PRZEDZIAŁY REFERENCYJNE TSH, FT4 I FT3 W ZALEŻNOŚCI OD PŁCI I WIEKU

Zenon Jakubowski1, Monika Zagórska1, Marlena Robakowska3, Grzegorz Mincewicz1, Grzegorz Krzykowski2,
Krzysztof Sworczak
1, Mirosława Szczepańska-Konkel1, Daniel Ślęzak4, Przemysław Żuratyński4, Anna Tyrańska-Fobke5,
Jerzy Robert Ładny
6, Klaudiusz Nadolny6,7

1 Medical University of Gdansk, Gdansk, Poland

2 Institute of Informatics, University of Gdansk, Gdansk, Poland

3 Department of Public Health & Social Medicine, Faculty of Health Sciences with Institute of Maritime and Tropical Medicine, Medical University of Gdansk, Gdansk, Poland

4 Emergency Medicine Workshop, Faculty of Health Sciences with Institute of Maritime and Tropical Medicine, Medical University of Gdansk, Gdansk, Poland

5 2nd Department of Radiology, Faculty of Health Sciences with Institute of Maritime and Tropical Medicine, Medical University of Gdansk, Gdansk, Poland

6 Department of Emergency Medicine and Disasters, Medical University of Bialystok

7 Voivodship Rescue Service in Katowice, Katowice, Poland

Abstract

Introduction: Basic laboratory diagnostics of thyroid gland disorders is based mainly on the level estimation of TSH (thyroid stimulating hormone) and thyroid hormones: free thyroxine (fT4) and free triiodothyronine (fT3) in the patient’s serum. When the hypothalamic – pituitary- thyroid axis functions properly, an increase of fT4 and fT3 levels causes a decrease of TSH level in the serum. Moreover, a decrease of fT4 and fT3 levels causes an increase of THS level.

The aim: Defining precise reference ranges of the thyroid profile may condition both early diagnosis and treatment of thyroid diseases. Such procedures may also influence the diagnostics of subclinical thyroid dysfunction (subclinical hypothyroidism and subclinical hyperthyroidism) before dysfunction symptoms occur.

Material and Methods: TSH, fT4 and fT3 level values in serum were used for the purpose of the research. Only the values of the period from October 8, 2010 to October 26, 2011 were considered and derived from the Central Clinical Laboratory of the Laboratory Medicine Center at the University Clinical Center (UCML UCK) in Gdańsk. A series of exclusion criteria was applied to obtain the reference values which were later used to define reference ranges. The primary material included the following data: 34 453 TSH measurements, 24 830 fT4 measurements and 8 147 fT3 measurements. It was tested if TSH , fT4 and fT3 estimation results distribution follows Gauss distribution but the distribution normality hypothesis was rejected in all cases (Cramer-von Mises normality test; p <0.001).

Results and Conclusions: The 2.5% quantile of TSH level in the subgroup of women under the age of 40 (0.46 μIU/mL) is statistically significantly higher than the 2.5% quantile of TSH level in the subgroup of women aged over 40 years (0.41μIU/mL) (Fisher exact test: p.value<0.001). The difference between 97.5% quantiles of TSH levels in these two female groups indicates no statistical significance (Fisher exact test: p.value = 0.142). The 2.5% quantile of TSH level in the subgroup of men under the age of 40 (0.52 μIU/mL) is statistically significantly higher than the 2.5% quantile of TSH level in the subgroup of men aged over 40 years (0.42μIU/mL)  (Fisher exact test: p.value <0.001). The difference between 97.5% quantiles of TSH levels in these two male groups indicates no statistical significance (Fisher exact test: p.value = 0.381).

