Keywords
Coronary Bifurcation Lesions, Crush Stenting, Culotte Stenting, Target Lesion Revascularization, Target Vessel Revascularization
-
Views
-
Downloads
How to cite this article
Komalasari I, Adji AS, Billah A et al. Long-Term Outcomes of Crush versus Culotte Stenting in Coronary Bifurcation Lesions: A Systematic Review and Network Meta-Analysis [version 1; peer review: awaiting peer review]. F1000Research 2025, 14:823 (https://doi.org/10.12688/f1000research.168528.1)
NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.
Export Citation
Sciwheel
EndNote
Ref. Manager
Bibtex
ProCite
Sente
▬
✚
Systematic Review
Long-Term Outcomes of Crush versus Culotte Stenting in Coronary Bifurcation Lesions: A Systematic Review and Network Meta-Analysis
[version 1; peer review: awaiting peer review]
PUBLISHED 26 Aug 2025
OPEN PEER REVIEW
REVIEWER STATUS
Abstract
Corresponding author:
Yudi Her Oktaviano
Competing interests:
No competing interests were disclosed.
Grant information:
The author(s) declared that no grants were involved in supporting this work.
Copyright:
© 2025 Komalasari I et al.
This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The author(s) is/are employees of the US Government and therefore domestic copyright protection in USA does not apply to this work. The work may be protected under the copyright laws of other jurisdictions when used in those jurisdictions.
How to cite: Komalasari I, Adji AS, Billah A et al. Long-Term Outcomes of Crush versus Culotte Stenting in Coronary Bifurcation Lesions: A Systematic Review and Network Meta-Analysis [version 1; peer review: awaiting peer review]. F1000Research 2025, 14:823 (https://doi.org/10.12688/f1000research.168528.1)
First published: 26 Aug 2025, 14:823 (https://doi.org/10.12688/f1000research.168528.1)
Latest published: 26 Aug 2025, 14:823 (https://doi.org/10.12688/f1000research.168528.1)
Introduction
Coronary artery disease (CAD) remains the leading cause of global mortality, accounting for approximately 17.9 million deaths annually, or about 31% of all deaths worldwide. One of the major challenges in percutaneous coronary intervention (PCI) is the treatment of coronary bifurcation lesions (CBL), which are found in about 15–20% of all PCI procedures.1 Bifurcation lesions, especially those involving the left main coronary artery, are associated with higher risks of restenosis, stent thrombosis, and major adverse cardiac events (MACE). In recent decades, dual-stent strategies such as the Crush and Culotte techniques have been developed to address the anatomical complexity of these lesions.2,3
Among the various dual-stenting strategies, Crush and Culotte techniques are the most frequently adopted. The Crush technique has evolved into several refined variants to improve outcomes and procedural success. These include Typical Crush, where the side branch stent is crushed by the main vessel stent without additional ballooning; Mini Crush, which reduces the amount of crushed strut by minimizing protrusion (<1–2 mm); Modified Crush, which integrates high-pressure balloon post-dilation to improve stent apposition; and Nano Crush, which utilizes ultra-thin struts and minimal crush to optimize flow dynamics and reduce metal burden in the carina.4 Meta-analysis data suggest that Double Kissing (DK) Crush is particularly effective, demonstrating a 60% relative risk reduction in major adverse cardiac events (MACE) compared to Culotte and other techniques.5 In the DKCRUSH-III trial, DK Crush reduced 3-year MACE rates to 8.2% compared to 23.7% in the Culotte group, driven mainly by lower myocardial infarction (3.4% vs. 8.2%) and target vessel revascularization (5.8% vs. 18.8%).6 In a real-world observational registry, DK Crush was also associated with a 72% reduction in MACE compared to Culotte during a 43-month follow-up, mainly due to significantly reduced target vessel revascularization (14.3% vs. 29.3%).7 Furthermore, Nano Crush has been shown to significantly reduce contrast usage, procedural time, and fluoroscopy exposure while maintaining superior clinical outcomes, including a three-year MACE rate of 6.2% compared to 24.2% with Culotte.4 A broader network meta-analysis covering over 10,000 patient-years identified DK Crush as having the highest probability (98%) of being the most effective two-stent strategy in reducing stent-related clinical outcomes.8
There is an urgent need for robust evidence to identify the optimal stenting technique for CBL, as MACE rates at 3 to 5 years post-intervention remain high—reaching up to 28% in some dual-stent approaches.6 For instance, in the DKCRUSH-III study, DK Crush was associated with an 8.2% MACE rate compared to 23.7% with Culotte stenting at 3-year follow-up.6 Additionally, a network meta-analysis found that DK Crush ranked highest in effectiveness for reducing MACE, myocardial infarction, stent thrombosis, and revascularization outcomes.5
Given this background, the present study aims to systematically evaluate and compare the one-, three-, and five-year clinical outcomes of the Crush versus Culotte techniques in patients with CBL. By applying a network meta-analysis of randomized controlled trials, this study seeks to provide comprehensive evidence regarding the long-term safety and effectiveness of both strategies, ultimately offering data-driven guidance for future clinical practice.
