Endoscopic biliary drainage

(1)

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

van Berkel, A.M.

Publication date

2003

Link to publication

Citation for published version (APA):

van Berkel, A. M. (2003). Endoscopic biliary drainage.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

(2)

PalliationPalliation of malignant

pancreaticobiliaiypancreaticobiliaiy obstruction

A.M.. van Berkel, P. Fockens, M.J. Bruno

Departmentt of Gastroenterology and Hepatology

Academicc Medical Center, Amsterdam, The Netherlands

In:In: Ginsberg G.G., Kochman M„ Norton L, Gostout C.J. (eds). ClinicalClinical Gastrointestinal Endoscopy.

(3)

::

4 4

(4)

INTRODUCTION N

Pancreaticobiliaryy malignancies include pancreatic head cancer, gallbladder carcino-mass and proximal cholangiocarcinomas also referred to as Klatskin tumors. Pancreaticc head carcinoma comprise of tumors which may originate from various tissuess and include pancreatic adenocarcinoma, distal cholangiocarcinoma, carcino-maa of the ampulla of Vater and duodenal carcinoma. Although there are marked dif-ferencess in biological behaviour and clinical outcome between these tumors, the overalll prognosis is dismal. At the time of presentation more than 90% of patients havee local irrcsectable disease or distant metastases which leaves only a minority of patientss suitable candidates for curative resection. Other treatment modalities such ass chemotherapy and radiotherapy have only little to no effect on survival. Unfortunately,, the majority of these patients can only be offered palliative treatment. Moree than 85% of patients with pancreaticobiliary malignancies will develop obstructivee jaundice in the course of their disease and often it is a presenting symp-tom.. Relieve of jaundice, besides pain management, is the mainstay of palliative therapy.. In the past, the golden standard treatment was surgical biliary diversion. Becausee of associated morbidity and mortality, surgical treatment has been chal-lengedd by endoscopic stent placement since the introduction of endoscopic retro-gradee cholangiopancreatography (ERCP) in 1980. Nowadays, endoscopic biliary drainagee has become the palliative treatment of choice to relieve biliary obstruction inn pancreaticobiliary malignancies.

EPIDEMIOLOGY Y

Off all pancreaticobiliary malignancies, pancreatic adenocarcinoma has the highest incidencee with around 30.000 new cases annually in the United States. It ranks fifth amongg the leading cause of cancer related deaths (1,2). Only 10% of patients are suit-ablee candidates for resection and the overall 5 year survival rate is less than 4%(3?4).

Thee incidence of gallbladder carcinomas is 1 per 100.000 person-years. The survival ratee is only slightly higher than that of pancreatic carcinoma (2). Most likely only the patientss who will survive are those in whom an early cancer was detected in a post-cholecystectomyy specimen.

Klatskinn tumors also have a poor prognosis with less than 10% of patients surviving 5 yearss after being diagnosed and with the vast majority of patients dying in the first year (5).. The number of rumors that are potentially resectable is low, ranging from 5-20%. Inn ampullary carcinoma biliary obstruction usually develops relatively early in the coursee of the disease. Therefore, tumors are usually small and radical resection is possiblee in the majority of cases with an overall 5 year survival rate up to 50% (6).

(5)

144 ENDOSCOPIC BILIARY DRAINAGE

PATHOGENESIS S

Althoughh a detailed discussion of the pathogenesis of pancreaticobiliary malignan-ciess is beyond the scope of this chapter it is interesting to note that several epidemi-ologicall studies have identified risk factors for the development of pancreaticobiliary malignancies. .

Tobaccoo smoking doubles the risk of pancreatic cancer (7,8), Patients with chronic pancreatitiss have an increased risk for developing pancreatic cancer which is esti-matedd at 4% per 20 years (9). The risk of developing pancreatic cancer in patients withh hereditary pancreatitis is as high as 50% with smoking being an important risk modifierr (10,11). Etiological factors for cholangiocarcinoma include primary scleros-ingg cholangitis and hepatolithiasis (12,13). Gallstone disease is the most important riskk factor for gallbladder cancer (14).

CLINICALL FEATURES

Thee most common presenting symptoms of pancreaticobiliary malignancies are painlesss jaundice with anorexia and weight loss, which are seen in the majority of patients.. If pain occurs, it is often located in the epigastric region or right upper quadrantt and may radiate to the back. Back pain usually indicates retroperitoneal infiltrationn with tumor and therefore irresectability. Other symptoms may include darkk urine, pale stools and pruritus. As much as 80% of patients with pancreatic cancerr have an impaired glucose tolerance or frank diabetes mellitus at the time of presentation.. Carcinoma of the body and tail of the pancreas presents with similar featuress although jaundice is usually absent or develop very late in the course of the disease. .

PATHOLOGY Y

Aboutt 90% of pancreaticobiliary malignancies are ductal adenocarcinoma. Most of thesee tumors arise from the pancreatic head. Other exocrine malignancies are muci-nouss cyst adenocarcinoma and acinar cell carcinomas. Endocrine tumors include gastrinomaa and insulinoma. Metastases of a primary tumor (mamma, lung, melanoma)) and a lymphoma should be considered because of important treatment implicationss (e.g. chemotherapy). Mesenchymal tumors are extremely rare.

Thee definitive diagnosis of malignancy depends on obtaining a tissue diagnosis. Althoughh a number of patients are palliated without definite confirmation of the tumor,, in cases of adjuvant therapies such as radiotherapy or chemotherapy, a cyto-logicall or histological biopsy proven malignancy is a prerequisite. In order to lower

(6)

thee number of costly and cumbersome ultrasound or CT guided punctures, it is advisablee to attempt obtaining a tissue diagnosis during the same ERCP procedure inn which a biliary endoprosthesis is inserted for palliation of jaundice. Various tech-niquess can be used to obtain tissue specimens during ERCP including cytologic brushings,, forceps biopsy, needle aspiration cytology and fluid collection from the bilee and/or pancreas.

Cytologicc brushings are relatively easy to obtain and widely used. Specificity approachess 100% but sensitivity is as low as 30-60% (15,16). The sensitivity in cho-langiocarcinomaa is higher compared to pancreatic carcinoma. Forceps biopsy or nee-dlee aspiration cytology requires endoscopic sphincleiotuiny and therefore carries an increasedd risk of complications, Ampullary tumours can be directly biopsied. FNA is superiorr to brush cytology and endobiliary forceps biopsy with a cancer detection ratee of 65% (17-19). Sampling of ductal fluid is a simple method, but its sensitivity is veryy low and therefore it is not used very often. Several studies have shown that sen-sitivityy can be increased by combining different techniques of tissue sampling (16). Endosonographicc fine needle aspiration biopsy results in an excellent sensitivity of 85-90%% and specificity of virtually 100% (20). Although these tests may be useful in makingg the diagnosis of carcinoma, a negative test cannot rule out malignant dis-ease.. Percutaneous fine needle aspiration biopsy is another accurate method for

con-firmationfirmation of malignancy with a sensitivity of 60-90% (21). However, needle track seedingg has been described and this technique should only be used for tissue

confir-mationn in the case of irresectable disease.

DIFFERENTIALL DIAGNOSIS

Thee most important discrimination is the differential diagnosis between benign and malignantt lesions. In the case of the former, surgery may not be indicated and may evenn cause harm to the patient while in the latter case it is the treatment of choice if aa lesion is resectable.

Ann enlarged pancreatic head may either be caused by pancreatitis or by carcinoma. Thee patients' history and clinical presentation contribute to making a diagnosis. Cysticc lesions of the pancreas may either be benign (pancreatic pseudocyst or serous cystadenoma),, pre-malignant (mucinous cystadenoma), or malignant (cystadenocar-cinoma).. Radiological imaging is used to characterize these lesions. Endoscopic ultrasoundd may further increase accuracy of the diagnosis in combination with fine needlee aspiration and fluid analysis.

Inn the case of a suspicious mid or proximal bile duct stricture a gallbladder carcino-maa should be included in the differential diagnosis. It is important to exclude benign causess of strictures such as Mirizzi's syndrome, primary and secondary sclerosing cholangitis,, and postoperative conditions. An algorithm of the diagnosis of pancre-aticobiliaryy cancer is presented in Figure 1.

