Original Contribution

Twelve-Month Outcomes With a Bioresorbable Everolimus-Eluting Scaffold: Results of the ESHC-BVS Registry at Two Australian Centers

Daniel Robaei, MBBS1,2,3;  Liam Back, MBBS1;  Sze-Yuan Ooi, MD1,3;  Mark Pitney, MBBS1,2,3;  Nigel Jepson, MBBS1,3

Daniel Robaei, MBBS1,2,3;  Liam Back, MBBS1;  Sze-Yuan Ooi, MD1,3;  Mark Pitney, MBBS1,2,3;  Nigel Jepson, MBBS1,3

Abstract: Background. The Absorb bioresorbable vascular scaffold (BVS; Abbott Vascular) is a relatively new type of coronary stent designed to provide temporary vessel scaffolding following percutaneous coronary intervention. International use of the device has grown despite a relative paucity of clinical data regarding the performance of the device and the optimal strategy for its use. We report 12-month clinical data on the Absorb BVS from a real-world registry in order to contribute to the overall understanding of the BVS device. Methods and Results. Absorb BVS implantation was attempted in 152 lesions in 100 patients at two Sydney hospitals, as part of the prospective ESHC-BVS registry. Patients selected harbored a range of complex lesions as encountered in real-world practice. Type-C lesions made up 37% of all lesions treated, with 64% of these being long lesions (>20 mm). Device success was achieved in 98.8% of cases. Predilation was performed in all scaffolds and postdilation was performed in 95% of scaffolds to a mean of 19.6 ± 4.6 atm. Twelve-month follow-up data were available for 99% of patients. At 12 months, the cumulative incidence of target-lesion revascularization was 4%, while the incidence of myocardial infarction was 2% and the incidence of scaffold thrombosis was 1%. There were no deaths in the follow-up period. Conclusion. In a cohort including complex lesions encountered in real-world practice, the Absorb BVS was associated with low rates of target-lesion revascularization, myocardial infarction, and scaffold thrombosis at 12 months when used with a strategy of meticulous lesion preparation, routine postdilation, and 12 months of dual-antiplatelet therapy. 

J INVASIVE CARDIOL 2016;28(8):316-322. Epub 2015 November 15.

Key words: bioresorbable scaffold, drug-eluting stent, percutaneous coronary intervention


Metallic stents provide scaffolding to coronary arteries in order to prevent acute vessel closure and vessel recoil following balloon angioplasty. The need for this mechanical scaffolding is temporary, as it is only advantageous in the first 6 months after balloon angioplasty.1 In the long term, the persistence of a permanent metallic stent within the coronary artery may have numerous disadvantages, including preventing positive vessel remodeling, preventing vasodilation in response to ischemia and antianginal therapy, acting as a persistent nidus for stent thrombosis and/or neoatherosclerosis, and hindering assessment of the stented segment with computed tomographic angiography.2,3

Conceptually, the Absorb bioresorbable vascular scaffold (BVS; Abbott Vascular) maintains the same advantageous drug-eluting properties of current-generation metallic stents, while addressing some of the disadvantages associated with the persistence of metallic struts and polymer coatings. 

The Absorb BVS consists of a structural framework of poly-L-lactide (PLLA) struts coated with a thin layer of poly-D, L-lactide (PDLLA) polymer containing the antiproliferative drug everolimus.4 Over time, the PLLA and PDLLA components of the BVS are hydrolyzed to lactic acid, leading to complete resorption of the device in approximately 2-4 years. The dose and release rate of everolimus integrated into the BVS is the same as in Abbott Vascular’s current-generation Xience line of metallic stents.1  

The conceptual advantages of a temporary vessel scaffold have driven increasing use of BVS internationally, despite a relative paucity of data on the performance of the device in real-world coronary disease. Regardless, excellent clinical results have been reported in non-complex disease at 12 months in the Absorb EXTEND and ABSORB II studies.5,6 Since December 2010, implantation of all Absorb BVS devices at our two institutions has occurred as part of the Eastern and Sutherland Heart Clinic BVS registry, a Human Ethics Research Committee approved, single-arm, prospective, open-label registry. As experience with the device has increased, its use has been expanded to include more complex disease. 

