| Backgrouds:It is critical of early prognosis of acute on chronic liver failure?ACLF?which has high mortality and acute deterioration.However,there are some inhernt limitations of the prognostic model which was widely used now in clinic because of the causes and factors of basic diseases,especially in the assessment of HBV-associated acute-on-chronic liver failure?HBV-ACLF?in China.Therefore,it is critical to establish an efficient prognostic model for the particular disease based on the clinical features and parameters of HBV-ACLF to guide and optimize the treatment of patients and improve the prognosis.Recently,in the study of prognostic markers of ACLF,inflammatory and immuno-related markers were most conspicuous.Therefore,this study based on the systemic inflammatory markers associated with chronic liver disease and the cytokines related to HBV-ACLF explored the potential clinical value of systemic inflammatory markers and cytokines in the prognostic evaluation of HBV-ACLF respectively.On this basis,a new Logistic regression model for the prognosis of HBV-ACLF was established based on the combined inflammatory markers and cytokines.The clinical value of the new model in evaluating the prognosis of HBV-ACLF was discussed.Firstly,we found the indicators which has statistical differences between three groups by analyzing these indicators and the levels of cytokines in the patients with HBV-ACLF,CHB and HC.Two multiariable Logistic regression models were established after applying differential inflammatory markers and cytokines into the single factor and multi-factor regression analysis.The clinical value of the prognostic model was evaluated through Receiver operating characteristic curve?ROC?analysis.Secondly,in order to improve the prediction performance,we established a new model to predict short-term prognosis in patients with HBV-ACLF through the integration of the cytokines and inflammatory markers based on blood routine parameters.Then it can guide and optimize the treatment of patients with HBV-ACLF which can provide powerful basis to improve the prognosis.Objective:Based on the viewpoint of immune inflammation,We intergrated Red blood cell distribution width?RDW??Neutrophil-to lymphocyte ratio?NLR??Red cell distribution width to platelet ratio?RPR??Lymphocyte monocyte ratio?LMR??Platelet to lymphocyte ratio?PLR??Tumornecrosis factor-??TNF-???Interleukin-6?IL-6??Interleuk in-8?IL-8??Interleukin-10?IL-10??Interleukin-17?IL-17?and traditional liver disease clinical parameters to establish model for the prognostic evaluation of patients with HBV–ACLF.The predictive value of the new model was discussed in this study.Methods:200 cases of HBV-related ACLF patients,100 cases of chronic hepatitis B patients who were hospitalized in the Second Affiliated Hospital of AnHui Medical University from April 2015 to March 2017 and 100 cases of healthy control were enrolled into this study.We retrospectively analyzed their clinical information and auxiliary examination data,and followed-up the prognosis of patients with ACLF at least 3 months after the time at admission.The expression of IL-8mRNA in Peripheral blood mononuclear cells?PBMC?of the patients with ACLF was analyzed by the application of fluorescence quantitative polymerase chain reaction?PCR?.At the same time the level of IL-8 in serum was analyzed using enzyme-linked immune sorbent assay technique?ELISA?.Then we can verify whether ELISA which we used to assay cytokine levels can reflect the immune status of the body.After that the ELISA method was applied to detect the level of TNF-??IL-6?IL-8?IL-10?IL-17 of the study subjects and the control.The relationship between inflammatory markers,the level of these cytokinesand the traditional clinical parameters,MELD score were comprehensive analyzed.The single factor and multiple factors analysis were used to evaluate the factors that may affect the prognosis of HBV-ACLF by the method of binary Logistic regression.Then we set up two multiariable Logistic regression model from the perspective of systemic inflammatory markers and cytokines respectively.The clinical value of the model in evaluating the prognosis of HBV-ACLF was evaluated by ROC analysis.Then,in order to further improve the prediction of model,we integrated the systemic inflammatory indexes based on blood routine parameters and the cytokines to establish a new model for the short-term prognostic evaluation of patients with HBV–ACLF.At the same time the cut-off point was determined.110 cases of HBV-related ACLF patients who were hospitalized in the Second Affiliated Hospital of AnHui Medical University from January 2010 to March 2015 were enrolled into the validation group.We retrospectively analyzed their clinical information and auxiliary examination data and we detected levels of cytokines of patients using ELISA method.We followed-up the prognosis of patients with ACLF at least 3 months after the time at admission and finally we verify the predictive performance of the model by ROC analysis.Results:Compared with CHB patients and HC,the RDW?16.69±2.49%?,NLR?5.56±5.51?,RPR?0.21±0.15?and LMR?3.89±2.56?values of HBV-ACLF were significantly increased.And the levels of