Can you score with balloons to enhance outcomes after drug coated balloon angioplasty? Insights from the Paris DCB Registry for in-stent restenosis

Objectives: The objective of this study was to assess the 12‐month clinical outcomes in patients with drug‐eluting stent in‐stent restenosis (DES‐ISR) who were either pre‐dilated with non‐compliant balloons (NCBA) and with additional scoring balloons (NCBA + SBA) prior to drug coated balloon (DCB) angioplasty. Methods: This monocentric, retrospective study included patients with DES‐ISR who were routinely treated over a 2‐year time span. Patients with stable angina and documented ischemia or selected forms of unstable angina due to a culprit DES‐ISR lesion were analyzed. The primary endpoint was the clinically driven target‐lesion revascularization (TLR) rate at 12 months. Secondary endpoints included post‐interventional lumen gain and late lumen loss (LLL) at 6 months. Results: The 12‐month TLR rates in 124 patients who underwent either NCBA + SBA or NCBA only group were not different (17.3%, 9/52 vs 11.6%, 8/69, P = 0.371) and low as compared to other comparable studies. The use of SBA led to equally high post minimal lumen diameters (MLD) in both treatment arms (NCBA 2.21 ± 0.33 vs NCBA + SBA 2.18 ± 0.41, P = 0.868). We did not find a significant difference in late lumen loss (LLL) between both groups (0.50 ± 0.62 mm vs 0.40 ± 0.46 mm, P = 0.468). Conclusions: Scoring Balloon Angioplasty can safely and effectively prepare DES‐ISR lesions to render them suitable for DCB angioplasty with acceptable TLR and MACE rates.

drug coated balloons; in‐stent restenosis; scoring balloon

There has been consensus on the benefit of drug coated balloon (DCB) angioplasty in patients with bare metal in‐stent restenosis (BMS‐ISR) resulting in a class IA recommendation.[1] Despite evidence in drug eluting stent in‐stent restenosis[2] (DES‐ISR), there were also reports with less favorable outcomes after DCB angioplasty as compared to the implantation of second generation drug eluting stents (DES) in DES‐ISR.[3] This initiated debates of the proper lesion preparation for DES‐ISR[4] which paid tribute to the acute lumen gain and the learning curve for this interventional strategy.[5]

The purpose of this study was to investigate the usefulness of cutting or scoring balloon angioplasty to accomplish sufficient lumen gain prior to DCB angioplasty in patients with DES‐ISR. Since the use scoring balloons is only indicated in lesions with insufficient lumen gain, a proper randomization with sufficient patient numbers for clinical endpoint testing was not within the realm of this observational study.

Initiated by the upcoming reimbursement for DCB angioplasty as the first non‐implant medical device[6] within the French health care system (Haute Autorité de Santé) and the availability of only one DCB study in an entirely French population,[5] our center conducted an retrospective analysis within the framework of routine care.

The primary endpoint was the rate of target lesion revascularization (TLR) at 12 months. Major adverse cardiac events (MACE) were defined as the composite of cardiac death, myocardial infarction, TLR or a target vessel total occlusion. The composite of TLR consisted of the rates for of percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG). Target vessel revascularization (TVR) rates were also recorded at 6 and 12 months. Secondary endpoints included post‐interventional lumen gain and late lumen loss (LLL) at 6 months.

Suitable patients had to be older than 18 years with symptoms of myocardial ischemia (MI) symptoms or evidence of MI with the presence of a DES‐ISR (diameter stenosis ≥50%). Furthermore, DES‐ISR patients with ST elevation infarction (STEMI) or non‐ST elevation infarction (NSTEMI), thrombus burden, documented history of neoplasm and chronic renal insufficiency were also included. It was the intention to pre‐dilate all lesions with non‐compliant balloon catheters (NCB). In case the initial lumen gain was not sufficient (>30% remaining stenosis) according to the established recommendations,[7] additional scoring balloon angioplasty (SCB) was done. Only one scoring balloon type was used in this assessment (Angiosculpt, Biotronik AG).

Patients with de novo lesions and/or patients with confirmed bare metal stent in‐stent restenosis (BMS‐ISR) were excluded from enrolling in this study. The presence of cardiogenic shock was also an exclusion criterion.

