Vitamin K is a cofactor in the post-translational carboxylation
of selective protein-bound glutamate residues which are con-
verted into gamma-carboxy glutamate (Gla) (1). To date, only a
dozen mammalian Gla-containing proteins (shortly: Gla-pro-
teins) have been identified. Vitamin K deficiency results in the
synthesis of under-carboxylated, biologically inactive Gla-pro-
teins. Examples of Gla-proteins are several blood coagulation
factors (all synthesized in the liver), osteocalcin (synthesised in
bone) and matrix Gla-protein (MGP), which is synthesized in a
number of non-hepatic tissues, notably cartilage and the arteri-
al vessel wall. Today MGP is regarded as a major inhibitor of
soft tissue calcification (2).
© 2004 Schattauer GmbH, Stuttgart
Beneficial effects of vitamins D and K on the elastic
properties of the vessel wall in postmenopausal women:
a follow-up study
Lavienja A. J. L. M. Braam
,Arnold P. G. Hoeks
, Fred Brouns
, Karly Hamulyák
Monique J.W. Gerichhausen
, Cees Vermeer
Departments of Biochemistry,
Internal Medicin, Cardiovascular Research Institute Maastricht (CARIM)
Department of Human Biology of the Nutrition and Toxicology Research Institute Maastricht (Nutrim), University of Maastricht,
Novartis Nutrition Research AG, Switzerland
Thromb Haemost 2004; 91: 373– 80
tics, including compliance coefficient (CC), distensibility coeffi-
cient (DC), intima-media thickness (IMT) and the Young’s
Modulus (E) were measured to assess the effect of the supple-
ments on the change of these parameters.The results showed
that the elastic properties of the common carotid artery in the
MDK-group remained unchanged over the three-year period,
but decreased in the MD- and placebo-group. Comparing the
MDK- and placebo-group, there were significant differences in
decrease of DC (8.8%; p<0.05), CC (8.6%; p<0.05), and in
increase of PP (6.3%; p<0.05) and E (13.2%, p<0.01).There were
no significant differences between the MD-group and placebo.
No significant differences were observed in the change of IMT
between the three groups. It is concluded that a supplement
containing vitamins K
and D has a beneficial effect on the elas-
tic properties of the arterial vessel wall.
Elastic properties, vessel wall, vitamin K, postmenopausal
Matrix-Gla Protein (MGP) is a strong inhibitor of vascular cal-
cification, the expression of which is vitamin D dependent. MGP
contains five γ-carboxyglutamic acid (Gla)-residues which are
formed in a vitamin K-dependent carboxylation step and which
are essential for its function. Hence vascular vitamin K-deficien-
cy will result in undercarboxylated, inactive MGP which is a
potential risk factor for calcification. In the present study we
describe the effects of vitamin K
and D supplementation on
vascular properties in postmenopausal women. In a randomized
placebo-controlled intervention study, 181 postmenopausal
women were given either a placebo or a supplement contain-
ing minerals and vitamin D (MD-group), or the same supple-
ment with vitamin K
(MDK-group). 150 participants com-
pleted the study and analysis was performed on 108 partici-
pants.At baseline and after three years, vessel wall characteris-
Cell Signalling and Vessel Remodelling
Department of Biochemistry, Maastricht University
P.O. Box 616, 6200 MD Maastricht,The Netherlands
Tel.: +31 433881680, Fax: +31 433884160
Received July 1, 2003
Accepted after revision December 1, 2003
This study was supported by Novartis Consumer Health SA (Nyon, Switzerland).
Prepublished online January 12, 2004 DOI: 10.1160/TH03-07-0423
Braam, et al.: Vitamin K and arterial elastic properties
Vascular calcification is an important factor that contributes
to considerable morbidity and mortality, notably in diabetics
and in hemodialysis and atherosclerotic patients (3).
Calcification may occur either in the tunica intima in associa-
tion with atherosclerosis or in the tunica media where it is
known as Mönckeberg’s sclerosis (4). Medial vascular smooth
muscle cells (VSMCs) synthesize most of the MGP in the ves-
sel wall (5). It was found that the overall arterial expression of
MGP was decreased in Mönckeberg’s sclerosis suggesting that
low levels of MGP may predispose to calcification (6).
However, in VSMCs adjacent to the calcium salt deposits, MGP
mRNA expression was substantially elevated which may repre-
sent a response to the locally increased calcium concentration in
order to enhance calcium clearance (5). Using monoclonal anti-
bodies against MGP Schurgers, et al. (7) demonstrated that
MGP also accumulates around the calcified areas.
