Sie sind auf Seite 1von 16

Pediatric Diabetes 2007: 8: 28–43 # 2007 The Authors

All rights reserved Journal compilation # 2007 Blackwell Munksgaard

Pediatric Diabetes

ISPAD Clinical Practice Consensus Guidelines 2006–2007

Diabetic ketoacidosis
Wolfsdorf J, Craig ME, Daneman D, Dunger D, Edge J, Lee Department of Paediatrics, University of Cambridge,
WRW, Rosenbloom A, Sperling MA, Hanas R. Diabetic Addenbrooke’s Hospital, Cambridge, UK
ketoacidosis. Department of Paediatrics, John Radcliffe Hospital, Oxford,
Pediatric Diabetes 2007: 8: 28–42. UK
Endocrinology Service Department of Paediatric Medicine,
KK Children’s Hospital, Singapore
Joseph Wolfsdorf a g
Division of Endocrinology, Department of Pediatrics,
Maria E Craig b University of Florida College of Medicine, Gainesville, FL,
Denis Daneman c USA
Department of Pediatric Endocrinology, Children’s Hospital,
David Dunger d University of Pittsburgh, PA, USA
Julie Edge e i
Department of Pediatrics, Uddevalla Hospital, Uddevalla,
WR Warren Lee f Sweden
Arlan Rosenbloom g
Mark A Sperling h and Corresponding author:
Ragnar Hanas i Ragnar Hanas, MD, PhD,
Department of Pediatrics, Uddevalla Hospital,
S-451 80 Uddevalla, Sweden.
Division of Endocrinology, Children’s Hospital Boston, MA, Tel: 146-522-92000;
USA e-mail:
School of Women’s and Children’s Health, University of New
South Wales, Sydney, Australia Editors of the ISPAD Clinical Practice Consensus Guidelines
University of Toronto, The Hospital for Sick Children, Toronto, 2006–2007; Ragnar Hanas, Kim Donaghue, Georgeanna
Canada Klingensmith, Peter Swift

Diabetic ketoacidosis (DKA) results from absolute utilization resulting in hyperglycemia and hyperosmo-
or relative deficiency of circulating insulin and the lality, and increased lipolysis and ketogenesis, causing
combined effects of increased levels of the counter- ketonemia and metabolic acidosis. Hyperglycemia that
regulatory hormones: catecholamines, glucagon, exceeds the renal threshold [approximately 10 mmol/L
cortisol, and growth hormone (1, 2). Absolute insulin (180 mg/dL), although the range in normal and diabetic
deficiency occurs in previously undiagnosed type 1 individuals is very wide)] and hyperketonemia causes
diabetes mellitus (T1DM) and when patients on osmotic diuresis, dehydration, and obligatory loss of elec-
treatment deliberately or inadvertently do not take trolytes, which often is aggravated by vomiting. These
insulin, especially the long-acting component of a changes stimulate further stress hormone production,
basal-bolus regimen. Patients who use an insulin which induces more severe insulin resistance and
pump can rapidly develop DKA when insulin delivery worsening hyperglycemia and hyperketonemia. If this
fails for any reason. Relative insulin deficiency occurs cycle is not interrupted with exogenous insulin and fluid
when the concentrations of counter-regulatory hor- and electrolyte therapy, fatal dehydration and metabolic
mones increase in response to stress in conditions such acidosis will ensue. Ketoacidosis may be aggravated by
as sepsis, trauma, or gastrointestinal illness with lactic acidosis from poor tissue perfusion or sepsis.
diarrhea and vomiting. The pathophysiology of DKA is characterized by severe depletion of water
DKA in children is summarized in Fig. 1. and electrolytes from both the intra- and extracellular
The combination of low serum insulin and high fluid (ECF) compartments; the range of losses is
counter-regulatory hormone concentrations results in shown in Table 1. Despite their dehydration, patients
an accelerated catabolic state with increased glucose continue to maintain normal blood pressure and have
production by the liver and kidney (via glycogenolysis considerable urine output until extreme volume
and gluconeogenesis), impaired peripheral glucose depletion leads to a critical decrease in renal blood
Consensus Guidelines

Pathophysiology of Diabetic Ketoacidosis

Absolute insulin deficiency
Stress, infection or insufficient insulin intake

Counter-regulatory hormones
Glucagon Cortisol
Catecholamines Growth Hormone

Lipolysis Glucose utilization Proteolysis Glycogenolysis

Protein synthesis

Gluconeogenic substrates
FFA to liver Gluconeogenesis

Ketogenesis Hyperglycemia

Alkali reserve Glucosuria (osmotic diuresis)

Acidosis Loss of water and electrolytes

Decreased fluid intake
Lactate Dehydration Hyperosmolarity

Impaired renal function

Fig. 1. Pathophysiology of diabetic ketoacidosis. Copyright # 2006 American Diabetes Association. From Diabetes Care, Vol. 29, 2006;
1150–1159. Reprinted with permission from The American Diabetes Association.

flow and glomerular filtration. At presentation, the ¤ Increased leukocyte count with left shift
magnitude of specific deficits in an individual patient ¤ Non-specific elevation of serum amylase
varies depending upon the duration and severity of ¤ Fever only when infection is present
illness, the extent to which the patient was able to
maintain intake of fluid and electrolytes, and the
content of food and fluids consumed before coming to Definition of DKA
medical attention. Consumption of fluids with a high- The biochemical criteria for the diagnosis of DKA
carbohydrate content (juices or sugar containing soft are (4):
drinks) exacerbates the hyperglycemia (3).
¤ Hyperglycemia (blood glucose . 11 mmol/L [200
Clinical manifestations of DKA mg/dL])
¤ Venous pH , 7.3 or bicarbonate , 15 mmol/L
¤ Dehydration ¤ Ketonemia and ketonuria
¤ Rapid, deep, sighing (Kussmaul respiration) Partially treated children and children who have
¤ Nausea, vomiting, and abdominal pain mimicking consumed little or no carbohydrates may have, on
an acute abdomen rare occasion, only modestly increased blood glucose
¤ Progressive obtundation and loss of consciousness concentrations (Ôeuglycemic ketoacidosis’) (5, 6).

Table 1. Losses of fluids and electrolytes in diabetic ketoacidosis (DKA) and maintenance requirements in normal
Average (range) losses per kg 24-h maintenance requirements

Water 70 mL (30–100) *10 kg 100 mL/kg/24 h

11–20 kg 1000 mL 1 50 mL/kg/24 h for each kg from 11–20
.20 kg 1500 mL 1 20 mL/kg/24 h for each kg .20
Sodium 6 mmol (5–13) 2–4 mmol†
Potassium 5 mmol (3–6) 2–3 mmol
Chloride 4 mmol (3–9) 2–3 mmol
Phosphate (0.5–2.5) mmol 1–2 mmol

Data are from measurements in only a few children and adolescents (41, 42, 80–82). In any individual patient, actual
losses may be less or greater than the ranges shown in the Table (E).
Three methods for determining maintenance water requirements in children are commonly used: *the Holliday–Segar
formula (143) (shown in this Table), a simplified Holliday–Segar formula† (see below), and a formula based on body
surface area for children more than 10 kg (1500 mL/m2/24 h) (144).
†Maintenance electrolyte requirements in children are per 100 mL of maintenance IV fluid (144, 145).
†Simplified method based on Holliday–Segar: ,10 kg 4 mL/kg/h; 11–20 kg 40 1 2 mL/kg/h for each kg between 11 and
20; .20 kg 60 1 1 mL/kg/h for each kg .20.

Pediatric Diabetes 2007: 8: 28–43 29

Wolfsdorf et al.

