Original Articles

Indian Pediatrics 1999; 36: 1107-1112

Fulminant hepatic failure: etiology, viral markers and outcome

Sachin V. Bendre, Ashish R. Bavdekar, Sheila A. Bhave, Anand N. Pandit, S.D. Chitambar* and V.A. Arankalle*

From the Department of Pediatrics, K.E.M. Hospital, Pune 411 011, India and *Hepatitis Department, National Institute of Virology, Pune, India.
Reprint requests: Dr. Shiela Bhave, Consultant in Pediatric Research, Department of Pediatrics, K.E.M. Hospital, Pune 411 011, India. E-mail: kemhrc@vsnl.com
Manuscript received: March 23, 1999; Initial review completed: May 7, 1999; Revision accepted: June 8, 1999.

Objective: To investigate the etiology and outcome of fulminant hepatic failure (FHF) in children. Setting: Hospital based descriptive. Methods: 36 children (22 males and 14 females) presenting with FHF over a period of one year were investigated. The ages ranged from 1.5 to 9 years. FHF was defined as occurrence of encephalopathy within eight weeks of onset of jaundice with no evidence of pre-existing liver disease. Detailed history, clinical examination, routine biochemical parameters and relevant diagnostic tests were carried out. Viral markers studied were anti HAV-IgM, HBsAg, anti HBc-IgM, anti-HCV and anti HEV-IgM. Results: A viral etiology could be established in 22 children (61.1%). Hepatitis A (n=12), Hepatitis B (n=3), Hepatitis A and B (n=2), and Hepatitis A and E (n=4). Two children had enteric fever (1 with associated HEV), 2 children had Wilson's disease, 1 child had Indian Childhood Cirrhosis (ICC) and 2 children had drug induced hepatitis. Etiological diagnosis was not possible in 8 children (22%). Fourteen children (39%) died. Poor outcome was associated with spontaneous bleeding, raised prothrombin time, lower transaminases and higher bilirubin on admission. Conclusion: Viral hepatitis is the commonest cause of FHF in children. HAV alone or in combination is responsible for upto 50% of all FHF in children. Chronic liver disease can also present as FHF. Etiological diagnosis is not possible to upto one-fourth of all cases.

Key words: Fulminant hepatic failure, Viral hepatitis.

Fulminant hepatic failure (FHF) is one of the leading causes of death in hospitalized children in India(1). The condition is particularly distressing as it occurs acutely, in previously `healthy' children and progresses rapidly inspite of all modern treatment. FHF is commonly presumed to be due to hepatotropic viruses, however, poisons, toxins and metabolic diseases can also cause the disease(2). Two recent developments make the etiology and outcome of FHF especially relevant: (a) the widespread availability of effective vaccines for hepatitis A and B; and (b) the successful use of liver transplantation in the treatment of terminal liver cell failure(3,4). The present study was therefore conducted to: (i) assess the frequency and clinical features of FHF in hospitalized children; (ii) ascertain if viral markers and metabolic tests can help clarify etiological differentiation; and (iii) assess the outcome in relation to etiology and clinical features.

Subjects and Methods

Definition: FHF was defined as occurrence of   encephalopathy within eight weeks of onset of jaundice in a child with no prior evidence of liver disease.

Patients: 36 children who presented with FHF to K.E.M. Hospital, Pune, over a period of one year from January 1997 to December 1997 were investigated. Detailed clinical evaluation included history, physical signs of liver cell failure and staging of hepatic encephalopathy. Routine investigations included renal function tests and liver function tests (S bilirubin, SGPT, serum alkaline phosphatase, prothrombin time). Special investigations like alpha fetoprotein, copper studies (S. ceruloplasmin, urinary copper, liver copper), Widal test, blood culture, drug levels, and other metabolic studies (serum and urine aminoacid levels), were carried out wherever indicated.

