Comparison of Titanium-Nitride-Oxide–Coated Stents With Zotarolimus-Eluting Stents for Coronary Revascularization

A Randomized Controlled Trial

Patient population

Patients who were at least 18 years of age with stable or unstable angina pectoris or non–ST-segment elevation myocardial infarction and signs of myocardial ischemia were eligible if they had at least 1 lesion with a diameter stenosis of 50% or more that was suitable for coronary stent implantation in a vessel with a reference vessel diameter ranging from 2.25 to 4.0 mm. The following exclusion criteria were applied: patients unable to provide informed consent; participation in another trial before reaching the primary endpoint; and patients with known intolerance to aspirin, clopidogrel, heparin, stainless steel, titanium, zotarolimus, or contrast material. The study complied with the Declaration of Helsinki and was approved by the institutional review boards of all participating institutions. Written informed consent was obtained from all patients. The study was an investigator-initiated study and there was no industry involvement in the design, conduct, financial support, or analysis of the study.

Study design and procedures

The study was a randomized assessor-blind, noninferiority trial performed in 3 institutions in Switzerland. Randomization was performed after diagnostic angiography and before percutaneous coronary intervention. Sequentially numbered, sealed, opaque, tamper-proof security envelopes (Envelock Safety Envelopes, Plasto-Sac Ltd., Yavne, Israel) were used, which were independently monitored by an academic clinical trials unit. The allocation schedule was based on computer-generated random numbers, stratified according to trial center and catheter laboratory, and blocked, with block sizes of 2, 4, and 6 varying randomly. Patients were assigned on a 1:1 basis to treatment with a TiNO-coated stent (Titan 2) or ZES (Endeavor).

TiNO-coated stents were available in diameters of 2.25 to 4.0 mm and in lengths of 7 to 28 mm. ZES were available in diameters of 2.25 to 4.0 mm and in lengths of 8 to 30 mm. Balloon angioplasty and coronary stent implantation were performed using standard techniques; direct stenting was allowed and performed at the discretion of the investigator. Full lesion coverage was attempted by implanting 1 or several stents. Crossover to a nonstudy stent was allowed to complete the procedure successfully, in the event that the assigned study stent could not be implanted. In case more than 1 lesion was treated in a given patient in the same session or in a staged procedure during the index hospitalization, the remaining lesions were still treated with the originally assigned study stent.

All patients were treated with at least 100 mg of acetylsalicylic acid and 300 to 600 mg of clopidogrel bisulfate before or at the time of the procedure. During percutaneous coronary intervention, unfractionated heparin in a dose of at least 5,000 IU or 70 to 100 IU/kg was administered to maintain an activated clotting time >250 s. The use of glycoprotein IIb/IIIa antagonists was left to the discretion of the operator. Creatine kinase (CK), CK-myocardial band, and troponin were assessed 6 h after the procedure and either at 18 h after the procedure or at hospital discharge, whichever came first. A 12-lead electrocardiogram was obtained before and within 24 h after the procedure or for any suspicion of acute ischemia. In case of elevated cardiac enzymes after the procedure, CK, CK-myocardial band, and troponin measurements were continued every 8 h until the peak of the biomarkers had been conclusively defined. All patients were discharged on acetylsalicylic acid 100 mg daily indefinitely and clopidogrel bisulfate 75 mg daily for an intended duration of 3 months.

Quantitative coronary angiography

Coronary angiograms were digitally recorded at baseline, immediately after the procedure, and at follow-up and were assessed at the angiographic core laboratory of Bern University Hospital. Readers of angiograms were blinded to the assigned study stent. The projections that best showed the stenosis were used for all analyses. Measurements were performed on the cineangiograms after maximum vasodilation with nitroglycerin and the contrast-filled, nontapered catheter tip was used for calibration (≥6-F guiding catheter). Digital angiograms were analyzed with the help of an automated edge-detection system (QAngio XA, Version 7/1/14.Zero, Medis Medical Imaging Systems, Leiden, the Netherlands). Quantitative measurements included the diameter of the reference vessel, the minimal luminal diameter, percent diameter stenosis (difference between reference vessel diameter and minimal luminal diameter/reference diameter × 100), and late lumen loss (difference between minimal lumen diameter after the procedure and minimal lumen diameter at follow-up). Binary restenosis was defined as stenosis of 50% or greater of the minimal lumen diameter in the target lesion at angiographic follow-up. All angiographic measurements of the target lesion were obtained in the stented area, within the margins 5 mm proximal and distal to each stent edge (in-stent), and over the entire segment (in-segment).

Study endpoints and definitions

Adverse events were assessed at 1, 6, and 12 months. All patients were asked to return for an angiographic follow-up study at 6 to 8 months. An independent clinical event committee whose members were unaware of the patient's treatment assignment adjudicated all clinical endpoints. The primary endpoint of the study was in-stent late loss at 6 to 8 months after stent implantation as assessed by quantitative coronary angiography. Secondary angiographic endpoints included in-segment late loss as well as in-stent and in-segment binary restenosis, minimal luminal diameter, and percent diameter stenosis. Secondary clinical endpoints included: death; cardiac death; myocardial infarction; clinically and nonclinically indicated TLR; and the composite of cardiac death, myocardial infarction, or clinically indicated target lesion revascularization at 30 days, 6 months, and 1 year.

