A recent survey on the use of direct oral anticoagulants (DOACs) revealed that 43% of patients with atrial fibrillation and renal impairment were potentially overdosed and had a hazard ratio for major bleeding of 2.19. In this review, we analyse and discuss the effect of renal failure on the pharmacokinetics and pharmacodynamics of DOACs and of strategies proposed to adjust doses according to the level of renal dysfunction. The pharmacokinetic characteristics of available DOACs (dabigatran, rivaroxaban, apixaban, edoxaban, betrixaban) differ substantially as regards oral bioavailability, plasma protein binding and the relative involvement of renal and non-renal elimination. In this respect, 80% of dabigatran is excreted as an unchanged drug in urine, whereas edoxaban, rivaroxaban, apixaban and betrixiban are excreted unchanged by, respectively, 50, 33, 27 and 11% of the dose. Therefore, drug exposure (the area under the concentration-time curve, AUC) is expected to increase to differing extents, depending on the residual renal function and the contribution of the kidneys to the excretion of each drug. Our analysis found that the increased AUC in patients with severe renal dysfunction was greater than expected in the case of dabigatran, betrixaban and rivaroxaban, indicating that other pharmacokinetic parameters may be altered besides renal clearance. Although DAOC pharmacodynamics do not seem to be altered by renal diseases (the correlation between plasma levels and anticoagulant effects overlaps that of healthy subjects), renal failure per se is associated with an increased risk of bleeding and thromboembolism. Guidelines on dose adjustments in patients with renal dysfunction have been published by three National Drug Agencies (FDA, EMA, HC), but many of their items do not match one another, reflecting our substantial paucity of knowledge in advanced renal failure. Routine monitoring of DOAC anticoagulant effects or plasma concentrations is not recommended, since no validated therapeutic ranges have been established. However, this approach may be useful in emergency situations such as bleeding or thrombotic events, urgent surgery, pharmacokinetic interactions, etc. We conclude that more experimental work is needed to improve our knowledge of DOAC pharmacology in renal failure and to provide clinicians with valid tools to adjust therapy.