Kidney Week

Abstract: FR-PO048

Chloride Abnormalities Are Independently Associated with Mortality in Critically Ill Children

Session Information

• AKI: Clinical, Outcomes, Trials - I
  October 26, 2018 | Location: Exhibit Hall, San Diego Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Acute Kidney Injury

• 102 AKI: Clinical, Outcomes, and Trials

Authors

• Barhight, Matthew, Children's Hospital Colorado, Aurora, Colorado, United States

Matthew Barhight,

• Brinton, John T., University of Colorado Denver School of Medicine, Aurora, Colorado, United States

John T. Brinton,

• Soranno, Danielle, Children's Hospital Colorado, Aurora, Colorado, United States

Danielle Soranno,

• Faubel, Sarah, University of Colorado Denver, Denver, Colorado, United States

Sarah Faubel,

• Griffin, Benjamin, University of Colorado, Aurora, Colorado, United States

Benjamin Griffin,

• Goebel, Jens W., Children's Hospital Colorado, Aurora, Colorado, United States

Jens W. Goebel,

• Mourani, Peter, University of Colorado School of Medicine, Aurora, Colorado, United States

Peter Mourani,

• Gist, Katja M., University of Colorado, Children''s Hospital Colorado, Aurora, Colorado, United States

Katja M. Gist,

Background

Chloride disturbances in critically ill adults are associated with mortality. The impact of chloride disturbances on outcomes in critically ill children is unknown. The purpose of this study was to determine if there is an association between mortality and 1) admission chloride levels and 2) changes in the chloride level following initial resuscitation in critically ill children.

Methods

We retrospectively studied all PICU patients (01/2014 – 12/2015) at Children’s Hospital Colorado, excluding those 1) aged < 90 days or > 25 years, 2) without admission laboratory tests, 3) end stage renal disease, 4) a disorder of chloride transport and 5) admitted for diabetic ketoacidosis. Patients were stratified based on admission chloride levels (hypochloremia, <100 mEq/L; normochloremia, 100-109 mEq/L and hyperchloremia ≥110 mEq/L). Additionally, they were dichotomized based on the change in chloride (ΔCl) in the first 24 hours (< 5 mEq/L, ≥ 5 mEq/L). Multivariate logistic regression and analysis of covariance were performed to determine the impact of chloride derangements on mortality and length of stay respectively.

Results

2022 patients were included; overall mortality was 4% (n = 77) and day 2 AKI occurred in 18% (n = 318). Hypochloremia, hyperchloremia, and ΔCl ≥ 5 mEq/L occurred in 7%, 21%, and 12% respectively. Hypochloremia and ΔCl ≥ 5 mEq/L were independently associated with a 4.57 (95% CI: 1.65 - 12.66) and a 3.17 (95% CI: 1.61 - 6.26) greater odds of mortality respectively, after adjusting for confounders. In ANACOVA models, hypochloremia and hyperchloremia were associated with an estimated increase in the expected length of hospital stay of 1.2 days (95% CI: 1.0 - 1.5; p=0.03), and 1.2 days (95% CI: 1.0 - 1.4 p=0.02) respectively. ΔCl ≥ 5 mEq/L was associated with an estimated increase in the expected time on mechanical ventilation of 1.3 days (95% CI: 1.0 - 1.4; p = 0.02).

Conclusion

Hypochloremia and ΔCl ≥ 5 mEq/L in the first 24 hours of admission are common and independent risk factors for mortality in critically ill children after adjusting for confounders. Further studies are needed to address the mechanism by which chloride disturbances increase mortality in critically ill children.