A major European trial suggests that reducing milk protein in infant formula may not lower BMI by age two, but may help put early growth on a more stable trajectory.
Trial: Infant Milk Intervention Trial (ToMI): A randomized controlled trial of the effect of protein content in infant milk on BMI and growth. Image credit: MIA Studio / Shutterstock
In a recent study published in the journal clinical nutritionresearchers present results from the Infant Milk Intervention (ToMI) trial, a multicenter (Spanish and German) randomized controlled trial (RCT) that investigated whether reducing protein intake during the first two years of life can reduce body mass index (BMI) at 24 months. In this study, 1,624 healthy children were randomly assigned to a low-protein (LP; 1.5 g/100 kcal) or high-protein (HP; 6.1 g/100 kcal) isocaloric regimen.
Study results showed no statistically significant differences in participants’ BMI z-scores (zBMI) at 24 months. However, in the high-protein group, we observed a significant upward bias in weight and length Z-scores compared to World Health Organization (WHO) standards. These findings suggest that although the critical period for obesity programs may peak during infancy, protein intake remains a major driver of body weight and linear growth throughout early childhood.
Background of the early protein hypothesis and infant nutrition
The Childhood Obesity Prevention (CHOP) trial previously demonstrated a causal link between high protein intake in infancy and increased risk of obesity in school age, an observation thought to be mediated by the ‘early protein hypothesis’, suggesting that high concentrations of branched chain amino acids (BCAAs) may activate the mammalian target of rapamycin (mTOR) pathway.
When children reach second grade, their diet changes markedly from monotonous (breast milk) intake to a variety of solid and semi-solid foods. However, pediatric records highlight that despite these changes, dairy products continue to provide approximately 50% of infants’ total protein intake.
These records further demonstrate that although US and European public health guidelines recommend that infants and young children consume moderate amounts of dairy products (often around 300-500 ml per day), actual intakes often exceed these levels, raising clinical concerns about unintended metabolic stimulation. Unfortunately, the evidence in early childhood is not as clear as in infancy.
Infant milk intervention trial design
The Infant Milk Intervention (ToMI) trial aimed to address these knowledge gaps and inform future pediatric recommendations by investigating whether altering the protein content of infant milk (YCF) could reduce hypothetical metabolic risks in populations already consuming high levels of animal protein.
The study used a randomized, double-blind, parallel-group design utilizing a cohort of 1,624 healthy children (12 months of age) recruited from Munich, Germany, and the Reus/Tarragona region of Spain. Study participants were randomly assigned to consume one of two isothermic milks (201 kJ/48 kcal per 100 mL) until 24 months of age.
The low protein (LP) cohort was provided with YCF intended to approximate the protein density of breast milk, 0.72 g protein/100 mL (1.5 g/100 kcal).
The high protein (HP) cohort was provided with YCF that approximates the protein density of regular cow’s milk, 2.95 g protein/100 mL (6.1 g/100 kcal).
This study used a standardized WHO protocol to track participants’ anthropometric measurements (particularly weight and recumbent length) at 12, 18, and 24 months of age. Statistical models were used to standardize these datasets into age- and sex-adjusted Z-scores.
In addition, dietary intake was monitored using parent-reported 24-hour recalls conducted during the day and utilizing photoportion guides. Statistical analyzes mainly consisted of analysis of covariance (ANCOVA) models, following an intention-to-treat (ITT) approach and adjusting for participants’ baseline anthropometry, gender, and country.
Effects of protein intake on infant growth
Study intervention analysis revealed cohort-specific dietary segregation at 18 months, with the LP group maintaining protein intake at 14% of total energy, whereas the HP group increased to 18% (p < 0.001). However, at 24 months, the mean zBMI scores of the two cohorts were statistically indistinguishable.
Of note, the longitudinal data revealed three statistically significant quadratic effects.
At 18 months, the LP group had significantly lower zBMI than the HP group (mean difference (MD) = -0.07; p = 0.007), suggesting a temporal effect on BMI at this age.
Both weight and length Z-scores were found to be significantly lower in the LP group at 18 and 24 months (p < 0.001). At 24 months, the Z-score for MD for weight = -0.11 and Z-score for MD for length = -0.13 for LP and HP.
Children with lower baseline protein intake, especially those with 15% or less of total energy intake from protein, had a more pronounced response to the study intervention (p < 0.001). Moreover, this response was significantly stronger in German children (p = 0.002) than in Spanish children (p = 0.911). The authors say this supports the hypothesis that lower pre-intervention protein intake enhances the intervention effect.
Infant milk recommendations and future follow-up
The ToMI trial shows that changing milk protein intake during the second year of life does not result in a sustained reduction in BMI at 24 months, but supports a more stable growth trajectory. Considering that most infants’ protein intake is reported to exceed growth requirements, the authors suggest that regular milk intake should generally not exceed approximately 200-400 mL per day to prevent unnecessary growth promotion.
However, the authors emphasize that further follow-up through the age of 6 years is needed to determine whether these early growth deviations manifest as long-term obesity. Additionally, the findings were limited to the milk protein present in the study formulation, and dietary intake was assessed using a single 24-hour recall.
