Pathogenic characteristics of ACLF combined with infection and establishment of a prediction model
MA Ya'nan
DONG Jingjing
WANG Sumei
GAO Yingtang
Abstract：Objective To explore the pathogenic characteristics and risk factors of infection in patients with acute-on-chronic liver failure(ACLF)and to construct a predictive model.Methods A total of 385 ACLF patients were en-rolled,and pathogens were identified before initiating antimicrobial therapy.Infection-related data were collected.The least absolute shrinkage and selection operator(LASSO)regression was used to select optimal predictive variables,which were then analyzed using multivariate Logistic regression to determine independent risk factors.A nomogram prediction model was developed based on these factors.The model's discriminative ability was assessed by calculating the area under the curve(AUC)of the receiver operating characteristic(ROC)curve.Calibration was evaluated using calibration curves and the Hosmer-Lemeshow test.Clinical utility was examined through decision curve analysis(DCA).Results Among 385 ACLF patients,169(43.90%)had concurrent infections,predominantly caused by Gram-negative bacteria(66.20%),with Escherichia coli being the most common(22.54%).Independent risk factors for infection included pro-longed hospitalization(OR=1.061,95%CI:1.030-1.095),ascites(OR=2.739,95%CI:1.579-4.826),procalcito-nin(PCT)≥0.5 ng/mL(OR=2.003,95%CI:1.200-3.374),C-reactive protein(CRP)≥8 mg/L(OR=5.125,95%CI:2.755-10.000),and elevated creatinine(Cr)levels(OR=1.004,95%CI:1.000-1.008)(all P<0.05).The no-mogram model based on these factors showed an AUC of 0.786,indicating good discrimination.Calibration curves aligned closely with the diagonal line,and the Hosmer-Lemeshow test confirmed good fit(P>0.05).DCA demonstrated superior net benefit within a threshold probability range of 0-0.91 compared to"intervention for all"or"intervention for none"strat-egies.Conclusions Infections in ACLF patients are primarily caused by Gram-negative bacilli(especially Escherichia coli).The nomogram model incorporating independent risk factors(prolonged hospitalization,ascites,PCT≥0.5 ng/mL,CRP≥8 mg/L,and elevated Cr)exhibits excellent discrimination,calibration,and clinical utility.
Keywords：acute-on-chronic liver failureinfectionrisk factorsprediction modelnomogram
Publication Date：2025-08-25
Online Publishing Date：2025-09-10(First online date of this platform, not the publication date of the document)
Pages：6( 36-41 )