EMERGING THERAPIES: DRUGS AND REGIMENS

Diabetes Care Volume 44, September 2021 2061

Effects of Fluid Rehydration Arleta Rewers,1 Nathan Kuppermann,2,3

Michael J. Stoner,4 Aris Garro,5
Strategy on Correction of Jonathan E. Bennett,6
Kimberly S. Quayle,7 Jeffrey E. Schunk,8
Acidosis and Electrolyte Sage R. Myers,9 Julie K. McManemy,10
Lise E. Nigrovic,11 Jennifer L. Trainor,12
Abnormalities in Children With Leah Tzimenatos,2 Maria Y. Kwok,13
Kathleen M. Brown,14 Cody S. Olsen,8
Diabetic Ketoacidosis T. Charles Casper,8 Simona Ghetti,15
Nicole S. Glaser,3 and
Diabetes Care 2021;44:2061–2068 | https://doi.org/10.2337/dc20-3113 the Pediatric Emergency Care Applied
Research Network (PECARN) FLUID Study
Group*

1
Division of Emergency Medicine, Department of
Pediatrics, Colorado Children’s Hospital, University
of Colorado–Denver School of Medicine, Aurora, CO
2
OBJECTIVE Department of Emergency Medicine, University of
California Davis Health, University of California,
Fluid replacement to correct dehydration, acidosis, and electrolyte abnormalities Davis, School of Medicine, Sacramento, CA
3
is the cornerstone of treatment for diabetic ketoacidosis (DKA), but little is known Department of Pediatrics, University of
about optimal fluid infusion rates and electrolyte content. The objective of this California Davis Health, University of California,
Davis, School of Medicine, Sacramento, CA
study was to evaluate whether different fluid protocols affect the rate of normali- 4
Department of Pediatrics, Nationwide Children’s
zation of biochemical derangements during DKA treatment. Hospital, Ohio State University College of Medicine,
Columbus, OH
5
RESEARCH DESIGN AND METHODS Departments of Emergency Medicine and
Pediatrics, Rhode Island Hospital, Warren Alpert
The current analysis involved moderate or severe DKA episodes (n = 714) in chil-
Medical School of Brown University, Providence, RI
dren age <18 years enrolled in the Fluid Therapies Under Investigation in DKA 6
Division of Emergency Medicine, Nemours/
(FLUID) Trial. Children were assigned to one of four treatment groups using a 2  Alfred I. duPont Hospital for Children, Sidney
2 factorial design (0.90% or 0.45% saline and fast or slow rate of administration). Kimmel Medical College at Thomas Jefferson
University, Philadelphia, PA
7
RESULTS Division of Emergency Medicine, Department
of Pediatrics, St. Louis Children’s Hospital,
The rate of change of pH did not differ by treatment arm, but PCO2 increased Washington University School of Medicine in
more rapidly in the fast versus slow fluid infusion arms during the initial 4 h of St. Louis, St. Louis, MO
8
treatment. The anion gap also decreased more rapidly in the fast versus slow infu- Department of Pediatrics, University of Utah
School of Medicine, Salt Lake City, UT
sion arms during the initial 4 and 8 h. Glucose-corrected sodium levels remained 9
Division of Emergency Medicine, Department of
stable in patients assigned to 0.90% saline but decreased in those assigned to Pediatrics, Children’s Hospital of Philadelphia,
0.45% saline at 4 and 8 h. Potassium levels decreased, while chloride levels Perelman School of Medicine at the University of
increased more rapidly with 0.90% versus 0.45% saline. Hyperchloremic acidosis Pennsylvania, Philadelphia, PA
10
Division of Emergency Medicine, Department
occurred more frequently in patients in the fast arms (46.1%) versus the slow of Pediatrics, Texas Children’s Hospital, Baylor
arms (35.2%). College of Medicine, Houston, TX
11
Division of Emergency Medicine, Department
CONCLUSIONS of Pediatrics, Boston Children’s Hospital,
Harvard Medical School, Boston, MA
In children treated for DKA, faster fluid administration rates led to a more rapid 12
Division of Emergency Medicine, Department of
normalization of anion gap and PCO2 than slower fluid infusion rates but were Pediatrics, Ann and Robert H. Lurie Children’s
associated with an increased frequency of hyperchloremic acidosis. Hospital of Chicago, Northwestern University
Feinberg School of Medicine, Chicago, IL
13
Division of Emergency Medicine, Department of
Pediatrics, New York Presbyterian Morgan Stanley
Diabetic ketoacidosis (DKA) is a life-threatening complication of type 1 diabetes. In Children’s Hospital, Columbia University College of
the U.S., DKA affects 30–60% of children and youth at diagnosis of type 1 diabe- Physicians and Surgeons, New York, NY
14
Division of Emergency Medicine, Department of
tes (1–3) and occurs at a rate of 7–10% per year among those with established
Pediatrics, Children’s National Medical Center,
type 1 diabetes (4–6). Brain injuries associated with DKA increase acute morbidity, George Washington School of Medicine and Health
mortality, and costs (7–9) and result in adverse long-term neurological outcomes Sciences, Washington, DC
2062 Rehydration and Correction of DKA in Children Diabetes Care Volume 44, September 2021

