Interobserver agreement in the histologic diagnosis of colorectal polyps. the experience of the multicenter adenoma colorectal study (SMAC)

J Clin Epidemiol. 2003 Mar;56(3):209-14. doi: 10.1016/s0895-4356(02)00587-5.

Abstract

Current clinical practice guidelines for patients with colorectal polyps are mainly based on the histologic characteristics of their lesions. However, interobserver variability in the assessment of specific polyp characteristics was evaluated in very few studies. The purpose of this study was to evaluate the interobserver agreement of four pathologists in the diagnosis of histologic type of colorectal polyps and in the degree of dysplasia and of infiltrating carcinoma in adenomas. A stratified random sample of 100 polyps was obtained from the 4,889 polyps resected within the Multicentre Adenoma Colorectal Study (SMAC), and the slides were blindly reviewed by the four pathologists. Agreement was analyzed using kappa statistics. A median kappa of 0.89 (range 0.79-1.0) was estimated for the interobserver agreement for the diagnosis of hyperplastic polyp vs. adenoma. The agreement in the diagnosis of tubular, tubulovillous, and villous type, was given by median kappa values of 0.50, 0.15, and 0.36, respectively. The median kappa for the diagnosis of infiltrating carcinoma was 0.78 (range 0.73-0.84). Agreement on diagnosis of adenoma histologic subtypes, degrees of dysplasia, or infiltrating carcinoma in adenoma was moderate. A simpler classifications might help to better identify patients at different risk of colorectal cancer.

Publication types

  • Multicenter Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adenocarcinoma / pathology
  • Adenoma / pathology*
  • Adult
  • Aged
  • Cohort Studies
  • Colorectal Neoplasms / pathology*
  • Disease Progression
  • Humans
  • Hyperplasia
  • Intestinal Polyps / pathology*
  • Middle Aged
  • Observer Variation
  • Risk Assessment