Since the first percutaneous transluminal coronary angioplasty in 1977, performed by Grüntzig [19], the materials and methods used have evolved from bare-balloon angioplasty to coronary stents, drug-eluting stents and balloons, DCB, and bioresorbable scaffolds [20, 21]. The principle of PTA was rapidly introduced in many vascular areas, in which surgical stenosis treatment was typically used, because it had already been demonstrated to be beneficial in the supraaortic, neurovascular area at the time when PTA was introduced [22, 23].

In 1983, PTA of the extracranial internal carotid artery (ICA) was first described in the literature [24], and 7 years later, suspected cerebral embolization through plaque damaging was addressed to in a groundbreaking way by the first cerebral protection device in a triaxial setup [25]. Despite these developments, major trials, such as SPACE (2004) and Crest (2010), did not demonstrate the superiority of the endovascular method to open surgery [2, 26].

However, it’s a meanwhile frequently performed procedure that’s not only carried out in surgically non-accessible anatomy but is regarded equally to surgical treatment in regard of different risk factors and conditions of both methods. But both methods stimulate neointimal proliferation, a mechanism that is believed to be primarily responsible for restenosis and instant restenosis in the neurovascular setting [27] increasing the risk of new ischemic events and death after successful treatment.

In intracranial arteries, owing to the high risk of complications, endovascular treatment of intracranial atherosclerotic stenosis (ICAS) has rarely been reported after an initial two-case report on PTA of high-grade basilary artery stenosis in 1980 [28]. Stents for intracranial use were first described in emergency or ultima ratio settings of SAH [29, 30] and surgically inaccessible major aneurysms rather than ICAS treatment. Stent-assisted treatment of ICAS was later incorporated into neurovascular interventions [31, 32, 33, 34, 35]. However, concerns regarding the mid- to long-term benefits of endovascular treatment of ICAS arose early [36]. Ultimately, no randomized multicenter trials to date have demonstrated an advantage of stent assisted ICAS treatment over the best medical treatment (SAMMPRIS [13], VISSIT [14]).

A major reason for the devastatingly poor outcomes of stent-treated ICAS is the high incidence of in-stent restenosis, which is again probably caused by neointimal proliferation [37, 38, 39].

The issue of neointimal proliferation and its negative effects on long-term outcomes have also been observed in the field of cardiology and were addressed in the early 2000s already after a retarding effect on neointimal proliferation after short-term exposure to taxane in a porcine model was described [40, 41]. Consequently, the development of DCB began.

DCB are now a part of cardiovascular guidelines [18, 42] and international consensus statements [43, 44], with the caveat that a class effect of DCB should not be presumed, because the results varied among different types of DCB [42, 45].

In the neurovascular setting, the problem of neointimal proliferation leading to in-stent restenosis and restenosis after angioplasty was initially countered by DES, too. However, the effects of DCB observed in non-neurovascular fields led to the introduction of DCB in the neurovascular setting in due course [44].

The primary use of DCB was the prevention and treatment of in-stent restenosis and restenosis of extracranial carotid artery arteriosclerotic stenosis (ECAS) and ICAS, and reported results indicated advantages over bare-balloon angioplasty as well as re-angioplasty, resembling findings in the field of cardiology. This method was pioneered in 2011 and 2012 by Vajda et al. [46, 47] with promising results and was only revisited in 2018 [48, 49, 50, 51, 52, 53, 54, 55, 56]. However, despite good results and reviews continuing to advocate for the evaluation of DCB in the treatment and prevention of ECAS and ICAS in in-stent restenosis [57, 58, 59] and restenosis, data addressing the use of DCB and the evaluation of its possible benefits in the neurovascular setting remain rare; consequently, we believe that this field is under-investigated, and publications are lacking.

Table 1 (Ref. [60]) displays commercially available DCB with a CE mark, intended for intracranial use or usable off-label in the neurovascular setting.

