The Influence of Nutrients Intake during Pregnancy on Baby’s Birth Weight: A Systematic Review

Abstract

Background and Objective

Maternal food intake during pregnancy can substantially interfere in the baby’s anthropometric measurements at birth. Our objective was to perform a systematic review that investigate the influence of nutrient intake via food during pregnancy on the baby’s anthropometric measurements at birth.

Methods

A search was performed without time limits on CINAHL, Embase, PubMed, Scopus and Web of Science databases and manual on studies references. All nutrients and baby’s anthropometric measurements at birth were included as descriptors.

Results

Thirty articles were included, the majority of prospective cohort studies, with 15.39 (2.36) quality points (maximum value: 22). Thirty-six results of associations were found, of which 17 studies had direct associations between nutrient intake and birth outcomes. Inverse associations were identified in 8 studies and 11 articles showed no significant associations in all analyses. Maternal food intake of vitamin C, calcium and magnesium during pregnancy seems to have a positive influence on the baby's birth weight, while carbohydrates intake have an inverse association with the same outcome.

Conclusion

It is suggested that vitamin C, calcium, magnesium and carbohydrates influence on baby’s birth weight. So, these specific nutrients need more attention to the consumption, in addition to carried out new studies, with robust methodologies for measuring maternal food consumption and considering the several factors that can interfere in this assessment.

Registration

This review has been registered to the PROSPERO (International prospective register of systematic reviews) (ID: CRD42020167889).

INTRODUCTION

Anthropometric measurements at birth act on the health prognosis of the child, as birth weight and length, head circumference and anthropometric index of weight/gestational age and their classifications. Such measurements have been used as a growth indicator to predict the development of future diseases, such as type 2 diabetes and hypertension [1, 2]. Taking that into consideration, maternal food intake during pregnancy represents one of the most important modifying factors for the child's health, and it can substantially interfere in the baby’s anthropometric measurements at birth [2].

Scientific evidence points out the importance of adequate nutrition during pregnancy. However, studies show that pregnant women do not consume a sufficient amount of micronutrients. It is estimated that around 32 million pregnant women worldwide lack especially iron, 19 million in vitamin A, and millions lack folate, zinc and iodine [3]. Also, the intake of excessive calories and macronutrients during pregnancy can be as harmful as the lack of them, as it increases the possibility of inadequate gestational weight gain, followed by a higher risk of miscarriage, pre-eclampsia and gestational diabetes [4].

In this perspective, several studies [5–9] have evaluated how maternal food consumption during pregnancy can influence the anthropometric measurements of the newborn, showing contradictory results.

In a prospective cohort study conducted in Canada, Stephens, et al. [6] showed an inverse association between maternal macronutrients intake and birth weight. A cross-sectional research [7] carried out in 20 Indian hospitals with 500 pregnant women showed that the lowest protein consumption was directly associated with low birth weight (LBW)—<1500 kcal/day and <40 g/day, respectively. For micronutrients, a longitudinal study [5], carried out with 169 Australian women, pointed out that the intake of magnesium, zinc and calcium was not associated with birth weight and head circumference. On the other hand, a cross-sectional design research [7] showed that the low intake of same minerals was directly associated with LBW.

The newborn's anthropometry represents a fundamental parameter related to adequate survival, growth and development in the public health care context. A better understanding of the influence of food intake of nutrients during pregnancy on baby’s anthropometric measurements may have important implications for maternal and child health, enabling effective prenatal care interventions.

Considering the scarcity of recent reviews addressing this issue and the divergences shown in the results of the studies, the influence of maternal macro and micronutrient intake via food during pregnancy, on baby’s anthropometric measurements at birth, was systematically reviewed.

MATERIALS AND METHODS

Article eligibility

This systematic review evaluated the influence of maternal macro and micronutrients intake during pregnancy on baby’s anthropometric measurements at birth. In line with the development of written production, we used the instruments Preferred Reporting Items for Systematic Reviews and Meta-Analyses and the Cochrane Handbook for Systematic Reviews of Interventions [10, 11].

