Preconception Cardiometabolic Markers and Birth Outcomes Among Women in the Hispanic Community Health Study/Study of Latinos

Abstract

Background:

Associations between preconception cardiometabolic markers and birth outcomes have been noted, but data are scarce for Hispanics/Latinos. We examined the association between preconception cardiometabolic markers, birthweight and preterm birth among U.S. Hispanic/Latina women.

Materials and Methods:

The Hispanic Community Health Study/Study of Latinos is a cohort study of U.S. adults 18–74 years of age, including 3,798 women of reproductive age (18–44 years) from four field centers representing Hispanic/Latino backgrounds of Cuban, Dominican, Mexican, Puerto Rican, Central American, and South American. A baseline clinic examination (2008–2011) and a second clinic examination (2014–2017), including ascertainment of birth outcomes, allowed for identification of 517 singleton live births between the exams. Preconception cardiometabolic markers included abdominal obesity (waist circumference ≥88 cm), body mass index >30 kg/m², high blood pressure (systolic ≥120 mmHg and diastolic ≥80 mmHg), elevated triglycerides (≥150 mg/dL), low high-density lipoprotein cholesterol (<50 mg/dL), elevated fasting glucose (≥100 mg/dL), and insulin. Complex survey linear regression modeled the association between cardiometabolic markers and birthweight-for-gestational age z-score; complex survey logistic regression modeled the association with preterm birth. Analyses adjusted for Hispanic/Latina background, field center, years between baseline and birth, age, and nulliparity.

Results:

In adjusted linear regression models, elevated fasting glucose was associated with higher birthweight z-scores (β = 0.56, 95% confidence interval [95% CI] 0.14 to 0.99), even after further adjustment for maternal percent body fat (β = 0.53, 95% CI 0.10 to 0.95). In adjusted logistic regression models, high blood pressure (odds ratio [OR] = 2.57, 95% CI 1.13 to 5.88) and increased insulin (OR = 1.50, 95% CI 1.06 to 2.14, for a 10 mU/L increase) were associated with higher odds for preterm birth.

Conclusions:

Infant birthweight and preterm birth may be influenced by selected cardiometabolic risk factors before pregnancy among Hispanic/Latina women.

Introduction

Women's cardiometabolic health has been shown to impact child health outcomes.¹ Prior studies suggest that a mother's metabolic environment during the prenatal period can adversely influence the health of her offspring, especially at birth.^2

–5 Specific cardiometabolic markers that appear especially relevant in the preconception period include elevated cholesterol,^6
–8 high blood pressure,^9,10 glucose,⁸ insulin,⁹ and obesity,¹ and are associated with an increased risk of adverse birth outcomes. However, a systematic review showed no associations between elevated triglycerides, elevated total and low-density lipoprotein (LDL), or low high-density lipoprotein (HDL) cholesterol and spontaneous preterm birth.¹¹ These data are limited due to the lack of standardized documentation of preconception health, and are especially scarce for Hispanics from diverse backgrounds.

Hispanic women have a higher prevalence of selected components of the metabolic syndrome than non-Hispanic women.¹² Between 2007 and 2012, U.S. Mexican American women had a higher prevalence of elevated waist circumference (67% vs. 59%) and lower levels of HDL cholesterol (52% vs. 46%) compared with non-Hispanic white women.¹² In the Hispanic Community Health Study/Study of Latinos (HCHS/SOL), a population-based prospective cohort study of Hispanic/Latino individuals living in the United States, abdominal obesity was the most prevalent component of the metabolic syndrome among women (18–74 years) in 2008–2011, followed by low HDL cholesterol.¹³

While several studies have examined the influence of prenatal maternal cardiometabolic markers on birth outcomes, the opportunity to intervene during pregnancy is limited and may come too late. Prospective data on preconceptional health parameters are rare, and few studies have reported associations between preconception maternal cardiometabolic health and birth outcomes.^6

–10 Improved cardiometabolic health before pregnancy could potentially decrease the risk of adverse pregnancy complications, certain birth outcomes, and future health problems for children. The HCHS/SOL provides a unique opportunity to prospectively investigate the association between preconceptional maternal health status and subsequent birth outcomes, due to baseline data collection that occurred before pregnancy among a subgroup of Hispanic/Latina women, a population that now represents nearly a quarter of births in the United States.¹⁴

Our objective was to examine the association between cardiometabolic markers and birthweight as well as preterm birth using data from the HCHS/SOL. We hypothesized that higher preconception levels of selected measures of cardiometabolic health (triglycerides, fasting glucose, insulin, waist circumference) and lower preconception levels of HDL cholesterol would be associated with higher birthweight z-scores. We further hypothesized that higher preconception levels of fasting glucose and insulin, lower preconception levels of HDL cholesterol, and high blood pressure would be associated with preterm birth. We further hypothesized that these associations would be attenuated by maternal body fat since maternal body mass index (BMI) is shown to be a significant determinant of these outcomes.¹⁵

Materials and Methods

Study sample and procedures

This study used data from the HCHS/SOL, a population-based prospective cohort study of 16,415 self-identified Hispanic/Latino individuals of diverse backgrounds living in four U.S. cities (Bronx, NY; Chicago, IL; Miami, FL; San Diego, CA). The cohort was selected using a complex sampling design that has been described previously.¹⁶ Briefly, a stratified two-stage area probability sample of household addresses was selected in each of the four field centers, with unequal probabilities of selection at each stage. Participants completed a baseline clinical examination (March 2008–June 2011) that included assessments of anthropometrics, blood pressure, and other biological measures. They were also asked to complete interviewer-administered questionnaires on a variety of sociodemographic and behavioral factors, acculturation, and medical history. Participants provided fasting blood samples based on standardized protocols. Pregnant women were asked to reschedule the clinical examination after delivery. Questionnaires were administered in either English or Spanish, depending on the participant's preference.

