Author + information
- Antonio Colombo, MD∗ ( and )
- Alaide Chieffo, MD
- Interventional Cardiology Unit, EMO GVM Columbus and San Raffaele Scientific Institute, Milan, Italy
- ↵∗Reprint requests and correspondence:
Dr. Antonio Colombo, Interventional Cardiology Unit, San Raffaele Scientific Institute, Via Olgettina 60, 20132 Milan, Italy.
It is now almost a decade that we have been discussing optimal treatment of bifurcation lesions, and invariably we come to the unanswered question: 1 or 2 stents?
Randomized clinical trials have been performed to assess the optimal stenting strategy for bifurcation lesions (1,2), and the answer is now clear: “if the lesion can be treated with 1 stent, there is no need to implant 2 stents (as intention-to-treat), provided crossover is possible” (3).
However, all of the randomized clinical trials evaluating 1- versus 2-stent strategies have been performed in non–left main (LM) bifurcation lesions (1,2). Therefore, we still do not have a clear answer on the optimal stenting strategy for bifurcations involving the LM. For this reason, the COBIS (Coronary Bifurcation Stenting Registry in South Korea) Registry II (4) results published in this issue of JACC: Cardiovascular Interventions are welcome. Before entering into a more specific analysis, an important point needs to be made: any registry, including COBIS, trying to compare 1 and 2 stents carries a major limitation due to the fact that patients with LM bifurcation lesions requiring 2 stents have a greater disease burden compared with patients who can be treated with 1 stent. The disease burden for patients requiring 2 stents can extend to other major epicardial vessels. In summary, patients needing 2 stents at the distal LM location have more atherosclerosis compared with patients who can be treated with 1 stent. In addition, baseline clinical and lesion characteristics of patients with bifurcation lesions treated with 2 stents are more adverse compared with those treated with 1 stent. Therefore, it is difficult to attribute worse clinical outcome to the stenting technique rather than unfavorable baseline characteristics. This digression is fundamental to critically evaluate the results of different registries including COBIS and represent the greatest limitation of observational registries compared with randomized clinical trials.
Important and positive attributes of the COBIS registry are the large number of consecutive patients with bifurcation lesions (2,897 patients, 853 of whom had LM bifurcation lesions), the multicenter design of the study, the exclusive use of drug-eluting stents, and minimal exclusion criteria. As expected, the 2-stent strategy was used more frequently in LM lesions compared with other bifurcations (40.3% vs. 20.8%; p < 0.01).
It is important to highlight that patients needing 2 stents were older; more frequently had a diagnosis of acute coronary syndrome; had more multivessel disease, more true bifurcations, and a higher SYNTAX score; more frequently required treatment of additional lesions besides the bifurcation; and had a greater disease burden in the main and side branch (more severe stenosis and longer lesions). Irrespective of the strategy used to treat these 2 groups of patients, good clinical judgment suggests that the prognosis of patients treated with 2 stents is likely to be less favorable than patients treated with 1 stent.
At a median follow-up of about 3 years, Song et al. (4) observed the following:
1. In non-LM bifurcations, patients treated with 2 stents had ∼3% to 4% (absolute) higher occurrence of target lesion revascularization (TLR) and target vessel revascularization (TVR) compared with patients treated with 1 stent without any difference in hard clinical events such as myocardial infarction and cardiac death.
2. In LM bifurcations, patients treated with 2 stents had more than double the occurrence of cardiac death, myocardial infarction, TLR, and TVR compared with patients treated with 1 stent.
3. The incidence of definite or probable stent thrombosis was higher in patients treated with 2 stents. In non-LM bifurcations, the difference did not reach statistical significance. Conversely, in LM bifurcations treated with 2 stents, definite or probable stent thrombosis was significantly higher (3.2% vs. 0.6%; p < 0.01).
The provocative finding is that, after appropriate and sophisticated statistical adjustment, the 2-stent strategy had a hazard ratio of 2.38 (95% confidence interval: 1.60 to 3.55; p < 0.01) for the primary and secondary endpoints in patients treated for LM bifurcation lesions.
How can we interpret these results?
First, these results have to be interpreted taking into account the important limitations of this registry, as also pointed out by the authors. There are remarkable differences in baseline clinical and lesion characteristics between patients treated with 1 versus 2 stents, and the ability of statistical adjustment to correct imbalances may be imperfect. Nevertheless, the hazard ratio of 2.38 with an acceptable confidence interval cannot be dismissed.
Our belief is that we are dealing with 2 sets of variables: patient/lesion complexity and higher metal burden (2 stents), leading to a higher risk of thrombosis and associated major adverse cardiovascular events such as death and myocardial infarction. In addition, as pointed out in some randomized studies and registries, the 2-stent strategy is technically more complex and sometimes may be associated with a suboptimal result (more frequently in unfavorable bifurcation angles), leading to a higher risk of stent thrombosis.
Why does the 2-stent strategy affect only lesions located in LM bifurcations?
Our interpretation is that major adverse cardiovascular events are rarely clinically silent in patients treated with LM lesions, whereas lesions located in other vessels, especially non–left anterior descending, may have vascular complications without a manifest clinical event.
What clinical and technical message can we take from these findings?
First, because of the limitations of the study design, we should be cautious in interpreting the results before dismissing the 2-stent strategy. In addition, we should consider the following points.
Every time we plan 2 stents in an LM bifurcation as intention-to-treat, we need to take into account the option of surgical revascularization, as correctly pointed out by the authors. If the surgical option is not viable, we should carefully evaluate the possibility of obtaining an optimal acute result, such as adequate stent expansion and lumen dimensions.
If we decide to proceed with 2 stents or we are crossing over from a 1-stent strategy, an optimal technique is a must. This includes lesion preparation, optimal stent expansion, and assessment of the result by intravascular ultrasound (5,6).
We need to accept that more metal means more foreign body and a higher risk of an imperfect result. Intermediate results should be evaluated with modern techniques such as fractional flow reserve (6), and crossover to 2 stents should not be performed to please “coronary aesthetics.”
When we implant 2 stents, especially in an LM bifurcation, we should be more careful to optimize antiplatelet therapy, and when prescribing clopidogrel, it may be important to evaluate platelet responsiveness (7) or to consider using prasugrel or ticagrelor.
The bottom line is this: when we are treating patients with more unfavorable baseline and lesion characteristics, especially if the target lesion is the LM bifurcation, more attention must be paid, and any decision should be carefully weighed, particularly when we need to implant 2 stents.
This study is telling us that 2 stents in LM bifurcations might be an effect of greater disease burden, but might also be the cause of more adverse clinical events.
↵∗ Editorials published in the JACC: Cardiovascular Interventions reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Interventions or the American College of Cardiology.
Dr. Colombo is a shareholder in Cappella Inc. and a consultant for CID Spa Saluggia. Dr. Chieffo has reported that she has no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Zimarino M.,
- Corazzini A.,
- Ricci F.,
- et al.
- Song Y.B.,
- Hahn J.Y.,
- Yang J.H.,
- et al.
- Kang S.J.,
- Ahn J.M.,
- Song H.,
- et al.
- Puri R.,
- Kapadia S.R.,
- Nicholls S.J.,
- et al.
- Aradi D.,
- Storey R.F.,
- Komócsi A.,
- et al.,
- on behalf of the Working Group on Thrombosis of the European Society of Cardiology