Hostname: page-component-6b989bf9dc-zrclq Total loading time: 0 Render date: 2024-04-13T11:08:47.066Z Has data issue: false hasContentIssue false

Prenatal dietary patterns in relation to adolescent offspring adiposity and adipokines in a Mexico City cohort

Published online by Cambridge University Press:  19 January 2023

Erica Fossee
Affiliation:
Department of Nutritional Sciences, University of Michigan School of Public Health, Ann Arbor, MI, USA
Astrid N. Zamora
Affiliation:
Department of Nutritional Sciences, University of Michigan School of Public Health, Ann Arbor, MI, USA
Karen E. Peterson
Affiliation:
Department of Nutritional Sciences, University of Michigan School of Public Health, Ann Arbor, MI, USA Department of Environmental Health Sciences, University of Michigan School of Public Health, Ann Arbor, MI, USA
Alejandra Cantoral
Affiliation:
Health Department, Universidad Iberoamericana, Mexico City, Mexico
Wei Perng
Affiliation:
Department of Nutritional Sciences, University of Michigan School of Public Health, Ann Arbor, MI, USA Lifecourse Epidemiology of Adiposity and Diabetes (LEAD) Center, University of Colorado Denver Anschutz Medical Campus, Aurora, CO, USA Department of Epidemiology, Colorado School of Public Health, University of Colorado Denver Anschutz Medical Campus, Aurora, CO, USA
Martha M. Téllez-Rojo
Affiliation:
Center for Research on Nutrition and Health, National Institute of Public Health, Cuernavaca, Mexico
Libni A. Torres-Olascoaga
Affiliation:
Center for Research on Nutrition and Health, National Institute of Public Health, Cuernavaca, Mexico
Erica C. Jansen*
Affiliation:
Department of Nutritional Sciences, University of Michigan School of Public Health, Ann Arbor, MI, USA
*
Address for correspondence: Erica C. Jansen, PhD, MPH, Department of Nutritional Sciences, University of Michigan School of Public Health, 3863 SPH 1, 1415 Washington Heights, Ann Arbor, MI 48109, USA. Email: janerica@umich.edu
Rights & Permissions [Opens in a new window]

Abstract

Maternal diet during pregnancy has been associated with obesity among offspring. The extent to which trimester-specific dietary patterns are associated with markers of adiposity during adolescence remains unclear. We examined associations between prenatal diet patterns with adolescent offspring measures of adiposity and adipokines in 384 mother–adolescent dyads from the Mexico City ELEMENT cohort. Trimester-specific diet patterns were derived from principal component analysis of food frequency questionnaire data. Adolescent anthropometry and serum leptin and adiponectin were measured at 10–17 years. Three maternal diet patterns were identified: Prudent Diet (PD), high in fish and vegetables, the High Meat and Fat Diet (HMFD), high in pork and processed meats, and the Transitioning Mexican Diet (TMD), high in corn tortillas and sugar-sweetened beverages. Multiple linear regression was used to estimate sex-stratified associations among quartiles of diet patterns with adiposity and adipokines, adjusting for maternal marital status, education, and parity. First trimester TMD was associated with greater anthropometric measures and higher leptin in females, while third trimester HMFD was associated higher body fat percentage, triceps thickness, waist circumference, and leptin, but lower adiponectin among males. Contrary to expectation, there were positive associations between the trimester 1 PD pattern and anthropometric measurements in females, and for trimester 2 HMFD and TMD patterns with adipokines among males. Findings suggest maternal diet patterns may influence offspring adiposity markers during adolescence in a sex-specific manner.

Type
Original Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
© The Author(s), 2023. Published by Cambridge University Press in association with International Society for Developmental Origins of Health and Disease

Introduction

Developmental programing research suggests that the in utero nutritional environment is an important determinant of offspring adiposity. The Dutch Famine study demonstrated that maternal malnutrition during the early stages of gestation was associated with offspring obesity in adulthood. Reference Roseboom, de Rooij and Painter1 Other human studies link overnutrition during pregnancy with higher adiposity; to illustrate, higher animal protein intake during pregnancy has been associated with higher body mass index (BMI) in offspring at age 20. Reference Maslova, Rytter and Bech2 Other studies have related a prenatal diet higher in carbohydrates and sugar to higher BMI in early childhood. Reference Chen, Aris and Bernard3

However, there are multiple limitations to the current body of literature. First, existing research on the links between the in utero dietary environment and offspring metabolic health has primarily focused on offspring BMI. Reference Maslova, Rytter and Bech2 BMI is an imperfect surrogate of adiposity and metabolic risk Reference Javed, Jumean and Murad4,Reference Karchynskaya, Kopcakova and Klein5 ; thus, other body composition markers and adiposity-related hormones are needed to uncover mechanisms and later disease risk. For example, adipokines such as leptin and adiponectin are a family of hormones secreted by fat tissue that play a role in obesity development and the relationship between obesity and other metabolic conditions. Reference Ghantous, Azrak, Hanache, Abou-Kheir and Zeidan6 These two adipokines are only produced by adipose tissue. Reference Park, Park and Kim7 In particular, leptin has been shown to suppress appetite, while adiponectin may protect against insulin resistance. Reference Koleva, Orbetzova and Atanassova8 Considering the role of these hormones is imperative as they provide insight into the development of cardiovascular disease during adolescence and may be more sensitive to changes during puberty. Reference Nieuwenhuis, Pujol-Gualdo, Arnoldussen and Kiliaan9

