Clinical Nutrition 33 (2014) 274e279

Contents lists available at SciVerse ScienceDirect

Clinical Nutrition
journal homepage: http://www.elsevier.com/locate/clnu

Original article

Factors associated with vitamin D deficiency in a population of 2044

HIV-infected patients
Maria Theodorou a, Thomas Sersté b, *, Marc Van Gossum b, d, Stéphane Dewit c, d
a
Internal Medicine, CHU Saint Pierre, Université libre de Bruxelles, Bruxelles, Belgium
b
Hepato-Gastroenterology and Clinical nutrition, CHU Saint Pierre, Université libre de Bruxelles, Bruxelles, Belgium
c
Infectious Diseases, CHU Saint Pierre, Université libre de Bruxelles, Bruxelles, Belgium

a r t i c l e i n f o s u m m a r y

Article history: Background and aims: The high prevalence of vitamin D deficiency in HIV-infected patients has been
Received 29 December 2012 demonstrated but there are still controversies regarding to the role of antiretroviral therapy (ART) in this
Accepted 22 April 2013 setting. The aims of this study was to validate factors associated with vitamin D deficiency in a large
cohort of HIV-infected patients.
Keywords: Methods: A retrospective analysis of 2044 consecutive patients from December 2005 to March 2011 was
Vitamin D
conducted. Factors independently associated with vitamin D deficiency (<30 ng/ml and <10 ng/ml) were
HIV
analyzed. Vitamin D levels were compared according to CD4 count, viral load and ART modalities.
Associated factors
Antiretroviral therapy
Results: vitamin D was <30 ng/ml in 89.2% and <10 ng/ml in 32.4%. The median value was 13.8 ng/ml (4
e102). Winter season, female sex, heterosexual acquisition of HIV, the need of second lines (complex and
sequential treatment modalities) and a longer duration of ART were independently associated with
vitamin D deficiency (<30 ng/ml). CD4 count <200/ml, advanced stages of disease and the current efa-
virenz use were independently associated with severe vitamin D deficiency (<10 ng/ml). Median vitamin
D levels was 14.1 ng/dl when CD4
200/ml, 11.5 ng/dl when CD4<200 (p ¼ 0.0003). The ART modalities
had a significant influence on vitamin D concentrations, the highest vitamin D level was observed in the
absence of treatment.
Conclusions: In HIV-infected patients, vitamin D deficiency is associated with ART modalities and
duration. The most severe vitamin D deficiencies are associated with low CD4 count, the use of efavirenz
and advanced stages of disease severity.
Ó 2013 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

1. Introduction malabsorption, multiple cardiovascular disease risk factors

Vitamin D plays a central role in the calcium and phosphate
metabolism. The role of vitamin D in reducing the risk of fractures
and increasing muscle strength is well known.1,2 Moreover, vitamin
D is also involved in cell differentiation, proliferation and immu-
nomodulation.3 It is commonly accepted that the normal serum
concentration of 25 (OH) vitamin D (25 (OH) D) is between 30 and
76 ng/ml and low vitamin D is generally defined as serum 25 (OH) D
concentrations below 30 ng/ml.4e8 vitamin D deficiency is associ-
ated with several pathological conditions such as the lack of
exposure to ultraviolet B (UVB), inadequate dietary intake,
* Corresponding author. Service d’Hépato-Gastroentérologie et Nutrition Clin-
ique, Hôpital Universitaire Saint-Pierre, 322 rue Haute, 1000 Bruxelles, Belgium.
Tel.: þ32 2 535 4109.
E-mail address: thomas.serste@skynet.be (T. Sersté).
d
These authors contributed equally to this work.
including diabetes mellitus, a body mass index (BMI) greater than
30 kg/m2 and current alcohol consumption.5,6 Female sex,
increasing age and dark skin pigmentation are also recognized as
risk factors for vitamin D deficiency in the general population.4
The relationship between chronic viral infections and hypo-
vitaminosis D is now well known. A high prevalence of vitamin D
deficiency has been specifically described within the HIV-positive
population,9 with rates as high as 83% in the EuroSIDA study.10,11
Several studies have focused on the parameters that influence
hypovitaminosis D in HIV-positive patients.10e12 The results of such
analyzes however remain discordant, mainly regarding to the
treatment modalities. Indeed, several studies have shown that
vitamin D concentrations are related with antiretroviral therapy
(ART) and with specific antiretroviral class use. Vitamin D con-
centration could be specifically lower in patients treated with non-
nucleoside reverse transcriptase inhibitors (NNRTIs) than in those
treated with protease inhibitors (PI).13,14 In particular efavirenz, an
NNRTI, has been described as independently associated with low

