Author + information
- Received April 17, 2008
- Revision received November 24, 2008
- Accepted December 17, 2008
- Published online March 1, 2009.
- Alfredo R. Galassi, MD, FACC, FSCAI, FESC⁎ (, )
- Salvatore D. Tomasello, MD,
- Davide Capodanno, MD,
- Giovambattista Barrano, MD,
- Gian Paolo Ussia, MD, FSCAI and
- Corrado Tamburino, MD, FSCAI, FESC
- ↵⁎Reprint requests and correspondence:
Prof. Alfredo R. Galassi, Via Antonello da Messina 75, Acicastello, 95021, Catania, Italy
Objectives This retrospective study sought to assess the clinical and angiographic long-term outcome after implanting drug-eluting stents in bifurcation lesions with the T-provisional (T-prov) technique and mini-crush (MC) technique.
Background The best option on the treatment of coronary bifurcation lesions is a subject of considerable debate. However, recent evidence suggests that bifurcation lesions might be treated by drug-eluting stent on both branches using the MC technique with a low rate of major adverse cardiac event and restenosis.
Methods From April 2004 to July 2006, 457 patients were consecutively treated with either MC technique (n = 199) or T-prov technique (n = 258). Of these latter, 170 patients were treated with 1 stent and 88 patients with 2 stents. The 9-month angiographic follow-up was completed in 188 of 229 (82.1%) bifurcation lesions of MC patients and in 207 of 266 lesions (77.8%) of T-prov patients.
Results After a propensity score adjustment, 2-year cumulative major adverse cardiac events were similar between groups (p = 0.16). The MC group compared with the T-prov 1-stent group had significantly lower main and side branches restenosis (hazard ratio [HR]: 0.52, 95% confidence interval [CI]: 0.27 to 0.99; p = 0.047; and HR: 0.41, 95% CI: 0.20 to 0.85; p = 0.016, respectively). However, the MC group compared with the T-prov-only group had significantly lower side branch restenosis (HR: 0.55, 95% CI: 0.37 to 0.82; p = 0.004).
Conclusions Both techniques of bifurcation treatment met high procedural success with low complication rates and similar major adverse cardiac event long-term outcome. However, the MC technique yields a lower restenosis rate at both main and side branches. These results may confirm the advantage of using prescheduled 2-stent technique to give a complete coverage of the side branches' ostium.
Treatment of bifurcation lesions remains a technical challenge for the interventionalist, but it is still hampered by an increased rate of restenosis as compared with nonbifurcation lesions (1–3). Part of this treatment complexity is related to the different anatomic patterns of the stenosis, considering the possible involvement with the proximal and the distal part of the main branch and/or the side branches, and the close proximity of side branches origins (4), besides the angle that the side branch takes off and the size of both main and side branches. Various studies on the drug-eluting stent era indicate the absence of a real advantage over restenosis considering the potential for both early procedural complications (stent distortion, plaque shift, dissection, side branching trapping) and late in-stent restenosis when using a complex strategy (5–8). Indeed, anatomical characteristics of bifurcation lesion may influence the treatment strategy, choosing between 1- and 2-stent approaches.
More recently further refinements of the 2-stent technique (9–11) have shown that the 2-stent strategy technique, when appropriately performed, may significantly reduce the restenosis rate and potentially target lesion revascularization (TLR) and major adverse cardiac event (MACE) in complex bifurcation lesions. This study aims to compare the long-term clinical and angiographic outcomes after implanting the drug-eluting stent in bifurcations lesions using the mini-crush (MC) and T-provisional (T-prov) techniques.
Materials and Methods
The ethics committee approved the protocol, and all patients gave written informed consent. Demographic and procedural data regarding all patients undergoing angioplasty with drug-eluting stents—either sirolimus-eluting (Cypher/Cypher Select, Cordis Johnson & Johnson, Warren, New Jersey) or paclitaxel-eluting (Taxus Liberte, Boston Scientific, Natick, Massachusetts)—were prospectively put into a dedicated database.
