Tannenbaum and metal stents in the palliative treatment of malignant distal bile duct obstruction: a comparative study of patency and cost effectiveness

Surg Endosc (2006) 20: 15871593 DOI: 10.1007/s00464-005-0778-1 Springer Science+Business Media, Inc. 2006Tannenbaum and metal stents in the palliative treatment of malignant distal bile duct obstruction: a comparative study of patency and cost eectivenessP. Katsinelos,1 D. Paikos,2 J. Kountouras,3 G. Chatzimavroudis,1 G. Paroutoglou,1 I. Moschos,2 A. Gatopoulou,2A. Beltsis,1 C. Zavos,3 B. Papaziogas11 Department of Endoscopy and Motility Unit, Central Hospital, 41 Ethnikis Aminis St., Thessaloniki, 546 35, Greece2 Department of Gastroenterology, Theagenio Anticancer Hospital, 2 Al. Symeonidi St., Thessaloniki, 540 07, Greece3 Department of Medicine, Second Medical Clinic, Aristotle University of Thessaloniki, Ippokration Hospital, 49 Konstantinoupoleos St., Thessaloniki, 546 42, GreeceReceived: 16 November 2005/Accepted: 2 April 2006/Online publication: 7 August 2006AbstractBackground: Stent clogging is the major limitation of palliative treatment for malignant biliary obstruction. Metal stents have much better patency than plastic stents, but are more expensive. Preliminary data suggest that the recently designed plastic (Tannenbaum) stent has better duration of patency than the polyethylene stent. This study aimed to compare the ecacy and cost eectiveness between the Tannenbaum stent without side holes and the uncovered metal stent for patients with malignant distal common bile duct obstruction.Methods: In this study, 47 patients (median age, 73 years, range, 5686 years) with inoperable malignant distal common bile duct strictures were prospectively randomized to receive either a Tannenbaum stent (n = 24) or an uncovered self-expandable metal stent (n = 23). The patients were clinically evaluated, and biochemical tests were analyzed if necessary until their death or surgery for gastric outlet obstruction. Cumulative rst stent patency and patient survival were compared between the two groups. Cost-eectiveness analysis also was performed for the two study groups.Results: The two groups were comparable in terms of age, gender, and diagnosis. The median rst stent patency was longer in the metal group than in the Tannenbaum stent group (255 vs 123.5 days; p = 0.002). There was no signicant dierence in survival between the two groups. The total cost associated with the Tannenbaum stents was lower than for the metal stents (17,700 vs 30,100 euros; p = 0.001), especially for patients with liver metastases (3,000 vs 6,900 euros; p < 0.001). Conclusions: Metal stent placement is an eective treatment for inoperable malignant distal common bile duct obstruction, but Tannenbaum stent placement is a cost- saving strategy, as compared with metal stent placement, especially for patients with liver metastases and expected short survival time.Key words: Malignant distal bile duct obstruction Metal stent Stent clogging Tannenbaum stent Endoscopic drainage Patency Cost effectiveness ERCPThe management of patients with malignant biliary obstruction is challenging for all specialists involved in their care. Evaluation should focus on potential surgical resection, which oers the principal chance of cure. At the time of diagnosis, however, more than 80% of patients with malignant biliary obstruction cannot be cured by operative resection because of locally advanced or metastatic disease [1, 2]. Patients who have inoperable tumors have a grim prognosis in terms of survival and quality of life. The average life expectancy of these patients is less than 1 year. Therefore, the therapeutic objective is short-term palliation of symptoms in an eort to enhance the quality of life. Palliative methods for bile duct malignancies include surgical bypass, intraluminal and external beam radiation therapy, chemotherapy, photodynamic therapy, and endoscopic interventions [36].The use of endoscopic retrograde cholangiopancreatography (ERCP) or percutaneous transhepatic cholangiography should be limited primarily to palliation of jaundice for patients with unresectable tumors, and to the establishment of tissue diagnoses in ambiguous