Evidence regarding therapy with drug-eluting stents in the left main coronary artery (LM) is based mostly on trials performed with first-generation drug-eluting stents. The aim of this study was to evaluate long-term clinical outcomes after treatment for unprotected LM disease with paclitaxel-eluting stents (PES) and everolimus-eluting stents (EES). The ESTROFA-LM is a multicenter retrospective registry including consecutive patients with unprotected LM disease treated with PES or EES. A total of 770 patients have been included at 21 centers, 415 with treated PES and 355 with EES. Treatment with 2 stents was more frequent with PES (17% vs 10.4%, p = 0.007), whereas intravascular ultrasound was more frequently used with EES (35.2% vs 26%, p = 0.006). The 3-year death and infarction survival rates were 86.1% for PES and 87.3% for EES (p = 0.50) and for death, infarction, and target lesion revascularization were 83.6% versus 82% (p = 0.60), respectively. Definite or probable thrombosis was 1.6% for PES and 1.4% for EES (p = 0.80). The use of 2 stents, age, diabetes, and acute coronary syndromes were independent predictors of mortality. In the subgroup of distal lesions, the use of intravascular ultrasound was an independent predictor of better outcome. Comparison of propensity score–matched groups did not yield differences between the 2 stents. In conclusion, the results of this multicenter registry show comparable safety and efficacy at 3 years for PES and EES in the treatment of LM disease. The use of bifurcation stenting techniques in distal lesions was a relevant independent predictor for events. The use of intravascular ultrasound appears to have a positive impact on patients treated for LM distal disease.
Evidence supporting percutaneous revascularization of unprotected left main coronary artery (LM) lesions is based primarily on results obtained with first-generation drug-eluting stents (DES), particularly with paclitaxel-eluting stents (PES). Second-generation DES, specifically everolimus-eluting stents (EES), have shown a superior efficacy and safety profile to PES, in clinical trials and in numerous registries. Therefore, it is crucial to assess outcomes with these new DES in LM lesions. There is not much information regarding second-generation DES in LM disease. A limited-size registry with medium-term follow-up has been published, which offers promising outcomes for EES. The objective of this multicenter study was to compare the long-term outcome of 2 large series of patients with LM disease treated with PES and EES in real-world practice.
Methods
This was a multicenter retrospective registry involving 21 centers throughout Spain. The study was acknowledged and supported by the Interventional Cardiology Spanish Working Group of the Spanish Society of Cardiology. This study is part of the ESTROFA Project and Study Network.
At each participating center, 2 groups of 10 to 30 consecutive patients were included, treated with PES or with EES in LM lesions. No cases with LM stenting as part of the treatment of the ostial left anterior descending or circumflex coronary arteries were included. To reduce relevant selection biases, the series of PES patients were obtained from the period before the introduction of EES at each center. Likewise, to allow the longest possible follow-up in the series with EES, the first patients treated with these stents were included. In any case, these should always be consecutive patient groups, and each center contributed with the 2 groups of stents. There were no clinical or angiographic exclusion criteria, except for the presence of cardiogenic shock at the time of the procedure. Patients who underwent treatment of lesions in other vessels were also included. Investigators were encouraged to include all consecutive cases. The prescribed time period of dual-antiplatelet therapy was systematically ≥12 months. All clinical, angiographic, and procedural data were recorded in a common database specifically designed for this study. At the same time, all information about clinical follow-up was also submitted, and this was adequately updated through registry and hospital database reviews, as well as through contact with patients by phone and clinical visits.
The primary objective of the study was to compare event-free survival (death, myocardial infarction, and target lesion revascularization [TLR]) at 3 years between PES and EES. Secondary end points included survival free of death and infarction, survival free of cardiac and infarction, survival free of TLR, and definite or probable incidence of thrombosis.
