Manual aspiration thrombectomy using a Penumbra catheter in patients with acute migrated MCA occlusion

Abstract

Background

We retrospectively assessed the efficacy and safety of manual aspiration thrombectomy (MAT) using a Penumbra catheter in patients with acute migrated middle cerebral artery (MCA) occlusion.

Methods

We conducted a retrospective review of patients who underwent MAT using Penumbra 4 or 5 MAX reperfusion catheters for the treatment of acute MCA occlusion between January 2012 and December 2015. A migrated thrombus was defined as distal migration greater than ≥1 cm on initial cerebral angiography compared with preprocedural angiographic findings. We evaluated immediate angiographic results and clinical outcomes through a review of electronic medical records and compared clinical outcomes between migrated and non-migrated MCA occlusions.

Results

During this period, 98 patients underwent MAT using Penumbra catheters for treatment of acute MCA occlusions. Of these, 19 (19.4%) had a migrated MCA thrombus on initial cerebral angiography compared with preprocedural angiographic findings. The overall rate of successful recanalization (TICI grade ≥ 2b) was 90.8%. The rate of successful recanalization in the migrated thrombus group was 94.7% (18/19). Overall procedure time for recanalization in the migrated thrombus group was significantly shorter than that in the non-migrated thrombus group (p = 0.001). Although the difference was not significant, there was a trend toward more favorable clinical outcomes in the migrated thrombus group compared to the non-migrated thrombus groups (78.9% versus 60.8%, p = 0.231).

Conclusion

MAT appears to be safe and can achieve a high rate of successful recanalization and favorable clinical outcomes with short procedure time and without serious complications in patients with acute migrated MCA occlusion.

Introduction

Early and efficient recanalization of large-vessel occlusion improves outcomes in select patient with acute ischemic stroke.^1–3 Manual aspiration thrombectomy (MAT) using a flexible aspiration catheter is an alternative technique for treating acute ischemic stroke, particularly cases refractory to stent-based thrombectomy (SBT).^4–7 Recent studies have suggested that MAT provides a high recanalization rate and favorable functional outcomes comparable to SBT.⁸ Combined use of SBT and MAT as a first-line and rescue treatment is also effective for vessel recanalization.^8–10 In our hospital, the combination of MAT with a Penumbra reperfusion catheter (Penumbra, Alameda, CA, USA) as a first-line therapy and SBT with a Solitaire stent (Covidien/ev3, Irvine, CA, USA) as a rescue therapy has been used in select patients with acute ischemic stroke since 2012.

We have occasionally observed that thrombi migrate between initial computed tomography (CT) or magnetic resonance imaging (MRI) and preprocedural cerebral angiography. We speculated that acute migrated thrombi might be removed more easily than non-migrated thrombi by MAT using a Penumbra catheter, which might improve technical and clinical outcomes in patients with an acute thrombus occlusion.

To date, however, there have been no systematic studies describing the technical success or clinical outcomes of acute migrated thrombus treatment in large cerebral artery occlusions. We assessed the efficacy and safety of MAT using a Penumbra in patients with acute migrated middle cerebral artery (MCA) occlusion.

Methods

The protocols followed in this study were approved by the local institutional review board.

Patients

We conducted a retrospective review of patients who underwent MAT using Penumbra 4 or 5 MAX perfusion catheters for treatment of an acute MCA occlusion as a first-line endovascular treatment from January 2012 to December 2015.

All patients at our institutions were initially evaluated via CT or MRI and their angiographic status was assessed routinely by MR angiography or three-dimensional CT angiography. Inclusion criteria for endovascular treatment (EVT) were initial National Institutes of Health Stroke Scale (NIHSS) ≥4, no evidence of intracerebral hemorrhage detected on the cranial CT or MRI, obvious major vessel occlusion of anterior circulation on CT or MR angiography, infarct core volume less than one-third of the MCA territory on the diffusion-weighted imaging (DWI) or non-enhanced CT, and/or perfusion deficit more than two-thirds of the MCA territory or perfusion deficit volume on MR more than 30% larger than lesion volume on DWI. At admission, the NIHSS score was assessed by a stroke neurologist. Eligible patients who met standard National Institute of Neurologic Disorders and Stroke (NINDS) criteria were treated with 0.9 mg/kg of intravenous (IV) recombinant tissue plasminogen activator (rtPA). Subsequent endovascular therapy was considered within one hour of IV rtPA for patients with no neurological improvement, which was defined as an NIHSS score unchanged from baseline or a worsening neurological deficit.

