CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2005;3:959 –963 EDUCATION PRACTICE Obscure Gastrointestinal Bleeding: The Role of the Tagged Red Blood Cell Scan, Enteroscopy, and Capsule Endoscopy DAVID R CAVE Division of Gastroenterology, University of Massachusetts Medical Center, Worcester, Massachusetts Clinical Scenario A 65-year-old African American man was referred for obscure gastrointestinal bleeding He initially had presented months previously with anemia and melena Both upper-gastrointestinal endoscopy and colonoscopy on occasions had not shown a bleeding source A tagged technetium-99m red blood cell bleeding scan was negative He had no weight loss or abdominal pain He had received a total of 25 units of blood Physical examination was unremarkable How should this patient be evaluated? What is the role of tagged red cell nuclear scans, capsule endoscopy, angiography, and intraoperative endoscopy in patients with obscure gastrointestinal bleeding? The Problem Obscure gastrointestinal bleeding, either occult or overt, refers to gastrointestinal bleeding that persists or recurs for which a source has not been defined by a thorough upper-endoscopy and colonoscopy Figure shows the constituent types of obscure bleeding Obscure bleeding usually is considered to represent about 5% of patients with gastrointestinal bleeding However, this figure usually is based on loose definitions as to the source, such as attributing gastric erosions or the presence of colonic diverticulosis as the bleeding source, without direct visualization of active bleeding If a more rigorous definition is used in which the bleeding source is defined as one that is bleeding actively or has evidence of stigmata of recent hemorrhage then up to 24% of bleeds are considered to be of unknown origin Clearly only a small portion of gastrointestinal bleeds will become recurrent or persistent The wider availability of capsule endoscopy has started to provide more accurate data because we now can examine the entire length of the small intestine noninvasively, albeit not the entire mucosal surface The timing of endoscopy clearly is important because 80% of acute gastrointestinal bleeding stops spontaneously Aggressive colonoscopy clearly provides a higher detection rate of active lower-gastrointestinal bleeding and capsule endoscopy has been shown to have a very high yield in identifying a source of obscure bleeding with concurrent active bleeding Management Strategies and Supporting Evidence Management of obscure GI bleeding depends on a variety of factors, some patient-related and others related to personnel and facilities When the patient presents for the first time with acute gastrointestinal bleeding it is not possible to predict whether or not they have obscure bleeding A careful history and physical examination does provide useful information and often can direct the most appropriate initial examination If the presenting complaint is hematemesis, upper endoscopy clearly is the diagnostic procedure of choice Similarly, bright red rectal bleeding or the passage of maroon stools usually is investigated initially by colonoscopy Patients presenting with melena provide more of a challenge because melena can originate all the way from the upper-gastrointestinal tract to the right side of the colon A frequent compounding variable is that it often is difficult to clarify whether or not the patient has passed red blood, maroon stool, melena, or a combination Although the use of color cards to identify stool color may be of some benefit, it should be noted that stool color can vary from stool to stool in the same patient depending on the rapidity and source of bleeding Bleeding scans are reported to be able to detect bleeding at the rate of 1–.5 mL/min Published data suggest scans are positive in 26%–72% of patients However, the study designs that established these data have considerable limitations They generally not include an appropriate denominator (ie, the total number of patients tested) and are reported selectively for severe bleeding © 2005 by the American Gastroenterological Association 1542-3565/05/$30.00 PII: 10.1053/S1542-3565(05)00716-0 960 DAVID R CAVE Figure Definitions of obscure gastrointestinal bleeding FOBT, fecal occult blood testing; IDA, iron-deficiency anemia There are no studies that relate the accuracy of technetium-99m red blood cell scans to obscure gastrointestinal bleeding The accuracy for detecting bleeding in the colon is quite good, particularly if the scan is positive almost immediately after injection In the small intestine, results of the red cell scans can be very misleading because of rapid transit of the isotope in the small intestine Not infrequently the isotope pools in the cecum when bleeding has originated in the small intestine Similar limitations with false localization of the bleeding source may occur with late scanning, which may suggest a bleeding source in the colon Capsule endoscopy now is available widely for the detection of obscure gastrointestinal bleeding and is becoming the examination of choice in this context It is noninvasive and generally does not require significant bowel preparation It currently is used predominantly in the outpatient setting in patients with a history of obscure bleeding However, there is accumulating evidence that the use of capsule endoscopy during or as close to the active bleeding event as possible may show a bleeding source in up to 92% of patients actively bleeding at the time of video capsule endoscopy This suggests that capsule endoscopy should be performed as early as possible, even during the night shortly after the patient has been admitted and stabilized Such an approach may minimize the chances of bleeding ceasing before testing is initiated The recent development of the double-balloon push enteroscope now has increased greatly the diagnostic and