Women typically present with coronary artery disease later than men with more unfavorable clinical and anatomic characteristics. It is unknown whether differences exist in women undergoing treatment for unprotected left main coronary artery (ULMCA) disease. Our aim was to evaluate long-term clinical outcomes in women treated with percutaneous coronary intervention (PCI) with drug-eluting stents versus coronary artery bypass grafting (CABG). All consecutive women from the Drug-Eluting stent for LefT main coronary Artery disease registry with ULMCA disease were analyzed. A propensity matching was performed to adjust for baseline differences. In total, 817 women were included: 489 (59.8%) underwent treatment with PCI with drug-eluting stents versus 328 (40.2%) with CABG. Propensity score matching identified 175 matched pairs, and at long-term follow-up there were no differences in all-cause (odds ratio [OR] 0.722, 95% confidence interval [CI] 0.357 to 1.461, p = 0.365) or cardiovascular (OR 1.100, 95% CI 0.455 to 2.660, p = 0.832) mortality, myocardial infarction (MI; OR 0.362, 95% CI 0.094 to 1.388, p = 0.138), or cerebrovascular accident (CVA; OR 1.200, 95% CI 0.359 to 4.007, p = 0.767) resulting in no difference in the primary study objective of death, MI, or CVA (OR 0.711, 95% CI 0.387 to 1.308, p = 0.273). However, there was an advantage of CABG in major adverse cardiovascular and cerebrovascular events (OR 0.429, 95% CI 0.254 to 0.723, p = 0.001), driven exclusively by target vessel revascularization (OR 0.185, 95% CI 0.079 to 0.432, p <0.001). In women with significant ULMCA disease, no difference was observed after PCI or CABG in death, MI, and CVA at long-term follow-up.
Only a few reports have evaluated optimal revascularization strategies in women with coronary artery disease, who typically present later than men with potentially more co-morbidities and unfavorable angiographic characteristics. Data are even more limited on outcomes of percutaneous coronary intervention (PCI) versus coronary artery bypass grafting (CABG) in women with complex coronary anatomy, including unprotected left main coronary artery (ULMCA) disease. In general, women are largely underrepresented in randomized clinical trials; specifically, in the Synergy between Percutaneous Coronary Intervention with TAXUS and Cardiac Surgery (SYNTAX) trial, in the ULMCA subgroup, women made up 10.3% of the overall population. Registries are hence the ideal setting to appraise the comparative effectiveness of different procedures in patients who are not adequately represented in randomized clinical trials. The aim of the present study was to evaluate if women had improved long-term clinical outcomes after ULMCA revascularization using PCI with drug-eluting stents (DES) compared with CABG, from the large Drug-Eluting stent for LefT main coronary Artery disease (DELTA) registry.
The DELTA registry included consecutive “all comers” with ULMCA disease treated in 14 multinational centers, by either PCI with DES or CABG, from April 2002 to April 2006. The W-DELTA is a subset analysis focusing on women from the DELTA registry.
Patients enrolled in the registry were evaluated by a multidisciplinary team including interventional cardiologists and cardiothoracic surgeons, and the choice of technique was deemed suitable to ensure complete revascularization. The decision was based on (1) the hemodynamic state, (2) lesion characteristics, (3) vessel size, (4) co-morbidities, (5) quality of arterial and/or venous conduits for grafting, and (6) patient and/or referring physician preference. Coronary angioplasty and stent implantation, including bifurcation strategy in the case of distal disease, were performed according to the operator’s preference with the aim of complete coverage of the diseased segment.
The use of dual antiplatelet therapy was recommended for at least 12 months in all patients undergoing PCI, consisting of aspirin 100 mg/day and clopidogrel 75 mg/day or ticlopidine 250 mg twice daily. Aspirin 100 mg/day was continued indefinitely thereafter. In the Korean center, cilostazol was additionally prescribed. Information regarding compliance was obtained in all patients. Angiographic follow-up was not mandatory unless there were clinical symptoms or subjective evidence of ischemia on functional testing.
All data relating to hospital admission, procedures, and follow-up were collected and adjudicated in each center according to local policy. Full written informed consent was obtained for the procedure and for subsequent data collection.
