Predictors of response to iron and erythropoietin stimulating agents in haemodialysis patients

Abstract

Renal anaemia affects a large proportion of haemodialysis patients. Treatment for this is with erythropoietin stimulating agents (ESA) and intravenous iron, response to which is unpredictable. These medications are titrated according to haemoglobin and ferritin levels which are measured monthly in these individuals. This study aims to evaluate the role of alternate biomarkers in predicting a treatment response. Method This study randomised individuals with a haemoglobin (Hb) between 90 and 104g/l and ferritin between 100 and 800ng/l to either an incremental increase in ESA dose or a course of intravenous iron sucrose, with bloods for biomarkers taken prior to the receipt of this medication. A positive treatment response was defined as an increase in haemoglobin of 5g/l within two months. Non-responders received the alternate treatment with response re-assessed. Results Over 2423 patient months amongst 196 individuals, there were 160 completed randomisation episodes. Response to ESA was seen in 61/84 (72.6%) and to iron in 53/76 (69.7%). C-reactive protein (CRP) was the strongest predictor of a positive treatment response (13.46mg/l versus 28.59mg/l in non-responders, p=0.038) with a lesser impact of a lower baseline ESA dose (2529.5 versus 2913.0 units/session, p=0.243). Successful iron treatment was associated with lower reticulocyte haemoglobin (33.8 vs 35.5pg, p=0.047), lower hepcidin (91.4 vs 131.0ug/ml, p=0.021) and higher C-reactive protein (29.5 vs 12.6mg/l, p=0.085). Interaction modelling indicated that CRP (p=0.08) and ESA dose (p=0.058) were the strongest differentiators of response to the two different treatments. Discussion Increased ESA and iron are equally effective, though treatment failure occurs in almost 30%. Baseline variables can predict treatment response, and a four-variable score shows promise in allowing directed treatment with improved response rates.