Abstract
Background
Evidence linking vitamin D deficiency (VDD) to varicose veins remains limited and largely derived from small or cross-sectional studies. We examined whether VDD is associated with incident varicose veins in a large real-world cohort.
Methods
Using the TriNetX Global Collaborative Network, we conducted a retrospective propensity score–matched cohort study of adults aged ≥40 years with at least two serum 25-hydroxyvitamin D [25(OH)D] measurements between 2010 and 2023. VDD was defined as two measurements ≤19.9 ng/mL, and controls had two measurements ≥30.0 ng/mL. A 1-year landmark design was applied. The primary outcome was 5-year incident varicose veins of the lower extremities. Secondary outcomes included varicose veins with ulceration, inflammation, and other complications. Cox proportional hazards models were used to estimate hazard ratios (HRs) with 95% confidence intervals (CIs).
Results
After matching, 257,651 patients were included in each cohort. VDD was associated with a higher risk of incident varicose veins (HR 1.62, 95% CI 1.50–1.75; p < 0.001). Increased risks were also observed for varicose veins with ulceration (HR 2.22), inflammation (HR 2.71), and complications (HR 1.45) (all p < 0.001). Findings were consistent across sensitivity and subgroup analyses. Vitamin D insufficiency (20.0–29.9 ng/mL) was also associated with increased risk (HR 1.56, p < 0.001).
Conclusion
VDD was associated with an increased risk of incident varicose veins in this large cohort. This association was also observed among older adults and individuals with diabetes, suggesting potential effect modification by age and diabetic status. These findings suggest that reduced vitamin D status may be relevant to venous disease risk, although causality cannot be inferred. Further prospective studies are needed to determine whether correction of VDD can modify this risk.
Keywords: chronic venous disease, cohort study, propensity score matching, TriNetX, varicose veins, vitamin D deficiency
1. Introduction
Varicose veins of the lower extremities are among the most common manifestations of chronic venous disease and affect approximately 23% of adults in the United States (1, 2). Although often perceived as a cosmetic condition, varicose veins are associated with leg symptoms, impaired quality of life, and progression to more advanced venous disease, including skin changes, superficial thrombophlebitis, and venous ulceration (3–5). Established risk factors include older age, female sex, pregnancy, obesity, prolonged standing, and genetic susceptibility (6–8). Despite the recognition of these risk factors, the potential contribution of nutritional and metabolic factors to varicose vein development remains incompletely understood.
Vitamin D deficiency (VDD), commonly defined as a serum 25-hydroxyvitamin D [25(OH)D] concentration below 20 ng/mL, is highly prevalent worldwide and has been estimated to affect more than one billion people (9, 10). Beyond its classical role in calcium homeostasis and skeletal health, vitamin D has been implicated in vascular biology through effects on endothelial function, oxidative stress, inflammatory signaling, and regulation of the renin-angiotensin system. These mechanisms are biologically relevant to venous wall remodeling and chronic venous disease. In addition, recent ex vivo evidence has suggested that active vitamin D may attenuate oxidative stress and modulate nitric oxide synthase-related pathways in varicose venous tissue (11), providing a plausible mechanistic rationale for a link between vitamin D status and venous disease.
However, clinical epidemiological evidence on the association between VDD and incident varicose veins remains limited. Existing studies appear to be predominantly small-scale, cross-sectional, or mechanistic in nature, and a Mendelian randomization study evaluating genetically predicted circulating 25-hydroxyvitamin D did not support a clear association with varicose vein risk (12). Accordingly, whether clinically defined VDD is associated with the subsequent development of varicose veins in real-world populations remains uncertain. To address this gap, we conducted a propensity score-matched cohort study using the TriNetX Global Collaborative Network to examine the association between VDD and the 5-year risk of incident varicose veins of the lower extremities, and to further explore whether a gradient pattern exists across vitamin D status categories.
2. Methods
2.1. Data sources and exposure definition
This retrospective cohort study was conducted using electronic health record data from the TriNetX Global Collaborative Network, a federated real-world data platform comprising 171 healthcare organizations across multiple countries (13). The TriNetX database has been widely used in real-world clinical research across multiple medical specialties (14–16). The study was approved by the Institutional Review Board of Chi Mei Medical Center, with a waiver of informed consent, and was conducted in accordance with the Declaration of Helsinki. Because all data were de-identified and HIPAA-compliant, individual patient consent was not required.
Eligible participants were adults aged ≥40 years who had at least two serum 25-hydroxyvitamin D [25(OH)D] measurements on record between January 1, 2010, and December 31, 2023. Two mutually exclusive cohorts were formed based on 25(OH)D status. The VDD cohort included individuals with two distinct serum 25-hydroxyvitamin D [25(OH)D] measurements of ≤19.9 ng/mL, with the second qualifying value designated as the index date. To confirm exposure stability, patients in the VDD group were excluded if any 25(OH)D value ≥30.0 ng/mL had been recorded in the 3 years preceding the index date. The reference cohort comprised patients with two separate 25(OH)D measurements of ≥30.0 ng/mL, with the index date assigned in the same manner. Individuals in the reference cohort were excluded if any 25(OH)D value of ≤19.9 ng/mL had been documented during the same three-year look-back period.
2.2. Exclusion criteria
Patients with a pre-existing diagnosis of varicose veins of the lower extremities were excluded to ensure incident-only ascertainment. A one-year landmark period was further applied, during which newly coded varicose vein diagnoses served as an additional exclusion criterion to reduce reverse causation. A 1-year landmark period was selected, given the chronic and slowly progressive nature of varicose vein development. Patients were also excluded if they had prior diagnoses of conditions that could independently influence vitamin D metabolism, venous hemodynamics, or venous disease risk, including advanced renal disease, prior bariatric surgery, severe hepatic dysfunction, inflammatory bowel disease, malabsorptive disorders, and selected cardiovascular or venous comorbidities. Complete code-level definitions for all exclusion criteria are provided in Supplementary Table 1.
2.3. Data collection and propensity score matching
Baseline characteristics were extracted across four domains: demographics, clinically relevant comorbidities, medication use, and laboratory indices. Variables were selected based on their known or biologically plausible role as confounders of the exposure–outcome relationship. Of particular relevance to varicose vein pathogenesis, the matching model explicitly incorporated obesity, pregnancy history, use of exogenous estrogens and systemic contraceptives, nicotine dependence, and anticoagulant use; variables known to affect vitamin D metabolism or status—including glucocorticoid use and vitamin D supplementation—were additionally included to ensure comparability of the underlying nutritional milieu. A full list of covariates is provided in Supplementary Table 1. One-to-one propensity score matching (PSM) was performed using a greedy nearest-neighbor algorithm with a caliper of 0.1 standard deviations of the logit-transformed propensity score. Covariate balance was assessed using standardized mean differences (SMDs), with values <0.10 indicating adequate balance.
