Clinical Chemistry 54:5
851–857 (2008)
Association between the UGT1A1
TA-Repeat Polymorphism and Bilirubin
Concentration in Patients with
Intermittent Claudication:
Results from the CAVASIC Study
Barbara Rantner,1,2 Barbara Kollerits,1 Marietta Anderwald-Stadler,1,3 Peter Klein-Weigel,4 Ingrid Gruber,2
Anke Gehringer,1 Markus Haak,1 Mirjam Schnapka-Köpf,5 Gustav Fraedrich,2 and Florian Kronenberg1*
BACKGROUND: Bilirubin has antioxidative and cytopro-
tective properties. Low plasma concentrations of bili-
rubin are reportedly associated with the development
of coronary and cerebrovascular disease, and bilirubin
concentrations are strongly correlated with the enzyme
activity of the hepatic uridine diphosphate glucurono-
syltransferase (UGT1A1). The activity of UGT1A1 is
influenced by a TA-repeat polymorphism in the pro-
moter of the UGT1A1 gene (UDP glucuronosyltrans-
ferase 1 family, polypeptide A1). In a case-control
study, we investigated the association between the
UGT1A1 polymorphism, bilirubin concentration, and
intermittent claudication.
METHODS: We included 255 consecutive male patients
presenting with intermittent claudication in the inves-
tigation and matched the patients by age and diabetes
mellitus with 255 control individuals.
RESULTS: Plasma bilirubin concentrations were signifi-
cantly lower in patients than in controls [mean (SD),
12.5 (5.3) ␮mol/L vs 15.4 (7.9) ␮mol/L; P ⬍ 0.001]. We
found a clear association between the number of TA
repeats and plasma bilirubin concentration. Consider-
ing the 6/6 TA-repeat genotype as the wild type, we
observed a slight increase in bilirubin concentration
individuals with the heterozygous 6/7 genotype and
pronounced increases for those with the homozygous
7/7 genotype. This association occurred in both con-
trols and patients; however, patients and controls were
not significantly different with respect to UGT1A1 TA-
repeat genotype frequencies.
1
Division of Genetic Epidemiology, Department of Medical Genetics, Molecular
and Clinical Pharmacology, Innsbruck Medical University, Innsbruck, Austria;
2
Department of Vascular Surgery, Innsbruck Medical University, Innsbruck,
Austria; 3 3rd Medical Department of Metabolic Diseases and Nephrology,
Hietzing Hospital, Vienna, Austria; 4 DRK-Klinik Mark Brandenburg, Department
of Angiology, Berlin, Germany; 5 Central Laboratory for Medical and Chemical
Laboratory Diagnostics, Innsbruck Medical University, Innsbruck, Austria.
* Address correspondence to this author at: Division of Genetic Epidemiology,
Molecular Diagnostics and Genetics
CONCLUSIONS: Our study of a well-phenotyped group of
patients with intermittent claudication and control in-
dividuals revealed a clear association between low bili-
rubin concentrations and peripheral arterial disease
but no association between the UGT1A1 polymor-
phism and the disease.
© 2008 American Association for Clinical Chemistry
The heme degradation product bilirubin has potent
antioxidative capacities that play a protective role in
various diseases, including coronary artery and vascu-
lar disease (1, 2 ). Studies on the relationship between
bilirubin concentration and cardiovascular disease
have shown an inverse relationship between bilirubin
concentration and atherosclerosis (3–13 ). Low biliru-
bin concentrations have also been independently and
inversely related to impaired flow-mediated vasodila-
tation and increased carotid intima-media thickness in
clinically healthy subjects (14, 15 ). These in vivo obser-
vations and experimental studies (16 ) suggest a protec-
tive effect of high physiological bilirubin concentra-
tions on early steps in atherogenesis.
By linkage analysis we recently identified a region
on chromosome 2q significantly linked to bilirubin
concentration (17 ), a finding that was subsequently
confirmed by investigators for the Framingham Heart
Study (18 ). The identified chromosomal region har-
bors the gene for the uridine diphosphate glucurono-
syltransferase (UGT1A1,6 UDP glucuronosyltrans-
ferase 1 family, polypeptide A1), the major enzyme of
Department of Medical Genetics, Molecular and Clinical Pharmacology, Inns-
bruck Medical University, Schöpfstr. 41, A-6020 Innsbruck, Austria. Fax (43)
512 9003-73560 or (43) 512 9003-73561; e-mail Florian.Kronenberg@i-med.
ac.at.
Received December 12, 2007; accepted February 15, 2008.
Previously published online at DOI: 10.1373/clinchem.2007.102046
6
Human Genes: UGT1A1 (aliases: GNT1, HUG-BR1, UDPGT, UGT1, UGT1A), UDP
glucuronosyltransferase 1 family, polypeptide A1.
851
bilirubin glucuronidation, which mainly determines
bilirubin elimination in humans. The activity of
UGT1A1 is substantially influenced by a TA-repeat
polymorphism in the UGT1A1 promoter (19 ). Indi-
viduals homozygous for 7 TA repeats (7/7) have lower
promoter activity and higher subsequent bilirubin
concentrations than heterozygous (6/7) or wild-type
homozygous (6/6) individuals (19 –21 ). In an investi-
gation of the correlation between the TA-repeat poly-
morphism and the incidence of cardiovascular and
