Sci Rep. 2021 May 13;11(1):10293. doi: 10.1038/s41598-021-89752-w.
Euglycemic diabetic ketoacidosis (EuDKA) secondary to Sodium-glucose co-transporter-2 inhibitors (SGLT2i) in type 2 diabetes mellitus (T2D) is a rare but increasingly reported phenomenon. Not much is known about the burden of EuDKA in patients on SGLT2i or the associated factors. This retrospective cohort study tries to delineate the differences in factors associated with the development of EuDKA as compared to hyperglycemic DKA. We conducted a multicentre, retrospective study across three tertiary care centers under Weill Cornell affiliated-Hamad Medical Corporation, Qatar. The cohort comprised of T2D patients on SGLT2i who developed DKA between January 2015 to December 2020. The differences between the subjects who developed EuDKA or hyperglycaemic DKA (hDKA) were analyzed. A total of 9940 T2D patients were on SGLT2i during 2015-2020, out of which 43 developed DKA (0.43%). 25 developed EuKDA, whereas 18 had hDKA. The point prevalence of EuDKA in our cohort was 58.1%. EuDKA was most common in patients using canagliflozin, followed by empagliflozin and Dapagliflozin (100%, 77%, and 48.3%, respectively). Overall, infection (32.6%) was the most common trigger for DKA, followed by insulin non-compliance (13.7%). Infection was the only risk factor with a significant point estimate between the two groups, being more common in hDKA patients (p-value 0.006, RR 2.53, 95% CI 1.07-5.98). Canagliflozin had the strongest association with the development of EuDKA and was associated with the highest medical intensive care unit (MICU) admission rates (66.6%). In T2D patients on SGLT2i, infection is probably associated with an increased risk of developing EuDKA. The differential role of individual SGLT2i analogs is less clear and will need exploration by more extensive prospective studies.