Assessment of an intraductal metallic stent in the management of postcholecystectomy biliary strictures

Abstract

Background and Aims

Postcholecystectomy bile duct strictures are currently managed by using multiple plastic stents. The use of metallic stents has been considered a contraindication due to fear that the metallic stent could block one of the main hepatic ducts. This study is the first to assess the use of an intraductal, fully covered metallic stent in postcholecystectomy strictures, including those with close proximity to the hepatic hilum.

Methods

This study was a retrospective analysis of 11 patients who underwent insertion of an intraductal metallic stent for the management of postcholecystectomy strictures. In those strictures close to the hepatic hilum (Bismuth types II, III, and IV), a 7F plastic stent was inserted in the contralateral hepatic duct to maintain its patency. Stent extraction was planned after 6 to 12 months.

Results

All stents were inserted successfully. After a median indwell time of 10 months (range, 3-15 months), all stents were successfully extracted. Strictures resolved in 10 patients (91%) at the time of stent extraction. One patient had premature extraction of the stent at 3 months due to stent obstruction. After a median follow-up of 24 months (range, 10-52 months), stricture recurrence occurred in 4 (40%) of 10 patients. No stent migration occurred, nor were there any cases of unilateral bile duct obstruction by the metallic duct.

Conclusions

In postcholecystectomy biliary strictures, intraductal, fully covered metallic stents were effective even in strictures close to or involving the biliary bifurcation when an additional 7F plastic stent was inserted. Effective long-term stricture resolution was achieved in a substantial proportion of patients using a low number of endoscopic procedures and no reported stent migration.

Postcholecystectomy bile duct strictures occur with an incidence of about 0% to 0.6% after laparoscopic cholecystectomy.1, 2, 3 Etiology is variable, with the main culprits including ischemia, common bile duct (CBD) ligation, or cystic duct ligation excessively close to the CBD. Endoscopic management using plastic stents has been established as the main therapeutic modality.⁴ The most commonly used protocol (multiple plastic stent [MPS] technique) is sequential balloon dilatations every 3 to 4 months with increases in the number of plastic stents reaching up to three or four 10F plastic stents to maintain lumen patency.^4,5 The technique has a high long-term success rate of about 74% to 89% with relatively low morbidity.^5,6 However, drawbacks include the need for repeated procedures over a long period and risk of stent migration.

The concept of using fully covered metallic stents offers the advantage of being able to completely dilate the stricture with a large-caliber stent in one session. Metallic stents have been used successfully in benign biliary strictures such as chronic pancreatitis.⁷ Postcholecystectomy strictures have an anatomical peculiarity rendering the use of metallic stents difficult and in general, prohibited; the fact that the strictures are always very close to the endoscopic hilum puts the risk of blocking one side of the hepatic ducts by the large metallic stent. To date, strictures less than 2 cm away from the hilum have been considered a contraindication for fully covered metallic stents.⁸ The other problem of fully covered metallic stents is the risk of migration and risk of blockage by duodenal food contents.⁷

A special stent was designed with specific features to overcome the usual drawbacks of conventional biliary metallic stents.⁹ The Kaffes stent (Taewoong Medical, Gyeonggi-do, South Korea) is a short, completely intraductal, fully covered metallic stent with a long retrieval lasso and a fusiform design aimed at reducing the risk of migration. To overcome the risk of blocking the contralateral bile ducts, a 7F, 12 cm plastic stent is inserted in the contralateral duct to avoid obstruction by the stent. The Kaffes stent has previously shown success in the management of some benign biliary strictures such as those developing after cadaveric liver transplantation.9, 10, 11 Studies on the use of metallic stents in postcholecystectomy strictures, however, remain sparse. The largest study assessed the use of conventional metallic stents in postcholecystectomy strictures in 18 patients⁸; it excluded any cases with strictures within 2 cm from the hepatic hilum, although, in reality, these compose the majority of postcholecystectomy strictures.

This is the first case series describing the use of an intraductal metallic stent in postcholecystectomy strictures, including cases in which the stent bypasses the hepatic hilum.

Methods

This was a retrospective analysis of all patients who underwent the insertion of a Kaffes stent for the management of postcholecystectomy strictures at our institution.

Inclusion and exclusion criteria

Inclusion criteria included the following: confirmed benign postsurgical biliary stricture with 3 criteria (US imaging or MRCP showing biliary dilatation with a proximal CBD stricture without depiction of a mass, raised serum bilirubin and alkaline phosphatase levels, and history of cholecystectomy within the previous year); had undergone an attempt at Kaffes stent insertion endoscopically; was naïve to endoscopic therapy before attempt at insertion of a Kaffes stent; had at least 6 months’ follow-up after extraction of the Kaffes stent; and age >18 years.

