Hostname: page-component-586b7cd67f-2brh9 Total loading time: 0 Render date: 2024-11-28T03:32:23.688Z Has data issue: false hasContentIssue false

Associations of maternal glucose markers in pregnancy with cord blood glucocorticoids and child hair cortisol levels

Published online by Cambridge University Press:  08 July 2022

Nathan Cohen
Affiliation:
College of Public Health, University of South Florida, Tampa, FL, USA
Sabrina Faleschini
Affiliation:
Division of Chronic Disease Research Across the Lifecourse, Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, MA, USA
Sheryl L. Rifas-Shiman
Affiliation:
Division of Chronic Disease Research Across the Lifecourse, Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, MA, USA
Luigi Bouchard
Affiliation:
Centre de recherche du Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Quebec, Canada Department of Biochemistry and Functional Genomics, Faculty of Medicine and Health Sciences, University of Sherbrooke, Sherbrooke, Quebec, Canada Clinical Department of Laboratory medicine, Centre intégré universitaire de santé et de services sociaux (CIUSSS) du Saguenay–Lac-St-Jean – Hôpital Universitaire de Chicoutimi, Saguenay, Quebec, Canada
Myriam Doyon
Affiliation:
Centre de recherche du Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Quebec, Canada
Olivier Simard
Affiliation:
Department of Biochemistry and Functional Genomics, Faculty of Medicine and Health Sciences, University of Sherbrooke, Sherbrooke, Quebec, Canada
Melina Arguin
Affiliation:
Centre de recherche du Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Quebec, Canada
Guy Fink
Affiliation:
Department of Medical Biology, University Health and Social Service Center of the Estrie, Fleurimont, Quebec, Canada
Amy C. Alman
Affiliation:
College of Public Health, University of South Florida, Tampa, FL, USA
Russell Kirby
Affiliation:
College of Public Health, University of South Florida, Tampa, FL, USA
Henian Chen
Affiliation:
College of Public Health, University of South Florida, Tampa, FL, USA
Ronee Wilson
Affiliation:
College of Public Health, University of South Florida, Tampa, FL, USA
Kimberly Fryer
Affiliation:
Department of Obstetrics and Gynecology, Morsani College of Medicine, University of South Florida, Tampa, FL, USA
Patrice Perron
Affiliation:
Centre de recherche du Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Quebec, Canada Department of Medicine, Faculty of Medicine and Health Sciences, University of Sherbrooke, Sherbrooke, Quebec, Canada
Emily Oken
Affiliation:
Division of Chronic Disease Research Across the Lifecourse, Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, MA, USA Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, MA, USA
Marie-France Hivert*
Affiliation:
Division of Chronic Disease Research Across the Lifecourse, Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, MA, USA Massachusetts General Hospital, Diabetes Unit, Boston, MA, USA
*
Address for correspondence: Marie-France Hivert, Division of Chronic Disease Research Across the Lifecourse, Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, MA, USA. Email: [email protected]
Rights & Permissions [Opens in a new window]

Abstract

Exposure to maternal hyperglycemia in utero has been associated with adverse metabolic outcomes in offspring. However, few studies have investigated the relationship between maternal hyperglycemia and offspring cortisol levels. We assessed associations of gestational diabetes mellitus (GDM) with cortisol biomarkers in two longitudinal prebirth cohorts: Project Viva included 928 mother–child pairs and Gen3G included 313 mother–child pairs. In Project Viva, GDM was diagnosed in N = 48 (5.2%) women using a two-step procedure (50 g glucose challenge test, if abnormal followed by 100 g oral glucose tolerance test [OGTT]), and in N = 29 (9.3%) women participating in Gen3G using one-step 75 g OGTT. In Project Viva, we measured cord blood glucocorticoids and child hair cortisol levels during mid-childhood (mean (SD) age: 7.8 (0.8) years) and early adolescence (mean (SD) age: 13.2 (0.9) years). In Gen3G, we measured hair cortisol at 5.4 (0.3) years. We used multivariable linear regression to examine associations of GDM with offspring cortisol, adjusting for child age and sex, maternal prepregnancy body mass index, education, and socioeconomic status. We additionally adjusted for child race/ethnicity in the cord blood analyses. In both Project Viva and Gen3G, we observed null associations of GDM and maternal glucose markers in pregnancy with cortisol biomarkers in cord blood at birth (β = 16.6 nmol/L, 95% CI −60.7, 94.0 in Project Viva) and in hair samples during childhood (β = −0.56 pg/mg, 95% CI −1.16, 0.04 in Project Viva; β = 0.09 pg/mg, 95% CI −0.38, 0.57 in Gen3G). Our findings do not support the hypothesis that maternal hyperglycemia is related to hypothalamic–pituitary–adrenal axis activity.

