The effect of delayed graft function and immunosuppressive drugs on posttransplant erythropoiesis was studied prospectively in 18 living-related (LR) and 84 cadaver-donor (CD) recipients. Eight of 18 LR and 20 of 84 CD recipients received antilymphoblast globulin (ALG) in addition to azathioprine and prednisone. Sixty-four CD recipients received cyclosporine (CsA) with prednisone. In the absence of rejection reticulocytosis began 6.7 +/- 0.2 days following graft implantation in azathioprine-only-treated LR recipients. This was lengthened by ALG to 9.4 +/- 0.3 and 9.9 +/- 0.7 days in LR and CD recipients, respectively, whose grafts functioned immediately. Delayed graft function prolonged onset of reticulocytosis to 15.9 +/- 0.9 days in ALG-treated but not in CsA-treated recipients (5.8 +/- 0.4 days). The shortest latency was noted in CsA-treated recipients (4.9 +/- 0.5 days) with immediately functioning grafts. The earlier onset of reticulocytosis of CsA-treated recipients was followed by statistically significant blunting of peak reticulocytosis, which correlated with a slower rate of correction of anemia (delta Hct = 0.19/day) compared with non-CsA-treated recipients (delta Hct = 0.34/day). Early rejection was associated with abrogation of reticulocytosis and correction of anemia without regard to immunosuppressive regimen) until rejection was reversed. Erythropoietin (EPO) was measured sequentially in 5 patients with immediate function. In 4 of 5 cases changes in EPO preceded those in reticulocytosis. EPO rose from a mean of 13 mU/ml pretransplant to a peak of 50 within 3 weeks and decreased to 18 mU/ml within 6 weeks of graft implantation. At six months posttransplant, normalized reticulocyte counts were only 55% higher (1.75 vs. 1.13%) but hematocrit had increased from 26 +/- 1% to 42 +/- 1%. Hematocrit varied inversely with serum creatinine, which was highest in CsA-treated patients with initial delayed graft function. We conclude that correction of anemia posttransplantation is driven by EPO but other factors may also be important, that neither ATN nor ALG-therapy have clinically important effects on erythropoiesis, and that CsA reduced "effective" erythropoiesis and influences correction of anemia--particularly if delayed graft function complicates the initial course posttransplantation.