We included all consecutive patients who underwent CTO percutaneous coronary intervention (PCI) with successful wire crossing of the occlusion, and a SB ≥ 2mm taking off at the proximal or distal cap, or within the occluded segment. Proximal and distal SB were included when the distance between the SB and the occluded segment was ≤ 5mm. The procedures were performed by experienced CTO-PCI operators at the 4 participating centers between January 2012 and June 2016. A total of 922 CTO in 905 patients were recanalized. Of these, 267 CTO in 267 patients involved a bifurcation lesion (29.0%). Twenty-nine (11%) patients were excluded due to the impossibility of wiring the SB before or after main vessel stent implantation despite the operator's intention. The remaining 238 patients (26.2%) were treated by the simple approach (n=201) or complex strategy (n=37). The study flowchart is shown in Figure 1. Written informed consent for treatment and data analysis was obtained from all patients.

Figure 1.

Flowchart of the study. CTO, coronary chronic total occlusion; SB, side branch.

(0.1MB).

The decision to use an antegrade or retrograde approach and the CTO crossing strategy was at the operator's discretion after a thorough study of the CTO anatomy using simultaneous double injection when applicable. Bifurcation lesions were divided into 3 types regarding the SB take-off from the main vessel: bifurcation lesions within the occluded segment, those located at the distal cap, and those at the proximal cap. The type of bifurcation treatment was also at the discretion of the operator.

The patients were pretreated with dual antiplatelet medication. In the cardiac catheterization laboratory, weight-adjusted heparin was administered to maintain an activated clotting time for > 300seconds and was monitored every 30minutes to determine whether an additional bolus of unfractionated heparin was necessary. After the procedure, all patients received 100mg of aspirin daily indefinitely and a maintenance dose of clopidogrel (75mg/d), prasugrel (10mg/d) or ticagrelor (90mg twice daily) for 6 to 12 months. In all patients, serial determinations of troponin levels were performed before and every 6hours after the procedure for the first 24hours.

Quantitative coronary analysis was performed before and after the procedure using the dedicated bifurcation software CAAS 5.11 (Pie Medical Imaging BV, Maastricht, The Netherlands). The following parameters were measured on the main vessel: reference vessel diameter, occlusion segment length, lesion segment length, final minimal lumen diameter, and final percentage of stenosis. In the SB, the parameters obtained were the reference diameter, minimal lumen diameter and percentage of baseline stenosis, final minimal lumen diameter, and final percentage of stenosis.

Coronary chronic total occlusion was defined as a 100% stenosis with thrombolysis in myocardial infarction (TIMI) flow grade 0 with an estimated duration of more than 3 months.9 The J-CTO score10 was calculated for each lesion. The baseline bifurcation anatomy was assessed according to the Medina classification.11 In SB located within the occluded segment or distally, the presence of ostium disease was studied by analyzing the filling of the SB by collaterals. True bifurcations were considered: (1,1,1), (0,1,1), and (1,0,1) of the Medina classification. Technical success of CTO recanalization was defined as a residual stenosis < 30% with TIMI flow grade 3 in the main vessel.12 Bifurcation technical success was considered to occur when a residual stenosis of < 30% in the main vessel and a final TIMI flow grade 3 in both branches were obtained.13 Procedural success was defined as angiographic success plus the absence of in-hospital adverse events (all-cause death, myocardial infarction [MI], stroke, recurrent angina requiring target-vessel revascularization with PCI or coronary artery bypass grafting, and tamponade requiring pericardiocentesis or surgery).12

Major adverse cardiac events (MACE) on follow-up were defined as a composite of cardiac death, MI, and clinically-driven target lesion revascularization. All deaths without a clear noncardiac cause were considered to be cardiac deaths. Periprocedural MI was defined as elevation of cardiac troponin values (> 5 × 99th percentile upper-reference limit) in patients with normal baseline values (≤ 99th percentile upper-reference limit) or as a rise in cardiac troponin values > 20% if baseline values were elevated and had been either stable or falling.14 Definite or probable stent thrombosis was adjudicated according to the Academic Research Consortium criteria.15

Clinical follow-up was performed by means of a review of hospital records, outpatient visit, or phone calls. Patients with symptom recurrence or with inducible ischemia were recommended to undergo angiographic evaluation.

Continuous variables are expressed as the mean ± standard deviation, if normally distributed, or the median [interquartile range: IQ25-75] if the distribution was nonnormal and were compared using the Student t test or the Mann-Whitney U test, respectively. Categorical variables are presented as counts and percentages and were compared using the chi-square test or the Fisher exact test, as appropriate. Major adverse cardiac events free survival was analyzed by the Kaplan-Meier method, and differences between groups were evaluated with the log-rank test. To account for selection bias between the simple and complex group, propensity score adjustment was performed using a genetic search algorithm and 1:1 matching. The propensity score was estimated with multivariable logistic regression for complex treatment probability including clinical, angiographic, and procedural variables potentially associated with complex PCI and the primary endpoint. The following covariates were included in the propensity score calculation: age, diabetes, SB diameter, true bifurcation lesions, dissection affecting bifurcation point, bifurcation location, and J-CTO score. Standardized differences were calculated for all covariates before and after matching to assess balance after matching (Figure 2). All statistical analyses were performed using SPSS version 19 and R version 3.4.0 for Windows. A P value < .05 was considered statistically significant.

Figure 2.

Standardized mean differences (SMDs) of the covariates used for propensity score modeling before and after adjustment. After adjustment, all covariates showed SMDs within the 10 cutoff (red vertical lines). PSM, propensity score modeling; SB, side branch.