EVT

One interventional neuroradiologist with 11 years of experience performed all EVTs. Written informed consent for cerebral angiography and EVT was obtained from all patients. All procedures were performed under local anesthesia via the percutaneous transfemoral route. Systemic anticoagulation was initiated with a 3000-U bolus of IV heparin followed by 1000-U/h infusion. A triple coaxial system assembled by combining an outermost 80-cm-long 8F sheath (Shuttle-SL; Cook, Bloomington, IN, USA), a middle 100-cm-long 8F guide catheter (Guider Softip; Stryker, Natick, MA, USA), and an inner 125-cm-long selection catheter (Headhunter; Cook, Bloomington, IN, USA), was placed in the common carotid artery and internal carotid artery for MAT. Once occlusion of the MCA was verified by cerebral angiography, a double-coaxial system assembled by combining an outer Penumbra catheter and an inner Rebar 18 microcatheter (Covidien/ev3, Irvine, CA, USA) was advanced to the level of the thrombus using a 0.014-inch micro-guidewire (Synchro; Stryker Neurovascular, Freemont, CA, USA). We gently advanced the Penumbra catheter into the thrombus until wedged tightly. Subsequently, the microcatheter and microguidewire were removed and a 20-ml syringe was connected to the proximal hub of the Penumbra catheter. Continuous manual aspiration was performed while maintaining the vacuum state between the tip of the Penumbra catheter and the thrombus while gently withdrawing the Penumbra catheter through the guide catheter. If angiographic recanalization was not noted after five trials, additional SBT was performed with a Solitaire stent. The details of the technique for SBT with a Solitaire stent have been described previously.¹¹

Image analysis

We defined a migrated thrombus as distal migration above ≥ 1 cm on initial evaluation with CT or MRI and angiographic findings compared with preprocedural angiographic findings. Location of the thrombus was considered the site of abrupt cut-off of the MCA on initial cerebral angiography and preprocedural angiography. One experienced neuroradiologist assessed differences in the occlusion site by comparing initial angiography with preprocedural angiography. We classified eligible patients into two groups: migrated thrombus versus non-migrated thrombus according to the presence of thrombus migration.

Outcome measures

For all patients, we retrospectively analyzed medical records to collect patient demographic, clinical, and angiographic data. We compared demographic and angiographic data and clinical outcomes between patients with migrated and non-migrated MCA occlusions. The NIHSS and modified Rankin scale (mRS) were checked on admission and at three months for all patients by a neurologist, and favorable outcome was defined as an mRS of ≤2. The degree of vessel occlusion after treatment was defined by the thrombolysis in cerebral infarction (TICI) scale and successful recanalization after treatment was defined as TICI scale ≥2b. Procedure duration was defined as the time from groin access to at least TICI 2b revascularization. Procedure-related symptomatic intracerebral hemorrhage (ICH) was defined as the presence of hemorrhage after treatment with any point of increasing NIHSS compared to NIHSS on admission. Mortality was defined as death occurring within 90 days from initial presentation.

Statistical analysis

Statistical analyses were performed with SPSS software V.20.0 (SPSS, Chicago, IL, USA). Differences between the two groups were tested using the Mann-Whitney test for continuous measures and χ² test for categorical measures. All tests were two sided and assessed at a significance level of 0.05.

Results

Patient demographics and preprocedural findings

From January 2012 to December 2015, 114 consecutive patients with acute ischemic stroke caused by MCA occlusion were treated with MAT using a Penumbra reperfusion catheter as first-line therapy. Of these, 16 were excluded for the following reasons: low-density area affecting more than two-thirds of the MCA territory (N = 8), disability (mRS ≥ 3) by previous infarction at admission (N = 4), technical failure caused by tortuous course of aorta, and/or common carotid artery (N = 2), and loss to follow-up and did not receive a clinical evaluation at three months (N = 2). Thus, 98 patients who underwent MAT using Penumbra as a first-line therapy were included in this study. Of these patients, 19 (19.4%) had a migrated MCA thrombus on initial cerebral angiography compared with preprocedural angiographic findings and the other 79 (80.6%) patients had no thrombus migration.

Baseline clinical characteristics and preprocedural findings between the two groups are summarized in Table 1. Patients were more often female (58.2%) and the median age was 73 years (age range, 18–92 years). Seventeen patients (17.3%) had a medical history of previous stroke and 36 (36.7%) had previous cardiac disease including atrial fibrillation. The time from symptom onset to admission ranged from 32–326 minutes (median, 123 minutes), and the time from admission to femoral puncture ranged from 36–303 minutes (median, 120 minutes). Median baseline NIHSS score was 11 (range, 2–29). Fifty patients (51%) received IV rtPA before angiography. There were no significant differences in patient demographics and preprocedural findings between the two groups except for the incidence of diabetes mellitus (42.9%, p = 0.014) and hypercholesterolemia (17.7%, p = 0.038).

