Stem cell transplantation is a potential curative treatment for hematologic malignancies. This treatment often requires repeated packed red blood cell (PRBC) transfusion requirements raising the risk of secondary hemochromatosis with each successive unit transfused. Erythropoietin-assisted phlebotomy is a potential treatment to alleviate iron overload in transfusion-dependent patients.
We carried out a retrospective review from 2000 to 2008 of records of both allogeneic and autologous stem cell transplant recipients treated at San Antonio Military Medical Center South for a total of 372 patients. We evaluated the maximum level of ferritin recorded for each patient and used an upper level of 1,000 ug/mL (approx 2.5x upper level of normal) to identify those patients at increased risk for complications of iron overload. 65 patients had ferritin levels <1,000ng/mL and served as a control group. There were 56 patients who met our criteria for iron overload. Their transfusion records were reviewed and further subdivided into those who received therapeutic erythropoietin-assisted phlebotomy and those who did not. They were also reviewed for evidence of any opportunistic infection during their treatment course.
Of the 56 patients with documented ferritin levels over 1,000ug/mL, the average ferritin level was 2,965ug/mL. The transfusion records for these patients were obtained and showed that they received an average of 26.6 units of transfused PRBCs during their transplant course. Ten of these patients underwent therapeutic erythropoietin-assisted phlebotomy of 500cc of whole blood approximately once every two weeks to decrease their ferritin levels with a stated goal of <500ugmL. All records were then reviewed for evidence of opportunistic infections complicating treatment and corresponding mortality. The non-phlebotomized group had an increased rate of infections and trended toward a higher mortality rate. Overall infections were significantly different, with the hyperferritinemia group who did not receive phlebotomy having the highest rates within our three groups at both the three and six month intervals (p<0.004 and p<0.008). These differences were not significant at one year (p<0.086). When we compared the two main groups (ferritin <1,000 ng/ml and >1,000 ng/ml), infection rate results were higher in the hyperferritinemia group at all three time intervals (p<0.003, p<0.008 and p<0.049). There was no significant difference between infection rates in the erythropoietin-assisted-phlebotomy group and the control group at either the 3 or 6 month intervals. (p<0.17 and p<0.107) Mortality differences between the two main groups (ferritin >1,000ng/ml and ferritin <1,000 ng/ml) were significant at both six months and one year (p<0.0004 and p<0.005 respectively). There was also a significant difference noted between the hyperferritinemia group who did not receive phlebotomy and the ferritin <1,000 ng/ml group at both six months and one year (p<0.00002 and p<0.0004 respectively). Two-tailed Fishers exact test showed that the mortality rate between the two subsets in the hyperferritinemia group was significant at six months (p=0.009 two sided; p=0.005 one sided) but not at one year (p=0.34, two-sided; p=0.19 one sided). There was no significant difference in mortality rates between the erythropoietin-assisted-phlebotomy group and the control group.
Iron overload can be insidious. Multiple transfusions raise the risk for ferritin levels ≥1,000 ug/mL and predispose patients to secondary hemochromatosis which is a risk factor for opportunistic fungal and bacterial infections and may lead to increased mortality rates. Erythropoietin-assisted phlebotomy is a potential treatment to prevent the complications of iron overload in HSCT patients.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.