Nephrol Dial Transplant (2016) 31: 698–705
doi: 10.1093/ndt/gfw082
Polar Views in Nephrology
Pro: Should we correct vitamin D deficiency/insufficiency in
chronic kidney disease patients with inactive forms of vitamin D
or just treat them with active vitamin D forms?
David J.A. Goldsmith
Guy’s and St Thomas’ Hospitals, London, UK
Correspondence and offprint requests to: David J.A. Goldsmith; E-mail: [email protected]
is we are trying to achieve here, and how best to balance potential benefits with potential harm.
A B S T R AC T
Evidence for the usefulness of using vitamin D to treat ‘renal
bone disease’ is now nearly six decades old. In regular clinical
practice, however, it is more like three decades, at most, that we
have routinely been using vitamin D to try to prevent, or reverse, the impact of hyperparathyroidism on the skeleton of
patients with chronic kidney disease (CKD). The practice
has been in the main to use high doses of synthetic vitamin
D compounds, not naturally occurring ones. However, the
pharmacological impacts of the different vitamin D species
and of their different modes, and styles of administration cannot be assumed to be uniform across the spectrum. It is disappointingly true to say that even in 2016 there is a remarkable
paucity of evidence concerning the clinical benefits of vitamin
D supplementation to treat vitamin D insufficiency in patients
with stage 3b–5 CKD. This is even more so if we consider the
non-dialysis population. While there are a number of studies
that report the impact of vitamin D supplementation on
serum vitamin D concentrations (unsurprisingly, usually reporting an increase), and some variable evidence of parathyroid hormone concentration suppression, there has been
much less focus on hard or semi-rigid clinical end point analysis (e.g. fractures, hospitalizations and overall mortality).
Now, in 2016, with the practice pattern changes of first widespread clinical use of vitamin D and second widespread supplementation of cholecalciferol or ergocalciferol by patients
(alone, or as multivitamins), it is now, in my view, next to impossible to run a placebo-controlled trial over a decent period
of time, especially one which involved clinically meaningful
(fractures, hospitalisation, parathyroidectomy, death) endpoints. In this challenging situation, we need to ask what it
© The Author 2016. Published by Oxford University Press
on behalf of ERA-EDTA. All rights reserved.
Keywords: hyperparathyroidism, mineral and bone disorder,
PTH, vitamin D
INTRODUCTORY REMARKS
For the purpose of this debate, I am confining my remarks to patients with chronic kidney disease (CKD) stage 3b, 4 and 5 (5ND
and 5D). I am not, in this essay, discussing earlier CKD (stages 1,
2 and 3a), mostly now managed by non-specialists, nor am I debating or discussing the complex matter of the ideal management
of post-transplantation bone pathology.
My esteemed and learned opponent and I have no quarrel
about the relevance of vitamin D to the health and well-being
of patients with clinically significant CKD; we both agree
that vitamin D plays an important role in the management of
skeletal, and possibly other, complications of CKD. It would be
hard indeed to find an opponent who could successfully rise to
the task of opposing the use of vitamin D at all in this clinical
situation, despite a lamentable lack of hard evidence to support
this practice.
Where my opponent and I do differ, however, is not so much
about what cards we need to have at our disposal, but the order
in which these cards are best played. Changing metaphors now,
my opponent uses the most powerful weapons in his armamentarium as soon as he can. I prefer a more nuanced, subtle,
approach to the same problem before deploying my heavier artillery, and only if I need to. This is indeed one of the (many)
distinguishing features between America and Great Britain.
698
It is important, therefore, to appreciate where the ‘fault line’
is between us. It is likely to reside around the issue of proportionality, and gradualness, as opposed to reliance on ‘shock and
awe’. Let us now examine our dossier of evidence, to support the
deployment of our weapons of mass distraction.
B AC K G R O U N D — V I TA M I N D
D E F I C I E N C Y — I T ’ S C O M P L I C AT E D !
•
PTH
•
Serum calcium and phosphate concentrations
•
Fibroblast growth factor 23 (FGF23)
Increased PTH secretion (most often due to a fall in the plasma
calcium concentration) and hypophosphataemia stimulate the
Vitamin D deficiency/insufficiency in CKD patients
699
P O L A R V I E W S I N N E P H R O LO G Y
Vitamin D, or calciferol, is a generic term and refers to a group of
lipid soluble compounds with a four-ringed cholesterol backbone structure. 25-Hydroxyvitamin D [25(OH)D] is the major
circulating form of vitamin D. It has a half-life of ∼3 weeks, compared with 24 h for parent vitamin D. It has activity at bone and
intestine, but has <100 times the potency, in terms of vitamin-D
receptor activation, as 1,25-dihydroxyvitamin D, the most ‘active’
form of vitamin D. The half-life of 1,25-dihydroxyvitamin D is
∼4–6 h. 1,25-Dihydroxyvitamin D binds to intracellular receptors in target tissues and regulates gene transcription. It appears
to function through a single vitamin D receptor (VDR), which is
nearly universally expressed in nucleated cells, and is able to activate perhaps ∼3% of the human genome. Its most important
biological action is to promote enterocyte differentiation and
the intestinal absorption of calcium. Other effects include stimulation of intestinal phosphate absorption, direct suppression of
parathyroid hormone (PTH) release from the parathyroid
gland, regulation of osteoblast function, and permissively allowing PTH-induced osteoclast activation and bone resorption.
