Diabetic ketoacidosis in children: Clinical features and diagnosis

INTRODUCTION — Diabetic ketoacidosis (DKA) is the leading cause of morbidity and mortality in children with type 1 diabetes mellitus. It occurs at the time of diagnosis of type 1 diabetes in approximately 30 to 40 percent of children in the United States [1-6]. In children with established diabetes, DKA occurs at rates of 6 to 8 percent per year [7,8]. DKA can also occur in children with type 2 diabetes (and particularly in African American adolescents with obesity), although at lower rates than those observed in type 1 diabetes [9-16].

The clinical features and diagnosis of DKA in children will be reviewed here. Topic reviews with related content include:

●(See "Overview of the management of type 1 diabetes mellitus in children and adolescents".)

●(See "Diabetic ketoacidosis in children: Treatment and complications".)

●(See "Diabetic ketoacidosis in children: Cerebral injury (cerebral edema)".)

●(See "Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Clinical features, evaluation, and diagnosis".)

DEFINITIONS

●Diabetic ketoacidosis (DKA) – DKA is defined by the presence of all of the following in a patient with diabetes, as outlined in a consensus statement from the International Society for Pediatric and Adolescent Diabetes in 2022 [17]:

•Hyperglycemia – Blood glucose >200 mg/dL (11 mmol/L)

•Metabolic acidosis – Venous pH <7.3 or serum/plasma bicarbonate <18 mEq/L (18 mmol/L)

•Ketosis – Presence of ketones in the blood (>3 mmol/L beta-hydroxybutyrate) or urine ("moderate or large" urine ketones)

The severity of DKA can be categorized according to the degree of acidosis as mild, moderate, or severe (table 1). (See 'Diagnosis' below and 'Assessment of severity' below.)

●Hyperglycemic hyperosmolar state (HHS) – HHS is a hyperglycemic emergency, which is distinguished from classic DKA by:

•Marked hyperglycemia (blood glucose >600 mg/dL [>33.3 mmol/L])

•Minimal acidosis (venous pH >7.25 or arterial pH >7.3 and serum/plasma bicarbonate >15 mmol/L)

•Absent to mild ketosis

•Marked elevation in serum osmolality (effective osmolality >320 mOsm/kg)

Altered consciousness occurs frequently in HHS [17,18]. HHS occurs most commonly in adults with poorly controlled type 2 diabetes but has also been reported in children, most often in African American adolescents with type 2 diabetes [19-21]. (See 'Hyperglycemic hyperosmolar state' below.)

EPIDEMIOLOGY — DKA and its complications are the most common cause of hospitalization, mortality, and morbidity in children with type 1 diabetes mellitus and is frequently present at diagnosis [22]. It can also occur in children with type 2 diabetes, although at lower rates than those observed in type 1 diabetes [9-16]. (See "Classification of diabetes mellitus and genetic diabetic syndromes", section on 'DKA in type 2 diabetes'.)

DKA at initial presentation of type 1 diabetes mellitus — DKA occurs at the time of diagnosis of type 1 diabetes in approximately 30 to 40 percent of children in the United States and Canada [1-6,23]. Factors that increase the risk that a child will have DKA at the initial presentation of type 1 diabetes include [17,23-27]:

●Young age (<5 years of age and especially <2 years)

●Delayed diagnosis of diabetes, including reduced access to medical care

●Low socioeconomic status, lack of health insurance

●Children living in countries with low prevalence of type 1 diabetes

The importance of socioeconomic status was illustrated in a review of 139 patients with newly diagnosed type 1 diabetes mellitus seen at a single center in the United States [28]. Children without private health insurance (a proxy for low socioeconomic status) had a 62 percent frequency of DKA at presentation compared with 34 percent among those with private insurance. The frequency of DKA at presentation of type 1 diabetes has been shown to vary inversely with the frequency of type 1 diabetes in the population, presumably reflecting a greater frequency of missed diagnoses of type 1 diabetes when clinicians are less familiar with the disorder [4].

