Does Urgent Colonoscopy Improve Outcomes in the Management of Lower Gastrointestinal Bleeding?

Abstract

Colonoscopy continues to be an essential diagnostic and therapeutic tool in the management of lower gastrointestinal bleeding (LGIB). Studies that have evaluated the role of urgent colonoscopy for treating LGIB have reached conflicting conclusions. We conducted a systematic review and meta-analysis to evaluate the role of urgent colonoscopy in several outcomes in patients with LGIB. We searched Medline, Embase, Scopus and Cochrane databases from inception to July 10, 2016 for comparative studies evaluating the role of urgent versus elective colonoscopy in the management of LGIB. We evaluated mortality, rate of rebleeding, length of stay in hospital, identification of bleeding source, stigmata of recent hemorrhage and need for surgery. Pooled odds ratios (OR) were calculated for dichotomous variables whereas standard mean differences were calculated for continuous variables. We assessed quality using the Cochrane tool and Newcastle Ottawa Scale for randomized controlled trials and observational studies, respectively. We used the GRADE framework to interpret our findings. A total of 6 studies (2 randomized controlled trials and 4 observational studies) with 23,419 patients (9,498 urgent colonoscopy and 13,921 elective colonoscopy) were included in this meta-analysis. Pooled ORs with 95% CI for mortality, rebleeding and identification of bleeding source were 0.84 (0.46-1.53), 1.18 (0.64-2.16) and 1.49 (0.86-2.59), respectively. Stigmata of recent hemorrhage were more readily identified with urgent colonoscopy OR 2.85 (1.90-4.28). There were no differences in requirement for surgery, length of hospital stay or rate of endoscopic intervention. However, these effect sizes were limited by considerable heterogeneity, which was probably due to studies being conducted in different countries having different criteria for discharge and on variations in the type of endoscopic therapy for stigmata of recent hemorrhage. In conclusion, among patients with acute LGIB, there is no evidence that urgent colonoscopy reduces mortality, rebleeding or requirement for surgery or that it improves the rate of identification of the bleeding source. However, urgent colonoscopy does increase the rate of detection of stigmata of recent hemorrhage.

Keywords: Colonoscopy; Hematochezia; Lower gastrointestinal bleeding.