Skip main navigation

Everolimus-Eluting Stents in Patients With Bare-Metal and Drug-Eluting In-Stent Restenosis

Results From a Patient-Level Pooled Analysis of the RIBS IV and V Trials
and under the auspices of the Interventional Cardiology Working Group of the Spanish Society of Cardiology
Originally publishedhttps://doi.org/10.1161/CIRCINTERVENTIONS.115.003479Circulation: Cardiovascular Interventions. 2016;9:e003479

    Abstract

    Background—

    Treatment of patients with drug-eluting stent (DES) in-stent restenosis (ISR) is more challenging than that of patients with bare-metal stent ISR. However, the results of everolimus-eluting stents (EES) in these distinct scenarios remain unsettled.

    Methods and Results—

    A pooled analysis of the RIBS IV (Restenosis Intra-Stent of Drug-Eluting Stents: Paclitaxel-Eluting Balloon vs Everolimus-Eluting Stent) and RIBS V (Restenosis Intra-Stent of Bare Metal Stents: Paclitaxel-Eluting Balloon vs Everolimus-Eluting Stent) randomized trials was performed using patient-level data to compare the efficacy of EES in bare-metal stent ISR and DES-ISR. Inclusion and exclusion criteria were identical in both trials. Results of 94 patients treated with EES for bare-metal stent ISR were compared with those of 155 patients treated with EES for DES-ISR. Baseline characteristics were more adverse in patients with DES-ISR, although they presented later and more frequently with a focal pattern. After intervention, minimal lumen diameter (2.22±0.5 versus 2.38±0.5 mm, P=0.01) was smaller in the DES-ISR group. Late angiographic findings (89.3% of eligible patients), including minimal lumen diameter (2.03±0.7 versus 2.36±0.6 mm, P<0.001) and diameter stenosis (23±22 versus 13±17%, P<0.001) were poorer in patients with DES-ISR. Results were consistent in the in-segment and in-lesion analyses. On multiple linear regression analysis, minimal lumen diameter at follow-up remained significantly smaller in patients with DES-ISR. Finally, at 1-year clinical follow-up (100% of patients), mortality (2.6 versus 0%, P<0.01) and need for target vessel revascularization (8 versus 2%, P=0.03) were higher in the DES-ISR group.

    Conclusions—

    This patient-level pooled analysis of the RIBS IV and RIBS V randomized clinical trials suggests that EES provide favorable outcomes in patients with ISR. However, the results of EES are less satisfactory in patients with DES-ISR than in those with bare-metal stent ISR.

    Clinical Trial Registration—

    URL: http://www.clinicaltrials.gov. Unique identifiers: NCT01239953 and NCT01239940.

    Introduction

    WHAT IS KNOWN

    • Treatment of patients with in-stent restenosis (ISR) is associated with good acute results but with high recurrence rates.

    • Currently, the use of drug-eluting balloons or drugeluting stents is recommended in patients with ISR.

    • Treatment of patients presenting with ISR of a drugeluting stent seems to be more challenging than that of patients with bare-metal stent ISR.

    WHAT THE STUDY ADDS

    • Everolimus-eluting stents provide excellent longterm clinical and angiographic results in patients presenting with ISR.

    • The long-term angiographic results of everolimuseluting stents are significantly worse in patients presenting with drug-eluting stent ISR than in those with bare-metal stent ISR.

    Stent implantation constitutes the default strategy during coronary interventions.1,2 Drug-eluting stents (DES) are widely used because of their unique ability to inhibit neointimal proliferation and reduce the restenosis risk.1,2 However, although rare, in-stent restenosis (ISR) still occurs after DES implantation.3,4 Of concern, several studies have suggested that treatment of patients with DES-ISR is more challenging than treatment of patients with bare-metal stent (BMS) ISR.59 Indeed, long-term clinical and angiographic results of patients treated for DES-ISR are significantly poorer than those seen in patients treated for BMS-ISR.59 These findings have been demonstrated in patients with ISR treated with first-generation DES.5,6 Recent studies suggest that drug-eluting balloons (DEB), another well-established therapy in this scenario, are also associated with poorer long-term clinical and angiographic outcomes in patients with DES-ISR than in those with BMS-ISR.79 The reasons accounting for the adverse results in patients with DES-ISR remain unsettled. Patients with ISR after DES may be more complex from a clinical and anatomic stand point and, therefore, prone to recurrences.4 These patients may also exhibit some drug resistance that would explain a poorer response to subsequent interventions, involving drug administration into the vessel wall.4 Finally, the presence of an untoward anatomic substrate, namely neoatherosclerosis, has also been advocated to account for these unfavorable findings.10,11 Indeed, neoatherosclerosis seems to be more prevalent in patients with DES-ISR, and treatment of this unique pathological substrate seems to be particularly challenging.4,10,11

    See Editorial by Waksman and Steinvil

    New-generation DES are safer and more effective than the first-generation DES.1214 Recently, the use of new-generation DES has also been advocated in patients with ISR.1517 The RIBS V (Restenosis Intra-Stent of Bare Metal Stents: Paclitaxel-Eluting Balloon vs Everolimus-Eluting Stent) trial demonstrated that in patients with BMS-ISR, everolimus-eluting stents (EES) provided superior late angiographic findings compared with DEB.16 Likewise, the RIBS IV (Restenosis Intra-Stent of Drug-Eluting Stents: Paclitaxel-Eluting Balloon vs Everolimus-Eluting Stent) trial demonstrated that in patients with DES-ISR, EES obtain not only better long-term angiographic results but also superior 1-year clinical outcomes compared with DEB.17 Accordingly, the use of EES emerges as a attractive therapeutic strategy for patients presenting with ISR.1517

    Whether the results obtained with EES are similar in patients with DES-ISR and BMS-ISR remain unknown. To elucidate the relative efficacy of EES in patients with DES-ISR and BMS-ISR, we performed this pooled analysis of the RIBS IV and RIBS V randomized trials using patient-level data.

