Efficacy of ursodeoxycholic acid as an adjuvant treatment to prevent acute cellular rejection after liver transplantation:a meta-analysis of randomized controlled trials

Chengdu,China

Introduction

Administration of ursodeoxycholic acid (UDCA)has been well recognized as a simple and costeffective option to treat various chronic liver diseases,such as primary biliary cirrhosis (PBC),primary sclerosing cholangitis (PSC),cysticfibrosis and intrahepatic cholestasis in pregnancy.[1,2]UDCA exhibits a cytoprotective activity which may be involved in reducing bile toxicity,anti-oxidation,anti-apoptosis and stabilizing membrane.[1-4]Additionally,in PBC patients,UDCA significantly reverses aberrant HLA class I molecules expression on hepatocytes,thereby suppressing T lymphocytes antigen recognition and cytotoxicity.[5,6]With the cytoprotective activities together with immunomodulatory effects,UDCA can act as an attractive adjuvant therapy in acute cellular rejection (ACR) after liver transplantation (LT).[7-9]

In 1990 Persson et al[10]reported that prophylactic use of UDCA after LT reduced the incidence of ACR significantly.Subsequently,several clinical trials validated this effect but others demonstrated contradictory results.[10-17]Therefore,we performed a meta-analysis of randomized controlled trials (RCTs)to determine the efficacy of UDCA in reducing the incidence of ACR in liver transplant recipients.

Methods

Literature search

All relevant studies regarding effects of UDCA in liver transplantation were searched independently by two researchers (DYL and XXZ) in electronic database,PubMed/MEDLINE,EMBASE,Cochrane Central Register of Controlled Trials,ScienceDirect databases and Web of Science,from January 1981 to March 2012.Relevant abstracts of international meetings were also searched.Finally,references of each included-study were searched manually.The following MeSH terms and key words were searched:"ursodeoxycholic acid" or "bile acids"combined with "liver transplantation" or "orthotopic liver transplantation".Any discrepancies were resolved by consensus.

Study selection

RCTs that evaluated the efficacy of UDCA as an adjuvant treatment to prevent ACR after LT were included without language limitation.Exclusion criteria were:(1) trials with non-parallel design; (2) trials with significant difference in baseline characteristics between treatment and control groups; (3) abstracts which were unable to obtain complete data; (4) multiple publications of the same database.

Data extraction and quality assessment

Data were extracted as: first author,publication year,number of participants,baseline characteristics of treatment and control groups (age,gender,original liver disease and immunosuppressive protocol),inventions(doses,starting and during),follow-up time,primary end-points (ACR,steroid-resistant rejection and multiple episodes of ACR) and the diagnosis of ACR and steroidresistant rejection.

The quality of each study was also assessed independently according to the Jadad scoring system by two reviewers (XXZ and CNS) and,a Jadad score of 3 to 5 points indicated high-quality research.[18]Additionally,allocation concealment and intention-to-treat (ITT)analysis were also considered,but did not contribute to the overall quality score.Any disagreements were resolved by discussion and consensus.

Statistical analysis

Meta-analyses were made using RevMan 5.1.6(The Cochrane Collaboration,Oxford,UK) software.Statistical heterogeneity between RCTs was assessed using the Chi-square test.A P value less than 0.1 was considered statistically significant,and the Higgins I2statistical method was quantitatively used to measure sources of heterogeneity.The I2value less than 25%indicated low heterogeneity,between 25% and 50%indicated moderate heterogeneity. Subsequently,thefixed effects model was applied to calculate the summary risk ratios (RRs) and the Mantel-Haenszel method was used to detect the 95% confidence intervals(CIs).When heterogeneity was high (I2value ＞50%),a random effects model was used and tried to explore the sources of heterogeneity by the subgroup or sensitivity analyses.[19]Funnel plots were constructed to screen potential publication bias.[20]Differences in proportions between two groups were examined by the Chi-square analysis,and P＜0.05 was considered statistically significant.

Results

Description of studies

The selection process flowchart of our study is shown in Fig.1.A total of 2350 records were identified by our initial literature search.After the primary screening of abstracts,2196 were rejected because of duplicity (n=857) and irrelevant (n=1339) records.Of the remaining 154 articles,137 records were excluded for the following reasons:review articles (n=14),irrelevant (n=98),nonhuman (n=23) and editorials (n=2).Subsequently,17 studies were selected for further analysis.Four records were excluded because of duplicate publications:three meeting abstracts and one letter[10]used the same data sources with other four trials.Two records were excluded because one was historical control trial(HCT)[11]and another did not have a control group.[13]Two records were excluded without our end-points of interest.[21,22]Three records were excluded for different intervention modalities.As one study evaluated the effect of pre-transplantation UDCA therapy on the outcome of LT,[23]and the other two trials evaluated the role of UDCA in liver transplantation recipients with total immunosuppression withdrawal (TIW).[24,25]One editorial was excluded as it has no experimental data.[26]One study was excluded as tauroursodeoxycholic acid was used instead of UDCA.[12]

Fig.1.The selection process flowchart.

