Secondary Logo

Journal Logo

Bone Marrow Failure in Children With Acute Liver Failure

Tung, John; Hadzic, Nedim; Layton, Mark*; Baker, Alastair J.; Dhawan, Anil; Rela, Mohamed; Heaton, Nigel D.; Mieli-Vergani, Giorgina

Journal of Pediatric Gastroenterology and Nutrition: November 2000 - Volume 31 - Issue 5 - p 557-561
Original Articles

Background Aplastic anemia is a rare but well-recognized complication of acute hepatitis and acute liver failure. The cause is unknown, and the condition is fatal without bone marrow recovery. Treatment includes immunosuppression regimens or bone marrow transplantation. The purpose of this study was to investigate the incidence, cause, treatment, and outcomes of this disorder in children.

Methods Retrospective chart review of 75 patients with acute liver failure in a major pediatric liver center.

Results Eight patients had evidence of bone marrow failure. Of those, six had aplastic anemia, and two had transient bone marrow suppression. There were five boys, median age 57 months (range, 36–132 months). Two had parvovirus B19, six had non-A, non-B, non-C hepatitis. Five underwent liver transplantation: auxiliary in one, orthotopic in four. The interval between initial symptoms and development of aplastic anemia and/or bone marrow suppression was 21 to 99 days (median, 39 days). Four patients with aplastic anemia received intravenous antithymocyte globulin (ATG) or antilymphocyte globulin (ALG). Median recovery period of granulopoiesis was 62 days (range, 27–115 days). Two made a full recovery, one had myelodysplasia, and one with unresponsive disease died of septic complications. Four did not receive ATG/ALG, two had aplastic anemia, and two had bone marrow suppression. Three underwent liver transplantation, and all four resumed granulopoiesis. One child who underwent liver transplantation died of sepsis with chronic rejection. Median recovery of granulopoiesis was 99 days (range, 20–153 days).

Conclusions Bone marrow failure occurs in 10.7% of children with acute liver failure. It sometimes occurs in association with non-A, non-B, non-C hepatitis and parvovirus B19 infection. Treatment with ATG/ALG is successful and is well tolerated in most cases.

Departments of Child Health, *Haematology, and †Transplant Surgery, King's College Hospital, United Kingdom

Received March 3, 2000;

revised August 1, 2000; accepted September 8, 2000.

Address correspondence and reprint requests to Dr. John Tung, AI Dupont Hospital for Children, PO Box 269, 1600 Rockland Road, Wilmington, DE 19899, U.S.A.

Aplastic anemia (AA), first described in 1888 by Paul Erlich, is defined as pancytopenia with a fatty or empty finding in bone marrow examination. The association of AA with acute hepatitis (AH) was first described in two patients in 1955 (1,2), and by 1974, 195 patients with acute liver failure (ALF) and AH had been reported (3). Effects on bone marrow range from AA to transient bone marrow suppression (BMS).

Patients with AA associated with AH (AHAA) and ALF (ALFAA) have been treated successfully with matched sibling donor bone marrow transplantation (BMT), which is considered the first-line therapy (4). However, only 30% of patients can be treated with this procedure (5), and in the absence of a matched, related bone marrow donor, immunotherapy with antithymocyte globulin (ATG), antilymphocyte globulin (ALG), corticosteroids, cyclosporin A (CyA), and granulocyte colony-stimulating factor (G-CSF) have also been successful (6–10).The causative agents responsible for AHAA and ALFAA are unknown, but the finding of parvovirus B19 in the liver and blood of some of these patients is of particular interest, because this virus causes ALF and BMS (11,12).

The objectives of this single-center study were to investigate the incidence, clinical features, cause, and outcomes of children in whom BMS or AA developed after ALF.

Back to Top | Article Outline


Data were obtained retrospectively from 75 children with ALF referred to King's College Hospital, London (ages, 1 week to 18 years; median, 36 months). Gender distribution was 36 boys and 39 girls. Acute liver failure was defined as the development of severe hepatocellular necrosis and coagulopathy, with or without encephalopathy and without pre-existing liver disease (13).