Few commonly imbibed foods naturally contain vitamin D
(oily fish livers are the main exception); dermal synthesis is the
major natural source of the vitamin. Previtamin D3 is synthesized
nonenzymatically in skin from 7-dehydrocholesterol during exposure to the ultraviolet (UV) rays in sunlight (most actively
with a wavelength of 311 nm, so in the ‘UV-B’ range). Previtamin
D3 undergoes a temperature-dependent rearrangement to form
vitamin D3 (cholecalciferol). The length of daily skin exposure
required to obtain the sunlight equivalent of oral vitamin D supplementation is difficult to predict on an individual basis and varies with the skin type, latitude, season, time of day and, of course,
the degree of nudity risked. Prolonged exposure of the skin to
sunlight does not produce toxic amounts of vitamin D3 because
of photoconversion of previtamin D3 and vitamin D3 to inactive
metabolites (lumisterol, tachysterol, 5,6-transvitamin D and suprasterols). In addition, prolonged sunlight induces the production of melanin, which through increasing pigmentation reduces
production of vitamin D3 in the skin.
Infants, disabled persons and older adults may have inadequate sun exposure, while the skin of those older than say
70 years of age also does not convert vitamin D so effectively.
In addition, at northern latitudes, there is not enough solar
radiation energy to convert vitamin D, particularly during the
winter. In some countries, milk, cereals, bread, baby mush and
other foodstuffs can be fortified with vitamin D. People with
chronic illness, infirmity or incarcerated in institutions are
typically severely vitamin D deficient.
Vitamin D from the diet or dermal synthesis is biologically
inactive and requires enzymatic conversion in the liver and
kidney to active metabolites. Dietary vitamin D travels to the
liver, bound to vitamin D-binding protein and in continued
association with chylomicrons and lipoproteins, where dietary
vitamin D and endogenously synthesized vitamin D3 are both
metabolized. The hepatic enzyme 25-hydroxylase places a
hydroxyl group in the 25 position of the vitamin D molecule,
resulting in the formation of 25(OH)D (calcidiol). 25(OH)D2
has a lower affinity than 25(OH)D3 for vitamin D-binding protein. Thus, 25(OH)D2 has a shorter half-life than 25(OH)D3,
and treatment with vitamin D2 may not increase serum total
25(OH)D levels as efficiently as vitamin D3 supplementation
can. 25(OH)D2 and D3 produced by the liver enter the circulation and travel to the kidney, again bound to vitamin D-binding
protein. This protein has a single binding site, which binds
vitamin D and all of its metabolites. Only 3–5% of the total
circulating binding sites are normally occupied; as a result,
this protein is not rate limiting in vitamin D metabolism unless
large amounts are lost in the urine, as is sometimes seen in
nephrotic syndrome. In the renal tubule, entry of the filtered
25(OH)D–vitamin D-binding protein complex into the cells
is facilitated by receptor-mediated endocytosis. At least two
proteins working together are involved in this process: cubilin
and megalin. Cubilin and megalin, expressed in the renal proximal tubule, are multi-ligand receptors that facilitate uptake of
extracellular ligands. Deficiency of either or both of these proteins results in increased 25(OH)D excretion in the urine.
Within the tubular cell, 25(OH)D is released from the
binding protein. The renal tubular cells contain two enzymes,
1-α-hydroxylase (CYP27B1) and 24-alpha-hydroxylase (CYP24),
that can further hydroxylate 25(OH)D, producing 1,25dihydroxyvitamin D, the most active form of vitamin D, or
24,25-dihydroxyvitamin D, an inactive metabolite. Both enzymes are members of the cytochrome P-450 system. In normal
human kidney, the distal nephron is the predominant site of
1-α-hydroxylase expression under conditions of vitamin D
sufficiency.
The 1-α-hydroxylase enzyme is also expressed in extrarenal
sites, including the gastrointestinal tract, skin, vasculature,
mammary epithelial cells, osteoblasts and osteoclasts. The
most widely recognized manifestation of extrarenal synthesis
of 1,25-dihydroxyvitamin D is hypercalcaemia and hypercalciuria in patients with granulomatous diseases, such as sarcoidosis. In this setting, PTH-independent extrarenal production of
1,25-dihydroxyvitamin D from 25(OH)D by activated macrophages occurs in lung parenchyma and in lymph nodes.
The plasma 1,25-dihydroxyvitamin D concentration is a
function of both the availability of 25(OH)D and the activities
of the renal enzymes 1-α-hydroxylase and 24-α-hydroxylase.
The renal 1-α-hydroxylase enzyme is primarily regulated by
the following factors:
Table 1. Vitamin D and derivatives
The established vitamin D compounds
Parent compound
Synonym
Product of first hydroxylation
Synonym
Product of second hydroxylation
Synonym
The newer vitamin D analogues
Full term
Synonym
Full term
Synonym
P O L A R V I E W S I N N E P H R O LO G Y
a
Vitamin D2 and derivatives
Vitamin D3 and derivatives
Vitamin D2
Ergocalciferol
25(OH)D2
Ercalcidiol
1,25-Dihydroxyvitamin D2
Ercalcitriol
Vitamin D3
Cholecalciferol
25(OH)D3
Calcidiol
1,25-Dihydroxyvitamin D3
Calcitriol
1-α-Hydroxyergocalciferol
Doxercalciferol
19-Nor-1,25-dihydroxyvitamin D2
Paricalcitola
22-Oxacalcitriol
Maxacalcitol
F6-1α,25-Dihydroxyvitamin D3
Falecalcitriol
In some literatures, paricalcitol is considered as the derivative of calcitriol.
enzyme and enhance 1,25-dihydroxyvitamin D production.
1,25-Dihydroxyvitamin D, in turn, inhibits the synthesis and
secretion of PTH, providing negative feedback regulation of
1,25-diydroxyvitamin D production. 1,25-Dihydroxyvitamin
D synthesis may also be modulated by VDRs on the cell surface;
down-regulation of these receptors may play an important role
in regulating vitamin D activation.
