Umeå University's logo

Background: Diet contains components that may exert broad immunoregulatory properties, including the promotion of tolerance. The timing of food introduction may influence the development of allergies. The aim of this study was to relate the timing of food introduction with doctor’s diagnosed allergies (food allergy, atopic eczema, and allergic asthma) at six years of age in a cohort of Swedish infants.

Methods: At six years of age, 430 children from the Swedish NICE birth cohort were assessed for food allergies, atopic eczema, and allergic asthma by a pediatric allergologist. Timing of introduction of 18 predefined foods (potatoes and roots, fruits and/or fruit juice, berries, nuts and/or almonds, peanuts and/or peanut oil, bread and/or biscuits, butter, margarine, vegetable oils, cow’s milk, ice cream, sour milk (“fil”), yoghurt, meat, fish, eggs, porridge, and gruel) was collected monthly via parent-completed web-based questionnaires. Associations were examined using logistic regression adjusted for allergic heredity.

Results: During the first year of life, 27% of the children received peanuts and/or peanut oil and 32% nuts and/or almonds. Earlier meat introduction was associated with 24% lower odds of food allergy per month earlier introduced [adjusted OR (95% CI): 0.76 (0.60–0.99), p = 0.028]. A similar tendency was observed for eggs, although not surviving adjustment for allergic heredity. Children with food allergies had been introduced to margarine earlier than non-allergic children [9.0 (5.0–12.0) vs. 12.0 (9.0–12.0) months, p = 0.021], although this finding likely reflects reverse causation.

Conclusions: Our results suggest that earlier introduction of meat may be associated with a lower risk of food allergies later in childhood, although this finding must be interpreted with caution as causality cannot be proven. Timing of food introduction did not appear to be associated with atopic eczema or allergic asthma. Furthermore, our results indicate that recommendations for introduction of potentially allergenic foods during the first year of life were not effectively implemented.

Trial registration: ClinicalTrials.gov, NCT05809479, 12 April 2023, Retrospectively registered.

Background/Objectives: Iron deficiency early in life can impair infant growth and cognitive development. Here, we follow infants’ plasma ferritin levels—an indicator of iron stores—over the first year of life and relate these to birth characteristics, maternal characteristics, and infant feeding.

Methods: Children and their mothers enrolled in the Swedish birth cohort NICE (ClinicalTrials.gov identifier: NCT05809479) were followed from pregnancy to twelve months postpartum. Plasma ferritin was quantified in umbilical cord blood at birth (n = 345), in venous plasma at four months after birth (mother–infant dyads, n = 133), and at twelve months of age (n = 158), using sandwich ELISA. Perinatal and postnatal growth, together with infant and maternal characteristics, were extracted from medical birth records. Breastfeeding and formula feeding were assessed using repeated monthly questionnaires during the first year. Longitudinal changes were analyzed using linear mixed-effects models, and factors associated with ferritin concentrations were examined using Spearman correlations, linear regression models, and segmented generalized additive models.

Results: The ferritin concentration declined over time (birth: 267 ng/mL; four months: 146 ng/mL; twelve months: 30 ng/mL). Boys had lower ferritin levels than girls at all timepoints. Ferritin status at four and twelve months was positively associated with ferritin concentrations in cord blood and with gestational age. Breastfeeding and formula feeding were not associated with ferritin concentrations.

Conclusions: Infant sex, cord ferritin concentrations, and maternal ferritin concentrations were independently associated with infant ferritin concentrations across the first year of life, whereas neither breastfeeding nor formula feeding was associated with ferritin concentrations in the present analyses. Infant sex, cord ferritin, and maternal ferritin measured four months postpartum may help identify children at risk of low iron stores, with maternal ferritin potentially offering a less intrusive alternative to repeated infant sampling. However, the clinical relevance and potential use of maternal ferritin as a proxy for infant ferritin concentrations require further investigation.

Dietary habits and nutrient intake early in life are important for long-term health. Here, we examine food and nutrient intake at 1 year of age in the Swedish NICE (Nutritional impact on Immunological maturation during Childhood in relation to the Environment) birth cohort in relation to dietary guidelines and family characteristics. Dietary data was collected using a web-based semi-quantitative food frequency questionnaire (TodMeal-Q). Our findings show that intakes of critical micronutrients such as selenium, iodine, and iron were substantially below recommended levels. Also, the dietary patterns observed, characterized by higher protein and lower fat intake compared to recommendations, highlight the need for nutritional guidance to mitigate long-term health risks. Moreover, family dynamics, including the presence of siblings and maternal smoking habits, should be considered in designing effective dietary interventions, as these factors may be indicative of the context in which feeding practices are established.

