2025 Volume 19 Issue 1 Article ID: cr.2025-0089
Objective: Mechanical thrombectomy is constantly advancing in terms of devices and techniques to enhance outcomes, and it is now commonly performed worldwide. Delayed stenosis is a potential complication that can occur following the procedure. However, its pathophysiology, risk factors, and time of onset remain unclear. We report 2 cases of delayed severe stenosis after mechanical thrombectomy.
Case Presentation: Case 1: A 70-year-old man was referred to our hospital with a right middle cerebral artery occlusion. We performed mechanical thrombectomy promptly. Complete recanalization was achieved after the 2nd thrombectomy attempt with no complications. Case 2: A 24-year-old man was sent to our hospital with an occlusion in his left middle cerebral artery. We quickly performed mechanical thrombectomy, successfully achieving recanalization after the 6th attempt. A post-procedural CT scan revealed an intracranial hemorrhage. In both cases, although no stenosis was observed the next day, significant delayed stenosis was discovered months later after mechanical thrombectomy. However, the patients remained asymptomatic.
Conclusion: Mechanical thrombectomy is a recognized treatment for acute ischemic stroke, but there are instances of delayed stenosis following the procedure. Excessive vessel injury caused due to thrombectomy may lead to delayed stenosis. This report emphasizes the importance of regular follow-up after mechanical thrombectomy, particularly in cases where there is a potential for substantial vascular damage due to thrombectomy.
The meta-analysis demonstrated the effectiveness of mechanical thrombectomy (MT) for acute ischemic strokes with large vessel occlusions in the anterior circulation.1) Following this result, MT is recommended as Grade A under the indicated conditions in the guideline.2) Devices and procedures of MT have matured, and optimizing therapeutic efficacy while reducing complications is crucial. Major MT-related complications include vasospasm, arterial dissection, and intracranial hemorrhage.3) Several reports indicate vessel occlusion and stenosis after recanalization, categorized as acute occlusion within 24 hours and delayed-onset vessel stenosis that follows.4,5)
Delayed stenosis following MT is an uncommon complication. Multiple reports indicate that vessel injury during MT could be associated with delayed stenosis. Nevertheless, there is still no consensus regarding the risk and timing of delayed stenosis following MT.
This report presents 2 cases of delayed severe stenosis following complete recanalization and describes the characteristics of such cases through a review of the literature.
Case 1 was a 70-year-old man who was referred to our hospital with motor impairment in his left upper and lower limbs 3 hours after the onset of symptoms. The neurological exam indicated left hemiparesis, left hemispatial neglect, and left facial paralysis. His National Institutes of Health Stroke Scale (NIHSS) score was 6. An MRI showed an acute cerebral infarction in the right middle cerebral artery region, with a diffusion-weighted imaging–Alberta stroke program early CT score (DWI–ASPECTS) of 8 points (Fig. 1A and 1B). MRA revealed an occlusion in the M1 segment of the right middle cerebral artery (Fig. 1C). We promptly conducted endovascular therapy to recanalize the occluded vessel.
Endovascular treatment commenced 185 minutes after symptom onset. An 8-Fr OPTIMO (Tokai Medical Products, Aichi, Japan) was placed in the right internal carotid artery. Angiography confirmed the occlusion of the right middle cerebral artery, as seen in the preoperative MRI (Fig. 1D). A Synchro select standard micro guidewire (Stryker, Kalamazoo, MI, USA) was guided to the occluded vessel, followed by a Phenom 21 catheter (Medtronic, Minneapolis, MN, USA). A Solitaire 6 × 40 mm (Medtronic) was placed at the occluded site. The Catalyst 6 (Stryker) was advanced to the occluded side using the stent as an anchor. The Aspiration-Retriever Technique for Stroke (ARTS) method was applied for thrombectomy (Fig. 1E). Following the 2nd thrombectomy pass, recanalization with a modified thrombolysis in cerebral infarction (mTICI) Grade 3 was achieved 222 minutes after symptoms began (Fig. 1F and 1G). Post-procedural angiography indicated no signs of vascular spasm, and the postoperative CT scans showed no complications, such as subarachnoid hemorrhage.