FGF23 inhibits renal production of 1,25-dihydroxyvitamin
D by limiting 1-α-hydroxylase activity in the renal proximal
tubule and by simultaneously increasing expression of 24-αhydroxylase and production of 24,25-dihydroxyvitamin D (an
inactive metabolite). 1,25-Dihydroxyvitamin D stimulates
FGF23, a phosphaturic hormone, creating a feedback loop.
Experimental data suggest that FGF23 decreases renal reabsorption of phosphate, and thereby counteracts the increased
gastrointestinal phosphate reabsorption induced by 1,25dihydroxyvitamin D, maintaining phosphate homeostasis.
Both 1,25-dihydroxyvitamin D and 25(OH)D are degraded
in part by hydroxylation by a 24-hydroxylase. The activity of the
24-hydroxylase gene is increased by 1,25-dihydroxyvitamin D,
which therefore promotes its own inactivation, and decreased
by PTH, thereby allowing more active hormone to be formed.
The interested reader can be directed to a number of source
references for the highly detailed and endlessly fascinating
biology of vitamin D [1].
I M P L I C AT I O N S F O R C K D P AT I E N T S O F
A B N O R M A L V I TA M I N D S TAT U S / V I TA M I N D
DEFICIENCY IN CKD
Vitamin D and its metabolites have a significant clinical role in
health, disease and especially in CKD because of their interrelationship with calcium homeostasis and bone metabolism.
Rickets (children) and osteomalacia (children and adults) due
to severe vitamin D deficiency are now uncommon except in
populations with unusually low sun exposure, lack of vitamin D
in fortified foods and malabsorptive syndromes. Subclinical
vitamin D deficiency, as measured by low serum 25(OH)D, is
by contrast very common. In the National Health and Nutrition
Examination Survey (NHANES) 2005–06, 41.6% of adult
participants (≥20 years) had 25(OH)D levels below 20 ng/mL
700
(50 nmol/L) [2]. This degree of vitamin D deficiency may contribute to the development of osteoporosis and an increased risk
of fractures and falls in older adults.
In CKD, significant vitamin D deficiency is pretty-well
ubiquitous; in a cross-sectional analysis of 825 consecutive
patients new to haemodialysis, Wolf et al. [3] reported that
78% of the cohort had some vitamin D problem [25(OH)D
<75 nmol/L], and that in ∼18%, there was true vitamin D deficiency (<20 nmol/L). Many other series in CKD, dialysis and
transplanted patients show that abnormal serum 25(OH)D concentrations are typical, again, nearly ubiquitous. Of course, where
there is deficiency, let us bring replacement (to paraphrase St
Francis of Assisi)—at least this is the typical nephrologist’s response, as we see from vitamins, iron, haemoglobin and other
parameters. But by now we know only too clearly that simply
by correcting what appear to be deficiencies, there is no guarantee of benefit at the level of an individual patient, at which level
we certainly also need to know much about safety concerns.
There are a tremendous number of uncertainties in this
highly complex field in 2016—for example, which ‘species’ of
vitamin D is best measured (or, indeed, whether we should be
measuring all of the assayable moieties)? What replacement
protocol works best—oral, intramuscular; daily, weekly, monthly,
less often? What end points should we be using to titrate the
effects and impacts? Just skeletal ones (if is so, just serum
PTH?)? What toxicity and harm signals should we be searching
for, and reacting to? What monitoring of vitamin D concentrations (quite expensive, unlike the vitamin itself ) should be
undertaken as screening and also during and after vitamin D
supplementation? I think it would be fair to comment that we
do not have the answers to any of these questions.
As a simple example, Table 1 lists the main different vitamin
D ‘species’, some of which will feature in this article by me (and
indeed by my opponent).
Patients with an estimated glomerular filtration rate (eGFR)
>30 mL/min who have no biochemical evidence for chronic kidney disease mineral and bone disorder (CKD-MBD, e.g. hyperparathyroidism, hyperphosphataemia) should have similar
vitamin D supplementation to that which patients with normal
renal function receive (so, in some cases, this is none at all, in
others, it is full repletion—again, evidence of benefit is scant).
As renal failure progresses (eGFR <30 mL/min), calcitriol (1,25-
D. Goldsmith
calcitriol, were excluded. This analysis reported improvements
in vitamin D and PTH levels associated with vitamin D supplementation, with mean differences of 19 ng/mL and −26 pg/mL,
respectively. A strikingly low incidence of hypercalcaemia and
hyperphosphataemia was reported. This is important, as we
shall see, as such complications are all too common with active
vitamin D preparations.
The efficacy of ‘native’ vitamin D supplementation on mineral metabolism and/or other parameters and outcomes (especially skeletal outcomes such as bone pain, and fracture) in
CKD has not been established [10]. To be fair, such studies
are worryingly absent from the literature for all forms of vitamin D intervention.
However, in vivo studies have added some modest support to
the in vitro observations that cholecalciferol therapy to increase
serum 25(OH)D concentrations results in upregulation of vitamin D response genes in monocytes of patients with end-stage
renal disease (ESRD), indicating that nutritional vitamin D
therapy has a biologic effect on circulating monocytes and
associated inflammatory markers in patients with ESRD [11].
After initiating treatment with any form of vitamin D, serum
calcium and phosphorus should be monitored quarterly, and
continued need for supplementation with vitamin D can be
re-evaluated annually. Seasonality in serum vitamin D concentrations (lowest in late winter, highest in late summer) is well
described and this needs to be taken into account in monitoring serum 25(OH)D levels and planning interventions. If the
serum level of corrected total calcium exceeds 10.5 mg/dL
(∼2.60 mmol/L), vitamin D therapy should certainly be discontinued. Again, my personal view on the matter.