Four prospective,randomized,double-blind and placebo-controlled trials with 234 patients werefinally selected for the analysis.All included patients had orthotopic liver transplantation (OLT) with the same surgical procedures divided into treatment (UDCA-treated) and control (placebo-treated) groups.Many factors like age,gender and especially primary liver disease can in fluence the incidence of ACR.[27-31]Autoimmune hepatitis (AIH),PBC[29,30]and hepatitis C[31]have been reported to be associated with higher incidence of ACR.Although all four studies claimed that there was no basic imbalance in baseline characteristics between the two groups included age,gender,and primary liver disease (Table 1),we still made comparative analysis of the proportion of patients with high ACR-risk diseases(AIH,PBC and hepatitis C) between the two groups in each trial to further clarify the homogeneity of underlying disease.However,no statistically significant difference was noted (Table 2).

The summary of four included trials is described in Table 3.In all the trials,the UDCA and the placebo had identical appearance and taste.In addition,the effects of different doses and duration of UDCA were also evaluated in these trials,10 to 15 mg/kg per day for 3 to 6 months in three RCTs[15-17]and 600 mg/d for 2 months in one RCT.[14]In Keiding et al's study,administration of UDCA was started on the first post-operative day whereas,in other three trials,[14,16,17]UDCA was startedon postoperative days 3 to 5.The primary end-points in all trials contained the incidence of ACR and steroidresistant rejection and the multiple episodes of ACR according to the similar diagnostic criteria.

Table 1.Baseline characteristics of included patients

Table 2.The proportion of patients with high ACR-risk diseases in the UDCA vs placebo groups

The quality assessment of four RCTs is summarized in Table 4.Each trial was randomized and double-blind(identical placebo),but only Barnes et al[16]provided the method to generate the sequence of randomization.There was a description of withdrawals and dropouts in reports on four trials.Two trials performed an ITT analysis.One trial had adequate allocation concealment with a statement on the use of sealed serial envelope.Therefore,we confirmed that all four trials were highquality without the source of heterogeneity.

Table 3.Summary of included trials

Table 4.Quality assessment of included studies

Data analysis

The four trials reported the number of patients who experienced one or more episodes of ACR and steroidresistant rejection in the UDCA and placebo groups,but only three trials reported the number of patients with the multiple episodes of ACR (Table 5).Prophylactic use of UDCA did not decrease the number of patients with ACR (RR:0.94,95% CI:0.77 to 1.16,P＞0.05)with nil heterogeneity (P=0.95,I2=0%).Use of UDCA also did not decrease the incidence of steroid-resistant rejection (RR:0.77,95% CI:0.47 to 1.27,P＞0.05) with nil heterogeneity (P=0.52,I2=0%).In addition,UDCA use did not decrease the number of patients with the multiple episodes of ACR (RR:0.60,95% CI:0.28 to 1.30,P＞0.05) with moderate heterogeneity (P=0.14,I2=49%)(Fig.2).

Subgroup analysis

According to the time of administration of UDCA,the four trials were divided into two main groups:early (the first postoperative day) and delayed(between postoperative days 3 and 5) intervention groups.The subgroup analysis showed that the early or delayed intervention of UDCA did not decrease the number of patients with ACR and steroid-resistant rejection (Fig.3).Thus,a subgroup analysis was basedon the dose of UDCA,which indicated that highdose UDCA (average 1000 mg/d) or low-dose UDCA(average 600 mg/d) also did not affect these results(Fig.4).In addition,comparison between the tripledrug immunosuppressive regimen and the doubledrug regimen showed that UDCA used as an adjuvant in LT patients treated with the triple-drug or doubledrug regimen did not decrease the number of patients with ACR and steroid-resistant rejection.In addition,the triple-drug regimen did not reduce the multiple episodes of ACR but the double-drug regimen tended to decrease this event (Fig.5).

Fig.2.Forest plot of the incidence of ACR,steroid-resistant rejection and multiple episodes of ACR after LT.

