Incidence of sodium–glucose cotransporter-2 inhibitor-associated perioperative ketoacidosis in surgical patients: a prospective cohort study
Hiroyuki Seki, Norifumi Kuratani, Toshiya Shiga, Yudai Iwasaki, Kanae Karita, Kazuki Yasuda, Natsuko Yamamoto, Yuko Nakanishi, Kenji Shigematsu, Kensuke Kobayashi, Junichi Saito, Ichiro Kondo, Nozomu Yaida, Hidenobu Watanabe, Midoriko Higashi, Tetsuro Shirasaka, Akira Doshu-Kajiura, Mitsutaka Edanaga, Satoshi Tanaka, Saori Ikumi, Shingo Ito, Masayuki Okada, Tomoko Yorozu, for the SAPKA Investigators
Abstract
Abstract Purpose Sodium-glucose cotransporter 2 inhibitors (SGLT2is) are commonly prescribed anti-diabetic medications with various beneficial effects; however, they have also been associated with ketoacidosis. The aim of this study was to determine the incidence of SGLT2i-associated perioperative ketoacidosis (SAPKA) in surgical patients. Methods We conducted a multicenter, prospective cohort study across 16 centers in Japan, enrolling surgical patients with diabetes who were prescribed SGLT2is between January 2021 and August 2022. Patients were monitored until the third postoperative day to screen for SAPKA, defined as urine ketone positivity with a blood pH of < 7.30 and HCO 3 level ≤ 18.0 mEq/L, excluding cases of respiratory acidosis. Results In total, 759 of the 762 evaluated patients were included in the final analysis. Among these, three patients (0.40%) had urine ketones with a blood pH of < 7.30; however, blood gas analysis revealed respiratory acidosis in all three, and none of them was considered to have SAPKA. The estimated incidence of SGLT2i-associated postoperative ketoacidosis was 0% (95% confidence interval, 0%–0.4%). Conclusions The observed incidence of SAPKA in our general surgical population was lower than expected. However, given that the study was observational in nature, interpretation of study results warrants careful considerations for biases.