ENDOSCOPIC TREATMENT OF POST-LIVER TRANSPLANTATION ANASTOMOTIC BILIARY STRICTURE: systematic review andmeta-analysis

INTRODUCTION

Orthotopic liver transplantation (OLT) is a well-accepted treatment of end-stage liver disease. Biliary fistula, anastomotic biliary strictures (ABS)^{(5, 17)} and ischemic biliary lesions⁽¹⁸⁾ figure among the most common complications after OLT.

Considering specifically ABS there are some treatment options including balloon dilation, the placement of multiple plastic stents (MPS) and the placement of self-expandable metal stents^{(6, 9, 10, 15, 18)}. These modalities can be employed individually or in association. Furthermore the biliary access can be achieved by the percutaneous route (PTC) or retrogradely by the endoscopic route (ERC). It is recognized that ERC is the approach of choice when the complication involves an end-to-end biliary anastomosis. However it is not clear which treatment modality presents the best result. On the other hand when less invasive treatment fails surgical hepaticojejunal anastomosis is usually indicated.

Despite the rather high incidence of ABS after OLT and the lack of a well-established treatment algorithm there is a paucity of randomized controlled trials and systematic reviews on the subject. Thus, the aims of this study were to systematically review the literature concerning ABS after OLT and to organize a meta-analysis with the results.

METHODS

Study identification and selection

A systematic review of the literature was performed on the management of ABS post-OLT. The Medline-PubMed, EMBASE, Scielo-LILACS, and Cochrane Databases were electronically searched from January 1966 to April 2013.

The terms for each group were sought in combination using the “OR” boolean operator. The results for the search terms which formed the “P” (Patients) group were combined with the result for the search which formed the “I” (Intervention) group, using the “AND” boolean operator, and subsequently with the “exclusion key words” using the “NOT” boolean operator (Figure 1).

FIGURE 1. The terms used on PubMed database search were developed using patient, intervention, comparison or control, outcome (PICO) structure 

The MEDLINE search was performed through PubMed (www.ncbi.nlm.nih.gov/pubmed) and was adapted by using the terms and Mesh-terms (Post-liver transplantation or Liver transplantation or Liver transplant or Hepatic Transplantation or Liver Grafting) and (Constrictions, Pathologic or Stricture* or Stenose* or Stenosis*) and (Liver Transplantation or Hepatic Transplantation or Liver Grafting) and (Stents or Cholangiopancreatography, Endoscopic Retrograde or ERCP or Cholangiography) or (Post-liver transplantation or Liver transplantation or Liver transplant or Hepatic Transplantation or Liver Grafting) and (Constrictions, Pathologic or Stricture* or Stenose* or Stenosis*) or ((Post-liver transplantation or Liver transplantation or Liver transplant or Hepatic Transplantation or Liver Grafting) and (percutaneous transhepatic biliary drainage or percutaneous interventions or percutaneous treatment or drainage or percutaneous). This initial search was followed by a more specific and detailed one in accordance with our patient, intervention, comparison or control, outcome (PICO) structure using these terms and Mesh-terms observed in Figure 1. The same strategy was used in the Embase (www.embase.com) and LILACS (http://lilacs.bvsalud.org) databases. The Cochrane Library Database was searched for both registered and already published systematic reviews on the treatment of ABS after OLT.

We included clinical studies involving adult and pediatric patients who were treated for ABS after OLT (either cadaveric or living donor organs). The quality of the studies was evaluated by two independent researchers (LSN and WMB). In case of disagreement the researchers held a consensus meeting to reach a final decision.

This study was approved by the Institutional Review Board.

RESULTS

Study selection

By using the search strategy, 3655 articles were initially retrieved. Their abstracts were reviewed and 13 studies were selected according to the inclusion criteria (Figure 2). Finally we extracted the data from 5 comparative studies to calculate the meta-analysis (Figure 3). No systematic reviews or meta-analysis on the treatment of ABS after OLT were found in the Cochrane Library.