coronary heart disease in the Framingham Offspring
Study, we observed a statistically significant decreased
risk for individuals with the 7/7 genotype (8 ).
Previous studies of bilirubin have primarily fo-
cused on patients with coronary artery disease. To our
knowledge, the present study is the first to investigate
an association between plasma bilirubin concentration
and the UGT1A1 TA-repeat polymorphism in a male
cohort with symptomatic peripheral arterial disease
(PAD)7 and an age- and diabetes-matched group of
control individuals.
Materials and Methods
STUDY PARTICIPANTS AND STUDY DESIGN
The CAVASIC (Cardiovascular Disease in Intermittent
Claudication) Study is a prospective case-control study
initiated in 2002 to identify cardiovascular risk factors
in patients with intermittent claudication. Patient and
controls were enrolled in 2 clinical centers, the Depart-
ment of Vascular Surgery, Medical University Inns-
bruck, and the 3rd Medical Department of Metabolic
Diseases and Nephrology, Hietzing Hospital, Vienna,
Austria. Participants are to be followed for at least 6
years. We describe our study of the association of bili-
rubin concentration and the UGT1A1 polymorphism
with the presence of symptomatic PAD at baseline.
Patients (n ⫽ 255) were consecutively included in
the study when they presented with or had a history of
intermittent claudication (PAD IIa or IIb, according to
the criteria of Fontaine), regardless of whether they had
already undergone a treatment procedure (bypass sur-
gery or intervention). Patients were excluded from the
study for any of the following reasons: presence of
acute or critical limb ischemia (Fontaine III or IV),
impaired liver function with increased enzyme concen-
trations (aspartate aminotransferase ⬎50 U/L, alanine
amino transferase ⬎25 U/L, ␥-glutamyltransferase
⬎60 U/L), impaired kidney function with serum creat-
7
Nonstandard abbreviations: PAD, peripheral arterial disease; UGT1A1, uridine
diphosphate glucuronosyltransferase; T2DM, type 2 diabetes mellitus; ABI,
ankle-brachial index.
852 Clinical Chemistry 54:5 (2008)
inine ⬎133 ␮mol/L, malignancy, past organ transplan-
tation, and therapy with nicotinic acid or cortico-
steroids.
We recruited 255 control individuals from the
same geographic region and matched them with the
patients with respect to age and presence of type 2 dia-
betes mellitus (T2DM). All members of the control
group had volunteered to participate in the study after
the publication of an invitation in newspapers. We ap-
plied the same exclusion criteria to the control group as
we used for the patients. Control individuals with
symptomatic PAD were excluded, but those with a his-
tory of cardiovascular disease were allowed to partici-
pate. Eighteen of the 255 controls either had a positive
cardiovascular history for angina pectoris according to
their responses on the Rose questionnaire (22 ) or had
had documented cardiovascular events or procedures,
such as myocardial infarction, aortocoronary bypass
surgery, percutaneous transluminal coronary angio-
plasty, and/or coronary angiography or stroke.
Neither the patients nor the controls had acute
illnesses or clinically detectable inflammatory pro-
cesses at the time of enrollment. All study participants
provided written informed consent, and the ethics
committee of the participating study centers approved
the examination protocol.
To minimize interobserver bias, we had a single
physician at each of the 2 clinical centers conduct all
of the interviews and examinations; these physicians
were specially trained in vascular examinations and
echocardiography.
BASELINE VASCULAR EXAMINATION
For the measurement of the ankle-brachial index
(ABI), the systolic brachial blood pressure was initially
measured once on both arms, and 2 additional blood
pressure measurements were made on the arm with the
higher systolic value. We used the mean of these 2 ad-
ditional measurements on the arm with the higher sys-
tolic value for further calculations. Systolic blood pres-
sures on the lower extremity were measured 3 times for
each artery (dorsalis pedis artery and tibialis posterior
artery, on both the left and right ankles). The mean of
the second and third measurements for each site was
used to calculate the ABI for each of the 4 lower-ex-
tremity sites. The ABI was calculated as the ratio of the
systolic blood pressure (mean of second and third mea-
surements) of each of the 4 sites to the mean systolic
blood pressure of the arm. We used the lowest ABI
value from the 4 sites for further data analysis to yield a
higher sensitivity of the ABI (23 ) and to be in line with
the Reduction of Atherothrombosis for Continued
Health (REACH) Registry (24 ).
The Edinburgh questionnaire was used to iden-
tify symptomatic intermittent claudication (25 ). Pa-
Bilirubin and Peripheral Arterial Disease
tients who presented with symptoms of intermittent
claudication and had an ABI ⬍0.90 were considered
to have PAD. Furthermore, we performed a pulse