(0.11MB).

The clinical characteristics of the study population are reported in Table 1. No differences were observed with regard to cardiovascular risk factors, chronic kidney disease, stroke, peripheral arterial disease, and prior MI, PCI, or coronary artery bypass grafting. However, the indication of CTO-PCI for acute coronary syndrome was more frequent in the simple approach group.

There were no differences between groups in terms of the target vessel CTO, J-CTO score, or location of the bifurcation lesion (Table 2). However, in the complex group, the SB was larger (2.53 ± 0.38 vs 2.30 ± 0.29mm; P < .01) and a true bifurcation was also encountered more frequently (92% vs 43%; P < .01) (Table 2). Additionally, after PCI, quantitative coronary data showed a larger minimal lumen diameter and lower percentage of stenosis at the SB in the complex group.

The procedural characteristics are summarized in Table 3. Recanalization techniques were similar between groups, with a predominance of antegrade techniques in both groups (78.6% vs 78.4%; P=ns). In the complex group, we observed a higher incidence of dissection affecting the bifurcation (48.6% vs 25.9%; P < .05) (Figure 3). Procedural metrics (contrast volume, radiation dose, and fluoroscopy time) were lower in the simple group.

Figure 3.

Left anterior descending artery CTO with a significant diagonal branch in the proximal cap (A) and distal filling by collaterals from the RCA (B). C: retrograde approach through a septal channel. D: retrograde guidewire directly crossed to the true lumen and introduced into the guide catheter. E: after predilation, an important dissection crossing bifurcation site was observed. F: this fact, together with the existence of a large lesion in the SB, dictated the strategy and a minicrush was performed with a good angiographic result. G: assessment of the enhancement stent visualization. CTO, coronary chronic total occlusion; RCA, right coronary artery; SB, side branch.

(0.18MB).

Regarding the type of 2-stent technique, T-stenting was performed in 17 (45.9%) patients, minicrush in 10 (27%), culotte technique in 9 (24.3%), and V-stenting in 1 (2.7%) patient.

Technical success of CTO was achieved in all patients. Bifurcation technical success was 94.5% in the simple and 97.3% in the complex groups, respectively (P=.48). In all cases, bifurcation technical failure was due to a TIMI flow at the SB < 3 and the SBs were mainly located in the occluded segment or in the distal cap (91.7%). Procedural success was also similar between groups (85.6% vs 81.1%; P=.49).

In the group of patients excluded due to the impossibility to wire the SB, TIMI flow at the SB was < 3 in 79.3% of them.

The incidence of procedural complications was higher in the complex group (24.3% vs 11.4%; P < .05). In this group, 7 patients had a non-Q periprocedural MI and 2 had contrast-induced nephropathy, while in the simple group 19 patients had a non-Q periprocedural MI, another had perforation with cardiac tamponade and early stent thrombosis 3 days later, and 4 developed contrast-induced nephropathy (Table 4).

Among the 29 excluded patients (no wiring the SB despite the operator's intention), the procedural complications rate was 27.6%: 7 patients (24.1%) developed a non-Q MI and another (3.4%) had a cardiac tamponade.

Follow-up was available for 236 patients (99%). The median follow-up was 25 months (interquartile range, 14-38). There were no significant differences in the unadjusted rates of events between groups. In particular, the MACE rate was 8% in the simple group and 10.8% in the complex group (P=.58) (Table 4). Kaplan-Meier curves of MACE-free survival at 3 years of follow-up are shown in Figure 4. There was a trend toward lower MACE-free survival in the complex group (80.1% vs 69.8%, P=.08).

Figure 4.

Unadjusted Kaplan-Meier curves of MACE-free survival in patients treated with the simple and complex strategy. MACE, major adverse cardiac events.

(0.1MB).

After propensity score matching, a total of 37 matched pairs were generated. After adjustment, all covariates showed standardized mean differences within the 10% cutoff, except dissection affecting bifurcation (10.7%) (Figure 2). The propensity score model showed the appropriate goodness-of-fit (Hosmer-Lemeshow P=.22). There were no differences in the clinical, angiographic, or procedural characteristics in the propensity score-matched population (Table 1, Table 2, and Table 3). Technical and procedural success were similar in the simple and complex groups (89.2% vs 97.3%; P=ns; and 83.8% vs 83.8%; P=ns). However, a complex strategy was associated with increased fluoroscopy time, X-ray dose, and contrast volume (Table 3). The incidence of periprocedural complications was 10.8% in the simple group and 24.3% in the complex group (P=.18) (Table 4).

After a median follow-up of 25 months, MACE rates did not differ between groups in the propensity score-matched cohort (13.5% vs 10.8%; P > .99) (Table 4). As in the overall analysis, MACE-free survival at 3 years was similar (78.3% vs 69.8%; P=.28) (Figure 5).

Figure 5.

Adjusted Kaplan-Meier curves of MACE-free survival at 3 years of follow-up in patients treated with the simple and complex strategy. MACE, major adverse cardiac events.

(0.11MB).

The present study has several limitations. First, this is not a randomized controlled trial. Despite the propensity score matching, we cannot rule out the effect of residual confounders. Second, the number of patients treated with the complex strategy was limited compared with those treated with the simple strategy and therefore the number of pairs generated was small, limiting the statistical power of the study. Therefore, the lack of difference in clinical outcomes between groups might reflect a type II error. Third, angiographic analyses were not performed by a core laboratory but by an experienced interventional cardiologist. Finally, our findings might not be directly extrapolable to other institutions without interventional cardiologists experienced in CTO-PCI.