Abstract
Background: Gastric cancer is prevalent in the Brazilian population, with diagnosis largely reliant on endoscopic biopsies. Current guidelines recommend a minimum of seven endoscopic biopsies. However, advances in high-definition imaging, chromoendoscopy technologies (such as narrow-band imaging [NBI]), and histopathological techniques suggest that fewer biopsies might suffice. Validating this hypothesis could reduce the duration of examinations, lessen the burden on pathologists, and minimize risks associated with multiple biopsies.
Methods: This prospective study evaluated the efficacy of various biopsy strategies in diagnosing gastric lesions. A total of 106 lesions from 104 patients were examined over 14 months. The study compared the diagnostic yield of the conventional seven biopsies (considered the gold standard) with a single white-light biopsy, one NBI-guided biopsy, and a combination of both white-light and NBI-guided biopsies.
Results: When compared to the gold standard of seven biopsies, the single white-light biopsy demonstrated a sensitivity of 90.1% (range 82.5-95.6%) and a specificity of 100% (range 47.8- 100%). The NBI-guided biopsy yielded a sensitivity of 83.2% (range 74.4-89.9%) and a specificity of 100% (range 47.8-100%). Combining both biopsy techniques achieved a sensitivity of 96.1% and a specificity of 100%.
Conclusions: A single white-light biopsy was found to be nearly equivalent to the gold standard of seven biopsies, using a 10% difference as the noninferiority criterion. Performing two biopsies-one conventional and one NBI-guided-provided even greater accuracy, closely approximating that of the gold standard.