(10–14). Children who present with DKA was to investigate the effect of IV fluid concentration. Children whose treatment
at the time of diagnosis of type 1 diabe- administration rate and NaCl content on was changed for clinical reasons (adverse
tes also have worse long-term glycemic the rate of change in pH, PCO2, glucose, events or other clinical conditions that
control (15,16), increasing the risk of anion gap, glucose-corrected sodium, required changes in treatment) were not
microvascular complications. chloride, and potassium levels as well as excluded from the current analyses. Finally,
Fluid replacement to correct dehydra- time to metabolic normalization in chil- we included only moderate or severe DKA
tion, acidosis, and electrolyte abnormali- dren randomized to one of four DKA fluid episodes defined as serum glucose >300
ties is the cornerstone of treatment of treatment protocols in the PECARN FLUID mg/dL and venous pH <7.20 or serum
DKA. However, the optimal volume, rate Trial. bicarbonate <10 mEq/L. Mild DKA epi-
of infusion, and sodium content of intra- sodes (522 of 1,389, 38%) were excluded
venous (IV) fluid replacement have been RESEARCH DESIGN AND METHODS because DKA often resolved rapidly, fre-
controversial (17–19). A recent large ran- Study Design quently ending before the 4- or 8-h time
domized controlled trial, the Pediatric The PECARN FLUID Trial evaluated the points, making analyses of biochemical
Emergency Care Applied Research Net- effects of fluid rehydration rate and NaCl trends difficult and irrelevant. The current
work (PECARN) Fluid Therapies Under content on neurocognitive outcomes in analyses, therefore, included 714 (51%)
Investigation in DKA (FLUID) Trial, found 1,255 children <18 years old with 1,389 moderate to severe DKA episodes.
that neither the rate of administration episodes of DKA between February 2011
nor the NaCl content of IV fluids signifi- and September 2016 (20). This random- Procedures
cantly influenced neurological outcomes ized controlled trial was conducted in 13 The PECARN FLUID Trial compared four
in children with DKA (20). Recommenda- children’s hospitals within PECARN. Writ- fluid rehydration protocols frequently
tions from the FLUID Trial are part of the ten informed consent was obtained from used in the U.S. to treat children with
current Clinical Practice Consensus Guide- the parents or guardians of all enrolled DKA (24) (Supplementary Table 1). In
lines of the International Society for Pedi- patients. This study was approved by the the current analysis, we compared bio-
atric and Adolescent Diabetes (21). The local institutional review board at each chemical trends among children ran-
guidelines suggest that fluid management study site. Assent was obtained from domized to each treatment arm. Fluid
should be based on the patient’s hydra- patients whose age met the minimum infusion rate was classified as either
tion status and may be administered at age for assent according to their local fast or slow, and NaCl content was
rates as high as those used in the FLUID institutional review board. Exclusion crite- classified as either 0.90% or 0.45%
Trial. ria were previously described (24) and NaCl according to randomization arm.
Liberalization of fluid treatment proto- consisted of conditions unrelated to DKA Patients received a fluid bolus of 0.90%
cols may have occurred as a result of the that may affect mental status or cognitive saline, with the amount of bolus fluid
FLUID Trial data; however, little is known abilities or receipt of substantial treatment dictated by the randomization group.
about the effects of IV fluid administration for DKA prior to transfer to the study cen- Fluid boluses could be repeated at the
rate and NaCl content on the rates of ter. DKA was defined as a blood glucose discretion of the clinician if additional
change in glucose and electrolytes and level of >300 mg/dL [16.7 mmol/L] and boluses were felt to be clinically neces-
metabolic normalization during treatment either a venous pH of <7.25 or a serum sary. For all enrolled children, insulin
of DKA in children. A small randomized bicarbonate level of <15 mEq/L. was administered as an IV infusion at
controlled trial in adults with DKA and In the current post hoc analysis of data 0.1 units/kg/h. Potassium replacement
modest fluid deficits suggested that from the PECARN FLUID Trial, we excluded was included in IV fluids after the initial
smaller volumes of IV fluid administration DKA episodes in which the child withdrew fluid boluses. Combinations of potas-
lead to faster normalization of bicarbon- (n = 43, 3%), was later found to be ineligi- sium salts used for replacement varied
ate concentrations and shorter hospitaliza- ble (n = 8, 1%), or did not receive treat- somewhat among sites as a result of a
tions (22). In contrast, a small randomized ment according to protocol for nonclinical shortage of potassium phosphate during
controlled trial in children with DKA (23) reasons (n = 102, 7%). Patients in this last the study period; however, analyses
showed faster metabolic normalization in group either 1) received <80% or >125% were stratified by site to account for
those receiving larger volumes of IV fluids. of the expected amount of total protocol- this variability. Dextrose was added to
The FLUID Trial was designed, in part, to prescribed fluid or 2) received a majority the IV fluids when the serum glucose
provide definitive answers in this area of (>50%) of nonbolus fluid that did not concentration was 200–300 mg/dL to
controversy. The goal of the current study match the protocol-prescribed sodium maintain the serum glucose concentra-