therapeutic capabilities of push enteroscopy However, the technology is only available in a few centers and is both capital- and labor-intensive The concept behind this technology is that the alternating inflation and deflation of balloons on the end of the endoscope and on CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol 3, No 10 the end of an overtube, and the change in the relationship of the 2, allows for pleating and hence shortening of the small intestine over the endoscope, thereby minimizing loop formation The complexities of this procedure suggest that capsule endoscopy and push enteroscopy should be performed before double-balloon enteroscopy, but the latter procedure does allow for nonsurgical intervention much further into the small bowel than previously was possible without intraoperative enteroscopy Indeed, in some patients complete endoscopy of the small bowel has been reported either by the oral route or by the combination of oral and anal approaches A therapeutic version of the double-balloon scope recently has become available and is fitted with a 2.8-mm channel This allows cauterization of vascular or bleeding lesions with bipolar or argon plasma coagulation, biopsy examination of tumors, and the snaring of polyps in the small intestine The injection of India ink may be used to tattoo the point of maximal insertion so that when the ileum is intubated retrograde at the time of colonoscopy the mark can be used to confirm visualization of the entire small intestine Obviously, if the bleeding site, which may include angioectasia, tumors, and ulcers, are within range of a standard push enteroscope, then the full range of diagnostic and therapeutic procedures are available to the clinician Intraoperative enteroscopy remains the gold standard for the detection of small intestinal bleeding An appropriate endoscope, such as an enteroscope, can be inserted via the mouth or via an enterotomy and passed along the small intestine to help the surgeon localize the source of bleeding This often is traumatic to the small-bowel mucosa and is by no means 100% effective Laparoscopy-assisted intraoperative endoscopy is still in the experimental phase Areas of Uncertainty There is a clear need to define and validate a costeffective and clinically appropriate algorithm for evaluating patients with obscure gastrointestinal bleeding Should Endoscopy Be Repeated? Repeating upper-gastrointestinal endoscopy and colonoscopy when previous studies have been negative is still frequently performed for the diagnosis of obscure gastrointestinal bleeding, but the yield is low It probably is not necessary to repeat these procedures if careful and complete examinations have been performed recently by a skilled endoscopist However, the aggressive use of colonoscopy with the initiation of bowel preparation as soon as the patient has been admitted to the emergency room has been reported to increase the diagnostic yield October 2005 OBSCURE GASTROINTESTINAL BLEEDING and therapeutic opportunity for patients with colonic bleeding Capsule Endoscopy or Push Enteroscopy? Assuming that the presenting patient has had at least one careful upper endoscopy and colonoscopy and they present for the second time with evidence of blood loss, the choice of the next test ideally should be limited to either capsule endoscopy or push enteroscopy Some investigators have preferred to use push enteroscopy before capsule endoscopy because therapy can be undertaken during the procedure This is not a current option with capsule endoscopy The choice of procedures depends on the availability of each technology, the skill of the local endoscopist, and the presentation of the individual patient The diagnostic yield of push enteroscopy for obscure gastrointestinal bleeding ranges from 38% to 75% However, in the studies comparing push enteroscopy with capsule endoscopy, the yield was considerably lower Sonde enteroscopy now largely has been abandoned and the use of surgical intervention and intraoperative enteroscopy now largely is determined by the results of capsule endoscopy The most common sources of bleeding, as detected by capsule endoscopy, are listed in Table Use of Red Blood Cell Tagged Scans Before Angiography If the patient is unstable hemodynamically and thought to have a lower-gastrointestinal bleed, a gamma camera bleeding scan possibly followed by angiography is a frequently used strategy However, this does require Table Most Common Causes of Obscure Gastrointestinal Bleeding Findings % Negative findings Positive findings Angioectasia Aphthoid ulcers ϩ serpiginous ulcers Active bleeding Ulcers ϩ/Ϫ stenosis Varices Tumors Other Suspicious findings Isolated angioectasia Other Gastric lesions Ulcers Gastric antral venous ectasia 38 Data from Pennazio et al.4 21 5 3 1 961 considerable resources and is difficult to organize particularly outside normal working hours The majority of radiology departments require a bleeding scan be performed before the use of angiography In practice there are many occasions when the time taken for the performance of the bleeding scan is associated with cessation of bleeding by the time angiography is performed and the opportunity for detection of the source of bleeding is lost Angiography rarely is indicated and it is restricted to patients with severe hemorrhage Angiography requires at least mL of blood loss per minute for the detection of extravasated contrast material Pharmacologic stress testing with the local infusion of heparin, tissue plasminogen activator, and a vascular dilating agent into mesenteric vessels has been reported with a yield of provoking bleeding in about 30% of patients No adverse events were reported Obviously, any bleeding that is provoked potentially can be controlled by embolic techniques Is the Visualized Lesion the Source of Bleeding? In the absence of bleeding, the endoscopist