Warfarin tedavisinin koroner kalsifikasyonda rolü var mıdır?

PDF Dosyası: 

arastirmax_1423_pp_26-30.pdf

Introduction
Warfarin is a widely used agent in cardiovascular
practice, especially in patients with prosthetic valves. Some recent reports have revealed that warfarin
inhibits the matrix γ-carboxyglutamic acid protein
(MGP), which acts as a calcification inhibitor in vivo
(1). In humans, defects in the MGP gene that predict
a nonfunctional MGP protein have been shown to
be responsible for Keutel syndrome (2). This inherited rare disease is characterized by abnormal calcification of cartilages, including costal, nasal, auricle,
tracheal, and growth plate cartilage, nasal hypoplasia
and brachtelephalangia; and by multiple peripheral
pulmonary artery stenoses (3,4). In mice, targeted deletion of the MGP-gene causes rapid calcification of
the elastic lamellae of the arterial media that begins
at birth and is sufficiently extensive by 3 to 6 weeks
of age that the arteries become rigid tubes that fracture, causing death by exsanguination in most of the
affected mice by 6 weeks of age (5). Treatment of rats
with the vitamin K antagonist, warfarin, at doses that
inhibit the γ-carboxylation of MGP causes rapid calcification of elastic lamellae of arteries and of aortic
heart valves and increased expression of MGP mRNA
in the calcifying artery (1). Calcification is a common
finding in the pathophysiology of the aging human
artery and heart valve and is associated with several
cardiovascular disease states. Warfarin-induced calcification in arteries is observed to occur only in the
growing animal, and neither seen in older, nongrowing rats nor in young rats whose growth was temporarily arrested by a calorically-restricted diet. Besides
more recent studies speculated that serum MGP levels
were inversely correlated with the severity of of coronary artery calcium load (6), and oral anticoagulation
might be associated with increased valvular and coronary calcium in patients with aortic valve disease (7).
Multidetector computed tomography was previously shown to be sensitive in determining coronary
* Department of Cardiology, Gulhane Military Medical Faculty
** Department of Cardiovascular Surgery, Gulhane Military Medical Faculty
*** Department of Radiology, Gulhane Military Medical Faculty
Reprint request: Dr. Mehmet Yokuşoğlu, Department of Cardiology, Gulhane
Military Medical Faculty, Etlik-06018, Ankara, Turkey
E-mail: myokusoglu@yahoo.com
Date submitted: February 10, 2010 • Date accepted: November 12, 2010Volume 53 • Issue 1 Role of warfarin in coronary calcifi cation • 27
artery calcification (8-10). Our aim was to investigate
the effect of warfarin use on coronary calcification
detected by multidetector computed tomography in
patients with mechanical prosthetic heart valves due
to rheumatic valve disease.
Material and Methods
In this cross-sectional observational study a total
of 67 subjects were studied. Study group (Group I)
was composed of 39 patients (11 men and 28 women) with a mean age of 54±15 years, who underwent
prosthetic heart valve surgery 66±46 months ago due
to rheumatic valve disease. All patients were under
warfarin therapy at least for 36 months, with targeted
INR value between 2.5-3.5, and no history of thromboembolism and bleeding attributed to warfarin
was present. Twenty eight age- and gender-matched
healthy subjects (9 male and 19 female) with a mean
age of 53±12 years comprised the control group
(Group II). All patients were provided informed consent before the investigation in accordance with local
ethics committee requirements.
Prior to the study enrollment, blood lipid analysis
was performed in all subjects. Patients with triglyceride levels higher than 400 mg/dl (due to the inability to calculate low-density lipoprotein values using
Friedewald formula), patients with lipid-lowering
medication and patients with parathyroid disease, increased serum calcium levels, and renal failure were
excluded.
Hypertension was defined as the blood pressure
over 140/90 mm Hg after 5 minutes rest or using antihypertensive medication, and diabetes mellitus was
defined as fasting plasma glucose level above 126 mg/
dl or using antidiabetic medication. Weight measurements were done with light clothes by using calibrated electronic scale, and height measurements with
naked foot. Smokers were defined as current smokers
or ex smokers.
Multidetector computed tomography (MDCT) was
performed in every subject using 16-slice CT scanner
(MX8000 IDT, Philiphs Medical Systems). To measure the calcium score, load and mass (prospective
electrocardiogram [ECG]-gated made, the collimation
space: 0.625 mm, rotation time 0.5 sec, scan time 250
msec, tube voltage 120 kV, pitch 0.275, tube current
165 mAS) with the workstation (Philiphs, Extended
Brilliance TM, workspace, Release 1.0.1.). The calcium score were measured according the method of
Agatston (11). On the basis of the Agatston score, the
patients were separated into the categories proposed
by Rumberger et al. (12): no calcification (calcium score 0), minimal calcification (calcium score 1-10), mild
calcification (calcium score 11-100), moderate calcification (calcium score 101-400), or severe calcification
(calcium score >400). Subjects with a heart rate over
70 were administered intravenous esmolol in order to
slow down the heart rate and obtain clear images.
