The results suggest that major life events, personality traits of hypochondriasis and depression, paranoia, mental fatigue, and daily problems aggravate the prognosis of antithyroid drug-treated hyperthyroidism. Escape from life events is virtually impossible; thus coping strategies suggested by the physician may be useful in improving prognosis in Graves' disease.

We report a 48-year-old woman who developed hyperthyroidism following primary hypothyroidism. The serum T4 level was initially low and serum TSH level was high with clinical signs of hypothyroidism. The thyroid gland was not enlarged. Therapy with L-T4 was started. Three years later she developed hyperthyroidism; serum free T4 increased to 29.1 pmol/l after cessation of L-T4 therapy. The 123I thyroid uptake was increased with no suppression by exogenous T3. When she was hypothyroid, the activity of thyroid stimulating antibodies (TSAb) in serum measured by cyclic AMP production in cultured porcine thyroid cells were negative at 93.4% (normal less than 140%), while thyroid stimulation-blocking antibodies (TSBAb) determined by inhibition of TSH-induced cyclic AMP increase were positive at 96.1% (normal less than 40%). When hyperthyroidism subsequently occurred, TSBAb became negative (30.9%), while TSBAb became positive (163.3%). The findings indicate that hypothyroidism due to the potent TSBAb activity is not always persistent, but can be changed when various types of thyroid-relating antibodies change in the course of the disease.

This study investigated 1) the relationship between thyroid volume and thyroid function in radioactive iodine (RAI) treatment for Graves' disease, and 2) the activity of thyroid-related Ig in serum on the responsiveness of thyroid tissue to RAI. The changes in thyroid volume per megabecquerel (MBq) of 131I retained in thyroid tissue was calculated by ultrasonography as a quantitative indicator of the effect of RAI on thyroid volume. Of the 52 patients treated with 131I (3.7 MBq retained/g thyroid tissue), 26 patients showed thyrotoxicosis, 20 patients became euthyroid, and 6 patients developed hypothyroidism 6 months after therapy. The change in thyroid volume per MBq 131I was lower (P < 0.01) in the hyperthyroid patients than in the euthyroid or hypothyroid patients. The activity of thyroid-stimulating antibody in serum immediately before the therapy was greater (P < 0.01) in the hyperthyroid patients than in the euthyroid patients and was greater (P < 0.05) in the euthyroid patients than in the hypothyroid patients; it was inversely correlated with the changes in thyroid volume per MBq 131I (r = -0.667; P < 0.01). Accurate measurement of changes in thyroid volume during the course of RAI treatment provides evidence of the responsiveness of Graves' disease thyroid tissue to RAI, which is related to the outcome of thyroid function. Thyroid-stimulating antibody determination may be useful in deciding the appropriate dose of RAI to obtain euthyroidism instead of hyperthyroidism.

We have studied serum concentrations of immunoassayable tumor necrosis factor-alpha (TNF alpha) and iodothyronines (T4, T3, and rT3) in normal subjects (n = 16) and patients with nonthyroidal illnesses (NTI; n = 13), hyperthyroidism (n = 10), and hypothyroidism (n = 9). The mean (+/- SEM; femtomoles per mL) serum concentration of TNF alpha was 45 +/- 4.3 in normal subjects, 84 +/- 38 in NTI, 54 +/- 6.0 in hyperthyroidism, and 50 +/- 10 in hypothyroidism; the various values did not differ significantly from one another. Serum TNF alpha was well within the normal range in all NTI patients, except one patient with a brain glioma and infection in whom it was elevated (540 fmol/mL). There was no significant correlation between serum TNF alpha and serum T4, T3, or rT3 levels in NTI patients. Similarly, there was no correlation between serum TNF alpha and serum thyroid hormone (T3 or T4) levels when data in normal subjects were combined with those in NTI patients. The dialyzable fraction of T3 and the free T3 concentration did not correlate with serum TNF alpha levels. However, there was a tendency toward a positive correlation between dialyzable fraction of T4 and the serum concentration of TNF alpha in NTI (r = 0.54; n = 11; 0.05 greater than P less than 0.1). The relationship between these two parameters became more clear when data in normal subjects and NTI patients were combined for statistical analysis (r = 0.59; n = 22; P less than 0.005). The free T4 concentration correlated positively with serum TNF alpha levels whether the data in NTI patients were analyzed alone (r = 0.93; P less than 0.001) or in combination with data from normal subjects (r = 0.85; P less than 0.001). Our data suggest that circulating TNF alpha may contribute to elevated free T4 in NTI. However, it is not a universal or common factor in the pathogenesis of other alterations in serum thyroid hormone levels in NTI (euthyroid sickness syndrome).