text
stringclasses 9
values | processed_text
stringclasses 9
values |
|---|---|
Page 1
Endocrine Reviews, 2024, 45, 625–654
https://doi.org/10.1210/endrev/bnae009
Advance access publication 27 April 2024
Review
ENDOCRINE
SOCIETY
OXFORD
6,7
8
9
10
Consensus Statement on Vitamin D Status Assessment
and Supplementation: Whys, Whens, and Hows
Andrea Giustina,¹® John P. Bilezikian, 2 Robert A. Adler,³ Giuseppe Banfi,4,5 D
Daniel D. Bikle, ℗ Neil C. Binkley, Jens Bollerslev, Roger Bouillon,
Maria Luisa Brandi, ①℗ Felipe F. Casanueva, 12 Luigi di Filippo, Lorenzo M. Donini,
Peter R. Ebeling, 14 Ghada El-Hajj Fuleihan, 15 Angelo Fassio, 16 Stefano Frara, 1 Glenville Jones, 17
Claudio Marcocci, 18 Adrian R. Martineau, 19 Salvatore Minisola, 20℗ Nicola Napoli,
Massimo Procopio, 22 René Rizzoli, 23 Anne L. Schafer, Christopher T. Sempos,2
Fabio Massimo Ulivieri, and Jyrki K. Virtanen 25D
11
6
1
24
21
13
ID
Institute of Endocrine and Metabolic Sciences, San Raffaele Vita-Salute University and IRCCS Hospital, Milan 20132, Italy
2Department of Medicine, Vagelos College of Physicians and Surgeons, New York, NY 10032, USA
³Richmond Veterans Affairs Medical Center and Virginia Commonwealth University, Richmond, VA 23284, USA
AIRCCS Galeazzi Sant'Ambrogio Hospital, Milano 20161, Italy
5 San Raffaele Vita-Salute University, Milan 20132, Italy
17
6 Department of Medicine, University of California and San Francisco Veterans Affairs Health Center, San Francisco, CA 94121-1545, USA
7Department of Endocrinology, University of California and San Francisco Veterans Affairs Health Center, San Francisco, CA 94121-1545, USA
8 School of Medicine and Public Health, University of Wisconsin-Madison, Madison, WI 53726, USA
9Faculty of Medicine, University of Oslo, Oslo 0313, Norway
10Laboratory of Clinical and Experimental Endocrinology, Department of Chronic Diseases, Metabolism and Ageing, KU Leuven, 3000 Leuven,
Belgium
11 Italian Foundation for the Research on Bone Diseases (F.I.R.M.O.), Florence 50129, Italy
12 Department of Medicine, Instituto de Investigación Sanitaria (IDIS), Complejo Hospitalario Universitario and CIBER de Fisiopatologia de la
Obesidad y Nutricion (CIBERobn), Santiago de Compostela University, Santiago de Compostela 15706, Spain
13 Department of Experimental Medicine, Sapienza University, Rome 00161, Italy
14 Department of Medicine, School of Clinical Sciences, Monash University, Clayton 3168, Australia
15 Calcium Metabolism and Osteoporosis Program, WHO CC for Metabolic Bone Disorders, Division of Endocrinology, American University of
Beirut, Beirut 1107 2020, Lebanon
16 Rheumatology Unit, University of Verona, Verona 37129, Italy
17 Department of Biomedical and Molecular Sciences, Queen's University, Kingston, Ontario, ON K7L 3N6, Canada
18 Department of Clinical and Experimental Medicine, University of Pisa, Pisa 56126, Italy
19 Faculty of Medicine and Dentistry, Queen Mary University of London, London E1 4NS, UK
20 Department of Clinical, Internal, Anesthesiologic and Cardiovascular Sciences, Sapienza University of Rome, Rome 00161, Italy
21 Unit of Endocrinology and Diabetes Campus Bio-Medico, University of Rome, Rome 00128, Italy
22 Division of Endocrinology, Diabetology and Metabolic Diseases, "Molinette" Hospital, University of Turin, Turin 10126, Italy
23 Geneva University Hospitals and Faculty of Medicine, Geneva 1205, Switzerland
24 Vitamin D Standardization Program (VDSP), Havre de Grace, MD 21078, USA
25 Institute of Public Health and Clinical Nutrition, University of Eastern Finland, Kuopio FI-70211, Finland
Correspondence: John P. Bilezikian, MD, Department of Medicine, PH 8E105G, Vagelos College of Physicians and Surgeons, 630 W 168th St, New York, NY
10032, USA. Email: Jpb2@cumc.columbia.edu.
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Abstract
The 6th International Conference, "Controversies in Vitamin D," was convened to discuss controversial topics, such as vitamin D metabolism,
assessment, actions, and supplementation. Novel insights into vitamin D mechanisms of action suggest links with conditions that do not depend
only on reduced solar exposure or diet intake and that can be detected with distinctive noncanonical vitamin D metabolites. Optimal 25-
hydroxyvitamin D (25(OH)D) levels remain debated. Varying recommendations from different societies arise from evaluating different clinical
or public health approaches. The lack of assay standardization also poses challenges in interpreting data from available studies, hindering
rational data pooling and meta-analyses. Beyond the well-known skeletal features, interest in vitamin D's extraskeletal effects has led to
clinical trials on cancer, cardiovascular risk, respiratory effects, autoimmune diseases, diabetes, and mortality. The initial negative results are
likely due to enrollment of vitamin D-replete individuals. Subsequent post hoc analyses have suggested, nevertheless, potential benefits in
Received: 1 September 2023. Editorial Decision: 7 March 2024. Corrected and Typeset: 27 April 2024
© The Author(s) 2024. Published by Oxford University Press on behalf of the Endocrine Society.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
Page 2
626
Endocrine Reviews, 2024, Vol. 45, No. 5
reducing cancer incidence, autoimmune diseases, cardiovascular events, and diabetes. Oral administration of vitamin D is the preferred route.
Parenteral administration is reserved for specific clinical situations. Cholecalciferol is favored due to safety and minimal monitoring requirements.
Calcifediol may be used in certain conditions, while calcitriol should be limited to specific disorders in which the active metabolite is not readily
produced in vivo. Further studies are needed to investigate vitamin D effects in relation to the different recommended 25(OH)D levels and the
efficacy of the different supplementary formulations in achieving biochemical and clinical outcomes within the multifaced skeletal and
extraskeletal potential effects of vitamin D.
Graphical Abstract
Non-canonical vitamin D
metabolites assessment
MM
25(OH)D assessment
Impaired metabolism and
genetic polymorphisms
Rx
Vitamin D
status
Vitamin D
deficiency
Reduced synthesis
and dietary intake
Use of vitamin D forms/
metabolites alternative to cholecalciferol
Skeletal
effects
Extra-skeletal effects
Reduced intestinal
absorption
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Cholecalciferol:
oral route
Cholecalciferol:
parenteral route
2024 Endocrine Society
Key Words: vitamin D, cholecalciferol, calcitriol, calcifediol, vitamin D assay, Vitamin D Standardization Program (VDSP)
Abbreviations: 1,25(OH)2D, 1,25 dihydroxyvitamin D; 7-DHC, 7-dehydrocholesterol; 25(OH)D, 25-hydroxyvitamin D; BMD, bone mineral density; BMI, body mass
index; CDC, Centers for Disease Control and Prevention; CKD, chronic kidney disease; CVD, cardiovascular disease; DBP, vitamin D binding protein; ES,
Endocrine Society; HAT, histone acetyltransferase activity; HDAC, histone deacetylase activity; ICU, intensive care unit; IL, interleukin; IOM, Institute of
Medicine; LC-MS, liquid chromatography-mass spectrometry; miRNA, microRNA; MR, mendelian randomization; OR, odds ratio; PTH, parathyroid hormone;
RCT, randomized controlled trial; SRC, steroid hormone receptor coactivator; T2D, type 2 diabetes; TSS, transcription start site; UVB, sunlight; VDR, vitamin
D receptor; VDSP, Vitamin D Standardization Program; vitamin D2, ergocalciferol; vitamin D3, cholecalciferol; VDT, vitamin D toxicity.
Page 3
Endocrine Reviews, 2024, Vol. 45, No. 5
627
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
ESSENTIAL POINTS
• Total serum 25-hydroxyvitamin D concentration is
the accepted biomarker of vitamin D status, but assay
methodology and standardization as well as desirable
levels, which may vary according to the underlying
condition, are still major issues
• Advances in knowledge about vitamin D have in-
cluded its metabolism, identification of noncanonical
metabolites, mechanisms of action, and genetic poly-
morphisms. These insights have added to our under-
standing of vitamin D's role in nutrition and in
disease
• Vitamin D deficiency reduces intestinal calcium ab-
sorption leading to secondary hyperparathyroidism,
bone loss, and increased risk of fractures in older
adults. Meta-analyses of clinical trials show that vita-
min D and calcium, together, decrease hip and other
fractures in nursing home residents
• Post hoc analyses of recent mega trials on extraskele-
tal effects of vitamin D suggest a link between vita-
min D status and immune system and development
of type 2 diabetes mellitus. Cardiovascular events
and mortality may be positively affected as well
• Daily vitamin D regimens seem to be the most effi-
cient and beneficial strategy to improve vitamin D
status but dosing schedules with longer intervals up
to 4 weeks have been proposed to overcome low
compliance with daily schedules
• Oral cholecalciferol (vitamin D3) remains the pre-
ferred form of vitamin D for supplementation, while
other vitamin D analogues (eg, calcifediol, calcitriol,
alfacalcidol) and parenteral administration should be
used in specific conditions
The 6th International Conference "Controversies in Vitamin
D" was held in Florence, Italy, September 21 to 24, 2022, as
part of this series that started in 2017 (1-10). The objective
of this conference, featuring international experts, was to re-
view and discuss controversial topics regarding vitamin
D. Before the event, participants reviewed the available litera-
ture on their assigned topic and presented their findings at the
conference. After each presentation, open sessions enabled full
discussion. On the last day of the conference, all participants
completed their discussion and agreed on a menu for addition-
al research. The 2 main topics addressed were recommenda-
tions on assessing vitamin D deficiency and vitamin D
supplementation. This paper summarizes the findings on the
"whys, whens, and hows" of these two topics.
Vitamin D Metabolism and Mechanism of
Action
Metabolism
Vitamin D3 is produced in the skin from 7-dehydrocholesterol
(7-DHC), while both vitamin D2 (ergocalciferol) and vitamin
D3 (cholecalciferol) can be present in the diet. Vitamin D2 and
D3 are hydroxylated first in the liver (and other tissues) to
25-hydroxyvitamin D (25(OH)D) and then in the kidney
Ꭰ
(and other tissues) to 1,25 dihydroxyvitamin
(1,25(OH)2D). Both 25(OH)D and 1,25(OH)2D are subse-
quently metabolized to their 24 (and for D3 23) hydroxy forms
24,25(OH)2D2/3, 23,25(OH)2D3, and 1,24,25(OH)3D2/3
(or 1,23,25(OH)3D3). Like other steroid hormones, vitamin D
is highly lipophilic and bound to protein carriers that
help maintain stable serum levels. The half-life of serum
25(OH)D is 2 to 3 weeks, and that of the more water-soluble
1,25(OH)2D is approximately 5 to 8 hours. The majority of
circulating 25(OH)D, including its metabolites, are bound
tightly by vitamin D binding protein (DBP) and more loosely
bound by albumin (4).
7-Dehydrocholesterol reductase
Although the production of vitamin D from 7-DHC under the
influence of sunlight (UVB) is a nonenzymatic step, the pro-
duction of 7-DHC is not. Its synthesis in the skin is a step in
the Kandutsch-Russell pathway. DHCR7 converts 7-DHC
to cholesterol, so its activity dictates how much 7-DHC is
available for vitamin D production. Inactivating mutations
of DHCR7 result in Smith-Lemli-Opitz syndrome, a develop-
mental disorder (11). These patients suffer primarily from the
consequences of too little cholesterol, steroids, or bile acids,
but they appear to be more sensitive to UVB light and may pre-
sent with higher serum 25(OH)D concentrations than normal
individuals. The regulation of DHCR7 is incompletely
understood. Cholesterol and vitamin D (but not 1,25(OH)2D)
increase proteasomal degradation of DHCR7, leading to
increased vitamin D production. AMPK (adenosine mono-
phosphate-activated protein kinase C), a key sensor and regu-
lator of cellular energy homeostasis and protein kinase A are
potent inhibitors of DHCR7 (12).
25-Hydroxylases
The liver is the major source of 25(OH)D production from
vitamin D. However, numerous enzymes within both mito-
chondria and microsomes have 25-hydroxylase activity.
Initial studies suggested that CYP27A1, a mitochondrial en-
zyme with substantial homology to CYP27B1 and
CYP24A1 (the 1a and 24-hydroxylases, respectively), was
the major 25-hydroxylase. However, patients with inactivat-
ing mutations in this enzyme develop cerebrotendinous xan-
thomatosis with abnormal bile and cholesterol metabolism
but not rickets (13). Current data support CYP2R1 as the ma-
jor 25-hydroxylase, at least in the liver (and testes), where it
resides in the microsomal compartment (13). When deleted
in mice, serum 25(OH)D levels fall by over 50%, but not
more. There is little effect on serum calcium and phosphate
levels, suggesting that other enzymes with 25-hydroxylase ac-
tivity compensate. Five functional mutations in CYP2R1 have
been described so far. Although these mutations result in little
or no 25-hydroxylase activity in vitro, individuals maintain
normal or even high 1,25(OH)2D levels and, in some cases, re-
spond both to vitamin D and 1a(OH)D with further increases
in 1,25(OH)2D. As children, these individuals develop classic-
al nutritional rickets responding to high doses of vitamin D or
small doses of 25(OH)D; as adults, they tend to lose their need
for vitamin D supplementation (14). Such data suggest that, as
in the mouse, CYP2R1 could not be the only enzyme with
25-hydroxylase activity (14).
Page 4
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
628
Previously, it had been thought that the 25-hydroxylation
of vitamin D was primarily substrate dependent. However, re-
cent evidence indicates that this is not the case. Roizen et al
(15) found that the serum concentration of 25(OH)D, but
not vitamin D, was decreased in mice fed a high-fat diet to in-
duce obesity associated with decreased expression of CYP2R1
in the liver. Aatsinki et al (16) found that a high-fat diet that
induced obesity and type 2 diabetes (T2D), as well as
streptozotocin-induced type 1 diabetes, both decreased the
hepatic messenger RNA and protein concentration of
CYP2R1. Thus, the concept that the low levels of 25(OH)D
in obesity and the limited response to vitamin D supplementa-
tion in these individuals are somehow related to increased
storage of vitamin D in fat is still controversial (17) and needs
further investigation.
CYP27B1―the 25-hydroxyvitamin D-1α-hydroxylase
Unlike the 25-hydroxylases, there is only a single 25(OH)
D-1α-hydroxylase, CYP27B1. This enzyme is found in the
mitochondrion along with CYP24A1. The kidney is the
main source of circulating 1,25(OH)2D, but many tissues, in-
cluding the epidermis and other epithelial tissues, bone, pla-
centa, and immune system cells, also express CYP27B1. The
product, 1,25(OH)2D, likely has paracrine or autocrine ac-
tions (18). Regulation of CYP27B1 in these extracellular sites
differs from that in the kidney. In the kidney, CYP27B1 is
regulated primarily by parathyroid hormone (PTH) and
insulin-like growth factor-1, which stimulate it, as well as by
fibroblast growth factor 23 (FGF23) and 1,25(OH)2D itself,
which inhibit it. In nonrenal tissues cells, such as keratinocytes
and macrophages, cytokines, such as, interferon-gamma
(IFN-y), tumor growth factor alpha (TNFα), and transforming
growth factor beta1 (TGFẞ1) are the major inducers of
CYP27B1. In peripheral blood mononuclear cells, interleukin
(IL)-1, IL-2, and IL-15 also stimulate CYP27B1 activity,
whereas IL-4 is suppressive (19-21). Thus, the induction of
CYP27B1 in these extrarenal tissues is by cytokines, and the
failure of CYP27B1 in these tissues to respond to the increased
circulating levels of 1,25(OH)2D and calcium account for the
hypercalcemia often found in granulomatous diseases, such as
sarcoidosis and lymphomas (22). Mutations in CYP27B1
cause a disease known as pseudovitamin D-deficiency rickets
or type 1A vitamin D-dependent rickets (23); both the renal
and extrarenal CYP27B1 have the same sequence, but their
differences in regulation occur because of differences in
tissue-specific multicomponent control modules within the
regulatory regions of the gene.
CYP24A1 and CYP3A-the 25-hydroxyvitamin D-24(23)
hydroxylases
These are the catabolic enzymes of vitamin D metabolism, with
both 25(OH)D and 1,25(OH)2D as their substrates (24-26).
CYP24A1 is the dominant 24-hydroxylase in most tissues,
but CYP3A4 likely plays a role in the liver and intestine, where
it is highly expressed. Both enzymes have 24-hydroxylase and
23-hydroxylase activity, although the relative proportions of
24-hydroxylase and 23-hydroxylase activity for CYP24A1
are species specific. Both enzymes are induced by
1,25(OH)2D and CYP24A1 is induced by 25(OH)D as
well (27) and the induction of CYP3A4 seems to be at least
as great as that for CYP24A1 in the intestine. To label
CYP24A1 as a purely catabolic enzyme in vitamin D
Endocrine Reviews, 2024, Vol. 45, No. 5
metabolism would appear to be a misnomer. 1,24,25(OH)3D
has a substantial affinity for the vitamin D receptor
(VDR), with approximately 10% of 1,25(OH)2D biological
activity. Moreover, a specific G protein-coupled membrane
receptor for 24,25(OH)2D, Fam57B2, has been identified
in bone and other tissues such as the skin, and through
this receptor, 24,25(OH)2D was found to be involved in
fracture repair (28). CYP24A1 is under the control of
1,25(OH)2D and FGF23 (both stimulatory) and calcium
(29). 5a-Dihydrotestosterone, via the progesterone receptor,
has also been reported to stimulate CYP24A1 (30). In humans,
inactivating mutations in CYP24A1 are now recognized as a
major cause of idiopathic infantile hypercalcemia, a syndrome
marked by severe hypercalcemia, hypercalciuria, and nephro-
calcinosis, decreased PTH, low 24,25(OH)2D, and inappro-
priately normal to high 1,25(OH)2D. Although initially
identified in children (31), more recent case reports indicate
that the diagnosis may not be made until adulthood, generally
following a condition of increased 1,25(OH)2D production
like pregnancy (32, 33). Such adults generally present with
early-onset nephrolithiasis and/or nephrocalcinosis.
Importantly, CYP3A4 mutations or drug-induced excess
CYP3A4 activity have recently been linked to vitamin D defi-
ciency and vitamin D-dependent rickets type 3, with affected
individuals demonstrating greatly accelerated inactivation of
vitamin D metabolites. This represents a novel mechanism
for vitamin D deficiency (34).
Mechanism of Action
The VDR is critical for most of the actions of vitamin D, with
1,25(OH)2D as its major ligand. VDR is a transcription factor
found in nearly all cells. Not surprisingly, vitamin D affects
many cellular processes via the VDR, with one of the most im-
portant being the regulation of intestinal calcium absorption
(Fig. 1) (4). In a recent ontology analysis (35), 11 031 putative
VDR target genes were identified, of which 43% were in-
volved with metabolism, 19% with cell and tissue morph-
ology, 10% with cell junction and adhesion, 10% with
differentiation and development, 9% with angiogenesis, and
5% with epithelial to mesenchymal transition. Furthermore,
VDR can regulate various microRNAs (miRNAs) and long
noncoding RNAs involving the expression of numerous pro-
teins directly or indirectly. As a result of the appreciation
that the VDR is so widespread along with the key vitamin D
metabolizing enzymes such as CYP27B1 and CYP24A1, inter-
est in understanding the role of vitamin D and the VDR in
nonclassic as well as classic target tissues regulating calcium
and phosphate homeostasis has been substantial. Although
most of the actions of VDR involve its role as a transcription
factor within the nucleus, the VDR has also been shown to
have nongenomic actions via its location in the plasma mem-
brane and perhaps even in mitochondria (4).
Regulation
The regulation of VDR expression is cell specific. For ex-
ample, 1,25(OH)2D regulates VDR expression in bone cells
but not in the intestine. Many factors in addition to
1,25(OH)2D regulate VDR expression, including growth fac-
tors, insulin, as well as PTH, glucocorticoids, estrogen, and
retinoic acid, in some cases acting via a variety of transcription
factors, such as AP-1, SP1, C/EBP, and CDX2, C/EBPẞ,
Runx2, cyclic adenosine monophosphate response element
Page 5
Endocrine Reviews, 2024, Vol. 45, No. 5
629
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Calcium
channel
Ca2+
Ca2+
Ca2+
Calbindin
Ca2+
VDR
Claudins
Ca2+
Vitamin D
Figure 1. Three-step mechanism of intestinal calcium absorption by
vitamin D. An important function of vitamin D is stimulating intestinal
calcium absorption by increasing the expression of calcium-permeable
claudins, apical membrane calcium channels, and calcium-binding
protein calbindins. The extrusion of calcium is across the basolateral
membrane. This process is especially enhanced when dietary calcium
intake is low.
binding protein (CREBP), retinoic acid receptor (RAR), and
glucocorticoid receptor (GR). Similarly, calcium upregulates
VDR expression in the parathyroid gland, presumably
through its calcium-sensing receptor. On the other hand,
SNAIL 1 and 2 (SLUG) downregulate VDR expression in sev-
eral cancer cell lines. MicroRNAs can regulate VDR levels, as
exemplified by the binding of miR-125b, miR-298, and
miR-27b to the 3'untranslated region to decrease VDR levels
(4,36).
Genomic actions
Carlberg (36) reported that the human genome contains more
than 23 000 VDR binding sites, most of which are cell specific.
Their locations varied with the duration of ligand exposure,
and only some were readily identified with a specific gene.
The VDR binding sites can be thousands of bases away
from the transcription start site (TSS) of the genes they regu-
late, and genes generally have multiple VDR binding sites,
the activity of which may vary in different cells and species.
An informative example of how this might work in different
cells is the regulation of the RANKL gene (Tnfsf11). This
gene is regulated by PTH and 1,25(OH)2D in osteoblasts
and by AP-1 factors, such as c-fos, in activated T cells (13).
The Pike laboratory identified 7 VDR binding sites in
RANKL up to 88 kb upstream of the TSS, of which the
-75-kb site proved most active in the mouse gene (37, 38),
whereas the proximal site was most active in the human
gene (39). However, in activated T cells, 3 additional sites
even further upstream of the TSS have been identified as sites
of RANKL induction by c-fos (13).
A similar example can be found for Cyp27b1. This gene is
negatively regulated by its product in the kidney but not in
other tissues (40). The VDR binding sites are generally situ-
ated in a region with other transcription factors that may
share regulation of that gene, potentially providing cell-
specific gene regulation. For example, the VDR binding region
of the RANKL gene contains several CREB sites responsible
for the PTH regulation of this gene (41).
Coregulators and epigenetic changes regulating vitamin D
receptor function
The sites of active transcription are marked by epigenetic
changes both in the gene itself and the histones that regulate
access of the transcriptional machinery to the gene. In humans
and mice, 1,25(OH)2D regulates these epigenetic changes by
affecting the binding of coregulators to the VDR, whether as
coactivators with histone acetyltransferase activity (HAT) or
as cosuppressors with histone deacetylase activity (HDAC).
More than 250 published coregulators interact with nuclear
hormone receptors. The best-studied coactivators with respect
to the VDR are the steroid hormone receptor coactivators
(SRC 1-3) and the Mediator complex. SRCs recruit HATs to
the VDR. The Mediator complex does not contain HAT activ-
ity but binds directly to RNA polymerase II to help form the
preinitiation complex along with basal transcription factors
such as TFIIB and several TAT-binding proteins. These coac-
tivators all bind to the AF2 domain of the VDR. On the other
hand, corepressors, such as SMRT and NCoR complexes,
have HDAC activity and bind to H3 to H5 in the absence of
a ligand. In the presence of 1,25(OH)2D and the conform-
ational change with H12, these corepressors are displaced, en-
abling the coactivators to bind to their sites on H12.
Hairless is a corepressor of VDR expressed primarily in the
brain and skin. It binds to the central region of the ligand-
binding domain of VDR, as does NCOR/SMRT. The role of
hairless is complex in that it represses ligand-dependent
VDR functions with respect to epidermal differentiation
(42) but is required for ligand-independent VDR regulation
of hair follicle cycling (43). In mice, VDR gene ablation elicits
both rickets and hair loss, while point mutations specifically
compromising either 1,25(OH)2D ligand or coactivator con-
tacts in human VDR result in rickets without hair cycle disrup-
tion. On the other hand, loss-of-function mutations in human
VDR results in disrupted VDR-DNA binding or VDR-RXR
heterodimerization; this impaired corepressor activity on
VDR-mediated transactivation, in part due to the attenuated
interaction of hairless with HDACs, can result in clinical con-
ditions, such as the rare autosomal recessive disease atrichia
with papular lesions or alopecia universalis congenita (42-
44). VDR interaction with its heterodimeric partner RXR is
probably pivotal to hair cycling, as the conditional inactiva-
tion of RXRa in mouse skin results in alopecia resembling
that in VDR-null mice. Similar to mutations in the
VDR-encoding gene, mutations in the mammalian hairless
gene result in congenital hair loss both in mice and humans.
Remarkably, the hair loss phenotype caused by the mutated
human VDR gene resembles the generalized atrichia caused
by mutations in the hairless gene (44).
In summary, new insights into the regulation of vitamin D-
related enzymes and the differential mechanism of action of
Page 6
630
Endocrine Reviews, 2024, Vol. 45, No. 5
Conditions in which vitamin D
status should be evaluated
due to reduced synthesis:
•
• Age >50 years
Low sun exposure
• Sunscreen
•
⋅
Air pollution
Phototype (dark skin)
Major source
(=80%)
SUNLIGHT
UVB
Minor source (20%)
DIETARY intake
Ergocaliferol: plants/supplements
Cholecalciferol: fish (cod liver oil),
meat, fortified milk, egg yolk, butter
Conditions in which vitamin D
status should be evaluated
due to reduced intake:
• Not fortified food
• Restrictive diet
• Gastrointestional malabsorption:
Bariatric surgery, celiac disease,
inflammatory bowel diseases
.
Season and latitude
Skin
(Previtamin D)
7-dehydrocholesterol
Liver
Cholecalciferol→→➤
25-hydroxylase
Clinical conditions in which vitamin D
status should be evaluated:
Low bioavailability: overweight and obese patients
• Increased consumption: patients treated with
interfering drugs
Low 25(OH)D synthesis: Liver dysfunction
25-hydroxyvitamin D
Kidney
25(OH)D renal loss or low 1,25(OH)2D synthesis: Parathyroid
kidney and parathyroid diseases
hormone
.
Small
intestine
Most used
biomarker
for Vitamin D
status
Osteometabolic conditions in which
vitamin D status should be evaluated:
• Parathyroid, kidney diseases and
intestinal disorders
Osteoporosis, osteopenia, skeletal
disorders
→ Active form
1,25-dihydroxyvitamin D
- Increases Ca absorption (small intestine)
- Increases urinary Ca and Phos reabsorption
Increases bone mineralization secondary
to the mineral absorptive and resorptive
functions
1-αhydroxylase
Useful biomarker
in kidney and
parathyroid diseases
Figure 2. Overview of vitamin D metabolism. The figure shows metabolism of vitamin D in physiologic and deficient status, with specific reference to
conditions in which vitamin D should be evaluated.
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
VDR have demonstrated important links between metabolic
disorders and vitamin D metabolism. A better understanding
of how the VDR interacts with other transcription factors in a
cell-specific fashion will provide a greater understanding of
how the same molecule can have such different actions in
many physiologic processes. In turn, more insights may lead
to more nuanced and/or specific uses of vitamin D and its me-
tabolites in clinical situations, as discussed next.
Assessment of Vitamin D Status
To date, total serum 25(OH)D, the sum of 25(OH)D3 and
25(OH)D2, is the accepted biomarker of vitamin D status
(Fig. 2). Observational studies have indicated the beneficial ef-
fects of an optimal vitamin D status on various outcomes not
directly associated with the classical target tissues for the
hormone the so-called pleiotropic effects (45). Based on
these studies, mostly using traditional radioimmunoassay
measurements, vitamin D guidelines issued by major organi-
zations worldwide recommend optimal 25(OH)D levels to
be in the range of 50 to 75 nmol/L (20-30 ng/mL) (46, 47).
However, optimal levels are still debated for several reasons
(48-50). Lack of assay standardization contributes to the
problem, and initiatives should be implemented to overcome
it (50, 51). In this perspective, the Endocrine Society (ES)
has asked a task force to review its 2011 guidelines.
Differences in the suggested optimal serum 25(OH)D levels
depend on several factors. It is essential to clarify what is
meant by optimal 25(OH)D level, that is, for whom and for
what, as it is essential to consider patients' clinical profiles
and the outcomes of interest. Many studies have been per-
formed with a focus on osteoporosis and bone metabolism.
Recently, several randomized controlled trials (RCTs) as-
sessed potential pleiotropic effects of 25(OH)D, in general,
with negative results (45, 52, 53). Another pivotal factor in de-
ciding optimal 25(OH)D levels is the perspective used. For ex-
ample, although the outcomes used to derive desirable
25(OH)D levels were similar for the Institute of Medicine
(IOM) (47) and the 2011 ES guidelines (46), most studies in-
cluded conclusions that differed. For example, in the case of
osteomalacia, although the same study was used, conclusions
differed. This is because the ES selected a cutoff above which
Page 7
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Endocrine Reviews, 2024, Vol. 45, No. 5
no individual had osteomalacia (clinical perspective), whereas
IOM chose a cutoff where 97.5% of the cohort did not have
osteomalacia (public health perspective). Finally, when dis-
cussing vitamin D status assessment, it is also important to dif-
ferentiate between screening, that is, a public health approach
undertaken in the general populations, and testing, that is, tar-
geted testing of high-risk individuals in the clinical setting.
Theoretically, obtaining a 25(OH)D level within an optimal
window for the general population will necessarily result in
overscreening and overtreatment of healthy individuals
(48, 49). While general screening for 25(OH)D deficiency/
insufficiency is not recommended, measurements could be
performed in patients with several risk factors for severe defi-
ciency or who are being evaluated for metabolic bone disease
(46, 53-56). This recommendation may help to mitigate the
dramatic increase in the number of 25(OH)D measurements
and the associated economic burden (45, 52, 54, 56).
Supporting this view, initiatives have been undertaken to re-
duce unnecessary 25(OH)D analyses in Australia (57, 58).
The change in recommended testing criteria halved the num-
ber of measurements but paradoxically increased the number
of unnecessary tests and decreased tests of patients at high
risk of deficiency, with only a small improvement in the detec-
tion of deficiency (56).
Screening and Testing for Vitamin D Status
Screening in the general population-public health approach
Levels of 25(OH)D in the general population vary consider-
ably depending on several factors, including the season, lati-
tude, cultural factors leading to reduced UVB light
exposure, skin pigmentation, body mass index (BMI), sex,
age, level of physical exercise, and food fortification with vita-
min D or use of vitamin D supplements, even among otherwise
comparable Western societies; moreover, genetic factors such
as gene polymorphisms may have major effects on serum
25(OH)D according to twin studies and mendelian random-
ization (MR) reports (4, 48).
In considering when to test for vitamin D deficiency, it is
well recognized that serum 25(OH)D levels vary by season.
This is not surprising, given that most vitamin D is generated
in the skin following UVB exposure with little vitamin D avail-
able from the average unfortified diet (59, 60). Given an ob-
served drop in 25(OH)D levels between seasons, a higher
target (~75 nmol/L) may be required at the end of summer
to allow for the anticipated 10 to 25 nmol/L drop during the
winter months (61). For example, in a sunny country, such
as Australia, the prevalence of vitamin D deficiency
(<50 nmol/L) is as high as 36% during winter and as low as
14% in summer (62). In Lebanon, another sunny country,
mean serum 25(OH)D levels were 12 to 15 nmol/L higher in
summer to fall compared to winter (63). In this regard, meas-
urement of serum 25(OH)D levels at the end of winter or in
early spring would increase the detection of low 25(OH)D lev-
els in the general population (61).
Regardless, in populations with a low prevalence of vitamin
D deficiency, screening of the general population is not cost-
effective, and the decision to assess an individual needs to be
made using a risk stratification approach for having vitamin
D deficiency. Prioritized screening for high-risk groups could
be useful, given the potentially adverse effects of vitamin D de-
ficiency on skeletal and overall health, particularly when the
serum 25(OH)D levels are less than 30 nmol/L (<12 ng/mL).
631
Boosted regression tree models have also been developed
from RCT data to predict the serum 25(OH)D concentration
(64). Several predictor variables of a deseasonalized serum
25(OH)D concentration less than 50 nmol/L have been iden-
tified from training and validation data sets in the D-Health
trial. The 2 strongest predictors were ambient UV radiation
and total intake of vitamin D. Other important predictors of
mild vitamin D deficiency were time spent outdoors, alcohol
consumption, BMI, quality of life measures, and physical ac-
tivity. Thus, a lack of ambient UVB radiation and lack of vita-
min D fortification of food or use of vitamin D supplements
will probably result in a poor vitamin D status, particularly
in individuals in whom other risk factors are also present.
In conclusion, screening for optimal vitamin D status in the
general population should be avoided as it is not informative
and has a considerable economic burden. Nevertheless, sev-
eral characteristics and pathological conditions in the general
population could place individuals at risk for severe deficits.
These populations, which should be recognized, are consid-
ered in the next section.
Testing populations at risk of vitamin D deficiency—clinical
approach
Measurement of 25(OH)D has been recommended in patients
at risk for deficiency (46, 47) (Table 1 and Fig. 2). Thus,
25(OH)D is widely measured in many of these high-risk
groups, for example, in older adults with decreased endogen-
ous vitamin D production and prone to develop osteoporosis,
in patients with parathyroid disorders and liver disease, and in
patients with obesity (46, 47, 54-56, 65-71). Patients with
class III obesity (BMI >40) present with low levels of
25(OH)D for various reasons, including nutritional factors,
psychological reasons leading to less sun exposure, decreased
hepatic expression of CYP2R1, and sequestration of the vita-
min in the excess adipose tissue (15, 72). Class III obesity may
be addressed by bariatric surgery, which, by itself, may lead to
malabsorption and, thereby, a further decrease of 25(OH)D
levels, potentially followed by a secondary hyperparathyroid-
ism (73). Other at-risk groups include those who are house-
bound, those working long hours indoors, dark-skinned
individuals, patients with a chronic disease, those taking med-
ications increasing vitamin D catabolism, etc (see Table 1).
Paradoxically, listing situations where it may be reasonable
to measure 25(OH)D accounts for most people. This would
again result in overtesting with high costs for the health care
system. Rather than testing in situations where it would be
reasonable to, it would be better to test only in situations
that actually warrant it. It comes down to the providers' judg-
ment in first recognizing these high-risk individuals and then
deciding to confirm with a measurement of 25(OH)D. There
is, in fact, little evidence for the scientific utility and cost-
effectiveness of testing for 25(OH)D deficiency, even in
some of these selected groups (53). For example, some guide-
lines have recommended against screening pregnant women
for vitamin D deficiency because of uncertainty about the ben-
efits of vitamin D supplementation for maternal and fetal out-
comes (47, 74). However, a case can be made for optimizing
vitamin D status in all pregnant or breastfeeding women
and their offspring, given the reemerging public health con-
cern of rickets in high-risk children (75) and potential benefits
on future peak bone mass (76). Evidence of a relationship be-
tween low 25(OH)D and adverse maternal outcomes together
Page 8
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
632
•
•
Table 1. Populations at risk of vitamin D deficiency according to a
clinical approach
•
Older people
Housebound people
Disabled people
Institutionalized people
People working long hours indoors
•
•
•
•
Office workers
Factory or warehouse workers
Taxi drivers
Night-shift workers
People with dark skin
Low levels of physical activity
People with a debilitating/chronic disease
• Diabetes
Chronic kidney disease
Gastrointestinal malabsorptive syndromes
Parathyroid disorders
•
Liver diseases
Obesity in particular those with highest levels of waist circumference
Patients after bariatric surgery
People taking medications increasing vitamin D catabolism:
•
•
Phenobarbitone
•
Carbamazepine
•
Dexamethasone
•
Rifampicin
•
Nifedipine
.
Spironolactone
• Ritonavir
Cyproterone acetate
Babies of vitamin D-deficient mothers
Sources: (15, 46, 47, 54-56, 65-70, 72).
with evidence that these adverse outcomes (eg, risk of pre-
eclampsia, gestational diabetes, low birthweight, and the
risk of severe postpartum hemorrhage) are reduced after vita-
min D supplementation have also been shown (77, 78). Poor
vitamin D status has also been associated with increased risk
of low birth weight (79-81), increased risk of
preterm birth
(81, 82), and offspring's adverse anthropometric and neuro-
developmental outcomes (81), while supplementation or suf-
ficient vitamin D status was found to be protective against
the risk of low birth weight, preterm birth, and small for ges-
tational age (80), and associated with improved offspring vita-
min D sufficiency status, reduced fetal or neonatal mortality,
and improved fetal and future linear growth (83, 84).
Screening and testing vitamin D status-conclusions
Screening the general population for vitamin D deficiency is
very expensive and does not result in practical clinical benefits.
25(OH)D measurements have primarily been indicated in pa-
tients with musculoskeletal disorders, but the increased
awareness of potential pleiotropic effects has widened the
interest in screening, although without any definitive evidence.
The most important risk factors in the general population,
as identified by recent studies, include low ambient UV radi-
ation, low vitamin D intake, and gene polymorphisms. Their
inclusion as primary risk factors in risk stratification ap-
proaches to assess vitamin D status will help effectively target
25(OH)D assessment in those most in need and at risk.
Finally, further studies-including those with health
Endocrine Reviews, 2024, Vol. 45, No. 5
economic measures—are warranted to best identify all the sit-
uations in which the assessment of vitamin D status is actually
needed or not.
Methods: Assays, Thresholds, and Standardization
Accuracy and precision of vitamin D measurements are crucial
to properly use the values obtained in biological fluids. The la-
boratory methods should be detailed in clinical trials, scientif-
ic papers, and even in the reports released to physicians and
patients. The measurements could be obtained by either
antibody-based methods (chemiluminescent or immunoenzy-
matic) or by liquid chromatography-mass spectrometry
(LC-MS or LC-MS/MS), with the latter giving more consistent
and accurate results; regardless, the reference material should
also be indicated in the report (85). The laboratory should de-
fine the reference values considering the method used for ana-
lyzing the molecule(s). The unit of measure (molar or mass)
should be clearly indicated. The mol/L unit should be pre-
ferred as the SI standard unit; alternatively, both mol/L and
ng/mL should be reported. Moreover, the critical difference
(or least significance change), or reference change values,
that is, the value (percentage) testifying to a real modification
of the molecule(s) concentration between 2 consecutive meas-
urements, assayed with the same method, in the same patient,
and, on the contrary, that the modification is not only depend-
ent from analytical and biological variability (natural oscilla-
tion of values in individuals), should be known and properly
considered (85).
Assay standardization remains a major challenge to inter-
preting data from various studies evaluating vitamin D and
its metabolites and analogues. It should be a priority to enable
rational pooling of data and implementation of meta-analyses
relating specific vitamin D metabolites to various outcomes of
interest (2). Indeed, it has been suggested that reporting stand-
ardized 25(OH)D results is required for funding and subse-
quent publication of vitamin D-related research data (51).
Mean bias between -5% and +5% is one of the two per-
formance thresholds used to define a 25(OH)D assay as being
standardized by the Vitamin D Standardization Program
(VDSP) (2, 86). However, a flaw in that threshold is that an
assay with a mean bias within that range may display enor-
mous variability outside those limits when, in actuality,
what is wanted is an assay with few measurements outside
them (87).
Data from the VDSP's Vitamin D Standardization
Certification Program conducted by the Centers for Disease
Control and Prevention (CDC) in the November 2019
report the last report before COVID-19-show the flaw in
the mean bias threshold in assays certified by the CDC to be
standardized (88). In the CDC's report, 20 immunoassays
and 17 LC-MS/MS assays were certified as being standardized
for serum total 25(OH)D measurement (Table 2).
The mean individual samples pass rate for LC-MS/MS as-
says (61%) was 2 times higher than the rate for immunoassays
(30%) (t=-7.2; P<.01). LC-MS/MS assays provided the
highest mean value (mean = 61%); however, there was con-
siderable overlap. The Fujirebio Lumpulse had the highest
pass rate for an immunoassay: 68%. The Boditech Ichroma
had a 65% pass rate, and 3 immunoassays had a 42% pass
rate, namely the Abbott Architect, IDS CLIA, and Siemens
Maglumi ones. The VDSP's definition of the mean bias thresh-
old should thus be revised. The criteria we suggest for any
Page 9
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Endocrine Reviews, 2024, Vol. 45, No. 5
Table 2. Mean individual samples pass rate for 40 serum samples by
Centers for Disease Control-certified standardized laboratories by
assay type
Assay type
No., certified Individual samples pass rate, %
Mean, % SD Minimum Maximum
Immunoassay
LC-MS/MS
20
17
30º
12.5 8
68
61ª
14.0 38
88
CDC individual samples pass rate is the percentage of individual samples out of 40
provided that met the certification criteria of ±5% bias. This information was
provided starting in February 2017. Data analyses by Prof Christopher Sempos.
Source: CDC (87).
Abbreviations: CDC, Centers for Disease Control and Prevention; LC-MS/MS,
liquid chromatography-tandem mass spectrometry.
at=-7.2; P=.00001.
revision are (1) consistent with the original guidelines, that is,
not an abrupt change; (2) easily calculated and easily under-
stood; (3) easy to operationalize; (4) easily modified to pro-
mote change over time; and (5) will promote competition
among assay manufacturers.
Importantly, we also suggest that 25(OH)D assays continue
to monitor their regular performance using an external quality
assessment scheme that provides target reference values from
a reference measurement procedure approved by the Joint
Committee for Traceability in Laboratory Medicine provided
by an LC-MS/MS standardized assay (eg, DEQAS, Charing
Cross Hospital, London UK). Such a process emphasizes the
importance of assay accuracy and, given that true concentra-
tion is available in such specimens, allows their use in retro-
spective standardization of 25(OH)D data (2, 89).
Assessment of Other Vitamin D Forms and Main
Metabolites
As discussed earlier, the vitamin D status assessment is based
on the 25(OH)D serum level measurement. However, other
forms of vitamin D such as free 25(OH)D, bioavailable
25(OH)D, DBP, or 1,25(OH)2D levels could be used as bio-
markers of vitamin D repletion, defined by effect on classical
and nonclassic vitamin D outcomes. As for 25(OH)D meas-
urements, these tests would also need to be standardized to en-
sure accuracy and replicability.
For circulating 25(OH)D, it is estimated that approximately
85% to 90% is bound by DBP and 10% to 15% by albumin;
therefore, free 25(OH)D levels are estimated to be less than
1% of the total and can vary according to DBP and albumin
polymorphisms and binding affinity (90-92). The free and
not the total 25(OH)D concentration in cell cultures affects
a biological response. While this is harder to assess in vivo,
some tissues with the megalin/cubilin complex, like the kidney
and parathyroid gland, can take up the vitamin D metabolites
bound to DBP (93). Nonetheless, free 25(OH)D may be highly
relevant to local intracellular (eg, osteoblasts, renal cells,
muscle cells) synthesis of 1,25(OH)2D, which can behave in
a paracrine and autocrine fashion (4). In normal populations,
total and free 25(OH)D, as well as free and calculated 25(OH)
D, are correlated (~60%-70% in healthy individuals), and
there is no clear evidence for a need to measure free metabo-
lites in healthy individuals and many clinical settings (92-
97). However, this may not hold true in conditions affecting
DBP such as pregnancy, cirrhosis, acute illness, conditions
633
that may affect the affinity of DBP or albumin to its ligands,
and even in aging nursing home residents (4, 69, 94), for
whom the free concentration is a better assessment than the
total.
Measurement of 1,25(OH)2D may contribute to the diag-
nosis of conditions with low calcitriol levels, such as
1α-hydroxylase deficiency, or those associated with high
1,25(OH)2D levels, such as hereditary vitamin D-resistant
rickets, granulomatous conditions (sarcoidosis and tubercu-
losis), and the hypophosphatemic syndromes (4, 69).
Available evidence to date is rather limited to determine
whether free or bioavailable 25(OH)D or 1,25(OH)2D is the
better biomarker of 25(OH)D availability to local tissues
and of its effect on target organs in special situations. The ex-
tremely low serum concentrations of 25(OH)D and 1,25(OH)
2D found in mice and humans with genetic absence of DBP
without implications on calcium homeostasis is the best argu-
ment for the free “vitamin D” hypothesis.
Assessment of Other Metabolites
Improvements in LC-MS/MS have triggered a revolution in
small-molecule clinical chemistry, particularly the analysis
of steroid hormones. The additional sensitivity and selectivity
provided by the recently emerged LC-MS/MS techniques
make it now feasible to measure most of the circulating vita-
min D metabolites of value to clinicians and physiologists in
human and animal studies. A comprehensive analysis can
now assay 8 metabolites simultaneously (cholecalciferol,
25(OH)D, 3-epi-25(OH)D, 24,25(OH)2D, 25,26(OH)2D,
1,25(OH)2D, 1,24,25(OH)3D, and 25(OH)D-26,23-lactone)
by judicious use of liquid-liquid-extraction and immune-
extraction steps. Besides specific clinical situations (98, 99),
a few previous reports have also highlighted a potential role
for vitamin D metabolites, in particular of the 24,25 to
25(OH)D ratio, in better-predicting fracture risk as compared
to only 25(OH)D levels (100, 101).
Infantile hypercalcemia, type 1, caused by defects in
CYP24A1
Despite the name of the disease, infantile hypercalcemia,
type 1 affects individuals throughout life, usually causing
nephrolithiasis. It is especially problematic in pregnant
females due to the placental production of 1,25(OH)2D3,
which cannot be efficiently metabolized. The utility of meas-
uring 24-hydroxylated forms, particularly the 25(OH)D to
24,25(OH)2D ratio, has been established as a useful screening
tool by groups worldwide. Ratios are elevated from 5 to 25 in
normal individuals to more than 80 in infantile hypercalce-
mia-affected individuals (102). It is important to recognize
that this same enzymatic defect can be identified in adults
with unexplained 1,25(OH)2D-dependent hypercalcemia.
These individuals present with hypercalcemia, hypercalciuria,
kidney stones, and suppressed levels of PTH.
Other hypercalcemias
Many causes of hypercalcemia can be distinguished by
their distinctive pattern of vitamin D metabolites. Kaufmann
et al (98) identified several patient groups by studying
the vitamin D metabolome. These include patients with
Williams syndrome exhibiting an elevated level of 25(OH)
D-26,23-lactone, a stable metabolite with high affinity for
DBP; patients with hypervitaminosis D taking toxic doses of
Page 10
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
634
vitamin D exhibiting very high 25(OH)D but suppressed
1,25(OH)2D, and where several other vitamin D metabolites
may contribute.
Chronic kidney disease
Many studies have documented a fall in serum 25(OH)D and
1,25(OH)2D with a decline in renal function. Studies of the
vitamin D metabolome over the 5 stages of chronic kidney dis-
ease (CKD) have revealed that the same phenomenon also ap-
plies to 24,25(OH)2D. Patients and animal models with
experimental kidney disease both show changes in the levels
of 24,25(OH)2D3 and 1,24,25(OH)3D3 with changes in glom-
erular filtration rates (103). The clinical consequences of these
changes remain to be elucidated.
Routine documentation of vitamin D metabolites in
randomized controlled trials
In most recent large RCTs, participants were monitored only
for health effects and serum 25(OH)D levels. One study—the
so-called Calgary study (or JAMA study) (104)—used doses
of up to 10 000 international units (IU) of vitamin D/day,
monitored only 25(OH)D, and reported deleterious effects
of the vitamin D on bone mineral density (BMD) (BMD was
assessed with high-resolution peripheral quantitative com-
puted tomography and not with dual-energy x-ray absorpti-
ometry). Although this JAMA study reported only 25(OH)D
data (105), by reanalyzing the serum from participants in
the study for the full vitamin D metabolome including
1,25(OH)2D3, 24,25(OH)2D3, and 1,24,25(OH)3D3, Burt
and colleagues (105) found that several vitamin D metabo-
lites, including 1,24,25(OH)3D3 but not 1,25(OH)2D3, were
elevated in individuals given the 10,000 IU of vitamin D/day
dose, a fact that could explain the bone loss observed at
high supplementation rates.
In conclusion, the study of a wider array of vitamin D me-
tabolites provides insight into a limited number of diseases.
It can potentially improve understanding of vitamin D status
and its relationship to multiple diseases.
Clinical Outcomes of Vitamin D Deficiency
Skeletal Outcomes
Skeletal outcomes of vitamin D deficiency are summarized in
Fig. 3. Vitamin D deficiency leads to a decrease in intestinal
absorption of calcium and phosphate. Other biochemical ab-
normalities, such as hypocalcemia, hypophosphatemia, and
an increase in alkaline phosphatase, become apparent when
serum 25(OH)D concentrations are lower than 25 nmol/L
(106). In milder forms of vitamin D deficiency, the lower
calcium concentration causes secondary hyperparathyroid-
ism, which increases the conversion of 25(OH)D into
1,25(OH)2D, increasing calcium absorption and correcting
serum calcium (4, 107). Secondary hyperparathyroidism
causes an increase in bone turnover, with relatively higher
bone resorption at cortical sites (107-112). More severe long-
standing vitamin D deficiency causes a decrease in the mineral-
ization of newly formed osteoid tissue. This is visible in bone
biopsies as an increase in osteoid surface and volume and in-
creased thickness of osteoid seams, leading to the clinical pic-
ture of osteomalacia (4, 107). Vitamin D deficiency and
related secondary hyperparathyroidism cause bone loss and
fractures in older adults. The incidence of hip fractures
Endocrine Reviews, 2024, Vol. 45, No. 5
attributable to vitamin D deficiency has been estimated at
5% to 10% (113). Meta-analyses of clinical trials with vita-
min D and calcium have demonstrated a decrease in hip and
other fractures of around 10% in nursing home residents,
whereas vitamin D alone was not effective (113-115). In these
studies, baseline mean serum 25(OH)D after cross-calibration
was found to be very low-namely less than 25 nmol/L—as
was the calcium intake (116). As almost all effective trials
used a calcium supplement in addition to vitamin D, the effect
on BMD of vitamin D supplements alone is difficult to deter-
mine, but it is considered to be less than 1% (113), and high
doses may even be harmful when administered to vitamin
D-replete individuals (104). Recent RCTs such as the VIDA,
VITAL, and D-Health studies do not show skeletal benefits
for mostly vitamin D-replete adults and older individuals;
for example, in the VITAL trial, cholecalciferol supplementa-
tion did not result in a significantly lower risk of fractures (to-
tal, nonvertebral, and hip fractures) than placebo among
generally healthy midlife and older adults not selected for vita-
min D deficiency, low bone mass, or osteoporosis (117, 118).
In the D-Health study, large bolus monthly doses (60 000 IU)
resulted in no increase nor decrease in fracture risk overall.
However, the hazard ratio appeared to decrease with increas-
ing follow-up time (119). Interestingly, in a recent retrospect-
ive longitudinal study (120), the use of cholecalciferol was
associated with reduced incidence of morphometric vertebral
fractures in high skeletal risk, such as acromegaly (121).
In a recent umbrella review of meta-analyses of vitamin D
RCTs, the only consistent significant findings were for calcium
and vitamin D, and not vitamin D alone, in reducing the risk
of hip fractures by 16% to 39%, in 8 of 13 meta-analyses, and
of any fracture, by 5% to 26%, in 8 of 14 meta-analyses.
These findings were driven by events in institutionalized older
and frailer individuals (122).
In children, the lack of calcium and phosphate causes the
expansion of the epiphyseal growth plates due to decreased
apoptosis of the hypertrophic chondrocytes, clinically visible
as thickening near the joints and radiologically as widening
of the growth plates (4). The weaker bone leads to typical
deformities, such as knock knees (genua valga) and bowlegs
(genua vara). The occurrence of rickets is mainly restricted
to the Middle East and some countries in Asia, such
as Mongolia and parts of China and India (123, 124),
while it is also observed in immigrants and refugees in other
countries (125).
Extraskeletal Outcomes
Putative extraskeletal outcomes of vitamin D deficiency are
summarized in Fig. 4. There are many preclinical data on
the extraskeletal effects of the vitamin D endocrine system, in-
cluding gene regulation, cellular function, and in vivo animal
studies. Indeed, about 3% of the mammalian genome is under
some control of vitamin D, and most cells express VDR or
can synthesize the active hormone 1,25(OH)2D locally.
Observational data largely align with these data as poor vita-
min D status is associated with many human diseases (4). To
complete the observational data, many large-scale trials that
evaluated the effects of vitamin D supplementation on several
extraskeletal health outcomes have been carried out recently,
including the large VITAL (USA) (118, 126-129) and
D-Health (Australia) (64, 129, 130) studies, as well as the
ViDA (New Zealand) (131-133), FIND (Finland) (134,
Page 11
Endocrine Reviews, 2024, Vol. 45, No. 5
Reduced synthesis
and dietary intake
Vitamin D
deficiency
Skeletal effects
Reduced intestinal
absorption
Clinical impairment on
metabolism and action
Osteopenia &
osteoporosis
(mild vitamin D deficiency)
Fragility
fractures
635
Osteomalacia &
Rickets
(severe vitamin D deficiency)
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Figure 3. Skeletal effects of vitamin D deficiency. A deficient vitamin D status can cause impairments in the skeletal system such as osteopenia, os-
teoporosis, osteomalacia, and rickets, resulting in high risk for fragility fractures. Clear boxes with dashed outlines refer to the risk factors for vitamin D
deficiency; dark boxes refer to the negative skeletal effects of vitamin D deficiency.
135), and the D2d (USA) (136, 137) trials (Table 3). Smaller
scale studies, such as the Calgary and the DO-Health
(Switzerland), provide additional data. Moreover, there are
now good genetic data on the prediction of serum 25(OH)
D, which resulted in about 100 MR studies (117).
Cancer
No effects of vitamin D supplementation on cancer risk were
observed in the large VITAL and ViDA trials, nor the FIND
trial using daily dosing in older participants, nor on cancer
mortality in the D-Health study, which used monthly dosing
―in line with prior trials and MR results (117, 130, 134).
Based on several MR studies, small changes in vitamin D sta-
tus are unlikely to affect cancer incidence (117). However, a
subanalysis of the VITAL trial (although not corrected for
multiple end point analysis) showed that vitamin D supple-
mentation could have some minor benefits in individuals
with normal BMI (128). In addition, several independent tri-
als have suggested, in post hoc analysis, the potential benefits
of vitamin D supplementation on cancer mortality, especially
when the follow-up is longer than 4 years (139). A meta-
analysis of RCTs suggested that vitamin D supplementation
decreased cancer mortality (140); an updated version of this
study specifically designed to examine whether results varied
by daily vs infrequent large-bolus dosing and by whether the
trial participants had obesity or not found that overall
benefit of vitamin D supplementation is lost when all the
studies are considered. However, when considering daily
regimens, vitamin D supplementation reduced total cancer
mortality and incidence in normal-weight individuals (141).
Therefore, a link between vitamin D status and cancer inci-
dence or mortality can be hypothesized, and supplementation
might be effective only with daily dosages, especially in people
with BMI within a normal range (117, 141).
Cardiovascular risk
Convergent evidence from MR studies and RCTs suggests that
vitamin D supplementation does not decrease the risk of car-
diovascular disease (CVD), especially in vitamin D-replete
adults. This conclusion may also apply to those with vitamin
D deficiency based on subgroup analyses of the ViDA and
VITAL trials. However, both studies recruited very few partic-
ipants with severe vitamin D deficiency (117), rendering these
conclusions uncertain. These null findings were corroborated
by a meta-analysis of 21 RCTs (142). Nonetheless, more recent
findings might suggest some small benefits. A detailed analysis
of the VIDA trial found some modest benefits on central (but
not peripheral) blood pressure, but the implications of this
Page 12
636
Autoimmune &
infectious diseases
Reduced synthesis
and dietary intake
Cardio-respiratory
diseases
(heart failure, hypertension,
pneumonia)
Vitamin D
deficiency
Impaired muscle
function & strength
(increased fall risk)
Reduced intestinal
absorption
Obesity & diabetes
Clinical impairment on
metabolism and action
Cancer incidence
& mortality
Endocrine Reviews, 2024, Vol. 45, No. 5
↑
Quality of life
Mortality
Acute COVID-19 severity
& Long COVID risk
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Figure 4. Putative extraskeletal effects of vitamin D deficiency and their implication in human health. A deficient vitamin D status is associated with
several extraskeletal effects. These include increased risk of diabetes and autoimmune, infectious, cardiovascular, and respiratory diseases, as well as
increase in cancer incidence and mortality. Such impairments result in lower quality of life and higher mortality, and can even increase acute COVID-19
severity and long COVID risk. Clear boxes with dashed outlines refer to the risk factors for vitamin D deficiency; dark boxes refer to the negative extra-
skeletal effects of vitamin D deficiency.
observation are limited because of the small scale of this VIDA
substudy (133). The FIND trial failed to note a reduction in the
number of major CV events, which was one of the two primary
end points (134); however, subsequent exploratory analyses
revealed that high-dose vitamin D supplementation might re-
sult in benefits in atrial fibrillation prevention in older individ-
uals, even in case of relatively high baseline 25(OH)D
concentrations (135). In the D-Health trial, the overall rate
of major CV and especially the rate of myocardial infarction
and coronary revascularization-was lower in the interven-
tion
group compared to the placebo group, although the abso-
lute risk difference was small, and the CI was consistent with a
null finding (hazard ratio 0.91; 95% CI, 0.81-1.01); moreover,
the protective benefits could be higher in those taking CV drugs
at baseline (138).
Respiratory effects
Vitamin D is known to influence the immune system. Most
immune cells express the VDR and vitamin D metabolism-
related enzymes; 1,25(OH)2D, in particular, induces innate
antimicrobial effector mechanisms such as the antimicrobial
peptides cathelicidin LL-37 and human beta-defensin 2 (4).
Indeed, clinical data regarding the effects of adequate vitamin
D status and supplementation on respiratory infections con-
firm, at least in part, its potential beneficial outcomes. Serum
25(OH)D levels of less than 25 nmol/L are associated (obser-
vationally and genetically) with an increased risk of bacterial
pneumonia (143). Individual participant data from a meta-
analysis of 25 trials showed a small but significant decrease
in the incidence of acute respiratory infections in the vitamin
D group compared with the control group when baseline vita-
min D status was poor (<25 nmol/L) (144). A more recent, up-
dated meta-analysis from the same group, including almost 50
RCTs, shows a protective but very small effect against respira-
tory infections following vitamin D supplementation with dai-
ly doses of 400 to 1000 IU; in contrast to their first
meta-analysis, baseline vitamin D status did not modify the re-
sults in this more recent one (145).
As respiratory tract infections are common in children,
some promising data are available also in this setting.
Children with poor vitamin D status were reported to be
more prone to developing respiratory infections, although a
conclusive association between the severity of respiratory
Page 13
Follow-up, Primary outcome(s)
Conclusions and comments
Table 3. Characteristics and results of the most recent and largest randomized controlled trials on vitamin D supplementation
Study
Participants
(n)
Age (mean
± SD), y
Sex (% of
women)
Mean
BMI
Ethnicity"
(% White
ethnicity)
Serum 25(OH)D, ng/mL
Baseline Final
Dose used
y
VITAL (117,
125-128)
25 871
67 ± 7
51
28
71
30.8 ±
10
42 ± 10
2000 IU/d +
omega-3
1 g/d
5.3
D-Health (64,
129, 130)
21 315
69.3
46
28
28
ViDA
(131, 132,
138)
5110
66 ± 8
58
58
96.5%
31 ± 10d 46 ± 12
60 000 IU/mo 5.7
All-cause mortality
Endocrine Reviews, 2024, Vol. 45, No. 5
Invasive cancers and major
CV events
Incidence of metastatic or
fatal cancer
Two or more falls and falls
resulting in a doctor or
hospital visit
All incident autoimmune
diseases
Incident total, nonvertebral,
and hip fractures
Risk of falling
Major CV events
Fractures and falls
Cancer incidence and
mortality
Incident major CVD and
invasive cancer
Atrial fibrillation risk
End point not met, but
reduction in total cancer
mortality when excluding
first 1-2 y of follow-up
VD reduced metastatic or fatal
cancers by 17%; strongest
reduction in normal BMI
End point not met
VD reduced autoimmune
diseases by 22%
End point not met; enrolled
individuals were generally
healthy and not selected for
VD deficiency, low bone
mass, or osteoporosis
End point not met; VD
increased cancer risk when
first 2 y of follow-up were
excluded
End point not met; VD
increased risk when BMI
<25, but not when BMI ≥25
End point not met; VD might
reduce CV events (small
absolute risk difference and
CI consistent with null
finding); VD reduced
myocardial infarction by 19%
End point not met (121); in one
substudy, VD lowered central
blood
pressure in deficient
participants
End point not met
End point not met; daily or
weekly dosing for longer
period may require further
study
End point not met; study failure
possibly due to sufficient VD
status in most participants at
baseline
VD reduced atrial fibrillation
risk by 27%-32%
29 ± 5.1
83
27 ± 9€
54 ± 16
200 000 IU + 3.3
Incident CVD and death
100 000 IU/
mo
FIND
(133, 134)
2495
685
43
43
27 ± 4
100
30 ± 7
40±9 (1600 IU/d
arm) 48 ± 9
(3200 IU/d arm)
1600 or
4.3
3200 IU/d
9/9/9////// w po
(continued)
637
33
Page 14
638
Conclusions and comments
Partly drafted with data from Bouillon et al (117).
T2D in adults with
Development of T2D
according to intratrial
serum 25(OH)D level
prediabetes
End point not met
VD resulting in 25(OH)D level
≥100 nmol/L reduces risk of
T2D
study included Asian people and a small number of indigenous Māori individuals.
"The VITAL and D2d studies included different American racial and/or ethnic groups including Black people and Hispanic people. The ViDA
Abbreviations: 25(OH)D, 25-hydroxyvitamin D; BMI, body mass index; CV, cardiovascular; CVD, cardiovascular disease; IU, international units; N.A., not available; T2D, type 2diabetes; VD, vitamin D.
"Final serum 25(OH)D concentration vitamin D-treated groups only.
Primary outcome(s) refers both to main trial and subsequent analyses.
"Evaluated in placebo group during follow-up.
Deseasonalized mean values.
Table 3. Continued
Study
(u)
± SD), y
Participants Age (mean
Sex (% of
women)
Mean
Ethnicity"
BMI
(% White
ethnicity)
Serum 25(OH)D, ng/mL
Baseline Final'
Dose used
Follow-up, Primary outcome(s)
y
D2d (135, 136)
2423
60 ± 10
45
32 ± 5
67
28 ± 10
54 ± 15
4000 IU/d
2.5
Endocrine Reviews, 2024, Vol. 45, No. 5
infections and low vitamin D levels was not clearly estab-
lished (146). RCTs show that vitamin D supplementation
can benefit infants, toddlers, and preschool children aged
0 to 5 years with a quicker recovery and fewer respiratory
symptoms (147). Unfortunately, study heterogeneity in
terms of design, vitamin D supplementation doses, and dur-
ation, along with participant characteristics, make it prob-
lematic to pool data and, thus, difficult to draw definitive
conclusions (79, 147).
There is also consistent evidence for an association between
low 25(OH)D levels and poor COVID-19 outcomes, al-
though the evidence supporting a beneficial effect of vitamin
D supplementation in decreasing the risk of COVID-19 com-
plications is conflicting (8, 148-151). An MR study found no
evidence that vitamin D is protective against SARS-CoV-2 in-
fection or COVID-19 severity (152). However, a meta-
analysis of several observational studies comprising almost
2 million adults suggests that inadequate vitamin D status in-
creases susceptibility to COVID-19 and severe COVID-19,
while the association with mortality was less robust. Of
note, the included studies were at high risk of bias and hetero-
geneity, and the heterogeneity in RCTs precluded their meta-
analysis (151). Furthermore, low 25(OH)D levels were also
recently associated with an increased risk for long COVID oc-
currence (150). However, a phase 3 RCT found no effect of
vitamin D supplementation on the risk of developing long
COVID after an episode of COVID-19 (142). Also, deficient
vitamin D status was recently reported to be associated with a
reduced long-term immune response to the anti-COVID-19
vaccination (153).
Vitamin D supplementation also seems effective in safely
and substantially reducing the rate of moderate/severe acute
exacerbations of chronic obstructive pulmonary disease in pa-
tients with baseline 25(OH)D levels less than 25 nmol/L—but
not in those with higher levels (154). A meta-analysis, con-
versely, found no role for vitamin D supplementation in im-
proving expiratory lung function (155).
Regarding asthma, there are insufficient RCTs to evaluate
the potential benefit of vitamin D or its hydroxylated metab-
olites in improving its control or reducing the risk of exacerba-
tions. However, as individuals with baseline 25(OH)D levels
less than 25 nmol/L and those with severe asthma were poorly
represented, and since one study investigating the effects of
calcidiol yielded positive results, further studies are warranted
in these populations and settings (156).
Autoimmune diseases
Conversely, from the innate immune system, the adaptive im-
mune system is downregulated by 1,25(OH) 2D in animal
models. Thus, vitamin D deficiency might predispose to auto-
immune diseases. Observational studies have suggested this
effect might apply to humans (4).
The VITAL RCT showed that vitamin D supplementation
decreased the risk of autoimmune diseases, especially rheuma-
toid arthritis and polymyalgia rheumatica, and at least 8 large
MR studies all agree that genetically predicted lower 25(OH)
D levels increased the risk of developing multiple sclerosis ei-
ther during adolescence or adulthood (117, 129). In any
case, the low number of intervention studies so far conducted
does not allow clarification of the relationship between vita-
min D and autoimmune diseases. However, these studies to
date seem promising.
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Page 15
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Endocrine Reviews, 2024, Vol. 45, No. 5
Diabetes
Despite observational studies consistently confirming lower
serum 25(OH)D concentrations in patients with T2D or meta-
bolic syndrome (4), most MR studies have not supported these
conclusions (117). In a small subgroup of individuals with obesity
and prediabetes, supplementation provided some modest benefit,
albeit lower than lifestyle modifications or metformin (157).
Of note, daily vitamin D supplementation (4000 IU) in the large
D2d trial did not retard the progression of prediabetes into
T2D. A post hoc and meta-analysis, however, suggested a pos-
sible beneficial effect in individuals with vitamin D deficiency
(<30 nmol/L) at baseline or in participants who were able to
achieve consistently high (≥100 nmol/L) serum 25(OH)D levels
(152). Furthermore, analysis of the combined results of the
D2d (US), Tromsø (Norway), and DPVD (Japan) RCTs―which
were specifically designed and conducted to test whether vitamin
Dreduces the risk of diabetes in adults with prediabetes-showed
that vitamin D supplementation reduced the risk of developing
T2D in people with prediabetes not selected for vitamin D defi-
ciency (158). In all 3 trials, the risk for diabetes was reduced in
the group assigned to vitamin D compared to the placebo group,
which did so in a remarkably similar way. The observed differen-
ces missed statistical significance in any trial because the reported
risk reductions were smaller than each trial was powered to
detect. An updated individual participant data meta-analysis of
the same trials (159) showed that vitamin D reduced the risk of
progression from prediabetes to diabetes by 15%. Also, vitamin
D increased the likelihood of regression to normal glucose regu-
lation by 30%, with no evidence of risk. In additional analyses,
participants in the vitamin D group who maintained intratrial
blood 25(OH)D of 50 ng/mL or greater (≥125 nmol/L)
had a 76% risk reduction in new-onset diabetes compared
to those who maintained blood 25(OH)D of 20 to 29 ng/mL
(50-75 nmol/L). All participants received and were encouraged
to follow the current lifestyle-based advice for diabetes preven-
tion. Based on the results of this meta-analysis, the benefit-to-risk
ratio of vitamin D to lower the risk of developing T2D in adults
with prediabetes is favorable. These results should not be extrapo-
lated to the general population at low or average risk for diabetes,
as the benefit-to-risk ratio of high doses for diabetes prevention
may not be favorable. Despite these promising results, some
questions remain, that is, the optimal vitamin D dose or for-
mulation and the specific blood 25(OH)D level to maximize
benefit with little or no risk of any side effects (159).
Thus, the evidence from large-scale MR studies and RCTs is
convergent and does not support vitamin D supplementation
to prevent T2D in the general population. However, vitamin
D supplementation benefits those with prediabetes and a pre-
disposition to T2D, especially those with vitamin D deficiency.
Additional studies or more in-depth analyses of the existing
studies are needed to validate these findings (117, 159).
Mortality
Observational data have repeatedly linked poor vitamin D sta-
tus with increased mortality. Large, older meta-analyses deal-
ing mostly with women older than 70 years (160, 161) showed
a 6% to 11% reduction in mortality. However, adding the
newest megatrials eliminated this effect, possibly because
they recruited a younger population. In these megatrials, over-
all mortality was much lower than shown in the previous meta-
analyses (160, 161), and no effect of vitamin D supplementa-
tion on overall mortality was observed (128).
639
A Cochrane meta-analysis of 56 randomized trials including
almost 100 000 participants, of whom were women older than
70 years, revealed that vitamin D, administered over 4 years,
decreased mortality; this effect was seen in 38 trials of vitamin
D3, but not with other forms of vitamin D (161). A newer meta-
analysis of 52 RCTs, including more than 75 000 individuals,
concluded that vitamin D (either vitamin D3 or D2) supplemen-
tation did not change mortality compared with no supplemen-
tation (162). Again, subanalyses found that vitamin D3
(instead of D2) supplementation tended to reduce mortality.
Some MR studies found a link between lower predicted serum
25(OH)D and mortality, especially in individuals with rather
poor vitamin D status (<16 ng/mL) (163-165). An individual
participant data meta-analysis of almost 27 000 study partici-
pants with 25(OH)D levels standardized per VDSP protocols
showed an association between low 25(OH)D and increased
risk of all-cause mortality (166). The positive but small effect
of vitamin D on mortality was confirmed by a recent umbrella re-
view of observational, randomized, and MR studies (167). In
conclusion, if vitamin D supplementation benefits extraskeletal
health outcomes and major diseases, it is likely to have some ef-
fects on mortality, especially in older adults with
poor vitamin
D status, but not in younger, replete individuals (117).
Summary of Vitamin D Deficiency-associated
Clinical Outcomes
The long-known skeletal benefits of vitamin D and calcium re-
lated to rickets or osteoporosis remain valid. Most reported
extraskeletal benefits of vitamin D were not confirmed by re-
cent, large RCTs (see Table 3). The gradual increase in vitamin
D levels in Western populations may explain these null find-
ings, and older trials and meta-analyses may be more likely
to show benefits because individuals were more likely to be
vitamin D deficient than they are nowadays. RCTs and meta-
nalyses published to date do not have adequate power to
evaluate important subgroups, such as individuals with low
25(OH)D levels, men, the very old, ethnic groups other than
White individuals, and those from low-income countries.
Moreover, most of the studies use adverse events data to identify
fractures and were performed in adults who were vitamin D re-
plete at baseline in whom benefit would be unlikely and toxicity
possible. Such studies confound the identification of possible
beneficial effects in vitamin D-deficient individuals who might
benefit from supplementation. Thus, when it comes to vitamin
D, it is advisable to "giveth to those who needeth” (168). In
fact, the benefit-to-risk ratio for vitamin D depends on the target
population and medical condition. It would be incorrect to ex-
trapolate vitamin D guidelines that apply to the general popula-
tion (such as those from the US National Academic of Medicine)
to avoid vitamin D deficiency (ie, rickets, osteomalacia) and pro-
mote bone health to special populations for whom the
benefit-to-risk ratio of vitamin D would be different.
Nonetheless, RCTs, MR studies, and metanalyses suggest a
link between vitamin D status with the immune system and
diabetes, as well as fleeting effects on some CV events and
some benefits on mortality risk when vitamin D3 is used.
Vitamin D Supplementation
Dosing Regimens
The term "dose" in relation to vitamin D is typically used to
signify the measured quantity of vitamin D (usually
Page 16
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
640
cholecalciferol, but other formulations such as ergocalciferol,
eldecalcitol, calcifediol, etc are also available) in a pill. It is ex-
pressed as μg or IU (where 10 µg is 400 IU). The dose of chole-
calciferol is considered an important measure as it correlates
with the change in blood 25(OH)D level, which is commonly
used to define vitamin D status and correlates with important
clinical outcomes. Doses can be considered as "loading" or
"maintenance." The most common use of a loading dose is
to rapidly improve a low blood 25(OH)D; however, the clin-
ical wisdom of this approach is questionable, especially given
studies that demonstrate adverse effects with very high doses
given infrequently, as discussed next. Intermittent administra-
tion of large doses is also used to optimize adherence. Daily
doses are generally preferred when vitamin D replacement is
considered necessary. The effect of a given dose on changing
blood 25(OH)D varies considerably from person to person
due to many factors, such as body weight, absorption, diet, de-
gree of adiposity, CYP2R1 activity, DBP. The recommended
dietary allowance for vitamin D by the National Academy
of Medicine is set at 400 to 800 IU per day, and the tolerable
upper intake level at 4000 IU per day; however, the "optimal"
dose of vitamin D varies by the desired outcome, and other au-
thors suggest that the upper limit of safety may be lower than
4000 IU per day (169-171). For example, 400 to 800 IU of
vitamin D per day may be adequate to avoid clinical
vitamin D deficiency and maintain calcium homeostasis in
healthy individuals. Doses of vitamin D higher than the rec-
ommended upper limit may be associated with toxicity; none-
theless, daily doses up to 10 000 IU have been used without
safety concerns (172). Careful and judicious use of vitamin
D will permit the realization of potential benefits and achiev-
ing optimal outcomes.
Generally, there is a lack of consensus about the recom-
mended vitamin D supplementation regimen (doses, adminis-
tration schedule, treatment duration, etc) (173). Such
heterogeneity can be explained, at least partly, by the scarcity
of comparative pharmacokinetics studies for different dosing
schedules (174-176). Moreover, different underlying condi-
tions (eg, obesity) might reduce the effect of vitamin D supple-
mentation (177, 178). Growing evidence suggests that the
treatment schedule itself (ie, bolus vs frequent administration)
may differently affect the effectiveness of the treatment (27,
179) and also clinical outcomes, with recent studies and a
few meta-analyses showing more promising results with fre-
quent administration schedules on skeletal and extraskeletal
outcomes (4, 141, 144, 180, 181).
In this perspective, vitamin D supplementation guidelines
should be specific for age group, body weight, ethnicity
(skin type), and latitude of residence. For example, differences
in serum 25(OH)D by BMI and absolute body weight have
been reported (182-185). Vitamin D dose per kilogram of
body weight per day could explain a 34.5% variation in circu-
lating 25(OH)D in multivariable regression analyses of data
pooled from several studies (184), leading to pronounced dif-
ferences across BMI categories. Obese and overweight indi-
viduals tend to have serum 25(OH)D levels that are, on
average, around 20 nmol/L lower and 8 nmol/L lower than
those of normal-weight individuals, requiring 2.6 and 1.47
times higher supplementation, respectively (185). This is
somewhat consistent with ES guidelines suggesting that the
vitamin D dosage for obese people is "three times" greater
than the recommended dose for individuals with normal
body weight (46).
course,
Endocrine Reviews, 2024, Vol. 45, No. 5
Another example of targeted, specific vitamin D dosing, of
, is in the pediatric setting. Infants and children have dif-
ferent upper tolerance limits compared to adults. To maintain
a desirable 25(OH)D concentration, the 2010 IOM guidelines
recommend 600 IU/d (15 µg) for children, adolescents, and
adults, and 400 IU/d (10 µg) for infants (47). ES guidelines
recommend 400 to 1000 IU/day (10-25 µg) for infants aged
up to 1 year and 600 to 1000 IU/day (15-25 µg) for children
older than 1 year to treat and prevent vitamin D deficiency
(46). These values are consistent with several guidelines issued
by other societies in the past several years. Of course, they can
be increased if a laboratory-confirmed vitamin D deficiency is
being treated (186).
Many studies investigated dosing regimens in pediatric pa-
tients. One trial comparing 4 different daily dosages (400,
800, 1200, 1600 IU) found that all dosages established
25(OH)D concentrations of 50 nmol/L or greater in 97% to
98% of infants at age 3 and 12 months, but only a dosage
of 1600 IU/d 25(OH)D levels to 75 nmol/L or greater in
97.5% of infants at age 3 months; nonetheless, this study
was discontinued prematurely because of elevated plasma
25(OH)D concentrations that have been associated with
hypercalcemia (187). Another study also found that 1600
IU/day given for 10 weeks to infants from 2 weeks to 3 months
of age maintained a 25(OH)D concentration above 80 nmol/L,
but without causing hypercalcemia or hypercalciuria (188).
Daily supplementation
From a physiological perspective, daily administration of
cholecalciferol seems to be most natural. Indeed, it appears
that a daily approach results in higher efficacy in terms of
25(OH)D exposure and extraskeletal benefits.
In a recent RCT comparing 3 different dosing regimens in
vitamin D-deficient participants with similar total
end-of-study cumulative doses (D3 daily 10 000 IU 8 weeks,
then 1000 IU for 4 weeks; 50 000 IU weekly for 12 weeks;
and 100 000 IU every 2 weeks for 12 weeks), the group receiv-
ing the daily supplementation was the quickest to reach suffi-
ciency (<2 weeks, although receiving a higher cumulative dose
in the first 8 weeks when compared to the other 2 arms) and
reached the highest serum 25(OH)D levels (172).
Importantly, daily administration was associated with higher
systemic exposure to 25(OH)D (greater area under the curve,
+23% and +27% compared to weekly and biweekly adminis-
tration, respectively), even when corrected for the cumulative
dose (172). The greater 25(OH)D exposure of daily regimens
could be due to lower activation of the 24-hydroxylase en-
zyme (CYP24A1). In an RCT of lactating women comparing
the effect of bolus (150 000 IU) vs daily vitamin D3 dosing
(5000 IU) on vitamin D3 catabolism, a single high-bolus
dose of vitamin D led to greater production of
24,25(OH)2D3, relative to the 25(OH)D3 value than did daily
vitamin D supplementation, with this effect persisting for at
least 28 days after supplementation (27). The greater thera-
peutic potential of daily regimens compared to other regimens
might be less relevant at lower doses (<2000 IU). Two studies
comparing 2000 IU/day vs 50 000 IU/month (189) and
800 IU/day vs 5600 IU/month (190) found no statistically sig-
nificant differences in the 2 areas under the curves.
