Stent-or-Not: ERCP Pancreatitis Prophylaxis Selector
A tool that tells doctors if a patient needs a pancreatic stent or just medication to prevent inflammation.
A point-of-care tool that scores each ERCP patient's risk of post-procedure pancreatitis and tells the endoscopist whether rectal indomethacin alone is enough or a protective pancreatic stent should be added, using the new trial's numbers. It flags the highest-risk cases where stent benefit is greatest so the team plans the procedure around guaranteed stent access.
Concept
A point-of-care decision-support screen used in the ERCP suite (tablet or EHR-embedded). The endoscopist enters — or the record auto-populates — the established predictors of post-ERCP pancreatitis (PEP): young age, female sex, prior PEP, suspected sphincter of Oddi dysfunction, normal bilirubin without stone, plus intraprocedural factors such as pancreatic-duct injection, difficult cannulation, or precut sphincterotomy. The tool assigns a risk tier and returns one recommendation with the numbers behind it:
- Average risk: rectal indomethacin alone — the stent's technical complexity, cost, and potential harms are not justified.
- High risk: rectal indomethacin plus a prophylactic pancreatic stent, per the new trial.
- Highest risk: a "maximum stent benefit" flag — the trial found the stent's benefit most prominent in the highest-risk patients — prompting the team to secure pancreatic-duct access early and stage the case so stent placement is not aborted. The screen displays the trial's absolute figures (PEP 14.9% on indomethacin alone vs 11.3% with the stent in high-risk patients) for the consent conversation, and writes the decision to the procedure record for quality reporting.
Why now
A 1,950-patient randomized trial across 20 US and Canadian referral centres (2015–2023) just closed the question of whether rectal indomethacin could replace the stent: it could not. PEP occurred in 14.9% of high-risk patients on indomethacin alone versus 11.3% with indomethacin plus stent (risk difference 3.6%, 95% CI 0.6–6.6), non-inferiority failed (p=0.18), and a post-hoc intention-to-treat analysis found indomethacin alone inferior (p=0.011), with the stent's benefit most prominent among the highest-risk patients. For years, preliminary evidence suggested stents could be skipped, leaving practice inconsistent; the trial now quantifies exactly where the stent's ~3.6% absolute benefit justifies its complexity — the precise input a risk-tiered selector needs.
Written by GLM · 11 Oct 2026
AI assessment
A highly focused clinical decision support tool that translates a specific, high-impact trial into a bedside workflow to reduce post-ERCP pancreatitis.
- Evidence strength 5/5
- The idea is directly derived from a large-scale (1,950 patient), multi-center randomized controlled trial with clear absolute risk reductions and failed non-inferiority, providing a robust quantitative basis for the tool's logic.
- Market pull 4/5
- High-volume endoscopy centers have a strong incentive to reduce PEP rates to avoid costly hospital readmissions and improve patient safety, though the buyer is likely the hospital system rather than the individual endoscopist.
- Novelty & moat 3/5
- While the clinical application is new, the mechanism is a standard decision-support calculator; it is novel in its specific application to this trial but follows a pattern seen in other cataloged EHR tools.
- Feasibility 5/5
- The tool is a simple logic-based calculator using established risk factors and trial data, making it extremely easy to build as a tablet app or EHR integration.
- Wedge clarity 5/5
- The wedge is exceptionally sharp: a single, specific clinical question (stent or no stent) for a specific procedure (ERCP) in a specific patient population (high-risk).
- Simplicity / focus 5/5
- The product avoids 'platform' creep, focusing entirely on one decision point and its associated documentation.
Scored by Gemma 4 · 11 Oct 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity), relative to the closest ideas already in the catalog. A prior estimate to compare ideas before real-world signal arrives.
Persona discussion
AI personas trained on real people's expertise debate this idea as it evolves.
View the discussion →Act on this idea
Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.
Who benefits
- Provationcompany
Its endoscopy documentation platform is used in thousands of procedure units; embedding the risk-tiered selector puts the recommendation directly into the workflow endoscopists already use at the scope.
- Mayo Clinicorganization
A very high-volume ERCP referral service sees a disproportionate share of the highest-risk patients in whom the trial showed stent benefit is most prominent; the tool standardizes tiering across many endoscopists.
- Kaiser Permanenteorganization
Its integrated GI service line can deploy the selector system-wide and measure post-ERCP pancreatitis rate changes across a defined population.
- Gastro Healthcompany
A large multi-state GI network whose community ERCP sites mix average- and high-risk cases; the tool prevents both unnecessary stenting and missed stents.
The society's quality and education programs can distribute a guideline-concordant selection tool to members as the trial updates practice standards.
Research it builds on
- Indomethacin with or without Prophylactic Pancreatic Stent Placement to Prevent Pancreatitis after ERCP: A Randomised Non-Inferiority TrialB. Joseph Elmunzer, B Joseph Elmunzer, Lydia D Foster et al. · 2026 · 114 citationsAll ideas from this paper →