Endoscopy 2017; 49(04): 342-350
DOI: 10.1055/s-0042-121005
Original article
© Georg Thieme Verlag KG Stuttgart · New York

Virtual chromoendoscopy (I-SCAN) detects more polyps in patients with Lynch syndrome: a randomized controlled crossover trial

Raf Bisschops
1  Department of Gastroenterology, University Hospital Leuven, Leuven, Belgium
,
Sabine Tejpar
1  Department of Gastroenterology, University Hospital Leuven, Leuven, Belgium
,
Hilde Willekens
1  Department of Gastroenterology, University Hospital Leuven, Leuven, Belgium
,
Gert De Hertogh
2  Department of Pathology, University Hospital Leuven, Leuven, Belgium
,
Eric Van Cutsem
1  Department of Gastroenterology, University Hospital Leuven, Leuven, Belgium
› Author Affiliations
Further Information

Publication History

submitted 08 December 2015

accepted after revision 06 October 2016

Publication Date:
20 January 2017 (eFirst)

Abstract

Background and study aims (Virtual) chromoendoscopy detects more polyps than standard white-light endoscopy in patients with Lynch syndrome. Previous back-to-back trials did not randomize for the examination order, creating a possible bias in favor of chromoendoscopy. We aimed to assess the difference in polyp detection between high definition white-light endoscopy (HD-WLE) and virtual chromoendoscopy with I-SCAN in patients with Lynch syndrome.

Patients and methods In this prospective, controlled trial, patients were randomized to either HD-WLE followed by I-SCAN (Group 1; n = 31) or I-SCAN followed by HD-WLE (Group 2; n = 30). Polyps found during the first pass were removed. The primary end point of the study was the difference in adenoma detection between HD-WLE and I-SCAN, expressed as the miss rate for adenomas for each technique.

Results In Group 1, I-SCAN detected four additional patients with at least one adenoma, whereas HD-WLE did not increase the adenoma detection rate in Group 2 (relative risk [RR] 0.4; P = 0.08). In Group 1, five adenomas were detected and removed with HD-WLE and a second pass with I-SCAN detected a further eight adenomas. In Group 2, I-SCAN detected 15 adenomas and subsequent HD-WLE detected 2 additional adenomas. The adenoma miss rate was significantly higher for HD-WLE (62 %) compared with I-SCAN (12 %; RR 0.44, 95 % confidence interval [CI] 0.21 to 0.87; P = 0.007). The miss rate for lesions was 57 % and 24 %, respectively, and was significantly different in favor of I-SCAN (RR 0.54, 95 %CI 0.3 to 0.85; P = 0.005). The mean inspection time in both groups was not significantly different during first (485 vs. 536 seconds; 95 %CI – 139.91 to 33.34) or second pass (421 vs. 387 seconds; 95 %CI – 32.24 to 104.89).

Conclusion Our data suggest that virtual chromoendoscopy with I-SCAN reduces the adenoma and polyp miss rate in patients with Lynch syndrome, independently of inspection time.

Trial registration: ClinicalTrials.gov (NCT01823471).