T H E R A P E U T I C D E P LO Y M E N T O F V I TA M I N D
IN CKD—CLINICAL EXPERIENCE
The administration of calcitriol or synthetic vitamin D analogues in pre-dialysis patients is not routine in clinical practice
except perhaps late in stage 4 CKD, when serum PTH concentrations are typically monitored (and are usually significantly
raised). Among patients with stage 3–5 CKD not yet on dialysis,
I feel that the administration of such agents can only be
warranted if correction of nutritional vitamin D deficiency,
administration of calcium supplementation and control of
serum phosphate with diet and binders have been collectively
ineffective in suppressing PTH levels to within the normal
range (itself a ‘best guess’ target, and entirely a surrogate [4]).
Treatment with either calcitriol or a synthetic vitamin D
analogue should in any case not be given to pre-dialysis patients
with stage 3–5 CKD unless the serum phosphate is in the normal range and the corrected serum total calcium concentration
is ideally no higher than the mid-point of the normal range [e.g.
<9.5 mg/dL (<2.37 mmol/L)], as there is a tendency to a modest
rise in serum calcium concentrations after correction of vitamin
D deficiency. Initiation of this sort of vitamin D supplementation requires outpatient follow-up to avert the small risk of
severe hypercalcaemia, with serum calcium and phosphate
being measured monthly for 3 months, and then at least
every 3 months. If the serum level of corrected total calcium
Evidence for the usefulness of using vitamin D to treat ‘renal
bone disease’ is now nearly six decades old [5]. In regular clinical practice, however, it is more like three decades, at most, and
even less time than that if we consider just interventions in CKD
patients not yet on dialysis.
It is disappointingly true to say that even in 2016 there is a
paucity of evidence concerning the clinical benefits of vitamin
D supplements to treat vitamin D insufficiency in patients with
stage 3b–5 CKD not yet on dialysis. While there are a number of
studies that report the impact of vitamin D supplementation on
serum vitamin D concentrations (unsurprisingly, usually
reporting an increase), there has been much less focus on
hard or semi-rigid clinical end point analysis (e.g. fractures,
hospitalizations). A few studies have found that vitamin D repletion therapy tends to normalize serum 25(OH)D levels and
also modestly to decrease PTH levels [6–8]. Some of the better
data are from a meta-analysis that included nine observational
studies comprising 286 non-dialysis patients [9]. Studies that
assessed vitamin D preparations at low doses (400–800 IU/day)
and those that assessed active vitamin D derivatives, such as
C A L C I T R I O L A N D S Y N T H E T I C V I TA M I N D
A N A LO G U E S
Vitamin D deficiency/insufficiency in CKD patients
701
P O L A R V I E W S I N N E P H R O LO G Y
dihydroxyvitamin D) production may be low due to diminished
glomerular filtration, loss of the 1-α-hydroxylase enzyme secondary to structural renal compromise and suppression of enzyme
activity secondary to hyperphosphataemia and rising FGF23 concentrations. The net result is a tendency for hypocalcaemia, secondary hyperparathyroidism (biochemically, as elevation of
serum PTH concentrations) and bone disease.
Here is perhaps is the ideal place to rehearse my well-known
and strong prejudice against over-easy, intellectually lazy, reliance on ‘PTH elevation’ as a substitute for careful evaluation of
a patient’s individual risk of complications and harm, either
from a disease or from its treatment. For now, we will both
(my opponent and I) have to accept for the purposes of this academic exercise that a chronic trend rise in serum PTH is a
marker of potential clinically relevant hyperparathyroidism.
Though this may simply not be the case, in certain situations.
As we have seen, 25(OH)D deficiency is a very common
finding in pre-dialysis and dialysis patients, is associated with
elevated PTH levels and may worsen the manifestations of secondary hyperparathyroidism in this setting. It seems reasonable
to me, therefore, that for those with stage 3–5 CKD not yet
on dialysis and with elevated plasma intact PTH, treatment
with ergocalciferol or cholecalciferol be initiated if co-incident
vitamin D deficiency exists. ‘Deficiency’ in this setting is demonstrated—in my personal view—by a 25(OH)D (calcidiol)
level of <20 ng/mL (50 nmol/L), and we have little current
choice but to accept a PTH in the normal range (as promulgated
by KDIGO in 2012) as being our physiological target, even
though to many, this seems over-zealous [4]. Of course, variation in the definitions and applicability of different target
serum 25(OH)D ranges—optimal, ideal, normal, insufficient and
deficient—only serve further to muddy already rather turbid
waters [1].
P O L A R V I E W S I N N E P H R O LO G Y
exceeds 10.5 mg/dL (2.60 mmol/L) after replacement therapy
has been started, all forms of vitamin D therapy should ideally
be discontinued (my opinion). Vitamin D therapy should also
be discontinued if intact PTH levels become persistently low
(trickier to define, but certainly ‘low’ values would be at or
near the lower end of the normal range), for fear of driving
even harder the modern epidemic of adynamic bone disease,
which owes much of its prevalence to chronic, widespread
and mechanized overuse of vitamin D compounds.
The comparative biological effects of calcitriol or the different synthetic vitamin D analogues in pre-dialysis patients with
CKD have not been established with certainty. This is in terms
of PTH reduction, or tendency to hypercalcaemia. As a result,
any one of the available oral agents (calcitriol, alfacalcidol, doxercalciferol or paricalcitol) may be administered.
Benefits with active vitamin D derivatives in patients with
mild to moderate chronic renal failure and secondary hyperparathyroidism have been shown in placebo-controlled randomized trials (RCTs). One of the earliest, and best, studies in this
field was done in the 1980s/1990s [12, 13], well before the modern era. In a prospective multicentre study of 176 patients [12]
with a creatinine clearance between 15 and 50 mL/min, 75%
had histologic evidence of high-turnover bone disease at baseline. Patients were randomly assigned to placebo or to the synthetic vitamin D analogue, alfacalcidol (1-hydroxyvitamin D),
at a dose of 0.25 μg/day, increasing to a maximum of 1 μg/day.
