Volume 14, Issue 11 pp. 2645-2650
Case Report
Free Access

Parvovirus Associated Fulminant Hepatic Failure and Aplastic Anemia Treated Successfully With Liver and Bone Marrow Transplantation. A Report of Two Cases

L. Bathla

Corresponding Author

L. Bathla

Methodist Specialty and Transplant Hospital, San Antonio, TX

Corresponding author: Lokesh Bathla, [email protected]Search for more papers by this author
W. J. Grant

W. J. Grant

Section of Transplantation, Department of Surgery, University of Nebraska Medical Center, Omaha, NE

Search for more papers by this author
D. F. Mercer

D. F. Mercer

Section of Transplantation, Department of Surgery, University of Nebraska Medical Center, Omaha, NE

Search for more papers by this author
L. M. Vargas

L. M. Vargas

Section of Transplantation, Department of Surgery, University of Nebraska Medical Center, Omaha, NE

Search for more papers by this author
C. L. Gebhart

C. L. Gebhart

Department of Pathology & Microbiology, University of Nebraska Medical Center, Omaha, NE

Search for more papers by this author
A. N. Langnas

A. N. Langnas

Section of Transplantation, Department of Surgery, University of Nebraska Medical Center, Omaha, NE

Search for more papers by this author
First published: 01 September 2014
Citations: 13

Abstract

Aplastic anemia (AA) has been observed in nearly a third of patients undergoing liver transplantation (LT) for non-A-E fulminant hepatic failure (FHF). Few of these patients have been successfully managed with sequential LT and bone marrow transplantation (BMT). No causative agent has been identified for the FHF or AA in these reported cases. At our center, two patients, aged 15 years and 7 years, respectively, underwent sequential living-related LT and living-unrelated BMT. These patients are 10/9 years and 5/4 years post-LT/BMT. Human parvovirus B19 (HPV-B19) was established as the causative agent for FHF in both these patients by polymerase chain reaction. This report presents the first two cases associating HPV-B19 with FHF and AA who underwent sequential LT and BMT with excellent outcomes.

Abbreviations

  • AA
  • aplastic anemia
  • ALG
  • anti-lymphocyte globulin
  • ALT
  • alanine aminotransferase
  • AST
  • aspartate aminotransferase
  • ATG
  • anti-thymocyte globulin
  • BMF
  • bone marrow failure
  • BMT
  • bone marrow transplantation
  • CMV
  • cytomegalovirus
  • FHF
  • fulminant hepatic failure
  • G-CSF
  • granulocyte-colony stimulating factor
  • GVHD
  • graft-versus-host disease
  • Hb
  • hemoglobin
  • HBV
  • hepatitis B virus
  • HCV
  • hepatitis C virus
  • HPV-B19
  • human parvovirus B19
  • HSV
  • herpes simplex virus
  • INR
  • international normalized ratio
  • IS
  • immunosuppression
  • IVIG
  • intravenous immunoglobulin
  • LT
  • liver transplantation
  • NMDP
  • National Marrow Donor Program
  • PCR
  • polymerase chain reaction
  • POD
  • postoperative day
  • SOS
  • sinusoidal obstruction syndrome
  • WBC
  • white blood cell
  • Introduction

    Aplastic anemia (AA) is a life-threatening complication seen occasionally in patients with fulminant hepatic failure (FHF). Although the incidence of AA in patients with acute viral hepatitis is estimated at 0.07%, AA has been observed in nearly one-third of the patients undergoing liver transplantation (LT) for FHF secondary to non-A-E hepatitis 1-4. Most of this data have been derived from the pediatric population. In most of these cases, the etiology is unknown. Over the years, AA in these patients has been managed with several treatment regimens with varying degrees of success. Early protocols included immunoablation with anti-thymocyte or anti-lymphocyte globulin (ATG/ALG) in combination with steroids and cyclosporine A along with intermittent use of granulocyte-colony stimulating factor (G-CSF). More recently, bone marrow transplantation (BMT) has been utilized with excellent outcomes in these patients. An extensive literature search identified only 10 cases that underwent BMT after LT for FHF and AA 5-13. No causative agent for the FHF or AA has been identified in any of these cases.

    We report two cases of FHF caused by human parvovirus B19 (HPV-B19), complicated with AA who underwent sequential LT and BMT. This is the first report associating HPV-B19 with FHF and AA and utilizing unrelated HLA matched donors for BMT with excellent outcomes.

