CVD is already the leading cause of death worldwide, claiming an estimated 17·9 million lives each year according to the WHO. In 2019 alone, 18·6 million people died and 34·4 million were disabled(Reference Roth, Mensah and Johnson1). Moreover, its prevalence is increasing, with the prevalence of high blood pressure estimated to increase from 51·2 % of USA adults in 2020 to 61·0 % in 2050. The prevalence of coronary artery disease (7·8 % to 9·2 %), heart failure (2·7 % to 3·8 %), stroke (3·9 % to 6·4 %), atrial fibrillation (1·7 % to 2·4 %) and total CVD (11·3 % to 15·0 %) will also increase(Reference Joynt Maddox, Elkind and Aparicio2).
Therefore, the prevention of CVD is particularly important, and the current measures are from the perspectives of nutrition and diet, exercise and physical activity(Reference Arnett, Blumenthal and Albert3). Among them, vitamin D regulates blood pressure, heart function, endothelial and smooth muscle cell function and therefore plays a role in cardiovascular health(Reference Wimalawansa4). A cross-sectional study showed that the prevalence of CVD in vitamin D deficiency was 74 %(Reference Kim, Sabour and Sagar5). Meanwhile, the D-Health trial concluded that vitamin D supplementation may reduce the incidence of major adverse cardiovascular events (MACE)(Reference Thompson, Waterhouse and English6). Also in a prospective cohort study, vitamin D supplementation in adults with hypertension reduced the risk of cardiovascular mortality(Reference Ye, Li and Liu7).
On the contrary, two recently published meta-analyses(Reference Mattumpuram, Maniya and Faruqui8,Reference Pei, Zhang and Peng9) showed no association between vitamin D supplementation and cardiovascular mortality, stroke, myocardial infarction (MI), arrhythmia or overall cardiovascular event risk. However, the effect sizes in these two articles were calculated using risk ratios, which are missing the time variable relative to the hazard ratio (HR). And they also had limitations included the relatively short follow-up time of the included literature (mostly less than 3 years), the fact that cardiovascular events were secondary outcomes in most trials and the absence of more detailed subgroup analyses. Incidentally, due to the long duration of the onset and progression of CVD itself and the fact that vitamin D supplements are primarily intended to improve the microenvironment and thus promote health, a longer follow-up is required to achieve this. Meanwhile, in the article by Rossello, X.(Reference Rossello, Pocock and Julian10), which summarises a number of studies on prophylactic medication in patients after MI, it is pointed out that most of the long-term follow-up considered in these articles is less than 3 years. By analogy, this study considers a follow-up of at least 3 years. According to the above reasons, the purpose of this study is to explore the relationship between vitamin D supplementation and incidence of MACE, based on high-quality randomised controlled trials (RCT).
Methods
Protocol and guidance
This study was conducted according to the guidelines of Preferred Reporting Items for Systematic Reviews and Meta-Analyse(Reference Page, McKenzie and Bossuyt11). In addition, this research protocol has been registered in the International Registry of Prospective Systems Reviews (PROSPERO identifier: CRD42024566447).
Literature retrieval
Pubmed, Web of science, Ovid, Cochrane Library and Clinical Trials were used to systematically search RCT on the topic of vitamin D intervention and the occurrence of cardiovascular events from the beginning of the database until July 2024, using the following headings and free words: ‘vitamin D’ AND (‘cardiovascular diseases’ OR ‘cardiovascular events’ OR ‘adverse cardiovascular’ OR ‘cardiovascular system’ OR ‘cardiac events’). Meanwhile, the references included in the literature are also searched. Grey literature was searched through Clinical Trials. Detailed strategies can be found in Supplemental Search Strategies. Finally, duplicate articles were removed by EndNote. Two authors (F. Y. Y. and Z. Q. F.) then independently screened all titles and abstracts of the remaining papers to identify potentially relevant articles. Subsequently, two authors assessed the full text of potentially relevant studies. Discrepancies were resolved by consensus or arbitrated by a third author (C. B. S.).
