http://www.kidney-international.
org
& 2011 International Society of Nephrology
see commentary on page 1016
Short-term vitamin D receptor activation increases
serum creatinine due to increased production with
no effect on the glomerular filtration rate
Rajiv Agarwal1, Jennifer E. Hynson1, Tyler J.W. Hecht1, Robert P. Light1 and Arjun D. Sinha1
1
Department of Medicine, Indiana University School of Medicine and Richard L. Roudebush Veterans Administration Medical Center,
Indianapolis, Indiana, USA
Vitamin D receptor activation has been associated with
increased serum creatinine and reduced estimated
glomerular filtration rates, raising concerns that its use may
be detrimental to kidney function. Here we studied the
effect of vitamin D receptor activation on serum creatinine,
creatinine generation, and its clearance. We measured
baseline serum creatinine and 24-h urine creatinine in
16 patients with chronic kidney disease. The measurements
were repeated every day for 7 days, during which time the
patients received 2 lg paricalcitol, an orally active vitamin D
receptor activator, every morning. At 4 days after stopping
the vitamin analog, measurements were continued for
3 days. Geometric mean parathyroid hormone levels
decreased from 77 pg/ml at baseline to 43 pg/ml at the end
of treatment and significantly rebounded to 87 pg/ml
following paricalcitol withdrawal, thereby supporting the
biological efficacy of the analog dose used. With this therapy,
the serum creatinine significantly increased at a rate of
0.010 mg/dl/day and urine creatinine at a rate of 17.6 mg/day.
Creatinine and iothalamate clearances did not change,
whereas urine albumin decreased insignificantly. Thus, short-
term vitamin D receptor activation increases creatinine
generation and serum creatinine, but it does not influence
the glomerular filtration rate.
Kidney International (2011) 80, 1073–1079; doi:10.1038/ki.2011.207;
published online 29 June 2011
KEYWORDS: albuminuria; chronic kidney disease; creatinine clearance; GFR;
serum creatinine; vitamin D
Correspondence: Rajiv Agarwal, Department of Medicine, Indiana
University and VAMC, 1481 West 10th Street, Indianapolis, Indiana 46202,
USA. E-mail: ragarwal@iupui.edu
Received 14 April 2011; revised 5 May 2011; accepted 10 May 2011;
published online 29 June 2011
Kidney International (2011) 80, 1073–1079
original article
For vitamin D to exert its endocrine effects, it is critically
important for the kidney to activate both nutritional and
sunlight-derived vitamin D.1 In the course of kidney disease,
the activation of vitamin D is abrogated; this occurs early in the
course of kidney disease, leads to reduction in calcium
absorption from the gut, increase in parathyroid hormone
(PTH) levels from the parathyroid gland, and culminates in
secondary hyperparathyroidism. To countervail these effects,
calcitriol, a vitamin D receptor (VDR) activator, has been
synthesized and administered to treat secondary hyperparathyr-
oidism among patients with chronic kidney disease (CKD).
Among patients with CKD, some unexpected benefits have
been observed. These relate to the antiproteinuric effects
of VDR activators that raise the possibility of renoprotection
in the long term.2–4 The best evidence for this derives from
the recently reported VITAL (Selective Vitamin D Receptor
Activator for Albuminuria Lowering) study in which
281 patients were studied and had measurements of
estimated glomerular filtration rate (eGFR) at baseline and
periodically over 6 months.5 The study demonstrated that
the VDR activator, paricalcitol, reduces albuminuria in
patients with CKD with diabetic nephropathy.6 Among these
patients, those randomized to high-dose (2 mg) paricalcitol
experienced a decline in eGFR.6 Although this decline in
eGFR reversed after stopping the drug, it raises concerns
of nephrotoxicity. If reduction in albuminuria was to be
accompanied by a true decline in GFR, then it can be
debated whether this drug would be useful in retarding
the progression of kidney disease.
There is some evidence that VDR activators may increase
serum creatinine and, consequently, the appearance of
a decline in eGFR.7 It is unclear whether the rise in serum
creatinine is because of a true decline in GFR or because of
alteration in creatinine handling, either through reduction
in secretion of creatinine or via increased creatinine
generation.8,9 Two studies report conflicting mechanisms.
One study suggests that VDR activators may reduce the
tubular secretion of creatinine.8 Another study suggests an
increase in creatinine production rates.9 Both these studies
evaluated patients with a wide range of GFR. Moreover, they
were not designed to directly examine creatinine generation
1073
original article R Agarwal et al.: Paricalcitol increases serum creatinine but not GFR

