Methods

The K-CTO registry is a retrospective cohort of 26 centers in South Korea. From January 1, 2007, to December 31, 2009, we enrolled patients treated with an SES (Cypher; Cordis Johnson & Johnson, Miami, Florida) or a PES (Taxus; Boston Scientific Corp., Natick, Massachusetts, or Pico Elite; AMG International GmbH, Raesfeld-Erle, Germany). From January 1, 2007 to December 31, 2011, we enrolled patients treated with an EES (Xience; Abbott Vascular, Santa Clara, California; or Promus; Boston Scientific Corp.). This was an all-comer analysis that included patients with stable angina, unstable angina, non–ST-segment elevation myocardial infarction (STEMI), and STEMI and those who underwent emergent percutaneous coronary intervention (PCI). This study protocol was approved by the institutional review board at each participating center and conducted according to the principals of the Declaration of Helsinki. All patients gave informed consent to participate in the study.

CTO was defined as a complete coronary obstruction with Thrombolysis In Myocardial Infarction flow grade 0 with an estimated duration ≥3 months with or without visible collateral flow, whether anterograde or retrograde. The duration of the occlusion was determined by the interval from the last episode of acute coronary syndrome, or in patients without a history of acute coronary syndrome, from the first episode of effort angina consistent with the location of the occlusion, or by a previous coronary angiography. All occlusions were located in a native vessel.

We established 2 study end points: primary and secondary end points. The primary end point was major adverse cardiovascular events (MACE) defined as a composite of cardiac death (CD), nonfatal MI, and target lesion revascularization (TLR). The secondary end points included individual components of MACE, all-cause mortality, target vessel revascularization (TVR), and stent thrombosis (ST).

All deaths were considered to have been from a cardiac cause unless a noncardiac origin was definitely documented. MI was defined according to the recommendations of the ESC/ACCF/AHA/WHF task force. TLR was defined as any repeat percutaneous intervention of the target lesion or bypass surgery of the target vessel performed for restenosis of the target lesion. TVR was defined as any repeat percutaneous intervention or surgical bypass of any segment of the target vessel. The presence of ST was assessed by the Academic Research Consortium definitions, and “definite or probable” categories were used to define ST.

All PCI procedures were performed using standard techniques. Various guidewires were used to cross the lesions. After pre-dilation, stents were deployed, and if necessary, adjunctive high-pressure balloon dilation was performed to achieve angiographic optimization. All patients were pretreated with aspirin and clopidogrel. Unfractionated heparin was administered either before or during the intervention to achieve adequate anticoagulation. The selection of equipment for the angioplasty, including the choice of DES, was decided by the operator during the procedure. The use of a glycoprotein IIb/IIIa inhibitor was also left to the operator’s discretion. After the procedure, all patients were given aspirin (at least 100 mg/d) indefinitely and clopidogrel (75 mg/d) for at least 12 months.

After index PCI, follow-ups were recommended at 1, 6, 9, and 12 months. Routine follow-up angiography was optional at 6 to 9 months. Clinical, angiographic, procedural, and outcome data were collected by independent nurses and researchers who were unaware of the purpose of the study. Patient data were retrospectively reviewed through electronic medical records.

The analysis was performed in 2 parts. First, analysis and comparison of the primary and secondary end points were conducted in the whole population. Second, comparison of the primary and secondary end points was re-analyzed by an inverse probability-weighted Cox proportional hazard model as a sensitivity analysis. Baseline and angiographic characteristics among the patients treated with the SES, the EES, or the PES were compared using a 1-way analysis of variance for continuous variables and the chi-square test for categorical variables. Cumulative event curves were plotted by means of the Kaplan-Meier method, and the differences among the 3 stent groups were assessed with the log-rank test. To reduce the impact of treatment selection bias, the difference in baseline characteristics and potential confounding bias, which were inherited limitations of an observational registry based study, we performed rigorous adjustment for the differences in baseline characteristics of patients using weighted Cox proportional hazards regression models with inverse probability of treatment weighting. In brief, propensity scores, which were the probabilities that patients received the EES, were calculated by a multiple logistic regression model, based on the 16 covariates. The adjusted variables of baseline characteristics were as follows: age (continuous), gender (men or women), hypertension (yes or no), diabetes mellitus (DM) (yes or no), smoking (current or noncurrent), dyslipidemia (yes or no), clinical indication of PCI (stable angina pectoris, unstable angina pectoris, NSTEMI, STEMI), left main (LM) disease (yes or no), number of disease vessel (1 vessel disease [VD], 2VD, and 3VD), lesion of CTO (left anterior descending [LAD], left circumflex [LC], right coronary, and LM), previous PCI (yes or no), previous coronary bypass (yes or no), previous MI (yes or no), history of congestive heart failure (CHF) (yes or no), total stent length in CTO lesion (continuous), and mean stent diameter in CTO lesion (continuous). After creating the inverse probability of treatment weighting, an inverse probability-weighted Cox proportional hazard model was fitted to compare the primary and secondary end points of the 3 groups. In addition, an inverse probability-weighted Cox proportional hazard model was used to identify independent predictors of the primary clinical outcome, MACE.

Additionally, pooled analysis for MACE was done to enhance statistical power for the rare clinical events. The pooled odds ratio was calculated with the methods by DerSimonian and Laird for random effects and Mantel-Haenszel for fixed effects. Heterogeneity was measured using the Cochran Q through the chi-square test and quantified using the I ² test. All statistical analyses were performed using SAS, version 9 (SAS Institute, Inc., Cary, North Carolina) and R programming, version 3.0.2 (The R Foundation for Statistical Computing, Vienna, Austria) using the meta-for and meta-commands. A 2-sided p value <0.05 was considered to be statistically significant.