SUN-571 Unmasking Euglycemic DKA: SGLT2 Inhibitor-Induced Ketoacidosis In A Non-Diabetic Patient

Abstract Disclosure: R.L. Moraes Dantas: None. C. Eliogu: None. D. Neri Rosario: None. I. Turcu: None. P. Kicker: None. Introduction: Euglycemic diabetic ketoacidosis (DKA) is a rare but serious complication of sodium-glucose cotransporter 2 inhibitors (SGLT2i), and particularly uncommon in non-diabetic individuals. CASE DESCRIPTION: A 75-year-old female with a medical history of hypertension, chronic kidney disease (CKD) stage III3b, gastroesophageal reflux disease, hypothyroidism, prediabetes, and chronic hyponatremia presented to the emergency department with generalized weakness, cough, reduced oral intake, diarrhea, and history of recent diagnosis of COVID-19 infection. She had been taking Empagliflozin, an SGLT2i, to slow the progression of her kidney disease.Initial laboratory tests revealed a sodium level of 122 mmol/L, serum osmolarity of 279 mOsm/kg, hemoglobin A1c of 5.8%, bicarbonate of 19 mmol/L, anion gap of 14.7, and a creatinine level of 3.0 mg/dL. Urinalysis showed 4+ glucose with negative ketones. She was initially admitted to telemetry for hyponatremia and acute kidney injury (AKI) on her CKD, then transferred to ICU the next day due to worsening hyponatremia (sodium 117 mmol/L) following repeated episodes of vomiting and diffuse abdominal pain. She was also diagnosed with pneumonia and ileus. During the ICU stay, her high anion gap metabolic acidosis (HAGMA) continued to worsen despite the initiation of a bicarbonate infusion. Her bicarbonate dropped to 7 mmol/L, pH was 7.25, her anion gap increased to 29.4, blood glucose was normal. Beta-hydroxybutyrate was significantly elevated at 73.74 mg/dL, while lactic acid remained normal at 0.6 mmol/L. The patient was diagnosed with euglycemic DKA induced by SGLT2i based on the presence of HAGMA, elevated beta-hydroxybutyrate, and absence of significant hyperglycemia. Other potential causes of high anion gap acidosis, including lactic acidosis, toxin ingestion, and starvation ketosis, were excluded. Insulin drip was started, and the diabetic ketoacidosis protocol of our institution was followed. Following treatment, her acidosis resolved, and her clinical status improved. DISCUSSION: DKA in prediabetic patients is rare and poorly studied, with few reported cases of euglycemic DKA. Our patient had several risk factors, including use of SGLT2i, infection, poor oral intake, and vomiting. Although sepsis with lactic acidosis was initially considered, lactate levels were normal. Starvation ketosis was also ruled out due to a lack of response to dextrose. Her unexplained high anion gap metabolic acidosis, along with home use of empagliflozin, strongly suggested euglycemic DKA. Empagliflozin was withheld during hospitalization and discontinued at discharge. Conclusion: Our case highlights the importance of considering euglycemic diabetic ketoacidosis in prediabetic patients taking SGLT2 inhibitors. Early recognition and appropriate insulin-based therapy are essential for effective management. Presentation: Sunday, July 13, 2025