Between April 2004 and July 2006, 3,480 patients underwent a percutaneous coronary intervention in our institution. Among them, 544 patients (16%) had de novo bifurcation lesions excluding left main trunk location, with a visually estimated percent diameter stenosis >50% involving the main vessel at or within 5 mm proximal or distal to the origin of a side branch >2 mm in diameter (12). All bifurcation lesions were classified by visual assessment according to a classification previously described by Medina et al. (12). The MC technique was performed in 199 patients (this group included 45 patients from a previous pilot study); T-prov in 170 patients with a single stent in the main vessel across the origin of the side branch (T-prov 1-stent); T-prov in the other 88 patients where an additional stent was required in the side branch because of the presence of severe impairment in the side branch during the angioplasty procedure (dissection or severe ostial stenosis due to plaque shift) (T-prov 2-stent). Other different techniques such as crush, culotte, and V-stenting were used in 87 patients. The bifurcation treatment strategy was chosen according to the operator's personal experience and preferences. All consecutive patients with a bifurcation lesion treated with either sirolimus- or paclitaxel-eluting stents by the MC or by a T-prov stenting technique with and without the additional implantation of a side branch stent were retrospectively identified in the operator's reports and checked through an independent review of all angiograms.
All patients were previously treated with aspirin 100 mg and either ticlopidine or clopidogrel. A 300-mg loading dose of clopidogrel before the index procedure was administered if patients were not pretreated. During the procedure, patients received intravenous unfractionated heparin (80 IU/kg) to maintain activated clotting time between 250 and 300 s. The administration of glycoprotein IIb/IIIa inhibitors was left to the operator's discretion. After the procedure, all patients were prescribed aspirin (100 mg daily) and clopidogrel (75 mg daily) or ticlopidine (500 mg daily) to be continued for at least 9 months after drug-eluting stent implantation. The MC and T-prov techniques have been previously described (9–13). Briefly, the MC technique as compared with the standard crush approach consisted of a minor retraction of the side branch stent into the main branch and using a balloon to crush the side branch stent instead of the main branch stent. Following this approach, a jailed wire technique and final kissing balloon would have been employed if possible, using wire recrossing into the side branch followed by high-pressure balloon inflation. As regards the T-prov approach, balloon dilatation of the side branch or kissing balloon were recommended before stent deployment in the parent vessel. The side branch was then rewired, and the balloon dilatation of the ostium was repeated across the metal structure of the stent; patients treated with this kind of approach were included in the T-prov 1-stent group. Stenting of the side branch origin was considered in the presence of a residual stenosis >50% or coronary Thrombolysis In Myocardial Infarction (TIMI) flow grade <3; patients treated with this approach were included in the T-prov 2-stent group. Consistent with the standard of care in our laboratory, final kissing balloon inflation was recommended in all patients.
All patients were monitored for any post-procedure events of chest pain, heart failure, bleeding, or any ischemic events. Creatine kinase-myocardial band and troponin I were measured at 6 to 8 and 12 to 24 h post-procedure.
Clinical definition and follow-up
Clinical follow-up was performed by office visit or by telephone interviews at 1 month, 6 months, 1 year, and 3 years. All events were confirmed by review of death certificates and hospital charts. Stress test or functional imaging test were recommended to be performed between 5 and 6 months, after the procedure for all patients. Angiographic follow-up was recommended as a laboratory standard of care, to be performed approximately at 9 months after the procedure unless indicated earlier by clinical suspicion or positive functional imaging test.
Major adverse cardiac events were defined as cardiac death, acute myocardial infarction (AMI), and target vessel revascularization, either percutaneous or surgical. All deaths were considered cardiac in origin unless alternatively documented. A non–Q-wave AMI was defined as creatine kinase-myocardial band enzyme elevation ≥3 times the upper limit of the normal value; when in addition to enzyme elevation there were new pathological Q waves in the electrocardiogram, the event was defined as a Q-wave AMI. Reintervention for restenosis was decided by taking into consideration the patient's symptoms, functional imaging testing, and angiographic findings. Target lesion revascularization was defined as a repeat revascularization with a stenosis ≥50% within the stent or in the 5-mm distal or proximal segments adjacent to the stent. Target vessel revascularization was defined as repeat revascularization within the treated vessel. Target bifurcation revascularization (TBR) was defined as a repeat revascularization with a stenosis ≥50% within 5 mm proximal or distal to the carina of bifurcation, both onto the main branch and/or side branch as previously described (9).
