Highlights
Over 24 years in the Swiss Neuropediatric Stroke Registry, the proportion of children with acute ischemic stroke diagnosed after the 4.5-hour intravenous thrombolysis window declined, but the majority were still diagnosed too late for reperfusion therapy.
Median time from stroke onset to diagnosis remained long at 26.9 hours, with significant improvement in continuous diagnostic delay seen mainly for in-hospital stroke rather than out-of-hospital presentations.
Shorter diagnostic delay was associated with older age, greater stroke severity, and facial palsy, whereas nonspecific symptoms markedly prolonged time to diagnosis. Posterior-circulation symptom patterns were linked to missed thrombolysis windows.
Among children with out-of-hospital stroke, direct presentation to a stroke center was associated with shorter time to diagnosis, supporting regionalized pediatric stroke pathways.
Why this study matters
Pediatric acute ischemic stroke is rare compared with adult stroke, but its consequences are substantial. Children may develop lifelong motor, cognitive, behavioral, language, and psychosocial impairments. Timely diagnosis matters even more now than in prior decades because reperfusion therapies, once largely confined to adult care, are increasingly considered in selected pediatric patients. As this treatment landscape evolves, the historical problem of delayed recognition becomes a practical barrier to intervention.
Unlike in adults, pediatric stroke often presents with lower pretest probability in the minds of families, emergency clinicians, and even inpatient teams. Mimics such as migraine, seizures, Bell palsy, intoxication, infection, or functional symptoms are common competing explanations. Very young children may also be unable to describe focal deficits clearly. These realities make any long-term analysis of diagnostic delay highly relevant for both clinical systems design and bedside decision-making.
The Swiss study by Brechbühl and colleagues addresses this issue using nationwide registry data collected over more than two decades. Its main contribution is not simply to document persistent delays, but to identify where modest progress has occurred and which clinical features still predict late diagnosis.
Study design and methods
This was a nationwide retrospective cross-sectional analysis of children enrolled in the Swiss Neuropediatric Stroke Registry between 2000 and 2023. The investigators included 314 children aged 28 days to 16 years with acute ischemic stroke.
The primary outcome was time from stroke onset to diagnosis, abbreviated as TOD. Because eligibility for reperfusion is time-sensitive, the authors also examined two clinically meaningful binary outcomes: diagnosis beyond the intravenous thrombolysis window, defined as 4.5 hours or more from onset, and diagnosis beyond the thrombectomy window, defined as 24 hours or more.
Analyses were stratified by stroke onset location: out-of-hospital and in-hospital. This is an important design choice because the causes of delay in those two settings are probably different. Out-of-hospital delay often reflects symptom recognition, transport, triage, and initial hospital destination. In-hospital delay may instead reflect diagnostic overshadowing by the child’s primary illness, postoperative status, sedation, or comorbid complexity.
The investigators used multivariable logistic regression to assess temporal trends for diagnoses beyond reperfusion windows, and robust linear regression to analyze continuous TOD. Prespecified covariates were entered and retained in multivariable models if they showed associations in univariable analyses. The abstract reports adjusted odds ratios, beta coefficients, and 95% confidence intervals.
Key results
Overall diagnostic delay remained substantial
Across the full 2000-2023 study period, the median time from stroke onset to diagnosis was 26.9 hours, with a very wide interquartile range of 10.1 to 91.5 hours. Clinically, this means that even though some children were diagnosed relatively quickly, many were recognized only after a full day or longer. That level of delay places most patients outside established or emerging reperfusion windows.
Some improvement occurred for the thrombolysis window
The proportion of children diagnosed beyond 4.5 hours fell significantly over time in the overall cohort, from 90.9% to 77.5%. The adjusted odds ratio per calendar year was 0.94, with a 95% confidence interval of 0.88 to 1.00. This suggests a gradual annual reduction in the odds of missing the thrombolysis window.
The trend was driven largely by children with out-of-hospital stroke. In that subgroup, diagnoses beyond 4.5 hours declined from 88.1% to 74.1%, with an adjusted odds ratio per year of 0.91 and a 95% confidence interval of 0.84 to 1.00. Although the confidence interval approaches the null, the direction and magnitude are consistent with improved recognition, transport, and referral over time.
Even so, the practical message is sobering: by the end of the study period, about three-quarters of children with out-of-hospital stroke were still diagnosed too late for intravenous thrombolysis.
No clear improvement for the thrombectomy window
By contrast, no significant change was observed over time in the proportion diagnosed beyond the 24-hour thrombectomy window. This finding is important because endovascular treatment may offer a broader rescue opportunity than thrombolysis in selected large-vessel occlusions. Yet the study suggests that system-level gains have not been sufficient to consistently bring more children into that broader therapeutic timeframe.
This may reflect the continuing challenge of pediatric stroke recognition rather than failure of advanced stroke centers per se. If recognition and routing are delayed at the front end, children will continue to arrive too late regardless of downstream technical capability.
