Pharmaceutical preparations, and also herbal products and dietary supplements, are emerging contributors to severe forms of liver disease. Although acetaminophen intoxication is still the reason for many cases of drug-induced liver injury (DILI) in Western countries, the bulk of hepatic reactions to drugs are idiosyncratic. Only a small fraction of individuals exposed to a drug associated with liver injury will develop hepatotoxicity. Indeed, the rarity of this serious adverse event prevents its detection in clinical trials. The pathogenesis of idiosyncratic DILI is not well known because of a lack of reliable animal models, although it probably involves the metabolism of the drug and/or activation of the immune system. Different databases have described antibiotics, NSAIDs and anticonvulsants as the main group of drugs incriminated in DILI. Clinical presentation of DILI includes predominantly a hepatocellular type of damage, yet cholestatic and mixed types are also common; the determinants of the type of damage induced by a given drug are poorly understood. Analysis of pooled data has recently underlined the influence of older age in the cholestatic/mixed expression of liver injury, as well as the independent association of female gender, older age, aspartate aminotransferase levels with hepatocellular type of damage and high bilirubin levels with the risk of fulminant liver failure/death. In the long term (providing the patient survives the initial episode), persistent damage may occur in at least 6% of patients, with the cholestatic mixed type of damage more prone to becoming chronic, while in the hepatocellular pattern the severity is greater, with further likelihood of evolution to cirrhosis. Cardiovascular and CNS drugs are the main groups leading to chronic liver damage. The diagnosis of hepatotoxicity remains a difficult task owing to the lack of reliable markers for use in general clinical practice. Diagnostic algorithms may add consistency to clinical judgment by translating a suspicion into a quantitative score. Currently, the Council for International Organizations of Medical Sciences/Roussel Uclaf Causality Assessment Method instrument is considered the gold standard in causality assessment of hepatotoxicity, although there is probably room for improvement. Current efforts in collecting bona fide cases will make refinements of existing scales feasible. Efforts should also be directed towards the development of an abridged instrument for use in evaluating suspected drug-induced hepatotoxicity at the very beginning of the diagnosis and treatment process when clinical decisions need to be taken. The treatment of idiosyncratic DILI is largely supportive. Early suspicion and withdrawal of the offending agent is the most important therapeutic measure.