Breast may not always be best: moderation of effects of postnatal depression by breastfeeding and infant sex

Background There is good evidence that female infants are particularly vulnerable to poor emotional outcomes following in utero glucocorticoid exposure. It is currently unclear whether such effects might persist into the postnatal period for breastfed infants, as maternal cortisol is expressed in breastmilk and is influenced by maternal psychological distress. We pre-registered hypotheses that maternal postnatal depression would be associated with infant negative emotionality, and that this effect would be moderated by breastfeeding status and infant sex. Methods We analysed data from the Wirral Child Health and Development Study (WCHADS), a prospective epidemiological study starting in pregnancy. Nine weeks after birth mothers self-reported depressive symptoms and breastfeeding status, and reported infant negative emotionality using the distress to limits subscale of the infant behaviour questionnaire (IBQ-R) when their infant was aged 9 weeks and 14 months. Maximum likelihood estimations made use of data from 857 mother-infant pairs. Results At 9 weeks of age, maternal postnatal depressive symptoms were positively associated with infant distress to limits; however, this effect was not moderated by infant sex or breastfeeding. At age 14 months, the association between postnatal depression symptoms and distress to limits was greatest in the breastfed females, whereas the association was smaller, but still significant, in the non-breastfed females. For males, the association was non-significant in both the breastfed and non-breastfed groups. A test of sex difference between breastfed males and females was significant. Conclusions We provide evidence that effects of maternal postnatal depression on child emotional outcomes are moderated by breastfeeding status and differ by infant sex. Female vulnerability to elevated maternal breastmilk glucocorticoids may, at least in part, explain these effects.

Characterizing Caregiver Verbalizations to Infants During the Introduction of a Novel Food

Objectives To characterize parental verbal prompts to infants during feeding and the relation of parent verbalizations to infant acceptance of a novel food. Methods Feeding interactions of parent-child dyads (n = 106; 54% boys; mean age 13.3 ± 4.9 mo) were video-recorded and all parental verbal communications to infants and infant acceptance/rejection of the bites were coded. Verbal prompts, defined a priori, were categorized as positive (8 codes, e.g., verbal encouragement), neutral (4 codes, e.g., comments about color of food) and negative (6 codes, e.g., pressure) prompts. Trained coders classified each utterance and recorded the number and rate of bites accepted (mean K = .81 for all codes). Frequency of verbal prompts were examined by infant sex (M/F), age (6–11.9, 12–17.9, & 18–24 mo), and over the course of the feeding using Wilcoxon rank sums & Kruskal Wallis tests (p < .05). Spearman's rho tested associations between verbalization frequency and both the number of accepted and rejected bites and the rate of bite acceptance. Results Total parent verbalizations increased with infant age (p = .002). No differences were noted by infant sex. Positive verbalizations declined over the course of the feed (p = .02), whereas neutral (p = .20) and negative (p = .09) verbalizations exhibited no significant change. Total verbalizations were positively correlated with the number of rejected bites (p = .0009). In contrast, fewer negative verbalizations were correlated with more accepted bites (p = .02) and a lower rate of acceptance (p = .002). The absence of verbalizations was correlated with more accepted bites (p = .02) and a higher rate of acceptance (p = .004). Conclusions During the course of offering a novel food, caregivers generally spoke more often with older infants and when children rejected the food. Caregivers appear to be more likely to employ verbal strategies to coax the acceptance of a food when children are reluctant to eat, rather than proactively engaging children throughout the feeding interaction. Furthermore, the nature and tone of caregiver verbalizations during feeding interactions may change over the course of the feeding session with positive verbalizations waning and the use of negative and neutral prompts increasing if children show reluctance to consume the food. Funding Sources The Sugar Association.

Abstract P171: A Longitudinal Study Of Maternal Thyroid Hormone Trajectories And Neonatal Anthropometric Measures

Introduction: Maternal thyroid hormones (TH) are essential for fetal growth and development. While prior studies have examined associations of TH with birthweight (BW), data on other neonatal anthropometric measures are scarce. Here we aimed to examine relations between TH and its trajectory across pregnancy with neonatal anthropometrics in a multi-racial pregnancy cohort. Hypothesis: We hypothesized that distinct TH trajectories across gestation would explain differences in neonatal anthropometrics including BW, length and sum of skinfolds (SSF). Methods: We used longitudinal data from 321 women who were included in a gestational diabetes (GDM) case-control study nested within the NICHD Fetal Growth Studies-Singletons Cohort (n=2802). Plasma free triiodothyronine (fT3), free thyroxine (fT4) and thyroid-stimulating hormone (TSH) were measured, and fT3/fT4 ratio was derived, using blood samples collected at gestational weeks (GW) 10-14, 15-26, 23-31 and 33-39. TH trajectories were identified using a latent class trajectory approach. BW was abstracted from medical records. Neonatal length and SSF were measured. To examine associations between TH levels and neonatal outcomes, we used weighted linear regression models, adjusted for maternal age, race/ethnicity, GW at blood collection, pre-pregnancy BMI, education, parity, infant sex, GDM diagnosis (33-39 GW only) and number of days post-delivery (length and SSF only). Results: Overall, neonates of women in the highest fT3 tertile at 33-39 GW had marginally higher BW (β: 205.5 g, 95% CI -23.4, 434.4) and BW z-score (β: 0.4, 95% CI 0.0, 0.8) than neonates of women in the lowest tertile. Significant interactions were observed by infant sex for associations of fT4 and BW, BW z-score and neonatal length (all p for interactions <0.05). Specifically, female neonates of women in the highest fT4 tertile had lower BW, BW z-score and length than those in the lowest tertile at 15-26 (β BW -416.3 g, 95% CI -753.7, -79.0; β BW zscore -0.5, 95% CI -0.9, -0.1; β length -1.5 cm, 95% CI -2.8, -0.1, respectively) and 23-31 GW (β BW -357.0 g, 95% CI -635.6, -78.4; β BW zscore -0.7, 95% CI -1.3, -0.1; β length -1.8 cm, 95% CI -3.3, -0.2, respectively). Across gestation, a low fT3 trajectory (class size 5.5%) was associated with lower neonatal SSF (β: -3.2 mm, 95% CI -6.0, -0.5) than an intermediate/referent trajectory (60.1%). A high fT3/fT4 ratio trajectory (16.2%) was associated with higher BW (β: 300.5 g, 95% CI 41.7, 559.4), BW z-score (β: 0.6, 95% CI 0.3, 0.8), and neonatal SSF (β: 3.3 mm, 95% CI 0.7, 5.9) than an intermediate/referent trajectory (44.3%). Conclusions: Our study suggests that both overall TH trajectory across gestation and TH levels at specific timepoints in pregnancy may be related to neonatal anthropometrics, with associations varying by infant sex. These novel data support the significance of monitoring thyroid hormones longitudinally in pregnancy.