Study participants were invited to a second clinical examination ∼6 years after baseline (October 2014–December 2017), which included many of the same assessments as the baseline exam, as well as questionnaires with information collected on reproductive history and all pregnancies that occurred since the baseline exam and lasted 6 months or longer. Informed consent was obtained at the baseline and second clinic visits, and all study protocols were approved by the Institutional Human Subjects Review Board at participating institutions.

There were 3,798 women of reproductive age (18–44 years) at baseline, of whom 2,559 participated in the second clinic visit. There were 679 pregnancies that resulted in live births (from 550 women) reported between baseline and the second clinic visit, of which 543 were singleton live births representing the first child born during this time frame. We excluded 26 women due to missing data on birth outcomes, resulting in a final sample of 517 women.

Measures

Preconception cardiometabolic markers

Maternal cardiometabolic markers were measured at the baseline clinic exam, before conception for the subsequent birth of interest. For this analysis, there were eight cardiometabolic markers of interest, including waist circumference, triglycerides, HDL cholesterol, systolic and diastolic blood pressure, glucose, insulin, and BMI. Blood samples were collected using a venous puncture after fasting for at least 8 hours before the clinic visit and analyzed to measure serum triglycerides (mg/dL), HDL cholesterol (mg/dL), and insulin (mU/L) as well as fasting plasma glucose (mg/dL). HDL cholesterol was measured using a direct magnesium/dextran sulfate method, and fasting plasma glucose was measured using a hexokinase enzymatic method (Roche Diagnostics, Indianapolis, IN). Serum triglycerides were measured on a Roche Modular P chemistry analyzer using a glycerol blanking enzymatic method (Roche Diagnostics).

Insulin was measured using the Mercodia Insulin enzyme-linked immunosorbent assay, a two-site enzyme immunoassay utilizing the direct sandwich technique (Mercodia AB, Uppsala, Sweden). The assay methodologies are described in HCHS/SOL Manual 7a (https://www2.cscc.unc.edu/hchs/system/files/protocols-manuals/UNLICOMMManual07AddendumCentralLaboratoryProceduresv1006222011.pdf).

Sitting blood pressure measurements were taken three consecutive times after a 5-minute rest period, and the average of the three measurements was calculated for both systolic and diastolic blood pressure (mmHg). Anthropometric measurements were taken, including waist circumference (cm), standing height (cm), weight (kg), and percent body fat. Weight and percent body fat were measured using the Tanita scale in the bioelectric impedance mode. BMI was calculated as weight in kilograms divided by height in meters squared. Obesity was defined as BMI ≥30 kg/m².¹⁷

Dichotomized versions of all cardiometabolic markers (except insulin) were also considered. Selected cardiometabolic markers that comprise the metabolic syndrome were classified according to the American Heart Association/National Heart, Lung, and Blood Institute 2009 Joint Scientific Statement¹⁸ for abdominal obesity (waist circumference ≥88 cm), elevated triglycerides (≥150 mg/dL), low HDL cholesterol (<50 mg/dL), and elevated fasting glucose (≥100 mg/dL); high systolic (≥120 mmHg) and diastolic blood pressure (≥80 mmHg) were classified based on the 2017 Hypertension Clinical Guidelines.¹⁹

Birth outcomes

Information about the index birth was reported by the mother at the second clinic visit, including gestational age at delivery (weeks), birthweight (g), infant sex, and pregnancy complications (gestational diabetes, gestational hypertension, preeclampsia). Birthweight z-scores were calculated based on gestational age at delivery and infant sex using an international reference that included Hispanic women^20,21; these calculations were done using the INTERGROWTH-21st web application.²² Specifically, a birthweight z-score of 0 corresponded to an average birthweight for the given gestational age at delivery and infant sex in the international reference population, and a birthweight z-score of 1 corresponded to a birthweight approximately one standard deviation above the average birthweight based on the given gestational age and infant sex. Gestational age at delivery was categorized as either preterm (<37 weeks) or not. Since this measure is based on self-report, preterm birth was considered a composite outcome and was not distinguished between spontaneous and indicated preterm birth.

Other measures

All models adjusted for baseline age, Hispanic/Latina background (Cuban, Dominican, Mexican, Puerto Rican, Central American, South American, and Mixed), field center, nulliparity (first child or not), and years between the baseline visit (i.e., timing of preconception cardiometabolic marker measurement) and birth. Additionally, the following baseline covariates were assessed for confounding, and included in the models if determined to be a confounder (see Statistical Analysis section): household size, education (no high school diploma/General Educational Development [GED], at most a high school diploma/GED, or greater than high school/GED), annual household income (< $30,000, ≥ $30,000, or not reported), employment status (employed or not), years in the mainland U.S. (<10 years, ≥10 years, or born in the mainland U.S.), cigarette use (never smoker, former smoker, or current smoker), and meeting the 2008 physical activity-level guidelines (yes or no).²³ Prepregnancy diabetes and hypertension were self-reported at baseline.