0261-5614/$ e see front matter Ó 2013 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
http://dx.doi.org/10.1016/j.clnu.2013.04.018
 M. Theodorou et al. / Clinical Nutrition 33 (2014) 274e279 275

vitamin D levels.15,16 But there are still discrepancies about the Table 1
deleterious role of NNRTI (and efavirenz in particular) on vitamin D Categories of treatment: their definitions, treatment combinations and their com-
positions (specific antiretroviral therapy in the category).
levels. In a recent work, Crushley et al. did not find any influence of
NNRTI on vitamin D while Welz et al. recently confirmed on a large Definition Treatment combinations Treatment composition
study the deleterious role of efavirenz on vitamine D concentra- of the category (% in the category)

tion.12,15 A larger study based on a wide cohort should then clarify First line 2 NRTI þ 1 NNRTI NRTI:
these uncertainties arising from previous works. treatment Lamivudine (30.46)
Abacavir (7.69)
In terms of HIV disease severity, vitamin D deficiency is asso-
Zidovudine (11.38)
ciated with a low CD4 count17 and with a viral load greater than 50 Stavudine (0.62)
copies/ml.18 Moreover, vitamin D deficiency is more frequent in Didanosine (5.23)
HIV-infected patients with advanced disease progressions.19,20 Emtricitabine (8.62)
NNRTI:
Finally, an increased risk of morbidity and mortality is observed
Darunavir (0.31)
in HIV-infected patients with low vitamin D levels.10 Up to now, no Efavirenz (15.08)
study takes into account in a multivariate analysis general variables Lopinavir (0.31)
(season, sex, ethnicity), parameters related to treatment (modality Nevirapine (8.00)
and duration of treatment) and data related to the disease severity Rilpivirine (0.31)
Ritonavir (0.31)
at once (viral load, CD4 count, stage of disease severity).
Tenofovir (11.69)
It was therefore important to conduct a large multivariable study First line 2 NRTI þ 1 PI NRTI:
aiming to identify risk factors independently associated with vitamin treatment Lamivudine (17,94)
D deficiency. Such a study should take into account previously Abacavir (3.52)
described variables and should lead to strong clinical conclusions. Zidovudine (3.75)
Stavudine (0.12)
The purpose of the present study was therefore to: 1) identify Didanosine (1.99)
factors associated with vitamin D deficiency in a HIV-infected Emtricitabine (10.08)
population, taking into account the vast majority of the variables PI:
that have been separately described in the literature; 2) remove Atazanavir(10.90)
Darunavir (0.12)
ambiguities that persist about the deleterious role of NNRTIs; 3)
Fosamprenavir (9.14)
compare median vitamin D concentrations according to the CD4 Indinavir (0.47)
count, the viral load and the antiretroviral class use. Lopinavir (9.38)
Ritonavir (16.30)
Saquinavir (0.70)
2. Materials and methods Tenofovir (15.59)
Second line Multiple combinations e
treatment and/or sequential treatments:
2.1. Population NRTI  NNRTI  PI  other
antiviral treatment
This was a single-centre retrospective study conducted in a No treatment Complete absence e
population of HIV-infected patients consecutively included from of antiretroviral therapy

December 2005 to March 2011. Patients were followed at the

Centre for Study and Treatment of Immunodeficiency (CETIM) of
the CHU Saint-Pierre (Brussels, Belgium), a high volume referential
center of infectious diseases. Biological and clinical data were
collected in this out patient clinic. Patients under the age of 18 years
and pregnant women were excluded from the study.
2.2. Determination of 25 (OH) vitamin D serum concentrations
Blood sampling for 25 (OH)D concentrations has been routinely
collected in all patients since 2005.21,22 For this study, the first

Fig. 1. Distribution of vitamin D levels in 2044 HIV-infected patients. (Legend: Effective of patients according to different categories of vitamin D concentration (ng/ml). Vitamin D
concentrations were not normally distributed and the median was 13.8 ng/ml (4e102).
276 M. Theodorou et al. / Clinical Nutrition 33 (2014) 274e279

