Reference Intervals of Thyroid Function During Pregnancy: Self-Sequential Longitudinal Study Versus Cross-Sectional Study

Abstract

Background:

A self-sequential longitudinal reference interval may be expected to minimize the inter-individual variation of thyroid function. Comparison between the self-sequential longitudinal reference interval (SLRI) and cross-sectional reference interval (CSRI) in pregnancy has not been well investigated. The objectives of this study were to establish a stringent SLRI of thyroid function in pregnant women and to compare it with the conventional CSRI.

Methods:

Three cohorts were enrolled: group 1, pregnant women for an SLRI (n = 99); group 2, pregnant women for a CSRI (n = 1318); group 3, non-pregnant control women (NC) as a control group (n = 301) according to the criteria of the National Academy of Clinical Biochemistry. Thyrotropin (TSH), total thyroxine (TT4), free thyroxine (fT4), total triiodothyronine (TT3), free triiodothyronine (fT3), serum ferritin (SF), and urine iodine concentration (UIC) were measured in the three groups.

Results:

Compared with CSRI, the reference interval of the SLRI group had narrower reference intervals of fT4 in the first and second trimesters (p < 0.05). The median of TSH was at a low level during the first trimester, and then gradually elevated in the second and third trimesters. The median of fT4 persistently decreased from 12 weeks, and did not return to the level of the NC group until 12 months postpartum. The TT4 increased to 131.4 nmol/L at gestational week 8, and reached a peak (170.0 nmol/L) at gestational week 12. In the first trimester, the prevalence of hypothyroxinemia was 9.1%, 4.0%, and 2.0% with a fT4 value below the 10th, 5th, and 2.5th percentile, respectively. In contrast, 29.3% of TT4 values were below the lower non-pregnancy reference limit multiplied by 1.5.

Conclusions:

No significant difference was found between a SLRI and a CSRI, even in a stringent self-sequential longitudinal reference interval of thyroid function in pregnant women. In addition, the limit of TT4 below the non-pregnant level multiplied by a factor 1.5 is not appropriate for diagnosing hypothyroxinemia in the first trimester.

Introduction

Adaptive physiological alterations cause many changes in thyroid hormone levels during pregnancy in the mother (1), ignorance of which leads to misdiagnosis of thyroid diseases. Stricker et al. (2) reported that 5.6–18.3% of misdiagnoses and missed diagnoses likely occur in clinical practice due to the use of non-pregnant reference intervals of thyroid function as the diagnostic criteria during pregnancy. Similar results were reported in China (3), Malaysia (4), and Australia (5). If subclinical hypothyroidism is defined as a thyrotropin (TSH) >2.5 mIU/L according to the American Thyroid Association (ATA) guidelines, >15% of pregnant women in the United States and Netherlands (6), and 27.8% women in China would be diagnosed with gestational thyroid disease (7), a prevalence that is much higher than the 2–3% estimated by the ATA guidelines (8). The prevalence of misdiagnosis for thyroid disorders in pregnancy decreases by using gestation-specific reference intervals (9). However, most published studies were based on a cross-sectional reference interval (CSRI) (2,10,11). A self-sequential longitudinal reference interval (SLRI) may be expected to minimize the inter-individual variation of thyroid function, which is difficult to establish. The present study attempted to establish a stringent SLRI and to compare its value with the conventional CSRI.

The gestation-specific reference intervals of thyroid function are probably also influenced by factors such as ethnicity, inter- or intra-individual variation, assay methods, iron deficiency, iodine deficiency, and body mass index (BMI) (12 –16). It was recently reported that iron deficiency is an independent risk factor for isolated hypothyroxinemia not only in pregnant women but also in non-pregnant women (16). Approximately 15% of women of reproductive age have iodine deficiency, even in the United States, which is generally an iodine-sufficient region (17). The incidence of isolated hypothyroxinemia in an overweight population was 3.64-fold compared with a normal-weight control population (18). Therefore, in the present study, a SLRI was established from eight weeks of gestation to 12 months postpartum in a pregnant population in whom iodine deficiency, iron deficiency, overweight, and positive thyroid antibodies were excluded compared with the conventional CSRI.