After testing of several drugs for the prevention of neointimal proliferation in stents and balloons [61, 62, 63, 64, 65], paclitaxel followed by sirolimus have become established in DCB and are currently the main drugs used.

Paclitaxel is an antimicrotubular drug that stabilizes microtubule polymerization during mitosis, thus inducing cell death [66, 67]. It effectively inhibits coronary neointimal proliferation in vitro and in vivo [68]. Paclitaxel is highly lipophilic and therefore virtually insoluble in water [69], thus decreasing drug loss during application and drug downstream—i.e., the effect of leaching of paclitaxel distal to the treated stenosis. However, the amount of paclitaxel effectively reaching the vessel wall during DCB angioplasty, is estimated to be only 10% (5–20%):

- 5–30% of loss en route to the destination.

- 40–70% loss during or directly after balloon inflation and deflation through washout.

- 0–30% of the drug may remain unapplied and may be recovered with the balloon after the intervention [70].

This proportion of lost paclitaxel led to suspicions of tissue harm and consequently to investigations of downstream effects, which culminated in a highly regarded, yet methodically and statistically limited meta-analysis connecting the application of paclitaxel-eluting devices to higher mortality in selected studies [71]. The results of this meta-analysis are controversial. Soon after, several contradictory meta-analyses and trials concerning this issue were published [72, 73, 74, 75, 76, 77]. Studies investigating the direct effects of drugs downstream from paclitaxel-coated devices nevertheless did not find relevant harm to (coronary) tissue distal to the point of application in the vessel [70, 78, 79, 80, 81]. Today, the effect of paclitaxel from paclitaxel-containing implants on tissue away from the site of application is considered irrelevant and is no longer directly related to increased mortality [82, 83].

Sirolimus is another name of rapamycin, which is derived from Streptomyces hygroscopius. Its immunosuppressive and antiproliferative effects stem from blocking the cell-cycle specific kinase Target of Rapamycin (TOR) and subsequently halting cell mitosis and inducing cytostasis [84, 85]. The antiproliferative effect of sirolimus discovered in early work [86] has been demonstrated to be highly effective in the treatment of coronary neointimal proliferation and in-stent restenosis and restenosis through DES [87, 88].

Further developments in coating led to the development of sirolimus-coated balloons (SCB), which can be applied in the same manner as paclitaxel-coated balloons (PCB) with sufficient sirolimus transfer to the vessel wall [89]. The advantages of sirolimus over paclitaxel are mainly seen in sirolimus having a broader therapeutic range, larger safety margin, greater success in restenosis treatment than other anti-restenosis drugs, and that it is cytostatic instead of cytotoxic [90, 91, 92].

Several trials have demonstrated that SCB, which emerged in the mid-2010s, are effective in the treatment of coronary neointimal proliferation [93, 94, 95]. The range of commercially available SCB remains narrow, and randomized trials are scant. Currently, no published data are available on the use of SCB in ECAS or ICAS. However, because SCBs are available in the cardiovascular setting, they may one day be used in the neurovascular setting as well.

In the current state of the art of DCB, the coatings contain a drug (e.g., paclitaxel) and a transfer enhancing excipient or spacer, because bare drug-coated devices achieve less transfer of the anti-proliferative drug [70, 96, 97]. The drug on the balloon is dosed in $\mu$g/mm${}^2$ balloon surface, and the amount of drug transferred to the vessel wall tissue is directly correlated with the inhibition of neointimal proliferation; however, the magnitude of drug transfer varies significantly [44].

The most common excipients in DCB are derived from modified plant and bacterial substances (e.g., n-butyryl tri-n-hexyl citrate, dextran, nordihydroguaiaretic acid, resveratrol, shellac, and urea), organic plasticizers (e.g., acetyltributylcitrate), magnesium stearate, and the iodine-containing contrast agent iopromide. Urea- and magnesium stearate-containing coatings have been found to strongly inhibit neointimal proliferation in a porcine model [98] and in patients. However, estimating the applied dose in vivo (i.e., in atherosclerotic human vessels) remains a challenge. The optimal technique and excipient may be yet unidentified, given that new coating techniques and excipients continue to be developed, and existing ones are being improved.

