AIM: To identify the predictive factors for the presence of small bowel lesions in patients with obscure gastrointestinal bleeding (OGIB). METHODS: A total of 242 patients with OGIB (overt 149: occult 93) were retrospectively included in the present study. Capsule endoscopy (CE) was carried out to investigate the small bowel, and detected lesions were classified according to the P0-P2 system. Only P2 lesions were defined as significant lesions. Univariate and multivariate logistic regression analyses were carried out to define the predictive factors for the presence of small bowel lesions. RESULTS: In patients with overt OGIB, chronic kidney disease (CKD) ≥stage 4 (odds ratio [OR] 4.03; 95% confidence interval [CI] 1.45-11.1, P = 0.007) was identified as an independent predictor of the presence of vascular lesions, and a history of non-steroidalanti-inflammatory drug (NSAID) use as that of erosive/ulcerated lesions (OR 4.73; 95% CI 1.47-15.2, P = 0.009). However, in patients with occult OGIB, no significant predictors of the presence of vascular lesions were identified, whereas a history of low-dose aspirin (LDA) (OR 3.57; 95% CI 1.21-10.5, P = 0.02) and proton pump inhibitor (PPI) use (OR 3.18; 95% CI 1.02-9.92, P = 0.05) were identified as independent predictors of the presence of erosive/ulcerated lesions. CONCLUSIONS: Our results indicated that bleeding pattern and clinical characteristics could contribute to predicting the origin of OGIB.
AIM: To identify the predictive factors for the presence of small bowel lesions in patients with obscure gastrointestinal bleeding (OGIB). METHODS: A total of 242 patients with OGIB (overt 149: occult 93) were retrospectively included in the present study. Capsule endoscopy (CE) was carried out to investigate the small bowel, and detected lesions were classified according to the P0-P2 system. Only P2 lesions were defined as significant lesions. Univariate and multivariate logistic regression analyses were carried out to define the predictive factors for the presence of small bowel lesions. RESULTS: In patients with overt OGIB, chronic kidney disease (CKD) ≥stage 4 (odds ratio [OR] 4.03; 95% confidence interval [CI] 1.45-11.1, P = 0.007) was identified as an independent predictor of the presence of vascular lesions, and a history of non-steroidalanti-inflammatory drug (NSAID) use as that of erosive/ulcerated lesions (OR 4.73; 95% CI 1.47-15.2, P = 0.009). However, in patients with occult OGIB, no significant predictors of the presence of vascular lesions were identified, whereas a history of low-dose aspirin (LDA) (OR 3.57; 95% CI 1.21-10.5, P = 0.02) and proton pump inhibitor (PPI) use (OR 3.18; 95% CI 1.02-9.92, P = 0.05) were identified as independent predictors of the presence of erosive/ulcerated lesions. CONCLUSIONS: Our results indicated that bleeding pattern and clinical characteristics could contribute to predicting the origin of OGIB.
Authors: Simon Nennstiel; Annkathrin Machanek; Stefan von Delius; Bruno Neu; Bernhard Haller; Mohamed Abdelhafez; Roland M Schmid; Christoph Schlag Journal: United European Gastroenterol J Date: 2017-04-07 Impact factor: 4.623