Materials and Methods

Subjects

The Subclinical Hypothyroid in Early Pregnancy (SHEP) study was conducted in Liaoning Province in China. The present study is part of the SHEP study. From August 2011 to June 2014, departments of obstetrics and gynecology and departments of endocrinology of 19 hospitals participated in this study. All participants lived in an iodine-sufficient region (median urine iodine concentration [UIC] = 160.2 μg/L in school-age children). Three cohorts were enrolled: group 1, pregnant women for a SLRI (n = 99); group 2, pregnant women for a CSRI (n = 1318) in whom 975 women were in the first trimester, 252 in the second trimester, and 91 in the third trimester; and group 3, non-pregnant women as a control group (n = 301). Recruitment criteria for pregnant women included: (i) aged 19–40 years; (ii) residing in the city for >10 years; and (iii) a singleton pregnancy at four to eight weeks. Exclusion criteria included: (i) multiple pregnancies; (ii) a personal or family history of thyroid disease; (iii) visible or palpable goiter confirmed by ultrasound; and (iv) patients on any medical regimen that may affect thyroid function, such as glucocorticoids, dopamine, or antiepileptic drugs. In the SLRI group, positive thyroid autoantibodies, overweight (BMI >25 kg/m²), iron deficiency (serum ferritin <12 μg/L), and iodine deficiency (median UIC <150 μg/L) were also exclusion criteria. In contrast to the strictly selected SLRI group, the above factors were not exclusion criteria in the CSRI group in order to reflect the real clinical situation.

Methods

In the SLRI group, samples were collected at nine time points (8, 12, 16, 20, 28, and 36 gestational weeks and 3, 6, and 12 months postpartum). The follow-up rate was 78.8% (78/99). In the CSRI group, samples were collected at different gestational weeks. In the non-pregnant group, only one sample was collected. Spot urine and blood samples were obtained in the morning after an overnight fast. Samples were stored at −20°C until shipped, and assayed in a centralized laboratory within one week. Additionally, 120 samples of school-age children were tested for UIC to assess the iodine status.

Serum TSH, free thyroxine (fT4), thyroperoxidase antibodies (TPOAb), thyroglobulin autoantibodies (TgAb), and serum ferritin (SF) were measured in all participants using electrochemiluminescence immunoassays with Cobas Elecsys 601 (Roche Diagnostics, Switzerland). The functional sensitivity of serum TSH was 0.002 mIU/L. The intra-assay coefficients of variation (CV) of serum TSH, fT4, TPOAb, TgAb, and SF were 1.57–4.12%, 2.24–6.33%, 2.42–5.63%, 1.3–4.9%, and 1.43–4.52%, respectively. The interassay CV values were 1.26–5.76%, 4.53–8.23%, 5.23–8.16%, 2.1–6.9%, and 3.52–7.91%, respectively. The laboratory reference ranges for TSH were 0.27–4.2 mIU/L, fT4 12–22 pmol/L, TPOAb 0–34 IU/mL, TgAb 0–115 IU/mL, and SF 12–150 μg/L. UIC was determined in all participants by the ammonium persulfate method based on the Sandell–Kolthoff reaction. The intra- and inter-assay CV for UIC were 3–4% and 4–6% at 66 μg/L and 2–5% and 3–6% at 230 μg/L, respectively.

Statistical analysis

The data were tested for normality using the Kolmogorov–Smirnov test. Categorical variables are expressed as frequencies and percentages. In the case of a normal distribution, continuous data are presented as mean value ± standard deviation. Otherwise, median and interquartile ranges are reported. The limits of the reference intervals were calculated as median, 2.5th percentile, 97.5th percentile. The Mann–Whitney U-test and Kruskal–Wallis test were performed for comparing continuous data that did not conform to normal distribution. Student's t-test and one-way analysis of variance were conducted to evaluate differences between quantitative variables that were normally distributed. Spearman's correlation test was used to examine the correlation between non-normality variables, and the Pearson's correlation test was used to examine the correlation between normality variables. All statistical analyses were performed using SPSS Statistics for Windows v20.0 (IBM Corp., Armonk, NY). A p-value of <0.05 was considered as statistically significant. G*Power was used to perform the power analysis, and most power factors were >80% in order to support the reliability of the results.

Ethics committee approval

All research protocols were approved by the Medical Ethics Committee of China Medical University, and were congruent with the declaration of Helsinki. Informed consent was obtained from every patient.

