Refining endoscopic surveillance beyond average-risk screening

Refining endoscopic surveillance beyond average-risk screening 
 
Dear colleagues, 

Screening and surveillance remain foundational to modern endoscopic practice, with colorectal cancer prevention through colonoscopy serving as a central focus for gastroenterologists. Yet beyond average-risk screening, there are several areas of surveillance where practice patterns vary, evidence continues to evolve, and uncertainty persists. In this issue, we turn our attention to two such areas: dysplasia surveillance in inflammatory bowel disease and gastric cancer prevention through the identification and management of gastric intestinal metaplasia. 

Dr. Samir A. Shah provides a practical and evidence-based approach to IBD surveillance, highlighting his use of chromoendoscopy with targeted sampling to improve dysplasia detection. He discusses risk stratification, patient selection, and procedural techniques that allow endoscopists to maximize yield while avoiding unnecessary over-surveillance. 

Complementing this, Drs. Isiah Gonzalez and Mimi Tan offer a thoughtful framework for the evaluation and surveillance of gastric intestinal metaplasia — an area often encountered incidentally and marked by low endoscopic sensitivity. They outline strategies to improve detection, emphasize systematic gastric mapping, and integrate current U.S. and international guidelines to help clinicians tailor surveillance based on individual risk. 

At a time when increasing attention is being paid to the risks and inefficiencies of over-surveillance, these perspectives aim to provide a balanced, pragmatic approach to identifying patients at increased malignancy risk while delivering high-quality, evidence-driven care. We hope these articles help inform and refine your surveillance practices, and we welcome your continued feedback and experience. 

Gyanprakash A. Ketwaroo, MD, MSc, is associate professor of medicine, Yale University, New Haven, and chief of endoscopy at West Haven VA Medical Center, both in Connecticut. He is an associate editor for GI & Hepatology News. 

Gastric intestinal metaplasia surveillance in U.S. practice 

By Isaiah T. Gonzalez, MD, and Mimi C. Tan, MD, MPH 

Gastric intestinal metaplasia (GIM) often remains outside the primary clinical focus of most gastroenterologists until the diagnosis appears unexpectedly on a pathology report, leaving the clinician uncertain about next steps. To improve detection of GIM, I focus on careful examinations using high-definition white-light endoscopy (HDWLE) and narrow band imaging (NBI). When GIM is suspected on the endoscopic examination, I routinely obtain five gastric biopsies per the Sydney protocol¹ and separate the biopsies into two jars (antrum and gastric body) which allows me to determine the anatomic distribution of GIM without a repeat endoscopy. When the location of GIM is unknown (random biopsies without location specified), repeat endoscopy with systematic gastric mapping biopsies should be performed to risk stratify and determine the surveillance interval. To guide surveillance of GIM, I use the AGA 2025 Clinical Practice Update on Screening and Surveillance in Individuals at Increased Risk for Gastric Cancer in the United States² and the 2025 Management of Epithelial Precancerous Conditions and Early Neoplasia of the Stomach (MAPS III) guideline update from the European Society of Gastrointestinal Endoscopy (ESGE)³. 

Gastric biopsies with downstream decisions in mind 

GIM and atrophic gastritis are difficult to detect endoscopically, particularly in the United States, where the diagnosis is infrequently encountered. Sensitivity of GIM detection on HDWLE is low (28%); even with NBI, the sensitivity remains suboptimal (53-80%). For this reason, I carefully assess for subtle endoscopic features of GIM and atrophic gastritis, such as cobblestoning/surface irregularity, red-white mucosa, mucosal pallor and greater visibility of submucosal vessels, and gastric fold disappearance while also assessing for the atrophic border per the Kimura-Takemoto classification⁴.  

When the mucosa appears suspicious for GIM or atrophy, I obtain a minimum of five biopsies according to the updated Sydney System (lesser curvature and greater curvature of antrum and gastric body along with incisura) and submit the specimens in separately labeled antrum and gastric body jars¹. This biopsy strategy gives me data on GIM severity and location should GIM be found, and I can then recommend a surveillance interval for endoscopy. This has helped me avoid the scenario of incidental GIM diagnosis on random biopsy histopathology without information on GIM location. 

Addressing ‘GIM on random gastric biopsy’ and unknown GIM extent  

A common clinical scenario is one where GIM is diagnosed on random gastric biopsies when the anatomic location of GIM is not known. In such cases, a repeat upper endoscopy with systematic gastric mapping is necessary to determine disease extent and to accurately risk-stratify the patient. Surveillance endoscopy interval and whether surveillance endoscopy is indicated is determined in part by GIM extent and location (antrum-restricted vs corpus-extended).  

