Suspicious macroscopic features of small malignant colorectal polyps

Abstract

Objective. The aim of this analysis was to retrospectively review video recordings of malignant polyps <10 mm in search for suspicious macroscopic features in white light endoscopy. Methods. Database entries and recordings of screening colonoscopies from a single tertiary referral center between June 2009 and December 2012 were reviewed. Malignant polyps <10 mm were analyzed. The recordings were reviewed by two expert endoscopists in search for suspicious morphological features: irregular contours, central depression, contact bleeding, shape deformity, central depression, chicken skin sign, circumscribed area with abnormal vascular and/or surface pattern. Then, six experienced endoscopists watched the recordings in search of listed features. Next, video recordings of these malignant polyps were mixed with randomly drawn video recordings of 20 non-malignant polyps matched by size and reviewed by 14 blinded endoscopists to assess the sensitivity and specificity for the diagnosis of malignant polyps. Results. Five of the 8651 (0.06%) subjects who underwent screening colonoscopy during the study period were diagnosed with a malignant polyp <10 mm. Only one of them was ad hoc identified by performing endoscopist as suspicious. On recordings review performed by the experts, each of the four remaining polyps presented at least one suspicious macroscopic feature. Presence of these features was confirmed by experienced endoscopists. The sensitivity and specificity for the diagnosis of malignant polyp were 73.21% and 85.35%, respectively, if at least two suspicious macroscopic features defined malignant polyp. Conclusions. On careful white light endoscopy examination small malignant colorectal polyps show suspicious macroscopic features, which were frequently unrecognized by examining endoscopists.

Key Words:

• colonoscopy
• colorectal cancer
• macroscopic features
• malignant polyp

Acknowledgements

We thank the following endoscopists for participation in blinded and unblinded evaluations: J. Berrueta, Montevideo, Uruguay, H. Cedron, Lima, Peru, D. Chiodi, Montevideo, Uruguay, A.Chwiesko, Bialystok, Poland, M. De Souza, Montevideo, Uruguay, N. Gonzalez, Montevideo, Uruguay, J. Guevara, Lima, Peru, M. Howe, Montevideo, Uruguay, W. Jalocha, Łodz, Poland, B. Kotowski, Warsaw, Poland C. Olano, Montevideo, Uruguay, K. Monkemuller, Birmingham, USA, J. Pietrzak, Warsaw, Poland, A. Piscoya, Lima, Peru, A. Rocca, Montevideo, Uruguay, A. Sanguinetti, Montevideo, Uruguay, V. Valenzuela, Lima, Peru.

Declaration of interest: This study was sponsored by Medical Center for Postgraduate Education, Warsaw, Poland. The authors report no conflicts of interest.The authors alone are responsible for the content and writing of the paper.