ABSTRACT: Background. To assess the optimal percutaneous coronary intervention (PCI) approach for coronary artery bifurcation lesions (CBL), we conducted a meta-analysis of randomized trials comparing provisional stenting (PS) to complex stenting strategy (CS). Data Sources. PubMed, Cochrane Register of Controlled Trials, conference proceedings, and internet-based resources of clinical trials.
Data Synthesis. Six randomized trials comparing the PS to the CS approach for CBL with a total of 1,641 patients met the selection criteria for meta-analysis. There was no difference in the clinical profile between the two groups. No significant heterogeneity was found across trials. There was no difference in the reference vessel diameter of the main vessel (MV) (2.73 ± 0.41 CS; 2.7 ± 0.44 PS; p =0.77) and side branch (SB) (2.31 ± 0.33 CS; 2.27 ± 0.34 PS; p = 0.30).There was no difference in the primary clinical outcome of major adverse cardiovascular events (MACE) between the two approaches (12.6% vs. 9.6%; relative risk [RR] 1.23, 95% CI, 0.91–1.68; p = 0.18). Similarly, no differences in other clinical endpoints including death (1% vs. 1.1%, RR 0.93, 95% CI, 0.37–2.33; p = 0.87), target lesion revascularization (TLR) (6% vs. 5.3%, RR 1.10, 95% CI, 0.73–1.64; p = 0.66), stent thrombosis (ST) (1.8% vs. 0.8%, RR 1.60, 95% CI, 0.65–3.91; p = 0.30), MV restenosis (4.9% vs. 5%; RR 0.74, 95% CI, 0.40–1.38; p = 0.34) and SB restenosis (13.8% vs. 13.8%; RR 1.00, 95% CI, 0.65–1.54); p = 0.99] were observed at a mean follow up of 10 months and a mean angiographic follow up of 7 months. Myocardial infarction (MI) was, however, significantly higher in the CS vs. the PS group (6.8% vs. 3.6%, RR 1.71, 95% CI, 1.02–2.88; p = 0.04).
Conclusion. A CS strategy for CBL had a significantly higher risk of MI compared to a PS strategy. Rates of death, ST, restenosis and TLR were similar.
J INVASIVE CARDIOL 2009;21:589–595
Key words: coronary bifurcation lesions, provisional stenting, complex stenting, outcomes
Coronary bifurcation lesions (CBL) remain among the most technically challenging segments in percutaneous coronary interventions (PCI). Interventions on bifurcation lesions are commonly associated with low procedural success and high complication and restenosis rates despite advances in device technology including the introduction of drug-eluting stents (DES).1–5 Stenting of a coronary artery stenosis involving a bifurcation may result in an “apparent” significant angiographic “jailing” of the side branch (SB).3,4 Thus, angioplasty or stenting of the SB is usually undertaken to improve blood flow into the SB, but this may increase subsequent risks of restenosis and/or stent thrombosis in the SB. Major adverse cardiac events (MACE) are higher in bifurcation lesions compared to non-bifurcation lesions and this is largely due to a higher target lesion revascularization (TLR) rate in the former.3–6
Coronary bifurcation lesions comprise 15–20% of all PCI procedures6–8 and these are expected to increase with technological advancement and refinement of techniques. This percentage takes on increasing significance when put into the larger context of more than 1,265,000 PCI procedures being performed annually in the United States, and an estimated 2 million procedures worldwide.9 However, there is a lack of an established and validated “optimal” treatment strategy for bifurcation lesions. This has largely been due to a lack of large randomized trial data and has resulted in the adoption of different strategies by different operators.1,2,6 Conventional approaches, largely based on the operator’s discretion, include stenting the main vessel (MV) with provisional intervention on the side branch (SB) (provisional strategy) versus deploying stents in both the MV and SB (complex strategy). The former approach has been more widely practiced,1,6 given the lack of established benefit of the systematic complex (two-stent) strategy (CS) over provisional stenting (PS), and partly because of the relative technical ease of the former.
