S1272 Yield of Esophagogastroduodenoscopy After Positive FOBT and Negative Colonoscopy: A Systematic Review and Meta-Analysis

Introduction: The role of esophagogastroduodenoscopy (EGD) in patients with a positive fecal occult blood test (FOBT) and negative colonoscopy remains unclear. We conducted a meta-analysis to assess the diagnostic yield of EGD in this patient population. Methods: We systematically reviewed studies reporting outcomes of EGD in patients with positive FOBT and normal colonoscopy, from PubMed, Embase and Scopus. Pooled prevalence and 95% confidence intervals (CI) were calculated using a fixed-effect model. Subgroup analyses were performed based on lesion type (e.g., ulcers, vascular, erosive, neoplastic). Results: Nineteen studies comprising a total of 7,773 patients were included. The mean age was 59.5 ± 5.6 years, and 50.6% females. The pooled prevalence of significant EGD findings in patients with positive FOBT but negative colonoscopy was 22.0% (95% CI: 20.8–23.4). In subgroup analysis, ulcerative lesions were the most frequently identified findings, with a pooled prevalence of 6.6% (95% CI: 6.1–7.2), including erosions (n = 85), stomach ulcers (n = 256), duodenal ulcers (n = 64), and esophageal ulcer/esophagitis (n = 195). Inflammatory lesions had a prevalence of 5.7% (95% CI: 5.2–6.2), predominantly gastritis (n = 455), gastropathy (n = 21), and duodenitis (n = 40). Vascular lesions were found in 1.1% (95% CI: 0.9–1.4), including gastroesophageal varices (n = 51), vascular ectasias (n = 18), and vascular malformations (n = 35). Upper gastrointestinal cancers were uncommon, with a pooled prevalence of 0.67% (95% CI: 0.53–0.87) across oral, esophageal, gastric, small bowel, and unspecified cancers. Barrett’s esophagus was identified in 0.64% (95% CI: 0.50–0.83) of patients. Structural findings such as esophageal strictures and other miscellaneous lesions were less common. Lesions unlikely to bleed, adenomas, and polyps were classified as “Other” and had a pooled prevalence of 2.7% (95% CI: 2.4–3.1). Egger’s test revealed no evidence of publication bias (P = 0.60). Conclusion: EGD identified clinically relevant pathology in approximately 1 in 5 patients with positive FOBT and negative colonoscopy. Malignancy was reported in approximately 7 in 1000 patients. These findings seem to support consideration of EGD in select patients, particularly those with additional risk factors or unexplained anemia.