Study at a glance
Do vitamin D supplements change blood signs of inflammation?
This page compares studies linking vitamin D blood levels with later heart and circulation problems and trials testing changes in blood signs of inflammation. The main analysis followed 1,054 older adults assigned vitamin D3 or placebo within the VITAL trial. Most did not have severe vitamin D deficiency. A separate VITAL analysis examined blood samples after one year.
Which inflammation measurements changed with vitamin D3?
One inflammation-related blood marker, C-reactive protein measured with a high-sensitivity test, was lower with vitamin D3 after two years. The other three markers did not show significant reductions. After four years, none of the four markers showed a significant benefit. The exact size of the two-year change in this marker is uncertain because two ways of expressing the published result give slightly different percentages.
How reliable was the two-year inflammation finding?
The two-year result remained significant after allowing for tests of four markers in the main analysis. A separate analysis excluding some participants based on pill use did not meet that same standard. Fewer blood samples were available after four years, and many measurements of one marker could not be used. The study did not establish a group of people who clearly benefited more based on their starting vitamin D level.
Do these blood changes show fewer heart attacks or strokes?
The separate one-year analysis found no overall reduction in its three markers, and one rose. These were different participant samples, so the results do not prove a sequence of effects that starts, reverses and stops. Observational links between low vitamin D and later heart and circulation problems do not prove cause and effect. Changes in blood markers alone do not show fewer heart attacks or strokes. This evidence does not establish a broad, lasting reduction in inflammation or a supplement dose or blood target to protect the heart and circulation.
Read the full article below ↓Does a low vitamin D blood concentration identify cardiovascular risk, and does raising vitamin D reduce inflammatory markers? Those are separate questions. This page principally examines the two: and four-year VITAL biomarker analysis by Dong and colleagues, alongside an earlier VITAL analysis and observational background. The findings vary by marker and follow-up time, and do not establish that a biomarker change prevents cardiovascular events. For the shorter account, read Vitamin D Status, Inflammation and Cardiovascular Risk Markers: Study Summary.
Three measurements that should not be confused
Vitamin D status is commonly assessed using circulating 25-hydroxyvitamin D, or 25(OH)D. Inflammatory biomarkers measure selected features of biological activity. Clinical outcomes, such as myocardial infarction or stroke, measure events experienced by people. The heart physiology overview provides the wider setting, but a result at one of these levels cannot automatically stand in for a result at another.
The distinction between vitamin D status and vitamin D effect is central here. People with low concentrations may differ in health, behaviour and other exposures from people with higher concentrations. A randomised trial instead asks what changes when an intervention is assigned. Even then, the answer is specific to the population, dose, duration and outcome studied.
Observational background: low status and later cardiovascular events
Wang and colleagues followed 1,739 Framingham Offspring participants without prior cardiovascular disease. Their mean age was 59 years, 55% were women and all were white. Over a mean 5.4 years, 120 participants developed a first cardiovascular event. A baseline 25(OH)D concentration below 15 ng/mL was associated with a multivariable-adjusted hazard ratio of 1.62, with a 95% confidence interval of 1.11-2.36, compared with concentrations at or above 15 ng/mL.
This is a relative association over follow-up. It does not mean that 62% of deficient participants experienced an event, and it is not an estimate of the benefit of supplementation. Adjustment for measured risk factors improves the comparison but cannot remove every unmeasured difference or establish that low vitamin D itself caused the events. The guide to correlation and causation explains why this observational finding motivates intervention research without determining its result.
The principal VITAL analysis: who was studied and what was assigned
Dong and colleagues analysed 1,054 Boston-area participants from VITAL who attended in-person assessments. VITAL was a randomised, double-blind, placebo-controlled trial using a two-by-two factorial design. Participants could receive vitamin D3, marine omega-3 fatty acids, both active agents or both placebos. The vitamin D comparison therefore pooled people assigned active vitamin D across both omega-3 assignments and compared them with those assigned vitamin D placebo.
Vitamin D3 was given at 2,000 IU daily. The separate marine omega-3 capsule contained 1 g of oil, including 460 mg EPA and 380 mg DHA. These were the study regimens, not personalised recommendations. Randomisation was centrally generated within sex, race and age strata. Participants, investigators, care providers and endpoint adjudicators were blinded to assignment.
