See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/49679775

Effect of Vitamin D Supplementation on Testosterone Levels in Men

Article  in  Hormone and Metabolic Research · December 2010

DOI: 10.1055/s-0030-1269854 · Source: PubMed

CITATIONS READS

174 15,433

8 authors, including:

Stefan Pilz Sabine Frisch

Medical University of Graz University Hospital Essen
475 PUBLICATIONS   23,675 CITATIONS    10 PUBLICATIONS   852 CITATIONS   

SEE PROFILE SEE PROFILE

Joachim Kuhn Jens Dreier

Herz- und Diabeteszentrum Nordrhein-Westfalen Herz- und Diabeteszentrum Nordrhein-Westfalen
126 PUBLICATIONS   3,281 CITATIONS    465 PUBLICATIONS   3,314 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Health economics of food fortification View project

EPATH Trial View project

All content following this page was uploaded by Stefan Pilz on 09 February 2014.

The user has requested enhancement of the downloaded file.

 Short Communication

Effect of Vitamin D Supplementation on Testosterone

Levels in Men

Authors S. Pilz1,*, S. Frisch2,*, H. Koertke2, J. Kuhn3, J. Dreier3, B. Obermayer-Pietsch1, E. Wehr1, A. Zittermann2

1
Affiliations Department of Internal Medicine, Division of Endocrinology and Metabolism, Medical University of Graz, Austria
2
Clinic for Thoracic and Cardiovascular Surgery, Heart Centre North Rhine-Westphalia, Ruhr University Bochum,
Bad Oeynhausen, Germany
3
Institute for Laboratory and Transfusion Medicine, Heart Centre North Rhine-Westphalia, Ruhr University Bochum,
Bad Oeynhausen, Germany

Downloaded by: Bibliothek der Medizinischen Universität Graz. Copyrighted material.

Abstract (9.09–55.28 nmol/l for males aged 20–49 years)
▼ in both groups. Mean circulating 25(OH)D con-
The male reproductive tract has been identified centrations increased significantly by 53.5 nmol/l
as a target tissue for vitamin D, and previous data in the vitamin D group, but remained almost
suggest an association of 25-hydroxyvitamin D constant in the placebo group. Compared to
[25(OH)D] with testosterone levels in men. We baseline values, a significant increase in total
therefore aimed to evaluate whether vitamin D testosterone levels (from 10.7 ± 3.9 nmol/l to
supplementation influences testosterone levels 13.4 ± 4.7 nmol/l; p < 0.001), bioactive testoster-
in men. Healthy overweight men undergoing a one (from 5.21 ± 1.87 nmol/l to 6.25 ± 2.01 nmol/l;
weight reduction program who participated in p = 0.001), and free testosterone levels (from
a randomized controlled trial were analyzed for 0.222 ± 0.080 nmol/l to 0.267 ± 0.087 nmol/l;
testosterone levels. The entire study included p = 0.001) were observed in the vitamin D supple-
200 nondiabetic subjects, of whom 165 par- mented group. By contrast, there was no signifi-
ticipants (54 men) completed the trial. Partici- cant change in any testosterone measure in the
pants received either 83 μg (3 332 IU) vitamin D placebo group. Our results suggest that vitamin
daily for 1 year (n = 31) or placebo (n = 23). Initial D supplementation might increase testosterone
25(OH)D concentrations were in the deficiency levels. Further randomized controlled trials are
range ( < 50 nmol/l) and testosterone values warranted to confirm this hypothesis.
were at the lower end of the reference range