Fig.3.Forest plot of the incidence of ACR and steroid-resistant rejection after LT:early intervention vs delayed intervention

Fig.4.Forest plot of the incidence of ACR and steroid-resistant rejection after LT:high-dose UDCA vs low-dose UDCA.

Sensitivity analysis

After excluding the Pageaux et al[14]study,there was still no difference in the incidence of ACR between the two groups (RR:0.95,95% CI:0.77 to 1.17,P＞0.05)with nil heterogeneity (P=0.83,I2=0%).In addition,this also did not alter the incidence of steroid-resistant rejection (RR:0.73,95% CI:0.44 to 1.21,P＞0.05) with nil heterogeneity (P=0.42,I2=0%) (Fig.6).

Discussion

Fig.5.Forest plot of the incidence of ACR,steroid-resistant rejection and multiple episodes of ACR after LT:triple-drug regimen vs double-drug regimen.

Fig.6.Sensitivity analysis.

Fig.7.Funnel plot of included trials.

ACR is regarded as one of the most common problems faced by liver transplant recipients in spite of advances in immunosuppressive therapy.[8,9]In addition,when high-dose steroid fails to treat the episode of ACR,steroid-resistant rejection occurs.OKT3,although helpful to further suppress the immune system,may cause infections.[32]The aberrant expression of HLA class I molecules on hepatocytes plays an important role in the process of ACR,thus providing a therapeutic window.Such abnormal expression could significantly be reversed by administration of UDCA in PBC patients.This immunomodulatory effect together with cytoprotective activities of UDCA made many clinical trials determine the efficacy of UDCA in the prevention of ACR after LT,but led to a contradictory conclusion.In view of limited reliability of retrospective studies and HCTs,we only included four high-quality RCTs (Jadad score 4 to 5) in our study.A sensitivity analysis showed no significant difference in the result which further confirmed the reliability and stability of our analysis.

This meta-analysis of RCTs showed that UDCA was beneficial as an adjuvant treatment for ACR in liver transplant recipients,including the incidence of ACR and steroid-resistant rejection.Moreover,no difference between different intervention programs (high vs lowdose UDCA,early vs delayed UDCA treatment,the triple-vs double-drug immunosuppressive regimen with UDCA) was found.Three of the four RCTs explored the effect of prophylactic UDCA on the multiple episodes of ACR,and showed no benefits with moderate heterogeneity.The subgroup analysis found that the double-drug regimen was superior to triple-drug regimen in decreasing the multiple episodes of ACR.

How does UDCA as an adjuvant prevent the episode of ACR? Persson et al[10]reported that 10 mg/kg/d of UDCA exerted this effect.But our analysis showed that two doses of UDCA (around 8 mg/kg per day and 15 mg/kg per day) did not reduce the incidence of ACR and even steroid-resistant rejection.However,perhaps 15 mg/kg of UDCA per day was still below the therapeutic range and a higher dose of UDCA might be effective.As ACR most likely occur around one week after transplantation,[9,27]it seemed that the difference in starting time of UDCA was likely to cause different results,but that was not able to be confirmed in our analysis.Whether the intervention started on the first postoperative day or between postoperative days 3 to 5 did not decrease the incidence of ACR,steroid-resistant rejection and the multiple episodes of ACR.In addition,the difference between the standard triple-drug and double-drug immunosuppressive regimen in included patients was not able to affect the conclusion that UDCA as an adjuvant did not reduce the incidence of ACR and steroid-resistant rejection,probably because both protocols contained the sufficient amount of cyclosporine.However,the double-drug immunosuppressive regimen with UDCA was able to reduce the multiple episodes of ACR.Because only three trials were included in this analysis,whether the doubledrug immunosuppressive regimen with UDCA is better to prevent the multiple episodes of ACR is inconclusive and more clinical evidences are required.

Only four RCTs were included in this meta-analysis,the risk of publication bias cannot be assessed well irrespective of a symmetrical funnel plot (Fig.7).[33]There may be a selection bias in our analysis because only one trial reported the methods of randomization and allocation concealment.Besides of age,gender,primary liver disease,the race of recipient,graft preservation and immunosuppressive protocol also seems to in fluence the incidence of ACR.[27,28,34,35]These factors were not entirely uniform across the four trials,which potentially contributed to limiting factors in our study.

Regardless of these limitations,this analysis confirmed that UDCA does not decrease ACR occurrence after LT.This conclusion seems to be frustrating.A recent study by Wang et al[36]indicated that UDCA administration early after LT not only significantly improved liver enzymes but also diminished the incidence of biliary sludge and casts within the first postoperative year; however,Wang et al[36]did not found the difference in the 5-year overall outcomes.Therefore,administration of UDCA still may be a cost-effective choice for improving quality of life in patients after LT.

