Post-Transplant Events

Intestinal thrombotic microangiopathy following reduced-intensity umbilical cord blood transplantation

Article metrics


Thrombotic microangiopathy (TMA) is a significant complication after hematopoietic stem-cell transplantation (HSCT); however, there is little information on it following reduced-intensity cord blood transplantation (RI-CBT). We reviewed the medical records of 123 adult patients who received RI-CBT at Toranomon Hospital between January 2002 and August 2004. TMA was diagnosed in seven patients based on intestinal biopsy (n=6) or autopsy results (n=1). While these patients showed some clinical symptoms such as diarrhea and/or abdominal pain, mental status alterations or neurological disorders were not observed in any of them. Laboratory results were mostly normal at the onset of TMA; >2% fragmented erythrocytes (n=1), <10 mg/dl haptoglobin (n=1), and >200 IU/dl lactic dehydrogenase (LD) (n=4). On endoscopic examination, TMA lesions, consisting of ulcers, erosions, and diffuse exfoliation, were distributed spottily from terminal ileum to rectum. Intestinal graft-versus-host disease (GVHD) and cytomegalovirus (CMV) colitis were confirmed in five and four patients, respectively. With therapeutic measures including supportive care (n=4), fresh frozen plasma (n=1), and a reduction of immunosuppressive agents (n=1), TMA improved in four patients. The present study demonstrates that intestinal TMA is a significant complication after RI-CBT. Since conventional diagnostic criteria can overlook TMA, its diagnosis requires careful examination of the gastrointestinal tract using endoscopy with biopsy.


Thrombotic microangiopathy (TMA) is a significant complication following hematopoietic stem-cell transplantation (HSCT). Endothelial injuries from multiple factors contribute to the formation of widespread platelet thrombi within the microvasculature, causing hemolytic anemia and damage to various organs.1, 2, 3, 4, 5, 6 Owing to the difficulty in making a definitive diagnosis of TMA in HSCT recipients, it is usually diagnosed based on clinical and laboratory findings, such as serum lactic dehydrogenase (LD) levels and the percentage of fragmented erythrocytes.4 However, these findings are frequently nonspecific,7 because they are influenced by many other clinical events. Nishida et al2 reported a 16-patient case series involving TMA with steroid-refractory diarrhea. They showed that TMA frequently involves the gastrointestinal tract in HSCT recipients. Pathological examination of the gastrointestinal tract may be essential in making a definite diagnosis of TMA in HSCT recipients with refractory diarrhea.

Reduced-intensity stem-cell transplantation (RIST) represents an attractive treatment for elderly patients or for those with organ dysfunction who have advanced hematologic malignancies. Shimoni et al1 recently reported that TMA frequently occurs in RIST as well as in myeloablative transplantation. We and other groups have demonstrated the feasibility of RIST using cord blood (reduced-intensity cord blood transplantation, RI-CBT) in adult patients.8, 9 However, it remains unknown whether TMA is a significant complication with RI-CBT. In our institution, HSCT recipients who develop persistent diarrhea regularly undergo endoscopic evaluation of their gastrointestinal tract, and this can sometimes lead to an early diagnosis of TMA of the gut. Thus, we retrospectively investigated the frequency and clinical features of gastrointestinal TMA in patients who had received RI-CBT.

Patients and methods

Patients and transplantation procedures

Between January 2002 and August 2004, 123 patients with hematologic diseases or solid tumors underwent RI-CBT at Toranomon Hospital, Japan. All patients had diseases that were incurable using conventional treatments and were considered to be inappropriate candidates for conventional HSCT due to the lack of an HLA-identical sibling or a suitable unrelated donor, being >50-years-old, and/or having organ dysfunction (generally attributable to previous intense chemo- and/or radiotherapy). All patients provided written informed consent in accordance with the requirements of the Institutional Review Board.

