Background: The Cockcroft-Gault formula is recommended to determine a renal indication for dose reduction of dabigatran, edoxaban, and rivaroxaban. Nephrology guidelines now recommend the Modification of Diet in Renal Disease (MDRD) and Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formulae as more accurate estimates of glomerular filtration rate. Methods: We analyzed anticoagulated patients with atrial fibrillation who were enrolled in the Prevention of Thromboembolic Events – European Registry in Atrial Fibrillation (PREFER in AF). The proportion of patients with dissimilar renal dosing indications was assessed when applying Cockcroft-Gault, MDRD, or CKD-EPI. Thromboembolic and major bleeding events at 1 year were compared in patients in whom Cockcroft-Gault and CKD-EPI provided concordant or discordant results around a threshold of 50 mL/minute. Results: Out of 1288 patients with atrial fibrillation with chronic kidney disease in whom Cockcroft-Gault suggested a dose reduction of dabigatran, edoxaban, or rivaroxaban (creatinine clearance ≤50 mL/minutes), 19% and 16% were reclassified to the respective higher doses, and 24% and 23% to the respective lower doses by applying the MDRD and CKD-EPI formulae, respectively. In patients potentially receiving a different dose of dabigatran, edoxaban, or rivaroxaban when using CKD-EPI, we observed an excess of thromboembolic events (4.1% versus 0.8%; odds ratio, 5.5 [95% CI, 1.5–20.8]; P =0.01). Major bleeding rates were nonsignificantly different in the discordance versus concordance group (5.7% versus 2.7%; odds ratio, 2.2 [95% CI, 0.9–5.6]; P =0.09). Conclusions: The MDRD and CKD-EPI formulae suggest a different dosing in up to a quarter of anticoagulated patients with atrial fibrillation. This seems to impact hard outcomes.