To guide the bibliographic search, the anagram PECOS (Population, Exposure, Comparator, Outcomes and Study design) [11] was used strategically to create the research question, which the subjects were pregnant women, the exposure was the intake of macro and micronutrients via food during pregnancy, the outcome was the baby's anthropometric measurements at birth: weight, length, head circumference and the classifications of the weight/gestational age index—small for gestational age (SGA, birth weight <10th percentile), appropriate for gestational age (AGA, birth weight between 10 and 90 percentile) and large for gestational age (LGA, birth weight >90 percentile) and the design was observational studies.

Search methods

This search was conducted with the assistance of a librarian and two reviewers on the bibliographic databases: Cinahl, Embase, Pubmed, Scopus, Web of Science and in the references of the main studies to identify additional works that were not indexed in those databases.

The descriptors included and text words were used in all titles, abstracts and keywords databases. The descriptors were: pregnancy, food consumption, food intake, prenatal nutrition, maternal diet, nutrients, macronutrients, micronutrients, carbohydrates, dietary fibers, proteins, dietary fats, saturated fatty acids, unsaturated fatty acids, docosahexaenoic acids, eicosapentaenoic acids, alpha-linolenic acid, omega-3 fatty acids, omega-6 fatty acids, calcium, copper, dietary iron, folic acid, dietary phosphorus, magnesium, manganese, potassium dietary, selenium, dietary sodium, zinc, vitamins A, B1, B2, B3, B5, B6, B7, B12, C, D and E, birth weight, LBW, macrosomia, weight for age, weight for gestational age, small for gestational age, appropriate for gestational age, large for gestational age, length at birth, head circumference and their respective translations and synonyms words. All the articles published until June 2020 were examined.

Selection of studies

Observational studies published in Portuguese, English or Spanish, which evaluated were considered eligible. Studies that did not specifically analyze maternal food intake of nutrients, such as those that dealt with isolated analysis of specific foods or food groups, supplementation, dietary pattern and diet quality index were excluded. Articles that investigated subject samples likely to skew the final result, such the ones composed of obese women or adolescents, or subjects who presented comorbidities during pregnancy, studies about animals and works that performed qualitative analyzes were also excluded. Conference papers, reviews, editorials and commentaries not published as an article, were desconsidered to the review. Moreover, studies that measured maternal food consumption through the analysis of biomarkers were excluded, due to the nutritional exposure [12] provided by the serum assessment.

Studies that evaluated food consumption stratifying the analyzes by period, the closest moment to delivery was chosen because maternal food intake [13] may be hindered by the frequent nausea and vomiting of the first trimester of pregnancy.

The references were exported to the EndNote Web^® program in order to organize them and eliminate duplicates. Two authors (R.C.V.S. and C.M.) read the title, abstract and keywords independently. After the initial selection, the Kappa test was performed to assess agreement between evaluators, for which the Statistical Package for the Social Sciences^® (SPSS) version 19.0 was used. The criteria of Byrt [14] were adopted to classify the result of the test. The articles with discordant evaluations (n = 24) were analyzed by a third reviewer (L.C.S.), to carry out the final analysis, which consisted of reading the full texts.

Analysis of quality and risk of bias

The quality of the selected articles was assessed by two independent reviewers (R.C.V.S and C.M.) using the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) tool [15], 22 points being the maximum score that can be achieved. Thus, according to the analysis of quality, after screening checks, studies that presented less than 50% [16, 17] of the essential items that should be described in observational studies were excluded.

Data extraction

Data extraction was performed in duplicate by two authors (R.C.V.S and C.M.) and in the case of disagreement, a consensus was reached by the third reviewer (L.C.S).