DKA in established type 1 diabetes mellitus — In children with established diabetes, DKA occurs at rates of 6 to 8 percent per year [7,8,23]. Groups with increased risk for DKA include [7,17]:

●Children with poor metabolic control (higher hemoglobin A1c [HbA1c] values and higher reported insulin requirements)

●Gastroenteritis with vomiting and dehydration

●Peripubertal and pubertal adolescent girls

●Children with a history of psychiatric disorders (including eating disorders) or unstable family circumstances

●Children with limited access to medical care (underinsured)

●Inadvertent or intentional omission of insulin, including dislodgement and occlusion of insulin pump infusion tubing

In a large prospective study in the United States, almost 60 percent of DKA episodes in children with established diabetes occurred in only 5 percent of children [7]. Similar findings were noted in the United Kingdom [22]. Patients who had four or more episodes of DKA (4.8 percent of patients) accounted for 35 percent of all episodes. Thus, a small group of patients consume a disproportionate amount of health care resources and costs [7].

DKA in type 2 diabetes mellitus — Although less common, ketosis and DKA can occur in children with type 2 diabetes, particularly in African American adolescents with obesity [9-16]. In a retrospective review of 69 patients (9 to 18 years of age) who presented with DKA, 13 percent had type 2 diabetes [13]. At presentation, there was no difference in pH levels, but patients with type 2 diabetes had higher blood glucose levels at presentation than those with type 1 diabetes. Mixed presentations of DKA and hyperglycemic hyperosmolar state (HHS) may also occur [18]. (See "Classification of diabetes mellitus and genetic diabetic syndromes", section on 'DKA in type 2 diabetes'.)

PRECIPITATING FACTORS — Common precipitating factors for DKA include the following; in many cases, multiple precipitating factors coexist:

●Poor metabolic control or missed insulin doses – Insulin omission and other diabetes mismanagement accounts for the majority of DKA episodes in children with established diabetes [7,29]. Omission of insulin injections (both intentional and unintentional) is particularly frequent among adolescents. (See "Complications and screening in children and adolescents with type 1 diabetes mellitus", section on 'Eating disorders'.)

●Illness – Intercurrent illnesses, particularly when associated with vomiting and dehydration, can precipitate DKA by increasing stress hormone levels (catecholamines, cortisol, and glucagon) that increase hepatic glucose output, cause peripheral insulin resistance, and promote ketogenesis. Illnesses involving vomiting are particularly problematic because they interrupt food intake, often requiring a reduction in the amount of insulin administered. Patients may completely omit insulin in an effort to avoid causing hypoglycemia. Anticipatory guidance for families on management of intercurrent illness can be helpful to avoid the need for hospitalization. (See "Management of type 1 diabetes mellitus in children during illness, procedures, school, or travel", section on 'Sick-day management'.)

●Medications – Certain medications, such as corticosteroids, atypical antipsychotics, tacrolimus, L-asparaginase, and diazoxide, have precipitated DKA in individuals not previously diagnosed with type 1 diabetes mellitus [30-33]. In children with established diabetes, use of corticosteroids can also lead to substantial insulin resistance with hyperglycemia and occasionally ketosis. Use of sodium-glucose cotransporter 2 (SGLT2) medications, which increase renal glucose excretion, have also been associated with increased risk of DKA in adults [34,35]. Patients using these medications may present with euglycemic DKA. (See "Sodium-glucose cotransporter 2 inhibitors for the treatment of hyperglycemia in type 2 diabetes mellitus", section on 'Diabetic ketoacidosis'.)

●Drugs and alcohol – In adolescents with type 1 diabetes, use of illicit drugs and alcohol may interfere with adherence to good medical management recommendations, resulting in poor metabolic control, which increases the risk for DKA. Acute ingestion of alcohol can cause hypoglycemia (rather than DKA) by decreasing hepatic gluconeogenesis. Case reports have described recurrent DKA in individuals with hyperemesis due to cannabis use [36]. (See "Cannabinoid hyperemesis syndrome".)

CLINICAL FEATURES — The clinical diagnosis of diabetes in a previously healthy child requires a high index of suspicion. Signs and symptoms of DKA are the result of acidosis, hyperglycemia, volume depletion, and electrolyte losses.

