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.

CLINICAL VALIDATION OF POLYDEEP A COMPUTER-AIDED DIAGNOSIS SYSTEM (CADE/X)

Pedro Davila-Piñón 1,2, Cristina Regueiro 1,2, Astrid Irene Díez Martín 1,2, Jorge Hernández-Camoiras 1,2, Rubén Domínguez-Carbajales 3, Jesús Herrero 3, Manuel Puga 3, Laura Rivas 3, Eloy Sánchez 3, Sara Zarraquiños Martínez 3, Noel Pin Vieito 3, Pablo Vega 3, Santiago Soto 3, David Remedios-Espino 3, Alba Nogueira-Rodríguez 4,5, Alejandro González-García 4,5, Florentino Fdez-Riverola 4,5, Hugo López-Fernández 4,5, Daniel Glez-Peña 4,5, Miguel Reboiro-Jato 4,5, Joaquín Cubiella 2,3

Affiliations

1 Galicia Sur Health Research Institute (IIS Galicia Sur), Vigo, Spain

2 Research Group in Gastrointestinal Oncology - Ourense, Ourense, Spain

3 Ourense University Hospital Complex, Ourense, Spain

4 CINBIO, Universidade de Vigo, Ourense, Spain

5 SING Research Group., Ourense, Spain

Summary

AI Generated

The Polydeep Advance 1 prospective trial compared the AI system Polydeep with experienced endoscopists for colorectal polyp detection and characterisation in 205 colonoscopy patients.

Polydeep demonstrated 91.5% sensitivity for polyp detection compared to 94.1% for endoscopists, with similar positive predictive values (89.7% vs 89.4%) but low specificity for both (14.3% vs 9.5%).
For adenoma detection, Polydeep achieved 92.6% sensitivity versus 94.4% for endoscopists, with comparable positive predictive values around 61% and negative predictive values of 51.4% versus 45.8%.
Both Polydeep and endoscopists achieved 98.6% sensitivity for advanced lesion detection.
For neoplasia characterisation, Polydeep showed 82.6% sensitivity and 31.1% specificity, while the endoscopists' NICE II classification achieved 76.5% sensitivity and 61.8% specificity.
The study enrolled patients with positive fecal immunochemical tests or undergoing surveillance after prior adenomas, and the authors state results will be confirmed in ongoing multicentre trials.

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

Polydeep is a CADe/x system that has demostrated a high diagnostic yield for in vitro detection and characterisation of colorectal polyps. We have designed a diagnostic accuracy study, Polydeep Advance 1 (NCT05514301) in order to confirm these preliminar results.

Aims & Methods

Polydeep Advance 1 is a unicentric prospective diagnostic test trial. Endoscopists performed the colonoscopy blinded to Polydeep. A second observer alerted the endoscopists when Polydeep detected a lesion not identified by the endoscopists. The main objective of this study was to compare Polydeep sensitivity for colorectal polyps (adenoma and serrated lesion with confirmed histology) detection with experienced endoscopists. Additionally, we evaluated the diagnostic accuracy for detection of colorectal adenomas, advanced colorectal lesions (advanced adenoma or serrated lesion) and lesions ≤ 5mm. Finally, we evaluated the characterization capacity of Polydeep for neoplasia detection (adenoma, Sessile Serrated Lesion-SSL, and Traditional Serrated Adenoma-TSA) and compared it with the optical diagnosis of endoscopists using the NICE classification.

Results

We included 205 patients (mean age 63.0 ± 6.2 years, 51.1% male) in the trial, between February and April 2023. The indications for colonoscopy were positive fecal immunochemical test (124) and surveillance after colorectal adenomas (81). We excluded 8 patients due to incomplete colonoscopy. Endoscopists detected 384 lesions (39 not detected by Polydeep). In contrast, Polydeep detected 409 additional lesions, but the endoscopists only confirmed 26 of these lesions. In total, 410 lesions were detected and resected/biopsed. Out of these lesions, 22 were not retreived, 4 were colorectal adenocarcinoma and 384 were colorectal polyps (231 adenomas, 35 SSL, 3 TSA, 73 hyperplastic polyps) and 42 were normal mucosa. We show in the attached table the results of the diagnostic accuracy of endoscopists and Polydeep for histologically confirmed colorectal polyps. With regard to the optical diagnosis, Polydeep classified as neoplastic (>50% probability) 273 (73.6%) of the lesions detected and gave no classification in 24 (6.4%) of the lesions. The endoscopists classified 242 as NICE II and 142 as NICE I. In the lesions with histological diagnosis, Polydeep showed a 82.6% sensitivity and 31.1% specificity for neoplasia detection. On the other hand, the sensitivity and specificity of the NICE II classification were 76,5% and 61,8% respectively.