Methods
This systematic review was conducted in accordance with the PRISMA guidelines.9 It has been registered in the International Prospective Register of Systematic Reviews (PROSPERO) under the identification number CRD420251064878.
Eligibility criteria
For this meta-analysis, we strictly followed predefined inclusion and exclusion criteria to identify relevant randomized controlled trials (RCTs) or observational studies evaluating stenting strategies in coronary bifurcation lesions (CBL). Patients, Interventions, Comparator, and Outcomes (PICO) were presented in Table 1. Inclusion criteria required studies that: (1) involved adult patients (aged ≥18 years) diagnosed with CBL, specifically classified using Medina 1,1,1 or 0,1,1; (2) compared Crush Stenting and Culotte Stenting techniques; (3) reported at least one relevant clinical outcome, including major adverse cardiovascular events (MACE) such as cardiac death (CD), myocardial infarction (MI), target lesion revascularization (TLR), coronary artery bypass grafting (CABG), and target vessel revascularization (TVR), as well as stent thrombosis (ST); (4) had a follow-up duration of at least 12 months and/or 36 months, with additional data up to 60 months considered when available; (5) were published in English. Conversely, we excluded studies that (1) involved non-adult populations or did not classify lesions using the specified Medina types; (2) did not compare both stenting techniques; (3) lacked adequate clinical outcome data or follow-up; and (4) case reports, reviews, editorials, or conference abstracts.
Table 1. PICOs framework.
Search strategy and selection of studies
A comprehensive search was conducted across several electronic databases to identify relevant studies up until June 2025. The databases included PubMed, ScienceDirect, Google Scholar, MEDLINE, Cochrane Library, Embase, and Europe PMC. The search strategy combined key terms related to coronary bifurcation stenting techniques and clinical outcomes, such as “Coronary Bifurcation Lesion,” “CBL,” “Crush Stenting,” “Culotte Stenting,” “Bifurcation PCI,” “Nano Crush,” “Mini Crush,” “Typical Crush,” “Modified Crush,” and “DK Crush.” Outcome-related terms, including “Major Adverse Cardiovascular Events,” “MACE,” “Cardiac Death,” “Myocardial Infarction,” “Target Lesion Revascularization,” “Target Vessel Revascularization,” and “Stent Thrombosis,” were also used. Boolean operators “AND” and “OR” were applied to combine and refine search terms for comprehensive coverage. In addition, the reference lists of all eligible articles were manually reviewed to identify further studies relevant to the meta-analysis.
Data extraction
Designated investigators (I.K., A.S.A., A.B., D.S., A.D.S., N.N.D., and Y.H.O.) meticulously extracted relevant data using a predefined data extraction form after identifying the eligible studies. The extracted data encompassed detailed study characteristics including the author names, publication year, study design, study location (country), study period, and population demographics. Specifically, we recorded the type of stent used (Crush or Culotte), mean age ± standard deviation (S.D.), number and percentage of male participants, bifurcation classification (Medina 1.1.1 and 0.1.1), and the number of patients with key cardiovascular risk factors such as diabetes mellitus (DM), hypertension (HT), smoking status, and hyperlipidemia. Information related to dual antiplatelet therapy (DAPT) duration, final kissing balloon dilation (FKBD) percentages, and stent characteristics such as mean vessel length (MVL), side branch length (SBL), mean vessel diameter (MVD), and side branch diameter (SBD) was also systematically recorded. To ensure data quality and consistency, a rigorous cross-checking process was conducted in which an independent investigator validated the extracted information, thereby ensuring accuracy and completeness for subsequent meta-analysis.
