Stem Cell Procurement

Safety of hematopoietic stem cell donation in glucose 6 phosphate dehydrogenase-deficient donors

Article metrics


Glucose-6-phosphate dehydrogenase (G6PD) deficiency is the most common RBC enzymatic disorder in humans capable of producing hemolytic events. Recently, concern has been raised about using G6PD-deficienct subjects as hemopoietic stem cell (HSC) donors. In a 10-year period, 101 consecutive HSC donors were submitted to donation procedures for transplantation inside their families in our Center. All donors were tested for G6PD and 19 (19%) turned out to be G6PD-deficient. The donors’ safety and the effectiveness of these transplant outcomes were compared with those of the remaining 82 donors. No difference could be observed in any safety parameter between the two groups. No difference was recorded in donors’ complications rates, in HSC production, in quantity of growth factor required, in Hb early drop or in Hb recovery. No difference was found in transplant outcome. From this retrospective analysis, we conclude that a G6PD-deficient but otherwise healthy volunteer can be selected as a HSC donor.


Glucose-6-phosphate dehydrogenase (G6PD) deficiency is an X-linked condition1 characterized by a markedly reduced capability to protect red blood cells from oxidative stresses. Carriers of this condition are considered healthy, but in certain circumstances, like exposure to a certain drugs, infections or contact with fava beans, they can develop severe hemolytic events. Worldwide, >400 million people are carriers of this condition, mainly in the malaria endemic area.2, 3 Recently, concern has been raised about the use of G6PD-deficient, but otherwise healthy subjects as hemopoietic stem cell (HSC) donors for transplantation. These concerns were for donor safety, use of granulocyte growth factors (G-CSF) and recipient security.

Apart from a limited experience with Chinese donors and recipients (five G6PD-deficient donors were studied),4, 5 no report has been published in the medical literature.

Our Center is the referring cancer and transplant center on the island of Sardinia, where the prevalence of G6PD-deficient subjects is reported to be as high as 12%.6 In our Center, G6PD screening for a G6PD-deficient healthy subject to be a HSC family donor has been included in standard operative procedures since 2001. Therefore, we had the opportunity to retrospectively investigate the use of G6PD-defective donors regarding donor safety and transplant outcome.

Subjects and methods

Since 2001, our transplant standard operative procedures included the possibility of using a G6PD-deficient but otherwise healthy family member as a HSC donor. In the case of multiple HLA-identical family donors, G6PD donor level was not considered as a selection criteria. In the case of the availability of an HLA-identical family member who was also G6PD-deficient, a donor search for a non-relative match was not taken into account.

Donor study

Complete donor evaluation and selection was performed following the European Bone Marrow Transplant guidelines.7 A full blood count was repeated just before starting HSC stimulation with G-CSF or harvesting procedure. Independently from source of stem cell, G6PD quantitative estimation was detected by standard spectophotometric method8 (Keylab automatic analyzer BPC+Biosed, Rome, Italy).

G6PD deficiency was defined as an enzyme dosage of <0.96 UI/g of Hb (U/g Hb) in the peripheral blood.

PBSC donors received 5 μg/kg granulocyte CSF (G-CSF), Filgrastim, twice a day for 4 days before HSC aphaeresis. A single dose was administered in the morning of the aphaeresis for a total of nine doses.9 CD34+ cell count was determined daily until collection. If necessary, additional doses of G-CSF were administered and further aphaeresis was arranged based on CD34+ cell production.

BM donors did not receive growth factor and received one or two autologous packed RBC (PRBC) units of support during and immediately after donation.

As per common practice, HSC donors were evaluated for hemolysis and all other complications commonly related to G-CSF administration following the The National Comprehensive Cancer Network guidelines for supportive care, myeloid growth factor in the available version following the periodic updating. For this report data have been analyzed according to the 2011 version ( A complete donor assessment was performed before marrow harvest, or before and during CD34+ cell stimulation. Subsequently, donors were evaluated at day+1 and at day+30 after collection independently from source of stem cell.

Patient study and transplant management.