Continuous time-to-diagnosis improved mainly for in-hospital stroke
When the authors analyzed TOD as a continuous variable, significant improvement over time was seen only for in-hospital acute ischemic stroke. The beta coefficient was -4.3, with a 95% confidence interval from -7.2 to -1.5. In practical terms, the data suggest that each advancing calendar year was associated with shorter diagnostic delay for strokes occurring in already hospitalized children.
This is a notable signal. It implies that inpatient monitoring, access to imaging, or growing awareness among hospital teams may have improved over time. It may also reflect better recognition of hospital-associated stroke risk in children with cardiac disease, hematologic disorders, vasculopathies, systemic illness, or peri-procedural complications.
However, the absence of a parallel improvement in continuous TOD for out-of-hospital stroke indicates that prehospital and early emergency care remain major bottlenecks.
Which children were diagnosed faster?
Several patient-level features were associated with shorter time to diagnosis. Older age was linked to faster diagnosis, with a beta coefficient of -1.8 and a 95% confidence interval from -2.9 to -0.8. This aligns with clinical experience: older children can report symptoms more clearly, and focal deficits may be interpreted more readily as vascular rather than developmental or behavioral.
Greater neurologic severity, measured by pedNIHSS, was also associated with shorter TOD, with a beta coefficient of -1.2 and a 95% confidence interval from -2.1 to -0.4. More severe deficits are simply harder to miss.
Facial palsy was another marker of earlier recognition, associated with a beta coefficient of -19.4 and a 95% confidence interval from -30.2 to -8.5. This is clinically plausible because facial asymmetry is a familiar stroke cue in both professional and public education campaigns.
Which children were diagnosed later?
Nonspecific symptoms strongly prolonged time to diagnosis. The reported beta coefficient was 110.0, with a 95% confidence interval from 72.5 to 147.5. That is a striking effect size, underscoring how nonfocal or ambiguous presentations continue to derail timely recognition.
In addition, posterior-stroke symptoms were associated with diagnosis beyond the thrombolysis window. Posterior circulation stroke in children may present with dizziness, vomiting, imbalance, headache, altered mental status, diplopia, or other symptoms that overlap with common benign diagnoses. These presentations remain a recurrent diagnostic blind spot.
The role of destination hospital
For out-of-hospital stroke, time to diagnosis was shorter when children presented to a stroke center rather than other sites. This is one of the most actionable findings in the study. It suggests that pediatric stroke outcomes may be improved not only by clinician education but by optimizing care pathways so that suspected cases reach the right facility first.
Although the abstract does not provide the exact effect estimate for this variable, the directional finding supports direct-to-stroke-center routing when pediatric stroke is suspected and geography allows.
Clinical interpretation
The study offers a realistic picture of progress: some improvement, but nowhere near enough. The modest reduction in missed thrombolysis windows is encouraging, especially for out-of-hospital stroke. Yet the central fact remains that most pediatric stroke diagnoses still occur after reperfusion opportunities have passed.
Several clinical lessons emerge. First, symptom pattern matters. Classic anterior-circulation signs such as facial droop and severe focal deficits trigger faster recognition; vague or posterior symptoms do not. Second, younger children remain especially vulnerable to delay. Third, systems of care are crucial. Stroke-center presentation appears to shorten diagnostic timelines, which strengthens the argument for pediatric-inclusive stroke triage protocols.
For inpatient care, the improvement in continuous diagnostic delay is also important. In-hospital stroke is often underappreciated, particularly in children with complex medical conditions in whom neurologic change may be attributed to sedation, metabolic problems, infection, or postoperative recovery. The observed improvement suggests that institutional awareness and response processes may be modifiable and effective.
How these findings fit with current practice and literature
Current pediatric stroke guidance increasingly emphasizes rapid recognition, emergent neuroimaging, and multidisciplinary triage at experienced centers. The American Heart Association scientific statement on stroke in neonates and children highlighted the need for organized systems, timely neurovascular imaging, and specialized expertise in pediatric stroke care. More recent international consensus statements on pediatric stroke management have similarly acknowledged growing use of hyperacute therapies in selected children despite the limited trial base compared with adults.
The Swiss data fit this broader trajectory. As treatment possibilities expand, delayed diagnosis becomes the limiting factor. A hospital cannot deliver thrombolysis or thrombectomy to a child whose stroke is not recognized until the next day. Thus, diagnostic delay is not simply a descriptive metric; it is the rate-limiting step in translating stroke science into pediatric practice.
The finding regarding posterior-circulation symptoms deserves special emphasis. In both adults and children, posterior strokes are more likely to present atypically and be missed. In pediatrics, where dizziness and vomiting are common and usually benign, the threshold for considering stroke may be even higher. This study therefore supports targeted educational interventions that go beyond the familiar FAST paradigm and include gait disturbance, diplopia, severe vertigo, dysarthria, reduced consciousness, and acute ataxia.
