Double Balloon Enteroscopy Procedure in Patients with Surgically Altered Bowel Anatomy: Analysis of a Large Prospectively Collected Database

Abstract

Introduction:

Patients with surgically altered bowel anatomy frequently undergo the double balloon enteroscopy (DBE) procedure to evaluate various small bowel diseases. There are limited data on the diagnostic yield and safety of DBE in these patients. We evaluated the diagnostic yield and complication rates of DBE in patients with surgical altered bowel anatomy. We also evaluated the success rate of DBE in achieving complete examination of the excluded segment of the small bowel and excluded stomach in these patients.

Subjects and Methods:

Our study was a single-center retrospective analysis of a large prospectively collected DBE database. Patients with a history of surgically altered bowel anatomy who had a DBE procedure performed between January 2006 and August 2011 were included in the study analysis. Patients' demographics, procedure indications, findings, endoscopic interventions, and postprocedural recovery data were recorded. We used frequency statistics to calculate the diagnostic yield and complication rates of DBE in these patients.

Results:

In total, 1215 DBEs were performed at our institution during the study period. Sixty-two patients with a history of altered bowel anatomy underwent 53 DBEs and 11 DBE-assisted endoscopic retrograde cholangiopancreatographies (ERCPs). The overall diagnostic yield of DBE was 61%, and that of DBE-assisted ERCP was 64%. No serious early or delayed DBE-associated complications were identified. In patients with surgically altered bowel anatomy containing excluded small bowel and excluded stomach, DBE success rate to achieve their complete examination was 92% (n=46).

Conclusions:

DBE including DBE-assisted ERCP is feasible, safe, and associated with reasonably high diagnostic yield in patients with surgically altered bowel anatomy.

Introduction

Patients with surgically altered bowel anatomy are frequently being referred for various endoscopy procedures. This may be due to a rise in application of bariatric surgery with a higher prevalence of morbid obesity, surgical interventions for incidental pancreatic lesions discovered during wide use of advanced imaging techniques, increasing numbers of liver transplants, etc.¹ Surgically altered bowel anatomy often excludes part of the gastrointestinal tract from conventional endoscopic access. Small bowel is difficult to visualize because of its length and tortuosity, especially in patients with surgically altered bowel anatomy.

Double balloon enteroscopy (DBE) is a device-assisted enteroscopy designed to visualize and treat various small bowel diseases. It has evolved rapidly since its introduction in 2001 to achieve complete visualization of the small bowel through its oral (antegrade) approach alone or combined oral (antegrade) and anal (retrograde) approach. This procedure appears as a “hand over hand” maneuver as the balloons are inflated and deflated alternately, pleating the bowel over the enteroscope and stepwise progression through small bowel.^2–4 It was designed to prevent the formation of redundant loops in the small intestine by using balloons that grip the intestinal wall, allowing the endoscope to achieve deeper intubation and easier negotiation through acute turns of the small bowel.⁵

Some common indications for DBE are gastrointestinal bleeding, abdominal pain, foreign body retrieval, chronic diarrhea, evaluation of small bowel lesions, etc. Patients with surgically altered bowel anatomy such as Roux-en-Y reconstruction are at increased risk for anastomotic stricture, gastrointestinal bleeding, small bowel perforation, etc. DBE allows the performance of multiple endoscopic interventions such as polypectomy, hemostasis with electrocauterization or hemoclip, confirmation of small bowel tumors, small bowel stricture dilation, etc.

Because of formation of the bilioenteric anastomosis there is a higher risk of biliary complications such as gallstones due to rapid weight loss, ascending cholangitis, bilioenteric anastomosis stricture, etc.¹ Surgically altered bowel anatomy prevents utilization of a standard side-viewing endoscopic retrograde cholangiopancreatography (ERCP) endoscope. ERCP performed with a conventional duodenoscope or pediatric colonoscope is associated with a high failure rate due to various technical difficulties.⁶ Instead, DBE-assisted ERCP is increasingly used in patients with surgically altered bowel anatomy because of ease of negotiating sharp turns of the small bowel.⁶ DBE-assisted ERCP allows various biliary and pancreatic interventions such as endoprosthesis, stone extraction, etc., in patients with surgically altered bowel anatomy who otherwise may require surgical intervention.^3,7

