Early fluid resuscitation improves outcomes in acute pancreatitis. Lactated Ringer’s (LR) may be a better fluid choice than normal saline (NS).

A retrospective cohort study¹ of 434 patients presenting with acute pancreatitis to a single tertiary medical center over 24 years looked at the optimal timing of IVF administration. Included were patients who presented directly to the medical center and whose primary admission diagnosis was acute pancreatitis. Patients were excluded if the IVF they received was incompletely documented. Using nursing administration documentation, patients were divided into “early” (receiving >1/3 of the total 72-hour fluid volume within the first 24 hours, 340 patients) and “late” (receiving <1/3 of the total 72-hour fluid volume within the first 24 hours, 94 patients) resuscitation groups. Groups
were well-balanced other than the early group having a lower proportion of patients who had post-ERCP (endoscopic retrograde cholangiopancreatography) acute pancreatitis (5% vs. 12%; P = .03) and were on TPN (26% vs. 46%; P = .01). Main outcomes included systemic inflammatory response syndrome (SIRS) or organ failure rates at 24, 48, and 72 hours, ICU transfers, duration of hospitalization, and mortality.

SIRS rates were comparable on admission but subsequently significantly lower in the early resuscitation group at all time points (24 hour: 15% vs. 32%, P = .001; 48 hour: 14% vs. 33%, P = .001; 72 hour: 10% vs. 23%, P = .01). Organ failure rates trended toward a difference only at 72 hours (5% vs. 10%; P = .05). Each of these differences persisted on multivariate adjustment. Length of stay (8 vs. 11 days; P = .01) and ICU transfer rates (6% vs. 17%; P = .001) were also lower in the early resuscitation group. Mortality rates did not differ (3% vs. 4%; P = .70). While the early resuscitation group received more IVF than the late group in the first 24 hours (mean 3.4 vs. 2.4 L; P = .001), they received less total fluid in the first 72 hours (mean 7.6 vs. 9.5 L; P = .003). Study caveats include confounding by indication, the lack of adjustment for secular trends, and the possibility of persistent unmeasured confounders.

A randomized controlled trial at three tertiary centers² compared goal-directed fluid resuscitation to standard therapy, in addition to use of LR versus NS, in 40 acute pancreatitis patients. Patients identified within 6 hours of initial presentation were included. Exclusion criteria were renal, cardiac, or liver disease, recent cardiovascular procedure or active chest pain, chronic obstructive pulmonary disease on home oxygen, active sepsis, electrolyte derangements, inflammatory diseases, cancer, or initial presentation to an outside hospital.

Patients were randomized to one of four treatment arms: goal-directed resuscitation with LR, goal-directed resuscitation with NS, standard resuscitation with LR, or standard resuscitation with NS. Goal-directed resuscitation consisted of following an algorithm of initial bolus and infusion with clinical reevaluation and blood urea nitrogen checks every 8 hours to assess fluid responsiveness, while standard resuscitation consisted of IVF given at the discretion of the treating physician. The primary outcome was presence of SIRS at 24 hours, while the secondary outcomes included serum CRP at 24 hours.

SIRS prevalence was lower than expected across all groups, leading to an underpowered primary outcome. Authors found that the goal-directed and standard resuscitation groups received similar amounts of fluid and that these strategies were not associated with any differences in SIRS or CRP at 24 hours. As compared to NS, use of LR led to lower rates of SIRS (84% vs. 0% relative reduction; P = .035) and CRP levels (51.5 vs. 104 mg/L; P = .02) at 24 hours.

A subsequent trial³ showed improvement in various inflammatory markers with LR over NS without any notable side effects. Both of these studies are limited by stringent exclusion criteria, small sample sizes, and surrogate outcomes. The 2013 ACG guidelines⁴ recommend early aggressive fluid resuscitation (strong recommendation, moderate quality of evidence), preferably with LR over NS (conditional recommendation, moderate quality of evidence), for patients without significant comorbidities such as renal or cardiac disease. The 2018 AGA guidelines⁵ suggest goal-directed resuscitation (conditional recommendation, very low quality of evidence) and make no recommendation about choice of crystalloid.