Abstract
Background
Thyroid autoimmunity (TAI) may be present in 1–17% of pregnant women. Monitoring of thyroid function in euthyroid pregnant women positive for anti-thyroperoxidase antibodies (TPOAb+) is recommended.
Objective
To determine the prevalence and possible clinical and biological risk factors of biochemical progression (rise in serum thyroid-stimulating hormone (TSH) > 2.5 mU/L) at second blood sampling during pregnancy, in euthyroid women (TSH ≤ 2.5 mU/L) according to their TPOAb status.
Methods
This study included demographic and biological data from two previously published cohorts (n = 274 women from August 1996 to May 1997 Copenhagen cohort, and n = 66 women from January 2013 to December 2014 Brussels cohort) having at least two measurements of TSH and free thyroxine (FT4) and at least one of TPOAb during spontaneously achieved singleton pregnancies.
Results
The majority of women studied did not show biochemical progression. Only 4.2% progressed, significantly more frequently among TPOAb+ women, as compared to TPOAb− group (9.4 vs 2.7%, P = 0.015). No rise in serum TSH > 4 mU/L at 2nd sampling was observed. Higher baseline TSH levels were associated with biochemical progression in both TPOAb+ (P = 0.05) and TPOAb− women (P < 0.001), whereas maternal age, BMI, multiparity, smoking, FT4, and TPOAb concentrations were not significantly different between women with and without progression.
Conclusions
Only a minority of euthyroid women with thyroid autoimmunity presented biochemical progression and none with a TSH > 4 mU/L. Larger studies are needed to better target the subset of women who would benefit most from repeated thyroid function monitoring during pregnancy.
Introduction
Thyroid autoimmunity (TAI), defined as the presence of increased levels of circulating thyroid antibodies (anti thyroperoxidase (TPOAb) and/or anti-thyroglobulin (TgAb) antibodies), may be present in up to 1−17% of pregnant women, depending on the iodine intake in the investigated area and the ethnic background (1, 2, 3, 4, 5, 6). TAI is the strongest predictor for the development of subclinical or overt thyroid dysfunction in the preconception, during pregnancy, and in the postpartum period (5, 7, 8). A significant association between thyroid autoimmunity and first-trimester pregnancy loss as well as preterm birth has been demonstrated, although without proof of causality or precise underlying pathophysiological mechanisms (9, 10). However, evidence of levothyroxine (LT4) treatment efficacy on adverse pregnancy outcomes seems more conflicting (11, 12, 13). Recently conducted randomized clinical trials in euthyroid TPOAb-positive women, showed that LT4 initiated preconceptionally and continued throughout pregnancy did not decrease the risk of pregnancy loss or preterm delivery as compared to placebo, neither in spontaneous nor in vitro fertilization (IVF) obtained pregnancies (14, 15, 16).
Current guidelines from the American Thyroid Association (published in 2017) recommend that TPOAb-positive euthyroid pregnant women should have measurements of serum thyroid-stimulating hormone (TSH) every 4 weeks through the late 2nd trimester of pregnancy (5). This recommendation was based on two studies showing that euthyroid women with TAI have an increased risk of developing higher serum TSH values during gestation (17, 18). A subsequent study, similarly reported that among TPOAb-positive women with serum TSH < 2.5 mU/L in early pregnancy (n = 58), around 40% developed serum TSH 2.5−5 mU/L later in pregnancy (19). However, none of the above-mentioned studies determined potential risk factors for biochemical deterioration in euthyroid women such as age, parity, BMI, TPOAb concentration, or initial TSH cutoff. Targeting the subgroup of women who could benefit the most from thyroid function monitoring during pregnancy could help to better identify ideal candidates for treatment as well as avoid overtreatment of others.
Therefore, the aim of this study was to determine the prevalence of biochemical progression, defined as developing serum TSH > 2.5 mU/L at the second blood sampling during pregnancy, in TPOAb-positive vs TPOAb-negative women and initial TSH levels ≤ 2.5 mU/L. This study aimed as well to determine potential clinical and biological risk factors associated with biochemical progression.
Materials and methods
This study included data from two previously published cohorts of pregnant women having at least two measurements of TSH and free thyroxine (FT4) and at least one of TPOAb during spontaneously achieved singleton pregnancies (3, 20, 21).
Cohorts
The first cohort was a prospective longitudinal cohort of pregnant women attending prenatal care at Copenhagen University Hospital (Rigshospitalet), Denmark, from August 1996 to May 1997. A total of 1605 women participating in a pregnancy and stress factors project, filled in a health questionnaire including information regarding socioeconomic status, smoking habits, prescription medications, and parity. A subset of 315 consecutively included women with no regular cigarette smoking, alcohol consumption or drug abuse, no psychiatric illness, and no known chronic diseases requiring daily prescription medications was enrolled for clinical and biochemical monitoring (20). At that time point, it was not recommended to routinely screen for thyroid dysfunction in pregnancy; known thyroid disease was an exclusion criterion for the study cohort. They were followed with questionnaires (self-reported pre-pregnancy weight and height, smoking, drug, alcohol habits, medication, and parity), ultrasound scans, and sequential blood sampling once each trimester during pregnancy. Blood samples were frozen and analyzed in batches. Informed consent was obtained from each woman, and the project was approved by the Danish Data Protection Agency as well as the local branch of the Danish Ethics Committee (reference number 01-077/96). After exclusion of multiple pregnancies (n = 2), thyroxine intake during pregnancy (n = 2), pregnancies obtained by assisted reproductive technology (n = 10), unknown gestational age at first visit (n = 11), no thyroid antibody measurement (n = 8) and significant non-thyroidal disease (n = 8) a total of n = 274 women from this cohort (in this study, referred to as the Copenhagen cohort) were analyzed (Fig. 1). For the purpose of this study, the first and second available blood sampling during pregnancy were analyzed.
