Abstract
Background:
Maternal thyroid hormones’ supply is crucial for fetal neurodevelopment; however, the role of maternal mild thyroid dysfunction is not clear. We aimed to assess the association of maternal mild thyroid dysfunction with child neuropsychological development from infancy to early childhood.
Methods:
We included 757 mother-child pairs from the prospective “Rhea” cohort on Crete, Greece. Maternal thyroid functioning was assessed by quantitative analysis of serum thyroid stimulating hormone (TSH), free thyroxine (fT4), thyroid peroxidase antibodies (TPO-Abs), and thyroglobulin antibodies (Tg-Abs) at early gestation (mean=14 weeks). Neuropsychological assessment was based on Bayley Scales of Infant Development (18 months of age), McCarthy Scales of Children’s Abilities (4 years of age), Raven’s Coloured Progressive Matrices, Trail Making Test, and Finger Tapping Test (6 years of age).
Results:
In multivariate adjusted linear regression analyses maternal hypothyroxinemia was associated with decreased verbal scores at 4 years and reduced motor speed at 6 years of age. Maternal thyroid autoimmunity was associated with decreased child perceptual and motor ability at 4 years of age. Four trajectories of longitudinal non-verbal cognitive development were identified and children exposed to maternal thyroid autoimmunity had increased risk for belonging to an adverse trajectory (“Low”: adjusted-RRR = 2.7 95%CI: [1.4, 5.2], “High-decreasing”: adjusted-RRR = 2.2 95%CI: [1.2, 4.0], “Low-increasing”: adjusted-RRR = 1.8 95%CI: [1.0, 3.2]).
Conclusion:
Maternal hypothyroxinemia is associated with reduced offspring verbal and motor ability. Maternal thyroid autoimmunity is associated with decreased offspring perceptual performance and motor ability and increased risk for adverse non-verbal cognitive development from infancy to childhood.
Keywords: thyroid hormones, hypothyroxinemia, thyroid autoimmunity, subclinical hypothyroidism, hyperthyroxinemia, neuropsychological development, cognitive development, motor ability
INTRODUCTION
Brain development begins shortly after conception and occurs in a chain of developmental events, which correspond to periods of increased vulnerability to environmental stressors [1]. During these critical periods thyroid hormones are essential for normal neural network construction, since the interactions of the active thyroid hormone triiodothyronine with nuclear receptors in the central nervous system regulate the expression of genes involved in cell differentiation, migration, signalling, and myelination [2].
Thyroxine of maternal origin represents the primary source of thyroid hormones until fetal thyroid gland functional maturation (18th–20th gestational week) and remains a complementary source of circulating thyroxine for the fetus until birth [3]. Magnetic Resonance Imaging (MRI) studies have shown that children of mothers with overt hypothyroidism during pregnancy have morphological alterations in the cortex [4] and the hippocampal volume [5]; brain regions that support various cognitive functions like perception, analytical thinking, executive functioning, and memory [6].
Evidence from observational studies suggest that low maternal free thyroxine (fT4) levels are associated with decreased mental performance [7] and intelligence [8], and maternal hypothyroxinemia [thyroid stimulating hormone (TSH) within the normal interval and decreased thyroxine (fT4)] is associated with decreased intelligence [9–11], decreased quantitative and general cognitive development [12], reduced cognitive ability, perceptual performance, and memory [13], decreased response speed [14], and increased risk for delay in expressive language and non-verbal cognition [15]. Furthermore, findings from a recent study have supported that both low and high concentration of maternal thyroxine is associated with reduced intelligence and decreased grey and total cortex volume [8]; however, the role of increased thyroxine levels in child cognitive development has not been studied in other populations and the role of hyperthyroxinemia is still unclear. Previous findings regarding subclinical hypothyroidism (TSH above the normal interval and fT4 within the normal interval) have been less conclusive. More specifically, subclinical hypothyroidism has been associated with neurodevelopmental delays in infancy [16], decreased intelligence in toddlerhood [10], and reduced verbal, memory and cognitive scores in prematurely born infants [12]. Conversely, results from two large observational studies did not support any association of maternal subclinical hypothyridism or decreased levels of TSH with offspring cognitive development [7, 15]. Furthermore, the heterogeneity of the findings regarding the role of maternal thyroid autoimmunity [10, 17–20] has underlined the need for further studies in order to clarify if there is any effect of maternal thyroid autoimmunity on child cognition, if the effect is transient [19], or whether it is dependent on the specific characteristics of the examined population (e.g. iodine status) [17].
In the present study we evaluated the impact of maternal mild thyroid dysfunction on child neuropsychological development. We hypothesized that maternal hypothyroxinemia, subclinical hypothyroidism, hyperthyroxinemia, and thyroid autoimmunity during early pregnancy are associated with decreased neuropsychological development from infancy to early childhood.
METHODS
Participants
This study is part of the “Rhea” study, a mother-child birth cohort on Crete, Greece, which follows up children from fetal life onward. Participants were recruited in early pregnancy, at the time of the first ultrasound examination. The inclusion criteria were residency within the study area, good understanding of the Greek language and maternal age greater than 16 years. Participants were invited to child neuropsychological follow-up assessments when the children were 18 months, 4 years and 6 years old. Data on maternal thyroid parameters were available for 1170 women with a live singleton birth. Of this population, 484, 695 and 488 participants provided data on neuropsychological development at 18 months, 4 years and 6 years of age, respectively (Supplementary Figure 1); while 757 children participated in at least one of these follow-up assessments. Further details on participant recruitment and follow up procedures have been previously described in detail [21].
