Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Stem Cell Procurement

Pelvic magnetic resonance imaging after bone marrow harvest – a retrospective study in 50 unrelated marrow donors


A total of 50 unrelated marrow donors were examined by pelvic magnetic resonance imaging (MRI) to investigate the morphological sequelae of bone marrow harvesting (BMH). Signal increase in T2-weighted sequences and contrast media enhancement in T1 sequences at the operative sites were found as typical MRI morphology 4 weeks after harvest (group A, n=16), corresponding to edema, hyperemia and proliferative activity. Although tissue repair was completed in the majority of donors 1 year after BMH, about 36% of donors in group B (n=16) had abnormal findings. These included a persistence of the ‘acute injury’ signal pattern (2/16, 12%), and signal alterations due to fatty marrow conversion (4/16, 24%). The proportion of MRI abnormalities increased to over 70% in two-time donors (group C, n=11), which might indicate a cumulation of tissue damage after repetitive harvests. If donors had experienced prolonged discomfort after BMH (group D, n=7), MRI revealed pathological signals in 86%. In conclusion, the MRI morphology reflects the pathophysiological reactions after BMH, including inflammation and tissue repair. A further prospective evaluation in a larger number of donors is necessary to confirm these results and to identify the factors which influence the extent and duration of tissue damage.


Bone marrow harvesting (BMH) from the posterior–superior iliac crest is a standard procedure to obtain hematopoietic stem cells (HSC) for transplantation from allogeneic donors. In addition to the collection of adequate numbers of hematopoietic progenitors, comfort and safety for the donors are of the uppermost importance, especially in unrelated volunteers. Several studies evaluated the risks of marrow harvesting in larger series and demonstrated BMH as a very safe procedure,1, 2, 3, 4, 5 with a reported incidence of major, life-threatening complications of less than 0.5%. However, a substantial number of bone marrow donors experience side effects resulting from general or spinal anesthesia, loss of oxygen-transporting erythrocytes or operative site morbidity. Pain at the collection site or in the lower back is the most frequent local problem, occurring with variable severity and duration in about 70–80% of the donors.3, 4, 6, 7 Furthermore, bleedings and hematomas at the collection site,3, 8 transient compressive neuropathies,2, 9 osteomyelitis10 as well as a fracture of the ileum11 were described as rare local complications of BMH. Nevertheless, there is no information available about acute or persisting morphological changes in the bone and soft tissues caused by the multiple punctures, which are needed to collect a sufficient volume of marrow.

The purpose of this retrospective study was to investigate the local alterations at the operative site after a BMH procedure by means of magnetic resonance imaging (MRI). Pelvic MRI was performed in 50 selected unrelated bone marrow donors with or without postsurgical complications and/or prolonged discomfort to assess the short- and long-term effects of BMH to the affected tissues.

Patients and methods

A total of 50 adult, unrelated bone marrow donors from the German Bone Marrow Donor Center (Deutsche Knochenmarkspenderdatei, DKMS, Tuebingen, Germany) were included in this MRI study. They had undergone BMH in one out of eight different German collection centers, either 4 weeks (group A, n=16) or 1 year (group B, n=16) before study entry. The donors in groups A and B were randomly selected and asked for study participation by the donor center. Furthermore, donors after a second BMH (group C, n=11), and donors with a postsurgical complication or prolonged discomfort (group D, n=7), were examined. These participants were all donors with these selection criteria (two-time donor, complication) from the entirety of 2447 volunteers of the DKMS, who underwent a BMH between May 1992 and July 2003. The protocol was approved by the Institutional Review Board of the University Hospital Dresden, Germany. Written informed consent was obtained from all participants.

The harvests were performed between December 1994 and January 2003. All collection centers used common harvest procedures. Marrow had been aspirated exclusively from the posterior–superior iliac crest, which was entered through a skin puncture or a small skin incision. A slightly different technique was used in one of the collection centers, making 16–33 punctures or incisions in a larger area above the posterior–superior iliac crests (six out of the 50 donors). During the harvests, donors were in the prone position and under general anesthesia. The volume of harvested marrow per single aspiration varied from a few to 20 ml. Demographic data and harvest parameters of the study population are listed in Table 1.

