Randomized Angiographic and Intravascular Ultrasound Comparison of Dual-Antiplatelet Therapy vs Triple-Antiplatelet Therapy to Reduce Neointimal Tissue Proliferation in Diabetic Patients
Abstract: Background. Previous studies have suggested a benefit of cilostazol in addition to standard dual-antiplatelet therapy (DAPT), reducing in-stent late luminal loss and restenosis after percutaneous coronary intervention (PCI) with bare-metal and drug-eluting stent (DES) implantation. However, there is a paucity of intravascular ultrasound (IVUS) assessment of neointimal tissue hyperplasia (NIH) after triple-antiplatelet therapy (TAPT), especially in diabetic patients treated with DES. Methods. This prospective, placebo-controlled trial was conducted in diabetic patients randomized (1:1) to receive either standard DAPT (aspirin and clopidogrel) vs TAPT with cilostazol for a minimum of 12 months after PCI with Endeavor zotarolimus-eluting stent (E-ZES). The primary endpoint was the 9-month comparison of percentage of NIH in both groups. Additionally, we compared in-stent late lumen loss, binary restenosis, major adverse cardiac event (MACE; cardiac death, non-fatal myocardial infarction, and restenosis) rates, and the incidence of vascular/bleeding complications. Results. In total, 133 diabetic patients were enrolled (cilostazol cohort = 65 patients) with 56.4% male and mean age of 60.8 years. Overall, the two cohorts were comparable in terms of baseline clinical and angiographic characteristics, except for the reference vessel diameter, which was smaller among patients randomized to cilostazol (2.48 ± 0.46 mm vs 2.69 ± 0.48 mm; P=.01). At 9 months, there was a non-significant trend toward less percentage of NIH obstruction in the TAPT cohort (33.2 ± 8.29% vs 35.1 ± 8.45%; P=.07). However, this finding did not impact angiographic late-lumen loss (0.60 ± 0.46 mm cilostazol group vs 0.64 ± 0.48 mm control group; P=.30) and binary restenosis (9.8% vs 6.8%; P=.99). MACE rate also did not significantly differ between the cohorts (13.8% cilostazol group vs 8.8% control group; P=.81). Of note, the addition of a third antiplatelet agent did not increase vascular and bleeding complications. Conclusion. In diabetic patients treated with E-ZES, TAPT with cilostazol did not add any significant benefit in terms of NIH suppression or MACE reduction.
J INVASIVE CARDIOL 2017;29(3):76-81
Key words: coronary artery disease, diabetes mellitus, cilostazol, intravascular ultrasound
Cilostazol is a selective inhibitor of phosphodiesterase III, which decreases the synthesis of extracellular matrix and smooth cell proliferation, potentially reducing neointimal tissue hyperplasia (NIH)1-3 after PCI with stent implantation. Recent studies have suggested a possible benefit of cilostazol in reducing in-stent late luminal loss (LLL) and restenosis when added to the standard dual-antiplatelet therapy (DAPT) with aspirin and clopidogrel following PCI with bare-metal (BMS) and drug-eluting stent (DES) implantation.4-10 Of note, Lee at al have demonstrated the angiographic benefit of this triple-antiplatelet therapy (TAPT) in patients with long lesions (≥30 mm) and diabetics treated with DES implantation.11,12
However, there is a paucity of intravascular ultrasound data showing the anti-restenotic benefits of cilostazol in western populations, especially among diabetic patients. Therefore, to determine whether cilostazol reduces neointimal hyperplasia after DES implantation, we undertook a randomized, double-blind, placebo-controlled study comparing TAPT (aspirin, clopidogrel, and cilostazol) and DAPT (aspirin, clopidogrel, and placebo) for 9 months in diabetics treated with Endeavor zotarolimus-eluting stent (E-ZES; Medtronic Vascular) implantation.
Methods
Study design and population. This is a prospective, double-blinded, placebo-controlled trial conducted in a single center in Brazil between April 2011 and August 2012. A total of 133 patients with diabetes mellitus and indication for PCI with E-ZES were randomly assigned (1:1) to receive either standard DAPT (aspirin + clopidogrel) or TAPT (aspirin + clopidogrel + cilostazol) for a minimum of 1 year after PCI. During the recruitment period, E-ZES was the only DES available at our public hospital.
