Infection-triggered encephalopathy syndromes: a meta-analysis of clinical characteristics and outcomes in 1946 cases.

Background: Infection-triggered encephalopathy syndromes (ITES) are acute, para-infectious disorders that can cause disability or death. Recognition is increasingly important in viral pandemics, but clinical features and outcomes remain poorly understood. Methods: PubMed search (inception to 6 March 2024) identified studies with published individual patient data (raw data were not collected from authors). The search was updated on 14 May 2026 using the same terms to identify reports and cases published after the data extraction cutoff. Data were extracted by paediatric neurologists using a standardised proforma. Major syndromes included acute encephalopathy with biphasic seizures and late reduced diffusion (AESD), acute necrotising encephalopathy (ANE), acute shock with encephalopathy and multiorgan failure (ASEM), hemiconvulsion-hemiplegia-epilepsy syndrome (HHE), febrile infection-related epilepsy syndrome (FIRES) and mild encephalopathy with reversible splenial lesion (MERS). We performed a pooled analysis of individual-level published data investigating patient characteristics, infections, and clinical outcomes measured by six-month modified Rankin Scale (mRS). Infection-syndrome associations were analysed with chi-square normalised residuals. Findings: 1946 patients from 656 studies (by ascending age of onset) included 164 ASEM (median age 0.78 years), 217 AESD (1.3 years), 95 HHE (2 years), 414 ANE (3.4 years), 562 FIRES (9 years), and 422 MERS (9.25 years). An updated search identified 658 cases from 171 studies that could be eligible for inclusion. AESD was linked to human herpesvirus 6 (z = 12.87); ANE with influenza A (z = 11.1), SARS-CoV-2 (z = 9.1) and influenza B (z = 4.3); MERS with rotavirus infection (z = 7.48); and FIRES with absence of microbiological identification (z = 18.2) (all p < 0.001).Neuroimaging was often delayed. Magnetic resonance imaging (MRI) brain restricted diffusion was the commonest finding, typically bilateral (except HHE). Characteristic patterns included thalamic involvement with or without haemorrhagic changes in ANE, corpus callosum involvement in MERS, and subcortical white matter involvement in AESD and HHE. Cerebrospinal fluid pleocytosis occurred mostly in FIRES and MERS, and elevated protein in ANE.Immunotherapy (commonly steroids, immunoglobulin) including biologics (anakinra, tocilizumab) was used most in ANE and FIRES. Acute mortality was 12% (227/1946), highest in ASEM (50%) and ANE (35%). There was good six-month outcome (mRS 0-2) in 52% (95% CI 50-55; 615/1174): MERS 99% (95% CI 97-100; 323/326), AESD 54% (95% CI 44-64; 50/93), FIRES 38% (95% CI 33-44; 120/314), ANE 33% (95% CI 27-38; 88/269), HHE 16% (95% CI 7-32; 5/32), and ASEM 15% (95% CI 9-24; 14/91). Interpretation: ITES showed distinct demographic, clinico-radiological features and outcomes. This largest ITES cohort to date highlights the importance of timely imaging, intervention, and future global collaboration to advance diagnosis, treatment, and pandemic preparedness. Funding: No funding was received for this work.