Type 2 diabetes mellitus (T2DM), associated with In children with established diabetes
increased rates and severity of obesity, in some centers (recurrent DKA)
now accounts for as much as one half of newly
diagnosed diabetes in children aged 10–21 yr, depend- The risk of DKA in established T1DM is 1–10% per
ing on the socioeconomic and ethnic composition of the patient per year (A, C) (26–29).
population (7). Acute decompensation with DKA has Risk is increased in (28):
been recognized to occur at the time of diagnosis in as
many as 25% of children with T2DM (7). This is more ¤ Children with poor metabolic control or previous
likely in those of African-American descent, less so in episodes of DKA
Hispanic, and least in Canadian First Nation teenagers ¤ Peripubertal and adolescent girls
(8–13). The majority of new cases of diabetes in ¤ Children with psychiatric disorders, including those
Japanese children and adolescents are detected in with eating disorders
asymptomatic individuals by routine urine screening ¤ Children with difficult or unstable family circum-
(14, 15); however, overall, approximately 5% of patients stances (e.g., parental abuse)
with T2DM have DKA at the time of diagnosis (16). ¤ Children who omit insulin (30) (C)
The severity of DKA is categorized by the degree of ¤ Children with limited access to medical services
acidosis (17): ¤ Insulin pump therapy (as only rapid-acting or
short-acting insulin is used in pumps, interruption
¤ Mild: venous pH , 7.3 or bicarbonate , 15 mmol/L of insulin delivery for any reason rapidly leads to
¤ Moderate: pH , 7.2, bicarbonate , 10 mmol/L insulin deficiency) (29) (C)
¤ Severe: pH , 7.1, bicarbonate , 5 mmol/L
Hyperglycemic hyperosmolar state (HHS) also may
occur in young patients with T2DM (18, 19). The Management of DKA
criteria for HHS include (2):
Emergency assessment
¤ Plasma glucose concentration . 33.3 mmol/L (600
¤ Perform a clinical evaluation to confirm the
diagnosis and determine its cause. Carefully look
¤ Arterial pH . 7.30
for evidence of infection. In recurrent DKA,
¤ Serum bicarbonate . 15 mmol/L
insulin omission or failure to follow sick day or
¤ Small ketonuria, absent to mild ketonemia (serum
pump failure management guidelines accounts for
b-hydroxybutyrate 1  0.2 (SEM) mmol/L (2))
almost all episodes, except for those because of
¤ Effective serum osmolality . 320 mOsm/kg
acute severe febrile or gastrointestinal illness.
¤ Stupor or coma
¤ Weigh the patient. (If body surface area is used for
It is important to recognize that overlap between the fluid therapy calculations, measure height or length
characteristic features of HHS and DKA may occur. to determine surface area.) This weight should be
Some patients with HHS, especially when there is very used for calculations and not the weight from
severe dehydration, have mild or moderate acidosis. a previous office visit or hospital record.
Conversely, some children with T1DM may have ¤ Assess clinical severity of dehydration.
features of HHS (severe hyperglycemia) if high- ¤ Clinical assessment of dehydration is imprecise,
carbohydrate-containing beverages have been used inaccurate, and generally shows only fair to
to quench thirst and replace urinary losses prior to moderate agreement among examiners. It should
diagnosis (3). Therapy must be appropriately modi- be based on a combination of physical signs. The
fied to address the pathophysiology and unique three most useful individual signs for assessing
biochemical disturbances of each individual patient. dehydration in young children and predicting at
least 5% dehydration and acidosis are:
Frequency of DKA
s prolonged capillary refill time (normal capillary
At disease onset
refill is 1.5–2 s)
There is wide geographic variation in the frequency of s abnormal skin turgor (Ôtenting’ or inelastic skin)
DKA at onset of diabetes; rates inversely correlate with s abnormal respiratory pattern (hyperpnea) (31)
the regional incidence of T1DM. Frequencies range
from approximately 15% to 70% in Europe and ¤ Other useful signs in assessing degree of dehydra-
North America (A) (20–24). DKA at diagnosis is more tion include dry mucus membranes, sunken eyes,
common in younger children (,5 yr of age), and in absent tears, weak pulses, and cool extremities.
children whose families do not have ready access to More signs of dehydration tend to be associated
medical care for social or economic reasons (A) (6, 24, 25). with more severe dehydration (31).
30 Pediatric Diabetes 2007: 8: 28–43
Consensus Guidelines

s 10% dehydration is suggested by the presence because the history of recent insulin administration
of weak or impalpable peripheral pulses, hypo- may not be available or accurate (37).
tension, and oliguria.
Supportive measures
¤ Assess level of consciousness [Glasgow coma scale
(GCS) – see Table 2] (32). ¤ Secure the airway and empty the stomach by
¤ Obtain a blood sample for laboratory measurement continuous nasogastric suction to prevent pulmo-
of serum or plasma glucose, electrolytes (including nary aspiration in the unconscious or severely
bicarbonate or total carbon dioxide), blood urea obtunded patient.
nitrogen, creatinine, osmolality, venous (or arterial ¤ A peripheral intravenous (IV) catheter should be
in critically ill patient) pH, pCO2, hemoglobin and placed for convenient and painless repetitive blood
hematocrit or complete blood count (an increased sampling. An arterial catheter may be necessary in
white blood cell count in response to stress is some critically ill patients managed in an intensive
characteristic of DKA and is not indicative of care unit.
infection (22). The white cell count need not be ¤ A cardiac monitor should be used for continuous
measured unless there is evidence of concurrent electrocardiographic monitoring to assess T waves
infection), calcium, phosphorus, and magnesium for evidence of hyper- or hypokalemia (35, 36).
concentrations (if possible), HbA1c. ¤ Give oxygen to patients with severe circulatory
¤ Perform a urinalysis for ketones. impairment or shock.
¤ Measurement of blood b-hydroxybutyrate concen- ¤ Give antibiotics to febrile patients after obtaining
tration, if available, is useful to confirm ketoaci- appropriate cultures of body fluids.
dosis and may be used to monitor the response to ¤ Catheterization of the bladder is usually not
treatment (33, 34). necessary, but if the child is unconscious or unable
¤ Obtain appropriate specimens for culture (blood, to void on demand (e.g., infants and very ill young
urine, throat), if there is evidence of infection. children) the bladder should be catheterized.
¤ If laboratory measurement of serum potassium is
Where should the child be managed? The child
delayed, perform an electrocardiogram (ECG) for
should receive care in a unit that has:
baseline evaluation of potassium status (35, 36).
¤ In the child with established diabetes (recurrent
¤ Experienced nursing staff trained in monitoring
DKA) and suspected insulin omission, it may be
and management
useful to obtain a blood sample for determination of
¤ Written guidelines for DKA management in
serum free insulin concentration (even when children
have been treated exclusively with biosynthetic
¤ Access to laboratories for frequent and timely
human insulin, anti-insulin antibodies may be present
evaluation of biochemical variables
in the serum and interfere with insulin immunoassays.
Thus anti-insulin antibodies must be eliminated A specialist/consultant pediatrician with training
before the concentration of free (active) insulin is and expertise in the management of DKA should
determined (27)) before insulin is administered direct inpatient management.

Table 2. Glasgow coma scale or score. The GCS consists of three parameters and is scored between 3 and 15 (32); 3
being the worst and 15 the best. One of the components of the GCS is the best verbal response, which cannot be
assessed in non-verbal young children. A modification of the GCS was created for children too young to talk.
Best eye response Best verbal response Best verbal response Best motor response
(non-verbal children)
1. No eye opening 1. No verbal response 1. No response 1. No motor response
2. Eyes open to pain 2. No words, only 2. Inconsolable, irritable, 2. Extension to pain
3. Eyes open to incomprehensible restless, cries (decerebrate posture)
verbal command sounds; moaning 3. Inconsistently consolable 3. Flexion to pain
4. Eyes open and groaning and moans; makes vocal (decorticate posture)
spontaneously 3. Words, but incoherent* sounds 4. Withdrawal from pain
4. Confused, disoriented 4. Consolable when 5. Localizes pain
conversation† crying and interacts 6. Obeys commands
5. Orientated, normal inappropriately
conversation 5. Smiles, oriented to
sound, follows objects
and interacts
*Inappropriate words, no sustained conversational exchange.
†Attention can be held; responds in a conversational manner, but shows some disorientation.

Pediatric Diabetes 2007: 8: 28–43 31

Wolfsdorf et al.

Immediate assessment
Clinical Signs Biochemical Features &
Assess dehydration
Deep sighing respiration (Kussmaul) Ketones in urine
Clinical History Elevated blood glucose
Smell of ketones
Polyuria Lethargy/drowsiness ± vomiting Acidemia
Polydipsia Blood gases, urea, electrolytes
Weight loss (Weigh) Other investigations as
Abdominal pain indicated
Diagnosis Confirmed
Confusion Diabetic Ketoacidosis (DKA)
Contact Senior Staff

Shock (reduced peripheral pulses) Dehydration >5% Minimal dehydration

Reduced conscious level/coma Not in shock Tolerating oral fluid
Acidotic (hyperventilation)

Resuscitation IV Therapy Therapy

Airway ± NG tube Calculate fluid requirements Start with SC insulin
Breathing (100% oxygen) Correct over 48 hours Continue oral hydration
Circulation (0.9% saline Saline 0.9%
10-20 ml/kg over 1-2 h. ECG for abnormal T-waves
& repeat until circulation is Add KCL 40 mmol per litre fluid
restored) but do not exceed 30
ml/kg No improvement

Continuous insulin infusion

0.1 unit/kg/hour

Critical Observations
Hourly blood glucose
Hourly fluid input & output
Neurological status at least hourly
Electrolytes 2 hourly after start of IV therapy
Monitor ECG for T-wave changes

Acidosis not improving Blood glucose 17 mmol/l Neurological deterioration

blood glucose falls >5 mmol/l/hour headache, slowing heart
rate, irritability,
decreased conscious
level, incontinence,
IV Therapy specific neurological
Re- ev aluate Change to 0.45% saline + 5% glucose signs
Adjust sodium infusion to promote an
IV fluid calculations increase in measured serum sodium
Insulin delivery system & dose Exclude hypoglycaemia
Need for additional resuscitation Is it cerebral edema?
Consider sepsis
Clinicallywell, tolerating oral fluids
Give mannitol 0.5-1 g/kg
Transition to SC Insulin Restrict IV fluids by one-third
Start SC insulin then stop IV insulin after an Call senior staff
appropriate interval Move to ICU
Consider cranial imaging
only after patient stabilised
Adapted from Dunger et al. Karger Pub.1999. NG, nasogastric, SC,subcutaneous

Fig. 2. Algorithm for the immediate assessment and management of diabetic ketoacidosis (DKA).

