Serum GPVI for Diagnostic Assessment and Short-Term Risk Stratification in Patients with Suspected Transient Ischemic...

Background/Objectives: Transient ischemic attack (TIA) is a neurological emergency associated with a substantial early risk of ischemic stroke, yet its diagnosis remains clinically challenging because objective biomarkers are limited. Glycoprotein VI (GPVI), a platelet collagen receptor involved in platelet activation and thrombus formation, may reflect prothrombotic activity in cerebrovascular ischemia. This study aimed to evaluate the diagnostic assessment and short-term risk stratification value of serum GPVI levels in patients presenting to the emergency department with suspected TIA. Methods: This prospective observational cohort study included 85 adult patients with transient focal neurological symptoms suggestive of TIA. Patients were classified according to the Precise Diagnostic Score (PREDISC), and early stroke risk was assessed using the ABCD2 score. Serum GPVI levels were measured using enzyme-linked immunosorbent assay. Patients were followed for cerebrovascular events at 2 and 7 days. Associations between GPVI levels, clinical scores, and early outcomes were analyzed, and diagnostic performance was assessed using receiver operating characteristic analysis. Results: Serum GPVI levels differed significantly across PREDISC categories and increased from the "TIA unlikely" group to the "TIA very likely" group (p < 0.001). GPVI levels were also higher in patients with ABCD2 scores ≥4 than in those with lower scores (p < 0.001). GPVI showed positive correlations with both PREDISCs (Spearman's r = 0.682, p = 0.001) and ABCD2 scores (Spearman's r = 0.469, p = 0.001). All early cerebrovascular events occurred in the high-risk ABCD2 group. Patients who experienced cerebrovascular events had higher baseline GPVI levels, with a small-to-moderate effect size for 2-day outcomes (r = 0.22) and a moderate effect size for 7-day outcomes (r = 0.36). ROC analysis demonstrated discriminative performance for identifying patients classified as "TIA very likely" according to the PREDISC system, with an area under the curve of 0.821 (95% CI: 0.71-0.91). Conclusions: Serum GPVI levels were associated with PREDISC-based clinical TIA likelihood, higher ABCD2-defined risk categories, and early cerebrovascular events in patients with suspected TIA. These findings suggest that GPVI may have a potential complementary role within structured emergency department-based diagnostic assessment and short-term risk stratification frameworks, pending validation in larger independent cohorts. However, because of the exploratory single-center design, modest sample size, limited number of outcome events, and absence of external validation, these findings should be interpreted as hypothesis-generating and require confirmation in larger multicenter studies.