The events analyzed during hospital stay and at clinical follow-up were death, both all-cause and cardiovascular, myocardial infarction (MI), cerebrovascular accident (CVA), target lesion revascularization (TLR), and target vessel revascularization (TVR). Major adverse cardiovascular and cerebrovascular event (MACCE) was defined as a composite of death, MI, CVA, and TVR. The primary study objective was the composite of death, MI, and CVA at long-term follow-up (1,185 days). The secondary study objectives were MACCE and each of the individual components of death, CVA, MI, and TVR at long-term follow-up.
Continuous variables are expressed as mean ± SD and were analyzed with the Student t test or Wilcoxon rank sum test depending on the variable distribution. Categorical variables were compared with the chi-square test with Yates’ correction for continuity or Fisher’s exact test, as appropriate.
Because of the nonrandomized nature of the study, to reduce the effect of treatment selection bias and potential confounding in this observational study, we performed rigorous adjustment for significant differences in the baseline characteristics of patients with propensity score matching. A propensity score was calculated by performing a parsimonious multivariate logistic regression using the following covariants: age, family history, hypertension, hypercholesterolemia, smoker, diabetes mellitus, unstable angina, left ventricular ejection fraction, chronic kidney disease, previous PCI, previous CABG, multivessel disease, right coronary artery disease, and distal disease. The C-statistic for the propensity score model was 0.77, confirming good discrimination, and the Hosmer-Lemeshow goodness of fit was 0.36, confirming good calibration. To identify matched pairs, we used the following algorithm: 1:1 optimal match with a ±0.03 caliper and no replacement. Clinical outcomes in the matched population were analyzed with Cox proportional hazards regression stratified on matched pairs. Results are reported as odds ratio (OR) with 95% confidence intervals (CI). Survival rate was recorded by Kaplan-Meier analysis and the log-rank method was used for comparison. The p for interaction between gender and revascularization modality assessed by chi-square analysis was p <0.001.
Statistical analysis was performed with Statistical Package for the Social Sciences, version 18.0 (SPSS Inc., Chicago, Illinois). A p value of <0.05 was considered statistically significant. The investigators had full access to and take full responsibility for the integrity of the data. All investigators have read and agree to the manuscript as written.
In total, 817 women were included in the W-DELTA registry, of which 489 (59.8%) underwent treatment with PCI with DES and 328 (40.2%) with CABG. The baseline clinical characteristics are illustrated in Table 1 and baseline lesion and procedural characteristics in Table 2 .
|Variable||PCI (n = 489)||CABG (n = 328)||p Value|
|Age (yrs)||67.4 ± 12.6||67.9 ± 11.6||0.562|
|Hypertension||359 (73.4)||240 (72.9)||0.111|
|Hypercholesterolemia||323 (66.1)||232 (70.5)||0.680|
|Smoker||122 (24.9)||55 (16.7)||0.041|
|Diabetes mellitus||161 (32.9)||101 (30.7)||0.504|
|Chronic kidney disease||28 (5.7)||9 (2.7)||0.044|
|Unstable angina pectoris||160 (32.7)||163 (49.5)||<0.001|
|Non–ST elevation myocardial infarction||66 (13.5)||34 (10.4)||0.181|
|ST elevation myocardial infarction||10 (2.0)||2 (0.6)||0.094|
|Previous CABG||51 (10.4)||13 (4.0)||0.001|
|Previous PCI||123 (25.2)||48 (14.6)||<0.001|
|Left ventricular ejection fraction||54.8 ± 12.2||54.5 ± 11.0||0.731|
|EuroSCORE||5.6 ± 4.0||5.4 ± 2.6||0.395|
|Variable||PCI (n = 489)||CABG (n = 328)||p Value|
|Multivessel coronary disease||381 (77.9)||310 (94.2)||<0.001|
