Thyroid status of patients receiving long-term anticonvulsant therapy assessed by peripheral parameters: a placebo-controlled thyroxine therapy trial.

Thyroid hormone concentrations and measures reflecting thyroid function were studied in sera from 35 patients receiving long-term phenytoin (PHT) or carbamazepine (CBZ) therapy. The mean concentrations of T4, FT4, FT3, and rT3, but not T3, of these patients were significantly lower than those of 19 controls of similar age and sex distribution. The mean serum thyrotropin (TSH) concentration was slightly but significantly higher in patients than in controls, but the serum TSH response to TRH was not significantly increased. In patients, the higher mean clinical diagnostic index of hypothyroidism (CDI-HT: -20.3 +/- 19.1 vs. -33.7 +/- 8.5, p < 0.05) and higher ratio of preejection period to left ventricular ejection time (PEP/LVET: 0.343 +/- 0.065 vs. 0.334 +/- 0.030, p < 0.05) than in controls were compatible with tissue hypothyroidism. However, comparison of the mean levels of alanine aminotransferase (ALAT), creatine kinase (CK), creatinine, triglycerides, cholesterol, high-density lipoprotein (HDL) cholesterol, osteocalcin, procollagen type III aminoterminal propeptide, and somatomedin-C showed no significant differences between patients and controls. The increased mean angiotensin convertase and sex hormone-binding globulin (SHBG) levels, typical of hyperthyroidism, were probably caused by drug effects. Fourteen patients with a subnormal FT4 concentration in serum participated in a double-blind thyroxine treatment cross-over study. Neither the mean CDI-HT score, nor the systolic time intervals were significantly different between the thyroxine and placebo periods. Five patients benefited subjectively from the treatment. On the basis of all data from the cross-sectional and thyroxine treatment studies, we conclude that patients receiving anticonvulsant drugs chronically are eumetabolic and do not need thyroxine supplementation.