Quality assessment
The second edition of the Cochrane Risk-of-Bias tool for randomized controlled trials (RoB 2) was utilized to assess the potential bias in the included RCT studies. The RoB 2 tool evaluates studies across five key domains, allowing for a thorough analysis of risk of bias. Additionally, for non-randomized controlled trials (N-RCTs), the ROBINS-I tool was employed to assess bias across seven domains.10,11 A total of 13 studies were included in the analysis, with 8 of them being non-RCTs assessed using the ROBINS-I tool and 5 RCTs evaluated using RoB 2. The quality assessment was performed by three researchers (A.S.A., A.B., and N.N.D). Any disagreements encountered during the evaluation process were resolved through collaboration and consensus among the researchers.
Outcome measure
This analysis focused on clinically relevant outcome measures to compare the effectiveness of Crush versus Culotte stenting techniques in the treatment of coronary bifurcation lesions over one- and three-year follow-up periods. The primary clinical endpoints assessed were myocardial infarction (MI), stent thrombosis (ST), cardiac death (CD), coronary artery bypass grafting (CABG), target lesion revascularization (TLR), and target vessel revascularization (TVR). Effect estimates were reported as risk ratios (RR) with 95% confidence intervals (CI), using a random-effects model to account for potential variation across studies. To determine the relative performance of each technique, Surface Under the Cumulative Ranking (SUCRA) values were calculated, providing a probabilistic ranking of treatment effectiveness. The analysis also included various subtypes of the Crush technique: DK Crush (Double Kissing Crush), which incorporates two kissing balloon inflations for optimal expansion; Mini Crush, which reduces metal overlap with shorter side branch protrusion; Typical Crush, the conventional form with full side branch coverage; Nano Crush, which minimizes protrusion to reduce restenosis; and Modified Crush, a procedural adaptation based on clinical need. Visual tools such as forest plots, network graphs, and league tables were used to compare relative effectiveness among these techniques over the defined time points.
Data synthesis and statistical analysis
Each outcome measure in this study was analyzed using pooled risk ratios (RR) and 95% confidence intervals (CI), calculated through a random-effects meta-analysis model. The primary clinical outcomes included myocardial infarction (MI), stent thrombosis (ST), cardiac death (CD), coronary artery bypass grafting (CABG), target lesion revascularization (TLR), and target vessel revascularization (TVR), assessed at one-year and three-year follow-up periods. The Surface Under the Cumulative Ranking (SUCRA) values were used to rank the effectiveness of each stenting strategy. Subgroup analysis was conducted to compare different types of Crush techniques, including DK Crush, Mini Crush, Typical Crush, Nano Crush, and Modified Crush, in relation to Culotte stenting. The I2 statistic was used to assess heterogeneity among studies. All statistical analyses were performed using R-Studio.12
Results
Study selection process and quality assessment
After conducting a comprehensive literature search across seven major databases—ScienceDirect, PubMed, Google Scholar, Europe PMC, Embase, MEDLINE, and Cochrane—a total of 7,358 records were identified. An additional 6 records were identified through citation searching, bringing the total to 7,364. Prior to screening, 24,865 records were removed due to duplication, ineligibility by automation tools, or other reasons. This left 4,308 records to be screened. Of these, 4,083 were excluded for reasons such as non-English language (44), incompatible designs (213), wrong publication type (180), irrelevant populations (450), unrelated stenting techniques (779), missing relevant clinical outcomes (947), or unrelated topics (787). A total of 243 reports were sought for retrieval, of which 43 were not retrieved due to abstract-only availability. Ultimately, 200 reports were assessed for eligibility. Among them, 193 were excluded for reasons including lack of bifurcation lesions (65), not involving both Crush and Culotte techniques (32), not reporting key outcomes (74), or irrelevance to coronary bifurcation PCI (22). From both database and citation sources, 13 studies were included in the meta-analysis, with 7 included in the final systematic review. The full selection process is summarized in the PRISMA flowchart (Figure 1).
Figure 1. PRISMA flow diagram of the study selection process.