cases. Palliation of jaundice and its complications is optimal with self-expanding stents. Safe and eective drainage can be achieved by using ERCP for targeted endoscopic placement of unilateral stents in most cases. EndoscopicCorrespondence to: J. Kountouras1588drainage may improve the patients sense of well-being, and stent patency therefore becomes an important factor aecting quality of life and cost of care [7, 8].The currently available biliary stents are classied into two main categories, according to their composition: plastic (polyethylene, Teon) and self-expanding metal stents (SEMS) (M.I. Tech, G.C. Medical Co Ltd., Pyungtaek, Korea). The major advantage of SEMS over plastic stents is their lower rate of occlusion (median patency duration, 812 vs 45 months), frequently caused by cancer ingrowth, cancer overgrowth, and, exceptionally, biliary sludge. On the other hand, the major disadvantages of SEMS are higher cost and diculty with repositioning or extraction once deployed.The major drawback of plastic stents is their propensity to clog because of bacterial biolm formation and sludge deposition. Clogging often necessitates repeated stent changes for maintenance of biliary drainage. Recently, Teon Tannenbaum stents (TT) (Sohendra ST-2; Wilson Cook, Winston-Salem, NC, USA) without side holes were reported to have longer duration of patency than conventional polyethylene stents with side holes and Teon pigtail stents, approximating the patency duration reported for SEMS [9].The data [10] comparing the patency and cost eectiveness between TT stents and SEMS in the palliative treatment of patients with unresectable malignant strictures of the common bile duct are limited. Therefore, we compare the clinical ecacy and cost eectiveness between TT stents and uncovered SEMS for patients with unresectable malignant distal common bile duct obstruction.Patients and methodsA prospective study conducted at two centers in the city of Thessaloniki, Greece, between January 2000 and September 2005, included 47 patients randomized to receive either a plastic TT stent, or a SEMS. Patients referred from various hospitals in northern Greece were included in the study if they were older than 18 years and had a life expectancy of at least 1 month, no encephalopathy, a malignant distant biliary stricture that was inoperable (because of either advanced age or major comorbidity) or unresectable (because of vascular invasion or distant metastases on imaging studies), no prior stent insertion, and no previous surgical or radiologic procedures performed on the biliary tree.Diagnostic procedures included hematologic and biochemical analyses and abdominal ultrasound, computed tomography (CT) scan, or magnetic resonance imaging (MRI) cholangiopancreatography. Exclusion criteria specied a lesion occupying the middle or the bifurcation of the common hepatic duct, duodenal obstruction, advanced metastatic disease making survival until stent exchange unlikely (less than 3 weeks), and American Society of Anesthesiology (ASA) general status grade 4 or 5. The study protocol was approved by the local ethics committees of both hospitals involved, and all the patients gave informed consent before entering the study.Fig. 1. Cholangiogram demonstrating a distal malignant biliary stricture in a 72-year-old man with cholangiocarcinoma.TreatmentWith the patients under midazolam plus pethidine sedation, experienced senior endoscopists performed ERCP using a duodenoscope with a large accessory channel (TJF-140; Olympus Optical Co, Ltd., Tokyo, Japan). A diagnostic ERCP was performed to assess the biliopancreatic anatomy. When a malignant distal common bile duct stricture was seen (Fig. 1) and deep cannulation was successful, patients were randomly assigned to receive either a metal (Fig. 2) or a TT stent (Fig. 3). Randomization was performed by computer-generated random numbers in sealed envelopes.In a few cases, dilation was necessary before stent placement. A small sphincterotomy also was performed before stent placement. Intravenous antibiotics (ciprooxacin 400 mg) were administered prophylactically 1 h before the ERCP or, in some cases, as treatment for cholangitis.Follow-up evaluationBiliary