Recanalization Treatments for Pediatric Acute Ischemic Stroke in France

Key Points Question In children with acute ischemic stroke, are intravenous thrombolysis (IVT) and/or endovascular treatment (EVT) safe and effective revascularization strategies? Findings In this cohort study of 68 children treated with IVT (n = 44) and/or EVT (n = 40), favorable neurologic outcome was observed in 78% at 1 year. Complication rates of revascularization treatments were low and main outcomes were not significantly different between the EVT and no-EVT groups, although the condition of patients who received EVT was initially more severe. Meaning The findings of this cohort study suggest that use of IVT and/or EVT may be safe in children with acute ischemic stroke.


Introduction
The incidence of acute arterial ischemic stroke (AIS) is known to increase with age, but AIS is not uncommon in children. 1 Acute arterial ischemic stroke during childhood, with an estimated incidence of 2 to 8 per 100 000 children-year, is an important burden of acquired disability and has long-term socioeconomic and psychosocial outcomes. 2 The field of AIS has seen major evolutions in adults in the past decades with the advent of intravenous recombinant tissue plasminogen activator (r-tPA) treatment and endovascular treatment (EVT) for select patients older than 18 years 3 ; however, to our knowledge, there is limited evidence that revascularization strategies are associated with improved functional outcome in children. 4 In turn, intravenous r-tPA and EVT are administered to children on a case-by-case basis, using indirect evidence derived from adult clinical trials. Beyond the undisputed pathophysiologic rationale of the benefits of blood flow restoration to salvageable hypoperfused brain tissue in the context of acute brain arterial occlusion, 5,6 many questions on the applicability, safety, and benefits of brain revascularization strategies in children remain. 5 Differences in stroke etiologies, coagulation system maturation, rarity of the condition leading to delayed diagnosis, and lack of experience may limit the applicability of treatments with proven benefit in adult stroke to pediatric stroke. Reports of intravenous thrombolysis (IVT) in children suggested a positive safety profile, despite the limited number of patients included. 7,8 The SaveChildS study 9 analyzed data from a multicenter cohort that included 73 children treated with EVT between 2000 and 2018. The investigators reported that EVT appeared to have a positive safety profile in children with large-vessel occlusion, similar to the results of the largest real-world prospective adult registries, and that long-term neurologic outcome was good in most patients. However, the generalizability of these results remains questionable because of the overrepresentation of cardioembolic events in this study and the risk of selection bias over a 220-month inclusion period in 27 centers. Recently, Dicpinigaitis et al 10 extracted data on a cohort of 190 children treated with EVT from the US National Inpatient Sample and supported the safety profile of EVT in a large pediatric sample. However, the generalizability of results remains limited by many approximations regarding the population characteristics and outcomes owing to the lack of key variables, such as the National Institutes of Health Stroke Scale (NIHSS) score or modified Rankin Scale score. At times of profound modifications in AIS treatment paradigms, attempts at optimizing systems of care toward pediatric stroke-ready networks have shown interest, 7 and both technical advances and better understanding of underlying etiologies of pediatric stroke have positioned the issue of pediatric AIS acute revascularization in the limelight. In a national multicenter retrospective study of consecutive children with AIS treated with IVT (using r-tPA) and/or EVT we aimed to provide an up-to-date evaluation of revascularization strategies in the pediatric population. children aged 28 days to 18 years with acute AIS who received a recanalization treatment and registered on ClinicalTrials.gov (NCT03887143).
This retrospective analysis received approval from the ethics committee of the French National Pediatrics Society. Patients and/or legal guardians were informed and gave oral consent for the children to be included in this study; no financial compensation was provided. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Patients were included in the study if they met the following criteria: (1) age 28 days to 18 years at the time of stroke onset, (2) radiologically confirmed acute AID (ie, patient had both radiographic evidence of an infarct in an arterial distribution and consistent neurologic signs and symptoms), (3) received IVT and/or EVT and/or digital subtraction arteriography in an intention-to-treat protocol with EVT at the acute phase of AIS, (4) residence in mainland France or overseas French territories, and (5) presented during the inclusion period(January 1, 2015, to May 31, 2018).

