Clinical Research Of Early Warning And Intervention Of HBV-related Acute-on-Chronic Liver Failure

There were 93 million chronic hepatitis B virus(HBV)carriers in our country,with about 30% appearing spontaneous acute exacerbation of chronic HBV each year.In addition,acute exacerbation of chronic HBV infected patients can caused by virus mutations induced by oral antiviral drugs,viral rebound after drug withdrawal and overlapping other hepatotropic virus infection.Disease progression and clinical manifestation of Hepatitis patients with acute exacerbation were complicated and various,some of these patients will progress to acute-on-chronic liver failure(ACLF)with high fatality rate.So it is necessary to carry out early warning and intervention in advance for dealing with risks of liver failure.In addition,there was a time-window of chronic HBV infection from occurrring acute exacerbation of hepatitis to ACLF,thus it left a certain possibility for early warning and active intervention of HBV related ACLF.Therefore,we call this time-window as acute on chronic pre-liver failure(pre-ACLF).Professors of our department in 2011 have put forward the concept and diagnostic criteria of the pre-ACLF firstly[1,2].However,there was still no accepted diagnostic criteria of the pre-ACLF both at home and abroad and lacking of sufficient evidences in evidence-based medicine.Currently internal comprehensive support treatments such as antiviral nucleoside analogues,liver protection and nutritional support for liver function were used in the early stage of hepatitis B liver failure,but these treatments cannot effectively prevent disease progression.Glucocorticoids have strong immunosuppressive function,anti-inflammatory effects and certain protective action for liver cells,which can effectively prevent occurrence and development of the hepatic failure in theory.We and Japanese scholars had ever reported that glucocorticoids can effectively prevent further development of hepatic failure in patients with HBV related pre-ACLF and severe acute exacerbation of chronic hepatitis B and improve prognosis of patients significantly [1,3].And glucocorticoids were used in the treatment of acute,sub-acute,and acute-on-chronic liver failure by physicians at home and abroad.But the efficacy of glucocorticoids was still controversial,which may mainly involve in treatment populations and difference timing of initial treatment with glucocorticoids [4].Earlier studies have shown that,HBV DNA replication is mainly active in the pathogenesis of the initial factor and lead to disease progression [5-6] One of the reasons,antiviral nucleoside analogues can be highly selective inhibition of HBV reverse transcriptase polymerase part,to inhibit replication of the virus,so as to achieve the role of hepatitis B treatment.In addition,several studies have shown that low HBV DNA and HBVDNA fast diminishing improve HBV-related ACLF [7] prognostic outcome.But the study of HBV-related pre-ACLF antiviral therapy impact on the progression of the disease is still small.This study aimed at retrospectively analyzing risks and relative factors of hepatic failure from 1279 cases with severe icterohepatitis of CHB and CHB patients in severe acute exacerbation stage,so as to work out diagnostic criteria of the pre-ACLF,providing evidence-based medical evidence for early prevention and treatment of ACLF.In addition,we have also analyzed curative effect and its related factors of glucocorticoids in the treatment of 56 patients with HBV related pre-ACLF to develop a forecast model and route diagram of the curative effect of glucocorticoids treatment,which was contribute to reasonable application of glucocorticoids in treating severe liver disease.Main results:1.There were 1279 cases with severe icterohepatitis of CHB and CHB patients in severe acute exacerbation stage.The relationship between different levels of serum alanine aminotransferase(ALT)and ACLF was analyzed to find that the baseline level of serum ALT for liver failure occurrence or not was 15.6±14.5×ULN and 14.8 ±14.5×ULN separately,it was no significant difference between them(P=0.430).Hierarchical cluster analysis showed that there was no significant difference in incidence rate of ACLF between patients with different levels of serum ALT(P=0.338).2.The baseline level of serum AST for liver failure occurrence or not between patients with different levels of serum AST was 16.5±15.2×ULN and 12.4±15.2×ULN separately,and there was a significant difference between them(P<0.001);Hierarchical cluster analysis showed that there was a significant difference in incidence rate of ACLF between patients with different levels of serum AST(P<0.001);the incidence rate of ACLF raised along with increasing AST level.The incidence rate of ACLF of patients with AST≥10×ULN was 28.3%(164/579),which was only 16.9%(118/700)in patients with AST<10×ULN and there was a significant difference(X2=23.584,P<0.001).3.The baseline level of serum TBil for liver failure occurrence or not between patients with different levels of serum TBil was 265.9±172.8μmol /L and 270.4± 172.5μmol separately,and there was no significant difference between them(P=0.634);Hierarchical cluster analysis showed that there was a significant difference in incidence rate of ACLF between patients with different levels of serum TBil(P<0.001).4.The baseline of international normalized ratio(INR)for liver failure occurrence or not between patients with different INR was 1.56±0.43 and 1.24±0.43 separately,there was a significant difference between them(P<0.001);Hierarchical cluster analysis showed that there was a significant difference in incidence rate of ACLF between patients with different INR(P<0.001);the incidence rate of ACLF raised along with increasing INR.5.We have put forward that the diagnostic criteria of the pre-ACLF as follows:(1)INR≥1.30;(2)Serum AST≥10×ULN and with obvious jaundice symptom(serum TBil≥51.3μmol/L),or serum TBil≥342.0μmol/L.The incidence rate of ACLF of patients met or