Comparison Of Diagnostic Efficiency Between APRI And Three Risk Assessment Scales In Predicting Esophagogastric Variceal Bleeding Recurrence

Objective By retrospectively analyzing the clinical data of viral cirrhosis patients with esophagogastric variceal bleeding(EVB),to evaluate the clinical predictive value of aspartate amino transferase-to-platelet ratio index(APRI)in recurrent esophagogastric variceal bleeding.By comparing the clinical data between APRI,AIMS65 score,Glasgow-Blatchford score(GBS)and pre-endoscopy Rockall Score(PRS),to get a more accurate evaluation index,therefore to provide a guiding basis and a new selection of prediction tool in recurrent esophagogastric variceal bleeding for clinicians.Method A total of 104 patients with acute upper vascular gastrointestinal bleeding from June 2013 to December 2017 were chosen.According to the inclusion and exclusion criteria,89 cases were finally included in this study,including 52 males(average age 60.26±15.61)and 37 females(average age 56.33±17.17).30 patients(including 19 males,average age 52.35±13.12;11 females,average age 54.31±16.23)with EVB recurrence were classified as the case group,and 59 patients(including 33 males,average age 58.79±16.67,11 females,average age 57.12±17.77)without EVB recurrence were classified as the control group.The APRI,AIMS65,GBS,and PRS scores were calculated.Statistical analysis were performed by SPSS19.0 statistical software,comparison of measurement data between case group and control group were performed using Mann-Whitney U test(non-normal distribution)or t test(normal distribution),X2 test was used for qualitative data,Spearman correlation analysis was used for correlation analysis.The Med Calc software was used to draw and analyze the APRI,AIMS65,GBS,and PRS subjects’ receiver operating characteristic curve(ROC)and evaluate their predictive value for EVB recurrence.Results1.The mean value of APRI in the case group was 2.55±1.15,which was significantly higher than that of the control group 1.7±1.67(P<0.05).The mean values of PRS in the case group and control group were 1.00±0.56 and 0.41±0.58 respectively.The mean values of AIMS65 in case group and control group were 1.56±0.87 and 0.64±0.87 respectively.The mean values of GBS in case group and control group were 10.63±3.29 and 8.59±1.43 respectively.Statistical differences in APRI and scores of three score scales between the two groups were found(P<0.05).2.The correlations between APRI with AIMS65 scores,PRS scores,and GBS scores were compared.APRI was positively correlated with AIMS65,PRS,and GBS scores with statistical significance(P<0.05).The correlation coefficients were 0.248,0.158,and 0.431,respectively.3.The area under ROC of APRI,AIMS65,GBS,and PRS were 0.737,0.735,0.721,and 0.714,and the sensitivity was 0.847,0.793,0.777,and 0.692 respectively.cut-off value of predicting EVB recurrence was 2.4 for APRI,2 points for AIMS65 score,10 points for GBS score,and 1 point for PRS score,respectively.As regard to the ROC areas of three scoring scales,it was P=0.487,Z=0.626 between AIMS65 and PRS,P=0.563,Z=0.573 between AIMS65 and GBS,and P=1.019,Z=0.308 between GBS and PRS,respectively.The ROC area between APRI with AIMS65,GBS,PRS was compared.The results between APRI and AIMS65 was P=0.428,Z=0.793,between APRI and GBS was P= 0.263,Z=1.118,between APRI and PRS was P=0.758,Z=0.308.There was no significant difference in the predictive ability of EVB recurrence among each evaluation indicators.Conclusion 1.APRI has effective diagnostic and predictive value for EVB recurrence.2.APRI,AIMS65,GBS,PRS scores can all be used for the prediction of EVB recurrence,The APRI and AIMS65 scores have similar predictions and are better than GBS and PRS scores.3.For patients with restricted conditions,APRI is a better non-invasive predictive index of EVB recurrence.