(7)

i66 ENDOSCOPIC BILIARY DRAINAGE

TREATMENT T

Sincee the introduction of endoscopic biliary stenting in 1980, the palliative treat-mentt of pancreaticobiliary malignancies has changed considerably. Nowadays, endo-scopicc stenting to relieve jaundice is well established and is considered the preferred treatmentt (Figure 2). Compared to percutaneous and surgical drainage it is associat-edd with lower morbidity and mortality rates (22-24). The main problem of endoscop-icc biliary drainage is late stent occlusion which necessitates stent exchange. The tech-nicall success rate of endoscopic biliary drainage is between 70-90% and is higher forr distal tumors compared to more proximal malignancies involving the bifurca-tion.. The complication rate of therapeutic ERCP ranges between 5-10% (25,26).

I N D I C A T I O N S / C O N T R A I N D I C A T I O N S S

Thee indications for an ERCP with a drainage procedure by stent placement are jaun-dicee and/or fever and/or pruritus. Biliary stenting has also been shown to improve symptomss of anorexia and quality of life (27,28). It has been suggested that preoper-ativee biliary drainage may improve surgical outcome after pancreaticoduodenecto-my.. This however, has not been substantiated in clinical trials (29-31). If preoperative drainagee is indicated because of cholangitis, drainage should be performed using plasticc stents. In this setting metal expandable stents are too expensive and might causee technical problems for the surgeon during the resection.

Theree are no absolute contraindications. Coagulation disorders are a relative con-traindicationn and should be corrected before ERCP.

OVERVIEWW OF STENTS FOR BILIARY DRAINAGE

Plasticc Stents

Thee median patency of a conventional 10 Fr plastic stent ranges between 3-6 months. Thee incidence of stent occlusion varies between 20-50% (32-34). The initial event in stentt blockage is adherence of proteins and bacteria to the inner wall of the stent and subsequentt formation of a biofilm. Bacteria are introduced into the biliary system duringg transpapillary placement of the stent. Sludge then forms from the accumula-tionn of bacteria which produce ^-glucuronidase and form calcium bilirubinate and calciumm palmitate (35-37). Many efforts have been made to prolong stent patency somee of which are discussed in the following paragraphs.

StentStent Diameter

Thee first biliary stents which were placed were only 7 Fr or 8 Fr in diameter due to limitationss of the diameter of the working channel of the endoscope (2.8 mm). Whenn side viewing endoscopes with large diameter working channels (4.2.mm) weree introduced in 1980 it became possible to insert large bore plastic stents (38). Largerr stents (10 Fr) perform better than smaller diameter stents (7 Fr) (39). This

(8)

appearss to be due to the higher flow rate, as predicted by Poisseuille's law and less stasiss with larger diameter stents. Theoretically, bile flow rate is proportional to the internall diameter raised to the fourth power, so even a small increase in diameter resultss in a substantial increase in flow capacity (40). Somewhat in contrast to this hypothesis,, use of larger diameter plastic stents of 11.5 Fr or 12 Fr did not result in furtherr improvement of stent patency (41-43).

StentStent Design

Thee first biliary stents had a pigtail configuration at the proximal end to provide bet-terr anchorage. Straight stents were then developed because of their improved bile flowflow characteristics compared to pigtail stents (40,44,45) (Figure 3). Huibregtse developedd the Amsterdam type stent: a straight design with two side holes to facili-tatee biliary drainage and two side flaps to prevent dislocation which has become the standardd type stent design since 1980 (46).

Sludgee in plastic stents mainly accumulates around side holes (35,47). This seems duee to higher intraluminal flow turbulence and decreased flow rates (40). Soehen-draa postulated that elimination of side holes might improve patency rates and designedd the so called teflon Tannenbaum stent: a straight stent without side holes andd multiple proximal and distal side flaps to prevent dislocation (48,49). First uncon-trolledd results were encouraging with patency rates comparable to metal stents but randomizedd trials could not confirm these initial results (50-52). Omitting side holes inn standard designed polyethylene stent also did not show improvement in stent patencyy (53).

StentStent Material

Differentt materials have been used for stent construction: polyethylene, polyurethanee and teflon. In vitro studies have shown a direct relation between the coefficientt of friction and the amount of encrusted material. Teflon has the lowest frictionn coefficient and therefore the best potential for preventing stent clogging (35). Initially,, teflon Tannenbaum stents showed a favourable patency rate (48,49). A ran-domizedd study comparing Amsterdam type stents made from polyethylene and teflonn did not show a difference in stent patency (54). Other controlled clinical trials couldd also not confirm the superiority of teflon material in a Tannenbaum designed stentt (50-52).

Scanningg electron microscopy of out-of-package biliary stents has shown that the innerr surface smoothness of plastic stents is highly variable. This is possibly due to thee manufacturing process of plastic stents by extrusion. Only the polyurethane stentt was found to have an extremely smooth surface (55).

Twoo new polymers were introduced with an ultrasmooth surface: Vivathane and Hydromer.. Both materials have been shown to reduce bacterial adherence in vitro (56,57).. In addition, the Hydromer stent not only has a smooth texture but also a

(9)

i88 ENDOSCOPIC BILIARY DRAINAGE

coatingg that absorbs water and provides a hydrophilic sheath. Because bacteria ini-tiallyy attach by hydrophobic interactions, this coating could potentially lower bacter-iall adhesion and therefore increase stent patency. However, the encouraging results off in vitro studies could not be confirmed in prospective clinical trials (58,59).

StentStent Coating

Primingg the inner surface of a stent with a coating with some form of anti-adhesion propertyy may reduce biofilm formation and hence stent clogging. Antibiotics, antitrombotics,, silver and hydrophilic coating were all effective in reducing bacterial colonisationn in vitro (57,60,61). Clinical studies however, using antibiotic-coated or hydrophilic-coatedd stents did not show any benefit (59).

StentStent Position

Placingg the stent entirely within the common bile duct has the theoretical advantage off preserving the barrier function of the sphincter of Oddi. This prevents duodenal refluxx of food and bacteria into the stent and biliary tree. This so-called 'inside stent' approachh can only be performed when a free margin of 1 to 2 cm is maintained betweenn the distal end of the stricture and the papilla. With this in mind, about one thirdd of patients with malignant obstructive jaundice are potential candidates for suchh treatment (62). However, in a randomized trial no difference in stent perform-ancee was found. In fact, in the 'inside stent group', stent migration occurred signif-icantlyy more frequent (63).

Antibiotics Antibiotics

Bacteriaa can enter the bile duct through the portal circulation but more easily direct-lyy from the duodenum. When an endoprosthesis is placed, the barrier function of thee sphincter of Oddi is lost and bacteria enter the biliary tract freely. Sludge may thenn form because these bacteria produce ^-glucuronidase and form calcium biliru-binatee and calcium palmitate. In order to prolong stent patency, prophylactic treat-mentt with antibiotics seemed a logical step.

InIn vitro studies showed that antibiotic treatment reduced bacterial adherence to

plas-ticc stents (64). In a prospective randomized study with ciproxin no difference in stentt patency was found (65). In another study rotating antibiotics (cycles of 2 weeks ampicillin,, metronidazol and ciprofloxacin) were combined with ursodeoxycholic acidd and no difference in stent patency was shown (66). Only one small pilot study showedd a reduced rate of stent blockage with norfloxacin plus ursodeoxycholic acid (67).. Other studies combining antibiotics and bile salts (ofloxacin and ursodeoxy-cholic,, ciprofloxacin and Rowachol) did not show a longer duration of stent patency (68,69).. m summary, at this point in time there is no compelling evidence that stent patencyy benefits from antibiotic prophylaxis.

(10)

Aspirin Aspirin

Animall studies with prairie dogs have shown that aspirin inhibits mucous glycopro-teinn secretion by blocking prostaglandin synthesis (70). In a clinical study, the use of aspirinn reduced the content of all sludge components although no effect on stent patencyy was shown {71). No further studies using aspirin have been performed.

BileBile Salts

Bilee salts have a potent antibacterial effect and may also stimulate bile flow. Because bacteriaa attach by hydrophobic interactions, hydrophobic bile salts (deoxycholate, taurodeoxycholate)) inhibit initial bacterial attachment as was shown in experimental studiess (72). However, hydrophobic bile salts are not well tolerated. Unfortunately, hydrophilicc bile salts like ursodeoxycholate, which are better tolerated, have minimal effectt on bacterial adhesion. Except for one small pilot study, different prospective clinicall studies using ursodeoxycholic acid alone or combining ursodeoxycholic acid withh antibiotics could not show a difference in stent patency (66-69).

StentStent Exchange

Somee endoscopists prefer to schedule patients for elective stent exchange every 3-4 months.. The optimal time interval remains an unanswered question (73,74). Prophylacticc stent exchange requires a repeat (clinically not indicated) endoscopy and hass to be compared to the risks of watchful waiting and the risk of (severe) cholangi-tis.. As the majority of patients will not develop stent occlusion before dieing of the underlyingg disease, most endoscopists favour an expectant management strategy.

