Endovascular treatment of acute ischemic stroke in patients with recurrent intracranial large vessel occlusion

Abstract

Background

Data on the treatment with recurrent mechanical thrombectomy of patients with acute ischemic stroke with recurrent large vessel occlusion are limited. We report our experience with recurrent mechanical thrombectomy for recurrent large vessel occlusion.

Methods

During the period between May 2013 and August 2018, data on patients with recurrent large vessel occlusion were collected. Baseline clinical characteristics, recanalization technique, recanalization rates and clinical outcomes of patients with recurrent large vessel occlusion treated with mechanical thrombectomy were analyzed. Patients with recurrent large vessel occlusion treated with mechanical thrombectomy were compared with patients who underwent single mechanical thrombectomy.

Results

During the study period, 7 of 474 patients (1.5%) were treated with mechanical thrombectomy for recurrent large vessel occlusion. The mean age of these patients was 64.4 (±7.9) years, and the mean time interval between thrombectomies was 47 (±48) h. The median baseline National Institutes of Health Stroke Scale (NIHSS) was 12 (range 5–24) before the first and 20 (range 3–34) before the second procedure; the mean NIHSS at discharge was 5 (range 2–25). Good clinical outcome after repeated mechanical thrombectomy defined as modified Rankin scale of 0–2 was achieved in 29% of patients at three months of follow-up.

Conclusions

Repeat mechanical thrombectomy is a rare procedure, but appears to be a feasible, safe and effective treatment option in patients with acute ischemic stroke and early recurrent large vessel occlusion.

Introduction

In recent years, endovascular treatment with mechanical thrombectomy in combination with intravenous thrombolysis has become the gold standard for patients with acute ischemic stroke (AIS) and large vessel occlusion (LVO) in the anterior circulation.¹ In rare cases (between 3% and 9%), re-occlusion of the target vessel can occur within 48 h of successful mechanical thrombectomy (MT) and is associated with poor outcomes.^2,3 More generally, of patients who have survived an initial stroke, the cumulative risk of stroke recurrence is near 25% within five years.⁴ Although repeated thrombolytic therapy can be safe and effective for selected patients with recurrent AIS, treatment risks include intracranial hemorrhage and are contraindicated in patients with a prior stroke within the last three months. While MT has become a standard-of-care for treating LVOs, treatment options for patients with recurrent stroke remain limited, as repeated MT for recurrent LVO is rare. Data regarding the outcome of patients treated with repeated MT for recurrent LVO are also limited, and only a few series and several case reports have been published to date.^5–14 These reports vary in the interval of re-thrombectomy, from early recurrence (<1 day) up to over a year.^5,9,11–14

In the present study, we report our experience with the endovascular treatment of patients with recurrent LVO, including three cases of target vessel re-occlusion. Our goal was to investigate the clinical and radiological outcomes of patients treated with re-thrombectomy.

Methods

We retrospectively reviewed data from all mechanical thrombectomies for AIS with intracranial LVO performed in a stroke center from May 2013 to August 2018. Selected patients included those with recurrent LVO in the anterior or posterior circulation treated with repeated MT. Recurrent LVO was defined as a recurrent thromboembolic occlusion following the initial procedure confirmed via CT angiography in patients with documented clinical worsening after first thrombectomy. Patients in which the recurrent embolism occluded a different site were not excluded.

Stroke severity was scored using the National Institute of Health Stroke Scale (NIHSS) upon admission, following recurrent stroke and at discharge. All patients received non-enhanced computed tomography (CT) and CT angiography imaging before the first or recurrent MT.

All patients were treated under local anesthesia or conscious sedation. After the procedure, control CT or magnetic resonance imaging (MRI) was performed 24 h after MT. Patient demographics, cardiovascular risk factors (hypertension, hyperlipidemia, diabetes mellitus, coronary artery disease), values of hemoglobin, fibrinogen, platelets, glycemia, and systolic and diastolic blood pressure were collected before the initial procedure. Medications administered prior to the first stroke, including antiplatelets or anticoagulants, were reviewed. Stroke etiology was classified using the TOAST (Trial of Org 10172 in Acute Stroke Treatment) classification. All endovascular procedures were performed by at least one experienced interventional radiologist.