received 04.08.2010
Introduction [2, 3]. These results are of particular interest
accepted 03.11.2010
▼ because both, vitamin D deficiency and hypogo-
Bibliography
Vitamin D deficiency is currently considered an nadism are associated with skeletal diseases (e. g.,
DOI http://dx.doi.org/ important public health problem being associ- osteoporosis or muscle weakness) as well as
10.1055/s-0030-1269854 ated with musculoskeletal diseases, cardiovascu- extra-skeletal disorders (e. g., cardiovascular dis-
Published online: 2010 lar disease, cancer, and infectious and ease or obesity) [1, 4, 5]. Recently, some of us [6]
Horm Metab Res autoimmune diseases [1]. The vitamin D receptor have shown in 2 299 men referred for coronary
© Georg Thieme Verlag KG (VDR), as well as key enzymes for vitamin D angiography that 25-hydroxyvitamin D [25(OH)D]
Stuttgart · New York
metabolism, are widely expressed in human tis- levels are significantly associated with testoster-
ISSN 0018-5043
sues and cells [2]. In this context, Blomberg one levels and that both hormones reveal similar
Correspondence
Jensen et al. [3] observed significant expressions seasonal variations with a peak at the end of
A. Zittermann of the VDR and vitamin D metabolizing enzymes summer. Whether there exists a causal link
Clinic for Thoracic and Cardio- in the male reproductive tract including Leydig between vitamin D and testosterone status is,
vascular Surgery cells of the testis. These data raised the question however, currently not known. Therefore, a sub-
Heart Centre North Rhine- whether vitamin D is able to influence male group analysis of a previously published prospec-
Westfalia reproductive hormone production. The existence tive, randomized vitamin D supplementation
Ruhr University Bochum
of such an effect is supported by previous studies trial was performed in overweight subjects [7].
Georgstraße 11
32545 Bad Oeynhausen suggesting that vitamin D deficiency may con- Here, we present results on serum testosterone
Germany tribute to reduced fertility and hypogonadism concentrations in the male participants of this
Tel.: + 49/5731/97 1912 study.
Fax: + 49/5731/97 2020
[email protected] * Both authors contributed equally to the present work.

Pilz S et al. Vitamin D and Testosterone … Horm Metab Res

 Short Communication

Table 1 Characteristics of the study groups at baseline and at the end of the study

Parameter Placebo group Vitamin D group p-Value

Baseline Study end Baseline Study end 2 vs. 4 2 vs. 3 4 vs. 5
Number 23 31 – – –
Age (years) 46.8 ± 12.0 49.4 ± 10.2 0.387 – –
Smokers ( %) 56.5 38.7 0.271 – –
Alcohol (g/d) 20.0 ± 19.5 14.1 ± 15.3 17.7 ± 15.1 15.3 ± 13.8 0.646 0.138 0.703
25(OH)D (nmol/l) 29.7 ± 23.7 35.5 ± 8.1 32.5 ± 20.0 86.4 ± 68.8 0.659 0.215 < 0.001
1,25(OH)2D (pmol/l) 77.0 ± 25.9 97.4 ± 32.9 96.0 ± 39.6 127.7 ± 94.3 0.053 0.027 0.100
PTH (pmol/l) 5.07 ± 3.63 4.13 ± 1.48 4.14 ± 1.98 3.54 ± 1.76 0.237 0.035 0.040
Body weight (kg) 105.7 ± 14.3 99.0 ± 13.5 109.9 ± 16.1 104.0 ± 17.2 0.323 < 0.001 < 0.001
BMI (kg/m2) 32.5 ± 3.8 30.5 ± 4.1 33.1 ± 3.9 31.2 ± 3.9 0.609 < 0.001 < 0.001
Albumin (mmol/l) 386 ± 182 302 ± 125 377 ± 194 297 ± 186 0.087 0.210 0.896
SHBG (mmol/l) 26.3 ± 13.7 29.5 ± 17.3 31.0 ± 10.3 35.3 ± 13.6 0.153 0.046 0.002
TT (nmol/l) 11.8 ± 4.0 12.7 ± 5.5 10.7 ± 3.9 13.4 ± 4.7 0.317 0.355 < 0.001
BAT (nmol/l) 6.39 ± 2.22 6.59 ± 2.33 5.21 ± 1.87 6.25 ± 2.01 0.040 0.626 0.001
fT (nmol/l) 0.264 ± 0.087 0.278 ± 0.097 0.222 ± 0.080 0.267 ± 0.087 0.067 0.532 0.001
Data are shown as means ± SD. Inter-group comparisons were performed by unpaired t-test and intra-group comparisons by paired t-test

Downloaded by: Bibliothek der Medizinischen Universität Graz. Copyrighted material.

