Cardiovascular Revascularization Medicine, 2018-07-01, Volume 19, Issue 5, Pages 493-497, Copyright © 2017 Elsevier Inc. Abstract Aims To analyse systematic isolated post-dilatation of the side branch as a part of provisional stent technique. Methods 1960 angioplasties performed in two centres were prospectively registered, of which 382 were coronary bifurcations with a side branch > 2 mm. In centre A, isolated post-dilatation of the side branch was performed regardless its impairment after main vessel stenting. In centre B, side branch post-dilatation was performed only if it was severely affected after stent implantation. Results There was no difference between the two centres in the rate of side branch affection after stent implantation (A: 44.6 vs B: 49.3%, p = 0.48) nor in the procedural success rate (A: 98.6% vs B: 96.7%, p = 0.45). After one-year follow-up, a reduction of cardiovascular events was observed in centre A (A: 4.4% vs B: 10.4%, p = 0.043) with a trend towards lower cardiac mortality (A: 2.2% vs B: 6.5%, p = 0.093) and stent thrombosis (A: 0% vs B: 2.6%, p = 0.077). There were no differences in the rate of myocardial infarction related to the treated artery (A: 1.4% vs B: 3.9%, p = 0.29), or target lesion revascularization (A: 1.4% vs. B: 3.2%, p = 0.45). Conclusions Systematic isolated post-dilatation of the side branch in the provisional stent technique was associated with a high angiographic success rate, and a low rate of cardiovascular events during follow-up. Although the study design does not allow definitive conclusions, this strategy could be considered a valid option in some cases or even as part of the provisional stent technique. Highlights Several options exist on side branch treatment in provisional stent technique. Systematic post-dilatation of the side branch was associated with a lower MACE rate. This technique showed a trend towards lower cardiac mortality and stent thrombosis. It was not associated with greater complications or increasing fluoroscopy time. This technique could be considered as a part of the provisional stent technique. 1 Introduction The most common strategy for percutaneous coronary intervention (PCI) of bifurcated lesions is the provisional stent technique (PST) in which a stent is implanted in the main vessel (MV) and the side branch (SB) is only treated if it is affected after the implantation. However there are no uniform criteria to consider as significant the SB impairment after MV stenting . The European Bifurcation Club established that the affection of the SB is significant when it presents an angiographic stenosis > 75% or flow impairment (TIMI < 3) after MV stenting. There is also no consensus about the intervention that must be performed in an affected SB after MV stenting. Most of the studies that have tackled this issue performed SB dilatation systematically and almost all of them using the Final Kissing balloon (FKB) technique. Although systematic FKB is mandatory after the implantation of 2 stents, its benefit in PST is not clear and has not been shown to be superior to the isolated dilation of SB . In addition, there is also no agreement on performing an isolated dilatation on an unaffected SB after stent implantation in the MV. Our objective was to analyse the benefit of the isolated systematic post-dilatation of the SB after MV stenting in the PCI of bifurcated lesions performed by PST. 2 Methods 2.1 Patients and study design A prospective registry was performed in two centres with a high volume of angioplasties. From March 2014 to April 2015, 1960 angioplasties were prospectively registered, of which 382 were coronary bifurcations with an SB > 2 mm. Patients older than 18 years with indication for PCI regardless of clinical presentation [stable coronary artery disease (SCAD), non-ST elevation myocardial infarction (NSTEMI), or ST elevation myocardial infarction (STEMI)] were included. All bifurcation types according to the Medina classification were included. We excluded those cases in which the procedure included two-stent technique, SB pre-dilatation or FKB. Restenosis or thrombosis of a previous stent in the treated lesion, cases not treated with a second-generation drug-eluting stent (DES) or patients with contraindications to antiplatelet therapy were also excluded. In the centre A, a systematic post-dilatation of the SB through the struts of the MV stent was performed regardless of the degree of SB stenosis or TIMI flow after its implantation. In the centre B, a selective post-dilatation of the SB after MV stenting was performed only in cases with a suboptimal result in the SB, according to the operator's decision ( Fig. 1 ). All patients received and signed an informed consent prior to the procedure. Fig. 1 Flowchart of CR12 study. PCI: percutaneous coronary intervention; SB: side branch; FKB: Final Kissing balloon; DES: drug eluting stent. 