Establishment of Early Pregnancy Related Thyroid Hormone Models and Reference Intervals for Pregnant Women in China Based on Real World Data

Abstract

Thyroid hormone reference intervals are crucial for diagnosing and monitoring thyroid dysfunction during early pregnancy, and the dynamic change trend of thyroid hormones during pregnancy can assist clinicians to assess the thyroid function of pregnant women. This study aims to establish early pregnancy related thyroid hormones models and reference intervals for pregnant women. We established two derived databases: derived database* and derived database#. Reference individuals in database* were used to establish gestational age-specific reference intervals for thyroid hormones and early pregnancy related thyroid hormones models for pregnant women. Individuals in database# were apparently healthy non-pregnant women. The thyroid hormones levels of individuals in database# were compared with that of individuals in database* using nonparametric methods and the comparative confidence interval method. The differences in thyroid stimulating hormone and free thyroxine between early pregnant and non-pregnant women were statistically significant (p<0.0001). The reference intervals of thyroid stimulating hormone, free thyroxine and free triiodothyronine for early pregnant women were 0.052–3.393 μIU/ml, 1.01–1.54 ng/dl, and 2.51–3.66 pg/ml, respectively. Results concerning thyroid stimulating hormone and free thyroxine reference intervals of early pregnancy are comparable with those from other studies using the same detection platform. Early pregnancy related thyroid hormones models showed various change patterns with gestational age for thyroid hormones. Early pregnancy related thyroid hormones models and reference intervals for pregnant women were established, so as to provide accurate and reliable reference basis for the diagnosing and monitoring of maternal thyroid disfunction in early pregnancy.

Establishment of discrete reference interval and next-generation reference interval for copper and zinc during pregnancy using real-world data

“…Conversely, next-generation RIs, or continuous RIs, enable a more detailed and more sensitive depiction of changes in analytes than discrete RIs, particularly age-dependent changes. Additionally, next-generation RIs have been widely used in recent years and are highly advantageous for various analytes [ [7] , [8] , [9] , [10] , [11] ] [ [7] , [8] , [9] , [10] , [11] ]. Among algorithms for estimating next-generation RIs, Generalized Additive Models for Location, Scale, and Shape (GAMLSS) employs parameters of the distribution (location, scale, skewness and kurtosis) to develop models and then estimate next-generation RIs [ 12 ].…”