Patients with ulcerative colitis (UC) have a significant risk of colorectal cancer (CRC) (1–3). Guidelines recommend regular surveillance colonoscopies for patients with long-standing UC for early detection of dysplasia at 1- to 5-yearly intervals depending on their risk profiles (4–6). Detection of dysplasia in UC has been traditionally difficult with standard white light (WL) endoscopy with consequent high rates of cancers occurring between scheduled surveillance visits (7,8). However, with advances in optical imaging techniques, most dysplasias and colitis-associated cancers may be visualized under endoscopy. A recent long-term study revealed an increased incidence rate of dysplasia, likely attributable to improved endoscopic imaging and the recent use of chromoendoscopy (CE) (9).
According to studies comparing surveillance techniques using SD colonoscopy, CE with targeted biopsy showed a superior diagnostic yield of dysplasia compared with random biopsy protocols with WL (10–15). The SCENIC meta-analysis study incorporating 8 randomized controlled trials (RCTs) using SD colonoscopy revealed that CE with targeted biopsy had a 1.8-fold increase in the dysplasia detection rate compared with WL endoscopy random biopsy protocols (6). These data indicate that CE should be the technique of choice when performing UC surveillance using SD colonoscopy.
What is unclear is whether CE should still be used when surveillance is performed with high-definition (HD) colonoscopy. The SCENIC consensus recommended CE when performing dysplasia surveillance with HD colonoscopy, but the strength of this recommendation was conditional due to low-quality evidence available (6). In a prospective, tandem study using HD colonoscopy for 75 patients with UC, dysplasia was identified in significantly more patients undergoing CE (n = 16, 21%) than WL colonoscopy alone (n = 7, 9%) (16). Conversely, Mooiweer et al. (17) in their evaluation of large retrospective surveillance colonoscopy data over a 13-year period reported that CE did not increase diagnostic yield compared with WL colonoscopy. Although they did not specify the type of colonoscopy devices used, their data were likely to include a considerable proportion of HD colonoscopic procedures (17). A recent Japanese RCT, in which HD colonoscopy was used for most patients with UC, showed that targeted biopsies were more cost-effective for detecting a similar proportion of neoplasia than random biopsies (18). More recently, a Spanish group completed a prospective surveillance study for patients with long-standing inflammatory bowel disease (19). Their colonoscopy protocol included 2 sequential passes, that is, a first pass with WL and a second pass with CE. They reported that CE-incremental detection yield was comparable for SD and HD colonoscopic surveillance (51.5% vs 52.3%, P = 0.30) (19). A recent guideline encourages HD scope-based surveillance (20), and another recent guideline suggests application of either narrow-band imaging (NBI) or CE to augment dysplasia detection, when HD scope-based surveillance is performed (21). However, it is still unclear whether endoscopists using HD colonoscopy have the advantage of improved dysplasia detection with or without CE.
Endoscopists may hesitate to select CE as a primary surveillance technique for UC because of the longer procedure time required, in addition to the lack of strong evidence favoring CE with HD colonoscopy. In 2 meta-analyses, CE for surveillance of UC increased the procedure time by 11 minutes on average (6,22). Most studies providing the procedure time data used a spraying catheter for panchromoendoscopy (10,12–14). One HD colonoscopy study used an automated water lavage pump for dye spraying, but it did not compare the withdrawal time of CE with that of WL colonoscopy (16).
The primary aim of this multicenter RCT was to determine whether CE-based targeted biopsy could improve the dysplasia detection rate compared with that of WL colonoscopic random biopsy when HD colonoscopy is used for surveillance of patients with UC. Second, we aimed to compare the withdrawal time between CE spraying with an automated lavage pump and WL colonoscopy.
This was a prospective RCT conducted at 9 institutions in South Korea to compare CE-based targeted biopsy using HD colonoscopy (HDCE-T) with WL-based random biopsy using HD colonoscopy (HDWL-R) for the surveillance of dysplasia or colitic cancer in patients with UC. The trial was approved by the institutional review boards of each institution and was registered at the Clinical Research Information Service (cris.nih.go.kr, identification number: KCT0001195: 4-2013-0622).