	Sensitivity (%)		Specificity (%)		Positive predictive value (%)		Negative predictive value (%)
	Polydeep	Endoscoopist	Polydeep	Endoscopist	Polydeep	Endoscopist	Polydeep	Endoscopist
Polyp	91.5%	94.1%	14.3%	9.5%	89.7%	89.4%	17.1%	16.7%
Adenoma	92.6%	94.4%	11.8%	7.2%	61.3%	60.6%	51.4%	45.8%
Advanced lesion	98.6%	98.6%
Polyp ≤ 5mm	89.5%	92.3%	14.6%	9.8%	86.4%	86.1%	18.8%	17.4%

Conclusion

We confirm the preliminar results of Polydeep in the clinical validation. These results will be confirmed in ongoing multicentric trials.

Conference

UEG Week Copenhagen 2023

Topics

Colorectal

Submission format

Abstract

Session

PP 06 Other lower GI disorders (Posters)

Citation

United European Gastroenterology Journal 2023; 11 (Supplement 8)

Published

2023

More Like This:

PREVENTION OF DELAYED POST-POLYPECTOMY HEMORRHAGE BY DOPPLER ENDOSCOPIC PROBE MONITORING AND FOCAL BLOOD FLOW ABLATION IN A RANDOMIZED CONTROLLED TRIAL

Jeffrey Gornbein

Exocrine pancreatic insufficiency: is there a pancreas-liver axis?

Ashkar Motaz

DETERMINANTS OF NEOPLASIA YIELD AT THE FIRST SURVEILLANCE COLONOSCOPY: THE IMPACT OF ENDOSCOPIST'S ADR

Carlo Senore

TEXTBOOK OUTCOMES OF EUS-GUIDED HEPATICOGASTROSTOMY ACCORDING TO INDICATION AND PRESENTING ANATOMY: A PROSPECTIVE COHORT STUDY

Paolo Giorgio Arcidiacono

Colorectal cancer metastasis: What did we learn from TransMet?

ARTIFICIAL INTELLIGENCE FOR PROCEDURAL GUIDANCE IN ERCP: A VIDEO DEMONSTRATION STUDY

Xiaofeng Zhang

UEG Presentation

New

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

PREVENTION OF DELAYED POST-POLYPECTOMY HEMORRHAGE BY DOPPLER ENDOSCOPIC PROBE MONITORING AND FOCAL BLOOD FLOW ABLATION IN A RANDOMIZED CONTROLLED TRIAL

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.

PREVENTION OF DELAYED POST-POLYPECTOMY HEMORRHAGE BY DOPPLER ENDOSCOPIC PROBE MONITORING AND FOCAL BLOOD FLOW ABLATION IN A RANDOMIZED CONTROLLED TRIAL

Dennis Jensen 1, Rome Jutabha 2, Gareth Dulai 3, Noam Jacob 4, Jeffrey Gornbein 5

Affiliations

1 University of California Los Angeles, Santa Monica, United States

2 Harbor-UCLA Medical Center, Torrance, United States

3 Kaiser Permanente, Downey, United States

4 West Los Angeles Veterans Administration Medical Center, Los Angeles, United States

5 University of California Los Angeles, Los Angeles, United States

Summary

AI Generated

Summary is not available for this content yet.

Download PDF

Was this helpful?

Thanks for your feedback.