Are the effects of air pollution on birth weight modified by infant sex and neighborhood socioeconomic deprivation? A multilevel analysis in Paris (France)

Adverse birth outcomes related to air pollution are well documented; however, few studies have accounted for infant sex. There is also scientific evidence that the neighborhood socioeconomic profile may modify this association even after adjusting for individual socioeconomic characteristics. The objective is to analyze the association between air pollution and birth weight by infant sex and neighborhood socioeconomic index. All birth weights (2008–2011) were geocoded at census block level. Each census block was assigned a socioeconomic deprivation level, as well as daily NO2 and PM10 concentrations. We performed a multilevel model with a multiple statistical test and sensible analysis using the spline function. Our findings suggest the existence of a differential association between air pollution and BW according to both neighborhood socioeconomic level and infant sex. However, due to multiple statistical tests and controlling the false discovery rate (FDR), all significant associations became either not statistically significant or borderline. Our findings reinforce the need for additional studies to investigate the role of the neighborhood socioeconomic which could differentially modify the air pollution effect.

Isolated fetal neural tube defects associate with increased risk of placental pathology: evidence from the Collaborative Perinatal Project.

Objective: To compare placental pathology and fetal growth in pregnancies with an isolated fetal neural tube defect (NTD; cases) to those without congenital anomalies (controls). We hypothesised that cases would be at an increased risk of placental pathology and poorer anthropometric outcomes at birth compared to controls Methods: We performed a matched case-cohort study using data from the Collaborative Perinatal Project. Cases (n=74) and controls (n=148) were matched (1:2 ratio) for maternal pre-pregnancy BMI, maternal race, infant sex, gestational age at birth and study site. Primary outcomes were placental characteristics (weight and size measurements, pathology). Secondary outcomes were infant birth outcomes. Subgroup analysis was done by type of NTD (spina bifida, anencephaly or encephalocele), infant sex, and preterm/term delivery. Data were analysed using adjusted generalized linear and nominal logistic regression models. Results are presented as adjusted β or adjusted odds ratio (aOR; 95% confidence interval). Results: Cases had lower placental weight (β=-22.2 g [-37.8 - -6.6]), surface area (β=-9.6 cm2 [-18.3 - -1.0]) and birth length z-scores (β=-0.4 [-0.7 - -0.001]) compared to controls. Cases were more likely to have a single umbilical artery (vs. two; 6 [8.1%] vs. 1 [0.7%]; aOR=301 [52.6 - 1726]), overall placental hypermaturity (9 [12.2%] vs. 5 [3.4%]; aOR=6.8 [3.1 - 14.7]), and many (vs. few) Hofbauer cells (9 [12.2%] vs. 7 [4.7%]; aOR=3.02 [1.2 - 7.3]), stromal fibrosis (9 [12.2%] vs. 10 [6.8%]; aOR=3.0 [1.4 - 6.3]) and pathological edema (11 [14.9%] vs. 12 [8.1%]; aOR=3.04 [1.4 - 6.7]) in placental terminal villi compared to controls. Placental pathology varied across NTD subtypes, infant sex, and preterms vs. term pregnancies. Conclusions: Fetuses with isolated NTDs may be at increased risk of placental pathology, which could be contributing to poor fetal growth in these pregnancies and subsequent postnatal morbidities.

Associations of Prenatal Exposure to Phthalates with Measures of Cognition in 4.5-Month-Old Infants

The association of prenatal phthalate exposure with physical reasoning was assessed in 159 (78 female; 81 male) 4.5-month-old infants from a prospective cohort. Phthalate metabolites were quantified in urine from 16–18 gestational weeks and a pool of five urines from across pregnancy. Infants’ looking times to physically impossible and possible events were recorded via infrared eye-tracking. Infants that recognize that one of the events is impossible will look at that event longer. Associations of phthalate biomarkers with looking time differences (impossible–possible) were adjusted for maternal age, infant sex, and order of event presentation, and effect modification by infant sex was assessed. Each interquartile range (IQR) increase of monoethyl phthalate in the pooled sample was associated with females’ increased looking time (β = 1.0; 95%CI = 0.3, 1.7 s) to the impossible event. However, for males, an IQR increase in monoethyl phthalate at 16–18 weeks (β = −2.5; 95%CI = −4.4,−0.6 s), the sum of di(isononyl) phthalate metabolites in the pooled sample (β = −1.0; 95%CI = −1.8, −0.1 s), and the sum of all phthalate metabolites in both samples (β = −2.3; 95%CI = −4.4, −0.2 s) were associated with increased looking to the possible event, suggesting that higher prenatal phthalate exposure is associated with poorer physical reasoning in male infants.