Trial investigates how quickly human milk responds to common plant based meat swaps

By Vijay Kumar MalesuReviewed by Susha Cheriyedath, M.Sc.Nov 18 2025

A tightly controlled crossover trial shows that short-term food swaps can reshape the fats infants receive through milk, even when mothers feel no difference in appetite, metabolism, or feeding patterns.

Study: The Effect of Consuming Diets Containing Beef Compared with Plant-Based Beef Substitute on Human Milk Composition in the Study of nUtrition in Postpartum and EaRly life (SUPER) Randomized Crossover Feeding Trial. Image Credit: Krysja / Shutterstock

In a recent study published in The American Journal of Clinical Nutrition, a group of researchers tested whether replacing beef with an ultra-processed plant-based beef substitute alters human milk fatty acid profiles and related maternal and infant measures in a double-masked, randomized, crossover feeding trial conducted as part of the SUPER study.

Dietary Shifts and Milk Lipid Sensitivity

For many families, the first “menu” a baby ever sees is human milk, which fuels rapid brain growth and immune development. What a lactating mother eats can shift the composition of milk nutrients, especially fats, but real-life food swaps are understudied. At the same time, ultra-processed foods (UPFs) now account for about 57 percent of daily calories in the United States, and higher UPF intake is linked to excess energy intake and poorer metabolic health. However, the study authors caution that these population-level associations do not establish direct causal effects on postpartum physiology. Determining whether plant-based meat substitutes and beef are biologically equivalent in human milk is practical for everyday meal planning and underscores the need for further research.

Controlled Feeding Design and Participant Criteria

The present study is a 25-day, double-blind, randomized, crossover-controlled feeding trial in lactating females, registered at ClinicalTrials.gov (NCT06082921) and approved by the Institutional Review Board (IRB). Each participant completed two 6-day diet phases (beef or plant-based substitute) separated by a 6-day washout; all foods other than the main (beef versus substitute) were identical and provided eucaloric meals aligned with Acceptable Macronutrient Distribution Ranges (AMDR). The main was 339 g (12 oz) of beef or a substitute per day, and compliance exceeded 95 percent. Evening human milk samples on day 6 of each phase were analyzed for 27 fatty acids.

Maternal and Infant Outcomes Assessed

Outcomes included human milk fatty acids, maternal weight, appetite, and satiety by visual analog scales (VAS), continuous glucose monitor (CGM) metrics (mean glucose, coefficient of variation, area under the curve (AUC)), and infant milk intake by weighed feeds. Fatty acid methyl esters (FAME) were quantified via gas chromatography flame ionization detection (GC FID) with structural confirmation by gas chromatography covalent adduct chemical ionization tandem mass spectrometry (GC CACI MS/MS). Eligibility targeted 6 to 12 weeks postpartum; exclusions included diabetes, metabolic medications, and neonatal intensive care unit (NICU) stays longer than 72 hours. Body mass index (BMI) was recorded at baseline; statistical comparisons used paired t-tests with false discovery rate control. The authors note the modest sample size (n = 17) as a limitation of the study.

Fatty Acid Shifts Following Diet Substitution

In a per-protocol analysis of 17 participants, human milk following the plant-based substitute phase contained more medium-chain saturated fatty acids (SFAs) typical of tropical oils and fewer long-chain polyunsaturated fatty acids (LCPUFAs) than after the beef phase. Specifically, lauric acid (12:0) was higher during the substitute diet (9.32 ± 1.8 percent versus 4.47 ± 1.82 percent), while arachidonic acid (ARA, 20:4n 6) was lower (0.35 ± 0.06 percent versus 0.41 ± 0.06 percent), both p < 0.001. These differences mirror the fatty acid profiles of the intervention foods, in which the substitute was higher in 10:0 to 14:0 SFAs and lacked several n-3 and n-6 LCPUFAs present in beef. However, the specific ingredient sources of these fatty acids are not identified in the paper. Total SFA in human milk was modestly higher with the substitute (35.75 percent ± 2.13 percent versus 34.75 percent ± 2.03 percent, p = 0.03).

Alignment of Milk Lipids With Dietary Profiles

During the substitute phase, human milk also shifted in several smaller fatty acids. Levels of several of these fats were lower than during the beef phase, including some n-6 fatty acids involved in inflammatory and immune pathways and one n-3 fatty acid important for cell membrane structure. A few other fatty acids showed small increases, reflecting the different fat composition of the two foods. Not all food differences translated into milk, suggesting physiological thresholds for transfer (for example, minimal change in oleic acid despite between-food differences). Overall, 13 milk fatty acids changed in the direction predicted by the foods’ profiles, while 14 showed no detectable difference.

Implications of LCPUFA Reductions for Infants

Clinically, reductions in LCPUFAs, especially ARA, may matter because ARA supports infant neurodevelopment and immune function and is typically accompanied by docosahexaenoic acid (DHA) in infant formulas. In this trial, the substitute diet resulted in about 15% lower ARA in human milk within 6 days. Contextualizing the dose, the beef diet supplied about 285 mg of ARA per day, consistent with prior supplementation studies that raise human milk ARA levels over weeks. Sustained avoidance of ARA-containing foods could plausibly reduce milk ARA further over time. However, the authors do not claim that these short-term biochemical shifts directly translate into clinical outcomes in infants. For households, this suggests that a “like-for-like” nutrient label match may not guarantee biological equivalence in milk lipids.

Maternal and Infant Measures Showing No Differences

Despite lipid shifts, no significant differences were observed between diets for maternal weight, subjective satiety (VAS), mean CGM glucose, glycemic variability, AUC, or infant milk intake by weighed feeds across days 4 to 6 of each phase. These stability signals are practical, while mothers might not feel different in fullness or energy, their milk lipids can still change quickly with common food swaps, and the authors note that carbohydrate-rich snacks in the controlled menus may have reduced the ability to detect subtle differences in postprandial glucose patterns.

Interpretation of Rapid Nutrient Shifts

In a tightly controlled, double blind crossover trial, replacing beef with a UPF plant based substitute rapidly altered human milk fatty acids, increasing medium chain SFAs (for example, lauric acid) and lowering LCPUFAs (notably ARA), these shifts occurred without measurable changes in maternal weight, appetite, CGM glucose patterns, or infant intake, suggesting that “nutrient matched” substitutes may not be biologically equivalent in early life nutrition and that full interpretation should consider the study’s sample size and funding context.

Note: This study was funded by the Beef Checkoff, a national program to increase beef demand. The study funder had no role in the study design, execution, or interpretation of the research, and placed no restrictions on publication.