Rescue angioplasty and stenting in refractory acute vertebrobasilar occlusion after mechanical thrombectomy: A single center experience
Article information
Abstract
Objective
Acute vertebrobasilar occlusion can led to a fatal outcome, but lack of established procedures poses many difficulties in its management. Although mechanical thrombectomy (MT) has shown positive outcomes recently, high reocclusion rate remains a hurdle. This study is to share experience and to review technical challenges of rescue angioplasty and/or stenting (RAS) for refractory occlusions after MT in posterior circulation ischemic stroke (PCIS).
Methods
Out of 494 patients with acute ischemic stroke from January 2014 to December 2022 in a Hospital, PCIS was identified in 50 patients. 2 extracranial vertebral artery occlusion patients were excluded. For 48 patients, MT was applied as the primary treatment. RAS was done for reocclusion after MT in 15 patients. We evaluated patient characteristics and clinical course, emphasizing the technical aspects of treatment.
Results
Compared to those without rescue procedures, RAS group had a higher percentage of large artery atherosclerosis as an etiology (p<0.001), long segment occlusions (p=0.03), and was more likely to involve posterior inferior cerebellar artery (p=0.007). There was no difference in functional outcome at 6 months between these two groups. Reopening could not achieve (N=2, 13.3%) and procedural complication rate (iatrogenic rupture) is 6.7% in RAS group. Rescue procedures were complicated with dissection, plaque rupture and migration, device damage, and misplacement of the balloon/stent. Avoiding these traps, finding true lumen, and reconstructing the flow by connecting the proximal and distal normal were the keys to the successful RAS.
Conclusions
RAS could be inevitable during endovascular treatment for PCIS and being aware of possible events and technical strategies would navigate interventionists to successful recanalization and the better outcomes.
INTRODUCTION
The posterior circulation of the brain plays a crucial role in supplying essential organs such as the brain stem, cerebellum, and thalamus [4]. Consequently, posterior circulation ischemic stroke (PCIS) presents distinct clinical features that differ from those observed in anterior circulation ischemic stroke (ACIS) [10]. However, there are not many studies (BAOCHE, ATTENTION trials) proving the efficacy of this treatment in PCIS and they mostly focus on basilar artery occlusion [6,12].
In certain instances, complete recanalization cannot be achieved solely through mechanical thrombectomy, particularly in the posterior circulation. The first-pass effect occurred only in 24.4% of patients with vertebrobasilar occlusion according to one recent report [11]. In addition, acute vertebrobasilar atherosclerotic reocclusion can occur even after mechanical thrombectomy due to the formation of new clots or the progression of underlying atherosclerosis [7]. This reocclusion can lead to a recurrence or worsening of the ischemic stroke symptoms and may require additional interventions beyond mechanical thrombectomy to restore blood flow and prevent further neurological damage.
Rescue angioplasty and/or stent insertion (RAS) are the possible options for these refractory occlusions. However, with regards to RAS, previous reports have not focused solely on PCIS and details of procedures were missing despite each step of the procedure must be modified according to the specifics of circumstances [5].
MATERIALS AND METHODS
Patient population
We conducted a retrospective study on hyperacute ischemic stroke patients who underwent consecutive mechanical thrombectomy procedures between January 2014 and December 2022 at a Hospital. A total of 50 PCIS patients over 18 years of age who underwent mechanical thrombectomy (IV thrombolysis, thromboaspiration, and/or stentriever) were included in the study. 2 patients were excluded as they were extracranial vertebral artery occlusion cases and the rest, 48 patients, was the final study population. Of this population, 15 patients exhibited in situ stenosis with reocclusion after initial mechanical thrombectomy and underwent RAS procedures. The study protocol was approved by our Institutional Review Board. (IRB No. 2021-10-006)
Stroke evaluation and endovascular procedure
PCIS was diagnosed with clinical symptoms after intracranial hemorrhage was excluded by brain computed tomography. Computed tomographic angiography (CTA) was performed routinely. Magnetic resonance (MR) was optional. These images showed the available access route and expected perfusion area. When the occlusion was confirmed by Allura Xper (Philips Medical System, Best, The Netherlands) biplane system, the first treatment modality was usually thromboaspiration. Stentriever was a second option. During stentriever, thromboaspiration was performed together as far as possible. Three passages were tried on average and rescue stenting was considered after trials. Contrast filling patterns were checked and further passage was decided according to change of occlusion patterns. Therapeutic alternatives were discussed within a multi-disciplinary team of neurology, neurosurgery, and neuro-intervention for the decision-making.
