The Accuracy of Video Capsule Endoscopy to Detect Upper Gastrointestinal Hemorrhage in the Emergency Department: A Systematic Review and Meta-Analysis

Gastrointestinal (GI) hemorrhage is a common presentation in US emergency departments (EDs), accounting for about 243,675 visits in 2014. The standard of care for patients presenting to the emergency room with symptoms of Upper Gastrointestinal Bleedings (UGIB) is admission to the hospital, followed by an in-patient esophagogastroduodenoscopy (EGD). Due to the inadequacy of current risk stratification tools, nasogastric lavage and Glasglow Blatchford Score, majority of the patients are inappropriately triaged for admission, resulting in increased patient load and cost of healthcare. Video Capsule Endoscopy (VCE) is an emerging technology which conducts live readings of patient\u27s gastrointestinal tract, thereby allowing detection of any lesions. This systematic review aims to synthesize data from pre-existing studies in the ED, to assess the accuracy of VCE in detecting upper gastrointestinal hemorrhage compared to the gold standard, EGD. A comprehensive systematic search following the PRISMA criteria for Diagnostic Tool Accuracy was conducted on PubMed, Scopus, and Cochrane CENTRAL independently by two authors. This process yielded 40 studies, and based on the inclusion and exclusion criteria, five studies were included in the review and meta-analysis, comprising of 193 patients total. All the studies must have occurred in the ED focusing on VCE as the index test and EGD as the reference test for the detection of UGIB. Further, data such as patient characteristics, type of study, sample size, sensitivity, specificity, and time to EGD was extracted from the studies. Methodological quality for the studies was found to be fair using the QUADAS-2 tool. The summary estimates of sensitivity and specificity were 0.724 and 0.748, respectively. The DOR was 6.293 (95% CI: 3.23-12.25), and using the Bivariate Model, the AUC was 0.782. This review was limited in part due to the quality and quantity of studies included. Three of the five studies included in the review had small sample sizes (less than 35 patients). Further, EGD should be ideally performed immediately after VCE due to the dynamic nature of gastrointestinal lesions; however, the variability in time may reflect the discrepancy between the two modalities. This review suggests that VCE has the potential to serve as a better risk stratification tool in the ED for UGIB; however, large-scale randomized trials will be needed to better understand its accuracy