Content uploaded by Hesham Keryakos
Author content
All content in this area was uploaded by Hesham Keryakos on Jan 13, 2021
Content may be subject to copyright.
Vol.:(0123456789)
1 3
International Urology and Nephrology
https://doi.org/10.1007/s11255-020-02753-y
NEPHROLOGY - ORIGINAL PAPER
Osteocalcin andvascular calcification inhemodialysis patients:
anobservational cohort study
HeshamKamalHabeebKeryakos1 · NagwaIsmailOkaily2· MariamAsaadYacoubBoulis1·
AhmedMohamedSaadeldinSalama1
Received: 26 May 2020 / Accepted: 15 December 2020
© The Author(s), under exclusive licence to Springer Nature B.V. part of Springer Nature 2021
Abstract
Background Vascular calcification contributes to morbidity and mortality in patients with ESRD on maintenance
hemodialysis.
Aims To study the relationship between osteocalcin and vascular calcification.
Methods 160 patients with ESRD on maintenance hemodialysis and 60 age-and sex-matched healthy controls were recruited.
Serum vitamin K2 and osteocalcin both intact and undercarboxylated were measured. Transthoracic echocardiography was
done for valvular calcification and thickening, and carotid duplex was done for carotid intimal medial calcification and
thickening.
Results Hemodialysis patients have higher median serum vitamin K2 (p < 0.001), higher undercarboxylated osteocalcin
(p < 0.001). Only older age, duration of hypertension, and duration of established cardiovascular disease are associated with
carotid media-intimal calcification. Old age is a strong predictor of carotid media intimal thickening. Female sex is associ-
ated with a valvular thickening.
Conclusions Functional vitamin K deficiency is present in maintenance hemodialysis patients and serum osteocalcin is not
associated with cardiovascular calcification.
Keywords Cardiovascular calcification· Hemodialysis· Osteocalcin· Vitamin K2· Valvular thickening
Introduction
Hemodialysis patients experience severe mixed intimal
and medial vascular calcification, which contributes to the
increased cardiovascular mortality and morbidity. Cardio-
vascular calcification contributes to approximately 50% of
all deaths in hemodialysis patients [1]. At the molecular
level, both matrix Gla protein (MGP) which is a central cal-
cification inhibitor in the arterial wall, and fetuin-A which is
a circulating inhibitor for vascular calcification are inhibitors
of vascular calcification. In hemodialysis patients, fetuin-A
level is decreased and MGP is undercarboxylated and their
contribution to cardiovascular calcification was studied
extensively but with different results [2]. The relationship
between cardiovascular calcification and fetuin-A is contro-
versial but with MGP is well established [3].
MGP is a vitamin K dependent protein. Vitamin K is
referred to a group of fat-soluble vitamins which act as a
cofactor for the enzyme γ-glutamyl carboxylase (GGCX),
which activates a number of vitamin K-dependent proteins
(VKDPs) mainly blood coagulation factors (II, VII, IX, and
X) in the liver which involved in hemostasis, matrix Gla pro-
tein (MGP) which is synthesized by chondrocytes and vas-
cular smooth muscle cells and involved in the prevention of
vascular calcification and osteocalcin which is synthesized
by osteoblasts and involved in bone formation [4].
Osteocalcin is a protein involved in bone metabolism. In
the general population, vitamin K deficiency and increased
uncarboxylated osteocalcin have been associated with
Supplementary Information The online version contains
supplementary material available at https ://doi.org/10.1007/s1125
5-020-02753 -y.
* Hesham Kamal Habeeb Keryakos
hesham.keryakos@mu.edu.eg
1 Internal Medicine Department, Faculty ofMedicine,
Minia University, Aswan-Cairo Agricultural Road,
El-Minia61111, Egypt
2 Clinical Pathology Department, Faculty ofMedicine, Minia
University, El-Minia, Egypt
International Urology and Nephrology
1 3
increased fracture rates. A meta-analysis based on 46 orig-
inal articles found conflicting inconsistent results between
osteocalcin and vascular calcification and atherosclero-
sis [5]. On the contrary, an inverse association between
circulating total osteocalcin and atherosclerotic outcomes
and CVD endpoints was found in another recent meta-
analysis of 33 observational studies that included 21,021
participants [6]. However, the relationship between under-
carboxylated osteocalcin and cardiovascular calcification
has not been studied well [5, 7, 8]. Therefore, we inves-
tigated serum vitamin K2 state in hemodialysis patients,
the relation between serum osteocalcin and cardiovascular
calcification, and to identify possible predictors of cardio-
vascular calcification in our cohort of patients.
