Euglycemic DKA (EDKA) is a clinical syndrome occurring both in type 1 (T1D) or type 2 (T2D) diabetes mellitus characterized by euglycemia (blood glucose less than 250 mg/dL) in the presence of severe metabolic acidosis (arterial pH less than 7.3 and serum bicarbonate less than 18 mEq/L) and ketonemia. DKA is one of the most severe and life-threatening complications of diabetes mellitus and can be seen in a variety of conditions. The incidence of EDKA, however, has grown with the introduction of sodium-glucose transporter 2 (SGLT2) inhibitors. It also presents a diagnostic challenge for physicians due to the variety of etiologies and normal blood glucose levels, often resulting in delayed diagnosis.
There are many known causes of EDKA in patients with diabetes. The overall mechanism is based on a general state of starvation, resulting in ketosis while maintaining normoglycemia. Therefore, conditions like anorexia, gastroparesis, fasting, ketogenic diet, and alcoholism can lead to states of carbohydrate starvation and resultant ketosis. Additional triggers for EDKA include pregnancy, pancreatitis, glycogen storage disorders, surgery, infection, cocaine toxicity, cirrhosis, and insulin pump use. T1D bariatric surgery patients experience DKA in over 20% of postoperative cases and may be especially prone to EDKA.
The newer oral antidiabetic SGLT2 inhibitors, canagliflozin, dapagliflozin, empagliflozin, or ertugliflozin, can also directly result in EDKA. EDKA may be more common in patients with diabetes on SGLT2 inhibitors with lower body mass index and decreased glycogen stores. Episodes can be triggered by surgery infection, trauma, a major illness, reduced food intake, persistent vomiting, gastroparesis, dehydration, and reduced insulin dosage.
Approximately 2.6% to 3.2% of DKA admissions are euglycemic. DKA-associated with SGLT2 inhibitors has rates ranging from 0.16 to 0.76 events per 1000 patient-years in patients with type 2 diabetes. Blau et al. estimate the SGLT2 inhibitors increase the risk of DKA in T2D patients 7-fold. Erondu et al. estimate an overall incidence of DKA from SGLT2 inhibitor use of approximately 0.1%. Data on a patient with type 1 diabetes who presented with DKA associated with SGLT2 inhibitors showed rates varying from 5% to 12%; however, euglycemia was not present in all cases. Data associated with other causes of euglycemic DKA is scarce.
The underlying mechanism of EDKA is secondary to a carbohydrate deficit resulting in generalized decreased serum insulin and excess counter-regulatory hormones like glucagon, epinephrine, and cortisol. The increased glucagon/insulin ratio leads to increased lipolysis, increased free fatty acids, and ketoacidosis. Ketone body production in EDKA is similar to DKA with acetoacetic acid, beta-hydroxybutyric acid (after acetoacetic acid reduction), and acetone (after acetoacetic acid decarboxylation). The resulting anion gap metabolic acidosis triggers respiratory compensation and sensation of dyspnea, as well as nausea, anorexia, and vomiting. Volume depletion resulting from decreased oral intake, vomiting, and osmotic diuresis from glucosuria, further exacerbates elevations in glucagon, cortisol, and epinephrine, worsening lipolysis and ketogenesis. Additionally, decreased gluconeogenesis by the liver, as in fasting where hepatic glycogen is already depleted, or increased glucosuria by the kidneys, contribute to EDKA.
Often the insulin-using diabetic patient does not recognize their symptoms as DKA because serum glucose is not elevated, and maintains or decreases their insulin dose. If insulin dosing is adequate for glucose levels, it will prevent gluconeogenesis, resulting in euglycemia. EDKA can be considered a “partially treated DKA” in this setting.
SGLT2 inhibitors (empagliflozin, canagliflozin, dapagliflozin) are a newer class of antidiabetic drugs that increase the risk of EDKA unrelated to the duration of exposure. The use of SGLT2 inhibitors in T1D is not recommended by the U.S. Food and Drug Administration but discouraged because the risk of ketone-associated effects can be as high as 9%.
