Variation in Diagnostic Yield of Back-to-Back Capsule Endoscopy in Obscure GI Bleeding: Preliminary Results

摘要

Objective: Although capsule endoscopy (CE) has revolutionized the evaluation of the small bowel, there are limited studies comparing the diagnostic yield of repeated CE. Missed lesions at CE may lead to unnecessary invasive diagnostic procedures such as intraoperative enteroscopy. The objective of this study is to prospectively compare the diagnostic yield of two back-to-back CE studies in patients presenting with obscure GI bleeding. Methods: Patients referred to our institution for CE to evaluate obscure GI bleeding were offered enrollment. From July 5, 2005 to October 31, 2005, eighteen consenting patients were enrolled. Fifteen patients completed the study. Four patients had obscure overt GI bleeding and the remainder had obscure occult GI bleeding. Of the 3 patients who did not complete the study, one required endoscopic placement of the capsule in the small bowel, and two did not return for the second capsule. For each patient, two CE studies were performed within 4-days. There were no complications. Based on prior results, which revealed acceptable interobserver variability, we assigned the two CE studies to independent readers. If the findings between the two studies varied, a third reader reviewed both capsule studies to confirm positive findings. Results: Of the 15 paired studies, eight were both normal. 3/15 of the first capsule studies detected new lesions that were not-seen on the second capsule study. Of those three, 2 were determined by the readers to. be clinically significant findings. 6/15 of the second capsule studies detected new lesidrts- that were not seen on the first capsule study. Of those six, 5 were determined to be clinically significant findings. Clinically significant lesions included angioectasia, erosions, and ulcers. Conclusion: This study suggests that there may be clinical utility in repeating a non-diagnostic CE study. A larger sample size is required to determine statistical significance. Patient recruitment is ongoing.This study is funded by a 2005 ASGE Research & Outcomes & Effectiveness Award.