Original Contribution

Paclitaxel Drug-Coated Balloon After Bare-Metal Stent Implantation, an Alternative Treatment to Drug-Eluting Stent in High Bleeding Risk Patients (The Panelux Trial)

Jerome Roncalli, MD, PhD1;  Matthieu Godin, MD2;  Kamel Boughalem, MD3;  John Shayne, MD4; Christophe Piot, MD, PhD5;  Bruno Huret, MD6;  Loic Belle, MD7;  Guillaume Cayla, MD, PhD8; Benjamin Faurie, MD9;  Max Amor, MD10;  Bernard Karsenty, MD11;  Florence Leclercq, MD, PhD12 

Jerome Roncalli, MD, PhD1;  Matthieu Godin, MD2;  Kamel Boughalem, MD3;  John Shayne, MD4; Christophe Piot, MD, PhD5;  Bruno Huret, MD6;  Loic Belle, MD7;  Guillaume Cayla, MD, PhD8; Benjamin Faurie, MD9;  Max Amor, MD10;  Bernard Karsenty, MD11;  Florence Leclercq, MD, PhD12 

Abstract: Background. Prolonged dual-antiplatelet therapy (DAPT) in high bleeding risk (HBR) patients undergoing percutaneous coronary intervention can be challenging. We assessed the clinical safety of bare-metal stent (BMS) implantation followed by drug-coated balloon (DCB) treatment in HBR patients for whom drug-eluting stent implantation could be problematic in maintaining low ischemic event rate without increasing hemorrhagic events. Methods. The study included patients with at least 1 de novo lesion who were either under long-term anticoagulant treatment or required semi-urgent non-coronary intervention. The strategy consisted of PRO-Kinetic Energy BMS stent (Biotronik AG) implantation followed by Pantera Lux DCB (Biotronik AG) and patients were followed for up to 12 months in 37 French centers. Results. Between October 2013 and April 2015, a total of 432 patients with 623 de novo lesions who were either under long-term anticoagulant treatment (n = 300) or required semi-urgent non-cardiac surgery (n = 132) were treated by BMS plus DCB. Mean patient age was 74.1 ± 9.1 years, 76.4% were men, and 38% were diabetic. The composite primary endpoint rate (defined as target-lesion failure at 12 months) was 5.6% (95% confidence interval, 3.3-7.9). Median duration for DAPT treatment was 33 days. Hemorrhagic events, as defined by the Bleeding Academic Research Consortium, occurred in 31 patients (7.2%) and definite stent thrombosis occurred in 5 patients (1.3%). Conclusions. The combination of BMS plus DCB intervention is safe even with a short duration of DAPT. This strategy might be an alternative to DES implantation in HBR patients if future randomized trials support this approach.

J INVASIVE CARDIOL 2019;31(4):94-100.

Key words: anticoagulant, drug-coated balloon, Panelux trial, percutaneous coronary intervention


Patients at high bleeding risk (HBR) who require percutaneous coronary intervention (PCI) are a challenging group, in need of careful evaluation of thrombotic and bleeding risks when selecting a stent and determining duration and intensity of antithrombotic management.1,2 Such patients are often excluded from clinical trials of antithrombotic therapy and PCI3-6 and their management is still uncertain. Additionally, concerns exist regarding long-term safety and efficacy after drug-eluting stent (DES) implantation, especially in patients with HBR.

Consequently, there is a need to search for new devices, procedures, and strategies to improve the safety/efficacy balance in patients who undergo PCI with stent implantation. Until recently, the perceived need for a very short dual-antiplatelet therapy (DAPT) often led operators to intentionally prefer a bare-metal stent (BMS) to a DES for such patients.7,8 However, BMS options have a higher rate of restenosis compared with DES.

Paclitaxel drug-coated balloons (DCBs) do not carry a polymer coating and could therefore represent an additional biological advantage compared with DES by preventing hypersensitivity reactions. Furthermore, these devices can theoretically complete re-endothelialization in a shorter time (<3 months). DCBs apply a more homogeneous drug impregnation to the arterial wall in comparison with DES options; thus, the entire surface of the DCB is involved in delivering the drug to the arterial wall and not just the stent struts.