In this scenario, I repeat upper endoscopy in approximately one year to complete gastric mapping biopsies and risk stratification. This approach allows identification of patients who warrant three-year surveillance due to high-risk features (multifocal, corpus-extended, or incomplete-type GIM) while recognizing that most patients with antrum-restricted GIM do not require further surveillance². 

GIM risk stratification 

GIM risk stratification involves assessment of the anatomic distribution (antrum-restricted vs corpus-extended), disease focality (focal vs multifocal), and the histologic severity and subtype (complete vs incomplete). I use the updated Sydney System for gastric mapping biopsies and place biopsies from each location in 5 separate jars when a GIM diagnosis is already known. Visible lesions are biopsied or removed and placed in separate jars. This approach allows the pathologist to document the Operative Link on Gastritis Assessment (OLGA) or Operative Link on Gastric Intestinal Metaplasia (OLGIM) stage, informed by disease severity and location⁵.  

AGA2 and ESGE3 support three-year intervals for patients with high-risk features, including moderate to severe atrophy, multifocal or corpus-extended GIM, and/or incomplete-type GIM. Conversely, no surveillance is advised for patients with mild, focal, and/or complete-type GIM, provided H. pylori has been eradicated, and there is no family history of gastric cancer. If there is a family history of gastric cancer in a first-degree relative, endoscopic surveillance is recommended in one to two years for high-risk GIM or atrophy and three to five years with mild, focal, or complete GIM. 

Primary and secondary prevention of gastric cancer  

While much attention is often directed towards endoscopic screening for gastric cancer and precancerous lesions, primary prevention remains critical for long-term gastric cancer control. Primary prevention includes H. pylori screening and eradication, smoking cessation, and promotion of healthy dietary habits. Groups at high-risk for gastric cancer should be testing for H. pylori even when asymptomatic, including first-generation immigrants from high-incidence regions, non-White racial and ethnic groups, individuals with a first-degree relative with gastric cancer, and those with select hereditary gastrointestinal polyposis or cancer syndromes². All confirmed H. pylori infections require treatment with post-treatment eradication testing. 

At the same time, it is important to acknowledge the current limitations of gastric cancer screening in the United States. Although mathematical and cost-effectiveness studies suggest that population-wide screen-and-treat strategies may be cost-effective in select high-risk populations, these approaches are not currently endorsed by the USPSTF or routinely covered by payers. Until this changes, the most impactful strategy remains vigilant H. pylori testing and eradication based on symptoms or risk factors combined with high-quality endoscopic examinations. 

Isaiah T. Gonzales, MD, is a physician in the Department of Medicine at Baylor College of Medicine in Houston, Texas. 

Mimi C. Tan, MD, MP, is a physician in the Section of Gastroenterology and Hepatology within the Department of Medicine at Baylor College of Medicine in Houston, Texas. 

References 

1. Dixon, M. F., Genta, R. M., Yardley, J. H., & Correa, P. (1996). Classification and grading of gastritis. The updated Sydney System. International Workshop on the Histopathology of Gastritis, Houston 1994. The American journal of surgical pathology, 20(10), 1161–1181. https://doi.org/10.1097/00000478-199610000-00001 

2. Shah, S. C., Wang, A. Y., Wallace, M. B., & Hwang, J. H. (2025). AGA Clinical Practice Update on Screening and Surveillance in Individuals at Increased Risk for Gastric Cancer in the United States: Expert Review. Gastroenterology, 168(2), 405–416.e1. https://doi.org/10.1053/j.gastro.2024.11.001 

3. Dinis-Ribeiro M, Libânio D, Uchima H, Spaander MCW, Bornschein J, Matysiak-Budnik T, Tziatzios G, Santos-Antunes J, Areia M, Chapelle N, Esposito G, Fernandez-Esparrach G, Kunovsky L, Garrido M, Tacheci I, Link A, Marcos P, Marcos-Pinto R, Moreira L, Pereira AC, Pimentel-Nunes P, Romanczyk M, Fontes F, Hassan C, Bisschops R, Feakins R, Schulz C, Triantafyllou K, Carneiro F, Kuipers EJ. Management of epithelial precancerous conditions and early neoplasia of the stomach (MAPS III): European Society of Gastrointestinal Endoscopy (ESGE), European Helicobacter and Microbiota Study Group (EHMSG) and European Society of Pathology (ESP) Guideline update 2025. Endoscopy. 2025 May;57(5):504-554. doi: 10.1055/a-2529-5025. Epub 2025 Mar 20. PMID: 40112834. 