Streszczenie

Wstęp: Podstawowa diagnostyka laboratoryjna schorzeń gruczołu tarczycowego opiera się głównie na oznaczeniu stężenia hormonu tyreotropowego (TSH) i stężeń hormonów tarczycowych: wolnej tyroksyny (fT4) i wolnej trijodotyroniny (fT3) w surowicy pacjenta. Przy prawidłowym funkcjonowaniu osi hormonalnej podwzgórze – przysadka – tarczyca, wzrost stężenia fT4 i fT3 powoduje spadek stężenia TSH w surowicy, zaś obniżone stężenia fT4 i fT3 skutkują jego podwyższeniem

Cel pracy:Wyznaczenie precyzyjnych przedziałów referencyjnych parametrów profilu tarczycowego, może warunkować wczesne rozpoznanie i leczenie chorób tarczycy. Postępowanie to może mieć ponadto związek ze zdiagnozowaniem podklinicznych dysfunkcji tarczycy (subkliniczna niedoczynność i subkliniczna nadczynność tarczycy) jeszcze przed wystąpieniem objawów u pacjenta.

Materiał i metody: Do badania posłużyły wyniki oznaczeń stężeń TSH, fT4 i fT3 w surowicy pochodzące z przedziału czasu od  ٨ października ٢٠١٠ roku do ٢٦ października ٢٠١١ roku. Zostały one pobrane z bazy danych Centralnego Laboratorium Klinicznego Uniwersyteckiego Centrum Medycyny Laboratoryjnej Uniwersyteckiego Centrum Klinicznego (UCML UCK) w Gdańsku. Wartości referencyjne, które posłużyły w dalszej kolejności do wyznaczenia przedziałów referencyjnych uzyskano dzięki zastosowaniu szeregu kryteriów wykluczania. Materiał pierwotny stanowiły dane: ٣٤ ٤٥٣ pomiarów TSH, ٢٤ ٨٣٠ pomiarów fT4 i 8 147 pomiarów fT3. Przeprowadzono test zgodności rozkładów referencyjnych dla TSH , fT4 i fT3 z rozkładem Gaussa i hipoteza o normalności rozkładów została we wszystkich przypadkach odrzucona (Cramer-von Mises normality test; p < 0.001).

Wyniki i wnioski: Kwantyl rzędu 2.5% dla stężenia hormonu tyreotropowego w grupie kobiet poniżej 40 r.ż (0.46 μIU/ml) jest statystycznie istotnie wyższy niż kwantyl rzędu 2.5% dla hormonu tyreotropowego w grupie kobiet powyżej 40 r.ż (0.41μIU/ml) (Fisher exact test: p.value<0.001). Różnica pomiędzy kwantylami rzędu 97.5% dla stężenia TSH u porównywanych powyżej grup kobiet okazała się nie być statystycznie znamienna (Fisher exact test: p.value = 0.142).

Kwantyl rzędu 2.5% dla stężenia TSH w grupie mężczyzn poniżej 40 r.ż (0.52 μIU/ml) jest statystycznie istotnie wyższy od kwantyla rzędu 2.5% dla TSH w grupie mężczyzn powyżej 40 r.ż (0.42μIU/ml)  (Fisher exact test: p.value<0.001). Kwantyl rzędu 97.5% dla stężenia hormonu tyreotropowego u mężczyzn poniżej 40 r.ż nie różni się statystycznie istotnie od kwantyla rzędu 97.5% dla TSH w podgrupie mężczyzn powyżej 40 r.ż (Fisher exact test: p.value = 0.381).

Public Health Forum 2018;IV(XII)2(45):89-94

 