Inclusion and exclusion criteria
Inclusion criteria were as follows: (1) including participants > 18 years old, (2) the intervention group received any dose of vitamin D, (3) the control group received placebo treatment, (4) studies reported MACE or related outcomes and (5) the study belongs to RCT.
Exclusion criteria were (1) the follow-up time of trials was less than 2 years, (2) the published literature was not written in English, (3) updated articles were published and only the most recent or comprehensive data were included and (4) the full text or the data were not available.
Primary outcome and secondary outcomes definition
The primary outcome of this study was MACE. The definition of adverse cardiovascular events is complicated. According to previous review, 15·5 % of the literatures considered MACE to include acute MI and stroke; 13·8 % of the literatures included acute MI, stroke and all-cause death and 8·6 % of the literatures considered to include acute MI, stroke and CVD(Reference Bosco, Hsueh and McConeghy12). This study combines these ideas by defining MACE as studies that were included as long as they included MI and stroke. At the same time, the expanded MACE refers to the addition to the original MACE that was not previously considered (including but not limited to cardiovascular death or coronary revascularisation).
The secondary outcome of this study was MACE for specific diseases such as MI, stroke and all-cause death.
Data extraction and quality evaluation
Full-text assessment was performed independently by two authors (C. Z. H. and C. X. Q.), and data were extracted from the full text by two authors (F. Y. Y. and Z. Q. F.). Discrepancies were resolved by consensus or arbitrated by a third author (X.W.).
Data were extracted from the included articles using pre-set tables, including the basic information of the articles (authors, publication year, region, medication regimen, follow-up time and MACE evaluation criteria) and the patient’s basic characteristics (underlying disease history, age, sex, BMI and primary outcome).
The Cochrane Code assessment was carried out on a line-by-line basis in the following seven areas: (1) random sequence generation (selection bias), (2) allocation concealment (selection bias), (3) blinding of participants and personnel (performance bias), (4) blinding of outcome assessment (detection bias), (5) incomplete outcome data (attrition bias), (6) selective reporting (reporting bias) and (7) other bias. The appeals items were compared with the included studies one by one, and the corresponding basis was given. The evaluation results were high risk, unknown risk and low risk.
Statistical analysis
Review Manager (RevMan, version 5.3) and Stata (version 12.0) were used for statistical analysis. HR and 95 % CI were used to evaluate the association of vitamin D interventions with cardiovascular events. Note that when the study had multiple groups of vitamin D supplementation, they were all experimental groups and were combined. When HR < 1, the results support the experimental group; when HR > 1, support the control group. Chi-square tests are used to assess heterogeneity, using I2 to indicate that I2 < 25 % is low heterogeneity, 25 % ≤ I2 < 50 % is moderate heterogeneity, and I2 > 50 % is high heterogeneity. When I2 ≤ 50 %, the fixed-effect model was employed, and when I2 > 50 %, the random effects model was employed.
Subgroup analyses were performed based on sex and BMI. All tests were two sides, and P < 0·05 was considered statistically significant. When more than ten studies were included, sensitivity analysis was carried out, and Egger’s test was used to test for publication bias.
Results
Basic characteristics of the study and quality assessment
A total of 26 278 studies were retrieved from PubMed, Web of science, Ovid, Cochrane Library databases and Clinical Trials. After removing duplications and no additional channel supplementation, 22 365 studies were remained. By reading the titles and abstracts, 22 302 studies were excluded. Then, after reading sixty-three full texts, fifty-eight studies were removed, of which five were excluded because they were not RCT, twenty one were excluded because the data were not available and thirty two were excluded because they had no interesting outcomes (its outcome measure is not the HR, or results do not include incidence of MACE). Finally, five RCT(Reference Thompson, Waterhouse and English6,Reference Virtanen, Nurmi and Aro13–Reference Desouza, Chatterjee and Vickery16) were included, as shown in Figure 1.