not met the above diagnosis criteria in this study was 45.9%(195/425)and 10.2%(87/854)respectively and there was a significant difference(X2=208.3,P< 0.001).The positive and negative predictive value of this criterion was 45.9% and 89.8% respectively and the sensitivity and speciality rate was 69.1% and 76.9% respectively.6.Multiple logistic regression analysis showed that the baseline level of INR,AST,TBil and ALT and age were independent predictors for developing ACLF.Thus,the constructed risk prediction model for ACLF was: PY=1= e X/(1+e X),X=-10.245+0.026× AST(ULN)–0.025×ALT(ULN)+0.046×TBil(mg/d L)+4.642×INR+0.049×age(years),Y=1 suggested progressing to ACLF in 4 weeks.7.A total of 56 patients with HBV related pre-ACLF were administered 10 mg dexamethasone intravenously(once per day in the morning)for five days.Compared with the baseline level of serum TBil,declining range of serum TBil that greater than 50% at the fifth day of treatment was known as rapid response and that ranged from 30% to 50% was called partial response and that less than 30% was called no response.Serum TBil level was decreased in 94.6% of patients with glucocorticoid treatment at the fifth day,among which there were 30 cases with rapid response(53.4%),11 cases with partial response(19.6%)and 15 cases with no response(26.8%).8.serum TBil level decreased continuously in the rapid response group after stopping hormone,which was significantly lower than that of the partial response group and no response group(both P<0.001).There were no statistical differences in serum TBil levels after stopping hormone between the partial response group and no response group at the 14 th,21th and 28 th day(P>0.05),except at the 10 th day(p=0.022).9.There was no case progress to ACLF or death at the end of hormone therapy in the rapid response group(n=30)and the partial response group(n=11),while the incidence and fatality rate of ACLF in the no response group was 33.3%(5/15)and 13.3%(2/15)respectively.10.Duration of Jaundice(P=0.013,OR 1.258,95% CI 1.050-1.508),PTA(P=0.029,OR 0.014,95% CI 0.000-0.650)and ALT/TBil ratio(P=0.013,OR 0.735,95% CI 0.577-0.963)were independent factors for curative effect of glucocorticoid,while serum ALT(P=0.722,OR 0.999,95% CI 0.996-1.003),AST(P=0.407,OR 1.001,95% CI 0.999-1.004),albumin(P=0.967,OR 0.997,95% CI 0.845-1.176)were not independent factors affecting curative effect of glucocorticoid.Mathematical prediction model of glucocorticoid were calculated as follows:Py=ex/(1+ex),X= 0.687 +0.125×duration of Jaundice-0.508×ALT/TBil ratio-4.241×PTA11.A total of 1063 patients with HBV related pre-ACLF were divided into anti-viral group(n=904)and conventional treatment group(n=159).Patients in the anti-viral group received nucleotide analogs(entecavir,telbivudine,lamivudine,adefovir + lamivudine)and traditional supporting treatments.The incidence rate of liver failure was compared between the two groups during 4 weeks.The significantly lower incidence of liver failure was observed in the anti-viral group than in the conventional treatment group(?2=10.554,P=0.001).12.For 297 patients with cirrhosis,there was no difference in the incidence rate of liver failure between the anti-viral group(n=244)and the non-antiviral group(n=53)(36.9% versus 47.2%,P=0.164).For 766 non-cirrhotic patients,the incidence rate of liver failure in the anti-viral group(n=660)was lower than in the non-antiviral group(n=106)(14.5% versus 24.5%,P=0.009).13.For 488 patients with ALT<10×ULN,there was significant difference in the incidence rate of liver failure between the anti-viral group(n=394)and the non-antiviral group(n=94)(17.0% versus 27.7%,P=0.018).For 575 patients with ALT ≥ 10 × ULN,there was significant difference in the incidence rate of liver failure between the anti-viral group(n=510)and the non-antiviral group(n=65)(23.3% versus 38.5%,P=0.008).For 552 patients with AST<10×ULN,there was no significant difference in the incidence rate of liver failure between the anti-viral group(n=98)and the non-antiviral group(n=65)(14.8% versus 22.4%,P=0.060).For 511 patients with AST ≥ 10 × ULN,there was significant difference in the incidence rate of liver failure between the anti-viral group(n=450)and the non-antiviral group(n=61)(26.4% versus 47.5%,P=0.013).14.For 353 patients with MELD score <17,there was no significant difference in the incidence rate of liver failure between the anti-viral group(n=319)and the non-antiviral group(n=34)(10.1% versus 14.7%,P=0.514).For 710 patients with MELD score ≥17,there was significant difference in the incidence rate of liver failure between the anti-viral group(n=585)and the non-antiviral group(n=125)(25.8% versus 36.8%,P=0.013).Conclusions:1.The baseline level of ALT,AST,TBil and INR as well as the age were independent risk factors for hepatic failure occurrence.Diagnostic criteria for the pre-ACLF was:(1)INR≥1.30;(2)Serum AST≥10×ULN and with obvious jaundice symptom(serum TBil≥51.3μmol/L),or serum TBil≥342.0μmol/L.Thus,the constructed risk prediction model for ACLF was: PY=1=e X/(1+e X),X=-10.245+0.026×AST(ULN)–0.025×ALT(ULN)+0.046×TBil(mg/d L)+4.642×INR+ 0.049×age(years),Y=1 suggested progressing to ACLF in 4 weeks.Higher PY resulted in higher rate of ACLF.2.Glucocorticoids can effectively prevent patients with HBV related pre-ACLF further progressing to hepatic failure and significantly improve the prognosis.The key factor of improving curative effect of glucocorticoid was to make a right treatment plan for different indications,treatment timing and course of treatment.And mathematical prediction model of glucocorticoid were calculated as follows:Py=ex/(1+ex),X=0.687+0.125×duration of Jaundice-0.508×ALT/TBil ratio-4.241×PTA.The lower Py value meant there was a higher effective rate of effectively preventing hepatic failure.3.Antiviral therapy can effectively prevent the progression of HBV-related acute-on-chronic pre-liver failure(pre-ACLF)to liver failure.Especially in patients with non-cirrhosis,higher AST levels and higher MELD scores.