Saline vs. Lactated Ringers for Emergency Department IV Fluid Resuscitation

This is a single-center randomized, single-blinded, controlled clinical trial testing superiority of LR compared to NS in patients receiving IV fluid before discharge.

The study site is the University of Iowa Emergency Department (annual patient volume is 60,000 patients/year).

The study population will include adult ED patients presenting with chief complaints associated with volume depletion. Figure 1 outlines the enrollment flowchart for the study with inclusion and exclusion criteria. Research assistants (RAs) are available in the ED from 9am - 10pm every day, and an automated paging system will alert the team when patients qualify, as the study team has done previously. All participants will receive 2 liters of blinded IV fluid as part of the study. The intervention group will receive lactated ringer's, and the control group will receive normal saline (0.9% sodium chloride). Nursing staff will establish the peripheral IV access, and subjects will be blinded to group.

The primary outcome, Quality of Recovery-40, will be analyzed using a t-test (if parametric) or Mann Whitney U Test (non-parametric) in an intention-to-treat (ITT) analysis. Normality will be assessed with a Kolmogorov-Smirnov (K-S) and visualization of data. Due to the nature of time-critical emergency department research, an additional analysis will be done by treatment received and compared to the primary ITT analysis as a sensitivity analysis. Secondary outcomes will be reported as a relative risk with 95%CI in an ITT analysis. A safety analysis of secondary outcomes will be performed by treatment received (see page 37). Baseline characteristics will be reported with summary statistics as appropriate (mean, SD, median, IQR) and differences in characteristics will be conducted using a t-test and chi-square test, as appropriate. The following variables will also be assessed: actual fluid administered(mL), pre-intervention QoR-40 score, and time from intervention to primary outcome. If differences in these characteristics exist between the control and intervention groups, multivariate linear regression and multivariate logistic regression (primary and secondary outcomes, respectively) will be used.

Inclusion criteria

• Emergency department patient
• Chief complaint of nausea, vomiting or emesis, diarrhea, abdominal pain, dehydration, heat stroke or exhaustion
• ED provider approves administration of two liters of fluid
• ED provider states patient is likely to be discharged
• Patient has followed up with PCP in the last two years
• Patient has access to phone for next two days following ED discharge

Exclusion criteria

• Prisoners
• Children
• Women known to be pregnant
• Jaundice
• Current chemotherapy