Cristina M. Estrada and Timothy G. Givens
Diabetic ketoacidosis (DKA) is caused by either an absolute or a relative insulin deficiency accompanied by increased action of four counterregulatory hormones: glucagon, epinephrine, cortisol, and growth hormone (1,2). Counterregulatory hormone activity increases the patient’s insulin resistance, thus compounding the effects of insulin deficiency. The results of this imbalance are hyperglycemia, dehydration, ketosis, and acidosis.
Precipitating events for this cascade include acute infection, inflammatory reactions, trauma, psychological stress, or medical noncompliance in patients known to have insulin-dependent diabetes mellitus (IDDM); recurrent DKA almost always occurs because of failure to take insulin as prescribed. Approximately 30% of cases of new-onset IDDM present as DKA (3); a high index of suspicion is required to make the initial diagnosis in infants and children. Mortality in DKA is most often associated with the development of cerebral edema, which occurs more frequently in children than adults (4–7).
CLINICAL PRESENTATION
Children with DKA may present variably making the diagnosis challenging in patients not previously known to be diabetic. Acutely, the child may appear moderately dehydrated and be misdiagnosed as having viral gastroenteritis or a urinary tract infection. Presenting symptoms include nausea, vomiting, abdominal pain, hyperpnea, fruity breath odor, dehydration, lethargy, or coma. Polyuria, polydipsia, and polyphagia, three classic features of diabetes, coupled with weight loss should alert physician to consider DKA.
DIFFERENTIAL DIAGNOSIS
All patients presenting with vomiting should be queried for underlying histories of polyuria and polydipsia because new-onset diabetes with DKA may easily be mistaken for a routine case of gastroenteritis or a urinary tract infection. Occasionally, dehydrated children with gastroenteritis will have hyperglycemia and glucosuria caused by stress but do not have IDDM; since they are usually nonketotic (8).
The degree of hyperglycemia appears to be significantly associated with the severity of illness. However, the evidence is unclear as to whether patients with stress-induced hyperglycemia are at risk for future development of IDDM (8,9). If a history of diabetes is unclear or unavailable in a child with altered mental status, the differential diagnosis should widen to include intoxication, inborn errors of metabolism, Reye syndrome, sepsis, meningoencephalitis, nonketotic hyperglycemic coma, or hypoglycemic coma (10). In children with known IDDM, the differential diagnosis should focus on potential precipitants of DKA, such as bacterial infections, viral gastroenteritis, appendicitis, pancreatitis, or other acute intra-abdominal disorders.
ED EVALUATION
The evaluation of any child presenting with suspected DKA must address the following issues in a timely manner: (1) establishing the diagnosis, (2) assessing and correcting dehydration and electrolyte imbalance, (3) providing adequate insulin to restore normal metabolism, (4) searching for the precipitant cause of DKA, and (5) avoiding complications of therapy including cerebral edema, hypoglycemia, and hypokalemia.
DKA, as the name implies, has three components; hyperglycemia, ketosis, and acidosis. Patients may present with hyperglycemia alone and with none of the other elements early in their course. Alternatively, they may present further along the continuum of illness and be hyperglycemic with ketonuria but have not yet become acidotic. It is therefore important to assess patients using clear criteria for hyperglycemia, ketonemia, and acidosis. Traditionally, patients have been deemed to be in DKA when the serum glucose level is greater than 300 mg/dL, ketonuria is detected by urine dipstick, and the pH measured via venous blood gas (VBG) is less than 7.30 or the serum bicarbonate is less than 15 mEq/dL. Patients who present earlier in their course of illness and are not in fulminant DKA may be candidates for less aggressive treatment. This may include additional doses of subcutaneous insulin on a sliding scale with close monitoring of glucose and urine ketones in consultation with an endocrinologist or primary care provider. It is important to appreciate the spectrum of presentations of DKA and how these may affect the ED evaluation and management.