15
Department of Psychology and the Center for Clinical trial reg. no. NCT00629707, clinicaltrials.gov © 2021 by the American Diabetes Association.
Mind and Brain, University of California, Davis, This article contains supplementary material Readers may use this article as long as the
Davis, CA online at https://doi.org/10.2337/figshare.14664 work is properly cited, the use is educational
Corresponding author: Arleta Rewers, arleta. 576. and not for profit, and the work is not altered.
[email protected] More information is available at https://www.
*A complete list of the PECARN FLUID Study
diabetesjournals.org/content/license.
Received 22 December 2020 and accepted 20 Group members can be found in the supplementary
May 2021 material online.
care.diabetesjournals.org Rewers and Associates 2063

SAS/STAT 9.4 statistical software (SAS

Enrolled in the PECARN FLUID Trial Institute, Cary, NC).
N=1389
RESULTS
Comparisons were based on 714 episodes
Excluded N=675 of moderate or severe DKA among 667
43 withdrew study participants. Baseline demographic
8 later found ineligible characteristics of participants and baseline
11 did not receive study treatment laboratory values (Table 1) did not differ
91 received treatment outside of protocol by treatment arm. Children in the slow
522 experienced mild DKA (pH ≥7.2) infusion arms received more than the
prescribed IV fluid bolus more often than
those in the fast infusion arms (P <
Included in Analysis N=714 0.001).
Fast arm, 0.45% NaCl N=170 Fast arm, 0.90% NaCl N=186 The primary study outcomes by treat-
Slow arm, 0.45% NaCl N=182 Slow arm, 0.90% NaCl N=176 ment assigned are compared in Table 2.
We did not find significant differences
in rates of change in pH among study
Figure 1—Study flow diagram. arms. Increases in PCO2 between treat-
ment initiation and 4 h after treatment
tion between 100 and 200 mg/dL. A age-specific bicarbonate level (in mEq/L, initiation were more rapid in children in
detailed description of patient enroll- <18 for age 1–3 years, <19 for 4–5 years, the fast infusion arms compared with
ment can be found in the study flow <20 for 6–7 years, <21 for $8 years) at those in the slow infusion arms. How-
diagram (Fig. 1). Primary outcomes any time between randomization and ever, rates of change in PCO2 between
included the rates of change in pH, DKA resolution or a reported adverse treatment initiation and 8 h after treat-
PCO2, anion gap, glucose, glucose-cor- event of hyperchloremic acidosis. Hyper- ment initiation were not significantly
rected sodium, chloride, and potassium natremia was defined as a glucose-cor- different among arms. Mean time to
during treatment. Rates of change in rected sodium level >155 mEq/L. normalization of pH and PCO2 also did
biochemical measures were determined not differ significantly among the arms
by calculating the differences between Statistical Analyses (Table 2).
the initial laboratory values and the val- We used the van Elteren test to analyze The rate of decrease in anion gap
ues measured 4- or 8-h after initiation treatment differences for the following was more rapid in children treated with
of protocol-prescribed IV fluids (±1 h). continuous outcomes: rate of change in fast fluid infusion compared with those
This difference was divided by the time pH, PCO2, anion gap, glucose, sodium, treated with slow fluid infusion at both
from DKA treatment initiation (typically chloride, and potassium concentrations 4 and 8 h after treatment initiation.
the first study fluid bolus) until the time between treatment initiation and 4 or There were no significant differences in
of the hour 4 or 8 laboratory measure- 8 h after treatment initiation as well as rates of change in anion gap between
ments to determine the hourly rate of time to normalization of pH, PCO2, anion the 0.90% and 0.45% NaCl arms. The
change. We used the first value mea- gap, and serum glucose concentration. mean time to normalization of anion
sured within 60 min of DKA treatment We used the Mantel-Haenszel test to gap was significantly shorter in children
initiation as the initial value. Data from assess differences in rates of hyperchlore- assigned to fast infusion arms compared
complete electrolyte panels were used mic acidosis and hypernatremia as well with those assigned to slow infusion
preferentially. Time to metabolic nor- as frequency of administration of addi- arms.
malization in hours was determined for tional fluid boluses beyond that dictated The rates of decline in glucose concen-
pH, PCO2, glucose, and anion gap. Those by the study protocol. Statistical analyses trations from treatment initiation to 4 and
outcomes were calculated as the time were stratified by study hospital. For 8 h after treatment initiation were more
from the initial laboratory measurement comparisons of fluid infusion rate, tests rapid in children treated with slow fluid
until the first normal value (pH $7.32; were further stratified by fluid NaCl con- infusion compared with those treated with
PCO2 $38 torr; anion gap <12 mEq/L; tent. For comparisons of fluid NaCl con- fast fluid infusion. In addition, the mean
glucose #200 mg/dL). Time of transi- tent, tests were further stratified by fluid time from treatment initiation to glucose
tion to subcutaneous insulin was used infusion rate. Both linear regression and normalization was significantly shorter
as a surrogate for normalization time if conditional logistic regression models among children assigned to slow infusion
the above-listed laboratory values were were used to test for treatment interac- arms compared with those assigned to
not achieved by 8 h of treatment. tions, adjusting for study hospital. Two- fast infusion arms. Rates of change in
Secondary outcome measures included tailed tests with a significance level of glucose concentrations did not differ
rates of adverse events related to changes 0.05 were used to evaluate outcomes for between the 0.45% and 0.90% NaCl
in glucose and electrolytes. Hyperchlore- all treatment comparisons. No correc- arms. In a sensitivity analysis, we removed
mic acidosis during DKA was defined as tions were made for multiple compari- 6% of the study subjects who did not
either an anion gap #12 mEq/L and low sons. We conducted all analyses using achieve glucose concentrations <200 mg/
2064 Rehydration and Correction of DKA in Children Diabetes Care Volume 44, September 2021

