A pregnancy food package grew the corpus callosum and shortened the femur

Design

Randomised controlled trial with weekly food delivery plus behaviour change counselling from 12 weeks' gestation to birth

Population

215 pregnant women in Ecuador in the first trimester: 111 intervention, 104 control

Primary outcome

Fetal brain and bone growth on serial ultrasound, with maternal dietary diversity

Effect

Dietary diversity threshold met by 74.5% vs 55.8% (P=0.004). From 21 to 35 weeks: corpus callosum length +0.19 cm (95% CI 0.02-0.35), gangliothalamic ovoid height +0.15 cm (0.03-0.26), femur length -0.10 cm (-0.19 to -0.02). At 35 weeks: corpus callosum Z +0.56 (0.03-1.09), femur length Z -0.21 (-0.42 to 0.00)

Whether improving a mother's diet changes her baby's brain has been argued mostly from observational data. This randomised trial in Ecuador tested a specific intervention: 215 women in the first trimester were assigned either to usual care or to Mikhuna — a weekly delivery of eight eggs, 500 g of fish and a range of locally sourced fruit and vegetables from 12 weeks to birth, with counselling to diversify the diet and limit ultra-processed food.

The intervention worked as an intervention: 74.5% of the Mikhuna group met the minimum dietary diversity threshold against 55.8% of controls. On fetal ultrasound between 21 and 35 weeks, corpus callosum length grew 0.19 cm more (95% CI 0.02-0.35) and gangliothalamic ovoid height 0.15 cm more (0.03-0.26) than in controls, with a corpus callosum Z score difference of 0.56 at 35 weeks.

Femur length, however, was 0.10 cm shorter in the intervention group (-0.19 to -0.02), with a Z score difference of -0.21 at 35 weeks. That is not a footnote. A nutritional intervention that increases brain structure measurements while reducing a standard measure of skeletal growth needs explaining before it can be recommended, and this trial does not explain it. The honest summary is that a diet intervention changed fetal ultrasound measurements in two directions, and what either means for the child is unknown.

• Do not translate this into dietary advice; the outcomes are ultrasound measurements, not child development
• Note the shorter femur length in the intervention group alongside the brain findings
• Dietary diversity counselling with actual food delivery is the intervention — counselling alone was not tested
• The setting is Ecuador; food availability and baseline diet differ substantially elsewhere
• Continue standard antenatal nutrition advice unchanged

Why it matters

It tests a maternal nutrition hypothesis with a randomised design and returns a genuinely mixed answer, which is rarer than it should be.

Don't overread it

These are fetal ultrasound measurements, not neurodevelopmental outcomes; no child outcome has been shown.

The statistics, in plain English

Differences of 0.15 to 0.19 cm on fetal ultrasound are close to the measurement error of the technique itself, and confidence intervals reaching 0.02 and 0.03 mean the smallest plausible effects are close to nothing. With multiple fetal structures measured at several timepoints, some differences reaching significance is expected by chance alone, and no correction for multiple comparisons is described. The femur result matters because it points in the opposite direction to the hypothesis, which is exactly the kind of finding that gets lost in a summary.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.