The following patients were excluded from attempting Kaffes stent insertion according to our institution protocol: patients with severe coagulopathy, liver transplant recipients or donors, and those unable to adhere to follow-up.

Procedure

Following receipt of patients’ written informed consent, all procedures were performed with the patient under propofol sedation by a single experienced endoscopist. Strictures were identified by cholangiography and then dilated by using a controlled radial expansion balloon (diameter 6-10 mm, at the discretion of the endoscopist). We routinely dilate for 2 reasons: first, to facilitate the insertion of the metallic stent; and second, we believe that dilatation and rupture of the fibrotic bands could ameliorate the long-term resolution of the strictures. A partial sphincterotomy was performed just to facilitate the passage of the stent. A 4 or 6 cm, fully covered expandable metallic stent with 8 mm central diameter and 10 mm peripheral diameter was then deployed at the level of the stricture (Kaffes stent) (Fig. 1).

Figure 1.

Kaffes stent.¹⁹

In cases close to the hepatic hilum in which the deployment of the stent was expected to reach 1 duct and possibly block another duct, a 7F stent was inserted before deployment of the metallic stent in the contralateral duct (Figs. 2 and 3). Patients were allowed oral intake 2 hours after the procedure and were discharged on the same or following day; prophylactic antibiotics were prescribed (ciprofloxacin 500 mg twice daily for 3 days). Stent removal was planned after 6 to 12 months. During the extraction procedure, the distal lasso of the stent was grasped by a forceps and pulled together with pulling of the endoscope. Cholangiography was conducted to confirm stricture resolution (occlusion cholangiogram using an extraction balloon), and the CBD underwent extraction balloon sweeps to expel any sludge or stones.

Figure 2.

A, Bismuth type II postcholecystectomy stricture. B, Kaffes stent inserted extending into the right hepatic duct with a 7F plastic stent in the left hepatic duct. C, Eight months after stent insertion, note the full expansion of the Kaffes stent and its completely intraductal position; note also the slight but insignificant migration of the 7F plastic stent. D, After stent removal, complete resolution of the stricture.

Figure 3.

A, Bismuth type I postcholecystectomy stricture. B, Dilatation of the stricture by a 6 mm balloon. C, Kaffes stent deployed with upper edge situated just below the bifurcation and thus with no need for plastic stent insertion. D, After stent removal, complete resolution of the stricture.

Follow-up

In cases in which the stricture appeared to have resolved at the extraction procedure, patients underwent follow-up laboratory tests and abdominal US at 1, 3, and 6 months and yearly thereafter to detect any recurrence. If at any point there was a suspicion of recurrence, MRCP and/or ERCP were performed for confirmation. In cases in which the stricture did not resolve at the extraction procedure, the patients were offered the option to undergo surgery or an attempt at multiple plastic stent placement.

Definitions

• •Technical success: successful insertion of the metallic stent in the required position (and the 7F plastic stent when necessary with the metallic stent).
• •Stricture resolution: disappearance of the stricture and free flow of contrast during ERCP at the time of stent removal.
• •Clinical success: absence of clinical or laboratory evidence of obstruction after stent insertion (absence of jaundice, cholangitis, normal bilirubin, and alkaline phosphatase).
• •Recurrence: clinically significant recurrence of the stricture (abnormal laboratory parameters and biliary dilatation) occurring after stricture resolution, confirmed by MRCP or ERCP.
• •Stent migration: displacement of the stent to a position where it is no longer covering the site of stricture.

Results

Study population

A total of 11 patients underwent insertion of a Kaffes stent for a postcholecystectomy stricture (Table 1). Eight (73%) had undergone a laparoscopic cholecystectomy, and 3 (27%) had an open one. The median interval between cholecystectomy and the stent insertion procedure was 20 days (range, 5-360 days). Cholangitis was the main presentation in 7 patients, and 4 presented with jaundice. Five patients (45%) underwent percutaneous transhepatic drainage before ERCP. Three (27%) had a concomitant leak with the biliary stricture. All strictures were in the proximal CBD and/or common hepatic duct. Seven strictures (55%) were ≤1 cm in length, and only four strictures were more than 1 cm.

Table 1.

Study population’s demographic characteristics and nature of strictures

Characteristic	Value
Age, y	47.5 (34-67)
Gender
Female	5 (45%)
Male	6 (55%)
Type of surgery
Laparoscopy	8 (73%)
Open	3 (27%)
PTD before endoscopy	5 (45%)
Timing of first endoscopic intervention (days after surgery)	20 (5-360)
Length of stricture
≤1 cm	7 (64%)
>1 cm	4 (36%)
Stricture classification
Bismuth type I	1 (9%)
Bismuth type II	5 (45%)
Bismuth type III	5 (45%)
Concomitant leak	5 (45%)
Plastic stent inserted with metallic stent	7 (64%)
Metal stent in-dwell time, mo	10 (3-15)

Values are median (range) unless otherwise indicated.

