Original. Jing Cai 1) *, Yujie Fang 1) *, Da Jing 2) *, Shaoyong Xu 1), Jie Ming 1), Bin Gao 1), Han Shen 3), Rong Zhang 1) and Qiuhe Ji 1)

Transcription

1 2016, 63 (4), Original Reference intervals of thyroid hormones in a previously iodine-deficient but presently more than adequate area of Western China a population-based survey Jing Cai 1) *, Yujie Fang 1) *, Da Jing 2) *, Shaoyong Xu 1), Jie Ming 1), Bin Gao 1), Han Shen 3), Rong Zhang 1) and Qiuhe Ji 1) 1) Department of Endocrinology, Xijing Hospital, Fourth Military Medical University, Xi an, China 2) Department of Biomedical Engineering, Fourth Military Medical University, Xi an, China 3) Department of Endocrinology, Beijing Tongren Hospital, Capital Medical University, Beijing, China Abstract. The aim of our study is to establish the reference intervals (RIs) of thyroid hormones in a previously iodinedeficient area but presently more than iodine-adequate area of Western China, and also to investigate the factors which affect thyroid function. The cross-sectional study conducted in Xi an, was based on China National Diabetes and Metabolic Disorders Survey. Among 1286 participating adults, 717 were finally included as reference population. Thyrotropin (TSH), total triiodothyronine (T 3 ), free triiodothyronine (FT 3 ), total thyroxine (T 4 ), free thyroxine (FT 4 ), thyroperoxidase antibody (TPO-Ab) and thyroglobulin antibody (Tg-Ab) were measured. Thyroid ultrasound examination was also performed. The present study established the new RIs of serum TSH ( ), FT 4 ( ), FT 3 ( ), T 4 ( ) and T 3 ( ), which were different from the data provided by the manufacturers. Significant differences among all the age groups were observed in FT 3, but neither in TSH nor in FT 4. The TSH levels in adults with pathologic results or positive thyroid autoantibody were significantly higher than those in reference adults. Our present results provide valuable references for the diagnosis of thyroid diseases in population of Western China. Considering that most inland areas of China have faced the challenge of the transition from iodine deficiency to adequacy or more than adequacy, we recommend physicians utilize our RIs to determine thyroid diseases in the similar areas with Xi an in China. Key words Thyroid hormone, Reference intervals, Population-based survey, Chinese adults THYROID hormones, including thyrotropin (TSH), total triiodothyronine (T 3 ), free triiodothyronine (FT 3 ), total thyroxine (T 4 ) and free thyroxine (FT 4 ), are regarded as regular biochemical indices to diagnose thyroid diseases [1]. TSH, an essential protein molecule secreted by the anterior pituitary, acts as the most sensitive marker for thyroid dysfunction and an essential index for the diagnosis of subclinical thyroid diseases [2]. The serum concentrations of all thyroid hormones can be affected by age, gender, race, medication, and particularly iodine nutritional status. Submitted Oct. 20, 2015; Accepted Jan. 4, 2016 as EJ Released online in J-STAGE as advance publication Feb. 3, 2016 Correspondence to Qiuhe Ji, Department of Endocrinology, Xijing Hospital, Fourth Military Medical University, 169 Changle Road West, Xi an , China. qiuheji@hotmail.com * JC, YF and DJ contributed equally to this work. The Japan Endocrine Society Thus, the establishment of accurate reference intervals (RIs) for specific population is of great importance [3]. It has been suggested by the National Committee for Clinical Laboratory Standards (now renamed as Clinical and Laboratory Standards Institute) that each laboratory should verify their own specific RIs for the population they serve. However, most clinical laboratories are still using the RIs provided by the commercial assay manufacturers due to time-consumption and high expenditures for the selection of a suitable largescale reference population. It might be a major limitation for the accurate and high-quality diagnosis of thyroid diseases. The association of iodine intake with the occurrence of thyroid disease is U-shaped [4]. Either iodine deficiency or iodine more than adequacy has great impact on determination of the RIs of thyroid hor-

