| ObjectiveA model was constructed to assess the risk of Esophageal and Gastric Varices Bleeding(EGVB)in patients with liver cirrhosis using Two-Dimension-Shearwave Elastography(2D-SWE)of the spleen,liver and spleen ultrasound indicators,and laboratory indicators.MethodsPatients with liver cirrhosis who underwent esophagogastroduodenoscopy at the Department of Hepatobiliary Medicine of the First Hospital of Jilin University between January 1 and December 31,2022 were included.Endoscopic results were used as the gold standard to divide patients into low and high risk groups for Esophageal and Gastric Varices Bleeding(EGVB).Clinical data(gender,age,etiology,presence of ascites,platelets and liver function laboratory indicators),liver and spleen ultrasound indicators,and spleen Two-Dimension-Shearwave Elastography(2D-SWE)data were collected.Statistical analysis was performed using R 4.1.2 package to screen for meaningful indicators for assessing EGVB risk.A multimodal diagnostic model was constructed through stepwise logistics regression.The diagnostic value of each statistically significant indicator and the newly established multimodal model for EGVB risk was evaluated using Receiver Operating Characteristic(ROC)curves.The Area Under the Curve(AUC)of each ROC curve and the optimal cut-off value for each indicator to determine EGVB risk were calculated,along with their corresponding sensitivity and specificity.Results1.Comparison of general data(gender,age,etiology,presence of ascites)between the high-risk and low-risk groups showed no statistically significant difference(P>0.05).;2.Comparison of the average value of spleen stiffness(SSmean)between the high-risk and low-risk groups showed a significant difference(P<0.001).Correlation analysis showed that platelet(PLT),cholinesterase,spleen length(SL),spleen width(SW),spleen size(SS),spleen vein diameter(SVD),and peak systolic velocity of splenic artery(SAS)were significantly correlated with SSmean(P<0.05),with correlation coefficients of-0.327,-0.306,0.255,0.292,0.272,0.320,and 0.373respectively;3.Comparison of liver and spleen ultrasound indicators between the high-risk and low-risk groups showed significant differences in spleen length(SL),spleen width(SW),spleen size(SS),spleen vein diameter(SVD),spleen vein velocity(SVV),and peak systolic velocity of splenic artery(SAS)(P<0.05);4.Comparison of laboratory indicators between the high-risk and low-risk groups showed significant differences in platelet(PLT),cholinesterase,albumin(ALB),and prothrombin time(PT)(P<0.05);5.Univariate logistics regression analysis showed that increased platelet and albumin were protective factors for high risk of EGVB;increased prothrombin time(PT),spleen width(SW),spleen size(SS),spleen vein diameter(SVD),spleen vein velocity(SVV),peak systolic velocity of splenic artery(SAS),and average value of spleen stiffness(SSmean)were risk factors for high risk of EGVB(P<0.05).Multivariate stepwise logistics regression showed that increased SSmean was an independent risk factor for high risk of EGVB.The optimal model included albumin(ALB),SVD,and SSmean,with the equation 1.9275×SSmean(Kpa)+ 0.6282×SVD(mm)-0.3891×ALB(g/L)-19.2074.An ROC curve was constructed for this model with a cut-off value of 0.7694 for predicting high-risk EGVB.The area under the curve was 0.9673,with a sensitivity and specificity of 92.2% and 91.7%,respectively.Conclusions1.Increased platelet and albumin were protective factors for high risk of EGVB(P<0.05),with regression coefficients of-0.019 and-0.137 respectively;increased average value of spleen stiffness(SSmean),prothrombin time(PT),spleen width(SW),spleen size(SS),spleen vein diameter(SVD),spleen vein velocity(SVV),and peak systolic velocity of splenic artery(SAS)were risk factors for high risk of EGVB(P<0.05),with regression coefficients of 0.628,0.096,0.070,0.398,0.211,0.029,and1.455 respectively,with SSmean being an independent risk factor for high risk of EGVB;2.For predicting high risk of EGVB using a single indicator,SSmean had the highest predictive value,with an area under the curve of 0.896 when using SSmean>12.89 Kpa to predict high risk EGVB,with a sensitivity and specificity of100% and 75% respectively;3.The multimodal model including spleen stiffness,spleen vein diameter,a nd albumin(1.9275×SSmean(Kpa)+0.6282×SVD(mm)-0.3891×ALB(g/L)-19.2074>0.7694)can serve as a good indicator for predicting EGVB risk,with an area under the curve of 0.9673 for predicting high-risk EGVB,with a sensitivity an d specificity of 92.2% and 91.7% respectively,superior to any single indicator prediction model. |
PDF Full Text Request