Results

Iodine status and characteristics of the reference population

The clinical characteristics of the reference population are summarized in Table 1. According to the historical data, the three cohorts lived in an iodine-sufficient region. The median UIC in school-age children was 160.2 μg/L; the median UIC of pregnant women before eight weeks of gestation was 167.78 μg/L; and the median UIC in child-bearing age women was 177.36 μg/L.

Table 1.

Clinical Characteristics of Different Groups

	N	Age, years	Gestational age, weeks	Singleton pregnancy, %	Iodized salt use, %	BMI, kg/m ²	UIC, μg/L	Tg, ng/mL	SF, μg/L
SLRI group	99	27 (25, 30)	7.71 (6.71, 8.14)	100	100	20.90 (19.22, 22.58)	207.90 (180.79, 285.41)	11.25 (6.24, 18.56)	75.81 (43.39, 100.20)
CSRI group
T1	975	29 (27, 31)	7.00 (6.00, 7.71)	100	100	—	167.78 (120.38, 234.19)	10.46 (6.10, 16.59)	68.64 (435.57, 102.75)
T2	252	29 (27, 31)	17.00 (14.28, 20.00)	100	100	—	132.52 (86.02, 179.64)	—	48.17 (29.91, 73.83)
T3	91	30 (28, 32)	30.43 (30.00, 31.29)	100	100	—	104.36 (73.89, 158.32)	—	19.22 (9.81, 27.75)
Non-pregnant group	301	28 (26, 31)	—	—	100	20.96 (19.72, 22.46)	177.36 (149.90, 208.23)	—	58.55 (33.21, 83.68)

Values shown as median and IQR (25th, 75th).

SLRI, self-sequential longitudinal reference interval; CSRI, cross-sectional reference interval; BMI, body mass index; UIC, urinary iodine concentration; Tg, thyroglobulin; SF, serum ferritin.

Gestation-specific reference intervals of thyroid function in the SLRI group

Table 2 shows the trends of the SLRI throughout gestation and postpartum, and compares it with non-pregnant levels. The median concentrations of TSH decreased significantly at eight weeks of gestation, reached a nadir (1.46 mIU/L) and then increased from then on before returning to the level of the non-pregnant state after childbirth (Fig. 1). fT4 levels decreased from 12 weeks of gestation, with a nadir in the third trimester, and did not return to the level of non-pregnancy until 12 months postpartum (Fig. 1B). Median concentrations of free triiodothyronine (fT3) decreased from eight weeks of gestation, and were lower than the non-pregnant levels during pregnancy at 12 months postpartum (p < 0.001; Fig. 1C). The median total thyroxine (TT4) increased slightly to 131.4 nmol/L (117% of the non-pregnant level) at eight weeks of gestation, and reached a level of 170.0 nmol/L (151% of the non-pregnant level) at 12 weeks of gestation. It decreased after delivery, and was lower than the non-pregnant level at 12 months postpartum (p < 0.001; Fig. 1D). The median total triiodothyronine (TT3) concentration rose continuously from eight weeks of gestation and remained stable during gestation. It decreased postpartum, and returned to the non-pregnant level at three months postpartum (Fig. 1E).

FIG. 1.

Dynamic changes of (A) TSH, (B) fT4, (C) fT3, (D) TT4, and (E) TT3 group. For all graphs, the top line is the upper reference limit (the 97.5th percentile), the middle line is the median level (the 50th percentile), the bottom line is the lower reference limit (the 2.5th percentile), and the gray dotted straight line is the non-pregnant level. SLRI, self-sequential longitudinal reference interval; M, median; 2.5th, 2.5th percentile; 97.5th, 97.5th percentile. TSH, thyrotropin; fT4, free thyroxine; fT3, free triiodothyronine; TT3, total triiodothyronine; TT4, total thyroxine. G8, 8th week of gestation; G12, 12th week of gestation; G16, 16th week of gestation; G20, 20th week of gestation; G28, 28th week of gestation; G36, 36th week of gestation. P3, 3rd month of postpartum; P6, 6th month of postpartum; P12, 12th month of postpartum.

Table 2.

