IBD Dysplasia Surveillance: New Approaches

Finding the needle before it becomes cancer

Gastroenterology · Seminar week 12 · released June 8, 2026 · includes a discussion video

Chromoendoscopy, AI-assisted polyp detection, and molecular biomarkers. Practical approaches to enhancing dysplasia detection in IBD patients.

Learning Objectives

By the end of this seminar, participants will be able to:

1. Understand the current dysplasia surveillance guidelines for IBD and risk-based stratification.
2. Evaluate the efficacy of high-definition white light endoscopy versus chromoendoscopy and narrow band imaging in detecting dysplasia.
3. Implement risk-stratified surveillance schedules in clinical practice.
4. Discuss the role of emerging technologies, including artificial intelligence, in enhancing dysplasia detection and management.

Section 1: Overview of Dysplasia in Ulcerative Colitis and Crohn's Disease

Duration: 12 minutes

MUST ACT: Patients with long-standing inflammatory bowel disease carry a significantly elevated risk of colorectal cancer compared to the general population. The cumulative probability of CRC in UC patients has been estimated at 2% at 10 years, 8% at 20 years, and 18% at 30 years of disease duration. In Crohn's colitis, the risk is comparable when accounting for extent of colonic involvement (PMID: 33592179). Every IBD patient with colonic involvement must be enrolled in a structured surveillance programme. Failing to do so represents a missed opportunity to prevent cancer.

Teaching Point: Dysplasia in IBD does not follow the classic adenoma-carcinoma sequence seen in sporadic CRC. Instead, chronic inflammation drives a distinct pathway: chronic inflammation leads to indefinite dysplasia, which progresses to low-grade dysplasia (LGD), then high-grade dysplasia (HGD), and finally invasive carcinoma. This inflammation-dysplasia-carcinoma sequence means that the entire colonic mucosa is at risk — a concept known as "field cancerisation" — rather than isolated polyps as in sporadic disease (PMID: 31884010).

The histological classification of dysplasia in IBD follows the Riddell classification, which distinguishes negative for dysplasia, indefinite for dysplasia, low-grade dysplasia, and high-grade dysplasia. Villanacci et al., in the position statement of the Italian IG-IBD and GIPAD groups, emphasised the importance of standardised biopsy protocols and expert GI pathology review to reduce interobserver variability, which remains a significant challenge (PMID: 31884010).

Nuance: The risk of CRC in IBD is not uniform. Key risk factors include duration of disease greater than 8-10 years, extensive colonic involvement (pancolitis carries higher risk than left-sided disease), coexisting primary sclerosing cholangitis (PSC), family history of CRC, severity and chronicity of inflammation, and prior history of dysplasia. PSC-IBD patients deserve particular attention: their CRC risk is 4-fold higher than IBD patients without PSC, and surveillance should begin at the time of PSC diagnosis regardless of disease duration (PMID: 34416977).

Say Out Loud: "Every IBD patient with colonic disease for more than eight years needs to be in a surveillance programme. If they also have PSC, surveillance starts at diagnosis. Full stop."

Audience Poll 1

> What do you consider the single most important risk factor for dysplasia in IBD? > A) Duration of disease > B) Extent of colonic involvement > C) Coexisting PSC > D) Family history of CRC > E) Severity of inflammation

Decision Point: When a patient with long-standing UC presents for their first surveillance colonoscopy and has never been risk-stratified, you must assess all five risk factors (duration, extent, PSC status, family history, inflammation severity) before determining the surveillance interval and modality. This is not optional — it is the foundation upon which the entire surveillance strategy is built.

The Molecular Basis of IBD-Associated Dysplasia

Teaching Point: Understanding the molecular underpinnings of IBD-associated neoplasia helps explain why surveillance approaches differ from sporadic CRC screening. In sporadic CRC, the classic Vogelstein model proceeds through sequential mutations: APC loss (initiating adenoma), KRAS activation (adenoma growth), and TP53 loss (malignant transformation). In IBD-associated neoplasia, the molecular sequence is fundamentally different. TP53 mutations occur early — often detectable even in non-dysplastic mucosa that is chronically inflamed — while APC mutations are late events. This early loss of the TP53 tumour suppressor, combined with ongoing oxidative stress from chronic inflammation, creates a mutational environment across the entire colonic mucosa rather than in discrete polyps.

Additionally, epigenetic changes are prominent in IBD-associated neoplasia. DNA methylation of tumour-suppressor gene promoters (including CDKN2A, MLH1, and CDH1) has been detected in inflamed but non-dysplastic mucosa, suggesting that the molecular "clock" toward cancer begins ticking long before morphological changes are visible. This molecular field effect reinforces the clinical concept of field cancerisation and justifies the intensive, pancolonic surveillance approach used in IBD.