2.4. Primary and secondary outcomes
The primary outcome was incident varicose veins of the lower extremities (ICD-10-CM I83), ascertained over the subsequent 4-year follow-up period (days 365–1825), within a total 5-year observation window from the index date. Secondary outcomes included clinically meaningful subtypes: varicose veins with ulceration, varicose veins with inflammation, and varicose veins with other complications. Detailed code-based definitions for these outcomes are provided in Supplementary Table 1.
2.5. Validation of study design
Two prespecified control outcomes were used to assess the validity of the study design. Osteoporotic fracture served as a positive control because vitamin D has a well-established role in calcium homeostasis and bone mineralization; therefore, an increased risk in the VDD cohort would support the internal validity of the analysis. Appendicitis was selected as a negative control because its pathogenesis is not biologically linked to vitamin D status; accordingly, a null association would reduce concern about substantial unmeasured confounding or systematic detection bias. In addition, we compared the overall rate of healthcare encounters between the matched cohorts to evaluate whether they had similar levels of medical contact, and thus comparable opportunities for outcome detection, during follow-up.
2.6. Sensitivity and subgroup analyses
Three prespecified sensitivity analyses were performed to examine the robustness of the primary findings against potential bias. Model I limited the study period to 2016–2023 to ensure coding consistency after full adoption of ICD-10-CM. Model II excluded patients who died during follow-up as a sensitivity analysis to assess whether competing mortality materially influenced the observed association. Model III included only patients with at least one ambulatory encounter during follow-up to address possible underascertainment of outcomes among clinically inactive individuals. Prespecified subgroup analyses were also conducted according to sex, age (40–65 vs. >65 years), obesity, hyperlipidemia, history of cancer, diabetes mellitus, and hypertension. Within each subgroup, propensity score matching and time-to-event analyses were repeated using the same analytic framework as in the primary analysis, and interaction testing was performed to assess potential effect modification.
2.7. Additional analysis
To examine whether a less severe reduction in vitamin D status was likewise associated with varicose vein risk, and to assess the presence of a possible exposure–response pattern, we performed an additional analysis using a vitamin D insufficiency (VDI) cohort. VDI was defined as two separate serum 25-hydroxyvitamin D [25(OH)D] measurements of 20.0–29.9 ng/mL. This cohort was compared with the same vitamin D–sufficient reference group, defined by serum 25(OH)D levels of at least 30.0 ng/mL. To preserve comparability with the primary analysis, the same eligibility criteria, exclusion criteria, propensity score matching variables, and statistical methods were applied.
2.8. Statistical analysis
Time-to-event relationships were examined using Cox proportional hazards regression models, and the results are presented as hazard ratios (HRs) with corresponding 95% confidence intervals (CIs). The proportional hazards assumption for each model was evaluated using Schoenfeld residuals. Kaplan–Meier analysis was used to illustrate event-free survival over time, and differences between groups were assessed with the log-rank test. In addition to the primary propensity score–matched analysis, a multivariable Cox regression model was performed in the unmatched cohort to examine the independent association between VDD and outcomes. For the primary outcome, E-values were additionally calculated to estimate the minimum strength of an unmeasured confounder that would be necessary to fully explain the observed association. Statistical significance for the primary analysis was defined as a two-sided p value <0.05. In contrast, analyses of secondary outcomes, sensitivity analyses, subgroup analyses, and design-validation outcomes were considered exploratory in nature and were therefore interpreted without adjustment for multiple comparisons. All statistical analyses were performed directly within the TriNetX platform, and missing data were handled as recorded in the source database, without imputation. Competing risk analysis was not performed due to platform limitations of TriNetX.
3. Results
3.1. Baseline characteristics
Following the application of eligibility and exclusion criteria, 291,819 patients with VDD and 999,076 individuals with sufficient vitamin D levels were identified (Supplementary Figure 1). After 1:1 propensity score matching, 257,651 patients were retained in each cohort. Prior to matching, the two groups differed substantially across multiple covariates, most notably age, race, obesity, and medication use. After matching, covariate balance was achieved across all variables, with all SMDs falling below 0.10 (Table 1). The median follow-up duration was 1,825 days in both cohorts, with mean follow-up of 1,573 days and 1,583 days in the VDD and control groups, respectively.
Table 1.
Baseline characteristics of patients with vitamin D deficiency and sufficient vitamin D levels.
| Variables | Before matching | After matching | ||||
|---|---|---|---|---|---|---|
| VDD group (n = 291,819) | Control group (n = 999,076) | SMD | VDD group (n = 257,651) | Control group (n = 257,651) | SMD | |
| Patient characteristics | ||||||
| Age at index (years) | 53.5 ± 14.1 | 60.8 ± 13.3 | 0.535 | 54.8 ± 14.1 | 55.2 ± 13.5 | 0.025 |
| BMI ≥ 30 (kg/m2) | 88,828 (30.4) | 221,276 (22.1) | 0.189 | 71,448 (27.7) | 74,372 (28.9) | 0.025 |
| Female | 197,915 (67.8) | 740,104 (74.1) | 0.138 | 177,282 (68.8) | 175,000 (67.9) | 0.019 |
| White | 133,258 (45.7) | 762,684 (76.3) | 0.663 | 131,917 (51.2) | 130,144 (50.5) | 0.014 |
| Black or African American | 68,619 (23.5) | 75,483 (7.6) | 0.452 | 47,038 (18.3) | 47,616 (18.5) | 0.006 |
| Asian | 13,039 (4.5) | 42,175 (4.2) | 0.012 | 12,645 (4.9) | 11,954 (4.6) | 0.013 |