4. Maric, L., Castaneda, D., Singh, H., Bejarano, P., Jimenez Cantisano, B., & Castro, F. J. (2022). Kimura-Takemoto Classification: A Tool to Predict Gastric Intestinal Metaplasia Progression to Advanced Gastric Neoplasia. Digestive diseases and sciences, 67(8), 4092–4099. https://doi.org/10.1007/s10620-021-07212-x 

5. Rugge, M., Meggio, A., Pennelli, G., Piscioli, F., Giacomelli, L., De Pretis, G., & Graham, D. Y. (2007). Gastritis staging in clinical practice: the OLGA staging system. Gut, 56(5), 631–636. https://doi.org/10.1136/gut.2006.106666 

Introduction to IBD dysplasia surveillance 

By Samir A. Shah, MD 

IBD patients have an elevated risk of colorectal cancer. Population based studies put the overall risk of colorectal cancer at about two times higher than that of the general population (lower than the purported risk in a 2001 meta-analysis of 8% by 20 years,¹² and 18% by 30 years).  

Risk factors include extent of colonic disease, number of years of disease, primary scloreosing cholangitis (PSC), family history of colorectal cancer (especially age < 50), degree of endoscopic/histologic inflammation, male gender, presence of post inflammatory pseudo-polyps, and history of previous dysplasia. The paradigm for surveillance in IBD has changed from multiple random biopsies and annual colonoscopies with standard definition scopes in the 1990s to the use of HD scopes with chromoendoscopy (dye spray or virtual) and targeted biopsies and longer intervals in between determined by findings and risk stratification.  

The optimal approach to dysplasia detection and colorectal cancer prevention in IBD continues to be debated. However, a strong case can be made for chromoendoscopy with targeted biopsies as the standard of care for IBD surveillance in 2026. 

When to initiate surveillance 

Surveillance colonoscopy should begin eight years after IBD diagnosis for patients with colonic disease extending beyond the rectum for ulcerative colitis and at least one-third of the colon involved for Crohn’s disease.^1-4,7,9 For patients with concomitant PSC, surveillance should start at diagnosis, regardless of disease duration or extent reflecting the substantially elevated cancer risk in this population.^{1-4, 9} 

Optimizing conditions for dysplasia detection 

The foundation of effective surveillance requires optimizing multiple procedural factors: (1) achieving disease remission before surveillance, as active inflammation significantly impairs dysplasia detection; (2) using high-definition (H-D) scopes with excellent bowel preparation and meticulous washing and inspection of all colorectal mucosa being essential; (3) performing a second look preferably with chromoendoscopy. 

The role of chromoendoscopy 

Chromoendoscopy maximizes dysplasia detection. The ASGE recommends chromoendoscopy with targeted biopsies as the preferred surveillance technique.^3-4 Dye chromoendoscopy should be considered in all IBD patients undergoing surveillance, but especially in higher risk patients such as patients with PSC or prior invisible dysplasia.^1-4 

A 2024 meta-analysis demonstrated that dye chromoendoscopy detected significantly more patients with dysplasia than HD white-light endoscopy (18.8% vs 9.4%, OR 1.94), with high certainty of evidence.⁶ This translates to nearly double the dysplasia detection rate, representing a clinically meaningful improvement in patient outcomes. 

A 2026 meta-analysis of studies using HD scopes found that virtual chromoendoscopy was not inferior to dye chromoendoscopy. Thus, the endoscopist can choose either virtual or dye chromoendoscopy. 

Virtual chromoendoscopy — including narrow-band imaging, Pentax Medical i-scan, and Fuji intelligent color enhancement — enhances vascular and surface architecture without requiring dye application, offering practical advantages in busy endoscopy units. The ACG-ASGE joint quality indicators recommend HD colonoscopy complemented with one modality of chromoendoscopy with targeted biopsy sampling of suspicious areas.⁸ 

My personal preference is dye spray over virtual. I believe that dye spray chromoendoscopy is well within the wheelhouse of endoscopists who perform screening and surveillance colonoscopies (all my partners utilize dye chromoendoscopy in our private practice). I incorporated dye-spray chromoendoscopy in my practice shortly after Professor Ralf Keisslich discussed his landmark 2003 publication showing the superior yield of chromoendoscopy at GI grand rounds at Brown University in 2024.  