Nuance: Microsatellite instability (MSI) is common in sporadic CRC but relatively uncommon in IBD-associated CRC. Chromosomal instability and aneuploidy are the predominant pathways in IBD-associated neoplasia. This distinction has implications for future molecular surveillance strategies — for example, p53 immunohistochemistry on surveillance biopsies has been explored as a biomarker to identify patients at highest risk of progression, though it has not yet been incorporated into routine clinical practice.

Crohn's Disease vs Ulcerative Colitis: Specific Considerations

Teaching Point: While much of the surveillance literature focuses on UC, Crohn's disease with colonic involvement carries a comparable CRC risk when matched for extent of disease. However, Crohn's presents unique challenges for surveillance. Strictures may be difficult to traverse endoscopically, and dysplasia arising within strictures is particularly concerning because it cannot be reliably excluded without tissue sampling. Active perianal or small bowel disease may complicate the timing and logistics of surveillance. Post-surgical anatomy (ileocolonic anastomoses, strictureplasties) creates areas that require focused attention during surveillance.

The IOIBD (International Organization for the Study of Inflammatory Bowel Disease) addressed the question of prioritising endoscopy in IBD patients, including dysplasia surveillance, in the context of resource limitations (PMID: 33085973). Their framework identifies patients with a past history of dysplasia as the highest priority for timely endoscopy, underscoring the clinical weight given to prior dysplasia findings.

Section 2: Guideline Updates — Risk Stratification and Surveillance Intervals

Duration: 12 minutes

MUST ACT: The 2021 AGA Clinical Practice Update on Endoscopic Surveillance and Management of Colorectal Dysplasia in IBD (Murthy et al.) represents the current standard of care. The key best practice advice statements include: (1) Surveillance colonoscopy should begin 8 years after symptom onset for patients with colonic IBD; (2) High-definition colonoscopy with chromoendoscopy is recommended over standard-definition white light endoscopy; (3) For patients at increased risk (PSC, prior dysplasia, family history, active inflammation, stricture), surveillance should be performed annually; (4) For average-risk patients, every 1-3 years is appropriate depending on additional factors (PMID: 34416977).

Teaching Point: Risk stratification is not a one-time assessment. It must be revisited at every surveillance colonoscopy because risk factors evolve. A patient who was initially average-risk may develop PSC, have a new family history of CRC, or show persistent inflammation that elevates them to the high-risk category. The GETECCU position paper from Spain proposed a practical risk-stratification algorithm that divides patients into high-risk, intermediate-risk, and low-risk categories with corresponding surveillance intervals of 1, 2-3, and 5 years respectively (PMID: 33592179).

Nuance: There is ongoing international variation in guideline recommendations. The AGA, BSG (British Society of Gastroenterology), ECCO (European Crohn's and Colitis Organisation), and GETECCU guidelines broadly agree on the importance of chromoendoscopy and risk-based intervals, but differ in specific details. For example, the BSG recommends beginning surveillance after 10 years of symptoms (rather than 8), while ECCO emphasises the importance of confirming disease extent at the initial screening colonoscopy before setting intervals. Clinicians should be familiar with their local guideline recommendations while understanding the shared evidence base.

The German S3 guideline on ulcerative colitis (Kucharzik, 2022) additionally emphasises that a proctocolectomy should be considered in cases of confirmed high-grade epithelial dysplasia or carcinoma, reinforcing the surgical decision-making that surveillance ultimately informs (PMID: 35166864).

Case 1: The Long-Standing UC Patient with Evolving Risk

Presentation: A 48-year-old woman with a 15-year history of left-sided ulcerative colitis presents for routine surveillance. She was diagnosed at age 33 and has had two prior surveillance colonoscopies — both negative for dysplasia — performed with standard-definition white light endoscopy. At her most recent clinic visit, she mentions that her father was diagnosed with colon cancer at age 62.

Decision Point: How does this new family history change her surveillance plan?

Teaching Point: This patient has moved from the average-risk to the higher-risk category due to first-degree family history of CRC. Her surveillance interval should be shortened to every 1 year, and the modality should shift from standard-definition WLE (which should no longer be used in any case) to high-definition colonoscopy with chromoendoscopy. Additionally, a conversation about optimising disease control and mucosal healing becomes critical, as persistent inflammation compounds her elevated risk (PMID: 34416977).

Say Out Loud: "Risk stratification is a living document. Every time you scope an IBD patient, reassess their risk factors. New family history? New PSC diagnosis? Ongoing inflammation? The surveillance interval must adapt."

