Euglycemic Diabetic Ketoacidosis Secondary to SGLT2-inhibitor Use in Combination With a Ketogenic Diet

Sodium-glucose cotransporter 2 (SGLT-2) inhibitors are cardiorenal protective agents increasingly used in patients with diabetes. Cases of euglycemic diabetic ketoacidosis (euDKA) have been reported particularly among patients with type 1 diabetes. Our case is an example that highlights the role SGLT-2 inhibitors play in the development of euDKA in a patient with type 2 diabetes with confounding factors of strict adherence to a ketogenic diet and ankle fracture. A 72-year-old female with a history of type 2 diabetes presented to the emergency department (ED) with right ankle pain and obvious deformity after a mechanical fall. Radiography of the right lower extremity confirmed distal fracture of the tibia and fibula. After reduction of her fracture in the ED, she was admitted in anticipation of an open reduction internal fixation. Her diabetes was controlled on empagliflozin monotherapy and adhering to a ketogenic diet. She monitored her blood ketones daily at home and reported values in the 3–4 mmol/L range. On admission, her basic metabolic panel (BMP) revealed a blood glucose of 148 mg/dL, bicarbonate of 20 mEq/L, anion gap of 18 mEq/L, Cr of 1.3 mg/dl, and eGFR of 40 mL/min. Her beta-hydroxybutyrate (BHB) was 5.09 mmol/L. The initial assessment was presumed starvation ketosis. On hospital day three, she complained of continued nausea, polydipsia, and abdominal pain. Chart review revealed nocturia with approximately 3–4 voids per night. Repeat BMP showed a blood glucose of 152 mg/dL, bicarbonate of 16 mEq/L, anion gap still at 18 mEq/L, Cr 1.4 mg/dl, and eGFR of 37 mL/min. Since admission, all of her blood glucose levels ranged between 118–178 mg/dL. She denied dyspnea but exhibited Kussmaul respirations on physical exam. Repeat labs revealed a BHB of 8.92 mmol/L, and arterial blood gas (ABG) showed pH 7.2, pCO2 23, pO2 100, bicarbonate 8.6 mEq/L, consistent with high-anion gap metabolic acidosis, confirming the diagnosis of euDKA. Her empagliflozin had been held since admission, but she had not received any insulin up to this point due to euglycemia. She was immediately started on a weight-based dose of 12 units of insulin glargine with subsequent improvement of her BHB and anion gap. This patient’s use of an SGLT2-inhibitor in combination with her being on a ketogenic diet, contributed to a nonregulated state of ketone production leading to euDKA in the perioperative period. As SGLT2-inhibitors become more readily available, it is important to educate physicians and patients about the risk of euDKA during fasting, ketogenic diets, and the perioperative period.