ORIGINAL ARTICLE: Clinical Endoscopy

Emergency video capsule endoscopy in patients with acute severe GI bleeding and negative upper endoscopy results Christoph Schlag, MD,1 Christoph Menzel,1 Simon Nennstiel, MD,1 Bruno Neu, MD,1 Veit Phillip, MD,1 Tibor Schuster, PhD,2 Roland M. Schmid, MD,1 Stefan von Delius, MD1 Munich, Germany

Background: In mid-GI bleeding, video capsule endoscopy (VCE) shows the best diagnostic yield for ongoing overt bleeding. To date, the utility of VCE in acute severe GI bleeding has been analyzed rarely. Objective: To evaluate the impact of VCE when performed on patients with acute severe GI bleeding immediately after an initial negative upper endoscopy result. Design: Prospective study. Setting: Tertiary-care center. Patients: Patients with melena, dark-red or maroon stool, hemodynamic instability, drop of hemoglobin level R2 g/dL/day, and/or need of transfusion R2 units of packed red blood cells per day were included. Interventions: After a negative upper endoscopy result, emergency VCE was performed by immediate endoscopic placement of the video capsule into the duodenum. Main Outcome Measurements: Rate of patients in whom emergency VCE correctly guided further diagnostic and therapeutic procedures. Results: Upper endoscopy showed the source of bleeding in 68 of 88 patients (77%). In the remaining 20 patients (23%), emergency VCE was performed, which was feasible in 19 of 20 patients (95%; 95% confidence interval [CI], 75%-99%). Emergency VCE correctly guided further diagnostic and therapeutic procedures in 17 of 20 patients (85%; 95% CI, 62%-97%) and showed a diagnostic yield of 75% (95% CI, 51%-91%). Limitations: Single-center study, small sample size. Conclusion: In patients with acute severe GI bleeding and negative upper endoscopy results, emergency VCE can be useful for the immediate detection of the bleeding site and is able to guide further therapy. (Clinical trial registration number: NCT01584869.) (Gastrointest Endosc 2015;81:889-95.)

Acute GI bleeding is a common clinical problem, with annual incidence rates of approximately 50 to 150 per 100,000 persons, with a significant mortality rate.1 GI bleeding includes cases of upper GI bleeding (bleeding source can be reached by EGD), lower GI bleeding (bleeding source can be reached by colonoscopy), and

mid-intestinal GI bleeding, which involves bleeding with an origin located between the papilla and the ileocecal valve.2 In acute GI bleeding, upper endoscopy is able to detect the bleeding lesion in the majority of patients. Early EGD within 24 hours has been recommended as the first diagnostic approach.3-5 However, if initial EGD cannot

Abbreviations: BAE, balloon-assisted enteroscopy; VCE, video capsule endoscopy.

0016-5107/$36.00 http://dx.doi.org/10.1016/j.gie.2014.09.035

DISCLOSURE: C. Schlag, B. Neu, and S. von Delius received speaker’s honoraria from Given Imaging. Given Imaging provided the video capsules but did not participate in the study design, data collection, data analysis, or manuscript preparation. No other financial relationships relevant to this article were disclosed.

Received June 20, 2014. Accepted September 10, 2014.

See CME section; p. 976.

Current affiliations: II. Medizinische Klinik, Klinikum rechts der Isar, Technische Universität München (1), Institut für Medizinische Statistik und Epidemiologie, Klinikum rechts der Isar der Technischen Universität München, Munich, Germany (2).

