- Poster presentation
- Open Access
Acute liver failure in children in Istanbul University: retrospective analysis of 30 cases
Critical Carevolume 10, Article number: P273 (2006)
Acute liver failure (ALF) can show different demographical characteristics in different countries. Liver transplantation is still the life-saving treatment modality if performed in a timely manner in patients who will not recover with supportive measures. Prognostic criteria for pediatric ALF to discriminate children who will require liver transplantation has not yet been established. We aimed to evaluate the case for pediatric patients in Istanbul retrospectively.
Materials and methods
The files of children with ALF diagnosed during a 15-year period between 1991 and 2005 were reviewed retrospectively. All the patients were re-evaluated to confirm the diagnosis according to the current definition of ALF by an INR ratio >2 and/or prothrombin time of 24 s or above and evidence of acute liver disease (without any clinical stigmata of cirrhosis) irrespective of the presence of encepalopathy. Thirty patients (17 boys, mean age 7.1 ± 4.9 years) with the diagnosis of ALF were entered into comprehensive analysis. Clinical and laboratory data were analysed as suggested in a previous European pediatric ALF meeting held in Hamburg, October 2003. Possible prognostic factors to determine the outcome, initial and worst laboratory values were entered into relevant statistical analyses (SPSS version 10.0, Chicago, IL, USA). The chi-square test, Fisher's exact test and Mann-Whitney U test were used to compare survivors and nonsurvivors where appropriate. Pearson or Spearman tests were used to investigate the correlations.
Jaundice was observed in 90% of the children. Hepato-megaly existed in 40%, splenomegaly in 30% and ascites in 40%. Fifty-three percent had grade I or grade II encephalopathy, whereas 20% already had grade IV encephalopathy on admission. The degree of encephalopathy correlated with initial hematocrit, platelet counts and prothrombin time as well as with the presence of splenomegaly (P < 0.05 for all). Encephalopathy did not exist in 20% initially at all. None of these clinical findings influenced the outcome statistically. Mortality seemed higher in girls (69.2% vs 52.9%) and those diagnosed before the year 2000 (67% vs 50%), although the difference was not statistically significant. Survival rates seemed higher in cases of hepatitis A, hepatitis B and toxic hepatitis, whereas mortality was highest in mushroom poisoning and fulminant Wilson's disease.
The initial leucocyte count was higher (14,474 ± 7368 vs 10,390 ± 6208/mm3, P = NS); total and conjugated bilirubin were increased (26.9 ± 14.9 vs 19.3 ± 19.3 mg/dl and 18.6 ± 12.1 vs 12.9 ± 14.0 mg/dl, P = NS) as well as BUN (16.1 ± 13.3 vs 10.3 ± 7.1 mg/dl, P = NS) among nonsurvivors. The INR was significantly lower in the survivor group (4.4 ± 1.1 vs 3.1 ± 0.9, P < 0.05). The maximal leucocyte count and maximal INR levels were significantly higher among the nonsurvivors (19,623 ± 14,917 vs 11,680 ± 4483/mm3 and 4.9 ± 0.7 vs 3.8 ± 0.9, P < 0.05) whereas minimal serum glucose levels were significantly lower (45.1 ± 28.2 vs 60.9 ± 14.5 mg/dl, P < 0.05).
Hepatitis A, toxic hepatitis and fulminant Wilson's disease are the leading etiological factors in Turkish children with ALF in Istanbul. The initial INR level seems to be the only prognostic factor on admission. However, the maximal leucocyte count and INR levels as well as minimal serum glucose levels during the hospital stay may also have a role in determining the outcome. The outcome of ALF seems to be improved after the year 2000 in our series, possibly due to the improvement of intensive care facilities.