32 Pediatric Diabetes 2007: 8: 28–43

Consensus Guidelines

Children with severe DKA (long duration of hematocrit, and blood gases should be repeated
symptoms, compromised circulation, or depressed 2–4 hourly, or more frequently, as clinically indi-
level of consciousness) or those who are at increased cated, in more severe cases
risk for cerebral edema (e.g., ,5 yr of age, severe ¤ Urine ketones until cleared or blood b-hydroxybu-
acidosis, low pCO2, high blood urea nitrogen) should tyrate (BOHB) concentrations, if available, every 2
be considered for immediate treatment in an intensive h (34)
care unit (pediatric, if available) or in a unit that has ¤ If the laboratory cannot provide timely results, a
equivalent resources and supervision, such as a child- portable biochemical analyzer that measures plasma
ren’s ward specializing in diabetes care (C,E) (4, 38). glucose, serum electrolytes and blood ketones on
In a child with established diabetes, whose parents have fingerstick blood samples at the bedside is a useful
been trained in sick day management, hyperglycemia adjunct to laboratory-based determinations.
and ketosis without vomiting or severe dehydration can ¤ Calculations:
be managed at home or in an outpatient health care
facility (e.g., emergency ward), provided an experienced s Anion gap ¼ Na 2 (Cl 1 HCO3): normal is 12 
diabetes team supervises the care (C,E) (17, 39, 40). 2 mmol/L
j In DKA the anion gap is typically 20–30 mmol/L;
Children and adolescents with DKA should be an anion gap .35 mmol/L suggests concomi-
managed in centers experienced in its treatment tant lactic acidosis (E)
and where vital signs, neurological status and
laboratory results can be monitored frequently
s Corrected sodium ¼ measured Na 1 2 3 ([plasma
glucose 2 5.6]/5.6) mmol/L
s Effective osmolality ¼ 2 3 (Na 1 K) 1 plasma
Clinical and biochemical monitoring glucose mOsm/kg
Successful management of DKA and HHS demands
meticulous monitoring of the patient’s clinical and Goals of therapy
biochemical response to treatment so that timely ¤ Correct dehydration
adjustments in treatment can be made when indicated ¤ Correct acidosis and reverse ketosis
by the patient’s clinical or laboratory data (E). ¤ Restore blood glucose to near normal
¤ Avoid complications of therapy
There should be documentation on a flow chart of ¤ Identify and treat any precipitating event
hour-by-hour clinical observations, IV and oral
medications, fluids, and laboratory results. Monitor-
Fluids and salt. Patients with DKA have a deficit in
ing should include the following:
ECF volume that usually is in the range 5–10% (C)
¤ Hourly (or more frequently as indicated) vital signs (41, 42). Shock with hemodynamic compromise is rare
(heart rate, respiratory rate, blood pressure) in pediatric DKA. Clinical estimates of the volume
¤ Hourly (or more frequently as indicated) neurolog- deficit are subjective and inaccurate (43, 44); there-
ical observations (Glasgow coma score) for warning fore, in moderate DKA use 5–7% and in severe DKA
signs and symptoms of cerebral edema (see below) 7–10% dehydration.
The effective osmolality (formula above) is fre-
s headache
quently in the range of 300–350 mmol/kg. Increased
s inappropriate slowing of heart rate
serum urea nitrogen and hematocrit may be useful
s recurrence of vomiting
markers of the severity of ECF contraction (40, 45).
s change in neurological status (restlessness, irri-
The serum sodium concentration is an unreliable
tability, increased drowsiness, incontinence) or
measure of the degree of ECF contraction for two
specific neurologic signs (e.g., cranial nerve
reasons: (1) glucose, largely restricted to the extracel-
palsies, abnormal pupillary responses)
lular space, causes osmotic movement of water into
s rising blood pressure
the extracellular space thereby inducing dilutional
s decreased oxygen saturation
hyponatremia (46, 47), and (2) the low sodium content
¤ Amount of administered insulin of the elevated lipid fraction of the serum in DKA.
¤ Hourly (or more frequently as indicated) accurate The latter is not a concern with most modern methods
fluid input (including all oral fluid) and output for measuring sodium. Therefore, it is important to
¤ Capillary blood glucose should be measured hourly calculate the corrected sodium (using the above
(but must be cross-checked against laboratory venous formula) and monitor its changes throughout the
glucose, as capillary methods may be inaccurate in the course of therapy. As the plasma glucose concentra-
presence of poor peripheral circulation and acidosis) tion decreases after administering fluid and insulin,
¤ Laboratory tests: serum electrolytes, glucose, blood the measured serum sodium concentration should
urea nitrogen, calcium, magnesium, phosphorus, increase, but it is important to appreciate that this
Pediatric Diabetes 2007: 8: 28–43 33
Wolfsdorf et al.

does not indicate a worsening of the hypertonic state. s Thereafter, deficit replacement should be with
A failure of measured serum sodium levels to rise or a solution that has a tonicity equal to or greater
a further decline in serum sodium levels with therapy than 0.45% saline with added potassium chlo-
is thought to be a potentially ominous sign of ride, potassium phosphate, or potassium acetate
impending cerebral edema (48–50). (see below under potassium replacement) (C,E)
The objectives of fluid and electrolyte replacement (45, 49, 53, 56, 57).
therapy are: s The rate of fluid (IV and oral) should be calculated
to rehydrate evenly over 48 h (C, E) (4, 45).
¤ Restoration of circulating volume s As the severity of dehydration may be difficult to
¤ Replacement of sodium and the ECF and intracel- determine and frequently is under- or over-
lular fluid deficit of water estimated (C) (44), infuse fluid each day at a rate
¤ Improved glomerular filtration with enhanced rarely in excess of 1.5–2 times the usual daily
clearance of glucose and ketones from the blood maintenance requirement based on age, weight,
¤ Reduction of risk of cerebral edema or body surface area (E) (4). See Table 3 for
examples of calculations.
Principles of water and salt replacement
¤ In addition to clinical assessment of dehydration,
Despite much effort to find the cause of cerebral edema
calculation of effective osmolality may be valuable
it remains a mystery. There is no convincing evidence of
to guide fluid and electrolyte therapy (E).
an association between the rate of fluid or sodium
¤ Urinary losses should not routinely be added to the
administration used in the treatment of DKA and the
calculation of replacement fluid, but may be
development of cerebral edema (51). No treatment
necessary in rare circumstances (E).
strategy can be definitively recommended as being
¤ The sodium content of the fluid may need to be
superior to another based on evidence. The principles
increased if measured serum sodium is low and
described below were developed after a comprehensive
does not rise appropriately as the plasma glucose
review of the literature and were accepted and endorsed
concentration falls (C) (48, 58).
by a panel of expert physicians representing the Lawson
¤ The use of large amounts of 0.9% saline has been
Wilkins Pediatric Endocrine Society (LWPES), the
associated with the development of hyperchloremic
European Society for Paediatric Endocrinology
metabolic acidosis (59, 60).
(ESPE), and the International Society for Pediatric
and Adolescent Diabetes (ISPAD) (4, 52).
Insulin therapy
Begin with fluid replacement before insulin therapy.
Volume expansion (resuscitation) is required only if
DKA is caused by a decrease in effective circulating
needed to restore peripheral circulation. Subsequent insulin associated with increases in counter-regulatory
fluid administration (including oral fluids) should hormones (glucagon, catecholamines, growth hor-
rehydrate evenly over 48 h at a rate rarely in excess of mone, cortisol). Although rehydration alone causes
1.5–2 times the usual daily maintenance. some decrease in blood glucose concentration (61, 62),
insulin therapy is essential to normalize blood glucose
¤ Water and salt deficits must be replaced (A). and suppress lipolysis and ketogenesis (A) (63).
¤ IV or oral fluids that may have been given in
another facility before assessment should be DKA is caused by either relative or absolute insulin
factored into calculation of deficit and repair (E). deficiency.
¤ If needed to restore peripheral circulation, volume Begin with 0.1 U/kg/h. 1–2 h AFTER starting fluid
expansion (resuscitation) should begin immediately replacement therapy.
with 0.9% saline (E).
s The volume and rate of administration depends Extensive evidence indicates that Ôlow dose’ IV
on circulatory status and, where it is clinically insulin administration should be the standard of care
indicated, the volume administered typically is (A) (64).
10–20 mL/kg over 1–2 h, and may be repeated, if
necessary (E). ¤ Start insulin infusion 1–2 h after starting fluid
s Use crystalloid not colloid (E). There are no data replacement therapy, i.e., after the patient has
to support the use of colloid in preference to received initial volume expansion (E,C) (65).
crystalloid in the treatment of DKA. ¤ Correction of insulin deficiency

¤ Subsequent fluid management (deficit replacement) s Dose: 0.1 unit/kg/h (for example, one method is
should be with 0.9% saline or Ringer’s acetate for to dilute 50 units regular [soluble] insulin in 50
at least 4–6 h (C,E) (45, 49, 53–55). mL normal saline, 1 unit ¼ 1 mL) (64, 66)
34 Pediatric Diabetes 2007: 8: 28–43
Consensus Guidelines

