Data on different direct oral anticoagulants (DOACs) in atrial fibrillation (AF) patients with renal impairment are insufficient. We aimed to perform pairwise and network meta-analysis comparing oral anticoagulants (OACs) in AF patients with renal impairment, including advanced chronic kidney disease (CKD) with creatinine clearance <30 mL/min.

PubMed, Embase, Cochrane Database, and references of related articles were searched up to April 2021. We included randomized trials and non-randomized studies using propensity-score or multivariable-model adjustments that compared clinical outcomes among OACs. Hazard ratios (HRs) for stroke or thromboembolism, major bleeding, and all-cause death were pooled using random-effects model.

From 19 studies, 124,628 patients were included. In patients with AF and CKD, DOACs presented significantly lower risks of stroke or thromboembolism [HR_pooled = 0.78, 95% confidence interval (CI) = 0.73–0.85, I² = 16.6%] and major bleeding [HR_pooled = 0.76 (0.64–0.89), I² = 85.7%] when compared with warfarin, regardless of the severity of renal impairment. Results were consistent in advanced CKD patients for stroke or thromboembolism [HR_pooled = 0.60 (0.43–0.85), I² = 0.0%] and major bleeding [HR_pooled = 0.74 (0.59–0.93), I² = 30.4%]. In the network meta-analysis, edoxaban and apixaban presented the highest rank probability to reduce the risk of stroke or thromboembolism (edoxaban, P-score = 94.5%) and major bleeding (apixaban, P-score = 95.8%), respectively. Apixaban remained the safest OAC with the highest rank probability for major bleeding (P-score = 96.9%) in patients with advanced CKD.

DOACs, particularly apixaban and edoxaban, presented superior efficacy and safety than warfarin in AF patients with CKD. Apixaban was associated with the lowest risk of major bleeding among OACs for patients with advanced CKD.

[PROSPERO], identifier [CRD42021241718].