Statistical analyses

We used multiple imputation to impute missing values in preconception cardiometabolic markers or covariates for 29 women from the 517 who had outcome data. Specifically, we used fully conditional specification to impute five values for continuous preconception cardiometabolic markers and potential confounders with missing data (household size, education, employment status, years in the mainland U.S., cigarette use, physical activity, percent body fat). In the imputation model we included age, Hispanic/Latina background, nulliparity, years between baseline and birth, income, and field center and sampling weight to account for the complex survey design.

Descriptive statistics (means, percents, medians, interquartile ranges [IQRs]) were calculated for all variables. Complex survey linear regression was used to separately estimate the association of each baseline maternal cardiometabolic marker with birthweight z-score. To assess whether to use the continuous or dichotomized cardiometabolic markers, the adjusted R ² was compared between a model with the continuous marker and a model with the dichotomized marker, and the form for each marker that produced the higher adjusted R ² was used in all final models.

Since there are no meaningful cut points for dichotomizing insulin in this population it was assessed as a continuous measure. The adjusted R ² for the continuous cardiometabolic markers ranged from 0.0068 (triglycerides) to 0.0141 (systolic blood pressure); the adjusted R ² for the dichotomous markers ranged from 0.0074 (triglycerides) to 0.0182 (glucose) (Supplementary Table S1). Based on these estimates, we chose dichotomized over continuous exposure variables for all cardiometabolic markers except for insulin. Similarly, the functional form for age (linear term, quadratic term, or categorized [18–24 years, 25–34 years, or 35–44 years]) in all models was selected by comparing the adjusted R ² between models with each functional form for age, and selecting the form that produced the highest adjusted R ² for most models. Based on the adjusted R ², we chose categorical age for all models.

For each cardiometabolic marker, we first adjusted for Hispanic/Latina background, field center, and years between baseline and the birth. We then further adjusted for age and nulliparity. To assess whether the association between each cardiometabolic marker and birthweight z-score differed by nulliparity, an interaction term was included in the model between the marker and nulliparity, and the interaction term was tested at a 0.05 significance level. Then, confounding was assessed by fitting the previous model for each cardiometabolic marker with and without each covariate in the model (i.e., household size, education, income, employment status, years in the United States, cigarette use, physical activity), and calculating the percent change in the regression coefficient for the marker when the covariate was added to the model.

All the covariates identified with >10% change in the regression coefficient for at least four markers (out of eight) were included to avoid having different adjusting covariates for each outcome. Lastly, we further adjusted for percent body fat, a more proximal determinant of disease risk than BMI.²⁴ Sensitivity analyses were conducted adjusting for all covariates identified with >10% change in the regression coefficient for at least two markers.

Complex survey logistic regression was used to separately estimate the association of each baseline cardiometabolic marker with preterm birth. The same functional form for each cardiometabolic marker, functional form for age, and set of confounders were used as for the models for birthweight z-score. Additional sensitivity analyses were conducted repeating the final model excluding women with selected prepregnancy conditions (prepregnancy diabetes and prepregnancy hypertension) as well as those who experienced selected pregnancy complications (gestational diabetes, gestational hypertension, and preeclampsia).

All analyses accounted for the complex survey design, including sampling weights, clustering, and stratification, and were performed using the complex survey procedures in SAS 9.4 or SAS-callable SUDAAN v10. The sampling weights accounted for the oversampling of population subgroups (e.g., over 45 years of age), for household- and person-level nonresponse at baseline and for person-level non-response at the second study visit. Weights were then trimmed to reduce the variability of the weights and the impact of extremely large weights values, and lastly calibrated to the age, sex, and Hispanic/Latino background distributions from the 2010 U.S. Census for the four study field centers. By using visit 2 sampling weights in the analyses the inference is to the original study population (all noninstitutionalized Hispanic/Latino adults 18–74 years of age and residing in the defined geographical areas (census block groups) across the four participating field centers).

Results

Study population

There were 517 women participating in HCHS/SOL included for this analysis. These women gave birth to their infant at a median of 3.0 years (IQR 1.7, 4.6) after the baseline examination. Half the women were 25–34 years old, 25.2% had less than high school education, 61.9% had a family income < $30,000, 30.6% were born in the United States, 79.9% never used cigarettes, and 33.3% were obese (Table 1). Half of these women were of Mexican background. Approximately 4% had prepregnancy diabetes, 2.3% had prepregnancy hypertension, 14.7% had gestational diabetes, 9.5% had gestational hypertension, and less than 5% had preeclampsia. The median (IQR) birthweight and gestational age of infants was 3349 g (2984.8–3676.4) and 39.0 weeks (37.7–39.6), respectively. Further detail on the distribution of gestational age at delivery and birthweight z-score are provided in Supplementary Figure S1.

Table 1.

