High Chloride Levels in the Blood Linked to Higher Death and Kidney Injury Risk in Critically Ill Children

Chloride is the most abundant anion in human blood plasma, a quiet workhorse of physiology that helps maintain acid-base balance, osmotic pressure, and the electrical gradients that keep cells functioning. Yet when its concentration climbs too high, a condition known as hyperchloremia, the consequences in the sickest patients may be far more serious than clinicians have traditionally assumed. A new systematic review and meta-analysis published in BMC Pediatrics by researchers at the First Affiliated Hospital of Xinjiang Medical University in Urumqi has now pulled together the available observational evidence on hyperchloremia in critically ill children, and its findings suggest that elevated serum chloride may be associated with roughly a doubling of the odds of short-term death and acute kidney injury in pediatric intensive care units. The work, registered prospectively as PROSPERO CRD420261400801, is among the first attempts to quantify this relationship specifically in children, a population in which fluid therapy and electrolyte management remain areas of intense debate.

The research team, led by Xianglan Shi and corresponding author Adilijiang Kari, systematically searched PubMed, Embase, and Web of Science from database inception through May 31, 2026. Their goal was to identify studies that enrolled critically ill patients aged zero to eighteen years admitted to a pediatric intensive care unit, or PICU, and that reported adjusted associations between hyperchloremia and either short-term mortality or acute kidney injury. The insistence on adjusted estimates is technically important: raw associations between an electrolyte abnormality and death are notoriously confounded, because children who are sicker to begin with are more likely to develop electrolyte disturbances in the first place. By restricting the quantitative synthesis to effect estimates that had been statistically corrected for illness severity and other covariates, the authors aimed to isolate a signal that could plausibly be attributed to chloride itself, or at least to something traveling alongside it.

After screening, the analysis settled on seven reports representing six independent observational cohorts and a combined 22,996 unique participants. That number sounds impressive, but the heterogeneity beneath it is considerable, and the authors were unusually candid about it. Hyperchloremia was defined using study-specific thresholds, including values greater than 109, at least 110, or greater than 110 millimoles per liter, reflecting the absence of any universally accepted pediatric cutoff. The chloride exposure itself was categorized in three distinct ways: hyperchloremia present at admission, the peak chloride concentration reached at any point during the PICU stay, or chloride measures averaged or aggregated across the ICU period. Outcomes were similarly variable, with short-term mortality encompassing PICU, hospital, or 28-day death, and acute kidney injury ascertained either through study-reported KDIGO criteria or through administrative ICD-9 and ICD-10 diagnostic codes.

The headline result concerns mortality. Because the studies contributing admission- and peak-chloride mortality estimates were not sufficiently comparable for quantitative pooling, the authors synthesized those findings narratively. However, an exploratory synthesis of two estimates evaluating mean chloride during the first seven PICU days produced a striking result: children with hyperchloremia by this definition had significantly higher odds of short-term mortality, with a pooled adjusted odds ratio of 2.34 and a 95 percent confidence interval of 1.57 to 3.48. The statistical heterogeneity across the pooled estimates was zero, as measured by the I-squared statistic, meaning the two studies agreed closely. The authors nonetheless graded the certainty of this evidence as low, a judgment that reflects the observational design of the underlying cohorts and the small number of studies feeding the pooled estimate.

For acute kidney injury, the pattern was similar. Single-study estimates for initial and maximum chloride were reported narratively because pooling was not feasible, but an exploratory synthesis of three ICU-period chloride estimates showed that hyperchloremia was associated with higher odds of AKI, with a pooled adjusted odds ratio of 2.10 and a 95 percent confidence interval of 1.58 to 2.79, again with zero heterogeneity. Here, however, the certainty of evidence was rated very low, one step below the mortality finding. The distinction matters because GRADE-style certainty ratings incorporate not only risk of bias and imprecision but also indirectness and inconsistency, and the AKI estimates drew on studies whose exposure windows and outcome definitions diverged more substantially. The reported prevalence of hyperchloremia across the included cohorts ranged dramatically, from 6.8 percent to 51.4 percent, depending on the population studied and the definition of the chloride exposure applied.