Signs and symptoms — The earliest symptoms of diabetes are related to hyperglycemia and are most apparent in children and adolescents. Symptoms include polyuria (due to glucose-induced osmotic diuresis), polydipsia (due to increased urinary water losses), and fatigue. Other findings include weight loss, nocturia, and enuresis. Occasionally, vaginal or cutaneous moniliasis may occur.

In infants, the diagnosis is more difficult because of lack of toilet training and difficulty expressing thirst. As a result, polyuria may not be appreciated and polydipsia is not apparent. However, decreased energy and activity, irritability, weight loss, and physical signs of dehydration are common findings. In addition, severe Candida diaper rash should raise suspicion for diabetes. (See "Diaper dermatitis".)

A number of other clinical findings may be seen in children with DKA:

●Children with DKA typically present with anorexia, nausea, vomiting, and abdominal pain. Focal pain may occur and can mimic appendicitis or other intraabdominal pathology. Polyphagia may be present early in the course of the illness. However, once insulin deficiency becomes more severe and ketoacidosis develops, appetite is suppressed. (See "Acute appendicitis in children: Clinical manifestations and diagnosis".)

●Hyperventilation and deep (Kussmaul) respirations represent the respiratory compensation for metabolic acidosis. Hyperpnea results from an increase in minute volume (rate × tidal volume) and can be increased by tidal volume alone without an increase in respiratory rate. In infants, hyperpnea may be manifested only by tachypnea. Patients may also have fruity breath odor secondary to exhaled acetone.

●Clinical signs of intravascular volume depletion such as tachycardia, poor peripheral perfusion, and decreased skin turgor occur in children with DKA but tend to be less prominent than in patients with the same degree of fluid loss from other conditions. This is because the intravascular volume is relatively preserved due to increased intravascular osmolality resulting from hyperglycemia. In addition, free water losses frequently exceed sodium losses. Finally, hyperglycemia-induced osmotic diuresis preserves urine output. Clinical estimates of the degree of dehydration are inaccurate [37] and unlikely to be helpful. Instead, patients should be rehydrated based upon a presumed fluid deficit and clinical response. (See "Diabetic ketoacidosis in children: Treatment and complications", section on 'Dehydration'.)

●Neurologic findings ranging from drowsiness, lethargy, and obtundation to coma are mainly related to the degree of acidosis [38]. Cerebral injury occurs in 0.3 to 0.9 percent of cases of DKA in children and is the leading cause of mortality. (See "Diabetic ketoacidosis in children: Cerebral injury (cerebral edema)".)

Fluid and electrolyte deficits — Children with DKA generally present with a 5 to 10 percent fluid deficit [1,39]. Measured weight loss provides the best estimate of volume depletion if an accurate recent measured weight (measured within a few weeks of presentation) is available for comparison. Otherwise, average fluid deficits of approximately 7 percent should be assumed [37,40,41] and fluid administration rates adjusted according to clinical response. (See "Diabetic ketoacidosis in children: Treatment and complications", section on 'Dehydration'.)

Total body sodium deficits in children with DKA range from 5 to 13 mmol/kg. Potassium deficits range from 3 to 6 mmol/kg [17].

Laboratory abnormalities

Blood glucose — A blood glucose level greater than 200 mg/dL (11 mmol/L) is generally required for the diagnosis of DKA [39,42]. This degree of hyperglycemia exceeds the renal tubular threshold for glucose reabsorption, resulting in osmotic diuresis with polyuria; volume depletion occurs when fluid losses are not sufficiently compensated by increased fluid intake.

In specific circumstances (pregnancy, use of sodium-glucose cotransporter 2 [SGLT2] inhibitor medications), DKA with normal or near-normal glucose levels can occur [34,35,43]. Also, children with established diabetes who have administered insulin prior to arrival in the emergency department may present with normal or near-normal glucose levels.

Acidosis — One of the diagnostic criteria for DKA is metabolic acidosis, defined as a venous pH <7.3 or serum bicarbonate concentrations <18 mEq/L. Venous pH is the most accurate measure of acidosis; however, measurements of serum bicarbonate may be used alone, especially in resource-limited settings, and are closely correlated with venous pH [44].

