Download Title page Title: Thyroid-stimulating hormone reference range and

1
Title page
2
3
Title: Thyroid-stimulating hormone reference range and factors affecting it in a nationwide
4
random sample
5
Authors’ names and full addresses:
6
Ville L. Langén1,2, Teemu J. Niiranen1, Juhani Mäki1, Jouko Sundvall3, Antti M. Jula1
7
1. Population Studies Unit, National Institute for Health and Welfare, Turku, Finland.
8
2. Operational Division of Medicine, Turku University Hospital, Turku, Finland.
9
3. Disease Risk Unit, National Institute for Health and Welfare, Helsinki, Finland.
10
Corresponding author’s postal and email address:
11
Address:
12
Ville Lauri Johannes Langén, M.D.
13
Population Studies Unit, National Institute for Health and Welfare
14
Peltolantie 3, 20720 Turku
15
Finland
16
Telephone: +358 29 524 6000, E-mail: [email protected]
17
Short title: TSH reference range and factors affecting it
18
A minimum of four keywords describing the manuscript: thyroid-stimulating hormone
19
(TSH); thyroid peroxidase antibody (TPOAb); reference range; reference interval; adult
20
population; thyroid dysfunction
21
The word count excluding references, captions and Tables: 3132 / Word count in
22
Abstract: 250 / no. of Keywords: 6 / no. of Tables: 2 / no. of Figures: 2 / References:
23
30 / Supplemental data: 1 table (Supplemental data, Table 1).
24
The Figure titles and legends have been placed in the end of this document.
25
1
26
Abstract
27
28
Objectives: Previous studies with mainly selected populations have proposed contradicting
29
reference ranges for TSH and have disagreed on how screening, age and gender affect
30
them. This study aimed to determine a TSH reference range on the Abbott Architect
31
ci8200 integrated system in a large, nationwide, stratified random sample. To our
32
knowledge this is the only study apart from the NHANES III that has addressed this issue
33
in a similar nationwide setting. The effects of age, gender, TPOAb-positivity and
34
medications on TSH reference range were also assessed.
35
36
Methods: TSH was measured from 6247 participants randomly drawn from the population
37
register to represent the Finnish adult population. TSH reference ranges were established
38
of a thyroid-healthy population and its subpopulations with increasing and cumulative
39
rigour of screening: screening for overt thyroid disease (thyroid-healthy population,
40
n=5709); screening for TPOAb-positivity (risk factor-free subpopulation, n=4586); and
41
screening for use of any medications (reference subpopulation, n=1849).
42
43
Results: The TSH reference ranges of the thyroid-healthy population, and the risk factor-
44
free and reference subpopulations were 0.4 – 4.4, 0.4 – 3.7 and 0.4 – 3.4 mU/L (2.5th –
45
97.5th percentiles), respectively. Although the differences in TSH between subgroups for
46
age (P=0.002) and gender (P=0.005) reached statistical significance, the TSH distribution
47
curves of the subgroups were practically superimposed.
48
49
Conclusions: We propose 0.4 – 3.4 mU/L as a TSH reference range for adults for this
50
platform, which is lower than those presently used in most laboratories. Our findings
2
51
suggest that intensive screening for thyroid risk factors, especially for TPOAb-positivity,
52
decreases the TSH upper reference limit.
53
3
54
Introduction
55
56
The biochemical definition of thyroid dysfunctions relies on the reference ranges of thyroid-
57
stimulating hormone (TSH) and thyroid hormones – thyroxine and triiodothyronine. The
58
definition of the TSH reference range is therefore of paramount importance especially for
59
diagnosing subclinical thyroid disorders.
60
61
In 2002 the National Academy of Clinical Biochemistry (NACB) published a
62
comprehensive guideline for diagnosing and monitoring of thyroid disease (1). The NACB
63
guideline instructed that TSH reference ranges should be established of rigorously
64
screened euthyroid volunteers and held it likely that by this approach the upper limit of
65
TSH reference range would be over time reduced from the current 4.0 – 4.5 mU/L to 2.5
66
mU/L. NACB considered that the current TSH reference range has been based on
67
populations containing individuals with occult mild hypothyroidism that even the sensitive
68
thyroid antibody immunoassays cannot always detect and that the failure to exclude these
69
volunteers has resulted in the skewed upper limit of TSH distribution. In contrast, Surks et
70
al. have suggested (2,3) that this skew could be reduced by dividing the TSH distribution
71
to age-specific curves as TSH has been shown to increase with age in several (3-7),
72
though not in all (8-10), studies. In addition to age-specificity, it has also been suggested
73
that TSH reference limits should be race- (2) and assay-specific (11).