The study enrolled patients with up to two de novo lesions in coronary arteries ≥2.5 mm in diameter. We excluded patients treated in the setting of ST-elevation myocardial infarction (MI; primary or rescue PCI), those with lesions in grafts and left main stem, as well as those with in-stent restenosis.
Procedures were performed in accordance with international guidelines. Antiplatelet regimen consisting of aspirin, clopidogrel, and cilostazol (or placebo) was given at least 24 hours prior to PCI.
By protocol, intravascular ultrasound (IVUS) evaluations were mandatory only at 9-month follow-up and were performed with the Atlantis Pro 40 MHz catheter (Boston Scientific Corporation).
Clinical follow-up was performed by office appointment at 1, 6, and 9 months. Invasive follow-up was scheduled for the 9-month visit.
Written informed consent was obtained from all patients and the local ethics committees approved this study.
Study endpoints and definitions. The primary endpoint was the comparison of IVUS-measured percentage of NIH obstruction between the two groups at 9-month follow-up.
Secondary endpoints included quantitative coronary angiography (QCA) assessment of lumen loss and binary restenosis as well as the occurrence of major adverse events (cardiac death, non-fatal MI, and restenosis) at 9 months. In-hospital vascular and bleeding complications (type 3 and 5 according to Bleeding Academic Research Consortium definition)13 as well as stent thrombosis (according to Academic Research Consortium definition)14 rates were also compared for both cohorts.
Percentage of NIH obstruction was defined as NIH volume divided by stent volume. This index was used to compensate for possible differences in reference vessel diameter and dimensions of deployed DES, which might result in differences in the absolute NIH volume.
Cardiac death was defined as any death due to cardiac cause (eg, MI, low-output failure, fatal arrhythmia). Unwitnessed death and death of unknown cause were also classified as cardiac deaths.
The MI classification and criteria for diagnosis were defined according to the per-protocol definitions. Q-wave MI was defined as the development of a new, pathological Q-wave. Non-Q wave MI was defined as the elevation of CK-MB to ≥3 times the upper limit of normal in the absence of new pathological Q-waves.
Binary restenosis was defined as the presence of ≥50% obstruction in the segment treated (stent + 5 mm proximal and distal). Ischemia-driven target-lesion revascularization (ID-TLR) was defined as any repeat percutaneous intervention of the target lesion or bypass surgery of the target vessel with either positive functional ischemia study or ischemic symptoms.
Device thrombosis was categorized as definite or probable, as well as acute (<1 day), subacute (1-30 days), and late (>30 days) according to the Academic Research Consortium guidelines.14
IVUS analysis. All IVUS analyses were performed according to the Expert Consensus Document on Standards for Acquisition, Measurement, and Reporting of Intravascular Ultrasound Studies.15 The visualization of at least two-thirds of the struts in consecutive frames was considered the landmark to define the beginning/end of the stent. After defining the stent beginning/end, a cross-sectional measurement was manually done at every 0.5 mm. Each cross-sectional measurement included the manual trace of vessel external elastic membrane (EEM), lumen, and stent.
Once all cross-sectional measurements were manually done, a volumetric assessment of vessel, lumen, and stent volumes was obtained using a mathematical validated model (Simpson’s rule). Other volumes were also derived from these main measurements (eg, NIH volume, plaque volume, volume, etc).
IVUS analysis was performed offline, by independent operators blinded to the patient allocation and using a validated software (Echoplaque version 3.0.28; Indec Medical Systems).
Statistical analysis. Based on previous studies with cilostazol, we estimated that 60 patients in each arm would provide 80% power to demonstrate the superiority of the TAPT over standard DAPT on reducing the percentage of NIH obstruction by IVUS at 9 months (α=0.05).
For the descriptive statistics, categorical data are presented as counts and percentages and continuous variables are presented as mean ± standard deviation. Categorical variables were compared by the Chi2-test or Fisher’s test when the Cochran’s rule was not met. Continuous variables were compared with the Student’s t-test.
Time-to-event variables are presented as Kaplan-Meier curves. A two-tailed P-value <.05 was considered statistically significant.
Results
Baseline clinical, angiographic, and procedure characteristics. Of the 133 diabetic patients enrolled in the trial, 65 (48.9%) were assigned to TAPT with cilostazol. Insulin-dependent diabetics represented only 7.8% of the total population.
