Antireflux Versus Conventional Plastic Stent in Malignant Biliary Obstruction: A Prospective Randomized Study

Abstract

Objective:

Endoscopic stents are used to relieve obstructive jaundice. The purpose of this prospective randomized study was to compare the patency of antireflux and conventional plastic biliary stent in relieving distal malignant biliary obstruction.

Materials and Methods:

All jaundiced patients admitted to hospital with suspected unresectable malignant distal biliary stricture between October 2009 and September 2010 were evaluated for the study. Eligible patients were randomized either to antireflux or conventional plastic stent arms. The primary endpoint was stent patency and the follow-up was continued either until the stent was occluded or until 6 months after the stent placement.

Results:

At an interim analysis, antireflux stents (ARSs; n = 6) had a significantly shorter median patency of 34 (8–49) days compared with the conventional stent (n = 7) patency of 167 (38–214) days (P = .0003). Based on these results, the study was terminated due to ethical concerns.

Conclusion:

According to these results, the use of this ARS is not recommended.

Introduction

Endoscopic biliary stenting is a widely used palliation in relieving obstructive jaundice in patients with unresectable malignant distal biliary stricture. Endoscopic stents are as effective as surgical biliary bypass with lower procedure-related mortality, morbidity, and hospitalization time.^1,2 Plastic stents (PSs) are relatively cheap, but their use is limited by the early occlusions requiring stent exchange every 3–5 months.^3,4 In case of liver metastases and expected short survival time, PSs are a good option for palliative treatment of biliary obstruction.^5,6 Self-expanding metal stents have longer patency and the major cause of stent occlusion is tumor ingrowth, but higher cost and inability to be removed restrict their use.^7,8

The exact mechanisms leading to PS occlusion remain unknown. Dietary fibers and plant materials have been found in PSs, suggesting that duodenobiliary reflux may be a major factor contributing to stent occlusion.^9,10 To eliminate retrograde flow from the duodenum, a PS with an antireflux valve has been developed (Fusion^® Marathon^™ Anti-Reflux Biliary Stent; Cook Endoscopy, Winston-Salem, NC). The aim of this study was to compare the patency of this novel antireflux stent (ARS) with the conventional PS.

Materials and Methods

Patients and data collection

All consecutive patients with jaundice or elevated liver enzymes admitted to our tertiary referral hospital with suspected unresectable malignant distal biliary stricture between October 2009 and September 2010 were to be evaluated for enrollment in the study. Eligible patients were randomized either to antireflux PS arm or conventional PS arm. No power analysis was performed, we had estimated to have ∼50 patients enrolled in the study, 25 for each arm. Inclusion criteria were jaundice or elevation of the liver enzymes secondary to an inoperable malignant distal common bile duct stricture. Exclusion criteria were¹ previous papillary procedures,² active cholangitis,³ altered route to papilla due to previous surgery,⁴ pregnancy, and⁵ age under 18 or over 89 years. The presence of a malignant stricture was confirmed with typical endoscopic retrograde cholangiopancreatography (ERCP) findings, computed tomography, and/or ultrasound. An informed consent was obtained from the patients. The study was approved by Ethics Committee of Turku University Hospital and registered in ClinicalTrials.gov (NCT00990366).

Study design

Patients were randomized by closed envelope method just before stent insertion to receive either a 10 Fr ARS (Fusion Marathon Anti-Reflux Biliary Stent; Cook Endoscopy) or a conventional 10 Fr PS (QuickPlaceV^™, DoubleLayer; Olympus Medical Systems Corp., Tokyo, Japan [n = 7] or Flexima^™; Boston Scientific Corp., Natick, MA [n = 1]). The ARS consisted of a plastic Tannenbaum (Cook Endoscopy) stent equipped with an antireflux “sock,” made of expanded polytetrafluoroethylene material, in the duodenal end of the stent. It has been designed to eliminate the duodenal reflux allowing the bile flow freely to the bowel.

All patients were given prophylactic antibiotic (cefuroxime 1.5 g) intravenously 1 hour before the procedure. An endoscopic sphincterotomy (ES) and dilatation of the stricture were performed, if needed. All the ERCP data, including patient demographics, findings, procedures, immediate complications, the stent type, and length, were prospectively recorded to a Microsoft Excel^® file, which has been created for study purposes in our ERCP unit.

The patients were followed by phone interview and laboratory tests at 1, 3, and 6 months after the stent placement or until the stent replacement or death. Hemoglobin (Hb), white blood cell (WBC) count, C-reactive protein (CRP), plasma alkaline phosphatase (AP), bilirubin (Bil), and carbohydrate antigen 19–9 (CA 19–9) were measured at each follow-up.