The present study was conducted according to the principles of the Helsinki Declaration. All procedures were approved by the Ethics Committee of the University Hospital of Heraklion. Written informed consent was obtained from all adult participants.
Maternal thyroid parameters
Maternal blood samples were collected at the first prenatal visit (mean gestational age 14.12 weeks, SD 3.6 weeks). Serum samples were collected in 10 ml vacutainer tubes, were centrifuged and stored in aliquots at −80° C until assayed. Maternal thyroid functioning was assessed by quantitative analysis of serum thyroid stimulating hormone (TSH), free thyroxine (fT4), thyroid peroxidase antibodies (TPO-Abs) and thyroglobulin antibodies (Tg-Abs), by Immulite 2000 immunoassay system (Siemens Healthcare Diagnostics, Los Angeles, CA). The inter- and intra-assay coefficients of variation were: TSH < 5.3 and < 6.4 (0.32 – 39 mIU/mL), fT4 < 7.8% and < 7.1% (0.51–4.82 ng/dL or 6.56–62.03 pmol/L), Tg-Abs < 4.9% and < 5.8% and TPO-Abs < 7.4% and 7.2%.
Population-based and trimester-specific reference intervals were applied for participants’ assignment to subclinical hypothyroidism, hyperthyroxinemia, and hypothyroxinemia categories. Further details regarding the estimation of the reference intervals are provided elsewhere [22]. Subclinical hypothyroidism was defined as TSH above the normal, trimester-specific reference interval but below 10 mIU/mL and fT4 within the normal range and the respective comparison group included women with TSH and fT4 levels within the normal trimester-specific reference ranges (1st trimester: TSH: 0.05 – 2.53 μIU/mL & fT4: 0.95 – 1.53 ng/dL; 2nd trimester: TSH: 0.18 – 2.73 μIU/mL & fT4: 0.87 – 1.45 ng/dL). Hypothyroxinemia was defined as TSH within the normal trimester-specific reference range and fT4 below the 5th percentile (corresponding to values of fT4 < 0.95 ng/dL); the respective comparison group included women with TSH within the normal trimester-specific reference range and fT4 above the 5th percentile and below the upper trimester-specific fT4 limit. Hyperthyroxinemia was defined as TSH within the normal trimester-specific reference range and fT4 above the 95th percentile (corresponding to values of fT4 > 1.54 ng/dL); the respective comparison group included women with TSH within the normal trimester-specific reference range and fT4 below the 95th percentile and above the lower trimester-specific fT4 limit. Thyroid autoimmunity was defined as elevated thyroid antibodies and TSH and fT4 within the normal range; thyroid-antibodies’ status was considered elevated if the concentration of thyroid peroxidase antibodies was ≥ 35 IU/mL and/or the concentration of thyroglobulin antibodies was > 40 IU/mL.
Child neuropsychological development
Child neuropsychological development was assessed by internationally validated and standardized scales administered by trained psychologists. Neuropsychological development assessment at 18 months [mean (SD): 1.5 (0.1) years] was conducted using the Bayley Scales of Infant and Toddler Development (Bayley-III) [23]. The Bayley-III provides indexes on five developmental domains (cognitive, receptive communication, expressive communication, fine motor and gross motor). Raw scores were standardized for child’s age and homogenized with a mean (SD) of 100 (15) [24]. Further information is provided elsewhere [25]. Cognitive and motor development assessment at 4 years of age [mean (SD): 4.3 (0.2) years], was conducted using McCarthy Scales of Children’s Abilities (MCSA) [26]. The MCSA provide indexes on five developmental domains (verbal, quantitative, memory, perceptual performance and motor) and a general cognitive development index. MCSA raw scores were standardized for child’s age and homogenized with a mean (SD) of 100 (15) [24]. Further details are provided elsewhere [27]. Cognitive development assessment at 6 years of age [mean (SD): 6.6 (0.3) years], was computerized and included the Raven’s Colored Progressive Matrices (RCPM) [28], which assess non-verbal intelligence, the Finger Tapping Test (FTT) [29], which assesses motor speed, and the Trail Making Test (TMT-Part A & Part B) [29], which assesses processing speed, mental flexibility and executive functioning.
Covariates
We selected the covariates of the analyses based on the Directed Acyclic Graphs (DAGs) approach (Supplementary Figure 2). The minimal sufficient adjustment set for estimating the total effect of maternal thyroid hormones on child cognition included maternal origin (Greek/other), parity (primiparous/multiparous), maternal smoking at early pregnancy (yes/no), maternal BMI in early pregnancy, maternal education (low level: ≤ 9 years of school, medium level: 9 to 12 years of school, high level: university or technical college degree), maternal age at birth (years), maternal marital status at pregnancy (married/other) and gestational age at blood sampling. Quality of assessment and child sex were included a priori in all models and child age was included a priori in the models involving TMT and FTT. The combined missing values of the covariates were < 6%.
Statistical Analysis
We conducted descriptive analyses on the characteristics of the study population and the distribution of the exposure. Multivariate linear regression models were used to estimate beta coefficients with 95% confidence intervals for the associations of maternal hypothyroxinemia, subclinical hypothyroidism, hyperthyroxinemia, and thyroid autoimmunity as well as the concentration levels of maternal TSH and fT4 with each score of offspring neuropsychological development. In addition, we constructed trajectories of longitudinal non-verbal cognitive development from infancy to early childhood using Group Based Trajectory Modelling (GBTM). The distribution percentiles of the Bayley’s cognitive scale, the MSCA’s perceptual performance, and the RCPM’s total score were calculated and used in GBTM, in order to homogenize the scales. The Stata traj-plugin was used to estimate the group-based trajectory model. Participants with data on neuropsychological development in, at least, two time-points were included (N=586). Probability of group membership, predicted trajectory of each group and posterior probabilities of group membership were estimated. 2–5 possible cognitive development trajectories were tested and a model with 4 classes was selected based on the Bayesian information criterion (BIC), the evaluation of average posterior probability (AvePP value > 0.65), the odds of correct classification (OCC > 5), and the number of participants in each group (Supplementary Table 1). The associations between maternal hypothyroxinemia, subclinical hypothyroidism, hyperthyroxinemia, and thyroid autoimmunity and the non-verbal cognitive trajectories of development were explored using multinomial linear regression models, weighted for each individual’s posterior probability of belonging to each of the trajectories. The estimated associations derived from the multinomial linear regression models are presented in terms of Relative Risk Ratios (RRR) and 95% confidence intervals.