A search for unrelated donors was made using the Japan Marrow Donation Program for patients without HLA-identical sibling donors.10 If no appropriate donor was identified, the Japan Cord Blood Bank Network11 was searched. Cord blood units were not depleted of T lymphocytes. Preparative regimens and prophylaxis of graft-versus-host disease (GVHD) are shown in Table 1. Supportive care and management of GVHD have been previously reported.9

Table 1 RI-CBT patient characteristics

Colonoscopic examination

We conducted a total colonoscopy from the terminal ileum to the rectum, including mucosal biopsies when patients developed diarrhea lasting 3 days or longer. Intermediate length scopes (CF230, CF240I, CFQ240I, PCF 230, PCF240; Olympus, Tokyo, Japan and EC410WN; Fujinon Toshiba ES Systems, Tokyo, Japan) were used for this examination.

Diagnostic criteria of TMA, GVHD, and cytomegalovirus enterocolitis

The diagnosis of TMA was based solely on the results of the pathological examination. Intestinal specimens obtained by biopsies or autopsies were retrospectively reviewed. All pathological evaluations were performed in a blinded fashion. Histopathological evaluation was performed on formalin-fixed paraffin-embedded sections. The pathological diagnosis of TMA was based on the following microscopic findings: (a) widening of the subendothelial space with myxoid degeneration (subintimal mucoid lesion); (b) intraluminal thrombi; and (c) fibrinoid necrosis of the terminal arterioles. A finding of apoptotic bodies in the bases of the crypts was the sole diagnostic criterion of GVHD. Cytomegalovirus (CMV) infection was evaluated using a monoclonal antibody for CMV (DAKO, Tokyo, Japan).

A single observer, blinded to the patient's clinical course, counted 3000 red blood cells per smear and calculated the percentage of fragmented erythrocytes.


Patient characteristics and transplantation outcomes

Patient characteristics are shown in Table 1. A total of 98 patients (80%) achieved primary neutrophil engraftment within 100 days of RI-CBT. The median day of engraftment was 20 (range, day 11–53). In all, 52 patients (42%) died due to transplant-related causes within 100 days of transplant. The primary disease progressed in 16 patients, and was fatal in five of them within 100 days. Of the patients, 28 developed grade II–IV acute GVHD with a median onset at day 22 (range, day 11–45). As of December 2004, the median follow-up of the surviving patients was 8.0 months (range, 3.9–31 months). Overall survival rates were 86% at day 30 and 53% at day 100.

Clinical features of intestinal TMA

Of the patients, 44 underwent 76 colonoscopies (1–7 times/patient). Autopsies were performed in 13 patients. Intestinal TMA was diagnosed in seven patients (5.4%) based on intestinal biopsies (n=6) or autopsy (n=1). The patient in whom TMA was diagnosed at autopsy had diarrhea with abdominal pain.

Clinical features of intestinal TMA are shown in Table 2. The median onset of TMA was day 51 (range, day 15–167). Laboratory results were normal in most of the patients at the time of onset of TMA (>2% fragmented erythrocytes, one patient; <10 mg/dl haptoglobin, one patient; >200 IU/dl LD, four patients). No patients had hypercoagulability. Five and four patients developed acute GVHD, and CMV colitis, respectively. The management of TMA comprised supportive care (n=4), fresh-frozen plasma (n=1) and reduction of the tacrolimus dose (n=1). Five patients with intestinal TMA died, but TMA was not the primary cause of death in any of these patients. TMA responded to the therapeutic measures in four patients. The other patients who had persistent TMA died of relapse or invasive aspergillosis.

Table 2 Clinical and laboratory findings of patients with intestinal TMA

Colonoscopic and histopathological findings of intestinal TMA

Colonoscopic and histopathological findings are shown in Figures 1 and 2, respectively. These findings are summarized in Table 3. TMA lesions, consisting of ulcers, erosions, and diffuse exfoliation, were distributed spottily from the terminal ileum to the rectum. Histopathological evidence of intestinal GVHD was confirmed in five patients, and CMV colitis was confirmed in four patients. In the patient (Patient 6) in whom TMA was diagnosed at autopsy, intraluminal thrombi were not observed in any specimens that had been obtained except for those obtained from the gut.

Figure 1

Colonoscopic intestinal findings. Spotty ulcer in rectum (Patient 2).

Figure 2

Pathological intestinal findings. (a) Intraluminal thrombi with extensive interstitial hemorrhage in Patient 6. (HE stain, × 200) (b) Subintimal mucoid lesion of the artery and red cell fragmentation (arrows). Immunohistological study with CD34 shows splitting of endothelial cells from the basement membrane (inset) in Patient 7. (c) Extensive necrosis of the rectal mucosa with thrombi (arrows) in the capillaries in Patient 4 (HE stain, × 150).

