Incidence of symptom-driven Coronary Angiographic procedures post-drug-eluting Balloon treatment of Coronary Artery drug-eluting stent in-stent Restenosis-does it matter?

Incidence of symptom-driven Coronary Angiographic procedures post-drug-eluting Balloon treatment of Coronary Artery drugeluting stent in-stent Restenosis-does it matter? Victor Voon1*, Dikshaini Gumani1, Calvin Craig2, Ciara Cahill1, Khalid Mustafa1, Terry Hennessy1, Samer Arnous1 and Thomas Kiernan1 1Cardiology Department, University Hospital Limerick, Dooradoyle, Limerick, Ireland 2Graduate Entry Medical School, University Hospital Limerick, Limerick, Ireland


INTRODUCTION
The clinical impact of drug-eluting balloon (DEB) treatment as a novel percutaneous coronary intervention strategy for drug-eluting stent (DES) in-stent restenosis (ISR) is a subject of ongoing interest [1]. Current knowledge of better DES safety pro iles have resulted in guidelines recommending DES by default regardless of clinical conditions and lesion subsets [2].
Nevertheless, the persistent and non-negligible problem of DES-ISR remains and little is known regarding optimal management of this entity [3]. Indeed, in recent times, the use of DEB to treat DES-ISR, has become an increasingly attractive non-

METHODS
This observational study was approved by the ethics committee of University Hospital Limerick, and conformed to the principles of the Helsinski Declaration. We retrospectively identi ied all patients treated with DEB for ISR, including DES-ISR from 2013-2015 within University Hospital Limerick database. Subsequently, baseline demographics, clinical presentations, medication, procedural and angiographic data as well as follow-up events were collected from the local database and medical records. Patterns of ISR were documented as previously described [10]. Paclitaxel-based Pantera Lux (Biotronik) DEB was used in all patients and post-DEB angiographic success was de ined as achievement of TIMI 3 low and inal residual stenosis <30%. Dual antiplatelet therapy was recommended for at least 1 year post-procedure.
The primary endpoint was incidence of SDCA performed. SDCA was de ined as coronary angiography performed on the basis of persistent symptoms reported by the patient and after clinical assessment by independent cardiologists. Symptoms were de ined as angina or angina-equivalent with/without exertion. Within this cohort, patients with and without target lesion revascularization (TLR) were identi ied. Patients presenting with acute coronary syndromes (ACS) without need for PCI were also observed within the SDCA without TLR group ( Figure 1). TLR was de ined as any repeat percutaneous or surgical coronary intervention due to ISR (diameter ≥ 70%) in the DEB-treated segment with associated symptoms and signs of ischemia. Secondary endpoints were major adverse cardiac event (MACE), de ined as all cause of death, heart failure, stroke, ACS requiring PCI and target vessel revascularization (TVR). TVR was de ined as any repeat percutaneous or surgical coronary intervention of any segment of the treated vessel.

STATISTICS
Data are presented as mean SD or frequencies (and percentages) for continuous normal or nominal/categorical variables, respectively. Analyses were carried out using SPSS version 18 statistical software (SPSS Inc, Chicago, Illinois).

RESULTS
A total of 28 patients treated with DEB were identi ied. Baseline demographics and medication pro iles are described in table 1 and 2, with 61% acute coronary syndrome clinical presentations due to ISR requiring DEB treatment. Pre-DEB treatment, all patients were on aspirin and 75% on a second anti-platelet drug. All patients had DES-ISR. Mean number of ISR per patient was 1.2±0.6 lesions ( Table 3). 69% had focal ISR lesions (Table 4). Baseline procedural pro iles are shown in table 5. Mean number of DEB per lesion was 1.2±0.6 balloons. Post-DEB, angiographic success was achieved in all patients. 33% of patients had second anti-platelet switched to an alternative antiplatelet drug.
Mean follow-up duration for all patients was 534±316 days. During this period, the primary endpoint of incidence of SDCA was 54% after mean follow-up duration of 179±241 days (Table 6). 67.8% of patients had follow-up data beyond 12 months. Within the irst year of follow-up, incidence of SDCA with and without TLR was 11% and 36%, respectively. 23% (3 out of 13) of patients requiring SDCA within the irst year, had TLR performed. Sub-analysis showed that the minority of patients had diffuse   pattern of DES-ISR in both SDCA with TLR (n=1) and SDCA without TLR (n=2) groups. The other patients with diffuse DES-ISR at baseline (n=5) had no further SDCA/TLR. In addition, 80% (8 out of 10 patients) requiring SDCA without TLR and 67% (2 out of 3 patients) requiring SDCA with TLR were already on dual anti-platelet therapy prior to DEB for DES-ISR at baseline. Interestingly, 30% (3 out of 10) of patients in SDCA without TLR group experience ACS not requiring PCI and all these occurred within the irst year of follow-up. During the total follow-up period, 3 patients had non-symptom driven coronary angiographic procedures performed but none required further PCI. No deaths, stroke, heart failure, target vessel revascularization nor ACS requiring PCI were observed during the total follow-up period.