INTRODUCTION

Basic laboratory diagnostics of thyroid gland disorders is based mainly on the level estimation of TSH (thyroid stimulating hormone) and thyroid hormones: free thyroxine (fT4) and free triiodothyronine (fT3) in the patient’s serum. When the hypothalamic – pituitary- thyroid axis functions properly, an increase of fT4 and fT3 levels causes a decrease of TSH level in the serum. Moreover, a decrease of fT4 and fT3 levels causes an increase of THS level [1]. Owing to the above described negative feedback mechanism, it is possible to diagnose endocrine thyroid dysfunction and identify its clinical presentation (hyperthyroidism or hypothyroidism). It is worth noting that country- and region specific reference ranges of TSH and other thyroid hormones are crucial in diagnostics of thyroid dysfunctions. The source of area-specific differentiation of reference range values is the varying level of iodine in water, plant and animal products. Iodine is an element that is crucial in the synthesis of thyroid hormones. Coastal areas are regions with the highest iodine levels. The longer the distance from the sea, the lower the level of iodine in soil [1, 8]. As it has been noted in other scientific research, in order to identify TSH, fT4 and fT3 reference range values, iodine level in food and water needs to be considered [10−12]. According to the guidelines of the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC ), each laboratory should define its own internal reference ranges [7]. It is especially important for doctors who are not endocrinologists. Defining precise reference ranges of the thyroid profile may condition both early diagnosis and treatment of thyroid diseases. Such procedures may also influence the diagnostics of subclinical thyroid dysfunction (subclinical hypothyroidism and subclinical hyperthyroidism) before dysfunction symptoms occur.

MATERIAL AND METHODS

Material

TSH, fT4 and fT3 level values in serum were used for the purpose of the research. Only the values of the period from October 8, 2010 to October 26, 2011 were considered and derived from the Central Clinical Laboratory of the Laboratory Medicine Center at the University Clinical Center (UCML UCK) in Gdańsk. A series of exclusion criteria was applied to obtain the reference values which were later used to define reference ranges. The following exclusion criteria were considered:

1. PESEL was missing (i.e. the unique 11-digit personal number that identifies Polish citizens and confirms their e.g. date of birth and gender);

2. the pre-analytical phase of the diagnostic process was performed outside University Clinical Center;

3. level values were provided by patients of endocrinology and maternity clinics and medical practices, as well as from intensive care units;

4. patients participated in the multi-center public health research called NATPOL PLUS [5], 5. level values were beyond the valid reference ranges [6];

6. the same parameter was tested twice or more frequently within the analyzed period;

7. there was a positive test result for thyroglobulin, thyroid peroxidase and thyrotropin hormone receptor antibodies.

Ethics

Consent was obtained from the Local Bioethics Committee of the Medical University of Gdańsk to conduct the research in question. Routine laboratory parameter results from a significant number of patients were used and, therefore, individual patient’s consent was not required.

Laboratory estimation methods

TSH, fT4 and fT3 levels in serum were estimated by means of the CMIA method (Chemiluminescent Microparticle Immunoassay). The reference ranges suggested by the reagent kit producer (Abbott Laboratories Polska) are the following: 0.35-4.94 μIU/mL for TSH, 9.01-19.05 pmol/L for fT4 and 2.63-5.7 pmol/L for fT3. To obtain these results, we used quantitative estimation results (AxSYM Ultrasensitive hTSH II, AxSYM Free T4 and AxSYM Free T3) of the thyroid profile parameters in serum of healthy individuals: a group of 549 individuals for TSH (99% confidence interval), another group of 411 individuals for fT4 (99% confidence interval) and a group of 436 individuals for fT3 (95% confidence interval) [2].

Statistical analysis

In order to test whether estimation results distribution of the analyzed parameters follows Gauss distribution, the Cramer-von Mises normality test was applied. Mean, median and quantile were calculated for the groups under research. Reference range end for all parameters were defined in subgroups of the reference group by means of the non-parametric method. The reference group was divided according to gender and age. The low end of the range was indicated by the percentile with 2.5% of results below it. The high end of the range was marked by the percentile with 97.5% results below it. Quantile testing in gender and age subgroups was performed using the Fisher’s Exact Test for the total number of exceedances of tested quantile values. The statistical analysis was conducted by means of the R 2.15.0 statistical computing package. The primary material included the following data: 34 453 TSH measurements, 24 830 fT4 measurements and 8 147 fT3 measurements. The analysis started with the application of a series of selection criteria. The result of this selection was a set of 9 726 patients’ thyroid profiles.

