Background: Although Sodium-glucose co-transporter-2 inhibitors (SGLT2i) were originally developed as therapeutic agents for diabetes mellitus, they are now widely used for chronic heart failure and kidney disease following the demonstration of their ...
Background: Although Sodium-glucose co-transporter-2 inhibitors (SGLT2i) were originally developed as therapeutic agents for diabetes mellitus, they are now widely used for chronic heart failure and kidney disease following the demonstration of their cardio- and reno-protective effects. SGLT2i offer clinical benefits, including renal protection by slowing the decline in glomerular filtration rate and inhibiting progression to end-stage renal disease (ESRD). However, this benefit is coupled with a risk of Diabetic Ketoacidosis (DKA). This duality presents a significant dilemma in the perioperative management of patients taking SGLT2i. To prevent DKA, guidelines, including those from the FDA, recommend discontinuing the medication 3-4 days before surgery. Surgery itself increases the risk of DKA, which can be fatal if not appropriately treated, potentially leading to decreased consciousness and death. Conversely, discontinuing SGLT2i eliminates their reno-protective effects, which, combined with factors like intraoperative hypotension and inflammation, may increase the risk of Acute Kidney Injury (AKI). Postoperative AKI occurs in a relatively high proportion (20%) of patients taking SGLT2i and increases the risk of other serious postoperative complications, such as infections, need for prolonged mechanical ventilation, and cardiovascular events. It also leads to extended hospital stays, increased mortality, and significant long-term impacts on renal health. Despite this trade-off, where discontinuation to prevent DKA may increase AKI risk, clinical evidence for the optimal discontinuation period is lacking, and institutional guidelines vary. Notably, previous studies have limitations, including unclear actual drug discontinuation periods and the inclusion of minor procedures. Therefore, research is needed to analyze the incidence of immediate postoperative complications by confirming SGLT2i discontinuation periods based on real-world clinical data, focusing specifically on major therapeutic surgeries.
Objective: This study aimed to provide evidence for the safe perioperative use of SGLT2i by comparing the incidence and risk of postoperative AKI and DKA between patients who discontinued SGLT2i for ≥ 3 days versus those who discontinued for < 3 days or continued the medication, among patients taking SGLT2i before major therapeutic surgery.
Methods: This was a single-center, retrospective cohort study targeting patients who underwent major therapeutic surgery at Seoul National University Hospital from June 1, 2014, to June 1, 2024. A total of 1,595 patients were included in the analysis. The group that discontinued SGLT2i for ≥ 3 days before surgery (including the day of surgery) comprised 318 patients, while the group that discontinued for < 3 days or continued consisted of 1,277 patients. The primary outcome was the incidence and risk of AKI, and the secondary outcome was the incidence and risk of DKA. To examine whether the effect of 'SGLT2i discontinuation' differed in specific patient groups, subgroup analyses for AKI were performed based on age, baseline chronic kidney disease (CKD), and cardiac surgery status. Subgroup analyses for DKA were performed based on BMI, heart failure (HF) status, and cardiac surgery status. To confirm the robustness of this analysis, sensitivity analyses were conducted using different SGLT2i discontinuation periods and DKA diagnostic criteria, and an evaluation for unmeasured confounders was performed. Incidence rates were analyzed using the Chi-squared test or Fisher’s exact test, while risk was analyzed using multivariable logistic regression to compare odds ratios. To control for confounding variables, Propensity Score Matching (PSM) and Inverse Probability of Treatment Weighting-Average Treatment effect on the Treated (IPTW-ATT) methods were used. All statistical analyses were performed using Python (ver. 3.12.12).
Results: Analysis of the primary outcome showed that the group discontinuing SGLT2i for ≥ 3 days had a significantly higher risk of postoperative AKI compared to the group discontinuing for < 3 days or continuing (OR 1.57, 95% CI 1.14-2.15, p=0.005). This increased AKI risk was particularly prominent in elderly patients aged 70 years or older. Analysis of the secondary outcome showed a tendency toward a reduced risk of DKA in the over 3 day discontinuation group; however, the statistical power was low (0.441) due to the small number of DKA events, and statistical significance was inconsistent across analytical methods (Multivariable logistic regression OR 0.296 vs. IPTW-ATT analysis OR 0.47).
Conclusion: This is the first study to simultaneously compare the risks of AKI and DKA in patients undergoing major therapeutic surgery based on the actual duration of SGLT2i discontinuation. We confirmed that the uniform preoperative discontinuation of SGLT2i to prevent DKA may unintentionally lead to a significant increase in the risk of AKI. Therefore, this suggests that future clinical guidelines need to be revised towards a personalized approach based on individual patient risk factors, comprehensively balancing the benefits of DKA prevention against the risk of AKI.