High-dose vitamin D in pregnancy did not leave children with stronger bones at 13, a Danish trial found

Six years ago a Danish research group reported that children whose mothers took a high dose of vitamin D in late pregnancy had more mineral in their bones at ages 3 and 6. When the same group scanned those children again at 13, the advantage was gone.

The follow-up, published on September 25 in The Lancet Regional Health - Europe, found no difference in bone mineral between teenagers whose mothers had taken 2,800 international units a day and those whose mothers had taken the standard 400. Fractures up to age 13 did not differ either. The trial was not designed around bone, so it cannot exclude a small effect in either direction, and every mother took at least the standard dose.

Why vitamin D in pregnancy was tested for child bone

Vitamin D helps the body take up calcium and phosphate, which are the main components forming bone mineral, and a baby’s skeleton is laid down before birth from what the mother supplies.

The longer-term hope concerns peak bone mass, the greatest amount of bone a person ever has, reached in early adulthood. The National Osteoporosis Foundation, a US body, has called raising it an important strategy aimed at reducing the risk of osteoporosis, a condition in which bones become thin and break easily, or of low bone mass later in life.

Small early gains attract attention because of a calculation the Danish authors cite. A 10% rise in peak bone mass around age 20, a mathematical analysis predicted, could delay osteoporosis risk in adulthood by 13 years. If a few months of extra vitamin D before birth could nudge the skeleton onto a higher track for life, it would be a cheap way to prevent fractures decades later.

Studies that simply measured mothers’ vitamin D levels had pointed both ways. One large study, the Avon Longitudinal Study of Parents and Children in England, followed 3960 mother-and-offspring pairs, most of them of white European origin. It found no meaningful link between the mother’s vitamin D status in pregnancy and the child’s bone mineral content, the weight of mineral in the skeleton, in late childhood. Only a randomized trial could separate the vitamin from everything else that differs between mothers with high and low levels.

How the pregnancy vitamin D trial worked

The trial sits inside a Danish study of childhood asthma called COPSAC2010, run from Copenhagen. Its entry on ClinicalTrials.gov, the US government’s public register of trials, gives its original aim as preventing asthma symptoms (recurrent wheeze) in childhood, with bone as a planned secondary question.

Between March 2009 and November 2010 the study enrolled 623 mother-child pairs in pregnancy week 24, leaving out women who had already taken more than 400 IU of vitamin D in the previous 6 months. The women were assigned at random to one of two groups. One took 2,400 international units (IU) of vitamin D3 a day, the international unit being the standard measure of vitamin D dose, with 400 IU equal to 10 micrograms. The other took 2 placebo tablets with no active substance, identical in look to the real ones.

Every woman was also advised to take the 400 IU a day that the Danish Health Authority sets as the standard intake in pregnancy, which made the trial a dose-comparison study of 2800 IU a day against 400 IU a day, with no group going without vitamin D. The women took the tablets from pregnancy week 24 until 1 week after birth.

Neither the families nor the researchers knew who was in which group until the asthma question had been answered: the study was unblinded at age 3 years, when the group assignments were revealed. The scans and fracture counts at 6 and 13 came after that point.

After withdrawals before birth, 584 children were available for analysis, and at 13 they were invited for a scan by dual-energy X-ray absorptiometry (DXA), a low-dose X-ray that measures how much mineral bones contain. Each scan took about 3 minutes in a single pass, and the team chose the method for its low radiation dose and short scan time. Two independent experienced specialists checked every scan without knowing which group the child was in.

Each scan gave two measures: bone mineral content, the total weight of mineral in grams, and bone mineral density, that mineral divided by the area of bone scanned. Both were predefined secondary outcomes, results the trial planned to report but was not sized to answer.

The team also counted broken bones from birth to 13, gathering them from interviews with parents and from medical records, and all fractures were radiologically verified by physicians, meaning doctors confirmed each one on an X-ray.

What the trial found in children’s bones at 13

At 13, 416 children (71%) had a whole-body DXA scan: 219 boys (53%) and 197 girls (47%), including 212 of the 297 children in the high-dose group and 204 of the 287 in the standard-dose group.

The two groups did not differ on any of the bone mineral outcomes. Total-body bone mineral content was 6.7 grams higher in the high-dose group, but the range ran from 28.3 grams lower to 41.6 grams higher, and total-body bone mineral density was 0.003 grams per square centimeter lower, with a range from 0.017 lower to 0.011 higher. Both ranges sit squarely across zero.

The researchers then looked for a group in which the vitamin might still have mattered, splitting the children by sex, by season of birth and by the mother’s vitamin D level before the tablets began, and found a difference in none of them. Puberty was another worry, since at 13 some children are well into it and others are not. When the team adjusted for Tanner stages, a standard five-step scale of how far puberty has progressed, the results were unchanged.

Fractures told the same story. The 550 children (94%) followed all the way to 13 had 114 fractures between them, spread among 99 children: 53 in the high-dose group and 61 in the standard-dose group. The rate of a first fracture was 15% lower in the high-dose group, but with a range from 43% lower to 25% higher the result is compatible with a real reduction, with no change and with a modest increase.

