Vitamin D and Mortality:
A Mendelian Randomization Study
Olivia Trummer,1Stefan Pilz,1Michael M. Hoffmann,2Bernhard R. Winkelmann,3Bernhard O. Boehm,4
Winfried März,5,6,7Thomas R. Pieber,1Barbara Obermayer-Pietsch,1*and Wilfried Renner5
BACKGROUND: Decreased circulating 25-hydroxy-vita-
min D (25-OH-vitamin D) concentrations have been
associated with mortality rates, but it is unclear
whether this association is causal. We performed a
3 common single-nucleotide polymorphisms (SNPs)
associated with 25-OH-vitamin D concentrations are
causal for mortality rates.
METHODS: Genotypes of SNPs in the group-specific
component gene (GC, rs2282679), 7-dehydro-
cholesterol reductase gene (DHCR7, rs12785878), and
cytochrome P450 IIR-1 gene (CYP2R1, rs10741657)
were determined in a prospective cohort study of 3316
male and female participants [mean age 62.6 (10.6)
years] scheduled for coronary angiography between
all-cause deaths, cardiovascular deaths, and noncar-
RESULTS: In a linear regression model adjusting for
centrations were predicted by GC genotype (P ?
0.001), CYP2R1 genotype (P ? 0.068), and DHCR7
genotype (P ? 0.001), with a coefficient of determina-
tion (r2) of 0.175. During a median follow-up time of
9.9 years, 955 persons (30.0%) died, including 619
deaths from cardiovascular causes. In a multivariate
Cox regression adjusted for classical risk factors, GC,
CYP2R1, and DHCR7 genotypes were not associated
with all-cause mortality, cardiovascular mortality, or
vitamin D concentrations do not predict mortality.
This suggests that low 25-OH-vitamin D concentra-
tions are associated with, but unlikely to be causal for,
higher mortality rates.
© 2012 American Association for Clinical Chemistry
25-Hydroxy-vitamin D (25-OH-vitamin D)8is known
for its crucial role in calcium and bone metabolism
(1, 2). In the past several years, attention has turned to
potential effects of 25-OH-vitamin D on cardiovascu-
syndrome (5, 6). We have previously shown that low
25-OH-vitamin D concentrations are associated with
all-cause and cardiovascular mortality (7). These re-
sults are in line with other studies (8, 9). It remains
unclear whether 25-OH-vitamin D is the cause or the
consequence of these effects (3).
Vitamin D deficiency has been shown to be highly
common among elderly individuals, probably due to
the decreasing capacity of skin to produce vitamin D
with aging (3).
location of alleles at the time of gamete formation. A
specific genotype carried by a person results from 2
such randomized transmissions, 1 from the paternally
inherited allele and the other from the maternally in-
herited allele. A logical consequence of these random-
izations is that genotypes are not expected to be asso-
ciated with known (measurable or not) or unknown
confounders for any outcome of interest, except those
lying on the causal pathway between the genotype and
the outcome. This allows the analysis of the genotype–
risk factor association and the genotype–outcome as-
sociation in an unconfounded manner. By combining
1Division of Endocrinology and Metabolism, Department of Internal Medi-
cine, Medical University of Graz, Graz, Austria;2Department of Clinical
Chemistry, University Medical Center, Freiburg, Germany;
Group Sachsenhausen, Frankfurt-Sachsenhausen, Germany;
Endocrinology and Diabetes, Ulm University, Ulm, Germany;5Clinical Insti-
tute of Medical and Chemical Laboratory Diagnostics, Medical University of
Graz, Graz, Austria;
Preventive Medicine, Medical Faculty Mannheim, University of Heidelberg,
Heidelberg, Germany;7Synlab Academy, Synlab Services LLC, Mannheim,
6Mannheim Institute of Public Health, Social and
* Address correspondence to this author at: Division of Endocrinology and
Metabolism, Department of Internal Medicine, Medical University of Graz,
Auenbruggerplatz 15, 8036 Graz, Austria. Fax ?43-316-385-13428; e-mail
Received July 16, 2012; accepted December 10, 2012.
Previously published online at DOI: 10.1373/clinchem.2012.193185
8Nonstandard abbreviations: 25-OH-vitamin D, 25-hydroxy vitamin D; CVD,
cardiovascular disease; SNP, single-nucleotide polymorphism; LURIC, Ludwig-
shafen Risk and Cardiovascular Health; CAD, coronary artery disease; r2,
coefficient of determination.