Back to Top | Article Outline


Other causes of ALF in children were excluded by appropriate investigations. Hepatitis B surface antigen (Murex kit, Abbott Laboratories, Maidenhead, UK), hepatitis B immunoglobulin (Ig)M anticore (IMX System, Abbott Laboratories), hepatitis A IgM, hepatitis C (HCV) antibody (positive results confirmed by HCV reverse transcription–polymerase chain reaction [RT-PCR]), antibodies to human immunodeficiency virus types 1 and 2, measles, cytomegalovirus, Epstein–Barr virus, varicella, adenovirus, and Leptospira antibody tests were performed. Parvovirus B19 serology was tested using an enzyme-linked immunosorbent assay (ELISA; Biotrin, DiaSorin, Ltd., Workingham, UK), and, when possible, results were confirmed with PCR (Central Public Health Laboratory, Colindale, London, UK). Bone marrow aspirates and trephine biopsy specimens were examined in all patients with pancytopenia.

Back to Top | Article Outline


Patients with ALF were classified according to the cellularity of the bone marrow aspirate and peripheral blood parameters into:(1) ALFAA and ALFBMS. Aplastic anemia was defined as hypocellular bone marrow associated with two of three of the following parameters (14) : neutrophils, less than 0.5 × 109/L; platelets, less than 20 × 109/L; and reticulocytes less than 1%.

Patients with neutropenia and thrombocytopenia who did not meet the criteria were classified as having BMS (ALFBMS). Patients with significant features of hemophagocytosis on the bone marrow aspirate were excluded.

Back to Top | Article Outline

Treatment With Antithymocyte and Antilymphocyte Globulins

All patients with ALFAA and their relatives underwent human leukocyte antigen (HLA) typing in case BMT became necessary. Patients who satisfied criteria for ALFAA were first treated with ATG or ALG. After 4 months, those with unresponsive disease were considered for a second course of ATG/ALG. All received G-CSF therapy. Patients who had undergone liver transplantation were treated either with combinations of CyA-azathioprine-prednisolone or tacrolimus-prednisolone which were continued according to liver transplant protocol. One patient received intravenous immunoglobulin for thrombocytopenia.

Doses of drugs used in treatment of ALFAA:

  • Recombinant human G-CSF (Amgen, Cambridge, UK), intravenously 5 to 10 μg/kg per day. The higher dose was administered when neutrophil counts were less than 1 × 109/L.
  • Antithymocyte globulin (rabbit, ATGAM; Upjohn, Kalamazoo, MI, U.S.A.), intravenously 10 to 20 mg/kg per day for 5 days.
  • Antileukocyte globulin (horse; Pasteur Mérieux Connaught, Lyon, France), intravenously 10 to 20 mg/kg per day for 5 days.
  • Patients not receiving CyA or tacrolimus for liver transplantation immunosuppression were treated with CyA beginning 1 week after ATG or ALG to maintain levels at 100 to 150 μg/L (AXSYM-specific [Abbott Laboratories] monoclonal fluorescence polarization immunoassay for CyA).
  • Methylprednisolone for prevention of serum sickness was administered (1 mg/kg per day) for 6 days after ATG/ALG was started, tapered after 7 days, and discontinued after 14 days.

Decision for listing for liver transplantation was based on the criteria of King's College Hospital (15) modified for the pediatric age group, because the presence of encephalopathy in children is often a late event.

Back to Top | Article Outline

Statistical Analysis

Data were analyzed by t-test on computer (Excel; Microsoft, Redmond, WA, U.S.A.).

Back to Top | Article Outline


Seventy-five children were referred with ALF from 1975 through 1998. Nineteen died without liver transplantation, and 27 underwent transplantation, 3 of whom died. Causes of ALF are listed in Table 1.



Eight (10.7%) of 75 children with ALF had ALFAA (n = 6) or ALFBMS (n = 2). There were five boys and three girls (age range, 36–132 months; median age, 57 months). Six had non-A, non-B, non-C hepatitis, and two had parvovirus B19. None had AA or BMS at the onset of liver disease (Table 2).



Four patients with ALFAA were treated with ATG/ALG. Three responded (two completely, one partially) and there was one death from infectious complications of AA unresponsive to ALG. In that patient, liver function had improved and liver transplantation was unnecessary. The patient with partial response (patient 4) had Noonan's syndrome and needed a prolonged course of G-CSF to maintain neutrophil counts. Myelodysplasia developed in that patient 296 days after initial examination which progressed to leukemia, but the disease was treated successfully with chemotherapy and BMT. Only two patients in this group underwent liver transplantation. All four tolerated treatment with ATG/ALG well.