Recently, several published studies, including six randomized trials addressing the issue of optimal treatment strategy for bifurcation lesions, have enhanced our understanding of bifurcation lesions.10–15 There is no evidence to support the superiority of routine stenting of both vessels compared to the provisional strategy. In fact, the use of stents in both limbs of a bifurcation lesion has been associated with longer procedure and fluoroscopy times, higher contrast volumes and increased frequency of procedure-related biomarker release. Additionally, stent thrombosis (ST) occurs more commonly in the stented SB.12,14,15
Our objective was to evaluate the clinical and angiographic outcomes, hence the optimal PCI strategy, by conducting a meta-analysis of the available randomized, controlled trials (RCT) comparing provisional versus complex strategies for coronary bifurcation lesions.
Methods
Search strategy. A literature search was conducted using PubMed, Embase, the Cochrane Central Register of Controlled Trials, Google Scholar and Internet-based sources of information on clinical trials (www.clinicaltrials.gov) in cardiology from January 1998 to September 2008. The Medical Subject Heading (MeSH) terms “coronary artery bifurcation lesions,” “PCI,” “stent” and “randomized trial” were used. No language restrictions were applied. Bibliographies of relevant studies and the “Related Articles” link in PubMed were used to identify additional studies. Published abstracts from the annual meetings of the American College of Cardiology (ACC), American Heart Association (AHA), European Society of Cardiology (ESC), TransCatheter Therapeutics (TCT), Society of Coronary Angiography and Intervention (SCAI) and Euro Percutaneous Coronary Revascularization (Euro PCR) were also identified. Reference lists of review articles and cited articles were used to locate additional studies. The following journals were searched from January 2002 to March 2008: Heart, Journal of the American College of Cardiology, American Journal of Cardiology, American Heart Journal, Circulation, Hypertension, Circulation Research, European Heart Journal, Journal of Invasive Cardiology, Journal of Interventional Cardiology, and International Journal of Cardiology.
Data extraction. Studies were selected and data were extracted independently by two reviewers (A.H. and F.K.). Disagreements were resolved by consensus. The studies were evaluated carefully for duplicate or overlapping data. In addition to the number of participating patients, we recorded the following clinical and angiographic characteristics: age, gender, diabetes mellitus, hypertension, left ventricular ejection fraction (LVEF), number of treated lesions, mean reference diameter of the vessel (MV/SB) where the lesion was located, and mean lesion length. The duration of clinical follow up and time point of follow-up angiography, as well as clinical follow-up rates, were also recorded. Raw data obtained from source information of the individual studies were used for all analyses.
We referred to the guidelines of the Cochrane Handbook for Systematic Reviews of Interventions 5.0.1 for our meta-analysis.16 We evaluated studies for effective randomization, performance of the analysis according to the intention-to-treat principle and blind assessment of the outcomes of interest.
Statistical analyses. All analyses of the trials were performed based on the intention-to-treat principle, except Colombo et al, where endpoint analysis was performed by actual treatment.
10 There was no need for data imputation. Risk ratios (RR) with 95% confidence intervals (CI) were computed as summary statistics. Categorical variables were reported as percentages and continuous variables are presented as means ± standard deviation. The pooled RR was calculated with the DerSimonian-Laird method for random effects.
17 To assess heterogeneity across trials, we used the Cochran Q-test based on the pooled RR by Mantel-Haenszel. Heterogeneity was also assessed by means of the I2 statistic as proposed by Higgins et al (determining the variance across groups as a result of heterogeneity instead of chance).
18 Results were considered statistically significant at p 16
Study selection. Randomized studies comparing provisional stenting (PS) to complex stenting (CS) for coronary bifurcation lesions (CBL) with at least > 30 patients in each group reporting the outcomes of interest: target lesion revascularization (TLR), angiographic restenosis of the main vessel (MV) and side branch (SB), stent thrombosis, myocardial infarction (MI), death, and the composite of death, TLR and MI (MACE) with at least 6 months of clinical and angiographic follow up.
Results
Six randomized trials comparing the PS versus CS strategies for coronary bifurcation lesions met the inclusion criteria for the meta-analysis. A total of 1,641 patients were included in these trials, with 841 randomized to the CS and 800 to the PS arms.