Clinical Chemistry 59:5
Endocrinology and Metabolism
the results of these 2 analyses appropriately, one can
estimate the risk factor–outcome association, which is
controlled trials, in which the random allocation of
ing factors across each group (11).
Results of twin studies have suggested a heritable
component of circulating 25-OH-vitamin D concen-
trations, with heritability rates ranging from 29% to
77% (12, 13, 14). Genomewide association studies
identified 3 common loci of genetic determinants for
vitamin D insufficiency: group-specific component
(GC, rs2282679),97-dehydrocholesterol reductase
(DHCR7, rs12785878), and cytochrome P450 IIR-1
synthesis, and hydroxylation (15, 16).
In the present study we analyzed the roles of these
three 25-OH-vitamin-D–associated single-nucleotide
polymorphisms (SNPs) in all-cause cardiovascular and
The Ludwigshafen Risk and Cardiovascular Health
evaluate the effects of genetic polymorphisms and
plasma biomarkers on cardiovascular health. Partici-
pants are consecutive white patients hospitalized for
coronary angiography between June 1997 and May
2001. The study was approved by the ethics review
committee at the “Landesärztekammer Rheinland-
Pfalz” (Mainz, Germany). Written informed consent
was obtained from each of the participants.
A detailed description of the LURIC study design and
baseline characteristics has been published previously
ticipants. According to the classification of the Ameri-
can Heart Association, coronary artery disease (CAD)
ing (?20% stenosis) in at least 1 of 15 coronary seg-
such as type 2 diabetes, hypertension, and smoking
were assessed. Type 2 diabetes mellitus was diagnosed
according to the criteria of the American Diabetes As-
sociation. Further, individuals with a history of type 2
diabetes or those receiving oral antidiabetics or insulin
were considered diabetic (19). Hypertension was de-
fined as a systolic and/or diastolic blood pressure ex-
habits were retrieved using questionnaires. To detect
“hidden” smokers, plasma cotinine concentrations
were determined using a commercial RIA (cotinine
RIA; DPC). Individuals suffering from acute illnesses
diac diseases, and a history of malignancy within the
past 5 years were not eligible.
A fasting blood sample was obtained in the morn-
ing before coronary angiography. Selected variables
were measured after samples were frozen and stored at
?80 °C. 25-OH-vitamin D concentrations were deter-
and interassay CVs of 8.6% and 9.2%, respectively.
local registries. Death certificates were used to classify
the diseased individuals into those who died from car-
diovascular vs noncardiovascular causes. This classifi-
cation was done independently by 2 experienced clini-
cians who were blinded to any data on the study
participants except the information that was required
to classify the causes of death.
Genomic DNA was prepared from EDTA anticoagu-
lated peripheral blood by using a common salting-out
5?-exonuclease assays (TaqMan™). Primer and probe
sets were designed and manufactured by Applied Bio-
systems (Life Tech). Assay IDs were C__26407519_10
phism). Endpoint fluorescence was measured in a
POLARstar plate reader (BMG Labtech). Fluorescence
data were exported into Excel format and analyzed as
scatter plots. As a QC measure, genotyping was re-
peated in 184 samples and no discrepancies were
Statistical analysis was done using PASW 18.0.0 soft-
ware (IBM). Continuous variables were compared be-
tween groups by univariate ANOVA. A linear regres-
25-OH-vitamin D concentrations. Cox regression in-
cluding GC, DHCR7, and CYP2R1 genotypes, age, sex,
CAD, smoking, and type 2 diabetes mellitus was used
ity (all cause, cardiovascular, and noncardiovascular).
For regression analyses, an allelic model based upon
9Human genes: GC, group-specific component; DHCR7, 7-dehydrocholesterol
reductase; CYP2R1, cytochrome P450 IIR-1.
Clinical Chemistry 59:5 (2013)
1 (heterozygous), or 2 (homozygous for minor allele).
Seasonal variations of vitamin D concentrations were
modeled using categorical variables for the month of
blood taking. The criterion for statistical significance
was P ? 0.05.
The investigated cohort consisted of 3316 persons, in-
cluding 2310 men (69.7%) and 1006 women (30.3%).
Demographic data for the study population are given
in Table 1. GC, DHCR7, and CYP2R1 genotypes
were successfully determined in 3130 (94.4%), 3109
(93.8%), and 2980 (89.9%) participants (Table 2). For
the GC and the DHCR7 polymorphism, minor alleles
(GC G-allele, DHCR7 G-allele) were associated with
lower 25-OH-vitamin D concentrations. For the
CYP2R1 polymorphism, the minor allele (CYP2R1
A-allele) was associated with higher 25-OH-vitamin D
concentrations. Genotype frequencies did not deviate
from the Hardy–Weinberg equilibrium.