As of now, besides balloon-assisted enteroscopy such as DBE, spiral enteroscopy has also been used to examine the small bowel. Compared with DBE, the spiral enteroscopy system was found to be faster, but its depth of insertion in the small bowel was much less, with limited visualization of the small bowel.⁸ DBE was found to have either similar or higher diagnostic yield compared with spiral enteroscopy during evaluation of the small bowel.^9,10 Multiple studies have compared capsule endoscopy and DBE, showing a similar diagnostic yield.^11–13 DBE is more advantageous over capsule endoscopy by allowing air insufflation, tissue rinsing, and directed tissue examination during transit and because of its ability to perform various therapeutic interventions.^4,14 In patients with surgically altered anatomy, capsule endoscopy is even less advantageous because of higher risk of retention and the inability to examine excluded small bowel limbs.³ Push enteroscopy is available as an alternative to DBE but may be difficult because of the force required to advance the instrument through the tortuous small intestine and carries a limited depth of insertion. Intraoperative enteroscopy is also used and is most reliable for total small bowel visualization. However, this process requires more time, cost, and staff effort. It is more invasive and has a substantial risk of complications, including mortality.^1,5,8

The diagnostic yield of DBE in surgically altered anatomy seems to be similar to the diagnostic yield of all DBEs.¹⁵ Studies examining DBE in patients with surgically altered anatomy have been small in size, and complication rates have been reported as high as 7%.^3,15 Scarce data have been reported regarding the risk of early or delayed complications of DBE in patients with surgically altered anatomy.^3,8

Therefore, the goal of our study was to document diagnostic yield and complication rates of DBE procedures in a cohort of patients with surgical altered bowel anatomy from a large prospectively collected endoscopy database. We also evaluated the success rate of DBE in achieving adequate examination of the bypass segment of the small bowel and/or excluded stomach in patients with surgically altered bowel anatomy.

Patients and Methods

Patients

Our study was a single-center retrospective analysis of a prospectively collected database of a cohort of patients who underwent DBE at our institution, the Mayo Clinic in Florida (Jacksonville). Electronic medical records of patients who underwent DBE between January 2006 and August 2011 were reviewed. Only patients with a history of surgical alteration of bowel anatomy were included in our study analysis. We obtained information such as patients' demographics, procedure indications, procedure findings, endoscopic interventions, and postprocedural recovery data and hospitalization records. DBE procedure-related details such as DBE route, procedure time, diagnostic yield, and therapeutic intervention were also recorded.

We analyzed the data using frequency statistics to calculate the diagnostic yield and complication rates of the DBE in these patients. Diagnostic yield was defined as the number of DBE procedures successfully able to identify the lesion/pathology to explain the pertinent indication of the procedure. Serious complications were defined as death, cardiac arrhythmias, ischemic cardiac events, need of cardiorespiratory resuscitation, significant respiratory distress, pneumonia, visceral perforation, or significant bleeding requiring hospitalization. We recorded complications during procedure as well as early (within 24 hours) postprocedure and delayed (after 24 hours) postprocedure. The study was approved by the Institutional Review Board of the Mayo Clinic in Florida.

DBE procedure

All DBEs were scheduled with an open access system in which the endoscopist performing the DBE screens the requests and approves procedure scheduling based on appropriateness of the procedure indication without having seen the patient. The decision of whether to perform the DBE through an oral and/or anal approach was dependent on the endoscopist's choice per the patient's clinical presentation and prior endoscopy and radiology findings. All DBEs were performed after appropriate informed consents were obtained. Immediately prior to the procedure, the endoscopist and the anesthesiologist obtained the clinical history and physical examination results. The decision of whether to proceed or to cancel the scheduled procedure was dependent on mutual agreement between the endoscopist and the anesthesiologist.

DBE was performed by two trained, expert gastroenterologists (M.E.S. and F.J.L.). The EN450-T5 enteroscope (Fujinon Inc., Wayne, NJ) was used for all DBEs. It includes a 230-cm-long endoscope, an overtube, and a double-barostatic pump (PB-10 balloon pump controller) that allows controlled inflation of the endoscope and overtube balloons. All DBEs were performed in a standard endoscopy unit with a portable fluoroscopy unit, and technicians assisting the process performed inflation–deflation of the balloons. Patients were required to have fasted for at least 6 hours prior to all endoscopy procedures. Sedation with general anesthesia or modified anesthesia care was used during all procedures. The elective placement of endotracheal intubation was dependent on the individual preference of the anesthesiologist and the endoscopist. During the DBE, vital signs, oxygen saturation, and end-tidal CO₂ were monitored by the anesthesiology team. The endoscopist performed therapeutic procedures as deemed necessary.

