Is Indomethacin a New Hope for Post-ERCP Pancreatitis?

Yu Bai,Duowu Zou,Zhaoshen Li

DOI: https://doi.org/10.1111/j.1572-0241.2007.01324_1.x

2007-01-01

Abstract:TO THE EDITOR: We read with great interest a randomized controlled trial by Sotoudehmanesh et al. (1) titled “Indomethacin May Reduce the Incidence and Severity of Acute Pancreatitis After ERCP” that has been published in the recent issue of the Journal. The authors concluded that rectal indomethacin given right before endoscopic retrograde cholangiopancreatography (ERCP) can reduce the incidence and severity of post-ERCP pancreatitis (PEP). However, before incorporating this inexpensive, easily available nonsteroidal anti-inflammatory drug (NSAID) into our clinical practice for PEP prevention, several issues need further clarification: (a) It had been pointed out from a prospective large sample study that only individuals with serum amylase levels fivefold the normal upper limits and persistent pancreaticlike pain for 24 h after ERCP strongly suggest possibility of PEP (2); however, the authors stated that besides the 2-h measurement, serum amylase was repeated the next morning, and it is not clear what the exact time interval was. This may cast some doubt on the accuracy of the diagnosis of PEP in this study. (b) In the Methods section, the authors claimed subgroup analysis was used to evaluate the effect of indomethacin related to different variables. Notwithstanding, due to the lack of relevant information, it is not clear whether these subgroup analyses were ad hoc or post hoc. If these subgroup analyses were performed retrospectively, it should be considered as “fishing” for positive results, and the results from subgroup analyses should be treated with caution (3). (c) No explanation was provided as to why none of the sphincter of Oddi dysfunction (SOD) patients developed PEP, which is contrary to most other reported trials, with PEP incidence of up to 24% for patients with suspicious SOD (4). (d) They found that pancreatic duct injection, pancreatic duct cannulation more than once, and age less than 60 yr were significant risk factors for developing PEP. Certainly, it is convenient to perform a univariate analysis for exploration of each variable’s influence on the primary outcome; nonetheless, up to now, available evidence suggests that the incidence of PEP was not determined by any single factor. Thus, a multivariate statistical analysis with logistic regression allowing estimation of the influence of covariates on the incidence of PEP seems more appropriate. To summarize, this study suggests indomethacin may be an attractive pharmacotherapy for the prevention of PEP because it is inexpensive, universally available, and relatively safe. Clarification on some raised concerns will better our understanding of the mystery of PEP and its pharmacotherapy prevention. Yu Bai, M.D. Duowu Zou, M.D. Zhaoshen Li, M.D.