PTD, Percutaneous transhepatic drainage.

Procedure

A Kaffes stent was inserted successfully in all patients (five 4 cm, six 6 cm). Plastic stents inserted in the contralateral duct were deemed necessary in 7 patients (64%), and all were successfully inserted (technical success, 100%). All patients showed clinical resolution of obstructive symptoms and laboratory parameters after stent insertion (clinical success 100%), and all 5 cases with biliary leak had immediate resolution of their leaks (Table 2).

Table 2.

Outcomes

Technical success	11 (100%)
Clinical success (resolution of obstruction after stent insertion)	11 (100%)
Stricture resolution at time of stent extraction	10 (91%)
Duration of follow-up after stent extraction, mo	24 (10-52)
Stricture recurrence	4/10 (40%)
No. of endoscopic procedures per patient	2 (2-3)
Adverse events
Cholangitis	4 (36%)
Obstruction of contralateral ducts by metal stent	0 (0%)
Stent migration	0 (0%)
Difficulty in stent extraction	1 (9%)
Bleeding	0 (0%)
Pancreatitis	0 (0%)
Perforation	0 (0%)

Values are median (range) unless otherwise indicated.

The median indwell time of the Kaffes stent was 10 months (range, 3-15 months). One patient had an early extraction of the stent at 3 months due to obstruction caused stones and cholangitis, and it was not possible to remove the stones without removing the stent. This patient underwent subsequent successful management with multiple plastic stents. In the remaining 10 patients, all stents were successfully extracted.

After the stents were extracted, cholangiography revealed stricture resolution in all patients (100%). No cases of stent migration were reported. Biliary stones were noted in 10 of the 11 patients (91%) at the time of stent removal. All were extracted successfully during ERCP; one of them, however, necessitated premature extraction of the Kaffes stent. The median number of procedures was 2 ERCPs per patient (range, 2-3), with only 3 patients requiring 3 endoscopic procedures (2 to remove stones clogging the metallic stents, and 1 ERCP 3 months after stent removal to clear residual stones).

Follow-up

Median follow-up duration after removal of the Kaffes stent was 24 months (range, 10-52 months). Of the 10 patients who had the stent in place for more than 6 months, 4 patients (40%) had recurrence of the stricture confirmed by imaging and laboratory findings after a median of 7.5 months (1, 5, 15, and 20 months). Three refused surgery and were managed by repeated multiple plastic stents, and 1 underwent a successful hepaticojejunostomy with no adverse events. An intention-to-treat approach was used, and the success of the Kaffes stent in achieving a durable stricture resolution was therefore 55% (6 durable successes without recurrence, 4 recurrences after stent removal, and 1 premature removal due to cholangitis).

Adverse events

Cholangitis occurred in 4 patients (36%). One patient experienced cholangitis immediately after stent insertion and was successfully treated conservatively with antibiotics. Three attacks of cholangitis occurred during the indwelling time of the stent at 3, 4, and 6 months. One of them (at 3 months) necessitated early stent extraction, and 2 underwent ERCP for extraction of the stones clogging the metallic stent. One case developed CBD stones 3 months after stent extraction, and these were successfully extracted endoscopically with no evidence of stricture recurrence. No cases of bleeding, perforation, or pancreatitis occurred. No cases of unilateral biliary obstruction caused by the Kaffes stent occurred. In 1 patient, stent extraction was difficult due to some adherence of the stent to the bile duct wall; it was, however, eventually extracted after some maneuvering of the endoscope and twisting of the stent together with forceful pulling.

Discussion

Management of postcholecystectomy strictures using multiple plastic stents has been established as the reference standard for such adverse events.¹² However, this technique has noticeable drawbacks, namely the necessity of at least 4 to 5 endoscopic procedures and the high rate of migration of the plastic stents.^4,12 Metal stents offer many theoretical advantages over multiple plastic stents, and this has been tested and proven in benign biliary strictures other than postcholecystectomy strictures.⁷ To date, only a few case reports, and 1 study with 18 patients, assessed the use of metallic stents in postcholecystectomy strictures.8, 9, 10, 11 The reason for this scarcity of studies on such a common medical condition is the fact that, to date, endoscopists have been reluctant to insert a fully covered metallic stent bypassing the hepatic confluence with the fear of blocking the biliary drainage from the contralateral duct. Postcholecystectomy strictures are almost always very close to the hepatic hilum, and it is thus seldom possible to insert a metallic stent while being confident it would not bypass the hepatic confluence. Even the single study that reported on the use of metallic stents in postcholecystectomy strictures excluded those strictures less than 2 cm from the hepatic hilum.⁸

The technique we endorsed in our patients to overcome the problems of metallic stents was the insertion of a plastic stent in the contralateral duct to avoid obstruction; a 7F stent was deemed sufficient as drainage would also occur in the grooves between the plastic and metallic stents. Our initial experience seems to confirm this hypothesis, as none of our patients developed unilateral bile duct obstruction by the stent, even though at least 7 patients had their metal stents definitely extending proximally above the hepatic duct confluence and required a plastic stent to be inserted simultaneously. There was also the concern that a 7F plastic stent is unlikely to remain patent for 6 to 12 months, but it seems that the grooves created between the plastic and metallic stent add drainage routes other than the lumen of the plastic stent and thus reduce the chance of biliary obstruction.

