CAPSULE ENDOSCOPY WITHOUT LESIONS SIGNIFICANT FINDINGS IN OBSCURE BLEEDING – IS POOR PREPARATION ASSOCIATED WITH GREATER BLEEDING RECURRENCE?

Aims: A false negative capsule endoscopy (CE) performed for obscure gastrointestinal bleeding (OGIB) can occur in about 20 – 30% of patients. Poor bowel preparation may impair CE's accuracy. The aim of this study was to evaluate bleeding recurrence in patients with OGIB with EC without significant findings, taking into account the bowel preparation. Methods: Retrospective study that included CE performed 2010 – 2014 for OGIB and with negative significant results (CE without findings or P0/P1 lesions according to Saurin classification). Bleeding recurrence was defined as need for transfusional support, presence of melena/haematochezia or haemoglobin drop of 2 g/dL. Bowel preparation was classified according to qualitative scale of Brotz. Results: Four hundred fifty-nine CE were evaluated and 86 were included (64% female, mean age 67 years). Of these patients 12% had manifest OGIB and 88% had occult OGIB. The CE showed no lesions in 63%, P0 in 7% and P1 in 30%. The CE preparation was rated as excellent in 7%, good in 29%, fair in 36% and poor in 28%. 15% of the patients had bleeding recurrence, which was not related with bowel preparation (p = 0.8). 12% of patients performed another method for the study small bowel and significant findings were found in 3 patients (2 angiectasias and 1 gastrointestinal stromal tumor) – these patients presented in the initial CE good or fair bowel preparations. Conclusions: In this sample, patients with CE with negative significant findings the quality of the preparation did not interfere in the bleeding recurrence that was 15%.