Received: 20 March 2017 | Revised: 30 May 2017 | Accepted: 27 June 2017

DOI: 10.1111/cen.13412

ORIGINAL ARTICLE

Maternal thyroid function, prepregnancy obesity and

gestational weight gain—The Generation R Study: A
prospective cohort study

Fernanda M. Collares1,2,3 | Tim I.M. Korevaar1,4,5 | Albert Hofman3 | Eric A.P. Steegers1,6 |
Robin P. Peeters4,5 | Vincent W.V. Jaddoe1,2,3 | Romy Gaillard1,2,3

1
The Generation R Study Group, Erasmus
University Medical Center, Rotterdam, Summary
The Netherlands Objective: Maternal prepregnancy obesity and excessive gestational weight gain are
2
Department of Paediatrics, Erasmus
associated with pregnancy complications. Thyroid function is related to differences in
University Medical Center-Sophia Children’s
Hospital, Rotterdam, The Netherlands body mass index (BMI) in adult populations. We examined the associations of maternal
3
Department of Epidemiology, Erasmus thyroid function in early pregnancy with maternal BMI and weight gain during
University Medical Center, Rotterdam,
pregnancy.
The Netherlands
4
Department of Internal Medicine, Erasmus Design and methods: In a population-­based prospective cohort study among 5726
University Medical Center, Rotterdam, mothers, we measured maternal TSH and FT4 levels at 13.5 weeks of gestation (95%
The Netherlands
5
range: 9.7-­17.6 weeks). Maternal weight was assessed before pregnancy and in each
Rotterdam Thyroid Center, Erasmus
University Medical Center, Rotterdam, trimester.
The Netherlands Results: Higher maternal TSH levels were associated with higher prepregnancy BMI
6
Department of Obstetrics and
(difference: 0.18 kg/m2 [95% CI: 0.01, 0.36] per SD increase in maternal TSH level)
Gynaecology, Erasmus University Medical
Center, Rotterdam, The Netherlands and higher total gestational weight gain (difference: 0.02 kg/wk [95% CI: 0.01, 0.03]
per SD increase in maternal TSH level). Higher maternal FT4 levels were associated
Correspondence
Romy Gaillard, The Generation R Study with lower prepregnancy BMI (difference: −0.44 kg/m2 [95% CI: −0.63, −0.26] per SD
Group (Na2918), Erasmus Medical Center,
increase in maternal FT4 level) and lower total gestational weight gain (difference:
Rotterdam, The Netherlands.
Email: r.gaillard@erasmusmc.nl −0.01 kg/wk [95% CI: −0.02, −0.01] per SD increase in maternal FT4 level). The asso-
Funding information ciations of maternal thyroid function with weight gain in early pregnancy were stronger
The Generation R Study is made possible by than those with weight gain in mid and late-­pregnancy. Maternal hypothyroidism was
financial support from the Erasmus Medical
Center, Rotterdam, the Erasmus University associated with higher prepregnancy BMI and early pregnancy weight gain, whereas
Rotterdam and the Netherlands Organization opposite effects were observed for maternal hyperthyroidism (P<.05).
for Health Research and Development.
Furthermore, this research has received Conclusions: Higher maternal TSH level and lower FT4 level in early pregnancy are
funding from the Netherlands Organization associated with higher prepregnancy BMI and higher gestational weight gain. Further
for Health Research and Development
NWOZonMw-­VIDI 016.136.361 to VWVJ; studies are needed to explore maternal and foetal consequences.
and a clinical fellowship project number
90700412 to RPP and the European KEYWORDS
Union’s Seventh Framework Programme
gestational weight gain, maternal prepregnancy body mass index, thyroid function
(FP7/2007-­2013), project Early Nutrition
under grant agreement no. 289346.

1 | INTRODUCTION higher BMI in adults.1,2 Differences in thyroid function may also influ-
ence the risk of overweight and obesity among women of reproduc-
Variations in thyroid function are related to differences in body mass tive age and subsequently affect maternal and childhood outcomes.
index (BMI). Higher TSH levels and lower FT4 levels are related to a Maternal obesity is associated with increased risks of gestational

