Volume 102, Issue 11 , Pages 1540-1544, 1 December 2008
Prevalence of Hypovitaminosis D in Cardiovascular Diseases (from the National Health and Nutrition Examination Survey 2001 to 2004)
Article Outline
This cross-sectional study examined the burden of cardiovascular diseases (CVDs) using serum 25-hydroxyvitamin D (25[OH]D) and prevalence of hypovitaminosis D in adults with CVDs using data from NHANES 2001 to 2004. Serum 25(OH)D levels were divided into 3 categories (≥30, 20 to 29, and <20 ng/ml), and hypovitaminosis D was defined as vitamin D <30 ng/ml. Of 8,351 adults who had 25(OH)D measured, mean 25(OH)D was 24.3 ng/ml, and the prevalence of hypovitaminosis D was 74%. The burden of CVDs increased with lower 25(OH)D categories, with 5.3%, 6.7%, and 7.3% coronary heart disease; 1.5%, 2.4%, and 3.2% heart failure; 2.5%, 2.0%, and 3.2% stroke; and 3.6%, 5.0%, and 7.7% peripheral arterial disease. Across all CVDs, hypovitaminosis D was more common in blacks than Hispanics or whites. Compared with persons at low risk for CVDs (68%), it was more prevalent in those at high risk (75%; odds ratio [OR] 1.32, 95% confidence interval [CI] 1.05 to 1.67), with coronary heart disease (77%; OR 1.48, 95% CI 1.14 to 1.91), and both coronary heart disease and heart failure (89%; OR 3.52, 95% CI 1.58 to 7.84) after controlling for age, race, and gender. In conclusion, hypovitaminosis D was highly prevalent in US adults with CVDs, particularly those with both coronary heart disease and heart failure.
Although a growing body of evidence suggested a role for hypovitaminosis D in the pathogenesis of cardiovascular diseases (CVDs), the importance of adequate vitamin D intake was often neglected.1 The prevalence of hypovitaminosis D in US adults was >40% in men and >50% in women.2 Inadequate vitamin D intake and serum 25-hydroxyvitamin D (25[OH]D) were associated with cardiovascular risk factors, such as hypertension, diabetes mellitus, obesity, and dyslipidemia.3, 4 In addition, hypovitaminosis D was highly prevalent in patients with various CVDs.5, 6, 7, 8 However, these observations were generally based on small samples. The prevalence of hypovitaminosis D in a broad representation of adults in the United States with a diagnosis of CVD is unknown. Therefore, we examined the burden of CVDs using 25(OH)D and prevalence of hypovitaminosis D in adults with CVDs using data from the National Health and Nutrition Examination Survey (NHANES) 2001 to 2004.
Methods
NHANES is a nationally representative survey of the noninstitutionalized population in the United States.9 It used a complex multistage cluster design to estimate the prevalence of chronic conditions. For this study, we combined the 2001 to 2004 cycles of NHANES for adequate sample size. This study was approved by the institutional review board at Thomas Jefferson University. Of 9,145 subjects ≥20 years who had serum 25(OH)D measured, races other than white, black, and Hispanic (n = 346) and pregnant women (n = 448) were excluded, leaving a final analytic sample of 8,351 subjects.
Serum 25(OH)D levels were measured using the DiaSorin 25(OH)D assay (DiaSorin S.p.A., Vercelli, Italy), which consisted of rapid extraction and radioimmunoassay. Results were divided into the categories of <20 (50), 20 to 29 (50 to 74.9), and ≥30 ng/ml (75 nmol/L). Hypovitaminosis D was defined as <30 ng/ml.1 The CVDs examined were coronary heart disease, defined as coronary heart disease, angina, or myocardial infarction; heart failure; stroke; and peripheral arterial disease, defined as ankle-brachial index (calculated as a ratio of mean systolic blood pressure in the respective ankle to that in the arm10; available for 4,818 subjects) <0.9. All CVDs except for peripheral arterial disease were self-reported as a response to the question “Has a doctor or other health professional ever told you that you had coronary heart disease; angina; a heart attack; congestive heart failure; and a stroke?”