Second, few studies have examined the role of the in utero dietary environment on offspring metabolic health during adolescence, a time of multiple developmental changes. Adolescence is a sensitive period for the development of obesity-related diseases. Reference Vickers, Breier, Cutfield, Hofman and Gluckman10 Specifically, puberty is characteristic of changes in adipose tissue distribution, placing some adolescents at a greater risk for cardiovascular disease than others. Reference Mihalopoulos, Holubkov, Young, Dai and Labarthe11 Moreover, studies have demonstrated sex differences in measures of total body fat (i.e., BMI and body fat percentage), peripheral and central fat during the progression of puberty. Reference Mihalopoulos, Holubkov, Young, Dai and Labarthe11

Third, previous research on offspring adiposity and prenatal diet has primarily focused on individual macronutrients Reference Gow, Ho and Burrows12 and micronutrients, Reference Wu, Zhang and Xiao13 or pre-specified dietary quality indices, such as glycemic index Reference Okubo, Crozier and Harvey14 or the Mediterranean diet. Reference Chatzi, Rifas-Shiman and Georgiou15 Data-driven techniques that identify naturally occurring diet patterns may offer a more comprehensive picture of prenatal diet. Further, the examination of diet patterns during each trimester allows for the possibility that diet may change throughout pregnancy.

To address gaps in the current literature, the present study aimed to investigate whether maternal trimester-specific dietary patterns were associated with measures of adiposity and adipokine levels of offspring during adolescence. Based on a priori knowledge, we hypothesized that healthier maternal dietary patterns would be associated with lower leptin, higher adiponectin, and lower adiposity. We further hypothesized that associations between maternal dietary patterns and adipokines would be sex-specific.

Methods

Study sample

The analytic sample included 384 mother–adolescent dyads from two sequential cohorts of the Early Life Exposures in Mexico to ENvironmental Toxicants (ELEMENT) study. Reference Perng, Tamayo-Ortiz and Tang16 Mothers were recruited during their first trimester of pregnancy between 1997 and 2003 from clinics in Mexico City. Research staff administered food frequency questionnaires (FFQs) and a sociodemographic questionnaire. This study is a secondary analysis of data from the original randomized controlled trial that began in 1994. Reference Hernandez-Avila, Gonzalez-Cossio and Palazuelos17

Data were available for the majority of the dyads (353) across all three trimesters. Since not all women enrolled in the study during the first trimester, the largest sample size available was during trimester 2 (n = 384; see Fig. 1 for more details).

Figure 1. Trimester 1 data were available for 379 mother–adolescent dyads. Eight dyads from trimester 1 did not have follow-up data for trimester 2, and 22 did not have follow-up data for trimester 3. Data from 384 dyads were available in trimester 2 and 13 of these dyads lacked trimester 1 data. Trimester 3 data were available for 366 dyads, 9 of which did not have trimester 1 data.

During a follow-up visit of the adolescent offspring in 2015–2017, adiposity measures and fasting leptin and adiponectin levels were collected. At the follow-up visit, adolescents ranged from 10 to 17 years old. Of the 554 adolescents measured at the follow-up visit, 379 had corresponding maternal FFQ data for at least one trimester of pregnancy. Figure 1 illustrates how this subcohort relates to the overall ELEMENT cohort, and Supplemental Figure S1 shows the overall timeline of the study. The Mexico National Institute of Public Health and the University of Michigan Human Subjects Committee approved all research protocols and procedures, and all participants provided informed consent.

Exposure assessment of maternal dietary patterns

Maternal prenatal diet was assessed each trimester via a semi-quantitative FFQ with 106 items validated for use among Mexican Spanish-speaking women of reproductive age. Reference Hernández, Aguirre and Serrano18,Reference Willett, Sampson and Stampfer19 The FFQ asked how often each standard portion size of a food item was consumed with nine possible responses: <1 time/month or never, 1–3 times/month, 1 time/week, 2–4 times/week, 5–6 times/week, 1 time/day, 2–3 times/day, 4–5 times/day, and ≥6 times/day.

Outcome assessment of offspring anthropometry and adipokines

Measures of adolescent adiposity included waist circumference (WC), triceps skinfold thickness (TS), body fat percentage (BF %), and BMI. Participants wore a clinical examination gown and were asked to remove hair ornaments, shoes, and socks according to the ELEMENT study protocol. Reference Perng, Tamayo-Ortiz and Tang16 Research assistants measured height to the nearest 0.5 cm with a Tanita stadiometer (Model WB-3000 m), weight to the nearest kg (InBody 270, Biospace, California, USA), WC to the nearest 0.1 cm at the iliac crest using a non-stretchable measuring tape (QM2000 QuickMedical; SECA model 201, Hamburg, Germany), and TS in mm (Lange calipers; Beta Technology, CA, USA) using standard anthropometry procedures. Reference de Onis, Onyango, Borghi, Siyam, Nishida and Siekmann20 BF% was estimated using bioelectrical impedance equipment (InBody 270, Biospace, CA, USA). Staff obtained duplicate measures for height, WC, and TS, and the average of two measurement values was used for analysis. Adolescent weight and height were used to calculate BMI (kg/m2) and standardized as a z score for sex and age using the World Health Organization (WHO) growth reference. Reference de Onis, Onyango, Borghi, Siyam, Nishida and Siekmann20

Fasting serum samples were collected from adolescents, frozen at −80°C, and shipped on dry ice to the University of Michigan Diabetes Research and Training Center Chemistry Lab (Ann Arbor, MI). Leptin and adiponectin levels were measured using RIA (Millipore).