74
75
In a 2005 German study that adhered very strictly to the NACB criteria – and even used
76
ultrasonographic assessment to exclude occult thyroid disease – the upper limit of TSH
77
reference range was comparable between the whole population and the rigorously
78
screened population (3.63 mU/L versus 3.77 mU/L) (12). Another German study,
4
79
conducted in a region of mild iodine deficiency, provided the same conclusion (13). This
80
has raised the question of whether or not a stringent screening is needed to select a
81
reference population for establishing a TSH reference range (14). However, some studies
82
have shown clear differences between the upper limits of the unscreened and screened
83
populations, especially when participants with thyroid peroxidase antibodies (TPOAb) are
84
excluded (4,8).
85
86
The debate continues on how and where to set the upper limit of the TSH reference range.
87
Previous studies with mainly selected populations have proposed contradicting reference
88
ranges and have disagreed on how population screening, age and gender affect them. Our
89
study aimed to determine a TSH reference range in a large, nationwide, stratified random
90
adult sample on the Abbott Architect ci8200 integrated system. We also assessed the
91
effect of age, gender, TPOAb-positivity and medications on TSH reference range. To our
92
knowledge, apart from the NHANES III (6), this is the only study that has addressed these
93
issues in a similar setting.
94
5
95
Participants and methods
96
97
Participants
98
This study is part of the Health 2000 Survey, which was a multi-disciplinary
99
epidemiological survey carried out in Finland from September 2000 to July 2001. A
100
nationwide stratified 2-stage cluster sample of 8028 persons was drawn randomly from the
101
national population register to represent the Finnish adult population aged 30 years and
102
older. A more detailed report on the sampling has been published previously (15). 6771
103
(84.3%) agreed to participate in a health interview and in a health examination. A blood
104
sample was available for TSH testing from 6247 (77.8%) participants.
105
106
Health interview and health examination
107
Interviews for information on health, illnesses, medications and functional capacity were
108
conducted at the participants’ homes by centrally trained interviewers. The participants
109
were examined one to four weeks thereafter by centrally trained physicians and nurses at
110
a local facility. In addition, blood samples were drawn from the participants.
111
112
Laboratory analyses
113
Plasma TSH was measured from all 6247 blood samples. Additionally, plasma TPOAb
114
was measured from all participants with even slightly abnormal TSH values (TSH < 0.4
115
mU/l or > 2.5 mU/l) . TPOAb was also measured from 480 (9.2%) randomly selected
116
participants with TSH between 0.4 and 2.5 mU/L (n=5221). All samples were stored in -
117
70°C and later analysed with an Abbot Architect ci8200 Analyzer (Abbott Laboratories, IL,
118
USA). During the course of measurements the between-batch coefficient of variation in
119
control samples was less than 4.0% for TSH and 3.6% for TPOAb. An ad hoc TPOAb
6
120
reference range for this study could not be determined because our study population did
121
not include participants who could meet the stringent criteria proposed by the NACB (1) for
122
establishing a TPOAb reference range (men under the age of 30, with TSH between 0.5
123
and 2.0 mU/L and without history of thyroid disease, goiter or auto-immune disease).
124
Thus, we opted to consider TPOAb concentrations ≥ 5.6 IU/mL positive (abnormal)
125
according to the reference limit provided by Abbott Laboratories.
126
127
Selection of study population and subpopulations
128
The TSH reference ranges were defined in the thyroid-healthy population and its
129
subpopulations with increasing rigour of screening for thyroid disease and factors affecting
130
thyroid function (Fig. 1).