Data on endovascular procedures including artery occlusion sites, endovascular treatment methods, number of passes, angiographic runs during endovascular procedures, time interval between the procedures, onset-to-door time, periprocedural complications, and patient outcome after 90 days of follow-up were analyzed. Successful endovascular recanalization was defined as Thrombolysis in Cerebral Infarction (TICI) score of 2b or 3. Good clinical outcome was defined using the modified Rankin Scale (mRS) 0–2 at 90 days. All angiograms of patients with recurrent LVO were evaluated for irregularities, vasospasms and intimal lesions. Patients treated for recurrent LVO were compared to a group of all patients consecutively treated with a single MT from December 2015 to December 2016. The study was approved by the local institution ethics committee.

Statistical analysis

Continuous variables were reported as mean ± SD or median (IQR), as appropriate; categorical variables were reported as proportions. P-values were calculated using Fisher's exact test for categorical variables, the Mann–Whitney U test (Wilcoxon rank-sum test) for non-normally distributed continuous variables and Welch's t-test for normally distributed continuous variables.

Results

During the study period, 7 (3 females and 4 males) of 474 patients underwent repeat MT for recurrent LVO (1.5%). Mean patient age was 64.4 (±7.9) years. LVO locations for both first and second stokes included the internal carotid artery (ICA), the middle cerebral artery (MCA) and the basilar artery (BA). Of the seven patients with recurrent LVO, six (86%) patients had recurrent LVO in the same vascular territory, while one patient had recurrent LVO in another vascular territory (Figure 1, Table 1). None of the patients had more than one recurrent LVO.

Figure 1.

Digital subtraction angiography of a 52-year-old patient with basilar artery occlusion before and after first thrombectomy (a,b) and with recurrent intracranial large vessel occlusion in left M2 MCA 30 h after first procedure before and after endovascular treatment (c,d).

Table 1.

Demographics, site of vessel occlusion in the first and recurrent LVO, the NIHSS before and after the first and second procedure, TICI recanalization grade after the first and subsequent thrombectomy, time interval between the procedures and patient outcome at 90 days.

Case	Age	Pre stroke mRs	Sex	TOAST	NIHSS before the first MT	Vessel occlusion in the first stroke	TICI after the first MT	NIHSS after the first MT	Time between MT (h)	NIHSS before the second MT	Vessel occlusion in the second stroke	MT technique	TICI after the second MT	NIHSS after the second MT	mRS 90 days
1	62	0	Male	CE	5	BA	3	3	9	15	BA	AT	3	3	2
2	61	0	Male	OE	18	ICA and M1 MCA	3	12	6	25	Carotid T	AT+SR	0	25	6
3	71	0	Female	ATS	11	BA	3	6	6	34	BA	AT	3	5	3
4	71	4	Female	CE	22	M2 MCA	3	12	139	20	M1 MCA	AT	3	7	4
5	52	5	Male	CE	9	BA	3	7	30	14	M2 MCA	AT	3	4	5
6	58	0	Female	CE	12	M1 MCA	3	0	4	3	M2 MCA	AT	3	2	0
7	76	0	Male	CE	24	M1 MCA	3	6	48	21	M1 MCA	AT	3	21	6

NIHSS: National Institutes of Health Stroke Scale; MT: mechanical thrombectomy; AT: aspiration thrombectomy; SR: stent-retriever; MCA: middle cerebral artery; BA: basilar artery; mRS: modified Rankin scale; ICA: internal carotid artery; TICI: thrombolysis in cerebral infarction; CE: cardioembolic; OE: other determined etiology; ATS: atherosclerosis; TOAST: trial of org 10172 in acute stroke treatment.

The mean time between repeat thrombectomies in our study group was 47 ± 48 (range 4–139) h. The median baseline National Institutes of Health Stroke Scale (NIHSS) was 12 (range 5–24) before the first procedure, 6 (range 3–12) after the first procedure, 20 (range 3–34) before the second procedure, and 5 (range 2–25) at discharge following the second MT.

Mean onset-to-door time in patients with recurrent LVO treated endovascularly was 242 ± 184 (range 60–605) min. Successful TICI 2b/3 recanalization after the first procedure was achieved in 100% of patients, and in 86% of patients treated with repeat MT after recurrent LVO. None of the patients with recurrent LVO had tandem occlusions, and no patient received an extracranially or intracranially implanted stent during or after MT (Table 2). No vessel wall irregularities, spasms or intimal flaps on angiograms were found after evaluation.

Table 2.

Comparison between patients treated with repeated thrombectomy for recurrent LVO and patients who underwent single thrombectomy.

