Pediatric Diabetic Ketoacidosis Management in the Era of Standardization

Ildiko H Koves; Catherine Pihoker


Expert Rev Endocrinol Metab. 2012;7(4):433-443. 

In This Article



In severe acidosis, total body sodium is markedly depleted and when sodium loss exceeds water loss, plasma sodium may also be low. Serum sodium concentration is an unreliable measure of the degree of extracellular fluid contraction as it is significantly influenced by hyperglycemia causing an osmotic shift; water movement into the extracellular space and induces dilutional hyponatremia. In addition, there is secondarily low sodium content with an elevated lipid fraction during DKA. It is recommended to calculate and monitor the corrected sodium level to encounter for the hyperglycemia-induced dilutional effect:

As fluid and insulin therapy is initiated, the serum sodium concentration should rise. The lack of this rise has been associated with an increased risk of CE.[35]


Total body potassium is generally considerably depleted during acidosis, even when the serum potassium concentration is normal or elevated. There is significant potassium loss with vomiting. Furthermore, the volume depletion leads to secondary hyperaldosteronism, which promotes further potassium loss via urinary excretion. Overall, these processes result in normal or even high serum potassium concentrations at presentation. Potassium moves from the intracellular to the extracellular space in the setting of acidosis, and there is lack of insulin-induced potassium entry into the cells. Renal dysfunction, by enhancing hyperglycemia and further reducing potassium excretion, contributes to the higher serum potassium level.[36] With exogenous insulin administration and glucose availability, potassium rapidly shifts back to the intracellular space, potentially resulting in life-threatening severe hypokalemia with arrhythmias. Therefore, with the commencement of insulin infusion, even with normal levels of potassium, replacement at 40 mEq/l is to be initiated,[3] or earlier if potassium levels are low. An ECG assessment (Box 3) is valuable to distinguish the presence of hypo- or hyper-kalemia,[37] if there is a delay in getting laboratory values, and should be performed in the setting of severe hypo- or hyper-kalemia. The authors recommend potassium replacement to be given in the form of potassium acetate and potassium phosphate. The use of potassium chloride may exacerbate hyperchloremia and hyperchloremic acidosis, and therefore should be avoided. Ongoing potassium replacement and monitoring is essential. In rare circumstances, 60 mEq/l potassium or higher is needed; however, the rate should not exceed 0.5 mEq/kg/h.[4] If severe hypokalemia persists, oral potassium replacement should be considered, and reduction of insulin infusion rate is only rarely needed.


Intracellular phosphate is generally depleted during DKA as phosphate is lost during the osmotic diuresis. Plasma phosphate levels fall with initiation of fluid therapy. The fall in plasma phosphate is further exacerbated by insulin therapy that promotes its intracellular shift. Profound, clinically significant phosphate deficiency may rarely occur with prolonged (>24 h) fasting that may lead to metabolic disturbance via the effect on oxygen–hemoglobin affinity and erythrocyte 2,3-diphosphoglycerate content.[38] Prospective studies have not shown clinical benefits from phosphate replacement.[39,40] Potassium phosphate salts may be safely used in combination with potassium acetate in a 50:50 ratio of total potassium salts, as intravenous phosphate administration may cause hypocalcemia.[41,42]


Plasma magnesium may be low. There is no role for magnesium replacement, as magnesium levels will normalize with resolution of acidosis and resuming oral diet. There is no proven benefit from replacing magnesium.

Acidosis & Bicarbonate

Severe DKA is reversed by fluid and insulin therapy. In the presence of insulin, no further ketoacids are produced, and the ketone bodies present become metabolized, which generates bicarbonate to reverse the acidosis. Exogenous bicarbonate use leads to rapid correction of acidosis and has been shown to cause hypokalemia,[43] and has also been associated with an increased risk of CE.[33,35] Therefore, in the routine management of DKA, bicarbonate use is not recommended, as it might not be safe.[44] It should be reserved only for extreme severe ketoacidosis, with hemodynamic instability or severe hyperkalemia given cautiously as a slow infusion over 60 min.

The best clinical indicator for resolution of ketosis is serum BOHB measurements. Traditionally, urine ketones have been used to assess for onset and resolution of ketosis during DKA. Current urine ketone strips use a nitroprusside reaction that tests for acetoacetate, and is significantly affected by hydration status, urine output and urine stagnating in the bladder, with the results only poorly reflecting the serum burden of ketones. Generally, in nonketotic states, there is a 1:1 ratio of BOHB:acetoacetate, whereas during DKA, this ratio increases by at least tenfold. BOHB is therefore a more accurate indicator of serum ketone burden and the resolution of ketosis.[45] The use of BOHB measurements as an end point for intravenous insulin therapy has been associated with earlier discharge from the intensive care unit, reduced hospital stay and overall cost savings.[46,47] Patients can be safely transitioned to subcutaneous insulin once the β-hydroxybutyrate level is less than 1 mmol/l.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.