Flow diagram of the selection process.
The five studies(Reference Thompson, Waterhouse and English6,Reference Virtanen, Nurmi and Aro13–Reference Desouza, Chatterjee and Vickery16) involved a total of 57 201 participants, including 29 019 in the intervention group and 28 182 in the control group, published between 2017 and 2023. The studies were conducted in two in the USA, one in New England, one in Australia and one in Finland. The intervention group received vitamin D supplementation (oral), while the control group received a placebo. The average follow-up period was 3–5·3 years. Additional baseline information for participants can be viewed in Table 1 and Supplemental Table 1.
Characteristics of all the studies included in the meta-analysis
RCT, randomised control trials; MACE, major adverse cardiovascular events.
For quality evaluation, among all RCT, 1 is at unclear risk of attrition bias, while others are low risk, as shown in online Supplementary Figures 1 and 2.
Primary outcome
Pooled results including all RCT(Reference Thompson, Waterhouse and English6,Reference Virtanen, Nurmi and Aro13–Reference Desouza, Chatterjee and Vickery16) showed no significant difference in the incidence of MACE between the vitamin D intervention group and the control group (HR: 0·96, 95 % CI: 0·89, 1·03; P = 0·77; I2 = 13·6 %; fixed-effect model), as shown in Figure 2(a).
Forest plot of the association between vitamin D supplementation and incidence of MACE ((a): MACE, P = 0·77, fixed-effect model; (b): expand MACE, P = 0·77, fixed-effect model). MACE, major adverse cardiovascular events.
In addition, the pooled results from three RCT(Reference Virtanen, Nurmi and Aro13,Reference Manson, Bassuk and Buring15,Reference Desouza, Chatterjee and Vickery16) , involving 30 789 participants, showed no significant difference between the vitamin D intervention group and the control group for expanded MACE incidence (HR: 0·96, 95 % CI: 0·86, 1·06; P = 0·77; I2 = 0 %; fixed-effect model), as shown in Figure 2(b).
Subgroup analysis
For sex, three RCT(Reference Thompson, Waterhouse and English6,Reference Virtanen, Nurmi and Aro13,Reference Manson, Bassuk and Buring15) were included, involving 49 668 participants. Subgroup analysis showed that among male participants, there was no significant difference in the incidence of MACE between the both groups (HR: 0·92, 95 % CI: 0·84, 1·02; P = 0·109; I2 = 14·1 %; fixed-effect model), as shown in Figure 3(a). Also, among female participants, there was no significant difference in the incidence of MACE between the both groups (HR: 0·95, 95 % CI: 0·82, 1·10; P = 0·468; I2 = 24·3 %; fixed-effect model), as shown in Figure 3(b).
Forest plots of the association between vitamin D supplementation and incidence of MACE in sex-specific subgroups ((a): male, P = 0·109, fixed-effect model; (b): female, P = 0·468, fixed-effect model). MACE, major adverse cardiovascular events.
For BMI, three RCT(Reference Thompson, Waterhouse and English6,Reference Virtanen, Nurmi and Aro13,Reference Manson, Bassuk and Buring15) were included, involving 49 668 participants. Subgroup analysis showed no significant difference in the incidence of MACE between the both groups among participants with BMI < 25 kg/m2 (HR: 1·04, 95 % CI: 0·80, 1·35; P = 0·782; I 2 = 51·7 %; random-effect model), as shown in Figure 4(a). However, among participants with BMI ≥ 25 kg/m2, the vitamin D intervention group had a lower incidence of MACE than the control group (HR: 0·91, 95 % CI: 0·83, 1·00; P = 0·055; I2 = 0 %; fixed-effect model), reduced MACE risk by 9 % but the difference was not statistically significant, as shown in Figure 4(b).
Forest plots of the association between vitamin D supplementation and incidence of MACE in BMI subgroups ((a): BMI < 25 kg/m2, P = 0·782, random-effect model; (b): BMI ≥ 25 kg/m2, P = 0·055, fixed-effect model). MACE, major adverse cardiovascular events.