FIGURE 2. Diagram of this systematic review showing the initial steps to select the articles and the included and excluded citations 

FIGURE 3. The PICO structure developed using a population, intervention, comparison and outcomeThe studies selected for meta-analysis are in bold.LT: liver transplantation; KMP: Kumpe; BD: balloon dilation; EP: plastic endoprosthesis; BS: biliary stricture; AS: anastomotic stricture; ERCP: Endoscopic retrograde cholangiopancreatography; LDLT: living donor liver transplantation; DDLT: deceased-donor liver transplantation; ABS: anastomotic biliary strictures; OLT: orthotopic liver transplantation; ECP: Endoscopic cholangioplasty; PCP: percutaneous cholangioplasty. 

Quality assessment of the studies

The data of the selected primary studies selection are described in Figure 3.

The quality assessment of the studies including study design, level of evidence and New Castle score (Ottawa Quality Assessment Cohort Studies)⁽¹⁶⁾ were demonstrated in Figure 4.

FIGURE 4. Quality assessment of the studies – study design, level of evidence and New Castle (Ottawa Quality Assessment Cohort Studies) 

DISCUSSION

Biliary strictures after OLT are recognized as the Achilles’ heel of OLT. They are usually classified in anastomotic biliary strictures (ABS) and non-anastomotic biliary strictures (NABS). The involved etiological factors seems to be the small bile duct diameter, a big disproportion between graft and recipient bile ducts, the ischemia time, tension on the anastomosis, infection and other technical issues⁽⁸⁾. On the other hand, ischemia related to hepatic artery problems seems to be more related to the occurrence of NABS^{(2, 13)}. NABS can present as a single or multiple stenosis and has a reserved prognosis when compared to ABS⁽¹⁸⁾. That is the reason why the present systematic review focused on ABS. We noticed that sometimes the two conditions were included in the cohort of patients but their results were not presented separately making impossible to evaluate the results of the treatment adequately.

Multiple sessions of balloon dilation followed by the endoscopic placement of plastic stents is the most commonly adopted strategy for the treatment of ABS after OLT⁽⁵⁾. The present systematic review makes clear that there is a paucity of well-designed randomized controlled trials to support this strategy. It is possible that several centers have used their expertise and experience in treatment of iatrogenic lesions of the biliary tree, and they may have brought the same concept for the treatment of ABS after OLT.

Concerning the route of access, there is only one retrospective comparative trial comparing the transhepatic and the endoscopic accesses⁽¹⁰⁾. In both groups the treatment was based on several sessions of dilation and stent insertion. There were no significant differences except for the fact that more sessions were needed in the transhepatic group. The authors attribute this occurrence to the accidental dislodgement of the external tube when the patients were at home. In this study, non-anastomotic biliary stricture cases occurred roughly in 30% of both groups. The authors recognized non-anastomotic biliary stricture as a predictor of failure of the dilation/stenting treatment. Considering that the transhepatic route is more invasive, and usually implicates in external drainage at some point of the treatment, it is easy to understand the widespread preference for the endoscopic access.