StentStent Cleaning

Somee endoscopists have proposed to leave an occluded stent in situ and clean the obstructedd lumen with a cytology brush or by flushing with saline instead of per-formingg stent replacement (75). This however, carries the risk of inducing biliary sepsiss by actively introducing the biofilm of the stent and bacteria from the duode-numm into the biliary tract. Therefore, stent cleaning is not recommended,

Metall Expandable Stents

Thee diameter of biliary stents was restricted by the size of the instrumentation chan-nell of the endoscope until the development of the expandable metal stents. All cur-rentlyy available expandable stents are made of metal. They differ in the way they are braided,, the size of the meshes, the type of metal and their rigidity. Presently, many differentt types of self expandable metal stents are available from different manufac-turerss (Figure 4). There are two expansion types: self-expandable stents with an intrinsicc expanding force and balloon expanding stents deployed by inflation of a bal-loon. .

(11)

Too date, the most experience has been gained with the self-expandable Wallstent. Thiss stent is delivered in a collapsed configuration on an 8 Fr delivery system. When deployedd it expands to a final diameter of 30 Fr (approximately 10 mm) and shortens aboutt 30% in length. The final diameter will be achieved after one week, when an equilibriumm is achieved between the dilating force of the stent and the resistance of thee bile duct wall/tumor.

Thesee large calibre metal expandable stents of 30 Fr provide a longer patency rate comparedd to plastic stents, but they do not prevent blockage indefinitely. Due to their design,, these metal expandable stents have much less surface upon which bacteria mayy adhere. The mechanism of stent blockage differs from plastic stents and includee rumor ingrowth through the interstices of the stent or overgrowth of the end off the stent and intima hyperplasia.

Severall studies have shown a median stent patency of about 6-9 months (33>34'74>76,77)) (Figure 5). Self-expandable stents are more difficult to insert, they cannott be removed after deployment and initial costs are high (about $1000). Variouss types of self expandable metal stents have been introduced, these are sum-marizedd below. Randomized trials with large groups of patients and long term fol-low-upp are only available for the Wallstent.

Wallstent® Wallstent®

Thee initial endoscopic placement experience was reported in 1989 (78). Wallstents (Bostonn Scientific, Boston, MN, USA) are made from stainless steel alloy filaments braidedd in a tubular mesh configuration. In the early phase of development, techni-call problems were mainly the restraining membrane failing to completely retract but thiss is rarely seen nowadays (79). The first randomized trial comparing plastic stents andd Wallstents was performed by Davids et al. Wallstent patency was superior to plasticc stents with median duration of 9 months (33). These results have been con-firmedfirmed in several other studies (34,74,80).

UltrajlexUltrajlex Diamond Stent®

Comparedd to the Wallstent, this stent is more flexible, employs a larger mesh design, hass less radial expandable force and is constructed from nitinol. The name of this stentt relates to the appearance of its mesh pattern. Non randomized studies compar-ingg the Ultraflex Diamond stent (Boston Scientific, Boston, MA, USA) to the Wallstentt suggested an equal or less durable patency (81,82), One prospective uncon-trolledd multicenter study showed a stent patency of the Diamond stent of almost 16 months,, a result which has never been documented with any type of biliary stent (83).. This exceptional outcome should be confirmed in a prospective randomized trial,, preferably in comparison to the Wallstent.

(12)

EndocoilEndocoil Instent®

Thiss 'removable' self-expanding stent is made of a coil spring of nickel-titanium alloy thatt expands radially (Instent, Eden Prairie, MN, USA). It can only be used for dis-tall stenosis because apposing coils prevent drainage of segmental ducts. Theoretically,, the problem of tumor ingrowth should be prevented by the apposing coils.. However, stent dysfunction by tumor ingrowth remains a problem and removall of the stent is not without risk (84,85). No long term follow-up or compara-tivee data are available.

GianturcoGianturco Z-Stenfö

Thee Gianturco Z-stent (Wilson Cook, Winston Salem, NC, USA) has wide gaps betweenn the zigzag bands, with greater potential for tumor ingrowth. Advantages of thiss stent are the fact that it does not shorten upon expansion and has no sharp edges att the ends. The Gianturco Z-stent is the second most used expandable stent and it iss mostly inserted via the percutaneous route. Patency rates are comparable to the Wallstentt (86-89).

StreckerStrecker Stent®

Thiss is a balloon expandable stent and there are only a few reported studies (90,91). Technicall failures occur in up to 27%. The main disadvantage of this stent is its diameterr of only 21 Fr and the absence of an intrinsic radial force (34,92). These unfavourablee features have prohibited the general use of the Strecker stent (Boston Scientific,, Boston, MA, USA). This stent is not commercially available any more.

CoveredCovered Self Expandable Stent

Tissuee ingrowth through the meshes of the stent is responsible for stent occlusion inn about 22-33% of patients (33,34). To overcome this problem, self expanding metal stentss have been covered with a polyurethane or silicone membrane, except for the proximall and distal 5 mm. Results of various stents (M.I. Tech Corporation, Seoul, Southh Korea; Wilson Cook, Winston Salem, NC, USA; Boston Scientific, Boston, MA,, USA) in various studies are contradictory (92-94). Major concerns are the risk off stent migration, cholecystitis and pancreatitis, although these complications have nott been reported with any significant frequency. Furthermore, these stent should nott be used intrahepatically because of occlusion of hepatic side branches by the cov-eringg membrane. The exact role of covered self expandable stent is still under inves-tigation. .

Plasticc or Metal Stent?

Selff expandable metal stents have a longer duration of patency compared to plastic stentss and ideally should be placed in all patients. The high initial costs have limit-edd their use in different health care settings worldwide.

(13)

222 ENDOSCOPIC BILIARY DRAINACE

Thereforee in a cost effective approach the choice between a plastic or metal stent dependss mainly on an estimate of patient survival. Tumor size seems a reliable pre-dictorr of survival. Prat et al. claim that in the case of a tumor greater than 30 mm a polyethylenee stent should be placed because of shorter expected survival (95). The presencee and number of liver metastases have also been shown to be independently relatedd to prognosis (96,97). Comparative studies did not show any benefit of self expandablee metal stents compared to polyethylene stents in the first three months afterr insertion (33,74). Therefore, it seems reasonable to insert a polyethylene stent inn patients with a life expectancy less than 3 months (Figure 6). If expected survival extendss 3 to 6 months an expandable metal stents should be considered (Figure 7). Differentt authors have shown this strategy as cost-effective (33.98,99). Patients who presentt with early clogging of a polyethylene stent (within one month after insertion) shouldd also receive a self expandable stent, irrespective of their life expectancy althoughh this has not been proven in prospective studies (100).

PROCEDUREE OF STENT PLACEMENT

Sedation n

ERCPP is usually performed under conscious sedation. This is achieved with intra-venouss administration of diazepam or midazolam. Fentanyl or pethidine may be usedd for control of pain. For reasons of safety, oxygen saturation is monitored by pulsee oximetry and heart rate and blood pressure are also registered. In some cen-terss more sophisticated sedating agents are used such as propofol. In most cases the usee of propofol is supervised by an anaesthesiologist although recently some reports suggestt that its use by a non-specialist anaesthesiologist such as endoscopists or nursingg staff is safe (101). Prior to sedation the pharynx is anesthetized with xylo-cainee spray.

Antibiotics s

Drainagee of the biliary tree is the mainstay of therapy for patients with cholangitis. Theree is controversy about routine use of pre-procedure antibiotic prophylaxis (102-104). .

Pre-operativee administration of antibiotics should definitively be started in a patient withh fever. Since failure to drain the entire biliary tree is the most important risk fac-torr associated with cholangitis after ERCP, antibiotic prophylaxis should also be administeredd in a highly selective group of patients in whom incomplete drainage is anticipatedd such as patients with a hilar malignancy or primary sclerosing cholangitis (105,106).. Prophylaxis can be given as a single, adequate dose shortly before the pro-cedure.. If contrast is injected in the biliary tract but obstruction can not be relieved, antibioticc therapy should be continued (or started) until drainage is established. Gramm negative bacteria are consistently the most common organisms in bile: Escherichiaa coli and to a lesser extent Klebsiella spp. and gram positive Enterococcus spp.. Therefore antibiotics in these cases should be bactericidal, aimed at

(14)

gram-neg-ativee bacteria, with good penetration in liver tissue and bile. Ciprofloxacin is current-lyy the first choice of antibiotic in our unit with the caveat that is does not cover Enterococci.. In case of fever despite ciprofloxacin, the addition of amoxicillin or switchh to piperacillin/ tazobactam is advisable.