Gestational Age–Specific Reference Intervals of Thyroid Function (Including the 3rd, 6th, and 12th Month Postpartum) Compared with Non-Pregnant Levels

		TSH, mIU/L					fT4, pmol/L					fT3, pmol/L					TT4, nmol/L					TT3, nmol/L
		Percentile					Percentile					Percentile					Percentile					Percentile
Gestational week or postpartum	N	2.5th	50th	97.5th	p^a	p^b	2.5th	50th	97.5th	p^a	p^b	2.5th	50th	97.5th	p^a	p^b	2.5th	50th	97.5th	p^a	p^b	2.5th	50th	97.5th	p^a	p^b
Non-pregnant	301	0.78	2.04	5.03	—	—	12.46	15.76	19.58	—	—	4.09	4.93	6.02	—	—	81.01	112.40	149.04	—	—	1.49	1.89	2.43	—	—
G8	99	0.35	1.65	4.13	—	<0.001	13.33	16.06	19.01	—	0.110	3.42	4.17	5.09	—	<0.001	89.11	131.40	195.75	—	<0.001	1.40	1.95	2.93	—	0.003
G12	87	0.06	1.46	3.33	0.187	<0.001	12.14	15.23	18.79	<0.001	0.043	3.45	4.23	5.31	0.618	<0.001	128.50	170.00	247.40	<0.001	<0.001	1.75	2.51	3.28	<0.001	<0.001
G16	85	0.37	1.87	4.78	<0.001	0.019	10.45	13.48	16.22	<0.001	<0.001	3.16	4.07	5.08	0.042	<0.001	103.36	168.95	222.58	0.728	<0.001	1.88	2.71	3.64	0.003	<0.001
G20	84	0.67	1.98	3.93	0.886	0.147	9.02	12.11	15.60	<0.001	<0.001	3.09	3.88	4.86	<0.001	<0.001	104.78	167.40	214.89	0.038	<0.001	1.82	1.67	2.15	0.485	<0.001
G28	84	0.55	1.79	3.75	0.438	0.017	7.47	10.86	14.97	<0.001	<0.001	2.90	3.52	4.53	<0.001	<0.001	99.70	155.90	198.35	0.002	<0.001	1.81	2.75	3.60	0.966	<0.001
G36	73	0.66	1.93	4.23	0.010	0.542	8.29	10.79	14.41	0.582	<0.001	2.93	3.52	4.39	0.094	<0.001	89.48	147.65	203.63	0.415	<0.001	1.81	2.71	3.76	0.131	<0.001
P3	65	0.16	1.89	5.12	0.423	0.051	10.56	14.55	30.63	<0.001	<0.001	3.66	4.39	9.66	<0.001	<0.001	73.78	106.80	177.16	<0.001	0.004	1.36	1.81	3.48	<0.001	0.837
P6	69	0.05	2.07	11.07	0.560	0.501	9.10	13.93	21.14	0.899	<0.001	3.28	4.29	5.76	0.593	<0.001	71.31	102.10	148.94	0.217	<0.001	1.37	1.81	2.38	0.074	0.123
P12	78	0.69	2.02	4.96	0.942	0.288	12.00	15.38	19.35	0.001	0.155	3.45	4.64	5.71	0.015	<0.001	77.21	102.10	144.11	0.627	<0.001	1.43	1.83	2.53	0.695	0.557

Compared with the upper gestation week.

Compared with the nonpregnant group.

TSH, thyrotropin; fT4, free thyroxine; fT3, free triiodothyronine; TT4, total thyroxine; TT3, total triiodothyronine.

Comparison of SLRI and CSRI

The reference intervals of thyroid function in the SLRI and CSRI groups (median, 2.5th percentile, 97.5th percentile) are listed in Table 3. No significant differences in reference intervals of TSH existed between the SLRI and CSRI groups (first trimester 0.35–4.13 mIU/L vs. 0.35–4.47 mIU/L; second trimester 0.37–4.78 mIU/L vs. 0.07–4.52 mIU/L; third trimester 0.55–3.75 mIU/L vs. 0.75–3.77 mIU/L, respectively; p > 0.05). In the first and second trimesters, the reference intervals of fT4 in the SLRI group were narrower than those in the CSRI group (first trimester 13.35–19.01 pmol/L vs. 13.10–20.31 pmol/L, p < 0.01; second trimester 10.45–16.22 pmol/L vs. 9.17–17.74 pmol/L, p < 0.05). However, in the third trimester, no difference was found in the reference intervals of fT4 between the SLRI and CSRI groups (7.47–14.97 pmol/L vs. 6.94–13.38 pmol/L, p > 0.05).

Table 3.

