Double-Balloon Enteroscopy and Capsule Endoscopy Have Comparable Diagnostic Yield in Small-Bowel Disease: A Meta-Analysis

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2008;6:671 676 Double-Balloon Enteroscopy and Capsule Endoscopy Have Comparable Diagnostic Yield in Small-Bowel Disease: A Meta-Analysis SHABANA F. PASHA, JONATHAN A. LEIGHTON, ANANYA DAS, M. EDWYN HARRISON, G. ANTON DECKER, DAVID E. FLEISCHER, and VIRENDER K. SHARMA Division of Gastroenterology & Hepatology, Department of Internal Medicine, Mayo Clinic, Scottsdale, Arizona Background & Aims: The aim of this study was to compare the diagnostic yield of capsule endoscopy (CE) with double-balloon enteroscopy (DBE) in small-bowel (SB) disease using meta-analysis. Methods: We performed a search of studies comparing CE with DBE in SB disease. Data on diagnostic yield of CE and DBE were extracted, pooled, and analyzed. The weighted incremental yield (IY W ) (yield of CE yield of DBE) of CE over DBE and 95% confidence intervals (95% CIs) for pooled data were calculated using a fixed-effect model (FEM) for analyses without, and a random-effect model (REM) for analyses with, significant heterogeneity. Results: Eleven studies compared CE and DBE; the pooled overall yield for CE and DBE was 60% (n 397) and 57% (n 360), respectively (IY W, 3%; 95% CI, 4% to 10%; P.42; FEM). Ten studies reported vascular findings; the pooled yield for CE and DBE was 24% (n 371) and 24% (n 364), respectively (IY W, 0%; 95% CI, 5% to 6%; P.88; REM). Nine studies reported inflammatory findings; the pooled yield for CE and DBE was 18% (n 343) and 16% (n 336), respectively (IY W, 0%; 95% CI, 5% to 6%; P.89; FEM). Nine studies reported polyps/ tumors; the pooled yield for CE and DBE was 11% (n 343) and 11% (n 336), respectively (IY W, 1%; 95% CI, 5% to 4%; P.76; FEM). Conclusions: CE and DBE have comparable diagnostic yield in SB disease, including obscure gastrointestinal bleeding. CE should be the initial diagnostic test because of its noninvasive quality, tolerance, ability to view the entire SB, and for determining the initial route of DBE. Because of its therapeutic capabilities, DBE may be indicated in patients with a positive finding on CE requiring a biopsy or therapeutic intervention, if suspicion for a SB lesion is high despite a negative CE, and in patients with active bleeding. Evaluation of the entire small intestine has been challenging, with limited success in the past because of a lack of endoscopic modalities. Common sources of bleeding in the small intestine, including angioectasias and ulcerations, are not detected easily on small-bowel (SB) imaging studies. 1 Conventional endoscopy is limited to evaluation of the proximal small intestine. 2 The introduction of capsule endoscopy (CE) in 2001, and subsequently double-balloon enteroscopy (DBE) in 2003, has revolutionized the diagnosis and management of patients with small-intestinal disorders, especially those with obscure gastrointestinal bleeding (OGIB). 3 5 CE has the advantage over conventional endoscopic modalities of allowing visualization of the entire small intestine, but is limited by a lack of therapeutic options. 6 DBE in addition to visualization of most of the small intestine, usually by a combined antegrade and retrograde approach, has the advantage of diagnostic biopsy capability and therapeutic interventions. 7 However, DBE is a more invasive procedure than CE, is often of prolonged duration, and needs to be performed under moderate/deep sedation or general anesthesia. 8 We have previously published 2 meta-analyses comparing the yield of CE with other diagnostic modalities that clearly showed the superiority of CE over both push enteroscopy and SB barium radiography in the diagnosis of patients with OGIB and nonstricturing Crohn s disease. 9,10 Several small studies have compared the yield of DBE and CE in the evaluation of patients with suspected small-intestinal disease, including OGIB. These studies have shown inconsistent results, and largely were limited by their small sample size. To overcome these limitations, we conducted a meta-analysis of studies comparing DBE with CE, to determine which of these 2 endoscopic modalities would have a higher diagnostic yield in patients with suspected small-intestinal disease. Methods A trained librarian conducted a thorough literature search through December 2006 for all published studies comparing CE and DBE for the evaluation of patients with suspected small-intestinal disease, using the MEDLINE, EMBASE, and Cochrane Central Trials databases. The MeSH subheadings used were double-balloon endoscopy, double-balloon enteroscopy, and