Objective: Immune methods those were used in the laboratories for serum TSH measurements, were standardized. Functional sensitivity of TSH tests is expected to be ²0.02 mIU/L. We aimed to compare the performances of three different immunassay devices for third generation TSH measurement.

Materials and Methods: Third generation TSH measurements were performed with Immulite 2000 (Siemens), DxI 800 (Beckman Co ulter), and Architect i2000 (Abbot Diagnostics) immunassay devices which used chemil uminescence method. MAS Liquimmune (Microgenics) was used as control serum. Low, mediate and high level control samples were measured with three devices concurrently for 20 times both in a day and interdays. 20 samples with low TSH level were evaluated with three devices concurrently. SPSS version 11 and Pearson correlation analysis were used for statistical evaluation.

Results: Variation coefficient for 3 level control serums during the reproducibility study were as follows: 3.8-2.5-2.4% for Architect i2000, 4.0-3.3-3.3% for Dxl 800 and 5.1-4.1-5.7% for Immulite 2000 devices and interday variation coefficients were 3.8-2.6-3.3%, 4.0-6.0-2.3% and 6.9-6.2-7.3% respectively. Evaluation of TSH results showed a positive and strong relationship between the correlation analysis of Architect i2000, Dxl 800 and Immulite 2000 devices (r=0.883, p<0.001 and r=0.979, p<0.001).

Conclusion: It was determined that all three immunassay devices give correct and reliable results for Third generation TSH measurements in clinical laboratories. It is concluded that in the selection of devices it is necessary to consider the working conditions and to evaluate quick results, cost and the other factors.