25-OH-vitamin D concentrations were available
from 3299 patients (99.5% of the entire study popula-
of blood sampling, age, and sex, vitamin D concentra-
tions were predicted by GC genotype (P ? 0.001),
CYP2R1 genotype (P ? 0.068), and DHCR7 genotype
(P ? 0.001), with a coefficient of determination (r2) of
(30.0% of the study population at baseline) had died.
Of these, 619 deaths were from cardiovascular causes
could not be classified because of insufficient data
about the cause of death.
In multivariate Cox regression models including
age, sex, type 2 diabetes, CAD, smoking habit, 25-OH-
vitamin D concentrations, GC genotype, CYP2R1 ge-
notype, and DHCR7 genotype, none of the genotypes
was significantly associated with all-cause mortality,
cardiovascular mortality, or noncardiovascular mor-
tality (Table 3).
Table 1. Demographic data of the LURIC study.
Age, years, mean (SD)
Height, cm, mean (SD)
Weight, kg, mean (SD)
25-OH-vitamin D, ng/mL, mean (SD)
nmol/L, mean (SD)
Vitamin D insufficiency,an (%)
CAD, n (%)
Current smoker, n (%)
Type 2 diabetes, n (%)
Hypertension, n (%)
aVitamin D insufficiency: 25-OH-vitamin D ? 20 ng/mL (? 49.9 nmol/L).
Table 2. 25-OH-vitamin-D–associated genotypes and 25-OH-vitamin-D concentrations.a
25-OH-vitamin D, ng/mL,bmean (SD)
GC genotype TT
GC G allele frequency
DHCR7 genotype TT
DHCR7 G allele frequency
CYP2R1 genotype GG
CYP2R1 A allele frequency
a25-OH-vitamin D concentrations according to GC (rs2282679), DHCR7 (rs12785878), and CYP2R1 (rs10741657) genotypes in the LURIC study.
bTo convert 25-OH-vitamin D concentrations into nanomoles per liter, multiply by 2.496.
cUnivariate linear regression.
Vitamin D and Mortality: A Mendelian Randomization Study
Clinical Chemistry 59:5 (2013)
We used a linear regression model including GC
genotype, CYP2R1 genotype, and DHCR7 genotype to
predict a genetically determined 25-OH-vitamin D
concentration for each study participant. In that
model, the genetically determined 25-OH vitamin D
concentration (ng/mL) was 18 158 ? (GC genotype ?
1.544) ? (DHCR7 genotype ? 0.981) ? (CYP2R1 ge-
notype ? 0.588). The mortality hazard ratio per 1
ng/mL (2.5 nmol/L) genetically determined 25-OH-
vitamin D concentration was 1.015 (95% CI, 0.962–
per 1 ng/mL (2.5 nmol/L) serum 25-OH-vitamin con-
centration was 0.951 (95% CI, 0.943–0.959; P ?
0.001). The point estimate of the effect of the geneti-
cally determined (i.e., causal) 25-OH-vitamin D con-
centrations was not within the 95% CI of the effect of
that the 2 estimates are truly different.
In the present study we aimed to test the hypothesis
that genetically lowered concentrations of 25-OH-
vitamin D causally increase mortality. Using a Men-
delian randomization method, we analyzed SNPs as-
sociated with 25-OH-vitamin D concentrations and
found that genetically determined differences in 25-
OH-vitamin D concentrations were not associated
with mortality rates. This lack of found association
The biological basis of the predictive power of 25-
ological conditions such as CVD and cancer remains
unclear. A potential explanation might be a reverse
in CVD or cancer could result in decreased 25-OH-
vitamin D. Interestingly, similar results have been
found for C-reactive protein, where C-reactive protein
plasma concentrations were predictive for specific dis-
eases, but Mendelian randomization studies found no
evidence for a causal role of this biomarker (20, 21).
