Author + information
- Received February 3, 2015
- Revision received June 3, 2015
- Accepted July 17, 2015
- Published online November 1, 2015.
- Harald Rittger, MD∗∗ (, )
- Matthias Waliszewski, PhD†,
- Johannes Brachmann, MD‡,
- Wolfgang Hohenforst-Schmidt, MD∗,
- Marc Ohlow, MD§,
- Andreas Brugger, MD‖,
- Holger Thiele, MD¶,
- Ralf Birkemeyer, MD#,
- Volkhard Kurowski, MD∗∗,
- Christian Schlundt, MD††,
- Stefan Zimmermann, MD‡‡,
- Sandra Lonke, MD∗,
- Moritz von Cranach, MD§§,
- Sinisa Markovic, MD‖‖,
- Werner G. Daniel, MD††,
- Stephan Achenbach, MD†† and
- Jochen Wöhrle, MD‖‖
- ∗Medizinische Klinik I, Klinikum Fuerth, Fuerth, Germany
- †Medical Scientific Affairs, B. Braun Vascular Systems, Berlin, Germany
- ‡Medizinische Klinik II, Klinikum Coburg, Coburg, Germany
- §Klinik für Kardiologie, Zentralklinikum Bad Berka, Bad Berka, Germany
- ‖Klinik für Kardiologie, Klinikum Kulmbach, Kulmbach, Germany
- ¶Medizinische Klinik II, Universitätsklinikum Lübeck, Lübeck, Germany
- #Herzklinik Ulm, Ulm, Germany
- ∗∗Abteilung für Innere Medizin/Kardiologie, Krankenhaus Mölln-Ratzeburg, Mölln-Ratzeburg, Germany
- ††Medizinische Klinik 2, Universitätsklinikum Erlangen, Erlangen, Germany
- ‡‡Klinik für Innere/Medizin Kardiologie, Diakonissenhaus Flensburg, Flensburg, Germany
- §§Klinik für Neurologie, Universitätsklinikum Freiburg, Freiburg, Germany
- ‖‖Klinik für Innere Medizin II, Universitätsklinikum Ulm, Ulm, Germany
- ↵∗Reprint requests and correspondence:
Dr. Harald Rittger, Medizinische Klinik I, Klinikum Fuerth, Jakob-Henle-Strasse 1, Fuerth D-90766, Germany.
Objectives The intention this PEPCAD-DES (Treatment of Drug-eluting Stent [DES] In-Stent Restenosis With SeQuent Please Paclitaxel Eluting Percutaneous Transluminal Coronary Angioplasty [PTCA] Catheter) study update was to demonstrate the safety and efficacy of paclitaxel-coated balloon (PCB) angioplasty in patients with DES-ISR at 3 years.
Background In the PEPCAD-DES trial late lumen loss and the need for repeat target lesion revascularization (TLR) was significantly reduced with PCB angioplasty compared with plain old balloon angioplasty (POBA) in patients with drug-eluting stent in-stent restenosis (DES-ISR) at 6 months. We evaluated whether the clinical benefit of reduced TLR and major adverse cardiac events (MACE) was maintained up to 3 years.
Methods A total of 110 patients with DES-ISR in native coronary arteries with reference diameters ranging from 2.5 mm to 3.5 mm and lesion lengths ≤22 mm were randomized to treatment with either PCB or POBA in a multicenter, randomized, single-blind clinical study. With a 2:1 randomization, 72 patients were randomized to the PCB group and 38 patients to the POBA group. At baseline, there were lesions with at least 2 stent layers in PCB (52.8%, 38 of 72) and POBA (55.3%, 21 of 38) patients.
Results At 36 months, the TLR rates were significantly lower in the PCB group compared with the POBA control group (19.4% vs. 36.8%; p = 0.046). Multiple TLRs in individual patients were more frequent in the POBA group compared with the PCB group (more than 1 TLR: POBA, 13.2%; PCB, 1.4%; p = 0.021). The 36-month MACE rate was significantly reduced in the PCB group compared with the POBA group (20.8% vs. 52.6%, log-rank p = 0.001).
