Has severe liver toxicity (necrosis, increased bilirubin) been reported as an adverse drug reaction

Fråga: Has severe liver toxicity (necrosis, increased bilirubin) been reported as an adverse drug reaction of valproic acid in adult patients? The question concerns a 66-year-old man with temporal glioblastoma and hypothyroidism. He started treatment with Absenor (valproic acid) 300 mg, 3x1 due to seizures related to the tumour and developed generalised jaundice with increased liver values six days later: bilirubin total 127 umol/L; AST 4.73 ukat/L; ALT 18.7 ukat/L; ALP 19.4 ukat/L; GT 29 ukat/L. Hepatitis B and C negative; HBA antigen positive (has been vaccinated). Valproic acid was discontinued another two days later. Eight days after withdrawal his liver values were: Bilirubin total 87; AST 9.66; ALT 38.63; ALP 27.2; GT 39.18.

Sammanfattning: Asymptomatic rises in liver enzymes can be observed in up to 45 per cent of patients treated with valproic acid. Severe liver toxicity has also been rarely associated in connection with valproic acid. In the literature there are over 130 fatal cases of liver toxicity due to valproic acid of which only a few concern adult patients. In Swedis there are 19 reports of liver damage of which six concern adults including two fatal cases of liver necrosis.

Svar: Valproic acid has been experimentally and clinically observed to induce hepatic dysfunction (1). Sub-clinical increases of liver values particularly liver transaminases can be observed in up to 45 per cent of patients treated with the drug during the first few weeks of treatment (2-4). Abnormal liver tests are also found and, seemingly similar frequency, among patients taking other anticonvulsants (3). Severe hepatic toxicity, occasionally with fatal outcome has also been associated with valproic acid. Since 1979 over 130 cases of fatal liver toxicity have been reported in the literature (2). The risk is reported to be highest (1:600) in children less than two years of age with complex neurological disorders and with multiple anticonvulsant therapy. In older children the incidence is around 1:37000 for monotherapy and 1:12000 for multi-drug therapy. The incidence in adults is unknown, but seems to be rare (2).

In an extensive literature search eight reports of severe liver injury in adults due to valproic acid were found (5-11). The patients were between 19-45 years old. The dose of valproic acid varied between 1200-2000 mg per day. Three patients were on multi-drug therapy with other anticonvulsants (Phenobarbital, carbamazepine or phenytoin); one patient was also taking chlorpromazine. The remaining patients were on monotherapy with valproic acid. Three patients died due to the reaction. For those who recovered, liver values were raised for several years after discontinuing valproic acid in some patients. Histological findings of the liver cells in biopsies or in autopsies showed zonal or massive necrosis of the liver associated with fatty acid changes and fibrosis. There was a rechallenge positive in at least one case (9). Two patients had hepatitis C virus infection (11).

In Swedis (12), there are 19 reports of liver damage in connection with valproic acid. Thirteen patients aged between 6 months and 17 years, while six aged between 23 and 68 years. Among the reports include two cases of liver necrosis. Both cases were women, 25 and 30 years of age. Both patients died after 2 months of treatment with valproic acid. The death was attributed to the treatment. Further details are missing.

The mechanism behind the liver toxicity due to valproic acid is not clearly known, but has been suggested to be an idiosyncratic metabolic effect of reactive metabolites of valproic acid, 4-en-valproic acid and 2,4-diene-valproic acid (13). Studies in children and adults receiving valproic acid have shown that the levels of 4-en-valproic acid increase with decreasing age, with concomitant therapy with other anticonvulsants, and with dose (13,14). These metabolites are suggested to act as a mitochondrial poison leading to depletion of acetyl CoA and carnithine (13). Carnithine is important for metabolism of long-chain fatty acids and for regulation of mitochondrial CoA.

We recommend this case be reported to the Swedish Adverse Drug Reactions Advisory Committee (SADRAC).