Font Size: a A A

Analysis Of Risk Factors Of Bleeding In Patients With Liver Failure And Evaluation Of Predictive Models

Posted on:2021-10-18Degree:MasterType:Thesis
Country:ChinaCandidate:B Y LvFull Text:PDF
GTID:2494306470476584Subject:Clinical Medicine
Abstract/Summary:PDF Full Text Request
Objective: Patients with liver failure are often accompanied by complex coagulation system dysfunction,and the corresponding reduction of procoagulant factors and anticoagulation factors in the patient’s coagulation system is considered to be in an unstable "rebalance" state.Although INR is one of the diagnostic criteria for liver failure,its predictive value for hemorrhage risk in patients with liver failure remains to be discussed.This study attempts to analyze hemorrhage risk factors in patients with liver failure and integrate various risk factors to establish a predictive model.Methods: This study retrospectively collected the results of 249 patients with liver failure admitted to the ICU ward of Tianjin Second People’s Hospital from January 2014 to December 2018.According to the inclusion criteria and exclusion criteria,65 cases were excluded.Among the remaining 184 cases,47 cases were randomly selected as the validation set,and the remaining 137 cases were divided into a bleeding group and a non-bleeding group.Collect the type of liver failure,etiology,whether hemorrhage,bleeding site,blood Na,TBIL,ALB,coagulation-related parameters,platelet-related parameters at admission or diagnosis of liver failure.In addition,patients in the bleeding group collected the above parameters within 24 hours of bleeding.Choose t test or non-parametric test according to the characteristics of the variables to compare the difference between the above parameters in the bleeding group and the non-bleeding group,and the difference between the bleeding group admission or diagnosis of liver failure and bleeding within 24 hours,through multivariate logistic regression Analyze the risk factors of bleeding in the course of liver failure and build a predictive model of bleeding in the course of liver failure.Result: The incidence of hemorrhage in acute(subacute)liver failure group,chronic plus(sub)acute liver failure group,and chronic liver failure group were 17.65%,38.37%,and 47.06%,respectively,and there were significant differences in the incidence of bleeding between the three types(P=0.049).There were statistically significant differences in ALB,TBIL,PT,TT,INR,PLT,and PCT between the bleeding group and the non-bleeding group(P <0.05).ALB and TBIL in the bleeding group at admission or diagnosis of liver failure were less than those in the non-bleeding group Decreased,PT,TT prolonged,INR increased,PLT,PCT decreased.There were 47 cases of hemorrhage in the course of liver failure,accounting for 34.3%.Compared with liver failure,PT,APTT and TT were prolonged,INR and PDW were increased,FIB,PLT and PCT were decreased and blood coagulation was increased.The parameters related to platelets have deteriorated.Incorporate ALB,TBIL,classification,TT,INR,PT,PLT,PCT into the regression equation for Logistic regression analysis(Forward: wald method),screen out the risk factors of bleeding in patients with liver failure,and build a joint prediction model Logit(P)= 0.041(PT)-0.111(ALB)+1.735.Regression analysis showed that PT,ALB and hepatic failure patients had a greater correlation with bleeding,but INR,PLT,PCT were not significantly correlated with bleeding.Plot the ROC curves of PLT,PT,INR,and Logit P.The area under the curve is 0.607,0.658,0.661,and 0.677.Among them,Logit P has the highest predictive power,with a sensitivity of 72.3%,a specificity of 56.7%,and a cutoff value of-0.833.The PLT,PT,INR and the area under the Logit P ROC curve were compared by the Z test,and the difference was not statistically significant(P> 0.05).Conclusion: Among the commonly used biochemical,platelet and coagulation parameters in clinic,prolonged PT and reduced ALB at the time of admission or diagnosis of liver failure are risk factors for bleeding in the course of liver failure;the regression equation established by PT and ALB as the main parameters for bleeding in patients with liver failure Compared with INR,the difference is not statistically significant,and the current commonly used clinical indicators are less predictive of hemorrhage in patients with liver failure.In addition,PT,APTT,TT,INR,FIB,PLT and other indicators of liver failure should be further worsened and heed bleeding.
Keywords/Search Tags:liver failure, bleeding, predictive model, coagulation dysfunction, anticoagulation
PDF Full Text Request
Related items