Controversies and technical issues in EUS-guided drainage: a report of three cases

Interest in EUS-guided biliary drainage is increasing and this approach is candidate to establish as an effective alternative to percutaneous transhepatic cholangiography, when endoscopic access to the papilla is  precluded because of duodenal stenosis, periampullary tumor invasion, or previous surgery; nevertheless this method is far from being standardized and  many technical issues need to be clarified. [1-6] We present hereby three cases from our experience which we think they can exemplify some of the choices and difficulties, that could be faced by the operators novice in EUS-guided drainage.

Case 1#. In August 2012, a 78 year old man, with a history of severe ischemic cardiomiopathy and peripheral vascular disease was referred to our institution for vomit and mild jaundice.

At the referring institution a duodenal post-bulbar tight stenosis and a partially cystic pancreatic head mass had been diagnosed. A complete endosonographic study of the pancreatic head was impossible, due to the stenosis: nevertheless EUS-FNA demonstrated the diagnosis of malignant IPMN. Although no signs of vascular invasion by the mass had been identified at computed tomography, the patient declined surgery, after thorough discussion of the high surgical  risk.

A trans-hepatic EUS-guided drainage was proposed. A dilated biliary duct in left liver lobe was punctured by a conventional 19G needle and the fistula was dilated up to 10 Fr, using graduated bougies over a 0.025 guidewire.  

The bile duct was dilated upstream of a short distal stenosis and it was easy to pass the guidewire into the duodenum.

A Cook 635, 10 mm, 6 cm, self-expanding stent was positioned anterogradically across the papilla over the guidewire; moreover a 10 Fr, 9 cm plastic stent was deployed bridging the hepato-gastric fistula. During the same session a duodenal 22 mm self-expanding stent was positioned (Figure 1, A-B).

 

The treatment was effective and the patient was discharged. One year later, the patient presented vomit again; the duodenal stent was partially occluded by ingrowth.The transhepatic fistula was sealed and the expulsed plastic stent was hitched onto the duodenal stent (Figure 1, C).

 

We removed the plastic stent and placed another duodenal stent inside the old one (Figure 1, D).

 

The patient is still alive and reasonably well (April 2014); the biliary stent is still patent, but two other duodenal stents had to be placed for recurrence of the duodenal stenosis.

Case 2# A 50 year old man underwent explorative laparotomy for duct cell pancreatic adenocarcinoma.

The neoplasm was known to have caused a tight stenosis of the post-bulbar duodenum, but despite the pre-operative work-up the vascular invasion had not definitely ascertained.  

At surgery, the tumor was confirmed to be not resectable and a gastro-jejunum, but not a bilio-digestive anastomosis was performed; palliative chemo-radiotherapy was started. Some months after, in January 2013, jaundice appeared and EUS-guided drainage was performed. The bile duct was punctured with a standard 19G needle by the duodenal bulb and the fistula was dilated up to 8.5 Fr, using graduated bougies. A partially covered expanding stent, diameter 10 mm, length  8 cm, was then placed (Figure 2, A-B).

 

Jaundice was resolved but from the placement of the stent to December 2013 (when the patient died), seven episodes of cholangitis occurred: in four episode the intrahepatic ducts were not dilated, with diffuse aerobilia at ultrasonography, thus endoscopy was not performed and the episodes were resolved by antibiotic therapy. In the other three episodes ERC was performed; in two cases the stent appeared obstructed by biliary sludge and alimentary residual, that were removed by extractor balloon (Figure 2, C-D);

 

 

in the latter case (9 months after first drainage procedure) a clear neoplastic ingrowth was diagnosed and a further stent was positioned.

Case 3#. A 64 year old man underwent duodeno-pancreatic radical resection for duct cell carcinoma, in  February 2012. Adiuvant chemotherapy was administrated, but in November 2013 a recurrence near the biliary anastomosis was demonstrated during a routine follow-up TC scan and on March 2014 the patient was referred for jaundice. A transhepatic EUS-guided drainage was attempted. We were not able to identify and to bridge the stenotic bilio-digestive anastomosis, thus we dilated the transgastric hepatic fistula by a 6 Fr dilator and subsequently by a 6 Fr cystotome (Endoflex) and positioned a half-covered self-expanding stent (Taewoong Niti-S,  GIOBOR type, diameter 10 mm, length  8 cm. Figure 3, A).

