Diabetic Ketoacidosis (DKA) is very common. Its treatment is seemingly straightforward, consisting of fluid/electrolyte replenishment and insulin administration. There are a few essential nuances, however, that clinicians and protocols frequently miss.
Insulin does not simply lower blood glucose. It facilitates glucose transfer from the bloodstream to the tissues. It also independently suppresses fat breakdown (lipolysis) associated with ketone production. When the body does not produce insulin in sufficient quantities (as in insulin-dependent diabetes), glucose remains in the bloodstream, causing hyperglycemia, leading to glucouresis and dehydration. Glucose cannot reach cells and serve as a substrate for energy production. As a short-term compensation mechanism, the body begins breaking down fats and producing ketones to use as an alternative fuel. Prolonged ketone production leads to anion gap metabolic acidosis.
It is essential to understand that insulin lowers blood glucose and inhibits ketones production independently. Therefore, normalizing blood glucose values (euglycemia) does not prevent ketone production. Historically, a similar approach was adopted to treat hyperglycemia and DKA – decreasing insulin infusion rate as blood glucose levels approach euglycemia. While it makes physiological sense in hyperglycemia, in DKA, blood glucose may normalize before ketone production is turned off. If insulin is discontinued at that point, DKA will not resolve. The alternative way of insulin delivery, which makes more physiological sense in DKA, is a fixed-rate weight-based infusion. When blood glucose decreases, dextrose is added to the intravenous fluids to prevent hypoglycemia.
Another aspect of traditional DKA management is fluid resuscitation with normal saline (NS). Not only it contributes to continuing acidosis (as NS is very acidic), but it is also nephrotoxic. A more up-to-date approach replaces normal saline with balanced intravenous fluids such as Lactated Ringer’s (LR).
Finally, traditionally basal insulin injections are administered after the anion gap is normalized, which requires a waiting period until the infusion is turned off. Giving long-acting insulin at the beginning (when the infusion is started) would shorten the treatment course and prevent the anion gap from reopening when the infusion is discontinued.