BACKGROUND & AIMS:

Drug-induced liver disease is the leading cause of acute liver failure in the United States. Accurate reporting of drug-induced liver injury is essential for early detection of hepatotoxicity and for developing reliable, interpretable literature. We assessed the extent to which published case reports of drug-induced liver disease include sufficient clinical data for interpreting the cause of toxicity.

METHODS:

We developed a list of 42 predetermined, specific minimal elements necessary in evaluating causality of drug-induced liver injury. We then analyzed 97 published case reports or series studies of hepatotoxicity from 6 drugs (from 3 classes): amoxicillin/clavulanic acid (n = 35), troglitazone (n = 32), rosiglitazone (n = 10), pioglitazone (n = 8), zafirlukast (n = 8), and montelukast (n = 4).

RESULTS:

Patient age, sex, primary disease, and drug name were reported in most, if not all, published case reports. However, many elements were underreported; some publications did not mention initial bilirubin levels (12%), many did not provide initial alkaline phosphatase levels (58%), and others provided vague descriptions of how certain diagnoses were excluded, that is, tests for hepatitis A, B, and C were negative. Data on abnormal results from serial liver tests frequently were absent. Exclusions of competing viral etiologies were reported in less than 50% of the studies.

CONCLUSIONS:

Reports of drug-induced liver diseases often do not provide the data needed to determine the causes of the adverse effects. Efforts to promote and include a list of essential diagnostic elements in research articles could increase the quality and clinical utility of published case reports of drug toxicity.