Study characteristics
Table 2 presents data from a total of 13 studies involving 3,777 patients with coronary bifurcation lesions. Among these, 8 studies are non-randomized controlled trials (non-RCTs) and 5 studies are randomized controlled trials (RCTs). The mean age of the participants is 65.93 ± 4.67 years. These studies were conducted across various regions, including 4 studies from China,6,13,14 1 study from the United States,7 1 study from Canada,15 3 studies from Italy,4,16,17 1 study from Japan,18 1 study from Latvia,19 1 study from Finland,20 and 1 study from Pakistan.21 The follow-up durations in these studies range from 12 months to 60 months, with most studies assessing clinical outcomes at 12 months. The patient populations are diverse, with some studies focusing on specific stenting techniques such as 1st-2nd DES (Drug-Eluting Stent), while others evaluate 3rd DES. The study characteristics show that 57.74% of patients have diabetes mellitus (DM), 57.33% have hypertension (HT), 48.15% are smokers, and 58.47% have hyperlipidemia. The mean number of patients per study is 310.63 (90.46%). Stent parameters include a mean vessel length (MVL) of 33.48 ± 14.01 mm for group 1 and 35.74 ± 15.99 mm for group 2, stent balloon length (SB) of 25.90 ± 13.83 mm and 26.72 ± 11.86 mm, mean vessel diameter (MVD) of 3.40 ± 0.34 mm and 3.34 ± 0.40 mm, and stent balloon diameter (SBD) of 3.04 ± 0.41 mm and 3.03 ± 0.41 mm.
Table 2. Data characteristics.
Risk of bias
Figure 2 shows the risk of bias assessments for a total of 13 studies, divided into 8 non-randomized controlled trials (N-RCTs) and 5 randomized controlled trials (RCTs). Panel A presents the risk of bias for the 8 non-RCT studies, assessed using the ROBINS-I tool, across seven domains, such as bias from confounding, participant selection, and classification of interventions. The majority of these studies show a moderate risk of bias (yellow), with about 62.5% of the non-RCT studies marked with moderate bias, and 37.5% showing low risk (green). The overall risk of bias for these non-RCT studies is summarized in the bar graph, with 62.5% of studies showing a moderate risk and 37.5% showing low risk. Panel B shows the 5 RCTs evaluated using the RoB 2 tool, which assesses five domains such as randomization and selection of the reported result. Most of the RCTs show a low risk of bias (green), with 80% of the RCT studies having a low risk of bias across the domains. However, there are some concerns in randomization, specifically in the study by Zheng 2016, which is marked with blue, indicating “no information.” The overall risk of bias for the RCTs is summarized in the bar graph, with 80% of studies showing low risk, and 20% showing some concerns.
Figure 2. Risk of bias summary and graph.
Network meta-analysis of coronary bifurcation stenting techniques: Crush versus Culotte
Table 3 offers a concise evaluation of the angiographic and clinical endpoints of DES compared to DCB in treating DES-ISR.
Table 3. Summary of results.
| End point | Crush vs Culotte in coronary bifurcation lesion | ||||
|---|---|---|---|---|---|
| Crush group | Culotte group | RR (95% CI) | |||
| One-Year | Three-Year | Five-Year | |||
| Clinical Outcomes | (n/total studies) | (n/total studies) | |||
| CD | (2015/12 studies) | (1590/12 studies) | 1.63 [0.51; 5.24], p=0.9440 | 0.80 [0.50; 1.27], p=0.8483 | 0.54 [0.17; 1.73], p=0.7563 |
| CABG | (841/4 studies) | (530/4 studies) | 4.98 [0.24; 103.03] | 2.25 [0.46; 11.03], p=0.9013 | 0.74 [0.29; 1.88] |
| MI | (1849/11 studies) | (1443/11 studies) | 0.93 [0.52; 1.67], p=0.3253 | 0.63 [0.39; 1.02], p=0.3994 | 0.39 [0.03; 5.12], p=0.1044 |
| TLR | (1376/9 studies) | (1250/9 studies) | 1.14 [0.30; 4.37], p=0.0225* | 0.68 [0.38; 1.21], p=0.0284* | 0.89 [0.56; 1.40] |
| TVR | (1682/9 studies) | (1352/9 studies) | 2.22 [0.40; 12.38], p=0.0014* | 0.62 [0.36; 1.10], p=0.0105* | 0.70 [0.37; 1.32] |
| ST | (1986/11 studies) | (1645/11 studies) | 0.81 [0.37; 1.78], p=0.6363 | 0.81 [0.51; 1.29], p=0.1860 | 0.33 [0.07; 1.52], p=0.2220 |
Crush vs Cullote in TVR
TVR outcomes were reported in 9 studies involving a total of 3002 patients. DK Crush significantly reduced the risk of target vessel revascularization compared to Culotte stenting (RR 0.39; 95% CI [0.27–0.57]), with a SUCRA value of 0.99, indicating the highest probability of being the most effective strategy. In contrast, Modified Crush showed a substantially higher risk of TVR (RR 14.28; 95% CI [1.91–106.68]) compared to Culotte, and ranked the lowest with a SUCRA of 0.00. Mini Crush (RR 0.77; 95% CI [0.43–1.39]) and Typical Crush (RR 0.89; 95% CI [0.61–1.29]) did not show statistically significant differences. The overall heterogeneity was low (τ2=0.0065; τ2=0.0809; I2=4.5%; 95% CI [0.0–75.8%]). Treatment rankings based on SUCRA values showed DK Crush as the most effective strategy in reducing TVR, while Modified Crush ranked the lowest. A net graph, forest plot, and league table comparing the effectiveness of Crush versus Culotte stenting in TVR outcomes are presented in Figure 3.
