Discontinuation of SGLT2 Inhibitors in Advanced CKD: An Inverse Probability of Treatment Weighting Analysis

Background: Sodium-glucose co-transporter 2 inhibitors (SGLT2i) slow the progression of chronic kidney disease (CKD). However, the evidence to support the continuation of SGLT2i when eGFR fell below 20 ml/min/1.73 m2is limited. Methods: Between 2015 and 2021, we identified 1011 CKD patients who received SGLT2i for at least 3 months and had eGFR progressed to ≤20 ml/min/1.73m2; 585 were continued with SGLT2i (Continuation group), and 426 had SGLT2i stopped (Discontinuation group). The primary outcome was dialysis-free survival. Inverse probability of treatment weighting (IPTW) was applied to minimize baseline differences between the two groups. Results: The median follow up was 14.3 months (IQR 8.7 – 21.8 months). The IPTW-adjusted two-year dialysis-free survival rates of the Discontinuation and Continuation groups were 41.7% and 58.9%, respectively (p < 0.0001). With the multi-variable Cox regression model to adjust for confounding factors, the Continuation group had a 39.9% (95%CI, 24.1% to 52.5%) reduction in risk of progression to dialysis or death. The IPTW-adjusted two-year patient survival rates of the Discontinuation and Continuation groups were 40.6% and 71.7%, respectively (p < 0.0001). With multi-variable Cox regression model, Continuation group had a 46.8% (95%CI, 29.8% to 59.7%) reduction in risk of death. The result remained similar when a propensity score-matched nested cohort was used for analysis. Conclusion: In CKD patients who received SGLT2i therapy, continuation of the SGLT2i therapy after the eGFR fell below 20 ml/min/1.73m2 was associated with a better dialysis-free survival and patient survival than discontinuation of SGLT2i. Our result support the recommendation of continuation with SGLT2i in this setting.