    Methods

    The RIBS V study was a multicenter randomized clinical trial that compared DEB versus EES in patients with BMS-ISR,16 (ClinicalTrials.gov identifier: NCT01239953; Data Supplement). The RIBS IV study was another multicenter randomized clinical trial that compared DEB versus EES in patients with DES-ISR,17 (ClinicalTrials.gov identifier: NCT01239940; Data Supplement). Both trials had identical inclusion/exclusion criteria and were conducted simultaneously in 25 Spanish sites.16,17 In RIBS V, 189 patients with BMS-ISR were randomly allocated to DEB (n=95) or EES (n= 94) from January 2010 to January 2012.16 In RIBS IV, 309 patients with DES-ISR were randomly assigned to DEB (n=154) or EES (n=155) from January 2010 to August 201317 (Figure 1). In both trials, patients with ISR (>50% diameter stenosis) presenting with angina or documented ischemia were eligible. Any type of BMS with BMS-ISR was eligible in RIBS V, whereas any type of DES with DES-ISR was eligible in RIBS IV. Patients with small vessels (≤2.0 mm in diameter), long lesions (>30 mm in length), or total occlusions (Thrombolysis in Myocardial Infarction [TIMI]=0) were excluded.16,17 Likewise, patients with early (<1 month) ISR, those presenting as an acute myocardial infarction, or showing a large thrombus on angiography were excluded. However, patients with recurrent ISR and those with edge-ISR were eligible. Finally, patients with contraindications for prolonged dual-antiplatelet therapy, those with serious medical conditions and short life expectancy (<1 year), and those with potential difficulties for a late angiographic follow-up were excluded.16,17 Written informed consent was obtained in all patients.

    Figure 1.

    Figure 1. Flow diagram depicting patients treated with everolimus-eluting stents (EES) and those undergoing final angiographic evaluation in the RIBS V (Restenosis Intra-Stent of Bare Metal Stents: Paclitaxel-Eluting Balloon vs Everolimus-Eluting Stent) and RIBS IV (Restenosis Intra-Stent of Drug-Eluting Stents: Paclitaxel-Eluting Balloon vs Everolimus-Eluting Stent) trials. Angio indicates angiographic; BMS, bare-metal stent; DES, drug-eluting stent; FU, follow-up; ISR, in-stent restenosis; Pts, patients; and QCA, quantitative coronary angiography.

    Data monitoring was organized from the coordinating center (Hospital Universitario Clínico San Carlos, Madrid, Spain).16,17 Both studies were independent, investigators-driven initiatives, developed under the auspices of the Working Group on Interventional Cardiology of the Spanish Society of Cardiology. Unrestricted research grants were obtained from B. Braun Surgical and Abbott Vascular. The Fundación Interhospitalaria de Investigación Cardiovascular was the sponsor. The studies were approved by the Institutional Ethics Committee of the corresponding centers. In both studies, the primary end point was the comparison of the in-segment minimal lumen diameter at late angiography between EES and DEB.16,17

    Interventions

    The interventional protocol was identical in the 2 trials.16,17 Briefly, all patients were pretreated with aspirin and clopidogrel and received intraprocedural heparin targeting for an activated clotting time >250 s. The protocol highlighted the importance of ensuring adequate les sion preparation but, at the same time, emphasized the prevention of any damage on the adjacent coronary segments. Whenever underexpanded stents were identified, aggressive high-pressure dilations with noncompliant balloons were recommended.16,17 Patients allocated to EES (Xience Prime, Abbott Vascular, IL) were treated using a 1.1:1 balloon:artery ratio and high (>14 bar) pressures. Systematic postdilation with noncompliant balloons at high pressures was recommended but eventually left to the operator's discretion according to procedural results. In both trials, a liberal use of intracoronary imaging was suggested, although again eventually left a discretion of the operator.16,17

    Serial assessment of cardiac enzymes and 12-lead electrocardiograms were obtained for 24 hours. Patients were asked to take aspirin indefinitely. The use of clopidogrel was recommended for 1 year after EES implantation.16,17

    Follow-Up

    In both studies, late angiographic follow-up was scheduled after 6 to 9 months or earlier when clinically indicated. Electronic case report forms were identical, and a dedicated uniform database was used in both studies.16,17 Events were adjudicated after the review of the corresponding source documents by a Clinical Events Committee that was blinded to treatment allocation. Deaths were classified as cardiac unless a clear noncardiac cause could be demonstrated. The diagnosis of myocardial infarction has remained unchanged in all the RIBS trials.1517 This required 3 of the following: (1) prolonged (>30 minutes) chest pain, (2) rise in creatine kinase levels more than twice the local upper normal values (with abnormal MB fraction), and (3) development of new persistent ischemic ECG changes (with or without new pathological Q waves).1517 Any revascularization during follow-up had to be clinically justified. All the angiograms of patients requiring target vessel revascularization were systematically reviewed to assess the precise location of the reintervention. The Academic Research Consortium definitions18 were used to classify stent thrombosis.

    Angiographic Analysis

    All angiograms were analyzed at the central core laboratory by trained personnel using a validated methodology.1517 Lesions were characterized using both the Mehran19 and the American College of Cardiology/the American Heart Association20 classifications. Optimal, carefully selected, orthogonal projections were obtained before the procedure, after interventions, and at late follow-up after the administration of intracoronary nitroglycerin. An automatic edge detection quantitative angiographic system (CAAS II System, Pie Medical, Maastricht, The Netherlands) was used.16,17 Angiographic analyses included both in-lesion and in-segment (lesion+treated segment+5-mm margins) measurements.

    Statistical Analyses

    Continuous data are presented as mean (SD) or median (25th to 75th percentile) according to the distribution. Normality was assessed using the Kolmogorov–Smirnov test. Continuous data were compared with the Student test or the Wilcoxon rank-sum test as required. Categorical variables are expressed as counts and percentages and were compared with the Pearson χ2 test or Fisher exact test. Results of the primary end point (in-segment minimal lumen diameter at follow-up) were analyzed in 10 prespecified clinically relevant patient subgroups.16,17 A multiple linear regression analysis was performed to account for the influence of potential confounders in the comparison of the minimal lumen diameter at follow-up in patients with BMS-ISR and DES-ISR. In this model, all variables showing an uneven distribution between the 2 groups (P<0.1) at univariate analysis (diabetes mellitus, hyperlipidemia, cigarette smoking, time to restenosis, >1 stent in the target lesion, left ventricular ejection fraction, edge ISR, length of the previous stent, ISR lesion length, and minimal lumen diameter before intervention) were included. The Kaplan–Meier method was used to determine the event-free survival in the 2 groups, and results were compared with the log-rank and Berslow exact tests. Time-related events were also compared with Cox proportional hazards regression analyses. The SPSS statistical package (version 15.00) was used. Because of the low number of events at follow-up, risk ratio and hazard ratio (95% confidence interval [CI]) were calculated using exact methods with the STATA 12.0 statistical package. A P value of <0.05 was considered as statistically significant.