A combination of a BMS with the aforementioned advantages of a DCB may provide a benefit in the treatment of patients who are not suitable for receiving DAPT for >1 month (HBR and semi-urgent extracardiac surgery). Consequently, the purpose of our study was to assess the safety of the combined treatment of a BMS plus a paclitaxel DCB during PCI in HBR patients with coronary artery disease (CAD).

Methods

Study design and population. Panelux was a multicenter, prospective, non-randomized study to demonstrate the safety and efficacy of the combination of PRO-Kinetic Energy BMS plus Pantera Lux DCB (Biotronik AG) for patients who cannot be treated with DAPT for several months, as it is indicated with DES options. The study was conducted in accordance with the Declaration of Helsinki. The study protocol and the patient data release form were approved by an independent ethics committee as per local regulations. Signed informed consent was obtained from all patients included in the study.

The main inclusion criteria were: (1) patient eligible for PCI and undergoing chronic oral anticoagulant treatment (vitamin K antagonist or novel oral anticoagulant) or awaiting a semi-urgent planned surgical intervention (from 4 weeks to 3 months after PCI); (2) patient eligible for DAPT with acetylsalicylic acid and clopidogrel for at least 3 weeks; and (3) de novo coronary lesions ≥50% and <100%, target-lesion length ≤26 mm, and target-vessel reference diameters ≥2.5 mm and ≤4.0 mm by visual estimation or quantitative coronary angiography (QCA). The main exclusion criteria were: in-stent restenotic lesion; chronic total occlusion; graft intervention; ejection fraction <30%; cardiogenic shock; and patients referred for ST-segment elevation myocardial infarction.

Study devices. The surface of the Pantera Lux DCB used in this study is coated with a homogenous dose of paclitaxel 3 µg/mm2 using Butyryl tri-n-hexyl citrate (BTHC) as an excipient. BTHC incorporates paclitaxel into a microcrystalline structure to improve drug uptake into the vessel wall.9,10 It degrades to citric acid and alcohol. Paclitaxel is a lipophilic antiproliferative substance that allows rapid drug absorption by the surrounding tissue. The Pantera Lux DCB is available in lengths of 10-30 mm with diameters of 2.0-4.0 mm.11 The BMS used in this study was the PRO-Kinetic Energy stent, a cobalt-chromium stent platform with a 60 μm strut thickness and double-helix stent design.12

Endpoints. The primary endpoint was target-lesion failure (TLF) at 12 months. TLF was defined as a composite of cardiac death, any target-vessel myocardial infarction (TV-MI), urgent coronary artery bypass graft (CABG), or clinically driven target-lesion revascularization (cd-TLR). Secondary endpoints were bleeding rate according to the Bleeding Academic Research Consortium (BARC) definition, clinically driven target vessel revascularization (cd-TVR), all deaths, Q-wave and non-Q wave myocardial infarction (MI), definite incidence of stent thrombosis (ST), cd-TLR, and major adverse cardiac and cerebrovascular event (MACCE). MACCE was defined as a composite of cardiac death, stroke, Q-wave and non-Q wave MI, and TVR by non-planned angioplasty or bypass graft. All endpoints were reviewed for the following patient subgroups: patients ≥80 years old, patients with chronic kidney failure, patients under anticoagulant therapy, and patients undergoing semi-urgent surgery.

Interventional procedure. Vascular access (radial or femoral) was left to the interventionalist’s discretion. The PCI was performed as shown in Figure 1A. Predilation with a conventional balloon was optional before PRO-Kinetic Energy BMS stent implantation. Postdilation with a conventional balloon could be performed to improve stent implantation and finally allow postdilation with a DCB for 30-60 seconds. The DCB diameter was selected to achieve a 1:1 ratio of the final BMS diameter according to the manufacturer’s pressure/diameter tables. The length of the DCB had to be equal to the length of the previously chosen stent or slightly longer, taking care to avoid balloon protrusion of >2 mm from each edge of the stent. Adjuvant medical therapy was left to the discretion of the operator and followed routine clinical practice.

Statistical analysis. The sample size calculation was based on the primary endpoint of TLF at 12 months, which consists on a two-sided comparison of the TLF relative proportion with a predetermined reference value p0=10% chosen from the literature. The TLF relative proportion was assumed to be 6% based on the fact that the 1-year TLF rate will be better with the addition of DCB than with BMS alone.12 With a type I error of 0.05 and 20% drop-out rate, 500 patients are needed to provide at least 80% power.