drainage was considered successful if serum bilirubin dropped more than 20% by 48 h after stenting and bilateral aerobilia was visible on a plain abdominal radiograph. Stent patency represented the interval between the time of stent insertion and the time of its replacement or the presence of both fever and jaundice at the time of death. Immediate follow-up assessment included evaluation for complications directly linked to stent placement within 1 week after the procedure. Complications of ERCP and endoscopic sphincterotomy were evaluated according to the criteria by Cotton et al. [11].ProsthesesThe TT stent is a straight biliary prosthesis consisting of a stainless steel mesh between an inner Teon coating and an outer polyamide layer without side holes and with multiple anchoring aps that do not perforate its wall.The metal prosthesis is pliable, self-expanding, and exible in the longitudinal axis. Mounted on a delivery system 222 cm long, it is constrained and moderately elongated by an invaginated membrane.The outer diameter, when constrained, is 9-Fr. Friction between the layers of the double-over membrane is reduced by injection of contrast medium at 5.0 atmospheres by a pressure device. After progressive withdrawal of the membrane, the prosthesis expands to a diameter of 8 to 10 mm (30 Fr) and shortens from 10 to 6.8 cm in length.The delivery system with the constrained stent is inserted over a 4-m, 0.35-mm guidewire after a small papillotomy. Transpapillary placement is performed under direct endoscopic view. The radio-opaque stent and two markers on the delivery catheter allow uoroscopic control of stent release and positioning across the stricture. Immediate removal of the metal stent after insertion is feasible, but after 2 to 7 days, stent removal becomes impossible.Table 1. Baseline patient characteristicsTreatmentTannenbaum stent n (range)Metallic stent n (range)No. of patients 24 23 Sex (male/female) 10/14 14/9 Median age (years) 72 (5682) 74 (5786) Median serum bilirubim(mg/dl)10.6 (4.819.3) 14.1 (5.3635.8)Diagnoses (n)Pancreatic carcinoma 13 12 Cholangiocarcinoma 4 4 Ampullary tumor 3 2 Compressive lymph nodes 4 5 No. of patients with livermetastases9 6ASA classication 2 11 9 3 13 14 1589occlusion of a TT stent occurred, it was removed with a snare, slit open with a scalpel to conrm occlusion, and replaced with another TT stent. If the patient died with a patent stent, the interval from insertion to death was recorded and analyzed. Patients were withdrawn from long-term analysis if they underwent surgical resection of the tumor.The primary end point in this study was the interval between stent insertion and rst exchange, or the presence of jaundice at death without a stent exchange. Secondary end points were the number of stent exchanges and the complications recorded in the two groups.Fig. 2. Final radiography demonstrating patency and correct position of a metal stent.Fig. 3. Endoscopic view of a well-functioning 10-Fr Tannenbaum endoprosthesis.Cost-eectiveness analysisThe cost associated with placement of the rst stent and treatment of stent occlusion was evaluated by recording the number of stents placed. The cost for both treatment strategies diered only with respect to stent price (i.e., 1,100 euros for a metal and 300 euros for a TT stent), because in the public health system of Greece, patients are charged only for the materials and not the operations per se. Eectiveness was expressed as the mean number of procedures required for each treatment strategy until death or the end of evaluation.Statistical analysisThe analysis was performed using the statistical software package SPSS (Statistical Package for the Social Sciences, Version 13.0, Chicago, IL, USA). Cumulative stent survival (patency) and patient survival were estimated according to the KaplanMeier technique, supplemented by the log-rank test for comparison between groups. Stent patency, calculated in days, represented the interval between the time of stent insertion and the time of rst replacement or the death of the patient with an unoccluded stent. All results are expressed as medians. The statistical analysis of the results was performed using nonparametric