Giant cell hepatitis (GCH) is a common descriptive histological finding in infants with neonatal cholestasis and is rare after the neonatal period. GCH has been associated with viral infections, drug-induced hepatotoxicity, autoimmune disorders, and other conditions and does not imply a specific cause or diagnostic entity. Autoimmune hemolytic anemia (AIHA) is characterized by the production of antibodies to red blood cell surface antigens with destruction of red blood cells by complement and reticuloendothelial system and is typically acute and self-limiting, with a good response to a short course of steroids (1). GCH and AIHA may present as independent conditions in children. However, the combination of GCH and AIHA is a rare distinct entity that carries poor response to immunosuppressive therapy and often progresses to fatal liver disease (1). We report on an infant with GCH and AIHA whose liver disease failed to respond to steroids, azathioprine, sirolimus, and intravenous immunoglobulin (IVIG) but improved clinically and biochemically after 4 courses of rituximab, a CD20 monoclonal antibody.
A male white infant, born at term, was referred to our hospital for evaluation of neonatal hepatitis. During the fifth week of life he became progressively pale and was brought to an emergency room because he had refused food. He was found to have congestive heart failure and a hemoglobin of 2 g/dL. A diagnosis of Coombs-positive hemolytic anemia was made, and the infant underwent transfusion with packed red blood cells. The family history was significant for a maternal grandmother with systemic lupus erythematosus and a paternal grandmother and aunt with hypothyroidism. He was treated with IVIG and oral prednisone 1.5 mg/kg. Despite receiving IVIG and steroids, he required 2 additional blood transfusions. The anemia improved, with hemoglobin levels ranging between 7 and 9 g/dL.
During a routine visit at 3 months of age, he was found to have elevated serum transaminases and was transferred to our institution for further evaluation. On admission he was pale and anicteric with hepatomegaly. His blood tests revealed a leukocyte count of 8100/mm3, hemoglobin 5.7 g/dL, positive Coombs test result, reticulocytes 2.9%, platelets 201,000/mm3, serum alanine aminotransferase (ALT) 2191 IU/L, serum aspartate aminotransferase (AST) 1867 IU/L, bilirubin 1.2 mg/dL, albumin 4.3 g/dL, and International Normalized Ratio (INR) 1.1.
Ultrasonography with Doppler revealed hyperechogenic hepatomegaly and mild splenomegaly. A liver biopsy revealed a mild increase in inflammatory infiltrate in the lobules with activation of Kupffer cells and scattered foci of giant cell transformation (Fig. 1). Electron microscopy did not reveal any evidence of infectious agents, and the hepatocytes seemed to be ultrastructurally normal. Serological evaluation was negative for hepatitis viruses A, B, and C; Epstein-Barr virus, cytomegalovirus, parvovirus B19, and HIV. The antinuclear antibody and antiliver-kidney microsomal antibody test results were negative. Anti–smooth muscle titer was initially negative and when repeated was 1:160. The total immunoglobulin G was below normal on 4 occasions (<270 mg/dL), IgA was low at 15 mg/dL, and IgM was normal at 88 mg/dL. Azathioprine 1.5 mg/kg was added to the steroid regimen. The AIHA was controlled, but the hepatitis was refractory to this conventional immunosuppressive regimen.
When the infant was 4 months old, his rising transaminase levels peaked at ALT 2092 IU/L and AST 704 IU/L, and his immunosuppressive therapy was therefore increased to prednisone 2.5 mg/kg and azathioprine 2.5 mg/kg based on 6-TG (thioguanine) levels (2). A second liver biopsy revealed giant cell transformation, enlarged hepatocytes with pale eosinophilic glassy cytoplasm suggesting rosette formations, and portal tract ductular proliferation with few interlobular damaged ducts accompanied by neutrophils. Trichrome stain showed mild portal fibrosis with focal extension beyond the portal areas (stage 1–2). On the basis of our previous experience in treating autoimmune liver disease after liver transplantation (3), sirolimus 1 mg (0.15 mg/kg) was substituted for azathioprine when transaminases were elevated, despite adequate 6-TG levels. This resulted in a partial response: transaminases declined to ALT 222 IU/L and AST 106 IU/L. The prednisone was reduced to 2 mg/kg (Table 1).
