Mediation Analysis of Serum Fibroblast Growth Factor 20 and Poor Prognosis After Ischemic Stroke

Abstract

Background We aimed to examine the relationship between serum FGF20 (fibroblast growth factor 20) levels and stroke prognosis in a multicenter cohort study. Methods and Results Patients with ischemic stroke/transient ischemic attack were prospectively recruited from 5 participating centers and followed up at 3 months and 1 year. FGF20 levels were measured using the ELISA method. The primary outcome was poor stroke functional outcome (modified Rankin Scale score of 3–6), and secondary outcomes included death and composite vascular events. Multivariable logistic regression analysis or Cox proportional hazards regression analysis was employed to estimate the relationship between FGF20 and study outcomes. Mediation analysis was conducted to examine the mediating effects of traditional risk factors on the association between FGF20 and stroke outcomes. A total of 1011 patients with ischemic stroke were included in the study. After adjusting for potential confounding factors, an elevated serum FGF20 level was associated with a reduced risk of the poor outcome and death. Multivariable adjusted spline regression analysis demonstrated a linear correlation between serum FGF20 levels and the stroke outcomes. The incorporation of FGF20 alongside conventional risk factors marginally enhanced the reclassification of adverse outcomes. Renal function and white blood cell count partially mediated the relationship between FGF20 and the prognosis of ischemic stroke. Conclusions Elevated FGF20 level is associated with decreased risks of adverse outcomes after ischemic stroke, which was partially mediated by renal function and white blood cells with a modest amount, indicating that serum FGF20 might serve as a promising biomarker for predicting stroke prognosis. Registration URL: https://www.chictr.org.cn ; Unique identifier: ChiCTR2100051104.