AIM: To analyze the risk factors for biliary stent migration in patients with benign and malignant strictures.

METHODS: Endoscopic stent placement was performed in 396 patients with bile duct stenosis, at our institution, between June 2003 and March 2009. The indications for bile duct stent implantation included common bile duct stone in 190 patients, malignant lesions in 112, chronic pancreatitis in 62, autoimmune pancreatitis in 14, trauma in eight, surgical complications in six, and primary sclerosing cholangitis (PSC) in four. We retrospectively examined the frequency of stent migration, and analyzed the patient factors (disease, whether endoscopic sphincterotomy was performed, location of bile duct stenosis and diameter of the bile duct) and stent characteristics (duration of stent placement, stent type, diameter and length). Moreover, we investigated retrieval methods for migrated stents and their associated success rates.

RESULTS: The frequency of tube stent migration in the total patient population was 3.5%. The cases in which tube stent migration occurred included those with common bile duct stones (3/190; 1.6%), malignant lesions (2/112; 1.8%), chronic pancreatitis (4/62; 6.5%), autoimmune pancreatitis (2/14; 14.3%), trauma (1/8; 12.5%), surgical complications (2/6; 33.3%), and PSC (0/4; 0%). The potential risk factors for migration included bile duct stenosis secondary to benign disease such as chronic pancreatitis and autoimmune pancreatitis (P = 0.030); stenosis of the lower bile duct (P = 0.031); bile duct diameter > 10 mm (P = 0.023); duration of stent placement > 1 mo (P = 0.007); use of straight-type stents (P < 0.001); and 10-Fr sized stents (P < 0.001). Retrieval of the migrated stents was successful in all cases. The grasping technique, using a basket or snare, was effective for pig-tailed or thin and straight stents, whereas the guidewire cannulation technique was effective for thick and straight stents.

CONCLUSION: Migration of tube stents within the bile duct is rare but possible, and it is important to determine the risk factors involved in stent migration.

Endoscopic biliary stenting using a tube stent is an accepted therapy for biliary obstruction developing secondary to malignant or benign disease[1]. The complication rate associated with biliary stent use reportedly ranges from 8% to 10%[2-4], and the described complications include cholangitis, cholecystitis, duodenal perforation, bleeding, pancreatitis, stent fracture, proximal stent migration, distal stent dislocation, and stent occlusion resulting in recurrent biliary obstruction[5-16]. Stent migration is a rare complication associated with biliary stenting[11,17,18]. Most tube stents are pig-tailed shaped with flaps at each end, to prevent proximal or distal migration. However, stents may migrate proximally or distally as a late complication of endoscopic stenting in 5%-10% of patients who have undergone biliary stenting[19]. Proximal stent migration can cause biliary obstruction and cholangitis, thus requiring retrieval or re-stenting. Therefore, it is important to determine the factors influencing biliary tube stent migration that have not yet been documented. A few studies have reported on the risks of stent migration[17]. Johanson et al[17] reported that stent migration occurs in cases of cholangiocarcinoma, larger diameter stents, and short stents. In addition, they concluded that the reason for migration was malignant stricture of the bile duct or, in cases of benign disease, advancement of a stricture that was previously present. In addition, the presence of multiple stents, a broken stent, and differences in physical constitution are proposed as factors causing stent migration. The retrieval of migrated stents is technically challenging, but may usually be achieved endoscopically, by using forceps, snare, or balloon techniques, and rarely requires surgical intervention[18,20-22]. In the present study, we aimed to determine the frequency of biliary stent migration; analyze the risk factors for proximal stent migration; and describe the methods used for retrieval of migrated stents.

Endoscopic biliary stenting using a tube stent is now a well-established therapy for biliary obstruction developing secondary to malignant or benign disease[1]. However, certain complications may develop with this technique and should be carefully considered. The complication rate for biliary stents reportedly ranges from 8% to 10%[2-4], and the complications include cholangitis, cholecystitis, duodenal perforation, bleeding, pancreatitis, stent fracture, proximal stent migration, distal stent dislocation, and stent occlusion resulting in recurrent biliary obstruction[5-16]. Stent migration, a rare complication associated with biliary stenting[7,8,12], is a late event following endoscopic stenting. It occurs in 5%-10% of patients who undergo biliary stenting, and may involve proximal or distal migration[7,8,15,17,18,20,23-28]. Previous studies have reported that 3.1%-4.9% and 3%-6% of patients undergoing biliary stenting experience proximal and distal stent migration, respectively[5,7,17,18,24,25,29,30]. In the present study, the frequency of stent migration was 3.5%, which is consistent with previous studies.

Stent migration can lead to symptoms of biliary obstruction, therefore, it is important to determine the risk factors of stent migration[17,20,31]. However, thus far, the risk factors for biliary stent migration have yet to be clarified. Arhan et al[32] reported that proximal stent migration occurs in cases with cholangiocarcinoma, short stents, or stents with large diameters. Moreover, the placement of multiple stents and breaking of stents may cause stent migration. In addition to these risk factors, we also analyzed other possible risk factors including subtypes of both benign and malignant biliary disease; EST use; location of the stenosis; duration of stent placement; and stent shape, diameter, and length.