Table 3 Summary of macroscopic and pathological findings

Risk factors for intestinal TMA

Univariate analysis of the incidence of TMA is shown in Table 4. Multivariate analysis failed to identify any significant independent factors.

Table 4 Univariate analysis of the incidence of TMA


This study demonstrates that intestinal TMA is a significant complication in RI-CBT as well as in myeloablative HSCT.12 It occurred in 5.4% of RI-CBT recipients. This frequency is lower than that noted in previous studies on post transplant TMA, in which it was diagnosed in 5–23% of HSCT recipients.1, 12 While the diagnostic criteria used in our study and in the previous reports were different, we used strict diagnostic criteria based on pathological diagnosis, demonstrating that intestinal TMA is common after RI-CBT. Systemic TMA was not considered in our study, and further research on this aspect of TMA is needed.

The most common criteria for a diagnosis of TMA following HSCT are the signs of microangiopathic hemolysis. Surprisingly, in the six evaluable patients, red cell fragmentation and serum LD elevation were mild or absent, and serum haptoglobin levels were detectable. Post-mortem studies failed to find any evidence of TMA other than in the intestine. Neither renal dysfunction nor neurologic abnormalities were present in any patients. Based on the conventional pentad of HUS/TTP, TMA was not diagnosed in any of them in our study. These findings suggest a difference in pathogenesis between TMA following RI-CBT and either classic TTP13 or TMA following conventional HSCT.2, 12

There is little information on the pathogenesis of intestinal TMA following RI-CBT. TMA after myeloablative HSCT has a multifactorial etiology that includes immunosuppressive agents,14, 15 total body irradiation (TBI),16 CMV infection,17 and acute GVHD.18 These factors injure the vascular endothelium of many organs.3 In contrast, our results suggest that particular factors specifically affecting the gastrointestinal system are largely involved in the etiology of TMA after RI-CBT. It should be noted that most patients with TMA after RI-CBT had overlapping gastrointestinal GVHD and/or CMV colitis. An animal study has demonstrated that the vascular endothelium is a target of alloimmunity.18 The present study supports this hypothesis. GVHD was associated with gastrointestinal TMA, and the association could partly explain why TMA was located in the gut. It is reasonable to assume that GVHD damages the gastrointestinal endothelium, leading to the development of intestinal TMA. Regimen-related toxicity (RRT) of the gut is known to increase the risk of intestinal GVHD.19 Gastrointestinal damage due to preparative regimens might contribute to the development of intestinal TMA in our study. CMV infection, which is another putative etiology of TMA,17 was documented in four patients, and all were located in the gut. CMV colitis might be associated with intestinal TMA following RI-CBT.

Intestinal TMA has been described in a few reports,2 and has drawn little attention.1, 12 The differences in the observations between our study and these previous reports can be explained by several reasons. First, intestinal TMA might be specific to RI-CBT; however, this is unlikely. After cord blood transplantation (CBT), intestinal damage by GVHD is rather mild due to the low incidence of GVHD. The conditioning agents, fludarabine, melphalan, and TBI, are also unlikely to specifically damage the intestinal vascular endothelium. The RI-CBT transplantation procedure is unlikely to cause intestinal TMA. Second, the patients included in our study were older. Advanced age tended to be a risk factor for TMA in our study. As RIST is used more commonly in older patients, the number of intestinal TMA cases may grow. Third, clinicians and pathologists were not commonly aware of TMA and could possibly have misinterpreted it as GVHD or infectious colitis. Finally, a pathological diagnosis of TMA can be difficult to make. Thrombolysis, which might occur after death, might have masked the pathological findings of TMA at autopsy.20 Further investigation into the frequency and clinical characteristics of intestinal TMA after conventional SCT is necessary to allow a proper interpretation of the various published reports.