 The risk of EDKA among T2D patients on SGLT2 inhibitors may be higher in patients with beta-cell insufficiency and perhaps predict those at greater risk for evolving to T1D. The mechanism of action of SGLT2 inhibitors is to enhance excretion and block reabsorption of filtered glucose from the proximal convoluted tubule. The loss of urinary glucose again creates a state of carbohydrate starvation and volume depletion, increasing the glucagon/insulin ratio, resulting in a state of severe dehydration and ketosis. Additionally, SGLT2 inhibitors have been found to directly stimulate the release of glucagon from the pancreas, further worsening the glucagon/insulin imbalance, as well as suppress the removal of beta-hydroxybutyrate and acetoacetate by the kidneys. Euglycemia is maintained due to the loss off urinary glucose  and SGLT2 inhibitor-triggered hypoinsulinemia. SGLT2 inhibitors also increase ketone reabsorption such that ketosis is common in patients taking SGLT2 inhibitors, and a trigger such as pregnancy, alcohol, surgery, infection, or starvation can be enough to trigger full EDKA.
Pregnancy is a risk factor for EDKA because of the physiologic state of hypoinsulinemia and increased starvation. Increased levels of cortisol and placental lactogen result in insulin resistance, and episodes of vomiting or fasting can lead to exaggerated starvation ketosis. Respiratory alkalosis leading to bicarbonate loss in the urine exacerbates the acidosis. These factors, taken together, can result in starvation, euglycemia, and ketoacidosis by the previous mechanisms described.
Alcoholic ketoacidosis may have a very similar presentation as EDKA, with anorexia, vomiting, dyspnea, and significant anion gap metabolic acidosis and ketonemia. Some consider alcoholic ketoacidosis as a subtype of euglycemic DKA, and both are associated with increased glucagon/insulin ratio. The differentiating factors are that alcoholic ketoacidosis patients do not have diabetes or using diabetic medications, more commonly present with hypoglycemia, and after a severe alcohol binge. Similarly, excessive alcohol consumption in a person with diabetes can destroy pancreatic beta cells, decreased gluconeogenesis, and decreased glycogen stores. Coupled with vomiting, and carbohydrate starvation results, again, accelerated lipolysis, ketoacidosis, and EDKA.
Signs and symptoms will vary on a case-by-case basis but will be similar in presentation with a hyperglycemic DKA presentation, although perhaps without polyuria, polydipsia, or as severe mental status changes. EDKA patients can present with nausea, vomiting, shortness of breath, generalized malaise, lethargy, loss of appetite, fatigue, or abdominal pain. Patients may or may not have polydipsia or polyuria since serum glucose is normal. The onset can be more insidious compared to hyperglycemic DKA, due to the mechanism of subacute starvation required to induce ketosis and dehydration. There may or may not be an inciting infection or stressor, such as pregnancy, surgery, pancreatitis, alcohol use, or fasting.
Patients may present with deep, rapid breathing known as Kussmaul respirations, which represents respiratory compensation for severe metabolic acidosis. They may have a fruity odor to their breath due to the loss of acetone. Tachycardia, hypotension, altered mentation, increased skin turgor, and delayed capillary refill, are all signs of total body fluid losses. In severe cases, severe dehydration and metabolic derangement can lead to hypovolemic shock, lethargy, respiratory failure, coma, and death.
The initial laboratory evaluation of EDKA includes basic electrolytes, glucose, calcium, magnesium, creatinine, BUN, serum and urine ketones, beta-hydroxybutyric acid, arterial, or venous blood gas analysis, lactic acid, chest radiograph, and electrocardiogram. Urine screening for ketones with nitroprusside reagent does not measure beta-hydroxybutyrate but does detect acetone and acetoacetate. Serum levels of beta-hydroxybutyrate are typically greater than 3 mmol/L in EDKA (normal values less than 0.5 mmol/L). If an infection is on the differential, consider CBC with differential white blood cell count and blood cultures. Serum osmolality, to assess for an osmolar gap, and toxic alcohols should be sent to rule out toxic alcohol ingestion when suspected in any patient with severe, unexplained anion gap metabolic acidosis. Close attention should be paid to the anion gap, to help narrow direct diagnosis, workup, and management.
As described previously, the patient will have normoglycemia (capillary blood glucose less than 250 mg/dL) in the presence of metabolic acidosis (pH less than 7.3), and a total decreased serum bicarbonate (less than 18 mEq/L). Serum and urine ketones must be elevated to make the diagnosis of EDKA. Lactic acid may be elevated, but should not account entirely for the elevation in anion gap. Leukocytosis may be present in the case of a concurrent infection; however, it is nonspecific and could also be due to hemoconcentration or stress, among other causes. Potassium levels will vary, but great attention should be paid to the level before starting therapy, as total body potassium is usually depleted. Hypomagnesemia and hypophosphatemia can be seen in the starvation state due to decreased total intake and increased losses. Mild hyponatremia may also be seen but is generally less severe than the “pseudo-hyponatremia” seen in profound hyperglycemic states.
