ESTABLISHMENT OF LOCAL POPULATION-BASED REFERENCE INTERVALS FOR THYROID FUNCTION TESTS IN PREGNANCY
Journal
Physioacta
Date Issued
2019
Author(s)
Abstract
Thyroid stimulating hormone (TSH) and free thyroxin (FT4) are first-line thyroid function tests (TFTs) for detecting thyroid abnormalities in general population and pregnant women (PW) as well. Our objective was to validate the proposed manufacturer’s reference range in pregnancy for TSH and FT4 and to obtain reliable reference intervals for our PW.
The reference population was selected from PW without personal or family history of thyroid disorders. For the validation of the manufacturer’s reference intervals we followed the 20-sample CLSI protocol and the large-sample protocol including 267 pregnant women from all trimesters. TFTs were assessed by IMMULITE 2000 chemiluminescence immunoassay system. Trimester values were expressed as median. To compare the trimester differences, Mann-Whitney U test was used. Reference intervals (RIs) were obtained by nonparametric method.
Following the CLSI protocol, we could not accept manufacturer’s RI for TSH, while our median was higher in comparison with manufacturer’s (1,9 mIU/L vs 1,1 mIU/L). For FT4 our findings indicated significant differences between the first and second / third trimester and lower medians (11.6 pmol/L vs 16.1 pmol/L and 10.3 pmol/L vs 13.9 pmol/L). Therefore, we established own RIs according to the large sample protocol. The median for TSH was 1,6 mIU/L and the new RI was 0,34– 3,4 mIU/L. For FT4, first trimester RI was 7.9-16.0 pmol/L, and second/third RI was 6.4-14.6 pmol/L.
These specific RI should be considered by clinicians for an accurate interpretation of TFTs in pregnancy and adequate patient treatment.
The reference population was selected from PW without personal or family history of thyroid disorders. For the validation of the manufacturer’s reference intervals we followed the 20-sample CLSI protocol and the large-sample protocol including 267 pregnant women from all trimesters. TFTs were assessed by IMMULITE 2000 chemiluminescence immunoassay system. Trimester values were expressed as median. To compare the trimester differences, Mann-Whitney U test was used. Reference intervals (RIs) were obtained by nonparametric method.
Following the CLSI protocol, we could not accept manufacturer’s RI for TSH, while our median was higher in comparison with manufacturer’s (1,9 mIU/L vs 1,1 mIU/L). For FT4 our findings indicated significant differences between the first and second / third trimester and lower medians (11.6 pmol/L vs 16.1 pmol/L and 10.3 pmol/L vs 13.9 pmol/L). Therefore, we established own RIs according to the large sample protocol. The median for TSH was 1,6 mIU/L and the new RI was 0,34– 3,4 mIU/L. For FT4, first trimester RI was 7.9-16.0 pmol/L, and second/third RI was 6.4-14.6 pmol/L.
These specific RI should be considered by clinicians for an accurate interpretation of TFTs in pregnancy and adequate patient treatment.