 

A month later, jaundice recurred with mild cholangitis; at endoscopy the stent appeared to have migrated proximally inside the fistula; the opening of the fistula was still partially patent and traces of purulent material were intermittently pouring from it (Figure 3, B).

 

It was very easy to put an uncovered stent, diameter 10 mm, length 8 cm, bridging the migrated stent and the fistula (Figure 3, C).

 

The episode of cholangitis was resolved rapidly and the patient was discharged the next day.

Our patients, affected by pancreatic neoplasm, had a relative long survival requiring an effective management of biliary and duodenal obstruction. In patient n. 1, despite the successful anterograde placement of a transpapillary self-expanding stent, we also deployed a plastic stent bridging the hepato-gastric fistula, as we feared the risk of bile leak; this approach was suggested in early experiences of EUS-guided anterograde transpapillary stenting. [7] At the beginning of 2012, the first large series of patients treated by a single-step EUS-guided biliary drainage with anterograde placement of a self-expanding stent was published [8]; in these series the bilio-digestive fistula had been dilated (without the use of cautery) sufficiently large so as to allow the introduction of the stent; at the end of the manoeuvre the guidewire was removed and the fistula was allowed to seal. Indeed in the published experiences, the incidence of biliary leak after anterograde placement of a transpapillary stent is low, probably because the pression in the bile ducts is reduced after successful placement of a stent downstream. Notwithstanding that the number of treated patients is still limited and larger experiences or comparative studies (with or without a second stent to protect the fistula) might be useful. In consequence of their low diameter and flexible introducer device (6 Fr) Zilver 635 stents seem to be attractive, when a EUS-guided anterograde placement of transpapillary stent is planned.