Figure 3. Net graph, forest plot and league table for target vascular revascularization.
Crush vs Cullote in TLR
In Figure 4, a total of 2,626 participants from 9 studies were analyzed. In the Forest Plot, DK Crush significantly reduces the risk of Target Lesion Revascularization (TLR) (RR: 0.30, 95% CI: 0.16-.055). DK Crush demonstrates a high SUCRA (Surface Under the Cumulative Ranking) score of 0.81, indicating it is the most effective treatment. Nano Crush also indicates a potentially effective treatment with a lower RR value (RR: 0.11, 95% CI: 0.01-2.04) and a strong SUCRA score of 0.89, but its broad confidence interval suggests there is some uncertainty. Mini Crush (RR: 0.93, 95% CI: 0.69–1.24) and Typical Crush (RR: 1.34, 95% CI: 0.76–2.35) showed no significant differences compared to Culotte, with lower SUCRA scores of 0.40 and 0.09. Overall, DK Crush is the most successful technique for lowering TLR, as evidenced by consistent performance across all comparisons and a high ranking probability.
Figure 4. Net graph, forest plot and league table for target lesion revascularization.
Crush vs Cullote in ST
A total of 3,353 patients from 10 studies of crush and cullote in ST were reported in Figure 5. DK Crush showed the lowest estimated risk (RR: 0.23, 95% CI: 0.05–1.08), indicating a possible reduction in ST, although its confidence interval exceeded 1, indicating statistical insignificance. Mini Crush (RR: 0.66, 95% CI: 0.08–5.34) and Modified Crush (RR: 0.50, 95% CI: 0.12–2.07) also tended to have a lower risk but with wider confidence intervals, indicating imprecision. Typical Crush showed the highest estimated RR value among the alternatives (RR: 0.86, 95% CI: 0.50–1.48) and showed a low SUCRA value (0.34), implying a lower position in reducing ST. The SUCRA scores place DK Crush (0.86) as the leading strategy, followed by Modified Crush (0.59) and Mini Crush (0.48). However, the wide confidence intervals between these comparisons warrant caution due to the potential imprecision involved. In summary, DK Crush appears to be quite effective in minimizing ST risk, ranking highest among the techniques evaluated.
Figure 5. Net graph, forest plot and league table for stent thrombosis.
Crush vs Cullote in MI
In Figure 6, MI outcomes were reported in total of 3,292 participants from 11 studies. All four alternative approaches—Mini Crush, Nano Crush, DK Crush, and Modified Crush—demonstrated lower risk estimates for MI compared to Culotte, although the confidence intervals for most comparisons were wide and crossed 1, suggesting a lack of statistical significance. Notably, DK Crush showed a better risk profile (RR: 0.50, 95% CI: 0.28–0.88) with the highest SUCRA value (0.60) among the techniques, indicating a relatively strong ranking. Mini Crush and Nano Crush followed with SUCRA values of 0.78 and 0.74, respectively, while Typical Crush had the lowest ranking (RR: 1.29, 95% CI: 0.75–2.23; SUCRA: 0.09). Overall, DK Crush appeared to offer the most consistent benefit in reducing MI risk, as supported by both ranking and statistical metrics.
Figure 6. Net graph, forest plot and league table for myocardial infraction.
