Timing of enteroscopy in overt-obscure gastrointestinal bleeding - a systematic review and meta-analysis.

BACKGROUND the impact of early enteroscopy on the outcome of overt-obscure gastrointestinal bleeding (OGIB) is still unclear. Our aim was to evaluate the impact of early enteroscopy on overt-OGIB. METHODS the PubMed-MEDLINE, Web of Science, and Scopus databases were systematically reviewed. Observational retrospective studies comparing early versus non-early enteroscopy in overt-OGIB were identified. Data on diagnosis, treatment, and rebleeding were extracted from each study, and a meta-analysis was performed. RESULTS fifteen studies (comprising 1,907 patients) were included. Early enteroscopy was performed in 470 patients and non-early enteroscopy in 1,437 patients. Early enteroscopy was associated with a significantly higher diagnostic yield (odds ratio [OR] = 3.2, 95 % CI: 1.9-5.3; p = 0.002) and therapeutic yield (OR = 4.9, 95 % CI: 1.2-20.5; p = 0.03). However, moderate and high heterogeneity was observed in both analyses (DY I2 = 60.4 %; p = 0.002; TY I2 = 83.1 %; p < 0.001). When considering only studies where enteroscopy was performed during ongoing bleeding or within ≤ 24 h, ≤ 48 h, and ≤ 72 h of bleeding, heterogeneity was removed while the positive effect on diagnostic yield was maintained (OR = 4.7, 95 % CI: 3.4-6.6, p < 0.001, I2 = 0 %). Early enteroscopy did not significantly influence rebleeding rate (OR = 0.87, 95 % CI: 0.40-1.89, p = 0.72) in our analysis. CONCLUSIONS in conclusion, early enteroscopy, especially when performed during ongoing bleeding or within 24 h, 48 h or 72 h of the bleeding episode, may increase diagnostic yield. Although an effect on therapeutic yield was observed, the value of early intervention has to be cautiously evaluated due to the high heterogeneity found among results. In our meta-analysis, early enteroscopy did not significantly influence rebleeding rate.