It should be mentioned that our data are limited
by the fact that we investigated a cohort of patients
referred to coronary angiography and our results may
not be generalizable to patients with other diseases.
tant factors predicting mortality were age, sex, smok-
ing, type 2 diabetes, and CAD. Although these factors
have been considered in the statistical analysis as con-
founders, we cannot rule out residual confounding by
Strengths of the present study are the in-depth
clinical and biochemical characterization of all pa-
tients, the high number of participants, and the long
time of follow-up, such that the primary end point
death was reached in almost one-third of the study
It should be kept in mind that the present study
was observational and not aimed to investigate the po-
tential benefits of vitamin D supplementation. Inter-
estingly, in a recent large metaanalysis of randomized
we would like to emphasize that the present results do
not invalidate the predictive power of 25-OH-vitamin
D concentrations for mortality. The biological path-
ways for the association between 25-OH-vitamin D
Table 3. 25-OH-vitamin-D–associated genotypes and risk for all-cause, cardiovascular, and
All-cause mortality Cardiovascular mortalityNoncardiovascular mortality
HR (95% CI), PHR (95% CI), P HR (95% CI), P
GC, G alleles
CYP2R1, A alleles
DHCR7, G alleles
25-OH-vitamin D, ng/mL
Type 2 diabetes yes/no
0.95 (0.86–1.06); 0.35
1.00 (0.91–1.10); 0.95
0.94 (0.84–1.05); 0.26
0.96 (0.95–0.97); ?0.001
1.07 (1.06–1.08); ?0.001
0.60 (0.51–0.71); ?0.001
1.49 (1.23–1.79); ?0.001
1.59 (1.39–1.82); ?0.001
1.49 (1.20–1.85); 0.001
0.90 (0.78–1.02); 0.106
0.94 (1.83–1.06); 0.313
0.96 (0.84–1.11); 0.604
0.96 (0.95–0.97); ?0.001
1.07 (1.06–1.08); ?0.001
0.63 (0.51–0.76); ?0.001
1.33 (1.04–1.69); 0.023
1.83 (1.54–2.18); ?0.001
1.68 (1.26–2.24); ?0.001
1.03 (0.86–1.22); 0.768
1.08 (0.92–1.27); 0.362
0.89 (0.73–1.07); 0.200
0.97 (0.95–0.98); ?0.001
1.08 (1.06–1.09); ?0.001
0.56 (0.43–0.74); ?0.001
1.81 (1.35–2.42); ?0.001
1.26 (1.00–1.58); 0.054
1.17 (0.84–1.63); 0.351
aA Cox regression model including GC, CYP2R1, and DHCR7 genotypes, age, smoking, type 2 diabetes, hypertension, and coronary artery disease to predict
Clinical Chemistry 59:5 (2013)
the intellectual content of this paper and have met the following 3 re-
quirements: (a) significant contributions to the conception and design,
acquisition of data, or analysis and interpretation of data; (b) drafting
or revising the article for intellectual content; and (c) final approval of
the published article.
uscript submission, all authors completed the author disclosure form.
Disclosures and/or potential conflicts of interest:
Employment or Leadership: None declared.
Consultant or Advisory Role: None declared.
Stock Ownership: None declared.
Honoraria: None declared.
Research Funding: LURIC funding from the Sixth Framework Pro-
gram (integrated project Bloodomics, grant LSHM-CT-2004-
503485) and the Seventh Framework Program (integrated project
Atheroremo, grant agreement number 201668) of the European
Union and BioPersMed (COMET K-project 825329), which is
funded by the Federal Ministry of Transport, Innovation and Tech-
nology (BMVIT) and the Federal Ministry of Economics and La-
bour/the Federal Ministry of Economy, Family and Youth (BMWA/
BMWFJ) and the Styrian Business Promotion Agency (SFG).
Expert Testimony: None declared.
Patents: None declared.
Role of Sponsor: The funding organizations played no role in the
of data, or preparation or approval of manuscript.
study and takes responsibility for the integrity of the data and the
accuracy of the data analysis.
1. Holick MF. Vitamin D deficiency. N Engl J Med
2. Bouillon R, Carmeliet G, Verlinden L, van Etten E,
Verstuyf A, Luderer HF, et al. Vitamin D and
human health: lessons from vitamin D receptor
null mice. Endocr Rev 2008;29:726–76.
3. Pilz S, Tomaschitz A, März W, Drechsler C, Ritz E,
Zittermann A, et al. Vitamin D, cardiovascular
disease and mortality. Clin Endocrinol 2011;75:
4. Pilz S, Tomaschitz A. Role of vitamin D in arterial
hypertension. Expert Rev Cardiovasc Ther 2010;
5. Wehr E, Trummer O, Giuliani A, Gruber HJ, Pieber
TR, Obermayer-Pietsch B. Vitamin D-associated
polymorphisms are related to insulin resistance
and vitamin D deficiency in polycystic ovary syn-
drome. Eur J Endocrinol 2011;164:741–9.