We performed sensitivity analyses, excluding mothers under thyroid medication during pregnancy (N = 96). We also repeated the main analyses in children exposed to elevated levels of maternal TPO-Antibodies (TPO-Abs ≥ 35 IU/mL) and to elevated levels of Tg-Antibodies (Tg > 40 IU/mL) separately.
The statistical analyses were conducted using STATA 13.1 (Statacorp, College Station, TX) and the DAG’s were designed using DAGitty-v. 3.0. All associations were tested assuming a p value < 0.05 significance level.
RESULTS
Maternal and child characteristics at 18 months, 4 and 6 years of age are presented at Table 1. Non-response analyses showed no differences between participants and non-participants in terms of maternal thyroid functioning (TSH, fT4), maternal smoking status during pregnancy, offspring gestational age at birth and birthweight. However, mothers who did not participate in the follow-up assessments were younger [mean difference = - 0.9 years; 95% CI (- 1.5, - 0.3) p = .004] and more often less educated [low educational level 29.8% versus 17.4%, χ2 = 28.85(2), p < 0.001], multiparous [65% versus 58%, χ2 = 5.14(1), p = 0.023) and non-Greek [14.3% versus 6.3%, χ2 = 19.84(1), p < 0.001] compared with mothers who participated to the follow-up assessments (Supplementary Table 2). We identified 4 trajectories of non-verbal cognitive development from infancy to early childhood (continuously low, continuously high, high at 18 months-decreasing over time, low at 18 months-increasing over time) (Figure 1). Individual trajectories based on GBTM analysis are presented at Supplementary Figure 3. The background characteristics between exposed and non-exposed participants are presented at Supplementary Table 4.
Table 1.
Baseline characteristics of the study participants at 18 months, 4 years and 6 years of age, Rhea mother-child study, Crete, Greece
| 18 months a (N = 467) |
4 years a (N = 658) |
6 years a (N = 460) |
|
|---|---|---|---|
| Maternal Characteristics | |||
| Age (years) | 30.3 (4.7) | 29.7 (5.0) | 30.1 (4.8) |
| Educational level | |||
| Low | 61 (13.1) | 110 (16.7) | 70 (15.2) |
| Medium | 228 (48.8) | 335 (51.0) | 232 (50.4) |
| High | 178 (38.1) | 212 (33.3) | 158 (34.4) |
| Parity | |||
| Primiparous | 196 (42.0) | 283 (43.0) | 199 (43.3) |
| Multiparous | 271 (58.0) | 375 (57.0) | 261 (56.7) |
| Origin | |||
| Greek | 449 (96.2) | 617 (93.8) | 435 (94.6) |
| Non-Greek | 18 (3.9) | 41 (6.2) | 25 (5.4) |
| Smoking at early pregnancy | |||
| Yes | 161 (34.5) | 233 (35.4) | 159 (34.6) |
| No | 306 (65.5) | 425 (64.6) | 301 (65.4) |
| Maternal thyroid parameters | |||
| TSH [median (IQR)] (μIU/mL) | 1.09 (0.95) | 1.08 (0.95) | 1.04 (0.91) |
| ft4 (ng/dL) | 1.22 (0.20) | 1.23 (0.21) | 1.23 (0.21) |
| Maternal iodine (median) (μg/L) | 178.6 | 172.4 | 168.5 |
| Hypothyroxinemia b | 22 (5.7) | 26 (4.7) | 22 (5.7) |
| Subclinical hypothyroidism b | 29 (7.0) | 38 (6.5) | 27 (6.7) |
| Hyperthyroxinemia b | 20 (4.8) | 28 (4.6) | 19 (4.5) |
| Thyroid autoimmunity b | 50 (13.1) | 83 (15.1) | 56 (14.8) |
| Gestational age-sampling (weeks) | 14.1 (3.6) | 14.1 (3.5) | 14.0 (3.5) |
| Thyroid medication | |||
| No medication | 395 (84.6) | 562 (85.4) | 390 (84.8) |
| Thyroxine | 68 (14.6) | 91 (13.8) | 65 (14.1) |
| Anti-thyroid medication | 4 (0.9) | 4 (0.6) | 4 (0.9) |
| Yes, no defined | - | 1 (0.2) | 1 (0.2) |
| Child Characteristics | |||
| Sex (female) | 247 (52.9) | 337 (51.2) | 251 (54.6) |
| Birth weight (grams) | 3,186.9 (435.2) | 3,215.4 (450.4) | 3,193.5 (448.8) |
| Gestational age (weeks) | 38.2 (1.4) | 38.2 (1.6) | 38.1 (1.6) |
| Age at cognitive assessment (years) | 1.5 (0.1) | 4.2 (0.2) | 6.6 (0.3) |
Data presented as mean (standard deviation) for continuous variables and as frequency (%) on each category for categorical variables, unless otherwise mentioned
Subclinical hypothyroidism definition: fT4 concentration levels within the population-based, trimester-specific reference ranges and TSH above the upper trimester-specific limit and below 10 mIU/mL; hypothyroxinemia definition: TSH concentration levels within the trimester-specific reference ranges & fT4 < 5th percentile; hyperthyroxinemia definition: TSH concentration levels within the trimester-specific reference ranges & fT4 > 95th percentile and above the lower trimester-specific limit; thyroid autoimmunity: TSH and fT4 concentration levels within the population-based, trimester-specific reference ranges & TPO-Abs ≥ 35 IU/mL &/or Tg-Abs > 40 IU/mL
Figure 1.