volume recording and evaluated walking distance with
a standardized constant-load treadmill examination
(12% acceleration and 3.0 km/h). If any further ther-
apy was planned, additional ultrasound scanning or
magnetic resonance imaging of the arteries of the
lower extremity was done. Conventional angiography
was performed in cases requiring further endovascular
treatment.
OTHER PHENOTYPIC CHARACTERIZATION
Demographic data, clinical history, smoking status, al-
cohol consumption, diet, amount of leisure, physical
activity during work, and atherosclerosis risk profile
were recorded via a standardized interview. Current
medications were recorded at the baseline exam.
All participants underwent a clinical examination
that was focused on the heart and arteries and included
electrocardiographic and echocardiographic evaluations.
Participants received a diagnosis of diabetes mel-
litus if their fasting plasma glucose concentration was
⬎6.99 mmol/L and/or they were being treated with an-
tidiabetes drugs. Participants were considered hyper-
tensive when the systolic blood pressure was ⱖ140
mmHg, if the diastolic blood pressure was ⱖ90 mmHg,
and/or if they were being treated with antihypertension
drugs.
LABORATORY MEASUREMENTS
Plasma and serum samples treated with EDTA and ci-
trate were obtained after an overnight fasting period of
10 –14 h, immediately processed, centrifuged, and
stored in aliquots at ⫺80 °C until laboratory measure-
ments were made. Bilirubin was measured immedi-
ately after blood collection via a colorimetric method
(diazotized sulfanilic acid reaction; Roche Diagnostics
reagent sets) in a Modular P800 analyzer.
GENOTYPING
DNA was extracted from whole-blood samples by a
salting-out method (Invisorb Blood Universal Kit; In-
vitek). The UGT1A1 promoter polymorphism was an-
alyzed as recently described (8 ). Samples were geno-
typed within the Genotyping Unit of the Gene
Discovery Core Facility at the Innsbruck Medical Uni-
versity, Innsbruck, Austria.
STATISTICAL ANALYSIS
Variables for the patient and control groups were eval-
uated statistically by means of unpaired Student t-tests,
nonparametric Wilcoxon rank sum tests, and the Pear-
son ␹2 test. Besides the 6 and 7 TA-repeat alleles for the
UGT1A1 polymorphism, we also observed 6 individu-
als with a 5 TA-repeat allele and 2 persons with an 8
TA-repeat allele. Because of the low frequencies of
these alleles and functional studies that revealed de-
creasing promoter activity with increasing numbers
of TA repeats, we combined the 5 and 6 TA-repeat
alleles and the 7 and 8 TA-repeat alleles for statistical
analyses. Variables contributing to PAD status were
identified by logistic regression analysis, and an ad-
justed general linear regression model was used to
estimate the proportion of the variation in bilirubin
concentration explained by different variables. The
full model included the UGT1A1 polymorphism
(coded as a recessive model for the 7 TA repeats),
age, LDL cholesterol, smoking status, T2DM, and glu-
cose. A submodel included these variables without the
UGT1A1 polymorphism. The difference in r2 between
the 2 models was considered to be the amount of the
variation in bilirubin concentration explained by the
UGT1A1 polymorphism.
Statistical analyses were performed with the Statisti-
cal Package for the Social Sciences for Windows (SPSS)
version 15.0 and SAS version 9.1 (SAS Institute).
Results
BASELINE CHARACTERISTICS
Between July 2002 and July 2006, we included 255 un-
related male patients (age range, 35–70 years) who pre-
sented with intermittent claudication graded as stage
IIa or IIb according to the Fontaine classification. Thir-
ty-eight of these patients had T2DM. The patients and
the age- and T2DM-matched controls had similar
T2DM durations [mean (SD), 11.7 (7.4) years vs
10.7 (8.4) years] and similar glycosylated hemoglobin
(HbA1c) values [7.5% (1.4%) vs 7.4% (1.1%)]. These
differences were not statistically significant.
Table 1 summarizes the characteristics of the pa-
tients and control individuals. The patient group
had a significantly higher frequency of smokers and
significantly higher triglyceride and creatinine con-
centrations and lower HDL cholesterol and albumin
concentrations than the control group. Hyperten-
sion frequency and blood pressures were markedly
higher in the patient group.
BILIRUBIN CONCENTRATIONS AND THE UGT1A1
POLYMORPHISM IN PATIENTS AND CONTROLS
Patients with PAD had significantly lower bilirubin
concentrations than the controls [mean (SD), 12.5 (5.3)
␮mol/L vs 15.4 (7.9) ␮mol/L; P ⬍ 0.001] (Table 2). Be-
cause the control group included 18 individuals with a
history of cardiovascular disease, we performed a sen-
sitivity analysis and excluded these 18 control individuals;
nevertheless, the mean bilirubin concentration for the
control group remained unchanged [15 (7.7) ␮mol/L].
Clinical Chemistry 54:5 (2008) 853
 Table 1. Baseline clinical and laboratory data for patients with peripheral arterial occlusive disease and for
control individuals matched by age and diabetes mellitus.a

