Neil Kaplowitz, MD; James H. Lewis, MD; Paul B. Watkins, MD
Disclosure: The authors are consultants to GlaxoSmithKline, the manufacturers of rosiglitazone.
Kaplowitz N, Lewis JH, Watkins PB. Did This Drug Cause My Patient's Hepatitis?. Ann Intern Med. 2003;138:159. doi: 10.7326/0003-4819-138-2-200301210-00025
Download citation file:
Published: Ann Intern Med. 2003;138(2):159.
TO THE EDITOR:
Dr. Nierenberg (1) makes several important points about causality assessment in drug-induced liver disease and the role of postmarketing surveillance. The “glitazones” are used to illustrate the issues. With troglitazone, roughly 2 million patients were treated, nearly 1000 developed acute hepatitis, and nearly 100 developed acute liver failure resulting in liver transplantation or death (that is, a ratio of approximately 1 per 20 000). In contrast, more than 3 million patients have been treated with rosiglitazone and pioglitazone, and no more than a handful of cases of drug-induced hepatitis (2-7), including three cases of acute liver failure, have been published (4-6), one of which was irreversible (5). Besides the fact that several of the cases are highly confounded (for example, by concomitant ischemic hepatitis  and by use of more likely hepatotoxins such as zafirlukast  and allopurinol plus amoxicillin–clavulanic acid ), it remains unclear whether these events are in fact due to rosiglitazone or pioglitazone or whether they merely represent “background noise.” Even when some degree of underreporting is accounted for, the background incidence of idiopathic acute hepatitis (a ratio of approximately 1 to 50 000) and cryptogenic acute liver failure (a ratio of 1 to 2 per million) makes it very difficult to conclude that the few reported hepatitis cases associated with rosiglitazone or pioglitazone are truly drug induced, particularly in the absence of a common clinical presentation, more incriminating features of systemic hypersensitivity, or positive rechallenge. Thus, there is no scientific basis to consider severe liver injury as a class effect of thiazolidenediones. Although we agree wholeheartedly with the importance of postmarketing surveillance, MedWatch reporting, and the publication of case reports, rare occurrences of hepatitis (a ratio of < 1 per 50 000) in a population treated with a certain drug should not be assumed to indicate a causal relationship.
Learn more about subscription options.
Register Now for a free account.
Copyright © 2016 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use
This PDF is available to Subscribers Only