Following a primary survey, the patient’s degree of dehydration must be assessed. Estimates of dehydration are often underestimated, given the hyperosmolarity of the extracellular fluid in DKA patients. Relying solely on physical signs that correlate with extracellular volume loss (e.g., hydration of mucous membranes, skin turgor, and temperature) can be misleading and in general, one should assume at least 10%, or moderate, dehydration in all patients with DKA (1).
The interpretation of serum electrolyte levels must take into account the underlying physiologic derangements that accompany DKA. Measured serum sodium levels are often low secondary to the hyperosmolar state caused by hyperglycemia, which osmotically draws fluid into the intravascular space resulting in pseudohyponatremia. Correction of the serum sodium can be expected to occur without specific treatment as glucose levels fall during treatment for DKA. Serum potassium levels, on the other hand, may be elevated despite the patient’s total body stores being depleted. In the presence of acidosis intracellular potassium ions are exchanged for extracellular hydrogen ions that may be lost via osmotic diuresis. Low-normal potassium levels in DKA may be indicative of extreme total-body potassium depletion and continuous electrocardiographic monitoring for potassium-related cardiac dysrhythmias may be warranted.
KEY TESTING
• Initial serum testing including glucose, electrolytes, creatinine and urea nitrogen, blood gases, and hematocrit.
• A urinalysis should document the presence of ketones.
• The diagnosis of DKA is confirmed by the findings of hyperglycemia BG >300 mg/dL), a metabolic acidosis (pH <7.3 or HCO3 <15 mEq/L), ketosis.
ED MANAGEMENT
Treatment protocols for pediatric DKA vary considerably (10). Fluid replacement and correction of dehydration should begin as soon as possible. Formerly, volume expansion with a bolus of isotonic fluid such as normal saline, 20 mL/kg, was recommended during the first 1 to 2 hours of therapy. Because overly vigorous administration of fluids may precipitate a rapid fall in plasma osmolarity and is associated with the potential development of cerebral edema; pediatric treatment protocols now advise more conservative fluid replacement, on the order of 10 mL/kg during the first hour (11). Unless the patient is severely hemodynamically compromised, larger fluid boluses should be avoided (4).
Once the diagnosis of DKA has been confirmed, insulin administration should begin. Although insulin administration has been effective using intramuscular or subcutaneous routes, continuous low-dose intravenous infusion allows for slower, more even correction of hyperglycemia and is associated with fewer episodes of hypoglycemia than other routes of insulin administration (4). There is no evidence an initial bolus of intravenous insulin is helpful, and it may be harmful (1). Thus, a continuous intravenous infusion of insulin at a rate of 0.05 to 0.1 U/kg/hr is recommended. When mixing the insulin for infusion, the concentration of insulin should be at least 0.5 U/mL, and the infusion line should be primed to overcome the effect of insulin binding to the intravenous infusion equipment. The addition of a single dose of the long-acting analogue insulin glargine to initial low-dose insulin infusion therapy may hasten resolution of acidosis and smooth the transition to subcutaneous insulin administration (12). Plasma glucose levels should be measured hourly and serum electrolyte concentrations and VBG every 2 to 4 hours. Calcium and phosphate concentrations should be checked every 8 hours. The rate of decrease in glucose should be no faster than 50 to 100 mg/dL/hr (1).
After the first hour, fluids should be adjusted to correct the free water deficit over the next 48 hours. This is accomplished with a solution of 0.45% sodium chloride (which closely approximates the fluid lost in DKA), with the addition of 40 mEq/L of potassium as either the chloride or phosphate salt. Some authors suggest that isotonic saline may be the most appropriate fluid for the first 24 hours of therapy (13). The overall guiding principle, independent of the particular choice of fluid, is the gradual correction of fluid and electrolyte deficits, with frequent monitoring of serum chemistries (1,14).
