Balanced Fluids or Saline? Major Review Weighs the Best Rescue Liquid for Children in Septic Shock

When a child arrives in the emergency department in septic shock, one of the first and most consequential decisions clinicians make is deceptively simple: which bag of intravenous fluid to hang. For decades, the default has been normal saline, a solution of sodium chloride in water that is cheap, universally available, and stored in virtually every hospital on Earth. Yet saline is not as neutral as its name suggests. Its chloride concentration is far higher than that of human plasma, and a growing body of research in adults has suggested that this so-called unbalanced composition may contribute to metabolic derangements and kidney stress. A new systematic review and meta-analysis published in BMC Pediatrics has now brought the sharpest available evidence to bear on this question in children, pooling randomized trial data on more than nine thousand pediatric patients to ask whether balanced crystalloids genuinely outperform saline when it matters most.

The study, conducted by a team of physicians and researchers affiliated with medical institutions in Pakistan and Afghanistan, set out to compare the efficacy and safety of balanced crystalloids against normal saline in pediatric patients with septic shock or suspected septic shock who required fluid resuscitation. Balanced crystalloids, which include solutions such as Ringer’s lactate and Plasma-Lyte, are formulated to more closely mimic the electrolyte profile of plasma, containing lower chloride concentrations and physiologic buffers. The researchers performed a systematic search of PubMed, Embase, Scopus, and ClinicalTrials.gov covering the entire period from database inception to May 17, 2026, and included only randomized controlled trials, the most rigorous form of clinical evidence available. Their primary outcomes were in-hospital mortality and acute kidney injury, while secondary outcomes spanned 90-day mortality, the need for renal replacement therapy, major adverse kidney events at 30 days, lengths of hospital and pediatric intensive care unit stay, mechanical ventilation, and electrolyte abnormalities.

After screening the literature, the team identified five randomized controlled trials encompassing 9,343 pediatric patients. That scale is notable, because pediatric septic shock trials are notoriously difficult to run: the condition is relatively uncommon compared with adult sepsis, enrollment windows are short, and consent must be obtained from distressed families in moments of acute crisis. Pooling data across trials with a random-effects meta-analysis allowed the investigators to extract statistical power that no single trial could achieve on its own, and they applied the GRADE framework to formally assess how much confidence clinicians and families should place in each finding.

The headline result is, on its face, a null one. Balanced crystalloids showed no significant difference compared with normal saline in in-hospital mortality, with a relative risk of 1.00 and a 95 percent confidence interval of 0.82 to 1.22, a range that comfortably spans the possibility of no effect in either direction. Acute kidney injury, defined by conventional criteria, likewise showed no statistically significant difference, with a relative risk of 0.94 and a wide confidence interval stretching from 0.52 to 1.70. The same pattern of no significant difference held for 90-day mortality, mechanical ventilation, major adverse kidney events at 30 days, and both hospital and pediatric intensive care unit lengths of stay. For a reader scanning only these numbers, the conclusion might seem to be that the two fluids are interchangeable.

But the authors are careful, and rightly so, to warn against that interpretation. The absence of a demonstrated difference is not the same as evidence of equivalence. Meta-analyses can only be as definitive as the trials they pool, and the confidence intervals around several outcomes were wide enough that a clinically important benefit or harm of either fluid cannot be excluded. This distinction matters enormously in clinical practice. A relative risk of 0.67 for renal replacement therapy, for example, would represent a one-third reduction in the need for dialysis or related therapies if real, yet the primary analysis in this review did not reach conventional statistical significance, with a confidence interval of 0.37 to 1.23. The finding is a signal, not a verdict.