Viral Studies: All serum samples were tested for IgM antibodies to Hepatitis A virus (IgM anti-HAV), Hepatitis B surface antigen (HBsAg), IgM antibodies to Hepatitis B core antigen (IgM anti-HBc) (Abbott USA), antibodies to Hepatitis C virus (anti HCV) (RIBA-3, Chiron USA) and IgM antibodies to Hepatitis E virus (IgM anti-HEV). HBsAg positive samples were tested for antibodies against Hepatitis delta virus (anti HDV, Abbott USA). All the samples were also screened for HCV RNA in nested PCR employing primers from highly conserved 5' non-coding region.

Management: All patients were nursed in the Pediatric Intensive Care Unit (PICU) with close monitoring and standard FHF protocols. Antibiotics, fresh frozen plasma, blood and anticerebral edema measures (mannitol) were used wherever indicated.

Results

Clinical Features: Of the 36 children with FHF, 22 were male and 14 female. Ages ranged from 1.5 years to 9 years with a mean of 4.2 years (SD 2.25 years). Twenty two children presented in Grades I-II encephalopathy and 14 children in Grades III-IV encephalopathy. At presenta-tion 8 (22%) children had bleeding (5 with upper GI hemorrhage, 3 with bleeding at multiple sites_GI, skin, epistaxis), 6 (16.5%) had ascites clinically and on USG. A diagnostic tap was done in 4 subjects and 3 showed transudate while 1 had exudate. Hypoglycemia was present in 3 (16.5%), seizures in 3 and severe infections in 4 (septicemia in 2, pneumonia in 1 and urinary tract infection in 1) (Table I).

Table I^__Clinical and Laboratory Features.

* p<0.05 (Student `t' test)  

Viral Markers and Etiology: One or more viral markers were positive in 22 (61.1%) of the 36 children studied. The distribution of the various viral markers is shown in Fig. 1.

Fig. 1. Positive viral markers in 36 patients of fulminant hepatic failure. HAV_Hepatitis A virus, HBV_Hepatitis B virus and HEV_Hepatitis E virus.

HAV infection alone or as co-infection was found in 50% of all children with FHF and in 82% of all with positive viral markers. HBV and HEV alone or in combination were seen in 10 cases. None were positive for Hepatitis E alone. None were positive for anti-HCV antibodies or HCV RNA.

Two children were diagnosed to have Wilson's Disease (serum ceruloplasmin 5.2 mg/dl and 4.6 mg/dl, urinary copper excretion 288.6 mg/day and 980 mg/day, hepatic copper levels 891.5 mcg/g and 557.6 mcg/g of dry liver weight, respectively). One child was diagnosed to have ICC clinically and by liver biopsy (postmortem). Two children had drug-related hepatitis. Both the children were known epileptics, one child was on sodium valproate plus carbamazepine therapy and the other child was on sodium valproate therapy alone. Serum sodium valproate levels were high in both (221 mg/dl and 182 mg/dl, respectively). Two children were diagnosed to have enteric fever (blood culture for S. typhi and Widal test positive, CSF was normal and sterile, CT scans were not done). One of them was also positive for anti-HEV IgM antibodies (Table II). No etiological diagnosis could be established in eight (22.2%) children with FHF.

Table II^__Mortality in Relation to Etiology of FHF

Outcome: Fourteen of 36 children admitted with FHF died (39%); 5 children died of hemorrhage (2 with upper GI and 3 with bleeding from  multiple sites), 4 of severe infection, 3 of resistant seizures and 2 of respiratory arrest. Most (n = 12) of these children died within five days of onset of encephalopathy. Two died on the seventh and ninth day of admission. Outcome in viral hepatitis was significantly better than in the other conditions (Table II). The non-survivors had significantly lower SGPT and higher serum bilirubin and prothrombin time than survivors (Table I). Only two children who presented with Grades III and IV encephalopathy survived, whereas only two children who presented in Grades I and II encephalopathy died.

Histology: Biopsies were possible in 16 children.Two showed features suggestive of Wilson's Disease (liver copper was confirma-tory). The liver biopsy was diagnostic of ICC in one (pericellular fibrosis and orcein stain positive). The biopsies in the two children with drug-induced FHF showed acute yellow atrophy and fatty infiltration. Widespread granulomas were seen in the post-mortem biopsy of the child with enteric fever. The diagnosis of enteric fever in this child was confirmed by a positive Widal test and blood culture for S. typhi. The rest of the biopsies showed features of acute massive necrosis.