Methodology of patients’ selection in the study.
Citation: Endocrine Connections 13, 9; 10.1530/EC-24-0151
The second cohort was a prospectively collected cohort of pregnant women attending prenatal care at Centre Hospitalier Universitaire Saint-Pierre in Brussels, Belgium from 2 January 2013 to 31 December 2014 (3). On a routine basis at this center, at the first antenatal consultation, thyroid function (TSH and FT4) and TPOAb are systematically analyzed, and demographic (including height and weight measured at the time of visit, smoking habits, medication) and obstetric (including parity) data are collected. The study was approved by the Centre Hospitalier Universitaire Saint-Pierre review board (AK/15-11-114/4568). At that time point, it was recommended to follow 2012 ATA guidelines, suggesting thyroxine treatment in TPOAb+ women with first-trimester TSH values > 2.5 mU/L (22). Of 1832 women who had a first antenatal visit during the above-mentioned period, multiple pregnancies and pregnancies obtained by assisted reproductive technology (n = 130), and TPOAb− women for whom no second blood sampling for thyroid function during pregnancy was performed (n = 1572) were excluded. Out of the remaining 130 TPOAb+ women, after the exclusion of n = 35 women treated with thyroxine before/during pregnancy, and n = 40 women without 2nd thyroid function sampling available during pregnancy (no explaining reason could be found in the medical file), a total of n = 55 TPOAb+ with a 2nd blood sampling for TSH, FT4 performed during pregnancy were included and analyzed in the Brussels cohort (Fig. 1).
For both cohorts, gestational age was based on ultrasound findings and expressed in full weeks from the first day of the last menstrual period. Multiparity was defined as two or more prior pregnancies. The data that were recorded and analyzed in the present study included maternal age at the first visit, height, weight, body mass index (BMI), gestational age at first blood sampling, parity, serum TSH (mU/L) and FT4 (pmol/L) at first and second sampling during pregnancy, and the first available TPOAb measurement.
The TSH was stratified as follows; TSH ≤ 2.5 mU/L, 2.5 < TSH ≤ 4 mU/L and TSH > 4 mU/L. Biochemical progression during pregnancy was defined as TSH at 2nd sampling > 2.5 mU/L in women who had TSH ≤ 2.5 mU/L at 1st sampling. A comparison of clinical and biological characteristics was conducted between women who showed biochemical progression at 2nd sampling versus those who remained below the TSH ≤ 2.5 mU/L threshold both at 1st and 2nd sampling.
Laboratory methods
Laboratory methods for both analyzed cohorts have been previously published (3, 20). For the Copenhagen cohort, the AutoDelfia automatic fluoro-immunoassay system (EEG, Turku, Finland) was used to analyze TSH and FT4, while TPOAb was analyzed by radioimmunoassay (DYNOtest, Brahms, Hennigsdorf, Germany). The inter-assay coefficients of variance (CV) were 4.8, 2.2, and 2.2% at the concentrations of 0.05, 0.9, and 17.6 mU/L, respectively, for TSH; and 5.3, 3.7, and 3.1% for 9.3, 15.9, and 19.5 pmol/L, respectively, for FT4. Reference ranges for TSH and FT4 according to the gestational age was calculated and previously published (i.e. 2.5−97.5% range for TSH for 15−20-week pregnancy was 0.43−3.95 mU/L, 2.5−97.5% range for FT4 for 15−20-week pregnancy was 7.62−11.56 pmol/L) (20). For the Brussels cohort, serum TSH, FT4, and TPOAb levels were measured by the Chemiluminescence Centaur XP Siemens immunoanalyser (Siemens Healthcare Diagnostics). The reference values for nonpregnant women were 0.3 to 4.0 mIU/L and 10.3 to 25.7 pmol/L, for TSH and FT4, respectively. The total imprecision coefficients of variation were 6.9, 4.2, and 7.6% for TSH, FT4, TPOAb, respectively. Antibody positivity was defined as a TPOAb level > 30 U/mL (based on the manufacturer’s functional assay sensitivity cut-off) for the Copenhagen cohort and as TPOAb > 59 U/mL (based on the manufacturer’s cut-off for TPOAb level positivity) for the Brussels cohort.
Statistical analysis
Statistical analyses were performed with GraphPad Prism, version 9.2.0. Continuous data were expressed as median (interquartile range (IQR)). Categorical data were presented as numbers (percentages) of cases. Differences between groups were analyzed by Fisher exact tests for categorical data and by a t-test or Mann–Whitney U test for continuous data. All statistical tests were considered significant whenever P < 0.05.
Results
Baseline characteristics of two cohorts
The baseline population characteristics at the first antenatal visit for the two cohorts studied are displayed in Table 1. The Copenhagen cohort was divided into two groups, TPOAb-negative (TPOAb−) women, n = 255 (93.1%) and TPOAb-positive (TPOAb+) women, n = 19 (6.9%). Women from the Brussels cohort were all TPOAb+ (n = 55, 100%) and constituted the third group.
