Real-time histology of colon polyps – is it ready for prime time?

“If at first, the idea is not absurd, then there is no hope for it.” – Albert Einstein

Up until a few years ago, the idea of real-time histology of colon polyps might have sounded absurd to many. However, recent innovations in optical technologies have now afforded gastroenterologists the unique capability of in vivo identification of polyp histology, in real time during colonoscopy. These technologies work on different principles of optics. They can accentuate the contrast between the normal and abnormal tissue and highlight the surface patterns on polyps, for example with dyes (chromoendoscopy) or by varying the wavelength of light by filters (narrow-band imaging [NBI]) or by color enhancement methods (Fujinon Intelligent Chromoendoscopy [FICE], i-scan). Other technologies can increase the magnification (confocal endomicroscopy and endocytoscopy) or help to characterize polyp histology by quantifying the color and intensity of the reflected light (autofluorescence, spectroscopy). Broadly speaking, these technologies can be divided into “large-field” and “small-field” technologies depending upon the area of mucosa imaged. Large-field technologies include chromoendoscopy, electronic chromoendoscopy (NBI, FICE, i-scan), and autofluorescence, whereas the small-field technologies include confocal endomicroscopy, endocytoscopy, and spectroscopy.

First, let us be clear about the rationale for real-time histology of colon polyps.

• The majority of polyps detected during screening colonoscopy are diminutive (i. e. ≤ 5 mm in diameter). Diminutive polyps/adenomas rarely harbor advanced histology such as high grade dysplasia, cancer or villous features [1]. Therefore, the current practice of resecting these polyps and submitting them for histopathology assessment mainly serves to determine whether they are adenomas or hyperplastic. This information is then utilized for surveillance interval recommendations [2]. However, there are several drawbacks to this practice, the major one being the substantial cost burden to the patient and society for a limited health benefit. One study estimated that the incremental cost-effectiveness ratio for submitting all polyps for histopathology was a staggering $377 460 per life – year gained [3]. Other drawbacks include the risk of complications from avoidable polypectomies (for hyperplastic polyps), and delays in informing patients of their results and recommending the future surveillance interval.

• We rely on pathologists for the histologic diagnosis of polyps; however, the accuracy of these assessments is not perfect. Studies have shown that pathologists are 85 % – 95 % accurate in polyp histology characterization. In one study, four pathologists reviewed slides of 100 polyps and a perfect agreement was seen in only 48 polyps [4]. The median kappa for interobserver agreement for the diagnosis of adenoma vs. hyperplastic polyp was 0.89 but not completely perfect (kappa = 1.0). In another study [5] involving eight pathologists who reviewed 100 polyps, the interobserver agreement with kappa statistics for discriminating between adenoma and hyperplastic polyps varied from 0.84 to 0.98. Moreover, when the initial histologic impression of a polyp is either no diagnostic abnormality or lymphoid aggregate, deeper sections of the polyp have been shown to change the diagnosis in up to 10 % of cases [6]. Thus, the surveillance interval recommendations are based on an imperfect gold standard (i. e. histopathology). Studies have also shown that the adherence to surveillance guidelines by gastroenterologists is suboptimal [7].

Consideration of these factors combined with the emergence of new technologies has led to the proposal of a paradigm shift in the management of colon polyps. This includes the “diagnose, resect, and discard” strategy for diminutive polyps and the “do not resect” strategy for diminutive hyperplastic polyps in the rectosigmoid [8] [9]. The major advantage of the resect and discard strategy would be huge cost savings that have been estimated to be approximately $1 billion annually in the USA [3]. Apart from making colonoscopy more cost-effective, these strategies would also decrease the risk of complications from avoidable polypectomies of distal hyperplastic polyps, as well as minimize delays in informing patients of the results and future surveillance interval recommendations, thus improving overall efficiency.