Audience Poll 2

> Which guideline most closely aligns with your current practice? > A) AGA 2021 > B) BSG > C) ECCO > D) I follow local institutional protocols > E) I'm not sure which guideline I follow

Section 3: Endoscopic Techniques — High-Definition White Light, Chromoendoscopy, and Narrow Band Imaging

Duration: 15 minutes

MUST ACT: The evidence is clear and the guidelines are unified: high-definition endoscopy with chromoendoscopy is superior to standard-definition white light endoscopy for dysplasia detection in IBD. The AGA Clinical Practice Update explicitly recommends chromoendoscopy-assisted colonoscopy over standard-definition WLE (PMID: 34416977). The GETECCU position paper independently reached the same conclusion, recommending chromoendoscopy as the preferred technique and describing in detail how its implementation should be carried out in endoscopic units (PMID: 33592179).

Teaching Point: Chromoendoscopy involves the topical application of contrast dyes — most commonly methylene blue (0.1%) or indigo carmine (0.03-0.3%) — to the colonic mucosa during colonoscopy. These dyes pool in mucosal grooves and highlight surface irregularities, making subtle flat lesions visible that would be missed by white light alone.

The technique adds approximately 10-15 minutes to the procedure and requires additional training, but the yield is substantial. Meta-analyses have consistently shown that chromoendoscopy detects 44% more dysplastic lesions compared to standard WLE, with a number needed to scope of approximately 14 to detect one additional case of dysplasia.

Decision Point: When choosing between chromoendoscopy and virtual chromoendoscopy techniques like narrow band imaging (NBI):

Chromoendoscopy advantages:

• Higher detection rate for flat and subtle lesions
• Better for detecting non-polypoid (Paris 0-IIb) dysplasia
• Recommended by all major guidelines as the preferred technique
• Can be performed with any high-definition endoscope

NBI/virtual chromoendoscopy considerations:

• Faster and more convenient (no dye preparation required)
• Some studies show non-inferiority when combined with high-definition endoscopes
• May be acceptable as an alternative when chromoendoscopy expertise is unavailable
• The GETECCU group specifically noted that chromoendoscopy remains superior to NBI (PMID: 33592179)

Nuance: The debate between dye-based chromoendoscopy and virtual chromoendoscopy (NBI, FICE, iScan) is not fully settled. The SCENIC consensus statement acknowledges that when high-definition endoscopy is used, the incremental benefit of dye-based chromoendoscopy over virtual techniques may be smaller than initially reported. However, in the specific context of IBD surveillance, where field cancerisation and flat lesions are prevalent, the majority of experts continue to recommend dye-based chromoendoscopy as the gold standard. The quality of the exam — including bowel preparation, withdrawal time, and endoscopist training — matters as much as the technique itself.

Practical Chromoendoscopy Technique

Teaching Point: For clinicians new to chromoendoscopy, the practical technique involves:

1. Preparation: Excellent bowel preparation is essential — target a Boston Bowel Preparation Scale (BBPS) score of 6 or higher. Poor preparation with residual stool or debris obscures the dye pattern and negates the benefit of chromoendoscopy. Consider using a split-dose preparation regimen and, if necessary, additional on-table washing during the procedure.

1. Dye selection and concentration: Methylene blue at 0.1% concentration or indigo carmine at 0.03-0.3%. Methylene blue is absorbed by normal mucosa and highlights abnormal areas as unstained. Indigo carmine is not absorbed but pools in mucosal grooves, providing topographic contrast. Both are effective; institutional preference and availability typically guide the choice.

1. Application technique: Use a spray catheter (e.g., Olympus PW-5V1 or similar) passed through the working channel. Apply dye segmentally during withdrawal — spray one colonic segment, wait 30-60 seconds for optimal contrast development, then inspect carefully under HD white light before moving to the next segment. Excess dye should be suctioned.

1. Inspection pattern: Examine the mucosa systematically, looking for surface pattern irregularities, subtle flat elevations (Paris 0-IIa), colour changes, and loss of normal vascular pattern. The Kudo pit pattern classification can help characterise lesion morphology: Type I-II pits are non-neoplastic, Type IIIS/IIIL and IV suggest adenomatous change, and Type V (irregular/non-structural) is highly suspicious for invasive disease.

1. Withdrawal time: A minimum withdrawal time of 20 minutes is recommended for chromoendoscopy surveillance. This is longer than the standard 6-minute minimum for screening colonoscopy and reflects the need for thorough segmental dye application and inspection.

MUST ACT: The minimum standards for an adequate IBD surveillance colonoscopy include: high-definition endoscope, chromoendoscopy technique, BBPS greater than or equal to 6, withdrawal time greater than or equal to 20 minutes, documentation of extent of inflammation, Paris classification of any lesions found, and targeted biopsies of all suspicious areas. If any of these standards cannot be met during a given procedure, the examination should be considered suboptimal and repeated.