Although the precise sequence of events in the association
between MGP-expression and calcification still needs to be elu-
cidated, a number of animal studies have established unequiv-
ocally that MGP is a potent inhibitor of calcification. The vas-
cular phenotype of the transgenic MGP null mouse showed
massive calcification of the large arteries within 4 weeks after
birth (8). The fact that comparable results were obtained in nor-
mal rats after treatment with the vitamin K-antagonist warfarin
demonstrated that Gla-residues are essential to the calcification
inhibitor function of MGP (2).
Whereas vitamin K is involved in the posttranslational pro-
cessing of MGP, vitamin D has a role in the regulation of MGP
gene expression. Fraser, et al. have shown that the MGP promo-
tor contains a vitamin D response element that is responsible for
a 2-3 fold enhancement of MGP expression after vitamin D
binding (9). However, to our knowledge, a direct association
between vitamin D deficiency and arterial media calcification
has not yet been demonstrated. With respect to vitamin K, it was
shown in postmenopausal women that low vitamin K
a risk factor for aortic calcification (10). In an independent pop-
ulation-based study among 4500 elderly subjects, an inverse
correlation was demonstrated between vitamin K
aortic calcification, myocardial infarction and sudden cardio-
vascular death (11). Based on these findings, it has been sug-
gested that in a substantial part of the population the vitamin K
status of the arterial vessel wall is inadequate to support full
MGP carboxylation (7). We wish to put forward the hypothesis
that local vitamin K deficiency forms a risk factor for vascular
hardening, increasing stiffness, and loss of elastic properties.
The present study forms part of a large intervention trial in
which we investigated the effects of minerals (calcium, magne-
sium, and zinc) and vitamins D and K on bone mineral density
and vascular properties. Here we report the effects of minerals
+ vitamin D, and minerals + vitamins D and K supplementation
on the vascular properties during a three-year follow-up period.
Measures of functional vascular properties are the distensibility
coefficient (DC) and compliance coefficient (CC) as they char-
acterize the arterial stiffness of the vessel wall, which may be
modified by medial calcification. The intima-media thickness
(IMT) is assumed to represent an endothelial adaptive response
to physiological and pathophysiological processes (12-14) and
is regarded as an indicator of atherosclerosis (15, 16). The ratio
of diameter to IMT corrects the functional parameter DC for the
tissue volume involved, resulting in a parameter for the structu-
ral elasticity, known as the Young’s Modulus (E) (17). The
present study also included the effects of vitamins K
and D on
changes in IMT and E. We selected a group of healthy postmen-
opausal women to assess the effects of vitamin K
and D sup-
plementation on changes in vessel wall characteristics because
it has been suggested that changes in vessel wall properties
occur faster in postmenopausal women between 45 and 60 years
than in men of the same age (18).
The participants were enrolled in a 3-year double-blind placebo-
controlled clinical trial in which the effects of minerals, vita-
mins D and K were investigated on bone mineral density and
vessel wall characteristics. The participants of this trial were
recruited by newspaper advertisements. Inclusion criteria were:
apparently healthy women, Caucasian race, between 50 and 60
years old and at least 2 years postmenopausal. Exclusion crite-
ria were: use or recent use (< 1 year) of oral anticoagulants, cor-
ticosteroids, hormone replacement therapy and alcohol con-
sumption of > 2 glasses/day. Furthermore, persons using multi-
vitamins or food supplements recently (< 1 year) were exclud-
ed from participation. In this way, the effects of the intervention
could be studied without wash-out effects. In total 188 women
met the criteria for participation and were randomized into our
study. Information on cardiovascular risk factors, current health
status, medical history, drug use and smoking behaviour was
collected before the start of the study. No medical history of
important systemic diseases was reported. Within this trial par-
ticipants were seen every 3 months to check for physical health
and compliance of the treatment by short questionnaires; on a
number of these occasions bone densitometry (DXA) and blood
sampling were performed. The vascular examination only took
place at baseline and at the end of the study after 3 years.
Baseline measurements were performed between November
1997 and March 1998, and the follow-up examination took
place between December 2000 and March 2001. All participants
gave written informed consent and the trial was approved by the
University Hospital medical ethics committee.