Insulin deficiency and increased plasma concentrations of glucagon, cortisol, and epinephrine increase glucose production, lipolysis, and ketogenesis, which collectively contribute to the development of both hyperglycemia and ketoacidosis. Acetoacetate is the initial ketone formed and is reduced to beta-hydroxybutyrate (BOHB) or decarboxylated to acetone. The latter will be detected as a ketone but does not contribute to the acidosis. (See "Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Epidemiology and pathogenesis".)

The severity of metabolic acidosis depends upon the rate and duration of increased ketoacid production, the rate of acid excretion in the urine, and adequacy of respiratory compensation [45,46]. Therefore, the acidosis will be more severe in a patient with delayed presentation or renal insufficiency.

Ketosis — The degree of ketosis can be estimated in several ways:

●Blood or serum BOHB – Measurement of serum BOHB is the most accurate clinically available index of ketosis and should be used whenever possible [47]. Most patients with DKA will have BOHB concentrations ≥3 mmol/L (31 mg/dL). BOHB in blood is available as a point-of-care test and is reasonably accurate in children and adults up to a blood BOHB concentration of 5 mmol/L (52 mg/dL) [48]. Measurement of BOHB in serum can also be performed in a clinical laboratory and is significantly more accurate than point-of-care testing at levels of BOHB >5 mmol/L.

●Urine ketones – Clinical testing with nitroprusside strips can be used to determine the presence of ketosis but is inaccurate for assessment of the degree of ketosis. In addition, this test may give a false impression of persistent ketoacidosis during recovery from DKA. This is because the nitroprusside test strip reacts with acetoacetate and acetone but not BOHB, which makes up 75 percent of circulating ketones. During recovery from DKA, BOHB is converted to acetoacetate and acetone, which are excreted in urine for many hours after the serum BOHB concentration has returned to normal.

●Anion gap – The anion gap is also useful in estimating the severity of ketosis, and the normalization of the anion gap is a helpful measure of the resolution of ketoacidemia. The normal anion gap in children is 12±2 mmol/L. The mean anion gap at presentation of DKA in children is 30±3 mmol/L [49].

When insulin is given to patients with DKA, metabolism of the ketoacid anions results in the regeneration of HCO₃^– and correction of the metabolic acidosis. For this reason, ketoacid anions have been called "potential bicarbonate," and their loss in the urine represents the loss of HCO₃^–. As a result of this precursor loss, as well as the high chloride content of intravenous fluids, a normal anion gap acidosis (hyperchloremic acidosis) is often seen during the treatment phase of DKA.

The anion gap is calculated in units of mEq/L or mmol/L, using a calculator (calculator 1) or the following formula:

Serum sodium — Patients with DKA have a total body sodium deficit ranging from 5 to 13 mmol/kg. Despite this deficit, their initial serum sodium concentrations can vary widely, ranging from mild hyponatremia (most patients) to mild hypernatremia. This is because the serum sodium concentration is influenced by two opposing effects of hyperglycemia:

●Hyperglycemia tends to lower the serum sodium concentration because it increases the plasma osmolality, resulting in movement of water from the intracellular to the extracellular space as a result of osmotic forces, thus lowering the serum sodium by dilution. In this setting, the patient's sodium status can be estimated by calculating a "corrected" plasma sodium concentration, which reflects the expected serum sodium concentration if the patient's glucose concentration were normal. The measured serum sodium is reduced by 1.6 mmol/L for every 100 mg/dL (5.5 mmol/L) increase in the blood glucose concentration above 100 mg/dL [50,51]. (See "Diabetic ketoacidosis in children: Treatment and complications", section on 'Serum sodium'.)

●Glucosuria-induced osmotic diuresis tends to raise the serum sodium concentration because of water loss in excess of sodium. This effect may partially correct the hyponatremia caused by hyperglycemia. If water intake is inadequate (which may be a particular problem in hot weather and in infants and young children who cannot independently access water), hypernatremia may occasionally occur.

If hyperlipidemia is present, the measured serum sodium concentration may be reduced due to a laboratory artifact. Hyperlipidemia can cause pseudohyponatremia by reducing the fraction of plasma that is water. As a result, the amount of sodium in the specimen is reduced and the measured plasma sodium concentration will be lower, even though the physiologically important plasma water sodium concentration and plasma osmolality are not affected [52]. With modern laboratory techniques, however, these erroneous sodium concentration measurements are uncommon. (See "Diagnostic evaluation of adults with hyponatremia".)