    Case 1

    A 15-year-old Native-American male presented with 3-week history of jaundice, dark urine and pruritus. He denied any other symptoms, sick contacts, foreign travel or exposure to any medication or chemicals. Work-up for Wilson's disease, autoimmune hepatitis and alpha-1 anti-trypsin deficiency was negative. Serologies for hepatitis A virus, hepatitis B virus (HBV), hepatitis C virus (HCV), human immunodeficiency virus, cytomegalovirus (CMV), human herpes virus 6, herpes simplex virus (HSV) 1 and 2, HPV-B19 DNA and Ebstein–Barr virus were also negative. Serum biochemistry at presentation showed a total bilirubin of 15.2 mg/dL, alkaline phosphatase 805 U/L, aspartate aminotransferase (AST) 2663 IU/L, alanine aminotransferase (ALT) 2669 IU/L and an international normalized ratio (INR) of 2.0. Liver biopsy showed features of acute hepatitis with sub-massive necrosis and architectural collapse. He initially improved with conservative management and was discharged to home. Three weeks later, he presented with symptomatic hypoglycemia. Exhaustive serological work-up was again negative. His bilirubin was 19.8, INR 2.5, white blood cell (WBC) 1100/cu mm, hemoglobin (Hb) 7.7 g/dL and platelets 37 000/cu mm. He was listed for LT, and underwent a living-related right lobe transplant from his sister 12 days after admission. Immuno-peroxidase staining for CMV and HSV were negative, but polymerase chain reaction (PCR), performed on native liver tissue, was positive for HPV-B19 DNA.

    His maintenance immunosuppression consisted of tacrolimus and prednisone. His postoperative course was remarkable for persistent pancytopenia and a bone marrow biopsy done on postoperative day (POD) 22 revealed markedly hypo-cellular bone marrow (<5%) with no cytogenetic abnormalities. He was started on G-CSF and erythropoietin therapy but continued to require packed red blood cell and platelet transfusions twice weekly with minimal response. Repeat bone marrow biopsy performed 4 months after liver transplant showed little evidence of regeneration with poor cellularity. He was listed for BMT and did not have any suitable matched siblings. A search through the National Marrow Donor Program (NMDP) identified a potential donor with 8 out of 10 HLA match with A and C locus mismatch, and the patient underwent a BMT 10 months after the liver transplant. He underwent preconditioning with cyclophosphamide 50 mg/kg/day for 4 days, 600-cGray total body radiation in 3 fractions over 36 h. Uncorrected cell count for BMT was 4.65 × 108/kg. The bone marrow transplant was successful with complete engraftment by POD 10. Nearly a month after the BMT, he presented with a generalized skin rash and diarrhea, suspicious for graft-versus-host disease (GVHD). Skin and colon biopsied confirmed grade-1 GVHD. He was placed under more frequent surveillance without any changes in the immunosuppression (IS) regimen. Over the course of next few weeks, the patient's symptoms resolved and repeat biopsies did not show any evidence of GVHD. One-year post-BMT, the patient was admitted to the hospital with weight loss, generalized weakness, recent history of pneumonia and ulcers on the lower extremities. Skin biopsies were consistent with chronic GVHD. No evidence of GVHD was found on bronchoscopy or in intestinal biopsies. He was treated with intravenous steroids and increase in his maintenance IS with complete recovery. His last follow-up was 115/106 months since the transplants, and he was doing well.

    Case 2

    A 7-year-old, otherwise healthy boy presented with FHF. Approximately 2 weeks prior to presentation, he developed flu-like symptoms, nausea, loss of appetite, right upper quadrant abdominal pain, clay colored stools and scleral icterus. On presentation his bilirubin was 11.0 mg/dL, albumin 3.2 g/dL, AST 2900 IU/L, ALT 3000 IU/L and alkaline phosphatase 520 U/L. His complete blood count at presentation showed WBC 11 900/cu mm, Hb 11.9 g/dL and platelets 169 000/cu mm. His serologic work-up (as above) was negative. His condition worsened with development of coagulopathy, cerebral edema, altered mental status and hypoglycemia requiring intubation and placement of an intracranial pressure monitoring device. He underwent a living-related LT using the left lateral segment from his mother 3 days after admission. The pathology of his explanted liver showed massive necrosis, and again, tissue PCR was positive for HPV-B19 DNA (see Figure 1).