The aim of therapy was to raise the serum calcium concentration to the upper limit of normal for the laboratory [mean 9.8–
10 mg/dL (2.45–2.50 mmol/L)], not in fact primarily to reduce
the serum PTH concentration (so a different clinical algorithm
from one we would use today).
After at least 2 years of follow-up, the following findings were
reported:
• Plasma PTH declined during the first 6 months of alfacalcidol therapy and then rebounded to pretreatment levels. In
comparison, there was more than a 2-fold increase in
PTH levels in the placebo group.
• Repeat bone biopsy showed improvement in 29% of patients
receiving alfacalcidol, whereas the bone disease worsened in
90% of placebo-treated patients.
• In patients without pre-existing histologic evidence of parathyroid bone disease, there was no evidence of progression
in either group.
• There was no difference in the rate of progression of renal
failure.
•
Hypercalcaemia, treated with dose reduction, was more
common in the alfacalcidol group (11 versus 3%).
Note that here bone biopsies were used to confirm the presence
of a real histological bone ‘disease’, in marked contrast to so
many later trials where predominantly shorter study time
points were employed, and also only using convenient surrogate biochemical end points.
Oral paricalcitol has also been examined in randomized
trials [14, 15]. In a Phase III trial of 220 patients with stage 3
and 4 CKD, compared with placebo, paricalcitol resulted in a
702
significant percentage of patients with at least two consecutive
decreases in PTH levels of ≥30% (91% on paricalcitol versus
13% on control therapy). Both groups had similar incidences
of hypercalcaemia, hyperphosphataemia and elevated calcium–
phosphorus products.
T W O R E C E N T S T U D I E S H AV E B E E N
U N D E R TA K E N T O A D D R E S S T H E S P E C I F I C
I M P AC T S O F U S I N G N U T R I T I O N A L
V I TA M I N D S U P P L E M E N TAT I O N I N C K D :
ARE THESE THE KILLER-PUNCH, THE
K N O C K - O U T B LO W ?
The first of these two recent (2015) trials was of the use of
ergocalciferol to answer the stated problem—namely that
while vitamin D [25(OH)D] deficiency is common in patients
initiating long-term haemodialysis, the safety and efficacy of
nutritional vitamin D supplementation in this population remains uncertain [16]. This comment/rationale was that which
the authors of the study used to justify their work—it is not really a valid uncertainty, at least in my opinion.
This randomized, placebo-controlled, parallel-group multicentre trial compared two doses of ergocalciferol with placebo
between October 2009 and March 2013. Haemodialysis
patients (n = 105) with 25(OH)D levels ≤32 ng/mL from 32
centres in the Northeast USA were randomly assigned to oral
ergocalciferol, 50 000 IU weekly (n = 36) or monthly (n = 33),
or placebo (n = 36) for a 12-week treatment period. The primary end point was the achievement of vitamin D sufficiency
[25(OH)D >32 ng/mL] at the end of the 12-week treatment
period. Survival was assessed through 1 year [16].
Baseline characteristics were similar across all arms, with
overall mean ± standard deviation (SD) 25(OH)D levels of
21.9 ± 6.9 ng/mL. At 12 weeks, vitamin D sufficiency [25
(OH)D >32 ng/mL] was achieved in 91% (weekly), 66%
(monthly) and 35% ( placebo) (P < 0.001). Mean 25(OH)D
was significantly higher in both the weekly (49.8 ± 2.3 ng/mL;
P < 0.001) and monthly (38.3 ± 2.4 ng/mL; P = 0.001) arms
compared with placebo (27.4 ± 2.3 ng/mL). Calcium, phosphate, PTH levels and active vitamin D treatment did not differ
between groups. All-cause and cause-specific hospitalizations
and adverse events were similar between groups during the
intervention period. The conclusions from this 2015 study
[16] were principally that oral ergocalciferol could increase
serum 25(OH)D levels in incident haemodialysis patients without significant alterations in blood calcium, phosphate or PTH
concentrations during a 12-week period [16].
Notwithstanding the excellence of the collaborative group
involved here, or the impressive three-way ‘high-low-no dose’
regime employed, the use of a 12-week assessment is barely
credible, when relating the impact of this intervention to matters such as infection, bone fractures, cardiovascular events and
overall mortality. Let alone the propensity to develop hypercalcaemia. Twelve months should be the shortest possible time to
consider in a study of this type, and much better still, 3–5 years
[noting that the dropout rate of potential trial participants in
D. Goldsmith
There have been a number of modestly sized cohorts, RCTs
and >8-week exposure vitamin D administration trials (reviewed nicely in [17]). These have many obvious shortcomings:
for example, the absence of any thought to consider what the
optimal serum 25(OH)D concentration might actually need
to be to effect a suitable reduction in serum intact PTH. It
has been assumed, on the basis of no evidence, just assertion,
that this must be ‘>75 nmol/L’ but what in fact if it were to be
‘>100 nmol/L’ or even ‘>150 nmol/L’? These concentrations are
safe, at least in many patient groups, and might be more
effective (and yes, maybe less safe) than the conventional target
we use.
C O M P A R I S O N S W I T H I N A S I N G L E S T U DY
C O M P A R I N G N AT I V E V I TA M I N D W I T H
SYNTHETIC VDRA IN THE SUPPRESSION
OF PTH
Published high-quality comparisons between ‘native’ or ‘natural’ and ‘synthetic’ vitamin D compounds in the treatment
of CKD-related bone disease (or more accurately, and often,
elevated serum PTH) are far too few in number to allow for
any confident elucidation of the relative merits and demerits
of both approaches—certainly in terms of the benefits of treatment on real skeletal or other clinically relevant end points.