push-and-pull enteroscopy. Our search included articles published in English and non-english languages. In addition, a hand search of reference lists and abstracts presented at the National Meetings of the American College of Gastroenterology and Digestive Diseases Week from 2003 to 2006 was performed. In case of articles presenting the same data, the most recent update was used in our analysis (Figure 1). All studies were reviewed independently by 2 reviewers (S.F.P. and V.K.S.) to determine eligibility for inclusion, and any differences were to be resolved by consensus. Both reviewers were expert in content matter. Trials that compared the yield of DBE and CE for SB findings were included in the analysis. For all Abbreviations used in this paper: CE, capsule endoscopy; CI, confidence interval; DBE, double-balloon enteroscopy; FEM, fixed-effect model; IY W, weighted incremental yield; OGIB, obscure gastrointestinal bleeding; REM, random-effect model; SB, small bowel. 2008 by the AGA Institute 1542-3565/08/$34.00 doi:10.1016/j.cgh.2008.01.005

672 PASHA ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 6 random-effect model (REM) was used for analyses with significant heterogeneity. The funnel plot visual inspection was performed to evaluate for the presence of any publication bias. Figure 1. Search strategy. studies that required clarification of data, we contacted and requested information from individual investigators. Data were extracted on the yield of all SB findings on CE and DBE. In addition, clinically significant findings were subdivided as vascular (angioectasias), inflammatory (ulcerations, erosions, and strictures), and mass lesions (tumors and polyps), when these data were available. Because of their small number, we included all prospective and retrospective studies, articles, and abstracts in our main analysis (Table 1). However, sensitivity analyses were performed using only prospective trials published as fulllength articles. Statistical Analysis Data on the diagnostic yield with CE and DBE were extracted, pooled, and analyzed using RevMan 4.2.9 software (The Cochrane Collaboration, Baltimore, MD). Heterogeneity was tested by the chi-square method and a P value of less than.1 was considered significant heterogeneity. Weighted incremental yield (IY W ) was defined as the weighted difference in the yield of CE over DBE, and 95% confidence intervals (95% CIs) for the pooled data were calculated. A fixed-effect model (FEM) was used for analyses without significant heterogeneity, and a Results A total of 113 studies on DBE initially were identified using the search strategy. This included 35 case reports, 32 original articles, 45 abstracts, and 1 editorial. Sixty-two articles not relevant to the study were excluded after preliminary review, leaving 51 for detailed evaluation. Of these, there were 12 studies (3 articles and 9 abstracts) that compared CE and DBE. One study (Nakamura et al 11 ) was published in both abstract and full-length article form, and hence the abstract was excluded from the analysis. All 3 articles were prospective studies. 8,12,13 Of the abstracts, 4 were prospective studies, 14 17 and 4 were retrospective 18 21 (Figure 1). All the selected studies were published in English. There was no disagreement between the 2 reviewers regarding inclusion of the studies in the analysis. The 11 studies (3 full-length articles and 8 abstracts) that met inclusion criteria included a total of 375 patients who underwent DBE for suspected small-intestinal disease, including 350 patients for obscure GI bleed. There was no heterogeneity among the studies (P.24), and the pooled overall yield for CE and DBE using a FEM was 60% (N 393) and 57% (N 356), respectively, with an IY W of 3% (P.42; 95% CI, 4% to 10%) (Figure 2). Ten studies reported the yield of vascular findings (angioectasias) on CE and DBE. 8,12 14,16 21 There was significant heterogeneity among the studies (P.0002), and the pooled yield for CE and DBE for vascular findings was 24% (N 371) and 24% (N 364), respectively, with an IY W of 0% (P.88; 95% CI, 5% to 6%) using a REM (Figure 3A). Nine studies reported the yield of inflammatory findings (ulcers, erosions, and strictures) on CE and DBE. 8,12 14,17 21 There was no heterogeneity among the studies (P.82), and the pooled yield for CE and DBE for inflammatory findings was 18% (N 343) and 16% (N 336), respectively, with an IY W of 0% (P.89; 95% CI, 5% to 6%) using a FEM (Figure 3B). Nine studies reported the yield of mass lesions (tumors and polyps) on CE and DBE. 8,12 14,17 21 There was no heterogeneity among the studies (P.19), and the pooled yield for CE and DBE for mass