Maternal Mediterranean Diet During Lactation and Infant Growth

Abstract

Background:

Human milk is considered the optimal source of nutrition for infants. Maternal diet is associated with the composition of human milk. The Mediterranean diet (MedDiet) has been studied in pregnancy and during lactation, and it has been associated with changes in milk composition, yet there is a lack of research on MedDiet during lactation and infant outcomes.

Methods:

Mother–infant dyads (n = 167) from ABC Baby, a prospective observational study, were included in this analysis. Maternal diet was obtained using an adapted version of the National Cancer Institute Diet History Questionnaire II, at 2 weeks or 2 months postpartum. Maternal MedDiet score was calculated using servings of vegetables, fruits, whole grains, nuts and seeds, legumes, fish, monounsaturated-to-saturated fatty acid ratio, red and processed meats, and added sugar. Infants’ length, weight, and flank skinfold thickness were measured at 6 months. Using World Health Organization standards, weight-for-age (WAZ), length-for-age (LAZ), and weight-for-length (WLZ) Z-scores were calculated. Multiple linear regression models were adjusted for potential confounders.

Results:

Higher maternal MedDiet score and intake of fruit and fish were associated with lower flank skinfold thickness (β = −0.33, −0.52, and −1.26, respectively). Intake of nuts and seeds was associated with higher WLZ (β = 0.29). Intake of red and processed meats was associated with lower WAZ (β = −0.18) and LAZ (β = −0.18). Energy-adjusted added sugar intake was associated with lower WLZ (β = −0.02).

Conclusions:

The maternal MedDiet score was associated with lower skinfold thickness, while its components were associated with differences in anthropometric Z-scores. Further research on the maternal MedDiet and corresponding human milk composition is needed to explore this relationship.

Introduction

Human milk is recommended as the primary source of infant nutrition, as it provides the ideal nutrient composition and nonnutritive factors that promote overall infant health.^1,2 It is recommended that infants are exclusively fed human milk for the first 6 months and continue human milk feeding along with complementary foods until 2 years.³

Human milk composition is complex and dynamic. The composition is influenced by a variety of factors including maternal genetics, health, diet, and infant sex.¹ Nutrient components in human milk are derived from three sources: synthesis in the lactocyte, maternal diet, and maternal stores in adipose tissue, bone, and teeth.¹ Two modifiable factors are maternal diet and maternal stores, with maternal diet more readily available for changes to impact human milk composition.⁴ Due to the impact of diet on human milk, it was mandated in 2014 that the Dietary Guidelines for Americans (DGA) must provide guidance for pregnant and lactating individuals and children from birth to 24 months.⁵ However, at present, there is a lack of knowledge on human milk composition and associated health outcomes in offspring to define recommended dietary intake guidelines during lactation.

Various diets related to optimal health outcomes for adults and children have been explored over the years. The Mediterranean diet (MedDiet) was highlighted initially in 1960 and drew attention to the health benefits of the diets found in Italy and Greece.⁶ Since then, the MedDiet has become a popular research topic given its associations with better adult cardiovascular health, lower incidence of and mortality from cancer, and greater cognitive health including lower incidence of Parkinson’s and Alzheimer’s disease.^7,8 In comparison with a Western-style diet, a MedDiet promotes the consumption of foods rich in monounsaturated fats, fiber, phytonutrients, whole grains, and omega-3 fatty acids while limiting intakes of omega-6 fatty acids, saturated fatty acids, and sugar.⁹ Greater maternal MedDiet score during pregnancy has been associated with less gestational weight gain, lower odds of gestational diabetes or fetal growth restriction, and healthier weight-for-height percentiles for children up to age 8 years.^10–12 Maternal MedDiet score during lactation has been examined in relation to human milk composition in a few studies.^13–15 Higher MedDiet score has been associated with higher antioxidant content, lower saturated fat, and higher monounsaturated fat content in human milk.^13,14 It has also been shown that mothers consuming Western diets have half the levels of milk omega-3 compared with non-Western diets.¹⁵ Maternal MedDiet during lactation has been associated with higher antioxidant content in infant urine¹⁴; however, further associations with health outcomes and growth are unknown.

While the association of MedDiet with health outcomes among adults and pregnant women has been described, the association of maternal MedDiet during lactation with infant anthropometrics has not been examined.¹⁶ This study aims to examine the association of the MedDiet during lactation with infant anthropometric outcomes. We hypothesize that a greater maternal MedDiet score during lactation will be associated with infant growth and lower adiposity.

Methods

Study design

This study utilized data from ABC Baby, a longitudinal observational cohort study.¹⁷ The study was designed to collect data on infant eating behavior across six time points in infancy: at 2 weeks and at 2, 4, 6, 9, and 12 months. Data collection included questionnaires, embedded experiments to interrogate eating behaviors, and anthropometric measurements. All data were collected via home visits by trained research assistants. This report focuses on data related to maternal diet during lactation and infant anthropometrics at 6 months.

Participants and recruitment

Mother–infant dyads were recruited from a community in the Midwest United States through flyers, postcards, and social media. Mothers provided written informed consent for themselves and assent for their infants. The study was approved by the University of Michigan Institutional Review Board (HUM00103575). A total of 336 dyads contacted the study team expressing interest and were screened for eligibility (Fig. 1). Given that some of the embedded experiments in the overall study protocol involved feeding with an artificial nipple, infants were required at enrollment to have consumed at least 2 oz in one feeding from an artificial nipple and bottle at least once per week; additional inclusion and exclusion criteria can be found in Reynolds et al.¹⁷ The study was designed to enroll infants beginning at age 2 weeks. To facilitate recruitment, dyads were also permitted to enroll at infant ages 2 or 4 months. Data collected at enrollment are referred to as “baseline.”