T E C H N I Q U EE OF STENT PLACEMENT

Thee procedure starts with the introduction of a large channel (4.2 mm) side viewing therapeuticc endoscope into the second portion of the duodenum. Standard cannula-tionn of the papilla of Vater is performed by a ball-tip or cone-tip catheter; eventually caiinulaiionn can be attempted with a guidewire inserted in the ball-tip catheter. If thiss approach fails, a double lumen sphincterotome with a guide wire (cannulotome) shouldd be used. Use of this device may aid in achieving an optimal angle for bile duct cannulation.. When this is also not successful a precut sphincterotomy is performed too obtain biliary access (107). With the use of all these different techniques, deep can-nulationn is achieved in up to 95% of patients.

Oncee a diagnostic cathether is inserted into the bile duct, contrast is injected. It is essentiall to define the exact anatomy, location and nature of the stenosis. To avoid postt procedural cholangitis in patients with complex hilar strictures, contrast filling off segments that will not be drained should be avoided. The next step is to pass a guidewiree through the stricture in order to facilitate introduction of the catheter and enablee exchange for other instruments. When passage of a guidewire through the stricturee cannot be accomplished, the direction of the guidewire can be changed by manipulatingg its position with movements of the endoscope similar to those made forr standard cannulation. The assistant can help to cross the stricture by moving the guidewiree in and out the catheter. The endoscopist can manipulate the guidewire by movingg the guiding catheter.

AA variety of guidewires are available with different flexibility, diameter and shape of thee tip. On the one hand, rigid guidewires facilitate introduction of instruments {suchh as an IDUS probe) and small diameter stents. On the other hand, very slippery guidewiress with a hydropolymer coating which follows bends easily and are used to passs asymmetric strictures. Once the guidewire is passed through the stricture a catheterr can be advanced and more complete filling can be obtained.

AA sphincterotomy is not routinely necessary for introduction of one biliary stent. Previouslyy it was believed that a sphincterotomy was necessary to facilitate introduc-tionn of different devices and also to avoid occlusion of the pancreatic duct by the endoprosthesis.. In clinical practice however, this did not prove to be a problem. Only inn cases when more then one prosthesis is placed, a sphincterotomy is indicated.

Plasticc Stents

Oncee the stricture is passed with a guidewire a stent can usually be inserted. First, a catheterr is introduced over the guidewire through the stricture to ensure a more rigidd introductory system in order to facilitate stent placement. If appropriate the

(15)

guidewiree can be exchanged for a cytology brush to obtain tissue samples.

Thee endoprosthesis is positioned over the guiding catheter and inserted into the instrumentationn channel. With a pusher tube the stent is further advanced towards thee tip of the endoscope with the elevator bridge closed. When the prosthesis reach-ess the tip of the instrumentation channel, the elevator bridge is opened and the stent iss pushed out of the endoscope by the pusher tube under endoscopic and fluoroscop-icc control. During further advancement of the stent it is important to keep the endo-scopee tip close to the papilla. The stent should be advanced one step at a time by pushingg it a little bit further each time into the duodenum. The stent is raised by closingg the elevator bridge, and the tip of the endoscope is moved closer to the papil-laa with the up-down knob hereby introducing the stent. These steps are repeated untill the distal side flap has reached the papilla. Finally, the assistant pulls out the catheterr and guidewire while the endoscopist keeps the prosthesis in position with thee pusher tube.

InIn most distal and mid common bile duct strictures it is usually possible to insert a 100 Fr endoprosthesis without prior dilatation. In proximal strictures however, it is nott rare that strictures have to be dilated in order to allow stent placement. This can bee achieved with the use of progressive dilating catheters which are introduced over aa rigid guidewire. Balloon catheters can be used as well to accomplish this goal. If it iss still not possible to insert a 10 Fr stent, a smaller calibre prosthesis {7 Fr) should bee inserted which can be exchanged for a 10 Fr prosthesis a few days later. When bothh right and left liver lobes have to be drained it is usually more convenient to placee the endoprosthesis draining the left side first, followed by the right side. Thee required length of the endoprosthesis can be measured by using guidewire as a measuringg device. First, under fluoroscopic control the proximal tip of the guidewire iss positioned at the level where the proximal tip of the endoprosthesis is projected. Then,, the endoscopy nurse fixes the guidewire between finger and thumb just where itt exits the catheter. Subsequently, under fluoroscopic control the guidewire is with-drawnn from the catheter until the proximal tip reaches the duodenum. The distance betweenn finger and thumb and the distal margin of the catheter is the required lengthh of the endoprosthesis.

Plasticc stents are available various widths (ranging from 5 to 12 Fr) and lengths (rang-ingg from 5 to 19 cm).

Managementt of Plastic Stent Occlusion

AA clogged plastic stent can be removed by means of a snare or dormia basket. It is importantt to keep the position of the endoscope in line with the common bile duct. Whenn a snare is used, the stent is caught in the snare and removed through the instrumentationn channel of the endoscope. When a dormia is used, the stent is pulledd close to the endoscope and both the endoscope and stent are withdrawn. Whenn massive tumor invasion is present in the duodenum and difficult stent

(16)

exchangee is anticipated because of a non-optimal scope position, it can be helpful to leavee the occluded stent in place and use it as a guidance for common bile duct can-nulationn and introduction of a second stent.

Soehendraa described a technique which enables removing a clogged stent while maintainingg the original pathway to the bile duct {108). A ball tip catheter is posi-tionedd at the distal end of the stent and the stent is cannulated with the guidewire. AA Soehendra retriever is introduced over the guidewire and the tip is screwed into thee distal end of the stent. Then the retriever with along the stent is pulled out while leavingg the guidewire in place.

Selff Expandable Stents

Forr introduction of a self-expandable stent a stiff guidewire is positioned through the stricturee by standard techniques. The insertion device with the constrained stent is thenn inserted through the instrumentation channel over the guidewire. When the insertionn device is in position with the help of radio-opaque markers, the prosthesis cann be released by removing the outer catheter while keeping the inner catheter in place.. Deployment follows gradually as the outer catheter is withdrawn and can be followedd fluoroscopically. If deployment is not according to plan and repositioning is required,, the expandable stent may be constrained again by pushing the outer catheterr inwards whenever the point of no return has not yet been passed. This point mayy vary with stent type but may extent to 83% of total stent deployment and is indi-catedd by a marker. Deployment reduces the length of the self expandable metal Wallstentt by about 30%. Therefore, it is important to constantly correct the position off the expandable stent under fluoroscopic control which usually means that one has too pull the insertion device outwards while deploying the stent.

Whenn the expandable metal stent bridges the papilla in the case of a distal stenosis, thee endoscopic image is used to keep a fixed distance of about 1 cm between the pap-illaryy orifice and the distal margin of the stent.

Stentt diameter expands to 8-10 mm and the available deployed lengths are 40, 60, 800 and 100 mm,

Inn case of a complex hilar stricture when both liver lobes are drained by two or more selff expandable endoprostheses the procedure is as follows (109). The procedure beginss with the introduction of two stiff guidewires, one in each liver lobe. If appro-priate,, dilation of a stricture is performed over one of the guidewires. Then, a expandablee metal stent is inserted over the guidewire into the left system and deployed.. Finally, an expandable metal stent is inserted into the right system along-sidee the first stent and deployed under fluoroscopic control {Figure 8).

Althoughh technically difficult, it is also possible to insert a second self expandable stentt through the meshes of a former placed self expandable stent (no). In that case aa guidewire is introduced and the mesh is dilated using a balloon catheter before passingg the second constrained stent and deploying it.

(17)

Managementt of Self Expandable Stent Metal Occlusion

Selff expandable stents can be removed within the first 2 to 3 days after deployment byy grasping it with a forceps or a snare. After this time the stent becomes embedded inn the tumour tissue and cannot be extracted.

Stentt obstruction is mainly due to tumor ingrowth through the interstices of the stentt or overgrowth of the ends of the stent. Management of stent occlusion consists off placement of a polyethylene stent or a second self expandable stent through the occludedd self expandable stent. Another strategy is mechanical cleaning by using a balloonn and flushing, but this is only effective in case of sludge formation.

