UEG's content platform for gastroenterology education bringing recordings, courses, guidelines, and more together in one place.

Re-live UEG Week 2026 on Gutflix Gut Guide online What's new on Gutflix

Trending on Gutflix

Rate for better recommendations!

Your feedback helps us show you more of what you like and less of what you dislike!

More of this!

We will show you more similar content.

Not for me!

We won't suggest this again.

Bookmark!

Save this content for later!

New to myUEG? Create an account! Log In with myUEG account

UEG Poster

Standard Poster

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

This content is part of Gutflix. Log in with your myUEG account, or create one free, to watch it.

Not sure what you can access? Learn more about account types.

DEVELOPMENT AND INTERNAL-EXTERNAL VALIDATION OF A DYNAMIC MULTIVARIABLE PREDICTION MODEL FOR ADVANCED COLORECTAL NEOPLASIA IN PATIENTS WITH INFLAMMATORY BOWEL DISEASE

Anouk M. Wijnands 1, Bas B. L. Penning de Vries 2, Maurice Lutgens 3, Zeinab Bakhshi 4, Ibrahim Al Bakir 5, Laurent Beaugerie 6, Charles N. Bernstein 7, Chang Ho Ryan Choi 8, Nayantara Coelho-Prabhu 9, Trevor A. Graham 10, Ailsa L Hart 11, Joren R. ten Hove 1, Unknown Unknown 12, Julien Kirchgesner 6, Erik Mooiweer 13, Seth R. Shaffer 7, Shailja C. Shah 14, Sjoerd G. Elias 2, Bas Oldenburg 1

Affiliations

1 University Medical Centre Utrecht, Utrecht, Netherlands

2 Department of Epidemiology, Julius Center for Health Sciences and Primary Care, University Medical Centre Utrecht, Utrecht University, Utrecht, Netherlands

3 Elisabeth-TweeSteden Hospital, Tilburg, Netherlands

4 Mayo Clinic, Rochester, United States

5 Chelsea and Westminster Hospital, London, United Kingdom

6 Hôpital Saint-Antoine, Paris, France

7 Inflammatory Bowel Disease Clinical and Research Centre and Department of Internal Medicine, Max Rady College of Medicine, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, Canada

8 Gastroenterology & Hepatology, St. George Hospital, Sydney, Australia

9 Department of Gastroenterology and Hepatology, Mayo Clinic, Rochester, United States

10 Evolution and Cancer Laboratory, Barts Cancer Institute, Queen Mary University of London and Genomics and Evolutionary Dynamics Laboratory, Centre for Evolution and Cancer, Institute of Cancer Research, London, United Kingdom

11 Inflammatory Bowel Disease Unit, St Mark's Hospital, London, United Kingdom

12 Icahn School of Medicine at Mount Sinai, New York, United States

13 Hospital St Jansdal, Harderwijk, Netherlands

14 Division of Gastroenterology, University of California San Diego; GI Section, VA San Diego Healthcare Center, San Diego, United States

Summary

AI Generated

A dynamic multivariable prediction model for advanced colorectal neoplasia in inflammatory bowel disease patients, based on eight clinical predictors and pooled data from six surveillance cohorts, demonstrated moderate to good discrimination and good calibration.

The model includes eight predictors: extensive disease, prior dysplasia, primary sclerosing cholangitis, male sex, IBD type, post-inflammatory polyps, age, and maximal endoscopic inflammation.
Among 3,731 IBD patients followed for a median of 5.7 years (26,336 patient-years), 146 individuals developed advanced colorectal neoplasia (57 high-grade dysplasia, 89 colorectal cancer).
The model achieved a c-statistic of 0.74 for 5-year and 0.75 for 10-year prediction windows, with 90% of 10-year predicted risks between 0.9% and 18.7%.
Model discrimination improved when restricted to data from 2010 onwards; generalizability results varied across cohorts and should be interpreted with caution due to varying cohort sizes.
The authors state that formal external validation, relating predicted risks to surveillance intervals, and evaluation in clinical practice are needed before clinical application.

Download PDF

Was this helpful?

Thanks for your feedback.

This summary was generated by an AI large language model based on the content transcript. It is for informational purposes only and should not be considered a substitute for clinical judgment. Always rely on your professional expertise and the full clinical context when making clinical decisions.

Abstract

Introduction

Surveillance is recommended in patients with colonic inflammatory bowel disease (IBD) given their increased risk of colorectal cancer (CRC). Present risk stratification strategies do not discriminate sufficiently between risk groups. We aimed to develop and validate a dynamic multivariable prediction model for advanced colorectal neoplasia (aCRN, including high-grade dysplasia and CRC) in IBD.

Aims & Methods

We pooled data from six existing surveillance cohort studies (two unpublished).^1-3 Patients with IBD and an indication for CRC surveillance were included if they underwent at least one follow-up procedure. Exclusion criteria included prior aCRN, prior colectomy, or an unclear indication for surveillance. Predictors were selected based on the results of a recent meta-analysis,⁴ frequency, and data availability: extensive disease, prior dysplasia (indefinite for dysplasia or low-grade dysplasia), primary sclerosing cholangitis, male sex, IBD type, post-inflammatory polyps, age, and maximal endoscopic inflammation. A dynamic prediction model was developed using a landmarking approach based on Cox proportional hazard modelling.⁵Model performance was assessed with Harrell’s c-statistic and by calibration curves, adjusted for optimism by internal cross-validation. In a sensitivity analysis model performance was assessed in data ≥2010 only (internal validation). Generalizability across the surveillance cohorts was evaluated by internal-external cross-validation.