RESULTS

The group under research with TSH level estimation in serum consisted of 8 252 individuals. The reference group included subgroups based on a division according to gender and age. The numbers of individuals in the particular subgroups are as follows: 1 821 female patients below the age of 40, 3 726 female individuals aged over 40 years, 636 male patients below the age of 40, and 2 069 male individuals aged over 40 years.

Free Thyroxine (fT4) level estimation in serum was performed in 4 916 individuals. There were 1 268 women below the age of 40 and 2 297 women aged over 40 years, as well 294 men below the age of 40 and 1057 male aged over 40 years.

Free Triiodothyronine (fT3) level estimation was performed in 3 312 patients. There were 651women below the age of 40 and 1 686 aged over 40 years, as well as 195 men below the age of 40 and 780 aged over 40 years.

It was tested if TSH , fT4 and fT3 estimation results distribution follows Gauss distribution but the distribution normality hypothesis was rejected in all cases (Cramer-von Mises normality test; p < 0.001). As a consequence, the reference ranges were estimated by means of the non-parametric method.

For reference ranges according to gender and age as well as descriptive statistics of TSH, fT4 and fT3 levels in serum, please consult tables 1−3.

Quantile testing in subgroups divided according to gender and age was performed by means of the Fisher’s Exact Test.

The 2.5% quantile of TSH level in the subgroup of women under the age of 40 (0.46 μIU/mL) is statistically significantly higher than the 2.5% quantile of TSH level in the subgroup of women aged over 40 years (0.41μIU/mL) (Fisher exact test: p.value<0.001). The difference between 97.5% quantiles of TSH levels in these two female groups indicates no statistical significance (Fisher exact test: p.value = 0.142).

The 2.5% quantile of TSH level in the subgroup of men under the age of 40 (0.52 μIU/mL) is statistically significantly higher than the 2.5% quantile of TSH level in the subgroup of men aged over 40 years (0.42μIU/mL)  (Fisher exact test: p.value<0.001). The difference between 97.5% quantiles of TSH levels in these two male groups indicates no statistical significance (Fisher exact test: p.value = 0.381).

No statistically significant differences have been observed between the Q2.5 quantile of TSH levels in women under the age of 40 and the Q2.5 quantile of TSH levels in men under the age of 40 (Fisher exact test: p.value = 0.053). Similarly, there is no statistically significant difference between the Q97.5 quantiles in the subgroup of women under the age of 40 and the subgroup of men under the age of 40 (Fisher exact test: p.value=0.767).

There is no statistically significant difference between the Q2.5 quantile of TSH level in serum in the subgroup of women aged over 40 years and the Q2.5 quantile of TSH level in the subgroup of men aged over 40 years (Fisher exact test: p.value=0.422). However, a statistically significant difference has been observed between Q97.5 quantiles of TSH levels in the subgroup of men and the subgroup of women both aged over 40 years (Fisher exact test: p.value < 0.002). The 97.5% quantile of TSH level in the subgroup of women aged over 40 years (3.89 μIU/mL) is statistically significantly higher than the 97.5% quantile of TSH level in the subgroup of men aged over 40 years (3.44μIU/mL).

As far as Free Thyroxine (fT4) levels in serum are concerned, there is no statistically significant difference between the 2.5% quantile in the subgroup of women under the age of 40 and the 2.5% quantile in the subgroup of women aged over 40 years (Fisher exact test: p.value = 0.91). The difference between 97.5% quantiles of fT4 levels in these two female groups indicates no statistical significance (Fisher exact test: p.value = 0.147).

There is no statistically significant difference between the Q2.5 quantile of fT4 level in serum in the subgroup of men under the age of 40 and the Q2.5 quantile of fT4 level in the subgroup of men aged over 40 years (Fisher exact test: p.value = 0.132).The difference between 97.5% quantiles of fT4 levels in these two male groups also indicates no statistical significance (Fisher exact test: p.value=0.521).