3.2.1.1 Internal Carotid Artery

Although in vivo experiments with temporal occlusion of the ICA have found that as much as 15% of mean cerebral perfusion is supplied by the ipsilateral external carotid artery [101], to date, no relevant advantage of sole external carotid artery-stenosis treatment—both endovascular and surgical—has been reported [102], apart from case reports focused on rare symptoms and constellations of external carotid artery stenosis. Treatment of the extracranial carotid artery is mainly focused on treatment of the proximal ICA; therefore, this section focuses on ECAS treatment of the ICA.

Use of PTA vs. TEA for the treatment of ECAS has been investigated intensively, yet the clear superiority of either remains unproven. Although TEA appears to raise the risk of a myocardial infarction, the risk of ipsilateral stroke is higher in PTA [103]. However, long-term outcomes may be similar [104]. The advantages and particularly the disadvantages of both treatment types are subject to an ongoing discussion and must be considered for each patient individually.

The use of DCB as an index treatment of external carotid artery stenosis is not common. To our knowledge, no trial or major case study has investigated the effects of using DCB in the PTA treatment of de novo ECAS.

The primary use of DCB is therefore the treatment of ECAS restenosis after TEA and PTA, which is a common issue: in the 2018 ICSS trial, the 5-year cumulative risk of restenosis after TEA and in-stent restenosis after PTA has been reported to be approximately 30% for TEA and 40% for PTA, thus raising the overall risk of ipsilateral stroke in the studied population of 1530 individuals [105].

Challenging ECAS in-stent restenosis (ISRS), the use of DCB was first reported in 2012 in two case series (IN.PACT™ Amphirion, Medtronic, Dublin, Ireland [106] and IN.PACT™ Admiral, Medtronic-Invatec, Dublin, Ireland [48]) describing the application of paclitaxel DCB in an ISRS after standard balloon predilation. At a median follow-up of 15 months, a total of ten treated patients showed not only arrested neointimal hyperplasia in the dilated vessel lumen but also a subsequent decrease in flow peaks on Doppler sonography.

This phenomenon was also reported in a case series of nine patients in 2014 with ISRS refractory to several preceding endovascular maneuvers (DIOR II; Eurocor) [107].

Piccoli et al. [108] have published a case series of 18 patients in 2015 with high-grade stenosis after TEA, who were treated with DCB before stent-PTA of the carotid artery. At a mean follow-up of 18 months, no restenosis over 50% of the vessel lumen occurred (IN.PACT™ Pacific balloon, Medtronic Vascular, Dublin, Ireland; Lutonix Balloon, Becton Dickinson, Franklin Lakes, USA).

In a retrospective single-center study, Pohlmann et al. [49] (SeQuent® Please, Braun Melsungen, Vascular Systems, Melsungen, Germany; IN.PACT™ Amphirion balloon, Medtronic, Dublin, Ireland) studied nine patients treated with DCB for ISRS (11 procedures in total due to bilateral ISRS in one patient and repeated DCB treatment due to restenosis after 19 months in another patient). In that study, the event-free rate was 100% after the first year and 83% after 5 years of follow-up. No periprocedural strokes or deaths were observed.

The first direct comparison of DCB angioplasty vs. bare balloon angioplasty in ISRS treatment was a single-center prospective registry evaluation, in a study published in 2019 by Tekieli et al. [109]. The DCB-group (n = 27) showed a second ISRS after DCB treatment earlier (mean 23 months) than the bare-balloon group (n = 19; mean 26 months). However, the rate of second ISRS was smaller in the DCB-treated group (23%) than in the group with bare balloon treatment (32%). The rate of periprocedural stroke or death was reported to be 2% for the bare-balloon (n = 19), DCB (n = 27) and stent-supported (n = 6) arms and is considered a DCB complication in this review.

