Diagnostic Value Of Narrow-Band Imaging For Polypoid Lesions Of Colorectal

Objective: To have knowledge of the morphology of various polypoid lesions of colorectal through a retrospective analysis of the clinical data of polypoid lesions of colorectal observed by high- definition NBI endoscopy in the digestive endoscopy center of First Hospital Affiliated to Medical College of Shihezi University from June 2013 to June 2015, to analyze its coincidence rate based on histopathological diagnosis, and to further discuss the diagnostic value of narrow-band imaging for polypoid lesions of colorectal.Methods: A total of 724 patients with colorectal polyps, including 398 males and 326 females(male and female ratio: 1.22:1; age: 18~89 years old; mean age: 55.8±11.7),who were diagnosed by common electronic colonoscopy in the digestive endoscopy center of First Hospital Affiliated to Medicine College of Shihezi University from June 2013 to June 2015.These patients undergoing NBI were retrospectively analyzed. The diagnostic sensitivity, specificity, PPV, NPV and accuracy of NBI international colorectal endoscopic [NICE] classification was calculated based on sizes, sites, profile, color, surface pattern and microvascular architecture of polypoid lesions of colorectal in these patients, which were compared with actual histopathological diagnosis, further guidance was provided on how to perform minimally invasive therapy or surgical treatment under endoscopy. An electron colonoscopy(CF-H260 AI, Olympus, Tokyo, Japan) and/or ultrasonic probes(UM-DP12-25 R and UM-DP20-25R) were employed to conduct experiment, with image processing apparatus(CV-260 SL,Olympus).Results: 1. Distribution of lesion site: ileocecus in 12 cases(1.5%), ascending in 91 cases(11.7%),transverse in 85 cases(11.0%), descending in 87 cases(11.2%), sigmoid in 233 cases(30.0%) and rectum in 268 cases(34.5%). 2. The lesion size: all size: 0.1~4.0cm; mean size: 0.86±0.81). 384 cases lesions of the size ≤0.5cm(49.5%), among which there were non-neoplastic in 272 cases, neoplastic in 112 cases; 200 cases between 0.5cm~1.0cm(25.8%), among which there were non-neoplastic in 56 cases, neoplastic in 144 cases; and 128 case of the size between 1.0cm~2.0cm(16.5%); and 64 cases of the size>2.0cm(8.2%);and lesions of the size >1.0cm there were non-neoplastic in 8 cases, neoplastic in 184 cases. 3. Distribution of disease entities:724 patients with polypoid lesions of colorectal who were undiagnosed by common electronic colonoscopy underwent NBI endoscopy. Among them, there were chronic inflammation of intestinal mucosa in 71 cases, hyperplastic polyps in 264 cases, schistosoma egg polyps in 1 cases, tubular adenoma in 171 cases, papillotubular adenoma in 4 cases, villous tubular adenoma in 49 cases, villous adenoma in 2 cases, serrated tubular adenoma in 1 cases, high grade intraepithelial neoplasia in 3 cases, local suspicious infiltration in 4 cases, not clearly classified in 206 cases.4. The coincidence rate between NBI and histopathology: Among 776 polypoid lesions of colorectal detected by NBI, among which there were non-neoplastic in 336 cases(43.3%), neoplastic in 440 cases(56.7%). The differential diagnostic sensitivity, specificity, PPV, NPV, and accuracy of NBI were 90.9 %(400/440), 85.7 %(288/336), 89.3%(400/448), 87.8%(288/328), 88.7 %(688/776), respectively. Diagnostic accuracy was 85.4%(328/384) in the 1 to 5 mm group, 88.0%(176/200) in the 6 to 10 mm group, 95.8%(184/192) in the more than 10 mm group, with statistically significant differences(P<0.05).Conclusion: To summarise, NBI endoscopy can not only distinctly show more lesions sizes, sites, profile, color, surface pattern and microvascular architecture but also effectively distinguish neoplastic from non-neoplastic colorectal lesions.