| Comorbidities and healthcare utilization | ||||||
| Essential (primary) hypertension | 84,602 (29.0) | 332,838 (33.3) | 0.093 | 73,077 (28.4) | 75,064 (29.1) | 0.017 |
| Hyperlipidemia | 72,866 (25.0) | 369,025 (36.9) | 0.261 | 66,595 (25.8) | 68,300 (26.5) | 0.015 |
| Encounter for general examination | 59,626 (20.4) | 278,385 (27.9) | 0.174 | 53,587 (20.8) | 53,394 (20.7) | 0.002 |
| Neoplasms | 51,994 (17.8) | 233,505 (23.4) | 0.138 | 46,976 (18.2) | 46,508 (18.1) | 0.005 |
| Other nutritional deficiencies | 48,161 (16.5) | 272,016 (27.2) | 0.262 | 45,501 (17.7) | 47,700 (18.5) | 0.022 |
| Overweight and obesity | 48,813 (16.7) | 109,354 (10.9) | 0.168 | 37,923 (14.7) | 38,945 (15.1) | 0.011 |
| Diabetes mellitus | 40,958 (14.0) | 119,645 (12.0) | 0.061 | 33,924 (13.2) | 34,811 (13.5) | 0.010 |
| Gastro-esophageal reflux disease | 37,706 (12.9) | 162,654 (16.3) | 0.095 | 33,197 (12.9) | 33,894 (13.2) | 0.008 |
| Disorders of thyroid gland | 33,577 (11.5) | 187,235 (18.7) | 0.203 | 31,299 (12.1) | 31,431 (12.2) | 0.002 |
| Nicotine dependence | 26,946 (9.2) | 46,522 (4.7) | 0.181 | 19,355 (7.5) | 19,470 (7.6) | 0.002 |
| Ischemic heart diseases | 13,196 (4.5) | 57,294 (5.7) | 0.055 | 11,780 (4.6) | 12,057 (4.7) | 0.005 |
| Diseases of liver | 12,043 (4.1) | 37,199 (3.7) | 0.021 | 10,181 (4.0) | 10,097 (3.9) | 0.002 |
| Iron deficiency anemia | 12,479 (4.3) | 30,781 (3.1) | 0.064 | 9,619 (3.7) | 9,777 (3.8) | 0.003 |
| Chronic kidney disease (CKD) | 9,647 (3.3) | 45,440 (4.5) | 0.064 | 8,673 (3.4) | 8,940 (3.5) | 0.006 |
| Cerebrovascular diseases | 8,962 (3.1) | 35,855 (3.6) | 0.029 | 7,731 (3.0) | 8,007 (3.1) | 0.006 |
| COPD | 8,684 (3.0) | 27,792 (2.8) | 0.012 | 7,391 (2.9) | 7,295 (2.8) | 0.002 |
| Pregnancy, childbirth and the puerperium | 8,188 (2.8) | 12,926 (1.3) | 0.107 | 5,729 (2.2) | 5,527 (2.1) | 0.005 |
| Alcohol related disorders | 6,545 (2.2) | 11,286 (1.1) | 0.087 | 4,569 (1.8) | 4,480 (1.7) | 0.003 |
| Systemic connective tissue disorders | 4,587 (1.6) | 21,368 (2.1) | 0.042 | 4,196 (1.6) | 4,206 (1.6) | 0.000 |
| COVID-19 | 3,798 (1.3) | 12,894 (1.3) | 0.001 | 3,293 (1.3) | 3,159 (1.2) | 0.005 |
| Malnutrition | 2,442 (0.8) | 5,130 (0.5) | 0.039 | 1887 (0.7) | 1898 (0.7) | 0.000 |
| Laboratory data | ||||||
| Hemoglobin ≥ 12 g/dL | 175,956 (60.3) | 570,513 (57.1) | 0.065 | 153,365 (59.5) | 154,975 (60.1) | 0.013 |
| Albumin ≥3.5 g/dL | 154,980 (53.1) | 530,652 (53.1) | 0.000 | 134,627 (52.3) | 136,699 (53.1) | 0.016 |
| HbA1c ≥ 9% | 14,900 (5.1) | 19,051 (1.9) | 0.175 | 10,261 (4.0) | 10,658 (4.1) | 0.008 |
| eGFR ≥ 60 mL/min/1.73 m2 | 158,902 (54.5) | 536,989 (53.7) | 0.014 | 136,283 (52.9) | 135,598 (52.6) | 0.005 |
| Medications | ||||||
| Cardiovascular system | 133,700 (45.8) | 500,274 (50.1) | 0.085 | 114,917 (44.6) | 116,036 (45.0) | 0.009 |
| Glucocorticoids | 72,782 (24.9) | 285,490 (28.6) | 0.082 | 63,351 (24.6) | 63,906 (24.8) | 0.005 |
| Vitamin d supplementation | 27,953 (9.6) | 172,356 (17.3) | 0.227 | 26,455 (10.3) | 27,681 (10.7) | 0.016 |
| Anticoagulants | 30,375 (10.4) | 89,069 (8.9) | 0.051 | 24,822 (9.6) | 25,337 (9.8) | 0.007 |
| Platelet aggregation inhibitors | 28,439 (9.7) | 117,345 (11.7) | 0.065 | 24,520 (9.5) | 24,952 (9.7) | 0.006 |
| Insulins and analogues | 20,345 (7.0) | 41,916 (4.2) | 0.121 | 15,582 (6.0) | 15,937 (6.2) | 0.006 |
| Contraceptives, systemic | 9,852 (3.4) | 26,574 (2.7) | 0.042 | 8,032 (3.1) | 7,698 (3.0) | 0.008 |
| Estrogens | 5,424 (1.9) | 47,989 (4.8) | 0.165 | 5,247 (2.0) | 5,311 (2.1) | 0.002 |
Data are presented as n (%) for categorical variables and mean ± standard deviation for continuous variables. Standardized mean differences (SMDs) were used to assess covariate balance between groups, with an absolute SMD <0.10 indicating adequate balance. BMI ≥30 kg/m2 was defined by measured body mass index, whereas “overweight and obesity” was defined by ICD-10-CM diagnostic codes. BMI, body mass index; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; COVID-19, coronavirus disease 2019; eGFR, estimated glomerular filtration rate; HbA1c, hemoglobin A1c; SMD, standardized mean difference; VDD, vitamin D deficiency.
3.2. Outcomes
Over the 4-year observation window following the 1-year landmark period, incident varicose veins of the lower extremities were recorded in 1,665 patients (0.65%) in the VDD group and 1,031 patients (0.40%) in the control group. VDD was associated with a significantly higher risk of incident varicose veins (HR 1.62, 95% CI 1.50–1.75; p < 0.001) (Table 2). The E-value for the point estimate was 2.62, and for the lower confidence interval boundary was 2.37. Kaplan–Meier analysis demonstrated a consistently lower event-free probability in the VDD cohort over time (log-rank p < 0.001) (Figure 1). With respect to secondary outcomes, VDD was similarly associated with varicose veins with ulceration (HR 2.22, p < 0.001), varicose veins with inflammation (HR 2.71, p < 0.001), and varicose veins with other complications (HR 1.45, p < 0.001).
Table 2.
Association between vitamin D deficiency and the 5-year risk of incident varicose vein.
| Outcome | VDD group (n = 257,651) | Control group (n = 257,651) | HR (95% CI) | p value |
|---|---|---|---|---|
| Events (%) | Events (%) | |||
| Primary outcome | ||||
| Varicose vein of LE | 1,665 (0.65%) | 1,031 (0.40%) | 1.62 (1.50–1.75) | <0.001 |
| Secondary outcomes | ||||
| Varicose vein with ulcer | 171 (0.07%) | 77 (0.03%) | 2.22 (1.70–2.91) | <0.001 |
| Varicose vein with inflammation | 276 (0.11%) | 102 (0.04%) | 2.71 (2.16–3.40) | <0.001 |
| Varicose vein with complications | 687 (0.27%) | 473 (0.18%) | 1.45 (1.29–1.63) | <0.001 |
CI, confidence interval; HR, hazard ratio; VDD, vitamin D deficiency; LE, lower extremities; Outcomes were assessed from 1 year after index date (days 365–1825).
Figure 1.
Kaplan–Meier curves for event-free probability of incident varicose vein over 5 years of follow-up in the propensity score-matched cohorts. The curves compare patients with vitamin D deficiency and matched controls with sufficient vitamin D status. A 1-year landmark period was applied, and between-group differences were assessed using the log-rank test. VDD, vitamin D deficiency.