25(OH)D: 25-hydroxyvitamin D; 1,25(OH)2D: 1,25-dihydroxyvitamin D; PTH: parathyroid hormone; BMI: body mass index; SHBG: sex-hormone binding

Subjects and Methods Baseline characteristics stratified by treatment group (vitamin D

▼ vs. placebo) are presented as means ± SD for continuous varia-
Study subjects were derived from a weight reduction program bles. Intra-group comparisons (paired t-test) rather than inter-
over 12 months, which included a daily supplementation of group comparisons were used at the end of the study because no
either 83 μg (3 332 IU) vitamin D or placebo as part of a double- sex-stratified randomization at baseline was performed. Statisti-
blind randomized controlled trial [7]. Details on the study design cal analyses were performed by SPSS 16.0 (SPSS Inc, Chicago,
and major outcomes of the trial have been published previously USA) and a p-value below 0.05 was considered statistically
[7]. Out of 200 nondiabetic individuals (62 men) who were significant.
included in the study, 165 (54 men) completed the trial. Only
male study subjects were analyzed for the present work. The
number of dropouts (3 men in the placebo and 5 men in the Results
vitamin D group) did not differ significantly between groups ▼
(p = 0.745). The original study was registered at clinical trials.gov Baseline characteristics of the 54 male patients who completed
as NCT004493012. the trial are shown in ● ▶ Table 1. At study entry, mean 25(OH)D

Participants were continuously recruited from December 2005 concentrations were in the deficiency range in both groups. Dur-
to October 2006 throughout the year. Fasting blood samples ing follow-up, weight loss was 5.9 ± 5.3 kg (p < 0.001) in the vita-
were drawn at study begin and after 1 year. Specimens were min D group and 6.6 ± 5.7 kg in the placebo group (p < 0.001), and
centrifuged at room temperature. Thereafter, serum aliquots thus similar in both groups. Circulating 25(OH)D increased by
were stored at − 80 ° C until analyses. To avoid inter-assay varia- 53.5 ± 65.3 nmol/l to 86.4 ± 68.8 nmol/l in the vitamin D group
tions, the samples of each participant were analyzed within the (p < 0.001), but increased only nonsignificantly in the placebo
same assay run. Concentrations of 25(OH)D were determined by group (increase by 5.8 ± 21.3 nmol/l to 35.5 ± 18.0 nmol/l;
means of a radioimmunoassay (DiaSorin, Stillwater, MN, USA) p = 0.215). PTH decreased in the placebo and vitamin D group
with an intra-assay CV of < 7 %. According to Holick [1], vitamin (decrease by 0.94 ± 3.09; p = 0.035, and 0.60 ± 1.67; p = 0.040,
D deficiency is defined as a 25(OH)D level of less than 50 nmol/l, respectively), whereas 1,25(OH)2D tended to increase in both
whereas a level of 52.5–72.5 nmol/l indicates a relative insuffi- groups (increase by 20.4 ± 41.0; p = 0.027 and 21.7 ± 106.0;
ciency, and a level of 75 nmol/l or greater indicates sufficient p = 0.100, respectively). At baseline, mean testosterone values
vitamin D. The serum concentrations of 1,25-dihydroxyvitamin were at the lower end of the reference range in both groups. By
D [1, 25(OH)2D] were measured by a test kit provided by Immun- comparing baseline testosterone values with follow-up values in
diagnostik (Bensheim, Germany). The serum levels of parathy- the placebo group no significant change in TT (11.8 ± 4.0 nmol/l
roid hormone (PTH), total testosterone (TT), and sex hormone vs. 12.7 ± 5.45 nmol/l, p = 0.355), BAT (6.39 ± 2.22 nmol/l vs.
binding globulin (SHBG) were analyzed by using the Immulite 6.59 ± 2.33 nmol/l, p = 0.626) or fT (0.264 ± 0.087 nmol/l vs.
2000 system (Siemens, Munich, Germany). The reference range 0.278 ± 0.097 nmol/l, p = 0.532) was found. In the vitamin D
for total testosterone is 9.09–55.28 nmol/l for males aged 20–49 group, however, a significant increase in all measures of testos-
years and 6.28–26.30 nmol/l for males aged ≥ 50 years. The SHBG terone status was observed. TT increased from 10.7 ± 3.9 nmol/l
reference range for males is 13–71 nmol/l. The within-run and to 13.4 ± 4.7 nmol/l (p < 0.001), BAT from 5.21 ± 1.87 nmol/l to
total coefficients of variation for SHBG and testosterone are 2.5 % 6.25 ± 2.01 nmol/l (p = 0.001) and fT from 0.222 ± 0.080 nmol/l to
and 5.2 %, respectively. Serum albumin was measured by using 0.267 ± 0.087 nmol/l (p = 0.001). In the placebo group, there were
the Architect autoanalyzer (Abbott, Wiesbaden, Germany). Bio- nonsignificant trends for seasonal differences in 25(OH)D and
active testosterone (BAT; reference range: 2.14–13.60 nmol/l) testosterone values. Compared with men recruited in the sum-
and free testosterone (fT; reference range: 0.090–0.580 nmol/l) mer half-year (mid April to mid October; n = 12), men recruited
were calculated according to Vermeulen et al. [8]. in the winter half-year (mid October to mid April; n = 11) had
lower values of 25(OH)D (21.8 ± 9.8 nmol/l vs. 37.4 ± 30.0 nmol/l;