Conclusions PCB angioplasty was superior to POBA for the treatment of DES-ISR patients in terms of MACE and TLR for up to 36 months. There was no late catch-up phenomenon. (Treatment of Drug-eluting Stent [DES] In-Stent Restenosis With SeQuent® Please Paclitaxel Eluting Percutaneous Transluminal Coronary Angioplasty [PTCA] Catheter [PEPCAD-DES]; NCT00998439)
In the PEPCAD-DES DES (Treatment of Drug-eluting Stent [DES] In-Stent Restenosis With SeQuent Please Paclitaxel Eluting Percutaneous Transluminal Coronary Angioplasty [PTCA] Catheter), late lumen loss (LLL) and the need for repeat target lesion revascularization (TLR) at 6 months were significantly reduced with paclitaxel-coated balloon (PCB) angioplasty as compared with plain old balloon angioplasty (POBA) for drug-eluting stent in-stent restenosis (DES-ISR) (1). However, LLL in the PCB population was higher than other reported LLL findings after PCB angioplasty in bare metal stent ISR (2–5) and DES-ISR (6–8). This difference may be explained with the inclusion of restenotic DES with different antiproliferative drugs. Moreover, the initial DES implantation in complex lesions with a high risk of restenosis such as multiple stent layers (Table 1) may also have contributed to the higher level of LLL. In this paper, the long-term clinical outcomes 3 years after initial treatment and additional findings are reported as well as whether the efficacy of PCB is different in paclitaxel-eluting stent (PES) restenosis or non-PES restenosis, whether PCB angioplasty is equally effective in single-layer versus multilayer DES-ISR, and whether there is a difference between diabetic and nondiabetic patients with DES-ISR when treated with PCB versus POBA.
From November 2009 to April 2011, 110 patients with restenosis of sirolimus-eluting stents, everolimus-eluting stents, or paclitaxel-eluting stents (PES) in a native coronary artery with the indication for percutaneous coronary intervention were randomized to treatment with either PCB or POBA in a multicenter, randomized, clinical single- blind study. With a 2:1 randomization, 72 patients were treated with PCB and 38 patients with POBA. Patients with sirolimus-eluting Cypher stent ISR (Cordis, Warren, New Jersey) or Yukon stent ISR (Translumina, Hechingen, Germany), everolimus-eluting Xience stent ISR (Abbott Vascular, Abbott Park, Illinois), or paclitaxel-eluting Taxus stent ISR (Boston Scientific, Natick, Massachusetts), with a reference vessel diameter of 2.5 to 3.5 mm and a lesion length less than 22 mm were included.
The inclusion and exclusion criteria, as well as the study design and the study procedure were reported previously (1). The protocol was approved by all ethics committees. All patients gave written informed consent.
Patients were scheduled for a 6-month angiographic follow-up and clinical follow-up at 12 and 36 months. Angiographic measurements were done with the CAAS 5.7 software (Pie Medical Imaging BV, Maastricht, the Netherlands) in the core lab of the University of Ulm, Ulm, Germany (9,10).
A major adverse cardiac event (MACE) was defined as a composite of cardiac death, myocardial infarction attributed to the target vessel, TLR, or a target vessel total occlusion. TLR was defined as the composite of TLR by means of percutaneous coronary intervention or coronary artery bypass grafting. Biochemical evidence of myocardial necrosis was defined as creatine-kinase myocardial band ≥3 times the upper normal limit or a total creatine kinase ≥3 times the upper normal limit whenever creatine-kinase myocardial band was not available. Clinical follow-up at 36 months was done by a telephone call with the patient, the patients' relatives, or the primary physician. A priori–defined subgroup analyses were based on paclitaxel-eluting stent in-stent restenosis (PES-ISR) versus non–PES-ISR, single-layer versus multilayer DES-ISR, and diabetic versus nondiabetic patients. Focal patterns of in-stent restenosis were defined as Mehran classes IA through ID, whereas diffuse patterns were described by Mehran classes II through IV (11).
