Fatal Risk Factors and the Efficacy of Glucocorticoid Therapy in Severe Fever with Thrombocytopenia Syndrome: A Multicenter Retrospective Cohort Study

Fang Zhong¹Shiyu Zhang²Chengxi Zheng¹Dilihumaer Zhayier²Shuting Liu²Qinyu You¹Hongming Huang²Bin Zhu²Jin Tian²Zhiliang Hu^3,4,5Xin Zheng²Baoju Wang^2*Zhihang Peng^1,6*

• ¹School of Public Health, Nanjing Medical University, Nanjing, China
• ²Department of infectious Disease, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei Province, China
• ³Department of Infectious Diseases, The Second Hospital of Nanjing, Nanjing, Liaoning Province, China
• ⁴Nanjing Hospital of Chinese Medicine Affiliated to Nanjing University of Chinese Medicine, Nanjing, Jiangsu Province, China
• ⁵Global Health Center, School of Public Health, Nanjing Medical University, Nanjing, Jiangsu Province, China
• ⁶Chinese Center For Disease Control and Prevention, Beijing, China

Background: Severe fever with thrombocytopenia syndrome (SFTS) is an emerging tick-borne infectious disease characterized by rapid progression and high mortality. Glucocorticoids (GCs) can be used as anti-inflammatory agents for SFTS, but no standardized protocols have been proposed.Methods: A total of 901 patients with SFTS diagnosed at two hospitals between July 2017 and October 2023 were included in this retrospective cohort study. Univariate and multivariate logistic regression were performed along with LASSO regression to identify independent risk factors of fatal outcomes and further develop mortality prediction model. A nomogram was used to visualize the predictive model. ROC curves, calibration curves, and DCA curves were conducted to assess model accuracy and clinical applicability. The efficacy of GC was assessed using survival analyses, and further subgroup analyses of the effects of different GC regimens on fatal outcomes and hospital-acquired infections (HAI) were performed. Propensity score matching (PSM) analyses were conducted to control confounding factors. Results: Older age (age > 69 years), consciousness disturbance, decreased monocyte counts, prolonged activated partial thromboplastin time (APTT), and high viral load were identified as strong predictors of fatal outcomes in patients with SFTS. Patients were classified into mild and severe groups according to risk scores calculated by the nomogram (cut-off value = 121.43). Survival analyses showed that GCs treatment may reduce the mortality in severe patients (p = 0.004). Further subgroup analyses indicated that relatively high doses and early treatment with GCs may increase mortality in SFTS patients [OR = 2.292 (1.071, 5.066); OR = 3.693 (1.710, 8.345) respectively]. GCs treatment was associated with an elevated risk of HAI in patients both with mild and severe SFTS (p = 0.024; p = 0.015, respectively). Initiation of GCs therapy at a low level of aspartate aminotransferase (AST < 189.75 U/L) reduced the mortality before and after PSM (p<0.001; p = 0.004, respectively).Conclusions: A new nomogram based on five independent risk factors effectively predicts the prognosis of SFTS. Severe patients and those with low AST levels might benefit from GCs therapy while early and relatively high doses of GCs therapy should be used with caution.