Crush vs Cullote in CD
CD outcomes were reported across six bifurcation stenting strategies in a network of studies, as visualized in the forest plot, network graph, and league table (Figure 7). The analysis included a total of 3,605 patients in 12 studies. Four alternative strategies; Nano Crush, Mini Crush, DK Crush, and Typical Crush showed lower RR for CD compared to Culotte, though none reached statistical significance, as all confidence intervals crossed 1. Notably, Nano Crush demonstrated the lowest RR (0.26, 95% CI: 0.03–2.18), followed by Mini Crush (RR: 0.59, 95% CI: 0.17–2.07) and DK Crush (RR: 0.84, 95% CI: 0.48–1.48). Modified Crush, in contrast, appeared to have an elevated risk compared to Culotte (RR: 2.52, 95% CI: 0.10–61.43), although the wide interval suggests substantial uncertainty. Ranking probability based on SUCRA values supported these findings, with Nano Crush achieving the highest SUCRA (0.86), suggesting the most favorable profile, followed by Mini Crush (0.66) and DK Crush (0.52). Typical Crush (0.37) and Modified Crush (0.23) ranked lower, with Culotte ranking the lowest in terms of efficacy. Overall, Nano Crush emerged as the top-ranked strategy for minimizing cardiac death, albeit with imprecise estimates due to limited sample size.
Figure 7. Net graph, forest plot and league table for cardiac death.
Crush vs Cullote in CABG
The CABG analysis included 1,371 patients from 4 studies. Typical Crush was associated with a lower risk of CABG compared to Culotte (RR: 0.74, 95% CI: 0.29–1.88), whereas DK Crush showed a higher risk (RR: 2.65, 95% CI: 0.65–10.82). However, both comparisons were statistically inconclusive due to wide confidence intervals that crossed the line of no effect. The SUCRA values shows Typical Crush ranked highest (SUCRA: 0.82), suggesting it had the most favorable profile in minimizing CABG risk. Culotte and DK Crush ranked lower, with DK Crush showing the least favorable outcome (SUCRA: 0.09). The network graph illustrates the available direct comparisons, with the most frequent being between Culotte and DK Crush (3 studies). DK Crush had a lower estimated risk than Typical Crush (RR: 0.38, 95% CI: 0.09–1.54), though the result was not statistically significant. Typical Crush also had a modestly better profile than Culotte (RR: 1.36, 95% CI: 0.53–3.47), supporting its top SUCRA ranking. In summary, Typical Crush appeared to be the most promising strategy in reducing CABG rates, although the wide uncertainty intervals suggest that further evidence is needed for definitive conclusions (Figure 8).
Figure 8. Net graph, forest plot and league table for coronary artery bypass graft.
Meta-analysis comparing Crush versus Culotte techniques in coronary bifurcation stenting
TVR and TLR in Crush vs Cullote
In Figure 9, we presented a comparative analysis of the Crush and Culotte stenting techniques for TVR and TLR outcomes. A total of Eight studies including 3,034 patients were evaluated for TVR at one- and three-year follow-up endpoints. For overall TVR in Figure 9A, pooled analysis showed significant different between the two treatments (RR 2.22, 95% CI: 0.48–1.33, I2 = 71.1%, p-value = 0.0005). In one- and three-years follow-up, we did not find significant difference between Crush and Culotte. In Figure 9B, comparison of the two techniques in terms of TLR also evaluated in 8 studies encompassing 2,626 patients. This analysis showed statistically significant difference (RR 0.79, 95% CI: 0.52–1.20, I2 = 57.6%, p-value = 0.00156). Consistent with the findings for overall TVR, no significant difference in TLR was observed at either the one-year or three-year follow-up. High heterogeneity was found in both TVR and TLR outcomes.
Figure 9. Forest plot of Crush vs Cullote in (A) TVR, and (B) TLR.
ST and MI in Crush vs Cullote
Figure 10 presented comparative analysis of the Crush and Culotte technique for the outcomes of ST and MI at one-, three-, and five-year follow-up periods. In Figure 10A, a total 11 studies comprising 3,631 patients were included in the analysis of ST. The pooled RR indicated no significant different between two techniques (RR 0.76, 95% CI: 0.52–1.12, I2 = 0%, p-value = 0.4416), with consistently non-significant results across all follow-up intervals. Figure 10B represented MI outcomes from 11 studies consisting of 3,292 patients. Pooled analysis showed that no significant result between two treatments (RR 0.71, 95% CI: 0.50–1.00, I2 = 8.2%, p-value = 0.1004).
Figure 10. Forest plot of Crush vs Cullote in (A) ST, and (B) MI.
