Daily News

Provisional or Double-Kissing for Bifurcation PCI: Interventionalists Review the Who, What, When, Why

World-renowned interventional cardiologists discuss strategies for left main and bifurcation PCI in E-training session

PCI for left main bifurcation lesions requires a high level of technical and strategic maneuvering – two aspects that experts broke down in concrete detail during a COMPLEX PCI 2020 virtual training session on Nov. 26.

Bifurcation PCI - a continually evolving field supplemented by ongoing research - requires technical expertise of the operator as well as strategic planning that tailors the approach to lesion complexity. The variability in choice ultimately results in heterogeneity in outcome across different operators, institutions, and even country borders.

To reduce variability in outcome, Shao-liang Chen, MD, PhD (Nanjing First Hospital, Nanjing Medical University, China) outlined the importance of correctly defining complex bifurcation lesions, choosing the better stenting technique particularly for the upfront two-stent route, and the question of how to treat the side branch (SB).

Interventional experts discuss strategies for left main and bifurcation PCI at COMPLEX PCI 2020 Virtual on Nov. 26. Top row, from left to right: Park Duk-woo (Asan Medical Center), Alan C. Yeung (Stanford School of Medicine), Sunao Nakamura (New Tokyo Hospital). Middle row: Park Seung-jeung (Asan Medical Center), Kenji Wagatsuma (Tsukuba Memorial Hospital), Shao-liang Chen (Nanjing First Hospital). Bottom row: Ahn Jung-min (Asan Medical Center), Koo Bon-kwon (Seoul National University Hospital), Teguh Santoso (Medistra Hospital, Indonesia).

Correct classification goes a long way

Accumulated data has shown a strong correlation between lesion complexity and clinical outcome as evidenced by the CACTUS, BBC-ONE, NORDIC trials that included CTO lesions, left main, and AMI, among others - emphasizing the need for correct lesion classification.

Furthermore, results from the NORDIC study indicated a strong correlation between lesion complexity and worse clinical outcomes and demonstrated that provisional stenting does not work equally across both simple and complex lesions.

Despite the need for a robust classification system, Dr. Chen pointed out the lack of a unifying and evidence-strong classification method for complex bifurcation lesions.

Current recommendations propose the use of the upfront two-stent approach where provisional stenting may not be the answer. For instance, the 2018 European Society of Cardiology (ESC) guidelines, states the two-stent approach may be preferable for complex coronary bifurcations that have an SB diameter greater than 2.75 mm, SB lesion length greater than 5 mm, and are difficult to access the SB after main vessel (MV) stenting.

Before the 2018 ESC guidelines, Chen and his team worked on defining bifurcation lesions for treatment with drug-eluting stents (DES) in the DEFINITION trial published in the Journal of the American College of Cardiology in 2014.1

“We had questions about the criteria for complex coronary bifurcation,” Chen said. “With this inquiry, we sought to address the issue of defining complex bifurcations.”

The research team built the definition criteria for differentiating simple bifurcation lesions from complex bifurcation lesions by pooling data from 1,500 patients with bifurcation lesions and then further validated the criteria by utilizing an external validation sample of another 3,660 patients. True bifurcation lesions with at least one Medina 1,1,1 or 0,1,1 coronary bifurcation lesion and an SB diameter of at least 2.5 mm were included.

The research team found eight confounding factors that correlated with one-year major adverse cardiac events (MACE) and thereupon established two major and six minor criteria for differentiating simple from complex bifurcation lesions.

Of the eight confounders, two parameters proved to be strongly correlated with MACE with the highest sensitivity (80 percent) and specificity (72~74 percent) and were consequently designated as major:

  • for distal left main bifurcation: SB diameter stenosis (DS) ≥70 percent and SB lesion length (LL) ≥10 mm
  • for non-left main bifurcation: SB diameter stenosis ≥90 percent and SB lesion length ≥10 mm

Another six parameters with p values <0.001 were classified as minor, which included moderate to severe calcification, multiple lesions, bifurcation angle <45°, main vessel reference vessel diameter (RVD) <2.5 mm, thrombus-containing lesions, MV lesion length ≥25 mm.

A combination of one major parameter and any two minor parameters indicated a complex bifurcation lesion.

This criteria were further tested in the DK CRUSH V2 study where the team found “very positive, very significant difference” between the DK crush group and the provisional stenting group, favoring DK crush in unprotected left main distal bifurcation with 1,1,1 and 0,1,1 lesions.

According to the DEFINITION criteria for patients in the complex bifurcation group, one-year target lesion failure (TLF) in the provisional group was 18.2 percent and 7.0 percent in the DK group (HR 0.68, 95% CI 0.05-0.54).

In the simple bifurcation lesion group, however, the absolute one-year TLF was 8.0 percent in the provisional group and 1.9 percent in the DK crush group, although there was no statistically significant difference between the two groups.

“The fact that the one-year TLF was nearly two percent in the DK crush group for simple lesions indicate the efficacy of these new definition criteria for differentiating simple from complex bifurcation lesions,” Chen said.

