Construction of predictive model for continuous renal replacement therapy in early stage of sepsis children
10.3969/j.issn.1673-9701.2025.32.004
- VernacularTitle:脓毒症患儿早期持续血液净化预测模型构建
- Author:
Xin YE
1
;
Qiyin CAI
1
;
Jiali HUANG
1
;
Qin LIU
1
;
Guanghui CHEN
1
;
Chaofu QIN
1
;
Ting LIANG
1
Author Information
1. 江门市妇幼保健院儿童重症监护室,广东江门 529000
- Publication Type:Journal Article
- Keywords:
Sepsis;
Continuous renal replacement therapy;
LASSO regression;
Prediction model
- From:
China Modern Doctor
2025;63(32):12-16
- CountryChina
- Language:Chinese
-
Abstract:
Objective To screen the predictors of early initiation of continuous renal replacement therapy(CRRT)in children with sepsis and construct a linear model,based on LASSO regression analysis.Methods A total of 55 children diagnosed with sepsis at Jiangmen Maternity and Child Health Care Hospital from April 2023 to February 2025.They were divided into CRRT group(n=17)and non-CRRT group(n=38)based on CRRT treatment usage.Using LASSO regression screening,predictive factors were identified and a Logistic regression model was established.The model's performance was evaluated using receiver operating characteristic(ROC)curve and calibration curve.Results There were significant differences in age,respiratory rate,body temperature,and mechanical ventilation between two groups(P＜0.05).Through LASSO regression analysis,four independent predictors of respiratory rate,body temperature,blood glucose,and C-reactive protein were identified.The constructed model demonstrated an area under the ROC curve of 0.94(95％CI:0.87-1.00),indicating good calibration accuracy.Conclusion The column line model based on body temperature,respiratory rate,blood glucose and C-reactive protein can effectively predict the need for early initiation of CRRT in sepsis children.