Background: Early-life lead exposure affects cognitive development and emerging evidence suggests similar effects of cadmium and fluoride.

Objective: To assess the impact of gestational and childhood exposure to lead, cadmium, and fluoride on cognitive abilities and behavioral and social communication problems.

Methods: We studied 470 pregnant women (gestational week 29) and their 4-year-old children from the NICE cohort in northern Sweden. Concentrations of erythrocyte lead and cadmium and urinary cadmium were measured using inductively coupled plasma mass spectrometry and urinary fluoride with an ion-selective electrode. Urinary concentrations were specific-gravity adjusted. Associations of log2-transformed exposure concentrations with cognitive abilities (full-scale IQ and verbal comprehension by Wechsler Preschool and Primary Scale of Intelligence-Fourth Edition), behavioral problems (Child Behavior Checklist), and social communication (Social Responsiveness Scale-Second Edition) were evaluated with multivariable-adjusted linear regression analysis.

Results: Both gestational and cord erythrocyte lead concentrations were non-significantly inversely associated with child cognitive abilities (full-scale IQ: B [95%CI]: −1.2 [-2.9, 0.5] and −1.6 [-3.7, 0.4], respectively; per doubling of exposure). Similarly, both gestational and child urinary cadmium were inversely associated with cognitive abilities (full-scale IQ: −1.1 [-2.5, 0.3] and −1.1 [-2.5, 0.4], verbal comprehension: −1.2 [-3.1, 0.6] and −1.4 [-3.4, 0.6], respectively). Urinary fluoride concentrations showed no association with cognitive abilities. However, gestational fluoride was associated with increasing externalizing problems (0.9 [-0.3, 2.0]) and ADHD raw scores (0.3 [0.0, 0.6]). Childhood erythrocyte lead and urinary cadmium were non-significantly associated with increased behavioral problems (lead with total problems: 1.2 [-0.4, 2.9] and internalizing problems: 1.5 [-0.4, 3.4]; cadmium with externalizing problems: 1.1 [-0.2, 2.4]).

Conclusion: Despite non-significant associations, both lead and cadmium exposure showed consistent inverse associations with cognitive abilities at 4 years, whereas associations with behavioral problems were less conclusive, especially for cadmium. Results on fluoride indicated association with externalizing problems, including ADHD, but prevalence of behavioral problems was low, increasing uncertainty.

Introduction: Allergies and other immune-mediated diseases are thought to result from incomplete maturation of the immune system early in life. We previously showed that infants’ metabolites at birth were associated with immune cell subtypes during infancy. The placenta supplies the fetus with nutrients, but may also provide immune maturation signals.

Objectives: To examine the relationship between metabolites in placental villous tissue and immune maturation during the first year of life and infant and maternal characteristics (gestational length, birth weight, sex, parity, maternal age, and BMI).

Methods: Untargeted metabolomics was measured using Liquid Chromatography-Mass Spectrometry. Subpopulations of T and B cells were measured using flow cytometry at birth, 48 h, one, four, and 12 months. Random forest analysis was used to link the metabolomics data with the T and B cell sub populations as well as infant and maternal characteristics.

Results: Modest associations (Q2 = 0.2–0.3) were found between the placental metabolome and kappa-deleting recombination excision circles (KREC) at birth and naïve B cells and memory T cells at 12 months. Weak associations were observed between the placental metabolome and sex and parity. Still, most metabolite features of interest were of low intensity compared to associations previously found in cord blood, suggesting that underlying metabolites were not of placental origin.

Conclusion: Our results indicate that metabolomic measurements of the placenta may not effectively recognize metabolites important for immune maturation.