Clinical Endocrinology. 2017;87:799–806. wileyonlinelibrary.com/journal/cen

© 2017 John Wiley & Sons Ltd | 799
800 | COLLARES et al.
hypertensive disorders, gestational diabetes, stillbirth and large size
3-6
for gestational age infants. In addition to maternal obesity, exces-
sive gestational weight gain also adversely affects maternal and foetal
pregnancy outcomes.4,5 Thus far, not much is known about the cor-
relations of maternal thyroid function during pregnancy with maternal
prepregnancy overweight and obesity and gestational weight gain. A
prospective cohort study among 1035 women showed that lower FT4
levels but not TSH levels are related to a higher maternal BMI through-
out pregnancy.7 Also, lower FT4 levels at mid-­pregnancy were related
to an increased risk of excessive gestational weight gain.7 However,
this previous study did not extensively adjust the analyses for potential
confounders and was unable to explore the effects of maternal thyroid
function on maternal weight gain during specific periods of pregnancy.
Therefore, we examined in a population-­based prospective cohort
study among 5726 pregnant women the associations of maternal TSH
and FT4 levels in early pregnancy with maternal prepregnancy BMI
and weight gain in early, mid and late-­pregnancy.
2 | METHODS
2.1 | Study design
This study was embedded in the Generation R Study, a population-­
based prospective cohort study from foetal life onwards in Rotterdam,
8
the Netherlands. It has been approved by the Medical Ethical
Committee of the Erasmus Medical Center in Rotterdam (MEC
198.782/2001/31). All mothers gave written consent. Pregnant
women with an expected delivery date from April 2002 to January
2006 were enrolled in the study. In total, 8879 women were enrolled
during pregnancy, of whom TSH and FT4 measurements in early preg-
nancy (<18 weeks) were available in 6013 subjects. As our interest
was in the effect of thyroid hormones within the normal variation in
low risk pregnancies, we excluded women with known thyroid disease
or thyroid medication usage, fertility treatment and nonsingleton live
birth (n=284). Women without data about weight before or during
pregnancy available were excluded (n=3). Thus, the population for
analysis included 5726 women (Fig. S1).
2.2 | Thyroid parameters
As previously described, maternal serum samples were obtained in
early pregnancy (median 13.5 weeks; 95% range: 9.7-­17.6 weeks).9
Maximally 3 h after sampling, plain tubes were centrifuged and serum
was stored at −80°C.10 TSH, FT4 and TT4 were determined in mater-
nal samples using chemiluminescence assays (Vitros ECI; Ortho Clinical
Diagnostics, Rochester, NY, USA). The intra-­ and interassay coefficients
of variation were <4.1% for TSH at a range of 3.97-­22.70 mIU/L, <5.4%
for FT4 at a range of 14.3-­25.0 pmol/L and <6.4% for TT4 at a range of
94.0-­151.0 nmol/L.11 As advocated by international guidelines, we used
population-­specific thyroid parameter reference ranges, details which
have been reported previously and applied to this same sample.9,12,13
Maternal reference ranges were 0.03-­4.04 mU/L for TSH and 10.4-­
22 pmol/L for FT4.9 We defined abnormal maternal thyroid function
as the following groups: overt hypothyroidism (TSH>4.04 mU/L and
FT4<10.4 pmol/L; n=23), subclinical hypothyroidism (TSH>4.04 mU/L
and FT4>10.4 and FT4<22 pmol/L; n=188) and overt hyperthyroidism
(TSH<0.03 mU/L and FT4>22 pmol/L, n=50). We combined overt hy-
pothyroidism and subclinical hypothyroidism in one category to study
the effects of abnormal lower thyroid activity.
2.3 | Maternal anthropometrics and weight gain
during pregnancy
Maternal anthropometrics were assessed in early (gestational age
<18 weeks), mid (gestational age 18-­25 weeks) and late-­pregnancy
(gestational age >25 weeks). Maternal height (cm) and weight (kg)
were measured without shoes and heavy clothing in each pregnancy
period, and BMI (kg/m2) was calculated. Information about maternal
weight just before pregnancy was obtained by questionnaire. As enrol-
ment in our study was in early pregnancy, we were not able to meas-
ure maternal weight before pregnancy. However, in our population for
analysis, 63% of all women were enrolled before a gestational age of
14 weeks. Correlation of prepregnancy weight, obtained by question-
naire, and weight measured at enrolment was 0.95 (P<.001) (regression
coefficient for this correlation: 0.93 [95% CI 0.93, 0.94]). Prepregnancy
BMI was categorized into four categories: underweight (<20 kg/m2),
normal weight (20-­24.9 kg/m2), overweight (25-­29.9 kg/m2) and obe-
sity (≥30 kg/m2). Weight gain until a gestational age of 30 weeks was
measured and available for 5428 mothers. Period-­specific weight gain
was calculated.14 Information about maximum weight during preg-
nancy was available in a subgroup of 2751 mothers and was assessed
by questionnaire two months after delivery. According to Institute of
Medicine guidelines, we defined excessive gestational weight gain in
relation to maternal prepregnancy BMI (for underweight and normal
weight mothers: total weight gain >16 kg; for overweight mothers: total
weight gain >11.5 kg and for obese mothers: total weight gain >9 kg).15
2.4 | Covariates
Information about maternal age, parity, educational level, ethnicity,
smoking, alcohol consumption and folic acid supplement use was
assessed by questionnaires.8 Total serum hCG levels were deter-
mined from blood samples drawn from the women at enrolment in
the study.16 First trimester maternal nutritional information was ob-
tained by food frequency questionnaire.17 Information on pregnancy
complications (gestational hypertension, preeclampsia and gestational
diabetes), gestational age at birth and birthweight was obtained from
medical records. Details of these procedures have been described
more specifically elsewhere.18
2.5 | Statistical analyses
We used logistic regression models to assess the associations of ma-
ternal TSH and FT4 levels with the risks of maternal prepregnancy un-
derweight, overweight and obesity and excessive gestational weight
gain. Next, we used linear regression models to assess the associations
COLLARES et al. | 801

of maternal thyroid hormone levels with maternal prepregnancy BMI, TABLE 1 Maternal characteristics (N=5726)a
total gestational weight gain and gestational weight gain in different pe-
Characteristics Value
riods of pregnancy in further detail. These regression models were first
Age, mean (SD), years 29.7 (5.0)
only adjusted for gestational age at thyroid function measurement, and
subsequently additionally adjusted for maternal socio-­demographic Height, mean (SD), cm 167.5 (7.4)

characteristics (maternal age, parity, educational level and ethnicity) Education, n (%)