The following major adverse cardiac events were defined: all-cause death and cardiac death (including any sudden death by undefined cause); ST-segment elevation myocardial infarction, defined as a typical increase and gradual decrease (troponin) or as a faster increase and decrease (creatine kinase-MB) of biochemical markers of myocardial necrosis, along with ST-segment elevation or the development of pathologic Q waves on electrocardiography or pathologic findings of transmural acute myocardial infarction; and non–ST-segment elevation myocardial infarction, defined as a typical increase and gradual decrease (troponin) or as a faster increase and decrease (creatine kinase-MB) of biochemical markers of myocardial necrosis with ≥1 of the following: ischemic symptoms, electrocardiographic changes indicating ischemia (ST-segment depression or T-wave changes) or pathologic findings of nontransmural acute myocardial infarction. TLR was defined as revascularization for LM restenosis (>50%), also including proximal or distal segments (5 mm) adjacent to the stent or stents used for treatment of the lesion and also the first distal 5 mm to the ostial left anterior descending or circumflex coronary artery. Therefore, this would consider TLR to be an intervention on a restenotic lesion in the distal edge of a stent in the LM that extended to the proximal left anterior descending coronary artery or on the proximal or distal edge of a stent implanted in the proximal left circumflex coronary artery. Any surgical revascularization due to restenosis as previously defined was also considered TLR. All revascularizations were defined as any revascularization performed in follow-up, including those over target and nontarget lesions. Definite and probable stent thrombosis were considered according to the definitions of the Academic Research Consortium. Definite stent thrombosis was determined when confirmed by angiography or when pathologic confirmation of acute thrombosis in patients with acute coronary syndromes was made. Probable stent thrombosis was defined as any unexplained death within 30 days or as target vessel infarction without angiographic confirmation of thrombosis or other identified culprit lesion.
The participating centers were requested to provide detailed description for the major events detected during follow-up. Although the aforementioned definitions for events were accepted and applied by all investigators, the final adjudication of events was conducted independently by 2 blinded independent investigators at the ESTROFA study coordination center (Hospital Universitario Marques de Valdecilla, Santander, Spain).
Continuous variables are presented as mean ± SD. Categorical variables are expressed as percentages. Continuous variables were compared using Student’s t tests if they followed a normal distribution and Wilcoxon’s tests when they did not (type of distribution was assessed using the Kolmogorov-Smirnov test). Categorical variables were compared using chi-square tests or Fischer’s exact tests, according to indication. Kaplan-Meier curves for event-free survival were obtained for each group or subgroup considered in the analysis and compared using log-rank tests. Cox proportional-hazards multiple regression analysis was used to determine those independent predictor variables of major cardiac adverse events during the follow-up period. The model included all variables that showed associations with the incidence of major adverse cardiac events in univariate analysis with p values <0.10. The use of intravascular ultrasound (IVUS) was included in the risk analysis, but only 26% to 35% of patients underwent IVUS investigation, which suggests the possibility of bias. We performed adjustment for differences in clinical, angiographic, and procedural characteristics using propensity score matching. The propensity scores were estimated using logistic regression in which treatment assignment was used as the outcome variable and the covariates were used as predictors. The factors included in the baseline propensity score model were age, gender, smoking habit, diabetes, hypertension, hypercholesterolemia, chronic renal failure, ejection fraction, previous infarction, previous percutaneous coronary intervention (PCI), previous coronary artery bypass graft surgery, stable angina, unstable angina, ST-segment elevation myocardial infarction, non–ST-segment elevation myocardial infarction, number of vessels with stenosis >50%, ostial LM lesion, midshaft LM lesion, distal LM lesion, diffuse lesion, ulceration, calcified, LM visual stenosis, LM stent length, LM stent diameter, complex technique (2 stents), PCI in non-LM lesions, number of lesions treated, number of stents implanted, IVUS use, and administration of glycoprotein IIb/IIIa inhibitors. A propensity score value, indicating the predicted probability of receiving a specific DES, was then calculated from the logistic equation for each patient. Discrimination and calibration ability of each propensity score model was assessed using the C-statistic and the Hosmer-Lemeshow statistic. Patients were then matched according to their individual propensity scores on a 1:1 “nearest neighbor” basis. To exclude bad matches, we imposed a caliper of 0.2 of the standard deviation of the logit of the propensity score. We planned in advance to evaluate the influence of IVUS guidance on outcomes. A p value <0.05 was considered statistically significant. SPSS version 15.0 (SPSS, Inc., Chicago, Illinois) was used.
