The study design is shown in Fig. 1. The CiLostazol administration before pErcutaneous coronAry intervention for Reduction of periprocedural myonecrosis trial (CLEAR trial) was a single center, randomized, prospective trial. Between June 2007 and May 2009, patients with typical angina, not taking statins and without elevated levels of cardiac enzymes, who were scheduled for drug-eluting stent (DES) implantation in de novo coronary artery lesions were randomly assigned 1 : 1 to pretreatment with cilostazol 200 mg/day for 7 days (Cilostazol group) or to no pretreatment (control). Stratified randomization was performed according to age (≥65 or <65 years), presence of bifurcation, and lesion length (≥28 or <28 mm). Inclusion criteria were stable angina with the presence of typical angina or positive stress test (electrocardiogram, treadmill test, or nuclear scan) and indications for stent implantation. Exclusion criteria were: any increase in CK-MB or cardiac troponin I (cTnI); acute myocardial infarction (MI) (<3 months); any increase in liver enzyme [aspartate aminotransferase (AST)/alanine aminotransferase (ALT)]; left ventricle ejection fraction <30%; renal failure with creatinine>3 mg/dL; history of liver or muscle disease; coronary lesions with left main disease, chronic total occlusion, in-stent restenosis, or bare-metal stent (BMS) implantation; current therapy with statins (for the exclusion of the effects of pretreatment with various statins); or needing PCI within 7 days of diagnostic coronary angiography. After undergoing diagnostic coronary angiography using radial artery access at an out-patient clinic, patients fulfilling the criteria were randomly assigned to each group. Randomization was stratified according to age (<65 or ≥65 years), presence of bifurcation lesions, and lesion length (<28 or ≥28 mm), which were regarded as the important predictors for periprocedural myonecrosis.3,4 Clinical follow-up of enrolled patients was performed for 6 months. The local ethics committee approved the study, and all patients gave written informed consent.

Stents were implanted according to current clinical practice guidelines. Angiographic success was defined as final angiographic residual stenosis of <20% by quantitative coronary angiographic analysis. All patients received daily aspirin 100-200 mg and clopidogrel 75 mg starting from 7 days before the elective PCI. Before PCI, all patients received a 60 IU/kg intravenous bolus of unfractionated heparin. Glycoprotein IIb/IIIa inhibitors were administered according to the operator's discretion. Angiographic analysis was performed with a computer-assisted automated edge-detection algorithm (Phillip INTEGRIS BH 5,000, Phillips Medical Systems, the Netherlands). Angiographic complications included: side branch occlusions (transient or permanent); abrupt intra-procedural vessel closure; major or minor dissection; thrombus formation; transient and/or prolonged slow-no reflow; and distal embolization. All patients received clopidogrel 75 mg/day for at least 6 months in addition to continued aspirin (100 mg/day).

Venous blood samples were drawn immediately after randomization (baseline), before (7 days later after pre-treatment), and 6 and 24 hours after PCI. CK-MB and cTnI levels were determined using a radioimmunoassay analyzer (Dimension RxL Analyzer; Dade Behring, Marburg, Germany). The UNL, which represents the 99th percentile of the distribution of a reference control group with an analytical imprecision of more than 10%, was 4 ng/mL for CK-MB and 0.2 ng/mL for cTnI. The peak enzyme ratio was calculated as the level of peak cardiac enzyme divided by the value of the upper normal limit. Before discharging the patients, we performed aspirin and clopidogrel resistance testing using the VerifyNow-ASA and -P2Y12™ assays (Accumetrics Inc., San Diego, CA, USA). Samples were obtained by antecubital venipunture using a 23-gauge syringe, and the initial 3 to 4 mm of blood was discarded. The second samples were collected in 4.5-mL plastic tubes containing 3.2% citrate for rapid platelet-function assay.8 The results of the aspirin and clopidogrel resistance tests were expressed as aspirin reaction unit (ARU) and percentage inhibition rate, respectively.8,11 The percentage inhibition of clopidogrel was calculated as (1-P2Y12 reaction unit/estimated baseline) ×100. Aspirin resistance was defined as an ARU of ≥550, and clopidogrel resistance as percentage inhibition of clopidogrel <20%.8

The primary end point was the occurrence of periprocedural myonecrosis, defined as any CK-MB elevation above the UNL.10 Secondary end points included: 1) the occurrence of periprocedural MI, defined as a postprocedural increase in CK-MB level ≥3 times above the UNL; 2) the occurrence of a large periprocedural MI, defined as a postprocedural increase of CK-MB ≥10 times above the UNL; 3) cTnI elevation above the UNL, 3 times, or 10 times above the UNL after PCI; 4) comparison of the rate of periprocedural myonecrosis according to the presence of clopidogrel resistance; and 5) rate of in-hospital and 6-month adverse events such as major adverse cardiac events (MACE; death, acute MI, need for bypass surgery or repeat PCI), stent thrombosis, cerebrovascular accidents (CVA), bleeding, or other serious side effects.