and lifestyle-­related characteristics (maternal smoking, diet and folic Primary 514 (9.7)
acid intake during pregnancy; prepregnancy BMI was included in ges- Secondary 2406 (45.3)
tational weight gain models only). Models with gestational weight gain Higher 2391 (45)
as outcomes were also adjusted for pregnancy-­related characteristics Race/Ethnicity, n (%)
(pregnancy complications and birth characteristics). For all analyses, Dutch or European 3340 (60.8)
we assessed the associations per standard deviation change in ma- Surinamese 477 (8.7)
ternal TSH and FT4 levels. Finally, we used similar models to assess
Turkish 448 (8.2)
the associations of clinically abnormal maternal thyroid function with
Moroccan 330 (6.0)
maternal prepregnancy BMI and gestational weight gain. Sensitivity
Cape Verdean or Dutch Antilles 390 (7.1)
analyses using maternal TT4 levels instead of FT4 levels and maternal
Others 510 (9.3)
BMI measured at enrolment, instead of maternal prepregnancy BMI,
Parity: no. of nulliparous (%) 3260 (57.4)
were performed. We also performed a sensitivity analysis in which we
additionally adjusted the analyses for total hCG to assess if the ob- Folic acid supplement use, n (%) 3247 (74.6)

served associations were influenced by this hormone. Missing data of Diet

the covariates were imputed using multiple imputation. The percent- Total energy intake, mean (SD), kcal 2039 (564)
ages of missing values within the population for analysis were lower Smoking, n (%) 1430 (28.1)
than 15%, except for information about folic acid supplementation use Pregnancy complications
(24%). All analyses were performed using the Statistical Package of Gestational hypertension, n (%) 98 (2.0)
Social Sciences version 21.0 for Windows (SPSS Inc, Chicago, IL, USA). Preeclampsia, n (%) 97 (1.7)
Gestational diabetes, n (%) 58 (1.1)

3 | RESULTS Thyroid hormones measures

TSH, median (95% range), mIU/L 1.4 (0.03, 4.5)
3.1 | Subject characteristics FT4, median (95% range), pmol/L 14.8 (9.4, 22.2)
Weight measures
Characteristics of the study population are shown in Table 1. Median
values (95% range) for TSH and FT4 were 1.4 mIU/L (0.03, 4.5) and Prepregnancy body mass index, mean (SD), kg/m2 24.5 (4.4)

14.8 pmol/L (9.4, 22.2), respectively. The mean (SD) prepregnancy Prepregnancy weight, mean SD, kg 68.7 (13.0)

BMI was 24.5 kg/m² (4.4), and total weight gain during pregnancy was Maximum weight gain (SD), kg 15.2 (5.6)
15.2 kg (5.6). Early pregnancy weight gain, mean (SD), kg/wk 0.17 (0.27)
Mid-­pregnancy weight gain, mean (SD), kg/wk 0.49 (0.24)
Late-­pregnancy weight gain, mean (SD), kg/wk 0.57 (0.39)
3.2 | Maternal thyroid function and the risks of
maternal prepregnancy underweight, overweight and Birth characteristics