Table 1.

Summary of patient demographics and preprocedural findings.

Patient demographics	All patients (n = 98)	Migrated thrombus (n = 19)	Non-migrated thrombus (n = 79)	p
Age, years, median (IQR)	73 (65–79)	68 (67–76)	73 (62–80)	0.589
Male, (%)	41 (41.8)	6 (31.6)	35 (44.3)	0.228
Right side, (%)	47 (48.0)	11 (57.9)	36 (45.6)	0.202
Stroke risk factors (%)
Diabetes mellitus	20 (20.4)	8 (42.9)	12 (15.2)	0.014
Hypertension	48 (50.0)	10 (57.1)	38 (48.1)	0.460
Hypercholesterolemia	14 (14.3)	0 (0.0)	14 (17.7)	0.038
Smoking	26 (26.5)	4 (28.6)	22 (17.8)	0.388
Previous stroke	17 (17.3)	3 (14.3)	14 (17.7)	0.572
Previous cardiac disease including AF	36 (36.7)	4 (21.4)	32 (40.5)	0.092
Preprocedural findings
IV rtPA use, (%)	50 (51.0)	13 (68.4)	37 (46.8)	0.075
Onset to door time, min, median (IQR)	123 (60–181)	108 (60–197)	123 (70–208)	0.243
Door to needle time, min, median (IQR)	55 (50–65)	55 (55–65)	55 (55–65)	0.942
Initial NIHSS, median (IQR)	11 (8–15)	12 (8–13)	11 (6–15)	0.708
Initial mRS, median (IQR)	4 (4–4)	4 (4–4)	4(4–4)	0.665

IQR: interquartile range; AF: atrial fibrillation; NIHSS: National Institutes of Health Stroke Scale; IV: intravenous; rtPA: recombinant tissue plasminogen activator; mRS: modified Rankin Scale.

Periprocedural findings, complications, and clinical outcomes

Table 2 summarizes periprocedural findings, complications, and clinical outcomes. The median procedure time was 30 minutes (range, 15–74 minutes). The procedure time and door to recanalization time in the migrated thrombus groups was significantly shorter than that of the non-migrated thrombus group. The rate of successful recanalization (TICI grade ≥2b) was 90.8% (89/98). Both the migrated thrombus group and non-migrated thrombus group had a high successful recanalization rate (94.7% versus 89.9%; Figure 1). The successful recanalization rate of a single Penumbra was 83.6% (82/98). Seven patients (7.1%) were treated in combination with a Solitaire stent as a rescue therapy. All of the seven patients using the Solitaire stent (migrated thrombus; n = 1, non-migrated thrombus; n = 6), showed successful recanalization. Three patients had a tandem occlusion in the anterior cerebral artery. Of these, two were treated by MAT using a Penumbra catheter and achieved TICI grade 2b recanalization of the anterior cerebral artery, and one was treated by mechanical thrombectomy using a microcatheter and urokinase infusion, but did not achieve recanalization. After the endovascular procedure, 66 of 98 patients (67.3%) were classified as TICI grade 3, 23 patients (25.8%) as TICI grade 2b, and nine patients (10.1%) as TICI grade 2a.

Table 2.

Summary of periprocedural findings, complications, and clinical outcomes after three months.

Patient demographics	All patients (n = 98)	Migrated thrombus (n = 19)	Non-migrated thrombus (n = 79)	p
Periprocedural findings
Procedure time, min, median (IQR)	30 (25–45)	25 (22–30)	35 (26–50)	0.001
Door to revascularization time, min, median (IQR)	90 (80–101)	85 (80–90)	95 (80–106)	0.007
Onset to revascularization time, min, median (IQR)	216 (161–298)	190 (145–285)	221 (165–314)	0.103
Discharge NIHSS, median (IQR)	4 (2–7)	4 (3–7)	4 (2–7)	0.278
Rescue therapy, n (%)	7 (7.1)	1 (5.3)	6 (7.6)	0.888
mTICI, 2b or 3 (%)	89 (90.8)	18 (94.7)	71 (89.9)	0.843
≤2a	9 (10.1)	1 (5.3)	8 (10.1)	0.843
2b	23 (25.8)	4 (21.0)	14 (17.7)	0.998
3	66 (67.3)	14 (73.7)	57 (72.2)	0.883
Complications, n (%)
Procedure-related SAH	4 (4.1)	2 (10.5)	2 (2.5)	0.352
ICH	9 (9.2)	2 (10.5)	7 (8.9)	0.828
Symptomatic ICH	4 (4.1)	0	4 (5.1)	0.720
Hemorrhage events	17 (17.3)	4 (21.1)	13 (16.5)	0.893
Favorable outcome (mRS 0–2)	64 (65.3)	15 (78.9)	48 (60.8)	0.231
Mortality	4 (4.1)	0	4 (5.1)	0.720