Stent thrombosis was defined as an acute coronary syndrome with angiographic documentation of either vessel occlusion or thrombus within or adjacent to a previously successful stented vessel or, in the absence of angiographic confirmation, either AMI in the distribution of the treated vessel or death not clearly attributable to any other causes (14). Stent thromboses were categorized according to the timing of the event into intraprocedural, subacute thrombosis (from the end of the procedure to 30 days), and late stent thrombosis (>30 days).
Quantitative coronary angiographic analysis
Coronary angiograms obtained at baseline, at completion of the stenting procedure, and at 9.6 ± 4.3 months for MC-treated patients, at 10.4 ± 3.7 months for T-prov 1-stent treated patients, and at 10.8 ± 5.6 months of T-prov 2-stent–treated patients (analysis of variance) (p = 0.147) were analyzed using CardiOp-B, a computer-based algorithm 3-dimensional reconstruction dedicated to bifurcations (Paieon Medical Ltd., Park Afek, Israel) (15). This system integrates information from at least 2 single-plane angiographic images taken from different angles of projection. One experienced technician who was blinded to patients' identities, outcomes, and sequence of the film, performed quantitative analyses of all angiographic data. Minimal lumen diameter and the nearest normal reference diameter were measured in millimeters using the catheter as a scaling factor. Percent stenosis was calculated as: 100 (1 − minimal lumen diameter/normal reference diameter). Binary angiographic restenosis was defined as >50% diameter stenosis of the target lesion. Late lumen loss was defined as the difference in minimal luminal diameter at completion of the stent procedure and during follow-up. Quantitative angiographic measurements of the target lesion were made at the “in-stent” zone (only the stented segment) and at the “in-segment” zone (stented segment and margins 5 mm proximal and distal to the stent). Angiographic success was defined as a final residual stenosis <30% with TIMI flow grade 3 in either the main branch or the side branch (16). Procedural success was defined as the achievement of angiographic success without in-hospital MACE. In patients who had restenosis, the pattern of restenosis was assessed according to the classification of Mehran et al. (17). Each patient's baseline, post-procedure, and follow-up coronary angiograms were analyzed. Patients for whom 1 of the angiograms was unreadable were cut out of the analysis: 15 from the MC group, 12 from the T-prov 1-stent group, and 5 from the T-prov 2-stent group. A total of 188 lesions out of 229 (82.1%) in 160 patients were analyzed for the MC group, 138 lesions out of 176 lesions (78.4%) in 132 patients were analyzed for the T-prov 1-stent group, and 69 lesions out of 90 lesions (76.7%) in 69 patients were analyzed for the T-prov 2-stent group
Continuous variables were presented as mean ± SD and were compared using Student unpaired t test or analysis of variance. Categorical variables were presented as counts and percentages and compared with the chi-square test when appropriate (expected frequency >5). Otherwise, the Fisher exact test was used. A 2-sided p value of <0.05 was considered to indicate statistical significance.
Propensity score was used to create a covariate that summarizes confounders within baseline demographic and angiographic characteristics and assesses the independent effect of treatment technique on all outcome measures. For each considered comparison (MC vs. T-prov 1-stent and MC vs. T-prov) a separate propensity score was derived. The details of the propensity-score method, with the resulting models and their predictive characteristics, are described in the Online Appendix. Calibration of the propensity score model was assessed using the Hosmer-Lemeshow test.
All data were processed using the Statistical Package for Social Sciences, version 15 (SPSS Inc., Chicago, Illinois).