Circulating food metabolites could improve dietary assessments by complementing traditional methods. Here, biomarker candidates of food intake were identified in plasma samples from pregnancy (gestational week 29, N = 579), delivery (mothers, N = 532; infants, N = 348), and four months postpartum (mothers, N = 477; breastfed infants, N = 193) and associated to food intake assessed with semi-quantitative food frequency questionnaires. Families from the Swedish birth cohort Nutritional impact on Immunological maturation during Childhood in relation to the Environment (NICE) were included. Samples were analyzed using untargeted liquid chromatography–mass spectrometry (LC-MS)-based metabolomics. Both exposure and outcome were standardized, and relationships were investigated using a linear regression analysis. The intake of fruits and berries and fruit juice were both positively related to proline betaine levels during pregnancy (fruits and berries, β = 0.23, FDR < 0.001; fruit juice, β = 0.27, FDR < 0.001), at delivery (fruit juice, infants: β = 0.19, FDR = 0.028), and postpartum (fruits and berries, mothers: β = 0.27, FDR < 0.001, infants: β = 0.29, FDR < 0.001; fruit juice, mothers: β = 0.37, FDR < 0.001). Lutein levels were positively related to vegetable intake during pregnancy (β = 0.23, FDR < 0.001) and delivery (mothers: β = 0.24, FDR < 0.001; newborns: β = 0.18, FDR = 0.014) and CMPF with fatty fish intake postpartum (mothers: β = 0.20, FDR < 0.001). No clear relationships were observed with the expected food sources of the remaining metabolites (acetylcarnitine, choline, indole-3-lactic acid, pipecolic acid). Our study suggests that plasma lutein could be useful as a more general food group intake biomarker for vegetables and fruits during pregnancy and delivery. Also, our results suggest the application of proline betaine as an intake biomarker of citrus fruit during gestation and lactation.

Our previous studies have linked gestational methylmercury exposure, originating from seafood, to changes in maternal thyroid hormones and infant birth size in a Swedish birth cohort. Herein we aimed to determine associations between maternal thyroid hormones and infant birth size and elucidate if maternal hormones could mediate the relationship between methylmercury and lower birth size. In 515 women, without known thyroid disease, we assessed metal exposure by erythrocyte mercury concentrations (mainly methylmercury, reflecting exposure over the past months) in early third trimester measured with inductively coupled plasma-mass spectrometry. Plasma concentrations of total and free thyroxine (tT4 and fT4) and triiodothyronine (tT3 and fT3), and thyroid-stimulating hormone (TSH) were measured at an accredited clinical laboratory. In multivariable-adjusted linear regression models, maternal tT3 (per 1 nmol/L increase) was positively associated with birth weight (B: 125 g; 95% CI 36, 214) and length (B: 0.59 cm; 95% CI 0.21, 0.97). Maternal fT4 was inversely associated with birth weight (B: − 33 g; 95% CI − 57, − 9.5), driven by obese women (n = 76). Causal mediation analyses suggested that a doubling of erythrocyte mercury (> 1 µg/kg; n = 374) was associated with a mean tT3-mediated decrease in birth weight of 11 g (95% CI − 25, − 1.6) and in birth length of 0.1 cm (95% CI − 0.12, − 0.01), both equivalent to about 12% of the total effect. To conclude, tT3 was positively associated with infant birth size. Reduced tT3 levels appeared to mediate a minor part of the inverse association between methylmercury exposure and birth size.

Background: Short-chain fatty acids (SCFAs) in intestinal contents may influence immune function, while less is known about SCFAs in blood plasma. The aims were to investigate the relation between infants’ and maternal plasma SCFAs, as well as SCFAs in mother's milk, and relate SCFA concentrations in infant plasma to subsequent sensitisation and atopic disease.

Methods: Infant plasma (N = 148) and corresponding mother's milk and plasma were collected four months postpartum. Nine SCFA (formic, acetic, propionic, isobutyric, butyric, succinic, valeric, isovaleric, and caproic acid) were analysed by UPLC-MS. At 12 months of age, atopic disease was diagnosed by a pediatric allergologist, and sensitisation was measured by skin prick test. All families participated in the Swedish birth cohort NICE (Nutritional impact on Immunological maturation during Childhood in relation to the Environment).

Findings: Infants with sensitisation, atopic eczema, or food allergy had significantly lower concentrations of five, three, and two SCFAs, respectively, in plasma at four months. Logistic regressions models showed significant negative associations between formic, succinic, and caproic acid and sensitisation [ORadj (95% CI) per SD: 0.41 (0.19–0.91); 0.19 (0.05–0.75); 0.25 (0.09–0.66)], and between acetic acid and atopic eczema [0.42 (0.18–0.95)], after adjusting for maternal allergy. Infants’ and maternal plasma SCFA concentrations correlated strongly, while milk SCFA concentrations were unrelated to both. Butyric and caproic acid concentrations were enriched around 100-fold, and iso-butyric and valeric acid around 3-5-fold in mother's milk, while other SCFAs were less prevalent in milk than in plasma.