FIG. 1.

Participant flow diagram. *Numbers may not add up to 100% due to potential subjects having more than one exclusion criteria during screening.

This analysis was limited to the subset of these 284 dyads who had complete and plausible (>500 and <5,000 kcal/day) maternal dietary intake at 2 weeks or 2 months, resulting in a base sample of 212 dyads (Fig. 1). An additional 45 dyads were excluded due to missing data for infant anthropometrics at 6 months, infant feeding practices at 2 months, and/or covariates, resulting in a final analytic sample size of 167. A secondary analysis excluded infants who received any formula at 2 months to understand if the associations differed among exclusively human milk-fed infants.

Measures

Demographics

Demographics were collected using a questionnaire and completed with a research assistant. Mothers reported their birth date (from which age was calculated), race/ethnicity, education level, and whether the pregnancy was a singleton or multiple gestation. Mothers also reported infant sex, birth date, and due date (from which gestational age was calculated).

Maternal race/ethnicity response options were those used by the National Institutes of Health (i.e., American Indian or Alaska Native, Asian, Black or African American, Hispanic or Latino, Native Hawaiian or Pacific Islander, White, Multiracial, and Other). Due to the small sample sizes in certain race/ethnicity groups, maternal race/ethnicity was collapsed into Black non-Hispanic, Hispanic any race, Other, and White non-Hispanic.

Anthropometry

Infant anthropometry was measured at 6 months of age, and the mother reported the birth weight. Research staff were certified in measurement techniques by a certified trainer and recertified annually. Infants were weighed without clothing or a diaper. Weight was measured on a BD-585 Digital Infant Scale (Tanita) in duplicate and averaged. The infant was weighed a third time if the weights differed by more than 0.1 kg. Recumbent length was measured using a pediatric stadiometer (M-PED LB 35-107-X; Ellard Instruments). The infant was aligned in the Frankfort horizontal plane, and the legs were positioned according to Shorr’s standards.¹⁸ Length was measured to the nearest 0.1 cm in duplicate and averaged. If measurements were not within 0.2 cm, a third measurement was obtained. Weight-for-age (WAZ), length-for-age (LAZ), and weight-for-length (WLZ) Z-scores were calculated based on the World Health Organization (WHO) growth charts.¹⁹ Flank skinfolds were measured using Harpenden skinfold calipers after calibration. All skinfold measurements were taken on the left-hand side of the infant’s body twice and recorded to the nearest 0.1 mm. A third measurement was taken if the measurements were more than 0.5 mm different. The measurements were averaged.

Mothers reported their pre-pregnancy weight. Maternal height was measured at baseline using a portable stadiometer (Seca 213/217).

Breastfeeding intensity

Breastfeeding intensity was calculated at age 2 months. Breastfeeding intensity is calculated by dividing the number of human milk feeds by the total number of liquid feeds (human milk feeds, formula feeds, and cow’s milk feeds).²⁰ The closer the ratio is to 1, the greater the intensity of human milk exposure. Select questions were used from age-appropriate questionnaires developed for the Infant Feeding Practices Study II (IFPS II) to assess infant dietary intake.^21,22 Responses to the question item “In the past 7 days, how often was your baby fed each food listed below?” were used to determine breastfeeding intensity.

Maternal diet

Food frequency questionnaire

Maternal diet was assessed using a 135-item food frequency questionnaire (FFQ) developed for the IFPS II.²² The IFPS II maternal FFQ was an adapted version of the National Cancer Institute (NCI) Dietary History Questionnaire (DHQ).^22,23 The IFPS II FFQ has yet to be validated to our knowledge; however, the DHQ has been validated against 24-hour recalls.²³ Overall, the DHQ performs as well or better than other FFQs with which it has been compared.²³

The IFPS II FFQ asked mothers to report intake within the past month. Because the IFPS II FFQ refers to the last month, responses at 2 weeks include some data regarding intake in the last week(s) of pregnancy.

Dietary intake values were generated using the NCI Diet*Calc program, with some adaptations required to account for changes in the DHQ for the IFPS II.²⁴ The Diet*Calc software generates intake of food groups based on the US Department of Agriculture’s pyramid servings. These variables represent standardized cup equivalent servings of food groups from all sources. In the Diet*Calc program, protein foods (meats, poultry, seafood, nuts, and seeds) were disaggregated into protein and fat fractions and summarized as ounces of lean meat equivalents.