Copyright ª 2015 by the American Society for Gastrointestinal Endoscopy

Reprint requests: Christoph Schlag, MD, II. Medizinische Klinik, Klinikum rechts der Isar, Technische Universität München, Ismaninger Str. 22, 81675 Munich, Germany.

www.giejournal.org

Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 889

Emergency video capsule endoscopy

identify the source of bleeding, there are no clear recommendations on the optimal approach for further diagnosis. Regardless whether mid-GI or lower GI bleeding is suspected, colonoscopy usually is performed as the next diagnostic procedure. However, colonoscopy provides only indirect information about mid-GI bleeding, when blood is seen in the terminal ileum. Moreover, for lower GI bleeding, the diagnostic yield of colonoscopy in the emergency setting is low.6,7 VCE of the small bowel is a minimally invasive method, which allows not only visualization of the entire mid-GI tract but also of the proximal parts of the lower GI tract.8 For the main indication of suspected mid-GI bleeding, the best diagnostic yield of VCE has been reported for active overt bleeding.9 However, the utility of VCE in the emergency setting of acute GI bleeding has been analyzed rarely. In the actual study, we evaluated a new diagnostic algorithm by using VCE immediately after a negative EGD result in patients with severe GI bleeding in order to determine the diagnostic yield of VCE and its impact on further therapeutic management.

Schlag et al

This prospective study was performed between December 2011 and February 2014 at a single university hospital (Klinikum rechts der Isar der Technischen Universität München, Munich, Germany). Written informed consent was obtained from all participants. The study was approved by the local ethics committee and registered on ClinicalTrials.gov (Identifier: NCT01584869). The patients we recruited experienced melena or dark red or maroon stools together with signs of severe bleeding, which was defined as hemodynamic instability (defined as mean arterial pressure !65 mm Hg and/or heart rate O110/minute) and/or drop of hemoglobin level R 2 g/dL/day (compared with previously known values) and/or need of transfusion R 2 units of packed red blood cells per day. Patients who exhibited hematemesis (suggestive for upper GI bleeding) or rectal fresh red blood (highly suggestive for lower GI bleeding) were not eligible for study inclusion. Patients aged !18 years and pregnant patients were excluded as well as patients with contraindications for EGD and contraindications for capsule endoscopy, which included known or suspected intestinal obstruction.

routinely received erythromycin (250 mg intravenously) shortly before gastroscopy. If EGD detected a probable bleeding source, which included the presence of fresh or old blood, varices with stigmata of hemorrhage, MalloryWeiss tears, severe reflux esophagitis (Los Angeles grade C or D), ulcers with stigmata of hemorrhage, angiodysplasias with stigmata of hemorrhage, Dieulafoy’s lesions, or tumors with stigmata of hemorrhage, endoscopic hemostasis was carried out as necessary. Patients in whom EGD did not show a probable bleeding source received immediate VCE (Pillcam SB 2-4, Given Imaging, Yoqneam, Israel). Before capsule placement, 0.5 L of polyethylene glycol cleansing solution was endoscopically instilled into the duodenum at the end of the initial EGD. To reduce the risk of aspiration, endoscopies were carried out while the patients were placed in a left lateral decubitus position with elevation of the upper part of the body. In the case of backflow of polyethylene glycol solution in the stomach, this excess fluid was sucked off. Thereafter, an 8-lead sensor array (Given Imaging) was applied to the still-sedated patient and connected to the capsule data recorder (DR-3, Given Imaging). After the connection was verified, the capsule was placed into the duodenum by using an endoscopic capsule delivery device (Pillcam Express, Given Imaging, which was available only until November 2012 [for patients 1-10] and AdvanCE, US Endoscopy, Mentor, Ohio since December 2012 [for patients 10-20]) as previously described.10 After endoscopic capsule placement, patients continued bowel cleansing with polyethylene glycol solution (application by mouth or by a nasogastric tube) as preparation for intended further endoscopic approach by enteroscopy or colonoscopy. VCE recording was stopped when either blood clearly could be detected by the live viewer function of the capsule data recorder, which was checked after 30 minutes and after 2 hours from capsule application, or when the capsule had clearly reached the colon, also monitored by the live viewer function of the capsule data recorder. Capsule recording finished 8 hours after application at the latest because of battery life. VCE images were transferred and analyzed immediately by using the manufacturer’s standard software (Rapid Reader, Given Imaging). Each capsule study was read by 1 experienced observer (C.S. or S.D.), and the suspected site of bleeding was documented. Depending on the suspected finding, patients received further therapeutic intervention (push enteroscopy, balloon enteroscopy, colonoscopy, surgery). If no bleeding source was detected by VCE, patients were referred for colonoscopy to exclude lower GI bleeding.