2.2 Procedure and stent implantation Patients received an antiplatelet and anticoagulant pre-treatment according to the recommendations of the clinical practice guidelines. During the procedure, unfractionated heparin (70–100 IU/kg) was used as an anticoagulant. Addition of Abciximab or use of Bivalirudin was left to the discretion of the operator. After angioplasty, double antiplatelet therapy was recommended, with indefinite aspirin and clopidogrel, prasugrel or ticagrelor for 6–12 months. During the angioplasty, avoiding pre-dilatation of the MV segments that were not going to be covered by a stent was recommended. Second generation DES or bioresorbable platforms of adequate length and diameter (according to the distal MV) were used to cover the diseased segment. The strategy of the procedure consisted of the following: crossing two guidewires to the MV and the SB, pre-dilatation of the MV (at the discretion of the operator) and implantation of the stent in the MV jailing the SB wire. Proximal optimization technique (POT) was also performed in the operator's judgment. In the centre A, the guidewires were systematically recrossed and the SB ostium was post-dilated with a small semi-compliant balloon (diameter:1.5–2 mm). If after this dilatation, a severe lesion or with deterioration of flow was maintained, a stent was implanted in the SB. In centre B, after stent implantation, it was assessed if there was angiographically significant impairment of the SB. If it was not, in the operator's opinion, the procedure was finished. If there was, guidewires were recrossed and the SB was post-dilated in a similar way to the centre A. The use of additional stents was allowed both in the MV and in the SB to completely cover the lesion to be treated. In both centres, when a second stent was implanted, FKB was mandatory. 2.3 Angiographic evaluation Quantitative coronary analysis was performed using the Xcelera® program (Philips Healthcare, Best, The Netherlands). Previously of the stent implantation we analysed: the percentage of stenosis of the proximal MV, distal MV and ostial SB, the lesion length of the MV and SB, the bifurcation angle and the TIMI flow. After the stent implantation we analysed: the percentage of stenosis of the MV and SB and the TIMI flow. Patients with an acute MV occlusion were also included, assessing the bifurcation characteristics after flow recovery with thrombectomy or balloon dilatation. 2.4 Follow-up and clinical definitions After PCI, 12-lead electrocardiograms and analytical determinations were performed including creatine-phospho-kinase (CPK), its MB fraction and troponin at 6 and 12 h after the procedure. Clinical assessment was performed after the procedure, at 1 month and 12 months. Information about major cardiovascular adverse events (MACE) was obtained through clinical visit, telephone call or review of the electronic medical records of the centres. Angiographic follow-up was only performed in those patients who underwent a new coronary angiogram due to clinical reasons (new symptoms, ischemia or deterioration of ventricular function in non-invasive tests) or those requiring a programmed PCI for other lesions than the index bifurcation. MACE [cardiovascular death, myocardial infarction (MI) related to the treated lesion, stent thrombosis (ST) or need for treated lesion revascularization (TLR)] at one-year follow-up was established as the primary endpoint. As secondary endpoints, the individual events of the composite endpoint were analysed independently. All deaths were considered cardiac unless another specific cause was documented. MI was defined according to the current recommendations and only those related to the treated lesion, whether periprocedural or during follow-up, were considered. TLR or ST were defined according to the Academic Research Consortium criteria . 