A multicenter randomized comparison of the Endocapsule and the Pillcam SB - 23/08/11

Doi : 10.1016/j.gie.2007.12.037

David R. Cave, MD, PhD ^⁎, David E. Fleischer, MD, Jonathan A. Leighton, MD, Douglas O. Faigel, MD, Russell I. Heigh, MD, Virender K. Sharma, MD, Christopher J. Gostout, MD, Elizabeth Rajan, MD, Klaus Mergener, MD, Anne Foley, BS, Michael Lee, BS, Kanishka Bhattacharya, MD
Current affiliations: University of Massachusetts Memorial Medical Center (D.R.C., A.F., K.B.), Worcester, Massachusetts, Mayo Clinic (D.E.F., J.A.L., R.I.H., V.K.S., E.R., M.L.), Scottsdale, Arizona, Mayo Clinic (C.J.G.), Rochester, Minnesota, Oregon Health and Science University (D.E.F.), Portland, Oregon, Digestive Disease Specialists (K.M.), Tacoma, Washington, USA

Reprint requests: David R. Cave, MD, PhD, University of Massachusetts Memorial Medical Center, 55 Lake Ave N, Worcester, MA 01655.

Worcester, Massachusetts, Scottsdale, Arizona, Rochester, Minnesota, Portland, Oregon, Tacoma, Washington, USA

Abstract

Background

Video capsule endoscopy has been shown to be the single most effective endoscopic procedure for identifying the source of obscure GI bleeding (OGIB).

Objective

Our purpose was to report on the Food and Drug Administration pivotal trial in which the Endocapsule (EC) (Olympus America, Allentown, Pa) was compared with the Pillcam SB (PSB) (Given Imaging, Yoqneam, Israel) in patients with OGIB.

Design

A novel trial design was used in which the EC and the PSB were swallowed by the same patient 40 minutes apart, in randomized order.

Setting

Four academic medical centers.

Patients

Patients with OGIB aged 18 to 85 years who had either been transfused or who had a hematocrit of <31% in males or <28% in females.

Interventions

Both video capsules were swallowed in random order. Videos from the PSB were read locally for patient management. All videos were then read by at least 2 independent readers for normal versus abnormal categorization, a diagnosis, capsule transit time, reading time, and a subjective assessment of image quality.

Main Outcome Measurement

Categorization of videos as either normal or abnormal.

Results

Data from 51 of 63 enrolled patients were analyzed. Nine patients were excluded for technical reasons and 3 for protocol violations. Twenty-four videos were read as normal and 14 as abnormal from both capsules. Disagreement occurred in 13. No adverse events were reported for either capsule. Overall agreement was 38 of 51 (74.5%) with a κ of 0.48, P = .008.

Limitations

Although ingestion order was randomized, the videos could not be read blind owing to a different shape of the image margin.

Conclusions

(1) Both devices were safe and had a comparable diagnostic yield within the range previously reported. (2) There was a subjective difference in image quality favoring the EC. (3) This study design provided unique information about capsule movement in the small intestine and the lack of electromechanical interference between 2 different capsules.

Le texte complet de cet article est disponible en PDF.

Abbreviations : FDA, MST, OGIB, EC, PSB