Background: Haematuria in atrial fibrillation (AF) patients taking oral anticoagulants (OACs) is usually viewed as less serious than intracranial and gastrointestinal bleeding. It is speculated that haematuria may result from renal excretion of active new oral anticoagulants (NOACs) causing a direct anticoagulating effect in the urinary tract. Vitamin K antagonists (VKAs) such as warfarin, on the other hand, undergo hepatic metabolism and may pose lower risk of haematuria. This large registry study investigated whether NOACs more likely cause haematuria compared with VKAs.

Purpose: To assess whether there is any difference in the incidence rate of haematuria in AF patients taking NOACs versus VKAs using data from the GARFIELD-AF registry.

Methods: GARFIELD-AF is an international prospective registry of nonvalvular AF patients with at least one additional risk factor for stroke, followed for at least 2 years. Macroscopic haematuria was identified by local investigators. Event rates were estimated by Poisson model. Adjusted hazard ratio (HR) for haematuria between treatment groups was calculated using overlap-weighted Cox model including a range of patient demographics and clinical parameters as variables. Only the first haematuria occurrence was considered. Patients who were not treated with either VKAs or NOACs were excluded.

Results: Among a registry population of 34,926 patients 24,079 were anticoagulated and 24,061 had available follow-up data. Baseline characteristics were evenly balanced between the VKAs and NOACs subgroups, except a somewhat higher proportion of VKA patients than NOAC patients received concomitant antiplatelet therapy. Rate of haematuria was similar between the two groups: VKAs, 115/12,307 cases (0.9% over study period; 0.55 [95% CI, 0.46–0.65] per 100 patient-years); NOACs, 119/11,754 cases (1.0% over study period; 0.49 [95% CI, 0.41–0.59] per 100 patient-years). Over 2 years cumulatively, adjusted HR for haematuria in NOAC group versus VKA was 0.85 (95% CI, 0.63–1.15; p=0.29). Most haematuria cases (approximately 94%) were minor or clinically relevant non-major bleeds, occurring at a similar rate in both subgroups. Major bleeds were very rare. No intervention was necessary in two thirds haematuria cases (65.2%); surgical procedures were performed in only 8.3%. No haematuria-related deaths were observed.

Conclusions: The incidence and severity of haematuria were not increased in AF patients taking NOACs versus VKAs. Haematuria may occur in approximately one in 100 AF patients on long-term OACs therapy and is usually non-serious.