Baseline population characteristics of the two studied cohorts. Data are shown as median (IQR: 25–75th), n = absolute number of patients for whom data were available, % = percentage of total population cohort. Values in bold indicate statistical significance.
| Population characteristics | Copenhagen cohort n = 274 | Brussels cohort n = 55 | Pc | |
|---|---|---|---|---|
| Thyroid TPOAb status | TPOAb– womenn = 255 (93.1%) | TPOAb+womenn = 19 (6.9%) | TPOAb+womenn = 55 (100%) | |
| Maternal age (years) | 29 (27−31) | 29 (26−32) | 31 (26−36) | 0.09 |
| Maternal height (cm) | 170 (165−174) | 170 (167−172) | 165 (160−169) | < 0.01 |
| Maternal weight (kg)a | 62 (57−67) | 64 (59−71) | 65 (59−72) | 0.05 |
| Maternal BMI (kg/m2)a | 21 (20−23) | 22 (20−24) | 24 (22−27) | < 0.01 |
| Multiparity (≥ 2 prior pregnancies)b | 16 (6.5%) | 2 (10.5%) | 4 (7.2%) | 0.79 |
| Smoking during pregnancy | 0 | 0 | 8 (14.5%) | < 0.01 |
| Gestational age (weeks) | ||||
| 1st sampling | 19 (18−20) | 19 (19−20) | 12 (11−15) | < 0.01 |
| 2nd sampling | 29 (28−29) | 29 (28−30) | 25 (23−28) | < 0.01 |
| TSH mU/L | ||||
| 1st sampling | 1.32 (0.95−1.86) | 1.74 (1.30−2.64) | 1.58 (1.12−1.99) | 0.02 |
| FT4 pmol/L | ||||
| 1st sampling | 9.35 (8.65−10.10) | 9.44 (8.86−9.96) | 14.16 (12.87−15.44) | < 0.01 |
aData available for n = 269 (98.2%) in the Copenhagen cohort; bData available for n = 268 (99.3%) in the Copenhagen cohort; cComparison was performed among all three groups.
Ab, antibody; TPO, thyroid peroxidase.
Comparisons were performed between all three groups. Median maternal age was similar (P = 0.09) among all three groups. Median maternal height was significantly higher (P < 0.01) in the Copenhagen cohort, both in TPOAb− (170 cm (IQR: 165−174)) and in TPOAb+ women (170 cm (IQR: 167−172)) compared to the Brussels cohort (165 cm (IQR: 160−169)). Maternal weight was not significantly different (P = 0.05) among the three groups. In consequence, maternal BMI (kg/m2) was significantly lower (P < 0.01) in the Copenhagen cohort both in TPOAb− (median 21 (IQR: 20−23)) and in TPOAb+ (median 22 (IQR: 20−24)) women compared to the Brussels Cohort (median 24 (IQR: 22−27)) (P < 0.01). The rate of multiparity was similar among the three groups (6.5% in TPOAb−, 10.5% in TPOAb+ in the Copenhagen cohort, and 7.2% in the Brussels cohort, P = 0.79). Only eight women in the Brussels cohort had a history of smoking during pregnancy, while none of the women in the Copenhagen cohort smoked (P < 0.01). Gestational age at the 1st and 2nd blood sampling was significantly lower (P < 0.01) in the Brussels cohort (median 12 weeks (IQR: 11−15) for the 1st sampling and median 25 weeks (IQR: 23−28) for the 2nd sampling) compared to the Copenhagen cohort (median 19 weeks (IQR: 18−20) in TPOAb− and in TPOAb+ (IQR: 19−20) for the 1st sampling and median 29 weeks in TPOAb− (IQR: 28−29) and in TPOAb+ (IQR: 28−30) for the 2nd sampling) Baseline TSH at 1st sampling was higher in TPOAb+ women both in Copenhagen and Brussels cohort (P = 0.02). Baseline FT4 was higher in the Brussels cohort (P < 0.01), a difference probably explained by the use of different assays and earlier gestational age for the 1st sampling in this cohort (Table 1).
Individual values and stratification in tiers of serum TSH at the 1st and the 2nd blood sampling in the total study population
Figure 2 illustrates the distribution of individual values and the stratification in tiers of serum TSH at the 1st and the 2nd blood sampling during pregnancy in both Copenhagen and Brussels cohorts, according to the TPOAb status. As expected, at baseline (1st sampling) and at 2nd sampling the median serum TSH in the TPOAb+ group was significantly higher (at 1st sampling: 1.32 mU/L (IQR: 0.95−1.86) in TPOAb− vs 1.62 mU/L (IQR: 1.24−2.08) in TPOAb+, P = 0.01 and at 2nd sampling: 1.26 mU/L (IQR: 0.93−1.66) in TPOAb− vs 1.66 mU/L (IQR: 1.16−2.29) in TPOAb+, P < 0.001) than in the TPOAb− group. Median TSH at 2nd sampling in TPOAb− group was lower (1.26 mU/L (IQR: 0.93−1.66) vs 1.32 mU/L (IQR: 0.95−1.86), P = 0.002) than at 1st sampling. No differences were observed in median TSH levels between the 1st and 2nd sampling in the TPOAb+ group (P = 0.46). After stratification of TSH in tiers, the only significant difference observed was the proportion (%) of women having serum 2.5 > TSH ≤ 4 mU/L at 2nd sampling which was more frequent in the TPOAb+ group as compared to the TPOAb− group (16.2% in TPOAb+ vs 4.7% in TPOAb− group, P = 0.02). No woman in either TPOAb+ or TPOAb− group presented TSH > 4 mU/L at 2nd sampling (Fig. 2).
Serum TSH (mU/L) values of pregnant women in TPOAb− (n = 255) and TPOAb+ (n = 74) groups in the total study population are illustrated at the 1st and the 2nd blood sampling. Data are presented as individual values (points) and median with IQR (25th−75th percentile). Median TSH was higher in the TPOAb+ group both at 1st (P = 0.01) and at 2nd sampling (P < 0.001) as compared to the TPOAb− group. No differences in median TSH levels were observed between the 1st and 2nd sampling in the TPOAb+ group (P = 0.46), whereas median TSH was lower in the 2nd sampling in TPOAb− group (P = 0.002). Stratification in three distinct tiers of serum TSH was performed and the proportion (%) of pregnant women distributed among the different tiers at both samplings is seen at the bottom of the x-axis. The bottom and upper dashed lines intersect the y-axis at 2.5 mU/L and 4 mU/L TSH levels, respectively. 16.2% of women in the TPOAb+ group had TSH 2.5 > TSH ≤ 4 mU/L as compared to 4.7% of women in the TPOAb− group at 2nd sampling during pregnancy (P = 0.02).
