UMHS LOGOUniversity of Michigan
Department of Pediatrics

Evidence-Based Pediatrics Web Site

Aggressive Fluid Resuscitation in Initial Treatment of Children Presenting in DKA May Cause Herniation


  • Does giving an IV fluid bolus to pediatric patients for the initial treatment of DKA put the patient at greater risk of developing cerebral edema/brain herniation?

Clinical Bottom Lines

  1. A single-center retrospective case-control study identified degree of acidosis (pH < 7.1) and degree of hypocapnea (PCO2 < 20 mm Hg) on presentation as the most reliable predictors of brain herniation during DKA tx in children.
  2. The most important treatment variable associated with brain herniation is the administration of > 50 cc/kg IVF over first 4 hrs tx. The relative odds of brain herniation in this series was 19.5, and the number needed to harm was 6 (for pts receiving > 50 cc/kg IVF vs < 50 cc/kg IVF over 1st four hrs of therapy).
  3. Of children admitted with severe DKA, those receiving > 50 cc/kg IVF over 1st four hrs were 19.5 times more likely to develop brain herniation. For every 6 children treated with this aggressive rate of hydration, one suffered brain herniation.
  4. Giving one 20cc/kg bolus, followed by standard protocol re-hydration does not place the patient in the "danger zone" identified by this study. Giving a second 20cc/kg bolus followed by standard re-hydration can place them into this zone.

Summary of Key Evidence

  1. Retrospective case-control study, reviewing charts from a 12-year period, from a single center in Seattle, WA, of patients aged 19 or less admitted with diagnosis of DKA. Numerous variables were evaluated to determine risk factors for brain herniation. Patients were divided into groups of "affected" and "non-affected" (with brain herniation). For comparison of lab values and clinical findings, affected patients were paired with unaffected controls, matched by age, race, and year of admission.
  2. Of the 153 children admitted for DKA during this time period, 119 of them suffered "severe DKA", as defined by pH < 7.1 or PCO2 < 20. None of the 34 children who maintained a pH > 7.1 and PCO2 > 20 suffered cerebral complications.
  3. Of the 119 children admitted with "severe" DKA: zero of 37 patients receiving less than 25 cc/kg IVF over 1st four hrs suffered herniation; 1 of 42 patients receiving between 25 and 50 cc/kg in this initial treatment phase suffered herniation; and 8 of the 40 patients receiving greater than 50cc/kg over the first 4 hours suffered brain herniation.

Additional Comments

  • There were other interesting trends noted in the study, which were not significantly significant (most likely due to small sample size.) The rate decline in both Na+ and glucose during treatment was greater in the affected group than in non-affected group.
  • The theory behind using caution during IVF administration is that overly rapid fluid re-hydration in DKA can cause a rapid decline in serum glucose, and thus cause fluid shifts from the extra-cellular to the intra-cellular compartment, causing cerebral edema, which can lead to brain herniation.
  • Other work has been done which suggests that a "negative sodium trend" (i.e., a decline in corrected serum Na+ level) during DKA management is also a predictor of brain herniation. In fact, in their series, by monitoring serum Na+, by using slow re-hydration, and by adjusting Na+ concentration in IVF to prevent a negative sodium trend, they had zero episodes of brain herniation (of 231 episodes of DKA).(2)
  • There was no increase in non-cerebral complications of dehydration in the patients who were re-hydrated modestly, as opposed to being re-hydrated aggressively.


  1. Mahoney CP. Vicek BW. DelAguila M. Risk factors for developing brain herniation during dibetic ketoacidosis. Pediatric Neurology 1999; 21(4):721-7.
  2. Harris GD. Fiordalisi I. Physiologic Management of diabetic ketoacidemia. A 5-year prospective pediatric experience in 2331 episodes. Archives of Pediatrics & Adolescent Medicine 1994; 148(10):1046-52.
  3. The Harriet Lane Handbook, 15th edition. Siberry GK. Iannone R. pp 207-209.

CAT Author: Patricia McNally , MD

CAT Appraisers: Jonathan Fliegel , MD

Date appraised: November 13, 2000

Last updated March 23, 2003
Department of Pediatrics and Communicable Diseases
© 1998-2002 University of Michigan Health System