Study procedure

Outcome measures

All patients were referred for emergency EGD within 24 hours after presentation in the emergency department or within 24 hours after notification of active bleeding signs (when the patients had already been hospitalized for other illness). A nasogastric tube lavage was not used in any of the patients before EGD. For gastric emptying, patients

The primary outcome measure was the rate of patients in whom VCE correctly guided further diagnostic and therapeutic procedures. The secondary outcome measures included diagnostic yield of VCE, quality of VCE (cleansing level), number of transfused packed red blood cells, time on intensive or intermediate care unit, length of hospital

890 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015

www.giejournal.org

PATIENTS AND METHODS Study design and patients

Schlag et al

stay, and mortality of the included patients. The cleansing level of the small bowel was assessed by using a subjective qualitative evaluation scale as previously described.11 If active bleeding was detected, the cleansing level was rated only up to the bleeding site.

Sample size estimation A priori power calculation suggested a sample size of 20 patients considering a prevalence of O10% for a clinical relevant finding detected by VCE in the studied patient population and assuming a power of 90%. Because the type II error was the primary controlling error, the power was set to 90% instead of the commonly anticipated 80%. In this noncomparative study, the statistical power (1 minus type II error) refers to the probability to detect at least 1 clinically relevant finding in the study sample. We assumed 10% to be the minimum relevant prevalence of such a critical condition in the target population. Sample size calculation was done by nQuery, version 7.0 (Statistical Solutions, Boston, Mass).

Data handling and statistical analysis Microsoft Excel (Microsoft Corporation, Redmond, Wash) was used for data handling. Categorical variables were illustrated as percentages, ordinal variables as medians, and interquartile ranges and numerical variables as means and standard deviations. Comparisons of variables were performed by chi square test, Mann-Whitney U test, and t test, respectively. For all statistical analyses, IBM SPSS Statistics for Windows Version 20.0 (IBM Corp, Armonk, NY) was used.

Follow-up All patients were followed-up from emergency VCE until discharge from hospital or death. Further clinical bleeding and the need of hospitalization were assessed 4 weeks after discharge from the hospital by telephone interview.

RESULTS Patients During the 26-months study period, 88 patients with acute GI bleeding were recruited for the study. Characteristics of these patients are listed in Table 1. In 68 patients (77%) EGD showed the probable source of bleeding, and endoscopic treatment was carried out if necessary. Of those, 19 patients were actively bleeding, and 49 patients showed lesions with stigmata of a recent hemorrhage. The 20 patients (23%) in whom EGD did not show a bleeding source were finally included in the study and received an emergency VCE.

Emergency video capsule endoscopy

the duodenum, and a sufficient capsule examination was performed according to the study protocol. In 1 patient, endoscopic capsule placement in the duodenum was not possible because of pyloric stenosis. In this patient, the capsule was placed in the stomach but failed to pass the pylorus during the complete capsule recording time. In 2 patients, capsule recording was stopped before the complete passage of the small bowel because definite bleeding was already detected by live view analysis. In the remaining 17 patients, a complete small-bowel transit was achieved (mean small-bowel transit time 4.6
2.2 hours).

Outcome measures Overall, emergency VCE correctly guided further diagnostic and therapeutic procedures in 17 of 20 patients (85%; 95% CI, 62%-97%) (Fig. 1). The secondary outcome measures are summarized in Table 2.