The analysis was performed on the intention-to-treat population. Continuous data were expressed as mean ± standard deviation or median (interquartile range [IQR], Q1-Q3) and categorical data as frequencies and percentages. Hypothesis tests for categorical data were made using either the Chi-square or Fisher’s exact test. Proportions were calculated using non-missing values. The t-test was used for continuous data. All tests had a significance level of 5%, which implies that a P-value of <.05 is statistically significant. SAS statistical software version 9.3 (SAS Institute, Inc) was used for all statistical calculations. The study was sponsored by Biotronik AG. The sponsor was involved in the design of the study, data collection, and monitoring. Statistical analysis was performed by an independent organization (Medpass International). The corresponding author had full access to all data in the study and together with the co-authors had final responsibility for the decision to submit for publication. The trial was registered at ClinicalTrials.gov (NCT01930903).

Follow up and DAPT. Clinical follow-up exams were performed at 1, 6, and 12 months after PCI (Figure 1B). Patients were medically treated according to current clinical practice guidelines. DAPT was prescribed for a minimum of 1 month for patients on anticoagulation and for at least 3 weeks when the surgery was scheduled at 1 month. According to the protocol, semi-urgent surgery was programmed up to 3 months after PCI and DAPT was stopped at least 5 days before the procedure. Patients were on simple antiplatelet therapy for the intervention. Patients were then on lifelong simple anticoagulant therapy, with the choice of antiplatelet agent according to standard of care at each site.

Source data verification and adjudication. Data quality was assured by full-source document verification for all patients included during monitoring visits. All serious adverse events were adjudicated by an independent clinical events committee (Dr Berland, Dr Moulin, and Dr François).

Results

Baseline characteristics and procedural data. Between October 2013 and April 2015, a total of 501 patients were enrolled in the Panelux study. Inclusion or exclusion criteria were not met in 69 patients, resulting in a total of 432 patients analyzed. Patient characteristics are shown in Table 1. Mean age was 74.1 ± 9.1 years and 76.4% of patients were male. Diabetes was present in 38.0% of patients and 20.8% of patients had a history of coronary angioplasty. Unstable angina was present in 26.6% of patients. A total of 300 patients were on anticoagulant therapy (69.4% of the population) and semi-urgent surgery was planned for 132 patients (30.6%).

Lesions (n = 623) were assessed by visual or online QCA and had a mean reference vessel diameter of 2.9 ± 0.4 mm and mean lesion length of 14.5 ± 4.4 mm. The majority of treated lesions (42.9%) were in the left anterior descending coronary artery and 27.1% of lesions were deemed B2 or C according to American College of Cardiology/American Heart Association classification. The mean number of treated lesions/patient was 1.5 ± 0.8 (range, 1-5 lesions; IQR, 1.0-2.0 lesions).

Procedure success rate was 98.9% in the 623 lesions. The DCB did not reach the lesion in 1 patient. The strategy of BMS plus DCB as a final postdilation was undertaken in 593 lesions (95.3%). Overall, the mean DAPT duration was 51.5 days (median, 33 days; IQR, 30-44 days) (Figure 2).

Clinical follow-up. TLF at 12 months occurred in 22 patients (5.6%; 95% confidence interval [CI], 3.3-7.9). The rates of the individual components of the TLF composite endpoint are summarized in Table 2. TLR was reported in 11 patients (2.9%; 95% CI, 1.2-4.6), cardiac death in 10 patients (2.6%; 95% CI, 1.0-4.1), and TV-MI in 7 patients (1.8%; 95% CI, 0.5-3.2). Definite ST occurred in 5 patients. One ST case was the result of a misunderstanding regarding the medication (the patient stopped both clopidogrel and acetylsalicylic acid; ST occurred 15 days after DAPT discontinuation). Another patient presented with ST 15 days after clopidogrel interruption. The remaining 3 ST cases occurred under DAPT, and at 6 days and 364 days post PCI. Bleeding occurred in 31 patients (7.2%) with a majority (5.4%) classified as BARC 3 (Table 2). TLF occurred while on DAPT in 10 patients.