In one randomized, open-label trial, 80 patients with CKD
(eGFR 15–60 mL/min/1.73 m2) were assigned to receive either
paricalcitol (1 or 2 μg/day) or ergocalciferol (50 000 IU weekly,
titrated to serum vitamin D concentrations >30 ng/mL) [19].
At 16 weeks of follow-up, more patients assigned to paricalcitol
achieved a 30% reduction in PTH compared with patients assigned to ergocalciferol (53 versus 18%, respectively). The
rates of hypercalcaemia and hyperphosphataemia were not different between groups. However, this type of trial is only too
reminiscent of the worst days of erythropoiesis-stimulating
agent (‘ESA’) trials, where ESA ‘A’ was pitted against ESA ‘B’
to reveal a slew of irrelevant but often wildly over-interpreted
differences in surrogate outcomes. Not very valuable information, except, perhaps, for the trial organizers. For example, as
with the previous 12-week study duration [16], just 16-week
therapy in this study is without common sense as a reasonable
time frame to correct a biochemical abnormality that likely has
taken months to years to develop.
P L E I O T R O P I C E F F E C T S O F V I TA M I N D —
E X T R A - S K E L E TA L E F F E C T S ( T H E
‘ E V E R Y T H I N G E L S E ’ F AC T O R )
Perhaps if we were prepared, or able, to start the correction of
vitamin D deficiency gently and early, using nutritional vitamin
D from CKD stages 3a and 3b, and not leaving this matter until
there is significant PTH elevation/resistance, the bone pathology associated with progressive CKD would be easier to manage (and patients would also have any benefit of pleiotropic
benefits—remembering that probably two-thirds of deaths in
CKD stage 3a are cardiovascular, not renal in origin).
Vitamin D deficiency/insufficiency in CKD patients
703
P O L A R V I E W S I N N E P H R O LO G Y
these types of trials is truly prodigious, from death (10–20% per
annum) and from organ transplantation].
While this was a larger than usual trial of this type, and that
too is to be commended, and some of the reported end points
were potentially relevant, and not simply biochemical (even if
here the biochemical end point was the primary outcome), the
very short duration of interventional therapy in this study
makes absolutely no biological sense to me.
The second recent study by Miskulin et al. [17] was a doubleblind, placebo-controlled, randomized clinical trial to assess the
effects of supplementation with ergocalciferol on epoetin utilization and other secondary outcomes in patients on haemodialysis with serum 25(OH)D <30 ng/mL. In all, 276 patients
were randomized to 6 months of ergocalciferol or placebo.
Mean ± SD serum 25(OH)D increased from 16 ± 5.9 ng/mL
at baseline to 39 ± 14.9 ng/mL in the ergocalciferol arm and
did not change (16 ± 6.4 and 17 ± 7.4 ng/mL, respectively) in
the placebo arm. There was no significant change in epoetin
dose over 6 months in the ergocalciferol or placebo arms and
no difference across arms (P = 0.78). No change occurred in
serum calcium, phosphorus, intact PTH or C-reactive protein levels, cinacalcet use, or phosphate binder or calcitriol dose
in either study arm. Rates of all-cause, cardiovascular and
infection-related hospitalizations did not differ by study arm, although statistical power was limited for these outcomes. The
authors concluded that 6 months of supplementation with ergocalciferol increased serum 25(OH)D levels in patients on haemodialysis with vitamin D insufficiency or deficiency, but had no
effect on epoetin utilization or secondary biochemical and clinical outcomes. This study was accompanied by a trenchant, almost apocalyptic, editorial by Graham Elder [18]—indeed, on
reading both the paper and the accompanying editorial, you
might think there was not ‘mush room’ for doubt on this topic,
so much so that some might think that I should accordingly leave
the field of battle waving a white flag of surrender [18].
However, I disagree! The study population was not a virgin
one at all. This second (2016) study was not a cross-over study,
but one where ‘therapy B’ (ergocalciferol) was overlain upon
therapy A (existing CKD-MBD treatments) [17]. The patients
had been on dialysis, and most of the active treatments associated with dialysis, for 3.5 years. In particular, 80–90% of patients were already chronically treated with VDR activators
(VDRAs), 90% were on oral phosphate binders and 25% were
on cinacalcet, and their starting serum PTH concentrations
were ∼450 pg/mL—so really quite well controlled ‘secondary
hyperparathyroidism’ (if they truly had this). There was no
‘wash out’ period between the end of the clinical and the start
of trial-based supervision periods. In practice, skeletal effects
would need at the very least 2–4 months (based on changes in
serum bone-specific alkaline phosphatase concentrations) of
withdrawal before subsequent or additional specific impacts
from ergocalciferol could reliably be determined. This trial
should have been conducted over a minimum of 12 and better,
24–36 months, to have any real validity.
In fact, I think that this trial, well-executed though it might
have been, adds very little indeed to the state of knowledge,
and nothing about the possible utility of 25(OH)D repletion
in dialysis except that it is easy, cheap and safe to achieve it.
P O L A R V I E W S I N N E P H R O LO G Y
It is often asserted, but not yet proven in a properly
conducted RCT, that vitamin D supplementation is associated
with better survival, in any patient cohort. Effects on endothelial function, micro-circulation, arterial stiffness, blood pressure
parameters, left ventricular hypertrophy (LVH) and function,
numerous proxies for inflammation, glucose tolerance, insulin
sensitivity, acute and chronic infection rates (bacterial, viral,
other) and acute asthma have all been reported [1]—but as
yet, these glimmers are all too readily extinguished by the icy
blasts of scepticism.