25(OH)D measurement performed for each patient was taken into
account. If patients were receiving ART, the concentration should
have been measured at least 6 months after the start of treatment.
For patients who changed the treatment during the 6 months pre-
ceding the measurement of 25(OH)D, the treatment that was being
taken prior to the change was considered. Patients did not receive
any oral vitamin D supplementation during the study period.
vitamin D concentrations were measured using an immunoassay
from the LiaisonÒ system (DiaSorin, Stillwater, Minnesota, USA). The
inferior detection limit of the vitamin D assay was 3 ng/ml, there was
no superior detection limit. According to the widely accepted defi-
nition, vitamin D deficiency4e6 was here defined as 25(OH)D con-
centrations below 30 ng/ml at entry and severe vitamin D deficiency
was identified as concentrations of 10 ng/ml or less.
2.3. Variables and study design
For each patient, the following data were collected: age, sex,
ethnicity, the season in which the measurement of vitamin D was
obtained, smoking habits, nutritional status (BMI, albumin, choles-
terol, LDL cholesterol, HDL cholesterol). Other biological markers
collected were: serum urea, creatinine, liver enzymes (AST, ALT, ALP)
and serum calcium. The mode of acquisition of HIV (heterosexual,
homosexual), the stage of disease severity according to the CDC
(Centre for Disease Control and Prevention) classification, the CD4
count, and the viral load was noted. The duration of ART and anti-
retroviral class use were carefully recorded. Different treatment
combinations were identified according to the classes of ART:
Nucleoside reverse transcriptase inhibitors (NRTIs), non-nucleoside
reverse transcriptase inhibitors (NNRTI), protease inhibitors (PI) and
other antiretroviral treatment. Those different combinations of ART
were classified as categories and are listed and defined in Table 1. The
“first line” treatment was the first therapeutic option. According to
the clinical practice in our centre during the study period, it was
systematically constituted with two types of combinations: either 2
NRTI and 1 NNRTI or 2 NRTI and 1 PI (Table 1). The “second line”
treatment was given in case of failure of the first line. This second
line was constituted of multiple, complex and/or sequential com-
bination of treatment, that were not corresponding to the first line
options (e.g. NRTI  NNRTI  PI  other antiviral treatment).
Aiming to identify the factors associated with vitamin D defi-
ciency, patients were divided into groups according to their vitamin
D concentrations, and their characteristics were compared. Two
different cut-offs were used to define the vitamin D deficiency:
first, below 30 ng/dL and second, below 10 ng/dL, as previously
defined. Two multivariable analyses were then performed to assess
independent variables associated with vitamin D deficiency.
Median 25(OH)D concentrations (used as a continuous variable)
were first compared according to the CD4 count (using a cut-off
defining low CD4 count fixed at 200 CD4/ml). Vitamin D concen-
trations were also compared according to the HIV viral load (using a
cut-off fixed at 50 copies/ml) and according to the treatment mo-
dalities as defined above (Table 1).
2.4. Statistical analysis
The distribution of variables was checked using the Kolmo-
goroveSmirnov test. Normally and non-normally distributed
continuous variables were reported as mean  SD and median and
range (minimum e maximum), respectively. Categorical variables
are reported as count and percentage. Group comparisons were
performed using a Student t test or Wilcoxon test for normally and
non-normally distributed continuous variables, respectively. Group
comparisons were made using the Chi2 test for categorical vari-
ables. Two multivariate analyses were performed to assess vari-
ables independently associated with vitamin D deficiency: for 25

Table 2
Patient characteristics, comparisons of variables according to vitamin D cut-off set at 30 ng/ml (univariate analysis).

Variable Whole group n ¼ 2044 Vit below 30 ng/ml n ¼ 1824 Vit D
30 ng/ml n ¼ 220 p Value