CD and CABG in Crush vs Cullote
In Figure 11A, 12 studies involving 3,605 patients were analyzed for cardiac death (CD). The pooled RR was 0.84 (95% CI: 0.56–1.25, I2 = 0%, p-value = 0.9440), indicating that Crush and Culotte has similar risk, with low heterogeneity. In Figure 11B, CABG outcome was evaluated based on 4 studies including 1,371 patients. The meta-analysis found no significant result between the two stenting strategies (RR 1.24, 95% CI: 0.58–2.62, I2 = 0%, p-value = 0.9013). At one-, three-, and five-year follow-up, no significant different was observed between Crush and Culotte treatment for both CD and CABG outcomes.
Figure 11. Forest plot of Crush vs Cullote in (A) CD, and (B) CABG.
Discussion
This study aims to evaluate the effectiveness of the Crush stenting technique compared to the Culotte technique in patients with coronary bifurcation lesions. By including 3,777 patients from various studies, our results show that the Double Kissing (DK) Crush technique provides the best outcomes in reducing the risk of target vessel revascularization (TVR), target lesion revascularization (TLR), and stent thrombosis (ST). This technique has a high SUCRA value, reflecting its strong likelihood of being the best strategy. In contrast, Modified Crush showed poor outcomes with the highest TVR risk and the lowest SUCRA. The Mini Crush and Typical Crush techniques did not show significant superiority over Culotte. Although there is variation in the effectiveness of the different techniques, the heterogeneity of the data in our study is low, strengthening the findings. Overall, DK Crush emerges as the most consistent and promising technique for managing complex coronary bifurcation lesions.
In this study, DK Crush significantly reduced the risk of TVR compared to the Culotte technique (RR 0.39; 95% CI [0.27–0.57]; SUCRA 0.99), while Modified Crush showed a drastic increase in risk (RR 14.28; 95% CI [1.91–106.68]; SUCRA 0.00). These results are supported by the study by Chen et al. (2013), which showed that DK Crush reduced TVR rates (4.3%) compared to Culotte (11.0%). A meta-analysis by Wang et al. (2021) also showed that DK Crush ranked the best for TLR (100%) compared to other techniques. Similarly, a study by Di Gioia et al. (2019) reported a significant decrease in TLR with DK Crush.5,14,22 Research by Chen et al. (2019) also identified a trend of decreasing TLR with the Crush technique, although the difference was not significant. Our study’s findings strengthen the evidence that DK Crush remains the most effective option to prevent revascularization in complex bifurcation lesions.2
Our study shows that DK Crush has the lowest risk of stent thrombosis (ST) (RR 0.23; 95% CI [0.05–1.08]) and a reduction in the risk of myocardial infarction (MI) (RR 0.71; 95% CI [0.50–1.00]), although neither reached statistical significance. A similar reduction was also observed in the DKCRUSH-III study by,6 which showed an ST rate of 0% for DK Crush compared to 3.4% for Culotte. A meta-analysis by2 showed lower risks of MI and ST with the Crush technique compared to Culotte. In the EVOLUTE-CRUSH IV study, DK Crush demonstrated much lower TLR rates compared to Mini Culotte (7.1% vs. 15.9%; p=0.035).3 These findings reinforce that DK Crush consistently performs better clinically in terms of MI and ST parameters, although the limitation of wide confidence intervals remains an important consideration.
For outcomes of cardiac death (CD) and the need for coronary artery bypass grafting (CABG), our study shows no significant difference between DK Crush and Culotte (CD: RR 0.84; 95% CI: 0.56–1.25; CABG: RR 1.24; 95% CI: 0.58–2.62). However, Typical Crush emerged as the strategy with the best CABG profile, based on the highest SUCRA value (0.86). This finding differs from the study by,14 which showed that DK Crush reduced CD compared to Culotte. A meta-analysis by Mou et al., 2022 found that DK Crush significantly lowered MI and CD rates compared to Culotte.23 In the PROGRESS-BIFURCATION registry, DK Crush showed a reduction in major adverse cardiovascular events (MACE) compared to Culotte (HR 0.28; p=0.001), driven by a reduction in TVR rates.7 Therefore, while our study did not show statistical significance in CD and CABG outcomes, the literature supports the superiority of DK Crush in reducing long-term risks.
