Blinded nonrandomized comparative study of gastric examination with a magnetically guided capsule endoscope and standard videoendoscope

Background: Passive video capsule endoscopy is the criterion standard for small-bowel exploration but cannot be used for the large gastric cavity. We report the first blinded comparative clinical trial in humans comparing a magnetically guided capsule endoscope (MGCE) and a conventional high-definition gastroscope. Objective: To assess the potential of gastric examination with a guided capsule. Design: Blinded, nonrandomized comparative study. Setting: Single endoscopy center. Methods: The trial involved 61 patients included in a blinded capsule and gastroscopy comparative study. MGCE examination was performed 24 hours after patients had undergone gastroscopy. To remove food residue or mucus, patients drank 900 mL of water in 2 portions. Then to provide the air-water interface required by the guidance system, they drank 400 mL of water at 35°C. Results: Visualization of the gastric pylorus, antrum, body, fundus, and cardia was evaluated as complete in 88.5%, 86.9%, 93.4%, 85.2%, and 88.5% of patients, respectively. Of gastric lesions, 58.3% were detected by both gastroscopy and MGCE at immediate assessment and review of recorded data. Capsule examination missed 14 findings and gastroscopy missed 31 findings seen with MGCE. Overall diagnostic yield was similar for both modalities. Limitation: Pilot study. Conclusions: Diagnostic results were similar for the 2 methods. After some technical difficulties related to gastric expansion or presence of mucus had been overcome, this study opened a new field for noninvasive gastric examination in countries where high gastric cancer incidence demands a screening tool.