Patients were evaluated for G6PD before transplant. A total of 25 of them were G6PD deficient. Our standard operative procedures did not include any difference in donor or patient management on the basis of G6PD level including the administration of sulfamethoxazole-trimethoprim at prophylaxis dose.10

As is common for transplant care, HSC recipients were evaluated daily for all possible complications during the transplant phase until discharge and 2–3 times weekly thereafter, at least up to day+100.

All recipients were submitted to a myeloablative-conditioning regimen. GVHD prophylaxis was performed with cyclosporine and MTX. Antiviral prophylaxis was performed with acyclovir, antifungal prophylaxis with fluconazole and anti Pneumocisti jiroveci with low sulfamethoxazole-trimethoprim prophylactic doses (5 mg trimethoprim/kg/day for 2 days/week). Details on the transplant procedure have been described elsewhere.10

In the post-transplant period, we evaluated other clinical complications in addition to transplant and disease: erythroid engraftment by the need of transfusions; indirect hemolytic signs such as lactate dehydrogenase and bilirubin following the differential diagnosis hemolysis criteria in transplanted patients by Gajewski et al.;11 adverse events following NCI Common Toxicity Criteria, version 3.0 9 August 2006 (; global engraftment by molecular biology markers or cytogenetic analysis.

Patients who had received hematopoietic stem cells from a G6PD-deficient donor were compared with all those who had G6PD-normal donors. Only first-time transplantations were considered and no selection was performed.

Statistical analysis

Data with a normal distribution are expressed as means±s.d.; those with a skewed distribution are expressed as medians and ranges (minimum–maximum). The unpaired Student’s t-test for means was used to assess the significance of difference between a comparison obtained in cases and controls for normally distributed values, and the Wilcoxon test was used for comparison of ordered values. The Fisher exact test was used for contingency tables. Survival probability was computed with the method of Kaplan and Meier. Survival curves were compared with the log-rank test. All statistical tests were two-tailed.

The study was approved by an internal review board.


From 19 September 2001 to 25 March 2011, 101 consecutive healthy HSC donors were submitted to donation procedures for related donor transplantation. There were 31 BM and 70 PBSC donors. All allogeneic healthy donors successfully mobilized CD34+ cells in this 10-year period.

All donors were tested for G6PD and 19 (19%) were found to be G6PD-deficient with a median value of 0.1 UI/g Hb. The remaining 82 donors served as the control. Table 1 reports G6PD values and other characteristics of the study and control groups. Figure 1 reports the G6PD value distribution in all 101 donors.

Table 1 Donors’ characteristics
Figure 1

Distribution plot of the G6PD enzyme levels in deficient and normal donors.

Donor safety

Table 1 reports peripheral blood CD34-positive cell production and nucleated cells harvested (BM). Delayed harvest with additional G-CSF administration (1 day=two G-CSF administrations) was necessary in 3/19 (15.7%) donors in the study group and in 11/82 (13.4%) in the control group (P=0.7).

In the group of normal donors there were 3 (3.5%) adverse events: a grade II allergic reaction, a grade III acute broncoconstriction leading to mild respiratory failure and an episode of grade I fever. All donors recovered in a few hours. All complications occurred in patients who underwent BM harvest and all were probably related to general anesthesia. In the 19 G6PD-deficient donors no adverse event was registered (P=1).

Differences between Hb levels (ΔHb) determined immediately before starting the donation procedure (baseline), at day +1 after donation and at day +30, were evaluated. In PBSC donors, ΔHb between baseline and day+1 was −0.26±0.7 g/dL and −0.2±0.6 g/dL in the study and control groups, respectively (P=0.48). ΔHb between day +1 and day+30 was +0.5±1 and +0.5±1 g/dL in the study and control groups, respectively (P=0.85). Similar data were reported for BM donors (data not shown). Figure 2 shows ΔHb between baseline and day +1 in G6PD-deficient and in normal PBSC donors.

Figure 2

Hb variations (Δ Hb) in PBSC donors before and after collection. Hb variation was calculated comparing Hb level at screening visit and at day +1. Mean difference values were –0.26±0.7 and 0.2±0.6 in the study and control groups, respectively (P=0.48). Transversal lines indicate mean values.

