Dynamics of erythropoietic recovery following bone marrow transplantation: role of marrow proliferative capacity and erythropoietin production in autologous versus allogeneic transplants.

The mechanisms of erythrocyte recovery after BMT are not known. We investigated the respective role of marrow function and erythropoietin production in 31 ABMT and 47 allogeneic BMT by analysing peripheral counts, serum erythropoietin levels, and serum transferrin receptor (TfR) levels which have been shown to be a quantitative measurement of erythropoiesis. Median times to complete neutrophil (25 vs 48 days, p < 0.0001) and platelet (45 vs 263 days, p < 0.001) recovery were faster after allogeneic BMT than ABMT, but complete erythrocyte recovery was slower (218 vs 101 days, p < 0.001). After ABMT, erythrocyte recovery paralleled that of neutrophils and platelets, and erythropoietin levels remained appropriate for the degree of anemia. After allogeneic BMT, erythrocytes developed independently of the other cell lines and defective erythropoietin production delayed recovery of adequate erythropoietic activity. This correlated with an alteration of renal function only in those patients remaining erythropoietin deficient beyond day 180. However, supranormal erythropoietin levels in interstitial pneumonia suggests that erythropoietin response to hypoxia is not abrogated. CMV infection could also affect erythropoiesis through erythropoietin production after ABMT as well as allogeneic BMT. It is concluded that after ABMT the development of erythropoiesis is determined by the overall marrow proliferative activity and erythropoietin plays only a facilitating role. After allogeneic BMT, erythropoiesis depends on erythropoietin levels which remain inadequate for prolonged periods of time. The results suggest that the administration of recombinant human erythropoietin could reduce transfusion requirements after BMT.

Circulating erythropoietin levels after bone marrow transplantation: inappropriate response to anemia in allogeneic transplants.

We studied 24 recipients of autologous bone marrow transplantation (ABMT) or allogeneic BMT (BMT) to determine whether impaired erythropoietin (Epo) response to anemia could delay full erythropoietic recovery. Observed Epo levels were compared with predicted levels based on the relationship between Epo and hematocrit in 125 control subjects. Circulating Epo levels were normal during conditioning and the early posttransplant period. Between days 21 and 180, Epo levels remained normal in ABMT patients but were inappropriately low for the degree of anemia in BMT patients. Median time to full erythropoietic engraftment was longer in BMT than in ABMT recipients. Circulating Epo returned to appropriate levels after day 180, except in patients with active cytomegalovirus infection. We conclude that impaired Epo response to anemia can contribute to delayed erythropoietic recovery after allogenic BMT. Renal toxicity of ciclosporin, interaction between host and donor marrow, and cytomegalovirus infection might play a role. This study could support the use of recombinant human Epo to accelerate erythropoietic engraftment after BMT.

Serum immunoreactive erythropoietin during pregnancy and in the early postpartum.

We studied 209 women during normal pregnancy, at delivery, or in the early postpartum, to determine whether erythropoietin (EPO) response was appropriate for the degree of anaemia. Serum immunoreactive EPO was measured in 74 nonpregnant women, including 33 normal subjects (16.4 +/- 4.1 mU/ml) and 41 women with hypoplastic, haemolytic, dyserythropoietic, or iron-deficient anaemia. An inverse linear relationship (R = -0.88, P less than 0.0001) between log(EPO) and Hct was observed. Predicted EPO values were derived for each Hct and an O/P ratio of observed/predicted log(EPO) was calculated for each sample (1.00 +/- 0.10, range 0.80-1.20). Serum EPO levels (mU/ml) were significantly higher during pregnancy (30 +/- 16, n = 142), at delivery (31 +/- 16, n = 41), and on day 7 postpartum (37 +/- 35, n = 26) than in normal women (P less than 0.001). EPO levels increased steadily from 18 +/- 6 mU/ml in the first, to 26 +/- 14 mU/ml in the second, and to 35 +/- 18 mU/ml in the third trimester (P less than 0.0001). The O/P ratio was normal on day 7 postpartum (1.01 +/- 0.16), at delivery (1.03 +/- 0.16), and in the third trimester (0.96 +/- 0.15), but was significantly reduced in the first two trimesters (0.88 +/- 0.15, P less than 0.001). A significant negative correlation between log(EPO) and Hct was lacking in the first two trimesters, was present but with a reduced slope during the third trimester and at delivery, and was normal postpartum. We conclude that EPO response to anaemia is impaired in early pregnancy, recovers in late pregnancy, and normalizes rapidly in the postpartum.