Online Poster Portal

Introduction 

A recent randomized controlled trial investigated the efficacy of ice-water irrigation of the ampulla (cryoprevention) following endoscopic cholangiopancreatography (ERCP) for the prevention of post-ERCP pancreatitis (PEP). Patients in this study did not receive rectal indomethacin, though in the United States this is a guideline-recommended prophylactic intervention, known to reduce the risk of PEP. The generalizability and added benefit of cryoprevention remain uncertain. We aimed to evaluate the efficacy of cryoprevention in a multicenter cohort, adjusting for use of indomethacin and other relevant confounders.  

Methods 

We conducted a multicenter, retrospective cohort study at five academic hospitals in the US. Data were reviewed for consecutive patients with native papillae undergoing ERCP between May and November 2025. The primary outcome was incidence of PEP. Secondary outcomes were the incidence of PEP in ‘high-risk’ patients (defined as having ≥ 1 traditional risk-factor for PEP). We used a propensity score model with inverse probability weighting (IPW) that included relevant demographic, clinical, and procedural covariates. A weighted logistic regression model was then fit to obtain adjusted odds ratios for the association between cryoprevention and PEP. 

Results 

Across five academic medical centers, 371 ERCPs were performed between May and November 2025; 129 patients received cryoprevention. The frequency of PEP in the entire cohort was 4.6%. Indomethacin was administered to 80% of the cohort. Table 1 summarizes patient characteristics in the cryoprevention and standard care groups. In the IPW analysis, cryoprevention was associated with numerically lower odds of PEP across all patients (OR 0.51, 95% CI 0.16–1.68, p=0.27) and among patients with any high-risk feature (OR 0.30, 95% CI 0.06–1.43, p=0.13), though differences did not reach statistical significance. Exploratory subgroup analyses suggested in patients with pancreatic duct cannulation (OR 0.85, 95% CI 0.74–0.98, p=0.02) or a history of pancreatitis (OR 0.77, 95% CI 0.59–1.00, p=0.05), use of cryoprevention was independently associated with reduced risk of PEP. Other subgroups, including those with difficult cannulation (OR 0.83, 95% CI 0.67–1.02, p=0.08) also showed modest trends in the same direction.  

Discussion 

Though cryoprevention was associated with a numerical trend toward reduced odds of post-ERCP pancreatitis, differences did not reach statistical significance. The beneficial effect of cryoprevention may be attenuated in cohorts receiving rectal indomethacin. Any potential protective effect of cryoprevention may have been further masked by the low overall rate of PEP in this study. Given signals for benefit in particular subgroups, larger, robustly powered studies are underway and will be necessary to characterize the utility of cryoprevention.