6. Chiu KC, Chu A, Go VL, Saad MF. Hypovitamin-
osis D is associated with insulin resistance and
beta cell dysfunction. Am J Clin Nutr 2004;79:
7. Dobnig H, Pilz S, Scharnagl H, Renner W, Seel-
horst U, Wellnitz B, et al. Independent association
of low serum 25-hydroxyvitamin d and 1,25-
dihydroxyvitamin d levels with all-cause and car-
diovascular mortality. Arch Intern Med 2008;168:
8. Pilz S, Dobnig H, Nijpels G, Heine RJ, Stehouwer
CD, Snijder MB, et al. Vitamin D and mortality in
older men and women. Clin Endocrinol 2009;71:
9. Melamed ML, Michos ED, Post W, Astor B. 25-
hydroxyvitamin D levels and the risk of mortality
in the general population. Arch Intern Med 2008;
10. Mithal A, Wahl DA, Bonjour JP, Burckhardt P,
Dawson-Hughes B, Eisman JA, et al. Global vita-
min D status and determinants of hypovitamino-
sis. Osteoporos Int 2009;20:1807–20.
11. Bochud M, Rousson V. Usefulness of Mendelian
randomization in observational epidemiology. Int
J Environ Res Public Health 2010;7:711–28.
12. Orton SM, Morris AP, Herrera BM, Ramagopalan
SV, Lincoln MR, Chao MJ, et al. Evidence for
genetic regulation of vitamin D status in twins
with multiple sclerosis. Am J Clin Nutr 2008;88:
13. Hunter D, De Lange M, Snieder H, MacGregor AJ,
Swaminathan R, Thakker RV, et al. Genetic con-
tribution to bone metabolism, calcium excretion,
and vitamin D and parathyroid hormone regula-
tion. J Bone Miner Res 2001;16:371–8.
14. Shea MK, Benjamin EJ, Dupuis J, Massaro JM,
Jacques PF, D’Agostino RB, et al. Genetic and
non-genetic correlates of vitamins K and D. Eur
J Clin Nutr 2009;63:458–64.
15. Wang TJ, Zhang F, Richards JB, Kestenbaum B,
van Meurs JB, Berry D, et al. Common genetic
determinants of vitamin D insufficiency: a
genome-wide association study. Lancet 2010;
16. Ahn J, Yu K, Stolzenberg-Solomon R, Simon KC,
McCullough ML, Gallicchio L, et al. Genome-wide
association study of circulating vitamin D levels.
Hum Mol Genet 2010;19:2739–45.
17. Winkelmann BR, März W, Boehm BO, Zotz R,
Hager J, Hellstern P, et al. Rationale and design
of the LURIC study–a resource for functional
genomics, pharmacogenomics and long-term
prognosis of cardiovascular disease. Pharmacog-
enomics 2001;2(1 Suppl 1):S1–73.
18. Austen WG, Edwards JE, Frye RL, Gensini GG,
Gott VL, Griffith LS, et al. A reporting system on
patients evaluated for coronary artery disease.
Report of the Ad Hoc Committee for Grading of
Coronary Artery Disease, Council on Cardiovas-
cular Surgery, American Heart Association. Circu-
lation. 1975;51(4 Suppl):5–40.
19. Expert Committee on the Diagnosis and Classifi-
cation of Diabetes Mellitus, report of the Expert
Committee on the diagnosis and classification of
diabetes mellitus. Diabetes Care 1997;20:1183–
20. Marott SC, Nordestgaard BG, Zacho J, Friberg J,
Jensen GB, Tybjaerg-Hansen A, et al. Does ele-
vated C-reactive protein increase atrial fibrillation
risk? A Mendelian randomization of 47,000 indi-
viduals from the general population. J Am Coll
21. Allin KH, Nordestgaard BG, Zacho J, Tybjaerg-
Hansen A, Bojesen SE. C-reactive protein and the
risk of cancer: a mendelian randomization study.
J Natl Cancer Inst 2010;102:202–6.
22. Bjelakovic G, Gluud LL, Nikolova D, Whitfield K,
Wetterslev J, Simonetti RG, et al. Vitamin D sup-
plementation for prevention of mortality in
adults. Cochrane Database Syst Rev 2011;7:
Vitamin D and Mortality: A Mendelian Randomization Study
Clinical Chemistry 59:5 (2013)