Subjects andmethods
Study participants
The study was reviewed and approved by the local ethics
committee and was conducted in accordance with the Hel-
sinki Declaration. Informed consent was obtained from all
participants. This study was an observational case–control
in-center hemodialysis study for investigating the correla-
tion between serum vitamin K2, serum osteocalcin, and
cardiovascular calcification in hemodialysis patients. The
study participants were recruited from the hemodialysis
unit of Minia University Hospital, in the period from
August 2018 to February 2019. The study enrolled one
hundred and sixty (160) patients with end-stage renal dis-
ease (ESRD) undergoing maintenance hemodialysis thrice
weekly; as well as sixty (60) age-and sex-matched healthy
controls. The inclusion criteria were adults 18years old
or more with ESRD patients on maintenance hemodi-
alysis for at least 6 months. The exclusion criteria for
ESRD patients were poor mobility (inability to walk 100
yards unaided), malabsorption (extensive bowel surgery,
short bowel), generalized carcinomatosis, glucocorticoid
therapy, inflammatory disorders (e.g. Active rheumatoid
arthritis, inflammatory bowel disease requiring oral gluco-
corticoids), endocrine diseases (e.g. Primary hyperparath-
yroidism, hyperthyroidism), chronic liver disease, current
treatment with teriparatide, strontium ranelate, participa-
tion in a trial with an investigational product within the
previous 3months, patients on anticoagulants such as
warfarin. The exclusion criteria for healthy controls were
serum creatinine > 1.2mg/dL. Plasma samples were col-
lected at the beginning of the dialysis session, after a 3-day
dialysis-free interval, and then stored at − 80°C until fur-
ther analysis.
Clinical data collection andlaboratory methods
Demographic data, history of cardiovascular risk factors,
any atherosclerotic cardiovascular disease (ASCVD), with
review to hemodialysis prescription with an assessment of
functional activity were obtained at enrollment. Abdomin-
opelvic ultrasound using General Electric ultrasound and
transducer with a frequency of 3.5 megahertz (MHz) (GE
Healthcare, USA) was performed to all patients. Tran-
sthoracic echocardiography using vivid S5 ultrasound
machine (GE Healthcare, USA) was used to check for val-
vular calcification, LVH, and pulmonary hypertension.
Carotid duplex using Philips HD5 ultrasound machine
(Philips Medical Systems Nederland B.V., The Nether-
lands) was used to assess for the presence and extent of
carotid atherosclerosis.
Complete blood count (CBC) was measured using
automated cell counter; Sysmex KX-21N (TAO Medi-
cal incorporation, Japan). Fasting blood glucose, liver
and renal function tests were measured using Konelab 20i
auto-analyzer (Thermo-electron incorporation, Finland).
Serum total calcium, serum phosphorus, serum iron, and
serum ferritin were measured using true semi-automated
clinical chemistry analyzer Microlab 300 (ELITechGroup
Biomedical Systems, Netherlands). Serum iPTH, serum
intact and undercarboxylated osteocalcin, and serum
human vitamin K2 (VK2) were determined using com-
mercially available ELISA kits (HumaReader Plus, Model:
3700; Germany).
Measurements of serum human vitamin K2 was per-
formed using a two-site second-generation enzyme-linked
immunosorbent assay (ELISA) kit (Bioassay Technology
Laboratory Co. Ltd., Shanghai, China) according to the
manufacturer’s instructions.
Measurements of human osteocalcin were performed
using a two-site second-generation enzyme-linked immu-
nosorbent assay (ELISA) kit (Bioassay Technology Labo-
ratory Co. Ltd., Shanghai, China) according to the manu-
facturer’s instructions.
Measurements of serum human undercarboxylated oste-
ocalcin (ucOC) were performed using a two-site second-
generation enzyme-linked immunosorbent assay (ELISA)
kit (Bioassay Technology Laboratory Co. Ltd., Shanghai,
China) according to the manufacturer’s instructions.
Statistical analysis
All statistical analyses were performed with IBM SPSS
Statistics software version 23 (IBM, New York, USA).
Non-parametric quantitative data are presented as range
and median while parametric quantitative data are reported
International Urology and Nephrology
1 3
as mean ± standard deviation (SD). On the other hand,
qualitative data are presented as number and percentage.
For the comparison between quantitative data, independ-
ent sample t test was used. Chi-square (χ2) test and Fish-
er’s exact test were applied for qualitative data analysis.
Kolmogorov–Smirnov for normality test was used to dif-
ferentiate between parametric data and non-parametric
data. Mann–Whitney test used for non-parametric quan-
titative data. Linear correlation was done using Pearson’s
correlation coefficient and Spearman’s rank correlation.
For all tests probability (p) was considered significant
if ≤ 0.05, highly significant if ≤ 0.01, very highly signifi-
cant if ≤ 0.001.
Results:
Demographic andlaboratory characteristics:
Table1 shows the demographic and laboratory characteris-
tics of the study participants. Both groups were age and sex
matched. Regarding cardiovascular risk factors 48 patients
were smokers (30.4%, p = 0.054), hypertension was pre-
sent in 96 patients (60%, p < 0.001), diabetes mellitus was
present in 26 patients (16.3%, p = 0.120). Established car-
diovascular disease was present in 50 hemodialysis patients
(31.3%, p = 0.009). Serum vitamin K2 was statistically sig-
nificantly higher in hemodialysis patients as compared to
controls (34.3ng/ml vs. 27ng/ml, p < 0.001); meanwhile
serum undercarboxylated osteocalcin was statistically signif-
icantly higher in hemodialysis patients as compared to con-
trols (21.8ng/ml vs. 3.9ng/ml, p < 0.001), and serum intact
osteocalcin was statistically significantly lower (2.5ng/ml
vs. 15ng/ml, p < 0.001) as shown in Fig.1. The ratio of
undercarboxylated osteocalcin to intact osteocalcin was sta-
tistically significantly higher in hemodialysis patients versus
controls (8.7 vs. 0.2, p < 0.001).