Type
Original Article
Copyright
© The Author(s), 2022. Published by Cambridge University Press in association with International Society for Developmental Origins of Health and Disease

Introduction

Gestational diabetes mellitus (GDM) is described by elevated glucose levels identified for the first time during pregnancy.Reference Casagrande, Linder and Cowie1,Reference Landon and Gabbe2 GDM affects approximately 7.6% of pregnancies in the United States,Reference Casagrande, Linder and Cowie1 and the prevalence has been rising in recent decades.Reference Chen, Wang and Ji3Reference Getahun, Nath, Ananth, Chavez and Smulian5 GDM is associated with complications at birth including macrosomiaReference He, Qin, Hu, Zhu, Tian and Li6,Reference Rice and Landon7 and long-term consequences in exposed offspring such as type 2 diabetesReference Damm, Houshmand-Oeregaard, Kelstrup, Lauenborg, Mathiesen and Clausen8 and greater adiposity later in life.Reference Wright, Rifas-Shiman, Rich-Edwards, Taveras, Gillman and Oken9,Reference Regnault, Gillman, Rifas-Shiman, Eggleston and Oken10

The developmental origins of health and disease hypothesisReference O'Donnell and Meaney11 suggests that long-term adverse outcomes may result from fetal programming due to exposures during in utero development. Prior studies found that hypothalamic–pituitary–adrenal (HPA) axis activity during childhood may be programmed during fetal development.Reference Reynolds12Reference Davis, Waffarn and Sandman21 Several animal studies also have implicated maternal hyperglycemia in stunted hippocampal neurodevelopment,Reference Hami, Sadr-Nabavi, Sankian, Balali-Mood and Haghir22,Reference Piazza, Segabinazi and de Meireles23 which could potentially dysregulate the inhibitory role of the hippocampus in HPA axis regulation.Reference Jacobson and Sapolsky24 However, few human studies have assessed offspring cortisol levels in the context of exposure to GDM or maternal hyperglycemia.Reference Jack-Roberts, Maples and Kalkan25Reference Mina, Lahti and Drake31 Furthermore, results of prior studies have been conflicting and have primarily been conducted in small cohorts.Reference Jack-Roberts, Maples and Kalkan25Reference Mina, Lahti and Drake31 Additionally, some of them have not accounted for potential confounding, such as maternal body mass index (BMI) and socioeconomic status.Reference Jack-Roberts, Maples and Kalkan25Reference Ma, Liu and Wu28,Reference Mina, Lahti and Drake31

The aim of this study was to examine the associations of maternal glucose markers in pregnancy with offspring cord blood glucocorticoids at birth and hair cortisol levels in mid-childhood and early adolescence using data from two longitudinal pre-birth cohorts. We hypothesized that high levels of maternal glycemia in pregnancy would be associated with higher levels of cord blood glucocorticoids and offspring hair cortisol levels.

Methods

Project Viva and Gen3G cohorts

Project Viva is an ongoing prospective prebirth cohort study of pregnant women and their children.Reference Oken, Baccarelli and Gold32 We recruited participants from obstetric clinics of Atrius Harvard Vanguard Medical Associates in eastern Massachusetts at their initial prenatal visits between 1999 and 2002. Inclusion criteria included singleton gestation, the ability to complete questionnaires written in English, plans to reside in eastern Massachusetts following delivery, and being at ≤22 weeks of gestation at enrollment. All study participants provided written informed consent at recruitment and postnatal follow-up visits, and children and teens provided verbal assent. The institutional review board at Harvard Pilgrim Health Care approved the study.Reference Oken, Baccarelli and Gold32

Of 2,128 live births included in Project Viva, we excluded 16 women with preexisting type 1 or type 2 diabetes and 45 with no data about pregnancy glucose status. We also excluded 1,139 mother–child pairs with missing values on all the cortisol outcomes. The 1,139 mother–child pairs included 251 nonwhite participants who had hair cortisol data available but no data available for any of the cord blood outcomes. Because of differences in hair growth rates and hair textures between individuals of different races that affect the cortisol assay,Reference Loussouarn33 hair cortisol analyses only included white children. In total, we included 928 pairs in the Project Viva study who were eligible for our current analyses (Supplementary Table S1).

The Genetics of Glucose Regulation in Gestation and Growth (Gen3G) cohortReference Guillemette, Allard and Lacroix34 is a prospective prebirth cohort of pregnant women and their children that began recruitment in 2010. We recruited Gen3G participants in Sherbrooke, Canada during a first trimester prenatal visit. Exclusion criteria included multiple pregnancies, known pregestational diabetes (type 1 or 2), diabetes diagnosed at first trimester, drug and/or alcohol abuse, and any other major medical conditions that could affect glucose regulation. All participants provided written informed consent prior to enrollment, and the institutional review board at Centre Hospitalier Universitaire de Sherbrooke approved the study. The Gen3G cohort is mostly of European descent and did not include enough nonwhite children with hair cortisol measurements available to be included in the present study. Therefore, analyses of hair cortisol in the Gen3G cohort only included white children. Gen3G analyses additionally excluded 39 children with undetectable hair cortisol values and those with missing second trimester oral glucose tolerance test (OGTT) results. In total, we included 313 children from the Gen3G cohort in the current analyses.

