Influenza-Associated Neurological Complications in Pediatric Patients: Clinical Characteristics, Outcomes, and Risk Stratification Using the Neuroinfluenza Severity Score  from the Multicenter TURK-INF Study

Technical Summary: Influenza-Associated Neurological Complications in Pediatric Patients

Problem Statement
Influenza-associated neurological complications (IANC) represent a heterogeneous spectrum of conditions in pediatric populations, ranging from benign febrile seizures to severe encephalopathy with high morbidity and mortality. While the prevalence of IANC in hospitalized children is estimated between 1.7% and 16.9%, existing literature is largely limited to single-center or small regional studies. Consequently, there is a lack of comprehensive data regarding the clinical characteristics, management strategies, and long-term outcomes of pediatric IANC. Furthermore, there is no validated, practical tool for early risk stratification to identify patients at high risk for poor neurological outcomes, long-term sequelae, or death.

Methodology
This study employed a retrospective, multicenter design involving 21 tertiary referral centers across 12 cities in Türkiye. The cohort consisted of 397 children (aged 1 month to 18 years) hospitalized with laboratory-confirmed influenza (via RT-PCR or rapid antigen testing) who presented with acute neurological manifestations or developed them within 48 hours of admission during the 2023–2024 and 2024–2025 influenza seasons.

Key methodological components included:

• Data Collection: Standardized extraction of demographic data, comorbidities, vaccination history, viral subtypes, clinical presentations, laboratory findings (including CSF analysis where available), neuroimaging (MRI/CT), and treatment protocols (antivirals, immunomodulators, antiepileptics).
• Definitions: IANC was defined as neurological symptoms unexplained by other causes in confirmed influenza cases. Complications were classified into seizures, encephalopathy, meningoencephalitis, Acute Disseminated Encephalomyelitis (ADEM), and others. Encephalopathy was further stratified into "mild" (recovery within 24 hours) and "severe" (prolonged impairment, ICU need).
• Outcome Measures: Primary outcomes were assessed at hospital discharge and at a six-month follow-up. A "good outcome" was defined as complete neurological recovery; a "poor outcome" included death or neurological sequelae (defined as a Pediatric Cerebral Performance Category [PCPC] score ≥ 2).
• Risk Stratification Tool: The authors developed the Neuroinfluenza Severity Score (NISS), a scoring system assigning one point for each of the following criteria: status epilepticus, encephalopathy lasting >24 hours, PICU admission, mechanical ventilation, high-risk neuroimaging features (e.g., thalamic/basal ganglia involvement, brainstem involvement, ANE pattern), and use of immunomodulatory therapy.
• Statistical Analysis: ROC curve analysis was utilized to evaluate the predictive performance of the NISS for unfavorable neurological outcomes.

Key Contributions

1. Large-Scale Multicenter Data: The study provides the most extensive characterization of pediatric IANC to date from the Turkish region, covering a cohort of 397 patients across diverse geographic and clinical settings.
2. Development of NISS: The primary novel contribution is the creation and validation of the Neuroinfluenza Severity Score (NISS). This tool integrates clinical, radiological, and therapeutic variables to stratify disease severity.
3. Long-Term Follow-Up: Unlike many previous studies focusing on acute outcomes, this research includes structured six-month follow-up data, offering insights into persistent neurological sequelae.
4. Phenotypic Characterization: The study delineates distinct clinical phenotypes, specifically identifying a "severe neuroinfluenza" phenotype characterized by delayed antiviral initiation, systemic inflammation, and specific neuroimaging patterns.

Results

• Epidemiology: Among 5,093 hospitalized children with influenza, 397 (7.8%) developed neurological complications. The median age was 45.6 months, with 70.8% of patients under five years old. Seizures were the most common manifestation (70.8%), followed by encephalopathy (10.9%) and meningoencephalitis (5.0%).
• Outcomes: Complete neurological recovery occurred in 90.2% of patients. However, 5.0% developed long-term neurological sequelae, and the overall mortality rate was 1.3%.
• Predictive Performance of NISS: ROC analysis demonstrated that the NISS has excellent predictive performance for unfavorable neurological outcomes. An optimal cutoff of NISS ≥ 2 yielded 82.1% sensitivity and 91.9% specificity (AUC 0.913).
• Severe vs. Mild Disease: Patients classified as having severe neuroinfluenza (NISS ≥ 2) were significantly associated with:
  • Older age (≥5 years) and higher rates of comorbidities.
  • Delayed initiation of oseltamivir.
  • Prolonged hospitalization and higher rates of PICU admission and mechanical ventilation.
  • Elevated inflammatory markers (leukocytes, ferritin, liver enzymes) and CSF abnormalities.
  • Significantly worse outcomes: 29.5% developed long-term sequelae (vs. 0.6% in non-severe) and 8.2% mortality (vs. 0% in non-severe).
• Clinical Nuances: Notably, 25.9% of patients presented without respiratory symptoms. While Influenza A was predominant, no significant difference in neurological outcomes was found between Influenza A and B. High-risk neuroimaging features, particularly bilateral thalamic involvement (ANE pattern), were strongly associated with severe disease.

Significance and Claims
The authors assert that IANC should be suspected in children with acute neurological manifestations even in the absence of respiratory symptoms. The study claims that the NISS serves as a practical, validated tool for early risk stratification, enabling clinicians to identify patients at increased risk for poor outcomes, long-term sequelae, and mortality. By identifying these high-risk patients early, the score facilitates timely escalation of care, including prompt antiviral therapy and consideration of immunomodulatory treatments.

The paper emphasizes that while the overall prognosis for IANC is generally favorable, the burden of severe disease is substantial. The findings highlight the critical need for early recognition, timely neuroimaging, and vaccination strategies, noting that the low vaccination rate represents a preventable gap in public health. The study concludes that the NISS provides a framework for improving clinical decision-making and patient management in the context of pediatric neuroinfluenza.