The outcome variable for this study was coronary
calcium score. The primary predictor variable was
the time period of warfarin therapy with covariates
of age, LDL-C, HDL-C, total cholesterol, body mass
index (BMI), diagnosis of hypertension, DM, current
and past tobacco use.
In statistical analysis parametric values were expressed as mean±1 standard deviation, and nonparametric values as percentage. Mann Whitney U and
Chi square for non parametric values, and independent samples t test for parametric values were used in
order to compare the groups. Multiple linear regression analysis was used to evaluate the patient contributions of age, gender, serum lipids and presence of
warfarin treatment for the amount of coronary calcium. A p value <0.05 was set significant.
Results
The average heart rates of the subjects were 68±7
and 66±6 for Group I, and Group II, respectively. The
prosthetic heart valves were mitral, aortic, and mitral
and aortic prosthesis in 20 (%52), 6 (15%), and 13
(33%) patients, respectively. Of these patients with
mitral prosthesis 15 were female, and 5 were male.
Aortic valve prosthesis was observed in 3 female and
3 male patients. All the patients with mitral and aortic prostheses were female.
The mean total calcium score of Group I and Group
II were 27±55 and 21±60, respectively. In Group I, 19
patients (48.7%) did not have any detectable calcium, 12 (30.8%) had minimal calcification, 5 (12.8%)
had mild calcification and 3 (7.6%) had moderate
calcification at their coronary arteries. None of the
cases had severe coronary calcification. On the other
hand, no calcium was detected in 16 subjects, minimal calcification in 3 subjects, and moderate calcification in 3 subjects in Group II. Neither total calcium
score difference between Group I and II, nor the risk
classification of the coronary calcification was significantly different between the groups. The data and the
statistical analysis of the groups are shown in Table I.
We found no correlation between the coronary calcification score and the duration of warfarin therapy
(r=0.09). The subgroup analysis of the study group revealed that mean warfarin therapy period was 80±53
months in patients with no coronary calcification,
and 44±35 months in patients with minimal calcification, and 50±36 months in patients with mild cal-28 • March 2011 • Gulhane Med J Sağ et al.
cification and 103±10 months in patients with moderate calcification, and the difference between the
subgroups was not significant (χ
2
=51.297; p=0.462).
Multiple linear regression analysis revealed that there
was no relationship between the length of warfarin
use and calcification score (p=0.299). There were also
no correlations between coronary calcification and
the BMI (r=0.01; p=0.647), triglyceride levels (r=0.04;
p=0.343), and age (r=0.08; p=0.217), whereas the
correlation coefficients between coronary calcification and gender, total cholesterol, LDL-cholesterol
and HDL-cholesterol were r=0.30 (p=0.01), r=0.26
(p=0.022), r=0.23 (p=0.02) and r=-0.22 (p=0.018),
respectively. Multiple linear regression analysis revealed that male gender (p=0.031), arterial hypertension (p=0.04), and LDL-C (p=0.022) were independent
predictors of the coronary calcification.
Discussion
This study has shown that long-term warfarin use
is not associated with excessive coronary calcification. Coronary calcification has been accepted as
a sensitive and specific marker for atherosclerosis.
Calcification of the intimal layer of the artery is observed in the setting of atherosclerotic plaque (13,14).
Several models have been proposed to explain the development of intimal calcification: a model of an active process in which there is formation of bone like
materials (15,16), a physicochemical model, and an
arterial osteoclast-like cell model (17). According to
the physicochemical model, there are inhibitors in
arteries that prevent precipitation of calcium, at least
in part by chelating calcium ions (18-22). Several proteins that might modulate calcium precipitation in
the extracelluler fluid contain γ-carboxyglutamic acid
(Gla) amino acid residues, which can chelate or bind
calcium ions. Gla-containing proteins also include
bone proteins such as matrix Gla protein (MGP) (23-
27). Numerous studies are consistent with the idea
that MGP prevents precipitation of calcium mineral
in arteries (1, 28-31).
The activity of MGP is dependent on carboxylation
using Vit K as a cofactor. Although in an animal model, low intake of Vit K has been shown to accelerate
vascular calcification via decreased MGP activity, it
has been found that dietary Vit K1 (phylloquinone)
intake was not correlated with coronary calcification
(32). The hypothesis that, warfarin, Vit-K antagonist,
inhibits the MGP activity and cause vascular calcification has been tested in animal models and rapid calcification of elastic lamellae and of aortic heart valves
have been observed (1,31). Although some case reports have claimed that warfarin therapy is associated with coronary and tracheobronchial calcification
(33,34), results of our study reveal that warfarin treatment is not associated with coronary calcification in
adult humans. Strikingly, children under warfarin
therapy are susceptible for tracheobronchial calcification (35). Similarly in an animal study (31), warfarin treatment caused massive focal calcification of
the artery media in 20-day-old rats and less extensive
focal calcification in 42-day-old rats, and finally no
calcification was found in 10-month-old adult rats.