Carotid media‑intimal calcification
Table2 shows the impact of study parameters on the pres-
ence of carotid media-intimal calcification. Only four
Table 1 Demographic and
laboratory characteristics of
study participants
Mann Whitney test for non-parametric quantitative data between the two groups. Chi-square test or Fisher
exact test for qualitative data between the two groups
eCVD established cardiovascular disease, Hb hemoglobin, ALP alkaline phosphatase, iPTH intact parathy-
roid hormone, vit K2 vitamin K2, ucOC undercarboxylated hemoglobin, iOC intact osteocalcin
*Significant level at P value < 0.05
Control Cases P value
N = 60 N = 160
Age (years) Median/IQR 48/(32–59) 50/(35.3–60) 0.814
Sex Male
Female
20 (33.3%)
40 (66.7%)
70 (43.8%)
90 (56.3%)
0.453
BMI Median/IQR 25/(23.8–27.2) 25.7/(21.2–32.8) 0.642
Residence Rural
Urban
36 (60%)
24 (40%)
88 (55%)
72 (45%)
0.720
Smoking No
Yes
60 (100%)
0 (0%)
112 (69.6%)
48 (30.4%)
0.054
Hypertension No
Yes
60 (100%)
0 (0%)
64 (40%)
96 (60%)
< 0.001*
Diabetes mellitus No
Yes
6 0(100%)
0 (0%)
134 (83.8%)
26 (16.3%)
0.120
eCVD No
Yes
60 (100%)
0 (0%)
110 (68.8%)
50 (31.3%)
0.009*
Hb (g/dL) Median/IQR 12.7/(11.5–13.5) 10/(9.1–11.1) < 0.001*
Ferritin (ng/ml) Median/IQR 89/(72–100) 206.5/(58.5–392) 0.049*
Calcium (mg/dL) Median/IQR 9/(8.5–9.6) 7.5/(7.2–7.8) < 0.001*
Phosphorus (mg/dL) Median/IQR 3.1/(2.8–3.9) 5.6/(5.2–6.7) < 0.001*
ALP (IU/L) Median/IQR 95/(90–110) 229.5/(162.8–306.3) < 0.001*
iPTH (pg/ml) Median/IQR 37/(19–42) 225/(200–400) < 0.001*
Vit K2 (ng/ml) Median/IQR 27/(23–32) 34.3/(29–41.8) < 0.001*
ucOC (ng/ml) Median/IQR 3.9/(3.3–4.1) 21.8/(14–38.4) < 0.001*
iOC (ng/ml) Median/IQR 15/(12–18) 2.5/(2–4) < 0.001*
ucOC/iOC ratio Median/IQR 0.2/(0.2–0.3) 8.7/ (5.5–13.2) < 0.001*
International Urology and Nephrology
1 3
parameters which are age, duration of hypertension, dura-
tion of diabetes mellitus, and duration of established car-
diovascular disease are statistically significant in patients
with carotid media-intimal calcification (60 year vs.
46year, p < 0.001; 11.5year vs 5.5year, p = 0.003; 20year
vs. 10year, p = 0.021; 6.5year vs. 3.0 year, p = 0.028,
respectively). This is supported in simple regression analy-
sis as showed in Table3. However, with simple regression
analysis only age, duration of hypertension and duration of
established cardiovascular disease were predictors of the
presence of carotid media-intimal calcification but not the
duration of diabetes mellitus duration.
Fig. 1 Serum vitamin K2 and
osteocalcin in study par-
ticipants. Despite higher serum
vitamin K2 in hemodialysis
patients, serum undercarboxy-
lated osteocalcin is higher
which reflects relative vitamin
K deficiency
0
5
10
15
20
25
30
35
40
Vitamin K2 (ng/ml)ucOC (ng/ml)iOC (ng/ml)ucOC/iOC rao
Vitamin K2 and Osteocalcin in study parcipants
Control Cases
*
*
*
*
p<0.001
p<0.001
p<0.001
p<0.001
Table 2 Impact of study
parameters on the presence
of carotid media-intimal
calcification
Mann Whitney test for non-parametric quantitative data between the two groups. Chi-square test or Fisher
exact test for qualitative data between the two groups
*Significant level at P value < 0.05
Carotid media-intimal calcification P value
No Yes
N = 126 N = 34
Age (years) Median/IQR 46/(32–59) 60/(53.5–64.5) < 0.001*
Sex Male
Female
54 (42.9%)
72 (57.1%)
16 (47.1%)
18 (52.9%)
0.757
Hypertension No
Yes
54 (42.9%)
72 (57.1%)
10 (29.4%)
24 (70.6%)
0.315
Diabetes mellitus No
Yes
110 (87.3%)
16 (12.7%)
24 (70.6%)
10 (29.4%)
0.136
eCVD No
Yes
92 (73%)
34 (27%)
18 (52.9%)
16 (47.1%)
0.113
HTN Duration (years) Median/IQR 5.5/(2.5–8.6) 11.5/(7–20) 0.003*
DM Duration (years) Median/IQR 10/(6.3–15) 20/(20–26.5) 0.021*
eCVD Duration (years) Median/IQR 3/(1–5) 6.5/(3–10) 0.028*
Dialysis Duration (years) Median/IQR 3.5/(1.5–6.5) 3/(1.5–4.5) 0.491
Serum Calcium (mg/dL) Median/IQR 7.6/(7.1–7.8) 7.3/(7.2–7.6) 0.128
Serum Phosphorus (mg/dL) Median/IQR 5.6/(5.2–6.6) 6/(5.2–7.4) 0.454
Serum ALP (IU/L) Median/IQR 234/(165–310) 215/(128.5–284) 0.455
Serum iPTH (pg/ml) Median/IQR 240/(130–400) 220/(200–355) 0.831
Vit K2 (ng/ml) Median/IQR 35/(30–42) 33/(28–41) 0.621
ucOC (ng/ml) Median/IQR 23/(14–41.6) 18/(10.6–25.4) 0.200
iOC (ng/ml) Median/IQR 3/(2–5) 2/(2–3) 0.275
ucOC/iOC ratio Median/IQR 9/(5.5–13.3) 8.3/(5.3–9.6) 0.522
International Urology and Nephrology
1 3
There was no relation between any of the laboratory
markers namely serum calcium, serum phosphorus, serum
ALP and serum iPTH with the presence of carotid media-
intimal calcification. Also, neither serum vitamin K2 nor its
vitamin K dependent protein osteocalcin showed any rela-
tion with the presence of carotid media-intimal calcification.