When he was 5 months old his transaminases began to rise again, resulting in augmentation of prednisone to 2.5 mg/kg and sirolimus to 2 mg (0.3 mg/kg). Despite escalation of medical treatment, the transaminases continued to mount, peaking at ALT 2807 IU/L and AST1539 IU/L, his bilirubin was 1.3 mg/dL, and the INR was 1.1. He received IVIG 400 mg/kg and IV solumedrol 20 mg/kg with only a transient response (ALT 454 IU/L and AST 177 IU/L) despite therapeutic sirolimus levels (7–18 ng/mL). Sirolimus was discontinued, azathioprine 3 mg/kg was reinstituted, and he was listed for liver transplantation.
At 9 months of age, he was admitted to the pediatric intensive care unit with decompensating liver disease: 2513 IU/L, AST 1768 IU/L, total bilirubin 6.2 mg/dL, direct bilirubin 5.2 mg/dL, albumin 1.8 g/dL, and INR 3.7 (synthetic liver function was previously normal). Coagulopathy was corrected with fresh frozen plasma and vitamin K, and IV n-acetylcysteine was empirically given. Increasing ascites developed, and diuretic treatment with spironolactone and furosemide was commenced. Empiric treatment with rituximab was commenced on the basis of a case report of AIHA responsive to rituximab in a patient with GCH and AIHA (1) and the experience of a colleague (Prof Giorgina Mieli-Vergani, King's College Hospital, London, personal communication). Weekly intravenous rituximab 375 mg/m2 was given for a total of 4 doses with no significant adverse events.
When the infant was 12 months old, the transaminases were nearly normal, allowing discontinuation of IVIG after 14 months, and diuretics. He was monitored for 2 years after diagnosis. At 21 months of age, he was weaned off steroids, and he is being allowed to outgrow his azathioprine (current dosage 3.5 mg/kg with a 6-thioguanine level of 37 U). His condition remains clinically stable, without recurrence of either GCH or AIHA to date.
The combination of GCH with AIHA is an uncommon distinct entity with poor prognosis. GCH is most commonly described in neonates and is associated with a variety of autoimmune disorders, drug reactions, viral infections, and other processes (1). The exact cause of this condition is unknown and is presumed to be an autoimmune disorder. There have been 20 reported cases of GCH with AIHA (1,4–11). All of the patients had disease onset between 4 and 24 months of age, with our patient being the youngest at 3 months. The AIHA usually precedes the diagnosis of GCH by 1 week to 15 months (usually 1–2 months), and the hepatitis presents more insidiously (9). Most patients initially respond to immunosuppression (7) but usually relapse to have an aggressive course. Immunosuppressants tried for the treatment of liver disease are steroids, azathioprine, cyclosporine A, tacrolimus, mycophenolate mofetil, sirolimus, vincristine, IVIG, and plasmapheresis with varying degrees of benefit, usually not sustained (4–6,8). In 11 of the 20 cases (60%) the outcome was fatal; 82% were secondary to liver failure (1).
Orthotopic liver transplantation was performed in 4 cases of GCH refractory to immunosuppression, resulting in the death of 3 (75%) and GCH recurrence in the survivor (6). Only 4 previously reported patients had positive autoimmune markers (4,7,9). Our patient had initially negative anti–smooth muscle titer that subsequently became positive, a family history of autoimmune diseases, and low immunoglobulin G. The autoimmune markers may develop later, necessitating episodic tests. It may also be that we have yet to define the specific autoimmune marker in this particular combination of GCH and AIHA. The histological findings, serology, ultrasonography, and relatively normal glutamyl transpeptidase and alkaline phosphatase were not consistent with autoimmune hepatitis with overlap syndrome or autoimmune sclerosing cholangitis. The failure of liver transplantation in GCH with AIHA is striking in comparison with other types of GCH found in children, strengthening the hypothesis of an underlying immune dysregulation.