excessive gestational weight gain Sex, male % 50.6

Gestational age, median (95% range), weeks 40.1 (35.6, 42.3)
Figure 1 shows that a higher maternal TSH level in early pregnancy
Birthweight, mean (SD), g 3413 (564)
was associated with a lower risk of maternal prepregnancy under-
a
weight (P<.05), but not with the risk of maternal overweight or obe- Values are means (standard deviation), medians (95% range) or numbers
(percentages).
sity. A higher maternal TSH level was associated with a higher risk of
excessive gestational weight gain (Odds Ratio [OR] 1.28 [95% CI: 1.13,
1.45] per SD increase in maternal TSH level). A higher maternal FT4
level in early pregnancy was associated with a higher risk of prepreg-
3.3 | Maternal thyroid function, prepregnancy body
nancy underweight, and with a lower risk of prepregnancy overweight
mass index and weight gain during pregnancy
or obesity (OR 1.30 [95% CI: 1.15, 1.46] and OR 0.83 [95% CI: 0.75,
0.92], respectively, per SD increase in maternal FT4 level). A higher Table 2 shows that a higher maternal TSH level was associated with
maternal FT4 level in early pregnancy was associated with a lower risk a higher maternal prepregnancy BMI in the model adjusted for socio-­
of excessive gestational weight gain (OR 0.83 [95% CI: 0.73, 0.94], per demographic and lifestyle-­related characteristics (difference: 0.18 kg/m2
SD increase in maternal FT4 level). [95% CI: 0.01, 0.36] per SD increase in TSH level). A higher maternal TSH
802 | COLLARES et al.
level was also associated with a slightly higher total gestational weight
gain and early pregnancy weight gain (differences: 0.02 kg/wk [95% CI:
0.01, 0.03] and 0.03 kg/wk [95% CI: 0.01, 0.04] per SD increase in TSH
level). These associations were not explained by socio-­demographic, life-
style and pregnancy-­related characteristics. No associations were pre-
sent between maternal TSH level and late-­pregnancy weight gain.
Table 3 shows that, in the basic model, a higher maternal FT4 level
was associated with a lower prepregnancy BMI (difference: −0.54 kg/
2
m [95% CI: −0.72, −0.35] per SD increase in FT4 level). Additional
adjustment for socio-­demographic and lifestyle-­related characteris-
tics did not explain these associations. A higher maternal FT4 level
was associated with a lower total gestational weight gain and early
pregnancy weight gain in the basic model (differences: −0.02 kg/wk
[95% CI: −0.02, −0.01] and −0.04 kg/wk [95% CI: −0.05, −0.03], per
SD increase in maternal FT4 level). These associations remained signif-
icant after adjustment for socio-­demographic, lifestyle and pregnancy-­
related characteristics. No associations of maternal FT4 with mid-­ or
late-­pregnancy weight gain were present.
Table S1 shows the associations of TT4 with maternal prepreg-
nancy BMI and gestational weight gain. Similar results were present
when we used maternal TT4 level instead of FT4 level. When we used
maternal BMI measured at enrolment instead of maternal prepreg-
nancy BMI, similar results were present (results not shown). Tables S2
and S3 show that additional adjustment for maternal hCG level in early
pregnancy did not explain the observed associations of maternal TSH
and FT4 level with maternal weight outcomes.
3.4 | Abnormal maternal thyroid function,
prepregnancy body mass index and weight gain
during pregnancy
Table 4 shows that, as compared to mothers with a normal thyroid
function, maternal hypothyroidism was associated with a higher
2
prepregnancy BMI (difference 0.66 kg/m [95% CI: 0.03, 1.29] in the
F I G U R E 1 Maternal thyroid function
and the risks of prepregnancy underweight,
overweight and excessive gestational
weight gain. Associations of maternal TSH
and FT4 levels in early pregnancy with the
risk of maternal prepregnancy underweight,
overweight or obesity and excessive
gestational weight gain, as compared to
maternal normal weight and nonexcessive
gestational weight gain, respectively.
Results are based on logistic regression
models and show the risks of the adverse
maternal outcomes per increase in standard
deviation of maternal TSH and FT4 levels in
the basic models
fully adjusted model), whereas hyperthyroidism was associated with
a lower prepregnancy BMI (difference −1.79 kg/m2 [95% CI: −3.11,
−0.47] in the fully adjusted model). Maternal hypothyroidism was
associated with higher early pregnancy weight gain, while hyperthy-
roidism was associated with lower early pregnancy weight gain (dif-
ferences 0.06 kg/wk [95% CI: 0.02, 0.10] and −0.23 kg/wk [95% CI:
−0.31, −0.15] as compared to women with a normal thyroid function,
in the fully adjusted models, respectively). Abnormal maternal thyroid
function was not related to total gestational weight gain and mid-­ or
late-­pregnancy weight gain.
4 | DISCUSSION
In this prospective cohort study, we observed that a higher maternal
TSH level in early pregnancy was associated with a higher prepreg-
nancy BMI and an increased risk of excessive gestational weight gain,
whereas a higher maternal FT4 level was associated with a lower
prepregnancy BMI and lower risk of excessive gestational weight gain.
Associations of maternal thyroid function with gestational weight gain
were strongest for early pregnancy weight gain. These associations
were not explained by maternal socio-­demographic characteristics,
lifestyle-­related characteristics or pregnancy-­related characteristics.
4.1 | Methodological considerations
One of the strengths of this study was the prospective data collec-
tion from early pregnancy onwards. To our knowledge, this is the first
study that examined the associations of maternal thyroid parameters
in early pregnancy with detailed maternal weight measurements
throughout pregnancy. We had a large sample size of 5726 pregnant
women. The response rate at baseline for participation in the study
cohort was 61%.8 Pregnant women who participated were higher
educated, healthier and more frequently of Dutch origin than were
COLLARES et al. | 803

T A B L E 2 Associations of maternal TSH level with maternal weight tends to be underestimated, so some misclassification might
prepregnancy body mass index (BMI) and gestational weight gaina have occurred. However, we observed similar results when we

Differences in maternal weight used maternal BMI measured at enrolment instead of self-­reported
measures (95% confidence prepregnancy BMI. In addition, several cohort studies have shown
interval [CI]) per SD change in that assessment of maternal prepregnancy BMI by questionnaire is
maternal TSH level
a valid method with an adequate concordance of prepregnancy BMI
Prepregnancy BMI (kg/m2) with measured BMI at the first study visit.20-22 Maternal prepregnancy
b
Basic Model 0.01 (−0.17, 0.19) BMI and maternal thyroid function were assessed at enrolment in the
Adjusted Model c
0.18 (0.01, 0.36)* study. Due to this cross-­sectional design, we were not able to draw
Total gestational weight gain (kg/wk) conclusions about directions and causality of these observed associa-