Patients with UC who underwent surveillance colonoscopy for dysplasia or CRC were eligible. Study participants were enrolled from November 2013 to December 2015. The inclusion criteria were as follows: age ≥19 years; diagnosis of UC based on clinical, endoscopic, and histologic findings; extensive colitis with ≥8 years or left-sided colitis with ≥10 years of disease duration; and clinical remission status (as defined by simple clinical colitis activity index ≤8 and a mild Truelove and Witt's disease activity score). Informed consent was obtained from all enrolled patients. Patients were excluded if they had any of the following conditions: a history of CRC, a history of any type of colectomy, coagulopathy (prothrombin time >50% or activated partial prothrombin time >50 seconds), or impaired renal function (serum creatinine >1.2 mg/dL).
Sample size calculation and randomization
The dysplasia detection rates of WL random biopsy ranged 0–21% according to the previous studies (10,12–16,23). After removing 2 outliers (15,23) from these studies, the pooled dysplasia detection rate was 6.5%. Thus, we assumed 6.5% to be the dysplasia detection rate for HDWL-R. The expected difference in the dysplasia detection rate between HDWL-R and HDCE-T was estimated as 12% on the basis of a previous study using HD colonoscopy (16). The type I error (1-sided) and power were set at 0.05 and 0.8. The calculated sample size was 93 in each group. Considering the possibility of withdrawal from study participation, we decided to enroll at least 100 cases for each group. Patients were randomized in a 1:1 ratio by consecutive numbering according to a computer-generated 4-block permuted randomization table developed by an independent statistician.
Surveillance techniques of each group
All surveillance colonoscopies were performed by 9 endoscopists (with at least 6 years of experience), using an HD colonoscope (CF-HQ260 or CF-HQ290, Olympus co., Tokyo, Japan). For the HDWL-R group, targeted biopsies were taken from any suspected dysplastic lesions visible under WL colonoscopy, and then, 4-quadrant random biopsies were taken at every 10-cm segment from the cecum to the rectum. NBI or CE for the focal area of interest was allowed for examining suspected dysplastic lesions detected under WL colonoscopy. For the HDCE-T group, a transparent cap containing a water supply tube (distal attachment cap; ERBE Elektromedizin GmbH, Tübingen, Germany) was attached to the distal end of the colonoscope. If the scope had its own water infusion channel, a conventional transparent cap was attached. After cecal intubation, a 0.05% diluted indigo carmine solution was sprayed onto the colonic segments through the water infusion channel. If any suspected dysplastic lesion was detected, a 0.16% diluted indigo carmine solution was sprayed, and then, at least 2 biopsy specimens were obtained. For the HDCE-T group, 2 biopsy specimens were taken from the cecum, transverse colon, sigmoid colon, and rectum, even in the absence of suspicion of dysplasia, to assess the microscopic extent of colitis.
Pathologic diagnosis for colitis-associated dysplasia (CAD) was categorized into negative for dysplasia, indefinite for dysplasia, low-grade dysplasia, and high-grade dysplasia. Discrimination between sporadic colorectal neoplasia (CRN) and CAD was determined on the basis of endoscopic findings and pathological features. Any discrete adenomatous lesion located at the noncolitic segment was categorized as sporadic adenoma. Pedunculated lesions with typical histological features of sporadic adenoma were classified as sporadic adenoma regardless of the location, as the study protocol was developed before publication of the SCENIC consensus statements (6). The pathology of targeted and random biopsy specimens was reviewed by board-certified pathologists at each institution, and each biopsy specimen suspicious for dysplasia was reviewed by a central pathologist (H.K.), who was blinded to the randomization.
Outcomes and variables
The primary outcome was comparison of the dysplasia detection rates between the 2 groups. The dysplasia detection rate in each group was defined as the number of patients having CAD over the number of patients who underwent surveillance colonoscopy. The secondary outcomes were withdrawal time, total number of biopsies, total number of lesions, and total number of patients with CAD. Time spent for biopsy or polypectomy was excluded from the withdrawal time. In addition to the primary and secondary outcomes, the following variables were recorded: demographic characteristics, medications, family history of CRC, C-reactive protein level, and erythrocyte sedimentation rate at the time of surveillance colonoscopy, Mayo endoscopic subscore (24), UC endoscopic index of severity (25), and the presence of pseudopolyps or strictures.
Descriptive statistics were used to describe the variables related to the patients, lesions, and procedure-related outcomes. The χ2 test or Fisher exact test was used to analyze noncontinuous variables. The Student t test was used to analyze the differences in withdrawal time between the 2 groups. Calculations were performed using SPSS version 21.0 for Windows software (SPSS Inc., Chicago, IL).