Covariates

Mothers reported sociodemographic characteristics at enrollment, including age at pregnancy, parity, education, and marital status. Maternal age was operationalized into four categories: 14–23, 23–26, 26–30, and 30–44 years old. Parity was categorized as 0 or 1, 2, ≥3. Maternal education was separated into the following categories: did not complete secondary (<9 years), completed some high school (9 to <12 years), completed high school (12 years), higher education (>12 years). Marital status was coded as a dichotomous variable: married or civil union and single, separated, divorced or widowed. Covariates were determined using a priori knowledge. Reference Mantzoros, Sweeney and Williams21

Statistical analysis

Principal component analysis (PCA) was used to identify dietary patterns in each trimester. Individual food items from the FFQ were put into 40 nutritionally similar food groups based on a nutrient profile. Reference Newby and Tucker22 Raw response frequencies of consumption values (1–9) were converted to servings per day and then adjusted for total energy intake using the residual method. Reference Willett, Howe and Kushi23 Next, PCA was performed and rotated orthogonally to obtain uncorrelated factors. The number of factors to keep was determined based on interpretability and visual inspection of scree plots and eigenvalues >1. Food groups with factor loadings >+0.30 or <−0.30 were considered relevant for the pattern. Scores were calculated by multiplying the factor loadings by the frequency of consumption in each group and then summing. Diet patterns were named in accordance with a prior publication for consistency, but factor loadings and food pattern scores are slightly different (due to differences in the analytic sample sizes). Reference Zamora, Peterson and Tellez Rojo24 Diet pattern scores were categorized into quartiles for analysis. Mothers received a score for each diet pattern, and scores were uncorrelated with one another.

We examined crude associations between maternal characteristics and the exposure (dietary patterns) as well as outcomes (leptin, adiponectin, and the adiposity measures). We estimated Spearman correlations between adiposity measures and adipokines. Adolescent anthropometric variables and adipokines were analyzed in relation to maternal prenatal diet pattern quartiles using multiple linear regression. All analyses were sex-stratified. Quartile 1 (Q1) was used as the reference group, and the analysis was adjusted for maternal marital status, education, and parity. In sensitivity analysis, we further adjusted for pubertal status (menarche status for girls and testicular volume for boys, measured as described elsewhere Reference Chavarro, Watkins and Afeiche25 ) and total energy intake. All analyses were conducted using R Software (version 3.5.0; Boston, MA). A 2-sided α level of 0.05 was considered statistically significant.

Results

The mean ± SD age of adolescents was 13.7 ± 1.9 years, and 49.3% of the sample was female. Average adolescent leptin and adiponectin levels were 24.3 ± 18.2 ng/ml and 11.7 ± 3.9 ng/ml. Average triceps thickness was 18.5 ± 6.9 mm, WC was 78.8 ± 11.7 cm, BMI-Z-score was 0.6 ± 1.3, and BF % was 26.8 ± 9.8%. The mean ± SD age of mothers at delivery was 26.6 ± 5.6 years. There were no significant associations between maternal demographic characteristics and adolescent anthropometry measures. Maternal characteristics were not associated with adipokines, except higher maternal parity was associated with lower leptin levels (Supplemental Table S1).

Three prenatal maternal dietary patterns for trimester 1 were derived using PCA (Table 1). The Prudent Diet (PD) pattern was high in fish, tomatoes, potatoes, fruit, cruciferous vegetables, yellow vegetables, leafy vegetables, other vegetables, legumes, and soup and accounted for 7% variance. The High Meat and Fat Diet (HMFD) was high in beef, pork, processed meat, chili, chips, spread, and Mexican foods, low in milk, and accounted for 5% variance. The third pattern, the Transitioning Mexican Diet (TMD), was high in chilis, corn tortillas, sugar beverages, coffee, low in beef, spreads, and high-fat dairy and accounted for 5% variance. Similar patterns were observed in trimesters 2 and 3 (Supplemental Tables S2 and S3, respectively). Across pregnancy, there was moderately high correlation for the PD (rho between 0.4 and 0.6), but the other diet patterns were not as well-correlated over time. The three diets account for 17% variance of the maternal diets, which is similar to the variance explained by diet patterns in other studies (e.g., 15% Reference Perng, Fernandez and Peterson26 , 21% Reference Schwedhelm, Iqbal, Knüppel, Schwingshackl and Boeing27 and 30% Reference Jansen, Zhou and Perng28 ).

Table 1. Trimester 1 principal component loadings of foods for selected principal components

* Foods considered meaningful if loadings were greater than an absolute value of 0.3.

a Factor loadings and diet pattern scores multiplied by (−1) for interpretability (i.e., to load positively with tortillas and sugar-sweetened beverages).

Associations between maternal sociodemographic characteristics and maternal diet patterns are shown in Table 2. Maternal age was positively associated with the PD pattern, and education was positively associated with the PD and TMD patterns.

Table 2. Associations between maternal sociodemographic characteristics and maternal prenatal diet patterns scores

* P < 0.05.

a P values calculated from one-way ANOVA.

Spearman correlations between the adipokines and anthropometric measures were statistically significant (Supplement Table S4). Leptin was most highly correlated with BF% (r = 0.88), and adiponectin was negatively correlated with WC (r = −0.32).