Valvular calcification
The presence of valvular calcification did not increase
with any of the study parameters; presence or duration of
traditional cardiovascular risk factors, laboratory markers
of BMD-CKD, serum vitamin K2 or VKDP as shown in
Table4.
Valvular thickening
The normal mitral valve thickening is 4–5mm. Valve thick-
ening was found in 46 hemodialysis patients. Most of the
hemodialysis patients were females (34 vs. 12) as shown
in Table5. With simple regression analysis, female sex
was found to increase the risk of valve thickening (Odds
ratio 2.935, 95% CI 1.011–8.519, p = 0.048) as shown in
Table6. Meanwhile, the presence of vascular thickening did
not increase with any of the study parameters; duration of
traditional cardiovascular risk factors, laboratory markers of
BMD-CKD, serum vitamin K2 or osteocalcin.
Carotid intimal‑medial thickness (CIMT)
Hemodialysis Patients were divided into two groups accord-
ing to an average CIMT; those with IMT ≤ 7.5mm and those
with IMT > 7.5mm. Patients with CIMT > 7.5mm tended
to be older (58 vs. 33, p = 0.008), had a higher incidence of
diabetes mellitus (p < 0.001), and longer duration of hyper-
tension (7 vs. 4, p = 0.033)(Table7). With simple regression
Table 3 Simple regression analysis for prediction of carotid media-
intimal calcification
OR odds ratio, CI confidence interval
*Significant level at P value < 0.05
OR 95% CI P value
Age (years) 1.072 1.024–1.123 0.003*
HTN duration (years) 1.168 1.045–1.305 0.006*
DM duration (years) 1.241 0.979–1.572 0.075
eCVD duration (years) 1.525 1.027–2.265 0.036*
Table 4 Impact of study
parameters on the presence of
valvular calcification
Mann Whitney test for non-parametric quantitative data between the two groups
Significant level at P value < 0.05
Valvular calcification p value
No Yes
N = 148 N = 12
Age (years) Median/IQR 50/(34.8–60.8) 55/(33–60.3) 0.898
Sex Male
Female
66 (44.6%)
82 (55.4%)
4 (33.3%)
8 (66.7%)
0.691
HTN No
Yes
60 (40.5%)
88 (59.5%)
4 (33.3%)
8 (66.7%)
1
Diabetes No
Yes
124 (83.8%)
24 (16.2%)
10 (83.3%)
2 (16.7%)
1
eCVD No
Yes
100 (67.6%)
48 (32.4%)
10(83.3%)
216.7%
0.660
HTN duration (years) Median/IQR 6.5/(3–10) 2.8/(2.1–12) 0.303
DM duration (years) Median/IQR 17.5/(10–22.3) 15/(15–15) 0.892
eCVD duration (years) Median/IQR 4/(2–6.5) 3/(2.5–4) 0.127
Dialysis duration (years) Median/IQR 3.5/(1.5–6) 3.8/(3.4–8.3) 0.206
Calcium (mg/dL) Median/IQR 7.6/(7.2–7.8) 7.4/(6.6–7.7) 0.453
Phosphorus (mg/dL) Median/IQR 5.7/(5.2–6.7) 5.2/(4.9–6.2) 0.156
ALP (IU/L) Median/IQR 232/(161–307.8) 184/(164.5–297.8) 0.584
PTH (pg/ml) Median/IQR 225/(187.5–400) 260/(200–400) 0.692
Vit K2 (ng/ml) Median/IQR 34.3/(29–41.3) 36/(32.8–45.8) 0.356
ucOC (ng/ml) Median/IQR 20.9/(13.5–38.1) 30.5/(21.3–46.4) 0.207
iOC (ng/ml) Median/IQR 2/(2–4) 3/(1.8–4.3) 0.829
ucOC/iOC ratio Median/IQR 8.4/(5.3–13.1) 12.2/(8.4–14.5) 0.149
International Urology and Nephrology
1 3
analysis, age was a strong predictor of CIMT > 7.5mm (odds
ratio 1.143, 95% CI 1.082–1.207, p < 0.001) (Table8).
