Double-balloon enteroscopy: ready for prime time? Since its introduction in 2003, double-balloon enteroscopy (DBE) has enabled endoscopic therapy in an area of the small intestine previously accessible only with the aid of laparotomy. Widespread use of this technology, however, has been limited by several technical and logistical hurdles. In this issue of Gastrointestinal Endoscopy, Gross and Stark1 report 200 consecutive DBEs performed in 137 patients by a single endoscopist. The largest single-center U.S. experience published to date, their study illustrates the promise of DBE as it grows beyond the academic setting and into everyday practice. It also addresses some of the fundamental questions that we must consider in evaluating the future of DBE in the United States: d Can it be done? (technical feasibility) d What are the benefits? (diagnostic yield and therapeutic impact) d What are the drawbacks? (complication rate and resource utilization) d What are the alternatives? (competing technologies)
However, this goal is not always achieved. Among the 45 patients in the current series in whom total enteroscopy was attempted, only 9 attempts (20%) were successful. The total enteroscopy rate of 86% initially reported by Yamamoto et al4 has not been replicated in subsequent studies. A large multicenter international study found a total enteroscopy rate of 15%, with the highest of any individual center being 41%,5 although it is unclear whether these numbers include patients for whom total enteroscopy was not attempted. May et al6 reported a slightly higher rate of 45%. The generally lower total enteroscopy rates in Western
The usefulness of double-balloon endoscopy in patients with obscure GI bleeding is unquestionable. When performed predominantly for this indication, its diagnostic yield is 75%, with remarkable agreement among multiple studies from around the world.
TECHNICAL FEASIBILITY Although the technical feasibility of DBE is well established, it remains a challenging procedure even for experienced endoscopists. The anal approach can be particularly difficult because of slippage of the scope during the first reduction maneuver inside the terminal ileum. In a previous study on the learning curve of DBE, procedure time for the oral approach decreased significantly after the first 10 DBEs, but the retrograde failure rate remained at 20%.2 This rate does, however, appear to decrease after 20 procedures.3 The current study by Gross and Stark reports excellent technical outcomes: the terminal ileum was intubated in all cases and mean insertion depths from the oral and anal approach were 220 and 124 cm, respectively. However, the procedure time did not decrease even after 150 procedures. These findings suggest that, although insertion depth and therapeutic yield can increase with experience, DBE remains an inherently time-consuming procedure. In theory, DBE from both the oral and anal approach allows for a complete examination of the small intestine.
Copyright ª 2008 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2007.10.033
countries compared with Japan may be partly due to a higher prevalence of obesity and adhesions from prior abdominal surgeries. These factors hinder pleating of the small bowel along the overtube, which is required for deep insertion of the enteroscope. However, experience and technical skill may also play important roles: the success rate of complete enteroscopy reported by Gross and Stark in the current series increased from 8% to 63% after the first 150 cases, as did the proportion of procedures deemed to be clinically helpful. It is also important to note that in most cases, complete enteroscopy is not required for diagnostic or therapeutic success. For example, Kaffes et al7 limited insertion time to 60 minutes and had a 0% total enteroscopy rate, yet still reported 80% cessation of bleeding after a mean follow-up of 10 months. Given the challenging nature of DBE, the optimal training method is unknown. Perez-Cuadrado et al8 have described a comprehensive training program that progresses from animal cases to supervised human procedures and culminates with a computerized simulator-based skills assessment. However, this type of hands-on mentorship is impractical for most physicians, and the current study demonstrates that DBE can be performed competently after training on an ex vivo porcine model and observation of
898 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 6 : 2008
human cases. This type of teaching could feasibly be carried out at training workshops, without the need for on-site mentorship. Future generations of gastroenterologists may be exposed to DBE as part of their training programs, but given the complexity of the procedure and the low volume of cases seen at any one institution, DBE may be best taught as part of an advanced endoscopy year. In summary, DBE is a challenging but learnable procedure. Technical expertise, as measured by insertion depth, therapeutic yield, and success rate of total enteroscopy appears to increase with experience, but the procedure remains lengthy. On-site mentorship is probably not required for training, but given the relatively small number of patients needing the procedure, DBE should be performed by a limited number of endoscopists with an adequate patient volume.
