Retrospective analysis of severe fever with thrombocytopenia syndrome and construction of a nomogram prediction model for mortality risk factors

Abstract Objective: To investigate high mortality risk factors in severe fever with thrombocytopenia syndrome (STFS) and to create a nomogram model for personalized prediction. Methods: 523 SFTS patients admitted to the Second Hospital of Nanjing, Nanjing University of Chinese Medicine, between January 2020 and December 2023 were retrospectively analyzed. 75 cases were classified in the death group (D group) and 448 cases in the survival group (S group). A predictive nomogram model was developed base on the independent risk factors that were stepwise screened through univariate analysis, least absolute shrinkage and selection operator (LASSO), and multivariate logistic regression analysis. Results: Based on stepwise variable screening by univariate analysis, LASSO, and multivariate logistic regression. Age(OR=1.06; 95%CI, 1.03–1.10; P＜0.001), hemorrhagic symptoms (OR=3.39; 95%CI; 1.31–8.78; P=0.012), neurologic symptoms (OR=4.89; 95%CI, 2.72–8.77; P＜0.001), platelet (OR=0.99; 95%CI, 0.98-0.99; P=0 .045), PT (OR=1.32; 95%CI;1.11-1.56; P=0.001), APTT (OR=1.02; 95%CI, 1.01–1.03; P=0.007) and viral load ≥107copies/ml(OR=2.66; 95%CI; 1.36 – 5.20; P =0.004) were independent mortality risk factors in patients with SFTS. The area under the curve (AUC) showed excellent predictive power (AUC = 0.87, 95% CI 0.832-0.909). Calibration curves showed the accuracy of the nomograms assessed. Decision curve analysis (DCA) results showed a greater net benefit when the threshold probability of patient death was between 0.02 and 0.75. Conclusions: A nomogram model consisting of seven risk factors was successfully constructed, which can be used to predict STFS mortality risk factors early.