In conclusion,this meta-analysis strongly suggested that UDCA as an adjuvant treatment did not prevent the episode of ACR and steroid-resistant rejection after LT.Further trials should be conducted in order to identify whether the higher dose of UDCA will exert such benefits,and whether the prophylactic use of UDCA to the double-drug immunosuppressive regimen rather than the triple-drug regimen could decrease the multiple episodes of ACR after LT.

Acknowledgements:We thank the professors of Key Laboratory of Transplant Engineering and Immunology,Ministry of Health at West China Hospital for guidance.

Contributors:DYL,XXZ and CNS designed the study and collected data.DYL and XXZ analyzed the data.DYL wrote the main body of the article.All authors helped with the manuscript revision.CNS is the guarantor.

Funding:This study was supported in part by a grant from the National Nature Science Foundation of China (30972923).

Ethical approval:Not needed

Competing interest:No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

1 Roma MG,Toledo FD,Boaglio AC,Basiglio CL,Crocenzi FA,Sánchez Pozzi EJ.Ursodeoxycholic acid in cholestasis:linking action mechanisms to therapeutic applications.Clin Sci (Lond) 2011;121:523-544.

2 Festi D,Montagnani M,Azzaroli F,Lodato F,Mazzella G,Roda A,et al.Clinical efficacy and effectiveness of ursodeoxycholic acid in cholestatic liver diseases.Curr Clin Pharmacol 2007;2:155-177.

3 Poupon R.Ursodeoxycholic acid and bile-acid mimetics as therapeutic agents for cholestatic liver diseases:an overview of their mechanisms of action.Clin Res Hepatol Gastroenterol 2012;36:S3-12.

4 Joutsiniemi T,Timonen S,Leino R,Palo P,Ekblad U.Ursodeoxycholic acid in the treatment of intrahepatic cholestasis of pregnancy:a randomized controlled trial.Arch Gynecol Obstet 2014;289:541-547.

5 Beuers U.Drug insight:Mechanisms and sites of action of ursodeoxycholic acid in cholestasis.Nat Clin Pract Gastroenterol Hepatol 2006;3:318-328.

6 Lindor K.Ursodeoxycholic acid for the treatment of primary biliary cirrhosis.N Engl J Med 2007;357:1524-1529.

7 Tippner C,Nashan B,Hoshino K,Schmidt-Sandte E,Akimaru K,Böker KH,et al.Clinical and subclinical acute rejection early after liver transplantation:contributing factors and relevance for the long-term course.Transplantation 2001;72:1122-1128.

8 Fisher LR,Henley KS,Lucey MR.Acute cellular rejection after liver transplantation:variability,morbidity,and mortality.Liver Transpl Surg 1995;1:10-15.

9 Wiesner RH,Demetris AJ,Belle SH,Seaberg EC,Lake JR,Zetterman RK,et al.Acute hepatic allograft rejection:incidence,risk factors,and impact on outcome.Hepatology 1998;28:638-645.

10 Persson H,Friman S,Scherstén T,Svanvik J,Karlberg I.Ursodeoxycholic acid for prevention of acute rejection in liver transplant recipients.Lancet 1990;336:52-53.

11 Friman S,Persson H,Scherstén T,Svanvik J,Karlberg I.Adjuvant treatment with ursodeoxycholic acid reduces acute rejection after liver transplantation.Transplant Proc 1992;24:389-390.

12 Angelico M,Tisone G,Baiocchi L,Palmieri G,Pisani F,Negrini S,et al.One-year pilot study on tauroursodeoxycholic acid as an adjuvant treatment after liver transplantation.Ital J Gastroenterol Hepatol 1999;31:462-468.

13 Clavien PA,Sharara AI,Camargo CA Jr,Harland RC,Fitz JG.Evidence that ursodeoxycholic acid prevents steroid-resistant rejection in adult liver transplantation.Clin Transplant 1996;10:658-662.

14 Pageaux GP,Blanc P,Perrigault PF,Navarro F,Fabre JM,Souche B,et al.Failure of ursodeoxycholic acid to prevent acute cellular rejection after liver transplantation.J Hepatol 1995;23:119-122.

15 Keiding S,Høckerstedt K,Bjøro K,Bondesen S,Hjortrup A,Isoniemi H,et al.The Nordic multicenter double-blind randomized controlled trial of prophylactic ursodeoxycholic acid in liver transplant patients.Transplantation 1997;63:1591-1594.

