Long-term outcome and factors associated with restenosis after combination therapy of balloon angioplasty and stenting for symptomatic intracranial stenosis

Intracranial atherosclerotic stenosis (ICAS) is an important etiological factor for ischemic stroke, which accounts for 8 to 10% of patients with ischemic stroke in the United States, 18% in Japan, and 46% in China [1]. The recurrence rate in patients with symptomatic ICAS is high despite optimal medical treatment, although no treatment strategy has been established [2]. A recent cohort study with a follow-up of 2.8 years reported that the recurrence rate after medical treatment in patients with symptomatic ICAS was 14.9% [3].

The Stenting and Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) trial demonstrated the periprocedural stroke rate and death rate of intracranial stenting to be significantly higher (14.7%) than aggressive medical treatment (5.8%) [4]. However, as limitations of this trial, the rate of acute-phase patients was high and the operators had insufficient experience. The recent Wingspan Stent System Post Market Surveillance (WEAVE) trial, a prospective on-label study that used the Wingspan stent (Stryker, Kalamazoo, MI, USA), reported periprocedural stroke in 2.6% of patients [5].

Primary balloon angioplasty for symptomatic ICAS has been reported since the 1990s [6]. However, periprocedural complications and restenosis remain unresolved. Recently, primary balloon angioplasty without stenting was reported to have a lower periprocedural complication rate than stenting. Recent meta-analyses showed that the incidence of stroke and death within 30 days of submaximal angioplasty were 4.9% and 3%, respectively [7, 8]. The optimal endovascular treatment strategy for patients with symptomatic ICAS who do not improve after best medical treatment has not been established.

Although balloon angioplasty alone for symptomatic ICAS has been performed in Japan since the 1990s, the Wingspan stent has been used in selected patients after its approval in July 2014 [9]. The indication of the Wingspan stent has been limited to advanced dissection or acute occlusion after balloon angioplasty in Japan. We used selective balloon angioplasty and stenting based on arterial access and lesion morphology after the initial balloon angioplasty. The purpose of this study was to evaluate the initial and long-term outcomes of a combination therapy of balloon angioplasty and stenting for symptomatic ICAS at our institution. In addition, we investigated the factors increasing the risk of restenosis associated with endovascular treatment.

The baseline characteristics of the patients (n = 160) are shown in Table 1. Their ages ranged from 39 to 88 years, with a mean of 67.6 years. There were 124 males and 36 females. The stenotic arteries consisted of the MCA in 68 patients (43%), ICA in 49 (31%), VA in 31 (19%), and BA in 12 (8%). Concerning ischemic symptoms, cerebral infarction was observed in 118 patients (74%) and TIAs were observed in 42 (26%). Seven out of 160 patients had multiple intracranial severe stenosis. In this study, only obvious symptomatic stenosis was treated. The median interval from an attack until treatment was 21 days. Aspirin at 100 mg and clopidogrel at 75 mg were administered to 133 patients (83%), and aspirin at 100 mg and cilostazol at 200 mg were administered to 27 (17%).

Currently, there is no consensus regarding endovascular treatment strategies for intracranial stenosis. The problems of periprocedural complications and restenosis after treatment have not yet been resolved. We performed this retrospective study focusing on minimally invasive combination therapy tailored to individual patients rather than prioritizing intracranial stenting, and evaluated the periprocedural safety, long-term outcome, and factors associated with restenosis.

An increasing number of studies have recommended balloon angioplasty alone with minimally invasive techniques, such as submaximal dilatation [8]. The advantage of balloon angioplasty alone without stenting has been reported to be the limited damage to the vessel and relatively easy navigation. A meta-analysis showed that the incidence of stroke and death within 30 days was 4.9%, and that after Day 30 was 3.7% [11]. Another meta-analysis revealed that the incidences of stroke and death within 30 days and after 1 year were 3 and 5%, respectively [12]. The incidence of perioperative complications related to balloon angioplasty alone for symptomatic stenosis of the MCA was reported to be 4.2% [13]. In this study, the incidence of perioperative intracranial complications was 5.0% in the balloon angioplasty group and 6.8% in the stenting group.

The usefulness of stenting for symptomatic ICAS was not confirmed in a randomized controlled trial with medical treatment. Among ischemic complications, the incidence of perforator infarction was the highest, and basilar artery stenosis, advanced age, and DM were significantly associated with ischemic complications [14]. However, the results of treatment in a high-volume center through strict patient selection, excluding patients with acute ischemic stroke, were favorable. Wang et al. reported that the incidence of complications related to Wingspan stent insertion for MCA stenosis was 1.4% in the phase of skilled techniques [15]. Ma et al. reported that the incidence of perioperative complications in the Wingspan-stent-treated group was 4% [16]. In the WEAVE trial, the incidence of complications related to on-label procedures was 2.6% [5]. The conditions for on-label procedures in the United States included an age of 22 to 80 years, history of cerebral infarction (≥ 2 episodes), interval of ≥ 8 days from the onset of cerebral infarction, 70–99% stenosis, lesion length of ≤ 14 mm, and vascular diameter of ≥ 2 mm. On the other hand, the specifications of the Wingspan stent system, which was approved by the FDA in the United States in 2005, have not been modified to date, which raises an issue as a thick, hard delivery system must be guided.