A previous study has indicated that stent displacement is less frequently noted in malignant than in benign biliary stenosis, which may be attributed to the tight covering of the stents by malignant tissue[32]. Similarly, in the present study, the frequency of migration was significantly higher in cases of benign than malignant disease (P = 0.030). One possible explanation for this phenomenon is that stenosis in benign disease is not as tight as that in malignant disease, which may be due to differences in the resolution of local inflammation. In cases of malignant stenosis, tumor growth may help to anchor the stent, thus preventing its migration. Arhan et al[32] reported that stent migration was observed less frequently in post-cholecystectomy strictures than in other benign biliary stenosis, except in cases of PSC[32]; the authors had indicated that the tightness of the fibrotic stricture could be a possible explanation for this finding. However, it was difficult to assess these hypotheses due to the low number of available cases in the present study.

Data on whether undergoing an EST before placement of a biliary stent affects the risk of migration are scarce. EST did not significantly affect the frequency of stent migration in previous studies[17,27,29,33]. Similarly, in the present study, the frequency of stent migration was not significantly higher in the cases who had undergone EST compared to those who had not undergone EST (P = 0.40). However, one study indicated a higher frequency of stent migration in patients who had undergone stent placement without prior EST compared to patients who had undergone stent placement after undergoing EST. The higher migration rate in these patients was primarily because of distal migration, whereas the incidence of proximal stent migration was not influenced by EST[34]. Thus, undergoing EST prior to stent placement cannot be considered as a proven risk factor for proximal stent migration.

With regard to stricture location, we generally believe that distal stent dislocation occurred more frequently in cases with stenosis of the upper bile duct, whereas proximal stent migration occurred more frequently in those with stenosis of the lower bile duct. In the present study, the frequency of migration was significantly higher in cases with stenosis of the lower common bile duct when compared with hilar stenosis or stenosis of the middle common bile duct (P = 0.031). However, in cases with stenosis of the lower common bile duct, it was unclear whether the migration was related to stent length.

We believe that a greater amount of space in the proximal bile duct is necessary for migration. Therefore, we hypothesized that the stent would be more likely to migrate proximally in cases with a larger bile duct diameter. In the present study, the frequency of migration was significantly higher in cases with a bile duct diameter > 10 mm compared with ≤ 10 mm (P = 0.023).

Moreover, in the present study, the frequency of migration was also significantly higher in cases with stent placement duration > 1 mo compared with ≤ 1 mo (P = 0.007). A longer duration of stent placement increased the risk of migration. Therefore, we suggest that the stents should be exchanged or removed after a short placement interval.

Biliary stent design may also affect the risk for migration. With regard to stent shape, we used pig-tailed shaped and straight tube stents only in the present study. Straight tube stents have been modified with side flaps or barbs to decrease the risk of migration[35]. Pig-tailed shaped stents were also used to decrease the risk of migration. In the present study, the frequency of stent migration was significantly higher in cases with straight-type stents when compared with pig-tailed stents (P < 0.001). We noted that the presence of side flaps or barbs was not sufficient to prevent stent migration. The diameter and length of the stents may also be associated with the risk of migration[17]. With regard to stent diameter, in the present study, the frequency of migration was significantly higher in the cases with 10-Fr stents compared with 7-Fr stents (P < 0.001). We believe that, because stenotic lesions of the bile duct were already improved in cases in which thick stents (such as 10-Fr stents) were used, these cases were more likely to experience migration. With regard to stent length, in the present study, no significant differences in the frequency of migration were noted among the patients who had 5-, 7- and 9-cm stents. Arhan et al[32] have reported that shorter stents tend to migrate proximally, whereas longer stents tend to migrate distally, in cases of benign biliary stenosis. Moreover, longer stents in the bile duct are less likely to migrate because a longer portion is fixed in the common bile duct, thus limiting proximal movement[17]. I think that we should pay attention to using shorter and straight stents in high-risk patients. I recommend using pig-tailed type stent for those cases.

Most of the stents that have migrated proximally can be successfully retrieved indirectly, through stone extraction with a balloon catheter, or directly by using various grasping accessories. All the stents were successfully retrieved in the present study, therefore, none of the patients required surgery for stent retrieval. In patients with benign strictures, this favorable finding could be a result of the dilation of the stricture due to further stenting, thereby facilitating the retrieval of the proximally migrated stent.

In conclusion, the risk of stent migration is higher in benign than in malignant biliary stenosis. Proximal migration of biliary or pancreatic stents is an infrequent occurrence. Proximal stent migration was found to be closely associated with benign stenosis, stenosis in the lower common bile duct, bile duct diameter > 10 mm, stent placement duration > 1 mo, straight-type stents, and 10-Fr stents. The migrated stents can be extracted endoscopically, with a high degree of success, using a variety of techniques involving baskets or balloons.