Abstract
PURPOSE: Severe fever with thrombocytopenia syndrome (SFTS) has a high mortality rate and is easily misdiagnosed as hemorrhagic fever with renal syndrome (HFRS), particularly in resource-limited rural areas where early diagnosis remains challenging. This study used routine laboratory parameters, epidemiology and clinical manifestations to develop a model for the early diagnosis of SFTS and identify fatal risk factors, ultimately reducing mortality of SFTS. PATIENTS AND METHODS: This retrospective cohort study included 141 SFTS and 141 HFRS patients. Of these, 94 patients with SFTS were allocated to the model cohort for mortality risk identification by using multivariable Cox regression analysis. Sensitivity, specificity, and predictive values were calculated from validation cohort to assess the clinical values. Then, we analyzed 62 SFTS and 113 HFRS using multivariable logistic regression to identify SFTS. Receiver operating characteristic (ROC) curve analysis was used to evaluate their diagnostic value. RESULTS: Multivariate Cox regression analysis showed that blood urea nitrogen (BUN) ≥10.22mmol/L activated partial thromboplastin time (APTT) ≥58.05s and D-dimer ≥4.68mg/L were the risk factors for death in SFTS. This combined indicators had an area under the curve (AUC) of 0.91 (95% CI: 0.847-0.973), with a sensitivity and specificity of 86%, respectively. Any indicator was achieved the cutoff, and sensitivity and specificity in the validation group were 93% and 54%. Multivariable logistic regression showed that age (OR: 1.10) and initial laboratory indicators including WBC (OR: 0.48), Cr (OR: 0.86), CK (OR: 1.01), and APTT (OR: 1.09) can be used to identify SFTS from HFRS. This model achieved an AUC value of 0.97 (95% CI: 0.977-0.999) and 0.98 (95% CI: 0.958-1.000) in validation cohort. CONCLUSION: In resource-limited rural hospitals, the integration of routine laboratory parameters with epidemiology and clinical manifestations demonstrates enhanced sensitivity for early SFTS identification and mortality risk stratification to reduce mortality rate.