Unprotected left main coronary artery (UPLMCA) patients with lower risk scores ((1),(2),(3),4) or lesions that are ostial or midshaft ((5),6) appear to have comparative outcomes after drug-eluting stent placement to the coronary artery bypass graft (CABG) ((1),(2),3). By contrast, distal UPLMCA bifurcation lesions are technically demanding, often requiring double stenting and resulting in less favorable long-term outcomes ((3),(5),(6),(7),8). Previous studies showed that double kissing (DK) crush and Culotte stenting were effective for coronary artery bifurcation lesions ((9),(10),(11),12), but, their durability and safety have never been randomly compared. Thus, we designed this DKCRUSH (DK Crush Versus Culotte Stenting for the Treatment of Unprotected Distal Left Main Bifurcation Lesions)-III study aiming to analyze the difference in the MACE between DK crush versus Culotte stenting for UPLMCA distal bifurcation lesions.

The DKCRUSH-III study was conducted in 18 centers. The protocol was approved by the ethics committee and institutional research board, and written consent was obtained from all patients.

Inclusion criteria were as follows: age ≥18 years, Medina (13) 1,1,1 or 0,1,1 de novo UPLMCA bifurcation lesions, and chronic total occlusion (2) in any parent vessel after successful recanalization. Clinical and angiographic exclusion criteria were demonstrated in (Figure 23_gr1). Additional exclusion criteria included pregnancy, a platelet count <10 × 10⁹/l and suspected intolerance to 1 of the study drugs.

The participants were randomly assigned in a 1:1 ratio to either the DK or Culotte group.

DK crush and Culotte stenting techniques (Figure 23_gr2) have been described previously ((11),14). The candidate stents were Firebird-2 (Microport Co., Shanghai, China) and Xience V (Abbott Vascular, California).

A 300-mg loading dose of clopidogrel was administered before the index procedure. After the intervention, all patients received 100 mg/day aspirin for life and clopidogrel (75 mg/day) for at least 12 months.

Clinical follow-up was performed with office visits or telephone contact at 1, 6, and 12 months. Follow-up coronary angiography was scheduled at 8 months after the index procedure unless clinical reasons indicated earlier.

Serial quantitative coronary analysis was analyzed according to our previous methods (14).

The primary endpoint was the occurrence of 1-year major adverse cardiac event (MACE) rate, including myocardial infarction, cardiac death, and/or target vessel revascularization (TVR). Secondary endpoint was in-stent restenosis (ISR). Stent thrombosis (ST) served as a safety endpoint. ST, myocardial infarction, cardiac death, target lesion revascularization (TLR), and TVR were defined according to the Academic Research Consortium definition (15).

We hypothesized that the rate of a 1-year MACE rate would be 5% in the DK crush and 15% in the Culotte groups. Accordingly, a total sample size of 358 was needed to detect a power of 0.8 (Type II error = 0.20, α = .05, 2-tailed). Because of the considerable uncertainty, the enrollment was extended to 420 patients (15% increment). The treatment-group differences were evaluated with a student t test or Wilcoxon rank sum scores for continuous variables when appropriate. The chi-square test or the Fisher exact test was used to analyze categorical variables. Rate-free survival from events were generated by Kaplan-Meier analysis, and they were compared using the log-rank test. Four pre-specified subgroup analyses by Forest plot were planned to be performed. Statistical significance was taken as a p value <0.05. All analyses were performed with the statistical program SPSS version 16.0 (SPSS Institute Inc., Chicago, Illinois).

Baseline clinical (Table 1), lesions (Table 2), and procedural (Table 3) characteristics were well matched between 2 treatment groups. All continuous variables were normally distributed.

