A-080 Overcoming Challenges of Equilibrium Dialysis-Based Free Thyroxine Measurements

Abstract Background Thyroid function tests are widely used to detect diseases of the thyroid. Concerns about the accuracy and reliability of free thyroxine (FT4) measurements have been stated by the clinical laboratory community. To address these concerns, CDC established a reference measurement procedure (RMP) for FT4 based on equilibrium dialysis (ED). ED-based methods have advantages over traditional direct immunoassays. As with all analytical methods, factors affecting accuracy and reliability of ED-based methods need identification and understanding. To investigate potential sources of error, the CDC FT4 RMP was used to evaluate T4 stability at different of storage conditions, adsorption of FT4 to common labware surfaces, and the effect of dialysis membrane materials on the measured FT4 in serum. Methods FT4 was measured in serum using the IFCC RMP (1). FT4 was separated from the protein-bound form during the equilibrium dialysis step of the RMP. Dialysates were spiked with internal standard (13C6-T4) and the analyte was further isolated by C18 solid phase extraction (SPE) and liquid-liquid extraction (LLE) in ethyl acetate prior to injection on an UHPLC system coupled with a triple quadrupole mass spectrometer. Bracketed calibration and primary reference materials were used to determine FT4 concentration in serum. This procedure was used to evaluate stability of serum FT4 during storage at −70 to 5 °C. Short- and long-term T4 stability of calibrator solutions at −70 °C to ambient temperature and the extent of adsorption of T4 calibrator solutions to the commonly used labware surfaces were evaluated. Furthermore, the suitability of 4 different dialysis membranes was assessed. In addition, three alternatives to the currently used SPE and LLE extraction steps were tested: LLE using cyclohexane and ethyl acetate, LLE with ethyl acetate, and supported liquid extraction with cyclohexane and ethyl acetate. Results T4 calibrator solution was stable in 1.7% ammonium hydroxide in ethanol for at least 2.9 years at −70 °C and up to 7 days at ambient temperature. Exposure of various calibrator solutions such as T4 in ethanol, ethanol with 1.7% ammonium hydroxide, 10% acetonitrile with 0.1% formic acid to common labware resulted in up to 10.1% loss of the analyte. FT4 in serum and T4 in dialysates and extracts were stable at −70 to 5 °C. Use of alternative dialysis membranes resulted in up to −8.8% mean bias to the reference membranes. Recoveries of the analyte using four extraction methods were 85.4%–95.0%. None of the tested extraction conditions significantly changed the results obtained with the original FT4 RMPs. Conclusion Although FT4 RMP must strictly adhere to the ED conditions described for the RMP, other parts of the sample preparation procedures can deviate from the originally published procedure. Careful selection of procedure for calibrator solvents, dialysis membranes, labware, and storage conditions is required to ensure accurate and reproducible FT4 measurements.