All patients experienced relief of their obstructive symptoms after insertion of the metallic stents. With the exception of 1 patient who had a premature extraction of the stent due to early cholangitis, all patients had stricture resolution at the time of stent extraction; this compares with 74% to 90% with the use of the MPS technique.^5,6,13 Perhaps the more important result, however, is the durability of the response; in this preliminary experience with a prolonged follow-up (median, 24 months), 60% of patients had no recurrence of their stricture (40% recurrence rate). This compares favorably (or perhaps slightly inferiorly) with the results reported with the MPS technique, in which the recurrence rate ranges between 0% and 33%.^5,6,13, 14, 15 It should be noted that we had a particularly difficult-to-treat study population; 5 of the 11 patients had very proximal strictures (Bismuth type III), which are known to have a worse outcome than more distal strictures.¹⁵ In the study by Draganov et al,¹⁵ more distal postoperative lesions (Bismuth types I and II) were associated with better success rates compared with proximal hilar lesions (Bismuth type III) (80% vs 25%). In addition, almost one-half of our patients had extremely tight strictures not passable endoscopically, and they necessitated percutaneous transhepatic drainage before the endoscopic approach. Collectively, these findings show that this was a very hard-to-treat study group, and the achieved results are therefore considered very encouraging.

The advantage of the metallic stent technique is shown in this preliminary experience with the lower number of endoscopic procedures per patient. The median number was 2 ERCPs per patient, with only 3 patients requiring 1 additional endoscopic procedure (to remove stones clogging the stents in 2 and one ERCP after stent removal to clear residual stones). This compares to an average of 4 to 5 endoscopic procedures with the standard MPS technique.4, 5, 6^,13, 14, 15, 16 Although number of procedures was not specifically assessed in this study, we would expect that the lower number of procedures will be reflected in shorter hospitalizations and probably a lower overall cost. The other advantage shown in this study is the absence of any stent migration; this compares to a migration rate of about 23% to 28% with the MPS technique,^6.13, 14, 15, 16 which also adds to the number of required ERCPs with the MPS technique.

Perhaps the only appreciable drawback of this technique is the high rate of development of biliary stones, as in almost all patients, multiple stones were detected at stent extraction, and stent obstruction by stones occurred in 2 patients. Why these stones develop more commonly with metallic stents compared with multiple plastic stents remains hypothetical. The MPS technique involves stacking the CBD with the stents, leaving no room for sizeable stones to develop; the nature of the nitinol metal mesh or silicone-covering membrane may also play a role in stimulating stone formation. One other explanation is that plastic stents are exchanged more frequently, which clears the CBD frequently from any sludge or gravels. This remains an area for further research and development of metal stents to reduce the rate of stone formation within. Adding oral ursodeoxycholic acid to the treatment regimen of patients with indwelling stents may reduce the risk of stone formation.¹⁷

The choice of Kaffes stent in particular seems to be advantageous in the setting of the proximal postcholecystectomy strictures. Being completely intraductal, the stent is less exposed to duodenal peristalsis and thus less likely to migrate. None of our metallic stents migrated compared with a transpapillary position with a rate of migration of 17% to 31% reported in the literature.^8,18 The preservation of the intact papillary sphincter also reduces the risk of stent obstruction by alimentary duodenal contents. It remains to be shown in larger studies if this advantage will be reflected in lower rates of stent obstruction and cholangitis, as currently the rate of stent obstruction (3 of 11 [27%]) seems to be similar to that with standard transpapillary metallic stents (27%-31%).^7,8,18

In conclusion, this study is the first to assess the use of metallic stents in postcholecystectomy strictures, including those in close proximity to the biliary confluence. Short intraductal, fully covered metallic stents seem to be feasible and effective even in strictures close to or involving the biliary bifurcation when an additional 7F plastic stent is inserted. Effective long-term resolution of the strictures is achieved in a substantial proportion of patients using a low number of endoscopic procedures and no reported migration of the metallic stents. Further prospective studies are required to validate these preliminary results as well as a randomized trial comparing the current approach with the MPS technique.

Disclosure

All authors disclosed no financial relationships.