Article Text

Download PDFPDF
Endovascular mechanical thrombectomy for acute stroke in young children
  1. Lisa R Sun1,
  2. Ryan J Felling1,
  3. Monica Smith Pearl2,3
  1. 1 Neurology, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
  2. 2 Radiology, Johns Hopkins Hospital, Baltimore, Maryland, USA
  3. 3 Radiology, Children’s National Medical Center, Washington, District of Columbia, USA
  1. Correspondence to Dr Lisa R Sun, Neurology, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA; lsun20{at}


Background Mechanical thrombectomy has emerged as a standard of care for acute stroke from large vessel occlusion in adults but remains controversial in children. Cerebral vessels are nearly adult size by 5 years of age but the technical feasibility of achieving recanalization in younger and smaller children with current endovascular tools remains unclear.

Objective To systematically review the literature on mechanical thrombectomy for stroke in children less than 5 years of age.

Results Mechanical thrombectomy for acute stroke has been reported in 11 children under the age of 5 years (range 9 months to 4 years). The mean time from symptom onset to groin puncture was 12 hours (range 4–50 hours). Complete recanalization was achieved in 7/12 (58%) vessels attempted, and partial recanalization in 4/12 (33%). Two procedure related complications were reported, with small vessel size felt to be contributory to basilar vasospasm in one case. Favorable neurological outcomes were reported in 7 cases (64%).

Conclusions Our review of the literature demonstrates that mechanical thrombectomy for acute ischemic stroke may be feasible in carefully selected infants and young children less than 5 years of age using currently available devices. Efficacy in promoting better neurologic outcomes remains unproven, and other questions persist, including whether complications such as vasospasm occur more frequently in young children compared with adults. Further study is needed to determine the safety and efficacy of pediatric mechanical thrombectomy. These data suggest that young children should not be excluded from future studies or clinical treatment on the basis of age alone.

  • pediatrics
  • stroke
  • thrombectomy

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Endovascular mechanical thrombectomy (MT) has revolutionized stroke care over the past several years. Multiple clinical trials have demonstrated the benefit of MT in carefully selected adults with acute stroke and large vessel occlusion.1–3 Although there are no similar randomized controlled data in children, an increasing number of reports suggest that endovascular therapy for acute childhood arterial ischemic stroke may be safe and effective.4–6 A recent recommendation from the American Heart Association/American Stroke Association suggests that MT with stent retrievers may be reasonable for some patients <18 years of age with large vessel occlusions if groin puncture can be achieved within 6 hours of symptom onset.7

While endovascular approaches are being increasingly used for the treatment of acute childhood stroke,4 the use of MT in young children is still controversial due to safety concerns regarding compatibility of devices developed for adults with the smaller pediatric intracranial vasculature and size of the femoral artery. A recent study of the angiographic diameters of the large cervical and cranial vessels found rapid vascular growth until age 5 years, at which time children’s vessels are nearly adult size.8 This suggests size compatibility of endovascular devices with the cerebral vasculature of children aged 5 years and older, but whether use of these devices in younger children is safe and feasible remains unclear. These technical concerns may result in MT being withheld from young children. In a recent survey of pediatric stroke neurologists regarding acute stroke treatment,9 most respondents did not have an established minimum age for considering thrombectomy. However, consonant with findings that cervicocerebral arterial diameter does not approximate adult size until age 5 years, of those who did report a minimum age, only 20% would consider MT in children <5 years of age (personal communication with Dr Jenny Wilson). Therefore, to address current concerns and technical considerations specific to this age group, we describe our systematic review of the literature on MT in children <5 years of age.


A systematic search of PubMed was performed to identify all studies published in English on or before May 1, 2018 on the use of MT for acute stroke in patients <5 years of age. We used the following search terms: ‘stroke’ AND ‘thrombectomy’ or ‘endovascular’ or ‘intra-arterial’ or ‘embolectomy’ AND ‘pediatric’ or ‘childhood’ or ‘child.’ Reference lists in all relevant articles were perused to identify additional relevant cases. Case series or reports on the use of MT in patients <5 years of age with acute ischemic stroke were included. Patients who received intra-arterial (IA) pharmacologic thrombolysis, such as IA alteplase, but did not have mechanical clot retrieval, were excluded. Duplicate cases were excluded.