93 were screened Table 1 | Characteristics of the study population

Clinical characteristic n (%) or mean (s.d.)
32 were not eligible
N 16 (100%)
61 were eligible Male 16 (100%)
45 did not consent
Race
16 consented White 12 (75%)
Black 3 (19%)
American Indian 1 (6%)
16 were enrolled Age (years) 69.6±9.8
Weight (kg) 97.6±35.6
2 subjects missed 3 visits out of 224 Height (cm) 166.7±26.5
(98.6% complete data) BMI (kg/m2) 29.9±3.3

0 doses of study medication were missed Etiology of CKD

Adult autosomal polycystic kidney disease 1 (6%)
Diabetes mellitus 5 (31%)
16 completed Hypertensive nephrosclerosis 6 (38%)
Ischemic nephropathy 1 (6%)
Figure 1 | Study flow. There was 100% compliance with the study Unknown 3 (19%)
medication, and 98.6% visits were completed over the duration of
the trial. One patient at day 7 had missing glomerular filtration Stage of CKD
rate (GFR) measurements due to technical issues with the infusion Stage III 12 (75%)
pump. Stage IV 4 (25%)

History of smoking
Never 2 (13%)
and excretion over a period of time when no change in true Past 11 (69%)
GFR is expected. Neither of these studies examined the Current 3 (19%)
reversibility of GFR change by stopping the VDR activator. The
results of these studies are unlikely to fully explain the VITAL History of
Myocardial infarction 6 (38%)
study results because in the VITAL study, patients had moderate Heart failure 2 (13%)
CKD and received a newer VDR activator, paricalcitol. Stroke 4 (25%)
In this study, we examined the notion that the VDR Diabetes mellitus 12 (75%)
Revascularization 8 (50%)
activator, paricalcitol, alters creatinine metabolism and has
Albumin (g/dl) 3.6±0.4
no effect on GFR. Furthermore, we evaluated whether these Creatinine (mg/dl) 1.9±0.4
effects were reversed upon stopping the drug. Hemoglobin (g/dl) 12.9±1.7
Estimated GFR (ml/min per 1.73 m2) 36.9±10.3
RESULTS Measured GFR (ml/min per 1.73 m2) 47.8±17.1
Calcium (mg/dl) 9.0±0.3
A total of 93 patients were assessed for eligibility, out of Phosphorous (mg/dl) 3.3±0.5
which 32 were excluded (Figure 1). Of the 61 eligible patients, PTH-intact (pg/dl) 90.3 (45.7, 101.7)
45 did not consent. Of the 16 patients who consented, Abbreviations: BMI, body mass index; CKD, chronic kidney disease; GFR, glomerular
all were enrolled and of the 224 possible visits, 221 (98.6%) filtration rate; PTH, parathyroid hormone.

were completed. Iothalamate GFR measurements were

available for all patients except one patient. In this patient,
on day 7, because of technical problems with the insulin
pump, only approximately two-thirds of the iothalamate
dose was delivered. This was discovered on examining the log
of the infused amount. Accordingly, this data point was
censored. Two subjects missed three visits because of winter
weather, but even these patients received the study medica-
tion, and hence no doses of medications were missed.
The baseline characteristics of the patients are listed in
Table 1. As may be expected of a veteran population,
all patients were older men. Although 12 patients had diabetes
mellitus, the etiology of kidney disease was thought to be
diabetic nephropathy in only five patients. Baseline eGFR was
36.9 ml/min per 1.73 m2, whereas measured iothalamate
clearance was 47.8 ml/min per 1.73 m2. Stage 3 CKD was
present in 12 patients and stage 4 in four patients. Median
PTH-intact was 90.3 pg/ml (interquartile range 45.7–101.7).
Figure 2 shows the decline in PTH-intact. The geometric
least square mean PTH declined from 77.2 to 43.2 pg/ml at
1 week after taking 2 mg paricalcitol. At 1 week after stopping
the drug, the PTH-intact level increased to 87.2 pg/ml or
near-baseline level. Each patient showed a decline in PTH
with a rebound in PTH after stopping the drug, making this
effect highly significant (Po0.0001). Least square mean
iothalamate GFR (right panel, Figure 2) was 49.1 ml/min per
1.73 m2, which was unchanged at 47.7 ml/min per 1.73 m2
after taking 2 mg paricalcitol for 7 days. After washout,
the GFR was unchanged at 47.0 ml/min per 1.73 m2.
Figure 3 shows the changes in serum and urine calcium
and phosphorus. For both calcium and phosphorus, the
urinary excretions and serum concentrations showed a
significant upward trend during exposure to the study drug,