Results

The surveillance cohorts encompassed time periods 1973-2021 and comprised 3,731 patients (54% male sex; 75% ulcerative colitis or IBD-unclassified and 25% Crohn’s disease) with a median follow-up of 5.7 years (interquartile interval 3.3-9.6; 26,336 patient-years of follow-up); 146 individuals were diagnosed with aCRN (see table 1 for patient characteristics). The final model contained eight predictors, with a cross-validation median c-statistic of 0.74 and 0.75 and calibration slope of 0.77 and 0.83 for, respectively, a 5- and 10-year prediction window in overall analyses. Calibration plots showed good calibration. Ninety percent of 10-year predicted aCRN risks were between 0.9 and 18.7%. Model discrimination improved in data ≥2010 and calibration results remained unchanged (48% of all procedures ≥2010). Internal-external cross-validation results showed medium discrimination and reasonable to good calibration.

	Total n=3.731	Cohort A n=874	Cohort B n=407	Cohort C n=1.163	Cohort D n=306	Cohort E n=811	Cohort F n=170
Hospital; country	-	St. Mark’s Hospital; UK	Mount Sinai; USA	Five tertiary and two secondary hospitals; NL	Four tertiary hospitals; NL	Mayo Clinic; USA	University of Manitoba; CA
Male sex, n (%) Missing	2.024 (54) 7	487 (56) 7	203 (50)	616 (53)	155 (51)	475 (59)	88 (52)
Age (y), median (interquartile interval (IQI))	47 (36-56)	45 (36-54)	41 (31-55)	46 (37-55)	49 (37-58)	51 (40-62)	42 (30-52)
IBD disease duration (y), median (IQI)	13 (9-21)	11 (9-17)	13 (9-21)	14 (10-21)	15 (9-24)	16 (10-24)	10 (8-14)
Extensive disease extent (E3 montreal or >50% colonic involvement), n (%)	2.946 (79)	874 (100)	262 (64)	936 (80)	261 (85)	485 (60)	128 (75)
Primary sclerosing cholangitis (study index or follow-up), n (%)	518 (14)	36 (4)	86 (21)	157 (13)	26 (8)	202 (25)	11 (6)
Prior dysplasia (indefinite for dysplasia and/or low-grade dysplasia), n (%)	558 (15)	41 (5)	64 (16)	175 (15)	64 (8)	211 (26)	6 (4)
Post-inflammatory polyps, n (%)	675 (18)	171 (20)	83 (20)	226 (19)	81 (26)	73 (9)	41 (24)
Events advanced colorectal neoplasia (prior to or at colectomy) High-grade dysplasia Colorectal cancer	57 89	12 36	15 14	14 15	1 4	13 19	2 1

Table 1: Characteristics of 3.731 included IBD patients at study index and procedure details (total cohort and per database)
* Summary data does not include all cohorts as variable is not present in one or more cohorts.

Conclusion

We generated a model based on clinical factors that predicts future aCRN risk, with moderate to good model discrimination and good calibration. However, generalizability results vary and are impacted by varying cohort sizes and should be interpreted with caution. Future research should focus on formal external validation, relate predicted aCRN risks to surveillance intervals, and investigate the impact of this new model in daily clinic prior to clinical application.

References

1. Choi CR, Al Bakir I, Ding NJ, et al. Cumulative burden of inflammation predicts colorectal neoplasia risk in ulcerative colitis: a large single-centre study. Gut 2019;68:414-422.
2. Mahmoud R, Shah SC, Ten Hove JR, et al. No Association Between Pseudopolyps and Colorectal Neoplasia in Patients With Inflammatory Bowel Diseases. Gastroenterology 2019;156:1333-1344 e3.
3. Ten Hove JR, Shah SC, Shaffer SR, et al. Consecutive negative findings on colonoscopy during surveillance predict a low risk of advanced neoplasia in patients with inflammatory bowel disease with long-standing colitis: results of a 15-year multicentre, multinational cohort study. Gut 2019;68:615-622.
4. Wijnands AM, de Jong ME, Lutgens M, et al. Prognostic Factors for Advanced Colorectal Neoplasia in Inflammatory Bowel Disease: Systematic Review and Meta-analysis. Gastroenterology 2021;160:1584-1598.
5. van Houwelingen, H., & Putter, H. (2012). Dynamic Prediction in Clinical Survival Analysis (1st ed.). CRC Press. https://doi.org/10.1201/b11311:

Disclosure

Laurent Beaugerie: Research grants: Biogen, Celltrion, Takeda, Ferring, Celltrion, Viatris; Personal fees: Takeda, Nordic Pharma, Janssen, BMS, and Viatris.
Charles N. Bernstein: Dr Bernstein is supported by the Bingham Chair in Gastroenterology. Dr. Bernstein has served on advisory Boards for AbbVie Canada, Amgen Canada, Bristol Myers Squibb Canada, Eli Lilly Canada, Roche Canada, Janssen Canada, Sandoz Canada, Takeda Canada, and Pfizer Canada; Educational grants from Abbvie Canada, Amgen Canada, Bristol Myers Squibb Canada, Ferring Canada, Organon Canada, Pfizer Canada, Takeda Canada, and Janssen Canada. Speaker’s panel for Abbvie Canada, Janssen Canada, Pfizer Canada, and Takeda Canada. Received research funding from Abbvie Canada, Amgen Canada, Sandoz Canada, Takeda Canada, and Pfizer Canada.
Steven H. Itzkowitz: Dr Itzkowitz has served on Advisory Board for Exact Sciences Corporation and a consultant for Geneoscopy. He has received research support from Exact Sciences Corporation and Freenome.
Julien Kirchgesner: Lecture fees from Janssen and Lilly, and consulting fees from Roche, Gilead, Pfizer, Janssen, Celltrion, Lilly, Takeda, and Abbvie.
Cyriel Y. Ponsioen: Received grant support from Gilead and Perspectum and speaker’s fees from Tillotts