In the subgroup analysis (Figure 3), TLF and BARC bleeding at 1 year were significantly higher in the ≥80 years cohort than in patients <80 years (8.5% vs 3.8% [P=.03] and 10.9% vs 5.7% [P=.04], respectively) (Figures 3A and 3B). Furthermore, TLF occurred more frequently in patients with renal insufficiency (9.3% vs 4.24%; P=.046), but did not result in more bleeding events (8.0% vs 7.1%; P=.61) (Figures 3C and 3D). No significant differences were observed in the different study populations (patients under anticoagulation vs those waiting for surgery) regarding the primary endpoint or bleeding (Figures 3E and 3F).

Discussion

The main finding of the Panelux trial is that an alternative approach of BMS plus DCB had positive outcomes in a high-risk population requiring stent implantation. These findings consolidate the efforts made by interventional cardiologists to manage patients with indications for revascularization (and especially stent implantation) in accordance with their comorbidities and others diseases.

Use of DES in patients referred to semi-urgent surgery or patients with curative anticoagulation is challenging because of the potential adverse events. The Panelux trial aimed to evaluate a safe alternative for risk/benefit ratio. The combination of BMS plus DCB was recently investigated in the PEBSI trial, a randomized study with ST-elevation MI patients. The objective was to evaluate the safety and efficacy of a combined treatment with BMS followed by DCB (DCB group) vs BMS alone (BMS group).14 Results showed a significant difference in terms of late lumen loss (LLL) at 9 months and a real benefit from the DCB. The primary endpoint of in-stent LLL at 9-month follow-up angiography was a median of 0.80 mm (IQR, 0.36-1.26 mm) in the BMS group vs 0.31 mm (IQR, 0.00-0.58 mm) in the DCB group (P<.001). Clinical outcomes at 1 year were also significantly different between the two groups, with lower MACE rate (12.5% vs 3.6%), TVF rate (11.6% vs 3.6%), and TVR rate (8.9% vs 1.8%) when DCB was used in combination with BMS. The DAPT duration in PEBSI was 12 months. The main finding of the PEBSI trial supports the results of the Panelux trial, and showed that lesion impregnation with a paclitaxel-eluting balloon after BMS implantation was more effective in reducing LLL than conventional treatment with BMS only.

There is no further evidence using this approach in the setting of acute MI.14 Up to now, another procedural sequence has been used where lesion preparation was first performed, followed by impregnation of the target segment with paclitaxel released from a balloon, and finally BMS implantation.15 In the current trial, we pursued a different strategy by applying paclitaxel after stent implantation. Application of paclitaxel after successful stent implantation ensures fast, atraumatic delivery of the paclitaxel to the target segment, which in turn appears to be fundamental for the effectiveness of these devices and the spreading of the drug. Furthermore, after stent implantation, the target segment can be accurately recognized by fluoroscopy, thus reducing the risk of geographical miss and consequently reducing the chances of DCB failure. Potential advantages are homogenous drug transfer to the entire vessel wall and not only on the struts, with a rapid and sustained release of high drug concentrations with the absence of a polymer.16 The Pantera Lux balloon chosen for this trial may have contributed to the results, because not all DCBs are equal.10 They differ in the type of folding, antiproliferative drugs, coating technology, excipients, kinetics of elution, transference to the vessel wall, and durability of the drug in the coronary vessel. European guidelines on myocardial revascularization clearly state that “one cannot assume a class effect for all drug-eluting balloons.”8 BTHC is a highly lipophilic compound that allows very quick, effective transfer of paclitaxel to the vessel wall. With the highly biocompatible BTHC excipient, Pantera Lux effectively delivers a proven antiproliferative drug to the lesion site.