Table 1—Demographic and baseline laboratory characteristics of the study participants

Treatment arm assigned
Fast 0.45% Fast 0.90% Slow 0.45% Slow 0.90% Overall
Characteristic (n = 170) (n = 186) (n = 182) (n = 176) (N = 714)
Age (years) 11.7 ± 4.1 12.3 ± 3.9 12.2 ± 3.8 12.1 ± 3.9 12.1 ± 3.9
Male sex 75 (44.1) 87 (46.8) 89 (48.9) 81 (46.0) 332 (46.5)
Race
White 110 (64.7) 111 (59.7) 123 (67.6) 116 (65.9) 460 (64.4)
Black or African American 40 (23.5) 55 (29.6) 43 (23.6) 37 (21.0) 175 (24.5)
Multiracial 10 (5.9) 6 (3.2) 5 (2.7) 10 (5.7) 31 (4.3)
American Indian/Alaska Native 0 (0.0) 2 (1.1) 2 (1.1) 2 (1.1) 6 (0.8)
Asian 1 (0.6) 0 (0.0) 1 (0.5) 1 (0.6) 3 (0.4)
Native Hawaiian or other Pacific Islander 1 (0.6) 0 (0.0) 1 (0.5) 2 (1.1) 4 (0.6)
Unknown 8 (4.7) 12 (6.5) 7 (3.8) 8 (4.5) 35 (4.9)
Ethnicity
Hispanic or Latino 26 (15.3) 36 (19.4) 26 (14.3) 36 (20.5) 124 (17.4)
Not Hispanic or Latino 138 (81.2) 142 (76.3) 150 (82.4) 135 (76.7) 565 (79.1)
Unknown 6 (3.5) 8 (4.3) 6 (3.3) 5 (2.8) 25 (3.5)
New-onset diabetes 72 (42.4) 77 (41.4) 77 (42.3) 69 (39.2) 295 (41.3)
pH 7.11 ± 0.07 (155) 7.10 ± 0.09 (168) 7.10 ± 0.08 (165) 7.09 ± 0.09 (163) 7.10 ± 0.08 (651)
PCO2 (mmHg) 25.0 ± 7.6 (155) 25.8 ± 7.0 (168) 25.2 ± 7.7 (164) 26.1 ± 7.4 (161) 25.5 ± 7.4 (648)
Glucose (mg/dL) 541 ± 161 (167) 540 ± 165 (179) 525 ± 166 (177) 537 ± 163 (175) 536 ± 163 (698)
Glucose-corrected sodium (mEq/L) 142.8 ± 5.0 (165) 141.5 ± 5.6 (177) 142.0 ± 5.2 (170) 142.3 ± 6.0 (170) 142.1 ± 5.5 (682)
Chloride (mEq/L) 99.4 ± 5.7 (165) 98.2 ± 7.1 (171) 99.3 ± 5.4 (165) 99.5 ± 6.4 (171) 99.1 ± 6.2 (672)
Bicarbonate (mEq/L) 7.4 ± 2.8 (164) 7.3 ± 2.5 (175) 7.6 ± 3.2 (170) 7.4 ± 2.8 (168) 7.4 ± 2.8 (677)
Potassium (mEq/L) 4.9 ± 0.9 (166) 5.1 ± 1.2 (173) 5.1 ± 1.1 (169) 5.1 ± 1.1 (171) 5.0 ± 1.1 (679)
Data are mean ± SD or n (%). Number of DKA episodes with responses are shown for characteristics with item nonresponse. No differences
by fast vs. slow or 0.45% vs. 0.90% NaCl were found.