3.3. Validation of study design and healthcare utilization
The positive control outcome, osteoporotic fracture, was more frequently observed in the VDD group (HR 1.45, 95% CI 1.28–1.64; p < 0.001), consistent with the known biological relationship between vitamin D and bone health. The negative control outcome, appendicitis, showed no significant difference between groups (HR 0.97, 95% CI 0.78–1.19; p = 0.734). Healthcare encounter rates were high in both cohorts (95.20% vs. 97.11%), with a small but statistically significant difference (HR 0.90; p < 0.001), supporting broadly comparable opportunity for outcome ascertainment across groups (Table 3).
Table 3.
Validation of study design and healthcare utilization.
| Outcome | VDD group (n = 257,651) | Control group (n = 257,651) | HR (95% CI) | p value |
|---|---|---|---|---|
| Events (%) | Events (%) | |||
| Positive control outcome | ||||
| Osteoporotic fracture | 607 (0.24%) | 419 (0.16%) | 1.45 (1.28–1.64) | <0.001 |
| Negative control outcome | ||||
| Appendicitis | 175 (0.07%) | 182 (0.07%) | 0.97 (0.78–1.19) | 0.734 |
| Healthcare utilization validation | ||||
| Healthcare visit | 245,295 (95.20%) | 250,210 (97.11%) | 0.90 (0.90–0.91) | <0.001 |
Data are presented as the number of events (percentage) and hazard ratios (HRs) with 95% confidence intervals (CIs). CI, confidence interval; HR, hazard ratio; VDD, vitamin D deficiency; Outcomes were assessed from 1 year after index date (days 365–1825).
3.4. Sensitivity analyses and subgroup analyses
The association between VDD and incident varicose veins was consistent across all three prespecified sensitivity analyses (Table 4). When the study period was restricted to 2016–2023 to ensure ICD-10-CM coding consistency (Model I), the association remained significant (HR 1.45, p < 0.001). Restriction to patients who survived throughout follow-up (Model II) yielded a comparable estimate (HR 1.53, p < 0.001), as did restriction to patients with at least one ambulatory encounter during follow-up (Model III; HR 1.59, p < 0.001). Findings for all secondary outcomes were likewise consistent across sensitivity models.
Table 4.
Sensitivity analyses of the association between vitamin D deficiency and the 5-year risk of incident varicose vein.
| Outcomes | Model I | Model II | Model III | |||
|---|---|---|---|---|---|---|
| HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | |
| Varicose vein of LE | 1.45 (1.34–1.58) | <0.001 | 1.53 (1.42–1.66) | <0.001 | 1.59 (1.47–1.71) | <0.001 |
| Varicose vein with ulcer | 2.03 (1.53–2.70) | <0.001 | 2.16 (1.62–2.87) | <0.001 | 2.04 (1.57–2.65) | <0.001 |
| Varicose vein with inflammation | 2.37 (1.79–3.14) | <0.001 | 2.39 (1.90–2.99) | <0.001 | 2.78 (2.22–3.48) | <0.001 |
| Varicose vein with complications | 1.36 (1.20–1.53) | <0.001 | 1.41 (1.25–1.58) | <0.001 | 1.46 (1.30–1.64) | <0.001 |
Model I restricted the study period to 2016–2023 to ensure coding consistency. Model II excluded patients who died during follow-up. Model III was restricted to patients with at least one ambulatory healthcare encounter during follow-up to address potential differences in outcome ascertainment. CI, confidence interval; HR, hazard ratio; LE, lower extremities.
Subgroup analyses revealed that the association between VDD and varicose vein risk was present across all prespecified subgroups (Table 5). Statistically significant effect modification was identified for age (p for interaction = 0.005), with a stronger association observed in patients older than 65 years (HR 1.77) than in those aged 40–65 years (HR 1.41). Significant interactions were also detected for obesity status (p = 0.048) and diabetes mellitus (p = 0.039), with higher HRs observed in the non-obese (HR 1.61) and diabetic (HR 1.80) subgroups, respectively. No significant interactions were observed for sex, hyperlipidemia, cancer history, or hypertension.
Table 5.
Subgroup analysis.
| Subgroup | HR (95% CI) | p value | p for interaction |
|---|---|---|---|
| Male | 1.62 (1.37–1.90) | <0.001 | Reference |
| Female | 1.58 (1.45–1.73) | <0.001 | 0.794 |
| 40–65 years | 1.41 (1.27–1.58) | <0.001 | Reference |
| >65 years | 1.77 (1.59–1.98) | <0.001 | 0.005 |
| Obesity | 1.37 (1.20–1.56) | <0.001 | Reference |
| No obesity | 1.61 (1.47–1.78) | <0.001 | 0.048 |
| Hyperlipidemia | 1.64 (1.47–1.83) | <0.001 | Reference |
| No hyperlipidemia | 1.45 (1.30–1.61) | <0.001 | 0.117 |
| Cancer history | 1.69 (1.48–1.93) | <0.001 | Reference |
| No cancer history | 1.43 (1.31–1.57) | <0.001 | 0.05 |
| DM | 1.80 (1.52–2.13) | <0.001 | Reference |
| No DM | 1.45 (1.33–1.59) | <0.001 | 0.039 |
| HTN | 1.65 (1.48–1.84) | <0.001 | Reference |
| No HTN | 1.44 (1.30–1.61) | <0.001 | 0.083 |
Prespecified subgroup analyses were also conducted according to sex, age (40–65 vs. >65 years), obesity, hyperlipidemia, history of cancer, diabetes mellitus, and hypertension.
3.5. Multivariable Cox regression analysis
In the multivariable Cox proportional hazards model, VDD remained independently associated with incident varicose veins after adjustment for all covariates (HR 1.66, 95% CI 1.56–1.76; p < 0.001) (Table 6). Among the remaining covariates, overweight/obesity (HR 1.56), White race (HR 1.54), and male sex (HR 0.70, protective) demonstrated the strongest independent associations with varicose vein risk. Modest but statistically significant associations were additionally observed for GERD (HR 1.20), liver disease (HR 1.17), iron deficiency anemia (HR 1.13), thyroid disorders (HR 1.12), hypertension (HR 1.11), neoplasms (HR 1.09), and older age (HR 1.02 per year).
Table 6.
Multivariable Cox proportional hazards regression analysis for incident varicose veins of the lower extremities.
| Variable | HR | (95% CI) | p-value |
|---|---|---|---|
| VDD vs. Control groups | 1.66 | (1.56–1.76) | <0.001 |
| Male vs. Female | 0.70 | (0.66–0.75) | <0.001 |
| Age at index | 1.02 | (1.02–1.02) | <0.001 |
| White | 1.54 | (1.45–1.63) | <0.001 |
| Essential hypertension | 1.11 | (1.04–1.17) | 0.001 |
| Hyperlipidemia | 1.01 | (0.95–1.07) | 0.836 |
| Neoplasms | 1.09 | (1.03–1.15) | 0.003 |
| Overweight / Obesity | 1.56 | (1.46–1.67) | <0.001 |
| Diabetes mellitus | 1.06 | (0.99–1.14) | 0.110 |
| GERD | 1.20 | (1.13–1.27) | <0.001 |
| Thyroid disorders | 1.12 | (1.06–1.19) | <0.001 |
| Nicotine dependence | 0.95 | (0.86–1.05) | 0.323 |
| Liver disease | 1.17 | (1.05–1.30) | 0.005 |
| Iron deficiency anemia | 1.13 | (1.00–1.27) | 0.043 |
| CKD | 1.07 | (0.96–1.19) | 0.241 |
| Alcohol-related disorders | 0.95 | (0.77–1.18) | 0.663 |
HR, hazard ratio; CI, confidence interval; CKD, chronic kidney disease; VDD: vitamin D deficiency; Age was modeled as a continuous variable (per 1-year increase).