The authors conclude that there was “no lasting effect of pregnancy supplementation with vitamin D on child bone mineralization and fracture risk”.

What the same children showed at 3 and 6

The new result undoes an earlier one from the same trial. In 2020 the group reported scans at ages 3 and 6 in JAMA Pediatrics, concluding that the high dose improved offspring bone mineralization through age 6 compared with the standard dose. Whole-body bone mineral content was then 11.5 grams higher in the high-dose group, with a range from 2.3 to 20.7 grams, and there had been 23 fractures in the high-dose group against 36 in the standard-dose group, a gap that fell just short of the usual test for ruling out chance.

On the strength of that, the 2020 paper suggested raising the recommended intake during pregnancy, which it said may influence peak bone mass, fracture risk and osteoporosis later in life.

A second paper went further by combining the mother’s dose with the child’s own vitamin D level at six months, counting at least 75 nanomoles per liter of blood as sufficient. The 179 children at or above that level whose mothers had taken the high dose had a 60% reduced incidence of fractures compared with the 93 children below it whose mothers had taken the standard dose.

Pooling all the scans from ages 3, 6 and 13, the new analysis put the overall difference at 3.79 grams of bone mineral content, with a range from 20.3 grams lower to 27.9 grams higher. The paper now calls the early gain a short-lived beneficial effect only observed in younger children.

One thread from the earlier work survived in weaker form. Among children whose mothers took the high dose, those who also had sufficient levels at 6 months, by the same definition, had overall higher bone mineral content at 13. That comparison was exploratory and made after the main analysis, and children with good vitamin D levels at six months may differ from other children in many ways.

What other trials of pregnancy vitamin D and bone found

Few trials have tested the question, and the Danish paper says it is the first randomized trial to look at long-term effects of prenatal high-dose vitamin D on children’s bone health.

The other large trial is British. The Maternal Vitamin D Osteoporosis Study, known as MAVIDOS, recruited pregnant women at three study sites in the UK and assigned 1,134 of them to 1,000 IU a day or placebo. Its main result, measured in newborns, was null: supplementation did not lead to more whole-body bone mineral content than placebo.

Later scans of some of those children looked better. When 454 children, all from one of the three sites, were followed up at ages 6 to 7, those whose mothers had taken vitamin D had greater bone mineral density and lean mass in mid-childhood than the placebo group, in an exploratory analysis that was not part of the original plan. The size of the gain was similar to the Danish result at the same age.

The British team has not yet reported beyond that age; their 2024 paper says follow-up during early adolescence is ongoing. The Danish authors say it will be interesting to see whether the vitamin D effects in that trial fade in the same way.

A third trial, in Dhaka, Bangladesh, found nothing at any age it examined. It followed 642 children to age 4 after their mothers had taken one of several weekly doses or placebo, and concluded that in a population where vitamin D deficiency is widespread, vitamin D supplementation did not increase the children’s bone mineral or grip strength. All the women had also been given calcium, which the authors suggest may have made up for the relative vitamin D deficiency among women in the control group.

The evidence is firmer for other influences on young bones. In a position statement grading what is known about building peak bone mass, the National Osteoporosis Foundation gave its best grade, A, to the positive effects of calcium intake and physical activity, especially in late childhood and around puberty.

What the vitamin D trial cannot show

It was not sized for bone. The number of women was chosen to detect an effect on wheeze, and the authors acknowledge that the trial may have been too small to pick up a modest effect on bone, because the study was powered for persistent wheeze and asthma. The ranges around the results leave room for a small benefit or a small harm.

Missing scans. Some 29% of the children did not have a DXA scan at age 13, a gap the paper lists as a limitation, though children with and without a scan did not differ in their characteristics at the start of the trial.

Few fractures, analyzed after the fact. The authors point to the relatively few fractures in an analysis done after the fact, and note that a child’s fracture risk also tracks physical activity and the accidents that follow, which the trial did not account for. An active child has more chances to break an arm, whatever their bone density.

Thirteen is an awkward age to measure. Bone density is hard to interpret around the teenage growth spurt, and the paper warns that areal bone density, calculated from the flat area of bone on the scan, may decrease despite normal growth. Two kinds of scanner were also used, though neither showed an effect on its own.

It compared two doses. Both groups took at least 400 IU a day, and women already taking more than that were excluded, so the trial says nothing about taking no vitamin D in pregnancy.

One country, one population. The 2020 report noted that all participants were Danish and white, so the results describe children in that population.

Thirteen is not adulthood. Peak bone mass comes around age 20. The authors call the question of whether vitamin D before birth changes it an important area of investigation for future research, while saying their own results do not support the idea.

The paper states that the funding agencies did not have any role in the study’s design or conduct, and the authors report no conflicts of interest.

Whether a vitamin D supplement, and at what dose, suits a particular person in pregnancy is assessed by their doctor.

A head start in bone mineral that was visible at 3 and 6 had disappeared by 13 in the one trial that has followed children that far.