1074 Kidney International (2011) 80, 1073–1079

R Agarwal et al.: Paricalcitol increases serum creatinine but not GFR original article

followed by a return to baseline after a week-long washout 0.05 mg/dl/day, urine calcium 10.2 mg/day, and urine phos-
period. Linear increases were significant and were as phorus 23.3 mg/day (all Po0.001). The mean level at
follows: serum calcium 0.055 mg/dl/day, serum phosphorus washout was not significantly different from baseline in any
instance. It is notable that exposure to a single dose of
250 100 paricalcitol caused the urine phosphorus to drop before
returning to baseline on the third day. This may be related to
200
initial suppression of PTH, followed by delayed increase in
80 intestinal absorption of phosphorus.
iGFR (ml/min per 1.73 m2) Figure 4 shows changes in serum creatinine, 24-h urine
PTH-intact (pg/ml)

150
100
77.2
50 43.2
0 20
Baseline Day 7
Visit
Figure 2 | Changes in PTH and iGFR over the course of the
study.The left graph shows the parathyroid hormone (PTH-intact)
concentrations from baseline visit to day 7 (final day of treatment)
and to day 14 (7 days after taking the last pill of paricalcitol). PTH
dropped 44% after 1 week of paricalcitol, but rebounded to
baseline 1 week after terminating the treatment. Dashed lines
show individual patients, whereas the solid line shows the
geometric modeled mean with its 95% confidence interval. Each
patient shows a decline in PTH when taking the treatment and
each shows a rebound after stopping therapy (Po0.0001). The
right graph shows the measured iothalamate GFR (iGFR) over the
trial. The solid line shows the arithmetic modeled mean with its
95% confidence interval. No paricalcitol-induced change in iGFR
was observed.
creatinine, and creatinine clearance (left panel) and changes
60
in serum urea nitrogen, 24-h urine urea nitrogen, and urea
87.2 49.1 47.7
clearance (right panel). A small but statistically significant
47.0
change in serum creatinine was noted averaging 0.01 mg/dl/
40
day. This was accompanied by an increase in 24-h urine
creatinine excretion rate averaging 17.6 mg/day. The creati-
nine clearance did not change. A small increase in serum urea
nitrogen was also seen. However, this was not accompanied
Day 14 Baseline Day 7 Day 14
by a change in urine urea nitrogen or urea clearance.
Visit
The pattern of change in serum urea nitrogen and urine urea
nitrogen was U-shaped. Therefore, we fitted a quadratic
model. This model confirmed the U-shaped nature of the
relationship for both serum urea nitrogen and 24-h urine
urea excretion rates (Supplementary Figure S1 online, please
see Supplementary data).
Supplementary Figure S2 online shows seated clinic blood
pressure and 24-h albumin excretion rate. The urine albumin
excretion rates dropped insignificantly at a rate of 2.2% per
day (P ¼ 0.18). After stopping the drug, the albumin
excretion rate returned to baseline. No changes were noted
in blood pressure.

