FIGURE 1.

FT4 (A and B) and TSH (C and D) results for specimens from euthyroid subjects in the phase 1 study: percentage differences. FT4 measurements were obtained using immunoassay (15 procedures) and ED in combination with ID-LC-MS/MS (2 procedures); these were compared with the mean of FT4 results obtained using the conventional RMP, which was based on ED with ID-LC-MS/MS. TSH measurements were obtained using immunoassay methods (16 procedures); these were compared with values assigned to the individual patients’ specimens as the all-method means of TSH results. In panels A and B, the y value of each data point represents the mean of 6 results for an individual FT4 assay as a percentage of the RMP mean value, and the x value represents the RMP mean value. FT4 data are shown before (A) and after (B) mathematical recalibration using the reverse relation (Deming regression of RMP mean values on individual-assay mean values). In panels C and D, the y value of each data point represents the mean of 6 results for an individual TSH assay as a percentage of the all-method mean value, and the x value represents the all-method mean value; the dashed lines are drawn at the total error goal based on intraindividual biological variation for TSH (±23%). TSH data are shown before (C) and after (D) mathematical recalibration. ED, equilibrium dialysis; FT4, free thyroxine; ID-LC-MS/MS, isotope dilution–liquid chromatography–tandem mass spectrometry; RMP, reference measurement procedure; TSH, thyroid-stimulating hormone. Adapted from references 11 and 12 with permission.