Among the four patients who did not have ATG/ALG, two had ALFAA and two had ALFBMS. All four recovered bone marrow function satisfactorily. Two patients with ALFAA were not treated with ATG/ALG, because in one the protocol did not exist at the time and in the second the bone marrow recovered spontaneously. Three had liver transplantation, but one experienced chronic rejection, necessitating retransplantation. The patient subsequently died of Klebsiella sepsis (Table 3).



The mean onset of neutropenia was 39 days after onset of jaundice (range, 21–99 days), and mean duration was 103 days (range, 20–163 days). The mean onset of thrombocytopenia was 26 days (range, 18 to 70 days), and mean duration was 165 days (range, 1–202 days). In the ATG/ALG–treated group, neutropenia persisted for 62 days (range, 27–115 days) compared with 120 days (range, 20–163 days) in the untreated group. Comparison of onset and end of neutropenia and thrombocytopenia using paired t-tests showed that these occurred at equivalent times (P = 0.31 and P = 0.13, respectively).

The unequivocal presence of parvovirus B19 was determined in one patient by using PCR analysis of a blood sample obtained before any blood products had been transfused. Serologic tests also confirmed an acute parvovirus B19 infection (Table 4).



A typical course of the disease is depicted in Figure 1 in patient 1 who had parvovirus B19 and in whom ALF and severe ALFAA developed. He underwent a successful auxiliary liver transplantation and was treated with ATG, and his bone marrow recovered.

FIG. 1.

FIG. 1.

Back to Top | Article Outline


Aplastic anemia associated with AH is relatively rare, with an incidence of 0.07% to 0.22% in patients with AH (16,17). In contrast, ALFAA was described in 6 (33%) of 18 children with ALF in a study by Cattral et al. (18). In our study ALFAA or ALFBMS developed in 8 (10.7%) of 75 children. Most studies on ALFAA and AHAA show that patients are children or young adults (3,19). It is possible that the causative agent, presumably a community virus, more commonly affects children, because many adults may already be immune.

Many viruses such as hepatitis A, B, C, and G and parvovirus B19 have been associated with AHAA and ALFAA, but in most patients the presumed viral infection remains undiagnosed, and the disease is designated non-A non-B non-C hepatitis (20–25). In this study non-A non-B non-C hepatitis was the major cause of ALFAA and ALFBMS, affecting six of eight patients.

The second most common cause was parvovirus B19, an erythrovirus, causing a common community illness (“slapped-cheek” syndrome) that affected two of eight of our patients. In both patients, severe ALFAA developed. Parvovirus B19 virus binds P-antigen, a sphingolipid, on the cell membranes of erythroid progenitor cells (26) and fetal liver cells (27). Currently, not much is known about the hepatotropism of this virus (28–30). In the current study, we tested serial blood samples from one patient for parvovirus B19 by PCR. The first specimen, obtained before any blood transfusions, was positive but all subsequent tests results were negative. Parvovirus B19 blood carriage can be as high as 1 in 167 during epidemics (31), which makes interpretation of results in patients who have received transfusions very difficult.

It is thought that ALFAA is caused by infection-triggered cytokine production that inhibits bone marrow stem cells (32). The therapeutic value of ALG was discovered after observations that a patient with AHAA recovered during ALG conditioning before BMT (33). Since then, combinations of CyA and ATG/ALG have been shown to be a combination superior to ATG/ALG monotherapy, with survival rates almost approaching that of BMT (6,34). Although BMT is an effective treatment for patients with ALFAA, the procedure may be associated with significant morbidity, mortality, and prolonged immunosuppression.

Our center now uses ATG/ALG, CyA, and G-CSF as a first-line treatment for ALFAA, while considering BMT for those in whom treatment fails. The response to ATG/ALG seen in three of our four patients is very promising. A multicenter study of patients with severe AA (without liver failure) reported a favorable response in 8 of 12 patients who received 1 dose and in 7 of 10 patients who received two doses of ATG (35).