Results

Baseline characteristics

In total, 1218 DBE procedures were performed during our study period. Of these, 64 DBEs including 11 DBE-assisted ERCPs were performed in 62 patients (73% female) with surgically altered bowel anatomy. Patients' mean age was 51 years (range, 26–77 years), and the mean body mass index was 28.2 kg/m² (range, 20.3–53.6 kg/m²).

Among the different types of altered bowel anatomies, bariatric gastric bypass surgery was the most common, present in 49 (79%) patients. Other types of altered bowel anatomies were non-bariatric Roux-en-Y reconstruction surgery in 9 (15%) patients, non–pyloric-preserving Whipple surgery in 2 (3%) patients, and Billroth II gastrojejunal surgery in 2 (3%) patients.

DBE without ERCP efficacy

In total, 57 DBEs without ERCP were performed among 50 patients. Of these, 53 (93%) DBEs were performed with an oral (antegrade) approach, and 4 (7%) DBEs were performed with an anal (retrograde) approach. The most common indication of DBE was abdominal pain. The other indications for DBE procedure are shown in Table 1.

Table 1.

Indications for Double Ballon Enteroscopy (Without Endoscopic Retrograde Cholangiopancreatography) in the Study Patients

Procedure, indication	Number (%)
Upper DBE (n)	53
Abdominal pain	16 (28%)
Overt GI bleeding	11 (19%)
Obscure GI bleeding	8 (14%)
Nausea, vomiting	7 (12%)
Abnormal imaging studies (CT or MRI of abdomen)	5 (9%)
Diarrhea	3 (5%)
Abnormal capsule endoscopy	2 (3.5%)
Small bowel adenoma surveillance	1 (2%)
Lower DBE (n)	4
Overt GI bleeding	2 (3.5%)
Obscure GI bleeding	1 (2%)
Abdominal pain	1 (2%)

CT, computed tomography; DBE, double ballon enteroscopy; GI, gastrointestinal; MRI, magnetic resonance imaging.

The overall DBE success rate for complete examination of biliopancreatic (excluded) limb and/or excluded stomach examination was achieved in 46 (92%) patients. Among the remaining 4 (8%) patients, DBE failed to reach the biliopancreatic limb anastomosis because of adhesions (n=3) or enteric stricture (n=1). Length of the alimentary limb up to the biliopancreatic limb anastomosis from above was found to be approximately 81 cm (range, 40–220 cm) during antegrade DBE. The average biliopancreatic limb length was 58 cm (range, 30–120 cm). During retrograde DBE, an average of 76 cm (range, 50–110 cm) length of small bowel was examined beyond the ileocecal valve. Overall, an average DBE procedure time was 89 minutes (range, 38–180 minutes), and average fluoroscopy exposure time was 164 seconds (range, 15–528 seconds).

The overall diagnostic yield of all DBEs without ERCP was found to be 61% (35 of 57 patients). Positive findings of 35 DBEs are shown in Table 2. The most common abnormal finding was excluded (bypassed) stomach ulceration in 11 (31%) patients. In patients with clinically suspected small bowel bacterial overgrowth (n=6), small bowel aspirate obtained during DBE confirmed its presence in all of them. Small bowel anastomotic ulceration was found in 5 (14%) patients, whereas distal small bowel (common channel) ulceration was found in 4 (11%) patients. Four patients had retained foreign bodies. Of these, 2 patients had retained a gastrostomy tube T fastener in the excluded stomach, 1 patient had a retained pancreatic duct stent, and 1 patient had a retained 4-cm metal suture with surrounding ulceration. During DBE, 2 patients were suspected to have an internal fistula arising from the stomach, and 1 patient was suspected to have a fistula located at the biliopancreatic limb anastomosis site.

Table 2.