Greater 25(OH)D exposure and lesser 24-hydroxylase ac-
tivity might be the rationale behind the potential extraskeletal
benefits of cholecalciferol supplementation. The already cited
Page 17
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Endocrine Reviews, 2024, Vol. 45, No. 5
metanalysis of 12 RCTs by Keum et al (141) found that the re-
duction in cancer mortality after vitamin D supplementation
was largely attributable to interventions with daily dosing
(as opposed to infrequent bolus dosing). Secondary analyses
of the VITAL trial giving 2000 IU/day of cholecalciferol found
a significant reduction in advanced cancers (metastatic or
fatal), especially among those with normal BMI (126).
However, the opposite was seen with monthly dosing in the
D-Health trial, where the risk of death from cancer was in-
creased (130). In the AMATERASU trial on patients with
digestive-tract cancers, 2000 IU/day of cholecalciferol pro-
vided a cumulative hazard ratio of relapse or death of 0.66,
significantly lower than the placebo group when adjusted by
age quartile (191). Regarding the prevention of autoimmune
diseases, supplementation with 2000 IU/day of cholecalciferol
for 5 years reduced autoimmune diseases by 22% in the
VITAL trial (129). Finally, a meta-analysis on the prevention
of acute respiratory infections after vitamin D supplementa-
tion found that vitamin D given daily had an odds ratio
(OR) of 0.78, compared to an OR of 0.97 to 0.98 if weekly
or bolus regimens (145). Protection was mainly associated
with administering daily doses of 400 to 1000 IU for up to
12 months and an age of 1.00 to 15.99 years at enrollment.
This result is particularly interesting as recommended preven-
tion strategies such as inactivated influenza vaccines in health
adults probably reduce acute respiratory infections from
21.5% to 18.1% with a relative risk of 0.84 (192).
In conclusion, daily cholecalciferol administration might be
the most efficient and beneficial strategy to increase serum
25(OH)D, at least from the biomedical (but not necessarily
bio-psycho-social) perspective. Indeed, most RCT data sug-
gesting extraskeletal benefits of cholecalciferol supplementa-
tion come from studies with daily dosing. Future studies
should investigate this observation in pathologic conditions
(ie, obesity).
Nondaily supplementation
Intermittent vitamin D dosing usually uses a greater amount to
reach equivalent doses with fewer administrations. The ration-
ale of this approach is to enhance adherence and ease manage-
ment of specific patient groups, such as children and
community-dwelling older people (193). Indeed, low adher-
ence to vitamin D prescription has often been reported, al-
though the topic is controversial. For example, Albrecht et al
(194) recently investigated adherence to bone health-promot-
ing lifestyle recommendations concerning osteoporosis status
in a cross-sectional database of community-dwelling older
adults (aged 65-75 years). In high-risk osteoporosis patients,
adherence to vitamin D intake, defined as regular consumption
of vitamin D-rich foods and/or vitamin D supplements, was
high, ranging from 85% (women) to 93% (men). In contrast,
in a cross-sectional study of pediatric outpatients affected by
various diseases, Arshad et al (195) found that adherence to
vitamin D prescription was quite low, particularly in those
with diseases where vitamin D deficiency presents as a high-
risk condition.
For these reasons, recurrent and protracted intervals of vita-
min D supplementation appear to be an effective and conveni-
ent way to achieve and maintain sufficient vitamin D status
and to increase patients' adherence, but there is no agreement
that treatment simplification with intermittent dosing signifi-
cantly improves compliance (196) and there is consistent
evidence to discourage the use of “megadoses” due to the
sible side effects (197).
Weekly and monthly regimens
641
pos-
With equivalent doses and large formulations, daily, weekly,
and monthly supplementation may lead to similar increases
and levels of 25(OH)D in middle-aged (198) and obese indi-
viduals (199), in older individuals with hip fractures (200),
and children with CKD (201). However, one study concluded
that a daily regimen was more efficient in circulating 25(OH)
D than weekly or monthly administration, but with different
formulations (179). As compared with a daily regimen, a bo-
lus dose is associated with a higher 24,25(OH)2D level and a
higher 24,25(OH)2D to 25(OH)D ratio (27). In a monocen-
tric, open-label randomized study in postmenopausal women,
weekly vitamin D was more efficient than monthly in improv-
ing muscular function (measured through the Sit-to-Stand and
Timed-Up-and-Go tests) (202). Monthly regimens have been
tested in several large trials with multiple outcomes.
Compared to a placebo, 100 000 IU monthly did not influence
the risk of CVD, falls, fracture, or cancer, and lung or arterial
functions in vitamin D-replete individuals (203). In those par-
ticipants with baseline 25(OH)D lower than 50 nmol/L, the
100 000 IU vitamin D regimen increased lumbar spine BMD
by 2.6% and improved lung and arterial functions (203). In
the D-Health trial including more than 21 000 individuals,
with 24% of them having a 25(OH)D level less than
50 nmol/L, 60 000 IU monthly did not influence all-cause
mortality (126) but was associated with a higher risk of falls
in those with a BMI of less than 25 (64). This observation
was in agreement with another trial in which a higher percent-
age of fallers was detected with 60 000 IU/month compared to
24 000 IU/month over 1 year (204). In small trials, few epi-
sodes of hypercalcemia were reported with weekly doses be-
tween 20 000 and 100 000 IU in various target populations
(205). Overall, trials with weekly or monthly vitamin D sup-
plementation regimens did not show significant effects on clin-
ical variables. This could be due to the recruited population
(vitamin D-replete or obese individuals) or too large
vitamin D doses leading to a U-shape dose-response relation-
ship. Currently, there is no evidence of a superiority in the
benefit/risk ratio of weekly or monthly vitamin D regimens
over daily supplementation.
Longer intervals
Although one study using high doses with prolonged intervals
(100 000 IU every 4 months) administered to community-
dwelling adults older than 50 years found a reduction in
fractures (206), other similar studies (500 000 IU every year
(193)/150 000 IU every 3 months (207)) did not show a reduc-
tion in hip/vertebral/nonvertebral/total fracture incidence.
This was also evidenced by the systematic review and meta-
analysis of Zhao et al (208). In studies on the efficacy of
vitamin D administration, the basal values of 25(OH)D are
often either not measured (206) or are at normal/high levels
(206), making it difficult to understand the real effect of sup-
plementation on 25(OH)D values. In a subgroup analysis of
Zhao's study, no differences in fracture incidence were found
between intermittent high doses given once every year and
other interval regimens (208). In Zhao's meta-analysis, refer-
ence is made to the study by Witham and colleagues (209) in
which no negative effects of longer intervals of high-dose
Page 18
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
642
vitamin D administration on blood pressure in older patients
with isolated systolic hypertension were reported.
Regarding the relation between long-term intervals of vita-
min D administration and CVD risk, falls, and fracture out-
comes in older and community-dwelling people, in a
systematic review with meta-analysis, Barbarawi et al (142)
did not find significant results favoring vitamin D intervention
(100 000 IU every 4 months (189)/500000 IU yearly (197)) in
preventing falls, fractures, or CVDs. Even in works cited in
this meta-analysis, the basal 25(OH)D values were either
not reported or sufficient.
In a systematic review with meta-analysis, Yang et al (210)
cited 2 works that investigated the effect of intermittent high
doses of vitamin D as adjuvant treatment in pneumonia in
children (100 000 IU every 3 months (211) and 300 000 IU
quarterly for 1 year (212)) on the incidence rate of repeated
episodes of pneumonia, rate of intensive care unit (ICU) hos-
pital admission, and complications rate. In both cases, no sig-
nificant definitively positive effects were found. Regarding the
safety of longer-interval vitamin D supplementation, in a re-
cent systematic review with meta-analysis on children,
Brustad et al (213) did not find any association with severe
side effects. This was also seen in other studies with protracted
intervals of vitamin D administration (211, 212, 214, 215).
Summary of Vitamin D Dosing Regimens
In conclusion, one of the major justifications for longer inter-
vals with high doses in vitamin D administration, namely, to
address low compliance with more frequent regimens, is con-
troversial. The rationale gains support in children and adoles-
cents rather than in older individuals. However, it has to be
taken into account that the cited meta-analyses underscored
the point that there is no evidence of efficacy in intermittent
high-dose and longer intervals of vitamin D administration
in reducing fracture rate, falls, CV events, or infectious dis-
eases. An increase in falls in older individuals has been ob-
served with large, intermittent dosing (197, 216) (the
literature regarding falls is somewhat controversial in part be-
cause there are no reliable methods to capture falls, as both
diaries and self-reports are flawed). These conclusions should
be tempered by inherent flaws in many reports in which the
baseline vitamin D dosage or pretreatment 25(OH)D levels
are not provided.
Routes of Administration
Oral supplementation of cholecalciferol is the most commonly
used approach. It is effective, simple, and generally safe.
Therefore, it is the preferred way to supplement vitamin
D. However, sometimes, the parenteral route may be a better
method for improving vitamin D status than oral administra-
tion of vitamin D, particularly in situations like intestinal mal-
absorption. Interestingly, a new transdermal route of vitamin
D administration is being proposed (217) but will not be dis-
cussed here due to the paucity of data.
Oral administration
Cholecalciferol (vitamin D3) and ergocalciferol (vitamin D₂)
are fat-soluble vitamins that are absorbed in the small intes-
tine. Because they are lipophilic compounds, their absorption
is similar to the absorption of lipids. Vitamin D is incorporated
into micelles with biliary salts on the micelle surface. On aver-
about 80% of vitamin D is absorbed, but the variation in
age,
Endocrine Reviews, 2024, Vol. 45, No. 5
absorption can be large (55%-99%) (218-220). Taking vita-
min D supplements with a fat-containing meal may improve
vitamin D absorption (218, 220). Cholecalciferol and ergocal-
ciferol are both rapidly absorbed, and the plasma levels peak
after about 24 hours of ingestion. Absorption into the entero-
cytes of the intestinal wall was thought to be a passive process,
but there is some evidence that vitamin D, especially in dietary
doses, is also actively transported through the enterocyte mem-
branes via cholesterol transporter proteins. However, passive
transport seems to occur with pharmacological doses of vita-
min D. From the enterocytes, vitamin D is exported in chylomi-
crons by the lymphatic route (218-220).
Bariatric surgery and intestinal malabsorption syndromes
that reduce fat absorption, such as inflammatory bowel dis-
eases, cystic fibrosis, and severe cholestasis, can also reduce
vitamin D absorption (10, 221). However, intestinal malab-
sorption does not seem to affect the absorption of calcidiol
as much, most likely because calcidiol is more water soluble,
thus not requiring bile salts for absorption, and because calci-
diol is absorbed by the portal route instead of the lymphatic
route (219).
As cholesterol transporters are involved in vitamin D ab-
sorption, factors that interfere with cholesterol absorption
could also affect vitamin D absorption. However, ezetimibe,
an inhibitor of cholesterol transport, does not seem to
affect vitamin D absorption despite the reduction in choles-
terol absorption. There is also no strong evidence that
phytosterols, plant sterols used to inhibit cholesterol absorp-
tion, impair vitamin D absorption. In contrast, there is some
evidence that drugs used to reduce intestinal fat absorption,
such as orlistat and olestra, may also reduce vitamin D
absorption (219).
Vitamin D supplements are available in different vehicles,
such as oil-containing gel capsules, oily drops, and hard pow-
der tablets. Although it could be hypothesized that vitamin D
would be better absorbed from oil-based vehicles, no convin-
cing evidence supports this premise. In fact, there is some evi-
dence that vitamin D may be better absorbed from a
powder-based vehicle than from an oil-based vehicle in cases
of intestinal fat malabsorption, such as in cystic fibrosis (222).
Parenteral administration
The optimal treatment of hypovitaminosis D in the general
population and disease states is still debated (7). Parenteral
administration of intermittent vitamin D boluses may be indi-
cated in patients with hypovitaminosis D who are not suitable
for oral intake or with intestinal malabsorptive diseases, in-
cluding inflammatory bowel disease, celiac disease, pancreatic
insufficiency, short-bowel syndrome, and post bariatric sur-
gery (10, 221). Based on advantageous pharmacokinetic prop-
erties and evidence-based clinical data, intramuscular
cholecalciferol may be the preferred form of vitamin D to be
used in these clinical settings. In fact, it has been shown that
cholecalciferol was able to reach higher serum 25(OH)D lev-
els more rapidly than ergocalciferol when both vitamin D
forms were administered as a single large intramuscular
dose (300 000 UI) in adult or older patients with hypovitami-
nosis D (223-225). Moreover, in the study by Romagnoli
et al (224), 2 months after administration of this large, intra-
muscular cholecalciferol dose, serum 25(OH)D levels were
higher than those obtained after the same oral dose.
Therefore, intermittent intramuscular cholecalciferol could
Page 19
Endocrine Reviews, 2024, Vol. 45, No. 5
Table 4. Characteristics of different forms/metabolites of vitamin D and when to use them
Vitamin D form
Holl...
H
643
Circulating
half-life
Around 1 day
(longer functional
half-life
in correlation
with its
slow release
from the
adipose tissue)
Features
Native form of human- and animal-produced vitamin D
Lipophilic, stored in fat and released on need
• Useful in clinical practice as it renders possible
intermittent administration regimes
• Wide safety range thanks to the predicted mechanisms
regulating its hydroxylation
When to use it
Most clinical situations where a vitamin D
deficiency must be addressed (see
below for exceptions)
Cholecalciferol
(vitamin D3)
Around 2 days
HO""
Ergocalciferol
(vitamin D2)
H
OH
HO"
Calcifediol
(25(OH)D)
OH
OH
H
Calcitriol
(1,25(OH)2D)
2-3 weeks
5-8 hours
Inferior to cholecalciferol in increasing 25(OH)D
serum levels
•Risk of over- or under-estimation of total 25(OH)D in
the presence of substantial amounts of 25(OH)D2
with subsequent risk of vitamin D toxicity in case of
dose increments
• Widely prescribed in the USA in high doses (50,000 IU)
•High doses alter vitamin D metabolism, increasing 24
hydroxylase activity
Hydrophilic, thus higher solubility in organic solvents,
less sequestration in adipose tissue, smaller
distribution volume and shorter half-life compared
to cholecalciferol
•⚫ Fast increase in 25(OH)D serum levels along with PTH
suppression
• Easier to manage than cholecalciferol in case of toxicity
• More efficient internalization in cells expressing the
megalin-cubilin system
•Promotes active intestinal calcium absorption and
suppresses PTH secretion
• Increases the activity of the CYP24A1, which stimulates
the degradation of 25(OH)D
•Risk of hypercalcemia and hypercalciuria
• To be administered daily or in lower doses distributed
over a 24-hour period
Only in cases of patients' ethical concern
(e.g., vegetarianism, veganism, others).
However, vegan D3 supplements (made
from lichen) are available and should
be considered instead of D2
supplements
Malabsorption syndromes, obesity,
CYP2R1 dysfunction, or in situations
in which a quick attainment of vitamin
D sufficiency is desirable
As a hormone replacement for patients
with limited/absent renal tubular
1-α-hydroxylase activity; vitamin D
resistant rickets type 1, X-linked
hypophosphataemic rickets, chronic
hypoparathyroidism, as an alternative
to the use of the native missing
hormone PTH, and moderate-to-severe
kidney failure
Consider replacing it with analogs with
less calcemic activity (maxacalcitol;
falecalcitriol; paricalcitol; doxercalciferol)
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
be useful in clinical conditions when rapid correction of hypo-
vitaminosis D is unnecessary and for long-term maintenance
of adequate serum vitamin D levels, as in some older patients,
to improve their adherence to vitamin D supplementation.
However, safety concerns limit the clinical use of intermittent,
excessive vitamin D doses. In fact, large intramuscular boluses
(300 000 IU) induce unwanted effects such as an increase in
falls and fracture events or enhance bone turnover (226, 227).
There is a consensus to administer vitamin D boluses not
higher than 100 000 IU (228). In conclusion, the therapeutic
regimen to recover from vitamin D deficiency should be tail-
ored to patients' characteristics, such as age, BMI, severity
of vitamin D deficiency, concurrent comorbidity, and use of
other drugs.
Different Forms of Vitamin D Supplementation
The main supplemental oral forms of vitamin D are cholecal-
ciferol (vitamin D3) and ergocalciferol (vitamin D2). Both are
readily available without a prescription. Cholecalciferol is the
most used form of supplemental vitamin D. Calcidiol (calcife-
diol, 25(OH)D), the inactive vitamin D metabolite produced
in the liver, and other vitamin D analogues, such as calcitriol
(1,25(OH)2D, the physiologically active form of vitamin D)
and alfacalcidol (1-hydroxyvitamin D), are used as prescrip-
tion medicines in some conditions (Table 4).
Ergocalciferol
Ergocalciferol does exist in nature (mainly in plants and fungi),
and low circulating levels of 25(OH)D2 are present in free-
ranging nonhuman primates and human population studies
(229, 230). The 2 forms of vitamin D, cholecalciferol (D3) and
ergocalciferol (D2), are often used interchangeably as supple-
mentation or treatment of vitamin D deficiency as, historically,
vitamins D2 and D3 were considered equally effective in treating
rickets (231). Similarly, previous recommendations consider vi-
tamins D2 and D3 interchangeable (46). Subsequently, however,
Page 20
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
644
multiple studies and meta-analyses comparing the effect of D2
and D3 on circulating 25(OH)D concentration have found
cholecalciferol to be superior (223-225, 232).
Challenges to 25(OH)D measurement are widely recog-
nized. The presence of 2 circulating 25(OH)D forms,
25(OH)D3 and 25(OH)D2, adds additional challenges, not-
ably for automated immunoassays. Importantly, it is possible
that the antibodies used in immunoassays may not detect
25(OH)D2 and 25(OH)D3 equally, and the proprietary ap-
proach to releasing 25(OH)D from DBP may not liberate the
2 forms equally (233). As such, it is perhaps unsurprising
that multiple reports find overestimation or underestimation
of total 25(OH)D in the presence of substantial amounts of
25(OH)D2 (233-236). This problem was corroborated by a re-
cent interlaboratory comparison performed at the behest of
the VDSP (87). Assay underestimation of total 25(OH)D in
the presence of substantial amounts of 25(OH)D2 can have
clinical consequences. A recent small report of patients receiv-
ing 50 000 IU of ergocalciferol every 2 weeks found 40%
(6/15) to have total 25(OH)D levels less than 30 ng/mL
when measured by immunoassay whereas all had values above
42 ng/mL when measured by LC-MS/MS (237). It is easy to im-
agine that such patients with “low” 25(OH)D values would
have their dose increased, with at least potential toxicity, or
undergo additional evaluation, such as for malabsorption.
Thus, these assay issues are not clinically inconsequential.
Issues surrounding ergocalciferol use are of note for the
United States, where 50 000 IU of vitamin D2 was the only
high-dose preparation available by physician prescription
and, therefore, ergocalciferol was widely prescribed. Now, in-
stead, 50 000 IU of vitamin D3 is available by prescription. In
addition to assay issues, widespread use of intermittent high-
dose ergocalciferol ("bolus" therapy) appears to alter vitamin
D metabolism, with increased 24-hydroxylase activity (27).
Other adverse consequences of high-dose vitamin D therapy,
notably increased fall risk (197), are reported and have led
to calls to critically assess daily vs bolus vitamin D therapy
(238).
To summarize, vitamins D2 and D3 are not equivalent in
raising circulating 25(OH)D, and bolus dosing may have ad-
verse effects on vitamin D metabolism and clinical outcomes.
As such, it is to be expected that calls for the use of only
cholecalciferol and avoidance of ergocalciferol have been
and continue to be published (233, 239, 240) with recent
osteoporosis-treatment guidance advising cholecalciferol
over ergocalciferol (241) (see Table 4). Despite these recom-
mendations, high-dose ergocalciferol remains widely pre-
scribed in the United States.
Calcifediol
Calcifediol is the intermediate metabolite between cholecalcif-
erol and calcitriol. Several pharmacokinetic studies performed
in the last 4 decades have demonstrated its hydrophilic prop-
erties, leading to higher solubility in organic solvents, less se-
questration in adipose tissue, smaller distribution volume, and
shorter half-life when compared to cholecalciferol (242-244).
By virtue of its hydrophilic properties, calcifediol is readily ab-
sorbed via the venous portal system and thus quickly increases
circulating concentrations of 25(OH)D 3. In contrast to chole-
calciferol, which is mostly stored in fat tissue, 25(OH)D tends
to be more evenly distributed throughout the body (20% in
muscle, 30% in circulation, 35% in fat, and 15% elsewhere)
Endocrine Reviews, 2024, Vol. 45, No. 5
(245). The administered dose will generally lead to predictable
25(OH)D levels and effective PTH suppression. In cases of
toxicity, this form of vitamin D is easier to manage than chole-
calciferol (244). Moreover, the greater affinity of calcifediol
for DBP allows for more efficient internalization in cells ex-
pressing the megalin-cubilin system of endocytic receptors,
such as the parathyroids and the renal tissue (246).
Such properties provide the rationale for using calcifediol in
specific clinical conditions. The clinical situations that make
use of calcifediol attractive are obesity, hepatic failure, pa-
tients with inactivating mutations of genes encoding
CYP2R1 (the principal enzyme that is responsible for vitamin
D 25-hydroxylation), or patients taking drugs that could in-
fluence the activity of cytochrome enzymes (ie, antiretroviral
or antitubercular). Calcifediol was shown to have the same
bioavailability in healthy adults with differing BMI and adults
with intestinal malabsorption compared to controls (247). In
an RCT on vitamin D-deficient, postmenopausal women,
weekly calcifediol was found to be more effective and faster
acting compared to cholecalciferol in increasing 25(OH)D se-
rum levels. This more favorable kinetics led to greater im-
provement in muscle function (202). In another RCT in 35
healthy women aged 50 to 70 years, calcifediol given daily,
weekly, or as a single bolus was about 2 to 3 times more potent
in increasing plasma 25(OH)D3 concentrations than cholecal-
ciferol (190).
New extended-release calcifediol formulations are more ef-
fective than cholecalciferol in raising serum 25(OH)D levels
even in overweight nondialytic CKD patients with secondary
hyperparathyroidism (248); nonetheless, it must be noted that
these data arise from observational, retrospective data and
subgroup post hoc analysis of RCTs.
Recently, retrospective studies have highlighted the role of
calcifediol administration on various end points related to
COVID-19 infection. To demonstrate a causative effect,
Nogues et al (249) investigated 2 cohorts of patients with
COVID-19, 1 of whom was untreated and 1 assigned to the
oral calcifediol group. The treatment regimen consisted of
oral calcifediol (0.532 mg the day of admission), followed
by doses of 0.266 mg on days 3, 7, 15, and 30. Out of 447 pa-
tients treated with calcifediol at admission, 20 (4.5%) re-
quired the ICU, and 21 (4.7%) died; this was significantly
lower compared to the untreated group of 391 patients, of
whom 82 (21%) required the ICU and 62 (15.9%) died
(both P<.01). Adjusted logistic regression of calcifediol treat-
ment on ICU admission indicates that patients treated with
calcifediol had a lower risk of ICU admission (OR 0.02;
95% CI, 0.07-0.23) and mortality (OR 0.21; 95% CI,
0.10-0.43), suggesting an effectiveness of calcifediol treatment
(249).
In summary, calcifediol seems to represent a form of vita-
min D that is useful for replenishing vitamin D status. Most
attractive clinical settings include malabsorption syndromes,
obesity, CYP2R1 dysfunction, or situations in which
quick attainment of vitamin D sufficiency is desirable (see
Table 4).
Calcitriol
Calcitriol is the active hormonal form of vitamin D and the nat-
ural VDR ligand. It promotes active intestinal calcium absorp-
tion and suppresses PTH secretion. Calcitriol has a short
half-life of around 5 to 8 hours; therefore, it should be
Page 21
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Endocrine Reviews, 2024, Vol. 45, No. 5
administered daily (or with intermittent regimens) and some-
times in lower doses distributed over a 24-hour period (240,
250). As calcitriol is not an organic micronutrient, its use in clin-
ical practice requires careful monitoring. Calcitriol increases the
activity of CYP24A1, which stimulates the degradation of
25(OH)D. This results in serum 25(OH)D not being useful as
a marker of adequate vitamin D supplementation and reduced
potential benefits of physiological extrarenal/local production
of calcitriol due to reduced substrate availability. Moreover,
some studies have reported a more significant incidence of ad-
verse events such as hypercalcemia and hypercalciuria. Thus,
there is a need to monitor serum and urine calcium and phos-
phate (240, 251, 252). Because of these safety and clinical prac-
ticality issues, there is consensus that calcitriol use should be
limited to hormone replacement for patients with limited/absent
renal tubular 1-a-hydroxylase activity, as their capacity to pro-
duce calcitriol is severely limited (240, 251, 252). Indeed, calci-
triol was first used to treat patients with vitamin D-resistant
rickets type 1 (23). Other indications are X-linked hypophospha-
temic rickets, chronic hypoparathyroidism, as an alternative to
the use of the native missing hormone PTH, and moderate-to-
severe kidney failure when calcitriol production is impaired
or to suppress excessive PTH secretion. This use helps to
control secondary hyperparathyroidism and resultant meta-
bolic bone diseases. However, as calcitriol use is associated
with frequent hypercalcemia, its use could be replaced by ana-
logues with less calcemic activity approved for use in patients
with secondary hyperparathyroidism in renal failure, in par-
ticular maxacalcitol (22-oxa-1,25(OH)2D3) and falecalcitriol
(1,25(OH)2-26,27-F6-D3), which are currently available in
Japan, and paricalcitol (19-nor-1,25(OH)2D2) and doxercalci-
ferol (1a(OH)D2), available in the United States (253, 254).
Calcitriol has also been proposed for the treatment of osteopor-
osis, but it is not approved in this setting (240, 251, 252).
In conclusion, calcitriol is not suitable for supplementation
or nutritional fortification, and none of many excellent
reviews, guidelines, and policy papers consider the use
of calcitriol in the nutritional context (supplementation
and fortification). However, guidelines suggest that vitamin
D supplementation is advised in patients with chronic
hypoparathyroidism, chronic kidney failure, and low
vitamin D status in addition to receiving therapeutic doses
of calcitriol (see Table 4). Such a recommendation is moti-
vated by the activity of extrarenal 1-a-hydroxylase, which is
compromised by reduced renal function (ie, not regulated
by PTH) and is not regulated by feedback mechanisms (240,
251, 252).
Vitamin D Safety and Monitoring
Vitamin D supplementation is generally a safe treatment with
minimal adverse events and no need for strict monitoring.
However, side effects of vitamin D treatment exist and can re-
sult in vitamin D toxicity (VDT).
Vitamin D toxicity
VDT is a clinical condition characterized by excess vitamin D
(hypervitaminosis D), resulting in severe hypercalcemia that
may persist for a prolonged period of time, leading to serious
health consequences. Signs and symptoms of VDT are related
primarily to hypercalcemia, with complications encompassing
adverse events in the CV, renal, gastrointestinal, neurological,
and musculoskeletal systems (255, 256). VDT prevalence is
645
unknown, but it is rare due to the wide therapeutic index of
vitamin D (255, 256). Evidence from systematic studies of
VDT in humans is missing for ethical reasons, and data mostly
stem from studies of VDT in animals and anecdotal reports.
The condition of infantile hypercalcemia was first described
in the United Kingdom and Switzerland, showing symptoms
such as failure to thrive, osteosclerosis, developmental delay,
and even death, but was not immediately associated with vita-
min D intake. Suggestions were made that excess vitamin D in-
take may be a causative factor (children received up to
35 000 IU daily). Eventually, the British Ministry of Health
suggested a substantial reduction in vitamin D allowance, re-
sulting in a marked decrease in infantile hypercalcemia cases
(257-260). As the prescriptions of vitamin D products are in-
creasing worldwide, so is the number of VDT reports, with
more than 75% published since 2010. Many of these cases re-
sult from inappropriate prescribing; moreover, the prescrip-
tion of high-dose unlicensed and poorly manufactured
treatments can be greater than 60%, as they are cheaper
(261, 262).
In healthy individuals, hypervitaminosis D is usually de-
fined as "exogenous" as it develops after uncontrolled use of
megadoses of vitamin D or its metabolites or analogues, as
in case of high dose of calcifediol leading to a faster increase
in 25(OH)D serum levels compared with cholecalciferol but
easier to manage than cholecalciferol in case of toxicity for
its hydrophilicity and lesser sequestration in adipose tissue.
On the other hand, excessive production of calcitriol in
granulomatous disorders, lymphomas, primary hyperpara-
thyroidism, and idiopathic infantile hypercalcemia results in
"endogenous" hypervitaminosis D (255, 256).
VDT is defined by a biochemical phenotype with
markedly elevated calcifediol concentrations (>150 ng/mL
or >375 nmol/L), along with dihydroxylated metabolites
(24,25(OH)2D3; 25,26(OH)2D3, 25(OH)D3-26,23-lactone),
unless the causal agents are vitamin D analogues, such as par-
icalcitol. Calcitriol levels may be in the normal reference range
or even reduced in exogenous VDT while elevated in endogen-
ous VDT. PTH levels can be very low or undetectable (263).
VDT thus results in severe hypercalcemia, hypercalciuria,
and hyperphosphatemia. Pathogenetically, hypercalcemia is
a consequence of high calcifediol levels in exogenous VDT
(with calcifediol at pharmacological concentrations overcom-
ing VDR affinity disadvantages and/or displacing 1,25(OH)
D2 from DBP (264)), while high calcitriol levels cause en-
dogenous VDT. Exogenous factors that interact with VDT
risk include dosage, calcium in the diet or as a supplement,
vitamin D intake with the diet, social status (ie, neglected pa-
tients), artificial UV light treatment sessions, quantity of sup-
plement use, and time of exposure. Endogenous risk factors
comprise age, sex, vitamin D status, hypersensitivity syn-
dromes, and the pharmacogenetics of the vitamin D response
and metabolism (253, 254, 263). This is why there is no clear
cutoff above which VDT occurs and below which it does not.
In conclusion, VDT is a rare but life-threatening event most-
ly caused by unintentional overdosing due to pharmaceutical
products. The prescriber and dispenser should avoid un-
licensed vitamin D products. VDT should always be consid-
ered a differential diagnosis when evaluating patients with
hypercalcemia. Future studies should encompass the evalu-
ation of concurrent conditions that increase the risk of VDT
and include the evaluation of classic and nonclassic adverse
events for VDT.
Page 22
646
Endocrine Reviews, 2024, Vol. 45, No. 5
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Monitoring vitamin D status during treatment
Monitoring treatments is important to assess their efficacy
and safety. Regarding vitamin D supplementations, there is
limited evidence for when to monitor response to therapy or
toxicity.
When it comes to achieving sufficiency in deficient patients, it
seems there is no need to monitor differently according to differ-
ent dosage regimens (dose and/or frequency) or baseline 25(OH)
D serum values. The increase in serum 25(OH)D concentration
after supplementation follows a curvilinear response with the in-
crease of the cumulative doses (265, 266). The delta increase
over 100 IU depends on baseline levels, and there is less increase
100 IU with high doses than low doses (267). Van Groningen
per
et al (268) calculated that the cholecalciferol loading dose re-
quired to reach the serum 25(OH)D target level of 75 nmol/L
can be calculated as dose (IU) = 40 × [75 - serum 25(OH)
D] body weight. Mean 25(OH)D levels over a 2-month period
are similar to daily, weekly, or monthly administrations (al-
though monthly dosing is associated with more variability),
and sufficiency can be reached independently from the baseline
25(OH)D values (200). In the study by Fassio et al (172), all par-
ticipants normalized 25(OH)D safely, regardless of dosing regi-
mens and including patients receiving 10 000 IU/day for the first
8 weeks; moreover, no cases of hypercalcemia were recorded.
With regard to recent megatrials results, no effects were found
on serum calcium or calciuria unless very high doses were
used, such as 4000 to 10 000 IU per day in the Calgary study
(104). Furthermore, these studies did not confirm the modestly
increased risk of kidney stones observed in the WHI trial
(400 IU per day) (117). However, there might be a need for mon-
itoring in case of other vitamin D metabolite use. As discussed
earlier, calcifediol acts much more rapidly than cholecalciferol
in increasing serum 25(OH)D levels, resulting in greater fluctu-
ation of 25(OH)D levels. For example, supplementation with
20 µg (800 IU) of cholecalciferol (vitamin D3) increases
25(OH)D concentrations toward 70 nmol/L (28 ng/mL) within
16 weeks, while supplementation with 10 or 15 µg calcifediol
(25(OH)D) increases 25(OH)D levels more than 75 nmol/L
(>30 ng/mL) in 8 and 4 weeks, respectively (269).
To summarize, cholecalciferol can maintain physiological
25(OH)D serum levels above 30 ng/mL (75 nmol/L) but be-
low 50 ng/mL (125 nmol/L) for a long time, regardless of
whether the dosage given is daily or intermittent (weekly, fort-
nightly, or monthly), due to its slow pharmacokinetic elimin-
ation caused by prolonged storage and release on demand
according to physiological needs (270). Routine monitoring
of 25(OH)D levels is generally unnecessary for patients on
long-term maintenance vitamin D doses of up to at least
2000 IU/day. Retesting after 8 to 12 weeks from the start of
supplementation may be appropriate when poor compliance
is suspected, in case of symptoms suggestive of vitamin D de-
ficiency, and for patients at risk of persistent 25(OH)D level
below 30 ng/mL (75 nmol/L). These comprise institutional-
ized or hospitalized individuals, people in whom vitamin D
therapy uncovers subclinical primary hyperparathyroidism,
obese individuals, individuals undergoing bariatric surgery,
individuals who use of certain concomitant medications (eg,
anticonvulsant medications, glucocorticoids), and patients
with malabsorption, including inflammatory bowel disease
and celiac disease. For patients on potent antiresorptive agents
(eg, denosumab or zoledronic acid), vitamin D levels should
be checked annually per protocol (71).
Conclusions
The metabolism, mechanisms of action, and pathophysiology
of vitamin D and its multifaceted implications in human
health have been extensively investigated for more than a cen-
tury. However, the role of vitamin D status assessment and the
detailed outcomes of vitamin D deficiency and its supplemen-
tation are still not completely understood. Thus, we extensive-
ly reviewed the literature on controversial vitamin D topics
to better clarify and summarize the "whys, whens, and
hows" of vitamin D assessment and supplementation in gener-
ally healthy populations and clinical conditions.
genes.
Vitamin D metabolism involves a different extensive panel of
enzymes, resulting in various hormonal metabolites. Moreover,
the VDR has been demonstrated to act as a key role transcrip-
tion factor in most cells and can regulate a plethora of
New insights into the regulation of vitamin D-related enzymes
and the differential mechanism of action of VDR have demon-
strated potential links between several metabolic disorders and
vitamin D effects. In this perspective, assessing a distinctive pat-
tern of noncanonical vitamin D metabolites may allow us to bet-
ter characterize different pathological conditions related to
vitamin D metabolism that do not depend only on reduced solar
exposure or vitamin D diet intake.
Besides the potential utility of the evaluation of noncanonical
vitamin D metabolites, 25(OH)D is nowadays the most widely
accepted biomarker to evaluate vitamin D status; however, its
optimal levels are still debated. Recommendations on optimal
25(OH)D levels deriving from international societies and guide-
lines can differ due to the different approaches used, including
clinical perspectives (level of cutoff at which no individual has
an undesirable outcome) or public health perspectives (level of
cutoff at which 97.5% of individuals do not have an undesirable
outcome). Another critical issue is the lack of an accepted labora-
tory test assay standardization, and this prevents a proper inter-
pretation of data reported by different studies, resulting in
enabling rational data pooling and implementation of meta-
analyses focused on vitamin D influence in various clinical out-
comes of interest. Thus, 25(OH)D laboratory assays should be
monitored in their performance through external quality assess-
ment plans providing target reference values from standardized
measurement procedures.
Vitamin D deficiency has been extensively related to the oc-
currence of skeletal disorders, such as rickets and osteomal-
acia. It can also be negatively implicated in osteopenia and
osteoporosis, which must be mandatory and managed with
vitamin D supplements. More recently, the interest in the
putative extraskeletal effects of vitamin D have resulted in
several clinical trials addressing vitamin D's influence on
cancer and CV risk, respiratory effects, autoimmune diseases,
diabetes, and mortality. The null results of some of these
RCTs especially the megatrials-hampered the enthusiasm
around these topics. However, these trials were progressively
revised, and their null results were mainly related to the enroll-
ment of vitamin D-replete adults in whom benefit would be
unlikely and the inhomogeneous methodologies in vitamin
D supplementation with different forms, metabolites, and
doses. Indeed, subsequent secondary analyses have progres-
sively shown that vitamin D might be useful in reducing cancer
incidence and mortality in the long term, in reducing auto-
immune diseases and CV events (in particular central arterial
hypertension, myocardial infarction, and atrial fibrillation)
occurrence, and the development of diabetes from prediabetes
Page 23
Downloaded from https://academic.oup.com/edrv/article/45/5/625/7659127 by guest on 02 April 2025
Endocrine Reviews, 2024, Vol. 45, No. 5
forms. Nonetheless, these RCTs and the following meta-
analyses were not powerful enough to evaluate these crucial
subgroups, and further studies with better methodological
conductions are warranted.
Regarding the different forms and metabolites used for vita-
min D supplementation, oral administration is the preferred
route, and parenteral administration should be reserved for spe-
cial clinical situations, such as in patients with severe gastrointes-
tinal malabsorption syndromes or after bariatric surgery.
Cholecalciferol remains the preferred choice, and it is generally
safe, requiring less strict monitoring. Ergocalciferol has been
demonstrated to be less effective in raising 25(OH)D serum lev-
els and, thus, should be reserved for specific clinical conditions.
Calcifediol could be recommended in patients with obesity, mal-
absorption syndromes, CYP2R1 dysfunction, or in situations in
which a quick, rapid achievement of vitamin D sufficiency is de-
sirable. Calcitriol use should be limited for patients with limited/
absent renal tubular 1-α-hydroxylase activity and in vitamin D—
resistant rickets type 1, X-linked hypophosphatemic rickets, and
chronic hypoparathyroidism.
Growing preclinical and clinical observations associating
vitamin D with many health clinical conditions have been pro-
gressively reported in recent years. However, the lack of rigor-
ous methodologies on patient enrollment, vitamin D
supplements, and standardized laboratory assays have limited
the ability to draw definitive conclusions about these data that
still need to be more clearly understood.
Thus, a "whys, whens, and hows" of vitamin D assessment
and supplementation derived from an international expert
panel discussion about controversial topics regarding vitamin
D metabolism, assessment, actions, and supplementation is
needed to help the scientific community in evaluating and con-
ducting future further studies with more rigorous methodolo-
gies, to better explore any clinical setting potentially
influenced by vitamin D, and to provide reliable data required
to update our international recommendations.
Acknowledgments
The authors wish to acknowledge Fabio Perversi (Polistudium
srl, Milan, Italy) for medical writing and Aashni Shah
(Polistudium srl, Milan, Italy) for linguistic and editorial
assistance.
Funding
This work was supported, in part, by the International
Vitamin D Expert Association (IDEA). The conference and
editorial assistance were supported by an unrestricted educa-
tional grant from Abiogen Pharma, Pisa, Italy. The sponsors
had no role in the selection of discussion topics, speakers, or
authors, preparation, or review of this paper. Work partially
supported by Glucocorticoid induced Osteoporosis Skeletal
Endocrinology Group (GIOSEG).
Disclosures
647
A.G. is a consultant for Abiogen and Takeda and received re-
search grant to institution from Takeda. J.P.B. is a consultant
for Abiogen. Travel and accommodation expenses for the
6th International Workshop on Controversies in Vitamin D
for R.B., C.M., and G.H.F. were covered by Abiogen Pharma
S.p.A. P.E. has received research funding from Amgen,
Sanofi, and Alexion, and honoraria from Amgen. A.F. has re-
ceived advisory board honoraria, consultancy fees, and/or
speaker fees from Boehringer Ingelheim, UCB Pharma, and
Abiogen. S.F. has received research grants to institution
from Abiogen Pharma. A.R.M. reports grants from Barts
Charity, The Fischer Family Foundation, DSM Nutritional
Products, The Exilarch's Foundation, The Karl R Pfleger
Foundation, The AIM Foundation, Thornton and Ross,
Warburtons, Hyphens Pharma, and Mr Matthew Isaacs (all
paid to his institution), and from UK National Institute for
Health Research Clinical Research Network and the HDR
UK BREATHE Hub (paid to Edinburgh University); consult-
ancy fees from DSM Nutritional Products; honoraria from
Oregon State University; payment for expert testimony from
Qiagen; and support for attending meetings from Abiogen
Pharma and Pharma Nord. A.R.M. is chair of the Data
Safety Monitoring Board for the VITALITY (vitamin D for
adolescents with HIV to reduce musculoskeletal morbidity
and immunopathology) trial, and a member of the data safety
monitoring board for a trial of vitamin D and zinc supplemen-
tation for improving treatment outcomes among COVID-19
patients in India; a program committee member for a vitamin
D workshop; and has received vitamin D capsules for clinical
trials from Pharma Nord, Synergy Biologics, and Cytoplan.
S.M. has served as speaker for Abiogen Pharma, Bruno
Farmaceutici, Diasorin, Geopharma, Sandoz, and UCB. He
also served on advisory boards for Abiogen, Eli Lilly, Kyowa
Kirin, Novo Nordisk, and UCB. R.R. has been a speaker or
participated on advisory boards for Abiogen, Naturex,
Nestlé, ObsEva, and Theramex. A.L.S. has received a research
grant from Amgen. N.N., M.P., C.T.S., F.M.U., J.V., G.J.,
P.L., R.A.A., G.B., D.D.B., N.C.B., J.B., M.L.B., F.F.C.,
L.d.F., and L.M.D. have nothing to declare.
| ### Consensus Statement on Vitamin D Status Assessment and Supplementation: Whys, Whens, and Hows
**Authors:**
Andrea Giustina, John P. Bilezikian, Robert A. Adler, Giuseppe Banfi, Daniel D. Bikle, Neil C. Binkley, Jens Bollerslev, Roger Bouillon, Maria Luisa Brandi, Felipe F. Casanueva, Luigi di Filippo, Lorenzo M. Donini, Peter R. Ebeling, Ghada El-Hajj Fuleihan, Angelo Fassio, Stefano Frara, Glenville Jones, Claudio Marcocci, Adrian R. Martineau, Salvatore Minisola, Nicola Napoli, Massimo Procopio, René Rizzoli, Anne L. Schafer, Christopher T. Sempos, Fabio Massimo Ulivieri, Jyrki K. Virtanen
**Abstract:**
The 6th International Conference, "Controversies in Vitamin D," was convened to discuss topics such as vitamin D metabolism, assessment, actions, and supplementation. Novel insights into vitamin D mechanisms of action suggest links with conditions that do not depend only on reduced solar exposure or diet intake and that can be detected with distinctive noncanonical vitamin D metabolites. Optimal 25-hydroxyvitamin D (25(OH)D) levels remain debated. Varying recommendations from different societies arise from evaluating different clinical or public health approaches. The lack of assay standardization also poses challenges in interpreting data from available studies, hindering rational data pooling and meta-analyses. Beyond the well-known skeletal features, interest in vitamin D's extraskeletal effects has led to clinical trials on cancer, cardiovascular risk, respiratory effects, autoimmune diseases, diabetes, and mortality. The initial negative results are likely due to enrollment of vitamin D-replete individuals. Subsequent post hoc analyses have suggested, nevertheless, potential benefits in reducing cancer incidence, autoimmune diseases, cardiovascular events, and diabetes. Oral administration of vitamin D is the preferred route. Parenteral administration is reserved for specific clinical situations. Cholecalciferol is favored due to safety and minimal monitoring requirements. Calcifediol may be used in certain conditions, while calcitriol should be limited to specific disorders in which the active metabolite is not readily produced in vivo. Further studies are needed to investigate vitamin D effects in relation to the different recommended 25(OH)D levels and the efficacy of the different supplementary formulations in achieving biochemical and clinical outcomes within the multifaceted skeletal and extraskeletal potential effects of vitamin D.
**Key Points:**
- **Total serum 25-hydroxyvitamin D concentration** is the accepted biomarker of vitamin D status, but assay methodology and standardization as well as desirable levels, which may vary according to the underlying condition, are still major issues.
- **Advances in knowledge about vitamin D** have included its metabolism, identification of noncanonical metabolites, mechanisms of action, and genetic polymorphisms. These insights have added to our understanding of vitamin D's role in nutrition and in disease.
- **Vitamin D deficiency** reduces intestinal calcium absorption leading to secondary hyperparathyroidism, bone loss, and increased risk of fractures in older adults. Meta-analyses of clinical trials show that vitamin D and calcium, together, decrease hip and other fractures in nursing home residents.
- **Post hoc analyses of recent mega trials** on extraskeletal effects of vitamin D suggest a link between vitamin D status and immune system and development of type 2 diabetes mellitus. Cardiovascular events and mortality may be positively affected as well.
- **Daily vitamin D regimens** seem to be the most efficient and beneficial strategy to improve vitamin D status but dosing schedules with longer intervals up to 4 weeks have been proposed to overcome low compliance with daily schedules.
- **Oral cholecalciferol (vitamin D3)** remains the preferred form of vitamin D for supplementation, while other vitamin D analogues (e.g., calcifediol, calcitriol, alfacalcidol) and parenteral administration should be used in specific conditions.
**Vitamin D Metabolism and Mechanism of Action:**
Vitamin D3 is produced in the skin from 7-dehydrocholesterol (7-DHC), while both vitamin D2 (ergocalciferol) and vitamin D3 (cholecalciferol) can be present in the diet. Vitamin D2 and D3 are hydroxylated first in the liver (and other tissues) to 25-hydroxyvitamin D (25(OH)D) and then in the kidney (and other tissues) to 1,25 dihydroxyvitamin D (1,25(OH)2D). Both 25(OH)D and 1,25(OH)2D are subsequently metabolized to their 24 (and for D3 23) hydroxy forms 24,25(OH)2D2/3, 23,25(OH)2D3, and 1,24,25(OH)3D2/3 (or 1,23,25(OH)3D3). Like other steroid hormones, vitamin D is highly lipophilic and bound to protein carriers that help maintain stable serum levels. The half-life of serum 25(OH)D is 2 to 3 weeks, and that of the more water-soluble 1,25(OH)2D is approximately 5 to 8 hours. The majority of circulating 25(OH)D, including its metabolites, are bound tightly by vitamin D binding protein (DBP) and more loosely bound by albumin.
**Assessment of Vitamin D Status:**
To date, total serum 25(OH)D, the sum of 25(OH)D3 and 25(OH)D2, is the accepted biomarker of vitamin D status. Observational studies have indicated the beneficial effects of an optimal vitamin D status on various outcomes not directly associated with the classical target tissues for the hormone, the so-called pleiotropic effects. Based on these studies, mostly using traditional radioimmunoassay measurements, vitamin D guidelines issued by major organizations worldwide recommend optimal 25(OH)D levels to be in the range of 50 to 75 nmol/L (20-30 ng/mL). However, optimal levels are still debated for several reasons. Lack of assay standardization contributes to the problem, and initiatives should be implemented to overcome it. In this perspective, the Endocrine Society (ES) has asked a task force to review its 2011 guidelines.
**Clinical Outcomes of Vitamin D Deficiency:**
**Skeletal Outcomes:**
Vitamin D deficiency leads to a decrease in intestinal absorption of calcium and phosphate. Other biochemical abnormalities, such as hypocalcemia, hypophosphatemia, and an increase in alkaline phosphatase, become apparent when serum 25(OH)D concentrations are lower than 25 nmol/L. In milder forms of vitamin D deficiency, the lower calcium concentration causes secondary hyperparathyroidism, which increases the conversion of 25(OH)D into 1,25(OH)2D, increasing calcium absorption and correcting serum calcium. Secondary hyperparathyroidism causes an increase in bone turnover, with relatively higher bone resorption at cortical sites. More severe long-standing vitamin D deficiency causes a decrease in the mineralization of newly formed osteoid tissue. This is visible in bone biopsies as an increase in osteoid surface and volume and increased thickness of osteoid seams, leading to the clinical picture of osteomalacia. Vitamin D deficiency and related secondary hyperparathyroidism cause bone loss and fractures in older adults. The incidence of hip fractures attributable to vitamin D deficiency has been estimated at 5% to 10%. Meta-analyses of clinical trials with vitamin D and calcium have demonstrated a decrease in hip and other fractures of around 10% in nursing home residents, whereas vitamin D alone was not effective. In these studies, baseline mean serum 25(OH)D after cross-calibration was found to be very low—namely less than 25 nmol/L—as was the calcium intake. As almost all effective trials used a calcium supplement in addition to vitamin D, the effect on BMD of vitamin D supplements alone is difficult to determine, but it is considered to be less than 1%, and high doses may even be harmful when administered to vitamin D-replete individuals. Recent RCTs such as the VIDA, VITAL, and D-Health studies do not show skeletal benefits for mostly vitamin D-replete adults and older individuals; for example, in the VITAL trial, cholecalciferol supplementation did not result in a significantly lower risk of fractures (total, nonvertebral, and hip fractures) than placebo among generally healthy midlife and older adults not selected for vitamin D deficiency, low bone mass, or osteoporosis. In the D-Health study, large bolus monthly doses (60 000 IU) resulted in no increase nor decrease in fracture risk overall. However, the hazard ratio appeared to decrease with increasing follow-up time. Interestingly, in a recent retrospective longitudinal study, the use of cholecalciferol was associated with reduced incidence of morphometric vertebral fractures in high skeletal risk, such as acromegaly. In a recent umbrella review of meta-analyses of vitamin D RCTs, the only consistent significant findings were for calcium and vitamin D, and not vitamin D alone, in reducing the risk of hip fractures by 16% to 39%, in 8 of 13 meta-analyses, and of any fracture, by 5% to 26%, in 8 of 14 meta-analyses. These findings were driven by events in institutionalized older and frailer individuals.
**Extraskeletal Outcomes:**
Putative extraskeletal outcomes of vitamin D deficiency are summarized in Fig. 4. There are many preclinical data on the extraskeletal effects of the vitamin D endocrine system, including gene regulation, cellular function, and in vivo animal studies. Indeed, about 3% of the mammalian genome is under some control of vitamin D, and most cells express VDR or can synthesize the active hormone 1,25(OH)2D locally. Observational data largely align with these data as poor vitamin D status is associated with many human diseases. To complete the observational data, many large-scale trials that evaluated the effects of vitamin D supplementation on several extraskeletal health outcomes have been carried out recently, including the large VITAL (USA) and D-Health (Australia) studies, as well as the ViDA (New Zealand), FIND (Finland), and the D2d (USA) trials. Smaller scale studies, such as the Calgary and the DO-Health (Switzerland), provide additional data. Moreover, there are now good genetic data on the prediction of serum 25(OH)D, which resulted in about 100 MR studies.
**Cancer:**
No effects of vitamin D supplementation on cancer risk were observed in the large VITAL and ViDA trials, nor the FIND trial using daily dosing in older participants, nor on cancer mortality in the D-Health study, which used monthly dosing—in line with prior trials and MR results. Based on several MR studies, small changes in vitamin D status are unlikely to affect cancer incidence. However, a subanalysis of the VITAL trial (although not corrected for multiple end point analysis) showed that vitamin D supplementation could have some minor benefits in individuals with normal BMI. In addition, several independent trials have suggested, in post hoc analysis, the potential benefits of vitamin D supplementation on cancer mortality, especially when the follow-up is longer than 4 years. A meta-analysis of RCTs suggested that vitamin D supplementation decreased cancer mortality; an updated version of this study specifically designed to examine whether results varied by daily vs infrequent large-bolus dosing and by whether the trial participants had obesity or not found that overall benefit of vitamin D supplementation is lost when all the studies are considered. However, when considering daily regimens, vitamin D supplementation reduced total cancer mortality and incidence in normal-weight individuals. Therefore, a link between vitamin D status and cancer incidence or mortality can be hypothesized, and supplementation might be effective only with daily dosages, especially in people with BMI within a normal range.
**Cardiovascular Risk:**
Convergent evidence from MR studies and RCTs suggests that vitamin D supplementation does not decrease the risk of cardiovascular disease (CVD), especially in vitamin D-replete adults. This conclusion may also apply to those with vitamin D deficiency based on subgroup analyses of the ViDA and VITAL trials. However, both studies recruited very few participants with severe vitamin D deficiency, rendering these conclusions uncertain. These null findings were corroborated by a meta-analysis of 21 RCTs. Nonetheless, more recent findings might suggest some small benefits. A detailed analysis of the VIDA trial found some modest benefits on central (but not peripheral) blood pressure, but the implications of this observation are limited because of the small scale of this VIDA substudy. The FIND trial failed to note a reduction in the number of major CV events, which was one of the two primary end points; however, subsequent exploratory analyses revealed that high-dose vitamin D supplementation might result in benefits in atrial fibrillation prevention in older individuals, even in case of relatively high baseline 25(OH)D concentrations. In the D-Health trial, the overall rate of major CV and especially the rate of myocardial infarction and coronary revascularization—was lower in the intervention group compared to the placebo group, although the absolute risk difference was small, and the CI was consistent with a null finding (hazard ratio 0.91; 95% CI, 0.81-1.01); moreover, the protective benefits could be higher in those taking CV drugs at baseline.
**Respiratory Effects:**
Vitamin D is known to influence the immune system. Most immune cells express the VDR and vitamin D metabolism-related enzymes; 1,25(OH)2D, in particular, induces innate antimicrobial effector mechanisms such as the antimicrobial peptides cathelicidin LL-37 and human beta-defensin 2. Indeed, clinical data regarding the effects of adequate vitamin D status and supplementation on respiratory infections confirm, at least in part, its potential beneficial outcomes. Serum 25(OH)D levels of less than 25 nmol/L are associated (observationally and genetically) with an increased risk of bacterial pneumonia. Individual participant data from a meta-analysis of 25 trials showed a small but significant decrease in the incidence of acute respiratory infections in the vitamin D group compared with the control group when baseline vitamin D status was poor (<25 nmol/L). A more recent, updated meta-analysis from the same group, including almost 50 RCTs, shows a protective but very small effect against respiratory infections following vitamin D supplementation with daily doses of 400 to 1000 IU; in contrast to their first meta-analysis, baseline vitamin D status did not modify the results in this more recent one.
**Autoimmune Diseases:**
Conversely, from the innate immune system, the adaptive immune system is downregulated by 1,25(OH)2D in animal models. Thus, vitamin D deficiency might predispose to autoimmune diseases. Observational studies have suggested this effect might apply to humans. The VITAL RCT showed that vitamin D supplementation decreased the risk of autoimmune diseases, especially rheumatoid arthritis and polymyalgia rheumatica, and at least 8 large MR studies all agree that genetically predicted lower 25(OH)D levels increased the risk of developing multiple sclerosis either during adolescence or adulthood. In any case, the low number of intervention studies so far conducted does not allow clarification of the relationship between vitamin D and autoimmune diseases. However, these studies to date seem promising.
**Diabetes:**
Despite observational studies consistently confirming lower serum 25(OH)D concentrations in patients with T2D or metabolic syndrome, most MR studies have not supported these conclusions. In a small subgroup of individuals with obesity and prediabetes, supplementation provided some modest benefit, albeit lower than lifestyle modifications or metformin. Of note, daily vitamin D supplementation (4000 IU) in the large D2d trial did not retard the progression of prediabetes into T2D. A post hoc and meta-analysis, however, suggested a possible beneficial effect in individuals with vitamin D deficiency (<30 nmol/L) at baseline or in participants who were able to achieve consistently high (≥100 nmol/L) serum 25(OH)D levels. Furthermore, analysis of the combined results of the D2d (US), Tromsø (Norway), and DPVD (Japan) RCTs—which were specifically designed and conducted to test whether vitamin D reduces the risk of diabetes in adults with prediabetes—showed that vitamin D supplementation reduced the risk of developing T2D in people with prediabetes not selected for vitamin D deficiency. In all 3 trials, the risk for diabetes was reduced in the group assigned to vitamin D compared to the placebo group, which did so in a remarkably similar way. The observed differences missed statistical significance in any trial because the reported risk reductions were smaller than each trial was powered to detect. An updated individual participant data meta-analysis of the same trials showed that vitamin D reduced the risk of progression from prediabetes to diabetes by 15%. Also, vitamin D increased the likelihood of regression to normal glucose regulation by 30%, with no evidence of risk. In additional analyses, participants in the vitamin D group who maintained intratrial blood 25(OH)D of 50 ng/mL or greater (≥125 nmol/L) had a 76% risk reduction in new-onset diabetes compared to those who maintained blood 25(OH)D of 20 to 29 ng/mL (50-75 nmol/L). All participants received and were encouraged to follow the current lifestyle-based advice for diabetes prevention. Based on the results of this meta-analysis, the benefit-to-risk ratio of vitamin D to lower the risk of developing T2D in adults with prediabetes is favorable. These results should not be extrapolated to the general population at low or average risk for diabetes, as the benefit-to-risk ratio of high doses for diabetes prevention may not be favorable. Despite these promising results, some questions remain, that is, the optimal vitamin D dose or formulation and the specific blood 25(OH)D level to maximize benefit with little or no risk of any side effects.
**Mortality:**
Observational data have repeatedly linked poor vitamin D status with increased mortality. Large, older meta-analyses dealing mostly with women older than 70 years showed a 6% to 11% reduction in mortality. However, adding the newest megatrials eliminated this effect, possibly because they recruited a younger population. In these megatrials, overall mortality was much lower than shown in the previous meta-analyses, and no effect of vitamin D supplementation on overall mortality was observed. A Cochrane meta-analysis of 56 randomized trials including almost 100 000 participants, of whom were women older than 70 years, revealed that vitamin D, administered over 4 years, decreased mortality; this effect was seen in 38 trials of vitamin D3, but not with other forms of vitamin D. A newer meta-analysis of 52 RCTs, including more than 75 000 individuals, concluded that vitamin D (either vitamin D3 or D2) supplementation did not change mortality compared with no supplementation. Again, subanalyses found that vitamin D3 (instead of D2) supplementation tended to reduce mortality. Some MR studies found a link between lower predicted serum 25(OH)D and mortality, especially in individuals with rather poor vitamin D status (<16 ng/mL). An individual participant data meta-analysis of almost 27 000 study participants with 25(OH)D levels standardized per VDSP protocols showed an association between low 25(OH)D and increased risk of all-cause mortality. The positive but small effect of vitamin D on mortality was confirmed by a recent umbrella review of observational, randomized, and MR studies. In conclusion, if vitamin D supplementation benefits extraskeletal health outcomes and major diseases, it is likely to have some effects on mortality, especially in older adults with poor vitamin D status, but not in younger, replete individuals.
**Summary of Vitamin D Deficiency-associated Clinical Outcomes:**
The long-known skeletal benefits of vitamin D and calcium related to rickets or osteoporosis remain valid. Most reported extraskeletal benefits of vitamin D were not confirmed by recent, large RCTs. The gradual increase in vitamin D levels in Western populations may explain these null findings, and older trials and meta-analyses may be more likely to show benefits because individuals were more likely to be vitamin D deficient than they are nowadays. RCTs and meta-analyses published to date do not have adequate power to evaluate important subgroups, such as individuals with low 25(OH)D levels, men, the very old, ethnic groups other than White individuals, and those from low-income countries. Moreover, most of the studies use adverse events data to identify fractures and were performed in adults who were vitamin D replete at baseline in whom benefit would be unlikely and toxicity possible. Such studies confound the identification of possible beneficial effects in vitamin D-deficient individuals who might benefit from supplementation. Thus, when it comes to vitamin D, it is advisable to "giveth to those who needeth." In fact, the benefit-to-risk ratio for vitamin D depends on the target population and medical condition. It would be incorrect to extrapolate vitamin D guidelines that apply to the general population (such as those from the US National Academic of Medicine) to avoid vitamin D deficiency (ie, rickets, osteomalacia) and promote bone health to special populations for whom the benefit-to-risk ratio of vitamin D would be different. Nonetheless, RCTs, MR studies, and meta-analyses suggest a link between vitamin D status with the immune system and diabetes, as well as fleeting effects on some CV events and some benefits on mortality risk when vitamin D3 is used.
**Vitamin D Supplementation:**
**Dosing Regimens:**
The term "dose" in relation to vitamin D is typically used to signify the measured quantity of vitamin D (usually cholecalciferol, but other formulations such as ergocalciferol, eldecalcitol, calcifediol, etc are also available) in a pill. It is expressed as μg or IU (where 10 µg is 400 IU). The dose of cholecalciferol is considered an important measure as it correlates with the change in blood 25(OH)D level, which is commonly used to define vitamin D status and correlates with important clinical outcomes. Doses can be considered as "loading" or "maintenance." The most common use of a loading dose is to rapidly improve a low blood 25(OH)D; however, the clinical wisdom of this approach is questionable, especially given studies that demonstrate adverse effects with very high doses given infrequently, as discussed next. Intermittent administration of large doses is also used to optimize adherence. Daily doses are generally preferred when vitamin D replacement is considered necessary. The effect of a given dose on changing blood 25(OH)D varies considerably from person to person due to many factors, such as body weight, absorption, diet, degree of adiposity, CYP2R1 activity, DBP. The recommended dietary allowance for vitamin D by the National Academy of Medicine is set at 400 to 800 IU per day, and the tolerable upper intake level at 4000 IU per day; however, the "optimal" dose of vitamin D varies by the desired outcome, and other authors suggest that the upper limit of safety may be lower than 4000 IU per day. For example, 400 to 800 IU of vitamin D per day may be adequate to avoid clinical vitamin D deficiency and maintain calcium homeostasis in healthy individuals. Doses of vitamin D higher than the recommended upper limit may be associated with toxicity; nonetheless, daily doses up to 10 000 IU have been used without safety concerns. Careful and judicious use of vitamin D will permit the realization of potential benefits and achieving optimal outcomes.
**Daily Supplementation:**
From a physiological perspective, daily administration of cholecalciferol seems to be most natural. Indeed, it appears that a daily approach results in higher efficacy in terms of 25(OH)D exposure and extraskeletal benefits. In a recent RCT comparing 3 different dosing regimens in vitamin D-deficient participants with similar total end-of-study cumulative doses (D3 daily 10 000 IU 8 weeks, then 1000 IU for 4 weeks; 50 000 IU weekly for 12 weeks; and 100 000 IU every 2 weeks for 12 weeks), the group receiving the daily supplementation was the quickest to reach sufficiency (<2 weeks, although receiving a higher cumulative dose in the first 8 weeks when compared to the other 2 arms) and reached the highest serum 25(OH)D levels. Importantly, daily administration was associated with higher systemic exposure to 25(OH)D (greater area under the curve, +23% and +27% compared to weekly and biweekly administration, respectively), even when corrected for the cumulative dose. The greater 25(OH)D exposure of daily regimens could be due to lower activation of the 24-hydroxylase enzyme (CYP24A1). In an RCT of lactating women comparing the effect of bolus (150 000 IU) vs daily vitamin D3 dosing (5000 IU) on vitamin D3 catabolism, a single high-bolus dose of vitamin D led to greater production of 24,25(OH)2D3, relative to the 25(OH)D3 value than did daily vitamin D supplementation, with this effect persisting for at least 28 days after supplementation. The greater therapeutic potential of daily regimens compared to other regimens might be less relevant at lower doses (<2000 IU). Two studies comparing 2000 IU/day vs 50 000 IU/month and 800 IU/day vs 5600 IU/month found no statistically significant differences in the 2 areas under the curves.
**Nondaily Supplementation:**
Intermittent vitamin D dosing usually uses a greater amount to reach equivalent doses with fewer administrations. The rationale of this approach is to enhance adherence and ease management of specific patient groups, such as children and community-dwelling older people. Indeed, low adherence to vitamin D prescription has often been reported, although the topic is controversial. For example, Albrecht et al recently investigated adherence to bone health-promoting lifestyle recommendations concerning osteoporosis status in a cross-sectional database of community-dwelling older adults (aged 65-75 years). In high-risk osteoporosis patients, adherence to vitamin D intake, defined as regular consumption of vitamin D-rich foods and/or vitamin D supplements, was high, ranging from 85% (women) to 93% (men). In contrast, in a cross-sectional study of pediatric outpatients affected by various diseases, Arshad et al found that adherence to vitamin D prescription was quite low, particularly in those with diseases where vitamin D deficiency presents as a high-risk condition. For these reasons, recurrent and protracted intervals of vitamin D supplementation appear to be an effective and convenient way to achieve and maintain sufficient vitamin D status and to increase patients' adherence, but there is no agreement that treatment simplification with intermittent dosing significantly improves compliance and there is consistent evidence to discourage the use of “megadoses” due to the possible side effects.
**Weekly and Monthly Regimens:**
With equivalent doses and large formulations, daily, weekly, and monthly supplementation may lead to similar increases and levels of 25(OH)D in middle-aged and obese individuals, in older individuals with hip fractures, and children with CKD. However, one study concluded that a daily regimen was more efficient in circulating 25(OH)D than weekly or monthly administration, but with different formulations. As compared with a daily regimen, a bolus dose is associated with a higher 24,25(OH)2D level and a higher 24,25(OH)2D to 25(OH)D ratio. In a monocentric, open-label randomized study in postmenopausal women, weekly vitamin D was more efficient than monthly in improving muscular function (measured through the Sit-to-Stand and Timed-Up-and-Go tests). Monthly regimens have been tested in several large trials with multiple outcomes. Compared to a placebo, 100 000 IU monthly did not influence the risk of CVD, falls, fracture, or cancer, and lung or arterial functions in vitamin D-replete individuals. In those participants with baseline 25(OH)D lower than 50 nmol/L, the 100 000 IU vitamin D regimen increased lumbar spine BMD by 2.6% and improved lung and arterial functions. In the D-Health trial including more than 21 000 individuals, with 24% of them having a 25(OH)D level less than 50 nmol/L, 60 000 IU monthly did not influence all-cause mortality but was associated with a higher risk of falls in those with a BMI of less than 25. This observation was in agreement with another trial in which a higher percentage of fallers was detected with 60 000 IU/month compared to 24 000 IU/month over 1 year. In small trials, few episodes of hypercalcemia were reported with weekly doses between 20 000 and 100 000 IU in various target populations. Overall, trials with weekly or monthly vitamin D supplementation regimens did not show significant effects on clinical variables. This could be due to the recruited population (vitamin D-replete or obese individuals) or too large vitamin D doses leading to a U-shape dose-response relationship. Currently, there is no evidence of a superiority in the benefit/risk ratio of weekly or monthly vitamin D regimens over daily supplementation.
**Longer Intervals:**
Although one study using high doses with prolonged intervals (100 000 IU every 4 months) administered to community-dwelling adults older than 50 years found a reduction in fractures, other similar studies (500 000 IU every year/150 000 IU every 3 months) did not show a reduction in hip/vertebral/nonvertebral/total fracture incidence. This was also evidenced by the systematic review and meta-analysis of Zhao et al. In studies on the efficacy of vitamin D administration, the basal values of 25(OH)D are often either not measured or are at normal/high levels, making it difficult to understand the real effect of supplementation on 25(OH)D values. In a subgroup analysis of Zhao's study, no differences in fracture incidence were found between intermittent high doses given once every year and other interval regimens. In Zhao's meta-analysis, reference is made to the study by Witham and colleagues in which no negative effects of longer intervals of high-dose vitamin D administration on blood pressure in older patients with isolated systolic hypertension were reported. Regarding the relation between long-term intervals of vitamin D administration and CVD risk, falls, and fracture outcomes in older and community-dwelling people, in a systematic review with meta-analysis, Barbarawi et al did not find significant results favoring vitamin D intervention (100 000 IU every 4 months/500000 IU yearly) in preventing falls, fractures, or CVDs. Even in works cited in this meta-analysis, the basal 25(OH)D values were either not reported or sufficient. In a systematic review with meta-analysis, Yang et al cited 2 works that investigated the effect of intermittent high doses of vitamin D as adjuvant treatment in pneumonia in children (100 000 IU every 3 months and 300 000 IU quarterly for 1 year) on the incidence rate of repeated episodes of pneumonia, rate of intensive care unit (ICU) hospital admission, and complications rate. In both cases, no significant definitively positive effects were found. Regarding the safety of longer-interval vitamin D supplementation, in a recent systematic review with meta-analysis on children, Brustad et al did not find any association with severe side effects. This was also seen in other studies with protracted intervals of vitamin D administration.
**Summary of Vitamin D Dosing Regimens:**
In conclusion, one of the major justifications for longer intervals with high doses in vitamin D administration, namely, to address low compliance with more frequent regimens, is controversial. The rationale gains support in children and adolescents rather than in older individuals. However, it has to be taken into account that the cited meta-analyses underscored the point that there is no evidence of efficacy in intermittent high-dose and longer intervals of vitamin D administration in reducing fracture rate, falls, CV events, or infectious diseases. An increase in falls in older individuals has been observed with large, intermittent dosing (the literature regarding falls is somewhat controversial in part because there are no reliable methods to capture falls, as both diaries and self-reports are flawed). These conclusions should be tempered by inherent flaws in many reports in which the baseline vitamin D dosage or pretreatment 25(OH)D levels are not provided.
**Routes of Administration:**
**Oral Supplementation:**
Cholecalciferol (vitamin D3) and ergocalciferol (vitamin D₂) are fat-soluble vitamins that are absorbed in the small intestine. Because they are lipophilic compounds, their absorption is similar to the absorption of lipids. Vitamin D is incorporated into micelles with biliary salts on the micelle surface. On average, about 80% of vitamin D is absorbed, but the variation in absorption can be large (55%-99%). Taking vitamin D supplements with a fat-containing meal may improve vitamin D absorption. Cholecalciferol and ergocalciferol are both rapidly absorbed, and the plasma levels peak after about 24 hours of ingestion. Absorption into the enterocytes of the intestinal wall was thought to be a passive process, but there is some evidence that vitamin D, especially in dietary doses, is also actively transported through the enterocyte membranes via cholesterol transporter proteins. However, passive transport seems to occur with pharmacological doses of vitamin D. From the enterocytes, vitamin D is exported in chylomicrons by the lymphatic route. Bariatric surgery and intestinal malabsorption syndromes that reduce fat absorption, such as inflammatory bowel diseases, cystic fibrosis, and severe cholestasis, can also reduce vitamin D absorption. However, intestinal malabsorption does not seem to affect the absorption of calcidiol as much, most likely because calcidiol is more water soluble, thus not requiring bile salts for absorption, and because calcidiol is absorbed by the portal route instead of the lymphatic route. As cholesterol transporters are involved in vitamin D absorption, factors that interfere with cholesterol absorption could also affect vitamin D absorption. However, ezetimibe, an inhibitor of cholesterol transport, does not seem to affect vitamin D absorption despite the reduction in cholesterol absorption. There is also no strong evidence that phytosterols, plant sterols used to inhibit cholesterol absorption, impair vitamin D absorption. In contrast, there is some evidence that drugs used to reduce intestinal fat absorption, such as orlistat and olestra, may also reduce vitamin D absorption. Vitamin D supplements are available in different vehicles, such as oil-containing gel capsules, oily drops, and hard powder tablets. Although it could be hypothesized that vitamin D would be better absorbed from oil-based vehicles, no convincing evidence supports this premise. In fact, there is some evidence that vitamin D may be better absorbed from a powder-based vehicle than from an oil-based vehicle in cases of intestinal fat malabsorption, such as in cystic fibrosis.
**Parenteral Administration:**
The optimal treatment of hypovitaminosis D in the general population and disease states is still debated. Parenteral administration of intermittent vitamin D boluses may be indicated in patients with hypovitaminosis D who are not suitable for oral intake or with intestinal malabsorptive diseases, including inflammatory bowel disease, celiac disease, pancreatic insufficiency, short-bowel syndrome, and post bariatric surgery. Based on advantageous pharmacokinetic properties and evidence-based clinical data, intramuscular cholecalciferol may be the preferred form of vitamin D to be used in these clinical settings. In fact, it has been shown that cholecalciferol was able to reach higher serum 25(OH)D levels more rapidly than ergocalciferol when both vitamin D forms were administered as a single large intramuscular dose (300 000 UI) in adult or older patients with hypovitaminosis D. Moreover, in the study by Romagnoli et al, 2 months after administration of this large, intramuscular cholecalciferol dose, serum 25(OH)D levels were higher than those obtained after the same oral dose. Therefore, intermittent intramuscular cholecalciferol could be useful in clinical conditions when rapid correction of hypovitaminosis D is unnecessary and for long-term maintenance of adequate serum vitamin D levels, as in some older patients, to improve their adherence to vitamin D supplementation. However, safety concerns limit the clinical use of intermittent, excessive vitamin D doses. In fact, large intramuscular boluses (300 000 IU) induce unwanted effects such as an increase in falls and fracture events or enhance bone turnover. There is a consensus to administer vitamin D boluses not higher than 100 000 IU. In conclusion, the therapeutic regimen to recover from vitamin D deficiency should be tailored to patients' characteristics, such as age, BMI, severity of vitamin D deficiency, concurrent comorbidity, and use of other drugs.
**Different Forms of Vitamin D Supplementation:**
The main supplemental oral forms of vitamin D are cholecalciferol (vitamin D3) and ergocalciferol (vitamin D2). Both are readily available without a prescription. Cholecalciferol is the most used form of supplemental vitamin D. Calcidiol (calcifediol, 25(OH)D), the inactive vitamin D metabolite produced in the liver, and other vitamin D analogues, such as calcitriol (1,25(OH)2D, the physiologically active form of vitamin D) and alfacalcidol (1-hydroxyvitamin D), are used as prescription medicines in some conditions.
**Ergocalciferol:**
Ergocalciferol does exist in nature (mainly in plants and fungi), and low circulating levels of 25(OH)D2 are present in free-ranging nonhuman primates and human population studies. The 2 forms of vitamin D, cholecalciferol (D3) and ergocalciferol (D2), are often used interchangeably as supplementation or treatment of vitamin D deficiency as, historically, vitamins D2 and D3 were considered equally effective in treating rickets. Similarly, previous recommendations consider vitamins D2 and D3 interchangeable. Subsequently, however, multiple studies and meta-analyses comparing the effect of D2 and D3 on circulating 25(OH)D concentration have found cholecalciferol to be superior. Challenges to 25(OH)D measurement are widely recognized. The presence of 2 circulating 25(OH)D forms, 25(OH)D3 and 25(OH)D2, adds additional challenges, notably for automated immunoassays. Importantly, it is possible that the antibodies used in immunoassays may not detect 25(OH)D2 and 25(OH)D3 equally, and the proprietary approach to releasing 25(OH)D from DBP may not liberate the 2 forms equally. As such, it is perhaps unsurprising that multiple reports find overestimation or underestimation of total 25(OH)D in the presence of substantial amounts of 25(OH)D2. This problem was corroborated by a recent interlaboratory comparison performed at the behest of the VDSP. Assay underestimation of total 25(OH)D in the presence of substantial amounts of 25(OH)D2 can have clinical consequences. A recent small report of patients receiving 50 000 IU of ergocalciferol every 2 weeks found 40% (6/15) to have total 25(OH)D levels less than 30 ng/mL when measured by immunoassay whereas all had values above 42 ng/mL when measured by LC-MS/MS. It is easy to imagine that such patients with “low” 25(OH)D values would have their dose increased, with at least potential toxicity, or undergo additional evaluation, such as for malabsorption. Thus, these assay issues are not clinically inconsequential. Issues surrounding ergocalciferol use are of note for the United States, where 50 000 IU of vitamin D2 was the only high-dose preparation available by physician prescription and, therefore, ergocalciferol was widely prescribed. Now, instead, 50 000 IU of vitamin D3 is available by prescription. In addition to assay issues, widespread use of intermittent high-dose ergocalciferol ("bolus" therapy) appears to alter vitamin D metabolism, with increased 24-hydroxylase activity. Other adverse consequences of high-dose vitamin D therapy, notably increased fall risk, are reported and have led to calls to critically assess daily vs bolus vitamin D therapy. To summarize, vitamins D2 and D3 are not equivalent in raising circulating 25(OH)D, and bolus dosing may have adverse effects on vitamin D metabolism and clinical outcomes. As such, it is to be expected that calls for the use of only cholecalciferol and avoidance of ergocalciferol have been and continue to be published with recent osteoporosis-treatment guidance advising cholecalciferol over ergocalciferol. Despite these recommendations, high-dose ergocalciferol remains widely prescribed in the United States.
**Calcifediol:**
Calcifediol is the intermediate metabolite between cholecalciferol and calcitriol. Several pharmacokinetic studies performed in the last 4 decades have demonstrated its hydrophilic properties, leading to higher solubility in organic solvents, less sequestration in adipose tissue, smaller distribution volume, and shorter half-life when compared to cholecalciferol. By virtue of its hydrophilic properties, calcifediol is readily absorbed via the venous portal system and thus quickly increases circulating concentrations of 25(OH)D 3. In contrast to cholecalciferol, which is mostly stored in fat tissue, 25(OH)D tends to be more evenly distributed throughout the body (20% in muscle, 30% in circulation, 35% in fat, and 15% elsewhere). The administered dose will generally lead to predictable 25(OH)D levels and effective PTH suppression. In cases of toxicity, this form of vitamin D is easier to manage than cholecalciferol. Moreover, the greater affinity of calcifediol for DBP allows for more efficient internalization in cells expressing the megalin-cubilin system of endocytic receptors, such as the parathyroids and the renal tissue. Such properties provide the rationale for using calcifediol in specific clinical conditions. The clinical situations that make use of calcifediol attractive are obesity, hepatic failure, patients with inactivating mutations of genes encoding CYP2R1 (the principal enzyme that is responsible for vitamin D 25-hydroxylation), or patients taking drugs that could influence the activity of cytochrome enzymes (ie, antiretroviral or antitubercular). Calcifediol was shown to have the same bioavailability in healthy adults with differing BMI and adults with intestinal malabsorption compared to controls. In an RCT on vitamin D-deficient, postmenopausal women, weekly calcifediol was found to be more effective and faster acting compared to cholecalciferol in increasing 25(OH)D serum levels. This more favorable kinetics led to greater improvement in muscle function. In another RCT in 35 healthy women aged 50 to 70 years, calcifediol given daily, weekly, or as a single bolus was about 2 to 3 times more potent in increasing plasma 25(OH)D3 concentrations than cholecalciferol. New extended-release calcifediol formulations are more effective than cholecalciferol in raising serum 25(OH)D levels even in overweight nondialytic CKD patients with secondary hyperparathyroidism; nonetheless, it must be noted that these data arise from observational, retrospective data and subgroup post hoc analysis of RCTs. Recently, retrospective studies have highlighted the role of calcifediol administration on various end points related to COVID-19 infection. To demonstrate a causative effect, Nogues et al investigated 2 cohorts of patients with COVID-19, 1 of whom was untreated and 1 assigned to the oral calcifediol group. The treatment regimen consisted of oral calcifediol (0.532 mg the day of admission), followed by doses of 0.266 mg on days 3, 7, 15, and 30. Out of 447 patients treated with calcifediol at admission, 20 (4.5%) required the ICU, and 21 (4.7%) died; this was significantly lower compared to the untreated group of 391 patients, of whom 82 (21%) required the ICU and 62 (15.9%) died (both P<.01). Adjusted logistic regression of calcifediol treatment on ICU admission indicates that patients treated with calcifediol had a lower risk of ICU admission (OR 0.02; 95% CI, 0.07-0.23) and mortality (OR 0.21; 95% CI, 0.10-0.43), suggesting an effectiveness of calcifediol treatment. In summary, calcifediol seems to represent a form of vitamin D that is useful for replenishing vitamin D status. Most attractive clinical settings include malabsorption syndromes, obesity, CYP2R1 dysfunction, or situations in which quick attainment of vitamin D sufficiency is desirable.