The 2.5% quantile of fT4 level in the subgroup of women under the age of 40 (11.34 pmol/L) is statistically significantly lower than the 2.5% quantile of fT4 level in the subgroup of men under the age of 40 (10.46 pmol/L) (Fisher exact test: p value <0.003). The difference between the 97.5% quantiles of fT4 level in the group of men and the group of women indicates no statistical significance (Fisher exact test: p value =0.411).

There is a statistically significant difference between the 2.5% quantile of fT4 level in women aged over 40 years (11.29 pmol/L) and the 2.5% quantile of fT4 level in men aged over 40 years (10.90 pmol/L) (Fisher exact test: p.value<0.012). However, there is no statistically significant difference between the 97.5% quantile of fT4 level in women aged over 40 years and the 97.5% quantile of fT4 level in men aged over 40 years.

As far as Free Triiodothyronine (fT3) is concerned, 2.5% quantile in women under the age of 40 (3.26 pmol/L) is statistically significantly higher than the 2.5% quantile in women aged over 40 years (2.88 pmol/L) (Fisher exact test: p.value <0.001). The 97.5% quantile of fT3 level is also statistically significantly higher in women under the age of 40 (5.49 pmol/L) than the 97.5% quantile of fT3 level in women aged over 40 years (5.35 pmol/L) (Fisher exact test: p value <0.029).

There is a statistically significant difference between the Q2.5 quantiles of fT3 level in the two male groups: one consisting of patients under the age of 40; the other group consisting of patients aged over 40 years (Fisher exact test: p.value < 0.008). The 2.5% quantile of fT3 level in the group of men under the age of 40 (3.36 pmol/L) is statistically significantly higher than the 2.5% quantile of fT3 level in the group consisting of patients aged over 40 years (2.82 pmol/L). A statistically significant difference has also been observed while testing 97.5 % quantiles. The 97.5 % quantile of fT3 level in the group of men under the age of 40 (5.61 pmol/L) is statistically significantly higher than the 97.5 % quantile of fT3 level in the group of men aged over 40 years (5.29 pmol/L) (Fisher exact test: p value <0.001).

There is no statistically significant difference between the 2.5% quantile of fT3 level in the group of women under the age of 40 and the 2.5% quantile of fT3 level in the group of men under the age of 40 (Fisher exact test: p.value= 0.797). There is a statistically significant difference between the 97.5% quantiles of fT3 level in these two groups, one consisting of men and the other one − of women (Fisher exact test: p.value < 0.037). The 97.5% quantile of fT3 level in the male group under the age of 40 (5.61 pmol/L) is statistically significantly higher than the 97.5% quantile of fT3 level in the female group under the age of 40 (5.49 pmol/L).

There is no statistically significant difference between the 2.5% quantile of fT3 level in the subgroup of women aged over 40 years and the 2.5% quantile of fT3 level in the subgroup of men aged over 40 years (Fisher exact test: p value =0.329). The statistically significant difference is also absent in the case of 97.5% quantiles of fT3 level in the two groups of men aged over 40 years and women aged over 40 years (Fisher exact test: p value =0.58).

DISCUSSION

The main goal of this article has been to identify the inter-laboratory reference ranges of basic thyroid profile parameters that take into account differences depending on the population under research. Precise exclusion criteria were applied to select reference individuals based on the archive patients database of the University Clinical Center. The results of TSH, fT4 and fT3 laboratory diagnostics that met the exclusion criteria served as reference values. According to the IFCC procedures regarding obtaining reference values [12], the reference ranges were identified by means of the non-parametric method because the collection of laboratory diagnostics result did not follow normal distribution.

In order to analyze the influence of gender and age on the values of reference range ends, different age groups were analyzed for both men and women. This means that there were always two female groups (under the age of 40 and over 40 years) and two male groups (under the age of 40 and over 40 years) for each parameter under research. The significance of differences between the particular groups was tested by means of the Fisher’s Exact Test. In both men and women, Q0.25 quantile levels of TSH and fT3 are statistically significantly lower in groups aged over 40 years. The Q97.5 quantile of TSH level does not follow this pattern but the Q97.5 quantile of fT3 level does. This means that the reference range end values of fT3 levels become lower with age.