Diagnosis of GDM

In Project Viva, participants completed a two-step diagnostic procedure as part of routine clinical practice, in which GDM was diagnosed based on the Carpenter-Coustan criteria.Reference Carpenter and Coustan35 The diagnostic procedure consisted of a 50-g non-fasting glucose challenge test (GCT) followed by a 100-g, 3-h OGTT for those who failed the GCT (1-h glucose > 140 mg/dL). We deemed women who passed the initial GCT to have normal glucose tolerance (NGT). We characterized women who had normal glucose measurements at all four time points during the OGTT after previously failing the GCT as having isolated hyperglycemia (IH). We categorized women with one abnormal glucose measurement during the OGTT as having impaired glucose tolerance (IGT), and women with two or more abnormal measurements with GDM.

In the Gen3G study, participants underwent a fasting, 2-h, 75-g OGTT between the 24th and 28th week of gestation to diagnose GDM based on the diagnostic criteria recommended by the International Association of the Diabetes and Pregnancy Study Groups (IADPSG).Reference Metzger, Gabbe and Persson36 In this study, we categorized women into NGT and GDM groups based on these criteria.

Cord blood glucocorticoids

Cord blood glucocorticoid levels were measured only in the Project Viva study.Reference Huh, Andrew, Rich-Edwards, Kleinman, Seckl and Gillman37 A research staff member collected a sample of umbilical cord blood from each participant at delivery. We used radioimmunoassay techniques to measure cortisol and cortisone in the cord blood in duplicates. We calculated the ratio of cortisol to cortisone in the cord blood using the mean of duplicate cortisol and cortisone assays.

Hair cortisol

We measured child hair cortisol in both studies: in Project Viva at mid-childhood and early adolescenceReference Petimar, Rifas-Shiman, Hivert, Fleisch, Tiemeier and Oken38,Reference Petimar, Rifas-Shiman, Hivert, Fleisch, Tiemeier and Oken39 and in Gen3G at approximately five years of age. In Project Viva, research staff collected a sample of hair measuring 3 cm from the posterior vertex region of the scalp (as close to the scalp as possible). In the laboratory, lab personnel measured cortisol using liquid chromatography tandem mass spectrometry.Reference Gao, Stalder, Foley, Rauh, Deng and Kirschbaum40 We did not assess the intra-assay coefficient of variation because the study did not collect duplicate hair samples, but previous research has shown that it tends to be low when this method is used.Reference Gao, Stalder, Foley, Rauh, Deng and Kirschbaum40 We assayed the proximal 3 cm so that the cortisol measurements reflected cumulative stress over a period of approximately 3 months based on the assumption that hair grows at approximately 1 cm per month.Reference Russell, Koren, Rieder and Van Uum41 Additional details regarding the measurement of hair cortisol in Project Viva are described elsewhere.Reference Petimar, Rifas-Shiman, Hivert, Fleisch, Tiemeier and Oken39

In the Gen3G study, research staff collected hair samples from the posterior vertex region of the scalp as close to the scalp as possible. For the same reason as above, hair samples were cut to be 3 cm long. Details regarding the measurement of hair cortisol in the Gen3G study are provided in the Supplementary Methods section. In the analyses of the Gen3G data, we imputed hair cortisol values below the detectable limit of 1.0 pg/mg to be 0.5 pg/mg. We used this approach to impute these values because it is a commonly used, straightforward approach.

Other variables

In Project Viva, research staff collected demographic data using questionnaires and interviews during the study visits in pregnancy and at delivery, including maternal age, education, smoking status, household income, and race/ethnicity. We categorized child race/ethnicity based on reported data by the mothers in early childhood. We imputed missing child race/ethnicity values with maternal race/ethnicity. We obtained data on child sex from medical records. We categorized maternal education as a dichotomous variable based on whether the woman graduated college. We dichotomized household income using a cut point of $70,000, which was determined based on the distribution. We calculated maternal prepregnancy BMI using values of height and weight that were reported at study enrollment. We categorized smoking status based on whether the woman never smoked, formerly smoked, or smoked during pregnancy.

In Gen3G, research staff collected baseline demographic data during pregnancy including maternal age, type of employment, smoking status, height, and weight. We obtained data on child sex from medical records. At the 5-year follow-up visit, Gen3G research staff collected maternal education and categorized it into three categories: high school or less, some college or university education, and vocational. We categorized maternal employment into four groups: (1) management, business, or finance, (2) sciences or healthcare, (3) education, law, and social, community, and government, and (4) other. We grouped maternal employment into these categories because of their similarities. We calculated maternal BMI based on height and weight measured at the first visit in pregnancy. We dichotomized maternal smoking status based on whether the woman smoked in pregnancy, based on self-report at the first pregnancy visit.

Statistical Analyses

Statistical modeling

In all analyses, we used linear regression to assess the associations of maternal prenatal glycemia with offspring cortisol biomarkers. We performed all analyses involving categorical exposures using women with NGT as the reference group. To address skewed distribution of the hair cortisol outcomes, we used a logarithmic transformation. We selected covariates for inclusion in these models based on prior research.Reference Petimar, Rifas-Shiman, Hivert, Fleisch, Tiemeier and Oken39,Reference Cohen, Schwartz, Epel, Kirschbaum, Sidney and Seeman42Reference Rosmalen, Oldehinkel, Ormel, de Winter, Buitelaar and Verhulst47 We assessed effect modification by child sex in the models with categorical exposures with the inclusion of an interaction term and via stratified analyses.

