Objective To construct a predictive model for post-hepatectomy liver failure(PHLF)in patients with hepa-tocellular carcinoma(HCC)based on the liver function imaging score(FLIS)obtained by gadoxetic acid-enhanced mag-netic resonance imaging(MRI)hepatobiliary phase images,and to validate the diagnostic efficacy for PHLF.Methods General clinical data,pathological information,and gadoxetic acid-enhanced MRI images were collected from a total of 927 HCC patients who underwent radical hepatectomy.The patients were randomly divided into training set(n=647)and vali-dation set(n=280)according to the proportion of 7∶3.Multivariate Logistic regression analysis was conducted to screen independent risk factors associated with the occurrence of PHLF,and a visual nomogram prediction model was established.Receiver operating characteristic curve(ROC)was plotted to evaluate the predictive performance of the model for PHLF.Results The Logistic regression analysis of the model revealed that decreased FLIS score(OR=0.456,P＜0.001),major resection(OR=1.702,P=0.023),prolonged prothrombin time(OR=1.229,P=0.009),and low platelet count(OR=0.996,P=0.005)were closely related to the occurrence of PHLF and were independent predictors of PHLF.The predictive model constructed based on these four independent factors achieved an AUC value of 0.791 in the training set and 0.772 in the validation set.Conclusion The PHLF prediction model based on FLIS score,major resection,prothrombin time,and platelet count demonstrates high diagnostic performance,and can effectively guide clinical preoperative evaluation and pre-operative decision-making.
Magnetic resonance imagingHepatocellular carcinomaPost hepatectomy liver failurePredictive model
NETL
NSTL
万方数据