tests. More specically, the MannWhitney test was used for quantitative variables and the chi-square test for qualitative variables. Signicance was set at a p value less than 0.05.Stents found to be occluded within 2 weeks after insertion were excluded from analysis because early stent blockage usually does not result from bacterial colonization, but rather from improper positioning, blood clots, or mucus plugs. Because palliation, and subsequently patient comfort and quality of life, is the essential objective of endoprosthesis insertion, we believed that the longer the patient lived without biliary symptoms, the better the palliation.Data were generated monthly by direct contact or by telephone interview with the patient or his or her primary care physician at monthly intervals until death or the end of the follow-up period (September 2005). During the interview, the patient was specically asked about the presence of jaundice. Stent occlusion was assessed on clinical (cholangitis or recurrent jaundice) and laboratory grounds. In the case of an occluded metallic stent, a plastic stent was placed inside it. WhenResultsBetween January 2000 and September 2005, 47 patients (24 men and 23 women) who met the inclusion criteria were recruited into the study. The median age of the patients was 73 years (range, 5686 years). The baseline characteristics of the patients receiving TT stents and SEMS are shown in Table 1. In terms of sex ratio, type1590Table 2. Details of the results from the randomized prospective studyTreatmentTannenbaum stent n (range) Metallic stent n (range) p ValueNo. of patients 24 23 Subsequent duodenal obstruction 4 1 NS Died with patent rst stent 8 16 0.013 Complications of rst stent procedure 4 0 NS Pancreatitis 3 0 NS Hemorrhage 1 0 NS Cholangitis 0 0 NS Perforation 0 0 NS Occlusion of stents Cholangitis 15 18 NS Anicteric cholestasis 9 5 NS Median rst stent patency (days) 123.5 (90144) 255 (30558) 0.002 Median survival (days) 207 (90571) 272 (30873) NS Median survival with liver metastases (days) 121 (90205) 80 (30460) NS Median survival without liver metastases (days) 431 (184571) 529 (97873) NS Number of stent exchanges 0 8 16 0.013 1 8 2 0.042 2 1 1 NS 3 3 4 NS 4 4 0 NS Total number of ERCPs required 59 39 0.025NS, not signicant, ERCP, endoscopic retrograde cholangiopancreatography1000 8006004002000Time (days)Type of stentMetallic stentTannenbaum stentof malignancy, location of the malignancy with respect to the common bile duct, and ASA classication, the two groups were similar.Cannulation of the common bile duct was achieved after a small sphincterotomy in 40 patients (85.1%) and after precut papillotomy in 7 patients (14.9%). The pancreatic duct was cannulated in all the patients. In 25 of the patients (53%), a stricture was visualized in both the pancreatic and distal common bile ducts (double duct sign). All SEMS and TT stents were successfully deployed. All were in a reasonable position, with a subsequent decrease in serum bilirubin or clinical resolution of jaundice.Three patients in the TT group experienced mild pancreatitis and one moderate hemorrhage (Table 2), which were treated with a 5-ml injection of adrenaline solution (1:10000) plus dextrose 50%. No other major, procedure-related complications such as cholangitis, perforation, early occlusion, or signicant bleeding occurred in either group.During the follow-up period, symptoms developed in ve patients (10.6%) because of duodenal obstruction (Table 2), which necessitated surgical biliary bypass for four of the patients and radiologic intervention for one patient. The main consequence of occlusion in the TT group was cholangitis in 15 patients and anicteric cholestasis in 9 patients, whereas in the SEMS group, the consequence was cholangitis in 18 patients and anicteric cholestasis in 5 patients (Table 2).There was no signicant dierence in overall patient survival between the two groups (Fig. 4). Median survival in the SEMS group was 272 days (range, 30873 days), as compared with 207 days (range, 90571 days) in the TT stent group (Table 2). The cumulative patency of the rst stent was signicantly longer (p = 0.002) in the SEMS group than in the TT stent group (Fig. 5).The