| BACKGROUND:Chronic hepatitis B virus(HBV)infection is a common liver disease,associated with high morbidity and death.Radix Sophorae flavescentis has been used for treating people with chronic hepatitis B,as pharmacological and clinical studies have found that it may have antiviral,anti-inflammatory,anti-tumor and immuno-suppressive effects.However,the latest Chinese guideline on HBV infection does not recommend any traditional Chinese therapy.This study is to evaluate the benefits and harms of Radix Sophorae flavescentis for chronic hepatitis B,and to use modified error matrix tool to overview the current clinical evidence.The thesis includes four parts:part 1 is literature review;part 2 and part 3 are Cochrane reviews which combine meta-analysis and trial sequential analysis(TSA);part4 is a methodological study on error matrix tool for evidence overview by applying to clinical evidence on Radix Sophorae flavescentis for chronic hepatitis B as an example.METHODS:(Study 1)We applied to Cochrane Hepato-Biliary Group for title registration,which was Radix Sophorae flavescentis for chronic hepatitis B.We submitted the research protocol and published it on Cochrane Library(NO.CD013089).We searched Chinese and English databases(e.g.Cochrane Hepato-Biliary trial register,CENTRAL,MEDLINE,Embase,Web of Science,CNKI,Wanfang,Chongqing VIP,Sinomed)to identify eligible randomized clinical trials(RCTs)on Radix Sophorae flavescentis for chronic hepatitis B.The last search was December 2018.Participants were diagnosed with chronic hepatitis B,as defined by the trialists or according to guidelines.Trial participants could also have had cirrhosis,hepatocellular carcinoma,concomitant HIV infection or AIDS,hepatitis C,hepatitis D,or any other concomitant diseases.Trials comparing Radix Sophorae flavescentis or its extractions(e.g.matrine,oxymatrine)at any dose,form,and regimen,with placebo or no intervention were included.We allowed co-interventions when the co-interventions were administered equally to all the intervention groups of a trial.Our primary outcomes were all-cause mortality,serious adverse events,and health-related quality of life.Our secondary outcomes were hepatitis B-related mortality,hepatitis B-related morbidity,and adverse events considered’not to be serious’.Our exploratory outcomes were proportion of participants with detectable HBV-DNA and proportion of participants with detectable HBeAg.We independently extracted data and double checked;we assessed the risk of bias,and analysed data according to Cochrane Handbook.We used RevMan5.3 to synthesize the evidence.We conducted TSA to calculate the required information size of meta analysis,and adjusted the significance level if the information size was not met.We observed forest plot and used I2 statistic to evaluate heterogeneity.We used funnel plot in combination with statistical tests to explore publication bias.We used GRADE methodology to evaluate our certainty in the evidence.(Study 2)We applied to Cochrane Hepato-Biliary Group for title registration,which was Radix Sophorae flavescentis versus other drugs or herbs for chronic hepatitis B.We submitted the research protocol and published it on Cochrane Library(No.CD013106).We searched Chinese and English databases(e.g.Cochrane Hepato-Biliary trial register,CENTRAL,MEDLINE,Embase,Web of Science,CNKI,Wanfang,Chongqing VIP,Sinomed)to identify eligible randomized clinical trials(RCTs).The last search was December 2018.Participants were diagnosed with chronic hepatitis B.Trial participants could also have had cirrhosis,hepatocellular carcinoma,concomitant HIV infection or AIDS,hepatitis C,hepatitis D,or any other concomitant diseases.Trials comparing Radix Sophorae flavescentis or its extractions(e.g.matrine,oxymatrine)at any dose,form,and regimen,with other drugs or herbs were included.We allowed co-interventions when the co-interventions were administered equally to all the intervention groups of a trial.The primary outcomes were all-cause mortality,serious adverse events,and health-related quality of life.Secondary outcomes were hepatitis B-related mortality,hepatitis B-related morbidity,and adverse events considered ’not to be serious’.Exploratory outcomes were proportion of participants with detectable HBV-DNA and proportion of participants with detectable HBeAg.The methods for data extraction,quality assessment,data synthesis,TSA analysis,sensitive analysis,subgroup analysis,and GRADE assessments were consistent with study 1.