    Results

    Baseline characteristics of patients with BMS-ISR and DES-ISR are summarized in Table 1. Patients with DES-ISR had more adverse baseline characteristics, including diabetes mellitus, hyperlipidemia, multiple previous interventions on the target lesion, longer initial stents, and more frequently edge ISR. However, patients with DES-ISR were treated later from initial stent implantation (Table 1). Procedural success was obtained in all patients. Qualitative and quantitative angiographic findings are presented in Table 2. In the DES-ISR group, lesions were more severe about both minimal lumen diameter and percent diameter stenosis. However, focal lesions were more frequently found in the DES-ISR group and required the use of shorter EES. In addition, although a smaller maximal balloon:artery ratio was used in patients with DES-ISR, eventually higher final pressures were required in this group (Table 1). Angiographic results immediately after intervention are presented in Table 2. The in-segment minimal lumen diameter after the procedure was smaller in the DES-ISR group (Figure 2).

    Table 1. Baseline Clinical, Angiographic, and Procedural Characteristics

    CharacteristicBMS-ISR (n=94)DES-ISR (n=155)P Value
    Age, y64±1266±100.20
    Female sex, n (%)12 (13)25 (16)0.47
    Risk factors, n (%)
     Diabetes mellitus19 (20)66 (42)<0.0001
     Insulin dependent6 (6)29 (19)0.0007
     Hyperlipidemia62 (66)121 (78)0.04
     Hypertension68 (72)121 (78)0.31
     Ever smoked70 (75)87 (56)0.004
    Clinical presentation, n (%)0.34
     Unstable angina42 (45)79 (51)
     Stable angina/Silent ischemia52 (55)76 (49)
    Previous myocardial infarction56 (60)77 (50)0.13
    Previous bypass surgery7 (7)17 (11)0.36
    >1 Stent on target lesion2 (2)18 (12)0.008
    Time-to-restenosis: d (median, 25th–75th percentile)350 (151–2680)550 (279–1670)0.001
    Ejection fraction, %59±1258±110.53
    Target artery, n (%)0.78
     Left anterior descending37 (39)71 (46)
     Left circumflex22 (23)34 (22)
     Right coronary32 (34)45 (29)
     Saphenous vein graft3 (3)5 (3)
    Procedural characteristics
     Length of the previous stent, mm18±621±70.001
     Length of the current EES, mm23±919±8<0.001
     Maximal pressure, atm19±320±40.08
     Inflation time, s66±4959±440.26
     Balloon:artery ratio1.23±0.21.17±0.20.04
     Angiographic success94 (100)155 (100)1

    BMS indicates bare-metal stent; DES, drug-eluting stent; EES, everolimus-eluting stents; and ISR, in-stent restenosis.

    Table 2. Angiographic Results

    VariableBMS-ISR (n=94)DES-ISR (n=155)P Value
    Qualitative features
     Mehran I, II, III–IV34 (36), 42 (45), 18 (19)99 (64), 44 (28), 12 (8)<0.001
     B2-C lesion51 (54)69 (46)0.14
     Edge-ISR7 (7)39 (25)<0.001
    Quantitative findings
     Before the procedure (in-segment)
      Reference vessel diameter, mm2.64±0.52.67±0.50.71
      Minimal lumen diameter, mm0.93±0.40.75±0.40.001
      Stenosis (% of lumen diameter)65±1372±15<0.001
      Lesion length, mm13.8±6.410.7±5.4<0.001
      Diffuse lesions (≥10 mm), n (%)66 (71)64 (42)<0.001
     Before the procedure (in-lesion)
      Reference vessel diameter, mm2.63±0.52.66±0.50.60
      Minimal lumen diameter, mm0.93±0.40.75±0.4<0.001
      Stenosis (% of lumen diameter)65±1372±15<0.001
     After the procedure (in-segment)
      Reference vessel diameter, mm2.68±0.52.55±0.50.05
      Minimal lumen diameter, mm2.38±0.52.22±0.50.01
      Stenosis (% of lumen diameter)11±1113±110.14
      Acute gain, mm1.45±0.51.48±0.60.72
     After the procedure (in-lesion)
      Reference vessel diameter, mm2.76±0.52.70±0.50.37
      Minimal lumen diameter, mm2.50±0.42.49±0.50.84
      Stenosis (% of lumen diameter)8±138±110.55
      Acute gain, mm1.57±0.51.75±0.60.01
     At follow-up (in-segment)(n=86)(n=133)
      Reference vessel diameter, mm2.74±0.52.66±0.60.33
      Minimal lumen diameter, mm2.36±0.62.03±0.7<0.001
      Stenosis (% of lumen diameter)13±1723±22<0.001
      Restenosis, n (%)4 (4.7)15 (11.3)0.09
      Late loss, mm (mean/median, 25th–75th percentile)0.04±0.4/0.001 (−0.21 to 0.19)0.18±0.6/0.06 (−0.20 to 0.40)0.15
      Loss index (mean/median, 25th–75th percentile)0.002±0.4/0.001 (−0.15 to 0.14)0.03±0.6/0.04 (−0.14 to 0.27)0.11
      Net gain, mm1.41±0.61.28±0.70.16
     At follow-up (in-lesion)
      Reference vessel diameter, mm2.81±0.52.76±0.50.48
      Minimal lumen diameter, mm2.45±0.62.20±0.70.009
      Stenosis (% of lumen diameter)13±1520±220.02
      Restenosis, n (%)4 (4.7)13 (9.8)0.17
      Late loss, mm (mean/median, 25th–75th percentile)0.08±0.5/0.01 (−0.21 to 0.25)0.28±0.7/0.15 (−0.20 to 0.59)0.04
      Loss index (mean/median, 25th–75th percentile)0.03±0.4/0.007 (−0.14 to 0.15)0.08±0.6/0.09 (−0.12 to 0.30)0.04
      Net gain, mm1.45±0.71.49±0.60.71

    BMS indicates bare-metal stent; DES, drug-eluting stent; and ISR, in-stent restenosis.

    Figure 2.

