Interobserver Variation among Gastroenterologists with “PillCam” Esophageal Capsule

Abstract

Purpose: Gastro-esophageal reflux disease (GERD) is a common entity. Erosive esophagitis, esophageal ulcers and Barrett's esophagus are found in 10 – 20% of patients with heartburn. As patients with Barrett's esophagus have 0.5% per patient-year risk of developing esophageal adenocarcinoma, endoscopy is recommended for surveillence. Although endoscopy is safe, there are rare risks including perforation and complications from sedation. A new ingestible “PillCam” esophageal capsule recently approved for the evaluation of esophageal disease may offer an alternative office-based approach to visualize the esophagus without sedation. However, unlike endoscopy which allows repeditive movements by the endoscopist to review an area, the esophageal capsule passes capturing selected images that are reviewed at a later period of time. The ability of gastroenterologists using the esophageal capsule to define pathology has not clearly been established. In addition, interpretation of the images may vary. Methods: In order to determine the accuracy and assess for interobserver variation of esophageal capsule endoscopy in the evaluation of GERD, 6 patients underwent esophageal capsule endoscopy within 24 hours of standard endoscopy, which served as the “gold standard”. The images were then reviewed by four gastroenterologists who were unaware of the original endoscopy findings. The gastroenterologists were asked to report the degree of esophagitis (LA Class), the presence of esophageal ulcers, Barrett's esophagus, hiatal hernia and nodules/submucosal masses. Results: Six patients were included in the study allowing 24 esophageal capsule reports to be collected from the gastroenterologists. Of the six patients included, 3 had normal exams, 2 had Barrett's esophagus and 1 had erosive esophagitis/esophageal ulcers. All four gastroenterologists were able to identify the presence of Barrett's esophagus and esophageal ulcers. However, there was no correlation in the degree of esophagitis (LA Class), the length of Barrett's esophagus and the size of the hiatal hernia. In addition, a nodule in Barrett's mucosa was not appreciated by 2/4 gastroenterologists. Conclusions: Although the “PillCam” esophageal capsule can assist in the identification of Barrett's esophagus, significant limitations exist which may effect the accuracy of this new technology, including difficulty in assessing the length of the hiatal hernia, the length of Barrett's mucosa and the degree of esophagitis.