The use of drug-coated balloons (DCB) to treat coronary artery disease has been expanding progressively in recent years. A nonnegligible rate of target vessel failure in the modern era of drug-eluting stents (DES) has created the need for different treatment strategies during percutaneous coronary intervention (PCI).1,2

When compared with DES, DCBs do not leave any metallic scaffolding in the vessel and provide fast and homogeneous transfer of antiproliferative drugs into the vessel wall upon balloon inflation. This treatment preserves coronary vasomotion, may promote positive vessel remodelling, and potentially translates into reduced late lumen loss at follow-up.3

The most widely accepted fields of application for DCB are in-stent restenosis (ISR), small vessels and combined strategies with DES during complex PCI (such as bifurcations).4–9 The available data mainly refer to paclitaxel-based platforms; it is only more recently that the introduction of new technologies for coating balloons including -limus family drugs has led to some preclinical and clinical studies using sirolimus-based platforms, thus opening the way to their use in clinical practice.

Overall, DCB are not routinely used to treat native large-vessel disease, since evidence is limited, and new-generation DES remain the gold standard following the international guidelines. Similarly, chronic total occlusions and acute coronary syndromes (ACS) still represent a gray zone for DCB use and further information from randomized clinical trials is needed.

As for the latter group, limited data are available. It has been questioned that drug delivery may be impaired in thrombus-containing lesions10; additionally, the need for mandatory lesion preparation before DCB, as well as for prolonged balloon inflation, may potentially discourage their use in acute settings where the risk of no-reflow is high. However, thrombus itself may also prevent correct vessel sizing, which paves the way to stent malapposition; in addition, the inflammatory milieu of ACS may favor delayed tissue coverage. Accordingly, both these conditions may benefit from the lack of stent struts, especially in multivessel coronary disease where multiple revascularizations are anticipated.

The all-comers EASTBOURNE Registry aims to evaluate the performance of a novel sirolimus-coated balloon in a large series of patients (NCT03085823).11 The PEACE study (Performance of a sirolimus-eluting balloon strategy in acute and chronic coronary syndromes) is a post-hoc analysis that compares DCB performance in chronic and acute settings.

Overall, 2123 patients treated with Magic Touch DCB were enrolled in the EASTBOURNE registry between September 2016 and November 2020. The clinical presentation at the index procedure was not reported in 40 patients, so that 2083 patients were finally included in the PEACE substudy. This population included 1115 (53.5%) CCS patients and 968 (46.5%) ACS patients treated with the study device. There were 364 patients with unstable angina (37.6%), 445 with NSTEMI (46%) and 159 with STEMI (16.4%, including early and late presenters belonged to the latter group) (figure 1A). A total of 79% of the lesions included in the ACS group were deemed culprit lesions by the investigator. The 12-month follow-up information was completed in 1927 patients (92.5%).

Figure 1.

Central illustration. PEACE substudy population with procedural and 12-month outcomes. A. overall population at baseline according to the clinical presentation at the index procedure. B. immediate procedural outcomes in terms of angiographic success (residual stenosis less than 50% and thrombolysis in myocardial infarction 3 coronary flow) compared between ACS and CCS patients. C. 12-month outcomes in terms of TLR and MACE compared between ACS and CCS patients. ACS, acute coronary syndrome; CCS, chronic coronary syndrome; MACE, major adverse cardiovascular event; NSTEMI, non–ST-elevation myocardial infarction; STEMI, ST-elevation myocardial infarction; TLR, target lesion revascularization; UA, unstable angina.

(0.19MB).

