Study objective: To compare the reported occurrence of liver failure in subjects in prospective trials with that in patients in retrospective reports after repeated use of therapeutic dosages of acetaminophen.
Design: Systematic review of the medical literature.
Data source: MEDLINE and EMBASE biomedical and pharmacologic databases.
Subjects: Adults who received repeated dosing of acetaminophen 4 g/day or lower for at least 24 hours.
Measurements and main results: Articles written in several languages were abstracted by trained personnel using a structured abstraction form. Data were categorized by methodology (prospective vs retrospective), acetaminophen dosage, and type of liver effect. A total of 791 articles were identified, which included 30,865 subjects in prospective studies and 9337 patients in retrospective reports. The prospective studies reported no cases of fulminant hepatic injury, liver transplantation, or death due to acetaminophen. Of the 30,865 subjects in these studies, 129 (0.4%) were identified who had a serum aminotransferase level that exceeded the upper limit of normal, including 61 subjects in randomized trials in which the proportion of serum aminotransferase level increase was the same as or less than that in the placebo group and 68 subjects in trials without a placebo group. In addition, 4263 (13.8%) received the maximum recommended therapeutic dosage (3.9-4 g/day). In the retrospective reports, 96 patients (1.0%) had a serum alanine aminotransferase level that exceeded the upper limit of normal, one (0.01%) underwent liver transplantation, and six (0.06%) died. Causality relationship of acetaminophen for each retrospective case was assessed with the Naranjo adverse drug reaction probability scale. The mean +/- SD Naranjo score for all 103 retrospective cases was 3.2 +/- 1.9, indicating a possible relationship between the increased aminotransferase levels and acetaminophen use. Some retrospective reports contained information suggesting that the patient had ingested an overdose despite a history of therapeutic use.
Conclusion: Prospective studies indicated that repeated use of a true therapeutic acetaminophen dosage may slightly increase the level of serum aminotransferase activity, but hepatic failure or death was not reported. Retrospective reports indicated a higher rate of increased serum aminotransferase levels, and several reported associated liver injury and death. The differing results and presence of evidence indicating inaccurate acetaminophen dosage information in some case reports suggests that these cases may be inadvertent overdoses, rather than true therapeutic dosages.