Post-ERCP Pancreatitis Prophylaxis: Rectal NSAIDs, Aggressive Fluids, or Both, in a High-Risk Patient?
A patient stacking three independent risk factors for post-ERCP pancreatitis turns a routine prophylaxis choice into a real question about whether one well-studied measure is enough, or whether her risk profile calls for the combination.
T.C. is scheduled for ERCP this afternoon to evaluate recurrent right-upper-quadrant pain that has brought her back to the emergency department twice this year, each time with mildly elevated liver enzymes that normalized within days without an identifiable stone ever being found — a pattern her gastroenterologist suspects is sphincter of Oddi dysfunction, one of the strongest known risk factors for post-ERCP pancreatitis in its own right. She is 67, which is not itself a risk factor the way younger age is, but she does carry two others that matter more here: a prior episode of post-ERCP pancreatitis, from a procedure done elsewhere four years ago, and today's planned intervention includes pancreatic duct injection as part of the sphincter of Oddi workup, itself an independent risk factor for the complication the team is trying to prevent.
Stacked together — suspected sphincter of Oddi dysfunction, a prior episode, and planned pancreatic duct injection — she carries three independently documented risk factors for post-ERCP pancreatitis in a single procedure, a combination her endoscopist has not seen this concentrated in one patient in recent memory. Prophylaxis itself is not in question. What her three factors cannot settle is whether stacking a second measure on top of the best-studied one has ever actually been shown to help a patient like her, or whether it only feels proportionate to how worried she makes the room.
Her prior post-ERCP pancreatitis, four years ago, was not a mild, self-limited episode — she spent four days in the hospital on that occasion, an experience she has mentioned unprompted at both of this year's emergency visits, clearly still shaping how she thinks about today's procedure. Her outside records from that admission, obtained specifically for today's planning, confirm the diagnosis was genuine post-ERCP pancreatitis rather than an unrelated complication, which is part of why her current team is treating that history as a real, weighty data point rather than an anecdote colored by memory.
Endoscopy suite, before the procedure
Rectal indomethacin, given now before the procedure starts. It's simple, requires no additional infrastructure, and the pivotal randomized trial (Elmunzer et al., NEJM, 2012) showed a real reduction in post-ERCP pancreatitis in a high-risk population that looks a lot like her.
Indomethacin's trial support is real, but it isn't the only prophylactic measure with trial evidence behind it. Aggressive periprocedural lactated Ringer's has its own separate randomized support (Buxbaum et al., 2014) for reducing post-ERCP pancreatitis, and if combination strategies genuinely add benefit in high-risk populations, then relying on a single measure may be under-treating a patient carrying three stacked risk factors at once.
I'd push back on that, because the trial built to answer it came out the other way. The additive signal you're describing comes from Mok et al. (2017), a four-arm study of fewer than 200 patients whose headline comparison was lactated Ringer's plus indomethacin against saline plus placebo — not against indomethacin alone. FLUYT (Sperna Weiland et al., Lancet Gastroenterology & Hepatology, 2021) then randomized 826 moderate-to-high-risk patients across 22 centers to aggressive periprocedural lactated Ringer's on top of a rectal NSAID or to the NSAID alone, and found no reduction in post-ERCP pancreatitis; the authors concluded the added fluid burden isn't justified. Her three stacked factors are real, but stacking risk doesn't convert a negative trial into a positive one for her. What her profile does argue for is the measure FLUYT never tested — a prophylactic pancreatic duct stent, which has its own evidence in precisely this population, and which is a decision for the endoscopist at the papilla rather than a drug we add now.
Agreed: rectal indomethacin before the procedure, lactated Ringer's at standard maintenance rather than aggressive rates, and a prophylactic pancreatic duct stent recommended to the endoscopist.
Not agreed: whether a risk profile as stacked as hers could ever justify departing from a negative trial's result. The pharmacologist holds that FLUYT settles the hydration question for everyone it enrolled, T.C. included; the endoscopist is less willing to treat a trial's average as binding on the most extreme patient in it, while conceding that nothing in the data supports his instinct.