Study characteristics are summarized in Table 1. In the pooled population, men constituted 79.5% of included patients, 20% had diabetes mellitus and 58% had hypertension. There were no significant differences between patients assigned to either the CS or PS strategy with regard to age, LVEF, prevalence of diabetes and hypertension. The mean duration of clinical follow up was 10.34 months (6–24 months), and the length of angiographic follow up (reported in 5 out of 6 studies) varied between 6 and 9 months after the index procedure. The overall angiographic follow-up rate was 86% (88% among patients treated with CS and 85% among patients treated with PS).
Table 2 shows the angiographic profile of the two arms in the six trials. There were no significant differences in the reference vessel diameters (RVD) of both MVs and SBs between the CS and PS groups. Five studies employed sirolimus-eluting stents (SES) and one used paclitaxel-eluting stents (PES). All 6 trials reported MACE as their primary outcome of interest. Additionally, five out of the six trials reported follow-up angiographic data. A funnel plot of the trials included is depicted in Figure 1; it shows a good conformation to the ideal funnel-shaped distribution.
Outcomes
Clinical endpoints. Risk ratios with 95% CIs for clinical and angiographic follow up are presented in Figures 2–8. The primary outcome of interest in our study was total MACE, which was reported in all six trials. The pooled incidence of MACE was 12.6% in the CS group and 9.6% in the PS arm, translating into a RR of 1.23 (95% CI, 0.91–1.68; p = 0.18) (Figure 2). In the pooled patient population, the incidence of MI was significantly higher in the CS arm compared to the PS arm: 6.8% vs. 3.6% (RR 1.71; 95% CI, 1.02–2.88; p = 0.04) (Figure 3). With respect to other clinical outcomes, no differences were observed between patients treated with CS strategy compared to those treated with the PS approach for CBL.
The mortality rate was 1% in the CS arm and 1.1% among patients in the PS arm (RR 0.93, 95% CI 0.37–2.33; p = 0.87) (Figure 4). Similarly, there was no difference in TLR between the two arms (6% vs. 5.3%, RR 1.1, 95% CI 0.73–1.64; p = 0.66) (Figure 5). The incidence of stent thrombosis was 1.8% with CS and 0.9% with PS (RR 1.6, 95% CI 0.65–3.91; p = 0.30) (Figure 6). No significant heterogeneity was found between trials with respect to any of these study outcomes.
Angiographic endpoints. Five of the six trials, except for BBC-ONE,10 reported angiographic outcomes at a mean follow-up of 7.2 months (6–12 months) in 85% of patients in the PS arm and 88% patients in the CS arms. With respect to the angiographic outcome of interest in this study, no difference was found in the binary restenosis rates of MV (4.9% vs. 5%; RR 0.74 [0.40–1.38]; p = 0.34) and SB (13.8% vs. 13.8%; RR 1.0 [0.65–1.54]; p = 0.99). There was no significant heterogeneity between trials with respect to this outcome (Figures 7 and 8).
Discussion
This meta-analysis of randomized, controlled trials comparing CS to PS strategies for coronary bifurcation lesions shows that routine stenting of both the MV and SB (complex strategy) offers no clinical or angiographic superiority over a more conservative provisional treatment approach to such lesions. In fact, patients who underwent CS had a significantly higher relative risk of myocardial infarction. This result was driven largely by the higher MI rate in the BBC-ONE trial.
10 Additionally, the use of a CS strategy was associated with significantly higher contrast agent use and longer total procedure and fluoroscopy times, as reported in three of the six randomized studies.
10,13,14 Importantly, there was no difference between the two strategies in terms of death, TLR, ST or restenoses of the MV or SB.
In all trials, there was exclusive use of drug-eluting stents (DES). Historical data describing the use of bare-metal stents (BMS) in bifurcation lesions suggest a TLR rate of 16– 38%,
2,4,22,23 with a tendency toward increased restenosis after stenting of both the MV and SB compared to single-vessel stenting. The only randomized study comparing DES to BMS for CBLs comes from a subanalysis of the SCANDSTENT trial in which implantation of SES was associated with significant reductions in restenosis rates at the MB (4.9% vs. 28.3%; p 19 This difference in MACE was driven by decreased TLR. The long-term outcome of the SCANDSTENT trial showed a sustained benefit of SES implantation at 3 years compared with BMS.
