The Value Of APRI,GP In The Diagnosis Of Esophageal Variceal Bleeding In Patients With Liver Cirrhosis

Background&objectiveEsophageal variceal bleeding(EVB)is one of the main causes of death in patients with liver cirrhosis.Gastroscopy is the most commonly used method for evaluating EVB,but it is invasive and has the possibility of inducing variceal bleeding.Therefore,it is necessary to explore the accuracy of non-invasive indicators for evaluating EVB.This study was to investigate it could be predicted the occurrence of esophageal varices(EV)and ruptured hemorrhage through the indexes such as aspartate aminotransferase/platelet(APRI),globulin/platelet(GP),gamma-glutamyltransferase/platelet(GPRI)and aspartate aminotransferase/alanine aminotransferase(AAR)ratio from the blood routine and liver function in patients with liver cirrhosis.Methods1.A retrospective analysis was used to collect 134 patients with liver cirrhosis who were hospitalized in our department from January 2016 to January 2019.Gender,age,etiology,blood routine,liver function and gastroscopy results were recorded.According to the grading criteria of esophageal varices,all patients were divided into three groups:non-variceal group,mild/moderate variceal group and severe variceal group.According to the diagnostic criteria of esophageal variceal bleeding,134 patients with liver cirrhosis were further divided into bleeding group and non-bleeding group.The values of APRI,GP,GPRI and AAR of each group were calculated,and ROC curves were drawn by statistical software to evaluate their diagnostic efficiency.2.The statistical software IBM SPSS 25.0 was used to describe the classification data by Chi-square test,and the metrological data was described by mean±standard deviation.T-test(two groups)or one-way ANOVA(multi-group)were used to compare metrological data in accordance with normal distribution,and rank sum test was used in skewness distribution data.P<0.05,the difference was statistically significant.The ROC curve was drawn according to the gold standard of gastroscopy,and the predictive value of(AUC)was evaluated by the area(AUC)under the ROC curve.The diagnostic value of AUC between 0.5 and 0.7 was lower than that between 0.7 and 0.9,and the diagnostic value was medium in the range of 0.7 to 0.9.The diagnostic value is higher than 0.9.Results1.Among 134 patients with cirrhosis,22 had no esophageal varices,59 had mild/moderate esophageal varices and 53 had severe esophageal varices.There was no significant difference in gender,age and etiology among the three groups(P>0.05).2.The average APRI of the non-mild/moderate and severe esophageal varices group was 0.69±0.24,1.29±0.52 and 1.83±0.92,with statistical significance among the three groups(P<0.001);the average GP was 2.77±1.42,3.77±1.72,5.00±1,95,with statistical significance among the three groups(P<0.001);the average GPRI and AR were 0.63±0.59,0.78±0.81,0.76±0.75 and 1.56±0.49,respectively.There were no significant differences among the three groups(P=0.719),(P=0.555).3.The ROC curves of APRI and GP were drawn respectively.APRI prediction of EV AUC 0.90.APRI had moderate diagnostic value for esophageal varices.We further used APRI to predict mild to moderate and severe EV,and the AUC is 0.85 and 0.75 respectively,which told that APRI had moderate diagnostic value for the mild to moderate and severe esophageal varices.Similar to APRI results,GP predicted AUC of EV,mild to moderate and severe EV were 0.76,0.70 and 0.72,respectively,which showed that GP had moderate diagnostic value for esophageal varices,mild to moderate and severe esophageal varices.Further,APRI and GP were combined to predict EV,mild to moderate EV and severe EV.The AUC values of APRI and GP were 0.91,0.85 and 0.78 respectively,suggesting that combined detection has high diagnostic value for esophageal varices and moderate diagnostic value for mild to moderate and severe esophageal varices.4.Among 134 patients with cirrhosis,64 were in the hemorrhage group and 70 were in non-hemorrhage group.There was no significant difference in age,sex and etiology between the two groups(P>0.05).5.In the hemorrhage group and the non-hemorrhage group,the average APRI was 1.81+0.88 and 1.03+0.45 respectively,with statistical significance between the two groups(P<0.001);the average GP was 5.08+1.92 and 3.17+1.45 respectively,with statistical significance between the two groups(P<0.001);the average GPRI and AR in the hemorrhage group and the non-hemorrhage group were 0.74±0.77,0.76±0.75 and 1.58±0.63,1.72±0.78,respectively.There was no statistical difference(P=0.878),(P=0.264).6.The ROC curves of APRI and GP were drawn respectively.APRI predicted that AUC of EVB was 0.79,and the AUC of GP was 0.79.The combined prediction of APRI and GP showed that the AUC of predicting esophageal variceal bleeding was 0.85,which indicated that the combined detection of APRI and GP was no significant improvement in diagnosis of esophageal variceal bleeding.Conlusion1.APRI,GP is of moderate value in evaluating esophageal varices,mild to moderate,and severe esophageal varices in cirrhotic patients,and the combination of APRI and GP is highly valuable in predicting esophageal varices in cirrhotic patients.2.The value of APRI,GP in predicting esophageal variceal bleeding in cirrhotic patients was moderate.The combination of APRI and GP was no significant improvement in diagnosis of esophageal variceal bleeding.