There are no statistically significant differences between the 97.5% quantiles of TSH level in both age groups under research in both men and women. To add to this, there is a statistically significant relationship between the age of men and women and the low range end value for TSH. Considering these two observations, it can be noticed that reference ranges become wider and the values of their low ends tend to be lower. It has also been observed that the upper reference range end for TSH in women aged over 40 years is statistically significantly higher than the upper reference range end in men of the same age group. One can conclude that upper reference range end values for women aged over 40 years should be higher than the values for men aged over 40 years.

As far as fT4 levels are concerned, there are no statistically significant differences between the Q2.5 and the Q97.5 quantiles for women under the age of 40 and aged over 40 years, as well as for men under the age of 40 and aged over 40 years. The conclusion here is that reference range end values of fT4 levels are independent of age. However, a relationship between reference range end values of fT4 levels and gender has been observed. Low reference range end values of fT4 levels in both age subgroups are statistically significantly higher in women than in men. It can be stated that fT4 level reference ranges in women move towards higher level values.

The observations have also pointed out to a relationship between gender and fT3 levels in serum. The Q97.5 quantile is statistically significantly higher in men under the age of 40 than in women of the same age group. The conclusion is that the upper reference range end in patients under the age of 40 depends on gender and is statistically significantly higher in men.

TSH, fT4 and fT3 level reference ranges identified depending on gender and age of patients present a more detailed description of the population under research. These reference ranges support clinicians of all departments in providing a proper reference system for laboratory diagnostic results. In this way, such reference ranges may to some extent support the differentiation of thyroid dysfunctions. Current universal reference ranges refer to the complete population and do not consider the age and gender of diagnosed patients [13].

This research involved the method of retrospective analysis of laboratory diagnostic results to identify particular reference ranges. However, because of the limits of this method as described in the first publication, the method should be verified by means of the ROC analysis.

References

1. Dembińska-Kieć A, Naskalski JW. Diagnostyka laboratoryjna z elementami biochemii klinicznej. Wrocław: Urban & Partner, 2010.

2. Standardowa procedura operacyjna oznaczania stężenia TSH/FT4/FT3 w surowicy Abbott Laboratories: ref. 7K62 49-4654/R03 B7K62P ref. 7K65 49-0272/R3B7K65P

3. R Development Core Team (2012). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org/.

4. Juergen Gross and bug fixes by Uwe Ligges (2012). nortest: Tests for Normality. R package version 1.0-2. http://CRAN.R-project.org/package=nortest

5. Zdrojewski T, Wyrzykowski B, Szczech R, Wierucki L, Naruszewicz M, Narkiewicz K, Zarzeczna-Baran M; Steering Committees of the Programmes NATPOL PLUS; SMS; Polish 400-Cities Project. Epidemiology and prevention of arterial hypertension in Poland. Blood Press Suppl. 200;2:10-16.

6. Standardowa procedura operacyjna oznaczania stężenia TSH/FT4/FT3 w surowicy Abbott Laboratories: ref. 7K62 49-4654/R03 B7K62P ref. 7K65 49-0272/R3B7K65P

7. Solberg HE. International Federation of Clinical Chemistry (IFCC), Scientific Committee, Clinical Section, Expert Panel on Theory of Reference Values, and International Committee for Standardization in Haematology (ICSH), Standing Committee on Reference Values. Approved recommendation on the theory of reference values. Part 1-4. The concept of reference values. J Clin Chem Clin Biochem 1987;25(5):337-342.

8. Ziemlański S, Bułhak-Jachymczyk B, Niedźwiecka-Kącik D, Panczewska−Kresowska B, Wartanowicz M. Normy Żywienia Człowieka. Warszawa: PZWL, 2001.

9. Knudsen N, Bülow I, Jorgensen T, Laurberg P, Ovesen L, Perrild H. Comparative study of thyroid function and types of thyroid dysfunction in two areas in Denmark with slightly different iodine status. Eur J Endocrinol 2000;143:485-491.