Findings of Double Ballon Enteroscopy with Positive Diagnostic Yield (n=35)

Finding	Number (%)
Excluded stomach significant erosions/ulceration	11 (31%)
Small bowel aspirate positive for small bowel bacterial overgrowth	6 (17%)
Small bowel anastomotic ulcerations	5 (14%)
Retained foreign body	4 (11%)
Distal common channel small bowel ulcerations	4 (11%)
Gastric/enteral fistula	3 (9%)
Excluded stomach contained perforation	1 (3%)
Roux limb tight stricture	1 (3%)

The diagnostic yield of DBE in patients with prior negative imaging studies and/or capsule endoscopy was found to be 73% (11 of 15 patients). The diagnostic yield of small bowel biopsy (targeted or random) was 9.4% (3 of 32 patients). Although most of the small biopsy results (targeted or random) showed nonspecific inflammatory changes, 1 patient was found to be positive for Helicobacter pylori infection in targeted biopsies of gastric ulcer, 1 patient was found to have small bowel malabsorption changes, and 1 patient was diagnosed with Crohn's disease. The diagnostic yield of small bowel aspirate in clinically suspected cases for small bowel bacterial overgrowth was found to be 100% (n=6).

DBE with ERCP efficacy

Among the total 11 DBE-assisted ERCPs, the most common indication was acute cholangitis with choledocholithiasis in 3 (27%) patients and common bile duct stricture in 3 (27%) patients. Two (18%) patients had recurrent acute pancreatitis, 1 (9%) patient had common bile duct dilation in the setting of recurrent abdominal pain, 1 (9%) patient had pancreatic duct dilation in the setting of recurrent abdominal pain, and 1 (9%) patient was found to have pancreatic pseudocyst with possible pancreatic duct disruption on imaging study.

The success rate of DBE-assisted ERCP with adequate examination of the pancreatobiliary tree pertinent to the indication and required therapeutic intervention was 64% (7 of 11 patients). Among patients with adequate pancreatobiliary examination, biliary stricture was confirmed in 4 (57%) patients, a common bile duct stone was retrieved in 1 (14%) patient, common bile duct stricture was ruled out in 1 (14%) patient, and pancreatic duct disruption was ruled out in 1 (14%) patient. Because of a significant stricture in the pancreatobiliary limb, ERCP evaluation was not achieved in 1 patient. Also, biliary duct cannulation was unsuccessful in 2 patients, and pancreatic duct cannulation was unsuccessful in 1 patient.

DBE safety

One patient was found to have contained perforation in the excluded stomach. Upon review of the patient's prior imaging studies done at an outside facility, the patient was discovered to have the changes of contained perforation prior to DBE. In regard to DBE complications, no serious early (within 24 hours) or delayed (after 24 hours) complications were identified after any DBE with a mean duration of 6 months of follow-up records.

Discussion

We obtained diagnostic yield of DBE among a specific patient population with a history of surgically altered bowel anatomy from our large prospectively collected database. In our study, we have found a 61% diagnostic yield of DBE in patients with surgically altered bowel anatomy, which is somewhat comparable to the 47%–68% in previous reports of diagnostic yield of DBE with normal bowel anatomy.^10,16 Because of development of adhesions and postsurgical changes, evaluation of the small bowel with DBE becomes challenging after altered bowel anatomy surgery. However, complete evaluation of the excluded segment of small bowel and stomach was possible achieved in 92% of the patients in our study. It is important to note that diagnostic yield may depend on the procedural indications. In prior case series and reports, gastrointestinal bleeding was the most common indication for patients undergoing DBE.^1,16,17 In contrast, abdominal pain was found to be the most common indication for DBE in patients with surgically altered bowel anatomy. The diagnostic yield has shown to be higher in patients with gastrointestinal bleeding and lower in patients with a single indication such as abdominal pain or diarrhea.¹⁸ We have found the diagnostic yield was comparable in patients with surgically altered bowel anatomy even with abdominal pain being the commonest indication for DBE in our study cohort. There are no known reports on diagnostic yield of tissue sampling (random or targeted) or small bowel aspirate obtained during DBE in surgically altered bowel anatomy. Such yield is largely dependent on the indication of the procedure and the endoscopist's decision on obtaining small bowel biopsies or aspirate. Most of the small biopsy results of targeted lesions showed nonspecific inflammatory changes. Our study reports relatively low yield (9.4%) of biopsies (random or targeted) and very high yield (100%) of small bowel bacterial overgrowth.