    Details are in the caption following the image
    Polymerase chain reaction (PCR) for human parvovirus B19 (HPV-B19). Lanes 1 and 8 present a 100-base pair (bp) ladder. Lane 7 is the negative control. Lane 4 is the positive control. Lanes 2–3 are patient #1 and lanes 5–6 are patient #2. Note that lanes 2–6 have a 104 bp product, indicating that each sample is positive for HPV-B19. Ten-micrometer-thick sections were cut, and DNA was extracted using the QIAamp DNA FFPE Tissue kit from Qiagen (Valencia, CA). One-microgram of DNA was added to the PCR. Primer sequences were 5′-CAA AAG CAT GTG GAG TGA GG-3′ and 5′-CCT TAT AAT GGT GCT CTG GG-3′. These primers amplify 104-bp region coding for the HPV-B19 nucleocapsid. PCR consisted of 45 cycles at 95°C for 30 s, 60°C for 30 s and 72°C for 30 s. The reaction product was run on a gel alongside appropriate positive and negative control reactions. DNA integrity and the absence of PCR inhibitors were assayed by performing a PCR-specific for the beta-heme gene.

    His postoperative course was complicated by a bile leak, which was managed successfully with percutaneous drainage and intravenous antibiotics. On POD 24, he was readmitted with neutropenia, thrombocytopenia and bacteremia. Broad-spectrum antibiotics were started and blood components were transfused as necessary. He underwent a bone marrow biopsy on POD 25, which revealed hypoplastic bone marrow consistent with AA.

    He received a course of intravenous immunoglobulin (IVIG) followed by rapid course of ATG (12.5 mg/kg, given over 5 days) and adjustment of his tacrolimus targets to a higher level (5–10 ng/mL). The patient received IVIG as a prophylaxis against infections for his pancytopenia. After an initial response to the G-CSF, his absolute neutrophil count dropped and stayed below 500/dL. He had minimal response to erythropoietin therapy with persistently low reticulocyte count and frequent transfusion requirements. Repeat bone marrow biopsy showed 15% cellularity in all cell lineages initially, but this response was not sustained. Bone marrow biopsy performed nearly 6 months after the liver transplant showed markedly decreased hematopoiesis. An unrelated 10-allele HLA matched CMV-negative donor was identified through the NMDP. He underwent preconditioning with cyclophosphamide 50 mg/kg/day for 4 days, 600-cGray total body radiation in 3 fractions over 36 h and 30 mg/kg ATG. Uncorrected cell count for BMT was 24.0 × 108/kg. The bone marrow transplant was successful with complete engraftment on POD 12.

    At 2 months after the BMT, the patient had a liver biopsy that showed marked hepatocellular iron deposition, with a serum ferritin level of 4394 ng/mL (normal 23–340 ng/mL). He was started on weekly phlebotomies after a limited response to iron chelation therapy. Nearly 14 months post-BMT, his serum ferritin normalized (<300 ng/mL) and his therapeutic phlebotomies were discontinued. At present, he is 59/48 months post-LT/BMT and doing well.

    Discussion

    HPV-B19 produces a spectrum of clinical manifestations in humans, including: hydrops fetalis and fetal demise in utero; erythema infectiosum (fifth disease) in children; an arthritis-like syndrome in adults; hematological disorders such as leukopenia, thrombocytopenia, transient aplastic crisis in patients with chronic hemolytic anemia and chronic anemia in immune-compromised patients. In most cases, HPV-B19 infection is benign and self-limited, requiring only symptomatic treatment. The hepatic manifestations of HPV-B19 infection range most commonly from transient transaminitis to rarely FHF.

