Day-to-Day Fluctuation in Micronutrient Content in Human Milk Relative to Maternal Diet

1. Introduction

Human milk (HM) is a dynamic fluid that changes depending on multiple factors, with maternal diet being one of the most relevant [1,2,3]. However, the effect of maternal diet on the composition of HM is not the same for all nutrients and in all circumstances [3,4].

A recently published systematic review [5] found strong evidence for the influence of maternal intake of docosahexaenoic acid (DHA); eicosapentaenoic acid; vitamins A, E, and K; iodine; and selenium on the concentration of these nutrients in HM. Weaker evidence of the impact of maternal intake of linolenic acid, group B vitamins, and vitamins C and D on their concentration in milk was found in the same systematic review [5]. However, it is notable that the investigation of micronutrients is much more limited when compared to that of fatty acids [5,6]. In addition, most of the studies evaluating micronutrients, as observed by the above-mentioned systematic review, were experimental studies, rather than observational studies, and mostly with doses of supplements higher than the recommended daily intakes [5]. In this regard, it is likely that in healthy women, the levels of some nutrients in HM, such as fat-soluble vitamins and vitamin C, respond only to high doses of supplementation and not to lower doses of supplementation or to variations within the habitual diet [4,5,7].

To our knowledge, the day-to-day changes in milk micronutrient composition attributed to daily fluctuations in usual diet and/or micronutrient intake in lactating women under non-experimental conditions and at a group level rather than individually have not yet been evaluated yet. However, this approach could be useful for better understanding the behavior of HM in response to dietary variations under real-life conditions. Such information could provide valuable support for establishing dietary recommendations for lactating women at the population level or in specific settings such as human milk banks.

Hence, the objective of this study was to evaluate the association between variations in maternal dietary and supplementation intake and variations in HM micronutrient composition over four consecutive days.

2. Materials and Methods

2.1. Study Design and Participants

This observational longitudinal study was carried out at the Aladina MGU Regional Human Milk Bank (RHMB) at the “12 de Octubre” University Hospital in Madrid, Spain. A group of lactating mothers was recruited from August 2017 to February 2020, comprising (1) HM donors from the RHMB, (2) lactating mothers of very preterm infants (i.e., ≤32 weeks of gestational age (GA)-admitted at the time of the study to the Neonatology Service of the "12 de Octubre" University Hospital), and (3) lactating mothers with vegetarian or vegan diet.

The lactation time had to be at least 3 weeks, so that all of the milk samples analyzed corresponded to mature milk.

Exclusion criteria were malabsorptive intestinal disease, the use of drugs that could modify appetite, nutrient absorption or metabolism (antineoplastic drugs, anorectics, anabolic agents, diuretics, etc.), and idiomatic barrier.

The study was conducted in accordance with the Declaration of Helsinki and was approved by the Clinical Research Ethics Committee of the "12 de Octubre" University Hospital (protocol code 15/269). All participants gave written informed consent.

2.2. Study Protocol

At the beginning of the study, socio-demographic, health and anthropometric data were collected from the participants and their breastfed children. Thee recruited lactating women were also weighed and measured, and their body mass indexes (BMIs) were calculated (electronic medical scale with BMI function Model 799 with measuring rod Model 220, CE approved—Class III, brand Seca®, Hamburg, Germany). Subsequently, they had to fill in a weighted dietary record for 5 consecutive days, including a holiday. Food intake and micronutrient supplementation were carefully reported. Detailed information of the recording method is described elsewhere [8]. Simultaneously, they collected milk for four consecutive days (from the second to the fifth day of the dietary record) (Figure 1).

Milk donors and lactating mothers with a vegetarian or vegan diet followed the same milk collection protocol: both groups of women collected at least one milk sample per day, without changing their milk expression routine and, therefore, without following a fixed schedule. The only requirement was to collect a complete milk expression. From each milk expression, after gently shaking, they collected 25 mL of milk with a sterile syringe, placing it in a sterile transparent glass bottle. The identification code as well as the date, time, and volume of the milk expression was labeled in the bottle before the sample was frozen to -20 °C. Mothers of very preterm infants admitted to our neonatal unit performed a 24-hour milk expression. Therefore, they collected a sample from each of the milk extractions carried out during the day to obtain a total volume of approximately 25 mL for each day. Otherwise, the procedure was the same as described above.

The five-day dietary record and milk samples were submitted to the RHMB within 15 days of study completion. The milk samples were transported to the RHMB in a portable cooler with cold accumulators. At RHMB, milk was thawed and, if available, milk samples from the same day were mixed, while milk from different days was not mixed, as the HM micronutrient composition of each day was analyzed individually. Twenty aliquots of 1 mL of the milk from each day were obtained. Each milk aliquot was labeled with the identification code and the day of the study to which it belonged. Subsequently, milk aliquots were kept frozen at -80 °C in the RHMB until they were sent for analysis.