Transplant outcome

Patients’ characteristics and transplant outcome data are shown in Table 2. The 101 consecutive patients analyzed in this study were divided into ‘standard prognosis’ (patients in their first or second remission or untreated patients with myelodysplastic syndrome) or ‘poor prognosis’ (active disease or in a remission later than the second). There were 8/19 and 32/82 patients with poor prognosis in the study and control groups, respectively (P=0.9).

Table 2 Patients’ characteristics and transplants outcome

The 100- and 365-day Kaplan–Meier probability of survival rates were 89% versus 84% and 52% versus 55% in the study group and in the control group, respectively (P=0.9). Figure 3 shows Kaplan–Meier overall survival in the two groups.

Figure 3

Kaplan–Meier survival curves of transplanted patients who received a G6PD-deficient donor (continuous line) and a normal donor (dotted line). Log-rank test: P=0.9.

No differences in erythroid recovery and supportive care were observed. In all, 10 patients presented serum bilirubin increase in the first 20 days after transplant. Only one of these events belonged to the study group (P=0.7). In this case, bilirubin peaked at day +8 and was predominantly un-conjugated (maximum level 17 mg/dL—grade IV toxicity). It spontaneously reversed in a few days without any clinical consequence. In all the other cases registered in the control group, bilirubin elevation was clearly related to transplant (that is, GvHD) or disease complication. The median value of lactate dehydrogenase up to day+14 was 450 u/L (range 250–776) in the study group and 400 u/L (range 148–1299) in the control group.

At day 30, full donor chimerism occurred in all 19 patients in the study group (median day +25) and in all but 1 of the 73 patients who could be evaluated in the control group (median day +35; P=1).

In patients who had received a G6PD-deficient donor, enzyme levels were evaluated after sustained engraftment was achieved. Nine recipients who were alive and well 1 year after transplant were studied (Figure 4). In this group, the median G6PD level before transplant was 1 U/g Hb (range 0.25–1.26) and 0.21 U/g Hb (range 0.02–0.93) at post-transplant control (P=0.07). The median value in these nine donors was 0.39 U/g Hb (range 0.05–0.94) (donor versus patients post-BMT: P=0.65). None of the patients who acquired G6PD deficiency presented G6PD-related complications in the long-term follow-up.

Figure 4

Distribution plot comparing G6PD enzyme levels in HSC donors and in recipients before and after transplant. All the nine recipients who were alive and well >1 year after and who have received a G6PD-deficient donor were included.


Glucose 6 phosphate dehydrogenase-deficiency is one of the most common genetic conditions, involving >400 million people worldwide. G6PD-deficient subjects are healthy apart from the risk of hemolytic crisis in response to oxidative injuries. Their life expectancy is comparable to that of the normal population.

Since the 1980s, practitioners of transfusion medicine have been divided on the use of a G6PD-deficient blood donor. Conflicting results have been reported on post-transfusion hemolytic risk in the recipient of a G6PD-deficient donor’s blood.12, 13

Recently, the Italian BM donor registry, answering an approval query for PBSC donation from a G6PD-deficient candidate match for an unrelated donor, disseminated a warning about this condition. The warning, and the refusal to use this donor, was prompted by the absence of safety data (particularly regarding use of G-CSF) in this group of volunteer donors.

We evaluated our 10-year consecutive population and could demonstrate that there were no differences to potentially bias the results of our analysis.

No statistically significant difference, nor any ‘trend’ could be observed in any safety parameter. No difference was recorded in HSC mobilization, in quantity of growth factor required, in Hb early drop, or in day 30 Hb recovery. Analysis of donation complications did not show any differences between the two groups, with mild complications reported only for BM donations.

G6PD-deficient donors did not present any increased hemolytic (or other) complications compared with a similar G6PD-normal population during the process of stimulation with G-CSF or during the collection of hematopoietic stem cells.

When we looked to recipients’ outcomes again, we could not find any difference in survival rates between the two groups. Even detailed analysis of engraftment and erythroid engraftment showed no difference despite the use of low doses of sulfamethoxazole-trimethoprim (a drug considered responsible for oxidative insult in G6PD deficiency). No differences in blood support or in indirect hemolytic signs could be detected. There was a single episode of pre-hepatic bilirubin increase that could possibly be related to the engrafted donor red cells. However, this event was transitory and spontaneously resolved itself without any specific medical intervention.

