Putting Propylthiouracil in Perspective

Abstract

The antithyroid drugs propylthiouracil (PTU) and methimazole (MMI) have played central roles in the management of hyperthyroidism for more than 50 yr. Although both drugs effectively control hyperthyroidism, observations over several decades have shown that MMI and its prodrug carbimazole are better than PTU in controlling more severe hyperthyroidism, having higher adherence rates, and causing less toxicity, especially when prescribed in lower doses (1).Thishas led to therecommendationthatMMIshouldbe the first-line drug when antithyroid drug therapy is initiated, either for primary treatment or to prepare a patient for radioiodine or surgery. An exception to this rule has been pregnancy, during which PTU has been preferred because of rare reports of birth defects associated with MMI(2).PTUhasalsobeenused inpatientswhohadminor reactions to MMI but, nonetheless, preferred to continue antithyroid drug therapy. PTU may also be preferable in patients with life-threatening thyrotoxicosis because of its additional inhibition of T4 to T3 conversion. It is in thiscontext thatcontinuedPTUuseasasecond-lineagent, a first-line agent in pregnancy, and routinely by some practitioners has been reevaluated at two meetings in the last 6 months. The first meeting, which was sponsored by the Eunice Kennedy Shriver National Institute of Child Health and Development on October 28, 2008, examined PTU safety in children because of accumulating reports of PTU-related liver failure and death in children (3). The second meeting, sponsored by the American Thyroid Association and the Food and Drug Administration (FDA) on April 18, 2009, reevaluated the role of PTU during pregnancy, given what is known about PTU-related hepatotoxicity and MMI-related birth defects (4). At both meetings, the world’s literature on PTU hepatotoxicity was reviewed. Representatives from the FDA provided data on currentPTUandMMIprescribingpractices. Informationpertaining to PTUhepatotoxicity fromtheFDAAdverseEventReportingSystem (AERS) MedWatch Program was examined. Data on hepatic transplantations for PTU-related hepatotoxicity, provided by The United Network for Organ Sharing (UNOS), were also reviewed. At the second meeting, MMI-related aplasia cutis and more severe teratogenesis were discussed; however, no new information about the frequency or the cause of this problem was presented. At both meetings, mechanisms of drug-related hepatic injury were reviewed, and the role of biochemical monitoring of liver integrity in patients taking drugs known to cause hepatic damage was discussed. A complex and incomplete but, nonetheless, worrisome picture emerged from these meetings. There are 33 published reports of severe PTU-related liver failure in adults and 14 in children (see Supplemental Table 1 and Supplemental Fig. 1, published as supplemental data on The Endocrine Society’s Journals Online web site at http://jcem.endojournals.org). UNOS reported 16 liver transplants in adults and seven in children between 1990 and 2007 due to PTU-induced liver failure (5, 6) (supplemental data). Although MMI can cause liver injury, too, it is typically characterized by serious cholestatic dysfunction rather than hepatocellular inflammation (1). Indeed, over the same 17-yr period when one to three PTU-related liver transplants occurred per year, there were no liver transplants in the United States attributed to MMI toxicity. The FDA AERS databases, which overlap published reports and are subject to underreporting, detail instances of severe liver injury in 22 adults over the past 20 yr, nine of whom died and five who received liver transplants. Over the same period, 12 pediatric patients sustained severe liver injury resulting in three deaths and six liver transplants. The average daily dose of PTU associated with liver failure was approximately 300 mg in both children and adults. Liver failure occurred after 6 to 450 d (median, 120 d) of treatment. In the AERS data, there were also two reports of serious maternal liver injury due to PTU during pregnancy and two reports of liver injury in fetuses whose mothers took PTU. Because the true incidence of severe liver injury among patients taking PTU is unknown, efforts were made at the two meetings to define this figure more precisely. Based on published age-specific annual incidence data for hyperthyroidism (7) and the age distribution of the U.S. population from the 2000 census, it can be calculated thatapproximately60,000adultsdevelophyperthyroidism each year. Based on data reported at the two meetings, PTU is prescribed to one fourth of patients treated with antithyroid drugs forhyperthyroidismintheUnitedStates (8).Approximately15,000 adults are, therefore, estimated tobeginPTUtherapyperyear. If the frequency of PTU-related severe liver damage is approximately