(Study 3)We applied error matrix tool to overview the clinical evidence in TCM by taking Radix Sophorae flavescentis for chronic hepatitis B as an example,and based on the results,we made suggestions on future clinical studies and methodological studies.The data were based on the two Cochrane reviews.Randomised clinical trials eligible for these two Cochrane reviews were included and analyzed.We also included the previous published systematic reviews with meta-analyses with similar inclusion criteria for trials.There are four steps for constructing error matrix:1)evaluating systematic error.Evidence type and methodological quality within each type of evidence were considered;2)evaluating random error.Standard error for effect estimate was calculated according to Cochrane statistical formula;3)evaluating design error.The priority of outcome measures was considered first.The adequacy of other eight design components were assessed,including participants,experimental intervention,control intervention,clinical setting,goal(exploratory/pragmatic),objective(superior/equivalent/non-inferior),study structure,and unit of analysis;4)we used EXCEL to construct the error matrix based on random error(x axis),systematic error(y axis)and design error(z axis).RESULTS:(Study 1)(1)Literature searching and selection:4044 literature were identified and 796 full-text papers were read.After excluding 323 papers,436 papers were put on the waiting lists due to insufficient information and data on adverse events were extracted if reported.Totally,35 RCTs with 3556 participants were included.(2)Quality assessment:all trials were assessed at high risk of bias.(3)Effect estimate:only one of the 35 RCTs assessed mortality;no deaths occurred.Ten trials assessed serious adverse events;no serious adverse events occurred.None of the trials reported health-related quality of life,hepatitis B-related mortality,or morbidity.Adverse events considered ’not to be serious’was an outcome in 19 trials;nine of the trials had zero events in both groups.Radix Sophorae flavescentis versus placebo or no intervention showed no difference in effect on adverse events considered’not to be serious’(RR 1.10,95%CI 0.76 to 1.59;12=49%;10 trials,1050 participants;very low certainty).Radix Sophorae flavescentis showed a reduction in the proportion of participants with detectable HBV-DNA(RR 0.61,95%Cl 0.55 to 0.68;I2=56%;29 trials,2914 participants;very low certainty)and proportion of participants with detectable HBeAg(RR 0.71,95%Cl 0.66 to 0.76;I2=19%;20 trials,2129 participants;very low certainty).Regarding our surrogate outcomes on adverse events considered ’not to be serious’,analysed separately,we found Radix Sophorae flavescentis was associated with a lower risk of abdominal distention and a higher risk of digestive symptoms,fatigue,yellow urine,and jaundice.Post-hoc meta analysis showed Radix Sophorae flavescentis group had a higher proportion of participants with normal ALT level(RR 1.32,95%CI 1.21 to 1.44;I2=15%;8 trials,739 participants)(4)TSA results:the Trial Sequential Analysis suggested that more information is needed for the outcome adverse events considered’not to be serious’.The Trial Sequential Analyses confirmed the results for outcomes proportion of participants with detectable HBV-DNA,detectable HBeAg and normal ALT level.(5)Sensitivity analysis and subgroup analysis:sensitivity analysis showed that the missing data had no influence on the effect estimate.Subgroup analysis showed that intravenous infusion might increase the risk of adverse events when compared to oral administration and to combined ways of administration of Radix Sophorae flavescentis.This might be a reason for the significant 49%heterogeneity shown in the meta-analysis.(6)GRADE assessments:downgraded due to within-study bias,unexplained heterogeneity,imprecision,and publication bias,the evidence is of very low quality.(Study 2)(1)Literature searching and selection:We identified 4044 references,and 119 papers were retrieved after excluding irrelevant studies.109 papers were put on the waiting lists and only data on adverse events,if reported,were extracted.10 randomised clinical trials with 898 participants were included.(2)Quality assessment:we judged all trials at high risk of bias.