Recruiting Centers and Validation Method
We hypothesized that care for every child with AIS receiving a recanalization treatment would be managed at a tertiary care center. Thus, stroke centers, pediatric neurology departments, and interventional neuroradiology departments were separately contacted at every academic center in France and asked whether they had treated at least 1 pediatric patient during the study period.
Duplicates across departments were identified and resolved. The Paris-Ile-de-France region, the most populated area of France, comprising Paris, served as a validation region for this method. In this region, every stroke unit was contacted, including nonacademic centers, to assess whether patients may be overlooked by focusing on academic centers.

Data Acquisition
Local investigators collected information on clinical and biological parameters at admission, including demographic data (age, sex), stroke presentation, clinical and biological assessment (including pediatric NIHSS [pedNIHSS] score, blood pressure value, coagulation parameters, and serum glucose level before treatment), and medical history and prestroke level of independence. The coordinating investigator (M.K.) centrally reviewed all score ratings, especially if the ratings were inferred from medical records. 11,12 A panel of 3 neuroradiologists (B.K., B.H., and O.N.) blinded to clinical data centrally reviewed initial, procedural, and follow-up imaging data: brain computed tomography (CT) and/or magnetic resonance imaging (MRI), CT or MR angiography, digital subtraction angiography, and posttreatment brain CT and/or MRI. The neuroradiologists rated the extent of the infarct in 2 ways. First, the infarct was rated according to the adult Alberta Stroke Program Early CT Score (ASPECTS) or posterior circulation ASPECTS depending on the vascular territory involved on initial CT and/or MRI. ASPECTS provides segmental assessment of the vascular territory, and 1 point is deducted from the initial score of 10 for every region involved (from 10 [no lesion] to 0 [maximum lesions]). 13 Second, the infarct was rated according to the ABC/2 method, which consists of 3 orthogonal linear measures to estimate the core volume. 14

Stroke Etiology
Parameters of baseline stroke workup were collected, including cardiac ultrasonography, CT and/or MRI, and CT angiography or MR angiography of the head and neck, and screening for infection, metabolic disease, coagulation abnormality, and vasculitis. 15 For each patient, a centrally performed multidisciplinary symposium determined stroke etiology according to the Childhood Arterial Ischemic Stroke Standardized Classification and Diagnostic Evaluation Classification measures, 15 which distinguishes the following subtypes: (1) unilateral or bilateral focal cerebral arteriopathy (FCA), (2) cardioembolic, (3) cervical artery dissection, (4) thrombotic, (5) other etiologies, and (6) multifactorial or unknown.

Recanalization Treatments and Pathway of Care
In this retrospective study, the final decision on treatment eligibility was at the discretion of the multidisciplinary team at included sites. Criteria for intravenous r-tPA and/or EVT eligibility in adults (eg, time from symptom onset, NIHSS score, absence of contraindication, and ASPECT score) were retrieved to estimate a posteriori the degree of adherence to and transposition from practice guidelines in adults.
We recorded the following variables when applicable: time metrics, including time from symptom onset (or last-seen well) to hospital presentation to first imaging to first recanalization treatment and, when appropriate, to groin puncture and recanalization, and technical characteristics of recanalization treatments (infusion characteristics, initial occlusion location on digital subtraction arteriography, EVT technical strategy, devices used, and number of passes performed during EVT).
Stroke pathway-of-care data were collected, including hospital-to-hospital transfers and management in a pediatric or adult ward.

Outcome Definitions
Patients treated with EVT had an immediate posttreatment recanalization assessment according to the extended modified Treatment in Cerebral Infarction (mTICI) score (mTICI 0 indicates no perfusion; grade 1, antegrade reperfusion past the initial occlusion but limited distal branch filling; grade 2a, antegrade reperfusion of less than half of the occluded target artery previously ischemic territory; grade 2b, antegrade reperfusion of more than half of the previously occluded target artery ischemic territory; grade 2c, near-complete perfusion except for slow flow in a few distal cortical vessels or presence of small distal cortical emboli; and grade 3, complete antegrade reperfusion, with an absence of visualized occlusion in all distal branches). A TICI score greater than or equal to 2b was considered to indicate successful recanalization. 16 Early clinical outcome was evaluated for all patients using the pedNIHSS at 24 hours and day 7 after admission and the modified Rankin Scale Clinical and radiologic complications related to stroke and/or stroke treatment were recorded: malignant cerebral artery infarct, hemorrhagic transformation, or stroke-related mortality.
Symptomatic intracerebral hemorrhagic transformation was defined according to the ECASS-II definition 17 as any intracranial hemorrhage on imaging control performed 24 hours after mechanical thrombectomy that was associated with neurologic deterioration (increase of Ն4 points in the pedNIHSS score from baseline).
Long-term outcome was evaluated using mRS at 90 days and 12-month follow-up and disability was scored using the pediatric stroke outcome measure (PSOM 0 [no deficit] to 10 [maximum deficit]). 18 For patients with missing PSOM values, the coordinating investigator (M.K.) established a retrospective scoring inferred from medical records in collaboration with the local investigators. 11,12