Findings of emergency VCE and further management In 15 of 20 patients (75%; 95% CI, 51%-91%), a positive finding (blood or a bleeding source) was detected by VCE (in 11 patients in the small bowel and in 4 patients in the cecum) (Table 3). Of the 15 patients with positive findings in VCE, 3 of 15 patients received push enteroscopy, and 7 of 15 patients received antegrade balloon enteroscopy; we performed argon plasma coagulation in 9 patients with angiodysplasias and epinephrine injection combined with hemoclip application in 1 patient with active jejunal ulcer bleeding. In 1 of 15 patients with severe active jejunal bleeding, which made an endoscopic approach impossible, surgical resection was carried out. Two of 15 patients with cecal angiodysplasias underwent therapeutic colonoscopy with argon plasma coagulation, and 1 of 15 patients with bleeding ulcers underwent therapeutic colonoscopy with hemoclip application. In 1 of 15 patients in whom VCE showed blood in the colon without identification of a lesion, colonoscopy was likewise not able to find a bleeding source. In this patient, EGD was repeated and revealed a small bleeding duodenal ulcer, which had been missed during the initial EGD. Five of the 20 patients who showed no positive findings in VCE underwent subsequent colonoscopy, which detected a presumed bleeding source in 3 of 5 cases (ulcer in 1 case, diverticula in 2 cases), which were all located in the left side of the colon. In the 2 remaining patients, a bleeding source was neither detected by subsequent colonoscopy nor by repeated EGD (including the patient in which VCE had failed).

Follow-up

In 19 of 20 patients (95%; 95% confidence interval [CI], 75%-99%) the capsule could be placed endoscopically in

One patient with diverticular bleeding died 9 days after VCE of a cardiac event, resulting in a mortality rate of 5%. Within the follow-up period of 4 weeks after discharge from the hospital, 1 patient with diverticular bleeding was readmitted to the hospital because of diverticular

www.giejournal.org

Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 891

Technical feasibility of emergency VCE

Emergency video capsule endoscopy

Schlag et al

TABLE 1. Characteristics of recruited patients (n [ 88) Patients without VCE (n [ 68)

Patients with VCE (n [ 20)

P value for difference

71.6
12.8

74.2
10.6

.37*

Sex, female/male

27/41

3/17

.04z

Melena/maroon stool, no. of patients

56/12

12/8

.04z

Arterial pressure, mean on presentation (mmHg), mean
SD

88
18

77
16

.01*

Heart rate on presentation, mean
SD, beats/min

87
22

81
16

.19*

Hemoglobin on presentation, mean
SD, g/dL

8.7
1.9

7.0
1.8

! .01*

2 (0-2)

2 (2-4)

.05y

1.53
.81

1.35
.40

.18*

Partial thromboplastin time on presentation, mean
SD, sec

38
20

33
7

.10*

Hemostasis-influencing drugs used, no. of patients (%)

40 (58.8)

16 (80.0)

.08z

Antiplatelet

26 (38.2)

14 (70.0)

.01z

Anticoagulant

19 (27.9)

6 (30.0)

.86z

5 (7.4)

4 (20.0)

.10z

NSAIDs used, no. of patients (%)

2 (2.9)

1 (5.0)

.54z

Comorbidities, no. of patients (%)

46 (67.7)

15 (75.0)

.53z

Chronic heart failure

9 (13.2)

4 (20.0)

.45z

Chronic kidney failure

24 (35.3)

8 (40.0)

.70z

Liver cirrhosis

13 (19.1)

3 (15.0)

.67z

Malignancy

11 (16.2)

3 (15.0)

.90z

12.4
9.0

9.8
6.6

.16*

Age, mean
SD, y

No. of units of packed red blood cells transfused prior to EGD, median (interquartile range) INR on presentation, mean þ/ SD

Both

Time from presentation to EGD, mean
SD, h

VCE, Video capsule endoscopy; SD, standard deviation; NSAIDs, nonsteroidal anti-inflammatory drugs. *t test. yMann-Whitney U test. zChi-square test.

