Abstract
Haemophagocytic lymphohistiocytosis (HLH) is a clinical syndrome of excessive inflammation and tissue destruction owing to abnormal immune activation. We report an unusual case of haemophagocytosis associated with hepatitis A virus (HAV) infection in a 21-year-old man. This was further complicated by haemolysis secondary to G-6-PD deficiency and fungal sepsis. Our patient was treated successfully with intravenous immunoglobulin (IVIg) and supportive care. A systematic review of all reported cases of HAV associated haemophagocytosis is presented.
Introduction
Haemophagocytic lymphohistiocytosis (HLH) is characterised by systemic hyper-inflammation resulting from hypercytokinaemia due to stimulated but ineffective immune process. 1 Common clinical manifestations include fever, organomegaly and cytopenia. HLH can be congenital or acquired; the congenital form is primarily a paediatric disease and associated with perforin 1 and hMunc13-4 gene mutations. 2 Acquired forms of HLH may occur in all ages and are usually encountered as a result of infection, autoimmune disease, malignancy or immunosuppression and are diagnosed by presence of any five of eight criteria which include fever, splenomegaly, cytopenia, hypertriglyceridaemia and/or hypofibrogenemia, elevated ferritin, decreased or absent natural killer (NK) cell activity, haemophagocytosis visualised in bone marrow, spleen or lymph nodes. 1 Infections causing HLH are mostly viral and include the Epstein–Barr virus, cytomegalovirus and human herpes virus (HHV-6 and HHV-8). 3 Despite a general high incidence of hepatitis A virus (HAV) infection, HLH is a rare consequence. 4 Owing to a lack of controlled studies, its treatment when associated with HAV infection is not well defined.
Case report
A 21-year-old man was admitted with symptoms of fever, jaundice and oliguria. He complained of nausea, vomiting, malaise and loss of appetite before the onset of jaundice. Laboratory tests showed markedly raised serum bilirubin of 133.4 µmol/L (direct = 80.4 µmol/L) with raised aspartate aminotransferase (AST; 1640 IU/L), alanine aminotransferase (ALT; 760 IU/L) and creatinine (564 µmol/L) levels. An initial coagulation international normalised ration (INR) was 2.3. Early blood counts showed no anaemia (haemoglobin [Hb] = 137 gm/L), no leukocytopenia (TLC; 4.2 × 109/L) and no thrombocytopenia (platelets = 130 × 109/L). The anti-HAV IgM serology was positive; serology for hepatitis B, C and E, human immunodeficiency virus (HIV), dengue, scrub typhus, leptospira, typhoid and smear for malaria parasite were all negative. During the hospital stay, the bilirubin began to rise reaching a maximum value of 889 µmol/L with a drop in Hb to a lowest value of 55 gm/L, leukocytopenia to 3.2 × 109/L and thrombocytopenia to 80 × 109/L. There were markedly raised levels of serum lactate dehydrogenase (LDH; 1447 U/L) and plasma Hb (1550 mg/L), suggestive of haemolysis, although a direct Coombs test was negative. There were also raised levels of triglycerides (7.24 mmol/L) and ferritin (19,710 pmol/L), but normal levels of fibrinogen (4.62 g/L). In addition, a G6PD deficiency was diagnosed. Six units of packed red blood cells were transfused and a bone marrow biopsy was performed which revealed an increased number of histiocytes with haemophagocytosis (Figure 1). The natural killer (NK) cell activity was 7% and perforin expression in NK cells was reduced to 54.4% (n = 84.9%). During admission, respiratory distress and acute kidney injury mandated mechanical ventilation and three sessions of haemodialysis. Intravenous meropenem and antifungal amphotericin B were infused; endotracheal tube aspirate cultures grew yeasts suggestive of aspergillae. In view of the ongoing haemophagocytosis, immunoglobulin at a dose of 400 mg/kg/day for five days was also administered. Immediate clinical improvement and normalisation of laboratory parameters were seen. The fever subsided, the anaemia improved, the bilirubin levels dropped, and liver function and renal function improved; we were able to discharge him in a stable condition after 25 days. Although fungal infection has been associated with HLH, most cases had associated systemic illness such as chronic hepatitis, liver cirrhosis, diabetes, HIV disease, chronic kidney disease and malignancy, none of which were present in our case.