    Figure 2. Cumulative frequency distribution curves of in-segment minimal lumen diameter (MLD) in patients treated with everolimus-eluting stents for bare-metal stent (BMS) in-stent restenosis (ISR; blue) and drug-eluting stent (DES)-ISR (red). The in-segment MLD was the primary end point of the RIBS IV (Restenosis Intra-Stent of Drug-Eluting Stents: Paclitaxel-Eluting Balloon vs Everolimus-Eluting Stent) and RIBS V (Restenosis Intra-Stent of Bare Metal Stents: Paclitaxel-Eluting Balloon vs Everolimus-Eluting Stent) trials. FU indicates at late angiographic follow-up (dashed lines); POST, immediately after intervention; and PRE, baseline.

    Late angiographic findings (obtained in 89.3% of eligible patients) were largely favorable in both groups. However, late angiographic results remained poorer in the DES-ISR group. At follow-up, patients treated for DES-ISR had smaller in-segment minimal lumen diameter (primary objective of this study; Figure 2). Of interest, despite a similar residual % diameter stenosis immediately after interventions, patients treated for DES-ISR developed a higher diameter stenosis at late follow-up (Figure 3). Consistent trends were found in all the remaining late angiographic parameters, including late loss, loss index, and restenosis rate. The poorer late angiographic results of the DES-ISR group were also confirmed in the in-lesion analysis (Table 2; Figures I and II in the Data Supplement).

    Figure 3.

    Figure 3. Cumulative frequency distribution curves showing in-segment percent diameter stenosis in patients treated with everolimus-eluting stents for bare-metal stent (BMS) in-stent restenosis (ISR; blue) and drug-eluting stent (DES)-ISR (red). FU indicates at late angiographic follow-up (dashed lines); POST, immediately after intervention; PRE, baseline; and RE, binary restenosis rate (<50% diameter stenosis at follow-up).

    The absolute mean difference in the in-segment minimal lumen diameter at follow-up was 0.34 mm (95% CI, 0.16–0.51; P<0.001) in favor of patients with BMS-ISR. A consistent pattern in this parameter was demonstrated across all prespecified relevant clinical and angiographic subgroups (Figure 4). Moreover, on multiple linear regression analysis, after adjusting for the potential influence of the differences in baseline characteristics, the absolute mean difference in minimal lumen diameter at follow-up remained statistically significant in favor of patients treated for BMS-ISR (0.26 mm; 95% CI, 0.007–0.52; P=0.04).

    Figure 4.

    Figure 4. Absolute mean difference (AMD, mm) in minimal lumen diameter at follow-up between patients with bare-metal stent (BMS) in-stent restenosis (ISR) and drug-eluting stent (DES)-ISR treated with everolimus-eluting stents according to the 10 prespecified subgroups. P values indicate the comparison of the difference (AMD) in mm between patients with BMS-ISR and DES-ISR. After formal testing all interactions were not statistically significant. B/A indicates balloon:artery ratio; LAD, left anterior descending coronary artery; RE, restenosis; and UA, unstable angina.

    Clinical results at 1 year were obtained in 100% of patients and, overall, were favorable in both groups (Table 3). However, mortality (cardiac and total) and the rate of target vessel revascularization at 9 months and at 1 year were significantly higher in the DES-ISR group compared with the BMS-ISR group (Table 3; Figure 5). There were 2 cases of definitive/probable stent thrombosis in the DES-ISR group, and none in the BMS-ISR group.

    Table 3. Major Adverse Clinical Events

    EventBMS-ISR (n=94)DES-ISR (n=155)P ValueRR/HR (95% CI)
    Hospital events, n (%)RR (95% CI)
     Death0 (0)0 (0)--
     Cardiac death0 (0)0 (0)--
     Myocardial infarction4 (4.3)1 (0.6)0.140.15 (0.01–1.52)
     TLR0 (0)0 (0)--
     TVR0 (0)0 (0)--
      Coronary angioplasty0 (0)0 (0)--
      Coronary surgery0 (0)0 (0)--
     Composite MACE4 (4.3)1 (0.6)0.140.15 (0.01–1.52)
    Events at 9 mo, n (%)HR (95% CI)
     Death0 (0)3 (1.9)<0.01-
     Cardiac death0 (0)1 (0.6)--
     Myocardial infarction4 (4)2 (1)0.160.30 (0.06–1.62)
     TLR0 (0)7 (5)<0.01-
     TVR1 (1)12 (8)0.017.55 (1.20–58.00)
     Composite MACE TLR4 (4)9 (6)0.621.35 (0.41–4.41)
     Composite MACE TVR5 (5)14 (9)0.321.69 (0.60–4.74)
    Events at 1 y, n (%)
     Death0 (0)4 (2.6)<0.01-
     Cardiac death0 (0)2 (1.3)<0.01-
     Myocardial infarction4 (4)2 (1)0.160.30 (0.06–1.61)
     TLR1 (1)7 (5)0.164.40 (0.55–35.3)
      Coronary angioplasty TLR1 (1)6 (4)0.223.77 (0.46–30.9)
      Coronary surgery TLR0 (0)1 (0.6)<0.05-
     TVR2 (2)13 (8)0.034.11 (1.08–12.22)
     Composite MACE (with TLR)5 (5)10 (7)0.731.21 (0.41–3.54)
     Composite MACE (with TVR)6 (6)16 (10)0.321.62 (0.63–4.17)

    Patients with >1 event are counted only once for the composite clinical end point, although each event is listed separately in the corresponding category. P values from Cox analysis. RR and HR from exact calculations from STATA. (-) indicates undefined; BMS, bare-metal stent; CI, confidence intervals; DES, drug-eluting stent; HR, hazard ratio (events at follow-up); ISR, in-stent restenosis; MACE, major adverse cardiac events (cardiac death, myocardial infarction, and target vessel revascularization [TVR]); RR, risk ratio (hospital events); and TLR, target lesion revascularization.

    Figure 5.

    Figure 5. Freedom from target vessel revascularization (TVR). BMS indicates bare-metal stent; DES, drug-eluting stent; FU, follow-up; and ISR, in-stent restenosis.