Citation: Endocrine Connections 13, 9; 10.1530/EC-24-0151
Illustration of the modifications below or above the TSH > 2.5 mU/L threshold at the 2nd vs 1st sampling in the total study population
The overall serum TSH modifications below or above 2.5 mU/L, defined as the threshold for biochemical progression between the 1st and 2nd sampling are presented in the descriptive Table 2. Total numbers and numbers according to TPOAb status are presented. The majority of women, 87.2% (89.8% in the TPO− group and 79.4% in the TPO+ group) had a serum TSH ≤ 2.5 mU/L and remained stable below ≤ 2.5 mU/L at the 1st and the 2nd sampling during pregnancy. Women who progressed to serum TSH > 2.5 mU/L while they had ≤ 2.5 mU/L at the 1st sampling represented 4.2% of the total population (2.7% in the TPOAb− group vs 9.4% in the TPOAb+ group). A decrease in TSH < 2.5 mU/L at the 2nd sampling with initial TSH level > 2.5 mU/L was observed in 4.2% of women. Finally, the remaining 4.2% of women remained above TSH > 2.5 m/L threshold at both samplings (Table 3).
Illustration of the modifications below and above the 2.5 mU/L TSH threshold at the 2nd vs 1st blood sampling during pregnancy in the total study population. Data are shown as n = absolute number, % = percentage of subjects in the study population.
| TSH 2nd sampling > 2.5 mU/L vs ≤ 2.5 at 1st sampling |
TSH 2nd sampling < 2.5 mU/L vs > 2.5 at 1st sampling |
TSH 1st and 2nd sampling ≤ 2.5 mU/L |
TSH 1st and 2nd sampling > 2.5 mU/L |
|
|---|---|---|---|---|
| Total population (n = 329) | 14 (4.2%) | 14 (4.2%) | 287 (87.2%) | 14 (4.2%) |
| Thyroid antibody status | ||||
| TPOAb− (n = 255) | 7 (2.7%) | 10 (3.9%) | 229 (89.8%) | 9 (3.5%) |
| TPOAb+ (n = 74) | 7 (9.4%) | 4 (5.4%) | 58 (79.4%) | 5 (6.7%) |
Ab, antibody; TPO, thyroid peroxidase.
Clinical and biological characteristics of women who showed biochemical progression (TSH at 2nd sampling > 2.5 mU/L vs ≤ 2.5 at 1st sampling) versus those who did not progress (TSH 1st and 2nd sampling ≤ 2.5 mU/L), in the total study population. Comparison according to thyroid antibody status. Data are shown as median (IQR: 25−75th), n = absolute number, % = percentage of subjects in the study population cohort. Values in bold indicate statistical significance.
| Clinical characteristics total population (n = 329) | Biochemical progression | P | |
|---|---|---|---|
| Yes | No | ||
| Maternal age (years) | |||
| TPOAb− | 29 (27−34) | 29 (27−31) | 0.91 |
| TPOAb+ | 31 (26−35) | 32 (26−35) | 0.61 |
| Maternal BMI (kg/m2) | |||
| TPOAb− | 21 (21−24) | 21.5 (20.0−23.0) | 0.70 |
| TPOAb+ | 25 (21−27) | 23.5 (21.8−27.0) | 0.19 |
| Multiparity (≥ 3) | |||
| TPOAb− | 1 | 14 | 0.37 |
| TPOAb+ | 0 | 5 | 0.99 |
| Smoking during pregnancy | |||
| TPOAb− | 0 | 0 | 0.99 |
| TPOAb+ | 0 | 8 | 0.58 |
| Thyroid antibody status | 0.02 | ||
| TPOAb− (n = 255) | 7 (2.7%) | 229 (89.8%) | |
| TPOAb+ (n = 74) | 7 (9.4%) | 58 (79.4%) | |
| TPOAb levels (U/ml) in TPOAb+ women (n = 74) | 378 (79−1427) | 281 (111−764) | 0.67 |
| Baseline TSH levels, 1st sampling (mU/L) | |||
| TPOAb− | 2.40 (1.86−2.47) | 1.25 (0.91−1.66) | < 0.01 |
| TPOAb+ | 2.00 (1.65−2.36) | 1.42 (1.09−1.86) | 0.01 |
| Baseline FT4 levels, 1stsampling (pmol/L) | |||
| TPOAb− (n .= 255) | 8.6 (8.3−9.1) | 9.4 (8.7−10.1) | 0.12 |
| TPOAb+ Copenhagen (n = 19) | 9.6 (9.6−9.6) | 9.3 (8.9−10.0) | 0.57 |
| TPOAb+ Brussels (n = 55) | 12.9 (12.9−14.2) | 14.2 (12.9−15.4) | 0.09 |
Ab, antibody; TPO, thyroid peroxidase.