Reference Intervals of TSH and fT4 Established by SLRI and CSRI Groups

	T1		T2		T3
	n	M (P2.5, P97.5)	n	M (P2.5, P97.5)	n	M (P2.5, P97.5)
TSH (mIU/L)
SLRI group	99	1.65 (0.35, 4.13)	85	1.87 (0.37, 4.78)	84	1.79 (0.55, 3.75)
CSRI group	957	1.84 (0.35, 4.47)	252	1.63 (0.07, 4.52)	91	1.71 (0.75, 3.77)
fT4 (pmol/L)
SLRI group	99	16.06 (13.35, 19.01)^*	85	13.48 (10.45, 16.22)^**	84	10.86 (7.47, 14.97)
CSRI group	957	16.52 (13.10, 20.31)	252	12.91 (9.17, 17.74)	91	10.92 (6.94, 13.38)

Compared with the CS group, p < 0.01.

Compared with the CS group, p < 0.05.

T1, first trimester; T2, second trimester; T3, third trimester; M, median; P2.5, 2.5th percentile; P97.5, 97.5th percentile.

Diagnosis of hypothyroxinemia in the first trimester

Hypothyroxinemia in the first trimester was defined as a fT4 value below the 10th, 5th, and 2.5th percentile (fT4 <14.15 pmol/L, <13.67 pmol/L, and <13.33 pmol/L, respectively, in the first trimester in the present study), or a TT4 value below the lower non-pregnant reference limit multiplied by 1.5 (TT4 <1.5 × 81.01 pmol/L in the present study) with a concomitant normal TSH level. In the SLRI group, the prevalence of hypothyroxinemia in the first trimester was 9.1%, 4.0%, and 2.0% based on fT4 criteria, respectively, which was much lower than a finding of 29.3% based on the TT4 criteria (p < 0.000, p = 0.000, and p = 0.000; Table 4).

Table 4.

Prevalence of Hypothyroxinemia on Different Centiles of fT4

Gestation week or postpartum	2.5th of T4 (pmol/L)	Hypothyroxinemia on 2.5th of fT4,% (n)	5th of T4 (pmol/L)	Hypothyroxinemia on 5th of fT4, % (n)	10th of T4 (pmol/L)	Hypothyroxinemia on 10th of fT4,% (n)	Hypothyroxinemia on TT4, % (n)
G8	13.33	2.0% (2/99)	13.67	4.0% (4/99)	14.15	9.1% (9/99)	29.3% (29/99)
G12	12.14	2.3% (2/87)	12.44	4.6% (4/87)	13.03	9.2% (8/87)	1.1% (1/87)
G16	10.45	2.4% (2/85)	10.83	4.7% (4/85)	11.44	9.4% (8/85)	2.4% (2/82)
G20	9.02	1.2% (1/84)	9.14	4.8% (4/84)	10.18	9.5% (8/84)	8.4% (7/83)
G28	7.47	2.4% (2/84)	8.32	4.8% (4/84)	8.72	9.5% (8/84)	15.7% (13/83)
G36	8.29	1.4% (1/73)	8.49	4.1% (3/73)	9.03	9.6% (7/73)	13.9% (10/72)
P3	10.56	1.5% (1/65)	11.22	4.6% (3/65)	12.06	9.2% (6/65)	83.6% (51/61)
P6	9.10	0% (0/69)	10.08	2.9% (2/69)	11.98	7.2% (5/69)	82.3% (56/68)
P12	12.00	1.2% (1/78)	12.65	3.8% (3/78)	13.11	9.0% (7/78)	82.7% (62/75)

Discussion

This study demonstrates a stringent SLRI of thyroid function throughout pregnancy (from eight weeks of gestation to 12 months postpartum), and compares it with the conventional CSRI in an iodine-sufficient region in the northeast of China. A SLRI may be expected to minimize the inter-individual variation of thyroid function. However, in this study, there was no significant difference between the SLRI and the CSRI, which is inconsistent with the findings of some other studies. Andersen et al. (19) reported that the intra-individual variation of thyroid function tests in non-pregnant individuals was smaller than the inter-individual variation. Boas et al. (20) drew a similar conclusion. Yu et al. (21) calculated longitudinal reference intervals in the Chinese population, and demonstrated narrower intervals compared with the cross-sectional studies. Wang et al. (3) compared a SLRI with a CSRI and suggested no significant differences in most reference intervals. However, the interquartile range of a SLRI was usually smaller than that of a CSRI, especially in the first trimester. In the present study, the reference intervals of fT4 were narrower in the first and second trimesters in SLRI than those in CSRI were. However, in the third trimester, there was no significant difference in the width of reference intervals between the two groups. Although a SLRI may be expected to minimize the inter-individual variation of thyroid function, the present study suggests that no significant differences exist between SLRI and CSRI, even in a strictly selected pregnant population.