    A sensitivity analysis was performed in the DES-ISR group to detect the potential implications of the type of underlying DES (first-generation DES [55%] versus second-generation DES [45%] and also EES DES [33%] versus non-EES DES [67%]) on the results obtained with EES. No significant differences were detected in any late angiographic variable (minimal lumen diameter, % diameter stenosis, restenosis rate, and late lumen loss) according to the type of underlying DES, although all late angiographic findings tended to be numerically worse in patients treated for second-generation DES-ISR and EES-ISR (difference not statistically significant). Likewise, clinical results (individual clinical end points and the main combined outcome measure [cardiac death, myocardial infarction, and target vessel revascularization]) were similar in patients with first-generation DES and second-generation DES and in patients with EES and non-EES ISR. Only the composite of cardiac death, myocardial infarction, and target lesion revascularization was lower in patients with first-generation DES-ISR compared with second-generation DES-ISR (2.5% versus 12%; P=0.02; hazard ratio, 0.2; 95% CI, 0.04–0.94) and in patients with non–EES-ISR versus EES-ISR (4% versus 12.5%; P=0.08; hazard ratio, 3.3; 95% CI, 0.09–1.16).

    Discussion

    This pooled, patient-level, analysis of the RIBS IV and V randomized studies confirms the efficacy of EES in patients with ISR. Our results also provide reassuring evidence of the safety profile of EES in this setting. This is the largest controlled study currently available assessing the value of this new-generation DES in this challenging anatomic scenario. The main novel finding of this study is that EES are less effective in patients with DES-ISR than in those with BMS-ISR. Importantly, the poorer results obtained in patients with DES-ISR were uniformly seen in most late angiographic parameters, including the minimal lumen diameter at follow-up (primary end point) and diameter stenosis. Notably, these angiographic results were consistent in the in-segment and in-lesion analyses, and both provided complementary information of the results of EES in this setting. In addition, this difference in the treatment effect was consistent among 10 prespecified patient subsets and was maintained after adjusting for potential confounders.

    Overall, baseline characteristics were more adverse in patients with DES-ISR than in those with BMS-ISR. However, as demonstrated in most previous studies,3,4 DES-ISR tended to present some benign features, including a focal pattern and a late presentation. Unexpectedly, minimal lumen diameter immediately after intervention was smaller in the DES-ISR arm. This is relevant as in our study optimization strategies, including high dilation pressures with noncompliant balloons, were systematically pursued in both arms. This finding might be a result of more severe and resistant lesions in DES-ISR. Importantly, however, after adjusting for potential clinical and angiographic baseline characteristics, minimal lumen diameter at follow-up remained significantly smaller in patients treated for DES-ISR. Notably, the adverse late angiographic findings seen in patients with DES-ISR eventually translated into a higher requirement of target vessel revascularization during follow-up. This underscores the clinical implications of the poorer late angiographic findings seen in these patients. The higher mortality rate seen in patients with DES-ISR should be interpreted with major caution because our study was clearly underpowered for this hard clinical end point. Further studies are required to fully ascertain the potential implications of treatment of DES-ISR on late clinical events. Finally, our findings were also inconclusive about whether results of EES differ according to the type of DES presenting with ISR. Therefore, further studies are also warranted to address this issue.

    From a methodological stand point, this is a patient-level pooled analysis of 2 randomized trials that used identical inclusion/exclusion criteria and interventional strategies.16,17 Moreover, a uniform methodology was used for the final angiographic analysis and for the assessment of clinical events.16,17 In addition, as a rigorous follow-up protocol, late angiographic results were available in 89.3% of eligible patients, whereas a complete clinical follow-up was obtained in 100% of patients. Furthermore, as interventions were performed in multiple centers and exclusion criteria were relatively limited, we believe that our results nicely represent those that should be expected during routine clinical practice. Altogether, these methodological considerations support the internal and external validity of our results.

    Previous Studies of EES in Patients With ISR

    Evidence on the value of EES in patients with ISR is limited and stems from retrospective observational studies, prospective registries, and scarce randomized clinical trials. In the large Xience V USA prospective multicenter registry,21 the results of EES in patients with ISR (n=383) were compared with those seen in patients with de novo lesions (n=4832). At 1-year target, lesion failure was 2-fold in the ISR group (10.9% versus 4.9%). A recent retrospective observational study including 322 patients with BMS-ISR22 suggested that those treated with EES (n=114) had similar 1-year clinical results compared with those treated with first-generation DES. In that study, however, rates of definitive stent thrombosis were numerically lower after treatment with EES. Another study of 174 patients with BMS-ISR23 compared results obtained with EES (n=79) and paclitaxel DES (n=95). At 1 year, rates of adverse events (4.5 versus 13.6%, P=0.06) and target lesion revascularization (1 versus 11.5%, P=0.02) were lower in the EES group. However, after 2-years, results were similar in the 2 groups. Another observational study by the same group in patients with DES-ISR24 compared the results of 40 patients treated with EES and 46 patients treated with DEB. Patients treated with EES were associated with higher rates of major adverse cardiac events (27.5% versus 8.6%, P=0.046) and target lesion revascularization (22.5% versus 4.3%, P=0.029) at 1 year compared with those treated with DEB. Conversely, in a series of 89 patients with recurrent ISR after the DEB treatment, Kubo et al25 demonstrated that EES were associated with better long-term clinical results than the repeated DEB treatment.

    Interestingly, Almalla et al26 compared the results of EES in patients with ISR according to the type of the underlying stent (64 BMS-ISR and 30 DES-ISR). In this small study, treatment of DES-ISR was associated with higher rates of recurrent target lesion revascularization (23.3% versus 1.6%), myocardial infarction (13.3% versus 0%), and total cardiac events (30% versus 4.6%) at 1 year. Nevertheless, no angiographic follow-up was obtained in this series, and because of the relatively small sample size, the study had no power to assess differences in clinical events.

    In all these previous studies assessing the value of EES in patients with ISR, late angiography was not systematically obtained. Actually, observational studies assessing late angiographic results of EES in patients with ISR are scarce. The XERES registry27 included 97 consecutive patients with BMS-ISR treated with EES at 20 sites in France. At 9-month follow-up, the in-stent late loss (primary end point) was 0.35±0.6 mm and the rate of binary ISR was 12.2%. In another study,28 results of EES (n=91) were compared with those of paclitaxel DES (n=107). In-stent late loss (0.20±0.39 versus 0.18±0.31) and binary restenosis rate (18.0% versus 16.7%) were similar for EES and paclitaxel-DES.