By 12-month follow-up, there were 34 (16.3%) composite MACEs in the Culotte group and 13 (6.2%) in the DK group (p = 0.001), mainly because of significantly increased TLR (6.7%) and TVR (11%) in the Culotte group (2.4%, p = 0.037) compared with the DK group (4.3%, p = 0.016) ((Table 4), Figs. (Figure 23_gr3), (Figure 23_gr4), and Figure 23_gr5). Among patients with distal bifurcation angle ≥70°, NERS (New Risk Stratification) score ≥20 and SYNTAX score ≥23, the incidence of the composite MACE in the DK group was significantly lower than that in the Culotte group (Figure 23_gr6).

By 12 months, the incidence of definite ST was 1.0% (n = 2) in the Culotte group and 0% in the DK group (p = 0.623).

There were 12 (6.8%) ISR at side branch (SB) in the DK group and 22 (12.6%) in the Culotte group (p = 0.037), mostly seen at ostial SB (Table 5).

The major finding was that Culotte stenting was associated with significantly increased 1-year composite MACE rate, mainly because of the increment of TVR rate.

For UPLMCA bifurcation lesions, single-stent strategy was superior to systematic double-stent (including V-stenting, kissing stenting) techniques ((3),(4),(5),6). Actually, Culotte stenting is a reverse T stenting, by which intraprocedural acute closure of the main vessel after stenting SB is unavoidable ((14),16), this rate was 1% in the present study, which might be catastrophic for distal left main, as we showed that 1 patient died soon after LAD closure. In the ISAR-LEFT MAIN study (12), 98% of 384 patients with distal UPLMCA lesions were treated by Culotte stenting, the 1-year MACE (15.8%) and ISR (19.4%) in the cypher group were compared well with the current study (14.4% vs. 12.6%, respectively). Except for the limitation of the cypher stent design (close cell) for bifurcation lesions (17), we might be of great courage to postulate that the ISR rate would become significantly higher if distal bifurcation subgroup was analyzed in that study. Our finding, restenotic lesions are most localized in the SB, is consistent with the previous results ((11),12). Thus, we have to say that DK crush is superior to Culotte stenting when bearing in mind that patients in the ISAR-LEFT MAIN study (12) were less at risk compared with the current study. Bench test (17) and intravascular ultrasound findings ((18),19) reported that a “napkin” or a gap or a metallic ridge is usually seen at the ostial SB after Culotte stenting, leading to the failure to fully cover the ostial SB and resulting in increased ISR and TLR. In contrast, DK crush introduced 2 times of kissing balloon inflation, allowed the full coverage of ostial SB, resulting in less ISR ((4),14).

The sustained durability of DK crush was demonstrated in patients at intermediate- and high-risk stratified by either NERS or SYNTAX score. Furthermore, clinical efficacy of DK crush is also maintained in the patients with distal bifurcation angle ≥70°. The controversy of whether there is a correlation of bifurcation angle with worse outcome has existed for a long time (20). We postulate that the abnormal hemodynamic change of shear stress induced by Culotte stenting might play a central role in the occurrence of ISR.

For bifurcation lesions, a complex stenting approach has been an independent factor for ST ((8),16), although there is no convinced data after stenting distal UPLMCA bifurcation lesions. Similar to ST rate (<1%) by the ISAR-LEFT MAIN trial (12), ST rate (1.0%) after Culotte stenting in the present study was slightly lower than the 1.9% reported by Erglis et al. (9) and 1.6% by Adriaenssens et al. (11). It might be plausible that both DK crush and culotte stenting techniques are safe for distal UPLMCA bifurcation lesions, and that optimizing the expansion of the SB stent, as did the DK crush technique, would have put patients at lower risk of ST.

Some kind of angle restriction should have been applied in the design of the study. We did not include a CABG group to contrast with the stenting techniques. However, the promising results achieved by the DK crush technique were comparable with those after CABG. Finally, the results were achieved in very high-volume operators performing these procedures. It remains unclear whether lower volume centers could reproduce these results.

Compared to the DK crush technique, Culotte stenting is associated with significantly increased MACEs in patients with UPLMCA bifurcation lesions.