Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
3308506 | Gastrointestinal Endoscopy | 2009 | 6 Pages |
BackgroundDespite its high sensitivity for detecting esophageal cancer, chromoendoscopy with iodine staining has low specificity and requires many biopsy specimens, which is undesirable. Dramatic color change after iodine staining (initially whitish yellow and then pink 2 to 3 minutes later) has come to be recognized as the pink-color sign in cancerous lesions.ObjectiveThe aim of the present study was to verify the accuracy of the pink-color sign for diagnosing esophageal squamous neoplasms by a quantitative analysis.DesignA quantitative analysis of endoscopic images was performed by using prospectively pooled data.SettingA cancer referral center.PatientsFrom December 2003 to September 2005, consecutive patients who underwent a procedure by an experienced endoscopist and had iodine-unstained lesions measuring 5 to 20 mm in diameter without obvious protrusions or ulcers were prospectively enrolled.Main Outcome MeasurementsThe diagnostic ability of the quantitative assessment of the color change at 2 minutes after iodine staining (late phase) was investigated by using a receiver operating characteristic analysis.ResultsA total of 45 iodine-unstained lesions in 32 patients were included in the study. Among these, 22 of 23 pink-color–positive lesions were diagnosed as high-grade intraepithelial neoplasias or invasive cancers, whereas 19 of 22 pink-color–negative lesions were diagnosed as nonneoplastic lesions or low-grade intraepithelial neoplasias. The area under the receiver operating characteristic curve was 0.94, indicating excellent validity of the test. A sensitivity of 88% and specificity of 95% were obtained.LimitationsSingle-center retrospective analysis.ConclusionsWe verified the accuracy of the pink-color sign for diagnosing esophageal squamous neoplasms by a quantitative analysis.