IQR: interquartile range; NIHSS: National Institutes of Health Stroke Scale; mTICI: modified thrombolysis in cerebral infarction; SAH: subarachnoid hemorrhage; ICH: intracerebral hemorrhage; mRS: modified Rankin Scale.

Figure 1.

A 72-year-old man with a migrated thrombus in the left middle cerebral artery (MCA) territory and score 1 on the modified Rankin Scale after three months. (a) Computed tomography angiography shows complete occlusion of the proximal MCA (arrow). (b) An initial angiogram shows occlusion of the left M2 inferior division (arrow) and recanalization of the M1 segment. (c) Positioning of the Penumbra 4MAX catheter in relation to the clot (arrow). (d) A final lateral angiogram after aspiration thrombectomy shows complete recanalization (thrombolysis in cerebral infarction (TICI) 3).

Seventeen patients (17.3%) experienced hemorrhage events after EVT during their hospital period. Among these, procedure-related subarachnoid hemorrhage was observed in four patients, and all showed complete absorption of the hemorrhage at discharge. Four patients showed symptomatic ICH and clinical deterioration, leading to death during hospitalization due to massive symptomatic hemorrhage, brain edema, and herniation. Thus, the overall mortality rate was 4.1%. There were no significant differences in periprocedural findings and complications between the two groups.

At discharge, the median NIHSS score of the patients who did not die was 4. Three months later, overall favorable clinical outcomes (mRS score at three months ≤2) were seen in 64 patients (65.3%). Although the difference did not reach statistical significance, there was a trend toward more favorable clinical outcomes in the migrated thrombus group compared to the non-migrated thrombus group (78.9% versus 60.8%, p = 0.231). Seven patients who had the Solitaire stent showed favorable clinical outcomes in three patients (42.8%).

Discussion

To the best of our knowledge, this is the first report describing the safety and outcomes of MAT as a first-line endovascular therapy for acute migrated thrombus occlusion. No previous studies have compared outcomes in patients according to whether they had a migrated thrombus or not with MAT using a Penumbra for acute MCA occlusion. In our study, successful recanalization (TICI grade 2b or 3) with MAT was achieved in 89 of 98 (90.8%) patients. Both groups had a high recanalization rate (94.7% versus 89.9%). The procedure time for recanalization in the migrated thrombus groups was significantly shorter than that of the non-migrated thrombus group. However, the migrated thrombus group showed a trend toward more favorable clinical outcome (mRS 0–2) than the non-migrated group, although the difference was not statistically significant (78.9% versus 60.8% p = 0.231). The shorter procedure time and favorable clinical outcome in the migrated thrombus group may result from several factors. First, migrated thrombi may be softer and smaller than those that do not migrate, making them easier to extract and aspirate by MAT as well as potentially more responsive to rtPA. Second, although both groups achieved recanalization in 25 minutes on average after groin puncture, the door to revascularization time was considerable faster in the migrated thrombus group. The substantially better outcome (mRS 0–2) in the migrated thrombus group may be explained in part by this rapid recanalization. The shorter procedure time in the migrated thrombus group compared to the non-migrated thrombus group also may have explained that additional rescue stent is less needed in the migrated group (migrated group versus non-migrated group; 5.3% versus 7.6%).

Fifty eligible patients (51%) who met NINDS criteria received IV rtPA before EVT. Although there was no statistical difference, the migrated thrombus group showed higher IV rtPA use than the non-migrated thrombus group (68.4% versus 46.8%). Although there is no exact patient’s medical record, some patients underwent MAT directly while rtPA was still running because of an unchanged initial NIHSS score or a worsening neurological symptom. After one-hour infusion and closed monitoring of rtPA was completed, however, some other patients underwent MAT. The differences in the rate of rtPA use and the time of rtPA action may have had an impact on the distal migration of thrombus in both groups.