Methods

All patients in whom treatment with an Absorb BVS was attempted at our two institutions between December 2010 and October 2013 are included in this series. During this period, the Absorb BVS was used as a nationally non-approved device and specific written informed consent was required from all patients both for use of the device and for follow-up of clinical outcomes. The decision regarding whether to implant a BVS was made at the discretion of the interventionalist. Patient and lesion complexity for BVS implantation featured a wide spectrum that was largely reflective of real-world clinical practice.

Factors that were considered not to favor BVS implantation included in-stent restenotic lesions, extreme proximal vessel tortuosity, extreme calcification, residual stenosis >50% after lesion preparation, contraindication to prolonged dual-antiplatelet therapy, high likelihood of non-compliance with dual-antiplatelet therapy, or planned major surgery within 6 months. Patients who were participating in another trial were excluded. No limits were set on the number of treated vessels or lesions, or on the lesion length. Use of a BVS was restricted to target vessels compatible with the available scaffold sizes, specifically, 2.5 mm, 3.0 mm, or 3.5 mm diameter scaffolds in 18 mm or 28 mm length, or 3.5 mm diameter scaffolds in a shorter 12 mm length. 

Procedures. The Absorb BVS was used for revascularization of de novo disease in the setting of stable angina or acute coronary syndrome. 

All patients were pretreated with aspirin and a P2Y12 inhibitor. Procedural anticoagulation was achieved with use of unfractionated heparin or bivalirudin, with or without a glycoprotein IIb/IIIa inhibitor, as per the discretion of the interventionalist. Lesion predilation was mandated in all cases and aggressive vessel preparation was strongly advocated prior to BVS implantation, including use of non-compliant balloons, cutting balloons, and rotational atherectomy where deemed necessary. 

The manufacturer’s guidelines for scaffold deployment, which specify the use of 2 atm pressure increase every 5 seconds, were adhered to in all cases. A strategy of covering 2 mm of non-diseased vessel proximal and distal to the target lesion was recommended. Intracoronary imaging with intravascular ultrasound (IVUS) or optical coherence tomography (OCT) was performed at the operator’s discretion, but was not considered mandatory. Twelve months of dual-antiplatelet therapy with aspirin in combination with clopidogrel, prasugrel, or ticagrelor was prescribed in all patients on discharge.

Outcomes. Clinical endpoints assessed were cardiac death, scaffold thrombosis (definite/probable/possible) as defined by the Academic Research Consortium criteria,7 myocardial infarction (MI) as defined by the universal criteria,8 as well as the need for target-lesion and non-target lesion revascularization. Target-lesion revascularization (TLR) was defined as any revascularization within 5 mm of the proximal or distal ends of the scaffold. Major adverse cardiac event (MACE) was defined as a composite of death, MI, or TLR. All cases of periprocedural MI were included in the MACE rate.  

Device success was defined as successful delivery and deployment of a BVS at the intended target lesion with a final residual stenosis <50% by quantitative coronary angiography (QCA). Lesion success was defined as successful delivery and deployment of a BVS or metallic stent at the intended target lesion with a final residual stenosis <50% by quantitative coronary angiography. Procedural success was defined as device success without any MACE within 7 days of the procedure. 

Evaluation by QCA was performed offline by an expert analyst using automated edge-detection algorithms. The following QCA parameters were measured: proximal Dmax and distal Dmax (maximum lumen diameter at proximal and distal ends of the target segment treated), lesion length, minimal lumen diameter, and percent diameter stenosis.

Postprocedure troponin levels were routinely measured on day 1 by means of a high-sensitivity assay. Collection of in-hospital, 30-day, and 12-month outcome data occurred primarily through phone follow-up and completion of a patient questionnaire, as well as review of clinical notes and reporting by the treating cardiologist. Researchers independent to interventionalists performing the procedures collated the central database. When necessary, data were verified through review of coronary angiograms, correspondence, and discharge summaries. 

Results

Patient, lesion and procedural data are summarized in Tables 1-3. One hundred patients with 152 lesions were treated with a total of 167 scaffolds in this registry. 

Patient age ranged from 19-83 years (mean age, 62 years). The majority of patients treated were male (68%), while 19% were diabetic and 13% were active smokers. The indication for BVS implantation was stable angina (or angina equivalent) in 56%, unstable angina in 25%, non-ST elevation MI in 15%, and ST-elevation MI in 4%. Difficulty in obtaining adequate informed consent prior to primary angioplasty limited the use of BVS during primary percutaneous coronary angioplasty (PCI) for ST-elevation MI. 