Season Winter, n (%) 1087 (53.2) 1005 (55.1) 82 (37.3) <0.0001

Sex Male, n (%) 1195 (58.5) 1029 (56.4) 166 (75.4) <0.0001
Age (years) 43 (20e85) 43.0 (20e85) 45 (22e73) 0.0509
Ethnic group Caucasian, n(%) 959 (46.9) 788 (43.2) 171 (77.7) <0.0001
African, (n%) 1008 (49.3) 964 (52.8) 44 (22.0) <0.0001
Tobacco use Presence, n (%) 969 (47.4) 847 (46.4) 122 (55.4) 0.0114
Body mass index (kg/m2) 24.4 (16.9e50.1) 24.55 (16.7e50.1) 23.6 (17.7e41.5) 0.0188
Serum albumin (g/dl) 4.3 (2.2e5.2) 4.3 (2.1e5.2) 4.45 (2.2e5.1) 0.2449
Cholesterol (mg/dl) 180 (85e430) 181 (85e430) 174 (83e305) 0.0027
HDL-cholesterol mg/dl) 50.1 (14.9e17.1) 50.7 (16.5e171.1) 46.4511.7e129.3) 0.0002
LDL-cholesterol (mg/dl) 102 (30e332) 102 (32e332) 99 (30e209) 0.2731
Serum calcium (mg/dl) 9 (7.6e11.8) 9 (7.6e11.8) 9.1 (7.9e10.0) 0.2901
Serum urea (mg/dl) 29 (12e161) 28 (12e161) 33 (11e74) <0.0001
Serum creatinine (mg/dl) 0.83 (0.5e5) 0.82 (0.5e5) 0.84 (0.49e1.69) 0.8231
Glomerular filtration rate (ml/min) 60 (39e190) 60 (39e177) 60 (44e190) 0.9991
ASAT (UI/l) 24 (12e1455) 24 (12e499) 25 (13e1455) 0.2004
ALAT (UI/l) 22 (7e1668) 22 (7e342) 24 (8e1668) 0.0152
PAL (UI/l) 187 (84e2348) 187 (83e2348) 189.5 (96e943) 0.9872
Mode of acquisition of HIV eHeterosexual 1168 (57.1) 1090 (59.7) 78 (35.4) <0.0001
eHomo/bisexual 612 (29.9) 489 (26.8) 123 (55.9)
CDC Stage A 1340 (65.5) 1187 (65.1) 153 (69.5) 0.1014
B 358 (17.5) 317 (17.4) 41 (18.6)
C 346 (16.9) 320 (17.5) 26 (11.8)
HIV Viral load (Copies/ml) 50 (20e32000000) 50 (20e10000000) 85 (20e32000000) 0.0046
CD4 count (n/ml) 495 (9e2315) 492 (9e2315) 521 (11e1702) 0.1357
Antiretroviral therapy Presence, n (%) 1500 (73.4) 1362 (74.7) 138 (62.7) 0.0002
Antiviral treatment 2 NRTI þ 1 NNRTI 475 (23.2) 433 (23.7) 42 (19.1) 0.0002
2 NRTI þ 1 PI 734 (35.9) 659 (36.1) 75 (34.1)
eSecond line 291 (14.2) 270 (14.8) 21 (9.5)
eNo treatment 544 (26.6) 462 (25.3) 82 (37.3)
Presence of efavirenz Presence, n (%) 324 (15.8) 300 (16.4) 24 (10.9) 0.0013
Duration of treatment (Month) 195 (0e728) 203.5 (0e728) 69.5 (0e681) 0.0004
 M. Theodorou et al. / Clinical Nutrition 33 (2014) 274e279 277

Table 3 (OH)D below 30 ng/ml and below 10 ng/ml. For both analyzes,
Results of multivariate analysis according to vitamin D cut-off set at 30 ng/ml and every variables associated with hypovitaminosis in univariate
10 ng/ml. OR: Odds Ratio; CI: confidence interval.
analysis (with a significance level set at 0.10) were entered into the
Variables Odds 95% 95% p-Value system. The results of the two multivariate analyses are expressed
ratios CI inf CI sup as odds ratios with 95% confidence intervals. Comparisons of me-
Vitamin D deficiency set at 30 ng/ml dian vitamin D concentrations (continuous variable) between
Winter season 1.92 1.37 2.71 0.0002 groups according to CD4 counts, viral loads and type of ART were
Female gender 1.85 1.07 3.33 0.0367
performed using ANOVA. Multiple comparisons of different treat-
Homosexual acquisition of HIV 0.35 0.18 0.71 0.0031
Duration of treatment 1.002 1.001 1.003 0.0115 ment groups were performed using the TukeyeKramer test.
Second lines treatment 1.96 1.02 3.70 0.0446 All statistical analyzes were performed with an SAS 9.2 program
Vitamin D deficiency set at 10 ng/ml (SAS Institute, Cary, NC, USA).
Winter season 2.86 2.16 3.79 0.0001
Black ethnicity 1.58 1.09 2.30 0.0138
High Serum creatinine (
1.2 mg/dl) 2.23 1.19 4.20 0.0121 3. Results
Low LDL cholesterol (below 100 mg/dl) 1.61 1.21 2.13 0.0009
Homosexual acquisition of HIV 0.47 0.32 0.69 0.0002
3.1. Patient characteristics
Low CD4 count (below 200/ml) 1.38 1.03 1.85 0.0278
CDC stage C 2.21 1.50 3.26 0.0001
Treatment with Efavirenz 2.18 1.47 3.22 0.0001 A total of 2044 consecutive patients were included in the study.
Vitamin D deficiency (25(OH)D < 30 ng/ml) was present in 1824
patients, which corresponds to a prevalence of 89.24%. As shown in
Fig. 1, vitamin D concentrations were not normally distributed and