131
132
1) Thyroid-healthy population: We strived to follow the NACB guideline for defining TSH
133
reference ranges (1), although information on family history of thyroid disease and
134
thyroglobulin antibody (TgAb) concentrations were not available. Out of the 6247
135
participants with an available TSH value, we excluded participants who were not
136
ambulatory (n=8), had a history of thyroid disease or goitre (n=365), had their blood
137
sample drawn before 8 AM or after 6 PM (n=90) or were using thyroid hormones or
138
antithyroid agents (n=245). In addition, pregnant or breast-feeding (n=55) women were
139
excluded as it has been demonstrated that pregnancy and the early puerperium period are
140
associated with alterations in TSH values (16,17). After excluding 15 participants with
141
extreme TSH values (± 5 SD from mean, > 12.1 mU/l) or with one or more exclusion
142
factors, the thyroid-healthy population consisted of 5709 participants.
143
7
144
2) Risk factor-free subpopulation (subset of the thyroid-healthy population): In accordance
145
with the NACB guideline, we excluded participants who had a positive TPOAb result
146
(n=364) from the thyroid-healthy population. Because TPOAb values were not available for
147
all participants who had a TSH value between 0.4 – 2.5 mU/L, we excluded 759 randomly
148
selected participants with a TSH value between 0.4 – 2.5 mU/L to prevent oversampling.
149
Thus, the total share of excluded participants within the TSH range 0.4 – 2.5 mU/L was
150
exactly equal to the true prevalence of TPOAb-positivity in it (16.3%), determined from the
151
available data. Finally, this subpopulation consisted of 4586 participants. We also formed a
152
subgroup of this subpopulation from which we excluded participants with use of
153
medications that have a potential effect on thyroid function tests. These medications have
154
been described in a previous review (18) and are listed online (see Supplemental data,
155
Table 1). This subgroup consisted of 3453 participants.
156
157
3) Reference subpopulation (subset of the risk factor-free subpopulation): We excluded
158
participants with any medications (n=2737) from the risk factor-free subpopulation. As
159
opposed to the NACB guidelines, we excluded even users of estrogen, as some estrogen
160
medications seem to affect TSH levels (19). The final reference subpopulation consisted of
161
1849 participants.
162
163
Ethics
164
The Health 2000 Survey protocol was approved by the Epidemiology Ethics Committee of
165
the Helsinki and Uusimaa hospital region and all the participants signed informed consent
166
according to the Declaration of Helsinki.
167
8
168
Statistical analyses
169
The TSH distribution was neither normal nor log-normal in the reference subpopulation.
170
We therefore established TSH reference ranges directly from the 2.5th and 97.5th
171
percentiles of the TSH measurements in the thyroid-healthy population and all of its
172
subsets, as has been done in several other notable studies (6,8,12,20). We controlled the
173
validity of this nonparametric method by defining the TSH reference range for the
174
reference subpopulation also from data on the 95% confidence limits of its TSH values that
175
were transformed by using a best suitable function to obtain a Gaussian distribution.
176
Differences in TSH between the subgroups for age and gender were compared using the
177
Kruskal-Wallis test. P < 0.05 was considered significant. Statistical analyses were
178
performed with SAS software (SAS Institute, Cary, NC), version 9.3.
179
180
181
182
9
183
Results
184
185
The characteristics of the thyroid-healthy population and its subpopulations are reported in
186
Table 1 and the TSH medians and reference ranges are summarised correspondingly in
187
Table 2.
188
189
Effect of TPOAb-positivity on TSH reference range
190
The exclusion of TPOAb-positive participants from the thyroid-healthy population resulted
191
in a marked reduction in the TSH reference range upper limit: the TSH upper limits for the
192
thyroid-healthy population and the risk factor-free subpopulation were 4.43 and 3.71 mU/L,
193
respectively (Table 2). The TSH lower limit did not change, as it was 0.41 mU/L for both
194
the thyroid-healthy population and the risk factor-free subpopulation.
195
196
Effect of medications on TSH reference range
197
The exclusion of participants with use of 1) thyroid function affecting and 2) any
198
medications lowered the TSH reference upper limit by 0.1 and 0.3 mU/L, respectively.
199
These exclusions had virtually no effect on the TSH lower limit (Table 2). The TSH
200
reference range for the reference subpopulation without any medications was 0.43 – 3.37
201
mU/L.