Baseline demographics were comparable between ACS and CCS patients, except for the prevalence of previous MI, which was higher for acute presentations (ACS 46.2% vs CCS 40.1%, P=.006) (table 1). De novo lesions represented more than half of the total study population (56.3%) and were equally distributed between the 2 groups (P=.683) (figure 2A). Procedural characteristics are reported in table 2. A DCB strategy was applied to a single lesion in most patients (89.3% vs 88.8% in ACS vs CCS patients respectively, P=.695), but up to 4 lesions were eventually treated with DCB in a few cases. As assessed by fluoroscopy, mildly and moderately calcified lesions were more common among ACS and CCS patients respectively, while severely calcified lesions were equally represented (mild 61.2% vs 9.4%; moderate 31.8% vs 84.4%; severe 7.1% vs 6.2%; P≤.001). Mean reference vessel diameter as well as mean lesion length were comparable between the 2 groups (diameter 2.56±0.68 vs 2.55±0.72 respectively, P=.881; length 18.49±9.20 vs 18.9±9.14 respectively, P=.302). Notably, predilatation was performed in a high proportion but not in all patients, despite strong recommendation per protocol, and was more commonly used in ACS than in CCS patients (93.5% vs 90.1%, P=.007). DCB were inflated at a standard pressure in both groups (mean atmospheres 9.76±3.66 vs 10.02±4.93, P=.189) with a moderately longer inflation time in acute coronary syndromes (mean duration expressed in seconds 58.74±15.84 vs 56.94±22.2, P=.036).

Figure 2.

Twelve-month outcomes of the PEACE cohort of patients (ACS and CCS) according to the type of lesion treated (de novo and ISR). A: overall population at baseline according to type of lesion treated. B: 12-month outcomes in terms of TLR and MACE according to the type of lesion treated and compared between ACS and CCS patients. ACS, acute coronary syndrome; CCS, chronic coronary syndrome; ISR, in-stent restenosis; MACE, major adverse cardiovascular event TLR, target lesion revascularization.

(0.21MB).

Angiographic success was high in both groups (ACS 97.4% vs CCS 97.7%, P=.820) (figure 1B) and stent implantation was limited to a low proportion of patients (ACS, n=68, 7% vs CCS, n=93, 8.3%; P=.592) for acute vessel recoil or flow-limiting dissection. Overall, a final dissection as detected on angiography was left in 73 patients (36 ACS, 3.7% vs 37 CCS, 3.3%; P=.707).

Table 3 and figure 1C report the overall 1-year unadjusted outcomes. At 12 months, the cumulative incidence of TLR was comparable irrespective of whether the index procedure had been performed in an acute or a chronic setting (6.6% vs 5.2% respectively, HR, 0.792; 95%CI, 0.528-1.186; P=.258) (figure 3); conversely, the cumulative incidence of MACE was higher for ACS presenters (10.4% vs 8.3%, HR, 0.656; 95%CI, 0.477-0.902; P=.009) and its occurrence seemed to diverge from the 6-month follow-up onwards (figure 4). The difference in the composite endpoint was mainly driven by a higher incidence of spontaneous MI (3.9% vs 1.2%, P<.001).

Figure 3.

Kaplan-Meier curve comparing the incidence of TLR between ACS and CCS patients over 12 months of follow-up. ACS, acute coronary syndrome; CCS, chronic coronary syndrome; TLR, target lesion revascularization.

(0.2MB).

Figure 4.

Kaplan-Meier curve comparing the incidence of MACE between ACS and CCS patients over 12 months follow-up. ACS, acute coronary syndrome; CCS, chronic coronary syndrome; MACE, major adverse cardiovascular event.

(0.21MB).

After adjustment of study outcomes for baseline and procedural characteristics, the overall study results were confirmed (TLR: HR, 0.76, P=.202; MACE: HR, 0.70, P=.036).

Subgroup analyses of the ISR and de novo lesions were also performed (patients with ACS treated for ISR 428, 47%; patients with CCS treated for ISR 482, 53%). At 12 months of follow-up, ISR treated with DCB showed a higher proportion of TLR and MACE occurrence compared with de novo lesions treated with a similar strategy (figure 5 and figure 6), independently of the type of presentation at the index procedure (for more details see table 4 and figure 2B). Similarly, it is noteworthy that recurrent spontaneous MI (a single component of the composite endpoint) was more common in ACS than CCS patients treated for ISR rather than de novo lesions (12-month cumulative rate of MI for de novo lesions: 1.2% vs 0.8% in ACS vs CCS, P=.748; 12-month cumulative rate of MI for ISR: 7.2% vs 1.8% in ACS vs CCS, P<.001).

