Etiology, Outcome and Prognostic Indicators of Childhood Fulminant Hepatic Failure in the United Kingdom *Way Seah Lee, †Patrick McKiernan, and †Deirdre Anne Kelly *Department of Paediatrics, University of Malaya Medical Centre, Kuala Lumpur, Malaysia; and †Liver Unit, Birmingham Children’s Hospital, Birmingham, United Kingdom ABSTRACT Objective: To study the etiology, outcome and prognostic indi- cators in children with fulminant hepatic failure in the United Kingdom. Design: Retrospective review of all patients ,17 years with fulminant hepatic failure from 1991 to 2000. Fulminant hepatic failure was defined as presence of coagulopathy (prothrombin time .24 seconds or International Normalized Ratio .2.0) with or without hepatic encephalopathy within 8 weeks of the onset of symptoms. Setting: Liver Unit, Birmingham Children’s Hospital, United Kingdom. Results: Ninety-seven children (48 male, 49 female; median age, 27 months; range, 1 day–192.0 months) were identified with fulminant hepatic failure. The etiologies were: 22 meta- bolic, 53 infectious, 19 drug-induced, and 3 autoimmune hepa- titis. The overall survival rate was 61%. 33% (32/97) recovered spontaneously with supportive management. Fifty-five children were assessed for liver transplantation. Four were unstable and were not listed for liver transplantation; 11 died while awaiting liver transplantation. Liver transplantation was contraindicated in 10 children. Of the 40 children who underwent liver trans- plantation, 27 survived. Children with autoimmune hepatitis, paracetamol overdose or hepatitis A were more likely to sur- vive without liver transplantation. Children who had a delay between the first symptom of liver disease and the onset of hepatic encephalopathy (median, 10.5 days versus 3.5 days), higher plasma bilirubin (299 mmol/L versus 80 mmol/L), higher prothrombin time (62 seconds versus 40 seconds) or lower al- anine aminotransferase (1288 IU/L versus 2929 IU/L) levels on admission were more likely to die of fulminant hepatic failure or require liver transplantation (P , 0.05). On multivariate analysis, the significant independent predictors for the eventual failure of conservative therapy were time to onset of hepatic encephalopathy .7 days, prothrombin time .55 seconds and alanine aminotransferase #2384 IU/L on admission. Conclusions: Children with fulminant hepatic failure with se- vere coagulopathy, lower alanine aminotransferase on admis- sion and prolonged duration of illness before the onset of hepatic encephalopathy are more likely to require liver trans- plantation. Early referral to a specialized center for consider- ation of liver transplantation is vital. JPGN 40:575–581, 2005. Key Words: Fulminant hepatic failure—Etiology—Outcome— Prognostic indicators. Ó 2005 Lippincott Williams & Wilkins INTRODUCTION The etiology of fulminant hepatic failure (FHF) in children differs around the world (1). In the developing world and in certain communities in developed countries, hepatitis A is the most important etiological agent caus- ing FHF in children (2–6). Hepatitis B virus infection is the most important cause of FHF in endemic areas (7). In developed countries the cause of FHF differs according to age and geographical location (1,8). In English infants younger than 2 years, hemophagocytic lymphohistiocy- tosis, cryptogenic hepatitis and metabolic diseases ac- count for 75% of cases of FHF (9). Infectious causes were rare, accounting for only two of the 45 cases studied (9). In contrast, in France, viral hepatitis and drug- induced liver injury are important causes (10,11). With the advent of liver transplantation (LT) as an effective therapy for FHF, the importance of early reli- able prognostic evaluation of patients has become critical (11–14). Etiology of FHF and patient age may be impor- tant prognostic factors. In metabolic diseases in which the liver is persistently exposed to a toxic insult, liver in- jury is considered irreversible and transplantation is often necessary. In an acute self-limited insult, such as hepa- titis A or paracetamol overdose, exposure to liver injury may be limited and liver failure is potentially reversible. In adult patients, laboratory tests of hepatic synthetic function (plasma albumin and prothrombin time) are important prognostic indicators of FHF (14). We reviewed 97 children admitted with FHF to one of two regional pediatric liver units in the United Kingdom over a period of 10 years to determine the etiology, outcome and prognostic indicators for survival. Received July 1, 2003; accepted January 23, 2005. Address correspondence and reprint requests to Dr. W. S. Lee, Department of Paediatrics, University of Malaya Medical Centre, 59100 Kuala Lumpur, Malaysia. (e-mail: leews@um.edu.my). Journal of Pediatric Gastroenterology and Nutrition 40:575–581 Ó May 2005 Lippincott Williams & Wilkins, Philadelphia 575