Background
Acute arterial ischemic stroke (AIS) is caused by an interruption of cerebral arterial blood flow due to thrombotic occlusion or arterial narrowing causing ischemic injury within an arterial territory. Reference Ferriero, Fullerton and Bernard1 Pediatric stroke occurs between 29 days and less than 18 years of age and affects approximately 2 in 100,000 children annually. Reference Ferriero, Fullerton and Bernard1,Reference deVeber, Kirton and Booth2 It is associated with an in-hospital mortality of 2.6%, with more than 50% of children with AIS suffering from long-term disability including persistent neurological deficits, seizures, and learning or developmental problems. Reference Ferriero, Fullerton and Bernard1,Reference Beslow, Dowling and Hassanein3–Reference Kirton, DeVeber and Pontigon5 Childhood AIS is diagnosed with a median delay of 23 hours post symptom onset. Reference Rafay, Pontigon and Chiang6 This has been attributed to several factors: a lack of awareness regarding childhood AIS among caregivers and health care professionals; barriers to accessing timely neuroimaging; and a high frequency of stroke mimics in children. Reference Rafay, Pontigon and Chiang6
Intravenous tissue-type plasminogen activator (IV-tPA) administered within 4.5 hours of symptom onset has been shown to significantly improve outcomes following acute ischemic stroke in adults. Reference Boulanger, Lindsay and Gubitz7 The use of tPA in children remains controversial. Currently, neither IV nor intra-arterial (IA) tPA is approved for use in children with AIS. Reference Bernard, Friedman and Stence8 In 2008, the American Heart Association Guidelines recommended that children with AIS should not receive tPA outside of a clinical trial, however consensus was not reached regarding tPA treatment for adolescents with AIS who otherwise met adult eligibility criteria. Reference Roach, Golomb and Adams9 Since that time, an increasing number of children have been treated off-label with tPA, with published case reports and case series reporting it to be well-tolerated in the vast majority. Reference Amlie-Lefond, deVeber and Chan10 The delay to diagnosis remains a barrier to tPA therapy given the recommendations for treatment within a 4.5-hour time window from symptom-onset. Accelerating the time to diagnosis is possible with the implementation of a stroke protocol in the pediatric emergency department, as shown in a 2016 study. Reference DeLaroche, Sivaswamy, Farooqi and Kannikeswaran11 Children presenting outside the time window for IV-tPA may still benefit from endovascular treatment (EVT). Reference Bernard, Friedman and Stence8 There is increasing evidence for EVT as a safe and effective treatment for childhood AIS. Reference Sporns, Sträter and Minnerup12
Little is known about the use or uniformity of stroke protocols and stroke management in pediatric centers across Canada. Reference Rivkin, deVeber and Ichord13,Reference MacKay, Wiznitzer and Benedict14 The aim of this study is to describe the current practice of hyperacute stroke care across different pediatric institutions in Canada, specifically in centers that have a protocol for the management of stroke patients, and to descriptively analyze differences and similarities between protocols.
Methods
This was a cross-sectional, exploratory study designed to understand pediatric stroke management in Canada. As the study was deemed to be a quality improvement study, a Research Ethics Board approval was not required. A standardized email inquiring about the existence of a formal pediatric stroke protocol was sent to either 1) a pediatric neurologist at each Canadian pediatric hospital who has expertise or clinical interest in pediatric stroke or 2) the division or section head of pediatric neurology. All 16 Canadian pediatric tertiary care centers were approached and pediatric neurologists from those centers were eligible to participate. There were no exclusion criteria. The contact information of pediatric neurologists was obtained through each hospital’s website.
We set out a priori to identify if there were center differences in eight pre-defined criteria. Pediatric AIS protocols were collected and analyzed for similarities and differences in the following domains: 1) initial suspected stroke screening questions; 2) lab workup and neuroprotective care for children presenting with suspected AIS; 3) pre-treatment neuroimaging; 4) age cut-off for IV-tPA treatment; 5) exclusion/inclusion criteria for IV-tPA; 6) IV-tPA dosage (mg/kg dosing and maximal dose); 7) alternatives to IV-tPA; and 8) post IV-tPA care. Neuroprotective care is defined as maneuvers prior to tPA administration to minimize the risk of expansion of the ischemic area (e.g. maintaining normoglycemia and normothermia) as well as to reduce the risk of complications with tPA therapy (avoiding excessively high or low blood pressures). Post IV-tPA care is defined as laboratory investigations, imaging, and patient care following IV-tPA treatment. Following the analysis, we interviewed pediatric neurologists from the seven institutions with a formal protocol to understand the site-specific rationale for differences in the above items. We contacted the same neurologists from our initial correspondence through a non-structured and non-standardized email. We posed open-ended questions to understand the reasoning behind the following protocol components: 1) pre-treatment neuroimaging choice; 2) age cut-off for IV-tPA treatment; 3) exclusion criteria for IV-tPA; 4) IV-tPA dosage; 5) alternatives to IV-tPA. Email interview questions were generated by study staff. A second, non-structured and non-standardized follow-up telephone interview was arranged to clarify ambiguous information.