Regarding safety, the 1-year definite and probable ST rate in our trial was lower within this strategy (1.63%) compared with the rate of 2.0% shown in the Leaders Free trial17 using a new-generation polymer-free and carrier-free drug-coated DES (Biosensors). This stent has been developed for patients with HBR with a 1-month regimen of DAPT, and this strategy was advantageous compared with BMS in respect to the primary safety and efficacy endpoints. Our strategy investigated another possibility to reduce bleeding in a population also requiring a rapid surgery. Thus, the Panelux trial is the first study to show safety in a high-risk population of patients who usually are not included in clinical trials and are at risk of MACCE and bleedings. The challenging investigation here was to minimize the event/bleeding ratio by combining this alternative approach of BMS with DCB and the decreased duration of DAPT. Since BMS options are less commonly used because of a high risk of restenosis in such a population, cardiologists may frequently implant a DES regardless of the DAPT duration recommendation. The Panelux trial has shown some safety of an alternative strategy to DES implantation, particularly in elderly patients and patients who are suffering from renal failure. In an aged population or renal failure population, with the risk of less medical observance, such an approach may improve the outcome.

Study limitations. The Panelux study was powered for clinical outcomes and is a non-randomized study. Lack of a control group, such as BMS alone, is a limitation to understand benefits of the combined use of BMS and DCB. However, findings from the Panelux trial show benefit for a new strategy of combining BMS and DCB for a limited population of HBR patients requiring semi-urgent surgery and patients who are treated with oral anticoagulants. This strategy has not been compared with new second-generation DES options that allow shorter DAPT duration, as recommended by the European guidelines.18 Nonetheless, the strategy proposed seems feasible and safe in clinical practice. BMS implantation or DES implantation are not the only solutions, and the combination of BMS and DCB can be a possible alternative strategy for cardiologists who are taking into account the bleeding risk of their patients and/or the need for a semi-urgent surgery. However, this strategy may be dependent on the DCB used due to the heterogeneity of the different DCBs in the suppression of neointimal growth, which was recently confirmed in a clinical trial.19 Despite the promising outcomes in our study, longer follow-up is necessary to determine whether the favorable safety and efficacy profile is maintained over 1 year.

Conclusion

Coronary angioplasty with paclitaxel DCB after BMS implantation is safe and could be especially useful in a challenging population in which the management of DAPT is difficult. Patient DAPT was safely stopped at 1 month, particularly in patients referred to semi-urgent surgery. Overall, this strategy might be an alternative to DES implantation in HBR patients if future randomized trials support this approach. Acknowledgment. The authors are grateful to Marie Gabillet, Biotronik France, for editorial and administrative support.