202
203
Effect of age and gender on TSH reference range
204
The TSH reference ranges were also determined in subgroups for age and gender in the
205
reference subpopulation. The TSH medians and reference ranges for these subgroups are
206
reported in Table 2. Although the small differences in TSH between subgroups for age
10
207
(P=0.002) and gender (P=0.005) reached statistical significance, the TSH distribution
208
curves of the subgroups were practically superimposed (Fig. 2).
209
210
TSH distribution curves
211
A skewed upper limit of TSH distribution was noticeable even in the reference
212
subpopulation (skewness 1.7, kurtosis 5.1). It was also noticeable in its age and gender
213
subgroups (Fig. 2). Log-transformation of the TSH values of the reference subpopulation
214
resulted in overcorrection (skewness -1.5, kurtosis 10.9). A well suitable function was the
215
square root transformation that yielded a better normal distribution (skewness 0.6, kurtosis
216
1.4) and by establishing from the 95% confidence limits a TSH reference range 0.41 –
217
3.22 mU/L that was well comparable with the result obtained by using the nonparametric
218
method.
219
220
11
221
Discussion
222
223
Our study showed that stringent population screening as proposed by the NACB
224
guidelines (1), in lieu of a more lenient approach, decreased the TSH upper reference limit
225
from 4.43 mU/L to 3.37 mU/L while the lower limit remained unchanged. The decrease in
226
the upper reference limit was mainly attributable to the exclusion of TPOAb-positive
227
participants. Screening for use of medications had a smaller effect on the TSH upper limit.
228
In the rigorously screened reference subpopulation, there were significant differences in
229
TSH between subgroups for age and gender, but the TSH distribution curves of these
230
subgroups were comparable.
231
232
In the present paper, out of the thyroid-healthy population and its subpopulations, the
233
exclusion criteria of the reference subpopulation followed the NACB guidelines most
234
closely. We therefore propose the TSH reference range of this subpopulation (0.4 – 3.4
235
mU/L) to be the TSH reference range for the adult population when the Abbott Architect
236
method is used. Juxtaposed with reference ranges proposed by other large population-
237
based studies with random or other type of probability sampling, the TSH upper limit has
238
been comparable (3.60 mU/L) in the Health Study of Nord-Trondelag conducted in Norway
239
(20) and higher in other studies from the United States, Australia, and the Netherlands (4.0
240
– 4.66 mU/L) (6,9,10). The TSH lower limits proposed by these studies have varied only
241
slightly (0.34 – 0.48 mU/L). In other noteworthy studies without random sampling, Kratzsch
242
et al. determined 3.77 mU/L (12) and Schalin-Jäntti et al. 3.6 mU/L (8) to be the TSH
243
upper reference limit, in line with our result. However, direct comparison between these
244
results is challenging due to different sampling and screening strategies. In addition,
245
differences in laboratory assays and in the iodine uptake of participants may cause
12
246
variation between results, as has been mentioned earlier in other papers (8,11,21).
247
Although there has been great overall improvement in TSH assay sensitivity over the
248
years, there is clear evidence of variability in clinical performance of the different TSH
249
immunometric assay methods (22). We think that these factors oblige caution in the
250
generalisation of the results of any single TSH reference range study.
251
252
TPOAb-positivity had a substantial effect on the TSH upper limit in the present study. The
253
TSH upper limit was 4.43 mU/L in the thyroid-healthy population and 3.71 mU/L in the risk
254
factor-free (TPOAb-negative) subpopulation. Previous studies have shown variation
255
regarding the effect of TPOAb-positivity on TSH upper limit. In some studies the effect
256
seems evident (4,8) yet in others nonexistent or less pronounced (7,12,23). In at least one
257
study the effect of TPOAb-positivity was apparent only among women (24). Overall, the
258
differing effects of TPOAb status on TSH upper limit observed in various studies may be
259
partially linked to differences in sampling and cut-off concentrations for a positive TPOAb
260
result. In the present study, the cut-off concentration of 5.6 IU/mL, provided by the
261
manufacturer of the assay, was relatively low but also provided a high sensitivity for
262
screening for thyroid disease.