Controls (n ⴝ 255) Patients (n ⴝ 255) P

Age, years 57.2 (9.2) 58.0 (7.0) NS

Body mass index, kg/m 26.8 (5.7) 26.8 (4.0) NS
Smoking (nonsmokers/former smokers/ 114/99/42 (44.7/38.8/16.5) 19/109/124 (7.5/43.3/49.2) ⬍0.001
current smokers), n (%)
Total cholesterol, mmol/L 5.39 (0.906) 5.34 (1.06) NS
LDL cholesterol, mmol/L 3.52 (0.855) 3.44 (0.958) NS
HDL cholesterol, mmol/L 1.54 (0.451) 1.28 (0.321) ⬍0.001
Triglycerides, mmol/Lb 1.51 (0.915) 1.94 (1.40) ⬍0.001
关0.904, 1.29, 1.75兴 关1.06, 1.51, 2.41兴
Glucose, mmol/L 5.83 (1.72) 6.00 (1.61) 0.006
Creatinine, ␮mol/L 90 (12) 88 (17) ⬍0.001
Albumin, g/L 45.5 (4) 44.6 (5) 0.014
Systolic blood pressure, mmHg 140 (17) 150 (20) ⬍0.001
Diastolic blood pressure, mmHg 82 (9) 83 (10) 0.180
Hypertension, n (%)c 153 (60.5) 214 (84.9) ⬍0.001
Ankle-brachial indexd 1.04 (0.15) 0.73 (0.24) ⬍0.001
a
Data are expressed as the mean (SD) except where noted.
b
Data are expressed as the mean (SD) 关25th, 50th, 75th percentiles兴.
c
Hypertension was defined as a systolic blood pressure ⱖ140 mmHg, a diastolic blood pressure ⱖ90 mmHg, and/or receiving antihypertension treatment.
d
The lowest ABI value from the 4 sites was used for data analysis (see Materials and Methods). Individuals with ABI values ⬎1.30 were excluded from this analysis.