PD pattern

There was a non-linear positive association between the trimester 1 PD pattern and WC in adjusted models (Table 3) among females only. To illustrate, females in Q2 had a 0.76 cm larger WC (95% CI: 0.32, 1.21) than the reference group, and quartile 4 (Q4) had a 0.26 cm larger WC (95% CI: −0.18, 0.70). In addition, PD was associated with BMIZ-score in females, with Q2 having a 0.58 greater BMIZ-score than the reference group (95% CI: 0.16, 0.99).

Table 3. Sex-stratified adjusted associations a between trimester 1 diet patterns and offspring adolescent anthropometric measures

* Denotes P < 0.05.

a From linear regression models adjusted for marital status, maternal education, and maternal parity.

b P for trend estimated by including a continuous ordinal variable representing quartiles of dietary pattern adherence into the linear regression models.

HMFD pattern

In trimester 2, there was a positive association between the HMFD pattern and male adiponectin, such that those in quartile 4 had 2.3 ng/mL higher adiponectin levels than those in quartile 1 (95% CI: 0.7, 3.9; Supplemental Table S6).

In trimester 3, there were positive associations between the HMFD pattern with body fat percentage, triceps thickness, and WC among males. To illustrate, those in quartile 4 versus quartile 1 had 0.7 (95% CI: 0.2, 1.1; p for trend = 0.05), 0.7 (95% CI: 0.2, 1.2; p for trend = 0.02), and 0.6 (95% CI: 0.1,1.0; p for trend = 0.21) higher body fat percentage, triceps thickness, and WC, respectively (Table 4). There was a positive association between the HMFD pattern and leptin among males, such that those in quartile 4 versus quartile 1 had 5.99 higher ng/mL leptin (95% CI: 0.07, 11.9; p for trend = 0.06) (Table 5). Finally, there was a non-linear inverse association with adiponectin; those in quartile 2 had −1.7 lower ng/mL adiponectin than those in quartile 1 (95% CI: −3.4, −0.02) (Table 5).

Table 4. Sex-stratified adjusted associations a between trimester 3 maternal prenatal diet patterns and offspring adolescent anthropometric measures

* Denotes P < 0.05.

a From linear regression models adjusted for marital status, maternal education, and maternal parity.

b P for trend estimated by including a continuous ordinal variable representing quartiles of dietary pattern adherence into the linear regression models.

Table 5. Adjusted associations between trimester 3 maternal diet patterns and offspring adolescent leptin and adiponectin levels by sex

* Denotes P < 0.05.

a From linear regression models adjusted for marital status, maternal education, and maternal parity.

b P trend estimated by including a continuous ordinal variable representing quartiles of dietary pattern adherence into the linear regression models.

TMD pattern

The trimester 1 prenatal TMD pattern was positively associated with fat distribution among female offspring (Table 3). To illustrate, Q4 of the TMD pattern was significantly greater than the reference for body fat percentage (0.48; 95% CI: 0.11, 0.85; p for trend = 0.09), triceps thickness (0.51 mm; 95% CI: 0.07, 0.95; p for trend = 0.15), and WC (0.52 cm; 95% CI: 0.05, 0.98; p for trend = 0.32). There was also a positive association between trimester 1 TMD pattern and leptin levels in females only (Table 6). To highlight, compared to female adolescents in the lowest quartile (Q1) of the prenatal maternal TMD pattern, the highest quartile (Q4) of the TMD pattern had 10.9 ng/mL higher leptin (95% CI: 3.8, 18.0; p for trend = 0.01).

Table 6. Adjusted associations between trimester 1 diet patterns and offspring adolescent leptin and adiponectin levels by sex

* Denotes P < 0.05.

a From linear regression models adjusted for marital status, maternal education, and maternal parity.

b P trend estimated by including a continuous ordinal variable representing quartiles of dietary pattern adherence into the linear regression models.

In trimester 2, the TMD pattern was associated with lower leptin among male offspring alone (p for trend 0.03; Supplemental Table S6), but it was not associated with other adiposity indicators.

In sensitivity analyses, there were no marked differences with respect to the direction, magnitude, or precision in the estimates after adjusting for pubertal status and total energy intake.

Discussion

In this cohort of Mexican adolescents, we observed a positive association between a trimester 1 TMD pattern and body fat percentage, triceps skinfolds, WC, and leptin levels in the female adolescent offspring. We also found positive associations between a trimester 3 HMFD pattern with anthropometric and adipokine adiposity indicators among boys only. There were also some associations in the unexpected direction, including positive associations between a trimester 1 PD pattern and female WC and BMIZ-score.