Results
A total of 770 patients were included: 415 treated with PES and 355 with EES. Clinical characteristics are listed in Table 1 . The patient profile indicates a real-world practice population with diverse clinical presentation. The arm treated with EES included more patients with diabetes and more patients with previous PCI, with the rest of baseline characteristics being similar. Angiographic and procedure characteristics are listed in Table 2 . Treatment with 2 stents was not common in any group, but it was more frequent with PES. In contrast, IVUS was more frequently used with EES. IVUS was primarily indicated for lesion severity assessment (10.7%) or PCI guidance (89.3%). After IVUS examination, additional postdilatation was done (40.8%) or new stents were implanted (5.6%).
| Variable | PES (n = 415) | EES (n = 355) | p Value |
|---|---|---|---|
| Age (yrs) | 69.4 ± 11 | 68.1 ± 12 | 0.10 |
| Women | 88 (21.2%) | 81 (22.8%) | 0.70 |
| Current smokers | 140 (33.7%) | 121 (34%) | 0.90 |
| Diabetes mellitus | 130 (31.3%) | 151 (42.5%) | 0.002 |
| Hypertension ∗ | 262 (63.1%) | 245 (69%) | 0.10 |
| Hyperlipidemia † | 224 (53.9%) | 207 (58.3%) | 0.30 |
| Chronic renal failure | 31 (7.5%) | 30 (8.5%) | 0.70 |
| Ejection fraction (%) | 54 ± 14 | 54.5 ± 15 | 0.90 |
| Previous myocardial infarction | 132 (31.8%) | 106 (30%) | 0.60 |
| Previous angioplasty | 75 (18%) | 97 (27.3%) | 0.003 |
| Previous coronary bypass | 31 (7.5%) | 20 (5.6%) | 0.30 |
| Stable angina pectoris | 127 (30.6%) | 108 (30.4%) | 0.90 |
| Non–ST-segment elevation myocardial infarction | 102 (24.6%) | 97 (27.3%) | 0.40 |
| ST-segment elevation myocardial infarction | 32 (7.7%) | 22 (6.2%) | 0.50 |
∗ Systolic pressure >140 mm Hg or diastolic pressure >90 mm Hg, physician-documented history of hypertension, or current use of antihypertensive medications.
† Physician-documented history of hyperlipidemia or current use of lipid-lowering medications.
| Variable | PES (n = 415) | EES (n = 355) | p Value |
|---|---|---|---|
| Number of narrowed coronary arteries ∗ | |||
| 3 | 151 (36.4%) | 137 (38.6%) | 0.60 |
| 2 | 129 (31%) | 125 (35.2%) | 0.30 |
| 1 | 88 (21.2%) | 59 (16.6%) | 0.10 |
| Ostial lesion | 100 (24%) | 92 (26%) | 0.60 |
| Midshaft lesion | 95 (23%) | 75 (21%) | 0.50 |
| Distal lesion | 221 (53.3%) | 188 (53%) | 0.90 |
| Diffuse lesion | 66 (16%) | 64 (18%) | 0.50 |
| Ulceration | 46 (11%) | 28 (8%) | 0.20 |
| Calcified | 174 (42%) | 172 (48.4%) | 0.10 |
| LM visual stenosis (%) | 73 ± 16 | 74.5 ± 15 | 0.20 |
| LM stent length (mm) | 16.2 ± 6 | 16.1 ± 5.5 | 0.80 |
| LM stent diameter (mm) | 3.6 ± 0.5 | 3.5 ± 0.4 | 0.80 |
| Complex technique (2 stents) | 71 (17%) | 37 (10.4%) | 0.007 |
| 2 stents/distal lesion | 32% | 19.7% | 0.008 |
| Crush technique | 39.4% | 54% | 0.20 |
| Second stent length (mm) | 16.7 ± 5 | 16 ± 4 | 0.40 |
| Second stent diameter (mm) | 3 ± 0.4 | 3 ± 0.4 | 1.00 |
| Angioplasty in non-LM lesions | 282 (68%) | 246 (69.3%) | 0.80 |
| Lesions treated | 1.9 ± 1.1 | 2 ± 1.1 | 0.20 |
| Stents implanted | 2.3 ± 1.2 | 2.2 ± 1.1 | 0.20 |
| IVUS | 108 (26%) | 125 (35.2%) | 0.006 |
| Glycoprotein IIb/IIIa inhibitors | 72 (17.3%) | 46 (13%) | 0.90 |
| Angiographic success | 408 (98.3%) | 348 (98%) | 0.90 |
| Dual-antiplatelet therapy ≥-12 months | 415 (100%) | 355 (100%) | 1.00 |
∗ Presence of lesions with ≥50% stenosis in vessels >2 mm other than the LM.