Table 3. Shows an alternative example of fluid volumes ¤ If the patient demonstrates marked sensitivity to
for the subsequent phase of rehydration insulin (e.g., some young children with DKA,
DKA: give maintenance 1 5% patients with HHS, and some older children with
of body weight/24 h established diabetes), the dose may be decreased to
0.05 unit/kg/h, or less, provided that metabolic
Body Maintenance acidosis continues to resolve.
weight kg mL/24 h mL/24 h mL/h
¤ During initial volume expansion the plasma glucose
4 325 530 22 concentration falls steeply (61) (C). Thereafter, and
5 405 650 27 after commencing insulin therapy, the plasma
6 485 790 33
7 570 920 38
glucose concentration typically decreases at a rate
8 640 1040 43 of 2–5 mmol/L/h, depending on the timing and
9 710 1160 48 amount of glucose administration (C) (69–75).
10 780 1280 53 ¤ To prevent an unduly rapid decrease in plasma
11 840 1390 58 glucose concentration and hypoglycemia, 5% glu-
12 890 1490 62
13 940 1590 66 cose should be added to the IV fluid (e.g., 5% glucose
14 990 1690 70 in 0.45% saline) when the plasma glucose falls to
15 1030 1780 74 approximately 14–17 mmol/L (250–300 mg/dL), or
16 1070 1870 78 sooner if the rate of fall is precipitous (B).
17 1120 1970 82
18 1150 2050 85
19 1190 2140 89 s It may be necessary to use 10% or even 12.5%
20 1230 2230 93 dextrose to prevent hypoglycemia while continuing
22 1300 2400 100 to infuse insulin to correct the metabolic acidosis.
24 1360 2560 107
26 1430 2730 114
28 1490 2890 120 ¤ If blood glucose falls very rapidly (.5 mmol/L/h)
30 1560 3060 128 after initial fluid expansion consider adding glucose
32 1620 3220 134 even before plasma glucose has decreased to 17
34 1680 3360 140
36 1730 3460 144 mmol/L (E).
38 1790 3580 149 ¤ If biochemical parameters of DKA (pH, anion gap)
40 1850 3700 154 do not improve, reassess the patient, review insulin
45 1980 3960 165 therapy, and consider other possible causes of
50 2100 4200 175
55 2210 4420 184
impaired response to insulin; e.g., infection, errors
60 2320 4640 193 in insulin preparation (E).
65 2410 4820 201 ¤ In circumstances where continuous IV administra-
70 2500 5000 208 tion is not possible, hourly, or 2-hourly subcutane-
75 2590 5180 216 ous (SC) or intramuscular (IM) administration of
80 2690 5380 224
a short- or rapid-acting insulin analog (insulin lispro
After initial resuscitation, and assuming 10% dehydration, or insulin aspart) is safe and may be as effective as IV
the total amount of fluid should be given over 48 h. The regular insulin infusion (C) (70, 76–79), but should
Table gives volumes for maintenance and rehydration per
24 h and per hour. If fluid has been given for resuscitation,
not be used in subjects whose peripheral circulation
the volume should not be subtracted from the amount is impaired (E).
shown in the Table. Fluids given orally (when patient has
improved) should be subtracted from the amount in the s Initial dose SC: 0.3 unit/kg, followed 1 h later by
table. The Table is based on maintenance volumes SC insulin lispro or aspart at 0.1 unit/kg every
according to Darrow (146). For body weights .32 kg,
the volumes have been adjusted so as not to exceed hour, or 0.15–0.20 units/kg every 2 h.
twice the maintenance rate of fluid administration. s If blood glucose falls to ,14 mmol/L (250 mg/dL)
before DKA has resolved (pH still ,7.30), add 5%
s Route of administration IV (A) glucose and continue with insulin as above.
s An IV bolus is unnecessary (67), may increase s When DKA has resolved and blood glucose is ,14
the risk of cerebral edema (65), and should not be mmol/L (250 mg/dL), reduce SC insulin lispro or
used at the start of therapy (C) aspart to 0.05 unit/kg/h to keep blood glucose 11
mmol/L (200 mg/dL) until resolution of DKA.
¤ The dose of insulin should usually remain at 0.1
unit/kg/h at least until resolution of DKA (pH .
7.30, bicarbonate . 15 mmol/L and/or closure of If the blood glucose concentration decreases too
quickly or too low before DKA has resolved,
the anion gap), which invariably takes longer than increase the amount of glucose administered. Do
normalization of blood glucose concentrations not decrease the insulin infusion.
(B) (68).
Pediatric Diabetes 2007: 8: 28–43 35
Wolfsdorf et al.

Potassium replacement potassium chloride and 20 mmol/L potassium

phosphate or 20 mmol/L potassium phosphate
Children with DKA suffer total body potassium deficits and 20 mmol/L potassium acetate (C,E).
of the order of 3–6 mmol/kg (41, 42, 80–82). The major ¤ Potassium replacement should continue through-
loss of potassium is from the intracellular pool. out IV fluid therapy (E).
Intracellular potassium is depleted because of trans- ¤ The maximum recommended rate of IV potassium
cellular shifts of this ion caused by hypertonicity replacement is usually 0.5 mmol/kg/h (E).
(increased plasma osmolality causes solvent drag in ¤ If hypokalemia persists despite maximum rate of
which water and potassium are drawn out of cells) and potassium replacement, then the rate of insulin
glycogenolysis and proteolysis secondary to insulin infusion can be reduced.
deficiency cause potassium efflux from cells. Potassium
is lost from the body from vomiting and as a conse- Phosphate
quence of osmotic diuresis. Volume depletion causes
secondary hyperaldosteronism, which promotes uri- Depletion of intracellular phosphate occurs in DKA and
nary potassium excretion. Thus, total body depletion of phosphate is lost as a result of osmotic diuresis (41, 42,
potassium occurs, but at presentation serum potassium 80). Plasma phosphate levels fall after starting treatment
levels may be normal, increased, or decreased (83). and this is exacerbated by insulin, which promotes entry
Renal dysfunction, by enhancing hyperglycemia and of phosphate into cells (87–89). Total body phosphate
reducing potassium excretion, contributes to hyper- depletion has been associated with a variety of metabolic
kalemia (83). Administration of insulin and the disturbances (90–92). Clinically significant hypophos-
correction of acidosis will drive potassium back into phatemia may occur if IV therapy without food intake is
the cells, decreasing serum levels (84). The serum prolonged beyond 24 h (41, 42, 80).
potassium concentration may decrease abruptly, pre-
disposing the patient to cardiac arrhythmias. ¤ Prospective studies have not shown clinical benefit
from phosphate replacement (A) (93–98).
Even with normal or high levels of serum potassium ¤ Severe hypophosphatemia in conjunction with
at presentation, there is always a total body deficit unexplained weakness should be treated (E) (99).
of potassium. Begin with 40 mmol potassium/L in the
infusate or 20 mmol potassium/L in the hypokalemic ¤ Administration of phosphate may induce hypocal-
patient receiving fluid at a rate .10 mL/kg/h. cemia (C) (100, 101).
¤ Potassium phosphate salts may be safely used as an
¤ Replacement therapy is required regardless of the alternative to or combined with potassium chloride
serum potassium concentration (A) (85, 86). or acetate, provided that careful monitoring of
¤ If the patient is hypokalemic, start potassium serum calcium is performed to avoid hypocalcemia
replacement at the time of initial volume expansion (C) (100, 101).
and before starting insulin therapy. Otherwise, start
replacing potassium after initial volume expansion
and concurrent with starting insulin therapy. If the
patient is hyperkalemic, defer potassium replace- Severe acidosis is reversible by fluid and insulin
ment therapy until urine output is documented (E). replacement; insulin stops further ketoacid production
¤ If immediate serum potassium measurements are and allows ketoacids to be metabolized, which
unavailable, an ECG may help to determine whether generates bicarbonate (A). Treatment of hypovolemia
the child has hyper- or hypokalemia (C) (35, 36). improves tissue perfusion and renal function increas-
Flattening of the T wave, widening of the QT interval, ing the excretion of organic acids.
and the appearance of U waves indicate hypokalemia. Controlled trials have shown no clinical benefit from
Tall, peaked, symmetrical, T waves and shortening of bicarbonate administration (B,C) (102–105). Bicarbon-
the QT interval are signs of hyperkalemia. ate therapy may cause paradoxical CNS acidosis (106,
¤ The starting potassium concentration in the in- 107); rapid correction of acidosis with bicarbonate
fusate should be 40 mmol/L. Subsequent potassium causes hypokalemia (106, 108, 109), and failure to
replacement therapy should be based on serum account for the sodium being administered and
potassium measurements (E). appropriately reducing the NaCl concentration of the
fluids can result in increasing osmolality (106). Never-
s If potassium is given with the initial rapid theless, there may be selected patients who may benefit
volume expansion, a concentration of 20 mmol/ from cautious alkali therapy. These include: patients
L should be used. with severe acidemia (arterial pH , 6.9) in whom
decreased cardiac contractility and peripheral vasodila-
¤ Potassium phosphate may be used together with tation can further impair tissue perfusion, and patients
potassium chloride or acetate; e.g., 20 mmol/L with life-threatening hyperkalemia (E) (110).
36 Pediatric Diabetes 2007: 8: 28–43
Consensus Guidelines

Other rare causes of morbidity and mortality include:

There is no evidence that bicarbonate is either
necessary or safe in DKA. ¤ Hypokalemia
¤ Hyperkalemia
¤ Bicarbonate administration is not recommended ¤ Severe hypophosphatemia
unless the acidosis is profound and likely to affect ¤ Hypoglycemia
adversely the action of adrenaline/epinephrine ¤ Other central nervous system complications (dis-
during resuscitation (A). seminated intravascular coagulation, dural sinus
¤ If bicarbonate is considered necessary, cautiously thrombosis, basilar artery thrombosis)
give 1–2 mmol/kg over 60 min (E). ¤ Peripheral venous thrombosis
¤ Sepsis
¤ Rhinocerebral or pulmonary mucormycosis
Complications of therapy ¤ Aspiration pneumonia
¤ Inadequate rehydration ¤ Pulmonary edema
¤ Hypoglycemia ¤ Adult respiratory distress syndrome (ARDS)
¤ Hypokalemia ¤ Pneumothorax, pneumomediastinum and SC
¤ Hyperchloremic acidosis
¤ Cerebral edema emphysema
¤ Rhabdomyolysis
¤ Acute renal failure
¤ Acute pancreatitis (115)
Introduction of oral fluids and transition to
SC insulin injections Cerebral edema
¤ Oral fluids should be introduced only when The incidence of cerebral edema in national popula-
substantial clinical improvement has occurred tion studies is 0.5–0.9% and the mortality rate is
(mild acidosis/ketosis may still be present) (E). 21–24% (50, 113, 114).
¤ When oral fluid is tolerated, IV fluid should be Demographic factors that have been associated with
reduced (E). an increased risk of cerebral edema include:
¤ When ketoacidosis has resolved, oral intake is ¤ Younger age (C) (116)
tolerated, and the change to SC insulin is planned, ¤ New onset diabetes (B) (112) (C) (116)
the most convenient time to change to SC insulin is ¤ Longer duration of symptoms (C) (117)
just before a mealtime (E).
These risk associations may reflect the greater
¤ To prevent rebound hyperglycemia the first SC
likelihood of severe DKA.
injection should be given 15–30 min (with rapid-
In addition, epidemiological studies have identified
acting insulin) or 1–2 h (with regular insulin) before
several potential risk factors at diagnosis or during
stopping the insulin infusion to allow sufficient
treatment of DKA. These include:
time for the insulin to be absorbed (E). With
intermediate- or long-acting insulin, the overlap ¤ Greater hypocapnia at presentation after adjusting
should be longer and the IV insulin gradually for degree of acidosis (C) (50, 118)
lowered. For example, for patients on a basal-bolus ¤ Increased serum urea nitrogen at presentation
insulin regimen, the first dose of basal insulin may (C) (50)
be administered in the evening and the insulin ¤ More severe acidosis at presentation (C) (65, 119)
infusion is stopped the next morning (E). ¤ Bicarbonate treatment for correction of acidosis
¤ The dose and type of SC insulin should be (C) (50, 120)
according to local preferences and circumstances. ¤ An attenuated rise in measured serum sodium
¤ After transitioning to SC insulin, frequent blood concentrations during therapy (C) (48–50)
glucose monitoring is required to avoid marked ¤ Greater volumes of fluid given in the first 4 h (65)
hyperglycemia and hypoglycemia (E). ¤ Administration of insulin in the first hour of fluid
treatment (65)

Morbidity and mortality Warning signs and symptoms of cerebral edema

In national population studies, the mortality rate from Headache & slowing of heart rate
DKA in children is 0.15–0.30% (C,B) (111, 112). Change in neurological status (restlessness, irritability,
Cerebral edema accounts for 60–90% of all DKA increased drowsiness, incontinence)
deaths (C,B) (50, 113). From 10% to 25% of survivors Specific neurological signs (e.g., cranial nerve palsies)
of cerebral edema have significant residual morbidity Rising blood pressure
Decreased O2 saturation
(C,B) (50, 113, 114).

Pediatric Diabetes 2007: 8: 28–43 37

Wolfsdorf et al.

Clinically significant cerebral edema usually devel- ¤ Elevate the head of the bed.
ops 4–12 h after treatment has started, but can occur ¤ Intubation may be necessary for the patient with
before treatment has begun (50, 114, 121–124) or, impending respiratory failure, but aggressive
uncommonly, may develop as late as 24–48 h after the hyperventilation (to a pCO2 , 2.9 kPa [22 mm
start of treatment (C,B) (50, 116, 125). Symptoms and Hg]) has been associated with poor outcome and is
signs are variable. A method of clinical diagnosis not recommended (C) (132).
based on bedside evaluation of neurological state is ¤ After treatment for cerebral edema has been
shown below (C) (126): started, a cranial CT scan should be obtained to
rule out other possible intracerebral causes of
Diagnostic criteria.
neurologic deterioration (10% of cases), espe-
¤ Abnormal motor or verbal response to pain cially thrombosis (133–136) or hemorrhage, which
¤ Decorticate or decerebrate posture may benefit from specific therapy.
¤ Cranial nerve palsy (especially III, IV, and VI)
¤ Abnormal neurogenic respiratory pattern (e.g.,
grunting, tachypnea, Cheyne-Stokes respiration,
Prevention of recurrent DKA
Management of an episode of DKA is not complete
Major criteria. until its cause has been identified and an attempt
¤ Altered mentation/fluctuating level of consciousness made to treat it.
¤ Sustained heart rate deceleration (decrease more
than 20 beats per minute) not attributable to All cases of recurrent DKA are preventable
improved intravascular volume or sleep state
¤ Age-inappropriate incontinence ¤ Insulin omission, either inadvertently or deliber-
ately, is the cause in most cases (C, A) (28, 30).
Minor criteria. ¤ The most common cause of DKA in insulin pump
users is failure to take extra insulin with a pen or
¤ Vomiting
syringe when hyperglycemia and hyperketonemia
¤ Headache
or ketonuria occur (E).
¤ Lethargy or not easily arousable
¤ Home measurement of blood BOHB concentra-
¤ Diastolic blood pressure . 90 mm Hg
tions, when compared with urine ketone testing,
¤ Age ,5 yr
decreases diabetes-related hospital visits (both
One diagnostic criterion, two major criteria, or one emergency department visits and hospitalizations)
major and two minor criteria have a sensitivity of 92% by the early identification and treatment of ketosis
and a false positive rate of only 4%. (137). Blood BOHB measurements may be espe-
A chart with the reference ranges for blood pressure cially valuable to prevent DKA in patients who use
and heart rate, which vary depending on height, a pump because interrupted insulin delivery rapidly
weight, and gender, should be readily available, either leads to ketosis (E).
in the patient’s chart or at the bedside.
s There may be dissociation between urine ketone
Have mannitol or hypertonic saline at the bedside and serum BOHB concentrations, which may be
and the dose to be given calculated beforehand. In increased to levels consistent with DKA when
case of profound neurological symptoms, mannitol a urine ketone test is negative or shows only trace
should be given immediately. or small ketonuria (138).

¤ There usually is an important psychosocial reason

Treatment of cerebral edema for insulin omission.
¤ Initiate treatment as soon as the condition is s an attempt to lose weight in an adolescent girl
suspected. with an eating disorder,
¤ Give mannitol 0.5–1 g/kg IV over 20 min and s a means of escaping an intolerable or abusive
repeat if there is no initial response in 30 min to 2 h home situation,
(C,E) (127–129). s clinical depression or other reason for inability of
¤ Reduce the rate of fluid administration by one-third. the patient to manage the diabetes unassisted.
¤ Hypertonic saline (3%), 5–10 mL/kg over 30 min,
may be an alternative to mannitol, especially if there ¤ An infection that is not associated with vomiting
is no initial response to mannitol (C) (130, 131). and diarrhea is seldom the cause when the patient/
s Mannitol or hypertonic saline should be avail- family is properly educated in diabetes manage-
able at the bedside. ment and is receiving appropriate follow-up care by
38 Pediatric Diabetes 2007: 8: 28–43
Consensus Guidelines