263
264
One of the additional aims of our study was to assess the effect of medications on TSH
265
reference range. In a study by Jensen et al., medications had no effect on TSH (25). In our
266
study, after applying all the other feasible exclusion criteria suggested in the NACB
267
guidelines we first excluded users of medications with a potential effect on thyroid function
268
tests. This caused only a 0.1 mU/L decrease in the TSH upper limit. In contrast, the
269
exclusion of users of any medications caused a more marked decrease of the TSH upper
270
limit by 0.3 mU/L but at the same time a 60% decrease in the population size. Medications
13
271
had no notable effect on the TSH lower limit. To sum up, the exclusion of participants with
272
any medications, albeit having a relatively minor effect on TSH reference range, seems to
273
be necessary. Our results raise questions of whether these observed changes in the TSH
274
upper limit were attributed to medications per se or rather to thyroid function affecting
275
comorbidities that users of medications could have.
276
277
The dilemma of a skewed upper limit of TSH distribution has been discussed earlier, most
278
notably in the NACB guideline (1) and by Surks et al. (2,3,26). In the present study, a
279
skewed upper limit was noticeable in all age and gender subgroups of the reference
280
subpopulation. Race neither explains the skewed TSH distribution here because the
281
Finnish population is ethnically homogeneous as only 1.9% of the population spoke a
282
foreign language as their mother tongue in the year 2000 (27). Furthermore, the same
283
assay and analyser were used throughout the study. Concerning the remaining plausible
284
explanations for the skewed TSH upper limit, emerging thyroid disorders that even the
285
sensitive TPOAb assays fail to detect are to be considered.
286
287
In several (3-7), though not in all (8-10) studies, TSH has been shown to increase with
288
age. Gender has been shown to have a slight effect on TSH reference limits in some
289
studies (4,24) whereas in others the effect has been inconsiderable or non-existent
290
(6,7,23,25). In the present study, there were small, but statistically significant differences in
291
TSH between subgroups for age and between men and women. The middle-aged
292
subgroup had the highest TSH median (1.41 mU/L) while the oldest subgroup had the
293
second highest levels (1.33 mU/L). Men also had slightly higher TSH median than women
294
(1.36 mU/L vs. 1.29 mU/L). However, the TSH distribution curves of all these subgroups
14
295
were comparable. Thus, we suggest that a universal TSH reference range may be
296
applicable for the adult population.
297
298
Our study has some limitations. First, iodine insufficiency was not excluded from
299
participants but the iodine intake in Finland has been shown to be sufficient in earlier
300
studies (28,29). Second, we could not establish an ad hoc TPOAb reference range for this
301
study and had to therefore rely on the cutoff limit provided by the manufacturer. Third, data
302
were not available for participants’ family history of thyroid diseases. Fourth, the serum
303
TgAb levels were not measured. However, previous studies have shown that the TPOAb
304
independent association between TgAb and TSH levels is minor or nonexistent (25,30).
305
Fifth, the TPOAb concentration was not measured in all participants with TSH in the range
306
of 0.4 – 2.5 mU/L. As a consequence of the last three limitations, we could not follow all
307
the stringent criteria of the NACB guidelines.
308
309
In conclusion, in our nationwide study that used stratified random sampling and adhered to
310
the NACB criteria as far as possible, the TSH reference upper limit (3.4 mU/L) on the
311
Abbott Architect ci8200 integrated system was lower than those presently used in most
312
laboratories. Exclusion of participants with any medications and TPOAb-positivity (cut-off
313
limit of 5.6 IU/mL) resulted in a marked decrease of the TSH upper limit. In the stringently
314
screened reference subpopulation, TSH did not increase with age, which is in contradiction
315
to some earlier papers and advocates further studies. Finally, it has to be emphasised that
316
the decision to treat subclinical thyroid dysfunction should not rely solely on the
317
established TSH reference limits but rather on the appropriate treatment guidelines and
318
comprehensive clinical assessment of the patient in question.