Table 1 Characteristics of study population

At the time of MRI, donors were asked to describe postsurgical complaints, especially severity and duration of pain, need for analgesic medication and/or limitations in normal life activities using a standardized questionnaire. Severity of maximum pain within 48 h after the harvest was classified by the donors on a scale from 0 to 4 (0 – no pain, 1 – mild, 2 – moderate, 3 – severe, 4 – very severe pain). In a nonstandardized free interview, the donors were explored for details of BMH-related problems. Physical examination included the inspection of the operative sites, an investigation of the mobility of the lower spinal column, as well as motility, sensibility and reflexes of the lower limbs. Medical history and examination were performed by a hematologist experienced in the field of marrow harvesting.

MRI was performed at a 1.5 T superconductive magnet (Magnetom Vision, Siemens, Erlangen, Germany) using a body phased-array coil. The study protocol included transverse T2- (turbo inversion recovery modulus (TIRM)) and T1-weighted (spin echo (SE)) sequences. The slice thickness was 5 mm, with an interslice gap of 0.5 mm. In patients, who underwent BMH 4 weeks before MRI, contrast medium (Gd-DTPA, Magnevist, Schering, Berlin, Germany) was administered intravenously via an antecubital vein at a dosage of 0.1 mmol/kg, followed by a saline flush of 20 ml. In this group, the axial T1-weighted sequence was repeated after contrast medium administration, applying a fat suppression technique. Total imaging time was 20 min approximately. MRI scans were assessed by two radiologists experienced in musculoskeletal diagnostics and the hematologist, who had also performed the clinical evaluation. The scans were specifically searched for bone and soft tissue injuries, sacroiliac joint (SIJ) damage, marrow morphology and signal alterations after application of the contrast medium. Primary data entry was performed electronically as free text using MedoraCOM software (GE Healthcare, Fairfield, USA). After completion of recruitment, morphology, frequency and interpretation of the observed MRI findings in the different groups were evaluated systematically. Mann–Whitney test was employed for the comparison of independent samples, and a P-value less than 0.05 was considered as statistically significant.


Group A (donors 4 weeks after BMH)

All 16 donors experienced pain on the day of marrow harvest and the day after the procedure. The maximum pain was described as severe or very severe in eight donors (50%), while the symptoms were mild to moderate in the remaining donors. None of the donors reported to have required analgesics. The median time with temporary discomfort at physical exertion or movement was 10 days (Table 1). Five donors reported persisting problems at the time of MRI. Apart from a positive Laségue's sign (70°) in one donor and a slight painful swelling at the operative site in another donor, the physical examinations did not reveal any pathological findings.

Morphological changes in the bone and/or overlying soft tissues could be seen in all of the 16 donors. MRI showed a signal increase in T2-weighted (Figure 1a), and a signal decrease in T1-weighted images (not shown) in the ilium at the posterior–superior spines, the adjacent muscles (gluteus maximus and gluteus minimus) and the connecting soft tissues. After administration of contrast medium, enhancement was observed in the T1-weighted sequences in the affected area (Figure 1b). In nine cases (56%) the signal alterations affected the sacroiliac joint or crossed to the lateral mass of the sacrum, but there was no increase in pain intensity or duration observed in these donors (Figure 4c).

Figure 1

Pelvic MRI from a 25-year-old female bone marrow donor 4 weeks after harvest. Typical hyperintense signal alterations at the posterior–superior iliac spines in the fat-suppressed T2-weighted sequence. There is only slight hyperintensity in the adjacent gluteus muscle (a). The fat-suppressed T1-weighted image after application of contrast medium demonstrates enhancement of numerous tubular structures marking the way of the aspiration needles. Note also minimal enhancement of overlying soft tissues (b).

Figure 4

Influence of harvest volume and SIJ affection on procedure-related symptoms. The harvest volume did not influence either the degree of pain (0 – no pain, 1 – mild, 2 – moderate, 3 – severe, 4 – very severe pain) within 48 h (a, n=50, all donors), or the number of days with any pain after the BMH (b, n=42, donors of groups A–C). An affection of the SIJ during the harvest procedure, which was detectable in a substantial number of donors, had also no impact on postsurgical pain and time for complete recovery (c). This analysis was carried out in all donors of groups A–C which had an abnormal MRI scan (n=30; without SIF affection, n=18; with SIJ affection, n=12). Boxes and lines represent medians and ranges.