**Calcitriol:**
Calcitriol is the active hormonal form of vitamin D and the natural VDR ligand. It promotes active intestinal calcium absorption and suppresses PTH secretion. Calcitriol has a short half-life of around 5 to 8 hours; therefore, it should be administered daily (or with intermittent regimens) and sometimes in lower doses distributed over a 24-hour period. As calcitriol is not an organic micronutrient, its use in clinical practice requires careful monitoring. Calcitriol increases the activity of CYP24A1, which stimulates the degradation of 25(OH)D. This results in serum 25(OH)D not being useful as a marker of adequate vitamin D supplementation and reduced potential benefits of physiological extrarenal/local production of calcitriol due to reduced substrate availability. Moreover, some studies have reported a more significant incidence of adverse events such as hypercalcemia and hypercalciuria. Thus, there is a need to monitor serum and urine calcium and phosphate. Because of these safety and clinical practicality issues, there is consensus that calcitriol use should be limited to hormone replacement for patients with limited/absent renal tubular 1-a-hydroxylase activity, as their capacity to produce calcitriol is severely limited. Indeed, calcitriol was first used to treat patients with vitamin D-resistant rickets type 1. Other indications are X-linked hypophosphatemic rickets, chronic hypoparathyroidism, as an alternative to the use of the native missing hormone PTH, and moderate-to-severe kidney failure when calcitriol production is impaired or to suppress excessive PTH secretion. This use helps to control secondary hyperparathyroidism and resultant metabolic bone diseases. However, as calcitriol use is associated with frequent hypercalcemia, its use could be replaced by analogues with less calcemic activity approved for use in patients with secondary hyperparathyroidism in renal failure, in particular maxacalcitol (22-oxa-1,25(OH)2D3) and falecalcitriol (1,25(OH)2-26,27-F6-D3), which are currently available in Japan, and paricalcitol (19-nor-1,25(OH)2D2) and doxercalciferol (1a(OH)D2), available in the United States. Calcitriol has also been proposed for the treatment of osteoporosis, but it is not approved in this setting. In conclusion, calcitriol is not suitable for supplementation or nutritional fortification, and none of many excellent reviews, guidelines, and policy papers consider the use of calcitriol in the nutritional context (supplementation and fortification). However, guidelines suggest that vitamin D supplementation is advised in patients with chronic hypoparathyroidism, chronic kidney failure, and low vitamin D status in addition to receiving therapeutic doses of calcitriol. Such a recommendation is motivated by the activity of extrarenal 1-a-hydroxylase, which is compromised by reduced renal function (ie, not regulated by PTH) and is not regulated by feedback mechanisms.
**Vitamin D Safety and Monitoring:**
**Vitamin D Toxicity:**
VDT is a clinical condition characterized by excess vitamin D (hypervitaminosis D), resulting in severe hypercalcemia that may persist for a prolonged period of time, leading to serious health consequences. Signs and symptoms of VDT are related primarily to hypercalcemia, with complications encompassing adverse events in the CV, renal, gastrointestinal, neurological, and musculoskeletal systems. VDT prevalence is unknown, but it is rare due to the wide therapeutic index of vitamin D. Evidence from systematic studies of VDT in humans is missing for ethical reasons, and data mostly stem from studies of VDT in animals and anecdotal reports. The condition of infantile hypercalcemia was first described in the United Kingdom and Switzerland, showing symptoms such as failure to thrive, osteosclerosis, developmental delay, and even death, but was not immediately associated with vitamin D intake. Suggestions were made that excess vitamin D intake may be a causative factor (children received up to 35 000 IU daily). Eventually, the British Ministry of Health suggested a substantial reduction in vitamin D allowance, resulting in a marked decrease in infantile hypercalcemia cases. As the prescriptions of vitamin D products are increasing worldwide, so is the number of VDT reports, with more than 75% published since 2010. Many of these cases result from inappropriate prescribing; moreover, the prescription of high-dose unlicensed and poorly manufactured treatments can be greater than 60%, as they are cheaper. In healthy individuals, hypervitaminosis D is usually defined as "exogenous" as it develops after uncontrolled use of megadoses of vitamin D or its metabolites or analogues, as in case of high dose of calcifediol leading to a faster increase in 25(OH)D serum levels compared with cholecalciferol but easier to manage than cholecalciferol in case of toxicity for its hydrophilicity and lesser sequestration in adipose tissue. On the other hand, excessive production of calcitriol in granulomatous disorders, lymphomas, primary hyperparathyroidism, and idiopathic infantile hypercalcemia results in "endogenous" hypervitaminosis D. VDT is defined by a biochemical phenotype with markedly elevated calcifediol concentrations (>150 ng/mL or >375 nmol/L), along with dihydroxylated metabolites (24,25(OH)2D3; 25,26(OH)2D3, 25(OH)D3-26,23-lactone), unless the causal agents are vitamin D analogues, such as paricalcitol. Calcitriol levels may be in the normal reference range or even reduced in exogenous VDT while elevated in endogenous VDT. PTH levels can be very low or undetectable. VDT thus results in severe hypercalcemia, hypercalciuria, and hyperphosphatemia. Pathogenetically, hypercalcemia is a consequence of high calcifediol levels in exogenous VDT (with calcifediol at pharmacological concentrations overcoming VDR affinity disadvantages and/or displacing 1,25(OH)2D from DBP), while high calcitriol levels cause endogenous VDT. Exogenous factors that interact with VDT risk include dosage, calcium in the diet or as a supplement, vitamin D intake with the diet, social status (ie, neglected patients), artificial UV light treatment sessions, quantity of supplement use, and time of exposure. Endogenous risk factors comprise age, sex, vitamin D status, hypersensitivity syndromes, and the pharmacogenetics of the vitamin D response and metabolism. This is why there is no clear cutoff above which VDT occurs and below which it does not. In conclusion, VDT is a rare but life-threatening event mostly caused by unintentional overdosing due to pharmaceutical products. The prescriber and dispenser should avoid unlicensed vitamin D products. VDT should always be considered a differential diagnosis when evaluating patients with hypercalcemia. Future studies should encompass the evaluation of concurrent conditions that increase the risk of VDT and include the evaluation of classic and nonclassic adverse events for VDT.
**Monitoring Vitamin D Status During Treatment:**
Monitoring treatments is important to assess their efficacy and safety. Regarding vitamin D supplementations, there is limited evidence for when to monitor response to therapy or toxicity. When it comes to achieving sufficiency in deficient patients, it seems there is no need to monitor differently according to different dosage regimens (dose and/or frequency) or baseline 25(OH)D serum values. The increase in serum 25(OH)D concentration after supplementation follows a curvilinear response with the increase of the cumulative doses. The delta increase per 100 IU depends on baseline levels, and there is less increase per 100 IU with high doses than low doses. Van Groningen et al calculated that the cholecalciferol loading dose required to reach the serum 25(OH)D target level of 75 nmol/L can be calculated as dose (IU) = 40 × [75 - serum 25(OH)D] body weight. Mean 25(OH)D levels over a 2-month period are similar to daily, weekly, or monthly administrations (although monthly dosing is associated with more variability), and sufficiency can be reached independently from the baseline 25(OH)D values. In the study by Fassio et al, all participants normalized 25(OH)D safely, regardless of dosing regimens and including patients receiving 10 000 IU/day for the first 8 weeks; moreover, no cases of hypercalcemia were recorded. With regard to recent megatrials results, no effects were found on serum calcium or calciuria unless very high doses were used, such as 4000 to 10 000 IU per day in the Calgary study. Furthermore, these studies did not confirm the modestly increased risk of kidney stones observed in the WHI trial (400 IU per day). However, there might be a need for monitoring in case of other vitamin D metabolite use. As discussed earlier, calcifediol acts much more rapidly than cholecalciferol in increasing serum 25(OH)D levels, resulting in greater fluctuation of 25(OH)D levels. For example, supplementation with 20 µg (800 IU) of cholecalciferol (vitamin D3) increases 25(OH)D concentrations toward 70 nmol/L (28 ng/mL) within 16 weeks, while supplementation with 10 or 15 µg calcifediol (25(OH)D) increases 25(OH)D levels more than 75 nmol/L (>30 ng/mL) in 8 and 4 weeks, respectively. To summarize, cholecalciferol can maintain physiological 25(OH)D serum levels above 30 ng/mL (75 nmol/L) but below 50 ng/mL (125 nmol/L) for a long time, regardless of whether the dosage given is daily or intermittent (weekly, fortnightly, or monthly), due to its slow pharmacokinetic elimination caused by prolonged storage and release on demand according to physiological needs. Routine monitoring of 25(OH)D levels is generally unnecessary for patients on long-term maintenance vitamin D doses of up to at least 2000 IU/day. Retesting after 8 to 12 weeks from the start of supplementation may be appropriate when poor compliance is suspected, in case of symptoms suggestive of vitamin D deficiency, and for patients at risk of persistent 25(OH)D level below 30 ng/mL (75 nmol/L). These comprise institutionalized or hospitalized individuals, people in whom vitamin D therapy uncovers subclinical primary hyperparathyroidism, obese individuals, individuals undergoing bariatric surgery, individuals who use of certain concomitant medications (eg, anticonvulsant medications, glucocorticoids), and patients with malabsorption, including inflammatory bowel disease and celiac disease. For patients on potent antiresorptive agents (eg, denosumab or zoledronic acid), vitamin D levels should be checked annually per protocol.
**Conclusions:**
The metabolism, mechanisms of action, and pathophysiology of vitamin D and its multifaceted implications in human health have been extensively investigated for more than a century. However, the role of vitamin D status assessment and the detailed outcomes of vitamin D deficiency and its supplementation are still not completely understood. Thus, we extensively reviewed the literature on controversial vitamin D topics to better clarify and summarize the "whys, whens, and hows" of vitamin D assessment and supplementation in generally healthy populations and clinical conditions. Vitamin D metabolism involves a different extensive panel of enzymes, resulting in various hormonal metabolites. Moreover, the VDR has been demonstrated to act as a key role transcription factor in most cells and can regulate a plethora of genes. New insights into the regulation of vitamin D-related enzymes and the differential mechanism of action of VDR have demonstrated potential links between several metabolic disorders and vitamin D effects. In this perspective, assessing a distinctive pattern of noncanonical vitamin D metabolites may allow us to better characterize different pathological conditions related to vitamin D metabolism that do not depend only on reduced solar exposure or vitamin D diet intake. Besides the potential utility of the evaluation of noncanonical vitamin D metabolites, 25(OH)D is nowadays the most widely accepted biomarker to evaluate vitamin D status; however, its optimal levels are still debated. Recommendations on optimal 25(OH)D levels deriving from international societies and guidelines can differ due to the different approaches used, including clinical perspectives (level of cutoff at which no individual has an undesirable outcome) or public health perspectives (level of cutoff at which 97.5% of individuals do not have an undesirable outcome). Another critical issue is the lack of an accepted laboratory test assay standardization, and this prevents a proper interpretation of data reported by different studies, resulting in enabling rational data pooling and implementation of meta-analyses focused on vitamin D influence in various clinical outcomes of interest. Thus, 25(OH)D laboratory assays should be monitored in their performance through external quality assessment plans providing target reference values from standardized measurement procedures. Vitamin D deficiency has been extensively related to the occurrence of skeletal disorders, such as rickets and osteomalacia. It can also be negatively implicated in osteopenia and osteoporosis, which must be mandatory and managed with vitamin D supplements. More recently, the interest in the putative extraskeletal effects of vitamin D have resulted in several clinical trials addressing vitamin D's influence on cancer and CV risk, respiratory effects, autoimmune diseases, diabetes, and mortality. The null results of some of these RCTs especially the megatrials—hampered the enthusiasm around these topics. However, these trials were progressively revised, and their null results were mainly related to the enrollment of vitamin D-replete adults in whom benefit would be unlikely and the inhomogeneous methodologies in vitamin D supplementation with different forms, metabolites, and doses. Indeed, subsequent secondary analyses have progressively shown that vitamin D might be useful in reducing cancer incidence and mortality in the long term, in reducing autoimmune diseases and CV events (in particular central arterial hypertension, myocardial infarction, and atrial fibrillation) occurrence, and the development of diabetes from prediabetes forms. Nonetheless, these RCTs and the following meta-analyses were not powerful enough to evaluate these crucial subgroups, and further studies with better methodological conductions are warranted. Regarding the different forms and metabolites used for vitamin D supplementation, oral administration is the preferred route, and parenteral administration should be reserved for special clinical situations, such as in patients with severe gastrointestinal malabsorption syndromes or after bariatric surgery. Cholecalciferol remains the preferred choice, and it is generally safe, requiring less strict monitoring. Ergocalciferol has been demonstrated to be less effective in raising 25(OH)D serum levels and, thus, should be reserved for specific clinical conditions. Calcifediol could be recommended in patients with obesity, malabsorption syndromes, CYP2R1 dysfunction, or in situations in which a quick, rapid achievement of vitamin D sufficiency is desirable. Calcitriol use should be limited for patients with limited/absent renal tubular 1-α-hydroxylase activity and in vitamin D—resistant rickets type 1, X-linked hypophosphatemic rickets, and chronic hypoparathyroidism. Growing preclinical and clinical observations associating vitamin D with many health clinical conditions have been progressively reported in recent years. However, the lack of rigorous methodologies on patient enrollment, vitamin D supplements, and standardized laboratory assays have limited the ability to draw definitive conclusions about these data that still need to be more clearly understood. Thus, a "whys, whens, and hows" of vitamin D assessment and supplementation derived from an international expert panel discussion about controversial topics regarding vitamin D metabolism, assessment, actions, and supplementation is needed to help the scientific community in evaluating and conducting future further studies with more rigorous methodologies, to better explore any clinical setting potentially influenced by vitamin D, and to provide reliable data required to update our international recommendations.
**Acknowledgments:**
The authors wish to acknowledge Fabio Perversi (Polistudium srl, Milan, Italy) for medical writing and Aashni Shah (Polistudium srl, Milan, Italy) for linguistic and editorial assistance.
**Funding:**
This work was supported, in part, by the International Vitamin D Expert Association (IDEA). The conference and editorial assistance were supported by an unrestricted educational grant from Abiogen Pharma, Pisa, Italy. The sponsors had no role in the selection of discussion topics, speakers, or authors, preparation, or review of this paper. Work partially supported by Glucocorticoid induced Osteoporosis Skeletal Endocrinology Group (GIOSEG).
**Disclosures:**
A.G. is a consultant for Abiogen and Takeda and received research grant to institution from Takeda. J.P.B. is a consultant for Abiogen. Travel and accommodation expenses for the 6th International Workshop on Controversies in Vitamin D for R.B., C.M., and G.H.F. were covered by Abiogen Pharma S.p.A. P.E. has received research funding from Amgen, Sanofi, and Alexion, and honoraria from Amgen. A.F. has received advisory board honoraria, consultancy fees, and/or speaker fees from Boehringer Ingelheim, UCB Pharma, and Abiogen. S.F. has received research grants to institution from Abiogen Pharma. A.R.M. reports grants from Barts Charity, The Fischer Family Foundation, DSM Nutritional Products, The Exilarch's Foundation, The Karl R Pfleger Foundation, The AIM Foundation, Thornton and Ross, Warburtons, Hyphens Pharma, and Mr Matthew Isaacs (all paid to his institution), and from UK National Institute for Health Research Clinical Research Network and the HDR UK BREATHE Hub (paid to Edinburgh University); consultancy fees from DSM Nutritional Products; honoraria from Oregon State University; payment for expert testimony from Qiagen; and support for attending meetings from Abiogen Pharma and Pharma Nord. A.R.M. is chair of the Data Safety Monitoring Board for the VITALITY (vitamin D for adolescents with HIV to reduce musculoskeletal morbidity and immunopathology) trial, and a member of the data safety monitoring board for a trial of vitamin D and zinc supplementation for improving treatment outcomes among COVID-19 patients in India; a program committee member for a vitamin D workshop; and has received vitamin D capsules for clinical trials from Pharma Nord, Synergy Biologics, and Cytoplan. S.M. has served as speaker for Abiogen Pharma, Bruno Farmaceutici, Diasorin, Geopharma, Sandoz, and UCB. He also served on advisory boards for Abiogen, Eli Lilly, Kyowa Kirin, Novo Nordisk, and UCB. R.R. has been a speaker or participated on advisory boards for Abiogen, Naturex, Nestlé, ObsEva, and Theramex. A.L.S. has received a research grant from Amgen. N.N., M.P., C.T.S., F.M.U., J.V., G.J., P.L., R.A.A., G.B., D.D.B., N.C.B., J.B., M.L.B., F.F.C., L.d.F., and L.M.D. have nothing to declare. |
# 补充钙和维生素D 防治骨质疏松症的全球临床指南进展
摘要: 补充钙和维生素D 防治骨质疏松症一直是全球权威学术机构临床指南的基本策略。然而，近来一些研究对过量补充
钙和维生素D 的健康骨骼益处提出不同的观点，从而干扰了国际公认的防治骨质疏松症的基本策略，混淆了临床医师的视
听。因此，笔者综述近几年来中国、 美国、 加拿大、 英国、 波兰、 日本、 韩国等权威学术机构的临床指南的观点，重点介绍骨质疏
松症的发病机制和病理生理学关注的骨重建的主要局部调节剂［受体激活剂核因子kb(ＲANK)及其配体ＲANKL 和诱饵受体
骨保护素(OPG);罗列各国为防治骨质疏松症推荐的钙和维生素D 摄入量;联合补充钙和维生素D 有利于防治骨质疏松症和
骨质疏松性骨折;分析补充钙和维生素D 的健康骨骼争议，一般健康人每天服用钙补充剂不应该超过1000 mg。 特别是基于亚
洲人(中国人、 日本人、 韩国人)的膳食钙摄入量、 血脂水平、 身体体重指数(BMI)和维生素D 营养状况，参考中国、 日本、 韩国
的补充钙和维生素D 防治骨质疏松症的临床指南，提出钙和维生素D 补充的方法:每日一次服用碳酸钙和维生素D3 补充剂
的剂量以元素钙500 ～ 600 mg 和维生素D3 200 IU 为宜，比较适合中国成年人群预防骨质疏松症。如果需要补充更多剂量的
钙和维生素D，必须分成多次服用。这些信息供我国临床医师参考使用。
关键词: 钙;维生素D;临床指南;骨质疏松症;骨质疏松性骨折
The progre ss in global clinical guide line s on the pre ve ntion and tre atme nt of oste oporosis using calcium and vitamin D supple me ntation
Abstract: Calcium and vitamin D supplementation has alw ays been the basic strategy of clinical guidelines for the prevention and
treatment of osteoporosis from global authoritative academic institutions． Ｒecently，how ever，several studies had different view s on
the benefits of excessive calcium and vitamin D supplementation on healthy bones，thus conflicted w ith the internationally recognized
basic strategy for the prevention and treatment of osteoporosis，and confused clinicians． Therefore，w e review ed the clinical
guidelines of authoritative academic institutions in China，USA，Canada，UK，Poland，Japan and Korea，and highlighted that the
receptor activator of nuclear factor-kb (ＲANK)，its ligand ＲANKL，and a decoy receptor，osteoprotegerin (OPG) have emerged
as major local regulators of bone remodeling in the pathogenesis and pathophysiology of osteoporosis． We listed the recommended
calcium and vitamin D intakes for the prevention and treatment of osteoporosis，and summarized that both calcium and vitamin D
supplementation are beneficial in the prevention and treatment of osteoporosis and osteoporotic fractures． The controversy of calcium
and vitamin D supplementation for healthy bones w as analyzed，and in general healthy people taking calcium supplements should not
exceed 1000 mg daily． Based especially on calcium intake，serum lipid levels，BM I and vitamin D nutritional status in Asians
(Chinese，Japanese and Korean )，referencing to Chinese，Japanese and Korean clinical guidelines on calcium and vitamin D
supplementation for the prevention and treatment of osteoporosis，it is suggested calcium and vitamin D supplementation as calcium
carbonate and vitamin D3 supplements (calcium 500 － 600 mg and vitamin D3 200 IU daily ) are suitable for the prevention of
osteoporosis in the Chinese population． If you need to add more calcium and vitamin D supplementation，you must divide them into
multiple doses． These information are provided to Chinese clinicians for reference and clinical application．
Ke y words: Calcium; Vitamin D; Clinical guidelines; Osteoporosis; Osteoporotic fracture
钙和维生素D 是骨骼健康的基本组分，补充钙
和维生素D 是防治骨质疏松症的重要的基本策略。
骨质疏松症的常见严重并发症是骨折， 2016 年美国
国家骨质疏松基金会的荟萃分析结论为中老年人服
用钙和维生素D 补充剂能够作为降低骨折风险的
干预措施［1 ］
。然而，近来对补充钙和维生素D 的健
康骨骼益处发生争议，甚至提出过量钙补充对心血
管健康发生潜在的有害影响，从而混淆了临床医师
的视听，干扰了国际公认的防治骨质疏松症的基本
策略。2016 年10 月25 日美国国家骨质疏松基金
会和美国预防心脏病学会的临床指南［2 ］明确指出:
一般健康成年人无论从食物或补充剂补充钙(不管
是否有/ 或没有维生素D)与其心脑血管疾病风险、
死亡率、 全因死亡率都没有任何关系(有益或有
害)。为此，笔者综述近几年来中国、 美国［2-4 ］
、 加拿
大［5 ］
、 英国［6 ］
、 波兰［7 ］
、 日本［8 ］
、 韩国［9 ］等权威学术
机构的补充钙和维生素D 防治骨质疏松症的临床
指南的观点，引用指南中的学术文献，以期介绍骨质
疏松症的发病机制和病理生理学进展，分析补充钙
和维生素D 的健康骨骼争议，指导正确的钙和维生
素D 补充剂量等等，供我国临床医师参考。
1 骨质疏松症的定义
骨质疏松症(osteoporosis，OP)是一种以骨量低
下，骨微结构损坏，导致骨脆性增加，易发生骨折为
特征的全身性骨病。2001 年美国国立卫生研究院
(NIH)提出骨质疏松症是以骨强度下降、 骨折风险
性增加为特征的骨骼系统疾病，骨强度反映骨骼的
两个主要方面，即骨密度(占70% )和骨质量(占
30% )［8 ］
。
2 骨质疏松症的流行病学
骨质疏松症是一种静悄悄的疾病，直到由于轻
微创伤后发生骨折或在某些情况下没有创伤就发生
骨折才知道患有骨质疏松症。
2016 年张智海等［10 ］通过对万方数据与中国知
网数据库检索，对国内已发表骨质疏松症发病率文
献中10011 例男性和12943 例女性分析骨质疏松发
病率的结果:(1)40 ～ 49 岁年龄段:女性为(7. 75 ±
\6. 38)% ，男性为(4. 00 ± 2. 90)% ;(2)50 ～ 59 岁年
龄段:女性为(28. 00 ± 15. 72)% ，男性为(15. 73 ±
\9. 49)% ;(3)60 ～ 69 岁年龄段: 女性为(52. 67 ±
\10. 76)% ，男性为(30. 55 ± 10. 79)% ;(4)70 ～ 79 岁
年龄段: 女性为(79. 45 ± 9. 53)% ，男性为(43. 46 ±
\7. 30)% ;(5 )80 岁以上年龄段: 女性为(89. 55 ±
\1. 04)% ，男性为(66. 19 ± 17. 56)% 。由此可见，中
国男性在各年龄段发病率均低于同年龄段女性发病
率，并随年龄增长骨质疏松症发病率逐渐增多，男性
与女性的每10 年骨质疏松症增长率分别约为15%
和20% ; 40 岁以上人群骨质疏松症发病率为
\24. 62% (约25% )，约1. 4 亿患病人群。
2013 年Svedbom 等报告，在欧盟每年约有350
万新的脆性骨折发生，仅仅在2010 年这些脆性骨折
就产生了37 亿欧元的花费，造成43000 人死亡［11 ］
。
3 骨质疏松症的发病机制和病理生理
\3. 1 骨质疏松症的发病机制
2010 年美国临床内分泌学家协会的临床实践
医学指南［4 ］提出，成年人的低骨量和骨骼脆性可能
是由于青少年时期的低峰值骨量、 以后的骨丢失过
多、 或两者都有的结果。人的一生骨骼总是在不断
变化。童年期和青春期是骨骼的成长时期，骨骼的
大小、 形状和成分都在改变。在青春期结束时骨骺
闭合，骨骼形状和大小的变化完成，随后5 ～ 10 年是
骨骼成长的巩固时期(取决于骨骼的部位)，一直到
达到成年人的峰值骨量，这通常发生在青少年或在
20 多岁时期［12-13 ］
。
大约70% ～ 80% 的峰值骨量是由遗传决定
的［14-17 ］
。许多非遗传因素也有助于骨骼成长，这包
括营养(例如钙、 磷酸盐、 蛋白质和维生素D)、 承重
的活动和儿童成长期和青春期相关的激素。
\3. 2 骨质疏松症的病理生理
一旦达到成年人的峰值骨量维持期后，就进入
骨骼的重建阶段，老骨被新骨替换。骨重建取决于
吸收老骨的破骨细胞和产生新骨的成骨细胞相互作
用。这些细胞的数量和活性，以及全身的激素和局
部的细胞因子参与。最近，受体激活剂核因子kb
(ＲANK)及其配体ＲANKL 和诱饵受体骨保护素
(OPG)，已经成为骨重建的主要局部调节剂［18 ］
。
ＲANKL 是由成骨细胞和基质细胞合成并存在于骨
微环境中，结合ＲANK，在骨髓的破骨细胞前体细胞
中表达，并促进破骨细胞生成。OPG 是也由成骨细中国骨质疏松杂志 2017 年3 月第23 卷第3 期 Chin J Osteoporos，March 2017，Vol 23，No． 3
胞和基质细胞合成并作为ＲANKL 的诱饵受体起作
用，防止ＲANKL 结合至ＲANK。破骨细胞活性的调
节至少部分取决于ＲANKL 和OPG 之间的平衡。
ＲANKL 和OPG 的相对含量是受到全身激素(如雌
激素)、 局部因子(如白细胞介素-6 和肿瘤坏死因
子)，也许还有其他因素控制。刺激能导致特异性
部位骨重建的级联活动触发机制尚不知道。然而，
研究表明健康个人骨重建过程至少50 年是平衡的
(也就是说，骨形成速率等于骨吸收速率)。直到50
岁，骨量一般几乎没有净丢失或净增加。Wnt 信号
是影响成骨细胞骨形成的重要途径。这是复杂的，
涉及许多超越骨骼的生理系统。
妇女整个围绝经期和绝经后初期的激素变化
(直接和间接地)刺激ＲANKL 产生，导致骨丢失加
速。大多数数据表明，骨转换率(和骨质丢失)在最
后一次月经前3 ～ 5 年加速和在最后一次月经后
3 ～ 5年又减缓。随着骨转换率的加速，骨平衡受到
干扰，因为每一次骨重建被激活，其骨量净丢失比增
加多。在这一时期平均骨质流失速率大约每年
1% ， 或者说，在绝经期的过渡阶段平均骨质流失速
率大约10% 。与绝经相关的骨丢失相反，男性和女
性在60 岁开始发生与年龄相关的骨丢失，从而使较
慢的骨质流失速率每年增加大约0. 5% 。虽然年龄
相关的骨丢失涉及的骨重建骨量与绝经相关的骨丢
失发生有相同的不平衡，但是起始过程尚不清晰。
结合由于绝经或老龄化的骨量丢失，骨质量也
有变化。骨质量变化包括松质骨的微结构单位(骨
小梁)破坏、 骨皮层变薄、 骨骼矿化程度降低、 以及
可能其他尚未知道的因素［19 ］
。许多因素包括营养、
维生素D、 运动、 吸烟以及患有其他疾病和使用药物
都可以影响骨丢失速率和骨折风险。在老龄化期间
和骨生长期间营养很重要。尤其是维生素D 缺乏，
无论是单独发生还是伴有更广泛的营养不良，几乎
已经在世界各地普遍发生。虽然严重维生素D 缺
乏会损害骨骼的矿化，但是即使轻度至中度维生素
D 缺乏也会减少钙吸收并可导致甲状旁腺激素
(PTH)介导的骨吸收增加。维生素D 缺陷也能够
引起肌肉力量和人体平衡的损害，导致跌倒风险增
加。大多数骨质疏松性骨折就是跌倒的结果，越来
越多的证据表明，低骨量的患者在更广泛的创伤后
骨折风险也增加［20 ］
。
4 补充钙防治骨质疏松症
\4. 1 钙的生物学作用
373
2014 年美国国家骨质疏松症基金会(NOF)的
预防和治疗骨质疏松症临床医生指南［3 ］认为，人体
99% 的钙储量在骨骼中，终身足够的钙摄入量对于
人体获得理想的峰值骨量和随后维持骨骼健康是必
要的。当外源性钙供应不足时，骨组织从骨骼吸收
出钙，释放到血液，以保持血清钙水平的恒定［3 ］
。
因此，足够的钙摄入量对骨骼健康是很重要的［21 ］
。
基于钙的生物学作用，几项流行病学研究已经证实
钙摄入量和骨密度或骨质量之间存在正相关关
系［22-23 ］
。因此，鼓励足够的钙摄入量或服用钙补充
剂已经成为治疗或预防骨质疏松症的基本策
略［6， 9 ］
。
\4. 2 防治骨质疏松症的钙推荐总摄入量
2014 年美国国家骨质疏松症基金会(NOF)的
预防和治疗骨质疏松症临床医生指南［3 ］认为，足够
的钙摄入量是任何骨质疏松症预防或治疗方案的基
本要求和任何年龄健康骨骼的生活方式。美国国家
骨质疏松症基金会(NOF)与美国医学科学院(IOM)
的膳食钙推荐摄入量一致［3， 24 ］
，2014 年的美国NOF
的临床医生指南［3 ］对绝经后妇女和50 岁及以上的
男性的推荐的膳食总钙摄入量:50 ～ 70 岁男性为
1000 mg / d，男性(70 岁以上)和女性(50 岁以上)为
1200 mg / d，如果饮食摄入量不足，就应该服用钙补
充剂。没有证据表明，超过这些钙的摄入量会赋予
额外的骨强度增加。也没有证据表明，超过1200 ～
1500 mg / d 的摄入量可能增加发生肾结石、 心血管
疾病和中风的风险［25-28 ］
。
2010 年加拿大骨质疏松症诊断和治疗的临床
实践指南［5 ］推荐， 50 岁以上中老年人每日摄入总的
元素钙(通过膳食和补充剂)应该为1200 mg。
2015 年韩国骨矿物研究学会的补充钙和维生
素D 指南［9 ］认为:研究明确显示韩国人钙摄入量增
加能够显著降低骨质疏松症的风险［29-30 ］
。钙摄入
量与所有骨骼部位的骨密度(BMD)值呈正相关关
系。 达到饮食膳食钙总摄入量 800 mg / d ～
1200 mg / d的水平，钙摄入量和骨密度之间的关联增
加［31-32 ］
。中国或日本妇女的短期和长期研究都已
经证明补充钙对骨质流失有预防作用［33-34 ］
。2015
年韩国的补充钙和维生素D 指南［9 ］提出:众所周知
韩国人膳食钙摄入量低。绝经后妇女和50 岁以上
男性每日钙摄入量为800 ～ 1000 mg / d。 当膳食钙摄
入不足时，应考虑服用钙补充剂。
2011 年日本骨质疏松学会、 日本骨矿物研究学
会和日本骨质疏松基金会的预防和治疗骨质疏松症473 的指南［8 ］建议补充钙和维生素D 作为防治骨质疏
松症的基本治疗，每天摄入钙700 ～ 800 mg，以优化
药物治疗的效果。
2013 年中国居民膳食每天钙推荐摄入量:18 ～
50 岁为800 mg，50 岁以上为1000 mg。 中晚期孕妇
和乳母为1000 mg［35 ］
。 2011 年中华医学会骨质疏松
和骨矿盐疾病分会的原发性骨质疏松症诊治指南推
荐，平均每日应补充的元素钙量为500 ～ 600 mg，应
注意避免超大剂量补充钙剂潜在增加肾结石和心血
管疾病的风险。2013 年中国居民膳食指南钙每天
可耐受最高摄入量(UL 值)为2000 mg(4 岁以上到
老年人)［35 ］
。
\4. 3 补充钙的安全问题
众所周知，补充钙具有健康骨骼益处。但是近
来全球各国的指南越来越多地关注过量钙补充对人
类健康(特别是与心血管疾病风险相关)的潜在有
害影响［36-38 ］
。最近几个流行病学调查或荟萃分析
研究提出了在奥克兰钙研究报告后较高钙摄入量增
加心血管事件的风险问题［37-38 ］
。在瑞典，61433 名
妇女乳房X 线照相队列研究中，与每天膳食钙摄入
量为600 ～ 1000 mg / d组，钙摄入量大于1400 mg / d
组全因死亡率和心血管死亡率更高。在包括11 项
前瞻性研究的荟萃分析数据中，每日膳食钙摄入量
高于1200 mg 时，心血管死亡率开始增加［39 ］
。来自
美国国立卫生研究院的AAＲP 饮食和健康前瞻性队
列研究的数据显示，388229 名50 岁以上男性和女
性的前瞻性队列研究，总心血管病死亡率与男性的
总钙摄入量呈U 型相关关联，在钙摄入量1500 mg /
d 和更高时能够观察到总心血管病死亡率增加［40 ］
。
一项关注钙补充与肾结石风险的大型研究［41 ］显示，
钙补充可能恶化高钙尿症，但是研究并没有评估参
与者的肾脏钙消耗。此外，肾结石的绝对风险差别
很小(钙补充剂组为2. 5% ，而对照组为2. 1% )。
另外，这些研究对象的平均总钙摄入量(饮食和补
充剂)高于目前的钙推荐摄入量。一般来说，健康
人不应该每天服用超过1000 mg 的钙补充剂。具有
肾结石病史的患者在决定服用钙补充剂前，应评估
肾结石形成的原因和是否患有高钙尿症。
另一方面，一些观察性研究报道了钙摄入量和
心血管病风险或死亡风险之间缺乏相关关联或甚至
负相关关联，研究表明较高水平的钙摄入量可能减
少心血管病发生或降低死亡率。妇女健康倡议的钙
加维生素D 补充(钙剂服用的是碳酸钙，元素钙量
每天1000 mg;维生素D3 为每天400 IU)的随机试
中国骨质疏松杂志 2017 年3 月第23 卷第3 期 Chin J Osteoporos，March 2017，Vol 23，No． 3
验报道，绝经后妇女服用钙和维生素D 补充剂的7
年期间，既不增加也不降低其冠心病或脑血管意外
的风险［42 ］
。在爱荷华州妇女健康研究中，超过
30000 名绝经后妇女，高膳食或补充剂的钙摄入量
与缺血性心脏病死亡率降低相关［43 ］
。
2016 年10 月25 日美国国家骨质疏松症基金
会和美国预防心脏病学会的临床指南［2 ］明确指出，
缺乏证据将一般健康成年人补充钙和维生素D / 单
独补充钙与心血管疾病联系在一起。临床指南认
为，钙是骨骼中存在的主要矿物质和美国膳食中短
缺钙。对于没有从其膳食中摄入足够量钙的人，笔
者已经推荐服用钙补充剂作为预防骨质疏松症和骨
质疏松性骨折的标准策略。临床指南的立场是:有
中等质量医学证据(B 级)显示一般健康成年人从
食物或钙补充剂补充钙和维生素D / 单独补充钙与
心脑血管疾病的发生风险、 和死亡率或者全因死亡
率都没有任何关系(有益的或有害的)。根据迄今
可获得的证据，从心血管观点来看，从食物和补充剂
摄入钙只要钙摄入量不超过可耐受最高摄入量(由
美国国家医学科学院制定为2000 ～ 2500mg / d)应该
被认为是安全的。
2013 年英国国家骨质疏松指南组(NOGG)更新
的骨质疏松症的诊断和治疗指南［6 ］指出，有研究提
示， 钙补充可能潜在与不良心血管结果相关［27， 37 ］
，
但这些研究已受到广泛批评，这种假定的关联需要
进一步澄清［44 ］
。虽然纵向队列研究也提示钙补充
可能心血管事件的风险，但是这个现象并没有在高
膳食钙摄入者中看到［45 ］
。因此，可能需要谨慎地增
加膳食钙摄入量和单独服用维生素D，应该考虑同
时补充钙和维生素D ［46 ］
。
2011 年日本骨质疏松学会、 日本骨矿物研究学
会和日本骨质疏松基金会的预防和治疗骨质疏松症
的指南［7 ］指出，据报道，钙补充剂可能增加心血管
疾病的风险。但是，相同的膳食钙摄入量尚未显示
增加心血管风险。此外，这些不良结果是从日本外
部报告的，其中钙摄入量、 血清脂质水平和体重指数
(BMI)与日本不同。此时，钙作为药物或补充剂，每
次剂量不应该超过500 mg。
2015 年韩国骨矿物研究学会的补充钙和维生
素D 指南［8 ］指出:膳食钙摄入量低也被认为与心血
管事件或死亡风险增加有关［47 ］
。亚洲人群的膳食
钙摄入量较低， 50 岁以上人群的平均饮食钙摄入量
约为470 mg / d，远低于可比的西方人群［48 ］
。此外，
一些提示较高钙摄入量伴有心血管病发生或死亡率中国骨质疏松杂志 2017 年3 月第23 卷第3 期 Chin J Osteoporos，March 2017，Vol 23，No． 3
增加的研究显示，钙摄入量较低的人群心血管病发
生率或死亡率明显增加或有增加的倾向［36 ］
。
2004 年我国卫生部、 科技部、 国家统计局发布
的中国居民营养与健康现状［49 ］指出:全国城乡居民
膳食钙摄入量仅为391 mg，相当于中国营养学会钙
推荐摄入量(800 ～ 1000 mg)的41% 。因此，中国居
民平均每日补充的元素钙量为500 ～ 600 mg。
5 补充维生素D 防治骨质疏松症
维生素D 在钙吸收、 骨骼健康、 肌肉性能、 人体
平衡和跌倒的风险中起主要作用。维生素D 在骨
骼和矿物质代谢中起着关键作用。维生素D 能增
加肠道吸收钙和磷酸盐，促进骨矿化。维生素D 也
对骨细胞有直接作用［50 ］
。因此，临床上维生素D 不
足会伴有骨质疏松症和骨折，维生素D 缺乏可导致
骨矿化缺陷，造成佝偻病和骨软化症。因此，维持充
足的维生素D 状态是骨骼健康的必要的先决条件。
同时，大量研究证明维生素D 对非骨骼疾病(包括
心血管疾病、 糖尿病、 癌症、 感染和自身免疫性疾
病)具有潜在预防作用［51-57 ］
。
2010 年美国临床内分泌学家协会的临床实践
医学指南［4 ］认为，儿童和成年人预防骨质疏松症，
重要的是要确保其维生素D 的充足。大多数“ 健
康” 成年人血清25(OH)D 低于所期望的值［58 ］
。大
多数天然食物不含有维生素D。主要含有维生素D
的食物:鱼油(包括鳕鱼肝油)、 强化的牛奶、 谷物和
面包。没有涂防晒剂的皮肤在阳光下能够产生维生
素D，但是在北部或南部的冬天是不能产生维生素
D。美国国家科学院推荐:51 ～ 70 岁正常成年人每
天需要维生素D 400 IU 和70 岁以上的正常成年人
需要600 IU。现在许多专家认为这些推荐摄入量太
低［59 ］
。美国国家骨质疏松症基金会推荐:50 岁或
以上的成年人每天需要维生素D 800 ～ 1000 IU，但
许多专家推荐更多:1000 ～ 2000 IU / d(维生素D 的
“ 安全上限” 是4000 IU / d)。居家行动受限的个人、
肠道吸收不良的患者、 或正在接受长期抗惊厥药或
糖皮质激素治疗的患者，尤其存在维生素D 缺乏的
风险的患者，需要相当多的维生素D 补充才能达到
所需的水平。25(OH)D 水平低于30 ng / mL 的患者
补充维生素D 能改善钙吸收分数，但是25 (OH)D
水平高于30 ng / mL 的患者补充维生素D 不能改善
钙吸收分数。一项绝经后妇女研究的荟萃分析发
现， 每天补充维生素D 剂量700 ～ 800 IU 以上的绝
经后妇女髋骨骨折和非椎骨骨折明显减少［60 ］
。
573
美国国家骨质疏松症基金会(NOF)的膳食维
生素D 推荐摄入量比美国医学科学院(IOM )的
高［3， 24 ］
，IOM 的膳食维生素D 推荐摄入量为600
IU / d(直到70 岁) 和800 IU / d (70 岁以上)［24 ］
。
2014 年的美国NOF 的临床医生指南［3 ］指出，适当
提高维生素D 摄入量:800 ～ 1000 IU / d (50 岁以上
中老年人)，如果需要应该服用维生素D 补充剂。
2010 年加拿大骨质疏松症诊断和治疗的临床
实践指南［5 ］指出，维生素D 缺乏风险低的健康成年
人每日常规补充维生素D3 400 ～ 1000 IU (10 ～ 25
μg);维生素D 缺乏中度风险的50 岁以上成年人每
日补充维生素D3 800 ～ 1000 IU(20 ～ 25 μg)。为了
达到最佳的维生素D 状态，可能需要每日补充维生
素D3 1000 IU(25 μg)以上。每日补充维生素D3 剂
量高达2000 IU(50 μg)是安全的，不需要监测。
2011 年日本骨质疏松学会、 日本骨矿物研究学
会和日本骨质疏松基金会的预防和治疗骨质疏松症
的指南［8 ］推荐，每天维生素D 摄入量为400 ～ 800
IU(10 ～ 20 μg)。
2015 年韩国骨矿物研究学会的补充钙和维生
素D 指南［9 ］提出:众所周知韩国人维生素D 缺乏。
建议维生素D 摄入量每天超过800 IU，这似乎可以
降低骨折的风险。
2011 年中华医学会骨质疏松和骨矿盐疾病分
会的原发性骨质疏松症诊治指南推荐每天维生素D
摄入量:成年人为200 IU(5μg)，老年人为400 ～ 800
IU(10 ～ 20 μg)，维生素D 用于治疗骨质疏松症时
为800 ～ 1000 IU(20 ～ 25 μg)。2013 年中国居民膳
食指南［35 ］维生素D 每天可耐受最高摄入量(UL
值)为2000IU(11 岁以上)。
6 同时补充钙和维生素D 防治骨质疏松症
2010 年美国临床内分泌学家协会的临床实践
医学指南［4 ］指出，
“ 骨健康” 的生活方式(足够的膳
食钙和维生素D、 运动、 避免烟草等)对每个人都很
重要(婴儿、 儿童、 青少年、 年轻的成人和骨质疏松
症患者)。其目标为:(1)骨骼成熟期优化骨骼发育
和最大化峰值骨量;(2)预防年龄相关性骨质流失
和继发性骨质流失的原因;(3)保持骨架的结构完
整性;(4)预防骨折。
2010 年美国临床内分泌学家协会的临床实践
医学指南［4 ］认为，老年人钙的需求量增加，因此，老
年人特别容易钙缺乏，其导致钙缺乏的因素包括肠
吸收钙和维生素D 都减少和肾功能不全导致维生673 素D 激活降低。患有胃肠道吸收不良的患者、 正在
服用大剂量糖皮质激素的患者、 胃酸产生减少的患
者(例如，有胃旁路术史、 伴有恶性贫血、 或使用质
子泵抑制剂)、 接受抗癫痫药物的患者，甚至那些无
症状的腹腔疾病患者，都特别容易发生钙和维生素
D 缺乏。那些需要药物治疗的候选患者应考虑实验
室检查来评估钙和维生素D 是否充足。研究已被
证明服用钙补充剂能够轻微增加BMD，但没有科学
证据支持单独服用钙补充剂(而不同时补充维生素
D)能够减少骨折风险。单独服用钙补充剂降低骨
折风险的证据缺乏可能部分归因于研究设计和患者
依从性问题［41， 61 － 63 ］
。
2014 年波兰骨质疏松症的诊断和处理指南［7 ］
指出， 2012 年中欧国家［64 ］
、 2011 年美国［65 ］和2012
年欧洲［66 ］的维生素D 补充的专家共识明确推荐，需
要同时补充钙和维生素D［血清25 (OH)D 水平＞
30ng / ml ］作为预防和药物治疗骨质疏松症必不可少
的标准。
2014 年国家骨质疏松症基金会的临床医生指
南［3 ］认为，所有骨质疏松症患者每日摄入充足的钙
和维生素D 是一种安全和廉价的防治方法，有助于
降低骨折风险。对照的临床试验已经证明，同时补
充钙和维生素D 可以降低骨折的风险［67 ］
。骨质疏
松症常见的并发症骨折，给老年人带来巨大的医疗
负担和个人痛苦，并对国家造成重大的经济损失。
2016 年美国国家骨质疏松基金会［68 ］从PubMed
检索2011 年7 月1 日至2015 年7 月31 日期间的
补充钙和维生素D 与预防骨折的随机对照研究，符
合纳入标准的8 项研究30970 例参与者中，有髋部
骨折195 例和所有骨折2231 例。Meta 分析结果显
示:补充钙和维生素D 能够显著降低总的骨折风险
15% ，其总体相对风险评估［SＲＲE］为0. 85 (95%
CI， 0. 73 ～ 0. 98)和显著降低髋骨骨折风险30% ，其
SＲＲE 为0. 70 (95% CI，0. 56 ～ 0. 87 )。结论:本项
ＲCT 的荟萃分析支持社区和养老院的中老年人服
用钙加维生素D 补充剂作为降低骨折风险的干预
措施。
2014 年Hiligsmann 等［69 ］对老年骨质疏松症患
者(女性和男性)补充钙和维生素D 与未补充治疗
所获得的每个质量调整生命年(QALY)的成本效益
进行比较。结果:补充钙和维生素D 组的成本小于
无补充组治疗骨质疏松性骨折的成本。结论:本研
究提示60 岁以上的骨质疏松症患者(女性和男性)
补充维生素D 和钙成本效益好的。从经济角度来
中国骨质疏松杂志 2017 年3 月第23 卷第3 期 Chin J Osteoporos，March 2017，Vol 23，No． 3
看，年龄超过60 岁的人群(包括正在采取其它骨质
疏松症治疗的患者)都应该服用钙和维生素D 补充
剂。
2010 年美国临床内分泌学家协会的临床实践
医学指南［4 ］关注的妇女健康倡议(WHI)研究显示，
每天补充钙1000 mg 和维生素D 400IU 组的患者髋
骨BMD 有小但显着的增加(1% )［41 ］
。除了维生素
D 的骨骼效应以外，研究也显示补充维生素D 能改
善肌肉力量、 人体平衡能力和降低跌倒风险［70 － 72 ］
，
还能改善存活率［73 ］
。
2013 年英国国家骨质疏松指南组(NOGG)更新
的骨质疏松症的诊断和治疗指南［6 ］指出， 50 岁以上
的绝经后妇女和男性应该纠正钙和维生素D 缺乏。
目前已经广泛向居家不出或生活在住宅区或养老院
的老年人推荐补充钙和维生素D，因为他们常见维
生素D 缺乏和膳食钙摄入量低。经常提倡补充钙
和维生素D 作为其他治疗骨质疏松症的辅助措施。
有研究提示，钙补充可能潜在与不良心血管结果相
关［37， 74 ］
，绝经后妇女的骨质疏松症患者治疗骨质疏
松症的主要药物是双磷酸盐、 狄诺塞麦、 雷奈酸锶、
雷洛昔芬和甲状旁腺激素肽。所有这些药物治疗已
经显示，当患者服用钙和维生素D 补充剂时，都能
够降低椎骨骨折的风险，有些也能够降低非椎骨骨
折的风险，在某些情况下特别降低髋骨骨折的风险。
2015 年韩国骨矿物研究学会考虑到50 岁以上
的韩国男性和绝经后妇女补充钙和维生素D 防治
骨质疏松症的重要性，研究表明，同时补充维生素D
与钙可以降低骨折和跌倒的风险，但单独补充维生
素D 可能无效［27， 75 ］;荟萃分析结果显示， 50 岁或以
上人群补充1200 mg 钙和800 IU 维生素D 的最小
推荐剂量防治骨质疏松症的效果最佳［76 ］
。
7 补充钙和维生素D 防治继发性骨质疏
松症
继发性骨质疏松症是由于疾病、 药物、 器官移植
等原因所致的骨量减少、 骨微结构破坏、 骨脆性增加
和易于骨折的代谢性骨病。
2013 年英国国家骨质疏松指南组(NOGG)更新
的骨质疏松症的诊断和治疗指南［6 ］指出，继发性骨
质疏松症的原因:类风湿关节炎、 男性和女性未治疗
性腺功能减退、 长时间不动、 器官移植、 I 型糖尿病、
甲状腺功能亢进、 胃肠道疾病、 慢性肝病、 慢性阻塞
性肺疾病。特别是糖皮质激素治疗(任何剂量、 口
服3 个月以上)。中国骨质疏松杂志 2017 年3 月第23 卷第3 期 Chin J Osteoporos，March 2017，Vol 23，No． 3
73
2006 年中华医学会骨质疏松和骨矿盐疾病分
会的继发性骨质疏松症诊疗指南(讨论稿)指出，继
发性骨质疏松症的基础治疗之一是适当补充钙和维
生素D 制剂。大部分继发性骨质疏松症(例如，糖
皮质激素性骨质疏松症、 制动性骨质疏松症、 长期肠
外营养支持性骨质疏松症、 糖尿病性骨质疏松症、 器
官移植后骨质疏松症等)除了原发疾病的治疗以
外，可以参考上述原发性骨质疏松症诊疗指南补充
钙和维生素D 制剂。但是，对于少数特殊疾病需要
特别注意:血液透析性骨质疏松症避免使用含铝透
析液和低磷低钙透析液;如果患者伴有高钙血症
(如肿瘤或甲状旁腺功能亢进症者)应该禁忌使用
钙剂及维生素D 制剂;如患者伴有肾结石及高尿
钙，则应慎用钙剂及维生素D 制剂。
2012 年巴西风湿性疾病学会与巴西医学会和
巴西风湿内科医师学会制定的预防和治疗糖皮质激
素性骨质疏松症指南［77 ］推荐，糖皮质激素性骨质疏
松症患者同时补充钙和维生素D 有益于预防骨量
丢失。绝经前妇女服用碳酸钙制剂(钙500 mg / d)
维持正在糖皮质激素治疗的妇女腰椎BMD。补充
钙和维生素D 被认为是低毒性和低成本治疗GIO
的第一步。补充钙和维生素D 能显著改善33% 正
在糖皮质激素治疗的患者腰椎和桡骨BMD;与安慰
剂组减少骨量比较，类风湿性关节炎患者和长期服
用糖皮质激素的患者组补充碳酸钙(1000 mg / d)和
维生素D(500 IU / d)能够增加骨量。
8 钙和维生素D 补充剂
2010 年美国临床内分泌学家协会的临床实践
医学指南［4 ］认为，膳食钙摄入量不足的患者应该改
变他们的饮食或服用钙补充剂。许多钙补充剂中碳
酸钙补充剂通常是最便宜，并且需要服用的片剂数
最少。然而，碳酸钙可能引起胃肠道(GI)不适(便
秘和腹胀)，在胃酸分泌缺乏的患者，必须进食后立
即服用才能充分吸收。柠檬酸钙补充剂价格通常比
碳酸钙补充剂贵，并需要服用更多的片剂才能达到
所需的剂量;然而，柠檬酸钙的吸收不依赖胃酸，并
且很少可能引起胃肠道不适。
2010 年美国临床内分泌学家协会的临床实践
医学指南［4 ］指出，每日补充维生素D2 和D3 似乎同
样有效，但是与间歇补充剂量(每周1 次或每月1
次)，维生素D3 似乎比维生素D2 更有效约3 倍［78 ］
。
2014 年英国国家骨质疏松学会(NOS)公布了
2013 年维生素D 和骨健康:实用临床患者管理指
南［79 ］强调，口服维生素D3 是治疗维生素D 缺乏的
首选方法。治疗采用口服维生素D3 而不是口服维
生素D2 是基于补充后达到的血清25 (OH)D 水平
证据的荟萃分析［80 ］
。
综上所述，近几年来中国、 美国［2-4 ］
、 加拿大［5 ］
、
英国［6 ］
、 波兰［7 ］
、 日本［8 ］
、 韩国［9 ］等的补充钙和维生
素D 防治骨质疏松症的临床指南介绍骨质疏松症
的发病机制和病理生理学进展，分析补充钙和维生
素D 的健康骨骼争议，推荐正确选择钙和维生素D
补充剂量和类型。依据中国、 日本［8 ］和韩国［9 ］指南
的相关部分，每日1 次服用钙剂量为500 ～ 600 mg
和维生素D3 剂量为200 IU 的碳酸钙和维生素D3
补充剂比较适合中国成年人群预防骨质疏松症。如
果需要补充更多剂量的钙和维生素D，必须分成多
次服用。 | # 补充钙和维生素D 防治骨质疏松症的全球临床指南进展
## 摘要
补充钙和维生素D 防治骨质疏松症一直是全球权威学术机构临床指南的基本策略。然而，近来一些研究对过量补充钙和维生素D 的健康骨骼益处提出不同的观点，从而干扰了国际公认的防治骨质疏松症的基本策略，混淆了临床医师的视听。因此，笔者综述近几年来中国、美国、加拿大、英国、波兰、日本、韩国等权威学术机构的临床指南的观点，重点介绍骨质疏松症的发病机制和病理生理学关注的骨重建的主要局部调节剂［受体激活剂核因子kb(ＲANK)及其配体ＲANKL 和诱饵受体骨保护素(OPG);罗列各国为防治骨质疏松症推荐的钙和维生素D 摄入量;联合补充钙和维生素D 有利于防治骨质疏松症和骨质疏松性骨折;分析补充钙和维生素D 的健康骨骼争议，一般健康人每天服用钙补充剂不应该超过1000 mg。特别是基于亚洲人(中国人、日本人、韩国人)的膳食钙摄入量、血脂水平、身体体重指数(BMI)和维生素D 营养状况，参考中国、日本、韩国的补充钙和维生素D 防治骨质疏松症的临床指南，提出钙和维生素D 补充的方法:每日一次服用碳酸钙和维生素D3 补充剂的剂量以元素钙500 ～ 600 mg 和维生素D3 200 IU 为宜，比较适合中国成年人群预防骨质疏松症。如果需要补充更多剂量的钙和维生素D，必须分成多次服用。这些信息供我国临床医师参考使用。
## 关键词
钙;维生素D;临床指南;骨质疏松症;骨质疏松性骨折
## 正文
钙和维生素D 是骨骼健康的基本组分，补充钙和维生素D 是防治骨质疏松症的重要的基本策略。骨质疏松症的常见严重并发症是骨折，2016 年美国国家骨质疏松基金会的荟萃分析结论为中老年人服用钙和维生素D 补充剂能够作为降低骨折风险的干预措施。然而，近来对补充钙和维生素D 的健康骨骼益处发生争议，甚至提出过量钙补充对心血管健康发生潜在的有害影响，从而混淆了临床医师的视听，干扰了国际公认的防治骨质疏松症的基本策略。2016 年10 月25 日美国国家骨质疏松基金会和美国预防心脏病学会的临床指南明确指出:一般健康成年人无论从食物或补充剂补充钙(不管是否有/ 或没有维生素D)与其心脑血管疾病风险、死亡率、全因死亡率都没有任何关系(有益或有害)。为此，笔者综述近几年来中国、美国、加拿大、英国、波兰、日本、韩国等权威学术机构的补充钙和维生素D 防治骨质疏松症的临床指南的观点，引用指南中的学术文献，以期介绍骨质疏松症的发病机制和病理生理学进展，分析补充钙和维生素D 的健康骨骼争议，指导正确的钙和维生素D 补充剂量等等，供我国临床医师参考。
### 1 骨质疏松症的定义
骨质疏松症(osteoporosis，OP)是一种以骨量低下，骨微结构损坏，导致骨脆性增加，易发生骨折为特征的全身性骨病。2001 年美国国立卫生研究院(NIH)提出骨质疏松症是以骨强度下降、骨折风险性增加为特征的骨骼系统疾病，骨强度反映骨骼的两个主要方面，即骨密度(占70% )和骨质量(占30% )。
### 2 骨质疏松症的流行病学
骨质疏松症是一种静悄悄的疾病，直到由于轻微创伤后发生骨折或在某些情况下没有创伤就发生骨折才知道患有骨质疏松症。
2016 年张智海等通过对万方数据与中国知网数据库检索，对国内已发表骨质疏松症发病率文献中10011 例男性和12943 例女性分析骨质疏松发病率的结果:(1)40 ～ 49 岁年龄段:女性为(7. 75 ± 6. 38)% ，男性为(4. 00 ± 2. 90)% ;(2)50 ～ 59 岁年龄段:女性为(28. 00 ± 15. 72)% ，男性为(15. 73 ± 9. 49)% ;(3)60 ～ 69 岁年龄段: 女性为(52. 67 ± 10. 76)% ，男性为(30. 55 ± 10. 79)% ;(4)70 ～ 79 岁年龄段: 女性为(79. 45 ± 9. 53)% ，男性为(43. 46 ± 7. 30)% ;(5 )80 岁以上年龄段: 女性为(89. 55 ± 1. 04)% ，男性为(66. 19 ± 17. 56)% 。由此可见，中国男性在各年龄段发病率均低于同年龄段女性发病率，并随年龄增长骨质疏松症发病率逐渐增多，男性与女性的每10 年骨质疏松症增长率分别约为15% 和20% ; 40 岁以上人群骨质疏松症发病率为24. 62% (约25% )，约1. 4 亿患病人群。
2013 年Svedbom 等报告，在欧盟每年约有350 万新的脆性骨折发生，仅仅在2010 年这些脆性骨折就产生了37 亿欧元的花费，造成43000 人死亡。
### 3 骨质疏松症的发病机制和病理生理
#### 3.1 骨质疏松症的发病机制
2010 年美国临床内分泌学家协会的临床实践医学指南提出，成年人的低骨量和骨骼脆性可能是由于青少年时期的低峰值骨量、以后的骨丢失过多、或两者都有的结果。人的一生骨骼总是在不断变化。童年期和青春期是骨骼的成长时期，骨骼的大小、形状和成分都在改变。在青春期结束时骨骺闭合，骨骼形状和大小的变化完成，随后5 ～ 10 年是骨骼成长的巩固时期(取决于骨骼的部位)，一直到达到成年人的峰值骨量，这通常发生在青少年或在20 多岁时期。
大约70% ～ 80% 的峰值骨量是由遗传决定的。许多非遗传因素也有助于骨骼成长，这包括营养(例如钙、磷酸盐、蛋白质和维生素D)、承重的活动和儿童成长期和青春期相关的激素。
#### 3.2 骨质疏松症的病理生理
一旦达到成年人的峰值骨量维持期后，就进入骨骼的重建阶段，老骨被新骨替换。骨重建取决于吸收老骨的破骨细胞和产生新骨的成骨细胞相互作用。这些细胞的数量和活性，以及全身的激素和局部的细胞因子参与。最近，受体激活剂核因子kb(ＲANK)及其配体ＲANKL 和诱饵受体骨保护素(OPG)，已经成为骨重建的主要局部调节剂。ＲANKL 是由成骨细胞和基质细胞合成并存在于骨微环境中，结合ＲANK，在骨髓的破骨细胞前体细胞中表达，并促进破骨细胞生成。OPG 是也由成骨细胞和基质细胞合成并作为ＲANKL 的诱饵受体起作用，防止ＲANKL 结合至ＲANK。破骨细胞活性的调节至少部分取决于ＲANKL 和OPG 之间的平衡。ＲANKL 和OPG 的相对含量是受到全身激素(如雌激素)、局部因子(如白细胞介素-6 和肿瘤坏死因子)，也许还有其他因素控制。刺激能导致特异性部位骨重建的级联活动触发机制尚不知道。然而，研究表明健康个人骨重建过程至少50 年是平衡的(也就是说，骨形成速率等于骨吸收速率)。直到50 岁，骨量一般几乎没有净丢失或净增加。Wnt 信号是影响成骨细胞骨形成的重要途径。这是复杂的，涉及许多超越骨骼的生理系统。
妇女整个围绝经期和绝经后初期的激素变化(直接和间接地)刺激ＲANKL 产生，导致骨丢失加速。大多数数据表明，骨转换率(和骨质丢失)在最后一次月经前3 ～ 5 年加速和在最后一次月经后3 ～ 5年又减缓。随着骨转换率的加速，骨平衡受到干扰，因为每一次骨重建被激活，其骨量净丢失比增加多。在这一时期平均骨质流失速率大约每年1% ， 或者说，在绝经期的过渡阶段平均骨质流失速率大约10% 。与绝经相关的骨丢失相反，男性和女性在60 岁开始发生与年龄相关的骨丢失，从而使较慢的骨质流失速率每年增加大约0. 5% 。虽然年龄相关的骨丢失涉及的骨重建骨量与绝经相关的骨丢失发生有相同的不平衡，但是起始过程尚不清晰。
结合由于绝经或老龄化的骨量丢失，骨质量也有变化。骨质量变化包括松质骨的微结构单位(骨小梁)破坏、骨皮层变薄、骨骼矿化程度降低、以及可能其他尚未知道的因素。许多因素包括营养、维生素D、运动、吸烟以及患有其他疾病和使用药物都可以影响骨丢失速率和骨折风险。在老龄化期间和骨生长期间营养很重要。尤其是维生素D 缺乏，无论是单独发生还是伴有更广泛的营养不良，几乎已经在世界各地普遍发生。虽然严重维生素D 缺乏会损害骨骼的矿化，但是即使轻度至中度维生素D 缺乏也会减少钙吸收并可导致甲状旁腺激素(PTH)介导的骨吸收增加。维生素D 缺陷也能够引起肌肉力量和人体平衡的损害，导致跌倒风险增加。大多数骨质疏松性骨折就是跌倒的结果，越来越多的证据表明，低骨量的患者在更广泛的创伤后骨折风险也增加。
### 4 补充钙防治骨质疏松症
#### 4.1 钙的生物学作用
2014 年美国国家骨质疏松症基金会(NOF)的预防和治疗骨质疏松症临床医生指南认为，人体99% 的钙储量在骨骼中，终身足够的钙摄入量对于人体获得理想的峰值骨量和随后维持骨骼健康是必要的。当外源性钙供应不足时，骨组织从骨骼吸收出钙，释放到血液，以保持血清钙水平的恒定。因此，足够的钙摄入量对骨骼健康是很重要的。基于钙的生物学作用，几项流行病学研究已经证实钙摄入量和骨密度或骨质量之间存在正相关关系。因此，鼓励足够的钙摄入量或服用钙补充剂已经成为治疗或预防骨质疏松症的基本策略。
#### 4.2 防治骨质疏松症的钙推荐总摄入量
2014 年美国国家骨质疏松症基金会(NOF)的预防和治疗骨质疏松症临床医生指南认为，足够的钙摄入量是任何骨质疏松症预防或治疗方案的基本要求和任何年龄健康骨骼的生活方式。美国国家骨质疏松症基金会(NOF)与美国医学科学院(IOM)的膳食钙推荐摄入量一致，2014 年的美国NOF 的临床医生指南对绝经后妇女和50 岁及以上的男性的推荐的膳食总钙摄入量:50 ～ 70 岁男性为1000 mg / d，男性(70 岁以上)和女性(50 岁以上)为1200 mg / d，如果饮食摄入量不足，就应该服用钙补充剂。没有证据表明，超过这些钙的摄入量会赋予额外的骨强度增加。也没有证据表明，超过1200 ～ 1500 mg / d 的摄入量可能增加发生肾结石、心血管疾病和中风的风险。
2010 年加拿大骨质疏松症诊断和治疗的临床实践指南推荐，50 岁以上中老年人每日摄入总的元素钙(通过膳食和补充剂)应该为1200 mg。
2015 年韩国骨矿物研究学会的补充钙和维生素D 指南认为:研究明确显示韩国人钙摄入量增加能够显著降低骨质疏松症的风险。钙摄入量与所有骨骼部位的骨密度(BMD)值呈正相关关系。达到饮食膳食钙总摄入量800 mg / d ～ 1200 mg / d的水平，钙摄入量和骨密度之间的关联增加。中国或日本妇女的短期和长期研究都已经证明补充钙对骨质流失有预防作用。2015 年韩国的补充钙和维生素D 指南提出:众所周知韩国人膳食钙摄入量低。绝经后妇女和50 岁以上男性每日钙摄入量为800 ～ 1000 mg / d。当膳食钙摄入不足时，应考虑服用钙补充剂。
2011 年日本骨质疏松学会、日本骨矿物研究学会和日本骨质疏松基金会的预防和治疗骨质疏松症的指南建议补充钙和维生素D 作为防治骨质疏松症的基本治疗，每天摄入钙700 ～ 800 mg，以优化药物治疗的效果。
2013 年中国居民膳食每天钙推荐摄入量:18 ～ 50 岁为800 mg，50 岁以上为1000 mg。中晚期孕妇和乳母为1000 mg。2011 年中华医学会骨质疏松和骨矿盐疾病分会的原发性骨质疏松症诊治指南推荐，平均每日应补充的元素钙量为500 ～ 600 mg，应注意避免超大剂量补充钙剂潜在增加肾结石和心血管疾病的风险。2013 年中国居民膳食指南钙每天可耐受最高摄入量(UL 值)为2000 mg(4 岁以上到老年人)。
#### 4.3 补充钙的安全问题
众所周知，补充钙具有健康骨骼益处。但是近来全球各国的指南越来越多地关注过量钙补充对人类健康(特别是与心血管疾病风险相关)的潜在有害影响。最近几个流行病学调查或荟萃分析研究提出了在奥克兰钙研究报告后较高钙摄入量增加心血管事件的风险问题。在瑞典，61433 名妇女乳房X 线照相队列研究中，与每天膳食钙摄入量为600 ～ 1000 mg / d组，钙摄入量大于1400 mg / d 组全因死亡率和心血管死亡率更高。在包括11 项前瞻性研究的荟萃分析数据中，每日膳食钙摄入量高于1200 mg 时，心血管死亡率开始增加。来自美国国立卫生研究院的AAＲP 饮食和健康前瞻性队列研究的数据显示，388229 名50 岁以上男性和女性的前瞻性队列研究，总心血管病死亡率与男性的总钙摄入量呈U 型相关关联，在钙摄入量1500 mg / d 和更高时能够观察到总心血管病死亡率增加。一项关注钙补充与肾结石风险的大型研究显示，钙补充可能恶化高钙尿症，但是研究并没有评估参与者的肾脏钙消耗。此外，肾结石的绝对风险差别很小(钙补充剂组为2. 5% ，而对照组为2. 1% )。另外，这些研究对象的平均总钙摄入量(饮食和补充剂)高于目前的钙推荐摄入量。一般来说，健康人不应该每天服用超过1000 mg 的钙补充剂。具有肾结石病史的患者在决定服用钙补充剂前，应评估肾结石形成的原因和是否患有高钙尿症。
另一方面，一些观察性研究报道了钙摄入量和心血管病风险或死亡风险之间缺乏相关关联或甚至负相关关联，研究表明较高水平的钙摄入量可能减少心血管病发生或降低死亡率。妇女健康倡议的钙加维生素D 补充(钙剂服用的是碳酸钙，元素钙量每天1000 mg;维生素D3 为每天400 IU)的随机试验报道，绝经后妇女服用钙和维生素D 补充剂的7 年期间，既不增加也不降低其冠心病或脑血管意外的风险。在爱荷华州妇女健康研究中，超过30000 名绝经后妇女，高膳食或补充剂的钙摄入量与缺血性心脏病死亡率降低相关。
2016 年10 月25 日美国国家骨质疏松症基金会和美国预防心脏病学会的临床指南明确指出，缺乏证据将一般健康成年人补充钙和维生素D / 单独补充钙与心血管疾病联系在一起。临床指南认为，钙是骨骼中存在的主要矿物质和美国膳食中短缺钙。对于没有从其膳食中摄入足够量钙的人，笔者已经推荐服用钙补充剂作为预防骨质疏松症和骨质疏松性骨折的标准策略。临床指南的立场是:有中等质量医学证据(B 级)显示一般健康成年人从食物或钙补充剂补充钙和维生素D / 单独补充钙与心脑血管疾病的发生风险、和死亡率或者全因死亡率都没有任何关系(有益的或有害的)。根据迄今可获得的证据，从心血管观点来看，从食物和补充剂摄入钙只要钙摄入量不超过可耐受最高摄入量(由美国国家医学科学院制定为2000 ～ 2500mg / d)应该被认为是安全的。
2013 年英国国家骨质疏松指南组(NOGG)更新的骨质疏松症的诊断和治疗指南指出，有研究提示，钙补充可能潜在与不良心血管结果相关，但这些研究已受到广泛批评，这种假定的关联需要进一步澄清。虽然纵向队列研究也提示钙补充可能心血管事件的风险，但是这个现象并没有在高膳食钙摄入者中看到。因此，可能需要谨慎地增加膳食钙摄入量和单独服用维生素D，应该考虑同时补充钙和维生素D。
2011 年日本骨质疏松学会、日本骨矿物研究学会和日本骨质疏松基金会的预防和治疗骨质疏松症的指南指出，据报道，钙补充剂可能增加心血管疾病的风险。但是，相同的膳食钙摄入量尚未显示增加心血管风险。此外，这些不良结果是从日本外部报告的，其中钙摄入量、血清脂质水平和体重指数(BMI)与日本不同。此时，钙作为药物或补充剂，每次剂量不应该超过500 mg。
2015 年韩国骨矿物研究学会的补充钙和维生素D 指南指出:膳食钙摄入量低也被认为与心血管事件或死亡风险增加有关。亚洲人群的膳食钙摄入量较低，50 岁以上人群的平均饮食钙摄入量约为470 mg / d，远低于可比的西方人群。此外，一些提示较高钙摄入量伴有心血管病发生或死亡率增加的研究显示，钙摄入量较低的人群心血管病发生率或死亡率明显增加或有增加的倾向。
2004 年我国卫生部、科技部、国家统计局发布的中国居民营养与健康现状指出:全国城乡居民膳食钙摄入量仅为391 mg，相当于中国营养学会钙推荐摄入量(800 ～ 1000 mg)的41% 。因此，中国居民平均每日补充的元素钙量为500 ～ 600 mg。
### 5 补充维生素D 防治骨质疏松症
维生素D 在钙吸收、骨骼健康、肌肉性能、人体平衡和跌倒的风险中起主要作用。维生素D 在骨骼和矿物质代谢中起着关键作用。维生素D 能增加肠道吸收钙和磷酸盐，促进骨矿化。维生素D 也对骨细胞有直接作用。因此，临床上维生素D 不足会伴有骨质疏松症和骨折，维生素D 缺乏可导致骨矿化缺陷，造成佝偻病和骨软化症。因此，维持充足的维生素D 状态是骨骼健康的必要的先决条件。同时，大量研究证明维生素D 对非骨骼疾病(包括心血管疾病、糖尿病、癌症、感染和自身免疫性疾病)具有潜在预防作用。
2010 年美国临床内分泌学家协会的临床实践医学指南认为，儿童和成年人预防骨质疏松症，重要的是要确保其维生素D 的充足。大多数“健康”成年人血清25(OH)D 低于所期望的值。大多数天然食物不含有维生素D。主要含有维生素D 的食物:鱼油(包括鳕鱼肝油)、强化的牛奶、谷物和面包。没有涂防晒剂的皮肤在阳光下能够产生维生素D，但是在北部或南部的冬天是不能产生维生素D。美国国家科学院推荐:51 ～ 70 岁正常成年人每天需要维生素D 400 IU 和70 岁以上的正常成年人需要600 IU。现在许多专家认为这些推荐摄入量太低。美国国家骨质疏松症基金会推荐:50 岁或以上的成年人每天需要维生素D 800 ～ 1000 IU，但许多专家推荐更多:1000 ～ 2000 IU / d(维生素D 的“安全上限”是4000 IU / d)。居家行动受限的个人、肠道吸收不良的患者、或正在接受长期抗惊厥药或糖皮质激素治疗的患者，尤其存在维生素D 缺乏的风险的患者，需要相当多的维生素D 补充才能达到所需的水平。25(OH)D 水平低于30 ng / mL 的患者补充维生素D 能改善钙吸收分数，但是25 (OH)D 水平高于30 ng / mL 的患者补充维生素D 不能改善钙吸收分数。一项绝经后妇女研究的荟萃分析发现，每天补充维生素D 剂量700 ～ 800 IU 以上的绝经后妇女髋骨骨折和非椎骨骨折明显减少。
美国国家骨质疏松症基金会(NOF)的膳食维生素D 推荐摄入量比美国医学科学院(IOM )的高，IOM 的膳食维生素D 推荐摄入量为600 IU / d(直到70 岁) 和800 IU / d (70 岁以上)。2014 年的美国NOF 的临床医生指南指出，适当提高维生素D 摄入量:800 ～ 1000 IU / d (50 岁以上中老年人)，如果需要应该服用维生素D 补充剂。
2010 年加拿大骨质疏松症诊断和治疗的临床实践指南指出，维生素D 缺乏风险低的健康成年人每日常规补充维生素D3 400 ～ 1000 IU (10 ～ 25 μg);维生素D 缺乏中度风险的50 岁以上成年人每日补充维生素D3 800 ～ 1000 IU(20 ～ 25 μg)。为了达到最佳的维生素D 状态，可能需要每日补充维生素D3 1000 IU(25 μg)以上。每日补充维生素D3 剂量高达2000 IU(50 μg)是安全的，不需要监测。
2011 年日本骨质疏松学会、日本骨矿物研究学会和日本骨质疏松基金会的预防和治疗骨质疏松症的指南推荐，每天维生素D 摄入量为400 ～ 800 IU(10 ～ 20 μg)。
2015 年韩国骨矿物研究学会的补充钙和维生素D 指南提出:众所周知韩国人维生素D 缺乏。建议维生素D 摄入量每天超过800 IU，这似乎可以降低骨折的风险。
2011 年中华医学会骨质疏松和骨矿盐疾病分会的原发性骨质疏松症诊治指南推荐每天维生素D 摄入量:成年人为200 IU(5μg)，老年人为400 ～ 800 IU(10 ～ 20 μg)，维生素D 用于治疗骨质疏松症时为800 ～ 1000 IU(20 ～ 25 μg)。2013 年中国居民膳食指南维生素D 每天可耐受最高摄入量(UL 值)为2000IU(11 岁以上)。
### 6 同时补充钙和维生素D 防治骨质疏松症
2010 年美国临床内分泌学家协会的临床实践医学指南指出，“骨健康”的生活方式(足够的膳食钙和维生素D、运动、避免烟草等)对每个人都很重要(婴儿、儿童、青少年、年轻的成人和骨质疏松症患者)。其目标为:(1)骨骼成熟期优化骨骼发育和最大化峰值骨量;(2)预防年龄相关性骨质流失和继发性骨质流失的原因;(3)保持骨架的结构完整性;(4)预防骨折。
2010 年美国临床内分泌学家协会的临床实践医学指南认为，老年人钙的需求量增加，因此，老年人特别容易钙缺乏，其导致钙缺乏的因素包括肠吸收钙和维生素D 都减少和肾功能不全导致维生素D 激活降低。患有胃肠道吸收不良的患者、正在服用大剂量糖皮质激素的患者、胃酸产生减少的患者(例如，有胃旁路术史、伴有恶性贫血、或使用质子泵抑制剂)、接受抗癫痫药物的患者，甚至那些无症状的腹腔疾病患者，都特别容易发生钙和维生素D 缺乏。那些需要药物治疗的候选患者应考虑实验室检查来评估钙和维生素D 是否充足。研究已被证明服用钙补充剂能够轻微增加BMD，但没有科学证据支持单独服用钙补充剂(而不同时补充维生素D)能够减少骨折风险。单独服用钙补充剂降低骨折风险的证据缺乏可能部分归因于研究设计和患者依从性问题。
2014 年波兰骨质疏松症的诊断和处理指南指出，2012 年中欧国家、2011 年美国 |
{
"text": "{\"text\":\"DOI : 10.3969/j. issn. 1674-2591. 2011. 02. 014\\n\\n《编者按》:由国际著名维生素 D 研究专家 Holick MF 教授牵头的专家组为美国内分泌学会 临床学术期刊 JCEM 撰写的\\\"维生素 D 缺乏的评价、预防及治疗\\\"(指南)一文,于2011年6月6 日首先以电子版发表,本文做了重点节译,供读者参考。这是一篇基于循证医学和流行病学研究 资料基础上所成之文,对内分泌临床及其他有关涉及骨矿盐代谢疾病领域的专业人员都有重要 的实际指导意义。但也应看到,该文所依据的资料大都来自对西方人群的研究,文内提出的对维 生素 D 缺乏诊断依据、标准,以及对预防及治疗时所设定的剂量范围,是否完全适合东方人群, 特别是我国人群,尚需我们自己的工作进一步阐明。又如,作者认为维生素 D2和Dg的实际应用 效果是一样的,但有研究资料显示,在调控人类骨钙代谢效能方面,D。较 D,强2~4倍。对此如 何看待?似值得考虑。\\n\\n# 维生素 D 缺乏的评价、预防及治疗 -- 内分泌学会临床实践指南\\n\\n近年来,维生素D不足或缺乏现象引起了大 家的关注与重视,如何规范地补充维生素 D 制剂, 也引起了学术界的广泛兴趣。2011年 Michael F. Holick 等多名专家联合制定了关于评价、治疗及预 防维生素 D 缺乏的内分泌学会临床实践指南,于 2011年6月6日发表在《临床内分泌代谢杂志》 ( Journal of Clinical Endocrinology and Metabotism , JCEM)。该临床实践指南主要包括了维生素 D 缺 乏的诊断、维生素 D 缺乏高危人群饮食摄入维生 素 D的建议、维生素 D 缺乏的防治策略及维生素 D 调节钙代谢以外的益处。现简要介绍如下:\\n\\n## 维生素 D 缺乏的诊断\\n\\n该指南建议应针对可能存在维生素 D 缺乏风 险的人群进行筛查,推荐采用可靠的分析方法来测 量血清25 羟D(250HD) 浓度,以评价患者是否 存在维生素 D 缺乏。维生素 D 缺乏是指 250HD 浓 度低于20 ng/mL(50 nmol/L)。1,25-双羟维生素 D 浓度检测只用于一定的疾病状态,如获得性及遗 传性维生素 D 代谢异常疾病及磷代谢异常的疾病。\\n\\n维生素 D 缺乏高危人群营养摄入维生素 D 的 建议 (表1)\\n\\n指南建议0~1 岁的婴幼儿应至少补充维生素 D 400 U/d (1 U=25 ng), 1 岁以上儿童,应至少 摄入600 U/d维生素 D,以使骨骼最大程度的获益。 是否此剂量能使婴幼儿与儿童获得足够的骨骼外益 处,尚不清楚。此外,如果要使血清250HD 浓度 持续大于 30 ng/mL (75 nmol/L),可能需要至少 补充1 000 U/d的维生素 D。\\n\\n·泽\\n\\n文 ·\\n\\n指南建议 19 ~ 50 岁的成年人应至少摄人 600 U/d的维生素 D,以使骨骼和肌肉最大程度的 获益。此剂量是否足以获得维生素D 相关的骨骼 外益处,尚不清楚。然而,如果要使得成年人血清 250HD 浓度持续大于30 ng/mL,可能需要至少补 充1500~2000 U/d的维生素 D。\\n\\n对于 50~70 岁及 >70 岁的人群,指南建议至 少分别摄人600U或 800 U/d的维牛素 D,600~ 800 U/d的维生素 D 是否能够获得维生素 D 相关的 骨骼外益处,尚不清楚。然而,要使得成年人血清 250HD 浓度持续大于30 ng/mL,可能需要至少摄 入1 500~2 000 U/d的维生素 D。\\n\\n对于妊娠期及哺乳期的妇女,指南建议至少补 充600 U/d的维生素 D,要使血清 250HD 浓度超 过 30 ng/mL,至少需要补充 1 500~2 000 U/d的 维生素 D。\\n\\n指南建议对于肥胖的儿童及成年人、以及服用 抗癫痫药物、糖皮质激素、酮康唑等抗真菌药物和 治疗艾滋病药物者,至少需要补充2~3 倍剂量的 维生素 D,以满足机体维生素 D 的需要量。\\n\\n指南建议应使维生素 D 水平保持在可接受范 围的上限,在缺乏医疗监督的情况下,维生素 D 剂量不能超过如下范围:6月龄以内的婴儿,最大 饮食补充量为1 000 U/d, 6个月~1 岁的婴儿不超 过1 500 U/d,1~3 岁儿童不超过2 500 U/d,4~ 8 岁儿童不超过3 000 U/d,8 岁以上儿童不超过 4 000 U/d。然而,为了校正维生素 D 缺乏,需要 补充更大剂量的维生素 D,0~1 岁的儿童可以补 充2 000 U/d的维生素 D, 1~18 岁的儿童可以补 充 4 000 U/d的维生素 D, > 19 岁的成人可以补充 10 000 U/d的维生素 D(表2)。\\n\\n## 维生素 D 缺乏的防治策略\\n\\n指南建议采用维生素 D2 或维生素 D2 治疗和预 防维生素 D 缺乏。\\n\\n指南建议对0~1岁婴幼儿维生素D 缺乏者, 采用维生素 D,或维生素 D2 2 000 U/d治疗,或使 用50000U维生素 D2或维生素 D3,每周1次,共 6周,使血250HD 水平达到30 ng/mL以上,之后 用 400 ~ 1 000 U/d维持。\\n\\n指南建议对1~18岁儿童维生素 D 缺乏者, 采用维生素 D,或维生素 D, 2 000 U/d治疗至少 6 周,或采用维生素 D,或维生素 D3 50 000 U,每 周1 次,共6周,使血250HD 水平达到30 ng/mL 以上,之后用600~1000 U/d维持。\\n\\n指南建议所有成人维生素 D 缺乏者,采用维 生素 D,或维生素 D2 50000U,每周1次,共8 周,或采用维生素 D,或维生素 D26000U/d 的等 同剂量,使血250HD水平达到30 ng/mL以上,之 后用1 500~2 000 U/d 维持。\\n\\n指南建议对肥胖、吸收不良综合征和服用影响 维生素 D代谢药物者采用大剂量维生素 D(2~3 倍剂量, 至少 6 000~10 000 U/d) 治疗维生素 D 缺乏,使血250HD 水平达到30 ng/mL以上,之后 用3 000~6 000 U/d 维持。\\n\\n某些患有结节病、结核病等患者,因其体内被 激活的巨噬细胞能以不被调控的方式将 250HD 转变 成1,25(OH)2D,使肠钙吸收和动员骨钙都增加, 产生高钙血症和高尿钙。指南建议对这种肾外异常 产生1,25(OH)2D的患者,在治疗维生素D 缺乏期 间应监测250HD和血清钙水平,预防高钙血症。\\n\\n对于原发性甲状旁腺功能亢进症合并维生素 D 缺乏者,指南建议进行维生素 D 治疗,同时监测 血清钙水平。\\n\\n# 维生素 D 调节钙代谢以外的益处\\n\\n针对维生素 D 的调节骨代谢和钙以外作用, 指南仅推荐使用维生素 D 预防跌倒,并不推荐使 用维生素 D 预防心血管事件、死亡和提高生活质 量。其他骨骼以外的益处尚需进一步评估。\\n\\n| 维生素 D 缺乏的风险人群 | | 生理需要量(U/d)<br>维持最佳骨骼和肌肉功能 | 维牛素 D 补充<br>上限(U/d) | 维持 250HD > 30 ng/mL<br>的补充量 | 纠正维生素 D 缺乏补充<br>上限(U/d) | |\\n|---------------|---------|---------------------------|---------------------|-----------------------------|-------------------------|--|\\n| 0~1 男 | 0~6个月 | 400 | 1 000 | 1 000 | 2 000 | |\\n| | 6~12 个月 | 400 | 1 500 | 1 000 | 2 000 | |\\n| | | 400 | | | | |\\n| 1 ~ 18 岁 | 1~3 岁 | 600 | 2 500 | 1 000 | 4 000 | |\\n| | 4~6岁 | 600 | 3 000 | 1 000 | 4 000 | |\\n| | 9~18 发 | 600 | 4 000 | 1 000 | 4 000 | |\\n| 19 ~ 50 % | | 600 | 4 000 | 1 500~2 000 | 10 000 | |\\n| 51 ~70 岁 | | 600 | 4 000 | 1 500~2 000 | 10 000 | |\\n| >70 岁 | | 800 | 4 000 | 1 500~2 000 | 10 000 | |\\n| 妊娠哺乳妇女 | | 600 | 4 000 | 1 500~2 000 | 10 000 | |\\n\\n营养摄入维生素 D 的建议 表 1\\n\\n(C)1994-2022 China Academic Journal Electronic Publishing House. All rights reserved. http://www.cnki.net\\n\\n· 146 ·\\n\\n| 表2 年生素D 缺乏的治疗剤量 | | | | | | |\\n|-----------------------------------|----------------------------|------|--------------------|--|--|--|\\n| 对象 | 治疗阶段<br>使 250HD > 30 ng/mL | | 维持阶段 | | | |\\n| 0~1 岁 | 2 000 U/d | 6 周 | 400 ~ 1 000 U/d | | | |\\n| | 50 000 U/周 | | | | | |\\n| 【~18岁 | 2 000 U/d | 至少6周 | 600 ~ 1000 U/d | | | |\\n| | 50 000 U/周 | | | | | |\\n| 所有成人 | 50 000 U/周 | 8 周 | 1 500~2 000 U/d | | | |\\n| | 6 000 U/d | | | | | |\\n| 肥胖老<br>吸收不良综合征<br>使用影响维生素 D 代谢的药物 | 6 000 ~ 10 000 U/d | | 3 000 ~ 6 000 U/d | | | |\\n| 伴有肾外1,25(OH)。D生成者 | 注意监测 250HD 和血钙水平 | | 防治维生素 D 治疗期间发生高钙血症 | | | |\\n| 原发性甲状旁腺功能亢进症 | 根据需要补充维生素 D | | 监测血钙水平 | | | |\\n\\n# 测定血 25OHD 的适应证(筛查候选)\\n\\n如果在临床上遇到以下情况,需要考虑进行维 生素 D 营养状态的判断:何偿病、骨软化症、骨 质疏松、慢性肾功能不全、肝衰竭、吸收不良综合 征、囊性纤维化、炎性肠病、Crohn 病、比罗式手 术、放射性肠炎、甲状旁腺功能亢进症、药物、抗 抽搐药、糖皮质激素、AIDS 药物、抗真菌药(如 酮康唑)、消胆胺、非裔美洲人、西班牙人、妊娠 期和哺乳期妇女、有跌倒史的老年人、有低创骨折 的老年人、肥胖的儿童和成人\\n\\nBMI 30 kg/m²、肉芽肿性疾病、结节病、结 核、组织胞浆菌病、球孢子菌病、镀中毒和部分淋 巴瘤。\\n\\n总之,维生素D缺乏非常普遍,对具有高风 险的人群可以考虑测定 250HD 水平判断维生素 D 的营养状态。针对维生素 D的补充,必需强调日\\n\\n光照射是机体维生素D合成的主要来源,如果日 照不足最好从食物中补充足够的维生素 D,但是富 含维生素 D 的食物太少。日照时为了避免黑色素 瘤或其他皮肤癌的风险,可以避开正午时分。居住 在纬度33 度以上地区的人群,需要注意补充维生 素 D。就预防你俊病和骨软化症而言,需要使血液 中维生素 D 水平 >20 ng/mL,为了使维生素 D 发 挥更大的调节钙、骨骼和肌肉的作用,需要将血液 维生素 D 水平维持在 > 30 ng/mlL水平。当维生素 D 水平在30 ng/mL以上,才可能发挥降低肿瘤、2 型糖尿病、心血管疾病和感染等疾病的风险。今后 需要加强研究,以评价补充较大量维生素 D(如 2 000~5 000 U/d)对钙和骨骼以外的益处。\\n\\n> (夏维波 李梅 摘译) (收稿日期:2011-06-10)\"}"
} | ### 维生素 D 缺乏的评价、预防及治疗 -- 内分泌学会临床实践指南
近年来，维生素 D 不足或缺乏现象引起了大家的关注与重视，如何规范地补充维生素 D 制剂，也引起了学术界的广泛兴趣。2011 年 Michael F. Holick 等多名专家联合制定了关于评价、治疗及预防维生素 D 缺乏的内分泌学会临床实践指南，于 2011 年 6 月 6 日发表在《临床内分泌代谢杂志》(Journal of Clinical Endocrinology and Metabolism, JCEM)。该临床实践指南主要包括了维生素 D 缺乏的诊断、维生素 D 缺乏高危人群饮食摄入维生素 D 的建议、维生素 D 缺乏的防治策略及维生素 D 调节钙代谢以外的益处。现简要介绍如下：
#### 维生素 D 缺乏的诊断
该指南建议应针对可能存在维生素 D 缺乏风险的人群进行筛查，推荐采用可靠的分析方法来测量血清 25 羟 D (25OHD) 浓度，以评价患者是否存在维生素 D 缺乏。维生素 D 缺乏是指 25OHD 浓度低于 20 ng/mL (50 nmol/L)。1,25-双羟维生素 D 浓度检测只用于一定的疾病状态，如获得性及遗传性维生素 D 代谢异常疾病及磷代谢异常的疾病。
#### 维生素 D 缺乏高危人群营养摄入维生素 D 的建议
指南建议 0~1 岁的婴幼儿应至少补充维生素 D 400 U/d (1 U=25 ng)，1 岁以上儿童，应至少摄入 600 U/d 维生素 D，以使骨骼最大程度的获益。是否此剂量能使婴幼儿与儿童获得足够的骨骼外益处，尚不清楚。此外，如果要使血清 25OHD 浓度持续大于 30 ng/mL (75 nmol/L)，可能需要至少补充 1 000 U/d 的维生素 D。
指南建议 19 ~ 50 岁的成年人应至少摄入 600 U/d 的维生素 D，以使骨骼和肌肉最大程度的获益。此剂量是否足以获得维生素 D 相关的骨骼外益处，尚不清楚。然而，如果要使得成年人血清 25OHD 浓度持续大于 30 ng/mL，可能需要至少补充 1 500~2 000 U/d 的维生素 D。
对于 50~70 岁及 >70 岁的人群，指南建议至少分别摄入 600 U 或 800 U/d 的维生素 D，600~800 U/d 的维生素 D 是否能够获得维生素 D 相关的骨骼外益处，尚不清楚。然而，要使得成年人血清 25OHD 浓度持续大于 30 ng/mL，可能需要至少摄入 1 500~2 000 U/d 的维生素 D。
对于妊娠期及哺乳期的妇女，指南建议至少补充 600 U/d 的维生素 D，要使血清 25OHD 浓度超过 30 ng/mL，至少需要补充 1 500~2 000 U/d 的维生素 D。
指南建议对于肥胖的儿童及成年人、以及服用抗癫痫药物、糖皮质激素、酮康唑等抗真菌药物和治疗艾滋病药物者，至少需要补充 2~3 倍剂量的维生素 D，以满足机体维生素 D 的需要量。
指南建议应使维生素 D 水平保持在可接受范围的上限，在缺乏医疗监督的情况下，维生素 D 剂量不能超过如下范围：6 月龄以内的婴儿，最大饮食补充量为 1 000 U/d，6 个月~1 岁的婴儿不超过 1 500 U/d，1~3 岁儿童不超过 2 500 U/d，4~8 岁儿童不超过 3 000 U/d，8 岁以上儿童不超过 4 000 U/d。然而，为了校正维生素 D 缺乏，需要补充更大剂量的维生素 D，0~1 岁的儿童可以补充 2 000 U/d 的维生素 D，1~18 岁的儿童可以补充 4 000 U/d 的维生素 D，> 19 岁的成人可以补充 10 000 U/d 的维生素 D。
#### 维生素 D 缺乏的防治策略
指南建议采用维生素 D2 或维生素 D3 治疗和预防维生素 D 缺乏。
指南建议对 0~1 岁婴幼儿维生素 D 缺乏者，采用维生素 D2 或维生素 D3 2 000 U/d 治疗，或使用 50 000 U 维生素 D2 或维生素 D3，每周 1 次，共 6 周，使血 25OHD 水平达到 30 ng/mL 以上，之后用 400 ~ 1 000 U/d 维持。
指南建议对 1~18 岁儿童维生素 D 缺乏者，采用维生素 D2 或维生素 D3 2 000 U/d 治疗至少 6 周，或采用维生素 D2 或维生素 D3 50 000 U，每周 1 次，共 6 周，使血 25OHD 水平达到 30 ng/mL 以上，之后用 600~1 000 U/d 维持。
指南建议所有成人维生素 D 缺乏者，采用维生素 D2 或维生素 D3 50 000 U，每周 1 次，共 8 周，或采用维生素 D2 或维生素 D3 6 000 U/d 的等同剂量，使血 25OHD 水平达到 30 ng/mL 以上，之后用 1 500~2 000 U/d 维持。
指南建议对肥胖、吸收不良综合征和服用影响维生素 D 代谢药物者采用大剂量维生素 D (2~3 倍剂量，至少 6 000~10 000 U/d) 治疗维生素 D 缺乏，使血 25OHD 水平达到 30 ng/mL 以上，之后用 3 000~6 000 U/d 维持。
某些患有结节病、结核病等患者，因其体内被激活的巨噬细胞能以不被调控的方式将 25OHD 转变成 1,25(OH)2D，使肠钙吸收和动员骨钙都增加，产生高钙血症和高尿钙。指南建议对这种肾外异常产生 1,25(OH)2D 的患者，在治疗维生素 D 缺乏期间应监测 25OHD 和血清钙水平，预防高钙血症。
对于原发性甲状旁腺功能亢进症合并维生素 D 缺乏者，指南建议进行维生素 D 治疗，同时监测血清钙水平。
#### 维生素 D 调节钙代谢以外的益处
针对维生素 D 的调节骨代谢和钙以外作用，指南仅推荐使用维生素 D 预防跌倒，并不推荐使用维生素 D 预防心血管事件、死亡和提高生活质量。其他骨骼以外的益处尚需进一步评估。
### 维生素 D 缺乏的风险人群及补充建议
| 维生素 D 缺乏的风险人群 | 生理需要量 (U/d) 维持最佳骨骼和肌肉功能 | 维生素 D 补充上限 (U/d) | 维持 25OHD > 30 ng/mL 的补充量 | 纠正维生素 D 缺乏补充上限 (U/d) |
|------------------------|---------------------------------------|-------------------------|---------------------------------|---------------------------------|
| 0~1 岁 | 400 | 1 000 | 1 000 | 2 000 |
| 1~18 岁 | 600 | 2 500 | 1 000 | 4 000 |
| 19~50 岁 | 600 | 4 000 | 1 500~2 000 | 10 000 |
| 51~70 岁 | 600 | 4 000 | 1 500~2 000 | 10 000 |
| >70 岁 | 800 | 4 000 | 1 500~2 000 | 10 000 |
| 妊娠哺乳妇女 | 600 | 4 000 | 1 500~2 000 | 10 000 |
### 维生素 D 缺乏的治疗剂量
| 对象 | 治疗阶段 使 25OHD > 30 ng/mL | 维持阶段 |
|--------------------------|-----------------------------|-----------------------|
| 0~1 岁 | 2 000 U/d | 400 ~ 1 000 U/d |
| 1~18 岁 | 2 000 U/d | 600 ~ 1 000 U/d |
| 所有成人 | 50 000 U/周 | 1 500~2 000 U/d |
| 肥胖者、吸收不良综合征、使用影响维生素 D 代谢的药物 | 6 000 ~ 10 000 U/d | 3 000 ~ 6 000 U/d |
| 伴有肾外 1,25(OH)2D 生成者 | 注意监测 25OHD 和血钙水平 | 防治维生素 D 治疗期间发生高钙血症 |
| 原发性甲状旁腺功能亢进症 | 根据需要补充维生素 D | 监测血钙水平 |
### 测定血 25OHD 的适应证 (筛查候选)
如果在临床上遇到以下情况，需要考虑进行维生素 D 营养状态的判断：佝偻病、骨软化症、骨质疏松、慢性肾功能不全、肝衰竭、吸收不良综合征、囊性纤维化、炎性肠病、Crohn 病、比罗式手术、放射性肠炎、甲状旁腺功能亢进症、药物、抗抽搐药、糖皮质激素、AIDS 药物、抗真菌药 (如酮康唑)、消胆胺、非裔美洲人、西班牙人、妊娠期和哺乳期妇女、有跌倒史的老年人、有低创骨折的老年人、肥胖的儿童和成人、BMI 30 kg/m²、肉芽肿性疾病、结节病、结核、组织胞浆菌病、球孢子菌病、镀中毒和部分淋巴瘤。
总之，维生素 D 缺乏非常普遍，对具有高风险的人群可以考虑测定 25OHD 水平判断维生素 D 的营养状态。针对维生素 D 的补充，必需强调日光照射是机体维生素 D 合成的主要来源，如果日照不足最好从食物中补充足够的维生素 D，但是富含维生素 D 的食物太少。日照时为了避免黑色素瘤或其他皮肤癌的风险，可以避开正午时分。居住在纬度 33 度以上地区的人群，需要注意补充维生素 D。就预防佝偻病和骨软化症而言，需要使血液中维生素 D 水平 >20 ng/mL，为了使维生素 D 发挥更大的调节钙、骨骼和肌肉的作用，需要将血液维生素 D 水平维持在 > 30 ng/mL 水平。当维生素 D 水平在 30 ng/mL 以上，才可能发挥降低肿瘤、2 型糖尿病、心血管疾病和感染等疾病的风险。今后需要加强研究，以评价补充较大量维生素 D (如 2 000~5 000 U/d) 对钙和骨骼以外的益处。
(夏维波 李梅 摘译) |
Page 1
2025/4/2 13:12
中国儿童维生素A、维生素D临床应用专家共识(2024版)
杂志2024年4月第32卷第4期
Chin J Child Health Care, Apr.2024. Vol. 32. No. 4
中国儿童维生素A、维生素D临床
应用专家共识(2024)
中华预防医学会儿童保健分会
关键词: 维生素A:维生素D:儿童
中图分类号:R179
文献标识码:C 文章编号:1008-6579【2024】04-0000-00
doi:10.11852/zgetbjzz2024-0279
.1.