319
15
320
Funding
321
322
The project organisation created for the Health 2000 Survey involved the Finnish Centre
323
for Pensions, the Social Insurance Institution, the National Public Health Institute, the
324
Local Government Pensions Institution, the National Research and Development Centre
325
for Welfare and Health, the Finnish Dental Society and the Finnish Dental Association,
326
Statistics Finland, the Finnish Work Environment Fund, the Finnish Institute for
327
Occupational Health, the UKK Institute for Health Promotion, the State Pensions Office
328
and the State Work Environment Fund.
16
329
Declaration of interest
330
331
None declared.
332
17
333
References
334
335
1. Demers LM, Spencer CA, editors. Laboratory Medicine Practice Guidelines: Laboratory
336
Support for the Diagnosis and Monitoring of Thyroid Disease. Washington, DC: National
337
Academy of Clinical Biochemistry, 2002.
338
2. Surks MI, Boucai L. Age- and race-based serum thyrotropin reference limits. J Clin
339
Endocrinol Metab 2010;95:496-502.
340
3. Surks MI, Hollowell JG. Age-specific distribution of serum thyrotropin and antithyroid
341
antibodies in the US population: implications for the prevalence of subclinical
342
hypothyroidism. J Clin Endocrinol Metab 2007;92:4575-4582.
343
4. Friis-Hansen L, Hilsted L. Reference intervals for thyreotropin and thyroid hormones for
344
healthy adults based on the NOBIDA material and determined using a Modular E170. Clin
345
Chem Lab Med 2008;46:1305-1312.
346
5. Atzmon G, Barzilai N, Hollowell JG, Surks MI, Gabriely I. Extreme longevity is
347
associated with increased serum thyrotropin. J Clin Endocrinol Metab 2009;94:1251-1254.
348
6. Hollowell JG, Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer CA, et al.
349
Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994):
350
National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab
351
2002;87:489-499.
352
7. Takeda K, Mishiba M, Sugiura H, Nakajima A, Kohama M, Hiramatsu S. Evaluated
353
reference intervals for serum free thyroxine and thyrotropin using the conventional outliner
354
rejection test without regard to presence of thyroid antibodies and prevalence of thyroid
355
dysfunction in Japanese subjects. Endocr J 2009;56:1059-1066.
18
356
8. Schalin-Jäntti C, Tanner P, Välimäki MJ, Hämäläinen E. Serum TSH reference interval
357
in healthy Finnish adults using the Abbott Architect 2000i Analyzer. Scand J Clin Lab
358
Invest 2011;71:344-349.
359
9. O'Leary PC, Feddema PH, Michelangeli VP, Leedman PJ, Chew GT, Knuiman M, et al.
360
Investigations of thyroid hormones and antibodies based on a community health survey:
361
the Busselton thyroid study. Clin Endocrinol (Oxf) 2006;64:97-104.
362
10. Hoogendoorn EH, Hermus AR, de Vegt F, Ross HA, Verbeek AL, Kiemeney LA, et al.
363
Thyroid function and prevalence of anti-thyroperoxidase antibodies in a population with
364
borderline sufficient iodine intake: influences of age and sex. Clin Chem 2006;52:104-111.
365
11. Waise A, Price HC. The upper limit of the reference range for thyroid-stimulating
366
hormone should not be confused with a cut-off to define subclinical hypothyroidism. Ann
367
Clin Biochem 2009;46:93-98.
368
12. Kratzsch J, Fiedler GM, Leichtle A, Brügel M, Buchbinder S, Otto L, et al. New
369
reference intervals for thyrotropin and thyroid hormones based on National Academy of
370
Clinical Biochemistry criteria and regular ultrasonography of the thyroid. Clin Chem
371
2005;51:1480-1486.
372
13. Zöphel K, Wunderlich G, Grüning T, Koch R, Döge H, Kotzerke J. Where does
373
subclinical hypothyroidism start? Implications for the definition of the upper reference limit
374
for thyroid stimulating hormone. Nuklearmedizin 2005;44:56-61.
375
14. Zöphel K, Wunderlich G, Kotzerke J. Should we really determine a reference
376
population for the definition of thyroid-stimulating hormone reference interval? Clin Chem
377
2006;52:329-330.
378
15. Laiho J, Djerf K, Lehtonen R. Sampling design. In: Heistaro S, editor. Methodology
379
report, Health 2000 Survey. Helsinki: National Public Health Institute, 2008:13-15.
