Clinical and laboratory features of 99 patients receiving long-term amiodarone therapy were analyzed to determine which individuals may be at a high risk for developing amiodarone-induced thyroid dysfunction. The group of 68 men and 31 women was followed up for an average of 27 months (range 3 to 60). There were no differences in age, sex, dose of amiodarone, type or severity of underlying heart disease or baseline serum thyroxine levels in patients who developed hypothyroidism (n = 32) or hyperthyroidism (n = 5) or remained euthyroid (n = 62). Baseline serum thyrotropin levels were statistically higher in patients who became hypothyroid, but there was considerable overlap with the other patient groups. Serum reverse triiodothyronine (reverse T3), which has been suggested to be a marker of amiodarone efficacy, correlated directly with serum thyroxine levels, and was not an independent variable. There was no pattern to the time course for development of thyroid dysfunction, which occurred in 49% of those followed up and developed as early as 1 month or, in one individual, as late as after 3 years of amiodarone therapy. There are few guidelines for replacement therapy in patients with amiodarone-induced hypothyroidism. L-thyroxine dosage was adjusted cautiously in these high risk individuals to achieve serum thyroxine levels within the reference range of euthyroid individuals taking amiodarone: the mean dosage required was 136 micrograms/day. Normalization of serum thyrotropin (TSH) would have required doses of L-thyroxine that were judged to be excessively high.(ABSTRACT TRUNCATED AT 250 WORDS)