Dye spray requires an excellent prep, resulting in a more thorough second look as the endoscopist can see where the dye went and did not go, and is forced to irrigate unstained areas and suction up puddles of excess dye for optimal visualization. With virtual chromoendoscopy, time pressure can make the second look too quick and not complete. Either methylene blue or FD&C#2 (a synthetic equivalent of indigo carmine) can be used and be efficiently delivered using the flushing apparatus.  

I prefer FD&C#2 mixed to a concentration of 0.1%, which I use routinely in our outpatient ASC. At the hospital, I use a very dilute concentration of 0.05% methylene blue (50mg/10ml in 1 liter of sterile water). We did a retrospective, uncontrolled comparison of chromoendoscopy to white light endoscopy (WLE) on my patients between January 2005 and August 2012 and reported 25 dysplastic lesions in 64 procedures with chromoendoscopy (39.1%) versus only 8 in 120 procedures (6.9%) with WLE (p<0.001, a 5-fold higher yield!).¹⁵ This changed my practice to use this technique for all my surveillance colonoscopies in IBD and now also for patients with serrated polyposis syndrome (SPS). 

A few caveats: Don’t use dye spray if the prep is not excellent (Boston Bowel Prep score = 9), or if there is significant inflammation. Wear scrubs and old sneakers, and warn staff and patients to wear old clothes that they don’t mind getting stained with blue dye. I book the cases for 45 minutes in our outpatient center and one hour for the hospital. Our approach to incorporating chromoendoscopy into practice is depicted below with an algorithm for clinicians to follow.¹⁴ 

Are biopsies still needed? 

When chromoendoscopy (dye-spray or virtual) is performed with high-definition endoscopy, targeted biopsies of suspicious lesions are sufficient as most dysplastic lesions are visible.2 However, non-targeted biopsies should still be considered in high-risk patients such as those with prior invisible dysplasia, multiple post-inflammatory pseudo-polyps, or primary sclerosing cholangitis.² 

Without chromoendoscopy, extensive non-targeted biopsies should be done approximately four adequately spaced biopsies every 10 cm from flat colorectal mucosa, totaling at least 32 samples.^2,5 This labor-intensive, costly and low yield approach underscores the efficiency gains achievable with chromoendoscopy-guided surveillance. 

Targeted biopsies dramatically outperform random sampling, with dysplasia found in 17.3% of targeted biopsies versus only 0.33% of random biopsies in comparative studies.² This 50-fold difference emphasizes the importance of enhanced visualization techniques. 

Lesion characterization and management 

All visible precancerous lesions should be described using the modified Paris Classification. Documentation should include size, morphology, border clarity, ulceration presence, location, and relationship to areas of current or past colitis. 

All clearly delineated dysplastic lesions without stigmata of invasive cancer or significant submucosal fibrosis should be considered for endoscopic resection. Endoscopic resection is preferred over biopsies when lesions are clearly demarcated. When resectability is uncertain or beyond the skill set of the endoscopist, referral to specialized advanced endoscopist and/or IBD center is recommended. 

Surveillance intervals and risk stratification 

After negative screening colonoscopy, surveillance intervals range from one to five years based on risk stratification.^1-2 High-risk patients — including those with prior dysplasia, strictures, primary sclerosing cholangitis, or extensive colitis with severe inflammation — require annual surveillance.^1-4,7,9 Intermediate-risk patients with extensive colitis and mild-to-moderate inflammation or family history of colorectal cancer should undergo surveillance every two to three years.^1-4,9 Low-risk patients with limited disease extent and minimal inflammation may extend to 5-year intervals.^1,9 

Future directions and emerging technologies 

Artificial intelligence-assisted detection systems show promise for enhancing surveillance performance. Computer-aided models have demonstrated 95% sensitivity and 99% specificity for detecting colorectal lesions in IBD patients.^10-11 

Conclusions 

Effective IBD surveillance requires integrating multiple evidence-based strategies: initiating surveillance at appropriate intervals, optimizing procedural conditions, employing chromoendoscopy, performing targeted biopsies of suspicious lesions, and risk-stratifying patients for appropriate follow-up intervals.  

While chromoendoscopy remains the preferred technique for maximizing dysplasia detection yield, high-definition white-light endoscopy with meticulous technique represents an acceptable alternative when chromoendoscopy expertise is unavailable.

Emerging technologies, including artificial intelligence and advanced imaging modalities, promise further improvements in dysplasia detection and patient outcomes. 