Conference

UEG Week Copenhagen 2023

Topics

IBD

Submission format

Abstract

Session

Endoscopy and beyond (Posters)

Citation

United European Gastroenterology Journal 2023; 11 (Supplement 8)

Published

2023

More Like This:

PERSISTENT TRANSMURAL DISEASE ACTIVITY ASSOCIATES WITH ABDOMINAL SURGERY AND DISEASE PROGRESSION EVEN IN ENDOSCOPICALLY-HEALED CROHN’S DISEASE

Nurulamin Noor

Introduction: Dilemmas in ulcerative colitis: Back to the basics or fast forward

Peter Bossuyt

FIRST MULTIDEVICE ARTIFICIAL INTELLIGENCE MODEL FOR INTRAOPERATIVE PREDICTION OF COLORECTAL ANASTOMOTIC LEAKAGE

Guilherme Macedo

INTESTINAL ULTRASOUND AND SHEAR-WAVE ELASTOGRAPHY: KINETICS IN ULCERATIVE COLITIS PATIENTS TREATED WITH FILGOTINIB

Krisztina Barbara Gecse

Education of endoscopy nurses in Europe – 30 years and still not standardised

Ulrike Beilenhoff

EVALUATION OF RECTAL WALL THICKNESS MEASUREMENT SENSITIVITY VIA TRANSABDOMINAL ULTRASOUND TO ASSESS MUCOSAL ACTIVITY IN ULCERATIVE COLITIS: COULD THIS BE A USEFUL NON-INVASIVE TOOL?

Elisa Marabotto

UEG Presentation

New

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

PERSISTENT TRANSMURAL DISEASE ACTIVITY ASSOCIATES WITH ABDOMINAL SURGERY AND DISEASE PROGRESSION EVEN IN ENDOSCOPICALLY-HEALED CROHN’S DISEASE

This content is part of Gutflix. Log in with your myUEG account, or create one free, to watch it.

Not sure what you can access? Learn more about account types.

PERSISTENT TRANSMURAL DISEASE ACTIVITY ASSOCIATES WITH ABDOMINAL SURGERY AND DISEASE PROGRESSION EVEN IN ENDOSCOPICALLY-HEALED CROHN’S DISEASE

Mohmmed Tauseef Sharip 1, Nick Kennedy 2, Sara Upponi 3, James C. Lee 4, Miles Parkes 1, Nurulamin Noor 1

Affiliations

1 Cambridge University Hospitals, Cambridge, United Kingdom

2 Royal Devon & Exeter Hospital, Exeter, United Kingdom

3 Barts Health NHS Trust, London, United Kingdom

4 The Francis Crick Institute, London, United Kingdom

Summary

AI Generated

Summary is not available for this content yet.

Download PDF

Was this helpful?

Thanks for your feedback.

Abstract

Introduction

Endoscopic healing is an important objective outcome measure in Crohn’s disease trials and a key treatment target in clinical practice. Despite growing focus on transmural healing, this is not currently recommended as a formal treatment target. Several, mostly retrospective, studies have reported that patients achieving transmural healing on cross-sectional imaging may have better outcomes than those with endoscopic healing alone.

Aims & Methods

In a prospective randomised controlled trial (PROFILE), we aimed to evaluate discordance between endoscopic remission and transmural activity, and whether residual transmural activity correlated with adverse outcomes over subsequent five year follow-up.
PROFILE was a multicentre, open-label, randomized controlled trial (RCT) comparing top-down (early combined immunosuppression) with accelerated step-up therapy in adults with newly-diagnosed active Crohn’s disease. Participants underwent ileocolonoscopy and MRE at baseline and at the end of the trial (week 48), and were subsequently followed up for up to five years. Endoscopic activity was quantified using the Simple Endoscopic Score for Crohn’s Disease (SES-CD). Adjudicated central SES-CD at week 48 was used, with endoscopic remission defined as a total SES-CD of ≤4. Week-48 MRE studies were scored centrally on a per-segment basis.
Transmural assessment was restricted to endoscopically accessible sites to allow like-for-like comparison. Patients were cross-classified (2×2) by week-48 endoscopic status (remission vs active) and end-of-trial MRE transmural status (remission vs active), yielding four groups. The primary outcome measure for this study was a composite of abdominal surgery and development of stricturing (B2) or penetrating (B3) disease. Time-to-event outcomes were summarised with Kaplan–Meier estimates and compared across the four groups using a log-rank test.

Results

250 patients had paired week-48 ileo-colonoscopy and end-of-trial MRE (both centrally scored) and were event-free at this timepoint. After exclusion of 43 patients who did not meet the criteria for inclusion (e.g. prior abdominal surgery or established B2 or B3 disease), 207 patients formed the primary analysis cohort, among whom 16 needed abdominal surgery or had progression to B2/B3 disease over five years. Within the endoscopically accessible sites, 137 patients were in endoscopic remission (SES-CD ≤4) at week 48. Of these, 46/137 (34%) had persistent transmural activity on MRE within the endoscopically-healed sites, indicating discordance between endoscopic and transmural healing. Residual disease activity was almost exclusively within the terminal ileum (43/46, 93%) and was dominated by mural oedema (61%) and grade 3 wall thickening (54%). Conversely, deep ulceration was uncommon (9%). Notably, 21 of the 46 discordant patients (46%) had a week-48 SES-CD of 0, despite evidence of ongoing transmural disease activity. Need for abdominal surgery or progression to B2/B3 disease occurred in 5/46 (11%) patients with residual transmural activity despite endoscopic healing compared with 1/91 (1%) patients who were in both endoscopic and transmural remission (odds ratio 10.8, 95% CI 1.2–525; Fisher’s exact p=0.016).