Fig. 2. Comparisons of vitamin D levels (medians) according to CD4 count, viral load, the presence of antiretroviral therapy and all combinations of treatments. (a). Comparison of
median vitamin D concentrations (ng/ml) according to CD4 count below 200/ml and CD4 count above 200/ml. (Legend (a): Patients with a CD4 count less than 200/ml had a
significantly lower median value of 25(OH)D than did patients with higher counts (11.5 ng/ml if CD4 below 200/ml versus 14.1 ng/ml, p ¼ 0.0003)). (b). Comparison of median
vitamin D concentrations (ng/ml) according to viral load (above 50 copies/ml and below 50 copies/ml). (Legend (b): There is a significant differences between the median 25(OH)D
values according to the viral load (14.4 ng/ml if viral load above 50 copies/ml versus 13.5 ng/ml, p ¼ 0.02)). (c). Comparison of median vitamin D concentrations (ng/ml) according to
the presence of antiretroviral therapy. (Legend (c): There is a statistical difference comparing medians of vitamin D levels according to the presence of ART (15.1 ng/ml without
treatment versus 13.3 ng/ml with treatment, p ¼ 0.0001). (d). Comparison of median vitamin D concentrations according to ART modalities: 2 NRTI þ 1 NNRTI, 2 NRTI þ 1 PI, second
line treatments (multiple and/or sequential treatment) and no treatment. (Legend (d): There is a statistical difference of vitamin D levels according the different modalities of
antiretroviral treatments (p < 0.0001). See full text for the results of multiple comparisons).
278 M. Theodorou et al. / Clinical Nutrition 33 (2014) 274e279
the median was 13.8 ng/ml (4e102). In 32.4% of patients, vitamin D
levels were less than 10 ng/ml (severe vitamin D deficiency). Clinical
characteristics, biochemical values, immunological status and an-
tiretroviral treatments are summarized in Table 2.
3.2. Factors associated with vitamin D deficiency
3.2.1. 25(OH)D cut off set at 30 ng/ml (univariate: Table 2,
multivariate: Table 3)
Patients were divided into two groups (<30 ng/ml and
30 ng/
ml) for comparison. Results of univariate analysis regarding factors
associated with a low serum vitamin D level were: the winter sea-
son, the female sex, the Black African ethnicity, the current use of
tobacco, a higher serum cholesterol, a higher HDL-cholesterol, a
lower serum urea, a lower ALAT level (Table 2). Vitamin D deficiency
was also related with a heterosexual acquisition of HIV, a lower viral
load, the presence of ART, the requirement of second line of ART and
a longer duration of ART. In the multivariate analysis, winter season,
female sex, heterosexual acquisition of HIV, the need of a second
line ART and duration of treatment were independently associated
with vitamin D deficiency <30 ng/ml (Table 3).
3.2.2. 25(OH)D cut off set at 10 ng/ml (multivariate: Table 3)
The multivariate analysis was performed selecting variables
associated with severe vitamin D deficiency (below 10 ng/ml).
Using this stricter definition of low vitamin D, independent factors
related with severe vitamin D deficiency were: Black African
ethnicity, the heterosexual acquisition of HIV, the winter season, a
higher serum creatinine, a lower serum LDL cholesterol (<100 mg/
dl), a lower CD4 count (<200 cells/ml), an advanced CDC stage
(stage C) and the use of efavirenz in the ART.
3.3. Vitamin D levels according to CD4 count, viral load (CV) and
treatment
Comparisons of vitamin D levels (medians) according to CD4
count, viral load, the presence of antiretroviral therapy and all
combinations of treatments are shown in Fig. 2aed respectively.
Patients with a CD4 count less than 200/ml had a significantly lower
median value of 25(OH)D than did patients with higher counts
(11.5 ng/ml if CD4 200/ml versus 14.1 ng/ml, p ¼ 0.0003, Fig. 2a).
There were significant differences in the median 25(OH)D values
according to the viral load (14.4 ng/ml if viral load above 50 copies/
ml versus 13.5 ng/ml, p ¼ 0.02, Fig. 2b). There was a statistical
difference comparing medians of vitamin D levels according to the
presence of ART (15.1 ng/ml without treatment versus 13.3 ng/ml
with treatment, p ¼ 0.0001, Fig. 2c). There was a statistical differ-
ence of vitamin D levels according the different combinations of
antiretroviral treatments (p < 0.0001, Fig. 2d). The multiple com-
parisons of these median vitamin D concentrations according to the
