Biliary leaks occur in 10 to 15% of patients after OLT and usually present with clinical symptoms earlier than strictures in the postoperative course [11
]. They are typically classified into anastomotic or nonanastomotic.
Endoscopy stands as a first-line treatment for post-OLT biliary leaks [11
]. However, many anastomotic leaks may require surgical repair [11
] and, therefore, have been defined as a complex leak [13
The main principle of endoscopic therapy for biliary leakage is to reduce transpapillary pressure gradient via transpapillary stenting with or without biliary sphincterotomy; this is conventionally performed using plastic stents [8
With the large diameter provided by metal stent and the ability to remove covered metal stents, the use of CSEMS in bile leaks was the logical next step.
This was initially described by Baron and Poterucha [15
], when they reported 3 cases of complex bile leaks successfully treated with CSEMS.
Promising results of CSEMS have also been further reported by our team with long-term leak control obtained in 87% (14/16) of patients with post-surgical bile leaks [13
]; however, long-term results were not available yet.
CSEMSs have mainly been used as a rescue therapy for patients who failed standard endoscopic therapy with plastic stenting. However, more recently has been used as a first-line measure in patients with complex and high-grade leaks [19
The rationale to deploy a SEMS through a leak is to grant the larger diameter possible allowing the patient to have the faster recovery time, with fewer sessions, and prevent further complications associated with the bile leak [21
The median indwelling of CSEMS in this study was 189, 102, and 98 days, respectively, for the PCSEMS, the FCSEMS with fins, and the FCSEMS with flare ends, respectively. The use of CSEMS for the treatment of bile leak has the theoretical advantage of decreasing the number of procedures needed to control the leakage in patients with complex or high-grade fistulas when compared to plastic stents. It is presumed that the initial increased cost associated with the use of CSEMS will be compensated by the reduction of sessions required as well as the days of hospitalization. This last point, however, remains to be proven.
All patients in the partially CSEMS group had previously failed plastic stenting (i.e., persistent leak after plastic stent placement), and, after placement of the metal stent, leak control was achieved in all of them (Tables and ).
Long-term evaluation of SEMS for the treatment of biliary leaks.
Wallstent was the first PCSEMS commercially available in the US market but led to mucosal hyperplasia at its uncovered portions and migration [12
]. Proximal migration is especially problematic since it can be associated with hyperplasia and secondary stricture after stent removal. Distal migration can lead to treatment failure.
Isayama et al. studied both radial (RF) and axial forces (AF) of Wallstent and Viabil [21
]. The results demonstrated Wallstent to have high AF, possibly related to biliary wall damage, kinking, and sludge formation and migration [21
]. In the present study, distal migration was noted in one patient (33.3%) from PCSEMS and one (10%) from FCSEMS with flare ends group. This migration rate has been reported in other studies [20
] and might be related to the respective force of the 2 CSEMSs, which are similar (Unpublished data from Isayama).
The FCSEMS with anchoring fins positioned at opposite ends was designed to prevent migration [14
]. It was found to have a very high RF in Isayama et al. study, what might cause an excessive high pressure to the biliary wall, resulting in increased tissue injury and eventual stricture [21
]. In this study, tissue injury was found in 4/5 (80%) patients when choledochoscopy was performed in this group and an increased number of post-stent-removal strictures (35%) that could be a consequence of biliary compression and ischemia. Even though migration rate was lower (5.6%), the increased number of postremoval stenosis is disencouraging.
A discussion is raised if the incidence of postremoval strictures was related to the high radial force or to the presence of fins that could stimulate tissue reaction and scarring.
Finally, FCSEMS with flare ends was recently released as fully covered SEMS, coated with premalume. It has a retrieval loop and flare ends to prevent migration. Although the leakage controlled has been encouraging in this study, we are still dealing with spontaneous distal migration (10%) and postremoval stricture (10%).
Interestingly there no statistically significant difference in either group in term of final outcome; this might be related to a type 2 error, which can be overcome by increasing the number of patients in each sample; however the complications' rate associated with all three CSEMS does not justify such a study.
Clearly, the ideal CSEMS for biliary leak is not available yet. It probably needs to be fully covered with an inert and resistant coating and has no fins, which seem to be associated to significant tissue reaction. Further CSEMS soon invading the market might offer those characteristics and need to be carefully evaluated.
In conclusion, temporary placement of CSEMS was effective to treat post-OLT biliary leaks. However, postremoval biliary stricture requiring further endoscopic treatment was seen especially with the FCSEMS with fins group. At the present time, CSEMS cannot be recommended in this patient population until major design changes have been made.