Flow Diverter Treatment of Ruptured Basilar Artery Perforator Aneurysms

Perforating artery aneurysms of the anterior circulation are a rare subtype of intracranial aneurysms (prevalence < 1%) [1, 2]. Basilar artery perforator aneurysms (BAPAs) are even less common, with a recent systematic literature review by Granja et al. [1] describing only 56 cases of non-trunk type BAPAs since their initial description in 1996 [3]. Categorizing BAPAs is a challenging task due to a potential overlap with so-called blister-like aneurysms of the basilar artery trunk. Aboukais et al. described BAPAs as those whose neck is entirely located on a perforator artery without direct involvement of the basilar trunk [4]. Recently, Satti et al. proposed a comprehensive classification system based on the angiographic description of four distinctive types of BAPAs, which also included small blister-like aneurysms arising from the basilar trunk adjacent to a perforating artery branch but not involving the origin of a perforating artery (type I) [2].

BAPAs are a rare cause of subarachnoid hemorrhage (SAH). Their diagnosis requires a high degree of suspicion as they are often easily missed due to their small size (85% ≤ 3 mm), partial thrombosis and small caliber feeding vessels on initial CTA, MRA, and even DSA studies [1, 2, 5]. Given the rarity of these aneurysms and their natural history, an ideal approach to treatment remains poorly defined. Various treatment options have been reported in case reports and small series, including microsurgical clipping, endovascular embolization using coils or liquid embolic agents, and flow diversion (FD) within the basilar artery, each with its own unique challenges and risks. Conservative treatment is an additional management option and multiple prior reports have suggested a high rate of spontaneous resolution of BAPAs [6‐9]; however, the rehemorrhage rate of BAPAs is reported in the order of 10–15%, which may result in death or severe disability [5, 6, 9, 10].

In this study, we present the multicenter experience derived from five large tertiary care neurovascular centers in the treatment of acutely ruptured BAPAs using FD stents.

We performed a retrospective review of treated BAPAs in five tertiary care, neurointerventional university centers in two countries. Patients were included when treated with a FD for a ruptured BAPA (according to the classification by Satti et al. [2]) between 2013 and 2020. Data on patient demographics, clinical and radiological presentation, procedural data and postprocedural course of each patient were analyzed. Angiographic outcome after FD treatment was analyzed according to the O’Kelly-Marotta grading scale [11]. We also collected clinical outcome and follow-up data at least 90 days following treatment. Due to the retrospective nature of this study, the diagnostic and therapeutic management of included patients was not standardized. The diagnostic work-up of all patients included a digital subtraction angiography (DSA) including a rotational DSA and 3D-reconstructions of the intracranial vessels, which have contact to the SAH or show any suspicious findings. The treatment of each patient was performed according to the local standards of each center and individual operators. Patient consent was obtained in retrospect, when possible, at most centers. At one center informed consent for patient data publication was obtained as part of a local prospective registry.

BAPA is a rarely reported cause of acute SAH. In this study, we aimed to pool a multicenter experience in the treatment of these aneurysms using FD stents. In our cohort, 18 patients were treated for a ruptured BAPA, which comprises to our knowledge the largest published case series of FD treatment for this type of aneurysm. There were no intraprocedural complications. Before discharge, four patients (Online Resource 3) suffered from FD-related ischemia. After discharge, another patient suffered from a delayed perforator ischemia on day 43 following treatment under single antiplatelet treatment. The overall long-term mortality rate was 13% (2/16).

BAPAs exhibit a small aneurysm size, small feeding vessels in cases of distal or non-trunk type, and the possibility of spontaneous thrombosis after rupture at the time of diagnosis. This makes them difficult to diagnose, which in turn may lead to an underreporting of these aneurysms as the cause of the acute SAH [12, 13]. In our series, we performed a rotational DSA with 3D-reconstructions of suspected vessel territories and a retrospective re-evaluation of the negative initial vascular imaging in all patients. The re-evaluation showed that all negative results in the initial DSA examinations were true negatives. Nonetheless, in our series, only 28% of the aneurysms were diagnosed on the initial angiographic imaging study, which is in line with recent systematic reviews of non-trunk BAPAs by Granja et al. [1] and of posterior circulation perforator aneurysms by Gardijan et al. [14] reporting diagnosis of the aneurysm on the first DSA study in 50% and 39%, respectively. In absence of a confirmed rebleeding, we tried to perform the repeat DSA after confirmation of resolution of SAH-related vasospasm, using transcranial Doppler. The decreased flow and image quality during vasospasm may potentially lead to a false negative result.

Early rebleeding of ruptured BAPAs prior to the diagnosis in the initial or repeat DSA was relatively common in 22% of our series (two patients prior to hospitalization and two patients following the initial negative vascular imaging). This rate is at the upper range compared to rebleeding rates reported in aneurysmal SAH patients in general (8–23%) [15, 16]. The latter may be in part attributed to the initial nondiagnosis and subsequent rehemorrhage that was triggering the diagnosis and eventually treatment of the ruptured BAPA. In a recent literature review of conservative treatment of BAPA a low rate of rebleeding was reported (approximately 10%). The difference is probably due to a possible underdiagnosis of these aneurysms, publication bias and/or due to the low number of cases in each collective.