The subjects were randomized into three groups. Of the total
group of 188 women, 7 women were referred for treatment out-
Braam, et al.: Vitamin K and arterial elastic properties
side the study because abnormalities were found during the
baseline measurements. In the first group (n=60) participants
received a placebo (maltodextrine, i.e. placebo group), in the
second group (n=58) participants received a supplement con-
taining 500 mg calcium (natural calcium complex derived from
milk), 10 mg zinc, 150 mg magnesium and 8 µg vitamin D
(minerals + vitamin D = MD-group), and in the third group
(n=63) participants received a supplement containing the same
constituents as the MD-group but with additional 1 mg vitamin
(minerals + vitamins D+K = MDK-group). In the present
study, we compared the changes of vessel wall characteristics
between treatment groups and placebo to elucidate the effects of
vitamins K and D on the vessel wall characteristics.
The randomization of the participants to the three groups
was performed according to a computer-generated randomiza-
tion list and the randomization codes were kept apart from the
study site during the trial. The participant randomization codes
were allocated sequentially in the order in which the partici-
pants were enrolled. One investigator who supervised the whole
study was responsible for the enrollment and assignment of par-
ticipants to the respective groups. Because the three different
types of supplements were similar in appearance and taste, par-
ticipants and investigators were not aware of group assignment.
Participants were allowed to choose between a supplement in
the form of a tasteless powder (to be mixed with water before
intake) or in the form of chocolate-coated tablets with a crunchy
malt core. The percentage of subjects who used the powder or
tablets was equally distributed across the three groups.
Participants were instructed to take one sachet with powder or
three tablets per day during evening hours, preferably after the
meal. Also, they were advised to maintain their usual diets and
to avoid taking supplements containing either calcium, vitamin
D, or K throughout the study. Novartis Consumer Health SA
(Nyon, Switzerland) prepared and provided all supplements.
After randomization, the women received the first batch of sup-
plements and were supplied with a new batch of supplements
every six months.
The primary outcome measures for the purpose of this study
were the vessel wall characteristics of the common carotid
artery measured with ultrasound (7.5 MHZ, ATL Mark V). The
ultrasonic vessel wall tracking system (WTS) to determine arte-
rial wall properties has been described in detail previously (19,
20). It provides estimates of the arterial end-diastolic adventitia-
adventitia diameter (d) and the change in diameter (distension)
from diastole to systole (∆d) normalized for the end-diastolic
diameter (∆d/d) for each captured heart beat. In parallel with
diameter change measurements, arterial blood pressure was
recorded at the level of the brachial artery by means of a semi-
automated oscillometric device (DINAMAP). Pulse pressure
(PP), defined as systolic minus diastolic blood pressure, was
determined by averaging the three measurements nearest to the
The intima-media thickness (IMT) was measured simulta-
neously at the same location (2-3 cm proximal to the bifurca-
tion) of the common carotid artery where the diameter and
diameter changes were measured. Only the IMT of the posteri-
or wall was assessed because here the reflections from the
blood-intima and media-adventitia transition are distinctly vis-
ible, whereas at the anterior wall, the trailing edge of the adven-
titial reflections may obscure the medial and intimal signal. At
the end of the session, recorded IMT-files were processed
employing the wall thickness program. The threshold for the
derivative was maintained at 0.025 (21). Each heart beat within
a recording resulted in an estimate of wall thickness; the medi-
an of the estimates per recording was used for further evalua-
tion. From d, ∆d, PP and IMT, we have calculated the vascular
distensibility coefficient (DC), the compliance coefficient (CC)
and the elasticity coefficient (E) according to the following
DC= (2d∆d + ∆d
CC= π(2d∆d + ∆d
E = d / (IMT*DC)
The examinations were performed in a quiet room with a con-
trolled temperature of 21 ± 2
C. The measurements were made
at the same moment of the day to avoid diurnal variations and
after a 10-15 minute supine rest to stabilize blood pressure lev-
els. Participants were asked to refrain from smoking and the
consumption of alcohol-, caffeine-, or quinine-containing bev-
erages at least 3 hours prior to the examination. To save time we
have investigated only the right common carotid artery. To our
knowledge no significant differences between the wall proper-
ties of the right and the left common carotid artery have ever
been reported. The same investigator performed all examina-
tions at the start and the end of the study and for each partici-
pant several repeated measurements (5-7) were made during
one session, the average of which was calculated and used for
the analysis of the data. The reproducibility of the method is
around 10% for the distension, DC, CC (20) and IMT (22).