Non-verbal cognitive development trajectories, Group-Based Trajectory Modelling, Rhea mother–child study, Crete, Greece.
Multivariate linear regression models showed that maternal hypothyroxinemia during pregnancy was associated with decreased verbal score at 4 years and motor score at 6 years of age (Table 2). Children exposed to maternal thyroid autoimmunity during gestation had decreased perceptual performance and motor score at 4 years (Table 2). We did not identify any association between maternal subclinical hypothyroidism or hyperthyroxinemia and child neuropsychological development. Children exposed to decreased concentration levels of maternal fT4 had decreased receptive communication scores at 18 months (Supplementary Table 3). Maternal concentration levels of TSH were not associated with any of the outcomes. The trajectory of continuously high non-verbal cognitive development from infancy to early childhood, was set as the base outcome for the multinomial logistic regression models. Children exposed to maternal thyroid autoimmunity during gestation had increased risk for belonging to suboptimal non-verbal cognitive development trajectories (Table 3).
Table 2.
Maternal hypothyroxinemia, subclinical hypothyroidism, hyperthyroxinemia and thyroid autoimmunity during pregnancy and child cognitive development at 18 months, 4 years and 6 years of age, Rhea mother-child study, Crete, Greece
| Hypothyroxinemia a | Subclinical Hypothyroidism a | Hyperthyroxinemia a | Thyroid Autoimmunity a |
|
|---|---|---|---|---|
| Neuropsychological development at 18 months cd | ||||
| Bayley Scales of Infant and Toddler Development-III | ||||
| Cognitive | 2.6 (−3.9, 9.0) | −0.9 (−6.4, 4.7) | 1.2 (−5.3, 7.7) | −3.6 (−7.9, 0.7) |
| Expressive communication | −4.1 (−10.6, 2.5) | −1.9 (−7.4, 3.6) | 4.3 (−2.5, 11.0) | 1.7 (−2.8, 6.1) |
| Receptive communication | −3.5 (−9.9, 2.9) | −0.2 (−5.6, 5.2) | 4.3 (−2.3, 10.8) | −0.7 (-5.1, 3.6) |
| Gross motor | 2.3 (−5.5, 10.1) | −0.5 (−7.1, 6.0) | 1.1 (−9.0, 6.8) | −2.3 (−7.5, 3.0) |
| Fine motor | −0.1 (−6.5, 6.3) | 1.4 (−4.1, 6.8) | 2.6 (−3.9, 9.0) | −3.1 (−7.4, 1.2) |
|
| ||||
| Neuropsychological development at 4 years ce | ||||
| McCarthy Scales of Children Abilities (MSCA) | ||||
| Verbal | −6.6 (−12.3, −0.9) b | −0.0 (−4.6, 4.5) | 3.3 (−2.2, 8.7) | 0.3 (-3.0, 3.5) |
| Perceptual | −0.5 (−6.5, 5.5) | 0.0 (−4.8, 4.8) | −0.7 (−6.4, 5.0) | −3.6 (−7.1, −0.2) b |
| Quantitative | −5.7 (−11.9, 0.4) | −2.5 (−7.4, 2.4) | −0.7 (−6.6, 5.1) | −1.5 (−5.1, 2.0) |
| General Cognitive | −4.9 (−10.6, 0.9) | −0.6 (−5.2, 3.9) | 1.0 (4.4, 6.5) | −1.4 (−4.7, 1.9) |
| Memory | −2.6 (−8.5, 3.3) | −2.0 (−6.6, 2.7) | 2.4 (−3.3, 8.0) | 1.7 (−1.7, 5.0) |
| Motor | 0.7 (−5.7, 7.1) | −0.1 (−5.2, 5.0) | −2.5 (−8.6, 3.6) | −4.5 (−8.2, −0.8) b |
|
| ||||
| Neuropsychological development at 6 years cf | ||||
| Raven’s Coloured Progressive Matrices (RCPM) | ||||
| Total score | −2.2 (−8.9, 4.5) | −0.2 (−6.2, 5.8) | −0.2 (−7.5, 7.0) | −3.3 (−7.5, 1.0) |
| Trail Making Test (TMT) | ||||
| Part A: log-transformed | 0.2 (−0.0, 0.5) | −0.2 (−0,4, 0.0) | −0.1 (−0.4, 0.2) | −0.1 (−0.3, 0.1) |
| Part B: log-transformed | 0.0 (−0.3, 0.3) | −0.0 (−0.3, 0.2) | 0.1 (−0.2, 0.4) | −0.0 (−0.2, 0.2) |
| Finger Tapping Test (FTP) | ||||
| Dominant hand | −2.0 (−10.0, 6.0) | −1.5 (−8.4, 5.4) | 1.8 (−7.0, 10.6) | −2.9 (−8.0, 2.2) |
| Non-dominant hand | −9.6 (−17.6, −1.5) b | −1.4 (−8.4, 5.6) | −3.0, (−11.7, 5.8) | −1.4 (−6.5, 3.8) |
Comparison group for hypothyroxinemia models: TSH concentration levels within the trimester-specific reference ranges & fT4 ≥ 5th percentile and below the upper trimester-specific limit; comparison group for subclinical hypothyroidism models: TSH and fT4 concentration levels within the population-based, trimester-specific reference ranges; comparison group for hyperthyroxinemia models: TSH concentration levels within the trimester-specific reference ranges & fT4 ≤ 95th percentile; comparison group for thyroid autoimmunity models: TSH and fT4 concentration levels within the population-based, trimester-specific reference ranges, TPO-Abs < 35 IU/mL & Tg ≤ 40 IU/mL
p < 0.05
Models adjusted for maternal age, maternal BMI, maternal origin, maternal educational level, maternal smoking status, maternal marital status at pregnancy, parity, child sex, gestational age at blood sampling-thyroid assessment & quality of assessment; additional adjustment for child’s age was included in the models with TMT and FTP outcomes
Hypothyroxinemia: N = 22, Subclinical Hypothyroidism: N = 29, Hyperthyroxinemia: N = 20, Thyroid autoimmunity: N = 50
Hypothyroxinemia: N = 26, Subclinical Hypothyroidism: N = 38, Hyperthyroxinemia: N = 28, Thyroid autoimmunity: N = 83
Hypothyroxinemia: N = 22, Subclinical Hypothyroidism: N = 27, Hyperthyroxinemia: N = 19, Thyroid autoimmunity: N = 56
Table 3.