The exact regimen for antiplatelet therapy with FD treatment in SAH was not standardized in our series. In most cases, a pretreatment loading with single or dual antiplatelet agents, alternatively a glycoprotein IIb/IIIa inhibitor was administered, followed by 6 months of dual antiplatelet therapy and then long-term monotherapy. 22% did not receive dual antiplatelet therapy. In our cohort, this heterogeneity did not seem to affect complications or outcomes following FD treatment of BAPAs. The heterogeneity of treatment regimens and its lack of effect on the clinical outcomes were similarly reported in a recent meta-analysis of FD stent treatment of acutely ruptured aneurysms [17]. A concern about using antiplatelet therapy in the acute phase is the potentially increased risk for hemorrhagic complications. This is of particular concern in cases requiring further operative procedures in the acute phase (e.g. for cerebrospinal fluid drainage). Seven% (16/223) hemorrhagic events following FD treatment of ruptured aneurysms were reported in the above-referenced meta-analysis [17], more than half of these hemorrhages (11/223) were not aneurysm related. Thus, the timing of external ventricular drainage should preferably precede the endovascular treatment. This seems to minimize hemorrhagic events associated with external ventricular drainage [18]. An alternative strategy maybe the use of thrombogenicity-reduced coated FD stents such as the PED Flex Shield [19] or the p48 hpc device [20] combined with antiplatelet monotherapy in acutely ruptured aneurysms. In our series, no rebleeding events were observed following antiplatelet therapy and as well as no in-stent thrombosis. This could be attributed to the application of antiplatelet therapy in standard dose and may be explained by the low number of included patients. In spite of an immediate angiographic incomplete occlusion, a direct protective effect of FD was observed in our cohort. Gardijan et al. [14] reported a similar effect.

FD-related ischemic events in the acute phase were encountered in 28% of our cohort. This rate is in the order of other published reports on the use of FD stents for ruptured posterior circulation aneurysms. In a recent meta-analysis, the total complication rate for posterior circulation ruptured aneurysms treated with flow diverter was 27% [17]. In another systematic review of non-saccular posterior circulation aneurysms treated by FD, the overall rate of periprocedural ischemia was 23% [21]. Likewise, high rates of ischemic complications were also encountered in other reports of ruptured perforator aneurysms of the posterior circulation treated conservatively (32%), or by endovascular therapy using FD stents and mostly stent-in-stent techniques (20%) [14]. The overall high strokes rates are most likely related to the large number of perforators in the posterior circulation. Covering these perforators with a FD stent may cause short-term or long-term ischemic complications due to perforator occlusion.

Our cohort showed an excellent occlusion rate on long-term follow-up angiography (100%) and favorable clinical outcome (mRS 0–2) in 81%. Similarly high occlusion rates for BAPAs (94%) were reported following endovascular treatment by the use of single stents (10%), telescoping stents (stent-in-stent, n = 45%) or FD stents (45%) in a recent review of the literature comprising only small retrospective case series (1–5 cases each) [14]. Overall, favorable outcome with these different stent/FD treatment strategies was 90% in their review.

The natural history of BAPAs remains unclear. Conservative treatment seems to be a valid alternative, as spontaneous resolution of these aneurysms has been described in many smaller series [1, 7, 14, 22]. Despite lower occlusion rates (70%) compared to endovascular treatment using stents/FD (94%), clinical outcome showed no statistically significant difference with favorable outcome of 77% (conservative) vs. 90% (endovascular therapy) in a recent systematic review of small case series [14]. In this review, ischemia was often observed with conservative treatment (32%); however, 56% of these were associated with a failed endovascular treatment attempt. Rebleeding may occur with a conservative approach (10%), and mortality was 6%. In view of a lack of evidence for a clear treatment decision between a conservative approach (watch and wait strategy) and endovascular FD stent treatment, future clinical research with larger sample sizes is definitely required. Such a study should ideally be designed as a randomized trial or a least as a prospective multicenter registry, which may be a difficult undertaking given that BAPA is such a rare disease and cause of SAH.

Our study was limited by the retrospective nature, small sample size and the heterogeneity of the treatment protocols that were left to the individual operator’s discretion. Moreover, angiographic and clinical outcomes were self-adjudicated by the contributing centers.

In our multicenter experience, FD treatment of ruptured BAPAs appears to have comparable safety and efficacy outcomes to FD treatment of other ruptured posterior circulation aneurysms as well as to the conservative management of BAPAs. This treatment strategy for a ruptured BAPA achieved a high rate of angiographic occlusion and favorable clinical outcome; however, as the conservative management also seems to offer similar clinical outcomes an individualized treatment decision is warranted. Future prospective studies comparing both approaches are required.