Discussion
In this study, we investigated the risk factors of cardiovascu-
lar calcification in hemodialysis patients with special empha-
sis on vitamin K2 and osteocalcin. Vitamin K is present
in two forms: K1 (phylloquinone) mainly found in green
vegetables, and in certain plant oils, and K2 (menaquinones)
derived from intestinal bacteria and fermented food with
a small amount produced from the systemic conversion of
phylloquinone to menaquinones. Vitamin K1 acts on the
liver, meanwhile vitamin K2 has extrahepatic actions. The
optimal daily intake of vitamin K required to activate all
vitamin K dependent proteins is not clear, and up till now
there is no known vitamin K toxicity. Measurement of serum
vitamin K is affected by dietary intake as it is a fat-soluble
vitamin. PIVKA II (protein induced by vitamin K absence
II) is a liver-derived vitamin K dependent protein with a con-
centration < 2ng/mL in the healthy population. PIVKA II is
a marker of vitamin K deficiency with circulating PIVKA II
levels are increased in states of vitamin K deficiency includ-
ing the use of vitamin K antagonists. In the normal popula-
tion, prothrombin is always carboxylated as the liver takes
all vitamin K in need. Also, vitamin K status is assessed by
measurement of dephosphorylated uncarboxylated MGP,
uncarboxylated osteocalcin in addition to PIVKA II and
prothrombin [9, 10].
MGP is activated by the vitamin K dependent γ-glutamyl
carboxylase enzyme in the endoplasmic reticulum. Uncar-
boxylated MGP formed because of vitamin K deficiency
is associated with cardiovascular disease. Mice deficient
in MGP survived less than 6weeks after birth because of
Table 5 Impact of study
parameters on the presence of
valvular thickening
Mann Whitney test for non-parametric quantitative data between the two groups. Chi-square test or Fisher
exact test for qualitative data between the two groups
*Significant level at P value < 0.05
Valvular thickening P value
No Yes
N = 114 N = 46
Age (years) Median/IQR 50/(36.5–63) 40/(34–60) 0.383
Sex Male
Female
58 (50.9%)
56 (49.1%)
12 (26.1%)
34 (73.9%)
0.043*
HTN No
Yes
50 (43.9%)
64 (56.1%)
14 (30.4%)
32 (69.6%)
0.267
Diabetes mellitus No
Yes
96 (84.2%)
18 (15.8%)
38 (82.6%)
8 (17.4%)
> 0.99
eCVD No
Yes
80 (70.2%)
34 (29.8%)
30 (65.2%)
16 (34.8%)
0.665
HTN Duration (years) Median/IQR 6.5/(3–12.3) 6/(2.6–10) 0.767
DM Duration (years) Median/IQR 20/(10–26.5) 12.5/(6.3–18.8) 0.309
eCVD Duration (years) Median/IQR 3.5/(1–6.1) 4/(3–6.9) 0.644
Dialysis duration (years) Median/IQR 3.5/(2–5.9) 3.5/(1.5–6.5) 0.551
Serum Calcium (mg/dL) Median/IQR 7.6/(7.1–7.9) 7.5/(7.2–7.6) 0.103
Serum Phosphorus (mg/dL) Median/IQR 5.7/(5.2–6.8) 5.6/(5.1–6.2) 0.519
Serum ALP (IU/L) Median/IQR 229/(139.5–320) 234/(180–286) 0.941
Serum iPTH (pg/ml) Median/IQR 300/(175–400) 200/(200–400) 0.577
Vit K2 (ng/ml) Median/IQR 36/(29.5–43.5) 33/(28–38) 0.172
ucOC (ng/ml) Median/IQR 22.7/(14.5–39.4) 18/(8.1–33.3) 0.148
iOC (ng/ml) Median/IQR 3/(2–6.5) 2/(1.2–3) 0.077
ucOC/iOC ratio Median/IQR 8.5/(6.3–13.1) 9/(4.1–14) 0.983
Table 6 Simple regression analysis for prediction of valvular thicken-
ing
OR odds ratio, CI confidence interval, Ref. Reference
*Significant level at P value < 0.05
OR 95% CI P value
Sex
Male Ref
Female 2.935 1.011–8.519 0.048*
International Urology and Nephrology
1 3
rupture of their heavily calcified arteries. Warfarin which is
vitamin K antagonist used in rats was shown to induce rapid
calcification of arteries and cardiac valves within 2weeks,
and this was associated with upregulation of uncarboxylated
MGP at sites of calcification [11].