MedDiet scoring

Maternal diet resemblance to the MedDiet was calculated using the alternate Mediterranean (aMed) score that was adapted from the MedDiet.²⁵ Two versions of the MedDiet score were calculated. One considered eight components: intake of vegetables (excluding potatoes), fruits, whole grains, nuts and seeds, legumes, fish, monounsaturated fat (MUFA)-to-saturated fat (SFA) ratio (MUFA:SFA), and red and processed meat. An alternative score included added sugar as a component. Added sugar is not traditionally part of the aMed scoring, but it is included in other MedDiet scoring indices.²⁶

For beneficial components (vegetables, fruits, whole grains, nuts and seeds, legumes, MUFA:SFA), mothers whose consumption was below the median were assigned a value of 0, and those at or above the median for the base population sample were assigned a value of 1. For red and processed meats and added sugar, which are considered detrimental, mothers whose consumption was below the median for the base population sample were assigned a value of 1, and persons whose consumption was at or above the median were assigned a value of 0. Total MedDiet score ranged from 0 (minimal resemblance) to 8 (maximal resemblance) without including added sugar and from 0 to 9 with added sugar included. Although alcohol consumption is traditionally a component of MedDiet scoring, it was not included in this analysis, given the breastfeeding sample. The U.S. Centers for Disease Control and Prevention recommends that breastfeeding mothers wait at least 2 hours after alcohol consumption to breastfeed. The time between consumption and infant feeding was unable to be determined; thus, infant exposure to alcohol in human milk was unable to be determined.²⁷ MedDiet components were adjusted for total energy intake using the residual method for individual component analysis.²⁸

Statistical analysis

Data analysis was performed by using SAS 9.4 statistical software. Univariate statistics were used to describe the sample. Multiple linear regression models were used to evaluate the association of diet variables with each of the anthropometric outcomes (WAZ, LAZ, WLZ, and flank skinfold thickness). Diet variable predictors include MedDiet score and individual continuous energy-adjusted MedDiet components (vegetables, fruits, whole grains, nuts and seeds, legumes, fish, MUFA:SFA, red and processed meats, and added sugar). The impact of added sugar on health outcomes is often thought to be due to the calories associated without nutrients added.²⁹ However, recent literature has shown that added sugar may also be related to ultra-processed food consumption,³⁰ associated with decreased vitamin content in milk,³¹ and infant overweight status.³² As such, added sugar adjusted for energy intake and added sugar unadjusted for intake were independently modeled in relation to infant anthropometrics.

Final models were adjusted for variables with the potential for confounding based on a priori knowledge, including maternal body mass index (BMI) at baseline (kg/m²), infant birth weight (kg), and breastfeeding intensity at 2 months (closer to 0 resembling exclusively formula-fed and closer to 1 resembling exclusively human milk-fed). Sensitivity analyses were performed exploring maternal education level and maternal age as potential covariates; the results did not change interpretations, and these covariates were not included in the final models.

A secondary analysis was performed, limiting the sample to only infants who were exclusively human milk-fed at 2 months. The secondary analysis models were adjusted for maternal BMI and infant birth weight.

Results

Characteristics of the 167 dyads’ included sample analysis are shown in Table 1. The mean maternal age was 31.6 (standard deviation [SD] 4.5) years. Mothers in the sample analysis were 8.5% Black non-Hispanic, 2.4% Hispanic, 12.1% “Other,” and 77.0% White non-Hispanic. Mothers in the sample were highly educated, with 38.3% having attained a 4-year college degree and 35.3% having more than a 4-year college degree. The cohort of infants was 49.1% male, the mean reported birth weight was 3.5 (SD 0.4) kg, and the mean gestational age was 39.6 (SD 1.1) weeks.

Table 1.

Demographic, Pregnancy, and Birth Characteristics (n = 167)

Mother characteristics	Value
Maternal age (years), mean (SD)	31.6 (4.5)
Maternal race/ethnicity, n (%), missing data for n = 2
Black, non-Hispanic	14 (8.5%)
Hispanic, any race	4 (2.4%)
Other (American Indian/Alaska Native, Asian, Native Hawaiian/Pacific Islander, multiracial, non-Hispanic)	20 (12.1%)
White, non-Hispanic	127 (77.0%)
Maternal education, n (%)
High school diploma or GED (or less)	9 (5.4%)
Some college to a 2-year degree	35 (21.0%)
Completed a 4-year degree	64 (38.3%)
Completed a postgraduate degree	59 (35.3%)
Maternal BMI, mean (SD)	29.2 (6.1)
Infant characteristics
Infant sex, male, n (%)	82 (49.1%)
Birth weight (kg), mean (SD)	3.5 (0.4)
Gestational age (weeks), mean (SD)	39.6 (1.1)
Breastfeeding intensity score, mean (SD)	0.8 (0.4)
Infant age at 6 month visit (months), mean (SD)	6.3 (0.5)

SD, standard deviation.

The 167 dyads in the analytic sample were compared with the 117 dyads excluded. Maternal race/ethnicity was significantly different (p = 0.0045); mothers excluded were 20.5% non-Hispanic Black, 6.9% Hispanic, 12.0% “Other,” and 60.7% non-Hispanic White. Maternal education level was significantly different between samples (p = 0.0004); 15.4% of the excluded sample attained a 4-year college degree compared with 38.3% of the included sample. Dyads included and excluded did not differ in terms of maternal age (p = 0.099), maternal BMI (p = 0.1293), infant birth weight (p = 0.2861), and gestational age (p = 0.1202).

Maternal food intake is summarized in Table 2. The DGA provides guidance on intake of vegetables (2.5 cup equivalents/day), fruits (2 cup equivalents/day), whole grains (≥3 oz equivalents/day), seafood (8 oz equivalents/week translating to 1.14 oz equivalents/day), nuts and seeds (5 oz equivalents/week translating to 0.71 oz equivalent/day), and added sugar (<10% of daily calories).²⁹ In general, the sample met the guidelines for vegetable and fruit intake but did not meet the guidelines for whole grain, fish, and nut intake. Half of the sample met the guidelines for added sugar, with the average participant consuming 12.6 teaspoons of added sugar a day. Current guidelines do not provide specific recommendations on legumes, MUFA:SFA ratio, or red and processed meat for lactating women.