9.6 200

9.4
Serum Ca (mg/dl)

Urine Ca (mg/day)

150

9.2

100
9 Slope = 0.055 Slope = 10.2
(0.032 to 0.076) (7.2 to 13.3)
P < 0.001 P < 0.001

8.8 50

3.8 1000
Urine Phos (mg/day)
Serum Phos (mg/dl)

3.6 900

3.4 800

3.2 700
Slope = 0.050 Slope = 23.3
(0.024 to 0.076) (12.1 to 34.6)
P < 0.001 P < 0.001
3 600
–3
–2
–1
0
1
2
3
4
5
6
7

12
13
14

–3
–2
–1
0
1
2
3
4
5
6
7

12
13
14

Days Days

Figure 3 | Serum calcium (Ca) and phosphorous (Phos) and 24-h urine excretion rates of calcium and phosphorous at baseline,
during drug treatment, and after washout. The gray columns represent exposure to the drug. The raw means (solid diamonds) and
their standard errors (whiskers) are shown. The dashed lines are the modeled values (from the mixed model as detailed in the text).
For either the urinary excretions or serum concentrations of both calcium and phosphorus, there was a significant upward trend during
exposure to the study drug, followed by a return to baseline after a week-long washout period. Slopes represent the rate of change
during exposure to the drug with the 95% confidence intervals in parentheses. The mean level at washout was not significantly different
from baseline in any instance.

Kidney International (2011) 80, 1073–1079 1075

original article R Agarwal et al.: Paricalcitol increases serum creatinine but not GFR

Slope = 0.010 Slope = 0.36

2 32
(0.001 to 0.019) (0.11 to 0.061)

Serum urea N (mg/dl)

P = 0.036 P =0.005
30

Serum Cr (mg/dl)
1.9
28

26
1.8
24

1.7 22

12
Slope = 17.6 Slope = 0.094

Urine urea N (g/day)

(0.5 to 34.6) (–0.049 to 0.237)
Urine Cr (mg/day)

1700 P = 0.044 P =0.19

11

10
1500
9

1300 8
Slope = 0.49 Slope = 0.01
(–0.26 to 1.23) (–0.37 to 0.40)

Urea clearance (ml/min)

P >0.2 32 P >0.2
Cr clearance (ml/min)

70
30
65
28
60
26
55 24
–3
–2
–1
0
1
2
3
4
5
6
7

12
13
14

–3
–2
–1
0
1
2
3
4
5
6
7

12
13
14
Days Days

Figure 4 | Serum creatinine (Cr) and urea nitrogen (N) and 24-h urine excretion rates and clearances of creatinine and urea nitrogen
at baseline, during drug treatment, and after washout. The gray columns represent exposure to the drug. The raw means (solid
diamonds) and the standard error of the means (whiskers) are shown. The dashed lines are the modeled values. Exposure to paricalcitol
caused a significant increase in serum creatinine and urea nitrogen values, as well as the daily urine excretion of creatinine. There was no
significant effect on creatinine or urea clearance.

One patient complained of dizziness on day 7 of the drug,
which was possibly medication related. Another reported
skin irritation at the site of iothalamate infusion on day 6 of
the study. It was thought to be because of irritation caused by
the tape used to secure the catheter. One patient reported
lightheadedness shortly after receiving 1 ml of iothalamate
intravenously on day 1 and day 12 of the study. This
resolved within 30 min. No other adverse events were noted.
DISCUSSION
The key findings of this interventional study are that
short-term exposure to paricalcitol results in the following:
(1) significant increase in serum creatinine, (2) significant
increase in 24-h urine creatinine output, but (3) no change in
clearances of creatinine, urea, or iothalamate. These changes
are reversible after stopping the drug. As expected, rapid
changes in PTH, calcium, and phosphorus were observed as
may be predicted from the physiology.
The effects of paricalcitol on both urea nitrogen excretion
rate and serum urea nitrogen were intriguing. There was an
initial decline followed by a steady increase in both these
measurements. Assuming that the protein intake of our
participants was consistent, the parallel initial decline in
urine and serum urea nitrogen suggests an initial anabolic
response. Given that VDR activators have diverse effects on
various organs, it may point to a heretofore unrecognized
effect of VDR activators on protein metabolism.1,10,11
Two studies have examined the effect of VDR activators on
creatinine metabolism. Perez et al.8 tested the safety and
efficacy of oral calcitriol for the treatment of psoriasis.
Among 85 patients studied, during the first 6 months of
therapy, the creatinine clearance decreased by 13.4%. GFR
(inulin clearance) was measured in a subset of eight patients.
In the first 6 months, these eight patients experienced a
decline in creatinine clearance of 22.5%. Maximal reduction
in creatinine clearance was seen in 2 months. Despite a
reduction in creatinine clearance, no change was observed in
the inulin clearance over the entire period of study. This may
indicate a reduction in tubular secretion of creatinine. In
contrast, Bertoli et al.9 studied 10 patients with stable renal
function who were given 0.5 mg calcitriol per day for
4 months. Creatinine and inulin clearance were performed
at the beginning and the end of study. Although serum
creatinine increased and creatinine clearance decreased,
inulin clearance remained unchanged. After stopping calci-
triol, serum creatinine decreased to baseline value within 60
days. The authors attributed the increase in serum creatinine
to the augmented release of creatinine from muscular tissue
as a result of the salutary effect of calcitriol on uremic
myopathy. Inulin clearance at baseline was 13.1 ml/min and
at 120 days was 11.7 ml/min, indicating that these patients
had quite advanced kidney disease. These studies were
conducted over several months where true changes in renal
function could occur. In the absence of time controls, these
effects over time may then obscure the effect of VDR
activation on creatinine metabolism. Our study clearly
demonstrates the effect of VDR activation on increased
creatinine generation and, consequently, increased serum