Basic Model b
0.02 (0.01, 0.02)* tions. We had detailed information about a large number of potential
confounding factors in this study, but as in any observational study,
Adjusted Modelc 0.02 (0.01, 0.03)*
residual confounding due to other socio-­demographic and lifestyle-­
Mediator Modeld 0.02 (0.01, 0.02)*
related determinants might still be an issue.
Early pregnancy weight gain (kg/wk)
Basic Modelb 0.02 (0.01, 0.04)*
Adjusted Modelc 0.03 (0.01, 0.04)* 4.2 | Interpretation of main findings
d
Mediator Model 0.03 (0.01, 0.04)* Studies among nonpregnant populations have reported that higher
Mid-­pregnancy weight gain (kg/wk) TSH and lower FT4 levels are associated with a higher weight and
Basic Modelb 0.01 (0.00, 0.02) higher BMI.1,2,23-26 Lower FT4 levels within the normal range have also
c
Adjusted Model 0.01 (−0.01, 0.02) been associated with higher waist circumference, waist-­to-­hip ratio,
Mediator Model d
0.01 (−0.01, 0.02) total body fat mass and abdominal subcutaneous fat mass.2,24,27-29
Late-­pregnancy weight gain (kg/wk) Less information is available about the associations of thyroid
Basic Model b
0.02 (0.00, 0.04) function with prepregnancy obesity and gestational weight gain. A
recent study from Pop et al. among 1035 women found a significant
Adjusted Modelc 0.02 (−0.01, 0.04)
correlation between lower FT4 level in early pregnancy and higher pre-
Mediator Modeld 0.02 (0.00, 0.05)
pregnancy BMI (r=−.14, P<.001).7 The same study did not observe a
a
Results are from linear regression analyses; values are regression coeffi-
correlation between TSH levels and prepregnancy BMI. A study among
cients (95% CI) and reflect the differences in maternal prepregnancy BMI
(kg/m2), maternal total gestational weight gain (kg/wk) and maternal 9209 euthyroid women and 306 treated hypothyroid women with
weight gain in early, mid and late pregnancy (kg/wk) per standard deviation singleton pregnancies showed that higher second-­trimester maternal
change of maternal TSH level. FT4 level was related to a lower maternal weight.30,31 These effects
b
Basic model is adjusted for gestational age at thyroid measurement.
c
were not present for maternal TSH level. We observed that higher
Adjusted model is additionally adjusted for socio-­demographic character-
maternal TSH and lower maternal FT4 levels in early pregnancy were
istics (maternal age, parity, educational level and ethnicity) and lifestyle
characteristics (maternal smoking, diet and folic acid intake during preg- associated with a higher maternal prepregnancy BMI. These associa-
nancy). For gestational weight gain models, prepregnancy BMI is also in- tions were independent of a large number of socio-­demographic and
cluded in lifestyle-­related characteristics. lifestyle-­related characteristics. In line with these previous studies, our
d
Mediator model is additionally adjusted for pregnancy-­related character-
observed associations of maternal FT4 levels in early pregnancy with
istics (pregnancy complications and birth characteristics).
*P<.05. prepregnancy BMI were stronger and more consistent as compared to
the associations for maternal TSH levels.
The same study from Pop et al. showed that higher median TSH
and lower median FT4 levels in all trimesters were correlated with a
those who did not participate.8 The nonresponse would lead to biased higher amount of total weight gain during pregnancy. Women with
effect estimates if the associations were different between those in- higher FT4 levels during mid-­gestation had a lower total weight
cluded and not included in the analyses. This seems unlikely because gain.7 In line with this study, we found that higher maternal TSH
biased estimates in large cohort studies mainly arise from loss to and lower maternal FT4 levels in early pregnancy were associated
19
follow-­up rather than from nonresponse at baseline. However, the with an increased risk of excessive gestational weight gain during
nonresponse may have led to the selection of a more healthy popula- pregnancy.
tion and may affect the generalizability of our results. Maternal thy- Gestational weight gain is a complex trait, which reflects multi-
roid parameters were assessed only once during pregnancy. As our ple components, including maternal nutritional status, fat storage,
study population is a relatively healthy and low risk population, the fluid expansion and growth of the foetus, placenta and uterus.5,14,15
variation in maternal thyroid parameters was relatively small as com- This also partly depends upon the timing of gestational weight gain.
pared nonpopulation-­based studies. Maternal prepregnancy BMI was Maternal weight gain in early pregnancy largely reflects maternal fat
assessed by questionnaire at enrolment in the study. Self-­reported deposition, whereas weight gain in mid-­ and late-­pregnancy reflects
804 | COLLARES et al.

T A B L E 3 Associations of maternal FT4 level with maternal
prepregnancy body mass index (BMI) and gestational weight gaina
Differences in maternal weight
measures (95% confidence
interval [CI]) per SD change in
maternal FT4 level
T A B L E 4 Associations of maternal abnormal thyroid function
with maternal prepregnancy body mass index (BMI) and gestational
weight gaina
Differences in maternal weight measures
(95% confidence interval [CI]) for maternal
hypothyroidism and hyperthyroidism