For the cord blood outcomes (in Project Viva), minimally adjusted models were adjusted for child race/ethnicity, sex, and gestational age at birth. Fully adjusted analyses were additionally adjusted for maternal prepregnancy BMI, education, and household income. We also assessed effect modification by mode of delivery in the cord blood analyses using tests for interaction and via stratified analyses.

For the hair cortisol outcomes (in both cohorts), minimally adjusted models were corrected for child age at hair samples collection and sex. Fully adjusted models were additionally corrected for maternal prepregnancy BMI, education, and socioeconomic status. For the analyses of the change in hair cortisol from mid-childhood to early adolescence in Project Viva, we used child age at mid-childhood as a covariate. We used household income as a measure of socioeconomic status in Project Viva, and maternal employment type in Gen3G.

Sensitivity and additional analyses

We conducted sensitivity analyses for the cord blood outcomes in Project Viva by including only white children, to include a similar population that was included in the hair cortisol analyses. Using Project Viva data, we performed additional analyses to assess the extent to which the continuous glucose result from the GCT was associated with offspring cortisol biomarkers. In the Gen3G cohort, we performed additional analyses with maternal HbA1c and glucose levels from the OGTT as continuous exposures. Sensitivity analyses including maternal smoking status in the fully adjusted models did not change the strength or the magnitude of the associations in either cohort, so we decided not to include it in our models.

Missing data

We used multiple imputation with SAS PROC MIANALYZEReference Yuan48 to handle missing data. We did not impute missing data in the creation of the figures or the tabulations of summary statistics. We performed all analyses in SAS (version 9.4), and we created all figures in R (version 3.6.1).

Results

Descriptive statistics

In Project Viva, 928 mother–child pairs had at least one child cortisol outcome available for analysis (Table 1). Of these 928 mothers, 83.3% had NGT, 8.8% had IH, 2.7% had IGT, and 5.2% had GDM. Mean (SD) maternal age was 33.1 (4.5) years and mean (SD) prepregnancy BMI was 24.3 (4.8) kg/m2. A total of 449 (48%) children were male. At the time of collection of the hair cortisol measurements at mid-childhood and early adolescence, mean (SD) child age was 7.8 (0.8) and 13.2 (0.9) years, respectively. The distribution of child hair cortisol at each of these time points is shown in Fig. 1.

Fig. 1. Distributions of hair cortisol in Project Viva and Gen3G by pregnancy glucose status. The boxplots are displaying the median, quartiles, minimum, and maximum values of hair cortisol in the Project Viva and Gen3G by pregnancy glucose status.

Table 1. Demographic characteristics of pregnant women and their children by pregnancy glucose status in Project Vivaa

a NGT = normal glucose tolerance; IH = isolated hyperglycemia; IGT = impaired glucose tolerance; GDM = gestational diabetes mellitus.

b Hair cortisol outcomes are restricted to white participants (N = 632 mid-childhood, N = 563 early adolescence, N = 452 change).

Analyses involving the Gen3G data included 313 women, 9.3% of whom had GDM (Table 2). Mean (SD) maternal age was 28.5 (4.3) years and mean (SD) prepregnancy BMI was 25.0 (5.8) kg/m2. A total of 180 (58%) children were male. Mean (SD) child age at the time of hair collection was 5.4 (0.3) years. The distribution of child hair cortisol in the Gen3G cohort is shown in Fig. 1.

Table 2. Demographic characteristics of pregnant women and their children by pregnancy glucose status in the Gen3G studya

a Data are presented as mean (SD) or median (IQR) for continuous variables and N (%) for categorical variables.

b Prepregnancy BMI was missing for 4 participants.

c N = 97 participants (N = 88 NGT, N = 9 GDM) had hair cortisol measurements below the detectable limit of 1.0 pg/mg. For these participants, we imputed the values to be 0.5 pg/mg.

Associations between maternal glucose status in pregnancy and offspring cortisol markers

We did not observe any associations between maternal pregnancy glucose status and cord blood outcomes in Project Viva (Table 3). For example, the mean cord blood cortisol to cortisone ratio was 0.12 units (95% CI −0.14 to 0.38) higher in women with GDM compared to women with NGT. Our sensitivity analyses for the cord blood outcomes including only white children showed similar results (Supplementary Table S2). We did not observe any evidence of effect modification by mode of delivery for the cord blood outcomes.

Table 3. Associations of maternal glucose status during pregnancy with cord blood glucocorticoids and child hair cortisol in Project Vivaa

a Reporting β (95% CI) comparing to the reference group, which is women with NGT.

b Adjusted for child age and sex. Cord blood outcomes additionally adjusted for child race/ethnicity. In the cord blood analyses, we adjusted for gestational age instead of child age.

c Minimally adjusted model additionally adjusted for maternal prepregnancy BMI, education (≥college graduate vs. <college graduate), and maternal household income at enrollment (>$70,000 per year vs. ≤$70,000 per year).

d We logarithmically transformed all hair cortisol outcomes in the analyses. Hair cortisol analyses included only white children.