Race was self-identified. Body mass index (calculated as weight in kilograms divided by the square of height in meters) was used to define overweight (25 to 29.9 kg/m2) and obesity (≥30 kg/m2). Leisure-time physical activity was defined as any participation in moderate or vigorous physical activities during the past 30 days and identified using the question: “Over the past 30 days, what vigorous or moderate activities did you do?” and MET intensity of 2.5 to 10. Vitamin D supplement use was defined as use of any supplement containing vitamin D for ≥20 of the past 30 days, and regular milk drinking was defined as using any type of milk for ≥5 days/week. Chronic kidney disease was defined as glomerular filtration rate (estimated using the Modification of Diet in Renal Disease formula11) <60 ml/min. Diabetes mellitus was assessed using self-report, fasting glucose ≥126 mg/dl (≥7.0 mmol/L), or insulin or oral hypoglycemic agent use. Hypertension was assessed using self-report, average systolic or diastolic blood pressure ≥140 or 90 mm Hg, respectively, or antihypertensive agent use, respectively. C-Reactive protein, fasting glucose, total–high-density lipoprotein cholesterol ratio, and urinary albumin-creatinine ratio were also examined.
Analyses were conducted incorporating appropriate sampling weights and stratification and clustering of the design using Stata 9.2 (StataCorp, College Station, Texas). Mean 25(OH)D levels were compared across various characteristics using linear regression after controlling for age, race, and gender. Next, subjects with no CVD were classified as high (vs low) risk if they had any of the risk factors of current smoking, obesity, hypertension, diabetes, fasting glucose 100 to 125 mg/dl (5.6 to 6.9 mmol/L), chronic kidney disease, urinary albumin-creatinine ratio ≥30, total–high-density lipoprotein cholesterol ratio ≥5, and C-reactive protein ≥0.30 mg/dl. The adjusted odds ratio and 95% confidence interval for each CVD comparing 25(OH)D categories and for hypovitaminosis D comparing subjects at low risk, high risk, and with CVDs were computed using logistic regression after controlling for demographic, lifestyle, and clinical characteristics. We examined various combinations of CVDs to identify a particular subgroup of subjects with a higher prevalence of hypovitaminosis D.
Results
Mean 25(OH)D was 24.3 ng/ml (61 nmol/L). The level decreased with older age (24.4 for 20 to 64 years, 24.3 for 65 to 74 years, and 23.6 for ≥75 years; p = 0.001) and black and Hispanic race (26.2 for white, 14.9 for black, and 21.2 for Hispanic; p <0.001), but did not differ by gender (24.6 for men and 24.0 for women; p = 0.115). Table 1 lists mean 25(OH)D by lifestyle and clinical characteristics. The burden of CVDs except stroke increased with lower 25(OH)D (Figure 1). The strength of association and dose–response relation attenuated with further adjustment for lifestyle and clinical characteristics.