专家共识
儿童中维生素A和维生素D缺乏很常见,可
能还存在较高比例的维生素A和维生素D同时缺
乏的问题,这些将影响儿童的健康状况、增加对感染
性疾病的易感性(24)。中华预防医学会儿童保健分
会于2021年组织国内相关专家撰写了《中国儿童维
生素A、维生素D临床应用专家共识》(以下简称《共
识》)。
近年的调查结果显示,我国学龄前儿童维生素
A缺乏和边缘缺乏率高,且有随年龄增长而增加的
趋势:0~18岁儿童中维生素D缺乏和不足率较普
遍,儿童中维生素A和维生素D的缺乏存在相关
性。因此,中华预防医学会儿童保健分会再次组织
专家就上述《共识》进行了修订,期待对我国儿童维
生素A、维生素D的临床应用和缺乏的预防提供
参考。
我国儿童维生素A、维生素D营养现况
(一) 维生素A
1 维生素A 营养状况判定指标与标准维生素A
的营养状况判定指标通常采用血清维生素A(视黄
醇)浓度,主要采用2011年世界卫生组织发布的“血
清视黄醇浓度用于确定人群维生素A缺乏的患病
率”中推荐值,即血清维生素A浓度<0.70umol/L
为缺乏, 0.70~1.05mol/L为边缘缺乏, ≥
1.05mol/L为正常。剂量反应试验可检测肝脏中
维生素A储备情况。
2 我国儿童维生素A营养状况 北京大学公共卫
生学院2017年的系统性评价结果显示,我国
0~12岁儿童中维生素A缺乏率为5.16%,边缘缺
乏率为24.29%。其中,5岁以下儿童维生素A缺
乏率为9.23%,边缘缺乏率为31.53%,高于
2010—2013年中国居民营养与健康状况监测报告
的结果(3~5岁儿童维生素A缺乏率为1.5%,边
缘缺乏率为27.8%)[10]。我国中西部地区横断面调
查结果显示,2~6岁儿童随年龄增长维生素A
未达1.05mol/L的比例分别为25.00%、28.92%、
38.38%、42.73%,维生素A边缘性缺乏的比例随
年龄的增长而增加。我国儿童维生素A边缘缺乏
率存在明显的地区和年龄差异。即使在北京这样的
一线城市的调查结果也与之相似。尽管学龄期
儿童维生素A缺乏和边缘缺乏率略低于学龄前期,
但依然需要引起重视。全国中小城市小学生维生素
A的总体缺乏率为9.77%,边缘缺乏率
20.56%[13]
(二) 维生素 D
1 维生素 D 营养状况判定指标与标准 血清
25-(OH)D水平被认为是反映机体维生素D营养状
况的最佳指标,血清25-(OH)D适宜浓度的判定可
参照2016年全球营养性佝偻病管理共识,即血
清25-(OH)D<30nmol/L为维生素D缺乏,30~
50nmol/L为维生素D不足,≥50nmol/L则为适宜。
2 我国儿童维生素D营养状况我国0~18岁健
康儿童中维生素D缺乏和不足较普遍,其中严重
缺乏率为2.46%(1.03%~4.47%)、缺乏率为
21. 57% (13.65% 30.72%), 28.71%
(20.83%~37.35%),婴幼儿组最低 11.06%
(6.07% 17. 30%). 56. 14% (39.54%
~72.07%)。我国7岁以下儿童维生素D缺乏和不
足检出率为14.0%,且随着年龄增加逐渐增高15]。
(三) 维生素A、维生素D缺乏共病问题
虽然多种维生素,如维生素C和E以及B族复
合物在免疫系统中以相对非特异性的方式发挥作
用(16-17),但是维生素A和D以高度特异性的方式影
响机体的免疫反应,因为它们各自的生物活性代谢物
视黄酸和1,25-二羟基维生素D[(1,25-(OH)D)]
具有激素样特性,而且这两种代谢物都是由体内不
同的组织和细胞从其维生素前体合成,并通过与核
激素受体结合远程对靶细胞发挥作用180。儿童中
维生素A和维生素D缺乏很常见,尤其是中低收入
国家的5岁以下患营养不良的儿童中[1.101,还同时
Page 2
2025/4/2 13:12
中国儿童维生素A、维生素D临床应用专家共识(2024版)
.2.
中国儿童保健杂志2024年4月第32卷第4期 Chin J Child Health Care, Apr. 2024, Vol. 32. No. 4
存在较高比例的维生素A和维生素D同时缺乏(共
病)的问题[24]。
维生素D的某些功能似乎与其他微量营养素
具有协同作用,特别是维生素A。这两种维生素
都共同调节机体的多种免疫反应,具有激素样特性
和通过核激素受体介导发挥作用(18)。维生素D缺
乏与小儿感染性疾病易感性增加有关,这与维生素
A的状态是一致的(2),这种情况在儿童营养不良和
疾病双重负担的低收入国家/地区5岁以下儿童中
可能特别严重。巴西涵盖 468名儿童[(13.4±
1.0)月龄]研究结果显示,血清视黄醇浓度与维生
素D呈显著正相关(P=0.001),维生素A浓度每增
加至 1ymol/L,维生素 D 充足率增加至 1.38 倍
(95%CI:1.18~1.61);控制所有协变量后,血清视
黄醇每增加1mol/L 与25-(OH)D浓度增加
0.43nmol/L 相关(95%CI:0.18~0.80)2)。有研
究结果显示,维生素A的存在可以使维生素D的生
物活性提高130%,增强免疫调节作用 2223)。鉴于
我国儿童普遍缺乏维生素A和维生素D,且维生素
A和维生素D存在协同作用,两者联合补充可能有
助于同时改善维生素A和维生素D的营养状况。
二我国儿童维生素A、维生素D缺乏原因
机体在某一阶段,对维生素A、D的需求和供应
平衡是维生素A、D的理想状态[24]
(一) 维生素A、D缺乏原因的分类
1 原发性缺乏原发性缺乏是由于机体摄入足
够的维生素A、D来满足机体的需要。
2 继发性缺乏 继发性缺乏是环境状况或生理状
况导致的对维生素A、D的吸收和利用的障碍,包括
维生素A、D摄取后不能充分吸收利用。
(二) 常见的维生素A、D缺乏原因 维生素A、D
水平的降低是由于有效供应的降低,或因为对维生
素A、D的需求增多所致。
1围生期储存不足维生素A和类胡萝卜素都很
难通过胎盘进入胎儿体内,新生儿血清和肝脏中维
生素A含量明显低于母体,围生期储存不足是导致
新生儿和婴儿期维生素A、D缺乏的重要原因。虽
然胎儿可通过胎盘获得维生素D,但脐血中25-
(OH)D水平仅为母亲的60%~85%,胎儿和新生
儿维生素D水平较低。母亲若有严重营养不良,肝
肾疾病、慢性腹泻时,则子代的维生素D水平更低。
2 生长发育迅速生长发育迅速是导致儿童维生
素A,D缺乏的重要原因。儿童时期体格生长是连
续而不均匀的,具有阶段性,是循序渐进的过程,婴
幼儿期和青春期是体格生长、器官成熟最快的两个
时期,对维生素A、D的需求量显著增加。维生素D
具有许多潜在的生物学功能,参与钙转运蛋白和骨
基质蛋白的转录以及细胞-周期蛋白转录的调节,良
好的维生素D 状态可提高机体的各种生理效能。
因此,及时充足地补充维生素A、D,可为儿童提供
体格生长、神经和心理行为发育的物质基础。维生
素A,D是生命早期“大脑构建营养素”,若供给不足
即可影响脑和神经系统的发育,即使在2周岁后足
量补充也无法完全逆转。儿童期维持充足的维生素
A、D营养水平,不仅是满足现阶段的营养保障,更
为重要的是为成年后潜能的发展奠定物质基础。
3 营养供给不足
3.1 母乳母乳中的维生素A、D具有较好的生物
活性,是婴儿期的重要来源,但受到乳母饮食和健康
的影响[26]。母乳中维生素D的含量无法满足婴儿
的每日需求量。早产儿、双胎儿、低出生体重儿由于
自身储存不足,为了满足后期追赶生长的需要,对维
生素A、D营养的需求明显增高。
3.2 天然食物维生素A存在于动物肝脏和红、
黄色蔬菜中,植物中的维生素A以维生素A原-类
胡萝卜素的形式存在,其转化率存在吸收有饱和现
象,可见膳食维生素A的摄入明显不能满足儿童生
长的需要。学龄前,学龄期及青春期儿童膳食维生
素A摄入量仍较低,同时,天然食物中维生素D的
含量通常较少,且其稳定性易受到高温、强光、强酸
碱等因素的破坏。
3.3 紫外线照射合成 紫外线照射皮肤合成的维
生素D是人体最主要的维生素D来源,其合成与季
节､海拔、纬度､照射强度、户外活动时间,暴露皮肤
的面积,以及空气污染、使用护肤品,不同衣着等因
素有关。
3.4 营养素补充 儿童维生素A边缘缺乏率存在
明显的地区和年龄差异,这与是否服用含维生素A
补充剂以及剂量有关。儿童维生素D缺乏和不足
也与维生素D摄入量不足有关。据调查,随着年
龄的增加,维生素D制剂补充率降低.0~<3岁儿
童维生素D补充率为65.0%,而3~<7岁儿童仅
有23.5%,可能是3~<7岁儿童维生素D缺乏和
不足率上升的原因。常规给予预防剂量维生素D
补充剂有助于改善儿童维生素D营养状况,降低发
生维生素D缺乏的风险[27280。每日持续服用小剂
量维生素A、D补充剂,可维持儿童更好的营养水
平。未规律服用,间断服用,长期停止服用等原因亦
是导致体内维生素A、D低水平的重要原因。
Page 3
2025/4/2 13:12
中国儿童维生素A、维生素D临床应用专家共识(2024版)
中国儿童保健杂志2024年4月第32卷第4期
Chin J Child Health Care, Apr.2024. Vol. 32. No.4
3.
4 疾病的影响
4.1 影响吸收 感染性疾病、消化道疾病、肝胆系
统疾病、肾脏疾病、甲状腺功能亢进等疾病均可影响
维生素A、D的吸收、代谢。而贫血、抽动障碍、孤独
症譜系障碍(autism spectrum disorder, ASD)等患
儿易合并维生素A的缺乏(29]。
4.2 消耗增加感染性疾病患病期间会导致维生
素A、D的大量消耗,尤其是麻疹、猩红热等患儿,体
内维生素A含量显著缺乏,且感染愈严重,维生素
A的耗损愈多。各种原因的发热状态,均增加维生
素A等多种维生素耗用,故而,患儿及时补充维生
素A,可改善疾病预后,并可降低再患感染的风险。
此外,肥胖儿童体内多余脂肪可吸收维生素D,使循
环中的维生素D含量下降,并降低其生物活性,从
而降低体内维生素D的功能。
5 药物的干扰长期服用某些药物,如新霉素､糖
皮质激素、抗惊厥、抗癫痫药物等,可导致维生素A、
D的吸收和代谢出现明显障碍,应充分引起重视。
6 其他因素
6.1 脂肪含量不足维生素A、D均为脂溶性维生
素,若膳食中的脂肪含量不足,可阻碍维生素A、D
的吸收。早产儿、中、重度消瘦患儿对脂肪的吸收能
力较差,易导致吸收不良。
6.2 维生素E的含量维生素E具有抗氧化作
用,能够防止维生素A在肠道内被氧化破坏,同时,
维生素E的低含量可降低维生素A的吸收。
6.3 维生素D受体的多态性(300 儿童青少年血液
中的维生素D含量可受基因多态性的影响,其不同
点位的基因多态性与机体各种代谢性疾病相关,携
带骨代谢障碍或佝偻病易感基因是导致机体内维生
素A,D缺乏的危险因素。
6.4 经济的影响经济欠发达的国家和地区中,许
多儿童未满足最低膳食量摄入和膳食的多样性,明
显降低维生素A、D含量的补充或互补〕。
三维生素A、维生素D缺乏对儿童健康的影响
(一) 维生素A缺乏对儿童健康的危害
维生素A是人体维持机体正常代谢和功能的
脂溶性维生素。维生素A具有维持暗视觉,促进上
皮细胞再生,维护皮肤、黏膜和其他组织表面的完整
性;促进生长发育,增强免疫功能、维护生殖功能、影
响造血的功能32。此外,最新研究发现维生素A还
可以参与肾脏损伤修复 33-340 以及抗癌作用[35-36] 。维
生素A缺乏的临床表现与缺乏阶段和程度有密切
关系。可疑和亚临床维生素A缺乏阶段主要表现
为非特异的临床表现,如肺部及消化道感染、生长发
育受阻、贫血,进而导致儿童发病率和死亡率增高;
临床型维生素A缺乏表现出特异症状,如干眼症、
角膜干燥溃疡,甚至失明。缺乏维生素A可能会导
致小儿麻疹风险增加,肝功能障碍和肾功能障碍。
1 视力障碍 维生素A的活性产物视黄醛是视网
膜视紫红质组成的重要成分,是维持正常暗视觉过
程的重要物质。维生素A缺乏会导致视觉色素缺
乏,早期会引起视杆细胞功能受损,影响暗环境适应
能力;严重者可导致夜盲症,补充维生素A后可恢
复;若维生素A缺乏进一步加重,将导致不可逆性
眼部疾病,如干眼症、毕脱氏斑、角膜炎甚至失
明[37-38]。总之,儿童在生长发育阶段,视力尚未完全
成熟,缺乏维生素A会影响其视力发育(39)。
2 免疫力下降维生素A是主要信号分子维甲酸
的前体,维甲酸是儿童免疫系统发育中不可或缺的
物质。维生素A缺乏会导致上皮受损,抗体生成减
少及免疫功能低下[40-41]。缺乏维生素A的儿童更
容易感染各种病原体,增加儿童肺炎、胃肠道疾
病以及麻疹的患病风险,从而影响其健康成长。
3 生长发育受阻 维生素A对儿童生长发育具有
促进作用,能增强早期成骨细胞分化和抑制骨矿化
来影响不同阶段的骨骼发育。维生素A缺乏不仅
会使儿童骨骼发育异常,如骨密度降低,骨折风
险增加,还会导致生长发育迟缓,身高、体重增长缓
慢,甚至导致发育不良。
4 皮肤病变 维生素A缺乏会抑制黏膜分泌,使
上皮分层和角化,导致皮肤病变,如皮肤干燥、角化
过度等。这些症状在儿童中表现为皮肤粗糙、角化
层增厚等。皮肤黏膜病变导致黏膜屏障受损,诱发
感染 [45-46]_
5 贫血 维生素A缺乏会限制造血基因表达、干
扰红系祖细胞的合成、影响铁调素表达,导致贫
血7)。而且维生素A缺乏导致严重慢性感染,也进
一步加重儿童贫血[48]。
6 心理行为发育研究表明,儿童心理行为发育与
维生素A水平密切相关。在诸多心理行为障碍中,
ASD,注意缺陷多动障碍(attention deficit and hy-
peractive disorder, ADHD)和抽动障碍儿童的维生
素A水平低于正常儿童。当这些儿童补充维生素
A后,其社交能力有明显提升。维生素A缺乏还影
响儿童的免疫系统,而免疫系统功能的改变也被认
为是ADHD 发病风险增加的一个因素[49]。
维生素A缺乏是儿童四大营养缺乏病之一。
目前,低水平的维生素A状态是新生儿中较常见的
Page 4
2025/4/2 13:12
中国儿童维生素A、维生素D临床应用专家共识(2024版)
中国儿童保健杂志2024年4月第32卷第4期 Chin J Child Health Care, Apr.2024. Vol. 32. No. 4
现象,且有较多研究指出新生儿体内维生素A水平
与孕晚期妇女体内的维生素A水平状态相关。
(二) 维生素D缺乏对儿童健康的危害
维生素D的生理功能包括内分泌功能和旁分
泌功能。内分泌功能主要包括小肠钙转运、骨代谢、
肾钙重吸收、胰岛素分泌等;旁分泌功能包括抑制细
胞增殖和免疫调节[50]。维生素D缺乏的危害不仅
限于骨骼系统疾病,如果在生命早期缺乏维生素D,
可能会对儿童和成人未来患病的风险产生长远影
响,如自身免疫性疾病和心血管疾病;近年研究发现
维生素D缺乏还会导致儿童免疫力下降,增加患过
敏性疾病、肺部疾病的风险;甚至会影响儿童心脑血
管的发育,导致儿童心理行为发育异常。
1 骨骼发育不良 维生素D是促进钙、磷吸收和
骨骼发育的关键营养素。维生素D缺乏会使儿童
骨骼钙化不良,导致佝偻病、骨质疏松等骨骼发育问
题,典型症状包括骨骼疼痛、肌肉无力、骨骼畸形(如
腿部弯曲)和生长发育迟缓,严重者可有消化和心肺
功能障碍(51-53_
2 免疫力下降维生素D在免疫系统的多个方面
都发挥着作用,包括调节免疫细胞的功能,增强抗菌
肽的产生、促进T细胞和B细胞的发育,以及增加
抗体的产生等。维生素D缺乏还与儿童呼吸道感
染(respiratory tract infections, RTIs)、肠道感染和
皮肤感染等的发生密切相关50。同时,感染过程
中,病原体和炎症反应会影响维生素D的代谢,进
一步导致维生素D水平降低。维生素D缺乏还与
自身免疫疾病的发生有关。
3 精神神经症状 维生素D在参与脑内神经免疫
调节、神经营养因子和神经传递方面具有重要作用,
因此其缺乏可能会影响与ADHD相关的神经系统
功能。近期研究表明,维生素D缺乏会增加认知功
能低下,社交情感发展受损和ASD的风险[58-59]。
4 过敏维生素D缺乏可能导致免疫系统失衡,
从而增加过敏性疾病的风险。维生素D缺乏的儿
童更容易患过敏性鼻炎、哮喘、湿疹等疾病(58)。
5 肥胖 维生素D在维持胰岛素分泌和胰岛素敏
感性方面具有重要作用,因此维生素D缺乏可能影
响血糖调节,增加糖尿病的风险。维生素D缺乏与
1型糖尿病(type 1 diabetes mellitus,T1DM)的发
生、发展及并发症密切相关。研究表明,患有T1DM
的儿童和青少年25-(OH)D水平较低[59]。
6 其他系统病变 维生素D参与血管内皮功能的
调节,对维持正常血压和心血管健康具有重要作用。
长期维生素D缺乏可能导致心血管疾病,如高血
压、动脉硬化等(60)。儿童期维生素D缺乏会增加成
年后患心血管疾病的风险。此外,维生素D有助于
促进胎儿及新生儿肺部的发展和成熟。所以维生素
D缺乏会导致新生儿发生呼吸窘迫综合征和支气管
肺发育不良]。
(三) 与维生素A和维生素D缺乏共病相关的
疾病
1 生长发育由于微量营养素在骨骼矿化中的作
用,长期严重维生素D缺乏可导致佝偻病[62]。维生
素A对细胞分化至关重要,影响生长、繁殖、免疫反
应和视觉功能,还与腹泻、RTIs 和麻疹等儿童感染
性疾病的发病率和死亡率增加有关[6360。因此维生
素A和维生素D是在生长发育中起重要作用的必
需微量营养素[6]。营养不良儿童通常同时存在维
生素A、维生素D等多种微量营养素缺乏问题[62],
严重影响儿童生长发育。对于这些儿童补充维生素
A、维生素D等多种复合微量营养素将有助于改善
其营养状况,预防营养缺乏病,使其生长发育潜能得
以发挥]。
2 反复呼吸道感染 全球范围儿童中RTIs 相当
常见,特别是在中低收入国家,而且RTIs也是5岁
以下儿童高发病率和死亡率的主要原因[6667)。其中
反复RTIs(recurrent RTI,RRTIs)是一年内多次发
病的常见儿科疾病,低年龄组多发,影响儿童身心健
康,并易导致并发症,如自身免疫性疾病和败血症
等68)。已有研究结果提示,儿童维生素A和维生素
D的营养状况与RRTIs的发生率密切相关 [67.69-72]
例如有调查结果显示,RRTIs患儿的血清维生素A
和维生素D含量显著低于同年龄健康对照组(P<
0.01)(7273),提示对于RRTIs患儿,及时进行膳食调
整且维生素A和维生素D同补对于疾病的恢复是
必要的。
3 孤独症譜系障碍 ASD是一种复杂的神经发育
障碍性疾病(7,遗传和环境因素与该疾病发生发展
有关,其中环境因素可能与某些微量营养素缺
乏有关,如维生素A和维生素D缺乏[7677]或很多情
况下两者同时缺乏(34.78-79),而且ASD患儿的维生素
A和维生素D缺乏率高于对照组,表现为膳食摄入
量和血清含量低于对照组,两者同时缺乏将会加重
ASD患儿的病情[78]
4 其他疾病 在疾病状态下,如生长发育障碍､感
染性疾病、过敏性疾病、神经发育障碍等疾病中,维
生素A、维生素D缺乏的风险更高,这些患儿的维生
素A和维生素D水平显著低于健康儿童,而且这两
种维生素同时缺乏可能会加重疾病病情[78.80-87]。
Page 5
2025/4/2 13:12
中国儿童维生素A、维生素D临床应用专家共识(2024版)
中国儿童保健杂志2024年4月第32卷第4期
Chin J Child Health Care, Apr.2024. Vol. 32. No.4
四维生素A、维生素D缺乏的防治措施
月,在
(一) 维生素A缺乏的防治措施
1 维生素A缺乏的预防
1.1 母乳喂养对孕哺期母亲进行健康教育,婴幼
儿每日膳食中的维生素A摄入量应达到推荐摄入
量(recommended nutrient intake,RNI)。提倡母乳
喂养,从出生后及时添加维生素A。
1.2 辅食添加建议按照辅食添加原则尽早指导
儿童多进食富含维生素A的食物。维生素A在动
物性食物(如乳类、蛋类、动物内脏)、深色蔬菜和水
果(南瓜、胡萝卜、西蓝花、菠菜、芒果和橘子等)中含
量丰富,注意调整膳食结构,适当增加这些食物的数
量是预防维生素A缺乏的有效措施。
1.3 维生素A 补充剂(88-9011)为预防维生素A缺
乏,婴儿出生后1周内应开始补充维生素A1500U/d
(450ug/d),0~1岁1500U/d(450ug/d),1~18岁
1500~2000U/d(450~600gg/d);针对高危因素可
采取维生素A补充、食物强化等策略提高维生素A
摄入量。2)早产儿、低出生体重儿、多胞胎应在出生
后几天补充口服维生素A制剂1500~2000U/d
(450~600ug/d),前3个月按照上限补充,3个月后
可调整为下限。3)RRTIs患儿每日应补充维生素
A2000U/d(600g/d),以促进儿童感染性疾病的
恢复,同时提高免疫力,降低RRTIs 发生风险。4)
慢性腹泻患儿每日应补充维生素A2000U/d
(600ug/d),以补充腹泻期间消耗掉的维生素A,有
利于腹泻症状的恢复,降低腹泻的发生风险。5)缺
铁性贫血及铁缺乏高危风险的儿童,每日应补充维
生素A1500~2000U/d(450~600ug/d),降低铁
缺乏的发生风险,提高缺铁性贫血的治疗效果。6)
其他罹患营养不良的慢性病患儿往往同时存在维生
素A缺乏,建议每日补充维生素A1500~2000U/
d(450~600gg/d),将有助于改善患病儿童的营养状
况、减少维生素A缺乏风险,改善慢性病的预后。
2 维生素A缺乏的治疗
2.1 维生素A缺乏症有临床维生素A缺乏的症
状时,应尽早补充维生素A进行治疗,可使大多数
病理改变逆转或恢复。见表1。
2.2 边缘型和亚临床型维生素A缺乏对于边缘
型和亚临床维生素A缺乏儿童,可采取以下两种方
法中的任何一种[88]:1)普通口服法:口服维生素A
1500~2000U/d(450~600ug/d)至血清维生素A
水平达正常;2)大剂量突击法:1年内口服维生素A
2次,每次10~20万U(30~60mg),间隔6个
人群
表1 维生素A缺乏的治疗与预防补充建议
Tab.1 Recommendations for the treatment and
prevention of vitamin A deficiency
出生后6~60个月
<6个月
6~12个月
>12月~成人
>干眼症
>麻疹
>蛋白质能量营养不良
>HIV母亲所生新生儿
>治疗性补充
>确诊后单剂量,
>24h/2 周各1次
>确诊后单剂量.
>24hl 次
>預防性补充
>每6个月补充1次
>5万 U(15mg)
>10万U(30mg)
>20万U(60mg)
>确诊后单剂量,
>此后每日需要量
>48h内单剂量,年龄
>段适宜的补充量
>此期间不应再投入其他维生素A 制剂。
>(二)维生素D缺乏的防治措施
>1 维生素D缺乏的预防[14.91]
>1.1 户外活动 建议尽早带婴儿到户外活动,逐步
>达到每天1~2h,以散射光为好,裸露皮肤,无玻璃
>阻挡;6个月以下的婴儿应避免在阳光下直晒;儿童
>户外活动时要注意防晒,以防皮肤灼伤。
>1.2 膳食摄入指导儿童多进食含钙丰富的食品,
>如乳类、奶制品、豆制品、海产品等。
1. 3 维生素D制剂1)为预防佝偻病,建议新生
儿出生后1周内开始补充维生素D,400~800U/d
(10~20ug/d),0~1岁400U/d,1~18岁每日400~
800U/d(10~20ug/d),以预防维生素D缺乏及不
足,保证婴幼儿及儿童青少年生长发育所需。针对
高危因素可采取主动阳光照射、维生素D补充、食
物强化等策略提高维生素D摄入量。2)自出生1周
开始,早产儿、低出生体重儿、多胎儿口服维生素D
制剂 800U/d(20ug/d),3个月后改用口服维生素D
制剂400U/d(10ug/d);如果用早产儿配方奶粉者可
口服维生素D 制剂 400U/d。3)RRTIs的患儿,维
生素D 能够有效促进患儿免疫功能的提高,减少
RTls 的发生次数,促进RTIs症状的恢复。建议
RRTIs 患儿应补充维生素D400~800U/d,以促进
疾病恢复,免疫力提高,降低RRTIs 发生风险。4)
建议腹泻病程期间,儿童应补充维生D400~800U/
d,以补充腹泻期间消耗掉的维生素D,有利于腹泻
症状的恢复,降低腹泻的发生风险。5)建议存在缺
铁性贫血及铁缺乏高危风险的儿童,应补充维生素
D 400~800U/d,降低铁缺乏的发生风险,提高缺铁
性贫血的治疗效果。6)营养不良等慢性疾病的儿童
Page 6
2025/4/2 13:12
中国儿童维生素A、维生素D临床应用专家共识(2024版)
.6.
中国儿童保健杂志2024年4月第32卷第4期 Chin J Child Health Care, Apr.2024, Vol. 32, No. 4.
易罹患维生素D缺乏的风险且病情严重程度与维
生素D缺乏程度呈正相关。建议补充维生素D 400~
800U/d,将有助于改善患病儿童的营养状况、减少维
生素D缺乏风险,改善慢性病的预后。
1.4 其他情况 早产儿、低出生体重儿、巨大儿、户
外活动少以及生长过快的儿童在使用维生素D制
剂的同时,根据膳食钙摄入情况酌情补充钙剂,达到
营养素推荐量要求。
2 维生素D缺乏的治疗符合维生素D缺乏性佝
偻病诊断标准的可采用维生素D补充治疗:
2.1 维生素D制剂的补充剂量上,可予每日
疗法或大剂量冲击疗法;在剂型上,可选用口服法或
肌肉注射法;治疗原则以口服为主,口服法比肌肉注
射法可更快提高25-(OH)D水平。维生素D2000U/d
为最小治疗剂量,强调同时补钙,疗程至少3个月。
见表2。
表2 维生素D缺乏性佝偻病的维生素D治疗量(U)[H]
Tab. 2 Therapeutic dose of vitamin D deficiency rickets (U)
年齡
每日剂量 单次剂量 每日维持剂量
不宜采用
<3月龄
2000
400
3月龄~
2000
5万
400
12月龄~
3 000-6 000
15万
600
12岁~
6 000
3万
600
注:治疗3个月后,评估治疗反应,确定是否需要进一步治疗;确
保钙最低摄入量为500mg/d。
口服困难或腹泻等影响吸收时,可肌肉注射维
生素D制剂15~30万U/次,并停用其他维生素D
制剂1个月,用药1个月后应随访。肌肉干预用药
一般只使用1次,如症状、体征均无改善时应考虑
其他疾病引起的佝偻病,需做进一步检查或转诊。
肌肉注射给药方法不宜应用于新生儿。任何一种疗
法之后都需要持续补充预防剂量的维生素D。
2.2 钙剂的补充 早产儿、低出生体重儿、巨大儿、
户外活动少以及生长过快的儿童在使用维生素D
制剂治疗的同时,联合补充钙剂更为合理。
2.3 增加户外活动与阳光照射 户外活动和阳光
照射可以增加皮肤维生素D的合成。夏秋季节多
晒太阳,主动接受阳光照射,这是防治佝偻病的简便
有效措施。强调平均户外活动时间应在1~2h/d。
2.4 膳食的调整 注意膳食结构的平衡,适当添加
和补充含钙丰富的食物,例如牛奶及奶制品、豆制
品､虾皮、紫菜、海带、海产品和蔬菜等,或加钙饼干
等钙强化食品。
五维生素A、维生素D临床应用的安全性
病发生率,促进儿童早期发展为目的,其重点在预
防,而不仅局限于对已发生营养素缺乏人群的矫正。
我国采取的维生素A、D每日补充的方式更适合我
国以维生素A边缘缺乏、维生素D不足为主要流行
病学特征的现况。每日补充的剂量基于中国营养学
会推荐的每日生理需要量,采取预防性补充措施不
会引起维生素A、D中毒的发生,文献中无因每日服
用小剂量维生素A、D补充剂而出现过量和中毒的
报道。
如需使用大剂量维生素A、维生素D用于治疗
时,应充分考虑安全性,并在医生指导、监测下使用。
(一) 维生素A 中毒 92957
人体摄入过量的维生素A可引起中毒综合征,
称维生素A中毒症(vitamin A toxicity)。中毒的原
因主要为一次性进食大量动物肝脏(如鳕鱼肝､狗
肝、鸡肝等)或一次性服用大剂量维生素A制剂(超
过30万U)引起,临床中也出现过因不遵医嘱意外
服用大剂量维生素A制剂而导致过量及中毒的
情况。
维生素A急性中毒的常见症状为颅内压增高,
多表现为头痛、呕吐、烦躁、囟门饱满或闭合时间延
迟等症状和体征。慢性中毒症状的严重程度与摄入
量及个体差异有关。临床表现为食欲下降、体重不
增或反降,可伴低热、多汗、烦躁等,皮肤干燥、鳞片
样脱屑、瘙痒、皮疹、口唇裂,毛发干枯、脱发等,骨
骼肌肉系统多表现为骨骼、肌肉疼痛,幼儿骨骼生长
过快、变脆,易折。神经系统有出现头疼,呕吐、眩
晕､视觉模糊等脑神经受压的症状,也有肝脾增大、
肝纤维化、肝硬化、门静脉高压等。除上述病史、症
状及体征外,X线检查对本病确诊有特殊价值,表现
为管状骨造型失常,骨质吸收,骨折;板改变及软
组织肿胀;骨干处骨膜下新骨形成;颅缝增宽,前囟
饱满扩大。脑脊液压力增高,可达2.55kPa
(260mmH2O),细胞和糖在正常范围,蛋白降低或正
常偏低值。血清维生素A水平显著增高,可达1000
~6000ug/L以上(婴幼儿正常水平为300~500ug/
L)。维生素A中毒的剂量参见表 3。
一旦确诊维生素A中毒症,应立即停服维生素
A,自觉症状常在1~2周内迅速消失;头颅X线征
象可在6周~2个月内恢复正常,血清维生素A可
于数月内维持较高水平,长骨X线征象恢复较慢,
常需半年左右,故应在数月内不再服维生素A,以免
症状复发。及时治疗,一般预后良好。
(二) 维生素D 中毒 93.95.97]
预防性补充干预是以预防营养素缺乏,降低疾
人体摄入过量的维生素D,出现高钙血症、血清
Page 7
2025/4/2 13:12
中国儿童维生素A、维生素D临床应用专家共识(2024版)
中国儿童保健杂志2024年4月第32卷第4期
Chin J Child Health Care, Apr.2024. Vol. 32.No.4.
25-(OH)D>250nmol/L,伴有高钙尿和低甲状旁
腺素血症,称维生素D中毒症(vitamin D toxicity)。
婴幼儿每天摄入2~5万U,连续数周或数月,可
引起中毒。部分患儿对维生素D敏感,每天服用
4000U,经1~3月后也可能出现中毒症状。常见于
未经诊断就给予大剂量突击治疗,滥用维生素D,误
服或误用。
维生素D 中毒的主要症状为恶心、呕吐、厌食、
腹痛、肠蠕动减少、便秘等胃肠道症状,烦渴、多尿、
脱水、血尿、高血钠、低血镁等肾脏代谢失衡的症状,
注意力不集中、头痛、嗜睡等中枢神经系统症状,有
的还表现为血压升高、心动过缓、心电图有ST段抬
高、房室传导阻滞等,肌肉骨骼出现肌肉无力、骨痛、
骨量减少等骨质疏松症。维生素D中毒的剂量参
见表3。
表3 维生素A、维生素D的中毒剂量[92-93]
Tab. 3 Toxic dose of vitamin A and vitamin D
维生素
维生素A
急性中毒
慢性中毒
多因不遵医嘱长期摄入
过量维生素A引起,如
婴幼儿每天摄入5~10
万U,超过6个月。
一次或短时间内连续数次
摄入超大剂量的维生素A,
如婴幼儿一次食入或注射
维生素A 30 万U以上。
维生素D 婴幼儿每天摄入2~5万U,连续数周或数月或每日
2000U/kg,连续1~3个月。
临床发现维生素D过量中毒时,应立即停服维
生素D,如血钙过高应限制钙的摄入,加速钙的排
泄,可口服氢氧化铝、依地酸二钠、泼尼松降低肠钙
的吸收,增加钙的排泄;亦可试用降钙素,同时注意
保持水、电解质的平衡。应避免长期、大剂量服用维
生素D,如需采取大剂量突击治疗,需监测血清 25-
(OH)D水平。
六维生素A、维生素D补充临床应用常见问题
(一) 维生素A 的补充方式
世界卫生组织建议在维生素A缺乏构成公共
卫生问题地区采取每半年一次的大剂量补充方式
(20万U的维生素A),并建议与多种维生素补充策
略一起施用。
理想的维生素A补充方式采用每日生理剂量
的补充,可以通过:1)提高饮食多样化来增加维生素
A的获取途径;2)食用维生素A强化食物或调味
品;3)辅助营养补充剂;4)每日/每周服用维生素A
补充剂等方式。
婴幼儿是预防维生素A缺乏的主要人群,即使
无临床症状的维生素A缺乏或边缘型维生素A缺
乏,也应尽早进行维生素A的补充,建议每日服用
维生素A1500~2000U。
(二) 阳光照射可以补充维生素D
人类皮肤中的7-脱氮胆固醇,经日光中紫外
线照射形成内源性维生素D,是人类维生素D的主
要来源。阳光照射是维生素D水平的重要决定因
素,皮肤合成的维生素D随着皮肤颜色、紫外线辐
射防护(如衣服、遮荫处,防晒霜)、户外时间、纬度、
季节,日照时间、云量、空气污染程度等的不同而变
化。户外活动应考虑到不同季节,不同气候,不同地
区特点进行,接受阳光的皮肤面积逐渐增加。提倡
阳光照射是防治维生素D缺乏的简便有效措施。
平均户外活动时间可在1~2h/d,6个月以下的婴儿
应避免阳光直射。但即使充分暴露在阳光下,儿童
也无法在深秋,冬季和早春期间在皮肤中合成足够
的维生素D。因此,补充安全剂量的维生素D仍是
有效预防儿童维生素D缺乏的主要方式。
(三) 我国儿童采取维生素A、D同补方式的原因
随着维生素A干预策略的开展,典型的维生素
A缺乏已经有了明显改善,但边缘型维生素A缺乏
的问题在我国儿童中依然严峻,对儿童生长发育和
健康造成影响。维生素A补充计划仍然是具有显
著成本效益的合适干预方法。维生素D缺乏的防
治依然是我国不容忽视的公共卫生问题,也是儿科
医疗保健工作者的重要任务。
维生素A和维生素D同为脂溶性维生素,在受
体层面也存在着密切的联系,9顺式-视黄酸可以促
进维生素D受体(VDR)-类视黄醇X受体(RXR)的
二聚体与维生素D反应元件的结合,使维生素 D
更好地发挥生物学活性,在免疫功能、骨骼发育、预
防贫血等诸多方面具有协同作用。RXR占维生素
D.依赖性基因表达的70%~80%,VR与维生素
D.结合要发挥转录的作用,RXR是主要的贡献者。
视黄酸与RXR结合后,更利于VDR-RXR二聚体的
形成,促进维生素D更好地发挥生物学活性。维生
素A的存在可以使维生素D在细胞核内过氧化氢
酶(catalase, CAT)蛋白的抗炎作用提高130%。维
生素A 的存在可以使维生素D的生物活性提高
130%,有研究显示维生素A水平增加1umol/L,维
生素D充足率提高至1.38倍。
在我国,选择维生素A、D同补的方式可以取得
事半功倍的效果,是方便、经济、可及的预防干预
措施。
(四) 维生素A、D如何补充
一项针对6月龄婴儿的调查结果显示,与每日
Page 8
2025/4/2 13:12
.8.
中国儿童维生素A、维生素D临床应用专家共识(2024版)
中国儿童保健杂志2024年4月第32卷第4期 Chin J Child Health Care, Apr.2024, Vol. 32. No. 4.