We evaluated baseline characteristics (age, sex, stroke risk factors), stroke presentation/severity, and imaging characteristics. Treatment related variables included time to treatment (time to groin puncture and time to recanalization), devices used, adjunct therapies, and procedure related complications. Radiographic and clinical outcomes were assessed. In accordance with prior studies,5 10 complete recanalization was defined as Thrombolysis in Cerebral Infarction (TICI) 2b and TICI 3, or notation of complete recanalization. Partial recanalization was defined as TICI 2a or notation of partial recanalization. No recanalization was defined as TICI 1, TICI 0, or notation of no recanalization. Clinical outcomes were based on the furthest available clinical endpoint. In congruence with prior studies, clinical outcome was classified as favorable if the patient was asymptomatic by clinical assessment, National Institute of Health Stroke Scale (NIHSS) 0–4, modified Rankin Scale (mRS) 0–2, pediatric stroke outcome measure (PSOM) <1, or Gross Motor Function Classification System (GMFCS) level 1.5 11 12


Eleven cases meeting the inclusion criteria were found, with dates of publication ranging from 2013 to 2018 (figure 1, table 1).4 6 13–21 In these cases, 14 vessels were occluded and MT was pursued in 13 vessels. In one case, two vessels were occluded but MT in only one vessel was attempted due to evolving infarction in the other territory.

Table 1

Reported cases of mechanical thrombectomy in children <5 years of age

The age range for thrombectomy in the cases reviewed was 9 months to 4 years of age, with a median age of 2.1 years. Of the 10 cases in whom patient sex was reported, 70% were male and 30% were female. The most commonly identified stroke etiology was cardiac disease, present in 60% of cases, in which specific stroke risk factors were identified. Arterial distributions involved included the anterior circulation in five cases, the posterior circulation in three cases, and both the anterior and posterior circulation in three cases. Presenting symptoms were variable and depended on the vascular territory affected. NIHSS at the time of the decision to proceed with thrombectomy was reported in six cases (ages 2.1–4 years) and ranged between 8 and 21.

Times to groin puncture and recanalization were highly variable. The mean time to groin puncture was 12 hours, ranging from 4 hours to 50 hours from the time the child was last seen well or symptom onset. Anterior circulation involvement was associated with a shorter time to groin puncture (5.2 hours in the anterior circulation vs 28 hours in the posterior circulation, P=0.009). Groin puncture occurred between 4 and 7.7 hours when the anterior circulation was involved and between 16 and 50 hours when only the posterior circulation was involved. The time from groin puncture to recanalization was only reported in a minority of cases but was, on average, <2 hours and similar in the anterior (two cases, mean 108 min) and posterior (three cases, mean 75 min) circulation.

Intravenous tissue plasminogen activator (tPA) was administered in two cases (18%, ages 3 years and 4 years) prior to MT. IA tPA was administered as part of the endovascular treatment in three cases (27%, mean age 3.4 years). The Solitaire stent retriever was the most commonly used device, with reported use in seven cases (64%). Multiple other devices were used in single case reports (table 2).

Table 2

Types of intervention and endovascular devices used

Two procedural complications were reported. Asymptomatic possible intracranial hemorrhage (could not be distinguished from contrast extravasation) was noted in a child who had also received IV tPA.19 In a toddler with a basilar artery occlusion, vasospasm was noted during the procedure and resolved after administration of IA nicardipine.14 The vasospasm was thought to be related to the large diameter stent, with the artery measuring 2.5 mm and the stent measuring 4 mm. The authors note that this procedure was performed prior to the availability of 3 mm stents.

Radiographic outcome was reported for 12 vessels. Complete recanalization (TICI 3 or 2b, or notation of complete recanalization) was achieved in seven vessels (58%), partial recanalization (TICI 2a or notation of partial recanalization) was achieved in four vessels (33%), and no recanalization (TICI 1 or 0, or notation of no recanalization) was reported in one vessel (8%). In the cases of posterior circulation thrombectomy, 80% (4 of 5) of attempts achieved complete recanalization as opposed to only 43% (3 of 7) of attempts in the anterior circulation.

There was no standard manner in which clinical outcomes were reported. NIHSS, mRS, PSOM, GMFCS level, assisting hand assessment score, and clinical description between discharge and 1 year were the outcome measures reported. A favorable outcome was achieved in seven cases (64%). In the remaining four cases (36%), poor outcome was reported and ranged from mRS 3 to 4 at 90 days, PSOM 1.5 at discharge and 24 months, and death in one case.