There was a clear association between the TA-re- the homozygous 7/7 genotype. This association was ap-
peat polymorphism in the UGT1A1 promoter and bil- parent in both the controls (12.3 ␮mol/L, 14.9 ␮mol/L,
irubin concentration. Considering the 6/6 TA-repeat and 26.7 ␮mol/L, respectively; P ⬍ 0.001) and the pa-
genotype as the wild type, we observed a slight increase tients (10.9 ␮mol/L, 12.5 ␮mol/L, and 16.6 ␮mol/L,
in bilirubin concentration in individuals with the het- respectively; P ⬍ 0.001) (Fig. 1). The proportion of the
erozygous 6/7 genotype and a pronounced increase for total variation in bilirubin concentration explained by
the TA-repeat polymorphism was 12.3% in the pa-
tients after adjusting for covariates, compared with
Table 2. Bilirubin concentrations and the UGT1A1 27.9% in the control group (Table 3).
TA-repeat polymorphism in controls and patients A comparison of the bilirubin concentrations for
with peripheral arterial occlusive disease. the patient and control groups by genotype revealed
lower bilirubin concentrations in the patients than in
Controls Patients the controls for each genotype (Fig. 1). The difference
(n ⴝ 255) (n ⴝ 255) P
was most pronounced for individuals with the 7/7 TA-
Bilirubin, ␮mol/La 15.4 (7.9) 12.5 (5.3) ⬍0.001 repeat genotype (16.6 ␮mol/L vs 26.7 ␮mol/L; P ⬍
关9.9, 13.7, 18.1兴 关9.1, 11.5, 14.7兴 0.001). The patients and control individuals showed no
UGT1A1 TA-repeat differences with respect to genotype frequencies, how-
genotype frequency, ever, regardless of whether the 18 control individuals
n (%)
with a history of cardiovascular disease were included
6/6 94 (37.0) 91 (36.8)
in the analysis (Table 2) or not (data not shown).
6/7 127 (50.0) 119 (48.2)
7/7 33 (13.0) 37 (15.0) 0.803 MULTIPLE LOGISTIC REGRESSION ANALYSIS
a
Data are expressed as the mean (SD) 关25th, 50th, 75th percentiles兴.
Finally, we investigated whether bilirubin concentra-
tion was associated with the case-control status. After

854 Clinical Chemistry 54:5 (2008)

Bilirubin and Peripheral Arterial Disease

diseases by protecting against oxidative stress, possibly

P < 0.001
30 Controls Patients 26.7
through the action of bilirubin or in conjunction with
bilirubin metabolism (2, 26 ). Several studies have de-
Bilirubin (µmol/L)

25

20
P < 0.001 scribed an association between low bilirubin concen-
P = 0.027 16.6
14.9 trations and coronary heart disease (3–13 ). The asso-
15 12.3 12.5
10.9
10
ciation of bilirubin with atherosclerosis might be
explained by in vitro findings that have shown bili-
5
rubin to have antioxidative and cytoprotective prop-
0
erties (27, 28 ). Bilirubin scavenges peroxyl radicals
6/6 6/7 7/7
and suppresses oxidation in liposomes more efficiently
UGT1A1 TA-repeat polymorphism
than ␣-tocopherol (27, 28 ). It also has antithrombotic
properties (29 ) and modulates macrophage activation
Fig. 1. Plasma bilirubin concentrations in control in- within atherosclerotic lesions (30 ). Because the en-
dividuals and patients with PAD and stratified by 3 zyme activity of uridine diphosphate glucuronosyl-
TA-repeat genotypes (6/6, 6/7, and 7/7) of the transferase is strongly influenced by the investigated
UGT1A1 promoter polymorphism. TA-repeat polymorphism (19 –21 ), the hypothesis of
an association between this polymorphism and athero-
sclerotic outcome is attractive.