RESULTS

About 1979 articles were found, via Cinahl (n = 156), Embase (n = 26), Pubmed (n = 270), Scopus (n = 1105), Web of Science (n = 220) and manual search (n = 2). After discarding duplicates (n = 254) and reading the title, abstract and keywords, 134 studies were included for a full-text reading. There was a almost perfect inter-rater agreement (Kappa = 0.893) [14]. Subsequently, 104 articles that did not meet the eligibility or quality criteria [16, 17] were excluded. Finally, the final sample consisted of 30 articles (Fig. 1).

According to Table 1, the average and SD of the quality score of the works was 15.39 (2.36) points. It is noteworthy that the studies that reached percentages above 80% according to the STROBE tool [15], were cross-sectional and population-based cohort studies. In addition, 20 studies presented a longitudinal design [5, 6, 8, 9, 19, 21, 23, 24, 26, 29, 31, 33, 35–42] and the 10 were cross-sectional [7, 18, 20, 22, 25, 27, 28, 30, 32, 34].

Table 2 shows the main characteristics extracted from the evaluated studies, published between 1999 and 2019. Regarding the origin of the studies, 13 of them were carried out in Asia [7, 18, 21, 23, 24, 27, 29, 30, 33–35, 37, 38], 6 in Europe [9, 19, 26, 32, 39, 42], 5 in North America [6, 8, 31, 40, 41], 4 in Oceania [5, 20, 28, 36] and 1 in Africa [22] and South America [25]. The sample size ranged from 77 [39] to 91 637 women [24], with nine studies showing sample sizes greater than 1000 participants [18, 19, 21, 23, 24, 26, 30, 31, 41].

Mean maternal age ranged from 22.3 (SD 4.9) [41] to 33.5 (SD 4.4) [20] years and 11 articles did not present this data [6, 8, 18, 19, 22, 24, 27, 30, 39, 40, 42].

As for the method of assessing food consumption, five different instruments were used: food frequency questionnaire (n = 15) [5, 6, 8, 18–20, 24, 27–30, 32, 35, 37, 40], 24-h food recall (n = 13) [7, 18, 21–23, 25, 26, 30, 31, 34, 36, 37, 41], food diary (n = 3) [9, 39, 42], food record (n = 2) [33, 36] and diet history of the previous month (n = 1) [38]. Most instruments (n = 13) were applied only one time during pregnancy [5, 7, 8, 18, 20, 22, 27, 29, 32, 34, 35, 40, 42], being the third gestational trimester the most frequent (n = 19) [5–7, 9, 18–23, 27, 31, 33, 36–39, 41, 42]. Two articles did not mention when maternal food consumption data were obtained [25, 28].

To assess the maternal intake of macronutrients and its fractions, three studies analyzed the relevant percentage of the total energy value [20, 24, 28], while the others presented them through daily food consumption, in grams or micrograms (data not shown).

In all selected articles, researchers [27], nurses [5, 9, 22, 31, 34, 37] obstetricians or collaborators [8, 40] took the baby’s anthropometric measurements or this information was obtained from medical records [6, 18–21, 23–26, 28–30, 33, 35, 36, 41]. Five studies [7, 32, 38, 39, 42] did not mention the professional who took the measurements.

The main one’s variables used as controlling confounders on the studies analysis were: newborn sex, gestational age, parity, maternal age, energy intake, level of education, pre-gestational body mass index (BMI), smoking and maternal height. Six articles [6, 7, 25, 29, 33, 37] did not mention that information.

Direct associations were identified among the main results, predominantly between the intake of protein [7, 39], vitamin C [22, 42], calcium [7, 35], and magnesium [7, 36] and birth weight. In the study carried out by Cuco, et al. [39], an increase of 1 g in daily protein intake was directly associated with an increase of 7.8 g in the baby's birth weight (95% CI 0.80–14.70; p < 0.05). Additionally, in the study by Khoushabi, et al. [7], protein consumption lower than 40 g/day was associated with LBW (p < 0.001).