Bleeding was defined based on the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries bleeding classification: severe or life-threatening bleeding was defined as intracerebral bleeding or bleeding with substantial hemodynamic compromise requiring treatment. Moderate bleeding was defined by the need for transfusion. Minor bleeding referred to other bleeding, not requiring transfusion or causing hemodynamic compromise.12 Bleeding at the percutaneous entry site was defined by the size of external hematoma (>10 cm for femoral and>2 cm for radial access) during or after the catheterization laboratory visit until discharge.13

According to recent studies, loading of clopidogrel before PCI reduced the occurrence of postprocedural ischemic complications by 45%.5,14 Assuming that the cilostazol pretreatment would be as potent as or more potent than high dose clopidogrel loading in the reduction of periprocedural myonecrosis and that the incidence of periprocedural myonecrosis of the control group would be 40-45%, we could expect the incidence of periprocedural myonecrosis to be 15% in the trial group. Thus, a sample size of 110 patients (55 patients in each group) would provide 80% power to detect a difference with an alpha (probability value) of 0.05. Based on a withdrawal rate of about 10%, sixty patients per each group, i.e., a total of 120 patients, were considered appropriate.

Statistical analysis was performed with Statistical Package for the Social Sciences (SPSS) software (SPSS Inc., Chicago, IL, USA). Categorical data, presented as numbers with frequencies, were compared with chi-square statistics or Fisher's exact test, while continuous data, presented as mean±SD, were compared with Student's t-test for normally distributed values; otherwise, the Mann Whitney U test was applied. To determine the independent predictors for periprocedural myonecrosis, multivariate logistic regression analysis was performed, and variables that were predictive at the 0.1 level by univariate analysis were entered into final multivariate analysis. A value of p<0.05 denoted significance.

Among all the enrolled patients, 10 were excluded, and the remaining 110 were ultimately eligible for our study; [cilostazol group (n=54)] versus [control group (n=56)] (Fig. 1). There were no significant differences in baseline clinical and biochemical variables between the two groups (Table 1). The angiographic and procedural characteristics were similar between the two groups (Table 2 and 3). The occurrence of angiographic complications during the procedure was similar between the two groups (Table 3).

After 7 days of pretreatment, there was no significant difference in ARU between the cilostazol and control groups (418±71 versus 435±46, p=0.170) (Fig. 2A). However, the % inhibition of clopidogrel was significantly higher in the cilostazol group (39±23%) than in the control group (25±22%, p=0.003) (Fig. 2B). The prevalence of clopidogrel resistance, defined by % inhibition of <20%, was significantly lower in the cilostazol group [12 patient (22%)] than in the control group [27 patients (48%), p=0.006] (Fig. 2C), which means the cilostazol group had significantly reduced clopidogrel resistance.

Periprocedural myonecrosis, defined as any postprocedural increase in CK-MB above the UNL, was detected in 13 (24%) patients in the cilostazol group and in 14 (25%) in control group, showing similar rates (p=1.000) (Fig. 3A). The incidences of periprocedural MI (CK-MB ≥3 times) or large periprocedural MI (≥10 times UNL) were not significantly different between the two groups. cTnI elevation above the UNL or ≥3 or 10 times above the UNL occurred at a similar rate between the two groups (Fig. 3B). In addition, there were no differences in the level of peak or peak ratio of CK-MB or cTnI after PCI (Fig. 4).

Using multivariate logistic regression analysis, the most significant independent predictor for any CK-MB elevation was the presence of multivessel disease [odds ratio (OR)=4.25, 95% confidence interval (CI) 1.49-10.09, p=0.010)]. However, cilostazol pretreatment was not a significant predictor (OR=0.92, 95% CI 0.36-2.40, p=0.814) (Table 4).

Both in the hospital and during the 6 months of follow-up, there was no significant difference in the occurrence of fatal events (Table 5). The occurrence of bleeding events was not significantly different between the two groups. However, the cilostazol group showed a tendency toward a slightly higher rate of any bleeding, chiefly due to the high incidence of entry site hematoma. Closure devices were used for all enrolled patients and the proportions of use of various types did not differ between the groups. There were no cases with retroperitoneal bleeding or requiring further intervention to control bleeding. All entry site hematomas spontaneously resolved before discharge. In addition, there was no relationship between bleeding episodes, including major and minor bleeding, and those resultant clinical outcomes.

To evaluate the effects of clopidogrel resistance on postprocedural enzyme level increases and clinical outcomes, we divided patients into two groups based on the value of % inhibition of clopidogrel: patients with clopidogrel resistance (% inhibition <20) [n=37 (34%)] and patients with normal response to clopidogrel (% inhibition ≥20) [n=73 (66%)] (Table 6). The incidences of any CK-MB or cTnI elevation were not different between the two groups. In addition, there were no significant differences in the incidence of postprocedural CK-MB or cTnI level ≥3 times or 10 times above the UNL between the two groups, respectively. The level of peak CK-MB or cTnI of the patients with clopidogrel resistance was not significantly different from those of the patients with normal response to clopidogrel (peak CK-MB; 5.4±19.7 vs. 5.1±10.1 ng/mL, p=0.904, peak cTnI; 0.7±3.1 vs. 0.7±2.5 ng/mL, p=0.910). The composite incidence of cardiac death, MI, target-lesion revascularization, or CVA in patients with clopidogrel resistance was slightly higher than that of patients with normal responses to the clopidogrel, but the difference did not achieve statistical significance (Table 6).