Abstract
Introduction:
Donor human milk is increasingly being utilized for both preterm and term infants when mother’s milk is unavailable. With the rising prevalence of maternal overweight and obesity, it is crucial to evaluate the relationship between maternal body mass index and the fat and energy content of donor human milk.
Objectives:
To assess the impact of maternal body mass index on human milk fat content.
Methods:
A cross-sectional study was carried out using retrospective data from women who made their first human milk donation at ≥15 days postpartum at a human milk bank in São Paulo, Brazil, from January 2018 to December 2020. Data of sociodemographic, obstetric, health, and anthropometric measures were collected by the human milk bank staff. Milk fat and energy content were determined using the crematocrit test. Analysis of variance and multiple linear regression were used to compare means of crematocrit and fat. We determined the p-values using a test of heterogeneity and linear trend and presented the one with the lower p-value.
Results:
Most donors were between 25 and 35 years old, had higher education, were employed, and lived with a partner. At the time of milk donation, 40.9% of women were overweight or obese. The fat (1.09 g/dL) and energy (9.83 kcal/dL) content of human milk were higher in obese donor compared with eutrophic donors.
Conclusions:
The fat and energy content of human milk were associated with maternal body mass index, suggesting the potential value for selective use of high fat and high calorie donor milk for very low birthweight or premature infants.
Introduction
Breastfeeding offers significant short-term benefits for infant survival, including a reduction in mortality from infectious diseases such as diarrhea and respiratory infection. 1 In the long term, breastfeeding may decrease the risk of asthma, type 2 diabetes, 2 and obesity.1,3 Additionally, it provides benefits for women by lowering the risk of developing breast cancer and type 2 diabetes.1,4
In Brazil, the human milk banks (HMB) were established as a public health initiative, 5 to promote and protect breastfeeding practices among premature infants. During birth hospitalization, low birth weight and premature infants who cannot receive their own mother’s milk are given pasteurized human milk (HM) from donors as an alternative. 6 This practice is associated with a lower incidence of infections and necrotizing enterocolitis, reduced likelihood of sepsis, 7 lower rates of bronchopulmonary dysplasia, 8 shorter durations of ventilatory support, and decreased cardiac complications and mortality. 9
Preterm infants require higher caloric and protein intake to support accelerated growth. 10 Studies have shown that HM from women who deliver preterm infants generally contains higher protein concentrations. 11 When a birth mother’s milk is unavailable, donor HM (DHM) is often recommended over formula feeding due to its recognized benefits. 12
DHM is undoubtedly a valuable resource for infants who cannot receive their mother’s milk. 12 However, DHM composition can vary due to several maternal factors. Immediately after birth, colostrum is rich in antibodies and proteins and has a lower fat content. 13 Maternal diet and bodyweight status significantly influence HM composition. 14 For example, a diet rich in omega-3 fatty acids can increase the levels of these fatty acids in breast milk. 15 Furthermore, the composition of HM, particularly its fat content, can change throughout the lactation period and even during a single feeding session. 16 Additionally, some medications may be transferred to breast milk and impact its composition. 17
HM from women delivering preterm infants may have an altered composition. In comparison to term milk, preterm milk has a notably higher fat and energy content during the first 2 weeks but contains less fat and energy from weeks 3 to 8.18,19 Some studies indicate that protein levels in the HM of women who deliver preterm infants can decrease by up to 50% within 10–12 weeks postnatal. Conversely, other studies suggest that HM from women delivering preterm initially has higher protein concentrations.
With the increasing prevalence of maternal obesity prior to and during pregnancy, it is becoming more common for HM donors to be overweight or obese. 20 Therefore, it is essential to evaluate whether a donor’s bodyweight status correlates with the fat content of their milk. The fat composition of HM plays a critical role in various aspects of infant heath, including long-term metabolic health, immune function, and gut health.21,22 Given the potential variations in HM composition among donors, monitoring fat levels in DHM is crucial to insure optimal infant growth and development. 23
Therefore, this study aimed to investigate the impact of maternal body mass index (BMI) on DHM fat content. Understanding this relationship is essential for gaining a comprehensive insight into how maternal BMI affects the nutritional variation of DHM and potentially aid in directing specific milk sources to infants at need of augmented caloric intake.
Methods
Study design
A cross-sectional study was carried out using data from women who donated HM to HMB in a Breastfeeding Center in São Paulo, Brazil. This HMB has been operational since 2003 and processes HM weekly from ∼40 donors per month, totalizing around 400 L per year. Donors typically pump or hand express HM during the day and store it in 120–200 mL containers. Frozen HM is transported in a dedicated −1°C cooler with additional ice packs to the HMB, ensuring transfer within 6 hours.
The HMB adheres to Brazilian guidelines, which recommend storing HM frozen at −20°C. Donated HM containers are kept in a −20°C freezer at the HMB and are processed following days, with a maximum storage period of up to 15 days from the time of pumping or expressing.
