Indirect estimation of reference intervals for thyroid parameters

The aim of this work was to determine indirect reference intervals from patients' results obtained during routine laboratory work. This could be an accurate alternative to the laborious and expensive job of producing reference intervals for populations according to international recommendations.

All the results for thyrotropin (TSH), total and free thyroxine, and triiodothyronine (T4, fT4, T3, and fT3) stored in our laboratory information system between 2008 and 2011 were included in this study. We used logarithmic transformation of the raw data to exclude outliers. After visual observation of the data distribution, we estimated non-parametric reference intervals. A standard normal deviation test was performed to test the significance of differences between subgroups.

There was no significant difference in the serum levels of the analyzed thyroid parameters, so we calculated combined reference values. However, we found a significant difference in TSH values between ambulatory and hospitalized patients, but only in 2011. Indirect reference values for TSH, T4, fT4, T3 and fT3 were 0.42 - 3.67 mIU/L, 66.0 - 136.10 nmol/L, 10.20 - 18.40 pmol/L, 1.10 - 2.39 nmol/L, and 3.17 - 5.59 pmol/L, respectively.

The indirect determination of laboratory-specific reference intervals using patients' laboratory data is a relatively easy and inexpensive method. Also, indirect reference limits will be more precise and true if skewness and kurtosis of the distribution are not too large.