Based on the Satterthwaite t test, a superiority test hypothesis was formulated. With a 2:1 randomization (PCB vs. POBA) and an assumed 6-month in-lesion LLL (primary endpoint) of 0.20 ± 0.30 mm in the PCB group (4) and 0.80 ± 0.80 mm in the POBA group (2), 64 patients were calculated for the PCB and 34 for the POBA group to achieve 90% power. Assuming a dropout rate of 10%, a total of 71 patients in the PCB group and 38 in the POBA group were estimated (1). For logistic regression analyses with 36-month MACE and 12-month TLR as the binary response variables, we considered a Nagelkerke R2 >0.2 as sufficient for selected explanatory variables to adequately explain the variance in the data. Categorical variables were compared using the Pearson chi-square test or Fisher exact test, as appropriate. All continuous variables are given as mean ± SD. Differences between proportions and t tests, logistic regression, and univariate analyses were computed with SPSS version 20.0 (IBM, Munich, Germany), whereas sample sizes were estimated with nQuery Advisor version 7.0 (Statistical Solutions Ltd., Cork, Ireland).
TLR and MACE rates at 36 months
Six-month clinical results showed a significant benefit of PCB compared with POBA (1). At 6 months, TLR rates were 15.3% (11 of 72) versus 36.8% (14 of 38) and MACE rates were 16.8% (12 of 72) versus 50.0% (19 of 38) (1).
At the 36-month follow-up (Table 2), the TLR rate remained significantly reduced (p = 0.046) in the PCB group compared with the POBA group (19.4% [14 of 72] vs. 36.8% [14 of 38], respectively). There was a highly significant difference in MACE between the 2 groups (20.8% vs. 52.6%, respectively, p = 0.001). Furthermore, our MACE-free survival analysis (Figure 1) revealed a sustained benefit for patients treated with PCB angioplasty (log-rank, p = 0.001).
Repeat TLR occurred more often in the POBA group (13.2% of patients [5 of 38] with more than 1 TLR vs. 1.4% in the PBC group [1 of 72]) during the 36 month follow-up.
There were 4 cases of possible late stent thrombosis in the POBA group (10.5%, 4 of 38) during the 3-year follow-up period. In the PCB treatment group, we recorded 1 possible late stent thrombosis (1.4%, 1 of 72) that led to myocardial infarction. The difference was borderline significant (p = 0.05).
Subgroup analyses in terms of 36-month TLR and MACE
Cardiovascular risk factors such as age and body mass index were analyzed with a univariate analysis of variance with the treatment group and various subgroups as fixed factors (Table 3). The only clinically and statistically significant difference between subgroups was the age difference between nondiabetic and diabetic patients in the POBA group (overall p = 0.005). Subgroup analyses in terms of 36-month TLR and MACE (Table 4) did not reveal any significant differences between non–PES-ISR and PES-ISR in each treatment group. In the PCB group, the 36-month TLR rates were 22.6% (12 of 51) in the non–PES-ISR and 9.5% (2 of 21) in the PES-ISR subgroups (p = 0.25). Likewise, the PCB subgroups of single-layer stents versus multilayer stents revealing 36-month TLR rates of 26.5% (9 of 34) and 13.2% (5 of 38), respectively, did not indicate a significant difference (p = 0.15). There was also no significant difference (p = 0.23) in the PCB subgroups of nondiabetic versus diabetic patients (15.2%, 7 of 46 vs. 26.9%, 7 of 26).
In contrast, in the POBA group, there was one significant difference (p = 0.03) in terms of 36-month TLR rate between the single-layer versus multilayer subgroups: 17.6% (3 of 17) versus 52.4% (11 of 21), respectively.
Subgroup analysis of focal versus diffuse DES-ISR patterns
To eliminate the factor of repeat TLR, which occurred more frequently in the POBA group between 12 and 36 months, an additional intragroup chi-square test was done to investigate the impact of the initial stenosis pattern on the 12-month TLR rates (Table 5). There were no significant differences between subgroups of focal versus diffuse patterns in either treatment group.
