A network meta-analysis of optimal strategies for preventing post-endoscopic retrograde cholangiopancreatography pancreatitis

Numerous interventions have been proposed to reduce the risk of post-endoscopic retrograde cholangiopancreatography (ERCP) pancreatitis (PEP), but it remains uncertain which approach is the most effective. The objective of this network meta-analysis was to evaluate and compare different preventive strategies for overall, mild, and moderate-to-severe PEP in both average- and high-risk patients. A systematic search was conducted for randomized controlled trials across the PubMed, Embase, and Cochrane Central databases. The inclusion criteria encompassed studies that featured at least two of the following interventions: pre-procedural administration of 100 mg of rectal diclofenac (D) or indomethacin (I), aggressive hydration utilizing normal saline (NS) or lactated Ringer's solution (LR), 5-french pancreatic duct stent (S), and either a placebo or an active control (C). The interventions were ranked according to their surface under cumulative ranking (SUCRA) values. The study protocol has been registered in PROSPERO (CRD42024538533). Of the total 42 (n = 11,493) identified studies, 23 studies (n = 5,410) were categorized into the average-risk group and 19 studies (n = 6,083) were categorized into high-risk group. Based on SUCRA, I + NS was ranked as the best preventive method for overall and mild PEP in average-risk group. However, I + LR was the most effective method of prevention of moderate-to-severe PEP in this group. In high-risk group, I + S was the best preventive method of overall and mild PEP. However, D was the most effective method of prevention of moderate-to-severe PEP in this group. The findings of this network meta-analysis indicate that endoscopists should consider all available prophylactic options when performing ERCP, rather than favoring one over the others.