INTRAHEPATICC BILIARY OBSTRUCTION

Stricturess at the level of the hepatic confluence account for about 20% of malignant bilee duct obstruction and mainly consist of primary cholangiocarcinoma, gallbladder neoplasmss and metastatic spread to hilar nodes. Cholangiocarcinoma arising at the hilarr level is also referred to as a Klatskin tumor and is classified according to the degreee of involvement of the intrahepatic bile ducts (in) (Figure 9). Stenting the proximall biliary tree is more challenging and associated with lower success rates thann stenting distal common bile duct stenosis. Drainage can be achieved either endoscopicallyy (retrograde) or percuntanously (antegrade).

Proceduree induced cholangitis caused by contrast injection in undrained biliary branchess is the main complication and occurs in up to 30% (112-114). The current managementt strategy (depending on local services available) is first to attempt endo-scopicc drainage; when this strategy is not successful percutaneous drainage offers additionall opportunities (115,116). When internal drainage fails, an external drain cann be left in situ minimizing the risk of cholangitis.

Uni-- or Bilateral Drainage?

Theree is controversy whether to drain one or both liver lobes in Bismuth type II, III andd IV strictures. In Bismuth type I, one stent always suffices because the left and rightt ducts communicate and drainage will be complete. Theoretically, at least 25% off the liver volume must be drained to achieve biochemical improvement and relief off symptoms (117). Concerns about unilateral drainage include the inability to relief jaundice,, as well as the potential for bacterial contamination in the undrained lobe. Indeed,, the worst treatment results seem to be obtained in patients with cholangio-graphicc opification of both lobes but drainage of only one (118). Recently, a prospec-tivee randomized trial compared unilateral to bilateral hepatic duct drainage (119). Unilaterall drainage was associated with a significantly higher rate of successful endoscopicc stent insertion. Bilateral stent placement was associated with a signifi-cantlyy higher rate of complications because of the higher rate of early cholangitis. In per-protocoll analysis the rate of successful drainage, complications, and mortality didd not differ between the two groups. MRCP-guided endoscopic stent placement in

(18)

Bismuthh III and IV malignancies was associated with a low morbidity and mortality inn an uncontrolled study (120). The intention was to place an unilateral stent in one off both lobes guided by the MRCP picture avoiding entry and contrast injection in thee contralateral lobe. In those patients in whom, by accident, guidewire entry (50%) orr contrast injection (20%) occurred in the contralateral liver lobe, stents were placed bilaterally.. This treatment strategy resulted in a very low cholangitis rate of only 6%. AA recent study evaluated selective unilateral MRCP or CT-targeted drainage and no episodess of cholangitis were observed (121). The message seems to emerge that uni-laterall drainage is appropriate when unilateral cannulation and opacification has beenn achieved. If the contralateral lobe is (unintentionally) opacified or probed, it shouldd also be drained to avoid cholangitis.

Plasticc or Self Expandable Metal Stent?

Byy design, expandable stents may be more suitable than plastic stents for draining hilarr tumors. The stent lumen is much wider and more importantly intrahepatic side branchess can drain through the metal meshes. Indeed, self expandable stents which weree inserted via the percutaneous route, showed a higher treatment efficacy com-paredd to plastic stents (115,122). There are no randomized studies available compar-ingg endoscopic and percutaneous insertions of self expandable metal stents in hilar strictures.. Additional proof of the superiority of self expandable stents over plastic stentss is suggested from a retrospective series of patients with non resectable hilar cholangiocarcinomaa by whom during stent treatment, plastic stents were replaced by metall expandable stents(i23). Successful palliation without the need for further bil-iaryy reintervention was achieved in the majority of patients (69%). A potential draw-backk of the placement of a metal stent is that, in the case of treatment failure, intro-ductionn of additional stents may become difficult. However, a technique of introduc-ingg a second stent through the wire mesh of the first stent has been described (no).

DUODENALL S T E N 0 5 I S

Duodenall stenosis due to pancreaticobiliary malignancies occurs in 10-20% of patientss (124). Presenting symptoms include nausea and vomiting due to gastric out-lett obstruction. Usually this is a late event and occurs in patients in poor general con-ditionn who have already received a biliary endoprosthesis (125). Surgical bypass has aa significant procedure related mortality of up to 10% as well as related morbidity andd prolonged hospital stay (24,126,127). Endoscopic stenting for duodenal obstruc-tionn together with bile duct stenting may be an effective alternative.

Placementt of duodenal stents has a high technical success rate without major proce-duree related complications (128-130). Stenting is carried out under simultaneous endoscopicc and fluoroscopic control. Preliminary dilatation of the duodenal stenosis cann be performed by balloon dilatation if necessary. Patients are usually able to tol-eratee a liquid diet immediately after stent placement. Full stent deployment may take aa few days during which time soft foods are allowed.

(19)

Onee study reported about simultaneous decompression of biliary and duodenal obstructionn with similar success rates as to duodenall stenting alone (131). Because of thee difficulty to endoscopically access the biliary tree through the mesh wall of a duo-denall stent, preferably an expandable metal biliary stent should be placed before the duodenall stent is introduced (Figure 10). If endoscopic biliary stenting fails the remainingg treatment options are percutaneous stenting, combined percutaneous andd endoscopic management or surgical bypass.

POSTPROCEDURALL CARE

Generall measures after conscious sedation include observation in a day care unit for severall hours with monitoring of blood pressure and oxygen saturation.

Whenn a patient develops fever post-ERCP, efforts should be made to obtain speci-menss for culture and administration of antibiotics should be started. If fever does nott subside, the accuracy of biliary drainage should be reassessed and migration and earlyy stent occlusion should be excluded. In the case of a complex malignant hilar stricturee it is important to check for undrained dilated intrahepatic segments and rulee out abscesses by transabdominal ultrasound or CT. Depending on the findings, ERCPP should be reattempted or percutaneous drainage achieved.

COMPLICATIONS S

Earlyy Complications

Earlyy complications are defined as those which occur less than 1 week after the con-clusionn of the procedure. The complications range between 5 and 10% for therapeu-ticc ERCP with a mortality rate of up to 1% (25,26,132). Cotton introduced a compli-cationn grading system in which complications are graded in mild, moderate and severee and these guidelines are still widely used (25).

Thee most frequent early complication is cholangitis, probably due to introduction of bacteriaa into the biliary tract during the procedure. This is reported in approximate-lyy 10-15% of patients in most series. It occurs more often following endoscopic pro-ceduress for complex hilar strictures when incomplete drainage is achieved. The samee holds true for patients with primary sclerosing cholangitis. In these 'high risk' proceduress antibiotics should be administered prophylactically and continued for a feww days after the procedure.

Postt ERCP-pancreatitis occurs in about 5-7% of patients. It is defined as new onset orr increased abdominal pain, lasting at least 24 hours after ERCP, with associated elevationn in serum amylase or lipase of at least 3-5 fold above normal (25,26,133). Mostt cases are mild, self-limited, and only require intravenous fluids and gut rest. Seriouss cases may evolve into (infected) necrotizing pancreatitis with multi organ failure. .

Thee rate of post sphincterotomy bleeding is about 0.2-5% withh an associated mortal-ityy rate less than 1% (134). Bleeding is usually obvious immediately after

(20)

sphinctero-tomyy but can be delayed for hours or even several days. Most episodes of delayed bleedingg are managed successfully by conservative measures and blood transfusions iff the haemoglobin level drops significantly. Post sphincterotomy bleeding usually occurss at the apex of the sphincterotomy site and can be managed endoscopically withh injection of adrenaline.

Retroperitoneall perforation occurs in less than 1% in most series. It may be caused byy standard sphincterotomy, precut sphincterotomy or by guide wire manipulation. Mostt cases are diagnosed or suspected during ERCP. These perforations mostly heal withh conservative measures and usually do not result in clinical symptoms (135). Conservativee treatment measures consist of nil by mouth, antibiotic treatment and nasogastricc suction. It is estimated that about 20-30% of these patients will require surgery. .

Inn cases of peritoneal perforation caused by the duodenoscope, prompt exploratory laparotomy,, with repair or oversewing of the defect in the duodenal wall, is manda-toryy (136).

Latee Complications

Thee primary late complication of stent placement is occlusion of the endoprosthesis, occurringg in up to 50% of cases (33,34). Clinically these patients present with a flu-likee syndrome with cholestasis, frank cholangitis or jaundice. Treatment consists of exchangee of the occluded stent or, in case of an occluded self expanding metal stent, insertionn of a polyethylene stent or second self expandable stent (see management off plastic stent occlusion and management of self expandable metal stent occlusion) throughh the obstructed expandable stent. Plastic stent migration, either proximally orr distally, may occur in up to 10% of cases (137).