In the majority of the cases a paclitaxel‐iopromide coated DCB (SeQuent® Please, B.Braun Melsungen AG) was used and supplemented with cases that utilized a paclitaxel‐butyryl‐tri‐hexyl citrate coated DCB (Pantera® Lux, Biotronik AG).

Whenever possible the radial approach was preferred. All recommendations[7] relative to DCB inflation times and overall handling were observed. At the discretion of the investigator intracoronary imaging was conducted if the gain of significant knowledge relative to the lesion morphology justified the additional effort and cost. Analogously, additional intracoronary imaging was done whenever the benefit/cost issue was permissible.

Patients received 500 mg of acetylsalicylic acid before the intervention in case they were not already on this long‐term treatment. A clopidogrel loading dose of 300‐600 mg was administered. Upon insertion of the sheath, heparin 50‐100 UI/kg was administered. Post‐interventionally, either clopidogrel (75 mg/d), ticagrelor (2 × 90 mg/d) or prasugrel (10 mg/d) were recommended for 3 months in stable patients and 12 months in patients with acute coronary syndrome (ACS) following the intervention. Acetylsalicylic acid 75‐160 mg/d was prescribed life long.

With the onset of the class IA recommendation of the European Society of Cardiology (ESC) recommendation, our institution commenced a dedicated in‐house data base for all patients that underwent DCB angioplasty. The first tranche of patients had lesions that were pre‐dilated with “NCBA only” whereas the 2nd part of our database consisted of patients who were treated with NCBA and additional scoring balloon angioplasty (NCBA + SBA). Under the scrutiny of our institution's ethics board, data were collected to protect the patients’ identity.

Within the framework of routine care and due to the high risk assessment of our DES‐ISR patients, control angiograms were scheduled at 6 months and analyzed in‐house by a dedicated staff member. If a patient reached any endpoint prematurely, that is, before 6 or 12 months, the follow‐up was considered complete at later time points.

Since our analyses are descriptive in nature, a sample size analysis was not done. Categorical variables were analyzed using Pearson's Chi[2] test or Fisheŕs exact test, when appropriate. All continuous variables are described with the mean values ± standard deviation or as median with range whenever applicable. Differences between proportions and t‐tests were either analyzed with the independent sample t‐test or with the Mann‐Whitney U test if deviations from the Gaussian distributions were detected with the Shapiro‐Wilk test. SPSS version 24.0 (IBM, Munich, Germany) was used for all analyses.

In total we treated 124 patients with the cardiovascular characteristics as detailed in Table . The rates of male gender were different between treatment arms (NCBA 91.7% vs NCBA + SBA 100.0%, P = 0.033) whereas there were no differences in the rate of diabetes between the two procedural groups (NCBA 48.6% vs NCBA + SBA 61.5%, P = 0.154).

Baseline data

1 NSTEMI, Non ST elevation myocardial infarction; STEMI, ST‐elevation myocardial infarction; MLD, Minimal lumen diameter; PES, paclitaxel‐eluting stent, SES, sirolimus eluting stent; EES, everolimus eluting stent; BES, biolimus eluting stent; ZES, zotarolimus eluting stent; BMS, bare metal stent; DES, drug eluting stent; RCA, right coronary artery; LCX, left circumflex coronary artery; LAD, left anterior descending artery.

2 aAll continuous variables are expressed as mean ± standard deviation and analyzed with the independent sample t‐test

A typical patient suitable for NCBA and SBA is shown in Figure . This patient had a focal DES‐ISR of overlapping DES that was pre‐dilated with NCBA and then treated with SBA before DCB angioplasty was done.

In terms of the treated lesion morphologies (Table ), the incidence of multiple stent layers were equally balanced in both treatment groups (20.8% vs 15.4%, P = 0.441). Moreover, the rates of at least one DES layer were also not different between groups (90.3% vs 94.2%, P = 0.425).The pre‐procedural minimal lumen diameter (MLD) was significantly lower in patients who were treated with scoring balloons (0.87 ± 0.26 mm vs 1.11 ± 0.24 mm, P = 0.011). The use of SBA led to equally high post MLD in both treatment arms (NCBA 2.21 ± 0.33 vs SBA 2.18 ± 0.41, P = 0.868). Bailout stenting was equally low in both groups (6.9% vs 5.8%, P =  0.793).