Two studies compared balloon dilatation versus balloon dilatation with plastic stenting for the treatment of ABS after OLT^{(9, 17)}. One of them⁽⁹⁾ was initiated as a prospective randomized trial (registration at ClinicalTrials.gov NCT00487201). As some of the patients were not randomized the authors changed the trial design to a prospective comparative study. The initial clinical success and sustained clinical success rates were similar, around 95% and 70%, respectively. More complication events were observed in the stent group. The main reason was the occurrence of stent migration and occlusion. The other study⁽¹⁷⁾ was a retrospective comparative one. Initial success rates were similar in both groups, around 90%. However in the balloon dilatation group stenosis recurrence rate was higher when compared to the combo treatment arm, roughly 60% and 30% respectively. The complication rates were similar between the groups. The present meta-analysis did not show any difference between the initial or sustained clinical success rates or the complication event rate when balloon dilatation and balloon dilatation plus stenting were compared. Some technical aspects may explain the above-mentioned results. In the study by Kulaksiz et al.⁽⁹⁾, the mean number of sessions of endoscopic treatment was similar in both groups (five sessions) while in the study by Zoepf et al.⁽¹⁷⁾, the mean number of sessions was double in the combo treatment group (four sessions) when compared to the balloon group (two sessions). Additionally, Kulaksiz et al.⁽⁹⁾ dilated the ABS up to 6 mm (18Fr) and placed just a 10Fr stent in the first session which could explain the higher complication rate which was due to stent migration in the combo treatment group. Finally, Zoepf et al.⁽¹⁷⁾ did not included stent migration as a complication. Taking in consideration these results it is possible that an intensive endoscopic treatment with several sessions of balloon dilation may achieve similar results when compared to balloon dilation and plastic stenting. If this were proved in a well-designed randomized trial, balloon dilation would be probably more cost effective in the treatment of ABS after OLT. On the other hand one could underline the fragility of the results obtained by the metanalytic model in experimental trials involving invasive procedures because the technical nuances (i.e., expertise of the operator, number of stents, dilatation pressure, duration of dilatation) are hardly controlled in that statistical model. In addition, another limitation of the results obtained by the present metanalysis was due to the scarce number of included studies and the small number of patients.

The need for several sessions of endoscopic treatment for the treatment of ABS after OLT make the rationale for the use of metallic stents which achieve a higher diameter and develop later biliary obstruction when compared to plastic stents. The first series employed partially covered stents which presented lower migration rates at the cost of embedding of the uncovered mesh into the duct wall making its removal difficult and causing irreversible inflammatory changes in the bile duct⁽⁵⁾. More recently fully covered metallic stents were employed with higher migration rates and variable results⁽⁵⁾. We found a systematic review comparing the use of multiple plastic stents with self-expandable metallic stents for the treatment of ABS after OLT⁽⁵⁾. The authors recognize that there were no well-designed randomized trials, but only small case series for both treatment strategies, making impossible a meta-analysis. They also conclude that the success rates with multiple plastic stents are possibly higher when the patients are treated for 12 months or longer. They found the same trend in patients treated with metallic stents for 3 months or longer.

We found two retrospective comparative studies comparing plastic and metallic stents^{(6, 15)}. Tarantino et al.⁽¹⁵⁾ compared the outcomes of patients with failure of the multiple plastic stenting strategy sent to metallic stenting (salvage therapy group) with patients sent directly to metallic stenting as a first line treatment (first line treatment group). The authors showed that the sustained clinical success rate was higher in the salvage therapy group when compared to the first line therapy group. The authors left the metallic stents for only 2 months in both groups. Kim et al.⁽⁶⁾ compared plastic and metallic stents both placed by percutaneously in patients with biliary strictures submitted to living donor liver transplantation. The sustained clinical success was higher in the plastic group (95 vs 70%, P<0.0005). It must be considered that the metallic stent was left in place for just 2 months.

These findings raise the concern about the lack of adequate evidence to support the early adoption of new technology in clinical practice based just on theoretical concepts and on previous experience with similar situation (i.e. endoscopic treatment of iatrogenic biliary strictures). As mentioned above, it is not proved that the combo treatment (balloon + plastic stent) is superior to balloon dilatation alone. It is possible that there are patients who would have resolution of the ABL with balloon dilatation alone. Well-designed randomized controlled trials could help to identify them.

It is also noteworthy that a search for ongoing randomized controlled trials at the site http://clinicaltrials.gov revealed that there are four trials comparing fully covered metallic stents with plastic stents for the treatment of ABS after OLT, three of them still recruiting patients. Most of them are supposed to collect the first data by the end of this year or the beginning of 2014.

In conclusion, there is a vacuum in the literature of the endoscopic management of ABL after OLT. Well-designed randomized trials comparing balloon dilation vs multiple plastic stenting vs fully covered metallic stents are warranted to define a more tailored approach to this population.