Maternal medications linked to changes in protein and fat levels in human milk

Breast milk from mothers on antidepressants or anti-inflammatory drugs shows reduced protein and fat levels

In a recent study published in the JAMA Network Open, a group of researchers investigated the association between maternal use of antidepressant and anti-inflammatory medications and the macronutrient composition of human milk.

Background

Breastfeeding provides significant health benefits for both mother and infant and is recommended as the exclusive source of nutrition for the first six months. Human milk contains carbohydrates (primarily lactose at 8 g/100 mL), fat (3.5-4 g/100 mL), and protein (1 g/100 mL), with a total energy content of 66 kcal/100 mL. Macronutrient levels naturally vary due to factors like feeding frequency, infant age, maternal body mass index (BMI), and diet.

Over 70% of breastfeeding mothers use medications, yet their impact on milk composition remains unclear. Further research is needed to understand how medications may influence macronutrient levels in human milk.

About the study

The present study collected exposure data and milk samples from October 2014 to January 2024 through the University of California, San Diego Human Milk Biorepository.

Participants included mothers of singleton infants, with milk samples collected between 2 weeks and 12 months postpartum. Of 3,366 samples meeting inclusion criteria, 310 were from mothers on long-term medication and 3056 from untreated mothers.

Medicated mothers were categorized into four groups based on exposure: selective serotonin reuptake inhibitors (SSRIs), monoclonal antibodies (MABs), systemic corticosteroids, and other anti-inflammatory drugs (ADs). Samples from mothers exposed to multiple medications were excluded.

Maternal health conditions and mood disorders were assessed via interviews and the Edinburgh Postnatal Depression Scale. Untreated mothers with inflammatory or mood disorders served as disease-matched controls (DMCs), while healthy controls were frequency-matched by infant age and sex.

Milk macronutrients (carbohydrates, fat, protein, and total energy) were measured using near-infrared spectroscopy. Statistical analysis adjusted for covariates such as maternal and infant age, BMI, breastfeeding exclusivity, feeding frequency, collection time, and maternal cannabis use.  

Study results

The study analyzed 384 milk samples, including 63 exposed to SSRIs, 63 to MABs, 33 to systemic steroids, 20 to other ADs, 141 from DMCs, and 64 from healthy, unmedicated mothers. Among the mothers, 7.0% were Asian, 12.5% non-Hispanic Black, 12.5% Hispanic, 75.8% non-Hispanic White, and 2.9% from other racial or ethnic groups.

The cohort included 50.5% female infants, with mean (SD) maternal and infant ages of 33.5 (4.4) years and 6.6 (5.4) months, respectively. Healthy control mothers were younger than exposed and DMC mothers, but no differences were found in infant age across groups.

Mothers treated with SSRIs had a significantly higher mean (SD) BMI of 29.9 (6.8) compared with other groups. Cannabis use was lower among mothers treated with ADs than in other groups, with no significant differences in race, parity, infant sex, exclusive breastfeeding, or socioeconomic factors. Missing data on sample collection time, maternal occupation, household income, and feeding frequency ranged from 3.6% to 12.5%.

Mood disorder prevalence was lower in SSRI-treated mothers compared to DMCs (92.1% vs. 100%), while inflammatory disorder prevalence was lower among AD-treated groups compared to DMCs.

Macronutrient analysis showed mean (SD) levels of carbohydrates, fat, protein, and total energy as 8.41 (2.19) g/100 mL, 4.06 (1.74) g/100 mL, 0.92 (0.53) g/100 mL, and 78.15 (22.35) kcal/100 mL, respectively.

Carbohydrate levels showed no significant differences between groups. Protein content was 15% to 21% lower in all exposed groups compared to healthy mothers, but after adjustments, the difference remained significant only for SSRIs and steroids.

Fat and energy levels were 10% to 22% lower in samples from mothers treated with other ADs compared to DMC and healthy controls. Adjusted analyses revealed significant differences only in fat content when compared to DMCs.

Conclusions

To summarize, this study is the first to investigate the relationship between long-term maternal medication use and macronutrient composition in human milk. While carbohydrate levels remained consistent, protein and fat levels showed significant variations among exposed groups. Although mean values were within reference ranges, some samples had notably low macronutrient levels.

The clinical implications of these differences are unclear, particularly for preterm or ill infants, who may face growth or health risks if nutrient deficits go undetected. Factors such as maternal diet, nutrient stores, and disease severity might influence milk composition and could represent unmeasured confounders.