Table 2.

Maternal Diet Characteristics

Diet variable per day	Median (Q1, Q3)
MedDiet score, without added sugar	4 (3, 5)
MedDiet score, with added sugar	5 (3, 6)
MedDiet components
Vegetables (excluding potatoes), cup equivalents	2.6 (2.0, 3.6)
Fruit, cup equivalents	2.1 (1.5, 3.2)
Whole grains, oz equivalents	1.1 (0.7, 1.6)
Nuts and Seeds, oz of lean meat equivalents	0.4 (0.2, 0.7)
Legumes, oz	0.1 (0.1, 0.2)
Fish, oz of lean meat equivalents	0.3 (0.1, 0.5)
MUFA:SFA ratio	1.2 (1.0, 1.3)
Red and processed meat, oz of lean meat equivalents	1.7 (1.1, 2.4)
Added sugar, energy adjusted, teaspoons	12.6 (9.3, 17.0)
Added sugar, teaspoons	12.5 (8.3, 18.3)

Table 3 summarizes infant anthropometric data. At 6 months, infants weighed 7.8 (SD 0.9) kg and were 66.4 (SD 2.3) cm in length. On an average, infants in this sample were marginally shorter compared with the WHO growth standards, with a LAZ score mean of −0.3, and marginally heavier based on a WLZ score mean of 0.3.

Table 3.

Infant Growth Characteristics at Age 6 Months, Mean (SD)

Weight (kg)	7.8 (0.9)
Length (cm)	66.5 (2.3)
Weight-for-age Z-score	0.0 (1.0)
Length-for-age Z-score	−0.3 (1.0)
Weight-for-length Z-score	0.3 (0.9)
Flank skinfold thickness (mm)	7.3 (3.1)

SD, standard deviation.

Results of linear models testing the association of MedDiet score with infant anthropometric outcomes in the total sample, adjusting for breastfeeding intensity, maternal BMI, and birth weight, are shown in Table 4. A higher MedDiet score was associated with lower flank skinfold thickness (β = −0.33). Individual components of MedDiet scoring were evaluated. Higher intakes of fruit and fish were both associated with lower flank skinfold thickness (β = −0.52 and −1.26, respectively); higher nuts and seed intake was associated with higher WLZ (β = 0.29); higher red and processed meat intake was associated with lower WAZ (β = −0.18) and LAZ (β = −0.18). When added sugar was adjusted for energy intake, a higher intake of added sugar was associated with lower WLZ (β = −0.02).

Table 4.

Associations of Maternal Diet with Anthropometric Outcomes (n = 167)^a

	WAZ		LAZ		WLZ		Flank (n = 152)
Diet components	Crude	Adjusted	Crude	Adjusted	Crude	Adjusted	Crude	Adjusted
MedDiet score	0.01 (−0.08, 0.09)	0.02 (−0.07, 0.11)	0.02 (−0.07, 0.10)	0.02 (−0.06, 0.11)	0.00 (−0.08, 0.09)	0.02 (−0.07, 0.10)	−0.29 (−0.58, −0.01)^*	−0.33 (−0.63, −0.02)^*
MedDiet score, with added sugar	0.02 (−0.06, 0.10)	0.03 (−0.05, 0.10)	0.02 (−0.06, 0.10)	0.02 (−0.06, 0.10)	0.02 (−0.06, 0.09)	0.02 (−0.06, 0.10)	−0.22 (−0.48, 0.04)	−0.25 (−0.53, 0.03)
Vegetables^b	0.07 (−0.02, 0.17)	0.06 (−0.03, 0.15)	0.08 (−0.02, 0.17)	0.06 (−0.03, 0.15)	0.04 (−0.05, 0.14)	0.04 (−0.05, 0.13)	0.13 (−0.18, 0.44)	0.13 (−0.18, 0.45)
Fruits^b	−0.06 (−0.16, 0.03)	−0.05 (−0.14, 0.05)	−0.06 (−0.16, 0.03)	−0.05 (−0.14, 0.04)	−0.03 (−0.12, 0.06)	−0.02 (−0.11, 0.07)	−0.48 (−0.78, −0.18)^*	−0.52 (−0.83, −0.21)^**
Whole grains^b	0.00 (−0.21, 0.21)	0.10 (−0.10, 0.305)	0.03 (−0.18, 0.24)	0.12 (−0.09, 0.32)	−0.01 (−0.22, 0.19)	0.05 (−0.15, 0.25)	0.52 (−0.23, 1.27)	0.57 (−0.20, 1.34)
Nuts and seeds^c	0.25 (−0.04, 0.54)	0.23 (−0.05, 0.50)	0.02 (−0.27, 0.31)	0.01 (−0.27, 0.29)	0.31 (0.03, 0.59)^*	0.29 (0.01, 0.56)^*	0.57 (−0.41, 1.55)	0.53 (−0.47, 1.53)
Legumes^d	0.09 (−0.29, 0.47)	−0.06 (−0.42, 0.31)	0.26 (−0.11, 0.66)	0.19 (−0.19, 0.55)	−0.08 (−0.45, 0.29)	−0.20 (−0.57, 0.17)	−0.23 (−1.67, 1.21)	−0.34 (−1.81, 1.14)
Fish^c	−0.19 (−0.55, 0.17)	−0.22 (−0.56, 0.12)	−0.01 (−0.37, 0.35)	−0.03 (−0.37, 0.31)	−0.24 (−0.59, 0.11)	−0.27 (−0.61, 0.07)	−1.20 (−2.41, 0.01)	−1.26 (−2.49, −0.04)^*
MUFA:SFA^e	−0.27 (−0.78, 0.24)	−0.24 (−0.73, 0.25)	−0.36 (−0.87, 0.14)	−0.33 (−0.83, 0.16)	−0.06 (−0.56, 0.43)	−0.05 (−0.54, 0.44)	−0.29 (−1.99, 1.41)	−0.37 (−2.12, 1.38)
Red and processed meat^c	−0.15 (−0.30, 0.01)	−0.18 (−0.33, −0.03)^*	−0.16 (−0.32, −0.01)^*	−0.18 (−0.33, −0.03)^*	−0.07 (−0.22, 0.09)	−0.09 (−0.25, 0.06)	−0.15 (−0.68, 0.37)	−0.17 (−0.71, 0.37)
Added sugar, energy adjusted^f	−0.01 (−0.03, 0.00)	−0.01 (−0.03, 0.01)	0.00 (−0.02, 0.02)	0.00 (−0.01, 0.02)	−0.02 (−0.03, 0.00)^*	−0.02 (−0.03, 0.00)^*	−0.01 (−0.07, 0.04)	−0.01 (−0.07, 0.04)
Added sugar^f	0.00 (−0.01, 0.01)	0.00 (−0.01, 0.01)	0.01 (−0.01, 0.02)	0.01 (0.00, 0.02)	−0.01 (−0.02, 0.01)	−0.01 (−0.02, 0.01)	−0.01 (−0.05, 0.03)	−0.01 (−0.05, 0.03)