Two secondary outcomes received less encouraging verdicts. Evidence linking hyperchloremia to the need for renal replacement therapy was rated very low certainty and was inconsistent across studies, offering no clear directional signal. Similarly, very-low-certainty evidence from a single study did not demonstrate a clear association between hyperchloremia and the duration of mechanical ventilation. These null or murky findings are a useful corrective to over-interpretation of the mortality and AKI results: the chloride signal, whatever its biological reality, does not extend uniformly to every clinically relevant outcome in the PICU, and the evidence base is simply too thin to support confident statements about RRT or ventilator days.

The biological plausibility of a chloride effect is worth taking seriously, even if this review cannot establish causation. In adults, a substantial literature has associated hyperchloremia, particularly hyperchloremia induced by large-volume resuscitation with normal saline, which contains a chloride concentration of 154 millimoles per liter compared with roughly 103 in plasma, with acute kidney injury, vasoconstriction of the renal afferent arteriole, and increased mortality. Proposed mechanisms include chloride-mediated renal vasoconstriction reducing glomerular perfusion, hyperchloremic metabolic acidosis, and inflammatory and coagulation disturbances. Children, however, are not small adults: their fluid requirements, renal maturation, and resuscitation practices differ, and pediatric-specific evidence has lagged far behind. The new analysis begins to close that gap, suggesting that the adult signal may have a pediatric counterpart, though the observational nature of the data leaves open the possibility that hyperchloremia is a marker of disease severity, greater fluid burden, or use of chloride-rich resuscitation fluids rather than a harmful agent in its own right.

The authors are explicit that these findings do not establish causality and should be interpreted cautiously. Few cohorts contributed to each pooled estimate, chloride-exposure and outcome definitions differed across studies, and residual confounding along with uncertainty about exposure-outcome timing remain serious concerns. A child whose mean chloride over the first week in the PICU is elevated may simply be a child who received more fluid, endured longer shock, or suffered more renal stress, and no amount of statistical adjustment in small observational cohorts can fully untangle those threads. The temporal question is particularly thorny: did hyperchloremia precede and contribute to kidney injury, or did emerging kidney injury cause chloride to rise? Without serial, protocolized chloride measurements aligned with protocolized creatinine and urine-output assessments, the direction of causality cannot be resolved from the existing literature.

What the review does establish is a research agenda. The authors call for prospective studies with standardized serial chloride assessment to clarify these associations, and, more ambitiously, for adequately designed randomized trials to evaluate the clinical effects of fluid strategies that differ in chloride load. Such trials already exist in adult medicine, most famously the comparisons of balanced crystalloids against saline in large emergency and critical care populations, but pediatric randomized evidence remains sparse. Until such trials are done, the practical takeaway for clinicians is one of heightened vigilance rather than changed practice: serum chloride, often an afterthought on the chemistry panel, may deserve closer attention in PICU patients, particularly when it persists or climbs over the first week of intensive care.

For a condition affecting somewhere between roughly one in fifteen and one in two critically ill children depending on how it is defined, hyperchloremia has long flown under the radar of pediatric critical care research. This meta-analysis, modest in its evidence base but rigorous in its transparency about limitations, elevates the question to where it belongs: as a testable, potentially modifiable risk factor with odds ratios that, if even partially causal, would translate into meaningful numbers of preventable deaths and kidney injuries among the thousands of children admitted to intensive care units worldwide each year. The certainty of the evidence is low to very low, the authors emphasize, but the direction of every well-adjusted estimate points the same way, and the next generation of prospective pediatric studies now has a clear benchmark to beat.

Subject of Research: The association between hyperchloremia and short-term mortality and acute kidney injury in critically ill children in the pediatric intensive care unit

Article Title: Association of hyperchloremia with mortality and acute kidney injury in critically ill children: a systematic review and meta-analysis

Article References: Shi, X., Zhu, H., Yahefu, A., & Kari, A. (2026). Association of hyperchloremia with mortality and acute kidney injury in critically ill children: a systematic review and meta-analysis. BMC Pediatrics. https://doi.org/10.1186/s12887-026-07540-9

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07540-9

Keywords: hyperchloremia, chloride, critically ill children, pediatric intensive care unit, acute kidney injury, mortality, meta-analysis, systematic review, fluid therapy, renal replacement therapy, KDIGO criteria, residual confounding