A total of 210 patients were enrolled (108 for the HDWL-R group and 102 for the HDCE-T group, Figure 1). Mean duration of UC at the surveillance colonoscopy was 14.1 years in the HDWL-R group and 15.2 years in the HDCE-T group (P = 0.121). The number of previous surveillance colonoscopies and the proportion of extensive colitis were not significantly different between the 2 groups. Overall, patients were well-matched, and there were no significant differences in demographic characteristics between the 2 groups (Table 1). There were no statistical differences in the bowel preparation status and the incidence of pseudopolyps or stricture between the 2 groups (Table 2).
Diagnostic yield of surveillance colonoscopy
The number of cases with any CRN, including both sporadic adenoma and CAD, was 13 in the HDWL-R group and 21 in the HDCE-T group (P = 0.093). All the 13 CRNs in the HDWL-R group were detected under WL. Of the 6 CADs identified in the HDWL-R group, targeted biopsies were taken for 2 CADs, as they were visible and suspected as CADs under WL. The remaining 4 CADs in the HDWL-R group were detected by biopsies for the visible lesions considered “not dysplastic, but abnormal” at the time of colonoscopy. In the HDCE-T group, 4 CADs were detected by target biopsies. There was no significant difference in the CAD detection rate between the 2 groups (P = 0.749). When the patients with moderate to severe inflammation were excluded, the CAD detection rates did not differ between the 2 groups (4.4% in the HDWL-R group vs 3.5% in the HDCE-T group, P = 0.538). More biopsies were taken in the HDWL-R group than in the HDCE-T group, but withdrawal time was similar between 2 groups (Table 3). For the detection of 1 CAD, 604 biopsies were taken in the HDWL-R group and 237 biopsies in the HDCE-T group. There were no procedure-related complications including bleeding and perforation in each group.
In the current RCT, we compared the efficacy of detecting CAD using HDWL with random biopsies and HDCE with targeted biopsies. Unlike SD scope-based comparative studies between CE and WL for surveillance in patients with colitis (6,22), any additional improvement in the dysplasia detection rate by using CE with concurrent use of an HD scope has yet to be established. A tandem study by Picco et al. (16) reported that HDWL followed by HDCE enhanced the dysplasia detection rate from 9.3% to 21.3% (P = 0.007). Recently, Iacucci et al. (26) compared the neoplasia (including dysplasia, tubular adenoma, serrated adenoma, and adenocarcinoma) detection rates of 3 arms: HDWL (n = 90), HDCE (n = 90), and HD virtual CE (n = 90) with the targeted biopsy strategy for long-standing colitic patients. Unlike the tandem study by Picco et al., this study reported that HDWL-based targeted biopsy was not inferior to HDCE-based targeted biopsy for detecting neoplasia (neoplasia detection rates, 46.7% and 30% in the HDWL and HDCE groups, respectively) and that dysplasia detection rates were similar among the 3 arms (P = 0.84) (26). Similar to the findings of the study by Iacucci et al., we did not find a statistically significant difference between dysplasia detection rates of the HDWL-R group (6/108) and those of the HDCE-T group (4/102). Six of 10 CADs were detected by targeted biopsies (2 in the HDWL-R group and 4 in the HDCE-T group) in this study. The remaining 4 CADs detected by random biopsy were also recognized as abnormal mucosal changes under WL colonoscopy. However, biopsies for these 4 CADs were considered random biopsies because the endoscopist did not highly suspect dysplasia at that time. Therefore, no dysplasia was detected under “blind” random biopsy in our study. Although the application of NBI or CE for abnormal mucosal changes in the HDWL-R group was permitted for each endoscopist, the implementation of NBI or CE for abnormal mucosal changes under WL colonoscopy was not clearly standardized and was not mandatory in our study protocol. Had meticulous protocols been designed and applied to describe the details of abnormal mucosal lesions in the HDWL-R group, some of the random biopsies for unsuspected dysplasia under WL colonoscopy might have potentially been reclassified as targeted biopsies. Thus, our study is in agreement with aforementioned studies (16,26), whereby a targeted biopsy should be the main strategy for HD scope-based surveillance of long-standing colitic patients irrespective of whether CE or WL colonoscopy was chosen. Instead of panchromoendoscopy, CE could potentially be applied to areas of suspicion under WL to enhance the characterization of suspected dysplasia or abnormal mucosal lesions when HD scopes are used. Any additive role of CE for dysplasia detection in the HD era requires further investigation.