Linear regression model results, crude and adjusted for maternal BMI at baseline, birth weight, and breastfeeding intensity score.

Vegetable, fruit, and whole grain intake expressed as cup equivalents/day according to MyPlate, intakes were energy adjusted.

Fish, nut and seed, and red and processed meats expressed as ounces of lean meat equivalents/day according to MyPlate, intakes were energy adjusted.

Legumes expressed as ounces/day, intake was energy adjusted.

MUFA:SFA refers to monounsaturated-to-saturated fat ratio, and the ratio was energy adjusted.

Added sugar expressed in teaspoons/day.

p < 0.05.

p < 0.01.

When the analysis was limited to infants (Table 5) who were exclusively human milk-fed at age 2 months as part of the secondary analysis (n = 109), higher MedDiet score (with and without added sugar) (β = −0.41, −0.34), higher intake of fruit (β = −0.55), and higher intake of fish (β = −1.40) remained associated with lower flank skinfold thickness.

Table 5.

Associations of Maternal Diet with Anthropometric Outcomes, Among Exclusively Human Milk-Fed Infants (n = 109)^a

	WAZ		LAZ		WLZ		Flank (n = 100)
Diet components	Crude	Adjusted	Crude	Adjusted	Crude	Adjusted	Crude	Adjusted
MedDiet Score	−0.01 (−0.13, 0.10)	0.00 (−0.11, 0.12)	0.02 (−0.09, 0.13)	0.02 (−0.09, 0.13)	−0.03 (−0.13, 0.07)	−0.01 (−0.11, 0.09)	−0.44 (−0.78, −0.09)^*	−0.41 (−0.78, −0.05)^*
MedDiet score, with added sugar	0.01 (−0.10, 0.11)	0.01 (−0.09, 0.12)	0.025 (−0.08, 0.13)	0.01 (−0.08, 0.12)	−0.01 (−0.10, 0.08)	0.01 (−0.10, 0.10)	−0.37 (−0.69, −0.05)^*	−0.34 (−0.68, 0.00)^*
Vegetables^b	0.05 (−0.07, 0.17)	0.05 (−0.07, 0.16)	0.05 (−0.07, 0.17)	0.04 (−0.07, 0.15)	0.03 (−0.08, 0.14)	0.03 (−0.07, 0.14)	0.07 (−0.30, 0.43)	0.08 (−0.29, 0.45)
Fruits^b	−0.10 (−0.21, 0.01)	−0.09 (−0.19, 0.02)	−0.08 (−0.19, 0.03)	−0.07 (−0.18, 0.01)	−0.07 (−0.16, 0.03)	−0.05 (−0.15, 0.04)	−0.56 (−0.87, −0.25)^**	−0.55 (−0.87, −0.22)^*
Whole grains^b	−0.08 (−0.36, 0.20)	0.04 (−0.23, 0.31)	−0.04 (−0.31, 0.24)	0.09 (−0.17, 0.36)	−0.08 (−0.32, 0.17)	−0.01 (−0.26, 0.24)	0.42 (−0.56, 1.40)	0.462 (−0.54, 1.46)
Nuts and Seeds^c	0.27 (−0.06, 0.60)	0.23 (−0.09, 0.54)	0.15 (−0.17, 0.48)	0.10 (−0.21, 0.41)	0.23 (−0.06, 0.52)	0.21 (−0.07, 0.50)	0.69 (−0.33, 1.72)	0.72 (−0.31, 1.75)
Legumes^d	0.04 (−0.40, 0.48)	−0.13 (−0.55, 0.30)	0.08 (−0.35, 0.51)	−0.10 (−0.51, 0.32)	−0.01 (−0.39, 0.40)	−0.10 (−0.49, 0.29)	−0.38 (−1.95, 1.19)	−0.37 (−1.97, 1.24)
Fish^c	−0.21 (−0.62, 0.20)	−0.22 (−0.61, 0.17)	0.01 (−0.40, 0.41)	−0.02 (−0.40, 0.37)	−0.29 (−0.64, 0.07)	−0.29 (−0.64, 0.07)	−1.41 (−2.68, −0.15)^*	−1.40 (−2.67, −0.14)^*
MUFA:SFA^e	−0.33 (−0.97, 0.32)	−0.31 (−0.93, 0.31)	−0.33 (−0.96, 0.31)	−0.34 (−0.94, 0.27)	−0.16 (−0.73, 0.41)	−0.13 (−0.70, 0.44)	0.37 (−1.64, 2.39)	0.47 (−1.55, 2.50)
Red and processed meat^c	−0.15 (−0.34, 0.05)	−0.18 (−0.37, 0.01)	−0.13 (−0.32, 0.06)	−0.15 (−0.33, 0.03)	−0.09 (−0.26, 0.08)	−0.12 (−0.30, 0.05)	−0.18 (−0.78, 0.43)	−0.21 (−0.82, 0.41)
Added sugar, energy adjusted^f	−0.01 (−0.03, 0.02)	0.00 (−0.03, 0.02)	−0.01 (−0.03, 0.02)	−0.01 (−0.03 0.02)	0.00 (−0.02, 0.02)	0.00 (−0.02, 0.02)	0.00 (−0.08, 0.07)	−0.01 (−0.08, 0.07)
Added sugar^f	0.01 (−0.01, 0.02)	0.01 (−0.01, 0.02)	0.00 (−0.02, 0.02)	0.00 (−0.01, 0.02)	0.01 (−0.01, 0.02)	0.01 (−0.01, 0.02)	0.00 (−0.05, 0.05)	0.00 (−0.06, 0.05)