lesions was 11% (N 343) and 11% (N 336), respectively, with an IY W of 1% (P.76; 95% CI, 5% to 4%) using a FEM (Figure 3C). Table 1. Summary of Trials Included for Meta-Analysis Study Country Article type Study design Number of patients Hadithi et al, 12 2006 The Netherlands Full-length article Prospective, nonblinded 35 Nakamura et al, 13 2005 Japan Full-length article Prospective, blinded 28 Matsumoto et al, 8 2005 Japan Full-length article Prospective, blinded 22 Arakawa et al, 19 2006 Japan Abstract Retrospective 43 Damian et al, 16 2006 Germany Abstract Prospective, blinded 28 Gonzalez Suarez et al, 17 2006 Spain Abstract Prospective 15 Kameda et al, 14 2006 Japan Abstract Prospective, blinded 24 Jang et al, 20 2006 Korea Abstract Retrospective 35 Mehdizadeh et al, 18 2006 United States Abstract Retrospective 130 Shiba et al, 21 2006 Japan Abstract Retrospective 26 Wi et al, 15 2006 Korea Abstract Prospective 11

June 2008 COMPARISON OF DBE WITH CE 673 Figure 2. Incremental yield (%) of individual trials and pooled data for total yield with CE compared with DBE for all SB findings on CE compared with DBE in all studied patients. Data on sensitivity analysis using the prospective studies published as full-length articles showed a significantly higher overall yield of CE over DBE (IY W, 19%; 95% CI, 5% 33%) for the diagnosis of clinically significant findings, (Figure 4) and a nonsignificantly higher yield for vascular lesions (IY W, 7%; 95% CI, 5% to 19%) and tumors (IY W, 7%; 95% CI, 3% to 17%). There was no difference in the yield for inflammatory lesions (IY W, 0%; 95% CI, 10% to 10%). There was no heterogeneity encountered in this analysis. Visual inspection of funnel plots of analyses revealed no evidence of publication bias (Figure 5). Discussion Very few studies have compared the diagnostic yield of CE and DBE, and these largely have been limited by their small sample size, in turn leading to inconsistent conclusions on the superiority of one endoscopic modality over the other. Because of the considerable differences in technique, patient tolerance, and advantages and disadvantages of these endoscopic modalities, it is important to determine which modality has a higher diagnostic yield, and hence should be the initial endoscopy in the evaluation of the small intestine. The main advantage of CE is its capacity to evaluate the entire small intestine in a noninvasive manner. It has the disadvantage of being a purely diagnostic modality without therapeutic capabilities, 6 and may be limited by inadequate bowel preparation, limited field of vision, and poor visual clarity. 22 In addition, because CE is not a real-time procedure, identification of patients with incomplete CE, and performance of timely interventions that would facilitate completion of the examination, are not feasible. Both false-positive and false-negative findings have been reported with CE, and the high detection rate of incidental findings reported in up to 23% of normal controls may lead to unnecessary endoscopic or surgical procedures. 23 The complications reported with CE are inability to swallow the capsule with impaction at the cricopharyngeus, 24 capsule retention often necessitating endoscopic removal or surgery, 25 and rare reports of disintegration of the capsule in the GI tract. 26 In addition to endoscopic evaluation of the entire small intestine, usually with a combination of antegrade and retrograde approaches, DBE allows diagnostic techniques including biopsies and India ink tattooing; and therapeutic procedures, which include polypectomy, hemostasis (epinephrine injection, thermocoagulation, and argon plasma coagulation), endoscopic mucosal resection, dilation of strictures, stent placement, and retrieval of foreign objects (including retained CE). 7,27,28 However, DBE is a more invasive procedure than CE, often requires general anesthesia, and can be time consuming. 8 Despite the best efforts by the endoscopist, evaluation of the entire small intestine may not be feasible in some patients. 29 The main complications reported with DBE are SB ileus, perforation, and pancreatitis. 