Dextrose is added to fluids when the plasma glucose levels approach 300 mg/dL; the desired goal is to maintain a glucose level between 200 and 300 mg/dL. At this point it is vital not to wean the insulin drip if the patient still has a clinically important anion gap. Glucose levels should be supported by the addition of dextrose-containing fluids and not by the tapering of insulin therapy. Some institutions use a “two-bag system” to easily titrate the dextrose concentration of intravenous fluids between 0% and 10% in response to the patient’s current serum glucose (15). In this technique, two bags of saline solution are prepared, one with 10% dextrose and the other 0% dextrose (simple normal saline solution). By regulating the rates of fluid administration from each bag, the amount of fluid can be maintained constant, whereas the glucose content of infusion can be adjusted to respond to changes in the patient’s serum glucose.
Hypophosphatemia and deficiency of 2,3-diphosphoglycerate are well-known features of DKA. They occur because of osmotic diuresis and phosphate competition with glucose for reabsorption at the renal tubules. Potassium phosphate has thus been proposed as an alternative to potassium replacement with the chloride salt. Correction of phosphate deficits has not been shown to clinically benefit children with DKA (2).
Both the benefits and the risks of the use of bicarbonate in the treatment of DKA remain unproven (16). The potential advantages of bicarbonate are improved myocardial function, reduction of potential for dysrhythmias, decreased insulin resistance, and a more rapid correction of acidosis. The major disadvantages of bicarbonate use are its potential for paradoxical central nervous system acidosis (caused by the increased diffusion of carbon dioxide across the blood–brain barrier), increased hemoglobin affinity for oxygen (which increases tissue hypoxia, because of reduced off-loading), and promotion of hypokalemia. Recent data show that bicarbonate therapy offers little benefit in terms of reducing the severity of acidosis, improving mental status, or correcting hyperglycemia (16). Rapid infusions of bicarbonate may produce hypokalemia and exacerbate the hyperosmolarity and have been associated with an increased incidence of cerebral edema in children (4). Therefore, bicarbonate, if is used at all, should be reserved only for severe and refractory acidosis or life-threatening hyperkalemia.
In spite of improvements in DKA management, mortality caused by cerebral edema has not changed (7). Data suggest a large percentage of patients in DKA have subclinical cerebral edema on computed tomography scans (3); however, clinically significant cerebral edema is reported in <1% of cases of DKA (5). Typically, signs of increased intracranial pressure present several hours into therapy, as the patient experiences a sudden decline in sensorium, headache, sluggish pupillary reflexes, or a change in vital signs despite improvement in metabolic and other clinical parameters.
Although cerebral edema has been thought to be related to aggressive fluid resuscitation in early DKA, there is no strong or consistent evidence implicating fluid administration as a primary cause (17). What is known, however, in spite of a paucity of mechanistic evidence, is that children with new-onset IDDM and/or a prolonged period of untreated DKA are at increased risk for ultimately developing cerebral edema (3,17).
Once clinically apparent, cerebral edema is associated with a mortality rate of 60% to 80% and long-term neurologic morbidity in the majority of the rest. Hence, if cerebral edema becomes clinically evident, the patient should receive rapid aggressive intervention to reduce intra-cranial pressure—with intravenous mannitol (1 g/kg) or with 2 to 5 mL/kg of 3% saline (11). Intubation with mild hyperventilation should also be considered. Because of the devastating outcomes in patients who develop this complication, the need for frequent neurologic reassessments and early intervention cannot be over emphasized.
CRITICAL INTERVENTIONS
• A history of polyuria and polydipsia should be sought in every case of a child who is vomiting, particularly in very young children.
• Pediatric treatment protocols advise conservative fluid replacement, 10 mL/kg during the first hour.
• An initial bolus of intravenous insulin is not recommended because of concern about lowering the blood sugar too rapidly. A continuous infusion of intravenous insulin at a rate of 0.05 to 0.1 U/kg/hr is recommended.
DISPOSITION
Children in moderate-to-severe DKA should be admitted to a pediatric intensive or intermediate care unit because of the frequency and extent of monitoring required (1). Patients with new-onset IDDM should be admitted to a facility where age-appropriate diabetes education can be given.
Patients who can retain oral intake without vomiting and who are not significantly dehydrated or acidotic (pH >7.25) may be discharged home from the emergency department if appropriate follow-up care can be arranged. Follow-up includes frequent telephone contact during the ensuing 24 hours with an individual knowledgeable in diabetes management.