In our patients n. 3. we placed half-covered self-expanding stents This stent was purposely designed for use in EUS-guided drainage. The covered half of the stent bridges the fistula, in order to avoid bile leak, while the uncovered proximal half should anchor the stent inside the bile duct; the distal covered end is flanged to reduce the risk of proximal migration. It is likely that the proximal stent migration in patient 3, had been favored by the use of cautery to dilate the fistula, but occurrence of migration with also this type of stent had been reported previously. [9] It is conceivable that the stents actually available for EUS-guided biliary drainage have a limited capacity to hold in contact the bile duct and the digestive tract, until the fistula is mature. A practical advice may be to deploy the stent with a relatively long tract (1,5-2 cm), outside the fistula in the stomach. In a very interesting review, dealing with practical tips in EUS-guided biliary drainage, Itoi et al. [5] observed that the distance between the puncture site on the gastric wall and the liver may change, depending on the position of the patient, the amount of gas, fluid and food in the stomach. Fabbri et al. did not observe stent migration in a 9 case series of choledocho-duodenostomies performed using partially covered nitinol stents (Wallflex); the fistula was dilated by needle-knife sphincterotome. [10] Recently a single case was reported of the use of a stent designed for pseudocyst drainage for bridging a choledo-duodenal fistula; it seems unlikely that this type of stent could be used in transhepatic drainage. [11] However, we guess that the available stents are still suboptimal and we await further improvements. Moreover to the best of our knowledge, comparative studies among different types of stents, used in different modalities of EUS-guided biliary drainage are lacking. Some authors prefer graduated bogies or balloon type dilators to cautery to dilate the fistula; use of cautery might be perceived to be more risky, but again comparative studies are scarce. [12]
In patient n. 2, a high number of cholangitis episodes after stent placement was observed. We hypothesized that the position of the stent in the duodenal bulb, in the same axis of the pylorus and immediately upstream to a digestive stenosis, might have favored food impaction and ascending cholangitis. Nevertheless data on long term follow-up of patients after EUS-guided drainage are scarce and no evidence is available to hypothesize a higher incidence of cholangitis after transduodenal drainage than after transgastric drainage.
In conclusion the discussion of our cases highlighted the need of further studies focusing on technical issues inherent to EUS-guided drainage, or investigating the long term results of these type of procedures. In particular comparative studies among different technical variants and accessories are eagerly awaited.
References
1. Kumta NA, Kedia P, Kahaleh M. Endoscopic ultrasound-guided biliary drainage: an update. Curr Treat Options Gastroenterol. 2014 Jun;12:154-68.
2. Takada J, Carmo AM, Artifon EL. EUS-guided biliary drainage for malignant biliary obstruction in patients with failed ERCP. J Interv Gastroenterol. 2013 Jul;3:76-81.
3. Kedia P, Gaidhane M, Kahaleh M. Endoscopic Guided Biliary Drainage: How Can We Achieve Efficient Biliary Drainage? Clin Endosc. 201;46:543-51.
4. Khashab MA, Dewitt J. EUS-guided biliary drainage: is it ready for prime time? Yes! Gastrointest Endosc. 2013 Jul;78(1):102-5.
5. Itoi T, Isayama H, Sofuni A, et al. Stent selection and tips on placement technique of EUS-guided biliary drainage: transduodenal and transgastric stenting. J Hepatobiliary Pancreat Sci 2011;18:664–72.
6. Püspök A, Lomoschitz F, Dejaco C, et al. Endoscopic ultrasound guided therapy of benign and malignant biliary obstruction: a case series. Am J Gastroenterol 2005;100:1743-47.
7. Michel Kahaleh, Everson LA Artifon, Manuel Perez-Miranda, et al. Endoscopic ultrasonography guided biliary drainage: Summary of consortium meeting, May 7th, 2011, Chicago. World J Gastroenterol 2013; 19: 1372-1379.
8. Shah JN, Marson F, Weilert F, at al. Single-operator, single-session EUS-guided anterograde cholangiopancreatography in failed ERCP or inaccessible papilla. Gastrointest Endosc. 2012;75:56-64.
9. Pesenti C, Bories E, Caillot F, et al. Hepaticojejunal gastric anastomoses by échoendoscopies: use of prostheses GIOBOR (Tae Wong ®) partially covered. Endoscopy 2012;44: A005808fr.
10. Fabbri C, Luigiano C, Fuccio L, et al. EUS-guided biliary drainage with placement of a new partially covered biliary stent for palliation of malignant biliary obstruction: a case series. Endoscopy. 2011;43:438-41.
11. Itoi T, Binmoeller KF. EUS-guided choledochoduodenostomy by using a biflanged lumen-apposing metal stent. Gastrointest Endosc. 2014;79:715.
12. Park do H, Jang JW, Lee SS, et al. EUS-guided biliary drainage with transluminal stenting after failed ERCP: predictors of adverse events and long-term results. Gastrointest Endosc. 201;74:1276-84.

Figure Captions.
Figure 1. Patient n. 1. A) The dilated bile ducts with self-expanding stent bridging the papilla; B) At the end of procedure: the plastic stent (arrow) is barely visible in the picture; C) One year later, the plastic stent was hitched onto the duodenal stent; D) A second duodenal stent was placed.

Figure 2. Patient n. 2. A) The choledochus was dilated and it was easy to target it from the duodenal bulb. B) The stent deployed at the end of the procedure. C) During one of the numerous episode of cholangitis, the stent was clearly occluded by biliary sludge and food stuffs; note that the end of the stent was exactly in the same axis of the pylorus; D) A extractor balloon was used to remove the occluding material inside the stent; no stenosis or ingrowth was visible.

Figure 3. Patient n. 3. A) An intrahepatic bile duct was punctured by the gastric corpus B) The stent at the end of the procedure was clearly protruding 1-2 cm from the gastric wall. C) The hepato-gastric fistula was still patent after the proximal migration of the stent. D) A second stent was inserted to bridge the fistula.