Key points are not available for this paper at this time.
Abstract Disclosure: A. Ribera: None. N. Vazquez: None. S. Knoblock: None. O. Sugahara: None. T. Buchannan: None. L. Collins: None. D. Spector: None. L. Zhang: None. U. Danilenko: None. H.W. Vesper: None. Thyroid function tests are among the most requested tests in patient care, and reliability of these measurements are crucial for appropriate diagnosis and management of thyroid disorders. In response to concerns from the clinical and research community about the quality of thyroid function tests, the CDC Clinical Standardization Programs (CSP) has created the hormone standardization (HoSt) program for free thyroxine (FT4) in partnership with the International Federation for Clinical Chemistry and Laboratory Medicine (IFCC). As a first step, CDC CSP has completed an interlaboratory comparison study of commercial immunoassays (IA) and lab-developed tests (LDT) to determine the current agreement among assays. Participants received 40 blinded, high quality individual donor serum samples with FT4 reference values assigned by the CDC FT4 reference measurement procedure (RMP). FT4 concentrations ranged between 11-32 pmol/L. Samples were analyzed in duplicate over 2 days and results were compared to the CDC FT4 RMP. To assess the potential impact of standardizing measurements to the CDC HoSt FT4 program, reported results were normalized to the CDC RMP results (recalibrated in silico using linear regression). Mean percent bias to the reference value was evaluated pre- and post-recalibration. In addition, inter-assay agreement of classification (assignment of hypo-, eu- or hyperthyroid status according to FT4 reference intervals) were assessed pre- and post-recalibration using the assay-specific and CDC RMP reference intervals, respectively. A total of 19 methods participated, consisting of 15 IA and 4 LDT. All participants showed a negative mean bias to the RMP, with IA showing a higher magnitude of bias (average: -20.7 ± 0.9%) than LDT (average: -4.7 ± 1.1%). Of the 40 samples measured, only 53% were classified uniformly by all 19 participants. Post-recalibration, mean percent bias improved profoundly (IA average: 0.0 ± 0.8%, LDT average: 0.1 ± 1.1%), and classification agreement improved by 33% to 86% of samples with consistent classification. The CDC CSP interlaboratory comparison among FT4 assays highlights the critical need for measurement standardization to improve accuracy of FT4 measurements and increase consistency in result interpretation. This study demonstrates that it is possible to develop common FT4 reference intervals via measurement standardization, which would allow for universal evidence-based clinical guidelines to be used in patient care. Presentation: 6/3/2024
Ribera et al. (Tue,) studied this question.