Parameter	Recurrent LVO MT (n = 7 pts)	Single MT (n = 99 pts)	P value
Age	64.4 ± 7.9	69.9 ± 12.4	0.116
Sex, male	57%	50%	1.000
Hypertension	57%	79%	0.190
Hyperlipidemia	43%	60%	0.446
Diabetes mellitus	29%	30%	1.000
CAD	14%	47%	0.127
Hemoglobin, g/l	119.4 ± 18.7	121.1 ± 18.0	0.567
Previous stroke/TIA	29%	12%	0.238
Fibrinogen, g/l	2.69 ± 0.46	2.93 ± 1.04	0.400
Glycemia, mmol/l	8.2 ± 2.1	8.2 ± 3.0	0.445
Systolic BP before procedure, mm Hg	141.3 ± 17.4	151.6 ± 21.3	0.176
Diastolic BP before procedure, mm Hg	76.3 ± 6.4	82.0 ± 11.4	0.159
Platelets, 109/l	284.7 ± 189.9	228.7 ± 67.2	0.499
TOAST (atherosclerosis)	14%	37%	0.417
TOAST (cardioembolic)	71%	54%	0.452
Thrombolysis with tPA	71%	76%	1.000
Admission NIHSS	12	15	0.934
Referral transfer	57%	41%	0.454
Anterior circulation	57%	89%	0.047
Posterior circulation	43%	11%	0.047
Onset-to-door time (min)	242 ± 184	71 ± 61.7	0.005
Previous medication before first stroke
Acetylsalicylic acid	29%	29%	1.000
Clopidogrel	29%	10%	0.179
DAPT	14%	5%	0.343
Warfarin	0%	6%	1.000
NOAC	0%	4%	1.000
Statin	29%	22%	0.665
Endovascular procedure
More than one pass	33%	20%	1.000
Aspiration thrombectomy	86%	84%	0.591
Combined treatment	14%	15%	0.590
Stent-retriever	0%	1%	1.000
TICI 2b-3	100%	84%	0.591
mRS 0–2	29%	32%	1.000
mRS 3–6	71%	68%	1.000

LVO: large vessel occlusion; TICI: thrombolysis in cerebral infarction; DAPT: dual antiplatelet therapy; mRS: modified Rankin scale; NOAC: novel oral anticoagulants; BP: blood pressure; CAD: coronary artery disease; TIA: transient ischemic attack; NIHSS: National Institutes of Health Stroke Scale; MT: mechanical thrombectomy; TOAST: trial of org 10172 in acute stroke treatment.

No access site complications were observed following the endovascular procedure, nor were re-occlusions observed during any procedure. One patient (14.3%) developed parenchymal hematoma type 2 after repeat MT. Good clinical outcomes (mRS 0–2) after re-thrombectomy were achieved in 28.6% of patients, while 71.4% of patients achieved mRS 3–6. Aspiration thrombectomy alone for recurrent LVO was performed in 85.7% of patients and combination treatment with aspiration thrombectomy and stent-retriever was performed in 14.3% of patients (Table 2).

The etiology of recurrent LVO was cardioembolic in 71.4%, followed by atherosclerosis in 14.3%, and stroke of other determined etiology (vasculitis) in 14.3% of patients using the TOAST classification. None of the patients with recurrent LVO received oral anticoagulation at the time of first stroke. Patient demographics, pre stroke mRS, site of vessel occlusion in the first and recurrent LVO, the NIHSS before and after the first and second procedure, TOAST classification, TICI recanalization grade after the first and recurrent thrombectomy, time interval between the procedures, endovascular technique and patient outcome at 90 days are summarized in Table 1.

Comparisons between patients treated with repeat MT for recurrent LVO and patients with single thrombectomy are shown in Table 2. Patients who underwent repeat MT had similar outcomes to those treated with single thrombectomy; however, patients with a posterior circulation stroke were associated with a higher probability of repeat MT (P < 0.05) (Table 2), and patients with a longer onset-to-door time were also associated with a higher probability of repeat MT (P < 0.001; Table 2).

Discussion

In this consecutive case series, we report on the safety and efficacy of repeat MT, a rarely used procedure for recurrent LVO. All but one patient treated with re-thrombectomy in this series achieved reperfusion rates of TICI 2b-3, and no vessel denudation or perforation was observed. Periprocedural complication rates were low (Table 1) and only one patient developed a type 2 parenchymal hematoma. When compared with patients undergoing single MT (Table 2), the number of patients achieving a functional outcome of mRS 0–2 after repeated MT was not significantly different.