Abstract

Introduction

Delayed post-polypectomy hemorrhage (DPPH) is caused by a submucosal artery in the ulcer base after polypectomy. Doppler endoscopic probe (DEP) is highly effective in detecting blood flow in medium or large sized arteries in PP induced ulcers (PPIUs). With focal treatment & arterial blood flow obliteration, DEP guided treatment has been reported to prevent severe DPPH throughout the colon (Ref). In contrast, mixed success rates with hemoclip (HC) closure are reported. The HC technique can achieve single-layer mucosal closure but buries submucosal arteries often without obliterating blood flow. It has been reported not to be cost effective, nor efficacious in the left colon or rectum. New treatments are needed to prevent DPPH - a common complication of polypectomy.

Aims & Methods

AIMS: To report final results of a randomized controlled trial (RCT) of blood flow monitoring & focal treatment to prevent DPPH.
METHODS: This was a multicenter RCT of high-risk outpatients having colonoscopies & prospective follow-up (FU). Polyps were removed by EMR &/or thermal coagulation. Patients were stratified by chronic anti-platelet or anti-thrombotic drug use & a PPIU of 10-40 mm; or for those without bleed drugs, PPIU’s were 15-40 mm. After polypectomy, randomization was to either standard management (e.g. following GI guidelines for bleed drugs) or DEP interrogation of PPIU’s & guided treatment with HCs in the PPIU until arterial blood flow was eradicated (1). Patients & their caregivers were blinded to treatments. Prospective FU was by a research coordinator at 7, 14, & 30 days to record complications (e.g. abdominal pain, vomiting, fever or dizziness); rectal bleeding & its severity (e.g. # bloody BM’s/day & # of days bleeding); & whether patients sought ER, clinic, or telemedicine care or hospitalization for bleeding. Major DPPH was diagnosed in patients with severe bleeding for 5 or more days &/or who needed hospitalization. Lesser bleeds were < 4 days, but without hospitalization. Demographic, laboratory, colonoscopic, & pathology results & 30 day outcomes were prospectively recorded on standard forms. Coded data were entered into HIPAA compliant computer files & managed with SAS. This is the final analysis by a biostatistician.

Results

For 180 high risk patients, 92 had standard treatment & 88 had DEP guided treatment. DEP monitoring & focal treatment was quick & easy. Treatment groups were well matched at baseline, including mean age of 68 years, 76% males; 55% on aspirin, 21 % on an anti-coagulant, & 9% on another anti-coagulant. Mean polyp size was 16 mm, most were sessile, & 57% were right colon. There were no drop-outs, severe adverse events, or deaths. However, DEP patients had significantly lower rates of DPPU bleeding – both major & total- as in Table 1.

Table 1. Delayed Post Polypectomy Induced Ulcer Bleeds by Treatment Group

DELAYED BLEEDS	Standard (N = 92)	Doppler (N = 88)	Risk Difference	SE	95% CI’s*	NNT**	****P-Value
Total Delayed PPIU Bleeds	11 (11.96 %)	2 (2.27 %)	9.68%	3.7%	2.36%;17.0%	10.3	0. 0127
Major Delayed Bleeds	***8 (8.7 %)	1 (1.14 %)	7.56 %	3.1%	1.39%;13.7%	13.2	0.0204
Lesser Delayed Bleeds	3 (3.26 %)	1 (1.14 %)	2.12%	2.2%	-2,13%;6.4%	47.1	0.5473

Legend: Treatment results shown as N’s & (%). * 95% CI’s is Confidence Intervals. **NNT is the number needed to treat. *** All were hospitalized except 1 patient with severe bleeding who refused hospitalization during the COVID pandemic. ****P values performed with Fisher exact test

Conclusion

DEP monitoring of arterial blood flow & guided focal treatment resulted in significantly lower rates of severe & total DPPHs than standard medical treatment. This is an easy, safe, & anatomically based method to prevent DPPH throughout the colon and rectum.

References

Jensen DM, Gornbein J, Fejleh P et al. Prediction and Prevention of Delayed Post-polypectomy-induced Ulcer Hemorrhage Based Upon an Anatomic Study and Multivariable Analyses. Dig Dis Sci 2026 Feb 10. Doi: 10.107/s10620-026-0970-2. Online ahead of print.

Disclosure

Study supported by a Veterans Administration Merit Review Grant - 5I01CX001403. Dr. Jensen has received research funding from Vascular Technology Inc.