Group B (donors 1 year after BMH)

The grading of postsurgical pain in the donors who were examined 1 year after BMH was similar to those in group A (Table 1). Six donors used analgesics (paracetamol) after the harvest. At study evaluation, all donors in this cohort were free of discomfort as well as motoric and/or sensory dysfunctions. MRI showed abnormal findings in six out of 16 donors (36%). Persisting hyperintense signal intensity in T2 sequences was found in two donors (12%) in the posterior–superior iliac spines and the sacroiliac joints, respectively. A different pattern with signal decrease in T2-weighted images with fat suppression and signal increase in T1-weighted images in the iliac bone was seen in four cases (24%) (Figure 2).

Figure 2

A 28-year-old asymptomatic male donor 5 years after the second harvest. The fat-suppressed T2-weighted image shows irregular areas of hypointense signal intensity in the iliac crest and the lateral mass of the sacrum bilaterally (a) that correspond to hyperintense signals in the plain T1-weighted image without fat suppression (b). The pattern of signal alteration indicates a crossing of the sacroiliac joint during the harvest.

Group C (donors after two harvest procedures)

The interval between the second BMH and MRI ranged from 11 months to 8.5 years. Interviews and physical examinations of the donors did not reveal any persisting problems related to the harvest procedures. The majority of donors in this group reported that there was no or only slight pain during the first 48 h after the harvest (Table 1). Only one of them required paracetamol. Four out of 11 donors described an increased discomfort after the second harvest compared to first BMH. One donor had suffered from a unilateral sciatic neuralgia with pain and prickle paresthesias due to a gluteal hematoma.

Morphological changes in MRI were seen in eight out of 11 (73%) donors, which is about twice as much as in the donors of group B. As in group B donors, there was a signal decrease in fat-suppressed T2-weighted images, and a signal increase in T1-weighted images in six out of 11 (55%) participants. In addition, scleroses were detected in two cases (18%) in the posterior–superior iliac crest or along the aspiration ducts. In four donors (36%), the alterations indicated an affection of the sacroiliac joint.

Group D (donors with prolonged discomfort)

Medical histories and MRI findings of the donors of this group are summarized in Table 2 and Figure 3. MRI scans showed abnormal findings in six donors (86%). A definite postsurgical complication had occurred in three donors (one sacroiliitis, two hematomas), whereas three other donors experienced prolonged pain and/or reduced daily activities. Only in the donor, in whom the sacroiliitis had been diagnosed, MRI demonstrated a specific (unilateral) morphology. In an obese female donor who has suffered from low back pain for about 2 years (UPN P32) after BMH, the MRI findings were compatible with a degenerative facet joint arthrosis.

Table 2 Donors with postsurgical complications and/or prolonged discomfort
Figure 3

Pelvic MRI of donors with postsurgical complications or prolonged discomfort (for medical history, see Table 2). UPN P33 (a, b) The fat-suppressed T2-weighted image shows a persisting marked signal increase in the right ilium and the lateral mass of the sacrum, as well as an irregular sacroiliac joint (a). After administration of contrast medium, enhancement occurs in the same area (T1-weighted, fat-suppressed sequence, b). UPN P35 (c, d): The fat-suppressed T2-weighted image (c) and the fat-suppressed contrast-enhanced T1-weighted scan (d) demonstrate signal increase of both iliac crests, affecting also the overlying soft tissues. There is a cystic lesion (arrow) of less than 10 mm in diameter in the left lateral mass of the sacrum, which was interpreted as a small hematoma or seroma.

Influence of donor weight and harvest volume on symptoms and MRI changes

Based on the analyses of donors in groups A, B and C, no correlation could be shown between harvest volume (Figure 4a and b) as well as body weight (data not shown) and intensity or duration of postsurgical discomfort. Furthermore, the comparison of the mean body weight of donors with and without MRI changes in cohorts B and C did not reveal a relevant difference (73 vs 78 kg, P=0.458). In contrast, the mean body weight of donors with an alteration of the SIJ (group A, B and C) was higher than those of donors without such a damage (83 vs 72 kg, P=0.008). In group B, the mean volume of harvested marrow was not different between donors with and without MRI changes (1123 vs 1073 ml, P=0.958), whereas the mean cumulative harvest volume in two-time donors was slightly higher in those with abnormal MRI findings compared to donors without signal alterations (1756 vs 1370 ml, P=0.376).