    Recently, some randomized trials confirmed the efficacy of EES in patients with ISR. In the mechanistic SEDUCE (Healing Responses After Treatment of Bare Metal Stent Restenosis With Implantation of an Everolimus-Eluting Xience V Stent Versus Use of a Paclitaxel-Eluting Balloon: Optical Coherence Tomography Study) randomized trial,29 optical coherence tomography was used to assess the healing characteristics of patients with BMS-ISR allocated to EES or DEB (25 patients per arm). DEB were associated with better healing characteristics (strut coverage on optical coherence tomography) but were less effective than EES. The RIBS V trial compared EES with DEB in patients with BMS-ISR,16 demonstrating that EES improved all late angiographic parameters. Alternatively, the TIS (Treatment of Coronary In-Stent Restenosis) trial,30 also in patients with BMS-ISR, recently suggested that late lumen loss was lower with DEB compared with EES, although minimal lumen diameter and % diameter stenosis at late follow-up were similar in both groups. Finally, the RIBS IV trial compared the same strategies in patients with DES-ISR.17 In this study, EES not only improved late angiographic findings but also reduced the need for repeat revascularization.

    BMS-ISR Versus DES-ISR

    Several previous studies suggest that treatment of patients with DES-ISR is more challenging than that of patients with BMS-ISR.59 Patients treated for DES-ISR tend to have higher rates of clinical and angiographic recurrences compared with those treated with DES-ISR.59 The reasons explaining these findings are not fully understood. Patients presenting DES-ISR represent a highly selected subset that developed ISR in spite of the use of a potent antiproliferative agent.3,4 Thus, they might remain vulnerable to recurrences in spite of reinterventions involving repeat drug administration into the vessel wall. Drug resistance has been advocated to explain these findings, but this concept remains largely speculative in the clinical arena.3,4 However, it has been suggested that the underlying substrate may be different in patients with DES-ISR and BMS-ISR.10,11 Indeed, pathological and optical coherence tomography findings suggest that neoatherosclerosis may be more prevalent in patients with DES-ISR.10,11 The therapy of choice for neoatherosclerosis remains unsettled,4 but the occurrence of neoatherosclerosis in patients with DES-ISR might explain their poorer outcomes. Further studies are warranted to elucidate whether the presence of neoatherosclerosis, as unraveled by optical coherence tomography, might be used to select and tailor optimal reinterventions in these complex patients.4

    Clinical Implications

    Currently, the use of both DES and DEB are supported by the same evidence (class I, level A) in patients with BMS-ISR and DES-ISR.14 The RIBS IV and V studies, however, suggest a potential additional benefit of EES, a new-generation DES, in these patients.16,17 Importantly, however, this study demonstrates that the efficacy of EES is reduced in patients with DES-ISR. We found that the poorer late angiographic results found in patients with DES-ISR were, at least in part, a result of their adverse baseline characteristics. This would explain the poorer immediate angiographic results in spite of a similar and aggressive interventional strategy. Therefore, any attempt to optimize final results seems justified in these patients. However, after adjusting for differences in baseline characteristics, the minimal lumen diameter at follow-up (primary end point) was significantly smaller in the DES-ISR group. This strongly suggests an adverse underlying substrate more resistant to treatment and prone to recurrences. Further studies are warranted to refine therapeutic strategies in patients with DES-ISR.

    Study Limitations

    Some important limitations should be acknowledged. First, the current polled study just represents a post hoc subanalysis of 2 previous randomized studies and, therefore, the effect of randomization on potential confounders is lost. Second, the late angiographic superiority of EES in BMS-ISR translated into a significant reduction on the need for target vessel revascularization but only showed a trend on target lesion revascularization. Likewise, caution is required when interpreting the differences in mortality. The lack of power for clinical events, as a result of the relatively small sample size, should be kept in mind. This factor also limits the power of the subgroup analyses within the DES-ISR group to unravel potential implications of the type of underlying DES. Further studies, including larger series with a longer follow-up, will be required to definitively establish the clinical benefit of EES in different subsets of patients with ISR. Finally, the attractiveness of EES compared with DEB in this setting should be balanced with the potential deleterious effects of adding multiple metal layers on the vessel wall.

    Conclusions

    The results of the current pooled analysis of RIBS IV and V confirm the value of EES in patients with ISR. However, our findings demonstrate that the results of EES are significantly poorer in patients with DES-ISR than in those with BMS-ISR.

    Footnotes

    Current address for A.C.: Universidad San Francisco de Quito USFQ, School of Medicine, Quito, Ecuador.

    The Data Supplement is available at http://circinterventions.ahajournals.org/lookup/suppl/doi:10.1161/CIRCINTERVENTIONS.115.003479/-/DC1.

    Correspondence to Fernando Alfonso, MD, Departamento de Cardiología, Hospital Universitario de La Princesa, IIS-IP, Universidad Autónoma de Madrid, c/ Diego de León 62, Madrid 28006, Spain. E-mail