a diabetes team with a 24-h telephone helpline (B) 8. PINHAS-HAMIEL O, DOLAN LM, ZEITLER PS. Diabetic
(139–141). ketoacidosis among obese African-American adolescents
with NIDDM. Diabetes Care 1997: 20: 484–486.
¤ A psychiatric social worker or clinical psychologist 9. SCOTT CR, SMITH JM, CRADOCK MM, PIHOKER C.
should be consulted to identify the psychosocial Characteristics of youth-onset noninsulin-dependent
reason(s) contributing to development of DKA (E). diabetes mellitus and insulin-dependent diabetes melli-
¤ Insulin omission can be prevented by schemes that tus at diagnosis. Pediatrics 1997: 100: 84–91.
provide education, psychosocial evaluation and VADHEIM CM. Early presentation of type 2 diabetes
treatment combined with adult supervision of in Mexican-American youth. Diabetes Care 1998: 21:
insulin administration (B) (142). 80–86.
11. SELLERS EA, DEAN HJ. Diabetic ketoacidosis: a com-
s Parents and patients should learn how to recognize plication of type 2 diabetes in Canadian aboriginal
youth. Diabetes Care 2000: 23: 1202–1204.
and treat impending DKA with additional rapid- 12. LIPTON R, KEENAN H, ONYEMERE KU, FREELS S.
or short-acting insulin and oral fluids (E) Incidence and onset features of diabetes in African-
s Patients should have access to a 24-h telephone American and Latino children in Chicago, 1985–1994.
helpline for emergency advice and treatment (B) Diabetes Metab Res Rev 2002: 18: 135–142.
(139) and course of Type 2 diabetes in youth in a large multi-
s When a responsible adult administers insulin ethnic city. Diabet Med 2004: 21: 1144–1148.
there may be as much as a tenfold reduction in 14. YOKOTA Y, KIKUCHI N, MATSUURA N. Screening for
frequency of recurrent DKA (B) (142). diabetes by urine glucose testing at school in Japan.
Pediatr Diabetes 2004: 5: 212–218.
This article is a Chapter in the ISPAD Clinical Practice 15. URAKAMI T, KUBOTA S, NITADORI Y, HARADA K,
Consensus Guidelines 2006–2007 of the International OWADA M, KITAGAWA T. Annual incidence and clinical
characteristics of type 2 diabetes in children as detected
Society for Pediatric and Adolescent Diabetes (ISPAD, by urine glucose screening in the Tokyo metropolitan The complete set of these Guidelines area. Diabetes Care 2005: 28: 1876–1881.
will later be published as a compendium. Additional 16. SUGIHARA S, SASAKI N, KOHNO H et al. Survey of
comments, clarifications or corrections should be current medical treatments for childhood-onset type 2
diabetes mellitus. Jpn Clin Pediatr Endocrinol 2005:
directed to the Corresponding Author. 14: 65–75.
The evidence grading system used in the ISPAD 17. CHASE HP, GARG SK, JELLEY DH. Diabetic ketoaci-
Guidelines is the same as that used by the American dosis in children and the role of outpatient manage-
Diabetes Association. See the Introduction of the ment. Pediatr Rev 1990: 11: 297–304.
ISPAD Clinical Practice Consensus Guidelines in 18. MORALES AE, ROSENBLOOM AL. Death caused by
hyperglycemic hyperosmolar state at the onset of type
Pediatric Diabetes 2006: 7: 341–342. 2 diabetes. J Pediatr 2004: 144: 270–273.
KIRK J. Childhood diabetes presenting with hyper-
References osmolar dehydration but without ketoacidosis: a report
of three cases. Diabet Med 2005: 22: 645–647.
1. FOSTER DW, MCGARRY JD. The metabolic derange- 20. LEVY-MARCHAL C, PAPOZ L, DE BEAUFORT C et al.
ments and treatment of diabetic ketoacidosis. N Engl J Clinical and laboratory features of type 1 diabetic
Med 1983: 309: 159–169. children at the time of diagnosis. Diabet Med 1992: 9:
2. KITABCHI A, UMPIERREZ G, MURPHY M et al. Manage- 279–284.
ment of hyperglycemic crises in patients with diabetes. 21. KOMULAINEN J, LOUNAMAA R, KNIP M, KAPRIO EA,
Diabetes Care 2001: 24: 131–153. AKERBLOM HK. Ketoacidosis at the diagnosis of type 1
3. MCDONNELL CM, PEDREIRA CC, VADAMALAYAN B, (insulin dependent) diabetes mellitus is related to
CAMERON FJ, WERTHER GA. Diabetic ketoacidosis, poor residual beta cell function. Childhood Diabetes
hyperosmolarity and hypernatremia: are high-carbo- in Finland Study Group. Arch Dis Child 1996: 75:
hydrate drinks worsening initial presentation? Pediatr 410–415.
Diabetes 2005: 6: 90–94. 22. LEVY-MARCHAL C, PATTERSON CC, GREEN A. Geo-
4. DUNGER DB, SPERLING MA, ACERINI CL et al. ESPE/ graphical variation of presentation at diagnosis of type
LWPES consensus statement on diabetic ketoacidosis I diabetes in children: the EURODIAB study. Euro-
in children and adolescents. Arch Dis Child 2004: 89: pean and Diabetes. Diabetologia 2001: 44(Suppl. 3):
188–194. B75–B80.
5. BURGE MR, HARDY KJ, SCHADE DS. Short-term fasting 23. HANAS R, LINDBLAD B, LINDGREN F. Diabetic ketoa-
is a mechanism for the development of euglycemic cidosis and cerebral edema in Sweden – a 2-year
ketoacidosis during periods of insulin deficiency. J Clin population study. Diabetes 2004: 53(Suppl. 2): A421.
Endocrinol Metab 1993: 76: 1192–1198. 24. REWERS A, KLINGENSMITH G, DAVIS C et al. Diabetic
6. PINKEY JH, BINGLEY PJ, SAWTELL PA, DUNGER DB, ketoacidosis at onset of diabetes: the SEARCH for
GALE EA. Presentation and progress of childhood diabetes in youth study. Diabetes 2005: 54(Suppl. 1):
diabetes mellitus: a prospective population-based A63.
study. The Bart’s-Oxford Study Group. Diabetologia 25. KOMULAINEN J, KULMALA P, SAVOLA K et al. Clinical,
1994: 37: 70–74. autoimmune, and genetic characteristics of very young
7. AMERICAN DIABETES ASSOCIATION. Type 2 diabetes in children with type 1 diabetes. Childhood Diabetes in
children and adolescents. (Consensus statement). Dia- Finland (DiMe) Study Group. Diabetes Care 1999: 22:
betes Care 2000: 23: 381–389. 1950–1955.

Pediatric Diabetes 2007: 8: 28–43 39

Wolfsdorf et al.

26. ROSILIO M, COTTON JB, WIELICZKO MC et al. Factors 45. HARRIS GD, FIORDALISI I. Physiologic management of
associated with glycemic control. A cross-sectional diabetic ketoacidemia. A 5-year prospective pediatric
nationwide study in 2,579 French children with experience in 231 episodes. Arch Pediatr Adolesc Med
type 1 diabetes. The French Pediatric Diabetes 1994: 148: 1046–1052.
Group. Diabetes Care 1998: 21: 1146–1153 (See 46. KATZ MA. Hyperglycemia-induced hyponatremia –
comments) calculation of expected serum sodium depression. N
27. SMITH CP, FIRTH D, BENNETT S, HOWARD C, CHISHOLM Engl J Med 1973: 289: 843–844.
P. Ketoacidosis occurring in newly diagnosed and 47. HILLIER TA, ABBOTT RD, BARRETT EJ. Hyponatremia:
established diabetic children. Acta Paediatr 1998: 87: evaluating the correction factor for hyperglycemia. Am
537–541. J Med 1999: 106: 399–403.
acute complications in children with type 1 diabetes. FINBERG L. Minimizing the risk of brain herniation
JAMA 15 2002: 287: 2511–2518. during treatment of diabetic ketoacidemia: a retrospec-
29. HANAS R, LINDBLAD B, LINDGREN F. Predisposing tive and prospective study. J Pediatr 1990: 117: 22–31.
conditions and insulin pump use in a 2-year population 49. HALE PM, REZVANI I, BRAUNSTEIN AW, LIPMAN TH,
study of pediatric ketoacidosis in Sweden. Diabetes MARTINEZ N, GARIBALDI L. Factors predicting cerebral
2005: 54(Suppl. 1): A455. edema in young children with diabetic ketoacidosis
30. MORRIS AD, BOYLE DI, MCMAHON AD, GREENE SA, and new onset type I diabetes. Acta Paediatr 1997: 86:
MACDONALD TM, NEWTON RW. Adherence to insulin 626–631.
treatment, glycaemic control, and ketoacidosis in 50. GLASER N, BARNETT P, MCCASLIN I et al. Risk factors
insulin-dependent diabetes mellitus. The DARTS/ for cerebral edema in children with diabetic ketoaci-
MEMO Collaboration. Diabetes Audit and Research dosis. The Pediatric Emergency Medicine Collabora-
in Tayside Scotland. Medicines Monitoring Unit. tive Research Committee of the American Academy of
Lancet 1997: 350: 1505–1510. Pediatrics. N Engl J Med 2001: 344: 264–269.
31. STEINER MJ, DEWALT DA, BYERLEY JS. Is this child 51. BROWN TB. Cerebral oedema in childhood diabetic
dehydrated? JAMA 2004: 291: 2746–2754. ketoacidosis: is treatment a factor? Emerg Med J 2004:
32. TEASDALE G, JENNETT B. Assessment of coma and 21: 141–144.
impaired consciousness. A practical scale. Lancet 1974: 52. DUNGER DB, SPERLING MA, ACERINI CL et al.
2: 81–84. European Society for Paediatric Endocrinology/Law-
33. WIGGAM MI, O’KANE MJ, HARPER R et al. Treatment son Wilkins Pediatric Endocrine Society consensus
of diabetic ketoacidosis using normalization of blood statement on diabetic ketoacidosis in children and
3-hydroxybutyrate concentration as the endpoint of adolescents. Pediatrics 2004: 113: e133–e140.
emergency management. A randomized controlled 53. ADROGUE HJ, BARRERO J, EKNOYAN G. Salutary effects
study. Diabetes Care 1997: 20: 1347–1352. of modest fluid replacement in the treatment of adults
34. HAM MR, OKADA P, WHITE PC. Bedside ketone with diabetic ketoacidosis. Use in patients without
determination in diabetic children with hyperglycemia extreme volume deficit. JAMA 1989: 262: 2108–2113.
and ketosis in the acute care setting. Pediatr Diabetes 54. MEL JM, WERTHER GA. Incidence and outcome of
2004: 5: 39–43. diabetic cerebral oedema in childhood: are there
35. MALONE JI, BRODSKY SJ. The value of electrocardio- predictors? J Paediatr Child Health 1995: 31: 17–20.
gram monitoring in diabetic ketoacidosis. Diabetes 55. WAGNER A, RISSE A, BRILL HL et al. Therapy of severe
Care 1980: 3: 543–547. diabetic ketoacidosis. Zero-mortality under very-low-
36. SOLER NG, BENNETT MA, FITZGERALD MG, MALINS JM. dose insulin application. Diabetes Care 1999: 22: 674–
Electrocardiogram as a guide to potassium replacement 677.
in diabetic ketoacidosis. Diabetes 1974: 23: 610–615. 56. ROTHER KI, SCHWENK WFN. Effect of rehydration
37. MALONE JI, ROOT AW. Plasma free insulin concen- fluid with 75 mmol/L of sodium on serum sodium
trations: keystone to effective management of diabetes concentration and serum osmolality in young patients
mellitus in children. J Pediatr 1981: 99: 862–867. with diabetic ketoacidosis. Mayo Clin Proc 1994: 69:
38. MONROE KW, KING W, ATCHISON JA. Use of PRISM 1149–1153.
scores in triage of pediatric patients with diabetic 57. FELNER EI, WHITE PC. Improving management of
ketoacidosis. Am J Manag Care 1997: 3: 253–258. diabetic ketoacidosis in children. Pediatrics 2001: 108:
Outpatient management of diabetic ketoacidosis. Am J 58. DUCK SC, WYATT DT. Factors associated with brain
Dis Child 1988: 142: 448–450. herniation in the treatment of diabetic ketoacidosis. J
40. LINARES MY, SCHUNK JE, LINDSAY R. Laboratory Pediatr 1988: 113: 10–14.
presentation in diabetic ketoacidosis and duration of 59. ADROGUE HJ, EKNOYAN G, SUKI WK. Diabetic
therapy. Pediatr Emerg Care 1996: 12: 347–351. ketoacidosis: role of the kidney in the acid-base
41. ATCHLEY D, LOEB R, RICHARDS D Jr, BENEDICT E, homeostasis re-evaluated. Kidney Int 1984: 25: 591–
DRISCOLL M. On diabetic ketoacidosis: a detailed study 598.
of electrolyte balances following the withdrawal and 60. OH MS, CARROLL HJ, URIBARRI J. Mechanism of
reestablishment of insulin therapy. J Clin Invest 1933: normochloremic and hyperchloremic acidosis in dia-
12: 297–326. betic ketoacidosis. Nephron 1990: 54: 1–6.
in severe diabetic ketosis. Q J Med 1952: 82: 225–248. BRATUSCH-MARRAIN P, NOWOTNY P. Severe hypergly-
43. MACKENZIE A, BARNES G, SHANN F. Clinical signs of cemia: effects of rehydration on endocrine derange-
dehydration in children. Lancet 1989: 2: 605–607. ments and blood glucose concentration. Diabetes 1979:
44. KOVES IH, NEUTZE J, DONATH S et al. The accuracy of 28: 577–584.
clinical assessment of dehydration during diabetic 62. OWEN OE, LICHT JH, SAPIR DG. Renal function and
ketoacidosis in childhood. Diabetes Care 2004: 27: effects of partial rehydration during diabetic ketoaci-
2485–2487. dosis. Diabetes 1981: 30: 510–518.