10. Bulow I, Knudsen N, Jørgensen T, Perrild H, Ovesen L, Laurberg P. Large differences in incidences of overt hyper- and hypothyroidism associated with a small difference in iodine intake: a prospective comparative register-based population survey. J Clin Endocrinol Metabol 2002;87:4462-4469

11. Guan H, Shan Z, Teng X, Li Y et al.Influence of iodine on the reference interval of TSH and the optimal interval of TSH: results of a follow-up study in areas with different iodine intakes. Clin Endocrinol. 2008 ;69(1):136-141.

12. Solberg H. Establishment and use of reference values. In: Burtis C, Ashwood E, Bruns D (eds). Tietz Textbook of Clinical Chemistry and Molecular Diagnostics, 6 edn. St. Louis: Saunders, 2006-2016.

13. Surks M, Boucai L. Age- and race-based serum thyrotropin reference limits. J Clin Endocrinol Metab 2010;95:496-502,

14. Boucai L, Surks M. Reference limits of serum TSH and free T4 are significantly influenced by race and age in an urban outpatient medical practice. Clin Endocrinol. 2009;70:788-793.

15. Boucai L, Hollowell J, Surks M. An approach for development of age-, gender- and ethnicity-specific thyrotropin reference limits. Thyroid. 2011;21(1):5-11.

Conflict of interest

All authors state that there is no conflict of interest.

Corresponding author

Zenon Jakubowski

Medical University of Gdansk, Faculty of Health Sciences,

Department of Public Health & Social Medicine,

al. Zwycięstwa 42A, 80-210 Gdańsk

tel: + 58 349 20 45

Received: 29.06.2018

Accepted: 20.07.2018

Table 1. Descriptive statistics of TSH levels (μIU/mL) in serum in the subgroups under research according to gender and age.

Gender

Age (years)

N

Mean

Median

2.5th centile

97.5th centile

female and male

≤40

2 457

1,58

1,39

0,48

3,73

 

>40

5 795

1,44

1,22

0,41

3,78

female

all

5 547

1,50

1,30

0,43

3,83

≤40

1 821

1,58

1,40

0,46a

3,72

 

>40

3 726

1,47

1,24

0,41a

3,89c

male

all

2 705

1,43

1,23

0,43

3,50

≤40

636

1,57

1,37

0,52b

3,75

 

>40

2 069

1,38

1,19

0,42b

3,44c

a, b, c – statistically significantly different values that amount to p <0.001

Table 2. Descriptive statistics of fT4 levels(pmol/L) in serum in the subgroups under research according to gender and age.

Gender

Age (years)

N

Mean

Median

2.5th centile

97.5th centile

female and male

≤40

1 562

14,50

14,42

11,21

18,41

 

>40

3 354

14,84

14,78

11,13

18,62

female

all

3 565

14,83

14,75

11,29

18,63

≤40

1 268

14,59

14,52

11,34a

18,43

 

>40

2 297

14,96

14,93

11,29b

18,66

male

all

1 351

14,47

14,39

10,74

18,35

≤40

294

14,10

14,04

10,46a

17,71

 

>40

1 057

14,58

14,50

10,90b

18,37

a, b – statistically significantly different values that amount to p <0.001

Table 3. Descriptive statistics of fT3 levels (pmol/L) in the subgroups under research according to gender and age.

Gender

Age (years)

N

Mean

Median

2.5th centile

97.5th centile

female and male

≤40

846

4,48

4,49

3,27

5,52

 

>40

2 466

4,17

4,20

2,86

5,34

female

all

2 337

4,26

4,28

2,95

5,40

≤40

651

4,44

4,45

3,26a

5,49b, e

 

>40

1 686

4,19

4,22

2,88a

5,35b

male

all

975

4,23

4,27

2,87

5,43

≤40

195

4,60

4,65

3,36c

5,61d, e

 

>40

780

4,14

4,17

2,82c

5,29d

a, b, c, d , e – statistically significantly different values that amount to p <0.001