TNF-??8.01±4.31pg/ml??IL-6?384.84±50.94pg/L??IL-8?366.66±119.07pg/ml??IL-10?6.50±3.73pg/ml??IL-17?204.05±49.86pg/ml?in peripheral blood were significantly increased in the patients with HBV-ACLF?p<0.01?.In patients with HBV-ACLF,compared with survivor group,the value of RDW?17.78±2.73%?,NLR?7.71±7.25?,the level of TNF-??10.43±5.09 pg/ml?and IL-6?416.46±55.87 pg/L?of the non-survivor group increased significantly.It suggested these indicators may be associated with the prognosis of patients with HBV-ACLF.Further analysis showed that TNF-?,IL-6 were positively correlated with RDW and NLR respectively.The correlation analysis of these four indexes and MELD scores was found to be positively correlated with each other.The prognostic factors of HBV-ACLF were further analyzed.The levels of IL-8 in serum of HBV-ACLF patients were significantly positively correlated with the expression of IL-8mRNA in PBMC,indicating that the level of cytokine in serum which was determined by ELISA method can reflect the immune status of the body.Single factor regression analysis showed that the serum albumin level was the protective factor for the prognosis of HBV-ACLF.Elevated age,Tbil,PT,International standardized ratio?INR?,NLR,RPR,RDW,MELD score and TNF-??IL-6?IL-8?IL-10?IL-17 were the risk factors for the prognosis of HBV-ACLF.In the first part,We joint the systemic inflammatory indexes and traditional clinical parameters in liver disease.We got five factors which include prothrombin time,INR,MELD score,RDW and NLR by multiple factors regression analysis.We used these indicators to establish the Logistic regression model YZ.The area under the receiver operating characteristic curve?AUC?of YZ was0.83?95%CI:0.773-0.886?in evaluating the short-term prognosis of HBV-ACLF.In the second part,We joint cytokines and traditional clinical parameters in liver disease.We got four factors which include prothrombin time,INR,TNF-?and IL-6 by multiple factors regression analysis.We used these indicators to establish the Logistic regression model MY.The AUC of MY was 0.872?95%CI:0.825-0.919?in assessing the mortality after 3 months of HBV-ACLF.In order to improve the prediction performance,in the third part,we integrated systemic inflammatory markers with cytokines.At the same time we combined these indicators with traditional clinical parameters in liver disease.Multivariate logistic regression analysis of independent risk factors for 3-month mortality in HBV-ACLF patients showed that PT?OR=1.066,95%CI:1.005-1.13,P=0.033?,the INR?OR=2.745,95%CI:1.123-6.709,P=0.027?,RDW?OR=1.285,95%CI:1.04-1.588,P=0.02?,the TNF-??OR=1.407,95%CI:1.113-1.777,P=0.004?and IL-6?OR=1.045,95%CI:1.017 1.075,P=0.002?were independent risk factors.So,the Logistic regression model was employed.ZH=-22.738+0.064×PT?s?+1.01×INR+0.251×RDW?%?+0.341×TNF-??pg/ml?+0.044×IL-6?pg/L?.According to the formula of P=1/1+e-y,wecalculated the cut-off point of the regression model was 0.988.The corresponding sensitivity and specificity were 76.74%and 83.33%respectively.The predictive value of ZH for the3-month mortality in HBV-ACLF patients was very high.It's AUC was 0.885?95%CI:0.841-0.929?,which was better than YZ,MY model and MELD score,indicating that the current model had higher ability to predict.To further verify the the predictive performance of the ZH model,we applied it to the validation group.The ROC analysis for the 3-month mortality in HBV-ACLF patients suggest that the AUC of the ZH model was 0.871?95%CI:0.802-0.939?,which was still higher than the YZ,MY model and MELD score.At the same time,the predictive ability of the ZH model for the different genotype in patients with HBV-ACLF was compared by ROC.In the 110HBV-ACLF patients with genotype B,the AUC of the ZH model was 0.922.While in the 72 HBV-ACLF patients with genotype C,the AUC area of the ZH model was only0.789.Conclusion:?Compared with CHB and HC,the RDW and NLR values of patients with HBV-ACLF were significantly increased and the levels of TNF-?and IL-6 in peripheral blood were increased significantly.Compared with survivor group,the value of RDW,NLR,the level of TNF-?and IL-6 of the non-survivor group were significantly increased in patients with HBV-ACLF.?The value of RDW,NLR,the level of TNF-?and IL-6 were positively correlated with MELD scores in patients with HBV-ACLF,indicating that these indicators may be related to the prognosis of disease.?RDW and NLR were independent predictors of 3-month mortality in patients with HBV-ACLF.YZ model was established based on systemic inflammatory markers.The multivariate logistic regression analysis based on cytokines showed that TNF-?and IL-6were independent predictors of short-term prognosis in patients with HBV-ACLF.We established MY model on this basis.?Finally,a new Logistic regression model named ZH was established to improve the predictive performance by integrating the systemic inflammatory markers based on blood routine parameters and cytokines.It had excellent predictive performance by analysis to validation group which can be used to judging the prognosis and evaluating the survival of conservative treatment at early stage in patients with ACLF.It also can improve the prognosis of patients by helping to judge whether need liver transplantation. |
PDF Full Text Request