2.3. Dietary Study

The weighed five-day dietary record was analyzed using DIAL-Software^® (DIAL.EXE version 3, February 2014, Alce Ingeniería, Madrid, Spain), and the nutrient intake for each day as well as the daily intake of each of the following food groups were obtained: (1) fruit; (2) vegetables and greens; (3) dairy; (4) grains, legumes, and nuts; and (5) eggs, meat, and fish. The daily intake from pharmacological supplements of each of the micronutrients investigated was also determined.

2.4. Micronutrient Analysis in HM

Micronutrient analyses in milk were performed by the NUTREN-Nutrigenomics Group of the Department of Experimental Medicine at the University of Lleida, Spain.

The following vitamins and minerals were analyzed in milk: free thiamine, free riboflavin, nicotinamide, pantothenic acid, pyridoxal, folic acid, cobalamin, ascorbic acid, dehydroascorbic acid, retinol, vitamin D₃, 25(OH)D₃, α-tocopherol, γ-tocopherol, iodine, calcium, phosphorous, and selenium. Table 1 shows the analytical techniques employed to determine each of the micronutrients studied. A full description of the laboratory studies is available in a previous manuscript [9].

2.5. Statistics

Statistical analysis was performed with the STATA 14 program.

For the descriptive analysis, qualitative variables were shown as absolute and relative frequencies and quantitative variables as medians and interquartile range or as means and the standard deviation depending on their distribution. Normality was determined by applying the Shapiro–Wilk normality test.

Longitudinal analyses were conducted to examine the possible association between daily changes in micronutrient levels in HM over a span of 4 days and the day-to-day intake of the equivalent micronutrients and supplements as well as different food groups. Hence, generalized estimating equations (GEE) models were employed using the vitamin and mineral concentration in HM as a dependent variable and the daily intake levels of different food groups, micronutrients, and supplements as independent variables. In this sense, the obtained results will inform about how the day-to-day fluctuations in the average micronutrient content in the HM of the investigated lactating mothers are related to the day-to-day fluctuations in their average micronutrient, supplement, and food group intake. A p-value of < 0.05 was considered statistically significant.

3. Results

A total of 146 lactating women were recruited, of whom 3 dropped out of the study early, resulting in 143 women completing the study. Of the 143 women who completed the study, 113 were milk donors, 15 were mothers of very preterm infants admitted to the neonatal service, and 20 were vegetarians or vegans. The groups were not mutually exclusive. Thus, three of the donors were also mothers of very preterm infants admitted to the neonatal service, and another two were also ovo-lacto-vegetarians. In total, the study included 123 omnivores, 9 vegetarians (5 ovo-vegetarians and 4 ovo-lacto-vegetarians), and 11 vegans. Regarding the breastfed infants, at the time of the study, there were 17 preterm infants not admitted to the neonatal service (gestational age from 22+6 weeks to 36+4 weeks; from 38+5 weeks of postmenstrual age to 19.2 months of corrected age) and 15 very preterm infants admitted at the time of the study to the neonatal service (gestational age range from 23+5 weeks to 30+3 weeks; postmenstrual age range from 27+4 weeks to 41+1 weeks).

Figure 2 shows the distribution of the participants by group.

3.1. Characteristics of the Studied Population

Results from the socio-demographic, health, and anthropometric data are presented in Table 2, Table 3, Table 4, Table 5 and Table 6. Furthermore, Table 7 shows the average vitamin and mineral concentration in human milk of lactating women collected during four successive days. In terms of nutrient supplementation reported in the socio-demographic and health survey, 141 of 143 lactating women (98.6%) declared using pharmacological micronutrient supplements during pregnancy, and 129 of 143 (90.2%) during lactation. Most participants took folic acid (n = 138 (96.5%)), vitamin B12 (n = 135 (94.4%)), and iodine (n = 136 (95,1%)) supplements during pregnancy.

3.1. Generalized Estimating Equations

Table 8 shows the association between day-to-day fluctuations in the average micronutrient content in HM across the entire sample of lactating women and the day-to-day fluctuations in their average food and supplement intake according to the dietary record.

For some of the micronutrients studied, the GEE analyses revealed some positive associations between lactating mothers’ intake and their concentration in milk over the days studied.

Regarding vitamins, (1) for each daily serving of meat, fish, and eggs, total vitamin D₃ levels in HM increased by 243.2 pg/mL (p=0.027); (2) for every 1 mg/day of riboflavin supplementation, free riboflavin levels in HM increased by 28.6 mcg/L (p=0.019); and (3) regarding vitamin B6, pyridoxal levels in HM increased by 6.3 mcg/L per 1 mg/day of vitamin B6 supplementation (p=0.001), and by 2 mcg/L per daily fruit ration (p=0.016), according to the adjusted coefficients of the model.

Concerning minerals, for each daily serving of meat, fish, and eggs, the daily selenium levels in HM increased by approximately 0.16 ppb (p=0.031).

