Should we adjust erythropoiesis-stimulating agent dosage to postdialysis hemoglobin levels? A pilot study

BACKGROUND: Predialysis hemoglobin (Hb) may overestimate the true erithropoiesis-stimulating agents (ESA) requeriments. We tested whether predialysis Hb is a reliable predictor of the postdialysis level to better control ESA dosage, and evaluated the relation between ESA, Hb and cardiovascular events (CVE). METHODS: Cohort study including 67 stable hemodialysis patients. Pre- and post-dialysis Hb concentrations were measured, and ESA doses were calculated. A model to predict post-dialysis Hb is proposed. During 18 months follow-up, CVE, hospitalizations and mortality were collected. RESULTS: After dialysis, Hb cocentration rise by 6.1 ± 5.6%. Using postdialysis Hb, the weight-adjusted ESA dosage would be lower respect to the prescription using predialysis Hb: 104 ± 120 vs 128 ± 124 U/kg/week (P < 0.001). Using predialysis Hb, 40.2% of subjects had a Hb level above 12 g/dL, whereas this percent increased to 70.1% using postdialysis Hb. During the follow-up, 15 patients had a CVE, without differences in Hb levels respect to subjects without CVE. However, patients with CVE had received higher ESA doses: 186 ± 180 vs 111 ± 98 U/Kg/week (P = 0.001). The prediction model is: Postdialysis Hb (g/dL) = 1.636 + 0.871 x predialysis Hb* (g/dL) + 0.099 x UF rate** (mL/kg/h) - 0.39 for women***. [R(2) = 0.74; *P < 0,001; **P = 0.001; ***P = 0.03). CONCLUSIONS: Postdialysis Hb can be a better reflect of the real Hb level in hemodialysis patients. Using postdialysis Hb would avoid the use of inappropriately high ESA doses. The prediction of postdialysis Hb with an adjusted model would help us to identify those patients at risk for ESA overdosification.