Another retrospective single-center study comparing different techniques of endovascular treatment vs. open surgery for ICA restenosis after TEA has not shown significant differences in a 1-year follow-up in terms of overall survival, survival without recurrent restenosis, and survival without reintervention for DCB (n = 18), bare balloon (n = 7), and CAS (n = 9) vs. open surgery (Lutonix, Becton Dickinson) [110]. Moreover, no significant differences among the three endovascular techniques were observed; however, because of the small number of cases, this result has limited validity.

Mihály et al. [52] published a comparison of ISRS treatment and re-ISRS treatments in a group of 46 patients with CAS (n = 7) vs. bare balloon (n = 37) vs. DCB (IN.PACT™ Falcon; Medtronic, Dublin, Ireland; Ranger™, Boston Scientific Corp., Marlborough, USA) angioplasty (n = 2) and found no restenosis in the DCB group in a mean follow-up of 30 months. The authors have concluded that DCB is superior to CAS and bare balloon angioplasty in the prevention of ISRS; however, owing to the small number of cases, this result has limited validity.

Table 2 (Ref. [48, 49, 52, 106, 108, 109, 110]) shows the study data in detail.

Table 2.Studies addressing the use of DCB for restenosis treatment of the extracranial carotid artery.

Author, year	DCB treatment N	Target	Follow-up	Reported outcome	Periprocedural stroke/death
Liistro et al. [106], 2012	3	In-stent restenosis	36 months for symptoms, 24 months for stent patency	100% asymptomatic, 100% stent patency	-
Montorsi et al. [48], 2012	7	In-stent restenosis	21 $\pm$ 19 months, mean 12 months	Significant decrease in in-stent peak systolic velocity	-
Piccoli et al. [108], 2015	18	Stenosis after TEA	13–22 months, mean 18 months	Stent patency $>$50% in 100% of cases	-
Pohlmann et al. [49], 2017	9, 11 procedures	In-stent restenosis	5 years	Event-free rate 83%	-
Tekieli et al. [109], 2019	27	In-stent restenosis	26 months	Earlier 2nd ISRS after DCB treatment than bare balloon, but at a lower rate in total	2% in total (DCB and bare balloon PTA)
Haupert et al. [110], 2020	18	Stenosis after TEA	12 months	Comparable results for DCB vs. CAS vs. bare balloon PTA vs. open surgery/14% restenosis	3% in total (DCB, CAS, and bare balloon PTA)
Mihály et al. [52], 2021	2	In-stent restenosis	9–53 months, mean 30 months	No restenosis in DCB-treated patients	Not reported

3.2.1.2 Vertebral Artery

Vertebral artery atherosclerotic stenosis (VAOS) may account for up to 25% of cerebral ischemia cases [111, 112] causing a high rate of mortality due to manifestation in the posterior circulation. The CAVEATS [113], VAST [114], and VIST [115] trials have investigated stent-assisted PTA of the vertebral arteries, and the results have suggested medical treatment as the first line of therapy. However, the ongoing debate about which treatment is better resembles that with ICAS.

Limited data are available regarding the use of DCB as primary treatment for VAOS. Wang et al. [116] reported the first case of DCB PTA in a patient with high-grade stenosis originating at a dominant vertebral artery, who did not qualify for stent-assisted PTA. The treatment was uneventful, as was the follow-up at 6 months, and no restenosis was observed on computed tomography angiography after that period.

In 2020, Gruber et al. [117] published a retrospective single-center study on 12 patients treated with DCB only for VAOS. No technical failures or instances of restenosis were found after a mean follow-up of 6.1 months, and no periprocedural major adverse events were recorded.