Hair cortisol measurements at mid-childhood and early adolescence in Project Viva did not differ between women with NGT and any of the other groups (Table 3). In the models for the hair cortisol outcomes in Project Viva, the effect estimates were close to zero and the confidence intervals all included the null (Table 3). For example, associations between GDM and child hair cortisol were null in fully adjusted models at mid-childhood (β = −0.56 pg/mg, 95% CI −1.16 to 0.04) and early adolescence (β = −0.24 pg/mg, 95% CI −0.79 to 0.31). We conducted additional analyses to investigate the extent to which continuous glucose values measured during the GCT were associated with offspring cortisol biomarkers. However, we only observed null associations (data not shown). Furthermore, we did not observe any evidence of effect modification by child sex for any of the outcomes that we assessed in Project Viva.

In the Gen3G study, we did not observe any associations between GDM and child hair cortisol at age five (β = 0.09 pg/mg, 95% CI −0.38 to 0.57) in fully adjusted analyses. We also did not observe an association of maternal prenatal HbA1c (β = −0.05%, 95% CI −0.54 to 0.44) with child hair cortisol. Results for any of the individual glucose values during OGTT (Table 4) were similarly null. We did not observe any evidence of effect modification by child sex in the Gen3G cohort.

Table 4. Associations of continuous measures of maternal glycemia with child hair cortisol at age five in the Gen3G studya

a Reporting β (95% CI).

b Adjusted for child age and sex.

c Minimally adjusted model additionally adjusted for maternal prepregnancy BMI, education, and maternal employment.

Discussion

Using data from two well phenotyped longitudinal prebirth cohorts, we examined associations of maternal glucose markers in pregnancy with cord blood glucocorticoids and offspring hair cortisol levels in mid-childhood and early adolescence. We did not find an association for any of these cortisol biomarkers in both Project Viva and Gen3G cohorts. These results do not support the hypothesis that prenatal exposure to maternal hyperglycemia is associated with dysfunction of the HPA axis in the child.

Few studies have examined the associations between maternal GDM or hyperglycemia and offspring cortisol. Our null findings are in line with some prior reports.Reference Jack-Roberts, Maples and Kalkan25,Reference Berglund, García-Valdés and Torres-Espinola27,Reference Ma, Liu and Wu28,Reference Krishnaveni, Veena and Jones30 For example, a prior study including 54 mother–child pairsReference Mina, Lahti and Drake31 found that maternal fasting glucose levels were not associated with child salivary cortisol reactivity in response to the Marshmallow Test, which is a test of delayed satisfaction that may be used to induce mild stress in young children. Our results are also consistent with a prior study of adverse maternal metabolic conditions which found that GDM was not associated with cord blood cortisol levels.Reference Berglund, García-Valdés and Torres-Espinola27 However, our results are in contrast with prior research that has identified lower cord blood cortisol levelsReference Chen, Guilmette and Luo26 and salivary cortisol levels at ages 11–13Reference Van Dam, Garrett and Schneider29 in children who were exposed to GDM. Notably, these results run counter to our hypothesis that GDM would be associated with higher child cortisol levels, but they are suggestive of potential dysregulation of HPA axis activity. These conflicting results may be due to differences in how cortisol was measured across studies. However, we are not aware of any previous studies that assessed child hair cortisol in relation to GDM. Hair cortisol is different from salivary cortisol because it measures long-term stress over a period of three months,Reference Russell, Koren, Rieder and Van Uum41 whereas salivary cortisol reflects acute stress assessed over a period of 20–30 min.Reference Puhakka and Peltola49 Given the limited amount of research in this area, our study expands upon prior knowledge regarding the relationship between maternal hyperglycemia and child cortisol levels.

This study has strengths and limitations. First, we used data from two prospective cohorts with follow-up into childhood and early adolescence. We also examined child hair cortisol at multiple time points to assess long-term stress axis activation. However, the small number of women with an abnormal pregnancy glucose status (in categorical exposures) is a major limitation that may have limited our statistical power. For example, our sample size in Project Viva would allow us to detect a mean difference of ≥109 nmol/L in cord blood cortisol from the GDM cases in contrast to NGT with 80% power under the assumption of a two-sided statistical test and a type I error rate of 5%. Additionally, we did not examine the association of GDM with diurnal cortisol variation, which could have produced different results because diurnal cortisol variation does not always reflect long-term HPA axis activity. We were unable to exclude participants on the basis of maternal prenatal glucocorticoid or scalp steroid use and were unable to control for shampoo usage in the hair cortisol analyses due to the lack of data on these factors. Moreover, GDM was diagnosed using different criteria in the Project Viva and Gen3G studies. Thus, we cannot conduct a pooled analysis combining data from both studies. However, we believe that a pooled analysis would have resulted in a null association, because we did not observe an association in either study. Finally, the hair cortisol analyses only included white children, so the results of those analyses cannot be generalized to children of other ethnicities.

In conclusion, the present study adds to the limited body of knowledge regarding child cortisol biomarkers in relation to in utero exposure to GDM and glycemic markers. We did not find any association between GDM and child cortisol levels assessed in childhood and early adolescence. Future studies using hair cortisol and other measures of HPA axis function should be conducted in other study populations to assess the generalizability of our results.