Say Out Loud: "If you are performing IBD surveillance with a standard-definition scope and no chromoendoscopy, you are performing substandard surveillance. The evidence has been clear for over a decade. Upgrade the scope, learn chromoendoscopy, and detect what you are currently missing."

Audience Poll 3

> Have you used chromoendoscopy in your clinical practice? > A) Yes, routinely for IBD surveillance > B) Yes, occasionally > C) No, but I plan to start > D) No, and I'm not planning to > E) I use NBI/virtual chromoendoscopy instead

Case 2: The Chromoendoscopy Conversion

Presentation: A 55-year-old man with 20-year pancolitis has had four prior surveillance colonoscopies with standard-definition WLE — all negative. He presents for his fifth surveillance. This time, the endoscopist uses HD colonoscopy with chromoendoscopy (0.1% methylene blue). A subtle 8mm flat lesion (Paris 0-IIa) is identified in the ascending colon that was not apparent on withdrawal with white light alone.

Teaching Point: Targeted biopsies of the lesion show low-grade dysplasia. This lesion was likely present on prior examinations but invisible to standard WLE. Chromoendoscopy identified it because the dye pooling highlighted the subtle mucosal irregularity. This case illustrates why the technique of surveillance matters as much as the frequency — four "negative" colonoscopies performed with the wrong technique provide false reassurance (PMID: 34416977).

Section 4: Classification and Management of Visible vs Invisible Dysplasia

Duration: 15 minutes

MUST ACT: The management of dysplasia in IBD hinges on whether the dysplasia is visible (endoscopically identifiable) or invisible (detected only on random biopsies). The 2021 AGA update fundamentally shifted the approach: visible dysplasia that is completely resected endoscopically can be managed with surveillance rather than colectomy, provided the resection margins are clear and subsequent biopsies of the surrounding mucosa are negative for dysplasia (PMID: 34416977).

Teaching Point: The Paris classification is essential for describing visible lesions:

Flat (0-IIb) and depressed (0-IIc) lesions are the most challenging to detect and carry the highest risk of submucosal invasion. These are the lesions that chromoendoscopy excels at identifying compared to white light endoscopy.

Decision Point: When dysplasia is found, the critical management decision depends on three factors:

1. Is the dysplasia visible or invisible?

• Visible: Can it be completely resected endoscopically?
• Invisible: Requires confirmation with repeat chromoendoscopy by an experienced endoscopist

1. What is the grade of dysplasia?

• Low-grade: Surveillance after complete endoscopic resection (if visible)
• High-grade: Strong consideration for colectomy, especially if multifocal or unresectable

1. Are there features suggesting non-resectability?

• Ill-defined borders
• Non-lifting sign (submucosal fibrosis)
• Located in area of active inflammation
• Multifocal lesions

MUST ACT: Invisible dysplasia — that is, dysplasia detected on random four-quadrant biopsies without a visible lesion — should prompt repeat colonoscopy with chromoendoscopy by an experienced endoscopist. If the repeat examination with optimised technique still fails to identify a visible lesion, multidisciplinary discussion (gastroenterologist, pathologist, surgeon) is mandatory. Confirmed invisible multifocal dysplasia or invisible high-grade dysplasia generally favours colectomy (PMID: 34416977).

Nuance: The concept of "invisible dysplasia" has been challenged in the chromoendoscopy era. Studies suggest that many lesions previously classified as invisible on standard WLE become visible when chromoendoscopy is used. This raises the question of whether truly invisible dysplasia is rarer than we thought, or whether it was primarily an artefact of inadequate examination technique. The practical implication is that before labelling dysplasia as "invisible," the patient must have been examined with optimal technique — HD endoscopy with chromoendoscopy in a well-prepared colon by an experienced endoscopist.

Endoscopic Resection Techniques in IBD

Teaching Point: When a visible dysplastic lesion is identified and deemed resectable, the choice of resection technique depends on lesion characteristics:

Endoscopic Mucosal Resection (EMR): The standard technique for lesions up to 20mm. Involves submucosal injection to create a fluid cushion beneath the lesion, followed by snare resection. En bloc resection is preferred for lesions under 20mm. For larger lesions, piecemeal EMR is acceptable but requires careful documentation of completeness and closer follow-up (typically 3-6 months) to assess for residual tissue.

Endoscopic Submucosal Dissection (ESD): Allows en bloc resection of larger lesions (>20mm) and lesions with morphological features suggesting submucosal invasion. ESD provides more complete pathological assessment because the specimen is intact. However, ESD in the context of IBD colitis is technically more challenging than in normal mucosa due to submucosal fibrosis from chronic inflammation, which can obscure the dissection plane and increase the risk of perforation.