Of the 167 scaffolds implanted between December 2010 and October 2013, 46% were implanted in the final 6 months. This increased uptake reflected increased operator experience with use of the device, and greater comfort in treating a wider spectrum of coronary lesions with a BVS. Despite this, a major factor limiting the uptake of BVS was financial constraints, with the treating institutions not receiving any reimbursement for implantation of the device.  

Patient selection to receive BVS reflected those in whom avoiding persistent metallic struts might provide greatest conceptual advantage. This included younger patients (<70 years), patients with long-segment disease (>28 mm), and those with disease involving the mid-portion of the left anterior descending coronary artery (the site of future attachment of a left internal mammary artery graft). 

The American College of Cardiology lesion classification in this series was 10% type-A lesions, 34% type-B1 lesions, 19% type-B2 lesions, and 37% type-C lesions. In the 56 lesions classified as type C, the majority (64%) were classified as type C on the basis of lesion length >20 mm, while 10 were chronic total occlusions and 4 were vein grafts. Moderate or severe calcification was present in 16% of lesions. Bifurcation lesions requiring a two-wire approach were treated on 6 occasions, while BVS-to-BVS overlap was performed in 18% of the lesions treated. A strategy of minimal scaffold overlap was used when overlapping scaffolds, due to concern regarding the higher device strut thickness. 

Predilation was performed in all cases and postdilation was performed in the vast majority (95%) of BVS implantations. The scaffolds were implanted at a mean of 13.9 ± 1.6 atm and postdilated at a mean of 19.6 ± 4.6 atm. Postdilation was performed with non-compliant balloons sized 0.25 mm and 0.5 mm larger than the scaffold diameter in 45% and 21%, respectively, while postdilation balloons were sized 1:1 with the scaffold in 33%. 

The antithrombotic administered during the procedure was weight-adjusted unfractionated heparin in 80% and bivalirudin in 20%. Intraprocedural glycoprotein IIb/IIIa inhibitors were used in 8%. QCA was used to guide scaffold selection and sizing in the majority of cases (91%). Intracoronary imaging with IVUS or OCT was employed in a minority of procedures (6.5% and 9.3%, respectively).

Procedural outcomes and cumulative clinical outcomes to 12 months of follow-up are summarized in Tables 4 and 5. BVS device success was achieved in 98.8%. The 2 cases of BVS device failure occurred in the setting of a tortuous and highly calcified right coronary artery, with 1 of these cases a CTO. On both occasions, a metallic drug-eluting stent was successfully deployed, which allowed lesion success in 100% of cases attempted. There were no cases of scaffold dislodgment. 

Thirty-day, 6-month, and 12-month follow-up data were available for 99% of patients. The 1 patient who was lost to follow-up after discharge from hospital was known to be alive at 12 months. 

Periprocedural (type-4) MI occurred in 4 cases, all of which were non-Q wave. Three of these cases occurred as a result of distal embolization following scaffold deployment leading to a minor rise in cardiac biomarkers, while 1 case resulted from compromise of a small, untreated side branch. As a result, the procedural success rate was 95.3%. There were no other cases of MI prior to hospital discharge. 

There were no deaths in the registry during the 12-month follow-up period. The overall MACE rate was 8%. This included the 4 cases of type-4 MI related to minor periprocedural rise in cardiac biomarkers. The incidence of TLR at 12 months was 4%, while the incidence of non-TLR was 2%. There was 1 case of scaffold thrombosis, resulting in an overall scaffold thrombosis rate of 1% per patient and 0.6% per scaffold. The incidence of MI at 12 months (excluding the periprocedural events) was 2%. 

Two cases of MI occurred following discharge in the 12-month follow-up period, with 1 case attributable to scaffold thrombosis (ST-elevation MI) and 1 case attributable to severe in-scaffold restenosis (non-ST elevation MI). In total, TLR occurred on 4 occasions (Table 6).