2.5 Statistic analysis The number of patients included in the study was based on previous trials similar to ours . Taking into account these data, we planned to include a minimum of 120 patients in each group to test our hypothesis with a power of 80%. Quantitative variables following a normal distribution were described as mean ± standard deviation. Those that did not follow were described by the median (range). Qualitative variables were described by absolute and relative frequencies of their categories. A statistical significance level of 0.05 was considered and the 95% confidence interval of the target analysis variables was calculated. For the bivariate analysis, the Student's t -test or the non-parametric Mann-Whitney U test were used for means comparison and Chi-square or the Fisher's exact test were used for the comparison of qualitative variables. For the multivariate analysis, Cox regression was used. Variables were considered as potential predictors of risk in the multivariate model when they presented a statistically significant association in the univariate analysis. The SPSS version 20 program was used for calculations. 3 Results Of the 382 PCIs of bifurcated lesions with an SB > 2 mm, 92 were excluded before the analysis because they met at least one of the exclusion criteria. The remaining 290 patients were analysed ( Fig. 1 ). 3.1 Baseline characteristics The baseline characteristics of the patients are shown in Table 1 . 79.1% were male and mean age was 65.5 ± 12.6 years. 33.2% were diabetics. There were no significant differences in baseline characteristics between the two centres except for a higher body mass index (BMI) and greater number of patients presenting as acute coronary syndrome (ACS) at centre A ( Table 1 ). Table 1 Baseline characteristics. A: SB systematic isolated post-dilation n = 136B: No SB systematic isolated post-dilation n = 154p Age (years) 64.6 ± 12.6 66.2 ± 12.6 0.28 Male 85 (62.5%) 120 (77.9%) 0.4 BMI 28.7 ± 4 27.4 ± 4.8 0.013 Smoker 45 (33.1%) 52 (33.7%) 0.5 Diabetes mellitus 41 (30.1%) 55 (35.7%) 0.19 Dyslipidaemia 62 (45.6%) 72 (46.7%) 0.9 Hypertension 91 (66.9%) 91 (59.1%) 0.11 Clinical presentation SCAD 55 (40.4%) 79 (51.3%) 0.003 NSTEMI 41 (30.1%) 34 (22.1%) STEMI 40 (29.4%) 41 (26.6%) LVEF 52.1 ± 11.8 54.7 ± 11 0.08 BMI: body mass index. LVEF: left ventricle ejection fraction. NSTEMI: non ST-elevation myocardial infarction. SB: Side branch. SCAD: stable coronary artery disease. STEMI: ST-elevation myocardial infarction. 3.2 Procedural characteristics The majority of bifurcations were located in the Left Anterior Descending-Diagonal (53.2%), being true bifurcations 39.3% of them. The rest of procedural characteristics are shown in Table 2 . Lesion's length in centre A was significantly higher, which implied the implantation of longer stents. In addition, the angle of the bifurcations of centre A was significantly smaller and POT was performed less frequently. In centre A, zotarolimus-eluting stents were used more frequently (42.8%), while in centre B the most used were everolimus-eluting stents (48.6%) ( Table 2 ). No patient required the implantation of a second stent in the SB and those in whom an initial strategy of two stents were proposed were excluded before the analysis. Table 2 Angiographic and procedural characteristics. A: SB systematic isolated post-dilation n = 136B: No SB systematic isolated post-dilation n = 154p Location of bifurcation LM 11 (8.1%) 19 (12.3%) 0.08 LAD-Diagonal 66 (48.5%) 88 (57.1%) Cx-OM 37 (27.2%) 23 (14.9%) RCA-AM 5 (3.7%) 10 (6,5%) PD-PL 14 (10.3%) 11 (7,1%) Other 3 (2.2%) 3 (1,9%) True bifurcations (1,1,1; 1,0,1; 0,1,1) 48 (35.3%) 66 (42.8%) 0.23 Proximal MV stenosis 64 ± 35.6% 68.7 ± 28.5% 0.13 Distal MV stenosis 62.5 ± 36.3% 69.3 ± 31.4% 0.08 Ostial SB stenosis 37.2 ± 28.2% 33 ± 33.4% 0.3 MV lesion length 22.2 ± 9.8 mm 19.3 ± 7.9 mm 0.006 SB lesion length 4.5 ± 6.1 mm 5.4 ± 7.5 mm 0.26 Bifurcation angle 38.4 ± 26.2° 51.2 ± 19.2° < 0.0001 Stent diameter 3 ± 0.4 mm 2.9 ± 0.4 mm 0.47 Stent length 26.4 ± 9.4 mm 23.3 ± 8.5 mm 0.011 SB significant impairment after MV stenting a 61 (44.8%) 76 (49.3%) 0.48 POT 47 (34.5%) 83 (53.9%) 0.001 SB