In rescue angioplasty and/or stent insertion, microwire always was gone through lesion with looping technique. Balloon size was underestimated under 70% of parent artery, angioplasty was performed several times, the location of angioplasty was just on main occlusion at first, and repetitive procedure was done back and forth if occlusion segment was long. The number of consecutive angioplasties was typically three on refractory circumstances. Stent size selection was also the same as balloon and open cell stent was preferred. Multiple stents insertion was not considered routinely if steno-occlusion involved focal lesion. Immediately after stent insertion, a loading dose of tirofiban was administered intravenously as soon as possible, followed by an additional maintenance dose. Poststenting angioplaty was applied only in flow compromise, the ballooning was performed leaving the microwire in the true lumen, and blood flow was checked after each angioplasty.
Factors affecting recanalization and outcomes
Several factors were considered for recanalization and outcome. The VA is divided into 4 segments and the BA is one trunk without segment. Short segment means that occluded lesion involved just one segment and long segment is 2 or more segments. Occlusion segments was concluded mainly by subtraction angiography and initial CTA or MRA was considered if insufficient images were presented in angiogram.
Recanalization may be affected by status of collaterals. To evaluate this, Yan L. et al presented the Posterior communicating artery(PCoA)-P1 score as follows: 1) 0 point: absence of a bilateral anterior-to-posterior circulation connection; 2) 1 point: presence of a unilateral PCoA-to-P1 connection, but with PCoA or P1 hypoplasia; 3) 2 points: presence of a unilateral normal PCoA-to-P1 connection or a bilateral PCoAto-P1 connection, but with either PCoA or P1 hypoplasia; 4) 3 points: presence of a bilateral PCoA-to-P1 connection, but with one side normal combined with contralateral PCoA or P1 hypoplasia; 5) 4 points: presence of bilateral connections between a normal PCoA and P1. Primary collaterals were dichotomized into good (2-4 points) and poor (0 or 1 point) [13]. Posterior inferior cerebellar artery (PICA) was defined to be involved in acute vertebrobasilar occlusion if the flow was compromised around vertebral artery (VA)-PICA junction or anterograde contrast filling pattern was heterogenous around the junction.
Clot migration is another factor for outcomes during thrombectomy and is divided as follows: 1) no thrombus migration through the procedure; 2) thrombus migration, large artery occlusion occurred, and additional procedure had to be performed; 3) thrombus migration, flow compromised at small vessels, another procedure was performed as it was functional area; 4) thrombus migration, small vessel occluded, no additional procedure. Time variables included in analysis were puncture to clot (PtC) and clot to end of procedure (CtE). PtC was defined as the time from arterial puncture to first access of the clot by microcatheter, and CtE was the time from first access of the clot by microcatheter to final angiogram.
Outcome was evaluated with modified thrombolysis in cerebral infarction (mTICI), post-procedural 3-month modified Rankin Scale (mRS), post-procedural 6-month mRS, post-procedural 12-month mRS.
RESULTS
1. Population characteristics
Initial presentation
Characteristics of the cohort are summarized in Table 1 and all variables were compared between RAS group and non-RAS group. Non-RAS group means that the recanalization by MT was achieved without RAS. Mean age was 68.3±11.3 (50-92) and male was predominant (n=35, 72.9%). The mean National Institutes of Health Stroke Scale (NIHSS) score at the time of admission was 11.67 for RAS group, and 20 for No RAS group (p=0.011).
Initial angiogram showed vertebrobasilar occlusion in 15 patients (31.3%) and basilar artery occlusion in 27 (56.3%). For 12 out of 48 patients (25%), PICA was involved, and collaterals were poor in 35 patients (75%). In RAS group, a higher percentage of long-segment occlusion was observed (60%, p=0.030).
According to the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) classification, the majority of strokes were caused by large-artery atherosclerosis (LAA) (47.9%) and cardioembolic (37.5%). There was a significant difference in the TOAST classification between RAS and No RAS groups (p<0.005): LAA was more common in the RAS group, while cardioembolic was more common in the No RAS group. One-third of No RAS group had underlying atrial fibrillation, while 2 out of 15 patients had it in RAS group.
