Introduction
	General internists, as well as specialist endocrinologists, need to know about drug effects on the pituitary-thyroid axis in order to distinguish spontaneous abnormalities of thyroid function from the often transient effects of medications. This review will outline how medications can influence thyroid function, with emphasis on substances that present special problems such as amiodarone, heparin, lithium, dopaminergics and phenytoin, as well as agents that alter immune function. An outline will also be given of how the estimation of free T4 and T3 can be influenced by substances that displace iodothyronines from plasma binding proteins. The conventional antithyroid drugs, the multiple effects of iodine itself and the potential for environmental substances to disrupt thyroid function have been reviewed elsewhere (see Roti and Braverman, 1997; Cooper, 1984; DeVito et al, 1999). The precise balance of the pituitary-thyroid axis, and the assays that are used to assess it, can be modified by many drugs - no other endocrine system is as susceptible to unintended drug interference. Already in 1981, Wenzel listed over 50 medications that could influence in vitro tests of thyroid function - the range of compounds that inhibit binding of T4 to TBG has more than doubled in the past 20 years (Lim et al 1988, Bishnoi et al 1994). Since this topic was reviewed in 1995 by Surks and Sievert, there have been advances in our understanding of the effects of cytokines, monoclonal antibodies, heparin, phenytoin and amiodarone. The major mechanisms by which medications can alter the function or testing of the pituitary-thyroid axis are listed in table 1. The effects can be complex because some important medications, for example amiodarone, glucocorticoids and diphenylhydantoin can influence the axis at more than one point (table 2).