Referring to our 36-month results, the TLR rate in the PCB group was significantly reduced as compared with the POBA group (15.3% vs. 36.8%, p = 0.046). The MACE rate was also significantly lower in the PCB group (20.8% vs. 52.6%, p = 0.001). Our survival analysis for MACE revealed a highly significant benefit of PCB over 36 months (log-rank p = 0.001). To date, there are no other 3-year follow-up results of PCB angioplasty in DES-ISR patients available so that a comparison is not possible. Nevertheless, a catch-up effect between 12 and 36 months could not be observed, which may be interpreted as sustained clinical benefits of PCB angioplasty. Upcoming long-term data with the same PCB technology in the ISAR-DESIRE-3 (Efficacy Study of Paclitaxel-eluting Balloon, -Stent vs. Plain Angioplasty for Drug-eluting Stent Restenosis; NCT00987324) (7) and PEPCAD China (A Safety and Efficacy Study of Paclitaxel-eluting Balloon to Paclitaxel-eluting Stent; NCT01622075) (8) trials will be available in the very near future to confirm our clinical results.
This study included patients with ISR of PES and non-PES and a significant number of lesions with more than 1 stent layer to elucidate the effect of PCB angioplasty for the treatment of “real-world” ISR. Therefore, the impact on clinical outcomes in subgroups of the DES type (PES, non-PES), the number of stent layers (single, multiple), and the risk factor of diabetes seemed worthy of further exploration. Our subgroup analyses for 36-month TLR and MACE did not show any differences in the PCB group, i.e., patients with PES-ISR or multiple stent layers and diabetic patients did not have less favorable clinical outcomes. There were significantly more TLRs in the multilayer-stent subgroup when treated with POBA (p = 0.03). Given the small sample sizes available to us, the aforementioned ongoing trials with a POBA control arm (6) may also confirm these findings for DES-ISR patients treated with PCB angioplasty.
To date, it remains unclear how the pathophysiological mechanisms of DES-ISR differ from those of bare-metal stent (BMS) ISR (12,13). Based on the current literature, DES-ISR is different from bare metal stent ISR in terms of its temporal presentation, underlying mechanism, morphological patterns, tissue composition, and response to treatment. There is evidence that restenosis patterns predict the outcome (14,15). However, there is also increased awareness that even a focal DES-ISR is a more resistant lesion associated with higher MACE rates (16). It can be suspected that those ISR patterns that sufficiently predict the outcomes after PCB treatment of bare metal stent ISR cannot be applied to the DES-ISR scenario. More precisely, in the latter lesion type, the classification of lesion types and their risk of repeat revascularizations may not be sufficiently meaningful for DES-ISR. To reiterate this statement, it can be stated that in bare metal stent ISR, the angiographic pattern plays an important role for the prediction of TLR (16). Latib et al. (15) confirmed these findings for DES-ISR, but there are important differences in contrast to bare metal stent ISR, whereas restenosis in DES is more often focal (17). The study reported by Latib et al. described a 1:3 ratio of nonfocal versus a 2:3 ratio of focal DES-ISR patterns. Our subgroup analysis in terms of the DES-ISR patterns (focal vs. diffuse) was not predictive of repeat TLR and MACE in the PCB group. Kastrati et al. (18) found that technical factors may have an important role in DES-ISR that are associated with a more focal pattern of restenosis (18). To investigate this aspect, the documentation of technical or procedural factors have to be defined a priori for a proper statistical analysis. Based on our subgroup analyses, we could not detect a difference in terms of PCB efficacy in any patient subgroup with diabetes, multilayer-stent ISR, or PES-ISR. Moreover, TLR rates were not significantly different between focal and diffuse ISR after treatment with PCB.
Our data imply that the initial pattern of ISR, focal or diffuse, does not appear as important as in the case of BMS ISR. Therefore, the classification of ISR, which is mainly based on findings of BMS ISR may not be attributable to DES-ISR by implication. Other factors such as stent expansion and different tissue composition might predict outcomes of DES-ISR more accurately than the pattern of restenosis. Xu et al. (8) commented that the use of PCB in ISR offers the clinical benefit of drug delivery to the lesion site without the need for additional metal layers. This, in turn, seems highly attractive for reasons such as less mechanical irritation of a permanent implant and especially reducing dual antiplatelet therapy duration. We believe that in the affected patient population, it is beneficial to reduce the dual antiplatelet therapy duration to a minimum, provided that there are no additional bare metal stents or DES implanted during the primary intervention. In case noncoronary interventions, vascular reconstructions, or other nonvascular surgical procedures are indicated, the risk of stent thrombosis associated with PCB angioplasty without additional stenting is very low or negligible in an all-comers patient population (18). The current dual antiplatelet therapy recommendation for PCBs with Paccocath technology is 4 weeks.