5
Consequently, we attribute HLH to hepatitis A infection.
Bone marrow biopsy showing haemophagocytosis.
Methods
Literature search
We conducted a systematic search of the English literature in the PubMed and Google Scholar databases till March 2018 of the following terms: ‘hepatitis’; ‘hepatitis A virus’, ‘hemophagocytosis’; ‘haemophagocytosis’; ‘hemophagocytic lymphohistiocytosis’; and ‘haemophagocytic lymphohistiocytosis’. This was further supplemented by the direct search of the references. Both abstracts and full text articles, where available, were reviewed. Data, regarding the clinical features, treatment, course and outcomes, were extracted in a pre-designed data extraction sheet. Our search yielded 27 cases (including our case) of HAV infection associated HLH. Full texts were available for 24 cases and only abstracts available for two cases in the articles reviewed.3,4,6–26
Results
Clinical features of reported patients of hepatitis A infection and HLH.
Bone marrow examination was not done in one patient.
HLH, haemophagocytic lymphohistiocytosis; LDH, lactate dehydrogenase.
Summary of all reported patients with hepatitis A infection-associated haemophagocytosis.
HCV, hepatitis C virus; HEV, hepatitis E virus; IVIg, intravenous immunoglobulin; SJIA, systemic juvenile idiopathic arthritis.
Discussion
Virus-associated haemophagocytic syndrome (VAHS) is a life-threatening disorder. Its clinical manifestations are diverse and non-specific. VAHS should be considered as a differential diagnosis for a patient with hepatitis, non-responsive fever and cytopenia. It can affect both children and adults with HAV infection. Most cases of VAHS due to HAV infection were diagnosed in their index admission, except three cases when the diagnosis was made within 2–6 weeks after improvement of initial hepatic illness.4,7,15 The presence of underlying autoimmune disorders probably increase the risk of HAV infection-related VAHS.11,21,22 Fever was the most consistent clinical feature and present in all cases. Thrombocytopenia is the most commonly reported cytopenia in VAHS with HAV infection, followed by leukocytopenia and, rarely, anaemia. 4 Hyperferritinemia and hypertriglyceridaemia are described in HAV infection, but such high levels, as in our case, were only seen if associated with VAHS.3,4 Data of soluble CD25 (interleukin-2 receptor) and NK cell activity were reported in only a limited number of cases and cannot be commented upon. 22
The treatment protocol of HAV infection associated VAHS is not well defined. The majority of reported cases were not treated with HLH 2004 treatment protocols. 27 Some improved without any definite treatment6,20 and a variety of steroids, cytotoxic agents, plasmapheresis and IVIg have been tried either as single or combination therapy, but cytotoxic agents were never used as single therapy. Its anti-inflammatory property is the likely mechanism of action of IVIg in the treatment of VAHS. 29 Based on HLH 2004 recommendations, immunosuppressive therapies are considered to be the main stay of therapy to control haemophagocytosis. 27 Although steroids have been commonly used, their use had not been associated with uniform success. As most patients developed haemophagocytosis after several days of admission, they were likely to harbour hospital-acquired infection. Considering the scenario, IVIg seems to be the better option in treatment of VAHS due to HAV infection, although more studies are necessary to verify the its real efficacy. Two patients were successfully treated with G-CSF along with IVIg and steroids.4,17 The theory that G-CSF increases haemophagocytosis is not proven, but it may be a useful drug nonetheless for leukocytopenia due to VAHS.
The overall mortality rate was 22.2% and the clinical outcome was found to compare relatively well with that of HLH associated with other secondary causes. 3 As some of the clinical criteria of HLH usually develop later in the course of the disease, diagnosis and early treatment are imperative.
Conclusion
There is currently no consensus in the literature regarding the optimal treatment of VAHS due to HAV infection.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
