Vertebral artery dissecting aneurysm (VADA) is a very rare subtype of intracranial aneurysm, which when ruptured, is associated with significant rates of morbidity and mortality. Although there have been many discussions and reports about VADAs, the optimal treatment method has not yet been clearly established. In the case of ruptured VADAs, a reconstruction method including stent coiling or multiple stents, or a deconstruction method such as parent artery trapping is performed according to the characteristics of the case. However, for the treatment of unruptured VADAs, conservative treatment using anticoagulation therapy is the first line of treatment; nevertheless, depending on the status, including the size of the dissecting segment and the change in shape or progression of related symptoms, endovascular treatment is also actively considered [25].

The treatment is determined according to the dominance of the parent artery and the inclusion of posterior inferior cerebellar artery (PICA) or various branches in the dissecting segment, and is not different from the treatment for ruptured VADAs. In recent times, with the development of endovascular devices, treatment of such cases using flow diverter devices (FDD) such as Pipeline Embolization Device (PED; Medtronic Inc., Minneapolis, MN, USA), Flow Redirection Endoluminal Device (FRED; Microvention, Aliso Viejo, CA, USA), and Surpass Flow-Diverter (Surpass; Stryker Neurovascular, Fremont, CA, USA) has been reported. However, such reports were individual cases with successful treatment outcomes [19,27,28]. The purpose of this study was to determine the radiological and clinical results of consecutive cases treated with FDD for unruptured VADA considered to be hemodynamically unstable lesions.

This retrospective study was approved by the Institutional Review Board of Pusan National University Yangsan Hospital (IRB 05-2021-177).

There were nine patients (seven females and four males), and the mean age was 54 years (range, 32-74) (Table 1). The exact location of the lesion was determined by diagnostic angiography and MR vessel wall images. Eight patients were V4 segment and one patient was V3 segment. In seven of these patients were PICA involving dissecting segment. The dominancy of the left and right side was compared through diagnostic angiography, and if the size was twice as thick, it was considered as a dominant. The lesion was in the dominant VA in two patients, in the non-dominant artery in two patients, and in five patients, the left and right dominance were not clear.

On admission, five patients had a lesion-related occipital headache, but four had no related symptoms. Therefore, the mRS score of five patients was 1 point.

Endovascular access and treatment were successful in all nine cases. All devices were successfully placed over the dissecting segment, providing complete coverage and demonstrating contrast stagnation in the dissecting segment on immediate angiography and 15 minutes delayed angiography. There were no peri-procedural or immediate post-procedural complications, except in one patient (case No. 6). The patient complained of mild facial palsy and numbness on the day after the procedure, and a small infarction was found in the medulla on diffusion-weighted magnetic resonance imaging (MRI) (Fig. 1C). The patient’s symptoms completely disappeared the next day, and he was discharged on the third day. All patients were discharged following the 3 days follow-up after the procedure. At the time of discharge, none of the patients complained of related symptoms or signs. None of the patients complained of discomfort during the outpatient clinical follow-up. All the patients who had a headache reported improvement in headache.

The data of follow-up conventional angiography at 3-6 months was available for all patients (Table 2). Complete/near-complete obliteration (grading scale 3 and 4) was seen seven patients. In one patient, the maximal aneurysm size, including the thrombus was decreased, but partial obliteration state. At the 12-month follow-up angiography, the results were not different from those of the first angiography. In one patient (case No. 6, Fig. 1), the first conventional angiography revealed occlusion of the dissecting segment in which the FDD was deployed.

All nine patients were able to confirm their clinical status as planned, at the outpatient clinic. Six months after the procedure, the antiplatelet medication was changed from dual to monotherapy.