The study was powered for the primary endpoint, i.e., in-lesion LLL and not for the subgroup analyses. Besides the pre-planned subgroup analyses for diabetic patients and the DES-ISR type (PES vs. non-PES), some post-hoc analyses could not be conducted (i.e., the single- vs. multiple-stent layer subpopulations). Consequently, some of the resulting subgroup patient numbers are low, in particular for the subgroup of PES-ISR. Hence, our results may be cautiously applied to clinical practice despite the fact that they are, from our point of view, hypothesis generating. The trial design was single blind and not double blind with the use of an uncoated balloon in the POBA group, resulting in different lengths of the study balloons. We did not use intravascular ultrasound to evaluate stent underexpansion or neointimal proliferation. It is important to point out that to perform meaningful multi- or univariate analyses or logistic regressions, our sample sizes were too small.
Long-term results after the treatment of DES-ISR with PCB angioplasty indicate that 3-year clinical event rates remain significantly lower compared with those associated with POBA. The presented subgroup analyses of the randomized, multicenter PEPCAD-DES trial revealed that treatment with PCB appears to be equally effective in PES-ISR and non–PES-ISR, in multilayer/single-layer DES-ISR, and in diabetic versus nondiabetic patients.
A careful projection of our results into clinical practice may also allow the statement that the pattern of DES-ISR is not predictive of repeat vascularization after PCB treatment at mid-term.
WHAT IS KNOWN? In PEPCAD-DES, PCB angioplasty was superior to POBA in terms of LLL, MACE, and TLR at 6 months for the treatment of DES-ISR.
WHAT IS NEW? This is the longest available follow-up of DES-ISR patients treated with PCBs. At 36 months, the clinical superiority (MACE, TLR) of drug-coated over uncoated balloons remained. There was no late catch-up phenomenon nor did any subgroup have significantly higher clinical event rates.
WHAT IS NEXT? A comparison with other trials in DES-ISR patients treated with PCB angioplasty or PES may confirm our findings in the very near future.
Dr. Rittger has received a research grant from B. Braun. Dr. Wöhrle has received speaker honoraria from B.Braun. Dr. Waliszewski is an employee of B. Braun. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- drug-eluting stent(s)
- drug-eluting stent in-stent restenosis
- late lumen loss
- major adverse cardiac event(s)
- paclitaxel-coated balloon catheter
- paclitaxel-eluting stent(s)
- paclitaxel-eluting stent in-stent restenosis
- plain old balloon angioplasty
- target lesion revascularization
- Received February 3, 2015.
- Revision received June 3, 2015.
- Accepted July 17, 2015.
- American College of Cardiology Foundation
- Rittger H.,
- Brachmann J.,
- Sinha A.M.,
- et al.
- Mehilli J.,
- Byrne R.A.,
- Tiroch K.,
- et al.
- Unverdorben M.,
- Vallbracht C.,
- Cremers B.,
- et al.
- Alfonso F.,
- Pérez-Vizcayno M.J.,
- Cárdenas A.,
- et al.,
- RIBS V Study Investigators, under the auspices of the Working Group on Interventional Cardiology of the Spanish Society of Cardiology
- Habara S.,
- Mitsudo K.,
- Kadota K.,
- et al.
- Xu B.,
- Gao R.,
- Wang J.,
- et al.
- Wöhrle J.,
- Birkemeyer R.,
- Markovic S.,
- et al.
- Mehran R.,
- Dangas G.,
- Abizaid A.S.,
- et al.
- Cosgrave J.,
- Melzi G.,
- Biondi-Zoccai G.G.,
- et al.
- Latib A.,
- Mussardo M.,
- Ielasi A.,
- et al.
- Douglas J.S. Jr..
- Kastrati A.,
- Dibra A.,
- Mehilli J.,
- et al.
- Wöhrle J.,
- Zadura M.,
- Möbius-Winkler S.,
- et al.