dL prior to initiation of subcutaneous insu- arms (P = 0.002). There was a smaller infusion rate on changes in potassium
lin. This analysis found similar results, with effect of NaCl concentration, with hyper- and glucose-corrected sodium concentra-
continued significant differences between chloremic acidosis developing in 157 of tions; however, these small differences
treatment groups in glucose normalization. 362 (43.4%) children in the 0.90% NaCl were unlikely to be clinically relevant.
Glucose-corrected sodium concentra- arms vs. 133 of 352 (37.8%) in the Our results demonstrate that ketoaci-
tions decreased gradually during the ini- 0.45% NaCl arms (P = 0.06). Rates of dosis resolves more quickly with more
tial 4 and 8 h of treatment in children hypernatremia did not differ by treat- rapid fluid infusion; however, hyper-
randomized to 0.45% NaCl infusion but ment arm. chloremic acidosis is also more frequent.
did not change appreciably in those ran- Our findings are similar to results of an
domized to the 0.90% NaCl infusion CONCLUSIONS earlier very small trial in children with
arms. The rate of rehydration had a smaller DKA (23) but contradict another report
In this large randomized controlled trial
effect on glucose-corrected sodium concen- (22) that showed more rapid correction
of fluid treatment in pediatric DKA, we
trations, but this effect was significant at of bicarbonate levels and shorter hospi-
analyzed differences in the rates of cor-
the 8-h time point. Chloride concentra- tal stay after receiving smaller amounts
rection of acidosis and normalization of of IV fluids in adults treated for DKA.
tions increased and potassium concen-
trations decreased more rapidly in the electrolytes with variable fluid infusion We found, rather unexpectedly, that
0.90% NaCl vs. 0.45% NaCl arms during rates and fluid NaCl concentrations. faster fluid administration led to a
both 4 and 8 h of treatment. Changes in More rapid fluid administration led to slower decline in glucose concentra-
concentrations of electrolytes and glu- more rapid increases in PCO2 during the tions, after adjusting for potential con-
cose during DKA treatment are pre- first 4 h of treatment and more rapid founders. The difference was small (3
sented in Fig. 2. normalization of the anion gap, which mg/dL/h on average), with children in
occurred 2–3 h earlier in the fast fluid the slower infusion arms reaching a
Secondary Outcome Measures infusion arms. In contrast, pH did not blood glucose concentration of #200
(Adverse Effects) normalize more rapidly with faster fluid mg/dL on average 1 h earlier than those
Hyperchloremic acidosis developed in infusion rates, possibly as a result of fre- in the fast infusion arms. This minor
164 of 356 (46.1%) children in the fast quent development of hyperchloremic effect may have been related to lower
fluid infusion arms compared with 126 acidosis in the fast fluid infusion arms. glucose load per hour in patients receiv-
of 358 (35.2%) in the slow fluid infusion We also observed modest effects of fluid ing dextrose-containing fluids at slower
care.diabetesjournals.org Rewers and Associates 2065

Table 2—Primary outcomes by treatment assigned

Treatment arm assigned P value
Fast 0.45% Fast 0.90% Slow 0.45% Slow 0.90%
(n = 170) (n = 186) (n = 182) (n = 176) Fast vs. slow 0.45% vs. 0.90%
4-h rate of change in pH 0.02 ± 0.02 (87) 0.02 ± 0.01 (102) 0.02 ± 0.02 (84) 0.03 ± 0.02 (90) 0.26 0.73
(units/h)
8-h rate of change in pH 0.02 ± 0.02 (79) 0.02 ± 0.01 (91) 0.02 ± 0.01 (79) 0.02 ± 0.02 (87) 0.13 0.98
(units/h)
Time from treatment 15.5 ± 7.1 (170) 15.2 ± 6.5 (186) 15.7 ± 8.2 (182) 16.6 ± 8.3 (176) 0.24 0.60
initialization until pH
reaches $7.32 (h)
4-h rate of change in PCO2 0.4 ± 1.2 (85) 0.6 ± 1.2 (102) 0.1 ± 1.3 (85) 0.0 ± 1.5 (88) 0.003 0.33
(mmHg/h)
8-h rate of change in PCO2 0.8 ± 1.3 (83) 0.7 ± 1.2 (89) 0.7 ± 0.9 (79) 0.7 ± 0.9 (84) 0.44 0.92
(mmHg/h)
Time from treatment 14.8 ± 7.6 (170) 14.9 ± 8.0 (186) 15.9 ± 8.3 (182) 16.0 ± 9.8 (176) 0.35 0.59
initialization until PCO2
reaches $38 (h)
4-h rate of change in anion 2.2 ± 0.8 (128) 2.5 ± 1.1 (116) 2.0 ± 0.9 (120) 2.0 ± 0.9 (109) <0.001 0.06
gap (mEq/L/h)
8-h rate of change in anion 2.1 ± 0.7 (120) 2.2 ± 1.1 (107) 1.9 ± 0.7 (121) 1.9 ± 0.9 (109) <0.001 0.89
gap (mEq/L/h)
Time from treatment 14.8 ± 6.3 (170) 14.4 ± 7.2 (186) 16.7 ± 9.3 (182) 17.8 ± 11.9 (176) 0.002 0.40
initialization until anion
gap <12 (h)
4-h rate of change in 51 ± 28 (116) 57 ± 39 (119) 57 ± 34 (120) 59 ± 32 (111) 0.02 0.10
glucose (mg/dL/h)
8-h rate of change in 46 ± 42 (93) 49 ± 43 (87) 51 ± 38 (86) 51 ± 39 (67) 0.04 0.36
glucose (mg/dL/h)
Time from treatment 8.8 ± 5.3 (170) 8.2 ± 4.5 (186) 7.6 ± 4.8 (182) 7.1 ± 4.1 (176) 0.003 0.26
initialization to glucose
#200 mg/dL or
subcutaneous insulin (h)
4-h rate of change in 0.5 ± 0.5 (140) 0.0 ± 1.1 (137) 0.3 ± 0.6 (135) 0.1 ± 0.6 (123) 0.07 <0.001
glucose-corrected
sodium (mEq/L/h)
8-h rate of change in 0.4 ± 0.4 (132) 0.0 ± 1.0 (134) 0.3 ± 0.5 (139) 0.1 ± 0.6 (125) 0.046 <0.001
glucose-corrected
sodium (mEq/L/h)
4-h rate of change in 1.7 ± 0.7 (137) 2.3 ± 1.0 (130) 1.7 ± 0.8 (126) 2.0 ± 0.8 (119) 0.32 <0.001
chloride (mEq/L/h)
8-h rate of change in 1.2 ± 0.6 (126) 1.7 ± 0.8 (124) 1.2 ± 0.5 (130) 1.4 ± 0.5 (122) 0.37 <0.001
chloride (mEq/L/h)
4-h rate of change in 0.06 ± 0.12 (137) 0.14 ± 0.18 (135) 0.11 ± 0.18 (133) 0.10 ± 0.21 (121) 0.15 0.001
potassium (mEq/L/h)
8-h rate of change in 0.07 ± 0.08 (132) 0.12 ± 0.12 (131) 0.10 ± 0.10 (137) 0.12 ± 0.11 (126) 0.04 <0.001
potassium (mEq/L/h)
Data are mean ± SD (n). Treatment interaction effects were present for rate of change in chloride at 4 h (P = 0.02) and 8 h (P = 0.02) as well
as for 4-h rate of change in potassium (P = 0.005) according to linear regression models. P values reported are from Van Elteren test stratified
by treatment and study hospital. Boldface type indicates significance at P < 0.05.