Regarding vitamins and minerals, Hjertholm, et al. [22] and Mathews, et al. [42] investigated the prospective effects of vitamin C intake on the birth weight (βa = 1.40; 95% CI 0.60 − 2.30; p < 0.01 and βa = 36.20; 95% CI 4.60 − 97.0; p = 0.03, respectively). In the research by Khoushabi and Saraswathi [7], observed that pregnant women who gave birth to newborns with appropriate weight had a higher consumption of calcium and magnesium than those who had babies with LBW (p = 0.001). Bawadi, et al. [35] found a direct association between calcium consumption and birth weight (βa = 0.08; p = 0.03), and similar findings were verified by another study [36].

Concerning the inverse associations, the consumption of carbohydrates [6, 36] and proteins [6, 24, 41] stand out. Watson and McDonald [36] pointed out a 14 g reduction in birth weight when the consumption of carbohydrates increased from the lowest to the highest quartile (p = 0.03) during the third gestational trimester. Stephens, et al. [6] also found an inverse correlation between birth weight and carbohydrates (r = −0.15; p < 0.05).

Morisaki, et al. [24] showed that maternal protein intake of less than or equal to 12% of the total energy value was associated with increased birth weight and a lower risk of SGA. Sloan, et al. [41] found a reduction in birth weight when protein intake was greater than 70 g/day (βa = −0.03; p = 0.01). Similarly, Stephens, et al. [6] reported an inverse correlation between maternal protein intake and birth weight (r = −0.22; p < 0.01).

It was also observed that 11 articles had no association in all analyses between nutrients intake during pregnancy and the outcomes [5, 9, 19, 26, 28, 31–33, 37, 38, 40].

Table 3 presents the systematic results of the associations (direct, inverse or without association) between macro and micronutrients intake during pregnancy and the anthropometric measurements of the newborn. In the 30 articles included, 36 associations were found. There is a predominance of direct associations (n = 17) between nutrients intake and the anthropometric measurements of the newborn. Inverse associations were found in 8 studies and 11 articles showed no significant association in all analyses (Table 3).

DISCUSSION

This study investigated observational studies that assessed the impact of maternal macro and micronutrient via food intake during pregnancy on the baby’s anthropometric measurements at birth. The results indicate that maternal intake of vitamin C, calcium, magnesium and carbohydrates, may influence on birth weight. Such results corroborate the authors' hypothesis that adequate maternal nutrients intake would result in better outcomes at birth, as demonstrated in a recently published review analysis [43].

The positive associations that lead to better results in birth weight can be explained by the essential roles of some nutrients, such as vitamin C, which acts in cellular defense against increased oxidative stress during pregnancy. Antioxidant defense systems, which are enhanced with food intake of some vitamins, as is primarily the case of vitamin C, are vital in protecting tissues and cells from damage caused by oxidative stress. An imbalance in the execution of these defense mechanisms can directly affect pregnancy outcomes, including delayed fetal growth [21]. Another hypothesis is that vitamin C acts as a co-factor in collagen biosynthesis and it benefits the development of the newborn’s cartilage and bones [21].

Magnesium has a fundamental function in physiological reactions, such as protein synthesis, in addition to being involved in several metabolic processes, such as bone formation and cellular energy development. The deficiency of this nutrient has unfavorable effects on the hematological system, reducing placental vascular flow and therefore, there are potential implications for fetal growth and development [36, 44].

The main feature of calcium metabolism during pregnancy is its active placental transport to the fetus. Complications may happen when this transport is not performed properly, such as restricted intrauterine growth, LBW, deficient bone mineralization and premature birth [45]. Studies suggest that ionized calcium is transferred from the mother to the fetus at a rate of 50 mg/day in early pregnancy to 330 mg/day in the last trimester [46], and it is directly related to the growth of the fetal skeleton, which positively impacts birth weight [35].

Carbohydrate consumption was associated with a higher risk of LBW, when in high proportions. One of the hypotheses postulated by the studies [36] is that diets with a high proportion of this macronutrient (about 71% of the total energy value) can overload the adenosine triphosphate system, increase the production of free radicals, DNA damage and thus, decrease cell division and growth. Another association studied [28] is that in addition to quantity, the quality of maternal carbohydrate intake during pregnancy can lead to high glycemic peaks, influence the supply of fetal glucose and the intrauterine growth.