Several RCTs are currently investigating the potential effect of
nutritional vitamin D on LVH (NCT01323712), insulin resistance (NCT00893451), erythropoietin dosing (NCT01395823),
proteinuria (NCT01426724), immunity (NCT00892099), arteriovenous fistula maturation (NCT00912782), and physical
and cognitive performance (NCT00511225, NCT01229878).
We will not be able to make a good judgement about this
critically important point until we have the results from a number of well-conducted epidemiological trials, examining overall,
cardiovascular and bone-related events. The VITAL study from
the USA is the largest, and most eagerly awaited, of these megatrials [1, 20]. Many of these trials are due to report in the next
24 months, and on these findings, we need to rest our case for
vitamin D supplementation across the board, or, just selective
patient treatment, or possibly even reservation of treatment
for existing actual disease states (such as bone diseases) rather
than just for a cluster of high-risk factors.
CONCLUSIONS
It is quite depressing seeing how little we actually know about
the genesis, implications and potential impacts of vitamin D
deficiency, and treatment, in CKD, even though for some decades, we have been busy deploying various vitamin D species in
an effort to render certain biochemical analytes more acceptable to scrutiny. Certainly, there is little evidence of true patientlevel benefit, in what is clearly a much more complex paradigm
than we perhaps had ever imagined it to be.
Thus, in the spirit of primum non nocere, knowing first
the propensity of active vitamin D compounds, especially
deployed in large doses over long periods of time, to engender
their own biochemical consequences—hypercalcaemia,
hyperphosphataemia—and potentially also accelerate vascular
calcification [21, 22], I think it is more prudent to start the
fight against ‘bone disease’ using cheap, natural and safe interventions, such as ergocalciferol and cholecalciferol, at the very
first signs of PTH elevation (CKD stage 3a/b). This might well
mandate earlier (in CKD terms) testing of serum PTH calcium
and 25(OH)D concentrations. This would need rigorous clinical testing of course, in different ways. I am happy to concede
that, with appropriate attempts to control phosphate, calcium
and PTH using diet, dialysis (where needed), phosphate binders
and ‘natural’ vitamin D species, if the PTH is still deranged, and
we feel compelled to continue to try to reduce its serum concentration, we can then swap the use of natural vitamin D species
for one of the many alien, unnatural, ones.
704
In the end, neither my opponent nor I has been able to administer a knock-out blow on the other. Neither of us has the evidence
at our disposal to do that. This is more like the First World War
than it is about the Nuclear Age. Progress is slow, grindingly so at
times, the terrain is unpromising, and the ‘promised land’ totally
out of sight. I truly believe in fact that we need to adopt a different
approach to future trial design for this clinical paradigm—replicating the complex sequential series of interventions we use in the
clinical situation—optimization of phosphate management, careful but sustained repletion of 25(OH)D, good dialysis for those
patients in CKD 5D—selecting ‘real’ patients, across many
units, and following their clinical progress in the real world to delineate real clinical outcomes of relevance—randomization under
these circumstances is still possible, and of course, so is stratification for a number of potential confounders. We just first have to
stop pretending that we have at our disposal a single miraculous
intervention—be it drug ‘A’, ‘B’ or ‘C’—for a ‘simple’ clinical
problem (elevated serum PTH), and that this wonder drug can
be given over a short space of time, has no side effects, and
thus that it is the nostrum, the panacea, the holy grail de nos jours.
C O N F L I C T O F I N T E R E S T S TAT E M E N T
D.G. has received speaking and consulting fees from Abbvie,
Amgen, Genzyme, Sanofi and Shire.
(See related articles by Agarwal and Georgianos. Con: Nutritional vitamin D replacement in chronic kidney disease and
end-stage renal disease. Nephrol Dial Transplant 2016; 31:
706–713; Zoccali and Mallamaci. Moderator’s view: Vitamin
D deficiency treatment in advanced chronic kidney disease: a
close look at the emperor’s clothes. Nephrol Dial Transplant
2016; 31: 714–716)
REFERENCES
1. Leckstrom DCT, Salzer J, Goldsmith DJA. The trials and tribulations of
vitamin D: time for the ‘sunshine’ vitamin to come in out of the cold—or
just more broken promises? Expert Rev Endocrinol Metab 2014; 9: 327–344
2. Forrest KY, Stuhldreher WL. Prevalence and correlates of vitamin D deficiency in US adults. Nutr Res 2011; 31: 48
3. Wolf M, Shah A, Gutierrez O et al. Vitamin D levels and early mortality
among incident hemodialysis patients. Kidney Int 2007; 72: 1004–1013
4. Kidney Disease: Improving Global Outcomes (KDIGO) CKD-MBD Work
Group. KDIGO clinical practice guideline for the diagnosis, evaluation,
prevention, and treatment of Chronic Kidney Disease-Mineral and Bone
Disorder (CKD-MBD). Kidney Int Suppl 2009; 113: S1–S130
5. Stanbury SW. Azotaemic renal osteodystrophy. Br Med Bull 1957; 13: 57–60
6. Al-Aly Z, Qazi RA, González EA et al. Changes in serum 25-hydroxyvitamin D and plasma intact PTH levels following treatment with ergocalciferol
in patients with CKD. Am J Kidney Dis 2007; 50: 59
7. Zisman AL, Hristova M, Ho LT et al. Impact of ergocalciferol treatment of
vitamin D deficiency on serum parathyroid hormone concentrations in
chronic kidney disease. Am J Nephrol 2007; 27: 36
8. Kooienga L, Fried L, Scragg R et al. The effect of combined calcium and
vitamin D on serum intact parathyroid hormone in moderate CKD. Am
J Kidney Dis 2009; 53: 408
9. Kandula P, Dobre M, Schold JD et al. Vitamin D supplementation in chronic kidney disease: a systematic review and meta-analysis of observational
studies and randomized controlled trials. Clin J Am Soc Nephrol 2011; 6: 50
D. Goldsmith
10. Kramer H, Berns JS, Choi MJ et al. 25-Hydroxyvitamin D testing and supplementation in CKD: an NKF-KDOQI controversies report. Am J Kidney
Dis 2014; 64: 499
11. Stubbs JR, Idiculla A, Slusser J et al. Cholecalciferol supplementation alters
calcitriol-responsive monocyte proteins and decreases inflammatory cytokines in ESRD. J Am Soc Nephrol 2010; 21: 353
12. Hamdy NA, Kanis JA, Beneton MN et al. Effect of alfacalcidol on natural
course of renal bone disease in mild to moderate renal failure. BMJ 1995;
310: 358
13. Rix M, Eskildsen P, Olgaard K. Effect of 18 months of treatment with
alfacalcidol on bone in patients with mild to moderate chronic renal failure.