different combinations of treatments showed a significant differ-
ence between patients treated with 2 NRTI þ 1 NNRTI and patients
2 NRTI þ 1 PI (12.5 ng/ml versus 14.3 ng/ml respectively,
p ¼ 0.0001). There was a statistical difference between patients
with 2 NRTI þ 1 NNRTI and untreated patients (12.5 ng/ml versus
15.1 ng/ml respectively, p ¼ 0.0002). There was also a significant
difference between the second line treatment group (multiple,
complex and/or sequential combinations) and the untreated pa-
tient group (12.8 ng/ml versus 15.1 ng/ml respectively, p ¼ 0.0003).
4. Discussion
Vitamin D deficiency has already been described in the HIV-
positive population. However, the previous studies were often
conflicting.12e15 It was necessary to confirm or disprove the
deleterious role of ART in this setting. It was also necessary to
introduce in a single multivariable analysis all variables previously
described separately as identified risk factors for low vitamin D
concentration.
This retrospective study focusing on the prevalence of vitamin D
deficiency in HIV-infected patients was performed on a very large
cohort of patients (more than 2000 patients). Moreover, the pre-
sent work was taking into account the immune status, the viral load
and the use of antiretroviral therapy aiming to correlate the
severity of HIV disease and vitamin D deficiency.
This work confirmed that vitamin D deficiency (<30 ng/ml) is
very common in HIV-infected patients, with a prevalence of 89.2%.
The prevalence of vitamin D deficiency was higher than that re-
ported in previous studies.13,14 However, it is comparable to the
rates described in the EuroSIDA study.10 The present study
emphasized that the season in which the assay is performed
(winter) and ethnicity (Black African) are important factors for
predicting vitamin D deficiency. Concerning factors related to the
HIV infection, a homosexual mode of acquisition was independently
associated with a lower risk of vitamin D deficiency. This work
suggested that the modalities of antiviral treatment (the need of a
second line with multiple, complex combinations and/or sequential
treatments) as well as the duration of treatment are independently
associated with vitamin D deficiency of less than 30 ng/ml. Inter-
estingly, in addition to the winter season and Black African
ethnicity, a low CD4 count (CD4 <200 cells/ml), a higher CDC stage
and a ART including efavirenz were here identified as values inde-
pendently associated with severe vitamin D deficiency (<10 ng/ml).
It is currently accepted that antiretroviral treatment, especially
with NNRTIs, had a significant influence on the vitamin D con-
centration. Although there are conflicting results,12 previous
studies showed that patients treated with NNRTIs had lower
vitamin D levels than did patients treated with protease in-
hibitors.10e13 The hypothesis generally adopted is that NNRTIs
could interact with cytochrome P450 enzymes, some of which may
affect vitamin D metabolism.11 Induction of these enzymes may
reduce the amount of 25(OH)D. The present work confirms that,
giving a first line ART, the use of NNRTI (2 NRTI þ 1 NNRTI) leads to
a lower vitamin D level than patients receiving PI (2 NRTI þ 1 PI).
Prospective studies specifically demonstrated that initiation of
efavirenz was associated with significantly lower vitamin D con-
centrations in the subsequent year.15,16 This study included 324
patients with efavirenz and confirmed that this therapy was inde-
pendently associated with severe vitamin D deficiency.
Parameters of disease severity caused by HIV infection are closely
related to vitamin D deficiency. The EuroSIDA study identified an
association between vitamin D deficiency and increased risk of
morbidity and mortality.10 Other studies have also suggested an
association between vitamin D deficiency and low CD4 count.17 The
association between vitamin D deficiency and a higher severity of
HIV disease could be explained by two different mechanisms. First,
infectious complications as a result of poor immunity require intra-
hospital cares, which significantly reduces the duration of sun
exposure for patients and UVB is known to be the main source of
vitamin D. Hospitalization and infectious complications may lead to
reduced oral intake of the few foods that contain vitamin D and can