持续补充维生素A、D相比,无论是维生素A、D与
维生素D交替补充,还是只补充维生素D,均不能改
善血清视黄醇水平。交替或间隔服用维生素AD
补充剂会导致服药依从性降低,增加服药难度,容易
出现漏服和错服的情况。每日持续补充维生素A、
D才能预防维生素A缺乏,每日需补充维生素A
1500~2000U,维生素D400~800U。
(五) 维生素A可以防治感染性疾病
维生素A被称为“抗感染维生素”,维生素A缺
乏会显著损害眼结膜以及呼吸道、胃肠道和泌尿生
殖道中的粘膜上皮屏障,削弱杯状细胞分泌粘液。
维生素A在非特异性免疫和特异性免疫中都发挥
积极的作用。如促进中性粒细胞的生长发育,维持
其杀菌作用;有助于巨噬细胞的吞噬和氧化活性;调
节白介素-12(IL-12)和降低肿瘤坏死因子-a(TNF-a)
的产生;参与辅助性T细胞1(Thl)和辅助性T细
胞2(Th2)的发育和分化等。众多研究发现,在肺
炎､RRTIs、腹泻、麻疹等感染性疾病患儿体内维生
素A水平低于健康儿童。感染性疾病会导致维生
素A被过度消耗,随尿液排出的维生素A的量远远
大于正常儿童,合并发热的患儿维生素A消耗更
多。补充维生素A可分别降低腹泻、麻疹、下呼吸
道感染的死亡率为28%、20%、22%,降低5岁以下
儿童腹泻发生率15%。在儿童肺炎期间服用维生
素A可明显缩短患儿发热、咳嗽、肺部罗音和异常
胸片持续时间,同时缩短住院时间。合理补充维生
素A可以降低儿童感染性疾病的发生和发展。
(六) 维生素D 补充剂优选维生素D制剂
维生素D是植物来源的维生素D,维生素D是
皮肤接受阳光照射合成的维生素D。维生素D 提
高血清 25-(OH)D的效果明显优于维生素D。每
天补充100U维生素D,提升的血清25-(OH)D水
平的效果,维生素D与维生素D相差16.4mmol/L。
另维生素D的半衰期比维生素D短,与维生素 D
结合蛋白的结合力弱。因此,推荐优选维生素D。
(七) 维生素A、D补充到什么年龄
调查显示围生期维生素A、D储存不足,导致
幼儿期维生素A缺乏,学龄前儿童维生素A缺乏/
边缘缺乏率高,且随年龄的增长而增加。中国疾病
预防控制中心开展的我国3~5岁儿童维生素A缺
乏率的调查结果显示:维生素A缺乏率为1.5%,农
村为2.1%,城市为0.8%;边缘缺乏率27.8%,农村
为34.7%,城市为21.4%。全国中小城市小学生维
生素A的总体缺乏率为9.77%,边缘缺乏率为
20.56%,Meta分析显示,我国0~18岁健康儿童维
生素D严重缺乏、缺乏、不足、适宜、过量、中毒发生
率依次为2.46%、21.57%、28.71%、55.80%、
0.30%、0.11%,缺乏率随着年龄增加逐渐增高。无
论维生素A还是维生素D,在我国儿童中的缺乏现
状均以不足(边缘缺乏)为主要形式。维生素A缺
乏(5.16%)及边缘缺乏(24.29%)在0~12岁儿童
均较为普遍;维生素D的缺乏及不足率随着年龄的
增长而增加(6.7%~54.8%)。维生素A和维生素
D在0~3岁婴幼儿中补充率较高,尤其是维生素
D:3岁以上儿童补充率不足,应重视3岁以上儿童
维生素A、维生素D营养水平。因此,建议补充至青
春期。
专家组成员:首都儿科研究所(戴耀华、王琳、李
涛),中国疾病预防控制中心营养与健康所(士
安),苏州大学儿科研究所(古桂雄),上海市妇幼保
健中心(彭咏梅),四川大学华西第二医院(吴康敏),
湖南省儿童医院(钟燕),山东省妇幼保健院(李燕)。
| ### 中国儿童维生素A、维生素D临床应用专家共识(2024版)
#### 专家共识
儿童中维生素A和维生素D缺乏很常见，可能存在较高比例的维生素A和维生素D同时缺乏的问题，这些将影响儿童的健康状况、增加对感染性疾病的易感性。中华预防医学会儿童保健分会于2021年组织国内相关专家撰写了《中国儿童维生素A、维生素D临床应用专家共识》(以下简称《共识》)。
近年的调查结果显示，我国学龄前儿童维生素A缺乏和边缘缺乏率高，且有随年龄增长而增加的趋势：0~18岁儿童中维生素D缺乏和不足率较普遍，儿童中维生素A和维生素D的缺乏存在相关性。因此，中华预防医学会儿童保健分会再次组织专家就上述《共识》进行了修订，期待对我国儿童维生素A、维生素D的临床应用和缺乏的预防提供参考。
#### 我国儿童维生素A、维生素D营养现况
##### 维生素A
1. **维生素A营养状况判定指标与标准**
维生素A的营养状况判定指标通常采用血清维生素A(视黄醇)浓度，主要采用2011年世界卫生组织发布的“血清视黄醇浓度用于确定人群维生素A缺乏的患病率”中推荐值，即血清维生素A浓度<0.70umol/L为缺乏，0.70~1.05mol/L为边缘缺乏，≥1.05mol/L为正常。剂量反应试验可检测肝脏中维生素A储备情况。
2. **我国儿童维生素A营养状况**
北京大学公共卫生学院2017年的系统性评价结果显示，我国0~12岁儿童中维生素A缺乏率为5.16%，边缘缺乏率为24.29%。其中，5岁以下儿童维生素A缺乏率为9.23%，边缘缺乏率为31.53%，高于2010—2013年中国居民营养与健康状况监测报告的结果(3~5岁儿童维生素A缺乏率为1.5%，边缘缺乏率为27.8%)。我国中西部地区横断面调查结果显示，2~6岁儿童随年龄增长维生素A未达1.05mol/L的比例分别为25.00%、28.92%、38.38%、42.73%，维生素A边缘性缺乏的比例随年龄的增长而增加。我国儿童维生素A边缘缺乏率存在明显的地区和年龄差异。即使在北京这样的一线城市的调查结果也与之相似。尽管学龄期儿童维生素A缺乏和边缘缺乏率略低于学龄前期，但依然需要引起重视。全国中小城市小学生维生素A的总体缺乏率为9.77%，边缘缺乏率20.56%。
##### 维生素D
1. **维生素D营养状况判定指标与标准**
血清25-(OH)D水平被认为是反映机体维生素D营养状况的最佳指标，血清25-(OH)D适宜浓度的判定可参照2016年全球营养性佝偻病管理共识，即血清25-(OH)D<30nmol/L为维生素D缺乏，30~50nmol/L为维生素D不足，≥50nmol/L则为适宜。
2. **我国儿童维生素D营养状况**
我国0~18岁健康儿童中维生素D缺乏和不足较普遍，其中严重缺乏率为2.46%(1.03%~4.47%)、缺乏率为21.57%(13.65%~30.72%)，28.71%(20.83%~37.35%)，婴幼儿组最低11.06%(6.07%~17.30%)，56.14%(39.54%~72.07%)。我国7岁以下儿童维生素D缺乏和不足检出率为14.0%，且随着年龄增加逐渐增高。
##### 维生素A、维生素D缺乏共病问题
虽然多种维生素，如维生素C和E以及B族复合物在免疫系统中以相对非特异性的方式发挥作用，但是维生素A和D以高度特异性的方式影响机体的免疫反应，因为它们各自的生物活性代谢物视黄酸和1,25-二羟基维生素D[(1,25-(OH)D)]具有激素样特性，而且这两种代谢物都是由体内不同的组织和细胞从其维生素前体合成，并通过与核激素受体结合远程对靶细胞发挥作用。儿童中维生素A和维生素D缺乏很常见，尤其是中低收入国家的5岁以下患营养不良的儿童中，还同时存在较高比例的维生素A和维生素D同时缺乏(共病)的问题。
维生素D的某些功能似乎与其他微量营养素具有协同作用，特别是维生素A。这两种维生素都共同调节机体的多种免疫反应，具有激素样特性和通过核激素受体介导发挥作用。维生素D缺乏与小儿感染性疾病易感性增加有关，这与维生素A的状态是一致的，这种情况在儿童营养不良和疾病双重负担的低收入国家/地区5岁以下儿童中可能特别严重。巴西涵盖468名儿童[(13.4±1.0)月龄]研究结果显示，血清视黄醇浓度与维生素D呈显著正相关(P=0.001)，维生素A浓度每增加至1ymol/L，维生素D充足率增加至1.38倍(95%CI:1.18~1.61)；控制所有协变量后，血清视黄醇每增加1mol/L与25-(OH)D浓度增加0.43nmol/L相关(95%CI:0.18~0.80)。有研究结果显示，维生素A的存在可以使维生素D的生物活性提高130%，增强免疫调节作用。鉴于我国儿童普遍缺乏维生素A和维生素D，且维生素A和维生素D存在协同作用，两者联合补充可能有助于同时改善维生素A和维生素D的营养状况。
#### 我国儿童维生素A、维生素D缺乏原因
机体在某一阶段，对维生素A、D的需求和供应平衡是维生素A、D的理想状态。
##### 维生素A、D缺乏原因的分类
1. **原发性缺乏**
原发性缺乏是由于机体摄入足够的维生素A、D来满足机体的需要。
2. **继发性缺乏**
继发性缺乏是环境状况或生理状况导致的对维生素A、D的吸收和利用的障碍，包括维生素A、D摄取后不能充分吸收利用。
##### 常见的维生素A、D缺乏原因
维生素A、D水平的降低是由于有效供应的降低，或因为对维生素A、D的需求增多所致。
1. **围生期储存不足**
维生素A和类胡萝卜素都很难通过胎盘进入胎儿体内，新生儿血清和肝脏中维生素A含量明显低于母体，围生期储存不足是导致新生儿和婴儿期维生素A、D缺乏的重要原因。虽然胎儿可通过胎盘获得维生素D，但脐血中25-(OH)D水平仅为母亲的60%~85%，胎儿和新生儿维生素D水平较低。母亲若有严重营养不良，肝肾疾病、慢性腹泻时，则子代的维生素D水平更低。
2. **生长发育迅速**
生长发育迅速是导致儿童维生素A,D缺乏的重要原因。儿童时期体格生长是连续而不均匀的，具有阶段性，是循序渐进的过程，婴幼儿期和青春期是体格生长、器官成熟最快的两个时期，对维生素A、D的需求量显著增加。维生素D具有许多潜在的生物学功能，参与钙转运蛋白和骨基质蛋白的转录以及细胞-周期蛋白转录的调节，良好的维生素D状态可提高机体的各种生理效能。因此，及时充足地补充维生素A、D，可为儿童提供体格生长、神经和心理行为发育的物质基础。维生素A,D是生命早期“大脑构建营养素”，若供给不足即可影响脑和神经系统的发育，即使在2周岁后足量补充也无法完全逆转。儿童期维持充足的维生素A、D营养水平，不仅是满足现阶段的营养保障，更为重要的是为成年后潜能的发展奠定物质基础。
3. **营养供给不足**
- **母乳**
母乳中的维生素A、D具有较好的生物活性，是婴儿期的重要来源，但受到乳母饮食和健康的影响。母乳中维生素D的含量无法满足婴儿的每日需求量。早产儿、双胎儿、低出生体重儿由于自身储存不足，为了满足后期追赶生长的需要，对维生素A、D营养的需求明显增高。
- **天然食物**
维生素A存在于动物肝脏和红、黄色蔬菜中，植物中的维生素A以维生素A原-类胡萝卜素的形式存在，其转化率存在吸收有饱和现象，可见膳食维生素A的摄入明显不能满足儿童生长的需要。学龄前，学龄期及青春期儿童膳食维生素A摄入量仍较低，同时，天然食物中维生素D的含量通常较少，且其稳定性易受到高温、强光、强酸碱等因素的破坏。
- **紫外线照射合成**
紫外线照射皮肤合成的维生素D是人体最主要的维生素D来源，其合成与季节、海拔、纬度、照射强度、户外活动时间，暴露皮肤的面积，以及空气污染、使用护肤品，不同衣着等因素有关。
- **营养素补充**
儿童维生素A边缘缺乏率存在明显的地区和年龄差异，这与是否服用含维生素A补充剂以及剂量有关。儿童维生素D缺乏和不足也与维生素D摄入量不足有关。据调查，随着年龄的增加，维生素D制剂补充率降低。0~<3岁儿童维生素D补充率为65.0%，而3~<7岁儿童仅有23.5%，可能是3~<7岁儿童维生素D缺乏和不足率上升的原因。常规给予预防剂量维生素D补充剂有助于改善儿童维生素D营养状况，降低发生维生素D缺乏的风险。每日持续服用小剂量维生素A、D补充剂，可维持儿童更好的营养水平。未规律服用，间断服用，长期停止服用等原因亦是导致体内维生素A、D低水平的重要原因。
4. **疾病的影响**
- **影响吸收**
感染性疾病、消化道疾病、肝胆系统疾病、肾脏疾病、甲状腺功能亢进等疾病均可影响维生素A、D的吸收、代谢。而贫血、抽动障碍、孤独症譜系障碍(autism spectrum disorder, ASD)等患儿易合并维生素A的缺乏。
- **消耗增加**
感染性疾病患病期间会导致维生素A、D的大量消耗，尤其是麻疹、猩红热等患儿，体内维生素A含量显著缺乏，且感染愈严重，维生素A的耗损愈多。各种原因的发热状态，均增加维生素A等多种维生素耗用，故而，患儿及时补充维生素A，可改善疾病预后，并可降低再患感染的风险。此外，肥胖儿童体内多余脂肪可吸收维生素D，使循环中的维生素D含量下降，并降低其生物活性，从而降低体内维生素D的功能。
5. **药物的干扰**
长期服用某些药物，如新霉素、糖皮质激素、抗惊厥、抗癫痫药物等，可导致维生素A、D的吸收和代谢出现明显障碍，应充分引起重视。
6. **其他因素**
- **脂肪含量不足**
维生素A、D均为脂溶性维生素，若膳食中的脂肪含量不足，可阻碍维生素A、D的吸收。早产儿、中、重度消瘦患儿对脂肪的吸收能力较差，易导致吸收不良。
- **维生素E的含量**
维生素E具有抗氧化作用，能够防止维生素A在肠道内被氧化破坏，同时，维生素E的低含量可降低维生素A的吸收。
- **维生素D受体的多态性**
儿童青少年血液中的维生素D含量可受基因多态性的影响，其不同点位的基因多态性与机体各种代谢性疾病相关，携带骨代谢障碍或佝偻病易感基因是导致机体内维生素A,D缺乏的危险因素。
- **经济的影响**
经济欠发达的国家和地区中，许多儿童未满足最低膳食量摄入和膳食的多样性，明显降低维生素A、D含量的补充或互补。
#### 维生素A、维生素D缺乏对儿童健康的影响
##### 维生素A缺乏对儿童健康的危害
维生素A是人体维持机体正常代谢和功能的脂溶性维生素。维生素A具有维持暗视觉，促进上皮细胞再生，维护皮肤、黏膜和其他组织表面的完整性；促进生长发育，增强免疫功能、维护生殖功能、影响造血的功能。此外，最新研究发现维生素A还可以参与肾脏损伤修复以及抗癌作用。维生素A缺乏的临床表现与缺乏阶段和程度有密切关系。可疑和亚临床维生素A缺乏阶段主要表现 |
{"text":"达成共识的议题\n\n1.营养性佝偻病\n\n营养性佝偻病,由单纯的维生素 D 或钙缺乏或两者兼而有之引起,仍然影响着全世界相当 数量的婴儿和儿童。维生素 D 缺乏性佝偻病可以通过补充维生素 D 治疗。人们一致认为, 婴儿和大多数儿童每天需要补充大约 400 国际单位(或大一点的儿童需要 600 国际单位)的 维生素 D 来预防佝偻病,这是因为通常要避免直接暴露在阳光下,并且也不推荐非常年轻 的人使用。然而,这种补充政策在许多国家要么没有得到执行,要么没有得到充分执行。\n\n尽管亚洲和中东国家最常受到营养维生素 D 缺乏的影响,但非洲和一些亚洲国家也发生钙 缺乏引起的佝偻病。对于 0 至 6 岁的新生儿和 6 至 12 个月大的婴儿,充足的钙摄入量分别 为 200 毫克/天和 260 毫克/天,而对于 12 个月以上的儿童,膳食钙摄入量<300 毫克/天会 增加佝偻病的风险,且与血清 25OHD 水平无关。对于 12 个月以上的儿童,根据膳食钙摄入 量可分为:充足=>500mg/天;不足=300-500mg/天,缺乏=<300mg/天。\n\n钙缺乏性佝偻病的发病机制可能比以前认为的更复杂。然而,我们确实知道减少钙摄入会增 加 PTH 分泌,进一步导致 FGF-23 分泌增加。PTH 和 FGF-23 的增加导致尿磷酸盐排泄增加。 这一病理生理现象导致血清磷酸盐减少,磷酸盐与 PTH 一起增加了 1,25(OH)D 水平。升 高的 1,25(OH)D 上调了许多基因,导致焦磷酸盐(一种已知的骨矿化抑制剂)以及骨桥 蛋白和小整合素结合配体 N 连接糖蛋白增加(SIBLINGS;图 1)。这些异常,以及低钙和低 磷酸盐水平,是钙缺乏性骨质软化症特征的主要原因。尽管这种病理生理序列已经在动物身 上得到证明,但很可能人类也受到同样的影响。\n\n图片\n\n图 1 慢性低钙摄入引起佝偻病的发病机制。\n\n注:ANK=强直蛋白;Ca=钙;ENPP=外核苷酸焦磷酸酶/磷酸二酯酶;FGF-23=成纤维细胞生 长因子 23;PPi=无机焦磷酸盐;PTH=甲状旁腺激素;SIBLINGS=小整合素结合配体,N-连接 糖蛋白;Tnap=组织非特异性碱性磷酸酶。\n\n目前,全球不同国家主要基于临床体征的软骨病发病率仍然很高(表 1)。基于软骨病在全 球的广泛流行,应成立一个工作队来处理这一问题。这样一个由国际内分泌学会、国际肌肉 骨骼研究学会联合会、儿科内分泌学会和欧洲儿科内分泌学会等学会的代表以及维生素 D 会议的代表组成的工作队应编写并向世卫组织提交一份计划,在 2030 年前根除佝偻病。 表 1 全球佝偻病流行情况\n\n图片\n\n目前,在世界各地的不同国家,佝偻病的发病率仍然很高,主要是根据临床症状(表 1)。 根据佝偻病在全球的普遍流行情况,应成立一个工作队来处理这一问题。该工作组由国际内 分泌学会、国际肌肉骨骼研究学会联合会、儿科内分泌学会和欧洲儿科内分泌学会等学会的 代表组成,以及来自维生素 D 会议的代表应制定并向世卫组织提交 2030 年前根除佝偻病的 计划。\n\n2.维生素 D 是由太阳的 UVB 光产生的\n\n维生素 D 是由太阳 UVB 光(波长约 280 至 310 nm)产生的 UVB 光打开 7-脱氢胆固醇的 B 环,这是胆固醇从头合成的最后一步,并生成维生素 D,在通过与血清维生素 D 结合蛋白(DBP) 结合转移到循环中之前,其经历热诱导异构化成维生素 D3。\n\nUVB 光(波长约为 280 至 310nm)打开 7-脱氢胆固醇的 B 环,这是胆固醇从头合成的最后 一步,并产生前维生素 D,在通过与血清维生素 D 结合蛋白(DBP)结合转移到循环中之前, 其经历热诱导异构化成维生素 D3。\n\n短时间暴露在阳光下有利于维生素 D 的产生,而长时间暴露在紫外线下会导致晒伤和 DNA 损伤。更大的剂量会导致更强烈的峰值反应,大致呈线性关系,线条的实际斜率由个体变异\n\n性定义,而个体变异性反过来又可能至少部分由遗传决定因素解释。随着紫外线剂量的增加, 简单的晒黑被更严重的晒伤所取代。相比之下,维生素 D 的形成是瞬间的,并且从非常小 的紫外线照射到非常大的紫外线照射以时间依赖性的方式呈线性增加。维生素 D 皮肤光合 作用的剂量反应在小剂量紫外线照射下呈线性增加,达到平台期的剂量远低于红斑阈值剂量, 与其他剂量反应曲线有显著不同;(图 2)。因此,短时间的紫外线照射可以增加维生素 D 光合作用。然而,有许多其他因素会影响维生素 D 皮肤光合作用,如年龄、肤色、防晒霜 的使用、纬度、一天中的时间和季节。因此,对于普通人群安全有效的阳光暴露剂量目前还 没有达成共识。此外,鉴于上述个体差异,尝试全面指导似乎是不明智的。\n\n图片\n\n图 2 紫外线照射的最小红斑剂量(minimal erythema dose,MED)与 DNA 损伤水平、晒黑/ 晒黑和维生素 D 合成之间的关系。\n\n3.维生素 D 缺乏流行\n\n尽管由于人类行为(室内工作、避光等)的原因,皮肤维生素 D3 的合成在阳光充足的情况 下会迅速发生,但维生素 D 缺乏的情况非常普遍。使用<20ng/mL(<50 nmol/L)的定义, 世界上多达三分之一的人口缺乏维生素 D,欧洲的比例高达 40%(表 2)(Lancet Glob Health. 2020;8(1):e134–42)。\n\n表 2 全球维生素 D 缺乏症\n\n图片\n\n全世界约 7%的人口患有严重的维生素 D 缺乏症,定义为<30nmol/L(或<12 ng/mL),不 同国家和人群之间存在显著差异。然而,全世界的高危人群中都存在严重的维生素 D 缺乏 ( Br JNutr. 2014;111(1):23–45)。维生素 D 缺乏的高危人群包括那些缺乏有效阳光照 射的人群。这可能是由于各种气候、文化或宗教原因,以及皮肤色素沉着。长期以来,维生 素 D 缺乏在非洲被认为是罕见的,但对非洲国家的系统分析表明,18%的非洲受试者存在严 重的维生素 D 缺乏,根据文化/行为实践,缺乏维生素 D 的群体普遍存在\n\n4.25OHD 是维生素 D 状态的\"最佳\"标志物\n\n目前已经广泛接受血清 25OHD 浓度作为反映个体维生素 D 状态的最佳指标,并已被许多机 构用于确定维生素 D 饮食要求和监测维生素 D 缺乏或不足的人群。然而,血清 25OHD 至少 在历史上,人们不认为维生素 D 具有生理调节作用,因此其他指标可能是维生素 D 状态的 更好指标。值得注意的是,关于游离 25OHD(未结合到载体蛋白)或 24,25(OH)2D:25OHD 的比值,是否更优于总的 25OHD。目前仍存在争议。\n\n25(OH)D3:24,25(OH)2D3 的比率已被作为 CYP24A1 突变引起的特发性婴儿高钙血症的 诊断工具。然而,维生素 D 缺乏患者这个比率也升高,这些患者因 CYP24A1 酶下调引起的 慢性肾病而接受透析。在某些情况下,1,25(OH)2D:25OHD 的比率也可能是 CYP27B1 活性 的有用标志。\n\n重要的是,维生素 D 研究数据受到血清总 25OHD 测定方法质量变化的困扰,该方法已经损 害并继续损害区分目前使用的不同指南的能力。同样,与血清总 25OHD 相比,游离 25OHD 测量质量的不确定性阻碍了其评估。对于 25OHD 和 24,25(OH)D2,可使用参考方法改进 这些分析物的标准化。维生素 D 标准化计划(VDSP)和维生素 D 外部质量保证计划(DEQAS) 鼓励标准化。DEQAS 在 CDC 标准化目标值的支持下,30 年来每季度监测 700 到 1000 个实验 室检测 25 OHD 的能力。几十年来,它记录了有问题的化验和试剂盒制造商。DEQAS 还通过 血清样本促进 24,25(OH)2D3 和 1,25(OH)2D 检测的准确性。\n\n目前,VDSP 正在协调一项通过开发\"真实\"控制来协调直接免费 25OHD 测量的努力(个人 交流,克里斯·塞姆波斯教授)。最后,作为 VDSP 的一部分,美国国家卫生研究院膳食补 充剂办公室正在赞助开发 1,25(OH)2D 的参考方法,这将有助于标准化其在维生素 D 研究\n\n中的测量,并使其作用更加清晰。这种标准化努力对于推进澄清什么是真正的维生素 D 依 赖至关重要。\n\n目前,VDSP 正在协调通过开发\"真实性\"控制(个人通信,Chris T Sempos 教授)来协调直 接免费 25OHD 测量的工作。最后,作为 VDSP 的一部分,国家卫生研究院膳食补充剂办公 室正在赞助 1,25(OH)2D 参考方法的开发。此类标准化工作对于进一步澄清维生素 D 缺乏 的真正原因至关重要。\n\n目前,VDSP 正在协调一项通过开发\"真实\"控制来协调直接免费测量 25OHD 的工作(个人 交流,克里斯·塞姆波斯教授)。最后,作为 VDSP 的一部分,美国国立卫生院膳食补充剂 办公室正在赞助开发 1,25(OH)2D 的参考方法,这将有助于使其在维生素 D 研究中的测量 标准化,并明确其作用。这种标准化努力对于进一步明确什么是真正的维生素 D 缺乏至关 重要。\n\n然而,在维生素 D 代谢物的测定中,标准化并不是唯一的分析挑战。在许多 25OHD 免疫分 析中,患者或矩阵相关偏差是一个众所周知的混杂因素,导致结果不准确,例如在孕妇或血 液透析患者中。此外,在免疫分析中,25(OH)D3 和 25(OH)D2 的药效或释放 DBP 的差 异导致服用 D2 补充剂的受试者在测定血清 25OHD 浓度时出现重要问题。这些问题无法通过 标准化措施解决,但在特定的免疫分析中是固有的;因此,这些免疫测定也必须改进。这在 普遍使用麦角钙化醇的地区以及可能选择避免使用胆钙化醇的素食者中尤为重要。\n\n5.维生素 D 缺乏的定义/阈值\n\n正如各专家组提出的不同建议所指出的那样,关于维生素 D 缺乏的定义仍存在争议。然而, 有两点是一致的:所有年龄段 25OHD 水平低于 12ng/mL(30nmol/L)提示明显不足,而高 于 30 ng/mL(75 nmol/L)提示 25OHD 充足。相反,对于在 12-30ng/mL(30-75nmol/L)之 间的水平存在争议。一些指南建议阈值定为 20ng/m(l 50nmol/L),而其他指南推荐≥30ng/ml (75nmol/L)。这些讨论很大程度上基于 25OHD 缺乏的标准化分析。\n\n这些切点对随机临床试验(RCT)具有关键意义。很少有临床试验纳入明确维生素 D 缺乏的 受试者;其中一个例子是 Chapuy 及其同事的工作)。研究补充维生素 D 对营养缺乏的受试 者的影响的重要性无论怎样强调都不为过,因为维生素 D 是一种临界营养素,这意味着生 理终点,如钙吸收,以剂量依赖性方式增强,达到阈值,继续高于该阈值,更高的水平不会 导致更大的效果。如果临床招募的受试者 25OHD 水平高于阈值,随机让受试者接受额外的 维生素 D 大大降低了显示补充益处的可能性。最近在主要同行评议期刊上发表的广为宣传 的随机对照试验很好地说明了这一混淆点。如果对照组和补充组都在基线时开始摄入足够水 平的 25OHD,则人们不会期望看到阈值营养素的效果。\n\n6.维生素 D 类似物是银屑病的首选局部疗法\n\n维生素 D 类似物治疗银屑病的益处已得到公认。外用维生素 D 类似物是银屑病治疗的首选, 可以单独使用,也可以与外用皮质类固醇联合使用。与皮质类固醇不同,皮质类固醇可能与 快速免疫有关,局部使用维生素 D 类似物治疗对所有年龄的患者都是长期有效的,没有副 作用。\n\n未达成共识的主题\n\n1.一般人群维持 25OHD 的正常水平,建议每天服用哪些剂量的维生素 D?\n\n如果婴儿、儿童/成人和老年人没有暴露在阳光下,医学研究所建议每天补充 400-600-800 国际单位的维生素 D。北欧和 DACH(德语)国家、澳大利亚和新西兰、欧洲食品安全局、 欧洲钙化组织协会和国际骨质疏松基金会制定的指南认可了这些建议。内分泌学会(美国) 建议对所谓的风险群体每天 600U 至 2000U。英国指南(营养科学咨询委员会)建议任何年 龄每天 400IU。其他一些组织建议更高的剂量(4000-10000U/天)。这些建议适用于没有骨 质疏松症或其他代谢性骨病的个人。不幸的是,这一点没有得到许多组织或从业者的赞赏。\n\n错误可能以两种方式出现。首先,如果遵循一些机构每天高达 10000 国际单位的建议,过度 关注骨骼健康的受试者可能会摄入过多。例如,据估计,美国有 3%的成年人每天服用超过 4000 国际单位的维生素 D 补充剂。这种剂量可能是有害的,因为这种剂量可能会降低而不 是增加骨密度或骨强度。另一方面,使用相对较低的剂量可能对那些要求较高的人有害(例 如,吸收不良或肥胖)。很明显,有必要维生素 D 不足定义并达成共识,以解决每日补充 剂量建议中存在的差异。\n\n2.代谢性骨病患者有哪些补充建议?\n\n在患有骨质疏松症或其他代谢性骨病的患者中,关于维生素 D 的讨论与一般人群不同。显 然,更强调首先保证 25OHD 水平完全高于阈值,以随后的阈值为准,无论是 20ng/ml 或 30ng/ml(50 或 75nmol/l)。此外,有证据表明,当维生素 D 和钙的充足性得到保证时,抗 骨质疏松药物的反应可能会增强。\n\n博兰及其同事最近进行的一项荟萃分析对这一共识提出了质疑。在他们的综述中,他们指出: \"我们的发现表明,补充维生素 D 不能防止骨折或跌倒,也不能对骨密度产生临床意义上 的影响。他们以以下声明结束了他们的讨论:\"使用维生素 D 补充剂来维持或改善肌肉骨骼 健康几乎没有正当理由\"和\"这个结论应该反映在临床指南中。\n\n首先,对他们的说法提出异议的其他专家质疑这种结论。Lips,Bilezikian 和 Bouillon 指出, 该荟萃分析排除了所有将钙加维生素 D 与双安慰剂进行比较的研究。布恩和他的同事们多 年前表明,有必要给予足量的钙和维生素 D 来观察骨折的减少情况。韦弗及其代表国家骨 质疏松症基金会的同事得出了相同的结论,姚及其同事在最近的荟萃分析中也得出相同的结 论。\n\n其次,超过 60%的研究是短期的,不足 1 年。在如此短的时间内期望抗骨质疏松药物对骨折 风险的有益效果是不合理的。\n\n第三,缺乏维生素 D 的个体(25OHD <12ng/ml 或 30nmol/l)在整个研究人群中只占极小 的百分比:<2.1%。\n\n第四,构成骨折风险最高个体(18%)试验因依从性差(约 50%)而受阻。该荟萃分析的另 一个缺陷是纳入了间歇性补充高剂量维生素 D 的研究,这可能会增加骨折风险。此外,这 项荟萃分析的两位主要作者分别发表了独立的荟萃分析,他们得出结论,维生素 D 和钙补 充剂联合使用可以降低老年人髋部和非脊椎骨折的风险。在他们最新的荟萃分析中没有提到 或讨论早期的综述。其他专家也得出了类似的结论。尽管如此,直到 2019 年 ASBMR 会议, 争论仍在激烈进行,相反的观点仍在表达。\n\n3.紫外线照射皮肤产生维生素 D\n\nHolick 及其同事领导的研究反复指出,一整天的阳光照射可以产生 10,000 到 25,000 国际 单位的维生素 D。在这一点上,在其他研究中已经表明,土著,皮肤非常黑的马赛人据说每 天能产生 10,000 或 20,000 国际单位的维生素 D。最近的研究提出了阳光对皮肤维生素 D 产生的影响程度的问题。在加那利群岛,年轻的丹麦妇女暴露在强烈的阳光下,其 25OHD 增加,相当于每天仅 600 至 1000IU。通过比较每周三次全身紫外线照射和每天仅口服 800 国际单位的维生素 D,发现阳光照射可以导致血清 25OHD 浓度增加。另一项来自加那利群 岛的丹麦年轻女性的研究显示,每天暴露在阳光下 1 周,血清中 25OHD 的水平仅增加 20nmol/L(8ng/mL),相当于每天口服 800IU 的维生素 D。在加那利群岛的另一项研究中, 几乎全身暴露在阳光下的波兰年轻志愿者的 25OHD 变化为 28nmol/L 或约 12ng/mL,相当于 每天约 600 至 1200IU(~15 至 30ug)的口服维生素 D。最后,疗养院居民每周三次背部暴 露 1000cm2,剂量为最小红斑剂量的一半,在三个月内,血清 25OHD 中位数从 7.2ng/mL 增 加到 24ng/mL(18nmol/L 到 60nmol/L),相当于每天补充 400IU/天。目前尚不清楚每天完 全暴露在阳光下会产生什么。是 1000IU 左右还是更接近 10000IU?这个问题的答案可能有\n\n助于解释很少暴露在阳光下的受试者每日维生素 D 的需要量。\n\n4.肝脏对 25OHD 代谢的调节\n\n肝脏对 25OHD 代谢的调节生化表明,组成性活性羟化酶 CYP2R1 使肝脏中的维生素 D 羟基 化为 25OHD。尽管这种酶是主要的转化酶,但它并不是羟化发生的唯一途径。最近的数据 也对这种反应的组成性质提出了质疑,证据表明这种酶受到几种不同的控制机制的影响。例 如,在禁食状态下,可以观察到 CYP2R1 表达显著降低。在糖尿病小鼠模型中,CYP2R1 的 mRNA 和蛋白质表达减少了 50%。本研究确定了糖尿病患者维生素 D 缺乏的新分子机制(包 括 PPARγ共激活因子 a1 和雌激素相关受体),并显示了一种新的负反馈机制,调控能量稳 态和维生素 D 之间的相互作用通路。糖皮质激素受体的激活(通过地塞米松或其他皮质类 固醇)也抑制 CYP2R1 的活性。因此,与其将肝脏视为通过不受调控的 CYP2R1 酶将维生素 D 定量转化为 25OHD 的组成工厂,不如将代谢和激素机制视为有效的。显然需要更多的研究 来更好地了解肝脏中 25OHD 的产生是如何调节的。\n\n5.维生素 D 过量的定义\n\n维生素 D 中毒的经典概念被认为是高钙血症可能发生的水平。血清 25OHD 水平超过 100 或 150ng/ml(250 或 375nmol/l)可能导致高钙血症,因此,这些切入点成为许多权威机构的 参考框架,如医学研究所、内分泌学会和参考实验室。\n\n虽然将高钙血症确定为毒性阈值似乎是合理的,但其他毒性指标,如高尿钙,可能发生的水 平要低得多。在 Gallagher 及其同事的研究中发现,30%的维生素 D 缺乏者在一年内每天仅 服用 800 至 2000IU 的维生素 D,而仅 9%患者发生了高钙血症。\n\n加拉格尔和他的同事进行了进一步的人体研究,并由考夫曼和他的同事进行了重新分析,研 究表明,一年内每天高达 4000IU 的维生素 D 摄入量导致血清 25OHD<90ng/mL。\n\n维生素 D 的服用量与尿钙排泄或高钙血症水平之间没有关系。此外,高钙尿是暂时的。然 而,这些数据并不能令人信服地支持这种低剂量方案可能有害的观点。维生素 D 的服用量 与尿钙排泄水平或高钙血症之间没有关系。此外,在这些受试者中,半数受试者的高尿钙是 暂时的。然而,进一步增加这些数据的不确定性的是,观察到服用安慰剂的患者也有同样的 高钙尿症发生率。这些数据并不能令人信服地支持这种低剂量方案可能有害的观点。事实上, 大多数专家都同意高达 4000IU 的剂量可能是安全的。更引人注目的讨论集中在与高剂量维 生素 D 相关的跌倒风险上。间歇性高剂量给药或定期给老年人服用维生素 D,与 25OHD> 45ng/mL(>113nmol/L)的水平相关,可能导致跌倒风险增加。需要进一步研究,以进一 步澄清这种 25OHD 水平是否确实会增加跌倒风险。\n\n骨骼健康也是与高剂量维生素 D 的副作用相关的最新研究的重点。在卡尔加里对无骨质疏 松症的健康志愿者进行的研究中,其平均基线 25OHD 约为 31 至 32ng/mL,与 400 IU/天相 比,以 4000IU/天或 10000IU/天的剂量使用维生素 D 治疗 3 年,在统计学上显著降低了桡骨 体积 BMD。然而,无论是桡骨还是胫骨,均未观察到显著的骨强度差异。Burt 及其同事从 这项研究得出结论,4000 IU 或更高剂量的维生素 D 作为骨骼健康的辅助手段没有任何益处。 6.急性疾病中维生素 D 缺乏\n\n在急性疾病的情况下,25OHD 的水平可能较低,因为循环 DBP 急剧减少。血管内急性液体 转移的稀释效应也可能是一个因素。此外,预先存在的维生素 D 营养状况也是一个因素。 后一点表明,纠正不良的维生素 D 状态可以降低发病率和死亡率。Christopher 及其同事建 议,在重症监护室的病人中,可能需要非常高剂量的维生素 D 才能获益。可能需要更高的 剂量,因为急性疾病患者可能分泌大量的皮质醇,这反过来会影响肝和肾脏对维生素 D 的 羟基化。\n\n7.生殖过程中维生素 D 的需求\n\n在生殖过程中使用维生素 D 的问题上缺乏共识。\n\n一方面,孕妇对维生素 D 的需求在怀孕或哺乳期间没有增加。达到的 25OHD 水平不受任何 一种生殖状态的影响,也没有证据表明妇女在怀孕或哺乳期间应保持高于健康非怀孕理想值 的 25OHD 水平。\n\n另一方面,孕期母体维生素 D 缺乏会影响胎儿和新生儿的健康;因此,确保孕妇在怀孕期 间的维生素 D 状态得到优化肯定是有意义的。这并不意味着妇女怀孕时比未怀孕时需要\"更 多\"的维生素 D。\n\n在哺乳期,母亲的维生素 D 状况并不直接重要,因为很少有维生素 D 进入母乳,特别是因 为随机对照试验表明,在低至高 25OHD 水平范围内,母乳中的钙含量与母亲的维生素 D 状 况无关。母乳喂养的婴儿需要补充维生素 D,而配方奶粉喂养的婴儿需要补充配方奶粉中的 维生素 D。\n\n尽管不同研究之间存在差异,但随机对照试验和系统评价的证据表明,维生素 D 摄入高达 2000IU/天对先兆子痫和妊娠期糖尿病以及新生儿结局都有好处。明确推荐在妊娠期维持生 素 D 的正常水平似乎是合理的。\n\n8.在维生素 D 受体的影响下,细胞和器官活动的广谱性潜力\n\n维生素 D 受体(VDR)几乎存在于所有细胞和组织中。1a‐羟化酶 CYP27B1 也存在于全身和 许多细胞类型中。据估计,脊椎动物(从斑马鱼到小鼠再到人类)所有基因的 3%至 10%受 1,25(OH)2D3 的直接或间接控制。这种进化的无处不在表明维生素 D 在所有器官的功能 中起着重要作用。在小鼠中以组织特异性方式删除该基因的实验证实了这一预期。对组织特 异性 KO 实验的取样表明,乳腺更容易患乳腺癌,心肌发生心肌肥大,肝脏变得肥胖(非酒 精性脂肪肝综合征),前列腺发生增生,动脉粥样硬化加速,小鼠对饮食诱导的肥胖产生抵 抗力。相反,VDR 的过度表达导致小鼠肥胖,但不会导致人类肥胖。需要更多的工作来理解 这些小鼠中的 KO 和过表达模型如何与人类病理生理学相关。\n\n最近的死亡率数据显示,低 25OHD 和全因死亡率风险增加之间存在关联。这些发现也在一 个欧洲联盟中观察到。几项对 25OHD 与总体死亡率和心血管死亡率的相关分析显示,呈 U 形曲线。一项基于 75,000 名患者,纳入 38 项补充试验的的荟萃分析也显示死亡率轻度下降, 但具有显著的统计学意义。\n\n孟德尔随机研究检验了维生素 D 对自身免疫性疾病的影响,三项独立的发现表明,维生素 D 水平的降低(比正常水平低 5%到 7%)显著增加了患多发性硬化症的易感性。最后,一项孟 德尔随机研究显示了维生素 D 和 1型糖尿病(T1DM)风险的相关性。\n\n关于小鼠癌症的数据也令人感兴趣。随着年龄的增长,1,25(OH)2D3 基因敲除小鼠发生癌 症的机会更大,而且肠和乳腺细胞的增殖率也增加。尽管 VDR 阴性小鼠通常不会自发地发 生更多的癌症,但当暴露于癌基因、抗癌基因缺失或暴露于致癌物或 UVB 光时,它们更可 能发生一系列恶性肿瘤,如乳腺癌、结肠癌、皮肤癌,\"癌症发展风险与多个事件相关。尽 管这些小鼠的数据看起来很有说服力,但孟德尔随机研究在人类中并没有得到支持。\" 9.维生素 D 和人类主要疾病之间的潜在联系\n\n许多横断面、观察性和回顾性研究都显示维生素 D 水平低下与许多人类疾病相关。总的来 说,这些报告表明维生素 D 对大多数人体器官系统的健康有广泛的影响。维生素 D 在免疫 系统调节中作用的临床前证据可能是最强的,因为 VDR 和 CYP27B1 在免疫系统的先天和适 应性臂细胞中都有表达。此外,免疫细胞中的 CYP27B1 表达受复杂的先天免疫和细胞因子 网络的调节。儿童和成人人群中都有广泛的临床证据表明,充足的维生素 D 可以减低病毒 或细菌感染感染率。积累的证据表明,维生素 D 在自身免疫疾病中的角色都将是预防性的, 而非治疗性的。补充维生素 D 可能获益的一种情况是在炎症性肠病克罗恩病的治疗中,对 一系列小规模临床试验的荟萃分析表明,补充维生素 D 可以降低疾病的严重程度。对克罗 恩病患者进行大规模 RCT 临床试验以巩固这些发现是很重要的。大规模 RCT 对于确定维生\n\n素 D 缺乏和疾病之间的关系是因果关系还是简单的关联至关重要。\n\n另一种与维生素 D 缺乏有关的疾病是糖尿病。已经证明,维生素 D 通过作用于有缺陷的抑 制细胞功能或通过调节细胞因子表达来预防胰岛炎和实验性糖尿病的进展。这些观察结果已 经得到临床研究证实,研究表明,在儿童早期补充维生素 D 可能会降低患 T1DM 的风险。 然而,进一步的研究没有发现补充骨化三醇对新诊断的 T1DM患者的胰岛素分泌、胰岛素敏 感性或胰岛素需求或骨转换的改善有任何显著影响。\n\n目前尚不确定妊娠期或出生时 25OHD 水平是否能降低儿童期 T1DM的风险。然而,当考虑 到与遗传变异的相互作用时,出生时较高的 25OHD 水平预示着发生 T1D 或胰岛自身免疫的 风险降低。儿童或母亲 VDR 单核苷酸多态性都可能降低 VDR 表达,从而抑制 T 细胞增殖, 从而增加自身免疫的风险。\n\n最近的维生素 D 评估(VIDA)、维生素 D 和ω-3(VITAL)以及维生素 D 和 2型糖尿病(D2d) 试验代表了将这些观察结果转化为临床相关性的尝试实例。\n\n10.心血管疾病\n\nVIDA 试验在 5110 名受试者中测试了平均 3.4 年内每月剂量为 100000 IU 的维生素 D3 与安慰 剂对心血管疾病的影响。两组之间没有统计学差异。在 VITAL 试验中,维生素 D 组和安慰剂 组在任何单个心血管事件(如心肌缺血)或复合心血管终点方面均无显著差异。 11.癌症风险和存活率\n\n更大规模的 VITAL 试验对 25,871 名 50 岁以上的男性和女性进行了测试,测试了 5 年内每天 2000 国际单位的维生素 D3 对心血管事件和癌症的影响。在患任何侵袭性癌症的风险上维生 素 D 组和安慰剂组没有显著差异,乳腺癌、前列腺癌或结直肠癌的风险两组之间也没有明 显统计学差异。然而,在体重指数<25 的人群中,任何侵袭性癌症的风险都有显著降低。 排除前两年的研究,癌症死亡率也有所降低。拉普及其同事的研究在这方面值得注意;他们 表明钙和维生素 D 似乎有降低新发癌症风险的作用,但没有达到统计学意义。\n\n参与 VITAL 试验的一些研究人员随后进行的荟萃分析非常值得注意。虽然 VITAL 试验显示一 个\"信号\",似乎补充维生素 D 可以提高生存率,但对 VITAL 和其他几项研究进行荟萃分析 发现,补充维生素 D 的受试者生存率显著提高,但进展中的癌没有获益。对于癌症总发病 率,纳入了 10 项试验(6537 例;3至 10 年随访;干预组达到的 25OHD 水平为 54-135nmol/L)。 在受试亚组中,癌症风险总结仍然为零,包括达到的 25OHD 水平超过 100 nmol/L。癌症总 死亡率包括五项试验(1591 例死亡;随访 3 至 10 年;54 至 135 nmol/L,达到干预组循环 25OHD 水平)。总 RR 为 0.87(95%CI,0.79-0.96;p=0.005),这在很大程度上归因于每日 给药的干预措施(与不经常给药相反)。因此,这项最新的随机对照试验荟萃分析表明,补 充维生素 D 显著降低了癌症总死亡率,但并未降低癌症总发病率。在 Torfadottir 研究中,目 的是探讨老年人诊断前循环 25OHD 水平是否与诊断后的总体和癌症特异性生存率相关。他 们使用了来自 AGES(基因/环境易感性)雷克雅未克研究的数据,该研究对进入时未患癌症 的参与者(n=4619)进行了研究,当时采集了血液样本进行 25OHD 标准化测量。在后来诊 断为癌症的患者中,采用 50-69.9nmol/L(20-28ng/mL)作为参考类别,比较四种 25OHD 类 别,评估与癌症风险、所有原因和癌症特异性死亡率的相关性。\n\n在初次抽样后平均 8.3 年,919 名参与者被诊断出患有癌症。对照组和其他 25OHD 组与总癌 症发病率之间未观察到相关性。平均诊断年龄为 80.9(5.7)岁。在确诊患者中,552 人在 随访期间死亡:67%死于癌症。重要的是,25OHD<30 nmol/L(<12 ng/mL)的诊断前低水 平与总死亡率增加显著相关(危险比[HR],1.39;95%可信区间,1.03–1.88),与癌症特异 性死亡率无显著相关性(HR,1.33;95%可信区间,0.93-1.90)。在诊断后存活 2 年以上的 患者中,诊断前高水平 25OHD(≥70 nmol/L)与总体低风险(HR,0.68;95%可信区间,0.46-0.99) 和癌症特异性死亡率相关(HR,0.47;95%可信区间,0.26-0.99)。在老年癌症患者中,诊\n\n断前低血清 25OHD 水平(<30 nmol/L[<12 ng/mL])与总体死亡率增加相关。\n\n12.糖尿病\n\nD2d 试验检测每天 4000 国际单位的维生素 D3 对显性糖尿病发展的影响,该试验纳入了男女 共 2423 名受试者,年龄大于 30 岁,伴有糖尿病危险因素。研究发现,维生素 D 组和安慰 剂组在此期间糖尿病发病率没有明显差异。然而,对受试者进行后续的分析时发现,基线 25OHD<12ng/m(l 或<30nmol/l )的受试者接受维生素 D 治疗组糖尿病发病率减少了 62%。 13.肺部、血压和其他影响\n\n此外,根据 VIDA 试验,补充维生素 D 的患者的中心血压显著降低(−7.5 mmHg,p=0.03), 服用非甾体抗炎药的患者数量显著减少(RR,0.87,p=0.01)。此外,向正常人群和维生素 D 缺乏人群给予维生素 D 可改善肺功能,并减少与年龄相关的骨质流失,和 meta 分析一致。 在 15 项随机对照试验的个体参与者数据荟萃分析中,维生素 D 缺乏者每日或每周补充维生 素 D,定义为血清 25OHD 水平<10 ng/mL,可将急性呼吸道感染风险降低 30%(优势比, 0.30;95%可信区间,0.17-0.53)。\n\n14.方法问题\n\n不幸的是,VIDA、VITAL 和 D2d 研究的共同缺点是大多数受试者中基线 25OHD 水平在并不确 定。25OHD 平均基线水平 VIDA(24.2ng/mL 或 60.5 nmol/L)、VITA(L 30.8 ng/mL 或 77.0 nmol/L) 和 D2d(28.2 ng/mL 或 70.5nmol/L)都在医学研究所定义的正常范围内。只有 33%的 VIDA、 12.7%的 VITAL 和 20.7%的 D2d 受试者 25OHD 水平低于 20ng/mL(50 nmol/L)。这些研究的 一个重要结论是,它们并没有显示维生素 D 缺乏的人群会因补充维生素 D 而受益,因为这 些人群已经补足了维生素 D。如前所述,如果受试者已经超过了阈值营养素的水平,给予更 多不一定会获益更多。因此,基于这些研究,声称维生素 D 对癌症、心血管系统或糖尿病 的发生没有影响是没有证据支持的。这一陈述重要性的线索来源是对 D2d 研究的事后分析, 在该研究中,维生素 D 缺乏的受试者,即基线水平为 25OHD<12 ng/mL(30 nmol/L)的受 试者,如果在维生素 D 治疗组中,发现患糖尿病的风险降低图 3。上面提到的其他线索与血 压和肺部感染有关,其中维生素 D 似乎确实具有有益的作用。\n\n图片\n\n图 3 可观察到积极影响的 25OHD 阈值\n\n另一个方法论问题是研究的持续时间。人们必须考虑在癌症、心血管疾病或糖尿病发展之前 多长时间必须进行干预以影响显性疾病的发展。例如,托法多蒂尔研究的平均持续时间为 8.3 年,从没有癌症迹象到癌症诊断。补充维生素 D 5 年或更短时间是否有可能改变癌症变 得明显的时间进程?致癌作用通常是一个缓慢的过程,在未确诊的情况下以逐步的方式进行, 甚至可能在确诊前很多年。因此,即使在 5 年的干预期内不能降低患癌症的风险,但是一旦 发生癌症,补充维生素 D 可以提高生存率的发现,也是证明保持足够维生素 D 水平有益的 一个主要因素。\n\n需要对缺乏维生素 D 的受试者群体进行严格的维生素 D 补充研究,并将其与足够水平的维 生素 D 进行比较,以解决围绕维生素 D 潜在/据称的非骨骼效应的争议。对于癌症和心血管 疾病等终点,需要超过 5 年的研究,以清楚地证明风险效益的存在与否。癌症存活的益处似 乎得到了确凿的证明。\n\n结论\n\n结论在这篇综述中,我们强调了维生素 D 作为一种营养素和细胞作用调节因子的共识和不 确定性。虽然营养性佝偻病在全世界范围内定义明确且高度流行,但需要全球共同努力根除 这种可治愈的疾病。更好地理解维生素 D 的内源性产生及其代谢调节,以及开发具有适当 质量控制的通用分析方法,仍然是有价值的追求目标。尽管动物数据为维生素 D 的非骨骼 效应的假说提供了有用的背景,但从一些荟萃分析以及最近的随机对照试验来看,人类研究\n\n并没有得出任何明确的结论。我们期待着未来精心设计的研究,能够清楚地确定维生素 D 的作用在多大程度上是无处不在的,并延伸到骨骼之外。"} | ### 达成共识的议题
#### 1. 营养性佝偻病
营养性佝偻病，由单纯的维生素 D 或钙缺乏或两者兼而有之引起，仍然影响着全世界相当数量的婴儿和儿童。维生素 D 缺乏性佝偻病可以通过补充维生素 D 治疗。人们一致认为，婴儿和大多数儿童每天需要补充大约 400 国际单位（或大一点的儿童需要 600 国际单位）的维生素 D 来预防佝偻病，这是因为通常要避免直接暴露在阳光下，并且也不推荐非常年轻的人使用。然而，这种补充政策在许多国家要么没有得到执行，要么没有得到充分执行。
尽管亚洲和中东国家最常受到营养维生素 D 缺乏的影响，但非洲和一些亚洲国家也发生钙缺乏引起的佝偻病。对于 0 至 6 岁的新生儿和 6 至 12 个月大的婴儿，充足的钙摄入量分别为 200 毫克/天和 260 毫克/天，而对于 12 个月以上的儿童，膳食钙摄入量<300 毫克/天会增加佝偻病的风险，且与血清 25OHD 水平无关。对于 12 个月以上的儿童，根据膳食钙摄入量可分为：充足=>500mg/天；不足=300-500mg/天，缺乏=<300mg/天。
钙缺乏性佝偻病的发病机制可能比以前认为的更复杂。然而，我们确实知道减少钙摄入会增加 PTH 分泌，进一步导致 FGF-23 分泌增加。PTH 和 FGF-23 的增加导致尿磷酸盐排泄增加。这一病理生理现象导致血清磷酸盐减少，磷酸盐与 PTH 一起增加了 1,25(OH)D 水平。升高的 1,25(OH)D 上调了许多基因，导致焦磷酸盐（一种已知的骨矿化抑制剂）以及骨桥蛋白和小整合素结合配体 N 连接糖蛋白增加（SIBLINGS）。这些异常，以及低钙和低磷酸盐水平，是钙缺乏性骨质软化症特征的主要原因。尽管这种病理生理序列已经在动物身上得到证明，但很可能人类也受到同样的影响。
目前，全球不同国家主要基于临床体征的软骨病发病率仍然很高。基于软骨病在 |
Page 1
The Journal of Clinical Endocrinology & Metabolism, 2024, 109, 1907-1947
https://doi.org/10.1210/clinem/dgae290
Advance access publication 3 June 2024
Clinical Practice Guideline
ENDOCRINE
SOCIETY
OXFORD
Vitamin D for the Prevention of Disease: An Endocrine
Society Clinical Practice Guideline
7
8
Marie B. Demay, 1℗ Anastassios G. Pittas,² Daniel D. Bikle,³ Dima L. Diab,4 Mairead E. Kiely,5
Marise Lazaretti-Castro, Paul Lips, Deborah M. Mitchell, M. Hassan Murad, Shelley Powers, 10
Sudhaker D. Rao,11,12 Robert Scragg, ¹³ John A. Tayek,14,15 Amy M. Valent, 16 Judith M. E. Walsh,17
and Christopher R. McCartney
18,19
13
17
'Department of Medicine, Endocrine Unit, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, USA
2Division of Endocrinology, Diabetes and Metabolism, Department of Medicine, Tufts Medical Center, Boston, MA 02111, USA
³Departments of Medicine and Dermatology, University of California San Francisco, San Francisco VA Medical Center, San Francisco, CA
94158, USA
*Department of Internal Medicine, Division of Endocrinology, Diabetes and Metabolism, University of Cincinnati, Cincinnati, OH 45267, USA
5Cork Centre for Vitamin D and Nutrition Research, School of Food and Nutritional Sciences and INFANT Research Centre, University College
Cork, Cork, T12 Y337, Ireland
6Department of Internal Medicine, Division of Endocrinology, Universidade Federal de Sao Paulo, Sao Paulo 04220-00, Brazil
Endocrine Section, Amsterdam University Medical Center, Internal Medicine, 1007 MB Amsterdam, Netherlands
Pediatric Endocrine Unit, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, USA
Evidence-Based Practice Center, Mayo Clinic, Rochester, MN 55905, USA
10 Bone Health and Osteoporosis Foundation, Los Gatos, CA 95032, USA
11Division of Endocrinology, Diabetes and Bone & Mineral Disorders, Henry Ford Health, Detroit, MI 48202, USA
12 College of Human Medicine, Michigan State University, Lansing, MI 48824, USA
13 School of Population Health, The University of Auckland, Auckland 1142, New Zealand
14 Department of Internal Medicine, Harbor-UCLA Medical Center, Torrance, CA 90509, USA
15 The Lundquist Institute, Torrance, CA 90502, USA
16 Department of Obstetrics & Gynecology, Oregon Health & Science University, Portland, OR 97239, USA
17 Division of General Internal Medicine, Department of Medicine, University of California San Francisco, San Francisco, CA 94143, USA
18 Department of Medicine, University of Virginia, Charlottesville, VA 22908, USA
19 Department of Medicine, West Virginia University, Morgantown, WV 26506, USA
Correspondence: Marie B. Demay, MD, Professor of Medicine, Department of Medicine, Endocrine Unit, Massachusetts General Hospital and Harvard Medical
School, 50 Blossom St, Thier 1113, Boston, MA 02114, USA. Email: mdemay@mgh.harvard.edu.
Co-Sponsoring Organizations: American Association of Clinical Endocrinology (AACE), European Society of Endocrinology (ESE), Pediatric Endocrine Society
(PES), American Society for Bone and Mineral Research (ASBMR), Vitamin D Workshop, American Society for Nutrition (ASN), Brazilian Society of
Endocrinology and Metabolism (SBEM), Society of General Internal Medicine (SGIM), Endocrine Society of India (ESI)
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
Abstract
Background: Numerous studies demonstrate associations between serum concentrations of 25-hydroxyvitamin D (25[OH]D) and a variety of
common disorders, including musculoskeletal, metabolic, cardiovascular, malignant, autoimmune, and infectious diseases. Although a causal link
between serum 25(OH)D concentrations and many disorders has not been clearly established, these associations have led to widespread
supplementation with vitamin D and increased laboratory testing for 25(OH)D in the general population. The benefit-risk ratio of this increase
in vitamin D use is not clear, and the optimal vitamin D intake and the role of testing for 25(OH)D for disease prevention remain uncertain.
Objective: To develop clinical guidelines for the use of vitamin D (cholecalciferol [vitamin D3] or ergocalciferol [vitamin D2]) to lower the risk of
disease in individuals without established indications for vitamin D treatment or 25(OH)D testing.
Methods: A multidisciplinary panel of clinical experts, along with experts in guideline methodology and systematic literature review, identified
and prioritized 14 clinically relevant questions related to the use of vitamin D and 25(OH)D testing to lower the risk of disease. The panel prioritized
randomized placebo-controlled trials in general populations (without an established indication for vitamin D treatment or 25[OH]D testing),
evaluating the effects of empiric vitamin D administration throughout the lifespan, as well as in select conditions (pregnancy and
prediabetes). The panel defined "empiric supplementation" as vitamin D intake that (a) exceeds the Dietary Reference Intakes (DRI) and (b) is
implemented without testing for 25(OH)D. Systematic reviews queried electronic databases for publications related to these 14 clinical
questions. The Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) methodology was used to assess the
certainty of evidence and guide recommendations. The approach incorporated perspectives from a patient representative and considered
patient values, costs and resources required, acceptability and feasibility, and impact on health equity of the proposed recommendations. The
process to develop this clinical guideline did not use a risk assessment framework and was not designed to replace current DRI for vitamin D.
Received: 8 April 2024. Editorial Decision: 24 April 2024. Corrected and Typeset: 3 June 2024
© The Author(s) 2024. Published by Oxford University Press on behalf of the Endocrine Society. All rights reserved. For commercial re-use, please contact
reprints@oup.com for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions
link on the article page on our site-for further information please contact journals.permissions@oup.com. See the journal About page for additional terms.
Page 2
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1908
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
Results: The panel suggests empiric vitamin D supplementation for children and adolescents aged 1 to 18 years to prevent nutritional rickets
and because of its potential to lower the risk of respiratory tract infections; for those aged 75 years and older because of its potential to lower
the risk of mortality; for those who are pregnant because of its potential to lower the risk of preeclampsia, intra-uterine mortality, preterm
birth, small-for-gestational-age birth, and neonatal mortality; and for those with high-risk prediabetes because of its potential to reduce
progression to diabetes. Because the vitamin D doses in the included clinical trials varied considerably and many trial participants were
allowed to continue their own vitamin D-containing supplements, the optimal doses for empiric vitamin D supplementation remain
unclear for the populations considered. For nonpregnant people older than 50 years for whom vitamin D is indicated, the panel suggests
supplementation via daily administration of vitamin D, rather than intermittent use of high doses. The panel suggests against empiric
vitamin D supplementation above the current DRI to lower the risk of disease in healthy adults younger than 75 years. No clinical trial
evidence was found to support routine screening for 25(OH)D in the general population, nor in those with obesity or dark complexion,
and there was no clear evidence defining the optimal target level of 25(OH)D required for disease prevention in the populations
considered; thus, the panel suggests against routine 25(OH)D testing in all populations considered. The panel judged that, in most
situations, empiric vitamin D supplementation is inexpensive, feasible, acceptable to both healthy individuals and health care
professionals, and has no negative effect on health equity.
Conclusion: The panel suggests empiric vitamin D for those aged 1 to 18 years and adults over 75 years of age, those who are pregnant, and
those with high-risk prediabetes. Due to the scarcity of natural food sources rich in vitamin D, empiric supplementation can be achieved
through a combination of fortified foods and supplements that contain vitamin D. Based on the absence of supportive clinical trial
evidence, the panel suggests against routine 25(OH)D testing in the absence of established indications. These recommendations are not
meant to replace the current DRIs for vitamin D, nor do they apply to people with established indications for vitamin D treatment or
25(OH)D testing. Further research is needed to determine optimal 25(OH)D levels for specific health benefits.
Key Words: vitamin D, 25-hydroxyvitamin D, vitamin D deficiency, mortality, pregnancy, infection, prediabetes, clinical practice guidelines, systematic reviews
Introduction
The role of vitamin D in the regulation of skeletal and mineral
ion homeostasis is well established. Epidemiologic evidence
has shown consistent associations of low vitamin D status
with increased risk of a variety of common disorders, includ-
ing musculoskeletal, metabolic, cardiovascular, malignant,
autoimmune, and infectious diseases (1-3,). However, obser-
vational studies are prone to confounding and various forms
of bias, and a causal link between low vitamin D status, as as-
sessed by serum 25-hydroxyvitamin D (25[OH]D) levels, and
many disorders has not been clearly established. Nonetheless,
these associations have led to widespread supplementation
and increased laboratory testing for 25(OH)D levels in the
general population. In the United States, the prevalence of
supplemental vitamin D use of 1000 IU (25 µg) or more per
day increased from 0.3% in the 1999-2000 National Health
and Nutrition Examination Survey (NHANES) to 18.2% in
the 2013-2014 NHANES (4). The use of 25(OH)D testing
in clinical practice has also been increasing; however, the cost-
effectiveness of widespread testing has been questioned, espe-
cially given the uncertainty surrounding the optimal level of
25(OH)D required to prevent disease.
Vitamin D is not a true vitamin (defined as a nutrient that
cannot be endogenously synthesized), as intake is not required
in those who have adequate sun exposure. However, seasonal
variation in UV-B availability and decreased sun exposure as-
sociated with clothing and limited time outdoors has resulted
in the general population being increasingly reliant on oral in-
take of vitamin D in a few natural sources, foods fortified with
vitamin D, and supplements containing vitamin D. Whether
ingested or synthesized in the skin, vitamin D is converted to
25(OH)D in the liver (5). This process is not tightly regulated;
therefore, the 25(OH)D concentration most accurately reflects
vitamin D status. A second hydroxylation step (1-alpha) leads
to the formation of the active metabolite, 1,25-dihydroxyvita-
min D in many tissues. Circulating 1,25-dihydroxyvitamin D is
thought to derive primarily from renal 1-alpha hydroxylation
in the absence of pathologic conditions (6). Although loss of
function mutations in vitamin D hydroxylases are rare, genetic
variants and several pharmacologic agents may affect their ac-
tivity (7-10). Vitamin D metabolites are secreted with bile acids
and reabsorbed in the terminal ileum; therefore, terminal ileal
disease, as well as general malabsorption and having a short
gut (including from Roux-en-Y gastric bypass), can lead to
low levels of serum 25(OH)D. There are other conditions
that place individuals at risk for low 25(OH)D levels. For ex-
ample, vitamin D metabolites bound to vitamin D-binding
protein and albumin are lost in the urine of those with nephrot-
ic syndrome. In addition, vitamin D metabolites are inacti-
vated primarily by the 24-hydroxylase, which is induced by
high levels of 1,25-dihydroxyvitamin D as well as by fibroblast
growth factor-23, as seen in chronic kidney disease (11).
Importantly, these guidelines do not apply to individuals
with such underlying conditions that substantially alter vita-
min D physiology.
The actions of vitamin D metabolites are mediated by the
vitamin D receptor (VDR), which is expressed in most tissues.
The VDR has been shown to regulate cellular differentiation
and target gene expression in many cell types, including those
of the immune system. The best-established physiologic role of
the VDR is promoting intestinal calcium absorption, which is
critical for maintaining skeletal and mineral ion homeostasis
(12, 13). The skeletal effects of vitamin D are dependent on ad-
equate calcium intake. The effects of vitamin D on the immune
system are due to local activation of 25(OH)D to
1,25-dihydroxyvitamin D and induction of VDR expression
(14). Thus, the optimal level of 25(OH)D to prevent disease
likely depends on the clinical outcomes being evaluated.
Similarly, the required duration of exposure to vitamin D for
specific outcomes is expected to vary, depending on the under-
lying pathophysiology (eg, acute [infections] vs chronic
[cancer]).
The Guideline Development Panel's primary goal for this
guideline, which replaces the previous guideline (15), was
to establish clinical guidelines for the use of vitamin D to
lower the risk of disease in individuals without established in-
dications for vitamin D treatment or 25(OH)D testing. The
panel recognized that there are numerous important clinical
questions regarding the use of vitamin D and 25(OH)D
testing in the general population; however, due to limited re-
sources, 14 of these clinical questions were prioritized and
4 to 6 outcomes were addressed for each question. Because
patient-important clinical outcomes are expected to differ ac-
cording to the target population, the panel proposed specific
outcomes for the pediatric population (ages 1 to 18 years),
Page 3
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
and for ages 19 to 49 years, 50 to 74 years, and 75 years and
older. Established guidelines recommend empiric vitamin D in
the first year of life, specifically to prevent nutritional rickets
(16-18); thus, this demographic was not addressed. Other
populations examined were pregnant individuals and those
with prediabetes, dark complexion, and obesity. The panel
also addressed whether daily supplementation with vitamin
D should be recommended rather than intermittent (non-
daily), higher-dose vitamin D, and whether supplementation
should be limited to those with circulating 25(OH)D levels be-
low a threshold.
Evidence from randomized controlled trials (RCTs) was pri-
oritized for the systematic reviews. Large (> 1000 participants)
longitudinal observational cohort studies were considered if
they included appropriate comparators (supplementation vs
no supplementation) and outcomes, but only when an insuffi-
cient number of RCTs was available. Trials where the interven-
tion was a vitamin D analog or metabolite other than vitamin
D2 or vitamin D3 were excluded because these compounds are
not globally available. Mendelian randomization studies were
excluded because they do not evaluate response to supplemen-
tation. The Grading of Recommendations, Assessment,
Development, and Evaluation (GRADE) methodology was
used to assess the certainty of evidence and inform recommen-
dations. The panel sought evidence relevant to all elements of
the Evidence-to-Decision (EtD) framework, which included
stakeholder values and preferences (including input from clin-
ical experts and a patient representative), costs and other re-
sources required, cost-effectiveness, acceptability, feasibility,
and impact on health equity. The panel did not identify robust
evidence pertinent to these EtD factors for most clinical
questions.
Limitations
In formulating this Guideline, several challenges were encoun-
tered that influenced the formulation of the final
recommendations.
1. Because those with lower baseline levels of 25(OH)D
are expected to benefit more from vitamin D supple-
mentation than those with higher levels (19), a major
limitation in formulating recommendations was the
paucity of RCTs addressing the efficacy and safety
of vitamin D supplementation in populations with
low baseline 25(OH)D levels. Average baseline levels
of 25(OH)D in many large trials were in a range
that most would consider adequate (eg, 31 ng/mL
[78 nmol/L] in the VITAL trial) (20). In such trials, a
lack of effect of vitamin D does not necessarily indicate
that vitamin D does not influence the relevant outcome,
but rather that the study populations had baseline lev-
els of 25(OH)D that were adequate for the desired
outcome.
2. Unlike typical trials for pharmacologic agents, in which
control participants are not exposed to the intervention,
all participants in vitamin D trials were routinely ex-
posed to vitamin D through sun exposure and dietary
sources. In addition, many trials allowed participants
to remain on their current supplements that contained
vitamin D which often reflected the DRI (eg,
600-800 IU [15-20 µg] daily for adults). Such
1909
circumstances may have biased trial results toward the
null hypothesis.
3. Most vitamin D trials did not include a specific 25(OH)
D level as an eligibility criterion, and no trials were de-
signed or powered to address the effect of vitamin D in
subgroups stratified by either baseline or achieved
25(OH)D levels. This prevented the panel from propos-
ing thresholds for 25(OH)D adequacy or providing tar-
get 25(OH)D levels for disease prevention, especially
since 25(OH)D thresholds are likely to vary by popula-
tion and outcome. Although many systematic reviews
include subgroup analyses according to the study aver-
age baseline 25(OH)D levels, such analyses are subject
to ecological fallacy in which inferences about individu-
als are based on aggregate group data. Therefore, the
commissioned systematic review informing this guide-
line does not include study subgroup analyses according
to average baseline 25(OH)D levels.
4. Many trials were considered to be of insufficient duration
to adequately assess the effect of the vitamin D interven-
tion on some outcomes, due to the long latency for the de-
velopment of chronic diseases such as cancer, diabetes,
cardiovascular disease (CVD) and osteoporosis.
5. Because the included trials used various doses and ad-
ministration schedules of vitamin D, specific dose rec-
ommendations for vitamin D could not be proposed
for specific populations. Instead, in the technical re-
marks, vitamin D doses used in the included trials are
summarized.
6. The trials that the panel considered were performed in
overall healthy populations at average risk for the out-
comes of interest; therefore, the recommendations are lim-
ited to generally healthy individuals without established
indications for vitamin D treatment or 25(OH)D testing.
7. In most trials, study participants were largely of European
ancestry or identified as non-Hispanic White, with very
few trials including large numbers of participants from
other races or ethnicities.
8. The panel developed clinical questions for different age
groups
s of adults (<50 years, 50 to 74 years, and 75 years
and older) to represent different stages of life. However,
the panel recognizes the somewhat arbitrary nature of
these categories and acknowledges that many trials in-
cluded populations that spanned these age categories.
As a result, it was challenging to directly apply study re-
sults to narrowly defined age groups.
9. Many trials in those aged older than 50 years combined
vitamin D with calcium, making it difficult to isolate the
effect of vitamin D from that of calcium. This is especially
relevant to outcomes related to skeletal health, for which
both vitamin D and calcium are considered essential.
10. Due to resource limitations, not all potential outcomes
of interest were addressed in all populations of interest.
The panel prioritized outcomes that they felt were most
relevant to the specific populations under consideration.
Thus, these clinical guidelines relate to the use of vitamin D to
lower the risk of disease in individuals without established indi-
cations for vitamin D treatment or 25(OH)D testing. The
Guideline Development Panel assumed that the Institute of
Medicine's (IOM, now known as the National Academy of
Medicine) DRIS for vitamin D (21) represent a baseline standard
for all individuals. Importantly, the panel's recommendations
Page 4
1910
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
should not be extrapolated to those with underlying medical
conditions that are known to negatively impact vitamin D physi-
ology. For those living in countries where food fortification with
vitamin D is not standard or where dietary supplements are not
routinely used, interventions may be required to insure a baseline
intake consistent with the IOM DRIS.
Recommendation 3
In the general adult population younger than age 50
years, we suggest against routine 25(OH)D testing.
(2|0000)
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
List of Recommendations
Question 1. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used for chil-
dren and adolescents (ages 1 to 18 years)?
Recommendation 1
In children and adolescents aged 1 to 18 years, we
suggest empiric vitamin D supplementation to pre-
vent nutritional rickets and potentially lower the risk
of respiratory tract infections. (2| OO)
Technical remarks
• Empiric vitamin D may include daily intake of forti-
fied foods, vitamin formulations that contain vitamin
D, and/or daily intake of a vitamin D supplement (pill
or drops).
• In the clinical trials included in the systematic re-
view, with respect to respiratory tract infections
in children, vitamin D dosages ranged from 300
to 2000 IU (7.5 to 50 μg) daily equivalent. The esti-
mated weighted average was approximately
1200 IU (30 μg) per day.
Question 2. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used for non-
pregnant adults < 50 years of age?
Question 3. Should vitamin D supplementation vs no vita-
min D supplementation be used for nonpregnant adults
<50 years of age only when 25(OH)D levels are below a
threshold?
Recommendation 2
In the general adult population younger than age
50 years, we suggest against empiric vitamin D sup-
plementation beyond the recommended Dietary
Reference Intake for this population. (2 | OOO)
Technical remarks
• In this population, 25(OH)D levels that provide
outcome-specific benefits have not been estab-
lished in clinical trials.
• The panel suggests against (a) routine screening for
a 25(OH)D level to guide decision-making (ie, vita-
min D vs no vitamin D) and (b) routine follow-up test-
ing for 25(OH)D level to guide vitamin D dosing.
• This recommendation relates to generally healthy
adults who do not otherwise have established in-
dications for 25(OH)D testing (eg, hypocalcemia).
Question 4. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used for
adults aged 50 to 74 years?
Question 5. Should vitamin D supplementation vs no vita-
min D supplementation be used for adults aged 50 to 74
years only when 25(OH)D levels are below a threshold?
Recommendation 4
In the general population aged 50 to 74 years, we
suggest against routine vitamin D supplementation
beyond the recommended Dietary Reference Intake
for this population. (2|0000)
Technical remark
• This recommendation relates to empiric vitamin D
supplementation that exceeds the DRIs established
by the IOM. Adults in this age group should follow
the Recommended Daily Allowance established by
the IOM (600 IU [15 µg] daily for those aged 50 to
70 years; 800 IU [20 µg] daily for those older than
70 years).
Recommendation 5
In the general population aged 50 to 74 years, we sug-
gest against routine 25(OH)D testing. (2 |
OOO)
Technical remark
• This recommendation relates to empiric vitamin D
supplementation that exceeds the DRIs established
by the IOM. Adults in this age group should follow
the Recommended Daily Allowance established by
the IOM (600 IU [15 µg] daily).
Technical remarks
• In this population, 25(OH)D levels that provide out-
come-specific benefits have not been established
in clinical trials.
Page 5
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
• The panel suggests against (a) routine screening for
a 25(OH)D level to guide decision-making (ie, vita-
min D vs no vitamin D) and (b) routine follow-up
testing for 25(OH)D level to guide vitamin D dosing.
• This recommendation relates to generally healthy
adults who do not otherwise have established in-
dications for 25(OH)D testing (eg, hypocalcemia).
Question 6. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used by adults
aged ≥75 years?
Question 7. Should vitamin D supplementation vs no vita-
min D supplementation be used by adults aged ≥75
years only when 25(OH)D levels are below a threshold?
1911
Question 8. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used during
pregnancy?
Question 9. Should vitamin D supplementation vs no vita-
min D supplementation be used during pregnancy only
when 25(OH)D levels are below a threshold?
Recommendation 8
We suggest empiric vitamin D supplementation dur-
ing pregnancy, given its potential to lower risk of pre-
eclampsia, intra-uterine mortality, preterm birth,
small-for-gestational-age (SGA) birth, and neonatal
mortality. (2| Ð○○)
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
Recommendation 6
In the general population aged 75 years and older, we
suggest empiric vitamin D supplementation because
of the potential to lower the risk of mortality.
(20000)
Technical remarks
• Empiric vitamin D may include daily intake of forti-
fied foods, vitamin formulations that contain vita-
min D and/or daily intake of a vitamin D supplement.
• For empiric supplementation, daily, lower-dose
vitamin D is preferred over nondaily, higher doses.
In the clinical trials included in the systematic re-
view that reported on the mortality outcome, vita-
min D dosage ranged from 400 to 3333 IU (10 to
83 µg) daily equivalent. The estimated weighted
average was approximately 900 IU (23 μg) daily.
Participants in many trials were allowed to remain
on their routine supplements, including up to 800
IU (20 µg) of vitamin D daily.
Technical remarks
• This recommendation is based on evidence from
trials conducted in healthy individuals during
pregnancy.
• Empiric vitamin D may include daily intake of forti-
fied foods, prenatal vitamin formulations that con-
tain vitamin D, and/or a vitamin D supplement
(pills or drops).
• In the clinical trials included in the systematic re-
view, the vitamin D dosages ranged from 600 IU
to 5000 IU (15 to 125 μg) daily equivalent, usually
provided daily or weekly. The estimated weighted
average was approximately 2500 IU (63 μg) per
day.
Recommendation 9
During pregnancy, we suggest against routine 25(OH)
D testing. (2 | OOO)
Recommendation 7
In the general population aged 75 years and older, we
suggest against routine testing for 25(OH)D levels. (2|
୫୦୦୦)
Technical remarks
• In this population, 25(OH)D thresholds that pro-
vide outcome-specific benefits have not been es-
tablished in clinical trials.
• The panel suggests against (a) routine screening for
a 25(OH)D level to guide decision-making (ie, vita-
min D vs no vitamin D) and (b) routine follow-up
testing for 25(OH)D level to guide vitamin D dosing.
• This recommendation relates to generally healthy
adults who do not otherwise have established in-
dications for 25(OH)D testing (eg, hypocalcemia).
Technical remarks
• In this population, 25(OH)D levels that provide
pregnancy outcome-specific benefits have not
been established in clinical trials.
• The panel suggests against (a) routine screening
for a 25(OH)D level to guide decision-making (ie,
vitamin D vs no vitamin D) and (b) routine follow-
up testing for 25(OH)D level to guide vitamin D
dosing.
• This recommendation relates to generally healthy
pregnant individuals who do not otherwise have
established indications for 25(OH)D testing (eg,
hypocalcemia).
Question 10. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used for adults
with prediabetes (by glycemic criteria)?
Page 6
1912
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
Recommendation 10
For adults with high-risk prediabetes, in addition to
lifestyle modification, we suggest empiric vitamin D
supplementation to reduce the risk of progression to
diabetes. (2000)
Technical remarks
• In healthy adults, 25(OH)D levels that provide out-
come-specific benefits have not been established
in clinical trials.
• This recommendation relates to adults who do not
otherwise have established indications for testing
with 25(OH)D levels (eg, hypocalcemia).
Technical remarks
• Lifestyle modification must be a routine manage-
ment component for adults with prediabetes.
• The clinical trials informing this recommenda-
tion primarily related to adults with high-risk pre-
diabetes, identified as meeting 2 or 3 American
Diabetes Association glycemia criteria (fasting
glucose, glycated hemoglobin [HbA1c], 2-hour
glucose after a 75-gram oral glucose challenge)
for prediabetes and those with impaired glucose
tolerance.
• In the clinical trials included in the systematic review,
the vitamin D dosages ranged from 842 to 7543 IU
(21 to 189 µg) daily equivalent. The estimated
weighted average was approximately 3500 IU
(88 µg) per day. Participants in some trials were al-
lowed to remain on their routine supplements, in-
cluding up to 1000 IU (25 µg) of vitamin D daily.
Question 11. Should a daily, lower-dose vitamin D vs non-
daily (ie, intermittent), higher-dose vitamin D be used
for nonpregnant people for whom vitamin D treatment
is indicated?
Question 13. Should screening with a 25(OH)D test (with
vitamin D supplementation/treatment only if below a
threshold) vs no screening with a 25(OH)D test be used
for adults with dark complexion?
Recommendation 13
In adults with dark complexion, we suggest against
routine screening for 25(OH)D levels. (2 | OOO)
Technical remarks
• This recommendation relates to generally healthy
adults with dark complexion who do not otherwise
have established indications for 25(OH)D testing
(eg, hypocalcemia).
• The panel did not identify any clinical trials that re-
lated clinical outcomes to skin complexion per se.
A secondary analysis did not clearly suggest net
benefit with vitamin D in those who self-identify
as Black. The panel recognized that self-identified
race is an inaccurate and otherwise problematic
proxy for dark complexion.
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
Recommendation 11
In adults aged 50 years and older who have indica-
tions for vitamin D supplementation or treatment,
we suggest daily, lower-dose vitamin D instead of
nondaily, higher-dose vitamin D. (2 | OO)
Question 14. Should screening with a 25(OH)D test (with
vitamin D supplementation/treatment only if below a
threshold) vs no screening with a 25(OH)D test be used
for adults with obesity?
Technical remark
• The panel did not identify evidence related to indi-
viduals younger than age 50 years.
Recommendation 14
In adults with obesity, we suggest against routine
screening for 25(OH)D levels. (2 | OOO)
Question 12. Should screening with a 25(OH)D test (with
vitamin D supplementation/treatment only if below a
threshold) vs no screening with a 25(OH)D test be
used for healthy adults?
Recommendation 12
In healthy adults, we suggest against routine screen-
ing for 25(OH)D levels. (2 | OOO)
Technical remarks
• In adults with obesity, 25(OH)D thresholds that pro-
vide outcome-specific benefits have not been es-
tablished in clinical trials.
• This recommendation relates to generally healthy
adults with obesity who do not otherwise have
established indications for 25(OH)D testing (eg,
hypocalcemia).
Page 7
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
Notes:
• The Guideline Development Panel did not find clinical tri-
al evidence that would support establishing distinct
25(OH)D thresholds tied to outcome-specific benefits in
the populations examined. Hence, the Endocrine Society
no longer endorses the target 25(OH)D level of 30 ng/
mL (75 nmol/L) suggested in the previous guideline (15).
Similarly, the Endocrine Society no longer endorses specif-
ic 25(OH)D levels to define vitamin D sufficiency, insuffi-
ciency, and deficiency.
• The current guideline suggests against routine 25(OH)D
screening (in the absence of well-established indications),
including in adults and children with obesity, in adults
and children with dark complexion, and during pregnancy.
This also represents a change from the 2011 guideline (15).
Methods of Development of Evidence-Based
Clinical Practice Guidelines
This guideline was developed using the process detailed on the
Endocrine Society website (https://www.endocrine.org/
clinical-practice-guidelines/methodology) and summarized
here. The Endocrine Society follows the GRADE (Grading
of Recommendations Assessment, Development and
Evaluation) methodology (22) (Tables 1 and 2). This method-
ology includes the use of evidence-to-decision (EtD) frame-
works to ensure all important criteria are considered when
making recommendations (23, 24). The process was facili-
tated by the GRADEpro Guideline Development Tool
(GRADEpro GDT) (25). This Guideline Development Panel
(GDP) consisted of content experts representing the following
specialties: adult endocrinology, general internal medicine,
obstetrics and gynecology, pediatric endocrinology, nutrition,
and epidemiology. A patient representative was also included
on the panel. Members were identified by the Endocrine
Society Board of Directors and the Clinical Guidelines
Committee (CGC) and were vetted according to the
conflict-of-interest policy (26), which was adhered to
throughout the guideline process to manage and mitigate
conflicts of interest. Detailed disclosures of panel members
and the management strategies implemented during the de-
velopment process can be found in Appendix A. In addition,
the group included a clinical practice guideline methodolo-
gist from the Mayo Evidence-Based Practice Center, who
led the team that conducted the systematic reviews and
meta-analyses, and a methodologist from the Endocrine
Society, who advised on methodology and moderated the
application of the EtD framework and development of the
recommendations.
From the Guideline Development Panel, 2 to 3 members
were assigned to lead each guideline question. The clinical
questions addressed in this guideline were prioritized from
an extensive list of potential questions through a survey of
the panel members and discussion; 14 questions were identi-
fied as most important. The Mayo Evidence-Based Practice
Center conducted a systematic review for each question and
produced GRADE evidence profiles that summarized the
body of evidence for each question and the certainty of the
evidence (29). The systematic searches for evidence were
conducted in February 2022 and updated in December
1913
2023. In parallel to the development of the evidence summar-
ies, the Guideline Development Panel members searched for
and summarized research evidence for other EtD criteria,
such as patients' values and preferences, feasibility, accept-
ability, costs/resource use, cost-effectiveness, and health
equity. Research evidence summaries noted in the EtD frame-
works were compiled using standardized terminology tem-
plates for clarity and consistency (30). During an in-person
panel meeting and a series of video conferences, the
Guideline Development Panel judged the balance of benefits
and harms, in addition to the other EtD criteria, to determine
the direction and strength of each recommendation (30, 31);
see Tables 1 and 2.
The draft recommendations were posted publicly for
external peer review and internally for Endocrine Society
members, and the draft guideline manuscript was reviewed
by the Society's Clinical Guidelines Committee, representa-
tives of co-sponsoring organizations, a representative of the
Society's Board of Directors, and an Expert Reviewer.
Revisions to the guideline were made based on submitted
comments and approved by the Clinical Guidelines
Committee, the Expert Reviewer, and the Board of
Directors. Finally, the guideline manuscript was reviewed be-
fore publication by the Journal of Clinical Endocrinology
and Metabolism's publisher's reviewers.
This guideline will be reviewed annually to assess the state
of the evidence and determine if there are any developments
that would warrant an update to the guideline.
Evidence-to-Decision Considerations Common
to Multiple Clinical Questions
Many of the EtD considerations were common to the clinical
questions addressing empiric vitamin D supplementation.
Most multivitamins contain 800 to 1000 IU (20-25 μg) of
vitamin D. Vitamin D is inexpensive and available without a
prescription, at costs varying from the equivalent of US $10
to $50 per year in North America, South America, New
Zealand, Europe, and India. Because empiric vitamin D sup-
plementation intervention would be limited to a daily supple-
ment that is readily available, the panel judged that the
intervention would be acceptable and feasible. Most vitamin
D3 on the market is from animal sources (lanolin), but vegan
vitamin D3 from lichen is also available. Vitamin D2, which is
plant-based, is widely available, and the costs are similar.
Evaluations of costs, acceptability, and feasibility refer to rou-
tine vitamin D use in the general population, and special con-
siderations that pertain to children and specific demographics
are discussed elsewhere.
When beneficial effects of empiric vitamin D were identi-
fied, the panel judged that empiric vitamin D will not likely
have a negative impact on health equity and may have a
favorable impact on improving health equity because low
vitamin D status is more prevalent in disadvantaged popula-
tions, including those with lower socioeconomic status. In
addition, disadvantaged persons tend to be at higher base-
line risk for many of the outcomes assessed (eg, poor
maternal-fetal outcomes, nutritional rickets, diabetes);
thus, whenever benefit is expected for such outcomes, disad-
vantaged populations would be expected to derive greater
absolute benefit.
Page 8
1914
Ages 1-18
Empiric vitamin D
supplementation*
To prevent nutritional
rickets and because
of the potential to
lower the risk of
respiratory tract
infections.
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
Ages 19-49 Ages 50-74 Ages ≥75
No empiric vitamin D
supplementation*
Follow the Institute of Medicine
Recommended Daily Allowance.
Empiric vitamin D
supplementation*
Because of the
potential to lower the
risk of mortality.
Pregnancy
Empiric vitamin D
supplementation*
Because of the
potential to lower the
risk of preeclampsia,
intrauterine mortality,
preterm birth, small for
gestational age birth
and neonatal mortality.
Prediabetes
Empiric vitamin D
supplementation*
Because of the
potential to lower the
risk of progression to
diabetes.
The panel assumed that all should follow the Recommended Dietary Reference Intakes (DRI) established by the US Institute of Medicine
(currently the National Academy of Medicine). The Recommended Daily Allowance (RDA) in the DRI is 600 IU (15 μg) for persons aged 1-70
years and 800 IU (20 μg) for persons older than 70 years. The RDA established by the the Institute of Medicine and the American College of
Obstetricians and Gynecologists (ACOG) is 600 IU (15 μg) during pregnancy.
* Empiric vitamin D supplementation refers to vitamin D (cholecalciferol [D] or ergocalciferol [D2]) intake (usually in pill or drop form) that
(a) exceeds the DRIs and (b) is implemented without testing for 25-hydroxyvitamin D. Vitamin D doses in the included clinical trials varied
considerably (see technical remarks under recommendations); hence, optimal doses remain unclear.
For people older than 50 years for whom vitamin D treatment is indicated, the panel suggests supplementation via daily administration of
vitamin D, rather than intermittent high doses.
The panel suggests against routine 25-hydroxyvitamin D testing for generally healthy individuals who do not otherwise have established
indications for 25-hydroxyvitamin D testing (e.g., hypocalcemia). The panel did not specifically address whether and how those who present
with low levels of 25-hydroxyvitamin D should be evaluated and/or treated.
* Importantly, this guideline does not address individuals with underlying conditions that substantially alter vitamin D physiology, including
various conditions associated with decreased absorption (e.g., short gut, gastric bypass, inflammatory bowel disease), increased
catabolism/decreased activation (e.g., some medications), and increased renal losses (e.g., nephrotic syndrome). In addition, this guideline
does not address persons known to be at high risk for fractures.
Figure 1. Vitamin D for the prevention of disease.
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
Table 1. GRADE certainty of evidence classifications
Certainty
of evidence
High
0000
Moderate
0000
Low
0000
Very Low
#000
Interpretation
We are very confident that the true effect lies close to that of
the estimate of the effect.
We are moderately confident in the effect estimate; the true
effect is likely to be close to the estimate of the effect, but
there is a possibility that it is substantially different.
Our confidence in the effect estimate is limited; the true effect
may be substantially different from the estimate of the effect.
We have little confidence in the effect estimate; the true
effect is likely to be substantially different from the estimate
of effect.
very
Source: Reprinted with permission from Schünemann HJ, Brożek J, Guyatt GH,
Oxman AD. GRADE Handbook. Handbook for grading the quality of evidence
and the strength of recommendations using the GRADE approach. Updated
October 2013 (27).
test results would be acceptable to many, the panel judged
that such an approach may be unacceptable to some. In add-
ition, access to accurate 25(OH)D testing is variable across the
globe, and an approach requiring such testing may not be feas-
ible in some settings.
Even if there were beneficial effects to screening with
25(OH)D and treating based on the results, the panel was un-
certain about the impact of such an approach on health equity.
While the panel acknowledged the increased prevalence of low
vitamin D status in disadvantaged populations, those with low
socioeconomic status, and those with dark complexion, the
costs and time commitment required to implement the inter-
vention I may limit its acceptability and feasibility in these pop-
ulations and those across the globe with poor access to health
care.
For each clinical question, additional details regarding all
EtD considerations are included in the supplemental materials
available online.
When the intervention involved testing for 25(OH)D prior
to treatment with vitamin D, the costs were felt to be moderate
and the intervention less acceptable. The cost of a 25(OH)D
assay varies from the equivalent of US $25 to $100 in North
America, South America, New Zealand, and Europe.
However, this does not include the cost of health care visits
for ordering the test, interpreting the test result, and the poten-
tial need for additional testing and health care visits. Thus,
while conditioning vitamin D supplementation on 25(OH)D
Vitamin D Use for Children Aged 1 to 18 Years
Background
The prevalence of low vitamin D status in childhood is high,
with marked variability across the globe. In the United
States, the population-based NHANES 2011-2014 survey
found 25(OH)D levels lower than 20 ng/mL (50 nmol/L) in
7% of 1- to 5-year-olds, 12% of 6- to 11-year-olds, and
23% of 12- to 19-year-olds (32). Much higher prevalence of
low vitamin D status is found in Northern Europe (33) and in
Page 9
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
Table 2. GRADE strength of recommendation classifications and interpretation
Strength of
recommendation
1-Strong
recommendation
for or against
2-Conditional
recommendation
for or against
Criteria
Desirable consequences
CLEARLY
OUTWEIGH the
undesirable
consequences in most
settings (or vice versa)
Desirable consequences
PROBABLY
OUTWEIGH
undesirable
consequences in most
settings (or vice versa)
Interpretation by patients
Most individuals in this
situation would want the
recommended course of
action, and only a small
proportion would not.
The majority of individuals
in this situation would
want the suggested course
of action, but many
would
not.
Decision aids may be useful
in helping patients make
decisions consistent with
their individual risks,
values and preferences.
Interpretation by health care
providers
Most individuals should follow
the recommended course of
action.
Formal decision aids are not
likely to be needed to help
individual patients make
decisions consistent with their
values and preferences.
Clinicians should recognize that
different choices will be
appropriate for each
individual and that clinicians
must help each individual
arrive at a management
decision consistent with the
individual's values and
preferences.
1915
Interpretation by policy makers
The recommendation can be
adopted as policy in most
situations.
Adherence to this recommendation
according to the guideline could
be used as a quality criterion or
performance indicator.
Policymaking will require
substantial debate and
involvement of various
stakeholders. Performance
measures should assess whether
decision-making is appropriate.
Source: Reprinted from Schünemann HJ et al. Blood Adv, 2018;2(22):3198-3225. © The American Society of Hematology, published by Elsevier (28).
low- and middle-income countries, where the majority of chil-
dren have 25(OH)D concentrations lower than 10 ng/mL
(25 nmol/L) (34). Particularly high-risk pediatric groups in-
clude children with limited exposure to sunlight, children
with dietary restrictions, and children with high skin melanin
content.
Several well-established guidelines recommend empiric
vitamin D in the first year of life, specifically to prevent nu-
tritional rickets (16-18); thus, in this guideline and the asso-
ciated systematic review, the panel did not address children
aged 0-1 years. However, nutritional rickets is not limited to
infancy. While rickets is often considered a historical dis-
ease, its incidence is rising in high-income countries.
Recent surveys indicate an incidence of up to 24 per 100
000 patient-years in North America, Australia, and Europe
(35). In Western countries, rickets mainly affects children
from racial and ethnic minority groups and non-Western im-
migrants and refugees (36, 37). In low- and middle-income
countries in the Middle East, South Asia, and Africa, the
burden of nutritional rickets is substantially higher, with re-
ported prevalence of 1% to 24% (35, 38). In Turkey, a sur-
vey of 946 children with rickets showed peak incidences at
ages 0 to 2 years and 12 to 15 years (39), indicating risk
throughout childhood and adolescence. Nutritional rickets
leads to pain, deformity, delayed milestone acquisition,
and poor growth, and can be complicated by seizure and di-
lated cardiomyopathy (40).
Vitamin D has been implicated in the prevention of respira-
tory infections, which are very common in children, with
pneumonia being the most common infectious cause of death
in the first 5 years of life (41-45). In addition, low vitamin D
status is associated with tuberculosis infection (46), another
major cause of childhood mortality, with an estimated
230 000 deaths annually (47). A potential role for vitamin
D in additional health outcomes affecting childhood, includ-
ing autoimmune disease, atopy, and diabetes, has also been
proposed. For example, several Mendelian randomization
studies have suggested an association between genetically
determined 25(OH)D levels and multiple sclerosis (48-51).
In addition, vitamin D is thought to play a role in immunity,
and childhood offers a unique window of opportunity to
train the immune system (52). Bone health is also important
during childhood, since peak bone mass accrual occurs dur-
ing this period, extending into early adulthood. Thus, inad-
equate vitamin D status in childhood may affect disease
vulnerability throughout the lifespan. The guideline panel
therefore addressed the question of whether empiric vitamin
D supplementation should be continued throughout child-
hood and adolescence.
Question 1. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used for chil-
dren and adolescents (aged 1 to 18 years)?
Recommendation 1
In children and adolescents aged 1 to 18 years, we
suggest empiric vitamin D supplementation to pre-
vent nutritional rickets and potentially lower the risk
of respiratory tract infections. (2 | OO)
Technical remarks
• Empiric vitamin D may include daily intake of
fortified foods, vitamin formulations that contain
vitamin D, and/or daily intake of a vitamin D sup-
plement (pill or drops).
• In the clinical trials included in the systematic re-
view, with respect to respiratory tract infections
in children, vitamin D dosages ranged from 300
to 2000 IU (7.5 to 50 µg) daily equivalent. The esti-
mated weighted average was approximately
1200 IU (30 µg) per day.
Page 10
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1916
Summary of Evidence
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
The evidence summaries, meta-analysis results, and a
detailed summary of the evidence and EtD tables can be
found online at https://guidelines.gradepro.org/profile/
gNMKfIPr5u4.
Benefits and Harms
The systematic review found no RCTs on the efficacy of vita-
min D in children and adolescents to prevent symptomatic nu-
tritional rickets. This was because vitamin D supplementation
was studied and implemented widely for prevention of rickets
long before clinical trial methodology was standardized (53),
and a placebo-controlled trial for nutritional rickets would
currently be considered unethical. Several lines of evidence,
however, indicate that vitamin D supplementation prevents
the development of nutritional rickets in children. In 1917,
Hess and Unger treated 49 infants and toddlers aged 1 month
to 17 months, who were at high risk of rickets, with cod liver
oil, the active ingredient of which is vitamin D, and then com-
pared them with 16 infants and children in the same commu-
nity. Eight of 49 infants in the treatment group and 15 of 16
in the control group developed rickets (odds ratio 0.18,
P=.002) (54). Chick and colleagues in Vienna compared in-
stitutionalized infants fed either a standard diet or one en-
riched with cod liver oil from 1920 to 1922 and observed
that 58% of the control group developed rickets compared
to none in the cod liver oil group (55). The institution of a
free vitamin D distribution program (400 IU/d [10 µg/d]) in
Turkey was associated with a reduction in the prevalence of
nutritional rickets from 6% in 1998 to 0.1% in 2008 (56).
These and other data were summarized in an earlier systemat-
ic review (18). While these interventions were primarily in in-
fants, the panel judged that these observations can be
reasonably generalized to all children with open growth plates
at risk for nutritional rickets.
The systematic review informing this guideline identified
12 RCTs (57-68) (12951 participants) reporting on the
effect of vitamin D on the incidence of respiratory infection,
with individuals experiencing any respiratory infection
representing the unit of analysis. Five of these trials were
conducted in South Asia (India and Bangladesh), 5 trials in
East Asia (Taiwan, Vietnam, Mongolia, and Japan), and 1
each in Afghanistan and Israel. Vitamin D regimens varied
greatly, ranging from daily dosing of 300 to 2000 IU (7.5
to 50 μg), weekly dosing of 10000 and 14 000 IU (250
and 350 µg), and a single dose of 100 000 (2500 µg) to
120 000 IU (3000 µg). The relative risk (RR) for developing
any respiratory tract infection was 0.94 (95% CI,
0.87-1.02), with an estimated absolute effect size of 43 few-
er respiratory infections per 1000 (93 fewer to 14 more).
Studies that had some concern for bias showed a lower
risk (RR 0.75 [95% CI, 0.61-0.94]) while studies with low
risk of bias showed no difference in risk (RR 0.99 [95%
CI, 0.92-1.07]) (P for heterogeneity 0.022). Study subgroup
analyses did not implicate vitamin D dosage or study partici-
pant age (younger vs older than 5 years) as significant pre-
dictors of outcomes. Among 6 trials (58, 60, 63, 64, 66,
68) that reported lower respiratory tract infection specifically
(10 356 participants), the RR for infection was 0.93 (95%
CI, 0.83-1.04), with an absolute effect size of 33 fewer lower
respiratory infections per 1000 (81 fewer to 19 more). Study
per
subgroup analysis suggested the possibility that higher vita-
min D dosages led to greater reductions in lower respiratory
tract infection risk (RR 0.82 [95% CI, 0.68-1.00]) compared
to standard dosages (RR 0.98 [95% CI, 0.94-1.03]), although
this was not a statistically significant interaction (P = .087).
The RR of developing tuberculosis (2 trials, 10533
participants) (68, 69) was 0.67 (95% CI, 0.14-3.11) in those
supplemented with vitamin D (10000 and 14 000 IU
[250-350 μg] weekly) with an absolute effect size of 1 fewer
1000 (from 2 fewer to 6 more). Three trials (58, 62, 70) re-
ported data on the total number of respiratory infections as the
unit of analysis. After combining data from these
trials, the incidence rate ratio (IRR) favored vitamin D (0.64
[95% CI, 0.51-0.82]). Supporting this finding was the trial in
which all patients had at least one acute respiratory infection
in the 6 months following the intervention, but the proportion
who had at least 3 infections was lower in the intervention
group (7.7% vs 32.4%) (67). Study subgroup analyses did
not strongly implicate study risk of bias or vitamin D dosage
as significant predictors of these outcomes.