Prolonged procedure times for CE compared with random biopsy have been reported in previous studies (6,22). Conventional CE was more time-consuming (additional 8–18 minutes) than standard colonoscopy because of the need for repeated insertion and withdrawal of the spraying catheter through the channel, in addition to the need for dye spraying (27,28). By contrast, we found no significant differences in either insertion or withdrawal times between the HDCE-T and HDWL-R groups in our study. Not only the small number of biopsies in the HDCE-T group but also the use of an automated lavage pump for dye spraying instead of spray catheters may explain this finding, as it may have had a significant impact on preventing prolongation of withdrawal time in the HDCE-T group in this study. If the prolonged withdrawal time represents a potential hurdle for endoscopists when selecting CE for surveillance, the routine use of an automated lavage pump should be considered.
In our study, there were a number of important additional findings. First, HDCE-T could significantly reduce the total number of biopsies required to detect dysplasia, as evidenced by the similar dysplasia detection rate when compared with HDWL-R (average number of biopsies taken per colonoscopy, 33.6 vs 9.2 for the HDCE-T and HDWL-R groups, respectively). Although the loss of the innominate line (19), neoplastic Kudo pit patterns (10,19,26,29), and protruded morphology (19) has been suggested as endoscopic findings predictive of dysplasia in the recent studies, some had not been reported at the time of our protocol development. Therefore, the relatively large number of biopsies required to detect 1 dysplasia in the HDCE-T group of our study (237 biopsies for 1 dysplasia detection) might have resulted from the lack of understanding of endoscopic features predictive of dysplasia. Better understanding and training for the identification of the endoscopic features of dysplasia may further maximize procedural efficacy of surveillance colonoscopy. Second, HDCE-T showed a trend toward an improved detection rate of any CRN (i.e., CADs and sporadic adenomas). This implies that CE or CE with a transparent cap for the HDCE-T group may have resulted in a statistically insignificant improvement in neoplasia detection rates compared with the HDWL-R group. However, the cap-assisted colonoscopy does not improve CRN detection rates significantly in the screening population compared with conventional colonoscopy (30), and thus, the net influence of the transparent cap in CRN detection for patients with UC should be evaluated in additional studies. Finally, the HDCE-T considerably improved some of the drawbacks of conventional CE with respect to labor intensity and time consumption. We used a cap with a built-in water infusion channel for dye spraying and could effectively reduce the total withdrawal time by up to 2 minutes. Since prolonged procedure time is a major disadvantage of CE, using this method could potentially alleviate reluctance of using CE by endoscopists.
There are several limitations to our study. First, the level of suspicion for dysplasia was not prospectively investigated in our target biopsy protocol. Prospectively collected data defining levels of suspicion of dysplasia by endoscopists would be useful to identify the characteristics of lesions which can guide the necessity of target biopsy. Second, the relatively small number of dysplasia lesions identified in our patients was another limitation for statistical analysis and interpretation. A relatively large proportion of patients having left-sided colitis and a relatively small proportion of patients having primary sclerosing cholangitis may be associated with a relatively low incidence of dysplasia in this study. Third, according to our study protocol, the patient's endoscopic severity of inflammation could not be determined at the time of randomization. Consequently, colitic patients with moderate to severe inflammation were not excluded from the study (15.7% of HDWL-R group and 16.2% of HDCE-T group). However, even after these patients were excluded from the analysis, dysplasia detection rates did not differ between the 2 groups.
In conclusion, when the HD colonoscope was used for the surveillance of patients with long-standing UC, CE targeted biopsy did not improve the CAD detection rate compared with WL random biopsy. Therefore, routine panchromoendoscopy may not be necessary when the HD scope is used for surveillance. Nonetheless, all CADs in the HDWL-R were detected by biopsies taken from abnormal looking areas. Furthermore, HDCE-T with automated infusion pumps did not increase withdrawal times or reduce the total number of biopsies needed to detect dysplasia compared with the HDWL-R. Therefore, in the HD endoscope era, the effectiveness and efficiency of a CE-based targeted biopsy strategy compared with WL-based targeted biopsy for the surveillance of colitic patients should be investigated through additional studies.
CONFLICTS OF INTEREST
Guarantor of the article: Hyun-Soo Kim, MD, PhD.