    References

    • 1. Stefanini GG, Holmes DR. Drug-eluting coronary-artery stents.N Engl J Med. 2013; 368:254–265. doi: 10.1056/NEJMra1210816.CrossrefMedlineGoogle Scholar
    • 2. García Del Blanco B, Hernández Hernández F, Rumoroso Cuevas JR, Trillo Nouche R.Spanish Cardiac Catheterization and Coronary Intervention Registry. 24th Official Report of the Spanish Society of Cardiology Working Group on Cardiac Catheterization and Interventional Cardiology (1990–2014).Rev Esp Cardiol (Engl Ed). 2015; 68:1154–1164.MedlineGoogle Scholar
    • 3. Dangas GD, Claessen BE, Caixeta A, Sanidas EA, Mintz GS, Mehran R.In-stent restenosis in the drug-eluting stent era.J Am Coll Cardiol. 2010; 56:1897–1907. doi: 10.1016/j.jacc.2010.07.028.CrossrefMedlineGoogle Scholar
    • 4. Alfonso F, Byrne RA, Rivero F, Kastrati A.Current treatment of in-stent restenosis.J Am Coll Cardiol. 2014; 63:2659–2673. doi: 10.1016/j.jacc.2014.02.545.CrossrefMedlineGoogle Scholar
    • 5. Steinberg DH, Gaglia MA, Pinto Slottow TL, Roy P, Bonello L, De Labriolle A, Lemesle G, Torguson R, Kineshige K, Xue Z, Suddath WO, Kent KM, Satler LF, Pichard AD, Lindsay J, Waksman R.Outcome differences with the use of drug-eluting stents for the treatment of in-stent restenosis of bare-metal stents versus drug-eluting stents.Am J Cardiol. 2009; 103:491–495. doi: 10.1016/j.amjcard.2008.09.107.CrossrefMedlineGoogle Scholar
    • 6. Byrne RA, Cassese S, Windisch T, King LA, Joner M, Tada T, Mehilli J, Pache J, Kastrati A.Differential relative efficacy between drug-eluting stents in patients with bare metal and drug-eluting stent restenosis; evidence in support of drug resistance: insights from the ISAR-DESIRE and ISAR-DESIRE 2 trials.EuroIntervention. 2013; 9:797–802.CrossrefMedlineGoogle Scholar
    • 7. Habara S, Iwabuchi M, Inoue N, Nakamura S, Asano R, Nanto S, Hayashi Y, Shiode N, Saito S, Ikari Y, Kimura T, Hosokawa J, Nakamura M, Kotani J, Kozuma K, Mitsudo K.A multicenter randomized comparison of paclitaxel-coated balloon catheter with conventional balloon angioplasty in patients with bare-metal stent restenosis and drug-eluting stent restenosis.Am Heart J. 2013; 166:527–533. doi: 10.1016/j.ahj.2013.07.002.CrossrefMedlineGoogle Scholar
    • 8. Wöhrle J, Zadura M, Möbius-Winkler S, Leschke M, Opitz C, Ahmed W, Barragan P, Simon JP, Cassel G, Scheller B.SeQuentPlease World Wide Registry: clinical results of SeQuent please paclitaxel-coated balloon angioplasty in a large-scale, prospective registry study.J Am Coll Cardiol. 2012; 60:1733–1738. doi: 10.1016/j.jacc.2012.07.040.CrossrefMedlineGoogle Scholar
    • 9. Habara S, Kadota K, Shimada T, Ohya M, Amano H, Izawa Y, Kubo S, Hyodo Y, Otsuru S, Hasegawa D, Tada T, Tanaka H, Fuku Y, Goto T, Mitsudo K.Late restenosis after paclitaxel-coated balloon angioplasty occurs in patients with drug-eluting stent restenosis.J Am Coll Cardiol. 2015; 66:14–22. doi: 10.1016/j.jacc.2015.05.002.CrossrefMedlineGoogle Scholar
    • 10. Nakazawa G, Otsuka F, Nakano M, Vorpahl M, Yazdani SK, Ladich E, Kolodgie FD, Finn AV, Virmani R.The pathology of neoatherosclerosis in human coronary implants bare-metal and drug-eluting stents.J Am Coll Cardiol. 2011; 57:1314–1322. doi: 10.1016/j.jacc.2011.01.011.CrossrefMedlineGoogle Scholar
    • 11. Otsuka F, Byrne RA, Yahagi K, Mori H, Ladich E, Fowler DR, Kutys R, Xhepa E, Kastrati A, Virmani R, Joner M.Neoatherosclerosis: overview of histopathologic findings and implications for intravascular imaging assessment.Eur Heart J. 2015; 36:2147–2159. doi: 10.1093/eurheartj/ehv205.CrossrefMedlineGoogle Scholar
    • 12. Alfonso F, Fernandez C.Second-generation drug-eluting stents. Moving the field forward.J Am Coll Cardiol. 2011; 58:26–29.CrossrefMedlineGoogle Scholar
    • 13. Palmerini T, Biondi-Zoccai G, Della Riva D, Stettler C, Sangiorgi D, D’Ascenzo F, Kimura T, Briguori C, Sabatè M, Kim HS, De Waha A, Kedhi E, Smits PC, Kaiser C, Sardella G, Marullo A, Kirtane AJ, Leon MB, Stone GW.Stent thrombosis with drug-eluting and bare-metal stents: evidence from a comprehensive network meta-analysis.Lancet. 2012; 379:1393–1402. doi: 10.1016/S0140-6736(12)60324-9.CrossrefMedlineGoogle Scholar
    • 14. Windecker S, Kolh P, Alfonso F, Collet JP, Cremer J, Falk V, Filippatos G, Hamm C, Head SJ, Jüni P, Kappetein AP, Kastrati A, Knuuti J, Landmesser U, Laufer G, Neumann FJ, Richter DJ, Schauerte P, Sousa Uva M, Stefanini GG, Taggart DP, Torracca L, Valgimigli M, Wijns W, Witkowski A.2014 ESC/EACTS Guidelines on myocardial revascularization: The Task Force on Myocardial Revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS) Developed with the special contribution of the European Association of Percutaneous Cardiovascular Interventions (EAPCI).Eur Heart J. 2014; 35:2541–2619.CrossrefMedlineGoogle Scholar
    • 15. Alfonso F, Pérez-Vizcayno MJ, Dutary J, Zueco J, Cequier A, García-Touchard A, Martí V, Lozano I, Angel J, Hernández JM, López-Mínguez JR, Melgares R, Moreno R, Seidelberger B, Fernández C, Hernandez R; RIBS-III Study Investigators (under the auspices of the Working Group on Interventional Cardiology of the Spanish Society of Cardiology). Implantation of a drug-eluting stent with a different drug (switch strategy) in patients with drug-eluting stent restenosis. Results from a prospective multicenter study (RIBS III [Restenosis Intra-Stent: Balloon Angioplasty Versus Drug-Eluting Stent]).JACC Cardiovasc Interv. 2012; 5:728–737. doi: 10.1016/j.jcin.2012.03.017.CrossrefMedlineGoogle Scholar