In addition, the CSRI was also influenced by different assay methods (22). The liquid chromatography tandem mass-spectrometry (LC/MS) method improved the accuracy of free measurements during pregnancy due to its higher reliability, reproducibility, and specificity (23). However, fT4 measured by LC/MS is not widely used because of technical difficulties and high costs (24). Compared with LC/MS, immunoassays are easier to implement. The free hormone index (fT4I) method indirectly estimates free hormone concentrations with two separate measurements and corrects the total hormone concentration for the influence of abnormal binding proteins, primarily thyroxine-binding globulin (TBG), based on a simple calculation (25). The fT4I method is less convenient than immunoassays and not widely used in clinical practice. However, some authors question the accuracy of immunoassays (12,22) because immunoassays do not test the serum fT4 directly, and they are influenced by changes in binding proteins such as albumin and free fatty acids during pregnancy (4). fT4 values measured by LC/MS correlate generally to a greater degree with logTSH values compared with values measured by immunoassays (26). Sapin et al. (27) reported that significant inaccuracies were observed in fT4 by immunoassays due to alterations in binding proteins. Kahric-Janicic et al. (23) suggested that fT4 concentrations determined by immunoassays were not in line with those determined by LC/MS. fT4 decreased with each successive trimester when measured by LC/MS, although it increased after the second trimester when measured by immunoassays. However, Anckaert et al. (28) compared fT4 results by three different immunoassays with those obtained by LC/MS assays, and proved that two of the immunoassays gave a consistent pattern with those by LC/MS assays. In the present study, fT4 concentrations measured by immunoassays decreased with progression of trimesters and displayed a consistent tendency with that obtained by LC/MS assays in the study by Kahric et al. (23) (Fig. 2). Moreover, fT4 measured by immunoassays negatively correlated with TSH in the first trimester in the present study (r = −0.201, p < 0.01), which suggests immunoassays could be used for measuring fT4 during pregnancy.

FIG. 2.

Comparison of fT4 by immunoassay with isotope dilution liquid chromatography tandem mass spectrometry (23). Box plot representation of the descriptive statistics for fT4. The square depicts the mean, the line shows the median, the top and bottom of the box represent the 75% and 25% percentile, respectively, and the whiskers show the 90th and 10th percentile.

Hypothyroxinemia is defined as a normal TSH concentration in conjunction with fT4 concentrations below the 10th, 5^th, or 2.5th percentile of the reference range (29). With a fT4 <10th percentile as the diagnostic criterion of hypothyroxinemia, the prevalence of mild hypothyroxinemia was 8.5% (30), which was consistent with 9.1% detected in this study. It has also been suggested to use TT4 to evaluate pregnant hypothyroxinemia, provided that the non-pregnant reference limits are adjusted by a factor of 1.5 (31). However, in the present study, with the lower limit of TT4 below the non-pregnant level multiplied by a factor 1.5 as the diagnostic criterion of hypothyroxinemia in the first trimester, the prevalence of hypothyroxinemia was 29.3%, which is much higher than that found in other reports (30). The explanation for the higher prevalence for hypothyroxinemia is that the median concentrations of TT4 values only increase by about 1.5 times by 12 weeks of gestation. However, in the present study, most pregnant women were screened before eight weeks of gestation, which led to an overdiagnosis for hypothyroxinemia.

In summary, a SLRI from eight weeks of gestation to 12 months postpartum in Chinese pregnant women with sufficient iodine intake, normal iron status, and normal BMI has been established, and a CSRI was collected as a control. Although a SLRI may be expected to minimize the inter-individual variation of thyroid function, there was no significant difference between the SLRI and the CSRI, even in a strictly selected pregnant population. fT4 measurements by immunoassay were consistent with that obtained by LC/MS in the diagnosis of hypothyroxinemia. In addition, the use of a lower TT4 reference limit multiplied by 1.5 is not appropriate for diagnosing hypothyroxinemia in the first trimester.