median duration of rst stent patency was 123.5 days (range, 90144 days) in the TT stent group and 255 days (range, 30558 days) in the SEMS group (Table 2).The overall cost for all the patients in the TT stent group was 17,700 euros, as compared with 30,100 euros in the SEMS group (p = 0.001) (Table 3). Further cost analysis showed that the total cost for all the patients for the rst stent placement was 7,200 euros in the TT stent group and 25,300 euros in the SEMS group. For patients with liver metastases, the total cost for all the patients was 3,000 euros for the TT stent group and 6,900 euros for the SEMS group (p < 0.001), whereas the cost for the patients free of liver metastases was 14,700 euros for TT stent group and 23,200 euros for SEMS group. The dierence is not signicant.Cumulative Patient Survival1,00,80,60,40,20,0Fig. 4. Overall patients survival.1591Cumulative Patency of the 1st stent1,00,80,60,40,20,0type of stentMetallic stentTannenbaum stent600 5004003002001000Time (days)Fig. 5. Cumulative patency of the rst stent.DiscussionObstructive jaundice attributable to inoperable biliary malignant disease is associated with complications such as malabsorption, coagulopathy, progressive hepato-cellular and renal dysfunction, and cholangitis, in addition to disabling pruritus, which greatly interferes with terminal patients quality of life. Endoscopic placement of biliary stents decreases the risk of these complications and is considered the palliation procedure of choice for patients with unresectable tumors.In this prospective randomized comparison between SEMS and the TT stent for palliation of malignant distal biliary obstruction, the success rate for transpapillary positioning, the initial drainage quality, and the number of early complications were similar in the two groups. One disadvantage of plastic stents is the risk of occlusion within 3 to 6 months after stent placement[12]. Despite the initial high ecacy in achieving biliary drainage, endoscopic TT stenting also has been limited by clogging within 3 to 5 months in our patients.Stent blockage appears to be a multifactorial process in which, stent design and materials aside, proteins, bile viscosity, and, most importantly, bacteria and the immunologic (immunoglobulins) defense system of the biliary system play a role [1317]. The sphincter of Oddi separates the colonized duodenum for the sterile biliary tract and acts as a mechanical barrier to microbial colonization [18]. All our patients underwent sphincterectomy, which abolished this barrier. In addition, disruption of this barrier by insertion of biliary stents across the sphincter into the duodenal lumen allows bacteria to ascend into the biliary tract. This leads to bacterial biolm and sludge formation in the biliary endoprosthesis.Specically, the process of stent blockage is initiated by bacteria attaching to the stent surface with their pili. Deconjugation of bilirubin diglucuronide and lecithin in the bile by bacterial b-glucuronidase and phospholipase leads to precipitation of calcium bilirubinate and calcium salts of fatty acids. With time, the growth of biolm and progressive agglomeration of the bile sediment forms biliary sludge, which results in occlusion of the stent lumen [1415]. Notably, the capacity of phagocytes and T-lymphocytes to clear bacterial pathogens is reduced in patients with untreated gastrointestinal cancers, including pancreatic cancers, and this dysfunction may increase their risk of becoming infected [19].In view of the aforementioned considerations, different strategies for the prevention of bacterial adhesion including new stent materials, prophylactic antibiotics, and bile salts have been attempted, but with variable success [2023]. Our patients prophylactically received intravenous ciprooxacin (400 mg/day) for a short poststenting time (3 days), possibly because long-term use of ciprooxacin does not seem to prevent blockage of biliary stents [20]. This regimen also was administered orally to our 33 patients (15 with TT stents and 18 with SEMS) who presented with stent occlusion leading to cholangitis for at least 10 days. The use of quinolones is recommended for patients with cholangitis because