A review of the mechanisms of valproate-induced liver injury with an emphasis on the role of reactive metabolites discovered by Tom Baillie

The anticonvulsant valproic acid (VA) is associated with a range of idiosyncratic drug reactions; the most common is liver injury. The liver injury can take several forms: hyperammonemia with encephalopathy, liver injury with hepatic microvesicular steatosis, liver injury without steatosis, and a Reye’s-like syndrome. The most common form leading to liver failure is the injury with hepatic microvesicular steatosis. This is different from most drug-induced idiosyncratic liver injury. Children under the age of 2 and those with inborn errors in metabolism are at increased risk. VA is a simple branched carboxylic acid, but it is metabolized to >20 metabolites. One metabolite, 2-n-propyl-4-pentenoic acid, resembles the structure of hypoglycin A, the cause of Jamaican Vomiting Sickness. However, there does not appear to be a good correlation between the level of this metabolite and liver injury. Tom Baillie discovered that the 4-ene metabolite is further metabolized to the reactive 2,4-diene. There appears to be a correlation between the amount of the N -acetylcysteine conjugate of the 2,4-diene in the urine of patients and the risk of liver injury. There are many hypotheses as to how VA and its metabolites cause liver injury. It obviously causes mitochondrial injury, which is the basis for the microvesicular steatosis and hyperammonemia. It decreases the levels of Co-enzyme A and carnitine, but that is not sufficient to explain the liver injury. It also causes oxidative stress and changes in DNA methylation, but it is not clear what is responsible for the idiosyncratic nature of valproate-induced liver failure.