20 Similarly, registry studies for DES have shown marked reductions in MACE and TLR rates compared with historical BMS controls. These reductions occurred regardless of whether a provisional stenting (MACE: 5.4% vs. 38%; TLR: 5.4% vs. 36%) or complex stenting (MACE: 13.3% vs. 51%; TLR: 8.9% vs. 38%) strategy was undertaken.
3,21–23
Of the 6 RCTs included in this meta-analysis, 5 used SES
11–15 and 1 used PES.
10 DES have become the preferred choice for the treatment of coronary bifurcation lesions. SES have been the most extensively studied in the area of CBL. In one randomized trial comparing SES with PES in treating CBL, patients with SES showed significantly lower rates of late lumen loss, restenosis and TLR compared to patients treated with PES.
24 Newer-generation stents, including zotarolimus and everolimus-eluting stents, may have additional advantages (e.g., lower rate of stent thrombosis with zotarolimus-eluting stents). While the efficacy and safety of these new stents have been demonstrated in multiple trials, the question of whether comparable clinical outcomes can be extended to lesion subsets of higher complexity, including bifurcation lesions, needs further investigation.
All studies used the intention-to-treat principle for data analysis except Colombo in 2004, whose results were reported by treatment received.
12 In that trial, 51% of patients in the PS group required stent placement in the SB. This was thought to be necessary to obtain an acceptable angiographic result. Among the included studies, the decision to stent the SB was based on a variety of factors that differed between trials. The crossover rate ranged from 2–18% in five of the six trials except in the Colombo 2004 trial,
12 where 50% of patients originally assigned to the provisional arm received stenting of both vessels. This major difference in the crossover rate may be attributable to the fact that in the Colombo 2004 study, relatively broad criteria for SB stenting were employed, i.e., suboptimal PS results. In all other trials,
10,11,13–15 SB stenting was only recommended (crossover) if there was a flow-limiting dissection or a significant SB residual stenosis in the PS arm.
While bifurcation lesions have been previously identified as independent predictors of ST, especially in the stented SB,
25 there was no difference in the incidence of ST between the two groups in our analysis. Compared to prior studies where the incidence of ST has been between 3.5% with T-stenting technique
26 and 4.5% with the crush-stenting technique,
27 the overall incidence of ST in the pooled population was relatively low (1.8% for CS and 0.8% for PS; p = 0.3).
A variety of approaches including crush, culotte, or T-stenting techniques, based on the operators’ discretion, were employed for SB intervention in the CS arm.
1 All of the trials’ protocols mandated kissing-balloon inflation at the end of each CS intervention. Ending the intervention with this strategy helps to realign stent geometry and minimize stent distortion after SB balloon inflation.
1,2,6 While the superiority of final kissing-balloon inflations for optimizing stent geometry in both the main and the SB is well recognized, there are no randomized data or formal consensus statements to evaluate balloon-sizing/overlap for the kissing technique.
Interestingly, a study by Koo et al using fractional flow reserve (FFR) to assess jailed SBs, revealed that only 27% of angiographically apparent significant SB stenoses were hemodynamically significant when using a FFR value 28 The outcomes were favorable for FFR > 0.75 in the initial SB, with no adverse events or target vessel revascularization at a mean follow up of 9.6 ± 4.6 months. This indicates the possibility that there may be overuse of SB interventions solely based on angiographic assessment.
One of the important findings in this study was the significant increase in MIs in patients undergoing a CS procedure. These included periprocedural as well as subsequent MIs. This could be explained at least in part by longer stented segments, overlapping stents (possibly compromising a small SB), longer procedure time and increased complexity of the procedure.
Another factor to consider is the use of glycoprotein (GP) IIb/IIIa receptor antagonists and their beneficial effect on reducing periprocedural MIs. Specifically, the Evaluation of Platelet IIb/IIIa Inhibitor for Stenting (EPISTENT) trial demonstrated a reduction in periprocedural MI and 1-year mortality with the use of GP IIb/IIIa inhibitors.
29 This effect cannot be studied, given the nonuniform reporting of the use of GP IIb/IIIa receptor antagonists in the analyzed trials. Additionally, the lack of patient-level data prohibits meta regression analysis, thereby precluding the assessment of GP IIb/IIIa receptor antagonists’ impact on outcomes in either of the two strategies.