The parent trial enrolled men aged at least 50 and women aged at least 55 without a history of major cardiovascular events or most cancers. In the biomarker cohort, mean age was 64.9 years, 49% were women and approximately 84% were white. Mean baseline 25(OH)D was 28.1 ng/mL, and 17% had concentrations below 20 ng/mL. This was not a trial restricted to people with severe vitamin D deficiency or an inflammatory disease.
Of these participants, 520 were assigned active vitamin D3 and 534 vitamin D placebo. Background medication use was common: about 41% used hypertension medication and 35% cholesterol medication. The results therefore concern older community-dwelling adults with varied risk factors, rather than a laboratory experiment that isolates a single inflammatory pathway.
Which inflammatory markers were measured?
The study measured four serum markers. They describe different parts of the inflammatory system and should not be merged into one score without justification. The inflammatory signalling overview explains the broader communication network within which such blood measurements sit.
| Marker | Meaning in this study |
|---|---|
| High-sensitivity C-reactive protein, hs-CRP | A sensitive measurement of a circulating acute-phase protein associated with systemic inflammation |
| Interleukin-6, IL-6 | A cytokine involved in inflammatory signalling |
| Tumour necrosis factor-α, TNF-α | A cytokine involved in several inflammatory pathways |
| Interleukin-10, IL-10 | A cytokine with important anti-inflammatory actions |
A blood concentration is not a direct reading of every tissue’s inflammatory activity. The experiment did not measure endothelial barrier leakage, plaque stability or heart-muscle contractility. The endothelial function pathway is a related physiological domain, but it requires its own measurements. An inflammatory-marker result cannot substitute for a vascular-function experiment.
Sample collection and analysis affect the interpretation
Fasting blood was collected at baseline, year two and year four, processed promptly and stored at −80°C. The analysis used 2,713 participant samples: 1,051 at baseline, 977 at year two and 685 at year four. These counts are blood samples at different visits, not 2,713 independently randomised people. There were also separate quality-control samples used to assess assay performance.
The assays did not yield usable values for every sample. Results outside the quantification limits were excluded: 0.5% for hs-CRP, 29% for IL-6, 2% for IL-10 and 0.1% for TNF-α. The particularly large IL-6 exclusion matters when judging precision and comparability between markers. A null estimate based on fewer usable measurements does not have exactly the same evidential weight as one based on almost all samples.
Because marker concentrations were skewed, the researchers analysed their natural logarithms. Mixed-effects models accounted for repeated samples from the same participant. The central treatment estimate was an interaction between treatment assignment and time, testing whether the trajectory differed between groups. The fully adjusted model also included the other randomised treatment, age, sex, race, body mass index, smoking and several medication categories.
These statistical adjustments do not turn the analysis into a different randomised trial. They address measured imbalance and improve the specified comparison within this cohort. The authors described the work as a post hoc biomarker analysis of a parent trial that was originally designed around cancer and cardiovascular outcomes. Its purpose and inferential limits should be understood at that level.
The positive result was confined to hs-CRP at two years
At year two, the vitamin D3 group had a lower hs-CRP trajectory than placebo. In the fully adjusted model, the log-scale treatment-by-time coefficient was −0.17 with a standard error of 0.06. The nominal P value was 0.007. After adjustment for testing four biomarkers, the reported P value was 0.028. The other three inflammatory markers did not show significant vitamin D effects at this time point.
The paper describes the hs-CRP finding as a 19% reduction. There is a numerical reporting nuance: directly exponentiating the rounded coefficient of −0.17 gives a ratio of approximately 0.84, or about 16% lower, under the usual natural-log interpretation. That calculation is an appraisal of the published coefficient, not a reanalysis of participant data. The precise percentage should therefore be treated cautiously; the supported finding is a lower hs-CRP trajectory at two years.
At year four, the fully adjusted hs-CRP coefficient was −0.02 with a standard error of 0.07 and P=0.753. No other marker showed a significant vitamin D effect at year four. The study consequently does not demonstrate a sustained reduction across all four years or a general lowering of every inflammatory signal. A statistically nonsignificant estimate also does not establish exact equivalence between groups.
| Vitamin D comparison | Fully adjusted hs-CRP coefficient, standard error | Interpretation |
|---|---|---|
| Two years | −0.17, 0.06; nominal P=0.007 | Lower trajectory; remained significant after the reported four-marker correction |
| Four years | −0.02, 0.07; nominal P=0.753 | No statistically significant separation in the reported analysis |
Why adherence and subgroup findings need care
The investigators also analysed data after censoring follow-up from participants reporting low study-pill use or substantial outside supplementation. The two-year hs-CRP result retained its direction, but its nominal P value was 0.027 and its multiple-comparison-adjusted P value was 0.11. Thus, this sensitivity analysis did not meet the same corrected significance standard. It should not be described as equally strong confirmation of the primary analysis.