VADA is a very rare disease; however, due to the high morbidity and mortality associated with it, careful management with neurosurgery is necessary. Since MRI was introduced as a diagnostic modality, unruptured VADA has been diagnosed more frequently [21]. VADA alters the shape of the parent artery leading to narrowing, widening, occlusion, or pseudo-aneurysm formation. As a result, it might cause sub-occipital headache, cerebral infarction, or SAH. In particular, in patients presenting with SAH, a poor outcome is predicted due to the high rate of re-bleeding [20,30]. Therefore, it is important for neurosurgeons to analyze the risk factors for rupture and decide on an appropriate treatment for unruptured VADA.

Anticoagulants are preferred for the conservative management of hemodynamic stable VADA [25]. However, there is a study showing that there is no difference in efficacy of antiplatelet and anticoagulant medications at preventing stroke and death in patients with symptomatic carotid and vertebral artery dissection [2]. Since we are more familiar with antiplatelet drugs for 1 year in VADA patients who needed conservative care.

Treatment for unruptured VADA is debatable, and there are no established guidelines about the optimal treatment. Kobayashi et al. [14] recommended treatment if the dissection segment length was more than 10 mm or if symptoms such as associated pain or mass effect were present. Nakagawa et al. [22] reported that treatment is necessary if an increase in the size of the dissecting segment is detected within a short period of time. Conversely, Kim and Lim [13] reported that spontaneous healing of unruptured VADA is more likely in females, nonsmokers, those with no involvement of the posterior-inferior cerebellar artery, and less luminal irregularity/steno-occlusion. Therefore, unruptured VADA should be treated considering the hemodynamic status, and the treatment should be carefully decided based on each individual case.

The well-known treatment methods include reconstructive and deconstructive treatment. The deconstructive treatment method completely occludes the dissecting segment using an endovascular or surgical approach. This method is considered a stable treatment in that there is no recurrence of the lesion [12]. However, in the case of dominant VA, there is a risk of ischemic symptoms due to the occlusion of a small branching artery and hemodynamic changes. In addition, there is a disadvantage that bypass needs to be considered for lesions involving PICA. Reconstructive treatment maintains the flow of the dissecting segment [5,31]. It includes methods such as stent-assisted coil embolization, deploying a single stent, or performing multiple stenting. Nam et al. [23] reported that internal trapping is a stable and effective treatment with no recurrence in ruptured VADA. Reconstructive treatment using stent and coil requires serial DSA follow-up for the risk of recurrence, but it could be a feasible alternative for ruptured VADA [23]. Li et al. [17] reported that in unruptured VADA, internal trapping increases the risk of ischemia, and multiple stenting with coil embolization could reduce recurrence and occlude the aneurysm. There are reports of successful treatment using multiple stents [18,26]; however, there are also reports of risks such as in-stent stenosis, thrombosis, and branch occlusion [7,16].

The use of FDDs is a relatively new and challenging endovascular treatment method that has redefined aneurysm treatment. Although there are differences in the structure and composition of each FDD, they provide high metal coverage, divert the blood inflow of the aneurysm and promote thrombus formation in the aneurysm [15]. At the same time, the parent artery should be reconstructed, and the patency of major and minor side branches must be maintained. In eight of our nine cases, blood flow in the side branches and parent artery flow were maintained well, except in the one case with segmental occlusion (case No. 6).

There are several reports of successful treatment of VADA using PED. Yeung et al. [32] reported excellent clinical and angiographic outcomes of four cases of unruptured dissecting VADA treated with PED. Although this report included long-term outcomes with a mean follow-up of 22 months, the dissection segment was not long. Cohen et al. [3] reported successful treatment using PED in eight cases with ischemic cerebrovascular conditions, including three cases of VADA. The cases of VADA were successfully treated, but another case of chronic internal carotid artery dissection had segment occlusion. Kühn et al. [15], reported that a 6-month angiographic follow-up showed complete occlusion in three cases, near-complete in two cases, and partial occlusion in one out of six cases of unruptured VADA treated using PED. In addition to these reports, Corley et al. [4] reported nine cases treated with PED, of which eight had complete occlusion. Of these nine cases, five presented with SAH, and one case presented with infarction. According to Corley et al. [4], in aneurysms with saccular components larger than 10 mm, using additional coils is recommended to promote aneurysm occlusion and reduce the risk of re-bleeding. Therefore, in our case No. 3, it is presumed that the aneurysm was not completely occluded because the saccular component was large at 13 mm.

