Pregnant Women Got Free Eggs and Fish. Parts of Their Babies’ Brains Grew More.

Every pregnancy comes with a flood of food advice. Eat this, avoid that, take these.

Most of that advice rests on studies that watch what people eat. Very few test it by changing a diet and waiting to see what happens.

A new trial in Ecuador did exactly that. It delivered eggs, fish, fruit and vegetables to pregnant women every week.

Then it scanned their babies’ brains before birth. Two deep brain structures grew more in the group that got the food.

It’s a genuinely interesting result. It’s also early, small, and easy to oversell, so let’s look at it carefully.

What is a healthy pregnancy diet?

Before the new study, here’s the baseline. The NHS guide to a healthy diet in pregnancy is a sensible place to start.

It advises at least 5 portions of a variety of fruit and vegetables every day. It also suggests 2 portions of fish a week, 1 of them oily, such as salmon, sardines or mackerel.

There are limits, though. The NHS says to have no more than 2 portions of oily fish a week, because it can contain pollutants.

It also says to avoid shark, swordfish and marlin. And tuna should be kept to no more than 4 cans or 2 tuna steaks a week.

Eggs are fine, with care. In the UK, the NHS says British Lion hen eggs can be eaten raw or partly cooked.

Other eggs should be cooked until the white and yolk are hard.

The NHS also recommends a folic acid supplement on top of food. So the big picture is simple: variety, plenty of plants, some fish, and a few sensible limits.

What’s been missing is strong proof that eating this way changes how a baby’s brain grows. That’s the gap this trial set out to fill.

The trial: a weekly food basket in Ecuador

Who took part

The study, published in PNAS in 2026, was led by researchers at Washington University in St. Louis and Universidad San Francisco de Quito.

It took place in El Quinche and Yaruquí, two low-income areas near Quito. The communities were chosen partly because of food insecurity.

The researchers recruited 215 healthy pregnant women, all in their first trimester. They were enrolled between April 2022 and July 2023.

How it worked

Each woman was randomly assigned to one of two groups. Randomisation is what makes a trial fair: the groups start out alike, so later differences are more likely caused by the programme.

• Control group (104 women): standard pregnancy care.
• Mikhuna group (111 women): standard care plus a weekly food delivery and nutrition coaching.

Mikhuna means “nourish” in Kichwa, a local Indigenous language.

Each weekly basket held 8 eggs, 500 grams of fish, and a variety of locally grown fruits and vegetables. The fish included trout, tilapia and canned tuna.

The coaching encouraged women to eat a wider range of foods and to limit highly processed ones. The programme ran from about 12 weeks of pregnancy until birth.

Why these foods?

The team based the basket on research into what humans ate for most of our evolution. Those diets were very varied, with fresh plants and fish or other animal foods.

The authors stress this is not the modern “Paleo” fad diet. It was adapted to foods that grow locally in Ecuador, including native fruits and squash.

What the trial found

Diets became much more varied

The programme clearly changed what women ate. By 35 weeks, 74.5% of the Mikhuna group met a standard target for dietary variety, against 55.8% of the control group.

The Mikhuna group also ate more distinct foods each day: 14.0 on average, against 9.0 in the control group.

Two deep brain structures grew more

Doctors scanned the babies by ultrasound at 21 and 35 weeks of pregnancy. The team then compared how much each structure grew between those scans.

The corpus callosum is the thick bundle of nerve fibres that connects the brain’s left and right halves. It’s the largest white matter tract in the brain.

The gangliothalamic ovoid, or GTO, is a cluster of deep structures that includes the thalamus and basal ganglia. These areas help relay signals and control movement.

The surprise: a shorter thigh bone

Not everything went in the expected direction. The thigh bone grew slightly less in the Mikhuna group.

The authors aren’t sure why. Milk and dairy intake didn’t differ between the groups, so that doesn’t explain it.

One idea is that nausea played a part. More women in the Mikhuna group had nausea and vomiting at 21 weeks, which could have reduced how much they ate.

Newborn size didn’t change

At birth, babies in both groups were about the same length and weight. The differences were small and not statistically significant.

Mothers gained an average of 8.47 kg between 12 and 35 weeks, with no difference between groups.

Why the result needs caution

Only one brain result held up to a stricter test

When a study checks many measurements, some will look significant by chance. Researchers use corrections to guard against this.

The team applied one, called the Benjamini–Hochberg procedure. After it, only the GTO result remained significant.