tients had OGIB). Perhaps more important than ‘‘diagnostic yield,’’ however, is whether the findings at DBE are helpful in the care of a patient. For example, a negative examination in a patient with unexplained anemia might adequately rule out GI blood loss and allow the clinician to focus elsewhere. Conversely, finding and treating a single small arteriovenous malformation (AVM) in a patient with recurrent massive hemorrhage and small bowel obstructions is unlikely to fix the problem. In the current series, Gross and Stark report that 71.5% of procedures were felt to be helpful, with an additional 11.7% of procedures deemed possibly helpful. However, an accurate determination of whether a procedure is helpful requires long-term follow-up.
THERAPEUTIC IMPACT Although no long-term studies on the clinical impact of DBE are available, the results of short-term follow-up studies are encouraging. Kaffes et al7 found that 80% of patients with OGIB who underwent DBE had cessation of bleeding after a mean follow-up of 10 months. A larger study with a mean follow-up of 16 months showed an 89% nonrebreeding rate among 96 patients, but 56 patients who had no findings at DBE or were lost to follow-up were excluded from the analysis.11 Because the natural history of AVMs is not well defined, it is unknown whether bleeding in these patients will recur after several years, and if so, at what interval repeat procedures will be required. Similarly, it is unknown at what interval patients with Peutz-Jegher’s polyposis need DBE to prevent small bowel obstruction. The therapeutic impact of DBE in patients with strictures requiring dilation, ulcerating lesions requiring biopsy, or Roux-en-Y anatomy requiring examination of the biliary ducts is immediate, but such indications account for a minority of cases. The major impact of DBE on the practice of gastroenterology will undoubtedly be in the treatment of obscure GI bleeding.
Because of the lack of an easily reproducible gold standard for small bowel pathology, ‘‘diagnostic yield’’ is often used as a surrogate for the sensitivity of small bowel examination techniques. DBE is itself likely to become the new gold standard, especially when total enteroscopy can be achieved. Even intraoperative enteroscopy and autopsy are not necessarily superior to DBE because vascular lesions may disappear with diminished perfusion to the small bowel, and trauma during intraoperative enteroscopy may be mistaken for vascular lesions. Although capsule endoscopy examines the entire small bowel more often than DBE, lesions can be missed as a result of the uncontrollable speed and orientation of the capsule. In the current study, Gross and Stark found that 3 of 4 patients with a ‘‘negative’’ capsule endoscopy had significant lesions found on DBE, although this likely represents a biased sample because only patients with continuing symptoms underwent DBE. The usefulness of DBE in patients with obscure GI bleeding (OGIB) is unquestionable. When DBE is performed predominantly for this indication, the diagnostic yield is 75%, with remarkable agreement among multiple studies from around the world.7,9-12 However, the findings themselves vary considerably, with vascular lesions being more common in Western countries and tumors being reported more often in Asia. This may reflect an underlying difference in the prevalence of these lesions. Other well-accepted indications for DBE include removal of polyps in Peutz-Jegher’s syndrome and biopsy of ulcerating lesions to distinguish lymphoma from Crohn’s disease or refractory celiac disease. Chronic diarrhea and chronic abdominal pain are also sometimes addressed with DBE, but the diagnostic yield falls to 30% to 40% under these indications.13 Although this yield is not insignificant, DBE must be used judiciously in such cases given the resources required. In the current series, the diagnostic yield of DBE was 80%, reflecting appropriate patient selection (74% of pa-