Prepregnancy BMI (kg/m2) Maternal abnormal thyroid function

Basic Modelb −0.54 (−0.72, −0.35)*
Hypothyroidism Hyperthyroidism
Adjusted Modelc −0.44 (−0.63, −0.26)* n=211 n=50
Total gestational weight gain (kg/wk)
Prepregnancy BMI (kg/m2)
Basic Modelb −0.02 (−0.02, −0.01)*
Basic Modelb 0.39 (−0.25, 1.04) −1.61 (−2.94, −0.28)*
Adjusted Modelc −0.01 (−0.02, −0.01)* c
Adjusted Model 0.66 (0.03, 1.29)* −1.79 (−3.11, −0.47)*
Mediator Modeld −0.01 (−0.02, −0.01)*
Total gestational weight gain (kg/wk)
Early pregnancy weight gain (kg/wk)
Basic Modelb 0.01 (−0.02, 0.04) −0.05 (−0.11, 0.01)
Basic Modelb −0.04 (−0.05, −0.03)*
Adjusted Modelc 0.01 (−0.01, 0.04) −0.05 (−0.11, 0.01)
Adjusted Modelc −0.02 (−0.04, −0.01)* d
Mediator Model 0.02 (−0.01, 0.05) −0.04 (−0.09, 0.02)
Mediator Modeld −0.02 (−0.04, −0.01)*
Early pregnancy weight gain (kg/wk)
Mid-­pregnancy weight gain (kg/wk)
Basic Modelb 0.04 (0.00, 0.08) −0.24 (−0.33, −0.16)*
Basic Modelb −0.01 (−0.02, 0.00)
Adjusted Modelc 0.06 (0.02, 0.10)* −0.24 (−0.32, −0.15)*
Adjusted Modelc −0.01 (−0.03, 0.00) d
Mediator Model 0.06 (0.02, 0.10)* −0.23 (−0.31, −0.15)*
Mediator Modeld −0.01 (−0.03, 0.00)
Mid-­pregnancy weight gain (kg/wk)
Late-­pregnancy weight gain (kg/wk)
Basic Modelb 0.00 (−0.04, 0.03) 0.01 (−0.06, 0.08)
Basic Modelb −0.01 (−0.03, 0.01)
Adjusted Modelc 0.00 (−0.03, 0.04) 0.03 (−0.05, 0.11)
Adjusted Modelc −0.01 (−0.04, 0.01) d
Mediator Model 0.00 (−0.03, 0.04) 0.04 (−0.03, 0.12)
Mediator Modeld −0.01 (−0.04, 0.02)
Late-­pregnancy weight gain (kg/wk)
a
Results are from linear regression analyses; values are regression coeffi- Basic  Modelb 0.02 (−0.06, 0.10) −0.09 (−0.25, 0.07)
cients (95% CI) and reflect the differences in maternal prepregnancy BMI
Adjusted Modelc 0.02 (−0.06, 0.10) −0.09 (−0.27, 0.09)
(kg/m2), maternal total gestational weight gain (kg/wk) and maternal
d
weight gain in early, mid and late pregnancy (kg/wk) per standard deviation Mediator Model 0.02 (−0.06, 0.10) −0.06 (−0.24, 0.11)
change of maternal FT4 level. a
b Results are from linear regression analyses; values are regression coeffi-
Basic model is adjusted for gestational age at thyroid measurement.
c cients (95% CI) and reflect the differences in maternal prepregnancy BMI
Adjusted model is additionally adjusted for socio-­demographic character-
(kg/m2), maternal total gestational weight gain (kg/wk) and maternal
istics (maternal age, parity, educational level and ethnicity) and lifestyle
weight gain in early, mid and late pregnancy (kg/wk) among different ab-
characteristics (maternal smoking, diet and folic acid intake during preg-
normal maternal thyroid function groups compared to euthyroid mothers.
nancy). For gestational weight gain models, prepregnancy BMI is also in- b
Basic model is adjusted for gestational age at thyroid measurement.
cluded in lifestyle-­related characteristics. c
d Adjusted model is additionally adjusted for socio-­demographic characteris-
Mediator model is additionally adjusted for pregnancy-­related character-
tics (maternal age, parity, educational level and ethnicity) and lifestyle charac-
istics (pregnancy complications and birth characteristics).
teristics (smoking, diet and folic acid intake). For gestational weight gain
*P<.05.
models, prepregnancy BMI is also included in lifestyle-­related characteristics.
d
Mediator model is additionally adjusted for pregnancy-­related character-
istics (pregnancy complications and birth characteristics).
maternal and amniotic fluid expansion and growth of foetus, pla- *P<.05.
centa and uterus.15 Previous studies have shown that abnormal
maternal thyroid function during pregnancy is associated with an in- The direction and the mechanisms underlying the associations of
creased risk of gestational hypertensive disorders and adverse birth maternal thyroid function with maternal weight gain during pregnancy
outcomes, such as low birthweight and small size for gestational age are unclear. A bidirectional effect has been proposed. Alterations in
at birth.32,33 We observed that additional adjustment for gestational thyroid function influence the basal metabolic rate, which may lead to
hypertensive disorders, birthweight and gestational age at birth did changes in energy expenditure and body weight.2,34 Although thyroid
not explain the observed associations of maternal thyroid function function has a clear influence in metabolism, it has also been sug-
with gestational weight gain. In addition, we observed the stron- gested that weight gain may lead to altered thyroid function. Recently,
gest associations of maternal thyroid function with maternal weight a study among 3014 children using a Mendelian Randomization ap-
gain in early pregnancy. These findings suggest that maternal thy- proach showed that children with a genetically higher BMI had higher
roid function may be related to maternal fat accumulation during FT3 but not FT4 levels, which suggests that a higher BMI may caus-
pregnancy. ally lead to alterations in thyroid function.35 It has been proposed that
COLLARES et al.
alterations in leptin levels related to weight gain may influence thy-
roid function via stimulation on the hypothalamus-­pituitary-­thyroid
axis by leptin.7,23,34 In our study, we were not able to explore the
direction of the observed associations due to the cross-­sectional de-