Removing participants according to behaviour after randomisation can also weaken the protection provided by random assignment. These analyses help assess robustness, but they are not automatically more reliable than the assigned-treatment comparison. The year-four result remained nonsignificant. The smaller number of available year-four samples adds another reason to examine uncertainty rather than assume a precise biological disappearance at a particular date.
The supplementary subgroup figure explored age, sex, race, smoking, body mass index, baseline nutrient status and medication use. Some within-subgroup comparisons reached nominal significance and others did not. However, the interaction tests did not establish that the vitamin D effect differed between the investigated subgroups. For baseline 25(OH)D below versus at least 20 ng/mL, the interaction P value was 0.47.
It would therefore be misleading to use that figure to identify a proven responder group or a blood concentration at which the intervention starts working. Statistical significance in one subgroup and nonsignificance in another is not itself a statistically significant difference between their effects. These analyses were exploratory and had limited power.
An earlier VITAL analysis found no overall anti-inflammatory reduction
Costenbader and colleagues examined a separate selected sample of 1,561 VITAL participants with paired baseline and one-year blood samples. They measured IL-6, hs-CRP and tumour necrosis factor receptor 2, or TNFR2. TNFR2 is a receptor-related marker of the TNF system; it is not the TNF-α concentration measured in Dong’s later analysis. The vitamin D regimen was again 2,000 IU daily within the factorial trial.
At one year, none of these three markers was significantly reduced by vitamin D. The reported net relative change in IL-6 was an increase of 8.19%, with a 95% confidence interval of 1.52%-15.31%. TNFR2 and hs-CRP did not show significant overall treatment effects. This finding does not establish clinical harm, just as a lower marker concentration would not by itself establish cardiovascular protection.
An exploratory pattern suggested a greater hs-CRP increase among participants with baseline 25(OH)D below 20 ng/mL. The study conducted numerous subgroup comparisons without a strict multiplicity correction. Such findings need confirmation and do not supply a treatment rule. The earlier samples were shipped for processing, whereas the Boston clinic samples were processed promptly on site; cohorts, assays and observation times also differed.
What these findings mean for cardiovascular interpretation
The one-, two: and four-year results are not a single uninterrupted measurement series in an identical sample. They should not be narrated as proof that vitamin D first raises inflammation, then lowers it, then stops working. Each analysis answers a defined question with its own participants, laboratory measurements and statistical uncertainty.
Biological hypotheses involving vitamin D receptor signalling remain relevant, but these trials did not directly demonstrate the molecular pathway responsible for an individual blood-marker change. They also did not establish that the hs-CRP finding mediated a change in cardiovascular-event risk. Demonstrating mediation would require additional evidence linking assignment, marker change and clinical outcomes within an appropriate analysis.
The evidence hierarchy helps keep the conclusions proportional. Observational low-status associations identify patterns worth investigating. Randomisation strengthens inference about assigned supplementation and the outcomes actually measured. Neither approach permits replacing a measured biomarker with an unmeasured clinical benefit.
The defensible conclusion is specific: a VITAL biomarker analysis found lower hs-CRP at two years, with qualifications about magnitude and sensitivity analyses, but not a sustained or broad anti-inflammatory effect across the measured markers. Together with the earlier null or upward findings, this evidence does not establish a cardiovascular supplement dose, a target vitamin D concentration or prevention of heart attacks and strokes.
Scientific sources
Dong Y, Zhu H, Chen L, Huang Y, Christen W, Cook NR, Copeland T, Mora S, Buring JE, Lee IM, Costenbader KH, Manson JE. Effects of Vitamin D3 and Marine Omega-3 Fatty Acids Supplementation on Biomarkers of Systemic Inflammation: 4-Year Findings from the VITAL Randomized Trial. Nutrients. 2022;14(24):5307. DOI: 10.3390/nu14245307. PMID: 36558465. Principal appraisal, including the supplementary subgroup figure.
Costenbader KH, MacFarlane LA, Lee IM, Buring JE, Mora S, Bubes V, Kotler G, Camargo CA Jr, Manson JE, Cook NR. Effects of One Year of Vitamin D and Marine Omega-3 Fatty Acid Supplementation on Biomarkers of Systemic Inflammation in Older US Adults. Clinical Chemistry. 2019;65(12):1508-1521. DOI: 10.1373/clinchem.2019.306902. PMID: 31699704. Earlier VITAL biomarker comparison.
