WATERFALL Study Should Have Us Rethinking Fluid Resuscitation in Acute Pancreatitis

David A. Johnson, MD


October 07, 2022

This transcript has been edited for clarity.

Hello. I'm Dr David Johnson, professor of medicine and chief of gastroenterology at Eastern Virginia Medical School in Norfolk, Virginia. Welcome back to another GI Common Concerns.

Fluid resuscitation has been our only approach for the initial treatment of patients with acute pancreatitis, given the lack of pharmacologic interventions. The goal of fluid resuscitation is to try to prevent these patients from developing moderately severe or severe disease, which occurs in approximately 35% of patients presenting with acute pancreatitis and is associated with worse outcomes.

We know from animal models that there is regional hypoperfusion of the pancreas, and this specifically correlates with necrosis. This may be corrected by fluid resuscitation.

National guidelines have mostly been staunch in recommending the use of aggressive fluid resuscitation, particularly during the first 24 hours. Randomized trials, however, have been somewhat mixed. A systematic review and meta-analysis suggested that the heterogeneity of these trials is such that we probably have not been getting the correct message, and it actually may be the converse, that overaggressive fluid resuscitation may have consequent adverse effects.

Dipping Into the Findings of WATERFALL

Into this ongoing discussion we now have the landmark publication of the WATERFALL study, which was recently published in The New England Journal of Medicine. This is an early weight-based comparison of aggressive vs nonaggressive goal-directed fluid resuscitation in the early phase of acute pancreatitis.

At 18 centers in India, Italy, Spain, and Mexico, patients were randomized in a 1:1 ratio after being diagnosed with acute pancreatitis using the Revised Atlanta Classification, which requires two of three criteria: typical abdominal pain, serum amylase or lipase level higher than three times the upper limit of the normal range, or signs of acute pancreatitis on imaging. Investigators excluded patients who had moderately severe or severe systemic disease, those with multiple comorbidities portending a worse outcome, and those who had symptom onset for more than 24 hours. Patients were randomized within at least 8 hours after the initial diagnosis was made.

Investigators then began assigning patients to receive either aggressive or moderate fluid resuscitation. Aggressive fluid resuscitation consisted of a bolus of 20 mL/kg of body weight over 2 hours, followed by 3 mL/kg per hour. Moderate fluid resuscitation consisted of a bolus of 1.5 mL/kg per hour. In this group, those who were hypovolemic also received a bolus of 10 mL/kg, whereas no bolus was given in those who were euvolemic.

Patient assessment was performed at 12, 24, 48, and 72 hours. The primary outcome was the development of moderately severe or severe pancreatitis during the hospitalization, whereas the primary safety outcome was fluid overload, defined by symptoms, physical signs, and imaging criteria of hypervolemia.

Initially, investigators estimated that they would need a sample size of 744 patients. The first interim safety analysis was planned for when the study had enrolled 248 patients. At that point, however, the trial was stopped. The reason for this is that there was no difference between the moderate and aggressive fluid resuscitation for outcomes; although, in the moderate resuscitation cohort, it was trending toward less development of moderately severe or severe disease, hospitalizations, and organ failure.

There was a major difference, however, in the primary safety outcome of fluid overload. The incidence of fluid overload was 20.5% in the aggressive-resuscitation arm and 6.3% in the moderate-resuscitation arm, resulting in an adjusted relative risk of 2.85 in favor of the latter arm. This prompted the data and safety monitoring board to stop the study.

Notably, the patients were all started on oral feedings within 12 hours if the intensity of abdominal pain — described using the validated, multilingual PAN-PROMISE instrument — was less than 5 out of a total score of 10.

PAN-PROMISE was derived to assess seven domains of patient-reported outcomes in acute pancreatitis trials. Of these domains, I think the most important are those related to abdominal pain, tolerance of bloating, and nausea and/or vomiting.

Nonetheless, once they were scored as less than 5, they were started on oral feedings, with fluid resuscitation able to stop if patients tolerated oral feeding for 8 hours. These oral feedings were continued, and the infusion was stopped as early as 20 hours after randomization in the moderate-resuscitation group and 48 hours in the aggressive-resuscitation group.

Game-Changing Takeaways

These results tell us a couple of things.

First, the fluid used is lactated Ringer's solution. We should consider that a standard of care.

Second, appropriate fluid resuscitation is critically important because the overload is where we really run into problems of patient safety. Again, this is something that's predicted based on hypoperfusion and hypovolemia. We need to do a bolus appropriately in some patients, but otherwise most patients who are euvolemic can just be started on a maintenance dose, with early initiation of oral feedings and stopping the infusion once these feedings are tolerated.

So, in summary, these results are a wake-up call, they're axiomatic, and they're a game-changer. Fluids are the answer, but we should use moderate resuscitation. In patients who are euvolemic, using maintenance fluids is important. Again, lactated Ringer's solution is the standard of care. And the even newer standard of care is appropriate fluids in acute pancreatitis.

These results are going to change my practice and should change your practice as well.

I'm Dr David Johnson. Thanks again for listening.

David A. Johnson, MD, a regular contributor to Medscape, is professor of medicine and chief of gastroenterology at Eastern Virginia Medical School in Norfolk, Virginia, and a past president of the American College of Gastroenterology. His primary focus is the clinical practice of gastroenterology. He has published extensively in the internal medicine/gastroenterology literature, with principal research interests in esophageal and colon disease, and more recently in sleep and microbiome effects on gastrointestinal health and disease.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.