Finally, recipients acquired G6PD donor deficiency status, but this fact did not imply any clinical significance.

On the basis of this analysis, we conclude that a G6PD-deficient healthy volunteer can be selected as a HSC donor. However, the size of this cohort is not large enough to detect rare side effects; therefore, an appropriate surveillance of donors and patients for possible hemolytic or other events must be carried out. At the same time, patients should be informed about this enzyme deficiency acquisition.


  1. 1

    Beutler E . Gene inactivation: the distribution of gene products among populations of cells in heterozygous humans. Cold Spring Harb Symp Quant Biol 1964; 29: 261–271.

  2. 2

    Beutler E . Glucose-6-phosphate dehydrogenase deficiency. N Engl J Med 1991; 324: 169–174.

  3. 3

    Beutler E . The genetics of glucose-6-phosphate dehydrogenase deficiency. Semin Hematol 1990; 27: 137–164.

  4. 4

    Au WY, Ma SK, Lie AK, Liang R, Cheng T, Kwong YL . Glucose-6-phosphate dehydrogenase deficiency and hematopoietic stem cell transplantation. Bone Marrow Transplant 2002; 29: 399–402.

  5. 5

    Au WY, So JC, Ma SK, Lie AK . Glucose-6-phosphate-dehydrogenase deficiency and haematopoietic stem cell transplantation in Chinese patients. Hong Kong Med J 2009; 15 (Suppl 3): 35–38.

  6. 6

    Testa U, Meloni T, Lania A, Battistuzzi G, Cutillo S, Luzzatto L . Genetic heterogeneity of glucose 6-phosphate dehydrogenase deficiency in Sardinia. Hum Genet 1980; 56: 99–105.

  7. 7

    Larghero J, Garcia J, Gluckman E . sources and procurement of stem cells. In: Haematology. ESo (ed) Hemopoietic Stem Cell Transplantation. The EBMT Handbook. 5th edn (Forum service editore: Genoa), 2008, pp 112–127.

  8. 8

    Beutler E . Red Cell Metabolism. A Manual of Biochemical Methods, 2nd edn. Grune & Stratton: New York, 1975.

  9. 9

    Ozer H, Armitage JO, Bennett CL, Crawford J, Demetri GD, Pizzo PA et al. 2000 update of recommendations for the use of hematopoietic colony-stimulating factors: evidence-based, clinical practice guidelines. American Society of Clinical Oncology Growth Factors Expert Panel. J Clin Oncol 2000; 18: 3558–3585.

  10. 10

    Baronciani D, Rambaldi A, Iori AP, Di Bartolomeo P, Pilo F, Pettinau M et al. Treosulfan/fludarabine as an allogeneic hematopoietic stem cell transplant conditioning regimen for high-risk patients. Am J Hematol 2008; 83: 717–720.

  11. 11

    Gajewski JL, Johnson VV, Sandler SG, Sayegh A, Klumpp TR . A review of transfusion practice before, during, and after hematopoietic progenitor cell transplantation. Blood 2008; 112: 3036–3047.

  12. 12

    Shalev O, Manny N, Sharon R . Posttransfusional hemolysis in recipients of glucose-6-phosphate dehydrogenase-deficient erythrocytes. Vox Sang 1993; 64: 94–98.

  13. 13

    Huang CS, Sung YC, Huang MJ, Yang CS, Shei WS, Tang TK . Content of reduced glutathione and consequences in recipients of glucose-6-phosphate dehydrogenase deficient red blood cells. Am J Hematol 1998; 57: 187–192.

Download references


This clinical research was supported by AIL Associazione Italiana contro le Leucemie—Cagliari section. We thank Carlo Brugnara, MD for critical reading and suggestions.

Author information

Correspondence to E Angelucci.

Ethics declarations

Competing interests

The authors declare no conflict of interest

Rights and permissions

Reprints and Permissions

About this article


  • G6PD deficiency
  • hemopoietic SCT
  • hemopoietic stem cell donors.

Further reading