Strengths of the study
The study has several strengths. It uses a nationwide registry, spans 24 years, and includes both out-of-hospital and in-hospital stroke, allowing a more nuanced analysis than many single-center reports. The use of clinically meaningful reperfusion thresholds enhances translational relevance. Stratification by onset location is particularly valuable because it distinguishes two fundamentally different diagnostic environments.
The registry-based design also captures real-world practice over time, which is exactly what clinicians and health systems need when assessing whether awareness campaigns and stroke pathways are changing care.
Limitations and cautions
As a retrospective observational analysis, the study cannot establish causality. Associations between symptoms or presentation site and diagnostic timing may reflect residual confounding. The abstract indicates that covariates were retained in multivariable models based on univariable associations, a pragmatic strategy but one that may not fully account for all clinically relevant confounders.
The rarity of pediatric stroke means sample size remains limited, especially for subgroup analyses across long calendar periods. Diagnostic and treatment practices likely evolved substantially between 2000 and 2023, including MRI availability, pediatric stroke awareness, telemedicine support, and endovascular expertise. These secular changes are part of the story but may also complicate interpretation.
Another limitation is that the abstract focuses on diagnosis time rather than imaging time, specialist consultation time, or treatment time. For hyperacute stroke workflows, each interval matters. A child diagnosed within 4.5 hours may still miss treatment if transfer or imaging is delayed. Likewise, the study does not address how many patients were ultimately eligible for or received reperfusion therapy.
Finally, the results come from Switzerland, a country with its own geography, transport systems, and hospital network. The general principles likely apply broadly, but specific effect sizes may differ in larger or more fragmented healthcare systems.
Implications for clinicians and health systems
For emergency and prehospital teams
Keep pediatric stroke on the differential for sudden focal deficits, acute speech disturbance, visual loss, ataxia, unexplained severe vertigo, or abrupt altered mental status, especially when symptoms are lateralizing or vascular in pattern. Posterior presentations should be treated with the same diagnostic seriousness as classic hemiparesis.
When feasible, suspected pediatric stroke should be discussed early with a stroke-capable center that has pediatric neurology and rapid neuroimaging access. The study’s findings support direct routing strategies rather than default presentation to the nearest nonspecialist facility when time and regional protocols permit.
For inpatient teams
Children already in the hospital can still have stroke, and diagnostic anchoring is a major hazard. Acute focal deficits after surgery, catheterization, extracorporeal support, severe infection, dehydration, sickle cell disease complications, or cardiac events should trigger immediate stroke evaluation. The observed improvement in in-hospital diagnostic timing suggests that local protocols, staff education, and activation pathways can work.
For pediatric neurologists and stroke program leaders
Education should focus not only on FAST-type symptoms but on pediatric-specific blind spots: younger age, nonspecific symptoms, and posterior circulation syndromes. Simulation, checklists, and pediatric stroke alert pathways may be useful. Quality improvement work should measure not just door-to-imaging time, but symptom-onset-to-recognition time across inpatient and outpatient settings.
Bottom line
This Swiss registry study shows that pediatric acute ischemic stroke diagnosis has become somewhat faster over the past 24 years, but not fast enough. Most children are still diagnosed after the window for thrombolysis, and many beyond the window relevant to thrombectomy. Younger children, those with nonspecific presentations, and those with posterior-circulation symptoms remain at greatest risk for delay. Direct presentation to stroke centers appears helpful, and in-hospital recognition has improved over time.
The practical takeaway is clear: if pediatric reperfusion care is to become more than a theoretical possibility, the next frontier is not only treatment capability but earlier recognition. Awareness of posterior-stroke signs, attention to stroke in younger children, and stronger direct-to-stroke-center pathways are likely to yield the greatest gains.
Funding and trial registration
The abstract provided does not report funding details or a ClinicalTrials.gov registration number. As this was a retrospective registry-based study, formal trial registration may not have been applicable. Readers should consult the full Stroke publication for complete funding and disclosure information.
References
Brechbühl D, Everts R, Goeggel-Simonetti B, Nava E, Wohlgemuth C, Bauder F, Oesch G, Fluss J, Bachmann M, Faignart N, Strozzi S, Muenger R, Darteyre S, Steinlin M. Diagnostic Delays in Pediatric Acute Ischemic Stroke: 24-Year Trends and Contributing Factors in Switzerland. Stroke. 2026-05-01. PMID: 42063408. Available at: https://pubmed.ncbi.nlm.nih.gov/42063408/
Ferriero DM, Fullerton HJ, Bernard TJ, Billinghurst L, Daniels SR, DeBaun MR, deVeber G, Ichord RN, Jordan LC, Massicotte P, Meldau J, Roach ES, Smith ER. Management of Stroke in Neonates and Children: A Scientific Statement From the American Heart Association/American Stroke Association. Stroke. 2019;50:e51-e96.
Mackay MT, Yock-Corrales A, Novak I, et al. Differentiating childhood stroke from mimics in the emergency department. Stroke. 2016;47:2476-2481.
Rivkin MJ, deVeber G, Ichord RN, et al. Thrombolysis in pediatric stroke study. Stroke. 2015;46:880-885.