40 Pediatric Diabetes 2007: 8: 28–43

Consensus Guidelines

DEFRONZO RA. Metabolic effects of low-dose insulin GREENMAN L. Studies in diabetic acidosis and coma,
therapy on glucose metabolism in diabetic ketoacido- with particular emphasis on the retention of adminis-
sis. Diabetes 1988: 37: 1470–1477. tered potassium. J Clin Invest 1949: 28: 1–9.
64. KITABCHI AE. Low-dose insulin therapy in diabetic 82. D ARROW D, P RATT E. Retention of water and
ketoacidosis: fact or fiction? Diabetes Metab Rev 1989: electrolyte during recovery in a patient with diabetic
5: 337–363. acidosis. J Pediatr 1952: 41: 688–696.
65. EDGE J, JAKES R, ROY Y et al. The UK case-control 83. ADROGUE HJ, LEDERER ED, SUKI WN, EKNOYAN G.
study of cerebral oedema complicating diabetic Determinants of plasma potassium levels in diabetic
ketoacidosis in children. Diabetologia 2006: 49: 2002– ketoacidosis. Medicine (Baltimore) 1986: 65: 163–172.
66. SCHADE DS, EATON RP. Dose response to insulin in Effect of graded doses of insulin on splanchnic and
man: differential effects on glucose and ketone body peripheral potassium metabolism in man. Am J Physiol
regulation. J Clin Endocrinol Metab 1977: 44: 1038– 1980: 238:E421–427.
1053. 85. TATTERSALL RB. A paper which changed clinical practice
67. LINDSAY R, BOLTE RG. The use of an insulin bolus in (slowly). Jacob Holler on potassium deficiency in
low-dose insulin infusion for pediatric diabetic ketoa- diabetic acidosis (1946). Diabet Med 1999: 16: 978–984.
cidosis. Pediatr Emerg Care 1989: 5: 77–79. 86. NABARRO JD, SPENCER AG, STOWERS JM. Treatment of
68. SOLER NG, FITZGERALD MG, WRIGHT AD, MALINS JM. diabetic ketosis. Lancet 17 1952: 1: 983–989.
Comparative study of different insulin regimens in 87. GUEST G. Organic phosphates of the blood and mineral
management of diabetic ketoacidosis. Lancet 1975: 2: metabolism in diabetic acidosis. Am J Dis Child 1942:
1221–1224. 64: 401–412.
69. MARTIN MM, MARTIN AA. Continuous low-dose 88. GUEST G, RAPOPORT S. Electrolytes of blood plasma
infusion of insulin in the treatment of diabetic and cells in diabetic acidosis and during recovery. Proc
ketoacidosis in children. J Pediatr 1976: 89: 560–564. Am Diabetes Assoc 1947: 7: 95–115.
ketoacidosis: low-dose insulin therapy by various infusion on plasma phosphate in diabetic patients.
routes. N Engl J Med 1977: 297: 238–241. Metabolism 1979: 28: 191–194.
intravenous insulin infusion in patients with diabetic TD. 2,3-Diphosphoglycerate and tissue oxygenation in
ketoacidosis: biochemical effects in children. Pediatrics uncontrolled diabetes mellitus. Lancet 26 1972: 2: 391–
1977: 60: 681–688. 395.
72. KAPPY MS, LIGHTNER ES. Low-dose intravenous 91. KNOCHEL JP. The pathophysiology and clinical char-
insulin in the treatment of diabetic ketoacidosis. Am acteristics of severe hypophosphatemia. Arch Intern
J Dis Child 1979: 133: 523–525. Med 1977: 137: 203–220.
MCENERY PT, KNOWLES HC. Low-dose intravenous hypophosphatemia on myocardial performance in
insulin infusion versus subcutaneous insulin injection: man. N Engl J Med 1977: 297: 901–903.
a controlled comparative study of diabetic ketoacido- 93. GIBBY OM, VEALE KE, H AYES TM, J ONES JG,
sis. Pediatrics 1977: 59: 733–738. WARDROP CA. Oxygen availability from the blood
74. EDWARDS GA, KOHAUT EC, WEHRING B, HILL LL. and the effect of phosphate replacement on erythrocyte
Effectiveness of low-dose continuous intravenous 2,3-diphosphoglycerate and haemoglobin-oxygen
insulin infusion in diabetic ketoacidosis. A prospective affinity in diabetic ketoacidosis. Diabetologia 1978:
comparative study. J Pediatr 1977: 91: 701–705. 15: 381–385.
75. BURGHEN GA, ETTELDORF JN, FISHER JN, KITABCHI 94. KELLER U, BERGER W. Prevention of hypophosphate-
AQ. Comparison of high-dose and low-dose insulin by mia by phosphate infusion during treatment of diabetic
continuous intravenous infusion in the treatment of ketoacidosis and hyperosmolar coma. Diabetes 1980:
diabetic ketoacidosis in children. Diabetes Care 1980: 29: 87–95.
3: 15–20. 95. WILSON HK, KEUER SP, LEA AS, BOYD AE 3rd,
76. SACKS HS, SHAHSHAHANI M, KITABCHI AE, FISHER JN, EKNOYAN G. Phosphate therapy in diabetic ketoacido-
YOUNG RT. Similar responsiveness of diabetic ketoa- sis. Arch Intern Med 1982: 142: 517–520.
cidosis to low-dose insulin by intramuscular injection 96. BECKER DJ, BROWN DR, STERANKA BH, DRASH AL.
and albumin-free infusion. Ann Intern Med 1979: 90: Phosphate replacement during treatment of diabetic
36–42. ketosis. Effects on calcium and phosphorus homeosta-
77. UMPIERREZ GE, LATIF K, STOEVER J et al. Efficacy of sis. Am J Dis Child 1983: 137: 241–246.
subcutaneous insulin lispro versus continuous intrave- 97. FISHER JN, KITABCHI AE. A randomized study of
nous regular insulin for the treatment of patients with phosphate therapy in the treatment of diabetic
diabetic ketoacidosis. Am J Med 2004: 117: 291–296. ketoacidosis. J Clin Endocrinol Metab 1983: 57: 177–
ketoacidosis with subcutaneous insulin aspart. Diabe- Effect of phosphate on oxygen-hemoglobin affinity,
tes Care 2004: 27: 1873–1878. diphosphoglycerate and blood gases during recovery
79. DELLA MANNA T, STEINMETZ L, CAMPOS PR et al. from diabetic ketoacidosis. Intensive Care Med 1989:
Subcutaneous Use of a Fast-Acting Insulin Analog: 15: 495–498.
An alternative treatment for pediatric patients with 99. BOHANNON NJ. Large phosphate shifts with treatment
diabetic ketoacidosis. Diabetes Care 2005: 28: 1856– for hyperglycemia. Arch Intern Med 1989: 149: 1423–
1861. 1425.
MACLACHLAN E. Metabolic studies in diabetic coma. HOPWOOD NJ, HAWKER CD. Hypocalcemia, hypomag-
Trans Assoc Am Phys 1947: 60: 102–109. nesemia, and transient hypoparathyroidism during