Total colonoscopy from the rectum to the terminal ileum with biopsy is required to make the diagnosis of intestinal TMA. Since our patients had focal TMA lesions of various distributions, biopsy of the rectum alone might have missed the diagnosis of TMA. Colonoscopic findings of TMA were diverse in our patients. It was difficult to differentiate TMA from intestinal GVHD21, 22 and CMV colitis.23 Furthermore, TMA was complicated by GVHD and CMV colitis in many patients. Macroscopic observation alone is not sufficient to make a diagnosis of TMA. Laboratory findings alone are also not useful. Clinically available risk factors were not identified in our study; laboratory data such as LD at the time of colonoscopy were not significantly different between patients with, and without TMA. Thus, biopsies and pathological examination extending from the rectum to the terminal ileum are probably necessary to make a definite diagnosis in patients with diarrhea.

While the appropriate treatment of intestinal TMA is unknown, a published series of cases suggests that reducing the dose of immunosuppressants may be effective for intestinal TMA as well as classic TMA.2 In our study, however, three of five patients with intestinal GVHD and TMA improved without immunosuppressant reduction (Table 2). This observation would indicate that the management of GVHD, rather than immunosuppressant reduction, is important in the treatment of intestinal TMA. In fact, the reduction of immunosuppressants to prevent vascular endothelial damage would aggravate GVHD, and increase the risk of TMA progression. Considering these possibilities, one should be vigilant when deciding on the dose of immunosuppressant for TMA after RI-CBT. The treatments used for classic TTP, such as fresh frozen plasma and plasma exchange, have been tried for TMA after bone marrow transplantation.12 However, the efficacy of these treatments in patients with intestinal TMA remains unclear. Minimizing damage to the intestinal mucosa and vascular endothelium would be more desirable for the management of intestinal TMA than would the treatments designed for classic TTP.

This study provides us with important information about TMA following RI-CBT; however, there are some limitations. First, this was a small, retrospective study. Unrecognized bias might have influenced the results. Second, some RI-CBT recipients did not have a total colonoscopy due to their poor physical condition. We might have underestimated the incidence of TMA. Third, there is the possibility that overwhelming infection of the intestine might have masked the clinicopathological features of TMA. Last, a post-mortem examination was conducted in only 13 of 76 patients. We cannot define the specificity of the pathological finding of intestinal TMA. Further studies are warranted to investigate a cause-effect relationship between TMA and bowel dysfunction, and to investigate whether TMA is present in organs other than the intestine.

In conclusion, the present study demonstrated that intestinal TMA is a significant complication after RI-CBT. When transplant recipients develop refractory diarrhea, TMA needs to be included in the differential diagnoses. However, conventional diagnostic criteria can fail to identify TMA. Thus, the diagnosis of TMA requires endoscopy with biopsy.


  1. 1

    Shimoni A, Yeshurun M, Hardan I et al. Thrombotic microangiopathy after allogeneic stem cell transplantation in the era of reduced-intensity conditioning: the incidence is not reduced. Biol Blood Marrow Transplant 2004; 10: 484–493.

  2. 2

    Nishida T, Hamaguchi M, Hirabayashi N et al. Intestinal thrombotic microangiopathy after allogeneic bone marrow transplantation: a clinical imitator of acute enteric graft-versus-host disease. Bone Marrow Transplant 2004; 33: 1143–1150.

  3. 3

    Pettitt AR, Clark RE . Thrombotic microangiopathy following bone marrow transplantation. Bone Marrow Transplant 1994; 14: 495–504.

  4. 4

    Zeigler ZR, Shadduck RK, Nemunaitis J et al. Bone marrow transplant-associated thrombotic microangiopathy: a case series. Bone Marrow Transplant 1995; 15: 247–253.

  5. 5

    Zeigler ZR, Shadduck RK, Nath R et al. Pilot study of combined cryosupernatant and protein A immunoadsorption exchange in the treatment of grade 3–4 bone marrow transplant-associated thrombotic microangiopathy. Bone Marrow Transplant 1996; 17: 81–86.

  6. 6

    Daly AS, Xenocostas A, Lipton JH . Transplantation-associated thrombotic microangiopathy: twenty-two years later. Bone Marrow Transplant 2002; 30: 709–715.

  7. 7

    Zomas A, Saso R, Powles R et al. Red cell fragmentation (schistocytosis) after bone marrow transplantation. Bone Marrow Transplant 1998; 22: 777–780.