Results
Childhood Stroke Protocols in Canada
Pediatric neurologists from each of the 16 pediatric tertiary care hospitals replied to our inquiry. Seven out of 16 centers (44%) have an established stroke protocol and 2/16 (17%) have a protocol that is under development. Of the seven institutions that do not have a protocol, two redirect their patients to adult neurology, and five rely on a collaborative local expert approach to manage stroke without a protocol. An overview of stroke management across Canadian pediatric institutions is illustrated in Figure 1. All seven established pediatric stroke protocols were included in our analysis and were kindly provided by the following institutions: Alberta Children’s Hospital (Calgary), British Columbia Children’s Hospital (Vancouver), Children’s Hospital of Eastern Ontario (Ottawa), Centre Hospitalier Sainte-Justine (Montréal), Children’s Hospital at London Health Sciences Center (London), The Hospital for Sick Children (Toronto), and Stollery Children’s Hospital (Edmonton).
Stroke management across 16 pediatric Canadian pediatric hospitals.
Similarities Among Pediatric Stroke Protocols
The key recommendations presented in each protocol are similar and are explained in Table 1.
Overview of similarities between seven protocols
Differences in Pediatric Stroke Protocols
The differences noted in the seven protocols described below are illustrated in Figure 2 and explained in Table 2. The center-specific rationale for differences in stroke protocol criteria are outlined in Table 3.
Differences in stroke protocol domains.
Overview of discrepancies between seven protocols
* depending on which imaging modality is available quicker to qualify a child for IV-tPA in the 4.5-hour treatment window;
** CT if MRI is contra-indicated or not accessible quickly enough to qualify a child for IV-tPA in the 4.5-hour treatment windows.
Center-specific rationale for differences in stroke criteria
Pre-Treatment Neuroimaging
The differences in neuroimaging are as follows: 3/7 centers (43%) use a computed tomography (CT) scan as their first-line choice; 2/7 centers (28%) use magnetic resonance imaging (MRI) as their initial neuroimaging; 2/7 centers (28%) use either a CT scan or MRI. The centers that use CT as first line do so because it is more accessible and does not require sedation (N = 3). The centers that use MRI as the first-choice do so based on the recommendations from the Thrombolysis in Pediatric Stroke (TIPS) study (N = 2). Reference Rivkin, deVeber and Ichord13 The sites that choose either imaging modality do so because although their stated preference was to obtain an MRI, after-hours (evenings, weekends) studies are more limited due to the availability of an MRI technologist and anaesthesiologist, making a CT head available in a more timely manner (N = 2).
Exclusion Criteria for IV-tPA
We identified 3/7 centers (43%) to consider the Alberta Stroke Program Early CT Score (ASPECTS) when determining if a child is eligible for IV-tPA. The other 4/7 institutions (57%) do not use ASPECTS. ASPECTS is a 10-point qualitative score validated for adult AIS to evaluate the extent of early ischemic changes in the anterior circulation on non-contrast CT head. Reference Pexman, Barber and Hill15 Two centers that do not use ASPECTS scoring state that they do not because of their preference to obtain an MRI brain instead of a CT head. One center feels that ASPECTS scoring is redundant as ASPECTS correlates with the PedNIHSS score. One center explains that physicians are not comfortable with ASPECTS scoring in children.
We notice that 3/7 centers (43%) use a Pediatric NIH Stroke Scale (PedNIHSS) score of <4 as their lower cut-off, one center (14%) uses a cut-off of <5, and 3/7 centers (43%) use a cut-off of <6. The PedNIHSS is a tool that objectively quantifies the impairment caused by stroke, ranging from 0 to 42. Reference Boulanger, Lindsay and Gubitz7 Institutions with <4 and <6 cut-offs indicate that they followed the TIPS study criteria when selecting this value. The TIPS trial initially indicated a <6 cut-off before revising it to <4. Reference Rivkin, deVeber and Ichord13 The institute with cut-off of <5 indicates that this limit is not absolute and based on the discretion of the treating physician.