3.6. Vitamin D insufficiency and risk of varicose vein
Among 317,566 matched pairs in the vitamin D insufficiency (VDI) analysis, VDI was likewise associated with a higher risk of incident varicose veins of the lower extremities (HR 1.56, 95% CI 1.47–1.67; p < 0.001). Associations were also observed for varicose veins with ulceration (HR 1.55, p < 0.001), inflammation (HR 1.79, p < 0.001), and other complications (HR 1.52, p < 0.001), suggesting a gradient pattern across vitamin D status categories (Table 7).
Table 7.
Association between vitamin D insufficiency (20.0–29.9 ng/mL) and the 5-year risk of incident varicose vein.
| Outcome | VDI group (n = 317,566) | Control group (n = 317,566) | HR (95% CI) | p value |
|---|---|---|---|---|
| Events (%) | Events (%) | |||
| Varicose vein of LE | 2,423 (0.76%) | 1,541 (0.49%) | 1.56 (1.47–1.67) | <0.001 |
| Varicose vein with ulcer | 195 (0.06%) | 125 (0.04%) | 1.55 (1.24–1.94) | <0.001 |
| Varicose vein with inflammation | 289 (0.09%) | 160 (0.05%) | 1.79 (1.48–2.17) | <0.001 |
| Varicose vein with complications | 1,057 (0.33%) | 690 (0.22%) | 1.52 (1.38–1.67) | <0.001 |
CI, confidence interval; HR, hazard ratio; VDI, vitamin D insufficiency; LE, lower extremities; Outcomes were assessed from 1 year after index date (days 365–1825).
4. Discussion
Despite the high global prevalence of varicose veins and the growing recognition of vitamin D’s vascular roles, large-scale longitudinal data examining the relationship between VDD and the incident risk of varicose veins have remained absent. Prior evidence has been confined to small mechanistic investigations or cross-sectional observations, leaving the epidemiological question unresolved. In the present propensity score-matched cohort of more than 515,000 matched individuals, VDD was associated with a significantly elevated five-year risk of incident varicose veins of the lower extremities (HR 1.62, 95% CI 1.50–1.75), with consistent associations observed across all secondary clinical subtypes, three prespecified sensitivity models, and a fully adjusted multivariable Cox regression model. These findings represent, to our knowledge, one of the largest real-world epidemiological analyses to date characterizing the association between VDD and varicose vein risk.
The present findings align with the broader body of evidence linking low vitamin D status to venous and vascular disease. Prior observational studies and meta-analyses have documented an association between VDD and venous thromboembolism, suggesting that low vitamin D status may be broadly relevant to venous vascular health (17, 18), though the underlying mechanisms likely differ across disease entities. At the tissue level, a recent ex vivo pilot study demonstrated that the active form of vitamin D attenuates reactive oxygen species generation and modulates nitric oxide synthase isoform expression in varicose venous samples from both obese and non-obese patients (11), providing direct molecular evidence of vitamin D’s involvement in venous wall biology.
A prior Mendelian randomization study did not demonstrate a significant association between genetically predicted vitamin D levels and varicose vein risk. Differences in exposure definition, including lifelong genetically determined levels versus clinically defined deficiency states, may partially account for this discrepancy. Genetic instruments for vitamin D explain only a modest fraction of its circulating variability, and the statistical power of MR analyses for disease endpoints with moderate event rates may be insufficient to detect associations of the magnitude observed here. Furthermore, MR does not capture the episodic or sustained clinical deficiency states that characterize real-world VDD. Taken together, these methodological distinctions suggest that the MR and observational findings address related but distinct scientific questions, and the null MR result does not necessarily negate an epidemiological association between clinically defined VDD and varicose vein risk.
Several biologically plausible mechanisms may explain the observed association between VDD and varicose vein risk. Vitamin D supports endothelial nitric oxide bioavailability and suppresses inducible nitric oxide synthase activity (19–21); therefore, reduced vitamin D status could plausibly contribute to oxidative stress, endothelial dysfunction, and impaired venous wall integrity, although these pathways were not directly examined in the present study. In addition, the vitamin D receptor is expressed in vascular smooth muscle and endothelial cells, where it modulates inflammatory signaling pathways such as NF-κB (22, 23). Reduced vitamin D signaling may be associated with pro-inflammatory cytokine activity, extracellular matrix degradation, venous valve dysfunction, and progressive venous dilation (11, 24); however, these mechanistic links remain inferential in the context of our observational design. Vitamin D also interacts with the renin–angiotensin system, and dysregulation of this pathway may adversely affect venous compliance and pressure (25, 26). Together, these mechanisms provide a biologically plausible explanation for the observed association, but they should be interpreted as hypothesis-generating rather than as evidence of a causal pathway.
An additional consideration is the potential role of magnesium in modifying the association between vitamin D status and varicose vein risk. Magnesium is a required cofactor for key enzymes involved in vitamin D metabolism, including hepatic 25-hydroxylase and renal 1α-hydroxylase, and magnesium deficiency has been shown to impair the conversion of 25-hydroxyvitamin D to its biologically active form, 1,25-dihydroxyvitamin D (calcitriol) (27, 28). Moreover, magnesium independently contributes to vascular smooth muscle regulation, endothelial function, and anti-inflammatory signaling (29–32), all of which are relevant to venous wall homeostasis. Therefore, concurrent magnesium deficiency could theoretically amplify the vascular consequences of VDD or exert independent effects on venous disease risk. As serum magnesium levels were not assessed in the present analysis, the potential confounding or mediating role of magnesium status could not be evaluated, and this represents an important avenue for future investigation.
In current study, although COVID-19 was included as a balanced covariate in the matching model, the low prevalence of COVID-19 in both cohorts (approximately 1.3%) precluded a dedicated subgroup analysis to determine whether the association between VDD and venous disease risk is modified by SARS-CoV-2 infection. Given the well-established hypercoagulable and pro-inflammatory state associated with COVID-19 (33, 34), the potential interaction between VDD and COVID-19 in promoting superficial venous thrombosis and varicose vein complications warrants investigation in future studies with sufficient sample sizes of infected individuals.