Other nutrients have non-consensual associations, such as protein. One study [39] suggests that protein is the macronutrient that has the greatest effect on a child's birth weight. On the other hand, the literature demonstrates [6, 41] that protein consumption higher than recommended, results in a reduction in the placental transport of amino acids, which can result in reduced protein synthesis and anthropometric measurements at birth. Because of these controversies, the importance of adequate protein consumption must be considered, since it is essential for implantation, placental growth, angiogenesis, the transfer of nutrients from the mother to the fetus and adequate embryonic growth and development [47].

The ideal balance of macro and micronutrients to achieve appropriate infant outcomes at birth can be influenced by several variables, such as the still-unknown interaction between nutrients, different methods of measuring food consumption, and adjustments due to confounding factors. In this sense, some explanations were postulated to clarify the inconsistencies and absence of associations found in some works.

The complexity in assessing individual or population food consumption may have interfered in the results, since the dietary variability by location, different forms of food preparations, and the metabolic interaction between the nutrients must be considered [43]. It is important to highlight that the synergistic relationships and interactions that happen between macro and micronutrients are little known in the short and long term [43] until the moment of this study.

Some studies tend to focus on analyze isolated nutrients [18, 20, 21, 23, 27, 28, 30, 34, 41]. Although some nutrients are important, they are rarely reported in the literature, such as complex B vitamins, manganese and selenium [48]. It is also worthy to note the limited number of studies that used the baby's anthropometric indexes (SGA, AGA and LGA), which are clinically very important, since they are directly related to neonatal mortality [49]. The difficulty in tracking studies using these nutrients and anthropometric measurements of the baby makes new studies of systematic review and meta-analysis prioritize specific outcomes in their performance, such as birth weight [43, 50].

During the evaluation of maternal food consumption, five different instruments were used and this variation can hinder the comparison between the studies. Furthermore, the lack of methodological care related to possible confounding factors associated with nutrients, such as maternal nutritional status, gestational weight gain and use of supplementation during pregnancy, are not always considered by the primary articles and can corroborate inconsistent results [43].

Strengths and limitations should be considered when interpreting the findings of this review. The methodological heterogeneity between the studies made it impossible to perform the quantitative synthesis, however, it must be emphasized that this is the first study in which a comprehensive search of nutrients and baby’s anthropometric measurements at birth was carried out through a peer-review protocol. This protocol enabled a critical assessment of the dietary methods and methodological quality of each selected study, allowing only reliable studies to be included and minimizing the bias associated with observational studies.

CONCLUSION

The findings of this systematic review suggest that maternal intake of vitamin C, calcium, magnesium and carbohydrate may influence on birth weight. A greater understanding of the mechanisms that influence the relationship between maternal intake of nutrients and birth weight may provide a basis for the development of dietary guidelines, prioritizing the consumption of foods which have greater bioavailability of these specific micronutrients, such as milk, fruits, dark green vegetables and nuts.

In case of difficulties in reaching the nutrient amount through diet, the possibility of supplementing these micronutrients should be considered, considering the positive outcomes for maternal and child health in the short and long terms. It is also important to evaluate the quality and quantity of carbohydrate intake, for better results at birth. To provide consistent new recommendations, randomized clinical trials and population-based longitudinal studies are suggested. Such studies should evaluate food consumption at different periods before and during pregnancy and its role in all anthropometric results at birth.

Funding

The current research was supported by Fundação de Amparo a Pesquisa do Estado de Minas Gerais (grant number APQ-01782-10) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (productivity scholarship—grant number 301555/2019-2). Fundação de Amparo a Pesquisa do Estado de Minas Gerais and Conselho Nacional de Desenvolvimento Científico e Tecnológico had no role in the design, analysis or writing of this article.

References