Nomogram Model Of Posthepatectomy Liver Failure For Hepatocellular Carcinoma

Objective Primary liver cancer is a common malignant tumor in the world,which is characterized by high heterogeneity and poor prognosis.Every year,more than half of the new cases and deaths of liver malignant tumors in China account for more than half of the world,most of which are hepatocellular carcinoma(HCC),which brings heavy medical and social burden to our country.Hepatectomy is the most important treatment for HCC.In China,most patients diagnosed with HCC are accompanied by varying degrees of liver cirrhosis or liver function damage caused by chronic infection of hepatitis B(HBV),which increases the risk of various liver-related complications after hepatectomy.The most serious complication after hepatectomy is posthepatectomy liver failure(PHLF).The existing clinical prediction systems such as Child-Pugh systems,model for end-stage liver disease(MELD),indocyanine green(ICG)metabolism test and albumin-bilirubin score(ALBI)all have some limitations.In this study,a PHLF prediction Nomogram model based on ICG-R15 was established by using the method of Nomogram and a variety of indexes.Methods The clinical data of 219 patients with hepatocellular carcinoma who underwent radical hepatectomy in the The Second Affiliated Hospital of Anhui Medical University from February 2017 to March 2022 were analyzed retrospectively,including133 cases of open surgery and 86 cases of endoscopic surgery.all the data were used with the written informed consent of the patients or their families,which is in line with medical ethics.Under the condition of excluding biliary obstruction,when the total bilirubin(TB)combined with INR increased on or after the 5th day after operation and was higher than that before operation,PHLF was diagnosed.The independent risk factors affecting the occurrence of PHLF were screened by Logistics regression,and the Nomogram model for predicting the occurrence of PHLF was established according to these factors.The prediction efficiency of the model was tested by drawing the subject operating characteristic curve(ROC)curve,calculating the C-index,drawing the internal verification of the calibration curve,decision curve analysis(DCA),and comparing with ICG-R15,ALBI score and MELD system.Results In this study,ICG-R15(OR=1.07,95%CI:1.01～1.14,P<0.05)、PLT(OR=0.99,95%CI:0.99～1.00,P<0.05)、INR(OR=1.50,95%CI:1.06～2.11,P<0.05)、HBV-DNA>1000IU(OR=2.26,95%CI:1.06～4.80,P<0.05)、AFP>400ng/L(OR=2.60,95%CI:1.20～5.65,P<0.05)、Open surgery(OR=0.26,95%CI:0.11～0.64,P<0.05)are independent risk factors for PHLF.The threshold of ROC analysis is 140.6,and the standard curve in the calibration chart fits well with the predictive calibration curve.C-index of Nomogram model(0.821,95%CI:0.760～0.881)is significantly higher than ICG-R15(C-index=0.656,95%CI:0.570～0.742),ALBI score(C-index=0.701,95%CI : 0.625 ～ 0.777)and MELD system(C-index=0.617,95%CI:0.526～0.709).DCA also shows that this model is more reliable than ICG-R15,ALBI score and MELD system.Conclusion In this study,ICG-R15,PLT,INR,HBV-DNA > 1000,AFP > 400ng/L and open hepatectomy are independent risk factors for the occurrence of PHLF.Based on these risk factors,a Nomogram model is established to predict the occurrence of PHLF.Compared with the traditional model,this model has stronger predictive ability in the population in this study.For patients with elevated ICG-R15,laparoscopic surgery is safer.After entering a large sample retrospective study of too many centers,the ICG-R15 safety threshold of laparoscopic hepatectomy can be further improved.