Pilz S et al. Vitamin D and Testosterone … Horm Metab Res

 Short Communication

p = 0.113), TT (11.5 ± 4.33 nmol/l vs. 13.29 ± 4.15 nmol/l; p = 0.336), mones do not exclude group-specific effects on the reproductive
BAT (6.04 ± 1.91 nmol/l vs. 7.37 ± 2.58 nmol/l; p = 0.173), and fT system, since nonclassical target tissues for vitamin D largely
(0.255 ± 0.078 nmol/l vs. 0.301 ± 0.104 nmol/l; p = 0.250). depend on circulating 25(OH)D levels [1], which differed mark-
In the 54 men, body mass index changes were inversely related edly between the vitamin D and placebo group.
to SHBG levels (r = –0.485; p < 0.001), but not to other indices of Our study has both strengths and limitations. Strengths are the
testosterone status. study design, the use of a daily vitamin D dose that was effective
to increase 25(OH)D values from the deficiency range into the
adequate range, and the fact that sample batching was per-
Discussion formed to avoid inter-assay variability. One limitation is the fact
▼ that the effect of vitamin D supplementation on testosterone
In overweight men with deficient vitamin D status a significant was not a prespecified study outcome and that we did not assess
increase in testosterone was observed after intake of 83 μg vita- testosterone-related functions such as libido, mood, or muscle
min D daily for 1 year whereas there was no significant change strengths. Another limitation is the relatively small number of
in men receiving placebo. This work is, to the best of our knowl- male study participants. In addition, future studies have to clar-
edge, the first study, which specifically addresses the effect of ify whether the vitamin D actions are mediated by a pituitary
vitamin D supplementation on androgens in men. The results of effect or a testicular one.
this study suggest that vitamin D supplementation might In conclusion, our study results suggest that vitamin D supple-
increase testosterone levels in men. Our data support several mentation might increase testosterone levels in men. Further