There exists a lack of consensus regarding the best endovascular treatment strategy for symptomatic ICAS. To select an optimal treatment method, the site of the stenotic artery, lesion shape, and access route to the lesion are examined, and comprehensive risk assessment regarding treatment procedures is conducted. The patient’s clinical factors and interval from the final ischemic attack are important factors to consider. At our institution, the indication and procedure of treatment are determined through strict preoperative risk assessment. For initial treatment, primary balloon angioplasty alone is performed, and stenting is considered in patients with insufficient dilation, marked acute dissection, or restenosis. We do not recommend stenting as initial treatment for lesions of the MCA or BA involving perforators or branching vessels, which have a high incidence of complications [14].

One of the disadvantages of balloon angioplasty is considered to be more restenosis than stenting. Ueda et al. reported that restenosis was observed in 31.9% of patients who underwent primary balloon angioplasty for symptomatic MCA stenosis with a median follow-up of 63 months [13]. Previous studies reported the incidence of restenosis after balloon angioplasty alone as 24 to 50% [17, 18]. Three recent meta-analyses demonstrated that the rate of restenosis was 8.9% [7], 18.4% [11], and 20% [12].

In the SAMMPRIS trial, the 3-year cumulative incidence of restenosis after stenting for symptomatic ICAS was 14% [19], Furthermore, restenosis was reported to be significantly associated with recurrent ischemic stroke [19, 20]. The Wingspan One-year Vascular Events and Neurologic Outcomes trial demonstrated a relatively low 8.5% 1-year stroke and death rate in Wingspan-stent-treated patients [21]. A meta-analysis demonstrated that the incidences of restenosis in the United States and Asia were 28.8 and 13.6%, respectively, demonstrating a regional difference [22]. Furthermore, this meta-analysis demonstrated that younger age was related to high in-stent restenosis rates [22]. A systemic review of symptomatic intracranial stenting reported that the rate of restenosis was 15.6% [23]. However, risks factors of restenosis after endovascular treatment for symptomatic intracranial stenosis have not yet been established.

A recent randomized trial reported that compared with bare metal stents (BMS), drug-eluting stents (DES) reduced the risks of in-stent restenosis (ISR) and ischemic stroke recurrence in patients with symptomatic high-grade ICAS [24]. Another randomized trial found that the use of a DES compared with a BMS resulted in a decreased risk of ISR, a similar rate of successful implantation of the stent, and similar adverse events [25]. A small single-center study reported that drug-eluting balloons may be a promising alternative treatment for patients with symptomatic high-grade ICAS showing a significantly lower rate of ischemic stroke recurrence or ISR in comparison with the Wingspan stent-treated patients with a similar safety profile [26].

In this study, the overall incidence of restenosis was 26.3% and 5.8% per year. Multivariate Cox regression analysis showed that DM, length of the lesion, and balloon angioplasty were independent predictors for restenosis. DM was reported to be an independent relevant factor for restenosis and recurrent ischemic events after balloon angioplasty alone for symptomatic MCA stenosis [13]. Several studies reported that DM was a predictive factor for restenosis after balloon-mounted stenting for ICAS [27, 28]. For percutaneous coronary intervention (PCI) for coronary lesions, DM is an important predictive factor for restenosis [29]. Neointimal hyperplasia at the lesion site may be primarily involved in the pathogenesis of restenosis. DM promotes neointima formation after PCI [30]. The following clinical factors related to restenosis after PCI have been reported: vascular diameter, lesion length, stent length, and degree of residual stenosis [31]. These characteristics may also be similar for endovascular treatment for intracranial artery stenosis.

This study has several limitations. First, this was a retrospective study at a single center; thus, there was a bias in patient/treatment selection and no control medical arm. Our study design did not compare the superiority or inferiority of treatment outcomes in the percutaneous transluminal angioplasty and stent groups. In the future, a randomized controlled trial with multiple large-volume centers is needed to demonstrate whether this combination therapy is superior to aggressive medical treatment. Second, the results were obtained from a small number of Japanese patients, which may limit the generalizability of our results. Third, the follow-up data on clinical events or images were limited and incomplete data were included.

Endovascular treatment for patients with symptomatic arteriosclerotic ICAS may have potential of better clinical outcome if patients properly selected and treated by an experienced neuro-interventionalist at a high-volume center. However, our results do not allow us to define a profile for patients with this disease to choose between medical treatment and endovascular treatment. Furthermore, well-designed randomized controlled trials are needed to determine the optimal treatment for symptomatic ICAS and the patients targeted.

A combination therapy of balloon angioplasty and stenting for patients with symptomatic arteriosclerotic intracranial artery stenosis is a feasible procedure with a low incidence of perioperative complications. It may have the potential for better initial and long-term outcomes for appropriately selected patients. Significant restenosis-associated factors were balloon angioplasty and DM. Additional prospective and randomized trials are required to confirm the efficacy of this treatment for recurrent ischemic strokes.