| BackgroundRecanalization for Budd-Chiari syndrome(BCS)refers to percutaneous angioplasty(PTA)alone or combined with elastic metal stent(EMS)placement,which could recanalize the obstructed hepatic venous outflow tract.According to the current guidelines,recanalization is the first-line invasive therapy for BCS patients with short-length stenosis.However,post-procedure restenosis rate can be as high as 50%within 1 year.A previous retrospective study indicated that addition of stent placement during PTA procedure could reduce the risk of restenosis by 80%.Nevertheless,the circumstances of stenting in the current clinical practice mainly includes rescue stenting for patients in whom restenosis have already occurred,or selective stenting in the initial procedure only for patients who might be susceptible to restenosis.The former practice could not prevent restenosis while the latter lacks support from clinical evidence and may not cover patients who could potentially benefit from stenting.In contrast,routine stenting for all patients might be a better solution,yet the rarity of BCS made it difficult to conduct randomized controlled trials to investigate the efficacy and safety of routine stenting.On the other hand,although multiple previous studies have attempted to develop prognostic models for outcome evaluation of BCS patients,these models did not originally target on patients receiving recanalization.Given that different treatment modalities for BCS patients have different optimal candidates,the prognosis of patients receiving recanalization might also differ from those receiving other treatments,and it remains unclear whether previous models could fit such population.Moreover,there lacks a recanalization-specific prognostic model to achieve individual outcome prediction and risk stratification for these patients.Objectives1)To investigate that compared to PTA alone,whether combination with routine stenting(PTA+EMS)could effectively improve long-term patency of recanalized veins,control symptom recurrence and maintain adequate safety with a randomized clinical trial.2)To analyze the utilization proportion of recanalization in Chinese patients with BCS,to validate the efficacy and safety difference between PTA+EMS and PTA alone,and to validate the robustness of these results with multiple sensitivity analyses.3)To examine the performance of previous prognostic scores in patients receiving recanalization,and to develop a recanalization-specific score to achieve individual outcome prediction and risk stratification.Methods1)Patients aged 18-75 and admitted to our center from July 2014 to September 2017 were screened,and eligible participants were randomly assigned to PTA+EMS or PTA group at a ratio of 1:1.PTA procedures were performed within 24 hours after randomization.Statisticians and investigators in charge of data collection and endpoint assessment were blind to randomization.Primary endpoint was restenosis,and secondary endpoints included recurrence of BCS-related symptoms and portal hypertension complications,death,and liver function,etc.Kaplan-Meier analysis was used to estimate cumulative probability of clinical events and its 95%confidence interval(CI).Cox regression analysis was used to investigate predictive factors of corresponding events.Results were further validated with competing risk analyses,recurrent event survival analyses,and subgroup analyses.2)Patient admitted to 6 participating centers in China from January 2010 to May 2019 were screened.Baseline information,treatment details,and follow-up data were acquired for eligible patients to analyze the utilization proportion of different treatment modalities and their contribution to patient survival;to compare between patients receiving PTA+EMS and PTA alone the efficacy parameters such as restenosis and symptom recurrence,and safety parameters such as complications;and to examine the robustness of these results with propensity score matching and subgroup analyses.Cox regression analyses were used to investigate predictive factors of patient outcome and to develop a prognostic model.The nomogram of the model was depicted,and calibration curve,C-index,and area under receiver operating characteristic curve(AUROC)were used to evaluate the performance of the model.Internal validations were also performed.A cut-off point was then selected for risk stratification.Finally,the model was compared with other previous models.Results1)Eighty-eight patients were included in the randomized controlled trial,among whom 43 patients were allocated to PTA+EMS group,and 45 to PTA group.Median follow-up was 27 months.Restenosis occurred in only 1 patient(2%)from PTA+EMS group and 18 patients(40%)from PTA group.According to survival analysis,PTA+EMS group had significantly lower cumulative probability of restenosis as compared to PTA group(3-year rates were 4%[95%CI 0-11.4%]vs.39.6%[95%CI 21.2-53.6%],respectively,log-rank p<0.001).The hazard ratio(HR)of restenosis was 0.04(95%Cl 0.01-0.31)for PTA+EMS group vs.PTA group,and the absolute risk difference was 35.6%(95%CI 24.2-55.0).Moreover,PTA+EMS group also had significantly better symptom recurrence-free survival than PTA group(3-year rates were 89.1%[95%CI 78.5-99.8%]vs.40.5%[26.9-61.0%],respectively,log-rank p<0.001,HR 0.23[95%CI 0.10-0.50]),with an absolute risk reduction of 39.2%(95%CI 19-55.1%).Furthermore,there was no statistical difference regarding the complication rates between the two groups(2%vs.0%,p=0.489),and no stent-related complications were observed.The survival rates and liver function were also similar between the two groups.Competing risk analyses,survival analyses for recurrent events,and subgroup analyses revealed consistent results.2)In the multicenter retrospective study,834 patients receiving recanalization were included,accounting for 83.1%of all patients being screened.Among these patients,87.1%had inferior vena cava involvement.The median follow-up of the cohort was 58 months,and the 5-year orthotopic liver transplantation(OLT)-free survival of the entire cohort was 91.1%(89.0-93.3%).Compared to PTA alone,PTA+EMS significantly reduced 5-year restenosis rate(1.2%[7.0-15.3%]vs.24.1%[20.1-27.8%],respectively,HR 0.50[0.340.73],log-rank p<0.001)and symptom recurrence rate(13.7%[9.0-18.2%]vs.20.5%[16.7-24.1%],respectively,HR 0.62[0.43-0.91],log-rank p=0.014).However,there was no statistical difference regarding 5-year OLT-free survival between patients receiving PTA+EMS and PTA alone(90.2%[86.4-94.1%]vs.91.6%[89.1-94.3%],respectively,HR 1.11[0.70-1.78],log-rank p=0.647).These results remained consistent in analyses after propensity score matching and in subgroup analyses.According to Cox regression analyses,variceal bleeding history,ascites,albumin and creatinine levels were independent predictors,and were used to develop the BCS-recanalization score,with C-index being(0.66-0.76),and time dependent AUROC at 1-year,3-year,and 5-year being 0.76(0.69-0.83),0.72(0.65-0.79),and 0.65(0.59-0.72),respectively,which were higher than previous models.Moreover,with a cut-off point,the model could be used for patient stratification,and high-risk patients identified by this score had worse outcome,lower patency and higher symptom recurrence rate.Conclusions1)As the first randomized controlled trial for BCS,the current study found that routine stenting at initial PTA procedure could effectively prevent restenosis and reduce symptom recurrence rate with adequate safety in BCS patients with short-length stenosis.2)In the real-world study,PTA+EMS could definitively improve treatment efficacy,and its benefit remained robust in analyses after propensity score matching and in subgroup analyses.Incidence of stent-related complications were low,none of which compromised long-term survival.3)BCS-recanalization is the first recanalization-specific prognostic score for patients with BCS,and could be used for individual outcome prediction and risk stratification with superior performance over other previous models.External validations of this model are needed in the future. |
PDF Full Text Request