Where the evidence did crystallize was in the electrolyte domain. Balanced crystalloids significantly reduced hyperchloremia, the elevation of serum chloride that results from infusing large volumes of high-chloride saline, with a relative risk of 0.65 and a tight confidence interval of 0.61 to 0.70. This is the most biologically predictable finding in the review, and its statistical robustness is reassuring. Hyperchloremia is not merely a laboratory curiosity; a strong physiological argument holds that chloride loading causes renal vasoconstriction, reduces kidney blood flow, and promotes a hyperchloremic metabolic acidosis that may confound clinical assessment in an already unstable patient. The review also found a consistent numerical reduction in hypernatremia with balanced crystalloids, with a relative risk of 0.59, although the primary Hartung-Knapp analysis did not reach statistical significance, with a confidence interval of 0.30 to 1.17. Notably, a sensitivity analysis using the DerSimonian-Laird method for the renal replacement therapy outcome produced the same point estimate as the primary analysis, with a confidence interval of 0.48 to 0.92 that crossed into significance, a concordance that strengthens the plausibility of a genuine kidney-protective effect even if it remains unproven.

The methodological choices underlying these analyses deserve attention from anyone seeking to understand why the same data can yield different conclusions. The Hartung-Knapp approach, used in the primary analysis, is generally regarded as more conservative because it accounts for uncertainty in the between-study variance, producing wider confidence intervals that are less likely to yield false-positive findings when only a handful of trials are available. With five trials, small-study effects and heterogeneity are real concerns, and running both primary and sensitivity analyses is a standard safeguard. The fact that the renal replacement therapy signal behaved consistently across both methods, while only crossing the significance threshold under one, illustrates the honest uncertainty at the heart of this field: the direction of effect favors balanced crystalloids, the magnitude could be clinically meaningful, but the evidence is not yet definitive.

So what should clinicians take away? The authors conclude that, in the absence of any demonstrated clinical advantage of normal saline, the totality of available physiological and clinical evidence supports a conditional preference for balanced crystalloids in fluid resuscitation of pediatric septic shock or suspected septic shock. The word conditional is doing real work here. The recommendation acknowledges that specific clinical circumstances may still favor saline, such as situations involving particular electrolyte disturbances, intracranial pathology, or liver failure, where the composition of balanced solutions could pose theoretical concerns. It also reflects the reality that balanced crystalloids remain less available and more expensive in many of the low- and middle-income settings where pediatric sepsis takes its greatest toll, a practical constraint that no meta-analysis can dissolve.

The broader significance of this work extends beyond the pediatric intensive care unit. Adult trials such as the SMART and SALT-ED studies have already shifted much of adult critical care toward balanced solutions, and pediatric practice has been waiting for evidence of comparable quality in children. This review provides the most comprehensive synthesis to date, but it also exposes the gap: five trials, however large in aggregate, are still too few to resolve effects on mortality or kidney injury with the precision that guideline writers and clinicians would want. The authors call explicitly for adequately powered randomized controlled trials to clarify the effects of fluid composition on mortality, acute kidney injury, renal replacement therapy, and other outcomes that matter to patients and families. Until those trials arrive, the choice of fluid in a crashing child will remain a judgment call, informed but not settled by evidence, and this review gives that judgment a firmer footing: no proven mortality benefit, a clear reduction in hyperchloremia, and suggestive signals on kidney protection that tilt the scales, gently but perceptibly, toward the balanced bag.

Subject of Research: Comparison of balanced crystalloids and normal saline for fluid resuscitation in pediatric septic shock

Article Title: Balanced crystalloids versus normal saline in pediatric septic shock or suspected septic shock requiring fluid resuscitation: a systematic review, and meta-analysis

Article References: Balanced crystalloids versus normal saline in pediatric septic shock or suspected septic shock requiring fluid resuscitation: a systematic review, and meta-analysis. (n.d.). https://doi.org/10.1186/s12887-026-07833-z

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07833-z

Keywords: pediatric septic shock, balanced crystalloids, normal saline, fluid resuscitation, meta-analysis, randomized controlled trials, acute kidney injury, renal replacement therapy, hyperchloremia, hypernatremia, pediatric intensive care, GRADE assessment