High-resolution colonoscopy and chromoendoscopy for classifying colonic polyps: A multicenter study

C. Y. Kim, D. E. Fleischer, R. A. Kozarek, D. L. Carr-Locke, T. C.M. Li, C. J. Gostout, S. J. Heller, E. A. Montgomery, F. H. Al-Kawas, M. I. Avigan, G. M. Eisen, J. H. Lewis, S. B. Benjamin

Research output: Contribution to journalArticlepeer-review

Abstract

Background: Visual, non-biopsy techniques for predicting histology of diminutive colorectal polyps have not proven reliable. We recently reported using an affordable conventional high-resolution colonoscope (HRC), the Fujinon EC-200LR, in combination with indigo carmine dye (ICD) for the diagnosis of small polyps. The sensitivity of our technique in distinguishing adenomatous from non-adenomatous polyps was 96%, and the negative predictive value was 96%. We now report on preliminary results of our five center trial (4 academic GI units and 1 primary care practice). Methods: Complete colonoscopy or flexible sigmoidoscopy using the Fujinon 200 series scopes (EC-200LR or ES-200ER) was performed in patients with polyps. All polyps <10mm were sprayed with 0.8% ICD. Histologic diagnosis was predicted based on the polyps's surface architecture ("pits" = hyperplastic (HP), "sulci" = adenomatous (AP)). All polyps were removed. Polyps seen and stained on flexible sigmoidoscopy were removed on a followup colonoscopy. Results: 120 polyps w/a mean diameter of 4.6mm were evaluated from 68 pts. 10 polyps were excluded because no prediction could be made. Actual Histology Predicted Histo:AP Predicted Histo:HP Diag Accuracy AP (N=42) 42 1 98% HP (N=39) 7 34 83% OT (N=25)* 4 22 NA * OT (Other) = colonic mucosa or lymphoid aggregate When attempting to separate AP from non-AP (HP+Other), this technique has a sensitivity of 98%, a specificity of 84%, and a negative predictive value of 98% (i.e. if a polyp is predicted to be hyperplastic, there is a 98% (56/57) likelihood that it is not an adenomatous polyp). Conclusions: Using a conventional HRC in combination with ICD, it is possible to separate non-AP from AP without biopsy. Since it is generally agreed that hyperplastic polyps do not require removal and are not markers for neoplastic diseases in the more proximal colon, these findings suggest that important reductions in the cost of colon cancer screening may be possible.

Original languageEnglish (US)
Pages (from-to)AB110
JournalGastrointestinal endoscopy
Volume45
Issue number4
DOIs
StatePublished - 1997
Externally publishedYes

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Gastroenterology

Fingerprint

Dive into the research topics of 'High-resolution colonoscopy and chromoendoscopy for classifying colonic polyps: A multicenter study'. Together they form a unique fingerprint.

Cite this