Association of plasma F2-isoprostanes and isofurans concentrations with erythropoiesis-stimulating agent resistance in maintenance hemodialysis patients

BACKGROUND In patients undergoing maintenance hemodialysis (HD), hyporesponsiveness to erythropoiesis stimulating agents (ESAs) is associated with adverse clinical outcomes. Systemic inflammation is highly prevalent in HD patients and is associated with ESA hyporesponsiveness. Oxidative stress is also highly prevalent in HD patients, but no previous study has determined its association with ESA response. This study assessed the association of plasma markers of oxidative stress and inflammation with ESA resistance in patients undergoing maintenance HD. METHODS We analyzed data from 165 patients enrolled in the Provision of Antioxidant Therapy in Hemodialysis study, a randomized controlled trial evaluating antioxidant therapy in prevalent HD patients. Linear and mixed-effects regression were used to assess the association of baseline and time-averaged high sensitivity F2-isoprostanes, isofurans, C-reactive protein (hsCRP), and interleukin-6 (IL-6) with ESA resistance index (ERI), defined as the weekly weight-adjusted ESA dose divided by blood hemoglobin level. Unadjusted models as well as models adjusted for potential confounders were examined. Predicted changes in ERI per month over study follow-up among baseline biomarker quartiles were also assessed. RESULTS Patients with time-averaged isofurans in the highest quartile had higher adjusted mean ERI compared with patients in the lowest quartile (β = 14.9 ng/ml; 95% CI 7.70, 22.2; reference group <0.26 ng/ml). The highest quartiles of hsCRP and IL-6 were also associated with higher adjusted mean ERI (β = 10.8 mg/l; 95% CI 3.52, 18.1 for hsCRP; β = 10.2 pg/ml; 95% CI 2.98, 17.5 for IL-6). No significant association of F2-isoprostanes concentrations with ERI was observed. Analyses restricted to baseline exposures and ERI showed similar results. Baseline hsCRP, IL-6, and isofurans concentrations in the highest quartiles were associated with greater predicted change in ERI over study follow-up compared to the lowest quartiles (P = 0.008, P = 0.004, and P = 0.04, respectively). There was no association between baseline F2-isoprostanes quartile and change in ERI. CONCLUSIONS In conclusion, higher concentrations of isofurans, hsCRP and IL-6, but not F2-isoprostanes, were associated with greater resistance to ESAs in prevalent HD patients. Further research is needed to test whether interventions that successfully decrease oxidative stress and inflammation in patients undergoing maintenance HD improve ESA responsiveness.