Dhruvil Radadiya, MD1, Bhaumik Brahmbhatt, MBBS2, Paris Charilaou, MD3, Chakradhar Reddy, MD1, Kalpit Devani, MD1; 1East Tennessee State University, Johnson City, TN; 2Mayo Clinic, Jacksonville, FL; 3Saint Peter's University Hospital, New Brunswick, NJ
Introduction: Post-endoscopic retrograde cholangiopancreatography pancreatitis (PEP) is a major adverse event after endoscopic retrograde cholangiopancreatography. No RCT has compared the efficacy of all American Society of Gastrointestinal Endoscopy (ASGE)- and European Society of Gastrointestinal Endoscopy (ESGE)-recommended interventions for PEP prevention. We assessed the effectiveness of these interventions and their combinations amongst each other using network meta-analysis approach. Methods: We conducted a literature search of PubMed, EMBASE, and Cochrane databases from inception to January 14, 2020, to identify RCTs comparing ASGE-and ESGE- recommended interventions -including rectal NSAIDs, pancreatic stent (PS), aggressive hydration (AH), sublingual nitrate or combinations of those - with each other or with the control group for PEP prevention (primary outcome). RCTs defining PEP only as per cotton's criteria or consensus criteria were included. Odds ratios (ORs) with 95% credible intervals (CrIs) were reported for network comparisons. We also did a subgroup network meta-analysis for high-risk patients. We used the surface under the cumulative ranking curve to rank interventions in terms of PEP prevention. Model consistency was assessed with the node splitting method. Results: We identified a total of 38 RCTs with ten different interventions (Figure 1). Each intervention was protective against PEP on network meta-analysis compared to controls as shown in figure 2. Except AH+diclofenac and NSAIDs+ sublingual nitrate, AH+indomethacin was associated with a significant reduction in risk of PEP compared to PS (OR:0.09; CrI:0.003-0.71), indomethcin+PS (OR:0.09; CrI:0.003-0.85), diclofenac (OR:0.09; CrI:0.003-0.65), AH (OR:0.09; CrI:0.003-0.65), sublingual nitrate (OR:0.07; CrI:0.002- 0.63), and indomethacin (OR:0.06; 0.002-0.43). AH with either rectal NSAIDs or sublingual nitrate had similar efficacy as shown in Figure 2. AH+indomethacin was the best intervention for preventing PEP with 95.3% probability of being ranked first (figure 3A). For high-risk patients, although the efficacy of PS and indomethacin were comparable, PS had an 80.8% probability of being ranked first(figure 3B). Node splitting with p-value < 0.05 for direct and indirect comparison confirmed model consistency. Discussion: Combination of rectal indomethacin with aggressive hydration seems the best intervention for preventing PEP. For high-risk patients, pancreatic stent seems the most effective strategy.
Network Plots Showing Direct Comparisons Available Among Interventions. A) All interventions. B) Interventions for high-risk patients. The thickness of the line connecting two interventions are proportional to available RCTs comparing them.
League table showing results of all possible comparison of all interventions. Odds ratio with 95% CrI in a particular cell represents a comparison of column defining treatment to raw defining treatment. All statistically significant comparisons are highlighted in green colour. AH: Aggressive hydration; I: Indomethacin; D: Diclofenac; N: Nitrate; S: Stent. "+" represents a combination of intervention
SUCRA Based on Cumulative Probabilities. A) All interventions. B) Interventions for high-risk patients. AH indicates aggressive hydration; D, rectal diclofenac; I, rectal indomethacin; N, sublingual nitrate; S, pancreatic stent; SUCRA, surface under the cumulative
Disclosures: Dhruvil Radadiya indicated no relevant financial relationships. Bhaumik Brahmbhatt indicated no relevant financial relationships. Paris Charilaou indicated no relevant financial relationships. Chakradhar Reddy indicated no relevant financial relationships. Kalpit Devani indicated no relevant financial relationships.