Comparison of Thyroid -Stimulating Hormone and Free Thyroxine Immunoassays Performed on Immulite 2000 and Maglumi 800 Automated Analyzers
OBJECTIVE: Our goal was to evaluate the analytical performance of the novel immunoassay platform and to compare the agreement between thyroid-stimulating hormone (TSH) and FT4 results, obtained by novel and currently used platform. MATERIALS AND METHODS: Both immunoassay platforms, current Immulite 2000 and novel Maglumi 800, are based on chemiluminecsence immunoassay method. Analytical performance was evaluated by the use of serum pools and commercial quality control samples. The comparison study was carried out with 80 serum samples. Obtained results were analyzed by descriptive statistics, Mann–Whitney U-test, and Paired t-test. Method comparison was performed with Passing-Bablok regression analysis and Bland-Altman plots. RESULTS: TSH Maglumi 800 showed better within-run precision for both concentration ranges (1.7–2.8 CV%) in comparison to Immulite 2000 (4.4–5.7 CV%). FT4 Maglumi 800 imprecision was higher compared with Immulite 2000 FT4 in both within-run (3.5–3.9 CV% vs. 4.9–6.6 CV%) and between-run (3.6–4.2 CV% vs. 4.6–5.9 CV%) tests. Mann–Whitney U-test for TSH revealed non-significant difference between data (p = 0.9011). Regression analysis showed no systematic (intercept = 0.01), nor proportional (slope = 0.9781) differences. Non-significant bias was observed in Bland-Altman Plots. For FT4, we found significant differences between methods using paired t-test (t39 = 10.5, p < 0.0001) and significant difference (p = 0.00745) with Mann–Whitney U-test. Bland-Altman plot revealed 22.8% average bias. CONCLUSION: TSH evaluation showed good precision and close agreement between Maglumi 800 and Immulite 2000 methods, which assures transferability of results. However, FT4 performance evaluation revealed higher imprecision of Maglumi 800 platform and significant differences of test results.