Statistical Analysis
Continuous variables were summarized using means (SDs) or medians (IQRs) as appropriate, and discrete variables were summarized using counts (percentages). Variables were summarized descriptively by treatment type for all patients included, and no imputation of missing variables was performed. We used the χ 2 or Fisher exact test, as appropriate, for comparisons of categorical variables between treatment groups and among age groups (0-6, 7-12, and 13-18 years) and we used the Wilcoxon rank sum test for continuous variables. All computations were performed with JMP, version 14 (SAS Institute Inc). The significance threshold was set at a 2-tailed value of P < .05 for all analyses. If needed, we derived 95% CIs by bootstrapping (2500 occurrences) statistical results. 19

Revascularization Treatments
Three procedures were performed at children's hospitals and 37 were conducted in adult hospitals.
All physicians performing the procedures were senior interventional neuroradiologists with at least 3 years of experience in EVT working in tertiary adult centers.   (Figure 1 and Figure 2).
In the EVT group, 1 patient developed intracranial arterial perforation (2.7%; 95% CI, 0%-16.6%), and 1 developed emboli in a new territory; both of these children had unfavorable outcomes at 1 year (mRs, 4). No other vascular complications, such as arterial dissection, periprocedural thrombosis, or puncture site complications were reported.

Long-term Clinical Outcome
Long-term clinical outcome was good, with an mRS score of 0 to 2 noted in 46 of 67 children (68.7%) at 3 months and 52 of 66 (78.8%) at 12 months after stroke onset. The median mRS score was 2 (IQR, 0-3) at 3 months and 1 (IQR, 0-2) at 12 months and did not differ significantly between the EVT and IVT-only groups. The median PSOM score was 2 (IQR, 0.5-4) after 3 months and 1.5 (IQR, 0.5-3) after 12 months-not significantly different between the EVT and IVT-only groups. Analysis by age group suggests that stroke recovery may be longer in children younger than 6 years without an association with longer-term prognosis with higher mRS scores in this subgroup at 3 months but comparable to scores in older children at 12 months ( Table 2).

Recruiting Centers and Method Validation
In mainland France and overseas territories, 57 academic hospitals were contacted through their respective stroke center, pediatric neurology, and interventional radiology (where applicable) departments. The response rate was 100%. Twenty-eight centers (49.1%) reported no revascularization treatment in pediatric patients with AIS during the study period, and 29 centers reported at least 1 treatment. A median of 2 patients (IQR, 1-3), with a maximum of 8, were treated at each center during the 42-month inclusion period. In the validation region, 21 centers were contacted (9 academic and 12 nonacademic centers).
Among the 17 patients included in this region, only 1 was treated in a nonacademic center, yielding a transposed estimated 94% (95% CI, 73.0%-98.9%) nationwide exhaustivity during the study period.