A notable finding of the present analysis is that vitamin D insufficiency (VDI; 20.0–29.9 ng/mL) was also associated with an increased risk of incident varicose veins (HR 1.56, 95% CI 1.47–1.67). Similar associations were observed for clinically relevant subtypes, including varicose veins with ulceration, inflammation, and other complications. Although the effect estimate for VDI was numerically lower than that observed for VDD, these analyses were conducted in separate matched cohorts and were not intended for direct comparison. Accordingly, the present findings should not be interpreted as establishing a formal dose–response relationship. Rather, they suggest that the association between reduced vitamin D status and varicose vein risk may extend beyond overt deficiency to include milder degrees of insufficiency. From a clinical perspective, this observation is potentially relevant because it suggests that even moderate reductions in vitamin D status may be associated with venous disease risk, although the underlying mechanisms and causal relevance require further investigation. The higher crude event rate observed in the VDI cohort (0.76%) (Table 7) than in the VDD cohort (0.65%) (Table 2) should be interpreted cautiously. Because the VDI and VDD analyses were performed as separate propensity score–matched comparisons, the matched populations may have differed in baseline characteristics and risk profiles, which could account for this numerical difference rather than indicating a truly higher risk in the VDI group.
Subgroup analyses suggested two clinically interpretable patterns. The association between VDD and varicose vein risk appeared more pronounced in patients older than 65 years (HR 1.77) compared with those aged 40–65 years (HR 1.41; p for interaction = 0.005), and in patients with diabetes mellitus (HR 1.80 vs. HR 1.45; p for interaction = 0.039). These findings may reflect increased biological vulnerability, as aging and diabetes are both associated with endothelial dysfunction and impaired vascular homeostasis (35–37), which could potentially amplify susceptibility to the vascular effects of VDD. In contrast, the association appeared attenuated in individuals with obesity (HR 1.37 vs. HR 1.61; p for interaction = 0.048). This pattern may reflect several non-mutually exclusive mechanisms. Given that obesity is a strong independent risk factor for varicose veins (38–40), the high baseline risk in this subgroup may reduce the relative contribution of VDD on a multiplicative scale. In addition, the known association between obesity and lower circulating 25(OH)D levels may narrow the effective exposure contrast between groups (41, 42), potentially attenuating the observed effect estimate. Consistent with this notion, Damay et al. (43) reported that anthropometric adiposity was associated with impaired endothelial function, further supporting the concept that obesity-related vascular dysfunction may modify vascular susceptibility to additional risk factors such as VDD. Alternatively, residual confounding or differences in healthcare utilization may also contribute. Importantly, the association between VDD and varicose vein risk remained statistically significant across all subgroups.
In the current study, the observed event rates were low (0.65% in the VDD group and 0.40% in the control group over four years) compared with the commonly cited cross-sectional prevalence of varicose veins in the general population, which is approximately 23% (44). This discrepancy is expected because our analysis assessed the incidence of newly diagnosed varicose veins among individuals without pre-existing disease, rather than the overall prevalence of existing varicose veins in the population. In addition, ascertainment based on electronic health records captures only clinically documented diagnoses; therefore, mild or asymptomatic varicose veins may not have been formally coded. Such underascertainment would likely be non-differential between groups and would tend to bias the association toward the null. Thus, the observed association may represent a conservative estimate of the relationship between VDD and incident varicose veins.
The present study benefits from several methodological features that enhance the validity of its findings. The large matched sample provides substantial statistical power, while the one-year landmark period, positive and negative control outcomes, E-value analysis, and three sensitivity models collectively support the robustness and internal validity of the primary association. The E-value of 2.62 for the point estimate indicates that an unmeasured confounder would need to be strongly associated with both VDD and varicose vein risk, beyond the level of most known confounders already accounted for, to fully explain the observed association.
Several limitations warrant acknowledgment. First, as with all electronic health record -based studies, the ascertainment of varicose veins depends on clinical documentation rather than systematic screening, and outcome underascertainment remains possible, particularly among patients with lower healthcare engagement or those managed in non-participating facilities. The slightly lower healthcare encounter rate in the VDD cohort may have reduced opportunities for varicose vein detection in this group. If present, such under-ascertainment would likely bias the association toward the null, suggesting that the observed HR may be conservative. Although the sensitivity analysis restricted to ambulatory encounters yielded consistent results, differential surveillance between cohorts cannot be fully excluded. Second, vitamin D status was defined using threshold-based categories rather than continuous measurements, which may introduce exposure misclassification and precludes examination of associations across the full spectrum of circulating 25(OH)D concentrations. Additionally, seasonal variation in 25(OH)D levels could not be standardized, although the requirement for two qualifying measurements and exclusion of individuals with discordant values within three years may mitigate the impact of transient seasonal fluctuations on exposure classification. Similarly, the TriNetX database does not capture detailed clinical severity information such as Clinical-Etiological-Anatomical-Pathophysiological (CEAP) classification, duplex ultrasound findings, or venous anatomical characteristics, although the secondary outcome analyses of varicose veins with ulceration, inflammation, and other complications provide a partial assessment of disease severity. Third, despite the rigorous propensity score matching framework, residual confounding from unmeasured variables cannot be entirely eliminated; in particular, physical activity level, occupational prolonged standing, sunlight exposure, socioeconomic status, dietary patterns, parity, and genetic predisposition to venous insufficiency were not captured in the TriNetX database and may have influenced the observed associations. Notably, inherited thrombophilia, including Factor V Leiden and prothrombin G20210A mutations, could not be reliably assessed because genetic thrombophilia testing is not routinely performed and is documented in only a small, non-representative subset of patients within the database; although anticoagulant use was included as a matching variable, residual confounding from undiagnosed thrombophilia cannot be excluded. Fourth, the causal directionality of the observed association cannot be established from observational data alone, and reverse causation, whereby subclinical venous disease reduces physical activity and sun exposure, cannot be completely excluded. Fifth, serum magnesium levels were not available for analysis; given that magnesium is a critical cofactor for vitamin D activation, unmeasured magnesium deficiency may have contributed to both the observed VDD and the associated varicose vein risk, and its potential confounding or mediating effect could not be assessed. Finally, as TriNetX predominantly comprises healthcare organizations in the United States, the generalizability of these findings to populations with different demographic profiles, dietary habits, or healthcare systems warrants caution.
5. Conclusion
In this large propensity score-matched cohort study, VDD was associated with an increased risk of incident varicose veins, including subtypes with ulceration and inflammation. The association was also observed among older adults and individuals with diabetes, suggesting that these groups may warrant particular attention in future research. These findings support the potential relevance of vitamin D status in the clinical evaluation of patients at risk for venous disease. However, given the observational design, the results should be interpreted as associations rather than evidence of causality. Whether correction of VDD can reduce the risk of varicose vein development remains uncertain and should be evaluated in prospective interventional studies.
Funding Statement
The author(s) declared that financial support was not received for this work and/or its publication.
Footnotes
Edited by: Muhammad Iqhrammullah, Universitas Muhammadiyah Aceh, Indonesia
Reviewed by: Garend Jafar, Syiah Kuala University, Indonesia
Meral Ekim, Bozok University, Türkiye
Sarah H. Shaalan, University of Kufa, Iraq
Data availability statement
The datasets presented in this article are not readily available because the data was obtained from the TriNetX Global Collaborative Network. Data access can be obtained through a direct agreement with TriNetX. Requests to access the datasets should be directed to https://trinetx.com.