Maternal hypothyroxinemia, subclinical hypothyroidism and thyroid autoimmunity during pregnancy and longitudinal trajectories of child non-verbal cognitive development (Group Based Trajectory Modelling) from 18 months to 6 years of age, Rhea mother-child study, Crete, Greece
| Non-verbal cognitive development trajectories cd |
||||||
|---|---|---|---|---|---|---|
| Low | High-Decreasing | Low-Increasing | ||||
|
|
||||||
| RRR | 95% CI | RRR | 95% CI | RRR | 95% CI | |
| Hypothyroxinemia a | 1.3 | (0.4, 3.7) | 1.7 | (0.7, 4.3) | 0.9 | (0.4, 2.2) |
| Subclinical Hypothyroidism a | 1.0 | (0.5, 2.2) | 0.6 | (0.3, 1.3) | 0.8 | (0.4, 1.7) |
| Hyperthyroxinemia | 1.4 | (0.5, 3.9) | 1.4 | (0.6, 3.7) | 1.6 | (0.6, 4.2) |
| Thyroid Autoimmunity a | 2.7 | (1.4, 5.2) b | 2.2 | (1.2, 4.0) b | 1.8 | (1.0, 3.2) b |
Comparison group for hypothyroxinemia models: TSH concentration levels within the trimester-specific reference ranges & fT4 ≥ 5th percentile and below the upper trimester-specific limit; comparison group for subclinical hypothyroidism models: TSH and fT4 concentration levels within the population-based, trimester-specific reference ranges; comparison group for hyperthyroxinemia models: TSH concentration levels within the trimester-specific reference ranges & fT4 ≤ 95th percentile and above the lower trimester-specific limit; comparison group for thyroid autoimmunity models: TSH and fT4 concentration levels within the population-based, trimester-specific reference ranges, TPO-Abs < 35 IU/mL & Tg ≤ 40 IU/mL
p < 0.05
Reference trajectory: continuously high non-verbal cognitive development from 18 months to 6 years
Models adjusted for maternal age, maternal BMI, maternal origin, maternal educational level, maternal smoking status, maternal marital status at pregnancy, parity, child sex, gestational age at thyroid assessment & quality of assessment
In sensitivity analyses, we repeated the main models excluding mothers who took thyroid medication during pregnancy (N = 96). We did not observe substantial differences between the results of sensitivity analyses and the results derived from the main analyses regarding the statistical significance of the associations and the magnitude of the associations (Supplementary Table 5). We also repeated the analyses in children exposed to elevated levels of maternal TPO-Antibodies and children exposed to elevated levels of maternal Tg-Antibodies, separately (Supplementary Table 6); we did not observe significant differences in comparison to the relevant main analyses (thyroid autoimmunity defined as elevated TPO-Abs &/or elevated Tg-Abs).
DISCUSSION
In this longitudinal cohort study, we found that maternal hypothyroxinemia is associated with decreased offspring verbal and motor ability scores. In addition, exposure to maternal thyroid autoimmunity was related with decreased perceptual performance and motor scores at 4 years of age. We also identified four trajectories of longitudinal non-verbal child cognitive development from infancy to early childhood and we demonstrated that maternal thyroid autoimmunity during pregnancy is associated with increased risk for children to belong in the “Low”, “High-Decreasing” and “Low-Increasing” trajectories, when compared to the continuously “High” non-verbal cognitive trajectory.
Current results confirm findings of previous observational studies which supported that maternal hypothyroxinemia is associated with decreased child neuropsychological development [7–10, 14, 15]. Even though thyroid hormones impact on fetal neurodevelopment through the interaction of T3 with nuclear receptors of the fetal nervous system cells, T3 is formed locally through the deiodination of maternal T4 by iodothyronine deiodinase enzymes in glial cells [2]. Furthermore, findings from animal studies have demonstrated that induced maternal hypothyroxinemia leads to atypical neuronal migration and structural alterations in the somatosensory cortex and hippocampus, regions which are important for learning, memory, basic perceptual skills and higher-order cognitive abilities [30, 31].