This was the first study that prospectively evaluated the utility of emergency VCE in patients with acute severe GI bleeding immediately after negative results from upper endoscopy. We could show that emergency VCE correctly guided further diagnostic and therapeutic procedures in 85% of cases and identified the site of suspected bleeding in 75% of patients. Previous VCE-bleeding studies have focused mainly on patients with obscure GI bleeding, which is defined by bleeding that persists or recurs without an obvious etiology after negative EGD and colonoscopy results, and obscure

GI bleeding has become the most frequent indication for VCE.12,13 It has been shown that VCE has a significantly higher diagnostic yield in patients with ongoing overt obscure GI bleeding than in patients with occult obscure GI bleeding and that VCE shows the highest diagnostic yield up to 92% when performed as close as possible to the bleeding episode.14-17 However, to date, only 1 small case series and 2 retrospective studies have evaluated the utility of VCE for acute severe bleeding. In 2009, Hogan et al18 described the successful use of VCE as an early diagnostic tool in acute life-threatening GI hemorrhage in 3 patients. In a study by Lecleire et al,19 emergency VCE allowed identification of the bleeding site in 89% of 55 patients with severe obscure overt GI bleeding. In another study on 15 patients by Almeida et al,20 the etiology of bleeding was correctly diagnosed by VCE in 73% of patients. In our study, patients with acute severe GI bleeding also were enrolled. However, the main difference was that

892 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015

www.giejournal.org

rebleeding. None of the other 18 remaining patients suffered from rebleeding or had to be readmitted to the hospital.

DISCUSSION

Schlag et al

Emergency video capsule endoscopy

Screened Patients n = 88 received EGD

No upper GI bleeding n = 20

Upper GI bleeding n = 68

received emergency VCE Failed examination

Sufficient examination

n=1

n = 19

Negative findings n=4

Positive findings n = 15

Colonoscopy n=4 positive: 3 of 4

Surgery n=1 positive: 1 of 1

Push-Enteroscopy n=3 positive: 3 of 3

Balloon-Enteroscopy n=7 positive: 7 of 7

Colonoscopy n=4 positive: 3 of 4

Figure 1. Study overview.

TABLE 2. Secondary outcome measures of patients who received emergency VCE (n [ 20) Positive findings in VCE, no. of patients (%)

Small bowel 1 (5.3)

Good

13 (68.4)

Fair

4 (21.1)

Poor

1 (5.3)

Overall no. of packed red blood cells transfused, median (interquartile range) Time in intensive or intermediate care unit, mean
SD, h Length of hospital stay, mean
SD, d

No. (%) 95% CI for %

15 (75.0)

Bowel cleansing level of VCE*, no. of patients (%) Excellent

TABLE 3. Findings of emergency VCE

4 (2-6) 77.9
11.7

11 (55) 32-77

Angiodysplasia

7 (35) 15-59

Active bleeding without lesion

4 (20) 6-44

Cecum

4 (20) 6-44

Angiodysplasia

2 (10) 1-32

Ulcer

1 (5) 0-25

Blood without lesion

1 (5) 0-25

No blood and no lesion

5 (25) 9-49

VCE, Video capsule endoscopy; CI, confidence interval.

8.9
5.8

we did not include patients with obscure GI bleeding after negative EGD and colonoscopy results but had already performed VCE immediately after negative EGD results. Thus, our study evaluated VCE as a real emergency procedure because patients received VCE within 24 hours of presen-

tation, whereas in the former studies patients received VCE after approximately 4 days.19,20 We could show that emergency VCE is feasible when directly performed after negative EGD results. Endoscopic capsule placement and a sufficient capsule examination, which was defined by either complete small-bowel passage or prior detection of the bleeding site, was achieved in 95% of cases. Usually, purgative bowel cleansing is recommended before VCE in order to improve visualization quality and diagnostic yield, but there is no consensus regarding the optimal dosage of the purgative.21 In our study, we

www.giejournal.org

Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 893

Mortality, no. of patients (%)

1 (5.0)

VCE, Video capsule endoscopy; SD, standard deviation. *Cleansing level of the small bowel could be assessed in only 19 patients because of non-passage in 1 patient.