    The association of FHF and AA in patients undergoing LT for FHF has been well described in the literature. In one of the earliest reported series, AA developed in nearly 30% of patients who underwent LT for FHF of “unknown” etiology 4. Hadzic et al 13 reported their experience of over 17 years in which they observed nearly 10% incidence of BMF after FHF in a cohort of 213 patients at a supra regional referral center. Only five of these patients received BMT for the management of AA. The overall mortality in their cohort was 20% with no mortality in patients undergoing BMT. Several other studies have suggested HPV B-19 as a possible cause of FHF and AA on the basis of PCR results 14-17. These findings have been challenged in a study by Wong et al 18 wherein, they did not observe any significant difference in the prevalence of HPV B19 DNA in the livers form patients with FHF or hepatitis associated AA, compared with liver tissue samples from patients with HBV or HCV infection. HPV-B19 DNA was detected in 4 of the 15 livers from patients with FHF and in 3 out of 22 livers from patients with nonviral diseases 18. They also observed positive HPV-B19 PCR in nearly one-third of patients with HBV or HCV infection. The presence of HPV-B19 DNA in the liver tissue of patients with nonviral and established hepatitis B or C infection, although interesting, fails to undermine causative association between HPV and FHF. Since no population-based data about the presence of HPV-B19 DNA in healthy subjects is available, studies with larger sample size are warranted before refuting association between HPV-B19 and FHF. Additionally, several reports in the literature have described AA secondary to HPV-B19 infection, including successful treatment with myeloablation and BMT 19-21. It would seem that in the appropriate clinical scenario of FHF with positive PCR for HPV-B19, especially without any other identifiable cause, causative association between HPV-B19 and FHF cannot be denied.

    It has been hypothesized that virally triggered immune-pathogenic mechanisms rather than direct stem cell injury are responsible for the development of AA in these patients with FHF 22. This is based on the findings that FHF associated AA can resolve with increased IS in nearly half the patients 23, 24.

    It has also been proposed that since the operative and postoperative management of these patients was no different from patients undergoing LT for other reasons, the AA was related to the antecedent hepatitis and not a complication of LT 2. Cattral et al 3 also made similar observations of AA in nearly one-third of patients who underwent LT for FHF. In their report, AA developed prior to transplantation in most of the patients, thereby strongly supporting the hypothesis that AA was a complication of the underlying disease and not otherwise. No causative agent was identified for FHF and AA in both these studies. In a retrospective analysis of patients with FHF and AA who had undergone LT, Langnas et al identified HPV-B19 DNA using PCR in the liver tissue of four out of six patients. This was the first study to provide evidence supporting the role of HPV-B19 in the development of FHF and associated AA 15. It is difficult to predict whether the identification of HPV-B19 as the underlying cause of FHF and AA would affect subsequent decision-making process regarding LT and BMT. This report besides adding to the literature of LT and BMT for FHF and AA, takes our understanding further by identifying an underlying cause for the disease process. It remains to be seen however, that if diagnosed early in the course, can the AA be managed differently and avoid BMT.

    Until 1991, a variety of treatment regimens ranging from supportive care only to IS with ALG/ATG or OKT3 were used for the management of AA patients with FHF and AA who had undergone LT 2, 25. The outcomes were dismal with improvement of AA seen in less than half the patients. In 1991, Kawahara et al reported the first case where allogeneic BMT was performed in a 6.5-year-old such patient 5. Since then, only nine cases have been reported in the literature (see Table 1). No cause for FHF or AA was identified in any of these patients. HPV-B19 serology was reported in three cases and was negative. HPV-B19 DNA PCR was performed in only one case and was negative 13. Both of our patients had negative serology with positive liver tissue PCR for HPV-B19.

    Table 1. Summary of all cases of fulminant hepatic failure and aplastic anemia treated with sequential liver transplant and bone marrow transplant
    Author (reference no.) Year Cause HPV-B19 antibody Pathology Parvo PCR Follow-up duration Outcome Age (years)
    Kawahara et al 5 1991 Unknown NR NR NR 2 years Well 6.5
    Dugan et al 6 1993 Unknown NR NR NR 2 years Well 29
    Trede et al 7 1997 Unknown Negative NR NR 1 year Well 8
    Hagglund et al 8 1997 Unknown NR Hepatic nec NR 3 years Well 2.5
    Perkins et al 9 2001 Unknown NR Massive nec NR 8 years Well 3.5
    Stachel et al 10 2002 Unknown Negative Fulminant NR 8 years Well 2.0
    Umeda et al 11 2002 Unknown NR NR NR 9 months Well 1
    Molina et al 12 2004 Unknown NR NR NR NR Well 7.9
    Molina et al 12 2004 Unknown NR NR NR NR Well 9
    Hadzic et al 13 2008 Unknown Negative NR Negative 10 months Well 5
    Present study 2013 HPV-B19 Negative Massive nec Positive 9.5 years Well 15
    Present study 2013 HPV-B19 Negative Massive nec Positive 5 years Well 7
    • HPV-B19, human parvovirus B19; nec, necrosis; NR, not reported; PCR, polymerase chain reaction.

