Spotlight on SGLT-2 inhibitors and DKA

Multiple recent studies looked at incidence of diabetic ketoacidosis (DKA) among patients taking sodium-glucose cotransporter-2 (SGLT-2) inhibitors.

The first study, published by the Journal of Hospital Medicine on Feb. 19, was a retrospective analysis of patients with DKA from 2018 to 2023 in a single health system. Of the 510 patients seen over the study period, 89.2% had traditional DKA and 10.8% had euglycemic DKA, but the percentage with euglycemic DKA rose from 7.1% in January 2018 to 15.5% in January 2023. The patients with euglycemic DKA were significantly more likely to be on SGLT-2 inhibitors (44.4% vs. 0.03%) and had a significantly longer time to diagnosis (388 vs. 166 min). However, their length of stay in the ICU and hospital was similar to patients with traditional DKA. “Given the ever-growing list of indications for [SGLT-2 inhibitor] use (e.g., type 2 diabetes, chronic kidney disease, heart failure), these findings highlight the potential downstream impact this changing pharmacoepidemiology may have on hospital practitioners,” said the study authors, calling for hospitalists in particular “to have heightened awareness of this issue and be aware of the nuanced presentation of [euglycemic] DKA.”

The second study, published by JAMA Surgery on Feb. 19, assessed the rate of postoperative DKA in patients taking SGLT-2 inhibitors. This retrospective cohort study included 34,671 U.S. patients with type 2 diabetes who underwent emergency surgery from 2016 to 2022; 7.5% of the patients were on SGLT-2 inhibitors. The unadjusted incidence of DKA was 4.9% with SGLT-2 inhibitors, compared to 3.5% without. After adjustment for diabetes severity and other comorbidities, the rates were 3.8% and 3.5%, respectively, for an average treatment effect of 0.2% (95% CI, −1.7% to 2.2%). “Findings from this study may provide reassurance that the absolute incremental risk of clinically meaningful postoperative diabetic ketoacidosis in the population of patients using [SGLT-2 inhibitor] medications undergoing surgery, including a subgroup of those who likely followed standard preoperative fasting guidelines, is extremely low,” said the study authors. They noted that withholding SGLT-2 inhibitors before surgery could potentially worsen glycemic control, hypertension, and cardiac and kidney function on the day of surgery and that these risks “must be weighed against the low risks of diabetic ketoacidosis estimated by this large-scale study.”

Finally, a study published by Diabetes, Obesity and Metabolism on Feb. 25 quantified the rate of DKA in patients eating very low-carbohydrate ketogenic diets. In an analysis of 50,404 patient-years of follow-up from patients with type 2 diabetes being treated by a telemedicine clinic specializing in ketogenic diets, the incidence rate of DKA was 1.01 per 1,000 person-years. In the 12,763 person-years of follow-up from the patients taking SGLT-2 inhibitors, DKA incidence was 2.90 per 1,000 patient-years. The study authors noted that the observed incidence rate of DKA in patients taking SGLT-2 inhibitors and following a ketogenic diet is comparable to that seen in other analyses of patients taking the medications and not following a ketogenic diet. They concluded that warnings against combining SGLT-2 inhibitors and such diets “may be able to be lightened if these findings are confirmed.”