Table 1 contains the main baseline clinical and angiographic data. The mean study population age was 60.8 years and most patients of both groups were male. Overall, the two cohorts did not significantly differ regarding baseline clinical aspects except for the predominance of patients with hypertension in the TAPT cohort (94.1% vs 80.5%; P=.03). The vast majority of patients (>90%) in both cohorts had stable clinical presentation including silent ischemia or stable angina.
Regarding baseline angiographic characteristics, 73% of the total population had single-vessel or double-vessel disease and the left anterior descending (LAD) artery was the coronary artery more often treated in both groups. Barely 86% of the entire population had preserved left ventricular (LV) function. One-half of the lesions in both groups were classified as B2/C according to the American College of Cardiology/American Heart Association classification for lesion complexity. Slightly more than one-half (54.9%) of all procedures were performed by radial approach, with no significant difference between the cohorts (53.8% in the cilostazol group vs 55.9% in the control arm; P=.81).
A total of 190 lesions were treated, with an average of 1.36 DESs/lesion (1.4 in the active arm vs 1.3 in the control arm; P=.39). Direct stenting was performed in 62.6% of the cases. Procedural success was achieved in 96.9% of the cases in the cilostazol arm and in 94.1% of the control group (P=.68).
Nine-month QCA and IVUS results. Tables 2 and 3 detail the most relevant QCA and IVUS data. Nine-month angiographic and IVUS follow-up were achieved in 130 (99.2%) and 123 (93.8%) out of the 131 eligible patients, respectively.
By QCA, there was a trend for shorter lesions in the cilostazol population (15.16 ± 6.58 mm in the cilostazol arm vs 17.01 ± 7.72 mm in the control arm; P=.07), while reference vessel diameter was markedly smaller in the TAPT cohort (2.48 ± 0.46 mm vs 2.69 ± 0.48 mm; P=.01). Preprocedure percentage of lesion stenosis was 71.0 ± 11.0%, with no difference between groups. Although acute luminal gain was smaller among cilostazol patients (1.91 ± 0.42 mm vs 2.07 ± 0.40 mm; P=.01), there was a trend for lower residual stenosis among them (3.2 ± 3.3% vs 4.0 ± 4.9%; P=.08). At 9-month QCA assessment, in-stent lumen loss did not differ between the cohorts (0.60 ± 0.46 mm in the cilostazol arm vs 0.64 ± 0.48 mm in the control group; P=.30) resulting in similar binary restenosis rates (9.8% in the active arm vs 6.8% in the placebo group; P=.99).
Since reference vessel diameter was bigger in the placebo group, larger diameter stents were deployed among those patients. As a consequence, at 9-month IVUS assessment, vessel and stent volumes were bigger in the placebo group, which resulted in higher NIH volume among them. However, when the NIH volume is indexed to the stent volume, there was no significant difference in the percentage of NIH obstruction between the two cohorts (33.2 ± 8.2% vs 35.1 ± 8.45%; P=.07).
Clinical outcomes. Table 4 details in-hospital and 9-month clinical events. In the in-hospital phase, there were 3.1% non-fatal periprocedural MIs in the TAPT cohort and 5.9% in the placebo group (P=.68). No cases of death, urgent revascularization, or stent thrombosis were observed in this period. Vascular complications (2.3%) and bleeding (1.5%) were infrequent and similar in both groups.
Complete clinical follow-up was achieved for all patients (average, 10.4 ± 1.9 months). During this period, 92.3% of the entire population remained asymptomatic. Two deaths (one cardiac and one non-cardiac) were observed, both in the cilostazol group. After hospital discharge, there was a single case of non-fatal MI (in the cilostazol arm) and 12 cases of TLR (9.2% in the cilostazol cohort and 8.8% in the placebo group; P=.70). Of note, there were 12 cases of non-TLR during the follow-up period (6 in each arm). Figure 1 shows the survival free of MACE curves for the groups.
A single case of (probable) stent thrombosis (sudden death 7 months after PCI) was observed in the cilostazol cohort. There were 2 cases (1.5%) of Blood Academic Research Consortium class 3 bleeding (upper digestive tract hemorrhage), one in each group. Hospitalization and blood transfusion were required in both cases.
Notably, the use of cilostazol was associated with higher incidence of sinus tachycardia at 1-month (27.7% vs 4.4%; P<.001) and 9-month follow-up (13.8% vs 2.9%; P=.02). Furthermore, migraine was more frequently described among patients in use of the TAPT (6.2% vs 0%; P=.05). Discontinuation of prescribed antiplatelet therapy occurred in 7% of the cases in each cohort.
Discussion
According to the present study, the addition of cilostazol to standard DAPT, in diabetic patients treated with DES, did not result in significant reduction in NIH formation either by QCA or IVUS assessment at mid-term invasive follow-up. As a consequence, its use did not impact the occurrence of binary restenosis or need for repeat lesion revascularization. Finally, from a safety perspective, patients randomized to this medication did not present with more hemorrhagic or vascular complications.
In 2005, Douglas et al published the results of the CREST trial, a randomized evaluation of cilostazol for restenosis prevention in patients treated with BMS. In that study, the use of TAPT resulted in 36% reduction in the relative risk of binary restenosis (22.0% vs 34.5%; P=.01). This observation was consistent among diabetic patients (17.7% vs 37.7%; P=.01).16 Similarly, Min et al, in a study with 59 patients (65 lesions) treated with BMS, demonstrated that the addition of cilostazol to current DAPT resulted in less NIH formation at 6-month IVUS evaluation (1.0 ± 0.5 mm³/mm vs 2.2 ± 1.4 mm³/mm; P=.01).17 Of note, this trial enrolled patients with larger reference diameter as compared with our current analysis (on average, 0.5 mm larger). In their study, like in the present analysis, there was no difference in binary restenosis rates.
Most of the positive trials on cilostazol come from the Asian experience with this drug. Of note, in 2008, Lee et al published a randomized comparison with DAPT vs TAPT in diabetic patients submitted to PCI with first-generation DESs (sirolimus or paclitaxel-eluting stents). In their experience, the use of cilostazol resulted in marked reduction in in-stent late luminal loss (0.25 ± 0.53 mm vs 0.38 ± 0.54 mm; P=.02) as well as in the restenosis rates (8.0% vs 15.6%; P=.03).12 In 2011, the same group published the results of the DECLARE-LONG II trial on 499 patients with long lesions (≥25 mm) treated with zotarolimus-eluting stents. As in their previous study, in this trial, the addition of cilostazol to the standard DAPT resulted in a significant reduction in in-stent luminal loss (0.32 ± 0.54 mm vs 0.47 ± 0.54 mm; P=.01) and binary restenosis (10.8% vs 19.1%; P=.02). In their IVUS subanalysis (n = 122 patients), the use of this drug reduced the percentage of NIH obstruction (22.1 ± 9.9% vs 27.1 ± 13.2%; P=.02).11 In their experience, the use of cilostazol did not markedly increase the occurrence of bleeding, but impacted the observation of cutaneous rash (7.5% vs 2.5%; P=.04). Part of the differences between these findings and those observed in our experience might be explained by the differences in inflammatory responses identified in patients from different ethnic groups.18
Study limitations. Despite the randomization, patients treated with cilostazol had smaller reference vessel diameters, which is a known independent predictor of higher restenosis rates and worse clinical outcomes. IVUS catheters are not reimbursed by our public health system and due to budget restraints, this imaging tool was only used at follow-up evaluations. The Endeavor zotarolimus-eluting stent is not available anymore for clinical use. Finally, the relatively small sample size precludes definitive conclusions on the clinical impact of cilostazol.
Conclusion
In diabetic patients treated with DES implantation, the addition of cilostazol to standard DAPT did not result in significant improvement in 9-month angiographic and IVUS assessments. Conversely, this phosphodiesterase III inhibitor demonstrated acceptable safety clinical profile, with low rates of bleeding and vascular complications.
References
1. Ishizaka N, Taguchi J, Kimura Y, et al. Effects of a single local administration of cilostazol on neointimal formation in balloon-injured rat carotid artery. Atherosclerosis 1999;142:41-46.
2. Brasil DP. Cilostazol após angioplastia coronária. In: Viveiros B, ed. Cilostazol e Aterosclerose em Múltiplos Territórios. São Paulo, BBS Editora. 2008:69-83.
3. Morishita R. A scientific rationale for the CREST trial results: evidence for the mechanism of action of cilostazol in restenosis. Atheroscler Suppl. 2005;6:41-46.
4. Biondi-Zoccai GG, Lotrionte M, Anselmino M, et al. Systematic review and meta-analysis of randomized clinical trials appraising the impact of cilostazol after percutaneous coronary intervention. Am Heart J. 2008;155:1081-1089.