That means the corpus callosum and thigh bone findings are best seen as leads to test again, not firm results. The authors say as much.

Hard-to-measure brain structures

Measuring small, deep brain structures by ultrasound is tricky. The study’s own reliability checks were low for these parts, including the corpus callosum and GTO.

That adds noise to the measurements. It doesn’t make the result wrong, but it’s another reason for caution.

People dropped out, and habits stuck

About 19.5% of participants were lost to follow-up. The most common reasons were miscarriage, moving away and lack of time.

Old habits also persisted. At 35 weeks, 96.7% of women were still eating rice and 92.8% ultra-processed foods, with no difference between groups.

Bigger isn’t automatically better

The trial measured brain structures, not how children think, learn or behave. Bigger structures on a scan don’t guarantee better development.

There are hints they might matter. A large Dutch study, Generation R, linked a smaller GTO in children to more emotional problems.

But that’s a different, observational study.

What earlier research found

A Spanish trial pointed the same way

The closest match is a trial in Barcelona called IMPACT BCN. It tested a Mediterranean diet, with free extra-virgin olive oil and walnuts, in pregnancies at risk of small babies.

An MRI analysis of 90 fetal brains compared the groups. Babies in the diet group had larger total brain volume and a larger corpus callosum than those in usual care.

The same trial then followed 626 children to age 2. Those in the Mediterranean diet group scored higher on standard tests of thinking and social-emotional development.

The wider picture: a small link

A 2017 systematic review and meta-analysis pooled 18 studies with 63,861 participants. It looked at mothers’ diet quality in pregnancy and children’s development.

It found “a small positive association” between better maternal diet and child development. The link was clearest for thinking skills.

But those studies were observational, and the review found signs of publication bias. Most also didn’t account for what the children themselves ate.

How it fits together

Two different diets, in two very different countries, were both linked to a bigger corpus callosum before birth. That’s encouraging.

Only the Spanish trial has so far followed children after birth. The Ecuador trial still has to show its brain differences translate into real-world benefits.

How much should you trust this?

Early. It’s a randomised trial, which is a strong design. But it’s one small study, measuring scans rather than children’s development.

What makes it convincing

• Women were randomly assigned, so the groups started out alike.
• The food basket clearly changed diets, so the programme did what it aimed to do.
• The trial was registered in advance, and the authors reported no competing interests.
• A separate Spanish trial found a similar brain difference with a different healthy diet.

What makes me cautious

• Only 215 women took part, and about a fifth were lost to follow-up.
• Only the GTO result survived the stricter statistical correction.
• Ultrasound measurements of deep brain structures had low reliability.
• The programme combined food, coaching and support, so we can’t tell which part mattered.
• Babies’ size at birth didn’t differ, and the thigh bone grew slightly less.
• No one has yet checked whether these children develop differently.

What this means for you

• Variety is the headline. The clearest effect was a more varied diet. That matches standard advice to eat plenty of different fruits, vegetables and proteins.
• Fish can be part of it, within limits. Follow local guidance, like the NHS limit of 2 portions of oily fish a week. Avoid high-mercury fish.
• Don’t treat this as a brain-boosting diet. The study measured scans, not intelligence. It’s not a reason to change your plan on its own.
• Talk to your midwife or doctor. Nutrition needs vary, especially with nausea, allergies or conditions like gestational diabetes.

The NHS page on foods to avoid in pregnancy is a practical checklist to keep handy.

In this short video, NHS Greater Glasgow and Clyde covers eating well in pregnancy and the right vitamins:

What we still don’t know

1. Do the brain differences last? The children need following into childhood.
2. Do they change development? No one has tested thinking, learning or behaviour yet.
3. Which part worked? Fish, eggs, vegetables and coaching were all delivered together.
4. Why did the thigh bone grow less? The authors could only offer possible explanations.
5. Would it work elsewhere? This was one community in Ecuador, with a locally designed basket.

My take: a promising basket, not a brain hack

What I like about this study is that it tested food, not pills. The researchers changed what real families ate, then measured what happened.

That’s much harder than surveying diets, and much more convincing.

I’m less convinced by the “ancestral diet” framing. The trial can’t tell us whether eating like our ancestors matters, or simply eating a varied, healthy diet.

And the brain findings are small and early. Only one survived a stricter check.

Still, it points the same way as the Spanish trial. Good food in pregnancy may shape a baby’s brain in ways we can start to see.

For now, the practical message hasn’t changed: eat a wide variety of real food, with sensible limits on fish.