However, the significance of maternal hypothyroxinemia as a marker of thyroid dysfunction in pregnancy has been challenged due to methodological issues in the relevant studies and due to concerns about the clinical relevance of hypothyroxinemia. The methodological problems that limit the interpretation of the studies include the differences in the definition of hypothyroxinemia, the diagnostic inaccuracy of fT4 measurements due to the physiological changes caused by pregnancy and due to the biases of the widely used automated immunoassays, the normal variability of fT4 concentration depending on the gestational age and the inadequate adjustment of the models for possible confounders of the explored associations [32–34] . Furthermore, observational studies that did not identify any association of maternal hypothyroxinemia with several pregnancy outcomes [35], as well as Randomized Controlled Trials (RCTs) that did not find any cognition benefit in offspring exposed to maternal hypothyroxinemia or other mild thyroid dysfunction after levothyroxine medication [34, 36–38] have resulted in concerns about the clinical importance of maternal hypothyroxinemia and in the no treatment-recommendation by the latest guidelines for thyroid dysfunction management in pregnancy [32].
Our results also suggest that maternal thyroid autoimmunity is associated with decreased perceptual performance and motor scores at 4 years and with increased risk for disadvantageous non-verbal cognitive development from infancy to early childhood. Previous findings regarding the role of maternal thyroid autoimmunity on child neuropsychological development have been diverse [10, 17–20]; and it has been suggested that the different iodine status between populations may be the underlying cause of this heterogeneity [17]. Current results support the adverse impact of maternal thyroid autoimmunity on child neuropsychological development in a Greek, iodine sufficient population.
The observed associations between maternal thyroid autoimmunity and offspring neuropsychological development can be explained through an impact of elevated of maternal thyroid antibodies on the concentration levels of maternal fT4. It has previously been shown that elevated maternal TPO-antibodies impair thyroidal stimulation caused by human chorionic gonadotropin (HCG), which is a pregnancy-specific hormone which binds to TSH receptors and ensures fT4 availability during pregnancy [17, 39]. Even though, we did not identify any significant differences of maternal fT4 concentration levels between mothers with and without elevated thyroid antibodies, later fT4 insufficiency during pregnancy cannot be excluded. In addition, maternal TSH concentration levels were higher in women with thyroid autoimmunity in comparison with those with normal levels of antibodies; this difference suggests an impact of the elevated antibodies on thyroidal function. This combination of factors (i.e. elevated thyroid antibodies with high TSH concentration), regardless of the presence or the absence of subclinical hypothyroidism, may synergistically increase the risk of adverse pregnancy outcomes [39].
Findings from animal studies support the observed associations, since they have demonstrated that the primary brain regions affected by decreased availability of maternal fT4 are the hippocampus, which is involved in memory and learning, the cortex, which is involved in perceptual skills and higher-order cognitive abilities, and the cerebellum, which is involved in motor abilities and motor coordination [40]. We have found no association of maternal subclinical hypothyroidism and offspring cognitive and motor development corroborating the results of previous large population-based studies [7, 15] and failing to replicate the findings of other studies [10, 12, 16]. Current findings support that thyroid autoimmunity regardless the presence of subclinical hypothyroidism are linked with adverse offspring outcomes. We have also explored the role of increased maternal fT4 concentration and hyperthyroxinemia in offspring cognitive and motor development, since a recent MRI study has suggested that both low and high concentration of maternal fT4 in early pregnancy are related with decreased intelligence and decreased grey matter and cortex volume [8]. We did not find any association of maternal hyperthyroxinemia with offspring cognitive and motor development. This null finding may be the consequence of low power in the specific analysis, since we have few cases with hyperthyroxinemia in our population and further research is necessary to elucidate if hyperthyroxinemia is linked with adversary offspring developmental outcomes.
The strengths of the present study include its population-based prospective design, the long follow-up period, and the reliable valid, and comprehensive psychometric instruments that were used to assess child cognition. The use of GBTM to identify non-verbal cognitive development trajectories and explore any longitudinal impact of mild maternal thyroid dysfunction on child non-verbal cognition is another strength and a novelty of this study. Thyroid hormones were measured at a single-timepoint during gestation for the current analysis, therefore the measurements might reflect a transient dysfunction, even though single-timepoint and longitudinal assessments are highly correlated. Bias due to non-participation and loss to follow up might influence the results; non-participants did not differ regarding maternal thyroid parameters but they were more likely to have younger, non-Greek, multiparous, and less educated mothers compared with the participants of this study. In addition, residual confounding effect of unmeasured variables cannot be excluded, even though the models have been adjusted for multiple possible confounders. Mothers who took thyroid medication during pregnancy were not excluded from the main analyses, however sensitivity analyses excluding these participants did not support substantially different results.
In conclusion, our findings support that maternal hypothyroxinemia during pregnancy is associated with decreased offspring verbal and motor ability in early childhood. We also demonstrated that maternal thyroid autoimmunity is associated with impaired perceptual performance and motor ability in preschool age and increases the risk for adverse non-verbal cognitive development longitudinally, from infancy to early childhood, in a Greek, iodine-sufficient population. Further studies are needed to explore the association of maternal thyroid autoimmunity and child neuropsychological development in other populations to evaluate the factors that impact thyroid autoimmunity and the factors that modify its relation with child neuropsychological development, as well as to identify the biological mechanisms of the observed associations.
Supplementary Material
What is already known on this subject
Maternal mild thyroid dysfunction during pregnancy is associated with adverse offspring neuropsychological development. However, previous research is focused on maternal hypothyroxinemia, hence the information regarding the relation of maternal subclinical hypothyrodism, hyperthyroxinemia, and thyroid autoimmunity with offspring development is relatively scarce and still inconclusive.
What this study adds
Current findings replicate previous research results regarding the association of maternal hypothyroxinemia with decreased child cognitive and motor development.