Therapy with ATG/ALG was well tolerated with no patient discontinuing treatment. The only patient in whom the disorder did not respond died of sepsis. In another patient, the disorder was partially responsive, but the patient underwent prolonged G-CSF therapy and subsequently had myelodysplasia and leukemia. It is unclear which of the underlying chromosomal problems in this patient led to the development of myelodysplastic changes (36,37). It is thought that late onset of such clonal proliferative disorders arising after recovery from AA is caused by persistence of defective stem cells. Whether this is an inherent defect occurring in AA or secondary to treatment with ATG/ALG is unknown (38,39). The risks of enhanced immunosuppression and prolonged bone marrow stimulation should be carefully balanced while treating already immunocompromised patients with AA after liver transplantation.

In conclusion, we have shown that ATG/ALG in combination with CyA and G-CSF is well tolerated in ALFAA and suggest that this protocol should be used as a first-line therapy for ALFAA with simultaneous active consideration of BMT for those with unresponsive disease.

Back to Top | Article Outline


The authors thank Dr. Bernard Cohen, Central Public Health Laboratory, Colindale, London, United Kingdom, for parvovirus B19 analysis.

Back to Top | Article Outline


1. Lorenz E. Panmyelopathienach hepatitis epidemica. Wien Klin Wochenschr. 1955; 105:19–22.
2. Fomina LG. K voprosu obizmenenii krovetvoreniia pri zabolevaniiakh pecheni. Sov Med. 1955; 19:28–31.
3. Hagler L, Pastore RA, Bergin JJ, et al. Aplastic anemia following viral hepatitis: report of two fatal cases and literature review. Medicine. (Baltimore) 1975; 54:139–64.
4. Camitta BM, Nathan DG, Forman EN, et al. Posthepatitic severe aplastic anemia—an indication for early bone marrow transplantation. Blood. 1974; 43:473–83.
5. Young N, Barrett AJ. The treatment of severe acquired aplastic anemia. Blood. 1995; 85:3367–77.
6. Pitcher LA, Hann IM, Evans JP, et al. Improved prognosis for acquired aplastic anaemia. Arch Dis Child. 1999; 80:158–62.
7. Bacigalupo A, Chaple M, Hows J, et al. Treatment of aplastic anaemia (AA) with antilymphocyte globulin (ALG) and methylprednisolone (MPred) with or without androgens: a randomized trial from the EBMT SAA working party. Br J Haematol. 1993; 83:145–51.
8. Hadzic N, Layton M, Heaton ND, et al. Anti-thymocyte globulin, cyclosporin A and granulocyte colony-stimulating factor for severe aplastic anaemia complicating paediatric liver transplantation. Eur J Pediatr. 1998; 157:107–8.
9. Rosenfeld SJ, Kimball J, Vining D, et al. Intensive immunosuppression with antithymocyte globulin and cyclosporine as treatment for severe acquired aplastic anemia. Blood. 1995; 85:3058–65.
10. Bacigalupo A, Broccia G, Corda G, et al. Antilymphocyte globulin, cyclosporin, and granulocyte colony-stimulating factor in patients with acquired severe aplastic anemia (SAA): a pilot study of the EBMT SAA Working Party. Blood. 1995; 85:1348–53.
11. Langnas AN, Markin RS, Cattral MS, et al. Parvovirus B19 as a possible causative agent of fulminant liver failure and associated aplastic anemia. Hepatology. 1995; 22:1661–5.
12. Pardi DS, Romero Y, Mertz LE, et al. Hepatitis-associated aplastic anemia and acute parvovirus B19 infection: a report of two cases and a review of the literature. Am J Gastroenterol. 1998; 93:468–70.
13. Bhaduri BR, Vergani GM. Fulminant hepatic failure: pediatric aspects. Semin Liver Dis. 1996; 16:349–55.
14. Camitta BM, Thomas ED, Nathan DG, et al. A prospective study of androgens and bone marrow transplantation for treatment of severe aplastic anemia. Blood. 1979; 53:504–14.