2 382 Cai et al. mones. China initiated the program of Universal Salt Iodization (USI) in 1996, and most areas of the country have become iodine sufficient since the year of 2000 [5]. As one of the largest cities in Western China, Xi an was once an iodine-deficiency area (median of water iodine concentration 3.99 μg/l; median of urinary iodine [MUI] concentration of school-age students in μg/l [6]) and has achieved the stage goal of elimination of iodine-deficiency disorder (IDD) since 2000 [7]. However, in 2013 it was reported that the MUI of school-age children in Xi an was μg/l [8], higher than the standard levels proposed by World Health Organization (WHO) [9], indicating that the iodine intake of population in Xi an was more than adequate. Actually, many inland areas of China have experienced the similar conversion from iodine-deficiency to iodine-sufficiency or more than adequacy during the past two decades [10], which necessitates the establishment of accurate RIs. More importantly, although China has three RIs of thyroid hormones so far, the three RIs were all from hospital- or communitybased studies or from studies not excluding patients with abnormal findings [11-13]. It might limit the generalization of the results to general population. Population-based studies that focus on establishing the RIs of thyroid hormones with rigorous eligibility criteria were thus required in the previously iodine-deficient but presently more than adequate areas, just like Xi an. Given the above background, we conducted a population-based cross-sectional survey, aiming to establish the RIs of thyroid hormones in a previously iodine-deficient but presently more than iodine-adequate area of Western China, and also to analyze the correlation between confounding factors and thyroid hormone levels. Materials and Methods Subjects This study was a subsection in the second stage of the China National Diabetes and Metabolic Disorders Survey (CNDMDS), a nationwide population-based cross-sectional survey launched in The details of the CNDMDS can be found elsewhere [14]. In brief, a multi-stage stratified sampling method was used to select a nationally representative sample of Chinese adults. Our present study was conducted in Xi an city, Shaanxi province, from July 2012 to September A total of 1286 adults aged years (male 501, female 785) were invited and participated in this study with an estimated response rate above 80%. Finally, we included 1181 subjects as the whole study population after excluding 105 subjects according to the following criteria (1) subjects with a history of thyroid disease; (2) pregnant or breastfeeding women; (3) subjects with moderate-to-severe ill health; (4) subjects who were taking medicine with potential influences on the thyroid function, such as estrogen, amiodarone, anti-epileptic drugs, glucocorticoids and excess iodine ingestion. In addition, to establish the RIs of thyroid hormones, we also excluded 457 subjects with abnormal results or thyroid antibodies (TPO-Ab or Tg-Ab) positive, and a total of 717 subjects (male 330, female 387) were included as reference population. The study was approved by the Ethics Committee of Xijing Hospital, Fourth Military Medical University. All participants signed written informed consent prior to data collection. Data and sample collection All participants filled out questionnaires requesting information on demographic characteristics, lifestyle risk factors, family history of diseases, and personal medical history [15]. The National Academy of Clinical Biochemistry (NACB) criteria for biochemical tests and regular thyroid were used to determine the eligibility for inclusion of subjects. Ultrasonography of thyroid was performed, and the definition of normal thyroid gland was referred as the thyroid with a homogenous echo pattern on thyroid images without any thyroid nodules and goiters. Venous blood samples were obtained, and centrifugation was performed within 30 min of blood drawing. The serum samples were stored at -80ºC until the measurement for TSH, FT 4, FT 3, T 3, T 4, TPO-Ab and Tg-Ab levels. All serum samples were measured by using electrochemiluminescence immunoassays (ADVIA Centaur XP, Siemens Healthcare Diagnostics Inc., Tarrytown, NY). The third generation assay, a two-site sandwich immunoassay based on direct chemiluminescent technology with functional sensitivity of 0.008, was employed for determining the TSH values. The intra- and interassay coefficients of variation (CV) were lower than 4.8% for the TSH values ranging from to 150. The FT 4, FT 3, T 3, T 4, TPO-Ab and Tg-Ab