The panel found limited RCT data on the impact of vitamin
D on the incidence of autoimmune disease, allergic disease,
and asthma, with too few events to analyze. The panel found
no RCT data on the effect of treating this population with
vitamin D to lower the risk of prediabetes and type 2 diabetes,
or fractures (in adulthood).
The systematic review did not find clear evidence that vita-
min D increases adverse events in children. Available trials
documented one case of symptomatic hypercalcemia in an in-
dividual assigned to vitamin D and one case of kidney failure
in an individual assigned to the control group; there were no
reported kidney stones.
Based on the panel's best estimates of treatment effects, the
panel judged that the anticipated desirable effects of empiric
vitamin D supplementation are likely to be beneficial for
many, and that the anticipated undesirable effects are likely
to be trivial for all.
Other Evidence-to-Decision Criteria and
Considerations
The cost of vitamin D supplementation is low, although vari-
able in different countries. Cost-effectiveness of universal vita-
min D supplementation for the prevention of rickets has been
addressed in economic modeling studies in the United
Kingdom. Two studies suggested that targeted vitamin D ad-
ministration to those with moderate-to-dark complexion (de-
fined in the study as having Afro-Caribbean ancestry) and
those with Asian ancestry would be either cost-saving or cost-
effective (71, 72). An additional study suggested that universal
vitamin D supplementation via flour fortification would be
cost-saving, while targeted supplementation of children
would be cost-effective (73). Given that the risk of nutritional
rickets is likely substantially increased among children with
darker complexion and among immigrants to high-income
countries (35, 40, 74-77)—populations that may experience
lower health equity as a group—the panel concluded that vita-
min D supplementation in children could potentially improve
health equity.
There is limited evidence regarding the acceptability of vita-
min D supplementation in children and among their care-
givers. In one trial in which children aged 9 to 12 years were
Page 11
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
offered various forms of vitamin D and calcium, 44% agreed
to continue fortified milk, 66% agreed to fortified orange
juice, and 95% agreed to supplements, suggesting that supple-
ment use may be the most accepted formulation (78). In one
small survey study in the United Kingdom, approximately
25% of caregivers aware of governmental recommendations
about vitamin D supplementation were not adherent to the
recommendations. Reasons for nonadherence included the
child's dislike of drops, low priority, and belief that other
strategies such as breastfeeding, outdoor play, and a varied
diet were sufficient (79).
Justification for the Recommendation
Given the high stakes of very low vitamin D status during skel-
etal growth the risk of nutritional rickets in particular―the
panel judged that empiric vitamin D supplementation may be
prudent in growing children/adolescents, especially for those
who are not otherwise expected to have adequate vitamin D
stores via sun exposure (for example, from adequate levels
of sun-safe outdoor physical activity) and ingestion of vitamin
D-containing or vitamin D-fortified foods, and those for
whom confidence is low that IOM DRIS are being achieved re-
liably. The panel agreed that low- to moderate-certainty evi-
dence suggests that vitamin D supplementation in children
may be beneficial for respiratory infections, which are a lead-
ing cause of mortality. The panel also concluded that supple-
mentation costs are generally low, that supplementation is
likely to be feasible and acceptable, and that empiric supple-
mentation may improve health equity. Given the low overall
certainty of evidence, and since net benefits may vary accord-
ing to individual circumstances, a conditional recommenda-
tion was issued.
Additional Considerations
The optimal dosage for prevention of respiratory tract infections
in children remains uncertain. In the trials included in this sys-
tematic review, the vitamin D dosages ranged from 300 to
2000 IU (7.5 to 50 µg) daily equivalent. The estimated median
vitamin D dosage used in these studies was 811 IU (20 µg) daily,
and estimated weighted average dosages were 1203 IU (30 μg)
per day for the any respiratory infection outcome and 1473 IU
(37 µg) per day for the lower respiratory tract infection outcome.
(Here and elsewhere in this document, the estimated weighted
average dosage for an outcome represents each relevant study's
vitamin D dosage weighted according to the study's weight in the
meta-analysis for that outcome.)
Research Considerations
Proposed areas for research include:
1. Adequately powered trials among children with appro-
priate controls to detect rare outcomes and long-term
follow-up should be conducted in specific populations
(eg, children with a history of asthma, risk of type 1 dia-
betes, new-onset type 1 diabetes) with outcomes specific
to these populations (eg, asthma exacerbations, incident
type 1 diabetes, progression of type 1 diabetes).
2. Since the majority of trials in children were conducted in
Asia, it is important to undertake studies examining the
effects of vitamin D on outcomes in other populations
1917
that may differ in terms of diet, sun exposure, and
complexion.
Vitamin D Use in Nonpregnant Adults Aged
< 50 Years
Background
While adults younger than age 50 years have lower health care
usage compared to older individuals (80), this is a critical time
during which many chronic diseases linked to environmental
and nutritional factors develop. A significant percentage of
adults in this age group have low vitamin D status. Levels of
25(OH)D lower than 12 ng/mL (30 nmol/L) were seen in
14% of Europeans and lower than 20 ng/mL (50 nmol/L) in
40% (81). In the United States, 24% and 6% of adults have
25(OH)D levels lower than 20 ng/mL (50 nmol/L) and
10 ng/mL (25 nmol/L), respectively (82). Numerous studies
have found associations between low 25(OH)D levels, low
BMD, and fractures. Low 25(OH)D levels have also been as-
sociated with fatigue and higher risks for respiratory infec-
tions, including COVID-19 (83).
The age span of 18 to 50 years is when peak bone mass oc-
curs, and the National Osteoporosis Foundation's systematic
review and implementation recommendations suggest that
vitamin D plays an important role in peak bone mass accrual
(84), which has implications for risk of osteoporotic fractures
later in life. Most pregnancies occur between ages 19 and 50
years, and, while pregnancy-specific recommendations are ad-
dressed elsewhere, those who are pregnant most often do not
present for care before the end of the first trimester, and hav-
ing adequate vitamin D status preconception may be import-
ant. Fatigue is also common in this age group and, like
respiratory infections, contributes to loss of productivity
and increased medical care.
Question 2. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used for non-
pregnant adults < 50 years of age?
Question 3. Should vitamin D supplementation vs no vita-
min D supplementation be used for nonpregnant adults
<50 years of age only when 25(OH)D levels are below a
threshold?
Recommendation 2
In the general adult population younger than age 50
years, we suggest against empiric vitamin D sup-
plementation beyond the recommended Dietary
Reference Intake for this population. (2 | OOO)
Technical remark
• This recommendation relates to empiric vitamin D
supplementation that exceeds the DRIS estab-
lished by the IOM. Adults in this age group should
follow the Recommended Daily Allowance estab-
lished by the IOM (600 IU [15 μg] daily).
Page 12
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1918
Recommendation 3
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
In the general adult population younger than age
50 years, we suggest against routine 25(OH)D testing.
(2❘ OOOO)
Technical remarks
• In this population, 25(OH)D levels that provide
outcome-specific benefits have not been estab-
lished in clinical trials.
• The panel suggests against (a) routine screening
for a 25(OH)D level to guide decision-making (ie,
vitamin D vs no vitamin D) and (b) routine follow-
up testing for 25(OH)D level to guide vitamin D
dosing.
• This recommendation relates to generally healthy
adults who do not otherwise have established in-
dications for 25(OH)D testing (eg, hypocalcemia).
Summary of Evidence
The evidence summaries, meta-analysis results, and a detailed
summary of the evidence and EtD tables can be found online
at https://guidelines.gradepro.org/profile/5NvU2k7Tigo and
https://guidelines.gradepro.org/profile/PdgmJZLRZTS.
Benefits and Harms
The systematic review identified 2 RCTs (85, 86) (17 074 par-
ticipants in New Zealand and Norway) reporting on the de-
velopment of a respiratory infection, with participants as the
unit of analysis. There was no significant difference between
the vitamin D and placebo groups (RR 1.02 [95% CI,
0.96-1.08]), with an estimated absolute effect size of 5 more
per 1000 (11 fewer to 22 more). In the New Zealand study
(85), the baseline mean 25(OH)D level was 29 ng/mL
(73 nmol/L), and vitamin D was given as 200 000 IU
(5000 µg) monthly for 2 months, followed by 100 000 IU
(2500 μg) monthly for 18 months. In the Norwegian study
(86), the baseline mean 25(OH)D level was not reported
and vitamin D was given as 400 IU (10 µg) of cod liver oil
daily.
The systematic review identified 4 studies (85, 87-89)
(1120 participants, New Zealand, Finland, Canada,
Australia) addressing the total number of respiratory infec-
tions as the unit of analysis; the IRR was 0.95 (95% CI,
0.83-1.07). The baseline mean 25(OH)D levels in these trials
were 24 to 30 ng/mL (60 to 75 nmol/L) (one trial did not re-
port baseline 25[OH]D). The intervention in 2 trials was
daily vitamin D (400 IU [10 μg] and 5000 IU [125 µg]),
whereas nondaily doses were administered in 2 other trials
(10 000 IU [250 μg] per week and 20 000 IU [500 μg] per
week).
The systematic review did not identify any trials examining
the effects of vitamin D on new-onset fatigue. One small RCT
(120 participants, Switzerland) (90) examined improvement
in fatigue among participants with fatigue and baseline
25(OH)D levels lower than 20 ng/mL (50 nmol/L) with a
mean level of 13 ng/mL (33 nmol/L). Participants were
randomized to receive a single dose of 100 000 IU (2500 μg)
of vitamin D or placebo. Four weeks later, those who received
vitamin D were more likely to report amelioration of fatigue
(72% vs 50%; RR 1.49 [95% CI, 1.08-1.94]), suggesting an
improvement in 245 per 1000 (40 fewer to 470 more). The im-
provement in fatigue was modest (change in the Fatigue
Assessment Scale [maximum score= 50] from 24.9 ± 5.4 to
21.6±5.8 in the intervention group vs 23.3 ± 5.4 to 22.5 ±
5.9 in the placebo group).
Studies examining the effects of vitamin D on BMD tested
different dosage regimens. Four studies (91-94) examined
lumbar spine BMD, 2 examined total hip BMD (93, 94), 2 ex-
amined femoral neck BMD (92, 94) and 2 (95, 96) reported on
volumetric tibial bone density by high-resolution peripheral
quantitative computed tomography (HR-pQCT) (Denmark,
Norway, Bangladesh, Austria, USA). Vitamin D was adminis-
tered either daily (400 IU [10 μg], 800 IU [20 μg], 1000 IU
[25 μg], 4000 IU [100 µg], or 7000 IU [175 µg]) or nondaily
(40 000 IU [1000 μg] per week, or 50000 IU [1250 μg]
twice monthly). Estimated mean differences in BMD were
0.003 g/cm² lower at the lumbar spine (0.042 lower to
0.036 higher), 0.049 g/cm² lower at the total hip (0.060 lower
to 0.038 higher), and 0.033 g/cm² higher at the femoral neck
(0.023 lower to 0.090 higher); volumetric bone density by
HR-PQCT was 6.862 mg/cm³ higher at the tibia (8.082 lower
to 21.805 higher). Some trials were felt to be of insufficient
duration (<1 year) to robustly evaluate the effects of vitamin
D on bone density.
The systematic review found no evidence of increased ad-
verse events (symptomatic hypercalcemia, nephrolithiasis,
and kidney disease/kidney failure) in trial participants as-
signed to vitamin D.
Based on the panel's best estimates of treatment effects (the
point estimates derived from meta-analyses), the panel judged
that the anticipated desirable effects of vitamin D are likely to
be small at best, and that the anticipated undesirable effects
are likely to be trivial.
Other Evidence-to-Decision Criteria and
Considerations
Considerations related to required resources, costs, accept-
ability, and feasibility have been previously addressed. A com-
prehensive review of studies addressing female patients' views
of osteoporosis therapy revealed that calcium and vitamin D
were viewed as safe and natural (97). Panel members judged
that empiric vitamin D would likely be acceptable to individ-
uals in this age group, especially females with risk factors for
developing osteoporosis.
Justification for the Recommendation
While vitamin D supplementation appears to be safe, inexpen-
sive, and readily available, the trials identified in the systemat-
ic review did not clearly show a substantive benefit of vitamin
D supplementation. For this reason, the panel issued a condi-
tional recommendation against routine vitamin D supplemen-
tation above what would be required to meet dietary reference
guidelines.
The panel was unable to recommend a 25(OH)D threshold
below which vitamin D administration provides outcome-
specific benefits, primarily due to the absence of large RCTs
designed to assess the effects of the intervention in those
with low baseline 25(OH)D levels. In addition, the financial
costs associated with both 25(OH)D testing and medical
Page 13
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
visits, as well acceptability of testing in this age group, where
routine phlebotomy is not typically indicated for healthy indi-
viduals, factored into the panel's judgment. The panel also
acknowledged that feasibility of 25(OH)D testing is variable
across the globe; and in the absence of evidence for benefit, a
recommendation for 25(OH)D testing could decrease health
equity. For all these reasons, the panel suggested against rou-
tine 25(OH)D testing in generally healthy adults who do not
otherwise have established indications for 25(OH)D testing
(eg, hypocalcemia).
Additional Considerations
The panel judged that healthy adults in this age group
could
rationally choose to take vitamin D supplements if they are
not expected to have adequate vitamin D status via sun expos-
ure and do not reliably meet DRI of vitamin D from vitamin
D-containing or fortified foods.
Testing to identify those with low 25(OH)D level, or to moni-
tor response to therapy, may be required in special populations
who are expected to require more than the DRI of vitamin D to
prevent/reverse low vitamin D status, including those with mal-
absorption (eg, from short gut syndrome, gastric bypass, inflam-
matory bowel disease), those with increased vitamin D
catabolism (eg, due to certain medications), and those with in-
creased renal losses of vitamin D (eg, nephrotic syndrome).
Research Considerations
1. Large clinical trials in populations with low baseline
25(OH)D levels will be required to determine if vitamin
D prevents disease and what dosages are required for the
desired outcomes. Although placebo-controlled trials in
those known to have low 25(OH)D levels may be viewed
as unethical, inclusion of various daily dosages and target-
ing several levels of 25(OH)D would inform the dosages
and target levels required for disease prevention.
2. Clinical trials must be designed to be of sufficient dur-
ation to address the outcomes being examined, consider-
ing the natural history and pathophysiology of the
diseases of interest (eg, acute infectious diseases vs frac-
tures or cancer).
Vitamin D Use in Adults Aged 50 to 74 Years
Background
Vitamin D status may decrease with age due to impaired bio-
synthesis (reduced biosynthesis capacity, lower sun exposure),
low dairy and fish consumption, and increased weight, al-
though the decrease is most marked above age 75 years.
Population-based data from the United States (NHANES) in
3377 adults aged 40 to 59 years and 3602 adults aged 60 years
and older indicate that 24% and 22%, respectively, had
25(OH)D concentrations lower than 20 ng/mL (50 nmol/L),
and 5.9% and 5.7%, respectively, had 25(OH)D concentra-
tions lower than 10 ng/mL (25 nmol/L), with similar values
for women and men (82). Population-based data from
Europe (ODIN) in children and adults (all ages) show a higher
prevalence of low vitamin D status, with 40% having values
lower than 20 ng/mL (50 nmol/L) and 13% having values low-
er than 12 ng/mL (30 nmol/L), with similar values for women
and men (81). The prevalence of low 25(OH)D levels is most
marked in housebound and institutionalized individuals (98).
1919
The period between 50 and 74 years of age corresponds to a
time of bone loss related to menopause and normal aging, de-
creasing muscle function, and increasing fall risk, all predis-
posing to increased risk of fractures. Importantly, some
studies suggest that these risks can be attenuated by vitamin
D and calcium (99). Vitamin D has also been hypothesized
to have a role in modifying the risk of CVD, diabetes, cancer,
acute respiratory infections, and mortality, all of which are
important outcomes relevant to this age group (100-102).
Many of the RCTs designed to address these questions in-
volved groups with mean baseline 25(OH)D levels that would
be considered adequate (approximately 25 ng/mL [63 nmol/
L]). This has contributed to uncertainty regarding whether
empiric vitamin D supplementation in those aged 50 to 74
years can reduce risk of chronic conditions common to this
population. Additionally, it is unclear whether this age group
should undergo screening to identify those with low levels of
25(OH)D who might be more likely to benefit from vitamin
D supplementation. For example, a meta-analysis of RCTs
suggests that vitamin D combined with calcium appears to de-
crease the incidence of fractures in the older and institutional-
ized population (103). However, several recent clinical trials
(104, 105) did not reveal similar findings, perhaps because
many participants in these trials did not have low baseline
25(OH)D levels. This suggests-but does not prove—that
the individuals most likely to benefit from vitamin D supple-
mentation are those at risk for low baseline 25(OH)D levels,
a group that is overrepresented in housebound and institu-
tionalized populations (106, 107). However, 25(OH)D
thresholds required to prevent disease may differ according
to the outcome, as suggested by epidemiological studies
(108). Trials specifically targeting people with low vitamin
D status and/or “treat-to-target” trials documenting the bene-
fit of achieving and maintaining specific 25(OH)D levels with
vitamin D have not been done.
Question 4. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used for
adults aged 50 to 74 years?
Question 5. Should vitamin D supplementation vs no vita-
min D supplementation be used for adults aged 50 to 74
years only when 25(OH)D levels are below a threshold?
Recommendation 4
In the general population aged 50 to 74 years, we
suggest against routine vitamin D supplementation be-
yond the recommended Dietary Reference Intake for
this population. (2000)
Technical remark
• This recommendation relates to empiric vitamin D
supplementation that exceeds the DRIs estab-
lished by the IOM. Adults in this age group should
follow the Recommended Daily Allowance estab-
lished by the IOM (600 IU [15 µg] daily for those
aged 50 to 70 years; 800 IU [20 μg] daily for those
older than 70 years).
Page 14
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1920
Recommendation 5
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
In the general population aged 50 to 74 years, we sug-
gest against routine 25(OH)D testing. (2 | OOO)
Technical remarks
• In this population, 25(OH)D levels that provide
outcome-specific benefits have not been estab-
lished in clinical trials.
• The panel suggests against (a) routine screening
for a 25(OH)D level to guide decision-making (ie,
vitamin D vs no vitamin D) and (b) routine follow-
up testing for 25(OH)D level to guide vitamin D
dosing.
• This recommendation relates to generally healthy
adults who do not otherwise have established in-
dications for 25(OH)D testing (eg, hypocalcemia).
Summary of Evidence
The evidence summaries, meta-analysis results, and a detailed
summary of the evidence and EtD tables can be found online
at https://guidelines.gradepro.org/profile/9FN6GJZdDJ4 and
https://guidelines.gradepro.org/profile/-NxdICB9sYc.
Benefits and Harms
The systematic review identified 13 RCTs (104, 105, 109-
119) (86 311 community-dwelling participants) comparing
vitamin D vs placebo with any fracture as the outcome. The
vitamin D dosages varied between 300 and 3500 IU/daily
equivalent (7.5 and 88 µg) with a median dosage of 1500 IU
(37.5 μg) daily. In many trials, the participants were allowed
to take a daily supplement that contained no more than 400
to 800 IU of vitamin D. The median of the average baseline
25(OH)D concentrations in these studies was 24 ng/mL (ran-
ging from 13 to 32 ng/mL) (60 nmol/L [32 to 80 nmol/L]).
The RR for any fracture with vitamin D was 0.97 (95% CI,
0.91-1.03), with an estimated absolute risk reduction of 2
fewer per 1000 (7 fewer to 2 more). Study subgroup analyses
suggested that the effect of vitamin D on fracture risk was not
modified by risk of bias, sex, dosage of vitamin D, or calcium
co-administration.
All-cause mortality was reported as an outcome in 13 RCTs
(20, 109, 111, 116, 119-127) (81 695 participants). The esti-
mated daily vitamin D dosage varied between 400 IU (10 μg)
and 4800 IU (120 µg), with a median of 2000 IU (50 μg). Most
trials allowed participants to take a supplement with vitamin
D between 400 to 800 IU/d. The median of the average base-
line 25(OH)D concentrations in these studies was 24 ng/mL
(ranging from 18 to 31 ng/mL) (60 nmol/L [45-78 nmol/L]).
The RR for mortality was 1.07 (95% CI, 0.95-1.20), translat-
ing to 2 more per 1000 (2 fewer to 6 more). The risk of mor-
tality in studies involving calcium co-administration (RR 0.90
[95% CI, 0.79-1.01]) appeared to be lower than those involv-
ing vitamin D alone (RR 1.12 [95% CI, 1.01-1.24]) (P for het-
erogeneity = .021). In addition, the risk of mortality appeared
to be higher with vitamin D in the studies involving high dos-
ages of vitamin D (RR 1.22 [95% CI, 1.06-1.39]) relative to
those involving standard dosages (RR 0.95 [95% CI,
0.86-1.04]) (P for heterogeneity = .003). Study subgroup ana-
lyses suggested that the effect of vitamin D on mortality risk
was not modified by risk of bias or sex.
Cancer was reported as an outcome in 15 RCTs (20, 109,
111, 119, 123, 125, 126, 128-135) (91 223 participants), us-
ing dosages of 300 to 4800 IU/daily equivalent, with a median
dosage of 2000 IU/d (50 µg/d). In many trials, the participants
were allowed to take a daily supplement that contained no
more than 400 to 800 IU of vitamin D. Mean baseline
25(OH)D ranged from 13 to 33 ng/mL (median 26 ng/mL)
(33 to 83 nmol/L [median 65 nmol/L]). The relative risk for
cancer with vitamin D was 1.00 (95% CI, 0.97-1.03) translat-
ing to 0 fewer patients with cancer per 1000 (4 fewer to 4
more). Study subgroup analyses suggested that the effect of
vitamin D on cancer outcomes was not modified by risk of
bias, sex, dosage of vitamin D or calcium co-administration.
Fourteen RCTs (20, 109, 111, 116, 118, 119, 122, 125-
127, 131, 134, 136, 137) involving 80 547 participants re-
ported on CVD events using dosages of 300 to 4800 IU/daily
equivalent, with a median dosage of 2000 IU/d (50 µg/d). In
addition, most trials allowed a vitamin D-containing supple-
ment from 400 to 800 IU/d. Mean baseline 25(OH)D ranged
from 13 to 31 ng/mL (mean 24 ng/mL) (33 to 78 nmol/L;
mean 60 nmol/L]). The relative risk for CVD with vitamin D
was 1.00 (95% CI, 0.93-1.08), translating to 0 fewer patients
with CVD per 1000 (2 fewer to 3 more). Seven RCTs (20, 109,
111, 119, 122, 125, 136) reported stroke with a summary RR
of 0.95 (95% CI, 0.83-1.09), translating to 1 fewer patient
with stroke per 1000 (2 fewer to 1 more). Myocardial infarc-
tion (MI) was an outcome in 7 RCTs (20, 109, 111, 119, 122,
125, 131), with a summary RR of 1.00 (95% CI, 0.83-1.20),
translating to 0 fewer patients with MI per 1000 (2 fewer to 2
more). Study subgroup analyses suggested that the effects of
vitamin D on cardiovascular events, stroke, and MI were
not modified by risk of bias, sex, dosage of vitamin D or cal-
cium co-administration.
Kidney stones were reported in 10 RCTs (20, 109-111,
118, 125, 129, 135, 138, 139) with a summary RR of 1.10
(95% CI, 1.00-1.19), translating to 2 more patients with kid-
ney stones per 1000 (0 fewer to 4 more). Kidney disease was
reported in 4 RCTs (20, 119, 127, 134) with a summary RR
of 1.04 (95% CI, 0.76-1.42), translating to 0 fewer patients
with kidney disease per 1000 (1 fewer to 2 more). Study
subgroup analyses suggested that the effects of vitamin D
on kidney stones and kidney disease were not modified by
risk of bias, sex, dosage of vitamin D, or calcium
co-administration.
The systematic review identified 3 RCTs that reported out-
comes specifically in participants with baseline serum 25(OH)
D below 20 ng/mL (50 nmol/L) (or the lowest quartile, ie,
<24 ng/mL [60 nmol/L]) receiving vitamin D vs placebo
(29). Meta-analysis of such data from 2 of these RCTs (20,
124) suggested an RR of 1.11 (95% CI, 0.85-1.46) for the
mortality outcome. Cancer was reported in 2 RCTs (20,
130), and vitamin D was associated with a RR of 0.91
(95% CI, 0.70-1.19) compared to placebo. Cardiovascular
disease events were reported in 3 RCTs (20, 122, 137) with
a RR of 1.02 (95% CI, 0.87-1.19) compared to placebo.
Subgroup analyses in single trials suggested no clear impact
on fractures (RR 1.01 [95% CI, 0.81-1.24]), stroke (RR
1.04 [95% CI, 0.39-2.75]), MI (RR 0.93 [95% CI,
0.38-2.29]), and adverse events (RR 1.26 [95% CI,
0.77-2.12]).
Page 15
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
Based on the panel's best estimates of treatment effects, the
panel judged that the anticipated desirable effects of vitamin
D, in addition to the anticipated undesirable effects, are likely
to be trivial.
Other Evidence-to-Decision Criteria and
Considerations
Considerations related to required resources (costs), accept-
ability, and feasibility of vitamin D have already been ad-
dressed. Prevention of hip fractures in older people at risk is
highly valued, as demonstrated by time-trade-off studies
(140). The effect of coronary artery disease on quality of life
may be small except for recurrent angina (141).
A cost-benefit analysis concluded that the costs of vitamin
D and calcium would be much lower than the costs of frac-
tures resulting from no supplementation. This result was
mainly driven by the age group older than 65 years (142).
Although a French study concluded that treatment based
on 25(OH)D concentrations was more cost-effective than
treating everybody (143), a systematic review of economic
evaluations concluded that there was insufficient economic
evidence to draw conclusions about the cost-effectiveness
of population strategies (144). The panel found these cost-
effectiveness studies difficult to contextualize given that the
commissioned systematic review of clinical trials did not dis-
close a substantive benefit of vitamin D on fractures in those
aged 50 to 74 years.
A comprehensive review of studies addressing women's
views of osteoporosis therapy revealed that vitamin D and
calcium were viewed as safe and natural and preferred to hor-
mones and other treatments (97). As such, vitamin D is likely to
be considered acceptable. The panel judged that empiric vitamin
D supplementation is feasible to implement, although condition-
ing vitamin D supplementation on 25(OH)D levels could re-
present an important barrier for some.
Justification for the Recommendations
Vitamin D supplementation appears to be safe when taken as
outlined in the IOM DRIS. Vitamin D is also inexpensive, readily
available, acceptable to patients, and relatively easy to imple-
ment. Adherence may be a challenge, because supplementation
typically involves lifelong use of vitamin D. Based on the meta-
analyses of the available trials, which yielded high certainty of
evidence for fractures, CVD events, cancer and mortality, the
panel judged that vitamin D supplementation appears to have lit-
tle or no beneficial impact on the outcomes analyzed in healthy
populations aged 50 to 74 years. There was therefore no compel-
ling rationale to recommend empiric vitamin D in this age group,
especially since supplementation would involve costs (admittedly
minor) and inconvenience.
Importantly, most of the recent trials were completed in
populations that were meeting their DRI and did not have
low vitamin D status at baseline. Given the well-established
harmful consequences of very low vitamin D status on skeletal
health and calcium homeostasis, the panel judges that some
subgroups in this age group could rationally choose to take
vitamin D supplementation, especially if they are not expected
to have adequate vitamin D status via sun exposure (dark
complexion, housebound, clothing style) or reliable
IOM-recommended intake via diet, supplements or ingestion
of vitamin D-fortified foods.
1921
Subgroup analyses did not provide evidence for benefit
with vitamin D in subgroups with 25(OH)D below 20 to
24 ng/mL (50-60 nmol/L). In addition, there are monetary
costs associated with both 25(OH)D testing and medical vis-
its, the panel judged that a recommendation for 25(OH)D
testing could decrease feasibility and health equity (especially
when compared to empiric vitamin D supplementation). For
all these reasons, the panel suggested against routine 25(OH)
D testing (eg, screening) in generally healthy adults aged 50
to 74 years.
Additional Considerations
These recommendations should not be extrapolated to
individuals with conditions known to substantially impact
vitamin D physiology, including malabsorption (eg, from
gastric bypass), increased vitamin D catabolism, renal
loss of vitamin D metabolites, and decreased vitamin D
activation.
With regard to 25(OH)D screening, the panel noted that 2
risk scoring systems can predict serum 25(OH)D concentra-
tions lower than 20 ng/mL and 12 ng/mL (<50 and
< 30 nmol/L), respectively, with reasonable accuracy, and
thus may be useful in clinical practice to identify persons
aged 55 to 85 years at high risk for low vitamin D without
the need for 25(OH)D testing (145). Risk factors in these scor-
ing systems include female sex, alcohol use, smoking, season,
medication use, no vitamin use, and limited outdoor activities
such as gardening and bicycling.
Research Considerations
1. The age group 65 to 74 years requires more attention,
since the risks of chronic diseases and the outcomes being
examined are higher than in those aged 50 to 64 years.
The age group of 50 to 74 years is a heterogenous popu-
lation in which some may be in excellent health, whereas
others may have chronic conditions and may be house-
bound. Thus, trials addressing the effect of vitamin D
on individuals with different health status are required.
2. RCTs specifically in those with low baseline 25(OH)D
levels are required to clarify the risks and benefits of vita-
min D and/or calcium supplementation.
3. Studies with longer follow-up may be needed, as some out-
comes may become apparent only after 5 years (117).
4. In secondary, exploratory analyses, vitamin D in this
age group has been implicated in the prevention of
autoimmune disease such as rheumatoid arthritis,
polymyalgia rheumatica, and autoimmune thyroid
disease (146). These data need confirmation by add-
itional RCTs.
5. Studies of the effect of vitamin D fortification on vitamin
D status in different populations at risk of low vitamin D
status are needed.
Vitamin D Use in Adults Aged ≥75 Years
Background
Low 25(OH)D levels are common among older people in the
United States. Recent results from NHANES surveys during
2001-2018 showed that the prevalence of low vitamin D
status (25[OH]D≤20 ng/dL [50 nmol/L]) in the US popula-
tion older than 80 years was 19.6% in females and 18.9%
Page 16
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1922
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
in males (147). Many observational studies have reported
inverse associations between 25(OH)D levels and adverse
health outcomes such as falls, fractures, and respiratory
disease (148-152). These conditions contribute significantly
to morbidity and mortality in older people. For example,
falls occur commonly in older people, with more than 14
million US adults 65 years and older falling one or more
times each year (153), resulting in an estimated 9 million
fall injuries annually (154). Falls are the leading cause of
injury-related death in this age group, which is an increasing
subset of the population (155). The annual health care costs
from fall injuries are about $50 billion (156). More than
95% of hip fractures are caused by falling (157), with
more than 300 000 people 65 years and older hospitalized
for a hip fracture each year in the United States (158-160).
Hip fractures are also associated with increased mortality
(161). Despite the importance of these conditions associated
with low vitamin D status in observational studies, it remains
unclear whether vitamin D supplementation lowers the risks
of such conditions: the data from randomized, placebo-
controlled trials of vitamin D supplementation are inconsist-
ent, and systematic reviews and meta-analyses of RCTs
have reported heterogeneous results for these outcomes
(162-165).
Question 6. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used by adults
aged ≥75 years?
Question 7. Should vitamin D supplementation vs no
vitamin D supplementation be used by adults aged
≥75 years only when 25(OH)D levels are below a
threshold?
Recommendation 6
In the general population aged 75 years and older, we
suggest empiric vitamin D supplementation because
of the potential to lower the risk of mortality. (2 |
0000)
Technical remarks
.
Empiric vitamin D may include daily intake of forti-
fied foods, vitamin formulations that contain vita-
min D and/or daily intake of a vitamin D
supplement.
• For empiric supplementation, daily, lower-dose
vitamin D is preferred over nondaily, higher doses.
• In the clinical trials included in the systematic re-
view that reported on the mortality outcome, vita-
min D dosage ranged from 400 to 3333 IU [10 to
83 μg] daily equivalent. The estimated weighted
average was approximately 900 IU (23 g) daily.
Participants in many trials were allowed to remain
on their routine supplements, including up to 800
IU (20 µg) of vitamin D daily.
Recommendation 7
In the general population aged 75 years and older,
we suggest against routine testing for 25(OH)D levels.
(2 | OOO)
Technical remarks
• In this population, 25(OH)D thresholds that pro-
vide outcome-specific benefits have not been es-
tablished in clinical trials.
• The panel suggests against (a) routine screening
for a 25(OH)D level to guide decision-making (ie,
vitamin D vs no vitamin D) and (b) routine follow-
up testing for 25(OH)D level to guide vitamin D
dosing.
• This recommendation relates to generally healthy
adults who do not otherwise have established in-
dications for 25(OH)D testing (eg, hypocalcemia).
Summary of Evidence
The evidence summaries, meta-analysis results, and a de-
tailed summary of the evidence and EtD tables can be found
online at https://guidelines.gradepro.org/profile/3knvwnb
vIkQ and https://guidelines.gradepro.org/profile/ySx1d8
ko_C4.
Benefits and Harms
The systematic review included 25 trials (20, 104, 121, 124,
166-186) (49 879 participants) that reported on the effect of
vitamin D on all-cause mortality. These trials involved partic-
ipants from community settings (n=17), nursing homes
(n = 6), and hospital clinics (n=2). Most trials assessed the
impact of vitamin D3 (cholecalciferol), commonly given as a
daily dose (13 trials), either alone or combined with calcium.
Follow-up durations ranged from 12 weeks to 7 years, with a
median of 2 years. Meta-analysis suggested that vitamin D
lowers mortality compared to placebo (RR 0.96 [95% CI,
0.93-1.00]), with an estimated absolute effect size of 6 fewer
deaths per 1000 people (from 11 fewer to 0 more). Study sub-
group analyses revealed no differences according to risk of
bias, gender, calcium co-administration, vitamin D dosage
(high vs standard), or setting (community, hospitalized, insti-
tutionalized). When restricting analysis to community-based
studies, vitamin D appeared to be associated with a similar re-
duction in mortality risk (RR 0.95 [95% CI, 0.90-0.99]).
Among study participants with low vitamin D status
(<20 ng/mL [50 nmol/L]), the results were consistent with
those observed in the broader population (RR of mortality
0.88 [95% CI, 0.46-1.67]).
The systematic review identified 14 trials (104, 117, 170,
171, 173, 177, 178, 180, 181, 183, 184, 187-190) that re-
ported the number of participants with a fracture as the unit
of measure (43 585 participants), and the RR for vitamin D
was 1.01 (95% CI, 0.94-1.08), with an estimated absolute ef-
fect size of 1 fewer per 1000 people (from 5 fewer to 6 more).
Fourteen trials (168, 172, 175, 191) [male and female,
Page 17
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
separately] (174, 180, 184, 185, 188, 189, 192-195) reported
the total number of fractures as the unit of measure, and the
IRR was 0.95 (95% CI, 0.82-1.10). Study subgroup analysis
suggested that estimated IRR may vary according to study
risk of bias, with IRR estimates appearing to be lower in stud-
ies with some concerns compared to those with either low or
high risk of bias. The IRR for number of fractures was lower
in studies involving calcium co-administration (0.78 [95% CI,
0.68-0.90]) vs no calcium co-administration (1.05 [95% CI,
0.88-1.28]) (P for heterogeneity .005), but a similar inter-
action was not observed when participants with fractures
served as the unit of analysis. Study subgroup analyses did
not implicate sex, vitamin D dosage, or setting (community
vs institutional) as significant predictors of fracture outcomes.
Data addressing fracture outcomes specifically in those with
25(OH)D levels <20 ng/mL (50 nmol/L) were unavailable.
The systematic review identified 16 trials (104, 166, 170,
171, 173, 174, 176, 184, 188, 189, 193, 194, 196-199) that
reported the number of participants with any fall as the unit
of measure (12 342 participants) and the RR for vitamin D
was 0.97 (95% CI, 0.91-1.03), with an absolute effects size
of 16 fewer people with falls per 1000 (from 48 fewer to 16
more). Fifteen trials (166, 173, 175, 184, 185, 187-190,
194, 195, 197-200) reported the number of falls as the unit
of measure, and the IRR was 0.91 (95% CI, 0.81-0.99). The
reduction in IRR for falls was confined mainly to studies
with high risk of bias, and no effect was seen in studies with
low risk of bias (IRR 1.03 [95% CI, 0.92, 1.11]). Study sub-
group analyses suggested that vitamin D reduced fall risk
more so in studies involving standard vitamin D dosages
(RR 0.93 [95% CI, 0.85-1.01]; IRR 0.88 [95% CI,
0.76-1.00]) compared to studies involving high vitamin D
dosages (RR 1.06 [95% CI, 1.01-1.11]; IRR 1.02 [95% CI,
0.86-1.10]) (P for interaction = .007 for RR and 0.033 for
IRR). The risk for falls appeared to be reduced by vitamin D
to a greater degree in studies involving calcium co-
administration (RR 0.85 [95% CI, 0.74-0.97]; IRR 0.73
[95% CI, 0.53-0.92]) vs studies without calcium co-
administration (RR 1.04 [95% CI, 1.01-1.08]; IRR 0.99
[95% CI, 0.91-1.07]) (P for interaction = .004 for RR and
0.007 for IRR). In addition, study subgroup analysis sug-
gested that vitamin D reduced total number of falls more so
in institutional-based studies (IRR 0.82 [95% CI,
0.69-0.94]) compared to community-based studies (IRR
0.96 [95% CI, 0.83-1.05]) (P for interaction = .024), but a
similar interaction was not observed when persons with falls
served as the unit of analysis. Analysis of 2 studies reporting
falls among participants with low vitamin D status (< 20 ng/
mL [50 nmol/L]) (194, 199), the RR for fall with vitamin D
was 0.65 (95% CI, 0.40-1.05).
The systematic review identified only 2 trials (168,201) that
reported on the effect of vitamin D on respiratory infections in
adults older than age 75 years. Both trials reported subgroup
analyses for both upper and lower respiratory tract infections
combined. The ViDA study compared monthly vitamin D3
with placebo, with number of participants experiencing re-
spiratory tract infection as the unit of measure, and the ad-
justed hazard ratio (HR) was 1.11 (95% CI, 0.94-1.30)
(201). In the DO-HEALTH trial, which evaluated the total
number of infections as the unit of measure, the adjusted
IRR was 1.15 (95% CI, 0.94-1.41) for daily 2000 IU
(50 µg) vitamin D3 (168). No trials reported subgroup ana-
lyses related to the impact of vitamin D on respiratory
1923
infections specifically for those with low 25(OH)D levels in
this age group.
Four trials reported possible undesirable outcomes in adults
aged 75 years and older (29). With the number of participants
as the unit of measure, the RR for nephrolithiasis among 6306
participants in 3 trials (20, 138, 168) was 0.94 (95% CI,
0.54-1.65) for vitamin D vs placebo, with an estimated abso-
lute effect size of 1 fewer per 1000 [7 fewer to 10 more]), and
the RR for kidney disease among 5634 participants in 3 trials
(20, 166, 168) was 0.76 (95% CI, 0.44-1.32) with an esti-
mated absolute effect size of 3 fewer per 1000 [6 fewer to 3
more]).
Based on the panel's best estimates of treatment effects (ie,
stipulating the veracity of point estimates), the panel judged
that the anticipated desirable effects of vitamin D are likely
small, and that the anticipated undesirable effects are likely
trivial. Among study participants with low vitamin D status,
the results were consistent with those observed in the broader
population.
Other Evidence-to-Decision Criteria and
Considerations
sup-
The panel concluded that the costs of empiric vitamin D
plementation were negligible because vitamin D is inexpen-
sive. Although the panel identified some cost-effectiveness
analyses related to falls and fractures, these were difficult to
apply because the systematic review suggested little to no
benefit for the fall and fracture outcomes. Regardless, given
minimal costs of vitamin D supplementation, the panel rea-
soned that vitamin D is likely to be cost-effective with regard
to its (likely) mortality benefit. Given that low vitamin D sta-
tus tends to be more prevalent among those with lower health
equity, assuming that vitamin D supplementation is most like-
ly to benefit those with low vitamin D status, and recognizing
that vitamin D supplementation is inexpensive, the panel rea-
soned that vitamin D probably improves equity, based on its
(likely) mortality benefit. The panel judged that empiric vita-
min D supplementation would be feasible and acceptable to
stakeholders.
The systematic review did not find evidence suggesting that
benefit with vitamin D is restricted to those with baseline
25(OH)D levels below a threshold. In addition, the panel con-
cluded that conditioning vitamin D supplementation/treat-
ment on 25(OH)D screening may create barriers for some
(eg, in places where access to laboratory testing is difficult).
Moreover, the addition of a 25(OH)D testing requirement
would increase costs, possibly decreasing acceptability for
some.
Justification for the Recommendations
Based on the systematic review, vitamin D probably results in
a slight decrease in all-cause mortality in this age group (high
certainty of evidence), and probably results in little to no dif-
ference in fractures (high certainty of evidence), or adverse
events (moderate certainty of evidence), including falls. The
panel had concerns that clinical trials using high dosages of
vitamin D may have masked improvement in fall risk, and
study subgroup analysis suggested that fall risk was likely re-
duced in trials employing standard vitamin D dosages.
While specific data related to respiratory infections were in-
adequate (low certainty of evidence), indirect data from gen-
eral populations suggest that vitamin D is unlikely to be
Page 18
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1924
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
harmful in this regard, and the panel prioritized the mortality
outcome. Given that the best available evidence suggests a
small but important benefit in terms of mortality risk and min-
imal to no harms, the panel judged that the balance between
desirable and undesirable effects probably favors empiric vita-
min D supplementation. In addition, the panel judged that em-
piric vitamin D supplementation is typically inexpensive, may
be cost-effective, may increase health equity, and is probably
both acceptable to key stakeholders and feasible to imple-
ment. For these reasons, the panel suggests empiric vitamin
D supplementation. In the absence of high overall certainty
of evidence, the panel issued a conditional recommendation
in this regard.
The systematic review did not find evidence suggesting that
net benefit is restricted to those with 25(OH)D below a thresh-
old, and the few available clinical trials that reported subgroup
results by 25(OH)D level did not clearly implicate baseline
25(OH)D level as a significant predictor of treatment effect;
however, data were judged to be sparse in this regard. In add-
ition, 25(OH)D testing and medical visits involve monetary
costs, and the panel judged that a recommendation for
25(OH)D testing could decrease feasibility and health equity (es-
pecially when compared to empiric vitamin D supplementation).
For these reasons, the panel suggests against routine 25(OH)D
testing (eg, screening) in adults aged 75 and older.
Additional Considerations
years
When considering all 25 clinical trials reporting mortality
data, the median (interquartile range) vitamin D dosage ap-
proximated 833 (800-1370) IU/day (21 µg/day [20-34 μg/
day]), and the estimated weighted average vitamin D dosage
(ie, each study's vitamin D dosage weighted according to the
study's weight in the meta-analysis for the mortality outcome)
was approximately 909 IU/day (23 µg/day). In many trials,
participants were permitted to remain on vitamin D supple-
ments up to 800 IU (20 µg)/day.
Vitamin D with calcium may be superior to vitamin D alone
at decreasing the risk of falls and fractures. Subgroup analysis
revealed that vitamin D significantly lowers fracture risk with
calcium co-administration when number of fractures was the
outcome; however, when the number of participants with
fracture was the unit of measure, the interaction was not stat-
istically significant. The median dosage of calcium used in the
included trials was 1000 mg per day (500-1500 mg/day).
Calcium supplementation does not appear to increase the
risk of CVD overall (202) nor mortality risk in the current
meta-analysis (29).
Research Considerations
1. Based on the known effects of vitamin D on the musculo-
may
skeletal system, it be unethical to keep a group of
people with low 25(OH)D levels on placebo for long pe-
riods to evaluate the effectiveness of vitamin D supple-
mentation on falls or fractures, both long-term
outcomes. However, studies using several different daily
dosages of vitamin D and targeting several achieved
25(OH)D levels are feasible and would define the
achieved levels that prevent adverse outcomes.
2. The great variability of protocols used in clinical trials may
have interfered in the evaluation of supplementation on
musculoskeletal health in this group of older individuals.
Future studies will require specific protocols, avoiding
bolus doses, and selecting individuals at risk for fractures
and falls to evaluate the effect of the intervention.
Vitamin D Supplementation During Pregnancy
Background
Nutritional status during pregnancy plays a critical role in
perinatal health, fetal growth, and infant development. The
fetus is dependent on maternal circulating 25(OH)D for pla-
cental metabolism and transfer of vitamin D metabolites
(203, 204). In pregnancy, very low vitamin D status
(25[OH]D < 10-12 ng/mL [<25-30 nmol/L]) is associated
with increased risk of neonatal hypocalcemic seizures, cardio-
myopathy, and neonatal rickets, with life-limiting and poten-
tially fatal outcomes (18, 205). Very low vitamin D status
during pregnancy is prevalent in both low- and high-income
settings (206, 207).
Many studies, for example (208), have described associa-
tions between 25[OH]D levels <20 ng/mL (<50 nmol/L)
and increased risk of hypertensive disorders of pregnancy
(gestational hypertension, preeclampsia, eclampsia, and
HELLP syndrome [Hemolysis, Elevated Liver enzymes and
Low Platelets]). Hypertensive disorders of pregnancy increase
risks for fetal growth restriction, small-for-gestational-age
(SGA) infants, and induced preterm delivery, with potentially
serious and lifelong consequences for infant bone and brain
development, as well as maternal and offspring long-term car-
diometabolic health (209). Economic costs of preeclampsia
have been estimated at twice those of healthy pregnancies
for maternal postnatal care (210). Hao et al (211) estimated
a 3-fold higher cost for pregnancies complicated by hyperten-
sive disorders relative to uncomplicated care when both ma-
ternal and infant costs were included.
Whether nutritional requirements for vitamin D change
during pregnancy is not known, and evidence for the role of
vitamin D in improving perinatal outcomes is conflicting
(212). Accordingly, preconception or pregnancy-specific rec-
ommendations for vitamin D are not universal, nor is there
a consensus on the dosage of vitamin D or 25(OH)D level
required to support a healthy pregnancy. While harmonized
global estimates do not yet exist, reported prevalence rates
for low and very low vitamin D status (25[OH]D<20 and
< 12 ng/mL [<50 and < 30 nmol/L], respectively) are high
among women of reproductive age and during pregnancy,
particularly among individuals with decreased skin synthesis
due to low exposure to UV-B light, low vitamin D intakes,
low nutrient-dense diets, and dark complexion (34, 213-
216). This, along with the fetal dependence on maternal vita-
min D and the inverse associations of low vitamin D status
with undesirable outcomes in the perinatal period, make it im-
portant to evaluate the role of vitamin D supplementation
during pregnancy. Additional high-priority clinical questions
relate to the potential utility of 25(OH)D testing during preg-
nancy and optimal maternal 25(OH)D concentrations during
pregnancy.
Question 8. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used during
pregnancy?
Question 9. Should vitamin D supplementation vs no vita-
min D supplementation be used during pregnancy only
when 25(OH)D levels are below a threshold?
Page 19
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
Recommendation 8
We suggest empiric vitamin D supplementation dur-
ing pregnancy, given its potential to lower risk of pre-
eclampsia, intra-uterine mortality, preterm birth, SGA
birth, and neonatal mortality. (2 | OO)
Technical remarks
• This recommendation is based on evidence from
trials conducted in healthy individuals during
pregnancy.
• Empiric vitamin D may include daily intake of forti-
fied foods, prenatal vitamin formulations that con-
tain vitamin D, and/or a vitamin D supplement
(pills or drops).
• In the clinical trials included in the systematic re-
view, the vitamin D dosages ranged from 600 to
5000 IU (15 to 125 μg) daily equivalent, usually pro-
vided daily or weekly. The estimated weighted aver-
age was approximately 2500 IU (63 µg) per day.
Recommendation 9
During pregnancy, we suggest against routine 25(OH)
D testing. (2000)
Technical remarks
• In this population, 25(OH)D levels that provide
pregnancy outcome-specific benefits have not
been established in clinical trials.
• The panel suggests against (a) routine screening for
a 25(OH)D level to guide decision-making (ie, vita-
min D vs no vitamin D) and (b) routine follow-up
testing for 25(OH)D level to guide vitamin D dosing.
• This recommendation relates to generally healthy
pregnant individuals who do not otherwise have
established indications for 25(OH)D testing (eg,
hypocalcemia).
Summary of Evidence
The evidence summaries, meta-analysis results, and a detailed
summary of the evidence and EtD tables can be found online
at https://guidelines.gradepro.org/profile/kZ8sir4uV7M and
https://guidelines.gradepro.org/profile/QSOmqUUCVGE.
Benefits and Harms
The systematic review identified 10 RCTs that met the inclu-
sion criteria (29). Due to the panel's a priori decision to include
only trials involving placebo-treated controls (rather than al-
lowing the control group to remain on routine supplements
or receive low-dose vitamin D), many RCTs were excluded, in-
cluding many from the United States, where 400 IU (10 µg)
was often given to the control group. Three included studies
were conducted in Europe; 2 in Bangladesh; 2 in India; 2 in
Iran and 1 in Pakistan. Of the 2979 participants, almost half
1925
(n=1298) came from the trial by Roth et al (217) in
Bangladesh. The included trials varied greatly in terms of
dose frequency (one-time vs daily vs intermittent dosing) and
dose ranges (600 to 200 000 IU [15 to 5000 μg]). The median
gestational age at which the intervention (vitamin D vs pla-
cebo) was initiated was about 20 weeks. Of the 7 trials that re-
ported baseline 25(OH)D concentrations, mean values were
below 12 ng/mL (30 nmol/L) in 4 (217-220).
When combined, data from 8 studies (217, 219, 221-226)
(2674 participants) suggest that vitamin D may reduce the
risk of preeclampsia (RR 0.73; 95% CI, 0.46-1.15]) with an
estimated absolute effect size of 23 fewer per 1000 (46 fewer
to 13 more).
Data from 4 trials (217-219, 223) (1738 participants) sug-
gest that vitamin D may reduce the risk of intra-uterine mor-
tality slightly (RR 0.70 [95% CI, 0.34-1.46]) with an
estimated absolute effect size of 6 fewer per 1000 (13 fewer
to 9 more). Similarly, data from 3 trials (217, 218, 223)
(1576 participants) indicate that vitamin D may reduce the
risk of neonatal mortality slightly (RR 0.57 [95% CI,
0.22-1.49]), with an estimated absolute effect size of 8 fewer
per 1000 (14 fewer to 9 more).
Data from 6 trials (217, 219, 222-225) (2085 participants)
suggest that vitamin D may reduce the risk of preterm birth
(RR 0.73 [95% CI, 0.39-1.36]) with an estimated absolute ef-
fect size of 28 fewer per 1000 (62 fewer to 37 more). Data
from 5 trials (217, 219, 220, 224, 225) (2355 participants)
suggest that vitamin D may reduce the risk of SGA birth
(RR 0.78 [95% CI, 0.50-1.20]) with an estimated absolute ef-
fect size of 41 fewer per 1000 (94 fewer to 38 more). SGA sta-
tus was variably defined in the different trials.
Adverse events of interest (nephrolithiasis, symptomatic hy-
percalcemia, kidney disease) were rare (one case of protein-
uria related to nephrotic syndrome in the vitamin D arm),
but most trials did not prespecify adverse events except for
the trials by Roth et al (217, 223), which reported no cases
of symptomatic hypercalcemia.
Study subgroup analyses did not implicate either risk of bias
or vitamin D dosage as a significant predictor of study out-
comes. Data were insufficient to address whether baseline
25(OH)D level was a significant predictor of treatment effects.
Based on the panel's best estimates of treatment effects (ie,
stipulating the veracity of point estimates), the panel judged
that the anticipated desirable effects of vitamin D during preg-
nancy for the outcomes specified are likely to be moderate.
Although the panel recognized that the 95% CIs included
the possibility for harm for each outcome, the panel noted
that all point estimates favored benefit and judged that the an-
ticipated undesirable effects are likely to be trivial.
The panel also considered a 2019 systematic review per-
formed by Palacios et al (227). According to this meta-
analysis, vitamin D supplementation during pregnancy re-
duced risks of preeclampsia (RR 0.48 [95% CI, 0.30-0.79]),
low birthweight (RR 0.55 [95% CI, 0.35-0.87]), and gesta-
tional diabetes (RR 0.51 [95% CI, 0.27-0.97]), with a non-
significant reduction in preterm birth (RR 0.66 [95% CI,
0.34-1.30]).
Other Evidence-to-Decision Criteria and
Considerations
Although the panel identified no direct evidence, the panel
judged that vitamin D supplementation would be acceptable
Page 20
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1926
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
and feasible to implement during pregnancy,
when health care
supervision is frequently available. The panel judged that pre-
venting low vitamin D status during pregnancy, particularly
among individuals most at risk for low vitamin D status
(206, 213), may improve health equity. Floreskul (71) re-
ported that free-of-charge provision of vitamin D supplements
to pregnant individuals and children younger than age 4 years
for rickets prevention in the United Kingdom would be clinic-
ally effective and cost-saving in participants with "dark and
medium skin tone," especially in regions with high incidence
of rickets.
Justification for the Recommendations
The systematic review suggested anticipated benefit with em-
piric vitamin D for all selected outcomes: preeclampsia
(2.3% anticipated absolute reduction with low certainty of
evidence), intra-uterine mortality (0.6% anticipated absolute
reduction with moderate certainty of evidence), preterm
birth (2.8% anticipated absolute reduction with low cer-
tainty of evidence), SGA birth (4.1% anticipated absolute re-
duction with low certainty of evidence) and neonatal
mortality (0.8% anticipated absolute reduction with moder-
ate certainty of evidence). The meta-analysis by Palacios et al
(227) showed benefits in the same direction (lower risk of
preeclampsia, low birthweight, gestational diabetes, and pre-
term birth). When taken together, and if stipulating the ver-
acity of these point estimates, the panel judged that these
desirable anticipated effects were moderately substantial.
However, for all the described outcomes, the 95% CIs in-
cluded the potential for harm, and available evidence for ma-
ternal mortality and maternal adverse events was not very
robust. Nonetheless, given that the best available evidence
(point estimates) suggested moderate benefit and minimal
harm, the panel judged that the balance between desirable
and undesirable effects probably favors empiric vitamin D
supplementation. In addition, the panel judged that empiric
vitamin D is typically inexpensive, may be cost-effective,
may increase health equity, and is probably acceptable to
key stakeholders and feasible to implement. Thus, the panel
suggests empiric vitamin D supplementation during preg-
nancy. Given the low overall certainty of evidence, the panel
issued a conditional recommendation.
Available evidence did not permit a well-supported judg-
ment about the net benefit of 25(OH)D testing during
pregnancy followed by vitamin D supplementation only
in those with low 25(OH)D levels. In addition, compared
to empiric vitamin D supplementation, adding the need
for 25(OH)D testing would add costs, and the panel judged
that testing could also decrease feasibility and health
equity. For all these reasons, the panel suggests that vita-
min D supplementation should generally proceed without
testing for baseline 25(OH)D levels and without the need
for subsequent monitoring of 25(OH)D levels to assess re-
sponse to supplementation, provided that vitamin D dos-
ages are within the tolerable upper intake level as
established by the IOM.
Additional Comments
This guideline is different from the World Health Organization
(WHO) guideline on vitamin D supplementation in pregnancy,
which was published in 2016 (228) and updated in 2020 (229).
Largely based on the systematic reviews by De-Regil (230),
which found a possible beneficial effect of vitamin D on redu-
cing preeclampsia, low birthweight, and preterm birth but a
potential adverse effect of calcium plus vitamin D supplemen-
tation on preterm birth, the guideline group did not recom-
mend vitamin D for pregnancy to improve maternal and
infant health outcomes (228). The updated WHO 2020 guide-
line (229), which also did not recommend vitamin D, was
largely based on the systematic review by Palacios et al
(227), which reported outcomes similar to those of the present
guideline for pre-eclampsia, preterm birth, low birth weight,
and adverse effects. There were some differences in the studies
selected for data synthesis, as Palacios et al (227), included a
larger number of studies, including trials that administered,
or allowed control participants to take, a low dosage of vita-
min D and trials that co-administered vitamin D and calcium.
The current guideline had access to more recent RCTs, includ-
ing Roth et al (217). Overall, the current panel found very little
evidence for harm with vitamin D supplementation, along with
some evidence for benefit.
The optimal dosage of vitamin D for the prevention of ma-
ternal and fetal complications remains unclear. In the studies
included in the commissioned systematic review, the estimated
median vitamin D dosage for preeclampsia evaluation was
3161 IU (79 μg) daily, and the estimated weighted average dos-
age was 2639 IU (66 µg) per day. The estimated median vita-
min D dosages in the studies assessing intra-uterine and
neonatal mortality were 3375 IU (84 µg) and 2750 IU
(69 µg) daily, respectively, and corresponding estimated
weighted average dosages were 2908 IU (73 µg) and 3052 IU
(76 µg) per day. For preterm birth and SGA birth studies, the
estimated median dosages were 3375 IU (84 µg) and 2750 IU
(69 µg) daily, respectively, while estimated weighted average
dosages were 2735 IU (68 µg) and 2642 IU (66 µg) per day.
Research Considerations
Proposed areas for research include:
1. Adequately powered clinical trials with prespecified
outcomes to address whether and to what degree vitamin
D impacts patient-important perinatal outcomes, in both
healthy individuals and those with high-risk pregnancies.
Particular attention should be paid to individuals at high
risks for adverse pregnancy and perinatal outcomes, with
medium and dark complexion, those with low UV-B
exposure, and those living with obesity. In future trials,
it will be critically important to assess baseline vitamin
D status and to gain a complete understanding of the
roles of vitamin D dosing strategies and calcium co-
supplementation.
2. Future trials should include umbilical cord blood 25(OH)
D analysis and a plan to follow the offspring throughout
early childhood.
Vitamin D for Adults With Prediabetes
Background
Diabetes mellitus poses a significant challenge to global
health care. Prediabetes increases the risk of developing
diabetes and CVD. In the United States, more than one
in three adults 18 years and older have prediabetes, and
only about 20% of these individuals have been informed
of their prediabetes status by a health care professional.
3. Page 1
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
Worldwide, diabetes affects more than 537 million people,
and this number is predicted to rise to 643 million by 2030
and 783 million by 2045 (231). In clinical trials, intensive
lifestyle changes focused on weight loss and increased
physical activity reduced the risk of developing diabetes
among adults with prediabetes who have impaired glucose
tolerance. However, these lifestyle modifications are
challenging to maintain over the long term. Even with
successful implementation, a residual risk remains, and
most individuals with prediabetes eventually progress to
diabetes. While certain medications approved for treating
type 2 diabetes have been shown to reduce diabetes risk
among people with prediabetes (232), the use of pharma-
cotherapy for diabetes prevention is not widely practiced
or generally recommended due to the associated burden
and cost. The search for weight-independent, easy-to-
implement, and low-cost interventions continues to be a
priority to lower diabetes risk. Over the last decade, sev-
eral studies have reported on the role of vitamin D in at-
tenuating the progression to type 2 diabetes in adults
with prediabetes.
Question 10. Should empiric vitamin D supplementation vs
no empiric vitamin D supplementation be used for adults
with prediabetes (by glycemic criteria)?
Recommendation 10
For adults with high-risk prediabetes, in addition to
lifestyle modification, we suggest empiric vitamin D
supplementation to reduce the risk of progression to
diabetes. (2000)
Technical remarks
• Lifestyle modification must be a routine manage-
ment component for adults with prediabetes.
• The clinical trials informing this recommendation
primarily related to adults with high-risk prediabe-
tes, identified as meeting 2 or 3 American Diabetes
Association glycemia criteria (fasting glucose,
HbA1c, 2-hour glucose after a 75-gram oral glu-
cose challenge) for prediabetes and those with im-
paired glucose tolerance.
• In the clinical trials included in the systematic re-
view, the vitamin D dosages ranged from 842 to
7543 IU (21 to 189 µg) daily equivalent. The esti-
mated weighted average was approximately
3500 IU (88 μg) per day. Participants in some trials
were allowed to remain on their routine supple-
ments, including up to 1000 IU (25 µg) of vitamin
D daily.
Summary of Evidence
The evidence summaries, meta-analysis results, and a de-
tailed summary of the evidence and EtD tables can be found
online at https://guidelines.gradepro.org/profile/zE0nx07
MCXw.
Benefits and Harms
1927
The commissioned systematic review included 11 RCTs
(233-243) that reported on the effect of vitamin D on new-
onset diabetes in adults with prediabetes (total of 5316 par-
ticipants). The trials were conducted in India (n=4), Iran
(n = 1), Greece (n = 1), Norway (n=1), Japan (n=1), and
the United States (n = 3). The panel also considered a recent-
ly published individual participant data meta-analysis
(IPD-MA) (101) of the 3 vitamin D trials (233, 234, 243)
that were specifically designed for diabetes prevention. In
contrast to aggregate data meta-analysis, an IPD-MA in-
creases the statistical power to detect benefits and risks;
avoids ecological fallacy in examining sources of between-
study heterogeneity; and, through data harmonization, im-
proves the precision of results and allows for additional
analyses.
Nine trials (233-239, 241, 242) used cholecalciferol (vita-
min D3), one trial (240) used both cholecalciferol and ergocal-
ciferol (D2), and one trial (243) used eldecalcitol, an active
vitamin D analog. While the panel did not specifically address
vitamin D analogs in its other questions, the panel recognized
the importance of including the second largest trial for dia-
betes prevention (DPVD) (243), which tested eldecalcitol,
when addressing the question about vitamin D and diabetes
prevention; consequently, the findings from the DPVD trial
were incorporated in the evidence synthesis. This approach
aligns the commissioned systematic review with 3 other recent
meta-analyses in this topic (101, 244, 245), ensuring consist-
ency of the evidence synthesis. The results of the commis-
sioned systematic review were similar with or without the
DPVD trial; however, to be consistent with the rest of the
guideline, we first present the meta-analysis results without
the DPVD trial, thereafter, presenting results with the
DPVD trial.
Participants in the included trials were at high risk for dia-
betes, based on having impaired glucose tolerance or meeting
2 or 3 glycemic criteria (fasting glucose, HbA1c, 2-hour glucose
after a 75-gram oral glucose challenge) for prediabetes. The
baseline mean 25(OH)D level in the 11 trials was 12 to
28 ng/mL (30-70 nmol/L). Among the 8 trials that did not in-
clude low baseline 25[OH]D as an eligibility criterion, the base-
line mean level of 25[OH]D was 18 to 28 ng/mL (45-70 nmol/
L). When combining data from the 10 trials (233-242) that
used either cholecalciferol or ergocalciferol, vitamin D reduced
the risk of developing diabetes (RR 0.90 [95% CI, 0.81-1.00]).
The estimated absolute effect size was 24 fewer per 1000 pro-
gressing to type 2 diabetes (46 fewer to 0 fewer). When the
DPVD trial (243) was included, the results were similar (RR
0.90 [95% CI, 0.81-0.99]). The IPD-MA of the 3 trials (233,
234, 243) that were specifically designed for diabetes preven-
tion (total of 4190 participants) showed a 15% reduction in
new-onset diabetes in adults with prediabetes randomized to
vitamin D compared to placebo (HR 0.85 [95% CI,
0.75-0.96]) (101). In these trials, the impact of vitamin D on
new-onset diabetes was in addition to participants receiving
lifestyle interventions for diabetes prevention.
In the commissioned systematic review, the beneficial effect
of vitamin D on diabetes risk was consistent across subgroups
by risk of bias or vitamin D dosage. In the IPD-MA, the effect
of vitamin D appeared to be more pronounced in the follow-
ing subgroups: age older than 62 years (HR 0.81 [95% CI,
Page 2
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1928
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
0.68-0.98]), baseline 25(OH)D level lower than 12 ng/mL
(30 nmol/L) (HR 0.58 [95% CI, 0.35-0.97]), and body mass
index (BMI) less than 30 kg/m² (HR 0.79 [95% CI,
0.66-0.95]) (101). However, the P values for these interac-
tions were not statistically significant.
The commissioned systematic review included 15 RCTs
(234-240, 242, 246-252) that reported the effect of cholecal-
ciferol or ergocalciferol on HbA1c in adults with prediabetes,
12 RCTs (234-238, 241, 242, 246-248, 251, 253) that re-
ported on fasting blood glucose, and 13 RCTs (234-238,
241, 242, 246, 248-251, 254) that reported on blood glucose
2 hours after a 75-gram oral glucose load. Compared to pla-
cebo, vitamin D lowered fasting blood glucose (mean differ-
ence -5.3 mg/dL [95% CI, −7.9 to −2.7]) and 2-hour blood
glucose after a 75-gram oral glucose tolerance test (mean dif-
ference -7.6 mg/dL [95% CI, −12.6 to −2.7]). There was a
trend for vitamin D to lower HbA1c (mean difference
-0.05% [95% CI, −0.10 to 0.01]). When the DPVD trial
(243) was included, the results were similar (mean difference
in fasting blood glucose -4.9 mg/dL [95% CI, −7.3 to −2.4;
2-hour blood glucose −6.6 mg/dL [95% CI, −11.2 to −2.1];
HbA1c -0.04% [95% CI, −0.90 to 0.00).
The commissioned systematic review also examined other
outcomes aside from the risk of diabetes in this population.
The Tromsø study (234) found no differences in upper re-
spiratory infections between those who took 20 000 IU
(500 µg) of vitamin D per week and those who took a placebo.
In the same study, men who received vitamin D had less reduc-
tion in BMD at the femoral neck compared to those who took
a placebo (0.000 vs −0.010 g/cm²; P = .008). There were no
differences in BMD at the femoral neck in women and no dif-
ferences in BMD at the hip in either gender. The study found
no difference in fractures between the vitamin D and placebo
groups; however, the data on fractures was sparse.
Meta-analyses of the 2 trials (234, 255) that used cholecal-
ciferol suggested no clear differences in all-cause mortality
with vitamin D (RR 0.75 [95% CI, 0.26-2.18]; estimated ab-
solute effect size of 1 fewer per 1000 [4 fewer to 6 more]) or
CVD events (234, 256) with vitamin D (RR 1.08 [95% CI,
0.33-3.57]; estimated absolute effect size of 1 more per
1000 [8 fewer to 31 more]). After including the DPVD trial,
results did not change.
The commissioned systematic review found no clear differ-
ence in nephrolithiasis (234, 255) with vitamin D (RR 1.20
[95% CI, 0.71-2.03]; estimated absolute effect size of 3
more per 1000 [5 fewer to 17 more]). There were no cases
of symptomatic hypercalcemia reported in any trial. In the
D2d study, there was 1 case of new-onset kidney disease in
the vitamin D group and 2 cases in the placebo group (RR
0.50 [95% CI, 0.05-5.51]) (255). In the IPD-MA, the fre-
quency of the prespecified adverse events of interest (nephro-
lithiasis, hypercalcemia, and hypercalciuria) was low, and
there were no differences between vitamin D and placebo
(101). In the D2d study, adverse events were overall less fre-
quent in the vitamin D group (4000 IU/day [100 μg/day] of
cholecalciferol) compared to placebo (IRR 0.94 [95% CI,
0.90-0.98]) (255).
Based on the point estimates derived from meta-analyses of
available clinical trials, the panel judged that the anticipated de-
sirable effects of vitamin D for diabetes prevention are likely
moderate, while the anticipated undesirable effects are likely
trivial.
Other Evidence-to-Decision Criteria and
Considerations
Vitamin D is generally available over the counter, and it is inex-
pensive. There are no cost-effectiveness studies of vitamin D
for preventing diabetes, fractures, all-cause mortality, cardiovas-
cular events, or respiratory infections in adults with prediabetes.
However, there is ample evidence of substantial economic value
in preventing the development of type 2 diabetes with non-vita-
min D interventions (eg, lifestyle, metformin) that are more ex-
pensive and burdensome to implement than vitamin D (257).
Therefore, the panel reasoned that there are likely cost savings
with using vitamin D for diabetes prevention.
The panel judged vitamin D use would be acceptable to
adults with prediabetes and to other stakeholders, such as
clinicians. Given ease of administration and low cost, the pan-
el judged empiric vitamin D to lower diabetes risk as a feasible
intervention for adults with prediabetes.
The risk of developing diabetes, the prevalence of diabetes,
and the burdens related to having diabetes are higher among ra-
cial and ethnic minority groups (primarily Hispanic and
non-Hispanic Asian populations) in the United States. In clinical
trials, intensive lifestyle changes have been found to lower the
risk of diabetes, regardless of race or ethnicity. However, access-
ing the necessary resources, such as nutritionists and exercise fa-
cilities, can be difficult, and there are disparities in access to these
resources. Racial and ethnic minority groups (in the United
States) are also at higher risk for having low vitamin D status,
and consumption of vitamin D supplements in these groups is
about half of that compared to non-Hispanic White groups, sug-
gesting differences in vitamin D use. Although vitamin D should
not be viewed as a replacement for lifestyle approaches to dia-
betes prevention, the panel judged that using vitamin D in adults
with prediabetes would likely have a favorable impact on health
equity, especially in low-resource environments.
Justification for the Recommendation
The panel justified a recommendation favoring empiric vita-
min D in adults with prediabetes based on moderate certainty
of evidence that vitamin D likely decreases progression to type
2 diabetes, likely without harm. In the commissioned systemat-
ic review, there was low certainty of evidence for the cardiovas-
cular and mortality outcomes with wide 95% CIs; however,
none of the included trials were designed or powered for car-
diovascular events or mortality, and only 3 trials (including
the DPVD trial) reported on these outcomes. Specific data re-
lated to fractures and respiratory infections were inadequate.
The benefits of vitamin D supplementation may preferentially
accrue to those at highest risk for vitamin D deficiency.
Although not addressed in the commissioned systematic
review, the IPD-MA suggested that the benefit may be great-
est for those with baseline 25(OH)D level lower than 12 ng/
mL (20 nmol/L) (HR 0.58 [95% CI, 0.35-0.97]) (101).
However, overall evidence did not support the net benefit of
25(OH)D testing in adults with prediabetes followed by vitamin
D supplementation in those with low 25(OH)D levels. Vitamin
D supplementation that leads to higher 25(OH)D levels may
further lower the risk of diabetes (101, 258), but it could poten-
tially increase the risk of adverse effects (hypercalcemia, hyper-
calciuria, kidney stones), although there was no evidence of this
in in the IPD-MA (101). In addition, compared to empiric vita-
min D supplementation alone, adding 25(OH)D testing would
Page 3
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
increase costs, thus decreasing feasibility and health equity.
Given these uncertainties, the panel did not recommend screen-
ing or routine monitoring with 25(OH)D in individuals with
prediabetes to guide vitamin D supplementation.
Additional Considerations
Ten trials (including the DPVD trial) reported on the effect of
vitamin D and regression to normal glucose regulation, defined
as having glycemic measures in the normal range, in people with
prediabetes. The commissioned systematic review did not com-
bine data on the effect of vitamin D on regression to normal glu-
cose regulation; however, other meta-analyses have synthesized
data on this outcome. Zhang et al combined aggregate data
from 5 trials totaling 1080 participants with prediabetes and
found a significant vitamin D benefit for regression to normal
glucose regulation by 48% compared to placebo (RR 1.48
[95% CI, 1.14-1.92]) (244). In the IPD-MA, vitamin D in-
creased the likelihood of regression to normal glucose regulation
by 30% (RR 1.30 [95% CI, 1.16-1.46]) (101).
The clinical trials informing this recommendation primarily
related to adults with high risk for diabetes, identified by meet-
ing 2 or 3 American Diabetes Association glycemia criteria (fast-
ing glucose, HbA1c, 2-hour glucose after a 75-gram oral glucose
challenge) for prediabetes or by having impaired glucose toler-
ance. The panel's use of the term “high-risk prediabetes” aligns
with the clinical trial evidence and aims to focus the recommen-
dation on adults at the highest risk for diabetes, not to mandate
specific testing methods.
The included trials used varying dosages of cholecalciferol or
ergocalciferol. The median (interquartile range) dosage em-
ployed was approximately 2663 (1410-3893) IU/day (67
[35-97] µg/day), and the estimated weighted average was
3520 IU (88 μg) per day. Due to this variability, the panel could
not recommend a specific dosage of vitamin D. In general, trial
participants in both active and placebo groups were allowed to
take vitamin D supplements on their own, up to a certain dosage
specific for their age.
While the absolute reduction in the risk of developing
new-onset diabetes may be relatively small, the panel considered
that such interventions with modest benefits could significantly
impact prevalent conditions like prediabetes. For example, the
absolute 3-year risk reduction in diabetes risk with vitamin D
(24 fewer per 1000 participants based on the systematic review
or 33 fewer per 1000 based on the IPD-MA) compares favorably
with metformin in the Diabetes Prevention Program in the
United States (70 fewer per 1000), especially when considering
that in the clinical trials, the vitamin D intervention was applied
in addition to recommended lifestyle changes.
Research Considerations
Proposed areas for research include:
1. Randomized controlled trials to evaluate a treat-to-target
strategy to define the 25(OH)D level that optimally re-
duces the risk of new-onset diabetes and increases time
spent in normoglycemia.
2. Randomized controlled trials designed to identify subpo-
pulations with prediabetes who are more likely to benefit
from vitamin D, focusing not only on biological varia-
bles, including body composition, but on environmental,
lifestyle, and dietary factors.
3. Cost-effectiveness analyses.
1929
4. Implementation studies to assess the practicality and ef-
fectiveness of vitamin D in real-world settings.
5. Studies on the effect of vitamin D in people at risk for or
with new-onset type 1 (autoimmune) diabetes.
Vitamin D Dosing
Background
There is uncertainty regarding the best approach to vitamin D
supplementation. Options range from daily intake to less fre-
quent regimens, such as weekly or monthly. While infrequent
dosing may improve adherence, large doses of vitamin D have
been associated with higher levels of inactive 24,25(OH)2
vitamin D (259), raising concerns about the benefit-risk ratio
of intermittent, high doses of vitamin D. Important questions
include the effect of nondaily dosing on clinical outcomes and
potential impact on the risk of adverse events.
Question 11. Should a daily, lower-dose vitamin D vs non-
daily (ie, intermittent), higher-dose vitamin D be used
for nonpregnant people for whom vitamin D treatment
is indicated?
Recommendation 11
In adults aged 50 years and older who have indica-
tions for vitamin D supplementation or treatment,
we suggest daily, lower-dose vitamin D instead of
nondaily, higher-dose vitamin D. (2 | OO)
Technical remark
• The panel did not identify evidence related to indi-
viduals younger than age 50 years.
Summary of Evidence
The evidence summaries, meta-analysis results, and a detailed
summary of the evidence and EtD tables can be found online
at https://guidelines.gradepro.org/profile/rzh7ywOCSRY.
Benefits and Harms
Two trials (260,261) with a total of 537 patients met the original
inclusion criteria, which specified a direct comparison between
intermittent high-dose vs daily lower-dose vitamin D supplemen-
tation. After expanding eligibility criteria to include trials that
compared high-dose intermittent doses vs placebo, the systemat-
ic review included 19 manuscripts derived from 15 studies (29)
involving 53 527 participants. In the included trials, daily vita-
min D doses ranged from 400 to 800 IU (10-20 µg). Doses given
at nondaily intervals included 50 000 IU (1250 µg) every 2
weeks, 60 000-100 000 IU (1500-2500 μg) monthly,
96 000-150 000 IU (2400-3750 μg) every 2 to 4 months, and
300 000 IU-500,000 IU (7500-12 500 μg) annually.
The systematic review identified 5 studies (104, 119, 180,
184, 190) that evaluated fractures with participants as the
unit of analysis. There was a trend for intermittent high-dose
vitamin D to increase fracture risk (RR 1.08 [95% CI,
0.98-1.19]), with an estimated absolute effect size of 5 more
participants with a fracture per 1000 (1 fewer to 11 more).
In subgroup analyses, studies involving doses higher than
Page 4
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1930
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
100 000 IU (2500 µg) may have had higher risk of fracture
(RR 1.14 [95% CI, 1.02-1.27]) than those involving lower
doses (RR 0.94 [95% CI, 0.79-1.12]) (P = .07 for interaction).
When examining the 7 studies (118, 174, 180, 184, 185, 260,
261) reporting the total number of fractures as the unit of ana-
lysis, the IRR for fractures was 0.96 (95% CI, 0.75-1.21) for
intermittent high-dose vitamin D. Studies involving doses
higher than 100 000 IU (2500 µg) had an IRR of 1.23 (95%
CI, 0.81-1.61) compared to an IRR of 0.86 (95% CI,
0.71-1.02) for studies involving doses 50 000-100 000 IU
(1250-2500 μg) (P = .026 for interaction). In study subgroup
analyses, dosing interval (every 1-12 weeks vs > 12 weeks for
intermittent high-dose vitamin D) was not a significant pre-
dictor of fracture risk.
In the meta-analysis of 6 studies (104, 118, 119, 174, 176,
184) reporting on falls with participants as the unit of ana-
lysis, the RR for intermittent high-dose vitamin D was 1.01
(95% CI, 0.93-1.10). Study subgroup analyses suggested the
possibility that doses greater than 100 000 IU (2500 μg)
may have higher fall risk (RR 1.04 [95% CI, 0.96-1.12]) com-
pared to lower doses (RR 0.79 [95% CI, 0.61-1.03]) (P = .056
for interaction). Studies employing a dosing interval greater
than every 12 weeks showed higher fall risk with vitamin D
(RR 1.08 [95% CI, 1.03-1.14]) compared to dosing intervals
of 1 to 12 weeks (RR 0.98 [0.92-1.04]) (P=.01 for inter-
action). Analysis of 6 studies (118, 184, 185, 190, 200, 260)
that reported on the number of falls as the unit of analysis re-
vealed an IRR of 1.05 (95% CI, 0.96-1.13) for intermittent,
high-dose vitamin D; subgroup analyses for falls as a unit of
analysis did not disclose significant study subgroup effects ac-
cording to dose or dosing interval.
For the 5 studies (85, 119, 123, 201, 260) reporting partic-
ipants with respiratory infections as the unit of analysis, there
were no differences between high-dose nondaily vitamin D vs
placebo (OR 1.00 [95% CI, 0.98-1.03]). Similarly, analysis of
4 studies (85, 123, 139, 260) that reported on the number of
respiratory infections as the unit of analysis revealed an IRR
of 0.98 (95% CI, 0.88-1.03) for intermittent, high-dose vita-
min D. Study subgroup analyses did not implicate vitamin D
dose as a predictor of these study outcomes.
The 3 studies (118, 124, 138) that reported on nephrolithia-
sis administered 50 000 to 100 000 IU (1250-2500 µg) vita-
min D every 2 to 4 weeks. The RR for nephrolithiasis was
1.00 (95% CI, 0.84-1.19) for intermittent, high-dose vitamin
D. Two studies (119, 166) did not disclose a clear difference
in kidney disease (RR 0.64 [95% CI, 0.28-1.47]), with an
estimated absolute effect size of 2 fewer per 1000 (3 fewer
to 2 more). No trials reported cases of symptomatic
hypercalcemia.