Serum potassium — Estimated potassium deficits in children with DKA are 3 to 6 mmol/kg. In spite of this total body deficit, serum potassium levels are usually normal or slightly elevated at presentation, reflecting the redistribution of potassium ions from the intracellular to the extracellular space. Regardless of the initial potassium level, therapy with insulin and fluids will lower the serum potassium concentration. As a result, potassium replacement, with careful monitoring of potassium levels, is essential. (See "Diabetic ketoacidosis in children: Treatment and complications".)

Some of the potassium deficit is due to urinary potassium loss from osmotic diuresis and ketoacid excretion. Elevated aldosterone concentrations in response to intravascular volume depletion also cause increased renal potassium loss. Additional potassium may be lost through vomiting (and through diarrhea in the case of DKA triggered by gastroenteritis). In addition, during DKA, potassium ions redistribute from the intracellular to the extracellular space as a result of the direct effects of low insulin concentrations (impairing potassium entry into cells), hypertonicity causing solvent drag, intracellular protein and phosphate depletion, and buffering of hydrogen ions in the intracellular fluid compartment. (See "Potassium balance in acid-base disorders".)

Serum phosphate — Children with DKA are typically in negative phosphate balance because of decreased phosphate intake and phosphaturia caused by glucosuria-induced osmotic diuresis. Despite the presence of phosphate depletion, at presentation, the serum phosphate concentration is usually normal or even slightly elevated because both insulin deficiency and metabolic acidosis cause a shift of phosphate out of the cells [53]. This transcellular shift is reversed and phosphate levels typically decline during DKA treatment. (See "Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Treatment", section on 'Phosphate depletion'.)

Blood urea nitrogen and creatinine — Patients with DKA often have elevated blood urea nitrogen (BUN) concentrations, which correlate with the degree of hypovolemia [54]. This finding at presentation may have predictive value since it is an important risk factor for cerebral injury during therapy [55]. (See "Diabetic ketoacidosis in children: Cerebral injury (cerebral edema)", section on 'Risk factors'.)

Many children with DKA have acute increases in serum creatinine compared with baseline, reflecting acute kidney injury (AKI). AKI occurs in 43 to 64 percent of children with DKA and is more common in children with more severe acidosis and circulatory volume depletion [56-58]. Within the group of children who have AKI, 33 to 65 percent have stage 2 or 3 AKI, suggesting renal tubular injury rather than prerenal dysfunction that is volume responsive. Children who have AKI during DKA have a higher long-term risk of diabetic kidney disease compared with those who recover from DKA without AKI [59].

EVALUATION — Initial laboratory testing should include the following [17,39,42]:

Immediate (point-of-care) tests — If DKA is suspected, the following tests should be performed to confirm the diagnosis:

●Blood glucose

●Blood beta-hydroxybutyrate (BOHB) or urine ketones (acetoacetate)

BOHB is the most direct and reliable measure of the degree of ketoacidemia. Measurement of urine ketones is less reliable but can be used to document the presence of ketosis and make a provisional diagnosis of DKA. (See 'Ketosis' above.)

Laboratory tests — At the same time, the following tests should be sent to the laboratory for more accurate measurements and to further characterize acid-base and electrolyte abnormalities:

●Blood glucose

●Electrolytes, including serum bicarbonate concentration

●Blood urea nitrogen (BUN) and creatinine

●Venous pH and partial pressure of carbon dioxide (pCO₂)

●Calcium, phosphorus, and magnesium – Abnormalities in these measures are unusual but occasionally require treatment

A complete blood count (CBC) is not essential for evaluation of a child with suspected DKA. If a CBC is performed in a child with DKA, typical findings include elevated white blood cell count with increased neutrophils. These findings are characteristic of DKA and do not help to identify children with infection.

Laboratory testing for specific clinical circumstances might also include:

●Serum BOHB concentration – This is useful when point-of-care testing for blood BOHB is not available or when urine ketone testing cannot be done or yields unclear results.

●Blood lactate concentration – This is useful in situations where the diagnosis of DKA is unclear and increased anion gap metabolic acidosis could be caused by lactic acidosis (eg, patients with very severe dehydration, shock, or suspected sepsis).