impede oral treatment with, for example, vitamin supplements.19,20
Second, vitamin D deficiency may also be a causal agent of the HIV
infection itself. Several studies provide evidence that vitamin D
deficiency is an increased risk factor to contract viral infections such
as influenza or HIV.23 A recent study showed an association between
vitamin D deficiency and a viral load of >50 copies/ml in HIV.18 A
direct antiviral effect of vitamin D on Hepatitis C virus production
was recently demonstrated in vitro suggesting that vitamin D could
have a role as natural antiviral mediator.24 In our study, vitamin D
 M. Theodorou et al. / Clinical Nutrition 33 (2014) 274e279
levels are lower when viral load is under 50 copies/ml. This has to be
related to the use and efficacy of ART that appears to be a strong
independent factor associated with vitamin D deficiency.
In the general population, vitamin D deficiency is a known risk
factor for many chronic diseases such as cancer, arterial hyperten-
sion and cardiovascular diseases.25e29 Our work confirmed that
HIV-infected patients have an increased risk of 25 (OH) D deficiency
compared with the general population. Because of the high prev-
alence of vitamin D deficiency in the present cohort, one should
suggest that all HIV-infected patients have to be systematically
screened for this condition and adequate vitamin D substitution
given if needed. It has been shown that oral supplementation of
vitamin D of 600e800 IU/day in patients with HIV with Afro-
American and Hispanic origin was not sufficient to reach vitamin D
levels> 30 ng/ml.17 A daily intake of vitamin D of at least 1000e
2000 IU/day may be needed to overcome the deficiency of vitamin
D in this population in particular. Large studies are necessary to
assess the impact of vitamin D supplementation on prevention of
co-morbidities among HIV-infected adults, such as osteoporosis,
osteopenia and cardiovascular disease.
In conclusion, the prevalence of vitamin D deficiency in the HIV-
infected population is high. Vitamin D deficiency (<30 ng/ml) in
HIV-infected patients is associated with known general factors
(female sex, winter season), but also specific factors related to HIV,
such as the duration of treatment and the presence of second line
treatments with multiple and complex combinations of ART. Severe
vitamin D deficiency (<10 ng/ml) is associated with a low CD4
count, a higher CDC stage and efavirenz therapy. Vitamin D con-
centration is significantly lower when ART is given especially with
NNRTIs, and when the CD4 count was less than 200/ml.
Source of funding
None.
Contributors and authorship
MVG and SD conceived and designed the study. MT obtained
data. TS did the data analysis and interpreted the data. MT and TS
drafted the report and MVG and SD critically revised the report.
All authors saw and approved the final version of this report.
This manuscript has not been published, nor is under consid-
eration, elsewhere.
Conflict of interest
The authors disclose no conflicts.
Acknowledgment
We thank Marc Delforge for his statistical help and knowledge.
References
1. Bischoff-Ferrari HA, Dietrich T, Orav EJ, Dawson-Hughes B. Positive association
between 25-hdroxy vitamin D levels and bone mineral density: a population-
based study of younger and older adults. Am J Med 2004;116:634e9.
2. Bischoff-Ferrari HA, Willett WC, Wong JB, Giovannucci E, Dietrich T, Dawson-
Hughes B. Fracture prevention with vitamin D supplementation: a meta-
analysis of randomized controlled trials. JAMA 2005;293:2257e64.
3. Etten Van, Mathieu C. Immunoregulation by 1,25-dihydroxyvitamin D3: basic
concepts. J Steroid Biochem. Mol Biol 2005;97:93e101.
279
4. Rosen CJ. Clinical practice. Vitamin D insufficiency. N Engl J Med 2011;364:
248e54.
5. Bischoff-Ferrari HA, Giovannucci E, Willett WC, Dietrich T, Dawson-Hughes B.
Estimation of optimal serum concentrations of 25-hydroxyvitamin D for mul-
tiple health outcomes. Am J Clin Nutr 2006;84:18e28.
6. Holick MF. Vitamin D deficiency. N Engln J Med 2007;357:266e81.
7. Heany RP. The vitamin D requirement in health and disease. J Steroid Biochem
Mol Biol 2005;97:13e9.
8. Steingrimsdottir L, Gunnarsson O, Indridason OS, Franzson L, Sigurdsson G.
Relationship between serum parathyroid hormone levels, vitamin D suffi-
ciency, and calcium intake. JAMA 2005;294:2336e41.