Discussion

During the study period of one year, FHF accounted for 4% of all our PICU admissions, and 9% of all deaths above the age of one year. Only a few studies on prevalence of FHF in India are available(1,5-7). Kapil et al. found 2% of all their PICU admissions to be due to FHF with a mortality of over 50%(1).

The etiology of FHF shows interesting  patterns in various age groups and various geographical areas in different countries and even within the same country(2). Viral hepatitis A and E are common causes of FHF in the developing world(1,5-7) whereas, hepatitis B is predominant in the far East and hepatitis C in Japan(2,8). Viral hepatitis as a cause of FHF is by and large uncommon in Western Europe and other developed countries, where the chief offenders are drugs and toxins, especially acetaminophen(2). Studies in adults show predominantly HBV and less frequently HCV as a cause of FHF(8,9). HEV is a significant cause of FHF in pregnant women(10). In children the predominant causative agent appears to be HAV alone or in combination with other infectious agents(1,5,6,8). In our present study HAV was responsible in 50% of all patients with FHF and in 82% of those with positive viral markers. A recent study on FHF in Delhi also highlights HAV as the major causative agent in children(6).

Apart from viral infections, many other infectious agents, notably Salmonella typhi and cytomegalovirus can cause FHF(2,11). Two of our patients with FHF were due to enteric hepatitis (one had co-infection with HEV). In fact, one of these children was diagnosed only on post-mortem biopsy which emphasises the need for early diagnosis of this treatable cause of FHF in children. Malarial hepatitis which also presents with icterus and encephalopathy should also be differetiated early, as specific therapy for malaria is associated with a good outcome(12).

Various hepatotropic drugs, notably antitubercular (ATT), antiepileptic and anti-malignancy drugs are known to cause FHF (13,14). In our study, two children who were known epileptics presented as FHF. They had high serum sodium valproate levels and could not be salvaged. We had no anti-tubercular treatment related FHF.

Our previous studies have shown that many chronic liver disorders such as Wilson's Disease and ICC, can present as FHF(15,16). The associated mortality is always very high.

Inspite of a concerted effort, we could not find any definite cause for FHF in 22% of our patients. Similarly, a recent Delhi study could not ascribe any cause to 12.5% patients of FHF(6). Obviously, hitherto unknown environ-mental toxins, metabolic disorders, non A-E viruses or other infectious agents have to be investigated as causal etiologies(17).

The importance of specific viral diagnosis is well-demonstrated in our study. The availability of kits for serological diagnosis have greatly enhanced our diagnostic abilities and management planning. However, these facilities are at present available only at select centres and are very expensive for routine use.

The occurrence of bleeding, higher grade of encephalopathy, grossly deranged PT (> 50 sec), high serum bilirubin and low SGPT are now established indicators of poor prog-nosis(18,19). Our study generally confirms these findings.

Conventional therapy of FHF is expensive and frustrating. Until the advent of liver transplantation the mortality ranged from 50 to 90% inspite of many heroic measures including artificial liver support systems(20). Liver transplantation in selected patients has greatly altered the mortality in FHF in developed countries(4). It is obvious that this modality though expensive is even more needed in our country where FHF is common.

Our study highlights that FHF is largely a preventable disease. Sixty one per cent of all our cases were caused by diseases which can be prevented by immunizations and hygienic measures. Hepatitis B and typhoid vaccination have already become popular and are now, widely available in our country.  Hepatitis A is generally considered a mild disease occurring mainly in young children in our country and vaccination against Hepatitis A is not recommended routinely. However, our study demonstrates that Hepatitis A is the commonest cause of FHF with high mortality in young children. Further, recent reports have suggested a change in the epidemiology making older children and young adults (especially of higher socioeconomic strata) now more vulnerable to Hepatitis A(21). These facts support the need of preventive immunization for Hepatitis A infection to those who can afford it.

References

1. Kapil D, Arvind B. The profile and outcome of patients admitted in Pediatric Intensive Care Unit. Indian J Pediatr 1993; 60: 5-10.