Based on the panel's best estimates of treatment effects in
adults aged 50 years and older, the panel judged that any de-
sirable effects of intermittent, high-dose vitamin D (compared
to lower-dose, daily vitamin D) are likely trivial, while the an-
ticipated undesirable effects are likely to be small.
Other Evidence-to-Decision Criteria and
Considerations
Vitamin D is relatively inexpensive and available over the
counter; however, higher dosages may require prescriptions,
which increase cost and burden. The panel did not identify
any cost-effectiveness studies addressing daily lower-dose
vitamin D vs intermittent, higher-dose vitamin D. The panel
did not identify any studies that addressed the potential im-
pact of intermittent high-dose vitamin D vs daily lower-dose
vitamin D on health equity, although any additional costs
and requirements for health care visits could decrease health
equity. The panel identified no studies that addressed the pos-
sibility of differential acceptability or feasibility of intermit-
tent high-dose vitamin D vs daily lower-dose vitamin
D. Nonetheless, the panel assumed that less frequent dosing
(weekly, monthly, or yearly) may be more acceptable to
some individuals and may possibly be associated with better
adherence, based on experience with medications like bi-
sphosphonates, for which nondaily administration improves
adherence (262).
Justification for the Recommendation
The available evidence (which is specifically pertinent to per-
sons age 50 years) suggests that, compared to daily lower-
dose vitamin D or placebo, intermittent high-dose vitamin D
offers no desirable effects, and may be associated with un-
desirable anticipated effects (namely, moderate certainty of
evidence suggests an estimated 0.5% absolute increase in frac-
ture risk). The panel judged that the potential convenience ad-
vantage of intermittent high-dose vitamin D may be
outweighed by the potential for undesirable anticipated ef-
fects. The panel identified no evidence to suggest material dif-
ferences in cost, equity, or feasibility, although cost likely
favors daily, lower-dose vitamin D, since the higher dosages
commonly require a prescription and thus involve the costs
of health care visits. Since overall certainty of evidence was
very low, and since individuals may value anticipated advan-
tages and disadvantages differently, the panel issued a condi-
tional recommendation.
Screening for Low Vitamin D Status With 25(OH)D
Testing
Vitamin D deficiency is traditionally defined clinically as hav-
ing symptoms and signs of rickets or osteomalacia. Although
these conditions are not uncommon, vitamin D “deficiency" is
more frequently defined based on circulating 25(OH)D levels.
However, the 25(OH)D level for defining deficiency has been
controversial, thus the prevalence of vitamin D deficiency var-
ies depending on the 25(OH)D threshold used. For example, if
vitamin D deficiency is defined as a 25(OH)D concentration
less than 20 ng/mL (50 nmol/L), 24% of US adults meet that
criterion, whereas if defined as a 25(OH)D concentration
less than 10 ng/mL (25 nmol/L), 6% of US adults would be
considered vitamin D-deficient (82).
Low vitamin D status has been associated with increased
risks for several common chronic conditions, such as osteopor-
osis (risk of fractures), CVD, and diabetes. However, whether
vitamin D supplementation lowers risk for developing such
outcomes in generally healthy populations has remained un-
clear. Nonetheless, rates of screening for low 25(OH)D levels
have increased in recent years. For example, in one study, test-
ing with 25(OH)D rose from 0.29 per 1000 person-years at
risk (95% CI, 0.27-0.31) in 2005 to 16.1 per 1000 person-
years at risk (95% CI, 15.9-16.2) in 2015 (263).
The panel prioritized 3 clinical questions related to screen-
ing for 25(OH)D levels and whether vitamin D should be giv-
en only to individuals who have 25(OH)D levels below a
threshold, recognizing that appropriate thresholds likely
vary based on the outcome of interest. In particular, the panel
Page 5
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
chose to address 25(OH)D screening in adults with dark com-
plexion, in adults with obesity, and in the general adult popu-
lation who do not have otherwise an established indication for
screening (eg, hypocalcemia). These screening questions relate
to whether vitamin D administration may be primarily―or
perhaps even exclusively-beneficial for adults with 25(OH)
D levels below a population- and condition-specific (and
thus far undetermined) threshold. If the net benefit of vitamin
D supplementation specifically accrues to those with low
25(OH)D levels, then it could be important to perform
25(OH)D testing to identify those individuals. In contrast, if
the net benefit of vitamin D supplementation does not specif-
ically accrue to those with 25(OH)D levels below a threshold
(ie, if net benefit is also realized in those with 25[OH]D levels
above that threshold), or if no net benefit of vitamin D admin-
istration is apparent, then 25(OH)D screening in these popu-
lations would presumably be unnecessary.
Importantly, for all 3 screening questions, no studies were
identified that compared a screening approach (testing for
25[OH]D levels followed by vitamin D treatment as indicated)
to a nonscreening approach. Therefore, the panel's approach
to the 3 screening questions followed a framework proposed
by Murad and colleagues (264). These criteria can be broadly
grouped into considerations related to the medical condition
in question, the test's characteristics, and the overall impact
on patient care. According to this framework, screening would
be justified when the following conditions are met:
• Importance: The condition is an important health prob-
lem in terms of prevalence and/or consequences.
The panel noted that low vitamin D status has been linked
to a number of important health problems.
• Natural history: The condition for which screening is
being performed has a well-understood natural history
that includes a latent (preclinical) phase.
The panel agreed that the adverse effects of low vitamin D
status may manifest only after a long latency period, and early
detection could plausibly lead to better long-term outcomes.
• Difference in management and treatment availability:
Persons with positive screening test results would be man-
aged differently from those with negative screening test
results.
The panel agreed that vitamin D supplementation is widely
available, inexpensive, and highly effective at raising 25(OH)
D levels.
• Test accuracy and safety: High- or moderate-certainty evi-
dence supports acceptable accuracy of the screening test
(eg, acceptable false-positive and false-negative rates).
There have been considerable efforts over the last
decade to standardize the 25(OH)D assays, and the assays
are significantly more reproducible than in the past, as
most large laboratories follow a standardization protocol
based on the work of the Vitamin D Standardization
Protocol (https://www.cdc.gov/labstandards/csp/pdf/hs/
vitamin_d_protocol-508.pdf). However, there is still
1931
considerable variability of 25(OH)D assays. The systemat-
ic review did not identify any studies showing that 25(OH)
D testing is harmful.
• Available treatment: Effective management is available
that improves patient-important outcomes when imple-
mented in the latent (preclinical) phase.
Vitamin D supplementation is highly effective at raising
25(OH)D levels. Questions regarding whether vitamin D
supplementation lowers the risks of patient-important out-
comes including in those with low 25(OH)D concentrations
specifically were the primary objective of the commissioned
systematic reviews described throughout this document.
• Difference in outcomes: The benefits of management ac-
cording to screening results outweigh the harms of screen-
ing (eg, overdiagnosis, unnecessary treatment for false
positives, anxiety, stigma, etc.).
The panel did not identify any harms related to screening
other than the financial costs associated with tests, health
care visits, and (potentially) unnecessary treatment.
• Other considerations: The screening strategy should be
cost-effective, acceptable to relevant stakeholders, and
feasible to implement.
Vitamin D supplementation and 25(OH)D testing
are judged to be acceptable and feasible. Data on imple-
mentation costs and cost-effectiveness considerations are
scant.
This section addresses whether to screen with a 25(OH)D
test in generally healthy populations. The panel did not specif-
ically address whether and how those who present with docu-
mented low levels of 25(OH)D should be evaluated and/or
treated.
Vitamin D Screening With a 25(OH)D Test for
Healthy Adults
Background
Recent trends have shown a rise in screening rates for vitamin D
status using serum 25(OH)D in the general population.
Specifically, 25(OH)D testing frequency rose from 0.29 per
1000 person-years at risk in 2005 to 16.1 per 1000 person-years
at risk by 2015, highlighting a growing interest by patients and
physicians in assessing vitamin D status (263, 265).
Advocating for the routine screening of 25(OH)D levels in
healthy adults is contingent upon demonstrating that such
screenings can effectively identify individuals with low 25(OH)
D who might not be detected through traditional risk factor as-
sessments, and that vitamin D supplementation, following the
identification of a low 25(OH)D level, leads to improvements
in clinical outcomes (eg, prevention of osteoporosis, CVD, dia-
betes, respiratory infections, overall mortality).
Screening for low 25(OH)D in generally healthy adults (ie,
those who are not at increased risk for vitamin D deficiency)
would involve testing large numbers of people, with import-
ant implications for health care systems.
The US Preventive Services Task Force (USPSTF) recently
concluded that there was insufficient evidence to inform a
Page 6
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1932
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
decision regarding the balance of benefits and harms of screen-
ing for vitamin D status with 25(OH)D in asymptomatic
adults (266). A recommendation against population screening
for vitamin D deficiency with 25(OH)D is included in the
"Choosing Wisely” campaign, an initiative by the American
Board of Internal Medicine to spark conversations between
clinicians and patients about the value of common tests
(choosingwisely.org).
Question 12. Should screening with a 25(OH)D test (with
vitamin D supplementation/treatment only if below a
threshold) vs no screening with a 25(OH)D test be used
for healthy adults?
Recommendation 12
In healthy adults, we suggest against routine screen-
ing for 25(OH)D levels. (2 | 000)
Technical remarks
• In healthy adults, 25(OH)D levels that provide
outcome-specific benefits have not been estab-
lished in clinical trials.
This recommendation relates to adults who do not
otherwise have established indications for testing
with 25(OH)D levels (eg, hypocalcemia).
Summary of Evidence
The evidence summaries, meta-analysis results, and a
detailed summary of the evidence and EtD tables can
be found online at https://guidelines.gradepro.org/profile/
7Pf6NRYV8TE.
Benefits and Harms
The benefits and harms of vitamin D supplementation in
nonpregnant adults with a 25(OH)D concentration below a
threshold are addressed in clinical questions 3, 5, and 7.
The available clinical trial data were insufficient to satisfactor-
ily assess whether net benefit varied according to baseline
25(OH)D level. When available, trial subgroup analyses
did not clearly indicate that net benefit of vitamin D specific-
ally accrues to those with low baseline 25(OH)D level.
However, conclusions drawn from subgroup analyses in
each individual trial are limited as subgroups lacked adequate
statistical power. Meta-analyses that combine aggregate study
data and perform subgroup analyses according to average
25(OH)D in each study are subject to ecological fallacy,
and thus were not included in the systematic reviews commis-
sioned for this guideline.
Other Evidence-to-Decision Criteria and
Considerations
The panel judged that screening for 25(OH)D would be ac-
ceptable to relevant stakeholders, assuming net benefit is
expected. While the panel judged that screening would be
feasible for many individuals, there are costs that accom-
pany screening, including the direct and indirect costs of
a visit for the test, the cost of the 25(OH)D test itself,
the time and cost for a health care provider visit to review
results, and potentially follow-up visits for consultation
and more testing. Variable access to 25(OH)D testing could
be an important barrier for some. In addition, screening en-
tire adult populations would involve substantial costs and
effort, thus feasibility from a societal perspective is unclear.
The panel did not identify studies that adequately ad-
dressed the cost-effectiveness of 25(OH)D screening in all
adults. One study estimated that among White adults
aged 65 to 80 years, screening would be slightly more ef-
fective than universal supplementation for reducing falls
and mortality (267). However, this modeling study relied
on trials published more than 15 years ago, so its current
relevance is unclear.
The effect of screening on health equity is unclear.
Screening with 25(OH)D may worsen health equity because
screening requires resources that may not be universally
available or accessible, but it could improve health equity if
screening led to the identification and effective treatment of
prevalent and important health conditions in disadvantaged
populations.
Justification for the Recommendation
The panel's conditional recommendation against routine
screening for 25(OH)D levels in generally healthy adults
primarily related to the lack of clinical trial-based evidence
regarding what 25(OH)D levels would inform a treatment
decision and the resultant effect of treatment with vitamin
D, compared with no screening. The panel also considered
the lack of clinical trial evidence that clearly supports the
hypothesis that net benefit specifically accrues to those
with a 25(OH)D level below a threshold. The panel was un-
certain that any putative benefits of screening would out-
weigh the increased burden and cost, and whether
implementation of universal 25(OH)D screening would be
feasible from a societal perspective. Importantly, the panel
recognized that it is possible that there is no single
threshold 25(OH)D level appropriate for the entire general
population.
Vitamin D Screening With a 25(OH)D Test for
Adults With Dark Complexion
Background
"Dark complexion" is defined by a phenotype that involves
the color of eyes, hair, and skin. In relation to vitamin D,
the panel was especially interested in skin pigmentation,
determined by the amount of melanin that can interfere
with the production of vitamin D in response to exposure
to UV-B rays. People at higher risk of low vitamin D status
include individuals with dark skin, generally those whose an-
cestors originated from sunnier regions of the planet,
including those with African heritage and descendants of in-
digenous peoples of the Americas, Oceania, and Asia. In add-
ition, some (268, 269) but not all (270) studies suggest that
the increase in 25(OH)D levels in response to vitamin D sup-
plementation is not as robust in those with dark complexion
compared to those with lighter complexion. Importantly,
many such studies assessed groups according to race/
ethnicity rather than skin complexion, introducing uncertainty
regarding the degree to which dark complexion per se impacts
circulating 25(OH)D concentrations. Nonetheless, since lower
Page 7
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
25(OH)D levels have been consistently observed in populations
who tend to have darker complexion (147), the panel judged
that it would be important to determine whether screening
for 25(OH)D levels is beneficial in persons with dark
complexion.
Question 13. Should screening with a 25(OH)D test (with
vitamin D supplementation/treatment only if below a
threshold) vs no screening with a 25(OH)D test be used
for adults with dark complexion?
Recommendation 13
In adults with dark complexion, we suggest against
routine screening for 25(OH)D levels. (2 | OOO)
Technical remarks
• This recommendation relates to generally healthy
adults with dark complexion who do not otherwise
have established indications for 25(OH)D testing
(eg, hypocalcemia).
• The panel did not identify any clinical trials that
related clinical outcomes to skin complexion per
se. A secondary analysis did not clearly suggest
net benefit with vitamin D in those who self-
identify as Black. The panel recognized that self-
identified race is an inaccurate and otherwise
problematic proxy for dark complexion.
Summary of Evidence
The evidence summaries, meta-analysis results, and a detailed
summary of the evidence and EtD tables can be found online
at https://guidelines.gradepro.org/profile/pHw68lfsrzU.
Benefits and Harms in Persons With Dark
Complexion
The systematic review did not identify any trials that exam-
ined whether screening with 25(OH)D (and vitamin D treat-
ment when 25[OH]D is found to be low) improves the
outcomes of interest in people with dark complexion per se.
The systematic review also did not identify any vitamin D tri-
als that assessed whether outcomes of interest vary according
to skin complexion per se. This absence of high-quality sup-
portive data was the primary reason why the panel suggested
against routine 25(OH)D screening in adults with dark skin
complexion.
Benefits and Harms in Persons Who Self-Identify as
Black
The panel specifically aimed to address screening for 25(OH)
D in persons with dark complexion given that melanin can
interfere with endogenous vitamin D production in response
to sun (UV-B) exposure. The panel also recognized that clinic-
al questions about the utility of 25(OH)D screening are fre-
quently posed for racial groups in which dark complexion is
common (although variable). The panel judged that such clin-
ical questions are not without merit, especially given differen-
ces in 25(OH)D levels in people of different races and
1933
ethnicities. In a recent analysis of the NHANES in the
United States, 25(OH)D levels lower than 10 ng/mL
(<25 nmol/L) were present in 1% of those who self-identified
as White and in 11% of those who self-identified as Black,
with levels of 12 to 20 ng/mL (25-50 nmol/L) in 14% and
49% of those self-identifying as White and Black, respectively
(147). However, the panel recognized that racial categories re-
present social rather than biological constructs, and self-
identified race is an inaccurate proxy for skin complexion
(271). Although as a group, persons who self-identify as
Black have darker skin complexion, they have highly variable
skin pigmentation. Accordingly, using race as a proxy for
skin complexion is subject to ecological fallacy and will mis-
classify many individuals. In addition, other factors (eg, social
determinants of health) may be associated with both self-
identified race and risk of low 25(OH)D levels, and outcomes
of interest (eg, risk of diabetes), yielding uncertainty regarding
the degree to which skin pigmentation per se predicts vitamin
D-related outcomes in clinical studies (271). Nonetheless, giv-
en that clinicians frequently pose similar clinical questions for
subgroups defined by race, the systematic review included a
secondary analysis that addressed the potential benefits and
harms of 25(OH)D screening in persons who self-identified
as Black or African American.
The systematic review did not identify any trials that exam-
ined whether outcomes are improved by screening with
25(OH)D (with vitamin D treatment when 25[OH]D is found
to be low) in people who self-identify as Black. Hence, the
panel gathered evidence from clinical trials that reported re-
sults for the prespecified outcomes of interest in subgroup ana-
lyses by self-identified race.
The systematic review identified 2 RCTs that reported
subgroup analyses on fracture risk in individuals who self-
identified as Black. The VITAL trial (105) reported no
difference in the incidence of total, nonvertebral, and hip frac-
tures among 5106 Black participants who received 2000 IU
(50 µg/day) of vitamin D daily vs placebo (HR 0.89 [95%
CI, 0.62-1.30]). The baseline 25(OH)D level among Black
participants was 25 ng/mL (62.5 nmol/L), and the cohort
was at low baseline risk for fractures. The Women's Health
Initiative (WHI) study (110) showed no statistically significant
benefit of low-dose vitamin D (400 IU/day; 10 µg/day)
(co-administered with calcium) over placebo (HR 0.73
[95% CI, 0.16-3.32]) on hip fractures in the subgroup of
3317 postmenopausal women who self-identified as Black.
One study (272) reported subset analyses on all-cause mor-
tality in women who self-identified as Black, showing no dif-
ference between vitamin D (co-administered with calcium)
and placebo (HR 0.97 [0.84-1.11]).
Three RCTs (20, 126, 272, 273) reported on the impact of
vitamin D on cardiovascular adverse events in Black persons.
In the VITAL trial (20), the risk of major adverse cardiovascu-
lar adverse events in those randomized to vitamin D vs pla-
cebo was similar in those who self-identified as Black (HR
0.91 [95% CI, 0.65-1.26], 5106 participants) and those
who self-identified as White (HR 0.93 [95% CI, 0.79-1.10]).
No differences in CVD between vitamin D vs placebo groups
were observed in the PODA trial (Physical Performance,
Osteoporosis and Vitamin D in African American Women tri-
al) (HR 2.23 [95% CI, 0.85-6.23]; 260 African American
female participants) (126) and the WHI study (vitamin D co-
administered with calcium, HR 0.99 [95% CI, 0.87-1.13];
3325 Black female participants) (272). Among Black women
Page 8
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1934
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
in the WHI (272, 273), there was no significant difference be-
tween vitamin D (co-administered with calcium) vs placebo on
the risk of MI (HR 0.89 [95% CI, 0.66-1.20]), heart failure
(HR 0.95 [95% CI, 0.73-1.23]), stroke (HR 0.87 [95% CI,
0.68-1.12]), transient ischemic attack (HR 0.99 [95% CI,
0.71-1.38]), or undergoing coronary artery bypass grafting
or percutaneous transluminal coronary angioplasty (HR
1.05 [95% CI, 0.80-1.38]).
The systematic review identified 2 RCTs (20, 272) and 1
observational study (274) that reported on the risk of devel-
oping cancer in Black participants. The VITAL trial (20) re-
ported a HR of 0.77 (95% CI, 0.59-1.01; 5106
participants), and the WHI (272) reported a HR of 0.99
(95% CI, 0.84-1.16) Among Black participants in the WHI
trial, vitamin D co-administered with calcium was not statis-
tically different than placebo for gastrointestinal cancer (HR
0.83 [95% CI, 0.60-1.15]), hematologic cancer (HR 0.72
[95% CI, 0.52-1.23]), lung cancer (HR 0.98 [95% CI,
0.63-1.51]), or breast cancer (HR 0.95 [95% CI,
0.74-1.23]) (272). In a 10-year observational cohort study
(274) of women with sisters who had breast cancer, no asso-
ciation was found between use of use of vitamin D supple-
ments and breast cancer among Black women (HR 0.89
[95% CI, 0.68-1.2]).
The systematic review did not identify trials addressing the
role of vitamin D in preventing respiratory infections in adults
who self-identify as Black.
No significant differences in kidney stones, symptomatic
hypercalcemia, kidney disease, or renal failure were observed
in the RCTs that performed subgroup analyses of adults who
self-identified as Black.
In summary, based on an assessment of the small number of
clinical trials that reported results according to self-identified
race, vitamin D did not clearly have a beneficial effect on
fractures, mortality, cardiovascular events, or cancer among
participants who self-identified as Black or African American.
Available studies only addressed US individuals who self-
identified as Black or African American, limiting generalizabil-
ity. Data were insufficient for other populations in which dark
complexion is common (eg, descendants of certain indigenous
populations of Asia, the Americas, or Oceania).
Other Evidence-to-Decision Criteria and
Considerations
The panel judged that testing for 25(OH)D levels (with vita-
min D supplementation/treatment as indicated) would be ac-
ceptable to many, although access to testing is variable
across the globe, which may limit feasibility for some.
Conditioning vitamin D supplementation on 25(OH)D test re-
sults would be expected to increase costs and burden, and the
panel did not identify any studies that adequately addressed
the cost-effectiveness of such an approach.
The panel did not identify any studies that adequately ad-
dressed the potential equity impact of 25(OH)D screening
for people with dark complexion, although the panel had con-
cerns that such a testing approach could negatively impact
health equity, especially given the absence of evidence for a
net benefit with vitamin D supplementation in those with
both dark complexion and low 25(OH)D. The panel also con-
sidered the potential equity impact of a 25(OH)D screening
strategy vs empiric vitamin D supplementation in those with
dark complexion. Similar to the general population, from
an equity standpoint, the panel judged that empiric vitamin D
supplementation could possibly be preferred to a screening
strategy assuming that net benefit is expected from vitamin
D supplementation—since it does not require healthcare access,
overall anticipated costs would be lower, and since vitamin D
supplementation is judged to be safe when kept within tolerable
upper intake levels as recommended by the IOM.
Justification for the Recommendation
The panel's conditional recommendation against routine
25(OH)D screening for those with dark complexion primarily
related to the lack of clinical trial evidence that would support
the benefit of 25(OH)D screening in addition to the lack of
clinical trial evidence that would support net benefit related
to vitamin D supplementation in those with dark complexion.
The panel was also uncertain that any putative benefits of
screening would outweigh potential downsides, including
the costs of 25(OH)D tests, and whether implementation of
25(OH)D screening for those with dark complexion would
be feasible from a societal perspective.
Research Considerations
1. Clinical trials should address whether the benefits and
harms of vitamin D screening (and treatment) vary ac-
cording to skin complexion per se (a biological character-
istic relevant to vitamin D), rather than using
self-identified race (a social construct) as a proxy for
skin complexion. At the same time, research is needed
to assess whether the benefits and harms of vitamin D
screening and/or treatment vary according to race/
ethnicity, as well as to define how social determinants
of health vs biological factors (eg, skin pigmentation) im-
pact clinical outcomes. Research should also address
whether advisable vitamin D intake (ie, DRIs) varies ac-
cording to skin complexion, race/ethnicity, or both.
2. It will be important to undertake studies to determine the
concentrations of 25(OH)D that are considered optimal
for disease prevention in individuals with dark complex-
ion, and what dosages of vitamin D are required to
achieve these levels.
3. People with dark complexion are overrepresented in im-
migrants to northern latitudes and in resource-poor set-
tings. The consequences of low vitamin D levels in this
population are not well studied.
Vitamin D Screening With a 25(OH)D Test for
Adults With Obesity
Background
Low serum 25(OH)D levels are common among people with
obesity. This is likely multifactorial, including insufficient
dietary intake of vitamin D; reduced sun exposure; diminished
25-hydroxylase activity (275); and changes in the
gut micro-
biome, which have been shown to affect vitamin D absorption
(276, 277). Notably, the absolute increase in 25(OH)D levels
observed after 2 years of vitamin D supplementation (2000 IU
[50 µg] per day) was attenuated in participants with obesity,
relative to those with a BMI <25 kg/m² (10.5 vs 13.5 ng/mL
[26 vs 34 nmol/L]) (278). After adjustment for other potential
predictors, adults with obesity in the United States were found
to have a 3-fold higher prevalence of 25(OH)D less than
20 ng/mL (50 nmol/L) and 2-times higher prevalence of
Page 9
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
25(OH)D between 20 and 30 ng/mL (50-75 nmol/L) than
adults without obesity (279).
While obesity is associated with higher bone density, this
does not necessarily translate into a reduced risk of fractures.
In fact, postmenopausal females with obesity were shown to
have a 50% higher risk of ankle and 70% higher risk of upper
leg fractures (280). Obesity has also been associated with an in-
creased risk of diabetes, all-cause mortality, CVD, cancer, and
lower tract respiratory infections. Notably, levels of 25(OH)D
lower than 20 ng/mL (50 nmol/L) are associated with an in-
creased risk of cardiometabolic mortality. Data from
NHANES suggest an additive effect of obesity and low vitamin
D status (25[OH]D less than 12 or 20 ng/mL [30 or 50 nmol/L])
on CVD, cancer mortality, and all-cause mortality (281).
Thus, if optimizing 25(OH)D levels lowers risk of these
conditions, high value could be realized at both individual
and health care system levels.
Question 14. Should screening with a 25(OH)D test (with
vitamin D supplementation/treatment only if below a
threshold) vs no screening with a 25(OH)D test be used
for adults with obesity?
Recommendation 14
In adults with obesity, we suggest against routine
screening for 25(OH)D levels. (2 | OOO)
Technical remarks
• In adults with obesity, 25(OH)D thresholds that
provide outcome-specific benefits have not been
established in clinical trials.
• This recommendation relates to generally healthy
adults with obesity who do not otherwise have es-
tablished indications for 25(OH)D testing (eg,
hypocalcemia).
Summary of Evidence
The evidence summaries, meta-analysis results, and a detailed
summary of the evidence and EtD tables can be found online
at https://guidelines.gradepro.org/profile/iNI2fGcamG8.
Benefits and Harms
The systematic review did not identify any trials that exam-
ined whether screening for 25(OH)D levels (with vitamin D
treatment when 25[OH]D is found to be low) in people with
obesity improves the prespecified outcomes of interest.
Thus, clinical trials in which subgroup analyses were per-
formed by baseline BMI were examined.
Vitamin D supplementation in adults with a BMI higher
than 30 kg/m² was not shown to have a significant effect on
fractures in 2 RCTs reporting on fractures in participants
with obesity. The VITAL RCT performed in the United
States (105) showed no reduction in fracture risk in adults
with a BMI higher than 30 kg/m² randomized to 2000 IU
(50 μg) vitamin D daily vs placebo (HR 1.17 [95% CI,
0.95-1.44]). However, the average baseline 25(OH)D level
in those with a BMI higher than 30 kg/m² was 28 ng/mL
1935
(72 nmol/L), the cohort was at low risk for fractures, and out-
comes based on baseline 25(OH)D in individuals with obesity
were not presented. The WHI, which was also performed in
the United States (110) showed a HR of 0.73 (95% CI,
0.49-1.09) for femoral fractures in female individuals with a
BMI higher than 30 kg/m² who were randomized to
1000 mg calcium and 400 IU (10 µg) vitamin D supplementa-
tion daily vs placebos. Baseline 25(OH)D levels were not
available in this subgroup. In the entire study cohort, the
risk of fracture decreased among those who were adherent
to calcium and vitamin D treatment, but there are no available
data among those with a BMI higher than 30 kg/m² who were
adherent to study medications.
The WHI also examined the effects of vitamin D (400 IU
[ug] with 1000 mg calcium daily) on all-cause mortality in in-
dividuals with obesity and did not show a statistically signifi-
cant effect (HR 0.93 [95% CI, 0.80-1.09]), including among
participants adherent to study medications (HR 0.87 [95%
CI, 0.73-1.04] (121).
Two RCTs (20, 125) examining the incidence of major car-
diovascular events in individuals with obesity found no over-
all benefit in those who were randomized to vitamin D. In the
VITAL study, participants with obesity who received 2000 IU
(50 μg) of vitamin D daily and those who received placebo had
a comparable risk of cardiac events (HR 0.98 [95% CI,
0.76-1.26] (20). Of interest, the FIND trial, performed in
Finland (125), suggested a reduction in the risk of developing
a major cardiovascular event in those with a BMI higher than
30 kg/m² receiving higher dosages of vitamin D (3200 IU/day
[80 µg/day]), which increased 25(OH)D levels from 30 ng/mL
(75 nmol/L) at baseline to 48 ng/mL (120 nmol/L) (HR 0.19
[95% CI, 0.04-0.82]).
Three RCTs and one observational study examined the ef-
fect of vitamin D on the development of cancer in adults
with obesity (29). None of these trials demonstrated a signifi-
cant effect of vitamin D on developing cancer. An HR of 1.13
(95% CI, 0.9-1.37); 2000 IU [50 µg] vitamin D/day) was
reported in the VITAL trial (20). In the FIND trial (125) nei-
ther the high (3200 IU/d, 80 µg/d) nor the lower dosage
(1600 IU/d; 40 µg/d) decreased cancer risk (HR 0.91 [95%
CI; 0.36-2.32] and HR 1.61 [95% CI; 0.72-3.59 respectively).
In the WHI, vitamin D plus calcium supplementation did not
alter the risk of colorectal cancer (282) (HR 1.07 [95% CI,
0.76-1.52]); invasive breast cancer (283) (HR 0.93 [95% CI,
0.77-1.12]), or in situ ductal breast cancer (284) (HR 0.81
[95% CI, 0.62-1.06]). In the Sister observational study
(274) use of vitamin D supplements did not decrease the
risk of breast cancer in obese women whose sisters had
breast cancer (HR 0.94 [95% CI, 0.82-1.10). Baseline
25(OH)D levels for individuals with obesity were not avail-
able for all these trials, but in the VITAL trial, the average
baseline level in those with a BMI higher than 30 kg/m²
was 28 ng/mL (72 nmol/L).
The ViDA trial, performed in New Zealand, reported no
beneficial effect of vitamin D on the risk of developing respira-
tory infections in adults with obesity. The baseline 25(OH)D
levels of participants with obesity were not presented.
The systematic review did not identify studies that reported
the risk of developing hypercalcemia in patients with obesity
receiving vitamin D. The risk of nephrolithiasis and a decline
in kidney function were examined in 1 RCT each, and no stat-
istically significant effect of vitamin D supplementation was
observed (120, 126).
Page 10
Downloaded from https://academic.oup.com/jcem/article/109/8/1907/7685305 by guest on 02 April 2025
1936
Other Evidence-to-Decision Criteria and
Considerations
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 8
Considerations related to required resources (costs), accept-
ability, and feasibility have been discussed. The panel did
not identify any studies that adequately addressed the cost-
effectiveness, or the potential equity impact of 25(OH)D
screening, for people with obesity. However, since obesity
has been associated with reduced health equity, if optimizing
25(OH)D levels were to preferentially improve outcomes in
persons with obesity and low 25(OH)D, then 25(OH)D
screening in those with obesity could possibly improve health
equity. The panel judged that screening would be acceptable
to most, assuming that a net benefit is expected. Although
there is variability in the cost and availability of testing for
25(OH)D levels across the globe, the panel judged that this
is feasible in many settings, as is the resultant intervention of
taking a daily nonprescription supplement. Given the very
high prevalence of obesity in many countries (eg, the preva-
lence of obesity in the United States is estimated to be
41.9% (285), a 25(OH)D screening strategy for all individuals
with obesity would require significant effort and resources,
which may not be feasible from a societal perspective.
Justification for the Recommendation
The panel's conditional recommendation against routine
25(OH)D screening for those with obesity is related primarily
to the lack of clinical trials examining the benefit of 25(OH)D
screening in those with obesity and treating only those with a
25(OH)D level below a threshold. Moreover, subgroup ana-
lyses from available clinical trials did not clearly demonstrate
a net benefit of vitamin D in individuals with obesity as a
group. . The panel was also uncertain that any putative benefits
of screening would justify the additional burden and costs
of 25(OH)D testing, including health care visits (cost-
effectiveness); and whether implementation of universal
25(OH)D screening for those with obesity would be feasible
from a societal perspective. In addition, the panel was uncer-
tain about what 25(OH)D level would necessitate subsequent
vitamin D administration.
Research Considerations
1. Large RCTs in participants with obesity will be required
to determine if vitamin D lowers the risk of disease,
whether benefit is limited to those with low baseline levels
(and defining what these levels are), what target levels are
required for optimal disease prevention, and what dos-
ages are required to achieve these target levels/desired
outcomes. Although placebo-controlled vitamin D trials
may be viewed as unethical for participants known to
have low 25(OH)D levels, inclusion of several daily dos-
ages and targeting several levels of 25(OH)D would in-
form the dosages and target levels required for disease
prevention.
2. Clinical trials must be designed to be of sufficient duration
to address the outcomes being examined, considering the
natural history and pathophysiology of the diseases of
interest (eg, acute infectious diseases vs cancer).
Acknowledgments
The Endocrine Society and the Guideline Development Panel
thank Marie McDonnell, MD, who served as Clinical
Guidelines Committee chair during the development of this
clinical practice guideline (CPG), for the contributions she
made through her leadership and expertise. The panel thanks
Maureen Corrigan, MA, Director for Clinical Practice
Guidelines for the Endocrine Society, and Elizabeth York,
MPH, Manager of Clinical Practice Guidelines for the
Endocrine Society, for their expert guidance and assistance
with all aspects of guideline development. We thank the nu-
merous contributors from the Mayo Evidence-Based Practice
Center, especially Vishal Shah, MD, MPH, for their contribu-
tion in conducting the evidence reviews for the guideline. We
also thank John Sluyter for his contribution in extracting the
data for the ViDA participants younger than and older than
75 years of age. We also thank the American College of
Physicians for their identification of John Tayek as a primary
care representative for this guideline.
Disclaimer
The Endocrine Society's clinical practice guidelines are
developed to be of assistance to endocrinologists by providing
guidance and recommendations for particular areas of practice.
The guidelines should not be considered as an all-encompassing
approach to patient care and not inclusive of all proper ap-
proaches or methods, or exclusive of others. The guidelines
cannot guarantee any specific outcome, nor do they establish
a standard of care. The guidelines are not intended to dictate
the treatment of a particular patient. Treatment decisions
must be made based on the independent judgment of healthcare
providers and each patient's individual circumstances. The
Endocrine Society makes every effort to present accurate and
reliable information. This publication is provided "as is" and
the Society makes no warranty, express or implied, regarding
the accuracy and reliability of these guidelines and specifically
excludes any warranties of merchantability and fitness for a
particular use or purpose, title, or noninfringement of third-
party rights. The Society, its officers, directors, members, em-
ployees, and agents shall not be liable for direct, indirect, spe-
cial, incidental, or consequential damages, including the
interruption of business, loss of profits, or other monetary dam-
ages, regardless of whether such damages could have been fore-
seen or prevented, related to this publication or the use of or
reliance on the information contained herein.
Funding
Funding for the development of this guideline was provided by
The Endocrine Society. No other entity provided financial
support.
Disclosures
Summary
Total number of Guideline Development Panel members = 14
• Percentage of total Guideline Development Panel mem-
bers with relevant (or potentially relevant) conflicts of
interest = 7%
Data availability
The data underlying this article are available in the article, in
its online supplementary material, and in the accompanying
systematic review publication.
| ### Vitamin D for the Prevention of Disease: An Endocrine Society Clinical Practice Guideline
#### Background
Numerous studies demonstrate associations between serum concentrations of 25-hydroxyvitamin D (25[OH]D) and a variety of common disorders, including musculoskeletal, metabolic, cardiovascular, malignant, autoimmune, and infectious diseases. Although a causal link between serum 25(OH)D concentrations and many disorders has not been clearly established, these associations have led to widespread supplementation with vitamin D and increased laboratory testing for 25(OH)D in the general population. The benefit-risk ratio of this increase in vitamin D use is not clear, and the optimal vitamin D intake and the role of testing for 25(OH)D for disease prevention remain uncertain.
#### Objective
To develop clinical guidelines for the use of vitamin D (cholecalciferol [vitamin D3] or ergocalciferol [vitamin D2]) to lower the risk of disease in individuals without established indications for vitamin D treatment or 25(OH)D testing.
#### Methods
A multidisciplinary panel of clinical experts, along with experts in guideline methodology and systematic literature review, identified and prioritized 14 clinically relevant questions related to the use of vitamin D and 25(OH)D testing to lower the risk of disease. The panel prioritized randomized placebo-controlled trials in general populations (without an established indication for vitamin D treatment or 25[OH]D testing), evaluating the effects of empiric vitamin D administration throughout the lifespan, as well as in select conditions (pregnancy and prediabetes). The panel defined "empiric supplementation" as vitamin D intake that (a) exceeds the Dietary Reference Intakes (DRI) and (b) is implemented without testing for 25(OH)D. Systematic reviews queried electronic databases for publications related to these 14 clinical questions. The Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) methodology was used to assess the certainty of evidence and guide recommendations.