Decision Point: Key factors determining endoscopic resectability include:

• Border definition: Well-defined borders are essential. Ill-defined or irregular borders suggest a non-resectable lesion.
• Lifting sign: Submucosal injection should create adequate lift. The non-lifting sign (failure to elevate with submucosal injection) suggests submucosal invasion and generally precludes endoscopic resection.
• Size and location: Lesions in areas of active inflammation are more challenging to resect and more likely to have unclear margins.
• Morphology: Depressed (Paris 0-IIc) lesions are more likely to harbour submucosal invasion and may warrant surgical referral even if technically resectable.

Teaching Point: After endoscopic resection of dysplasia in IBD, the follow-up protocol includes: confirmation of clear histological margins, biopsies of the surrounding flat mucosa to exclude synchronous invisible dysplasia, follow-up colonoscopy with chromoendoscopy at 3-6 months to assess the resection site, and then annual surveillance colonoscopies with chromoendoscopy thereafter. The resection site should be tattooed to allow identification on follow-up examinations.

Say Out Loud: "Before you tell a patient they have invisible dysplasia and need a colectomy, make sure they've had a proper chromoendoscopy. What is invisible under white light may be perfectly visible under dye."

Case 3: Visible Dysplasia — Endoscopic Resection vs Surgery

Presentation: A 52-year-old man with 18-year pancolitic UC is found to have a 15mm Paris 0-Is sessile lesion in the sigmoid colon during surveillance chromoendoscopy. The borders are well-defined, the lesion lifts with submucosal injection, and it is completely resected en bloc by EMR. Histopathology confirms LGD with clear resection margins. Biopsies of the flat mucosa surrounding the resection site and from each colonic segment are negative for dysplasia.

Decision Point: Does this patient need a colectomy?

Teaching Point: No. Per AGA 2021 recommendations, this patient can continue with intensified endoscopic surveillance (every 1 year) rather than proceeding to colectomy. The key criteria are met: (1) the lesion was visible and well-defined, (2) it was completely resected with clear margins, (3) surrounding mucosa biopsies are negative, and (4) the patient has no other high-risk features such as PSC or multifocal dysplasia. This represents a major shift from older guidelines that recommended colectomy for any dysplasia in IBD (PMID: 34416977).

Case 4: The Invisible Dysplasia Dilemma

Presentation: A 60-year-old woman with 25-year pancolitic UC undergoes surveillance colonoscopy with standard-definition WLE (the endoscopy unit has not yet adopted chromoendoscopy). Random four-quadrant biopsies reveal LGD in the transverse colon. No visible lesion was identified.

Decision Point: What is the next step?

Teaching Point: This patient must not be sent directly to the surgeon. The first step is repeat colonoscopy with HD chromoendoscopy by an experienced endoscopist to determine whether a visible lesion can be identified at the site of the dysplastic biopsy. If a visible lesion is found, it should be assessed for endoscopic resectability. Only if repeat chromoendoscopy confirms truly invisible dysplasia should multidisciplinary discussion about colectomy proceed (PMID: 34416977; PMID: 33592179).

Section 5: Future Directions — Artificial Intelligence and Evolving Biopsy Protocols

Duration: 10 minutes

Teaching Point: Artificial intelligence is rapidly emerging as a potential tool to augment dysplasia detection in IBD surveillance. AI-based computer-aided detection (CADe) and computer-aided diagnosis (CADx) systems have shown promising results in the detection and characterisation of sporadic colorectal polyps. The application to IBD-associated dysplasia is an area of active research, though significant challenges remain.

Current AI systems for colonoscopy have been trained predominantly on sporadic polyps. IBD-associated dysplasia presents unique challenges: flat morphology, inflammatory background that mimics and obscures dysplasia, post-inflammatory polyps (pseudopolyps) that generate false positives, and the need to assess the entire mucosal surface rather than discrete lesions. An AI system optimised for IBD surveillance would need to integrate mucosal pattern recognition with contextual information about inflammation and disease extent.

Nuance: The potential of AI in IBD dysplasia surveillance extends beyond simple lesion detection. Future applications may include: (1) real-time guidance for optimal biopsy sites, reducing the number of random biopsies needed; (2) quantitative assessment of inflammation severity to predict dysplasia risk; (3) automated quality metrics for surveillance colonoscopy (withdrawal time, mucosal inspection coverage, preparation quality); (4) virtual chromoendoscopy enhancement through AI post-processing of white-light images.

However, significant barriers exist. Regulatory approval for IBD-specific AI tools lags behind general polyp detection. Training datasets for IBD dysplasia are far smaller than for sporadic polyps. And there is legitimate concern that over-reliance on AI could reduce the endoscopist's skill in pattern recognition — a problem if the AI fails or is unavailable.