1076 Kidney International (2011) 80, 1073–1079

R Agarwal et al.: Paricalcitol increases serum creatinine but not GFR
creatinine. The clearance of any of the filtration markers such
as creatinine or serum urea nitrogen or the reference standard
marker of GFR, iothalamate, was unaltered by VDR activation.
Thus, VDR activation seems to alter creatinine metabolism.
As we did not study the mechanism of change in
creatinine metabolism, we can only speculate how VDR
activation may change creatinine metabolism. Myopathy
associated with rickets is well known. This can be rapidly
reversed with vitamin D supplementation. Muscle is
considered a non-traditional target of VDR activators.12
In fact, VDR is widely expressed in the skeletal muscle, and
its deletion results in abnormal skeletal development
and myopathy.13 In a variety of diseases, alterations in both
skeletal muscle structure and function have been associated
with low levels of vitamin D. For example, in a murine model
of alcoholism, low levels of vitamin D are associated with
skeletal muscle atrophy.14 In patients with chronic obstruc-
tive pulmonary disease, VDR genotypes have been associated
with quadriceps strength.15 Among participants of the
Longitudinal Ageing Study in Amsterdam, low vitamin D
levels were a determinant of both loss of muscle strength and
muscle mass.16 Among the elderly, fatty degeneration of thigh
muscles, impaired balance, and gait were associated with
vitamin D deficiency.17 Supplementation with VDR activa-
tors may result in direct improvement in muscle strength and
reduced falls.18 VDR supplementation for 3–6 months, given
to 11 patients with bone loss caused by aging, reversed myopathy
and restored the creatinine phosphate stores to normal in
muscle biopsies from vastus lateralis.19 Finally, osteomalacia
treated with nutritional vitamin D supplementation resulted in
increased serum creatinine.20 Given these data on the effect of
VDR activation on muscle structure and function, it is possible
that paricalcitol had a direct effect on muscle metabolism,
resulting in increased creatinine generation.21
In an earlier trial reported by our group, we found no
changes in measured GFR using continuous iothalamate
infusion at baseline, after treatment with paricalcitol, and
after washout.3 In the same trial, ambulatory blood pressure
did not change, but a decrease in albuminuria was noted. In
this study, we were again not able to demonstrate any
meaningful changes in GFR on treatment or withdrawal of
the drug or changes in seated clinic blood pressure. Although
we found a trend toward a decline in albuminuria, this
was not significant. Reduction in albuminuria may take
longer than a week because reparative mechanisms may take
much longer to manifest. As an example, supplementation of
active vitamin D in a subtotal nephrectomy model in mice
leads to less podocyte injury, decreased podocytes loss,
and abrogation of podocyte hypertrophy.22 These repair
mechanisms are unlikely to be clinically detectable in 1 week.
The major strength of our study is that it used repeated
measurements in the same patient to observe small changes
in creatinine metabolism. In this era of large multicenter
trials, this study demonstrates that small studies have a place
in uncovering mechanistic insights that are made possible
with repeated measurements. The limitations of our study
Kidney International (2011) 80, 1073–1079
original article
include the small sample size and the limited duration of
exposure to VDR activator. This study did not examine the
long-term effects of VDR activators. Future studies may
explore these effects.
In conclusion, VDR activation may alter creatinine
metabolism. An increase in creatinine generation may lead
to an increase in serum creatinine and, if eGFR is used to
measure kidney function, it may give the appearance that
kidney function is declining when truly it may not be altered.
Paricalcitol may be added to the list of drugs that alter serum
creatinine without affecting renal clearance. The long-term
effect of VDR activators on creatinine and protein metabo-
lism may reveal new mechanisms on how VDR activators
may modulate metabolic pathways in the uremic milieu.
MATERIALS AND METHODS
Participants
Between 1 September 2010 and 16 February 2011, we recruited
16 patients from the renal clinic at Richard L Roudebush VA
Medical Center, Indianapolis, Indiana. Patients were considered
eligible for the study if they were older than 18 years, and had CKD
with an eGFR between 15 and 60 ml/min per 1.73 m2 and had stable
renal function. We excluded patients with the following character-
istics: hyperphosphatemia (46 mg/dl), hypercalcemia (410 mg/dl),