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:

PREVENTION OF DELAYED POST-POLYPECTOMY HEMORRHAGE BY DOPPLER ENDOSCOPIC PROBE MONITORING AND FOCAL BLOOD FLOW ABLATION IN A RANDOMIZED CONTROLLED TRIAL

Jeffrey Gornbein

Exocrine pancreatic insufficiency: is there a pancreas-liver axis?

Ashkar Motaz

DETERMINANTS OF NEOPLASIA YIELD AT THE FIRST SURVEILLANCE COLONOSCOPY: THE IMPACT OF ENDOSCOPIST'S ADR

Carlo Senore

TEXTBOOK OUTCOMES OF EUS-GUIDED HEPATICOGASTROSTOMY ACCORDING TO INDICATION AND PRESENTING ANATOMY: A PROSPECTIVE COHORT STUDY

Paolo Giorgio Arcidiacono

Colorectal cancer metastasis: What did we learn from TransMet?

ARTIFICIAL INTELLIGENCE FOR PROCEDURAL GUIDANCE IN ERCP: A VIDEO DEMONSTRATION STUDY

Xiaofeng Zhang

UEG Presentation

New

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

Exocrine pancreatic insufficiency: is there a pancreas-liver axis?

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.

Exocrine pancreatic insufficiency: is there a pancreas-liver axis?

Ashkar Motaz 1

Affiliations

1 Mayo Clinic, Rochester, United States of America

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

Endoscopy IBD

Session

Hot topics in European Gastroenterology

Citation

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

Published

2026

More Like This:

PREVENTION OF DELAYED POST-POLYPECTOMY HEMORRHAGE BY DOPPLER ENDOSCOPIC PROBE MONITORING AND FOCAL BLOOD FLOW ABLATION IN A RANDOMIZED CONTROLLED TRIAL

Jeffrey Gornbein

Exocrine pancreatic insufficiency: is there a pancreas-liver axis?

Ashkar Motaz

DETERMINANTS OF NEOPLASIA YIELD AT THE FIRST SURVEILLANCE COLONOSCOPY: THE IMPACT OF ENDOSCOPIST'S ADR

Carlo Senore

TEXTBOOK OUTCOMES OF EUS-GUIDED HEPATICOGASTROSTOMY ACCORDING TO INDICATION AND PRESENTING ANATOMY: A PROSPECTIVE COHORT STUDY

Paolo Giorgio Arcidiacono

Colorectal cancer metastasis: What did we learn from TransMet?

ARTIFICIAL INTELLIGENCE FOR PROCEDURAL GUIDANCE IN ERCP: A VIDEO DEMONSTRATION STUDY

Xiaofeng Zhang

UEG Presentation

New

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

DETERMINANTS OF NEOPLASIA YIELD AT THE FIRST SURVEILLANCE COLONOSCOPY: THE IMPACT OF ENDOSCOPIST'S ADR

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.

DETERMINANTS OF NEOPLASIA YIELD AT THE FIRST SURVEILLANCE COLONOSCOPY: THE IMPACT OF ENDOSCOPIST'S ADR

Stefano Rousset 1,2, Gianluigi Ferrante 1, Marco Silvani 1, Marco Calcagno 1, Cristiano Piccinelli 1, Annalisa Castella 3, Dario Mazzucco 4, Arrigo Arrigoni 5, Fabiana Venezia 6, Lorenzo Marcello Orione 7, Vittoria Grammatico 8, Paola Cassoni 9, Maria Caterina Maconi 10, Cesare Hassan 11, Carlo Senore 1

Affiliations

1 University Hospital Città della Salute e della Scienza, Torino, Italy

2 School of Specialization in Statistics and Biometry, University of Turin, Torino, Italy

3 ASL TO3 Piedmont, Collegno (To), Italy

4 Ospedale di Rivoli, Rivoli (To), Italy

5 IRCCS Candiolo, Candiolo (To), Italy

6 ASL TO4 Piedmont, Chivasso (To), Italy

7 ASL CN1 Piedmont, Cuneo, Italy

8 ASL TO5 Piedmont, Chieri (To), Italy

9 University of Turin, Torino, Italy

10 ASL AL Piedmont, Alessandria, Italy

11 Humanitas University, Pieve Emanuele MI, Italy

Summary

AI Generated

Summary is not available for this content yet.