Characteristics of Hispanic Community Health Study/Study of Latinos Hispanic/Latina Women Who Delivered a Live-Born Infant Following the Baseline Examination (N = 517), 2008–2017

	Sample size	Statistic^a
Maternal characteristics
Household size, mean (SE)	517	3.7 (0.11)
Age, years, % (SE)
18–24	218	44.1 (2.6)
25–34	244	50.4 (2.8)
35–44	55	5.5 (1.1)
Field center
Bronx	137	33.3 (3.2)
Chicago	154	17.8 (1.8)
Miami	87	20.2 (2.7)
San Diego	139	28.9 (3.1)
Hispanic/Latina background
Cuban	42	13.5 (2.2)
Dominican	62	14.7 (2.3)
Mexican	259	48.2 (3.1)
Puerto Rican	58	10.5 (1.5)
Central American	56	7.4 (1.3)
South American	23	2.8 (0.8)
Mixed	17	2.9 (0.8)
Education
Less than high school	139	25.2 (2.4)
High school graduate	174	31.4 (2.6)
Greater than high school	203	43.4 (2.8)
Annual household income
< $30,000	326	61.9 (2.8)
> $30,000	147	27.9 (2.6)
Not reported	44	10.2 (1.8)
Employed	259	47.1 (2.8)
U.S. residence, years
<10	187	37.5 (3.0)
≥10	162	32.0 (2.9)
Born in U.S.	164	30.6 (2.6)
Cigarette use
Never	405	79.9 (2.1)
Former	42	6.8 (1.2)
Current	68	13.3 (1.9)
Meeting the 2008 PA guidelines (self-report)
Yes	337	68.2 (2.6)
Body mass index, kg/m² ^b
Under or normal weight	186	37.1 (2.7)
Overweight	158	29.6 (2.4)
Obese	171	33.3 (2.6)
Percent body fat^c	517	34.3 (0.5)
Prepregnancy diabetes (yes)	21	3.7 (1.0)
Prepregnancy hypertension (yes)	15	2.3 (0.8)
Pregnancy characteristics
Gestational diabetes (yes)	81	14.7 (1.9)
Gestational hypertension (yes)	55	9.5 (1.6)
Preeclampsia (yes)	26	4.3 (0.9)
Nulliparity (yes)	195	39.0 (2.8)
Birth outcomes, median (IQR) or %
Birthweight, g	517	3349.3 (2984.8, 3676.4)
Birthweight z-score	517	0.5 (−0.3, 1.3)
Preterm birth (<37 weeks; yes)	58	10.6
Gestational age at delivery, weeks	517	39.0 (37.7, 39.6)
Years between baseline and birth	517	3.0 (1.7, 4.6)

Mean (SE), % (SE) or median (first quartile, third quartile), as noted.

Body mass index categorized as under or normal weight (≤24.9 kg/m²), overweight (25.0–29.9 kg/m²), and obese (≥30 kg/m²).

Measured using a bioimpedance scale.

IQR, interquartile range; PA, physical activity; SE, standard error.

In this study population, as shown in Table 2, abdominal obesity was the most prevalent preconception cardiometabolic risk factor (57.5%), followed by low HDL cholesterol (53.7%), obesity (33.3%), elevated triglycerides (12.4%), and elevated fasting glucose (10.1%). Nearly 12% of women had high blood pressure.

Table 2.

Preconception Cardiometabolic Markers Among Hispanic Community Health Study/Study of Latinos Hispanic/Latina Women Who Delivered a Live-Born Infant Following the Baseline Examination (N = 517)

Cardiometabolic markers	Median (first quartile, third quartile)
Waist circumference, cm	90.4 (81.2, 100.8)
Triglycerides, mg/dL	80.9 (55.2, 111.7)
HDL cholesterol, mg/dL	47.8 (40.5, 57.6)
Systolic blood pressure, mmHg	104.2 (97.2, 111.2)
Diastolic blood pressure, mmHg	64.8 (58.8, 71.4)
Fasting glucose, mg/dL	87.8 (84.2, 92.9)
Insulin, mU/L	10.7 (6.9, 16.6)
Body mass index, kg/m²	26.8 (23.4, 32.1)

Dichotomized cardiometabolic markers^a	% (SE)
Abdominal obesity	57.5 (2.8)
Elevated triglycerides	12.4 (2.0)
Low HDL cholesterol	53.7 (3.0)
Elevated fasting glucose	10.1 (1.6)
High blood pressure	11.9 (1.8)
Obesity	33.3 (2.6)

Components of the metabolic syndrome were classified according to the American Heart Association/National Heart, Lung, and Blood Institute 2009 Joint Scientific Statement¹⁸ for abdominal obesity (waist circumference ≥88 cm), elevated triglycerides (≥150 mg/dL), low HDL cholesterol (<50 mg/dL), and elevated fasting glucose (≥100 mg/dL); high systolic (≥120 mmHg) and diastolic blood pressure (≥80 mmHg) were classified based on the 2017 Hypertension Clinical Guidelines.¹⁹ Obesity was defined as body mass index ≥30 kg/m².¹⁷

HDL, high-density lipoprotein.

Preconception cardiometabolic markers and birth outcomes

There were no statistically significant interactions observed between any of the cardiometabolic markers and nulliparity, and thus the models were not stratified. Additionally, there were no covariates identified with >10% change in the regression coefficient for at least four markers in the models for birthweight z-score or preterm birth.