Management and outcomes
Recombinant tissue Plasminogen activator (rt-PA) was administered to 11 (22.9%) patients mainly in No RAS group (10 out of 11 patients, p=0.030).
Mechanical thrombectomy was performed in all patients, but RAS group had a higher percentage of combined thromboaspiration and stentriever procedures (73.3%) than No RAS group (24.2%) (p<0.005). In No RAS group thromboaspiration only was the most common. PtC was an average of 22.3 min and there was no difference between two groups. CtE was significantly longer in RAS group (p=0.002). Among them, 12 patients achieved recanalization with RAS while 2 patients could not achieve reopening. 1 patient underwent parent artery occlusion due to iatrogenic rupture by balloon angioplasty. Hemorrhagic transformation was observed in 10.0% of the patients, with no significant difference between the two groups. At the 3-month and 6-month follow-up, there was no statistically significant difference in mRS between these groups. But for No RAS group, 18.2% of patients were expired.
2. Technical notes of RAS
Successful rescue angioplasty and/or stent insertion (Fig. 1)

Usual case of rescue angioplasty and stenting. (A) Subtraction angiography revealed occlusion of basilar artery. (B) Mechanical thrombectomy was performed with stent retriever in combination with thrombus aspiration and recanalization was achieved. (C) Shortly, reocclusion of basilar artery was observed. (D) Rescue balloon angioplasty and stent insertion were done for reocclusion. (E) Because of progressive re-stenosis, post-stenting balloon was done. (F) The final angiogram showed stable plaque with good patency
Of 15 patients (RAS group), 12 patients obtained successful reperfusion, defined as mTICI grade 2b or 3. The recanalization was achieved by routine thromboaspiration and/or stentriever technique. After recanalization, serial angiogram presented progressive occlusion. Pre-stenting balloon angioplasty preceded stents insertion in 9 cases and stent was inserted without pre-ballooning in 3 cases. Balloon angioplasty was performed after stenting in 11 patients and 1 case showed only stenting due to small caliber parent vessel. A larger balloon was selected only when intrastent stenosis was less than 50%. Serial angiogram was performed, and post-stenting ballooning was tried several times if angiogram showed compromised flow pattern. An exemplary case is depicted in Fig. 1.
Refractory rescue angioplasty and/or stent insertion (Fig. 2)

Disastrous case. (A) vertebrobasilar occlusion identified. (B) The first trial was applied with stentriever with thromboaspiration and flow reconstructions were partially obtained. (C) The stent was inserted due to progressive occlusion. However, angiography after stenting showed no flow to vertebrobasilar artery. (D) Balloon angioplasty had been performed several times. (E) Main blood flow was restored slightly but plaque was migrated to occlude basilar top. (F) Basilar artery was reoccluded progressively and could not be reopened despite multiple ballooning and thromboaspiration were done.
Final recanalization was not achieved in 2 patients despite multiple attempts to restore blood flow. Both cases showed LAA in TOAST classification, long segment involvement, and main occlusion was initiated just above PICA orifice. Every step of procedures was difficult, even a simple pass could not be obtained with a few trials and errors. Temporary anterograde perfusion was achieved by routine thromboaspiration and stentriever in one case and not in another case. Stent was inserted first due to difficult navigation and then balloon angioplasty was performed several times in one case. In another case, angioplasty was done first in multiple segments and multiple stenting was performed. Results were not improved despite intra-arterial pharmacologic thrombolysis, in-stent thromboaspiration, and several times of angioplasty or multiple stents.
Unclassified case of rescue angioplasty (Fig. 3)

Case with complication. (A) Mid-basilar occlusion was identified on angiography. (B, C) Mechanical thrombectomy was performed to restore the blood flow. (D) Progressive occlusion occurred. (E) Balloon angioplasty was performed for basilar artery lesion but iatrogenic basilar perforation occurred. (F) Torn basilar artery was occluded with coils and further treatment was no longer possible.
One case presented basilar artery occlusion and stentriever was chosen as the first treatment modality. Balloon angioplasty was performed for mid-basilar artery lesion and recanalization was achieved. However, basilar artery was re-occluded spontaneously, and during navigation by microwire, basilar top was torn. Hemorrhage was stopped with coils, but further treatment was no longer possible.