The trimester 1 TMD was associated with higher leptin levels in females and higher adiposity (body fat percentage, WC, and triceps skinfolds). This dietary pattern was high in sugar-sweetened beverages, and of the dietary patterns identified, it was most similar to a Western diet pattern. From this perspective, the study findings align with some previous childhood studies. For example, previous research in a Singaporean population found a positive association between higher maternal sugar intake and infancy and early childhood BMI. Reference Chen, Aris and Bernard3 Similarly, a previous study among a Dutch population found a positive association between maternal intake of sugar-sweetened beverages and childhood BMI and fat mass. Reference Jen, Erler and Tielemans29 Further, an animal study found that mice fed a diet rich in fats, sugar, and salt (a “Western” diet) during pregnancy gave birth to pups that went on to gain more weight and have more adipose tissue compared to the pups born from mothers eating a control diet. Reference Parlee and MacDougald30

In trimester 3, a maternal HMFD pattern was associated with higher levels of adiposity in male offspring. In this trimester, the HMFD pattern loaded highly on high-fat dairy, chicken, processed meat, spread, and low on tortillas and sugar-sweetened beverages. This association also has biologic plausibility, especially given that the late third trimester of intra-uterine life is a period of fat accumulation for the offspring. Reference Herrera and Amusquivar31 Further, a meta-regression analysis of animal model studies found that maternal high-fat diet was associated with increased adiposity during adulthood in male offspring, Reference Ribaroff, Wastnedge, Drake, Sharpe and Chambers32 aligning with our study finding. A separate animal study revealed that exposure to a maternal high-fat diet rich in monounsaturated fatty acids programed male offspring fatty acid metabolism and predisposed to greater adiposity in adulthood. Reference Seet, Yee, Jellyman, Han, Ross and Desai33

To our knowledge, no other studies have examined the links between prenatal dietary patterns and offspring adiposity during adolescence. Yet, results from human epidemiologic studies generally support our first and third trimester findings in that poor diet quality throughout gestation has been associated with increased neonatal offspring adiposity, Reference Shapiro, Kaar and Crume34 (although sex-specific findings were not reported). A separate US study reported inverse associations between maternal diet quality during pregnancy and offspring adiposity in early postnatal life. Reference Tahir, Haapala and Foster35

Some of the statistically significant associations were not in the hypothesized directions. First, the finding that higher trimester 1 PD was non-linearly associated with higher adiposity among females was not in line with expectations, given that the PD pattern was high in fruits and vegetables. A few previous studies have found that a healthier prenatal maternal diet was associated with a lower prevalence of overweight and obesity in adolescents. Reference Chen, Aubert and Shivappa36,Reference Fernández-Barrés, Romaguera and Valvi37 In contrast, higher adherence to a Healthy Eating Index during pregnancy was associated with lower adiponectin levels in male offspring at 4–7 years old. Reference Francis, Dabelea, Shankar and Perng38 Other studies report null associations between pregnancy diet and adolescent adiposity after accounting for confounders. Reference Chen, Aubert and Shivappa36 Thus, it is important to note that our study results may be impacted by unmeasured confounding such as sociodemographic and lifestyle factors specific to the present Mexico City cohort. One possible non-causal explanation for our results is that the PD pattern is associated with higher economic stability in the home, which could be associated with higher weight gain. We have previously shown that higher economic stability is associated with larger, more regular meals, potentially leading to higher overall caloric intake in the Mexican working-class context. Reference Jansen, Marcovitch and Wolfson39 Nonetheless, results remained the same when adjusted for energy intake in our analysis. A few other associations were also not in the expected direction, namely the trimester 2 associations among boys between HMFD and TMD diet patterns with adiponectin and leptin, respectively. The possibility of chance findings should not be discounted, given that statistically significant associations were not observed across the anthropometric indicators.

Sex differences in our associations are potentially due to variation in body composition and hormones impacting adipokine and anthropometric variables across sexes. For example, higher subcutaneous fat levels in females compared to males have been related to higher leptin levels, possibly due to greater mRNA expression of leptin in subcutaneous fat compared to visceral fat. Reference Roemmich, Clark and Berr40 In addition, sex steroid differences could play a role. Higher testosterone levels have been associated with lower leptin levels, irrespective of body fat percentage. Reference Elbers, Asscheman, Seidell, Frölich, Meinders and Gooren41

Many of the relationships between diet patterns and adipokines were null, which is also in line with a few other studies. Reference Mantzoros, Sweeney and Williams21,Reference Volberg, Harley and Aguilar42,Reference Sen, Rifas-Shiman and Shivappa43 To illustrate, when adipokine levels were measured in 9 year olds, no associations between prenatal maternal calorie, protein, total fat, saturated fat, fiber, sugar-sweetened beverage consumption, and offspring adipokine levels were found. Reference Volberg, Harley and Aguilar42 Similarly, there were no associations between blood cord adipokine levels and maternal adherence to the Mediterranean diet. Reference Mantzoros, Sweeney and Williams21 Additionally, another study found no association between child adiposity or leptin levels and a higher pro-inflammatory prenatal maternal diet. Reference Sen, Rifas-Shiman and Shivappa43 It is possible that only one measurement of adipokines during adolescence was not adequate to detect true underlying differences in metabolism.

The present study had various strengths and limitations. The longitudinal study design was one strength that allowed for inter-generational analysis. We also assessed pregnancy diet multiple times, which allowed us to consider the possibility that associations could vary by trimester. It was interesting to note that the non-PD patterns were not well-correlated over time, suggesting that women’s diets changed over pregnancy, as reported previously. Reference Ancira-Moreno, O’Neill and Rivera-Dommarco44 Other strengths included the relatively large sample size and objective measurements. The fact that the participants are from Mexico City means that the results may not be generalizable to other populations. The FFQ is subject to measurement error and could lead to bias, notably if measurement error differed with respect to maternal weight status. Finally, specific unmeasured confounders such as maternal physical activity during pregnancy could have influenced the results.