At 6 months the TLR rate in the NCBA was numerically lower without reaching statistical significance (NCBA 4.3% vs NCBA + SBA 11.5%, P = 0.136). The target lesion LLL were not significant different either (0.50 ± 0.62 mm vs 0.40 ± 0.46 mm (P = 0.468). Our 12‐month follow‐up rates were 95.8% in the NCBA only group and 100% in the NCBA + SBA group (Table ). Our primary endpoint, the 12‐month TLR rates in the NCBA only and the NCBA +SBA groups were not different (11.6%, 8/69 vs 17.3%, 9/52, P = 0.371). The 12‐month MACE rate in the NCBA group was numerically lower as compared to the NCBA + SBA group without reaching significance (13.0%, 9/69 vs 19.2%, 10/52, P = 0.354).

Six‐month angiographic and clinical outcomes at 6 and 12 months

• 3 TLR, target lesion revascularization; TVR, target vessel revascularization; MACE, major adverse cardiac events.
• 4 aIf an endpoint was reached prematurely, the follow‐up was analyzed “as completed.”
• 5 bAnalyzed with the Mann‐Whitney U test.
• 6 cValues denote median (minimum, maximum).

In our routine follow‐up, we did not find a difference in late lumen loss (LLL) between both groups, that is, in the SBA group the LLL was 0.50 ± 0.62 mm while the NSBA group had a LLL of 0.40 ± 0.46 mm (P = 0.468).

As summarized by Auffret et al,[5] post‐procedural diameter stenosis (DS) were in the range of 10.5% in the PEPCAD China trial[8] to 15.6% in the GARO registry[5] and were as high as 30.0% of the RIBS IV study.[3] In our series we found DS in the 21.0‐22.0% range which led to a late lumen loss (LLL) less than the values reported by Rittger et al.[9] [10]

The adequate lesion preparation was discussed by Scheller et al[4] and Waksman.[11] While both sides of this discussion agree on the necessity of initial lumen gain, the treatment strategies seem to fundamentally differ. While Waksman stresses the need to provide radial support properties to eliminate dissections and acute recoil by implanting some sort of scaffold, Scheller et al[4] emphasize the clinical experience to judge whether a dissection is benign so it can be left without metal caging. Obviously, there are two worlds that clash, however, do we really need an additional stent within a DES‐ISR?

Given our single center experience, we sense that additional stenting in DES‐ISR can be avoided in the majority of our patients if the lesion preparation is done with care. In addition, the presented evidence is suggesting that the lesion preparation in terms of lumen gain strongly promotes acceptably low LLL and favorable 6‐month TLR rates in the 5‐12% range even in multiple layer DES‐ISR. We can only agree with the necessity for proper lesion preparation which is certainly part of the learning curve. It is our conviction that albeit implanting an additional layer of stents in a DES‐ISR is more “convenient” and effective from a time and certainly from the French reimbursement perspective, the potentially longer lesion preparation in case of DCB angioplasty will bring long‐term benefits for the patient, that is, a lower clinical event rate over years following the intervention with SBA and DCB.

Within the descriptive framework of our routine care, our data suggest that SBA can safely and effectively prepare DES‐ISR lesions to render them suitable for drug‐coated balloon angioplasty resulting in acceptable rates of TLR and MACE at 12 months. Larger studies should confirm these early findings in patients in whom randomization is ethically acceptable. For future studies, one can expect a TLR rate of 12‐17% at 12 months and a lumen gain with scoring balloon angioplasty and non‐compliant balloons of approximately 1.3 mm in case future study designs involve an angiographic study endpoint.

This is an observational single center study with the intention to investigate the impact of lesion preparation independent of the used DCB technology to revascularize DES‐ISR lesions. Furthermore, this study was not powered for a clinical endpoint with supportive angiographic data which was obtained in the framework of routine care. Furthermore, it would have been desirable to have a higher angiographic follow‐up rate but due to our clinical routine, we could not conduct routine control angiography.

Within the nature of a monocentric descriptive study, our findings represent the learning curve of our dedicated staff and may not be easily transferable to other centers. Furthermore, it would have been also useful to know all DES involved in our assessment since the percentage of unknown DES type was significant (see Table , lesion characteristics).

Moreover, it would have been preferable that an independent core lab conducted the quantitative coronary angioplasty. On a positive note our clinical follow‐up rates were high (>95%) which is, however, not uncommon for monocentric studies.

PHOTO (COLOR): Drug eluting stent in‐stent restenosis in an 82 years old ACS patient