Skip to main content

The clock starts ticking early: circadian variation in preterm breastmilk

January 27, 2026

Human breast milk is surprisingly adaptive, consistently changing to best meet the baby’s needs. Even the time of day has been known to affect its composition. But what happens when a baby is born very prematurely? 

A new study by BioClock researchers Demy van Gilst  and Inês Chaves aimed to find out if 24-hour variations are already present in breastmilk following a very premature birth. Researchers collected around 15 milk samples over 72 consecutive hours, analyzing macronutrients such as fat, protein, and carbohydrates. They then analyzed whether these nutrients had predictable day-night fluctuations.

Nutrient variation

The study found that 55% of milk series displayed a circadian rhythm in at least one macronutrient. Fat showed the strongest rhythmicity: 41% of the series exhibited clear day‑night patterns, although the timing of peaks varied between mothers, likely reflecting individual differences in their biological rhythms. Protein, carbohydrates, and total energy content displayed rhythmicity less often (14%, 18%, and 27%, respectively).

Gestational age influences rhythmicity

Interestingly, the longer the pregnancy had lasted, the higher the likelihood that the milk showed a circadian rhythm, despite all infants being very preterm. This suggests that circadian signals may be increasingly embedded in maternal milk as pregnancy progresses.

What does this mean?

Now that we know that breastmilk composition varies, even after very preterm births, this can help adapt neonatal care. For example, in the Neonatal Intensive Care Unit (NICU), this knowledge could inform feeding times and composition so that they better align with the biological rhythms present in the milk. Incorporating this concept of ‘chronolactation’ in care, may help give preterm babies some extra assistance while they adjust to life outside the womb.

Read the full publication here.


NWO logo



The BioClock Consortium is funded by the NWA-ORC programme of the Dutch Research Council (NWO; project number 1292.19.077).