A Comparative Study Of The Prognostic Effects Of Anatomic Versus Nonanatomic Hepatectomy On Hepatocellular Carcinoma Based On Microvascular Invasion Prediction

backgroundMicrovascular invasion(MVI)of HCC can significantly affect the long-term survival of patients,and is the most important factor leading to postoperative recurrence.Preoperative accurate prediction of MVI status is very important for the selection of surgical methods,which is conducive to improving tumor recurrence and long-term prognosis.However,the diagnosis of MVI can only be made by pathological examination of the specimen after surgery,which is difficult to be detected even by preoperative puncture biopsy,thus limiting its clinical use.The use of serum markers alone or imaging data(including CT,MRI and PET-CT)to predict MVI status in patients with HCC has insufficient diagnostic efficacy and lack of effective clinical validation.Therefore,this study combined serum markers and liver cancer imaging features to establish a predictive model of MVI,which was transformed into an MVI risk scoring system,and patients were treated with either anatomical hepatectomy(AR)or non-anatomical hepatectomy(NAR)to provide the best evidence.Purpose1.Construct a mixed microvascular invasion prediction model based on the enhancement of NMR hepatocellular carcinoma specific signs and serum markers based on disodium gadolinium cetonate.2.To further analyze the scope of application of anatomical liver resection(AR)and non-anatomical liver resection(NAR)under known MVI states through predictive models.MethodAccording to the following criteria,228 patients with primary liver cancer surgery in the Department of Hepatobiliary Surgery of Zhujiang Hospital from January 2012 to June 2018 were included.Fifty-two patients were randomly selected as the verification cohort,and the remaining 172 patients were the experimental cohort.After being approved by the Ethics Committee of Zhujiang Hospital of Southern Medical University,the clinical data of patients were collected retrospectively.Single-factor and multi-factor logistic regression analysis was used to screen out the risk factors related to MVI,and to construct a predictive model and verify the model.Then,Kaplan-Meier survival curve analysis was used to compare the recurrence rate and survival rate of HCC in the AR group and the NAR group based on the MVI risk in the prediction model.ResultUsing laboratory hematology test indicators and gadoxetate disodium to enhance the logistic regression analysis of the specific signs of nuclear magnetic hepatocellular carcinoma,four risk factors related to MVI were screened:alpha-fetoprotein(AFP)>15 ng/mL(OR:5.411;95%CI:2.093,13.990;P<0.001);Neutrophil lymphocyte ratio(NLR)>3.8(OR:3.97;95%CI:1.689,9.368;P=0.002),coronary enhancement(CE)(OR:6.183;95%CI:2.478,15.429;P<0.001),low uptake around hepatobiliary specific tumors(PH-HBP)(OR:8.754;95%CI:3.355,22.843;P<0.001);MVI prediction model MVI risk=1.5 ×AFP+1 × NLR+2 × CE+2 × PH-HBP,the AUROC of the prediction model in the original cohort and validation cohort are 0.884(95%CI:0.829,0.938)and 0.899(95%CI:0.821,0.967).Taking the MVI prediction model score of 3.75 as the cutoff value,they were divided into MVI high-risk group and MVI low-risk group.The median recurrence time in the high-risk group was 18 months,and the median recurrence time in the low-risk group was 28 months.The difference was statistically significant(P=0.003)The AR group had higher risk of MVI or low risk of MVI but tumor diameter>5 cm compared with NAR The recurrence rate and mortality rate of patients in the group is low;and when the patient is in a low-risk MVI state with a tumor size?5cm,there is no difference in the recurrence rate and mortality rate between AR and NAR.Conclusion1.A simple and accurate MVI prediction model for liver cancer was established in this study.2.The stratified analysis of the prognosis of AR and NAR in patients with HCC by MVI prediction showed that AR was beneficial to improve the clinical prognosis for patients with high-risk MVI(regardless of tumor diameter)or low-risk MVI but tumor diameter>5 cm.3.For patients with low-risk MVI and tumor diameter?5cm,AR or NAR has no significant difference in clinical prognosis.