The role of vitamin K deficiency in vascular calcifica-
tion is supported in a prospective study that assessed the
impact of the use of vitamin K antagonist in 157 patients
with low-risk AF patients on the coronary artery calcifica-
tion (CAC) score. Patients on vitamin K antagonist had a
significant increase in CAC score. This increase was more
evident with increasing duration of vitamin K antagonist
use and with age > 65years [12]. To test the role of vita-
min K in preventing vascular calcification, an experimental
study performed on 30 rates fed warfarin and vitamin K1
to induce vascular calcification (termed W&K group) for
6weeks and compared with 18 control rats that received a
normal dose of vitamin K1 and no warfarin. Vitamin K1 was
used in the W&K group to prevent bleeding. After 6weeks
rats in the W&K group were divided into 4 groups each
of 6 rats; one group continued warfarin and vitamin K1
diet, whereas warfarin was discontinued in the remaining
3 groups: one group received normal vitamin K1, second
group received high vitamin K1 diet, and the third group
received high vitamin K2 for another 6weeks. Rats fed
high vitamin K showed a reduction of calcification by about
37% as compared to the group continued on warfarin and
the group fed low vitamin K. The reduction in calcification
was associated with increased carboxylated MGP expres-
sion in the vessel wall [13]. These findings were supported
in adenine-induced CKD rat models in which vitamin K
status was modified either by vitamin K antagonist or by
increasing dietary vitamin K intake. Rats who received
warfarin showed increased vascular calcification, while
rats who received dietary vitamin K showed decreased
Table 7 Study parameters
associated with CIMT > 7.5mm
Mann Whitney test for non-parametric quantitative data between the two groups. Chi-square test or Fisher
exact test for qualitative data between the two groups
*Significant level at P value < 0.05
CIMT P value
≤ 7.5 > 7.5
N = 60 N = 100
Age (years) Median/IQR 33/(24–38.5) 58/(48–64) 0.008*
Sex Male
Female
20 (33.3%)
40 (66.7%)
50 (50%)
50 (50%)
0.164
HTN No
Yes
24 (40%)
36 (60%)
40 (40%)
60 (60%)
> 0.99
Diabetes No
Yes
60 (100%)
0 (0%)
74 (74%)
26 (26%)
0.001*
eCVD History No
Yes
42 (70%)
18 (30%)
68 (68%)
32 (32%)
0.852
HTN Duration (years) Median/IQR 4/(1.9–7.9) 7/(3.8–13.5) 0.033*
DM Duration (years) Median/IQR – 15/(10–21.5) –
CVD Duration (years) Median/IQR 3/(1–5.8) 4/(2–7) 0.168
Dialysis duration (years) Median/IQR 4/(1.5–7.1) 3.5/(1.5–5.5) 0.066
Calcium (mg/dL) Median/IQR 7.6/(7.1–7.8) 7.5/(7.2–7.8) 0.458
Phosphorus (mg/dL) Median/IQR 5.7/(5.2–7.4) 5.6/(5.2–6.3) 0.973
ALP (IU/L) Median/IQR 187.5/(143.3–272.5) 265.5/(180.8–320) 0.205
iPTH (pg/ml) Median/IQR 255/(145–400) 225/(200–400) 0.341
Vit K2 (ng/ml) Median/IQR 35.5/(30–40.3) 34/(28–44.5) 0.721
ucOC (ng/ml) Median/IQR 21/(12.7–39) 22.2/(15–38.1) 0.733
iOC (ng/ml) Median/IQR 3/(2–7) 2/(2–3.3) 0.081
ucOC/iOC ratio Median/IQR 7.1/(3.6–11.7) 9/(7.6–13.6) 0.268
Table 8 Simple regression analysis for prediction of carotid intimal-
medial thickening > 7.5mm
OR odds ratio, CI confidence interval, NA not applicable
*Significant level at P value < 0.05
OR 95% CI P value
Age (years) 1.143 1.082–1.207 < 0.001*
Diabetes history NA NA NA
HTN duration (years) 1.096 0.983–1.221 0.099
International Urology and Nephrology
1 3
vascular calcification in a dose-dependent manner [14].
The role of vitamin K was further supported in humans in
an interventional randomized non-placebo-controlled trial
designed to study the effect of vitamin K2 supplementation
on functional vitamin K deficiency in hemodialysis patients
which enrolled 53 stable hemodialysis patients divided into
3 groups and received increasing doses of vitamin K2 for
6weeks (45, 135, and 360μg/day, respectively) with 50 age-
matched healthy controls. Hemodialysis patients had higher
levels of PIVK-II, uncarboxylated MGP, and uncarboxylated
osteocalcin. Vitamin K2 induced dose- and time-dependent
decrease in uncarboxylated vitamin-K dependent proteins.
Serum fetuin-A was not affected in this study and a high
dose of vitamin K2 did not affect adversely routine labora-
tory parameters [15].
These observational data were tested in two big clinical
trials. The first was an interventional randomized study, the
vitamin K1 to slow Vascular Calcification in Hemodialysis
Patients (VitaVasK study) in Europe, and the second was a
placebo-controlled study inhibiting the progression of arte-
rial calcification with vitamin K in hemodialysis patients
(iPACK-HD) trial in Canada, with the results of both have
not yet published. VitaVasK study assessed the impact of
vitamin K2 treatment on the cardiovascular calcification
with the progression of coronary artery calcification as the
primary outcome, and the progression of aortic and valvular
calcification as secondary outcomes. However, in a recent
1-year interventional randomized trial to assess the impact
of vitamin K2 in the prevention of vascular calcification
progression that enrolled 102 hemodialysis patients with
measurement of uncarboxylated MGP as a marker of vita-
min K deficiency, serum uncarboxylated MGP was found
to be reduced without an effect on the progression of aortic
calcification at the end of the trial [16].
In our study, we found that serum vitamin K2 in hemo-
dialysis patients was significantly higher than in healthy
controls. This was associated with a significant increase
in serum undercarboxylated osteocalcin that reflected vita-
min K deficiency. Combined the two findings together, this
is explained by subclinical vitamin K deficiency in hemo-
dialysis patients. This was proved in patients with stage 3
to 5 CKD [10] as well as dialysis patients [10, 15, 17–19].