Download PDF

Was this helpful?

Thanks for your feedback.

Abstract

Introduction

Previous studies showed an inverse association with interval CRC (PCCRC) and the adenoma detection rate (ADR) of the endoscopist performing the colonoscopy (TC) following a positive FIT result (FIT+ index TC). While an inverse association between the characteristics of the index lesion and the risk of neoplasia at surveillance TC has been documented, there is limited evidence about the impact of endoscopist's ADR on this key intermediate outcome.

Aims & Methods

Aims: To analyse the relative weight of patient-, procedure-, and provider-level determinants of advanced adenoma (AAd) and CRC detection rate (DR) at the first surveillance TC.
Methods: Population: cohort of FIT+ screenees, within the Piedmont region screening program, who performed FIT between 2004 and 2022 and were referred to 3-year surveillance, following the excision of high-risk (HR) adenomas (> 2 adenomas or 1 tubular adenoma >9 mm or 1 adenoma with villous component > 20%, or with high-grade dysplasia). The ADR of the endoscopist performing the FIT+ index TC was measured as the proportion of TCs with at least one adenoma, out of the total number of TCs performed within the program between 2004 and 2022. ADR was categorized according to an approximation of the quintile distribution (≤39.9%; 40% - 44.9%, 45% - 49.9%; 50% - 54.9%, ≥55%). Operators who had performed < 100 TCs were excluded. Primary outcome: AAd DR; secondary outcome CRC DR. The association between explanatory variables and outcomes was assessed using multilevel logistic regression models with random intercepts for endoscopists.

Results

Out of 47,245 FIT-positive screenees undergoing TC, 14,423 (31%) were detected with HR adenomas, 11,100 (77%) were referred to 3-year surveillance and 8,063 of them (73%) underwent a surveillance TC. After excluding subjects with missing information about the endoscopist, inadequate preparation, or incomplete exam, 6909 HR patients (median age 65 years, 63% male) examined by 163 operators were included in the analysis: 936 were detected with an AAd (13.5%) and 31 with a CRC (0.4%). In multilevel models, older age (65-70 years: OR=1.18; 95% CI: 1.02–1.36), male sex (OR=1.63; 95% CI: 1.39–1.91), proximal polyp location (OR=2.08; 95% CI: 1.70-2.55 ), and suboptimal quality of bowel preparation at index TC (OR 1.29; 95% CI: 1.11–1.50) were associated with increased AAd risk at surveillance TC, while the AAd DR at surveillance TC showed a decreasing trend with increasing ADR of the endoscopist performing the FIT+ index TC, reaching a plateau for ADR ≥50% (OR=0.67; 95% CI: 0.50-0.89).

AAd risk at surveillance TC by endoscopist’s ADR at index FIT+ TC Adjusted by sex, age, polyp size, interval between index and surveillance TC, polyp location, quality of preparation of index TC
	OR*	95% CI
ADR ≤39.9	1.00	-
ADR 40-44.9	0.86	0.66 - 1.12
ADR 45-49.9	0.73	0.55 - 0.96
ADR 50-54.9	0.67	0.50 - 0.89
ADR ≥55	0.65	0.47 - 0.89

Conclusion

Endoscopist's competence (measured using ADR metrics) is affecting not only the PCCRC risk, but also the DR of neoplasia at the surveillance TC. These findings suggest that a substantial proportion of the observed risk of AN recurrence following excision of AAds might be attributable to the miss rate of lesions associated with low ADR at the index TC. Consistent with previous studies, we observed a plateau for the protective effect of ADR above 45%. Endoscopist' ADR should be regularly monitored, to optimize surveillance strategies among FIT+ screenees and it might be considered among the criteria for defining the surveillance interval following HR adenoma detection.