3 Reference intervals of thyroid hormones 383 assays were competitive, chemiluminescent immunoassays. The intra-assay and inter-assay CV for all these values were lower than 6.2% and 7.1%, respectively. The ADVIA Centaur TSH3-UL method was traceable to the WHO 3 rd International Reference Preparation (IRP) 81/565, and the FT 4, FT 3, T 3 and T 4 assays were traceable to internal standard manufactured using the United States Pharmacopeia (U.S.P.) material. The TPO-Ab and Tg-Ab reference calibrators were traceable to the MRC 66/387 and MRC 65/093 reference preparations, respectively. The corresponding normal levels of serum TSH, FT 4, FT 3, T 4, T 3, TPO-Ab and Tg-Ab provided by the manufacturers were , , , , , 0-60 U/mL and 0-60 U/mL, respectively. Statistical analysis The data were analyzed by using IBM SPSS statistics (version 19.0). All data were expressed as mean ± standard deviation (SD), mean ± 1.96 SD, or median with interquartile range as appropriate. The Kolmogorov-Smirnov method was used to test for normality. The FT 4, FT 3, T 4 and T 3 levels in reference population were normal distribution. After log-transformed, the TSH levels were adhering to the Gaussian distribution. Therefore, the 95% intervals of TSH, FT 4, FT 3, T 4 and T 3 levels were calculated by mean ± 1.96 SD (for TSH, the data was prepared conventionally by using log-transformed TSH values). Comparisons of the TSH and other hormones for different age and gender groups were performed by t-test or one-way ANOVA. Comparisons of the TSH in normal and abnormal groups were performed by Mann Whitney U test. A two-way P-value less than 0.05 was considered as significant. Results Clinical characterization of the study subjects The clinical characteristics of the study participants are shown in Table 1, and the age distributions of the 717 reference subjects are shown in Table 2. Gender was comparably distributed within all age groups (data not shown). On average, the reference subjects with normal results of thyroid ultrasound and negative thyroid antibodies were younger as compared with the whole subjects. No significant difference was found between the reference subjects and the whole subjects except for the proportion of cigarette smoking and alcohol drinking. Table 1 Comparison of anthropometric and other relevant data between the whole subjects and the reference subjects Whole subjects Reference subjects Number 1, Age, years 53.1 ± ± 12.9* Gender (Male), N (%) 459 (38.9) 330 (46.0) Ethnics (Han), N (%) 1159 (98.7) 705 (98.3) Cigarette smoking, N (%) 232 (19.6) 173 (24.1)* Alcohol drinking, N (%) 226 (19.1) 167 (23.3)* Height, cm ± ± 8.4 Weight, kg 64.5 ± ± 11.4 Body mass index, kg/m ± ± 3.5 Systolic blood pressure, mmhg ± ± 21.1 Diastolic blood pressure, mmhg 77.1 ± ± 10.9 Heart rate, beats/min 73.7 ± ± 11.1 * P<0.05 Table 2 Means and 95% intervals of serum TSH, FT 4, FT 3, T 4 and T 3 levels according to age and gender N TSH () FT 4 () FT 3 () T 4 () T 3 () Age group, year (ref.) ( ) 16.1 ( ) 4.89 ( ) ( ) 1.66 ( ) ( ) 16.0 ( ) 4.76 ( ) ( ) 1.66 ( ) ( ) 15.6 ( ) 4.71 ( ) ( ) 1.63 ( ) ( ) 15.8 ( ) 4.69 ( )* ( ) 1.65 ( ) ( ) 15.5 ( ) 4.63 ( )* ( ) 1.66 ( ) > ( ) 15.6 ( ) 4.31 ( )* ( ) 1.53 ( )* Gender Men (ref.) ( ) 16.3 ( ) 4.88 ( ) ( ) 1.69 ( ) Women ( )* 15.3 ( )* 4.51 ( )* ( ) 1.60 ( )* Total ( ) 15.7 ( ) 4.68 ( ) ( ) 1.64 ( ) * P<0.05. The TSH (after log-transformed), FT 4, FT 3, T 4 and T 3 levels were expressed as mean since they were all normally distributed.

