[Clinical significance of the ultrasensitive TSH assay]

The authors attempted to answer the question whether the low thyroid stimulating hormone (TSH) levels measurable by the TSH ultrasensitive DELFIA kit have any clinical significance and whether they are more informative than the results obtained by the supersensitive TSH assay. No measurable TSH was detected in 111 sera among 896 random specimens, by using a supersensitive fluorimetric kit. These 111 sera were further investigated, TSH was measured by an ultrasensitive assay, in addition, the levels of the peripheral hormones (total T4, total T3, T3-uptake, free T4, free T3), were also determined. On basis of the latter, the patients were classified as having subclinical (n = 28) or manifest (n = 80) hyperthyroidism. The TSH levels of the patients affected by manifest hyperthyroidism were found significantly (p < 0.0001) lower than those encountered in subclinical hyperthyroidism. The groups were then further divided to homogeneous clinical subgroups (patients treated with thyrostatic drugs, untreated patients, toxic adenoma, Graves' disease) and the results were analyzed. It can be stated that the ultrasensitive test safely distinguishes manifest and subclinical disease in all subgroups (range of sensitivity: 90.0-94.7%). Specificity for the diagnosis of subclinical hyperthyroidism was 66.7% for the untreated subgroups, irrespective of aetiology, while in treated patients the value of specificity was 10%. In Graves' disease, specificity was 100%, in toxic adenoma 0% (the number of patients, however, was very small in these homogeneous subgroups). These results suggest that although the ultrasensitive method furnishes more information than the supersensitive test, its exclusive application would not be appropriate in characterizing thyroid function because of the broad range of individual scatter.