Evaluation of a new chemiluminescence technique for human thyrotropin (BeriLux hTSH): diagnostic value of five immunometric assay methods.

A new commercially available human thyrotropin immunochemiluminometric assay (ICMA) kit was evaluated. The BeriLux assay (Hoechst Co., Germany) was compared with two other non-radioisotopic methods (AIA-1200 and IMx) and two other immunoradiometric assays (RIA-gnost TSH IRMA and EIKEN IRMA kits) in 32 normal subjects and 104 patients with Graves' disease, divided into seven groups: 1) untreated hyperthyroidism; 2) hyperthyroidism during treatment; 3) euthyroid with negative thyroliberin test (subclinical hyperthyroidism); 4) euthyroid with low thyroliberin test; 5) euthyroid with normal thyroliberin test; 6) euthyroid with high thyrotropin level (subclinical hypothyroidism); and 7) primary hypothyroidism. Patients in groups 2-6 were undergoing treatment with mercazole and propylthiouracil. The new immunoluminometric assay (ILMA) BeriLux kit was shown to have a remarkably improved analytical and clinical sensitivity. The minimal detectable level of thyrotropin in the assay was 0.006 mU/l. The precision was 2.8% and 6.1% at 0.093 +/- 0.003 mU/l and 0.028 +/- 0.002 mU/l, respectively, whereas the precision of the other methods was above 17.2% and 59.4% respectively. Seven patients from the untreated hyperthyroid group were given 500 micrograms thyroliberin i.v. (the thyroliberin test). The thyrotropin pattern before and after thyroliberin administration was always less than 0.006 mU/l with the BeriLux kit, whereas the other methods showed random fluctuations indicating their low accuracy at this concentration. Using the BeriLux kit, 7 of the 16 overt hyperthyroid patients undergoing treatment showed a measurable thyrotropin level below 0.01 mU/l but a negative thyroliberin test.(ABSTRACT TRUNCATED AT 250 WORDS)