The only case of scaffold thrombosis in the registry during the 12 months was a case of definite scaffold thrombosis (as defined by ARC criteria) occurring in the setting of premature interruption to dual-antiplatelet therapy. The patient was originally treated with two overlapping 3.0 x 18 mm BVSs for long-segment disease in the proximal left anterior descending coronary artery, with repeat coronary angiography after 3 months showing the scaffolds to be widely patent. Ticagrelor was temporarily ceased at 4 months for non-cardiac surgery, at which time the patient experienced anterior ST-elevation MI. Coronary angiography confirmed scaffold thrombosis within the proximal left anterior descending and IVUS demonstrated suboptimal apposition of the proximal scaffold. Successful thrombus aspiration was achieved and balloon angioplasty was performed, followed by implantation of a 3.0 x 12 mm bare-metal stent within the proximal scaffold. Postdilation was performed with a 3.5 mm non-compliant balloon to high pressure. There were no further events during a further 18 months of follow-up. 

The other case of type-1 MI during the 12 months occurred in a patient treated with a 3.0 x 28 mm BVS to the mid right coronary artery. The patient experienced chest pain 10 months after BVS implantation and was diagnosed with a non-ST elevation MI. Coronary angiography revealed severe in-scaffold restenosis of the BVS with 95% stenosis, while OCT confirmed diffuse neointimal hyperplasia. There was no malapposition or underexpansion of the BVS demonstrated. The restenotic segment was predilated with a 3.0 mm AngioSculpt scoring balloon (AngioScore) prior to implantation of a 3.0 x 38 mm metallic zotarolimus-eluting stent across the entire length of the BVS. Postdilation was performed using a 3.5 mm non-compliant balloon to 18 atm. The patient remained asymptomatic, with no further events recorded for a further 12 months of follow-up. 

The third case requiring TLR – ie, the other case of in-scaffold restenosis – involved a patient treated with a total of 4 BVSs to the posterior descending artery and proximal, mid, and distal right coronary artery who re-presented to hospital with unstable angina 7 months after right coronary artery intervention. Coronary angiography revealed restenosis of the proximal right coronary artery BVS as well as progression of previously moderate circumflex atheroma. The remaining scaffolds in the posterior descending artery and mid/distal right coronary artery were patent. After discussion of various revascularization options, the patient underwent coronary artery bypass surgery.

The final case of TLR involved a patient who was treated with a BVS to the mid left anterior descending who experienced a recurrence of angina and was found to have a positive stress test 5 months after the intervention. Coronary angiography revealed a patent BVS, but significant focal disease at the proximal edge of the scaffold, as well as progression of disease in the circumflex. The patient underwent coronary artery bypass graft surgery. 

Non-TLR occurred in 2 cases over the 12-month period, both of which were due to residual disease remote to the target segment.  

Discussion

Bioresorbable scaffolds such as the Absorb BVS are a recent advance in the field of PCI, aiming to prevent early vessel reocclusion by reducing elastic vessel recoil and scaffolding intimal tissue flaps, while avoiding the detrimental effects of a permanent vessel cage. Potential benefits of a temporary scaffold include a reduction in the incidence of very late stent thrombosis, the risk of which is essentially eliminated following complete resorption of the bioresorbable scaffold after 2-4 years.9 A further major advantage may be late positive remodeling of the vessel and late luminal gain following resorption of the polymeric material.3,10 This, combined with drug elution to prevent neointimal hyperplasia, may reduce the incidence of late lumen loss and target-vessel failure when compared with traditional metallic drug-eluting stents. Other advantages of a bioresorbable scaffold include improved arterial vasomotor function, reduced limitations in future revascularization options for the same vessel segment, and improved ability to assess patency of the scaffolded segment via non-invasive means.2 

These conceptual advantages have driven increasing uptake of the Absorb BVS internationally despite a relative paucity of data, particularly randomized clinical data comparing the device with current-generation metallic drug-eluting stents. While the technology holds promise, the performance of the device and optimal strategies for its use are not fully known. 

Use of the Absorb BVS at our two centers has occurred exclusively as part of a prospective registry with written consent obtained from all patients. While the decision to implant a BVS was left to the discretion of the interventionalist, the absence of established superiority of the BVS over current class-leading metallic stents resulted in the device being used for cases where a temporary scaffold could confer the greatest theoretical benefit. This included younger patients, given the greater cumulative reduction in the risk of late stent thrombosis with bioresorbable versus permanent metallic stents in these individuals. Long-segment disease was preferentially treated in order to avoid the use of long-segment metallic stents, which could act as a nidus for late stent thrombosis and TLR, and which could limit future revascularization options. Mid left anterior descending lesions were seen as an indication for use of a BVS over a metallic stent in order to maintain the possibility of future revascularization by left internal mammary artery grafting.