To our knowledge, this is the first study which analyzed the morphological alterations in the tissues affected by BMH. Multiple marrow aspirations inevitably result in major trauma to the (sacro)iliac bones and the adjacent soft tissues.

There are no investigations regarding the specific pathophysiology of bone injury due to BMH. Nevertheless, by analogy with (noninfectious) osteomyelitis,12 the consequences might include a destruction of bone trabeculae, vasodilatation, hematoma, edema, migration of cellular immune effectors, cytokine release, as well as the activation of osteoclasts and osteoblasts.

MRI is a suitable modality for the detection of acute bone injury due to its capability to demonstrate anatomical details and pathological changes of bone, bone marrow and surrounding soft tissues with high sensitivity and excellent spatial resolution. Gadolinium-enhanced imaging is helpful in the discrimination of fibrovascular scars from inflammatory active foci.13, 14

Edema pattern on T2-weighted sequences and contrast media enhancement after intravenous gadolinium due to hyperemia and increased endothelial permeability were found as a typical MRI morphology in donors 4 weeks after harvest (group A, Figure 1) and corresponds to the suggested inflammatory and regenerative changes (‘acute injury pattern’).

The tissue repair has resulted in a restitutio ad integrum in the majority of donors 1 year after BMH. Nevertheless, about 36% of the study participants in group B had abnormal MRI findings. In two cases, there was a persisting increase of signal intensity in T2-weighted sequences, which was interpreted as an ongoing bone repair. After a surgical intervention due to fracture or osteomyelitis, similar signal intensities and enhancement pattern may also persist for up to 12 months.15, 16 In four other donors (24%), MRI showed a different morphology with signal decrease in T2-weighted and signal increase in T1-weighted images (Figure 2), reflecting a local replacement of hematopoietic tissue by fat marrow. The proportion of donors with fatty marrow conversion and scleroses as residual MRI abnormalities (‘long-term pattern’) increased to over 70% in two-time donors. Although data from MRI changes after the first harvest in these donors do not exist for a direct comparison, the results might indicate a cumulation of tissue damage caused by repetitive harvest procedures.

The appearance of normal human bone marrow in MRI is heterogeneous and the distribution of cellular and fatty components is variable due to age and gender.17, 18, 19 Nevertheless, we think that the detection of fat marrow conversion is a real sequel of harvest but not a physiological phenomenon. We conclude this from the size and morphology of the alterations and their localization in the areas, which are typically affected by the harvesting procedure.

Although the assessment of postsurgical pain and recovery time might be biased in part by a long period between procedure and MRI, the results are comparable to what is known from other reports.3, 4, 5, 6 Therefore, the study includes a representative population of bone marrow donors. The number of study participants in each cohort is too small for a grading of morphological changes and a statistical correlation of the experienced discomfort, but we do not assume a strong association between complaints and tissue injury. This is based on the observation that there were donors in whom MRI demonstrated extensive signal alterations without having a higher-than-average discomfort and vice versa.

The persistence of an abnormal MRI morphology in donors of groups B and C was independent of the donor's body weight and the volume of harvest. However, the mean cumulative volume of harvested marrow in two-time donors was higher in those with MRI changes compared to donors with normal scans, but this difference was not statistically significant. MRI findings indicated an affection of the SIJ in a substantial number of donors, with an increased risk in donors with a higher body weight. Nevertheless, SIJ damage did not lead to an increase in pain intensity or duration (Figure 4c).

In addition to the examination of nature and course of tissue damage after BMH, the study also included all DKMS donors with harvest-related complications and/or prolonged side effects (group D). Although there might be a (small) number of unreported cases, the proportion of seven out of 2447 donations (0.3%) confirms again that BMH is, in general, a very safe procedure. Only in one donor of this cohort, MRI revealed a specific finding which was compatible to the diagnosis of a unilateral sacroiliitis. In the other examinations, MRI detected a bone and soft tissue morphology similar to those in groups A, B and C, or even failed to detect abnormal findings. These results and the fact that age and harvest volumes in group D donors were not different from the other cohorts might reflect an interindividual different tissue vulnerability and pain sensation. That intraoperative events do not alone account for the postsurgical pain is also suggested by a study which reported significantly more pain after BMH in related than in unrelated donors.20