Emergency video capsule endoscopy

Schlag et al

administered only 0.5 L polyethylene glycol solution in the duodenum before endoscopic placement of the capsule. By using this protocol, we observed sufficient cleansing levels in the vast majority of patients, which allowed detection of potential bleeding without any difficulties. In the routine diagnostic approach of acute GI bleeding, colonoscopy usually follows negative EGD results as the next diagnostic procedure. However, cleansing the colon of stool and blood before colonoscopy is necessary for achieving complete examination to the cecum, which might be difficult to accomplish and impedes urgent colonoscopy.22 Thus, the optimal timing of colonoscopy in GI bleeding remains uncertain, and urgent colonoscopy has shown no evidence of improving clinical outcomes in comparison to delayed colonoscopy.23 According to our study data, the time needed for preparation of the colon can optimally be used to perform emergency VCE. By using this strategy, we could identify patients with mid-GI bleeding more quickly, allowing patients to undergo prompt therapeutic interventions of the small bowel. Depending on the VCE-suspected bleeding site also, the best way of access can be chosen. There is also no crucial delay for patients requiring a colonoscopy when lower GI bleeding is assumed after VCE. This approach seems particularly reasonable if mid-GI bleeding is suspected by clinical presentation. In our study, only patients presenting with melena or maroon stool and negative EGD resultsd suggestive for mid-GI bleedingdreceived emergency VCE, whereas patients presenting with rectal fresh red blood, which is highly suggestive for lower GI bleeding, were not included. In this selected sample of patients, VCE showed a diagnostic yield of 75%. However, it has to be mentioned that in only 55% of patients the lesions detected by VCE were located in the small bowel, whereas in 20% the findings were located in the cecum, which could have been in the reach of a colonoscopy in a routine diagnostic approach. General drawbacks of VCE have to be considered, particularly when it is used in the emergency setting. Retention rates up to 5% have been reported when VCE was used in the evaluation of GI bleeding.24 Patients with known Crohn’s disease or previous intestinal surgery carry a higher risk and were excluded from our study, in which no retention was observed. Endoscopic placement by a capsule delivery device might be impeded by an altered anatomy. In our study, endoscopic capsule placement was not possible because of pyloric stenosis in one patient. Similar difficulties may exist in patients with pharyngeal or esophageal strictures or patients with histories of gastric surgery. Another limitation of VCE is its lack of possible therapeutic intervention, so it has to be mentioned that alternative procedures in patients with suspected acute mid-GI bleeding exist. Balloon-assisted enteroscopy (BAE) provides diagnostic yields similar to those of VCE,25,26 with the advantage of therapy for bleeding sites in the same session. However, total enteroscopy may be