vitamin D use (4400 units/day), VDR activator use, use of drugs
that block creatinine secretion (cimetidine, triamterene, or
trimethoprim), or allergy to radiocontrast dye.
The study protocol was approved by the Institutional Review
Boards and the VA Research and Development Committee, and all
patients provided written informed consent. The trial was registered
at clinicaltrials.gov (NCT01163162).
Study protocol
At baseline, we collected three consecutive 24 h urine specimens to
measure creatinine, urea, calcium, phosphorus, and albumin. Each
urine collection was bracketed by a blood draw for measurement of
creatinine, urea, calcium, phosphorus, and other analytes. The
patients were asked not to consume meat before the blood draw.
Following these collections we measured GFR by a continuous
infusion of iothalamate as reported below. PTH-intact was
measured once.
After these baseline measurements, each patient received 2 mg
of paricalcitol once a day for 7 consecutive days. This was witnessed
daily by the study coordinator. On each one of these days,
the patient collected 24 h urine, and blood was collected in the
fasting state. Seated blood pressure was measured at each visit in
triplicate using a validated oscillometric device (HEM 907, Omron
Healthcare, Bannockburn, IL). After 7 days, GFR and PTH-intact
was again measured and paricalcitol stopped. At 4 days after
stopping the drug, 24 h urine specimen and blood were collected
again for 3 consecutive days. GFR and PTH-intact were then
measured for the final time.
GFR measurement
An outpatient continuous infusion of iothalamate was used to
measure GFR.23 On the first visit, a baseline plasma sample was
drawn. Further, a 1 ml bolus of iothalamate (Conray 60; Mallinckrodt
Medical, St Louis, MO) was injected intravenously. A subcutaneous
catheter for infusion of iothalamate was inserted in the abdominal
1077
original article
subcutaneous tissue. Iothalamate was infused at a constant rate of
125 ml/h through an insulin pump (MiniMed model 506 or 507;
Medtronic, Northridge, CA). Approximately 24 h after the start of
the infusion, the subjects returned to the research laboratory. The
insulin pump was refilled with iothalamate if needed. An
intravenous catheter was placed in the forearm and three blood
samples were obtained at 30 min intervals for measurements
of plasma iothalamate by high-performance liquid chromato-
graphy.24,25 Plasma clearance of iothalamate was calculated by
dividing the infusion rate by plasma iothalamate concentration.
Plasma iothalamate clearances have a coefficient of variation of 7.7%
when measured 4 months apart.23
Biochemical assays
Biochemical assays were performed in our hospital laboratory.
Serum creatinine was calibrated to the isotope dilution mass
spectroscopic standard. Urine albumin was measured with a
particle-enhanced turbidimetric inhibition immunoassay adapted
to the Dimension clinical chemistry system that allows direct
quantification of albumin in urine samples (Siemens Healthcare
Diagnostics, Deerfield, IL).
Statistical analysis
Sample size estimation was based on the following considerations.
Two previous studies that have investigated the effect of VDR
activators on renal function measured inulin clearances in 8 and 10
patients, respectively.8,9 Each of these studies could detect changes in
creatinine metabolism even when assessments were made using
single 24-h urine collections. Neither of the studies stopped VDR
activators; thus, they did not explicitly test the reversibility of
creatinine metabolism. The duration of those studies were long.
Long duration of the study could result in deterioration in renal
function that could obscure the effect of the drug from the effect of
time on creatinine metabolism. Using a more intense protocol with
repeated measurements, we reasoned that 16 patients would be
sufficient to detect an effect size similar to, or smaller than, that seen
with previous studies. A short term was chosen to obviate changes in
GFR due to disease progression.
The primary method of data analysis was by mixed models.26
The hypothesis being tested was that paricalcitol alters creatinine
generation and increases serum creatinine, and that these changes in
creatinine metabolism are reversible and unaccompanied by a
reduction in GFR or creatinine clearance. We had 13 measurements
of creatinine clearance per patient: three at baseline, seven during
paricalcitol therapy, and three after washout. All these data were
analyzed using a composite model. In this model, we specified fixed
intercept effects for baseline, treatment, and washout. In addition,
we specified a fixed slope effect for treatment. The random effects
included in the model were subject and visits. Covariance was