Although there is a steady increase in the use of peripheral blood progenitor cells (PBPC) for allogeneic transplantation, it is improbable that classical BMH will become completely obsolete in the future. Certainly, two recently published randomized studies21, 22 dispelled the initial concern of a higher incidence of graft-versus-host disease after PBPC allografting, but priming with granulocyte colony-stimulating factor (G-CSF) before BMH was described to expand the number of progenitor cells in harvested marrow, resulting in a more rapid engraftment after transplantation.23, 24, 25 Furthermore, in randomized comparisons symptom burden in BM donors was not significantly different from those reported by PBPC donors.5, 7, 26

In conclusion, the results of this study should be interpreted cautiously due to the retrospective design, the limited number of donors and the lack of a control sample. However, the detection of significant bone and soft tissue alterations after BMH, which might reflect the pathophysiological reaction induced by this procedure, and their persistence for a longer time period in some donors represent an important information for all physicians involved in the care of marrow donors. A further prospective evaluation in a larger number of donors is necessary to confirm these results and to identify the factors which influence the extent and duration of tissue damage.


  1. 1

    Bortin MM, Buckner CD . Major complications of marrow harvesting for transplantation. Exp Hematol 1983; 11: 916–921.

    CAS  Google Scholar 

  2. 2

    Buckner CD, Clift RA, Sanders JE et al. Marrow harvesting from normal donors. Blood 1984; 64: 630–634.

    CAS  Google Scholar 

  3. 3

    Stroncek DF, Holland PV, Bartch G et al. Experiences of the first 493 unrelated marrow donors in the National Marrow Donor Program. Blood 1993; 81: 1940–1946.

    CAS  Google Scholar 

  4. 4

    Gandini A, Roata C, Franchini M et al. Italian Bone Marrow Donor Registry (IBMDR) Veneto and Trentino-Alto Adige Regional Registry. Unrelated allogeneic bone marrow donation: short- and long-term follow-up of 103 consecutive volunteer donors. Bone Marrow Transplant 2001; 28: 369–374.

    CAS  Article  Google Scholar 

  5. 5

    Favre G, Beksac M, Bacigalupo A et al. European Group for Blood and Marrow Transplantation (EBMT). Differences between graft product and donor side effects following bone marrow or stem cell donation. Bone Marrow Transplant 2003; 32: 873–880.

    CAS  Article  Google Scholar 

  6. 6

    Ordemann R, Hölig K, Wagner K et al. Acceptance and feasibility of peripheral stem cell mobilisation compared to bone marrow collection from healthy unrelated donors. Bone Marrow Transplant 1998; 21 (Suppl 3): S25–S28.

    Google Scholar 

  7. 7

    Kennedy GA, Morton J, Western R et al. Impact of stem cell donation modality on normal donor quality of life: a prospective randomized study. Bone Marrow Transplant 2003;31: 1033–1035.

    CAS  Article  Google Scholar 

  8. 8

    Drake PA . Hemorrhage after bone marrow harvest: a case presentation. Clin J Oncol Nurs 2000; 4: 29–31.

    CAS  PubMed  Google Scholar 

  9. 9

    Irving I, Cooper M, Durrant S . Sciatic nerve compression following bone marrow harvest (in process citation). Bone Marrow Transplant 2000; 26: 705–706.

    CAS  Article  Google Scholar 

  10. 10

    Riley D, Evans TG . Osteomyelitis complicating bone marrow harvest. Clin Infect Dis 1992; 14: 980–981.

    CAS  Article  Google Scholar 

  11. 11

    Klumpp TR, Mangan KF, MacDonald JS, Mesgarzadeh M . Fracture of the ilium: an unusual complication of bone marrow harvest. Bone Marrow Transplant 1992; 9: 503–504.

    CAS  PubMed  Google Scholar 

  12. 12

    Tsukayama DT . Pathophysiology of posttraumatic osteomyelitis. Clin Orthop 1999; 360: 22–29.

    Article  Google Scholar 

  13. 13

    Gross T, Kaim AH, Regazzoni P, Widmer AF . Current concepts in posttraumatic osteomyelitis with new imaging options. J Trauma 2002; 52: 1210–1219.