adjustment for age, we found a 1.7-␮mol/L increase in The highly significant association between biliru-
bilirubin concentration decreased the probability of bin concentration and PAD noted in our study is con-
being a patient by 11.5% (Table 4, model 1). This asso- vincing and in line with the majority of the other stud-
ciation was still significant after we adjusted for other ies on coronary heart disease (3–13 ). Furthermore, we
risk factors, such as smoking status, HDL cholesterol, found a strong association between the UGT1A1 poly-
glucose, creatinine, and hypertension (Table 4). morphism and bilirubin concentration; however, we
were surprised that we could find no association be-
Discussion tween the UGT1A1 polymorphism and PAD. In a re-
cent Framingham Heart Study investigation, a strong
To our knowledge, this study is the first to investigate association was observed between the 7/7 TA-repeat
plasma bilirubin concentration and the UGT1A1 pro- genotype and a lower risk of cardiovascular disease.
moter TA-repeat polymorphism in a cohort of patients During a 24-year follow-up, carriers of this geno-
with PAD and an age- and diabetes-matched control type had about one third the risk for cardiovascular
group. We observed significantly lower bilirubin con- disease and coronary heart disease as carriers of the 6
centrations in patients than in controls, and the allele (8 ). This finding is in contrast to the longitudinal
UGT1A1 polymorphism was strongly associated with population– based Rotterdam Study of elderly indi-
bilirubin concentration in both patients and controls. viduals, which found no association of either the TA-
Interestingly, the polymorphism was not associated repeat polymorphism or bilirubin concentration with
with PAD. coronary heart disease (31 ). Similarly, the ECTIM
Activation of heme oxygenase and the heme cata- case-control study of myocardial infarction found no
bolic pathway are both strongly influenced by genetics association of the UGT1A1 7 TA-repeat allele with cor-
and have been proposed to have beneficial effects on onary heart disease (32 ).
The differences between these studies and our
study of PAD patients in comparison with the Fra-
Table 3. Percentage of total variation in bilirubin
mingham Heart Study are obvious: the Framingham
concentration explained by the UGT1A1 TA-repeat
Heart Study is a population-based prospective cohort
polymorphism.
study of individuals free of cardiovascular disease at
baseline. Although the Rotterdam Study was a prospec-
r 2, full r 2, r 2 explained by
tive study, participants had a mean age of nearly 70
modela submodelb polymorphism years at baseline, which makes a survival bias highly
Controls 35.2% 7.3% 27.9% likely. We cannot exclude a survival bias in our study
because concomitant PAD and cardiovascular disease
Patients 18.6% 6.3% 12.3%
is the rule rather than the exception.
a
Full model is adjusted for age, LDL cholesterol, smoking status, T2DM, In this context we consider the following observa-
glucose, and UGT1A1 TA-repeat polymorphism. tion highly interesting: Although we observed an asso-
b
Submodel is the same as the full model but without adjustment for the
UGT1A1 TA-repeat polymorphism.
ciation between the TA-repeat polymorphism and
bilirubin concentration in both the controls and the

