Reference intervals for thyroid hormones on the architect analyser.

The objective of this study was to establish reference intervals for thyroid stimulating hormone (TSH), free thyroxine (FT4), free triiodothyronine (FT3), total thyronine (TT4) and total triiodothyronine (TT3) on the Architect i2000 analyser (Abbott). Serum samples were obtained from apparently healthy adults (n=217, age 18-90 years) excluding individuals taking oral contraceptives or under hormone replacement therapy. The second group were ambulatory euthyroid patients (n=323) excluding those with a history of thyroid disorders. We also investigated thyroid hormones in sera from euthyroid hospitalised patients (n=490) excluding those with severe non-thyroidal illness. The reference intervals for the healthy adults were for TSH 0.17-4.23 mIU/l, for FT4 11.24-26.86 pmol/l, for FT3 2.56-6.36 pmol/l, for TT4 55.8-155.1 nmol/l and for TT3 0.90-2.54 nmol/l. TSH and TT3 concentrations were similar in males and females. However, FT4, FT3 and TT4 levels exhibited significant differences between females and males. No significant differences were observed between the concentrations of TSH, FT3, TT3, FT4 and TT4 in healthy subjects and in euthyroid ambulatory patients aged 18-90 years. TSH levels in healthy subjects were the same in younger and older individuals. In contrast, in outpatients and in hospitalised patients TSH concentrations were significantly lower (20%) in subjects older than 50 years compared to those younger than 50 years. For FT3 and TT3 we consistently observed in all three study groups 6-7% and 8-12% higher concentrations in the younger (< 50 years) compared to the older (> 50 years) subjects. For FT4 and TT4 no consistent pattern of correlation with age was detectable when the three study groups were analysed independently. The reference intervals for thyroid hormones determined in this study differ considerably from values found in other European and non-European countries. This underlines the need for population-specific reference ranges.