modeled as unstructured that allows the intercepts and slopes to
change independently of each other. Maximal likelihood estimations
were used to model the data.
Statistical analysis was performed using Stata 11.0 (Stata, College
Station, TX). Significance was set at two-sided P-value of o0.05.
DISCLOSURE
RA serves as a consultant for and on the speaker bureau of Abbott
Pharmaceuticals. He also serves or has served on steering committees
of clinical trials using paricalcitol supported by Abbott
Pharmaceuticals. All the other authors declared no competing
interests.
1078
R Agarwal et al.: Paricalcitol increases serum creatinine but not GFR
ACKNOWLEDGMENTS
We dedicate this work in general to all veterans who selflessly
participate in clinical research and in particular to the 16 participants
of this arduous study. This study was supported by a grant from
Abbott.
SUPPLEMENTARY MATERIAL
Figure S1. Serum urea nitrogen and 24-h urine urea nitrogen at
baseline, during drug treatment, and after washout.
Figure S2. Seated systolic and diastolic blood pressure (BP) and
24-hour urine albumin excretion rate over the study.
Supplementary material is linked to the online version of the paper at
http://www.nature.com/ki
REFERENCES
1. Holick MF. Vitamin D deficiency. N Engl J Med 2007; 357: 266–281.
2. Agarwal R, Acharya M, Tian J et al. Antiproteinuric effect of oral
paricalcitol in chronic kidney disease. Kidney Int 2005; 68:
2823–2828.
3. Alborzi P, Patel NA, Peterson C et al. Paricalcitol reduces albuminuria and
inflammation in chronic kidney disease: a randomized double-blind pilot
trial. Hypertension 2008; 52: 249–255.
4. Fishbane S, Chittineni H, Packman M et al. Oral paricalcitol in the
treatment of patients with CKD and proteinuria: a randomized trial. Am J
Kidney Dis 2009; 54: 647–652.
5. Lambers Heerspink HJ, Agarwal R, Coyne DW et al. The selective
vitamin D receptor activator for albuminuria lowering (VITAL) study:
study design and baseline characteristics. Am J Nephrol 2009; 30:
280–286.
6. de Zeeuw D, Agarwal R, Amdahl M et al. Selective vitamin D receptor
activation with paricalcitol for reduction of albuminuria in patients with
type 2 diabetes (VITAL study): a randomised controlled trial. Lancet 2010;
376: 1543–1551.
7. Christiansen C, Rodbro P, Christensen MS et al. Deterioration of renal
function during treatment of chronic renal failure with 1,25-
dihydroxycholecalciferol. Lancet 1978; 2: 700–703.
8. Perez A, Raab R, Chen TC et al. Safety and efficacy of oral calcitriol
(1,25-dihydroxyvitamin D3) for the treatment of psoriasis. Br J Dermatol
1996; 134: 1070–1078.
9. Bertoli M, Luisetto G, Ruffatti A et al. Renal function during
calcitriol therapy in chronic renal failure. Clin Nephrol 1990; 33:
98–102.
10. Dusso AS, Brown AJ, Slatopolsky E. Vitamin D. Am J Physiol Renal Physiol
2005; 289: F8–F28.
11. Agarwal R. Vitamin D, proteinuria, diabetic nephropathy, and progression
of CKD. Clin J Am Soc Nephrol 2009; 4: 1523–1528.
12. Demay M. Muscle: a nontraditional 1,25-dihydroxyvitamin D target tissue
exhibiting classic hormone-dependent vitamin D receptor actions.
Endocrinology 2003; 144: 5135–5137.
13. Endo I, Inoue D, Mitsui T et al. Deletion of vitamin D receptor gene in
mice results in abnormal skeletal muscle development with deregulated
expression of myoregulatory transcription factors. Endocrinology 2003;
144: 5138–5144.
14. Gonzalez-Reimers E, Duran-Castellon MC, Lopez-Lirola A et al. Alcoholic
myopathy: vitamin D deficiency is related to muscle fibre atrophy in a
murine model. Alcohol Alcohol 2010; 45: 223–230.
15. Hopkinson NS, Li KW, Kehoe A et al. Vitamin D receptor genotypes
influence quadriceps strength in chronic obstructive pulmonary disease.
Am J Clin Nutr 2008; 87: 385–390.
16. Visser M, Deeg DJ, Lips P. Low vitamin D and high parathyroid hormone
levels as determinants of loss of muscle strength and muscle mass
(sarcopenia): the Longitudinal Aging Study Amsterdam. J Clin Endocrinol
Metab 2003; 88: 5766–5772.
17. Tagliafico AS, Ameri P, Bovio M et al. Relationship between fatty
degeneration of thigh muscles and vitamin D status in the elderly:
a preliminary MRI study. AJR Am J Roentgenol 2010; 194: 728–734.
18. Gallagher JC. The effects of calcitriol on falls and fractures and
physical performance tests. J Steroid Biochem Mol Biol 2004; 89–90:
497–501.
19. Dluhy RG, McMahon GT. Intensive glycemic control in the ACCORD and
ADVANCE trials. N Engl J Med 2008; 358: 2630–2633.
20. Fonseca V, Mohiuddin J, Weerakoon J et al. Plasma creatinine and
creatinine clearance in nutritional osteomalacia. Lancet 1984; 1:
1093–1095.
Kidney International (2011) 80, 1073–1079
R Agarwal et al.: Paricalcitol increases serum creatinine but not GFR original article