The ERCP procedure is known to be more difficult in surgically altered anatomy compared with normal anatomy. This is due to the variable orientation of the biliary and pancreatic opening after different surgeries altering normal bowel anatomy, the technical difficulty using a side-viewing standard ERCP endoscope, the limited diameter (2.8 mm) of the double balloon enteroscope's accessory channel, the long length of the enteroscope, and the absence of appropriately modified ERCP accessories such as a longer through the scope pancreatobiliary stent delivery system. In prior reports, the ERCP success rate was found to be more than 90% in normal bowel anatomy compared with little more than 50% in surgically altered bowel anatomy.^19,20 Several case series have reported the diagnostic yield of DBE-ERCP to be around 60%–90%.^21–23 In our study, other than cases with enteral strictures, biliopancreatic duct opening was reached in all the rest of the DBE-assisted ERCP procedures. The diagnostic yield of DBE-assisted ERCP was found to be 64%, which is comparable to previously reported study results.^21–23 Significant variability in diagnostic yield is dependent on types of altered bowel anatomy and the indication for ERCP, such as biliary versus pancreatic indication. The success rate of DBE-ERCP was much lower for pancreatic indication compared with biliary indication.¹⁹ In our study, we did not have a sufficient number of patients with pancreatic indication to arrive at a such conclusion.

DBE does come with more inherent risks than conventional endoscopic procedures. These include the position deep inside the small bowel with limited flexibility because the small bowel is pleated over the overtube and that it is more time consuming and more invasive than conventional endoscopic procedures.^8,24 A longer procedure time increases the anesthesia time and the endoscopist's fatigue. In addition, some studies of patients with surgically altered anatomy have suggested that the presence of adhesions and fragile areas of anastomosis may make perforation a greater risk.^1,14 DBE complication rates in studies of patients with normal anatomy have varied and range from 0% to 7%, with most citing a range up to 1.5%.^1,8,14,16,22 Another large study of nine U.S. centers revealed a DBE complication rate of 0.9%, but the DBE complications rate increased to 3% in those patients with surgically altered anatomy.¹ In our study, we did not experience any early or delayed serious complication in the cohort of patients with altered bowel anatomy undergoing DBE. It would account for referral bias from the type of patient population, an experienced endoscopist, and careful management of anesthesia during DBE. Variability in complication rates could be indicative of variables such as recently created anastomosis, length and complexity of afferent limbs, patient's health status, need for ERCP therapeutics, and many other factors. Overall, DBE in patients with altered bowel anatomy is a relatively safe procedure with a less than 1% complication rate in those experienced with the procedure.

Compared with various conventional endoscopy procedures, DBE is less common because of its limited availability. On top of that, DBE in specific patient populations such as those with altered bowel anatomy is more uncommon. The number of patients with altered bowel anatomy undergoing DBE in our study is reasonably high compared with its utilization status. We gave an accurate report of serious complication rates associated with DBE in a prospective fashion. There are some limitations to our study, including that it is a single-center analysis with no comparison group. In addition, DBEs were performed by two experienced endoscopists, limiting the reproducibility of the study results. Our study patients were predominantly white and female. It can be explained by referral bias, and higher number of white women undergoing bariatric surgeries.

Device-assisted enteroscopy such as DBE is the gold standard endoscopic procedure for the exploration of the small bowel, allowing complete evaluation and therapeutic interventions of various small bowel diseases.⁷ In our study, we demonstrated that DBE is feasible and safe and has a reasonably high diagnostic yield in patients with surgically altered bowel anatomy. It is effective in diagnostic and therapeutic procedures including ERCP that would otherwise require surgery in this class of patients. Drawbacks to the DBE procedures are invasiveness, prolonged procedure time, and less frequent but known endoscopy-associated risks. There is a need to evaluate DBE procedure techniques, especially DBE-assisted ERCP, to identify the factors affecting the diagnostic yield in surgically altered anatomy. Although demand for DBE is expanding because of emerging new indications,¹⁴ it is important to evaluate the role of the endoscopist's training for complex DBE, technical limitation of the DBE endoscope (increasing accessory channel, incorporating a side-vision optical system, adding an elevator system), and DBE accessories for ERCP (longer pancreatobiliary stent delivery system) on the diagnostic yield of DBE in patients with surgically altered bowel anatomy. The optimal role and cost-effectiveness of DBE in various small bowel pathologies, especially in patients with surgically altered bowel anatomy, need to be determined further by prospective studies.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Gerson

et al. Complications associated with double balloon enteroscopy at nine US centers. Clin Gastroenterol Hepatol, 2009; 7:1177–1182–1182.e1–3.

Akerman

, Cantero

. Spiral enteroscopy and push enteroscopy. Gastrointest Endosc Clin North Am, 2009; 19:357–369.