    Article  Google Scholar 

  14. 14

    Kaim AH, Gross T, von Schulthess GK . Imaging of chronic posttraumatic osteomyelitis. Eur Radiol 2002; 12: 1193–1202.

    Article  Google Scholar 

  15. 15

    Kaim A, Ledermann HP, Bongartz G et al. Chronic posttraumatic osteomyelitis of the lower extremity: comparison of magnetic resonance imaging and combined bone scintigraphy/immunoscintigraphy with radiolabelled monoclonal antigranulocyte antibodies. Skeletal Radiol 2000; 29: 378–386.

    CAS  Article  Google Scholar 

  16. 16

    Deely D, Schweitzer M . MR imaging of bone marrow disorders. Radiol Clin North Am 1997; 35: 193–212.

    CAS  PubMed  Google Scholar 

  17. 17

    Vande Berg BC, Malghem J, Lecouvet FE, Maldaque B . Magnetic resonance imaging of normal bone marrow. Eur Radiol 1998; 8: 1327–1334.

    CAS  Article  Google Scholar 

  18. 18

    Ricci C, Cova M, Kang YS et al. Normal age-related patterns of cellular and fatty bone marrow distribution in the axial skeleton: MR imaging study. Radiology 1990; 177: 83–88.

    CAS  Article  Google Scholar 

  19. 19

    Duda SH, Laniado M, Schick F et al. Normal bone marrow in the sacrum of young adults: differences between the sexes seen on chemical-shift MR imaging. Am J Roentgenol 1995; 164: 935–940.

    CAS  Article  Google Scholar 

  20. 20

    Chang G, McGarigle C, Spitzer TR et al. A comparison of related and unrelated marrow donors. Psychosom Med 1998; 60: 163–167.

    CAS  Article  Google Scholar 

  21. 21

    Bensinger WI, Martin PJ, Storer B et al. Transplantation of bone marrow as compared with peripheral blood cells from HLA-identical relatives in patients with hematologic cancers. N Engl J Med 2001; 344: 175–181.

    CAS  Article  Google Scholar 

  22. 22

    Couban S, Simpson DR, Barnett MJ et al, for the Canadian Bone Marrow Transplant Group. A randomized multicenter comparison of bone marrow and peripheral blood in recipients of matched sibling allogeneic transplants for myeloid malignancies. Blood 2002; 100: 1525–1531.

    CAS  Article  Google Scholar 

  23. 23

    Damiani D, Fanin R, Silvestri F et al. Randomized trial of autologous filgrastim-primed bone marrow transplantation versus filgrastim-mobilized peripheral blood stem cell transplantation in lymphoma patients. Blood 1997; 90: 36–42.

    CAS  Google Scholar 

  24. 24

    Isola L, Scigliano E, Skerret D et al. A pilot study of allogeneic bone marrow transplantation using related donors stimulated with G-CSF. Bone Marrow Transplant 1997; 20: 1033–1037.

    CAS  Article  Google Scholar 

  25. 25

    Isola L, Scigliano E, Fruchtman S . Long-term follow-up after allogeneic granulocyte colony-stimulating factor-primed bone marrow transplantation. Biol Blood Marrow Transplant 2000; 6: 428–433.

    CAS  Article  Google Scholar 

  26. 26

    Rowley SD, Donaldson G, Lilleby K et al. Experiences of donors enrolled in a randomized study of allogeneic bone marrow or peripheral blood stem cell transplantation. Blood 2001; 97: 2541–2548 (erratum in: Blood 2001; 98:1301).

    CAS  Article  Google Scholar 

Download references


The study was initiated and financially supported by the German Bone Marrow Donor Center (DKMS), Tuebingen, Germany. We thank Catrin Theuser for carefully reading the manuscript.

Author information



Corresponding author

Correspondence to F Kroschinsky.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kroschinsky, F., Kittner, T., Mauersberger, S. et al. Pelvic magnetic resonance imaging after bone marrow harvest – a retrospective study in 50 unrelated marrow donors. Bone Marrow Transplant 35, 667–673 (2005).

Download citation


  • bone marrow harvest
  • unrelated donor
  • magnetic resonance imaging

Further reading


Quick links