these agents eectively penetrate an obstructed biliary tree and can be administered orally [23].Although it has been suggested that white bile (largely devoid of bilirubin and bile acids) is associated with signicantly worse survival for patients with malignant bilirary obstruction [24], this prognostic feature was not observed in our patients.For more than 15 years, SEMS has been used clinically for malignant biliary strictures, and many studies [2, 9, 10, 25] have documented that SEMS has better patency rates than plastic stents. In our study, the overall patency rate for the rst SEMS was signicantly longer statistically (p = 0.002) than the rate for TT stents. Patients treated with SEMS experience fewer symptomatic biliary complications, have a longer symptom-free survival, undergo fewer ERCPs, and have a shorter hospital stay than those treated with a plastic stent [26, 27]. This superiority of SEMS is attributable to both its internal diameter (30 Fr or 10 mm, as opposed to the 10-Fr or 3-mm diameter of plastic stents) and the mechanism by which it becomes obstructed. Occlusion develops more slowly, essentially because of tumor ingrowth or overgrowth, and not because of the biolm formation and sludge deposition [28]. Studies have reported that tumor growth over the ends of SEMS occurs frequently, indicating that adequate stent length is of critical importance [7, 8, 29]. However, the use of SEMS is limited by high cost relative to that of plastic stents (as much as a 10-fold dierence). Another disadvantage is the diculty removing SEMS after insertion in the event of incorrect positioning, tumor overgrowth, or occlusion.The TT stent is reported to be associated with a remarkable increase (almost double) in the patency rate of the rst stent and a signicantly lower number of stent exchanges, as compared with conventional plastic stents [30, 31]. Our prospective randomized trial, with its small number of patients (n = 24) in the TT group, showed a median rst stent patency of 123.5 days (range, 90144 days), and did not conrm the expecta-1592Table 3. Cost analysis (in euros)TreatmentTannenbaum stent Metallic stent p ValueTotal cost for all patients (median) 17,700 (738) 30,100 (1,309) 0.001 Total cost of rst stent for all patients 7,200 25,300 Total cost for patients with liver metastases (median) 3,000 (300) 6,900 (1,100) <0.001 Total cost for patients free of liver metastases (median) 14,700 (900) 23,200 (1,100) NSNS, not signicanttions of previous nonrandomized studies [13, 32, 33]. Our ndings, however, are supported by recent reports [2224] with a larger number of patients and results similar to ours.This study is the second investigation to compare TT stents with SEMS for endoscopic palliation of malignant distal common bile duct obstruction. In the rst study, Kaassis et al. [10] compared the patency and cost eectiveness of TT and SEMS in the palliative treatment of 118 patients with unresectable malignant strictures of the common bile duct. These authors found that SEMS placement was the most eective treatment for inoperable malignant common bile duct strictures. This approach was especially cost eective for patients without hepatic metastases, whereas a TT stent should be placed in patients with spread of the tumor to the liver. In this regard, there is a consensus that patients with longer expected survival should be treated by insertion of SEMS.Yeoh et al. [34] demonstrated that the cost of stent strategies for managing malignant common bile duct strictures depends on the survival of the patients. The use of plastic stents was less costly for patients surviving less than 4 months, whereas SEMS was relatively less expensive for patients with a longer survival. The current study, reporting a median survival of 329 days (range, 97873 days) for patients free of liver metastases, as compared with 80 days (range, 30460 days) for patients with liver metastases, justies the use of SEMS and is in agreement with the study by Kaassis et al. [10]Our study diers from previous investigations in the method used for the cost-eectiveness analysis between the groups (TT stent vs SEMS). We did not calculate the cost of the ERCPs and the days of hospitalization