Supplementary materials

For supplementary material for this article, please visit https://doi.org/10.1017/S2040174422000381

Acknowledgments

The authors thank the participants and research staff in the Project Viva and Gen3G studies. The work presented in this article was performed as a partial fulfillment of the requirements for the PhD in Public Health with a concentration in Epidemiology for Nathan Cohen.

Financial support

This work was supported by the National Institutes of Health (E.O., grant number R01 HD 034568), (E.O., grant number UH3 OD023286); Fonds de recherche du Québec – Santé (M.F.H., grant number 20697); Canadian Institute of Health Research (M.F.H., grant number MOP-115071), (L.B., grant number PJT-152989); Programme d'aide financière interne (PAFI) from the Centre de recherche du Centre Hospitalier Universitaire de Sherbrooke (P.P., no grant number); a Diabète Québec grant (P.P.); and a Fonds de Recherche du Québec – Société et culture (FRQSC) postdoctoral fellowship award (S.F.). L.B. is a research scholar from the Fonds de recherche du Québec – Santé.

Conflicts of interest

None.

Ethical standards

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national guidelines on human experimentation (The U.S. Department of Health and Human Services, Office for Human Research Protections and the Tri-Council Policy Statement in Canada) and with the Helsinki Declaration of 1975, as revised in 2008, and has been approved by the institutional committees (Harvard Pilgrim Health Care Institutional Review Board and Centre Hospitalier Universitaire de Sherbrooke Institutional Review Board).