Ethics statement
The studies involving humans were approved by Institutional Review Board of Chi Mei Medical Center. The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants’ legal guardians/next of kin because this retrospective cohort study only used de-identified, HIPAA-compliant electronic health record data from the TriNetX Global Collaborative Network. No identifiable patient information was accessed, and the study posed minimal risk to participants.
Author contributions
K-ML: Conceptualization, Data curation, Investigation, Methodology, Project administration, Resources, Writing – original draft, Writing – review & editing. W-TW: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft. Y-JC: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Writing – original draft, Writing – review & editing.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was used in the creation of this manuscript. AI-assisted tools were used only to improve grammar and language clarity. No AI tools were used to generate scientific content, perform data analysis, or draw conclusions. All authors reviewed and approved the final manuscript and accept full responsibility for its content.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnut.2026.1870207/full#supplementary-material
References
- 1.Hamdan A. Management of varicose veins and venous insufficiency. JAMA. (2012) 308:2612–21. doi: 10.1001/jama.2012.111352, [DOI] [PubMed] [Google Scholar]
- 2.Rochon PJ, Reghunathan A, Kapoor BS, Kalva SP, Fidelman N, Majdalany BS, et al. ACR appropriateness criteria® lower extremity chronic venous disease. J Am Coll Radiol. (2023) 20:S481–s500. doi: 10.1016/j.jacr.2023.08.011, [DOI] [PubMed] [Google Scholar]
- 3.Gawas M, Bains A, Janghu S, Kamat P, Chawla P. A comprehensive review on varicose veins: preventive measures and different treatments. J Am Nutr Assoc. (2022) 41:499–510. doi: 10.1080/07315724.2021.1909510, [DOI] [PubMed] [Google Scholar]
- 4.Atkins E, Mughal NA, Place F, Coughlin PA. Varicose veins in primary care. BMJ. (2020) 370:m2509. doi: 10.1136/bmj.m2509 [DOI] [PubMed] [Google Scholar]
- 5.Anuforo A, Evbayekha E, Agwuegbo C, Okafor TL, Antia A, Adabale O, et al. Superficial venous disease-an updated review. Ann Vasc Surg. (2024) 105:106–24. doi: 10.1016/j.avsg.2024.01.009, [DOI] [PubMed] [Google Scholar]
- 6.He QF, Cai JY, Cheng M, Feng SJ, Lu QY, Wang F. Global prevalence and risk factors of varicose veins among health care workers: a systematic review and meta-analysis. BMC Nurs. (2025) 24:550. doi: 10.1186/s12912-025-03155-0, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ahmed WU, Kleeman S, Ng M, Wang W, Auton A, Lee R, et al. Genome-wide association analysis and replication in 810,625 individuals with varicose veins. Nat Commun. (2022) 13:3065. doi: 10.1038/s41467-022-30765-y, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Laurikka JO, Sisto T, Tarkka MR, Auvinen O, Hakama M. Risk indicators for varicose veins in forty- to sixty-year-olds in the Tampere varicose vein study. World J Surg. (2002) 26:648–51. doi: 10.1007/s00268-001-0283-1, [DOI] [PubMed] [Google Scholar]
- 9.Dunlop E, Pham NM, Van Hoang D, Mazahery H, Neo B, Shrapnel J, et al. A systematic review and meta-analysis of circulating 25-hydroxyvitamin D concentration and vitamin D status worldwide. J Public Health (Oxf). (2025) 47:e520–9. doi: 10.1093/pubmed/fdaf080, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cui A, Zhang T, Xiao P, Fan Z, Wang H, Zhuang Y. Global and regional prevalence of vitamin D deficiency in population-based studies from 2000 to 2022: a pooled analysis of 7.9 million participants. Front Nutr. (2023) 10:1070808. doi: 10.3389/fnut.2023.1070808, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Rațiu S, Mariș MI, Furdui-Lința AV, Stanciu-Lelcu T, Borza C, Olariu S, et al. Vitamin D alleviates oxidative stress in varicose veins: a pilot study in obese and non-obese patients. Mol Cell Biochem. (2025) 480:4797–807. doi: 10.1007/s11010-025-05292-1, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yuan S, Bruzelius M, Damrauer SM, Larsson SC. Cardiometabolic, lifestyle, and nutritional factors in relation to varicose veins: a Mendelian randomization study. J Am Heart Assoc. (2021) 10:e022286. doi: 10.1161/JAHA.121.022286, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ludwig RJ, Anson M, Zirpel H, Thaci D, Olbrich H, Bieber K, et al. A comprehensive review of methodologies and application to use the real-world data and analytics platform TriNetX. Front Pharmacol. (2025) 16:1516126. doi: 10.3389/fphar.2025.1516126, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hung KC, Yu TS, Lai YC, Chen IW. Sugammadex and postoperative respiratory failure in head and neck surgery: a cohort study. Drug Des Devel Ther. (2026) 20:580072. doi: 10.2147/DDDT.S580072, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Chang YJ, Weng HL, Hung KC, Ho CN, Lai YC, Wu JY, et al. Vitamin D deficiency and risk of heart failure in patients with obstructive sleep apnea: a cohort analysis. Front Nutr. (2026) 13:1755607. doi: 10.3389/fnut.2026.1755607, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chen IW, Chang LC, Chang YJ, Lai YC, Hung KC. Association between iron deficiency anemia and the risk of new-onset tuberculosis infection: a matched cohort analysis. Front Nutr. (2026) 13:1727992. doi: 10.3389/fnut.2026.1727992, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wan J, Yuan J, Li X, Bao Y, Hou Y, Li Z, et al. Association between serum vitamin D levels and venous thromboembolism (VTE): a systematic review and meta-analysis of observational studies. Complement Ther Med. (2020) 54:102579. doi: 10.1016/j.ctim.2020.102579, [DOI] [PubMed] [Google Scholar]
- 18.Hung K-C, Yang S-H, Chang C-Y, Wang L-K, Lin Y-T, Yu C-H, et al. Is circulating vitamin D status associated with the risk of venous thromboembolism? A meta-analysis of observational studies. Nutrients. (2023) 15:1113: 1113. doi: 10.3390/nu15051113, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Siervo M, Hussin AM, Calella P, Ashor A, Shannon OM, Mendes I, et al. Associations between aging and vitamin D status with whole-body nitric oxide production and markers of endothelial function. J Nutr. (2024) 154:469–78. doi: 10.1016/j.tjnut.2023.12.002, [DOI] [PubMed] [Google Scholar]