Breastfeeding women who made their first HM donation at least 15 days postpartum (indicating mature milk) between 2018 and 2020 were eligible for inclusion in this study. Exclusion criteria included mothers of preterm babies (<37 weeks gestation), those classified as underweight (BMI <18.5 kg/m2), or those for whom postpartum anthropometric data were unavailable. In this study, we used HM samples from the first donation when the women were enrolled as donors.
Data collection
Data on sociodemographic variables (age, education level, marital status, and employment), obstetric factors (parity, type of delivery, gestational age at delivery, and days postpartum), maternal anthropometry (prepregnancy and postpartum BMI), and HM crematocrit (%) were collected from the HMB database as part of routine service operations. Maternal prepregnancy and postpartum weights, as well as height, were retrieved from prenatal medical records, and BMI was calculated in kg/m2. Postpartum BMI was recorded on the first day of donation. Donors were classified as eutrophic (BMI ≥18.5 and <25.0 kg/cm2), overweight (≥25 and <30 kg/cm2), or obese (BMI ≥30 kg/cm2). 24
HM fat (grams per deciliter) and energy (kilocalories/dL) contents were estimated using the crematocrit (%) test, following the technical standards of the Global Network of HMB. 5 This test estimates fat and energy contents by centrifuging 75 µL of HM samples for 15 minutes in a microhematocrit capillary tube to separate cream from aqueous fraction.
The HM sample used for the crematocrit test was taken from homogenized HM prior to the pasteurization process. Briefly, DHM was defrosted and homogenized, and a 10 mL sample was collected using a sterile syringe from the middle of each container. The collected HM samples were placed in glass test tube and reheated in a water bath at 40°C for 15 minutes. After reheating, the crematocrit test was conducted at a temperature of 20°C.
The crematocrit (%) value was obtained by measuring the length of the cream column relative to the total HM column length in three individual tubes using a millimeter rule. 5 The fat concentration is linearly correlated with the crematocrit (%) of HM and was estimated using the formula: Fat (g/dL) = (crematocrit % − 0.59)/1.46.5,25 The formula used for energy calculation was: kcal/dL = 6.20 × crematocrit (%) + 35.1 (frozen milk).26,27
Statistics
Analyses were carried out using Stata version 15.0, and ANOVA was used to compare the means of crematocrit, fat, and energy according to categories of education level, maternal prepregnancy BMI, maternal postpartum BMI, maternal age, parity, type of delivery, and infant birthweight. Multiple linear regression was used to compare crematocrit, fat, and energy estimates according to maternal postpartum BMI categories, adjusting for maternal age, parity, and infant birthweight. For ordinal variables, we estimated the p-value for the difference across the categories and for linear trend.
Ethical statement
The study was approved by the Research Ethics Committee under number 4.201.145, in accordance with norms and guidelines outlined in Resolution No. 466/2012 of the National Health Council.
Results
In this study, we identified 475 donors who met the criteria for enrollment. Of these, 277 were excluded for reasons detailed in Figure 1, resulting in 198 women included in the analysis.

Flowchart of sample selection.
Table 1 shows that most women were aged between 25 and 35 years (62.1%), had higher education (85.5%), lived with a partner (91.2%), and had paid employment (93.2%). Additionally, most donors were primiparous (66%) and the majority had undergone cesarean delivery (58.6%). Donors averaged 39 (±1.2) weeks of gestation at delivery and were 125 days postpartum (range 16–610 days) at the time of donation. The prevalence of maternal overweight increased from 19.9% in the pregestational period to 31.3% in the postpartum period.
Sociodemographic, Obstetric Characteristics, and Body Weight Status of Donors
Number of samples may vary because of missing values.
BMI, body mass index.
Maternal postpartum BMI, but not prepregnancy BMI, was significantly correlated with HM crematocrit, fat, and energy contents. In contrast, maternal age, parity, and birth weight did not show a clear pattern of association with HM fat content (see Table 2).
Crematocrit, Fat, and Energy Data According to Maternal Postpartum BMI at First Human Milk Donation Adjusted for Maternal Age, Parity, and Infant Birthweight
L, test for linear trend.
C, test for categorical.
BMI, body mass index; CI, confidence interval.
Table 3 illustrates that, even after controlling for confounding variables (e.g., maternal age, parity, and birth weight), obese donors exhibited significantly higher crematocrit (1.59%), fat content (1.09 g/dL), and energy content (9.83 kcal/dL) compared with eutrophic donors. Additional analysis revealed that each 1 kg/m2 increase in BMI was associated with a 0.62 kcal/dL increase in HM energy.
Human Milk Crematocrit, Fat, and Energy Contents According to Maternal Postpartum Body Mass Index, Adjusted for Maternal Age, Parity, and Infant Birthweight
L, test for linear trend.
C, test for categorical.
