The objective of this study is to determine whether early video capsule endoscopy (VCE) immediately after a negative gastroscopy in the setting of suspected upper gastrointestinal bleeding (UGIB) improves patient outcomes as compared to the standard approach which typically involves performing colonoscopy prior to small bowel investigations. We aim to examine the differences in diagnostic yield, total number of procedures, length of hospitalization, mortality rates, and healthcare cost between the two groups.

UGIB is common medical emergency with an annual hospitalization rate of 67 cases per 100,000 and costing in excess of 1 billion dollars in the US.1 In Alberta, Canada, annual incidence of UGIB due to peptic ulcers alone is between 35.4 to 41.2 cases per 100,000 with an overall in-hospital mortality rate of 8.5%.2 The presentation of acute UGIB are variable, but most have visible signs of bleeding such as hematemesis, coffee-ground-emesis or melena.3 Clinically, these findings are used to guide management plans. For instance, a patient with suspected UGIB based on clinical presentation will typically undergo gastroscopy as the initial diagnostic test.4,5 However, it is becoming increasingly clear that clinical signs of bleeding do not always correlate with location of bleeds. Despite melena being the most common presenting symptom for UGIB, more than 10% of melena arise from the small bowel or right colon.3,4 Similarly, 15% of hematochezia, which are thought to be signs of colonic bleeding, in fact have bleeding sources in the upper GI tract.6 In essence, bleeding can be anywhere along the gastrointestinal tract regardless of patient's presentation. This poses a significant challenge for clinicians in determining which investigations to pursue in patients with suspected UGIB after a negative gastroscopy. Currently, there are no clear guidelines on this subject matter.4,5 In clinical practice, such patients often undergo colonoscopy to rule out colonic sources of bleed, before small bowel investigations with cross-sectional imaging and VCE. This is a time-consuming approach that requires multiple invasive procedures, leading to patient discomfort, prolonged hospitalization, and increased cost. The utility of colonoscopy in suspected UGIB have also been questioned by previous study that reported a diagnostic yield of less than 5%.7 Furthermore, proceeding with colonoscopy first will cause delays in small bowel investigation which may lower the detection rate of small bowel lesions.8 With these in mind, we propose the study of an alternative diagnostic approach which prioritizes small bowel investigation over colonic investigation. We propose early VCE following negative gastroscopy instead of colonoscopy. VCE is a pill shaped camera that is well tolerated, non-invasive, and offers superior mucosal imaging of the small bowel compared to cross-sectional imaging in the setting of GI bleeding.9 This approach would likely have higher diagnostic yield than the traditional approach of conducting colonoscopy first.

At the present, patients with suspected UGIB and negative gastroscopy pose significant diagnostic challenges for clinicians. This study aims to help determine the importance of prioritizing small bowel investigation in this clinical setting. If this study can demonstrate the safety and efficacy of the early small bowel investigation, the results will fundamentally change the approach and management of patients with acute GI bleeding, potentially leading to better patient outcomes, reduced hospitalization, and healthcare savings.

Inclusion Criteria:

• Age ≥ 18 years

• New onset of acute gastrointestinal bleeding defined as:

  1. The presence of melena/maroon stool on physical exam OR history of melena/maroon stool within 48 hours of emergency room presentation PLUS
  2. Hemoglobin drop of ≥ 20 g/L from patient's baseline

• Have capacity to consent

• Hemodynamically stable (i.e. blood pressure >100/60 or pulse <110 at the time of consent)

• Initial gastroscopy negative for active bleeding or potential bleeding source

Exclusion Criteria:

• Inability to provide consent due to lack of capacity, language barrier or other reasons
• Pregnant women
• Prior history of small bowel surgery
• Prior history of Crohn's disease
• Prior history of small bowel or colonic strictures
• Prior history of abdominal radiation