In our meta-analysis comparing Crush vs Cullote in CBL, MACE represented a frequent postoperative complication associated with significant morbidity and mortality. In this meta-analysis, we defined MACE as including cardiac death (CD), stent thrombosis (ST), myocardial infarction (MI), coronary artery bypass grafting (CABG), target lesion revascularization (TLR), and target vessel revascularization (TVR). We conducted an analysis focusing on the revascularization outcomes of TLR and TVR. In this network meta-analysis, DK-Crush demonstrated the most favorable and statistically significant outcomes in reducing the risk of TVR and TLR (RR 0.39 and 0.30, respectively) compared to other treatment strategies. However, in the pairwise analysis between Crush and Culotte, both outcomes showed no statistically significant differences (RR 0.80 [95% CI: 0.48–1.33] and 0.79 [95% CI: 0.52–1.20], respectively). Additionally, no significant differences were observed between one-year and three-year follow-up periods. These findings are consistent with two previous meta-analyses by Chen et al., (2020) which reported similar results.2 TVR is frequently employed as the primary endpoint for assessing long-term efficacy, as it more precisely reflects the anatomical site of stent deployment compared to (TVR). Consequently, TLR is considered a more clinically relevant measure and is preferred for evaluating the performance of stenting procedures.
For the take-home message in this research, we conclude that our study included 3,777 patients with coronary bifurcation lesions who underwent either the Crush or Culotte stenting technique. The DK Crush technique demonstrated superior performance in terms of TVR, TLR, ST, and MI compared to other techniques. However, for the outcomes of CD and CABG, no significant differences were found between the groups. The results of this study support DK Crush as the most promising intervention strategy for complex bifurcation lesions. Going forward, large-scale randomized trials are needed to confirm the superiority of DK Crush in the global population. Further evaluation with IVUS or OCT guidance is also important to understand the mechanism behind the success of this technique. New techniques, such as Nano Crush, are worth testing further in head-to-head studies against DK Crush. An individualized strategy based on bifurcation anatomy complexity and the SYNTAX score needs to be developed to optimize bifurcation invasive therapy.
This study has several limitations. First, the overall sample size included in the analysis was relatively small (n = 3,777), which may limit the statistical power of the findings. Second, some of the included studies showed a high degree of heterogeneity in their results, potentially affecting the robustness of the pooled estimates. Third, variations in patient populations, especially in terms of comorbidities and the severity of coronary bifurcation lesions, may influence the generalizability of the findings to broader clinical settings. Fourth, variations in patient populations, especially in terms of comorbidities and the severity of coronary bifurcation lesions, may influence the generalizability of the findings to broader clinical settings. Lastly, there was no direct comparison between certain subtypes of stenting techniques, such as DK Crush versus Nano Crush, which limits our ability to draw definitive conclusions regarding the relative efficacy among different variations of the Crush and Culotte techniques.
Conclusion
This systematic review and network meta-analysis comparing Crush and Culotte stenting techniques in coronary bifurcation lesions found that Culotte stenting had better short-term outcomes at one year, particularly in reducing target lesion and vessel revascularization. However, by three years, Crush stenting—especially the DK Crush subtype—demonstrated superior effectiveness in lowering TLR and TVR, with SUCRA rankings further supporting the dominance of various Crush subtypes in other outcomes like stent thrombosis and cardiac death. Although no significant differences were observed in myocardial infarction, stent thrombosis, cardiac death, or CABG between the two techniques, the overall evidence supports DK Crush as a more effective long-term strategy for complex PCI in bifurcation lesions.
Download
Export To
metrics
| Views | Downloads | |
|---|---|---|
| F1000Research | - | - |
|
PubMed Central
Data from PMC are received and updated monthly.
|
- | - |
Citations
CITE
how to cite this article
Komalasari I, Adji AS, Billah A et al. Long-Term Outcomes of Crush versus Culotte Stenting in Coronary Bifurcation Lesions: A Systematic Review and Network Meta-Analysis [version 1; peer review: awaiting peer review]. F1000Research 2025, 14:823 (https://doi.org/10.12688/f1000research.168528.1)
NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.
track
receive updates on this article
Track an article to receive email alerts on any updates to this article.
Open Peer Review
Current Reviewer Status:
AWAITING PEER REVIEW
AWAITING PEER REVIEW
Key to Reviewer Statuses
VIEW
HIDE
ApprovedThe paper is scientifically sound in its current form and only minor, if any, improvements are suggested
Approved with reservations
A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.
Not approvedFundamental flaws in the paper seriously undermine the findings and conclusions
Comments on this article Comments (0)