Multi-biomarker score model for predicting fatal outcomes in severe fever with thrombocytopenia syndrome: a multicenter cohort study.

Background: Severe fever with thrombocytopenia syndrome (SFTS) is a rapidly progressive disease with high mortality. This study aims to identify mortality risk factors in SFTS and create a prognostic model for early high-risk patient identification. Methods: A total of 301 SFTS patients were enrolled between May 2012 and July 2023 from nine clinical centers across the East China region. Using principal component analysis and cox regression, we identified independent risk factors for mortality and constructed the multi-biomarker score model (1 point was assigned when age ≥60 years, AST/ALT ratio ≥2.23, BUN ≥5.6 mmol/L, ALP ≥ 68 U/L, or ALB <33.3 g/L). Results: Of the 301 patients with SFTS, 57 (18.9%) experienced fatal outcomes during hospitalization. The risk of mortality escalated with each additional point on the multi-biomarker score, with a HR of 2.04 (95% CI, 1.60-2.61). Patients were stratified into low (0-1), intermediate (2-3), and high (4-5) risk groups based on their multi-biomarker scores. Notably, those in the high-risk category were at an over eightfold increased risk of mortality (HR, 8.59; 95% CI, 2.63-28.05). High scores (4-5) were also predictive of adverse outcomes, including secondary bacterial infection, meningitis, ICU admission, heart failure, respiratory failure, and renal failure. The receiver operating characteristic (ROC) curve analysis for the multi-biomarker score revealed an AUC of 0.769 (95% CI, 0.632-0.787), suggesting a cutoff value above 3 as the threshold for optimal discrimination. Conclusions: SFTS patients with more than three of the following criteria-age ≥60 years, AST/ALT ratio ≥2.23, BUN ≥5.6 mmol/L, ALP ≥68 U/L, or ALB <33.3 g/L-show a significantly elevated mortality risk.