Use of MT for acute stroke in young children is increasingly reported in the literature, perhaps related to the emergence of a plethora of data regarding the safety and efficacy of MT in a subset of adults with acute stroke.1–3 Successful recanalization has been reported in patients as young as 9 months of age and in 11 cases of children <5 years of age.4 6 13–21

Radiographic outcomes after MT for acute stroke in the pediatric population have been reported to be excellent, with consistently high rates of recanalization, although this may be subject to reporting bias.5 11 In the current analysis of children <5 years of age, some recanalization was achieved in 11 of 12 vessels for which thrombectomy was attempted, and complete recanalization was achieved in seven vessels. These data suggest that MT is technically feasible in at least some infants and young children. While overall rates of recanalization in young children are similar when compared with older children (table 3), the rate of complete recanalization of 58% is lower than previously reported complete recanalization rates in children of all ages (76–79%) (table 3)5 22 and adults (71%)23 who underwent MT for acute stroke, although direct comparison between these groups is not possible given the small sample size and likelihood of reporting bias.

Table 3

Comparison of two case reviews of mechanical thrombectomy for acute stroke

Favorable clinical outcomes were seen in 64% of patients in this review, in comparison with 80–87% rates of favorable outcomes reported in similar analyses that included the complete pediatric population.5 22 Interestingly, the proportion of pediatric patients with favorable outcomes is higher than what has been reported in the adult stroke literature, with 46% of adults receiving MT within 6 hours of stroke onset achieving an mRS score of 0–2 at 90 days.1 This may be related to reporting bias, as the adult data come from randomized controlled trials as opposed to the pediatric data which come from case reports and case series, or it may be related to better stroke outcomes in children compared with their adult counterparts. Alternatively, differences in outcomes may be attributable to variability among the outcome measures used and difficulties inherent in applying adult outcome measures to children. In the absence of randomized trial data, it is not possible to conclude that MT is effective in the pediatric population.

Procedure related complications were reported in two cases in this analysis, yielding a complication rate of 18%, which is similar to the 14% rate in a previous analysis of all pediatric patients who underwent MT for acute stroke.5 The first was asymptomatic possible intracranial hemorrhage (could not be distinguished from contrast extravasation) in a child who had also received IV tPA.19 The other was intraprocedural cerebral vasospasm felt to be related to the relatively large size of the 4 mm stent in comparison with the vessel.14 Importantly, this case was performed prior to the development of a smaller 3 mm stent. The vasospasm resolved with administration of IA nicardipine. Both of these children had a good clinical outcome despite these complications. Notably, one additional child died, but the death was felt to be unrelated to the procedure. The complication rates reported in this analysis and in the larger review of all pediatric patients are similar to the reported complication rate of 12.5–15% in adult trials.24 25 Although the reported complication rates of MT for acute stroke in young children are similar to complication rates in older children and adults, caution is advised in interpreting these results given the small number of published cases as well as the likelihood of publication bias in retrospective pediatric reports, with potential underreporting of cases with poor outcomes or severe complications.

Evidence from large randomized trials of adults with acute stroke and causative large vessel occlusion without significant early ischemic changes on imaging supports consideration of MT in patients within 6 hours of stroke onset and in a subset of patients with a mismatch between core volume and clinical deficit or perfusion abnormality up to 24 hours from stroke onset. Interestingly, in the survey of pediatric stroke neurologists regarding MT practices, notably performed prior to the recent studies demonstrating the efficacy of MT in an extended time window in selected patients, 33% of respondents used a time window of 6 hours for anterior circulation stroke in accordance with the adult data and the American Heart Association/American Stroke   Association recommendations, but 24% used a time window of 8 hours and 38% had no set time window.8 In this analysis, only one case of anterior circulation vessel occlusion was treated with MT outside of the traditional 6 hour window, with groin puncture at 7.7 hours from stroke onset. The outcome in that case was unfavorable. However, three cases of basilar thrombectomy were performed 16 hours or more after stroke ictus. All three of these patients had favorable outcomes. The good outcomes in this review despite prolonged time to recanalization must be interpreted in the context of a more favorable natural history of basilar artery occlusion in children compared with adults. In an analysis of the Swiss Neuropaediatric Stroke Registry, a prospective multicenter registry which included 97 cases of childhood basilar artery occlusion, 92% of children with a basilar artery occlusion survived and 46% had a good outcome, as defined by an mRS score of 0–2.26 In contrast, in the BASICS study of 600 adults with basilar artery occlusions, fewer than 25% of patients had an mRS score of 0–2 at follow-up despite treatment, and the case fatality rate was 36%.27 This suggests a significantly better outcome after basilar artery occlusion in children compared with adults, and warrants caution in the interpretation of uncontrolled studies of stroke treatment in children. Regardless, some authors have suggested that the therapeutic window for MT in children may extend beyond the established window for the adult population,6 but there is little evidence yet to directly support this.