Conclusion

Our study demonstrates that transmural activity can persist in approximately one in three patients with Crohn’s disease who have achieved endoscopic remission. Ongoing transmural activity was associated with higher risk of disease progression and abdominal surgery over a five-year period.

References

Noor NM, Lee JC, Bond S, et al. A biomarker-stratified comparison of top-down versus accelerated step-up treatment strategies for patients with newly-diagnosed Crohn’s disease (PROFILE): a multicentre, open-label randomised controlled trial. Lancet Gastroenterology & Hepatology. 2024. PMID 38402895.
Noor NM, Davies N, Tahir W, et al. Anti-TNF treatment from diagnosis is more effective and less costly than conventional ‘step-up’ care for patients with active Crohn’s disease: a cost-effectiveness analysis from the PROFILE trial. Journal of Crohns and Colitis. 2025. PMID 41125135.

Disclosure

MTS reports grant from AstraZeneca; support for attending meetings and travel from Takeda; and fees from advisory board membership of Dr Falk, Teva. NAK reports grants from Pharmacosmos; consulting fees from AbbVie, BMS, Falk, J&J, Pharmacosmos, Sandoz; honoraria from AbbVie, J&J, Pharmacosmos, Tillotts; support for attending meetings from AbbVie, J&J, Lilly. JCL reports grants or contracts from Kenneth Rainin Foundation Research Grant, MRC/MSD Alliance Research Grant, Lifearc ‘idea-to-innovation’ research grant; has received consulting fees from J&J, Enhanced Genomics, Pasithea, Isomorphic Labs; honoraria from J&J, AbbVie; patents planned, issued or pending GB2304900.0: “Methods of Treatment of Inflammatory Diseases”. MP reports travel grants from J&J, Takeda; and grants from AstraZeneca, Celltrion, Galapagos, Gilead, Lilly, Pfizer. NMN reports educational/travel grants and/or speaker fees from AbbVie, BMS, Celltrion, Falk, Ferring, J&J, Lilly, Medfyle, Pfizer, Pharmacosmos, Takeda and Tillotts; all outside the submitted work. All other authors report no conflicts of interest.

Event

UEG Week Barcelona 2026

Topics

Histopathology Surgery Radiology & Imaging Mechanisms & Personalised Medicine Immunology IBD

Submission format

Late-Breaking Abstract

Session

Late-Breaking in IBD: From translational discoveries to treatment

Citation

2026: United European Gastroenterology Journal 2026; 14 (S1)

Published

2026

More Like This:

PERSISTENT TRANSMURAL DISEASE ACTIVITY ASSOCIATES WITH ABDOMINAL SURGERY AND DISEASE PROGRESSION EVEN IN ENDOSCOPICALLY-HEALED CROHN’S DISEASE

Nurulamin Noor

Introduction: Dilemmas in ulcerative colitis: Back to the basics or fast forward

Peter Bossuyt

FIRST MULTIDEVICE ARTIFICIAL INTELLIGENCE MODEL FOR INTRAOPERATIVE PREDICTION OF COLORECTAL ANASTOMOTIC LEAKAGE

Guilherme Macedo

INTESTINAL ULTRASOUND AND SHEAR-WAVE ELASTOGRAPHY: KINETICS IN ULCERATIVE COLITIS PATIENTS TREATED WITH FILGOTINIB

Krisztina Barbara Gecse

Education of endoscopy nurses in Europe – 30 years and still not standardised

Ulrike Beilenhoff

EVALUATION OF RECTAL WALL THICKNESS MEASUREMENT SENSITIVITY VIA TRANSABDOMINAL ULTRASOUND TO ASSESS MUCOSAL ACTIVITY IN ULCERATIVE COLITIS: COULD THIS BE A USEFUL NON-INVASIVE TOOL?

Elisa Marabotto

UEG Presentation

New

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

Introduction: Dilemmas in ulcerative colitis: Back to the basics or fast forward

This content is part of Gutflix. Log in with your myUEG account, or create one free, to watch it.

Not sure what you can access? Learn more about account types.

Introduction: Dilemmas in ulcerative colitis: Back to the basics or fast forward

Peter Bossuyt 1

Affiliations

1 Imelda General Hospital, Bonheiden, Belgium

Summary

AI Generated

Summary is not available for this content yet.

Download PDF

Was this helpful?

Thanks for your feedback.

Abstract

Event

Postgraduate Teaching Programme Barcelona 2026

Topics

Mechanisms & Personalised Medicine IBD

Session

Dilemmas in ulcerative colitis: Back to the basics or fast forward

Citation

2026: United European Gastroenterology Journal 2026; 14 (S1)

Published

2026

More Like This:

PERSISTENT TRANSMURAL DISEASE ACTIVITY ASSOCIATES WITH ABDOMINAL SURGERY AND DISEASE PROGRESSION EVEN IN ENDOSCOPICALLY-HEALED CROHN’S DISEASE

Nurulamin Noor

Introduction: Dilemmas in ulcerative colitis: Back to the basics or fast forward

Peter Bossuyt

FIRST MULTIDEVICE ARTIFICIAL INTELLIGENCE MODEL FOR INTRAOPERATIVE PREDICTION OF COLORECTAL ANASTOMOTIC LEAKAGE

Guilherme Macedo

INTESTINAL ULTRASOUND AND SHEAR-WAVE ELASTOGRAPHY: KINETICS IN ULCERATIVE COLITIS PATIENTS TREATED WITH FILGOTINIB

Krisztina Barbara Gecse

Education of endoscopy nurses in Europe – 30 years and still not standardised

Ulrike Beilenhoff

EVALUATION OF RECTAL WALL THICKNESS MEASUREMENT SENSITIVITY VIA TRANSABDOMINAL ULTRASOUND TO ASSESS MUCOSAL ACTIVITY IN ULCERATIVE COLITIS: COULD THIS BE A USEFUL NON-INVASIVE TOOL?