References

M. F. Kaminski et al., ‘Performance measures for lower gastrointestinal endoscopy: a European Society of Gastrointestinal Endoscopy (ESGE) quality improvement initiative’, United Eur. Gastroenterol. J., vol. 5, no. 3, pp. 309–334, 2017.
E. Dekker, et al. Post-polypectomy colonoscopy surveillance: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy 2013; 45: 842–851’, pp. 842–851
N. D. Pilonis et al., ‘Adenoma Detection Rates by Physicians and Subsequent Colorectal Cancer Risk.’, Jama, vol. 333, no. 5, pp. 400–407, 2024.
M. Zorzi et al., ‘Adenoma Detection Rate and Colorectal Cancer Risk in Fecal Immunochemical Test Screening Programs’, Ann. Intern. Med., vol. 176, no. 3, pp. 303–310, 2023.

Disclosure

No COI to disclose

Event

UEG Week Barcelona 2026

Topics

Colorectal Endoscopy Digestive Oncology

Submission format

Abstract

Session

Beyond colonoscopy: Improvements to come!

Citation

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

Published

2026

More Like This:

PREVENTION OF DELAYED POST-POLYPECTOMY HEMORRHAGE BY DOPPLER ENDOSCOPIC PROBE MONITORING AND FOCAL BLOOD FLOW ABLATION IN A RANDOMIZED CONTROLLED TRIAL

Jeffrey Gornbein

Exocrine pancreatic insufficiency: is there a pancreas-liver axis?

Ashkar Motaz

DETERMINANTS OF NEOPLASIA YIELD AT THE FIRST SURVEILLANCE COLONOSCOPY: THE IMPACT OF ENDOSCOPIST'S ADR

Carlo Senore

TEXTBOOK OUTCOMES OF EUS-GUIDED HEPATICOGASTROSTOMY ACCORDING TO INDICATION AND PRESENTING ANATOMY: A PROSPECTIVE COHORT STUDY

Paolo Giorgio Arcidiacono

Colorectal cancer metastasis: What did we learn from TransMet?

ARTIFICIAL INTELLIGENCE FOR PROCEDURAL GUIDANCE IN ERCP: A VIDEO DEMONSTRATION STUDY

Xiaofeng Zhang

UEG Presentation

New

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

TEXTBOOK OUTCOMES OF EUS-GUIDED HEPATICOGASTROSTOMY ACCORDING TO INDICATION AND PRESENTING ANATOMY: A PROSPECTIVE COHORT STUDY

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.

TEXTBOOK OUTCOMES OF EUS-GUIDED HEPATICOGASTROSTOMY ACCORDING TO INDICATION AND PRESENTING ANATOMY: A PROSPECTIVE COHORT STUDY

Giuseppe Vanella 1,2, Rosa Claudia Stasio 1,3, Rukaia Barà 1,2, Giuseppe Marzocca 1,2, Diego Palumbo 2,1, Francesco Frigo 1,3, Andrea Casadei-Gardini 2,1, Francesco De Cobelli 2,1, Michele Reni 2,1, Massimo Falconi 2,1, Paolo Giorgio Arcidiacono 1,2

Affiliations

1 IRCCS San Raffaele Scientific Institute, Milan, Italy

2 Vita-Salute San Raffaele University, Milan, Italy

3 University of Turin, Turin, Italy

Summary

AI Generated

Summary is not available for this content yet.

Download PDF

Was this helpful?

Thanks for your feedback.

Abstract

Introduction

Endoscopic Ultrasound-Guided HepaticoGastrostomy (EUS-HGS) has emerged as a valuable complement to retrograde biliary drainage. However current evidence is focused on technical success and constrained by small sample sizes, retrospective design, heterogeneous techniques and indications.

Aims & Methods

Single center prospective study conducted between April-2021 and October-2025. All consecutive patients undergoing EUS-HGS with a dedicated partially covered metal stent (Giobor, Taewoong) for a malignant indication were enrolled in a prospective registry of Therapeutic EUS (PROTECT; ClinicalTrials.gov NCT04813055) with prospective scheduled bimonthly assessment of outcomes. 30-days Net Clinical Benefit (NCB30) was defined as Clinical Success (CS) without recurrent biliary obstruction (RBO) or 30-days mortality, while Textbook Outcome (TBO30) was defined as NCB30 plus no 30-days procedure-related adverse events (AEs).

