Groove area involvement predicts post-ERCP pancreatitis after 8-mm fully-covered metal stent placement in resectable pancreatic cancer

Abstract Fully covered self-expandable metal stents (FCSEMSs) provide durable preoperative biliary drainage in pancreatic cancer but may increase risk of post-endoscopic retrograde cholangiopancreatography (ERCP) pancreatitis (PEP). We evaluated whether groove involvement was an independent anatomical PEP risk factor and compared PEP incidence after 8-mm FCSEMS and plastic stent (PS) placement using propensity-score–based inverse probability of treatment weighting (IPTW). Sixty-two consecutive patients with resectable or borderline resectable pancreatic cancer and distal biliary strictures with naïve papillae underwent ERCPs between February 2015 and August 2024. An 8-mm FCSEMS or PS (7–11.5F) was placed. Independent PEP predictors were identified using multivariable Firth logistic regression. PEP incidence was compared between stent types after IPTW adjustment for age, sex, clinical stage, groove involvement, main pancreatic duct diameter, and prophylactic pancreatic-stent placement. Mean age was 73.3 ± 8.2 years (62.9% male). Groove-area extension was present in 21.0% of tumors. PEP occurred in six patients (9.7%), all after FCSEMS placement. Groove involvement independently predicted PEP (adjusted odds ratio, 14.7; 95% confidence interval, 2.26–95.9; P = 0.005). After IPTW, the weighted PEP rate remained higher with FCSEMS than PS (13.4% vs 0%; P = 0.011). Groove-area tumor extension is an independent imaging-detectable PEP risk factor. Even after baseline difference adjustment, 8-mm FCSEMS placement was associated with a higher pancreatitis risk than PS placement. Pre-procedural groove involvement identification may guide stent selection and support selective prophylactic pancreatic stenting. However, further confirmation through larger prospective studies is required.