As shown in the current study, DBE is a safe procedure. The most common adverse effect from DBE is minor abdominal pain from gas in the small bowel. This can be ameliorated by insufflation with carbon dioxide rather than air because carbon dioxide is absorbed more rapidly through the intestinal wall. The reduced distention seen with carbon dioxide may also facilitate stricture dilation14 and increase insertion depth by allowing for better pleating of the small bowel. Bleeding and perforation are uncommon in DBE. The largest study of DBE-related complications to date, encompassing more than 2300 procedures at 10 high-volume centers, reported a major complication rate of 8 per 1000 procedures.5 This is comparable to the complication rate seen with other endoscopic procedures. Risk factors for
Volume 67, No. 6 : 2008 GASTROINTESTINAL ENDOSCOPY 899
perforation include recently created anastomoses2 and ulceration caused by lymphoma10; it is also felt that EhlersDanlos syndrome and small bowel diverticulosis may increase the risk associated with DBE,15 but this has not been reported. In cases where polypectomy, hemostatis, or dilation is performed, the complication rate of DBE is higher (4.3%), and this needs to be considered when deciding between endoscopic therapy and surgical intervention for small bowel findings. Another well-described complication of DBE is pancreatitis. The mechanism for this may either be papillary edema from repeated passage of the overtube or direct pressure on the pancreas. The rate of pancreatitis in the large series by Mensink et al5 was low (0.3%), but only patients who required 2 or more days of hospitalization were included. Asymptomatic hyperamylasemia may occur in as many as 46% of DBE cases,16 but the significance of this is unclear because a minor rise in amylase can be seen with small bowel injury. The estimate by May et al6 of a 1% incidence of pancreatitis is probably the most realistic. All the above risks need to be considered when a patient is evaluated for DBE. It must be kept in mind, however, that alternate procedures such as intraoperative enteroscopy or heparin-provocation angiograms may confer even greater risk with no added benefit.
overtube or enteroscope while the other instrument was manipulated. DBE has also been performed successfully without fluoroscopy; it is likely that as endoscopists become more adept at recognizing looping and slippage in the small bowel, most procedures will be done in this manner. In the current study, general anesthesia with endotracheal intubation was used for the oral approach because of the perceived risk of aspiration. This is probably not necessary because thousands of procedures have been safely performed with the patient under moderate (‘‘conscious’’) sedation. The single reported episode of aspiration from vomiting was during a retrograde procedure.15 The use of moderate sedation rather than general anesthesia does not appear to affect procedure time,2 but it may shorten the mean recovery time of 96 minutes seen in the current study. Where available, propofol may also be a useful anesthetic for DBE. In the United States, almost all DBEs are performed on an outpatient basis, and this is a safe practice. Although 87% of cases in the large international series by Mensink et al5 were inpatients, hospitalization is generally not required either before or after DBE.