In 2015, several randomized controlled trials showed the clear clinical benefit of MT over medical therapy alone, in the treatment of patients with AIS and LVO in the proximal anterior circulation.¹ However, data regarding the incidence and outcome of endovascular treatment in patients with AIS and recurrent LVO are scarce. To the best of our knowledge, there are five series of 15, 23, 25, 27 and 35 patients (Table 3) and several case reports regarding endovascular treatment of recurrent LVO.^5–14 In the available series to date, Bouslama et al.⁵ found that the incidence of recurrent LVO was 2% with the median time interval between the first and last procedure of recurrent LVO being 18 days. Similarly, Weber et al.¹¹ found the incidence of recurrent LVO as 1.4%; however, the mean time interval between the first and second MT was 147 days. Ikenberg et al.¹³ had 1.5% incidence, found 5 patients (21.7%) with recurrent thrombectomy within 24 h and 14 patients (60.9%) within two weeks. Mosimann et al. noted that the incidence of early reocclusion after 48 h was 2.3% in patients after successful thrombectomy and Marto et al. found that the incidence of early reocclusion after 24 h was 6.6%.^2,3 In the study of Bhogal et al.,¹² only four patients had recurrent stroke in the first 7 days with the median time interval between strokes being 71 days. Pirson et al.¹⁴ found in their large nationwide cohort of patients that 0.7% of patients underwent recurrent thrombectomy with 78 days median time interval between thrombectomies. In our single center case series, the incidence of recurrent LVO treated with recurrent thrombectomy was 1.5%; however, all of them occurring in the early period between 4 and 139 h follow the initial procedure.

Table 3.

List of studies with repeat thrombectomy for recurrent LVO.

Study	Number of patients	TOAST cardioembolic	Number of patients with interval between thrombectomies within two days	mRS 0–2 at three months
Bouslama et al.5	15	66%	4 (27%)	9 (60%)
Weber et al.11	35	43%	7 (20%)	NA (49% mRS 0–1 at discharge)
Bhogal et al.12	25	86.5%	2 (8%)	9 (36%)
Ikenberg et al.13	23	52%	NA(5 within 24 h)	NA
Pirson et al.14	27	67%	2 (7%)	12 (44%)

TOAST: trial of org 10172 in acute stroke treatment; mRS: modified Rankin scale; NA: not available.

Two main endovascular strategies during endovascular treatment of AIS with LVO are used. The first option is MT with stent-retrievers, and the other is revascularization by aspiration thrombectomy. The optimum first-line endovascular thrombectomy technique for LVO is still a matter of debate. In a recent systematic review and meta-analysis of comparative studies on stent retriever-first and aspiration-first thrombectomy, both endovascular strategies were associated with comparable final reperfusion rates and functional outcomes. Stent-retriever thrombectomy was superior in achieving reperfusion as a stand-alone first-line technique, with lower use of rescue devices but a longer groin-to-reperfusion time.¹⁵ Bouslama et al.⁵ used stent-retriever alone for recurrent thrombectomy in 9 of 15 patients, thromboaspiration alone in 2 patients and combined stent-retriever and thromboaspiration in 4 patients. Bhogal et al. used stent-retrievers for recurrent thrombectomy in all patients and Chiti et al. treated patients with recurrent basilar artery occlusion first with mechanical and pharmacological thrombolysis, and then with stenting.⁶ Meza et al. and Fandler et al. utilized the stent-retrievers in the first and in repeat endovascular procedures.^7,10 In the study of Ikenberg et al.,¹³ Solumbra technique was utilized in 73.9% of repeat thrombectomies. In our study, aspiration thrombectomy alone was performed in six patients for recurrent thrombectomy (86%) with TICI 3 recanalization and combination treatment with aspiration thrombectomy and stent-retriever was performed in one patient (14%). To the best of our knowledge, this is the largest series of patients treated with aspiration thrombectomy for recurrent LVO.

Data regarding patient outcome after repeat endovascular thrombectomy are limited. Bouslama et al.⁵ found no difference in outcome between patients treated with repeat thrombectomy and those treated with single thrombectomy. In contrast, Mosimann et al.³ showed that patients with early reocclusion without repeat thrombectomy had a worse clinical outcome compared to patients with sustained recanalization after first thrombectomy, and that three patients with early reocclusion treated with repeat thrombectomy showed no clinical improvement. Weber et al.¹¹ found that 49% of patients had a NIHSS score 0 or 1 at discharge after repeat MT. Bhogal et al.¹² found nine patients (36%) with good clinical outcome mRS 0–2 at 90 days after repeat thrombectomy. Similarly, Pirson et al.¹⁴ found 44% of patients with good clinical outcome mRS 0–2. In our series, despite the small number of patients, 29% of patients treated with repeat MT for early recurrent LVO in the early period between 4 and 139 h after initial procedure had good clinical outcome after three months, comparable to outcomes from our group of single MT patients.

