Episode 56- “Glucose and ‘gliflozins”: Managing euglycemic DKA

Hello and welcome to Episode 56 of ER-Rx. This week, we’re going to talk about how to manage euglycemic diabetic ketoacidosis or DKA. Although it seems straightforward enough, there are some diagnostic and management pearls that we have to be aware of in these patients.

Euglycemic DKA is a syndrome that occurs in both type I and II diabetics. It’s associated with the presence of severe metabolic acidosis (meaning a pH < 7.3 and/or a serum bicarb < 18 mEq/L), an anion gap > 10, and usually the presence of ketones in the blood or urine. These are all diagnostic criteria for DKA, but where euglycemic DKA differs is that patients will have a glucose level of < 250 mg/dL, instead of our typical DKA presentation of a glucose > 250 mg/dL.

Seeing a normal glucose in our patients will sometimes lead to automatically excluding DKA in the setting of acidosis, or at the very least it will delay diagnosis. Although there are a number of known causes of euglycemic DKA, such as anorexia, fasting, a ketogenic diet, and alcoholism- euglycemic DKA still makes up only around 2% of all DKA cases. But this will probably change. We now have a relatively new class of oral antidiabetic medications called SGLT2 inhibitors (such as canagliflozin and empagliflozin, and a few others ending in “-gliflozin” that are just as difficult to say, and I’m not even sure I pronounced them right in the first place). These agents work by inhibiting the sodium-glucose transporter protein 2 and the reabsorption of glucose in the kidney. Apart from causing your patient to pee out all glucose and lower their glucose levels, they have also been shown to have significant cardiovascular benefits, making their way into cardiovascular and diabetic guidelines. On the flip side, they do cause some adverse effects, such as more urinary tract infections and an increased risk of DKA, especially in type 1 diabetics.

The mechanism by which SGLT2 inhibitors cause euglycemic DKA is complex, but it boils down to the fact that they create a state of carbohydrate deficiency, decreased insulin, and excess counter-regulatory hormones like glucagon and cortisol. This in turn leads to increased lipolysis and the release of ketones. They’ve also been shown to directly stimulate the release of glucagon and inhibit the removal of ketones by the kidneys.

Patients usually present as they would if they had hyperglycemic DKA; with nausea, vomiting, fatigue, or abdominal pain. They may or may not have polydipsia or polyuria since the glucose level is normal, and they tend to have slower onsets of DKA than patients with hyperglycemic DKA. They also don’t necessarily need to have an inciting infection or stressor- just being on the SGLT2 is enough.

Treatment for the most part is similar to how we treat old-fashioned DKA with fluid resuscitation and continuous IV insulin infusions. DKA insulin infusion protocols vary from site to site, and some are better than others. Most start with a weight-based, unit/kg/hour rate. Others, like mine, use strange calculators and multipliers based on glucose levels to arrive at an insulin dose. Whatever you have at your site, in this particular situation, it may be best to start the insulin infusion at a rate of 0.05- 0.1 units/kg/hr. But this is where the big difference comes in when treating euglycemic DKA. Given the glucose will already be < 250 mg/dL, a 5% or 10% dextrose solution should be started at the same time as the insulin infusion. Why? Because your patient’s glucose is low, some protocols, or even treating providers and nurses, will want to turn off the insulin infusion when the patient inevitably gets hypoglycemic. This is bad since we all know that insulin is essential in clearing ketones and restoring acid/base balance. By giving the D5 or D10 at the same time- although counter-intuitive, you make sure the insulin infusion stays on at the ordered rate without leading to hypoglycemia. If patients get hypoglycemic, it’s best just to leave the insulin at the current rate, and simply increase the rate of the dextrose infusion. If they are hyperglycemic, just turn down the dextrose infusion, or give a higher dose of insulin. At my site, we leave the insulin infusion at a set weight-based rate and titrate the dextrose infusion to maintain a glucose level between 150- 200. We also adjust our orders and have clear communication with the team to say that the typical hyperglycemic DKA orderset or order-panel will not be adequate for these patients.

In conclusion, with the advent and increasing use of those very-difficult-to-pronounce SGLT2 inhibitors we may see more euglycemic DKA. If you have an acidotic patient with normal glucose levels, don’t be quick to rule out DKA. Check the home medication list for any drug ending in “-gliflozin” to start. Also, be ready to adjust your usual DKA orderset or order-panel by ordering a continuous IV insulin infusion with a titratable order for D5 or D10 at the same and maintain glucose levels between 150-200 until your patient is safe to transition off of the insulin drip.

As always, thank you so much for your time. Feel free to add any comments or suggestions for any episode you hear on errxpodcast.com or the @errxpodcast Instagram page.