Comparative Study of the Magnetically Guided Capsule Endoscopy Versus High Definition Gastroscopy

The sudy takes place in a private hospital (Institute Arnault Tzanck, St. Laurent du Var) with an annual number of 4000 upper GI endoscopies. Patients will be recruited from this center and from the cooperating Department of Gastroenterology at the University of Nice by about half. After given written informed consent at least 24 hours prior the examination patients are included into the study. The examiners are blinded to the patients origin.

Participants will be recruited in two groups A) enriched population group (n=50): In both institutions, patients presenting for upper GI endoscopy with suspected or diagnosed focal lesions as follows; gastric tumors, submucosal tumor, gastric ulcer, polyp a)adenoma, b)hyperplastic polyp or angioectasia,ulcers for diagnostic confirmation, follow-up or planning of therapy will be informed and asked to participate in the study.

B) standard gastroscopy group (n=150): consecutive patients without prior gastroscopy presenting with the main indications reflux, upper abdominal complaints, anemia and anorexia. Here, recruitment is based on presenting symptoms.

6 endoscopists experienced in upper GI endoscopy (> 1000 examinations) and specifically trained in capsule gastroscopy (simulator, 10 gastric capsule cases) will perform capsule gastroscopy. 2 well trained endoscopists (> 1000 upper GI endoscopies), different from the capsule gastroscopists will perform subsequent gastroscopies. They receive standard information about gastroscopy indication from a list of indications. They are accompanied by a study nurse who knows patient details but ascertains examiner blindness and takes care of secondary unblinding during gastroscopy.

Gastroscopy is always performed after MGCE in this study with a maximum delay of 1 day but a minimum delay of 4 h due to water filling of the stomach with the capsule examination.

The study is a comparative study of capsule gastroscopy accuracy with conventio¬nal gastroscopy serving as gold standard as described above. Disease prevalence is enriched to include 25-30% of significant pathology such as tumors and ulcers. Confidence intervals will be calculated, e.g. for a disease prevalence of about 30% and 200 study cases, a sensitivity of 85% will have a 95% CI of 73-93%. Cross tabulation of the results of the index tests (including indeterminate and missing results) will be performed by the results of the reference standard. Study statistics will be done by Prof. Dr. Karl Wegscheider, University Hospital Hamburg/Eppendorf.

All patient data will be collected at Institut Arnault Tzanck in St. Laurent du Var. Data will be analyzed in pseudonymity way (code, age, sex) at the University of Hamburg. Original data will be stored for 10 years at Institut Arnault Tzanck in St. Laurent du Var

Inclusion criteria

• Patients with gastric focal lesions (Group A)
• Patients with gastric symptoms and indicated for Upper GI endoscopy (Group B)
• Age of 18 years to 75 years"

Exclusion criteria

• Patients with dysphagia or symptoms of gastric outlet obstruction
• Patients with suspected or known intestinal stenoses
• Overt GI bleeding
• Known large and obstructing tumors of the upper GI tract
• Patients after upper GI surgery or abdominal surgery altering GI anatomy
• Patients under full anticoagulation
• Patient in poor general condition (ASA III/IV)
• Patients with claustrophobia
• Patients using equipment that may be affected by radio transmission
• Patients using equipment that may be affected by magnetic field
• Pregnancy or suspected pregnancy "