Pediatric Diabetes 2007: 8: 28–43 41

Wolfsdorf et al.

therapy with potassium phosphate in diabetic ketoa- bicarbonate infusion in diabetic acidosis. J Pediatr
cidosis. Diabetes Care 1979: 2: 265–268. 1980: 96: 968–973.
101. WINTER RJ, HARRIS CJ, PHILLIPS LS, GREEN OC. 121. DEEB L. Development of fatal cerebral edema during
Diabetic ketoacidosis. Induction of hypocalcemia and outpatient therapy for diabetic ketoacidosis. Pract
hypomagnesemia by phosphate therapy. Am J Med Diab 1989: 6: 212–213.
1979: 67: 897–900. 122. GLASGOW AM. Devastating cerebral edema in diabetic
102. HALE PJ, CRASE J, NATTRASS M. Metabolic effects of ketoacidosis before therapy. Diabetes Care 1991: 14:
bicarbonate in the treatment of diabetic ketoacidosis. 77–78 (Letter)
BMJ (Clin Res Ed) 1984: 289: 1035–1038. 123. COUCH RM, ACOTT PD, WONG GW. Early onset fatal
103. MORRIS LR, MURPHY MB, KITABCHI AE. Bicarbonate cerebral edema in diabetic ketoacidosis. Diabetes Care
therapy in severe diabetic ketoacidosis. Ann Intern 1991: 14: 78–79.
Med 1986: 105: 836–840. 124. FIORDALISI I, HARRIS GD, GILLILAND MG. Prehospital
104. OKUDA Y, ADROGUE HJ, FIELD JB, NOHARA H, cardiac arrest in diabetic ketoacidemia: why brain
YAMASHITA K. Counterproductive effects of sodium swelling may lead to death before treatment. J Diabetes
bicarbonate in diabetic ketoacidosis. J Clin Endocrinol Complications 2002: 16: 214–219.
Metab 1996: 81: 314–320. 125. EDGE JA. Cerebral oedema during treatment of
105. GREEN SM, ROTHROCK SG, HO JD et al. Failure of diabetic ketoacidosis: are we any nearer finding
adjunctive bicarbonate to improve outcome in severe a cause? Diabetes Metab Res Rev 2000: 16: 316–324.
pediatric diabetic ketoacidosis. Ann Emerg Med 1998: 126. MUIR AB, QUISLING RG, YANG MC, ROSENBLOOM AL.
31: 41–48. Cerebral Edema in Childhood Diabetic Ketoacidosis:
106. ASSAL JP, AOKI TT, MANZANO FM, KOZAK GP. Natural history, radiographic findings, and early
Metabolic effects of sodium bicarbonate in man- identification. Diabetes Care 2004: 27: 1541–1546.
agement of diabetic ketoacidosis. Diabetes 1974: 23: 127. FRANKLIN B, LIU J, GINSBERG-FELLNER F. Cerebral
405–411. edema and ophthalmoplegia reversed by mannitol in
107. OHMAN JL Jr, MARLISS EB, AOKI TT, MUNICHOODAPPA a new case of insulin-dependent diabetes mellitus.
CS, KHANNA VV, KOZAK GP. The cerebrospinal Pediatrics 1982: 69: 87–90.
fluid in diabetic ketoacidosis. N Engl J Med 1971: 128. SHABBIR N, OBERFIELD SE, CORRALES R, KAIRAM R,
284: 283–290. LEVINE LS. Recovery from symptomatic brain swelling
108. SOLER NG, BENNETT MA, DIXON K, FITZGERALD MG, in diabetic ketoacidosis. Clin Pediatr (Phila) 1992: 31:
MALINS JM. Potassium balance during treatment of 570–573.
diabetic ketoacidosis with special reference to the use 129. ROBERTS MD, SLOVER RH, CHASE HP. Diabetic
of bicarbonate. Lancet 1972: 2: 665–667. ketoacidosis with intracerebral complications. Pediatr
109. LEVER E, JASPAN JB. Sodium bicarbonate therapy in Diabetes 2001: 2: 109–114.
severe diabetic ketoacidosis. Am J Med 1983: 75: 263– 130. CURTIS JR, BOHN D, DANEMAN D. Use of hypertonic
268. saline in the treatment of cerebral edema in diabetic
110. NARINS RG, COHEN JJ. Bicarbonate therapy for ketoacidosis (DKA). Pediatr Diabetes 2001: 2: 191–194.
organic acidosis: the case for its continued use. Ann 131. KAMAT P, VATS A, GROSS M, CHECCHIA PA. Use of
Intern Med 1987: 106: 615–618. hypertonic saline for the treatment of altered mental
111. CURTIS JR, TO T, MUIRHEAD S, CUMMINGS E, DANE- status associated with diabetic ketoacidosis. Pediatr
MAN D. Recent trends in hospitalization for diabetic Crit Care Med 2003: 4: 239–242.
ketoacidosis in Ontario children. Diabetes Care 2002: 132. MARCIN JP, GLASER N, BARNETT P et al. Factors
25: 1591–1596. associated with adverse outcomes in children with
112. EDGE JA, FORD-ADAMS ME, DUNGER DB. Causes of diabetic ketoacidosis-related cerebral edema. J Pediatr
death in children with insulin dependent diabetes 1990– 2002: 141: 793–797.
96. Arch Dis Child 1999: 81: 318–323. 133. KANTER RK, OLIPHANT M, ZIMMERMAN JJ, STUART MJ.
113. EDGE JA, HAWKINS MM, WINTER DL, DUNGER DB. Arterial thrombosis causing cerebral edema in associ-
The risk and outcome of cerebral oedema developing ation with diabetic ketoacidosis. Crit Care Med 1987:
during diabetic ketoacidosis. Arch Dis Child 2001: 85: 15: 175–176.
D. Population-based study of incidence and risk children with diabetic ketoacidosis. J Diabetes Com-
factors for cerebral edema in pediatric diabetic plications 1996: 10: 100–108.
ketoacidosis. J Pediatr 2005: 146: 688–692. 135. KEANE S, GALLAGHER A, ACKROYD S, MCSHANE MA,
115. HADDAD NG, CROFFIE JM, EUGSTER EA. Pancreatic EDGE JA. Cerebral venous thrombosis during diabetic
enzyme elevations in children with diabetic ketoacido- ketoacidosis. Arch Dis Child 2002: 86: 204–205.
sis. J Pediatr 2004: 145: 122–124. 136. ROSENBLOOM AL. Fatal cerebral infarctions in diabetic
116. ROSENBLOOM AL. Intracerebral crises during treat- ketoacidosis in a child with previously unknown
ment of diabetic ketoacidosis. Diabetes Care 1990: 13: heterozygosity for factor V Leiden deficiency. J Pediatr
22–33. 2004: 145: 561–562.
117. BELLO FA, SOTOS JF. Cerebral oedema in dia- 137. LAFFEL LM, WENTZELL K, LOUGHLIN C, TOVAR A,
betic ketoacidosis in children. Lancet 1990: 336: 64 MOLTZ K, BRINK S. Sick day management using blood
(Letter) 3-hydroxybutyrate (3-OHB) compared with urine
118. MAHONEY CP, VLCEK BW, DELAGUILA M. Risk factors ketone monitoring reduces hospital visits in young
for developing brain herniation during diabetic ketoa- people with T1DM: a randomized clinical trial. Diabet
cidosis. Pediatr Neurol 1999: 21: 721–727. Med 2006: 23: 278–284.
119. DURR JA, HOFFMAN WH, SKLAR AH, EL GAMMAL T, 140. LAFFEL L. Sick-day management in type 1 diabetes.
STEINHART CM. Correlates of brain edema in uncon- Endocrinol Metab Clin North Am 2000: 29: 707–723.
trolled IDDM. Diabetes 1992: 41: 627–632. 139. HOFFMAN WH, O’NEILL P, KHOURY C, BERNSTEIN SS.
120. BUREAU MA, BEGIN R, BERTHIAUME Y, SHAPCOTT D, Service and education for the insulin-dependent child.
KHOURY K, GAGNON N. Cerebral hypoxia from Diabetes Care 1978: 1: 285–288.

42 Pediatric Diabetes 2007: 8: 28–43

Consensus Guidelines

140. DROZDA DJ, DAWSON VA, LONG DJ, FRESON LS, 143. HOLLIDAY MA, SEGAR WE. The maintenance need for
SPERLING MA. Assessment of the effect of a compre- water in parenteral fluid therapy. Pediatrics 1957: 19:
hensive diabetes management program on hospital 823–832.
admission rates of children with diabetes mellitus. 144. FRIEDMAN AL. Pediatric hydration therapy: historical
Diabetes Educ 1990: 16: 389–393. review and a new approach. Kidney Int 2005: 67: 380–
WV. Coping skills training for youth with diabetes 145. HENDRICKS K, DUGGAN C., eds. Manual of Pediatric
mellitus has long-lasting effects on metabolic control Nutrition, 4th edn. Hamilton, Ontario: BC Decker,
and quality of life. J Pediatr 2000: 137: 107–113. 2005
142. GOLDEN MP, HERROLD AJ, ORR DP. An approach to 146. DARROW DC. The physiologic basis for estimating
prevention of recurrent diabetic ketoacidosis in the requirements for parenteral fluids. Pediatr Clin North
pediatric population. J Pediatr 1985: 107: 195–200. Am 1959: 6: 29–41.

Pediatric Diabetes 2007: 8: 28–43 43