Table 1. Mean 25-hydroxyvitamin D (25[OH]D) across lifestyle and clinical characteristics in the National Health and Nutrition Examination Survey 2001 to 2004
| Characteristics | % Population | Mean 25(OH)D (ng/ml) | p Value⁎ |
|---|---|---|---|
| Body mass index (kg/m2)† | <0.001 | ||
 <25.0 | 32.4% | 26.7 | |
| 25.0–29.9 | 34.0% | 24.9 | |
| ≥30.0 | 30.7% | 21.5 | |
| Current smoking status† | 0.021 | ||
| No | 74.7% | 24.5 | |
| Yes | 25.2% | 23.8 | |
| Leisure-time physical activity | <0.001 | ||
| No | 34.8% | 21.8 | |
| Yes | 65.2% | 25.6 | |
| Vitamin D supplement use | 0.002 | ||
| No | 86.0% | 24.0 | |
| Yes | 14.0% | 26.4 | |
| Regular milk consumption† | <0.001 | ||
| No | 50.2% | 23.2 | |
| Yes | 49.7% | 25.5 | |
| Chronic kidney disease | 0.974 | ||
| No | 91.7% | 24.3 | |
| Yes | 8.3% | 24.4 | |
| Diabetes mellitus | <0.001 | ||
| No | 91.1% | 24.6 | |
| Yes | 8.9% | 21.3 | |
| Hypertension | <0.001 | ||
| No | 64.2% | 24.9 | |
| Yes | 35.8% | 23.2 | |
| C-Reactive protein (mg/dl)† | 0.002 | ||
| <0.10 | 29.9% | 25.5 | |
| 0.10–0.29 | 32.8% | 24.6 | |
| ≥0.30 | 37.3% | 23.1 | |
| Fasting glucose (mg/dl)† | <0.001 | ||
| <100 | 80.8% | 24.8 | |
| 100–125 | 13.6% | 22.8 | |
| ≥126 | 5.2% | 20.4 | |
| Total/high-density lipoprotein cholesterol ratio† | <0.001 | ||
| <3.5 | 38.7% | 25.5 | |
| 3.5–4.9 | 37.2% | 23.9 | |
| ≥5.0 | 23.9% | 23.0 | |
| Urine albumin-creatinine ratio† | <0.001 | ||
| <30 | 89.8% | 24.6 | |
| 30–299 | 7.5% | 22.0 | |
| ≥300 | 1.2% | 19.1 |
⁎Controlled for age, race, and gender. |
†Data were not available for 306 subjects for body mass index; 9 for current smoking status, regular milk consumption, and C-reactive protein; 41 for fasting glucose; 20 for total//high-density lipoprotein cholesterol ratio; and 158 for urine albumin-creatinine ratio. |
Figure 1.
Prevalence and adjusted odds ratios (ORs) of CVD by 25(OH)D in NHANES 2001 to 2004. *Controlled for age, race, and gender. †Controlled for age, race, gender, current smoking, leisure-time physical activity, vitamin D supplement use, and regular milk drink. ‡Controlled for body mass index category, chronic kidney disease, hypertension, and diabetes mellitus in addition to the mentioned demographic and lifestyle characteristics. §Subjects with ankle-brachial index measured were included (n = 4,818). CHD = coronary heart disease; CI = confidence interval; HF = heart failure; PAD = peripheral arterial disease.
The prevalence of hypovitaminosis D was 74% overall; 68% for whites, 97% for blacks, 88% for Hispanics, and 74% for both men and women. Across various CVDs, the prevalence of hypovitaminosis D was higher in blacks than in Hispanics or whites, but no consistent pattern was observed by age and gender (Table 2). Hypovitaminosis D was especially more common in subjects with coronary heart disease and heart failure (Figure 2). Additional adjustment for lifestyle and clinical characteristics attenuated these associations.
Table 2. Prevalence of hypovitaminosis D in subjects with cardiovascular diseases in the National Health and Nutrition Examination Survey 2001–2004
| White | Black | Hispanic | Total | |
|---|---|---|---|---|
| Coronary heart disease | (n = 519) | (n = 112) | (n = 122) | (n = 753) |
| 20–64 yrs | 77% | 94% | 97% | 81% |
| 65–74 yrs | 72% | 97% | 98% | 75% |
| ≥75 yrs | 78% | 96% | 74% | 79% |
| Men | 75% | 91% | 98% | 77% |
| Women | 77% | 97% | 90% | 81% |
| Total | 76% | 95% | 94% | 79% |
| Heart failure | (n = 198) | (n = 48) | (n = 43) | (n = 289) |
| 20–64 yrs | 73% | 100% | 100% | 80% |
| 65–74 yrs | 85% | 100% | 98% | 87% |
| ≥75 yrs | 85% | 100% | 51% | 84% |
| Men | 77% | 94% | 99% | 80% |
| Women | 85% | 100% | 81% | 87% |
| Total | 81% | 98% | 90% | 83% |
| Stroke | (n = 191) | (n = 65) | (n = 53) | (n = 309) |
| 20–64 yrs | 67% | 94% | 69% | 72% |
| 65–74 yrs | 72% | 96% | 89% | 77% |
| ≥75 yrs | 72% | 100% | 60% | 74% |
| Men | 77% | 89% | 71% | 78% |
| Women | 65% | 100% | 73% | 71% |
| Total | 70% | 96% | 72% | 74% |
| Peripheral arterial disease⁎ | (n = 242) | (n = 87) | (n = 71) | (n = 400) |
| 20–64 yrs | 78% | 100% | 95% | 83% |
| 65–74 yrs | 91% | 96% | 98% | 92% |
| ≥75 yrs | 69% | 100% | 87% | 73% |
| Men | 83% | 100% | 87% | 85% |
| Women | 75% | 98% | 99% | 80% |
| Total | 79% | 99% | 94% | 82% |
⁎Subjects with ankle-brachial index measured were included (n = 4,818). |
Figure 2.
