Background-Accurate assessment of the pregnant woman's thyroid status is critical, for both the initiation of thyroid hormone therapy and for the adjustment of thyroid hormone dose in those already receiving thyroid hormone. Trimester-specific intervals are especially important during pregnancy when thyroid insufficiency may be associated with adverse obstetric outcome and fetal neurodevelopmental deficits. We defined pregnancy-specific reference intervals for thyroxine (T4) and 3,5,3′-triiodothyronine (T3). We used a novel isotope dilution tandem mass spectrometry (LC/MS/MS) method, and compare these to reference intervals obtained by immunoassays (IAs) performed on the same samples.Methods-Concentrations of circulating T4 and T3 were measured simultaneously during first, second and third trimesters and postpartum in iodine-sufficient, healthy, singleton pregnancies using API-3000 LC/MS/MS with deuterium-labeled internal standard (L-thyroxine-d 2 ). Immunoassays were conducted on the same samples (T4 Dade Behring RxL, T3 DPCImmunolite).Results-Linear regression is reported for method comparisons; for T4, the slope decreased from r=0.900 in nonpregnant women to 0.802-0.820 during pregnancy. For T3, correlations between LC/MS/MS and immunoassays were weaker in all cases (r=0.407-0.574).Conclusion-In this longitudinal study, we established trimester-specific reference intervals for T4 and T3 by LC/MS/MS and compare these to intervals obtained by immunoassays.