March 12, 2019
2 min read

Inpatient diabetic ketoacidosis more frequent with SGLT2 inhibitor use

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Adults with type 2 diabetes prescribed an SGLT2 inhibitor were more likely to develop diabetic ketoacidosis during a hospital admission vs. nonusers of the therapy, according to findings from a retrospective medical records audit in Australia.

Shane (Peter) Hamblin

SGLT2 inhibitors overall were associated with an increased odds ratio of developing [diabetic ketoacidosis] in type 2 diabetes in the community and in hospitalized patients,” Shane (Peter) Hamblin, MBBS (Hons), FRACP, head of endocrinology and diabetes at Western Health in Melbourne, Australia, and clinical associate professor in the department of medicine at the University of Melbourne, told Endocrine Today. “The risk is small but significant: The incidence of diabetic ketoacidosis was 1.02 per 1,000 patients in SGLT2 inhibitor users vs. 0.69 per 1,000 adults in non-SGLT2 inhibitor users. All doctors and patients who use these drugs should be made aware of the risk and have a ‘sick day’ management plan. These are good drugs for diabetes, but this adverse event risk needs to be properly managed.”

Hamblin and colleagues analyzed medical records from 162 patients with physician-adjudicated type 2 diabetes and verified diabetic ketoacidosis (DKA), admitted to one of 11 public hospital networks in Australia from September 2015 to October 2017. Researchers identified users (n = 37) and nonusers of SGLT2 inhibitors (n = 125) and used those patients as a comparator group.

Researchers recorded any surgeries that occurred in the cohort and other precipitating factors for the development of DKA, including presence of infection, myocardial infarction and stroke. When available, researchers also assessed duration of planned fasting for surgical procedures or unplanned fasting due to illness. Researchers calculated ORs using incidence data.

Within the cohort, DKA developed during the course of inpatient admission in 14 SGLT2 inhibitor users (38%) vs. two (2%) nonusers, for an OR of 37.4 (95% CI, 8-175.9).

Across hospital networks, researchers found that 35,294 patients had at least one prescription dispensed for dapagliflozin (Farxiga, AstraZeneca) or empagliflozin (Jardiance, Boehringer Ingelheim) among 217,164 adults with type 2 diabetes, for an estimated non-SGLT2 inhibitor user population of 181,870. Using those figures, researchers determined the incidence of DKA was 1.02 per 1,000 patients among SGLT2 inhibitors vs. 0.69 per 1,000 patients in non-SGLT2 inhibitor users (OR = 1.48; 95% CI, 1.02-2.15).

Hamblin said SGLT2 inhibitor use should be ceased when a patient with type 2 diabetes is admitted to a hospital for surgery or when the patient is medically unwell and not eating properly.


“We did not observe any DKA cases in patients treated with an SGLT2 inhibitor admitted for day procedures, such as coronary angiography, gastroscopy or colonoscopy,” Hamblin said. “It would be good to see other studies closely examine these type of hospital admissions to see if patients undergoing these very common procedures can safely remain on their SGLT2 inhibitors at the time of these procedures.” – by Regina Schaffer

For more information:

Shane (Peter) Hamblin, MBBS (Hons) , FRACP, can be reached at Western Health, Department of Endocrinology and Diabetes, 176 Furlong Road, St. Albans 3021, Victoria, Australia; email:

Disclosures: The authors report no relevant financial disclosures.