AUTHOR=Bolado Federico , Buxbaum James L. , Vaillo-Rocamora Alicia , Cárdenas-Jaén Karina , Maisonneuve Patrick , de-Madaria Enrique 

TITLE=Early Weight-Based Aggressive vs. Non-Aggressive Goal-Directed Fluid Resuscitation in the Early Phase of Acute Pancreatitis: An Open-Label Multicenter Randomized Controlled Trial (The WATERFALL Trial), Design, and Rationale

JOURNAL=Frontiers in Medicine

VOLUME=Volume 7 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2020.00440

DOI=10.3389/fmed.2020.00440

ISSN=2296-858X

ABSTRACT=Treatment options are limited for acute pancreatitis (AP). Early aggressive fluid resuscitation (AFR) has been widely considered beneficial due to theoretical improvement in end-organ perfusion, including the pancreas and gut, with pancreatic necrosis and bacterial traslocation as consequences of ischemia. There is scarce direct evidence for its association to improved outcomes. Furthermore, it has been described that AFR may be associated to poor outcomes in severe AP. 
WATERFALL is an investigator-initiated international multicenter open-label randomized controlled trial comparing AFR versus moderate fluid resuscitation (MFR) in acute pancreatitis. The main outcome variable will be the incidence of moderate-to-severe AP (a clinically relevant outcome that has been validated). AFR will consist in Lactated Ringer Solution (LR) 20 ml/kg bolus (administered over 2 hours) followed by LR 3 ml/kg/h. Patients randomized to MFR will receive a LR bolus 10 ml/kg in case of hypovolemia or no bolus in patients with normal volemia, followed by LR 1.5 ml/kg/h. The patients will be assessed at 3 (±1), 12 (±4), 24 (±4), 48 (±4) and 72 (±4) hours from recruitment, and fluid resuscitation will be adjusted to the patient´s clinical and analytical status according to a protocol. 
Based on a prospective multicenter study, the incidence of moderate to severe AP is 35%. Sample sizes of 372 patients per group (overall 744) achieve 80% power to detect a difference in the incidence of moderate to severe AP of 10%, at a significance level (alpha) of 0.05 using a two-sided z-testNA, assuming a 10% drop-out rate. These results assume that 3 sequential tests are made using the O'Brien-Fleming spending function to determine the test boundaries.