Clinical Chemistry 54:5 (2008) 855

 Table 4. Logistic regression analysis for predicting PAD.

Model 1 Model 2 Model 3

Variable (increment) Odds ratio (95% CIa) P Odds ratio (95% CI) P Odds ratio (95% CI) P

Bilirubin (1.7 ␮mol/L) 0.885 (0.840–0.933) ⬍0.001 0.913 (0.864–0.964) 0.001 0.914 (0.862–0.970) 0.003
Age (1 year) 1.014 (0.992–1.037) 0.218 1.025 (1.001–1.050) 0.042 1.019 (0.990–1.049) 0.193
Smoking (current smoking) — 4.510 (2.936–6.927) ⬍0.001 4.105 (2.526–6.672) ⬍0.001
HDL cholesterol (0.026 mmol/L) — — 0.960 (0.946–0.975) ⬍0.001
Creatinine (88.4 ␮mol/L) — — 0.271 (0.078–0.939) 0.039
Glucose (0.056 mmol/L) — — 0.998 (0.991–1.006) 0.641
Hypertension (yes) — — 4.500 (2.632–7.696) ⬍0.001
a
CI indicates confidence interval.

cases, the relative and absolute differences in bilirubin
concentration between individuals with the 6/6 and 7/7
TA-repeat genotypes was much more pronounced in
the controls than in the patients. Control individuals
with the 7/7 TA-repeat genotype had bilirubin concen-
trations that were about 120% higher than controls
with the 6/6 TA-repeat genotype. This difference in
PAD patients was only about 52%, however (Fig. 1).
We speculate that the lower relative difference in bili-
rubin concentration in PAD patients with the 7/7 ge-
notype could be explained by a selection bias between
the patients and controls within the 7/7 genotype
group. If this speculation is accurate, we expect the
bilirubin concentrations of patients with the 7/7 geno-
type to fall in the lower range of expected values for this
genotype, whereas the bilirubin concentrations in the
control group would fall more toward the center of the
expected range. Such a phenomenon might account for
the observation that the TA-repeat polymorphism ex-
plained only about half of the bilirubin variance in
PAD patients compared with controls (12.6% vs
25.6%, Table 3). If we consider that only 12.6% of the
variance in bilirubin concentration is explained by the
TA-repeat polymorphism, it is no longer surprising
1. Morita T. Heme oxygenase and atherosclerosis.
Arterioscler Thromb Vasc Biol 2005;25:1786 –
95.
2. Vitek L, Schwertner HA. The heme catabolic path-
way and its protective effects on oxidative stress-
mediated diseases. Adv Clin Chem 2007;43:1–57.
3. Schwertner HA, Jackson WG, Tolan G. Associa-
tion of low serum concentration of bilirubin with
increased risk of coronary artery disease. Clin
Chem 1994;40:18 –23.
4. Breimer LH, Wannamethee G, Ebrahim S, Shaper
AG. Serum bilirubin and risk of ischemic heart
disease in middle-aged British men. Clin Chem
1995;41:1504 – 8.
5. Hopkins PN, Wu LL, Hunt SC, James BC, Vincent
that we cannot detect an association of this polymor-
phism with PAD outcome, although the association of
bilirubin concentration with PAD is relatively strong. It
is possible that thousands of patients and controls
would be needed in a case-control study to detect such
an association, if it exists. The clear association ob-
served in the Framingham Heart Study might be ex-
plained by its having a prospective study design that is
less confounded by most kinds of selection and survival
bias than case-control designs.
In summary, our study of a well-phenotyped
group of PAD patients and control individuals shows a
clear association between low bilirubin concentrations
and peripheral arterial occlusive disease.
Grant/Funding Support: This study was supported by
grants from the “Austrian Nationalbank” (Project
9331) and the Austrian Heart Fund, and by the
“Genomics of Lipid-associated Disorders—GOLD” of
the “Austrian Genome Research Programme GEN-
AU” to F.K. B.R. was supported by a DOC-fFORTE
scholarship from the Austrian Academy of Sciences.
Financial Disclosures: None declared.
References
GM, Williams RR. Higher serum bilirubin is asso- 8. Lin J-P, O’Donnell CJ, Schwaiger JP, Cupples A,
ciated with decreased risk for early familial cor- Lingenhel A, Hunt SC, et al. Association between
onary artery disease. Arterioscler Thromb Vasc the UGT1A1*28 allele, bilirubin levels, and coro-
Biol 1996;16:250 –5. nary heart disease in the Framingham Heart
6. Hunt SC, Kronenberg F, Eckfeldt JH, Hopkins PN, Study. Circulation 2006;114:1476 – 81.
Myers RH, Heiss G. Association of plasma biliru- 9. Troughton JA, Woodside JV, Young IS, Arveiler D,
bin with coronary heart disease and segregation Amouyel P, Ferrieres J, et al. Bilirubin and coro-
of bilirubin as a major gene trait: the NHLBI nary heart disease risk in the Prospective Epi-
Family Heart Study. Atherosclerosis 2001;154: demiological Study of Myocardial Infarction
747–54. (PRIME). Eur J Cardiovasc Prev Rehabil 2007;14:
7. Djousse L, Levy D, Cupples LA, Evans JC, 79 – 84.
D’Agostino RB, Ellison RC. Total serum bilirubin 10. Novotný L, Vitek L. Inverse relationship between
and risk of cardiovascular disease in the Framing- serum bilirubin and atherosclerosis in men: a
ham Offspring Study. Am J Cardiol 2001;87: meta-analysis of published studies. Exp Biol Med
1196 –200. (Maywood) 2003;228:568 –71.