Footnotes

Acknowledgments

We gratefully acknowledge the residents who participated in this study. This work was supported by National Science and Technology Support Program (2014BAI06B02); Chinese National Natural Science Foundation (81170730); Research Foundation of Key Laboratory of Endocrine Diseases, Department of Science and Technology, Shenyang government, Liaoning province, China (F11-244-1-00); Liaoning science public research funds (2014001001/GY2014-A-001); key platform foundation of science and technology for the universities in Liaoning Province (16010).

Author Disclosure Statement

The authors have no potential conflict of interest to declare.

References

Glinoer

. 1997. The regulation of thyroid function in pregnancy: pathways of endocrine adaptation from physiology to pathology. Endocr Rev, 18:404–433.

Stricker

, Echenard

, Eberhart

, Chevailler

, Perez

, Quinn

, Stricker

. 2007. Evaluation of maternal thyroid function during pregnancy: the importance of using gestational age-specific reference intervals. Eur J Endocrinol, 157:509–514.

Wang

, Yu

, Huang

, Cao

, Zhu

, Sun

, Zhou

. 2011. Assessment of thyroid function during pregnancy: the advantage of self-sequential longitudinal reference intervals. Arch Med Sci, 7:679–684.

Shan

, Chen

, Teng

, Yu

, Li

, Zhou

, Gao

, Zhou

, Ding

, Ma

, Wu

, Liu

, Xu

, Liu

, Li

, Wang

, Li

, Fan

, Wang

, Guo

, Zhang

. 2009. A study for maternal thyroid hormone deficiency during the first half of pregnancy in China. Eur J Clin Invest, 39:37–42.

Thevarajah

, Chew

, Lim

, Sabir

, Sickan

. 2009. Determination of trimester specific reference intervals for thyroid hormones during pregnancy in Malaysian women. Malays J Pathol, 31:23–27.

Negro

, Stagnaro-Green

. 2014. Diagnosis and management of subclinical hypothyroidism in pregnancy. BMJ, 349:g4929.

, Shan

, Mao

, Wang

, Xie

, Zhou

, Li

, Xu

, Bi

, Meng

, Du

, Zhang

, Gao

, Zhang

, Yang

, Fan

, Teng

. 2014. Assessment of thyroid function during first-trimester pregnancy: what is the rational upper limit of serum TSH during the first trimester in Chinese pregnant women?. J Clin Endocrinol Metab, 99:73–79.

Casey

, Dashe

, Wells

, McIntire

, Byrd

, Leveno

, Cunningham

. 2005. Subclinical hypothyroidism and pregnancy outcomes. Obstet Gynecol, 105:239–245.

Soldin

, Hilakivi-Clarke

, Weiderpass

, Soldin

. 2004. Trimester-specific reference intervals for thyroxine and triiodothyronine in pregnancy in iodine-sufficient women using isotope dilution tandem mass spectrometry and immunoassays. Clin Chim Acta, 349:181–189.

10.

Soldin

, Soldin

, Sastoque

. 2007. Gestation-specific thyroxine and thyroid stimulating hormone levels in the United States and worldwide. Ther Drug Monit, 29:553–559.

11.

La'ulu

, Roberts

. 2007. Second-trimester reference intervals for thyroid tests: the role of ethnicity. Clin Chem, 53:1658–1664.

12.

Duntas

, Biondi

. 2013. The interconnections between obesity, thyroid function, and autoimmunity: the multifold role of leptin. Thyroid, 23:646–653.

13.

Rotondi

, Magri

, Chiovato

. 2011. Thyroid and obesity: not a one-way interaction. J Clin Endocrinol Metab, 96:344–346.

14.

Biondi

. 2010. Thyroid and obesity: an intriguing relationship. J Clin Endocrinol Metab, 95:3614–3617.

15.

Shi

, Han

, Li

, Mao

, Wang

, Xie

, Li

, Xu

, Meng

, Du

, Zhang

, Gao

, Zhang

, Fan

, Shan

, Teng

. 2015. Optimal and safe upper limits of iodine intake for early pregnancy in iodine-sufficient regions: a cross-sectional study of 7,190 pregnant women in China. J Clin Endocrinol Metab, 100:1630–1638.

16.