●Hemoglobin A1c (HbA1c) – This is useful in patients with known diabetes to evaluate the degree of metabolic control. HbA1c levels within the target range suggest that DKA was precipitated by an acute event. Elevated HbA1c levels suggest that chronic poor adherence to insulin treatment is a contributing factor. For patients with new onset of diabetes, HbA1c measurements are unnecessary for management of DKA.

●Diabetes-associated antibodies – Diabetes-associated antibodies (glutamic acid decarboxylase antibodies, insulin auto-antibodies, islet cell antibodies, and zinc transporter 8 antibodies) are not useful for management of DKA. However, they are useful in children with new onset of diabetes because positive results confirm the diagnosis of type 1 diabetes (diabetes of autoimmune origin). It should be noted, however, that approximately 10 to 15 percent of children with type 1 diabetes do not have detectable autoantibodies. (See "Epidemiology, presentation, and diagnosis of type 1 diabetes mellitus in children and adolescents", section on 'Distinguishing type 1 from type 2 diabetes'.)

DIAGNOSIS

Diabetic ketoacidosis — DKA is diagnosed when patients with diabetes mellitus exhibit all of the following:

●Hyperglycemia (blood glucose >200 mg/dL [11 mmol/L])

●Metabolic acidosis (venous pH <7.3 or serum bicarbonate <18 mEq/L [18 mmol/L])

●Ketosis (presence of ketones in the blood or urine)

Beta-hydroxybutyrate (BOHB) in blood or serum is the most accurate measure of ketosis and should be used whenever available. Serum BOHB concentrations ≥3 mmol/L (31 mg/dL) are present in most children with DKA [48]. Furthermore, BOHB ≥5.3 mmol/L is a strong indicator of DKA (sensitivity 77 percent, specificity 97 percent) [47]. Standard measurements of urine ketones should not be used to determine the severity of ketonemia, because this test only measures acetoacetate (see 'Ketosis' above). Venous pH is the most accurate measure of metabolic acidosis in patients with DKA. (See 'Acidosis' above.)

Hyperglycemic hyperosmolar state — DKA must be distinguished from hyperglycemic hyperosmolar state (HHS), which is characterized by:

●Marked hyperglycemia (plasma glucose >600 mg/dL [>33.3 mmol/L])

●Minimal acidosis (venous pH >7.25 or arterial pH >7.3 and serum bicarbonate >15 mmol/L)

●Absent to mild ketosis

●Marked elevation in serum osmolality (effective osmolality >320 mOsm/kg)

HHS occurs most commonly in adults with poorly controlled type 2 diabetes but has also been reported in children, most often in African American adolescents with type 2 diabetes [19-21]. Recognition of HHS is important because it is associated with more severe dehydration and a higher frequency of complications than is typical of DKA. Management of HHS is discussed in a separate topic review, as well as in published pediatric guidelines [17,18]. (See "Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Treatment", section on 'Fluid replacement'.)

ASSESSMENT OF SEVERITY — At presentation, the following clinical and laboratory findings may be used to estimate the severity of DKA and risks for complications:

●Acid-base status – The venous pH and serum bicarbonate concentration directly reflect the severity of the acidosis (table 1). The respiratory rate and partial pressure of carbon dioxide (pCO₂) also may be helpful since the magnitude of the respiratory compensation is related to the severity of the acidosis.

●Ketosis – Measurement of blood or serum beta-hydroxybutyrate (BOHB), rather than urine ketone concentration, is the optimal method for monitoring the degree of ketoacidemia. The magnitude of the anion gap is another measure of the severity of the ketosis and can be a helpful estimate of acidosis. A very large anion gap may also reflect decreased renal perfusion, which limits ketoacid excretion. (See 'Ketosis' above.)

●Neurologic state – Severe alterations in mental status at presentation or during DKA treatment may be indicative of DKA-related cerebral injury, a complication with high rates of morbidity and mortality. The pathophysiology, treatment, and prognosis of cerebral injury in children with DKA are discussed in detail separately. (See "Diabetic ketoacidosis in children: Cerebral injury (cerebral edema)".)

