Hepatotoxicity to sodium valproate: a review.

Abstract

Progress reportHepatotoxicity to sodium valproate: a review Since 1978 a number of reports have indicated a possible association between sodium valproate therapy and the occurrence of hepatitis.In this review we have analysed, with particular reference to the clinical presentation, course and coadministration of other drugs, the 42 cases with fatal hepatitis either known to the manufacturers or reported elsewhere,'-11 and the more frequent cases with milder forms of hepatitis varying in severity from a transient rise in aminotransferase activities to symptomatic cases with reversible hepatic failure.12-30 We have also considered the three reported cases with a Reye's-like syndrome2 9 31 and 22 instances of hyperammonaemia32-38 which is usually found in the absence of clinical evidence of hepatic injury but can easily be confused with fulminant hepatic failure.Possible mechanisms for the hepatotoxicity in relation to recently reported experimental work and the value of liver function tests in monitoring valproate therapy are also considered. Hepatitis-like syndromeThe incidence of abnormal serum aminotransferase activities ranges from 0-44% in 19 clinical trials,'2-30 with an overall incidence of 11% in the 1197 patients monitored.Only two out of 57 patients with raised amino- transferase activities in three of the series'2 4 16 were symptomatic, with malaise, lethargy, and anorexia which rapidly resolved on reducing the dose or stopping the drug.Aminotransferase activities in non-fatal cases were usually between one and three times the upper limit of normal and not usually accompanied by rises in serum bilirubin or alkaline phosphatase, except in the most severe cases.9The clinical, biochemical, and histological findings in the 42 cases with a fatal outcome largely agree with those presented by Zimmerman and Ishak39 and Zafrani and Berthelot,40 in part because of the duplication of a proportion of the cases.The age at presentation ranged from 21 months to 34 years with 29 of the cases (69%) aged 10 years or less.Below the age of 15 years the proportion of males was 62-5%, but above this age the proportion was lower at 30% (Table).The overall male incidence of 57% compares with figures of 57% and 35% in the publications by Zimmerman and Ishak39 and Zafrani and Berthelot40 respectively.The disproportionate vulnerability of young individuals, particularly boys, does not appear to be a reflection of prescribing habits in that age group.Indications for valproate therapy comprised focal or generalised epilepsy, except for one patient with febrile convulsions.Evidence for structural neurological defects were present in 29% of patients.There were two instances of familial involvement.In one family two sibs of 17 and 10 years 673