In conclusion, a trimester 1 TMD pattern was associated with higher leptin and adiposity among female offspring, while a trimester 3 High Meat and Fat Diet was associated with higher adiposity indicators among males within this Mexican cohort. There were also a few associations in the unexpected direction, including for the trimester 1 Prudent diet pattern with female WC and BMIZ-score, and trimester 2 Transitioning Mexican and High Meat and Fat diets with male adipokines. These findings point to the potential role of pregnancy diet patterns on the long-term anthropometry and metabolic health of offspring within the Mexican population.

Supplementary materials

For supplementary material for this article, please visit https://doi.org/10.1017/S2040174422000678

Acknowledgments

We gratefully acknowledge the American British Cowdray Hospital for the use of their facilities.

Financial support

US Environmental Protection Agency grant RD 83543601; National Institute of Environmental Health Sciences grants: P01 ES02284401, P30 ES017885, R24 ES028502, and R24 ES028502 Supplement; National Heart, Lung, Blood Institute grant K01 HL151673. This study was also supported and partially funded by the National Institute of Public Health/Ministry of Health of Mexico.

Conflicts of interest

The authors have no conflicts of interest to disclose.

Ethical standards

The study was approved by the Mexico National Institute of Public Health (INSP) and the University of Michigan Human Subjects Committee in accordance with relevant national and international guidelines for medical research.