13,30,31 The optimal route of DBE (antegrade or retrograde) in the evaluation of SB disease is based largely on the clinical discretion of the gastroenterologist regarding the location of the bleeding source in the proximal or distal SB, as well as any positive findings on prior CE or imaging studies. Our meta-analysis revealed a comparable yield of SB findings for both CE and DBE in patients with suspected SB disease, predominantly OGIB. The yield was comparable for vascular and inflammatory lesions, and SB tumors. However, data using the highest-quality prospective studies published as full-length articles, described later, 8,12,13 showed a 19% higher overall diagnostic yield of CE over DBE for clinically significant findings. The statistically significant higher yield on CE was caused by a numerically higher yield for vascular lesions and tumors, and a higher number of miscellaneous findings including blood/ blood clots reported with CE. Hadithi et al 12 evaluated 35 consecutive patients with OGIB prospectively using CE and DBE. The diagnostic yield of CE was 80% compared with 60% with DBE. DBE could not confirm findings detected by CE in 8 patients (angioectasia [3], polyps [2], and fresh blood [3]), and, conversely, 1 patient had angioectasia detected on DBE that were not detected on CE. Although both examinations were well tolerated, CE was more acceptable to the patients. DBE facilitated the ability to diagnose and perform therapeutic procedures in 77% of patients. Seventy-four percent of patients who underwent DBE remained clinically stable without further need for blood transfusions. Nakamura et al 13 similarly evaluated 28 patients with OGIB using both endoscopic methods. Thirteen patients had A1 lesions (requiring immediate hemostatic procedures), and 6 had A2 lesions (requiring close observation) detected on CE. In comparison, 11 had A1 lesions and 1 had an A2 lesion with DBE. The diagnostic yield of CE was 59.4% compared with 42.9% with DBE, which was not a statistically significant dif-

674 PASHA ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 6 Figure 3. (A) Incremental yield (%) of individual trials and pooled data for yield of vascular lesions with CE compared with DBE. (B) Incremental yield (%) of individual trials and pooled data for yield of inflammatory lesions with CE compared with DBE. (C) Incremental yield (%) of individual trials and pooled data for yield of neoplasms with CE compared with DBE. ference. In the evaluation of A1 lesions there were concordant findings in 8 patients. There was a higher discordance for A2 lesions, with more lesions detected on CE than DBE. The difference in detection rate may have been related to the fact that the entire small intestine was visualized in 90.6% of patients on CE and only in 62.5% of patients on DBE. DBE allowed diagnostic and therapeutic interventions in 82% of patients with A1 lesions. Ninety percent of patients had no recurrent bleeding at the 6-month follow-up evaluation. Matsumoto et al 8 evaluated 22 patients using DBE, followed by CE, and found that the overall diagnostic yield of DBE was similar to CE. The patients included 13 with OGIB and 9 with

June 2008 COMPARISON OF DBE WITH CE 675 Figure 4. Sensitivity analysis of incremental yield (%) of individual high-quality trials and pooled data for all SB findings on CE compared with DBE in all studied patients. suspected polyposis. The yield of DBE was 54.5%, and CE detected positive findings in 36.4% of patients in the area of the small intestine evaluated by DBE, and in 50% of patients in the area not explored by DBE. The findings were concordant in 12 of 13 patients with OGIB, with 1 SB ulcer missed on CE. Similarly, in the patients with polyposis on DBE, CE failed to detect any polyps in 3 patients, and detected a smaller number of polyps than DBE in 2 patients. A recent cost-benefit analysis that examined potential strategies for the diagnosis and management of OGIB suggested that DBE might be a more cost-effective modality than CE, especially when AVMs were the underlying cause for the bleeding. The analysis, however, lacked data on important outcomes such as quality of life, and was limited by the strong assumptions of the analytic model. 32 There are reports of SB tumors missed on CE that have been detected by push enteroscopy or DBE. 22,33,34 In the study by Matsumoto et al 8 CE failed to detect any polyps in 3 patients who were found to have polyps on DBE, and a smaller number of polyps as compared with DBE in 2 additional patients. It could be postulated that CE may be limited in the diagnosis of solitary lesions in the small intestine, including tumors, because of the variability in its position and direction, and its inability to provide a circumferential view as it passes through the small intestine. Although our sensitivity analysis showed that CE had a higher yield as compared with DBE for mass lesions, the majority of the tumors/polyps detected on CE were located in the portion of SB that was not evaluated by DBE, and were not truly missed on DBE. 8 Therefore, DBE still would be of value in patients with suspected SB lesions, despite a negative CE, because it may have an advantage over CE