infusion rates while continuing to treated with 0.45% NaCl had declines in increased less in the 0.45% NaCl arms
receive insulin at the same dosage as glucose-corrected sodium concentra- compared with the 0.90% NaCl arms,
those receiving higher fluid infusion tions in contrast to those receiving resulting in lower rates of hyperchlore-
rates. 0.90% NaCl, whose average glucose-cor- mic acidosis. Finally, potassium levels
The choice of 0.45% vs. 0.90% NaCl rected sodium concentrations did not decreased less rapidly in the 0.45%
solution as the rehydration fluid had a decrease during the 8-h observation NaCl arms. This effect was more sub-
substantial effect on the rate of change period. These findings are similar to stantial in children receiving fast fluid
in glucose-corrected sodium, chloride, those reported in a previous small ret- replacement rates, where a significant
and potassium concentrations. Children rospective study (25). Chloride levels interaction between treatment effects
2066 Rehydration and Correction of DKA in Children Diabetes Care Volume 44, September 2021

Figure 2—Changes in electrolyte and glucose concentrations during DKA treatment. Means and 95% CIs are shown by treatment group assigned.

was noted. These differences likely Although hyperchloremic acidosis resolves administration resolved DKA more quickly
result from increased sodium delivery spontaneously, this metabolic derange- (normalization of anion gap) but with
to the renal tubules, with more rapid ment may mask recognition of resolution an increased risk of hyperchloremia, espe-
infusion or higher sodium content fluids of ketoacidosis when total base deficit or cially if 0.90% NaCl was used. Potassium
causing increased aldosterone-stimu- bicarbonate levels are used to monitor replacement in our study was given as an
lated sodium reabsorption and potas- biochemical improvement (28,29). To avoid equal mixture of potassium chloride and
sium secretion. this misinterpretation, the International potassium phosphate, which further
Preferential renal excretion of ketones Society for Pediatric and Adolescent Diabe- increased the chloride load. The develop-
over chloride ions in DKA, in addition to tes guidelines recommend measurement ment of hyperchloremic acidosis could be
infusion of large amounts of NaCl, may of bedside β-hydroxybutyrate concentra- mitigated by using 0.45% NaCl rather than
lead to hyperchloremic metabolic acidosis tions to monitor DKA resolution (21). In 0.9% NaCl and by using potassium salts
during treatment for DKA (25–27). the current study, more rapid fluid other than potassium chloride.
care.diabetesjournals.org Rewers and Associates 2067