- 20.Kim D-H, Meza CA, Clarke H, Kim J-S, Hickner RC. Vitamin D and endothelial function. Nutrients. (2020) 12:575. doi: 10.3390/nu12020575, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ojaroodi AF, Jafarnezhad F, Eskandari Z, Keramat S, Stanek A. Recent updates and advances in the association between vitamin D deficiency and risk of thrombotic disease. Nutrients. (2024) 17:90: 90. doi: 10.3390/nu17010090, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Duddu S, Katakia YT, Suresh N, Chakrabarti R, Majumder S, Shukla PC. Vitamin D attenuates PCSK9-driven phenotypic switch of vascular smooth muscle cells and neointimal hyperplasia via SIRT6 activation. Biochim Biophys Acta. (2026) 1872:168135: 168135. doi: 10.1016/j.bbadis.2025.168135, [DOI] [PubMed] [Google Scholar]
- 23.Cimmino G, Morello A, Conte S, Pellegrino G, Marra L, Golino P, et al. Vitamin D inhibits tissue factor and CAMs expression in oxidized low-density lipoproteins-treated human endothelial cells by modulating NF-κB pathway. Eur J Pharmacol. (2020) 885:173422. doi: 10.1016/j.ejphar.2020.173422, [DOI] [PubMed] [Google Scholar]
- 24.Maggio M, De Vita F, Lauretani F, Ceda GP, Volpi E, Giallauria F, et al. Vitamin D and endothelial vasodilation in older individuals: data from the PIVUS study. J Clin Endocrinol Metab. (2014) 99:3382–9. doi: 10.1210/jc.2014-1536, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Latic N, Erben RG. Interaction of vitamin D with peptide hormones with emphasis on parathyroid hormone, FGF23, and the renin-angiotensin-aldosterone system. Nutrients. (2022) 14:5186. doi: 10.3390/nu14235186, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Wang L, Cook NR, Manson JE, Gaziano JM, Buring JE, Sesso HD. Associations of vitamin D-related biomarkers with hypertension and the renin–angiotensin system in men and women. Am J Hypertens. (2024) 37:953–61. doi: 10.1093/ajh/hpae103, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Risco F, Traba ML. Influence of magnesium on the in vitro synthesis of 24,25-dihydroxyvitamin D3 and 1 alpha, 25-dihydroxyvitamin D3. Magnes Res. (1992) 5:5–14. [PubMed] [Google Scholar]
- 28.Dai Q, Zhu X, Manson JE, Song Y, Li X, Franke AA, et al. Magnesium status and supplementation influence vitamin D status and metabolism: results from a randomized trial. Am J Clin Nutr. (2018) 108:1249–58. doi: 10.1093/ajcn/nqy274, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kudryavtseva O, Lyngsø KS, Jensen BL, Dimke H. Nitric oxide, endothelium-derived hyperpolarizing factor, and smooth muscle-dependent mechanisms contribute to magnesium-dependent vascular relaxation in mouse arteries. Acta Physiol. (2024) 240:e14096. doi: 10.1111/apha.14096, [DOI] [PubMed] [Google Scholar]
- 30.AlShanableh Z, Ray EC. Magnesium in hypertension: mechanisms and clinical implications. Front Physiol. (2024) 15:1363975. doi: 10.3389/fphys.2024.1363975, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Zhu D, You J, Zhao N, Xu H. Magnesium regulates endothelial barrier functions through TRPM7, MagT1, and S1P1. Adv Sci. (2019) 6:1901166. doi: 10.1002/advs.201901166, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Marques B, Klein M, da Cunha MR, de Souza Mattos S, de Paula Nogueira L, de Paula T, et al. Effects of Oral magnesium supplementation on vascular function: a systematic review and meta-analysis of randomized controlled trials. High Blood Press Cardiovasc Prev. (2020) 27:19–28. doi: 10.1007/s40292-019-00355-z [DOI] [PubMed] [Google Scholar]
- 33.Conway EM, Mackman N, Warren RQ, Wolberg AS, Mosnier LO, Campbell RA, et al. Understanding COVID-19-associated coagulopathy. Nat Rev Immunol. (2022) 22:639–49. doi: 10.1038/s41577-022-00762-9, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Gómez CA, Sun CK, Tsai IT, Chang YP, Lin MC, Hung IY, et al. Mortality and risk factors associated with pulmonary embolism in coronavirus disease 2019 patients: a systematic review and meta-analysis. Sci Rep. (2021) 11:16025. doi: 10.1038/s41598-021-95512-7, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Pacinella G, Ciaccio AM, Tuttolomondo A. Endothelial dysfunction and chronic inflammation: the cornerstones of vascular alterations in age-related diseases. Int J Mol Sci. (2022) 23:15722. doi: 10.3390/ijms232415722, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Jiang Y, Yabluchanskiy A, Deng J, Amil FA, Po SS, Dasari TW. The role of age-associated autonomic dysfunction in inflammation and endothelial dysfunction. Geroscience. (2022) 44:2655–70. doi: 10.1007/s11357-022-00616-1, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Zhang W, Sun Y, Yang Y, Chen Y. Impaired intracellular calcium homeostasis enhances protein O-GlcNAcylation and promotes vascular calcification and stiffness in diabetes. Redox Biol. (2023) 63:102720. doi: 10.1016/j.redox.2023.102720, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Petrascu F-M, Matei S-C, Margan M-M, Ungureanu A-M, Olteanu G-E, Murariu M-S, et al. The impact of inflammatory markers and obesity in chronic venous disease. Biomedicine. (2024) 12:2524. doi: 10.3390/biomedicines12112524, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.de Medeiros Alves Fernandes WR, Lins EM, Rocha FA, de Oliveira Buril G, de Andrade Souza Caldas RP, de Souza Leão Câmara F, et al. Effect of obesity and bariatric surgery on saphenous veins. Obes Surg. (2023) 33:821–5. doi: 10.1007/s11695-023-06469-7 [DOI] [PubMed] [Google Scholar]
- 40.Millen RN, Thomas KN, Versteeg MPT, van Rij AM. Popliteal vein compression, obesity, and chronic venous disease. J Vasc Surg Venous Lymphat Disord. (2022) 10:200–8.e2. doi: 10.1016/j.jvsv.2021.05.013, [DOI] [PubMed] [Google Scholar]
- 41.Naganuma J, Koyama S, Arisaka O, Yoshihara S. Low serum 25-hydroxyvitamin D level is associated with obesity and atherogenesis in adolescent boys. Ann. Pediatric Endocrinol. Metabol. (2022) 27:30–6. doi: 10.6065/apem.2142112.056, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Mirza I, Mohamed A, Deen H, Balaji S, Elsabbahi D, Munasser A, et al. Obesity-associated vitamin D deficiency correlates with adipose tissue DNA hypomethylation, inflammation, and vascular dysfunction. Int J Mol Sci. (2022) 23:14377. doi: 10.3390/ijms232214377, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Damay VA, Ivan I, Islami NA, Rubismo KY. Impact of anthropometric adiposity and excessive daytime sleepiness on endothelial function in healthcare workers: a cross-sectional analysis. Narra J. (2025) 5:e2003. doi: 10.52225/narra.v5i2.2003, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Piazza G. Varicose veins. Circulation. (2014) 130:582–7. doi: 10.1161/CIRCULATIONAHA.113.008331, [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets presented in this article are not readily available because the data was obtained from the TriNetX Global Collaborative Network. Data access can be obtained through a direct agreement with TriNetX. Requests to access the datasets should be directed to https://trinetx.com.