both for the detection and tissue diagnosis of SB tumors. Figure 5. Funnel plot analysis comparing CE with DBE in all studied patients. Note is made of an outlying trial (Wi et al 15 ). Statistical analysis confirmed no evidence of publication bias. One of the limitations of our meta-analysis was the inclusion of data from abstracts, which usually are considered to be of poor quality because of incomplete data. These were included because of the small number of studies comparing DBE and CE so that we could obtain a pooled sample size that would reach statistical significance. Other limitations included differences in inclusion criteria of patients in individual studies, small sample sizes of the studies, and differences in the interpretation of findings on CE and DBE by different investigators. There was significant heterogeneity between study results for vascular findings, which was addressed statistically using a REM. In addition, we communicated directly with the primary investigators to obtain necessary information that was not reported in the abstracts. Although the diagnostic yield of both these novel endoscopic modalities appears to be higher than push enteroscopy, there are relatively scant data on their actual impact on patient management and overall outcome. In addition, it is difficult to ascertain their true diagnostic yield in the absence of comparison with a gold standard such as intraoperative enteroscopy or surgery. In summary, CE and DBE appear to have a comparable diagnostic yield in patients with suspected SB disease, including OGIB. However, CE is often the preferred initial diagnostic test because of its noninvasive quality, better tolerance, and ability to view the entire SB. It also can be very helpful in determining the initial approach for DBE. Because of its diagnostic and therapeutic capabilities, DBE would be indicated in patients with a positive finding on CE requiring endoscopic biopsy or therapeutic intervention, and patients in whom suspicion for a SB lesion is high, despite a negative CE. DBE also may be the preferred initial approach in those patients who are actively bleeding, but more data are needed to confirm this. References 1. Hara AK, Leighton JA, Sharma VK, et al. Small bowel: preliminary comparison of capsule endoscopy with barium study and CT. Radiology 2004;230:260 265. 2. Lewis BS. The history of enteroscopy. Gastrointest Endosc Clin N Am 1999;9:1 11. 3. Buchman AL, Wallin A. Videocapsule endoscopy renders obscure gastrointestinal bleeding no longer obscure. J Clin Gastroenterol 2003;37:303 306. 4. Yamamoto H, Yano T, Kita H, et al. New system of double-balloon enteroscopy for diagnosis and treatment of small intestinal disorders. Gastroenterology 2003;125:1556. 5. May A, Nachbar L, Ell C. Double-balloon enteroscopy (push-andpull enteroscopy) of the small bowel: feasibility and diagnostic and therapeutic yield in patients with suspected small bowel disease. Gastrointest Endosc 2005;62:62 70. 6. Ell C, Remke S, May A, et al. The first prospective controlled

676 PASHA ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 6 trial comparing wireless capsule endoscopy with push enteroscopy in chronic gastrointestinal bleeding. Endoscopy 2002;34:685 689. 7. 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 1016. 8. Matsumoto T, Esaki M, Moiyama T, et al. Comparison of capsule endoscopy and enteroscopy with the double-balloon method in patients with obscure bleeding and polyposis. Endoscopy 2005; 37:827 832. 9. Triester SL, Leighton JA, Leotiadis GI, et al. A meta-analysis of the yield of capsule endoscopy compared to other diagnostic modalities in patients with obscure gastrointestinal bleeding. Am J Gastroenterol 2005;100:2407 2418. 10. Triester SL, Leighton JA, Leontiadis GI, et al. A meta-analysis of the yield of capsule endoscopy compared to other diagnostic modalities in patients with non-stricturing small bowel Crohn s disease. Am J Gastroenterol 2006;101:954 964. 11. Nakamura M, Niwa Y, Ohmiya N, et al. A preliminary comparison of capsule endoscopy and double-balloon enteroscopy in patients with suspected small bowel bleeding. Gastrointest Endosc 2005;61:AB176. 12. Hadithi M, Heine GDN, Jacobs MAJM, et al. A prospective study comparing video capsule endoscopy with double-balloon enteroscopy in patients with obscure gastrointestinal bleeding. Am J Gastroenterol 2006;101:52 57. 13. Nakamura M, Niwa Y, Ohmiya N, et al. Preliminary comparison of capsule endoscopy and double-balloon enteroscopy in patients with suspected small-bowel bleeding. Endoscopy 2006; 38:59 66. 