Among the risk factors from patient demographics in both groups, the rate of diabetes mellitus was much higher in the migrated thrombus group (42.9%, p = 0.014), but there were more patients with hypercholesterolemia in the non-migrated thrombus group. Diabetes mellitus and hypercholesterolemia are well-recognized risk factors of atherosclerosis, which results in vascular wall injury rather than circulatory stasis or a hypercoagulable state.^16–18 Nonetheless, diabetes mellitus and hypercholesterolemia showed opposite results in the present study. In addition, atrial fibrillation, a well-known cause of embolic stroke and more likely to be associated with migrated thrombi, occurred more often in the non-migrated thrombus group than the migrated thrombus group.

Recent randomized trials have shown improved outcomes of endovascular therapy using stent retrievers.^4,12–14 The MAT technique we used shows promising results in recanalization rate, favorable functional outcome, and safety compared with the most recent outcomes of stent-retriever thrombectomy.^4–8,10 In the recently reported Solitaire with the Intention for Thrombectomy as Primary Endovascular Treatment (SWIFT PRIME) trial,⁴ TICI 2b–3 recanalization was achieved with SBT in 88% of cases, and 60% had good functional outcomes (mRS 0–2). Kang et al.¹⁰ reported a series of 74 patients with acute ischemic stroke who were treated with a combination of MAT and SBT, resulting in successful primary recanalization (TICI 2b or 3) rate of 53% (39/74). Jankowitz et al.⁷ reported a 59% (66/112) recanalization rate in a retrospective examination of 112 stroke cases with MAT as the first-line therapy and an adjunctive device such as the Merci retriever, Solitaire or Trevo as rescue therapy.

Recently introduced large-bore aspiration catheters have yielded improved recanalization rates in MAT. In a study by Turk et al.,⁹ MAT with a large-bore aspiration catheter as a first-line endovascular therapy achieved TICI 2b or 3 recanalization in 78% of cases. The additional use of SBT improved the TICI 2b/3 recanalization rate to 95%.We confirmed the efficacy of the newest generation of large-bore aspiration catheters and extended their application to acute MCA occlusion. The MAT-only successful recanalization (TICI 2b-3) rate was high in our study (83.6%) and greater than that observed by Turk et al.⁹ (78%). Additional use of SBT in our study also showed an improved recanalization rate up to 90.8%. The frequency of good outcomes (mRS 0–2) was 64% in our study, which was greater than that of Turk et al.⁹ (40%) and similar to that observed in the SWIFT PRIME trial⁴ (60%). In our study, the mortality was 4.1% (4/89), which is among the lowest reported in the literature.

Vascular tortuosity is a major cause of difficulty or failure treating intracranial lesions in endovascular therapy of acute ischemic stroke. Thus, a proper catheter system for accessing the target lesion and providing stable support is essential for successful mechanical thrombectomy. Multiple coaxial catheter systems are safe and efficient for mechanical thrombectomy in acute ischemic stroke.¹⁵ Therefore, the selection of the appropriate catheter and wire for MAT is at the operator’s discretion. In our study, the combination of an 80-cm-long 8F (0.113 inner diameter) sheath and 100-cm-long 8F guide catheter (0.106 outer diameter and 0.086 inner diameter) provided appropriate proximal mechanical support from the groin to the level of the common carotid or internal carotid artery. In addition to the mechanical support of this system, the 8F guide catheter provided easy access to the proximal part of various Penumbra catheters, which is larger than the distal part. Moreover, the 8F guide catheter also allowed access for stent retrieval when the MAT technique failed. The combination of a Rebar 18 microcatheter and Synchro 0.014-inch wire was effective for advancing more distally through the thrombus up to distal M2 of the MCA.

Our study had several limitations. First, this study was based on a single-center experience, and therefore lacks strong power to confirm the efficacy and safety of MAT. Second, the total number of patients in the migrated thrombus group was relatively small, preventing reliable comparisons of the subgroup. Third, absence of a core laboratory to adjudicate angiographic results confers a bias to better angiographic outcomes than may truly be present. Fourth, we did not control the effects of IV rtPA on thrombus during MAT in both groups. Fifth, our study did not include histology analysis of the aspirated thrombi or whether patients were taking anticoagulant or antiplatelet agents, which undermines the validity of the comparison for variables between the two groups and decreases confidence in the assumptions made regarding study results. As such, further study with a larger patient group and histological analysis should be performed to produce reliable data.

Conclusion

MAT using a Penumbra catheter appears to be safe and can achieve a higher rate of successful recanalization and more favorable clinical outcomes with short procedure time and without serious complications in patients with acute migrated MCA occlusion than non-migrated MCA occlusion. Evidence suggests that MAT using a Penumbra catheter provides more favorable clinical outcomes in acute migrated MCA occlusion.

Declaration of conflicting interests

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

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This paper was supported by research funds from Chonbuk National University in 2017.