Cerebral edema is one of the most feared complications in pediatric patients with diabetic ketoacidosis (DKA). It is the leading cause of death from this disease process. Thankfully cerebral edema is a rare complication, occurring in less than 1% of patients¹. Although there is a paucity of evidence to support this practice, traditional teaching recommends cautious use of IV fluids due to the risk of inducing cerebral edema. These recommendations come from retrospective reviews and case studies not randomized controlled trials.
Does IV fluid infusion rate or sodium content of crystalloids influence neurologic outcomes of children with DKA?
Kuppermann et al. Clinical Trial of Fluid Infusion Rates for Pediatric Diabetic Ketoacidosis. New England Journal of Medicine 2018. [paper]
This study was a prospective, multicenter, randomized control trial. Children <18 years old who presented in DKA to 13 different emergency departments within the PECARN network were randomly assigned to one of four treatment groups as seen below. The primary outcome assessed was the number of episodes of decline in GCS to less than 14. Two secondary outcomes were also assessed including the incidence of clinically apparent brain injury and short term memory deficits. Clinically apparent brain injury was defined as clinical deterioration requiring intubation, hyperosmolar therapy, or result of death. Memory was assessed during treatment with digit span recall testing and memory testing at 3 months. Children presenting with GCS less than 14 at time of enrolment were not included in the primary analysis.
A total of 1,389 episodes of DKA occurred in 1,255 patients who underwent randomization and
1,361 episodes were included in the primary analysis. There was a total of 48 episodes of GCS
decline to <14 (3.5%). There was no statistically significant difference in the frequency,
magnitude, or duration of GCS decline between the groups. There was also no statistically
significant difference between the treatment arms in regards to secondary outcomes of
clinically apparent brain injury and memory/neurocognitive function. The number of episodes
of clinically apparent brain injury and GCS decline to <14 occurred fewer times in the fast
rehydration groups compared to the slow groups, however, these results were not statistically
In this prospective randomized controlled trial the rate of IV fluid administration and sodium chloride content did not affect neurologic outcomes in patients presenting with a GCS of 14 or greater.
This is the first randomized controlled trial to evaluate the effect of IV fluid infusion rate and sodium chloride content on neurologic outcomes in patients presenting with DKA. The strengths of the study are that it is a large multicenter study including over 1,200 patients and the study groups had similar characteristics and demographics. The clinical dogma surrounding fluids in pediatric DKA was well challenged by this study. Current practice is to limit fluids and resuscitate slowly to prevent cerebral edema. This study has shown that we should be providing appropriate fluid resuscitation.
While these results did not demonstrate a statistically significant difference in neurologic outcomes, clinically apparent brain injury and decline of GCS to <14 occurred less frequently in the fast rehydration groups. This may suggest more aggressive fluid resuscitation may be of benefit while treating DKA.
For me, the biggest limitations of this study is the exclusion of patients with an initial GCS <14 from the primary analysis. These results may not apply to a more critically ill population of patients who also may be at higher risk for complications including cerebral edema. Another limitation is the rate of IV fluids used in the study. These rates are typically the upper and lower limits of fluid replacement protocols at most institutions. The rapid rate of fluid resuscitation used in this study is comparable to the protocol at UIC. While I don’t think this literature will necessarily change my practice, I will be more comfortable with more aggressive fluid resuscitation protocols in DKA.
¹ Walls, R. M. (2018). Rosen’s Emergency Medicine Concepts and Clinical Practice.
Philadelphia, PA: Elsevier.
Ryan Gluth is an EM Resident, Class of 2020
Knowledge Bomb! is presented during Resident Educational Conference and discussed among residents and attendings in the UIC EM program. Elspeth Pearce, MD editor.