Curves for event-free survival are shown in Figures 1 and 2 . No significant differences were observed between groups in terms the primary objective and secondary end points. The rates of survival free of major adverse cardiac events at 3 years for the 2 groups are listed in Table 3 . The incidence of definite or probable thrombosis at 3 years was comparable, 1.6% with PES and 1.4% with EES (p = 0.80), with a 62% rate of probable thrombosis. As listed in Table 4 , Cox regression analysis showed that age, diabetes, acute coronary syndromes, and the use of 2 stents were independent predictors of major adverse cardiac events.
| Variable | PES (n = 415) | EES (n = 355) | p Value |
|---|---|---|---|
| Death and infarction | 86.1 ± 1.7% | 87.3 ± 2.3% | 0.50 |
| Cardiac death and infarction | 91.8 ± 1.4% | 91.3 ± 1.8% | 0.90 |
| TLR | 96 ± 0.9% | 94 ± 1.5% | 0.10 |
| All revascularizations | 84 ± 1.5% | 86 ± 2.5% | 0.10 |
| Death, infarction, and TLR | 83.6 ± 1.8% | 82 ± 2.6% | 0.60 |
| Predictor | Hazard Ratio | 95% Confidence Interval | p Value |
|---|---|---|---|
| Overall population | |||
| 2-stent technique | 2.1 | 1.3–3 | 0.001 |
| Age | 1.04 | 1.02–1.06 | 0.0001 |
| Acute coronary syndromes | 1.8 | 1.1–2.7 | 0.01 |
| Diabetes | 1.7 | 1.14–2.2 | 0.006 |
| Distal lesions subgroup | |||
| 2-stent technique | 2.2 | 1.3–3.5 | 0.001 |
| Age | 1.03 | 1.01–1.06 | 0.006 |
| Diabetes | 1.6 | 1.05–2.6 | 0.03 |
| IVUS | 0.5 | 0.23–0.99 | 0.04 |
Exploratory analysis was conducted in 3 subgroups: patients with distal lesions, patients with diabetes, and patients without multivessel disease. In the first group, the use of 2 stents showed a significantly worse evolution in terms of death and infarction as well as TLR ( Figure 3 ). The incidence of definite or probable thrombosis was higher in this group (4.8% with 2 stents and 1.4% with 1 stent, p = 0.06). There were no differences between PES and EES in bifurcations treated with 2 stents. In terms of the diabetic subgroup, no differences between groups were observed ( Figure 4 ). In the subgroup without multivessel disease, no differences were found for the primary objective (event-free survival 89% with EES and 83.5% with PES, p = 0.20). The use of IVUS was associated with better outcomes, but it was not independent predictor in the multivariate analysis for the overall group. Survival curves in the subgroup with distal lesions and in the subgroup with bifurcations treated with two stents showed wide differences whether IVUS was used or not ( Figure 5 ). The multivariate analysis in the subgroup with distal LM lesions identified IVUS as an independent predictor for major adverse cardiac events, together with diabetes, age, and the use of 2 stents ( Table 4 ).
