Department of Clinical Laboratory, The First Affiliated Hospital of Anhui Medical University, Hefei, Anhui, China; Department of Pathogen Biology and Provincial Laboratories of Pathogen Biology and Zoonoses, Anhui Medical University, No. 81 Meishan Rd, Hefei, China.
Department of Clinical Laboratory, The First Affiliated Hospital of Anhui Medical University, Hefei, Anhui, China.
J Infect Public Health. 2023 Mar;16(3):393-398. doi: 10.1016/j.jiph.2023.01.007. Epub 2023 Jan 16.
Severe fever with thrombocytopenia syndrome (SFTS) is an emerging infectious disease with high mortality. Early identification of patients who may advance to critical stages is crucial. This investigation aimed to establish models to predict SFTS before it reaches the critical illness stage.
Between January 2016 and September 2022, 278 cases have been included in this study. There were 87 demographic and systemic chosen variables. For selecting the predictive variables from the cohort, the LASSO was utilized, and for identifying independent predictors, multivariate logistic regression was performed. Based on these factors, a nomogram was established for critical illness. Concordance index values, decision curve analysis and the area under the curve (AUC) were also examined.
Multivariate logistic regression demonstrated the most important differentiating factors as;> 65 years old (P < 0.001, OR 3.388, 95 % CI 1.767-6.696), elevated serum PT (P = 0.011, OR 6.641, 95 % CI 1.584-31.934), elevated serum TT (P = 0.005, OR 3.384, 95 % CI 1.503-8.491), and elevated serum bicarbonate (P = 0.014, OR 0.242, 95 % CI 0.070-0.707). The C-index of the nomogram was 0.812 (95 % CI: 0.754-0.869), representing good discrimination. The model also showed excellent calibration. The AUC of the nomogram established based on four factors, as mentioned earlier, was 0.806. Furthermore, the model had the excellent net benefit, as revealed by the decision curve analysis.
An accurate risk score system built on manifestations noted in patients with SFTS upon admission to hospital, might be advantageous in managing SFTS.
严重发热伴血小板减少综合征（SFTS）是一种具有高死亡率的新发传染病。早期识别可能进展为危急阶段的患者至关重要。本研究旨在建立预测 SFTS 达到危急病期之前的模型。
本研究纳入了 2016 年 1 月至 2022 年 9 月期间的 278 例病例。选择了 87 个人口统计学和系统变量。为了从队列中选择预测变量，使用了 LASSO，为了确定独立预测因子，进行了多变量逻辑回归。基于这些因素，建立了一个用于确定危急病期的列线图。还评估了一致性指数值、决策曲线分析和曲线下面积（AUC）。
多变量逻辑回归表明最重要的区分因素为：年龄>65 岁（P<0.001，OR 3.388，95%CI 1.767-6.696）、血清 PT 升高（P=0.011，OR 6.641，95%CI 1.584-31.934）、血清 TT 升高（P=0.005，OR 3.384，95%CI 1.503-8.491）和血清碳酸氢盐升高（P=0.014，OR 0.242，95%CI 0.070-0.707）。列线图的 C 指数为 0.812（95%CI：0.754-0.869），表示具有良好的区分度。该模型还显示出良好的校准度。基于前面提到的四个因素建立的列线图的 AUC 为 0.806。此外，决策曲线分析表明，该模型具有优异的净获益。
基于 SFTS 患者入院时的表现建立的准确风险评分系统，可能有利于 SFTS 的管理。
