| Objective:To perform a meta-analysis and a systematic review of the related research on total ankle arthroplasty versus ankle arthrodesis for the treatment of ankle arthritis by collecting and retrieving data.To perform a comprehensive comparison about the clinical curative effect between total ankle arthroplasty and ankle arthrodesis.Methods:We conducted a comprehensive search in the Pub Med、Embase,and Cochrane Central Register of Controlled Trials(CENTRAL databases.The following search terms were ‘ankle arthritis’,‘ankle arthroplasty’,‘ankle replacement’,‘ankle arthrodesis’ and ‘ankle fusion’,all searches as of Jan,2017.After duplicate studies were excluded using End Note X7,the title and abstract of each identified study were screened.The methodological quality of each study was estimated using criteria set by the Journal of Bone & Joint Surgery.The observation indexes included clinical score(AOFAS score,VAS score,SF-36 score,AOS score,FAAM score).We also investigated the patient satisfaction rate,the prevalence of complications and the re-operation rate.The clinical score was a continuous variable,the standardized mean difference(SMD)or the mean difference(MD or WMD)were chosen as the combined effects.The patient satisfaction rate,the prevalence of complications and the re-operation rate were discontinuous variable,the relative risk(RR)or odds ratio(OR)was calculated by combining the effects of 95%(95%CI).Statistical heterogeneity was assessed using the value of the I2 statistic and Q test.P<0.1 I2>50% was considered to be of high statistical heterogeneity,we chose the random effects model.P>0.1,I2<50%was considered to be of low statistical heterogeneity,we chose the fixed effects model.Sensitivity analysis was performed by altering the effect model,omitting one study in each turn to estimate the stability of the results.We used STATA,version12.0 software(Stata Corporation,College Station,TX,USA)to performed all thestatistical analyses.We used the Begg and Egger tests to evaluate the potential publication.Results:There were 562 reports identified in Pub Med、Embase and CENTRAL database,338 reports in Pub Med database,201 reports in Embase database,23 reports in CENTRAL database.There were 12 original studies including in the article.The original studies included 4 prospective cohort studies,8 retrospective cohort studies,5multicenter studies,and 7 single center studies.There were 7 studies with the evidence of level II(medium quality cohort study),3 studies with the evidence of level III(low quality cohort study),but no studies with the evidence of level IV.The original research was published from 2009 to 2016;The most sample size of studies was687,the least was 30.The average follow-up was seventy-four months,the minimum was six months.The result of statistical analysis:AOFAS score(WMD = 7.34,CI95% [-1.80,16.48],P=0.115),VAS score(WMD =-0.27,95% CI [-1.24,0.70],P=0.583),SF-36 score(WMD = 12.22,CI 95% [6.93,17.51],P=0.000),AOS score(WMD = 0.317,CI [-1.14,1.77],P=0.669),FAAM score(ADL:WMD = 11.95,CI[5.33,18.56],P=0.000;Sport:WMD = 15.97,95% CI [7.68,24.26],P=0.000);the patient satisfaction rate(RR = 1.025,95% CI [0.935,1.125],P=0.596),the prevalence of the re-operation rate(RR = 1.928,95% CI [1.462,2.544],P=0.000),the prevalence of complications(WMD = 2.19,95% CI [1.47,3.26],P=0.000)。Sensitivity analysis results suggest that the conclusion is stable,Begg and egger test showed no publication bias.Conclusion:In the clinical score,there was no significant difference between the TAA group and AA group,however,TAA is superior to AA in terms of quality of life,daily activity and motor function.There was no significant difference between the TAA group and AA group in the patient satisfaction rate,however,the incidence of re-operation and complication of TAA was higher than AA.Due to the limited number of the sample size,lacking of long-term follow-up,there were heterogeneity and potential bias in individual observation indicators.We would be careful of the conclusion,do a long-term follow-up about more high quality randomized,and large sample multicenter randomized controlled trial,then we could make a furthervalidation and a more comprehensive analysis and a meta-analysis. |