DBE is a resource-intensive procedure. In the current study, the mean procedure time was 101 minutes for oral DBE and 96 minutes for anal DBE. The total room use time for DBE is even longer, given the need to assemble the endoscope balloon immediately before the procedure. The long duration of DBE is largely due to the lack of a clear end point; the procedure can continue for several hours, especially when complete enteroscopy is desired. However, most endoscopists performing DBE have noticed diminishing returns after the first hour, and some have set arbitrary end points for insertion time with good outcomes. In the future, the procedure time will likely become more predictable as endoscopists gain a better sense of the point at which further attempts at advancement are futile. However, DBE will remain time consuming compared with other endoscopies. This is to be expected, given the length of intestine that is examined and the fact that most patients undergoing DBE have had easily accessible lesions excluded by multiple previous procedures. Other elements of resource utilization include staffing, use of fluoroscopy, and mode of sedation. Many early DBE reports described a 2-physician technique done with fluoroscopic guidance with the patient under general anesthesia. This is in some ways reminiscent of the early days of colonoscopy. DBE has already shown signs of becoming simpler and more streamlined. In the current study, only one gastroenterologist was needed, with a technician holding the
Before DBE, push enteroscopy (PE) was a standard procedure for evaluating patients with OGIB. Although technically simpler and less time consuming than DBE, PE allows for examination of only 60 to 100 cm of small bowel beyond the pylorus compared with the 220 cm examined in the current DBE study. In many institutions where DBE is available, it has replaced PE as the initial endoscopic test in patients with OGIB. Intraoperative enteroscopy, long considered the gold standard for examining the small bowel, has also in many places been replaced by DBE because of a comparable diagnostic yield at a much lower cost and risk to the patient. Capsule endoscopy is likely to remain an important part of the algorithm for OGIB, mostly as a complementary procedure to DBE. The capsule is more widely available (at least in the United States), safer, and better tolerated than DBE. Capsule endoscopy also examines the entire small bowel more often than bidirectional DBE. However, as a purely diagnostic modality, it should be used only if its findings are likely to change subsequent management, either by allowing one to choose the optimal route of subsequent exploration (eg, oral DBE for proximal lesions vs rectal DBE for distal lesions)17 or by obviating the need for further investigation if the capsule is negative. Patients with OGIB and negative findings on capsule endoscopy have only a 5.6% chance of rebleeding after at least 12 months of follow-up.18 The consensus report from the First International Workshop on Double Balloon Endoscopy presents an algorithm for evaluating patients with OGIB. It suggests proceeding directly to oral DBE in patients with continuing overt
900 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 6 : 2008
obscure bleeding. For other patients, it suggests initial capsule endoscopy, with positive results prompting a DBE (direction determined by location of the lesion) and negative results prompting observation only. In the latter group of patients, a repeat capusule endoscopy could be done at some interval to exclude an intermittently bleeding lesion.19 It might also be reasonable to proceed directly to DBE when AVMs have been detected and treated on upper endoscopy and colonoscopy but bleeding continues. DBE was invented by Dr Yamamoto and developed in conjunction with the Fujinon Corporation. Olympus has recently released a single-balloon enteroscope, which consists of a similar balloon-tipped overtube as the Fujinon system but lacks the second balloon on the endoscope tip. Slippage of the enteroscope during manipulation of the overtube is avoided by hooking the scope within the intestine, similar to a reduction maneuver during colonoscopy. Given that the larger overtube balloon is responsible for most of the pleating of the small bowel, this system may allow for a similar depth of insertion as DBE with a simpler technique. However, the absence of haustra in the small bowel may make holding the scope position difficult. Further studies are required to evaluate the diagnostic and therapeutic yield of single-balloon enteroscopy compared with DBE. If it the two are found to be comparable, then the term ‘‘push-and-pull enteroscopy’’ or ‘‘balloon-assisted enteroscopy’’ might be appropriate to designate the procedure, with double-balloon and single-balloon enteroscopes being 2 instruments that can be used for the procedure. DBE is revolutionizing the way we deliver therapy to the small intestine, and its use in the United Sates will hopefully continue to grow. Despite the technical challenges of the procedure and the resources required, it is clearly the preferred method for treating bleeding lesions in the small bowel and obtaining tissue for definitive diagnosis. It has also been used safely and effectively for dilation of strictures, retrieval of impacted capsules, and a variety of other indications. As DBE moves beyond the academic setting, it is likely that the procedure will become more streamlined and predictable, but it should still be performed only by endoscopists with adequate training and patient volume to maintain their skills. The further growth of DBE in the United States also requires an appropriate Current Procedural Terminology code that reflects the difficulty of the procedure and the time required. To the extent that it can replace intraoperative enteroscopy and prevent repeated admissions for ongoing bleeding, it adds immense value to the health care system.
Ahmad Kamal, MD, MSc Santa Clara Valley Medical Center Stanford University School of Medicine Stanford, California, USA Abbreviations: AVM, arteriovenous malformation; DBE, double-balloon enteroscopy; OGIB, obscure GI bleeding; PE, push enteroscopy.