Elisa Marabotto

UEG Presentation

New

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

FIRST MULTIDEVICE ARTIFICIAL INTELLIGENCE MODEL FOR INTRAOPERATIVE PREDICTION OF COLORECTAL ANASTOMOTIC LEAKAGE

This content is part of Gutflix. Log in with your myUEG account, or create one free, to watch it.

Not sure what you can access? Learn more about account types.

FIRST MULTIDEVICE ARTIFICIAL INTELLIGENCE MODEL FOR INTRAOPERATIVE PREDICTION OF COLORECTAL ANASTOMOTIC LEAKAGE

Miguel Mascarenhas-Saraiva 1, Francisco Mendes 1, André Martins dos Santos 1, Filipa Fonseca 1, Larissa Mercadante de Assis 2, Felix Bonilla 2, Gonçalo Soares 3, Daniela Cavadas 4, Miguel Martins 1, Maria Almeida 1, Joana Mota 1, Ali Murtada 5, Marley Feitosa 6, Matheus Carvalho 2, Filipe Marques 7, João Ferreira 7, Ahsan Javed 5, Shakil Ahmed 5, Maisa Vasconcellos 6, Carlos Frederico Sparapan Marques 2, Manuel Limbert 4, Guilherme Macedo 1

Affiliations

1 Centro Hospitalar São João, Porto, Portugal

2 Hospital das Clínicas da Faculdade de Medicina da USP, São Paulo, Brazil

3 Centro Hospitalar Universitário de Santo António, Porto, Portugal

4 Instituto Português de Oncologia de Lisboa, Lisboa, Portugal

5 Royal Liverpool University Hospital NHS Foundation Trust, Liverpool, United Kingdom

6 Hospital das Clínicas da Faculdade de Medicina da USP de Ribeirão Preto, Ribeirão Preto, Brazil

7 Faculdade de Engenharia da Universidade do Porto, Porto, Portugal

Summary

AI Generated

Summary is not available for this content yet.

Download PDF

Was this helpful?

Thanks for your feedback.

Abstract

Introduction

Anastomotic leakage (AL) is the postoperative complication with the greatest impact on both patients and healthcare systems after colorectal surgery, carrying substantial morbidity and mortality. Current assessment remains suboptimal and relies exclusively on clinical, radiological, and analytical parameters, often leading to delayed diagnosis and worse outcomes. Artificial intelligence (AI) in digestive healthcare has enabled the development of several detection (CADe) and characterization (CADx) models, although CADex models remain scarce. This study aimed to develop a multidevice deep learning model for intraoperative risk stratification, enabling real time prediction of AL.

Aims & Methods

An AI model based on a convolutional neural network was developed to predict AL using laparoscopic video frames, with a bounding box highlighting the region contributing most to leak prediction. AL diagnosis was established according to clinical, analytical, and imaging criteria. Normal anastomosis required stoma closure and at least 6 months of clinical follow up. To prevent overfitting and selection bias, dataset splitting was performed at the case level, ensuring that frames from the same surgery were not represented in both training and testing sets. The dataset was divided into training and testing sets using an 80 percent and 20 percent split. The detection component (CADe) was evaluated with mAP50, whereas the characterization component (CADx) was evaluated with accuracy and AUC ROC.

Results

This multicenter study included 71 colorectal surgeries and 17,801 frames collected across 4 reference centers in Europe and the Americas, using seven different laparoscopy systems, including conventional and robotic platforms. Thirteen patients developed CAL during postoperative follow up, whereas the remaining patients had an uneventful recovery without CAL. The AI model detected anastomoses with an mAP50 of 0.973. Frames associated with AL were identified with 99.0 percent accuracy and an AUC ROC of 0.997.

Conclusion

This is the first multidevice, interoperable AI model for intraoperative risk stratification enabling automatic prediction of AL development through a CADe mechanism for anastomosis identification, in a cohort including patients from both Europe and the Americas. The development of prognosis driven AI models is essential to support intraoperative decision making, allowing risk reduction through targeted preventive or active measures. Ultimately, these tools may reduce the incidence of this postoperative complication, with meaningful benefit for both patients and healthcare systems.

Event

UEG Week Barcelona 2026

Topics

Endoscopy Surgery Mechanisms & Personalised Medicine Colorectal

Submission format

Abstract

Session

Lower GI disorders: A multidisciplinary approach

Citation

2026: United European Gastroenterology Journal 2026; 14 (S1)

Published

2026

More Like This:

PERSISTENT TRANSMURAL DISEASE ACTIVITY ASSOCIATES WITH ABDOMINAL SURGERY AND DISEASE PROGRESSION EVEN IN ENDOSCOPICALLY-HEALED CROHN’S DISEASE

Nurulamin Noor

Introduction: Dilemmas in ulcerative colitis: Back to the basics or fast forward

Peter Bossuyt

FIRST MULTIDEVICE ARTIFICIAL INTELLIGENCE MODEL FOR INTRAOPERATIVE PREDICTION OF COLORECTAL ANASTOMOTIC LEAKAGE

Guilherme Macedo

INTESTINAL ULTRASOUND AND SHEAR-WAVE ELASTOGRAPHY: KINETICS IN ULCERATIVE COLITIS PATIENTS TREATED WITH FILGOTINIB

Krisztina Barbara Gecse

Education of endoscopy nurses in Europe – 30 years and still not standardised

Ulrike Beilenhoff

EVALUATION OF RECTAL WALL THICKNESS MEASUREMENT SENSITIVITY VIA TRANSABDOMINAL ULTRASOUND TO ASSESS MUCOSAL ACTIVITY IN ULCERATIVE COLITIS: COULD THIS BE A USEFUL NON-INVASIVE TOOL?

Elisa Marabotto

UEG Presentation

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

INTESTINAL ULTRASOUND AND SHEAR-WAVE ELASTOGRAPHY: KINETICS IN ULCERATIVE COLITIS PATIENTS TREATED WITH FILGOTINIB

This content is part of Gutflix. Log in with your myUEG account, or create one free, to watch it.

Not sure what you can access? Learn more about account types.

INTESTINAL ULTRASOUND AND SHEAR-WAVE ELASTOGRAPHY: KINETICS IN ULCERATIVE COLITIS PATIENTS TREATED WITH FILGOTINIB

Maarten Pruijt 1, Christoph Teichert 1, Floris de Voogd 1, Reimer Janssen 1, Mark Lowenberg 1, Rogier Goetgebuer 1, Geert R. D'Haens 1, Krisztina Barbara Gecse 1

Affiliations

1 Amsterdam UMC, Amsterdam, Netherlands

Summary

AI Generated

Summary is not available for this content yet.

Download PDF

Was this helpful?

Thanks for your feedback.

Abstract

Introduction

A reduction of bowel wall thickness (BWT) on intestinal ultrasound (IUS) is an early predictor of endoscopic response in ulcerative colitis (UC).¹ Shear-wave elastography (SWE) may offer additional prognostic information.

Aims & Methods

We aimed to identify early IUS parameters predicting treatment response to filgotinib, a selective JAK-1 inhibitor. This prospective observational study included UC patients with endoscopically active disease starting on filgotinib 200 mg daily. IUS parameters, including SWE, were assessed at baseline (T0), week 4 (T1) and at second endoscopy (T2). The endoscopic Mayo score (EMS) was assessed at T0 and at T2 in both the most-affected segment and the sigmoid. Endoscopic response was defined as a decrease in EMS of at least 1 point.

Results

A total of 23 patients were included. 10 patients were female (10/23, 44%), the mean disease duration was 14.4 years (SD±10.4), and 14 patients (14/23, 61%) had failed 2 or more biologics. Endoscopic response was observed in 6 of the 21 patients (29%) who underwent a second endoscopy after a median of 14 weeks (IQR 12-17). The submucosal thickness in the sigmoid decreased significantly more between T0 and T1 in endoscopic responders compared to non-responders (-40±23.9% vs -11.5±24.5%, p=0.022). At T1, a BWT decrease of ≥1.33 mm or ≥24.7% (OR 32.5 [2.38-443.15], p=0.009 and OR: 13.8 [1.21-156.6], p=0.035) and a submucosal thickness decrease of ≥20.8% (OR: 13.8 [1.21-156.6], p=0.035) predicted endoscopic response at T2. Additionally, improvement in Colour Doppler Signal (CDS) at T1 predicted endoscopic response (OR: 20.0 [1.66-241.7], p=0.018). SWE values in the sigmoid at T0, T1 and T2 were not significantly different between endoscopic responders and non-responders in the sigmoid. While the change in SWE values at T2 compared to baseline was significantly different between the two groups (9.9± 11.5 kPa vs. -8.1 ± 11.4 kPa, p=0.002), the difference at T1 was not significant and did not predict endoscopic response in the sigmoid (OR: 1.07 [0.99-1.16], p=0.088).

Logistic regression for endoscopic response (EMS ≥1 decrease) based on IUS B-mode parameters at week 4 (T1) in the sigmoid	Univariable Odds ratio (95% CI)	P-value
BWT (per mm increase)	0.34 [0.99-1.17]	0.087
BWT (per % increase)	0.97 [0.92-1.01]	0.118
BWT decrease ≥1.33 mm	32.5 [2.38-443.15]	0.009
BWT decrease ≥ 24.72%	13.8 [1.21-156.6]	0.035
Submucosa (per % increase)	0.95 [0.89-1.00]	0.049
Submucosa decrease ≥20.76%	13.8 [1.21-156.6]	0.035
CDS (per one category increase)	0.26 [0.09-0.77]	0.015
CDS (≥1 decrease in mLimberg classification)	20.0 [1.66-241.7]	0.018
CDS (mLimberg of 0)	28.0 [1.99-394.4]	0.014

Conclusion

On IUS, BWT, submucosal thickness and CDS predict endoscopic response after four weeks of filgotinib treatment. SWE values in the sigmoid changed differently among endoscopic responders and non-responders, but did not predict treatment response.

References

1. de Voogd FA et al. Early Intestinal Ultrasound Predicts Clinical and Endoscopic Treatment Response and Demonstrates Drug-Specific Kinetics in Moderate-to-Severe Ulcerative Colitis. Inflamm Bowel Dis 2023. DOI: 10.1093/ibd/izad274.

Disclosure

MP: none;
CT: none;
FV: received speaker’s fees from AbbVie, Pfizer and Takeda;
RJ: none,
ML: Consultancy/lecture fees from Abbvie, Bristol Myers Squibb, Eli Lilly, Galapagos, Janssen-Cilag, Johnson & Johnson, Medtronic, Pfizer, Takeda, Tillotts. Grants received from Alfasigma, NFU transformation deal, ZonMW and TKI.
RG: received speakers free from Janssen-Cilag BV and Pfizer Inc.
GD: has served as advisor for AbbVie, Abivax, Alimentiv,, Astrazeneca, Bristol Meiers Squibb, Boehringer Ingelheim, Celltrion, Eli Lilly, , GlaxoSmithKline, Johnson and Johnson, Merck Sharp Dome, Pfizer, Polpharm, Sanofi, Takeda, Tillotts, and received speaker fees from AbbVie, Eli Lilly, Johnson and Johnson, Merck Sharp Dome, , Pfizer, Takeda, Tillotts,.
KG: KB Gecse has received grants from AbbVie, Pfizer Inc, Celltrion and Galapagos; consultancy fees from AbbVie, Arena Pharmaceuticals, ImmunicTherapeutics, Janssen Pharmaceuticals, Novartis, Pfizer Inc., and Takeda and speaker’s honoraria from Abbvie, Celltrion, Eli Lilly, Ferring, Janssen Pharmaceuticals, Pfizer Inc, Takeda and Tillotts.

Event

UEG Week Berlin 2025

Topics

IBD Radiology & Imaging

Submission format

Abstract

Session

Monitoring IBD by imaging

Citation

United European Gastroenterology Journal 2025; 13 (Supplement 8)

Published

2025

More Like This:

PERSISTENT TRANSMURAL DISEASE ACTIVITY ASSOCIATES WITH ABDOMINAL SURGERY AND DISEASE PROGRESSION EVEN IN ENDOSCOPICALLY-HEALED CROHN’S DISEASE

Nurulamin Noor

Introduction: Dilemmas in ulcerative colitis: Back to the basics or fast forward

Peter Bossuyt

FIRST MULTIDEVICE ARTIFICIAL INTELLIGENCE MODEL FOR INTRAOPERATIVE PREDICTION OF COLORECTAL ANASTOMOTIC LEAKAGE

Guilherme Macedo

INTESTINAL ULTRASOUND AND SHEAR-WAVE ELASTOGRAPHY: KINETICS IN ULCERATIVE COLITIS PATIENTS TREATED WITH FILGOTINIB

Krisztina Barbara Gecse

Education of endoscopy nurses in Europe – 30 years and still not standardised

Ulrike Beilenhoff

EVALUATION OF RECTAL WALL THICKNESS MEASUREMENT SENSITIVITY VIA TRANSABDOMINAL ULTRASOUND TO ASSESS MUCOSAL ACTIVITY IN ULCERATIVE COLITIS: COULD THIS BE A USEFUL NON-INVASIVE TOOL?

Elisa Marabotto

UEG Presentation

New

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

Education of endoscopy nurses in Europe – 30 years and still not standardised

This content is part of Gutflix. Log in with your myUEG account, or create one free, to watch it.

Not sure what you can access? Learn more about account types.

Education of endoscopy nurses in Europe – 30 years and still not standardised

Ulrike Beilenhoff 1

Affiliations

1 ESGENA Scientific Secretary, Ulm, Germany

Summary

AI Generated

Summary is not available for this content yet.

Download PDF

Was this helpful?

Thanks for your feedback.

Abstract

Event

UEG Week Barcelona 2026

Topics

Nurses

Session

Education of patients and staff

Citation

2026: United European Gastroenterology Journal 2026; 14 (S1)

Published

2026

More Like This:

PERSISTENT TRANSMURAL DISEASE ACTIVITY ASSOCIATES WITH ABDOMINAL SURGERY AND DISEASE PROGRESSION EVEN IN ENDOSCOPICALLY-HEALED CROHN’S DISEASE

Nurulamin Noor

Introduction: Dilemmas in ulcerative colitis: Back to the basics or fast forward

Peter Bossuyt

FIRST MULTIDEVICE ARTIFICIAL INTELLIGENCE MODEL FOR INTRAOPERATIVE PREDICTION OF COLORECTAL ANASTOMOTIC LEAKAGE

Guilherme Macedo

INTESTINAL ULTRASOUND AND SHEAR-WAVE ELASTOGRAPHY: KINETICS IN ULCERATIVE COLITIS PATIENTS TREATED WITH FILGOTINIB

Krisztina Barbara Gecse

Education of endoscopy nurses in Europe – 30 years and still not standardised

Ulrike Beilenhoff

EVALUATION OF RECTAL WALL THICKNESS MEASUREMENT SENSITIVITY VIA TRANSABDOMINAL ULTRASOUND TO ASSESS MUCOSAL ACTIVITY IN ULCERATIVE COLITIS: COULD THIS BE A USEFUL NON-INVASIVE TOOL?

Elisa Marabotto

UEG Presentation

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

EVALUATION OF RECTAL WALL THICKNESS MEASUREMENT SENSITIVITY VIA TRANSABDOMINAL ULTRASOUND TO ASSESS MUCOSAL ACTIVITY IN ULCERATIVE COLITIS: COULD THIS BE A USEFUL NON-INVASIVE TOOL?

This content is part of Gutflix. Log in with your myUEG account, or create one free, to watch it.

Not sure what you can access? Learn more about account types.

EVALUATION OF RECTAL WALL THICKNESS MEASUREMENT SENSITIVITY VIA TRANSABDOMINAL ULTRASOUND TO ASSESS MUCOSAL ACTIVITY IN ULCERATIVE COLITIS: COULD THIS BE A USEFUL NON-INVASIVE TOOL?

