[Capsule endoscopy and obscure gastrointestinal bleeding: does the form of presentation matter?]

Obscure gastrointestinal bleeding (OGIB) is defined as bleeding from the gastrointestinal tract with no obvious cause after assessment with upper and lower gastrointestinal endoscopy. In these cases, the source is suspected to be in the small bowel. Obscure bleeding can be occult or overt. The aim of this study was to analyze the clinical and analytical characteristics and findings on capsule endoscopy in patients with OGIB and to determine the factors related to the detection of lesions in both forms of presentation.

We performed a retrospective study of capsule endoscopies carried out between November 2009 and November 2012 for OGIB.

We analyzed 284 capsule endoscopies in 272 patients. Initially, 12 procedures could not be evaluated and were repeated. A total of 272 procedures were finally included in the analysis. The results of 114 (41.9%) capsule endoscopies were normal. Compared with patients with occult OGIB, those with overt OGIB were significantly older (70.2 vs. 67.5 years; p = 0.04), consumed more NSAID (24.2% vs. 11.9%; p = 0.01), had higher hemoglobin levels (9.3 vs. 10.4; p < 0,001) and more frequently required transfusion (64.5% vs 32.2%; p < 0.001). No differences were found between the two forms of presentation in the detection of canker sores-ulcers and polyps-masses. Vascular lesions were more frequently detected in overt than in occult OGIB (40.3% vs. 25.7%, respectively), (p < 0.05). When the total number of diagnoses carried out by capsule endoscopy was analyzed, no differences were found in diagnostic yield between overt OGIB (57%) and occult OGIB (54%), (p = 0.6). In overt OGIB, multivariate analysis showed that the variables that significantly predicted the detection of lesions on capsule endoscopy were consumption of medication NSAID (OR 2.75; p = 0.01), antiplatelets and anticoagulants (OR 2.64; p = 0.03) and analytical data hemoglobin (OR 3.23; p < 0.001) and INR (OR 1.8; p = 0.02). In occult OGIB, multivariate analysis showed that the factors significantly related to the detection of lesions on endoscopy were age (OR 1.9; p = 0.04) and NSAID consumption (OR 2.1; p = 0.01).

Capsule endoscopy is essential in the assessment of OGIB. Although the diagnostic yield was similar in both forms of presentation, vascular lesions were more frequently detected in overt OGIB. The diagnostic yield of capsule endoscopy could be optimized by taking into account the form of presentation (overt vs. occult) and certain clinical and analytic data (age, drug consumption, hemoglobin).