Results

83 patients were included (median age 67 [60-74], female 61.4%, pancreatic adenocarcinoma 50.6%, cholangiocarcinoma 20.5%, metastatic diseases 78.3%, ascites 41%).
Indication for EUS-HGS were: 1) distal malignant biliary obstruction (MBO) with duodenal invasion (43.4%); 2) drainage completion of hilar MBO (39.8%); 3) surgically-altered anatomy (16.9%).
Technical success and CS rate were 95.2% and 70.9% respectively. AEs were registered in 43.4% of patients (mild/moderate 19.5%, severe/fatal 25.6%) and were deemed procedure-related in 28.4%. 30-days mortality was 21.8% (cancer-related in 83%). After a median follow-up of 63 [32-113] days, RBO was registered in 6.8%.
NCB30 and TBO30 were 54.2% and 45.8% respectively. Best Supportive Care (BSC) was advised in 60.9% of patients.
CS was significantly lower when the stenosis was hilar versus distal/anastomotic (p=0.001). Accordingly, the NCB30 was significantly lower when EUS-HGS drained only one segment versus the left hemiliver versus the whole liver (36.4% versus 45.5% versus 73.5%, p=0.02). Since the AEs were also higher in patients undergoing partial versus complete liver drainage (p=0.001), the TBO30 was accordingly different, being 9.1% versus 39.4% and 67.6% respectively (p=0.002) as well as advice for BSC (p=0.0008).

Conclusion

Despite excellent technical success, only ≈50% of patients experience a net clinical benefit or a textbook outcome after EUS-HGS, whilst the rest receive a recommendation for BSC. Outcomes are deeply influenced by the level of the stenosis and actual liver drainage, claiming for personalized imaging-guided patients’ selection.

Event

UEG Week Barcelona 2026

Topics

Radiology & Imaging Hepatobiliary Endoscopy

Submission format

Abstract

Session

Endoscopy: Taming the biliary tree

Citation

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

Published

2026

More Like This:

PREVENTION OF DELAYED POST-POLYPECTOMY HEMORRHAGE BY DOPPLER ENDOSCOPIC PROBE MONITORING AND FOCAL BLOOD FLOW ABLATION IN A RANDOMIZED CONTROLLED TRIAL

Jeffrey Gornbein

Exocrine pancreatic insufficiency: is there a pancreas-liver axis?

Ashkar Motaz

DETERMINANTS OF NEOPLASIA YIELD AT THE FIRST SURVEILLANCE COLONOSCOPY: THE IMPACT OF ENDOSCOPIST'S ADR

Carlo Senore

TEXTBOOK OUTCOMES OF EUS-GUIDED HEPATICOGASTROSTOMY ACCORDING TO INDICATION AND PRESENTING ANATOMY: A PROSPECTIVE COHORT STUDY

Paolo Giorgio Arcidiacono

Colorectal cancer metastasis: What did we learn from TransMet?

ARTIFICIAL INTELLIGENCE FOR PROCEDURAL GUIDANCE IN ERCP: A VIDEO DEMONSTRATION STUDY

Xiaofeng Zhang

UEG Presentation

New

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

Colorectal cancer metastasis: What did we learn from TransMet?

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.

Colorectal cancer metastasis: What did we learn from TransMet?

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

Surgery Hepatobiliary Digestive Oncology Colorectal

Session

Expanding the boundaries of transplant oncology

Citation

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

Published

2026

More Like This:

PREVENTION OF DELAYED POST-POLYPECTOMY HEMORRHAGE BY DOPPLER ENDOSCOPIC PROBE MONITORING AND FOCAL BLOOD FLOW ABLATION IN A RANDOMIZED CONTROLLED TRIAL

Jeffrey Gornbein

Exocrine pancreatic insufficiency: is there a pancreas-liver axis?

Ashkar Motaz

DETERMINANTS OF NEOPLASIA YIELD AT THE FIRST SURVEILLANCE COLONOSCOPY: THE IMPACT OF ENDOSCOPIST'S ADR

Carlo Senore

TEXTBOOK OUTCOMES OF EUS-GUIDED HEPATICOGASTROSTOMY ACCORDING TO INDICATION AND PRESENTING ANATOMY: A PROSPECTIVE COHORT STUDY

Paolo Giorgio Arcidiacono

Colorectal cancer metastasis: What did we learn from TransMet?