21. Dirks-Naylor AJ, Lennon-Edwards S. The effects of vitamin D on skeletal
muscle function and cellular signaling. J Steroid Biochem Mol Biol 2011;
125: 159–168.
22. Kuhlmann A, Haas CS, Gross ML et al. 1,25-Dihydroxyvitamin D3
decreases podocyte loss and podocyte hypertrophy in the
subtotally nephrectomized rat. Am J Physiol Renal Physiol 2004; 286:
F526–F533.
23. Agarwal R. Ambulatory GFR measurement with cold iothalamate
in adults with chronic kidney disease. Am J Kidney Dis 2003; 41:
752–759.
24. Agarwal R, Vasavada N, Chase SD. Liquid chromatography for
iothalamate in biological samples. J Chromatogr B Analyt Technol Biomed
Life Sci 2003; 785: 345–352.
25. Agarwal R. Chromatographic estimation of iothalamate and
p-aminohippuric acid to measure glomerular filtration rate and
effective renal plasma flow in humans. J Chromatogr B Biomed Sci Appl
1998; 705: 3–9.
26. Holden JE, Kelley K, Agarwal R. Analyzing change: a primer on
multilevel models with applications to nephrology. Am J Nephrol 2008;
28: 792–801.

Kidney International (2011) 80, 1073–1079 1079