Initial management should be directed toward fluid resuscitation, as patients usually present profoundly dehydrated. Begin with the administration of isotonic saline or lactated ringers solution. The American Diabetes Association (ADA) recommends 1 to 1.5 L/hr isotonic fluids during the first 1 to 2 hours. Continuous insulin infusion should follow fluid replacement, contingent on serum potassium levels greater than 3.3 mEq/L, starting at a rate of 0.05 to 0.1 U/kg/hr. In contrast to DKA management, since serum glucose in EDKA is less than 250 mg/dL, dextrose 5% should be added initially to the fluids to avoid hypoglycemia and hasten clearance of ketosis. Consider increasing the amount of dextrose to 10% if ketoacidosis persists on D5%. Potassium should be carefully monitored as well, as total body potassium levels will likely be depleted, and IV supplementation of potassium and other electrolytes may be needed. Blood glucose levels should be checked hourly and electrolytes every four hours at a minimum, as is the standard for the treatment of DKA. Patients taking SGLT2 inhibitors should have these medications discontinued as soon as the diagnosis is recognized and held until recovery from the acute illness. Sodium bicarbonate infusions are not indicated and even use in the setting of severe acidemia of pH less than 6.9 is controversial. Patients will generally require ICU admission for close hemodynamic and laboratory monitoring as well as frequent titration of insulin infusion. Treatment with IV fluid resuscitation should continue until the anion gap closes, and acidosis has resolved.
In patients who present with anion gap metabolic acidosis, a variety of considerations must be employed early on. Infections, including pneumonia, genitourinary infection, bacteremia, must be ruled out early. In patients presenting with abdominal pain, consider intraabdominal infection and pancreatitis. Consider toxic alcohol (methanol, ethylene glycol) or paraldehyde ingestion, salicylate overdose, lactic acidosis, starvation ketosis, and pregnancy in the appropriate clinical setting. Note that patients may have also recently administered insulin, contributing to the euglycemic presentation. The presentation is very similar to alcoholic ketoacidosis (AKA), except EDKA patients have diabetes, and AKA patients present after an alcohol binge, more commonly have hypoglycemia and can be successfully resuscitated with crystalloid and dextrose without the requirement for insulin.
Most patients with EDKA recover well with prompt recognition and treatment. Delayed diagnosis and inadequate treatment, especially involving hydration without dextrose/insulin infusion, can lead to persistent acidosis, vomiting, and prolonged hospitalization. Prognosis is worse for small children and pregnant women. Rarely, severe cases, respiratory failure, hypovolemic shock, coma, and death. Death is rare in most EDKA cases; however, pregnant women are at greater mortality risk than the general population.
Consider consultation with a critical care intensivist and endocrinologist for severe cases. Pregnant women suffering from EDKA benefit from involvement with obstetric and maternal-fetal medicine consultants.
Vigilant monitoring of capillary or urine ketones by patients with diabetes, especially if on SGLT2 inhibitors, or during episodes of nausea or illness, can diagnose EDKA before it becomes severe.
Patients with T1D should not take SGLT2 inhibitors because of the high risk of EDKA.
Successful diagnosis is dependent on early screening with serum or urine ketones, even when serum glucose is normal, whenever EDKA is suspected.
After volume expansion with crystalloid, the foundation of therapy is a combination of dextrose (5% to 10%) and insulin (0.05 to 0.1 u/kg/hr) infusion until acidosis resolves.
Insulin infusion should not be avoided due to normal glucose levels, but instead are essential to successful treatment.
Ketosis does not resolve with intravenous crystalloid hydration alone.
SGLT2 inhibitor treatment should be discontinued as soon as EDKA is diagnosed.
Sodium bicarbonate infusion is not indicated.
Euglycemic DKA is becoming more prevalent with the appearance of the new SGLT2 inhibitors. However, it is important to recognize the variety of etiologies of a potentially fatal condition. Early diagnosis and initiation of treatment can significantly improve morbidity and mortality. In patients presenting with a euglycemic anion gap acidosis, great care must be taken to rule out other causes, including sepsis, toxic alcohol ingestion, alcoholic ketoacidosis, among others. Early IV crystalloid and prompt initiation of insulin and dextrose infusion are the primary treatment.