In multivariable linear regression models (Table 3), preconception elevated fasting glucose (≥100 mg/dL) was associated with birthweight z-scores that were 0.56 higher (95% confidence interval [95% CI] 0.14 to 0.99) compared with fasting glucose <100 mg/dL, after adjusting for Hispanic/Latina background, field center, years between baseline visit and birth, age, and nulliparity. Further adjustment for percent body fat did not substantially change the association between elevated fasting glucose and birthweight z-scores (β = 0.53, 95% CI 0.10 to 0.95). No other associations were observed between any other preconception cardiometabolic markers and infant birthweight.

Table 3.

Linear Regression Coefficients (95% Confidence Interval) for the Association Between Preconception Cardiometabolic Markers and Birthweight z-Scores, Hispanic Community Health Study/Study of Latinos (N = 517)

Cardiometabolic markers^a	Model 1, ^b β (95% CI)	Model 2, ^c β (95% CI)	Model 3, ^d β (95% CI)
Abdominal obesity, Y/N	0.25 (−0.03 to 0.53)	0.23 (−0.05 to 0.52)	0.13 (−0.36 to 0.63)
Elevated triglycerides, Y/N	0.22 (−0.25 to 0.69)	0.18 (−0.28 to 0.64)	0.12 (−0.37 to 0.60)
Low HDL cholesterol, Y/N	0.04 (−0.24 to 0.31)	0.03 (−0.25 to 0.30)	−0.02 (−0.30 to 0.26)
High systolic or diastolic blood pressure, Y/N	−0.25 (−0.75 to 0.26)	−0.32 (−0.82 to 0.18)	−0.42 (−0.93 to 0.08)
Elevated fasting glucose, Y/N	0.55 (0.11 to 0.99)	0.56 (0.14 to 0.99)	0.53 (0.10 to 0.95)
Insulin (10-unit increment), mU/L	0.04 (−0.10 to 0.18)	0.04 (−0.10 to 0.19)	−0.04 (−0.23 to 0.15)
Obesity, Y/N	0.29 (−0.01 to 0.59)	0.25 (−0.04 to 0.55)	—

Bold font indicates statistical significance (p < 0.05).

Adjusted for Hispanic/Latina background, field center, and years between baseline visit and birth.

Adjusted for model 1 covariates plus age (categorical), and nulliparity.

Adjusted for model 2 covariates plus percent body fat.

95% CI, 95% confidence interval.

High blood pressure (systolic blood pressure ≥120 mmHg or diastolic blood pressure ≥80 mmHg) was associated with increased odds of preterm birth compared with normal blood pressure based on logistic regression models adjusted for Hispanic/Latina background, field center, years between baseline visit and birth, age, and nulliparity (odds ratio [OR], 2.57; 95% CI 1.13 to 5.88, Table 4). Further adjustment for percent body fat did not substantially change the association between high blood pressure and preterm birth (OR = 2.54, 95% CI 1.07 to 6.00). In addition, increased insulin was associated with increased odds of preterm birth in models adjusted for Hispanic/Latina background, field center, years between baseline visit and birth, age, and nulliparity (OR for a 10 mU/L increase in insulin, 1.50, 95% CI 1.06 to 2.14, Table 4). Further adjustment for percent body fat slightly strengthened the association between increased insulin and preterm birth (OR = 1.68, 95% CI 1.09 to 2.58). No significant associations were observed between preterm birth and other cardiometabolic markers assessed.

Table 4.

Odds Ratios (95% Confidence Interval) for the Association Between Preconception Cardiometabolic Markers and Preterm Birth, Hispanic Community Health Study/Study of Latinos (N = 517)

Cardiometabolic markers^a	Model 1, ^b OR (95% CI)	Model 2, ^c OR (95% CI)	Model 3, ^d OR (95% CI)
Abdominal obesity, Y/N	1.21 (0.60 to 2.44)	1.18 (0.58 to 2.43)	1.01 (0.33 to 3.10)
Elevated triglycerides, Y/N	0.94 (0.35 to 2.54)	0.89 (0.32 to 2.48)	0.84 (0.31 to 2.31)
Low HDL cholesterol, Y/N	0.73 (0.37 to 1.45)	0.74 (0.37 to 1.45)	0.70 (0.35 to 1.37)
High systolic or diastolic blood pressure, Y/N	2.79 (1.27 to 6.16)	2.57 (1.13 to 5.88)	2.54 (1.07 to 6.00)
Elevated fasting glucose, Y/N	1.80 (0.69 to 4.73)	1.53 (0.56 to 4.21)	1.47 (0.54 to 4.05)
Insulin (10-unit increment), mU/L	1.52 (1.08 to 2.14)	1.50 (1.06 to 2.14)	1.68 (1.09 to 2.58)
Obesity, Y/N	1.47 (0.74 to 2.95)	1.47 (0.73 to 2.96)	—

Bold font indicates statistical significance (p < 0.05),

Adjusted for Hispanic/Latina background, field center, and years between baseline visit and birth.

Adjusted for model 1 covariates plus age (categorical), and nulliparity.

Adjusted for model 2 covariates plus percent body fat.

OR, odds ratio.