required to exclude mid-GI bleeding, which often may be achieved only through a combination of antegrade and retrograde approaches.27 Most notably, BAE is an invasive and time-consuming procedure with increased risk of adverse events, which might be particularly harmful for critically ill patients with severe GI bleeding. Alternatively, less-invasive push enteroscopy can be discussed as a first-line procedure.28 However, for push enteroscopy, distinct lower diagnostic yields compared with VCE have been demonstrated in GI bleeding.29 Radiologic angiography also can be used for diagnosis and therapy of acute GI bleeding,30 but a lower diagnostic yield in comparison to VCE has been reported recently.31 The advantage of our applied approach, which favors VCE as the first-line procedure after negative EGD results, is to avoid unnecessary invasive and potentially harmful procedures. Depending on the VCE findings, specific therapeutic interventions can be initiated as underlined by the results of our study. We performed push enteroscopy in 15%, antegrade BAE in 35%, small-bowel surgery in 5%, and colonoscopy in 45% of patients. Overall, VCE correctly guided further diagnostic and therapeutic procedures in 85% of cases. The main limitations of our study are the small sample size and the non-randomized study design. We could show that emergency VCE performed immediately after negative EGD results is feasible and can be helpful for the early detection of the bleeding site in the selected cohort of patients with suspected mid-GI bleeding. However, only randomized controlled studies allow conclusions to be made on whether such an algorithm may improve patient clinical outcomes and help to reduce the amount of transfused blood units and whether the need for monitoring in the intensive and/or intermediate care unit can be shortened, leading to a shorter hospital stay and lower costs. Furthermore, we relied on a short follow-up period of only 4 weeks. This period was chosen to ensure that there was no other bleeding site instead of the suspected one and that no acute rebleeding occurred. However, only long-term follow-up data, which was not assessed in our study, could have given information if emergency VCE followed by immediate therapeutic interventions might have positive effects on patient long-term outcomes. In conclusion, VCE is safe and feasible in severe acute GI bleeding when performed directly after negative EGD results. VCE is not necessarily restricted to patients with chronic obscure GI bleeding and can be used in acute severe GI bleeding when mid-GI bleeding is suspected. Here, VCE shows a high diagnostic yield and is able to correctly guide further therapy in the majority of patients.

894 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015

www.giejournal.org

REFERENCES 1. Hearnshaw SA, Logan RF, Lowe D, et al. Use of endoscopy for management of acute upper gastrointestinal bleeding in the UK: results of a nationwide audit. Gut 2010;59:1022-9.

Schlag et al

Emergency video capsule endoscopy

2. Ell C, May A. Mid-gastrointestinal bleeding: capsule endoscopy and push-and-pull enteroscopy give rise to a new medical term. Endoscopy 2006;38:73-5. 3. Hwang JH, Fisher DA, Ben-Menachem T, et al. The role of endoscopy in the management of acute non-variceal upper GI bleeding. Gastrointest Endosc 2012;75:1132-8. 4. Laine L, Jensen DM. Management of patients with ulcer bleeding. Am J Gastroenterol 2012;107:345-60; quiz 361. 5. Barkun AN, Bardou M, Kuipers EJ, et al. International consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding. Ann Intern Med 2010;152:101-3. 6. Angtuaco TL, Reddy SK, Drapkin S, et al. The utility of urgent colonoscopy in the evaluation of acute lower gastrointestinal tract bleeding: a 2-year experience from a single center. Am J Gastroenterol 2001;96:1782-5. 7. Laine L, Shah A. Randomized trial of urgent vs. elective colonoscopy in patients hospitalized with lower GI bleeding. Am J Gastroenterol 2010;105:2636-41; quiz 2642. 8. Iddan GJ, Swain CP. History and development of capsule endoscopy. Gastrointest Endosc Clin N Am 2004;14:1-9. 9. Hartmann D, Schmidt H, Bolz G, et al. A prospective two-center study comparing wireless capsule endoscopy with intraoperative enteroscopy in patients with obscure GI bleeding. Gastrointest Endosc 2005;61:826-32. 10. Holden JP, Dureja P, Pfau PR, et al. Endoscopic placement of the smallbowel video capsule by using a capsule endoscope delivery device. Gastrointest Endosc 2007;65:842-7. 11. Brotz C, Nandi N, Conn M, et al. A validation study of 3 grading systems to evaluate small-bowel cleansing for wireless capsule endoscopy: a quantitative index, a qualitative evaluation, and an overall adequacy assessment. Gastrointest Endosc 2009;69:262-70, 270.e1. 12. Committee ASoP, Fisher L, Lee Krinsky M, et al. The role of endoscopy in the management of obscure GI bleeding. Gastrointest Endosc 2010;72:471-9. 13. Ladas SD, Triantafyllou K, Spada C, et al. European Society of Gastrointestinal Endoscopy (ESGE): recommendations (2009) on clinical use of video capsule endoscopy to investigate small-bowel, esophageal and colonic diseases. Endoscopy 2010;42:220-7. 14. Carey EJ, Leighton JA, Heigh RI, et al. A single-center experience of 260 consecutive patients undergoing capsule endoscopy for obscure gastrointestinal bleeding. Am J Gastroenterol 2007;102:89-95. 15. Pennazio M, Santucci R, Rondonotti E, et al. Outcome of patients with obscure gastrointestinal bleeding after capsule endoscopy: report of 100 consecutive cases. Gastroenterology 2004;126:643-53. 16. Yamada A, Watabe H, Kobayashi Y, et al. Timing of capsule endoscopy influences the diagnosis and outcome in obscure-overt gastrointestinal bleeding. Hepato-gastroenterology 2012;59:676-9.

