Sodium-glucose co-transporter 2 inhibitors induced euglycemic diabetic ketoacidosis within four days of initiation.

Euglycemic diabetic ketoacidosis (EDKA) is a well-known complication of sodium-glucose co-transporter 2 inhibitors, and many cases with variable onset following the initiation of these agents are reported before, with a median onset of approximately 2 wk. This letter discusses a 45-year-old lady who initially presented with ischemic stroke but developed EDKA 4 d after starting empagliflozin, a rare occurrence. The patient had severe metabolic acidosis that necessitated admission into the intensive care unit. Prompt discontinuation of empagliflozin and DKA management resulted in clinical recovery. ©The Author(s) 2022. Published by Baishideng Publishing Group Inc. All rights reserved.

Core Tip: With a steady surge in the prescription of sodium-glucose co-transporter 2 inhibitors (SGLT2-i) in medical conditions including type 2 diabetes mellitus (DM), type 1 DM, and heart failure, there are increasingly reported cases of euglycemic diabetic ketoacidosis (EDKA) with their use. EDKA in the context of SGLT2-i use is reported in various patients with different precipitating factors, some even with no inciting event. One of the rarely reported inciting events is stroke. Another aspect of SGLT2-i induced EDKA which remains relatively less understood is the time of initiation of the drug to the development of EDKA. In our patient, severe EDKA developed within 4 d of empagliflozin initiation, necessitating intensive care and discontinuation of empagliflozin, resulting in complete recovery regarding the EDKA.

TO THE EDITOR

With great interest, we read the recent article "Euglycemic diabetic ketoacidosis: A missed diagnosis" by Nasa et al[1]. The authors have described various factors that can potentiate euglycemic diabetic ketoacidosis (EDKA) in sodium-glucose co-transporter 2 inhibitor (SGLT2-i) use. There is a steady increase in EDKA reports secondary to SGLT2-i. Most of the articles mention a precipitating factor behind the development of EDKA in patients taking SGLT2-i. More extensive studies mention no or unknown precipitating factor in 16%-51% of cases[2,3]. This creates a need to explore a possible direct link of SGLT2-i in the development of EDKA in an otherwise healthy patient with diabetes. Acute vascular events such as stroke are infrequent inciting events for EDKA in the setting of SGLT2-i use.

We recently encountered an interesting case of a patient with type 2 diabetes mellitus (T2DM) who was admitted with acute stroke and developed EDKA within 4 d of initiation of empagliflozin.

A 45-year-old woman presented with sudden onset left arm weakness and slurred speech with facial droop. Magnetic resonance imaging revealed a right basal ganglia acute infarction, in addition to left parietal subcortical microangiopathic changes. The patient had a history of breast carcinoma, treated with mastectomy and maintenance tamoxifen. She also had T2DM and was prescribed sitagliptin/metformin 50/1000 mg two tablets daily. However, the patient was non-compliant with the medication and was not checking her blood sugar regularly.

The patient was started on dual antiplatelet therapy (aspirin 100 mg and clopidogrel 75 mg once daily) after establishing the diagnosis of an acute stroke. Her HbA1c was 14%, confirming a poor control of her diabetes. To manage her poorly controlled diabetes mellitus, sitagliptin/metformin was continued, with the addition of insulin glargine 12 units at bedtime and empagliflozin 10 mg once daily.

Four days later, the patient developed vomiting and generalized fatigue. Arterial blood gas showed severe metabolic acidosis with a pH of 6.9 and bicarbonate level of 3 mEq/L (reference range: 22-26 mEq/L). Serum B-hydroxy butyrate was higher than the reported threshold of 9.60 mmol/L (reference range: 0.03-0.3). Her blood glucose level at the time was 10.3 mmol/L (reference range: 3.3-5.5), and her urine dipstick showed +4 ketone. She was diagnosed with severe EDKA and was shifted to the medical intensive care unit for further management and treatment.

Regular insulin infusion and intravenous fluids were initiated, and a right internal jugular line was inserted for monitoring and resuscitation. Her arterial blood gas was measured every 2 h, and serum ketones were measured daily. Forty-eight hours later, the patient's condition improved, and she started tolerating oral feed. Her ABG results showed significant improvement with the closure of the anion gap (Table 1). She was started on subcutaneous glargine 20 units daily and insulin as part 7 units three times a day.

Table 1

Laboratory investigations of the patient during euglycemic diabetic ketoacidosis

Investigation	Onset	24 h	48 h	Reference
PH	6.9	7.32	7.42	7.35-7.45
HCO3 (mmol/L)	3	12.5	20.6	22-29
Glucose (mmol/L)	10.3	10.8	6.4	3.3-5.5
Sodium (mmol/L)	140	137	139	133-146
Potassium (mmol/L)	3.8	3.2	3.4	3.5-5.3
Chloride (mmol/L)	101	111	110	95-108
Anion Gap	36	13.5	8.4	10-12
Lactate (mmol/L)	1.3	0.7	0.9	0.36-1.6
B-hydroxybutyrate (mmol/L)	> 9.60	1.22	0.11	0.03-0.3

The patient was consequently shifted back to the care of the general medicine team, where her glycemic control was monitored closely. After ensuring the patient's fitness and stability, she was transferred to a physical and occupational therapy rehabilitation facility.

Our case highlights that in the presence of a precipitating factor, SGLT2-i drugs can cause an early and severe EDKA. We recommend that wherever other choices are available, initiation of SGLT2-i should be delayed until patients are otherwise healthy and not admitted with an acute event. SGLT2-i medications should ideally be started in an outpatient setting, and the patients should be counseled not to rely on blood glucose and seek immediate medical attention when experiencing symptoms of DKA.