BMI, body mass index; CI, confidence interval; R, regression coefficient.
Discussion
Our findings indicate that fat and energy contents of HM are higher among obese donors. Notably, >40% of the mothers included in this study were classified as overweight or obese during the postpartum period. This prevalence aligns with other Brazilian participants studies. 28 The increase in the prevalence of overweight and obesity within the first 4 months postpartum, despite ongoing breastfeeding, highlights concern about the impact of pregnancy weight gain on long-term overweight and obesity. Interventions aimed at preventing excessive weight gain during pregnancy and facilitating postpartum weight normalization may be crucial in mitigating the risk of overweight and obesity associated with pregnancy.29,30
HM contains >200 identified components, with considerable individual variability in caloric content.31,32 The primary contributors to milk caloric content are lactose and fat. Among the macronutrients, milk fat exhibits significant variability, while the content of milk protein and carbohydrates remains relatively stable. 33 Systematic reviews and meta-analyses have observed a positive association between maternal adiposity or BMI with HM fat concentration across various lactation periods, particularly in mature milk.34,35 In addition to BMI, HM fat content energy is influenced by maternal diet, breastfeeding frequency, lactation stage beyond 16 weeks postpartum, 36 and milk volume produced. 37
Recent studies, including our own, suggest that the higher milk fat content in overweight or obese breastfeeding women may be due in part to increased serum triglycerides and insulin levels. 38 Furthermore, the number of milk fat globules increased as the mammary lobe empties progressively. 39
DHM represents a selective population of postpartum women who may have unique motivations, breastfeeding histories, milk productivity, and diets. 40 These factors can influence the composition of DHM. Our results regarding DHM fat content align with previous reports studies, indicating that DHM samples can exhibit considerable variability. A recent systematic review reported that the mean fat content of DHM ranges from 1.8 to 4.1 g/dL, 41 and this fat content is positively associated with both maternal BMI and the volume of DHM. The review also highlighted the potential for up to a two-fold difference in the fat, protein, and energy composition of DHM, 42 raising concerns about both excessive nutrients and deficiencies.
High caloric intake is necessary for the growth of premature infants,42,43 with feeding goals of at least 110–135 kcal/kg/day to match the growth of term infants.44,45 Cooper et al. 46 noted significant implications for the growth of premature infants based on differences in energy intake between nutrient-poor (3rd percentile) and nutrient-rich (75th percentile) DHM (88 kcal/kg/day). Notably, DHM analysis is often performed on a sample of the provided HM volume, which may represent a mix of foremilk and hindmilk. As fat concentration increases approximately four-fold from foremilk to hindmilk, 39 composite milk samples may obscure individual differences.
The use of high-fat and high-energy DHM may support weight gain in very low birthweight, especially if the birth mother’s milk is unavailable. Conversely, large gestational age infants might benefit from DHM with lower caloric content. These findings suggest that a targeted DHM program could enhance nutritional efficacy based on milk composition differences.
Among the study’s limitations, we did not evaluate the donors’ diet, so we could not evaluate whether diet mediates the association between maternal adiposity and HM fat and energy. 47 Additionally, 58.3% of the women were excluded, primarily due to incomplete information or failure to meet the criteria for mature milk (see Fig. 1), though this is unlikely to affect the association between BMI and HM fat and energy contents. Due to the small number of underweight donors in our HMB, we have decided not to include this specific bodyweight group in the study. Importantly, the timing and volume of the milk samples (∼200 mL per container) were not assessed. As HM fat content increases markedly from foremilk to hindmilk, this may obscure the relationship between BMI and fat content.
Conclusions
The fat and energy content of HM were associated with postpartum maternal BMI, suggesting potential value for very low birthweight or premature infants.
HMB might consider assessing the composition of DHM and potentially implementing a program of “personalized” milk caloric content to benefit infants requiring higher calorie intake, such as premature infants. However, this was a cross-sectional retrospective study, highlighting the need for additional prospective, rigorously controlled studies. Future research should address factors such as lactational stage, timing of milk collection, storage conditions, and maternal influences (e.g., preterm versus term birth, medications) to better understand the variability in HM components, including protein, lactose, immunoglobulins, and oligosaccharides.
Footnotes
Authors’ Contributions
L.S.d.C. and K.P.C.: were responsible for the conceptualization of the study. L.S.d.C., A.C.L.R., R.d.F.P., and K.P.C.: were responsible for the data collection. L.S.d.C. and B.L.H.: analyzed the data. L.S.d.C., K.P.C., B.L.H., M.D., and M.G.R.: prepared the original draft. All authors reviewed and edited the draft and last version article.
Disclosure Statement
No competing financial interests exist.
Funding Information
R.d.F.P. received financial support by the nutritionist multi-professional residence program of Universidade Federal de São Paulo (UNIFESP). No specific funding was received for this analysis. This project was supported in part by National Institutes of Health’s grant no. R01HD099813 and