Secondary outcomes
For MI and stroke, four RCT(Reference Thompson, Waterhouse and English6,Reference Virtanen, Nurmi and Aro13–Reference Manson, Bassuk and Buring15) were included, involving 54 778 participants. Subgroup analysis showed that the vitamin D intervention group had a lower incidence of MI than the control group, but the difference was not statistically significant (HR: 0·88, 95 % CI: 0·77, 1·01; P = 0·061; I2 = 0 %; fixed-effect model), as shown in Figure 5(a). In addition, there was no significant difference in the incidence of stroke between the both groups (HR: 0·97, 95 % CI: 0·84, 1·12; P = 0·675; I2 = 0 %; fixed-effect model), as shown in Figure 5(b).
Forest plots of the association between vitamin D supplementation and incidence of different cardiovascular outcome in cardiovascular event subgroups ((a): MI, P = 0·061, fixed-effect model; (b): stroke, P = 0·675, fixed-effect model; (c): cardiovascular death, P = 0·422, fixed-effect model). MI, myocardial infarction.
For cardiovascular death, two RCT(Reference Virtanen, Nurmi and Aro13,Reference Manson, Bassuk and Buring15) were included, involving 28 366 participants. Subgroup analyses showed no significant difference in cardiovascular death rates in the vitamin D intervention group compared with the control group (HR: 1·10, 95 % CI: 0·87, 1·38; P = 0·422; I2 = 0 %; fixed-effect model), as shown in Figure 5(c).
Discussion
This meta-analysis was based on five high-quality RCT that assessed the effect of vitamin D supplementation on the incidence of MACE for primary prevention. Some results of this study were consistent with previous meta-analyses, which showed that vitamin D supplementation had no significant effect on MACE incidence, extended MACE incidence, cardiovascular mortality or stroke incidence. However, subgroup analysis found that the effect of vitamin D supplementation on the incidence of MACE was influenced by BMI. Overweight/obese participants (BMI ≥ 25 kg/m2) benefit, while men, women and non-overweight/obese participants (BMI < 25 kg/m2) do not. In addition, the study also found a possible association between vitamin D supplementation and a reduction in the incidence of MI.
Previous studies have shown that vitamin D plays an important protective role in cardiovascular effects. The mechanism may be as follows: (1) in the renin-angiotensin system, angiotensin II causes significant constriction of small arteries throughout the body and increases blood pressure, which can lead to hypertension and increased cardiac workload. Vitamin D promotes the expression of angiotensin converting enzyme-2, which breaks down angiotensin II into angiotensin 1–7. Vitamin D also counteracts excess angiotensin II, thereby enhancing the antihypertensive, antifibrotic and anti-inflammatory properties of angiotensin 1–7(Reference Cui, Xu and Li17). In one experiment, renin expression and plasma angiotensin II production were increased several-fold in vitamin D receptor (VDR)-deficient mice, while 1,25(OH)2D3 injections resulted in renin inhibition(Reference Li, Kong and Wei18). (2) In the immune system, 1,25(OH)2D3 inhibited IL-12 production by dendritic cell while inducing IL-10 production. 1,25(OH)2D3 was also observed to act in T cells, which inhibits the production of IL-2, IL-17 and interferon-ϒ and attenuates the cytotoxic activity and proliferation of cluster of differentiation 4 and cluster of differentiation 8 T cells(Reference Mora, Iwata and von Andrian19). Thereby maintaining the stability of endothelial and vascular smooth muscle cells. (3) In the fibrinolytic system, it has also been shown that plasminogen activator inhibitor-1 is a major mediator involved in CVD, and 1,25(OH)2D3 reduces CVD by blocking NF-κB activation to downregulate plasminogen activator inhibitor-1(Reference Chen, Kong and Sun20).