3. Other procedure-related complications
Stent thrombosis occurred within 24 hours because dual antiplatelet medication preparation or tirofiban injection was missed in one case. Parent vessel perforation and tearing occurred in one case. Atheroma migration and dissection were presented in 2 cases.
DISCUSSIONS
Within our cohort, this treatment modality demonstrated variability in success, linked to patient characteristics and clot dynamics. We scrutinized the technical challenges inherent to RAS and endeavor to delineate the procedural steps modified for the unique demands of PCIS.
As described in representative cases above, there are technical challenges of navigation for the reocclusion cases after mechanical thrombectomy. Under angiographic occlusion, device navigation follows a blind pattern and must be performed instinctively. The microwire looping technique has been always introduced. Contrast injection through the microcatheter might help distinguish between scattered and continuous contrast-filling patterns, aiding in the separation of the true lumen from the false lumen.
Failure to achieve proper recanalization and inadvertent passage through undesirable lumens can result in device-induced plaque disruption and migration [9]. Typically, occlusion arises from the most fragile fibrous cap, followed by the detachment of the thrombus from the atheroma, initiating the coagulation cascade (Fig. 4).

Schematic drawing (A) This figure is the best optimal navigation of microwire without contact with plaque. (B) As the microwire penetrates or (C) permeates the entire plaque, a variety of events occur. (illustrated by Joo Hyun Park)
There might not be definite radiological findings that make next procedure clear. The more thrombectomy was performed, the worse lesion got. During 3 times of thrombectomy, the shape of the luminal irregularity continued to change in the direction of the flow impairment, concurrent occlusion was seen on angiography, and rescue procedures were introduced right away.
Short vs long segment occlusion
While short-segment occlusions may carry the better prognosis due to a smaller clot burden and the potential for flow reconstruction, long-segment occlusions in atherosclerosis present a completely different scenario. These occlusions are characterized by a large burden of thrombus formation, making it difficult to identify the initial location of the coagulation cascade. Long segment occlusions indicate in situ stenosis along an extended segment, with the fibrous cap being disrupted at various points. Thrombus formation occurs rapidly within the confined space, propagating and solidifying around the site of the lesions. Navigating through undesirable lumens can inadvertently dig into stable plaques or lesions, potentially leading to plaque migration during angioplasty or stenting. Plaque migration caused by rescue procedures is irrecoverable, making it challenging to remove the migrated plaque and resulting in the continuous formation of massive thrombus.
Flow reconstruction in rescue procedure
In short, to achieve successful rescue procedures, neurointerventionists must focus on finding the true lumen and avoiding contact with stable plaques as much as possible. During rescue procedures, it is essential to remove forming thrombus, deposit and stabilize plaques against the vessel walls, and employ in-stent aspiration and angioplasty repeatedly throughout the entire process. In cases where recanalization cannot be achieved despite concerted efforts, the number of passes through the occlusion site should be minimized [1,3], and new passages should be avoided. Utilizing undersized balloons and stents covering the entire lesion, single stent instead of double stents might also help recanalization. Especially in poststening angioplasty, a much smaller balloon size may be preferable in multiple stents for long segmental occlusions and full size may be better in single stenting.
In comparison to ACIS, re-occlusion is more prone to occur in the posterior circulation [8]. This susceptibility can be attributed to factors such as blood volume, main atherosclerotic rupture, and so on. These procedures did not involved simple angioplasty and stent insertion, next procedures after intervention were inevitable, and repetitive procedures followed. In ischemic stroke, atherosclerotic occlusion is initiated by plaque rupture and thrombosis is propagated. These lesions may be very advanced at the time of admission, and the procedure may result in further thrombotic lesions. Although there were no definite criteria for using Tirofiban, the infusion of glycoprotein IIb/IIIa inhibitors might enhance the efficacy of rescue procedures prior to the initiation of rescue procedures [14].
Limitations
The study was conducted at a single institution and relied on retrospective data, which raises concerns about selection bias and the lack of randomization. The mTICI scores were obtained for overall territory reperfusion and are yet to be validated in PCIS [2].
CONCLUSIONS
RAS may be inevitable in refractory mechanical thrombectomy for PCIS and we are aware of possible events. The aim of technical strategies is to avoid repetitive passage through main lesions and stabilize the plaque for successful recanalization and better outcomes.
Notes
ACKNOWLEDGEMENTS
This work was supported by a research grant from Jeju National University Hospital in 2022.
Disclosure
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.