Before the vascular examination, height and weight of each
participant were measured with standardized equipment to esti-
mate the body mass index (weight/height
The sample size was calculated on the assumption that the
desired minimal detectable effect was a 15% reduced decrease
in distensibility of the MDK-group compared to the placebo
group with a 90% power and a 0.05 level of significance. With
the assumption of a dropout rate of 10% per year we calculated
that 180 subjects had to be included. Statistical analysis was
performed using the Statistical Package SPSS (SPSS Corp,
Braam, et al.: Vitamin K and arterial elastic properties
Chicago, IL). Results are presented as means ± standard devia-
tion (SD), unless indicated otherwise. Only participants who
had completed the study were included in the primary outcome
analysis. Furthermore, participants who had started to use med-
ication directly affecting the vessel wall during the study and
participants with atherosclerotic plaques in the common carotid
artery, were excluded from analysis. Also, participants with a
high variability in their results were excluded (arterial transla-
tion of > 2 mm and beat-to-beat variation in distension of
> 20%). Namely, a beat-to-beat variation in distension of > 20%
would only be explainable by difficulties in the recording tech-
nique rather than by changes in blood pressure. Furthermore, for
an arterial translation of > 2 mm it is highly unlikely that this
motion is restricted to the plane of observation. Out of plane
motion induces false diameter and distension measurements.
The primary outcome analyses were repeated to exclude the
possibility of bias by drop-outs and excluded subjects. In these
repeat analyses, the missing values of drop-outs and excluded
subjects were replaced by mean values of outcome variables of
the total population of 108 participants. A paired t-test was used
to evaluate the change in the vessel wall characteristics over the
three years within each group. We considered a level of p<0.05
to be statistically significant. For every participant, the percent-
age change from baseline in all parameters was calculated and
the mean change from baseline was calculated per group.
Primary outcome analysis consisted of comparison of the
change in DC, CC, PP, IMT, d/IMT and E between the MD-
group and placebo and between the MDK-group and placebo
using linear regression analysis. In this analysis, the change in
vascular parameters relative to baseline was used as dependent
variable and the treatment groups and several covariates were
used as explanatory variables. Baseline values of age, weight,
smoking (yes or no), heart rate and mean arterial pressure were
chosen as covariates because their influence on the change in
vascular properties or response to the supplementation could
not be excluded.
Baseline characteristics of the selected participants are present-
ed in Table 1. The MD group differed slightly but significantly
from the placebo and the MDK group with respect to age and
the number of years since menopause. No significant differ-
ences were observed in the baseline values of the vessel wall
characteristics. Figure 1 shows the flow of participants through
each stage for the separate groups. Of the 181 subjects who
entered the study, 31 discontinued their participation during the
course of the study and were not available for the follow-up
measurement. Of the remaining 150 participants who complet-
ed the study, we excluded 42 subjects from the analysis:
4 women started to use medication known to have direct effects
on the vessel wall (ACE-inhibitors, β-blockers or Calcium-
antagonists), in 11 subjects atherosclerotic plaques in the com-
mon carotid artery were clearly visible and they were referred
for treatment outside the study, and 27 subjects had a high var-
iability in their repeated measurements (arterial translation of
> 2 mm and beat-to-beat variation in distension of > 20%).
Analysis was performed on 108 participants who completed the
study and who did not have any of the above mentioned inter-
fering factors. There were no significant differences in baseline
values of the vessel wall characteristics between the 108
selected participants and those who discontinued treatment or
Table 1: Baseline characteristics
(mean ± standard deviation) in the
placebo group (n=40), MD-group
(n=30) and MDK-group (n=38).
*significant different from placebo
(p<0.05), ** significant different from
placebo and MDK-group (p<0.05).
Abbreviations used: body mass index
(BMI), compliance coefficient (CC),
distensibility coefficient (DC), intima-
media thickness (IMT), Elasticity (E).
Braam, et al.: Vitamin K and arterial elastic properties
were excluded (n=73). The only side-effect of the allocated
intervention reported to the investigator were complaints of
mild constipation in a few participants of the MD-group (n=4)
and the MDK-group (n=3). No further adverse events occurred
during the study.
