Background: Serum free thyroxine (FT4) testing in pregnancy is known to be challenging for immunoassays (IAs). We verified the reliability of FT4 results by 3 commercial IAs throughout pregnancy, by comparison of the pattern to that obtained with an equilibrium dialysis isotope dilution-mass spectrometry (ED ID-MS) candidate reference measurement procedure.
Methods: Pregnant females (107) and age-matched non-pregnant controls (26) were enrolled. The IAs tested were those performed on the Cobas 6000 (Roche Diagnostics), ARCHITECT i2000SR (Abbott Diagnostics) and Immulite 2000 (Siemens Healthcare) platforms.
Results: Compared to the controls (mean FT4: 18.2pmol/L), ED ID-MS gave in the late first trimester pregnancy an 8.8% lower (p<0.05) mean; in the second trimester it was 29.1% lower (p<0.001), to stabilize in the third trimester (p=0.99). Similar observations were made for the Cobas and Immulite IAs. The ARCHITECT IA showed no significant decrease in the late first trimester (mean 13.5pmol/L versus 13.6pmol/L in controls), but a significant, less pronounced, decrease in the second and third trimesters (15% and 14.4%, respectively). All IAs were susceptible towards alterations in T4 binding proteins during pregnancy.
Conclusion: We proved that IAs may give a FT4 pattern during pregnancy similar to that obtained by ED ID-MS.
Copyright 2010 Elsevier B.V. All rights reserved.