sign of this study and we did not have information on maternal leptin
levels available. Maternal hCG level is a pregnancy-­specific hormone
that has a high homology with TSH and therefore has a weak affinity
for the TSH receptor. Via stimulation of the TSH receptor, increasing
levels of hCG stimulate the thyroid and this leads to an increase in
FT4 and a decrease in TSH levels during pregnancy. However, mater-
nal hCG level may also affect maternal weight during pregnancy, and
it has been suggested that the influence of hCG may partly explain
the weaker correlation of maternal TSH level with maternal weight
during pregnancy.7,36,37 When we additionally adjusted our analyses
for maternal hCG level, our results did not materially change which
suggests that the effects of thyroid hormones on maternal weight
gain are mediated via different biological pathways. Further studies
are needed to obtain more insight into the direction of the observed
associations and their complex underlying mechanisms, especially
among pregnant women.
The observed effect estimates for the associations of maternal
thyroid function with prepregnancy BMI and gestational weight
gain were small and are mainly of interest from an aetiological per-
spective. However, previous studies from the same cohort have
shown that already small increases in maternal prepregnancy BMI
and gestational weight gain are associated with increased risks of
gestational hypertensive disorders, gestational diabetes, delivering a
large size for gestational age infant and obesity in the offspring.5,14,38
Thus, maternal weight during pregnancy should be considered as an
important covariate in studies focused on the influence of mater-
nal thyroid function during pregnancy on pregnancy and offspring
outcomes.
5 | CONCLUSIONS
This population-­based prospective cohort study showed that higher
maternal TSH levels and lower FT4 levels in early pregnancy are as-
sociated with a higher prepregnancy body mass index and gestational
weight gain, with strongest associations present for early pregnancy
weight gain. Further studies are needed to assess the consequences
for maternal and offspring health.
ACKNOWLE DG E MEN TS
The Generation R Study is conducted by the Erasmus Medical Center
in close collaboration with the School of Law and Faculty of Social
Sciences of the Erasmus University Rotterdam, the Municipal Health
Service Rotterdam Area, Rotterdam, the Rotterdam Homecare
Foundation, Rotterdam and the Stichting Trombosedienst and
Artsenlaboratorium Rijnmond (STAR), Rotterdam. We gratefully ac-
knowledge the contribution of participating mothers, general practi-
tioners, hospitals, midwives and pharmacies in Rotterdam.
| 805
CO NFL I C T O F I NT ER ES T
The authors declare no conflict of interest.
REFERENCES
1. Knudsen N, Laurberg P, Rasmussen LB, et al. Small differences in
thyroid function may be important for body mass index and the
occurrence of obesity in the population. J Clin Endocrinol Metab.
2005;90:4019‐4024.
2. Svare A, Nilsen TI, Bjoro T, Asvold BO, Langhammer A. Serum TSH
related to measures of body mass: longitudinal data from the HUNT
Study, Norway. Clin Endocrinol (Oxf). 2011;74:769‐775.
3. Aune D, Saugstad OD, Henriksen T, Tonstad S. Maternal body
mass index and the risk of fetal death, stillbirth, and infant death
a systematic review and meta-­analysis. J Am Med Assoc. 2014;311:
1536‐1546.
4. Poston L, Harthoorn LF, van der Beek EM, Workshop CIE. Obesity
in pregnancy: implications for the mother and lifelong health of the
child. A consensus statement. Pediatr Res. 2011;69:175‐180.
5. Gaillard R, Durmus B, Hofman A, Mackenbach JP, Steegers EA,
Jaddoe VW. Risk factors and outcomes of maternal obesity and
excessive weight gain during pregnancy. Obesity (Silver Spring).
2013;21:1046‐1055.
6. McDonald SD, Han Z, Mulla S, Beyene J, Knowledge Synthesis Group.
Overweight and obesity in mothers and risk of preterm birth and
low birth weight infants: systematic review and meta-­analyses. BMJ.
2010;341:c3428.
7. Pop VJ, Biondi B, Wijnen HA, Kuppens SM, Lvader H. Maternal
thyroid parameters, body mass index and subsequent weight gain
during pregnancy in healthy euthyroid women. Clin Endocrinol (Oxf).
2013;79:577‐583.
8. Jaddoe VW, van Duijn CM, Franco OH, et al. The Generation R Study:
design and cohort update 2012. Eur J Epidemiol. 2012;27:739‐756.
9. Medici M, de Rijke YB, Peeters RP, et al. Maternal early pregnancy and
newborn thyroid hormone parameters: the Generation R study. J Clin
Endocrinol Metab. 2012;97:646‐652.
10. Jaddoe VW, Bakker R, van Duijn CM, et al. The Generation R Study
Biobank: a resource for epidemiological studies in children and their
parents. Eur J Epidemiol. 2007;22:917‐923.
11. Korevaar TIM, Medici M, de Rijke YB, et al. Ethnic differences in ma-
ternal thyroid parameters during pregnancy: the Generation R Study.
J Clin Endocrinol Metab. 2013;98:3678‐3686.
12. Stagnaro-Green A, Abalovich M, Alexander E, et al. Guidelines of
the American Thyroid Association for the diagnosis and manage-
ment of thyroid disease during pregnancy and postpartum. Thyroid.
2011;21:1081‐1125.
13. De Groot L, Abalovich M, Alexander EK, et al. Management of thy-
roid dysfunction during pregnancy and postpartum: an Endocrine
Society clinical practice guideline. J Clin Endocrinol Metab. 2012;
97:2543‐2565.