FUTUREE TRENDS

P H O T O D Y N A M I CC T H E R A P Y

Photodynamicc therapy (PDT) involves the administration of a photos ensitizer which iss activated with a laser light and causes necrosis of the exposed tissue. Preliminary resultss suggest prolonged survival and stent patency for PDT in cholangiocarcinoma att the hilum (138-140). Controlled trials are in progress. PDT cannot be combined withh uncovered expandable stents, since PDT generates necrotic tumor tissue, which sloughss into and occludes the lumen (141). However, replacement of plastic stent by aa self-expandable stent one month after PDT could be promising (138).

DRUGG COATED BILIARY STENTS

Futuree prospects include development of chemotherapy impregnated expandable stents.. Covering biliary stents with chemotherapeutic agents, delivering

(21)

chemother-300 ENDOSCOPIC BILIARY DRAINAGE

apyy directly to the tumor tissue, at least in theory, should give protection against tumorr ingrowth, overgrowth, or both. For optimal therapeutic effects these drugs shouldd be released over a longer period of time with good penetration in tissue and withoutt systemic toxicity.

Carboplatinn and placlitaxel have shown to inhibit cell proliferation in in vitro studies (142,143).. Carboplatin coated plastic stents have been used with promising prelimi-naryy results in a limited number of patients {143). Further controlled trials are war-ranted. .

EUSS G U I D E D PLEXUS NEUROLYSIS

Celiacc plexus neurolysis can improve pain control in patients with pancreatic cancer. Thee injected agent usually includes a local anaesthetic (bupivacaine or lidocaine) and aa neurolytic (phenol or alcohol). A meta-analysis showed long-lasting benefit for 70-90%% of patients and adverse effects were common but generally transient and mild (144).. For best results it is recommended to perform celiac plexus neurolysis not too latee in the course of the disease when pain becomes unbearable. This may be explainedd by central effects of chronic pain leading to hypersensitization and unre-sponsivenesss to anti pain treatments. EUS may reduce neurological complications (becausee of the anterior approach) compared to the percutaneous technique althoughh no comparative studies have been performed (145). Side effects of celiac plexuss neurolysis are usually mild and include postprocedural pain and transient diarrhoea. .

(22)

Obstructivee jaundice and suspicion pancraticobiliaryy malignancy ERCPP (brush) and d intraductal l ultrasound d

Figuree 1. Algorithm of diagnosis of pancreaticobiliary cancer.

Figuree 2. Stenosis of both common bile duct and pancreatic duct, also called a doublee duct sign (A), caused by a pancreatic adenocarcinoma and a 10 Fr 9 cm plasticc endoprosthesis inserted through a distal bile duct sticture (B).

(23)

Figuree 3. Different types of plastic endoprosthesis (fromm top downwards): a double pigtail stent, an Amsterdamm type stent (one side hole and one side flapp at each end) and a Tannenbaum type stent (with-outt side holes and multiple side flaps at each end).

Figuree 4. Different types of self expanding metal stentss (from top downwards): Wallstent (Boston Scientific,, Boston, MA, USA), Gianturco Z-stent (Wilsonn Cook, Winston Salem, NC. USA), Hanaro stentt and covered Hanaro stent (M.I. Tech Corporation,, Seoul, South Korea).

(24)

Davids s (33) ) Carr--Locke(77) ) Knyrim m (34) ) Numberr of patients s PE E 49 9 78 8 31 1 SEMS S 56 6 86 6 31 1 Drainagee (%) PE E 95 5 95 5 100 0 SEMS S 96 6 98 8 100 0 Occlusionn rate (%) ) PE E 54 4 13 3 43 3 SEMS S 33 3 13 3 22 2 Mediann stent patencyy (days) PE PE 126 6 62 2 140* * SEMS S 273 3 111 1 189* * ** Mean

PE:: polyethylene stent, SEMS: selfexpandable metal stent

Figuree 5. Results of trials comparing selfexpandable stents with plastic stents.

Figuree 6. Mid common bile duct stricture caused by gallbladder carcinoma (A) with

(25)

B B

Figuree 7. Distal common bile duct stricture caused by a pancreatic adenocarcinoma (A)) with a self expandable metal stent inserted (B).

(26)

Figuree 8.

A.. Klatskin type II tumor (irresectable because of vascular involvement) B.. Guidewires inserted to both the left and right biliary sytem.

C.. A self expandable metal stent has been inserted into the left system and deployed. D.. Two sided self expandable metal stent drainage.

(27)

II II UIA IIIB IV

Figuree 9. Bismuth classification:

II Stricture involving the common hepatic duct III Stricture involving both right and left hepatic duct

IIII A Stricture extending proximally to the right secondary intrahepatic ducts IIIBB Stricture extending proximally to the left secondary intrahepatic ducts IVV Stricture involving secondary intrahepatic ducts bilaterally

Figuree io. Pancreatic adenocarcinoma growing into the duodenum with a self expandablee metal stent (not yet fully deployed) in the biliary tract and a self expand-ablee metal stent in the duodenum.

(28)

REFERENCES S

i.. Ries LA, Wingo PA, Miller DS, Howe HL, Weir HK, Rosenberg HM et al. The annual reportt to the nation on the status of cancer, 1973-1997, with a special section on colorec-tall cancer. Cancer 2000;

88(io):2398-2424-2.. Michaud DS. The epidemiology of pancreatic, gallbladder, and other biliary tract cancers. Gastrointestt Endosc 2002; 56(6 Suppl):Sl95-S20o.

3.. Warshaw AL, Fernandez-del Castillo C. Pancreatic carcinoma. N Engl } Med 1992; 326{7}:455-465. .

4.. Rosewicz S, Wiedenmann B. Pancreatic carcinoma. Lancet 1997; 349(905o):485-489-5.. Bismuth H, Castaing D, Traynor O. Resection or palliation: priority of surgery in the

treatmentt ot hilar cancer. World J Surg 1988; i2(i):39-47.

6.. Monson JR, Donohue JH, McEntee GP, Mcllrath DC, van Heerden JA, Shorter RG et al. Radicall resection for carcinoma of the ampulla of Vater. Arch Surg 1991; ^^(})'-353-^57-7.. Bueno de Mesquita HB, Maisonneuve P, Moerman CL Runia S, Boyle P. Life-time his-toryy of smoking and exocrine carcinoma of the pancreas: a population-based case-control studyy in The Netherlands. Int J Cancer 1991; 49(6):8i6-822.

8.. Gold EB, Goldin SB. Epidemiology of and risk factors for pancreatic cancer. Surg Oncol Clinn N Am 1998; 7(i):67-9i.

9.. Lowenfels AB, Maisonneuve P, Cavallini G, Ammann RW, Lankisch PG, Andersen JR et al.. Pancreatitis and the risk of pancreatic cancer. International Pancreatitis Study Group. NN Engl J Med 1993; 328(2o):i433-i437.

10.. Lowenfels AB, Maisonneuve P, Whitcomb DC, Lerch MM, DiMagno EP. Cigarette smok-ingg as a risk factor for pancreatic cancer in patients with hereditary pancreatitis. JAMA 2001;; 286(2):r69-i7o.

11.. Ghadirian P, Lynch HT, Krewski D. Epidemiology of pancreatic cancer: an overview. Cancerr Detect Prev 2003;

27(2):87-93-12.. Bergquist A, Ekbom A, Olsson R, Kornfeldt D, Loof L, Danielsson A et al. Hepatic and extrahepaticc malignancies in primary sclerosing cholangitis. J Hepatol 2002; 3Ö(3):32i-327. .

13.. Okuda K, Nakanuma Y, Miyazaki M. Cholangiocarcinoma: recent progress. Part 1: epi-demiologyy and etiology. J Gastroenterol Hepatol 2002; I7(IO):IO49-IO55.

14.. Lowenfels AB, Maisonneuve P, Boyle P, Zatonski WA. Epidemiology of gallbladder can-cer.. Hepatogastroenterology 1999; 46(27):i529-i532.

15.. Mansfield JC, Griffin SM, Wadehra V, Matthewson K. A prospective evaluation of cytol-ogyy from biliary strictures. Gut 1997; 4o(5):67i-677.

16.. Jailwala J, Fogel EL, Sherman S, Gottlieb K, Flueckiger J, Bucksot LG et al. Triple-tissue samplingg at ERCP in malignant biliary obstruction. Gastrointest Endosc 2000; 51(4 Pt i):383-390. .