We conducted sensitivity analyses, including all covariates with >10% change in the regression coefficient for at least two markers in the models for birthweight z-score or preterm birth. For both outcomes, the results were similar to the primary results (data not shown). Additional sensitivity analyses excluded women with prepregnancy diabetes or hypertension as well as those who experienced selected pregnancy complications. For birthweight z-scores, the association with fasting glucose was modestly attenuated after excluding those with gestational diabetes (β = 0.50, 95% CI 0.10 to 0.95). In contrast, the association between high blood pressure and lower birthweight was strengthened and became statistically significant after excluding women with prepregnancy diabetes (β = −0.53, 95% CI −1.01 to −0.04), gestational diabetes (β = −0.60, 95% CI −1.15 to −0.05), or gestational hypertension (β = −0.64, 95% CI −1.18 to −0.10). For preterm birth, the associations with high blood pressure and insulin did not persist after excluding women with prepregnancy conditions or pregnancy complications (data not shown).

Discussion

This study documents associations between preconception elevated fasting glucose and higher birthweight z-scores. High blood pressure and increased insulin were associated with preterm birth and this association persisted after adjustment for confounders. Nulliparity did not modify these associations, and adjusting for percent body fat, as measured by bioelectrical impedance analysis, did not eliminate these associations. After exclusion of women with selected prepregnancy conditions and pregnancy complications, associations between elevated fasting glucose and birthweight were attenuated while the association between high blood pressure and lower birthweight was strengthened. The associations between high blood pressure or insulin and preterm birth no longer remained.

Elevated fasting glucose during the preconception period signifies prediabetes and has been shown to be associated with higher birthweight in a Scandinavian population.¹⁰ In contrast, one study assessing racial disparities in the association between cardiovascular risk factors and birth outcomes, elevated fasting glucose was not associated with large for gestational age among white or black women; Hispanics were not included in the sample.⁷ Hispanics/Latinos in the United States are more likely to have prediabetes and type 2 diabetes than non-Hispanic whites, and this risk varies considerably by country of origin.²⁵ Our data support this observation with 10% of the reproductive-aged women having had elevated glucose, 33% being classified as obese, and more than half (58%) had abdominal obesity before pregnancy, known risk factors for diabetes. These findings together with the multivariable results for elevated fasting glucose, provide evidence for the importance of addressing lifestyle behaviors in the preconception period to help break the cycle of diabetes risk among Hispanics/Latinos living in the United States.

Previous studies conducted in non-Hispanic populations have observed associations between elevated blood pressure and lower birthweight.^9,10 Although these associations were not statistically significant in our primary analysis, the association between high blood pressure and lower birthweight was strengthened and statistically significant after excluding women with prepregnancy diabetes, gestational diabetes, or gestational hypertension. While the mechanism for this association is unclear, given the importance of blood flow regulation from the mother to the fetus through the placenta, an organ responsible for critical nutrient transport, the association with lower birthweight may suggest placental dysfunction.²⁶ Further research is needed to better understand these associations.

In addition, our findings that high blood pressure and increased insulin are associated with an increased risk of preterm birth are supported by prior studies.^8,9 In sensitivity analyses that excluded women who developed metabolic complications of pregnancy, these associations did not persist. These findings suggest that selected pregnancy complications may mediate the associations that were observed in the primary analysis.

Our finding that elevated triglycerides are not associated with preterm birth is supported by some,⁹ but not all studies.⁸ Furthermore, the lack of an association observed between elevated triglycerides and birthweight has also been previously found.⁹ While prior studies have noted associations among other lipids (e.g., total cholesterol, HDL cholesterol, or LDL cholesterol) and both gestational age^3,8,9 and birthweight,¹⁰ we did not observe an association between low HDL cholesterol and birthweight in our population of Hispanic/Latina women. In general, there is limited evidence on the association between cholesterol and birthweight from studies, including women from various racial/ethnic backgrounds suggesting that cholesterol is regulated by hormonal changes and less influenced by lifestyle behaviors.

In contrast to previous studies,^15,27 preconception maternal obesity (BMI ≥30 kg/m²) measured on average 3 years before pregnancy, was not associated with higher birthweight z-score in the unadjusted or adjusted models. Women in this sample were relatively young, with a high percentage of women having obesity (33%) and a similar percentage having overweight. Given this prevalence of higher BMIs, a cut point of 30 kg/m² may not be indicative of higher birthweight.

Strengths and limitations

The HCHS/SOL is a unique cohort study that allowed for the assessment of cardiometabolic markers clinically measured prepregnancy among a diverse population of Hispanic/Latinas in the United States.¹⁶ The data collection methods with fasting serum samples measured at baseline minimize the potential for measurement error. This study also included detailed information on pregnancy-related complications (e.g., gestational diabetes, preeclampsia) and prepregnancy conditions (e.g., diabetes), thus enabling these measures to be accounted for in sensitivity analyses.

This study also has limitations, including reliance on maternal self-report of perinatal outcomes and pregnancy complications, the time lapse between preconception measures and birth, sample size, and generalizability. Information on birthweight, gestational age at delivery, and pregnancy complications was based on maternal self-report at the second clinic visit and is subject to measurement error. However, previous studies suggest that the reliability of self-reported information on birth outcomes is very good (kappa ranges of 0.6 to 0.9) for birthweight, gestational age, previous live births, pregnancies, and miscarriages.²⁸ Other studies have reported moderate validity of maternal self-reported preeclampsia^29,30 and gestational diabetes,³⁰ and good concordance between self-reported and objectively measured hypertension and/or proteinuria.³¹ There may also be unmeasured risk factors associated with both preconception cardiometabolic markers and birth outcomes. Residual confounding by these factors, which is apparent in any observational cohort study, may confound the observed associations.