Moreels

, Hubens

, Ysebaert

, Op de Beeck

, Pelckmans

. Diagnostic and therapeutic double-balloon enteroscopy after small bowel Roux-en-Y reconstructive surgery. Digestion, 2009; 80:141–147.

Teshima

, May

. Small bowel enteroscopy. Can J Gastroenterol, 2012; 26:269–275.

Yamamoto

et al. Total enteroscopy with a nonsurgical steerable double-balloon method. Gastrointest Endosc, 2001; 53:216–220.

Raithel

et al. Double-balloon-enteroscopy-based endoscopic retrograde cholangiopancreatography in post-surgical patients. World J Gastroenterol, 2011; 17:2302–2314.

Moreels

. Altered anatomy: Enteroscopy and ERCP procedure. Best Pract Res Clin Gastroenterol, 2012; 26:347–357.

May

, Manner

, Aschmoneit

, Ell

. Prospective, cross-over, single-center trial comparing oral double-balloon enteroscopy and oral spiral enteroscopy in patients with suspected small-bowel vascular malformations. Endoscopy, 2011; 43:477–483.

Kogure

et al. Spiral enteroscopy for therapeutic ERCP in patients with surgically altered anatomy: Actual technique and review of the literature. J Hepatobiliary Pancreat Sci, 2011; 18:375–379.

10.

Frieling

, Heise

, Sassenrath

, Hulsdonk

, Kreysel

. Prospective comparison between double-balloon enteroscopy and spiral enteroscopy. Endoscopy, 2010; 42:885–888.

11.

Akahoshi

et al. Double-balloon endoscopy in the diagnosis and management of GI tract diseases: Methodology, indications, safety, and clinical impact. World J Gastroenterol, 2006; 12:7654–7659.

12.

Sidhu

et al. Evaluating the role of small-bowel endoscopy in clinical practice: The largest single-centre experience. Eur J Gastroenterol Hepatol, 2012; 24:513–519.

13.

Shimatani

et al. Effective "short" double-balloon enteroscope for diagnostic and therapeutic ERCP in patients with altered gastrointestinal anatomy: A large case series. Endoscopy, 2009; 41:849–854.

14.

Sunada

, Yamamoto

. Double-balloon endoscopy: Past, present, and future. J Gastroenterol, 2009; 44:1–12.

15.

Kim

et al. Usefulness of double balloon endoscopy in patients with surgically distorted intestinal anatomy. J Clin Gastroenterol, 2009; 43:737–742.

16.

Xin

, Liao

, Jiang

, Li

. Indications, detectability, positive findings, total enteroscopy, and complications of diagnostic double-balloon endoscopy: A systematic review of data over the first decade of use. Gastrointest Endosc, 2011; 74:563–570.

17.

May

, Nachbar

, Ell

. Double-balloon enteroscopy (push-and-pull enteroscopy) of the small bowel: Feasibility and diagnostic and therapeutic yield in patients with suspected small bowel disease. Gastrointest Endosc, 2005; 62:62–70.

18.

Moschler

, May

, Muller

, Ell

. Complications in and performance of double-balloon enteroscopy (DBE): Results from a large prospective DBE database in Germany. Endoscopy, 2011; 43:484–489.

19.

Chahal

et al. Endoscopic retrograde cholangiopancreatography in post-Whipple patients. Endoscopy, 2006; 38:1241–1245.

20.

Farrell

et al. Endoscopic retrograde cholangiopancreatography after pancreaticoduodenectomy for benign and malignant disease: Indications and technical outcomes. Endoscopy, 2006; 38:1246–1249.

21.

Iwamoto

et al. Double balloon endoscope facilitates endoscopic retrograde cholangiopancreatography in Roux-en-Y anastomosis patients. Dig Endosc, 2010; 22:64–68.

22.

Itoi

et al. Long- and short-type double-balloon enteroscopy-assisted therapeutic ERCP for intact papilla in patients with a Roux-en-Y anastomosis. Surg Endosc, 2011; 25:713–721.

23.

Aabakken

, Bretthauer

, Line

. Double-balloon enteroscopy for endoscopic retrograde cholangiography in patients with a Roux-en-Y anastomosis. Endoscopy, 2007; 39:1068–1071.

24.

Mensink

et al. Complications of double balloon enteroscopy: A multicenter survey. Endoscopy, 2007; 39:613–615.