Treatment requires a team of interprofessional healthcare workers, possibly including consultation with a critical care intensivist as well as an endocrinologist. Emergency and critical care nurses monitor patients, administer ordered treatments, and report changes to the physicians so treatment can be optimized. The clinicians can also consult with the pharmacy regarding appropriate interventions, as well as having them run a full medication reconciliation to check for drug interactions or agents that may contribute to EDKA. These interprofessional interventions are crucial to achieving better patient outcomes. [Level 5]
|||Peters AL,Buschur EO,Buse JB,Cohan P,Diner JC,Hirsch IB, Euglycemic Diabetic Ketoacidosis: A Potential Complication of Treatment With Sodium-Glucose Cotransporter 2 Inhibition. Diabetes care. 2015 Sep; [PubMed PMID: 26078479]|
|||Modi A,Agrawal A,Morgan F, Euglycemic Diabetic Ketoacidosis: A Review. Current diabetes reviews. 2017; [PubMed PMID: 27097605]|
|||Rawla P,Vellipuram AR,Bandaru SS,Pradeep Raj J, Euglycemic diabetic ketoacidosis: a diagnostic and therapeutic dilemma. Endocrinology, diabetes [PubMed PMID: 28924481]|
|||Guo RX,Yang LZ,Li LX,Zhao XP, Diabetic ketoacidosis in pregnancy tends to occur at lower blood glucose levels: case-control study and a case report of euglycemic diabetic ketoacidosis in pregnancy. The journal of obstetrics and gynaecology research. 2008 Jun; [PubMed PMID: 18588610]|
|||Madaan M,Aggarwal K,Sharma R,Trivedi SS, Diabetic ketoacidosis occurring with lower blood glucose levels in pregnancy: a report of two cases. The Journal of reproductive medicine. 2012 Sep-Oct; [PubMed PMID: 23091997]|
|||Dowsett J,Humphreys R,Krones R, Normal Blood Glucose and High Blood Ketones in a Critically Unwell Patient with T1DM Post-Bariatric Surgery: a Case of Euglycemic Diabetic Ketoacidosis. Obesity surgery. 2019 Jan; [PubMed PMID: 30328578]|
|||Yu X,Zhang S,Zhang L, Newer Perspectives of Mechanisms for Euglycemic Diabetic Ketoacidosis. International journal of endocrinology. 2018; [PubMed PMID: 30369948]|
|||Goldenberg RM,Berard LD,Cheng AYY,Gilbert JD,Verma S,Woo VC,Yale JF, SGLT2 Inhibitor-associated Diabetic Ketoacidosis: Clinical Review and Recommendations for Prevention and Diagnosis. Clinical therapeutics. 2016 Dec; [PubMed PMID: 28003053]|
|||Joseph F,Anderson L,Goenka N,Vora J, Starvation-induced true diabetic euglycemic ketoacidosis in severe depression. Journal of general internal medicine. 2009 Jan; [PubMed PMID: 18975036]|
|||Burge MR,Garcia N,Qualls CR,Schade DS, Differential effects of fasting and dehydration in the pathogenesis of diabetic ketoacidosis. Metabolism: clinical and experimental. 2001 Feb; [PubMed PMID: 11229425]|
|||Legaspi R,Narciso P, Euglycemic Diabetic Ketoacidosis Due to Gastroparesis, A Local Experience. The Journal of the Arkansas Medical Society. 2015 Sep; [PubMed PMID: 26390536]|
|||Jenkins D,Close CF,Krentz AJ,Nattrass M,Wright AD, Euglycaemic diabetic ketoacidosis: does it exist? Acta diabetologica. 1993; [PubMed PMID: 8180418]|
|||Blau JE,Tella SH,Taylor SI,Rother KI, Ketoacidosis associated with SGLT2 inhibitor treatment: Analysis of FAERS data. Diabetes/metabolism research and reviews. 2017 Nov; [PubMed PMID: 28736981]|
|||Erondu N,Desai M,Ways K,Meininger G, Diabetic Ketoacidosis and Related Events in the Canagliflozin Type 2 Diabetes Clinical Program. Diabetes care. 2015 Sep; [PubMed PMID: 26203064]|
|||Taylor SI,Blau JE,Rother KI, SGLT2 Inhibitors May Predispose to Ketoacidosis. The Journal of clinical endocrinology and metabolism. 2015 Aug; [PubMed PMID: 26086329]|