No associations were found for the rest of the micronutrients studied (free thiamine, nicotinamide, pantothenic acid, folic acid, cobalamin, ascorbic acid, dehydroascorbic acid, retinol, α-tocopherol, γ-tocopherol, iodine, calcium, and phosphorous).

4. Discussion

To the best of our knowledge, this is the first study to longitudinally assess the effect of variations in the habitual diet and supplementation on daily variations in milk micronutrients in a sample of lactating women with different dietary patterns, gestational ages, and a wide range of lactation durations. Compared to other investigations regarding changes over time in HM micronutrient concentration, this was an observational study conducted under real-life conditions. That is, no intervention or modification of diet or regular intake of micronutrient supplements was made. A detailed study of the diet was carried out through a weighted dietary record compiled for 5 days in a row. Concomitantly, women expressed milk for 4 consecutive days, and the micronutrient content was analyzed on each of the days studied.

Under these conditions, we found that the usual day-to-day variations in the diet or the intake of commonly used supplements of lactating women had an impact on the fluctuation in the concentration of some micronutrients in their milk over the 4 days studied.

A positive association between the intake of daily servings of meat, fish, and eggs and the concentration of total vitamin D and selenium in milk was found. Likewise, the intake of vitamin B2 and B6 supplements contained in multi-nutrient supplements designed for lactating women in Spain was related to an increase in the content of free riboflavin and pyridoxal in milk. In addition, a higher daily intake of fruit was associated with an increased pyridoxal content in milk.

This should encourage medical professionals to emphasize the importance of women's nutrition during lactation; at the level of human milk banks, this knowledge may lead to the implementation of measures to optimize donors’ intake of micronutrients in order to increase the quality of donor HM. We should be aware that the diet of lactating mothers is a modifiable factor that can be improved with simple health advice.

The associations found are consistent with the limited knowledge available from studies with a cross-sectional design as well as with a longitudinal design but with an experimental design. In the literature, vitamin B2 and B6 concentrations in HM are strongly associated with maternal intake [2,3,4,30]. Concerning selenium, the levels of this mineral are usually high in eggs, meat, poultry, fish, and seafood [31]. Indeed, a previous study conducted in Italy [32] found a positive association between mother’s egg intake during gestation and fresh fish intake during lactation and the selenium content in HM. A recent systematic review concluded that there is a strong response of selenium content in milk to maternal selenium intake [5]. A transversal study in Galicia, a region of Spain, showed a positive correlation between adherence to the Mediterranean diet of lactating women (compared to the Atlantic diet) and the concentration of selenium in their milk [33]. Regarding vitamin D, there is some evidence of the response of vitamin D in HM to the mother's diet [4,5]. Eggs, meat, and fish are also good dietary sources of vitamin D, especially eggs, tuna, herring, and sardines [34]. However, concerns about the mercury content of fish can lead to restrictive fish consumption behaviors by pregnant and lactating women. Nevertheless, in a previous study, no significant relationship was found between mercury levels in HM and maternal consumption of any category of fish. Interestingly, the multiple linear regression models showed a negative association between maternal intake of freshwater fish and mercury levels in HM [35]. Small oily fish, such as salmon, sardines, mackerel, trout, anchovy, and herring, are valuable foods for breastfeeding mothers as they are a good source of selenium, vitamin D, and also omega-3, while being lower in mercury than other species [36,37].

On the other hand, we were not able to demonstrate the effect of habitual changes in mothers' diet on the content in the milk of other micronutrients such as vitamin A, E, B1, C, and iodine. However, the concentration of all these nutrients in HM has been shown to be influenced by dietary or supplemental intake in cross-sectional and/or longitudinal experimental studies [3,5,7,8,38]. One possible explanation is that small daily variations in the usual maternal diet or low-dose supplementation of these nutrients are probably not sufficient to influence daily milk concentration in well-nourished women. In fact, in a systematic review, higher doses of supplementation showed greater effects [7]. However, we cannot rule out that we were not able to detect associations that truly exist, as we did not control for possible confounding factors such as circadian rhythm, or the time elapsed between feeding and expression of milk. In a study conducted by Hampel et al. [39], afternoon and evening expressions provided the most representative milk samples for the micronutrient study, and acute effects of supplementation were observed for vitamins B1, B2, B6, and A at 2-4 h after administration. Nevertheless, in our study, we did not determine a schedule for milk sampling, but the women extracted their milk at the time that suited them best. It is also feasible that a larger sample size would have yielded significant results for some of the micronutrients not associated with diet in the present study.

5. Conclusions

Observational longitudinal analysis of successive days allows the quantification of the impact of habitual variations in diet and/or common supplement intake on day-to-day HM vitamin D, free riboflavin, pyridoxal, and selenium variations in a population of lactating women. This should encourage medical professionals to emphasize the importance of women's nutrition during lactation and promote measures to improve their nutrient intake.