Graves' ophthalmopathy in the absence of elevated free thyroxine and triiodothyronine levels: prevalence, natural history, and thyrotropin receptor antibody levels.

The aims of this study were to (a) determine the prevalence of patients without elevated thyroid hormone levels in Graves' ophthalmopathy (GO) using current generation free thyroid hormone assays, (b) measure the prevalence of thyrotropin receptor antibodies (TRAb) in these cases, and (c) identify possible predictors of hyperthyroidism. Over a 30-month period, 1020 cases of thyroid eye disease were evaluated, of which only 19 (1.9%) met the diagnostic criteria. Ten (1%) had subclinical thyrotoxicosis, 7 (0.7%) were euthyroid, and 2 (0.2%) were hypothyroid as determined by a third-generation thyrotropin (TSH) assay. TRAb levels were measured in 16 of these 19 patients. The prevalence of TRAb varied according to the assay used. Polyethylene glycol-extracted thyroid-stimulating immunoglobulin (PEG-TSI), unfractionated thyroid-stimulating immunoglobulin (uTSI), first-generation porcine TSH-binding inhibitory immunoglobulin (pTBII), and second-generation human TSH-binding inhibitory immunoglobulin (hTBII) assays were positive in 93.8%, 50%, 18.8%, and 81.3% of patients, respectively. TRAb was detected by at least one method in all patients. Patients were followed up for 15 to 45 months. Hyperthyroidism developed in 4 patients (25%). Suppressed TSH levels and elevated TBII were predictors of hyperthyroidism. When sensitive assays are used, the prevalence of GO patients without elevated thyroid hormone levels is extremely low. The sensitivities of assays for TRAb detection differ substantially in these cases. PEG extraction improves the detection rate of TSI (p = 0.02), and hTBII assays improve the detection of TBII in these patients (p = 0.002). The high prevalence of TRAb in such cases supports a role for these antibodies in the pathogenesis of thyroid-associated eye disease.