|||Rosenstock J,Ferrannini E, Euglycemic Diabetic Ketoacidosis: A Predictable, Detectable, and Preventable Safety Concern With SGLT2 Inhibitors. Diabetes care. 2015 Sep; [PubMed PMID: 26294774]|
|||Henry RR,Thakkar P,Tong C,Polidori D,Alba M, Efficacy and Safety of Canagliflozin, a Sodium-Glucose Cotransporter 2 Inhibitor, as Add-on to Insulin in Patients With Type 1 Diabetes. Diabetes care. 2015 Dec; [PubMed PMID: 26486192]|
|||Pfützner A,Klonoff D,Heinemann L,Ejskjaer N,Pickup J, Euglycemic ketosis in patients with type 2 diabetes on SGLT2-inhibitor therapy-an emerging problem and solutions offered by diabetes technology. Endocrine. 2017 Apr; [PubMed PMID: 28303514]|
|||Bonner C,Kerr-Conte J,Gmyr V,Queniat G,Moerman E,Thévenet J,Beaucamps C,Delalleau N,Popescu I,Malaisse WJ,Sener A,Deprez B,Abderrahmani A,Staels B,Pattou F, Inhibition of the glucose transporter SGLT2 with dapagliflozin in pancreatic alpha cells triggers glucagon secretion. Nature medicine. 2015 May; [PubMed PMID: 25894829]|
|||Ferrannini E,Muscelli E,Frascerra S,Baldi S,Mari A,Heise T,Broedl UC,Woerle HJ, Metabolic response to sodium-glucose cotransporter 2 inhibition in type 2 diabetic patients. The Journal of clinical investigation. 2014 Feb; [PubMed PMID: 24463454]|
|||Crespo L,McConnell B,Wick JY, Euglycemic Diabetic Ketoacidosis: Hidden in Plain Sight. The Consultant pharmacist : the journal of the American Society of Consultant Pharmacists. 2016; [PubMed PMID: 27535077]|
|||Umpierrez GE, Diabetes: SGLT2 inhibitors and diabetic ketoacidosis - a growing concern. Nature reviews. Endocrinology. 2017 Aug; [PubMed PMID: 28621337]|
|||Bader N,Mirza L, Euglycemic Diabetic Ketoacidosis in a 27 year-old female patient with type-1-Diabetes treated with sodium-glucose cotransporter-2 (SGLT2) inhibitor Canagliflozin. Pakistan journal of medical sciences. 2016 May-Jun; [PubMed PMID: 27375734]|
|||Sinha N,Venkatram S,Diaz-Fuentes G, Starvation ketoacidosis: a cause of severe anion gap metabolic acidosis in pregnancy. Case reports in critical care. 2014; [PubMed PMID: 24963418]|
|||Lucero P,Chapela S, Euglycemic Diabetic Ketoacidosis in the ICU: 3 Case Reports and Review of Literature. Case reports in critical care. 2018; [PubMed PMID: 30364093]|
|||Dhatariya K, Blood Ketones: Measurement, Interpretation, Limitations, and Utility in the Management of Diabetic Ketoacidosis. The review of diabetic studies : RDS. 2016 Winter; [PubMed PMID: 28278308]|
|||Munro JF,Campbell IW,McCuish AC,Duncan LJ, Euglycaemic diabetic ketoacidosis. British medical journal. 1973 Jun 9; [PubMed PMID: 4197425]|
|||Nyenwe EA,Kitabchi AE, The evolution of diabetic ketoacidosis: An update of its etiology, pathogenesis and management. Metabolism: clinical and experimental. 2016 Apr; [PubMed PMID: 26975543]|
|||Dai Z,Nishihata Y,Kawamatsu N,Komatsu I,Mizuno A,Shimizu M,Toya N,Ishimatsu S,Komiyama N, Cardiac arrest from acute myocardial infarction complicated with sodium-glucose cotransporter 2 inhibitor-associated ketoacidosis. Journal of cardiology cases. 2017 Feb; [PubMed PMID: 30546697]|
|||Dalfrà MG,Burlina S,Sartore G,Lapolla A, Ketoacidosis in diabetic pregnancy. The journal of maternal-fetal [PubMed PMID: 26461169]|
|||Tarif N,Al Badr W, Euglycemic diabetic ketoacidosis in pregnancy. Saudi journal of kidney diseases and transplantation : an official publication of the Saudi Center for Organ Transplantation, Saudi Arabia. 2007 Nov; [PubMed PMID: 17951948]|