In another study conducted on 188 hemodialysis patients
to assess the correlation of circulating inactive vitamin
K-dependent proteins with all-cause or cardiovascular
mortality, 64% of hemodialysis patients was found to have
profound vitamin K deficiency as indicated by increased
PIVKA-II. Patients on vitamin K antagonist in this study
had the highest level of PIVKA-II [17]. In all these studies
measurement of serum vitamin K was not correlated with
levels of inactivated vitamin-K dependent proteins. Func-
tional vitamin K deficiency in most hemodialysis patients
is attributed to the dietary restrictions as a low phosphorus
diet is deficient in vitamin K2 and low potassium diet is
deficient in vitamin K1. Vitamin K1 can be converted to
vitamin K2 after ingestion so both diets can affect vitamin
K2 levels. In addition, uremic milieu inhibits γ glutamyl
carboxylase which is essential for vitamin K recycling.
We assessed serum ucOC/iOC ratio in our study cohort.
We found that the median ratio was 8.7 (5.5–13.2). In a
previous study which enrolled 189 hemodialysis patients
together with 89 pre-dialysis CKD patients, the serum
ucOC/iOC ratio > 1.0 was observed in about 71.4% of
hemodialysis patients especially those with high bone
turnover [20]. However, we found no correlation between
serum osteocalcin and cardiovascular calcification whether
carotid media-intimal calcification or thickening, nor val-
vular thickening or calcification. Our findings are sup-
ported by data in a recent meta-analysis of 46 studies
which showed no definitive correlation between OC and
cardiovascular calcification with mixed positive and nega-
tive correlations [5].
Our study showed that age, hypertension duration and
established cardiovascular disease duration were the pre-
dictors of carotid media-intimal calcification; but only age
was a strong predictor of carotid intimal-medial thick-
ness > 7.5mm. Female sex was associated with a valvular
thickening. Carotid intimal-medial thickness is an inde-
pendent predictor of cardiovascular events and all-cause
mortality in chronic hemodialysis patients [21–23]. This
is in accordance with an old study that found no relation-
ship between traditional cardiovascular risk factors nor
the duration of hemodialysis and CIMT or plaques with
age is the only significant determinant of plaques [23].
In a prospective study to evaluate the utility of CIMT for
evaluation of coronary artery disease in pretransplant
ESRD, CIMT > 0.75 was found to be a strong predictor of
CAD with sensitivity and specificity of 90.47% and 73%,
respectively [24].
Although vascular calcification is more common in dia-
betic patients than the general population [25], we found
no correlation between diabetes mellitus and the presence
of carotid media-intimal calcification, valvular calcifica-
tion, and valvular thickening. This may be attributed to the
small number of diabetic patients in the study participants.
However, all diabetic patients had CIMT > 7.5mm which
reflected the impact of diabetes on vascular calcification.
In conclusion, we found that there is functional vitamin
K deficiency in hemodialysis patients, but no correlation
between serum osteocalcin and vascular calcification in
maintenance hemodialysis patients unlike MGP. Traditional
cardiovascular risk factors are not associated with cardio-
vascular calcification in hemodialysis patients. The cur-
rent study is limited by the small number of patients, being
single-center study, and future larger studies are needed to
confirm our findings.
International Urology and Nephrology
1 3
Acknowledgements We thank our colleagues from the Internal
Medicine department who provided insight and expertise that greatly
assisted the research.
Author contributions AMSS and HKHK equally contributed to the
conception and design of the research; NIO contributed to the design
of the research; MAYB contributed to the acquisition and analysis of
the data; AMSS and NIO contributed to the interpretation of data;
and HKHK and MAYB drafted the manuscript. All authors critically
revised the manuscript, agree to be fully accountable for ensuring the
integrity and accuracy of the work, and read and approved the final
manuscript.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflicts of
interest.
Statement of Ethics The research was conducted in accordance with
the ethical standards of the Minia University committee on human
experimentation and with the Helsinki Declaration of 1975, as revised
in 2008. Informed consent was obtained from all patients for being
included in the study.