The lesions that were treated encompassed a wide spectrum, including complex disease. Type-B2 and type-C lesions comprised 56% of all 152 lesions treated. Despite the inherently higher crossing profile and reduced deliverability of the BVS, device success was achieved in 98.8% of cases due to meticulous lesion preparation prior to BVS delivery.11 There were no cases of scaffold dislodgment despite prior observation in previous cohorts.12 

Excellent procedural and clinical outcomes were achieved in our cohort. There were no deaths in the registry over the 12 months of follow-up. MI and TLR rates at 12 months were low (2% and 4%, respectively). These figures were similar to trials of current-generation metallic stents,13,14 although the differing patient populations preclude direct comparison. 

The potentially higher thrombogenicity of the BVS has been an area of recent concern, with registry data demonstrating rates of scaffold thrombosis exceeding what would be anticipated for second-generation drug-eluting stents (up to 3.4% at 6 months).15,16 Despite these findings, the rate of scaffold thrombosis in our cohort was very low, with only 1 case during the 12-month follow-up period, equating to an annualized rate of 1% per patient and 0.6% per BVS implanted. This was similar to results achieved in ABSORB II, albeit with a significantly higher percentage of type-C lesions, particularly long lesions and lesions requiring scaffold overlap, treated in our cohort.6 The 1 case of scaffold thrombosis that occurred in our registry followed premature interruption in dual-antiplatelet therapy. 

Potential factors that may have contributed to this low rate of scaffold thrombosis include the high rates of postdilation, as well as the recommendation for 12 months of dual-antiplatelet therapy. The frequency of postdilation in our cohort was 95%, which is higher than in previous studies.6,15,17 Further investigation is required to determine the relationship of this high frequency of postdilation with the low rate of scaffold thrombosis. In our view, the higher strut thickness of the Absorb BVS and the difficulty in visualizing strut apposition angiographically warrants consideration of either a strategy of routine postdilation or assessment by intracoronary imaging in order to ensure adequate scaffold expansion and apposition. Our strategy of routine postdilation allowed successful BVS implantation with a low overall rate of IVUS and OCT use of 15.8%. This included use of these modalities during the early learning phase for BVS use, where intracoronary imaging may play an important role. 

The optimal duration of dual-antiplatelet therapy following BVS implantation remains unknown and may differ from current-generation drug-eluting stents. While clinical experience with the device is expanding and further results of trials are pending, our approach has been to recommend 12 months of dual-antiplatelet therapy in all patients treated with BVS, and to avoid BVS implantation in patients judged to be at particularly high risk of curtailed dual-antiplatelet therapy.  

The ideal strategy for treatment of cases of scaffold failure is not known, and may differ from traditional metallic stents. Two of the 4 patients in our cohort who required TLR were treated with a metallic stent within the BVS. In the absence of evidence, the use of a metallic stent within a BVS is a sound strategy, although this may pose the risk of malapposed, albeit neoendothelialized, metallic struts following resorption of the scaffold. Further long-term clinical and imaging data are required to evaluate metallic stents implanted within BVS.

Study limitations. Our findings are limited by relatively small patient numbers and the inherent nature of a prospective single-arm registry, which does not allow randomization to a control group in the form of current-generation drug-eluting stents. While lesions treated included a range of complex disease, the results cannot be generalized to other cohorts, such as routine use of BVS in ST-elevation MI patients.

Conclusion 

When used with a strategy of meticulous lesion preparation, systematic postdilation, and a recommendation for 12 months of antiplatelet therapy, the Absorb bioresorbable scaffold is associated with low rates of scaffold thrombosis, MI, and TLR at 12 months in a study population indicative of real-world coronary artery disease. 

References

1.    Serruys PW, Ong AT, Piek JJ, et al. A randomized comparison of a durable polymer everolimus-eluting stent with a bare-metal coronary stent: the SPIRIT First trial. EuroIntervention. 2005;1:58-65.