References

Casagrande, SS, Linder, B, Cowie, CC. Prevalence of gestational diabetes and subsequent Type 2 diabetes among U.S. women. Diabetes Res Clin Pract. 2018; 141(8), 200208.CrossRefGoogle ScholarPubMed
Landon, MB, Gabbe, SG. Gestational diabetes mellitus. Obstet Gynecol. 2011; 118(6), 13791393.CrossRefGoogle ScholarPubMed
Chen, P, Wang, S, Ji, J, et al. Risk factors and management of gestational diabetes. Cell Biochem Biophys. 2015; 71(2), 689694.CrossRefGoogle ScholarPubMed
Lavery, JA, Friedman, AM, Keyes, KM, Wright, JD, Ananth, CV. Gestational diabetes in the United States: temporal changes in prevalence rates between 1979 and 2010. BJOG. 2017; 124(5), 804813.CrossRefGoogle ScholarPubMed
Getahun, D, Nath, C, Ananth, CV, Chavez, MR, Smulian, JC. Gestational diabetes in the United States: temporal trends 1989 through 2004. Am J Obstet Gynecol. 2008; 198(5), 525.e521–525.CrossRefGoogle ScholarPubMed
He, XJ, Qin, FY, Hu, CL, Zhu, M, Tian, CQ, Li, L. Is gestational diabetes mellitus an independent risk factor for macrosomia: a meta-analysis? Arch Gynecol Obstet. 2015; 291(4), 729735.CrossRefGoogle ScholarPubMed
Rice, MM, Landon, MB. What we have learned about treating mild gestational diabetes mellitus. Semin Perinatol. 2016; 40(5), 298302.CrossRefGoogle ScholarPubMed
Damm, P, Houshmand-Oeregaard, A, Kelstrup, L, Lauenborg, J, Mathiesen, ER, Clausen, TD. Gestational diabetes mellitus and long-term consequences for mother and offspring: a view from Denmark. Diabetologia. 2016; 59(7), 13961399.CrossRefGoogle Scholar
Wright, CS, Rifas-Shiman, SL, Rich-Edwards, JW, Taveras, EM, Gillman, MW, Oken, E. Intrauterine exposure to gestational diabetes, child adiposity, and blood pressure. Am J Hypertens. 2009; 22(2), 215220.CrossRefGoogle ScholarPubMed
Regnault, N, Gillman, MW, Rifas-Shiman, SL, Eggleston, E, Oken, E. Sex-specific associations of gestational glucose tolerance with childhood body composition. Diabetes Care. 2013; 36(10), 30453053.CrossRefGoogle ScholarPubMed
O'Donnell, KJ, Meaney, MJ. Fetal origins of mental health: the developmental origins of health and disease hypothesis. Am J Psychiatry. 2017; 174(4), 319328.CrossRefGoogle ScholarPubMed
Reynolds, RM. Glucocorticoid excess and the developmental origins of disease: two decades of testing the hypothesis--2012 Curt Richter Award Winner. Psychoneuroendocrinology. 2013; 38(1), 111.CrossRefGoogle ScholarPubMed
Moisiadis, VG, Matthews, SG. Glucocorticoids and fetal programming part 1: outcomes. Nat Rev Endocrinol. 2014; 10(7), 391402.CrossRefGoogle ScholarPubMed
Vieau, D, Sebaai, N, Léonhardt, M, et al. HPA axis programming by maternal undernutrition in the male rat offspring. Psychoneuroendocrinology. 2007; 32 Suppl 1, S16S20.CrossRefGoogle ScholarPubMed
Sullivan, EL, Riper, KM, Lockard, R, Valleau, JC. Maternal high-fat diet programming of the neuroendocrine system and behavior. Horm Behav. 2015; 76(Suppl. 1), 153161.CrossRefGoogle ScholarPubMed
Kajantie, E, Feldt, K, Räikkönen, K, et al. Body size at birth predicts hypothalamic-pituitary-adrenal axis response to psychosocial stress at age 60 to 70 years. J Clin Endocrinol Metab. 2007; 92(11), 40944100.CrossRefGoogle ScholarPubMed
Krishnaveni, GV, Srinivasan, K. Maternal nutrition and offspring stress response-implications for future development of non-communicable disease: a perspective from India. Front Psychiatry. 2019; 10, 795.CrossRefGoogle ScholarPubMed
Jones, A, Godfrey, KM, Wood, P, Osmond, C, Goulden, P, Phillips, DI. Fetal growth and the adrenocortical response to psychological stress. J Clin Endocrinol Metab. 2006; 91(5), 18681871.CrossRefGoogle ScholarPubMed
Nazzari, S, Fearon, P, Rice, F, et al. Beyond the HPA-axis: exploring maternal prenatal influences on birth outcomes and stress reactivity. Psychoneuroendocrinology. 2019; 101(10), 253262.CrossRefGoogle ScholarPubMed
Laurent, H. Early calibration of the HPA axis by maternal psychopathology. Psychoneuroendocrinology. 2017; 78, 177184.CrossRefGoogle ScholarPubMed
Davis, EP, Waffarn, F, Sandman, CA. Prenatal treatment with glucocorticoids sensitizes the hpa axis response to stress among full-term infants. Dev Psychobiol. 2011; 53(2), 175183.CrossRefGoogle ScholarPubMed
Hami, J, Sadr-Nabavi, A, Sankian, M, Balali-Mood, M, Haghir, H. The effects of maternal diabetes on expression of insulin-like growth factor-1 and insulin receptors in male developing rat hippocampus. Brain Struct Funct. 2013; 218(1), 7384.CrossRefGoogle ScholarPubMed
Piazza, FV, Segabinazi, E, de Meireles, ALF, et al. Severe uncontrolled maternal hyperglycemia induces microsomia and neurodevelopment delay accompanied by apoptosis, cellular survival, and neuroinflammatory deregulation in rat offspring hippocampus. Cell Mol Neurobiol. 2019; 39(3), 401414.CrossRefGoogle ScholarPubMed
Jacobson, L, Sapolsky, R. The role of the hippocampus in feedback regulation of the hypothalamic-pituitary-adrenocortical axis. Endocr Rev. 1991; 12(2), 118134.CrossRefGoogle ScholarPubMed
Jack-Roberts, C, Maples, P, Kalkan, B, et al. Gestational diabetes status and dietary intake modify maternal and cord blood allostatic load markers. BMJ Open Diabetes Res Care. 2020; 8(1), e001468.CrossRefGoogle ScholarPubMed
Chen, L, Guilmette, J, Luo, ZC, et al. Placental 11β-HSD2 and cardiometabolic health indicators in infancy. Diabetes Care. 2019; 42(5), 964971.CrossRefGoogle ScholarPubMed
Berglund, SK, García-Valdés, L, Torres-Espinola, FJ, et al. Maternal, fetal and perinatal alterations associated with obesity, overweight and gestational diabetes: an observational cohort study (PREOBE). BMC Public Health. 2016; 16(1), 207.CrossRefGoogle ScholarPubMed