The current study was a post hoc Andrew Bremer [Eunice Kennedy Shriver the study and take responsibility for the integrity
analysis, and the results should there- National Institute of Child and Human Devel- of the data and the accuracy of the data
opment, Bethesda, MD], Thomas Cook analysis.
fore be interpreted with several limita- [Department of Biostatistics and Medical Prior Presentation. Parts of this study were
tions in mind. The generalizability of our Informatics, University of Wisconsin, Madison, presented in abstract form at the 79th Scien-
results may be somewhat limited for WI], and Beth Slomine [Department of Psychi- tific Sessions of the American Diabetes Associ-
patients at the extremes of the spec- atry, Johns Hopkins University School of Medi- ation, San Francisco, CA, 7–11 June 2019.
trum of pediatric DKA. We excluded a cine, Baltimore, MD]), and the members of
the study outcome adjudication committee References
small subset of patients presenting with (Kathleen Meert [Department of Pediatrics, 1. Rewers A, Dong F, Slover RH, Klingensmith GJ,
Glasgow Coma Scale scores #11. We Children’s Hospital of Michigan, Detroit, MI], Rewers M. Incidence of diabetic ketoacidosis at
also excluded patients with mild DKA. It Jerry Zimmerman [Center for Clinical and diagnosis of type 1 diabetes in Colorado youth,
is possible that results may have dif- Translational Research, Seattle Children’s Hos- 1998-2012. JAMA 2015;313:1570–1572
pital, Seattle, WA], and Robert Hickey [Divi- 2. Alonso GT, Coakley A, Pyle L, Manseau K,
fered for these subsets of patients. In
sion of Pediatric Emergency Medicine, Thomas S, Rewers A. Diabetic ketoacidosis at
addition, the fluid infusion rates used in Children’s Hospital of Pittsburgh, Pittsburgh, diagnosis of type 1 diabetes in Colorado children,
our study were selected to represent PA]). 2010-2017. Diabetes Care 2020;43:117–121
the upper and lower ends of DKA proto- Funding. This study was supported by a grant 3. Jensen ET, Stafford JM, D’Agostino RB Jr., et al.
cols used in the U.S. Results may differ from the Eunice Kennedy Shriver National Insti- Rapid increase in prevalence of DKA at diagnosis
tute of Child Health and Human Development among youth with type 1 diabetes: the SEARCH
for other protocols previously proposed
(U01-HD-062417). This project was also sup- for Diabetes in Youth study. Oral presentation at
for this patient population (30). ported in part by the Health Resources and the 55th European Association for the Study of
To our knowledge, this is the largest Services Administration, Maternal and Child Diabetes Annual Meeting, 16–20 September
randomized clinical trial evaluating the Health Bureau, and Emergency Medical Services 2019, Barcelona, Spain
effect of different fluids rates and for Children Network Development Demonstra- 4. Rewers A, Chase HP, Mackenzie T, et al.
tion Program under cooperative agreement Predictors of acute complications in children with
sodium content on changes in electro-
numbers U03-MC-00008, U03-MC-00001, U03- type 1 diabetes. JAMA 2002;287:2511–2518
lyte levels and metabolic normalization MC-00003, U03-MC-00006, U03-MC-00007, U03- 5. Cengiz E, Xing D, Wong JC, et al.; T1D
in children treated for DKA. We found MC-22684, and U03-MC-22685. Exchange Clinic Network. Severe hypoglycemia
that faster fluid administration normal- This information, content, and conclusions of and diabetic ketoacidosis among youth with type
ized the anion gap 2–3 h earlier than this article are those of the authors and should 1 diabetes in the T1D Exchange clinic registry.
not be construed as the official position or pol-
slower fluid infusion rates. Previous icy of, and no endorsements should be inferred
Pediatr Diabetes 2013;14:447–454
6. Maahs DM, Hermann JM, Holman N, et al.;
analyses also demonstrated that faster by, the Health Resources and Services Adminis- National Paediatric Diabetes Audit and the Royal
fluid infusion rates were not associated tration, Department of Health and Human Serv- College of Paediatrics and Child Health, the DPV
with greater risk of mental status ices, or the U.S. government. Initiative, and the T1D Exchange Clinic Network.
changes or clinical diagnoses of cerebral Duality of Interest. No potential conflicts of Rates of diabetic ketoacidosis: international
interest relevant to this article were reported. comparison with 49,859 pediatric patients with type
injury in the full study population and Author Contributions. A.R. conceived and 1 diabetes from England, Wales, the U.S., Austria,
resulted in improved mental status dur- designed the study, contributed to the data anal- and Germany. Diabetes Care 2015;38:1876–1882
ing DKA treatment in some patient sub- ysis, supervised patient enrollment and data 7. Glaser N, Barnett P, McCaslin I, et al.; Pediatric
groups (20). Although more rapid fluid abstraction, drafted the initial manuscript, and Emergency Medicine Collaborative Research
approved the final manuscript. N.K. conceived Committee of the American Academy of
rates were associated with an increased
and designed the study, obtained grant funding, Pediatrics. Risk factors for cerebral edema in
frequency of hyperchloremic acidosis, supervised training of study personnel, super- children with diabetic ketoacidosis. N Engl J Med
this complication is generally benign vised patient enrollment and data abstraction, 2001;344:264–269
and could be mitigated by using 0.45% contributed to the data analysis, contributed to 8. Edge JA, Jakes RW, Roy Y, et al. The UK case-
NaCl and potassium salts other than the initial draft, and revised and approved the control study of cerebral oedema complicating
final manuscript. M.J.S., A.G., J.E.B., K.S.Q., J.E.S., diabetic ketoacidosis in children. Diabetologia
potassium chloride for potassium
S.R.M., J.K.M., L.E.N., J.L.T., L.T., M.Y.K., and 2006;49:2002–2009
replacement. In conclusion, on the basis K.M.B. supervised patient enrollment and data 9. Patel A, Singh D, Bhatt P, Thakkar B, Akingbola
of our findings, we recommend treating abstraction, contributed to the study design, and OA, Srivastav SK. Incidence, trends, and outcomes
pediatric DKA using fluid infusion rates revised and approved the final manuscript. of cerebral edema among children with diabetic
similar to the fast infusion rates used in C.S.O. conducted the data analyses, drafted the ketoacidosis in the United States. Clin Pediatr
tables and figures for the final manuscript, and (Phila) 2016;55:943–951
this study.
revised and approved the final manuscript. T.C.C. 10. Ghetti S, Lee JK, Sims CE, Demaster DM,
supervised data analyses and drafting of tables Glaser NS. Diabetic ketoacidosis and memory
and figures for the final manuscript and revised dysfunction in children with type 1 diabetes. J
Acknowledgments. The authors thank Marci and approved the final manuscript. S.G. super- Pediatr 2010;156:109–114
Fjelstad and Amy Watson from the PECARN vised training of study personnel and neurocog- 11. Cameron FJ, Scratch SE, Nadebaum C, et al.; DKA
Data Coordinating Center for their assistance; nitive data collection, contributed to the study Brain Injury Study Group. Neurological consequences
the research coordinators in PECARN, without design, and revised and approved the final man- of diabetic ketoacidosis at initial presentation of type 1
whom this trial would not have been possi- uscript. N.S.G. conceived and designed the study, diabetes in a prospective cohort study of children.
ble; the clinicians in PECARN who enrolled obtained grant funding, supervised training of Diabetes Care 2014;37:1554–1562
children into this study; the members of the study personnel, supervised patient enrollment 12. Semenkovich K, Bischoff A, Doty T, et al. Clinical
data and safety monitoring board (Roger and data abstraction, contributed to the data presentation and memory function in youth with
Lewis [Department of Emergency Medicine, analysis, contributed to the initial draft, and type 1 diabetes. Pediatr Diabetes 2016;17:
Harbor-UCLA Medical Center, Los Angeles, revised and approved the final manuscript. 492–499
CA], Jeffrey Blumer [School of Medicine, Case C.S.O. and T.C.C. are the guarantors of this work 13. Aye T, Mazaika PK, Mauras N, et al.; Diabetes
Western Reserve University, Cleveland, OH], and, as such, had full access to all the data in Research in Children Network (DirecNet) Study
2068 Rehydration and Correction of DKA in Children Diabetes Care Volume 44, September 2021