ARTIFICIAL INTELLIGENCE FOR PROCEDURAL GUIDANCE IN ERCP: A VIDEO DEMONSTRATION STUDY

Xiaofeng Zhang

UEG Presentation

New

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

ARTIFICIAL INTELLIGENCE FOR PROCEDURAL GUIDANCE IN ERCP: A VIDEO DEMONSTRATION STUDY

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.

ARTIFICIAL INTELLIGENCE FOR PROCEDURAL GUIDANCE IN ERCP: A VIDEO DEMONSTRATION STUDY

Xiaofeng Zhang 1, jing Yang 2

Affiliations

1 Hangzhou First People's Hospital, Hangzhou, China

2 Affiliated Hangzhou First People's Hospital, School of Medicine, Westlake University, Hangzhou, China

Summary

AI Generated

Summary is not available for this content yet.

Download PDF

Was this helpful?

Thanks for your feedback.

Abstract

Introduction

Selective biliary cannulation during endoscopic retrograde cholangiopancreatography (ERCP) is technically challenging. Successful cannulation depends not only on papillary orientation but also on the overall configuration of the endoscope. We developed an artificial intelligence (AI)-assisted guidance systemand present a three-case video demonstration to illustrate its potential value in guiding novice ERCP operators during selective biliary cannulation.

Aims & Methods

The system integrates real-time analysis of both endoscopic and fluoroscopic images. In the endoscopic view, the system automatically identifies the papillary axis and displays the predicted direction for biliary cannulation: a thin yellow line indicates the papillary axis, while a thicker orange-pink line represents the suggested cannulation path. In the fluoroscopic view, the system analyzes duodenoscope configuration, including shaft course, luminal curvature, and bending segment angle. These parameters characterize the endoscopic course within the stomach and duodenum, thereby reflecting procedural compliance and providing additional guidance for assessing case suitability for junior endoscopists.

Results

During cannulation attempts in all three cases, the system generated continuously updated real-time overlays. The endoscopic module provided visual guidance for alignment with the presumed biliary direction, while the fluoroscopic module supplemented information on scope configuration and distal bending. This integrated visualization helped interpret the relationship between local papillary orientation and global endoscope geometry during selective biliary access.

Conclusion

Real-time AI-assisted guidance during ERCP is feasible and may offers practical support for selective biliary cannulation. By combining papillary axis recognition with fluoroscopic assessment of scope configuration, this approach may improve cannulation positioning accuracy and enhance understanding of cannulation difficulty. Further prospective evaluation is needed to evaluate its impact on cannulation success rates, procedure time, and training outcomes.

References

Event

UEG Week Barcelona 2026

Topics

Endoscopy Hepatobiliary

Submission format

Video Case

Session

Lights, camera, lumen! High-impact endoscopy video cases

Citation

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

Published

2026

More Like This:

PREVENTION OF DELAYED POST-POLYPECTOMY HEMORRHAGE BY DOPPLER ENDOSCOPIC PROBE MONITORING AND FOCAL BLOOD FLOW ABLATION IN A RANDOMIZED CONTROLLED TRIAL

Jeffrey Gornbein

Exocrine pancreatic insufficiency: is there a pancreas-liver axis?

Ashkar Motaz

DETERMINANTS OF NEOPLASIA YIELD AT THE FIRST SURVEILLANCE COLONOSCOPY: THE IMPACT OF ENDOSCOPIST'S ADR

Carlo Senore

TEXTBOOK OUTCOMES OF EUS-GUIDED HEPATICOGASTROSTOMY ACCORDING TO INDICATION AND PRESENTING ANATOMY: A PROSPECTIVE COHORT STUDY

Paolo Giorgio Arcidiacono

Colorectal cancer metastasis: What did we learn from TransMet?

ARTIFICIAL INTELLIGENCE FOR PROCEDURAL GUIDANCE IN ERCP: A VIDEO DEMONSTRATION STUDY

Xiaofeng Zhang