9. Villamor E. A potential role for vitamin D on HIV infection? Nutr Rev 2006;64:
226e33.
10. Viard JP, Souberbielle JC, Kirk O, Reekie J, Knysz B, Losso M, et al. Vitamin D and
clinical disease progression in HIV infection: results from the EuroSIDA study.
AIDS 2011;25(10):1305e15.
11. Mueller NJ, Fux CA, Ledergerber B, Elzi L, Schmid P, Dang T, Magenta L, et al.
High prevalence of severe vitamin D deficiency in combined antiretroviral
therapy-naïve and successfully treated Swiss HIV patients. AIDS 2010;24:
1127e34.
12. Crutchley RD, Gathe Jr J, Mayberry C, Trieu A, Abughosh S, Garey KW. Risk
factors for vitamin D deficiency in HIV-infected patients in the south central
United States. AIDS Res Hum Retroviruses 2012;28(5):454e9.
13. Van Den Bout-Van Den Beukel CJ, Fievez L, Michels M, Sweep FC, Hermus AR,
Bosch ME, et al. Vitamin D deficiency among HIV type 1-infected individuals in
the Netherlands: effects of antiretroviral therapy. Aids Res Hum Retrovirus
2008;24:1375e82.
14. Dao CN, Patel P, Overton ET, Rhame F, Pals SL, Johnson C, et al. Low vitamin
D among HIV-infected adults: prevalence of and risk factors for low
vitamin D Levels in a cohort of HIV-infected adults and comparison to
prevalence among adults in the US general population. Clin Infect Dis
2011;52:396e405.
15. Welz T, Childs K, Ibrahim F, Poulton M, Taylor CB, Moniz CF, et al. Efavirenz is
associated with severe vitamin D deficiency and increased alkaline phospha-
tase. AIDS 2010;24(12):1923e8.
16. Brown TT, McComsey GA. Association between initiation of antiretroviral
therapy with Efavirenz and decreases in 25-Hydroxyvitamin D. Antivir Ther
2010;15:425e9.
17. Stein EM, Yin MT, McMahon DJ, Shu A, Zhang CA, Ferris DC, et al. Vitamin
D deficiency in HIV-infected postmenopausal Hispanic and African-
American women. Osteoporos Int 2011 Feb;22(2):477e87.
18. Kim JH, Gandhi V, Psevdos Jr G, Espinoza F, Park J, Sharp V. Evaluation of
vitamin D levels among HIV-infected patients in New York city. AIDS Res Hum
Retroviruses 2012;28(3):235e41.
19. Mehta S, Giovannucci E, Mugusi FM, Spiegelman D, Aboud S, Hertzmark E, et al.
Vitamin D status of HIV-infected women and its association with HIV disease
progression, anémia, and mortality. PLoS One 2010;5(1):8770.
20. Mehta S, Mugusi FM, Spiegelman D, Villamor E, Finkelstein JL, Hertzmark E,
et al. Vitamin D status and its association with morbidity including wasting and
opportunistic illnesses in HIV-infected women in Tanzania. AIDS Patient Care
STDS 2011;25(10):579e85.
21. Knobel H, Guelar A, Vallecillo G, Nogués X, Díez A. Osteopenia in HIV-infected
patients. Is it the disease or is it the treatment. AIDS 2001;15(6):807e8.
22. Libois A, Clumeck N, Kabeya K, Gerard M, De Wit S, Poll B, et al. Risk factor of
osteopenia in HIV-infected women: no role of antiretroviral therapy. Maturitas
2010 Jan;65:51e4.
23. Haug CJ, Aukrust P, Haug E, Mørkrid L, Müller F, Frøland SS. Severe deficiency of
1.25 dihydroxyvitamin D3 in human immunodeficiency virus infection: asso-
ciation with immunological hyperactivity and only minor changes in calcium
homeostasis. J Clin Endocrin Metab 1998;83:3832e8.
24. Gal-Tanamy M, Bachmetov L, Ravid A, Koren R, Erman A, Tur R, et al. Vitamin
D: an innate antiviral agent suppressing Hepatitis C virus in human Hepato-
cytes. Hepatology 2011;54:1570e9.
25. Jenab M, Bueno-de-Mesquita HB, Ferrari P, van Duijnhoven FJ, Norat T,
Pischon T, et al. Association between pre-diagnostic circulating vitamin D
concentration and risk of colorectal cancer in European populations: a nested
case-control study. BMJ 2010;340:b5500.
26. Pfeifer M, Begerow B, Minne HW, Nachtigall D, Hansen C. Effects of a short-
term Vitamin D (3) and calcium supplementation on blood pressure and
parathyroid hormone levels in elderly women. J Clin Endocrinol Metab 2001;86:
1633e7.
27. Martins D, Wolf M, Pan D, Zadshir A, Tareen N, Thadhani R, et al. Prevalence of
cardiovascular risk factors and the serum levels of 25-Hydroxyvitamin D in the
United States. Arch Intern Med 2007;167:1159e65.
28. Wang TJ, Pencina MJ, Booth SL, Jacques PF, Ingelsson E, Lanier K, et al.
Vitamin D deficiency and risk of cardiovascular disease. Circulation
2008;117:503e11.
29. Wang TJ, Pencina MJ, Booth SL, Jacques PF, Ingelsson E, Lanier K, et al. Vitamin
D and risk of cognitive decline in elderly persons. Arch Intern Med 2010;170:
1135e41.