Sirolimus (SRL) has a concentration-related effect on hematopoiesis. In this study, 430 renal transplant recipients were randomized (1:1) 3 months post-transplantation to continue SRL-cyclosporine (CsA)-steroids (ST) or to have CsA withdrawn (SRL-ST). Over 5 years, on therapy calculated glomerular filtration rate (GFR), hematological indices, erythropoietin (EPO) use, and rates of mild, moderate, and severe anemia were determined. Longitudinal analyses using linear mixed models examined covariates predicting hemoglobin (Hgb) levels. Mean Hgb was significantly lower with SRL-ST at 6 months; but subsequently became significantly higher (at 2 years, 129 vs. 135 g/l, SRL-CsA-ST vs. SRL-ST, P<0.001). Mean corpuscular volume was low with both therapies, and significantly lower with SRL-ST. EPO use was similar in the two groups, approximately 30% during the first year and 10% thereafter. The incidence of anemia was significantly higher with SRL-CsA-ST>or=2 years. At year 5, only 39.1% of SRL-CsA-ST patients had normal Hgb vs. 68.5% of SRL-ST patients. GFR and recipient age as well as the interaction term x treatment time were significant covariates predicting Hgb. CsA withdrawal followed by SRL immunotherapy resulted in significantly less anemia than SRL-CsA-ST, despite twofold higher SRL exposure. This suggests that the improvement in GFR accompanying CsA withdrawal may mitigate the effect of SRL on hematopoiesis. (ClinicalTrials.gov number: NCT00428064).