Figure 5.

Kaplan-Meier curve comparing the incidence of TLR between ISR and de novo lesions over 12 months of follow-up. ISR, in-stent restenosis; TLR, target lesion revascularization.

(0.21MB).

Figure 6.

Kaplan-Meier curve comparing the incidence of MACE between ISR and de novo lesions over 12 months of follow-up. ACS, acute coronary syndrome; CCS, chronic coronary syndrome; MACE, major adverse cardiovascular event.

(0.2MB).

Percutaneous reperfusion strategies have dramatically improved outcomes in patients presenting with acute MI, and stent implantation has progressively overcome plain old balloon angioplasty by reducing the need for TLR at follow-up.14 Nevertheless, even new generation DES do not significantly reduce the incidence of hard endpoints such as mortality or recurrent MI during long-term follow-up, which has been explained by the possible occurrence of stent-related complications (late and very-late thrombosis).15

As an alternative, a DCB strategy combines the advantages of an antiproliferative drug with the absence of a metallic permanent scaffolding of the vessel.

The main findings of the present study can be summarized as follows: a) this is the first study investigating the performance of a sirolimus DCB in acute coronary syndromes, given that only paclitaxel-based platforms have previously been validated in this setting; b) the Magic Touch sirolimus DCB shows similar performance in terms of acute and 1-year outcomes whether used for PCI during ACS or CCS; c) similar predilatation rates and DCB inflation times between ACS and CCS suggest that longer procedural times, as expected when using DCB, do not prevent operators from applying this strategy even in the acute setting.

As already mentioned, evidence on the application of DCB to ACS patients are limited to some prospective registries and a few randomized trials, all based on the use of paclitaxel-eluting balloons.16–19The REVELATION (Revascularization with paclitaxel-coated balloon angioplasty vs drug-eluting stenting in acute MI) trial enrolled 120 STEMI patients eligible for primary PCI who were randomized 1:1 to either treatment with DCB (Pantera Lux, Biotronik, Germany) or DES (Orsiro, Biotronik; or Xience, Abbott, Abbott Park, United States). The angiographic follow-up at 9 months showed that the primary endpoint fractional flow reserve did not differ between the 2 groups (P=.27); similarly, there was no difference in the clinical secondary composite endpoint MACE (P=1.0), which also included TLR (3% for DEB vs 2% for DES).19 A similar MACE rate between the 2 groups was confirmed at 2 years of follow-up (hazard ratio, 2.86; 95%CI, 0.30-27.53; P=.34). Only 1 additional event was reported between the 9-month and 2-year follow-up. The specific patient was initially allocated to DCB angioplasty but required bailout stenting during the index procedure and TLR occurred due to ISR leading to an acute coronary syndrome at 561 days after implantation.20

The PEP-CAD NSTEMI trial enrolled 210 NSTEMI patients with de novo coronary lesions randomized to treatment with either a paclitaxel-coated DCB (SeQuent Please, B. Braun Melsungen AG, Germany) or a standard bare metal stent (BMS). DCB showed noninferiority to BMS since both the primary endpoint, target lesion failure, and the secondary endpoint, MACE, were comparable between the 2 groups (target lesion failure 3.8% vs 6.6% for DCB and BMS respectively, P=.53; MACE 6.7% vs 14.2% for DCB and BMS respectively, P=.11).18