380
Available at: http://www.terveys2000.fi/doc/methodologyrep.pdf. Accessed 7 June, 2014.
19
381
16. O'Leary PC, Boyne P, Atkinson G, Mileham KJ, James I. Longitudinal study of serum
382
thyroid hormone levels during normal pregnancy. Int J Gynaecol Obstet 1992;38:171-179.
383
17. Kurioka H, Takahashi K, Miyazaki K. Maternal thyroid function during pregnancy and
384
puerperal period. Endocr J 2005;52:587-591.
385
18. Kundra P, Burman KD. The effect of medications on thyroid function tests. Med Clin
386
North Am 2012;96:283-295.
387
19. Wiegratz I, Kutschera E, Lee JH, Moore C, Mellinger U, Winkler UH, et al. Effect of
388
four oral contraceptives on thyroid hormones, adrenal and blood pressure parameters.
389
Contraception 2003;67:361-366.
390
20. Bjoro T, Holmen J, Krüger O, Midthjell K, Hunstad K, Schreiner T, et al. Prevalence of
391
thyroid disease, thyroid dysfunction and thyroid peroxidase antibodies in a large,
392
unselected population. The Health Study of Nord-Trondelag (HUNT). Eur J Endocrinol
393
2000;143:639-647.
394
21. Guan H, Shan Z, Teng X, Li Y, Teng D, Jin Y, et al. Influence of iodine on the
395
reference interval of TSH and the optimal interval of TSH: results of a follow-up study in
396
areas with different iodine intakes. Clin Endocrinol (Oxf) 2008;69:136-141.
397
22. Spencer CA, Takeuchi M, Kazarosyan M, MacKenzie F, Beckett GJ, Wilkinson E.
398
Interlaboratory/intermethod differences in functional sensitivity of immunometric assays of
399
thyrotropin (TSH) and impact on reliability of measurement of subnormal concentrations of
400
TSH. Clin Chem. 1995;41:367-374.
401
23. Chan AO, Iu YP, Shek CC. The reference interval of thyroid-stimulating hormone in
402
Hong Kong Chinese. J Clin Pathol 2011;64:433-436.
403
24. Eskelinen S, Suominen P, Vahlberg T, Löppönen M, Isoaho R, Kivelä SL, et al. The
404
effect of thyroid antibody positivity on reference intervals for thyroid stimulating hormone
20
405
(TSH) and free thyroxine (FT4) in an aged population. Clin Chem Lab Med 2005;43:1380-
406
1385.
407
25. Jensen E, Hyltoft Petersen P, Blaabjerg O, Hansen PS, Brix TH, Kyvik KO, et al.
408
Establishment of a serum thyroid stimulating hormone (TSH) reference interval in healthy
409
adults. The importance of environmental factors, including thyroid antibodies. Clin Chem
410
Lab Med 2004;42:824-832.
411
26. Surks MI. TSH Reference Limits: New Concepts and Implications for Diagnosis of
412
Subclinical Hypothyroidism. Endocr Pract 2013;19:1066-1069.
413
27. Statistical Yearbook of Finland 2012. Helsinki: Statistics Finland, 2012:97. Available
414
from:
415
http://www.stat.fi/tup/julkaisut/tiedostot/julkaisuluettelo/yyti_stv_201200_2012_6270_net.p
416
df. Accessed 7 June, 2014.
417
28. Varo P, Saari E, Paaso A, Koivistoinen P. Iodine in Finnish foods. Int J Vitam Nutr Res
418
1982;52:80-89.
419
29. Lamberg BA, Haikonen M, Mäkelä M, Jukkara A, Axelson E, Welin MG. Further
420
decrease in thyroidal uptake and disappearance of endemic goitre in children after 30
421
years of iodine prophylaxis in the east of Finland. Acta Endocrinol (Copenh) 1981;98:205-
422
209.
423
30. Bülow Pedersen I, Laurberg P, Knudsen N, Jørgensen T, Perrild H, Ovesen L, et al. A
424
population study of the association between thyroid autoantibodies in serum and
425
abnormalities in thyroid function and structure. Clin Endocrinol (Oxf) 2005;62:713-720.