Baseline and procedural characteristics
Baseline clinical characteristics are shown in Table 1 among the 3 groups of the patients. Table 2 shows the angiographic locations and types of treated bifurcation lesions. It is interesting to note that type 1,1,1 bifurcation was present in the great majority of cases in all the groups of patients and as type 1,1,1 more frequently in MC groups than in T-prov groups. Procedural details are shown in Table 3.
Clinical and angiographic outcomes are shown in Table 4. A higher rate of hospital MACE albeit small was shown in both T-prov groups of patients, which may be explained by the higher rates of unstable angina as clinical presentation in this cohort of patients. Successful delivery of stent to both main and side branch, in the case of MC was accomplished adequately in each case. An additional stent in the main branch to cover the full lesion length or to seal significant dissections at edges of study stent was required in 58 of 229 lesions (25.3%) in the MC group, in 58 of 176 lesions (32.9%) in the T-prov 1-stent group, and in 15 of 90 lesions (16.6%) in the T-prov 2-stent group. Two additional stents in the main vessel were implanted in 10 out of 229 lesions (4.4%) in the MC group, in 4 out of 176 lesions (2.3%) in the T-prov 1-stent group, and in 3 out of 90 lesions (3.3%) in the T-prov 2-stent group. An additional stent in the side branch was required in 8 of 229 (3.5%) in the MC group, but in only 1 case (1.1%) in the T-prov 2-stent group. Stress test or functional imaging test were performed in 96% of MC patients, in 93% of T-prov 1-stent patients, and in 92% of T-prov 2-stent patients.
The 9-month angiographic follow-up was completed in 188 of 229 (82.1%) bifurcation lesions of MC patients, in 138 of 176 (78.4%) lesions of T-prov 1-stent patients, and in 69 of 90 (76.7%) lesions of T-prov 2-stent patients. Angiographic quantitative coronary angiography results for both the main and the side branch at baseline, after procedure, and at follow-up are summarized in Table 5.
Clinical follow-up was available for all patients at 752 ± 445 days (25 ± 15 months). Based on the average follow-up time, no significant differences were observed among the groups. Cumulative MACE at follow-up was similar between MC and both T-prov patient groups (20.6% vs. 25.9% vs. 26.1%, p = NS among all 3 groups) and TLR (16% vs. 20.3% vs. 26.1%, p = NS among all 3 groups). However, it is interesting to note that TBR was significantly lower in MC patients as compared with T-prov 2-stent patients (8.5% vs. 17.4%, p < 0.05). The MC group compared with both T-prov groups showed a similar main branch restenosis (11.7% vs. 17.4% vs. 18.6%, p = NS among all 3 groups) but significantly lower side branch restenosis (8.5% vs. 18.8% vs. 25.7%, p < 0.01 of MC vs. T-prov 1-stent and p < 0.001 of MC vs. T-prov 2-stent, respectively). There were 37 focal restenosis at the ostium of side branch, 6 in the MC group, 17 in the T-prov 1-stent group, and 14 in the T-prov 2-stent group. There were 2 cases of multifocal and 3 of edge restenosis at the side and main branches in the T-prov 2-stent patients. There were also 24 cases of segment restenosis: 6 in the main branch of T-prov patient groups, and 6 in the side branch of all patients groups.
After adjusting for the baseline differences between the MC and T-prov 1-stent groups using propensity score, treatment type was a significant predictor of main and side branch restenosis at 9-month angiographic follow-up (hazard ratio [HR]: 0.52, 95% confidence interval [CI]: 0.27 to 0.99; p = 0.047, and HR: 0.41, 95% CI: 0.20 to 0.85; p = 0.016, respectively). Conversely, TBR, TLR, and long-term cumulative MACE were similar between the 2 groups (Fig. 1).
After adjusting for the baseline differences among the MC and both T-prov groups using propensity score, treatment type was a significant predictor of side branch restenosis at 9-month angiographic follow-up (HR: 0.55, 95% CI: 0.37 to 0.82; p = 0.004) (Fig. 2). Conversely, TBR, TLR, and long-term cumulative MACE were similar between the 2 groups.