Clinical and biological factors for biochemical progression (TSH > 2.5 mU/L) in pregnancy in the total study population
Finally, in Table 3, clinical and biological characteristics in women with (TSH > 2.5 mU/L at 2nd sampling while having TSH ≤ 2.5 mU/L at 1st sampling) and without (TSH ≤ 2.5 mU/L at 1st and 2nd sampling) biochemical progression during pregnancy are displayed. Comparisons were performed, taking into account the TPOAb status (Table 3). Maternal age, BMI, multiparity (≥ 2 prior pregnancies), and smoking during pregnancy were not significantly different between women with and without biochemical progression, in neither TPOAb+ nor in TPOAb− groups. A significantly higher proportion of TPOAb+ women showed biochemical progression, as compared to TPOAb− women (9.4% in TPOAb+ vs 2.7% in TPOAb−, P = 0.02). Among TPOAb+ women, circulating antibody concentrations were similar (P = 0.67) between women who showed biochemical progression (median TPOAb levels 378 U/mL (IQR: 79−1427) and those who did not (median TPOAb levels 281 U/mL (IQR: 111−764)). FT4 concentrations at 1st sampling (baseline) were analyzed within the same cohort, because of significant differences in the laboratory reference range used (see methods section). In the TPOAb− group (n = 255), consisting entirely of Copenhagen cohort women, FT4 levels were similar (P = 0.12) between women who showed progression and those who did not. In the TPOAb+ Copenhagen cohort group (n = 19) FT4 levels were similar (P = 0.57) between women who showed progression (n = 2) and those who did not (n = 17). In the TPOAb+ Brussels cohort group (n = 55), serum FT4 levels were lower in women who progressed (n = 10) but this did not reach statistical significance (12.9 pmol/L (IQR: 12.9−14.2) vs 14.2 pmol/L (IQR: 12.9−15.4), P = 0.09).
Finally, serum TSH levels at the 1st sampling were significantly higher in women who showed biochemical progression compared to those who did not, and this observation was performed in both TPOAb− (median TSH 2.40 mU/L (IQR: 1.86−2.47) in women with progression vs median 1.25 mU/L (IQR: 0.91−1.66) in women with no progression, P < 0.01) and in TPOAb+ (median TSH 2.0 mU/L (IQR: 1.65−2.36) in women with progression vs median 1.42 mU/L (IQR: 1.09−1.86) in women with no progression, P = 0.01) groups (Table 3).
The delta TSH (serum TSH level of 2nd sampling – serum TSH level of 1st sampling) was calculated and compared between TPOAb+ and TPOAb− women (data not shown). Delta TSH was 0.07 mU/L (IQR: −0.29−0.38) for TPOAb+ women and −0.09 mU/L (IQR: −0.35−0.20) for TPOAb− women, which was just below the statistical significance level (P = 0.03). However, given the very slight absolute difference in the TSH levels between the two samplings during pregnancy in both groups this result is unlikely to be clinically significant.
Discussion
In this observational study based on two population cohorts with spontaneously achieved singleton pregnancies, we observed that 4.2% of women in the total study population presented a rise from baseline TSH ≤ 2.5 mU/L levels to > 2.5 mU/L at a subsequent blood sampling. TPOAb positivity and higher baseline TSH concentrations were significantly associated with this biochemical progression.
Thyroid autoimmunity has been associated with adverse maternal pregnancy outcomes; in TPOAb+ pregnant women, the thyroidal response to human chorionic gonadotropin stimulation appears to be impaired, and so, these women may fail to meet the pregnancy-related increased demand for thyroid hormone synthesis (23, 24).
It is worth noting that most studies on thyroid and pregnancy outcomes are of cross-sectional nature with thyroid function tests typically performed in the late first trimester. A more recent longitudinal study in apparently healthy pregnant women suggested that longitudinal trajectories may correlate better with pregnancy outcomes and that multiparity and fetal sex as potential thyroid trajectory modifiers should be addressed in future studies (25).
The most recent 2017 American Thyroid Association (ATA) guidelines recommend repeated blood samplings every 4 weeks of gestation through the late second trimester in TPOAb+ women, suggest consideration of treatment in TPOAb+ women with TSH > 2.5 mU/L, and recommend treatment when TSH is above the pregnancy-specific range or, if not available, when TSH is above 4 mU/L (5). These guidelines are based on the studies by Glinoer et al. in 1994 in which euthyroid women who were TPOAb+ or TgAb+ (n = 87) had a higher risk of developing serum TSH > 4 mU/L during pregnancy (17). Also, Negro et al. showed that euthyroid TPOAb+ women (n = 57) had higher TSH values throughout the entire pregnancy compared to euthyroid TPOAb-negative women (n = 869) (18). Subsequently, in 2017, Nazarpour et al. reported that among TPOAb+ women with serum TSH < 2.5 mIU/L in early pregnancy (n = 58), 40.9% developed serum TSH 2.5−5 mU/L in the second trimester of pregnancy and 36.4 % in the third trimester (19). Important mentioning is that in the Nazarpour et al. study, a total of 39.4% of women in the second trimester had TSH levels > 5.0 mU/L, although it is not clear which women progressed from TSH values 2.5−5.0 mU/L to > 5.0 mU/L and which variables contributed to this progression.