Interpretation: Butyric and caproic acid might be actively transported into breast milk to meet the needs of the infant, although mechanistic studies are needed to confirm this. The negative associations between certain SCFAs on sensitisation and atopic disease adds to prior evidence regarding their immunoregulatory potential.

Funding: Swedish Research Council (Nr. 2013-3145 and 2019-0137 to A-S.S.), Swedish Research Council for Health, Working Life and Welfare FORTE, Nr 2018-00485 to A.W.), The Swedish Asthma and Allergy Association's Research Fund (2020-0020 to A.S.).

Introduction: Placental function is essential for fetal development, but it may be susceptible to malnutrition and environmental stressors.

Objective: To assess the impact of toxic and essential trace elements in placenta on placental function.

Methods: Toxic metals (cadmium, lead, mercury, cobalt) and essential elements (copper, manganese, zinc, selenium) were measured in placenta of 406 pregnant women in northern Sweden using ICP-MS. Placental weight and birth weight were obtained from hospital records and fetoplacental weight ratio was used to estimate placental efficiency. Placental relative telomere length (TL) and mitochondrial DNA copy number (mtDNAcn) were determined by quantitative PCR (n = 285). Single exposure-outcome associations were evaluated using linear or spline regression, and joint associations and interactions with Bayesian kernel machine regression (BKMR), all adjusted for sex, maternal smoking, and age or BMI.

Results: Median cadmium, mercury, lead, cobalt, copper, manganese, zinc, and selenium concentrations in placenta were 3.2, 1.8, 4.3, 2.3, 1058, 66, 10626, and 166 μg/kg, respectively. In the adjusted regression, selenium (>147 μg/kg) was inversely associated with placental weight (B: −158; 95 % CI: −246, −71, per doubling), as was lead at low selenium (B: −23.6; 95 % CI: −43.2, −4.0, per doubling). Manganese was positively associated with placental weight (B: 41; 95 % CI: 5.9, 77, per doubling) and inversely associated with placental efficiency (B: −0.01; 95 % CI: −0.019, −0.004, per doubling). Cobalt was inversely associated with mtDNAcn (B: −11; 95 % CI: −20, −0.018, per doubling), whereas all essential elements were positively associated with mtDNAcn, individually and joint.

Conclusion: Among the toxic metals, lead appeared to negatively impact placental weight and cobalt decreased placental mtDNAcn. Joint essential element concentrations increased placental mtDNAcn. Manganese also appeared to increase placental weight, but not birth weight. The inverse association of selenium with placental weight may reflect increased transport of selenium to the fetus in late gestation.

Studies have indicated that early-life exposure to toxic metals and fluoride affects the immune system, but evidence regarding their role in allergic disease development is scarce. We aimed to evaluate the relations of exposure to such compounds in 482 pregnant women and their infants (4 months of age) with food allergy and atopic eczema diagnosed by a paediatric allergologist at 1 year of age within the Swedish birth-cohort NICE (Nutritional impact on Immunological maturation during Childhood in relation to the Environment). Urinary cadmium and erythrocyte cadmium, lead, and mercury concentrations were measured by inductively coupled plasma mass spectrometry (ICP-MS), urinary inorganic arsenic metabolites by ICP-MS after separation by ion exchange chromatography, and urinary fluoride by an ion-selective electrode. The prevalence of food allergy and atopic eczema was 8 and 7%, respectively. Gestational urinary cadmium, reflecting chronic exposure, was associated with increased odds of infant food allergy (OR [95% CI]: 1.34 [1.09, 1.66] per IQR [0.08 μg/L]). Both gestational and infant urinary fluoride were associated, albeit at a statistically non-significant level, with increased atopic eczema odds (1.48 [0.98, 2.25], 1.36 [0.95, 1.95], per doubling, respectively). By contrast, gestational and infant erythrocyte lead was associated with decreased odds of atopic eczema (0.48 [0.26, 0.87] per IQR [6.6 μg/kg] and 0.38 [0.16, 0.91] per IQR [5.94 μg/kg], respectively), and infant lead with decreased odds of food allergy (0.39 [0.16, 0.93] per IQR [5.94 μg/kg]). Multivariable adjustment had marginal impact on the estimates above. After additional adjustment for fish intake biomarkers, the methylmercury associated atopic-eczema odds were considerably increased (1.29 [0.80, 2.06] per IQR [1.36 μg/kg]). In conclusion, our results indicate that gestational cadmium exposure might be associated with food allergy at 1 year of age and, possibly, early-life exposure to fluoride with atopic eczema. Further prospective and mechanistic studies are needed to establish causality.