Outcome measures

The primary endpoint of the study was stent patency. Stent occlusion was defined as jaundice that improved after stent replacement. The patency of the stent was recorded in days from the stent insertion to the replacement of an occluded stent. If the patient died jaundiced, the stent was regarded as obstructed. In case of stent occlusion, the old stent was removed and a new stent was inserted. The follow-up period was considered the patency time, if the stent was still functioning after 6 months of insertion of the stent. If the patient died with a functioning stent, the time from the stent insertion to death was regarded as the patency time.

The secondary endpoint was acute cholangitis (an episode with jaundice or elevated liver enzymes and hyperthermia above 38°C for over 24 hours).

Statistics

Continuous variables were characterized using medians and range of values and categorical variables using frequencies and percents. Differences between stent groups in normally distributed continuous variables were tested using independent sample t-test and Mann–Whitney U-test when appropriate. Kaplan-Meier curves between stents were compared using log-rank test. P-values <.05 were considered statistically significant. Statistical analyses were performed using SAS System for Windows, Version 9.2 (SAS Institute, Inc., Cary, NC).

Results

Between October 2009 and May 2010, altogether 15 out of 34 patients were randomized to either ARS (n = 7) or PS (n = 8) arms. Stent insertion was technically successful in all, but 1 patient, who underwent repeated unsuccessful insertion attempts of an ARS and finally had a conventional PS inserted. He was later excluded from the study since the stricture proved to be benign. One patient was also excluded from the PS arm as the tumor was operable by pancreaticoduodenectomy. All analyzed patients (n = 13) were diagnosed with pancreatic cancer located in the head of the pancreas causing distal biliary obstruction. None of the randomized patients had liver cirrhosis or renal dysfunction. Detailed patient demographics and results are presented in Table 1. Mean age of the analyzed patients with ARS was 72 years (range 60–86 years) and 67 years (range 42–80 years) in the PS group (P = ns). Six patients were not evaluated for the enrollment in the study due to misconception regarding the study protocol. Eleven patients were not included in the randomization protocol due to prior procedures involving the papilla,⁵ gastrointestinal obstruction,² inability to give informed consent,² active cholangitis,¹ and a tumor considered to be operable.¹ Two of the patients were not randomized due to multiple liver metastasis causing icterus¹ and prior laparotomy within 10 days, affecting the CRP value.¹ The flow diagram is shown in Figure 1.

FIG. 1.

Flow diagram of the randomized patients.

Table 1.

Detailed Patient Demographics and Results

Patients	Gender/age (years)	Stent type (1) QuickPlaceV ^™ (2) Flexima ^™	ES	Stent patency (days)	Stent exchange	Survival (days)/alive at last follow-up
1	M/62	ARS	+	32	PS	52
2	M/64	PS (1)	+	125	SEMS	Alive
3	F/86	ARS	+	34	None, TC	36
4	M/67	PS (2)	+	167	SEMS	Alive
5	M/61	PS (1)	+	38 (P)		38
6	M/42	PS (1)	+	137 (P)		137
7	M/66	ARS	+	Technical failure, excluded
8	M/80	ARS	+	8	SEMS	Alive
9	F/67	ARS	+	18	PS	Alive
10	F/75	ARS	+	45	PS	82
11	M/60	ARS	−	49	SEMS	Alive
12	F/74	PS (1)	+	214 (P)		Alive
13	M/44	PS (1)	+	Excluded
14	F/80	PS (1)	+	187 (P)		Alive
15	M/80	PS (1)	+	176 (P)		Alive

ARS, antireflux stent; ES, endoscopic sphincterotomy; F, female; M, male; P, patent; PS, plastic stent; SEMS, self-expanding metal stent; TC, terminal care.

All, but 1 patient, with ARS underwent endoscopic sphincterotomy (ES) before the insertion of the ARS and ES was performed in all PS patients (Table 1). Altogether, 7 patients underwent dilatation of the stricture before stent placement. The insertion of the ARS was regarded technically similar to PS insertion by our experienced ERCP endoscopists.

After the enrollment of these 15 patients, a clinical suspicion of early occlusions in the ARS arm was raised during the study period and an unplanned interim analysis of the remaining 13 patients was performed in May 2010. This interim analysis showed significantly shorter (P = .0003) stent patency in the ARS arm. The median patency was 34 (8–49) days in the ARS arm compared to the median patency of 167 (38–214) days in the PS arm. One patient with an ARS needed an early reintervention (irrigation) due to persistent jaundice and was considered a primary failure. This intervention time was considered the patency time. Based on these interim results, the use of this ARS was considered unethical and the study was prematurely closed with no further enrollment permitted.