Intravenous thrombolysis (IVT) is now contraindicated in patients with stroke within the previous three months, which is mainly based on the hypothesis of increased bleeding risk. However, in the largest case series on repeat IVT in early recurrent stroke within the previous three months, 19 of 7537 patients received repeat IVT for early recurrent stroke.¹⁶ In another series of repeat thrombolysis for ischemic stroke, 7 of 437 patients (1.6%) received repeat thrombolysis.¹⁷ Theoretically, MT can lead to vessel wall damage with an increased risk of reocclusion or disruption of the hematoencephalic barrier with the risk of hemorrhage. Both endovascular strategies of aspiration thrombectomy and stent-retriever led to acute damage to vessel walls in an analysis of damage to swine extracranial arterial walls.¹⁸ In our case series, five of seven patients treated with repeat MT (72%) received IVT prior to first endovascular therapy and none of the patients received repeat IVT after recurrent LVO before repeat MT, and all our patients received MT and one symptomatic intracranial hemorrhage after repeat MT occurred.

The prognosis, outcome and management of patients with AIS are influenced by the etiology of ischemic stroke. The TOAST classification denotes five subtypes of ischemic stroke.¹⁹ In the study of Weber et al., a cardioembolic cause was found to cause the initial and recurrent LVO in 43% of patients; Mosimann et al. found cardioembolic origin in 18.8% of patients with early reocclusion.^3,11 Bhogal et al.¹² found that 86.5% of the strokes were deemed as being cardioembolic in nature. Ikenberg et al.¹³ reported 69.6% of patients, where the etiology of the first stroke and its recurrence was considered as likely being the same, mostly of cardioembolic or unknown origin. Similarly Pirson et al.¹⁴ found that 67% of patients receiving recurrent MT had cardioembolic source of stroke. In our series, five out of seven of patients with recurrent LVO had cardioembolic cause of stroke occurring in the early period less than two days after first stroke. In a clinical point of view, in patients suffering from atrial fibrillation, the early effective anticoagulation treatment could be the prevention of this early recurrent symptomatic LVO. However, although the risk of early recurrent ischemic stroke is high in this population, early oral anticoagulation is suspected to increase the risk of potentially harmful intracranial hemorrhage, including hemorrhagic transformation of the infarct and deciding when to initiate oral anticoagulation in patients with non-valvular atrial fibrillation is a longstanding, common, and unresolved clinical challenge.²⁰

Predictors influencing potential early reocclusion in patients with AIS are not well known. Mosimann et al.³ identified four factors associated with early reocclusion after 48 h: elevated platelet count on admission >220 g/L, missed residual thrombus or stenosis on angiography after thrombectomy, M2 occlusion as the initial occlusion site and stroke of undetermined cause. In the study of Marto et al.,² reocclusion within 24 h of successful MT was independently associated with statin pretreatment, occlusion site, more complex procedures, atherosclerotic cause, and residual thrombus or stenosis after recanalization. In our series, the group of patients with recurrent LVO was too small to predict independent factors associated with recurrent LVO.

There are several limitations in the current investigation. First, the very small number of patients and retrospective design of the investigation are clear limitations. Second, the number of patients with recurrent LVO might have been higher, as we did not routinely perform control CTA or MRA studies in all patients after MT during the study period, and the recurrent endovascular procedure was performed in patients with documented clinical worsening and CT angiography confirmed the recurrent LVO consistent with the new onset of symptoms. Third, there is a potential selection bias due to selecting the control group of patients receiving single MT not during the entire study period but only during one year of the study.

Conclusion

The present study showed that repeated MT appears to be a safe and effective treatment option for early recurrent LVO and should not be withhold according to the current data. Data from patients who underwent repeated thrombectomy showed good clinical outcomes that were comparable to single thrombectomy outcomes. Endovascular therapy, including aspiration-only procedures, should be considered in patients with AIS and recurrent LVO after previous MT. Due to the rarity of this procedure, larger studies or registries collecting data from repeated endovascular procedures following recurrent LVO are necessary.

Authors' contributions

AK and JH designed the model of the study. MM, MR, JC carried out the implementation and performed the statistical analysis. AK, MH, JH wrote the manuscript with input from all authors. AK and GK were in charge of overall direction and planning.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval

Written informed patient consent for publication of this case series has been obtained.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Andrej Klepanec https://orcid.org/0000-0002-7460-2974