Fluid Resuscitation in Children—Better to be “Normal” or “Balanced”?

Intravenous fluid resuscitation was first described by Thomas Latta, who in a letter to the Lancet published on June 2, 1832, noted that “improvement in the pulse and countenance is almost simultaneous, the cadaverous expression gradually gives place to appearances of returning animation, the livid hue disappears, the warmth of the body returns” (1). The exact composition of the fluid infused by Dr. Latta remains unknown, but fluids of various composition were subsequently described. Use of 0.9% saline emerged in the 1880’s when Hartog Joakob Hamburger recognized that erythrocytes did not lyse when placed in this fluid and concluded that “the blood of man was isotonic with a NaCl solution of 0.9%” (2). Although these observations conceivably led to the “normal” moniker for 0.9% saline, human plasma is actually closer to 0.6% NaCl. Lactated Ringer’s (LR) was also born in the 1880’s when Sydney Ringer observed that addition of calcium and potassium to saline preserved the rhythmicity of frog heart muscle ex vivo better than saline alone (2). In 1932, the pediatrician Alexis Hartmann modified Ringer’s original formula in order to reduce the acidosis observed with diarrheal dehydration (2). More recently, Plasma-Lyte was developed to address the slight hypotonicity and calcium-induced incompatibility with other medications in LR and Hartmann’s solutions.

Although crystalloids have been commonly used for nearly 200 years, there remains a lack of clear, evidenced-based support for the most effective and safest fluid type. Although historical preference has been to use 0.9% saline for fluid bolus therapy (2-6), saline contains a supra-physiologic concentration of chloride and low strong ion difference (SID) that contributes to hyperchloremic metabolic acidosis, decreases renal blood flow, and promotes inflammation (7). In contrast, more balanced/buffered crystalloids (LR, Hartmann’s, PlasmaLyte) have electrolyte compositions that better approximate human plasma, including a higher SID due to an anion buffer. Observational studies in children have associated use of balanced/buffered fluids with decreases in acidemia, acute kidney injury (AKI), and death compared to 0.9% saline (8-10). Interventional studies, largely focused on adults, have demonstrated either a small benefit for buffered/balanced fluids or no difference between fluid types (11-13). However, well-powered interventional studies are lacking in children.

In this issue of Pediatric Critical Care Medicine, Dr. Lehr and colleagues present the largest systemic review and meta-analysis of the effects of balanced/buffered fluid bolus therapy in critically ill children (14). Due to a low rate of adverse outcomes, the authors chose as a primary endpoint the change in serum bicarbonate or blood pH, along with several secondary endpoints. A total of 13 studies with 11,848 patients ages 28 days to 18 years treated with ≥20 mL/kg fluid bolus therapy were included. Nine of the studies were randomized controlled trials (RCTs), though only three (162 total patients) were included in the meta-analysis of the primary endpoint. Most studies focused on septic/dengue shock or diarrheal dehydration, but two studied diabetic ketoacidosis. LR was the most common balanced/buffered fluid.

The meta-analysis of three RCTs—all in children with diarrheal dehydration in which stool bicarbonate loss is a prominent component of the acidemia—found a small increase in serum bicarbonate (1.60, 95% CI 0.04, 3.16 mmol/L) and blood pH (0.03, 95% CI 0.00, 0.06) with balanced/buffered fluids compared to 0.9% saline. Additional analyses for hyperchloremia, AKI, renal replacement therapy, vasopressors, mechanical ventilation, length of stay, and mortality did not reveal differences between fluid types.

Although well-conducted, the analyses by Dr. Lehr and colleagues were unable to provide clear, evidenced-based support for the most effective and safest crystalloid fluid type for pediatric fluid bolus therapy. It is unlikely that the small increase in serum bicarbonate/blood pH with balanced/buffered fluid is clinically relevant. One possible explanation for their findings is that there is not a true difference in patient outcomes between 0.9% saline and balanced/buffered fluids. With this interpretation, the historical preference to use saline is justified as this fluid is generally the cheapest, most readily available, and best compatible with other medications.

Alternatively, the inability to delineate an important differential impact between fluid types may reflect a type II error due to a combination of low sample size, selection bias, small volume of fluid administration, or wrong outcomes. Several studies suggest that evidence to preferentially use balanced/buffered fluids, rather than 0.9% saline, is waiting to be found. First, hyperchloremia ≥110 mEq/L is associated with AKI, organ dysfunction, and mortality in children with septic shock (15, 16). Although hyperchloremia was not significantly different between fluid types in the report by Lehr et al, other studies have demonstrated that chloride levels rise higher and faster with 0.9% saline (9, 15, 17). Second, hyperchloremic metabolic acidosis partially explained the excess mortality induced by unbuffered fluids (including 0.9% saline) in the FEAST trial of febrile children (18). Third, the largest observational study of children with sepsis demonstrated lower length of stay, vasoactive infusion days, AKI, and mortality when only balanced fluids were used (8). Finally, interventional studies in adults have found a lower incidence of major adverse kidney events with balanced/buffered fluids, especially among adults with sepsis (11, 12).

Notably, the Surviving Sepsis Campaign recently suggested use of balanced/buffered crystalloids, rather than 0.9% saline, for the initial resuscitation of children with septic shock (19). This weak recommendation was based on “very low quality of evidence” that the “desirable consequences of balanced/buffered crystalloids probably outweigh the undesirable consequences (including cost), especially in those who require large volume fluid resuscitation.” However, in acknowledgement of ongoing uncertainty, the Campaign also called for a clinical trial to establish clear, high-quality evidence. Such a trial, the PRagMatic Pediatric Trial of Balanced vs nOrmaL Saline FlUid in Sepsis (PRoMPT BOLUS, www.clinicaltrials.gov/NCT04102371), is now actively enrolling with a target of 8,800 children with suspected septic shock from >40 sites in the United States, Canada, Australia, and New Zealand (20). PRoMPT BOLUS should provide the missing high-quality evidence for whether resuscitation with balanced/buffered fluids can safely improve outcomes compared to 0.9% saline. After all, with nearly 200 years of fluid administration, an answer is due.