17.. Howell DA, Beveridge RP, Bosco J, Jones M. Endoscopic needle aspiration biopsy at ERCPP in the diagnosis of biliary strictures. Gastrointest Endosc 1992; 38(5):53i-535-18.. Pugliese V, Conio M, Nicolo G, Saccomanno S, Gatteschi B. Endoscopic retrograde

for-cepss biopsy and brush cytology of biliary strictures: a prospective study. Gastrointest Endoscc 1995; 42(6):52o-526.

19.. Ponchon T, Gagnon P, Berger F, Labadie M, Liaras A, Chavaillon A et al. Value of endo-biliaryy brush cytology and biopsies for the diagnosis of malignant bile duct stenosis: resultss of a prospective study. Gastrointest Endosc 1995; 42(6):5&5-572.

(29)

20.. Williams DB, Sahai AV, Aabakken L, Penman ID, van Velse A, Webb J et al. Endoscopic ultrasoundd guided fine needle aspiration biopsy: a large single centre experience. Gut 1999;; 44(5)720-726.

21.. Linder S, Blasjo M, Sundelin P, von Rosen A. Aspects of percutaneous fine-needle aspi-rationn biopsy in the diagnosis of pancreatic carcinoma. Am } Surg 1997; i74(3):303-3o6. 22.. Speer AG, Cotton PB, Russell RC, Mason RR, Hatfield AR, Leung JW et al. Randomised

triall of endoscopic versus percutaneous stent insertion in malignant obstructive jaun-dice.. Lancet 1987; 2{855o):57-62.

23.. Andersen JR, Sorensen SM, Kruse A, Rokkjaer M, Matzen P. Randomised trial of endo-scopicc endoprosthesis versus operative bypass in malignant obstructive jaundice. Gut 1989;; 3o(8):ii32-ii35.

24.. Smith AC, Dowsett JF, Russell RC, Hatfield AR, Cotton PB. Randomised trial of endo-scopicc stenting versus surgical bypass in malignant low bileduct obstruction. Lancet 1994;; 344(8938):i655-i66o.

25.. Cotton PB, Lehman G, Vennes J, Geenen JE, Russell RC, Meyers WC et al. Endoscopic sphincterotomyy complications and their management: an attempt at consensus. Gastrointestt Endosc 1991;

37(3):383-393-26.. Freeman ML, Nelson DB, Sherman S, Haber GB, Herman ME, Dorsher PJ et al. Complicationss of endoscopic biliary sphincterotomy. N Engl } Med 1996; 335(131:909-918. .

27.. Ballinger AB, McHugh M, Catnach SM, Alstead EM, Clark ML. Symptom relief and qualityy of life after stenting for malignant bile duct obstruction. Gut 1994; 35(4):4Ö7-470. 288 Abraham NS, Barkun JS, Barkun AN. Palliation of malignant biliary obstruction: a prospectivee trial examining impact on quality of life. Gastrointest Endosc 2002;

56(6):835-84i. .

299 Lai EC, Mok FP, Fan ST, Lo CM, Chu KM, Üu CL et al. Preoperative endoscopic drainage forr malignant obstructive jaundice. Br J Surg 1994; 8i(8):ii95-ii98.

30.. Sewnath ME, Birjmohun RS, Rauws EA, Huibregtse K, Obertop H, Gouma DJ. The effectt of preoperative biliary drainage on postoperative complications after pancreatico-duodenectomy.. } Am Coll Surg 2001; 192(6)726-734.

31.. Martignoni ME, Wagner M, Krahenbuhl L, Redaelli CA, Friess H, Buchler MW. Effect of preoperativee biliary drainage on surgical outcome after pancreatoduodenectomy. Am J

Surgg 2 0 0 1 ; I 8 I { I ) : 5 2 - 5 9 .

32.. Shepherd HA, Royle G, Ross AP, Diba A, Arthur M, Colin-Jones D. Endoscopic biliary endoprosthesiss in the palliation of malignant obstruction of the distal common bile duct: aa randomized trial. Br J Surg 1988; 75(i2):n66-ii68.

33.. Davids PH, Groen AK, Rauws EA, Tytgat GN, Huibregtse K. Randomised trial of self-expandingg metal stents versus polyethylene stents for distal malignant biliary obstruc-tion.. Lancet 1992; 34o(8834-8835):i488-i492.

34.. Knyrim K, Wagner HJ, Pausch J, Vakil N. A prospective, randomized, controlled trial of metall stents for malignant obstruction of the common bile duct. Endoscopy 1993;

25(3):207-2i2. .

35.. Coene PP, Groen AK, Cheng J, Out MM, Tytgat GN, Huibregtse K. Clogging of biliary endoprostheses:: a new perspective. Gut 1990; 3i(8):9i3-9i7.

36.. Leung JW, Ling TK, Kung JL, Vallance-Owen J. The role of bacteria in the blockage of bil-iaryy stents. Gastrointest Endosc 1988; 34(i):i9-22.

(30)

37.. Speer AG, Cotton PB, Rode J, Seddon AM, Neal CR, Holton J et al. Biliary stent block-agee with bacterial biofilm. A light and electron microscopy study. Ann Intern Med 1988; io8(4):546-553. .

38.. Huibregtse K. Endoscopic biliary and pancreatic drainage. Stuttgart: Thieme, 1988. 39.. Speer AG, Cotton PB, MacRae KD. Endoscopic management of malignant biliary

obstruction:: stents of 10 French gauge are preferable to stents of 8 French gauge. Gastrointestt Endosc 1988; 34(5):4i2-4i7.

40.. Rey JF, Maupetit P, Greff M. Experimental study of biliary endoprosthesis efficiency. Endoscopyy 1985; I7(4):i45-I48.

41.. Pereira-Lima JC, Jakobs R, Maier M, Benz C, Kohier B, Riemann JF. Endoscopic biliary stentingg far the palliation cf pancreatic cancer results, survival pieuicLive factors, and comparisonn of 10-French with 11.5-French gauge stents. Am } Gastroenterol 1996; 9i(io):2i79-2i84. .

42.. Kadakia SC, Starnes E. Comparison of 10 French gauge stent with 11.5 French gauge stentt in patients with biliary tract diseases. Gastrointest Endosc 1992; 38(4):454-459. 43.. Siegel JH, Pullano W, Kodsi B, Cooperman A, Ramsey W. Optimal palliation of

malig-nantt bile duct obstruction: experience with endoscopic 12 French prostheses. Endoscopy 1988;; 2o(4):i37-i4i.

44.. Scheeres D, O'Brien W, Ponsky L, Ponsky J. Endoscopic stent configuration and bile flow ratess in a variable diameter bile duct model. Surg Endosc 1990; 4(2):9i-93.

45.. Leung JW, Del Favero G, Cotton PB. Endoscopic biliary prostheses: a comparison of materials.. Gastrointest Endosc 1985; 3i(2):93>95.

46.. Huibregtse K, Tytgat GN. Palliative treatment of obstructive jaundice by transpapillary introductionn of large bore bile duct endoprosthesis. Gut 1982; 2.^):yji-y]^.

47.. Dowidar N, Kolmos HJ, Matzen P. Experimental clogging of biliary endoprostheses. Rolee of bacteria, endoprosthesis material, and design. Scand J Gastroenterol 1992; 27(1)77-80. .

48.. Seitz U, Vadeyar H, Soehendra N. Prolonged patency with a new-design Teflon biliary prosthesis.. Endoscopy 1994; 26(5):478-482.

49.. Binmoeller KF, Seitz U, Seifert H, Thonke F, Sikka S, Soehendra N. The Tannenbaum stent:: a new plastic biliary stent without side holes. Am J Gastroenterol 1995; 9o(io):i764-i768. .

50.. Catalano MF, Geenen JE, Lehman GA, Siegel JH, Jacob L, McKinley MJ et al. "Tannenbaum"" Teflon stents versus traditional polyethylene stents for treatment of malignantt biliary stricture. Gastrointest Endosc 2002; 55(3):354"358.

51.. England RE, Martin DF, Morris J, Sheridan MB, Frost R, Freeman A et al. A prospective randomisedd multicentre trial comparing 10 Fr Teflon Tannenbaum stents with 10 Fr polyethylenee Cotton-Leung stents in patients with maJignant common duct strictures. Gutt 2000; 46(3)1395-400.

52.. Terruzzi V, Comin U, De Grazia F, Tori GL, Zambelli A, Beretta S et al. Prospective ran-domizedd trial comparing Tannenbaum Teflon and standard polyethylene stents in distal malignantt biliary stenosis. Gastrointest Endosc 2000; 5i(i):23-27.