Francesco Calabrese 1, ANDREA PASTA 1, Graziana Santamorena 1, Elisa Marabotto 1, Manuele Furnari 1, Edoardo G. Giannini 1, Giorgia Bodini 1

Affiliations

1 Università Degli Studi Di Genova/irccs Policlinico San Martino, Genova, Italy

Summary

AI Generated

Summary is not available for this content yet.

Download PDF

Was this helpful?

Thanks for your feedback.

Abstract

Introduction

Ulcerative colitis (UC) is a chronic inflammatory bowel disease where accurate assessment of disease activity is essential for effective patient management. Endoscopy is the gold standard for evaluating mucosal inflammation, but non-invasive alternatives, such as transabdominal ultrasound (TAUS), are garnering interest.

Aims & Methods

This study aimed to evaluate the sensitivity of rectal wall thickness measurements using TAUS in assessing disease activity in ulcerative colitis, and to identify an ideal cut-off threshold by comparing results with endoscopic and histological findings.
In a prospective, observational study, we enrolled consecutive patients with UC who were referred to our tertiary center from May 2024 to April 2025. Eligibility required a scheduled TAUS within 30 days of a staging colonoscopy, without intervening therapy adjustments. Exclusion criteria included technically inadequate TAUS readings (e.g., due to anatomical variations or excessive gas interposition) and minor-aged patients. Each patient underwent TAUS performed by a single, experienced operator. A second experienced operator blind assessments were recorded to evaluate inter-observer variability. Rectal wall thickness was measured and compared with clinical, laboratory (CRP, fecal calprotectin), and endoscopic (MAYO score) indicators of disease activity. Histological status (active/inactive) and additional clinical data, including disease duration, extent, and therapy, were recorded.

Results

A total of 74 patients were analyzed, of whom 51.4% were female, with a median age of 55 years (IQR: 42–68). The majority presented with left-sided colitis (Montreal E2, 62.2%), followed by extensive colitis (E3, 25.7%) and proctitis (E1, 12.1%). The median disease duration was 12 years (IQR 5–23). Patients with an endoscopic MAYO score of 0 or 1 had a median rectal wall thickness of 4.95 mm (IQR 4.5–5.88), while those with a score of 2 or 3 had a median of 6.15 mm (IQR 5.3–6.7; p=0.003). Receiver Operating Characteristic (ROC) analysis for TAUS rectal wall thickness showed good discriminatory ability for endoscopic activity (cut-off: 5.55 mm, AUC 0.725, 95CI 0.597-0.853, Se 0.80, Sp 0.71), comparable to fecal calprotectin (cut-off: 277 µg/g, AUC 0.717, 95CI 0.589-0.845, Se 0.70, Sp 0.66; p=0.256) and superior to CRP (cut-off: 1 mg/L, AUC 0.511, 95CI 0.357-0.664, Se 0.50, Sp 0.50; p<0.0001). For histological activity, TAUS again performed well (cut-off: 5.55 mm, AUC 0.794, 95CI 0.682-0.906, Se 0.76, Sp 0.80), closely matching fecal calprotectin (cut-off 302, AUC 0.744, 95CI 0.627-0.861, Se 0.69, Sp 0.73; p=0.425) and outperforming CRP (cut-off 1.5, AUC: 0.631, 95CI 0.497-0.765, Se 0.62, Sp 0.63; p<0.0001). Inter-observer agreement for TAUS rectal wall thickness was excellent, with an intraclass correlation coefficient of 0.923 (p<0.0001), underscoring the reproducibility of this non-invasive method.

Conclusion

TAUS measurement of rectal wall thickness demonstrates promising sensitivity and reproducibility in detecting mucosal and histological activity in UC, aligning closely with established markers such as fecal calprotectin. A cut-off value of 5.55 mm showed good discriminatory performance in this cohort, suggesting potential utility as a non-invasive marker of disease activity..Given its non-invasive nature, low cost, and strong inter-observer agreement, TAUS is a valuable adjunct for routine UC monitoring. Combined with clinical and biochemical data, it may reduce the need for repeated endoscopy, minimizing patient burden, healthcare costs, and environmental impact.

Event

UEG Week Berlin 2025

Topics

IBD Radiology & Imaging

Submission format

Abstract

Session

Monitoring IBD by imaging

Citation

United European Gastroenterology Journal 2025; 13 (Supplement 8)

Published

2025

More Like This:

PERSISTENT TRANSMURAL DISEASE ACTIVITY ASSOCIATES WITH ABDOMINAL SURGERY AND DISEASE PROGRESSION EVEN IN ENDOSCOPICALLY-HEALED CROHN’S DISEASE

Nurulamin Noor

Introduction: Dilemmas in ulcerative colitis: Back to the basics or fast forward

Peter Bossuyt

FIRST MULTIDEVICE ARTIFICIAL INTELLIGENCE MODEL FOR INTRAOPERATIVE PREDICTION OF COLORECTAL ANASTOMOTIC LEAKAGE

Guilherme Macedo

INTESTINAL ULTRASOUND AND SHEAR-WAVE ELASTOGRAPHY: KINETICS IN ULCERATIVE COLITIS PATIENTS TREATED WITH FILGOTINIB

Krisztina Barbara Gecse

Education of endoscopy nurses in Europe – 30 years and still not standardised

Ulrike Beilenhoff

EVALUATION OF RECTAL WALL THICKNESS MEASUREMENT SENSITIVITY VIA TRANSABDOMINAL ULTRASOUND TO ASSESS MUCOSAL ACTIVITY IN ULCERATIVE COLITIS: COULD THIS BE A USEFUL NON-INVASIVE TOOL?

Elisa Marabotto