  8. 8

    Barker JN, Weisdorf DJ, DeFor TE et al. Rapid and complete donor chimerism in adult recipients of unrelated donor umbilical cord blood transplantation after reduced-intensity conditioning. Blood 2003; 102: 1915–1919.

  9. 9

    Miyakoshi S, Yuji K, Kami M et al. Successful engraftment after reduced-intensity umbilical cord blood transplantation for adult patients with advanced hematological diseases. Clin Cancer Res 2004; 10: 3586–3592.

  10. 10

    Kodera Y, Morishima Y, Kato S et al. Analysis of 500 bone marrow transplants from unrelated donors (UR-BMT) facilitated by the Japan Marrow Donor Program: confirmation of UR-BMT as a standard therapy for patients with leukemia and aplastic anemia. Bone Marrow Transplant 1999; 24: 995–1003.

  11. 11

    Nishihira H, Kato K, Isoyama K et al. The Japanese cord blood bank network experience with cord blood transplantation from unrelated donors for haematological malignancies: an evaluation of graft-versus-host disease prophylaxis. Br J Haematol 2003; 120: 516–522.

  12. 12

    Allford SL, Bird JM, Marks DI . Thrombotic thrombocytopenic purpura following stem cell transplantation. Leuk Lymphoma 2002; 43: 1921–1926.

  13. 13

    Furlan M, Robles R, Galbusera M et al. von Willebrand factor-cleaving protease in thrombotic thrombocytopenic purpura and the hemolytic-uremic syndrome. N Engl J Med 1998; 339: 1578–1584.

  14. 14

    Trimarchi HM, Truong LD, Brennan S et al. FK506-associated thrombotic microangiopathy: report of two cases and review of the literature. Transplantation 1999; 67: 539–544.

  15. 15

    Pham PT, Peng A, Wilkinson AH et al. Cyclosporine and tacrolimus-associated thrombotic microangiopathy. Am J Kidney Dis 2000; 36: 844–850.

  16. 16

    Ballermann BJ . Endothelial cell activation. Kidney Int 1998; 53: 1810–1826.

  17. 17

    Takatsuka H, Wakae T, Mori A et al. Endothelial damage caused by cytomegalovirus and human herpesvirus-6. Bone Marrow Transplant 2003; 31: 475–479.

  18. 18

    Ertault-Daneshpouy M, Leboeuf C, Lemann M et al. Pericapillary hemorrhage as criterion of severe human digestive graft-versus-host disease. Blood 2004; 103: 4681–4684.

  19. 19

    Goldberg J, Jacobsohn DA, Zahurak ML et al. Gastrointestinal toxicity from the preparative regimen is associated with an increased risk of graft-versus-host disease. Biol Blood Marrow Transplant 2005; 11: 101–107.

  20. 20

    Iwata H, Kami M, Hori A et al. An autopsy-based retrospective study of secondary thrombotic thrombocytopenic purpura. Haematologica 2001; 86: 669–670.

  21. 21

    Martin PJ, McDonald GB, Sanders JE et al. Increasingly frequent diagnosis of acute gastrointestinal graft-versus-host disease after allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant 2004; 10: 320–327.

  22. 22

    Iqbal N, Salzman D, Lazenby AJ, Wilcox CM . Diagnosis of gastrointestinal graft-versus-host disease. Am J Gastroenterol 2000; 95: 3034–3038.

  23. 23

    Meyers JD, Flournoy N, Thomas ED . Risk factors for cytomegalovirus infection after human marrow transplantation. J Infect Dis 1986; 153: 478–488.

Download references


We are grateful to Dr Yoshihisa Morishita (JA Aichi Showa Hospital, Aichi, Japan) for his critical reading of the manuscript.

Author information

Correspondence to M Kami.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Narimatsu, H., Kami, M., Hara, S. et al. Intestinal thrombotic microangiopathy following reduced-intensity umbilical cord blood transplantation. Bone Marrow Transplant 36, 517–523 (2005) doi:10.1038/sj.bmt.1705099

Download citation


  • diarrhea
  • colonoscopy
  • allogeneic hematopoietic stem-cell transplantation
  • vascular endothelial cell

Further reading