Background: We conducted a multicenter prospective cohort study to assess the safety of warfarin therapy in Japanese hemodialysis (HD) patients.
Methods: Chronic HD patients on warfarin therapy (warfarin users) were recruited from 111 HD centers in Japan. Two dialysis-vintage-matched warfarin non-users (non-users) were selected from the same HD center as each warfarin user. Clinical data were collected upon registration and every 12 months thereafter for up to 36 months.
Results: The final cohort consisted of 365 warfarin users and 692 non-users and was followed for an average of 27.7 months. The mean age of warfarin users (68.8 ± 10.6 years) was significantly higher than that of non-users (66.9 ± 11.0 years, p < 0.001). The analyses by multivariate Cox proportional-hazard models showed that the age [hazard ratio (HR) = 1.07 for each 1-year increase, 95 % confidence interval (CI) 1.05-1.08, p < 0.001] was significantly associated with the death from any cause, but warfarin use (1.08, CI 0.75-1.57, p = 0.68) was not when being adjusted for sex, diabetes mellitus, antiplatelet use, and atrial fibrillation. The risk of composite events, which included death from any cause, stroke, cardiovascular disease, and peripheral arterial disease, was also associated with age but was not associated with warfarin use.
Conclusion: The results of this study suggested that warfarin use by HD patients might not be harmful in chronic state, while the safety for the initiation of warfarin therapy in HD patients remained to be determined.
Keywords: Anticoagulation; Atrial fibrillation; Calciphylaxis; Cardiovascular disease; Hemodialysis.