Mechanical thrombectomy in pediatric acute ischemic stroke: Clinical outcomes and literature review

Abstract

There are limited data on outcomes of mechanical thrombectomy for pediatric stroke using modern devices. In this study, we report two cases of pediatric acute ischemic stroke treated with mechanical thrombectomy, both with good angiographic result (TICI 3) and clinical outcome (no neurological deficits at 90 days). In addition, we conducted a literature review of all previously reported cases describing the use of modern thrombectomy devices. Including our two cases, the aggregate rate of partial or complete vessel recanalization was 100% (22/22), and the aggregate rate of favorable clinical outcome was 91% (20/22). This preliminary evidence suggests that mechanical thrombectomy with modern devices may be a safe and effective treatment option in pediatric patients with acute ischemic stroke.

Introduction

Several recent trials have demonstrated statistically significant outcome benefit of mechanical thrombectomy in adult patients with acute ischemic stroke (AIS) due to large-vessel occlusion.^1–5 However, the role of mechanical thrombectomy in the pediatric population is not clearly established. Since the publication of a review article on mechanical thrombectomy in pediatric AIS,⁶ numerous case reports on the topic have been published. Based in part on these reports, recently updated American Heart Association guidelines suggest that mechanical thrombectomy may be considered in pediatric patients with AIS due to large-vessel occlusion.⁷ Here, we describe two new cases of mechanical thrombectomy using modern devices in pediatric AIS and review the existing literature to investigate in aggregate the outcomes of this treatment in the pediatric population.

Methods

Institutional review board approval was obtained for retrospective review and reporting of cases treated at our institution. No approval was required for literature review.

Patient 1

This is a previously normal 16-year-old boy with no pertinent family history who had sudden onset of fluctuating dysarthria and ataxia. Upon presentation to the hospital, his National Institutes of Health Stroke Scale (NIHSS) score was 7. Noncontrast computed tomography (CT) of the head demonstrated no intracranial hemorrhage, and aspirin 325 mg was administered orally. Magnetic resonance (MR) imaging of the head and neck demonstrated acute bilateral pontine and right cerebellar infarct with occlusion of the basilar artery and right superior cerebellar artery (SCA) (Figure 1(a)). As the patient was outside the window for intravenous tissue plasminogen activator (tPA), he was taken for catheter angiography seven hours after symptom onset. This study demonstrated non-occlusive thrombus in the basilar artery. During this procedure, the patient had spontaneous clinical improvement, with resolution of all neurological deficits except for minimal dysarthria. In view of this improvement, thrombectomy was deferred and medical management initiated with continuous intravenous heparin infusion. Later, the patient developed worsening neurologic deficits (NIHSS 9) in the pediatric intensive care unit, and intervention was performed 10 hours after symptom onset. Under general anesthesia, a Neuron Max 088 sheath (Penumbra Inc., Alameda, CA, USA) was positioned in the proximal left vertebral artery. Angiography from the left vertebral artery (Figure 1(b) and (c)) demonstrated occlusion of the distal basilar artery and no filling of the posterior cerebral arteries or SCAs. Next, a 5 Max ACE reperfusion catheter (Penumbra Inc., Alameda, CA, USA) preloaded with a Velocity microcatheter (Penumbra Inc., Alameda, CA, USA) and Transend Standard microwire (Stryker Neurovascular, Fremont, CA, USA) was advanced through the sheath and into the thrombus in the distal basilar artery. The microcatheter and microwire were then exchanged for a Penumbra 035 separator (Penumbra Inc, Alameda, CA, USA), which was intermittently advanced and withdrawn within the distal basilar artery as aspiration was applied to the 5 Max ACE reperfusion catheter. Angiography performed after this first attempt at thrombectomy demonstrated restoration of flow in the distal basilar artery and both posterior cerebellar arteries, but persistent occlusion of the right SCA. Two additional attempts at thrombectomy using the 5 Max ACE reperfusion catheter and 035 separator were performed. These efforts resulted in Thrombolysis In Cerebral Infarction scale (TICI) 3 flow restoration, with a small residual filling defect in the proximal right SCA but normal flow in the distal right SCA (Figure 1(d) and (e)). After treatment, the patient was closely monitored in the pediatric intensive care unit. No additional imaging was performed after treatment.

Figure 1.

Teenage boy with dysarthria and ataxia. (a) Brain MRI/MRA shows scattered restricted diffusion in the brainstem and right cerebellar hemisphere, as well as a filling defect in the basilar artery consistent with occlusion. ((b), (c)) Pre-treatment angiogram demonstrates focal long-segment occlusion of the distal basilar artery (arrows) with no filling of the posterior cerebral and superior cerebellar arteries. ((d), (e)) Post-thrombectomy angiogram demonstrates TICI 3 flow. MRI: magnetic resonance imaging; MRA: magnetic resonance angiography; TICI: Thrombolysis In Cerebral Infarction scale.