Recent analysis by Niccoli et al
30 pooled three prospective observational and three randomized studies to address the issue of optimal stenting strategy for CBL. The authors concluded that while CS does not offer any advantages over the simple stenting technique, there was a trend toward lower early MI with simple stenting. They underlined the need for further randomized trials to address this issue.
Since their study,
29 results from three new well-designed randomized trials have become available (which have been included in our meta-analysis). These trials reaffirm that simple stenting for bifurcation lesions offers overall better clinical outcomes (largely driven by a lower incidence of MI) compared to a CS strategy. Our meta-analysis summarizes all the available randomized data (six trials) and establishes the safety, efficacy and overall clinical superiority of a simple stenting technique for CBL. We believe this report adds incremental value to the current evidence and will have a pronounced impact on the percutaneous approach to bifurcation lesions.
Limitations of the analysis. This study has the inherent limitations present in all meta analyses. These include the inherent limitations in the analyzed trials even though they were all well-designed randomized, controlled studies. There was lack of detail regarding specific criteria for SB intervention in PS, which may account for minor systematic differences. The large proportion of crossovers in some of the studies also may have contributed to the general lack of significant differences noted between the strategies, other than MI risk. There is little information on the use of GP IIb/IIIa receptor antagonists. It is possible that the disparity in the use of these agents may have affected the periprocedure MI rate between different studies. Moreover, the definition of MI was nonuniform across trials. Additionally, details regarding specific types of bifurcation lesions (based on Medina/Lefevre Classification) and stenting strategy are lacking. The mean vessel diameter was Implications for clinical practice. PS as an initial strategy is superior to CS in treating CBL and is associated with a lower incidence of MI. This suggests that a CS strategy should not be the routine approach and should only be reserved for specific cases that have severe stenosis involving the ostium of a large SB. The major clinical implication of this analysis is that routine stenting of the SB is not warranted and may have untoward clinical outcomes, particularly increased rate of MI. Further research in this area, including randomized evaluation of a FFR-based approach to guide SB intervention, is needed.
References
1. Iakovou I, Ge L, Colombo A. Contemporary stent treatment of coronary bifurcations. J Am Coll Cardiol 2005;46:1446–1455.
2. Lefevre T, Louvard Y, Morice MC, et al. Stenting of bifurcation lesions: Classification, treatments, and results. Catheter Cardiovasc Interv 2000;49:274–283.
3. Yamashita T, Nishida T, Adamian MG, et al. Bifurcation lesions: Two stents versus one stent-immediate and follow-up results. J Am Coll Cardiol 2000; 35:1145–1151.
4. Suwaidi JA, Berger PB, Rihal CS, et al. Immediate and long-term outcome of intracoronary stent implantation for true bifurcation lesions. J Am Coll Cardiol 2000;35:929–936.
5. Suwaidi JA, Yeh W, Cohen HA, et al. Immediate and one-year outcome in patients with coronary bifurcation lesions in the modern era (NHLBI dynamic registry). Am J Cardiol 2001;87:1139–1144.
6. Latib A, Colombo A. Bifurcation disease: What do we know, what should we do? J Am Coll Cardiol Interv 2008;1;218–226.
7. Meier B, Gruentzig AR, King SB III, et al. Risk of side branch occlusion during coronary angioplasty. Am J Cardiol. 1984;53:10–14.
8. Myler RK, Shaw RE, Stertzer SH, et al. Lesion morphology and coronary angioplasty: Current experience and analysis. J Am Coll Cardiol 1992;19:1641–1652.
9. Heart Disease and Stroke Statistics —2008 Update American Heart Association available at: https://www.americanheart.org/presenter.jhtml?identifier=3000090.
10. Hildick-Smith D; BBC ONE (British Bifurcation Coronary study: Old, new and evolving) strategies trial, presented at TCT 2008, Washington, D.C.
11. Colombo A, Bramucci E, Saccà S, et al. Randomized study of the crush technique versus provisional side-branch stenting in true coronary bifurcations: The CACTUS (Coronary Bifurcations: Application of the Crushing Technique Using Sirolimus-Eluting Stents) study. Circulation 2009;119:71–78.
