant/research support from Abbott Vascular, Biosensors International, and Boston Scientific.
  • The Tryton Bifurcation Study was sponsored by Tryton Medical.
  • Dr. Leon reports receiving institutional research support from Abbott, Boston Scientific, and Medtronic.