The American Society for Gastrointestinal Endoscopy (ASGE) recently developed a Preservation and Incorporation of Valuable Endoscopic Innovations (PIVI) initiative that has set the thresholds for real-time endoscopic assessment of the histology of diminutive colorectal polyps [9]. In the PIVI statement, the committee of experts appointed by the ASGE determined that “for colorectal polyps ≤ 5 mm in size to be resected and discarded without pathologic assessment, endoscopic technology (when used with high confidence) used to determine histology of polyps ≤ 5 mm in size, when combined with histopathologic assessment of polyps > 5 mm in size, should provide a ≥ 90 % agreement in assignment of post-polypectomy surveillance intervals when compared with decisions based on pathology assessment of all identified polyps.” In addition, “for a technology to be used to guide the decision to leave suspected rectosigmoid hyperplastic polyps ≤ 5 mm in size in place (without resection), the technology should provide ≥ 90 % negative predictive value (when used with high confidence) for adenomatous histology.” Once an endoscopic technology meets these PIVI thresholds and appropriate training in the technology has been completed successfully, ASGE will support the use of that technology for incorporation into clinical practice. These clinically relevant benchmarks are now serving as a guide to researchers and investigators in the field of real-time histology of colon polyps.

In this issue of Endoscopy there are two studies on this theme that have evaluated NBI and FICE for characterizing colon polyps in real time during colonoscopy. The study by Longcroft-Wheaton et al. [10] from the UK compared standard-definition and high-definition colonoscopy with white light imaging and FICE for characterization of small colonic polyps (< 10 mm). The study was conducted at a tertiary referral center by one experienced investigator. With white light imaging there were no significant differences observed between the diagnostic characteristics of standard- and high-definition colonoscopes. With FICE, however, high-definition colonoscopes showed a superior sensitivity for neoplasia (93 % vs. 83 %; P = 0.048) and similar specificity, accuracy, and negative predictive value (NPV) (85 % vs. 73 %; P = 0.156) compared with standard-definition colonoscopes. The accuracy of surveillance intervals based on the real-time prediction was 100 % with FICE and high-definition colonoscopes compared with 88 % – 91 % with standard-definition scopes depending upon which society guidelines (ASGE or British Society of Gastroenterology [BSG]) were followed. Thus, FICE with high-definition endoscopes performed significantly better than standard-definition colonoscopes both in terms of diagnostic characteristics and the accuracy of surveillance intervals, highlighting the fact that advancements in optical technology do improve the ability to characterize colon polyps. The authors did not present the data for accuracy in surveillance intervals if only the diminutive polyps (≤ 5 mm) were characterized during colonoscopy, as has been proposed by the ASGE PIVI. Moreover, the NPV for adenomatous histology did not reach the recommended threshold of 90 % for the “do not resect” strategy for diminutive rectosigmoid hyperplastic polyps. Nonetheless, this study showed that by using FICE with high-definition colonoscopes, the surveillance intervals can be accurately predicted in individuals with small colon polyps.

The second study by Paggi et al. [11] from Italy evaluated NBI in a community hospital setting with six staff endoscopists who were experienced in the technology. The sensitivity, specificity, and accuracy for diagnosing adenoma was 95 %, 66 %, and 86 %, respectively, in small polyps (< 10 mm) and 94 %, 68 %, and 84 %, respectively, in diminutive polyps (≤ 5 mm). The authors estimated the surveillance interval based on the endoscopic prediction of polyp histology and compared it with the actual surveillance intervals that were based on histopathology. The concordance rate was 83 % in patients with small polyps and 85 % in patients with only diminutive polyps, which was lower than that reported in the UK study. This fell short of the 90 % benchmark set forth by the PIVI. The results of this study, however, appear to be more generalizable given its community hospital setting employing multiple endoscopists.

We have previously published our experience from three prospective trials conducted at two tertiary referral centers involving six endoscopists [12]. We found that in vivo optical diagnosis for all diminutive polyps with NBI combined with histologic assessment of the rest of the polyps predicted the correct surveillance interval in 86 – 94 % of individuals, depending upon whether those with 1 – 2 small tubular adenomas had a repeat colonoscopy at 5 years or 10 years. The NPV for adenomatous histology in diminutive rectosigmoid polyps was 95 %, thereby achieving both PIVI thresholds. I am optimistic that in the near future we will have more data from studies focusing on achieving the PIVI thresholds in different clinical practice settings and utilizing the various technologies available for in vivo histology of polyps.