We also demonstrate that maternal thyroid autoimmunity is associated with decreased perceptual performance and motor development in preschoolers, as well as with increased risk for suboptimal non-verbal cognitive development longitudinally, from infancy to early childhood, in an iodine-sufficient population.
We did not find any association between maternal subclinical hypothyroidism or hyperthyroxinemia with offspring cognitive and motor development.
Acknowledgements
The authors would like to thank all Rhea mother-child cohort participants for their generous collaboration.
Role of the funding source
The “Rhea” project was financially supported by European projects [EU FP6–2003-Food-3-NewGeneris], [EU FP6. STREP Hiwate], [EU FP7 ENV.2007.1.2.2.2. Project No 211250 Escape], [EU FP7–2008-ENV-1.2.1.4 Envirogenomarkers], [EU FP7-HEALTH-2009 - single stage CHICOS], [EU FP7 ENV.2008.1.2.1.6. Proposal No 226285 ENRIECO], [EU FP7.2007–2013- Project No 308333-The Helix Project] and the Greek Ministry of Health [Program of Prevention of obesity and neurodevelopmental disorders in preschool children, in Heraklion district, Crete, Greece: 2011–2014], [“Rhea Plus”: Primary Prevention Program of Environmental Risk Factors for Reproductive Health, and Child Health: 2012–15].
The funding bodies had no involvement in the production of this article.
Footnotes
Competing Interest: None declared.
REFERENCES
- 1.Rice D, Barone S Jr. Critical periods of vulnerability for the developing nervous system: evidence from humans and animal models. Environmental health perspectives 2000;108:511. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bernal J Thyroid hormone receptors in brain development and function. Nature Reviews Endocrinology 2007;3:249. [DOI] [PubMed] [Google Scholar]
- 3.De Escobar GM, Obregón MJ, Del Rey FE. Role of thyroid hormone during early brain development. European Journal of Endocrinology 2004;151:U25–U37. [DOI] [PubMed] [Google Scholar]
- 4.Lischinsky JE, Skocic J, Clairman H, et al. Preliminary findings show maternal hypothyroidism may contribute to abnormal cortical morphology in offspring. Frontiers in Endocrinology 2016;7:16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Samadi A, Skocic J, Rovet JF. Children born to women treated for hypothyroidism during pregnancy show abnormal corpus callosum development. Thyroid 2015;25:494–502. [DOI] [PubMed] [Google Scholar]
- 6.Fuster JM. Cortex and mind: Unifying cognition: Oxford university press; 2003. [Google Scholar]
- 7.Julvez J, Alvarez-Pedrerol M, Rebagliato M, et al. Thyroxine levels during pregnancy in healthy women and early child neurodevelopment. Epidemiology 2013;24:150–7. [DOI] [PubMed] [Google Scholar]
- 8.Korevaar TI, Muetzel R, Medici M, et al. Association of maternal thyroid function during early pregnancy with offspring IQ and brain morphology in childhood: a population-based prospective cohort study. The Lancet Diabetes & Endocrinology 2016;4:35–43. [DOI] [PubMed] [Google Scholar]
- 9.Ghassabian A, El Marroun H, Peeters RP, et al. Downstream effects of maternal hypothyroxinemia in early pregnancy: nonverbal IQ and brain morphology in school-age children. The Journal of Clinical Endocrinology & Metabolism 2014;99:2383–90. [DOI] [PubMed] [Google Scholar]
- 10.Li Y, Shan Z, Teng W, et al. Abnormalities of maternal thyroid function during pregnancy affect neuropsychological development of their children at 25–30 months. Clinical endocrinology 2010;72:825–9. [DOI] [PubMed] [Google Scholar]
- 11.Pop VJ, Brouwers EP, Vader HL, et al. Maternal hypothyroxinaemia during early pregnancy and subsequent child development: a 3‐year follow‐up study. Clinical endocrinology 2003;59:282–8. [DOI] [PubMed] [Google Scholar]
- 12.Williams F, Watson J, Ogston S, et al. Mild maternal thyroid dysfunction at delivery of infants born≤ 34 weeks and neurodevelopmental outcome at 5.5 years. The Journal of Clinical Endocrinology & Metabolism 2012;97:1977–85. [DOI] [PubMed] [Google Scholar]
- 13.Suárez-Rodríguez M, Azcona-San Julián C, de Aguilar VA. Hypothyroxinemia during pregnancy: the effect on neurodevelopment in the child. International Journal of Developmental Neuroscience 2012;30:435–8. [DOI] [PubMed] [Google Scholar]
- 14.Finken MJ, van Eijsden M, Loomans EM, et al. Maternal hypothyroxinemia in early pregnancy predicts reduced performance in reaction time tests in 5-to 6-year-old offspring. The Journal of Clinical Endocrinology & Metabolism 2013;98:1417–26. [DOI] [PubMed] [Google Scholar]
- 15.Henrichs J, Bongers-Schokking JJ, Schenk JJ, et al. Maternal thyroid function during early pregnancy and cognitive functioning in early childhood: the generation R study. The Journal of Clinical Endocrinology & Metabolism 2010;95:4227–34. [DOI] [PubMed] [Google Scholar]
- 16.Su P-Y, Huang K, Hao J-H, et al. Maternal thyroid function in the first twenty weeks of pregnancy and subsequent fetal and infant development: a prospective population-based cohort study in China. The Journal of Clinical Endocrinology & Metabolism 2011;96:3234–41. [DOI] [PubMed] [Google Scholar]