Wang TJ, Pencina MJ, Booth SL, Jacques PF, Ingelsson E, Lanier K, Benjamin EJ, D’Agostino RB, Wolf M, Vasan RS. Vitamin D Deficiency and Risk of Cardiovascular Disease. Circulation. 2008;117(4):503-511. DOI: 10.1161/CIRCULATIONAHA.107.706127. PMID: 18180395. Observational background; the cohort finding is used to distinguish association from intervention evidence.
Related reading
3′‐Sialyllactose Prevents Atherosclerosis by Attenuating Chronic Inflammation via the Gut‐Immune‐Cardiovascular Axis in LDLR −/− Mice
Herein, 40 male low‐density lipoprotein receptor knockout (LDLR −/−) mice were randomized into five groups: normal diet (ND), high‐cholesterol diet (HCD), and three 3′‐SL groups (low dose SLL, medium dose SLM, high dose SLH).
Original paper ↗Read more →Association Between Vitamin D Deficiency and Cardiovascular Disease Risk Factors in the MENA Population: A Systematic Review and Meta-Analysis
The Middle East and North Africa (MENA) region faces a high cardiovascular disease (CVD) burden alongside endemic serum 25(OH)D (vitamin D) deficiency.
Original paper ↗Read more →Association of Oxidative Stress Markers with Cardio-Kidney-Metabolic Parameters and Cardiovascular Disease in Patients with Type 2 Diabetes Mellitus
We aimed to investigate the association between oxidative stress (OS), inflammation, and kidney function and the predictive ability of OS for mortality and cardiovascular disease in 143 patients with type 2 diabetes (T2DM) and various degrees of kidney function.
Original paper ↗Read more →Association of vitamin D status and body mass index with cardiometabolic risk and cardiovascular disease in adults
This study aimed to evaluate the association between serum vitamin D concentrations, body mass index (BMI), and cardiometabolic risk factors in adults.
Original paper ↗Read more →Can Vitamin D Reduce Inflammation? The Influence of Supplementation on Selected Immunological Markers
The aim of this study was to evaluate the effect of different doses of VitD on selected markers of inflammation in patients with relapsing: remitting MS (RRMS).
Original paper ↗Read more →Cardiac tissue oxidative stress and inflammation after vitamin D administrations in high fat: diet induced obese rats
The current study was aimed to evaluate the effects of vitamin D administration on the markers of oxidative stress and inflammation in the cardiac tissue of high-fat diet induced obese rats.
Original paper ↗Read more →Cumulative Inflammation and HbA1c Levels Correlate with Increased Intima-Media Thickness in Patients with Severe Hidradenitis Suppurativa
Hidradenitis suppurativa (HS) is a chronic inflammatory skin disease that has been associated with a greater risk of metabolic and cardiovascular comorbidities.
Original paper ↗Read more →Effects of Biologic Therapies and Narrowband UVB Phototherapy on Vascular Inflammation and Systemic Inflammatory Biomarkers in Psoriasis: A Systematic Review and Narrative Synthesis of Prospective Studies
We aimed to systematically synthesize prospective evidence on treatment-associated changes in vascular inflammation and systemic inflammatory biomarkers in adults with moderate-to-severe psoriatic disease.
Original paper ↗Read more →Effects of personalized vitamin D3 on inflammation in colorectal cancer patients: a randomized trial
In an ongoing randomized double-blind, placebo-controlled trial in Germany, CRC patients who underwent surgery in the past year and had serum 25-hydroxyvitamin D levels.
Original paper ↗Read more →IL-33, IL-37, and Vitamin D Interaction Mediate Immunomodulation of Inflammation in Degenerating Cartilage †
Additionally, in vitro studies with normal human chondrocytes were conducted to investigate the effect of calcitriol on the expression of IL-33, IL-37, IL-6, TNF-α, TLRs, DAMPs, and MMPs.
Original paper ↗Read more →Inflammation and Vitamin D Receptor Polymorphism: Impact on All-Cause and Cardiovascular Mortality in Mexican Women on Dialysis
This study aimed to determine how these factors affect mortality.
Original paper ↗Read more →Investigation of the Relationship Between Glycemic Control and Inflammation-Nutrition Indices in Older Adults with Type 2 Diabetes
Objective: To investigate the relationship between glycemic control and inflammation: nutrition indices in older adults with type 2 diabetes mellitus and to evaluate their prognostic value for 30-day mortality.