856 Clinical Chemistry 54:5 (2008)

Bilirubin and Peripheral Arterial Disease
11. Levinson SS. Relationship between bilirubin, apo-
lipoprotein B, and coronary artery disease. Ann
Clin Lab Sci 1997;27:185–92.
12. Schwertner HA, Fischer JR Jr. Comparison of var-
ious lipid, lipoprotein, and bilirubin combinations
as risk factors for predicting coronary artery dis-
ease. Atherosclerosis 2000;150:381–7.
13. Schwertner HA. Association of smoking and low
serum bilirubin antioxidant concentrations. Ath-
erosclerosis 1998;136:383–7.
14. Erdogan D, Gullu H, Yildirim E, Tok D, Kirbas I,
Ciftci O, et al. Low serum bilirubin levels are
independently and inversely related to impaired
flow-mediated vasodilation and increased carotid
intima-media thickness in both men and women.
Atherosclerosis 2006;184:431–7.
15. Vitek L, Novotný L, Sperl M, Holaj R, Spacil J. The
inverse association of elevated serum bilirubin
levels with subclinical carotid atherosclerosis. Ce-
rebrovasc Dis 2006;21:408 –14.
16. Öllinger R, Bilban M, Erat A, Froio A, McDaid J,
Tyagi S, et al. Bilirubin: a natural inhibitor of
vascular smooth muscle cell proliferation. Circu-
lation 2005;112:1030 –9.
17. Kronenberg F, Coon H, Gutin A, Abkevich V,
Samuels ME, Ballinger DG, et al. A genome scan
for loci influencing anti-atherogenic serum biliru-
bin levels. Eur J Hum Genet 2002;10:539 – 46.
18. Lin JP, Cupples LA, Wilson PW, Heard-Costa N,
O’Donnell CJ. Evidence for a gene influencing
serum bilirubin on chromosome 2q telomere: a
genomewide scan in the Framingham study. Am J
Hum Genet 2003;72:1029 –34.
19. Bosma PJ, Chowdhury JR, Bakker C, Gantla S, de
Boer A, Oostra BA, et al. The genetic basis of the
reduced expression of bilirubin UDP-glucurono-
syltransferase 1 in Gilbert’s syndrome. N Engl
J Med 1995;333:1171–5.
20. Monaghan G, Ryan M, Seddon R, Hume R, Burch-
ell B. Genetic variation in bilirubin UPD-glucu-
ronosyltransferase gene promoter and Gilbert’s
syndrome. Lancet 1996;347:578 – 81.
21. Beutler E, Gelbart T, Demina A. Racial variability
in the UDP-glucuronosyltransferase 1 (UGT1A1)
promoter: a balanced polymorphism for regula-
tion of bilirubin metabolism? Proc Natl Acad Sci
U S A 1998;95:8170 – 4.
22. Rose G. The diagnosis of ischaemic heart pain
and intermittent claudication in field surveys. Bull
World Health Organ 1962;27:645–58.
23. Schröder F, Diehm N, Kareem S, Ames M, Pira A,
Zwettler U, et al. A modified calculation of ankle-
brachial pressure index is far more sensitive in
the detection of peripheral arterial disease. J Vasc
Surg 2006;44:531– 6.
24. Ohman EM, Bhatt DL, Steg PG, Goto S, Hirsch AT,
Liau CS, et al. The REduction of Atherothrom-
bosis for Continued Health (REACH) Registry: an
international, prospective, observational in-
vestigation in subjects at risk for atherothrom-
botic events-study design. Am Heart J 2006;151:
786.e1–10.
25. Leng GC, Fowkes FG. The Edinburgh Claudication
Questionnaire: an improved version of the WHO/
Rose Questionnaire for use in epidemiological
surveys. J Clin Epidemiol 1992;45:1101–9.
26. Exner M, Minar E, Wagner O, Schillinger M. The
role of heme oxygenase-1 promoter polymor-
phisms in human disease. Free Radic Biol Med
2004;37:1097–104.
27. Stocker R, Glazer AN, Ames BN. Antioxidant ac-
tivity of albumin-bound bilirubin. Proc Natl Acad
Sci U S A 1987;84:5918 –22.
28. Neuzil J, Stocker R. Free and albumin-bound
bilirubin are efficient co-antioxidants for alpha-
tocopherol, inhibiting plasma and low density
lipoprotein lipid peroxidation. J Biol Chem 1994;
269:16712–9.
29. Lindenblatt N, Bordel R, Schareck W, Menger MD,
Vollmar B. Vascular heme oxygenase-1 induction
suppresses microvascular thrombus formation in
vivo. Arterioscler Thromb Vasc Biol 2004;24:
601– 6.
30. Nakayama M, Takahashi K, Komaru T, Fukuchi
M, Shioiri H, Sato K, et al. Increased expression
of heme oxygenase-1 and bilirubin accumula-
tion in foam cells of rabbit atherosclerotic le-
sions. Arterioscler Thromb Vasc Biol 2001;21:
1373–7.
31. Bosma PJ, van der Meer IM, Bakker CT, Hofman
A, Paul-Abrahamse M, Witteman JC. UGT1A1*28
allele and coronary heart disease: the Rotterdam
Study. Clin Chem 2003;49:1180 –1.
32. Gajdos V, Petit FM, Perret C, Mollet-Boudjemline
A, Colin P, Capel L, et al. Further evidence that
the UGT1A1*28 allele is not associated with
coronary heart disease: the ECTIM Study. Clin
Chem 2006;52:2313– 4.
Clinical Chemistry 54:5 (2008) 857