Three randomized controlled trials, subsequent to the publication of 2017 ATA guidelines, investigated the outcomes of treatment with LT4 and did not find a beneficial effect on pregnancy loss nor on preterm birth in euthyroid women with autoimmunity (TABLET, T4LIFE, Chinese randomized trial) (14, 15, 16). Moreover, an increase in FT4 during pregnancy in overtreated mothers has been associated with adverse maternal and offspring outcomes (gestational hypertension, lower birth weight, lower IQ, and attention deficit disorder in offspring) (26, 27, 28, 29). In light of the above-mentioned trial results, routine LT4 treatment in euthyroid TPOAb+ women seeking pregnancy does not seem to be recommendable; other factors (i.e. immunological) beyond thyroid function alone might contribute to the adverse pregnancy outcomes in (euthyroid) TPOAb+ women (30). In a secondary analysis of TABLET trial data, an overall of n = 70 (7.4%) out of n = 940 randomized euthyroid TPOAb-positive participants developed subclinical (SCH, n = 63) or overt (n = 7) hypothyroidism (OH); the great majority (n = 60, 85.7%) did so in the preconception period. Thus, thyroid function testing in TPOAb+ women when trying to conceive is relevant to promptly diagnose and appropriately treat thyroid dysfunction before pregnancy. If progression occurs, it seems to occur early in pregnancy, since in the placebo group (n = 470) progression to SCH during pregnancy was observed in n = 7 patients at 6−8 weeks, in n = 2 patients at 16−18 weeks, and in none during the third trimester; the only woman who developed OH during pregnancy in the placebo group did so at early first trimester (6−8 weeks) (7). Furthermore, baseline serum TSH was higher in women who developed SCH/OH (3 mU/L vs 2 mU/L) as compared to those who did not, a finding which was also observed in this study (7).
Accurate diagnosis of thyroid function in pregnancy by applying the proper reference intervals (assay vs institutional, non-pregnancy vs pregnancy trimester-specific intervals) is challenging per se, and prone to misdiagnosis, especially in the diagnostic category of subclinical hypothyroidism (31). Multiplication of blood samplings to monitor thyroid function and/or LT4 treatment, when added value of LT4 in euthyroid women does not seem evident, and intervention of different specialists (endocrinologists, gynecologists) for the accurate interpretation of results may lead to excessive costs, stress, and overmedicalization during pregnancy. Knowledge of clinical and biological determinants, other than TPOAb, to better target the group of women that should benefit the most from regular biochemical monitoring during pregnancy is currently lacking.
Indeed, in a very recently published large metanalysis, maternal age, BMI, smoking status, parity, and gestational age at blood sampling were shown to have poor predictive ability to detect thyroid dysfunction during pregnancy on an individual basis; TPOAb positivity alone or combined with TgAb positivity were relevant as expected for predicting subclinical/overt hypothyroidism and indication for treatment (32).
In the current study, the majority (87.2%) of pregnant women did not show biochemical progression, defined as a rise in serum TSH > 2.5 mU/L. Progression was significantly more frequent (9.4%) among TPOAb+ women; however, progression of TSH > 4 mU/L was not observed. Noteworthy, 2.4% of TPOAb− women also showed progression. In the TPOAb+ group, there was no significant difference in antibody concentrations between women with and without progression. Besides antibody status, the only other factor associated with progression was higher baseline serum TSH concentrations at the first antenatal visit.
There are limitations to this study. The first one was the limited number of subjects and in consequence the limited number of observed events (biochemical progression). As a result, the study was underpowered to discern independent predictive factors for biochemical progression during pregnancy, and definitive conclusions cannot be drawn at this stage. A possible selection bias was that only TPOAb+ women in the Brussels cohort had a second blood sampling during pregnancy, in consistence with clinical practice guidelines, so the progression of TSH in TPOAb− women of this cohort could not be studied.
In conclusion, in spontaneously pregnant women with initial TSH < 2.5 mU/L and thyroid autoimmunity, our data showed progression of serum TSH > 2.5 mU/L in only a minority (9.4%) of them and never above 4 mU/L. Higher baseline TSH at the first sampling and TPOAb positivity were associated with progression. Despite this, our data seems in favor of declining systematic high-frequency follow-up of TSH during pregnancy in these women, which would also be in line with the absence of a beneficial impact of LT4 on pregnancy outcomes in TPOAb+ euthyroid women. Future prospective studies in larger cohorts with longitudinal assessment of thyroid function could help to identify factors such as a relevant baseline TSH cutoff, TPOAb cutoff, parity, hCG levels and others, for biochemical progression and potential association of this progression with pregnancy outcomes. This could help to better target women who would benefit the most from repeated thyroid function testing during pregnancy and at which interval.
Declaration of interest
All authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the study reported.
Funding
UFR’s research salary was sponsored by a grant from the Kirsten and Freddy Johansen's Fund. SB research was supported by grants from the Novo Nordisk Foundation (NNF22OC0077221, NNF23OC0087269). DHP’s research was supported by a grant from The Danish Research Council (registration no. 28809).
Acknowledgements
We would like to thank patients and their families for their trust and our colleague Dr Flora Veltri for her help in the data collection for Brussels cohort. AK, BC, and UFR are members of the European Reference Network for Rare Endocrine Diseases (ENDO-ERN).
