22 Jul 2020

Multicentre, randomised comparison of two-stent and provisional stenting techniques in patients with complex coronary bifurcation lesions: the DEFINITION II trial

Selected in European Heart Journal by N. Ryan

DEFINITION-II adds to the evidence that an upfront 2-stent strategy may be most appropriate in true complex bifurcation lesions.

References

Authors

Zhang JJ, Ye F, Xu K, Kan J, Tao L, Santoso T, Munawar M, Tresukosol D, Li L, Sheiban I, Li F, Tian NL, Rodríguez AE, Paiboon C, Lavarra F, Lu S, Vichairuangthum K, Zeng H, Chen L, Zhang R, Ding S, Gao F, Jin Z, Hong L, Ma L, Wen S, Wu X, Yang S, Yin WH, Zhang J, Wang Y, Zheng Y, Zhou L, Zhou L, Zhu Y, Xu T, Wang X, Qu H, Tian Y, Lin S, Liu L, Lu Q, Li Q, Li B, Jiang Q, Han L, Gan G, Yu M, Pan D, Shang Z, Zhao Y, Liu Z, Yuan Y, Chen C, Stone GW, Han Y, Chen SL

Reference

European Heart Journal, ehaa543, https://doi.org/10.1093/eurheartj/ehaa543

Published

26 June 2020

Link

Read the abstract

Reviewer

Nicola Ryan

@NicolaRyanI1

Interventional cardiologist / Cardiologist

Aberdeen, United Kingdom

Latest contributions

TAILORED-CHIP: Tailored antiplatelet therapy for complex high-risk PCI PCRonline @ ESC Congress 2025: the interventional cardiology perspective! One- versus three-month DAPT after everolimus-eluting stent implantation in diabetic patients at high-bleeding risk: results from the XIENCE Short...

My Comment

Why this study? – the rationale/objective

DEFINITION II was a randomised control trial comparing PCI in complex bifurcation lesions using a two-stent technique versus a provisional stenting strategy.

The optimal PCI strategy in bifurcation lesions is a topic of debate with provisional stenting (PS) considered the standard approach in the majority of cases. Recently DKCRUSH-V showed lower rates of target lesion failure (TLF) at one and three years in true Left Main bifurcations using a double-kissing crush technique compared to provisional stenting.

The DEFINITION criteria were developed to identify complex bifurcations however they have yet to be assessed in a prospective fashion.

How was it executed? – the methodology

DEFINITION II was a randomised control trial in patients undergoing PCI for symptoms or proven ischaemia with a de-novo complex bifurcation stenosis (Medina 1,1,1 or 0,1,1) as defined by the DEFINITION criteria.

The DEFINITION criteria require one major and two minor criteria to assess a bifurcation lesion as complex. (Major: SB lesion length ≥10mm with ≥70% stenosis for LM or ≥90% stenosis for non-LM, Minor: moderate-severe calcification, multiple lesions, bifurcation angle <45° or > 70°, MVRD <2.5mm, thrombus containing lesion, MV lesion length ≥25mm.)

The primary endpoint was TLF (a composite of cardiac death, target vessel MI, or clinically driven TLR) at 12 months, with a safety endpoint of stent thrombosis.
DK Crush or Culotte was strongly recommended for the two-stent group, with the bailout strategy left to the operators' discretion in the provisional group.

What is the main result?

DEFINITION-II randomised 660 patients, 329 to PS and 331 two-stent techniques with 12-month follow up available for 325 and 327 respectively. In both the two-stent and PS groups, the majority of bifurcations were LAD/Diagonal followed by distal LMS (62.5% vs. 60.6% and 28.7% vs. 28.9% p=0.552, respectively). IVUS was used in slightly over a quarter of cases (27.7%).

TLF was significantly higher in the PS group compared to the two-stent group (11.4% vs. 6.1%, HR 0.52; 95% CI 0.30-0.90, p=0.019), mainly driven by TLR (5.5% vs. 2.4%, HR 0.43,95% CI 0.19-1.0, p=0.049) and TVMI (7.1% vs. 3.0%, HR 0.43, 95% CI 0.2-0.9, p=0.025)
Periprocedural MI was higher in the PS group (5.8% vs. 2.1%, HR 0.37, 95%CI 0.15-0.87, p=0.022), with the majority of TVMI occurring within the first 30 days (PS 6.5% vs two-stent 2.4%, HR 0.38, 95%CI 0.17-0.85, p=0.018).
Of note, 22.5% of the PS group required an SB stent for suboptimal results, with 46.2% of SB predilated mainly for SB compromise after MV predilation.
In the two-stent group, DK-Crush was the most common technique (77.8%), whilst a TAP (64.4%) was the most common bailout strategy in the provisional group.

Critical reading and the relevance for clinical practice

The results of this study show that a two stent technique is superior to a provisional approach in true complex bifurcations. The initial increased rates of TVMI in the PS approach appears counterintuitive however there was side-branch occlusion in 28 cases (8.6%), at some time during the procedure, in the PS group with 3 not recoverable, as well as numerically more intra-procedural complications in the PS group, which may account for some of this difference. Long term durability of the two techniques is of interest particularly as TLR appears to increase at 12 months in the provisional group. Overall rates of IVUS were low and its use for pre-procedural planning as well as stent optimisation may improve outcomes.

A note of caution when interpreting the results, one must bear in mind that all bifurcations were complex with significant stenoses in important side-branches therefore not generalisable to all bifurcations. DK Crush was the most commonly used two-stent technique however the study was not powered to demonstrate differences between two-stent techniques.

In summary, DEFINITION-II adds to the evidence that an upfront 2-stent strategy may be most appropriate in true complex bifurcation lesions. The ongoing EBC MAIN trial as well as the BBK-3 trial will further add to our knowledge with regard to the most appropriate treatment strategy for complex bifurcations.