Safety and outcomes of dapagliflozin initiation in critically ill patients with acute kidney injury: A post-hoc analysis of the defender trial.

Abstract

BACKGROUND: SGLT2 inhibitor use in acute kidney injury (AKI) is controversial due to concerns about hemodynamic instability. We evaluated dapagliflozin initiation in critically ill patients with AKI enrolled in the DEFENDER trial. METHODS: Among 212 patients with AKI at enrollment (100 dapagliflozin, 112 control), we compared 28-day mortality, kidney replacement therapy (KRT), and composite death/KRT. Adjusted risk differences were estimated controlling for age, sepsis, baseline vasopressor use, and creatinine. Physiological trajectories (creatinine, urine output, fluid balance, acid-base parameters) over days 1-5 were analyzed using mixed models. Likelihood ratios quantified compatibility with clinically meaningful harm or benefit. RESULTS: Event rates were similar: 28-day mortality 38% vs 40%, KRT 12% vs 18%, composite 41% vs 42% (dapagliflozin vs control). Adjusted risk differences were - 1.9% (95% CI -14.5 to 10.7) for death, -7.4% (-16.2 to 1.5) for KRT, and - 0.9% (-13.6 to 11.8) for the composite. Physiological trajectories showed no divergence suggestive of hemodynamic or metabolic instability. Likelihood ratios provided limited separation: at 5% absolute effect threshold, LR against harm was 1.47 and against benefit 1.19. CONCLUSIONS: Dapagliflozin initiation in critically ill patients with AKI was not associated with excess mortality, KRT, or physiological derangement. The near-neutral evidential profile indicates neither moderate harm nor benefit can be excluded, supporting feasibility of dedicated trials of SGLT2 inhibitors in AKI.