Meanwhile, vitamin D deficiency may be a risk factor for CVD. A meta-analyse involving up to 180 000 individuals have shown that low vitamin D levels are associated with a significantly increased risk of hypertension, cardiovascular events and cardiovascular mortality(Reference Zhang, Li and Gao21). And observational studies have shown associations between serum 25(OH)D levels and general health, CVD risk factors and CVD mortality(Reference Mandarino, Júnior and Salgado22,Reference Park and Lee23) . In several mouse studies, matrix metalloproteinase, a protein responsible for abnormal remodelling of cardiomyocytes after injury and atherosclerosis, was found to be upregulated in VDR knockout mice(Reference Rahman, Hershey and Ahmed24), and cardiac diastolic and systolic forces were impaired and left ventricular hypertrophy developed in VDR knockout mice(Reference Tishkoff, Nibbelink and Holmberg25). Finally, it has also been shown that dietary vitamin D is insufficient to accelerate calcification in the vascular system of low-density lipoprotein receptor knockout and wild-type mice, and that vascular calcification is reduced after increased vitamin D supplementation(Reference Schmidt, Brandsch and Schutkowski26). It was also found that in patients with chronic kidney disease, active vitamin D may have a beneficial effect on vascular calcification by increasing the secretion of the anti-ageing factor Klotho and arterial medial bone bridging protein(Reference Lau, Leaf and Hu27).
However, the present study found that vitamin D supplementation was not helpful for the incidence of MACE in the general population or for the extended MACE incidence. Previous meta-analysis(Reference Zhang, Li and Gao21,Reference Wang, Song and Manson28) of dose–response analyses of circulating 25(OH)D concentrations found that a non-linear relationship was observed between 25(OH)D and CVD, with a plateau between 20 and 30 ng/ml (50 and 75 nmol/l), and that in populations below 30 ng/ml, the lower the vitamin D level, the greater the risk of CVD and the greater the benefit of vitamin D supplementation. However, vitamin D supplementation had little effect on people with levels above 30 ng/ml. This may be because oversupplementation does not convert all the vitamin D into active vitamin D. Experiments have shown that 1,25(OH)2D3 inhibits the hepatic synthesis of its precursor 25-OHD, preventing further conversion of vitamin D into 1,25(OH)2D3 (Reference Bell, Shaw and Turner29). It is also possible that because there is a limited amount of VDR in the body, in people who are not vitamin D deficient, the excess vitamin D does not have receptors to bind to efficiently and is converted to 24,25(OH)2D3, which has little binding to the VDR(Reference Tebben, Milliner and Horst30). It is also possible that when the concentration of 1,25(OH)2D3 in the body is high, the level of vitamin D-24-hydroxylase(Reference Hahn, Baker and Laslo31), the enzyme that breaks down and metabolises it, increases and converts it to inactive vitamin D, thus maintaining the dynamic balance of active vitamin D in the body. In our study, mean baseline 25(OH)D levels ranged from 24·4 to 30·9 ng/ml (non-vitamin D-deficient population), suggesting that additional supplements in the non-vitamin D-deficient population may not reduce the risk of MACE.
