Association Between Febuxostat Use and the Incidence of Cardiovascular Events, Mortality, and Kidney Events in Patients With Chronic Kidney Disease Compared to Allopurinol: A Study Using a Japanese Nationwide Database

Background Patients with chronic kidney disease (CKD) were excluded in most trials that investigated the effects of urate-lowering agents, such as febuxostat and allopurinol, in hyperuricemic patients. This exclusion leads to uncertainty regarding the efficacy of febuxostat in patients with CKD. Due to the high prevalence of hyperuricemia in patients with CKD, we aimed to assess the effect of febuxostat on improving patient outcomes concerning cardiovascular events and survival compared with those treated with allopurinol among patients with CKD. Methods We conducted a retrospective cohort study using Japanese nationwide administrative data from Jan 1, 2013, to Sep 30, 2020. Patients aged over 60 years diagnosed with CKD were included in this study if they were prescribed either febuxostat or allopurinol. The primary outcome was the occurrence of cardiovascular events including myocardial infarction, stroke, unstable angina requiring urgent revascularization, and all-cause deaths. We estimated hazard ratios (HR) and 95% CI using a Cox proportional hazard regression model adjusted for comorbidities, medications, and laboratory data. We also assessed defined starting kidney replacement therapy as a secondary endpoint treating death as a competing risk using the Fine & Gray regression model. Results A total of 21,015 patients included those with febuxostat (n=17,796) and those with allopurinol (n=3,219). The association between the type of drug and the occurrence of cardiovascular events did not show a significant difference (0.107 vs. 0.116 events per patient-year; adjusted HR 0.953, 95% CI: 0.854 to 1.062, P=0.381). Similar results were seen for all-cause deaths (0.060 vs. 0.068 events per patient-year; adjusted HR 0.877, 95% CI: 0.760 to 1.012, P=0.073). Regarding the secondary endpoint, the association between the type of drug and the timing of starting kidney replacement therapy did not show a significant difference (0.118 vs. 0.097 events per patient-year; adjusted HR 0.953, 95% CI: 0.854 to 1.062, P=0.425). Conclusion The use of febuxostat was neither associated with a decreased risk of cardiovascular events or deaths nor with the timing of starting kidney replacement therapy compared to the use of allopurinol in patients with CKD.