426
427
21
428
429
430
431
Figure 1 Flow-chart of inclusion-exclusion of participants in the present study. aExclusion
432
of nonambulatory participants and those without a blood sample drawn between 8 AM – 6
433
PM window or with any of the following: use of thyroid hormones or antithyroid agents;
434
personal history of thyroid disease or goiter; or ongoing pregnancy or breast-feeding. bSee
435
methods.
436
22
437
438
439
Figure 2 TSH distribution by age (A) and gender (B) in the reference subpopulation.
23
440
Table 1 Characteristics of the thyroid-healthy population and its subpopulations.
441
Study population
Thyroid-healthy populationa
Risk factor-free subpopulationb
No thyroid affecting medicationsc
Reference subpopulationd
n
Age (yr)
Women
BMI
Medications
Daily smoking
5709
52.2 ± 14.8
2964 (51.9%)
26.9 ± 4.6
3422 (59.9%)
1278 (22.4%)
4586
52.4 ± 14.8
2330 (50.8%)
26.9 ± 4.7
2737 (59.7%)
1045 (22.8%)
3453
50.6 ± 14.3
1508 (43.7%)
26.7 ± 4.6
1604 (46.5%)
841 (24.4%)
1849
46.2 ± 11.7
745 (40.3%)
26.2 ± 4.2
0 (0%)
504 (27.3%)
442
443
Values are mean ± SD for continuous data and number and percentage for categorical data. BMI = body mass index (kg/m2).
444
aAmbulatory
445
hormones or antithyroid agents; personal history of thyroid disease or goiter; or ongoing pregnancy or breast-feeding. bThyroid-
446
healthy population after exclusion of TPOAb-positive participants. See methods for description. cSee Supplemental data, Table 1 for
447
a list of these medications. dRisk factor-free subpopulation after exclusion of participants using any medications.
448
24
participants with blood sample drawn between 8 AM – 6 PM window without any of the following: use of thyroid
449
Table 2 TSH reference ranges established from the thyroid-healthy population and its
450
subpopulations formed with an increasing rigour of screening for thyroid disease and
451
factors affecting thyroid function.
452
TSH (mU/L)
Percentile
Study population
n
Median
2.5
97.5
5709
1.37
0.41
4.43
4586
1.34
0.41
3.71
3453
1.35
0.41
3.60
1849
1.33
0.43
3.37
Men
1104
1.36
0.50
3.55
Women
745
1.29
0.39
3.22
30-44
953
1.29
0.42
3.07
45-59
626
1.41
0.44
4.13
≥60
270
1.33
0.38
3.41
Thyroid-healthy populationa
Risk factor-free subpopulationb
No thyroid affecting medicationsc
Reference subpopulationd
Gender
Age (years)
453
participants with blood sample drawn between 8 AM – 6 PM window without
454
aAmbulatory
455
any of the following: use of thyroid hormones or antithyroid agents; personal history of
456
thyroid disease or goiter; or ongoing pregnancy or breast-feeding. bThyroid-healthy
457
population after exclusion of TPOAb-positive participants. See methods for description.
458
cSee
459
subpopulation after exclusion of participants using any medications.
25
Supplemental data, Table 1 for a list of these medications. dRisk factor-free
460
Supplemental data, Table 1 Use of medications that are known to have a potential effect
461
on thyroid function tests among the participants in the risk factor-free subpopulation.
462
Medication
n (%) participants
Amiodarone
3 (0.07%)
Pituitary hormones and analogues
1 (0.02%)
Sex hormones
488 (10.6%)
Glucocorticoids
91 (2,0%)
Non-steroidal anti-inflammatory drugsa
420 (9.2%)
Opioids
145 (3.2%)
Drugs for acid related disordersa
85 (1.9%)
Furosemide
113 (2.5%)
Phenobarbital
0 (0%)
Rifampisin
0 (0%)
Phenytoin
3 (0.07%)
Carbamazepin
27 (0.6%)
Antineoplastic agents
29 (0.6%)
Mycophenolic acid
0 (0%)
Tacrolimuse
0 (0%)
Bexarotene
0 (0%)
463
464
aIf
465
466
26
used within one week prior blood test.