Specific author contributions: Dong-Hoon Yang, MD, PhD and Soo Jung Park, MD, PhD, contributed equally to this work. D.-H.Y., S.J.P., and H.-S.K.: planning and conducting the study, collecting and analyzing data, interpreting the results, drafting the manuscript, and approval of the final draft. Y.S.P., D.I.P., K.-M.L., S.-A.J., J.S.K., J.H.C., S.-K.Y., and W.H.K.: planning and conducting the study, collecting data, and approval of the final draft. C.H.C. and C.-H.R.C.: analyzing the collected data, interpreting the results, drafting the manuscript, and approval of the final draft. J.K. and H.K.: planning and conducting the study, review of pathology, and approval of the final draft.
Financial support: This study was supported by grants from the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea. (HI10C2020).
Potential competing interests: None.
Clinical trials registration: This study was registered at cris.nih.go.kr (identification number: KCT0001195: 4-2013-0622) and approved October 29, 2013.
WHAT IS KNOWN
- ✓ Standard-definition colonoscopy is used for surveillance of long-standing UC; CE improves the dysplasia detection rate compared with WL endoscopy.
- ✓ Guidelines recommend CE-based target biopsies over WL-random biopsies for surveillance colonoscopy in patients with long-standing UC.
- ✓ Benefits of HD colonoscopy in CE vs WL for surveillance of UC are inconclusive.
- ✓ Prolonged procedure time may discourage the endoscopists from performing CE-based targeted biopsies in real practice.
WHAT IS NEW HERE
- ✓ When HD scopes were used, CE did not improve CAD detection compared with WL in surveillance of patients with UC.
- ✓ CADs in the HDWL-based random biopsy group were visible but not considered dysplasia under WL colonoscopy.
- ✓ Despite dye spraying throughout the large intestine, withdrawal time of HDCE-based targeted biopsy was similar to HDWL-based random biopsy.
- ✓ Further studies evaluating dye application exclusively to suspicious areas under HD scope to increase surveillance efficiency are warranted.
1. Colman RJ, Rubin DT. Histological inflammation increases the risk of colorectal neoplasia in ulcerative colitis: A systematic review. Intest Res 2016;14:202–10.
2. Jess T, Rungoe C, Peyrin-Biroulet L. Risk of colorectal cancer in patients with ulcerative colitis: A meta-analysis of population-based cohort studies. Clin Gastroenterol Hepatol 2012;10:639–45.
3. Eaden JA, Abrams KR, Mayberry JF. The risk of colorectal cancer in ulcerative colitis: A meta-analysis. Gut 2001;48:526–35.
4. Annese V, Daperno M, Rutter MD, et al. European evidence based consensus for endoscopy in inflammatory bowel disease. J Crohns Colitis 2013;7:982–1018.
5. Cairns SR, Scholefield JH, Steele RJ, et al. Guidelines for colorectal cancer screening and surveillance in moderate and high risk groups (update from 2002). Gut 2010;59:666–89.
6. Laine L, Kaltenbach T, Barkun A, et al. SCENIC international consensus statement on surveillance and management of dysplasia in inflammatory bowel disease. Gastroenterology 2015;148:639–51 e28.
7. Rutter MD, Saunders BP, Wilkinson KH, et al. Thirty-year analysis of a colonoscopic surveillance program for neoplasia in ulcerative colitis. Gastroenterology 2006;130:1030–8.
8. Mooiweer E, van der Meulen-de Jong AE, Ponsioen CY, et al. Incidence of interval colorectal cancer among inflammatory bowel disease patients undergoing regular colonoscopic surveillance. Clin Gastroenterol Hepatol 2015;13:1656–61.
9. Choi CH, Rutter MD, Askari A, et al. Forty-year analysis of colonoscopic surveillance program for neoplasia in ulcerative colitis: An updated overview. Am J Gastroenterol 2015;110:1022–34.
10. Kiesslich R, Fritsch J, Holtmann M, et al. Methylene blue-aided chromoendoscopy for the detection of intraepithelial neoplasia and colon cancer in ulcerative colitis. Gastroenterology 2003;124:880–8.
11. Kiesslich R, Goetz M, Lammersdorf K, et al. Chromoscopy-guided endomicroscopy increases the diagnostic yield of intraepithelial neoplasia in ulcerative colitis. Gastroenterology 2007;132:874–82.
12. Rutter MD, Saunders BP, Schofield G, et al. Pancolonic indigo carmine dye spraying for the detection of dysplasia in ulcerative colitis. Gut 2004;53:256–60.
