Endocytoscopy can provide additional diagnostic ability to magnifying chromoendoscopy for colorectal neoplasms

• Published in: Journal of gastroenterology and hepatology 2014-01, Vol.29 (1), p.83-90
• Main Authors: Kudo, Shin-ei, Mori, Yuichi, Wakamura, Kunihiko, Ikehara, Nobunao, Ichimasa, Katsuro, Wada, Yoshiki, Kutsukawa, Makoto, Misawa, Masashi, Kudo, Toyoki, Hayashi, Takemasa, Miyachi, Hideyuki, Inoue, Haruhiro, Hamatani, Shigeharu
• Format: Article
• Language: English
• Subjects: Aged; Colonoscopy - methods; colorectal cancer; Colorectal Neoplasms - diagnosis; Colorectal Neoplasms - pathology; Colorectal Neoplasms - surgery; comparative study; endoscopy; Female; Gentian Violet; Humans; Male; Methylene Blue; Middle Aged; Retrospective Studies; Sensitivity and Specificity; Staining and Labeling
• ISSN: 0815-9319; 1440-1746
• DOI: 10.1111/jgh.12374

Background and Aim Pit pattern (PIT) diagnosis with magnifying chromoendoscopy is effective diagnostic method for predicting a massively invasive submucosal colorectal cancer (SMm) which has possibility of metastasis, whereas endocytoscopy (EC) is recently reported to provide excellent diagnostic ability by enabling in vivo cellular visualization. The aim was to assess the additional diagnostic value of EC to PIT for diagnosing colorectal lesions. Methods We conducted a retrospective comparative analysis using a prospectively recorded database in a referral hospital. The subjects were 538 patients who were detected of a colorectal lesion with use of a magnifying colonoscope with EC capability. Each detected lesion was initially diagnosed by PIT findings followed by EC diagnosis by the on‐site endoscopist. The diagnostic abilities in predicting neoplastic change and SMm were compared between PIT and PIT plus EC. Results Overall, 514 lesions from 455 patients were available for analysis. Of them, there were 58 non‐neoplastic lesions, 352 adenomas, 15 slightly invasive submucosal cancers, and 89 SMm. The diagnostic abilities of predicting neoplastic change were comparable between PIT and PIT plus EC: sensitivity was 97.8% versus 97.4%, specificity was 91.4% versus 89.7%, and accuracy was 97.1% versus 96.5%. Regarding those of predicting SMm, PIT plus EC showed additional specificity and accuracy to PIT: specificity was 99.1% versus 97.6% (P = 0.041), and accuracy was 96.3% versus 93.8% (P = 0.004). Conclusions Though PIT has feasible diagnostic ability for predicting both neoplastic change and SMm, EC provides additional diagnostic value to PIT diagnosis for predicting SMm.