|Right coronary artery disease||150 (30.9)||231 (73.1)||<0.001|
|Left system coronary artery disease||314 (64.2)||305 (93.0)||<0.001|
|SYNTAX score||26.8 ± 13.0||37.1 ± 12.8||<0.001|
|Distal location||280 (57.6)||185 (58.5)||0.794|
|Cutting balloon||29 (5.9)|
|Intra-aortic balloon pump||31 (8.4)||10 (14.3)||0.117|
|Intravascular ultrasound||207 (42.3)|
|Intravascular ultrasound guided||47 (9.6)|
|Intravascular ultrasound controlled||160 (32.7)|
|Mean stent diameter||3.34 ± 0.341|
|Mean stent length||20.40 ± 15.5|
|2-Stent technique||168 (34.4)|
|Mini crush||17 (3.5)|
|T stenting||28 (5.7)|
|V stenting||22 (4.5)|
|Maximum diameter||3.69 ± 0.53|
|Maximum pressure||15.43 ± 3.86|
|Final kissing balloon inflation||202 (41.3)|
|Vessels treated||1.51 ± 0.871||2.30 ± 0.86||<0.001|
|Lesions treated||1.81 ± 1.29|
|CABG beating heart||18 (5.5)|
|Mean arterial grafts||1.97 ± 1.06|
|Mean venous grafts||1.82 ± 1.23|
|Complete revascularization||276 (94.8)|
|Unintentional incomplete||1 (0.5)|
|Mean hospital stay||4.1 ± 4.0||14.5 ± 9.4||<0.001|
During hospitalization, all-cause mortality in PCI versus CABG occurred in 4.1% versus 2.7% patients and cardiovascular mortality in 3.5% versus 1.8% patients, respectively. Periprocedural MI (as defined by elevation of serum creatine kinase-myocardial band exceeding 5 times the upper reference limit) was observed in 5.9% versus 18.2% patients and CVA in 0.6% versus 1.5% patients. Overall in-hospital MACCE was 9.6% versus 22.5%. Of note, there were 4 episodes (0.8%) of in-hospital TVR in the PCI group and none in the CABG group. In those patients with distal disease treated with PCI, 55.4% underwent a single-stent strategy. The in-hospital MACCE was 9.0% in patients treated with a single-stent strategy versus 13.0% in those who underwent implantation of 2 stents.
Clinical follow-up was obtained at a median of 1,185 days (interquartile range [IQR] 628 to 1,548) in 98.8% of patients in the PCI group and 99.1% in the CABG group. With regard to all-cause mortality, this was 14.1% versus 7.0%, and cardiovascular mortality was 7.0% versus 4.6% in the PCI versus CABG group, respectively. Elective PCI mortality rate was 11.9% versus 20.2% for urgent PCI cases. With regard to MI, such an event was reported in 4.3% versus 1.5%, with TLR 10.2% versus 3.8%, and TVR rates 15.1% versus 5.1%. MACCE was adjudicated at 30.5% versus 15.7% at long-term follow-up. In patients with distal disease treated with a single-stent strategy, the long-term MACCE was 28.4% versus 35.2% in those requiring 2 stents. Figure 1 illustrates survival curves. Furthermore, definite stent thrombosis (ST) occurred in 6 of the women (1.2%) treated with PCI: 2 subacutely and 4 late. Probable ST was adjudicated in 4 (0.8%) and possible in 3 patients (0.6%).
After propensity score matching, there were 175 matched pairs of patients in both treatment groups. The baseline characteristics of the matched groups are listed in Table 3 .
|Variable||PCI (n = 175)||CABG (n = 175)||p Value|
|Age (yrs)||67.1 ± 12.0||67.5 ± 10.3||0.736|
|Hypertension||120 (68.6)||125 (71.0)||0.617|
|Hypercholesterolemia||117 (66.9)||127 (72.2)||0.281|
|Smoker||34 (19.4)||33 (18.8)||0.872|
|Diabetes mellitus||50 (28.6)||55 (31.3)||0.584|
|Chronic kidney disease||9 (5.1)||5 (2.8)||0.271|
|Unstable angina pectoris||72 (41.1)||75 (42.6)||0.780|
|Previous PCI||35 (20.0)||31 (17.6)||0.567|
|Left ventricular ejection fraction||55.2 ± 11.9||54.3 ± 11.0||0.445|
|EuroSCORE||5.1 ± 2.5||5.6 ± 4.2||0.270|
|Multivessel disease||160 (91.4)||162 (92.0)||0.834|
|Right coronary artery disease||104 (59.1)||104 (59.1)||1.000|
|Distal location||92 (52.6)||101 (57.4)||0.365|
|Intra-aortic balloon pump||13 (8.8)||2 (5.9)||0.579|
|SYNTAX score||26.6 ± 11.1||34.0 ± 13.5||<0.001|