post-dilatation 119 (87.5%) 3 (1.9%) < 0.0001 MV stent type Everolimus 35 (25.7%) 48,6% < 0.0001 Zotarolimus 55 (40.4%) 26,7% Sirolimus 40 (29.4%) 0% Biolimus 3 (2.2%) 22,5% Bioresorbable (everolimus) 3 (2.2%) 2,1% Angiographic success 134 (98.5%) 149 (96.7%) 0.45 Fluoroscopy time 17.8 ± 9.7 min 16.1 ± 11.1 0.17 Angiographic follow up 23 (16.9%) 16 (10.4%) 0.17 AM: acute marginal; Cx: circumflex artery; LAD: left anterior descending; LM: left main; MV: main vessel; OM: obtuse marginal; PD: posterior descending; PL: posterolateral; POT: proximal optimization technique. RCA: right coronary artery; SB: side branch. a SB significant impairment after MV stenting: SB stenosis > 75% or TIMI < 3. There were no significant differences between both centres neither in the percentage of stenosis nor in the lesion length of the SB prior to the implantation of the MV stent. There were also no significant differences in the rate of SB impairment (stenosis > 75% or TIMI flow < 3) after stent implantation (A: 44.6 vs B: 49.3%; p = 0.48). SB post-dilatation was performed in 87.5% of cases in centre A (45.4% without SB involvement) and in 1.9% of cases in centre B (p < 0.0001). In centre A, only in one patient the reason because SB post-dilatation was not performed was the impossibility to recross the SB after MV stenting. In the other 16 patients in which the SB was not post-dilated, there was no severe impairment of SB and the decision was done by the main operator. In the centre B, only 3 cases underwent SB post-dilatation. All of them after an occlusive SB dissection after MV stenting. There was no significant difference in fluoroscopy time between the two centres. Procedural success rate was high and similar in both centres ( Table 2 ). 3.3 Adverse events during follow-up After one-year follow-up, a lower MACE rate was observed in centre A (A: 4.4% vs B: 10.4%, p = 0.043) as well as a trend towards lower cardiac mortality (A: 2.2% vs. B: 6.5%, p = 0.093) and ST rate (A: 0% vs B: 2.6%, p = 0.077). There were no significant differences between the two centres regarding the rate of MI related to the treated artery (A: 1.4% vs. B: 3.9%, p = 0.29) or TLR rate (A: 1.4% vs. B: 3.2%; p = 0.45) ( Table 3 ). In the centre A the three cardiac deaths were patients with extensive MI who died during the initial admission due to cardiogenic shock. In the centre B, 5 patients died due to cardiogenic shock during initial admission (two had recovered cardiac arrest), 3 due to advanced heart failure at 4, 7 and 16 months after the initial procedure, 1 due to cardiac rupture after extensive MI and 1 for a probable ST. In centre A, one MI occurred after the occlusion of a SB during the procedure that could not be recrossed, and another MI occurred in relation to a restenosis of the proximal edge of the stent that required new revascularization. In the centre B, 3 of the MI were produced during the initial procedure by occlusive dissection of the SB that was resolved after recrossing and SB post-dilatation, and in other 3 cases by ST (two acute and one late ST, eleven months after the initial procedure). In the centre A the 2 TLR cases were due to severe restenosis at the proximal edge of the stent, without involvement of the SB. In the centre B, 3 were due to ST, one to restenosis of the proximal edge of the stent without involvement of the SB and another to diffuse restenosis of the stent without involvement of the SB. Table 3 Major Adverse Cardiovascular events after one-year follow-up. A: SB systematic isolated post-dilation n = 136B: No SB systematic isolated post-dilation n = 154p Cardiac death 3 (2.2%) 10 (6.5%) 0.093 MI related to the treated lesion 2 (1.4%) 6 (3.9%) 0.29 TLR 2 (1.4%) 5 (3.2%) 0.45 ST 0 (0%) 4 (2.6%) 0.077 MACE 6 (4.4%) 16 (10.4%) 0.043 MACE: Major Adverse Cardiovascular events; MI: myocardial infarction; SB: side branch; ST: stent thrombosis; TLR: target lesion revascularization. Apart from the differential treatment strategy of the SB of each centre, the variables associated with the presentation of MACE at one year were the presence of a true bifurcation (5.1% vs 11.4%, p = 0.042), the percentage of MV stenosis (65.9 ± 32.7% vs 84.6 ± 10.4%, p < 0.0001), the percentage of initial stenosis in the ostial SB (33.8 ± 35.6% vs 49.9 ± 34.6%, p = 0.042), the SB