The definition of cardiac death included any death due to immediate cardiac cause (e.g., myocardial infarction, low-output failure, fatal arrhythmia); procedure-related deaths, including those related to concomitant treatment; unwitnessed death; and death of unknown cause. Myocardial infarction was defined using the electrocardiographic criteria of the Minnesota Code Manual, or as an elevation of CK levels to more than 2 times the upper limit of normal with positive levels of CK-myocardial band or troponin. TLR was defined as any repeat PCI within the stent or within the 5-mm borders adjacent to the stent, or bypass surgery of the target vessel. Revascularization of the target lesion and vessel was regarded as clinically indicated if the stenosis on any target lesion or vessel was at least 50% of the diameter of the vessel on the basis of quantitative coronary angiography in the presence of recurrent angina or objective signs of ischemia, or if the stenosis was at least 70% of the diameter of the vessel even in the absence of ischemic signs and symptoms. Stent thrombosis was defined according to the Academic Research Consortium definition as definite, probable, or possible (15).

Statistical analysis

This was a noninferiority trial, which was powered for noninferiority on the primary endpoint in-stent late lumen loss at 6 to 8 months. Based on angiographic outcomes reported for the ENDEAVOR II (Randomized Controlled Trial to Evaluate the Safety and Efficacy of the Medtronic AVE ABT-578 Eluting Driver Coronary Stent in De Novo Native Coronary Artery Lesions) trial (11) and the TINOX (Randomized Comparison of a Titanium-Nitride-Oxide–Coated Stent With a Stainless Steel Stent for Coronary Revascularization) study (9), we postulated the following in-stent late lumen loss values: ZES 0.61 ± 0.46 mm; TiNO 0.55 ± 0.63 mm. Assuming a noninferiority margin of 0.15 mm as the acceptable difference between TiNO and ZES to claim TiNO noninferiority to ZES, an average number of 1.5 lesions per patient, a design factor of 1.1 to take into account the correlation of lesions within patients, and a 1-sided type-1 error of 0.05, we estimated that 234 patients (117 patients in each group) will yield >80% power to detect noninferiority. To account for patients not undergoing repeat angiography 6 to 8 months after the procedure (typically 20% to 25% of the overall patient population), we pre-specified a total enrollment of 300 patients.

Baseline clinical characteristics, pre-procedural, procedural, and follow-up angiographic results (including the primary endpoint in-stent late lumen loss) are reported as mean ± SD for continuous variables and absolute numbers and percentages for categorical variables. For the angiographic results, the differences (including 95% confidence interval [CI] and p value) between the 2 treatment groups were calculated using multilevel mixed-effects linear or logistic models that allowed for correlation of multiple lesions within patients. Stratified analyses were performed for the primary outcome according to the presence or absence of diabetes, lesions located in the left anterior descending artery, multivessel intervention, small vessel disease, long lesions, and age (>65 years vs. ≤65 years). To determine whether there was an interaction between treatment group and these characteristics, we used a likelihood-ratio test. We used the Mantel-Cox model and the corresponding log-rank test for between-group comparison of clinical outcomes occurring up to 1 month, up to 6 months, and up to 1 year. The duration of clopidogrel prescription between both treatment groups was compared using the Wilcoxon rank sum test. All patients who underwent randomization were included in the analysis in the groups to which they were originally allocated to (intention-to-treat principle). Analyses of angiographic outcomes were restricted to lesions from patients who attended follow-up angiography at 6 to 8 months (±2 months). In a sensitivity analysis, only the first lesion of each patient was analyzed on angiographic outcomes. Analyses were performed by a statistician (A.L.) who was blinded to the allocated treatment. No adjustments were made for multiple comparisons in secondary analyses; all p values and 95% CI are 2-sided. All analyses were done in Stata (version 11.0, StataCorp, College Station, Texas).

Angiographic outcomes

Angiographic follow-up was completed according to protocol in 239 of 302 patients (79.1%) with 358 of 451 lesions (79.4%) at a median follow-up of 7.9 months (IQR: 7.4 to 8.9 months) (Table 3). A total of 122 patients (80.3%) allocated to TiNO and 117 patients (78.0%) allocated to ZES underwent follow-up angiography (p = 0.63). Patients with angiographic follow-up were younger (63.9 vs. 67.6; p = 0.008) and were less frequently diabetic (15.5% vs. 33.3%; p = 0.001) and hypertensive (69.5% vs. 82.5%, p = 0.04). Among patients undergoing angiographic follow-up, clinical baseline characteristics were similar between TiNO and ZES, with the exception of age (65.5 vs. 62.3 years; p = 0.01).