Patient 2

This is a 17-year-old boy who presented with a gunshot wound to the left face. Head and neck CT angiography demonstrated occlusion of the right internal carotid artery (ICA) above the bifurcation (Figure 2(a)). Initially, he was neurologically intact. Upon transfer to the intensive care unit, his neurologic exam deteriorated with complete left hemiparesis (NIHSS 13). The patient was taken for endovascular therapy four hours after symptom onset.

Figure 2.

Teenage boy status post-gunshot wound to left face. (a) Gas in the carotid space (solid arrow) and non-opacification of the right internal carotid artery on neck CTA. (b) Pre-treatment angiogram demonstrates occlusion of the inferior division of the right MCA (dashed arrow). ((c), (d)) Post-thrombectomy angiogram demonstrates TICI 3 flow. CTA: computed tomography angiography; TICI: Thrombolysis In Cerebral Infarction scale.

Under general anesthesia, a Neuron Max 088 sheath was positioned in the right common carotid artery. Initial angiography demonstrated occlusion of the proximal right ICA and a tandem embolic occlusion in the inferior division of the right middle cerebral artery (MCA) (Figure 2(b)). A 3 Max reperfusion catheter (Penumbra Inc, Alameda, CA, USA) preloaded with a Transend Standard microwire was then advanced through the sheath and into the distal cervical right ICA. The microwire was then removed, and aspiration was then performed through the 3 Max reperfusion catheter. Angiography performed after thrombectomy in the cervical right ICA demonstrated restoration of flow in the distal right ICA, but persistent occlusion of the inferior division of the right MCA. Next, a 5 Max ACE reperfusion catheter preloaded with a 3 Max reperfusion catheter and Transend Standard microwire was advanced through the sheath; the 3 Max reperfusion catheter was positioned in inferior division while the 5 Max ACE reperfusion catheter was positioned in the distal right ICA. Then, the 3 Max and 5 Max ACE reperfusion catheters were sequentially removed from the patient, with suction applied to each catheter as it was withdrawn. Post-treatment angiography demonstrated TICI 3 restoration of normal flow in the right MCA, but interval reocclusion of the cervical right ICA (Figure 2(c) and (d)). Stenting of the ICA was considered but deferred because of robust collateral filling of the right MCA through the anterior and posterior communicating arteries. After treatment, the patient was closely monitored in the pediatric intensive care unit. On post-procedure day 1, a non-contrast head CT was normal.

Literature review

A PubMed search was performed using the following terms individually or in combination: “endovascular treatment,” “pediatric stroke” and “mechanical thrombectomy.” Patient-level data were extracted from individual case reports and case series; these data included occluded vessel location, mechanism of stroke, pre-procedural imaging findings, time to treatment (either reported as time to groin puncture or time to vessel recanalization after symptom onset), device type used, usage of adjunctive thrombolytics, target vessel recanalization (partial or complete), length of clinical follow-up, and favorable clinical outcome (asymptomatic, NIHSS 0–4, modified Rankin Scale (mRS) 0–2). Cases that used balloon angioplasty or Merci devices were excluded from this review.

Results

Case series

In patient 1, angiography demonstrated occlusion of the distal basilar artery that was treated by mechanical thrombectomy comprising three passes of a 5 Max ACE reperfusion catheter and Penumbra 035 separator. Post-treatment angiography demonstrated restoration of normal flow in the basilar artery (TICI 3) 10 hours after symptom onset. At 90-day clinical follow-up, mRS was 0.

In patient 2, angiography demonstrated occlusion of the proximal right ICA and a tandem occlusion in the inferior division of the right MCA that were treated with mechanical thrombectomy comprising one pass with a 3 Max reperfusion catheter in the cervical ICA and one pass with 5 Max ACE and 3 Max reperfusion catheters in the inferior division of the right MCA. Post-treatment angiography demonstrated restoration of normal flow in the right MCA (TICI 3) five hours after symptom onset. At 90-day clinical follow-up, mRS was 0.

Literature review

Including the two cases described above, a total of 22 cases of mechanical thrombectomy in pediatric AIS using modern devices have been reported (Table 1). These data indicate partial or complete vessel recanalization in 100% (22/22) of cases and favorable clinical outcome in 91% (20/22) of cases. Treatment in each of these cases was with modern reperfusion devices (aspiration catheters and stent-retrievers), sometimes used alone or in combination with other devices. Specifically, the Solitaire stent-retriever was used in 11/22 (50%) cases, Penumbra aspiration in 4/22 (18%) cases, and CAPTURE, Phenox, Revive, and Trevo thrombectomy devices being used in only 1/22 (5%) case each. In the 3/22 (14%) cases in which multiple devices were needed, the second-line device was Penumbra in each case.^8,9

Table 1.

