N. Fleming, PS. Hamblin, D. Story and EI. Ekinci (2020). Evolving Evidence of Diabetic Ketoacidosis in Patients Taking Sodium-Glucose Cotransporter 2 Inhibitors. The Journal of clinical endocrinology and metabolism, 105(8). https://doi.org/10.1210/clinem/dgaa200

PubMed 32302001

[Abstract]
INTRODUCTION: Sodium glucose cotransporter 2 inhibitors (SGLT2i) have emerged as an important class of blood glucose-lowering medications, due to cardiovascular, metabolic, and renal benefits. However, there is a small but significant risk of diabetic ketoacidosis (DKA) associated with their use.

METHODS: A literature search was conducted in Ovid MEDLINE and Embase to July 2019 using variants on the key search terms sodium-glucose cotransporter 2, diabetic ketoacidosis, and type 2 diabetes. A broad spectrum of evidence was incorporated to facilitate a comprehensive narrative review. Further sources were identified through hand searching of reference lists.

DISCUSSION: Although cardiovascular outcome trials demonstrated mixed evidence of SGLT2i associated DKA, increasing evidence from case reports and cohort studies has identified an increased risk. SGLT2i use is associated with a ketotic state caused by an increased glucagon:insulin ratio and stimulated by factors including stress-induced hormonal changes, insufficient insulin, decreased glucose, increased ketone resorption, and hypovolemia. Atypical presentations of DKA with lower-than-expected blood glucose levels are possible with SGLT2i use, so clinical and biochemical monitoring is vital for early identification and management. DKA risk is particularly increased with precipitating factors, therefore optimization of risk factors is vital. Recommendations for perioperative and sick day management of patients taking SGLT2i have been suggested based on available evidence.

CONCLUSION: SGLT2i are an excellent class of drug in the physician's toolkit for managing type 2 diabetes. However, both clinicians and patients must be aware of the potential for DKA and the need for increased monitoring, both clinically and biochemically, when potential precipitating factors are present. In acutely unwell patients, these medications should be withheld to reduce the risk of DKA.