The best treatment for bifurcation lesions has not been established yet. In the absence of proven efficacy of dedicated devices, drug-eluting stents and various techniques, generally divided into 2 basic strategies with or without side branch stenting, are generally used to treat patients with bifurcation lesions (12,18). Because some recent randomized studies of stenting of both branches failed to demonstrate superiority over main vessel stenting and balloon dilatation with provisional stenting of the side branch (6–8), the routine usage of 2 stents is actually not recommended. However, the Colombo trial utilizing sirolimus-eluting stents is difficult to interpret, as the number of patients is limited and there was a very high crossover rate from simple to complex strategy (6). The CORPAL (Drug-Eluting Stents for Complex Lesions: Randomized Rapamycin Versus Paclitaxel) trial is also dealing with a limited number of cases and the complex strategy was performed with a technique that did not have much success when it employed T-prov but not as double-stent intention-to-treat technique (7). The Nordic trial considered as the end point of an angiographic diameter stenosis >50% of the main vessel or an occlusion of the side branch. Indeed, in this series, restenosis rate in the side branch was significantly higher in the group using a simple strategy (19.2%) as compared with those of using a complex strategy (10.9%) (8). Furthermore, as recently stated, it could be impossible to do randomized trials in bifurcation lesions as each patient's anatomy may favor a particular technique, but trial patients will be overall heterogeneous (12) and still a remarkable subgroup of patients with bifurcation lesions (with larger side branches and/or large myocardium at risk supplied by the side branch and/or more diseased vessels and/or with suboptimal result of the side branch) may be treated with double stenting even if the intention is to try to avoid it.
During the last few years, various researchers have tried to find alternative strategies to the crush technique, one of the complex techniques with higher safety profile on long-term outcomes (19), but still with a significantly higher restenosis rate at the side branch (20) despite the kissing balloon after stenting, which is mandatory to reduce side branch restenosis, is performed (21,22). Recently, Jim et al. (10) has reported a modified approach of the crush technique. Using the so-called sleeve technique, these investigators split the final kissing balloon inflation into 2, so that the wire and balloon have to cross only 1 layer of stent struts at every time. Although, this approach has the advantage of avoiding manipulation or positioning of 2 bulky stents simultaneously, it has the drawback of being a more cumbersome and likely prolonged procedure. In many aspects this technique is similar to the “step crush technique” that Colombo et al. (23) have described in the past with a main difference due to the performance of recrossing, balloon inflations, and kissing inflations before the implantation of a stent into the main branch. A less cumbersome approach is illustrated by Sianos et al. (24), by Lim et al. (25), and more extensively in the study by Galassi et al. (9), where the MC technique described is a further refinement of the previous 2 approaches. Using this technique, which consists of crushing the 1- to 2-mm proximal side branch stent by a balloon instead of crushing the 3- to 4-mm proximal side branch stent by another stent, as in the standard crush technique, Galassi et al. (9) showed excellent in-hospital outcomes with low MACE and restenosis rates especially at the side branch in a pilot study. Following this approach it seems that there is less risk that the main-vessel stent will be deformed and potentially unopposed to the vessel wall and thus predisposed to restenosis or stent thrombosis of the side branch, as is further suggested by Ormiston and colleagues by bench studies (26,27).
Employing this MC technique in a consecutive series of patients with bifurcation lesions, the results of this study showed that this 2 drug-eluting stent technique provides excellent long-term outcomes and rates of MACE at long-term follow-up similar to those obtained by the T-prov approach using a single or a double stent. Similar rates of acute, subacute, and late thrombosis are found following the 2 different treatment strategies. The overall results obtained by the MC technique of a TLR of 16.0%, TBR of 8.5%, main branch restenosis of 11.7%, and side branch restenosis of 8.5% confirmed the results previously obtained in a small pilot study by Galassi et al. (9) where TLR was 12.2%, TBR 4.1%, main branch restenosis 12.2%, and side branch restenosis 2.0%.
Moreover, after adjusting for baseline differences between the MC and T-prov 1-stent groups using propensity score MC technique was found to provide significantly lower main and side branches restenosis, while when after adjusting for baseline differences between the MC and both T-prov groups using propensity score, main branch restenosis lost significance albeit small.
Comparison with previous studies
The overall T-prov TLR rate (22.2%) is higher than the rate previously reported in other studies (Table 6) (6–9,13,19–21,28). This was primarily due to a higher restenosis rate of the side branch as was recently shown in the Nordic study where restenosis rate at the side branch was 26.4% (8); however, in this previous study, the TLR rate remained extremely low (1.9%), mainly because the primary angiographic end point was restenosis in the main vessel or occlusion of the side branch. In our studies, potential differences include the longer duration of clinical follow-up (9 vs. 6 months) and the higher rate of assessing patients using stress test or functional imaging tests after procedure and treatment of silent ischemia. Indeed, following up this approach, 61 lesions in 93 cases of restenosis (65.6%) in our patients were treated as compared with only 7 of 17 (41%) in the Colombo study (6), 5 of 10 (50%) in the Tanabe study (27), and 4 of 36 (11%) of the Nordic study (8). Finally, in the study of Pan et al. (13), only 50% of patients completed the angiographic follow-up, thus reducing potential number of TLR and silent restenosis rate even further.
This study has some limitations due to its retrospective nature. The most important is the lack of randomization with regard to stenting strategy left to the operator's discretion. However, as pointed out by a recent document of consensus, randomized trials are difficult to carry out because each patient's anatomy may favor a particular technique, but trial patients overall will be heterogeneous (12). Furthermore, we cannot exclude a selection bias due to treatment unbalance among the surgeons.
Although the study consisted of a consecutive group of patients who all had routine angiographic follow-up and in all cases the stringent criteria of the MC technique or T-prov approach were used, it only reflected the experience of a single institution.
Both techniques of bifurcation treatment met high procedural success with low complication rates and similar MACE long-term outcomes. However, the MC technique, a double-stent intention-to-treat technique, yields a lower restenosis rate at side branches when compared with the T-prov technique. These results may confirm the advantage of using a pre-scheduled 2-stent technique to give complete coverage of the side branches' ostium as compared with a provisional technique whether or not a second bail-out stent is needed.
For propensity score and sensitivity analyses, please see the online version of this article.
- Abbreviations and Acronyms
- acute myocardial infarction
- confidence interval
- hazard ratio
- major adverse cardiac event
- target bifurcation revascularization
- target lesion revascularization
- T-provisional stenting
- Received April 17, 2008.
- Revision received November 24, 2008.
- Accepted December 17, 2008.
- American College of Cardiology Foundation
- Lemos P.A.,
- Serruys P.W.,
- van Domburg R.T.,
- et al.
- Garot P.,
- Lefevre T.,
- Savage M.,
- et al.
- Colombo A.,
- Moses J.W.,
- Morice M.C.,
- et al.
- Steigen T.K.,
- Maeng M.,
- Wiseth R.,
- et al.,
- NORDIC PCI Study Group
- Jim M.H.,
- Ho H.H.,
- Chan A.O.,
- Chow W.H.
- Burzotta F.,
- Gwon H.C.,
- Hahn J.Y.,
- et al.
- Cutlip D.E.,
- Windecker S.,
- Mehran R.,
- et al.,
- Academic Research Consortium
- Sheehan F.H.,
- Braunwald E.,
- Canner P.,
- et al.
- Mehran R.,
- Dangas G.,
- Abizaid A.,
- et al.
- Iakovou I.,
- Ge L.,
- Colombo A.
- Hoye A.,
- Jakovou J.,
- Ge L.,
- et al.
- Ge L.,
- Airoldi F.,
- Iakovou I.,
- et al.
- Costa R.A.,
- Mintz G.S.,
- Carlier S.G.,
- et al.
- Tanabe K.,
- Hoye A.,
- Lemos P.A.,
- et al.