MACE include a wide range of cardiovascular outcomes, mainly MI, stroke and cardiovascular death. Subgroup analyses have been performed that show a protective effect of vitamin D against MI. Higher serum 25(OH)D levels were associated with a lower prevalence of MI than lower levels in both case–control studies and showed that the optimal 25(OH)D level should be at least 30 ng/ml to reduce the risk of MI(Reference Scragg, Jackson and Holdaway32). This phenomenon may be mediated by vasodilatory effects, improved endothelial and vascular smooth muscle cell function(Reference Wakasugi, Noguchi and Inoue33,Reference Artaza and Norris34) . Also, through a number of pathways vitamin D inhibits matrix metalloproteinase, the enzymes that are important in plaque destabilisation, especially matrix metalloproteinase-9 and matrix metalloproteinase-2 which are increased in MI(Reference Timms, Mannan and Hitman35). However, vitamin D does not reduce cardiovascular death or stroke. A meta-analysis of all-cause mortality found no further reduction in cardiovascular mortality when vitamin D levels exceeded 36 ng/ml(Reference Garland, Kim and Mohr36). This may be due to the existence of competition for death, with MI accounting for only about 10 % of all-cause deaths(Reference Benjamin, Virani and Callaway37), and the difference was not statistically significant due to the limited sample size. Stroke is divided into ischaemic and haemorrhagic strokes, with haemorrhagic strokes mainly associated with cerebral haemorrhage and subarachnoid haemorrhage and ischaemic strokes mainly associated with atrial fibrillation and carotid plaque. No significant association has been found between vitamin D supplementation and cerebral and subarachnoid haemorrhage, so vitamin D supplementation is not associated with haemorrhagic stroke. There are conflicting conclusions regarding the association of vitamin and others suggesting that there is no correlation(Reference Chen, Liu and Xu38,Reference Blondon, Sachs and Hoofnagle39) . Studies have also reported a non-linear positive association between serum 25OHD levels and carotid intima-media thickness in participants with adequate serum 25(OH)D (≥ 50 nmol/l)(Reference van Dijk, Sohl and Oudshoorn40). Similarly, the effect of vitamin D supplementation on atrial fibrillation is inconclusive, with some studies finding no association between low levels of 25(OH)D (< 20 ng/ml) and the incidence of atrial fibrillation(Reference Vitezova, Cartolano and Heeringa41). Therefore, the effect of vitamin D supplementation on ischaemic stroke is not clear.
Similarly, since vitamin D supplementation has no significance for the total population, there is no point in distinguishing between the sexes. However, it is worth exploring the fact that the included studies did not differentiate between female subjects with menopause. Then again, it is well known that oestrogen has a protective effect on lipids(Reference Haffner and Valdez42) and blood pressure(Reference Xue, Pamidimukkala and Lubahn43,Reference Vlachovsky, Di Ciano and Oddo44) . This could then interfere with the experimental results and lead to some bias. Second, the studies included different races, which may also affect the role of vitamin D.
Although vitamin D supplementation had no effect on the overall population or by sex, BMI subgroup analyses showed some significant results. It is well known that obesity is one of the cardiovascular risk factors(Reference Ng, Fleming and Robinson45). Our study analysed BMI in subgroups, and it is noteworthy that vitamin D supplementation reduced the risk of MACE in overweight/obese people. The reasons for this are (1) obesity itself is an independent cardiovascular risk factor(Reference Afshin, Forouzanfar and Reitsma46), and obese patients often have comorbid metabolic diseases such as hypertension, type 2 diabetes mellitus(Reference Martos-Moreno, Martínez-Villanueva and González-Leal47), hyperuricaemia(Reference Kubota48) and fatty liver disease(Reference Anderson, Howe and Jones49). In this high-risk group, vitamin D supplementation has helped to reduce the risk of MACE. (2) Obesity is associated with vitamin D deficiency(Reference López-Galisteo, Gavela-Pérez and Mejorado-Molano50), which may be related to four aspects: first, obese people have short exercise time, short sunshine and limited synthesis of vitamin D(Reference Gutiérrez Medina, Gavela-Pérez and Domínguez-Garrido51); second, volume dilution(Reference Drincic, Armas and Van Diest52), obese people need more vitamin D, vitamin D is relatively insufficient and then it may be associated with people living with obesity throughout. The decrease in vitamin D intake is due to a decrease in their intake of oily fish and dairy products(Reference Benedik53); finally, the bioavailability of vitamin D is lower in obese people than in normal-weight people because they have more deposits in fat tissue that they cannot use(Reference Wortsman, Matsuoka and Chen54). It has also been shown experimentally that the expression levels of vitamin D 25-hydroxylase CYP2J2 and vitamin D 1α-hydroxylase CYP27B1 in the subcutaneous adipose tissue of obese women are 71 % and 49 % lower(Reference Wamberg, Christiansen and Paulsen55), respectively, than in lean women. Therefore, vitamin D deficiency is more common in the obese population.