Wang et al. [118] have reported a restenosis rate of 10% at 6 months after DCB treatment (Orchid, Acotec Scientific, Beijing, China; Dhalia, Acotec Scientific, Beijing, China) of VAOS in 26 patients. No periprocedural strokes/deaths were reported. The trial was a pilot for a larger, prospective randomized single-center trial comparing VAOS treatment through DCB (PCB, Acotec Scientific, Beijing, China) with distal embolic protection (n = 49) vs. bare metal stenting without embolic protection (n = 46), which was published in 2021 [119]. The technical success rate was comparable in both groups (94% for DBC vs. 96%), and fewer embolic DWI lesions were observed in the DCB group. The restenosis rate was comparable in both groups after the mean follow-up of 12 months (10% DCB vs. 13% bare metal stent). The reported rate of periprocedural major adverse events was 2% in total, with one case in each group.

Table 3 summarizes the results of the described studies.

Table 3.Results of studies investigating DCB as a primary treatment for VAOS.

Number of studies; period	Patients n	Follow-up	Restenosis rate	Periprocedural complication rate
4; 2018, 2020, 2021	88	7.5 months (IQR 6–12)	5% (IQR 0–10)	0.5% (IQR 0–2)

Of note, the role of DCB in the treatment of subclavian artery stenosis is unknown. In the current literature, only a few case reports have addressed this topic, including one case report of successful treatment of ISRS in the subclavian artery [120, 121, 122].

SAMMPRIS (2011/12) and VISSIT (2015) put the use of intracranial bare-metal stents for ICAS treatment on the second line. High-intensity statin and dual- to mono-platelet inhibition [13] has since then been considered first-line. Endovascular treatment is considered in cases of a second stroke in the same vessel territory under medical therapy. However, as devices and strategies have evolved, the idea of “leaving nothing behind” that originated in cardiology has prompted the investigation of the use of DCB as a primary treatment approach for atherosclerotic stenotic vessel disease in the neurovascular setting, too.

In 2018, Gruber et al. [50] were the first to publish a comparative retrospective cohort study, investigating the effect of ICAS treatment with a DCB (Neuro Elutax sV, Elutax, Aachen scientific, Aachen, Germany) vs. a bare-metal stent for intracranial stenosis treatment (Wingspan® Stent system, Boston Scientific, Marlborough, USA). In eight patients treated with DCB and 11 patients treated with the stent system, a significantly lower rate of the endpoint combination of ischemic re-event and/or restenosis was observed in the DCB-group (n = 1 DCB vs. n = 7 stent system). The follow-up time was 9.5 months in the DCB group and 10 months in the stent group. No peri- or intraprocedural complications were reported for the DCB-group, but one intraprocedural in-stent thrombosis and one case of temporary hyperperfusion syndrome were reported in the stent group.

In addition, in 2018, Han et al. [51] published a retrospective single-center analysis of 30 patients with 31 vessel segments treated for symptomatic high-grade ICAS with DCB (SeQuent® Please, B. Braun Melsungen, Melsungen, Germany). Two patients with periprocedural stroke were reported, and after a mean follow up of 10 $\pm$ 3 months, no new ischemic symptoms were observed. Angiographic follow-up revealed one case of asymptomatic restenosis (3%).

Also in 2018, Gruber et al. [53] described the primary DCB treatment (SeQuent® Please Neo, B. Braun Melsungen, Melsungen, Germany) of ten patients with high-grade, symptomatic ICAS. They received a median stenosis reduction grade of 28% (IQR 75–80% to IQR 45–53%) without restenosis during the follow-up period of 3 months. No periprocedural adverse events were recorded, and no new ischemic events or deaths were recorded in the follow-up.

Zhang et al. [54], in 2020, reported a retrospective study comparing the treatment of high-grade, symptomatic intracranial stenosis with bare-balloon PTA (n = 73) and DCB PTA (SeQuent® Please, B. Braun Melsungen, Melsungen, Germany; n = 42). Both groups achieved favorable outcomes; however, the restenosis rate in the DCB group was significantly lower than that in the bare-balloon group in a 6-month follow-up (DCB restenosis rate 5% vs. 34%). The rate of periprocedural complications was 3% in the DCB-group and 11% in the bare balloon–group.

