It is well known that thyroid dysfunction increases with age. This study is aimed to determine reference intervals, in males and females, suitable for thyroid disease exploration during adult life using routinely collected serum thyrotropin (TSH) data in a tertiary center from 2007 to 2018. Over 11 years, 295,775 TSH levels were measured in a single lab. Among the 156,025 TSH results available for analysis, 90,538 values were from female subjects, 82,019 were from patients aged >60 years and 26,825 were from patients aged >80 years. By using an indirect approach, we determined reference values of TSH adapted to age and sex, and we then evaluated the proportion of patients who would have been reclassified with these reference values. The median TSH ranged from 1.2-1.4 mUI/L during the study period. The upper limit of reference range of TSH increased with age; in females the median to 97.5th percentile values increased continuously from the age of 30 years to the oldest age group. Using new calculated reference values in patients with TSH above the conventional upper-limit reference value (4 mUI/L), the proportion of results reclassified as within the reference interval among patients aged >60 years ranged, according to age group, from 50.5% to 65.1% of females and from 33.0% to 37.7% of males. The use of TSH age-specific and sex-specific upper-limit reference values led to the reclassification of a great number of samples, notably among women. This suggests that age-specific TSH upper-limit reference intervals in daily practice should be used in order to avoid misclassification.

Purpose: Measurement of prolactin in clinical laboratories is an important part of the management of patients with pituitary adenoma. Prolactin is known to be sensitive to the high-dose hook effect, in the presence of extremely high concentrations. This interference is mentioned in most recent articles discussing prolactin assay and management of pituitary prolactin adenomas [1][2][3]. The objective of our study was to evaluate if the mention of the high dose hook effect is relevant in actual practice.Methods: A serum from a patient with giant macroprolactinoma was assayed by using all the different reagents available in France in 2020 on native serum and after dilution. Technical inserts from manufacturers were reviewed to study information of analytical principles, numbers of steps, and mention to high dose hook effect if necessary.Results: Fourteen reagents were studied; all were two-site immunometric assays, mostly in 1 step (11/14). One tested reagent was sensitive to high dose hook effect leading to falsely low prolactin concentration when measured on native serum. Conclusion:The high-dose hook effect still exists. The evolution of reagents may lead to new reagents sensitive to this effect in the future. We therefore advise that this hook effect be mentioned in the recommendations.