4 384 Cai et al. The normal RIs of serum TSH and other thyroid hormones The statistical distributions of TSH, FT 4 and FT 3 are shown in Fig. 1. The RIs of serum TSH, FT 4, FT 3, T 4 and T 3 were , , , and , respectively (Table 2). The effects of age and gender on TSH and other thyroid hormones in the reference population The reference population was further categorized into various age groups and gender groups as shown in Table 2. No statistically significant difference was observed in the TSH levels among all the age groups. The mean FT 3 levels in age groups of years (4.69 ), years (4.64 ) and over 71 years (4.31 ) were lower than that in age group of (4.89, P<0.05), and the elderly group (over 71 years old) had lower mean T 3 as compared with age group (1.53 vs. 1.66, P<0.05). In addition, males had higher mean FT 3, FT 4 and T 3 levels but lower mean TSH level (1.38 vs. 1.69, P<0.05) than females. The effects of risk factors on serum TSH determination Thyroid ultrasound abnormalities The prevalence of ultrasound abnormalities in all subjects was 28.9% and women had higher prevalence than men (32.4% vs. 23.3%, P<0.001). The median TSH values in the ultrasound abnormal group were higher than those in normal subjects (P<0.01, Table 3). Thyroid autoantibodies The prevalence of positive thyroid autoantibodies (either TPO-Ab or TG-Ab) in all subjects was 16.9%, and women showed significantly higher positive rate in thyroid autoantibodies than men in each group (P<0.001). Moreover, higher median TSH value were observed in women who were autoantibody-positive as compared with the normal female group (2.08 vs. 1.77, P<0.001). Fig. 1 Histograms of TSH (A), FT 4 (B) and FT 3 (C) samples in reference population Horizontal ordinates represent TSH, FT 4 and FT 3 values, and vertical ordinates represent the frequencies. The curves are the normal distribution curves. Table 3 The effects of different risk factors on TSH level Factors N (%) TSH () Women Men N (%) TSH () N (%) TSH () Normal (ref.) 717 (61.3) 1.65 (1.07) 387 (51.0) 1.77 (1.29) 330 (71.9) 1.44 (0.95) Abnormal ultrasound 341 (28.9) 1.81 (1.42)* 234 (32.4) 2.03 (1.74) 107 (23.3) 1.49 (1.14) Thyroid antibody positive 200 (16.9) 2.00 (1.78)* 165 (22.9) 2.08 (1.78)* 35 (7.6) 1.49 (1.07) TPO-Ab+ 136 (11.5) 2.10 (1.89)* 111 (15.4) 2.53 (1.79)* 25 (5.4) 1.28 (1.01) Tg-Ab+ 162 (13.7) 1.99 (1.88)* 139 (19.3) 2.05 (1.84)* 23 (5.0) 1.49 (0.92) Both 98 (8.3) 2.22 (2.14)* 85 (11.8) 2.59 (2.07)* 13 (2.8) 1.28 (1.00) Whole subjects 1, (1.26) (1.51) (0.98) * P<0.01. The TSH levels were expressed as median with interquartile range (the contents in brackets represent interquartile range) since they were not normally distributed.

