Vitamin D in Pregnancy: A Bone Marker Moves, Outcomes Don’t

Vitamin D during pregnancy can modestly reduce a marker of maternal bone resorption. That is a real randomized-trial finding. It is not proof that supplementation preserves a mother's bone density through breastfeeding, prevents fractures, or changes long-term skeletal health.

The distinction matters because pregnancy asks a lot of calcium metabolism. Bone turnover changes across gestation, and the body adapts to support fetal mineral needs. A lower resorption marker may be biologically meaningful, but markers are not the same thing as bone density, symptoms, or fractures. In this case, the strongest evidence stops at the marker.

A randomized trial found a modest CTX effect

The most directly relevant study is the 2021 MAVIDOS trial, a randomized, double-blind, placebo-controlled study. Participants took 1,000 IU of cholecalciferol daily from 14 weeks of pregnancy until delivery, or placebo. Researchers measured second-void urinary C-terminal telopeptide of type I collagen, or CTX, at 14 and 34 weeks. CTX reflects bone resorption: higher values generally indicate more breakdown of bone collagen.

Among 372 women with paired urinary CTX and 25-hydroxyvitamin D data, median urinary CTX rose during pregnancy in both groups. It increased from 223.6 to 449.7 micrograms per millimole of creatinine in the placebo group and from 222.3 to 419.3 micrograms per millimole in the vitamin D group. At 34 weeks, the between-group difference was 30.4 micrograms per millimole of creatinine, about 6.8% lower with vitamin D. The study reported that the difference was statistically detectable.

That is a modest biochemical effect. Vitamin D did not stop the pregnancy-related rise in CTX; it slightly reduced its size. The trial offers good evidence that this dose changed this urinary marker under its study conditions. It does not establish the size of any benefit that a patient would feel or notice.

CTX tracked with early-postpartum bone measures, but that was an association

MAVIDOS also measured maternal bone mineral density and bone mineral content by DXA within two weeks after delivery. Women with higher CTX at 34 weeks tended to have lower early-postpartum lumbar-spine bone measures. This association makes the CTX finding more interesting: the marker was related to a measurement of maternal bone status.

But association does not turn CTX into a validated stand-in for every skeletal outcome. Most importantly, MAVIDOS did not show a randomized treatment effect on maternal postpartum bone mineral density or bone mineral content. The supplement changed the biomarker, while the trial did not demonstrate that it changed the harder bone endpoint.

A second trial narrows the claim further

A 2024 study by Krupa and colleagues found the same general direction: prenatal vitamin D supplementation reduced maternal CTX. Continuing vitamin D after delivery did not improve that signal, and the study found no effect on other biomarkers of bone metabolism.

Replication of the CTX result helps. Yet the pattern is selective rather than comprehensive. It does not show that vitamin D produces broad anabolic bone remodeling in pregnancy or postpartum. It shows that prenatal supplementation can nudge one resorption marker.

Pregnancy cohort research gives that marker some context. O'Brien and colleagues' 2021 study linked bone resorption and dietary calcium during pregnancy with later maternal bone-health signals. That supports the idea that pregnancy bone resorption deserves attention. It cannot prove that lowering CTX with vitamin D causes better future bone health, because it was observational rather than a randomized test of that clinical outcome.

Why CTX cannot settle the bone-health question

CTX is useful, but it is not a direct scan of the skeleton. A 2021 study by Breasail and colleagues illustrates the problem. In a population with habitually low calcium intake, CTX rose 23.0% during pregnancy, with a 95% confidence interval of 15.09% to 29.29%. Parathyroid hormone rose 13.2% and 1,25-dihydroxyvitamin D rose 21.0%. Despite those biochemical changes, peripheral pQCT did not show evidence of skeletal mobilization.

This does not mean CTX is meaningless. It means a rise or fall in CTX alone cannot prove that maternal bone was lost or saved. Bone turnover markers capture part of a changing physiology; DXA, pQCT, follow-up through lactation, and clinical outcomes answer different questions.

Vitamin D has randomized evidence for a modest late-pregnancy CTX effect. It does not yet have pregnancy-specific randomized evidence for preserving maternal bone density across lactation.

The postpartum window is where the unanswered question lives

DXA within two weeks of delivery is too early to describe the larger arc of lactation. A 2023 study of bone density during and after lactation reported areal bone mineral density losses of 2% to 8% over the first two to six months of lactation, followed by partial recovery after weaning. A single immediate-postpartum scan cannot tell us whether a prenatal intervention changes that later loss or recovery.

That trajectory can also vary by health status and treatment exposure. In Ugandan women with HIV who initiated tenofovir disoproxil fumarate-based antiretroviral therapy during pregnancy, Nabwire and colleagues reported elevated bone resorption during lactation, accentuated declines in areal bone mineral density, and only partial recovery after lactation compared with women without HIV. Those findings should not be generalized to every pregnant person. They show why maternal bone biology cannot be reduced to one late-pregnancy urine result.

Vitamin D is plausible, but plausibility is not an outcome

Outside pregnancy, vitamin D3 can affect skeletal measures under some conditions. A 2023 systematic review and meta-analysis examined vitamin D3 supplementation, bone mineral density, and bone-turnover outcomes in adults and reported beneficial effects on bone mineral density with dose-response analysis. That evidence supports biological plausibility. It does not answer whether 1,000 IU daily during pregnancy protects maternal bone through breastfeeding.

Other pregnancy interventions also suggest maternal bone-related biology can respond to behavior. In the BHIP randomized trial, an individualized high-dairy-protein diet plus walking program supported bone-health biomarker profiles during pregnancy and postpartum recovery. That was a combined nutrition-and-exercise intervention, not a test of vitamin D alone, so it cannot be used as proof for prenatal vitamin D.

What to do with this evidence

For someone considering vitamin D during pregnancy, the evidence supports a narrow statement: 1,000 IU per day of cholecalciferol from 14 weeks until delivery modestly attenuated the rise in urinary CTX in a randomized trial. A second study found a prenatal CTX reduction as well. Higher late-pregnancy CTX was associated with worse early-postpartum DXA measures in MAVIDOS.

The evidence does not show that vitamin D supplementation prevents maternal bone loss during lactation, improves persistent postpartum recovery, reduces bone pain, or prevents fractures. Calcium intake, vitamin D status, breastfeeding intensity, ethnicity, medical conditions, and medications may all affect maternal skeletal responses, and the available studies do not settle how each factor changes the response.

Vitamin D is therefore a reasonable pregnancy nutrient with a measurable effect on a resorption marker, not established insurance against maternal bone loss. The trial result is worth taking seriously. So is its boundary: the clinically important outcomes remain unproven.