Vitamin D3 for Life

Live Enlightened

Physiology

Early-Life Vitamin D and Childhood Immune Trajectories: Full Evidence Review

Previous related pageEarly-Life Vitamin D and Childhood Immune Trajectories: Study SummaryNext related pageEarly-Life Vitamin D and Childhood Immune Trajectories: Study Summary
Study at a glance

What was studied?

Does vitamin D in early life affect developing immune cells or later childhood illness? The review compares an animal experiment with five independent human research programmes.

What did the studies find?

A mouse study found changes in developing blood-forming cells and their descendants. Two pregnancy trials did not show a clear breathing-related benefit at six years from higher supplementation. A higher infant regimen did not reduce overall reported infections. Two observational programmes gave different results for vitamin D status and an immune response linked to type 1 diabetes.

What does this mean?

Experimental cell changes, childhood illnesses and disease-risk markers answer different questions. The evidence does not establish that higher supplementation improves every part of immune development or prevents childhood immune disease.

What are the limits?

The human studies did not directly measure normal immune maturation. An association with vitamin D status does not prove prevention. Animal results do not establish the same effect in children. Some supplementary material was unavailable.

Read the full article below ↓

Early-life vitamin D research spans several different questions: whether vitamin D availability changes developing immune cells, whether supplementation changes childhood illness, and whether measured vitamin D status predicts later autoimmunity. These questions cannot be answered by one blood test or one disease outcome. This review brings together experimental developmental evidence and five independent human research programmes to examine what is known about immune-related trajectories from early life into childhood. For the shorter account, read Early-Life Vitamin D and Childhood Immune Trajectories: Study Summary.

Development, illness and risk are different outcomes

Immune development concerns how immune cells and their functions emerge and change. A childhood infection count, asthma diagnosis or autoantibody result can be relevant to immune health without directly measuring that development. A convincing explanation must distinguish an experiment on developing cells from an association with a later diagnosis. Vitamin D and Innate Responses: Recognition, Peptides and Function: Full Evidence Review examines a related distinction between immune recognition signals, measured defence proteins and functional responses.

The human evidence examined here includes the COPSAC and VDAART pregnancy trials, the VIDI infant supplementation trial, and the TEDDY and DAISY observational programmes. Each contributes a different comparison. COPSAC and VDAART tested additional vitamin D during pregnancy and followed respiratory outcomes. VIDI compared two infant regimens and counted reported infections. TEDDY and DAISY examined measured status or intake in relation to islet autoimmunity, an immune response directed against insulin-producing pancreatic tissue.

An additional mouse study by Oh and colleagues addresses developmental immune-cell programming more directly. Its experimental results help explain what such a proposition means. They do not convert the five human programmes into measurements of normal immune maturation, nor do they prove that a supplement prevents disease in children. The relevant biochemical background is available in Vitamin D Activation and Vitamin D Enzyme Pathways (CYP27B1 / CYP24A1). These explain processing of vitamin D, which is a different question from whether an intervention changes a childhood illness outcome.

All doses below describe research interventions. These interventions do not establish that a higher dose is better. For current UK infant and child guidance, see NHS Vitamins for children. Infants born early require advice from their neonatal team.

Experimental evidence for developmental immune-cell programming

Oh and colleagues investigated whether maternal vitamin D deficiency could affect developing blood-forming cells and their descendants. Their experiments used mice and fetal-liver cell preparations containing blood-forming stem cells and progenitors. Such progenitors give rise to more specialised cells, including cells involved in immune responses. The investigators also studied cell behaviour in a small set of human cord-blood samples, but this was not a trial following children’s health.

In the mouse experiments, cells from the different maternal nutritional conditions were transferred into adult recipients. The recipients received vitamin D sufficient conditions. Later experiments included transfer through a further recipient generation. The investigators observed differences in descendant macrophage characteristics and in metabolic outcomes in the recipient animals. Macrophages are immune cells with roles that extend beyond responses to infection, including activity within tissues such as fat. Vitamin D Receptors introduces receptor-mediated vitamin D signalling without establishing any particular developmental outcome.

This provides a specific developmental proposition: an early nutritional environment can be associated with altered blood-forming cell populations and later properties of their descendants in this experimental system. It is stronger evidence for a developmental effect than simply finding a relationship between a child’s vitamin D blood level and illness. The experimental manipulation and persistence across transfers are material to that interpretation.