References

  1. Urban P, Meredith IT, Abizaid A, et al. Polymer-free drug-coated coronary stents in patients at high bleeding risk. N Engl J Med. 2015;373:2038-2047.
  2. Ariotti S, Adamo M, Costa F, et al. Is bare-metal stent implantation still justifiable in high bleeding risk patients undergoing percutaneous coronary intervention? A pre-specified analysis from the ZEUS trial. JACC Cardiovasc Interv. 2016;9:426-436.
  3. Stone GW, Rizvi A, Newman W, et al. Everolimus-eluting versus paclitaxel-eluting stents in coronary artery disease. N Engl J Med. 2010;362:1663-1674.
  4. Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med. 2007;357:2001-2015.
  5. Kim BK, Hong MK, Shin DH, et al. A new strategy for discontinuation of dual antiplatelet therapy: the RESET trial (real safety and efficacy of 3-month dual antiplatelet therapy following Endeavor zotarolimus-eluting stent implantation). J Am Coll Cardiol. 2012;60:1340-1348.
  6. Feres F, Costa RA, Abizaid A, et al. Three vs twelve months of dual antiplatelet therapy after zotarolimus-eluting stents: the OPTIMIZE randomized trial. JAMA. 2013;310:2510-2522.
  7. Levine GN, Bates ER, Bittl JA, et al. 2016 ACC/AHA guideline focused update on duration of dual antiplatelet therapy in patients with coronary artery disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2016;68:1082-1115.
  8. Windecker S, Kolh P, Alfonso F, et al. 2014 ESC/EACTS guidelines on myocardial revascularization: the task force on myocardial revascularization of the European Society of Cardiology (ESC) and the European Association of 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.
  9. Joner M, Radke PW, Byrne RA, et al. Preclinical evaluation of a novel drug-eluting balloon in an animal model of in-stent stenosis. J Biomater Appl. 2013;27:717-726.
  10. Joner M, Byrne RA, Lapointe JM, et al. Comparative assessment of drug-eluting balloons in an advanced porcine model of coronary restenosis. Thromb Haemost. 2011;105:864-872.
  11. Toelg R, Merkely B, Erglis A, et al. Coronary artery treatment with paclitaxel-coated balloon using a BTHC excipient: clinical results of the international real-world DELUX registry. EuroIntervention. 2014;10:591-599.
  12. Erbel R, Eggebrecht H, Roguin A, et al. Prospective, multi-center evaluation of a silicon carbide coated cobalt chromium bare metal stent for percutaneous coronary interventions: two-year results of the ENERGY Registry. Cardiovasc Revasc Med. 2014;15:381-387.
  13. Serruys PW, Silber S, Garg S, et al. Comparison of zotarolimus-eluting and everolimus-eluting coronary stents. N Engl J Med. 2010;363:136-146.
  14. García-Touchard A, Goicolea J, Sabaté M, et al. A randomised trial of paclitaxel-eluting balloon after bare metal stent implantation vs. bare metal stent in ST-elevation myocardial infarction (the PEBSI study). EuroIntervention. 2017;12:1587-1594.
  15. Liistro F, Porto I, Angioli P, et al. Elutax paclitaxel-eluting balloon followed by bare-metal stent compared with Xience V drug-eluting stent in the treatment of de novo coronary stenosis: a randomized trial. Am Heart J. 2013;166:920-926.
  16. Speck U, Cremers B, Kelsch B, et al. Do pharmacokinetics explain persistent restenosis inhibition by a single dose of paclitaxel? Circ Cardiovasc Interv. 2012;5:392-400.
  17. Garot P, Morice MC, Tresukosol D, et al. 2-year outcomes of high bleeding risk patients after polymer-free drug-coated stents. J Am Coll Cardiol. 2017;69:162-171.
  18. Valgimigli M, Bueno H, Byrne RA, et al. 2017 ESC focused update on dual-antiplatelet therapy in coronary artery disease developed in collaboration with EACTS. Eur Heart J. 2018;39:213-260.
  19. Assadi-Schmidt A, Mohring A, Liebsch E, et al. SeQuent Please vs. Pantera Lux drug coated balloon angioplasty in real life: results from the Düsseldorf DCB registry. Int J Cardiol. 2017;231:68-72.

From the 1Department of Cardiology, Institute CARDIOMET, University Hospital of Toulouse, Toulouse, France; 2Department of Cardiology, CHU Charles Nicole, Rouen, France; 3Department of Cardiology, Clinique Alleray Labrouste, Paris, France; 4Department of Cardiology, Clinique du Diaconat, Mulhouse, France; 5Department of Cardiology, Clinique du Millénaire, Montpellier, France; 6Department of Cardiology, Hopital Privé Saint Martin, Caen, France; 7Department of Cardiology, CH Annecy-Genevois, Metz-Tessy, France; 8Department of Cardiology, CHU Carémeau, Université de Montpellier, Nimes, France; 9Department of Cardiology, Institut Mutualiste, Grenoble, France; 10Department of Cardiology, Clinique Louis Pasteur, Essey Les Nancy, France; 11Department of Cardiology, Hopital Privé Saint Martin, Pessac, France; and 12Cardiology and vascular Pole, Arnaud de Villeneuve Hospital, Montpellier, France.

Funding: This research was supported by Biotronik France, Rungis, France. ClinicalTrials.gov (NCT01930903).

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Cayla reports grants to his institution from the Fondation Coeur et Recherche and Medtronic; personal fees from Amgen, AstraZeneca, Bayer, Biotonik, Bristol Myers Squibb, Europa, MSD, Pfizer, and Sanofi; non-financial support from Boston Scientific. Dr Roncalli reports personal fees and non-financial support from Biotronik France. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript submitted October 26, 2018, provisional acceptance given November 1, 2018, final version accepted November 6, 2018.

Address for correspondence: Professor Jerome Roncalli, Department of Cardiology, Institute CARDIOMET, University Hospital of Toulouse Rangueil 1, Avenue Jean Poulhès TSA50032, 31059 Toulouse Cedex 9, France. Email: roncalli.j@chu-toulouse.fr

/sites/invasivecardiology.com/files/articles/images/94-100%20Roncalli%20JIC%20April%202019%20wm.pdf