Downloaded by: Bibliothek der Medizinischen Universität Graz. Copyrighted material.

experimental and clinical findings: First, VDR knockout mice randomized controlled trials are needed to confirm this hypoth-
suffer from hypergonadotropic hypogonadism [2]. Second, vita- esis and to evaluate whether vitamin D driven increases in testos-
min D status is directly associated with testosterone levels in terone levels contribute to the vitamin D effects on various
men [6]. Third, the male reproductive tract is a target tissue for health outcomes.
vitamin D effects [3]. The nonsignificant trend for seasonal dif-
ferences in both 25(OH)D and testosterone in the placebo group
supports our hypothesis of a vitamin D effect on testosterone. Acknowledgements
In our study participants, mean baseline 25(OH)D values were in ▼
the deficiency range and mean testosterone values were at the We would like to thank Mrs. Marlen Ewald for excellent techni-
lower end of the reference range. Traditionally, low solar ultra- cal assistance.
violet B irradiation of the skin is a major cause of vitamin D defi-
ciency [1]. Both, vitamin D [1] and testosterone [5, 9] show References
beneficial effects on the musculoskeletal system. From an evolu- 1 Holick MF. N Engl J Med 2007; 357: 266–281
tionary point of view it would make sense that an active lifestyle 2 Bouillon R, Carmeliet G, Verlinden L, van Etten E, Verstuyf A, Luderer
HF, Lieben L, Mathieu C, Demay M. Endocr Rev 2008; 29: 726–776
(leading to an adequate skin synthesis of vitamin D) also has
3 Blomberg Jensen M, Nielsen JE, Jorgensen A, Rajpert-De Meyts E, Kris-
beneficial effects on muscle function, bone health, and the male tensen DM, Jorgensen A, Skakkebaek NE, Juul A, Leffers H. Hum Reprod
reproductive system. We are aware that no final conclusions can 2010; 25: 1303–1311
be drawn from our study regarding the effect of vitamin D sup- 4 Haring R, Völzke H, Steveling A, Krebs A, Felix SB, Schöfl C, Dörr M, Nauck
M, Wallaschofski H. Eur Heart J 2010; 31: 1494–1501
plementation on testosterone in men but we do believe that our 5 Allan CA, Strauss BJ, Burger HG, Forbes EA, McLachlan RI. J Clin Endo-
work is of great importance because it provides a reasonable crinol Metab 2008; 93: 139–146
rationale for future studies. Besides the marked increase in 6 Wehr E, Pilz S, Boehm BO, März W, Obermayer-Pietsch B. Clin Endocri-
25(OH)D levels in the vitamin D group, there was also a slight nol (Oxf) 2010; 73: 243–248
7 Zittermann A, Frisch S, Berthold HK, Götting C, Kuhn J, Kleesiek K, Stehle
(nonsignificant) increase in 25(OH)D in the placebo group dur- P, Koerke H, Koerfer R. Am J Clin Nutr 2009; 89: 1321–1327
ing follow-up. We assume that the similar decrease in PTH and 8 Vermeulen A, Verdnock L, Kaufman JM. J Clin Endocrinol Metab 1999;
the similar trend for an increase in 1,25(OH)2D in both study 84: 3666–3672
9 Krasnoff JB, Basaria S, Pencina MJ, Jasuja GK, Vasan RS, Ulloor J, Zhang
groups is due to a nonlinear association of these 2 calciotropic
A, Coviello A, Kelly-Hayes M, D’Agostino RB, Wolf PA, Bhasin S, Mura-
hormones with increasing circulating 25(OH)D levels [10], with bito JM. J Clin Endocrinol Metab 2010; 95: 2790–2799
a pronounced effect at low and virtually no effect at high 10 Zittermann A. Prog Biophys Mol Biol 2006; 92: 39–48
25(OH)D levels. Nevertheless, the similar changes in these hor-

Pilz S et al. Vitamin D and Testosterone … Horm Metab Res

View publication stats