Nevertheless, the cell preparations were not identical populations of purified stem cells. Supplementary measurements showed differences in progenitor composition. Consequently, the later outcomes cannot be assigned with certainty to an intrinsic change in one otherwise identical stem-cell type. The transplantation setting also differs substantially from ordinary development in an intact child.

The paper investigated a network involving gene regulation and macrophage metabolism. Its findings do not establish that a particular methylation change was the sole causal mechanism. The observed metabolic phenotype in mice must also remain separate from clinical type 2 diabetes in children, despite the broader wording of the publication’s title.

The human cord-blood component included 30 term mother: infant pairs and laboratory cell measurements. It did not measure the subsequent maturation of children’s immune systems or establish a later disease benefit from supplementation. Nor does this review interpret the animal dietary-switch comparisons as proof that early effects are irreversible. The study supports experimental developmental immune-cell programming; it does not establish a universal or permanent human trajectory. Vitamin D and Lasting Gene Regulation: Full Evidence Review examines persistence and the distinction between a lasting response and a molecular mark.

Pregnancy supplementation and asthma at six years

COPSAC and VDAART tested whether increasing vitamin D exposure during pregnancy could reduce childhood respiratory outcomes. Both are relevant to early-life trajectories because the intervention preceded the childhood assessment. However, asthma outcomes are not a direct assay of immune-cell maturation, and the intervention periods do not cover every stage of pregnancy and childhood.

In the COPSAC pregnancy trial, 623 women were assigned from approximately 24 weeks of pregnancy to an additional 2,400 international units daily or its comparator, with both groups receiving the usual 400 international units daily. The six-year publication assessed a follow-up of the same trial, rather than a new independent experiment. Asthma was recorded in 23 of 274 children in the higher-dose group and 18 of 268 in the comparison group.

The adjusted odds ratio was 1.21, with a 95% confidence interval from 0.63 to 2.32. This did not demonstrate a preventive benefit. The interval includes materially different possibilities, so the result should not be turned into proof of harm or exact equivalence. An odds ratio describes a comparison of odds, not a direct percentage-point change in the chance of disease.

Interpretation also depends on follow-up conditions. The trial had fewer recruits than planned, treatment allocation was unblinded after the three-year assessment, and losses were related to some social and smoking characteristics. The analysis accounted for the separate omega-3 allocation. These features do not erase the randomised comparison, but they affect precision and the extent to which the follow-up represents all originally enrolled families.

VDAART enrolled families at increased allergic risk because of parental allergic disease. Pregnant participants were assigned to a total daily study intake of 4,400 or 400 international units. There were 881 randomised pregnancies, with 806 children included in the six-year analysis. The outcome combined asthma or recurrent wheeze, so it should not be described as an identical measure to the COPSAC asthma endpoint.

By six years, that combined outcome occurred in 176 of 405 children in the higher-dose group and 184 of 401 in the comparison group. The trial did not show a statistically significant difference in its primary analysis, with P = .25. It also did not show a spirometric benefit. Spirometry measures aspects of airflow during breathing; it is a functional respiratory test rather than a direct immune measurement.

Caregiver reports contributed to the respiratory outcome, and the children’s later vitamin D exposure was not randomly assigned by this pregnancy intervention. Results therefore describe the consequences of the prenatal regimens under the observed follow-up conditions. They cannot answer whether a different start time, a different population or a sustained childhood intervention would produce the same result.

The full VDAART principal paper, its material tables and its first two figures were examined for this review. Its appendix was not accessible, so detailed supplementary sensitivity or safety claims are not made here. The six-year overall result is supported by the principal report. Neither trial’s earlier follow-up should be counted as another independent confirmation, and this review does not reconstruct detailed three-year trajectories from sources that were not fully assessed.

Infant supplementation and reported infections

VIDI tested vitamin D3 during infancy rather than pregnancy. The Finnish trial compared 400 with 1,200 international units daily from about two weeks until 24 months. It analysed 975 infants after 12 post-randomisation exclusions. Almost all infants were vitamin D sufficient at birth according to the study’s definition: 95.7%. This was therefore mainly a comparison in healthy, initially replete term infants, not a trial of treating severe deficiency.

The higher-dose group achieved higher vitamin D blood concentrations. That confirms a difference in exposure, but it does not itself establish a functional advantage. Vitamin D Testing: 25-OH-D vs 1,25-OH₂-D explains the blood measurements; From Cellular Cytokine Responses to Systemic Inflammation examines the related problem of connecting cellular responses to outcomes measured across the body. The clinically relevant question was whether outcomes also differed. Parent diaries contributed infection information from 449 infants in one group and 448 in the other, with 8,458 reported infection episodes in total.

