Euglycemic Diabetic Ketoacidosis in a Patient With Type 2 Diabetes After Treatment With Empagliflozin

Sodium–glucose cotransporter 2 (SGLT2) inhibitors have been associated with euglycemic diabetic ketoacidosis (eDKA). All reports to date have involved canagliflozin (Invokana; Janssen Pharmaceuticals), with the exception of one case associated with ipragliflozin (1). It has been anticipated that eDKA is a class effect, but no case reports of eDKA with other SGLT2s have been reported. Here, we report a case of eDKA in a patient with type 2 diabetes treated with empagliflozin.

A 64-year-old woman with a 15-year history of type 2 diabetes and a 5-day history of treatment with empagliflozin (Jardiance; Boehringer Ingelheim) presented to the emergency room (ER) for evaluation of shortness of breath. She had been treated with insulin for 10 years. At the time she started empagliflozin, she was taking liraglutide 1.8 mg per day. She had been taking NPH 40 units twice daily and regular insulin 20 units with meals but had independently discontinued insulin 3 weeks prior to presentation to determine if her blood glucose could be controlled with liraglutide alone. Capillary glucose measurements were in the low 200 mg/dL range on liraglutide alone, so treatment with empagliflozin 10 mg per day was initiated. Within 24 h of starting empagliflozin, she developed symptoms of weakness, joint pain, and mild confusion. After 4 days of treatment, she began to experience dyspnea on exertion and presented to the ER the next day. She had had only one alcoholic drink between starting empagliflozin and presenting to the ER.

Laboratory evaluation in the ER showed CO₂ of 11 mmol/L, anion gap of 21 mEq/L, and blood glucose of 161 mg/dL. Over the next 4–6 h, CO₂ fell to 6 mmol/L, anion gap increased to 24 mEq/L, and nausea and vomiting developed. Having been informed of the risk of DKA with empagliflozin, she asked that her endocrinologist be consulted. He was suspicious of eDKA and suggested a measurement of serum ketones. β-Hydroxybutyrate was found to be 8.22 mmol/L (normal up to 0.27 mmol/L), and arterial pH was 7.07. Venous lactate was normal. She was treated with intravenous fluids, insulin, and glucose, and the eDKA resolved over 24–36 h.

This is the first report of eDKA during treatment with empagliflozin. Potential risk factors based on previous reports include recent discontinuation of insulin and alcohol intake. In a recent case series of eDKA associated with canagliflozin, two patients with type 2 diabetes developed eDKA in the postoperative setting (1). A recent analysis of eDKA in the canagliflozin clinical trial program indicated that the incidence of DKA in people with type 2 diabetes treated with canagliflozin was comparable to that of the general population (2). The authors commented that most subjects had precipitating factors for DKA or possibly had been misdiagnosed with type 1 diabetes or had latent autoimmune diabetes of adulthood. The patient reported here had type 2 diabetes diagnosed on clinical grounds. She had a history of gestational diabetes mellitus followed by the diagnosis of diabetes 5 years later and the initiation of insulin 5 years after diagnosis. Her BMI was 36.5 kg/m², she was poorly controlled while taking 140 units of insulin daily (∼1 unit per kg), and her blood glucose had been maintained in the low 200 mg/dL range on liraglutide alone for 3 weeks. In conclusion, this is the first report of eDKA during treatment with empagliflozin, further indicating that eDKA is a class effect and that patients with type 2 diabetes may develop eDKA during treatment with SGLT2 inhibitors.

• Received August 17, 2015.
• Accepted September 21, 2015.

• © 2016 by the American Diabetes Association. Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered.