WATERFALL: Aggressive or Moderate Fluids in Acute Pancreatitis

Aggressive or Moderate Fluid Resuscitation in Acute Pancreatitis.

de-Madaria E, Buxbaum JL, Maisonneuve P, et al.; ERICA Consortium.

N Engl J Med. 2022 Sep 15;387(11):989-1000.

NEJM Video Summary

Aggressive fluid resuscitation for acute pancreatitis has been a long standing practice and it was thought to prevent complications such as necrosis. But as guidelines and others have noted, the evidence base supporting such as practice is “paltry” and of poor quality – so much so that fluids in general for acute pancreatitis are “anybody’s guess” [1]. The best quality evidence available prior to the current study, a that found that “moderate-rate fluid infusion should be preferred over high-rate infusion” though evidence was of low-certainty [2].

WATERFALL was designed as a large enough study to investigate the safety and efficacy of aggressive fluid resuscitation as compared with moderate fluid resuscitation in a diverse sample of patients with acute pancreatitis of varying severity.

Patient population and Design

This was a multicenter, open-label, , randomized, controlled trial from 18 centers across four countries (India, Italy, Mexico, and Spain). Patients ≥18 years with a acute pancreatitis according to the Revised Atlanta Classification (see below) were eligible as long as they presented within 24 hr of pain onset.

Atlanta Criteria for acute pancreatitis
(1) serum lipase or amylase >3 times upper limit of normal
(2) characteristic findings on CT, and
(3) epigastric abdominal pain

Key exclusion criteria included severe disease at baseline (shock, respiratory failure, and acute renal failure) or a history of heart failure (NYHA II, III, or IV), uncontrolled hypertension, hypernatremia, hyponatremia, hyperkalemia, hypercalcemia, chronic pancreatitis, chronic renal failure, or decompensated cirrhosis.

Patients were randomly assigned in a 1:1 ratio to receive aggressive fluid resuscitation (aggressive-resuscitation group) or moderate fluid resuscitation. In the aggressive-resuscitation group, a bolus of 20 mL/kg (70kg -> 1.4L) LR over 2 hr, followed by 3 mL/kg/hr (70 kg -> 210 mL/hr).

In the moderate-resuscitation group, patients were bolused 10mL/kg (700mL) LR only if hypovolemic and then all received 1.5 mL/kg/hr (105 mL/hr). Hypovolemia was defined as follows:

Hypovolemia was defined as follows
– Cr >1.1 mg/dL or BUN>20 mg/dL
– Hct >44%
– Increase in Cr and/or BUN from baseline
– UOP <0.75 mL/kg/hr
– SBP <90 mmHg without other explanation than hypovolemia
– Signs and/or symptoms of dehydration (intense thirst, dehydrated oral mucosa, decreased skin turgor–skin pinch)

Oral feeding was started at 12 hours if the intensity of abdominal pain, as measured on the Patient-Reported Outcome Scale in Acute Pancreatitis (PAN-PROMISE), was <5.


The primary outcome was the development of moderately severe or severe acute pancreatitis, according to the Revised Atlanta Classification:

Revised Atlanta Classification for severity of acute pancreatitis, 1 or more of:
(a) local complications
(b) exacerbation of a preexisting coexisting condition
(c) Cr >1.9 mg/dL
(d) SBP < 90 mmHg despite IVFs
(e) Pao2/Fio2 of ≤300

Secondary outcomes included organ failure and local complications occurring after randomization and during the hospitalizatio, length of hospital stay; ICU, use of nutritional support or invasive treatment, and the presence of SIRS.

The main safety outcome was fluid overload, which required two of the following (1) symptoms, (2) physical signs, or (3) imaging evidence of hypervolemia.


During an interim analysis it was found that aggressive fluid resuscitation was associated with a significantly higher incidence of fluid overload than moderate fluid resuscitation (20.5% vs. 6.3%; aRR 2.85; 95% CI, 1.36-5.94). For this reason the study was stopped early before reaching its enrollment goal.