The overall infection incidence-rate ratio was 1.00, with a 95% confidence interval from 0.93 to 1.06. An incidence-rate comparison accounts for events over observation time rather than merely the proportion of children who ever became ill. The result did not support fewer overall reported infections with the higher regimen in this setting.

These were reported episodes, not a complete set of laboratory-confirmed infections. The mostly Northern European, healthy term population also limits application to infants born early, severely deficient infants or settings with different nutritional and infection conditions. Post hoc seasonal findings cannot replace the trial’s overall comparison as the principal conclusion.

VIDI illustrates a central distinction for this subject: a higher status biomarker and an early-life intervention can be demonstrated without a measurable improvement in the selected illness outcome. It does not show that vitamin D has no role in immune biology. It shows that this higher dose did not improve the overall infection outcome compared with the standard regimen in the studied population.

Nor does the infection count describe every aspect of immune development. The trial provides a functional illness outcome during an important early period. It does not map the emergence of immune-cell subsets, the quality of each immune response or every later immune-related disease. Those would require their own measurements and follow-up.

Vitamin D status and islet autoimmunity

TEDDY examined children selected for increased genetic risk of type 1 diabetes. Its analysis related plasma 25-hydroxyvitamin D, a marker of vitamin D status, to the development of persistent islet autoantibodies. Autoantibodies are proteins produced by the immune system that recognise the body’s own components. Persistent islet autoantibodies indicate an autoimmune process, but they are not identical to a diagnosis of clinical diabetes.

The nested case: control analysis included 376 children with persistent autoantibodies and 1,041 controls from the larger cohort. The infant measurement analysis involved a somewhat smaller sample. The first blood samples were collected at three, six or nine months, rather than at birth. This matters when interpreting the study as evidence about timing: it cannot directly identify the effects of fetal exposure from those infant measurements.

Higher childhood vitamin D concentrations were associated with lower odds of autoimmunity. The adjusted odds ratio was 0.93 per five nanomoles per litre higher concentration, with a 95% confidence interval from 0.89 to 0.97. The corresponding infancy association was also inverse. The analysis accounted for genetic risk and ancestry, among other design features, but adjustment cannot guarantee removal of all confounding.

This was an observational association. Families were not randomly assigned to achieve different blood concentrations, and the estimate does not define a treatment target. A higher concentration can accompany differences in diet, supplements, health or other exposures. Even when measured before autoantibodies appear, it does not by itself show that raising the concentration would prevent autoimmunity.

DAISY provides important contrary evidence from a separate high-risk cohort in Denver. Investigators examined vitamin D intake and repeated blood measurements in relation to islet autoimmunity and progression to type 1 diabetes. The overall cohort included 2,644 children, with smaller subsets contributing particular dietary and blood analyses. These different denominators should not be combined into one implied complete dataset.

The longitudinal vitamin D status analysis did not show a clear association with autoimmunity: the hazard ratio was 1.12 per 20.27 nanomoles per litre, with a 95% confidence interval from 0.88 to 1.43. The nine-month analysis was also inconclusive and had a wider interval. Among children with autoimmunity, the analysis of progression to diabetes likewise did not establish an association.

A hazard ratio concerns the rate of an event occurring over follow-up and is not directly interchangeable with TEDDY’s odds ratio. Differences in exposure timing, sampling, assays, populations and statistical precision complicate a numerical comparison. The DAISY findings do not prove that an effect is impossible, but they prevent the TEDDY association from being presented as consistently confirmed across independent cohorts.

The DAISY principal report was assessed, but its missing-sample supplement was not available. This review therefore makes no claim that missing blood samples were demonstrably harmless to interpretation. A separate TEDDY paper on maternal supplement use was identified but not fully appraised. It belongs to the same broader programme and examines a different exposure; the child’s biomarker association must not be translated into a claim that maternal supplements prevent autoimmunity.

What can be concluded about a childhood trajectory?

The assessed evidence supports different conclusions at different levels. The mouse work demonstrates a developmental immune-cell programming phenomenon under experimental conditions. The pregnancy trials do not show a clear six-year respiratory prevention benefit from their higher regimens. The infant trial does not show fewer overall reported infections with its higher dose. The observational cohorts disagree about the relationship between measured status and islet autoimmunity.