Discussion
This study provides data on acute revascularization strategies for AIS in select children, using a retrospective nationwide study initiated in 2015, when mechanical thrombectomy became the standard of care in adults with AIS with large-vessel occlusion, alone or in addition to IVT. Our main findings are that (1) 22 In this work, the authors noted that no child experienced a symptomatic hemorrhagic transformation following r-tPA infusion and that asymptomatic intracerebral hemorrhage occurred very infrequently compared with the incidence rates in adult samples. Similarly, in our sample, none of the children treated with PA developed intracranial hemorrhage or neurologic deterioration following r-tPA administration. Together, our data are in line with previous reports that the risk of ICH after intravenous r-tPA is low in in appropriately selected children, 4,22 suggesting that for similarly selected children, the benefits of thrombolysis may outweigh the risks.
Moving forward, as discussed in the recently published study on the Save ChildS registry, 9 the main drawback of the off-label use of EVT in children is concern regarding its safety in the absence of solid evidence regarding the clinical benefits of EVT. 5   One important safety question is whether certain underlying etiologies bear a higher risk for EVT procedure complications or futility. In our study, as in the SaveChildS study, 9 the representation of stroke etiologies is different from classically reported causes of stroke, with fewer children with focal cerebral arteriopathy (25% in KidClot, 10% in SaveChilds, 30% in overall pediatric strokes) and more children with a cardioembolic stroke. 5 This variability may reflect the widely expressed concern about EVT in the setting of FCA, relating to the inflammatory nature of FCA, which may represent a risk factor for artery perforation or treatment futility. In our study, patients with FCA displayed a low procedural hemorrhagic risk comparable to patients with cardioembolic stroke. the rate of reocclusion or persisting more than 50% stenosis on follow-up imaging was substantially elevated in this subgroup. These results suggest a higher possibility of EVT treatment failure, but with safety, in patients with FCA, which should be further investigated.
Furthermore, included patients were older than overall pediatric patients with AIS in both reports, potentially suggesting that (1) treating physicians were likely more inclined to use IVT or EVT in older children, who resemble adults, (2) treating physicians made an implicit choice toward patients with cardioembolic stroke (who are older than those with FCA), and (3) younger children may have been deemed either ineligible or presented outside the standard therapeutic windows owing to increased care delays.
Although our data contribute to the accumulating evidence on the role of emergent revascularization treatments for children with acute cerebral arterial occlusion, there remain issues that will likely require long-term international collaboration on the way to optimizing therapeutic strategies in pediatric patients. One key issue regards the benefit of emergent revascularization therapies in this population.
If the proportion of children with AIS who could theoretically benefit from revascularization is largely unknown, we believe the children in our cohort who were treated during the study period represent the overall population among a pool of 15 million individuals younger than 18 years. 25 Considering an incidence rate of 2 to 8 per 100 000 children-years for all pediatric AIS, 1 this rate translates to a proportion of 1.5% to 5.9% of children with AIS receiving revascularization. This figure compares unfavorably with the rates of more than 40% of eligibility for revascularization found in adults, following the initiation of both telestroke networks and expansion in EVT indications. 26 It is arguable that in the absence of robust evidence for the efficacy of these newer strategies, this proportion will not steadily increase, but our study aimed to provide data for both clinicians admitting children with suspected AIS and policy makers to inform future treatment recommendations. Yet, as moving forward, the community needs to address the question as to why revascularization strategies are proposed at such low rates in children, limiting both the opportunities to optimize systems of care locally 7 and the feasibility of a randomized evaluation of

JAMA Network Open | Neurology
Recanalization Treatments for Pediatric Acute Ischemic Stroke treatment strategies. 4 In this work, we found that 75% of included centers (encompassing all the largest volume stroke centers nationally) treated fewer than a median of 1 pediatric patient per year with IVT, EVT, or both strategies. This point mandates a careful analysis of potential means for improvement both locally and globally that could include standardized imaging pathways for children with acute neurologic symptoms and increase awareness at adult stroke centers and nontertiary pediatric emergency departments.

Strengths and Limitations
The strengths of our study include (1) a relatively large sample size of recanalization treatment in pediatric patients, observed after mechanical thrombectomy became standard of care in adults with AIS and large-vessel occlusion, using modern devices and, in some cases, an optimized pathway; (2) a centralized review of imaging data; and (3) a centralized review of stroke etiology by a multidisciplinary symposium. The study has limitations. These limitations are inherent to the retrospective design, including missing data, selection bias, heterogeneity of the population, lack of a control group and, inability to perform direct comparisons between treatment groups.

Conclusions
The findings of this study suggest that use of IVT, EVT, or both in a population of children with AIS may be safe. The 1-year neurologic outcome was good in three-quarters of the study sample.
Endovascular treatment proved to be feasible, with rates of successful revascularization and procedural duration of similar magnitude to those found in adult populations. These results may help clinical decision-making for the use of revascularization treatment in pediatric AIS and encourage international collaboration to optimize therapeutic strategies and pathways in pediatric patients.