5 Reference intervals of thyroid hormones 385 The effects of risk factors on other thyroid hormones The FT 3 values in ultrasound abnormal subjects were lower than those in the normal controls (P<0.05). The T 4 values in males with thyroid-autoantibody-positive or TPO-Ab-positive were found to be higher than those in the normal controls (P<0.05). Discussion The present study established the new RIs of serum TSH ( ), FT 4 ( ), FT 3 ( ), T 4 ( ) and T 3 ( ) in Xi an of Western China, and the RIs were different from the data (e.g. TSH ) provided by the manufacturers. In addition, significant differences among all the age groups were observed in FT 3, but neither in TSH nor in FT 4. We also showed that the TSH levels were significantly higher in adults with pathologic results or positive thyroid autoantibodies than those in reference population (1.81 or 2.00 vs ). To our knowledge, this is the first populationbased study to establish the RIs of thyroid hormones with rigorous eligibility criteria in a previously iodinedeficient but presently more than iodine-adequate area of Western China. Our results showed that the RIs were different from the data provided by the manufacturers. For example, the upper-limit of serum TSH level in our results was slightly higher, and the upper-limit of serum FT 4 level was slight lower than those provided by the manufacturers (5.51 vs and 20.4 vs. 22.7, respectively). We speculated with caution that apparently thyroid healthy individuals were likely to be recruited without rigorous exclusion criteria and iodine nutritional status was also not considered when the manufactures established the RIs of thyroid hormones. Table 4 shows the RIs of thyroid hormones reported by the investigators in various countries and regions during the past decade [11-13, 16-22]. Our upper-limit of serum TSH level was marginally higher than Guan s results but similar with Li s results [11, 13]. It was partly due to the fact that the reference population in the former study was not only from mildly iodine deficient areas but also from excessive iodine intake areas, while the reference population in the latter was only recruited in iodine-sufficient areas. The major strength of our study was the populationbased sample and rigorous eligibility criteria in the selection of reference population. Generally speaking, the RIs and 95% confidence intervals of thyroid hormones should be established in the thyroid disease-free reference population, and the methods of reference population selection influence the results. Therefore, according to the results from the National Health and Nutrition Examination Survey (NHANES) III studies, the NACB proposed rigorous criteria for selecting reference population [23]. However, most previous studies in China were all from hospital- or community-based studies or from studies not excluding patients with abnormal findings. It might thus limit the generalization of the results to general population. Our results showed that significant differences among all the age groups were observed in FT 3, but neither in TSH nor in FT 4. The findings were in accord with substantial previous studies [12, 21, 22, 24, 25], but were inconsistent with several other studies [13, 17, 20]. The possible explanations might be associated with the differences of race and iodine nutritional status [17, 24, 26, 27]. In addition, our data showed that the TSH levels in the ultrasound abnormal group were higher than those in the normal subjects, but no statistically significant differences were observed after stratification by gender. These findings were inconsistant with the results reported by Li et al. [12]. Some limitations in the present study should be addressed. Firstly, liquid chromatography-tandem mass spectrometry (LC-MS/MS) is the gold standard method for thyroid hormones detemination [28-33], and the immunoassay for FT 4, FT 3 and total T 3 in our study is inaccurate at low concentrations. For example, for the individuals classified as having subclinical hypothyroidism with a standard immunoassay, 65% of them are actually hypothyroid and have low FT 4 levels by LC-MS/MS measurements [33]. However, it s still reasonable to establish the RIs determined by immunoassays, since it is not possible to expect each laboratory in China to use LC-MS/MS nowadays. Secondly, urinary iodine was not assessed and the association of urinary iodine with the RIs of thyroid hormones was not analyzed in the present study. Thirdly, as the numbers of young subjects with age less than 30 years and elderly subjects with age over 71 years were relatively low, the accuracy of the estimate would be relatively limited. Further studies with larger numbers of reference subjects, particular young and elderly subjects, are needed. Lastly, due to the examination deviation among differ-