Age Cut-off for IV-tPA Administration
Five out of seven centers (72%) use IV-tPA starting at the age of 2 years, one center (14%) only administers IV-tPA starting at the age of 3 years, and one center (14%) only starting at age 10 years old. The five centers using a 2-year cut-off reference the TIPS study and the International Pediatric Stroke Study (IPSS) as their guideline. Avoidance of CT scans in younger children and lack of safety evidence for treating young children with IV-tPA are the rationale for protocols that have 3-year and 10-year cut-off ages, respectively.
IV-tPA Dosage
During our initial protocol analysis, 3/7 centers (43%) listed using an IV-tPA dose of 0.9 mg/kg, with a maximum dose of 60 mg. Two centers (28%) use a dose of 0.9 mg/kg, with a maximum dose of 90 mg. One center (14%) adjusts its dose based on age; children younger than 12 years old receive a dose of 0.75 mg/kg (maximum dose 75 mg) and children 12 years and older are administered 0.9 mg/kg (maximum dose 90 mg). However, during our teleconference discussion regarding protocol differences, 3/7 centers (43%) updated their maximum dosage from 60 mg to 90 mg to align with the best available evidence recently published in the medical literature. Reference Medley, Miteff and Andrews16 This yields 6/7 centers (86%) using a dose of 0.9 mg/kg, with a maximum dose of 90 mg. All seven institutions referred to the TIPS study criteria and adult dosing when choosing their IV-tPA dosage.
Alternatives to IV-tPA
Only 3/7 centers (43%) use intra-arterial thrombolysis (IA-tPA) for a subset of their pediatric stroke patients. All seven institutions consider mechanical thrombectomy. Three centers (43%) will perform mechanical thrombectomy within 6 hours of symptom onset. Two centers (28%) will extend their timeline of mechanical thrombectomy to 24 hours from symptom onset. Two centers (28%) will perform mechanical thrombectomy if the child presents outside of the 4.5-hour IV-tPA treatment timeframe, with no definite upper time limit for the thrombectomy procedure, ultimately leaving the decision to the treating team.
Discussion
This qualitative study shows that seven out of 16 Canadian pediatric tertiary care centers have established stroke protocols. Although these protocols have high concordance, several key differences were noted.
All protocols share initial neuroprotective measures, which require the normalization of temperature, blood glucose, oxygenation, blood pressure, and blood volume.
There were differences in age cut-off for IV-tPA use, ranging between 2 and 10 years of age. As there is no standardized guideline for IV-tPA use in children, centers referred to the TIPS and IPSS trial age criteria when creating their protocols. Reference Rivkin, deVeber and Ichord13,Reference MacKay, Wiznitzer and Benedict14 Although there is evidence of successful IV-tPA use in children as young as 2 years, the majority of studies report IV-tPA use in older children, which may explain the higher age cut-off at some Canadian centers. Reference Benedict, Ni, Schloesser, White and Bale17
The variability in IV-tPA dosage may be explained by the absence of randomized controlled trials for IV-tPA in the pediatric population. Recent published guidelines for childhood stroke management recommend a maximum dose of 0.9 mg/kg, with a maximum dose of 90 mg. Reference Medley, Miteff and Andrews16 Notably, during our preliminary investigation, 3/7 centers initially listed a maximum dose of 60 mg, which was then changed to 90 mg after our first teleconference.