Linear regression model results, crude and adjusted for maternal BMI at baseline, birth weight, and breastfeeding intensity score.

Vegetable, fruit, and whole grain intake expressed as cup equivalents/day according to MyPlate, intakes were energy adjusted.

Fish, nut and seed, and red and processed meats expressed as ounces of lean meat equivalents/day according to MyPlate, intakes were energy adjusted.

Legumes expressed as ounces/day, intake was energy adjusted.

MUFA:SFA refers to monounsaturated-to-saturated fat ratio, and the ratio was energy adjusted.

Added sugar expressed in teaspoons/day, energy adjusted and raw.

p < 0.05.

p < 0.01.

Discussion

In this analysis, we demonstrated associations between maternal diet during lactation and infant anthropometrics at 6 months. MedDiet score, adapted to the lactation period, was negatively associated with flank skinfold thickness in the full sample and among infants who were exclusively human milk-fed. The addition of added sugar to the MedDiet scoring was not associated with growth in the full sample but was associated in the exclusively human milk-fed sample. Additionally, we found individual MedDiet components were associated with infant anthropometrics at 6 months.

In terms of specific MedDiet components, maternal intake of fruit and fish was negatively associated with flank skinfold thickness. This association held for primary and secondary models. Additionally, in the full sample models, red and processed meat intake was negatively associated with WAZ and LAZ, and added sugar (not energy adjusted) was negatively associated with WLZ.

There is limited research evaluating MedDiet score during lactation and infant anthropometrics. However, maternal added sugar intake during lactation and infant anthropometrics has been examined in a prospective cohort of breastfeeding mother–infant dyads. Nagel et al. found that maternal added sugar intake was positively associated in the crude model with infant percent body fat at 6 months, as measured by dual x-ray absorptiometry (DEXA), but attenuated when adjusted for covariates, such as energy intake.³³ Interestingly, we found the opposite, with added sugar being associated with WLZ only when adjusted for energy intake but attenuated when energy adjusted. This shows that it may not be the calories from added sugar influencing infant growth but rather another mechanism, such as modifying overall intake and replacing nutrient-rich foods in the diet. It could also be related to the foods with which added sugar is often associated, such as ultra-processed foods.^30,34 However, we recognize that our ability to estimate infant body fat by WLZ is inferior to a direct measurement by DEXA.³⁵

The results related to maternal intake of fruits, nuts and seeds, fish, and red and processed meats are complicated to explain due to a lack of understanding of these factors on milk composition. Very few studies have explored food groups in relation to infant anthropometrics or human milk composition. In a study on maternal adherence to the dietary approaches to stop hypertension (DASH) diet while lactating, mothers on the diet consuming more fruits, nuts, legumes, seeds, and whole grains were found to have higher levels of antioxidants in milk and lower levels of triglycerides.³⁶ Another study found that maternal intake of cereals and legumes during lactation was positively associated with human milk saturated fatty acid levels.³⁷ This indicates these food groups may be impacting milk composition and thus impacting infant anthropometrics, as our results have indicated. Further research is needed to examine the connection between human milk composition due to maternal diet and infant growth in exclusively human milk-fed infants.

