Diagnostic and therapeutic yields of early capsule endoscopy and device-assisted enteroscopy in the setting of overt GI bleeding: a systematic review with meta-analysis

• Published in: Gastrointestinal endoscopy 2022-04, Vol.95 (4), p.610-625.e9
• Main Authors: Estevinho, Maria Manuela, Pinho, Rolando, Fernandes, Carlos, Rodrigues, Adélia, Ponte, Ana, Gomes, Ana Catarina, Afecto, Edgar, Correia, João, Carvalho, João
• Format: Article
• Language: English
• Subjects: Capsule Endoscopy - methods; Gastrointestinal Hemorrhage - diagnosis; Gastrointestinal Hemorrhage - etiology; Gastrointestinal Hemorrhage - therapy; Humans; Intestine, Small - pathology
• ISSN: 0016-5107; 1097-6779
• DOI: 10.1016/j.gie.2021.12.009

Small-bowel capsule endoscopy (SBCE) and device-assisted enteroscopy (DAE) are essential in obscure GI bleeding (OGIB) management. However, the best timing for such procedures remains unknown. This meta-analysis aimed to compare, for the first time, diagnostic and therapeutic yields, detection of active bleeding and vascular lesions, recurrent bleeding, and mortality of “early” versus “nonearly” SBCE and DAE. MEDLINE, ScienceDirect, and Cochrane Central Register of Controlled Trials were searched to identify studies comparing early versus nonearly SBCE and DAE. Random-effects meta-analysis was performed; reporting quality was assessed. From 1974 records, 39 were included (4825 patients). Time intervals for the early approach varied, within 14 days in SBCE and 72 hours in DAE. The pooled diagnostic and therapeutic yields of early DAE were superior to those of SBCE (7.97% and 20.89%, respectively; P < .05). The odds for active bleeding (odds ratio [OR], 5.09; I2 = 53%), positive diagnosis (OR, 3.99; I2 = 45%), and therapeutic intervention (OR, 3.86; I2 = 67%) were higher in the early group for SBCE and DAE (P < .01). Subgroup effects in diagnostic yield were only identified for the early group sample size. Our study failed to identify differences when studies were classified according to time intervals for early DAE (I2 < 5%), but the analysis was limited because of a lack of data availability. Lower recurrent bleeding in early SBCE and DAE was observed (OR, .40; P < .01; I2 = 0%). The role of small-bowel studies in the early evaluation of OGIB is unquestionable, impacting diagnosis, therapeutic intervention, and prognosis. Comparative studies are still needed to identify optimal timing.