The MRI conducted the day after the procedure showed the recanalization of the vessel without vascular stenosis (Fig. 2A). Noncontrast black blood imaging showed no high-intensity lesion in the right M1 segment, suggesting no vascular dissection caused due to MT (Fig. 2B). Contrast-enhanced black blood imaging revealed vessel wall enhancement in the right M1 segment (Fig. 2C).
Atrial fibrillation was detected, and he was diagnosed as having cardiogenic cerebral embolism. Apixaban was initiated for secondary prevention. The patient was discharged without any neurological deficit.
One month following the procedure, the MRA revealed slight stenosis in the recanalized vessel (Fig. 2D). However, 3 months later, the MRA indicated severe stenosis in that vessel (Fig. 2E). DSA revealed 99% stenosis using the warfarin–aspirin symptomatic intracranial disease (WASID) method, while distal perfusion beyond the stenosis was maintained via collateral circulation through the anterior cerebral artery (Fig. 2F and 2G). Single-photon emission CT demonstrated preserved cerebral blood flow at rest but noted a decline in cerebrovascular reactivity. As the patient was asymptomatic, surgical intervention was deemed unnecessary, and aspirin therapy was commenced. At the 1-year follow-up, the patient remained asymptomatic, with the vascular stenosis unchanged.
Case 2 involved a 24-year-old male who was admitted to our hospital due to consciousness disturbances. His medical history revealed no significant issues. He exhibited a right hemiparesis and aphasia, and his NIHSS score was recorded at 19. MRI and MRA findings revealed an occlusion in the M1 segment of the left middle cerebral artery, accompanied by acute cerebral infarction in the left hemisphere, as indicated by a DWI–ASPECTS score of 7 (Fig. 3A–3C). We swiftly performed endovascular therapy to recanalize the occluded vessel.
Endovascular treatment began 201 minutes after symptoms arose. An 8-Fr OPTIMO catheter was positioned in the left internal carotid artery. Angiography revealed an occlusion in the left middle cerebral artery (Fig. 3D). A Synchro select standard micro guidewire was introduced to the blocked vessel, followed by the insertion of a Phenom 21 catheter. An Embotrap 5 × 37 mm (CERENOVUS, Johnson & Johnson Medical Devices, Irvine, CA, USA) was deployed at the occlusion site. An EMBOVAC (CERENOVUS) was navigated to the blocked side using the stent as support. Thrombectomy was performed using the ARTS technique (Fig. 3E). Unfortunately, the thrombus shifted distally, causing the occlusion of the M2 segment. We switched to Solitaire 4 × 20 mm (Medtronic) and a Catalyst 6, which are smaller-profile devices, and continued thrombectomy. Recanalization with an mTICI Grade 3 was achieved following the 6th thrombectomy pass, occurring 270 minutes after symptom onset (Fig. 3F). Postoperative CT scans indicated subarachnoid hemorrhage in the Sylvian fissure (Fig. 3G). There was no penetration by the guidewire during the procedure, and the hemorrhage was presumed to have resulted from avulsion of a pial vessel.
The patient was diagnosed with mitral valve prolapse and infective endocarditis, with embolic cerebral infarction determined as the underlying mechanism. He underwent cardiac surgery and started warfarin therapy.
The following day, the MRA revealed stenosis in the left M2 segment but showed no stenosis in the left M1 segment (Fig. 4A). A month later, the MRA revealed stenosis in the left M1 segment (Fig. 4B). As the stenosis in the M2 segment was reversible, it was considered to have been caused due to vasospasm. By 3 months post-procedure, the stenosis had progressed (Fig. 4C). Contrast-enhanced black blood imaging revealed vessel wall enhancement in the left M1 segment (Fig. 4D).
However, antegrade blood flow distal to the stenotic segment remained intact, and the patient stayed asymptomatic. To prevent further stenosis, aspirin therapy was initiated. A year after the procedure, there was no progression of stenosis following the start of aspirin therapy (Fig. 4E).
Informed consent for research and publication was obtained from the patients included in this study.
