Search for a command to run...
Stroke remains a leading cause of disability and mortality worldwide, with early and reliable prediction of stroke severity remaining a critical clinical challenge. This systematic review aims to evaluate the prognostic value of admission peripheral eosinophil counts in predicting stroke outcomes, including poor functional outcome, mortality, post-stroke infection, and symptomatic intracranial hemorrhage. We conducted a Preferred Reporting Items for Systematic reviews and Meta-Analyses-compliant search of PubMed, Embase, Web of Science, and Google Scholar up to February 2025 using keywords and MeSH terms for “eosinophil” and “stroke,” supplemented by reference screening. Adult cohort studies reporting admission eosinophil counts with relevant outcomes were included. Studies involving non-stroke populations, without eosinophil data, and non-original research were excluded. Data extraction captured study characteristics, eosinophil metrics (absolute count vs . percentage), cut-offs, timing of blood draw, and clinical outcomes. Methodological quality was assessed using the Newcastle–Ottawa Scale and Risk Of Bias In Nonrandomized Studies of Interventions (ROBINS-I), with most studies rated as high quality on Newcastle–Ottawa Scale and moderate risk of bias overall. Random-effects meta-analyses were conducted in R using the Hartung–Knapp–Sidik–Jonkman method and restricted maximum likelihood. Primary outcomes were poor functional outcome (modified Rankin Scale score ≥ 3), mortality, symptomatic intracranial hemorrhage, and post-stroke infection; effect estimates were reported as odds ratios (ORs) with 95 % confidence intervals (CIs), and 95 % prediction intervals were calculated. Subgroup analyses compared absolute eosinophil counts versus percentages. Sensitivity analyses were performed restricting to studies reporting adjusted estimates, and leave-one-out influence analyses assessed robustness of pooled results. A total of 21 cohort studies ( n = 22,930), predominantly ischemic stroke patients, met the inclusion criteria. Eosinopenia was not consistently associated with poor functional outcomes (OR = 2.05; 95 % CI: 0.59–7.11; I ² = 90 %) or mortality (OR = 3.07; 95 % CI: 0.86–10.93; I ² = 81 %), and no clear association was observed with post-stroke infection (OR = 1.41; 95 % CI: 0.21–9.36; I ² = 83 %). However, low eosinophil levels were significantly associated with symptomatic intracranial hemorrhage (OR = 4.53; 95 % CI: 4.23–4.85; I ² = 0 %). Sensitivity analyses using adjusted estimates and subgroup analyses by eosinophil metric yielded consistent findings. Admission eosinopenia is a potential adjunctive biomarker for identifying patients at risk of hemorrhagic complications, particularly after reperfusion therapy. Evidence for associations with functional outcomes, mortality, or infection is of low certainty, reflecting heterogeneity and methodological limitations. Eosinophil counts should be used to complement, not replace, established predictors, and further prospective, standardized studies are warranted to validate their clinical utility. The protocol for this review was registered with the International Prospective Register of Systematic Reviews under registration number CRD420251007761 on March 09, 2025.