    • 16. Alfonso F, Pérez-Vizcayno MJ, Cárdenas A, García Del Blanco B, Seidelberger B, Iñiguez A, Gómez-Recio M, Masotti M, Velázquez MT, Sanchís J, García-Touchard A, Zueco J, Bethencourt A, Melgares R, Cequier A, Dominguez A, Mainar V, López-Mínguez JR, Moreu J, Martí V, Moreno R, Jiménez-Quevedo P, Gonzalo N, Fernández C, Macaya C; RIBS V Study Investigators, under the auspices of the Working Group on Interventional Cardiology of the Spanish Society of Cardiology. A randomized comparison of drug-eluting balloon versus everolimus-eluting stent in patients with bare-metal stent in-stent restenosis: the RIBS V Clinical Trial (Restenosis Intra-Stent of Bare Metal Stents: Paclitaxel-Eluting Balloon vs Everolimus-Eluting Stent).J Am Coll Cardiol. 2014; 63:1378–1386. doi: 10.1016/j.jacc.2013.12.006.CrossrefMedlineGoogle Scholar
    • 17. Alfonso F, Pérez-Vizcayno MJ, Cárdenas A, García del Blanco B, García-Touchard A, López-Minguéz JR, Benedicto A, Masotti M, Zueco J, Iñiguez A, Velázquez M, Moreno R, Mainar V, Domínguez A, Pomar F, Melgares R, Rivero F, Jiménez-Quevedo P, Gonzalo N, Fernández C, Macaya C; RIBS IV Study Investigators (under auspices of Interventional Cardiology Working Group of Spanish Society of Cardiology). A prospective randomized trial of drug-eluting balloons versus everolimus-eluting stents in patients with in-stent restenosis of drug-eluting stents: the RIBS IV randomized clinical trial.J Am Coll Cardiol. 2015; 66:23–33. doi: 10.1016/j.jacc.2015.04.063.CrossrefMedlineGoogle Scholar
    • 18. Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, van Es GA, Steg PG, Morel MA, Mauri L, Vranckx P, McFadden E, Lansky A, Hamon M, Krucoff MW, Serruys PW; Academic Research Consortium. Clinical end points in coronary stent trials: a case for standardized definitions.Circulation. 2007; 115:2344–2351. doi: 10.1161/CIRCULATIONAHA.106.685313.LinkGoogle Scholar
    • 19. Mehran R, Dangas G, Abizaid AS, Mintz GS, Lansky AJ, Satler LF, Pichard AD, Kent KM, Stone GW, Leon MB.Angiographic patterns of in-stent restenosis: classification and implications for long-term outcome.Circulation. 1999; 100:1872–1878.LinkGoogle Scholar
    • 20. Alfonso F, Cequier A, Angel J, Martí V, Zueco J, Bethencourt A, Mantilla R, López-Minguez JR, Gómez-Recio M, Morís C, Perez-Vizcayno MJ, Fernández C, Macaya C, Seabra-Gomes R; Restenosis Intra-stent Balloon Angioplasty Versus Elective Stenting (RIBS) Investigators. Value of the American College of Cardiology/American Heart Association angiographic classification of coronary lesion morphology in patients with in-stent restenosis. Insights from the restenosis intra-stent balloon angioplasty versus elective stenting (RIBS) randomized trial.Am Heart J. 2006; 151:681.e1–681.e9. doi: 10.1016/j.ahj.2005.10.014.CrossrefMedlineGoogle Scholar
    • 21. Lee MS, Yang T, Mahmud E, Park KW, Kim HS, Kim MH, Dangas G, Hermiller J, Krucoff M, Rutledge D.Clinical outcomes in the percutaneous coronary intervention of in-stent restenosis with everolimus-eluting stents.J Invasive Cardiol. 2014; 26:420–426.MedlineGoogle Scholar
    • 22. Ota H, Mahmoudi M, Torguson R, Satler LF, Suddath WO, Pichard AD, Waksman R.Safety and efficacy of everolimus-eluting stents for bare-metal in-stent restenosis.Cardiovasc Revasc Med. 2015; 16:151–155. doi: 10.1016/j.carrev.2015.02.005.CrossrefMedlineGoogle Scholar
    • 23. Almalla M, Schröder JW, Pross V, Stegemann E, Marx N, Hoffmann R.Everolimus-eluting versus paclitaxel-eluting stents for treatment of bare metal stent restenosis.Am J Cardiol. 2011; 108:518–522. doi: 10.1016/j.amjcard.2011.03.080.CrossrefMedlineGoogle Scholar
    • 24. Almalla M, Schröder J, Pross V, Marx N, Hoffmann R.Paclitaxel-eluting balloon versus everolimus-eluting stent for treatment of drug-eluting stent restenosis.Catheter Cardiovasc Interv. 2014; 83:881–887. doi: 10.1002/ccd.25072.CrossrefMedlineGoogle Scholar
    • 25. Kubo S, Kadota K, Otsuru S, Hasegawa D, Habara S, Tada T, Tanaka H, Fuku Y, Katoh H, Goto T, Mitsudo K.Everolimus-eluting stent implantation versus repeat paclitaxel-coated balloon angioplasty for recurrent in-stent restenosis lesion caused by paclitaxel-coated balloon failure.EuroIntervention. 2015; 10:e1–e8. doi: 10.4244/EIJV10I9A180.CrossrefMedlineGoogle Scholar
    • 26. Almalla M, Pross V, Marx N, Hoffmann R.Effectiveness of everolimus-eluting stents in the treatment of drug-eluting stent versus bare-metal stent restenosis.Coron Artery Dis. 2012; 23:492–496. doi: 10.1097/MCA.0b013e328358a58f.CrossrefMedlineGoogle Scholar
    • 27. Carrié D, Delarche N, Piot C, Berland J, Menneveau N, Bonello L, Py A, Teiger E, Leborgne L, Bayet G, Wittenberg O, Schiele F.Everolimus-eluting stent for the treatment of bare metal in-stent restenosis: clinical and angiographic outcomes at nine-month follow-up of XERES (Xience evaluation in bare metal stent restenosis) trial.EuroIntervention. 2014; 10:700–708. doi: 10.4244/EIJV10I6A122.CrossrefMedlineGoogle Scholar
    • 28. Markovic S, Paliskyte R, Rottbauer W, Wöhrle J.Everolimus-eluting stents compared with paclitaxel-eluting stents for treatment of coronary in-stent restenoses.Cardiovasc Revasc Med. 2012; 13:307–310. doi: 10.1016/j.carrev.2012.09.003.CrossrefMedlineGoogle Scholar
    • 29. Adriaenssens T, Dens J, Ughi G, Bennett J, Dubois C, Sinnaeve P, Wiyono S, Coosemans M, Belmans A, D’hooge J, Vrolix M, Desmet W.Optical coherence tomography study of healing characteristics of paclitaxel-eluting balloons vs. everolimus-eluting stents for in-stent restenosis: the SEDUCE (safety and efficacy of a drug eluting balloon in coronary artery restenosis) randomised clinical trial.EuroIntervention. 2014; 10:439–448. doi: 10.4244/EIJV10I4A77.CrossrefMedlineGoogle Scholar
    • 30. Pleva L, Kukla P, Kusnierova P, Zapletalova J, Hlinomaz O.Comparison of the efficacy of paclitaxel-eluting balloon catheters and everolimus-eluting stents in the treatment of coronary in-stent restenosis: the treatment of in-stent restenosis study.Circ Cardiovasc Interv. 2016; 9:. doi: 10.1161/CIRCINTERVENTIONS.115.003316.LinkGoogle Scholar