Perioperative discontinuation of SGLT2-inhibitors and cardiac complications after non-cardiac surgery

Background Based on case reports, health authorities and clinical practice guidelines recommend to discontinue sodium–glucose cotransporter-2 inhibitors (SGLT2i) 3-4 days before surgery to avoid euglycemic diabetic ketoacidosis (EDKA). However, major uncertainties remain regarding the risk-benefit-ratio of discontinuation of SGLT2i, as it may increase postoperative cardiac complications. Methods Discontinuation of SGLT2i was assessed and quantified in consecutive high-risk patients on chronic SGLT2i therapy undergoing major non-cardiac surgery enrolled in two prospective cohort studies. The primary outcome was an ordinal composite of acute heart failure hospitalisation and cardiovascular death within 90 days, analysed using a partial proportional-odds ordinal logistic regression with adjustment for prespecified confounders. The secondary outcome was EDKA within 7 days. Results Among 451 patients (22% women; mean age 72 years), 95.1% had diabetes mellitus, and 36.9% chronic heart failure. SGLT2i were preoperatively discontinued in 87.1% of patients (39.2% for 1 day, 34.4% for 2 days, 13.5% for ≥3 days). The primary outcome occurred in 1.7% of patients without discontinuation, 5.7% with discontinuation for 1 day, 8.4% with discontinuation for 2 days, and 11.5% with discontinuation for 3 or more days. This association persisted after multivariable adjustment: cumulative odds ratio 1.58; 95% CI, 1.08–2.30 per discontinued day. One mild case (0.22%) of EDKA was observed without discontinuation of SGLT2i. Conclusions Perioperative discontinuation of SGLT2i was associated with a substantially increased risk of 90-day cardiac complications, clearly outweighing the very low rate of EDKA. This suggests that the current recommendation for discontinuation may need reconsideration.CIF of AHF and CV-death within 90 daysAdjusted cOR of primary outcome by pred.