5. Tamhane U, Meier P, Chetcuti S, et al. Efficacy of cilostazol in reducing restenosis in patients undergoing contemporary stent based PCI: a meta-analysis of randomized controlled trials. EuroIntervention. 2009;5:384-393.
6. Friedland SN, Eisenberg MJ, Shimony A. Meta-analysis of randomized controlled trials on effect of cilostazol on restenosis rates and outcomes after percutaneous coronary intervention. Am J Cardiol. 2012;109:1397-1404.
7. Jang JS, Jin HY, Seo JS, et al. A meta-analysis of randomized controlled trials appraising the efficacy and safety of cilostazol after coronary artery stent implantation. Cardiology. 2012;122:133-143.
8. Zhang Y, Tang HQ, Li J, Fu ZX. Efficacy and safety of triple-antiplatelet therapy after percutaneous coronary intervention: a meta-analysis. Chin Med J (Engl). 2013;126:1750-1754.
9. Chen Z, Qian J, Chen Y, Ma J, Ge J. Addition of cilostazol to conventional dual antiplatelet therapy reduces the risk of cardiac events and restenosis after drug-eluting stent implantation: a meta-analysis. J Clin Pharmacol. 2013;53:532-539. Epub 2013 Feb 22.
10. Chen J, Meng H, Xu L, et al. Efficacy and safety of cilostazol-based triple-antiplatelet treatment versus dual-antiplatelet treatment in patients undergoing coronary stent implantation: an updated meta-analysis of the randomized controlled trials. J Thromb Thrombolysis. 2015;39:23-34.
11. Lee SW, Park SW, Kim YH, et al. A randomized, double-blind, multicenter comparison study of triple antiplatelet therapy with dual-antiplatelet therapy to reduce restenosis after drug-eluting stent implantation in long coronary lesions: results from the DECLARE-LONG II (Drug-Eluting Stenting Followed by Cilostazol Treatment Reduces Late Restenosis in Patients with Long Coronary Lesions) trial. J Am Coll Cardiol. 2011;57:1264-1270.
12. Lee SW, Park SW, Kim YH, et al. Drug-eluting stenting followed by cilostazol treatment reduces late restenosis in patients with diabetes mellitus. (The DECLARE-DIABETES trial). J Am Coll Cardiol. 2008;51:1181-1187.
13. Mehran R, Rao SV, Bhatt DL, et al. Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the Bleeding Academic Research Consortium. Circulation. 2011;123:2736-2747.
14. Cutlip DE, Windecker S, Mehran R, et al. Academic Research Consortium (ARC). Clinical endpoints in coronary stent trials: a case for standardized definitions. Circulation. 2007;115:2344-2351.
15. Mintz GS, Nissen SE, Anderson WD, et al. American College of Cardiology Clinical Expert Consensus Document on Standards for Acquisition, Measurement, and Reporting of Intravascular Ultrasound Studies (IVUS). A report of the American College of Cardiology Task Force on Clinical Expert Consensus Documents. J Am Coll Cardiol. 2001;37:1478-1492.
16. Douglas JS Jr, Holmes DR Jr, Kereiakes JD, et al. Coronary stent restenosis in patients treated whit cilostazol. For the Cilostazol for Restenosis Trial (CREST) Investigators. Circulation. 2005;112:2826-2832.
17. Min PK, Jung JH, Ko YG, Choi D, Jang Y, Shim WH. Effect of cilostazol on in-stent neointimal hyperplasia after coronary artery stenting: a quantitative coronary angiography and volumetric intravascular ultrasound study. Circ J. 2007;71:1685-1690.
18. Ma MK, Woo MH, McLeod HL. Genetic basis of drug metabolism. Am J Health Syst Pharm. 2002;59:2061-2069.
From the 1Hospital da Beneficência Portuguesa, São Paulo, Brazil; and 2Instituto Dante Pazzanese de Cardiologia, São Paulo, Brazil.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. All institutions involved received a research grant to conduct the trial. However, none of the authors have any individual conflicts of interest regarding this publication.
Manuscript submitted September 9, 2016, final version accepted September 19, 2016.
Address for correspondence: Maria Fernanda Zuliani Mauro, MD, PhD, Department of Interventional Cardiology, Hospital Beneficência Portuguesa, R. Maestro Cardim, 769 - Bela Vista, São Paulo, SP – Brazil 01323-001. Email: fzuliani@uol.com.br