Decision Point: The evolving biopsy protocol question: random biopsies versus targeted biopsies alone. The traditional protocol of 33 or more random four-quadrant biopsies (every 10 cm) has been challenged. When chromoendoscopy is used, targeted biopsies of visible lesions may be sufficient, as the yield of random biopsies in the absence of visible abnormalities is extremely low (approximately 0.1-0.2% per biopsy). However, guidelines have not yet fully endorsed abandoning random biopsies, and many expert endoscopists continue to perform both targeted and random biopsies. This is an area where clinical practice is evolving faster than formal guideline updates.

Molecular Biomarkers and the Future of Non-Endoscopic Surveillance

Teaching Point: Beyond AI-enhanced endoscopy, the field is exploring molecular biomarkers that could complement or even partially replace endoscopic surveillance. Stool DNA testing, which has been validated for sporadic CRC screening (Cologuard), is being investigated for IBD-associated neoplasia, though the inflammatory background creates significant noise that reduces specificity. Blood-based circulating tumour DNA (ctDNA) assays and methylation panels are in early-phase studies. The concept of a "liquid biopsy" for IBD dysplasia surveillance — drawing blood instead of performing colonoscopy — remains aspirational but is the subject of active research.

Nuance: Confocal laser endomicroscopy (CLE) represents another advanced imaging modality that provides real-time histological-level imaging during colonoscopy. CLE can distinguish neoplastic from non-neoplastic tissue at the cellular level and has shown high sensitivity and specificity for dysplasia detection in IBD. However, the technology requires specialised equipment and training, is time-consuming, and has limited field of view, making it more suitable as an adjunctive tool for characterising suspicious lesions rather than as a primary screening modality.

Say Out Loud: "AI will not replace the trained endoscopist in IBD surveillance anytime soon. But it may make all of us better by reducing the lesions we miss. The technology is coming — stay informed, but stay skilled."

Audience Poll 4

> How confident are you in the potential of AI to improve dysplasia detection in IBD? > A) Very confident — it will be transformative > B) Cautiously optimistic > C) Sceptical — too many challenges > D) I need to learn more about this > E) Not applicable to my practice

Section 6: Interdisciplinary Approaches for Comprehensive Surveillance

Duration: 10 minutes

Teaching Point: Dysplasia surveillance in IBD is not a solo act. It requires coordinated input from gastroenterologists, GI pathologists, colorectal surgeons, and IBD nurse specialists. The ACR Appropriateness Criteria for Crohn disease (Moore et al., 2022) emphasise the importance of imaging and clinical collaboration across specialties for optimal disease monitoring (PMID: 35550801).

The pathologist plays a particularly critical role. Interobserver variability in dysplasia grading remains significant — studies have shown agreement rates of only 40-70% for LGD. Expert GI pathologist review can change the diagnosis in up to 30% of cases. Any dysplasia diagnosis that will drive a management decision (surveillance interval change, consideration of colectomy) should be confirmed by a second expert GI pathologist (PMID: 31884010).

Nuance: The multidisciplinary team (MDT) meeting is the gold standard for complex dysplasia management decisions. Cases that should be discussed in MDT include: invisible dysplasia confirmed on repeat chromoendoscopy, multifocal LGD, any HGD, dysplasia in the setting of a stricture, dysplasia found in a patient with PSC, and discordant pathology opinions. The MDT allows integration of endoscopic findings, pathological diagnosis, radiological assessment, patient preferences, and surgical risk to arrive at an individualised management plan.

Say Out Loud: "If you find dysplasia in an IBD patient and you are not sure what to do, the answer is always 'bring it to the MDT.' These are not decisions to make alone."

The Role of the GI Pathologist: Overcoming Interobserver Variability

Teaching Point: The GI pathologist is arguably the most critical member of the dysplasia management team, because the histological diagnosis drives every subsequent decision. Yet dysplasia grading in IBD is subjectively challenging. The distinction between "reactive atypia" (inflammation-induced cellular changes that are not dysplastic) and true low-grade dysplasia can be extremely difficult. Studies of interobserver agreement show kappa values of only 0.4-0.6 for LGD, meaning that even experienced pathologists disagree in a substantial proportion of cases.

The practical implications are significant. A diagnosis of LGD may trigger intensified surveillance, repeat chromoendoscopy, or referral for MDT discussion. A diagnosis of reactive atypia changes nothing. And yet these two diagnoses may be separated by the subjective interpretation of a single pathologist examining a few millimetres of tissue. This is why expert review is not a luxury — it is a clinical necessity.

The Italian position statement by Villanacci et al. recommended that all dysplasia diagnoses in IBD should be confirmed by at least two pathologists, ideally with expertise in GI pathology (PMID: 31884010). Immunohistochemistry for p53 can assist — overexpression or complete loss of p53 staining supports true dysplasia over reactive change — but it is not definitive on its own.