These cases experienced delayed severe stenosis that occurred after MT, despite complete recanalization. We performed a literature review and identified 17 cases of delayed stenosis following MT (Table 1).6–11) A retrospective study reported that 3.4% of patients developed de novo stenosis, while 0.9% experienced vessel occlusion 3 months post-MT.6) Another retrospective study revealed significant diffuse stenosis of target vessels in 8.3% of patients 3 months post-MT.4) Nonetheless, the cases detailed in this report utilized the Merci device (Concentric Medical, Mountain View, CA, USA), which is believed to cause more significant vascular injury than the latest stent retrievers.12) Marto et al. found that repeat occlusion happens in 6.6% of cases within 24 hours.5) In our present cases, an MRI conducted the day after the MT revealed no stenosis or occlusion of the target vessel. Eugène et al. noted that 10.3% of cases showed delayed de novo stenosis during the 1-year follow-up MRA, and delayed stenosis may illustrate device-related arterial wall damage.8) Research involving animals has indicated that iatrogenic vascular injuries can occur following MT, which includes the loss of endothelium and subsequent intimal proliferation.13,14) Thus, intimal proliferation arises as a consequence of arterial wall damage caused by the use of thrombectomy devices, which leads to delayed stenosis.
| No. | Authors | Year | Age | Sex | Region | Stent retrievers | Aspiration catheters | Number of device passes | Detection of the stenosis | Symptom |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Kurre et al.6) | 2013 | NA | NA | VA | Solitaire 4 × 20 | None | 4 | 3 months | None |
| 2 | 60 | NA | VA | BONnet, Solitaire 4 × 20 | None | 3 | 222 days | None | ||
| 3 | NA | NA | M1 | BONnet, short BONnet | None | 2 | 3 months | None | ||
| 4 | NA | NA | M2 | BONnet | None | 1 | 3 months | None | ||
| 5 | 75 | NA | M2 | BONnet, Solitaire 4 × 20, pREset | None | 3 | 154 days | None | ||
| 6 | Enomoto et al.7) | 2015 | 78 | M | M1 | Merci | None | NA | 3 months | None |
| 7 | 79 | M | M1 | Merci | Penumbra | NA | 3 months | None | ||
| 8 | 76 | M | M1 | None | Penumbra | NA | 3 months | None | ||
| 9 | Eugène et al.8) | 2015 | 47 | NA | M1 | Solitaire FR 4 × 15 or 4 × 20 | None | 4 | 1 year | None |
| 10 | 53 | NA | M2 | Solitaire FR 4 × 15 or 4 × 20 | None | 4 | 1 year | None | ||
| 11 | 68 | NA | M1 | Solitaire FR 4 × 15 or 4 × 20 | None | 1 | 1 year | None | ||
| 12 | 53 | NA | BA | Solitaire FR 4 × 15 or 4 × 20 | None | 1 | 1 year | None | ||
| 13 | Kim et al.9) | 2016 | 73 | M | M1 | Trevo provue | None | 1 | 77 days | Leukoencephalopathy |
| 14 | Yamaguchi et al.10) | 2018 | 48 | F | M2, ICA | Solitaire 6 × 30 | Penumbra 5MAX ACE | 4 | 90 days | None |
| 15 | Ito et al.11) | 2021 | 37 | F | M1 | REVIVE, Solitaire 4 × 20 | Penumbra ACE68 | 6 | 3 months | None |
| 16 | Our cases | 70 | M | MCA | Solitaire 6 × 40 | Catalyst6 | 2 | 3 months | None | |
| 17 | 24 | M | MCA | Embotrap 5 × 37, Solitaire 4 × 20 | EMBOVAC, Catalyst6 | 6 | 3 months | None |
BA, basilar artery; BONnet, phenox, Bochum, Germany; Catalyst 6, Stryker, Kalamazoo, MI, USA; Embotrap, CERENOVUS, Johnson & Johnson Medical Devices, Irvine, CA, USA; EMBOVAC, CERENOVUS; F, female; ICA, internal carotid artery; M, male; MCA, middle cerebral artery; Merci, Concentric Medical, Mountain View, CA, USA; NA, not available; Penumbra, Penumbra, Alameda, CA, USA; pREset, phenox; REVIVE, Codman Neuro, Raynham, MA, USA; Solitaire, Medtronic, Minneapolis, MN, USA; Trevo provue, Stryker, Kalamazoo, MI, USA; VA, vertebral artery
Recent MRI imaging studies demonstrate that MRI vessel wall imaging can detect vascular damage following MT. One clinical study noted greater endothelial damage when the thrombectomy device was significantly larger than the target vessel.15) Another clinical study using vessel wall imaging revealed a correlation between the extent of arterial damage and the number of thrombectomy attempts or devices employed.16) In our cases as well, contrast enhancement of the vessel wall indicates that thrombectomy induced endothelial injury. In previous reports, contrast enhancement of the vessel wall was observed in more than half of the cases following MT using stent retrievers.15,16) Furthermore, Lindenholz et al. reported that contrast enhancement of the vessel wall can be observed even with aspiration thrombectomy alone.17) This suggests that vessel wall enhancement is a common finding after MT. However, it remains unclear whether the presence or absence of vessel wall enhancement can predict delayed stenosis.