17. Apostolopoulos P, Liatsos C, Gralnek IM, et al. Evaluation of capsule endoscopy in active, mild-to-moderate, overt, obscure GI bleeding. Gastrointest Endosc 2007;66:1174-81. 18. Hogan RB 3rd, Pareek N, Phillips P, et al. Video capsule endoscopy in life-threatening GI hemorrhage after negative primary endoscopy (with video). Gastrointest Endosc 2009;69:366-71. 19. Lecleire S, Iwanicki-Caron I, Di-Fiore A, et al. Yield and impact of emergency capsule enteroscopy in severe obscure-overt gastrointestinal bleeding. Endoscopy 2012;44:337-42. 20. Almeida N, Figueiredo P, Lopes S, et al. Urgent capsule endoscopy is useful in severe obscure-overt gastrointestinal bleeding. Dig Endosc 2009;21:87-92. 21. Rokkas T, Papaxoinis K, Triantafyllou K, et al. Does purgative preparation influence the diagnostic yield of small bowel video capsule endoscopy? A meta-analysis. Am J Gastroenterol 2009;104:219-27. 22. Lhewa DY, Strate LL. Pros and cons of colonoscopy in management of acute lower gastrointestinal bleeding. World J Gastroenterol 2012;18:1185-90. 23. Jang BI. Lower gastrointestinal bleeding: is urgent colonoscopy necessary for all hematochezia? Clin Endosc 2013;46:476-9. 24. Mishkin DS, Chuttani R, Croffie J, et al. ASGE Technology Status Evaluation Report: wireless capsule endoscopy. Gastrointest Endosc 2006;63:539-45. 25. Pasha SF, Leighton JA, Das A, et al. Double-balloon enteroscopy and capsule endoscopy have comparable diagnostic yield in small-bowel disease: a meta-analysis. Clin Gastroenterol Hepatol 2008;6:671-6. 26. Teshima CW, Kuipers EJ, van Zanten SV, et al. Double balloon enteroscopy and capsule endoscopy for obscure gastrointestinal bleeding: an updated meta-analysis. J Gastroenterol Hepatol 2011;26:796-801. 27. Manabe N, Tanaka S, Fukumoto A, et al. Double-balloon enteroscopy in patients with GI bleeding of obscure origin. Gastrointest Endosc 2006;64:135-40. 28. Foutch PG, Sawyer R, Sanowski RA. Push-enteroscopy for diagnosis of patients with gastrointestinal bleeding of obscure origin. Gastrointest Endosc 1990;36:337-41. 29. de Leusse A, Vahedi K, Edery J, et al. Capsule endoscopy or push enteroscopy for first-line exploration of obscure gastrointestinal bleeding? Gastroenterology 2007;132:855-62; quiz 1164-5. 30. Walker TG, Salazar GM, Waltman AC. Angiographic evaluation and management of acute gastrointestinal hemorrhage. World J Gastroenterol 2012;18:1191-201. 31. Leung WK, Ho SS, Suen BY, et al. Capsule endoscopy or angiography in patients with acute overt obscure gastrointestinal bleeding: a prospective randomized study with long-term follow-up. Am J Gastroenterol 2012;107:1370-6.

www.giejournal.org

Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 895