#### Results
The panel suggests empiric vitamin D supplementation for:
- **Children and adolescents aged 1 to 18 years** to prevent nutritional rickets and because of its potential to lower the risk of respiratory tract infections.
- **Adults aged 75 years and older** because of its potential to lower the risk of mortality.
- **Pregnant individuals** because of its potential to lower the risk of preeclampsia, intra-uterine mortality, preterm birth, small-for-gestational-age birth, and neonatal mortality.
- **Individuals with high-risk prediabetes** because of its potential to reduce progression to diabetes.
The panel suggests against empiric vitamin D supplementation above the current DRI to lower the risk of disease in healthy adults younger than 75 years. No clinical trial evidence was found to support routine screening for 25(OH)D in the general population, nor in those with obesity or dark complexion, and there was no clear evidence defining the optimal target level of 25(OH)D required for disease prevention in the populations considered; thus, the panel suggests against routine 25(OH)D testing in all populations considered.
#### Conclusion
The panel suggests empiric vitamin D for those aged 1 to 18 years and adults over 75 years of age, those who are pregnant, and those with high-risk prediabetes. Due to the scarcity of natural food sources rich in vitamin D, empiric supplementation can be achieved through a combination of fortified foods and supplements that contain vitamin D. Based on the absence of supportive clinical trial evidence, the panel suggests against routine 25(OH)D testing in the absence of established indications. These recommendations are not meant to replace the current DRIs for vitamin D, nor do they apply to people with established indications for vitamin D treatment or 25(OH)D testing. Further research is needed to determine optimal 25(OH)D levels for specific health benefits.
#### Key Words
Vitamin D, 25-hydroxyvitamin D, vitamin D deficiency, mortality, pregnancy, infection, prediabetes, clinical practice guidelines, systematic reviews.
---
### Recommendations by Population
#### Children and Adolescents (Ages 1-18)
- **Empiric Vitamin D Supplementation**: To prevent nutritional rickets and potentially lower the risk of respiratory tract infections.
- **Dosage**: In clinical trials, dosages ranged from 300 to 2000 IU (7.5 to 50 µg) daily equivalent, with an estimated weighted average of approximately 1200 IU (30 µg) per day.
#### Adults Aged 19-49
- **No Empiric Vitamin D Supplementation**: Follow the Institute of Medicine Recommended Daily Allowance (600 IU [15 µg] daily).
#### Adults Aged 50-74
- **No Empiric Vitamin D Supplementation**: Follow the Recommended Daily Allowance established by the IOM (600 IU [15 µg] daily for those aged 50 to 70 years; 800 IU [20 µg] daily for those older than 70 years).
- **No Routine 25(OH)D Testing**: The panel suggests against routine screening for 25(OH)D levels.
#### Adults Aged ≥75
- **Empiric Vitamin D Supplementation**: Because of the potential to lower the risk of mortality.
- **Dosage**: In clinical trials, dosages ranged from 400 to 3333 IU (10 to 83 µg) daily equivalent, with an estimated weighted average of approximately 900 IU (23 µg) daily.
- **No Routine 25(OH)D Testing**: The panel suggests against routine testing for 25(OH)D levels.
#### Pregnancy
- **Empiric Vitamin D Supplementation**: Given its potential to lower the risk of preeclampsia, intra-uterine mortality, preterm birth, small-for-gestational-age birth, and neonatal mortality.
- **Dosage**: In clinical trials, dosages ranged from 600 to 5000 IU (15 to 125 µg) daily equivalent, with an estimated weighted average of approximately 2500 IU (63 µg) per day.
- **No Routine 25(OH)D Testing**: The panel suggests against routine 25(OH)D testing.
#### Prediabetes
- **Empiric Vitamin D Supplementation**: In addition to lifestyle modification, to reduce the risk of progression to diabetes.
- **Dosage**: In clinical trials, dosages ranged from 842 to 7543 IU (21 to 189 µg) daily equivalent, with an estimated weighted average of approximately 3500 IU (88 µg) per day.
#### Adults with Dark Complexion
- **No Routine 25(OH)D Testing**: The panel suggests against routine screening for 25(OH)D levels.
#### Adults with Obesity
- **No Routine 25(OH)D Testing**: The panel suggests against routine screening for 25(OH)D levels.
---
### Additional Considerations
- **Daily vs Intermittent Dosing**: For nonpregnant people older than 50 years for whom vitamin D is indicated, the panel suggests supplementation via daily administration of vitamin D, rather than intermittent use of high doses.
- **Health Equity**: The panel judged that, in most situations, empiric vitamin D supplementation is inexpensive, feasible, acceptable to both healthy individuals and health care professionals, and has no negative effect on health equity.
---
### Research Considerations
Further research is needed to determine:
1. Optimal 25(OH)D levels for specific health benefits.
2. The efficacy of vitamin D supplementation in populations with low baseline 25(OH)D levels.
3. The long-term effects of vitamin D supplementation on chronic diseases such as cancer, diabetes, and cardiovascular disease.
4. The impact of vitamin D fortification on vitamin D status in different populations at risk of low vitamin D status.
---
### Conclusion
The panel suggests empiric vitamin D for specific populations based on the potential benefits and the lack of significant harm. However, routine 25(OH)D testing is not recommended in the absence of established indications. These guidelines aim to provide evidence-based recommendations for the use of vitamin D in disease prevention, while acknowledging the need for further research to refine these recommendations. |
# 维生素 Ｄ 及其类似物临床应用共识
自 ２０ 世纪初对佝偻病的研究发现维生素 Ｄ 以来ꎬ
维生素 Ｄ 与钙磷代谢和骨骼健康的重要关联被不断发
现ꎮ 我国内分泌学先驱刘士豪教授和朱宪彝教授针对
维生素 Ｄ 与钙磷代谢和骨软化的研究ꎬ 以及由他们提
出并命名的“ 肾性骨营养不良” 得到了国际学者的广
泛认可ꎮ 维生素 Ｄ 及其相关制剂 ( 或称类似物) 的应
用从根本上遏制了全球范围内佝偻病 / 骨软化症的广
泛流行趋势ꎮ 然而ꎬ 维生素 Ｄ 缺乏和营养不足在人群
中仍普遍存在ꎬ 全球约有超过 １０ 亿人群的血清 ２５ 羟维
生素 Ｄ (２５ ｈｙｄｒｏｘｙｖｉｔａｍｉｎ Ｄꎬ ２５ＯＨＤ) 水平达不到维持
骨骼肌肉健康所推荐的 ３０ μｇ / Ｌ (７５ ｎｍｏｌ / Ｌ) 水平ꎬ 因
此维生素 Ｄ 作为基本健康补充剂用于骨质疏松症的防
治ꎮ 维生素 Ｄ 在体内经过两步羟化后形成 １ꎬ２５ 双羟维
生素 Ｄ [ １ꎬ２５￣ｄｉｈｙｄｒｏｘｙｖｉｔａｍｉｎ Ｄꎬ １ꎬ２５(ＯＨ)２ Ｄ] ꎬ 后
者是体内维生素 Ｄ 的主要活性形式并发挥重要的生理
作用ꎬ 又被称为“ Ｄ 激素” 或“ 活性维生素 Ｄ”
ꎮ 近年
来ꎬ 有许多与维生素 Ｄ 结构相似且具有活性维生素 Ｄ
样作用的化学物质 ( 活性维生素 Ｄ 类似物) 被不断开
发并应用于临床ꎬ 特别是用于骨质疏松症、 佝偻病、 慢
性肾 脏 病 和 皮 肤 病 等 疾 病ꎮ 随 着 维 生 素 Ｄ 受 体
(ｖｉｔａｍｉｎ Ｄ ｒｅｃｅｐｔｏｒｓꎬ ＶＤＲ) 和 ２５ＯＨＤ￣１α￣羟化酶 (１α￣
ｈｙｄｒｏｘｙｌａｓｅꎬ ＣＹＰ２７Ｂ１) 在许多骨外组织中被发现ꎬ 维
生素 Ｄ 的作用已不再囿于调节钙磷代谢和维护骨骼健
康ꎬ 其在肌肉、 心血管疾病、 糖尿病、 癌症、 自身免
疫和炎性反应等中的作用也逐渐被关注ꎬ 维生素 Ｄ 已
成为临床及基础研究的热点ꎮ
１ 维生素 Ｄ 概述
学名: ９ꎬ １０￣开环胆甾￣ ５ꎬ ７ꎬ １０ (１９) ￣三烯￣３β￣醇] 和
植物来源 的 维 生 素 Ｄ２ [ 麦 角 固 醇ꎬ ｅｒｇｏｃａｌｃｉｆｅｒｏｌꎬ 化
学名: ９ꎬ １０￣开 环 麦 角 甾￣ ５ꎬ ７ꎬ １０ ( １９ ) ꎬ ２２￣四 烯￣
３β￣醇] ꎬ 化学结构见图 １ꎮ 维生素 Ｄ 在体内经 ２５ 羟化
酶的催化合成 ２５ＯＨＤꎬ 是体内的主要贮存形式ꎬ 反映
体内维生素 Ｄ 的营养状态ꎮ ２５ＯＨＤ 经过 １α 位羟化成为
１ꎬ２５(ＯＨ)２ Ｄꎬ 是体内维生素 Ｄ 的主要活性代谢物ꎬ 与
组织中广泛存在的维生素 Ｄ 受体结合ꎬ 发挥激素样作
用ꎬ 又称 Ｄ 激 素ꎮ 因 此ꎬ 维 生 素 Ｄ 亦 被 看 作 是 激 素
原[１]
ꎮ 维生素 Ｄ 及其代谢物的主要生理作用是促进钙
和磷在肠道 中 吸 收ꎬ 并 抑 制 甲 状 旁 腺 素 ( ｐａｒａｔｈｙｒｏｉｄ
ｈｏｒｍｏｎｅꎬ ＰＴＨ) 释放ꎬ 维持血钙和磷水平正常ꎬ 进而
保证骨骼健康和神经肌肉功能正常ꎮ 维生素 Ｄ 的骨骼
外作用 包 括 对 肌 肉、 心 血 管、 代 谢、 免 疫、 肿 瘤 发
生、 妊娠和胎儿发育等多方面的影响[２]
ꎮ
维生素 Ｄ 的发现源自 １９２０ 至 １９３０ 年对佝偻病的
研究ꎬ 维生素 Ｄ 缺乏、 代谢异常或过量主要影响骨代
谢和钙、 磷稳态ꎬ Ｄ 激素属于“ 钙调激素” 之一
ꎮ 严
重维生素 Ｄ 缺乏和代谢异常会导致佝偻病 / 骨软化症ꎻ
维生素 Ｄ 是一种脂溶性的开环固醇类物质ꎬ 包括
动物来源的维生素 Ｄ３ [ 胆骨化醇ꎬ ｃｈｏｌｅｃａｌｃｉｆｅｒｏｌꎬ 化
图 １ 维生素 Ｄ 的化学结构
Ｖｏｌ ９ Ｎｏ ２ １２７协 和 医 学 杂 志
维生素 Ｄ 不足与骨质疏松症及其骨折密切相关ꎮ 推荐
维持 骨 骼 健 康 的 循 环 ２５ＯＨＤ 水 平 应 达 到 ３０ μｇ / Ｌ
(７５ ｎｍｏｌ / Ｌ) 以上ꎮ 同时ꎬ 维生素 Ｄ 缺 乏 和 作 用 不 足
还与多种疾病的发生发展相关联[３]
ꎮ 老年人可能存在
维生素 Ｄ 营养缺乏、 活性维生素 Ｄ 的生成减少和作用
不足ꎬ 并容易并发肌少症、 虚弱症和跌倒风险增加ꎮ
随着社会 经 济 发 展 和 生 活 方 式 变 化ꎬ 特 别 是 户
外生活、 工作时 间 的 减 少ꎬ 维 生 素 Ｄ 缺 乏 已 经 成 为
全球性的公共 健 康 问 题[ ４]
ꎮ 流 行 病 学 资 料 表 明 维 生
素 Ｄ 缺乏在我国人群中普遍存在[ ５￣ ６]
ꎮ 近年来ꎬ 维生
素 Ｄ 与骨骼健康及多种疾病的联系受到了广泛重视ꎬ
维生素 Ｄ 检测、 补充和活性维生 素 Ｄ 及 其 类 似 物 使
用日趋频 繁ꎬ 但 所 采 用 的 制 剂 和 方 法 各 异ꎮ 为 指 导
公众科学获得 充 足 的 维 生 素 Ｄ 营 养、 合 理 补 充 维 生
素 Ｄ 和规范使用维生素 Ｄ 类似 物ꎬ 中 华 医 学 会 骨 质
疏松和骨矿 盐 疾 病 分 会 组 织 本 领 域 的 相 关 专 家ꎬ 著
此共识ꎬ 期作指引ꎮ
２ 维生素 Ｄ 代谢及其生理作用
为限速反应ꎬ 主要受 ＰＴＨ 的调控ꎬ ＰＴＨ 刺激 １α￣羟化
酶的合 成[７ꎬ９]
ꎮ １ꎬ２５(ＯＨ)２ Ｄ 被 ＤＢＰ 运 输 到 靶 器 官 组
织ꎬ 如肠道、 肾脏和骨骼ꎬ 与这些组织细胞内的 ＶＤＲ
结合后ꎬ 上调或下调靶基因的转录ꎬ 从而发挥其经典作
用ꎬ 包括促进肠道内钙和磷的吸收ꎬ 以及促进肾小管内
钙 的 重 吸 收[７ꎬ９ꎬ１１]
ꎬ 从 而 有 利 于 骨 骼 矿 化ꎮ 此 外ꎬ
１ꎬ２５(ＯＨ)２ Ｄ还直接作用于成骨细胞ꎬ 并通过成骨细胞
间接作用于破骨细胞ꎬ 从而影响骨形成和骨吸收ꎬ 并维
持骨组织与血液循环中钙、 磷的平衡[１２]
ꎮ
ＶＤＲ 除存在于肠道、 肾脏和骨骼以外ꎬ 还存在于
许多其他组织ꎬ １ꎬ２５( ＯＨ) ２ Ｄ作用于这 些 组 织 细 胞 内
的 ＶＤＲ 后ꎬ 发挥许多非经 典 作 用ꎬ 包 括 抑 制 细 胞 增
生、 刺激 细 胞 分 化、 抑 制 血 管 生 成、 刺 激 胰 岛 素 合
成、 抑制肾素合成、 刺激巨噬细胞内抑菌肽合成、 抑
制 ＰＴＨ 合成和促进骨骼肌细胞钙离子内流等[１ꎬ１３￣ １５]
ꎮ
１ꎬ２５(ＯＨ)２ Ｄ在发挥这些作用的同时ꎬ 还激活靶细胞
内的维生素 Ｄ￣ ２４￣羟化酶 (２４￣ｈｙｄｒｏｘｙｌａｓｅꎬ ＣＹＰ２４Ａ１) ꎬ
使 ２５ＯＨＤ 和１ꎬ２５( ＯＨ) ２ Ｄ 转 变 为 无 活 性 的 代 谢 产 物ꎬ
这是１ꎬ２５( ＯＨ) ２ Ｄ 的 一 种 重 要 自 身 调 节 机 制[７ꎬ１６]
ꎬ 其
目的是防止１ꎬ２５( ＯＨ) ２ Ｄ在靶细胞内 的 作 用 过 强ꎮ 此
外ꎬ一些 肾 外 组 织 也 具 有 产 生 １ꎬ２５( ＯＨ) ２ Ｄ 的 能 力ꎬ
这 些 组 织 表 达 ＣＹＰ２７Ｂ１ꎬ 能 将 ２５ＯＨＤ 转 变 为
１ꎬ２５( ＯＨ) ２ Ｄꎬ 与来源于血液循环中的１ꎬ２５( ＯＨ) ２ Ｄ共
同调节局部组织细胞的功能[７ꎬ１４] ( 图 ２) ꎮ
３ 维生素 Ｄ 缺乏及其危险因素
人体维生 素 Ｄ 主 要 来 源 于 表 皮 中 的 ７￣脱 氢 胆 固
醇ꎬ 在表 皮 经 阳 光 中 的 紫 外 线 ( 波 长 ２９０ ~ ３１５ ｎｍ)
照射后转变为维生素 Ｄ３ 前体ꎬ 经温促作用转换为维
生素 Ｄ３
[７]
ꎮ 维生素 Ｄ 的另一来源是食物ꎬ 包括植物
性食物和动物性食物ꎬ 含维生素 Ｄ２ 或 Ｄ３ 的食物种类
很少ꎬ 植物性食物 ( 如受阳光照射后的蘑菇) 含有较
丰富的维生素 Ｄ２ ꎬ 而动物性食物 ( 如野生多 脂 肪 海
鱼) 含有较丰富的维生素 Ｄ３ ꎮ 与外源性维生素 Ｄ２ 或
Ｄ３ 相 比ꎬ 内 源 性 维 生 素 Ｄ３ 在 血 液 中 的 半 衰 期 更
长[８]
ꎮ 维生素 Ｄ２ 和 Ｄ３ 为无活性形式ꎬ 两者不能互相
转化ꎬ 统称为维生素 Ｄꎮ 维生素 Ｄ 需经两次羟化才能
转变为 １ꎬ２５( ＯＨ) ２ Ｄꎬ 成 为 具 有 生 物 活 性 的 Ｄ 激 素ꎮ
第一步羟化主要在肝脏完成ꎬ 维生素 Ｄ 通过维生素 Ｄ
结合蛋白 ( ｖｉｔａｍｉｎ Ｄ ｂｉｎｄｉｎｇ ｐｒｏｔｅｉｎꎬ ＤＢＰ) 的运输到
达肝 脏ꎬ 在 肝 细 胞 内 经 维 生 素 Ｄ￣２５￣羟 化 酶 ( ２５￣
ｈｙｄｒｏｘｙｌａｓｅꎬ ＣＹＰ２Ｒ１ 和 ＣＹＰ２７Ａ１ ) 催 化 转 变 为
２５ＯＨＤꎬ 该过程为非限速反应[７ꎬ９]
ꎮ 约 ８５％ ~ ９０％ 的
２５ＯＨＤ 在血液 循 环 中 与 ＤＢＰ 结 合ꎬ １０％ ~ １５％ 与 白
蛋白结合ꎬ 游离部分不足 １％ ꎮ 由于与白蛋白结合部
分容易解离ꎬ 与 游 离 部 分 一 起 被 称 为 生 物 可 利 用 的
２５ＯＨＤ[ １０]
ꎮ 第二步羟化主要在肾脏完成ꎬ 肾 小 球 滤
液中的 ２５ＯＨＤ 在 ＤＢＰ 协助下进入肾小管细胞ꎬ 在细
胞内 ＣＹＰ２７Ｂ１ 催化下ꎬ 转变为１ꎬ２５( ＯＨ) ２ Ｄꎬ 该过程
３ １ 维生素 Ｄ 缺乏的诊断标准
体内可检测 到 的 维 生 素 Ｄ 代 谢 物 约 有 ４０ 多 种ꎬ
其中 ２５ＯＨＤ 是循环中存在最多的代谢物ꎬ 可反映机
体维生素 Ｄ 的营养水平ꎮ 血清 ２５ＯＨＤ 水平检测已被
公认为反映维生素 Ｄ 状态的最合理指标ꎮ 目前国际、
国内多 数 机 构 和 专 家 认 为: 血 清 ２５ＯＨＤ < ２０ μｇ / Ｌ
(５０ ｎｍｏｌ / Ｌ) 为 维 生 素 Ｄ 缺 乏 ( ｄｅｆｉｃｉｅｎｃｙ ) ꎬ ２０ ~
３０ μｇ / Ｌ( ５０ ~ ７５ ｎｍｏｌ / Ｌ ) 为 维生素 Ｄ 不 足 ( ｉｎｓｕｆｆｉｃｉ￣
ｅｎｃｙ)ꎬ >３０ μｇ / Ｌ ( > ７５ ｎｍｏｌ / Ｌ) 为 维 生 素 Ｄ 充 足ꎬ <
１０ μｇ / Ｌ( < ２５ ｎｍｏｌ / Ｌ) 为严重缺乏[１７]
ꎮ 按照该标准ꎬ
全球维生 素 Ｄ 不 足 或 缺 乏 相 当 普 遍ꎬ 约 占 总 人 口 的
５０％ ~ ８０％ [１７￣ １８]
ꎮ 中国不同纬度城市的调查显示ꎬ 人
群普遍存在维生素 Ｄ 不足或缺乏[１９￣ ２１]
ꎮ 我国五大城市
１４３６名健康人群横断面研究显示: 血清 ２５ＯＨＤ 平均水
平 为 ( １９ ８７ ± ８ １４ ) μｇ / Ｌꎬ 其 中 ２５ＯＨＤ > ３０ μｇ / Ｌ、
２０ ~ ３０ μｇ / Ｌ及<２０ μｇ / Ｌ的比例分别为 １１ ７％、 ３１ ３％和
１２８ Ｍａｒｃｈꎬ ２０１８维生素 Ｄ 及其类似物临床应用共识
图 ２ 维生素 Ｄ 的代谢途径
５７ ０％ [６]
ꎮ 上海地区 ２５８８ 名２０ ~ ８９ 岁人群中维生素 Ｄ
不足者男、 女性分别为 ８４％和 ８９％ꎬ 维生素 Ｄ 缺乏者
分别为 ３０％和 ４６％ꎬ ２５ＯＨＤ<１０ μｇ / Ｌ (２５ ｎｍｏｌ / Ｌ) 分
别为 ２％和 ３ ６％ [２２]
ꎮ 对 １０ ０３８ 名 ４５ ~ ７５ 岁的兰州城
市居民检测ꎬ 发现维生素 Ｄ 缺乏人群占 ７５ ２％ [２３]
ꎮ
３ ２ 血清 ２５ＯＨＤ 测定方法
血清 ２５ＯＨＤ 浓度测定最早使用氚 ( ３ Ｈ) 标记的
２５ＯＨＤ 进行竞争性蛋白结合分析 ( ｃｏｍｐｅｔｉ￣ｔｉｖｅ ｐｒｏｔｅｉｎ
ｂｉｎｄｉｎｇ ａｓｓａｙꎬ ＣＰＢＡ) ꎬ 但 ３ Ｈ 标记繁杂ꎬ 且需 要 处 理
放射性 废 物 ( ３ Ｈ 的 半 衰 期 很 长) ꎬ 故 该 方 法 已 被 放
弃ꎮ 免疫层析法及酶联免疫法都是使用 ２５ＯＨＤ 的特
异性抗体 进 行 分 析ꎬ 前 者 步 骤 简 单ꎬ 可 以 用 于 快 速
筛查ꎻ 后者所获得的抗原￣抗体复合物 的 信 号 经 过 酶
联信号放 大 作 用ꎬ 比 前 者 提 高 了 灵 敏 度ꎮ 化 学 发 光
法 ( ｃｈｅｍｉｌｕｍｉｎｅｓｃｅｎｔ ｉｍｍｕｎｏａｓｓａｙꎬ ＣＬ) 本 质 上 也 属
于竞争性蛋白结合分析ꎬ 无放射性污染ꎬ 且仪器可自
动化分析ꎬ 避 免 了 手 工 操 作 误 差ꎬ 节 约 成 本ꎬ 国 内
较 广 泛 使 用 该 方 法ꎮ 液 相 色 谱 与 质 谱 串 联 分 析
(ｌｉｑｕｉｄ ｃｈｒｏｍａｔｏｇｒａｐｈｙ￣ｍａｓｓ ｓｐｅｃｔｒｏｍｅｔｒｙ / ｍａｓｓ ｓｐｅｃｔｒｏｍｅ￣
ｔｒｙꎬ ＬＣ￣ＭＳ / ＭＳ) 能够区分 ２５ＯＨＤ３ 、 ２５ＯＨＤ２ 及其他与
２５ＯＨＤ 分子结构相似的物质ꎬ 该检测方法的特异性最
高ꎬ 且 灵 敏 度 也 极 高ꎬ 可 以 认 为 是 ２５ＯＨＤ 检 测 的
“ 金” 标准ꎬ 但是检测耗费较大ꎮ
目前临床常用的 血 ２５ＯＨＤ 测 定 方 法 为 化 学 发 光
法ꎬ 该方 法 检 测 的 成 分 不 仅 包 括 ２５ＯＨＤ３ ꎬ 也 包 括
２５ＯＨＤ２ 及 ２５ＯＨＤ 代谢产物ꎬ 但不包括１ꎬ２５( ＯＨ) ２ Ｄꎬ
所检测的成分能够代表体内维生素 Ｄ 营养状态ꎮ
３ ３ 维生素 Ｄ 缺乏风险因素和筛查
维生素 Ｄ 缺乏与环境和遗传因素有关ꎮ 影响维生
素 Ｄ 水平的因素很多ꎬ 包括年龄、 肤色、 季节、 地理
纬度、 海拔、 日照时间、 着装习惯、 防晒措施、 饮食
习惯、 空气污染、 肥胖以及影响维生素 Ｄ 代谢的药物
等ꎮ 老年人皮肤维生素 Ｄ 合成量显著减少ꎬ 同等程度
日照合成维生素 Ｄ 的能力只有年轻人的 ３０％ [２４]
ꎬ 老
年女性维生素 Ｄ 缺乏的风险比男性高 １ ５ 倍[２５] ꎻ 黑色
素通过吸收紫外线降低皮肤维生素 Ｄ 合成ꎬ 故肤色黑
的人维生素 Ｄ 缺乏的风险更高[２６] ꎻ 而季节、 纬度和
海拔不同ꎬ 紫外线的照射时间与强度差异大[２７]
ꎬ 秋冬
季比春夏季维生素 Ｄ 合成少[２８] ꎻ 随着海拔升高ꎬ 皮
肤合成维生素 Ｄ 会相应的增加[２８]
ꎮ 另外ꎬ 着装习惯
和户外活动时间也影响皮肤维生素 Ｄ 的合成[２９￣ ３０]
ꎮ 使
用防晒霜可使皮肤合成维生素 Ｄ 的总量明显下降[３１]
ꎮ
食用多脂鱼可补充维生素 Ｄꎬ 且野生三文鱼维生素 Ｄ
的含量比人工饲养者高[２８]
ꎮ 多项研究显示ꎬ 城市对流
层臭氧量高于农村ꎬ 致城市居民维生素 Ｄ 缺乏风险明
显升 高[３２]
ꎬ 农 村 女 性 维 生 素 Ｄ 水 平 比 城 市 女 性
高[３３￣ ３４]
ꎮ 同时ꎬ 空气污染地区维生素 Ｄ 缺乏现象明显
高于无污染地区[３５￣ ３７]
ꎬ 若孕期暴露于污染空气ꎬ 可致
胎儿脐带血维生素 Ｄ 水平降低ꎬ 导致婴儿维生素 Ｄ 缺
乏[３８]
ꎮ 超重 和 肥 胖 与 维 生 素 Ｄ 缺 乏 风 险 存 在 相 关
性[３９]
ꎮ 长期使用某些药物ꎬ 如苯妥英钠、 苯巴比妥、
利福平等ꎬ 可加快体内维生素 Ｄ 分解代谢而致维生素
Ｄ 缺乏[４０]
ꎮ
此外ꎬ 遗传因素也是影响维生素 Ｄ 代谢的原因之
一[４１￣ ４３]
ꎮ 通过 不 同 人 种 或 者 地 区 的 研 究ꎬ 明 确 了 血
２５ＯＨＤ 水平的差异与维生素 Ｄ 结合蛋白、 ７ 脱羟胆固
醇还原酶 ( ７￣ｄｅｈｙｄｒｏｃｈｏｌｅｓｔｅｒｏｌ ｒｅｄｕｃ￣ｔａｓｅ) 和 ２５￣羟化
酶的编码基因 ( 上述蛋白质编码基因分别为 ＧＣ、 ＤＨ￣
ＣＲ７ 和 ＣＹＰ２Ｒ１ 基因) 的遗传变异具有高度相 关 性ꎮ
已发现影响 ２５ＯＨＤ 水平的 ３ 个重要风险位点ꎬ 即 ＧＣ
Ｖｏｌ ９ Ｎｏ ２ １２９协 和 医 学 杂 志
基因 ｒｓ２２８２６７９、 ＤＨＣＲ７ 基因 ｒｓ７９４４９２６ 和 ＣＹＰ２Ｒ１ 基
因 ｒｓ１０７４１６５７ 位点ꎻ 单个遗传变异位点可解释 ２５ＯＨＤ
水平群体间差异的 １％ ~ ４％ꎬ 携带上述 ３ 个风险基因
型位点的人群ꎬ 更容易出现 ２５ＯＨＤ 缺乏[４４￣ ４８]
ꎮ 此外ꎬ
药物基因组学研究提示上述基因的遗传变异影响普通
维生素 Ｄ 补充后的疗效[４５]
ꎮ
虽然维生 素 Ｄ 缺 乏 或 不 足 在 全 球 人 群 中 非 常 普
遍ꎬ 但并不主张在人群普遍筛查血 ２５ＯＨＤ 水平ꎬ 而推
荐在具有维生素 Ｄ 缺乏风险及需要维持合理维生素 Ｄ
营养状态的人群中进行血 ２５ＯＨＤ 水平的筛查 ( 表 １) ꎮ
４ 维生素 Ｄ 与佝偻病 / 骨软化症
表 １ 建议筛查血 ２５ＯＨＤ 水平的人群
建议筛查人群
特殊人群 妊娠和哺乳期女性
有跌倒史和 / 或有非创伤性骨折史的老年人
缺乏日照的人群 ( 室内工作、 夜间工作等)
肥胖儿童和成人 ( ＢＭＩ≥３０ ｋｇ / ｍ２ )
接受减重手术的人群
疾病状态 佝偻病
骨软化症
骨质疏松症
甲状旁腺功能亢进症
慢性肾脏病
肝功能衰竭
小肠吸收不良综合征
急 / 慢性腹泻、 脂肪泻
胰腺囊性纤维化
炎性反应性肠病 ( 克罗恩病、 溃疡性结肠炎)
放射性肠炎
淋巴瘤
形成肉芽肿的疾病
结节病
结核病
艾滋病
组织胞浆菌病
球孢子菌病
慢性铍中毒
某些药物 抗癫痫药物: 苯妥英钠、 苯巴比妥等
糖皮质激素: 泼尼松等
抗结核药物: 利福平等
唑类抗真菌药物: 酮康唑等
某些调脂药物: 考来烯胺 ( 消胆胺) 等
ＢＭＩ: 体质量指数
佝偻病 ( ｒｉｃｋｅｔｓ) / 骨软化症 ( ｏｓｔｅｏｍａｌａｃｉａ) 是生
长板软骨和类骨质矿化 障 碍 所 致 代 谢 性 骨 病[４９]
ꎮ １７
世纪中叶佝偻病在欧洲工业化国家广泛流行ꎬ一度成
为危害儿童生命的主要原因ꎬ 直到 ２０ 世纪初维生素 Ｄ
的发现ꎬ 才阐明了维生素 Ｄ 缺乏是佝偻病的病因[５０]
ꎮ
在儿童时期ꎬ 骨骺生长板尚未闭合ꎬ 生长板软骨矿化
障碍导致特征 性 的 骨 骼 畸 形ꎬ 称 为 佝 偻 病ꎻ 成 年 后ꎬ
生长板 已 闭 合ꎬ 骨 重 建 部 位 新 形 成 的 类 骨 质 矿 化 障
碍ꎬ 骨矿物质含量减少ꎬ 致使骨骼易于变形和发生骨
折ꎬ 称为骨软化症ꎮ 佝偻病多见于 ６ 个月至 ２ 岁的婴
幼儿ꎬ 可出现方颅ꎬ 手镯、 脚镯征ꎬ 肋串珠ꎬ 严重时
出现鸡胸、 Ｏ 型腿或 Ｘ 型腿ꎬ 生长缓慢ꎮ 骨软化症早
期症状不明显ꎬ 逐渐出现乏力、 骨痛、 行走困难ꎬ 严
重者出现四肢长骨、 肋骨、 骨盆和脊柱等处的病理性
骨折ꎬ 身高变矮ꎬ 甚至卧床不起[５１]
ꎮ 维生素 Ｄ 缺乏、
维生素 Ｄ 代谢异常及作用异常是佝偻病和骨软化症的
重要病因ꎮ
４ １ 维生素 Ｄ 缺乏性佝偻病 / 骨软化症
维生素 Ｄ 缺乏和 / 或钙缺乏可导致佝偻病 / 骨软化
症ꎬ 又称营养性佝偻病ꎮ 主要原因是患者缺少日照导致
皮肤维生素 Ｄ３ 合成不足ꎮ 其他原因尚有膳食中维生素
Ｄ 缺乏和消化道疾患致维生素 Ｄ 吸收不良ꎬ 如胃肠切
除、 小肠吸收不良ꎬ 肝胆疾病ꎬ 慢性胰腺炎等[７ꎬ１７]
ꎮ 通
常认为当血 ２５ＯＨＤ 水平低于 １０ μｇ / Ｌ (２５ ｎｍｏｌ / Ｌ) 时ꎬ
就可能引起维生素 Ｄ 缺乏性佝偻病 / 骨软化症[５０]
ꎮ 此
时ꎬ 体内１ꎬ２５( ＯＨ) ２ Ｄ 相 对 不 足ꎬ 肠 道 钙 吸 收 减 少ꎬ
血 液 中 的 离 子 钙 水 平 偏 低ꎬ 刺 激 甲 状 旁 腺 素 分
泌[７ꎬ１７ꎬ５２]
ꎬ 从而增加肠钙吸收及肾小管对钙的重吸收ꎬ
但升高的甲状旁腺素抑制了肾小管对磷的重吸收ꎬ 使
血磷水平下降ꎬ 骨骼矿化不良[５３]
ꎮ
４ ２ 维生素 Ｄ 代谢异常致佝偻病 / 骨软化症
１α￣羟化酶缺 陷: 常 见 于 慢 性 肾 功 能 不 全 所 致 肾
性骨营养不良ꎬ 另外可见于假性维生素 Ｄ 缺乏性佝偻
病 Ⅰ 型 (ｐｓｅｕｄｏ￣ｖｉｔａｍｉｎ Ｄ ｄｅｆｉｃｉｅｎｃｙ ｒｉｃｋｅｔｓ ｔｙｐｅⅠ ꎬ ＰＤＤＲ
Ⅰ 型)ꎬ 亦称维生素 Ｄ 依赖性佝偻病 Ⅰ 型 ( ｖｉｔａｍｉｎ Ｄ￣ｄｅ￣
ｐｅｎｄａｎｔ ｒｉｃｋｅｔｓ ｔｙｐｅⅠ ꎬ ＶＤＤＲⅠ 型)ꎬ 为常染色体隐性遗
传疾病[５４]
ꎮ 因编码 １α￣羟化酶的 ＣＹＰ２７Ｂ１ 基因突变ꎬ
使酶功能缺陷ꎬ 导致１ꎬ２５( ＯＨ) ２ Ｄ合成减少ꎬ 肠道钙、
磷吸收减少ꎬ 出现低钙血症、 继发性甲状旁腺功能亢
进、 轻微的 低 磷 血 症ꎮ 血 清１ꎬ２５( ＯＨ) ２ Ｄ 显 著 降 低 是
１３０ Ｍａｒｃｈꎬ ２０１８维生素 Ｄ 及其类似物临床应用共识
该病的特征性生化改变[５５]
ꎮ 活性维生素 Ｄ 对该病具
有良好的治疗效果ꎮ
２５￣羟化酶缺乏: 主要见于严重的肝功能损伤、 药
物诱导 ２５￣羟 化 酶 缺 乏 和 遗 传 性 ２５￣羟 化 酶 缺 乏ꎬ 使
２５ＯＨＤ 生成障碍ꎬ 导致佝偻病 / 骨软化症[５６]
ꎮ
４ ３ 维生素 Ｄ 作用异常致佝偻病 / 骨软化症
维生素 Ｄ 依赖性佝偻病Ⅱ型 ( ｖｉｔａｍｉｎ Ｄ￣ｄｅｐｅｎｄａｎｔ
ｒｉｃｋｅｔｓ ｔｙｐｅ Ⅱꎬ ＶＤＤＲ Ⅱ型) ꎬ 又称遗传性维生素 Ｄ 抵
抗性佝偻病ꎬ 为常染色体隐性遗传疾病ꎬ 因编码维生
素 Ｄ 受体的 ＶＤＲ 基因突变ꎬ 导致１ꎬ２５( ＯＨ) ２ Ｄ不能发
挥正常 的 生 理 功 能ꎮ 血 液 中 １ꎬ２５( ＯＨ) ２ Ｄ 显 著 升 高ꎬ
约有 ２ / ３ 患者并发禿发[５７]
ꎮ
４ ４ 低血磷性佝偻病 / 骨软化症的维生素 Ｄ 代谢异常
成纤维 细 胞 生 长 因 子 ２３ ( ｆｉｂｒｏｂｌａｓｔ ｇｒｏｗｔｈ ｆａｃｔｏｒ
２３ꎬ ＦＧＦ２３) 相关低血磷性佝偻病 / 骨软化症包括遗传
性低血磷性佝偻病[５８] 和肿瘤相关的骨软化症[５９]
ꎮ 此
类疾病患者血 液 循 环 中 高 水 平 的 ＦＧＦ２３ 抑 制 肾 小 管
钠磷共转运 蛋 白 的 表 达 和 功 能ꎬ 使 尿 磷 排 出 增 加ꎬ
血磷下 降ꎮ 同 时 ＦＧＦ２３ 抑 制 １α￣羟 化 酶 的 合 成 与 活
性ꎬ 使血１ꎬ２５( ＯＨ) ２ Ｄ水平不适当降低ꎬ 肠道钙磷吸
收减少ꎬ 加 重 低 磷 血 症ꎬ 导 致 矿 化 障 碍ꎬ 发 生 佝 偻
病 / 骨软化症[ ５８]
ꎮ
５ 维生素 Ｄ 与骨质疏松症
等ꎬ 影 响 骨 构 建、 重 建 和 矿 化[６４]
ꎮ 此 外ꎬ 维 生 素 Ｄ
还调节骨骼肌细胞的增生、 分化、 肌管的大小ꎬ 对肌
肉量与肌功能发挥重要影响[６５￣ ６６]
ꎮ
骨质疏松症的发生ꎬ 取决于年轻时获得的峰值骨
量和中老年阶段的骨丢失速率ꎮ 研究表明ꎬ Ｄ 激素是
调节骨骼生长发育的重要内分泌激素ꎬ 青少年阶段ꎬ
合成及摄取足量的维生素 Ｄꎬ 能够促进骨骼构建与矿
化ꎬ 有助于获得较高的峰值骨量[６７]
ꎮ 此 后ꎬ 充 足 的
维生素 Ｄ 帮助 维 持 正 钙 平 衡ꎬ 减 少 骨 转 换 失 衡 和 骨
丢失加速ꎮ 多项 研 究 显 示 维 生 素 Ｄ 缺 乏 与 中 老 年 人
甲状旁腺 素 水 平 增 高、 骨 吸 收 增 加、 骨 量 丢 失、 跌
倒和骨折风 险 升 高 相 关[ ４８ꎬ６８￣ ６９]
ꎮ 随 机 安 慰 剂 对 照 研
究显示ꎬ 补充维生素 Ｄ ８００ ~ １０００ ＩＵ / ｄꎬ 能够降低骨
转换水 平ꎬ 减 少 骨 丢 失 率ꎬ 增 加 腰 椎 和 髋 部 骨 密
度[７０￣ ７１]
ꎮ 另外ꎬ 跌倒是骨质疏松性骨折的主要诱因ꎬ
我国 ７０ 岁以上人群ꎬ 跌倒的发生率超过 ２０％ [７２]
ꎬ 每
天补充 ７００ ~ １０００ ＩＵ 的维生素 Ｄꎬ 可明显降低老年人
群跌倒发生[７３￣ ７４]
ꎮ Ｍｅｔａ 分析表明ꎬ 维 生 素 Ｄ 及 钙 剂
联合治疗能够降低老年人群骨折风险[７５]
ꎮ
活性维生素 Ｄ 及其类似物在骨质疏松症治疗中也
发挥重要作用ꎮ 临 床 上 常 用 阿 法 骨 化 醇 ( １α 羟 维 生
素 Ｄ) 或骨化三醇 [ １ꎬ２５( ＯＨ) ２ 维生素 Ｄ] ꎮ 由于老
年人群皮 肤 合 成 维 生 素 Ｄ 能 力 下 降ꎬ 肾 脏 对 ２５ＯＨＤ
的 １α 羟化能力减弱ꎬ 活性维生素 Ｄ 尤其适用于老年
患者或有肾脏疾病的患者ꎮ 研究显示ꎬ 骨化三醇和阿
法骨化醇能够改善肌肉功能和平衡能力ꎬ 降低老年人
跌倒风险[７６￣ ７７]
ꎮ 老年人群的系统综述显示ꎬ 骨化三醇
和阿法骨化醇能够降低骨吸收生化指标水平ꎬ 增加骨
密度[７８]
ꎮ 我国研究 也 显 示ꎬ 骨 化 三 醇 联 合 碳 酸 钙 治
疗ꎬ 明显增加腰 椎 和 股 骨 颈 骨 密 度[７９]
ꎮ 也 有 研 究 表
明ꎬ 骨化三醇和阿法骨化醇能够降低跌倒和非椎体骨
折的风险[７８]
ꎮ
骨质疏 松 症 是 以 骨 密 度 降 低、 骨 组 织 微 结 构 损
伤ꎬ 导致骨强度下降、 骨折危险性增加为特征的全身
性骨骼疾病[６０￣ ６１]
ꎮ 随着增龄ꎬ 骨质疏松性骨折风险显
著增加ꎬ 同 时 易 伴 发 肌 少 症ꎬ 使 患 者 的 生 活 质 量 下
降ꎬ 甚至死亡风险增加 [６２￣ ６３]
ꎮ 由于维生素 Ｄ 对骨骼
和肌肉均发挥着重要的调节作用ꎬ 其在骨质疏松症的
发生、 发展和防治中的作用不容忽视ꎮ
１ꎬ２５( ＯＨ) ２ Ｄ是重要的钙调 节 激 素 之 一
ꎬ 增 加 肠
道及肾脏钙 吸 收ꎬ 促 进 正 钙 平 衡ꎮ 维 生 素 Ｄ 可 通 过
升高血钙水 平 或 直 接 作 用 于 甲 状 旁 腺ꎬ 抑 制 甲 状 旁
腺素分泌ꎬ 减少继发性甲状旁腺功能亢进症的发生ꎬ
进而减轻后者引起的过度骨吸收[ ６￣ ７]
ꎮ 此外ꎬ 维生素
Ｄ 通过结合于 成 骨 细 胞 和 骨 细 胞 核 的 ＶＤＲꎬ 作 用 于
维生素 Ｄ 反 应 元 件ꎬ 能 够 调 节 多 种 基 因 的 表 达ꎬ 包
括骨钙素、 骨形态发 生 蛋 白、 ＦＧＦ￣ ２３、 同 源 Ｘ 染 色
体连锁的 磷 酸 盐 调 节 内 肽 酶 ( ｐｈｏｓｐｈａｔｅ ｒｅｇｕｌａｔｉｎｇ ｅｎ￣
ｄｏｐｅｐｔｉｄａｓｅ ｈｏｍｏｌｏｇ Ｘ￣ｌｉｎｋｅｄꎬ ＰＨＥＸ) 、 低 密 度 脂 蛋 白
相关 蛋 白￣ ５ ( ＬＤＬ ｒｅｃｅｐｔｏｒ ｒｅｌａｔｅｄ ｐｒｏｔｅｉｎ ５ꎬ ＬＲＰ￣ ５)
６ 维生素 Ｄ 骨骼外作用
６ １ 维生素 Ｄ 与 ２ 型糖尿病
人群研究表明维生素 Ｄ 不足与 ２ 型糖尿病 ( ｔｙｐｅ
２ ｄｉａｂｅｔｅｓ ｍｅｌｌｉｔｕｓꎬ Ｔ２ＤＭ) 发生率增加有关[８０￣ ８１]
ꎬ 维
生素 Ｄ 缺乏是 Ｔ２ＤＭ 患病的潜在危险因素[８２￣ ８３]
ꎮ 横断
面研究提示在 Ｔ２ＤＭ 和代谢综合征人群中ꎬ 维生素 Ｄ
水平与胰岛素敏感性及胰岛 β 细胞功能具有独立相关
性[８４￣ ８５]
ꎮ 具有较高 ２５ＯＨＤ 浓度者ꎬ 空腹及糖负荷后
２ ｈ血糖水平均较低[８６]
ꎮ 纵向队列研究和 Ｍｅｔａ 分析均
Ｖｏｌ ９ Ｎｏ ２ １３１协 和 医 学 杂 志
显示较高维生素 Ｄ 摄入与 Ｔ２ＤＭ 发生风险降低存在关
联性[８７￣ ８８]
ꎮ 观察性研究也揭示维生素 Ｄ 缺乏与糖尿病
慢性并发症ꎬ 如糖尿病性视网膜病变的发生有关[８９]
ꎮ
然而ꎬ 目前对于补充维生素 Ｄ 是否能降低或预防
Ｔ２ＤＭ 的发生并无肯定结果ꎮ 多项双盲随机对照药物
试验研究表明维生素 Ｄ 缺乏的糖尿病前期患者ꎬ 短期
或长期甚至是大剂量补充维生素 Ｄ 均不能降低其进展
为 Ｔ２ＤＭ 的风险[９０￣ ９２]
ꎮ
有关维生素 Ｄ 糖调节作用的机制包括: 通过直接
激活 ＶＤＲ 或通过干扰胰岛素受体启动基因区域的维生
素 Ｄ 反应元件影响胰岛 β 细胞的功能[９３￣ ９４] ꎻ 通过增强
胰岛素受体与胰岛素反应ꎬ 改善胰岛素敏感性和葡萄
糖转运[９５￣ ９７] ꎻ 诱导胰岛素原向胰岛素转换增加ꎮ 此外
维生素 Ｄ 也 可 通 过 间 接 调 节 钙 稳 态 影 响 疾 病 的 进
展[９８]
ꎮ 最新研究表明活性维生素 Ｄ 通过下丘脑室旁
核 ＶＤＲꎬ 起到减少摄食、 降低体重、 改善糖耐量和胰
岛素敏感性的作用[９９]
ꎮ
６ ２ 维生素 Ｄ 与心血管疾病
心脏及血管平滑肌中表达 ＶＤＲ 及１α￣羟化酶和 ２４￣
羟化酶[１００]
ꎬ 维生素 Ｄ 可能通过影响或调节炎性反应
细胞因子[１０１]
、 血管钙化[１０２]
、 肾素￣血管紧张素￣醛固
酮系统[１０３] 参与心血管保护作用ꎮ 维生素 Ｄ 可通过抑
制 ＰＴＨ 的分泌 ( ＰＴＨ 可降低脂肪分解) [１０４] 和增加血
钙水平 ( 钙会降低肝脏三酰甘油的形成和分泌) [１０５] 等
途径改善血脂代谢ꎮ
低维生素 Ｄ 水平与动脉粥样硬化[１０６￣ １０７]
、 冠状动脉
疾病、 心肌梗死[１０８]
、 心力衰竭[１０９]
、 卒中[１１０]
、 心血管
病死率[１１１] 和全因病死率等相关ꎬ 是心血管疾病的独立
危险因素[１１２]
ꎮ 而补充维生素 Ｄ 对于血压、 人群总体的
全因病死率及心血管病发病率和病死率的影响存在争
议ꎻ 少量随机对照临床试验表明补充维生素 Ｄ 可以改
善心衰患者的预后[１１３]
ꎬ 但需要进一步确定维生素 Ｄ 缺
乏与冠心病、 心肌梗死之间是否存在因果联系ꎮ
６ ３ 维生素 Ｄ 与肌力和跌倒
肌力下降是跌倒的诱因之一
ꎬ 已有研究发现补充
维生素 Ｄ 可以改善肌力、 降低跌倒风险ꎮ 维生素 Ｄ 既
可通过 ＶＤＲ 调节靶基因的转录ꎬ 直接促进 肌 细 胞 发
育ꎬ 又可通过快速跨膜通路促进钙离子内流ꎬ 增强肌
肉收缩功能[６４ꎬ１１４￣ １１７]
ꎬ 也可调节血钙和磷水平间接影
响肌肉的功能[１１８￣ １２０]
ꎮ 观察性研究提示ꎬ 严重维生素
Ｄ 缺乏者肌力下降ꎬ 跌 倒 风 险 增 加[７２ꎬ１２１]
ꎮ 随 机 对 照
试验结果表明ꎬ 适量补充维生素 Ｄ 可以改善肌力、 降
低跌倒风 险ꎬ 尤 其 是 对 基 础 维 生 素 Ｄ 水 平 较 低 的 人
群ꎬ 若同时补充钙剂效果更显著[１２２]
ꎬ 但具体剂量和
疗程尚无定论ꎮ 老年人群跌倒发生率高且后果严重ꎬ
因此对老年人群补充适量的维生素 Ｄ 对改善肌力和减
少跌倒更具价值[１２０]
ꎮ
６ ４ 维生素 Ｄ 与免疫和肿瘤
１ꎬ２５( ＯＨ) ２ Ｄ能够抑制 Ｔ 淋巴细胞的活化和增生ꎬ
影响细胞因子的表达、 诱导单核细胞的分化、 增加中
性粒细胞和单核细胞的抗菌肽生成、 抑制树突状细胞
的成熟 和 分 化ꎮ 动 物 实 验 显 示１ꎬ２５( ＯＨ) ２ Ｄ 可 预 防 １
型糖尿 病、 自 身 免 疫 性 脑 脊 髓 炎￣多 发 性 硬 化 的 发
生[１２３￣ １２５]
ꎬ一项来 自 缺 乏 日 光 照 射 地 区 的 研 究 提 示ꎬ
婴儿期 补 充 维 生 素 Ｄ 能 降 低 今 后 １ 型 糖 尿 病 的 风
险[１２６]
ꎮ 补 充 维 生 素 Ｄ 有 助 于 缓 解 系 统 性 红 斑 狼
疮[１２７]
、 炎性反应性肠病的病情[１２８]
ꎮ 动物实验和体外
细胞培养研究均表明１ꎬ２５( ＯＨ) ２ Ｄ促进细胞分化和抑
制肿瘤细胞增生ꎬ 且具有抗炎、 促凋亡、 抑制血管生
成的特性[１２８]
ꎮ 多个对结、 直肠癌患者的观察性研究
表 明ꎬ 血 清 ２５ＯＨＤ 水 平 与 癌 症 发 生 率 呈 负 相
关[１２９￣ １３０]
ꎮ 但维生素 Ｄ 与肿瘤的相关性仍需要大规模
的随机对照研究证实ꎬ 特别是以防治肿瘤作为主要观
察终点的研究ꎮ
７ 维生素 Ｄ 临床应用
７ １ 维生素 Ｄ 缺乏
预防维生素 Ｄ 缺乏的一般措施: 增加日照和富含
维生素 Ｄ 食物的摄入是预防维生素 Ｄ 缺乏 / 不足的经
济有效 的 方 法ꎮ 通 常ꎬ 春、 夏 和 秋 季１１ ∶ ００ － １５ ∶ ００
将面部和双上臂暴露于阳光 ５ ~ ３０ ｍｉｎ ( 取决 于 多 因
素) ꎬ 每周 ３ 次即可达到预防目的[７ꎬ１３１￣ １３２]
ꎮ 缺少日照
时建议补充维生素 Ｄꎬ 维生素 Ｄ２ 或维生素 Ｄ３ 均可ꎬ
二者在疗效和安全性方面无显著差别[１３３]
ꎮ
对维生素 Ｄ 缺乏高危人群ꎬ 维生素 Ｄ 摄入量的推
荐见表 ２[１７]
ꎮ
表 ２ 维生素 Ｄ 缺乏高危人群维生素 Ｄ 补充推荐
年龄 建议补充剂量
( ＩＵ / ｄ) 年龄 可耐受摄入上限
( ＩＵ / ｄ)
０ ~ １ 岁 ４００ ~ １０００ ０ ~ ６ 个月 １０００
\>１ ~ １８ 岁 ６００ ~ １０００ >６ 个月~ １ 岁 １５００
\>１８ ~ ５０ 岁 １５００ ~ ２０００ >１ ~ ３ 岁 ２５００
\>５０ ~ ７０ 岁 １６００ ~ ２０００ >３ ~ ８ 岁 ３０００
\>７０ 岁 １６００ ~ ２０００ >８ 岁 ４０００
１３２ Ｍａｒｃｈꎬ ２０１８维生素 Ｄ 及其类似物临床应用共识
建 议 妊 娠 和 哺 乳 期 妇 女 补 充 维 生 素 Ｄ １５００ ~
２０００ ＩＵ / ｄꎬ 而具有维生素 Ｄ 缺乏高风险者可耐受上限
是 １０ ０００ ＩＵ / ｄꎻ 建议肥胖儿童和成人及用抗惊厥药、
糖皮质激素、 抗真菌药和抗艾滋病药物的儿童和成人
至少需要同年龄段 ２ ~ ３ 倍的维生素 Ｄ 方能满足需要ꎮ
维生素 Ｄ 缺乏的防治策略[１７]
: 对维生素 Ｄ 缺乏
的防治ꎬ 建议用普通维生素 Ｄ２ 或 Ｄ３ 制剂ꎮ 不建议单
次超大剂量补充维生素 Ｄ 的用法ꎬ 不推荐用活性维生
素 Ｄ 或其类似物纠正维生素 Ｄ 缺乏ꎮ
对 ０ ~ １ 岁维生素 Ｄ 缺乏婴幼儿建议用维生素 Ｄ２
或 Ｄ３ ２０００ ＩＵ / ｄ 或 ５０ ０００ ＩＵ / 周ꎬ 用 ６ 周 以 使 血 清
２５ＯＨＤ 水平达到 ３０ μｇ / Ｌ (７５ ｎｍｏｌ / Ｌ) 以上ꎬ 继而以
４００ ~ １０００ ＩＵ / ｄ 维持ꎻ 对>１ ~ １８ 岁的维生素 Ｄ 缺乏儿
童和 青 少 年ꎬ 建 议 用 维 生 素 Ｄ２ 或 Ｄ３ ２０００ ＩＵ / ｄ 或
５０ ０００ ＩＵ / 周ꎬ 用 ６ 周以使血清 ２５ＯＨＤ 水平达 ３０ μｇ / Ｌ
(７５ ｎｍｏｌ / Ｌ) 以上ꎬ 继而以 ６００ ~ １０００ ＩＵ / ｄ 维持ꎻ 对
维生素 Ｄ 缺 乏 所 有 成 年 人ꎬ 建 议 用 ５０ ０００ ＩＵ / 周 或
６０００ ＩＵ / ｄ的维生素 Ｄ２ 或 Ｄ３ ８ 周以使血清 ２５ＯＨＤ 水平
达 ３０ μｇ / Ｌ (７５ ｎｍｏｌ / Ｌ) 以上ꎬ 继而以１５００~ ２０００ ＩＵ / ｄ
维持ꎻ 对肥胖患者、 小肠吸收不良综合征患者和正在
使用影 响 维 生 素 Ｄ 代 谢 药 物 的 患 者ꎬ 建 议 用 高 剂 量
(常规剂量的 ２ ~ ３ 倍ꎬ 至少 ６０００ ~ １０ ０００ ＩＵ / ｄ) 的维生
素 Ｄ 治疗维生素 Ｄ 缺乏ꎬ 以达到血清 ２５ＯＨＤ 水平在
３０ μｇ / Ｌ (７５ ｎｍｏｌ / Ｌ) 以 上ꎬ 继 而 以 ３０００ ~ ６０００ ＩＵ / ｄ
维持ꎮ
在有“肾外” 产生 １ꎬ２５( ＯＨ)２ Ｄ 的疾病 ( 如结节
病、 结核病) 的患者ꎬ 用维生素 Ｄ 治疗期间ꎬ 建议监
测血清 ２５ＯＨＤ 水平和血、 尿钙水平ꎬ 以防止高钙血症ꎮ
对有原发性甲状旁腺功能亢进症并发维生素 Ｄ 缺
乏的患者ꎬ 建议酌情考虑维生素 Ｄ 治疗ꎬ 且建议监测
血清钙水平ꎮ
启动 维 生 素 Ｄ 治 疗 后 ３ ~ ６ 个 月ꎬ 再 检 测 血 清
２５ＯＨＤ 水平ꎬ 以判断疗效和调整剂量ꎮ
７ ２ 佝偻病 / 骨软化症
预防营养缺乏性佝偻病 / 骨软化症需保证足够的维
生素 Ｄ 与钙的营养ꎮ 充足日照是预防维生素 Ｄ 缺乏最
为安全和经济有效的办法[１３４]
ꎮ 缺乏日照时建议补充维
生素 Ｄ 预防维生素 Ｄ 缺乏ꎬ 补充剂量参见上节内容ꎮ
维生素 Ｄ 缺乏佝偻病 / 骨软化症患者给予充足的
普通维生素 Ｄ 和 钙 剂 干 预 后ꎬ 常 有 显 著 疗 效ꎮ 维 生
素 Ｄ 缺乏的治 疗 剂 量 参 见 上 节 内 容ꎮ 胃 肠 吸 收 不 良
的患者口服维 生 素 Ｄ 的 需 要 量 更 大ꎬ 或 采 用 肌 肉 注
射方 式ꎮ 建 议 将 ２５ＯＨＤ 水 平 至 少 提 高 到 ２０ μｇ / Ｌ
( ５０ ｎｍｏｌ / Ｌ) ꎬ 最好达到 ３０ μｇ / Ｌ (７５ ｎｍｏｌ / Ｌ) 以上ꎮ
应适当补充钙剂保证人体每日钙需要量ꎮ 可监测血钙
以及尿钙排出量ꎬ 以便调整维生素 Ｄ 剂量ꎮ
ＰＤＤＲ Ⅰ型或 ＶＤＤＲ Ⅰ型的治疗在活性维生素 Ｄ
及其类似物上市之前ꎬ 通常用普通维生素 Ｄ ２ 万~ １０
万 ＩＵ / ｄꎬ 但效果不好且易导致体 内 大 量 维 生 素 Ｄ 蓄
积ꎮ 目前常采用阿法骨化醇 ０ ５ ~ １ ５ μｇ / ｄ或骨化三
醇 ０ ５ ~ １ ０ μｇ / ｄ 治疗ꎬ 同时补充适量钙剂ꎮ 活性维
生素 Ｄ 治疗能使 ＰＤＤＲ Ⅰ型佝偻病痊愈ꎬ 患儿的生长
速度趋于正常[５５]
ꎮ ＶＤＤＲ Ⅱ型患者ꎬ 由于体内维生素
Ｄ 受体 抵 抗ꎬ 需 要 更 大 剂 量 的 阿 法 骨 化 醇 或 骨 化 三
醇ꎬ 甚至需要静脉补充钙剂维持血钙稳定[５７]
ꎮ
ＦＧＦ２３ 相关的低血磷性佝偻病 / 骨软化症的治 疗
需要补充磷和使用活性维生素 Ｄꎮ 采用中性磷酸盐溶
液补充磷ꎬ 骨化三醇剂量为儿童 ２０ ~ ３０ ｎｇ / ( ｋｇｄ) ꎬ
成人 ０ ５０ ~ ０ ７５ μｇ / ｄꎬ 分两次服用[１３５]
ꎮ 如使用阿法
骨化醇ꎬ 其剂量约为骨化三醇的 １ ５ 倍ꎮ 此类患者通
常无需补充钙ꎬ 除非存在显著的钙缺乏ꎮ 治疗中需注
意监测血钙磷、 尿钙磷和血甲状旁腺素的水平ꎮ
７ ３ 甲状旁腺功能减退症
甲状旁 腺 功 能 减 退 症 ( ｈｙｐｏｐａｒａｔｈｙｒｏｉｄｉｓｍꎬ 简 称
甲旁减) 是一种少见的内分泌疾病ꎬ 因甲状旁腺素产
生减少导致钙、 磷代谢异常ꎬ 以低钙血症、 高磷血症
伴 ＰＴＨ 水平降低或在不适当的正常范围内为特征ꎬ 临
床上可表现为手足搐搦、 癫痫发作ꎬ 可并发颅内钙化
及低钙性白内障等慢性并发症[１３６]
ꎮ 甲状腺手术是成
年起病甲旁减的最常见病因ꎬ 其经典治疗为长期口服
钙剂和维生素 Ｄ 制剂[１３６￣ １３８]
ꎮ
除了每日补充元素钙 １ ０ ~ ３ ０ ｇ ( 分次服用) 外ꎬ
维生素 Ｄ 及其类似物可促进肠道的钙吸收ꎬ 在甲旁减
的长期治疗中具有重要地位ꎮ 各种维生素 Ｄ 制剂在甲
旁减患者中的使用剂量如下[１３７￣ １３８]
:
骨 化 三 醇 [ １ꎬ２５(ＯＨ)２ Ｄ ]: 常 用 剂 量 为 ０ ２５ ~
２ ０ μｇ / ｄꎬ 但也有患者需要更大的剂量ꎮ 由于半衰期
短ꎬ 剂量超过 ０ ７５ μｇ / ｄ 时建议分次服用ꎻ 停药后作
用消失也较快 (２ ~ ３ ｄ) ꎮ
阿法骨化醇 [１α ( ＯＨ) Ｄ３ ] : 常用剂量为 ０ ５ ~
４ ０ μｇ / ｄꎬ 其升高血钙的作用弱于骨化三醇ꎬ 剂量大
约为骨化三醇的 １ ~ ２ 倍ꎬ 半衰期长于骨化三醇ꎬ 可每
日一次服用ꎻ 停药后作用消失约需 １ 周ꎮ
普通维生素 Ｄ ( 维生素 Ｄ２ 或 Ｄ３ ) : 由于 ＰＴＨ 作
用缺乏ꎬ 单独用于甲旁减治疗时需要很大的剂量ꎬ 且
不同患者间剂量变异范围较大ꎬ 治疗剂量 １ 万~ ２０ 万
Ｖｏｌ ９ Ｎｏ ２ １３３协 和 医 学 杂 志
Ｕ / ｄꎬ 维生素 Ｄ３ 作用或强于维生素 Ｄ２ ꎮ 普通维生素 Ｄ
半衰期长 (２ ~ ３ 周) ꎬ 使用剂量较大时可在人体脂肪
组织内蓄积ꎬ 停药后需要更长的时间 ( ２ 周~ ４ 个月)
才失效ꎬ 尤需警惕高钙血症的风险ꎮ 此外ꎬ 对于以活
性维生素 Ｄ 或 ＰＴＨ１￣ ８４ 为主要治疗方案的患者ꎬ 推荐
每日补充普通维生素 Ｄ ４００ ~ ８００ ＩＵꎬ 也可根据 血 清
２５ＯＨＤ 维生素 Ｄ 水平补充普通维生素 Ｄ 以避免维生
素 Ｄ 缺乏或不足[１３７ꎬ１３９]
ꎮ
双氢速变固醇 ( ｄｉｈｙｄｒｏｔａｃｈｙｓｔｅｒｏｌ) : 常用 治 疗 剂
量为 ０ ３ ~ １ ０ ｍｇ / ｄ ( 每日一次) ꎬ 停用后作用消失时
间约为 １ ~ ３ 周ꎮ 国内目前无此制剂ꎮ
钙剂和维生素 Ｄ 制剂的剂量应个体化ꎬ 必须定期
监测血钙磷水平以及尿钙排量ꎬ 治疗目标为维持血钙
水平轻度低于正常或位于正常低值范围ꎬ 同时避免高
钙尿症ꎮ 噻 嗪 类 利 尿 剂 可 以 促 进 肾 小 管 对 钙 的 重 吸
收ꎬ 减少尿钙的排出ꎬ 联合低盐饮食适用于尿钙水平
明显升高的患者ꎮ
７ ４ 骨质疏松症
普通维生 素 Ｄ 常 作 为 骨 骼 健 康 的 基 本 营 养 补 充
剂ꎬ 但补充普通维生素 Ｄ 在不同人群中增加骨密度、
降低骨 折 和 跌 倒 风 险 的 作 用 尚 存 争 议[７０￣ ７１ꎬ１２２ꎬ１４０￣ １４１]
ꎮ
建议骨质疏松症患者接受充足的阳光照射ꎬ 促进皮肤
合成内源性维生素 Ｄꎮ 日照不足者可每天补充 ６００ ~
１０００ ＩＵ的普通维生素 Ｄꎬ 也有研究表明对于维生素 Ｄ
缺乏患者每天补充更大剂量的维生素 Ｄ ( >２０００ ＩＵ / ｄ)ꎬ
可增加骨密度[６９]
ꎮ 建议定期监测患者血清 ２５ＯＨＤ 和
甲状旁腺素水平ꎬ 以指导调整普通维生素 Ｄ 的补充剂
量[１７ꎬ６０]
ꎬ 建议至少将血清 ２５ＯＨＤ 浓度调整到２０ μｇ / Ｌ
( ５０ ｎｍｏｌ / Ｌ) ꎬ 最好在 ３０ μｇ / Ｌ (７５ ｎｍｏｌ / Ｌ) 以上ꎬ 以
防止维生素 Ｄ 缺乏引发的继发性甲状旁腺功能亢进症
和骨密度 的 降 低[６ꎬ１７ꎬ１４２]
ꎮ 部 分 研 究 显 示ꎬ 每 天 补 充
７００ ~ １０００ ＩＵ普通 维 生 素 Ｄꎬ 能 够 降 低 跌 倒 风 险[７４]
ꎮ
就骨质疏松症的防治而言ꎬ 不建议患者常规单次补充
超大剂量 ( > ５００ ０００ ＩＵ / 年) 的普通维生素 Ｄꎬ 有研
究显示其可导致老年人跌倒风险升高[１４３￣ １４４]
ꎮ 若患者
血清 ２５ＯＨＤ 浓度超过１５０ μｇ / Ｌ (３７５ ｎｍｏｌ / Ｌ) 则可能
出现维生素 Ｄ 中毒[１７]
ꎮ
活性维生素 Ｄ 及其类似物是经过羟基化的维生素
Ｄ 类似物ꎬ 属于骨质疏松症的治疗药物ꎬ 推荐用于年
龄在 ６５ 岁以上或血清肌酐清除率小于 ６０ ｍｌ / ｍｉｎ 者ꎮ
临床应用的活性维生素 Ｄ 及其类似物包括骨化三醇和
阿法骨化醇等[１４５]
ꎮ 活性维生素 Ｄ 能够增加肠钙吸收ꎬ
减少继发性甲状旁腺功能亢进ꎬ 抑制骨吸收ꎬ 轻度增
加患者骨密度、 降低跌倒风险、 减少椎体或非椎体骨
折风险[７７￣ ７８]
ꎮ 活性维生素 Ｄ 可以与其他抗骨质疏松药
物联合使用ꎮ 建议骨质疏松症患者服用骨化三醇的剂
量通常为 ０ ２５ ~ ０ ５ μｇ / ｄꎬ 阿法骨化醇的剂量为 ０ ２５ ~
１ ０ μｇ / ｄꎮ 对于 明 显 缺 乏 维 生 素 Ｄ 的 骨 质 疏 松 症 患
者ꎬ 必要时可予普通维生素 Ｄ 以纠正维生素 Ｄ 的营养
缺乏ꎬ 同时给予活性维生素 Ｄ 以发挥其对骨质疏松症
的治疗作用ꎮ 需要注意的是ꎬ 使用活性维生素 Ｄ 的患
者ꎬ 不能根据血清 ２５ＯＨＤ 浓度调整药物剂量ꎬ 而可依
据血清 ＰＴＨ 水 平 及 骨 转 换 生 化 指 标ꎬ 评 估 药 物 的
疗效ꎮ
无论使用 普 通 维 生 素 Ｄꎬ 还 是 活 性 维 生 素 Ｄ 制
剂ꎬ 或者两者联合使用ꎬ 都建议定期监测患者血清钙
及 ２４ ｈ 尿钙浓度ꎬ 根据其水平调整药物剂量ꎬ 以避免
药物过量所引发的高钙血症或高尿钙的发生ꎬ 以保证
治疗的安全性[７８ꎬ１４５]
ꎮ
艾地骨化醇 ( ｅｌｄｅｃａｌｃｉｔｏｌꎬ ＥＤ￣ ７１) 是新型维生素
Ｄ 类似物ꎬ 其与１ꎬ２５( ＯＨ) ２ Ｄ相比ꎬ 血清半衰期更长ꎬ
抑制破骨细胞的活性更强ꎬ 使骨密度增加的幅度更明
显ꎬ 已在国外上市用于骨质疏松症防治[１４６￣ １４７]
ꎮ
７ ５ 慢性肾脏病￣矿物质和骨异常
维生素 Ｄ 代谢异常是慢性肾脏病￣矿物质 和 骨 异
常 ( ｃｈｒｏｎｉｃ ｋｉｄｎｅｙ ｄｉｓｅａｓｅ￣ｍｉｎｅｒａｌ ａｎｄ ｂｏｎｅ ｄｉｓｏｒｄｅｒꎬ
ＣＫＤ￣ＭＢＤ) 发生机制中的关键 环 节ꎮ 维 生 素 Ｄ 代 谢
异常及 ＣＫＤ￣ＭＢＤ 可发生于 ＣＫＤ 早期ꎬ 并贯穿于肾功
能减退全过程ꎬ 其与患者并发症及病死率增加密切相
关ꎮ 合理应用 活 性 维 生 素 Ｄ 及 其 类 似 物 有 助 于 治 疗
ＣＫＤ￣ＭＢＤꎬ 改善患者生存质量ꎮ
活性维生 素 Ｄ 及 其 类 似 物 在 ＣＫＤ￣ＭＢＤ 的 应 用:
活性维生素 Ｄ 及其类似物主要用于 ＣＫＤ 继发甲状旁
腺功能亢进症的治疗ꎮ 由于 ＣＫＤ￣ＭＢＤ 患者钙、 磷和
甲状旁腺素间关系复杂且相互影响[１４８]
ꎬ 因此应用活
性维生素 Ｄ 时ꎬ 应动态观察钙、 磷和 ＰＴＨ 变化ꎬ 并综
合判断[１４９]
ꎮ 目前常用于 ＣＫＤ￣ＭＢＤ 的 维 生 素 Ｄ 制 剂
主要有骨化三醇、 阿法骨化醇和帕立骨化醇 ( ｐａｒｉｃａｌ￣
ｃｉｔｏｌ) 等ꎬ 应用上述药物将血 ＰＴＨ、 钙、 磷等维持在
目标范围ꎮ ＣＫＤ ３ ~ ５期非透析患者适宜的 ＰＴＨ 水平目
前尚不清楚ꎬ 需对患者 ＰＴＨ 水平进行早期监测和动态
评估ꎬ 建议将升高的血磷降至接近正常范围ꎬ 且避免
高钙血症[１４９]
ꎮ ＣＫＤ ５Ｄ 期 (５ 期且已透析) 患者 ＰＴＨ
水平应维持于正常上限的 ２ ~ ９ 倍ꎮ
对 ＣＫＤ ３ ~ ５ 期非透析患者ꎬ 如 ＰＴＨ 水平高于正
常ꎬ 建议首先积极控制高血磷、 低血钙和维生素 Ｄ 缺
１３４ Ｍａｒｃｈꎬ ２０１８维生素 Ｄ 及其类似物临床应用共识
乏等因素ꎮ 对 ＣＫＤ ４ ~ ５ 期非透析患者ꎬ 当发生严重且
进展性甲状旁腺功能亢进时ꎬ 可使用骨化三醇或其类
似物[１４９]
ꎮ 近来研究显示帕立骨化醇对 ＣＫＤ 非透析患
者心血管事件无益ꎬ 且增加高血钙风险[１５０￣ １５１]
ꎮ 鉴于
维生素 Ｄ 类似物缺乏随机对照研究证明其在 ＣＫＤ 预
后中的获益ꎬ 且有增加高钙血症的风险ꎬ ＰＴＨ 水平轻
中度升高可能是机体的适应性反应ꎬ 故建议骨化三醇
或其类似物主要用于 ＣＫＤ ４ ~ ５ 期并发重度进展性继发
性甲状旁腺功能亢进者ꎮ
ＣＫＤ ３ ~ ５Ｄ 期患者使用活性维生素 Ｄ 或其类似物ꎬ
建议从小剂量开始ꎬ 如骨化三醇 ０ ２５ μｇ / ｄ 或阿法骨化
醇 ０ ２５ μｇ / ｄ 或帕立骨化醇 １ ０ μｇ / ｄꎬ 并根据 ＰＴＨ、
钙、 磷水平调 整 剂 量 ( 增 加 或 减 少 原 剂 量 的 ２５％ ~
５０％) ꎮ ＣＫＤ ５Ｄ 期患者ꎬ 如 ＰＴＨ 水平超过目标值或在
目标范围 内 进 行 性 升 高ꎬ 建 议 使 用 活 性 维 生 素 Ｄ 制
剂ꎬ 如骨化三醇 ０ ２５ ~ ０ ５ μｇ / ｄ 或阿法骨化醇 ０ ２５ ~
１ ０ μｇ / ｄ 或帕立骨化醇 １ ０ ~ ２ ０ μｇ / ｄꎮ 如使用活性
维生素 Ｄ 并 调 整 剂 量 后ꎬ ＰＴＨ 仍 超 过 目 标 值ꎬ 可 间
断使用较大剂 量 活 性 维 生 素 Ｄ 冲 击 治 疗ꎬ 如 骨 化 三
醇 ２ ０ ~ ４ ０ μｇ / 次ꎬ 每周 ２ ~ ３ 次ꎬ 并根据 ＰＴＨ 水平
调整剂量ꎮ
使用活性维生素 Ｄ 制剂治疗前后应监测患者钙磷
水平: ＣＫＤ ３ ~ ５ 期非透析患者ꎬ 使用活性维生素 Ｄ 制
剂后每个月监测血钙磷水平连续 ３ 个月ꎬ 以后每 ３ 个
月 １ 次ꎻ 每 ３ 个月监测 １ 次 ＰＴＨ 水平ꎮ ＣＫＤ ５Ｄ 期初
始或大剂量使用活性维生素 Ｄ 制剂者ꎬ 建议第 １ 个月
每 ２ 周监测 ｌ 次血钙、 磷水平ꎬ 以后每个月 １ 次ꎻ 全
段 ＰＴＨ 水平在开始 ３ 个月每个月监测 １ 次ꎬ 以后每 ３
个月 １ 次ꎮ 如 ＰＴＨ 水平低于正常上限的一半ꎬ 或出现
高钙、 高磷 血 症 时ꎬ 建 议 活 性 维 生 素 Ｄ 制 剂 减 量 或
停用ꎮ
活性维生素 Ｄ 与 ＣＫＤ￣骨质疏松: ＣＫＤ ３ ~ ５Ｄ 期患
者ꎬ 容易 并 发 骨 质 疏 松ꎬ 甚 至 骨 质 疏 松 性 骨 折ꎬ 髋
部、 股骨颈和桡骨远端低骨密度可预测 ＣＫＤ ３ ~ ５Ｄ 期
患者的骨折风险[１５２￣ １５３]
ꎬ 故建议 ＣＫＤ 患者重视骨密度
测定ꎬ 以帮助治疗决策ꎮ
并发骨质疏松和 / 或高骨折风险的 ＣＫＤ 患者ꎬ 可
考虑使用骨化三醇或其类似物ꎬ 但需结合 ＣＫＤ 分期ꎬ
并综合考虑血钙、 磷和 ＰＴＨ 水平: ＣＫＤ １ ~ ２期患者ꎬ
可参照普通人群ꎬ 给予钙剂联合骨化三醇或维生素 Ｄ
类似物ꎻ ＣＫＤ ３ ~ ５ 期非透析患者ꎬ 如 ＰＴＨ 在正常范
围ꎬ 应参照骨代谢状态或骨活检结果ꎬ 决定是否予活
性维生素 Ｄꎻ ＣＫＤ ３ ~ ５ 期非透析患者ꎬ 如果 ＰＴＨ 进行
性升高或高于正常值上限ꎬ 建议使用活性维生 素 Ｄꎻ
ＣＫＤ ５Ｄ 期患者ꎬ 根据 ＰＴＨ 水平ꎬ 调整活性维生素 Ｄ
的剂量及给药方法ꎬ 使 ＰＴＨ 水平达到目标值ꎮ
７ ６ 维生素 Ｄ 类似物在皮肤疾病中的应用
人工合成的维生素 Ｄ 类似物卡泊三醇 (ｃａｌｃｉｐｏｔｒｉｏｌ)
是一种选择性维生素 Ｄ 受体激动剂 ( ｖｉｔａｍｉｎ Ｄ ｒｅｃｅｐ￣
ｔｏｒ ａｇｏｎｉｓｔꎬ ＶＤＲＡ) ꎬ 与 ＶＤＲ 结合发挥一系列生物学
效应ꎬ 如控制炎性反应、 调节免疫应答、 抑制角质形
成细胞过度增生、 诱导表皮正常分化成熟等ꎮ 其外用
制剂被广泛应用于银屑病、 鱼鳞病、 掌跖角化病等皮
肤病的治疗ꎮ 卡泊三醇软膏一般用于头皮、 面部、 皮
肤皱褶处以外部位的皮损ꎬ 搽剂则主要用于头皮部位
病灶ꎮ 卡泊三醇治疗银屑病的使用方法是取软膏 / 搽
剂少量涂于患处皮肤ꎬ 早晚各一次ꎮ一般用药 ２ 周起
效ꎬ ６ ~ ８ 周疗效最佳ꎬ 可使半数以上寻常型银屑病患
者皮损完全消退或显著改善ꎮ 若患者单用卡泊三醇搽
剂ꎬ 则每周用量应少于 ６０ ｍｌꎮ 当患者单用软膏ꎬ 或同
时使用 软 膏 和 搽 剂 时ꎬ 每 周 卡 泊 三 醇 总 量 不 应 超 过
５ ｍｇꎬ 按 ０ ００５％ 浓 度 计 算ꎬ 即 １００ ｇ 卡 泊 三 醇 软 膏
(１ ｍｌ 卡泊三醇搽剂相当于 １ ｇ 软膏) ꎮ
安全性方面ꎬ 按照规范的方法ꎬ 在合适的部位外
用适当剂型、 剂量的卡泊三醇不会导致高钙血症ꎮ一
旦发生高钙血症ꎬ 停药 ３ ｄ 后即可缓解[１５４]
ꎮ 并发肾功
能不全、 与环磷酰胺合用易导致高钙血症[１５５￣ １５７]
ꎬ 绝
大多数成年患者ꎬ 每周外用 １００ ｇ 以内的软膏或搽剂
不会引起血钙升高ꎮ 卡泊三醇禁用于高钙血症者ꎮ
有研究提出孕妇及哺乳期妇女外用该药应控制在
３ ~ ４ 周、 每周 ２５ ~ ５０ ｇ 以内[１５８]
ꎮ 儿童外用卡泊三醇
的安全性目前尚未完全确定ꎬ 故儿童外用卡泊三醇时
应更为谨慎ꎮ 老年患者使用剂量可参照成人水平ꎮ
８ 维生素 Ｄ 的安全性
普通维生素 Ｄ 安全剂量范围宽ꎬ 人群中极少会长
期使用超过最大耐受剂量的维生素 Ｄꎬ 少有因普通维
生素 Ｄ 摄入过量导致中毒的报道[１７]
ꎮ 生理剂量补充
普通维生素 Ｄ 导致高钙血症的风险非常小ꎬ 不需常规
监测血钙及尿钙ꎮ 尿钙升高可能是监测维生素 Ｄ 过量
较为敏感的指标ꎬ一般认 为 ２４ ｈ 尿 钙 大 于 ７ ５ ｍｍｏｌ
(３００ ｍｇ) 为高钙尿症[１５９]
ꎮ 尿钙受多种因素的影响ꎬ
在服用维生素 Ｄ 的人群中ꎬ 不能简单地认为尿钙升高
就是维生素 Ｄ 中毒ꎬ 典型的维生素 Ｄ 中毒通常表现为
高血钙及其相关症状ꎬ 如烦渴、 多尿、 呕吐、 食欲下
Ｖｏｌ ９ Ｎｏ ２ １３５协 和 医 学 杂 志
降、 肾结石等[１６０￣ １６２]
ꎮ
通常可通过检测血清 ２５ＯＨＤ 浓度判断是否存在维
生素 Ｄ 中毒ꎮ 尽管不同研究间差异很大ꎬ 导致维生素 Ｄ
中毒 的 血 ２５ＯＨＤ 水 平 常 在 ２２４ μｇ / Ｌ ( ５６０ ｎｍｏｌ / Ｌ )
以上[１６３]
ꎬ 其对 应 的 维 生 素 Ｄ 补 充 剂 量 多 超 过 每 天
３０ ０００ ＩＵꎬ 且 应 用 时 间 较 长[１６４]
ꎮ 对 于 健 康 人 群ꎬ
２５ＯＨＤ 水平不宜超过 １５０ μｇ / Ｌ ( ３７５ ｎｍｏｌ / Ｌ) ꎬ 否则
中毒风险增加ꎮ
过量补充维生素 Ｄ 可能导致尿钙升高ꎬ 尿钙持续
超过 １０ ｍｍｏｌ / ｄ ( ４００ ｍｇ / ｄ) 可能增加肾结石和肾脏
钙盐沉着的风险[１６５]
ꎮ 然而ꎬ 由于普通维生素 Ｄ 的安
全剂量范围很广ꎬ 常规剂量补充普通维生素 Ｄ一般不
增加 肾 结 石 和 肾 钙 盐 沉 着、 进 而 损 害 肾 功 能 的 风
险[１６１]
ꎮ 同时常规剂量补充维生素 Ｄ 也不增加心脑血
管事件风险ꎬ 甚至可能有保护作用[１６６￣ １６７]
ꎮ 维生素 Ｄ
缺乏和过量都可能与血管钙化的发生相关ꎮ
活性维生 素 Ｄ 及 其 类 似 物 ( 骨 化 三 醇、 阿 法 骨
化醇和帕立骨 化 醇 等) 导 致 高 尿 钙 的 风 险 明 显 高 于
普通维生素 Ｄꎬ 特别是联合补充钙剂时[１６８￣ １６９]
ꎮ 活性
维生素 Ｄ 剂量越大ꎬ 发生高钙血症的风险越高ꎮ
活性维生素 Ｄ 的 半 衰 期 短ꎬ一 旦 发 现 用 药 期 间
出现高尿 钙 或 高 血 钙ꎬ 应 立 即 减 量 或 停 药ꎬ 特 别 需
要注意同时 减 少 钙 剂 和 含 钙 食 物 的 摄 入ꎬ 血 钙 水 平
多数能很快恢复[ １６８]
ꎮ 对于需要长期使用活性维生素
Ｄ 治疗的患者ꎬ 建议在启动治疗后的 １、 ３ 及 ６ 个 月
分别监测尿 钙 磷 及 血 钙 磷 水 平ꎬ 此 后 建 议 每 年 监 测
两次血钙 磷、 尿 钙 磷 及 肾 功 能ꎬ 以 确 定 长 期 治 疗 方
案的安 全 性[ １７０]
ꎮ 慢 性 肾 功 能 不 全 需 持 续 透 析 的 患
者ꎬ 无法测定尿 钙 磷ꎬ 使 用 活 性 维 生 素 Ｄ 期 间 需 动
态监测 血 ＰＴＨ、 血 钙、 血 磷 是 否 控 制 达 标ꎬ 并 每 年
监 测 异 位 钙 化 情 况ꎬ 根 据 结 果 及 时 调 整 药 物
剂量[ １７１￣ １７２]
ꎮ
综上所述ꎬ 维生素 Ｄ 总体安全性好ꎮ 使用常规剂
量普通维生素 Ｄ一般不需要监测血钙和尿钙ꎻ 在长期
使用活性维生素 Ｄ、 维生素 Ｄ 联合钙剂及其用于 ＣＫＤ
患者时ꎬ 则需要更加关注其安全性ꎮ
防治骨质疏 松 症 的 基 本 健 康 补 充 剂ꎮ 活 性 维 生 素 Ｄ
及其类似物 在 临 床 上 也 常 被 用 于 佝 偻 病 / 骨 软 化 症、
骨质疏松症、 甲旁减、 ＣＫＤ￣ＭＢＤ 和皮肤疾病等ꎮ 维
生素 Ｄ 及其类 似 物 的 使 用 需 要 注 意 其 安 全 性ꎬ 监 测
血、 尿钙水 平ꎬ 防 止 维 生 素 Ｄ 中 毒ꎮ 尽 管 维 生 素 Ｄ
对钙、 磷代谢 调 节 和 骨 骼 以 外 的 作 用 被 不 断 发 现ꎬ
但其在糖 尿 病、 肿 瘤、 免 疫 疾 病 和 感 染 性 疾 病 防 治
中所需的剂 量 和 效 果 尚 不 确 定ꎮ 随 着 未 来 研 究 的 深
入ꎬ 期待更多新 型 维 生 素 Ｄ 制 剂 和 新 的 药 物 适 应 证
被不断开发和应用ꎮ | # 维生素 D 及其类似物临床应用共识
自 20 世纪初对佝偻病的研究发现维生素 D 以来，维生素 D 与钙磷代谢和骨骼健康的重要关联被不断发现。我国内分泌学先驱刘士豪教授和朱宪彝教授针对维生素 D 与钙磷代谢和骨软化的研究，以及由他们提出并命名的“肾性骨营养不良”得到了国际学者的广泛认可。维生素 D 及其相关制剂（或称类似物）的应用从根本上遏制了全球范围内佝偻病/骨软化症的广泛流行趋势。然而，维生素 D 缺乏和营养不足在人群中仍普遍存在，全球约有超过 10 亿人群的血清 25 羟维生素 D（25 hydroxyvitamin D，25OHD）水平达不到维持骨骼肌肉健康所推荐的 30 μg/L（75 nmol/L）水平，因此维生素 D 作为基本健康补充剂用于骨质疏松症的防治。维生素 D 在体内经过两步羟化后形成 1,25 双羟维生素 D [1,25-dihydroxyvitamin D，1,25(OH)2 D]，后者是体内维生素 D 的主要活性形式并发挥重要的生理作用，又被称为“D 激素”或“活性维生素 D”。近年来，有许多与维生素 D 结构相似且具有活性维生素 D 样作用的化学物质（活性维生素 D 类似物）被不断开发并应用于临床，特别是用于骨质疏松症、佝偻病、慢性肾脏病和皮肤病等疾病。随着维生素 D 受体（vitamin D receptors，VDR）和 25OHD-1α-羟化酶（1α-hydroxylase，CYP27B1）在许多骨外组织中被发现，维生素 D 的作用已不再囿于调节钙磷代谢和维护骨骼健康，其在肌肉、心血管疾病、糖尿病、癌症、自身免疫和炎性反应等中的作用也逐渐被关注，维生素 D 已成为临床及基础研究的热点。
## 1 维生素 D 概述
维生素 D 是一种脂溶性的开环固醇类物质，包括动物来源的维生素 D3 [胆骨化醇，cholecalciferol，化学名：9,10-开环胆甾-5,7,10(19)-三烯-3β-醇] 和植物来源的维生素 D2 [麦角固醇，ergocalciferol，化学名：9,10-开环麦角甾-5,7,10(19),22-四烯-3β-醇]。维生素 D 在体内经 25 羟化酶的催化合成 25OHD，是体内的主要贮存形式，反映体内维生素 D 的营养状态。25OHD 经过 1α 位羟化成为 1,25(OH)2 D，是体内维生素 D 的主要活性代谢物，与组织中广泛存在的维生素 D 受体结合，发挥激素样作用，又称 D 激素。因此，维生素 D 亦被看作是激素原。维生素 D 及其代谢物的主要生理作用是促进钙和磷在肠道中吸收，并抑制甲状旁腺素（parathyroid hormone，PTH）释放，维持血钙和磷水平正常，进而保证骨骼健康和神经肌肉功能正常。维生素 D 的骨骼外作用包括对肌肉、心血管、代谢、免疫、肿瘤发生、妊娠和胎儿发育等多方面的影响。
维生素 D 的发现源自 1920 至 1930 年对佝偻病的研究，维生素 D 缺乏、代谢异常或过量主要影响骨代谢和钙、磷稳态，D 激素属于“钙调激素”之一。严重维生素 D 缺乏和代谢异常会导致佝偻病/骨软化症。
维生素 D 不足与骨质疏松症及其骨折密切相关。推荐维持骨骼健康的循环 25OHD 水平应达到 30 μg/L（75 nmol/L）以上。同时，维生素 D 缺乏和作用不足还与多种疾病的发生发展相关联。老年人可能存在维生素 D 营养缺乏、活性维生素 D 的生成减少和作用不足，并容易并发肌少症、虚弱症和跌倒风险增加。
随着社会经济发展和生活方式变化，特别是户外生活、工作时间的减少，维生素 D 缺乏已经成为全球性的公共健康问题。流行病学资料表明维生素 D 缺乏在我国人群中普遍存在。近年来，维生素 D 与骨骼健康及多种疾病的联系受到了广泛重视，维生素 D 检测、补充和活性维生素 D 及其类似物使用日趋频繁，但所采用的制剂和方法各异。为 |
维生素D 缺乏性佝偻病中医诊疗指南
丁 樱，任献青，韩改霞，刘莎莎，郭庆寅
(河南中医学院第一附属医院儿科医院，河南 郑州 450000)
说 明
1)本指南的编写目的在于规范中医儿科的临床诊断和
治疗， 为临床医师提供中医标准化处理的策略与方法，
促进中医儿科临床诊疗和科研水平的提高。
2)本指南是根据现代中医儿科学的发展状况
和临床需要，在文献研究、 专家调查问卷分析、 专家
论证会的基础上形成的。
3)本指南内容涵盖了维生素D 缺乏性佝偻病
的中医诊断、 辨证、 治疗方法，适用于儿童维生素D
缺乏性佝偻病患者的中医诊疗。
4)本指南由中华中医药学会儿科分会组织实
施，河南中医学院第一附属医院为负责起草单位。
5)本指南主要起草人:丁樱、 任献青、 韩改霞、
刘莎莎、 郭庆寅。
6)本指南研究经费由国家中医药管理局提供，
与其他任何组织或个人无潜在利益冲突。
7)本指南的形成过程
(1) 文献检索
文献检索主要利用检索工具，采取人工检索、 计
算机检索和网络检索相结合的方法查询相关文献。
其中古代文献资料主要通过《 中医儿科古代文献数
据库》《 中华医典》《 古今图书集成－医部全录》 以及
大学图书馆检索查找。
现代期刊文献主要在中国清华大学制作的“ 中
国期刊全文数据库” 和美国国立医学图书馆制作的
生物医学文献数据库“ MEDLINE” 中检索查找，为保
证查全率，统一以维生素D 缺乏性佝偻病病名作为
检索词，从多种途径查找。
现代其他文献主要参考了国家中医药管理局
1994 年6 月28 日发布的《 中华人民共和国中医药行
业标准·中医病证诊断疗效标准》、 多版中医儿科全
国教材以及西医学的相关指南、 标准及主要教材。
(2)专家调查
本指南依据文献检索的结果，从诊断、 辨证、 治
法、 方药、 预防护理等方面综合古今见解，按Delphi
法(即专家调查法)制作了问卷，向以中医儿科医师
为主的专家(高级职称者)群体征求建议，共制作了
3 轮专家问卷，总回收率在84． 52% 以上，从而形成
了专家共识。
(3)证据选择
证据的采集主要为随机临床试验，还包括:同期
对照研究、 历史对照研究、 病例报道、 非对照研究和
专家意见。
(4)证据及推荐建议级别
本指南采用2001 年国际感染论坛(ISF)提出的
Delphi 分级标准，确定推荐建议的级别。见表1。
(5)Delphi 分级标准的有关说明
①推荐级别或推荐强度分为A，B，C，D，E 5 级。
强度以A 级为最高，并依次递减。
②所谓推荐级别或推荐强度只是指文献的支持
程度，并不代表特别建议。
③该标准的“ 研究课题分级” 中，大样本指≥
100 例的高质量的单篇随机对照试验报道或系统综
述报告，小样本指＜100 例的高质量的单篇随机对照
试验报道。
④研究课题分级(即临床报道文献)之Ⅳ、 Ⅴ级
均有专家意见，分别定义为Ⅳ级:为专家共识，例如
本项目所实施的专家调查问卷结果。Ⅴ级:为个别
或其他报道中的专家意见。
⑤本指南工作中，在采用Delphi 分级标准时，
其研究课题中的Ⅰ 级、 Ⅱ 级随机对照试验，依据
Jadad 量表评分，只采用质量在3 分(包括3 分)以
上的文献。如只有低于3 分的文献，则不作为随机
对照试验采纳。
Jadad 评分量表
a 随机分组序列的产生方法。2 分:通过计算机
产生的随机序列或随机数字表产生的序列。1 分:
试验提到随机分配，但产生随机序列的方法未予交
待。0 分:半随机或准随机试验，指采用交替分配病
例的方法如入院顺序、 出生日期单双数。
收稿日期:2011－10－13
基金项目:国家中医药管理局中医药标准化项目［ZYYS－2009(0004)－34］。
作者简介:丁樱(1951－)，女，教授，主任医师，博士生导师，主要从事中医儿科临床和研究工作。
—
1
—第8 卷第1 期
2012 年1 月
中医儿科杂志
J. PEDIATRICS OF TCM
Vol. 8 No. 1
Jan. 2012
表1 Delphi 分级标准
推荐级别 研究课题分级
A 至少有2 项I 级研究结果支持 Ⅰ 大样本，随机研究，结果清晰，假阳性或
假阴性的错误很低
B 仅有1 项I 级研究结果支持 Ⅱ 小样本，随机研究，结果不确定，假阳性
和/ 或假阴性的错误较高
C 仅有II 级研究结果支持 Ⅲ 非随机，同期对照研究
D 至少有1 项III 级研究结果支持 IV 非随机，历史对照和专家意见
E 仅有IV 级或V 研究结果支持研究 V 病例报道，非对照研究和专家意见
b 双盲法。2 分:描述了实施双盲的具体方法
并且被认为是恰当的，如采用完全一致的安慰剂等。
1 分:试验仅提及采用双盲法。0 分:试验提及采用
双盲，但方法不恰当，如比较片剂与注射剂而未提及
使用双伪法。
c 退出与失防。1 分:对退出与失防的病例数和
退出理由进行了详细的描述。0 分:没有提到退出
与失防。
(6)关于本指南诊断与辨证部分的推荐级别的
标注说明
本指南中的诊断和辨证部分，不依据“ Delphi 分
级标准” 标注证据分级和推荐意见级别。若有高质
量的诊断性试验或中医证候辨证分型研究的文献，
可直接作为参考文献标注在相关的内容处。
(7)形成指南
综合专家问卷调查结果和文献检索分析结果，
由丁樱教授、 任献青副教授执笔撰写指南初稿，之后
召开了专家论证会，征求中医儿科专家以及中华中
医药学会、 国家中医药管理局、 国家标准局等专家及
管理干部的意见，最终形成了本指南。
(8)本指南计划2 年更新1 次，通过文献研究
与专家研讨会相结合的方式实施更新计划。
1 范围
本指南提出了维生素D 缺乏性佝偻病的诊断、
辨证、 治疗建议。本指南适用于维生素D 缺乏性佝
偻病的诊断和治疗。
2 术语和定义
下列术语和定义适用于本指南。
维生素D 缺乏性佝偻病(rickets of vitamin D
deficiency)是由于儿童体内维生素D 不足，致使钙
磷代谢失常的一种慢性营养性疾病，临床以正在生
长的骨骺端软骨板不能正常钙化，造成骨骼改变为
—
2
—
主要特征。本病常发于冬春季，主要见于婴幼儿，尤
以6 ～ 12 月婴儿发病率较高。北方发病率高于南方
地区，工业城市高于农村，人工喂养的婴儿发病率高
于母乳喂养者。本病轻证如治疗得当，预后良好;重
者如失治、 误治，易导致骨骼畸形，留有后遗症，影响
儿童正常生长发育。
古代医籍中的夜惊、 鸡胸、 龟背、 龟胸、 汗证、 五
软、 五迟等病证，有与本病相关的论述。
3 诊断［1－10 ］
3． 1 临床表现
本病根据症状、 体征可以分为4 期:
3． 1． 1 初期 有多汗、 烦躁、 睡眠不安、 夜间惊啼。多
汗与室温及季节无关， 常因多汗及烦躁而摇头擦枕， 出
现枕秃及脱发圈。还可见囟门迟闭、 牙齿迟出等。
3． 1． 2 激期 除早期症状加重外，还可见乒乓头、
方颅、 肋串珠、 肋外翻、 鸡胸、 漏斗胸、 龟背、 手脚镯、
下肢弯曲等骨骼改变。
3． 1． 3 恢复期 经治疗后，症状逐渐好转而至消
失，体征逐渐减轻、 恢复。
3． 1． 4 后遗症期 多见于3 岁以后的小儿，经治疗
或自然恢复，症状逐渐消失，骨骼改变不再进展，但
遗留不同程度的骨骼畸形，无其他临床症状。
3． 2 实验室检查
1)初期血钙正常或稍低，血磷明显降低，钙磷
乘积小于30，血清碱性磷酸酶增高。X 线片可正常
或钙化带稍模糊，血清25- (OH)2 D3 下降。
2)激期血清钙、 磷均降低，碱性磷酸酶明显增
高，腕部X 线片见临时钙化带模糊，干骺端增宽，边
缘呈毛刷状或杯口状改变。
3)恢复期X 线片临时钙化带重现，血生化恢复
正常。
4)后遗症期理化检查均正常。第8 卷第1 期
2012 年1 月
中医儿科杂志
J. PEDIATRICS OF TCM
Vol. 8 No. 1
Jan. 2012
3． 3 需与维生素D 缺乏性佝偻病鉴别的病种
肾性佝偻病、 肾小管性酸中毒、 软骨营养不良、
维生素D 依赖性佝偻病、 先天性甲状腺功能低下、
低血磷抗维生素D 佝偻病。
4 辨证［1－10 ］
4． 1 肺脾气虚证
形体虚胖，肌肉松软，面色少华，纳呆，大便不调，
多汗， 睡眠不宁，囟门开大，头发稀疏易落，可见枕秃，
易反复感冒， 舌淡、 苔薄白， 指纹淡， 脉细软无力。
4． 2 脾虚肝旺证
烦躁夜啼，惊惕不安，面色少华或面色萎黄，头
部多汗，发稀枕秃，囟门迟闭，出牙延迟，纳呆食少，
坐立行走无力，夜啼不宁，易惊多惕，甚则抽搐，舌
淡、 苔薄，指纹淡青，脉细弦。
4． 3 肾精亏损证
面白虚烦，形瘦神疲，纳呆乏力，多汗肢软，筋骨
萎软，立迟、 行迟、 齿迟，头颅方大，肋骨串珠，手镯脚
镯，鸡胸龟背，下肢畸变，舌淡、 苔少，指纹淡紫，脉细
无力。该期已有明显骨骼畸形后遗症。
5 治疗
5． 1 治疗原则(推荐级别:D)［1－10 ］
本病治疗，重在调补脾肾，多用补益之法，先天
不足者补肾为先，后天失调者补脾为先，脾肾俱虚，
病程迁延者，脾肾兼顾，同时注意益肾填精壮骨。根
据脾肾亏损轻重，采用不同的治法。初期以脾虚为
主，用健脾益气为主法;激期多属脾肾两亏，当予脾
肾并补;恢复期、 后遗症期以肾虚为主，当补肾填精，
佐以健脾。本病在调补脾肾的同时，还要注意到补
肺益气固表、 平肝清心安神等治法的配合使用。
此外，在预防护理上鼓励母乳喂养，科学合理添
加辅食。注意维生素D 及钙、 磷的补充，增加小儿
户外活动、 多晒太阳。勿使患儿过早或过多坐立和
行走，提倡穿背带裤。已有骨骼严重畸形后遗症患
儿可手术矫正。
5． 2 分证论治
5． 2． 1 肺脾气虚证(推荐级别:E)［1－10 ］ 治法:健
脾益气，补肺固表。主方:人参五味子汤加味。常用
药:人参、 白术、 茯苓、 五味子、 麦冬、 天门冬、 黄芪、 炙
甘草等。加减:汗多者加煅龙骨、 煅牡蛎固涩止汗，
夜惊、 睡眠不宁、 烦躁者加炒酸枣仁、 夜交藤，大便不
实者加苍术、 山药、 白扁豆。
5． 2． 2 脾虚肝旺证(推荐级别:E)［1－10 ］ 治法:健
脾柔肝，平肝息风。主方:益脾镇惊散加减。常用
药:人参、 白术、 茯苓、 朱砂、 钩藤、 煅龙骨、 煅牡蛎、 炙
甘草、 灯心草。加减:多汗者加五味子、 瘪桃干，睡中
惊惕者加远志、 珍珠母、 僵蚕，抽搐者加全蝎、 蜈蚣，
夜啼不宁者加蝉蜕、 竹叶。
5． 2． 3 肾精亏损证(推荐级别:E)［1－10 ］ 治法:补
肾填精，佐以健脾。主方:补肾地黄丸加减。常用
药:紫河车、 熟地黄、 肉苁蓉、 巴戟天、 菟丝子、 山茱
萸、 枸杞子、 山药、 酸枣仁、 远志。加减:多汗者加黄
芪、 煅龙骨、 煅牡蛎，乏力肢软者加黄芪、 党参，纳呆
者加砂仁、 陈皮、 佛手，面白者加当归、 白芍。
5． 3 常用中成药
1)龙牡壮骨冲剂(党参、 黄芪、 麦冬、 醋制龟甲、
炒白术、 山药、 醋制五味子、 龙骨、 煅牡蛎、 茯苓、 大
枣、 甘草、 乳酸钙、 炒鸡内金、 维生素D2 、 葡萄糖酸
钙):2 岁以下每服半包， 2 ～ 7 岁每服1 包， 7 岁以上
每服2 包， 1 日3 次。可用于各证型。(推荐级别:
C)［9 ］
2)玉屏风口服液(蜜炙黄芪、 炒白术、 防风):每服
1 支， 1 日3 次。用于肺脾气虚证。(推荐级别:E)
3)六味地黄丸(熟地黄、 山茱萸、 山药、 牡丹皮、
茯苓、 泽泻):每服3 ～ 6 g，1 日3 次。用于肾精亏损
证。(推荐级别:E)
| # 维生素D 缺乏性佝偻病中医诊疗指南
## 作者
丁樱，任献青，韩改霞，刘莎莎，郭庆寅
(河南中医学院第一附属医院儿科医院，河南 郑州 450000)
## 说明
1. 本指南的编写目的在于规范中医儿科的临床诊断和治疗，为临床医师提供中医标准化处理的策略与方法，促进中医儿科临床诊疗和科研水平的提高。
2. 本指南是根据现代中医儿科学的发展状况和临床需要，在文献研究、专家调查问卷分析、专家论证会的基础上形成的。
3. 本指南内容涵盖了维生素D 缺乏性佝偻病的中医诊断、辨证、治疗方法，适用于儿童维生素D 缺乏性佝偻病患者的中医诊疗。
4. 本指南由中华中医药学会儿科分会组织实施，河南中医学院第一附属医院为负责起草单位。
5. 本指南主要起草人:丁樱、任献青、韩改霞、刘莎莎、郭庆寅。
6. 本指南研究经费由国家中医药管理局提供，与其他任何组织或个人无潜在利益冲突。
## 指南的形成过程
1. **文献检索**
文献检索主要利用检索工具，采取人工检索、计算机检索和网络检索相结合的方法查询相关文献。其中古代文献资料主要通过《中医儿科古代文献数据库》《中华医典》《古今图书集成－医部全录》以及大学图书馆检索查找。现代期刊文献主要在中国清华大学制作的“中国期刊全文数据库”和美国国立医学图书馆制作的生物医学文献数据库“MEDLINE”中检索查找，为保证查全率，统一以维生素D 缺乏性佝偻病病名作为检索词，从多种途径查找。现代其他文献主要参考了国家中医药管理局1994 年6 月28 日发布的《中华人民共和国中医药行业标准·中医病证诊断疗效标准》、多版中医儿科全国教材以及西医学的相关指南、标准及主要教材。
2. **专家调查**
本指南依据文献检索的结果，从诊断、辨证、治法、方药、预防护理等方面综合古今见解，按Delphi 法(即专家调查法)制作了问卷，向以中医儿科医师为主的专家(高级职称者)群体征求建议，共制作了3 轮专家问卷，总回收率在84.52% 以上，从而形成了专家共识。
3. **证据选择**
证据的采集主要为随机临床试验，还包括:同期对照研究、历史对照研究、病例报道、非对照研究和专家意见。
4. **证据及推荐建议级别**
本指南采用2001 年国际感染论坛(ISF)提出的Delphi 分级标准，确定推荐建议的级别。
5. **Delphi 分级标准的有关说明**
- 推荐级别或推荐强度分为A，B，C，D，E 5 级。强度以A 级为最高，并依次递减。
- 所谓推荐级别或推荐强度只是指文献的支持程度，并不代表特别建议。
- 该标准的“研究课题分级”中，大样本指≥100 例的高质量的单篇随机对照试验报道或系统综述报告，小样本指＜100 例的高质量的单篇随机对照试验报道。
- 研究课题分级(即临床报道文献)之Ⅳ、Ⅴ级均有专家意见，分别定义为Ⅳ级:为专家共识，例如本项目所实施的专家调查问卷结果。Ⅴ级:为个别或其他报道中的专家意见。
- 本指南工作中，在采用Delphi 分级标准时，其研究课题中的Ⅰ级、Ⅱ级随机对照试验，依据Jadad 量表评分，只采用质量在3 分(包括3 分)以上的文献。如只有低于3 分的文献，则不作为随机对照试验采纳。
6. **关于本指南诊断与辨证部分的推荐级别的标注说明**
本指南中的诊断和辨证部分，不依据“Delphi 分级标准”标注证据分级和推荐意见级别。若有高质量的诊断性试验或中医证候辨证分型研究的文献，可直接作为参考文献标注在相关的内容处。
7. **形成指南**
综合专家问卷调查结果和文献检索分析结果，由丁樱教授、任献青副教授执笔撰写指南初稿，之后召开了专家论证会，征求中医儿科专家以及中华中医药学会、国家中医药管理局、国家标准局等专家及管理干部的意见，最终形成了本指南。
8. **本指南计划2 年更新1 次，通过文献研究与专家研讨会相结合的方式实施更新计划。**
## 范围
本指南提出了维生素D 缺乏性佝偻病的诊断、辨证、治疗建议。本指南适用于维生素D 缺乏性佝偻病的诊断和治疗。
## 术语和定义
下列术语和定义适用于本指南。
维生素D 缺乏性佝偻病(rickets of vitamin D deficiency)是由于儿童体内维生素D 不足，致使钙磷代谢失常的一种慢性营养性疾病，临床以正在生长的骨骺端软骨板不能正常钙化，造成骨骼改变为主要特征。本病常发于冬春季，主要见于婴幼儿，尤以6 ～ 12 月婴儿发病率较高。北方发病率高于南方地区，工业城市高于农村，人工喂养的婴儿发病率高于母乳喂养者。本病轻证如治疗得当，预后良好;重者如失治、误治，易导致骨骼畸形，留有后遗症，影响儿童正常生长发育。
古代医籍中的夜惊、鸡胸、龟背、龟胸、汗证、五软、五迟等病证，有与本病相关的论述。
## 诊断
### 临床表现
本病根据症状、体征可以分为4 期:
1. **初期**
有多汗、烦躁、睡眠不安、夜间惊啼。多汗与室温及季节无关，常因多汗及烦躁而摇头擦枕，出现枕秃及脱发圈。还可见囟门迟闭、牙齿迟出等。
2. **激期**
除早期症状加重外，还可见乒乓头、方颅、肋串珠、肋外翻、鸡胸、漏斗胸、龟背、手脚镯、下肢弯曲等骨骼改变。
3. **恢复期**
经治疗后，症状逐渐好转而至消失，体征逐渐减轻、恢复。
4. **后遗症期**
多见于3 岁以后的小儿，经治疗或自然恢复，症状逐渐消失，骨骼改变不再进展，但遗留不同程度的骨骼畸形，无其他临床症状。
### 实验室检查
1. 初期血钙正常或稍低，血磷明显降低，钙磷乘积小于30，血清碱性磷酸酶增高。X 线片可正常或钙化带稍模糊，血清25- (OH)2 D3 下降。
2. 激期血清钙、磷均降低，碱性磷酸酶明显增高，腕部X 线片见临时钙化带模糊，干骺端增宽，边缘呈毛刷状或杯口状改变。
3. 恢复期X 线片临时钙化带重现，血生化恢复正常。
4. 后遗症期理化检查均正常。
### 需与维生素D 缺乏性佝偻病鉴别的病种
肾性佝偻病、肾小管性酸中毒、软骨营养不良、维生素D 依赖性佝偻病、先天性甲状腺功能低下、低血磷抗维生素D 佝偻病。
## 辨证
1. **肺脾气虚证**
形体虚胖，肌肉松软，面色少华，纳呆，大便不调，多汗，睡眠不宁，囟门开大，头发稀疏易落，可见枕秃，易反复感冒，舌淡、苔薄白，指纹淡，脉细软无力。
2. **脾虚肝旺证**
烦躁夜啼，惊惕不安，面色少华或面色萎黄，头部多汗，发稀枕秃，囟门迟闭，出牙延迟，纳呆食少，坐立行走无力，夜啼不宁，易惊多惕，甚则抽搐，舌淡、苔薄，指纹淡青，脉细弦。
3. **肾精亏损证**
面白虚烦，形瘦神疲，纳呆乏力，多汗肢软，筋骨萎软，立迟、行迟、齿迟，头颅方大，肋骨串珠，手镯脚镯，鸡胸龟背，下肢畸变，舌淡、苔少，指纹淡紫，脉细无力。该期已有明显骨骼畸形后遗症。
## 治疗
### 治疗原则(推荐级别:D)
本病治疗，重在调补脾肾，多用补益之法，先天不足者补肾为先，后天失调者补脾为先，脾肾俱虚，病程迁延者，脾肾兼顾，同时注意益肾填精壮骨。根据脾肾亏损轻重，采用不同的治法。初期以脾虚为主，用健脾益气为主法;激期多属脾肾两亏，当予脾肾并补;恢复期、后遗症期以肾虚为主，当补肾填精，佐以健脾。本病在调补脾肾的同时，还要注意到补肺益气固表、平肝清心安神等治法的配合使用。
此外，在预防护理上鼓励母乳喂养，科学合理添加辅食。注意维生素D 及钙、磷的补充，增加小儿户外活动、多晒太阳。勿使患儿过早或过多坐立和行走，提倡穿背带裤。已有骨骼严重畸形后遗症患儿可手术矫正。
### 分证论治
1. **肺脾气虚证(推荐级别:E)**
治法:健脾益气，补肺固表。主方:人参五味子汤加味。常用药:人参、白术、茯苓、五味子、麦冬、天门冬、黄芪、炙甘草等。加减:汗多者加煅龙骨、煅牡蛎固涩止汗，夜惊、睡眠不宁、烦躁者加炒酸枣仁、夜交藤，大便不实者加苍术、山药、白扁豆。
2. **脾虚肝旺证(推荐级别:E)**
治法:健脾柔肝，平肝息风。主方:益脾镇惊散加减。常用药:人参、白术、茯苓、朱砂、钩藤、煅龙骨、煅牡蛎、炙甘草、灯心草。加减:多汗者加五味子、瘪桃干，睡中惊惕者加远志、珍珠母、僵蚕，抽搐者加全蝎、蜈蚣，夜啼不宁者加蝉蜕、竹叶。
3. **肾精亏损证(推荐级别:E)**
治法:补肾填精，佐以健脾。主方:补肾地黄丸加减。常用药:紫河车、熟地黄、肉苁蓉、巴戟天、菟丝子、山茱萸、枸杞子、山药、酸枣仁、远志。加减:多汗者加黄芪、煅龙骨、煅牡蛎，乏力肢软者加黄芪、党参，纳呆者加砂仁、陈皮、佛手，面白者加当归、白芍。
### 常用中成药
1. **龙牡壮骨冲剂(党参、黄芪、麦冬、醋制龟甲、炒白术、山药、醋制五味子、龙骨、煅牡蛎、茯苓、大枣、甘草、乳酸钙、炒鸡内金、维生素D2、葡萄糖酸钙):2 岁以下每服半包，2 ～ 7 岁每服1 包，7 岁以上每服2 包，1 日3 次。可用于各证型。(推荐级别:C)**
2. **玉屏风口服液(蜜炙黄芪、炒白术、防风):每服1 支，1 日3 次。用于肺脾气虚证。(推荐级别:E)**
3. **六味地黄丸(熟地黄、山茱萸、山药、牡丹皮、茯苓、泽泻):每服3 ～ 6 g，1 日3 次。用于肾精亏损证。(推荐级别:E)** |
一、足月新生儿和婴儿
（1）0~6 个月：无论何种喂养方式，从出生后第一天起，予维生素 D3 400 IU/天（10 μg/天）。
（2）6~12 个月：根据饮食摄入的维生素 D 总量，予维生素 D3 400~600 IU/天（10~15 μg/天）。
（3）不推荐足月新生儿和健康婴儿使用骨化二醇。
二、儿童（1~10 岁）
（1）在 1~3 岁的健康儿童中，应选取维生素 D3 作为补充剂，每日剂量为 600 IU（15 μg/天），由于年龄较小而日光浴时间受限，建议全年补充。
（2）4~10 岁的健康儿童，从 5 月到 9 月底，建议在 10:00~15:00 进行 15~30 分钟的日光浴，期间应裸露前臂和腿部，无需涂抹防晒霜。这期间不需要补充维生素 D3，但仍然推荐补充且是安全的。
（3）对于不符合上述指南的 4~10 岁健康儿童，建议根据体重和膳食中维生素 D 摄入量补充维生素 D3，剂量为 600~1000 IU/天（15~25 μg/天）。
（4）对于 1~10 岁的健康儿童，不推荐使用骨化二醇。
三、青少年（11~18 岁）
（1）健康青少年预防维生素 D 缺乏，维生素 D3 是补充剂的首选，骨化二醇是次选。
（2）健康青少年建议在从 5 月到 9 月底的 10:00~15:00 之间裸露前臂和腿部进行 30~45
分钟日光浴，这期间不需要补充维生素 D3，但仍然推荐补充且是安全的。
（3） 如果不符合上述指南， 建议根据体重和膳食中维生素 D 摄入量， 全年补充维生素 D3，剂量为 1000~2000 IU/天（25-50 μg/天）。
（4）如果不符合上述指南，推荐全年补充骨化二醇，每日剂量为 10 ug（口服液），并在开始 6~8 天后进行血清 25(OH)D 测定，与开始服用补充剂前进行对比。
四、19 岁至 65 岁的成年人
（1）健康成人中，以维生素 D3 为首选补充剂和骨化二醇为第二选择。
（2）健康成人建议在从 5 月到 9 月底的 10:00~15:00 之间裸露前臂和腿部进行 30~45分钟日光浴，这期间不需要补充维生素 D3，但仍然推荐补充且是安全的。（3） 如果不符合上述指南， 建议根据体重和膳食中维生素 D 摄入量， 全年补充维生素 D3，
剂量为 1000~2000 IU/天（25-50 μg/天）。
（4）如果不符合上述指南，推荐全年补充骨化二醇，每日剂量为 10 ug（口服液），并在开始 6~8 天后进行血清 25(OH)D 测定，与开始服用补充剂前进行对比。
五、年龄较小的老年人(65-75 岁)
（1）健康成人中，以维生素 D3 为首选补充剂和骨化二醇为第二选择。
（2）由于皮肤合成的功效下降，建议根据体重和膳食中维生素 D 摄入量，全年补充维生素 D3，剂量为 1000~2000 IU/天（25-50 μg/天）。
（3）如果不符合上述指南，推荐全年补充骨化二醇，每日剂量为 10 ug（口服液），并在开始 6~8 天后进行血清 25(OH)D 测定，与开始服用补充剂前进行对比。
六、老年人（75-89 岁）和最年长的老年人（≥90 岁）
（1）健康成人中，以维生素 D3 为首选补充剂和骨化二醇为第二选择。
（2）由于皮肤合成的潜在吸收不良和代谢改变，建议根据体重和膳食中维生素 D 摄入量，
全年补充维生素 D3，剂量为 2000~4000 IU/天（25-50 μg/天）。
（3）如果不符合上述指南，推荐全年补充骨化二醇，每日剂量为 10 ug（口服液），并在
开始 6~8 天后进行血清 25(OH)D 测定，与开始服用补充剂前进行对比。
七、妊娠期和哺乳期
（1）计划怀孕的妇女应与普通成年人一样补充足够的维生素 D3，使血清 25(0H)D 浓度维持在正常范围。
（2）当确认妊娠至哺乳结束时，补充维生素 D3，使血清 25(0H)D 浓度维持在 30-50ng/ml。
（3）如果不能评估血清 25(0H)D 的浓度，建议在妊娠和哺乳期间使用 2000IU/日的维生素D3。 | ### 足月新生儿和婴儿
- **0~6 个月**：无论何种喂养方式，从出生后第一天起，予维生素 D3 400 IU/天（10 μg/天）。
- **6~12 个月**：根据饮食摄入的维生素 D 总量，予维生素 D3 400~600 IU/天（10~15 μg/天）。
- **不推荐**足月新生儿和健康婴儿使用骨化二醇。
### 儿童（1~10 岁）
- **1~3 岁**：应选取维生素 D3 作为补充剂，每日剂量为 600 IU（15 μg/天），由于年龄较小而日光浴时间受限，建议全年补充。
- **4~10 岁**：从 5 月到 9 月底，建议在 10:00~15:00 进行 15~30 分钟的日光浴，期间应裸露前臂和腿部，无需涂抹防晒霜。这期间不需要补充维生素 D3，但仍然推荐补充且是安全的。
- **不符合上述指南的 4~10 岁健康儿童**：建议根据体重和膳食中维生素 D 摄入量补充维生素 D3，剂量为 600~1000 IU/天（15~25 μg/天）。
- **不推荐**1~10 岁的健康儿童使用骨化二醇。
### 青少年（11~18 岁）
- **健康青少年**：预防维生素 D 缺乏，维生素 D3 是补充剂的首选，骨化二醇是次选。
- **建议**：在从 5 月到 9 月底的 10:00~15:00 之间裸露前臂和腿部进行 30~45 分钟日光浴，这期间不需要补充维生素 D3，但仍然推荐补充且是安全的。
- **不符合上述指南**：建议根据体重和膳食中维生素 D 摄入量，全年补充维生素 D3，剂量为 1000~2000 IU/天（25-50 μg/天）。
- **推荐**：全年补充骨化二醇，每日剂量为 10 ug（口服液），并在开始 6~8 天后进行血清 25(OH)D 测定，与开始服用补充剂前进行对比。
### 19 岁至 65 岁的成年人
- **健康成人**：以维生素 D3 为首选补充剂和骨化二醇为第二选择。
- **建议**：在从 5 月到 9 月底的 10:00~15:00 之间裸露前臂和腿部进行 30~45 分钟日光浴，这期间不需要补充维生素 D3，但仍然推荐补充且是安全的。
- **不符合上述指南**：建议根据体重和膳食中维生素 D 摄入量，全年补充维生素 D3，剂量为 1000~2000 IU/天（25-50 μg/天）。
- **推荐**：全年补充骨化二醇，每日剂量为 10 ug（口服液），并在开始 6~8 天后进行血清 25(OH)D 测定，与开始服用补充剂前进行对比。
### 年龄较小的老年人(65-75 岁)
- **健康成人**：以维生素 D3 为首选补充剂和骨化二醇为第二选择。
- **建议**：由于皮肤合成的功效下降，建议根据体重和膳食中维生素 D 摄入量，全年补充维生素 D3，剂量为 1000~2000 IU/天（25-50 μg/天）。
- **不符合上述指南**：推荐全年补充骨化二醇，每日剂量为 10 ug（口服液），并在开始 6~8 天后进行血清 25(OH)D 测定，与开始服用补充剂前进行对比。
### 老年人（75-89 岁）和最年长的老年人（≥90 岁）
- **健康成人**：以维生素 D3 为首选补充剂和骨化二醇为第二选择。
- **建议**：由于皮肤合成的潜在吸收不良和代谢改变，建议根据体重和膳食中维生素 D 摄入量，全年补充维生素 D3，剂量为 2000~4000 IU/天（25-50 μg/天）。
- **不符合上述指南**：推荐全年补充骨化二醇，每日剂量为 10 ug（口服液），并在开始 6~8 天后进行血清 25(OH)D 测定，与开始服用补充剂前进行对比。
### 妊娠期和哺乳期
- **计划怀孕的妇女**：应与普通成年人一样补充足够的维生素 D3，使血清 25(0H)D 浓度维持在正常范围。
- **确认妊娠至哺乳结束时**：补充维生素 D3，使血清 25(0H)D 浓度维持在 30-50ng/ml。
- **不能评估血清 25(0H)D 的浓度**：建议在妊娠和哺乳期间使用 2000IU/日的维生素 D3。 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.