53.. Sung JJ, Chung SC, Tsui CP, Co AL, Li AK. Omitting side-holes in biliary stents does not improvee drainage of the obstructed biliary system: a prospective randomized trial. Gastrointestt Endosc 1994; 4o(3):32i-325.

(31)

prospectivee randomized trial of Teflon versus polyethylene stents for distal malignant bil-iaryy obstruction. Endoscopy 1998; 3o(8):68i-686.

55.. van Berkel AM, van Marie }, van Veen H, Groen AK, Huibregtse K. A scanning electron microscopicc study of biliary stent materials. Gastrointest Endosc 2000; 5i(i):i9-22. 56.. McAllister EW, Carey LC, Brady PG, Heller R, Kovacs SG. The role of polymeric surface

smoothnesss of biliary stents in bacterial adherence, biofilm deposition, and stent occlu-sion.. Gastrointest Endosc 1993; 39(3):422-425.

57.. Jansen B, Goodman LP, Ruiten D. Bacterial adherence to hydrophilic polymer-coated polyurethanee stents. Gastrointest Endosc 1993; 39(5):67<>673.

58.. Costamagna G, Mutignani M, Rotondano G, Cipolletta L, Ghezzo L, Foco A et al. Hydrophilicc hydromer-coated polyurethane stents versus uncoated stents in malignant biliaryy obstruction: a randomized trial. Gastrointest Endosc 2000; 5i(i):8-n.

59.. van Berkel AM, Bruno MJ, Bergman JJ, Van Deventer SJ, Tytgat GN, Huibregtse K. A prospectivee randomized study of hydrophilic polymer-coated polyurethane versus poly-ethylenee stents in distal malignant biliary obstruction. Endoscopy 2003; 35(6):478-482. 60.. Leung JW, Liu Y, Cheung S, Chan RC, Inciardi JF, Cheng AF. Effect of antibiotic-loaded hydrophilicc stent in the prevention of bacterial adherence: a study of the charge, dis-charge,, and recharge concept using ciprofloxacin. Gastrointest Endosc 2001; 53(4):43i-

437--61.. Leung JW, Lau GT, Sung JJ, Costerton JW. Decreased bacterial adherence to silver-coat-edd stent material: an in vitro study. Gastrointest Endosc 1992; 38(3):338-34o.

62.. Liu Q, Khay G, Cotton PB. Feasibility of stent placement above the sphincter of Oddi ("inside-stent")) for patients with malignant biliary obstruction. Endoscopy 1998; 3o(8):687-6go. .

63.. Pedersen FM, Lassen AT, Schaffalitzky de Muckadell OB. Randomized trial of stent placedd above and across the sphincter of Oddi in malignant bile duct obstruction. Gastrointestt Endosc 1998;

48(6):574-579-64.. Leung JW, Liu YL, Desta TD, Libby ED, Inciardi JF, Lam K. In vitro evaluation of antibi-oticc prophylaxis in the prevention of biliary stent blockage. Gastrointest Endosc 2000;

51(3)^96-303--65.. Sung JJ, Sollano JD, Lai CW, Ismael A, Yung MY, Tumala I et al. Long-term ciprofloxacin treatmentt for the prevention of biliary stent blockage: a prospective randomized study. Amm J Gastroenterol 1999; 94(n):3i97-320i.

66.. Ghosh S, Palmer KR. Prevention of biliary stent occlusion using cyclical antibiotics and ursodeoxycholicc acid. Gut 1994; 35{i2):i757-i759.

67.. Barrioz T, Ingrand P, Besson I, De L, V, Silvain C, Beauchant M. Randomised trial of pre-ventionn of biliary stent occlusion by ursodeoxycholic acid plus norfloxacin. Lancet 1994; 344(8922):58i-582. .

68.. Halm U, Schiefke, Fleig WE, Mossner J, Keim V. Ofloxacin and ursodeoxycholic acid ver-suss ursodeoxycholic acid alone to prevent occlusion of biliary stents: a prospective, ran-domizedd trial. Endoscopy 2001; 33(6):49i-494.

69.. Luman W, Ghosh S, Palmer KR. A combination of ciprofloxacin and Rowachol does not preventt biliary stent occlusion. Gastrointest Endosc 1999; 49(3 Pt 1)1316-321.

70.. Lee SP, Carey MC, LaMont JT. Aspirin prevention of cholesterol gallstone formation in prairiee dogs. Science 1981; 2ii(4489):i429-i43i.

(32)

placebo-con-trolledd study on the efficacy of aspirin and doxycycline in preventing clogging of biliary endoprostheses.. Gastrointest Endosc 1989;

35(6):485-489-72.. Sung JY, Shaffer EA, Lam K, Rususka I, Costerton JW. Hydrophobic bile salt inhibits bac-teriall adhesion on biliary stent material. Dig Dis Sci 1994; 39(5):999-ioo6.

73.. Frakes JT, Johanson JF, Stake ƒƒ. Optimal timing for stent replacement in malignant bil-iaryy tract obstruction. Gastrointest Endosc 1993; 3 9 { 2 ) : I 6 4 - I 6 7 .

74.. Prat F, Chapat O, Ducot B, Ponchon T, Pelletier G, Fritsch J et al. A randomized trial of endoscopicc drainage methods for inoperable malignant strictures of the common bile duct.. Gastrointest Endosc 1998; 47(I):I-7.

75.. Matsuda Y, Shimakura K, Akamatsu T. Factors affecting the patency of stents in malig-nantt biliary obstructive disease: univariate and multivariate analysis. Am J Gastroenterol 1991;; 86(7):843-849.

76.. Lammer }. Biliary endoprostheses. Plastic versus metal stents. Radiol Clin North Am

I 9 9 0 ;; 28{6):i2II-I222.

77.. Carr-Locke D, Ball TJ, Conners PJ. Multicenter randomized trial of Wallstent biliary pros-thesiss versus plastic stents. Gastrointest Endosc 1993; 39:310-316.

78.. Huibregtse K, Cheng J, Coene PP, Fockens P, Tytgat GN. Endoscopic placement of expandablee metal stents for biliary strictures—a preliminary report on experience with 333 patients. Endoscopy 1989; 21(6)1280-282.

79.. Bethge N, Wagner HJ, Knyrim K, Zimmermann HB, Starck E, Pausch J et al. Technical failuree of biliary metal stent deployment in a series of 116 applications. Endoscopy 1992; 24(5):395"400. .

80.. Lammer J, Hausegger KA, Fluckiger F, Winkelbauer FW, Wildling R, Klein GE et al. Commonn bile duct obstruction due to malignancy: treatment with plastic versus metal stents.. Radiology 1996; 2oi(i):i67-i72.

81.. Dumonceau JM, Cremer M, Auroux }, Delhaye M, Deviere J. A comparison of Ultraflex Diamondd stents and Wallstents for palliation of distal malignant biliary strictures. Am J Gastroenteroll 2000; 95(31:670-676.

82.. Ahmad ) , Siqueira E, Martin J, Slivka A. Effectiveness of the Ultraflex Diamond stent for thee palliation of malignant biliary obstruction. Endoscopy 2002; 34(10)793-796. 83.. Ferlitsch A, Oesterreicher C, Dumonceau JM, Deviere J, Leban T, Born P et al. Diamond

stentss for palliation of malignant bile duct obstruction: a prospective multicenter evalu-ation.. Endoscopy 2001; 33(8):645-650.

84.. Goldin E, Beyar M, Safra T, Globerman O, Craciun I, Wengrower D et al. A new self-expandable,, nickel-titanium coil stent for esophageal obstruction: a preliminary report. Gastrointestt Endosc 1994; 40(i):64-68.

85.. Smits M, Huibregtse K, Tytgat G. Results of the new nitinol self-expandable stents for distall biliary structures. Endoscopy 1995; 27(7):505-5o8.

86.. Rossi P, Bezzi M, Rossi M, Adam A, Chetty N, Roddie ME et al. Metallic stents in malig-nantt biliary obstruction: results of a multicenter European study of 240 patients. J Vase Intervv Radiol 1994;

5(2):279-285-87.. Coons HG. Self-expanding stainless steel biliary stents. Radiology 1989; 170(3 Pt 2):979-983. .

88.. Irving JD, Adam A, Dick R, Dondelinger RF, Lunderquist A, Roche A. Gianturco expand-ablee metallic biliary stents: results of a European clinical trial. Radiology 1989; i72(2):32i-326. .