Price et al. have concluded that warfarin-induced calcification may be promoted by increase in serum calcium or phosphate, which are found in higher serum
levels in younger rats or by metabolic processes that
are activated by growth and by vitamin D (31). This
Table I. The data of the study group and control group and their statistical comparison
Group I
(Study group)
Group II
(Control)
p value
Gender
Male
Female
11 (28.2%)
28 (71.8%)
9 (32.1)
19 (67.9)
0.730
Age (year) 54±15 53±12 0.238
Hypertension 5 (12.8%) 0
Diabetes mellitus 4 (10.3%) 0
Body mass index (kg/m
2
) 25.54±3.56 24.71±3.77 0.728
Cholesterol (mg/dl) 186.66±42.39 192.79±33.07 0.380
Triglyceride (mg/dl) 129.14±46.74 134.62±42.05 0.588
HDL-C (mg/dl) 46.7±11.37 41.34±12.84 0.640
LDL-C (mg/dl) 112.35±38.34 118.03±34.24 0.944
Total coronary calcium score 26.63±54.60 20.90±59.70 0.408
Risk classification of coronary calcification 19 (48.7%) no calcification
12 (30.8%) minimal calcification
5 (12.8%) mild calcification
3 (7.6%) moderate calcification
16 (57.1%) no calcification
6 (21.4%) minimal calcification
3 (10.7%) mild calcification
3 (10.7%) moderate calcification
0.666Volume 53 • Issue 1 Role of warfarin in coronary calcifi cation • 29
finding may explain the absence of relation between warfarin use and coronary calcification. All of our
patients were adult, whose bone development were
completed. Besides, the patients with disease that
might increase serum calcium level were excluded.
Another explanation might be that arterial calcifying effect of warfarin may be dose dependent. The
doses used in animal studies are higher than the therapeutic dose for humans, for example, Price et al.
administered 15 mg warfarin/100 g body weight/day
to rats (31). In our study, maximum daily dose was
0.028 mg/100 g body weight. This is very low compared to the study of Price et al.
The lack of correlation might be also explained by
the different structure of the coronary arteries. In
experimental models, warfarin-induced vascular calcification is attributed to precipitation of calcium at
elastic lamellae of the media of artery (1,31). Because
human coronary arteries have thin media and elastic lamellae, the calcifying effect might have been so
small to be detected by tomography.
Previous studies indicated a relationship between hyperlipidemia and vascular, and also cardiac
valve calcification independent of age (36-39). Our
study also showed that coronary calcification is
positively correlated with gender, total cholesterol,
LDL-cholesterol and negatively correlated with HDLcholesterol, which is similar to those papers previously reported. There was no correlation found between
coronary calcification and BMI, triglyceride levels,
and age in our study. These results are coherent to
that of those papers.
A very recent study done by multidetector computerized tomography found a relationship between
warfarin and aortic valvular, and coronary calcification (7). In that study while treatment of warfarin was
only independent predictor for aortic valvuler calcification, male gender, hypertension, and warfarin
treatment were independent predictors for coronary
calcification. In our study we evaluated only coronary calcification, and determined that male gender,
hypertension, and LDL-C but not warfarin treatment
were independent predictors for coronary calcification. The discrepancy between two studies may be
explained by our younger patients and shorter duration of warfarin treatment. However, more important
difference was the patient population of our study,
which was composed of those with all heart valve replacement due to rheumatic heart disease. The
study group of the paper published by Koos et al. included those with old aortic valve stenosis, probably
due to aortic calcification (7). Wang et al. declared
that gene polymorphism of vitamin K epoxide reductase subunit 1 is associated with stroke, coronary
artery disease, and aortic dissection (40). The risk of
calcification induced by warfarin may be associated
with this pronounced gene polimorphism. Readily
there are no data about this idea. We specially chose
the patients with heart-valve prosthesis because warfarin is widely used in this population and coronary
calcification associated with warfarin use is an important concern in these patients. Our study revealed
that long term usage of warfarin is safe with regard to
coronary calcification.
In conclusion the data of our study imply that warfarin use is not associated with increased coronary calcification when compared to the control group. The
lack of difference could be attributed to susceptibility to calcification. However, the lack of relationship
between the duration of warfarin use and coronary
calcification score also supports the conclusion that
warfarin use in therapeutic doses is not associated
with coronary calcification, at least in adult humans.