The median length of time between the baseline measurements for preconception cardiometabolic risk factors and birth was ∼3 years, which may have implications for maternal health status at the time of birth for those with a longer time lapse between measurements. Further examination of characteristics listed in Table 1 comparing women who had a pregnancy >3 years from baseline versus those with a more recent pregnancy yielded no significant differences (results available upon request). It also limited our ability to assess differences in the associations by Hispanic/Latina background. In addition, the generalizability of study results is limited to the four urban communities sampled, including the Bronx, Miami, Chicago, and San Diego. However, these areas have some of the largest populations of Hispanics/Latinos in the United States,³² and the HCHS/SOL study design is more rigorous than the simple convenience samples typical of most epidemiological cohort studies with its use of probability sampling within these preselected regions.

Conclusions

Our study adds to the literature demonstrating the importance of the preconception period for selected factors associated with the metabolic syndrome (i.e., elevated fasting glucose, high blood pressure, increased insulin) on infant birthweight and preterm birth for Hispanic/Latina women. In addition, we reported in a prior study that better diet quality during the preconception period was associated with higher birthweight adjusted for gestational age, among this population of women.³³ Thus, the preconception period represents a highly sensitive phase in the life course of reproductive-aged women, which would benefit from more targeted messaging concerning lifestyle behaviors.³⁴ More personalized medical advice for individual- and community-level interventions has the potential to encourage women to optimize dietary intake and lifestyle behaviors to improve birth outcomes and reduce maternal cardiometabolic risks.

Footnotes

Acknowledgments

The authors thank the staff and participants of HCHS/SOL for their important contributions.

Disclaimer

The views expressed in this publication are solely the opinions of the authors and do not necessarily reflect the official policies of the U.S. Department of Health and Human Services, the Health Resources and Services Administration, the National Institute on Minority Health and Health Disparities, or the National Institutes of Health nor does mention of the department or agency names that imply endorsement by the U.S. Government.

Data Sharing

Data collection instruments described in the article are publicly available at: https://sites.cscc.unc.edu/hchs/. Analytic code will be made available on request.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

The HCHS/SOL (www.cscc.unc.edu/hchs/) was carried out as a collaborative study supported by contracts from the National Heart, Lung, and Blood Institute (NHLBI) to the University of North Carolina (N01-HC65233), University of Miami (N01-HC65234), Albert Einstein College of Medicine (N01-HC65235), Northwestern University (N01-HC65236), and San Diego State University (N01-HC65237). The following Institutes/Centers/Offices contribute to the HCHS/SOL through a transfer of funds to the NHLBI: National Institute on Minority Health and Health Disparities, National Institute on Deafness and Other Communication Disorders, National Institute of Dental and Craniofacial Research, National Institute of Diabetes and Digestive and Kidney Diseases, National Institute of Neurological Disorders and Stroke, NIH Institution-Office of Dietary Supplements.

Additional support was provided by the Life Course Methodology Core (LCMC) at the New York Regional Center for Diabetes Translation Research (DK111022-8786 and DK111022) through funds from the National Institute of Diabetes and Digestive and Kidney Diseases.

Supplementary Material

Supplementary Figure S1

Supplementary Table S1

References

Oken

. Maternal and child obesity: The causal link. Obstet Gynecol Clin North Am, 2009; 36:361–377, ix–x.

Metzger

, Lowe

, Dyer

, et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med, 2008; 358:1991–2002.

Catov

, Bodnar

, Kip

, et al. Early pregnancy lipid concentrations and spontaneous preterm birth. Am J Obstet Gynecol, 2007; 197:610.e611–e617.

Catov

, Bodnar

, Ness

, Barron

, Roberts

. Inflammation and dyslipidemia related to risk of spontaneous preterm birth. Am J Epidemiol, 2007; 166:1312–1319.

Grace

, Vladutiu

, Nethery

, et al. Lipoprotein particle concentration measured by nuclear magnetic resonance spectroscopy is associated with gestational age at delivery: A prospective cohort study. Br J Obstet Gynecol, 2018; 125:895–903.

Catov

, Ness

, Wellons

, Jacobs

, Roberts

, Gunderson

. Prepregnancy lipids related to preterm birth risk: The coronary artery risk development in young adults study. J Clin Endocrinol Metab, 2010; 95:3711–3718.

Harville

, Myers

, Shu

, Wallace

, Bazzano

. Pre-pregnancy cardiovascular risk factors and racial disparities in birth outcomes: The Bogalusa Heart Study. BMC Pregnancy Childbirth, 2018; 18:339.

Magnussen

, Vatten

, Myklestad

, Salvesen

, Romundstad

. Cardiovascular risk factors prior to conception and the length of pregnancy: Population-based cohort study. Am J Obstet Gynecol, 2011; 204:526.e521–e528.

Harville

, Viikari

, Raitakari

. Preconception cardiovascular risk factors and pregnancy outcome. Epidemiology, 2011; 22:724–730.

10.

Romundstad

, Davey Smith

, Nilsen

, Vatten

. Associations of prepregnancy cardiovascular risk factors with the offspring's birth weight. Am J Epidemiol, 2007; 166:1359–1364.

11.