lesion length (4.6 ± 6.4 mm vs 9.3 ± 9.8 mm, p = 0.002) and the presence of SB impairment after MV stent implantation (3.9% vs 11.8%). The multivariate analysis included all variables associated with the presentation of MACE in the univariate analysis and those that were distributed in a significantly different way in both centres. The only independent predictor of MACE at one year was the systematic post-dilatation strategy of SB [HR 0.1 (0.019–0.476); p = 0.004]. Although an adjustment for the variables that were distributed differently in both centres was made through multivariate analysis, since there is no randomization, it cannot be ruled out that other not studied factors may have influenced the result. Because of that, and trying to understand if there were significant differences between both centres, we compared the angiographic result and the MACE rate in the whole population (bifurcation and no bifurcation PCI). Analyzing the initial 1960 PCIs, there was no significant differences nor in the initial angiographic success rate (A: 96.3%; B: 97.1% p = 0.7) neither in the one-year MACE rate (A: 6.1%; B: 6.4%; p = 0.23). 4 Discussion The main finding of this study is that in PCI of coronary bifurcations performed by PST, the systematic SB post-dilatation was associated with a lower MACE rate at one-year follow-up. In addition, there was a trend towards lower cardiac mortality and ST rate and a numerical, but not statistically significant, reduction in the rate of MI related to treated lesion and TLR. There were no differences in angiographic success rate or fluoroscopy time. PST is recommended for the PCI of most bifurcated lesions. However, there are no uniform criteria to consider as significant the SB impairment after MV stenting nor on what type of intervention should be performed on it. The systematic SB post-dilatation may have potential benefits by reducing the percentage of jailing struts in the SB ostium, improving the flow dynamics towards the SB, the endothelial coverage of struts and decreasing the thrombosis and restenosis rate . Although an intracoronary analysis of these details was not performed in our study, a trend towards a reduction of ST that could be related to this fact was observed. In addition, this dilatation may facilitate the recrossing towards the SB in future procedures. Among the potential disadvantages, it should be noted that balloon dilatation of the SB in areas not covered with stents could produce acute dissections and favour restenosis and thrombosis. In our study there were no significant differences regarding TLR rate between the two groups and there was even a trend towards a lower rate of TLR and ST in the SB systematic post-dilatation group. On the other hand, the isolated post-dilatation of the SB could displace the carina towards the MV, producing asymmetry of the stent . However, it has been suggested that this asymmetry may not be related to adverse events , something that agrees with our results. In addition, the FKB technique that could solve this asymmetry has not demonstrated clear benefits in PST. Finally, another potential disadvantage of the technique is the increase in duration and complexity of the procedure. However, training in the routine implementation of this technique causes it to be performed in a few minutes and in fact, in our study, the fluoroscopy time was not significantly different in both centres. Two of the main studies that have studied the benefits of routine dilatation of SB in PST have been CORPAL Kiss and Nordic Baltic Bifurcation Study III. The CORPAL Kiss study compared in 293 patients the FKB strategy against isolated dilatation of SB in a sample similar to ours. The authors found no difference in clinical events between the two groups at one year of follow-up. In the Nordic Baltic Bifurcation Study III, the authors analysed in 474 patients the benefits of FKB technique after PST and found no clinical benefit at 6 months with FKB. However a higher rate of restenosis was observed in the non-FKB group (7.9% vs 15.4%, p = 0.0039), especially in the true bifurcations subgroup (7.6% vs 20%, p = 0.024). Patients were randomized if, after stent implantation, the SB presented a TIMI flow 3, and no intervention could be performed on the SB if they were assigned to the non-FKB