By this early termination point a total of 249 patients were enrolled with 122 assigned to the aggressive-resuscitation group and 127 to the moderate-resuscitation group. Patient characteristics at baseline were similar except for more gallstone pancreatitis and SIRS more common in the aggressive fluid group

Patients in the aggressive-resuscitation group received a median of 7.8 liters of LR during the first 48 hours, while the moderate-resuscitation group received a median of 5.5 liters of LR.

There was no significant difference in the primary outcome, development of moderately severe or severe acute pancreatitis, occurring in 22.1% of the patients in the aggressive-resuscitation group and in 17.3% of those in the moderate-resuscitation group (aRR 1.30; 95% CI, 0.78-2.18; P=0.32).

For multiple secondary outcomes there was a trend towards harm in the aggressive-resuscitation group. Organ failure occurred in 7.4% in the aggressive-resuscitation group and 3.9% in the moderate-resuscitation group (aRR 1.23; 95% CI, 0.47-3.23). Necrotizing pancreatitis was similar in the the aggressive-resuscitation group compared to the moderate group, 13.9% vs, 7.1% (aRR 1.95; 95% CI, 0.87-4.38). ICU admission was statistically similar as well, 6.6% vs. 1.6% (aRR 2.71; 95% CI, 0.64-11.51)


This trial showed that aggressive fluid resuscitation increased the risk of volume overload in acute pancreatitis, even with patients excluded that might be at risk for overload (i.e. h/o CHF). Additionally, there was no difference in the primary outcome, development of moderately severe or severe pancreatitis.

Multiple secondary outcomes had a trend towards harm in the aggressive-resuscitation group, for example development of organ failure, necrotizing pancreatitis or ICU admission. We are left to speculate what the findings would be had the study be allowed to reach its pre-specified enrollment goal.

This study does have limitations, perhaps most importantly is its open label nature, but also how fluids could be adjusted in the moderate group but had to continue to 48 hr unchanged in the aggressive group, and that groups were unbalanced at baseline. These limitations, however, likely represent real world practice of aggressive fluids in the first 48 hrs. While it could be argued that gallstone pancreatitis carries different risks, it should not be inherently more prone to fluid overload.

As the existing evidence in support of aggressive fluids has been previously described as “paltry” [1], this study is the most robust available to date. That the harm was clear enough to stop the trial early and that there was no difference in the primary outcome should be sufficient to dispel the dogma of aggressive fluids in acute pancreatitis.

One could consider the moderate fluid strategy: to bolus 10 mL/kg only if hypovolemic and then order 1.5 mL/kg/hr continuous fluids until the patient is tolerating sufficient oral intake. Conversely, one could titrate fluids to the physical exam.

When and how to feed (enteral not necessary), how much IV fluid (less) and which type of IV fluid (maybe LR) are sufficiently answered with the current evidence. Keep in mind that about 20% of patients in either group still developed criteria for more severe pancreatitis and many still develop complications like necrotizing pancreatitis. As TB Gardner points out in the accompanying editorial, it is time to move past studying the supportive care aspects of acute pancreatitis and find more targeted therapies [3].

F: Follow upYes
R: RandomizationYes, stratified by center, presence or absence of SIRS, presence or absence of baseline hypovolemia
I: Intention to treatYes
S: Similar at baselineGallstone pancreatitis and SIRS more common in the aggressive fluid group
B: BlindingNo, open label
E: Equal treatmentYes
S: Source (funding)Instituto de Salud Carlos III and others

  1. Haydock MD, Mittal A, Wilms HR, Phillips A, Petrov MS, Windsor JA. Fluid therapy in acute pancreatitis: anybody’s guess. Ann Surg. 2013 Feb;257(2):182-8.
  2. Di Martino M, Van Laarhoven S, Ielpo B, et al.; Systematic review and meta-analysis of fluid therapy protocols in acute pancreatitis: type, rate and route. HPB (Oxford). 2021 Nov;23(11):1629-1638.
  3. Gardner TB. Fluid Resuscitation in Acute Pancreatitis – Going over the WATERFALL. N Engl J Med. 2022 Sep 15;387(11):1038-1039. doi: 10.1056/NEJMe2209132.