Incidence and Predictors of Worsening Renal Function in Edoxaban-Treated Atrial Fibrillation Patients Within ETNA-AF-Europe Registry

Background: Managing patients with atrial fibrillation (AF) and worsening renal function (WRF) remains a clinical challenge due to the need of dose adjustment of non-vitamin K antagonist oral anticoagulants. Objectives: To determine the incidence of WRF in patients with AF treated with edoxaban, the association of WRF with clinical outcomes, and predictors of WRF and clinical outcomes in these patients. Methods: This is a subanalysis of the Edoxaban Treatment in routiNe clinical prActice for patients with non-valvular Atrial Fibrillation in Europe study (NCT02944019), an observational study of edoxaban-treated patients with AF. WRF was defined as a ≥25% reduction in creatinine clearance between baseline and 2 years. Results: Of the 9,054 patients included (69% of the total 13,133 enrolled), most did not experience WRF (90.3%) during the first 2 years of follow-up. WRF occurred in 9.7% of patients. Patients with WRF had significantly higher rates of all-cause death (3.88%/y vs 1.88%/y; P < 0.0001), cardiovascular death (2.09%/y vs 0.92%/y; P < 0.0001), and major bleeding (1.51%/y vs 0.98%/y; P = 0.0463) compared with those without WRF. Rates of intracranial hemorrhage (0.18%/y vs 0.18%/y) and of any stroke/systemic embolic events were low (0.90%/y vs 0.69%/y; P = 0.3161) in both subgroups. The strongest predictors of WRF were a high CHA ₂DS ₂-VASc score, high baseline creatinine clearance, low body weight, and older age. Most predictors of WRF were also predictors of clinical outcomes. Conclusions: WRF occurred in approximately 10% of edoxaban-treated AF patients. Rates of death and major bleeding were significantly higher in patients with WRF than without. Stroke events were low in both subgroups.