Nephrol Dial Transplant 2004; 19: 870
14. Coyne D, Acharya M, Qiu P et al. Paricalcitol capsule for the treatment of
secondary hyperparathyroidism in stages 3 and 4 CKD. Am J Kidney Dis
2006; 47: 263
15. Kovesdy CP, Lu JL, Malakauskas SM et al. Paricalcitol versus ergocalciferol
for secondary hyperparathyroidism in CKD stages 3 and 4: a randomized
controlled trial. Am J Kidney Dis 2012; 59: 58.
16. Bhan I, Dobens D, Tamez H et al. Nutritional vitamin D supplementation in dialysis: a randomized trial. Clin J Am Soc Nephrol 2015; 10: 611–619
17. Miskulin DC, Majchrzak K, Tighiouart H et al. Ergocalciferol supplementation in hemodialysis patients with vitamin D deficiency: a randomized
clinical trial. J Am Soc Nephrol 2015; Epub ahead of print
18. Elder G. Mushroom clouds for vitamin D? J Am Soc Nephrol 2016. doi:
10.1681/ASN.2015111279
19. Kovesdy CP, Lu JL, Malakauskas SM et al. Paricalcitol versus ergocalciferol
for secondary hyperparathyroidism in CKD stages 3 and 4: a randomised
controlled trial. Am J Kidney Dis 2012; 59: 58–66
20. http://www.vitalstudy.org/ (8 March 2016, date last accessed)
21. Zhang Q, Li M, Zhang T et al. Effect of vitamin D receptor activators on
glomerular filtration rate: a meta-analysis and systematic review. PLoS
One 2016; 11: e0147347
22. Brown RB, Haq A, Stanford CF et al. Vitamin D, phosphate, and vasculotoxicity. Can J Physiol Pharmacol 2015; 93: 1077–1082
Received for publication: 21.3.2016; Accepted in revised form: 21.3.2016
Nephrol Dial Transplant (2016) 31: 705 –
doi: 10.1093/ndt/gfw082a
Rajiv Agarwal1 and Panagiotis I. Georgianos2
1
Department of Medicine, Indiana University School of Medicine and Richard L. Roudebush Veterans Administration Medical Center,
Indianapolis, IN, USA and 2Division of Nephrology and Hypertension1st Department of Medicine, AHEPA Hospital, Aristotle University of
Thessaloniki, Greece
Correspondence and offprint requests to: Rajiv Agarwal; E-mail: [email protected]
Nutritional vitamin D is cheap and natural, but it is as good as
administering a placebo to treat chronic kidney disease (CKD) patients with secondary hyperparathyroidism. Unlike the general
population, people with CKD do not have the enzymes to activate
vitamin D. If CKD patients had the activating enzymes, there
would be little reason not to repair the deficiency of vitamin D
first with native vitamin D. I agree with my opponent that vitamin
D serum levels are corrected upon supplementation of vitamin D;
however, the deficiency remains uncorrected. The objective of
vitamin D supplementation in CKD is not to simply restore plasma levels of this substance, but to rectify the manifestations of
vitamin D deficiency. Our meta-analysis shows that the native
vitamin D has no meaningful effect on parathyroid hormone
(PTH). This is the most ubiquitous manifestation of vitamin D
deficiency; yet native vitamin D supplementation cannot repair
it. It is no surprise that vitamin D supplementation has no effect
on other putative outcomes such as inflammation, insulin resistance, infections and anaemia management in those with CKD.
Thus, in CKD, vitamin D supplementation restores the levels,
but not the deficiency.
My opponent states: ‘among patients with stage 3–5 CKD not
yet on dialysis, I feel that the administration of such agents can
© The Author 2016. Published by Oxford University Press
on behalf of ERA-EDTA. All rights reserved.
only be warranted if correction of nutritional vitamin D
deficiency, administration of calcium supplementation and control of serum phosphate with diet and binders have been collectively ineffective in suppressing PTH levels to within the normal
range’. We do not wait for the blood pressure to fall with dietary
sodium restriction, nor wait for cholesterol to decline with diet
and exercise before using drug therapies. If we want to prevent
secondary hyperparathyroidism, we should measure PTH and
not wait until the patient reaches dialysis. Neither should we
wait till all other measures described by my opponent have failed.
My opponent states: ‘I prefer a more nuanced, subtle,
approach to the same problem, before deploying my heavier artillery, and only if I need to. This is indeed one of the (many)
distinguishing features between America and Great Britain.’ In
treating our patients, we should not be guided in our actions
by the British way or the American way, but the right way.
Activated vitamin D is approved for the prevention and treatment of secondary hyperparathyroidism, is evidence based and
supported by science. Until such time that we perform a randomized trial to show independent benefit of native vitamin D supplementation, among those with manifestations of vitamin D
deficiency and CKD, using activated vitamin D is the right way.
705
P O L A R V I E W S I N N E P H R O LO G Y
Opponent’s comments