“DK Crush the better two-stent approach for complex bifurcation”

DEFINITION results served as the foundation for the DEFINITION II3 trial published in the European Heart Journal in June.

Chen and colleagues, noting that the DEFINITION criteria were not prospectively utilized in a randomized study, designed the international, multicenter, randomized DEFINITION II trial to examine the outcomes of the routine two-stent method compared with provisional stenting approaches in patients with bifurcation lesions as classified by the DEFINITION criteria.

The primary endpoint was one-year TLF and all patients were classified by complex bifurcation with 330 patients in each group. Only DK crush and culotte stenting techniques were recommended in the two-stent group.

Results showed that at one-year follow-up, TLF rate was 11.4 percent in the provisional group and 6.1 percent in the two-stent group (HR 0.52, 95% CI 0.30–0.90; P=0.019), a difference driven mainly by the lower one-year rates of target vessel myocardial infarction (TVMI) (HR 0.43, 95% CI 0.20–0.90; P=0.025) and clinically driven target lesion revascularization (TLR) (HR 0.43, 95% CI 0.19–1.00; P=0.049) in the two-stent arm and the presence of MI in the provisional arm.

“Regarding the provisional stenting technique, almost 25 percent of bifurcation lesions could be classified by complex bifurcation lesions according to the DEFINITION criteria,” Chen said. “If approaching these complex bifurcation lesions with the two-stent approach, I am very confident that DK crush is the better technique.”

“For the remaining 75 percent of simple bifurcation lesions with either Medina 1,1,0 or 1,0,1 or 1,0,0, I would recommend the provisional technique with one- or two- stents as the primary treatment,” he added.

Chen also noted the importance of the SB treatment strategy for provisional stenting with either one- or two-stents. In a recent analysis where Chen and colleagues combined four databases for insight, results showed that one-year TLF was 16.7 percent in the SB pretreatment group and 11.5 percent in the non-pretreatment group (p=0.015).

“Interventional cardiologists must decide on whether or not to handle the side branch,” he said. “And after side branch treatment, we still have to decide on a provisional technique with either one- or two-stents as well as considering whether the provisional and two-stent technique should be considered as a complication.”

Tips and questions

Ongoing research in the field will help craft a uniform and standardized technique in bifurcation PCI, but operators will have to continue to assess their level of expertise as well as the anatomical and physiological characteristics of the patient’s lesion for better outcomes.

“It would be nice to know which patients need an upstream two-stent technique, but I don’t think this is a big question for us - if we can regard ourselves as experts in bifurcation,” Bon-kwon Koo, MD, PhD (Seoul National University Hospital, Korea) said. “But for the beginners and learners, whether they start with either upstream two-stent or provisional is important since not choosing the right strategy spells big trouble in left main.” v“In any case, it’s reassuring to know that double stenting is as good as one stenting when done by experts, but this result was produced by Chinese doctors who also stated that less experienced operators in left main PCI demonstrated outcomes associated with higher mortality,” he added. “So all in all, the proper selection and targeting for two-stenting are what we need now and in the future.”

Panelists also drew from the current literature to aid in strategy selection.

“We don’t have a right answer for when we should go with two-stents and that is something that future works will have to define,” Ahn Jung-min, MD (Asan Medical Center, Korea) said. “Our previous data4 showed that FFR measurement in jailed side branch being less than 0.80 was seven percent, so we have two criteria to predict the side branch jail, namely the plaque burden and the minimal lumen area (MLA).”

“It’s important to select either the one- or two-stent technique based on the two factors of plaque burden and MLA. This will aid in roughly predicting side branch jailing after single crossover stenting,” Ahn continued. “If circumflex ostium plaque burden is more than 50 percent and MLA is less than 4 mm, then the upfront two-stent technique is preferred over simple crossover stenting. The provisional combined with the two-stent procedure should be considered for acute complications.”

“There is a difference in MLA between our and Chen’s datasets so all-in-all, the effective stent area should be maximized to the vessel size, regardless of the lesion you are dealing with” Ahn added.

Park Seung-Jung, MD, PhD (Asan Medical Center, Heart Institute Chairman, Korea) agreed by saying, “To put simply, if free of disease and with normal circumflex, then go with a single stent crossover for almost every time. But there are some disease states where the two- to three-stent strategy is preferred - namely in cases of complex true bifurcation lesions with diseased circumflex artery.”

Wrapping up the session, Alan C. Yeung, MD (Stanford School of Medicine, U.S.A) said, “I would like to add that imaging is very important – meaning we get fooled a lot. If you want to do left main bifurcation stenting, you have to know imaging. Knowing the plaque percentage is impossible without imaging. MLA also can’t be known without imaging. Double check whether everything is FFR negative, checking for whether you got the right spot, whether there isn’t anything left behind, and if everything is addressed.”


  1. https://www.sciencedirect.com/science/article/pii/S1936879814012357?via%3Dihub
  2. https://www.sciencedirect.com/science/article/pii/S1936879819314189?via%3Dihub
  3. https://academic.oup.com/eurheartj/article/41/27/2523/5862959
  4. https://www.e-kcj.org/Synapse/Data/PDFData/0054KCJ/kcj-41-304.pdf