In our series, times to groin puncture were significantly longer in cases of posterior circulation involvement alone compared with cases with anterior circulation involvement. This may be related to the tendency of clinicians to pursue basilar thrombectomy under less stringent time criteria based on anticipated poor outcomes after basilar artery occlusion. Interestingly, 27% of the reported MT cases in children <5 years of age had involvement of both the anterior and posterior circulation. This is higher than the 5–7% of reported cases in the full pediatric population with vascular occlusions of both the anterior and posterior circulation.5 22 This more frequent involvement of multiple vascular territories might be related to the higher proportion of cardioembolic stroke etiologies in the young group (60%) compared with 35–40% of the full pediatric group. The presence of multiple vascular occlusions could present a technical challenge for thrombectomy and increases time to recanalization.

In this review of published cases of MT for acute stroke in children <5 years of age, we report similar rates of recanalization, favorable outcome, and procedure related complications compared with the full pediatric population and the adult population. However, there are several considerations specific to this population. First, children <5 years of age have smaller cerebral vasculature than adults.8 As seen in one case in this review, when adult devices are used, pediatric size vessels may be at increased risk for vasospasm. Another important technical consideration is the size of the femoral artery, limiting femoral arterial sheath size. Fortunately, current stent retrievers can be delivered through microcatheters that will fit through a 4 F guide catheter and 4 F femoral arterial sheath. In the authors’ opinion, in the young age group, every attempt should be made to use the smallest delivery system that achieves successful recanalization. Adjunct devices, specifically aspiration catheters used alone or in conjunction with stent retrievers, require a larger guide catheter and larger femoral arterial sheath. In our opinion, the decision to use these additional devices and a larger delivery system must be made carefully and should be reserved for older children, particularly when consistently high rates of recanalization are reported with stent retrievers, which can be successfully delivered with a 4 F system. Additionally, although no complications resulting in significant blood loss were reported in our population, the volume of blood aspirated, whether by hand or by an aspiration pump, can be considerable and has a larger impact in young children than the same volume in older children and adults.

As there are no randomized controlled data to support the use of MT in children with stroke, the decision to pursue endovascular therapy must be made on a case by case basis. Procedural risks must be weighed against the natural history of the disease. The level of risk should be assessed by a multidisciplinary team, including both pediatric stroke neurologists and interventionalists,28 and may depend on medical comorbidities, time from stroke onset, parent preference, and the experience of the treating team. Safety data on pediatric diagnostic cerebral angiography suggest very low complication rates when performed by experienced operators.29 30 Therefore, when feasible, consideration should be given to emergent transfer to centers with extensive pediatric neurointerventional experience.

The relative rarity of pediatric stroke is a significant barrier to obtaining class I evidence for MT in children. The Thrombolysis in Pediatric Stroke (TIPS) trial, a prospective study to evaluate IV tPA in pediatric arterial ischemic stroke, was unfortunately terminated early in 2013 due to lack of patient accrual.31 It is likely that a randomized trial of MT would suffer from the same difficulties with recruitment. Therefore, it is imperative that all pediatric thrombectomy cases be included in registries or published case series. The current analysis highlights the need for a higher level of uniformity when reporting endovascular cases. At a minimum, we suggest that reports include the pediatric NIHSS at presentation (validated in children 2–18 years of age obtained prospectively or retrospectively),32 33 vessels occluded, time to groin puncture, time to recanalization, adjunct use of IV thrombolysis, endovascular devices including femoral sheath size, procedural complications, radiographic outcome (TICI score), and clinical outcome (specifically PSOM and mRS at 3 months).

This review demonstrates that MT for acute ischemic stroke may be feasible in carefully selected infants and young children. In the hands of an experienced operator, the devices currently used for adult MT can be utilized to achieve recanalization in even the infant vasculature. Therefore, we suggest that the decision to pursue MT should be made based on the individual patient and treating team instead of a predetermined lower age limit. Further study is needed to determine if this treatment modality is safe and effective for acute stroke in this young age group.

Figure 1

Number of reported cases of mechanical thrombectomy in children <5 years of age by year.



  • Contributors LRS conceived of the research idea and designed the systematic review, performed the systematic review, performed the initial analysis and interpretation of the data, drafted the manuscript, revised the manuscript, and crafted and submitted the final version. She agrees to be accountable for all aspects of the work. RJF contributed meaningfully to the study design, interpretation of the data, data analysis, and critical review of the manuscript, including the initial submission and revision. He agrees with the final version of the manuscript and agrees to be accountable for all aspects of the work. MSP contributed meaningfully to the study design, interpretation of the data, data analysis, drafting sections of the manuscript, and critical review of the remainder of the manuscript, including the initial submission and revision. She agrees with the final version of the manuscript and agrees to be accountable for all aspects of the work.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement There are no additional unpublished data from this study.

  • Patient consent for publication Not required.