LAUSR

OBJECTIVE To determine the incidence of postoperative ketoacidosis after elective cardiac surgery and to evaluate its association with sodium-glucose cotransporter 2 (SGLT2) inhibitor therapy and perioperative biochemical parameters. DESIGN A prospective observational study. SETTING Cardiac intensive care unit of a tertiary academic hospital in Madrid, Spain, between July 2024 and May 2025. PARTICIPANTS Adult patients undergoing elective cardiac surgery. Patients undergoing emergency procedures or those without informed consent were excluded. INTERVENTIONS Standard perioperative care. Arterial blood gases and capillary ketone measurements were obtained at intensive care unit (ICU) admission (defined as postoperative hour 0), at postoperative hour 6, and at postoperative hour 16. MEASUREMENTS AND MAIN RESULTS Postoperative ketoacidosis was defined as arterial pH <7.30 and ketonemia ≥0.6 mmol/L in the absence of preoperative ketoacidosis. A total of 107 patients were included (28 SGLT2 inhibitor users and 79 nonusers). Postoperative ketoacidosis occurred in 17 of 28 patients (54.8%) receiving SGLT2 inhibitors, compared with 8 of 79 (10.1%) in the non-SGLT2 inhibitor group (p < 0.001). Diabetes mellitus was independently associated with ketoacidosis (odds ratio, 5.13; p = 0.001). Patients with ketoacidosis had lower arterial pH (7.31 v 7.35, p < 0.001), lower bicarbonate (21.6 v 22.8 mmol/L, p = 0.012), and more negative base excess (-4.2 v -2.6, p = 0.03). No significant differences were observed in ICU length of stay, duration of mechanical ventilation, reintervention, or mortality. CONCLUSIONS Postoperative ketoacidosis, frequently euglycemic, is relatively common after cardiac surgery and strongly associated with SGLT2 inhibitor therapy. Careful monitoring and tailored perioperative management should be considered for this high-risk group.