, Shan

, Li

, Mao

, Wang

, Xie

, Liu

, Teng

, Zhou

, Li

, Xu

, Bi

, Meng

, Du

, Zhang

, Gao

, Zhang

, Yang

, Fan

, Teng

. 2015. Iron deficiency, an independent risk factor for isolated hypothyroxinemia in pregnant and non-pregnant women of childbearing age in China. J Clin Endocrinol Metab, 100:1594–1601.

17.

Caldwell

, Miller

, Wang

, Jain

, Jones

. 2008. Iodine status of the U.S. population, National Health and Nutrition Examination Survey 2003–2004. Thyroid, 18:1207–1214.

18.

Gowachirapant Melse-Boonstra

, Winichagoon

, Zimmermann

. 2014. Overweight increases risk of first trimester hypothyroxinaemia in iodine-deficient pregnant women. Matern Child Nutr, 10:61–71.

19.

Andersen

, Pedersen

, Bruun

, Laurberg

. 2002. Narrow individual variations in serum T(4) and T(3) in normal subjects: a clue to the understanding of subclinical thyroid disease. J Clin Endocrinol Metab, 87:1068–1072.

20.

Boas

, Forman

, Juul

, Feldt-Rasmussen

, Skakkebaek

, Hilsted

, Chellakooty

, Larsen

, Petersen

, Main

. 2009. Narrow intra-individual variation of maternal thyroid function in pregnancy based on a longitudinal study on 132 women. Eur J Endocrinol, 161:903–910.

21.

, Wang

, Huang

, Cao

, Zhu

, Sun

, Zhou

, Zhang

. 2010. Establishment of self-sequential longitudinal reference intervals of maternal thyroid function during pregnancy. Exp Bio Med, 235:1212–1215.

22.

Springer

, Zima

, Limanova

. 2009. Reference intervals in evaluation of maternal thyroid function during the first trimester of pregnancy. Eur J Endocrinol, 160:791–797.

23.

Kahric-Janicic

, Soldin

, West

, Gu

, Jonklaas

. 2007. Tandem mass spectrometry improves the accuracy of free thyroxine measurements during pregnancy. Thyroid, 17:303–311.

24.

Negro

, Soldin

, Obregon

, Stagnaro-Green

. 2011. Hypothyroxinemia and pregnancy. Endocr Pract, 17:1–24.

25.

Roberts

, La'ulu

, Roberts

. 2007. Performance characteristics of seven automated thyroxine and T-uptake methods. Clin Chim Acta, 377:248–255.

26.

Jonklaas

, Kahric-Janicic

, Soldin

. 2009. Correlations of free thyroid hormones measured by tandem mass spectrometry and immunoassay with thyroid-stimulating hormone across 4 patient populations. Clin Chem, 55:1380–1388.

27.

Sapin

, D'Herbomez

, Schlienger

. 2004. Free thyroxine measured with equilibrium dialysis and nine immunoassays decreases in late pregnancy. Clin Lab, 50:581–584.

28.

Anckaert

, Poppe

, Van Uytfanghe

, Schiettecatte

, Foulon

, Thienpont

. 2010. FT4 immunoassays may display a pattern during pregnancy similar to the equilibrium dialysis ID-LC/tandem MS candidate reference measurement procedure in spite of susceptibility towards binding protein alterations. Clin Chim Acta, 411:1348–1353.

29.

Stagnaro-Green

, Abalovich

, Alexander

, Azizi

, Mestman

, Negro

, Nixon

, Pearce

, Soldin

, Sullivan

, Wiersinga

. 2011. Guidelines of the American Thyroid Association for the diagnosis and management of thyroid disease during pregnancy and postpartum. Thyroid, 21:1081–1125.

30.

Henrichs

, Bongers-Schokking

, Schenk

, Ghassabian

, Schmidt

, Visser

, Hooijkaas

, de Muinck Keizer-Schrama

, Hofman

, Jaddoe

, Visser

, Steegers

, Verhulst

, de Rijke

, Tiemeier

. 2010. Maternal thyroid function during early pregnancy and cognitive functioning in early childhood: the generation R study. J Clin Endocrinol Metab, 95:4227–4234.

31.

Lazarus

, Brown

, Daumerie

, Hubalewska-Dydejczyk

, Negro

, Vaidya

. 2014. European Thyroid Association Guidelines for the management of subclinical hypothyroidism in pregnancy and in Children. Eur Thyroid J, 3:76–94.