6 386 Cai et al. Table 4 Reference intervals of thyroid hormones obtained from recent studies and manufacturers Authors Country Published year Kratzsch J, et al. [16] Volzke H, et al. [17] O Leary P C, et al. [18] Guan H, et al. [11] Hamilton TE, et al. [19] Quinn FA, et al. [13] Li C, et al. [12] Yoshihara A, et al. [20] Marwaha R. K, et al. [21] Sriphrapradang C, et al. [22] Germany 2005 Germany 2005 Including criteria Iodine status Not mentioned Previously iodinedeficient Australia 2006 NACB Iodine replete China 2008 U.S.A 2008 China 2009 NACB China 2011 Japan 2011 India 2013 The present study China MUI μg/l Methods for TFT Age Sample size , ,026 T 4 T 3 FT 3 FT 3 RI , Iodine-sufficient ELISA Borderline sufficient MUI μg/l More than sufficient MUI μg/l Thailand 2014 NACB Iodine sufficient Previously deficient, now iodine replete TSH FT 4 FT 3 T 4 T , , , , FT 3 FT 3 T 4 T ng/ml The manufacturers FT (Siemens 4 FT Healthcare 3 T Diagnostics Inc.) 4 T 3 TFT, thyroid function test; NACB, The National Academy of Clinical Biochemistry;, chemiluminescence immunoassay; MUI, median urinary iodine. ent assay systems, the RIs of thyroid functions might be different to some extent in differenct laboratories. In conclusion, the present study established the new RIs of serum thyroid hormones in a previously iodinedeficient but presently more than iodine-adequate area of Western China, which were different from the data provided by the manufacturers. Our present results provide valuable references for the diagnosis of thyroid diseases in populations of Western China. Considering the fact that most inland areas of China have faced the challenge of the transition from iodine deficiency to adequacy or more than adequacy, we recommend physicians utilize our RIs to determine thyroid diseases in the similar areas with Xi an in China. Acknowledgment We thank all physicians, laboratorians and participants of the study, for their co-operations and generous participations. Disclosure This study was supported by the Chinese Medical Association Foundation and Chinese Diabetes Society. QJ was supported by the Natural Science Foundation of Shaanxi Province, China (Grant No.2013KTZB02-01). The authors declare that they have no competing financial interests.

7 Reference intervals of thyroid hormones 387 References 1. Brabant G, Beck-Peccoz P, Jarzab B, Laurberg P, Orgiazzi J, et al. (2006) Is there a need to redefine the upper normal limit of TSH? Eur J Endocrinol Laurberg P, Andersen S, Carle A, Karmisholt J, Knudsen N, et al. (2011) The TSH upper reference limit where are we at? Nat Rev Endocrinol Taylor PN, Razvi S, Pearce SH, Dayan CM (2013) Clinical review A review of the clinical consequences of variation in thyroid function within the reference range. J Clin Endocrinol Metab Zou S, Wu F, Guo C, Song J, Huang C, et al. (2012) Iodine nutrition and the prevalence of thyroid disease after salt iodization a cross-sectional survey in Shanghai, a coastal area in China. PLoS One 7 e Sun X, Shan Z, Teng W (2014) Effects of increased iodine intake on thyroid disorders. Endocrinol Metab (Seoul) Yu Z (1988) The prevention and treatment survey of iodine deficiency disorders in Shaanxi province. Endemic Diseases Bulletin (In Chinese). 7. Wei C, Jun-feng G, Ling J, Ping L, Gang Y, et al. (2013) Analysis on the iodine nutrition status of key population in iodine deficiency disorders areas in Xi an. Modern Preventive Medicine (In Chinese). 8. Ling J, Wei C, Gang Y, Ping L, Lu D (2015) Analysis on monitoring results of iodine deficiency disorders in Xi an city in Clin J Endemiol (In Chinese). 