- 17.Derakhshan A, Korevaar TI, Taylor PN, et al. The Association of Maternal Thyroid Autoimmunity During Pregnancy With Child IQ. The Journal of Clinical Endocrinology & Metabolism 2018;103:3729–36. [DOI] [PubMed] [Google Scholar]
- 18.Pop VJ, de Vries E, van Baar AL, et al. Maternal thyroid peroxidase antibodies during pregnancy: a marker of impaired child development? The Journal of Clinical Endocrinology & Metabolism 1995;80:3561–6. [DOI] [PubMed] [Google Scholar]
- 19.Wasserman EE, Pillion JP, Duggan A, et al. Childhood IQ, hearing loss, and maternal thyroid autoimmunity in the Baltimore Collaborative Perinatal Project. Pediatric research 2012;72:525. [DOI] [PubMed] [Google Scholar]
- 20.Ghassabian A, Bongers-Schokking JJ, De Rijke YB, et al. Maternal thyroid autoimmunity during pregnancy and the risk of attention deficit/hyperactivity problems in children: the Generation R Study. Thyroid 2012;22:178–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Chatzi L, Leventakou V, Vafeiadi M, et al. Cohort Profile: The Mother-Child Cohort in Crete, Greece (Rhea Study). International journal of epidemiology 2017;46:1392–3k. [DOI] [PubMed] [Google Scholar]
- 22.Karakosta P, Chatzi L, Bagkeris E, et al. First-and second-trimester reference intervals for thyroid hormones during pregnancy in “Rhea” mother-child cohort, Crete, Greece. Journal of thyroid research 2011;2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Weiss LG, Oakland T, Aylward GP. Bayley-III clinical use and interpretation: Academic Press; 2010. [Google Scholar]
- 24.Royston P, Wright EM. A method for estimating age‐specific reference intervals (‘normal ranges’) based on fractional polynomials and exponential transformation. Journal of the Royal Statistical Society: Series A (Statistics in Society) 1998;161:79–101. [Google Scholar]
- 25.Koutra K, Chatzi L, Roumeliotaki T, et al. Socio-demographic determinants of infant neurodevelopment at 18 months of age: Mother-Child Cohort (Rhea Study) in Crete, Greece. Infant Behav Dev 2012;35:48–59. [DOI] [PubMed] [Google Scholar]
- 26.D M. Manual for the McCarthy Scales of Children’s Abilities New York: The Psychological Corporation; 1972. [Google Scholar]
- 27.Kampouri M, Kyriklaki A, Roumeliotaki T, et al. Patterns of Early-Life Social and Environmental Exposures and Child Cognitive Development, Rhea Birth Cohort, Crete, Greece. Child Dev 2017. [DOI] [PubMed]
- 28.Raven JC, Court JH, Raven JE. Raven’s coloured progressive matrices: Harcourt Assessment; 1998. [Google Scholar]
- 29.Lezak M, Howieson D, Loring D. Neuropsychological Assessment. Oxford University Press; New York: 2004. Orientation and Attention 2004:351–60. [Google Scholar]
- 30.Ausó E, Lavado-Autric Ra, Cuevas E, et al. A moderate and transient deficiency of maternal thyroid function at the beginning of fetal neocorticogenesis alters neuronal migration. Endocrinology 2004;145:4037–47. [DOI] [PubMed] [Google Scholar]
- 31.Lavado-Autric R, Ausó E, García-Velasco JV, et al. Early maternal hypothyroxinemia alters histogenesis and cerebral cortex cytoarchitecture of the progeny. The Journal of clinical investigation 2003;111:1073–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Alexander EK, Pearce EN, Brent GA, et al. 2017 Guidelines of the American Thyroid Association for the diagnosis and management of thyroid disease during pregnancy and the postpartum. Thyroid 2017;27:315–89. [DOI] [PubMed] [Google Scholar]
- 33.Mandel SJ, Spencer CA, Hollowell JG. Are detection and treatment of thyroid insufficiency in pregnancy feasible? Thyroid 2005;15:44–53. [DOI] [PubMed] [Google Scholar]
- 34.Moleti M, Trimarchi F, Vermiglio F. Doubts and concerns about isolated maternal hypothyroxinemia. Journal of thyroid research 2011;2011. [DOI] [PMC free article] [PubMed]
- 35.Casey BM, Dashe JS, Spong CY, et al. Perinatal significance of isolated maternal hypothyroxinemia identified in the first half of pregnancy. Obstetrics & Gynecology 2007;109:1129–35. [DOI] [PubMed] [Google Scholar]
- 36.Hales C, Taylor PN, Channon S, et al. Controlled Antenatal Thyroid Screening II: Effect of Treating Maternal Suboptimal Thyroid Function on Child Cognition. The Journal of Clinical Endocrinology & Metabolism 2018;103:1583–91. [DOI] [PubMed] [Google Scholar]
- 37.Lazarus JH, Bestwick JP, Channon S, et al. Antenatal thyroid screening and childhood cognitive function. New England Journal of Medicine 2012;366:493–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Momotani N, Iwama S, Momotani K. Neurodevelopment in Children Born to Hypothyroid Mothers Restored to Normal Thyroxine (T4) Concentration by Late Pregnancy in Japan: No Apparent Influence of Maternal T4 Deficiency. The Journal of Clinical Endocrinology & Metabolism 2012;97:1104–8. [DOI] [PubMed] [Google Scholar]
- 39.Korevaar TI, Medici M, Visser TJ, et al. Thyroid disease in pregnancy: new insights in diagnosis and clinical management. Nature Reviews Endocrinology 2017;13:610. [DOI] [PubMed] [Google Scholar]
- 40.Min H, Dong J, Wang Y, et al. Maternal hypothyroxinemia-induced neurodevelopmental impairments in the progeny. Molecular neurobiology 2016;53:1613–24. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.