Vascular parameters of elasticity
Table 2 summarizes per group the differences between the mean
values at baseline and the end of the study for all vascular
parameters with their paired-levels of significance. Changes in
percentages between t=0 and t=3 yr are given as well. The DC
and CC in the placebo group decreased significantly with 10%
and 6%, respectively. The PP, on the other hand increased by
7%, but the increase did not reach the level of significance. In
the MD-group, DC decreased significantly with 7% and CC
decreased with 4%, while the PP increased with 6%, however
these latter two changes did not reach the level of significance.
In the MDK-group, the DC, CC and PP remained constant over
the three years period, with even a tendency for the CC to
increase (+3%). Figure 2 illustrates the change in DC and CC of
the three groups. After adjustment for baseline heart rate, mean
arterial pressure, age, weight and smoking, there were signifi-
cant differences between the MDK- group and the placebo with
respect to DC (8.8%, 95% CI: 1.9 to 21.4), CC (8.6%, 95% CI:
1.8 to 20.3), and PP (-6.3%, 95% CI: -17.1 to –0.7). In the same
analysis no differences were found between the MD- and place-
bo-group with respect to DC (2.5%, 95% CI: -6.3 to 14.8), CC
(2.2%, 95% CI: -6.3 to 13.8), and PP (-0.11%, 95% CI: -12.1 to
5.6). The repeat analyses in which mean values were allocated
to drop-outs and excluded subjects, showed the same trends as
those described above; as to be expected differences between
the MDK-group and placebo in change of DC, CC and PP were
smaller, but they remained statistically significant.
Changes of the intima-media thickness relative to baseline are
also portrayed in Table 2. The IMT increased in all three groups:
9% in the placebo (p<0.01), 10% in the MDK-group (p<0.01),
Figure 1: Flow diagram of participants through the different stages of the study.
Braam, et al.: Vitamin K and arterial elastic properties
but only 4% in the MD-group (p=0.32). The d/IMT ratio
decreased significantly in the placebo and MDK-group by 3.8%
and 6.5% respectively, while in the MD-group the ratio
remained constant. The Young’s Modulus (E) increased in the
placebo and MD-group by 13.2% and 13.7% respectively, how-
ever, these changes did not reach the level of significance. In the
MDK-group, E remained constant. Figure 2 illustrates the
change in IMT and E of the three groups. In the multivariate
analysis with adjustments for baseline heart rate, mean arterial
pressure, age, weight, and smoking, the difference in increase of
IMT between the MDK-group and placebo was 1.3% (95% CI:
-3.2 to 9.2) and between the MD-group and placebo the differ-
ence was -4.5% (95% CI: -8.9 to 4.2). Hence a significant ben-
eficial effect on the age-related increase of the IMT could not be
demonstrated for either the MD- nor the MDK-supplement.
However, there was a significant difference in the change of E
between the MDK-group and placebo (-13.2%; 95% CI: -35.8
to -5.3), while no significant difference was observed between
the MD-group and placebo (0.46%; 95% CI: -23.2 to 9.4) and
no significant differences were observed between the groups in
change of the d/IMT ratio.
In the present investigation, we have demonstrated a long-term
beneficial effect of a supplement containing vitamins K
on the elastic properties of the carotid artery. In contrast, no
effect was found on the IMT. Vitamin D alone did not influence
these variables. To our knowledge, this is the first study which
shows a longitudinal beneficial effect of vitamins K
+ D sup-
plementation on vascular properties.
The occurrence of vitamin K-dependent proteins in the arte-
rial vessel wall was discovered recently, and only limited data
are available on a possible relation between vitamin K intake
and vessel wall properties. In animals it was found that phar-
macological doses of vitamin K
prevent the progression of ath-
erosclerosis by supression of plaque formation, intima-thicken-
ing and pulmonary atherosclerosis (23). One of the few studies
in humans showed an inverse correlation between dietary vita-
intake and aortic calcification in postmenopausal
women (10). In a recent population-based study an inverse cor-
relation was found between vitamin K
intake and aortic calcifi-
cation, myocardial infarction and sudden cardiovascular death
(11). The latter study suggests that vitamin K
may be a more
powerful inhibitor of arterial calcification than vitamin K
Table 2: Change in vessel wall characteristics (mean ± standard deviation and change in percentages) in the placebo group (n=40),
MD-group (n=30) and MDK-group (n=38) after three years of treatment. *significant different from placebo (p<0.05); P-value in
parenthese with respect to baseline. Abbreviations used: compliance coefficient (CC), distensibility coefficient (DC), intima-media
thickness (IMT), Elasticity (E).