    The treatment of FHF associated AA is yet to be standardized. Hadzic et al reported their experience of over 17 years observed a total of 20 patients who developed BMF after FHF 13. Only five of these patients received BMT for the management of AA. The overall mortality in their cohort was 20% with no mortality in patients undergoing BMT. The success rate of BMT for AA is reported from 75% to 90% with the best outcome being observed when a fully matched sibling donor is available. Additionally, BMT has the advantage of elimination of possible late hematological complications which are seen in nearly 10% of patients treated with IS for the management of AA. No mortality has been reported so far among patients who have undergone LT and BMT for FHF and associated AA.

    There is always concern for the development of GVHD, sinusoidal obstruction syndrome (SOS) in the transplanted liver, drug hepatotoxicity and serious infections in patients undergoing BMT after LT. GVHD has been reported in one patient so far who underwent BMT after LT for FHF and associated AA 6. One of our patients developed GVHD that was successfully treated with corticosteroids.

    SOS is a clinical syndrome characterized by the triad of hepatomegaly, fluid retention, weight gain and elevated bilirubin. The reported incidence of SOS varies from zero to as high as 60% especially following high-dose myeloablative conditioning therapy. Several risk factors for the development of SOS have been identified, including patient age, disease status, number of transplants, anastomotic stricture of the hepatic veins and liver status just prior to BMT. Nearly 70% of patients with SOS recover spontaneously with supportive care, including management of sodium and water balance, preservation of renal blood flow and repeated paracentesis to relieve abdominal discomfort. It has been suggested that prophylaxis with ursodiol prevents SOS, but the largest randomized trial of ursodiol that specifically tracked SOS as an end point found no evidence of protection 26. Infusion of defibrotide has also been shown to be effective prophylaxis against SOS in children 27. Other options for the prevention of SOS focus primarily on the modifications of dose and type of chemotherapy drugs used in the myeloablative regimens. Treatment of established SOS that fails to recover is difficult. Although multiple drugs have reportedly been used for the management of SOS, none is recommended as standard treatment at this time. Development of SOS has been reported in only one patient who underwent both LT and BMT for FHF and associated AA 11. This is also the only reported case besides the two in this report where the allograft was obtained from a living donor.

    Patients, especially children who present with features of FHF should be closely monitored for early signs of BMF. HPV-B19 should always be on the list of differential diagnosis in these patients. Use of myelotoxic drugs should be avoided in these patients. Pancytopenia should be managed aggressively with broad-spectrum antibiotics and antifungals. These patients should be evaluated for BMT early in their disease course. In the meantime, immune-modulatory treatment with the use of ATG can be initiated while awaiting a suitable donor for BMT, keeping in mind that infectious complications are responsible for a majority of deaths in patients who fail to respond to such regimen (see Figure 2). The authors are not aware of any failed cases of BMT either in their personal experience or the literature in this setting except for one patient 10 who received a second BMT and did well.

    Details are in the caption following the image
    Algorithm for management of patients who have fulminant hepatic failure and associated aplastic anemia. FHF, fulminant hepatic failure; AA, aplastic anemia; OLTx, orthotopic liver transplantation; ATG, anti-thymocyte globulin; BMT, bone marrow transplantation (see text for details).

    In summary, FHF with AA requiring LT and BMT is a rare disease entity with limited understanding of the pathophysiology of simultaneous liver and bone marrow failure. All cases reported in the literature that were treated with LT and BMT failed to identify the underlying cause for the FHF and AA. This is the first report associating FHF and AA to HPV-B19. The management of these patients has evolved over time. Our experience of the clinical outcome of AA is only limited to patients who underwent LT. The severity of AA in patients with FHF is not a contraindication for LT at our center. Although some of these patients may recover with myeloablative therapy, it is not possible to predict which patients will recover with such therapy. As illustrated in Figure 2, we treat AA in patients post-OLTx with ATG. The patients who fail to respond to myeloablative therapy have a grim prognosis. The current literature, including this report supports the role of BMT in patients who have FHF and associated AA who fail to respond to conservative measures.

    Disclosure

    The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

      The full text of this article hosted at iucr.org is unavailable due to technical difficulties.