because they are oered free of charge in the public health institutions of Greece. This resulted in a signicantly lower total cost for all the patients in the TT group (p = 0.001), especially those with liver metastases (3,000 euros for the TT group vs 6,900 euros for the SEMS group; p < 0.001), because the total number of ERCPs needed in each group (59 procedures in the TT stent group vs 39 in the SEMS group) was not taken into consideration. On the contrary, in countries such as the United States, in which the cost of an ERCP is particularly high and included in the nal cost-eectiveness analysis, the total cost of SEMS for all patients is signicantly lower because long-term patency results in fewer ERCPs for replacement of occluded stents, especially in patients without liver metastases and thus longer life expectancy.In conclusion, the current study demonstrated that the presence or absence of liver metastases is useful in determining the optimal therapeutic approach for patients with malignant distal common bile duct obstruction. The use of SEMS for patients without liver metastases is cost eective. Placement of a plastic stent is a cost-saving strategy, as compared placement of SEMS, for patients with liver metastases because expected survival is shorter.References1. Shepherd HA, Royle G, Ross AP, Diba A, Arthur M, Colin-Jones D (1988) Endoscopic biliary endoprosthesis in the palliation of malignant obstruction of the distal common bile duct: a randomized trial. Br J Surg 75: 116611682. Prat F, Chapat O, Ducot B, Ponchon T, Pelletier G, Fritsch J, Choury AD, Buet C (1998) A randomized trial of endoscopic drainage methods for inoperable malignant strictures of the common bile duct. Gastrointest Endosc 47: 173. Lichtenstein DR, Carr-Locke DL (1995) Endoscopic palliation for unresectable pancreatic carcinoma. Surg Clin North Am 75: 969 9884. Smith AC, Dowsett JF, Russell RC, Hateld AR, Cotton PB (1994) Randomised trial of endoscopic stenting versus surgical bypass in malignant low bile duct obstruction. Lancet 344: 1655 16605. Mado DC, Wallace MJ (2002) Palliative treatment of unresec-table bile duct cancer: which stent? Which approach? Surg Oncol Clin North Am 11: 9239396. Pinol V, Castells A, Bordas JM, Real MI, Llach J, Montana X, Feu F, Navarro S (2002) Percutaneous self-expanding metal stents versus endoscopic polyethylene endoprostheses for treating malignant biliary obstruction: randomized clinical trial. Radiology 225: 27347. Eisen GM, Dominitz JA, Faigel DO, Goldstein JL, Kalloo AN, Petersen BT, Raddawi HM, Ryan ME, Vargo JJ III, Young HS, Fanelli RD, Hyman NH, Wheeler-Harbaugh J, American Society for Gastrointestinal Endoscopy. Standards of Practice Committee (2001) An annotated algorithmic approach to malignant biliary obstruction. Gastrointest Endosc 53: 8498528. Van Steenbergen W (2001) Treatment of malignant biliary stenosis: which stent to use? Acta Gastroenterol Belg 64: 3093139. Levy MJ, Baron TH, Gostout CJ, Petersen BT, Farnell MB (2004) Palliation of malignant extrahepatic biliary obstruction with plastic versus expandable metal stents: an evidence-based approach. Clin Gastroenterol Hepatol 2: 27328510. Kaassis M, Boyer J, Dumas R, Ponchon T, Coumaros D, Delcenserie R, Canard JM, Fritsch J, Rey JF, Burtin P (2003) Plastic or metal stents for malignant stricture of the common bile duct? Results of a randomized prospective study. Gastrointest Endosc 57: 17818211. Cotton PB, Lehman G, Vennes J, Geenen JE, Russell RC, Meyers WC, Liguory C, Nickl N (1991) Endoscopic sphincterotomy complications and their management: an attempt at consensus. Gastrointest Endosc 37: 383393159312. Schilling D, Rink G, Arnold JC, Benz C, Adamek HE, Jakobs R, Riemann JF (2003) Prospective, randomized, single-center trial comparing 3 dierent 10F plastic stents in malignant mid and distal bile duct strictures. Gastrointest Endosc 58: 545813. England RE, Martin DF, Morris J, Sheridan MB, Frost R, Freeman A, Lawrie B, Deakin M, Fraser I, Smith K (2000) A prospective randomised multicentre trial comparing 10-Fr Teon Tannenbaum stents with 10-Fr polyethylene Cotton-Leung stents in patients with malignant common duct strictures. Gut 46: 395 40014. Chan FKL, Suen