13. Marion JF, Waye JD, Present DH, et al. Chromoendoscopy-targeted biopsies are superior to standard colonoscopic surveillance for detecting dysplasia in inflammatory bowel disease patients: A prospective endoscopic trial. Am J Gastroenterol 2008;103:2342–9.
14. Hlavaty T, Huorka M, Koller T, et al. Colorectal cancer screening in patients with ulcerative and Crohn's colitis with use of colonoscopy, chromoendoscopy and confocal endomicroscopy. Eur J Gastroenterol Hepatol 2011;23:680–9.
15. Matsumoto T, Nakamura S, Jo Y, et al. Chromoscopy might improve diagnostic accuracy in cancer surveillance for ulcerative colitis. Am J Gastroenterol 2003;98:1827–33.
16. Picco MF, Pasha S, Leighton JA, et al. Procedure time and the determination of polypoid abnormalities with experience: Implementation of a chromoendoscopy program for surveillance colonoscopy for ulcerative colitis. Inflamm Bowel Dis 2013;19:1913–20.
17. Mooiweer E, van der Meulen-de Jong AE, Ponsioen CY, et al. Chromoendoscopy for surveillance in inflammatory bowel disease does not increase neoplasia detection compared with conventional colonoscopy with random biopsies: Results from a large retrospective study. Am J Gastroenterol 2015;110:1014–21.
18. Watanabe T, Ajioka Y, Mitsuyama K, et al. Comparison of targeted vs random biopsies for surveillance of ulcerative colitis-associated colorectal cancer. Gastroenterology 2016;151:1122–30.
19. Carballal S, Maisterra S, López-Serrano A, et al. Real-life chromoendoscopy for neoplasia detection and characterisation in long-standing IBD. Gut 2018;67:70–8.
20. Magro F, Gionchetti P, Eliakim R, et al. Third european evidence-based consensus on diagnosis and management of ulcerative colitis. Part 1: Definitions, diagnosis, extra-intestinal manifestations, pregnancy, cancer surveillance, surgery, and ileo-anal pouch disorders. J Crohns Colitis 2017;11:649–70.
21. Rubin DT, Ananthakrishnan AN, Siegel CA, et al. ACG clinical guideline: Ulcerative colitis in adults. Am J Gastroenterol 2019;114:384–413.
22. Subramanian V, Mannath J, Ragunath K, et al. Meta-analysis: The diagnostic yield of chromoendoscopy for detecting dysplasia in patients with colonic inflammatory bowel disease. Aliment Pharmacol Ther 2011;33:304–12.
23. Günther U, Kusch D, Heller F, et al. Surveillance colonoscopy in patients with inflammatory bowel disease: Comparison of random biopsy vs. targeted biopsy protocols. Int J Colorectal Dis 2011;26:667–72.
24. Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosalicylic acid therapy for mildly to moderately active ulcerative colitis. A randomized study. N Engl J Med 1987;317:1625–9.
25. Travis SP, Schnell D, Krzeski P, et al. Reliability and initial validation of the ulcerative colitis endoscopic index of severity. Gastroenterology 2013;145:987–95.
26. Iacucci M, Kaplan GG, Panaccione R, et al. A randomized trial comparing high definition colonoscopy alone with high definition dye spraying and electronic virtual chromoendoscopy for detection of colonic neoplastic lesions during IBD surveillance colonoscopy. Am J Gastroenterol 2018;113:225–34.
27. Kahi CJ, Anderson JC, Waxman I, et al. High-definition chromocolonoscopy vs. high-definition white light colonoscopy for average-risk colorectal cancer screening. Am J Gastroenterol 2010;105:1301–7.
28. Hüneburg R, Lammert F, Rabe C, et al. Chromocolonoscopy detects more adenomas than white light colonoscopy or narrow band imaging colonoscopy in hereditary nonpolyposis colorectal cancer screening. Endoscopy 2009;41:316–22.
29. Bisschops R, Bessissow T, Dekker E, et al. Pit pattern analysis with high-definition chromoendoscopy and narrow-band imaging for optical diagnosis of dysplasia in patients with ulcerative colitis. Gastrointest Endosc 2017;86:1100–6.e1.
30. Facciorusso A, Del Prete V, Buccino RV, et al. Comparative efficacy of colonoscope distal attachment devices in increasing rates of adenoma detection: A network meta-analysis. Clin Gastroenterol Hepatol 2018;16:1209–19 e9.