Figure 2: Mean percent change (± standard error of the mean)
from baseline (0%) in distensibility coefficient, compliance coeffi-
cient, intima-media thickness and the Young’s Modulus (E) of the
common carotid artery in the placebo group (n=40), MD-group
(n=30) and MDK-group (n=38) after three years of treatment.
*significant different from placebo (p< 0.05).
Braam, et al.: Vitamin K and arterial elastic properties
Furthermore, vitamin K
was shown to decrease the total circu-
lating cholesterol concentration (24). Most of the studies pub-
lished so far have focused on the effects of vitamin K
of vitamin K
When hypothesizing about the mechanism underlying our
observations, we would like to focus on the vitamin K-depen-
dent protein MGP. It is synthesized by the vascular smooth mus-
cle cells (5), and accumulates in or around the elastic fibres in
the tunica media (7). In this respect it is worth mentioning that
transgenic MGP-deficient mice developed arterial calcifications
starting in the media, but even at later stages there was no neo-
intima formation or atherosclerosis. Rather, the type of calcifi-
cation was similar to that found in Mönckeberg’s sclerosis of
the media such as is often seen in diabetics and hemodialysis
patients. In these patients calcification starts from the elastic
lamellae of the media and occurs without inflammation. DC and
CC are related to the functions of elastin and collagen in the
vascular media as they represent the elastic properties of the
vessel wall. IMT, on the other hand, is regarded as endothelial
response to pathophysiological processes as in atherosclerosis.
While DC and CC represent functional characteristics of the
vessel wall, the Young’s Modulus (E) expresses the structural
characteristics of the tissue. In this study it was shown that sup-
plementation with vitamins D+K exerted beneficial effects on
both structural and functional characteristics of the vessel wall.
In many papers it has been suggested that the vitamin K
requirement of extra-hepatic tissues is substantially higher than
that of the liver. Although these conclusions are mainly based
on the bone Gla-protein osteocalcin, which was found to be
undercarboxylated in the majority of the population (notably
elderly) (25), substantial undercarboxylation has also been
reported for MGP (26). Assuming that at baseline part of the
MGP was synthesized in an undercarboxylated (i.e.: inactive)
form, this must be expected to form a risk factor for media
sclerosis and vascular stiffening, but not for atherosclerosis.
With regard to this, it is conceivable that increased vitamin K
intake will increase the vascular vitamin K status and hence the
production of active MGP, thus contributing to the inhibition of
calcification and protection against arterial stiffening. Although
this is precisely the effect observed in our study, it cannot be
excluded that other vitamin K-dependent proteins are involved
in maintaining vascular elasticity.
Since the MGP promotor contains a vitamin D-responsive
element, it is theoretically possible that a low vitamin D status
also affects the level of MGP expression. To make certain that
the participants were sufficient in vitamin D, the treatment
included supplementation with vitamin D (1 RDA) together
with the vitamin K
(8 RDA). In a third arm of the study it
was demonstrated that vitamin D alone had no effect on any
of the vessel wall characteristics measured. No significant
differences were observed in longitudinal changes between the
placebo and the vitamin D group.
To which extent vessel wall characteristics of the carotid
artery represent a measure for risk on cardiovascular disease
remains to be investigated. In various studies, increased IMT of
the carotid artery was shown to be a risk factor for atheroscler-
osis, especially if the IMT is > 1 mm. Since in our study popu-
lation the IMT in all three groups remained between 0.6 –
0.7 mm, it cannot be excluded that under different conditions
(e.g. in older age groups, or after prolonged treatment) an effect
of vitamin K on IMT will also be seen. This may especially
occur at later stages of atherosclerosis in which calcification of
the lesions forms an end stage process. Less information is
available on the correlation of distensibility and compliance
with cardiovascular disease, although in previous studies arte-
rial stiffness was also shown to be associated with myocardial
infarction and coronary artery disease (27, 28). In a recent
study of 110 end-stage renal disease patients, a strong correla-
tion was found between calcification score and arterial stiff-
ness, especially with E, but also with DC, CC and IMT (29). It
has been suggested in this study that measurement of arterial
parameters, exploring both structural and functional properties,
could be helpful in the assessment of risk on cardiovascular
disease and in the evaluation of risk reduction by treatment.