Common Pitfalls
• Failure to administer adequate insulin, which may result from errors in insulin dilution or infusion. Concerns about therapy-related hypoglycemia are addressed by adding dextrose to the fluid replacement, not by slowing or stopping the insulin infusion.
• Failure to treat the precipitating process. Patients with refractory DKA should be thoroughly reexamined for a source of infection or intra-abdominal disorder, such as pancreatitis or appendicitis.
• Inadequate volume expansion from failure to monitor the patient’s fluid input and output.
• Too-rapid correction of the hyperosmolarity may lead to cerebral edema and brainstem herniation.
• Inadequate potassium replacement during correction of acidosis, which leads to a shift of potassium into the cells and resultant hypokalemia.
REFERENCES
1. Rosenbloom AL. The management of diabetic ketoacidosis in children. Diabetes Ther. 2010;1(2):103–120.
2. Rewers A. Current concepts and controversies in prevention and treatment of diabetic ketoacidosis in children. Curr Diab Rep. 2012;12(5):524–532.
3. Klingensmith GJ, Tamborlane WV, Wood J, et al; Pediatric Diabetes Consortium. Diabetic ketoacidosis at diabetes onset: Still an all too common threat in youth. J Pediatr. 2013;162(2):330–334.
4. Dunger DB, Sperling MA, Acerini CL, et al. ESPE/LWPES consensus statement on diabetic ketoacidosis in children and adolescents. Arch Dis Child. 2004;89(2):188–194.
5. Wolfsdorf J, Glaser N, Sperling MA; American Diabetes Association. Diabetic ketoacidosis in infants, children, and adolescents: A consensus statement from the American Diabetes Association. Diabetes Care. 2006;29(5):1150–1159.
6. Levitsky LL. Symptomatic cerebral edema in diabetic ketoacidosis: The mechanism is clarified but still far from clear. J Pediatr. 2004;145(2):149–150.
7. Glaser N. Pediatric diabetic ketoacidosis and hyperglycemic hyperosmolar state. Pediatr Clin North Am. 2005;52(6):1611–1635.
8. Weiss SL, Alexander J, Agus MS. Extreme stress hyperglycemia during acute illness in a pediatric emergency department. Pediatr Emerg Care. 2010;26(9):626–632.
9. Otto-Buczkowska E, Dworzecki T, Mazur U. Alterations of blood glucose homeostasis in critically ill children - hyperglycemia. Pediatr Endocrinol Diabetes Metab. 2007;13(1):43–46.
10. Glaser NS, Ghetti S, Casper TC, et al; Pediatric Emergency Care Applied Research Network (PECARN) DKA FLUID Study Group. Pediatric diabetic ketoacidosis, fluid therapy, and cerebral injury: The design of a factorial randomized controlled trial. Pediatr Diabetes. 2013;14(6):435–446..
11. Glaser N, Kuppermann N. The evaluation and management of children with diabetic ketoacidosis in the emergency department. Pediatr Emerg Care. 2004;20(7):477–481.
12. Shankar V, Haque A, Churchwell KB, et al. Insulin glargine supplementation during early management phase of diabetic ketoacidosis in children. Intensive Care Med. 2007;33:1173–1178.
13. Cooke DW, Plotnick L. Management of diabetic ketoacidosis in children and adolescents. Pediatr Rev. 2008;29(12):431–435.
14. Inward CD, Chambers TL. Fluid management in diabetic ketoacidosis. Arch Dis Child. 2002;86:443–444.
15. Poirier MP, Greer D, Satin-Smith M. A prospective study of the “two-bag system’’ in diabetic ketoacidosis management. Clin Pediatr (Phila). 2004;43(9):809–813.
16. Chua HR, Schneider A, Bellomo R. Bicarbonate in diabetic ketoacidosis – a systematic review. Ann Intensive Care. 2011;1(1):23.
17. Hom J, Sinert R. Is fluid therapy associated with cerebral edema in children with diabetic ketoacidosis? Ann Emerg Med. 2005;45:550–551.