For the five RCT(Reference Thompson, Waterhouse and English6,Reference Virtanen, Nurmi and Aro13–Reference Desouza, Chatterjee and Vickery16) included in the study, some of the research designs and research results can still be thought about for their impact on this study. First, adherence was high in all five studies, with most adherence around 80 %. However, there were still some trial patients who did not participate in the surveys or had poor adherence, which could have biased the results. Second, the D2d trial(Reference Desouza, Chatterjee and Vickery16) included patients with pre-diabetes, which may have affected the results. Finally, the D-Health trial(Reference Thompson, Waterhouse and English6) did not measure baseline vitamin D levels, but rather created a model to predict them, which could also have biased the results. Based on these analyses, it is possible that these confounding factors and the effect of adherence contributed to the fact that the primary outcome of this study was not statistically significant.
However, when in the BMI subgroup, the presence of the weight factor modifies the degree of association between vitamin D supplementation and MACE, making it present a benefit of supplementation in the overweight/obese population and vice versa no benefit. This may be related to the previously stated idea that obesity itself is one of the CVD risks, volume dilution, poor diet, limited vitamin D synthesis and low vitamin D utilisation in these populations. In contrast, in the sex subgroup, the sex factor relatively did not change the association between vitamin D supplementation and MACE, so the results were still not statistically significant. So more and larger, multicentre RCT are needed to troubleshoot these confounding factors and further explore and validate ideas.
For all we know, this is the first meta-analysis to comprehensively and systematically explore the relationship between vitamin D supplementation and the incidence of MACE using HR and 95 % CI as an indicator for evaluating the relationship, which could be informative about the value of vitamin D supplementation in the prevention of MACE. Importantly, meta-analysis was based on recent (2017–2023), high-quality RCT. However, there are some limitations to this meta-analysis. First, the limited number of studies did not allow for further subgroup analyses. Meanwhile, the statistical efficacy of the study may also be weak due to the small number of included literature, which makes it difficult to perform publication bias assessment and sensitivity analysis. Second, this meta-analysis did not have the means to examine the dose–effect relationship and the effect of supplementation duration on the incidence of MACE. Finally, a small number of unavailable articles and non-English articles were excluded, which may have an increased risk of publication bias and selection bias.
Conclusion
The meta-analysis showed that vitamin D supplementation did not significantly contribute to the risk reduction of MACE and expanded MACE in the general population, as well as to the reduction of stroke risk, and cardiovascular death. Notably, vitamin D supplementation was found to reduce the incidence of MACE in overweight/obese populations (BMI ≥ 25 kg/m2), but not in non-overweight/obese populations (BMI < 25 kg/m2). Vitamin D should be recommended for overweight/obese people. Vitamin D supplementation was also found to be potentially helpful in reducing the incidence of MI in the general population. Unfortunately, vitamin D supplementation did not have a significant protective effect for the entire population, but it is still worth continuing to study the effects for different populations. Finally, due to the limited number of RCT included and the restricted experimental population, the findings may not be fully applicable to the entire population, more multicenter RCT with larger-scale were needed to further confirm the conclusion.
Supplementary material
For supplementary material/s referred to in this article, please visit https://doi.org/10.1017/S0007114525103954
Acknowledgements
None
No financial support was received for this study.
Each author contributed significantly to the conception and development of the present paper. X. C. and Z. C. designed the research process. Y. F. and Q. Z. searched the database for corresponding articles and extracted useful information from the articles above. B. C. and W. X. used statistical software for analysis. Z. P. and L. G. drafted the meta-analysis. All authors have read and approved the final manuscript.
The authors declare no conflict of interest.
Ethics approval and consent to participate: Not applicable, as this paper is based on research from global databases.
The datasets supporting the conclusions of this article are included within the article. For further details, please contact the corresponding author.