2.    Ormiston JA, Serruys PW. Bioresorbable coronary stents. Circ Cardiovasc Interv. 2009;2:255-260.

3.    Strandberg E, Zeltinger J, Schultz DG, Kaluza GL. Late positive remodeling and late lumen gain contribute to vascular restoration by a non-drug eluting bioresorbable scaffold. Circ Cardiovasc Interv. 2012;5:39-46.

4.    Ormiston JA, Serruys PW, Regar E, et al. A bioabsorbable everolimus-eluting coronary stent system for patients with single de-novo coronary artery lesions (ABSORB): a prospective open-label trial. Lancet. 2008;371:899-907.

5.    Serruys PW, Chevalier B, Dudek D, et al. A bioresorbable everolimus-eluting scaffold versus a metallic everolimus-eluting stent for ischaemic heart disease caused by de-novo native coronary artery lesions (ABSORB II): an interim 1-year analysis of clinical and procedural secondary outcomes from a randomised controlled trial. Lancet. 2015;385:43-54.

6.    Abizaid A, Ribamar Costa J Jr, Bartorelli AL, et al. The ABSORB EXTEND study: preliminary report of the twelve-month clinical outcomes in the first 512 patients enrolled. EuroIntervention.  2015;10:1396-1401. 

7.    Cutlip DE, Windecker S, Mehran R, et al. Clinical endpoints in coronary stent trials. Circulation. 2007;115:2344-2351.

8.    Thygesen K, Alpert JS, Jaffe AS, Simoons ML, Chaitman BR, White HD, Third universal definition of myocardial infarction. Eur Heart J. 2012;33:2551-2567.

9.    Serruys PW, Ormiston JA, Onuma Y, et al. A bioabsorbable everolimus-eluting coronary stent system (ABSORB): 2-year outcomes and results from multiple imaging methods. Lancet. 2009;373:897-910.

10.    Serruys PW, Onuma Y, García-García HM, et al. Dynamics of vessel wall changes following the implantation of the ABSORB everolimus-eluting bioresorbable vascular scaffold: a multi-imaging modality study at 6, 12, 24 and 36 months. EuroIntervention. 2014;9:1271-1284.

11.    Robaei D, Back LM, Ooi SYM, Pitney MR, Jepson N. Everolimus-eluting bioresorbable vascular scaffold implantation in real world and complex coronary disease: procedural and 30-day outcomes at two Australian centres. Heart Lung Circ. 2015;24:854-859. Epub 2015 Feb 23.

12.    Ishibashi Y, Onuma Y, Muramatsu T, et al. Lessons learned from acute and late scaffold failures in the ABSORB EXTEND trial. EuroIntervention. 2014;10:449-457.

13.    Von Birgelen C, Basalus MW, Tandjung K, et al. A randomized controlled trial in second-generation zotarolimus-eluting Resolute stents versus everolimus-eluting Xience V stents in real world patients: the TWENTE trial. J Am Coll Cardiol. 2012;59:1350-1361.

14.    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.

15.    Capodanno D, Gori T, Nef H, et al. Percutaneous coronary intervention with everolimus-eluting bioresorbable scaffolds in routine clinical practice: early and midterm outcomes from the European multicentre GHOST-EU registry. EuroIntervention. 2015;10:1144-1153.

16.    Ishibashi Y, Nakatani S, Onuma Y. Definite and probable bioresorbable scaffold thrombosis in stable and ACS patients. EuroIntervention. 2015;11:e1-e2. 

17.    Wohrle J, Naber C, Schmitz T, et al. Beyond the early stages: insights from the ASSURE registry on bioresorbable vascular scaffolds. EuroIntervention. 2015;11:149-156.


From the 1Eastern Heart Clinic, Prince of Wales Hospital, Sydney, Australia; 2Sutherland Heart Clinic, Sutherland Hospital, Sydney, Australia; and 3the University of New South Wales, Sydney, Australia.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Jepson reports speaker fees from Abbott Vascular. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript submitted May 11, 2015, provisional acceptance given July 17, 2015, final version accepted August 21, 2015.

Address for correspondence: Dr Nigel Jepson, Eastern Heart Clinic, Prince of Wales Hospital, Level 3 Campus Centre Building, Barker St, Randwick NSW 2031, Australia. Email: nigel.jepson@ehc.com.au

/sites/invasivecardiology.com/files/wm%20316-322%20Robaei%20JIC%20Aug%202016.pdf