References

Roseboom, T, de Rooij, S, Painter, R. The Dutch famine and its long-term consequences for adult health. Early Hum Dev. 2006; 82(8), 485491. DOI 10.1016/j.earlhumdev.2006.07.001.CrossRefGoogle ScholarPubMed
Maslova, E, Rytter, D, Bech, BH, et al. Maternal protein intake during pregnancy and offspring overweight 20 y later. Am J Clin Nutr. 2014; 100(4), 11391148. DOI 10.3945/ajcn.113.082222.CrossRefGoogle ScholarPubMed
Chen, LW, Aris, IM, Bernard, JY, et al. Associations of maternal macronutrient intake during pregnancy with infant BMI peak characteristics and childhood BMI. Am J Clin Nutr. 2017; 105(3), 705713. DOI 10.3945/ajcn.116.148270.CrossRefGoogle ScholarPubMed
Javed, A, Jumean, M, Murad, MH, et al. Diagnostic performance of body mass index to identify obesity as defined by body adiposity in children and adolescents: a systematic review and meta-analysis. Pediatr Obes. 2015; 10(3), 234244. DOI 10.1111/ijpo.242.CrossRefGoogle ScholarPubMed
Karchynskaya, V, Kopcakova, J, Klein, D, et al. Is BMI a valid indicator of overweight and obesity for adolescents? Int J Environ Res Public Health. 2020; 17(13), 4815. DOI 10.3390/ijerph17134815.CrossRefGoogle ScholarPubMed
Ghantous, CM, Azrak, Z, Hanache, S, Abou-Kheir, W, Zeidan, A. Differential role of leptin and adiponectin in cardiovascular system. Int J Endocrinol. 2015; 2015, e534320. DOI 10.1155/2015/534320.CrossRefGoogle ScholarPubMed
Park, KG, Park, KS, Kim, MJ, et al. Relationship between serum adiponectin and leptin concentrations and body fat distribution. Diabetes Res Clin Pract. 2004; 63(2), 135142. DOI 10.1016/j.diabres.2003.09.010.CrossRefGoogle ScholarPubMed
Koleva, DI, Orbetzova, MM, Atanassova, PK. Adipose tissue hormones and appetite and body weight regulators in insulin resistance. Folia Med (Plovdiv). 2013; 55(1), 2532. DOI 10.2478/folmed-2013-0002.CrossRefGoogle ScholarPubMed
Nieuwenhuis, D, Pujol-Gualdo, N, Arnoldussen, IAC, Kiliaan, AJ. Adipokines: a gear shift in puberty. Obes Rev. 2020; 21(6), e13005. DOI 10.1111/obr.13005.CrossRefGoogle ScholarPubMed
Vickers, MH, Breier, BH, Cutfield, WS, Hofman, PL, Gluckman, PD. Fetal origins of hyperphagia, obesity, and hypertension and postnatal amplification by hypercaloric nutrition. Am J Physiol Endocrinol Metab. 2000; 279(1), E837. DOI 10.1152/ajpendo.2000.279.1.E83.CrossRefGoogle ScholarPubMed
Mihalopoulos, NL, Holubkov, R, Young, P, Dai, S, Labarthe, DR. Expected changes in clinical measures of adiposity during puberty. J Adolesc Health. 2010; 47(4), 360366. DOI 10.1016/j.jadohealth.2010.03.019.CrossRefGoogle ScholarPubMed
Gow, ML, Ho, M, Burrows, TL, et al. Impact of dietary macronutrient distribution on BMI and cardiometabolic outcomes in overweight and obese children and adolescents: a systematic review. Nutr Rev. 2014; 72(7), 453470. DOI 10.1111/nure.12111.CrossRefGoogle ScholarPubMed
Wu, Y, Zhang, Q, Xiao, X. The effect and potential mechanism of maternal micronutrient intake on offspring glucose metabolism: an emerging field. Front Nutr. 2021; 8, 813. DOI 10.3389/fnut.2021.763809.CrossRefGoogle ScholarPubMed
Okubo, H, Crozier, SR, Harvey, NC, et al. Maternal dietary glycemic index and glycemic load in early pregnancy are associated with offspring adiposity in childhood: the Southampton Women’s Survey. Am J Clin Nutr. 2014; 100(2), 676683. DOI 10.3945/ajcn.114.084905.CrossRefGoogle ScholarPubMed
Chatzi, L, Rifas-Shiman, SL, Georgiou, V, et al. Adherence to the Mediterranean diet during pregnancy and offspring adiposity and cardiometabolic traits in childhood. Pediatr Obes. 2017; 12(S1), 4756. DOI 10.1111/ijpo.12191.CrossRefGoogle Scholar
Perng, W, Tamayo-Ortiz, M, Tang, L, et al. Early Life Exposure in Mexico to ENvironmental Toxicants (ELEMENT) project. BMJ Open. 2019; 9(8), e030427. DOI 10.1136/bmjopen-2019-030427.CrossRefGoogle ScholarPubMed
Hernandez-Avila, M, Gonzalez-Cossio, T, Palazuelos, E, et al. Dietary and environmental determinants of blood and bone lead levels in lactating postpartum women living in Mexico City. Environ Health Perspect. 1996; 104(10), 10761082.CrossRefGoogle ScholarPubMed
Hernández, M, Aguirre, J, Serrano, L. Alimentación de Obreros y sus Familias, 1983.   División de Nutrición de Comunidad, Instituto Nacional de la Nutricion Salvador Zubirán, Mexico.Google Scholar
Willett, WC, Sampson, L, Stampfer, MJ, et al. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol. 1985; 122(1), 5165. DOI 10.1093/oxfordjournals.aje.a114086.CrossRefGoogle ScholarPubMed
de Onis, M, Onyango, AW, Borghi, E, Siyam, A, Nishida, C, Siekmann, J. Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ. 2007; 85(9), 660667. DOI 10.2471/blt.07.043497.CrossRefGoogle ScholarPubMed
Mantzoros, CS, Sweeney, L, Williams, CJ, et al. Maternal diet and cord blood leptin and adiponectin concentrations at birth. Clin Nutr. 2010; 29(5), 622626. DOI 10.1016/j.clnu.2010.03.004.CrossRefGoogle ScholarPubMed
Newby, PK, Tucker, KL. Empirically derived eating patterns using factor or cluster analysis: a review. Nutr Rev. 2004; 62(5), 177203. DOI 10.1111/j.1753-4887.2004.tb00040.x.CrossRefGoogle ScholarPubMed
Willett, WC, Howe, GR, Kushi, LH. Adjustment for total energy intake in epidemiologic studies. Am J Clin Nutr. 1997; 65(4 Suppl), 1220S1228S; discussion 1229S–1231S. DOI 10.1093/ajcn/65.4.1220S.CrossRefGoogle ScholarPubMed
Zamora, AN, Peterson, KE, Tellez Rojo, MM, et al. Third trimester maternal dietary patterns are associated with sleep health among adolescent offspring in a Mexico City cohort. J Nutr. 2022; 152(6), 14871495.CrossRefGoogle Scholar
Chavarro, JE, Watkins, DJ, Afeiche, MC, et al. Validity of self-assessed sexual maturation against physician assessments and hormone levels. J Pediatr. 2017; 186, 172178.e3.CrossRefGoogle ScholarPubMed
Perng, W, Fernandez, C, Peterson, KE, et al. Dietary patterns exhibit sex-specific associations with adiposity and metabolic risk in a cross-sectional study in urban Mexican adolescents. J Nutr. 2017; 147(10), 19771985. DOI 10.3945/jn.117.256669.CrossRefGoogle Scholar
Schwedhelm, C, Iqbal, K, Knüppel, S, Schwingshackl, L, Boeing, H. Contribution to the understanding of how principal component analysis-derived dietary patterns emerge from habitual data on food consumption. Am J Clin Nutr. 2018; 107(2), 227235. DOI 10.1093/ajcn/nqx027.CrossRefGoogle Scholar
Jansen, EC, Zhou, L, Perng, W, et al. Vegetables and lean proteins-based and processed meats and refined grains–based dietary patterns in early childhood are associated with pubertal timing in a sex-specific manner: a prospective study of children from Mexico City. Nutr Res. 2018; 56, 4150. DOI 10.1016/j.nutres.2018.04.021.CrossRefGoogle Scholar
Jen, V, Erler, NS, Tielemans, MJ, et al. Mothers’ intake of sugar-containing beverages during pregnancy and body composition of their children during childhood: the Generation R Study. Am J Clin Nutr. 2017; 105(4), 834841. DOI 10.3945/ajcn.116.147934.CrossRefGoogle ScholarPubMed
Parlee, SD, MacDougald, OA. Maternal nutrition and risk of obesity in offspring: the Trojan horse of developmental plasticity. Biochim Biophys Acta. 2014; 1842(3), 495506. DOI 10.1016/j.bbadis.2013.07.007.CrossRefGoogle ScholarPubMed
Herrera, E, Amusquivar, E. Lipid metabolism in the fetus and the newborn. Diabetes Metab Res Rev. 2000; 16(3), 202210. DOI 10.1002/1520-7560(200005/06)16:3<202::AID-DMRR116>3.0.CO;2-#.3.0.CO;2-#>CrossRefGoogle ScholarPubMed
Ribaroff, GA, Wastnedge, E, Drake, AJ, Sharpe, RM, Chambers, TJG. Animal models of maternal high fat diet exposure and effects on metabolism in offspring: a meta-regression analysis. Obes Rev Off J Int Assoc Study Obes. 2017; 18(6), 673686. DOI 10.1111/obr.12524.CrossRefGoogle ScholarPubMed
Seet, EL, Yee, JK, Jellyman, JK, Han, G, Ross, MG, Desai, M. Maternal high-fat-diet programs rat offspring liver fatty acid metabolism. Lipids. 2015; 50(6), 565573. DOI 10.1007/s11745-015-4018-8.CrossRefGoogle ScholarPubMed
Shapiro, ALB, Kaar, JL, Crume, TL, et al. Maternal diet quality in pregnancy and neonatal adiposity: the Healthy Start Study. Int J Obes. 2016; 40(7), 10561062. DOI 10.1038/ijo.2016.79.CrossRefGoogle ScholarPubMed
Tahir, MJ, Haapala, JL, Foster, LP, et al. Higher maternal diet quality during pregnancy and lactation is associated with lower infant weight-for-length, body fat percent, and fat mass in early postnatal life. Nutrients. 2019; 11(3), 632. DOI 10.3390/nu11030632.CrossRefGoogle ScholarPubMed
Chen, LW, Aubert, AM, Shivappa, N, et al. Maternal dietary quality, inflammatory potential and childhood adiposity: an individual participant data pooled analysis of seven European cohorts in the ALPHABET consortium. BMC Med. 2021; 19(1), 33. DOI 10.1186/s12916-021-01908-7.CrossRefGoogle ScholarPubMed
Fernández-Barrés, S, Romaguera, D, Valvi, D, et al. Mediterranean dietary pattern in pregnant women and offspring risk of overweight and abdominal obesity in early childhood: the INMA birth cohort study. Pediatr Obes. 2016; 11(6), 491499. DOI 10.1111/ijpo.12092.CrossRefGoogle ScholarPubMed
Francis, EC, Dabelea, D, Shankar, K, Perng, W. Maternal diet quality during pregnancy is associated with biomarkers of metabolic risk among male offspring. Diabetologia. 2021; 64(11), 24782490. DOI 10.1007/s00125-021-05533-0.CrossRefGoogle ScholarPubMed
Jansen, EC, Marcovitch, H, Wolfson, JA, et al. Exploring dietary patterns in a Mexican adolescent population: a mixed methods approach. Appetite. 2020; 147, 104542. DOI 10.1016/j.appet.2019.104542.CrossRefGoogle Scholar
Roemmich, JN, Clark, PA, Berr, SS, et al. Gender differences in leptin levels during puberty are related to the subcutaneous fat depot and sex steroids. Am J Physiol Endocrinol Metab. 1998; 275(3), E543E551. DOI 10.1152/ajpendo.1998.275.3.E543.CrossRefGoogle Scholar
Elbers, JMH, Asscheman, H, Seidell, JC, Frölich, M, Meinders, AE, Gooren, LJG. Reversal of the sex difference in serum leptin levels upon cross-sex hormone administration in transsexuals*. J Clin Endocrinol Metab. 1997; 82(10), 32673270. DOI 10.1210/jcem.82.10.4284.Google ScholarPubMed
Volberg, V, Harley, KG, Aguilar, RS, et al. Associations between perinatal factors and adiponectin and leptin in 9-year-old Mexican-American children. Pediatr Obes. 2013; 8(6), 454463. DOI 10.1111/j.2047-6310.2012.00127.x.CrossRefGoogle ScholarPubMed
Sen, S, Rifas-Shiman, SL, Shivappa, N, et al. Associations of prenatal and early life dietary inflammatory potential with childhood adiposity and cardiometabolic risk in Project Viva. Pediatr Obes. 2018; 13(5), 292300. DOI 10.1111/ijpo.12221.CrossRefGoogle ScholarPubMed
Ancira-Moreno, M, O’Neill, MS, Rivera-Dommarco, JÁ., et al. Dietary patterns and diet quality during pregnancy and low birthweight: the PRINCESA cohort. Matern Child Nutr. 2020; 16(3), e12972. DOI 10.1111/mcn.12972.CrossRefGoogle ScholarPubMed
Figure 0