Nuance: The category "indefinite for dysplasia" deserves specific attention. This diagnosis means the pathologist cannot determine whether the cellular changes represent true dysplasia or reactive atypia from active inflammation. The recommended response is to optimise medical therapy to achieve mucosal healing, then repeat surveillance colonoscopy with chromoendoscopy in 3-6 months. If the "indefinite" diagnosis persists after inflammation is controlled, the risk of underlying true dysplasia is higher, and the case should be discussed at MDT.

Surgical Decision-Making: When to Recommend Colectomy

Teaching Point: Indications for recommending proctocolectomy in the setting of IBD-associated dysplasia include: confirmed invisible multifocal LGD (after adequate chromoendoscopy), any confirmed invisible HGD, visible HGD that is not amenable to endoscopic resection, dysplasia within a stricture that cannot be adequately sampled or visualised, and patient preference after informed discussion of the endoscopic surveillance alternative. The German S3 guideline explicitly recommends colectomy for confirmed HGD or carcinoma (PMID: 35166864).

The surgical options include total proctocolectomy with ileal pouch-anal anastomosis (IPAA), which preserves continence and is the preferred approach in younger patients, or total proctocolectomy with end ileostomy. The choice depends on patient factors including age, comorbidities, sphincter function, and personal preference. Post-IPAA, the ileal pouch mucosa itself is at low but non-zero risk of dysplasia, and annual pouch surveillance is recommended.

Case 5: The PSC-IBD Patient — When Everything is High-Risk

Presentation: A 35-year-old man with Crohn's colitis and primary sclerosing cholangitis presents for his annual surveillance colonoscopy. He was diagnosed with PSC at age 28 and IBD at age 30. HD chromoendoscopy reveals two subtle flat lesions (Paris 0-IIa, 6mm and 9mm) in the right colon. Targeted biopsies show LGD in both lesions. Random biopsies from remaining segments are negative.

Decision Point: This patient has multiple high-risk features: PSC (4-fold increased CRC risk), Crohn's colitis with extensive involvement, young age with decades of future risk, and now multifocal LGD. What is the management approach?

Teaching Point: This case illustrates the importance of risk-factor aggregation. The endoscopic findings alone (two small, resectable LGD lesions) might suggest that endoscopic resection and surveillance is appropriate. However, the PSC-IBD combination fundamentally changes the calculus. The AGA update recommends annual surveillance for PSC-IBD patients, and the finding of multifocal LGD in this high-risk context warrants urgent MDT discussion including the hepatologist, pathologist, colorectal surgeon, and gastroenterologist. Options range from complete endoscopic resection with 3-6 month follow-up chromoendoscopy, to proctocolectomy with ileal pouch-anal anastomosis. The patient's values, hepatological status (liver transplant candidacy), and surgical risk must all be weighed (PMID: 34416977).

Section 7: Practical Implementation — Making It Work in Your Endoscopy Unit

Duration: 8 minutes

Teaching Point: Knowledge of guidelines is necessary but not sufficient. The real challenge is implementing evidence-based surveillance in your local endoscopy unit. Practical steps include:

1. Equipment: Ensure all colonoscopes used for IBD surveillance are high-definition. Budget for dye spray catheters and methylene blue or indigo carmine dye.
2. Training: Identify 2-3 endoscopists per unit who will develop expertise in chromoendoscopy. Attend hands-on training courses. Review SCENIC and GETECCU guidance on technique.
3. Protocol: Develop a written IBD surveillance protocol with risk-stratification criteria, required bowel preparation quality (Boston Bowel Preparation Scale greater than or equal to 6), minimum withdrawal time (greater than or equal to 20 minutes for chromoendoscopy), and standardised reporting.
4. Pathology: Establish a pathway for expert GI pathologist review of all dysplastic biopsies. Create a reflex second-opinion policy for all LGD and indefinite for dysplasia diagnoses.
5. MDT: Set up a regular IBD dysplasia MDT meeting, even if initially only monthly.

MUST ACT: The GETECCU position paper provides a detailed implementation framework for chromoendoscopy in endoscopy units, describing exactly how the procedure should be carried out, in which patients, and at what intervals. This level of practical guidance is invaluable for units transitioning from WLE-based surveillance to chromoendoscopy (PMID: 33592179).

Quality Assurance and Audit

Teaching Point: Implementation without audit is incomplete. Key quality metrics for IBD surveillance that should be tracked include:

• Chromoendoscopy completion rate: What percentage of IBD surveillance colonoscopies in your unit are performed with chromoendoscopy? Target: 100% for patients meeting criteria.
• Bowel preparation adequacy rate: What percentage achieve BBPS 6 or higher? Poor preparation should trigger rescheduling rather than proceeding with a substandard examination.
• Withdrawal time: Are you consistently achieving 20 minutes or more? This can be tracked through endoscopy reporting software.
• Dysplasia detection rate: What is your unit's dysplasia detection rate per surveillance colonoscopy? This metric helps identify whether your technique is finding what is expected.
• Paris classification documentation: Are all lesions being described using standardised terminology?
• Expert pathology review rate: What percentage of dysplasia diagnoses are reviewed by a second expert? Target: 100%.
• Surveillance interval adherence: Are patients returning at their recommended interval, or are they overdue? A recall system is essential.