References
- 1↑
Valdés S, Maldonado-Araque C, Lago-Sampedro A, Lillo JA, Garcia-Fuentes E, Perez-Valero V, Gutierrez-Repiso C, Ocon-Sanchez P, Goday A, Urrutia I, et al.Population-based national prevalence of thyroid dysfunction in Spain and associated factors: Di@bet.es study. Thyroid 2017 27 156–166. (https://doi.org/10.1089/thy.2016.0353)
- 2↑
Pedersen IB, Knudsen N, Carlé A, Vejbjerg P, Jørgensen T, Perrild H, Ovesen L, Rasmussen LB, & Laurberg P. A cautious iodization programme bringing iodine intake to a low recommended level is associated with an increase in the prevalence of thyroid autoantibodies in the population. Clinical Endocrinology 2011 75 120–126. (https://doi.org/10.1111/j.1365-2265.2011.04008.x)
- 3↑
Veltri F, Belhomme J, Kleynen P, Grabczan L, Rozenberg S, Pepersack T, & Poppe K. Maternal thyroid parameters in pregnant women with different ethnic backgrounds: do ethnicity-specific reference ranges improve the diagnosis of subclinical hypothyroidism? Clinical Endocrinology 2017 86 830–836. (https://doi.org/10.1111/cen.13340)
- 4↑
Korevaar TIM, Medici M, Visser TJ, & Peeters RP. Thyroid disease in pregnancy: new insights in diagnosis and clinical management. Nature Reviews. Endocrinology 2017 13 610–622. (https://doi.org/10.1038/nrendo.2017.93)
- 5↑
Alexander EK, Pearce EN, Brent GA, Brown RS, Chen H, Dosiou C, Grobman WA, Laurberg P, Lazarus JH, Mandel SJ, et al.2017 Guidelines of the American Thyroid Association for the diagnosis and management of thyroid disease during pregnancy and the postpartum. Thyroid 2017 27 315–389. (https://doi.org/10.1089/thy.2016.0457). Erratum: Thyroid 27 1212. (https://doi.org/10.1089/thy.2016.0457.correx)
- 6↑
Kyrilli A, Unuane D, & Poppe KG. Thyroid autoimmunity and pregnancy in euthyroid women. Best Practice and Research. Clinical Endocrinology and Metabolism 2023 37 101632. (https://doi.org/10.1016/j.beem.2022.101632)
- 7↑
Gill S, Cheed V, Morton VAH, Gill D, Boelaert K, Chan S, Coomarasamy A, & Dhillon-Smith RK. Evaluating the progression to hypothyroidism in preconception euthyroid thyroid peroxidase antibody-positive women. Journal of Clinical Endocrinology and Metabolism 2022 108 124–134. (https://doi.org/10.1210/clinem/dgac525)
- 8↑
Bliddal S, Derakhshan A, Xiao Y, Chen L-M, Männistö T, Ashoor G, Tao F, Brown SJ, Vafeiadi M, Itoh S, et al.Association of thyroid peroxidase antibodies and thyroglobulin antibodies with thyroid function in pregnancy: an individual participant data meta-analysis. Thyroid 2022 32 828–840. (https://doi.org/10.1089/thy.2022.0083)
- 9↑
Thangaratinam S, Tan A, Knox E, Kilby MD, Franklyn J, & Coomarasamy A. Association between thyroid autoantibodies and miscarriage and preterm birth: meta-analysis of evidence. BMJ 2011 342 d2616. (https://doi.org/10.1136/bmj.d2616)
- 10↑
Consortium on Thyroid and Pregnancy—Study Group on Preterm Birth, Korevaar TIM, Derakhshan A, Taylor PN, Meima M, Chen L, Bliddal S, Carty DM, Meems M, Vaidya B, et al.Association of thyroid function test abnormalities and thyroid autoimmunity with preterm birth: a systematic review and meta-analysis. JAMA 2019 322 632–641. (https://doi.org/10.1001/jama.2019.10931)
- 11↑
Rao M, Zeng Z, Zhou F, Wang H, Liu J, Wang R, Wen Y, Yang Z, Su C, Su Z, et al.Effect of levothyroxine supplementation on pregnancy loss and preterm birth in women with subclinical hypothyroidism and thyroid autoimmunity: a systematic review and meta-analysis. Human Reproduction Update 2019 25 344–361. (https://doi.org/10.1093/humupd/dmz003)
- 12↑
Sun X, Hou N, Wang H, Ma L, Sun J, & Liu Y. A meta-analysis of pregnancy outcomes with levothyroxine treatment in euthyroid women with thyroid autoimmunity. Journal of Clinical Endocrinology and Metabolism 2020 105 dgz217. (https://doi.org/10.1210/clinem/dgz217)
- 13↑
Di Girolamo R, Liberati M, Silvi C, & D’Antonio F. Levothyroxine supplementation in euthyroid pregnant women with positive autoantibodies: a systematic review and meta-analysis. Frontiers in Endocrinology 2022 13 759064. (https://doi.org/10.3389/fendo.2022.759064)
- 14↑
Dhillon-Smith RK, Middleton LJ, Sunner KK, Cheed V, Baker K, Farrell-Carver S, Bender-Atik R, Agrawal R, Bhatia K, Edi-Osagie E, et al.Levothyroxine in women with thyroid peroxidase antibodies before conception. New England Journal of Medicine 2019 380 1316–1325. (https://doi.org/10.1056/NEJMoa1812537)
- 15↑
van Dijk MM, Vissenberg R, Fliers E, van der Post JAM, van der Hoorn M-LP, de Weerd S, Kuchenbecker WK, Hoek A, Sikkema JM, Verhoeve HR, et al.Levothyroxine in euthyroid thyroid peroxidase antibody positive women with recurrent pregnancy loss (T4LIFE trial): a multicentre, randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Diabetes and Endocrinology 2022 10 322–329. (https://doi.org/10.1016/s2213-8587(2200045-6)
- 16↑
Wang H, Gao H, Chi H, Zeng L, Xiao W, Wang Y, Li R, Liu P, Wang C, Tian Q, et al.Effect of levothyroxine on miscarriage among women with normal thyroid function and thyroid autoimmunity undergoing in vitro fertilization and embryo transfer a randomized clinical trial. JAMA 2017 318 2190–2198. (https://doi.org/10.1001/jama.201-7.18249)
