<p>BACKGROUND Intravenous fluid therapy is one of the most common interventions used in acute medicine. Worldwide, there are variations in the prescription of intravenous fluids and no consensus on the best intravenous fluid to use. Currently, 0.9% saline is the most frequently prescribed intravenous fluid; however, there is emerging evidence to suggest that 0.9% saline may be associated with an increased risk of acute kidney injury (AKI), increased blood transfusion requirements and gastrointestinal dysfunction when compared to a buffered crystalloid fluid, such as Plasma-Lyte 148. METHODS Study one: A prospective, multicentre, randomised, double-blind, cluster, double crossover study conducted over 28 weeks in four New Zealand intensive care units (ICU) comparing 0.9% saline to Plasma-Lyte 148. The primary outcome was the proportion of patients with either AKI or renal failure according to the RIFLE criteria definitions based on serum creatinine levels. Study two: An exploratory subgroup analysis of cardiac surgical patients enrolled in study one to compare the effects of 0.9% saline vs. Plasma-Lyte 148 on the need for blood transfusions. The primary outcome was the proportion of patients receiving blood or blood products in the ICU. Additionally, an in-depth single-centre nested cohort study comparing the effect of 0.9% saline vs. Plasma-Lyte 148 on blood loss and blood products was also conducted. Study three: A single-centre nested study within study one comparing the effect of 0.9% saline vs. Plasma-Lyte 148 on gastrointestinal dysfunction in patients expected to be mechanically ventilated for >48 hours and receiving enteral nutrition via the nasogastric route. The primary outcome was the proportion of patients with gastrointestinal intolerance (high gastric residual volumes (GRV), vomiting, and diarrhoea). RESULTS Study one: 2262 patients were enrolled and analysed, 1152 patients were allocated to receive Plasma-Lyte 148 and 1110 participants were allocated to receive 0.9% saline. In the Plasma-Lyte 148 group, 102 of 1067 patients (9.6%) developed AKI, compared with 94 of 1025 patient (9.2%) in the 0.9% saline group (RR 1.05; 95% CI, 0.78 to 1.41; p=0.76). Overall, 87 of 1152 patients (7.6%) in the Plasma-Lyte 148 group and 95 of 1110 patients (8.6%) in the 0.9% saline group died in hospital (RR, 0.88; 95% CI 0.67 to 1.17; p=0.40). Study two: 954 cardiac surgical patients were included, 475 patients were allocated to receive Plasma-Lyte 148 and 479 were allocated to receive 0.9% saline. 128 of 475 patients (26.9%) in the Plasma-Lyte 148 group received blood or a blood product compared with 94 of 479 (19.6%) patients in the 0.9% saline group (OR [95% CI], 1.51 [1.11-2.05]; p=0.008). Within the nested cohort (n=251, 131 assigned to Plasma-Lyte 148 and 120 assigned to 0.9% saline), there were no differences between groups in chest drain losses at 12 hours. Study three: 69 patients were enrolled and analysed, with 35 allocated to receive Plasma-Lyte 148 and 34 allocated to receive 0.9% saline. In the Plasma-Lyte 148 group, 21 of 35 patients (60.0%) developed gastrointestinal feeding intolerance, compared with 22 of 34 patients (64.7%) in the 0.9% saline group (OR, 0.82; 95% CI, 0.31–2.17; p=0.69). A high GRV was seen in 4 of 35 patients (11.4%) in the Plasma-Lyte 148 group, and in 11 of 34 patients (32.4%) in the 0.9% saline group (OR, 0.27; 95% CI, 0.08–0.96; p=0.04). CONCLUSION This research programme provides data from the first interventional study conducted in critically unwell patients on the comparative effectiveness of 0.9% saline compared to a buffered crystalloid solution for intravenous fluid therapy. Among patients receiving crystalloid fluid in the ICU, allocation to Plasma-Lyte 148 compared to 0.9% saline did not influence the risk of AKI. Within a subgroup of patients undergoing cardiac surgery allocation to Plasma-Lyte 148 was associated with an increased requirement for blood or blood products. Among mechanically ventilated patients receiving nasogastric feeding, the use of Plasma-Lyte 148 compared to 0.9% saline did not reduce the proportion of patients developing gastrointestinal feeding intolerance but was associated with a decreased incidence of high GRV. These findings suggest that Plasma-Lyte 148 as compared with 0.9% saline may have differing effects within the biological/organ systems and sub-groups of patients admitted to the ICU. Further large randomised clinical trials (RCT) are needed to assess the efficacy in higher-risk populations and to measure ongoing areas of uncertainty in clinically important outcomes such as mortality.</p>