Ma, R, Liu, J, Wu, L, et al. Differential expression of placental 11β-hydroxysteroid dehydrogenases in pregnant women with diet-treated gestational diabetes mellitus. Steroids. 2012; 77(7), 798805.CrossRefGoogle ScholarPubMed
Van Dam, JM, Garrett, AJ, Schneider, LA, et al. Reduced cortical excitability, neuroplasticity, and salivary cortisol in 11-13-year-old children born to women with gestational diabetes mellitus. EBioMedicine. 2018; 31(1), 143149.CrossRefGoogle ScholarPubMed
Krishnaveni, GV, Veena, SR, Jones, A, et al. Exposure to maternal gestational diabetes is associated with higher cardiovascular responses to stress in adolescent indians. J Clin Endocrinol Metab. 2015; 100(3), 986993.CrossRefGoogle ScholarPubMed
Mina, TH, Lahti, M, Drake, AJ, et al. Maternal lipids in pregnancy are associated with increased offspring cortisol reactivity in childhood. Psychoneuroendocrinology. 2017; 83(2), 7983.CrossRefGoogle ScholarPubMed
Oken, E, Baccarelli, AA, Gold, DR, et al. Cohort profile: Project Viva. Int J Epidemiol. 2015; 44(1), 3748.CrossRefGoogle ScholarPubMed
Loussouarn, G. African hair growth parameters. Br J Dermatol. 2001; 145(2), 294297.CrossRefGoogle ScholarPubMed
Guillemette, L, Allard, C, Lacroix, M, et al. Genetics of Glucose regulation in Gestation and Growth (Gen3G): a prospective prebirth cohort of mother-child pairs in Sherbrooke, Canada. BMJ Open. 2016; 6(2), e010031.CrossRefGoogle ScholarPubMed
Carpenter, MW, Coustan, DR. Criteria for screening tests for gestational diabetes. Am J Obstet Gynecol. 1982; 144(7), 768773.CrossRefGoogle ScholarPubMed
Metzger, BE, Gabbe, SG, Persson, B, et al. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care. 2010; 33(3), 676682.CrossRefGoogle ScholarPubMed
Huh, SY, Andrew, R, Rich-Edwards, JW, Kleinman, KP, Seckl, JR, Gillman, MW. Association between umbilical cord glucocorticoids and blood pressure at age 3 years. BMC Med. 2008; 6(1), 25.CrossRefGoogle ScholarPubMed
Petimar, J, Rifas-Shiman, SL, Hivert, MF, Fleisch, AF, Tiemeier, H, Oken, E. Childhood hair cortisol concentration and early teen cardiometabolic outcomes. Pediatr Obes. 2020; 15(3), e12592.CrossRefGoogle Scholar
Petimar, J, Rifas-Shiman, SL, Hivert, MF, Fleisch, AF, Tiemeier, H, Oken, E. Prenatal and childhood predictors of hair cortisol concentration in mid-childhood and early adolescence. PLoS One. 2020; 15(2), e0228769.CrossRefGoogle ScholarPubMed
Gao, W, Stalder, T, Foley, P, Rauh, M, Deng, H, Kirschbaum, C. Quantitative analysis of steroid hormones in human hair using a column-switching LC-APCI-MS/MS assay. J Chromatogr B Analyt Technol Biomed Life Sci. 2013; 928(7), 18.CrossRefGoogle ScholarPubMed
Russell, E, Koren, G, Rieder, M, Van Uum, S. Hair cortisol as a biological marker of chronic stress: current status, future directions and unanswered questions. Psychoneuroendocrinology. 2012; 37(5), 589601.CrossRefGoogle ScholarPubMed
Cohen, S, Schwartz, JE, Epel, E, Kirschbaum, C, Sidney, S, Seeman, T. Socioeconomic status, race, and diurnal cortisol decline in the Coronary Artery Risk Development in Young Adults (CARDIA) study. Psychosom Med. 2006; 68(1), 4150.CrossRefGoogle ScholarPubMed
Castro-Diehl, C, Diez Roux, AV, Seeman, T, Shea, S, Shrager, S, Tadros, S. Associations of socioeconomic and psychosocial factors with urinary measures of cortisol and catecholamines in the Multi-Ethnic Study of Atherosclerosis (MESA). Psychoneuroendocrinology. 2014; 41, 132141.CrossRefGoogle ScholarPubMed
Liao, J, Brunner, EJ, Kumari, M. Is there an association between work stress and diurnal cortisol patterns? Findings from the Whitehall II study. PLoS One. 2013; 8(12), e81020.CrossRefGoogle ScholarPubMed
Bleker, LS, Roseboom, TJ, Vrijkotte, TG, Reynolds, RM, de Rooij, SR. Determinants of cortisol during pregnancy - the ABCD cohort. Psychoneuroendocrinology. 2017; 83(2), 172181.CrossRefGoogle ScholarPubMed
Gray, NA, Dhana, A, Van Der Vyver, L, Van Wyk, J, Khumalo, NP, Stein, DJ. Determinants of hair cortisol concentration in children: a systematic review. Psychoneuroendocrinology. 2018; 87(3–4), 204214.CrossRefGoogle ScholarPubMed
Rosmalen, JG, Oldehinkel, AJ, Ormel, J, de Winter, AF, Buitelaar, JK, Verhulst, FC. Determinants of salivary cortisol levels in 10-12 year old children; a population-based study of individual differences. Psychoneuroendocrinology. 2005; 30(5), 483495.CrossRefGoogle ScholarPubMed
Yuan, YC. Multiple imputation for missing data: concepts and new development. In: Proceedings of the Twenty-Fifth Annual SAS Users Group International Conference, 2000, Cary (NC).Google Scholar
Puhakka, IJA, Peltola, MJ. Salivary cortisol reactivity to psychological stressors in infancy: a meta-analysis. Psychoneuroendocrinology. 2020; 115(12), 104603.CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1. Distributions of hair cortisol in Project Viva and Gen3G by pregnancy glucose status. The boxplots are displaying the median, quartiles, minimum, and maximum values of hair cortisol in the Project Viva and Gen3G by pregnancy glucose status.

Figure 1

Table 1. Demographic characteristics of pregnant women and their children by pregnancy glucose status in Project Vivaa

Figure 2

Table 2. Demographic characteristics of pregnant women and their children by pregnancy glucose status in the Gen3G studya

Figure 3

Table 3. Associations of maternal glucose status during pregnancy with cord blood glucocorticoids and child hair cortisol in Project Vivaa

Figure 4

Table 4. Associations of continuous measures of maternal glycemia with child hair cortisol at age five in the Gen3G studya

Supplementary material: File

Cohen et al. supplementary material

Cohen et al. supplementary material

Download Cohen et al. supplementary material(File)
File 14.8 KB