This analysis has several strengths, including a prospective cohort design and longitudinal data. Our study also had several limitations that should be addressed in future research. The sample primarily included White mothers with relatively high education. There was attrition, and the sample of infants contributing data for this analysis differed from the overall sample in maternal race/ethnicity and education level. These findings may, therefore, not be generalizable to mother–infant dyads without these characteristics. There was significant missing data, primarily due to the longitudinal commitment required of the dyads. Study participation required that infants had taken at least 2 oz from an artificial nipple at least once per week by 2 months, which excludes infants who are exclusively breastfed from the breast through 2 months, limiting generalizability to this group. However, of mothers who breastfeed, 85% express milk and feed the infant from an artificial nipple in the first 4 months of infancy, with most doing so within the first week.³⁸ In our measurement of infant length, the slightly lower than average Z-scores suggest that despite extensive training and certification of our data collectors, there could have been some degree of error. Finally, regarding infant and maternal intake, the measurement of infant breastfeeding intensity and maternal diet is inherently challenging. In terms of infant breastfeeding intensity, while we did have prospectively collected data, we measured this in the number of feedings per day. We were not able to account for volume per feeding, which may have introduced some degree of error. Measuring diet intake is prone to measurement error,³⁹ and the FFQ instrument utilized for maternal diet was an adapted version of a validated DHQ FFQ that is yet to be validated.^22,23 This could introduce measurement error into our diet estimations.

Conclusion

In summary, we found that higher maternal MedDiet score was associated with lower flank skinfold thickness, while specific MedDiet components were also associated with anthropometric Z-scores at 6 months, in differing directions based on component. Continued research and investigation into the impact of maternal diet and overall dietary patterns on infant anthropometrics during lactation is an important future direction.

Footnotes

Authors’ Contributions

A.G., A.B., L.E., J.C.L., and B.G.: designed research study objectives and aims. A.G., A.B., J.R., N.K., and J.S.: conducted the research. J.S., A.L.M., A.N.G., N.K., and J.C.L.: are the primary investigators on the original research cohort, overseeing the data collection processes and methodology. A.G.: wrote the article, with aid from L.E., J.C.L., and B.G.: All authors have read and approved the final article.

Data Availability

Data described in the article, code book, and analytic code will be made available upon request.

Disclosure Statement

The authors have no financial relationships or conflicts of interest to disclose.

Funding Information

This work was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (grant no. R01HD084163).

References

Ballard

, Morrow

. Human milk composition: Nutrients and bioactive factors. Pediatr Clin North Am, 2013; 60(1):49–74; doi: 10.1016/j.pcl.2012.10.002

Oftedal

. The evolution of milk secretion and its ancient origins. Animal, 2012; 6(3):355–368; doi: 10.1017/S1751731111001935

Anonymous. What are the recommendations for breastfeeding? | NICHD—Eunice Kennedy Shriver National Institute of Child Health And Human Development. 2017. Available from: https://www.nichd.nih.gov/health/topics/breastfeeding/conditioninfo/recommendations [Last accessed: April 26, 2024].

Bravi

, Wiens

, Decarli

, et al. Impact of maternal nutrition on breast-milk composition: A systematic review. Am J Clin Nutr, 2016; 104(3):646–662; doi: 10.3945/ajcn.115.120881

Casavale

, Ahuja

JKC

, Wu

, et al. NIH workshop on human milk composition: Summary and visions. Am J Clin Nutr, 2019; 110(3):769–779; doi: 10.1093/ajcn/nqz123

Anonymous. About the seven countries study. n.d. Available from: https://www.sevencountriesstudy.com/about-the-study/ [Last accessed: April 4, 2022].

Martínez-González

, Gea

, Ruiz-Canela

. The Mediterranean diet and cardiovascular health. Circ Res, 2019; 124(5):779–798; doi: 10.1161/CIRCRESAHA.118.313348

Sofi

, Cesari

, Abbate

, et al. Adherence to Mediterranean diet and health status: Meta-analysis. BMJ, 2008; 337:a1344; doi: 10.1136/bmj.a1344

Panagiotakos

, Kastorini

C-M

, Pitsavos

, et al. The current Greek diet and the Omega-6/Omega-3 balance: The Mediterranean diet score is inversely associated with the Omega-6/Omega-3 ratio. World Rev Nutr Diet, 2011; 102:53–56; doi: 10.1159/000327791

10.

H Al Wattar

, Dodds

, Placzek

, et al. ESTEEM study group. Mediterranean-style diet in pregnant women with metabolic risk factors (ESTEEM): A pragmatic multicentre randomised trial. PLoS Med, 2019; 16(7):e1002857; doi: 10.1371/journal.pmed.1002857

11.

Chatzi

, Mendez

, Garcia

, et al. INMA and RHEA study groups. Mediterranean diet adherence during pregnancy and fetal growth: INMA (Spain) and RHEA (Greece) mother-child cohort studies. Br J Nutr, 2012; 107(1):135–145; doi: 10.1017/S0007114511002625

12.

Gonzalez-Nahm

, Marchesoni

, Maity

, et al. Maternal Mediterranean diet adherence and its associations with maternal prenatal stressors and child growth. Curr Dev Nutr, 2022; 6(11):nzac146; doi: 10.1093/cdn/nzac146

13.

Di Maso

, Bravi

, Ferraroni

, et al. Adherence to Mediterranean diet of breastfeeding mothers and fatty acids composition of their human milk: Results from the Italian MEDIDIET study. Front Nutr, 2022; 9:891376; doi: 10.3389/fnut.2022.891376

14.

Karbasi

, Mohamadian

, Naseri

, et al. A Mediterranean diet is associated with improved total antioxidant content of human breast milk and infant urine. Nutr J, 2023; 22(1):11; doi: 10.1186/s12937-023-00841-0

15.

Jensen

. Lipids in human milk. Lipids, 1999; 34(12):1243–1271; doi: 10.1007/s11745-999-0477-2

16.