Rituximab, an anti-CD 20 monoclonal antibody, inhibits B cell proliferation and prevents further antibody production. It has been effectively used in chronic refractory AIHA (12). In 1 case report rituximab was used successfully to treat AIHA with GCH in a patient whose liver disease responded to initial steroid therapy, but the AIHA was refractory to immunosuppression (1). In our case report the AIHA responded to steroids; however, the liver disease was refractory to steroids, azathioprine, IVIG, and sirolimus. Once the liver disease decompensated, rituximab was used as rescue therapy. The rituximab infusion was well tolerated and resulted in significant sustained improvement of the liver function after the first dose.
In summary, GCH with AIHA is a rare and distinct entity, predominately seen in young children. It often has poor response to immunosuppression and a fatal outcome. We report the first use, as far as we are aware, of effective rituximab therapy in a case of GCH refractory to maximal conventional immunosuppression. The treatment was well tolerated and resulted in sustained dramatic improvement of the liver function. Given the poor outcome after liver transplantation secondary to recurrence of the disease, we conclude that rituximab should be considered for patients with GCH who do not respond to conventional immunosuppression.
1. Gorelik M, Debski R, Frangoul H. Autoimmune hemolytic anemia with giant cell hepatitis: case report and review of the literature. J Pediatr Hematol Oncol 2004; 26:837–839.
2. Rumbo C. Azathioprine metabolite measurements: its use in current clinical practice. Pediatr Transplant 2004; 8:606–608.
3. Kerkar N, Dugan C, Rumbo C, et al
. Rapamycin successfully treats post-transplant autoimmune hepatitis. Am J Transplant 2005; 5:1085–1089.
4. Hartman C, Berkowitz D, Brik R, et al
. Giant cell hepatitis with autoimmune hemolytic anemia and hemophagocytosis. J Pediatr Gastroenterol Nutr 2001; 32:330–334.
5. Hadzic N, Portmann B, Lewis I, et al
. Coombs positive giant cell hepatitis—a new feature of Evans' syndrome. Arch Dis Child 1998; 78:397–398.
6. Melendez HV, Rela M, Baker AJ, et al
. Liver transplant for giant cell hepatitis with autoimmune haemolytic anaemia. Arch Dis Child 1997; 77:249–251.
7. Weinstein T, Valderrama E, Pettei M, et al
. Early steroid therapy for the treatment of giant cell hepatitis with autoimmune hemolytic anemia. J Pediatr Gastroenterol Nutr 1993; 17:313–316.
8. Imgrueth M, Wagner HP, Pipczynski-Suter K, et al
. Plasma exchange: an important part of the therapeutic procedure in a small child with autoimmune hemolytic anemia. Acta Paediatr Scand 1986; 75:1037–1041.
9. Kashyap R, Sarangi JN, Choudhry VP. Autoimmune hemolytic anemia in an infant with giant cell hepatitis. Am J Hematol 2006; 81:199–201.
10. Choulot JJ, Parent Y, Etcharry F, et al
. Giant cell hepatitis and autoimmune hemolytic anemia: efficacy of splenectomy on hemolysis]. Arch Pediatr 1996; 3:789–791.
11. Vajro P, Migliaro F, Ruggeri C, et al
. Life saving cyclophosphamide treatment in a girl with giant cell hepatitis and autoimmune haemolytic anaemia: case report and up-to-date on therapeutical options. Dig Liver Dis 2005; 38:846–850.
12. Gottardo NG, Baker DL, Willis FR. Successful induction and maintenance of long-term remission in a child with chronic relapsing autoimmune hemolytic anemia using rituximab. Pediatr Hematol Oncol 2003; 20:557–561.