SUN-673 Development of Euglycemic Diabetic Ketoacidosis in a Patient with Well-Controlled Type 2 Diabetes Mellitus on a Sodium-Glucose Transporter 2 Inhibitor

Background: We describe the case of a patient with euglycemic diabetic ketoacidosis (euDKA), in the setting of sodium-glucose cotransporter-2 (SGLT2) inhibitor use, complicated by non-anion gap metabolic acidosis and low-carbohydrate diet leading to admission. Presentation: A 43-year-old woman with a history of type 2 diabetes mellitus treated with Metformin, with no prior history of DKA, was admitted with progressive dizziness, nausea, vomiting, malaise, palpitations, and dyspnea starting 3 days prior to admission. Her other history includes anemia due to uterine fibroids, hypertension, and hyperlipidemia. Hemoglobin A1C was 6.7%, however in the setting of anemia. She denied polyuria and polydipsia, and alcohol and drug use. She was started on a low-dose SGLT2i and pioglitazone 1 week prior to admission. Labs revealed mild hyperglycemia (blood glucose 145 mg/dL), with mixed anion-gap and non-anion-gap metabolic acidosis, and respiratory alkalosis [arterial pH 6.97 (rr7.35-7.45), PCO2 <13.0 mmHg (rr 32-45 mmHg), bicarbonate 5mm/dL (rr 24-33 mg/dL), anion gap 22, B-hydroxybutyrate 12.52 mmol/L (rr 0-0.3 mmol/L), and chloride 108 mEq/L (rr 98-109 mEq/L)], with normal renal function, hepatic function, and lactate. Infectious work-up was negative, including chest x-ray and urinalysis. She was diagnosed with euDKA due to SGLT2i. The SGLT2i was stopped and she was treated with insulin drip, intravenous fluids, and temporary bicarbonate drip given combined acidoses and severely low bicarbonate level, until her acidosis cleared. The patient noted that she had lately been eating a very low-carbohydrate diet in order to improve her glycemic control and promote weight loss. Discussion: In this case, DKA was likely precipitated by ketogenesis from low-carbohydrate diet for 1 week while taking a low-dose SGLT2i. Additionally, the dual anti-hyperglycemic therapy with Metformin and SGLT2i contributed to high anion-gap metabolic acidosis, along with the presence of a non-anion-gap metabolic acidosis. The patient was successfully transitioned to Metformin and pioglitazone upon discharge. As the use of SGLT2i is becoming widespread across multiple disciplines, recognizing euDKA in the setting of profound acidemia and very low carbohydrate diet in patients who are overall lower risk is particularly important.