Figure 1. Trimester 1 data were available for 379 mother–adolescent dyads. Eight dyads from trimester 1 did not have follow-up data for trimester 2, and 22 did not have follow-up data for trimester 3. Data from 384 dyads were available in trimester 2 and 13 of these dyads lacked trimester 1 data. Trimester 3 data were available for 366 dyads, 9 of which did not have trimester 1 data.

Figure 1

Table 1. Trimester 1 principal component loadings of foods for selected principal components

Figure 2

Table 2. Associations between maternal sociodemographic characteristics and maternal prenatal diet patterns scores

Figure 3

Table 3. Sex-stratified adjusted associationsa between trimester 1 diet patterns and offspring adolescent anthropometric measures

Figure 4

Table 4. Sex-stratified adjusted associationsa between trimester 3 maternal prenatal diet patterns and offspring adolescent anthropometric measures

Figure 5

Table 5. Adjusted associations between trimester 3 maternal diet patterns and offspring adolescent leptin and adiponectin levels by sex

Figure 6

Table 6. Adjusted associations between trimester 1 diet patterns and offspring adolescent leptin and adiponectin levels by sex

Supplementary material: File

Fossee et al. supplementary material

Tables S1-S6 and Figure S1

Download Fossee et al. supplementary material(File)
File 54.4 KB