Multicenter evaluation of the new immunoassay method for TSH measurement using the automated DxI platform

Abstract Aim of the study Recently, Beckman Coulter Diagnostics set up a new TSH immunoassay for the automated DxI platform. The aim of this study was to evaluate and compare the analytical performance and clinical results of this method with those of previous method. Material and methods A multicenter study (named TSH ELAS Study) was organized using 593 serum samples, collected from healthy subjects and patients with thyroid disorders, and 13 control samples, circulated in an External Quality Assessment (EQA) scheme. Results The values of LoB and LoD, and LoQ at 20% CV were 0.0004 mIU/L, 0.001 mIU/L and 0.0023 mIU/L, respectively. Moreover, TSH concentrations > 0.01 mIU/L actually show imprecision values lower than 5% CV. This new TSH assay showed a systematic underestimation (on average of 6.25%) compared to old method, which is mainly due to larger differences between methods for samples with low TSH concentrations, related to the better analytical sensitivity of new compared to old method. In a reference population, including 279 apparently healthy adult subjects, Caucasian volunteers (mean age 43.6 years, age 20–63 years, 138 women and 141 males) the distribution of TSH concentrations was: mean (CI 95%) 1.694 mIU/L (1.588–1.779), median 1.495 mIU/L (1.412–1.588 mIU/L), 97.5th percentile 3.707 mIU/L. Conclusions The new TSH immunoassay for DxI platform shows some relevant improvements compared to the previous one: use of the most recent WHO 3rd IRP 81/563 standard and monoclonal antibodies (instead of polyclonal antibodies of the old method), and better analytical sensitivities and reproducibility.