There were no statistical differences in laboratory values between the two study arms before stent insertion. At 1-month follow-up, CRP and AP values were significantly higher (P = .007 and P = .03, respectively) in the ARS arm compared with the PS arm; detailed laboratory values are shown in Table 2.

Table 2.

Laboratory Values Before Stent Insertion (−0) and at 1-Month Follow-Up (−1) and P Values Between Antireflux and Conventional Plastic Stents

Parameter	ARS, median, range (No. analyzed)	PS, median, range (No. analyzed)	P	Normal range
B-Hb-0	113, 82–138 (6)	129, 101–152 (7)	.14	117–167 g/L
B-Hb-1	121, 88–131 (5)	140, 94–147 (7)	.17
fB-WBC-0	6.1, 4.4–8.8 (6)	6.0, 3.3–7.5 (7)	.44	3.4–8.2 E9/L
fB-WBC-1	6.6, 5.3–8.9 (5)	7.4, 3.1–12.6 (7)	.90
P-CRP-0	5, 1–14 (6)	10, 2–32 (7)	.06	<10 mg/L
P-CRP-1	27, 25–69 (4)	3, 1–8 (7)	.007
P-AP-0	389, 256–1079 (6)	507, 240–957 (7)	.86	35–105 U/L
P-AP-1	530, 222–943 (4)	109, 75–173 (7)	.03
P-Bil-0	162, 21–243 (6)	109, 52–297 (7)	.77	<21 μmol/L
P-Bil-1	30, 13–260 (5)	14, 10–34 (7)	.06
S-CA 19-9-0	92, 6–48430 (5)	265, 26–1727 (6)	.86	<27 kU/L
S-CA 19-9-1	137, 87–10000 (3)	123, 13–8585 (6)	.44

B-Hb, blood hemoglobin; fB-WBC, fasting blood white blood cell; P-AP, plasma alkaline phosphatase; P-Bil, plasma bilirubin; P-CRP, plasma C-reactive protein; S-CA 19–9, serum carbohydrate antigen 19–9.

Discussion

In our trial, the patency of the studied ARS was significantly lower compared to that of the conventional PS, resulting in premature termination of the study. The sphincter of Oddi controls the flow between the common bile duct and duodenum preventing the retrograde flow of duodenal material into the biliary ducts. Transpapillary insertion of a stent destroys this natural barrier allowing intestinal reflux. Several studies have suggested that dietary fibers may play a significant role in stent clogging.^9,10 Dua et al. reported the results of a prospective, randomized study comparing this novel ARS with PS, and the results were promising with median patency of 145 days for ARS and 101 days for PS (P = .002).¹¹ To our knowledge, their study has the largest number of patients (n = 48); the median patency analysis included 35 patients as 7 PS patients and 6 ARS patients died of disease progression without evidence of biliary obstruction.¹¹

The results of our randomized trial using the same ARS were exactly opposite as the ARS was associated with a significantly shorter stent patency compared to a conventional PS. Our results are consistent with a recently published similar randomized study by Leong et al.¹² In their study with 16 patients, the antireflux biliary stent (n = 9) had significantly shorter patency time compared with plastic biliary stent (n = 7) in an unplanned interim analysis. All of the ARS were occluded within 30 days.¹²

The reason for early occlusions in the ARS group remains unclear. The antireflux mechanism consists of a windsock-shaped tubular valve (“sock” consisting of expanded tetrafluoroethylene attached to the duodenal end of the stent). In our study, the valves of the occluded ARSs were filled with sludge and often twisted 180°, obstructing the bile flow mechanically, similarly as described in the recently published study by Leong et al.¹² In the study of Dua et al.,¹¹ the valve was fully extended in those patients with repeat ERCP. In this study, ES was performed to all patients except one. However, ES has not been shown to deteriorate PS patency¹³ and our study results with conventional PSs are in accordance with earlier studies.^3,4 There were no differences in deployment of the ARSs between the studies.

