Narrow‐band imaging with dual focus magnification in differentiating colorectal neoplasia

Abstract

Background: Real-time optical diagnosis of colorectal polyps may lead to substantial time and cost savings and could potentially reduce complications associated with polypectomy. We prospectively assessed the utility of a novel narrow-band imaging (NBI) system with dual focus magnification (DF) in differentiating colorectal polyps in consecutive patients undergoing colonoscopy. Patients and Methods:All procedures were carried out with a prototype 190 series Exera III NBI system with Dual Focus capability. Histology of each polyp was predicted in real time with NBI-DF based on the modified Sano's classification with a confidence level (low/high). NBI-DF diagnosis was then compared to the final (blinded) histopathology results. Primary endpoint was the accuracy of endoscopic prediction with the modified Sano's classification of all polyps when they were diagnosed with high confidence. Secondary endpoints were the accuracy of post-polypectomy surveillance intervals for diminutive polyps (Յ5 mm). A total of 164 polyps in 87 patients (53 males) with a mean age of 63 (range 28-86) years were evaluated.Results: 149 polyps were diagnosed with high confidence on endoscopic prediction, out of which 80 were diminutive, 31 small (6-9 mm) and 38 large (>10 mm). Overall accuracy of NBI-DF compared to final histopathology was 97%. The accuracy for postpolypectomy surveillance interval based on the endoscopic prediction was 97%. The NPV for diminutive rectosigmoid polyps for adenomatous histology was 100%. Conclusion:In this preliminary feasibility study, NBI-DF permitted prediction of histology of colorectal polyps with high accuracy. In addition, both of the secondary endpoints exceeded the ASGE PIVI thresholds for the management of diminutive polyps.

ASGE Technology Committee systematic review and meta-analysis assessing the ASGE PIVI thresholds for adopting real-time endoscopic assessment of the histology of diminutive colorectal polyps

“…[7][8][9][10][11][12]15,18,20,26 The pooled percentage of agreement using the random-effects model was 89% (95% CI, 85-93) (Fig. 6).…”

“…[7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25] These studies collectively evaluated 4013 in situ small or diminutive colorectal polyps in real-time by using NBI and compared it with criterion standard histopathology. The median prevalence of adenomas among these polyps was 48.5% (range 18%-88%).…”

A prospective dual-center proof-of-principle study evaluating the incremental benefit of narrow-band imaging with a fixed zoom function in real-time prediction of polyp histology. Can we resect and discard?

“…17 When incorporating these features in the prediction of SSA, we found them to be acceptably accurate. Our study is the largest to date to accurately assess SSAs with the NBI-NF with a high accuracy, having been previously demonstrated by Singh et al 27 A larger study by Vu et al 28 incorporated serrated polyps into their study but did not achieve the accuracy required by the PIVI statements. Basford et al 29 achieved the PIVI requirements by using i-Scan technology (Pentax, Tokyo, Japan); however, they identified only two serrated polyps.…”

“…The main limitation of the study is that we used somewhat small numbers, which was also an issue for Singh et al 27 Our study was a proof-of-concept study to determine whether the NBI-NF can be used with confidence in detecting adenomatous pathology. This became more of an issue when we looked at subgroups such as diminutive SSAs, of which we had only 13 cases.…”

Novel computer-aided diagnostic system for colorectal lesions by using endocytoscopy (with videos)

“…These have been proven to have excellent diagnostic performance, most of them achieving accuracies of greater than 90%; however, such accuracy can only be achieved by expert endoscopists. [10][11][12][21][22][23] The same is true of EC, an emerging endoscopic technology that allows real-time cellular imaging of alimentary mucosa and provides very accurate predictions of pathology (eg, accuracies of 94.1% 17 and 96.5% 25 for neoplastic change). 14,17,[24][25][26] However, with this technology as well, such accuracy is achieved only by trained and experienced expert endoscopists.…”