Nuance: The concept of a "quality IBD surveillance colonoscopy" parallels the quality metrics developed for screening colonoscopy (adenoma detection rate, caecal intubation rate, withdrawal time). However, IBD surveillance quality metrics are less well-defined and less widely reported. Developing and reporting these metrics at the unit level is an opportunity for quality improvement and patient safety.

Patient Communication and Shared Decision-Making

Teaching Point: Surveillance is a lifelong commitment that requires patient understanding and engagement. Patients should understand why they need surveillance (elevated cancer risk), what the procedure involves (including chromoendoscopy), what the possible findings are (no dysplasia, LGD, HGD), and what the implications of each finding would be. Shared decision-making is particularly important when dysplasia is found and the choice is between continued endoscopic surveillance and surgery. The IBD nurse specialist plays a vital role in patient education, coordination of appointments, and ongoing support.

Say Out Loud: "It is not enough to know that chromoendoscopy is better. You have to make it happen in your unit. Buy the dye, train the endoscopists, write the protocol, and audit the outcomes."

Audience Poll 5

> What is the biggest barrier to implementing chromoendoscopy in your endoscopy unit? > A) Lack of training > B) Equipment/dye costs > C) Added procedure time > D) Institutional resistance to change > E) Unclear local guidelines

Tonight on Shift: Actionable Checklist

1. Identify and risk-stratify every IBD patient with colonic disease. Duration >8 years, extent, PSC status, family history of CRC, inflammation severity, prior dysplasia. This determines the surveillance interval and modality.

1. Use high-definition endoscopy with chromoendoscopy for all IBD surveillance. Standard-definition WLE alone is substandard surveillance. If your unit does not offer chromoendoscopy, refer the patient to a centre that does.

1. Classify visible lesions using the Paris classification and assess resectability before deciding between endoscopic management and surgery. Complete resection with clear margins permits ongoing surveillance.

1. Do not label dysplasia as "invisible" until the patient has had HD chromoendoscopy. What is invisible under WLE may be visible under dye.

1. Bring complex dysplasia cases to the MDT. PSC-IBD, multifocal LGD, any HGD, invisible dysplasia, discordant pathology — these are not solo decisions.

1. Get expert pathology review for all dysplasia diagnoses. Interobserver variability is significant. A second expert opinion can change the diagnosis in up to 30% of cases.

1. Reassess risk at every surveillance colonoscopy. Risk factors evolve. Family history changes, PSC can develop, inflammation may persist. The surveillance plan must be dynamic.

1. Implement chromoendoscopy in your unit. Buy the dye, train the endoscopists, write the protocol, audit outcomes. The evidence has been clear for over a decade.

References

1. Murthy SK, Feuerstein JD, Nguyen GC, et al. AGA Clinical Practice Update on Endoscopic Surveillance and Management of Colorectal Dysplasia in Inflammatory Bowel Diseases: Expert Review. Gastroenterology. 2021;161(3):1043-1051. PMID: 34416977

1. Sicilia B, Vicente R, Arias L, et al. Recommendations of the Spanish Working Group on Crohn's disease and Ulcerative Colitis (GETECCU) on dysplasia screening in inflammatory bowel disease patients. Gastroenterologia y hepatologia. 2021;44(6):435-447. PMID: 33592179

1. Villanacci V, Reggiani-Bonetti L, Caprioli F, et al. Histopathology of inflammatory bowel disease — Position statement of the Pathologists of the Italian Group for the Study of Inflammatory Bowel Disease (IG-IBD) and Italian Group of Gastrointestinal Pathologists (GIPAD-SIAPEC). Dig Liver Dis. 2020;52(3):262-267. PMID: 31884010

1. Kucharzik T. Living guideline on ulcerative colitis. Der Chirurg. 2022;93(3):236-248. PMID: 35166864

1. Moore MM, Gee MS, et al. ACR Appropriateness Criteria: Crohn Disease — Child. J Am Coll Radiol. 2022;19(5S):S137-S152. PMID: 35550801

1. Ng SC, Mak JWY, Hitz L, et al. COVID-19 Pandemic: Which IBD Patients Need to Be Scoped. J Crohns Colitis. 2020;14(Suppl 3):S714-S720. PMID: 33085973

Read this seminar as Markdown · All seminars · Lecture library · Question bank