Only three centers use IA-tPA in pediatric stroke patients. The lack of randomized controlled trials for IA-tPA in the pediatric population and the need for an interventional radiologist may explain why not all centers offer this treatment option. Reference Boulanger, Lindsay and Gubitz7 Numerous reports over the past two decades describe successful IA-tPA use in children. Reference Arnold, Steinlin and Baumann18 Of note, a recent consensus statement recommends thrombectomy over the use of IA-tPA for adults. Reference Powers, Rabinstein and Ackerson19
Our analysis revealed that the seven protocols differ in their choice of CT or MRI as the first-line imaging test to diagnose stroke. Given that AIS is a radiological diagnosis and the frequency of stroke mimics in children, timely neuroimaging findings must support the diagnosis of AIS prior to tPA administration. Reference Amlie-Lefond and Wainwright20 Although the American Academy of Neurology guidelines recommend the use of a diffusion MRI to detect AIS due to its increased sensitivity, CT imaging is more readily available in most hospitals. Reference Boulanger, Lindsay and Gubitz7 Access to an urgent MRI may be challenging in time-sensitive settings, particularly given the potential need for sedation in non-cooperative patients, including young, aphasic or anxious children. A non-contrast CT therefore remains the neuroimaging of choice if a diffusion-weighted MRI cannot be obtained within the appropriate timeframe for IV-tPA administration. Reference Bernard, Friedman and Stence8 This difference in pre-treatment neuroimaging among pediatric centers reflects the current stroke imaging debate in the adult population; specifically, what is “best” (MRI) versus what is “more practical” (CT). Reference Boulanger, Lindsay and Gubitz7
All institutions indicate the use of mechanical thrombectomy in their protocols, though the timeframe for the procedure varies between 6 and 24 hours. Although mechanical thrombectomy has not been extensively studied in children, it is successfully performed within 24 hours of symptom onset in adults. Reference Nogueira, Jadhav and Haussen21 A recent systematic review with data from the last 20 years suggested that mechanical thrombectomy is an effective option for the treatment of AIS in patients aged 1–18 years. Reference Bhatia, Kortman and Blair22 Ultimately, pediatric neurologists we contacted agreed that mechanical thrombectomy is patient- and situation-dependent, and the decision should be at the discretion of the treating team.
There are several limitations to our study. We did not inquire about the healthcare team and services involved in the initial triage of a patient with suspected AIS, the healthcare staff that trigger the stroke pathway, or the target time to neuroimaging at each institution. Additionally, our study did not examine the procedures performed post-tPA treatment to determine the cause of stroke. Furthermore, we did not investigate the frequency of stroke protocol use and the number of patients meeting IV-tPA criteria at the centers with a formal protocol. Future analyses of stroke protocols and response to management will be required.
Conclusion
Our study is unique as it describes the current landscape of childhood stroke management at Canadian pediatric hospitals. It is assuring that seven pediatric tertiary care centers have formal stroke pathways with extensive areas of agreement such as post-tPA care, though they differ in important aspects of diagnostic workup and treatment. Our work has already yielded amendments in stroke protocols at some centers – specifically regarding the maximum dosage of IV-tPA at three centers – and has sparked discussion regarding the need for further collaboration among Canadian institutions to optimize childhood stroke management. We advocate for a Canadian consensus regarding IV-tPA and EVT treatments with the goal of implementing and, where possible, standardizing pediatric stroke protocols.
Acknowledgements
We would like to acknowledge Dr. Gabrielle de Veber who provided the foundation for our understanding of pediatric AIS in Canada, and who provided training and mentorship for many pediatric neurologists.
We would like to thank all pediatric neurologists that responded to our emails and the questionnaire. We would also like to thank the following individuals who collaborated with us for this project: Ms. Sara Ieradi, Children’s Hospital of Eastern Ontario; Dr. Philippe Major, Centre Hospitalier Sainte-Justine; Dr. Christopher Marquis, Centre Hospitalier Sainte-Justine; Dr. Laurence Ducharme-Crevier, Centre Hospitalier Sainte-Justine; Dr. Nicolas Chrestian, Centre Hospitalier de L’Universitaire Laval; Dr. Hélène Dubreuil, Centre Hospitalier Universitaire Sherbrooke; Dr. Paula Brna, IWK Health Centre; Dr. David Buckley, Janeway Children’s Hospital and Rehabilitation Centre; Dr. Craig Campbell, London Health Sciences; Dr. David Callen, McMaster Children’s Hospital; Dr. Sébire Guillaume, Montreal Children’s Hospital; Dr. Salah Almubarak, Royal University Hospital; Dr. Mubeen Rafay, Winnipeg Children’s Hospital; Dr. Samantha Marin, Winnipeg Children’s Hospital.
Disclosures
This study was not sponsored and did not receive funding. MG received a summer scholarship to work at the Children’s Hospital of Eastern Ontario. AA has received grants from AMOSO, Brain Canada, and Canadian Institutes of Health Research, and has membership on board advisory for UCB Canada and speaker bureau for Eisai Canada Ltd. The remaining authors have no conflicts of interest to declare.
Statement of Authorship
MG wrote the first draft of the manuscript and was responsible for study design.
HJM, DP edited the manuscript, provided data and were responsible for study design.
AA, CB, IB, JM, AM, MM, SP, and JS provided data and a critical review of the manuscript.
All authors reviewed and approved the manuscript in its final form.