There are reports of successful treatment using FDD; however, there are also reports about some necessary cautions. Typically, hemorrhagic complications, including delayed or remote hemorrhage and thromboembolic events, including in-stent thrombosis, are reported. Tan et al. [29] reported that five of 74 patients (6.8%) with PED had a thromboembolic event, and one patient (1.4%) developed delayed intracranial hemorrhage. Moreover, 50.9% of patients exhibited changes on post-procedural MRI diffusion-weighted imaging, and up to 14% experienced symptomatic infarctions [29]. Delgado Almandoz et al. [6] reported that among 48 PEDs used in 44 patients, they experienced four (8.3%) thromboembolic events and four events (8.3%) of hemorrhagic complications. Acute thromboembolic events might occur due to direct perforator blockage from a tine or strut of the FDD; chronic thromboembolism might occur due to excessive neo-intimal proliferation [4]. Fiorella et al. [8] reported a case wherein a patient developed very late fetal thrombosis a year after multiple PEDs with coils were placed for fusiform aneurysms of the dominant VA. The patient’s contralateral VA was occluded before the procedure, and eventually, the patient died due to a large brain stem infarction [8]. They also could not confirm the cause of delayed thrombosis. In our illustration case No 3, it was difficult to ascertain at what point the occlusion developed. Fortunately, the contralateral VA was patent, and bilateral PCoA was very well developed; hence, no clinical symptoms occurred.

Although there are few reports about the results of using FDD as a treatment for VADA, there are also few reports about its complications; nevertheless, there has been a regular concern about the safety of FDD use. In addition to the thromboembolic events mentioned above, there are complications related to hemorrhage. These are complications in which an aneurysm ruptured despite treatment with FDD, or intracerebral hemorrhage (ICH) occurred in the ipsilateral hemisphere where the FDD was used. Although it is a relatively early data from a retrospective study on PED among 793 patients, spontaneous rupture occurred in five patients (0.6%), and intra-parenchymal hemorrhage occurred in 19 patients (2.4%) [10]. In a pipeline for uncoilable or failed aneurysms trial, 107 patients treated with pipeline, ICH was reported in five patients (4.6%) [1]. The exact mechanism of the hemorrhage complications has not been elucidated. However, as a possible theory, the most probable cause of ipsilateral hemispheric intraparenchyme hemorrhage is procedural embolization of some type of material-small thrombus, air bubbles, catheter coating that result in microvascular damage, transient occlusion with microinfarction, or hemorrhagic reperfusion [1].

However, in the recently published The Pipeline Flex with SHield Technology Embolization-An International MulticEnter ObservationaL Post-Market StuDy (SHIELD), a multi-center, prospective study on improved PED, ICH did not occur and major stroke occurred in six patients (2.9%) [28]. Nevertheless, it needs to be noted that about half the patients (102 patients) had small aneurysms of size 7 mm or less, and in 47 patients (23%) PED was used as re-treatment [28].

We reported the angiographic and clinical results of treatment with FDD in nine cases of unruptured VADA. Within 1 year after the procedure, seven cases showed complete/near-complete occlusion, one case with a 13 mm sac showed partial occlusion, and one case showed segmental occlusion. Although this is a report of only nine cases, it is valuable in that it reports the treatment outcomes of FDD for long segment VADA that was considered hemodynamic unstable. The cause of segmental occlusion in one case was unknown; however, it indicates that more caution is necessary in such cases than when FDD is used for dominant VA. Another limitation is that we used FRED more than PED, which is the most widely used FDD. However, we believe that our study contributes to the evidence regarding the safety and effectiveness of FDD in the treatment of unruptured VADA. We consider the FDD to be a feasible treatment option for unruptured VADA.