M, Li JYW, Sung JJY (1998) Bile immunoglobulins and blockage of biliary endoprosthesis: an immunohistochemical study. Biomed Pharmacother 52: 40340715. Leung JW, Ling TK, Kung JL, Vallance-Owen J (1988) The role of bacteria in the blockage of biliary stents. Gastrointest Endosc 34: 192216. Dowidar N, Kolmos HJ, Matzen P (1992) Experimental clogging of biliary endoprostheses: role of bacteria, endoprosthesis material, and design. Scand J Gastroenterol 27: 778017. Jansen B, Goodman LP, Ruiten D (1993) Bacterial adherence to hydrophilic polymer-coated polyurethane stents. Gastrointest Endosc 39: 67067318. Sung JY, Olson ME, Leung JWC, Lundberg MS, Costerton JW (1990) The sphincter of Oddi is a boundary for bacterial colonization in the feline biliary tract. Micro Ecology Health Dis 3: 199 20819. Amati L, Caradonna L, Greco B, Leo S, Caccavo D, Jirillo E (1998) Impairment of phagocytic and T-cell-mediated antibacterial activity and plasma endotoxins in patients with untreated gastrointestinal cancer. Scand J Gastroenterol 33: 84785220. Sung JJ, Sollano JD, Lai CW, Ismael A, Yung MY, Tumala I, Chung SC (1999) Long-term ciprooxacin treatment for the prevention of biliary stent blockage: a prospective randomized study. Am J Gastroenterol 94: 3197320121. Halm U, Schiefke, Fleig WE, Mossner J, Keim V (2001) Ooxacin and ursodeoxycholic acid versus ursodeoxycholic acid alone to prevent occlusion of biliary stents: a prospective, randomized trial. Endoscopy 33: 49149422. Galandi D, Schwarzer G, Bassler D, Allgaier HP (2002) Ursodeoxycholic acid and/or antibiotics for prevention of biliary stent occlusion. Cochrane Database Syst Rev 3:CD00304323. Rerknimitr R, Fogel EL, Kalayci C, Esber E, Lehman GA, Sherman S (2002) Microbiology of bile in patients with cholangitis or cholestasis with and without plastic biliary endoprosthesis. Gastrointest Endosc 56: 88588924. Ahuja V, Garg PK, Kumar D, Goindi G, Tandon RK (2002) Presence of white bile associated with lower survival in malignant biliary obstruction. Gastrointest Endosc 55: 18619125. Wagner HJ, Knyrim K, Vakil N, Klose KJ (1993) Plastic endoprostheses versus metal stents in the palliative treatment of malignant hilar biliary obstruction: a prospective and randomized trial. Endoscopy 25: 21321826. Isayama H, Komatsu Y, Tsujino T, Sasahira N, Hirano K, Toda N, Nakai Y, Yamamoto N, Tada M, Yoshida H, Shiratori Y, Kawabe T, Omata M (2004) A prospective randomised study of covered versus uncovered diamond stents for the management of distal malignant biliary obstruction. Gut 53: 72973427. Knyrim K, Wagner HJ, Pausch J, Vakil N (1993) A prospective, randomized, controlled trial of metal stents for malignant obstruction of the common bile duct. Endoscopy 25: 20721228. Moesch C, Sautereau D, Cessot F, Berry P, Mounier M, Gainant A, Pillegand B (1991) Physicochemical and bacteriological analysis of the contents of occluded biliary endoprostheses. Hepatology 14: 1142114629. Shah RJ, Howell DA, Desilets DJ, Sheth SG, Parsons WG, Okolo P III, Lehman GA, Sherman S, Baillie J, Branch MS, Pleskow D, Chuttani R, Bosco JJ (2003) Multicenter randomized trial of the spiral Z-stent compared with the Wallstent for malignant biliary obstruction. Gastrointest Endosc 57: 83083630. Binmoeller KF, Seitz U, Seifert H, Thonke F, Sikka S, Soehendra N (1995) The Tannenbaum stent: a new plastic biliary stent without side holes. Am J Gastroenterol 90: 1764176831. Seitz U, Vadeyar H, Soehendra N (1994) Prolonged patency with a new-design Teon biliary prosthesis. Endoscopy 26: 47848232. van Berkel AM, Boland C, Redekop WK, Bergman JJ, Groen AK, Tytgat GN, Huibregtse K (1998) A prospective randomized trial of Teon versus polyethylene stents for distal malignant biliary obstruction. Endoscopy 30: 68168633. Terruzzi V, Comin U, De Grazia F, Toti GL, Zambelli A, Beretta S, Minoli G (2000) Prospective randomized trial comparing Tannenbaum Teon and standard polyethylene stents in distal malignant biliary stenosis. Gastrointest Endosc 51: 232734. Yeoh KG, Zimmerman MJ, Cunningham JT, Cotton PB (1999) Comparative costs of metal versus plastic biliary stent strategies for malignant obstructive jaundice by decision analysis. Gastrointest Endosc 49: 466471