Ellsworth

, Harman

, Padmanabhan

, et al. Lactational programming of glucose homeostasis: A window of opportunity. Reproduction, 2018; 156(2):R23–R42; doi: 10.1530/REP-17-0780

17.

Reynolds

LAF

, McCaffery

, Appugliese

, et al. Capacity for regulation of energy intake in infancy. JAMA Pediatr, 2023; 177(6):590–598; doi: 10.1001/jamapediatrics.2023.0688

18.

Development UD of TC for, Office US, Programme NHSC. How to weigh and measure children: Assessing the nutritional status of young children in household surveys. Department of Technical Co-operation for Development and Statistical Office; 1986.

19.

Anonymous. WHO child growth standards: Length/Height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age: Methods and development. n.d. Available from: https://www.who.int/publications-detail-redirect/924154693X [Last accessed: June 5, 2023]

20.

Piper

, Parks

. Use of an intensity ratio to describe breastfeeding exclusivity in a national sample. J Hum Lact, 2001; 17(3):227–232; doi: 10.1177/089033440101700306

21.

Grummer-Strawn

, Scanlon

, Fein

. Infant feeding and feeding transitions during the first year of life. Pediatrics, 2008; 122(,Suppl 2):S36–S42; doi: 10.1542/peds.2008-1315d

22.

Fein

, Labiner-Wolfe

, Shealy

, et al. Infant feeding practices study II: Study methods. Pediatrics, 2008; 122(Suppl 2):S28–S35; doi: 10.1542/peds.2008-1315c

23.

Subar

, Thompson

, Kipnis

, et al. Comparative validation of the Block, Willett, and National Cancer Institute food frequency questionnaires : The eating at America’s Table study. Am J Epidemiol, 2001; 154(12):1089–1099; doi: 10.1093/aje/154.12.1089

24.

Anonymous. Diet history questionnaire II (DHQ II) for U.S. & Canada. Government. 2023. Available from: https://epi.grants.cancer.gov/dhq2/ [Last accessed: April 26, 2024].

25.

Fung

, Rexrode

, Mantzoros

, et al. Mediterranean diet and incidence of and mortality from coronary heart disease and stroke in women. Circulation, 2009; 119(8):1093–1100; doi: 10.1161/CIRCULATIONAHA.108.816736

26.

Aoun

, Papazian

, Helou

, et al. Comparison of five international indices of adherence to the Mediterranean diet among healthy adults: Similarities and differences. Nutr Res Pract, 2019; 13(4):333–343; doi: 10.4162/nrp.2019.13.4.333

27.

Anonymous. Alcohol and breastfeeding. 2022. Available from: https://www.cdc.gov/breastfeeding/breastfeeding-special-circumstances/vaccinations-medications-drugs/alcohol.html [Last accessed: May 3, 2023].

28.

Willett

. Nutritional Epidemiology. OUP: USA; 2013.

29.

Anonymous. Dietary Guidelines for Americans, 2020-2025. n.d.

30.

Martínez Steele

, Baraldi

, Louzada

MLdC

, et al. Ultra-processed foods and added sugars in the US diet: Evidence from a nationally representative cross-sectional study. BMJ Open, 2016; 6(3):e009892; doi: 10.1136/bmjopen-2015-009892

31.

Amorim

NCM

, da Silva

AGCL

, Rebouças

, et al. Dietary share of ultra-processed foods and its association with vitamin E biomarkers in Brazilian lactating women. Br J Nutr, 2022; 127(8):1224–1231; doi: 10.1017/S0007114521001963

32.

de Melo

JMM

, Dourado

, de Menezes

RCE

, et al. Early onset of overweight among children from low-income families: The role of exclusive breastfeeding and maternal intake of ultra-processed food. Pediatr Obes, 2021; 16(12):e12825; doi: 10.1111/ijpo.12825

33.

Nagel

, Jacobs

, Johnson

, et al. Maternal dietary intake of total fat, saturated fat, and added sugar is associated with infant adiposity and weight status at 6 mo of age. J Nutr, 2021; 151(8):2353–2360; doi: 10.1093/jn/nxab101

34.

DiNicolantonio

, Berger

. Added sugars drive nutrient and energy deficit in obesity: A new paradigm. Open Heart, 2016; 3(2):e000469; doi: 10.1136/openhrt-2016-000469

35.

Demerath

, Fields

. Body composition assessment in the infant. Am J Hum Biol, 2014; 26(3):291–304; doi: 10.1002/ajhb.22500

36.

Karbasi

, Bahrami

, Hanafi-Bojd

, et al. Maternal adherence to a Dietary Approaches to Stop Hypertension (DASH) dietary pattern and the relationship to breast milk nutrient content. Matern Child Health J, 2023; 27(2):385–394; doi: 10.1007/s10995-022-03552-w

37.

Ramiro-Cortijo

, Herranz Carrillo

, Gila-Diaz

, et al. Association between adherence to the healthy food pyramid and breast milk fatty acids in the first month of lactation. Nutrients, 2022; 14(24):5280; doi: 10.3390/nu14245280

38.

Labiner-Wolfe

, Fein

, Shealy

, et al. Prevalence of breast milk expression and associated factors. Pediatrics, 2008; 122(,Suppl 2):S63–S68; doi: 10.1542/peds.2008-1315h

39.

Kipnis

, Subar

, Midthune

, et al. Structure of dietary measurement error: Results of the open biomarker study. Am J Epidemiol, 2003; 158(1):14–21; doi: 10.1093/aje/kwg091