●Volume status – Measured weight loss provides the best estimate of volume depletion if an accurate very recent measured weight is available for comparison. Signs concerning for severe volume depletion include decreased urine output during hyperglycemia and the presence of hypotension. Of note, some patients with DKA present with paradoxical hypertension, although they are hypovolemic, or hypertension may develop during treatment [60]. This finding is associated with more severe DKA and/or alterations in mental status.

Markedly elevated blood urea nitrogen (BUN) or creatinine (or hemoglobin and hematocrit) also suggests significant dehydration, although elevated creatinine concentrations can also indicate DKA-related acute kidney injury (AKI). (See 'Blood urea nitrogen and creatinine' above.)

Together, these clinical features help to determine the appropriate clinical setting in which to treat the child. For example, mild DKA without vomiting in a child with established diabetes can often be safely managed in the outpatient setting, under close supervision and monitoring by an experienced diabetes team. On the other hand, a patient with severe DKA should be managed in a pediatric intensive care unit [39,42]. (See "Diabetic ketoacidosis in children: Treatment and complications".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Diabetes mellitus in children" and "Society guideline links: Hyperglycemic emergencies".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Diabetic ketoacidosis (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Epidemiology – Diabetic ketoacidosis (DKA) is the leading cause of morbidity and mortality in children with type 1 diabetes mellitus. DKA also can occur in children with type 2 diabetes mellitus. (See 'Epidemiology' above.)

DKA is the presenting feature of new-onset type 1 diabetes in approximately 30 to 40 percent of cases in the United States. Very young children and those of lower socioeconomic status are more likely to have DKA when they present with type 1 diabetes. (See 'DKA at initial presentation of type 1 diabetes mellitus' above.)

●Clinical presentation – Presenting symptoms of diabetes in children and adolescents include polyuria, polydipsia, and fatigue. Other findings include weight loss, nocturia (with or without secondary enuresis), daytime enuresis, and vaginal or cutaneous moniliasis. Infants tend to present with decreased energy and activity, irritability, weight loss, and physical signs of dehydration; a severe Candida diaper rash is common. (See 'Signs and symptoms' above.)

●Evaluation – Initial laboratory testing should include blood glucose, blood or serum beta-hydroxybutyrate (BOHB), serum electrolytes, creatinine, blood urea nitrogen (BUN), and venous blood gases. The diagnosis of DKA is confirmed by the findings of hyperglycemia, a high anion gap acidosis, and significant ketonemia. (See 'Evaluation' above and 'Diagnosis' above.)

●Diagnosis

•Diagnostic criteria – DKA is diagnosed when patients with diabetes mellitus exhibit all of the following (see 'Diagnosis' above):

-Hyperglycemia (blood glucose of >200 mg/dL [11 mmol/L])

-Metabolic acidosis (venous pH <7.3 or serum bicarbonate <15 mEq/L [15 mmol/L])

-Ketosis (presence of ketones in the blood or urine)

•Differential diagnosis – DKA must be distinguished from hyperglycemic hyperosmolar state (HHS), which is a hyperglycemic emergency characterized by:

-Marked hyperglycemia (blood glucose >600 mg/dL [>33.3 mmol/L])

-Minimal acidosis (venous pH >7.25 or arterial pH >7.3 and serum bicarbonate >15 mmol/L)

-Absent to mild ketosis

-Marked elevation in serum osmolality (effective osmolality >320 mOsm/kg)

HHS occasionally occurs in pediatric patients, most often adolescents with type 2 diabetes. Recognition of HHS is important because it is associated with more severe dehydration than DKA and requires a different management approach. (See 'Hyperglycemic hyperosmolar state' above and "Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Clinical features, evaluation, and diagnosis".)

●Assessment of severity – The venous pH and serum bicarbonate concentration directly reflect the severity of the acidosis (table 1). Neurologic status should also be formally assessed at presentation and periodically during treatment because cerebral injury is an important cause of morbidity and mortality in patients with DKA. (See 'Assessment of severity' above and "Diabetic ketoacidosis in children: Cerebral injury (cerebral edema)".)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Morey W Haymond, MD, and George S Jeha, MD, who contributed to earlier versions of this topic review.