A limitation of this study is that the occluded stents were not examined microscopically. Biofilm formation consisting of bacterial material is assumed to be the initial event in stent clogging.^14,15 We speculate that the polytetrafluoroethylene antireflux valve itself increases the friction and turbulence within the stent and thus possibly provokes the attachment of bacteria and other duodenal contents leading to stent occlusion. As the stent itself did not have any side holes, the malfunctioning of the valve resulted in stent occlusion. Good preliminary results have been achieved in a retrospective study with 23 patients having a novel antireflux metal stent with a silicone valve inserted (median duration of stent patency 14 months).¹⁶

Main limitation to this study is the small sample size, which is explained by the early termination of the study due to the ethical concerns. We acknowledge the fact that the unplanned interim analysis might attribute to type I error. Considering the study of Leong et al.¹² with similar results and conclusions, we strongly feel that the difference between these groups cannot be explained only by type I error. Even though the sample size was small, the patency of the ARSs was consistently and significantly shorter compared to the patency of the PSs. A strong element of this study is that this is a randomized, prospective study performed in a highly experienced ERCP unit, reducing the risk of variable expertise.

In conclusion, the patency of the studied ARS was significantly lower compared to that of the conventional PS, resulting in premature termination of the study. According to the results of our study, this ARS should not be used in clinical practice.

Footnotes

Acknowledgment

The writing of this article was supported by government research grant (EVO foundation) awarded to Turku University Hospital.

Authors' Contributions

All authors have participated in conception and design of the study, analysis and interpretation of the data, drafting of the article, critical revision of the article, and approved the final version of the article.

Disclosure Statement

No competing financial interests exist.

References

Andersen

, Sorensen

, Kruse

, et al. Randomised trial of endoscopic endoprosthesis versus operative bypass in malignant obstructive jaundice. Gut, 1989; 30:1132–1135.

Smith

, Dowsett

, Russell

, et al. Randomised trial of endoscopic stenting versus surgical bypass in malignant low bileduct obstruction. Lancet, 1994; 344:1655–1660.

Kaassis

, Boyer

, Dumas

, et al. Plastic or metal stents for malignant stricture of the common bile duct? Results of a randomized prospective study. Gastrointest Endosc, 2003; 57:178–182.

Tringali

, Mutignani

, Perri

, et al. A prospective, randomized multicenter trial comparing DoubleLayer and polyethylene stents for malignant distal common bile duct strictures. Endoscopy, 2003; 35:992–997.

Gronroos

, Gullichsen

, Laine

, et al. Endoscopic palliation of malignant obstructive jaundice in extremely elderly patients: Plastic stent is enough. Minim Invasive Ther Allied Technol, 2010; 19:122–124.

Katsinelos

, Paikos

, Kountouras

, et al. 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; 20:1587–1593.

Davids

, Groen

, Rauws

, et al. Randomised trial of self-expanding metal stents versus polyethylene stents for distal malignant biliary obstruction. Lancet, 1992; 340:1488–1492.

Prat

, Chapat

, Ducot

, et al. A randomized trial of endoscopic drainage methods for inoperable malignant strictures of the common bile duct. Gastrointest Endosc, 1998; 47:1–7.

van Berkel

, van Marle

, Groen

, et al. Mechanisms of biliary stent clogging: Confocal laser scanning and scanning electron microscopy. Endoscopy, 2005; 37:729–734.

10.

Weickert

, Venzke

, König

, et al. Why do bilioduodenal plastic stents become occluded? A clinical and pathological investigation on 100 consecutive patients. Endoscopy, 2001; 33:786–790.

11.

Dua

, Reddy

, Rao

, et al. Impact of reducing duodenobiliary reflux on biliary stent patency: An in vitro evaluation and a prospective randomized clinical trial that used a biliary stent with an antireflux valve. Gastrointest Endosc, 2007; 65:819–828.

12.

Leong

, Shen

, Au

, Luo

, Lau

, Wu

, Chan

, Sung

. A prospective, randomized study of the patency period of the plastic antireflux biliary stent: An interim analysis, Gastrointest Endosc, 2016; 83:387–393.

13.

Giorgio

, Luca

. Comparison of treatment outcomes between biliary plastic stent placements with and without endoscopic sphincterotomy for inoperable malignant common bile duct obstruction. World J Gastroenterol, 2004; 10:1212–1214.

14.

Leung

, Ling

, Kung

, et al. The role of bacteria in the blockage of biliary stents. Gastrointest Endosc, 1988; 34:19–22.

15.

Speer

, Cotton

, Rode

, et al. Biliary stent blockage with bacterial biofilm. A light and electron microscopy study. Ann Intern Med, 1988; 108:546–553.

16.

, Wang

, Shi

, et al. A novel antireflux metal stent for the palliation of biliary malignancies: A pilot feasibility study (with video). Gastrointest Endosc, 2011; 73:143–148.