14. Gaillard R, Steegers EA, Franco OH, Hofman A, Jaddoe VW. Maternal
weight gain in different periods of pregnancy and childhood cardio-­
metabolic outcomes. The Generation R Study. Int J Obes (Lond).
2015;39:677‐685.
15. Rasmussen KM, Yaktine AL, (eds), & Committee to Reexamine IOM
Pregnancy Weight Guidelines, Food and Nutrition Board, Board on
Children, Youth and Families. Weight Gain During Pregnancy: Reexamining
the Guidelines. Washington, DC: The National Academies Press; 2009.
16. Korevaar TI, Steegers EA, de Rijke YB, et al. Reference ranges and
determinants of total hCG levels during pregnancy: the Generation R
Study. Eur J Epidemiol. 2015;30:1057‐1066.
17. Heppe DH, Medina-Gomez C, Hofman A, Franco OH, Rivadeneira F,
Jaddoe VW. Maternal first-­trimester diet and childhood bone mass:
the Generation R Study. Am J Clin Nutr. 2013;98:224‐232.
806 | COLLARES et al.
18. Coolman M, de Groot CJ, Jaddoe VW, Hofman A, Raat H, Steegers
EA. Medical record validation of maternally reported history of pre-
eclampsia. J Clin Epidemiol. 2010;63:932‐937.
19. Nohr EA, Frydenberg M, Henriksen TB, Olsen J. Does low participa-
tion in cohort studies induce bias? Epidemiology. 2006;17:413‐418.
20. Natamba BK, Sanchez SE, Gelaye B, Williams MA. Concordance
between self-­reported pre-­pregnancy body mass index (BMI) and
BMI measured at the first prenatal study contact. BMC Pregnancy
Childbirth. 2016;16:187.
21. Shin D, Chung H, Weatherspoon L, Song WO. Validity of prepreg-
nancy weight status estimated from self-­reported height and weight.
Matern Child Health J. 2014;18:1667‐1674.
22. Tomeo CA, Rich-Edwards JW, Michels KB, et al. Reproducibility and
validity of maternal recall of pregnancy-­related events. Epidemiology.
1999;10:774‐777.
23. Biondi B. Thyroid and obesity: an intriguing relationship. J Clin
Endocrinol Metab. 2010;95:3614‐3617.
24. De Pergola G, Ciampolillo A, Paolotti S, Trerotoli P, Giorgino R. Free
triiodothyronine and thyroid stimulating hormone are directly asso-
ciated with waist circumference, independently of insulin resistance,
metabolic parameters and blood pressure in overweight and obese
women. Clin Endocrinol. 2007;67:265‐269.
25. Iacobellis G, Ribaudo MC, Zappaterreno A, Iannucci CV, Leonetti F.
Relationship of thyroid function with body mass index, leptin, insulin
sensitivity and adiponectin in euthyroid obese women. Clin Endocrinol.
2005;62:487‐491.
26. Reinehr T. Obesity and thyroid function. Mol Cell Endocrinol.
2010;316:165‐171.
27. Alevizaki M, Saltiki K, Voidonikola P, Mantzou E, Papamichael
C, Stamatelopoulos K. Free thyroxine is an independent predic-
tor of subcutaneous fat in euthyroid individuals. Eur J Endocrinol.
2009;161:459‐465.
28. Marzullo P, Minocci A, Tagliaferri MA, et al. Investigations of thy-
roid hormones and antibodies in obesity: leptin levels are associ-
ated with thyroid autoimmunity independent of bioanthropometric,
hormonal, and weight-­related determinants. J Clin Endocrinol Metab.
2010;95:3965‐3972.
29. Shin JA, Mo EY, Kim ES, Moon SD, Han JH. Association between
lower normal free thyroxine concentrations and obesity phenotype in
healthy euthyroid subjects. Int J Endocrinol. 2014;2014:104318.
30. Haddow JE, Craig WY, Neveux LM, et al. Implications of high free
thyroxine (FT4) concentrations in euthyroid pregnancies: the FaSTER
trial. J Clin Endocrinol Metab. 2014;99:2038‐2044.
31. Haddow JE, Neveux LM, Palomaki GE, Lambert-Messerlian G, Malone
FD, D’Alton ME. An inverse relationship between weight and free thy-
roxine during early gestation among women treated for hypothyroid-
ism. Thyroid. 2015;25:949‐953.
32. Budenhofer BK, Ditsch N, Jeschke U, Gartner R, Toth B. Thyroid (dys-­)
function in normal and disturbed pregnancy. Arch Gynecol Obstet.
2013;287:1‐7.
33. Medici M, Korevaar TI, Schalekamp-Timmermans S, et al. Maternal
early-­pregnancy thyroid function is associated with subsequent hy-
pertensive disorders of pregnancy: the generation R study. J Clin
Endocrinol Metab. 2014;99:E2591‐E2598.
34. Feldt-Rasmussen U. Thyroid and leptin. Thyroid. 2007;17:413‐419.
35. Taylor PN, Richmond R, Davies N, et al. Paradoxical relation-
ship between body mass index and thyroid hormone levels: a
study using Mendelian randomization. J Clin Endocrinol Metab.
2016;101:730‐738.
36. Glinoer D, De Nayer P, Robyn C, Lejeune B, Kinthaert J, Meuris S.
Serum levels of intact human chorionic gonadotropin (HCG) and its
free alpha and beta subunits, in relation to maternal thyroid stimula-
tion during normal pregnancy. J Endocrinol Invest. 1993;16:881‐888.
37. Laurberg P, Andersen SL, Pedersen IB, Andersen S, Carle A. Screening
for overt thyroid disease in early pregnancy may be preferable
to searching for small aberrations in thyroid function tests. Clin
Endocrinol. 2013;79:297‐304.
38. Gaillard R, Steegers EA, Duijts L, et al. Childhood cardiometabolic out-
comes of maternal obesity during pregnancy: the Generation R Study.
Hypertension. 2014;63:683‐691.
S U P P O RT I NG I NFO R M AT I O N
Additional Supporting Information may be found online in the sup-
porting information tab for this article.
How to cite this article: Collares FM, Korevaar TIM, Hofman
A, et al. Maternal thyroid function, prepregnancy obesity and
gestational weight gain—The Generation R Study: A
prospective cohort study. Clin Endocrinol (Oxf). 2017;87:799–
806. https://doi.org/10.1111/cen.13412