Samir A. Shah, MD, is chief of gastroenterology, The Miriam Hospital, Providence, RI; and clinical professor of medicine, Warren Alpert School of Medicine at Brown University. He reports disclosures as a speaker for AbbVie and JNJ, a consultant for Roche Information Systems, and an advisory board member for Pfizer. 

References 

1.East JE, Gordon M, Nigam GB, et al. British Society of Gastroenterology guidelines on colorectal surveillance in inflammatory bowel disease. Gut Published Online First: 30 April 2025. doi: 10.1136/gutjnl-2025-335023 

2. AGA Clinical Practice Update on Endoscopic Surveillance and Management of Colorectal Dysplasia in Inflammatory Bowel Diseases: Expert Review. Gastroenterology. 2021. Murthy SK, Feuerstein JD, Nguyen GC, Velayos FS. 

3. The Role of Endoscopy in Inflammatory Bowel Disease. Gastrointestinal Endoscopy. 2015. American Society for Gastrointestinal Endoscopy Standards of Practice Committee, Shergill AK, Lightdale JR, et al. 

4. ASGE IBD Endoscopy Consensus Panel; Shen B, Abreu MT, Cohen ER, Farraye FA, Fischer M, Feuerstadt P, Kapur S, Ko HM, Kochhar GS, Liu X, Mahadevan U, McBride DL, Navaneethan U, Regueiro M, Ritter T, Sharma P, Lichtenstein GR. Endoscopic diagnosis and management of adult inflammatory bowel disease: a consensus document from the American Society for Gastrointestinal Endoscopy IBD Endoscopy Consensus Panel. Gastrointest Endosc. 2025 Feb;101(2):295-314. doi: 10.1016/j.gie.2024.08.034. Epub 2024 Oct 18. PMID: 39425706. 

5. Colorectal Cancer Screening. National Comprehensive Cancer Network. Updated 2025-06-24. 

6. Dye Chromoendoscopy Outperforms High-Definition White Light Endoscopy in Dysplasia Detection for Patients With Inflammatory Bowel Disease: An Updated Meta-Analysis of Randomized Controlled Trials. The American Journal of Gastroenterology. 2024. Mohamed MFH, Marino D, Elfert K, et al. 

7. ACG Clinical Guideline: Management of Crohn's Disease in Adults. The American Journal of Gastroenterology. 2018. Lichtenstein GR, Loftus EV, Isaacs KL, et al. 

8. Quality Indicators for Colonoscopy. The American Journal of Gastroenterology. 2024. Rex DK, Anderson JC, Butterly LF, et al.  

9. Ulcerative Colitis. Lancet. 2023. Le Berre C, Honap S, Peyrin-Biroulet L. 

10. Artificial Intelligence and Endo-Histo-Omics: New Dimensions of Precision Endoscopy and Histology in Inflammatory Bowel Disease. The Lance. Gastroenterology & Hepatology. 2024. Iacucci M, Santacroce G, Zammarchi I, et al. 

11. Artificial Intelligence in Advancing Inflammatory Bowel Disease Management: Setting New Standards. Cancers. 2025. Labarile N, Vitello A, Sinagra E, et al. 

12. Eaden JA, Abrams KR, Mayberry JF. The risk of colorectal cancer in ulcerative colitis: a meta-analysis. Gut. 2001 Apr;48(4):526-35. doi: 10.1136/gut.48.4.526. PMID: 11247898; PMCID: PMC1728259. 

13. Kiesslich R, Fritsch J, Holtmann M, Koehler HH, Stolte M, Kanzler S, Nafe B, Jung M, Galle PR, Neurath MF. Methylene blue-aided chromoendoscopy for the detection of intraepithelial neoplasia and colon cancer in ulcerative colitis. Gastroenterology. 2003 Apr;124(4):880-8. doi: 10.1053/gast.2003.50146. PMID: 12671882. 

14. Kochar B, Mao EJ, Shah SA. Optimal Dysplasia Detection and Management in IBD: Now and in the Future. Am J Gastroenterol. 2023 Nov 1;118(11):1905-1908. doi: 10.14309/ajg.0000000000002302. Epub 2023 May 4. PMID: 37104667. 

15. Niren Jasutkar, Samir Shah, Steven Reinert, Murray Resnick, P-102 Chromoendoscopy in Community GI: A Single Physician Experience, Inflammatory Bowel Diseases, Volume 18, Issue suppl_1, 1 December 2012, Page S55, https://doi.org/10.1097/00054725-201212001-00135