has to make a decision as to whether the presence of an erosion, ulcer, angioectasia, or tumor may be the actual source of bleeding This may be referred to as endoscopic rationalization An example of this dilemma was a patient with melena who on endoscopic examination had a small gastric ulcer without stigmata of recent hemorrhage, large polypoid masses on colonoscopy, carcinoma of the cecum, and a 4-cm polyp in the sigmoid On capsule endoscopy, active bleeding was found to be originating in the small intestine The colonic or gastric lesion could have been rationalized as the source of bleeding if the capsule endoscopy had not been performed This type of rationalization is carried to the extreme in the colon, where the conventional colonoscopic work-up of acute lower-gastrointestinal bleeding often shows no active bleeding The presence of diverticulosis usually is regarded as the origin of the bleeding in the absence of other overt pathology Published Guidelines The American Gastroenterological Association published a technical review in 2000 based on the evaluation and management of occult and obscure gastric intestinal bleeding based on a review of data published in 1998 or earlier This review contained an extensive analysis of the evaluation of occult bleeding and of obscure bleeding Etiology, diagnostic techniques, management, and outcomes were reviewed in detail, but recommenda- 962 DAVID R CAVE tions were not specified, either descriptively or by algorithm This review also predated the introduction of capsule endoscopy and double-balloon enteroscopy by several years The development of these techniques requires us to reconsider our approach to diagnose obscure gastrointestinal bleeding Recommendations Figure shows a practical view of the management of obscure gastrointestinal bleeding and it also applies to the evaluation of both iron-deficiency anemia and overt obscure gastrointestinal bleeding This algorithm is an update to one published earlier It should be noted that conventional barium radiology, specifically a small-bowel series and enteroclysis, are not included in the algorithm because extensive data have shown that their application in the context of obscure bleeding is of minimal benefit The algorithm does not include the initial evaluation and stabilization of the patient, which are part of routine care The severity of the patient’s obscure bleeding will dictate whether or not evaluation is pursued as an outpatient or inpatient CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol 3, No 10 Case Resolution and Follow-Up Evaluation Capsule endoscopy was performed in our patient and bright red blood without specific source was found 14 minutes after the capsule had passed the pylorus, which was estimated to be in the proximal jejunum The area of active bleeding was seen on the capsule images to be coincident with a lymphangiectasia Push enteroscopy found neither active bleeding, nor another possible source of bleeding, nor the lymphangiectasia An angiogram of the superior mesenteric artery was unremarkable and showed no extravasation of dye The patient continued to bleed in an accelerated manner and underwent double-balloon enteroscopy A tiny angioectasia was found in juxtaposition to a lymphangiectasia This was treated with argon plasma coagulation No other lesion was found However, some bright red blood accumulated close to the site of initial cautery, although the source of this could not be identified The site was marked with India ink for subsequent surgical resection on the presumption that either there was an angioectasia that could not be visualized or there was a Dieulafoy’s lesion that eluded identification At laparotomy the India ink mark Figure Algorithm for the detection and management of obscure gastrointestinal bleeding TPA, tissue plasminogen activator; NTG, nitroglycerine; VCE, video capsule endoscopy; P/E, push enteroscopy October 2005 was only 15 cm beyond the ligament of Treitz A nodule was palpable, which on histology was a submucosal angioectasia The patient continued to bleed after surgery Capsule endoscopy and double-balloon enteroscopy were repeated because the ligament of Treitz had been taken down at surgery to facilitate subsequent procedures Another active bleeding site was found by both procedures 50 cm distal to the anastomosis, which was cauterized successfully No further bleeding has occurred for months Suggested Reading Zuckerman GR, Prakash C, Askin MP, et al AGA technical review on the evaluation and management of occult and obscure gastrointestinal bleeding Gastroenterology 2000;118:201–221 Vreeburg EM, Snel P, de Bruijne JW et al Acute upper gastrointestinal bleeding in the Amsterdam area: incidence, diagnosis, and clinical outcome Am J Gastroenterol 1997;92:236 –243 OBSCURE GASTROINTESTINAL BLEEDING 963 Jensen DM, Machicado GA, Jutabha R, et al Urgent colonoscopy for the diagnosis and treatment of severe diverticular hemorrhage N Engl J Med 2000;342:78 – 82 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– 653 Yamamoto H, Kita H, Sunada K, et al Clinical outcomes of double-balloon endoscopy for the diagnosis and treatment of smallintestinal diseases Clin Gastroenterol Hepatol 2004;2:1010 –1016 Cave D Video capsule endoscopy Clin Perspect Gastroenterol 2002;5:203–207 David R Cave, MD, PhD, Director of Clinical Gastroenterology Research, Division of Gastroenterology, University of Massachusetts Medical Center, 55 Lake Avenue North, Worcester, Massachusetts 01655 e-mail: caved@ummhc.org; fax: (508) 856-3981 Dr Cave has been a speaker for and has received research grants from Given Imaging, and he has received research support from Olympus Corporation ... R, et al Urgent colonoscopy for the diagnosis and treatment of severe diverticular hemorrhage N Engl J Med 2000;342:78 – 82 Pennazio M, Santucci R, Rondonotti E, et al Outcome of patients with