The author reports the following conflict: A. Kamal has served on the Speaker’s Bureau for Novartis, received research support from Gilead, and is a consultant for NeoGuide.
1. Gross S, Stark ME. Initial experience with double-balloon enteroscopy at a U.S. center. Gastrointest Endosc 2008;67:890-7. 2. Mehdizadeh S, Ross A, Gerson L, et al. What is the learning curve associated with double-balloon enteroscopy? Technical details and early experience in 6 U.S. tertiary care centers. Gastrointest Endosc 2006;64: 740-50. 3. Mehdizadeh S, Han NJ, Cheng DW, et al. Success rate of retrograde double-balloon enteroscopy. Gastrointest Endosc 2007;65:633-9. 4. Yamamoto H, Sekine Y, Sato Y, et al. Total enteroscopy with a nonsurgical steerable double-balloon method. Gastrointest Endosc 2001;53: 216-20. 5. Mensink PB, Haringsma J, Kucharzik T, et al. Complications of double balloon enteroscopy: a multicenter survey. Endoscopy 2007;39:613-5. 6. May A, Nachbar L, Ell C. 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. 7. Kaffes AJ, Siah C, Koo JH. Clinical outcomes after double-balloon enteroscopy in patients with obscure GI bleeding and a positive capsule endoscopy. Gastrointest Endosc 2007;66:304-9. 8. Perez-Cuadrado E, Latorre R, Carballo F, et al. Training and new indications for double balloon endoscopy. Gastrointest Endosc 2007;66: S39-46. 9. May A, Nachbar L, Pohl J, et al. Endoscopic interventions in the small bowel using double balloon enteroscopy: feasibility and limitations. Am J Gastroenterol 2007;102:527-35. 10. Yamamoto H, Kita H, Sunada K, et al. Clinical outcomes of double-balloon endoscopy for the diagnosis and treatment of small-intestinal diseases. Clin Gastroenterol Hepatol 2004;2:1010-6. 11. Sun B, Rajan E, Cheng S, et al. Diagnostic yield and therapeutic impact of double-balloon enteroscopy in a large cohort of patients with obscure gastrointestinal bleeding. Am J Gastroenterol 2006;101:2011-5. 12. Safatle-Ribeiro AV, Kuga R, Ishida R, et al. Is double-balloon enteroscopy an accurate method to diagnose small-bowel disorders? Surg Endosc 2007;21:2231-6. 13. Zhong J, Ma T, Zhang C, et al. A retrospective study of the application on double-balloon enteroscopy in 378 patients with suspected smallbowel diseases. Endoscopy 2007;39:208-15. 14. Hirai F, Matsui T, Yao K, et al. Efficacy of carbon dioxide insufflation in endoscopic balloon dilation therapy by using double balloon endoscopy. Gastrointest Endosc 2007;66:S26-9. 15. Lo SK. Technical matters in double balloon enteroscopy. Gastrointest Endosc 2007;66:S15-8. 16. Honda K, Itaba S, Mizutani T, et al. An increase in the serum amylase level in patients after peroral double-balloon enteroscopy: an association with the development of pancreatitis. Endoscopy 2006;38: 1040-3. 17. Gay G, Delvaux M, Fassler I. Outcome of capsule endoscopy in determining indication and route for push-and-pull enteroscopy. Endoscopy 2006;38:49-58. 18. Lai LH, Wong GL, Chow DK, et al. Long-term follow-up of patients with obscure gastrointestinal bleeding after negative capsule endoscopy. Am J Gastroenterol 2006;101:1224-8. 19. Suzuki T, Matsushima M, Okita I, et al. Clinical utility of double-balloon enteroscopy for small intestinal bleeding. Dig Dis Sci 2007;52:1914-8.
Volume 67, No. 6 : 2008 GASTROINTESTINAL ENDOSCOPY 901