Body fat mass and serum leptin levels influence epoetin sensitivity in patients with ESRD

Background. Dose requirements of epoetin vary considerably among patients with end-stage renal disease (ESRD), whereas determinants of epoetin sensitivity are poorly understood. Fat mass is an important source of adipokines, including interleukin 6 (IL-6), which is associated with decreased epoetin sensitivity. Furthermore, the adipokine leptin stimulates human erythroid development in vitro. In the present study, we investigate the impact of fat mass and leptin level on epoetin sensitivity in patients with ESRD. Methods: One hundred sixty-six patients with ESRD (107 men; 64%) with a mean age of 56.9 +/- 0.9 years were studied in a post hoc cross-sectional analysis. Body composition was analyzed by dual-energy X-ray absorptiometry and correlated with serum markers of inflammation and leptin (analyzed by enzyme-linked immunosorbent assays), as well as with epoetin dose, in international units administered per week (IU/wk). To correct for differences in body mass and hemoglobin (Hb) levels, epoetin sensitivity was approximated as epoetin/Hb ratio, le, epoetin dose per unit of Hb (IU/wk/g/dL) and epoetin/Hb/kg ratio, ie, epoetin dose per unit of Hb and kilogram of patient body weight (IU/wk/Hb/kg). Results Patients were divided into 3 groups according to epoetin/Hb/kg ratio (no-epoetin group, low-epoetin group, and high-epoetin group). The 3 groups had significantly different serum levels of high-sensitivity C-reactive protein (hsCRP; median, 8.6 versus 3.1 and 8.0 mg/L, respectively; P < 0.05), neopterin (median, 112.4 versus 94.3 and 96.1 ng/L, respectively; P < 0.05), and IL-6 (median, 6.8 versus 4.1 and 6.5 ng/mL, respectively; P < 0.05). Significant between-group differences also were found in fat mass and leptin levels (median, 14.8 versus 10.5 and 7.9 ng/mL, respectively; P = 0.02). In univariate analyses, significant relationships between epoetin sensitivity indices, leptin levels, and levels of the inflammatory markers hsCRP and IL-6 were found. In a multivariate stepwise regression model, log ferritin, parathyroid hormone, log leptin, log IL-6, and polycystic kidney disease were significantly associated with the epoetin/Hb ratio. Conclusion: The present study shows that leptin level may be a predictor of epoetin sensitivity. The effect could be either direct stimulation of erythropoiesis or indirect stimulation by associated adipokines. Although truncal fat is associated with secretion of proinflammatory cytokines, this secretion appears not to have inhibitory effects on epoetin sensitivity in the presence of high leptin levels.