However, extrapolation of these results to other groups may not
be justified because of the particular characteristics of these
patients. The fact that our data suggest a more pronounced
effect of vitamin K1 supplements on (medial) calcification than
on IMT thickening warrants further studies in other popula-
tions, for instance diabetics or patients with severe atheroscle-
One limitation of our study is the use of the brachial pulse
pressure instead of the carotid pulse pressure. The brachial pulse
pressure tends to deviate from the pulse pressure in the common
carotid artery. One should realize, however, that intra-subject
errors due to differences in the locations of blood pressure
measurement and artery wall property assessment will be
methodological in nature and of the same order of magnitude at
the various moments of determination. Changes in heart rate
and the associated changes in aorta pulse pressure will modify
the brachial and carotid pulse pressure to the same extent. The
basic question is whether for the 3 subgroups the brachial and
carotid pulse pressure will vary in parallel and to the same
extent. For young adults, the carotid pulse pressure is about
70% of that of the brachial artery, while the difference gradual-
ly vanishes with advancing age. For the considered age group
the difference will be considerably less. Using the brachial
pulse pressure the carotid wall parameters will be underestimat-
ed. If in the control group the pulse pressure percentage
increases more than in the MDK-group, then the DC of the con-
trol group would deviate even more, thus supporting our main
conclusion. Since both the control and MD-group developed in
parallel, there is no reason why the carotid arteries of the MDK-
group should age faster.
Braam, et al.: Vitamin K and arterial elastic properties
A further limitation of our study is the fact that we observed
a higher drop-out rate in the treatment groups than in the place-
bo group. Given the specific reasons for drop-out as mentioned
in the results, it is unlikely that the treatment assignment was
responsible for the higher drop-out rate. However, it cannot be
excluded that the higher drop-out rate was due to minor differ-
ences in taste of the different supplements. Another limitation of
this study was the exclusion of part of our randomized popula-
tion. Extensive analysis of the collected ultrasound data lead to
necessary exclusions because of technical difficulties in meas-
uring diameter changes within regular heart cycles, as exhibited
by example arterial translations of > 2 mm and beat-to-beat
variations in distension of > 20%. Additional analyses were per-
formed to exclude the possibility that the high drop-out rate and
exclusion of subjects had seriously biased our results. This
turned out not to be the case. The study presented in this paper
is a first step to corroborate the role of vitamin K in vascular tis-
sue. Clinical trials measuring other endpoints and progression
of calcification are needed to provide further evidence that vita-
min K contributes to the prevention of cardiovascular disease.
The statistical advice of Dr. A. Kester (Department of Biostatistics, UM) and
the technical assistance of J.M. Willigers (Department of Biophysics) is grate-
fully acknowledged. Finally the authors wish to thank Professor J. Rosing for
critically reading this manuscript.
1. Vermeer C. Gamma-carboxyglutamate-
containing proteins and the vitamin K-depen-
dent carboxylase. Biochem J 1990; 266:
2. Price PA, Faus SA, Williamson MK. Warfarin
causes rapid calcification of the elastic lamel-
lae in rat arteries and heart valves. Arterioscler
Thromb Vasc Biol 1998;18:1400-7.
3. Shanahan CM, Proudfoot D, Farzaneh-Far A,
et al.. The role of gla proteins in vascular cal-
cification. Crit Rev Eukaryot Gene Expr 1998;
4. Shearer MJ. Role of vitamin K and Gla pro-
teins in the pathophysiology of osteoporosis
and vascular calcification. Curr Opin Clin
Nutr Metab Care 2000; 3: 433-8.
5. Shanahan CM, Cary NR, Metcalfe JC, et al..
High expression of genes for calcification-reg-
ulating proteins in human atherosclerotic
plaques. J Clin Invest 1994; 93: 2393-402.
6. Shanahan CM, Cary NR, Salisbury JR, et al..
Medial localization of mineralization-regulat-
ing proteins in association with Monckeberg’s
sclerosis: evidence for smooth muscle cell-
mediated vascular calcification. Circulation
1999; 100: 2168-76.
7. Schurgers LJ, Dissel PEP, Spronk HMH, et al..
Role of vitamin K and vitamin K-dependent
proteins in vascular calcification. Z Kardiol
2001; 90: 57-63.
8. Luo G, Ducy P, McKee MD, et al..
Spontaneous calcification of arteries and car-
tilage in mice lacking matrix GLA protein.
Nature 1997; 385: 78-81.