Group. Impact of early diabetic ketoacidosis on 19. White PC. Optimizing fluid management design of a factorial randomized controlled trial.
the developing brain. Diabetes Care 2019;42: of diabetic ketoacidosis. Pediatr Diabetes Pediatr Diabetes 2013;14:435–446
443–449 2015;16:317–319 25. Basnet S, Venepalli PK, Andoh J, Verhulst S,
14. Ghetti S, Kuppermann N, Rewers A, et al.; 20. Kuppermann N, Ghetti S, Schunk JE, et al.; Koirala J. Effect of normal saline and half normal
Pediatric Emergency Care Applied Research Network PECARN DKA FLUID Study Group. Clinical trial of fluid saline on serum electrolytes during recovery
(PECARN) DKA FLUID Study Group. Cognitive function infusion rates for pediatric diabetic ketoacidosis. N phase of diabetic ketoacidosis. J Intensive Care
following diabetic ketoacidosis in children with new- Engl J Med 2018;378:2275–2287 Med 2014;29:38–42
onset or previously diagnosed type 1 diabetes. 21. Wolfsdorf JI, Glaser N, Agus M, et al. ISPAD 26. Oh MS, Carroll HJ, Uribarri J. Mechanism of
Diabetes Care 2020;43:2768–2775 Clinical Practice Consensus Guidelines 2018: diabetic normochloremic and hyperchloremic acidosis in
15. Fredheim S, Johannesen J, Johansen A, ketoacidosis and the hyperglycemic hyperosmolar diabetic ketoacidosis. Nephron 1990;54:1–6
et al.; Danish Society for Diabetes in Childhood state. Pediatr Diabetes 2018;19(Suppl. 27):155–177 27. Adrogue HJ, Wilson H, Boyd AE 3rd, Suki WN,
and Adolescence. Diabetic ketoacidosis at the 22. Adrogue HJ, Barrero J, Eknoyan G. Salutary Eknoyan G. Plasma acid-base patterns in diabetic
onset of type 1 diabetes is associated with future effects of modest fluid replacement in the ketoacidosis. N Engl J Med 1982;307:1603–1610
HbA1c levels. Diabetologia 2013;56:995–1003 treatment of adults with diabetic ketoacidosis. 28. Taylor D, Durward A, Tibby SM, et al. The
16. Duca LM, Wang B, Rewers M, Rewers A. Use in patients without extreme volume deficit. influence of hyperchloraemia on acid base
Diabetic ketoacidosis at diagnosis of type 1 diabetes JAMA 1989;262:2108–2113 interpretation in diabetic ketoacidosis. Intensive
predicts poor long-term glycemic control. Diabetes 23. Bakes K, Haukoos JS, Deakyne SJ, et al. Effect Care Med 2006;32:295–301
Care 2017;40:1249–1255 of volume of fluid resuscitation on metabolic 29. von Oettingen JE, Rhodes ET, Wolfsdorf JI.
17. Duck SC, Wyatt DT. Factors associated with normalization in children presenting in diabetic Resolution of ketoacidosis in children with new
brain herniation in the treatment of diabetic ketoacidosis: a randomized controlled trial. J onset diabetes: evaluation of various definitions.
ketoacidosis. J Pediatr 1988;113:10–14 Emerg Med 2016;50:551–559 Diabetes Res Clin Pract 2018;135:76–84
18. Harris GD, Fiordalisi I, Harris WL, Mosovich LL, 24. Glaser NS, Ghetti S, Casper TC, Dean JM; 30. White PC, Dickson BA. Low morbidity and
Finberg L. Minimizing the risk of brain herniation during Pediatric Emergency Care Applied Research Network mortality in children with diabetic ketoacidosis
treatment of diabetic ketoacidemia: a retrospective (PECARN) DKA FLUID Study Group. Pediatric diabetic treated with isotonic fluids. J Pediatr 2013;163:
and prospective study. J Pediatr 1990;117:22–31 ketoacidosis, fluid therapy, and cerebral injury: the 761–766