15. O'Grady JG, Alexander GJ, Hayllar KM, et al. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology. 1989; 97:439–45.
16. Pikis A, Kavaliotis J, Manios S. Incidence of aplastic anemia in viral hepatitis in children. Scand J Infect Dis. 1988; 20:109–10.
17. Bottiger LE, Westerholm B. Aplastic anemia: Part I. Incidence and aetiology. Acta Med Scand. 1972; 192:315–8.
18. Cattral MS, Langnas AN, Markin RS, et al. Aplastic anemia after liver transplantation for fulminant liver failure. Hepatology. 1994; 20:813–8.
19. Bottiger LE, Westerholm B. Aplastic anaemia. Part III. Aplastic anaemia and infectious hepatitis. Acta Med Scand. 1972; 192:323–6.
20. de Silva HJ, de Mel WC, Kularatne WN. Aplastic anaemia complicating hepatitis A virus infection. Ceylon Med J. 1993; 38:88–90.
21. Chong RS, Ng HS, Ong YY, et al. A case of aplastic anaemia associated with fulminant hepatitis B. Singapore Med J. 1990; 31:75–7.
22. Hibbs JR, Frickhofen N, Rosenfeld SJ, et al. Aplastic anemia and viral hepatitis: non-A, non-B, non-C? JAMA. 1992; 267:2051–4.
23. Bachwich D, Dienstag J. Aplastic anemia and hepatitis C: molecular biology exonerates another suspect. Hepatology. 1993; 17:340–2.
24. Brown KE, Tisdale J, Barrett AJ, et al. Hepatitis-associated aplastic anemia [see comments]. N Engl J Med. 1997; 336:1059–64.
25. Kiem HP, Storb R, McDonald GB. Hepatitis-associated aplastic anemia. N Engl J Med. 1997; 337:424–5.
26. Brown KE, Hibbs JR, Gallinella G, et al. Resistance to parvovirus B19 infection due to lack of virus receptor (erythrocyte P antigen). N Engl J Med. 1994; 330:1192–6.
27. Yoto Y, Kudoh T, Haseyama K, et al. Human parvovirus B19 infection associated with acute hepatitis. Lancet. 1996; 347:868–9.
28. Tsuda H. Liver dysfunction caused by parvovirus B19 [letter]. Am J Gastroenterol. 1993; 88:1463.
29. Yoto Y, Kudoh T, Asanuma H, et al. Transient disturbance of consciousness and hepatic dysfunction associated with human parvovirus B19 infection. Lancet. 1994; 344:624–5.
30. Naides SJ, Karetnyi YV, Cooling LL, et al. Human parvovirus B19 infection and hepatitis. Lancet. 1996; 347:1563–4.
31. Cohen B. Fortnightly Review: Parvovirus B19: an expanding spectrum of disease. BMJ. 1995; 311:1549–52.
32. Young NS, Maciejewski J. The pathophysiology of acquired aplastic anemia. N Engl J Med. 1997; 336:1365–72.
33. Mathe G, Amiel JL, Schwarzenberg L, et al. Bone marrow graft in man after conditioning by antilymphocytic serum. Transplant Proc. 1971; 3:325–32.
34. Frickhofen N, Kaltwasser JP, Schrezenmeier H, et al. Treatment of aplastic anemia with antilymphocyte globulin and methylprednisolone with or without cyclosporine. The German Aplastic Anemia Study Group [see comments]. N Engl J Med. 1991; 324:1297–304.
35. Matloub YH, Smith C, Bostrom B, et al. One course versus two courses of antithymocyte globulin for the treatment of severe aplastic anemia in children. J Pediatr Hematol Oncol. 1997; 19:110–4.
36. Jacoby RF, Brasitus TA. Myelodysplastic syndrome and pancytopenia in a patient with non-A, non-B chronic active hepatitis. J Clin Gastroenterol. 1991; 13:211–3.
37. Wilcox WD. Occurrence of myeloproliferative disorder in patients with Noonan syndrome. J Pediatr. 1998; 132:189–90.
38. Tooze JA, Marsh JC, Gordon-Smith EC. Clonal evolution of aplastic anaemia to myelodysplasia/acute myeloid leukaemia and paroxysmal nocturnal haemoglobinuria. Leuk Lymphoma. 1999; 33:231–41.
39. Socie G. Could aplastic anaemia be considered a pre-pre-leukaemic disorder? Eur J Haematol Suppl. 1996; 60:60–3.

Acute liver failure; Aplastic anemia; Hepatitis; Parvovirus B19

© 2000 Lippincott Williams & Wilkins, Inc.