Summary of case reports using modern devices for pediatric mechanical thrombectomy.

First author	Age	Location	Etiology	MRI	TTT	Device	Thrombolytic	Recanalization	Length of follow-up (months)	Favorable outcomea
Current study	17	R MCA	GSW	None/CTA	5	Penumbra	None	Yes	3	Yes
Current study	16	BA	Cryptogenic	BA territory	10	Penumbra	None	Yes	3	Yes
Savastano10 (2015)	1.8	BA	CE	BA territory	16	Solitaire	None	Yes	6	Yes
Huded11 (2015)	6	BA	VAD	BA territory	26	Solitaire	None	Yes	3	Yes
Mittal12 (2015)	17	L MCA	CE	L MCA	NR	Solitaire	IV tPA	Yes	NR	Yes
Vega9 (2015)	11	R MCA	CE/Myxoma	None/CTA	3.1	Trevo/ Penumbra	None	Yes	0.3	Yes
Stidd13 (2014)	2	R MCA	CE	R MCA	7	Trevo	None	Yes	1	Yes
Bodey14 (2014)	10	BA	VAD	BA territory	36	Revive	None	Yes	6	No
Bodey14 (2014)	5	BA	Dehydration	No MR, HDAS on NCT	6	Solitaire	None	Yes	6	Yes
Bodey14 (2014)	6	BA	VAD	BA territory	4	Solitaire	None	Yes	6	Yes
Bodey14 (2014)	15	L MCA	Cryptogenic	No MR, HDAS on NCT	6	Solitaire	None	Yes	6	Yes
Ladner15 (2014)	5	BA	VAD	BA territory	9	Solitaire	None	Yes	3	Yes
Rhee16 (2014)	9	L MCA	CE	None/CTA	7	Solitaire	None	Yes	11	Yes
Sainz de la Maza17 (2014)	12	R TICA	CA	None/CTA	8	Solitaire	None	Yes	3	Yes
Fink18 (2013)	11	BA	Cryptogenic	BA territory	4	Solitaire	IV tPA	Yes	3	Yes
Hu8 (2014)	9	L TICA	Cryptogenic	L MCA	NR	Solitaire/ Penumbra	IV tPA	Yes	6	Yes
Hu8 (2014)	7	R TICA	CE	None	NR	Solitaire/ Penumbra	None	Yes	3	Yes
Dubedout19 (2013)	7	BA	Cryptogenic	BA territory	6	CAPTURE	None	Yes	1	Yes
Taneja20 (2011)	14	BA	Cryptogenic	No MR, HDAS on NCT	24	Solitaire	None	Yes	1	Yes
Grunwald21 (2010)	16	BA	Cryptogenic	BA territory	8	Penumbra	IA tPA	Yes	1	No
Grunwald21 (2010)	7	L ICA	CE	L ICA	3.2	Penumbra	None	Yes	1	Yes
Grunwald21 (2010)	16	L MCA	OCPs	L MCA	3.5	Phenox	IA tPA	Yes	1	Yes

MRI: magnetic resonance imaging; BA: basilar artery; GSW: gunshot wound; MCA: middle cerebral artery; TTT: time to treatment; TICA: terminal internal carotid artery; VAD: vertebral artery dissection; IV: intravenous; IA: intra-arterial; tPA: tissue plasminogen activator; CE: cardioembolic; CA: cerebral arteriopathy; HDAS on NCT: hyperdense artery sign on non-contrast head computed tomography (CT); OCPs: oral contraceptive pills; L: left; R: right. ^aFavorable clinical outcome defined as asymptomatic. National Institutes of Health Stroke Scale (NIHSS): 0–4; modified Rankin Scale (mRS): 0–2, Penumbra: 3 Max or 5 Max reperfusion catheter system (Penumbra Inc., Alameda, CA, USA); Solitaire: Solitaire FR revascularization device (Medtronic, Minneapolis, MN, USA); Trevo: Trevo Provue Retriever (Stryker, Kalamazoo, MI, USA); Revive: Revive SE clot retrieval device (Codman & Shurtleff Inc., Raynham, MA): CAPTURE: CAPTURE catheter system (Medtronic, Minneapolis, MN, USA); Phenox: Phenox clot retriever system (phenox GmbH, Bochum, Germany).