References
1. Ruderman I, Holt SG, Hewitson TD, Smith ER, Toussaint ND
(2018) Current and potential therapeutic strategies for the man-
agement of vascular calcification in patients with chronic kidney
disease including those on dialysis. Semin Dial 31(5):487–499
2. Mizuiri S, Nishizawa Y, Yamashita K, Ono K, Naito T, Tanji
C etal (2019) Relationship of matrix Gla protein and vitamin
K with vascular calcification in hemodialysis patients. Ren Fail
41(1):770–777
3. Kirkpantur A, Altun B, Hazirolan T, Akata D, Arici M, Kirazli
S etal (2009) Association among serum fetuin-A level, coronary
artery calcification, and bone mineral densitometry in mainte-
nance hemodialysis patients. Artif Organs 33(10):844–854
4. Fusaro M, Gallieni M, Rizzo MA, Stucchi A, Delanaye P, Cavalier
E etal (2017) Vitamin K plasma levels determination in human
health. Clin Chem Lab Med 55(6):789–799
5. Millar SA, Patel H, Anderson SI, England TJ, O’Sullivan SE
(2017) Osteocalcin, vascular calcification, and atherosclerosis:
a systematic review and meta-analysis. Front Endocrinol (Laus-
anne) 8:183
6. Seidu S, Kunutsor SK, Khunti K (2019) Association of circulating
osteocalcin with cardiovascular disease and intermediate cardio-
vascular phenotypes: systematic review and meta-analysis. Scand
Cardiovasc J 53(6):286–295
7. Kim KM, Lim S, Moon JH, Jin H, Jung KY, Shin CS etal (2016)
Lower uncarboxylated osteocalcin and higher sclerostin levels
are significantly associated with coronary artery disease. Bone
83:178–183
8. D’Onofrio L, Maddaloni E, Buzzetti R (2019) Osteocalcin and
sclerostin: background characters or main actors in cardiovascular
disease? Diabetes Metab Res Rev 2:e3217
9. Holden RM, Iliescu E, Morton AR, Booth SL (2008) Vitamin
K status of Canadian peritoneal dialysis patients. Perit Dial Int
28(4):415–418
10. Holden RM, Morton AR, Garland JS, Pavlov A, Day AG, Booth
SL (2010) Vitamins K and D status in stages 3–5 chronic kidney
disease. Clin J Am Soc Nephrol 5(4):590–597
11. Price PA, Faus SA, Williamson MK (1998) Warfarin causes rapid
calcification of the elastic lamellae in rat arteries and heart valves.
Arterioscler Thromb Vasc Biol 18(9):1400–1407
12. Weijs B, Blaauw Y, Rennenberg RJ, Schurgers LJ, Timmermans
CC, Pison L etal (2011) Patients using vitamin K antagonists
show increased levels of coronary calcification: an observa-
tional study in low-risk atrial fibrillation patients. Eur Heart J
32(20):2555–2562
13. Schurgers LJ, Spronk HM, Soute BA, Schiffers PM, DeMey
JG, Vermeer C (2007) Regression of warfarin-induced medial
elastocalcinosis by high intake of vitamin K in rats. Blood
109(7):2823–2831
14. McCabe KM, Booth SL, Fu X, Ward E, Adams MA, Holden RM
(2017) Vitamin K metabolism in a rat model of chronic kidney
disease. Am J Nephrol 45(1):4–13
15. Westenfeld R, Krueger T, Schlieper G, Cranenburg EC, Magde-
leyns EJ, Heidenreich S etal (2012) Effect of vitamin K2 sup-
plementation on functional vitamin K deficiency in hemodialysis
patients: a randomized trial. Am J Kidney Dis 59(2):186–195
16. Oikonomaki T, Papasotiriou M, Ntrinias T, Kalogeropoulou
C, Zabakis P, Kalavrizioti D etal (2019) The effect of vitamin
K2 supplementation on vascular calcification in haemodialysis
patients: a 1-year follow-up randomized trial. Int Urol Nephrol
51(11):2037–2044
17. Schlieper G, Westenfeld R, Kruger T, Cranenburg EC, Magde-
leyns EJ, Brandenburg VM etal (2011) Circulating nonphospho-
rylated carboxylated matrix gla protein predicts survival in ESRD.
J Am Soc Nephrol 22(2):387–395
18. Cranenburg EC, Schurgers LJ, Uiterwijk HH, Beulens JW,
Dalmeijer GW, Westerhuis R etal (2012) Vitamin K intake and
status are low in hemodialysis patients. Kidney Int 82(5):605–610
19. Pilkey RM, Morton AR, Boffa MB, Noordhof C, Day AG, Su
Y etal (2007) Subclinical vitamin K deficiency in hemodialysis
patients. Am J Kidney Dis 49(3):432–439
20. Nagata Y, Inaba M, Imanishi Y, Okazaki H, Yamada S, Mori K
etal (2015) Increased undercarboxylated osteocalcin/intact osteo-
calcin ratio in patients undergoing hemodialysis. Osteoporos Int
26(3):1053–1061
21. Sato M, Ogawa T, Sugimoto H, Otsuka K, Nitta K (2012) Relation
of carotid intima-media thickness and silent cerebral infarction to
cardiovascular events and all-cause mortality in chronic hemodi-
alysis patients. Intern Med 51(16):2111–2117
22. Nishizawa Y, Shoji T, Maekawa K, Nagasue K, Okuno S, Kim
M etal (2003) Intima-media thickness of carotid artery predicts
cardiovascular mortality in hemodialysis patients. Am J Kidney
Dis 41(3 Suppl 1):S76–S79
23. Hojs R (2000) Carotid intima-media thickness and plaques in
hemodialysis patients. Artif Organs 24(9):691–695
24. Modi N, Kapoor A, Kumar S, Tewari S, Garg N, Sinha N (2006)
Utility of carotid intimal medial thickness as a screening tool for
evaluation of coronary artery disease in pre-transplant end stage
renal disease. J Postgrad Med 52(4):266–270
25. Chen NX, Duan D, O’Neill KD, Moe SM (2006) High glucose
increases the expression of Cbfa1 and BMP-2 and enhances the
calcification of vascular smooth muscle cells. Nephrol Dial Trans-
plant 21(12):3435–3442
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
A preview of this full-text is provided by Springer Nature.
Content available from International Urology and Nephrology
This content is subject to copyright. Terms and conditions apply.