Methods
Study Design and Eligibility Criteria
The design and methods of the SIRTAX LATE study have been reported previously. A total of 1,012 patients with ≥1 lesion in a vessel with a reference diameter between 2.25 and 4.00 mm were randomly assigned to treatment with sirolimus-eluting stent (SES) (Cypher; Cordis, Johnson & Johnson, Miami Lakes, FL) or paclitaxel-eluting stent (PES) (Taxus; Boston Scientific, Natick, MA) without limitations on the number of lesions or vessels. The study complied with the Declaration of Helsinki and was approved by the institutional ethics committees at the University Hospital Bern in Zurich, Switzerland. All patients provided written, informed consent.
Data Collection and Clinical and Angiographic Follow-up
Adverse events were assessed in-hospital at 1, 6, and 9 months and on an annual basis up to 5 years. All patients were asked to return for repeat angiography at 8 months. The results of the primary clinical end point at 9 months and the principal angiographic end point at 8 months have been previously reported. All patients who had at least 1 study lesion without intervening revascularization were invited to undergo an angiographic study between 4 and 5 years of follow-up.
All patients who had at least 1 lesion without intervening revascularization during long-term follow-up were invited to undergo repeat angiography at 5 years of follow-up. Patients undergoing clinically indicated revascularization of the target lesion beyond 8 months (time point of the first angiographic follow-up) were included into the angiographic long-term cohort and contributed to the assessment of delayed late loss at 5 years.
All patients were advised to take acetylsalicylic acid indefinitely and clopidogrel for the duration of 1 year.
Study End Points and Definitions
Diabetes was defined by the presence of antidiabetic medical therapy at baseline.
An independent clinical events committee unaware of the patients' assignments adjudicated all clinical end points. The primary end point was a composite of major adverse cardiac events (MACEs): cardiac death, MI, and ischemia-driven target lesion revascularization (TLR) at 9 months. Secondary end points included the individual components of MACE as well as overall mortality, any TLR, target vessel revascularization (TVR), target vessel failure (composite of cardiac death, MI, and ischemia-driven TVR), and stent thrombosis (ST). All STs were adjudicated post hoc according to the Academic Research Consortium criteria. Definitions of clinical end points have been previously reported.
The principal secondary end point of the angiographic substudy was delayed late lumen loss (LL) between 8 months and 5 years among patients undergoing paired angiography. Lumen loss was defined as the difference between the minimal luminal diameter (MLD) after the procedure and MLD at follow-up. Delayed LL was defined as the difference between MLD at 8 months and MLD at 5 years. Secondary angiographic end points were percent diameter stenosis and binary restenosis. Binary restenosis was defined as stenosis of at least 50% of the MLD in the target lesion at angiographic follow-up.
Quantitative Coronary Angiography
Digital angiograms were analyzed with the use of an automated edge detection system (CAAS II; Pie Medical Imaging, Maastricht, The Netherlands). Angiographic readers were unaware of the type of stent implanted. Quantitative coronary angiography from patients returning for repeat angiography in the setting of ST was not included during the first 30 days. However, events beyond 30 days were part of the angiographic analysis because the need for repeat revascularization could no longer be attributed to a short-term response of the lesion to the procedure.
Statistical Analyses
Baseline characteristics were compared between diabetic and nondiabetic patients using χ test without taking into account the random allocation to SES or PES. We used a Cox proportional hazards model to compare clinical outcomes between groups. P values for differences in clinical outcomes between diabetic and nondiabetic patients were derived from Cox proportional hazards models adjusted for treatment allocation. To determine whether there was an interaction between treatment effect and diabetic status, we used likelihood ratio tests. Angiographic data were analyzed for all patients undergoing paired angiography at baseline, 8 months, and 5 years. Study lesions requiring revascularization by either percutaneous coronary intervention (PCI) or coronary artery bypass grafting between 8 months and 5 years were assessed by quantitative coronary angiography and contributed to the 5-year angiographic analysis. A patient could have had >1 lesion in which a stent was implanted. Therefore, in the analysis of the quantitative angiographic data, we used maximum likelihood logistic and linear regression models, crude and adjusted for MLD at baseline, which were based on robust standard errors that allowed for the correlation of multiple lesions within a patient to compare the characteristics of lesion between groups at baseline and follow-up. Trial data were held by CTU Bern, University of Bern, Bern, Switzerland. Analyses were performed with the use of Stata 11.1 (College Station, TX). No adjustments were made for multiple comparisons in secondary analysis. All P values are 2 sided.