In the prespecified subanalysis of the BASKET-SMALL 2 trial, which included lesions in vessels with diameter<3 mm, at 1-year follow-up there was no significant difference in the incidence of the primary endpoint, MACE, by randomized treatment in patients with ACS (HR, 0.50; 95%CI, 0.19-1.26 for DCB vs DES) or CCS (HR, 1.29; 95%CI, 0.67-2.47) for DCB vs DES and there was no interaction between clinical presentation and treatment effect (P for interaction .088).21

Nevertheless, a class effect for DCB is not to be expected given the huge heterogeneity in formulations, doses, release kinetics and drug-tissue interaction. Antiproliferative drugs belonging to the limus family usually have a worse transfer rate to tissues compared with paclitaxel. To overcome these limitations, several technologies have been implemented to provide sirolimus delivery, including those using a phospholipid drug carrier, on which the Magic Touch platform is based.3

The present study shows for the first time the performance of a sirolimus-eluting balloon in the setting of acute coronary syndromes. By direct comparison between ACS and CCS patients, we show that there was no significant difference in procedural and 1-year outcomes in terms of angiographic success and TLR, thus supporting its use in both settings without distinction. The higher incidence of MACE observed in ACS patients was mainly driven by the occurrence of new spontaneous MI (not procedurally related). This result could be expected, regardless of the device used, since recurrences are not uncommon after ACS (in this series ACS patients also had a higher rate of previous MI among baseline characteristics).

In addition, our study is based on an all-comers prospective registry that includes both small and large vessels, de novo and ISR lesions, single vessel, and multiple vessels disease.

Despite such a heterogenous population, the overall outcomes at 12 months of follow-up after Magic Touch use in ACS patients were good and not different from those of SCB use in CCS patients; notably, in this all-comers registry, the rates of TLR and MACE in the cohort of patients with de novo lesions (1.9% and 5.9%, respectively) are comparable to those reported for paclitaxel-coated balloons used for selected populations in the previously mentioned trials (such as the 3% TLR rate in the REVELATION trial or the 6.7% MACE rate in the PEP-CAD NSTEMI trial). In addition, current generation DES used in the acute setting yield similar results on follow-up. For example, the TLR incidence was 2.7% for a biodegradable polymer sirolimus-eluting stent in the MASTER study (which compared a DES with a BMS), while in the BIOSTEMI trial, comparing a biodegradable polymer sirolimus-eluting stent with a durable polymer everolimus-eluting stent, the rate was 2% and 3%, respectively.22,23

The apparently worse outcomes for the Magic Touch DCB used in the cohort with ISR are still in line with previous studies on paclitaxel balloons used to treat ISR, which showed a TLR rate of 10%-15% at 12 months.24,25 Notably, our results are independent of the indication for PCI (ACS vs CCS) and confirm that treatment strategies for ISR remain a challenge.

Finally, our study shows similar predilatation rates and DCB inflation time between ACS and CCS. Moreover, we observed that even the presence of thrombus (detected in 91 ACS patients and preventively addressed with aspiration in a small proportion of these, 18.7%) did not prevent operators from using a DCB strategy.

The Magic Touch sirolimus DCB shows good performance in terms of acute and 1-year outcomes whether used for PCI during ACS or CCS. In addition, despite often being time-consuming, a DCB strategy is feasible even in the acute setting where operators usually feel more comfortable with quicker and rapidly resolutive procedures.

The original EASTBOURNE study was an independent, investigator-driven study that received funding from Envision Scientific; this society played no role in the protocol definition, selection of centers, conduct of the study, or interpretation of the results. The PEACE substudy did not receive any funding.

The EASTBOURNE study received the approval of the central ethics committee of the coordinating center; data from the PEACE substudy are derived from the main registry. Each patient enrolled in the EASTBOURNE registry signed an informed consent that was archived. The sex variable is reported in accordance with the SAGER guidelines, and no impact on outcomes was expected. The STROBE form was completed.

No artificial intelligence has been used in the preparation of this work.

All authors equally contributed to the study conception and design, data collection, analysis and interpretation, and manuscript preparation.

None to disclose.