Original paper ↗Read more →Longitudinal associations between vitamin D status and biomarkers of inflammation in a pan-European cohort of children and adolescents
To investigate longitudinal associations between the vitamin D status and inflammatory markers in children and adolescents.
Original paper ↗Read more →Measuring Vitamin D Status in Chronic Inflammatory Disorders: How does Chronic Inflammation Affect the Reliability of Vitamin D Metabolites in Patients with IBD?
Evidence gained from recent studies has generated increasing interest in the role of vitamin D in extraskeletal functions such as inflammation and immunoregulation.
Original paper ↗Read more →New circulating biomarkers for predicting cardiovascular death in healthy population
Of 82 participants, 17 died by CVD causes.
Original paper ↗Read more →Predictors of coronary artery calcification and its association with cardiovascular events in patients with chronic kidney disease
CAC was evaluated via computed tomography (CT).
Original paper ↗Read more →Sex disparities in vitamin D status and the impact on systemic inflammation and survival in rectal cancer
Here we investigated whether the association between the vitamin D status and prognosis pertains to the general, unselected population of rectal cancer patients.
Original paper ↗Read more →The influence of persistent organic pollutants in the traditional Inuit diet on markers of inflammation
We aimed to assess the association between exposure to POPs from the marine diet and inflammation, taking into account other factors such as vitamin D.
Original paper ↗Read more →The Role of Vitamin D and Its Molecular Bases in Insulin Resistance, Diabetes, Metabolic Syndrome, and Cardiovascular Disease: State of the Art
In the last decade, an increasing awareness was directed to the role of Vitamin D in non-skeletal and preventive roles for chronic diseases.
Original paper ↗Read more →The vitamin D-cardiovascular nexus unraveling mechanistic promise and trial reality
In contrast, large-scale randomized trials and their meta-analyses have uniformly failed to demonstrate significant cardiovascular benefit from vitamin D supplementation in the general population.
Original paper ↗Read more →Vitamin D-rich marine Inuit diet and markers of inflammation: a population-based survey in Greenland
Participants were divided into three groups based on degree of intake of the traditional Inuit diet.
Original paper ↗Read more →Original sources
Effects of Vitamin D3 and Marine Omega-3 Fatty Acids Supplementation on Biomarkers of Systemic Inflammation: 4-Year Findings from the VITAL Randomized Trial
Background: The VITAL study was a nationwide, randomized, double-blind, placebo-controlled, 2 × 2 factorial trial of vitamin D3 (2000 IU/day) and marine n-3 FAs (1 g/day) supplements.
Publication details
Dong Y, Zhu H, Chen L, Huang Y, Christen W, Cook NR, Copeland T, Mora S, Buring JE, Lee IM, Costenbader KH, Manson JE. Effects of Vitamin D3 and Marine Omega-3 Fatty Acids Supplementation on Biomarkers of Systemic Inflammation: 4-Year Findings from the VITAL Randomized Trial. Nutrients. 2022;14(24):5307. DOI: 10.3390/nu14245307. PMID: 36558465. Principal appraisal, including the supplementary subgroup figure.
Effects of One Year of Vitamin D and Marine Omega-3 Fatty Acid Supplementation on Biomarkers of Systemic Inflammation in Older US Adults
Published in Clinical Chemistry (2019).
Publication details
Costenbader KH, MacFarlane LA, Lee IM, Buring JE, Mora S, Bubes V, Kotler G, Camargo CA Jr, Manson JE, Cook NR. Effects of One Year of Vitamin D and Marine Omega-3 Fatty Acid Supplementation on Biomarkers of Systemic Inflammation in Older US Adults. Clinical Chemistry. 2019;65(12):1508-1521. DOI: 10.1373/clinchem.2019.306902. PMID: 31699704. Earlier VITAL biomarker comparison.
Vitamin D Deficiency and Risk of Cardiovascular Disease
Published in Circulation (2008).
Publication details
Wang TJ, Pencina MJ, Booth SL, Jacques PF, Ingelsson E, Lanier K, Benjamin EJ, D’Agostino RB, Wolf M, Vasan RS. Vitamin D Deficiency and Risk of Cardiovascular Disease. Circulation. 2008;117(4):503-511. DOI: 10.1161/CIRCULATIONAHA.107.706127. PMID: 18180395. Observational background; the cohort finding is used to distinguish association from intervention evidence.