The Value Of Noninvasive Test To Predicting Risk Of Esophageal Varices In Patients With Hepatitis B Cirrhosis

BackgroundCirrhosis is a chronic,damaging alteration of the liver that progresses to the advanced stages of many liver diseases and is a chronic disease with high mortality.Portal hypertension(PHT)as major complications of liver cirrhosis,esophageal varices(EV)is it important and serious complications.They are often clinically diagnosed due to bleeding from acute digestive tract varices.Digestive endoscopy is currently the"gold standard"for the diagnosis of esophageal varices.However,most patients undergoing screening endoscopy may have no varicose veins and no preventive treatment is required.Therefore,to seek an accurate and noninvasive means to predict esophageal varices can make most patients,especially low-risk patients with cirrhosis of the liver can effectively avoid unnecessary invasive examination.Currently,a variety of noninvasive methods have been proposed,including laboratory indicators and ultrasound technology,but also some new models proposed.Acoustic Radiation Force Impulse(ARFI)is an emerging ultrasound elastography technique that is immune to obesity and ascites and displays liver ultrasound in real time.However,the independent measurement parameters can not achieve satisfactory results in predicting the esophageal varices in patients with cirrhosis.Therefore,in recent years,new models have been proposed to better predicted the occurrence of cirrhosis with EV.In this study,we not only used the new ultrasound technique such as ARFI,but also combined it with laboratory indexes and ultrasound index.We compared the ROC curves of patients with cirrhosis have EV,and get the most critical value of its diagnostic performance,at present there is no systematic report about this.AimAssess whether LS measured with ARFI elastography can be used as a predictor of the presence of EV in patients with cirrhosis.Verification combined with other ultrasound and laboratory indicators,can more accurately assess the risk of EV in patients with cirrhosis.Logistic regression analysis was performed on the collected data to screen independent risk factors influencing the occurrence of cirrhotic EV.To explore the clinical value of ultrasound techniques,laboratory indexes and the combination of the two in predicting the risk of developing cirrhosis EV.The ROC curve was used to develop Non-invasive monitoring of cirrhosis of the EV the best critical value for clinical reference.MethodsA total of 142 patients with hepatitis B cirrhosis who meet the diagnostic criteria of cirrhosis of the chronic hepatitis B guidelines were enrolled,of which 15 were excluded.A total of 127 patients were enrolled,the gender,age,height,and weight of each patient were recorded and according to the endoscopy results of the subjects,patients with gastroesophageal varices were assigned to the esophageal varices group(EV group,n=50).patients without esophageal varices Patients were assigned to non-esophageal varices group(NEV,n=77).All subjects were strictly fasting 8-12hours after taking venous blood for the detection of liver function,coagulation routine,blood and conventional immunization.The same sonographer used the Siemens Acuson S2000 ultrasound system to record the measurement data by performing routine sonography and ARFI elastography on each patient.Analyze the data using SPSS 23.0 for Windows and MedCalc Software.Comparison of groups using independent samples t-test continuous variables,and the use ofχ~2 test to compare classification variables.Endoscopy results as the gold standard,multivariate Logistic regression analysis including univariate Logistic regression analysis of P<0.05 variables,screening out the risk factors for the occurrence of esophageal varices.The discriminative ability of the different noninvasive methods and models for identification of EVs and NEVs was assessed with receiver operating characteristic(ROC)curve analysis and expressed as the area under the ROC curve(AUROC).Comparison between AUROCs was made using the DeLong test,choose the best cut-off value is better to exclude and regulate EV patients.P<0.05 is defined as statistically significant.Result1,EV and NEV of parameters comparisonThe Child-Pugh grading of NEV group was significantly lower than EV group.The hepatic portal vein diameter,spleen portal vein diameter,spleen diameter were significantly larger in EV group than those in NEV group.The diameter of right lobe of EV group was significantly smaller than that of NEV group.The values of SWV of spleen and SWV of liver in EV group were significantly higher than those in NEV group.The AST in EV group was significantly higher than that in NEV group,and the platelet count in EV group was significantly lower than that in NEV group.There was no significant difference between NEV group and EV group in gender,age and BMI.2,Logistic regression analysis of the risk factors influencing the occurrence of EVThe diameter of spleen,SWV of liver,SWV of spleen and platelet count had statistical significance,and the OR values were 2.459,4.142,1.964 and 0.942respectively.3,The Value of Noninvasive test to Predicting Risk of Esophageal Varices in CirrhosisThe ROC curve of SWV of spleen,SWV of liver and spleen diameter and APRI、ASPS were plotted.The area under the curve(AUC)and 95%confidence interval were 0.952(95%CI:0.899-0.982),0.893(95%CI:0.826-0.941),0.886 and0.989(95%CI:0.952-0.999).The best cutoff value of Youden index was determined as the critical point,which were 11.8cm,1.89m/s and 2.81m/s,0.702 and 2.87respectively.Conclusion1.Child-Pugh classification of liver function,diameter of right hepatic lobe,diameter of hepatic portal vein,diameter of splenic portal vein,diameter of spleen,diameter of spleen,SWV of liver,SWV of spleen,AST,platelet count plays a role in determining the risk of developing cirrhosis of the liver.2.Logistic regression analysis showed that spleen diameter,SWV of liver and SWV of spleens were independent predictors of EV.Platelet count was protective factor of EV.3.According to ROC curve analysis,ASPS showed great efficacy in predicting the risk of EV in cirrhosis.However,when the conditions limited,APRI is also a good choice for diagnosis of EV.It is suggested that ASPS model can be a new and effective model to diagnose the risk of developing EV,and its clinical application should be further discussed.