A Chinese single-center retrospective cohort study published in 2020 [56] described DCB (SeQuent® Please, B. Braun Melsungen, Melsungen, Germany ) for the primary treatment of ICAS with bare balloon predilation. A total of 39 vessel segments in 35 patients were treated, and the angiographic follow-up was 11 $\pm$ 4 months. The authors reported a restenosis rate of 8%, with a disproportionately high incidence in intracranial ICA, thus suggesting that the risk of restenosis varies among vascular segments. The reported rate of periprocedural complications was 6% for two major complications.

Remonda et al. [123] have described two types of DCB (SeQuent® Please, B. Braun Melsungen, Melsungen, Germany; ELUTAX-SV, Aachen Scientific, Aachen, Germany) for the treatment of high-grade ICAS in a 2021 study with a total of 33 patients and 35 vessel segments treated. In a mean follow-up of 9 months (IQR 3–22 months), restenosis occurred at a rate of 12%. The authors reported a 6% rate of periprocedural complications. The results were not reported according to the balloons used.

In 2020, Yang et al. [124] investigated the safety and short-term efficacy of DCB for symptomatic intracranial vertebrobasilar artery stenosis exclusively. A total of 16 treated patients showed good to very good results (0–20% residual stenosis), and no restenosis and new symptoms were observed in the 6-month angiographic and clinical follow-up. The rate of periprocedural complication was 6%.

Table 4 summarizes the results of the described studies.

DCB for the treatment of ICAS restenosis and ICAS ISRS is another field of DCB use in the neurovascular setting. However, data on this topic are scarce. The need for this treatment is based on the frequent rate of restenosis of ICAS after balloon and stent treatment. A current review from China on risk factors and the incidence of intracranial ISRS has pooled 5043 patients from 51 studies on the subject and reported a 12–18% rate of relevant restenosis in intracranial stents after the first ICAS treatment at a mean follow-up time of 18 months [125]. Among the patients with ISRS, 29% were symptomatic.

As in cardiology, neointimal proliferation is suspected to be a main factor influencing intracranial ISRS [27], among other factors including thrombus reorganization, elastic return, vascular remodeling, or platelet aggregation [126]. Regarding this pathophysiology, data published in 2011 and 2012 by Vajda et al. [46, 47] on the treatment and prevention of intracranial ISRS were promising. In the 2011 study, among 51 patients with intracranial ISRS, 63 maneuvers of restenosis treatment were performed, 41 using a DCB (SeQuent® Please, B. Braun Melsungen, Melsungen, Germany) vs. 20 using a bare balloon. After a mean follow up of 7.5 months with a maximum of 17 months, a restenosis rate of 9% was reported for DCB vs. 50% for bare balloons. The rate of major periprocedural events was 3%.

The data published in 2012 addressed the prevention of ISRS by investigating the effects of using a DCB (SeQuent® Please, B. Braun Melsungen, Melsungen, Germany) before applying a self-expanding stent in symptomatic ICAS. This maneuver, performed in 51 patients, resulted in a restenosis rate of 3% after a mean follow-up of 9 months and a maximum of 16 months. The rate of major adverse events was 5%.

However, both studies have described technical problems with the applied DCB relating to the stiffness of the device, which complicated control and led to technical failure in a minority of cases.

Since then, only Xu et al. [55] further investigated the use of DCB (SeQuent® Please, B. Braun Melsungen, Melsungen, Germany) of intracranial ISRS treatment in a study of 11 patients with symptomatic ISRS with DCB. No symptomatic strokes or deaths were reported for the supplied territory of the treated vessel in a 3-month follow-up in this preliminary study.

Table 5 summarizes the results of both studies addressing ISRS treatment with DCB.

To date, there have been no investigations or recommendations on the use of DCB in cervical or intracranial dissected vessels, nor for occluded vessels (via embolus) or comparable intracranial artery diseases with limited indications for revascularization.