12. Colombo A, Moses JW, Morice MC, et al Randomized study to evaluate sirolimus-eluting stents implanted at coronary bifurcation lesions. Circulation 2004;109:1244–1249.
13. Ferenc M, Gick M, Kienzle RP, et al. Randomized trial on routine vs. provisional T-stenting in the treatment of de novo coronary bifurcation lesions. Eur Heart J 2008;29:2859–2867.
14. Steigen TK, Maeng M, Wiseth R, et al. Randomized study on simple versus complex stenting of coronary artery bifurcation lesions: The Nordic bifurcation study. Circulation 2006;114:1955– 1961.
15. Pan M, de Lezo JS, Medina A, Romero M, et al. Rapamycin-eluting stents for the treatment of bifurcated coronary lesions: A randomized comparison of a simple versus complex strategy. Am Heart J 2004;148:857–864.
16. Higgins JPT, Green S (eds). Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.1 [updated September 2008]. The Cochrane Collaboration, 2008. Available from www.cochrane-handbook.org.
17 DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7:177–188.
18. Higgins JPT, Thompson SG. Quantifying heterogeneity in a meta-analysis. Statistics in Medicine 2002;21:1539–1558.
19. Thuesen L, Kelbaek H, Klovgaard L, et al. Comparison of sirolimus eluting and bare metal stents in coronary bifurcation lesions: Subgroup analysis of the Stenting Coronary Arteries in Non-Stress/Benestent Disease Trial (SCANDSTENT). Am Heart J 2006;152:1140–1145.
20. Kelbaek H, Klovgaard L, et al. Long-term outcomes in patients treated with sirolimus eluting stents in complex coronary artery lesions. 3-year result of the scandstent (Stenting coronary arteries in non-stress/Benestent Disease) trial. J Am Coll Cardiol 2008;51:2011–2016.
21. Pan M, Suarez de Lezo J, Medina A, et al. Simple and complex stent strategies for bifurcated coronary arterial stenosis involving the side branch origin. Am J Cardiol 1999;83:1320–1325.
22. Anzuini A, Briguori C, Rosanio S, et al. Immediate and long-term clinical and angiographic results from Wiktor stent treatment for true bifurcation narrowings. Am J Cardiol 2001;88:1246–1250.
23. Ge L, Tsagalou E, Iakovou I, et al. In-hospital and nine-month outcome of treatment of coronary bifurcational lesions with sirolimus-eluting stent. Am J Cardiol 2005;95:757– 760.
24. Pan M, Suárez de Lezo J, Medina A, et al. Drug-eluting stents for the treatment of bifurcation lesions: A randomized comparison between paclitaxel and sirolimus stents. Am Heart J 2007;153:15. e1–e7.
25. Iakovou I, Schmidt T, Bonizzoni E, Ge L, et al. Incidence, predictors, and outcome of thrombosis after successful implantation of drug-eluting stents. JAMA 2005;293:2126–2130.
26. Colombo A, Stankovic G, Orlic D, et al . Modified T-stenting technique with crushing for bifurcation lesions: Immediate results and 30-day outcome. Catheter Cardiovasc Interv 2003;60:145–151.
27. Ge L, Airoldi F, Iakovou I, et al. Clinical and angiographic outcome after implantation of drug-eluting stents in bifurcation lesions with the crush stent technique: Importance of final kissing balloon post-dilation. J Am Coll Cardiol 2005;46:613–620.
28. Koo BK, Kang HJ, Youn TJ, et al. Physiologic assessment of jailed side branch lesions using fractional flow reserve. J Am Coll Cardiol 2005;46:633–637.
29. Topol EJ, Mark DB, Lincoff AM, et al. Outcomes at 1 year and economic implications of platelet glycoprotein IIb/IIIa blockade in patients undergoing coronary stenting: results from a multicentre randomised trial. EPISTENT Investigators. Evaluation of Platelet IIb/IIIa Inhibitor for Stenting. Lancet 1999;354:2019–2024.
30. Niccoli G, Ferrante G, Porto I, et al. Coronary bifurcation lesions: To stent one branch or both? A meta-analysis of patients treated with drug eluting stents. Int J Cardiol. 2008 Nov 21 [Epub ahead of print].