The TiNO group failed to reach the pre-specified noninferiority margin (difference: 0.16, upper 1-sided 95% CI: 0.26; p_{noninferiority} = 0.54) (Table 3). In-stent late loss was significantly lower with ZES than TiNO (0.47 ± 0.48 mm vs. 0.64 ± 0.61 mm; p_superiority = 0.02). In-segment late loss was also lower with ZES than TiNO (0.36 ± 0.44 mm vs. 0.48 ± 0.55 mm; p_superiority = 0.04). Consistent results in favor of ZES were observed for measurements of in-stent and in-segment minimal lumen diameter and percent diameter stenosis. As a result, a significant reduction of in-segment binary restenosis was observed with ZES versus TiNO groups (11.1% vs. 20.5%; p_superiority = 0.04). The cumulative distribution of in-stent late lumen loss for the 2 stent groups is shown in Figure 2A. The cumulative frequencies of in-stent percent diameter stenosis for the 2 stent groups before and after the procedure, and at follow-up angiography are shown in Figure 2B. Figure 3 shows the results of stratified analyses of the primary endpoint. Results were consistent among patients with or without lesions located in the left anterior descending artery, multivessel intervention, and long lesions. Conversely, differences between TiNO and ZES appeared more pronounced among diabetic compared with nondiabetic patients, patients with or without small vessel disease, and patients aged below or above 65 years. However, a p value for interaction reached conventional levels of significance only in the case of diabetic mellitus (p = 0.04).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Cumulative Distribution of In-Stent Late Loss and Cumulative Frequency Distribution of In-Stent Percent Diameter Stenosis

(A) Cumulative distribution of in-stent late loss shown separately for the 2 stent types. The cumulative distribution curve of in-stent late loss is shifted to the right for titanium-nitride-oxide (TiNO) relative to zotarolimus-eluting stent (ZES), indicating that neointimal hyperplasia was more effectively reduced with ZES. (B) Cumulative frequency distribution curves of percent in-stent diameter stenosis shown separately for the stent types. There was no difference in measurements before and immediately after the procedure between the 2 stent types. At follow-up angiography, the cumulative distribution curve of percent in-stent diameter stenosis is shifted to the right for TiNO compared with ZES, indicating that in-stent diameter stenosis was more effectively reduced with ZES. TiNO = continuous lines; ZES = dotted lines.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Stratified Analyses for Primary Outcome

Subgroup analysis of the primary endpoint, in-stent late loss, among patients randomized to treatment with TiNO or ZES. *p values for interaction between treatment group and corresponding stratification factor. CI = confidence interval; other abbreviations as in Figures 1 and 2.

Clinical outcomes

At 30 days, adverse events were rare, and there were no differences with respect to ischemic and revascularization endpoints among patients randomized to TiNO or ZES (Table 4). At 6 months, revascularization procedures by any definition (clinically indicated TLR, any TLR, clinically indicated target vessel revascularization, any target vessel revascularization, any repeat revascularization procedure) tended to be more common among patients randomized to TiNO than to ZES (Table 4). This trend remained sustained up to 1 year, but there were no differences with respect to death, cardiac death, or myocardial infarction. Similarly, major adverse cardiac events—the composite of cardiac death, myocardial infarction. and clinically indicated target vessel revascularization—showed no significant differences between TiNO and ZES up to 1 year (Fig. 4A). Cumulative incidence curves of clinically indicated TLR began to diverge as early as 4 months after the procedure in favor of ZES (Fig. 4B). Rates of definite, probable, and possible stent thrombosis during the early and late time periods were low and comparable for both stent types (Table 5).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Composite of Cardiac Death, Myocardial Infarction, or Clinically Indicated Target Vessel Revascularization

(A) Kaplan-Meier curves show the cumulative incidence of the composite endpoint of cardiac death, myocardial infarction, and clinically indicated target vessel revascularization up to 12-month follow-up for TiNO and ZES. (B) Kaplan-Meier curves show the cumulative incidence of just clinically indicated target lesion revascularizations up to 12-month follow-up for TiNO and ZES. TiNO = continuous lines; ZES = dotted lines. Abbreviations as in Figures 1 and 2.

Study limitations

The present study was designed for an angiographic endpoint and, therefore, is too small to qualify as a clinical endpoint study. Although late loss was significantly lower with ZES than TiNO, clinical outcomes through 1 year showed similar results in terms of major adverse cardiac events for both stent platforms, and differences in revascularization procedures were likely driven by protocol-mandated angiographic follow-up. Another limitation is the lack of longer-term clinical and angiographic follow-up beyond 1 year. It cannot be excluded that the 2 stent platforms have a differential impact on long-term vessel remodeling. Thus, longitudinal angiographic follow-up series of patients treated with bare-metal stents suggest late improvements in lumen diameter beyond the early period of neointimal proliferation, whereas a late catch-up phenomenon has been observed with early generation DES (26). Notwithstanding, ZES represents a newer generation DES, and long-term clinical follow-up indicates low rates of repeat revascularization up to 3 years (13). Finally, the overall number of patients studied with TiNO is rather small, and therefore, the available evidence is limited as compared to that for established DES platforms.