Types of erythropoiesis-stimulating agents and risk of end-stage kidney disease and death in patients with non-dialysis chronic kidney disease.

Abstract

Despite the widespread use of erythropoiesis-stimulating agents (ESAs) to treat anaemia, the risk of adverse outcomes associated with the use of different types of ESAs in non-dialysis chronic kidney disease (CKD) is poorly investigated. From a pooled cohort of four observational studies, we selected CKD patients receiving short-acting (epoetin α/β; n = 299) or long-acting ESAs (darbepoetin and methoxy polyethylene glycol-epoetin β; n = 403). The primary composite endpoint was end-stage kidney disease (ESKD; dialysis or transplantation) or all-cause death. Multivariable Cox models were used to estimate the relative risk of the primary endpoint between short- and long-acting ESA users. During follow-up [median 3.6 years (interquartile range 2.1-6.3)], the primary endpoint was registered in 401 patients [166 (72%) in the short-acting ESA group and 235 (58%) in the long-acting ESA group]. In the highest tertile of short-acting ESA dose, the adjusted risk of primary endpoint was 2-fold higher {hazard ratio [HR] 2.07 [95% confidence interval (CI) 1.37-3.12]} than in the lowest tertile, whereas it did not change across tertiles of dose for long-acting ESA patients. Furthermore, the comparison of ESA type in each tertile of ESA dose disclosed a significant difference only in the highest tertile, where the risk of the primary endpoint was significantly higher in patients receiving short-acting ESAs [HR 1.56 (95% CI 1.09-2.24); P = 0.016]. Results were confirmed when ESA dose was analysed as continuous variable with a significant difference in the primary endpoint between short- and long-acting ESAs for doses >105 IU/kg/week. Among non-dialysis CKD patients, the use of a short-acting ESA may be associated with an increased risk of ESKD or death versus long-acting ESAs when higher ESA doses are prescribed.