• References

• 1 Gupta N, Bansal A, Rao D et al. Prevalence of advanced histological features in diminutive and small colon polyps. Gastrointest Endosc 2012; 75: 1022-1030
• 2 Levin B, Lieberman DA, McFarland B et al. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. Gastroenterology 2008; 134: 1570-1595
• 3 Kessler WR, Imperiale TF, Klein RW et al. A quantitative assessment of the risks and costs savings of forgoing histologic examination of diminutive polyps. Endoscopy 2011; 43: 683-691
• 4 Constantini M, Sciallero S, Giannini A et al. Interobserver agreement in the histologic diagnosis of colorectal polyps: the experience of the multicenter adenoma colorectal study (SMAC). J Clin Epidemiol 2003; 56: 209-214
• 5 Cross SS, Betmouni S, Burton JL et al. What levels of agreement can be expected between histopathologists assigning cases to discreet nominal categories? A study of the diagnosis of hyperplastic and adenomatous colorectal polyps. Mod Pathol 2000; 13: 941-944
• 6 Wu ML, Dry SM, Lassman CR. Deeper examination of negative colorectal biopsies. Am J Clin Pathol 2002; 117: 424-428
• 7 Schoen RE, Pinsky PF, Weissfeld JL et al. Utilization of surveillance colonoscopy in community practice. Gastroenterology 2010; 138: 73-81
• 8 Ignjatovic A, East JE, Suzuki N et al. Optical diagnosis of small colorectal polyps at routine colonoscopy (Detect InSpect ChAracterise Resect and Discard; DISCARD trial): a prospective cohort study. Lancet Oncol 2009; 10: 1171-1178
• 9 Rex DK, Kahi C, O’Brien M et al. The American Society for Gastrointestinal Endoscopy PIVI (Preservation and Incorporation of Valuable Endoscopic Innovations) on real-time endoscopic assessment of the histology of diminutive colorectal polyps. Gastrointest Endosc 2011; 73: 419-422
• 10 Longcroft-Wheaton G, Brown J, Cowlishaw D et al. High-definition vs standard-definition colonoscopy in the characterization of small colonic polyps: results from a randomized controlled trial. Endoscopy 2012; 44: 905-910
• 11 Paggi S, Rondonotti E, Amato A et al. Resect and discard strategy in clinical practice: a prospective cohort study. Endoscopy 2012; 44: 899-904
• 12 Gupta N, Bansal A, Rao D et al. Accuracy of in vivo optical diagnosis of colon polyp histology by narrow-band imaging in predicting colonoscopy surveillance intervals. Gastrointest Endosc 2012; 75: 494-502
• 13 Rex DK, Fennerty MB, Sharma P et al. Bringing new endoscopic imaging technology into everyday practice: what is the role of professional GI societies? Polyp imaging as a template for moving endoscopic innovation forward to answer key clinical questions. Gastrointest Endosc 2010; 71: 142-146
• 14 Raghavendra M, Hewett DG, Rex DK. Differentiating adenomas from hyperplastic colorectal polyps: narrow-band imaging can be learned in 20 minutes. Gastrointest Endosc 2010; 72: 572-576
• 15 Ignjatovic A, Thomas-Gibson S, East JE et al. Development and validation of a training module on the use of narrow-band imaging in differentiation of small adenomas from hyperplastic colorectal polyps. Gastrointest Endosc 2011; 73: 128-133
• 16 Rastogi A, Rao DR, Gupta N et al. Impact of a computer based teaching module on characterization of diminutive colon polyps using narrow band imaging (NBI) by non-experts in academics and community practice: a video based study. Gastrointest Endosc 2012; 75 : AB152-AB153
• 17 Hogan RB, Brill JV, Littenberg G et al. Predict, resect, and discard … really?. Gastrointest Endosc 2012; 75: 503-505