(3)Effect estimate:the trials covered oral capsules,intravenous infusion,intramuscular injection,and acupoint injection of Radix Sophorae flavescentis with a follow-up period from 1 to 12 months.The drugs being used as a comparator were lamivudine,adefovir,interferon,thiopronin,thymosin,or other Chinese herbs.None of the trials reported all-cause mortality,health-related quality of life,serious adverse events,hepatitis B-related mortality,or morbidity.We are uncertain as to whether Radix Sophorae flavescentis has a beneficial or harmful effect on adverse events considered ’not to be serious’(RR 0.86,95%CI 0,42 to 1.75;I2=0%;2 trials,163 participants;very low certainty)as well as if it decreases or increases the proportion of participants with detectable HBV DNA(RR 1.14,95%CI 0.81 to 1.63;12=92%;8 trials,719 participants;very low certainty).Radix Sophorae flavescentis showed a reduction in the proportion of participants with detectable HBeAg(RR 0.86,95%CI0.75to0.98;i2=43%;7 trials,588 participants;very low certainty).Post-hoc meta-analysis showed Radix Sophorae flavescentis had a higher proportion of participants with normal ALT level(RR 1.40,95%CI 1.13 to 1.73;I2=24%;4 trials,375 participants).(4)TSA results:we performed Trial Sequential Analysis for all three outcomes,and the monitoring boundaries were not crossed as well as the required information size was not reached for any of the outcomes.(5)Sensitivity analyses and subgroup analyses:sensitivity analyses showed that there was no impact of missing data on the results.Regarding the reduction effect on proportion of participants with detectable HBV DNA,subgroup analyses showed that forms and ways of administration Radix Sophorae flavescentis,and different comparators being used may affect the effect estimate.(6)GRADE assessments:the evidence is of very low quality as downgraded due to within-study bias,unexplained heterogeneity,imprecision,and publication bias.(Study 3)According to the error risks assessment,all the reviews with meta-analyses were assessed at critically low quality and all the randomized trials were assessed at high risk of bias in terms of systematic error.On this systematic error level,they found there was substantial risk of random error regarding all-cause mortality(SE 0.36),moderate risk regarding serious adverse events(SE 0.22),substantial risk on nonserious adverse events(SE 0.35),and small to moderate risk regarding surrogate outcomes such as HBeAg and HBV-DNA(SE 0.16and0.21).Data were lacked on patient-centered outcomes(primary and secondary outcomes),while surrogate outcomes as hypothesis-generating were reported mostly in the articles.The quick glance at the evidence using a conceptualized visual matrix tool showed that the results from available meta-analysis and randomized trials supporting Radix Sophorae flavescentis for chronic hepatitis B seemed to be of low certainty.The gray color of the charts reflected that certain design error risks were occurred,mainly reflected as unclear or inadequate participants selection,inadequate experimental or control intervention,single center design,insufficient data to judge the studies as superiority and pragmatic design.CONCLUSION:Implications for clinical practice:the current clinical evidence lacked data on clinical relevant outcomes,such as all-cause mortality,health-related quality of life,serious adverse events,hepatitis-B related mortality,and hepatitis-B related morbidity.Compared with no treatment or placebo,Radix Sophorae flavescentis may help reduce the serum detectable HBV-DNA and detectable HBeAg,and increase the proportion of normal ALT;the effect of Radix Sophorae flavescentis on the adverse events considered ’not to be serious’ is still unclear.Compared with other drugs or herbs,Radix Sophorae flavescentis may help reduce the serum detectable HBeAg and may increase the proportion of normal ALT;the effect of Radix Sophorae flavescentis on the adverse events considered ’not to be serious’ and on the detectable HBV-DNA is still unclear.We caution the results of Radix Sophorae flavescentis showing a inhibition effect of viral replication and improvement of liver function because the trials are at high risk of bias,the required information size of meta analysis is not met,and because of the non-validated surrogate character of the outcome(hypothesis generating).Implication for further research:future more qualified trials on clinical important outcomes are needed to get knowledge on its effect for preventing disease progress,improving quality of life,and decreasing mortality.Using error matrix tool to overview the clinical evidence,systematic,random,and design error risks were found.Future studies could be designed to minimise the three error risks to improve the validity of the evidence.Error matrix tool could be used in commentary to synthesis method and GRADE assessment,and the methodology should be improved through future studies. |
PDF Full Text Request