• Department of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvani 15213-2582, USA. rschoen@pitt.edu

BACKGROUND & AIMS: Observational screening of the colon with subsequent referral for colonoscopy raises questions about the threshold of polyp size that necessitates referral. To examine the yield at colonoscopy when a given size lesion is observed, we assessed the yield of advanced adenoma and cancer at colonoscopy based on the size of the abnormality detected at flexible sigmoidoscopy (FSG).

METHODS: We used data from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial, a randomized, controlled, community-based study of FSG.

RESULTS: Subsequent colonoscopy was performed on 10,850 subjects (60.4% male; mean age, 62.9 years) with a polyp visualized on screening FSG. For women with a polyp 0.5-0.9 cm on FSG (n = 1426), the yield in the distal colon on colonoscopy was 0.6% for cancer (number needed to screen [NNS] = 166) and 14.5% for advanced adenoma (NNS = 7). In men (n = 2183), the yield was 0.7% (NNS = 142) for cancer and 15.9% (NNS = 6) for advanced adenoma. Among persons with polyps 0.5-0.9 cm identified on FSG, 5.5% (198/3609) had distal advanced adenomas that measured <1.0 cm but had villous histology or high-grade dysplasia, and 9.9% (357/3609) had adenomas > or =1 cm.

CONCLUSIONS: The yield for a distal advanced adenomatous lesion when a polyp 0.5-0.9 cm is observed at FSG is substantial and is due to the presence of advanced histology in polyps <1 cm and to detection of polyps that measure > or =1.0 cm on colonoscopy. Establishing thresholds for observation versus evaluation will require careful assessment of the overall yield.