DISCUSSION
In this single centre registry, we have for the irst time, reported on a high incidence of symptom-driven coronary angiography in patients treated with DEB for DES-ISR. Over half (54%) of the patients within 2 years and nearly half (46%) of patients within 1 year of follow-up, required SDCA. This represents real-life data compared to studies to date, which have often reported on scheduled follow-up coronary angiographic procedures regardless of presence of symptoms. Furthermore, recent observations demonstrate lack of correlation between current metrics by scheduled routine coronary angiography with TLR and/or lesion quanti ication, and the prediction of major adverse cardiovascular events [4][5][6]8]. Scheduling serial transradial coronary angiography increases the risk of radial/ulnar artery occlusion, a rare complication,  3 (11%) Mean follow-up duration to SDCA with TLR within fi rst year, days 134 ± 141 Secondary endpoint All cause of death, (%) 0 (0) Stroke, (%) 0 (0) Heart failure, (%) 0 (0) ACS requiring PCI, (%) 0 (0) TVR, (%) 0 (0) SDCA, symptom-driven coronary angiography; TLR, target lesion revascularization; ACS, acute coronary syndrome; PCI, percutaneous coronary intervention; ISR, in-stent restenosis; TVR, target vessel revascularization. Data are expressed as mean ± SD and %.
Published: June 29, 2017 40/41 but that may limit vascular access for these higher risk patients who will likely require further PCI. In our baseline cohort, 29% of patients required DEB treatment via femoral access due to poor radial/ulnar access, thus increasing peri-procedural bleeding risk. Therefore, if possible and appropriate, complementary non-invasive stress imaging should be considered to risk stratify such symptomatic patients.
However, it may be inevitable that certain patients require SDCA with TLR as demonstrated in our study at an incidence of 11% within the irst year. This igure represents nearly a third of all SDCA within the irst year and is comparable to previous studies estimating TLR incidence of 2.9% at 6 months [11] and 22.1% at 12 months [7] in patients with DEB-treated DES-ISR. This may not be surprising as a potential at-risk cohort as 61% of patients with DES-ISR at baseline presented with an acute coronary syndrome in our study. Interestingly, we only observed 10% of diabetics in our cohort compared to previous studies reporting incidence 33-42.9% [4,6,7]. Despite that, diabetes has not been shown to signi icantly in luence the impact of DEB in DES-ISR [6].
Furthermore, MACE incidence was 0% in our cohort during the total followup duration. This may be explained in part by speci ic re-classi ication of ACS not requiring PCI/TLR within the SDCA without TLR group. Indeed, no ACS requiring PCI nor TVR were observed throughout follow-up. We regard this to be a more informative de inition of follow-up events in our patient cohort. The incidence of MACE in several major studies have been reported as follows: MACE 7.7% (including myocardial infarction (MI) 5.1%, TLR 2.6%) at 9 months in the GARO registry [4]; MACE 16.8% (including 0% MI, 15.3% TLR) in PEPCAD-DES study [6]; and MACE 23.5% (which included MI 2.8%, TLR 22.1%) at 12 months in the ISAR-DESIRE 3 trial [7]. However, the generalized description of such outcomes may result in lack of useful inferences relating to ISR. Thus, SDCA may itself be a more clinically meaningful marker of a higher risk cohort but this requires clari ication in larger studies.
Indeed, patients with SDCA may represent a subgroup lacking optimization of medical therapy, a subgroup at risk of coronary artery spasm [12], a subgroup with microvascular disease that may be at higher risk of requiring repeat TLR and/or a subgroup of patients with a lower threshold for ischemia. Post-DEB for DES-ISR, patients requiring SDCA without TLR may be at higher risk of coronary disease progression. Symptoms may be a re lection of ongoing microvascular disease downstream which is stimulated by vasoactive and proin lammatory markers released by DES-ISR site beyond the effects of homogenous local anti-proliferative drug delivery by DEB. Such changes may in turn contribute to the propagation of neointimal negative remodeling, thus accelerating coronary disease. This may be explained by our data demonstrating that 30% of patients with SDCA without TLR experienced acute coronary syndrome during the irst year of follow-up.
Finally, diffuse-type ISR was recently suggested to cause early DES-ISR and several studies have reported on incidence over 40% of such lesions in DES-ISR [1,4,6]. However, the impact of such lesions after DEB for DES-ISR remains unknown. Post-DEB, diffuse-type DES-ISR at baseline did not appear to predict TLR as demonstrated in our study. In fact, the majority of patients having such angiographic lesions did not have further SDCA nor MACE during the follow-up period. Therefore, this supports the potential of SDCA being more useful clinical barometer in larger studies to quantify and predict risk of disease progression in DEB-treated DES-ISR.

LIMITATIONS
We acknowledge several limitations to this study. First, this was a retrospective, observational, single-centre study with small sample size, which lacked a comparator group. However, as aforementioned, it also represents real-life data that can be extrapolated into daily clinical practice. Therefore, despite being hypothesis generating, such results should be cautiously applied. Second, we were unable to obtain data of several variables such as time course between DES implantation and ISR, generationtype of DES used, previous DES size, mechanical factors (eg. stent under-expansion), use of intracoronary imaging post-DEB, and non-invasive stress imaging prior to SCDA. These parameters would be informative in future observational studies performed. Third, we acknowledge the possibility of coronary vasospasm is a contributor towards symptoms reported and lack provocation testing in this regard.

CONCLUSION
A high incidence of symptom-driven coronary angiography was observed, particularly within the irst 12 months after DEB-treated DES-ISR. This under-reported metric may represent a cohort at higher cardiovascular risk but requires further con irmation in larger studies.