Atrial fibrillation (AF) and chronic kidney disease (CKD) share common risk factors and associations with other conditions, such as metabolic syndrome and cardiovascular diseases [1]. In the general Japanese population, the odds ratio of the presence of AF in subjects with a reduced estimated glomerular filtration rate (eGFR: ≤59 mL/min/1.73 m2) is 2.10 (95% confidence interval 1.21–3.86) compared with those with an eGFR ≥ 90 mL/min/1.73 m2 [2]. Previous studies have demonstrated that the presence of CKD was associated with a greater increase in the risk of stroke and thromboembolic events in patients with AF [3, 4].

In patients with AF and mild to moderate CKD, direct oral anticoagulant (DOAC) therapy showed noninferiority and even mild superiority to warfarin therapy for reducing the risk of stroke and thromboembolic and bleeding events [1, 5, 6]. A retrospective cohort study suggested that DOAC therapy was associated with a delayed decline in renal function [7]. Rivaroxaban is a selective inhibitor of factor Xa that catalyzes thrombin formation in the coagulation cascade. Thrombin has established nonhematologic functions beyond blood coagulation that include endothelial cell senescence and inflammatory responses [8, 9]. Ichikawa et al. [10] recently reported that rivaroxaban attenuated angiotensin II-induced renal injuries partly via inhibition of the protease-activated receptor-2-mediated inflammatory pathway, suggesting that rivaroxaban has the potential to exert renoprotective effects. However, the clinical effects of rivaroxaban on renal function in patients with AF and CKD remain uncertain. Therefore, we investigated the effects of rivaroxaban, compared with those of warfarin, on urinary albumin excretion (UAE) in those patients.

The X-NOAC was an investigator-initiated, multicenter, prospective, randomized, open-label, and blinded-endpoint trial. The protocol of the study has been reported previously [11]. Briefly, patients aged 30 years or older diagnosed with nonvalvular AF and CKD grade G2 to G3b were eligible for enrollment in the study. Detailed exclusion criteria are shown in Supplementary Text 1. Eligible patients were allocated randomly to either a warfarin or rivaroxaban therapy group at a 1:1 ratio. The rivaroxaban group received 15 mg orally once daily. In patients with a creatinine clearance (calculated by the Cockcroft–Gault) of 30–49 mL/min, a lower dose of 10 mg of rivaroxaban was given orally once daily. In the warfarin group, the dose was adjusted to maintain an international normalized ratio (INR) between 2.0 and 3.0 in patients younger than 70 years and between 1.6 and 2.6 in patients older than 70 years, according to the relevant local guidelines. The study was approved by the individual sites’ institutional review boards or independent ethics committees, in compliance with the Declaration of Helsinki and the current Japanese regulations for clinical trials. All patients provided written informed consent prior to study initiation.

The primary endpoint was the magnitude of change in the urine albumin-to-creatinine ratio (UACR) from baseline to after 3 months of treatment. The secondary endpoints were changes in several renal function biomarkers from baseline to 3 months. We compared the intergroup and intragroup differences in outcomes using the Wilcoxon rank sum test and Wilcoxon signed-rank test, respectively. All statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC). P < 0.05 was considered to be statistically significant.

A total of 25 patients on rivaroxaban and 26 patients on warfarin were included in the analyses (Supplementary Fig. 1). The baseline demographics and clinical characteristics were almost balanced between the two treatment groups, although the use of statins and antiplatelet agents was more frequent in the rivaroxaban group than in the warfarin group (Table 1).

Table 1 Baseline characteristics
Full size table

In the rivaroxaban group, six patients received rivaroxaban at a dose of 10 mg daily, while the remaining patients received rivaroxaban at 15 mg daily throughout the study period. In the warfarin group, the median dose of warfarin was 2.5 mg daily [interquartile range (IQR): 2.0, 3.0] at baseline and 2.5 mg daily [IQR: 2.0, 3.25] at 3 months, with a median INR value of 1.99 [IQR: 1.61, 2.40] at 3 months.

The changes in the UACR in the overall population are shown in Fig. 1a. After 3 months of treatment, the change in median UACR were −1.4 mg/g.Cr [−23.5, 0.9] in the rivaroxaban group and −0.4 mg/g.Cr [−12.2, 4.6] in the warfarin group. There was no significant intergroup difference in the magnitude of these changes (p = 0.52). When the patients were stratified into subgroups according to their baseline UACR, the subgroup with a baseline UACR ≥ 30 mg/g.Cr had a median change in the level of UACR at 3 months of −33.4 mg/g.Cr [−119.0, −0.7] in the rivaroxaban group and −16.3 mg/g.Cr [−59.0, 42.9] in the warfarin group (Fig. 1b). However, there was no significant difference in the magnitude of these reductions in the two groups (p for intergroup difference = 0.32). In the subgroup with a baseline UACR < 30 mg/g.Cr (Fig. 1c), the changes after 3 months were smaller in both groups. The reduction in UACR was also similar in the two groups (p for intergroup difference = 0.58).

Fig. 1
figure 1

Change in urinary albumin excretion. No significant difference in the magnitude of the changes in UACR after 3 months of treatment was observed between the two treatment groups in a the overall population (p = 0.52), b the subgroup with a baseline UACR ≥ 30 mg/g.Cr (p = 0.32), and c the subgroup with a baseline UACR < 30 mg/g.Cr (p = 0.58). The p values in the figures are for baseline vs. 3 months. Rivaroxaban treatment significantly reduced UACR only in the subgroup with a baseline UACR ≥ 30 mg/g.Cr (p = 0.02). UACR urine albumin-to-creatinine ratio

Full size image

The results of the secondary endpoints are summarized in Supplementary Table 1. Overall, no significant differences in the values at each time point or the magnitude of changes after 3 months were observed between the two treatment groups for any biomarker measured in the study. In the subgroup with a baseline UACR ≥ 30 mg/g.Cr, the rivaroxaban treatment did not affect the urinary liver-type fatty acid-binding protein, a marker of renal proximal tubular injury, while warfarin treatment significantly increased it (Supplementary Fig. 2).

Our current findings suggest that 3 months of rivaroxaban treatment, relative to warfarin therapy, had neutral effects on UAE and relevant biomarkers in patients with AF and mild to moderate CKD. However, our additional analyses suggest that rivaroxaban is effective for reducing UAE and attenuating tubular injury in patients with AF and micro/macroalbuminuria CKD. Importantly, we did not observe significant changes in blood pressure from baseline to 3 months, even in the subgroup with a baseline UACR ≥ 30 mg/g.Cr (Supplementary Table 2). To our knowledge, this is the first preliminary report that rivaroxaban may have beneficial effects on UAE in patients with AF and micro/macroalbuminuria. Nevertheless, the major limitation in our study was the relatively small sample size, with the actual number of participants not reaching the anticipated number (total 160) [11], which was partly due to the rapid spread in the clinical use of DOACs and slow enrollment. Therefore, further research using a larger-scale and longer-term design is needed to assess the detailed nonhematologic effects of DOACs on renal function in patients with AF and CKD.