Comprehensive meta-analysis of randomized controlled trials of Lactated Ringer’s versus Normal Saline for acute pancreatitis

Introduction Fluid resuscitation is the keystone of treatment for acute pancreatitis. Though clinical guidelines and expert opinions agree on large volume resuscitation, debate remains on the optimal fluid type. The most commonly used fluids are Lactated Ringer's (LR) and Normal Saline (NS), but the studies published to date comparing LR vs NS yield conflicting results. We aimed to identify and quantitatively synthesize existing high quality data of the topic of fluid type or acute pancreatitis resuscitation. Methods In collaboration with the study team, an information specialist performed a comprehensive literature review to identify reports addressing type of fluid resuscitation. Studies were screened using the Covidence system by two independent reviewers in order to identify Randomized controlled trials comparing LR versus NS. The main outcome was the development of moderately severe or severe pancreatitis and additional outcomes included local complications, ICU admission, and length of stay. Pooled odds ratios were estimated using the random effects model and standardized mean difference to compare continuous variables. Results We reviewed 7964 abstracts and 57 full text documents. Four randomized controlled trials were identified and included in our meta-analyses. There were a total of 122 patients resuscitated with LR versus 126 with NS. Patients resuscitated with LR were less likely to develop moderately severe/severe pancreatitis (OR 0.49; 95 % CI 0.25–0.97). There was no difference in development of SIRS at 24 or 48 h or development of organ failure between the two groups. Patients resuscitated with LR were less likely to require ICU admission (OR 0.33; 95 % CI 0.13–0.81) and local complications (OR 0.42; 95 % CI 0.2–0.88). While there was a trend towards shorter hospitalizations for LR (SMD -0.18, 99 % CI -0.44-0.07), it was not statistically significant. Conclusion Resuscitation with LR reduces the development of moderately severe-severe pancreatitis relative to NS. Nevertheless, no difference in SIRS development or organ failure underscores the need for further studies to verify this finding and define its mechanism.