Metagenomic sequencing of 4,526 longitudinal fecal samples from 714 Dutch mother-infant pairs (Lifelines NEXT cohort), spanning pregnancy through 12 months postpartum, reveals the maternal gut — not vaginal or breast milk microbiomes — as the dominant source of infant gut strain colonization. Transmission probability scales with maternal species abundance and is time-dependent. Critically, maternal gut microbiome composition independently predicts infant eczema. Delivery mode and feeding mode exerted the strongest influence on infant gut functional profiles, while home versus hospital birth showed only modest microbiome differences.
This is a landmark dataset — 474 integrated clinical variables with ultra-deep multi-site sequencing rarely achieved at this scale in human cohorts. The finding that vaginal and breast milk microbiomes contribute only occasionally to infant gut colonization challenges the long-held "seeding" narrative around cesarean section recovery protocols and vaginal seeding interventions, which deserve reassessment. The maternal gut-to-eczema link is particularly actionable: it opens a causal pathway hypothesis where prenatal maternal microbiome optimization — via diet, avoidance of smoking, infection management — might reduce atopic disease burden in offspring. However, this remains observational; confounding by shared genetics and household environment cannot be excluded. The subtle pregnancy-related changes in the maternal microbiome influenced by diet and infections suggest a relatively stable transmission source, but also a modifiable one. For researchers and clinicians, this shifts the intervention target upstream — to maternal gut health before and during pregnancy — rather than postnatal supplementation strategies.