14. Kameda N, Higuchi K, Shiba M, et al. A prospective trial comparing wireless capsule endoscopy and double-balloon enteroscopy in patients with obscure gastrointestinal bleeding (abstr). Gastrointest Endosc 2006;63:AB162. 15. Wi JH, Kim JO, Eun SH, et al. A prospective comparison of capsule endoscopy and double balloon enteroscopy in small bowel disease (abstr). Gastrointest Endosc 2006;63:AB175. 16. Damian U, Schilling D, Hartmann D, et al. Double-balloon enteroscopy (push and pull enteroscopy) of the small bowel: comparison with video capsule endoscopy and magnetic resonance imaging (abstr). Gastrointest Endosc 2006;63:AB174. 17. Gonzalez-Suarez B, Galter S, Espinos JC, et al. Comparative study of capsule endoscopy, mesenteric angiography and double balloon enteroscopy in obscure gastrointestinal bleeding: preliminary results (abstr). Gastroenterology 2006;130:AB476. 18. Mehdizadeh S, Ross AS, Leighton J, et al. Double balloon enteroscopy (DBE) compared to capsule endoscopy (CE) among patients with obscure gastrointestinal bleeding (OGIB): a multicenter US experience (abstr). Gastrointest Endosc 2006;63: AB90. 19. Arakawa D, Ohmiya N, Nakamura M, et al. Usefulness of doubleballoon enteroscopy (DBE) for diagnosis and treatment of obscure gastrointestinal bleeding-in comparison with capsule endoscopy (abstr). Gastrointest Endosc 2006;63:AB175. 20. Jang HJ, Choi MH, Park CH, et al. Comparison of double balloon enteroscopy and capsule endoscopy in patients with suspected small bowel diseases (abstr). Gastrointest Endosc 2006;63: AB181. 21. Shiba M, Higuchi K, Kameda N, et al. Wireless capsule endoscopy and double-balloon enteroscopy in Japanese patients with obscure gastrointestinal bleeding (abstr). Gastrointest Endosc 2005;61:AB182. 22. Chong AKH, Chin BWK, Meredith CG. Clinically significant smallbowel pathology identified by double-balloon enteroscopy but missed by capsule endoscopy. Gastrointest Endosc 2006;64: 445 449. 23. Goldstein J, Eisen G, Lewis B, et al. Abnormal small bowel findings are common in healthy subjects screened for a multicenter, double blind, randomized, placebo controlled trial using capsule endoscopy (abstr). Gastroenterology 2003;124:AB337. 24. Fleischer DE, Heigh RI, Nguyen CC, et al. Video capsule impaction at the cricopharyngeus: a first report of this complication and its successful resolution. Gastrointest Endosc 2003;57:427 428. 25. Barkin J, Friedman S. Wireless capsule endoscopy (WCE) retention requiring surgical intervention: the world experience (abstr). Am J Gastroenterol 2002;97:S928. 26. Fry LC, De Petris G, Swain JM, et al. Impaction and fracture of a video capsule in the small bowel requiring laparotomy for removal of the capsule fragments. Endoscopy 2005;37:674 676. 27. Ell C, May A, Nachbar L, et al. Push-and-pull enteroscopy in the small bowel using the double-balloon technique: results of a prospective European multicenter study. Endoscopy 2005;37: 613 616. 28. May A, Nachbar L, Ell C, et al. Extraction of entrapped capsules from the small bowel by means of push-and-pull enteroscopy with the double balloon technique. Endoscopy 2005;37:591 593. 29. 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 US tertiary care centers. Gastrointest Endosc 2006;64:740 750. 30. Attar A, Maissiat E, Sebbagh V, et al. First case of paralytic intestinal ileus after double balloon enteroscopy. Gut 2005;54: 1823 1824. 31. Groenen MJM, Moreels TGG, Orlent H, et al. Acute pancreatitis after double-balloon enteroscopy: an old pathogenetic theory revisited as a result of using a new endoscopic tool. Endoscopy 2006;38:82 85. 32. Kamal A, Gerson LB. Cost-benefit analysis of double balloon enteroscopy compared to other diagnostic modalities for obscure gastrointestinal hemorrhage (abstr). Gastrointest Endosc 2006; 63:AB9055. 33. Madisch A, Schimming W, Kinzel F, et al. Locally advanced smallbowel adenocarcinoma missed primarily by capsule endoscopy but diagnosed by push enteroscopy. Endoscopy 2003;35:861 864. 34. Ross A, Mehdizadeh S, Tokar J, et al. Double balloon enteroscopy for the detection of small bowel mass lesions presenting as obscure gastrointestinal bleeding: do we have a new gold standard (abstr)? Gastrointest Endosc 2006;63:AB159. Address requests for reprints to: Shabana F. Pasha, MD, 13400 E Shea Boulevard, Mayo Clinic, Scottsdale, Arizona 85259. e-mail: pasha. shabana@mayo.edu; fax: (480) 301-8673. J.A.L. has received research support from Fujinon, Inc. D.E.F., J.A.L., and V.K.S. have received research support from Given Imaging. J.A.L. and V.K.S. are consultants for Given Imaging. Presented in abstract form at Digestive Diseases Week, Washington, DC, May 19 24 2007. Previously published in abstract form in Gastrointestinal Endoscopy 2007;65:A364.