The average time to treatment in this cohort was 10.1 hours. In patients with anterior circulation occlusion, the average time to treatment was 5.35 hours (range 3.1–8 hours). In patients with basilar artery occlusion, the average time to treatment was 13.5 hours (range 4–36 hours), including four outliers treated 16 or more hours beyond symptom onset. Excluding these outliers, the average time to treatment in patients with basilar artery occlusion was 6.7 hours.

Discussion

Recently, five consecutive randomized control trials (RCTs) have shown the therapeutic benefit of mechanical thrombectomy in patients with acute large-vessel occlusions in the anterior circulation.^1–5 A pooled meta-analysis of these five RCTs indicated that the number needed to treat for good clinical outcome using mechanical thrombectomy is approximately five.²² This therapeutic benefit compared to earlier, negative trials is likely related to improved patient selection, time to treatment, and device efficacy.

There is no reliable RCT data on the use of intravenous tPA or mechanical thrombectomy in the pediatric population, ostensibly because of the low incidence of AIS and high incidence of stroke mimics that may delay recognition of AIS within established therapeutic windows.^6,23 An open-label multicenter international safety and dose-finding study (TIPS trial) to evaluate the use of intravenous tPA in children was attempted but terminated because of poor enrollment.²⁴ Available data on the use of mechanical thrombectomy in the pediatric population are even more limited. Ellis et al. described endovascular treatment of pediatric AIS and reported a 74% rate of partial or complete recanalization and a complication rate of 29%,⁶ though this review included 20 cases of intra-arterial thrombolysis and had comparatively low utilization of modern stent-retrievers or aspiration catheters.⁶ The literature relating to endovascular treatment of pediatric AIS using modern devices is limited to small case series.

Patient demographics dictate the role for endovascular intervention in children, including especially the time since symptom onset. Current guidelines advocate for treatment of adults with large-vessel occlusion in the anterior circulation if treatment can be initiated within six hours of symptom onset,⁷ with some data suggesting a role for treatment up to eight hours.⁵ All of the patients in this review with large-vessel occlusion in the anterior circulation received treatment within eight hours. Time windows for thrombectomy in the posterior circulation are not clearly established, though many of the patients described here underwent thrombectomy beyond eight hours and still had good clinical outcome.

Reported intra- and post-procedural complication rates of mechanical thrombectomy in children are low. Ellis et al. described post-procedural intracranial hemorrhage (ICH) in only one patient (9.1%; 1/11) undergoing mechanical thrombectomy (this patient also received intra-arterial tPA), compared to 7 out of 23 patients (30.4%) receiving intra-arterial tPA only.⁶ In terms of device-related complications, there is one report of an older-generation device (Merci retriever) becoming fractured and permanently lodged in the left MCA, leading to poor clinical outcome.²⁵ Our aggregated data set includes no cases of post-procedural ICH or device-related complications.

Several factors may guide device selection in cases of pediatric AIS. In general, the underlying lack of vessel tortuosity in these patients allows for easy catheter maneuverability and deployment of thrombectomy devices.⁸ Smaller pediatric vessel size in the range of 2.0–3.0 mm may better accommodate the 0.032-inch Penumbra aspiration system;²⁶ the smallest stent-retrievers commercially available at the time of prior published reports (Trevo, Solitaire) were sized at 4 mm in diameter. These larger stent-retriever devices may induce severe vasospasm, especially in patients under the age of 2.¹⁰ Smaller systems like the Penumbra 3 Max cerebral reperfusion catheter²⁷ or 3 mm stent retriever (“low-profile Trevo”)²⁸ may be helpful in these cases, though there have been instances in which upsizing the stent-retriever led to vessel recanalization when initial passes with smaller devices failed.^13,14 Aside from these considerations, it is difficult at present to directly compare the efficacy of different thrombectomy devices because of small sample sizes, non-blinding of operators, and crossover between devices or thrombolytics. Notably, we had good angiographic and clinical outcomes in our two patients using aspiration, though it should be noted that there is currently no level I evidence to support the use of aspiration catheters in AIS.

There are several limitations to this study. First, the sample size is small, though this reflects the relatively low incidence of large-vessel ischemic stroke in the pediatric population. Second, the retrospective nature without a control group imparts selection bias toward endovascular intervention. Finally, the use of observational studies limited to case series and case reports allows for reporting bias and lack of standardized imaging follow-up. Nevertheless, these preliminary results are encouraging and should prompt consideration of thrombectomy in appropriately selected pediatric patients.

Conclusion

Based on preliminary reports, endovascular thrombectomy may be a safe and effective treatment for large-vessel occlusion in the pediatric population.

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