- 17↑
Glinoer D, Riahi M, Grün JP, & Kinthaert J. Risk of subclinical hypothyroidism in pregnant women with asymptomatic autoimmune thyroid disorders. Journal of Clinical Endocrinology and Metabolism 1994 79 197–204. (https://doi.org/10.1210/jcem.79.1.8027226)
- 18↑
Negro R, Formoso G, Mangieri T, Pezzarossa A, Dazzi D, & Hassan H. Levothyroxine treatment in euthyroid pregnant women with autoimmune thyroid disease: effects on obstetrical complications. Journal of Clinical Endocrinology and Metabolism 2006 91 2587–2591. (https://doi.org/10.1210/jc.2005-1603)
- 19↑
Nazarpour S, Ramezani Tehrani F, Simbar M, Tohidi M, Alavi Majd H, & Azizi F. Effects of levothyroxine treatment on pregnancy outcomes in pregnant women with autoimmune thyroid disease. European Journal of Endocrinology 2017 176 253–265. (https://doi.org/10.1530/EJE-16-0548)
- 20↑
Bliddal S, Feldt-Rasmussen U, Boas M, Faber J, Juul A, Larsen T, & Precht DH. Gestational age-specific reference ranges from different laboratories misclassify pregnant women’s thyroid status: comparison of two longitudinal prospective cohort studies. European Journal of Endocrinology 2014 170 329–339. (https://doi.org/10.1530/EJE-13-0672)
- 21↑
Veltri F, Kleynen P, Grabczan L, Salajan A, Rozenberg S, Pepersack T, & Poppe K. Pregnancy outcomes are not altered by variation in thyroid function within the normal range in women free of thyroid disease. European Journal of Endocrinology 2018 178 189–197. (https://doi.org/10.1530/EJE-17-0628)
- 22↑
Stagnaro-Green A, Abalovich M, Alexander E, Azizi F, Mestman J, Negro R, Nixon A, Pearce EN, Soldin OP, Sullivan S, et al.Guidelines of the American Thyroid Association for the diagnosis and management of thyroid disease during pregnancy and postpartum. Thyroid 2011 21 1081–1125. (https://doi.org/10.1089/thy.2011.0087)
- 23↑
Korevaar TIM, Steegers EAP, Chaker L, Medici M, Jaddoe VWV, Visser TJ, de Rijke YB, & Peeters RP. Thyroid function and premature delivery in TPO antibody negative women: the added value of hCG. Journal of Clinical Endocrinology and Metabolism 2017 102 3360–3367. (https://doi.org/10.1210/jc.201-7-00846)
- 24↑
Korevaar TIM, Steegers EAP, De Rijke YB, Visser WE, Jaddoe VWV, Visser TJ, Medici M, & Peeters RP. Placental angiogenic factors are associated with maternal thyroid function and modify hCG-mediated FT4 stimulation. Journal of Clinical Endocrinology and Metabolism 2015 100 E1328–E1334. (https://doi.org/10.1210/jc.2015-2553)
- 25↑
Pop V, Broeren M, Wijnen H, Endendijk J, Van Baar A, Wiersinga W, & Williams GR. Longitudinal trajectories of gestational thyroid function: a new approach to better understand changes in thyroid function. Journal of Clinical Endocrinology and Metabolism 2018 103 2889–2900. (https://doi.org/10.1210/jc.2017-02556)
- 26↑
Maraka S, Mwangi R, McCoy RG, Yao X, Sangaralingham LR, Singh Ospina NM, O'Keeffe DT, De Ycaza AE, Rodriguez-Gutierrez R, Coddington CC 3rd, et al.Thyroid hormone treatment among pregnant women with subclinical hypothyroidism: US national assessment. British Medical Journal 2017 25 i6865. (https://doi.org/10.1136/bmj.i6865)
- 27↑
Hales C, Taylor PN, Channon S, McEwan K, Thapar A, Langley K, Muller I, Draman MS, Dayan C, Gregory JW, et al.Controlled antenatal thyroid screening II: effect of treating maternal suboptimal thyroid function on child behavior. Journal of Clinical Endocrinology and Metabolism 2020 105. (https://doi.org/10.1210/clinem/dgz098)
- 28↑
Korevaar TIM, Muetzel R, Medici M, Chaker L, Jaddoe VWV, de Rijke YB, Steegers EAP, Visser TJ, White T, Tiemeier H, et al.Association of maternal thyroid function during early pregnancy with offspring IQ and brain morphology in childhood: a population-based prospective cohort study. Lancet. Diabetes and Endocrinology 2016 4 35–43. (https://doi.org/10.1016/S2213-8587(1500327-7)
- 29↑
Derakhshan A, Peeters RP, Taylor PN, Bliddal S, Carty DM, Meems M, Vaidya B, Chen L, Knight BA, Ghafoor F, et al.Association of maternal thyroid function with birthweight: a systematic review and individual-participant data meta-analysis. Lancet. Diabetes and Endocrinology 2020 8 501–510. (https://doi.org/10.1016/S2213-8587(2030061-9)
- 30↑
Korevaar TIM, & Dhillon-Smith R. Levothyroxine treatment in euthyroid women positive for thyroid peroxidase antibodies and recurrent pregnancy loss. Lancet. Diabetes and Endocrinology 2022 10 299–301. (https://doi.org/10.1016/S2213-8587(2200079-1)
- 31↑
Osinga JAJ, Derakhshan A, Palomaki GE, Ashoor G, Männistö T, Maraka S, Chen L, Bliddal S, Lu X, Taylor PN, et al.TSH and FT4 reference intervals in pregnancy: a systematic review and individual participant data meta-analysis. Journal of Clinical Endocrinology and Metabolism 2022 107 2925–2933. (https://doi.org/10.1210/clinem/dgac425)
- 32↑
Osinga JAJ, Liu Y, Männistö T, Vafeiadi M, Tao FB, Vaidya B, Vrijkotte TGM, Mosso L, Bassols J, López-Bermejo A, et al.Risk factors for thyroid dysfunction in pregnancy: an individual participant data meta-analysis. Thyroid 2024 34 646–658. (https://doi.org/10.1089/thy.2023.0646)