16 Barnes D,Talenti D,Cammell G,Goormastic M,Farquhar L,Henderson M,et al.A randomized clinical trial of ursodeoxycholic acid as adjuvant treatment to prevent liver transplant rejection.Hepatology 1997;26:853-857.

17 Fleckenstein JF,Paredes M,Thuluvath PJ.A prospective,randomized,double-blind trial evaluating the efficacy of ursodeoxycholic acid in prevention of liver transplant rejection.Liver Transpl Surg 1998;4:276-279.

18 Jadad AR,Moore RA,Carroll D,Jenkinson C,Reynolds DJ,Gavaghan DJ,et al.Assessing the quality of reports of randomized clinical trials:is blinding necessary? Control Clin Trials 1996;17:1-12.

19 Higgins J,Green S.Cochrane Handbook for Systematic Reviews of Interventions,Version 5.0.1 (updated March 2011).Available from:www.cochrane-handbook.org.

20 Sterne JA,Egger M.Funnel plots for detecting bias in metaanalysis:guidelines on choice of axis.J Clin Epidemiol 2001;54:1046-1055.

21 Söderdahl G,Nowak G,Duraj F,Wang FH,Einarsson C,Ericzon BG.Ursodeoxycholic acid increased bile flow and affects bile composition in the early postoperative phase following liver transplantation.Transpl Int 1998;11:S231-238.

22 Maboundou CW,Paintaud G,Vanlemmens C,Magnette J,Bresson-Hadni S,Mantion G,et al.A single dose of ursodiol does not affect cyclosporine absorption in liver transplant patients.Eur J Clin Pharmacol 1996;50:335-337.

23 Heathcote EJ,Stone J,Cauch-Dudek K,Poupon R,Chazouilleres O,Lindor KD,et al.Effect of pretransplantation ursodeoxycholic acid therapy on the outcome of liver transplantation in patients with primary biliary cirrhosis.Liver Transpl Surg 1999;5:269-274.

24 Assy N,Adams PC,Myers P,Simon V,Minuk GY,Wall W,et al.Randomized controlled trial of total immunosuppression withdrawal in liver transplant recipients:role of ursodeoxycholic acid.Transplantation 2007;83:1571-1576.

25 Assy N,Adams PC,Myers P,Simon V,Ghent CN.A randomised controlled trial of total immunosuppression withdrawal in stable liver transplant recipients.Gut 2007;56:304-306.

26 Keiding S.Does prophylactic ursodeoxycholic acid treatment of liver transplant patients prevent acute rejection? Hepatology1998;27:1748-1749.

27 Neuberger J.Incidence,timing,and risk factors for acute and chronic rejection.Liver Transpl Surg 1999;5:S30-36.

28 Kelly DA,Sibal A.Liver transplantation in children.Indian Pediatr 2006;43:389-391.

29 Seiler CA,Dufour JF,Renner EL,Schilling M,Büchler MW,Bischoff P,et al.Primary liver disease as a determinant for acute rejection after liver transplantation.Langenbecks Arch Surg 1999;384:259-263.

30 Liberal R,Zen Y,Mieli-Vergani G,Vergani D.Liver transplantation and autoimmune liver diseases.Liver Transpl 2013;19:1065-1077.

31 McTaggart RA,Terrault NA,Vardanian AJ,Bostrom A,Feng S.Hepatitis C etiology of liver disease is strongly associated with early acute rejection following liver transplantation.Liver Transpl 2004;10:975-985.

32 Pillai AA,Levitsky J.Overview of immunosuppression in liver transplantation.World J Gastroenterol 2009;15:4225-4233.

33 Macaskill P,Walter SD,Irwig L.A comparison of methods to detect publication bias in meta-analysis.Stat Med 2001;20:641-654.

34 Maggard M,Goss J,Ramdev S,Swenson K,Busuttil RW.Incidence of acute rejection in African-American liver transplant recipients.Transplant Proc 1998;30:1492-1494.

35 Jin H,Dahmen U,Liu A,Huang H,Gu Y,Dirsch O.Prolonged cold ischemia does not trigger lethal rejection or accelerate the acute rejection in two allogeneic rat liver transplantation models.J Surg Res 2012;175:322-332.

36 Wang SY,Tang HM,Chen GQ,Xu JM,Zhong L,Wang ZW,et al.Effect of ursodeoxycholic acid administration after liver transplantation on serum liver tests and biliary complications:a randomized clinical trial.Digestion 2012;86:208-217.