Prevalence and adjusted ORs of hypovitaminosis D in subjects with various CVDs in NHANES 2001 to 2004. *Hypovitaminosis D defined as serum 25(OH)D <30 ng/ml (<75 nmol/L); peripheral arterial disease was not included because ankle-brachial index was measured in only a subset of subjects, and subjects with heart failure and stroke could not be assessed because of small sample size (n = 12). †Low risk indicated the absence of the following risk factors and clinical CVDs, such as CHD, HF, and stroke: current smoking, obesity, hypertension, diabetes, impaired fasting glucose, chronic kidney disease, micro- or macroalbuminuria, total/high-density lipoprotein cholesterol ratio ≥5, and C-reactive protein ≥0.30. High risk indicated the presence of any of the mentioned risk factors without clinical CVDs. ‡Controlled for age, race, and gender. §Controlled for age, race, gender, current smoking, leisure-time physical activity, vitamin D supplement use, and regular milk drink. ∥Controlled for body mass index category, chronic kidney disease, hypertension, and diabetes mellitus in addition to the mentioned demographic and lifestyle characteristics. Abbreviations as in Figure 1.
Discussion
In NHANES 2001 to 2004, CVDs were more prevalent in adults with lower 25(OH)D, and hypovitaminosis D was also more common in those with certain CVDs. Although our findings were consistent with previous reports of the association of hypovitaminosis D with cardiovascular risk factors5 and such CVDs as coronary heart disease,7 heart failure,8 stroke,6 and peripheral arterial disease,5 our study provided national data about the burden of CVDs by 25(OH)D levels and the prevalence of hypovitaminosis D by age, race, and gender in adults with CVDs. We also identified that subjects with both coronary heart disease and heart failure had the highest prevalence of hypovitaminosis D.
Several lines of evidence suggested that hypovitaminosis D may contribute to CVDs by stimulating renin expression,12 proliferation of cardiomyocytes13 and smooth muscle cells,14 secondary hyperparathyroidism,15 and inflammation.16 Attenuation of the association between hypovitaminosis D and CVDs after controlling for cardiovascular risk factors suggested that such factors were responsible in part for the observed association. The association of hypovitaminosis D appeared stronger with CVDs than with risk factors, which was in line with earlier findings that vitamin D supplementation was associated more with hard endpoints than less hard endpoints.17 In addition, higher prevalence of hypovitaminosis D in those with both coronary heart disease and heart failure may have been caused by limited physical activity and sunlight exposure as a result of CVDs. However, it has been shown that patients with heart failure and healthy controls differed in vitamin D–associated lifestyle factors in their earlier lives, suggesting that hypovitaminosis D may precede CVDs.18 Furthermore, 25(OH)D of 20 to 29 ng/ml (50 to 74.9 nmol/L) was not significantly associated with increased prevalence of CVDs. Mild hypovitaminosis D was not associated with incident hypertension19 or cardiovascular events20 in recent prospective studies.