group regardless of whether this vessel presented severe stenosis. In fact, SB dilatation was only performed in 1.7% of patients in the non-FKB group, and the crossover rate of the non-FKB group to the FKB group was only 0.8%. In our study, the low percentage of SB dilatation in centre B (1.9%) is also noteworthy, especially observing that 49.3% of patients presented angiographically significant impairment of the SB after stent implantation. This fact, as in the work of Niemela et al. , could have influenced the results. However, it may reflect normal practice in PST after the work of Koo et al. in which it was shown that only 27% of SB lesions with > 75% of angiographic stenosis were functionally significant. Another remarkable issue was the low POT rate in both centres. However this is somewhat logical since until the recommendations of the 11th consensus document of the European Bifurcation Club of 2016, this technique was not considered as a standard step in the PST and was reserved for cases with large differences in reference diameter between proximal and distal MV or to aid difficult recrossing into a SB . As far as we know, our work compares for the first time the systematic or selective isolated SB post-dilatation in PST. MACE rate in our study was low in both groups and consistent with that of other previous studies . This is relevant considering that the bifurcations included had a SB of sufficient entity to produce myocardial damage in case of affectation. In addition, it is noteworthy the high rate of presentation as an acute coronary syndrome (55%) with 26.8% of STEMI that were treated with primary PCI. In fact, most cardiac deaths in both centres were related to complications of extensive infarctions and the strategy on SB may not have a direct influence on them. In contrast, the Nordic-Baltic study excluded STEMI patients and only 25% of the patients had unstable angina. In addition, in that study, the MIs related to the procedure were not considered, which makes it difficult to extrapolate their results. We believe that including the MIs related to the procedure better reflects the safety of the technique and that it is more appropriate to consider the MI related to the treated artery, since considering MI related to another vessel may bias the assessment of the technique performed in the initial lesion. 4.1 Limitations It is a prospective observational registry with all the limitations that are supposed to this type of studies. Although an adjustment for the variables that were distributed differently in both centres was made through multivariate analysis, since there is no randomization, it cannot be ruled out that other not studied factors may have influenced the result. Although the operators were knowledgeable about the treatment group, two independent researchers performed follow-up and review of the events. The authors are aware that it is difficult to draw definitive conclusions due to the study design, however, we believe that this work reflects that systematic isolated post-dilatation of the SB is a new variant of PST with good initial angiographic result and a low MACE rate at one-year follow-up compared to conventional strategy. It can even pose some advantages and might be considered in some cases as a valid alternative. The exclusion of cases with SB pre-dilatation or FKB was done in order not to interfere in the evaluation of the isolated SB post-dilatation technique. This could have selected patients with lower initial SB impairment. However, there was no difference in the percentage of stenosis and SB flow after stent implantation in the MV between both groups, so we believe that this fact should not significantly interfere with the results. 5 Conclusions In the PCI of coronary bifurcations performed by PST, the systematic isolated post-dilatation of the SB achieved a high initial angiographic success rate, and was associated with a low MACE rate at one-year follow-up. This technique was not associated with greater complications during the procedure nor implied an increase in fluoroscopy time. Although the study design does not allow definitive conclusions to be drawn, this strategy could be considered a valid option in some cases or even as part of the PST in the percutaneous treatment of coronary bifurcations.