BMC infectious diseases · 2025
Prediction models for encephalitis and in-hospital death associated with severe fever with thrombocytopenia syndrome.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Wu Xiaoxin, Tan Juntao, Zhao Yuxi, Niu Jianhua, Cen Xinting, Feiz Mohammadjavad, Tang Songjia, Zhang Zhengyu
Аннотация
BACKGROUND: Severe fever with thrombocytopenia syndrome (SFTS) is an emerging infectious disease associated with encephalitis and high in-hospital death rates. This study aimed to develop and validate two predictive models for the early identification of encephalitis and prevention of in-hospital death in SFTS patients.
METHODS: This retrospective study included SFTS patients admitted to the First Affiliated Hospital of Zhejiang University School of Medicine between January 2017 and July 2024. To construct the prediction models, we utilized least absolute shrinkage and selection operator (LASSO) coupled with logistic regression, subsequently presenting the results through a comprehensive nomogram. The performance of the models was rigorously evaluated across multiple dimensions: discrimination ability was measured by the area under the receiver operating characteristic curve (AUROC), calibration accuracy was assessed through calibration curves and Brier scores, and clinical applicability was determined using decision curve analysis (DCA) and a clinical impact curve (CIC).
RESULTS: The analysis of data from 196 SFTS patients revealed significant clinical outcomes, with an in-hospital death rate of 16.84% (33 patients) and an encephalitis incidence rate of 45.41% (89 patients). The in-hospital death prediction model and the encephalitis prediction model demonstrated robust performance, achieving AUROCs of 0.935 (95% confidence interval [CI]: 0.884-0.986) and 0.895 (95% CI: 0.839-0.952), respectively. Additionally, the Brier scores for these models were 0.066 and 0.120, respectively, indicating a high level of accuracy in their predictions. The DCA and CIC further revealed that both models achieved a relatively greater net benefit, suggesting their potential clinical applicability. Notably, age and ferritin levels were identified as the most significant variables coselected by both models, highlighting their importance in predicting outcomes.
CONCLUSION: The predictive models developed in this study demonstrated robust performance in predicting both in-hospital death and encephalitis incidence among SFTS patients. Notably, age and ferritin levels emerged as significant independent risk factors, showing strong associations with both clinical outcomes.
METHODS: This retrospective study included SFTS patients admitted to the First Affiliated Hospital of Zhejiang University School of Medicine between January 2017 and July 2024. To construct the prediction models, we utilized least absolute shrinkage and selection operator (LASSO) coupled with logistic regression, subsequently presenting the results through a comprehensive nomogram. The performance of the models was rigorously evaluated across multiple dimensions: discrimination ability was measured by the area under the receiver operating characteristic curve (AUROC), calibration accuracy was assessed through calibration curves and Brier scores, and clinical applicability was determined using decision curve analysis (DCA) and a clinical impact curve (CIC).
RESULTS: The analysis of data from 196 SFTS patients revealed significant clinical outcomes, with an in-hospital death rate of 16.84% (33 patients) and an encephalitis incidence rate of 45.41% (89 patients). The in-hospital death prediction model and the encephalitis prediction model demonstrated robust performance, achieving AUROCs of 0.935 (95% confidence interval [CI]: 0.884-0.986) and 0.895 (95% CI: 0.839-0.952), respectively. Additionally, the Brier scores for these models were 0.066 and 0.120, respectively, indicating a high level of accuracy in their predictions. The DCA and CIC further revealed that both models achieved a relatively greater net benefit, suggesting their potential clinical applicability. Notably, age and ferritin levels were identified as the most significant variables coselected by both models, highlighting their importance in predicting outcomes.
CONCLUSION: The predictive models developed in this study demonstrated robust performance in predicting both in-hospital death and encephalitis incidence among SFTS patients. Notably, age and ferritin levels emerged as significant independent risk factors, showing strong associations with both clinical outcomes.