A few limitations should be noted. The causal relation could not be established because of the cross-sectional nature of this study. Clustered sampling may underestimate SEs, leading to false-positive associations. Although we combined the 2 cycles of NHANES for adequate sample size, there were small numbers of subjects in certain age, race, and gender-specific groups and certain combinations of CVDs. Finally, there was a possibility of misclassification because CVDs were self-reported.
Nevertheless, our study provided valuable data about the prevalence of hypovitaminosis D for clinicians who treat patients at high risk or with CVDs, as well as for researchers who design clinical trials on the protective effect of vitamin D supplementation against CVDs.
References
- . Vitamin D in preventive medicine: are we ignoring the evidence?. Br J Nutr. 2003;89:552–572
- . The prevalence of hypovitaminosis D among US adults: data from the NHANES III. Ethn Dis. 2005;15(suppl 5):S5–S101
- . Vitamin D is related to blood pressure and other cardiovascular risk factors in middle-aged men. Am J Hypertens. 1995;8:894–901
- . Prevalence of cardiovascular risk factors and the serum levels of 25-hydroxyvitamin D in the United States: data from the Third National Health and Nutrition Examination Survey. Arch Intern Med. 2007;167:1159–1165
- . Vitamin D deficiency and secondary hyperparathyroidism are common complications in patients with peripheral arterial disease. J Gen Intern Med. 2002;17:663–669
- . Reduced vitamin D in acute stroke. Stroke. 2006;37:243–245
- . Myocardial infarction is inversely associated with plasma 25-hydroxyvitamin D3 levels: a community-based study. Int J Epidemiol. 1990;19:559–563
- . Low vitamin D status: a contributing factor in the pathogenesis of congestive heart failure?. J Am Coll Cardiol. 2003;41:105–112
- . National Health and Nutrition Examination survey data. Hyattsville, MD: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention. 2001–2004. http://www.cdc.gov/nchs/nhanes.htmAccessed December 17, 2007
- . Assessment of peripheral vascular disease in diabetes (Report and recommendations of an international workshop sponsored by the American Heart Association and the American Diabetes Association 18–20 September 1992, New Orleans, Louisiana). Diabetes Care. 1993;16:1199–1209
- . Using standardized serum creatinine values in the Modification of Diet in Renal Disease Study equation for estimating glomerular filtration rate. Ann Intern Med. 2006;145:247–254
- . 1,25-Dihydroxyvitamin D(3) is a negative endocrine regulator of the renin-angiotensin system. J Clin Invest. 2002;110:229–238
- . 1,25-Dihydroxyvitamin D3 regulation of cardiac myocyte proliferation and hypertrophy. Am J Physiol. 1997;272:H1751–H1758
- . 1,25-Dihydroxyvitamin D3 modulates growth of vascular smooth muscle cells. J Clin Invest. 1991;87:1889–1895
- . Parathyroid hormone has a prosclerotic effect on vascular smooth muscle cells. Kidney Blood Press Res. 2003;26:27–33
- . Regulation of lymphokine production and human T lymphocyte activation by 1,25-dihydroxyvitamin D3 (Specific inhibition at the level of messenger RNA). J Clin Invest. 1987;79:1659–1664
- . Vitamin D supplementation and total mortality: a meta-analysis of randomized controlled trials. Arch Intern Med. 2007;167:1730–1737
- . Patients with congestive heart failure and healthy controls differ in vitamin D-associated lifestyle factors. Int J Vitam Nutr Res. 2007;77:280–288
- . Plasma 25-hydroxyvitamin D levels and risk of incident hypertension. Hypertension. 2007;49:1063–1069
- . Vitamin D deficiency and risk of cardiovascular disease. Circulation. 2008;117:503–511
PII: S0002-9149(08)01275-7
doi:10.1016/j.amjcard.2008.06.067
© 2008 Elsevier Inc. All rights reserved.
Volume 102, Issue 11 , Pages 1540-1544, 1 December 2008