These results should be integrated without treating them as votes on one universal question. A transplant experiment, an infant infection diary and a persistent autoantibody outcome measure different processes. A negative intervention result in initially replete infants cannot rule out every role of deficiency. A positive observational association cannot override a null trial addressing a different exposure and outcome.

The most important unresolved connection is between demonstrated experimental developmental effects and ordinary human immune maturation. The human studies here do not directly establish that connection. They constrain claims about particular illness or risk outcomes, while the experimental work supplies a limited biological proposition. Neither line of evidence justifies saying that vitamin D determines a child’s immune future.

A useful interpretation remains specific about timing, population and outcome. Early vitamin D biology can be relevant to developing immune cells, yet clinical benefit from a particular intervention still requires direct evidence. Childhood risk also remains a trajectory with multiple influences, rather than a fixed consequence inferred from a single vitamin D measurement.

Scientific sources

Oh J and colleagues. Embryonic vitamin D deficiency programs hematopoietic stem cells to induce type 2 diabetes. Nature Communications. 2023;14:3278. DOI: 10.1038/s41467-023-38849-z. Principal paper and material supplementary findings assessed; experimental and limited cord-blood evidence.

Brustad N and colleagues. High-Dose Vitamin D Supplementation During Pregnancy and Asthma in Offspring at the Age of 6 Years. JAMA. 2019;321:1003 to 1005. DOI: 10.1001/jama.2019.0052. Six-year report from COPSAC2010.

Litonjua AA and colleagues. Six-Year Follow-up of a Trial of Antenatal Vitamin D for Asthma Reduction. New England Journal of Medicine. 2020;382:525 to 533. DOI: 10.1056/NEJMoa1906137. Principal report assessed for overall results; appendix not accessible.

Rosendahl J and colleagues. Effect of Higher vs Standard Dosage of Vitamin D3 Supplementation on Bone Strength and Infection in Healthy Infants: A Randomized Clinical Trial. JAMA Pediatrics. 2018;172:646 to 654. DOI: 10.1001/jamapediatrics.2018.0602. VIDI principal report and material supplementary tables assessed.

Norris JM and colleagues. Plasma 25-Hydroxyvitamin D Concentration and Risk of Islet Autoimmunity. Diabetes. 2018;67:146 to 154. DOI: 10.2337/db17-0802. TEDDY principal report and supplementary tables assessed.

Simpson M and colleagues. No association of vitamin D intake or 25-hydroxyvitamin D levels in childhood with risk of islet autoimmunity and type 1 diabetes: the Diabetes Autoimmunity Study in the Young (DAISY). Diabetologia. 2011;54:2779 to 2788. DOI: 10.1007/s00125-011-2278-2. Principal report assessed; missing-sample supplement not accessible.

Silvis and colleagues. Maternal dietary supplement use and development of islet autoimmunity in the offspring: TEDDY study. Pediatric Diabetes. 2019;20:86 to 92. DOI: 10.1111/pedi.12794. Complementary maternal-exposure report identified but not fully appraised; not an additional independent cohort.

Related reading

Original sources

Source

Vitamins for children

All doses below describe research interventions.

Publication details

All doses below describe research interventions. These interventions do not establish that a higher dose is better. For current UK infant and child guidance, see NHS Vitamins for children. Infants born early require advice from their neonatal team.

Read more →
Source

Effect of Higher vs Standard Dosage of Vitamin D3 Supplementation on Bone Strength and Infection in Healthy Infants: A Randomized Clinical Trial

Published in JAMA Pediatrics (2018).

Publication details

Rosendahl J and colleagues. Effect of Higher vs Standard Dosage of Vitamin D3 Supplementation on Bone Strength and Infection in Healthy Infants: A Randomized Clinical Trial. JAMA Pediatrics. 2018;172:646 to 654. DOI: 10.1001/jamapediatrics.2018.0602. VIDI principal report and material supplementary tables assessed.

Read more →
Source

No association of vitamin D intake or 25-hydroxyvitamin D levels in childhood with risk of islet autoimmunity and type 1 diabetes: the Diabetes Autoimmunity Study in the Young (DAISY)

Published in Diabetologia (2011).

Publication details

Simpson M and colleagues. No association of vitamin D intake or 25-hydroxyvitamin D levels in childhood with risk of islet autoimmunity and type 1 diabetes: the Diabetes Autoimmunity Study in the Young (DAISY). Diabetologia. 2011;54:2779 to 2788. DOI: 10.1007/s00125-011-2278-2. Principal report assessed; missing-sample supplement not accessible.

Read more →