9. Fraser JD, Price PA. Induction of matrix Gla
protein synthesis during prolonged 1,25- dihy-
droxyvitamin D3 treatment of osteosarcoma
cells. Calcif Tissue Int 1990; 46: 270-9.
10. Jie KS, Bots ML, Vermeer C, et al.. Vitamin K
intake and osteocalcin levels in women with
and without aortic atherosclerosis: a popula-
tion-based study. Atherosclerosis 1995; 116:
11. Geleijnse J, Vermeer C, Schurgers LJ, et al..
Inverse association of dietary vitamin K2-
intake with cardiac events and aortic atheros-
clerotic: The Rotterdam Study. Thromb
Haemost 2001; Supplement ISTH XVIII:
12. Salonen R, Salonen JT. Progression of carotid
atherosclerosis and its determinants: a popula-
tion-based ultrasonography study. Atheroscle-
rosis 1990; 81: 33-40.
13. Salonen JT, Salonen R. Ultrasonographically
assessed carotid morphology and the risk of
coronary heart disease. Arterioscler Thromb
1991; 11: 1245-9.
14. O’Leary DH, Polak JF, Kronmal RA, et al..
Thickening of the carotid wall. A marker for
atherosclerosis in the elderly? Cardiovascular
Health Study Collaborative Research Group.
Stroke 1996; 27: 224-31.
15. Bots ML, Hofman A, De Jong PT, et al..
Common carotid intima-media thickness as an
indicator of atherosclerosis at other sites of the
carotid artery. The Rotterdam Study. Ann
Epidemiol 1996; 6: 147-53.
16. Bots ML, Hofman A, Grobbee DE. Increased
common carotid intima-media thickness.
Adaptive response or a reflection of athero-
sclerosis? Findings from the Rotterdam Study.
Stroke 1997;28: 2442-7.
17. Hoeks APG, Reneman RS. Biophysical princi-
ples of vascular diagnosis. J Clin Ultrasound
1995; 23: 71-9.
18. Hansen F, Mangell P, Sonesson B, et al..
Diameter and compliance in the human com-
mon carotid artery–variations with age and
sex. Ultrasound Med Biol 1995; 21: 1-9.
19. Hoeks APG, Brands PJ, Smeets FA, et al..
Assessment of the distensibility of superficial
arteries. Ultrasound Med Biol 1990; 16:
20. Kool MJF, van Merode T, Reneman RS, et al..
Evaluation of reproducibility of a vessel wall
movement detector system for assessment of
large artery properties. Cardiovascular
Research 1994; 28: 610-4.
21. Hoeks APG, Willekes C, Boutouyrie P, et al..
Automated detection of local artery wall
thickness based on M-line signal processing.
Ultrasound Med Biol 1997; 23: 1017-23.
22. Willekes C, Hoeks APG, Bots ML, et al..
Evaluation of off-line automated intima-media
thickness detection of the common carotid
artery based on M-line signal processing.
Ultrasound Med Biol 1999; 25: 57-64.
23. Kawashima H, Nakajima Y, Matubara Y, et al..
Effects of vitamin K2 (menatetrenone) on ath-
erosclerosis and blood coagulation in hyper-
cholesterolemic rabbits. Jpn J Pharmacol
1997; 75: 135-43.
24. Nagasawa Y, Fujii M, Kajimoto Y, et al..
Vitamin K2 and serum cholesterol in patients
on continuous ambulatory peritoneal dialysis.
Lancet 1998; 351: 724.
25. Knapen MHJ, Hamulyak K, Vermeer C. The
effect of vitamin K supplementation on circu-
lating osteocalcin (bone Gla protein) and uri-
nary calcium excretion. Ann Intern Med 1989;
26. Price PA, Williamson MK. Primary structure
of bovine matrix Gla protein, a new vitamin
K-dependent bone protein. J Biol Chem 1985;
27. Dart AM, Lacombe F, Yeoh JK, et al.. Aortic
distensibility in patients with isolated hyper-
cholesterolaemia, coronary artery disease, or
cardiac transplant. Lancet 1991; 338: 270-3.
28. Hirai T, Sasayama S, Kawasaki T, et al..
Stiffness of systemic arteries in patients with
myocardial infarction. A noninvasive method
to predict severity of coronary atherosclerosis.
Circulation 1989; 80: 78-86.
29. Blacher J, Guerin AP, Pannier B, et al..
Arterial calcifications, arterial stiffness and
cardiovascular risk in end-stage renal disease.
Hypertension 2001; 38: 938-42.