2. Williams R. Classification, etiology and considerations of outcome in acute liver failure. Semin Liver Dis 1996; 16: 343-348.

3. Van Damme P, Thoelen S, Cramm M, De Groota K, Safory A, Mehens A. Inactivated hepatitis A vaccine: Reactogenicity, immunogenicity and long term antibody persistence. J Med Virol 1994; 44: 446-451.

4. Devictor D, Desplanques L, Debray D, Ozier Y, Dubousset AM, Valayer J, et al. Emergency liver transplantation for FHF in infants and children. Hepatology 1992, 16: 1156-1162.

5. Acharya SK, Dasarathy S, Kumar TL. Sushma S, Prasanna KS, Tandon A. Fulminant hepatic failure in tropical population: Clinical course, cause and early predictors of outcome. Hepatology 1996; 23: 1148-1155.

6. Arora NK, Nanda SK. Acute viral hepatitis types E, A and B singly and in combination in acute liver failure in children in North India. J Med Vir 1996; 48: 215-221.

7. Arankalle VA, Jha J, Favorov MO, Chaudhari A, Fields HA, Banerjee K. Contribution of HEV and HCV in causing fulminant non-A, non-B hepatitis in Western India. J Viral Hep 1995; 2: 189-193.

8. Shapiro CN, Margoles HS. World epidemiology of hepatitis A virus infection. J Hepatology 1993; 18 (Suppl 2): S11-S14.

9. Jaiswal S, Chitnis DS, Asolkar MV, Naik G, Artwani KK. Etiology and prognostic factors in hepatic failure in central India. Trop Gastroenterol 1996; 17: 217-220.

10. Khuroo MS, Rustgi VK, Dawson GJ, Mushahwar IK, Yattoo GN, Kamili S, et al. Spectrum of hepatitis E virus infection in India. J Med Virol 1994; 43: 281-286.

11. Bhutta ZA. Fulminant hepatic failure with typhoid fever in childhood. J Pakistan Med Assoc 1991; 41: 123-126.

12. Anand AC, Ramji C, Narula AS, Singh W. Malarial hepatitis: A heterogeneous syndrome? Natl Med J India 1992; 5: 59-62.

13. Tsagar OS, Mataki ET. Hepatotoxic reactions in children with severe tuberculosis treated with isoniazid-rifampin. Pediatr Infec Dis 1983; 4: 270-273.

14. Dreifuss FE, Langer DH, Moline KA, Maxwell JE. Valproic acid hepatic fatalities. Neurology 1989; 39: 201-207.

15. Bhave SA, Pandit AN. Liver disease in India. Arch Dis Child 1982; 57: 922-927.

16. Bhave SA. Pediatric Wilsons Disease in India. J Western India 1990; 18: 14-17.

17. Mutimer D, Shaw J, Neuberger J, Skidmore S, Martin B, Hubscher S. Failure to incriminate hepatitis B, hepatitis C and hepatitis E viruses in the etiology of fulminant non-A non B hepatitis. Gut 1995; 36: 433-436.

18. Srivastava KL, Mittal A, Kumar A, Gupta S, Natu SM, Kumar R, Govil YC. Predictors of outcome in FHF in children. Indian J Gastroenterol 1998; 17: 43-45.

19. Dhiman RK. Seth AK. Jain S, Chawla YK, Dilawari JB. Prognostic evaluation of early indicators in FHF by multivariate analysis. Dig Dis Sci 1998; 43: 1311-1316.

20. Demetriou AA, Rozga J, Podesta L, Lepage E, Morsiani E, Moscioni AD, et al. Early clinical experience with a hybrid bioartifical liver. Scand J Gastroenterol 1995; 30: (Suppl 208): 111-117.

21. Hadler SC. Global impact of Hepatitis A virus infection changing patterns. In: Viral Hepatitis and Liver Disease: Proceedings of the International Symposium on Viral Hepatitis and Liver Disease. Contemporary Issues and Future Prospects. Eds. Hollinger FB, Lemon SM, Margolis H. Baltimore, Williams and Wilkins, 1990; pp 14-19.