Endovascular treatment for acute ischemic stroke offers various procedural options, such as aspiration or stent retriever, which can be used together when necessary. However, the best choice of techniques and devices to consistently ensure total or near-total recanalization safely remains unclear.15) A study conducted with a swine model found that thrombectomy using a stent retriever resulted in significantly more severe vascular injuries than thrombectomy performed with aspiration.13) To the best of our knowledge, only 1 case of delayed stenosis has been reported following MT performed with aspiration alone.7) Therefore, MT with a stent may carry an increased risk of delayed stenosis compared to MT utilizing aspiration alone. Although the combined technique is commonly employed, the variety of methods incorporating stent retrievers and aspiration devices adds complexity to the assessment of vascular injuries.
Several studies have indicated that using a larger and longer stent retriever can achieve 1st-pass reperfusion.18,19) Although these reports indicate no significant differences in complication rates, they do not include delayed stenosis as a measured outcome. Nonetheless, an oversized device for the target vessel can increase endothelial injury.15) In our case 1, employing a 6-mm oversized stent retriever for the M1 segment, with a vessel diameter of about 2–3 mm, induced greater vascular injury.20) In our case 2, we performed 6 thrombectomy trials, which caused intracranial hemorrhage. Thus, it is likely that the substantial damage to the arterial wall was a result of the procedure. Multiple passes were also a contributing factor to the more extensive vascular injury. Therefore, the use of a relatively larger and longer stent retriever in the target vessels, as well as repeated procedures with thrombectomy devices, carries a high risk of delayed stenosis.
Typically, cases of delayed stenosis after MT are asymptomatic, like our case.4,6,8) The development of collateral circulation and compensatory vasodilation helps ensure sufficient blood flow.21) In our case 1, angiography indicated the development of collateral circulation from the anterior cerebral artery. For cases of delayed stenosis post-MT, standard medications for intracranial artery stenosis, which include managing blood pressure, controlling lipids, and administering antiplatelet therapy, are essential.21) Delayed stenosis following MT has been frequently reported between 3 months and 1 year post-procedure.7,8) A case has been reported where percutaneous transluminal angioplasty was performed due to symptomatic arterial stenosis after MT.9) In this case, stenosis was detected on postoperative day 77, which is somewhat earlier than that reported in other cases. This suggests that a few cases with delayed stenosis that progress rapidly can present symptoms due to insufficient development of collateral circulation. Even if complete recanalization is achieved, follow-up evaluation is essential to confirm the development of symptoms or any progression of stenosis. MRI should be performed 3 months after MT. If stenosis is detected at this time, follow-up should be continued for approximately 1 year postoperatively. Delayed stenosis detected more than 3 months post-procedure may remain asymptomatic over the course of follow-up.
We present 2 cases of delayed stenosis that developed months after successful recanalization following MT. Delayed stenosis is induced by device-related arterial damage. The use of a relatively larger and longer stent retriever in the target vessels, as well as repeated procedures with thrombectomy devices, carries a high risk of delayed stenosis. This report emphasizes the importance of regular follow-up after MT, particularly in cases where there is a potential for substantial vascular damage due to thrombectomy.
All authors have no conflicts of interest to declare.