Moayeri

, Heida

, Franx

, Spiering

, de Laat

, Oudijk

. Maternal lipid profile and the relation with spontaneous preterm delivery: A systematic review. Arch Gynecol Obstet, 2017; 295:313–323.

12.

Moore

, Chaudhary

, Akinyemiju

. Metabolic syndrome prevalence by race/ethnicity and sex in the United States, National Health and Nutrition Examination Survey, 1988–2012. Prev Chron Dis, 2017; 14:E24.

13.

Vladutiu

, Siega-Riz

, Sotres-Alvarez

, et al. Parity and components of the metabolic syndrome among US Hispanic/Latina Women: Results from the Hispanic Community Health Study/Study of Latinos. Circ Cardiovasc Qual Outcomes, 2016; 9(Suppl 1):S62–S69.

14.

Osterman

MJK

, Hamilton

, Martin

, Driscoll

, Valenzuela

. Births: Final Data for 2020. Nat Vital Stat Rep, 2022; 70:1–49.

15.

Poston

, Caleyachetty

, Cnattingius

, et al. Preconceptional and maternal obesity: Epidemiology and health consequences. Lancet Diabetes Endocrinol, 2016; 4:1025–1036.

16.

Lavange

, Kalsbeek

, Sorlie

, et al. Sample design and cohort selection in the Hispanic Community Health Study/Study of Latinos. Ann Epidemiol, 2010; 20:642–649.

17.

Clinical Guidelines on the

Identification

, Evaluation, and Treatment of Overweight and Obesity in Adults—The Evidence

Report

. National Institutes of Health. Obes Res, 1998; 6(Suppl 2):51s–209s.

18.

Alberti

, Eckel

, Grundy

, et al. Harmonizing the metabolic syndrome: A joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation, 2009; 120:1640–1645.

19.

Whelton

, Carey

, Aronow

, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation, 2018; 138:e426–e483.

20.

Villar

, Cheikh Ismail

, Victora

, et al. International standards for newborn weight, length, and head circumference by gestational age and sex: The Newborn Cross-Sectional Study of the INTERGROWTH-21st Project. Lancet, 2014; 384:857–868.

21.

Villar

, Giuliani

, Fenton

, Ohuma

, Ismail

, Kennedy

. INTERGROWTH-21st very preterm size at birth reference charts. Lancet, 2016; 387:844–845.

22.

INTERGROWTH-21st Applications and Calculators. Available at: https://intergrowth21.tghn.org/global-perinatal-package/intergrowth-21st-very-preterm-size-application-tool/ Accessed September 1, 2021 .

23.

U.S. Department of Health & Human Services. 2008 Physical Activity Guidelines for Americans. Published 2008. Available at: www.health.gov/PAGUidelines/guidelines/default.aspx Accessed September 1, 2021 .

24.

Byambasukh

, Eisenga

, Gansevoort

, Bakker

, Corpeleijn

. Body fat estimates from bioelectrical impedance equations in cardiovascular risk assessment: The PREVEND cohort study. Eur J Prev Cardiol, 2019; 26:905–916.

25.

Centers for Diease Control and Prevention. Hispanic/Latino Americans and Type 2 Diabetes. Available at: https://www.cdc.gov/diabetes/library/features/hispanic-diabetes.html Accessed September 1, 2021 .

26.

Lim

, Lee

, Tan

, et al. The association between maternal blood pressures and offspring size at birth in Southeast Asian women. BMC Pregnancy Childbirth, 2014; 14:403.

27.

El-Chaar

, Finkelstein

, Tu

, et al. The impact of increasing obesity class on obstetrical outcomes. J Obstet Gynaecol Can, 2013; 35:224–233.

28.

Olson

, Shu

, Ross

, Pendergrass

, Robison

. Medical record validation of maternally reported birth characteristics and pregnancy-related events: A report from the Children's Cancer Group. Am J Epidemiol, 1997; 145:58–67.

29.

Coolman

, de Groot

, Jaddoe

, Hofman

, Raat

, Steegers

. Medical record validation of maternally reported history of preeclampsia. J Clin Epidemiol, 2010; 63:932–937.

30.

Beekers

, Jamaladin

, van Drongelen

, Roeleveld

, van Gelder

. Data from web-based questionnaires were valid for gestational diabetes and preeclampsia, but not gestational hypertension. J Clin Epidemiol, 2020; 125:84–90.

31.

Falkegård

, Schirmer

, Løchen

, Øian

, Acharya

. The validity of self-reported information about hypertensive disorders of pregnancy in a population-based survey: The Tromsø Study. Acta Obstet Gynecol Scand, 2015; 94:28–34.

32.

U.S. Census Bureau. The Hispanic Population: 2010. 2010 Census Briefs Web site. Published 2011. Available at: https://www.census.gov/prod/cen2010/briefs/c2010br-04.pdf Accessed September 1, 2021 .

33.

Siega-Riz

, Vladutiu

, Butera

, et al. Preconception adherence to the Healthy Eating Index is associated with birthweight for gestational age among Hispanic/Latina women in the Hispanic Community Health Study/Study of Latinos. J Acad Nutr Diet, 2021; 121:458–466.

34.

Stephenson

, Heslehurst

, Hall

, et al. Before the beginning: Nutrition and lifestyle in the preconception period and its importance for future health. Lancet, 2018; 391:1830–1841.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB

0.01 MB