ABSTRACT NUMBER: ESOC2026A2091 HUMAN SYSTEMIC BIOMARKERS FOR THE EARLY DIAGNOSIS OF ACUTE ISCHEMIC AND HEMORRHAGIC STROKE: A SYSTEMATIC REVIEW AND META-ANALYSIS

Abstract Background and aims Accurate early differentiation between acute ischemic stroke (AIS) and intracerebral hemorrhage (ICH) is essential to guide management and improve outcomes. Numerous blood-based systemic biomarkers have been investigated, but the evidence is fragmented and rapidly evolving. We performed an updated systematic review and meta-analysis of diagnostic blood biomarkers in acute stroke to support emergency triage decisions. Methods We included studies enrolling adults (≥18 years) with acute neurological symptoms suggestive of stroke, where at least one systemic biomarker was measured within 24 hours of symptom onset or last known well. Biomarkers with sufficient data were pooled with random-effects models for two diagnostic questions: (1) ICH vs AIS; and (2) stroke (ICH and/or AIS) vs non-stroke. Results From 14,406 records, 614 full texts were assessed and 146 studies included. Across included studies, 128 distinct biomarkers were evaluated for ICH vs AIS discrimination and 47 for stroke vs non-stroke. However, only 16 biomarkers, including GFAP, S100B, MMP-9, D-dimer, IL-6, and NT-proBNP, had sufficient data for meta-analysis. These were pooled using random-effects meta-analyses of standardized mean differences, yielding 19 comparisons (10 for ICH vs AIS; 9 for stroke vs non-stroke) from 34 studies. GFAP also had sufficient threshold data for HSROC analysis (ICH vs AIS). Conclusions Despite substantial literature, no blood-based biomarker is routinely used in clinical stroke care, underscoring the need for stronger validation and ongoing discovery of novel biomarkers and multimarker panels. Future diagnostic accuracy studies should prioritize head-to-head comparisons, standardized reporting (sampling time, assay platform, thresholds), and robust external validation across cohorts. Conflict of interest All authors: nothing to disclose.