9. WHO/UNICEF/ICCIDD (2007) Assessment of iodine deficiency disorders and monitoring their elimination, a guide for program managers (3 rd ed). GenevaWHO. (http// 10. Zhao W, Han C, Shi X, Xiong C, Sun J, et al. (2014) Prevalence of goiter and thyroid nodules before and after implementation of the universal salt iodization program in mainland China from 1985 to 2014 a systematic review and meta-analysis. PLoS One 9 e Guan H, Shan Z, Teng X, Li Y, Teng D, et al. (2008) 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 (Oxf) Li C, Guan H, Teng X, Lai Y, Chen Y, et al. (2011) An epidemiological study of the serum thyrotropin reference range and factors that influence serum thyrotropin levels in iodine sufficient areas of China. Endocr J Quinn FA, Tam MC, Wong PT, Poon PK, Leung MS (2009) Thyroid autoimmunity and thyroid hormone reference intervals in apparently healthy Chinese adults. Clin Chim Acta Yang W, Lu J, Weng J, Jia W, Li J, et al. (2010) Prevalence of diabetes among men and women in China. N Engl J Med Ming J, Xu S, Yang C, Gao B, Wan Y, et al. (2014) Metabolic syndrome and chronic kidney disease in general Chinese adults results from the China National Diabetes and Metabolic Disorders Study. Clin Chim Acta Kratzsch J, Fiedler GM, Leichtle A, Brugel M, Buchbinder S, et al. (2005) New reference intervals for thyrotropin and thyroid hormones based on National Academy of Clinical Biochemistry criteria and regular of the thyroid. Clin Chem Volzke H, Alte D, Kohlmann T, Ludemann J, Nauck M, et al. (2005) Reference intervals of serum thyroid function tests in a previously iodine-deficient area. Thyroid O Leary PC, Feddema PH, Michelangeli VP, Leedman PJ, Chew GT, et al. (2006) Investigations of thyroid hormones and antibodies based on a community health survey the Busselton thyroid study. Clin Endocrinol (Oxf) Hamilton TE, Davis S, Onstad L, Kopecky KJ (2008) Thyrotropin levels in a population with no clinical, autoantibody, or ultrasonographic evidence of thyroid disease Implications for the diagnosis of subclinical hypothyroidism. J Clin Endocr Metab Yoshihara A, Noh JY, Ohye H, Sato S, Sekiya K, et al. (2011) Reference limits for serum thyrotropin in a Japanese population. Endocr J Marwaha RK, Tandon N, Ganie MA, Mehan N, Sastry A, et al. (2013) Reference range of thyroid function (FT3, FT4 and TSH) among Indian adults. Clin Biochem Sriphrapradang C, Pavarangkoon S, Jongjaroenprasert W, Chailurkit LO, Ongphiphadhanakul B, et al. (2014) Reference ranges of serum TSH, FT4 and thyroid autoantibodies in the Thai population the national health examination survey. Clin Endocrinol (Oxf) Demers LM, Spencer CA (2003) Laboratory medicine practice guidelines laboratory support for the diagnosis and monitoring of thyroid disease. Clin Endocrinol (Oxf) Surks MI, Boucai L (2010) Age- and race-based serum thyrotropin reference limits. J Clin Endocrinol Metab Kussmaul T, Greiser KH, Haerting J, Werdan K, Thiery J, et al. (2014) Thyroid analytes TSH, FT3 and FT4 in serum of healthy elderly subjects as measured by the Roche modular system do we need age and gender dependent reference levels? Clin Lab Surks MI, Hollowell JG (2007) Age-specific distribu-

8 388 Cai et al. tion of serum thyrotropin and antithyroid antibodies in the US population implications for the prevalence of subclinical hypothyroidism. J Clin Endocrinol Metab Spencer CA, Hollowell JG, Kazarosyan M, Braverman LE (2007) National Health and Nutrition Examination Survey III thyroid-stimulating hormone (TSH)- thyroperoxidase antibody relationships demonstrate that TSH upper reference limits may be skewed by occult thyroid dysfunction. J Clin Endocrinol Metab Jonklaas J, Sathasivam A, Wang H, Gu J, Burman KD, et al. (2014) Total and free thyroxine and triiodothyronine measurement discrepancies, particularly in inpatients. Clin Biochem Jonklaas J, Davidson B, Bhagat S, Soldin SJ (2008) Triiodothyronine levels in athyreotic individuals during levothyroxine therapy. JAMA Jonklaas J, Kahric-Janicic N, Soldin OP, Soldin SJ (2009) Correlations of free thyroid hormones measured by tandem mass spectrometry and immunoassay with thyroid-stimulating hormone across 4 patient populations. Clin Chem van Deventer HE, Mendu DR, Remaley AT, Soldin SJ (2011) Inverse log-linear relationship between thyroidstimulating hormone and free thyroxine measured by direct analog immunoassay and tandem mass spectrometry. Clin Chem van Deventer HE, Soldin SJ (2013) The expanding role of tandem mass spectrometry in optimizing diagnosis and treatment of thyroid disease. Adv Clin Chem Gounden V, Jonklaas J, Soldin SJ (2014) A pilot study subclinical hypothyroidism and free thyroid hormone measurement by immunoassay and mass spectrometry. Clin Chim Acta