Review Article

The Diagnosis and Investigation of Obscure Gastrointestinal Bleeding

K. Liu; A. J. Kaffes


Aliment Pharmacol Ther. 2011;34(14):416-423. 

In This Article

Approach to OGIB

In the evaluation and treatment of OGIB, capsule endoscopy and double balloon enteroscopy are considered complementary procedures.[25,27,55–57] The inability of capsule endoscopy to obtain biopsies or administer therapy is made possible with DBE. Conversely, the low rates of achieving total enteroscopy in patients undergoing DBE is remedied by CE which has complete examination rates of 83.6% in the setting of OGIB.[6] Furthermore, Bar-Meir et al. reported that 20–35% of patients with severe IDA (Hb < 10 g/dL) and negative initial CE had abnormalities detected on a second CE.[56] These lesions can be detected (and treated) by a DBE, which can discover an additional 30% of patients with OGIB when initial CE was negative.[26] A suggested algorithm for evaluation of OGIB is shown in Figure 1.

Figure 1.

Proposed approach to diagnosis and management of obscure gastrointestinal bleeding. DAE, device assisted enteroscopy; PE, push enteroscopy; CE, capsule endoscopy; routine endoscopy = oesophagogastroduodenoscopy and/or colonoscopy. * This direction compared with capsule endoscopy is preferred in many countries where CE is not readily available or is too expensive for routine use. This pathway has also been shown in clinical studies to be more effective in high volume centres.62

Unless contraindicated, CE is usually the initial diagnostic test in haemodynamically stable patients with suspected OGIB because of its minimally invasive nature, tolerance and ability to visualise the entire small bowel. DBE is indicated if CE detects a lesion requiring biopsy or endoscopic intervention or in patients whom suspicion of small bowel bleeding is high despite a negative initial CE.[25,27,31,55,57–59] This approach leads to a resolution of bleeding and normal Hb in greater than 75% of patients and also a reduction in transfusion and iron requirements.[60] However, from a cost minimisation perspective, initial DBE is the least expensive strategy when the need for therapeutic intervention or definitive diagnosis is highly probable (>25–30%).[61,62] If only visual identification is required, initial CE may be preferred.[61] A recent study by Albert et al. suggested that an initial DBE strategy is more cost effective in high volume centres (>80–100 investigations per annum) where a substantial number of patients present with small bowel bleeding.[62]

As the majority of lesions responsible for IDA are located in the proximal small bowel, it is reasonable to start with an anterograde DBE, unless other investigations show a lesion beyond the proximal two-thirds to three-quarters of the small bowel.[60,63] However, Hakamura et al. point out that there is currently no consensus for choosing the DBE route.[64]

Currently no clear guidelines exist for further investigation of patients with a negative initial CE. It is reasonable to observe clinically stable patients and treat medically with iron therapy, if necessary. However, patients with evidence of ongoing or recurrent OGIB (e.g. overt bleeding, iron deficiency anaemia or positive FOBT) should have further work-up. The options include repeating routine endoscopies, repeating the CE, performing radiographic or nuclear medicine scans, angiography, DBE, PE or even intraoperative enteroscopy. Which option to pursue should be decided on a case-by-case basis determined by the clinical scenario, diagnostic yield, risks involved, availability and patient preference. Repeat oesophagogastroduodenoscopy and/or colonoscopy should be considered in patients with ongoing overt bleeding or if there is a suspicion of a missed lesion on the initial examination because of suboptimal visibility or bowel preparation.

Patients with acutely bleeding lesions should undergo a therapeutic endoscopic procedure (e.g. PE or DBE) or angiography +/− embolisation depending on local availability and expertise. These should be performed only after appropriate resuscitation has been implemented.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: